ASNT Industry Handbook

Aerospace
NDT

American Society for Nondestructive Testing

 Aerospa
ASNT Industry Handbook

ce NDT
Technical Editor
Richard H. Bossi

The American Society for

Nondestructive Testing, Inc.
ASNT Industry Handbook: Aerospace Nondestructive Testing
Richard H. Bossi, Technical Editor

Copyright © 2014 by The American Society for Nondestructive Testing, Inc.

The American Society for Nondestructive Testing, Inc. (ASNT) is not responsible for the authenticity or accuracy of information
herein. Published opinions and statements do not necessarily reflect the opinion of ASNT. Products or services that are advertised or
mentioned do not carry the endorsement or recommendation of ASNT.

No part of this publication may be reproduced or transmitted in any form, by means electronic or mechanical including
photocopying, recording or otherwise, without the expressed prior written permission of The American Society for Nondestructive
Testing, Inc.

IRRSP, NDT Handbook, The NDT Technician, and [Link] are trademarks of The American Society for Nondestructive Testing,
Inc. ACCP, ASNT, Level III Study Guide, Materials Evaluation, Nondestructive Testing Handbook, Research in Nondestructive
Evaluation, and RNDE are registered trademarks of The American Society for Nondestructive Testing, Inc.

First printing 10/14

Errata, if available for this printing, may be obtained from ASNT‘s website, [Link], or as hard copy by mail, free on request
from ASNT at the address below.

ISBN 978-1-57117-339-3 (print)

ISBN 978-1-57117-340-9 (ebook)

Library of Congress Control Number: 2014949019

Printed in the United States of America

Published by:
The American Society for Nondestructive Testing, Inc.
1711 Arlingate Lane
PO Box 28518
Columbus, OH 43228-0518
[Link]

Edited by: Patrick O. Moore, Handbook Editor

Assisted by: Robert B. Conklin, Educational Materials Editor
Hollis Humphries, Technical Publications Supervisor
Joy Grimm, Production Manager
Timothy E. Jones, Senior Manager of Publications

ASNT exists to create a safer world by promoting the profession and technologies of nondestructive testing.

ii
Preface

This ASNT Industry Handbook on Aerospace NDT applications. The chapters on each method
was conceived by ASNT‘s Aerospace Committee as a include more details about the applications. This
means to consolidate, from the large method arrangement will hopefully prove to be practical
volumes of the NDT Handbook, information for the book‘s users in the industry. It is the hope
pertinent to the aerospace community into one of ASNT‘s Aerospace Committee and ASNT‘s
volume. In Spring 2004, Gary L. Workman (chair) Handbook Development Committee that the
reported to the Handbook Development Committee Aerospace NDT handbook will be a useful reference
that members in the Industrial Division of ASNT‘s and addition to the bookshelves of those working in
Technical and Education Council envisioned new, the aerospace industry.
industrially oriented handbooks. This Aerospace NDT A very largenumber of contributors worked
handbook is thefirst. Although it is not intended to on this handbook over ten years, and their names
be comprehensively detailed, the handbook‘s vision are listed on the title pages of particular chapters.
is to provide the fundamental material that would be There were also many advisors and reviewers who
useful to inspectors, engineers, and managers in the contributed to the handbook during its development
aerospace industry who have responsibilities for and helped greatly to shape the content. Their
decisions related to the selectionand implementation efforts, both large and small, are greatly appreciated
of nondestructive testing (NDT) for their products. although their names may not [Link]
The handbook has pulled material from the NDT support of the handbook did come from the
Handbook and from other sources to provide current Aerospace Committee leadership, including Kevin
information on a wide range of NDT methods and Smith, Lisa Brasche, N. David Campbell Jr., B. Boro
applications useful to the aerospace industry. The Djordjevic, and Shant Kenderian. A special thank
challenge in implementing the vision is mainly the you is due to Eric v.K. Hill for his detailed editorial
decision of how to organize the information. review of every chapter, and to Eugene Mechtly for
Material types or components often use multiple his review of metric units. Special thanks are also
NDT methods, and any single NDT method may due to Patrick O. Moore, ASNT handbook editor, and
be used in a variety of aerospace materials or the ASNT staff. I also thank the Boeing Company for
components. As a solution, this volume offers the its support of my activities.
crossreference tablesin Chapter 4. There, the reader
can look up a component or material and see which Richard H. Bossi
methods are applicable. Likewise a method can be Senior Technical Fellow (retired)
reviewed in the cross reference tables for aerospace The Boeing Company

iii
Contributors

Richard H. Bossi, The Boeing Company, Seattle, Matthew J. Golis, Columbus, Ohio (introduction,
Washington (introduction, visual, electromagnetic, glossary)
ultrasonic, radiologic, bond, in-situ NDT for
structural health, microwave, X-ray diffraction) Trey Gordon, The Boeing Company, Seattle,
Washington (radiologic, X-ray diffraction)
Lisa Brasche, Pratt and Whitney, Hartford,
Connecticut (introduction, reliability, magnetic Eric v.K. Hill, New Smyrna Beach, Florida (acoustic
particle, penetrant, ultrasonic) emission)

John C. Brausch, United States Air Force, Wright- David K. Hsu, Iowa State University, Ames, Iowa
Patterson Air Force Base, Ohio (electromagnetic) (bond)

Byron B. Brenden, Richland, Washington Dwight L. Isenhour Jr., Newport News, Virginia
(ultrasonic) (radiologic)

David J. Brown, VM Products, Puyallup, Leanne M. Jauriqui, Vibrant NDT, Albuquerque,

Washington (electromagnetic) New Mexico (resonance)

Clifford Bueno, General Electric, Niskayuna, New Timothy E. Kinsella, Dassault Falcon Jet, Teterboro,
York (radiologic) New Jersey (radiologic)

Donald R. Christina, The Boeing Company, Doron Kishoni, Centennial, Colorado (ultrasonic)
Charleston, South Carolina (visual) Victoria A. Kramb, University of Dayton, Dayton,
Michele D. Dorfman, Lockheed Martin, Fort Worth, Ohio (ultrasonic)
Texas (acoustic emission) Jocelyn A. Langlois, Sigma Transducers, Kennewick,
Thomas C. Dreher, Rolls-Royce, Indianapolis, Washington (ultrasonic)
Indiana (magnetic particle) Mark P. Lessard, Thermo Scientific Portable
John C. Duke, Virginia Polytechnic and State Analytical Instruments, Tewksbury, Massachusetts
University, Blacksburg, Virginia (in-situ tests for (X-ray fluorescence)
structural health) Glenn M. Light, Southwest Research Institute, San
Charles W. Eick, Horizon NDT Services, Cabot, Antonio, Texas (in-situ tests for structural health)
Arizona (penetrant, magnetic particle) Eric A. Lindgren, Wright-Patterson Air Force Base,
Barry A. Fetzer, The Boeing Company, Renton, Dayton, Ohio (ultrasonic)
Washington (ultrasonic) Chester Lo, Iowa State University, Ames, Iowa
Michael D. Fogarty, The Boeing Company, Seattle, (barkhausen)
Washington (in-situ tests for structural health) Paul J. Lomax, Fischer Technology, Windsor,
David S. Forsyth, Texas Research Institute, Austin, Connecticut (beta backscatter)
Texas (reliability) E.I. Madaras, National Aeronautics and Space
Joseph J. Gabris, The Boeing Company, Saint Louis, Administration Langley Research Center, Hampton,
Misssouri (reliability) Virginia (in-situ NDT for structural health)

Gary E. Georgeson, The Boeing Company, Seattle, Xavier [Link], University Laval, Quebec,
Washington (ultrasonic, in-situ tests for structural Quebec, Canada (thermographic)
health) Kane M. Mordaunt, The Boeing Company, Seattle,
Valery F. Godinez-Azcuaga, Shaw Pipeline Services, Washington (leak)
Houston, Texas (acoustic emission) William P. Motzer, The Boeing Company, Seattle,
Neil J. Goldfine, Jentek Sensors, Waltham, Washington (ultrasonic)
Massachusetts (in-situ tests for structural health)

v
 John W. Newman, Laser Technology, Norristown, Robert E. Stevens, United Airlines, San Francisco,
Pennsylvania (shearographic) California (visual)

David A. Raulerson, Pratt and Whitney, West Palm Jeffrey G. Thompson, Boeing, Kent, Washington
Beach, Florida (electromagnetic) (electromagnetic, ultrasonic)

Alan J. Rein, Agilent Technologies, Danbury, Jeffrey A. Umbach, Pratt and Whitney, Palm Beach
Connecticut (fourier transform infrared) Gardens, Florida (X-ray diffraction)

Dennis P. Roach, Sandia National Laboratories, James L. Walker II, National Aeronautics and Space
Albuquerque, New Mexico (bond) Administration, Huntsville, Alabama (acoustic
emission)
Gregory C. Sayler, Presto Casting Company,
Phoenix, Arizona (visual) Andrew P. Washabaugh, Jentek Sensors, Chula
Vista, California (electromagnetic, in-situ NDT for
Karl F. Schmidt, Jr., Evisive, Baton Rouge, structural health, dielectrometry)
Louisiana (microwave)
Gregory J. Weaver, Vibrant NDT, Albuquerque, New
John A. Seelenbinder, Agilent Technologies, Mexico (resonance)
Danbury, Connecticut (fourier transform infrared)
Gary L. Workman, University of Alabama,
Steven M. Shepard, Thermal Wave Imaging, Huntsville, Alabama (ultrasonic)
Ferndale, Michigan (thermographic)
Steven M. Younker, The Boeing Company, Seattle,
Surendra Singh, Honeywell Aerospace, Phoenix, Washington (magnetic particle)
Arizona (resonance)

Flynn Spears, Laser Technology, Seattle,

Washington (electromagnetic)

vi AEROSPACE NONDESTRUCTIVE TESTING

Contents

Chapter 1. Introduction to Aerospace Chapter 4. Cross Reference Tables

Nondestructive Testing for Test Method Selection
1. Nondestructive Testing ..........................1.2 1. Introduction ............................................ 4.2
2. Purpose of Aerospace Handbook .......... 1.3 2. Materials, Structure, and Processes
3. Development ....................................... 1.5 Tables ................................................. 4.3

References .............................................. 1.7 3. Method Tables ........................................4.10

References ................................................. 4.20
Chapter 2. Aerospace Structures
1. Introduction........................................ 2.2 Chapter 5. Visual Testing
2. Types of Vehicles and Missions ............ 2.3 1. Introduction ............................................ 5.2

3. Propulsion Systems............................. 2.6 2. Optical Principles ..................................... 5.3

4. Importance of Nondestructive Testing ... 2.8 3. Optical Techniques ..................................... 5.8

5. Structural Failure Modes ......................2.10 4. Applications .......................................... 5.16

6. Evolution of Structural Design and References ................................................. 5.24

Fatigue Considerations ..................... 2.13
7. Static Life, Infinite Life, and Safe Life .... 2.16 Chapter 6. Liquid
8. Damage Tolerant Design.................... 2.18 1. Penetrant Testing ..................................... 6.1

9. Structural Design and Damage Tolerance 2. Introduction ............................................ 6.2

for Composite Materials ........................2.24 3. Penetrant Principles .................................... 6.3
10. Bonded Assemblies .......................... 2.32 4. Penetrant Techniques ...................................... 6.5
11. Inservice Nondestructive Testing ....... 2.36 5. Safety .................................................. 6.9
References ............................................ 2.37 6. Standards and Specifications .....................6.10
7. Applications ........................................... 6.11
Chapter 3. Reliability References ................................................. 6.16
1. Reliability of Nondestructive Testing ......3.2
2. Probability of Detection and Other Chapter 7. Magnetic Particle Testing
Performance Measures in Practice ........ 3.5 1. Magnetic Particle Testing Principles ............. 7.2
3. Mistakes in Estimation of Nondestructive 2. Magnetic Particle Testing Standards
Test Reliability .................................... 3.7 and Specifications ...................................7.10
4. Human Factors and Nondestructive 3. Magnetic Particle Testing Applications ....... 7.13
Testing .................................................... 3.8
References ................................................. 7.17
5. Model Assisted Probability of Detection 3.11
References ............................................ 3.12

vii
 Chapter 8. Electromagnetic Testing 5. Thermographic Test Standards ......... 11.16

1. Principles of Electromagnetic Testing .... 8.2 6. Application Examples ...................... 11.17

2. Eddy Current Coil Techniques ............. 8.4 References .......................................... 11.21

3. Nonconventional Electromagnetic
Methods ............................................... 8.18 Chapter 12. Shearographic and
4. Relevant Standards and Specifications... 8.25 Holographic Testing
5. Aerospace Applications of Eddy 1. Introduction ..................................... 12.2
Current Testing................................. 8.27 2. Shearography Operation .................. 12.10
References ............................................ 8.48 3. Relevant Standards .........................12.16
4. Applications ....................................12.19
Chapter 9. Ultrasonic Testing 5. Laser and Acoustic Excitation Safety ..12.29
1. Principles of Ultrasonic Testing ............. 9.2 References ..........................................12.32
2. Basic Ultrasonic Techniques .............. 9.14
3. Specialized or Emerging Techniques .... 9.21 Chapter 13. Acoustic Emission Testing
4. Relevant Standards and Specifications ... 9.28 1. Acoustic Emission Principles............. 13.2
5. Application Examples........................ 9.30 2. Acoustic EmissionTest Techniques .... 13.7
References ............................................ 9.37 3. Acoustic Emission Standards
and Specifications ............................ 13.9
Chapter 10. Radiologic Testing 4. Acoustic Emission Test Applications ... 13.11
1. Radiologic Test Principles .................. 10.2 References ..........................................13.17
2. Radiologic Test Techniques ...............10.10
3. Factors Affecting Image Quality ........ 10.19 Chapter 14. Bond Testing
4. Sensitivity Measurement ................. 10.29 1. Introduction ..................................... 14.2
5. Radiologic Test Interpretation ...........10.31 2. Bond Testing Methods ...................... 14.5

6. Specialized and Emerging References .......................................... 14.10

Radiation Techniques...................... 10.33
7. Radiologic Testing Standards ........... 10.36 Chapter 15. In-Situ Tests for
8. Applications of Radiologic Testing. .... 10.37 Structural Health Monitoring
References .......................................... 10.43 1. Principles ......................................... 15.2
2. Methods ........................................... 15.4
Chapter 11. Thermographic Testing 3. Applications ......................................15.5
1. Thermographic Principles .................. 11.2 References ..........................................15.14
2. Thermographic Instrumentation ........ 11.7
3. Interpretation and Analysis of Chapter 16. Leak Testing
Thermographic Results .................... 11.11 1. Introduction ..................................... 16.2
4. Emerging Thermographic Techniques ... 11.14 2. Leak Test Techniques and Applications ... 16.3
References ..........................................16.12

viii AEROSPACE NONDESTRUCTIVE TESTING

Chapter 17. Other Methods 3. Magnetic Particle Testing .................. 18.2

1. Barkhausen Effect ............................ 17.2 4. Electromagnetic Testing .................... 18.3

2. Microwave Testing ............................. 17.5 5. Ultrasonic Testing............................. 18.3

3. Dielectrometry: Capacitance Sensing ... 17.12 6. Radiologic Testing ............................. 18.4

4. X-Ray Diffraction ............................ 17.15 7. Thermographic Testing ..................... 18.5

5. X-Ray Fluorescence Spectroscopy .... 17.18 8. X-Ray Diffraction .............................. 18.6

6. Fourier Transform Infrared Testing ... 17.20

Chapter 19. Glossary
7. Beta Backscatter ............................. 17.22
1. Introduction ..................................... 19.2
8. Resonance and Vibration Testing ...... 17.24
2. Terms ............................................... 19.3
References .......................................... 17.27
References.......................................... 19.22

Chapter 18. Formulas

Index...................................................... 20.1
1. Visual Testing ................................... 18.2
2. Penetrant Testing .............................. 18.2
Figure Sources .............................. 20.12

ix
 X
1
CHAPTER

Introduction to
Aerospace
Nondestruct
ive Testing

Contents
Part 1. Nondestructive
Testing, 1.2
Part2. Purpose of Aerospace
Handbook, 1.3
Part 3. Development, 1.5
References, 1.7

Contributors
Richard H. Bossi
Lisa Brasche
Matthew J. Golis 1.1
PART1

Nondestructive testing is a materials science Nondestructive testing asks, ―Is there something
concerned with many aspects of quality and wrong with this material?‖ In contrast, performance
serviceability of materials and structures. The science and proof tests ask, ―Does this component work?‖ It
of nondestructive testing incorporates all the is not considered nondestructive testing when an
technology for process monitoring and for detection inspector checks a circuit by running electric current
and measurement of significant properties, including through it. Hydrostatic pressure testing is a form of
discontinuities, in items ranging from research test proof testing that sometimes destroys the test object.
objects to finished hardware and products in service. A gray area in the definition of nondestructive
Nondestructive testing examines materials and testing is the phrase future usefulness. Some material
structures without impairment of serviceability and investigations involve taking a sample of the test
reveals hidden properties and discontinuities. object for a test that is inherently destructive. A
Nondestructive testing is becoming increasingly noncritical part of a pressure vessel may be scraped
vital in the effective conduct of research, or shaved to get a sample for electron microscopy,
development, design, and manufacturing programs. for example. Although future usefulness of the vessel
Only with appropriate nondestructive testing can the is not impaired by the loss of material, the procedure
benefits of advanced materials science be fully is inherently destructive and the shaving itself — in
realized. The information required for appreciating one sense, the true test object — has been removed
the broad scope of nondestructive testing is available from service permanently.
in many publications and reports. The idea of future usefulness is relevant to the
quality control practice of sampling. Sampling (that
is, less than 100 percent testing to draw inferences
Definition about the unsampled lots) is nondestructive testing if
the tested sample is returned to service. If steel bolts
Nondestructive testing (NDT) has been defined as are tested to verify their alloy and are then returned
those methods used to test a part or material or to service, then the test is nondestructive. In contrast,
system without impairing its future usefulness (ASNT even if spectroscopy in the chemical testing of many
2012). The term is generally applied to nonmedical fluids is inherently nondestructive, the testing is
investigations of material integrity. destructive if the samples are discarded after testing.
Nondestructive testing is used to investigate Nondestructive testing is not confined to crack
specifically the material integrity or properties of a detection. Other anomalies include porosity, wall
test object. A number of other technologies — for thinning from corrosion, and many sorts of
instance, radio astronomy, voltage and amperage disbonds. Nondestructive material characterization is
measurement, and rheometry (flow measurement) — a field concerned with properties including material
are nondestructive but are not used specifically to identification and microstructural characteristics —
evaluate material properties. Radar and sonar are such as resin curing, case hardening, and stress —
classified as nondestructive testing when used to that directly influence the service life of the test
inspect dams, for instance, but not when used to object.
chart a river bottom.
 dbook PART2

This ASNT Industry Handbook on aerospace in 1908 and the tragic death of Thomas Selfridge.
nondestructive testing is intended to provide the Figure 1 shows the wreckage.
practitioner of nondestructive testing in the From the very beginnings of flight, lightweight
aerospace industry a useful reference to the variety and durable materials have been the key to
of principles, techniques, and methods that may be successful aerospace structures. Validation of the
needed to inspect aerospace components and quality of the materials and assembly is of utmost
structures. The goal is to include the fundamental importance. Nondestructive testing plays a
physical principles and tables of constants used in paramount role in ensuring that high quality.
aerospace nondestructive testing. The content of the Inspection criteria for aerospace structures tend to be
handbook is intended to cover applications and the most critical of the industries that employ
techniques of interest to NDT Level II and III nondestructive test methods. The restrictions on
technical staff, nondestructive evaluation research weight to create effective and economical structures
engineers, manufacturing engineers, aerospace limit the ability to have large safety margins and
structures engineers, and management. The redundancy. Safety margins are often substantial for
handbook provides a number of cross reference critical components in other industries, such as
tables to associate applicable nondestructive test factors of 4 or higher. But for aerospace applications,
methods and techniques to material evaluations and the factors are much lower, typically from 1.15 to
measurements of possible interest. The objective is to 1.5. The cost of structural weight to deliver high
provide the reader with the baseline approach most safety margins would be prohibitive to the mission,
commonly used for the evaluation of the common especially in space applications. Thus, aerospace
aerospace materials and structures, and also to nondestructive testing is concerned with a small
include alternative possibilities for those situations discontinuity size that will not grow to critical size
where the baseline approach is insufficient or not within the operational cycle. Additionally, aircraft
applicable. While the details of advanced methods structure and engine components require that
are not expanded upon in this handbook, the intent inspections be performed during their service life to
is to provide sufficient insight for the reader to be ensure that discontinuities are detected while below
able to seek additional information from other critical size during their lifetimes. Nondestructive
sources. testing developments play a critical role in aerospace
ASNT provides the inspection and testing economics because the minimum detectable
community with nondestructive testing method discontinuity size sets the inspection interval. Poor
handbooks. Each method handbook covers the discontinuity detectability increases the frequency of
technical information in great detail for engineers inspection with added cost while high sensitivity and
and researchers that are working in that discipline. reliable nondestructive testing methods can extend
This aerospace industry handbook draws together the
basic method information as it applies to the
materials and structures used in aerospace. This Figure 1. Fatal crash of the Wright Flyer, September 17, 1908.
handbook addresses the material problems of interest
in the aerospace industry and identifies the
applicability of the different nondestructive testing
methods to those problems. The information is aimed
at providing approaches and solutions to inspection
issues in the aerospace industry. The method
chapters in this handbook summarize the basic
physics of the method, the variations of its
applicability, and examples of its usage.

Importance of Inspection to
Aerospace Industry
The aerospace industry is particularly sensitive to
the quality condition of components because of the
severity of the consequences of failure during service
and the costs of the equipment. It was a cracked
propeller that led to the first crash of a Wright flyer

1.3
 Table 1. Structure issues for nondestructive testing.

Issue Comments for Nondestructive Testing

Thin structure Small discontinuity size criteria result in high resolution/sensitivity requirements for
nondestructive testing capability.
Reinforcement Complex geometries have limited access and boundary edges. Multilayered materials,
multiple materials, and nonuniformity are challenges to nondestructive test schemes.
Shape/contour Aerodynamic, nonparallel and nonflat surfaces require contour following capability.

Temperature High cost, heat resistant materials can challenge inspection systems for penetration and
sensitivity to detail.
Coatings Paint and coatings can have special requirements for detection sensitivity and thin layer
evaluation.
Bonds Adhesive bond quality validation can be very challenging depending on the design and
bonding process.
Propulsion systems Jetenginecomponents and propellant rocketmotorsystemsrequirehighreliability verified
by sensitive nondestructive testing.
Thick components Casting and forging geometries can have thick sections complicated by microstructural
noise.

inspection intervals, thereby increasing the strength and stiffness as a function of position. This
operational efficiency. nonuniformity greatly complicates nondestructive
Aerospace structures such as aircraft, rotorcraft, testing. Efficient aerodynamics requires special
spacecraft, rockets, and missiles also have some other contours and shapes that challenge nondestructive
characteristics that can make them particularly test equipment to maintain proper inspection
challenging for nondestructive test methods. Jet orientations. Aerospace structures have high and low
engines and rocket motors have additional temperature characteristics that require specific
challenges. Table 1 lists some of these structural and materials with their subsequent nondestructive test
materials issues. First, to maintain light weight, the limitations. Coatings and bonds in aerospace systems
structures must be thin but stiffened for strength. present their own quality issues that affect
Discontinuity size limitation in thin structures must nondestructive test operations. Finally, propulsion
be small, demanding fine resolution and sensitivity systems such as jet engines or rocket motors must
for nondestructive test systems. Stiffening of the have exceedingly high reliability, requiring
structure results in complex geometries that can sophisticated and sensitive nondestructive test
challenge nondestructive test methods for sensitivity methods. The success of nondestructive testing is
near edges or radius geometries. Internally, aerospace reflected in the reduction of the number of engines
structures are designed with frequent changes in that are used for modern commercial aircraft.
thickness and layers of materials to create optimized
 pment
PART3

Early Years systems, whereas ferromagnetic parts could be tested

with the magnetic particle systems of the time.
Even as early as the Wright brothers‘ efforts at Although the equipment was bulky and
achieving powered flight, it was recognized that the somewhat crude by twenty-first century standards,
materials in vehicles heavier than air needed to be of it represented a major step in ensuring the integrity
the utmost quality if the venture was to be a success. of materials being used in both aircraft structures
Early concerns were therefore focused on the and engines. It became clear that the assurance of
integrity of aircraft component materials. In most discontinuity free materials significantly enhanced
cases, a close visual check was all that could be done the capability of designers to develop structures and
to ensure that no weak spots existed in materials systems with ever declining levels of conservatism
being incorporated into the early experimental and thus leading to higher performance aircraft and
aircraft. With the advent of radiography, it became reduced bulk materials costs (Hagemaier 1985).
possible to see both surface blemishes and internal
conditions that could lead to unexpected and
premature failure. Particularly in cast metal parts,
internal discontinuities became visible and Organization and
radiography became an integral part of screening for
discontinuity-free materials. Early records show that
Standardization
In 1959, the original version of the
the military, as early as 1919, within the Material
Nondestructive Testing Handbook edited by
Section based at McCook Field in Dayton, Ohio, was
Robert C. McMaster was published by Ronald Press
charged with the task of testing fabricated parts and
(ASNT 1959). It was created using contributions
to make routine inspection tests for their
from the world‘s experts in the various
procurement section. Of the three dozen or so
nondestructive test methods and rapidly became the
nondestructive test methods used or investigated in
resource that the entire industry relied on for the
the twenty-first century, nearly all have appeared
theory and application of inspection practices. It
since the 1920s. And most have appeared since the
was the first time that the disparate collections of
1940s (Straw1984).
techniques used in nondestructive testing were
presented in a cogent and practical format. It
explained the philosophy of nondestructive testing:
World War II how it supported the goals of controlling the quality
As with many industrial needs, the period of materials being used in industry, especially in
between 1940 and 1945 resulted in the discovery aerospace applications because of their intolerance
and application of numerous technologies that of failure and tighter safety margins. This aerospace
served the needs of the war effort. For aircraft specific handbook builds on the earlier work by
structures, the materials of most concern were expanding on the operational nondestructive testing
aluminum, magnesium, and stainless steel. During details used to ensure the integrity of materials
this period, the need for improved inspection of being used in today‘s contemporary air vehicles.
these nonmagnetic structural materials resulted in As with all evolving disciplines, there comes a
the development of ultrasonic systems to inspect time when the exploratory stage of invention needs
aircraft component materials using both contact and stabilization, often through the mechanisms of
immersed transducer scanning systems. This standardization and regulations. Throughout the
permitted the detection of ever finer discontinuities 1950s and into the 1960s, nondestructive testing
through improved resolution in comparison to that came into its own when its technologies were
achieved with contact techniques on heavier steel incorporated into national and international
structures and components. standards for the many industrial sectors. In the
The concepts of eddy currents were expanded case of the aerospace industry, nondestructive
from simply using encircling coils for screening testing was divided into the military set of directives
discontinuities in drawn wire and tubes and sorting (the MIL-STD series) and those formulated by the
small components for size and alloy consistency to commercial aircraft industry (supported by the
the application of probe based systems for directives of the FAA). It was a time when
determining material alloy content, detecting professional groups such the American Society for
inservice discontinuities, and gaging the thicknesses Testing and Materials (ASTM, later to become ASTM
of thin metals and nonmetallic coatings on International), the Society of Automotive Engineers
conductive substrates. Surface discontinuities were (SAE, later to become SAE International), the
being found with fluorescent liquid penetrant American Society for Metals (ASM, later to become
the American Society for Materials and then ASM
International), the American Society of Mechanical Explosion in Equipment
Engineers (ASME), the American Society for Welding
(AWS), the American Petroleum Institute (API), and Capabilities and Modern
others, through the use of consensus committees,
formulated their individual approaches to how
Materials
nondestructive test methods were to be applied in With the discovery of the transistor and other
their respective systems and products. emerging avenues for consolidating enormous
These standards and regulations were the results operational capabilities into a small space, the bulky
of intensive laboratory and field investigations and instruments of yesteryear became an endangered
demonstrations of the capabilities of the species. With this vast increase in performance and
nondestructive test methods. The stage was set for computational possibilities, the equipment used in
the wide ranging standards documents that now nondestructive testing has continued to shrink in
guide the ways that nondestructive tests are to be size and has absorbed many of the operational
performed. The technical issues of nondestructive details associated with calibration, standardization,
test practices and acceptance criteria rested mostly data gathering, and results analysis.
with the military and ASTM standards. In general, Concurrently, more sophisticated and complex
the various industry sectors adapted these basic materials are being developed to serve the needs of
performance documents in accordance with their the aerospace industry. The continual development
specific tolerance for risk of failure (ANSI 2010). of new materials emphasizes the point of having to
determine how these materials (including advanced
composites, ceramics, and thermal or
electromagnetic shielding systems) are to be
Fracture Mechanics and examined for their mission critical characteristics.
Quantitative Nondestructive And with these new materials comes the need for
nondestructive testing standards development to
Evaluation ensure that those critical characteristics are met.
While stress-strain curves and safety factors have As a better understanding of material and
played a key role in the design of aerospace structural failure modes has evolved in recent years,
structures, advances in fracture mechanics concepts the role of nondestructive testing for inservice
made in the 1960s placed a new emphasis on applications has become integral to the ways that
nondestructive testing, requiring approaches that fleets of aircraft and critical components are
went beyond simple detection to include crack managed. The effectiveness and reliability of
sizing. With the introduction of fracture mechanics nondestructive testing processes remain of utmost
and damage tolerance concepts into the design importance to the aerospace community, both now
process came the need to quantify inspection and in the future.
capability.
Even with the somewhat controlled studies into
the limits of the nondestructive test techniques in
use, it was suspected that the broad array of
inspection techniques being used throughout the
many classes of inspection tasks, particularly those
done manually, might vary between diverse
inspection situations (locations, personnel,
equipment, and instructions). A landmark study was
conducted in the late 1960s and launched the use of
a statistical inferences approach to determining the
reliability of nondestructive tests (Haviland 1973).
The concept of probability of detection (for
nondestructive testing purposes) was born.
ANSI. 2010. NSSN: A National Resource for Global Standards. Website. Haviland, G.P., and C. Tiffany. 1973. AIAA Paper 73-18, ―The USAF
New York, NY: American National Standards Institute. Aircraft Structural Integrity Program (ASIP). Proceedings of the
ASNT. 1959. Nondestructive Testing Handbook, first edition, 2 vols. New American Institute of Aeronautics and Astronautics 9th Annual
York, NY: Ronald Press, reprinted Columbus, OH: American Society Meeting and Technical Display [Washington, DC, January 1973].
for Nondestructive Testing. Reston, VA: American Institute of Aeronautics and Astronautics.
ASNT. 2012. Nondestructive Testing Handbook, third edition; Vol. 10, Straw, R. 1984. ―Voices in the Air — The Early Days of Aircraft NDT.‖
Nondestructive Testing Overview. Columbus, OH: American Society Materials Evaluation 42(2): 152-160.
for Nondestructive Testing.
Hagemaier, D.J. 1985. ―Aerospace Radiography — The Last Three
Decades.‖ Materials Evaluation 43(10): 1262-1264+.

1.7
 2
CHAPTER

Aerospa
ce
Structure
s
Contents
Part 1. Introduction, 2.2
Part 2. Types of Vehicles and
Missions, 2.3
Part 3. Propulsion Systems, 2.6
Part 4. Importance of
Nondestructive Testing, 2.8
Part 5. Structural Failure
Modes, 2.10
Part 6. Evolution of Structural
Design and Fatigue
Considerations, 2.13
Part 7. Static Life, Infinite Life,
and Safe Life, 2.16
[Link] Tolerant
Design, 2.18
Part 9. Structural Design and
Damage Tolerance for Composite
Materials, 2.24
Part 10. Bonded Assemblies, 2.32
Part 11. Inservice Nondestructive
Testing, 2.36
 Introduction
PART1

Aerospace vehicles include air vehicles, control, communication, navigation, weather,

spacecraft, rockets, and missiles. Ensuring the weapons, and anticollision systems. Aerospace
integrity ofthe vehicle structure and associated systems maybe broken into many categories, such
avionics and systems is essential when considering aspropulsion, environmental control, hydraulics,
manufacturing costs, the value ofitemsthat are and armament.
transported, and human safety. Some typical components of aircraftstructure are
Avionics are generally the aviation electronics, illustrated in Figure 1.
computers, and software that govern guidance and

Figure 1. Basic aircraft components (Boeing 737).

Vertical stabilizer tip

Rudder
Vertical stabilizer trailing edge

Vertical stabilizer leading edge Tailcone

Dorsal fin Elevator

Horizontal stabilizer
trailing edge

Horizontal stabilizer tip

Horizontal stabilizer leading edge

Wing-to-body fairings

Main flaps
Aft flap Aft flap
Door (typical)
Spoilers
Aileron
Nose radome Wing Fixed trailing edge

Krueger flaps Wing tip

Slats Wing fixed
Nose landing leading edge
gear (NLG)
Airstair Engine strut fairing
NLG doors
Engine strut
Main landing gear (MLG)
Nozzle and plug

MLG doors Fan duct cowl and thrust reverser

Fan cowl
Inlet cowl
Legend
= primary structure
= secondary structure

2.2 AEROSPACE NONDESTRUCTIVE TESTING

 Types of Vehicles and Missions PART2

The structure is the aerospacevehicle‘s frame that Civil aviation may be broken into two categories:
carries the loads duringinitiation of flight, during (1) commercial scheduled air transport, which
execution of the mission, and in some cases during includes all passenger and cargo flights operating on
recovery of the vehicle. Aerospace structure is regularly scheduled routes; and (2) general aviation,
commonly divided into primary structure, the failure which includes all other civil flights, private or
of which will cause loss of life or loss of the vehicle, commercial. General aviation is the category for
and secondary structure. The integrity of aerospace recreational civilian aircraft as well as most
structures and systems is determined by using some helicopters and gliders. All of these craft are
level of nondestructive testing (NDT) during typically smaller than commercial aircraft, have
manufacturing and throughout the life cycle of the shorter ranges, fly at lower altitudes, with lower
vehicle. speeds, and require shorter runways for takeoff and
Aerospace vehicles are commonly segregated into landing. General aviation aircraft account for around
four categories: aircraft, spacecraft, missiles, and 80 percent of all aircraft in the United States.
rockets. The intended use or mission of the air Title 14 in the United States Code of Federal
vehicle provides the basis for system and structure Regulations establishes Federal Requirements for
design. Aeronautics and Space, which are administered by
the Federal Aviation Administration in the
Department of Transportation. The manufacturer of
Aircraft the aircraft is the design authority and seeks air
worthiness certification for flight from the Federal
Most aircraft fly at altitudes less than 14 km Aviation Administration before makingcommercial
(45000 ft) and return to Earth for subsequent flights. deliveries.
From an aerodynamics perspective, aircraft fall into Military aircraft fall into four general categories:
two categories — fixed wing or rotor wing. Typical (1) fighter/attack, (2) transport, (3) bombers, and
missions for fixed and rotor wing aircraft are shown (4) specialized mission (including tanker,
in Table 1. From a usage perspective, aircraft can be reconnaissance, and electronic warfare) that dictate
placed in one of two classifications: military and the important characteristics of military aircraft
civilian. design and performance. Aircraft are predominately
manned vehicles; however, there is an emerging
class of unmanned air vehicles that are not rated for
Table 1. Fixed wing and rotor wing aircraft passenger safety, which is a significant design
missions. consideration. In the United States, all aircraft
intended for use by the Department of Defense (DOD)
must be approved and funded by the legislative
Fixed Wing Rotor Wing branch of the United States federal government. The
design, airworthiness certification for flight,
schedule, and fiscal management aspects are
Civilian Civilian
administered by the specific branch of the
Commercial transport Transport Department of Defense with program progress
Commercial freight Law enforcement reported to the legislative and executive branches of
Agriculture Medical the United States government.
Recreation Recreation In the United States, Federal Law has established
Aerobatics News/traffic the International Traffic in Arms Regulations.
Essentially all aircraft, the equipment, and the
Construction technology to develop, manufacture, support, and
Firefighting maintain theaircraft must have United States
Military Military government approval before export.
Fighter Fighter
Attack Attack Aircraft Structure
Bomber Air support Aircraft design and support is normally divided
Reconnaissance Reconnaissance into three commoncategories: structure, systems,
Electronic warfare Troop transport and avionics. Although numerous tests that are
Troop transport Supply nondestructive in nature are applied to avionics
Equipment transport Medivac equipment, this handbook will be predominantly
focused on the traditional nondestructive tests that
Medivac are performed on structure and systems. Some
typical components of aircraft structure are
illustrated in Figure 1. 2.3
 Aircraft Coordinate Systems up and is called thewater line (WL). The units ofthe
coordinate system aretypically in inches withthe
During design, the manufacturer of the aircraft origin at an arbitrary point on each aircraft: there is
establishes a coordinate system forthe airframe no convention for the origin. It may be at the nose
structure. Because an airframe is three-dimensional, ofthe aircraft, thecenterofgravity, onthe ground,
a cartesian coordinate system (also known as a or at a point in front of the nose. The value of the
rectangular coordinate system) is used (Figure 2). coordinate system is to locate and label structural
The aerospaceindustrydoes not havea convention components. For example, a fuselage bulkhead may
for labeling the XYZ axes or for the origin for the be identified as FS Y500.500 (meaning fuselage
axes. Although Figure 2 shows the axis in the station at 500.500 inches from the 0,0,0 reference
forward/aft direction to be the Y axis and the axis in point along the longitudinal axis). These coordinates
the inboard/outboard direction as the X axis, it is are not only important in the initial design,
not unusual to see the X and Y axes switched fabrication, and assembly of the structure but are
depending on the convention established by the also used in maintenance manuals and inspection
aircraft designer. bulletins foraircraftservice.
The axis that points aft is typically called the
fuselage station (FS). The axis that points outboard
from the fuselage is typically called the butt line (BL)
but is sometimes referred to as the wing station Table 2. Spacecraft missions.
(WS). The Z axis is almost always the axis pointing
Manned Spaceflight

Figure 2. Aircraft structural coordinate system (Boeing F-18). Earth orbiting vehicles and shuttles
Space station
Lunar and planetaryspace missions
Z
Tourism
Unmanned Spaceflight
Yaw Earth Orbiting Space Exploration
Vertical axis

Communications Fly-by observers

Weather Orbiters
Navigation Landers
Roll Reconnaissance Rovers
Pitch Intelligence Sensors
Y Science
Astronomical observation
X

Figure 3. Components of the International Space Station (ISS).

Docking compartment(DC) 1
Zarya control
Zvezda service module module
S1truss Port photovoltaic arrays
Pressurized mating
adaptor (PMA) 1 segment SO truss P1 truss segment
Research module (RM) segment
S3/4 truss Mobile P6trusssegments
segments servicing
system
Multipurpose laboratory module (MLM)
MLM outfitting
S6 trusssegment S5 truss
segment
P5 truss segment
Canadarm 2 P3/4 truss segments
Special purpose dexterous manipulator (SPDM)
Kibo ELM-PS
Airlock
Starboard photovoltaic arrays
Z1 truss Node 1 Kibo remote manipulator system (RMS)
Mobile remote servicer base system segment
Legend (MBS), Mobile transporter (MT) and exposed facility
Kibo
Node3
Elements in orbit in 2007 Destiny PMA 2
PMA 3 Node 2
Elements from United States Cupola
Elements from Russia Columbus

2.4 AEROSPACE NONDESTRUCTIVE TESTING

Spacecraft Figure 4. Westar® satellite, example of a satellite system.
Spacecraft normally fly in low Earth orbit, which
is commonly accepted to be 160 to 2000 km (100 to
1240 mi) above the Earth‘s surface. Some spacecraft
are designed to leave the Earth‘s atmosphere and Antenna
travel through the galaxy and universe. Spacecraft Antenna reflector
can be classified as manned, Earth orbiters, and
space exploration (planetary/lunar/galactic/universe).
Table 2 lists spacecraft missions. Figures 3 and 4
illustrate the International Space Station (ISS) and a
satellite. Space structures are designed to take
significant loads at launch and to last for years Antenna feeds
without service (Sarafin 1995).
Despun repeater shelf

Missiles and Rockets

Missiles are a single-use item, as are most rockets. Thermal radiator
They are munitions that may use ballistic trajectory
or may be guided by sensors. Some missiles travel at Control thruster Battery pack
very low altitudes, whereas others may travel into
low Earth orbit before reaching their target. Desired
range, launch medium, and type of damage to be Propellant tanks
incurred distinguish classes of missiles. Rockets are Solar cell arrays
generally used to assist spacecraft in reaching low
Earth orbit and so serve as both an air vehicle and a
propulsion system (Fleeman 2001). Figure 5 shows
Kick motor
the relative size of heavy lift rocket system launch
vehicles comparing the National Aeronautics and
Space Administration Ares ICrew Launch Vehicle
(CLV) and the Ares V heavy lift Cargo Launch
Vehicle (CaLV) systems to the Space Shuttle and
Saturn V systems.

Figure 5. National Aeronautics and Space Administration diagram comparing rocket launch vehicles (Cook 2011).

122 m (400 ft)

Ares V
Saturn V

Ares I

91 m (300 ft)

Space Shuttle
61 m (200 ft)

30 m (100 ft)

AEROSPACE STRUCTURES 2.5

 Propulsion Systems
PART3

During the initial design phase of any aerospace reciprocating pistons, which turns a crankshaft that
vehicle, the propulsion system is the key to rotates the propeller. Most commercial and military
achieving themission requirements for payload, aircraft utilize gas turbine engines because of their
range, speed, and altitude. Propulsion systems, ability to achieve higher levels of thrust by using jet
classified as an engine ora rocket motor, account technology in the form of a turbo jet, turbofan, or
for nearly all means of vehicle propulsion. In special turboprop. The thrust of the turbojet results from
instances, there are alternative means of propulsion the action of a fluid jet leaving the engine; hence,
such as wind for gliders, hot air for balloons, and the name ―jet engine‖ is often applied to these
low molecular weight gases for balloons and blimps. engines. The turbofan, turboprop, and turboshaft
engines are adaptations of the turbojet to supply
thrust orpower through fans, propellers, and shafts.
Civilian and Military Aircraft These types of engines involve rotating machinery
with bearings, cases, combustion chambers, and
Propulsion numerous internal blades. Figure 6 shows a
The vast majority of aircraft uses either a gas turboprop engine cutaway and Figure 7 shows a
turbine engine or a piston engine with a propeller. turbofan engine cutaway. The various components
Most general aviation airplanes are piston type ofthe engines require significant nondestructive
internal combustion engines. In a piston powered testing at manufacture and in service to maintain
aircraft, internal combustion drives the reliable operation. Engine materials include high

Figure 6. Parts of turboprop engine. Figure 7. Parts of a turbofan engine.

Hotgases
Nozzle

Nozzle
Mixer
Turbine

Combustor

Combustion chamber
Turbine
Fuel injector

Compressor

Compressor

Inlet GearBox

Fan
Prop

2.6 AEROSPACE NONDESTRUCTIVE TESTING

temperature metal alloys and ceramics. Some
composites are used in lower temperature regions of Figure 8. X-ray computed tomography slice of
the engines, such as large fan blades. Propulsion solid rocket motor.
systems are also categorized by the criticality of their
potential failure. Primary rotating parts are deemed
more critical because ofthe potential damage that
can occur should they fail during operation. For this
reason, rotating parts such as engine disks have Feature
more strenuous inspection requirements than the Propellant
stationary components inanengine.

Missile and Rocket Propulsion Hole

Air breathing missiles typically use variations of Liner
jet engines sized for the missile system. Rocket
engines use the expulsion of gas at high velocity Case
from a nozzle for propulsion. The gas is generated at
high temperature and pressure by the combustion of
either solid or liquid propellants (Mattingly 2006).
Solid propellant is usually a combination of an
oxidizer material mixed with a solid fuel. Once
ignited, the propellant burns in accordance with the Figure 9. BoeingTM RS-68 liquid rocket engine
shape of the propellant surface until it is totally (Wood 2002).
consumed. Nondestructive testing is performed to
validate the condition of the solid propellant
assembly. Anomalies in the propellant grain or liner
can lead to catastrophic failure of the motor.
Figure 8 shows an X-ray computed tomography scan
of a solid rocket motor. The computed tomography
image cross section through the body of the motor
shows the interior configuration commonly used in
solid propellant systems. The liner condition and the
propellant grain inspections are performed at
manufacture and may also be required after
significant storage time.
Liquid motors contain the chemical oxidizer and
fuel in separate tanks. The engine mixes the two
chemicals, burning the fuel to create the propulsion
gas. Liquid fuel engines can be turned on oroff on
control. The engine components often have
complicated geometric shapes. Welds are typically
critical issues in liquid fuel engine components.
Figure 9 shows the Boeing RS68 liquid engine
configuration for the Delta IV rocket that can
produce 3.37 MN (758 000 lbf) with 45 to
1 thrust-to-weight ratio.

AEROSPACE STRUCTURES 2.7

 Importance of Nondestructive Testing
PART4

Nondestructive testing, inspection, evaluation, design process and throughout the life cycle of
and characterization (NDT, NDI, NDE, and NDC) are aerospace vehicles. The use of lower safety factors
all terms that relate to using processes that check for typically requires nondestructive testing to identify
quality and suitability for service withoutaltering smaller discontinuities with higher probabilities of
the product being examined. Nondestructive testing detection to ensure that rogue anomalies are not
and inspection are usually used synonymously for present in the structure.
the examination and measurement process, and
comparisons to specification based accept/reject
criteria. Nondestructive evaluation and Conceptual and Initial Design
characterization are terms that are used for relating
the nondestructive measurements to fitness of the Phases
object for service. In aerospace applications, the The design of an aircraft is an iterative process. It
consequences of failure of a component are often begins with a conceptual phase that defines the
dire in terms of risk to life or cost. Thus, mission and operational parameters. This phase
nondestructive testing, in all its forms, becomes a triggers an initial design phase complemented with
very fundamental part of every aerospace product. trade studies and test programs, including
In the first edition of the NDT Handbook, Robert (1) preliminary weights and loads (static and
McMaster described the importance of fatigue), (2) aerodynamics, (3) propulsion concepts,
nondestructive testing. Table 3 lists his critical (4) proposed materials and construction geometry,
points. All these points are useful to keep in mind (5) integration of new fabrication and assembly
when designing for and implementing processes, (6) exploration of tooling concepts for
nondestructive testing in aerospace products. fabrication and assembly, and (7) performing effects
of anomalies study programs.
This phase may lastseveral years, with the results
Table 3. Importance of nondestructive testing
of the various analytical and laboratory testing
(ASNT 1959). reported to the design team to create a refined
definition of all aspects of the air vehicle. Prime
To ensure product reliability contractors of the vehicle typically partner with
To prevent accidents and save humanlives subcontractors of systems and propulsion systems to
ensure a parallel initial design phase to be integrated
To make a profit for the user (add value) in the overall air vehicle. Nondestructive testing
To ensurecustomersatisfaction and to maintain plays an important role during the initial design
the manufacturer’s good reputation phase, where there is a great deal of interaction with
To aid in better product design structural and design engineers on activities.
To control manufacturing processes 1. Evaluation of various inspection methods to
To lower manufacturing cost provide the optimum inspection for the
material/geometry.
To maintain a uniform quality level 2. Determination of the type, size, and effects of
anomalies that can be found in a material or
geometry. Where inspections cannot be
Because of the nature of aerospace products, the performed, structural engineers will have to
nondestructive test procedures and acceptance assume an anomaly could be present in their
criteria are among the most stringent of all analysis, which can result in heavier structure to
industries. This can be understood bythe safety compensate.
factors that are used in aerospace design. They may 3. Evaluation of specimens for material property
run as low as 1.15 for some military applications tests (before, during, and after testing).
and up to as high as 1.5 to 2.5 for commercial Structural and design engineers use the results of
aircraft. This is compared to safety factors of these studies to determine allowable loads, refine
between 4 and 10 that are commonly applied to design geometries, select fabrication processes, and
other industrial products and construction. Lower develop inservice inspection concepts during the
safety factors allow the design and use of thinner primary design phase. Nondestructive test personnel
and/or lower strength structure, thereby reducing involved in these activities need to be knowledgeable
weight that may reduce propulsion needs and about the variety of nondestructive test techniques
increase range and payload. As such, nondestructive and alternatives that may be applied as well as the
testing engineering plays an important role in the consequences ofvariations intheirapplication.

2.8 AEROSPACE NONDESTRUCTIVE TESTING

 techniques, (2) design of associated tooling for
Primary Design and nondestructive testequipment and to hold objects
Fabrication during nondestructive test operations,
The primary design and fabrication has also been (3) identificationand procurement of nondestructive
known as the full scale development (FSD) or the test equipment, (4) design and fabrication of
engineering manufacturing development (EMD) reference standards, and (5) determination of
phase of an aircraft program. This phase begins with inspection sequence during the manufacturing and
the primary design of the vehicle followed by the assembly process.
manufacture and assembly of the first several air A critical design review (CDR) is normally
vehicles that will be used in the certification process. performed when most ofthe structural design has
During primary design, the basic external applied been established and at the early stages of
loads are established, which allows calculation of fabrication. As fabrication of parts begins, raw
internal element loads. External loads are then material manufacturers perform nondestructive
distributed into the allowable internal element loads testing on raw material and starting stock for key
that will meet strength requirements. Factors of structural components as required by engineering
safety are determined for the primary structure that drawings and models. After raw material is
serves as the major load path. With this knowledge, fabricated into parts and assembled, nondestructive
selection ofmaterial and geometry ofthe structure testing is performed at various stages to verify that
may be defined and the parts sized for the aircraft discontinuities do not exceed theacceptance criteria.
structure. Initial sizing is based uponstaticloads, but Typically, the first few fully assembled aircraft, along
fatigue loading becomes the primary consideration withseveralofthe majorsubassemblies and systems,
in the final design of the structure. are subjected to ground, flight, and functional
Systems designoccurs inparallel withstructure testing as part of flight certification.
design and as the general shape and size of the Ground tests mayinclude adestructive testarticle
structure is refined, so are the systems. Although and tests of small assemblies as well as full scale
systems do not carry primary structural aircraft tests. Destructive tests typically allow coupons to be
loads, system parts are designed and sized based on cut and tested to verify performance and critical
reliable functioning ofthe system forthelifeofthe features in a part. They also provide an opportunity
vehicle. to verify anomalies seen during nondestructive
After parts are sized, initial vehicle weights are testing by excising pieces and performing section
adjusted and power plant requirements for lift, cuts. Ground testing for static loads is normally
thrust, and range are refined. Production drawings completed before initiating a flight test program.
(or models) are then completed. During this period of Static test articles areusually disassembled and
sizing and design, nondestructive testing engineers evaluated with nondestructive testing to determine if
develop process specifications that establish any damage was incurred during statically applied
requirements forthe various nondestructive test loading.
methods, including the acceptance/rejection criteria A flight readiness review is typically performed
for features or anomalies that may be detected by around the time that static and fatigue testing begin.
the selected nondestructive test method. These The review must be completed before any flight
criteria are established bytesting or analysis and operations. Fatigue ground tests and flight testing
coordinated with the structural and materials usually run concurrently and can last from several
engineers. Nondestructive testing engineers work months to several years. Oneofthe key elements of
with the design engineers to identify the flight testing is to verify flight loads and increase the
nondestructive testing process specifications and operation envelope of the aircraft. Additional full
acceptance criteria to be called out on the scale tests may be performed that are related to
engineering drawings and models. mission performance of structure and systems, such
Nondestructive testing engineers also participate as corrosion, drop, flutter, mechanical vibration,
in the evaluation and qualification of subcontractors environmental, and leak tests.
that will perform nondestructive testing on Nondestructive testing plays a key role during
production and test parts. Close communication ground, flight, and functional test programs to
between the nondestructive testing engineer and identify system and structural integrity and key
quality assurance nondestructive testing specialists is failure points. Nondestructive testing is typically
necessary early in the design process to develop the performed before the start of these tests to verify the
nondestructive testing philosophy and provide initial condition, at key points during the tests, and
adequate time to plan nondestructive test activities at the conclusion of the test to make a final
before fabrication of parts. Some of the determination of the integrity of the structure or
nondestructive test activities that require advance systems. This information is used torefine design
planning are (1) development of detailed inspection features of structure and systems before beginning
low rate initial production or full scale production.

AEROSPACE STRUCTURES 2.9

 PART5
Structural Failure Modes

One of the primary responsibilities of a structural

engineer is to design structures that will allow the
Elastic Deformation
aircraft to perform its mission for the designed Elastic deformation occurs when a load is applied
service life. All of the various inservice structural to a part and the part changes shape, but when the
failure modes must be evaluated and the design load is removed, the part returns to its original
adjusted to prevent these types of failures from shape. Elastic deformation itself does not normally
occurring during the design life of the structure. lead tofailure; however, elastic deformation ofone
Historically, these failure modes and analytical part may impact another part by changing the
approaches were developed for metallic materials clearance or interference that could lead to failure.
and were based on static and fatigue loads. As
composite material usage has increased in aircraft, a
modified setoffailure modes had tobe considered Plastic Deformation
because of the significant differences in material Plastic deformation occurs when sufficient load is
properties and response to various types of loadings. applied to a material such that it changes shape;
In parallel to the structural activities, systems however, when the load is removed, the part does
engineers evaluate potential system failures and not return to its original shape. A permanent set has
refine the design to prevent the failure of aircraft been created in the material that may also be called
systems during the design life of the air vehicle. ―inelastic deformation.‖ The amount of stress that is
required to create the inelastic deformation is known
as the yield stress. This failure mode does not lead to
Figure 10. Stress-to-strain curve for aluminum. immediate catastrophic failure of a part; in fact, the
part may carry additional load for some period of
time before fracture actually occurs. Plastic and
elastic deformations are easily illustrated on a
stress-to-strain curve as shown in Figure 10 for
u aluminum.
As load is applied to a part, the stress  is
y
pl measured in units of force P per cross sectional
area A.
f
P
(1)  
A
Stress 

1
0.2offset Strain e measures how much the part has
e1
elongated (or changed shape) measured in change in
length L divided byunit length L.

Elastic region
Yielding region L
(2)  
L
Strength 
Elastic deformation Plastic deformation
occurs when stress is occurs whenHecking
stress isregion
below proportional limit above proportional limit Typical strains before yielding in aerospace
materials are commonly in the range of 1 to
Legend 20 percent and are commonly reported as 1000 to
u  ultimate strength 20 000 microstrain where 1 microstrain is a factor of
y  0.2 offset yield strength 0.000001 change or 0.0001 percent. The elastic
pl  proportional limit modulus, also known as Young‘s modulus, is given
f  fracture strength as

(3) E  




2.10 AEROSPACE NONDESTRUCTIVE TESTING
 Table 4. Material characteristics.

Density Tensile Strength Yield Strength Modulus Elongation

Material
g/cm3 MPa (103 lb/in.2) MPa (103 lb/in.2) GPa (106 lb/in.2) percent
Graphite epoxy reinforced fiber
glass AS4/3501-6 — 0 degree
tape (MIL-HBK-17-2 F) 1.59 2000 (291) 135 (19.6) 1.2
Graphite epoxy reinforced fiber
glass AS4/3501-6 —
quasi-isotropic tape layup
(MIL-HBK-17-2 F) 1.59 738 (107) 56 (8.1)
IM7/5250-4 BMI composite
(ASM 2001b) 1.6 2830 (410) 161 (23) 0.9
Fiberglass — electronic glass
(ASM 2001b) 1.9 484 (69) 26 (4) 2.5
UNS A82024 wrought
aluminum alloy clad with
same, heat treatable,
temper 3 (ASM 1984) 2.78 483 (70) 345 (50) 73 (11) 18
UNS A87075 wrought
aluminum alloy clad with
same, heat treatable,
temper 6 (ASM 1984) 2.81 572 (83) 503 (73) 72 (10.4) 11
UNS R56406 titanium
alloy 6Al-4V (ASM 1984) 4.43 950 (138) 880 (128) 114 (16.5) 14
High strength steel (UNS
K92580 martensitic age
hardenable alloy) 7.89 1960 (285) 1720 (250) 193 (28) 14
High temperature steel
(UNS N07718 nickel
chromium alloy precipitation
hardenable) 8.82 1400 (200) 1170 (170) 208 (29.6) 20

Material stiffness is defined by Young‘s

modulus E that is derived from the relationship of
Fatigue
stress and strain in the elastic region (see When failure occurs in a structure that has been
stress-to-strain curve). Table 4 lists some material subjected to repeated loading (at levels below the
characteristics such as modulus and yield strength. yield strength of the material), the failure mode is
known as fatigue. Loads may be constant or variable
amplitude. Fatigue failures account for a high
percentage of all mechanical fractures in the
Buckling (Compression Failure) aerospace industry; therefore, it is very important
In aerospace structural analysis buckling that it be a main consideration in design of the
describes the failure of a structure when a portion of structure (Schijve 1994, 2007).
the structure has plasticdeformation perpendicular
to the primary direction of loading. This compressive
or shear loading greatly diminishes the ability of the
part to carry load when enough load is applied to a
Creep
material to make it change shape. In buckling, Creep is the failure phenomenon associated with
permanent set occurs, which means that when the thelong term effect of constant load (below theyield
load is removed, the part does not return to its strength). Creep failure is more severe at elevated
original shape. Buckling typically occurs in thin temperatures. During this type of loading, elongation
members loaded in compression or shear mode and continuously occurs at a very slow rate until failure
is controlled by the thickness, the distance between eventually occurs. This type of failure is the major
support, and moment of inertia of the cross section design driver in propulsion systems that operate at
(Grandt 2003). elevated temperatures for extended periods of time.

AEROSPACE STRUCTURES 2.11

 in most instances by performing a baking operation
Corrosion after exposure to the hydrogen environment. Baking
Degradation of metallic materials because of the metal at a low temperature allows the hydrogen
chemical or electrochemical attack is known as to migrate out of the crystalline structure that has
corrosion. Many forms of corrosion are a concern in been expanded because of the higher temperature.
the aerospace industry; some of them are: dissimilar Some high strength fasteners can still develop
metal (galvanic), pitting, intergranular corrosion, hydrogen embrittlement failures after baking when
microbial, crevice corrosion, exfoliation, filiform, placed under a high load, which strains themetal
and stress corrosion cracking. Corrosion locally lattice structure and allows a local area in the lattice
reduces the thickness of the material, which can be to be an accumulation point for small amounts of
detrimental in several ways. First, the decrease in dissolved hydrogen. Eventually, sufficienthydrogen
crosssectionofastructural memberrequiresthat the migrates to the region that a rupture occurs.
same load be carried over a smaller area, thereby Hydrogen embrittlement can also be an issue in
increasing the stress in the member. Another titanium alloys, particularly those used at elevated
detrimental effect of corrosion is that the metal temperatures inpropulsion applications. To prevent
surface is no longer smooth and stress risers are inadvertent contamination, guidelines are in place
created where concentrated stress can be the regarding acceptable service adhesives, coatings, and
initiation site for a fatigue crack. Most corrosion other surface treatments. Adherence to cleaning
concerns are handled during the design phase by: guidelines should also be followed to prevent
proper material selection, adequate drainage (to contamination.
prevent water entrapment), controlling dissimilar
metal contact, using nonconductive barriers, and
applyingprotective coatings andfinishes.
Wear
Wear is the erosion of material of one solid
surface by another material (typically solid.) Three
Hydrogen Embrittlement principal wear processes are adhesive wear, abrasive
Hydrogen embrittlement is a phenomenon that wear, and fretting wear.
occurs when hydrogen is dissolved in a metal and is Effects of wear may be greatly reduced during the
therefore trapped, allowing the hydrogen pressure to design of aircraft components. Some of the most
build up to a point of localized rupture within the common ways to improve wear resistance are by
crystalline structure. When a metalhas beenexposed choosing proper materials for faying surfaces, use of
to a process that allows hydrogen to migrate into the lubricants, improved surface coatings and finishes,
crystalline structure, steps mustbe taken to allowthe use of hard coatings, reduced pressures between
hydrogen to escape. faying surfaces, and reducedspeedbetweensurfaces
High strength steels are most susceptible to that are in contact and in motion. Wear surfaces are
hydrogen embrittlement. These materials are detrimental in several respects: reduced cross section
typically used for high strength steel fasteners, for carrying load, localized surface imperfections for
landing gear, arresting gear, and to attach hardware. corrosion to begin, and stress risers for the initiation
One of the most common ways in which hydrogen point of cracks.
can be trapped is during electrochemical plating
operations. Hydrogen embrittlement can be avoided

2.12 AEROSPACE NONDESTRUCTIVE TESTING

 Evolution of Structural Design PART6
and Fatigue Considerations

During the past 100 years of aerospace was split into two agencies: the Civil Aeronautics
development, new demands by mission requirements Administration (CAA) and the Civil Aeronautics
have optimized performance in every manner. Board (CAB). The CAA was responsible for air traffic
Although improved materials, construction methods, control, pilot licensing, safety enforcement, airway
and analysis can produce very long service lives, development, and aircraft certification. Civil Air
engineers continue to be challenged to avoid Regulations (CARs) were created by the CAA to
overdesigning structures so that range, speed, replace ABs and became the focal point of the
payload, and service lives may be optimized and original certification of aircraft from the 1940s
balanced. through the early 1960s. As such, the CARs may still
The evolution of structural design and fatigue in be needed as a reference for older aircraft or as a
both the military and commercial sectors drove the standard for minor changes to older aircraft design.
need for and implementation of nondestructive CARs established the requirements for air vehicle
testing to assist in the certification ofnew designs structural design and strength that were applicable
and to ensure the continued airworthiness of to all types of civil aviation, both fixed and rotor
inservice aircraft. Rotorcraft, spacecraft, missile, and wing.
rocket design experienced a similar evolution of In 1958, the United States Congress passed the
design and airworthiness requirements Federal Aviation Act, which transferred the CAA‘s
(Lazzeri 2007). functions and responsibilities to a newly created
body named the Federal Aviation Agency. The
responsibilities included authority to minimize
Regulatory Evolution aviation hazards, creation of safety regulations
(previously handled bythe CAB), and creation ofa
In the early days offlight, theairframe wasmade shared civil and military system of air navigation
of wood, cloth fabric, and wire. Designers were and air traffic control. Congress found it necessary
primarily concerned with getting the aircraft to pass additional legislation in 1966 to create a
airborne. Damage that occurred to the structure cabinet level position named the Department of
during any aspect of the flight or recovery was Transportation (DOT), which combined major federal
easily repaired. In 1926, the United States Congress transportation authority and responsibility to a
passed legislation known as the Air Commerce Act, central body. The Federal Aviation Agency was
which began the regulation of civil aviation. With moved under DOT and its name was changed to the
the full support of the rapidly growing aviation Federal Aviation Administration (FAA). Another new
industry, the Aeronautics Branch of the Department body named the National Transportation Safety
of Commerce was formed. The department initially Board (NTSB) was created to investigate all
concentrated on issuing requirements for air traffic transportation accidents and effectively assume the
safety rules, licensing pilots, and certifying aircraft. CAB responsibility. On 1 February 1965, the Federal
Aeronautical regulations used between 1926 and Aviation Administration migrated all previous
1938 were the Aeronautical Bulletins (ABs), one of requirements and regulation documents from CARs
which, 7a in October 1934, provided ―Airworthiness and CABs to Federal Aviation Regulations (FARs) in
Requirements for Aircraft.‖ This document requireda an initiative known as recodification. Table 5
stress analysis supplemented by test data for all identifies the main airworthiness standards.
primary structural members. It established a
minimum safety factor of 1.5, required allowable
loads to be based on guaranteed physical properties
Table 5. Federal Aviation Administration recodification of regulations
of the material, and required that the applied stress
in any material must not exceed the yield point. (February 1965).
Between World War I and the early 1930s,
airframe and engine technology continued to Federal Aviation
Airworthiness Standard
advance along with metal technology. The initial Regulation
commercial application for aircraft was transport of
the United States mail. One of the first all metal FAR 23 Normal, utility, acrobatic, and commuter category
airplanes
aircraft, built by Ford Motor Company in 1927, was
the 5-AT-B. It carried mail but was also the first FAR 25 Transport category airplanes
plane designed to carry passengers, primarily to view FAR 27 Normal category rotorcraft
Niagara Falls. FAR 29 Transport category rotorcraft
The Civil Aeronautics Act of 1938 changed FAR 33 Aircraft engines
authority for civil aviation from the Commerce
FAR 35 Propellers
Department to a new government agency named the
Civil Aeronautics Authority. In 1940, the Authority
2.13
 maneuverability (fighters). In addition, new missions
Fatigue Effects for Aircraft on aircraft carriers required short takeoffs and harder
Design landings with a pitching ship deck and had increased
From the 1930s to early 1940s, civil and risk for corrosion. These higher mission and
commercial air transport became more widespread performance requirements drove materials to have
with the introduction of the Boeing 247 and the higher static strengths; however, little consideration
Douglas DC-1 and DC-3. These aircraft were was initially given to fatigue strength.
significant in shifting moreload carrying ability to As technology advanced in flight science,
the exterior surface of the plane. As payload became materials, and aircraft missions (both type and
an important consideration, static and ultimate frequency), the structural load spectrum became
strength became the focus of the airframe whereas much more cyclic and strenuous. Unfortunately,
fatigue aspects of design were limited to the engine. catastrophic fatigue failures played a key role in
These were the first aircraft to take advantage of the advancing airframe design technology to accurately
strengthened 2000 series aluminum alloys that account for fatigue loading.
coincidentally had improved fatigue resistance In 1952, the De Havilland Comet was the first
(Roylance 2001). commercial jet transport to enter service (Figure 11).
During the era that covered World War II and the The aircraft was capable of flying at higher speeds
Korean conflict, military needs required improved and altitudes than contemporary propeller aircraft
and varied performance of aircraft. Aircraft were and was subject to an extensive series of flight tests
asked to fly longer distances (transports), and to and certifications in 1949-1952, before entering
carry heavier payloads (bombs) and have increased service. However, in early 1954, two aircraft were
lost while ascending to cruise altitude. Design
modifications were incorporated on remaining fleet
aircraft; however, several months later, a third
Figure 11. De Havilland Comet™ (British Airways Heritage Centre).
aircraft was lost, which prompted a grounding of the
remaining fleet. An in-depth investigation was
initiated to identify the failure mechanism. One of
thefleet aircraft that had accumulated 1200 flight
hours and over 700 cabin pressurization cycles was
removed from service for use in a full scale
pressurization test. Catastrophic fatigue failures were
observed on an escape hatch and at several windows
on the test article because of stress concentrations at
the square corners (Atkinson 1960). Comparison of
the pressurization test to failure analysis of
recoveredstructures fromthe fleet failuresconfirmed
the cause ofthe accidents. Results ofthis accident
investigation fundamentally changed structural
design philosophy for commercial transport aircraft
and resulted in the CAA issuing CAR 4B Amendment
3 in which the concept of safe life was first
introduced.
Fatigue evaluation of flight structure was defined
in FAR 25.571 with the release of the recodified
regulations in February 1965. Numerous
Figure 12. B-47 Stratojet™ landing with drag chute.
amendments have been issued since then to add
damage tolerant design and to modify fail safe
requirements.
The first new United States Air Force (USAF)
aircraft design following World War II was the swept
wing, multijet engine aircraft known as the
B-47 Stratojet (Figure 12). Over 2000 aircraft were
produced between 1947 and 1956 by three
contractors: Boeing, Douglas, and Lockheed. Aircraft
designers used a strength based design and
consequently did not predict a usage life for the
B-47. The original primary mission of the aircraft
was a medium range, high altitude bomber for Air
Table 6. Increase in thrust. Force postwar, ―Cold War‖ strategy. During the
production run of three models, the gross weight and
engine thrust requirements grew asshown in
Gross Mass Thrust per Engine
Model Table 6. The aircraft mission evolved to a low
Mg (103 lb) kN/kg (103 lbf/lb) altitude bomber to avoid Soviet surface-to-air
missiles. A hard popup maneuver had to be
B-47A 56.7 125.0 39.2 4.0 incorporated in the bombing run to clear the bomb
B-47B 83.9 185.0 56.9 5.8 blast. These new missionrequirements significantly
changed the loading spectrum for the B-47 airframe.
B-47E 93.8 206.7 58.8 6.0 In 1958, two B-47 B-models, with less than
2500 flight hours, broke up during flight in separate

2.14 AEROSPACE NONDESTRUCTIVE TESTING

events. During theongoing investigations ofthese hours. The failure occurred in pulling 3.5 G, less
incidents, three Emodel aircraft, all withless than than half the design limit load. The accident
1500 flight hours, experienced catastrophic failures investigation attributed the failure to a forging lap
in separate incidents. With the early failures on five that was about 0.9 in. (2.3 cm) long by 0.2 in.
aircraft, the United States Air Force was faced with a (0.5 cm) deep in the lower plate of the left wing
difficult challenge to ensure the continued pivot box. This discontinuity was missed during
airworthiness of the aircraft. A focused team of production level inspections. Fractography revealed
airframe contractors was tasked with determining the a very short fatigue crack before the catastrophic
cause of the failures. New approaches were overload failure. This accident, along with additional
developed to define the various parameters that fatigue problems during the airframe test programs,
affected aircraft loads including number of takeoffs prompted an investigation focused on D6AC high
and landings, high gravity and rolling pullup strength steel in several critical components. The
maneuvers at low altitude and wind gust/turbulent material was found to have a high variability in
weather. Armed with the new analytical approach, fracture toughness, was extremely notch sensitive,
the load spectrum was verified on three separate and had a short crack growth life. To clear the
fatigue tests. Early failures occurred in the longerons aircraft for flight, a cold proof test was accompanied
on two test articles. Structural repairs were by detailed nondestructive testing. The result of the
incorporated and tests were resumed. Subsequent F-111 investigation was the development of the
failures occurred on the wings, which were repaired, MIL-STD-1530, ―Aircraft Structural Integrity
and once again the tests resumed. Eventually, major Program, Airplane Requirements,‖ that recommended
fuselage cracking occurred on all three test articles damage tolerant design methodology for future
and testing was halted. The results of these tests aircraft (USDOD 1996, 2002). ―The Military
drove major design changes that were retrofitted on Specification, Airplane Damage Tolerance
the aircraft to allow them to fly well into the 1960s Requirements, MIL-A-83444 (USAF) was issued in
(Wanhill 2002). July 1974 and presented detailed damage tolerance
This entire activity became the cornerstone for requirements as a function of design concept and
the USAF Aircraft Structural Integrity Program degree of inspectability‖ (USAF Handbook for
(ASIP), which established baseline design Damage Tolerant Design).
requirements for all future USAF aircraft via a As technical concepts regarding damage tolerant
United States Air Force technical memorandum. In design progressed, United States Air Force
1960, the United States Air Force issued the requirements documents were revised and additional
MIL-A-88xx series of specifications, which more ones issued. The United States Department of
thoroughly documented and clarified all aspects of Defense (DOD) created a uniform approach to
ASIP. damage tolerant design in 1998 with the issuance of
General Dynamics Corporation was awarded a the DOD Joint Services Specification Guide,
contract by the United States Air Force to develop JSSG-2006. This document has replaced all previous
and manufacture the F-111 fighter bomber in 1962 Air Force ASIP and damage tolerance requirements
with first flight in 1964. F-111 Ship #94 lost the left for future aircraft design. The chronology of United
wing in alowlevel training flight in December 1969. States Air Force damage tolerance requirements is
The aircraft was lost and both pilots died. At the summarized in Figure 13.
time of the accident, the aircraft had only 107 flight

Figure 13. Evolution of United States Air Force damage tolerance design requirements.

1998. JSSG-2006, Department of Defense — Joint Service

Specification Guide

1996. MIL-HDBK-1530, Military Handbook — General Guidelines for

Aircraft Structural Integrity Program

1990. AFGS-87221, Air Force Guide Specification — General

Specifications for Aircraft Structures

1985. MIL-A-87221, General Specifications for Aircraft Structures

1970. MIL-A-83444, Aircraft Damage Tolerant Requirements

and MIL-STD-1530 Aircraft Structural Integrity Program

1960. MIL Spec - 8800, Series for Airplane Strength and Rigidity

1958. ARCD-AMC, Program Requirements for the Structural

Integrity Program for High Performance Aircraft

AEROSPACE STRUCTURES 2.15

 Static Life, Infinite Life, and Safe Life
PART7

The design philosophy of aerospace structures

plays a critical role for nondestructive testing. The
Infinite Life Design
structures community and the nondestructive testing As aircraft began to age and operational
community must interact because the structural requirements increased, structural engineers began to
design philosophy for safety is driven by the observe failures below theultimate strength ofthe
discontinuity sizethat can betolerated and which material. It was quickly recognized that fatigue
must be detected by some means of inspection. failure ofamaterial had tobe considered in aircraft
Although nondestructive testing is closely tied to design. Initial engineering knowledge of fatigue was
materials and manufacturing issues, it is the based on failures found in the railroad industry in
structural engineer who in the end determines the rail car axles. Early fatigue data were developed on
viability of the structure based on any anomalies steels using a constant amplitude stress level, so by
that may be discovered during manufacture or cycling numerous specimens to failure at different
inservice inspection. Once an aircraft is in service, constant load levels, a plot of the stress level versus
fatigue, corrosion, and damage that can occur over the number of cycles to failure (on a logarithmic
the life of an aircraft are sources of anomalies or can scale) would produce a fatigue curve (also known as
affect the growth of anomalies that may have an SN curve) as shown in Figure 14.
existed below the detection limit at the time of Althoughinitial studies had not identified the
manufacture. The design philosophy is critical to the microstructural mechanisms that cause fatigue, the
interpretation of anomalies for continued use or plot of results from constant amplitude fatigue tests
repair. provided an important observation that addressed
fatigue failures. The key to the plot was that the
SN curve flattens out for ferrous materials so that,
below a certain stress, failure does not occur no
Static Life Designs matter how many load cycles are applied. This point
The earliest aircraft designs were based on the is known as the endurance limit. Fatigue specimens
ability of the structure to carry a static load. Later have a wide scatter band; therefore, a large number
design concepts required aircraft to fly longer and of specimens at various loads is required to develop
faster, carry more payload, and have greater a curve witha high statistical confidence, making
durability. One way to accomplish this was to SN curves very costly to develop. When engineers
strengthen thestructure to carrythe heavier loads; began developing SN curves for nonferrous alloys
however, this required larger, heavier propulsion (particularly aluminums), it was observed that the
systems to get the heavier structure airborne. The materials did not have an endurance limit, as
developmentofthe stronger 2000 series aluminum illustrated in Figure 14. A common approach to
alloys helped reduce structural weight. identifying resistance to fatigue for nonferrous
materials is defined bythe stress required to fail a
specimen at 108cycles. There is also a limit to the
Figure 14. Fatigue curve. rate of cycling because of limits of the test machine
components and heating ofthe samples. In some
instances, it can take morethan 11days toobtain
107 cycles.
Although the endurance limit approach to fatigue
Stress (relative scale)

life was reasonable for industries that used ferrous

materials and had constant amplitude loading, it was
not considered a practical solution for the aerospace
industry. The problem with this approach is that the
SN curve is developed with pristine specimens that
Ferrous materials do not have corrosion, nicks, dings, or scratches and
Endurance limit a constant amplitude stress is used during cycling.
Because surface imperfections were found to have a
significant impact on fatigue failures along with
Nonferrous materials variable amplitude loading and residual stresses, a
reliable approach to design using the concept of
108 cycles to failure was quickly abandoned for
aerospace applications.

Number of cycles (logarithmic scale)

2.16 AEROSPACE NONDESTRUCTIVE TESTING

Safe Life Design Figure 15. Safe life approach.
The key principle of safe life design is that
primary structural components of the air vehicle will
not experience crack initiation during the design life

Crack length
of the air vehicle. ―The United States Air Force
(USAF) initiated the Aircraft Structural Integrity
Program (ASIP) in November 1958 using a
probabilistic approach for establishing the safety of Crack initiation life Crack growth life Critical
their aircraft. This process was called safe life and crack
length
relied upon the results of a laboratory test of a full
scale airframe. Aircraft failures in service arising acr
from the safe life process demanded a fundamental
change be made in the approach to design,
qualification, and inspection of aircraft‖ (Gallagher
1984). A safe life program for an air vehicle may be
set up as follows: (1) have an analytical design life Initial crack length
for primary structural components and assemblies
that is typically four times the air vehicle design life ai
and (2) perform full scale testing to prove that the Life
structure can survive two design lifetimes without Service life Aircraft tested Aircraft analytical
crack initiation. Figure 15 shows the safe life of aircraft to twiceservice design life to four
approach. life times service life
It has been common to use a safety factor of four
against the mean fatigue life of a material to account
for the large scatter band of fatigue test results when approach called damage tolerant design, based on
determining design life. The use of safety factor for present United States Air Force and Federal Aviation
design and test are intended to assist in creating a Administration requirements. However, the safe life
low probability of fatigue failure. Design fatigue approach is still used in somestructures such as
spectrums used in analysis and tests are based on landing gears and helicopter components. Landing
severe usage, typically representing the 85th to gear components are typically made of high strength
90thpercentile of planned fleet usage, representing steels that are notch sensitive and have short crack
aircraft like the United States Air Force growth lives. Nondestructive testing plays a key role
Thunderbirds. in safe life structure because of the need to find
The safe life approach to structure design nearly all detectable anomalies to optimize fatigue
provides for a significant safety margin in the life. Unexpected failures found in test articles and
fatigue life of a structure, such that cracks that for inservice air vehicles require numerous
might develop will not become critical during the inspection bulletins as well as structural and systems
useful life of the structure. One of the ways to assist changes to preclude early retirement of the air
the safe life design is to define and establish critical vehicle or component.
parts lists that would define the parts or assemblies With the extraordinary cost to design and launch
that, if failed, could cause loss of life, aircraft, or anew air vehicle, therearetwo approaches tolook
mission, or cause uncommanded release of at extending the life of the air vehicle beyond the
munitions. Safe life design methods would require original design objectives. First is stretching the
that an aircraft be removed from service after the initial fatigue test; the second is a safe life
design life, even though the structure may very well assessment program that examines the results of
be tolerant of extended use. On the other hand, there fatigue tests and fleet usage data to construct a safe
have been cases of safe life designs such as the life extension program. Structural analysts realized
KC-135, F-111, and F-5 where unanticipated that there were a limited number of fracture critical
anomalies were presentand the structure failed well parts that established the design life, so by
below the design service life of the air vehicle. inspecting and/or replacing them, the life of the
There are two primary concerns with a strict safe structure could be extended. The essential parts of a
life philosophy: one concern is that unanticipated life extension program include performing major
structural damage and environmental effects can disassembly of fracture critical structures coupled
expedite the crack initiation portion ofthefatigue with an extensive nondestructive testing program.
process; the second concern is that safe, useful Failures found during nondestructive testing require
structures can beremoved from service, which is structural repair to meet the desired life extension.
uneconomical. Some of the most critical structures may be replaced
Most recent designs for air vehicles have with new structures while retaining the less critical
abandoned the safe life approach in favor of a new structures for the duration of the vehicle life.

AEROSPACE STRUCTURES 2.17

PART8 Damage Tolerant Design

Damage tolerant design is just that — a design Today, it is well known that new production parts
that can tolerate damage. After years of using a safe have the potential of having an existing defect that
life approach in which the design depended upon is below the detectable threshold or was missed
crack initiation not occurring until after the air entirely during nondestructive testing. Likewise,
vehicle had reached its design life and theoriginal inservice air vehicles may have unplanned damage
fail safe approach that was based on redundant load suchasscratches, dings, dents, wear, and corrosion,
paths, structural engineers realized that existing which could serve as a stress concentration point
anomalies must be considered in the design phase. and reduce the crack initiation life. In addition,
The fail safe concept for aircraft structures began when nondestructive testing is performed on
in the late 1950s and was initially based on inservice air vehicles, an anomaly may be below the
redundancy in structural load paths and crack detectable size or inadvertently missed. Facing up to
arrestment features in part design. When a failure these hard truths resulted in the birth of a drastic
occurred in one structural member, the load was change in design philosophy. The new approach
taken up by another structural member. One of the assumes that a discontinuity is initially present in all
concerns became that, even though the load was primary structural components at the most critical
diverted to another structural member, the member locations and to account for that discontinuity in the
that took up the load would have its fatigue life determination of air vehicle life. This approach was
reducedbecauseofthe additionalloading. Therefore, an attempt to ensure that initial damage would not
it became important to perform inspections to detect grow to a critical size that would result in a
cracks at earlier stages and to implement a repair of catastrophic failure that compromised flight safety
the failed structural member. during the life of the aircraft.
Damage tolerant design philosophy continued to
mature through the 1960s to thepresent. Key factors
Figure 16. Residual strength and damage growth guidelines: that contributed to the evolution of damage
(a) strength; (b) damage (USDOD 2002a). tolerance philosophy are the following: full scale
fatigue testing of major assemblies and complete
aircraft using load spectra morerepresentative of
(a) service; advances in the field of fracture mechanics;
Indicates that strength capability continued development of computers in terms of
decreases with increased damage size memory and processing speeds for computations and
Residual static strength

analytical models along with model based

definitions; advances and growth in the field of
nondestructive testing; and improved probability of
Minimum required detection (POD) studies for establishing crack
strength load detection capabilities of nondestructive testing
equipment.

Minimum period of unrepaired service

Service exposure time
Preexisting Damage Growth
Damage tolerance program guidelines apply to all
safety of flight structure, whose failure (or
(b) undetected failure) could cause direct loss of aircraft.
The damage tolerance program begins with design
Damage size for existing damage, by using redundant load paths,
growth limit crack arresting features, and low stress concentration
Damage size

points for all critical components. Design features

Indicates that damage that have been found to improve fatigue life, such as
increases in time edge breaks, generous radii, and compressive surface
Assumed initi stress (because of peening), are valuable; however,
minim they are difficult to quantify and are generally not
dama allowed in analytical approaches to show fatigue life
al
um improvement. The damage tolerance program must
ge Minimum period of unrepaired service assume a crack existing before service induction and
Service exposure time then analyze residual strength and crack growth
rates for the load spectrum and the materials to
determine the critical crack length when rapid

2.18 AEROSPACE NONDESTRUCTIVE TESTING

growth will occur. This is shown graphically in anomalies are undetectable by standard
Figure 16. It is recommended that, because ―residual nondestructive test methods. It is logical that the
static strength generally decreases with increased smaller the size of an anomaly, the more anomalies
damage size, the residual strength and growth that can be present before there is a critical size or
guidelines are coupled through the maximum quantity for detection. Examples of this are lattice
allowable damage size, that is, the damage size anomalies in thecrystalline structure, internal
growth limit established by the minimum required microporosity, and internal microcracks. It is
residual strength load. The safe growth period assumed that these anomalies willbe cracks or
(period of unrepaired service usage) is coupled to initiation points for cracks. Figure 17 defines the
either the design life requirement for the aircraft or maximum crack size that could exist in the as
to the scheduled inservice inspection intervals‖ fabricated structure as ao. During initial production
(Gallagher 2005). (or spares‘ production), it is assumed that all cracks
Analytical design life must be verified by (or anomalies) larger than ao will be detected and
performing full scale fatigue and component tests. repaired. Anomalies that are smaller than ao may
The results of the analysis and the full scale test become initiation sites for cracks during the
programs willestablishareliable inspection interval. inservice life of the part. For a variety of reasons,
The program will also establish a strong damage inservice inspections may not be capable of
detection program using nondestructive testing to detecting cracks as small as production inspections.
improve the probability of crack detection and to Consequently, a probability of detection program
provide a clear decision process for the may be instituted to determine aNDI (the minimum
determination of repair, replacement, or retirement crack size that can be reliably detected by
nondestructive testing). Probability of detection is
of components that are damaged. Damage tolerance
discussed elsewhere in this handbook.
programs that institute and maintain a disciplined
Slow crack growth theory assumes that the
approach for all of these issues will achieve an
preexisting anomalies have already initiated a crack
optimum economical life for the air vehicle yet
but that crack will not propagate to a critical size
maintain an airworthiness program that ensures a
where rapid crack growth occurs before the next
high level of safety. When attempting to qualify
scheduled inspection. Slow stable crack growth rates
structure for airworthiness, two approaches have
must be ensured for a specific period of time.
been found acceptable: (1) slow crack growth and
Figure 18 shows the stable crack growth within the
(2) fail safe.
planned service life. This period depends on the
degree of inspectability for inservice aircraft.
Components must retain some minimum level of
Slow Crack Growth residual strength even with a preexisting crack that
In the slow crack growth approach, it is assumed is below the critical crack size acrit for catastrophic
there are preexisting anomalies in the material for all failure. As the initial small crack experiences loading
critical structure at the most critical locations. during the crack growth period, it will increase in
Figure 17 graphically illustrates an example of the size and become easier to detect. The challenge is to
size and quantity of anomalies that are either establish an inspection interval that will not be so
inherent in a material or induced during fabrication. long as to allow the crack to grow to failure.
Below some threshold size (and/or quantity), these

Figure 17. Discontinuity detectability theory based on size and quantity of anomalies (USDOD 2002a,
Walker 1979).

Discontinuities
inherent as Maximum
fabricated crack size that
could exist as Critical
Number of discontinuities

fabricated crack
Maximum crack size size
that will not grow Detectable
appreciably in service crack size
in service
Discontinuities
introduced Discontinuities detected by inspection
during and eliminated during production
fabrication
Discontinuities
introduced in service

aNDT acr
ai ao

AEROSPACE STRUCTURES 2.19

 This approach is normally applied to monolithic
Figure 18. Slow crack growth. structure (structure fabricated froma single piece of
metal with no joints or seams) and components that
only have a single load path.
Crack length The United States Air Force damage tolerance
program differs from the Federal Aviation
Administration for structure that is considered
Catastrophic uninspectable. The Federal Aviation Administration
Planned service life
failure requires uninspectable structure to bequalified with
safe life. However, the United States Air Force allows
a slow crack growth methodology to qualify
structure (even those with multiple load paths) if the
structure is considered (a) uninspectable, (b) difficult
ao to inspect without major disassembly, or (c) too
Size of complex to predict initial damage locations
preexisting
crack (Gallagher 1984).

Life
Fail Safe
A fail safe design approach allows for failure of a
primary structural component in a safe manner by
taking advantage of the redundant load path and/or
Figure 19. Fail safe approach. crackarresting featuresso that acatastrophic failure
of the entire structure does not occur. The structure
that remains intact picks up additional load from the
Inspection and failed member and is assumed to have a preexisting
crack detection
must occur during
crack, so there must be a slow enough crack growth
rate to allow the structure to make it to the next
Crack length

this period
Failure of inspection interval while carrying the increased load.
Size of backup It is important to set up the inspection interval based
preexisting structure on accurate analytical tools along with the results of
Failure of first crack in full scale testing and flight testing. The remaining
component backup structure, although capable of carrying the
structure additional load, will have a reduced life because of
ao the higher loads, so it is important to perform
Size of inspections and repairs at the appropriate intervals
preexisting to avoid catastrophic failure. Figure 19 shows the
crack failsafeapproach.
Life

Fatigue life of Fatigue life of backup

Institution of Modern Damage
first component structure (after load Tolerance Programs
(load iscarried by redistribution, load is
primary and only carried by Although the general approach to damage
backup structure) backup structure) tolerance is the same for civil and military aircraft
design, there are some differences in the
requirements documents. The Federal Aviation
Administration issues Federal Aviation Regulations
(FARs), which establish minimum performance
Figure 20. View of Aloha Airlines Flight 243 pressure cabin skin requirements for damage tolerance in aircraft
departure looking aft on left-hand side of fuselage (FAA training). structure and allows the airframe manufacturer to
specify the details of the requirement for the specific
aircraft design. The design is qualified by performing
analysis and full scale testing to gain acceptance
from the Federal Aviation Administration and
ultimately obtain an airworthiness certificate for the
type aircraft. The United States Air Force (USAF)
issued the Damage Tolerant Design Handbook that
provides guidelines and analysis methods to aid in
compliance with JSSG 2006, USAF Airplane Damage
Tolerant Guidelines for Metallic Structures. A
secondary purpose is to provide specific background
data and justification for the detailed guidelines.
The United States Department of Defense (DOD)
issued the Joint Services Specification Guide —
Aircraft Structures (JSSG 2006) in October 1998.
This document provides a unified Department of
Defense approach to setting specific requirements for

2.20 AEROSPACE NONDESTRUCTIVE TESTING

the design category (slow crack growth or fail safe), Aloha Airlines, Flight 243. Although a 5.5 m (18 ft)
degree of inspectability, inspection intervals, long section of the pressure cabin skin and the
assumptions for damage size (initial, inservice, and substructure above the cabin floor departed the
continuing damage), minimum required residual aircraft, the pilot was able to return the aircraft to
strength, damage size growth limits, period of ground with an emergency landing before the
unrepaired service usage, and remaining structure aircraft structure could completely break up.
damage sizes. Figure 20 shows a photograph of the aircraft. The
Design philosophy for commercial engines has
predominantly been based on safe life concepts. Like
airframe methodologies, the engine design
methodology uses knowledge of the number of Figure 21. Numerous small fatigue cracks grow and interconnect:
cycles for a given stress level. Upon reaching the (a) fuselage; (b) closeup; (c) detail (NTSB 1989).
design life, the component is removed from service.
Beginning in the 1990s, the Federal Aviation
Administration required damage tolerant concepts in (a)
jet engine design as described in FAA Advisory
Circular, 33.14-1, Damage Tolerance for High Energy
Turbine Engine Rotors (FAA 2006). To improve
safety, airlines are now required to perform
inspections of opportunity when engine disks are
off-wing for other routine maintenance reasons.
Engine manufacturers have developed additional
eddy current inspection techniques to perform these
inspections of opportunity.

Aging Aircraft
In the early stages of an aircraft program, a Frame station
design life goal is established and all analysis and
testing is based upon this initial goal. When faced
with the cost of a new aircraft, economics require
that almost every air vehicle extend its actual life (b)
well beyond the design life. As the aircraft ages, an
honest evaluation must bemade ofthe continued
airworthiness of the structure, the ability to perform
mission requirements, and the cost of inspections, Upper skin
maintenance, and structural repairs. Even with the
best analytical tools, unexpected damage occurs and Failure safe tear
accumulates with usage and time. In addition, the strap connection
interaction of cracks in multiple components
becomes extraordinarily complex, which necessitates
aprobabilistic risk analysis instead ofadeterministic
approach.
As the fleet of a type of aircraft accumulates
usage, damage, and wear data or if the aircraft is
desired to be used beyond the original design life, Adhesive
the original damage tolerance plan must be Upper row
fasteners
re-evaluated and adjusted. To sustain operational Lower skin
capabilities, the updated damage tolerance plan
becomes an essential element of an aging aircraft Typical skin lap joint
sustainment plan, which also includes logistics and
asset allocation plans.
Typical fatigue
(c) crack Outer of upper skin
Widespread Fatigue Damage Adhesive
Aviation design and sustainment concepts have
continued to evolve as new failure mechanisms have
been reported from inservice data. New cases of Stringer
catastrophic failures can drive the entire aviation
industry into a flurry of activity — including failure
analysis, fleet nondestructive testing, and analysisof UP

nondestructive test data, plus reevaluation of design

concepts and maintenance plans. One of the most
recentexamplesofthis is widespreadfatiguedamage.
A Boeing 737-200 had an explosive Inner or lower skin
decompression failure in the passenger cabin during
ascent to cruise altitude in 1988 while in service for

AEROSPACE STRUCTURES 2.21

 investigation attributed the failure mechanism to a probability of a specific location having a
rapid linking of numerous fatigue cracks in the same detectable multiple-site or multiple-element
skin of a longitudinal skin splice as shown in fatigue damage condition.
Figure 21, along with low cold bond durability and 2. The structural modification point is the moment
corrosion. This type of failure was termed in an aircraft‘s operational life when operations
widespread fatigue damage. should be ceased after consideration of
The ensuing investigation and subsequent report widespread fatigue damage on a two-lifetime
(Accident Report NTSB/AAR-89-03) from the fatigue test. Aircraft operation shall be
National Transportation Safety Board (NTSB) cited terminated until modification or part
the failure of the carrier‘s maintenance program to replacement has been performed to address the
detect the disbonding and fatigue damage, which led area of potential widespread fatigue damage.
to the failure of the lap joint and subsequent Ultimately, the Federal Aviation Administration
separation of the upper fuselage lobe (NTSB 1989). released Amendment 96 to 14CFR 25.571 in 1998 to
The safety issues raised in the NTSB report included provide new regulations, which require
the quality of air carrier maintenance programs and demonstration that widespread fatigue damage
the Federal Aviation Administration surveillance of would not occur within the design service objective
those programs as well as the human factors of of the airplane as proven by full scale fatigue test
maintenance and inspection procedures, training, (FAA 2010).
certification, and qualification. Another new concept that evolved from the study
The Aloha Airlines incident initiated worldwide of widespread fatigue damage was the limit of
activities to ensure flight safety and continued validity (LOV) of an airplane model maintenance
airworthiness of aging aircraft. The Technical program (Eastin 2007). Limit of validity is the
Oversight Group for Aging Aircraft (TOGAA) was moment in the airplane‘s operational life when there
chartered by the Federal Aviation Administration. is insufficient engineering data to support continued
The National Aeronautics and Space Administration, operation because of fatigue concerns. Currently, the
the Federal Aviation Administration, and the Federal Aviation Administration is considering
Department of Defense teamed together to begin an regulations to mandate inspection programs and the
annual conference on Aging Aircraft. Nondestructive determination ofthelimitof validity for each model
testing research programs were established at Sandia of aircraft. Additional data such as new fatigue tests,
National Laboratory and Iowa State University to service experience, and inspection results can be
address aviation inspection issues, particularly those used to extend the limit of validity fora model.
associated with aging aircraft. Air Transport
Association of America (ATA) and the Aerospace
Industries Association of America (AIAA) presented
the Federal Aviation Administration with a proposal Corrosion, Surface Finish,
to form the Aging Aircraft Task Force, an Wear, and Erosion
international group composed of technically
qualified individuals representing the airlines and Corrosion, surface finish, wear, and erosion are
aircraft manufacturing industry, to investigate aging all contributing factors in widespread fatigue
aircraft issues. The Federal Aviation Administration damage analysis and can add a degree of difficulty
accepted the proposal and the group‘s activities were and uncertainty. As an aircraft begins to age, the
shared with worldwide civil aviation authorities. In potential for corrosion increases, which may lead to
1992, this group became the Airworthiness cracks that may not have been anticipated in a
Assurance Working Group (AAWG.) damage tolerance assessment. Corrosion creates a
Because of the Aloha Airlines accident, several nucleation site for cracks that can reduce the fatigue
new failure mechanisms became the focus of the life ofacomponent (ASM 2001a). Stress corrosion
Federal Aviation Administration and thetask groups: cracking is particularly notorious for slow
environmentally induced nucleation and growth of
1. Widespread fatigue damage is the simultaneous crackswithsustained mechanical or residual stress
presence of cracks at multiple structural that is usually below the yield stress of the material.
locations that are of sufficient size and density To minimize the potential damaging effects of
such that the structure will no longer meet the corrosion it is important to have a robust corrosion
requirements of Federal Aviation Regulation control program. This begins in the design stage with
25.571(b). material selection, where consideration must be
2. Multiple-site fatigue damage is a source of given to a material‘s susceptibility to stress corrosion
widespread fatigue damage characterized by the cracking as well as dissimilar materials that will be
simultaneous presence of fatigue cracks in the in contact with each other. Finish systems, sealant,
same structural element. and drain holes can minimize the potential for
3. Multiple-element fatigue damage is a source of corrosion. Inservice aircraft should be periodically
widespread fatigue damage characterized by the inspected for corrosion by both visual and other
simultaneous presence of fatigue cracks in nondestructive testmethods aswell as missing or
similar adjacent structure. damaged finish systems that willlead to corrosion
Two new concepts have been incorporated into (ASM 2001a).
aircraft maintenance plans to address widespread Surface finish, wear, and erosion are also
fatigue damage: important factors in improving fatigue life of
structure. The fatigue life will be improved with a
1. The inspection starting point is the moment in
smoother surface finish that is free of deep scratches.
the aircraft‘s operational life when special
Removal of finish systems may also lead to
inspections are initiated because of the

2.22 AEROSPACE NONDESTRUCTIVE TESTING

unanticipated damage that can contribute to High strength materials are particularly susceptible
multiple-site fatigue damage or multiple-element to hydrogen embrittlement. Consequently,
fatigue damage failures. If material is inadvertently nondestructive test methods that do not require
thinned during mechanical removal of a finish removal of the finish system may reduce the
system, a crack can develop earlier than anticipated potential for unintended damage. This must be
in the original damage tolerance assessment. balanced with the improved sensitivity of a
Chemical methods for removal of finish systems can nondestructive testmethod when the finish system is
cause delayed cracking (or hydrogen embrittlement) removed.
if subsequentembrittlementrelief is not performed.

AEROSPACE STRUCTURES 2.23

 Structural Design and Damage
PART9 Tolerance for Composite Materials

Development and use of composite materials in plastics, on the other hand, differ in that after the
aerospace applications originated during World initial flow and formation of a shape, they cannot be
War II. Material scientists were attempting to softened and reshaped by increasing temperature —
develop materials with good specific strength and in essence their shape is set (ASM 2001b).
stiffness that surpassed existing structural materials. Initial applications of composite materials in
One of the earliest uses of composite material was in aircraft structures were primarily simple shapes that
glass fiber reinforced polymers for rocket motors were not primary structure. Eventually composite
(ASM 2001b). Plastics were a material type that materials migrated into doors, fairings, and control
consisted of organic polymers combined with surfaces and finally into primary load carrying
hardeners, fillers and reinforcements. The polymer structure. Before considering damage tolerance
and its additions are typically processed at elevated characteristics of composite materials, it is necessary
temperatures that produce flow of the polymer to to provide a brief overview of composite materials,
help achieve the final shape, which is retained upon their product forms, properties, and the typical
cooling to a solid state. Plastics are categorized into anomalies that occur during manufacturing and
either thermoplastics or thermosets. Thermoplastic assembly as well as inservice damage.
materials are capable of being repeatedly softened by
increasing temperature and then hardened with a
decrease in temperature. When heated, Composite Materials
thermoplastics can flow and be molded or extruded
into a new shape and, after decreasing the Numerous industry groups have provided
temperature, the material hardens and retains the definitions of composite material that are similar but
shape until heated again. The sequence of not identical. This text uses a hybrid definition
heating/softening/flowing/shaping can be performed compiled from Composite Materials Handbook 17,
repeatedly and the material characteristics always ASTM D 3878, and SAE AIR 4844 (ASTM 2007,
remain the same after cooling/hardening. Thermoset SAE 2009).

Table 7. Organic matrix composites (Campbell 2004).

Classification Polymer Resin Characteristics and Usage

Thermosets Bismaleimide High temperature resin matrix, 135 to 175 °C (275 to 350 °F); processing similar to epoxies
but requiring post cure (carbon and glass prepreg composite used on F-22 aircraft)
Cyanate Ester High temperature resin matrix for 175 to 315 °C (350 to 600 °F); processing similar to
epoxies and BMI but higher cost; used for space applications,
Epoxy High performance matrix common in aerospace structure usage, 120 to 135 °C (250 to
275 °F)
Phenolic High temperature resin system with good smoke and fire resistance, used for aircraft interior
Polyester Commercial use, inexpensive, processing flexibility, continuous or discontinuous
reinforcement
Polyimide Very high temperature, 260 to 315 °C (500 to 600 °F); difficult to process, noncombustible
Vinyl ester Similar to polyesterbutbettercorrosion and moistureresistance, highercost
Thermoplastics Polyether etherketone Highglass transition temperature(heat resistance) withgoodmechanical properties, most
widely used, crystalline morphology
Polyetherketone ketone Similar to polyether etherketone with slighter better properties and environmental resistance
Polyetherimide Amorphous morphology, lower temperature resin, low cost
Polyphenalsulphide Semicrystalline morphology, high heat resistance and good mechanical properties
Carbon-carbon not applicable Veryhigh temperature applications suchasrocketnozzles and brackets

2.24 AEROSPACE NONDESTRUCTIVE TESTING

 Composite material is a substance consisting of
two or more materials, which are combined to form Figure 22. Reinforcement fiber types, continuous. More typically,
a useful engineering material possessing certain unidirectional (or tape) in prepreg form is not laced but just held
properties not possessed by the constituents. The together by resin (Campbell 2004).
individual materials differ in formor composition
Unidirectional Cloth Roving
and typically consist of reinforcing elements, fillers,
and a matrix binder. A composite material is
inherently inhomogeneous on a microscopic scale
but can often be assumed to act in a homogeneous
manner on a macroscopic scale. The materials are
insoluble in one another but when combined can be
physically identified and exhibit an interface
between one another. The properties of the
individual components are typically inferior to the ±30° (helical
properties of the composite material as a whole 0°(laced) ±45° (laced) ±90°(woven) filament wound)
(USDOD 2002b; ASTM 2007;SAE 2009).
Composite materials are significantly different
from metals because when two different elements are
combined, the resulting metal alloy has a
Figure 23. Reinforcement fiber types,
homogenous structure onamacroscopicscale. Metal
alloys result from combining materials that are discontinuous (Campbell 2004).
soluble in a metallic host matrix. The metal alloy has Mat Chopped
new chemical, physical, and mechanical properties
that differ from the original constituents. Metals and
their alloys may exhibit isotropic or anisotropic
properties depending onthegrain size, structure,
shape, and fabrication method. Isotropic material
properties are about the same in all directions
whereas anisotropic material properties vary by
direction. Although metal alloys may have some
anisotropic properties, the directional variance is of the vapor deposition process, where fibers must be
minimal compared to most composite materials, made individually. The fiber has a high modulus and
whichhave highly directional properties. extremely good compression properties. However, it
Composite materials may be developed and used is difficult to cut and machine and is not easy to
for structural applications as well as electrical, fabricate into complicated shapes because of its large
thermal, wear, and corrosion properties. Because a diameter (Campbell 2004).
high proportion of aerospace nondestructive testing Because oftheir low cost and goodmechanical
is applicable to structural load carrying elements and properties, glass fibers are the most common
assemblies, most of this discussion will be focused reinforcement material (Campbell 2004). Glass fibers
on structural composites. In structural applications, come in two types: E glass, or electrical glass, and
composite materials rely on a matrix with S glass, or structural glass. Glass/epoxy composites
reinforcement materials. The matrix is typically are more commonly known as fiberglass. Fiberglass
softer, whereas the reinforcement material is stronger cannot be used for high performance primary
and has a higher stiffness. structure because of its low stiffness properties:
The primary classes of matrix materials are wings that are made of fiberglass tend to flex too
organic matrix composites, metal matrix composites, much ordroop. However, S glass is widely used in
and ceramic matrix composites. Generally, aerospace aircraft radomes because of its excellent strength
structural composites use organic matrices. Some properties and transparency to high frequency radar
common matrix materials and their characteristics signals.
are shown in Table 7. Aromatic polyamide or aramid fibers are also
Reinforcement materials improve strength and known by the DuPont trade name Kevlar®. These
stiffness of the composite and may be used in several fibers have low density and are extremely tough
different forms (Figures 22 and 23). with very high damage tolerance properties. Aramid
Continuous reinforcing fibers are capable of can experience degradation from long term exposure
carrying much more load than particles or to ultraviolet radiation, is weak in both shear and
discontinuous fibers and are therefore the primary compression, and mustbeused attemperaturesless
reinforcement means for structural composites. than 177 °C (350 °F) (Campbell 2004).
Although fibers may have a high tensile strength, Material scientists and structural engineers
they are generally brittle and exhibit virtually no generally consider carbon fibers to have the best
yielding with a low strain to failure. The most overall properties compared to aramid or glass fibers.
common fibers used in composite structures are Carbon fibers have a wide range of tensile strengths
glass, carbon, aramid, and boron. and stiffnesses (as described by the modulus of
Early composite usage on the USAF F-111, F-14, elasticity). As such, carbon fibers are divided into
and F-15 aircraftstructuresusedboronfibers. Thisis three categories: standard modulus, intermediate
a relatively large diameter (125 m) fiber that is modulus, and highmodulus fibers aslisted in
manufactured by pulling a fine piece of tungsten Table 8. Generally, all three categories have low
wire through a reactor in which a chemical density, a low coefficient of thermal expansion, are
deposition of boron is applied to a uniform reasonably good in compression, and are conductive.
thickness. This is also a very expensive fiber because

AEROSPACE STRUCTURES 2.25

 resin in theindividual plies togointo a gel state and
Table 8. Carbon fiber properties. flow together, whereupon cooling to room
temperature creates a continuous structure.
Tensile Strength Elastic Modulus Material selection during design must balance
Modulus strength, stiffness (elastic modulus), fatigue, and
GPa (103 lbf/in.2) kN/kg (103 lbf/in.2)
weight along withother key properties associated
Standard 3.4 to 4.8 (500 to 700) 220 to 240 (32 to 35) with the service application. One of the ways to
Intermediate 4.1 to 6.2 (600 to 900) 280 to 296 (40 to 45) compare metals to composite material is tolook at
High 4.1 to 5.5 (600 to 800) 345 to 450 (50 to 65) the specific strength (strength/density) and specific
modulus (modulus/density). High strength carbon
composites can have higher specific strength and
There has been a tendency in the nondestructive modulus than commonly used metal alloys. Using a
testing community to improperly use the terms material with a higher specific strength and modulus
carbonand graphite interchangeably. In fact, there is can reduce weight of the structure which results in
a difference: carbon fibers contain 95 percent carbon
improved range and fuel savings along with
and are carbonized at 982 to 1482 ºC (1800 to improved performance and increased payloads
2700 ºF); graphite fibers are 99 percent carbon and (Campbell 2004).
are carbonized at 1982 to 3038 ºC (3600 to 5500 ºF).
Carbon fibers result in standard and intermediate
modulus properties whereas the graphite fibers result
in high modulus properties. Mechanical Properties of
Both carbon and graphite fibers are made from Composite Fiber Laminates
two different precursor materials, the mostcommon
being polyacrylonitrile and the other being pitch. Mechanical properties of composite materials are
These fibers range in diameter from 6 to 8 m (about established by coupon tests using static or cyclic
the diameter ofahuman hair) and are produced in loadings. Static testing may be short duration or
bundles known as tows. Tows are classified as 1k, long term loading. The majority of coupon testing
3k, 6k, 12k, and 24k where k means 1000 fibers. applies tensile, compressive, or shear loads. Cyclic
loading may be either constant amplitude or varying
amplitude (spectrum loads). Some of the more
common types of load applications are
Carbon Fiber Reinforced (1) tension/tension, (2) tension/compression, and
Polymer (3) reverse shear. Impact testing is also performed to
measure the material‘s ability to withstand impacts
Carbon fiber composites are often referred to as that occur during fabrication and maintenance
carbon fiber reinforced polymer (CFRP). Partsmay be operations or in service from projectiles, bird strikes,
fabricated as dry fiber mats over which an epoxy or debris. Polymer matrix composites are affected by
resin may be flowed or with the fibers precoated with environmental conditions such as temperature and
the resin (the material is known asprepreg) where the moisture. During the material selection phase of the
prepreg plies are stacked to form the part. Acurve design process, various environmental effects on
illustrating the general behavior of the properties of material properties of composites must be
the individual components and the effect of investigated. Even moreso than metals, temperature
combining the components is shownin Figure 24a. and moisture can have an impact on the properties
Prepreg is cut into pieces and collated to match the of composites. Moisture can affect metals in the
pattern of the part. The individual plies are typically form of corrosion, which tends to lower static and
laid up on a metal tool with a low coefficient of fatigue strength. Composite materials typically
thermalexpansion. The plies in the layup may be absorb moisture, which affects mechanical properties
eitheraunidirectionalorwovenmaterial; however, by degrading the fiber/matrix bond. Temperature
when unidirectional material is used, the fiber and moisture effects on mechanical properties may
directionofeachply mustbepositionedtoachievea not be linear and when combined can result in
balancedloadcarryingcapabilityforstrengthand significant property degradation.
stiffness. Usually, individualplies are orientedin a Becauseof this degradation, polymer composites
sequence of 0°, +45°, –45°and 90°called quasi- are typically exposed to different fluids before or
isotropic,shownin Figure 25. during testing. Fluids are generally placed into two
Prepreg typically has a backing material and a groups. The first is made up ofmaterials that have
cover material to keep layers from being the potential to pool or be in contact with the
contaminated or sticking together. These materials composite for extended periods of time such as fuels,
can tear during removal resulting in some of the hydraulic fluid, lubricating oils, methylene chloride,
material being inadvertently left in the layup during and ethylene glycol/urea (deicer) salt water solutions.
the cure process. When this occurs, the nonflyaway The second group consists ofmaterials that are in
consumable material is said to beincluded. brief contact with the composite, usually something
After all of the plies have been laid up on the that is wiped on and off such as isopropyl alcohol or
tool, the ply stack is covered with a plastic bag and alkaline and hydrocarbon based cleaning solutions.
sealed around the edges with vacuum putty so that a Tests for fluid effects generally involve some
vacuum can be pulled on the plies during cure. The period of exposure before mechanical testing. In
purpose of the vacuum is to facilitate compaction some circumstances, the coupons are subject to
and to assist in removing unwanted air and volatiles nondestructive testing between exposure and
that could cause porosity in thefinished part. The mechanical testing. When the mechanical tests are
tool is then placed in an autoclave for exposure toa not taken to failure, a nondestructive test may be
cure cycle specifically designed to control performed after subjecting the test coupons to the
temperature and pressure. The cure cycle allows the potential contaminants. Finally, hot/wet conditions

2.26 AEROSPACE NONDESTRUCTIVE TESTING

are usually simulated bysoaking the composite test structure is called cocured; when an adhesive is
coupons in hot water before testing. placed between the skin and stiffener and the
structure is cured simultaneously, the structure is
described as cobonded. Common structural shapes
Fatigue and Composites are shown in Figure 26. Similar approaches to
Generally, aerospace composite materials have
good fatigue properties. However, there is some Figure 25. Fiber direction and stackup sequence
variation between carbon, aramid (Kevlar®), and glass for quasi-isotropic layup (Campbell 2004).
fibersystems, where carbon is superior to aramid and
glass is inferior to aramid. Carbon/epoxy laminates
0°
arethe mosttypical composites incritical structures.
It is significant to note that their fatigue strength is
far superior to metallic materials used in aerospace
Bar indicates
structures, as shown in Figure 24b (Campbell 2004). +45°
plane of
symmetry
runs through
Composite Structure –45° this ply.
Composite laminates may be used in many forms,
the simplest being a flat laminate. In most instances,
the flat laminate requires special geometry or some 90° [0/±45°/9—0]s
type of structure to provide stiffness. Geometry
approaches may utilize a sine wave or corrugated
pattern. The laminate skin may also be mechanically
attached to a composite or metal substructure. The –45°
cost and weight efficiency of composites has seen
riveted metal substructure increasingly being
replaced by bonded assemblies and cocured +45°
composite structures.
Integrally stiffened composite skins use the epoxy
resin to cocure the skin to a structural shape. If a 0°
thin layer of adhesive is used between the precured
composite skin and a precured composite stiffener,
it is considered a bonded structure. When the
composite skin and the stiffening structure are in
direct contact and cured simultaneously, the Figure 26. Common cocured or cobonded
stiffener shapes.
Figure 24. Strain behavior of aerospace
materials: (a) stress versus strain in composites;
(b) fatigue properties (Campbell 2004).

(a)
Fiber
Tensile stress (relative scale)

Composite
laminate
C channel J stiffener

Matrix
Strain (relative scale)
I beam T stiffener
(b)

Carbon
Fatigue strength/density

epoxy

Titanium

Steel
Aluminum
Cycles to failure (relative scale)

AEROSPACE STRUCTURES 2.27

 stiffening are possible by mechanically attaching the tometals. Another challenge is in development of
stiffening shapes to the skin with fasteners or by analytical models to account for initiation and
using an adhesive to bond or cobond the stiffener to growth characteristics of manufacturing and service
the skin. induced anomalies (Lincoln 1986).
Nondestructive testing is used to identify the Development of material property data and
various types of anomalies and damage that occur design criteria for composites actually began with
during fabrication and assembly. Some typical issuance of Air Force-Navy-Civil Committee (ANC)
anomalies include porosity, disbonds, included Bulletin 17, Plastics for Aircraft. Initially the United
material, plydistortion, wrinkling, matrix cracking, States Army wastheprimary sponsor of ANC
splintering, and delaminations. Inservice damage to Bulletin 17 and the replacement MIL-HBK 17. The
composite laminates can occur from numerous document has evolved into the present day
stimuli, some of which are low and high velocity Composite Materials Handbook 17, with the
impacts, thermal exposure, fluid exposure, sponsorship role transferred to the Federal Aviation
wear/erosion, lightning strike and static overload, Administration. The evolution of Composite
fatigue, and compressive buckling or crippling. Materials Handbook 17 is found in Figure 27
(Ilcewicz 2002; Foedinger 2009).
Composite Materials Handbook 17 is the primary
Certification of Composite source for composite material data in modern
aerospace design. Contents of Composite Materials
Structure Handbook 17 is periodically reviewed and updated
Certification authorities have been challenged to by a joint committee of industry and government
establish structural integrity requirements for representatives. Currently, Composite Materials
composite structures to be used in aerospace Handbook 17 is contained in six volumes: Volume 1,
applications. One of the reasons for the challenges is Polymer Matrix Composites: Guidelines for
that there tends to be a greater degree of scatter in Characterization of Structural Materials; Volume 2,
strength and fatigue data forcomposites compared Polymer Matrix Composites: Material Properties;

Figure 27. Evolution of Composite Materials Handbook 17 (USDOD 2002b).

2009-10. Release of Rev. G - CMH-17 Handbooks

2006. Transition from Army to FAA as Primary Sponsor established

Roadmap to New Composite Materials Handbook “Release G”

2004. Joint Meetings with CACRC, SAE-P17

2002. MIL-HDBK-17B Vol. 1F, 2F, 3F, 4A, 5 commercial publication through ASTM

1998. Joint meetings with ASTM D-30

1996. CMC Coordination Group formed

1993. MMC Coordination Group formed

1990. First PMC Data Set approved

1988. MIL-HDBK-17B Vol. 1 Release

1971. MIL-HBDBK-17A Plastics for Aerospace Vehicles

1959. MIL-HDBK-17 Plastics for Air Vehicles

1943. ANC Bulletin 17 Plastics for Aircraft

2.28 AEROSPACE NONDESTRUCTIVE TESTING

Volume 3, Polymer Matrix Composites: Material that achieve a certain property value. The testing
Usage, Design and Analysis; Volume 4, Metal Matrix begins withpristine specimens toestablishabaseline
Composites; Volume 5, Ceramic Matrix Composites; performance, and then various anomalies are added
Volume 6, Structural Sandwich Composites. (one type per specimen) to determine the durability
Besides providing material property data, of the material and effects of anomalies (Campbell
Composite Materials Handbook 17 establishes 2004).
procedures and protocols for material property Nondestructive testing plays a key role by
development testing as well as guidelines for design determining the initial condition, damage initiation,
and analysis of composites. and growthof anomalies in themechanical testing
Certification of composite structures requires both of coupons. This is particularly useful in determining
analytical data and testing to verify design and the knockdown for various levels of porosity relative
structural integrity. The Federal Aviation to thickness in a material. Porosity is one of the most
Administration issued Advisory Circular 20-107A in common manufacturing anomalies, and the
1984 to establish certification procedures for acceptable level of porosity relative to retained
composite structure in commercial aircraft strength is one of the essential test programs that
(FAA 1984). Department of Defense programs have must be performed at the beginning of the design
typically used customized approaches to certification process. The relationship of porosity to retained
of composite structure, which are aligned with the strength is typically tied to a volume percent
basic premise of damage tolerance and structural porosity and decibels of attenuation during an
integrity programs yet also incorporate some or all ultrasonic examination ofthe material.
aspects of the building block approach described in During the design value development phase of
Composite Materials Handbook 17 and shown in the building block approach, other nondestructive
Figure 28 (MIL-HBK-17-F). testing initiatives such as effects of anomalies and
The building block approach begins with material detectability programs are performed, including radii
property development. Typically, generic coupon anomalies, cracks, impact damage, disbonds, and
specimens are tested to establish material allowables. included material in the layup. All aspects of the
This testing is performed with room temperature dry testing should be documented including fabrication
conditions (to establish a baseline for the material) of test specimens, test parameters, and results. Some
and is then followed by environmental testing (to of the key elements of fabrication are: configuration
account for variables such as moisture and control of specimen geometry and construction,
temperature). Numerous specimens are tested to material type, layup and debulking, cure, and
establish a scatter band that is then used to bagging details, as well as size and location of
statistically determine the percentage of specimens embedded anomalies. Important aspects of the test

Figure 28. Building block approach of Composite Materials Handbook 17 (USDOD 2002b).

Block 6
Group C
Analysis verification
Component
tests

Block 5
Sub-component tests
Group B
Design value
Block 4 development
Structural element tests

Allowable development
Block 3

Group A
Material property
Block 2 Materials and processes specification development development

Block 1 Material screening and selection

AEROSPACE STRUCTURES 2.29

 parameters are: test fixtures; loading schemes; Consideration must be given to both damage
environmental conditions; and strain sensor tolerance and damage resistance during the design
location, orientation, and recording. Besides a phase because there can be a competition in
written summary of the test results, it may be achieving these attributes. Material selection and
important to retain the actual specimens and raw structural configuration canaffect fabricationand
data as evidence to support the certification of the operational costs.
composite. Records may include photographs of the For example, toughened resin material systems
test part before and after test, nondestructive test typically improve damage resistance relative to
data, mechanical property tests, and section cuts untoughened systems, which results in reduced
through the test specimen. When actual test maintenance costs associated with damage from low
specimens are retained, they may be used at a future severity impact events. However, these cost savings
date in several ways: (1) to evaluate technical compete with the higher per-pound material costs
advancements in nondestructive testequipmentand for the toughened systems. In addition, these
methods, (2) to compare to new conditions observed materials can also result in lower tensile capability
from production or inservice parts, and (3) to be used of the structure with large damages or notches,
for stretch testing if the specimen was not testedto which might require the addition of material to
failure or section cut for evaluation. Most of this satisfy structural capability requirements at limit
testing is performed on structural elements and load. This extra materialand associated weightresult
subcomponents using static, dynamic and acoustic in higher material and fuel costs, respectively
loadings. (USDOD 2002b).
Final analysis verification in the building block The most common form of inservice damage for
approach is performed on component and full scale composite material is impact damage, which is
tests of the aerospace structure. Ground testing may categorized as either readily detectable impact
be done withorwithout environmental conditions damage (RDD) or barely visible impact damage
depending on contractual requirements. (In some (BVID). Damage level may be determined from a
instances, testing is performed at room temperature walk-around visual inspection, a general visual
dry, and then knockdown factors obtained during inspection, a detailed visual inspection, or a special
element and coupon testing are applied to account detailed inspection for nonvisible impact damage
for potential environmental effects.) Ground testing using nondestructive test procedures. Impact damage
is typically performed on several full scale structures. can occur from dropped tools and equipment during
One full scale article receives statically applied maintenance, runway debris, bird strikes, hail,
ultimate loads and a second is dynamically loaded to ballistics, departureorfailureofstructureorengines,
a spectrum representative of the anticipated design and collisions with ground vehicles or other aircraft.
loadsduringvarious flightmaneuvers. Configuration Based on the certification program, periodic
control ofthetest article, thetest parameters, and inspections may be required at defined intervals of
posttestevaluation and documentation are essential the service life of the composite component. Because
in the certification process. Nondestructive testing of salvage dispositions during the initial fabrication
plays an important rolethroughout thetesting life or during the service life interval, it may be useful to
cycle by evaluating structure after significant events have access to past inspection data and repair
are observed by strain sensors placed in strategic instructions. Inspection data may be in the form of a
locations throughout the test article. At the hand drawn illustration or electronic data. Because
conclusion of the test, the airframe is disassembled the life of some air vehicles may span decades, data
and evaluated with nondestructive testing and storage, software, and hardware become relevant
failure analysis to identify failures that may have issues and must be periodically considered as
occurred during the test. technology advancements occur during the service
life.
Damage tolerant design philosophy commonly
Structure Durability, Damage establishes three distinct damage categories
(Figure 29):
Resistance, and Damage 1. Damage beyond the critical damage threshold
typically occurs in flight from a discrete source
Tolerance event and is immediately obvious (such as bird
As with parts fabricated from metallic materials, strike, departure of an airframe component,
composite parts must show durability, damage engine fatigue failure, hail damage, or lightning
resistance and damage tolerance to meet the safety strike). Safe flight (with limited maneuvers and
and design requirements for aerospace applications. envelope) can continue with damage larger than
Damage resistance is a quantification of the damage the critical damage threshold to return the air
caused by a specific event, and the durability is vehicle to ground. This type of damage requires
related to the amount of time a component requires immediate repair before returning to flight. On
repair or replacement because ofdamage thatoccurs rare occasions, the airworthiness authority may
duringthe service life (USDOD 2002b). grant permission for a very limited envelope or
Damage tolerance addresses the structure‘s ability a one-time flight to transport the vehicle to a
to tolerate a specific damage condition during repair station.
service without compromising safe operation.

2.30 AEROSPACE NONDESTRUCTIVE TESTING

 Figure 29. Load versus damage size (Tomblin1999).
Fuselage pressure
Ultimate critical structure
Damage load requirement

Maximum load per airplane lifetime

Continued safe flight

Damage occurring in flight which

Damage may never Damage must be repaired is immediately obvious
be discovered when discovered

Allowable damage limit Critical damage threshold

Increasing damagesize

2. When damage larger than the allowable damage 3. Nonvisible or barely visible impact damage is
limit is discovered, it must be repaired. This type generally not detected by production or
of damage may require special detailed inservice inspections (including nondestructive
inspections for detection but should be easily testing). The air vehicle must withstand ultimate
detectable with that inspection method. load with no adverse impact on operation for
Consequently, the design should allow for thelifeofthe structure withthistype of damage
damage larger than allowable damage limit to (USDOD 2002b; Tomblin 1999).
withstand repeated loads and limit load (a once
per lifetime load) until the air vehicle reaches
the next inspection interval.

AEROSPACE STRUCTURES 2.31

 Bonded Assemblies
PART10

Bonded assemblies are used in manufacturing and

in field repairs to reduce weight and cost. Some of
Laminate Bonds
the advantages and limitations of adhesive bonding One ofthe simplestbonded constructionmethods
are listed in Table 9. Bonded assemblies maybe all is skin-to-skin, which may be accomplished by
metal, metal composite or all composite materials. In cocure, cobond, or secondary bonding. Some typical
composite materials, the adhesive bonding is broken adhesively bonded flat skin joints are shown in
into one of the following three methods: (a) a Figure 30. Load is transferred from one component
secondary bonding process, (b) acocure process or of the bonded assembly to another through the
(c) a cobonding process. The secondary bonding adhesive bondline. Shear load transfer is the optimal
occurs when an adhesive is used to bond the two use of adhesive bonded structure. Loading of bonded
materials after both have been cured (if it is ametal joints in cleavage or peel is not recommended
to composite, then it is after the composite has been because failures usually occur quite rapidly
cured). When the bonding is performed (Campbell 2004).
simultaneously with the cure of the composites and
the resin is used as the adhesive, the resulting
bonded structure is described asintegrally cocured. Sandwich Structure
If an adhesive other than the composite resin is used
Adhesive bonding is also used extensively for
and the bonding operation is performed at the same
sandwich structure. In this type of assembly, a core
time as the composite cure of one of the
material is adhesively bonded to two relatively thin
components, then it is known as cobonded structure.
skins. The core material may be metal, glass fabric,
Because of the advantages of bonding (Table 9),
aramid paper, aramid fabric, carbon fabric, foam or
there is great desire to use bonding throughout
syntactic core. Skins may be metal or a polymer
aerospace structures. Nondestructive testing supports
matrix composite. The adhesive is usually a film
bonding by ensuring that the bonds are secure.
adhesive.

Table 9. Advantages and limitations of bonded assemblies.

Advantages Limitations

Greatly reducesneed for mechanical fasteners, and Disassembly is very difficult and typically creates damage to
thereby eliminates cost of fasteners, assembly adherends to such an extent that one or both pieces may
costs, and tooling associated with mechanically require significant repair or replacement.
fastened joint. Surface preparation (including cleaning and fitup) is critical to
Eliminates stressconcentration atmechanical creation of sound structural joint.
fasteners and creates more uniform stress There are no reliable nondestructivetestingmethods to evaluate
distribution in joint. bond strength or degree of bonding. Weak bonds or kissing
Reduces weight because fewer fasteners are used. bondscan’tbedetectedwithcurrentcommonnondestructive
Reduces potential corrosion from dissimilar test methods, however adhesive voids and porosity may be
material contact. detected by nondestructive testing.
In mechanically stiffened structures that are Adhesive materials have a finite shelf life, working life, and cure
susceptible to buckling, bonded structure can time and require special storage procedures.
improve stiffness uniformly over bonded surface Strength of cured adhesives may be reduced by moisture
which improves structure’s buckling strength. absorption, temperature extremes, and attack from chemicals
Allows assembly or repair of structure with only such as cleaning agents and paint strippers.
single side access to structure (Campbell 2004). Adhesive bondlines are susceptible to having high residual
stresses or even cracking because of differences in rates of
thermalexpansion duringthecuringprocess(Campbell 2004).

2.32 AEROSPACE NONDESTRUCTIVE TESTING

 Figure 30. Some typical adhesively bonded joints Figure 31. Some common core shapes (Hexcel 1986).
(Campbell 2004).

Bonded doubler Hexagonal core

T
Single lap

W direction
Single strap

Tapered single lap

L direction
Double lap

Over-expanded core
Double strap
T

Tapered strap
W direction

Stepped lap

Scarf
L direction

Flexible core
Double scarf
T

W direction
The most common honeycomb shape is
hexagonal core, although other patterns such as
overextended core (OX-core) and FlexCore® are also
available (Figure 31). The L direction is usually
referred to as the ribbon direction. Honeycomb core
L direction
blocks may be spliced together to obtain larger
blocks than those available from the expansion or
corrugation processes. Splicing may be performed
withdifferent density core to create acustomized
block that is better suited for inservice loads.
Splicing is normally accomplished by using a altitudes; and (4) water in cells during high
foaming adhesive that may be in either a sheet or temperatureexposure (in equatorialzones orduring
paste form. The sheet is the more prevalent form and repair), which causes gas expansion in the
is typically 1.0 to 1.5 mm (0.04 to 0.06 in.) thick honeycomb cells, resulting inskin-to-core disbonds
with no carrier fabric support. It is common that the or large node bond failures known as blown core.
adhesives be loaded with a fine aluminum powder to Moisture ingress is largely because of poor
make them more detectable by X-ray and a blowing sealing of the exposed edges of the structure;
agent to cause expansion during cure. Foaming however, the natural condensation from
adhesives are also used to bond the closure structure temperatures at the dew point may also be a factor.
(ribs and spars) to the core blocks (ASM 2001b; Water in the core is typically detected by
Campbell 2004). radiography or thermography approaches.
Honeycomb core has excellent strength and Filleting, as shown in Figure 32, has a major
stiffness compared to the foam and syntactic foam impact on the strength of the sandwich assembly.
cores. However, one of the major drawbacks of using Before bonding, the face sheets are prefit and a
honeycomb core is the inservice damage because of verification film is made to verify that appropriate
moisture intrusion. There are four prominent issues gaps between the core block and face sheets will be
with moisture in honeycomb core structure: filled byone ortwo layers offilm adhesive. Finally,
(1) corrosion caused by moisture in contact with the face sheets, film adhesive, and core blocks are
metallic core; (2) moisture absorption in the adhesive assembled in a bonding tool and then cured under
resulting in reduction of node bond adhesive elevated pressure and temperature to complete the
strength; (3) the long term effects of numerous bonded assembly.
freeze/thaw cycles as the moisture freezes at high

AEROSPACE STRUCTURES 2.33

 Figure 32. Adhesive filleting on honeycomb core (Campbell 2004).

Good fillets, One-sided

wetting cell wall Uneven High core
fillet Small fillet
fillet

Short core, Excessive

no fillet flow

Figure 33. Common honeycomb closeout configurations Closure configurations at the periphery of the
(Campbell 2004). honeycomb core box have features that require
unique nondestructive testing techniques to identify
the various discontinuities that may occur in the
Outward facing channel bonded components. Some of the more common core
closeout configurations are found in Figure 33. In
many cases, a development standard must be
constructed to verify the ability to find discontinuity
size at various locations in the assembly.
Bonded core assemblies can have anomalies
occurring in the materials before bonding, during
bonding and assembly, and in service. Some of the
Z closure common anomalies in honeycomb core bonded
assemblies are identified in Table 10.
Closed cell foam cores made with polyurethane,
polyvinyl chloride (PVC), and polymethacrylimide
(PMI) are another material used in sandwich
structure. Polyurethane and PVC can be fabricated as
either secondarily bonded or cocured structure and
have lower mechanical properties than honeycomb
Inward facing channel
material. PMI structure has good mechanical
properties relative to the other foam materials (but
still lower than honeycomb) and is used for some
high performance aerospace structures. Water
absorption is common with the foams and can lead
to skin-to-core disbonds. The primary use offoam
sandwich structure is in lightweight aircraft.
Integral Y closure Syntactic core is a core material supplied as either
a paste or film form consisting of an epoxy resin
system as the matrix, which is filled with
microballoons. The epoxy resin is typically capable
of service temperatures ranging from 121 °C (250 °F)
to 177 °C (350°F). Microballoons may be made of
Integral Z closure glass, quartz, phenolic, or carbon. Syntactic core has
a significantly higher density (480 to 1281 kg/m3, or
30 to 80 lb/ft3) than honeycomb core (128 to
640 kg/m3, or 8 to 40 lb/ft3). Lower density can be
obtained by increasing the quantity of microballoons
in the resin. In spite of the higher density, syntactic
core is agoodreplacement in secondary structure
made from thin gage honeycomb core because of the
Table 10. Typical honeycomb core anomalies. difficulty in machining thin honeycomb blocks and
Location Anomalies creating knife edges.

Core Crushed core

Voidsorlack of adhesive between coresplices Bond Certification
Lack of fillet Certification of bonded structure for airworthiness
is performed in the same manner as composite
Facesheet Porosity laminates by using the building block approach.
Delamination When composite skins and/or nonmetallic cores are
used in the assembly, the testing is normally

2.34 AEROSPACE NONDESTRUCTIVE TESTING

performed ina dry, room temperature environment. over the repair. The bond line of doublers bonded
After a baseline is established, the tests may be over repaired core can usually be inspected using
repeated for environmental conditions or a scatter pulse echo ultrasonics or resonant ultrasonic
factor and a knockdown for environmental effects methods. It is strongly recommended to create
may be applied to the original data. development standards of a pristine repair and of a
Typical repair schemes to bonded core assemblies repair with simulated anomalies during
involve excising the damaged area, then splicing in a nondestructive test technique development.
new piece of core and bonding a tapered doubler

AEROSPACE STRUCTURES 2.35

 Inservice Nondestructive Testing
PART11

The initial design objectives for the air vehicle are astronauts, so the vessel mustremainleak tight and
the primary drivers for postproduction inspections. the systems must continuously function without
Some items such as missiles and satellites are a one- failure. In contrast, the fleet of space shuttles were
time use vehicle — for instance, missiles are typically multiuse vehicles, which were thoroughly inspected
destroyed when they reach their target and satellites upon return from each mission. There is no higher
are not typically inspected in orbit nor are they visibility aerospace program in the world than the
returned to Earth forserviceand reused. International Space Station, so failures of any type
Some items such as the International Space are unacceptable. Primary structure is under close
Station (ISS) are passenger rated systems and have surveillance to ensure safe flight.
long service lives. Once in orbit, the vehicle does not All inservice aircraft have scheduled and
experience extreme flight loads; however, the entire unscheduled inspections. Some scheduled inspections
external structure is exposed to high levels of are based on events (such as number of landings or
ultraviolet radiation and constant high or low number of pressure cycles), whereas others may be
temperatures. Internally, the system must generate based on aircraft structural integrity programs.
an adequate supply of oxygen to sustain the

2.36 AEROSPACE NONDESTRUCTIVE TESTING

References
ASM. 1984. Aluminum Properties and Physical Metallurgy. Materials NTSB. 1989. Accident Report AAR 89/03, Aloha Airlines, Flight 243,
Park, OH: ASM International, for the Aluminum Association. Boeing 737-200, N73711, Near Maui, Hawaii, April 28, 1988.
ASM. 2001a. ASM Handbook 13A. Corrosion: Fundamentals, Testing Washington, DC: National Transportation Safety Board.
and Protection. Materials Park, OH, ASM International. Roylance, D. 2001. Fatigue. Cambridge, MA: Massachusetts Institute of
ASM. 2001b. ASM Handbook 21. Composites. Materials Park, OH: ASM Technology, Department of Materials Science and Engineering.
International. SAE. 2009. SAE AIR 4844, Composites and Metal Bonding Glossary,
ASNT. 1959. Nondestructive Testing Handbook, 1st edition. 2 volumes. Warrendale, PA: SAE International.
New York, NY: Ronald Press. Rpt. Columbus, OH: American Society Sarafin, T.P., and W.J. Larson. 1995. Spacecraft Structures and
for Nondestructive Testing. Mechanisms: From Concept to Launch. Torrance, CA: Microcosm
ASTM. 2007. D 3878, Standard Terminology for Composite Materials. Press.
West Conshohocken, PA: ASTM International. Schijve, J. 1994. ―Fatigue of Aircraft Materials and Structures.‖
Atkinson, R.J., W.J. Winkworth, and G.M. Norris. 1960. Reports and International Journal of Fatigue 16(1): 21-32.
Memoranda No. 3248, Behaviour of Skin Fatigue Cracks at the Schijve, J. 2007. ―Fatigue Damage in Aircraft Structures: Not Wanted,
Corners of Windows in a Comet I Fuselage. London, United But Tolerated?‖ International Conference on Damage Tolerance of
Kingdom: Ministry of Aviation, Aeronautical Research Council. Aircraft Structures [September]. Delft, Netherlands: Delft University
Campbell, F.C.2004. Manufacturing Processesfor Advanced Composites. of Technology, Faculty of Aerospace Engineering.
Kidlington, Oxford, UK: Elsevier. Tomblin, J., T. Lacy, B. Smith, S. Hooper, A. Vizzini, and S. Lee. 1999.
Cook, S.A., and T. Vanhooser. [Link] Next Giant Lap: NASA’s Ares DOT/FAA/AR-99/49, Review of Damage Tolerance for Composite
Launch Vehicles Overview. Web page. Washington, DC: National Sandwich Airframe Structures: Final Report. Washington, DC:
Aeronautics and Space Administration. United States Department of Transportation, Federal Aviation
Eastin, R.G. 2007. ―‗WFD‘ — What Is It and What‘s LOV Got To Do With Administration, Office of Aviation Research (August).
It?‖ First International Conference on Damage Tolerance of Aircraft USDOD. 1996. MIL-HBK-1530, General Guidelines for Aircraft
Structures [Delft, Netherlands]. Structural Integrity Program. Arlington, VA: United States
FAA. Training literature, untitled. No date. Washington, DC: Federal Department of Defense.
Aviation Administration. USDOD. 2002a. Handbook for Damage Tolerant Design. Washington,
FAA. 1984. Advisory Circular 20-107A. Washington, DC: Federal DC: United States Department of Defense.
Aviation Administration. USDOD. 2002b. MIL-HBK-17-F. Composite Materials Handbook.
FAA. 2006. Advisory Circular No. 120-YY. 2006 draft. Widespread Arlington, VA: United States Department of Defense.
Fatigue Damage on Metallic Structure. Washington, DC: Federal Wanhill, R.J.H. 2002. NLR-TP-2002-521, Milestone Case Histories in
Aviation Administration. Aircraft Structural Integrity. Amsterdam, Netherlands: National
FAA. 2010. Web database on Title 14 of the Code of Federal Regulations. Aerospace Laboratory.
History files of aviation regulations and circulars. Washington, DC: Wood, B.K. 2002. AIAA 2002-4324. ―Propulsion for the 21st Century —
Federal Aviation Administration. RS-68.‖ Proceedings of the 38th Joint Liquid Propulsion Conference
Fleeman, E.L. 2001. Tactical Missile Design. Reston, VA:American [Indianapolis, Indiana]. Reston, VA: American Institute of
Institute of Aeronautics and Astronautics. Aeronautics and Astronautics.
Foedinger, R., and R. Andrulonis. 2009. ―Composite Materials Handbook
— Introduction and Overview.‖ SAMPE 2009 [Chicago, IL, August].
Gallagher, J.P., F.J. Giessler, and A.P. Berens. 1984. USAF Damage
Tolerant Design Handbook: Guidelines for the Analysis and Design Bibliography
of Damage Tolerant Aircraft Structures, Revision B, Final Report. Baker, A.A., S. Dutton, and D.W. Kelly. 2004. Composite Materials for
Dayton, OH: Dayton University Research Institute. Aircraft Structures. 2nd ed. Reston, VA:American Institute of
Gallagher, J.P., C.A. Babish, and J.C. Malas. 2005. Damage Tolerant Risk Aeronautics and Astronautics.
Analysis Techniques for Evaluating the Structural Integrity of Ball, D. 2003. ―The Role of Nondestructive Testing in Aircraft Damage
Aircraft Structures. Structural Integrity in Transportation Systems Tolerance.‖ Materials Evaluation 61(7): 814-818.
Symposium: 11th International Conference on Fracture [Turin, Italy, Chun-Yung N.M. 1988. Airframe Structural Design. Los Angeles, CA:
March]. Wright Patterson Air Force Base, OH: United States Air Technical Book Company.
Force Research Laboratory, Materials and Manufacturing DeGarmo, E.P., J. Temple Black, and R.A. Kohser. 1984. Materials and
Directorate. Processes in Manufacturing, 6th ed. New York, NY: Macmillan.
Grandt, A.F., Jr. 2003. Fundamentals of Structural Integrity: Damage DeGarmo, E.P., J.T. Black, and R.A. Kohser. 2007. DeGarmo’s Materials
Tolerant Design and Nondestructive Evaluation. Hoboken, NJ: and Processes in Manufacturing, 10th ed. New York, NY: Wiley.
Wiley. Goranson, U.G. 1993. ―Damage Tolerance Facts and Fiction,‖
Hexcel. 1986. TSB 120, Mechanical Propertiesof Hexcel Honeycomb Proceedings of the 17th Symposium of the International Committee
Materials. Stamford, CT. on Aeronautical Fatigue [Stockholm, Sweden].
Ilcewicz, L. 2002. ―Key Characteristics of Composite Material Control.‖ Goranson, U.G. 2007. ―Damage Tolerance Facts and Fiction.‖ In
FAA/NASA Workshop on Composites [Chicago, IL, August]. Proceedings of the First International Conference on Damage
Lazzeri, L., and U. Mariani. 2007. ―Application of Damage Tolerant Tolerance of Aircraft Structures [Delft, Netherlands].
Principles to the Design of Helicopters.‖ First International Keyser, Carl A. Materials Science in Engineering. Columbus, OH: Charles
Conference on Damage Tolerance of Aircraft Structures. Delft, E. Merrill Publishing (1980).
Netherlands. Maxwell, R.D.J. 1973. ―Fail-Safe Philosophy: An Introduction to the
Lincoln, J.W. 1986 ―Certification of Composites for Aircraft.‖ Symposium.‖ RAE (TR 73183). Proceedings of the 7th ICAF
Proceedings of United States Air Force Conference on Aircraft Symposium.
Structural Integrity [Sacramento, CA]. MMPDS-04, Metallic Materials Properties Development and
Mattingly, J.D. 2006. Elements of Propulsion: Gas Turbines and Rockets. Standardization. 2008. Washington, DC: Federal Aviation
AIAA Education Series. Reston, VA: American Institute of Administration.
Aeronautics and Astronautics. Mohaghegh, M. 2005. ―Validation and Certification of Aircraft
Structures.‖ 46th AIAA/ASME/ASCE/AHS/ASC Structures,
Structural Dynamics and Materials Conference [Austin, TX, April].

AEROSPACE STRUCTURES 2.37

O‘Brien, K., M.B. Benoy, C. Torkington, R.B. Douglas. 1973. RAE Schmidt, H.J., B. Schmidt-Brandecker, and G. Tober. 1999. ―Design of
(TR 73183). ―The Impact of Long Service on the Fatigue of Modern Aircraft Structure and the Role of NDI.” 7th ECNDT 4(6),
Transport Aircraft.‖ Proceedings of the 7th ICAF Symposium. Proceedings Online [Copenhagen, Denmark, May 2008]. Kirchwald,
Preston, E. 2011 accessed. The Federal Aviation Administration and Its Germany: [Link].
Predecessor Agencies. Web page. Washington, DC: Federal Aviation Walker, E.K., J.C. Ekvall, and J.E. Rhodes. 1979. ―Design for Continuing
Administration, United States Centennial of Flight Commission. Structural Integrity.‖ Structural Integrity Technology. New York, NY:
Razi, H., and S. Ward. 1996. Principles for Achieving Damage Tolerant American Society of Mechanical Engineers.
Primary Composite Aircraft Structures. 11th DoD/FAA/NASA
Conference on Fibrous Composites in Structural Design [Fort Worth,
TX, August].

2.38 AEROSPACE NONDESTRUCTIVE TESTING

 X
3
CHAPTER
CHAPTER

Reliability
Contents
Part 1. Reliability of
Nondestructive Testing, 3.2
Part 2. Probability of Detection
and Other Performance
Measures in Practice, 3.5
Part 3. Mistakes in Estimation
of Nondestructive Test
Reliability, 3.7
Part 4. Human Factors and
Nondestructive Testing, 3.8
Part 5. Model Assisted
Probability of Detection, 3.11
References, 3.12

Contributors
David S. Forsyth
Lisa Brasche

3.1
PART1 Reliability of Nondestructive Testing

Because nondestructive testing (NDT) is used in size, delaminations, and other measurable anomaly
critical roles for process control and the inspection dimensions.
of safety critical physical systems such as aircraft, The exact definition of probability of detection
pressure vessels, nuclear reactor components, and and the statistical methods used to calculate it have
pipelines, the measurement of the performance of evolved over time. The first probability of detection
nondestructive testing has become very important. It curves were constructed using moving averages or
is no longer sufficient in many cases to simply by averaging the response of all cracks in an interval
assume that an inspection is a perfect process of and by manually fitting a curve through these points
unbounded capability, rather, it is imperative to (Pettit 1972; Anderson 1973; Rummel 1973). The
know the probability of finding (or of missing) probability of detection at each point on the moving
discontinuities of interest. This measure of reliability average, or for each interval, was used to calculate
is usually called the probability of detection (POD). confidence intervals.
An example of a probability of detection curve is Early work was sponsored by the National
shown in Figure 1 where the reliability, shown on Aeronautics and Space Administration (NASA), and
the vertical axis, decreases as the crack size becomes the United States Air Force (USAF) was moving
smaller. While probability of detection is often along a similar path at the same time. In both cases,
discussed in terms of cracks as in Figure 1, it is the motivation was damage tolerance philosophies in
important to note that the probability of detection design and maintenance: it would be assumed that
approach is not limited to cracks and has in fact parts contained discontinuities when they left
been applied to other discontinuities such as manufacturing and that these discontinuities would
corrosion loss, impact damage, inclusion size, void grow as cracks under the expected operation of the

Figure 1. Probability of detection curve from transverse ultrasonic testing of welds for longitudinal cracks
in flush grounded, gas tungsten arc welds made of Unified Numbering System A92219, heat treatable,
wrought aluminum alloy. Three operators detected 291 indications in 345 opportunities for combined
probability of detection of 90 percent at 7.54 mm (0.030 in.) (NTIAC 1997).

100

90

80
Probability of detection (percent)

70

60

50

40

30

20

10

0
0 1.27 2.54 3.81 5.08
(0.05) (0.10) (0.15) (0.20)
Actual crack depth mm (in.)
Legend
= probability of detection
= hit/miss data

3.2 AEROSPACE NONDESTRUCTIVE TESTING

part. The other technology advance that enabled this
approach was the maturing of fracture mechanics Figure 2. Damage tolerance approach to structural integrity used by
approaches that could be used to model crack National Aeronautics and Space Administration and United States Air
growth in metals. Force (ASNT 2007, 531).
Later NASA adopted a damage tolerance Unstable growth
approach for the Space Shuttle Program. The USAF until failure
adopted fracture mechanics and damage tolerance in
response to structural failures on the F-111 and
C-5 aircraft that were virtually new (Haviland 1973). acrit
The USAF released MIL-STD-1530 in 1972 to revise
their Aircraft Structural Integrity Program (ASIP)

Discontinuity size
to include damage tolerance (USDOD 2010).
MIL-A-83444 was released in 1974 detailing the
damage tolerance requirements for airplane safety
offlight structure (USDOD 1987; USAF 1999). The
essence of these approaches is illustrated in Figure 2.
Assuming a severe initial cracklike discontinuity aNDT
(sometimes called the rogue flaw), with length Inspection
denoted by a0, using fracture mechanics principles, interval
a0
the time for this to grow a crack to failure
(length acrit) can be estimated. Inspections are
scheduled to have one or more opportunities in this
time to detect the crack and repair/replace the part Usage or time (relative scale)
before failure. Legend
The approach to damage tolerance illustrated in acrit = discontinuity of critical size
Figure 2 defines a crack size detection threshold aNDT aNDT = discontinuity at moment of test
above which all cracks would be found. Given a a0 = discontinuity before propagation
probability of detection curve, the simple answer for
picking this size for nondestructive test capability
would be at the point where probability of detection
is 100 percent. However, the binomial statistics in 3. The 90/95 criterion was consistent with the
the early efforts required very large numbers of trials USAF MIL-HDBK-5 B basis materials allowables,
to obtain high values of probability of detection and and thus familiar to the people working with the
high statistical confidence levels. At some time in damage tolerance approaches.
the mid 1970s, the lower limit thresholds were set at A published standard or guideline on how to
90 percent probability of detection and 95 percent perform an experiment to estimate a probability of
confidence level. This is often referred to as a90/95 or detection curve was not available immediately.
simply the ―90/95‖ value. The acceptance of the Within the American Society for Nondestructive
90/95 criteria as a threshold for inspection capability Testing (ASNT), an effort was initiated by W.H. Lewis
seems to have been due to a confluence of factors. to develop a recommended practice document.
1. A threshold was needed for deterministic Although the final version was finished in 1976,
fracture mechanics approaches. The USAF and publication did not occur until 1982 (Rummel 1982).
NASA approaches to damage tolerance were This recommended practice is based on binomial
such that a crack growth analysis was started methods. In the meantime, the American Society for
from a ―rogue flaw‖ and the time to crack Materials‘ ASM Handbook 11: Nondestructive
instability was calculated from this starting Inspection and Quality Control, 8th edition (1976)
point. Inspections are used to detect the crack included a section from Packman describing
before instability or critical crack growth to binomial based methods for probability of detection
failure. To reduce risk 100 percent probability of estimation (Packman 1976). It is noted that the ASM
detection at crack sizes greater than the ―rogue document was developed with the ASNT authors.
flaw‖ size is preferred. Because 100 percent The most common method of estimating
probability of detection is not likely at any size probability of detection curves from inspection data
of crack, a lesser threshold had to be chosen. today is based on USAF MIL-HDBK-1823, first
2. Binomial statistics were in use to calculate released in 1992 as a NATO document. The Federal
probability of detection and confidence bounds. Aviation Administration (FAA) has also published a
Using binomial statistics to demonstrate high descriptive series of documents on how to conduct a
probability of detection numbers requires many probability of detection study, based on the
tests. Twenty-nine successes out of 29 trials are statistical methods described in MIL-HDBK-1823
required for a 95 percent confidence. Many (Spencer 1993a; Spencer 1993b; Spencer 1995).
more trials are required to increase the These statistical methods were developed by Berens
confidence level. and Hovey at the University of Dayton in the 1980s

3.3
 (Berens 1981, Berens 1983). There are two key
Figure 3. Illustration of log linear relationship observed between observations underpinning the current approach to
nondestructive testing signal magnitude and crack size (ASNT 2007, probability of detection:
532).
1. Different cracks of the same size will not be
detected with the same probability of detection.
2. In many cases, the relationship between the
Shadedareaindicatesproportion signal from the nondestructive test system and
detected at that crack size = POD the size of the crack are linearly related if
plotted on a logarithmic chart.
Log (NDT system signal)

These ideas are illustrated in Figure 3, which

instigates the log normal equation to fit the
Hit relationship of probability of detection to crack size
Decision as documented in MIL-HDBK-1823.
threshold
Miss

Log (discontinuity size)

3.4 AEROSPACE NONDESTRUCTIVE TESTING

 Probability of Detection and Other PART2
Performance Measures in Practice

Estimation of Probability of replicate the service discontinuities as closely as

possible or to use engineering judgment whether
Detection safety factors are needed to account for the
Both the USAF and the Federal Aviation Agency difference between the probability of detection
(FAA) have published guidelines that describe in experiment and inservice conditions.
detail the experiments required to estimate the
probability of detection of an inspection system.
These documents are in the public domain, and can False Call Rates
be obtained for free from the respective government The number of false positives or false calls made
agencies as well as the Department of Defense‘s during inspections is very important in practical
Advanced Materials, Manufacturing, and Testing applications due to the cost implication and the trust
Information Analysis Center (AMMTIAC). The in the nondestructive test procedure. The detection
USAF MIL-HDBK-1823 was updated in 2009 threshold may always be lowered to capture smaller
(USDOD 2009). NASAhasalsopublished guidelines discontinuities, but in general lowering it increases
on nondestructive testing and its capability the likelihood of false calls. The appropriate level
assessment including NASA-STD-5009, will depend on specific situations; key factors to
NASA-STD-5013, and NASA-HDBK-5014 consider include the cost of false calls versus the
(NASA 2008; NASA 2009; NASA 2009 draft). cost of missing discontinuities.
Requirements for other industries appear in their The tradeoff between false calls and the detection
respective codes, such as the ASME Boiler and of discontinuities of a single size can be visualized
Pressure Vessel Code. The reference to be used in a using the receiver operating characteristic (ROC)
particular application should be based on the curve, described in the next section.
regulating body that requires this assessment.
The process of probability of detection estimation
requires a number of inspections to be performed
within the following general guidelines. Receiver Operating
The complete, predefined inspection system that
is being assessed should be used including
Characteristic
representative equipment, procedures, inspectors, The receiver operating characteristic (ROC) is a
and target parts. measure of the separation of the noise signals in a
The study is performed using parts with system from the signals due to the item of interest
discontinuities representative of the actual inspection (for example, cracks). This measure originated in the
targeted by the study. Machined notches or flat evaluation of radar signals in World War II. It shows
bottom holes can provide a useful measure of how the decision threshold affects the tradeoff
capability, but should not be assumed to be between false calls and detection of discontinuities
representative of cracks or other natural at a single size. This is illustrated in Figure 4 where
discontinuities. Means to assess the difference or a the distance between the distributions is a measure
transfer function is evaluated to compensate the
difference and evaluate the probability of detection
for nonartificial discontinuities. Figure 4. Illustrations of signal distribution due to noise and
The study is performed using an inspection discontinuities: (a) idealized; (b) realistic (ASNT 2007, 533).
procedure and environment typical of the deployed
environment for the actual inspection. Human (a) (b)
factors studies have shown that the relationship
of factors such as environment (lighting and Signal from
discontinuities of
temperature), training, experience, motivation, and Noise particular size Noise Signal
Probability
Probability

others are not simple and often not intuitive.

It is crucial to understand the physical parameters
and the interaction between the applied
nondestructive test system and the discontinuity to
be able to execute a representative probability of
detection estimate. Although rarely the case, parts Amplitude Amplitude
from service with discontinuities arising from use are
preferred for a valid probability of detection. Ideal case: Realistic case:
However, in most cases, this is not possible. perfect detection and higher detection rate also gives
zero false calls. higher false callrate.
Therefore every reasonable effort should be made to

3.5
 Figure 5. Various receiver operating characteristic (ROC) curves for nondestructive tests of different
capability: (a) best and worst case; (b) realistic case (ASNT 2007, 533).
(a) (b)

100 100
Heavy line is best case

Fraction of correct detections

Fraction of correct detections

Realistic case

0 Dashed line is 0
0 worst case 100 0 100
Fraction of false calls Fraction of false calls

of the capability of the inspection, and can be A set of distribution curves is shown in Figure 4.
represented in a receiver operating characteristic The perfect curve is one in which the noise signal
curve. and discontinuity signal distributions do not
To plot the receiver operating characteristic curve overlap (Figure 4a). This makes a curve as shown
requires making measurements of the signal in Figure 5a. A more realistic curve is shown in
distributions from noise and from discontinuities, as Figure 5b, corresponding to the situation shown
shown in Figure 4. The receiver operating in Figure 4b. The worst possible receiver operating
characteristic curve is then plotted by changing the characteristic curve occurs when the noise and
decision threshold and plotting the fraction of discontinuity signal distributions are identical and
detected discontinuities as a function of the fraction completely overlap: the inspection cannot
of false calls. Starting with a decision threshold at distinguish between the two. This is a line at a
positive infinity, one would find zero discontinuities 45 degree angle on the plot, as illustrated in
and have zero false calls. As the decision threshold Figure 5a. The plot applies to a single discontinuity
goes lower, the detection rate goes up with few false size and provides complementary information to the
calls. As the threshold is lowered, the detection rate probability of detection curve about the performance
will increase slowly and the false call rate will capability of the testing equipment.
increase rapidly.

3.6 AEROSPACE NONDESTRUCTIVE TESTING

 Mistakes in Estimation of PART3
Nondestructive Test Reliability

The methods used to assess nondestructive testing required to do a study as described in the USAF
reliability have evolved significantly over the past MIL-HDBK-1823 may be prohibitive, and the
40 years, and the statistical basis of the current temptation may be strong to apply data from
methods is often at a level beyond the statistical previous studies. However, this can only be done
knowledge of most nondestructive testing with careful engineering judgment and an
practitioners. There are a few common errors in the understanding of all the factors that affect
estimation of nondestructive test reliability worth nondestructive testing capability in a particular
mentioning to help the reader avoid them: situation. Residual stresses are only one example
1. Often the customer for nondestructive test of factors that are often undocumented but can
reliability information is the design engineer have an enormous impact on the detectability of
who may have very little knowledge of cracks.
nondestructive testing. It is crucial to know 3. Another common mistake is to assume that
what the end use of the information will be in repeated inspections will increase probability of
order to provide the best estimate of reliability. detection. Recognizing the statistical nature of
Often designers will ask for ―the smallest flaw the inspection process, many people have then
that can be found.‖ The nondestructive test assumed that inspections are ―independent‖ and
practitioner is cautioned to avoid simply therefore repeating the inspections will improve
replying with the smallest discontinuity ever the probability of detection. It must be
found. Rather, the question needs to be clarified understood that the concept of statistical
so that the designer understands the statistical independence applies to random events like a
nature of nondestructive testing. This will help coin toss, and an inspection is not a random
the designers meet their needs and provide the event.
inspector with reasonable expectations. 4. Finally, proper technique documentation,
2. The most common mistake in nondestructive calibration procedures, and training are essential
testing reliability is the use of preexisting to ensure that the carefully controlled data
probability of detection information beyond its acquired in a probability of detection study can
scope of applicability. The cost and time be reproduced in the actual inspection
environment.

RELIABILITY 3.7
 Human Factors and Nondestructive
PART4 Testing

An individual, working within the operational

parameters of the inspection process, can have a
Inspection Guidance
huge impact on the reliability of any given Inspectors can get their directions for performing
inspection. The major human related factors that an inspection in many ways, but mostly they get
might degrade the inspection process fall into the them in the form of written instructions or
following general categories: traditional human procedures. These instructions are usually disbursed
behavior characteristics, guidance related to the using one set of guidelines, assuming all parts are
inspection, preparation for conducting the inspection, generally the same. They are often created and
physical environment where the inspection is taking issued by persons who are not production inspectors,
place, equipment suitability (user-to-machine but who may be nondestructive testing engineers
interface), management‘s understanding and based in a location well removed from the usual
appreciation of the inspection, and ability to remain inspection setting. Thus it is necessary to verify if
vigilant over long periods of inactivity. When all of the procedure can realistically be used by individuals
these factors are positive, one can expect relatively who make up the inspector corps. The instructions
high levels of performance as exhibited by high should be of adequate detail to leave very little to
detection probabilities. As the nature of these chance on the part of the inspector. The more
categories tends toward less supportive conditions, assumptions left in the written instructions, the
the performance reliability can deteriorate to a higher the likelihood that the inspector will
completely unacceptable level in some cases. The misinterpret the instructions in unflagged areas, will
challenge for those involved in managing use unauthorized approaches, and will produce
nondestructive test operations is to seek to maximize questionable interpretations. Thus, it is quite possible
satisfaction of all of these human factors‘ needs if that an ―inspection‖ error on the job may in fact be a
high levels of inspection reliability are to be reached procedural ―instruction‖ error not caused by the
and sustained. inspector at all, but still the result of a human
factor‘s error.

Traditional Human Behavior

We all have personality attributes that may lend
Inspector Preparedness
themselves to the necessary characteristics for an Inspectors need to be suitably trained in order to
effective inspector. Included among these are develop a command of the tasks they are required to
personal integrity, patience, ability to concentrate perform. The training should convey a thorough
(ignore distractions), tolerance of exterior conditions, understanding of the basics of the inspection process
and manual dexterity. A person equipped with these but also an intimate familiarity with the manner by
favorable characteristics generally has the which the physical aspects of the inspection must be
disposition best suited for working in the inspector carried out. This hands-on aspect of the training is
role. However, that is not enough. Successful people vital to their success. In most situations, it is the
are motivated to achieve the best that they can hands-on familiarity with the process that carries the
regardless of the operational conditions. They set day in the real world of inspections. In those cases
their personal goals within the context of the where an unfamiliar test item or process is involved,
organizations for which they work, they are specific supplemental training is required to assure
respected for what they do and what they know, and that the inspector is well aware of any subtle
their compensations are considered fair for their line differences that may exist between their routine
of work. They must be confident in what they are experiences and that of the newly introduced
doing in order to stand up to others who may inspection scenario.
challenge their findings and conclusions. Because of
the physical aspects of many inspections, inspectors
need to be in good physical condition and mentally Inspector’s Environment
alert.
The physical environment such as temperature,
In general, it is difficult to find individuals with
humidity, noise, lighting, exposure to hazards, and
all of these attributes. When errors are made during
encumbrance by protective gear can all affect the
inspections, it is easy to assign blame to the
performance of an inspector. However, studies have
inspector and ignore other factors that may have
shown that with the proper motivation, superior
contributed to errors. The following sections discuss inspectors will continue to be able to conduct
factors that lead to degradation of inspection effective inspections, contrary to the expectations of
performance.

3.8 AEROSPACE NONDESTRUCTIVE TESTING

casual observers of the process. This is attributed to
their ability to focus on the task at hand, and to not
Inspector Vigilance
be easily distracted by such environmental factors. Numerous studies have indicated that when a
Fatigue, on the other hand, can be accelerated by person spends long periods of time with no positive
these factors and must be carefully managed. Tactics feedback that their efforts are still functioning as
used to counteract the effects of fatigue include expected, they become complacent and less engaged
limiting the numbers of hours inspectors must (or are with the inspection process. If there is little
allowed to) remain on the job during a given shift. expectation for something to be found, then the
Similarly, working in awkward or restricted positions inspector tends to be less vigorous in their attempts
aggravates the onset of fatigue and must be limited to find objectionable conditions. This reduces the
if the inspections are to be considered valid. likelihood of detecting anomalies and the integrity of
Besides the physical environmental issues, the entire process can be severely degraded.
interpersonal issues can affect the mental state in
which a person is conducting inspection tasks. These
issues tend to fall into the categories of problems at
home and problems in the workplace. The personal
What Can Be Done?
What can be done about all the human factors
problems resulting from instability of the home
that degrade inspections? One popular approach to
setting can carry over to the work place and, in the
reduce human factors in inspection is to remove the
extreme, can intrude on the mental processes
human from the inspection process as much as
necessary to conduct a proper inspection. The
possible. In the extreme, total automatic inspection
personal problems resulting from instability within
is used for particularly critical components that have
the workplace can have the same effect, but these
a high degree of consistency from one item to the
conditions tend to be longer ranging and involve
next. For example, jet engine disks call for the
misunderstandings between coworkers that can
detection of extremely small surface cracks that can
persist for extended periods of time.
be found using eddy current techniques. In order to
assure consistent and thorough eddy current
inspection, the inspections of the highly similar
Appropriate Equipment surfaces and holes found in such parts have been
In settings involving humans interacting with made largely automatic using automatic robotic
different types of equipment (computers, scientific inspection stations such as the so called retirement
components, and mechanical sensors), particularly for cause systems used by some engine maintenance
when decisions must be made through visual facilities. For large area inspections, portable
recognition of subtle changes in display conditions scanning systems equipped with nondestructive
while manipulating some type of probe or sensor, testing sensors can be used that will assure
inspection performance can be greatly enhanced if consistent coverage of a given area. Some emerging
the overall inspection system is designed with the techniques yield direct images of internal conditions
inspector in mind. Improved approaches to the that make the spatial relations of geometrical
user-to-machine interface in systems such as those features of a test article quite clear, discontinuities
found with cockpit controls and nondestructive test being easily recognized as deviations from the
instruments promise to aid in reducing fatigue and normal image conditions. Unfortunately, most
thus enhancing an operator‘s ability to remain alert components being inspected nondestructively are
over extended periods of time with undiminished somewhat unique in their fabrication and assembly
inspection reliability. and thus are not ideal candidates for inspection by
such expensive automated inspection approaches.
A simpler but expensive approach to reducing
human factors is to use various types of dedicated
Management’s Involvement tools that assist the inspector in consistently
Attitudes of senior management can impact the maintaining the relative position of a sensor with
effectiveness of inspection operations because the respect to an inspection zone of interest. In some
workforce looks to management to set the overall cases, manual scanning is being assisted through
tone for the organization. If in those cases where templates and probe designs that are made
nondestructive testing leads to the successful specifically to maintain a fixed position with respect
detection of unacceptable conditions and if to test item features such as edges and other known
management is pleased that the inspection process is inspection zones of interest. Thus, in those cases
working, then the inspector corps will strive to where exact location of a probe must be maintained
repeat the experience. On the other hand, when in order to keep the inspection results consistent,
management is displeased that production levels fall there are simple devices available to remove some of
off as a result of such findings, then the inspector the uncertainty of the inspection process.
will tend to be much more cautious when The electronics based instruments being used for
questionable calls must be made. Such caution tends nondestructive testing have been greatly improved in
to increase the likelihood that significant their ability to display comparative information and
discontinuities will be missed. to store information needed for repeated calibrations
and reports preparation. Such devices have relieved
some of the administrative aspects of the inspection
process, thereby freeing up more time for actual
inspections and thus represent an increase in
inspector productivity. But they do not necessarily
improve inspection reliability. In fact, some of the

3.9
 added complexity of such instruments tends to be inspections, as a general practice, the directional
confusing to the average inspector. Instrument materials issued to inspectors should include a
storage of critical factors, such as calibrations, can minimum of uncertainties. Any modifications to the
also be a source of confusion. inspection process should also be validated against
The self confidence and pride of achievement the original test conditions to ensure that the
needed to keep inspectors at the top of their game suggested approach is providing the needed
comes from positive feedback from successfully capability.
detecting and characterizing objectionable When selecting candidates for the position of
conditions in test articles, be they routine production inspector, special care needs to be taken to use
items or proficiency test articles intended to measure individuals who have a temperament that supports
the effectiveness of the inspector corps. Such the needs of the inspection tasks. Patient, focused,
successes need to be recognized by management physically fit, with a curiosity for the unknown and
when they occur, and must not be perceived as those who are thorough when attacking problems
costly intrusions into a profitable production process. tend to make the best inspectors. They must be able
One way to achieve ongoing proficiency is through to resist external pressures and still be confident in
frequent exercises that refresh understanding of the their decisions regarding the presence of possible
inspection settings such as refresher classes and objectionable conditions within items being
competitive exams that test the intellectual and examined.
physical capabilities of each inspector. Where There is little that can be done about the
deficiencies are found, it becomes an opportunity to emotional distress an inspector may be working
upgrade to a more capable inspector or to reassign under. Colleagues and managers should always be on
an individual to a more fitting line of responsibilities. the lookout for signs of distress and to support
To assure the preparedness of inspectors the affected individuals in any way possible. Assistance
instructions used in the course of maintaining may be in the form of time off, counseling,
component integrity must be thoroughly prepared supplemental financial support, or any other
and validated before they are made the basis for assistance that may be available to them. This factor
inspections. Although it is often helpful to take is considered a commonly hidden danger that can
advantage of the ingenuity and skill of an inspector unexpectedly affect the effectiveness of an inspector
to overcome unexpected conditions during difficult if not recognized in a timely manner.

3.10 AEROSPACE NONDESTRUCTIVE TESTING

 Model Assisted Probability of
PART5
Detection

The previous parts of this chapter have primarily

focused on empirically based probability of detection
Summary
approaches that require sufficient samples and The measurement of the performance of
associated discontinuities to provide a statistically nondestructive testing is important for maintaining
valid curve. However, in even the simplestinspection safety, for scheduling inspections and other
situations, generating a valid sample set can be a maintenance actions, and for the selection of the
very costly process, often exceeding the optimum nondestructive testing technique for a
$100 000 mark. Creating samples with actual fatigue particular application. The most common method
cracks of the desired size relevant to the probability for assessing nondestructive testing performance
of detection study can sometimes be impossible. is the probability of detection curve. The USAF
Parts with actual cracks can be used. However, when MIL-HDBK-1823 providesdetailedinformation on
the part being tested is ―one of a kind,‖ having a set the experiments required to estimate the probability
of test articles with naturally occurring anomalies of detection for a specific inspection technique and
may not be feasible. In the early 2000s, the Air application. Human factors play a key role in the
Force, Federal Aviation Administration, and NASA reliability of nondestructive testing. High levels of
joined together to foster the development of performance, as exhibited by high detection
supplemental approaches to probability of detection probabilities, require vigilance in maintaining
that take advantage of physics based models of the inspector awareness to procedures and details.
inspection process. The Model Assisted Probability of
Detection (MAPOD) Working Group was established
with the objective to promote the increased
understanding, development, and implementation of
the methodologies. A website was established to
provide a repository of the efforts of the working
group and provide electronic access to prior
empirical probability of detection studies. The most
recent update to MIL-HDBK 1823 includes a chapter
describing the model assisted approach.

RELIABILITY 3.11
References
Anderson, R.T., T.J. DeLacy, and R.C. Stewart. 1973. NASA CR-128946,
Detection of Fatigue Cracks by Nondestructive Testing Methods. Bibliography
Washington, DC: National Aeronautics and Space Administration. ASME. N.d. ASME Boiler and Pressure Vessel Code. New York, NY:
Berens, A.P., and P.W. Hovey. 1981. Evaluation of NDE Reliability American Society of Mechanical Engineers.
Characterization. USAF Report No. AFWAL-TR-81-4160. Arlington, Berens, A.P. 1988. ―NDE Reliability Data Analysis.‖ In Metals Handbook
VA: United States Department of Defense. 17: Nondestructive Evaluation and Quality Control. 9th ed.
Berens, A.P., and P.W. Hovey. 1983. "Statistical Methods for Estimating Materials Park, OH: ASM International, 659-701.
Crack Detection Probabilities." In STP 798, Probabilistic Fracture Fahr, A., D.S. Forsyth, and M. Bullock. 1993. Report
Mechanics and Fatigue Methods: Applications for Structural Design LTR-SMPL 1993-0102. A Comparison of Probability of Detection
and Maintenance. West Conshohocken, PA: ASTM International, Data Determined Using Different Statistical Methods. Ottawa,
79-94. Ontario, Canada: National Research Council Canada.
Haviland, G.P., and C. Tiffany. 1973. AIAA Paper No. 73-18. ―The USAF Fahr, A., D.S. Forsyth, M. Bullock, and W. Wallace. 1994.
Aircraft Structural Integrity Program (ASIP).‖ In Proceedings of the NRCC-LTR-ST-1961. NDT Techniques for Damage Tolerance-Based
American Institute of Aeronautics and Astronautics 9th Annual Life Prediction of Aero-Engine Turbine Disks. Ottawa, Ontario,
Meeting and Technical Display [Washington, DC, January 1973]. Canada: National Research Council Canada.
Reston, VA: American Institute of Aeronautics and Astronautics. Forsyth, D.S., A.P Berens, and W.D. Rummel. 2006. ―Nondestructive
NTIAC. 1997. NTIAC DB-97-02. Rummel, W.D., and G.A. Matzkanin. Inspection Reliability and Risk in the Field and Depot.‖ In
Nondestructive Evaluation (NDE) Capabilities Data Book. 3rd ed. Proceedings of the USAF Aircraft Structural Integrity Program
Austin, TX: Nondestructive Testing Information and Analysis (ASIP) [San Antonio, TX, November 2006]. Dayton, OH: Universal
Center. Technology Corporation, for the United States Air Force Research
Packman, P.F., H.S. Pearson, J.S. Owens, and G.B. Marchese. 1968. Laboratory (2006).
AFML-TR-68-32. The Applicability of a Fracture Mechanics — Hyatt, R.W., G.E. Kechter, and R.G. Menton. 1991. ―Probability of
Nondestructive Testing Design Criterion. Detection Estimation for Data Sets with Rogue Points.‖ Materials
Packman, P.F., S.J. Klima, R.L. Davies, J. Malpani, J. Moyzis, W. Walker, Evaluation 49(11): 1402-1408.
B.G.W. Yee, and D.P. Johnson. 1976. ―Reliability of Flaw Detection Lewis, W.H., B.D. Dodd, W.H. Sproat, and J.M. Hamilton. 1978. USAF SA-
by Nondestructive Inspection.‖ Metals Handbook 11: Nondestructive ALC/MEE 76-6-38-1. Reliability of Nondestructive Inspections —
Inspection and Quality Control, 8th edition. Materials Park, OH: Final Report (Have Cracks, Will Travel). Arlington, VA: United
ASM International, 414-424. States Department of Defense.
Pettit, D.E., and D.W. Hoeppner. 1972. NASA CR NAS 9-11722 NASA. 2008. NASA-STD-5009, Nondestructive Evaluation Requirements
LR 25387. Fatigue Flaw Growth and NDT Evaluation for Preventing for Fracture Critical Metallic Components. Washington, DC:
Through Cracks in Spacecraft Tankage Structures. Washington, DC: National Aeronautics and Space Administration.
National Aeronautics and Space Administration. NASA. 2009. NASA-STD-5013, Pyrovalve Applications and Performance
Rummel, W.D. 1982. "Recommended Practice for Demonstration of Handbook. Washington, DC: National Aeronautics and Space
Nondestructive Evaluation (NDE) Reliability on Aircraft Production Administration.
Parts." Materials Evaluation 40 (August): 922-932. NASA. 2009 draft. NASA-HDBK-5014, Nondestructive Evaluation (NDE)
Rummel, W.D., P.H. Todd Jr., and W.L. Castner. 1973. ―Detection of Implementation Handbook for Fracture Control Programs.
Fatigue Cracks by Nondestructive Evaluation Methods.‖ Spring Washington, DC: National Aeronautics and Space Administration.
Conference [Los Angeles, CA, March 1973]. Columbus, OH: Rummel, W.D., P.H. Todd Jr., S.A. Frecska, and R.A. Rathke. 1974. NASA
American Society for Nondestructive Testing. CR-2369. The Detection of Fatigue Cracks by Nondestructive Testing
Spencer, F., G. Borgonovi, D. Roach, D. Schurman, and R. Smith. 1993a. Methods. Washington, DC: National Aeronautics and Space
DOT/FAA/CT-92/12.1. Reliability Assessment at Airline Inspection Administration.
Facilities 1: A Generic Protocol for Inspection Reliability Rummel, W.D., R.A. Rathke, P.H. Todd Jr., and S.J. Mullen. 1975.
Experiments. Washington, DC: United States Department of MCR-75-212, NASA-CR-144639. The Detection of Tightly Closed
Transportation. Flaws by Nondestructive Testing Methods. Washington, DC:
Spencer, F., G. Borgonovi, D. Roach, D. Schurman, and R. Smith. 1993b. National Aeronautics and Space Administration.
DOT/FAA/CT-92/12, 2. Reliability Assessment at Airline Inspection Spencer, F.W. 1998. ―Identifying Sources of Variation for Reliability
Facilities 2: Protocol for an Eddy Current Inspection Reliability Analysis of Field Inspections.‖ RTO-MP-10, AC/323 (AVT) TP/2.
Experiment. Washington, DC: United States Department of RTO Meetings Proceedings 10: Airframe Inspection Reliability under
Transportation. Field/Depot Conditions.
Spencer, F.W., and D. Schurman. 1995. DOT/FAA/CT-92/12,3. Reliability USAF. 2006. MIL-HDBK-5 B, Metallic Materials and Elements for
Assessment at Airline Inspection Facilities 3: Results of an Eddy Aerospace Vehicle Structures. Wright Patterson Air Force Base, OH:
Current Inspection Reliability Experiment. Washington, DC: United Aeronautical Systems Center.
States Department of Transportation. Yee, B.G.W., F.H. Chang, J.C. Couchman, and G.H. Lemon. 1975. NASA
USAF. 1999. AFGS 87221, Aircraft Structures, General Specification for. CR-134991. Assessment of NDE Reliability Data. Washington, DC:
Wright Patterson Air Force Base, OH: Aeronautical Systems Center. National Aeronautics and Space Administration.
USDOD. 2009. MIL-HDBK-1823, Nondestructive Evaluation System
Reliability Assessment. Arlington, VA: United States Department of
Defense.
USDOD. 2010. MIL-STD-1530, Aircraft Structural Integrity Program.
Arlington, VA: United States Department of Defense.
USDOD. 1987. MIL-A-83444, Airplane Damage Tolerance Requirements.
Superseded by USAF AFGS 87221. Arlington, VA: United States
Department of Defense.

3.12 AEROSPACE NONDESTRUCTIVE TESTING

 X
4
CHAPTER
CHAPTER

Cross Reference
Tables for
Test
Metho
d
Selection
Contents
Part 1. Introduction, 4.2
Part 2. Materials, Structures,
and Processes Tables, 4.3
Part 3. Method Tables, 4.10
Bibliography, 4.20

4.1
 Introduction
PART1

This chapter contains reference tables for both the subtleties that can affect some tests, nor do they
material applications and the methods of include all variations of technologies that could be
nondestructive testing (NDT) that may be applied in applied in some circumstances. Nevertheless, they
industry. The intent of the tables is to provide the provide valuable input such that the reader can find
user with a starting point for selecting and applying relevant approaches to solve problems, appreciate
technology to solve nondestructive testing issues. why certain techniques are used, or serve as a
Tables 1 through 15 are cross references between starting point for further technique discoveries.
the material applications and nondestructive test Tables 16 through 26 are cross references between
methods. The reader can select the table for the type nondestructive test methods and material tests. These
of material of interest and find inspection issues, tables represent the main chapters of this handbook
applicable nondestructive test techniques, comments, for common nondestructive test methods. The tables
and references to other chapters in this handbook. list the principle of the method, the typical
The comments suggest the most commonly applied application, its advantages, and limitations. Again,
approaches for testing and the primary issues that these tables are not exhaustive, but are intended to
the applicable nondestructive test techniques address provide basic information on the physics of the
for the materials of interest. The tables are not methods and where the methods are commonly
exhaustive, nor do they contain all the details and applied in the aerospace industry.

4.2 AEROSPACE NONDESTRUCTIVE TESTING

 Materials, Structure, and PART2
Processes Tables

Tables 1 through 15 contain examples of the product fabrication, assembly, or during the service
structures and materials that are common in the life. The tables provide insight into common testing
aerospace industry. Nondestructive testing for issues and the nondestructive test techniques that
various types of conditions is required during can be applied.

Table 1. Fiber reinforced polymer (FRP) composite laminate: graphite fiber in epoxy polymer; glass/epoxy composite.

Inspection Nondestructive Comments on Method

Issues Test Techniques Nondestructive Testing Chapters
Consolidation, Graphite fiber/epoxy
porosity,
fiber-to-resin ratio, Visual testing Visually inspect surface for dry plies, wrinkles, and other 5
inclusions, surface damage.
delaminations,
wrinkles, Eddy current testing Eddycurrent testing is notcommonly applied to composites 8
surface layers, but does have sensitivity to fiber and matrix variations.
measurement of
curved surface radii Ultrasonic testing Through-transmission or pulse echo ultrasonic testing are 9
noodles (through-transmission, primarytechniques for compositelaminate materialsusing
pulse echo, automated, hand automated scanning with water coupled piezoelectric
held, laser, air coupled) transducers and acceptance standards.
Manyvariations ofultrasonic testing are applicable depending 9
on material, configuration, thickness, and sensitivity
requirements. Techniques include hand held, resonance, laser,
air coupled, lamb waves, spectroscopy, roller probes, and
others.

Radiologic testing Radiologic testing can be used for engineering evaluation for 10
cracking, density variations, and tracer yarn detection for ply
orientation.

Radioscopic ormicrofocus radiologic testing for discontinuities 10

in small components.
X-ray computed tomography is used for complex structure 10
configurations for assembly, density, and delaminations. Micro
computed tomography can measure porosity in small samples.

Thermography Thermography is sensitive to delaminations and inclusions. It is 11

applicable to thin (<2.5 mm, or <0.1 in.) materials.

Shearography Shearography isused for impact damage, wrinkles, 12

delaminations, fiber bridging.

Acoustic emission testing Acoustic emission testing is used for pressure vessel testing 13
and structural health monitoring.

Microwave testing Microwave (radar) testing and other electromagnetic 15

techniquescan besensitive to moistureorresindegradation
effects.

Glass/epoxy
Backlighted visual testing Intensely backlit visual testing is often used for fiberglass for 5
detection of voids, resin pockets, and cracking.

Ultrasonic testing Through-transmission or pulse echo ultrasonic testing can be 9

used, but attenuation ismuchgreaterthan in graphite/epoxy.

CROSS REFERENCE TABLES FOR TEST METHOD SELECTION 4.3

 Table 2. Foam sandwich composite: open or closed cell; variable density; polymer or graphite.

Inspection Nondestructive Comments on Method

Issues Test Techniques Nondestructive Testing Chapters

Cracking, voids, Ultrasonic testing Through-transmission ultrasonic testing is most common 9

density, bonding to technique, with standard water squirter systems at 1 MHz
skin, inclusions, fluid looking primarily for delaminations and porosity. Pulse echo
ingress, skin quality, ultrasonic testing may be used for composite skin evaluation.
porosity
Air coupled ultrasonic testing may be acceptable. Resonance 9
testing may apply.
Radiologic testing Radiologic testing can be used for internal features. 10

X-ray computed tomography for complex configurations for 10

assembly, density, and delaminations.
Thermography Thermography may be applicable to thin skin structure. 11

Shearography Shearography for skin-to-core bond, cracked core, 12

core-to-core disbonds, voids.
Bond testing Bonding of skin to core may be tested with resonance 14
technique.
Microwave testing Microwave testing for material moisture uptake. 17

Table 3. Honeycomb sandwich structures: aluminum or composite, including septumized structures.

Inspection Nondestructive Comments on Method

Issues Test Techniques Nondestructive Testing Chapters

Bonding, Ultrasonic testing Through-transmission ultrasonic testing is most common test 9

crushed/damaged technique with standard water squirter systems at 1 MHz
core, filled core, lookingprimarily for delaminations and porosity. Aircoupled
inclusions, skin ultrasonic testing may be acceptable. Pulse echo ultrasonic
quality, fluid ingress testingmay be used for composite skinevaluation.

Radiologic testing Radiologic testing can be used for internal features. 10

X-ray computed tomography is used for complex 10

configurations for assembly, density, and delaminations.

Thermography Thermography may be applicable to thin skin structure and for 11

water detection.

Shearography Shearography is applicable to skin-to-core bond, 12

core-to-septum disbonds, sheared core, crushed/damaged,
filled core, inclusions, fluid ingress.

Resonance testing; bond testing Bonding of skin to core may be tested with bond testing or 14
resonance techniques.

Tap testing Tap testing may be used for inservice check and some 14
acceptance of noncritical structure.

4.4 AEROSPACE NONDESTRUCTIVE TESTING

Table 4. Nonmetals: plastic, rubber, ceramics, propellant.

Inspection Nondestructive Comments on Method

Issues Test Techniques Nondestructive Testing Chapters
Voids, cracks, Visual testing Visual testing for plastic on surface features and for internal 5
porosity, density crazing.

Penetrant testing Penetrant testing for surface cracks. 6

Ultrasonic testing Ultrasonic testing for detection of cracks and anomalies. 9

Radiologic testing Radiologic testing for internal features such as voids. 10

X-ray computed tomography for volumetric anomalies and 10
density uniformity.

Microwave testing Microwave testing for ceramics and dielectrics. 17

Table 5. Castings: aluminum, steel, magnesium, titanium, nickel alloy; investment, sand mold, permanent mold.

Inspection Nondestructive Comments on Method

Issues Test Techniques Nondestructive Testing Chapters

Dimensional Visual testing Visual testing for surface anomalies and dimensional 5
tolerance, porosity, measurements.
shrinkage, inclusions,
cracks, slag Penetrant testing Penetrant testing for surface cracks. 6
Magnetic particle testing Magnetic particle testing for surface and near-surface 7
anomalies in ferromagnetic parts.
Electromagnetic testing Electromagnetic testing for inservice cracking. 8

Ultrasonic testing Ultrasonic testing with pulse echo and phased array 9
techniques for internal cracks.
Radiologic testing Radiologic testing is main method of casting inspection for 10
internal anomalies including porosity, shrinkage, inclusions,
and cracking.
Computed tomography is used on some castings such as 10
turbine blades for dimensions and anomalies.
Neutron imaging Neutron radiologic testing may be used for special doped 10
inclusion detection.
X-ray diffraction X-ray diffraction is used to determine crystallographic 15
orientation of single-crystal turbine blades.

CROSS REFERENCE TABLES FOR TEST METHOD SELECTION 4.5

 Table 6. Forgings: aluminum, steel, titanium, nickel.

Inspection Nondestructive Comments on Method

Issues Test Techniques Nondestructive Testing Chapters

Cracks, inclusions, Penetrant testing Penetrant testing for surface cracks. 6

grain size, tearing,
residual stress Magnetic particle testing Magnetic particle for surface and near-surface anomalies. 7

Electromagnetic testing Electromagnetic testing for inservice cracking. 8

Ultrasonic testing Ultrasonic testing of billets and sonic shapes for inclusions or 9
internal anomalies.

Radiologic testing Radiologic testing for inclusions. 10

X-ray diffraction X-ray diffraction for residual stress. 17

Barkhausen noise analysis Barkhausen noise analysis for residual stress in magnetic 17
materials.

Table 7. Machined parts: aluminum, steel, titanium, magnesium, rolled stock, forgings, castings.

Inspection Nondestructive Comments on Method

Issues Test Techniques Nondestructive Testing Chapters

Dimensional Visual testing Visual testing for surface anomalies and dimensional 5
tolerance, cracks, measurements.
residual stress,
repairs Penetrant testing Penetrant testing for surface cracks. 6

Magnetic particle testing Magnetic particle for surface and near-surface anomalies. 7

Electromagnetic testing Electromagnetic testing for inservice cracking. 8

Ultrasonic testing Ultrasonic testing for weld repairs. 9

Radiologic testing Radiologic testing for weld repairs. 10

Barkhausen noise analysis Barkhausen noise analysis for residual stress in magnetic 17
materials.

Table 8. Fastened structures: aluminum, steel, titanium, composite; rivets, bolts, fasteners.

Inspection Nondestructive Comments on Method

Issues Test Techniques Nondestructive Testing Chapters

Cracks, corrosion, Visual testing Visual testing for surface anomalies, cracks, and surface 5
alloy type displacement due to corrosion.

Electromagnetic testing Electromagnetic testing for cracking around fasteners in 8

layered structure and for loss of material by corrosion.

Ultrasonic testing Ultrasonic testing for cracks around fasteners and for loss of 9
material by corrosion.

Radiologic testing Radiologic testing for corrosion and cracking. 10

X-ray diffraction X-ray diffraction for alloy type fastener check. 17

X-ray fluorescence Material and alloy identification. 17

4.6 AEROSPACE NONDESTRUCTIVE TESTING

Table 9. Welded joints: aluminum, steel, titanium, plastic, fusion (gas, arc, others), friction, induction, laser, electron beam,
thermal plastic.

Inspection Nondestructive Comments on Method

Issues Test Techniques Nondestructive Testing Chapters
Voids, porosity, Visual testing Visual testing for surface anomalies, weld buildup, undercut. 5
shrinkage, lack of
fusion, undercut, Penetrant testing Penetrant testing for smooth weld or weld repair. 6
lack of penetration,
cracks, slag, Electromagnetic testing Electromagnetic testing for inservice cracking. 8
inclusions, residual
stress Ultrasonic testing Ultrasonic testing may be used for metallic and plastic welds 9
depending on conditions.

Radiologic testing Radiologic testing ismost commonlyrequirednondestructivetest 10

for anomalies in metallic weldments and weld repairs.

Acoustic emission testing Acoustic emission weld monitoring. 13

X-ray diffraction X-ray diffraction for residual stress. 17

Table 10. Bonded joint assemblies: aluminum, titanium, composite, core, foam, primary structure bonds, secondary structure
bonds, covalent bonds, hot bonds, room temperature bonds, patch repairs, diffusion bonds.

Inspection Nondestructive Comments on Method

Issues Test Techniques Nondestructive Testing Chapters

Disbonds, voids, Visual testing Visual testing for surface preparation, process control, and 5
degradation, after bond adhesive squeezeout.
bond strength,
contamination, Ultrasonic testing Ultrasonic test techniques are most common bond testing. 9
surface wetting Inspectionisprimarily for disbonds, delamination, and voids
usingstandard pulseechoultrasonic testing, resonance, or
advanced ultrasonic testtechniques. Ultrasonictestingmaybe
used for metallic and plastic welds depending on conditions.
Through-transmission technique is applicable.

Radiologic testing Radiologic testing for voids in some types of bonds. 10

Thermography Thermography for disbonds. 11

Shearography Shearography for disbonds, metal-to-metal bonds, repairs, 12

diffusion bonds, foam and honeycomb core bonding.

Shock waves Shock wave techniques can be used for bond strength 14
evaluation.
14
Tap testing Tap testing for disbonds and delamination.

Microwave testing Microwave testing for foam bonds and bond degradation. 17

Dielectric testing; capacitance Dielectric or capacitance testing for bond degradation. 17

testing

Spectroscopy Spectroscopy for material identification. 17

Fourier transform infrared testing Fourier transform infrared testing for surface chemistry. 14

CROSS REFERENCE TABLES FOR TEST METHOD SELECTION 4.7

 Table 11. Coatings: paint, conductive layers, thermal coatings, insulation, low observable, abradable seals.

Inspection Nondestructive Comments on Method

Issues Test Techniques Nondestructive Testing Chapters
Thickness, Visual testing Visual testing for surface condition. 5
adhesion, damage,
degradation, Electromagnetic testing Electromagnetic testing for conductive coatings or thickness of 8
disbonds coatings on conductive substrate.

Ultrasonic testing Ultrasonic testing with high frequency pulse echo technique, 9
surface wave (lamb wave) technique, or guided wave
technique.

Thermography Thermography for uniformity and disbonds of some types of 11

coatings.

Shearography Shearography for coating adhesion, damage, degradation, 12

disbonds.

Beta backscatter technique Beta backscatter for coating thickness. 17

Microwave testing Microwaves for thickness of some coating types and for 17
coating degradation.

Table 12. Engines and motors: jet, piston, liquid rocket, solid rocket.

Inspection Nondestructive Comments on Method

Issues Test Techniques Nondestructive Testing Chapters

Inservice cracking, Visual testing Visual testingwith borescope for blade damage (foreign object 5
corrosion volumetric damage to compressor blades; erosion in turbine blades).
features, assembly
configuration, Fluorescent penetrant testing Fluorescent penetrant testing for surface connected 6
propellant aging, discontinuities, especially cracking.
inhibitor disbonds,
Magnetic particle testing Magnetic particle for cracking in ferromagnetic components. 7
thermal insulation,
erosion Eddy current testing Eddy current testing for cracking of metallic components and 8
conductivity testing for heat damage.

Ultrasonic testing Ultrasonic testing for internal discontinuities and cracking 9

especially in air seal disk forgings.

Ultrasonic thickness measurement for wall thickness in hollow 9

turbine blades, evaluating blends, and assessing serviceability
from wear.

Radiologic testing Radiologic testing for internal discontinuities especially in 10

castings.

Computed tomography inspection for assembly and propellant 10

aging.
Thermography
Thermography for internal discontinuities in composites. 11
Shearography/holography
Holography for brazed seals. 12

Shearography for composite stators fan cases, liner and 12

insulation disbonds.
X-ray diffraction
X-ray diffraction to measure crystal orientation in 15
single-crystal turbine blades.

4.8 AEROSPACE NONDESTRUCTIVE TESTING

 Table 13. Subsystems: landing gears, tires, hydraulics, motors, actuators, pressure vessels, others.

Inspection Nondestructive Comments on Method

Issues Test Techniques Nondestructive Testing Chapters

Corrosion, cracks, Visual testing Visual testing for surface condition. 5

residual stress,
surface condition, Electromagnetic testing Electromagnetic testing for near-surface cracking and 8
assembly, failure corrosion.
analysis
Radiologic testing Radiologic testing for internal failed components and 10
corrosion.
Computed tomography for internal configuration and assembly. 10

Shearography Shearography for pressure vessels. 12

Acoustic emission testing Acousticemissiontesting for pressurevessels(includingrocket 13

motor cases).
Barkhausen noise analysis Barkhausen testing for residual stress in ferromagnetic 17
materials.

Table 14. Inservice or damaged structure: metals, composites, assemblies, systems.

Inspection Nondestructive Comments on Method

Issues Test Techniques Nondestructive Testing Chapters
Fatigue cracks, Visual testing Visual testing for surface condition. 5
corrosion, impact
damage, heat Penetrant testing Penetrant testing for fatigue cracks. 6
damage, moisture
ingress, disbonds/ Magnetic particle testing Magnetic particle testing for fatigue cracks in ferromagnetic 7
delamination materials.
Electromagnetic testing
Electromagnetictesting for fatigue cracking, corrosionmaterial 8
loss, or aluminum heat damage.
Ultrasonic testing
Ultrasonic testing for composite impact damage, cracks, and 9
corrosion.
Radiologic testing
X-ray inspection for corrosion and assembly internal 10
configuration change.
Thermography
Thermography for degradation or impact damage and for 11
water ingress detection in honeycomb.
Shearography
Shearography for degradation, impact damage, water, ingress, 12
or damaged core.
Acoustic Emission testing
Acoustic emission testing for structural health monitoring. 13
Microwave testing
Microwave testing for moisture or damage. 17

Table 15. Electronics: wiring, connectors, circuit boards, computer chips.

Inspection Nondestructive Comments on Method

Issues Test Techniques Nondestructive Testing Chapters
Connectivity, Visual testing Visual testing for connections. 5
corrosion damage
Electrical checks. 5
Ultrasonic testing Acoustic microscopy for electronic components. 9

Radiologic testing X-ray inspection for solder and damage. 10

Computed tomography inspection of connector assembly. 10

Thermography Thermography for electric hot spots. 11

CROSS REFERENCE TABLES FOR TEST METHOD SELECTION 4.9

 Method Tables
PART3

Tables 16 through 26 describe the nondestructive

test methods covered in the chapters of the
Liquid Penetrant Testing
handbook. These tables indicate applications of the Liquid penetrant testing is an important
nondestructive test methods for various materials, nondestructive test method for aerospace parts to
structures, and systems. detect surface cracks that are below visible detection
limits. Aerospace structures are often fatigue critical,
requiring tests for surface anomalies. Table 17 lists
characteristics of liquid penetrant testing and
Visual Testing applicable material types for tests.
Visual testing is an essential part of all
nondestructive testing activity. Table 16 lists
characteristics of visual testing and applicable
material types for inspection.

Table 16. Visual testing.

Characteristics Comments

Physical Principles
Optical, visible light reflection, or transmission Use glancing light, magnification, borescopes for internals, and video
recording.
Applications
Surface feature detection and measurement. Distorted structure, Always visually check product as part of nondestructive testing
cracks, voids, discoloration, roughness, pitting activity and correlate observations with other nondestructive test
data. Use enhanced lighting and visual aids as appropriate.
Advantages
Easy to apply, high resolution, can be applied through small access Requires proper lighting and vision acuity.
holes Very subtle surface changes can be detected that may indicate
internal discontinuities.
Limitations
Must have access to surface, subsurface anomalies only inferred As magnification is increased for sensitivity field of view becomes
by changes in the surface, image contrast affects sensitivity smaller and time to inspect increases.
Material or Structure Inspection Issues
Fiber reinforced polymer (FRP) composite laminate Surface porosity, dry plies, wrinkling, contamination, impact damage,
crazing, matrix cracking, resin changes
Nonmetals Voids, crazing, cracks
Castings Cracks, surface porosity, shrinkage
Forgings Cracks, dimensions
Machined parts Cracks, dimensions
Fastened structure Distorted structure, missing fasteners, cracks, corrosion, damage
Welded joints Surface porosity, shrinkage, lack of fusion, cracks, undercut
Bonded joint assemblies Surface preparation conditions, bond line thickness at edges,
squeeze out
Coatings Thickness, cracks, damage, adhesion
Engines and motors Blade damage and corrosion
Subsystems Cracks, corrosion, fitup
Inservice or damaged structure Impacts, cracks, corrosion
Electronics Heat damage, burns, damaged components, wiring

4.10 AEROSPACE NONDESTRUCTIVE TESTING

Magnetic Particle Testing Radiologic Testing
Magnetic particle testing is applicable to Radiologic testing, sometimes called radiographic
ferromagnetic materials. Table 18 lists characteristics testing, is one of the oldest nondestructive testing
of magnetic particle testing and applicable material methods. Table 21 lists characteristics of radiologic
types for inspection. testing and material types for inspection.

Eddy Current Testing Thermography

Eddy current testing is one of the most widely Infrared thermography is applicable to aerospace
used techniques for aerospace materials because of sandwich structures and other material
its application to inservice testing for aluminum configurations. Table 22 lists characteristics of
structure. Table 19 lists characteristics of eddy infrared thermography and applicable material types
current testing and applicable material types for for inspection.
inspection.

Shearography and Holography

Ultrasonic Testing Shearographic and holographic methods use
Ultrasonic testing is one of the most commonly optical imaging interferometers combined with
used methods for aerospace structure (especially stressing mechanisms and may be very sensitive to
composities) and is applied in a wide variety of subsurface anomalies and structures. Table 23 lists
applications. Table 20 lists characteristics of characteristics of shearographic and holographic
ultrasonic testing and applicable material types testing and applicable material types for inspection.
for inspections.

Table 17. Liquid penetrant testing.

Characteristics Comments

Physical Principles
Liquid capillary action fills cracks and is bled back out to indicate Dye and fluorescent penetrants are designed with surface energies
anomalies. and viscosities appropriate for material applications.
Applications
Surface connected anomalies Select penetrant testing for surface breaking discontinuities, such as
Cracks, pinholes, laps, seams, cold shuts, leaks. fatigue crack or processing surface anomalies in fatigue critical
components.
Advantages
Easy to apply, no power, applicable to manymaterials, detects Covers large parts rapidly, magnifies effects of tiny surface
cracks that cannot be resolved with naked eye. discontinuities for detection.
Limitations
Anomalies must be open to nonporous surface. Dirt, contamination surface roughness, or processes that smear
metal may mask discontinuities.
Chemical cleanup, environmental concerns. Penetrant rinsingtechniquescan overwashand removepenetrant
before bleedout, missing indications.
Material or Structure Inspection Issues
Castings Cracks, laps, seams, cold shuts, leaks
Forgings Cracks
Machined parts Cracks
Welds Cracks
Inservice Fatigue cracks

CROSS REFERENCE TABLES FOR TEST METHOD SELECTION 4.11

 Acoustic Emission Testing Barkhausen Noise Analysis
Acoustic emission testing is a passive method that The barkhausen effect refers to discontinuous
detects elastic stress waves, which may be audible or changes in magnetization of magnetic materials
ultrasonic, that are released in a structure at the subject to a time varying magnetic field. Barkhausen
onset of failure. Table 24 lists characteristics of noise analysis is used for nondestructive testing of
acoustic emission testing and applicable material ferromagnetic materials, mainly steel, in various
types for inspection. areas including aerospace, pipeline, automotive,
power plant, and steel industries. Table 26 lists
characteristics of barkhausen noise analysis and
Microwave Testing applicable material types for inspection.
Microwave testing is used primarily for moisture
detection in nonmetal structure. Table 25 lists
characteristics of microwave testing and applicable
material types for inspection.

Table 18. Magnetic particle testing.

Characteristics Comments

Physical Principles
Magnetic field distortions due to anomalies in test object are Magnetic field lines are applied in appropriate directions on an
detected by alignment of magnetic particles on surface. article to test.
Magnetic particles (wet or dry) are dispersed on part and align with
magnetic field.
Anomalies distort field lines, causing magnetic particles to indicate
presence of discontinuity.

Applications
Near-surface anomalies in ferromagnetic materials Select magnetic particle testing for surface or near-surface
Cracks, seams, pores, inclusions anomalies in fatigue critical components.

Advantages
Detects cracks and near-surface features that cannot be resolved Covers large parts rapidly, magnifies effects of tight anomalies for
with naked eye. detection.
Can detect anomaly that is not necessarily surface breaking.
Limitations
Material must be ferromagnetic and anomaly within magnetic Only works on ferromagnetic materials.
field near to surface. Magnetic fields must be properly oriented.
Large parts may require high currents for magnetization.
Demagnetizing after testing may leave residual.
Material or Structure Inspection Issues
Castings Cracks, laps, seams, hottears, shrinkage, coldshuts, near-surface
inclusions
Forgings Cracks, near-surface inclusions
Machined parts Cracks, grinding near-surface inclusions
Inservice Fatigue cracks

4.12 AEROSPACE NONDESTRUCTIVE TESTING

Table 19. Eddy current testing.

Characteristics Comments

Physical Principles
Induced electric field lines (10 Hz to 10 MHz) are Eddycurrentsaresensitivetoconductivity,permeability,and
perturbed by anomalies creating magnetic fields geometry.
that affect inductive reactance of testing probe. Single or multiple eddy current coils and sensors are used to induce
fields and detectchanges.
Applications
Surface and subsurface cracks, corrosion, Choose eddycurrent for conductivematerial inspections for cracks,
inclusions, sorting, thickness, and damage in corrosion, damage, coatings, or changes that are at surface or
materials. near-surface.
[Link]
frequency requires larger coils but penetrates deeper.
Resolution affected by frequency. Higher frequency withsmaller
coils increases resolution.
Material electrical conductivity measurement for
alloy detection and heat damage effects
Advantages
Very fast, low cost method for surface and Noncontact technology, but mostly surface riding.
subsurface discontinuity detection.
Coils can be designed for many configurations Eddy current technique can be varied through coil designs and
including tubes. magnetic sensors.
Limitations
Must be conductive material. Magnetic materials affect sensitivity.
Depthof penetration for detectiondependson Liftoff from surface and surface roughness variations affect
frequency and conductivity. readings.
Material or Structure Inspection Issues
Metals Conductivity, heat damage, alloy type, cracking
Castings Cracks, laps, seams, hottears, shrinkage, coldshuts, near-surface
inclusions
Forgings Cracks, inclusions
Machined parts Cracks, inclusions
Fastened structure Cracks, corrosion
Welded joints Cracks, lack of fusion, porosity
Coatings Thickness, degradation
Engines and motors Cracks
Subsystems Cracks
Inservice Fatigue cracks
Fiber reinforced polymer (FRP) composite laminate Not commonly applied to fiber reinforced polymer composites, but
there is an eddy current effect that can be used to detect changes in
composite characteristics.

CROSS REFERENCE TABLES FOR TEST METHOD SELECTION 4.13

 Table 20. Ultrasonic testing.

Characteristics Comments

Physical Principles
Acoustic waves, typically in the 0.5 to 15 MHz range, with Through-transmission, pulse echo, and resonance techniques.
longitudinal, transverse, lamb, and raleigh waves Some applications using acoustic microscopy can use ultrahigh
frequencies up to or exceeding 100 MHz.
Applications
Detection of cracks, delaminations, inclusions, voids, thickness, Apply to wide range of aerospace products during production and in
porosity, laps, seams service to validate material consolidation and to detect planar lying
anomalies.
Advantages
Very sensitive to planar anomalies Can be applied in wide range of configurations.
Relatively low cost Sensitivity can be adjusted to meet test needs.
Limitations
Material must transmit acoustic energy. Ultrasound must be coupled to test article. Techniques include
Multilayer materials must be in intimate contact or bonded for contact, water dribblers, water jet, air coupled, laser, and
transmission. electromagnetic.
Orientation of ultrasonic beams, part geometry, and discontinuity
orientation may limit inspectability.
Anisotropic materials can cause counterintuitive results.
Material or Structure Inspection Issues
Fiber reinforced polymer (FRP) composite laminate Consolidation, porosity, delaminations, thickness variation
Foam composite Voids, cracks, disbonds, crushed core
Honeycomb composite Disbonds, inclusions, crushed core
Nonmetals Voids, cracks, delaminations, inclusions
Castings Cracks, inclusions, porosity, shrinkage
Forgings Cracks, inclusions
Fastened structure Distorted/damaged structure, corrosion, cracks
Welded joints Cracks, porosity, inclusions, shrinkage, lack of fusion
Bonded joint assemblies Adhesive fill, voids/porosity
Coatings Thickness, cracks
Engines and motors Volumetric anomalies in forging, cracks, corrosion
Subsystems Cracks, corrosion
Inservice or damaged structure Impact damage, cracks, corrosion

4.14 AEROSPACE NONDESTRUCTIVE TESTING

Table 21. Radiologic testing.

Characteristics Comments

Physical Principles
Transmission of radiation through materials is function of
attenuation coefficient (absorption and scatter) and thicknessof Transmission or scatter of penetrating radiation is used to image
material. internal condition of test objects for changes in material, density, or
thickness.
Applications
Detection of voids, porosity, inclusions, cracks in a wide range of
materials and structures Radiologic testing is used to detect volumetric feature changes in
Use X-rays, gamma rays, or neutrons. objects.
Advantages
Penetrates materials, noncontact, dimensional measurements
Area inspection
Can be applied in a wide range of configurations.
Versatile physical testing schemes are possible.
Limitations Applies to structure and systems composed of multiple components.
Sensitivity depends on material radiation attenuation coefficient.
Anomalies must be oriented to radiation beam for detection.
Crack detection depends on orientation.
Radiation safety concerns and regulations
Penetrating radiations are health hazards requiringsafety
precautions.
Setting safe boundaries and working off shift hours are among the
difficulties associated with field tests.
Material or Structure
All materials Inspection Issues
Fiber reinforced polymer (FRP) composite laminate Cracks, delaminations, uniformity, inclusions
Foam composite Tracer yarn orientation, matrix cracking
Honeycomb composite Voids, cracks
Castings Crushed core, fluid ingress
Forgings Cracks, inclusions, porosity, shrinkage
Machined parts Cracks, inclusions dimensions
Fastened structure Cracks, dimensions
Welded joints Cracks, corrosion, damage
Bonded joint assemblies Cracks, porosity, shrinkage, lack of fusion
Coatings Disbonding, bondlinethicknessvariation, voids
Inservice or damaged structure Component location, damage
Electronics Impact damage, delamination, cracks, corrosion
Solder connections, wiring

CROSS REFERENCE TABLES FOR TEST METHOD SELECTION 4.15

 Table 22. Thermography.

Characteristics Comments

Physical Principles
Rate of diffusion of heat through sample is function of material Typically, heat lamp illuminates surface and diffusion of absorbed
conditions. energy from surface is monitored. Features or conditions below
Detected image is function of surface emissivity surface affect heat flow.
Passive infrared imaging detects hot spots on surfaces.
Applications
Detection ofsubsurfacediscontinuitiessuchasdelaminations, Thermography is useful for quick surveys of large materials for
voids, porosity, inclusions or disbonds internal planar damage orchanges.
Advantages
Area inspection Can be applied in wide range of configurations.
Noncontact Can be used actively with heat lamps or passively to observe
surfaces.
Image display of results Flash thermography helps detect anomalies close to the surface.
Limitations
Surface emissivity affects results. Low emissivity limits detectability.
Heat uniformity and directionality depends on the material. Lateral dispersion of heat limits depth of sensitivity
Material or Structure Inspection Issues
Fiber reinforced polymer (FRP) composite laminate Delaminations, porosity, inclusions
Foam composite Thin skin inspection for porosity
Honeycomb composite Skininspection, delamination, fluid ingress
Bonded joint assemblies Disbonding, voids
Engines or motors Hot spots
Coatings
Inservice or damaged structure Impact damage, delamination, cracks, corrosion
Electronics Overheated components

4.16 AEROSPACE NONDESTRUCTIVE TESTING

Table 23. Shearography and holography.

Characteristics Comments

Physical Principles
Surface displacements on order of 100 to 500 nm are detectable Bystressing structurewith heat, pressure, partial vacuumorother
by phase stepping optical interference. means, the surface strain is measured which relates to localized
stiffness of structure. Anomalies are detected as local
out-of-plane deformation.
Applications
Detectionof subsurfacediscontinuitiessuchasdelamination, Shearography and holography are used on composite or metal
disbonds, voids, porosity, inclusions, damage, and water. bonded structures, for internal damage, disbonds, and delamination,
damage or changes in material properties.
Advantages
Area inspection Can be applied to wide range of material and structure
configurations.
High throughput Excellent on thin skin structures.
Noncontact
Image display of results with easy interpretation
Quantitative measurements
Fast testing
Limitations
Requires uniform change in stress over field of view. Minimumdiscontinuitysizedetectableincreases withdepth and
Environmental effects can degrade data. increased modulus. Field and production shearography and
holography systems, properly used, substantially overcome most
Requiresdiffusive surfaceswithreflectivity typically greater limitations.
than 3 percent.
Sensitivity of method may require multiple tests.
Thickness and stiffness of structure can limit sensitivity to deeper
anomalies.
Material or Structure Inspection Issues
Fiber reinforced polymer (FRP) composite laminate Delaminations, porosity, inclusions, impact damage
Foam composite Thin skin inspection for delamination from foam
Honeycomb composite Skin inspection, delamination, skin to core, far side skin to core
disbonds impact damage, water, composite to liner disbonds, fiber
bridging
Sprayed-on foam insulation (SOFI) Voids, delaminations, crushed foam
Bonded joint assemblies Disbonding
Braze or diffusion bonded metal structures Disbonding
Composite over-wrapped pressure vessels (COPV) Impactdamage, cracks, fiber bridging, linerdisbonds, broken fibers,
and changes in material properties
Inservice or damaged structure Impact damage, delamination, disbonds, cracking, corrosion

CROSS REFERENCE TABLES FOR TEST METHOD SELECTION 4.17

 Table 24. Acoustic emission testing.

Characteristics Comments

Physical Principles
Acoustic signals are detected at various locations in test object Array of acoustic sensors placed at distributed locations on part
under stress. under load will detect noise signals from localized flaw growth
activity withinstructure.
Applications
Senses onset of failure in structure under load. Apply during proof testing or other loading operations of structure
Provideslocalization of failure points. to sense onset of critically active acoustic emission leading to
failure.
Advantages
Dynamic inspectionofstructureunderstress Multiple sensors can be used to determine location of failure.
Limited access required.
Limitations
Requires system to be in some state of stress. Noises from both structure and environment require interpretation
Background noise of their significance.

Material or Structure Inspection Issues

Fiber reinforced polymer (FRP) composite laminate Used during proof testing to monitor amount of damage in structure
and to predict failure loads.
Bonded joint assemblies Disbond sensing under load.
Inservice or damaged structure Detection ofcracking, otherdamage detection, oroverloading
during service.
Pressure vessels Used during proof testing to monitor onset and amount of damage
in structure.

Table 25. Microwave testing.

Characteristics Comments

Physical Principles
Electromagnetic waves (microwaves, millimeter waves, and Microwave guide and detector are used.
terahertz waves) in the 0.3 to 30 mm wavelength range are
scattered, reflected, or attenuated.
Applications
Changes in dielectricproperties, thickness, ordiscontinuities are Consider microwaves for dielectric materials uninspectable with
detected. ultrasound whereconcerninchange in dielectric propertiesordepth
Cracks, porosity, disbonds, moisture ingress, or absorption ofpenetration withothermethods is limited.

Advantages
Noncontact technique Very sensitive to moisture
Applies to dielectric materials.
Deep penetration
Limitations
Does not apply to conductive materials such as metals. Sensitivity limitations due to reflection coefficient of anomalies of
interest
Material or Structure Inspection Issues
Fiber reinforced polymer (FRP) composite laminate Moisture, delaminations
Foam composite Cracks, debonds, moisture
Honeycomb structure Moisture
Sprayed-on foam insulation (SOFI) Millimeter and terahertz waves are used to detectvoids.
Bonded joints Disbonds
Coatings Cracks, disbonding
Inservice or damaged structure Moisture

4.18 AEROSPACE NONDESTRUCTIVE TESTING

Table 26. Barkhausen noise analysis.

Characteristics Comments

Physical Principles
Inductivelydetectingnoisegenerated by the abrupt movementsof
Barkhausen noise analysis is sensitive to presence and distribution of
magnetic domain walls forced to move by application of
elastic stresses, and metallurgical microstructure, often related to
alternating magnetic field, typically around 100 kHz
hardness.
Applications
Primarily used on heat treated, high strength, low alloy steel parts
Technique involves detecting evidence of microstructural changes,
stresses, or combinations of both conditions in ferromagnetic
materials.
Advantages
Can be applied in wide range of manufacturing and inservice
configurations. Detects metallurgical conditions of future failures.
May be performedbefore or after chromiumplating orother
nonmagnetic protective coating process. Sensitivity can be adjusted to meet inspection needs.

Limitations
Material must be ferromagnetic.
Technique is not applicable for parts that have plated metal
ferromagnetic deposit(suchasnickel) betweensubstrate and
chromium plating.
Material or Structure Inspection Issues
Landing gear Elastic stress: normalloading stress; localized inducedstresses
(impact damage); overload yielding stress
Bearings Microstructural features: hardness (improper heat treatment); shot
peen quality; high temp thermal conditions, including friction burns
including grinding burns
Bearing housings External heat sources

CROSS REFERENCE TABLES FOR TEST METHOD SELECTION 4.19

References

Bibliography
ASNT. 2012. Nondestructive Testing Handbook, 3rd ed.: Vol. 10,
Nondestructive Testing Overview. Columbus, OH: American Society
for Nondestructive Testing.
Vary, A. 1973. N73-31441, Nondestructive Evaluation Technique Guide.
Springfield, VA: National Technical Information Service.

4.20 AEROSPACE NONDESTRUCTIVE TESTING

 X
5
CHAPTER
CHAPTER

Visual
Testing

Contents
Part 1. Introduction, 5.2
Part 2. Optical Principles, 5.3
Part 3. Optical Techniques, 5.8
Part 4. Applications, 5.16
References, 5.24

5.1
 Introduction
PART1

Visual testing is the most fundamental attack are also of interest. Detailed (small area)
nondestructive test method. Eighty percent of inspections are conducted using optical aids, such as
transport aircraft inspections are visual magnifiers and borescopes, often very effective
(FAA 43-204 1997). Visual tests can take a variety of where access is poor for other nondestructive test
forms depending on the level of detail required for methods. Additional visual techniques are used for
the task. Visual testing of part condition and surface quantitative measurements of features or surfaces.
detail is often critical for the interpretation of results Proper visual testing is a combination of
of other nondestructive test methods and should be a observations of the global condition of the structure,
first consideration when any inspection is to be where attitude and appearance might indicate an
performed. For aircraft and engine maintenance, anomalous condition, and close visual inspection for
visual testing is the primary method, both basic microscopic anomalies that could affect safety and
observation and enhanced visual techniques. performance. Visual testing can also include surface
Generally, visual inspections cover a broad area of measurements where surface indications and surface
the aircraft structure and engines. Cracks, corrosion, displacements may be indicative of subsurface
and disbonds are primary interests. Environmental concerns. Table 1 summarizes visual testing
effects such as sunlight discoloration or chemical applications, advantages, and limitations.

Table 1. Visual testing advantages, limitations, and applications.

Characteristics Comments

Physical Principles
Light (visible wavelength radiation) is Reflected or scattered light intensity provides information on color or contrast.
reflected from or transmitted through Transmission of lightmeasures opacity. Illumination andmagnification are used for
surfaces. enhancement. Surface displacement measurements can be made.

Advantages
Easy to apply; rapid; high resolution; low Requires proper lighting. Enhancement toolsaid access. Very subtle surfacechangescan be
cost; accessible to internal regions through detectedthat mayindicate internal anomalies.
small openings.

Limitations
Operatormusthave access to surface; Surfaces must be clean. As magnification is increased for sensitivity, field of view becomes
subsurface discontinuities are inferred only smaller and time to inspect increases. Operator must have good vision acuity and
through changes at the surface; image knowledge of structure.
contrast affects sensitivity.

Applications Inspection Features

General The method detects and measures surface features: distorted structure, cracks, voids,
discoloration, roughness, and pitting. Surfacedistortionmeasurement can be enhancedby
glancing angleandinterference fringe methods.
Fiber reinforced polymer composite laminate Surfaceporosity,dryplies,wrinkling,contamination,impactdamage,crazing,cracking,resinchanges.
Nonmetals Voids, crazing, cracks.
Castings Cracks, surface porosity, shrinkage.
Forgings and machined parts Cracks, dimensions.
Fastened structures Distorted structure, missing fasteners, cracks, corrosion, dents, damage.
Welded joints Surface porosity, shrinkage, lack of penetration, cracks, undercut.
Bonded joint assemblies Surface preparation conditions, bondline thickness at edges, squeeze out.
Coatings Uniformity, cracks, damage.
Engines and motors Cracks and corrosion.
Subsystems Cracks, corrosion, fitup.
Inservice or damaged structures Impacts, cracks, corrosion.
Electronics Heat damage, burns, damaged components, wiring.
Surface preparation For other methods.

5.2 AEROSPACE NONDESTRUCTIVE TESTING

 Optical Principles PART2

Light Specular reflection of light occurs when it strikes

a smooth surface. The direction of the reflected beam
Visual testing is based on the use of light photons can be determined by constructing a line
in the visible range of the electromagnetic spectrum perpendicular to the reflecting surface. The angle of
where the wavelength is generally considered to be reflection is the same as the angle of incidence
between 400 and 780 nm (between 4 × 10-7 and compared to the perpendicular, or normal line
7.8 × 10-7 m). Table 2 shows the range of visible (Figure 1).
light in the electromagnetic spectrum. Diffuse reflection is caused when light strikes an
The important variables in light propagation uneven surface. A rough surface has many different
include wavelength, frequency, and the coefficients plane surfaces. Each incident light beam strikes a
of reflection and refraction. All forms of different reflecting plane and is reflected at an angle
electromagnetic radiation travel through a vacuum that corresponds to the relative angle of the plane
at the same speed, 299 793 km·s–1 (186 282 mi·s–1).
When light travels through any other medium, the
velocity is altered. The change of velocity at
interfaces causes refraction of the waves. The Figure 1. Angle of incidence equals angle of reflection (ASNT 2010).
difference in the time it takes light to travel through
different media is responsible for the operating
principles of optical instruments. While the velocity
is variable, the frequency remains fixed and is
independent of the medium. The velocity v,
frequency f, and wavelength  are related:
(1) v  f  
where v is the velocity of light waves in the medium.
The energy contained in the photon is given:

(2) E  hf Legend
 = Angle of incidence
where E is energy and h is Planck‘s constant,  = Angle ofreflection
6.63 × 10-34 J·s (4.14 × 10–15 eV·s).

Table 2. Electromagnetic spectrum.

Common Name Wavelength Frequency Energy

Long waves 103 m (1 km) 3 × 105 Hz (30 kHz) 10–9 eV (1.2 neV)
Amplitude modulation (AM) radio 102 m (100 m) 3 × 106 Hz (3 MHz) 1.2 × 10–8 eV (12 neV)
Radio and television 10 m 3 × 107 Hz (30 MHz) 1.2 × 10–7 eV (0.12 µeV)
Radio and television 1m 3 × 108 Hz (300 MHz) 1.2 × 10–6 eV (1.2 µeV)
Radar 10–1 m (0.1 m) 3 × 109 Hz (3 GHz)1 1.2 × 10–5 eV (12 µeV)
Microwaves 10–2 m (1 cm) 3 × 1010 Hz (30 GHz) 1.2 × 10–4 eV (0.12 meV)
Millimeter waves 10–3 m (1 mm) 3 × 1011 Hz (300 GHz) 1.2 × 10–3 eV (1.2 meV)
Terahertz waves, far infrared radiation 10–4 m (0.1 mm) 3 × 1012 Hz (3 THz) 1.2 × 10–2 eV (12 meV)
Infrared radiation 10–5 m (10 µm) 3 × 1013 Hz (30 THz) 1.2 × 10–1 eV (0.12 eV)
Near infrared radiation 10–6 m (1 µm) 3 × 1014 Hz (0.3 PHz) 1.2 eV
Light (visible radiation) 10–7 m (0.1 µm) 3 × 1015 Hz (3 PHz) 12 eV
Ultraviolet radiation 10–8 m (10 nm) 3 × 1016 Hz (30 PHz) 1.2 × 102 eV (0.12 keV)
X-rays 10–9 m (1 nm) 3 × 1017 Hz (0.3 EHz) 1.2 × 103 eV (1.2 keV)
X-rays 10–10 m (0.1 nm) 3 × 1018 Hz (3 EHz) 1.2 × 104 eV (12 keV)
X-rays (including gamma rays) 10–11 m (10 pm) 3 × 1019 Hz (30 EHz) 1.2 × 105 eV (120 keV)
X-rays (including gamma rays) 10–12 m (1 pm) 3 × 1020 Hz (0.3 ZHz) 1.2 × 106 eV (1.2 MeV)

VISUAL TESTING 5.3

 surface. The difference from a perfect normal object‘s surface. If the optical density of the new
reflection forms the basis of some optical techniques. medium is less than that of the original, the light is
Refraction is the bending or pivoting of light bent away from the normal. The amount of
from its original direction at the interface of two refraction depends on the angle of the incident light
different media. If the optical density of the new and the index of refraction. The index of refraction
medium is greater than the density of the original, is given by the ratio of the velocity of light in a
the path of light is bent toward a line normal to the vacuum to the velocity of light in the medium:

vv
(3) n 
Figure 2. Refraction of light (ASNT 2010). vm
where n is index of refraction, vm is the velocity in
a medium, and vv is the velocity in a vacuum. The
actual angle of refraction is determined by a
mathematical relation called Snell‘s law:
0
Air
(4) n0 sin 0  n1 sin1

Glass where n0 is the index of refraction in the initial

1 medium, 0 is the initial angle of incidence, n1 is the
index of refraction in the new medium, and 1 is the
angle of refraction. Figure 2 shows the refraction of
Air light in glass.
Lenses use the process of refraction to bend light
for focusing and magnifying. Converging lenses
Legend focus light on a single point while divergent lenses
0= angle of incidence disperse light. Convex lenses are converging lenses;
1= angle of refraction they are thicker in the center than at the edge.
Concave lenses are divergent; they are thinner in the
center than at the edge. Figure 3 shows examples of
Figure 3. Lens types: (a) biconvex; (b) plano-convex; these lenses. Thin lenses are those where the
(c) convexo-concave; (d) biconcave; (e) plano-concave; thickness of the lens is small compared to the focal
(f) concavo-convex (ASNT 2010).
(a) (b) (c) (d) (e) (f)
Figure 5. Compound lenses: (a) combination;
(b) doublet; (c) triplet (ASNT 2010).

(a)

(b)
Figure 4. Converging lens geometry (ASNT 2010).

Principal Focal
plane

Object
Optical axis
(c)
Image

Focal length

5.4 AEROSPACE NONDESTRUCTIVE TESTING

length. The properties of thin lenses are described focal distance when the focal distance is measured in
using the lens law that relates the image distance, meters. A change of 1 diopter will have the effect of
object distance, and focal length of a lens as follows: enlarging the object‘s image by 25 percent.
1 1 1
(5)  
f i o
Illumination
where f is focal length, i is image distance, and o is Radiometry is the measurement of radiant energy,
object distance. The focal length or focal distance of both visible and invisible. Radiometry and
a lens is the distance from the principal plane to the photometry have the same principles but different
focal plane. The principal focus or focal plane is units of measure (Table 3). The intensity of light was
where parallel incident light rays converge after formerly measured in footcandles (ftc) but is now
being refracted by the lens. Single converging lens expressed in lux (lx): 1 ftc  10 lx. A typical indoor
magnifiers, as shown in Figure 4, allow the eye to office has illumination of about 400 lx. Daylight
be positioned closer to the test object when the ranges from 1 to 25 klx. Direct sunlight is several
retinal focus plane is at the focal plane of the lens. times more. Light sources are measured in candela
Magnification is the size of the image divided by the (cd), defined as the luminous intensity in a given
size of the object: direction of a source that emits monochromatic
Di radiation
1.46 mW ofper540 THz at The
steradian. a radiant intensity
luminous flux of
in a
(6) M   Si
S
Do o
steradian (sr) is measured in lumens (lm). The
measurement in lumens is the product of candela
and steradian (1 lm = 1 cd·sr). A light flux of one
where Di is image distance from optical axis, Do is lumen (1 lm) striking one square meter (1 m2) on the
object distance from optical axis, M is
magnification, Si is image size, and So is object size. surface of the sphere around the source illuminates it
Thick lenses may consist of a single thick lens, with one lux (1 lx). If the source itself is scaled to
combinations of thin lenses, or compound one square meter (1 m2) and emits one candela
arrangements of thin lenses. Compound lenses (1 cd), then the luminance of the source is 1 cd/m2.
shown in Figure 5 are used in most optical systems Table 4 shows the many characteristics of light, light
from doublet and triplet magnifiers to sophisticated sources, lighting materials, and lighting installations
zoom camera lenses (Sayler 1998). Compound lenses that may be measured, including (1) illuminance,
provide for high magnification and close control of (2) luminance, (3) luminous intensity, (4) luminous
the focal plane. When properly designed, compound flux, (5) contrast, (6) color appearance and color
lenses also can correct the chromatic and spherical rendering, (7) spectral distribution, (8) electrical
aberrations inherent to a single lens. Chromatic characteristics, and (9) radiant energy (ASNT 2010).
aberration occurs when lenses are unable to focus The purpose of lighting in a visual inspection
different wavelengths of light at the same focal point. area is to provide adequate contrast so that relevant
This results from a change in refractive index objects or discontinuities are detected. Contrast
(decreasing) with wavelength (increasing). Spherical detection is the most basic of visual tasks. It is a
aberrations occur due to improper focusing of the property of the difference between an object and its
refracted light from the edges of lenses relative to background of either luminance or color. Luminance
near the center. contrast is the difference in reflected light between
The refracting power of lenses can be expressed the discontinuity and its background. The contrast
by the diopter. The diopter is the inverse of the value is constant for any value of luminance but,
distance (in meters) from the lens to an image of a like reflection, varies with the position of the subject
distant object. This distance is the inverse of the (observer) and object. Chromatic contrast is the

Table 3. Photometric quantities and corresponding radiometric quantities (ASNT 2010).

Radiometric
Photometric Quantity Photometric Unity Radiometric Unit
Quantity
Color temperature kelvin (K) radiance temperature kelvin (K)
Illuminance lux (lx)b irradiance watt per square meter (W/m2)
Luminance candela per square meter (cd/m2)c radiance watt per steradian per square meter (W·sr–1·m–2)
Luminous energy lumen second (lm·s) radiant energy joule (J)
Luminous exitance lumen per square meter (lm/m2) radiant exitance watt per square meter (W/m2)
Luminous exposure lux second (lx·s) radiant exposure watt per square meter second (W·m–2·s)
Luminous flux lm (lumen) radiant flux watt (W)
Luminous intensity candela (cd)a radiant intensity watt per steradian (W/sr)
a. 1 cd = 1 lm/sr
b. 1 lx = 1 lm/m2
c. 1 cd/m2 = 1 lm·sr –1·m–2

VISUAL TESTING 5.5

 Table 4. Measurable characteristics of light, light sources and lighting materials.

Characteristic Unit Instrument

Light
Colora none spectrophotometer and colorimeter
Illuminance (flux density)b lux photometer
Polarization degreea percent (dimensionless ratio) polarization photometer
Polarization orientationa degree (angle) analyzing prism
Wavelengtha meter spectrometer

Light Sources
Angle, direction or zonal distributiona lumen or candela per direction goniophotometer
Color temperatureb kelvin colorimeter or filtered photometer
Energy radiateda joule per square meter calibrated radiometer
Luminanceb candela per square meter photometer or luminance meter
Luminous flux (light output)a lumen integrating sphere photometer
Luminous intensityb candela photometer
Power consumptionb watt watt meter, orvoltmeter and ammeterc
Spectral power distributiona watts per nanometer spectroradiometer

Lighting Materials
Optical density dimensionless number densitometer
Reflectanceb percent (dimensionless ratios) reflectometer
Spectral reflectance and transmittancea percent (at specific wavelengths) spectrophotometer
Transmittanceb percent (dimensionless ratios) photometer

a. Can be measured in laboratory.

b. Can be measuredin field or laboratory.
c. For direct current and for unity power factor
alternating current.

difference in hue and saturation between an object

Table 5. Recommended luminance ratios.
and its background.
Chromatic contrast may augment or detract from
Ratio Environment the achromatic luminance contrast, depending on the
perceptive ability of the eye to detect the colors
involved. Luminance contrast will generally provide
3 to 1 betweentestsurface and adjacent darker surfaces up to a 20 percent increase in contrast relative to
1 to 3 betweentest surfaceand adjacent lighter surfaces chromatic contrast. Quality of lighting or
20 to 1 between test surface and distant darker surfaces illumination in the inspection area refers to the
distribution of the light sources in the area and
1 to 20 betweentestsurface and distantdarker surfaces
implies that these devices aid visual performance and
comfort. Quality illumination is composed of area
lighting and specific test lighting. To avoid inspector
Table 6. Recommended illuminance.
eye fatigue and to enhance the probability of
detection due to size, the luminance ratios of the
Activity Illuminance (lx) inspection area should be controlled. Table 5 lists the
maximum recommended luminance ratio between
the test object and the environment and for different
General Lighting areas of the environment. This table recommends
Occasional visual tasks 100 to 200 maximum luminance ratios for areas where
Tasks with high contrast or large size 200 to 500 reflectance in the work area can be controlled when
Task Lighting control of the surrounding area is limited. Table 6
Tasks with medium contrast or small size 500 to 1000 lists recommended illuminance levels as a function
of the task.
Tasks with low contrast or very small size 1000 to 2000

5.6 AEROSPACE NONDESTRUCTIVE TESTING

 The measurement of light is photometry. The angle of peripheral vision is not a primary
Instruments used to measure light values are called consideration when performing detailed visual tests.
photometers. Measurement of visible light using a It is significant under certain inspection conditions;
photometer should correlate with the visual such as when surveying large areas for a
sensation of a normal human observer. To achieve discontinuity indication that has a high contrast
the aim of photometry, the characteristic of human ratio with the background and is observed to one
vision must be taken into account. A photometer side of the normal lens axis. The inspector‘s
should have a spectral response that matches that of attention is drawn to this area and it can then be
the human eye. Hand held photometers that provide scrutinized by focusing the eyes on the normal plane
digital measurements of the ambient direct light in of the lens axis. The viewer should in all cases
lux are called light meters when used for attempt to observe the target on the center axis of
photography. the eye. The angle of view should not vary more
than 45 degrees from normal. Viewing distance
varies with individuals and falls within a range of
30 to 60 cm (12 to 24 in.) from the object
Vision Acuity (Good 2004). The inspector should be positioned at
The identification of objects across the field of an angle not less than 30 degrees to the inspection
vision requires several different elements of surface as shown in Figure 6. The same principle
discrimination. In visual testing tasks, vision applies to objects being viewed through accessories
discriminates primarily three elements: space, such as mirrors or borescopes. The field of view
contrast, and color. should be maintained much in the same way that it
Vision acuity is the ability of the eye to resolve is when viewed directly. If the examination surface is
fine detail. It is usually measured by determining the immovable and situated so that the eye cannot be
smallest letter that can be correctly identified. An placed within this region, suitable visual aids, such
eye that can resolve a letter with only 5 minutes of as mirrors, must be used.
arc is said to have 6/6 (20/20) vision acuity,
expressed as a ratio of distances in meters (or feet).
The snellen fractions such as 6/6 or 6/12 (20/20 or
20/40) are used to express vision acuity. The Figure 6. Viewing angle recommendations for visual testing
numerator expresses testing distance at which an (ASNT 2010).
individual can distinguish adjacent objects while the
denominator is the distance at which the object
subtends 5 minutes of arc. The included angle of 5 Eye
minutes of arc is optimum, the average in which an
individual can encompass a sharp image (W. Bailey
1993; ASNT 2010).
Normal vision examinations typically include
vision acuity testing at both far and near distances.
Young workers who are well corrected for far vision
should have equivalent vision acuity measures for No closer than
the two test distances. If 6/6 (20/20) vision acuity is 150 mm (6 in.)
30 degrees 30 degrees
present at 3 m (10 ft) or greater, the same acuity
should be present for near working distances.
Workers under 40 years of age should have natural
focusing ability sufficient to get a clear retinal Test surface Test site
image, and hence, good vision acuity at normal near
working distances.
Viewing angle is an important consideration in
visual testing (W. Bailey 1993; ASNT 2010).

VISUAL TESTING 5.7

 Optical Techniques
PART3

The duty of visual inspection is the identification magnify a surface from 10×to several hundred times
of discontinuities in aerospace structures when magnification. The distance from lens to object is
viewed primarily at near and intermediate distances. adjusted until the object is in focus. As the
In addition to simple visual observation with the magnification becomes greater, the lens will be
human eye there are a number of enhancements that placed closer to the object and the field of view
can be employed to improve visual testing. Table 7 becomes smaller. At very high magnification, only a
lists some common approaches to visual testing and small area of an object can be inspected. This area is
comments on their use. known as the field of view. Also with magnification,
The most important prerequisite in visual testing the depth of field will be limited. Objects with highly
is that illumination must be sufficient for the proper varying surface features can be difficult to view at
detection and interpretation of visual indications. high magnification as the near and far portions of
Magnification, lighting aids, and orientation that the surface features will require different focus
provide measurable levels of sensitivity are critical to depths.
the proper evaluation of the area to be inspected. Mirrors can be used to provide viewing access to
Cleaning before visual testing is often necessary. the back of parts or in small regions, hard to access.
The dental mirror or mirrors that rotate on handles
are typically part of a visual inspector‘s tool kit.
Human Eye and Digital
Photographic Recording Digital Video Magnifiers
Visual testing is performed with adequate lighting Digital video imagers offer an easy method to
intensity and with the light at an angle to the view surfaces. Figure 7 shows an example of such as
surface. Normal inspection is performed within a system for hand held visual testing. The probe head
foot or two of the surface. For aircraft walk-around contains a light source and zoom magnification
inspection and in particular barely visible impact capability. Such systems can be used freehand or
damage detection, the viewing distance is nominally they can be mounted on fixtures.
1.5 m (5 ft). In addition to visual observation,
cleaning, moving, tilting, and adjusting of parts are
important in visual testing.
The visual observation of features for the Figure 7. Digital video imager.
detection of anomalies can be recorded using digital
photography. The ability to use digital imaging
systems to photograph conditions allows easy
documentation of features for reports. Measurements
are also important and can be documented in digital
image records. Including scales in the photographic
record is useful for reporting and recordkeeping. The
issues and techniques of photography for visual
testing are discussed in detail in the NDT Handbook,
third edition, Volume 9 on Visual Testing
(ASNT 2010). Digital imaging technology has
revolutionized visual testing practice and should be
implemented as part of routine visual inspections.

Magnifying Lenses,
Microscopes, andMirrors Borescopes
A borescope is an optical device resembling a
Magnification is used to improve the visibility of medical endoscope. It is a precise optical instrument
details on a surface. Single lens magnifiers provide with built-in illumination. Figure 8 shows a
between 1.5× and 10× magnification and may be borescope configuration and the physical parameters
hand held or fixed. The loupe is a term for some of that define it. It can be used to visually check
the simple magnifiers used in various industries. internal areas and deep holes and bores. Borescopes
Optical microscopes use a combination of lenses to are available in rigid and flexible models.

5.8 AEROSPACE NONDESTRUCTIVE TESTING

Rigid Borescopes designed to look in any direction from forward to
sideways to slightly backward. The tips may be at a
Rigid borescopes use a relay lens or fiber system fixed angle or interchangeable. There are many
inside a solid tube. The tube can vary widely in factors involved when selecting a borescope for an
diameter and length to suit the function. Commonly, application. In general, the largest diameter that the
the borescopes used are within the range of 4 to geometry of the application will allow should be
10 mm in diameter, but they can be much smaller as used. This provides for more light into the inspection
well as larger for longer length. The manufacturer area and a brighter, higher contrast image, presented
should be contacted for all the variations and back through the imaging system to the user. This
options in diameter and length that are available. allows smaller details to be resolved.
For viewing with a rigid borescope, the tip can be

Table 7. Visual testing methods.

Method Comments

Human eye and photography Use proper illumination.

Operator must have vision acuity.
Photographic recording.
Magnifying lens and mirrors Simple hand magnifiers.
Binoculars.
Microscopes.
Use mirrors for looking around and behind obstructions.
Borescopes Uses lenses in tubes to be inserted into locations with difficult access.
Tips for viewing at any angle can be attached.
Flexible fiber optic borescopes Uses fiber optics to allow flexible snaking of scope into tight locations for inspection.
Video borescopes Imaging chips can be at the tip of inserted scope.
Display is on video screen.
Stereo imaging.
Allows for image closeups.
Diffracted light Uses glancing light to highlight surface indentation through changes in low angle scatter.
Optical profiler Optical aid for scratch measurement that uses line source of light viewed from angle to
triangulate the depth and display on measurement scale.
Photogrammetry Uses multiple cameras to triangulate on surfaces for measurement of position, orientation, or
dimensions.
Interferometric fringe analysis Uses optical interference patterns to indicate small changes in surface under load.
Gloss Uses gloss meters to measure surface reflectance.
Color Uses color meters or spectrophotometers to measure surface color spectra.

Figure 8. Rigid borescope configuration and design parameters.

F

VA
DW

VC

Legend LT
DW = working diameter
LGA
LO = overall length LW
LT = tip length
LW = working length
F = angle of field
LGA = angle, light guide, axial
VA = angle of view, axial LO
VC= angle of view, circumferential

VISUAL TESTING 5.9

 Figure 9. Diagram of a representative flexible borescope (ASNT 2010).
Eyepiece Focusing ring
lens
Image
guide Objective lens

Light guide Protective sheath

Diopter ring
Light guide exit
Light source

Projection lamp

Flexible Borescopes Figure 10. Example of modern video borescope.

Fiber optics allows borescopes to be flexible.
Flexible fiber optic borescopes (also called
fiberscopes) permit manipulation of the test
instrument around corners and through passages
with several directional changes. Woven stainless
steel sheathing protects the image relay bundle
during repeated flexing and maneuvering. These
devices are designed to provide sharp, clear images
of components and interior surfaces normally
impossible to inspect. Remote end tip deflection
allows the viewer to thread the fiber optic borescope
through a complex series of bends. The end tip is
deflected with a rotating control mounted on the
handle. Most of these devices have a wide angle
objective lens that can provide a 100 degree field of
view. The lens also provides a depth of field that
allows placing the viewing tip very close to the
inspection area for a greater degree of magnification.
Tip deflection of ±90 degrees or greater allows for
scanning a wide area and looking around the Figure 11. Video probe: (a) tip; (b) attachments
contours of internal parts. Most fiber optic image (ASNT 2010).
bundles are equipped with a diopter ring at the
eyepiece to bring the subject into sharp focus over a
wide range of viewing distances. Figure 9 shows a (a)
diagram of a flexible borescope.

Video Borescopes
Video borescopes use miniature digital camera
chips embedded in the movable tip of flexible
borescopes. Illumination may be transmitted by fiber
optic or direct light emitting diode lighting while the
image is obtained at the tip by a camera chip and (b)
transmitted electronically to a display. The small Grabber
8 mm 127 µm
optical tips allow variable depths of field and magnet Brush comparison
magnification. Figure 10 shows an example of a measurement
Alligator gage
video viewing borescope system. These systems are
often called video probes or video scopes to Hook
distinguish them from fiber optic borescopes, which
they resemble. Figure 11 shows the tip and options
for tools that can be added to tips. The video
borescope method offers a bright, high resolution,
color image. Video viewing systems should be
checked before use with a resolution target to ensure
that there has been no imaging system degradation
and that there is no distortion or image artifacts that

5.10 AEROSPACE NONDESTRUCTIVE TESTING

would interfere with an evaluation. The device does
not have an eyepiece like other borescopes. Images Figure 12. Spatial frequency resolution test
are displayed on a viewing monitor. Most video standard; 1951 United States Air Force
borescopes incorporate a freeze frame feature that resolution test chart pattern conforms to
allows image capturing for closer viewing, saving MIL-STD-150A.
the image in an electronic medium. The image can
be electronically transferred for permanent
documentation. Grids or measurement references
may be entered into the margin of the image and
can become part of the permanent record. Software
assisted digital measurements can be taken and
recorded with some types of video borescope
systems. Many of these systems come with
interchangeable tips with fields of view from 60 to
120 degrees. Video borescope systems may also
include four-way articulation that can be designed
to articulate from 0 to as much as 160 degrees. A
wide angle of view tip allows scanning from a fixed
point across a plane of 360 degrees. The technology
for video borescopes is changing rapidly and
Figure 13. Enhanced surface inspection of
manufacturers should be contacted for details on
sizes, sensitivity, and cost. A number of aircraft lap joint: (a) photograph;
measurements such as dimension, depth, and area (b) edge-of-light image (ASNT 2010).
are possible with these systems (ASNT 2010,
148-149). (a)

Spatial Frequency Targets

Before using a borescope or any type of remote
visual testing system, it is important to have a
reference standard for visual sensitivity. Spatial
frequency resolution targets should be used to
establish the capability of the viewing system. (b)
Figure 12 shows an example of a resolution chart.
A resolution target provides a reference standard
that allows the comparison of imaging systems and
establishes a relative correlation with probable size
of discrepancy that can be reliably imaged.

Diffracted Light
When surfaces are viewed at shallow angles, the
light glancing off the surface is sensitive to small Figure 14. Setup for diffracted light method of visual testing
surface changes such as depressions or wrinkles. (ASNT 2010).
This diffracted or edge-of-light technique has been
developed for visualizing surface corrosion and Screen
impact damage effects on surfaces as small as
10 µm (0.0004 in.) (Komorowski 1991; Komorowski
2000). Raking light from an oblique angle creates
shadows from irregularities on the test surface.
Figure 13 shows the result of edge of inspection of
an aircraft lap joint. In Figure 13a the lap joint is
shown. In Figure 13b the corresponding edge-of- 90 degrees to screen
light scan shows the bright and dark regions present
Surface
as the surface deformation. This deformation implies
the potential for hidden corrosion between layers.
The optical setup for the diffracted light Lens
technique consists of a light source, a retroreflective 1.2 m (4 ft) Lamp
screen, and the object being inspected, as shown in
Figure 14. The surface being inspected must be
reflective. Both flat and moderately curved surfaces 30 degrees 1.2 m (4 ft)
can be inspected using this technique. The diffracted
light effect can be explained with geometric optics. 1.5 m (5 ft)
If a flat surface with an indentation is inspected, the
light striking the indentation is deflected. It then
strikes the retroreflective screen at a point removed
from the light rays reflected from the area

VISUAL TESTING 5.11

 surrounding the indentation. The retroreflective of diffracted light equipment have ranged from a
screen attempts to return all these rays to the points manual hand held system, in which the operator
on the inspected surface from which they were first directly views the inspected part, to systems with
reflected. However, the screen, consisting of video cameras and computer based image processors.
numerous glass beads, returns a cone of light to the
surface. This imperfection of the retroreflective
screen creates the diffracted light effect. By
backlighting the discontinuity, the technique Optical Profilers
increases the light intensity on one side of the Optical surface profiling instruments can be used
indentation and reduces it on the opposite side. As a to measure scratches, dents, or other surface
real time technique particularly applicable to rapid anomalies as an aid to visual testing. Figure 15
inspection of large surfaces, the diffracted light shows how these tools work (Jones 2008). The
technique has been used to inspect automobile body surface is illuminated by a line light source. The
panels and metal working dies. Commercial versions illuminated region is then viewed at an angle. Most

Figure 15. Laser profiler method for scratch or dent measurement (Jones 2008).

Sensor

Light source

Z-axis Receiver

Target

X-axis

Figure 16. Data trace across scratch for depth measurement.

–25 (–1)
Depth of scratch, µm (10–3 in.)

–50 (–2)

–75 (–3)

–100 (–4)

–125 (–5)

–150 (–6)

–175 (–7)

–200 (–8)
–2.0 –1.5 –1.0 –0.5 0 0.5 1.0 1.5 2.0
(–0.08) (–0.06) (–0.04) (–0.02) (0) (0.02) (0.04) (0.06) (0.08)
Distance across scratch, mm (in.)

5.12 AEROSPACE NONDESTRUCTIVE TESTING

instruments today use a camera to collect the data spray paint, the motion of the features can be
for display on a screen. When a scratch or dent is monitored by a pair of digital cameras during
encountered, the return light path is changed on the surface disturbance. The standoff distance and
screen and so the depth of the feature can be camera-to-camera separation determine the field of
measured. Figure 16 shows an example of a data view and spatial resolution. The fields of view are
output of depth as a function of linear position. The solely optics dependent.
technique can measure scratches down to 10 μm Photogrammetry is capable of performing
(0.0004 in.). measurements from submicron volumes (crystalline
structures) (Woodard 2008) to hundreds of meters
(wind turbines and bridges) (Tyson 2008).
Calculation of the movement of the feature spots by
Photogrammetry triangulation measurements in real time allows
Photogrammetry (and digital image correlation) is display of in-plane and out-of-plane displacement
the science of obtaining quantitative measurements for the full field of view. Typically, the undeformed
of physical objects through the process of recording, reference measurement image is compared to the
measuring, and interpreting photographic images deformed images, and a correlation algorithm
(ASPRS 2004; Linder 2009; Luhmann 2006; measures the relative movement and shape change
Mikhail 2001). The process uses multiple cameras of the random or regular pattern that had been
(usually two) to view objects for the triangulation of applied to the surface of the test object
the location. The data can provide accurate (Avitabile 2010). The result is the calculated local
positional information and surface displacement displacements and surface strain. The displacement
measurements. measurement noise is on the order of 1/30 000 the
One particularly useful implementation of digital field of view or about 33 µm/m. The surface strain
image correlation photogrammetry for aerospace is noise is on the order of 100 µm/m, independent of
surface strain measurement (Schmidt 2003). By the field of view. Such measurements of surface
marking a surface with an array of features (random strain values provide quantitative measures for stress
or ordered), such as those left by a light sputtering of analysis and nondestructive testing of structures
under loading conditions such as mechanical testing,
pressurization, heating, or cooling. Figure 17 shows
an example of a full field strain measurement
Figure 17. Photogrammetry for surface photogrammetry system. The resulting surface strain
displacement measurement: (a) dual-camera maps on a motor segment are shown in Figure 17b.
system with illumination lights; (b) display of Such measurements are usually used to correlate
surface displacement model response from point with structural models, but they will indicate
vibration source. anomalies if the test structure strains in a manner
(a) inconsistent with a good structure.

Grid and Moire Measurements

Light is a wave phenomenon that is subject to
diffraction and interference effects. Moire
measurement is a nondestructive method for full
field deformation measurement and strain analysis.
The beating of two superimposed gratings (that is,
the animated variance between them), one being
deformed and the other a reference, is used to detect
surface variations when a load is applied. This
process can detect surface and subsurface damage by
variations in the fringe patterns that result from the
grating interference effects. The moire method is
(b) discussed in detail in the Nondestructive Testing
Handbook, Volume 9, Special Nondestructive Testing
Methods (ASNT1995).

Visual Testing Standards

Although visual testing is the oldest test method,
its practice was not widely standardized before 1993
(ASNT 2010, vi). Some published standards for visual
testing are listed in Table 8.

VISUAL TESTING 5.13

 Table 8. Some standards for visual testing.

Organization Standard

NAS 410, National Aerospace Standard [for] Certification and Qualification of

Aerospace Industries Association Nondestructive Test Personnel
ANSI B3.2, Rolling ElementBearings— AircraftEngine, Engine Gearbox, and Accessory
American National Standards Institute Applications—SurfaceVisualInspection (1999)
API SPEC 5CT [ISO 11960], Specification for Casing and Tubing (2006)
American Petroleum Institute
API STD 5T1, Imperfection Terminology (2003)
API STD 653, Tank Inspection, Repair, Alteration, and Reconstruction (2008)
API STD 1104, Welding of Pipelinesand Related Facilities (2005)
ASNT Recommended Practice No. SNT-TC-1A: Personnel Qualification and Certification in
American Society for Nondestructive Testing Nondestructive Testing
AWS B1.11, Guide for the Visual Examination of Welds (2000)
American Welding Society AWS D1.1M, Structural Welding Code — Steel (2008)
AWS D8.1M, Specification for Automotive Weld Quality — Resistance Spot Welding of Steel
(2007)
AWS D18.2, Guide to Weld Discoloration Levels on Inside of Austenitic Stainless Steel Tube
(1999)
AWS G1.6, Specification for the Qualification of Plastics Welding Inspectors for Hot Gas, Hot
GasExtrusion, andHeatedToolButtThermoplasticWelds (2006)
AWS QC1, Standard for AWS Certification of Welding Inspectors (2007)
ASME Boiler and Pressure Vessel Code: Section V, Nondestructive Examination. Article 9,
Visual Examination (2009)
ASME Boilerand Pressure Vessel Code: Section XI, Rules for Inservice Inspection of Nuclear
Power Plant Components. Article IWA-2000, “Examination and Inspection”
ASME International ASTM A 802M, Standard Practice for Steel Castings, Surface Acceptance Standards, Visual
Examination (2006)
ASTM D 2562, Standard Practice for Classifying Visual Defects in Parts Molded from
Reinforced Thermosetting Plastics (2008)
ASTM D 2563, Standard Practice for Classifying Visual Defects in Glass-Reinforced Plastic
ASTM International Laminate Parts (2008)
ASTM D 4385, Standard Practice for Classifying Visual Defects in Thermosetting Reinforced
PlasticPultrudedProducts (2008)
ASTM E 1799, Standard Practice for Visual Inspections of Photovoltaic Modules (1999)
ASTM F 584, Standard Practice for Visual Inspection of Semiconductor Lead-Bonding Wire
(2006)
ASTM F 1236, Standard Guide for Visual Inspection of Electrical Protective Rubber Products
(2007)
IPC-OI-645, Standard for Visual Optical Inspection Aids. (1993)
CGA C-13, Guidelines for Periodic Visual Inspection and Requalification of Acetylene
Cylinders (2006)
CGA C-6, Standards for Visual Inspection of Steel Compressed Gas Cylinders (2007)
CGA C-6.1, Standards for Visual Inspection of High Pressure Aluminum Compressed Gas
Cylinders (2006)
CGA C-6.2, Guidelines for Visual Inspection and Requalification of Fiber Reinforced High
Pressure Cylinders (2005)
CGA C-6.3, Guidelines for Visual Inspection and Requalification of Low Pressure Aluminum
Association Connecting Electronics Industries Compressed Gas Cylinders (1999)
Compressed Gas Association CGA C-6.4, Methods for External Visual Inspection of Natural Gas Vehicle (NGV) and
HydrogenVehicle(HV)FuelContainersandTheirInstallations (2007)
EIA 699, Test Method for the Visual Inspection of Quartz Crystal Resonator Blanks (1997)
EN 970, Non-Destructive Examination of Fusion Welds— Visual Examination
EN 1330-1, Non-Destructive Testing — Terminology — Part 1, List of General Terms
EN 1330-2, Non-Destructive Testing— Terminology— Part 2,Terms Common tothe Non-
Destructive Testing Methods
EN 1330-10, Non-Destructive Testing — Terminology — Part 10, Terms used in Visual Testing
EN 1370, Founding — Examination of Surface Condition
EN 3841-201, Circuit Breakers — Test Methods — Part 201, Visual Inspection (2005)
EN 4179, AerospaceSeries— Qualification andApproval of Personnelfor Non-Destructive
Testing
EN 12454, Founding — Visual Examination of Surface Discontinuities — Steel Sand Castings
Electronic Components Industry Association EN 13018, Non-Destructive Testing — Visual Testing — General Principles (2007)
European Committee for Standardization
continued on next page

5.14 AEROSPACE NONDESTRUCTIVE TESTING

Table 8. Some standards for visual testing (continued).

Organization Standard

EN 13100-1, Non Destructive Testing of Welded Joints of Thermoplastics Semi-Finished

Products—Part1,VisualExamination (2000)
EN 13927, Non-Destructive Testing — Visual Testing — Equipment
FAA AC 43-204, Visual Inspection for Aircraft (1997)
Federal Aviation Administration IEC 60748-23-2, Semiconductor Devices — Integrated Circuits — Part 23-2, Hybrid
International Electrotechnical Commission Integrated Circuits and Film Structures — Manufacturing Line Certification – Internal
Visual Inspectionand Special Tests (2002)
ISO 580, Plastics Pipingand Ducting Systems—Injection-Moulded Thermoplastics Fittings—
Methods for Visually Assessing the Effects of Heating (2005)
International Organization for Standardization
ISO 3058, Non-Destructive Testing— Aids to Visual Inspection — Selection of Low-Power
Magnifiers (1998)
ISO 9712, Non-Destructive Testing — Qualification and Certification of NDT Personnel
ISO 11960 [API SPEC 5CT], Petroleumand Natural Gas Industries — Steel Pipes for Use as
Casing or Tubing for Wells (2006)
ISO 17637, Non-Destructive Testing of Welds — Visual Testing of Fusion-Welded Joints
(2003)
JIS H 0613, Non-Ferrous Metalsand Metallurgy— Visual Inspection for Sliced and Lapped
SiliconWafers (1978)
JIS H 0614, Non-Ferrous Metalsand Metallurgy— Visual Inspection for Silicon Wafers with
Specular Surfaces (1996)
Japanese Institute of Standards JIS Z 3090, Visual Testing Method of Fusion-Welded Joints (2005)
MSS SP-55, Quality Standard for Steel Castings for Valves, Flangesand Fittings and Other
Piping Components —Visual Method for Evaluation of SurfaceIrregularities (2006)

Manufacturers Standardization Society

VISUAL TESTING 5.15

 Applications
PART4

What Can Go Wrong? cracks may occur in both the web and doubler or in
each member separately. The cracks originate at the
When performing visual testing, operators should lower edge of both members shown in Figure 18.
be concerned about the issues that are identified in Direct access to the cracked areas of the web and
Table 9. doubler requires removal of the trapezoidal fitting. If
both members are cracked, fuel may leak from under
the fitting, indicating through-thickness cracks.
Aircraft Other methods of nondestructive testing cannot be
The visual test is a viable and economical method used because of the poor access caused by the
that helps monitor the structural integrity of fitting. The doubler and web are made from clad,
in-service aircraft. Magnifiers and borescopes are wrought aluminum sheet about 4 mm (0.16 in.)
typically used to detect corrosion and cracks where thick. Access to the area is aft of the rear spar,
access is limited. inboard and outboard of the trapezoidal panel and
fitting.
The visual test can be performed with a 0 degree
Wing Cracks under Panel borescope along with a 70 to 90 degree borescope,
Fatigue cracks, for example, may occur in the 300 to 480 mm (12 to 19 in.) in length, and 4 to
wing rear spar cap web and doubler under the 5 mm (0.16 to 0.20 in.) in diameter. The test area is
trapezoidal panel attachment fitting. These fatigue cleaned using a cotton swab wetted with solvent.

Table 9. What can go wrong with visual testing.

Issue Comments

Access Clean the surface for access to the features of interest.

Use a mirror to look around corners or behind barriers.
Obtain proper size borescope.
Flexibleborescopescan be articulated to lookback onthemselvesandoffer considerable
enhancement to access.
Observation Posture affects the way an object is observed.
Appropriate posture and viewing angle minimize fatigue, eyestrain, and distraction.
The viewer’s posture should make it easy to maintain the optimum view on the axis of the lens.
Scattered light On reflective backgrounds, the viewing angle should be off normal but not beyond 30 degrees.
This angle is maintained so that the light reflected off the surface is not directed toward the
eyes, reducing the contrast image of the surface itself.
Glare Glare reduces visibility and visual performance:
(1) decrease the intensity of the light source,
(2) reduce the area of the light source,
(3) increase the angle between the light source and the field of view, or
(4) decrease reflections by using a light source with a larger area and lower luminance.
Low contrast Provide adequate illumination.
Alter the angle of viewing.
Interpretation of depth Lighting angle can improveinterpretationofdepth.
Stereo viewing in fiberscopes can be an aid.
Apparent image quality Use enhancements such as optical profiling to measure feature depth.
Viewingsystemslikeborescopesandvideoscopesshouldbetested against areference standard
for resolution. Apparent image quality can be deceiving without a test process for a
quantifiable measurement. An image of the area of interest can look acceptable but not have
adequate sensitivity to resolveadiscontinuity. Glare reduction, for example, can reduceimage
sensitivity.

5.16 AEROSPACE NONDESTRUCTIVE TESTING

The area is first viewed using the 0 degree borescope
to check for cracks in the radius areas of the doubler Figure 18. Drawings from industry specification (Pratt 1995) for wing
and web (Figure 18, position 1). If a crack exists and rear spar doubler and web crack location: (a) from above; (b) from side.
it does not run under the fitting, its length in the
doubler may be determined by a liquid penetrant or (a)
high frequency eddy current test using a shielded
surface probe. To measure the length of a crack in Forward Tee cap
the web (foremost member) under an uncracked
doubler requires removal of the fitting and
application of a low frequency eddy current test. Rear spar web
Detecting a crack in the doubler or web hidden by Inboard
the fitting (Figure 19b, positions 2 and 3) requires
the 70 or 90 degree borescope. The area is cleaned
with a cotton swab wetted with solvent, and the
borescope is inserted through the small opening Doubler or Fitting
between the forward side of the fitting and the aft web crack
side of the doubler as shown in Figure 19. To Doubler
measure the length of doubler cracks under the
(b)
fitting at position 2 or 3 requires ultrasonic
Grabber
transverse wave (angle beam) techniques. Up

Rudder Flange Cracks

Radiographic tests can reveal cracks developing
in the rib flanges of the rudder. In one case, analysis Inboard
determined that cracking of the rib flanges resulted
from acoustically induced vibration. It was also
determined that installation of stiffeners on the
rudder ribs strengthens the rudder and inhibits
further cracking. A service bulletin was issued giving
criteria for flyable crack lengths based on the Crack position 1 Edge of doubler and web
number of cracked ribs and the length of the cracks.
Unrepaired rib flange cracks may cause cracks in the
rudder skins, thereby requiring more extensive
repairs. Figure 19. Borescopic visual test of wing rear spar doubler and web for
A radiographic test is first conducted and if cracks under fitting: (a) view looking inboard; (b) view looking forward
cracks are detected in the rib flanges at or adjacent at lower portion of rear.
to the skin attachment fastener holes, their lengths (a)
must be determined. The cracks may run upward Trapezoidal panel
into the flange radius and progress into the rib area Rear spar
as shown in Figure 20. Their lengths may be difficult Up
to determine from radiologic images alone. If cracks
occur within or progress into the flange upper
radius, their lengths must be determined by use of a
3.2 mm (0.125 in.) maximum diameter rigid Forward
borescope or flexible borescope.
To perform this test, the fastener common to the
rudder skin and the rib flange (opposite the cracked
flange) must be removed. A 0 degree borescope is
inserted through the open hole, and the opposite Cracks
flange radius and web are inspected to determine Borescope
crack position and length. An alternative approach Wing skin
requires removal of a fastener common to the rudder
skin and the cracked flange judged to be about Trace
50 mm (2 in.) beyond the crack image on the (b) Fitting
radiograph. A flexible borescope or rigid retrograde
borescope is inserted through the open hole and
articulated to allow viewing of the rib flange, radius,
and web to determine crack position and length. If
cracks exceed flyable length, the ribs are repaired. If
cracks are of tolerable length, an easily replaced Up
fastener is installed in the open holes to allow for
repetitive evaluation.

Inboard
Landing Gear Pitting Corrosion Crack position 2
Borescope Crack position 3
Main landing gear truck beam assembly failures
resulted in major secondary damage to an aircraft.
Investigation revealed that failure was a result of

VISUAL TESTING 5.17

 Figure 20. Determining crack length in rudder ribs by visual test: (a) from side; (b) from above.
(a)

Rigid
Flange upper borescope
radius
Typical rib web
D
Fiber optic
(b) Fiber optic light source
light source

Rudder skin

D
Typical
Flange upper cracks
radius Flexible
borescope
Rigid
borescope Cracked Flexible
flange borescope
Up

stress corrosion fracture that initiated next to the borescope and a 70 or 90 degree (lateral) borescope
intersection of the lubrication hole and the pivot to evaluate whether the component can continue in
bore shown in Figure 21. The stress corrosion service at a reduced inspection cycle or be removed
fracture in the lubrication hole results from severe and repaired. Beams showing slight corrosion may
pitting caused by inadequate lubrication, which continue in service for a limited time provided that
could result in part failure. Removing corrosion or periodic borescope and ultrasonic transverse wave
pitting from the surface of the four pivot bore tests are made to detect possible stress corrosion
lubrication holes and increasing the frequency of cracks. Figure 22a shows an acceptable condition at
lubrication minimizes the possibility of failure and the inner bore chamfer. Figure 22b shows a typical
extends the service life of the beam assemblies. heat treat pit — an acceptable condition away from
Inservice testing for corrosion pitting requires the inner bore chamfer. Figure 22c shows
removal of the lubrication fitting and grease from unacceptable corrosion pitting at the inner pivot
each of four holes in the subject beam. The internal bore intersection. Figure 22d shows acceptable
surface of each bore is checked using a 0 degree machining marks (rifling) in the bore outer edge.
(forward looking) 2.8 mm (0.11 in.) diameter

Figure 21. Case history of corrosion leading to component failure: (a) main landing gear truck beam;
(b) fractured surface of failed component caused by failure to apply lubricant. Lubrication hole is visible
opening on inside surface.
(a) (b)

Lube holes

5.18 AEROSPACE NONDESTRUCTIVE TESTING

Wing Spar Cap Cracks X-ray, ultrasonic transverse wave, and other tests.
The ultrasonic and borescope tests are made through
The purpose of this test is to check a spar cap an access hole in the wing leading edge, as shown in
tang for fatigue cracks at the fastener locations. Figure 23, and a smaller access hole in the footstool
The total inspection for fatigue cracks includes fitting, so that the fasteners under the fitting can be

Figure 22. Surface conditions possible in typical main landing gear truck beam: (a) acceptable bore
chamfer; (b) typical heat treat pit; (c) severe bore chamfer corrosion; (d) machining marks (rifling).

(a) (c)

(b) (d)

Figure 23. Borescopic visual test of wing lower forward spar cap tang for fatigue cracks.
Footstool fitting
Forward
Footstool fitting

Pylon fitting
Inboard
Fastener

Typical crack Spar cap Aft

forward edge
Access hole Flexible borescope
Inboard

Spar forward tang

Access hole
Inboard side of footstool fitting
J tube guide
Pylon fitting and footstool fitting left side
Light source

VISUAL TESTING 5.19

 visually inspected. Inspection of the six fasteners at rolling marks, are found first in the manufacturing
the inboard side of the fitting is accomplished using process; other anomalies, such as corrosion and
a flexible borescope. fatigue damage, may be found by maintenance
The inspector positions the tip of the 90 degree inspectors during routine inspections.
flexible borescope forward and aft of each fastener The operating environment of engines is a factor
to detect fatigue cracks in the spar cap forward tang. to be taken into account during inspection. Many
The area under the footstool fitting is inspected engine parts are expected to survive intense heat.
using two techniques: (1) the seven fastener Some superficial clues to engine condition are leaks
locations are inspected using a 0.6 m, 90 degree, and evidence of smoke. Peening tends to remove
rigid borescope or a flexible borescope supported evidence of heat discoloration and to smooth
by semirigid plastic tubing, and (2) the forward edge damage, making it less apparent.
of the spar cap is inspected with a flexible borescope
in a semirigid tube bent into a J shape shown in
Figure 23. These tests have been successful in Compressor and Turbine Blades
detecting small cracks at the forward and aft sides Blades are evaluated from a standpoint of
of the fasteners and larger cracks that propagate structural integrity. The blade is engineered to
to the leading edge (forward) or vertical leg (aft) transfer combustion energy to the engine compressor
of the cap. (providing mostly air as the driving force) and, in the
process, experiences severe thermal and mechanical
loading. In an aerospace engine, the material
integrity of turbine blades must be high to sustain
Engines the service life. All blade surfaces must be smooth,
Details observed in engine inspection differ from and all evidence of previous leading and trailing
airframe inspection and generally relate to surface edge repairs should be smoothed, or blended, with a
finish and the means for inferring underlying minimum edge specified for that component.
conditions. Inspections within protective cowls and Blades may be damaged by impact with foreign
housings often require complex visual aids such as objects. The use of a substantial number of blades
light guides, borescopes, and special inspection that have been repaired to or near maximum limits,
fixturing. The engine components have surface or the use of blades that have many repaired areas,
textures different from that of the rolled and formed may adversely affect compressor efficiency and
surfaces in airframes and so are inspected with therefore engine performance. If two or more
different criteria. Many engine anomalies are related blended blades are found grouped, an attempt should
to surface features of the material, and engine be made to redistribute these blades evenly
manuals describe visible surface characteristics in throughout the engine.
terms of the process that caused them and the At the same time that blades are checked for
process‘s effect on satisfactory performance of the foreign object damage, they should be visually tested
part. Some anomalies, such as forging cracks or for arc burn. Arc burn is evidenced on the blade
surface by a small circular or semicircular heat
affected area that may contain shallow pitting,
remelting, or cracking. Blades with arc burn are not
Figure 24. Borescopic access ports, right side. acceptable for further service.
Blending is intended to ensure that all subsurface
as well as surface damage is removed. It is critically
important that the blending procedures be followed
carefully whether in the shop or on the wing. Proper
blending must be performed in successive stages:
233 degrees 215 degrees
Second stage vane removal of the visible damage, confirmation by eddy
Fifth stage stator current or fluorescent penetrant testing that all
surface damage has been removed, and finally
removal of additional material to ensure the removal
of subsurface deformed material.
After refurbishment or repair, the blade should be
visually tested with bright light, mirrors, a 3×
magnifier, fingertips, and fingernails to look
specifically for small depressions in the surface
contour. Observed conditions may be compared with
standard documentation and reference photographs.
Diffuser case
Access ports

Fourth stage stator

338 degrees Borescopy
(bleed valve opening) 277 degrees
Borescopic testing of the engine is a significant
Sixth stage stator 225 degrees aid to effective maintenance. Borescope access ports
are incorporated at many locations along the engine
Tenth stage stator
245 degrees gas path, allowing detailed examination of critical
internal engine areas as shown in Figure 24.
231 degrees 214 degrees

5.20 AEROSPACE NONDESTRUCTIVE TESTING

 Indirect visual testing with video borescopes or Straight-on lighting may be needed to see the
other electric viewing devices must be performed in features at the bottom of a void; oblique lighting
a protected area. In wet weather, precautions must may be needed to distinguish between a protruding
prevent damage to equipment or electric shock to the or concave surface feature.
operator. Also, operation of visual test equipment Because of heterogeneity, it is important to
beyond its recommended limits may be economically choose an appropriate scale factor for viewing
unjustifiable because of impaired performance or composite materials with magnification. Many visual
reduced servicelife. tests are performed with 5× to 10×power. This
The regular inspection interval is part of the relatively low power evaluation is a balance of close
written maintenance procedure. A reduced viewing of the small surface imperfections and
inspection interval may be half that value. Visual retention of the ability to view a significant
test intervals may be adjusted on the basis of surrounding area for perspective and comparison.
maintenance protocols and of consultation with Figure 25 shows a questionable area on a cross
the local regulatory agency. section of an aramid polyester composite. Without
Engines get very hot. Temperatures of parts to be proper lighting and scale factor, it appears that the
borescopically tested should be permitted to cool area in the center of the figure has two matrix cracks.
after engine operation. Damage may result to Figure 26 shows the same area, with higher
borescopic equipment if equipment is exposed to gas magnification and with better lighting. The two lines
path or metal temperatures above 66 °C (150 °F). A are not matrix cracks. Figure 27 shows actual matrix
rule of thumb is that, if adjacent cases are too hot to cracks (notice the prominentcrevice).
touch, then visual testing should be delayed.
Borescopic equipment for visual testing should
satisfy the standards, procedures, and specifications Figure 25. Possible matrix cracks in carbon fiber reinforced polymer
in force for the engine and workplace. Equipment material (ASNT 2010).
and associated hardware includes power source, light
cables, and adapter required during borescopy. As an
example, low magnification borescope resolution
measured at a 25 mm (1 in.) distance from the target
should be 16 line pairs per millimeter, while high
magnification borescope resolution should be 36 line
pairs per millimeter. Optional borescope equipment
enhances the test capability and eases the task for
the inspector. These items range from clamping
equipment to hold the borescope in position to
optical video equipment to record the visual test for
later review.
For some engines, full 360 degree borescopic
testing of the high pressure rotor can be
accomplished by rotor cranking of the gearbox. A
pneumatic motor actuated by foot or hand will free
the operator to use both hands to position and adjust
the borescope. It is recommended that rotation be 1 mm (0.04 in.)
stopped as each rotor blade is positioned properly for
inspection. Tilting will allow thorough appraisal of
each blade‘s condition.
Figure 26. Magnified area from Figure 25, showing coloration change
only and no cracks (ASNT 2010).
Composites
Visual testing of carbon fiber reinforced polymer
composite materials is more difficult than that of
most other engineering materials because the carbon
fiber reinforced polymer is heterogeneous (Bailey
1993). A perfectly good structure may have
variations in resin, causing color variations on the
exterior surface, surface roughness, or waviness due
to woven materials used to apply pressure during the
layup or due to benign surface features such as small
voids or wrinkles.
Preparation of the part to be viewed is important
in visual testing of composites. Often, structures
cannot be viewed immediately after manufacture.
Edges and bag side surfaces must be trimmed
extensively to remove excess resin. After trimming
operations there may be dust and residue on the
surfaces. Sanding or polishing of the edge may be
necessary if the edge of the structure must be viewed.
After the cleaning, good lighting is essential.

VISUAL TESTING 5.21

 Because a visible light image can be readily
Figure 27. Actual matrix cracks (ASNT 2010). viewed, documentation of an evaluation is important.
Often the visual test will be requested and
documentation required because of an anomaly
detected with another form of nondestructive testing.
Low magnification imaging is useful to document
surface discontinuities as shown in Figure 28.
Microscope photography is useful in imaging
discontinuities that have been cross sectioned and
polished as shown in Figure 29.
Visual testing is the first evaluation a carbon
fiber reinforced polymer structure receives after
manufacture. This may be simply a quick evaluation
of the condition of the structure after removal from
the autoclave. After removal of the tooling and
bagging materials, the surfaces of the component
may be viewed for excessive wrinkles, holes, gaps
between adjacent plies, surface voids, cracks,
buckling, and other conditions. This visual check is
normally done without magnification. Just as the
actual part itself is viewed, the tooling and bond
forms may be viewed for discontinuities and
alignment problems. These evaluations are normally
performed by manufacturing personnel. A more
thorough evaluation may be required by engineering
Figure 28. Voids exposed on composite surface (ASNT 2010).
personnel. This may entail looking for resin rich or
resin starved areas, blisters or delaminations in the
exterior plies, surface voids, and edge separations. If
the matrix material is semitransparent, this
evaluation may also include looking for internal
voids by shining a high intensity light through the
structure and viewing in a darkened area.
Visual testing is performed on large aerospace
structures to identify damage at the visible impact
damage detection level. Visual testing is the first
opportunity to detect damage to aircraft from hail
impact, runway debris, or lightning strikes. Looking
at the structure for dents, buckled areas, or
discoloration is part of preflight inspection.
Depending on the size of visible damage, more
extensive nondestructive inspections may be applied
25 mm (1 in.) to determine the required repairs. Some aircraft
radomes (antenna housings) have had excessive
static discharges that have resulted in burns to the
composite. These burns are visually detected and
Figure 29. Voids in cross section of carbon fiber reinforced polymer then the composite structure is measured by
(ASNT 2010). ultrasound and radiography. These visual evaluations
often are required at fixed inservice intervals. The
visual indication is the first sign of something wrong
that the follow-up nondestructive testing reveals.

5.22 AEROSPACE NONDESTRUCTIVE TESTING

 Magnifiers, light sources, inking dyes, and
wetting agents such as alcohol are all suitable tools Figure 30. Dye enhanced surface cracks (ASNT 2010).
for performing visual testing on composite materials.
A common light source is a small, bright flashlight.
It is portable and its beam is easily directed to the
viewing area. The alcohol is used as a visible
penetrant to enhance the contrast on edge
discontinuities and voids. After wiping the alcohol
onto the composite surface, the surface will dry very
quickly whereas the alcohol trapped in cracks or
voids will take longer to dry. The dark marks go
away as the alcohol evaporates. Inking dyes are
useful in imaging small surface breaking cracks or
voids, particularly when trying to document the
cracks with photography. A small amount of dye is
poured onto a cloth and wiped over the area. After
the residual surface dye is wiped away, some dye
stays trapped in the cracks, contrasting them against
the background color. Figure 30 shows surface
cracks in polyester resin after an impact damage
event using dye enhancement. Of course, contacting
materials must be compatible so as not to trigger
material degradation.

10 mm (0.4 in.)

VISUAL TESTING 5.23

References
ASNT. 1995. "Grid and Moire Nondestructive Testing." Nondestructive Komorowski, J.P., D.L. Simpson, and R.W. Gould. 1991. ―Enhanced
Testing Handbook 9, Special Nondestructive Testing Methods, 2nd Visual Technique for Rapid Inspection of Aircraft Structures.‖
edition. Columbus, OH: American Society for Nondestructive Materials Evaluation 49(12): 1486-1490.
Testing, 19-93. Krauss, D.G. 1998. ASNT Level II Study Guide: Visual and Optical
ASNT. 2010. Nondestructive Testing Handbook 9, Visual Testing, 3rd Testing Method. Columbus, OH: American Society for
edition. Columbus, OH: American Society for Nondestructive Nondestructive Testing.
Testing. Linder, D. 2009. Digital Photogrammetry: A Practical Course. New York,
ASPRS. 2004. Manual of Photogrammetry. 5th edition. Bethesda, MD: NY: Springer.
American Society for Photogrammetry and Remote Sensing. Luhmann, T., S. Robson, S. Kyle, and I. Harley. 2006. Close Range
Avitabile, P., C. Niezrecki, M. Helfrick, C. Warren, and P. Pingle. 2010. Photogrammetry: Principles, Methods and Applications. Dunbeath,
―Noncontact Measurement Techniques for Model Correlation.‖ Caithness, United Kingdom: Whittles.
Sound and Vibration (January): 8-12. Mikhail, E.M., J.S. Bethel, and J.C. McGlone. 2001. Introduction to
Bailey, D. 1993. ―Visual Testing of Composite Materials.‖ Nondestructive Modern Photogrammetry. New York, NY: Wiley.
Testing Handbook 8, Visual and Optical Testing, 2nd edition. Pratt. 1995, rev. 2005. CTE 6181, Borescope Kit Assembly. Hartford, CT:
Columbus, OH: American Society for Nondestructive Testing, Pratt and Whitney.
328-330. Sayler, G.C. 1998, revised 2006. ASNT Level III Study Guide: Visual and
Bailey, W.[H.] 1993. ―Vision and Light.‖ Nondestructive Testing Optical Testing Method. Columbus, OH: American Society for
Handbook 8, Visual and Optical Testing, 2nd edition. Columbus, Nondestructive Testing.
OH: American Society for Nondestructive Testing, 9-21. Schmidt, T., J. Tyson, and G. Konstantin. 2003. ―Full-Field Dynamic
FAA 43-204. 1997. Visual Inspection for Aircraft. Washington, DC: Displacement and Strain Measurement Using Advanced 3D Image
United States Government Printing Office, for Federal Aviation Correlation Photogrammetry.‖ Experimental Techniques, Part 1,
Administration. 27(3): 47-50. Part 2, 27(4): 44-47.
Good, G.W., J.J. Nichols, M.V. Subbaram, V.B. Nakagawara, and R.W. Tyson, J., T. Schmidt, A. Gilat, A. Walker, and J. Seidt. 2008. ―3D Image
Montgomery. 2004. ―An Analysis of the Visual Demands Associated Correlation Studies of Geometry and Material Property Effects
with Aviation Maintenance Inspectors.‖ Washington, DC: Federal During Split Hopkinson Bar Experiments.‖ Proceedings of the XIth
Aviation Administration. International Congress and Exposition [Orlando, FL, June]. Bethel,
Jones, C. 2008. "Non-Contact Displacement Sensor Technology." Power CT: Society for Experimental Mechanics.
in Motion (October): Mechelen, Belgium: Thomas Industrial Woodard, N., E. Johnson, J. Demers, C. Chen, J. Nielsen, and R. Logan.
Media, 12-13. 2008. ―Low-Cost Electrical Resistance Spot Welding Technique for
Komorowski, J.P., and D.S. Forsyth. 2000. ―The Role of Enhanced Visual Micro-Optical Component Mounting with Sub-Micron Tolerance."
Inspections in the New Strategy for Corrosion Management.‖ Pasadena, CA: Phasebridge.
Aircraft Engineering and Aerospace Technology 72(1): Bradford,
West Yorkshire, United Kingdom: Emerald, 5-13.

5.24 AEROSPACE NONDESTRUCTIVE TESTING

 X
6
CHAPTER
CHAPTER

Liquid
Penetrant
Testing
Contents
Part 1. Introduction, 6.2
Part 2. Penetrant Principles, 6.3
Part 3. Penetrant Techniques, 6.5
Part 4. Safety, 6.9
Part 5. Standards and
Specifications, 6.10
Part 6. Applications, 6.11
References, 6.16

Contributors
Charles W. Eick
6.1
Lisa Brasche
 Introduction
PART1

Penetrant testing is one of the oldest discontinuities to the surface. Inservice inspection of
nondestructive test methods. The earliest use was on many aerospace components, especially jet engine
iron railroad wheels and was referred to as the oil parts, includes penetrant testing to detect the onset
and whiting process. Since those very early uses, of surface breaking fatigue cracks that can develop
penetrant testing has evolved into a very sensitive during service.
and useful process for detecting surface Two methods of penetrant testing are available:
discontinuities. fluorescent and dye. Fluorescent penetrant testing is
Penetrant is used in all aspects of aerospace the required method of choice for most aerospace
products, such as raw castings, forgings, and applications, in particular jet engine applications.
weldments. The primary applications are after Visible (dye) penetrant is less sensitive than
machining processes and in maintenance and fluorescent penetrant and therefore is prohibited on
overhaul facilities for detection of service induced most aircraft applications but is often used on welds
cracking. This inspection checks for cracks that may for space launch vehicles, spacecraft, and ground
be induced by machining or the opening of internal based equipment.

6.2 AEROSPACE NONDESTRUCTIVE TESTING

 Penetrant Principles PART2

Physics penetrant entering these areas first. The amount of

liquid entering the crack is determined by a balance
Penetrant testing is based on the physics of of forces for any trapped gas pressure in the crack
surface tension and capillary action. There are other and the tension forces pulling in the penetrant. If the
factors that affect the ability of penetrant to enter a crack is too small, there may not be a sufficient
crack, but surface tension is a major factor. The key volume of penetrant that enters the discontinuity for
element of a good penetrant is the ability of the subsequent detection. A more detailed description of
liquid to wet the part. If the penetrant can wet the the physics principles that govern penetrant testing
part, then the surface tension of the liquid-to-gas can be found in the Nondestructive Testing
interface is lower than the surface tension of the Handbook 2: Liquid Penetrant Testing (ASNT 1999).
liquid-to-solid material interface on the part.
Consider a bead of water on a freshly waxed car;
the surface tension at the liquid to gas (air) interface Basic Penetrant Process
is higher than at the liquid to material (waxed paint) The basic penetrant process (Figure 1) requires a
interface causing the water to bead and not spread liquid penetrant, a removing agent, and a developer
over the surface of the car. This is exactly what is material. The penetrant is applied to a clean, dry
not wanted in penetrant testing. In penetrant testing,
the liquid-to-material interface must have a higher
surface tension than the liquid to gas interface such Figure 1. Postemulsifiable (type 1, method D) penetrant process:
that the liquid will spread across the surface, wetting
(a) clean, dry part to which penetrant is applied; (b) after dwell time,
the surface of the material to be tested. When the
liquid reaches a crack on the surface of the material, spray rinse to remove excess penetrant; (c) emulsification bath;
capillary action will draw the liquid into the crack. (d) rinsing after emulsification; (e) developer application in developer
Capillary action occurs when the surface tension in a chamber; (f) interior of booth darkened for inspection under ultraviolet
narrow gap is strong enough to overcome the liquid radiation.
to gas surface tension of the penetrant. The forces (a) (d)
involved in capillary action are given:
2
(1) P 
R

where P is the capillary pressure, R is the radius of

curvature of the liquid surface, and  is the surface
tension of the liquid. For a liquid in a capillary, it is
found that R  r/(cos ) where r is the radius of the
tube and  is the angle of contact between the (b) (e)
liquid and the tube (ASNT 1999, 84). The capillary
pressure becomes:
cos
(2) P  2 
r
showing that as the capillary gets smaller the
pressure for the fluid to enter increases. The pressure
can be great. For example, for a 0.01 mm
(0.0004 in.) radius the pressure can be about
5000 Pa (700 lbf in2) for a fluid with a surface (c) (f)
tension of 0.025 Nm (1.7 lbf in).
The liquid-to-gas and liquid-to-solid tension
factors, coupled with the size and roughness of the
interior walls of the crack, are the driving factors
that control how the penetrant will seep into the
crack. Large gaps of the material decrease capillary
pressures and decrease the ability of the penetrant to
enter the crack. Small or narrow areas of the crack
increase the capillary action, which results in the

LIQUID PENETRANT TESTING 6.3

 surface (aerospace applications typically use to draw penetrant from discontinuities. The
fluorescent penetrant in a dip tank) and seeps, developer also provides a contrasting background to
through capillary action, into any exposed surface view indications. Adequate coverage of all test
discontinuities. The dwell time is critical: for most surfaces is critical. Indications are then evaluated to
aerospace applications, 1200 s (20 min) or longer. (If assess the test object. Fluorescent penetrant requires
the postemulsifiable process is used, the test object is ultraviolet (UV-A) irradiation in a dark room or
dipped in an emulsification bath to make water booth. The inspector must ensure that eyes adapt to
washable the oil based penetrant.) The penetrant is the dark.
then rinsed off with a spray of acceptable
temperature and pressure. To ensure that removal of
penetrant is not excessive, specifications define the
contact time for rinsing and emulsification. After the
surface dries, developer is applied to the test surface

6.4 AEROSPACE NONDESTRUCTIVE TESTING

 Penetrant Techniques PART3

Penetrant Materials postemulsifable, lipophilic method, but there are

notable differences. The hydrophilic method uses
Types a remover material that is made up of
surfactants that when activated by water
There are two basic types of penetrant: type 1, displaces the exposed surface penetrant material
fluorescent penetrant, and type 2, visible penetrant. on the part. The remover material is normally
The fluorescent penetrant materials have mineral and mixed with water in a controlled mixture and
chemical ingredients that provide a greenish yellow either sprayed onto the parts or the parts are
glow when exposed to ultraviolet radiation of 320 to dipped into an emulsifier bath to remove the
400 nm. This band of ultraviolet radiation is penetrant. This allows for a very controlled
sometimes called black light (a misnomer because it method of penetrant removal and is generally
is not black and is not light). The type 2 penetrant the most sensitive method, all other process
contains a red dye, readily visible with normal conditions being equal. This method is required
vision. Because of the sensitivity requirements for for most rotating engine components.
aerospace applications, most penetrant inspection
used by the aerospace industry is type 1.
Sensitivity
Methods There are five levels of sensitivity of fluorescent
Penetrant materials are also divided into four (type 1) penetrant materials: level 0.5, very low
groups named for the removal method of the sensitivity; level 1, low sensitivity; level 2, medium
penetrant material. The four methods are: method A, sensitivity; level 3, high sensitivity; level 4, very
water washable; method B, postemulsifiable high sensitivity.
lipophilic; method C, solvent removable; method D, Visible dye (type 2) penetrant does not have
postemulsifiable hydrophilic. sensitivity levels. The lowest sensitivity level of type
1 penetrant materials is higher than the sensitivity
A. As their name implies, water washable
level of visible penetrants. Type 2, visible dye
penetrants can be removed from a part with
penetrants are not used on aerospace products for
water washing. The penetrant material has
final determination of serviceability. The post
emulsifiers that when exposed to water mix with
emulsifier methods are the most sensitive.
the penetrant material, letting it be removed
The correlation between sensitivity and crack size
with a water spray. This method is not permitted
is very difficult. The length and width of a crack is
for rotating engine components.
not as important as the depth of the crack. A long
B. Postemulsifiable, lipophilic penetrant is not
wide crack with no depth will not be able to store
water washable. These materials must have an
penetrant, thereby resulting in a very weak
emulsifier applied to the surface penetrant to
indication whereas a short, narrow crack with depth
render the penetrant water washable. Adding
would store penetrant and produce a very strong
emulsifier adds another step to the penetrant
indication. The key to choosing the penetrant
process but provides a more sensitive inspection
sensitivity to use for an inspection is to understand
than water washable penetrant materials. This
the type and size of discontinuities that need to be
method is also not permitted for rotating engine
found. Too high a sensitivity penetrant on a
components.
noncritical part can result in overinspection and
C. Solvent removable penetrant is removed with a
possible needless rejection of parts whereas too low
solvent. This removal method is actually named
sensitivity penetrant on a critical part could result in
more for the type of process than the type of
missed discontinuities and potential failure of the
penetrant. Any of the penetrants in the other
part. For most jet engine applications, level 4 (very
methods can be removed with solvent.
sensitive) penetrant is typically used. For other
Generally, this process is for portable
aerospace parts, levels 0.5 to 3 may be used
applications: the materials in this method are (Larson 2002).
normally in aerosol containers. Used correctly, If the inspection of a part requires the detection
this method can be very sensitive in detecting of shallow, tight cracks, or cracks possibly
surface discontinuities. contaminated with foreign material, then the
D. To remove the excess penetrant, penetrant testing should be replaced with eddy
postemulsifiable, hydrophilic penetrant also current testing or other proven method in critical
requires multiple steps that include prerinsing, areas. It is important to remember that penetrant
emulsification, and postrinsing. The method is testing includes a visual testing step: if the inspector
very similar in application to the cannot see the area, it cannot be inspected.
ThisLIQUID PENETRANT TESTING 6.5
 includes cavities, bolt holes, and so on. It is Solvent Removers
important that design engineers be mindful of this
situation when specifying the inspection method for Solvent removers are normally used in method C,
a component. the portable process. The solvents can be obtained in
aerosol cans or bulk materials. Solvent removers are
divided into three classes: class 1, halogenated
Developers (nonflammable); class 2, nonhalogenated
Developers serve two important functions in (flammable); and class 3, special removers, for
penetrant testing. First, they help the penetrant to specific applications.
bleed out of any surface discontinuities that the
penetrant entered during the penetrant dwell time. Summary of Materials
Second, the developer provides a contrasting
background for the penetrant bleedout. The contrast Table 1 summarizes the material configurations in
helps inspection personnel see indications in the penetrant testing. The question is often asked, ―Why
process. all the different methods, classes, sensitivities, and
Developers are divided into five categories: forms?‖ The answer may seem contradictory but
(a) dry powder developer, (b) aqueous soluble they are used to simplify the process. For example, if
developer, (c) aqueous suspendible developer, a procedure requires a type 1, method D penetrant
(d) nonaqueous wet developer, and (e) nonaqueous process using a form a developer, a qualified
type 2 visible dye (solvent based) developer. To these penetrant test technician knows exactly what to use.
five forms a sixth may be added: (f) special This means of identifying the penetrant materials is
developers, for specific applications. much shorter and simpler than stating that the
Developers are composed of small powder technician should use a fluorescent postemulsifiable
particles that perform their function as powder on penetrant material with a hydrophilic base to be
the inspected surface. The penetrant is pulled out of removed with a remover material and to be
the surface breaking discontinuity by capillary developed using a dry developer. Opinions vary
action of the interstices of the particles; the particle among penetrant professionals, but the generally
surface energy is such that the penetrant will wet the accepted designation in the industry is type, method,
particles. The size and concentration of particles and form. The most common penetrant material
affect the performance. Care must be taken to ensure configurations for aerospace applications are as
adequate developer coverage to all inspected follows:
surfaces. If the developer layer is too thin, or does Type 1, method A, sensitivity 2, using a form a
not contain the appropriate constituents to draw the developer — for castings, forgings, welds.
penetrant from the indication, a successful Type 1, method D, sensitivity 3 or 4, using a
inspection is unlikely. If the developer layer is too form a developer — for critical parts or
thick, it can mask smaller indications. hardware.
Type 1, method C, sensitivity 3 or 4, using a
form d developer — for localized spot
inspections.

The exact type, method, and form for a given

Table 1. Penetrant testing material options. application should be included in the procedure
document and should be adhered to.

Ratio Environment
Penetrant Processes
Penetrant visibility 1, fluorescent
(types) 2, visible dye Cleaning
Penetrant removal A, water washable Penetrant testing is designed to detect surface
indications. If the penetrant cannot reach the surface
(methods) B, postemulsifiable, lipophilic
C, solvent removable of the part, the inspection will not be effective.
D, postemulsifiable, hydrophilic Therefore, cleaning and drying of the surface of the
part is critical. All substances that could prevent the
Sensitivity (levels) 0.5, very low penetrant from seeping into a discontinuity must be
1, low removed. However, this cleaning or removal process
2, medium must be controlled and not damage the part.
3, high There are three types of cleaning processes:
4, ultra high solvent, chemical, and mechanical. Of the three,
Developers (forms) a, dry powder developer solvent is the least aggressive to the surface of the
b, aqueous soluble developer part. Solvent cleaning is simply the application of a
c, aqueous suspendible developer solvent, such as alcohol, to a cloth and wiping of the
d, nonaqueous wet developer (NAWD) part surface. The solvent cleaners dissolve the
e, special, for specific applications contaminant on the part surface and allow it to be
wiped clean.
Solvent removers 1, halogenated (nonflammable) Chemical cleaning uses stronger chemicals, such
(classes) 2, nonhalogenated (flammable) as alkaline cleaners, which can be very effective in
3, special removers, for specific applications removing contaminants that cannot be dissolved by

6.6 AEROSPACE NONDESTRUCTIVE TESTING

solvent cleaners. Caution must be used in that some Removing Penetrants
chemical cleaning methods can harm the surface of
parts. For example, some alkaline cleaners can attack Depending on the method used, the penetrant can
the surface of aluminum parts. Parts must always be be washed or wiped off. Extreme care should be used
thoroughly rinsed with water to ensure that all not to overclean the part. A very slight remaining
chemicals are removed from the part after chemical glow from fluorescent penetrant is considered
cleaning methods have been used. Adherence to acceptable after penetrant removal. This very slight
rinse times and temperatures using clean water, remaining glow is preferred to a complete lack of
uncontaminated with carryover from the cleaning any indication that penetrant was on the part, which
process, is critical to a successful inspection. could indicate overwashing, or in the case of
Residual cleaning fluids can affect the fluorescent Methods B and D, overemulsification. For complex
properties of the penetrant. Parts must also be dried parts with internal cavities, care should be taken to
such that cleaning solutions and/or rinse water does ensure that washing and emulsifier fluids have been
not remain in the discontinuities, thereby preventing completely removed.
the entry of penetrant and lowering the inspection
effectiveness. Drying time
Mechanical cleaning is used when solvent or
chemicals will not remove the contaminant. Exact drying time is difficult to determine. There
Mechanical cleaning includes grinding, wire are many factors that control how fast a part will
brushing, mechanical blasting using aluminum oxide, dry. Factors such as humidity, drying temperature,
plastic media, walnut shells, sanding, and other part geometry, and air circulation affect the length
methods that physically ―work‖ the surface of the of time needed to dry a part. Most fluorescent
material. While effective in removing contaminants penetrant testing facilities include an oven station as
from the surface the danger is that the working of the part of the fluorescent penetrant test system.
surface metal may deform the surface and impede Technicians should ensure that factors, such as dry
the ability of the penetrant to seep into surface temperature, are controlled and then monitor the
discontinuities. Mechanical cleaning methods have drying to ensure that parts do not overdry. Extended
been shown to smear metal over cracks. In some exposure to heat or air flow can lead to fading of
applications, a chemical etch is applied to the part fluorescent penetrant indications and thereby affect
surface after mechanical cleaning; however, the detection of critical discontinuities.
chemical etches should not be applied without being
specifically addressed in the penetrant procedure Developer Application
controlling the inspection or without authorized
engineering approval as the etching step also The application of the developer depends on the
changes the surface and, if abused, can lead to type of developer. Dry powder developers are
unacceptable tolerance variations. It is also normally dusted or sprayed onto the part or the parts
important to rinse and dry the parts after mechanical are put into closed chambers and air is applied to the
cleaning and/or etching to ensure that surface developer to form a dust cloud around the part.
contaminants from the cleaning process do not Either way can be effective in applying dry
interfere with the penetrant process. developers. It is important to ensure that the
developer is applied to all surfaces being inspected.
It has been shown that the bottom surfaces of parts
Applying Penetrants can have insufficient quantities of developer, leading
Penetrants can be applied by brushing, dipping, to inadequate detection sensitivity. Consideration
spraying, or pouring on the part. Care should be should also be given to fixtures and rollers that may
used in portable (method C) applications, for what is block the developer from reaching all areas of the
put on must also be removed. inspection surface. Supplemental developer is often
added to the shielded region using a bulb or spray
wand.
Penetrant Dwell Times Wet developers are normally applied by dipping
Selection of penetrant dwell times is affected by the part in the developer solution or pouring or by
several factors. Such factors include the following: spraying the solution over the part. Efforts should be
the size of the discontinuity sought, penetrant type, made to ensure that the solution does not pool in
test object material, test surface condition (rough or any one location. Rotation of parts during the drying
smooth), and the temperature of the test object and process normally prevents pooling. Wet developer
of the environment. solutions should be checked daily for the
Insufficient dwell time will not let enough concentration of particles to ensure quality.
penetrant seep into open discontinuities to produce Nonaqueous developers are applied by spraying the
detectable indications. The procedure for the developer onto the parts. Care should be exercised to
inspection must address the dwell time as determined ensure that only a very slight dusting of the part is
through procedure development. Dwell times mustbe produced from the spraying operation. Too much
controlled and monitored to ensure satisfactory developer will result in masking of small
inspection results. Typical dwell times for aerospace discontinuities.
components range from 20 to 120 min.

LIQUID PENETRANT TESTING 6.7

 Developer Dwell Time Inspection Area and Conditions
Just as penetrant dwell time is critical to the Inspection areas for fluorescent penetrants must
penetrant process, so is the developer dwell time. A be clean. Fluorescent penetrant residual material in
general rule of thumb is that the developer dwell the area tends to contaminate parts and can result in
time should be half the penetrant dwell time. Some unsatisfactory inspections. The inspection area must
standards indicate that the minimum developer dwell have the proper viewing conditions. In the case of
time should not be less than 600 s (10 min). The fluorescent penetrants, there must be ultraviolet
correct answer lies in the best developer dwell time radiation of correct intensity in the area. Ultraviolet
found during the procedure development process. In lamps must be checked daily to ensure that the
some cases a part may need to be visually checked radiation source is providing the specified UV-A
during the entire developer process to ensure that intensity level. The ambient light level should also be
discontinuities are not missed. An example is in the checked daily to ensure that the area is sufficiently
inspection of welds where crater cracks are dark for satisfactory interpretation of indications.
suspected. The bleed out from the center of the crater Refer to Table 2 for specifications regarding light
can at times mask the cracks and lead the inspector levels. It is also important that the inspector allow
to believe that the indication is from a pit. time for dark adaptation of the eyes before
After the developer time has lapsed, indications inspection of the test object.
can be verified by carefully wiping the indication
with a clean lint free cloth or cotton swab dampened
with penetrant remover under ultraviolet radiation.
Typically isopropyl alcohol or acetone is used.
Developer is then reapplied and the dwell time is
restarted. If the indication returns it should be
evaluated using the specified accept/reject criteria.
Most procedures will define the ―rebleed procedure‖
and criteria for acceptance/rejection of indications.
Multiple rebleed steps are typically not allowed as
this has been shown to lead to lost indications.

6.8 AEROSPACE NONDESTRUCTIVE TESTING

 Safety PART4

Penetrant testing should always be accomplished parts that are used in oxygen systems should not be
with safety of the inspector and test object in mind. inspected with regular penetrant materials since the
Individuals should wear rubber gloves to protect materials are normally petroleum based. Special
their hands. Eye protection should ensure that penetrant materials are available for inspecting parts
penetrant materials do not enter the eyes. However, used in oxygen systems. Plastic and composite
it is also important that photosensitive glasses not be materials should be closely checked to ensure that
worn if they impair the desired detection sensitivity. the penetrant materials will not harm the surface of
Caution should be used when operating the part. The use of penetrant materials on damaged
ultraviolet lamps. Never look directly into an composite materials is not recommended since the
ultraviolet lamp. Older style irradiation sources use penetrant can seep into the plies and interfere with
mercury vapor lamps and can become very hot to repair actions. When developer materials are used to
the touch. Care should be exercised to prevent the detect leaks, such as on lower wing panels and rear
hot surfaces from touching the inspector or spars, the developer material should be removed to
remaining near penetrant spray cans. prevent creating a corrosive environment. Developer
Parts should always be checked to ensure that the materials are not corrosive; however, if left on
penetrant materials will not harm the part. Penetrant surfaces the developer residual will attract moisture
materials do not harm most metal parts. However, that can lead to corrosion.

LIQUID PENETRANT TESTING 6.9

PART5
Standards and Specifications

Penetrant inspection methods and procedures are responses. Any difference in appearance such as size,
contained in a number of available specifications. shape, color, or quantity of indications, between the
Table 2 lists a few relevant specifications. Both SAE results of the daily system performance check and
AMS 2647 Revision C and ASTM E 1417 provide the recorded baseline results, requires a system
guidance on known discontinuity standards (KDS) evaluation to determine and resolve the cause.
panels for monitoring the performance of penetrant The rationale, complexity, and analytical methods
inspection systems. The most common KDS panel for a full probability of detection demonstration are
features five starburst indications of increasing size beyond the scope of this publication. The reader is
from 0.38 to 6.35 mm (0.015 to 0.250 in.) with sizes referred to MIL-STD-1823 for requirements and
defined in relevant specifications. The panels should methodologies for a full probability of detection
be processed with unused penetrant to establish demonstration (SAE 1995; USDOD 1999).
baseline responses for comparison to the in-use

Table 2. Some industry standards for fluorescent penetrant testing.

Sstandards Title Purpose

ASTM E 165-02 Standard Practice for Liquid Penetrant Procedures for penetrant examination of materials.
Examination for General Industry
ASTM E 1135 (R03) Standard Test Method for Comparing the Describes techniques for comparing brightness of penetrants used
Brightness of Fluorescent Penetrants in fluorescent penetrant process.
ASTM E 1220 Standard Test Method for Visible Penetrant Procedures for visible penetrant testing with solvent removable
Examination Usingthe Solvent Removable process.(Thisspecification is notused in aerospace industry.)
Process
ASTM E 1417 Standard Practice for Liquid Penetrant Establishes minimum requirements for conducting liquid
Testing penetrant examination of nonporous metal and nonmetal
components.
ASTM E 1418 Standard Test Method for Visible Penetrant Describes procedures for visible liquid penetrant examination with
Examination Usingthe Water-Washable water washable process. (This specification is not used in
Process aerospace industry.)
ASTM E 2297 Standard Guide for Use of UV-A and Visible Used for illumination of penetrant and magnetic particle
Light Sources inspections.
SAE AIR 4964 Exposure Levels of UV Radiation in Long wave ultraviolet (UV-A) radiation (320 to 400 nm) for
Nondestructive Inspection Processes fluorescent inspections in magnetic particle and liquidpenetrant
examinations is generated from either fluorescent or high
intensity discharge lamps, stationary or portable. Commercially
available UV-A lamps vary greatly in intensity output, a legitimate
concern for health hazard. This document reviews nature of UV-A
radiation emitted by ultraviolet sources, acceptable dosagelimits,
and safe practices when working with UV-A radiation.
SAE AMS 2647 Rev. C Fluorescent Penetrant Inspection Aircraft Requirements and procedures for detection of discontinuities in
and Engine Component Maintenance aircraft and enginecomponentsduringmaintenance and
overhaul.
SAE AMS 2644 Rev. D Penetrant Inspection Materials Qualified products list for penetrants, developers, and emulsifiers.

6.10 AEROSPACE NONDESTRUCTIVE TESTING

 Applications PART6

Advantages and Limitations

Table 3. Advantages and limitations of penetrant testing.
of Penetrant Testing
Table 3 lists the advantages and limitations of
penetrant testing. The most important advantage is Advantages Limitations
that it can detect very small surface breaking fatigue
or machining cracks not found by other methods Inexpensive. Discontinuities must be open to the
and therefore is a critical method to prevent surface.
inservice failure of components. Covers large parts rapidly.
Magnifies the effects of tiny Cleaning is necessary both before and after
surface discontinuities for inspection.
What Can Go Wrong detection; detects cracks Dirt or contamination may mask
Penetrant testing requires careful adherence to that cannot be resolved with discontinuity.
process parameters through steps for proper the naked eye. Penetrant rinsing techniques can overwash
sensitivity to be obtained. Table 4 lists what can go Indications are directly on and remove penetrant before bleed out,
wrong in penetrant testing and corrective actions.
the surface. missing discontinuity.
Easy to apply. Environment and safety issues.
Examples No power is required for Temperature limitations.
portable methods.
Automated Liquid Penetrant Spray Operator dependent.
Applicable to many materials Properlightingandintensity are necessary.
Processing System in Aircraft and complex sizes and
Manufacturing shapes. Thin coatings adversely affect sensitivity.
The automated aerospace liquid penetrant spray Sensitivity depends on process.
Portable.
processing and testing system (Figures 2 to 7)
developed for a major aircraft company is an Process materials can degrade or become
installation 60 m (200 ft) long, 6 m (20 ft) high, and contaminated.
4.5 m (15 ft) wide. It uses the electrostatic liquid

Table 4. What can go wrong with penetrant testing.

Problem Consequence Corrective Action

Part surface is not clean enough. Penetrant cannot entersurfacediscontinuities, Clean and dry surface with acceptable
Penetrant that beads or separates resulting in loss of sensitivity and in missed technique to remove all containments and
on surface indicates inadequate discontinuities. rinsing fluids before penetrant application.
cleanliness.
Part surface is wet from cleaning Rinse water or cleaning fluids thatremain in Use approved dryingtechniquessuchasoven
operation, particularly in crevices cracks prevent penetrant from entering, leading dry or flash dry.
and corners. to loss of sensitivity and missed discontinuities.
Part is too hot when penetrant Fluorescent penetrant materials will flash, Ensure part and penetrant are within
material is applied. degrading fluorescent material’s ability to acceptable temperatures, not to exceed 49 °C
fluoresce under ultraviolet irradiation. Result is (120 °F). In general, if the part cannot be
missed discontinuities. handled with bare hands, it is too hot.
Part or penetrant is too cold. Penetrant materials lose effectiveness to seep Warm penetrant and/or part to acceptable
into open surface discontinuities during normal temperatures. Generally acceptable lower
dwell time. Result is missed discontinuities. temperature is 4 °C (40 °F).
Penetrant will not seep into small open Dwell time indicated in procedure must be met
Penetrant dwell time is discontinuities, resultingin missed for successful inspection. When dwell time is
insufficient. discontinuities. notmet, part mustbecleaned and reprocessed.

continued on next pagLIQUID PENETRANT TESTING 6.11

 Table 4. What can go wrong with penetrant testing (cont.).

Problem Consequence Corrective Action

Type 2 dye penetrant leads to Result will be failed inspections and missed Ensure dye penetrant is not used on test objects
ineffective subsequent inspections discontinuities. that will be inspected later using fluorescent
by Type 1 penetrants penetrants.

Emulsifier time exceeded. Overremoval of penetrant results in missed Clean and reprocess test object.
discontinuities.

Overremoval of penetrant results in missed Correct water temperature or pressure and

Wash temperature or pressure is discontinuities. completely reprocess test object.
incorrect.
Overremoval of penetrant results in missed Do not apply remover directly on test object. Spray
Overapplication of solvent discontinuities. or dampen clean lint free cloth and wipe penetrant
remover in method C. off test object. Clean and reprocess test object.
Penetrant will undergo fluorescent Clean and reprocesstest object. Inspecttest object
degradation (heat fade), resulting in missed for damage.
Drying oven temperature too discontinuities. Possible damage to test object.
high, or test object is left in oven
too long. If concentration was too high, result will be Bringdeveloper to correctstrength. Clean and
too much developer on test surface, resulting reprocess test object.
Developer at incorrect strength in possible masking of discontinuities. If
(concentration of developer concentration was too low, developer coverage
particles in solution is too high or may be inadequate to produce indications.
too low).
Less bright indications may result in missing Apply supplemental developer in affected areas.
discontinuities.
Inadequate developeronbottom
orsidesoftestobject. Inadequate
developer because of fixtures or
rollers blocking application. If concentration was too high, result will be Completely reprocess test object and apply only
too much developer on test surface, resulting enough nonaqueous developer to coat test object
in possible masking of discontinuities. If lightly.
Too much or too little concentration was too low, developer coverage
nonaqueous developer applied to may be inadequate to produce indications.
test object.
Fluorescentpenetrant indications will be too Check ultraviolet lamp with calibrated meter.
dim, resulting in missed Change bulb or filter as needed to bring radiation
indications/discontinuities. to proper intensity.
Ultraviolet radiation below Excessive light in inspection area decreases Check brightness of white light in inspection area
allowable minimum intensity. visibility of fluorescent penetrant indications. with calibrated meter. White light level should not
Result will be missed indications and exceed specification. Darken as needed.
discontinuities.
Light contamination in inspection
area. Missed indications and discontinuities. Allow 60 to 300 s (1 to 5 min) for technician’s eyes
to adapt to darkness.
Missed indications and discontinuities. As system performance check, verify under
ultraviolet radiation that discontinuity standards
Inspection before technician’s are clean and do not fluoresce beforeprocessing.
eyes have adapted to darkness.

With extended use,

contamination can occur of
known discontinuities such as low Result will be failed inspections and missed Was process correctly followed? Was standard
cycle fatigue cracks and artificial discontinuities. acceptable? Was penetrant and/or developer dwell
discontinuities on standard time correct? Was penetrant properly removed from
reference panels. standard? Was ultraviolet lamp checked for
radiation intensity? Correct problem and reprocess
standard.
Penetrant process fails system
performance test.

6.12 AEROSPACE NONDESTRUCTIVE TESTING

penetrant spray technique and consists of spray, conductor lugs to provide the necessary ground
wash, drying, and near ultraviolet radiation viewing potential for the electrostatic liquid penetrant spray.
modules. Cracked test specimens are loaded on the first load
A semidarkened observation area with near bar through the liquid penetrant system after every
ultraviolet radiation is located on the outlet end of shutdown, no matter how brief. Testing of these
the liquid penetrant spray module to view parts for known anomaly specimens confirms proper system
full liquid penetrant coverage. When developer is operation. The conveyor speed through the liquid
required, the nonaqueous wet aerosol developer is
applied in the ultraviolet radiation viewing booth.
The viewing booth is equipped with an exhaust Figure 2. Automated aerospace liquid penetrant
system to protect personnel during the application of spray processing and inspection system.
developer.
Automation of parts handling during liquid
penetrant testing produces uniformity of processing.
All systems are adjusted to uniform and consistent
part travel speed, which provides uniformity in
liquid penetrant coverage, liquid penetrant dwell
time, wash time, and drying time. This consistency
prevents inadequate dwell periods, overwashing, and
overdrying — common discrepancies of manually
operated systems.
The automated chemical spray system can handle
large parts in less space than open tank systems
require. Parts of unrestricted length can be processed.
Long spars of 24 m (80 ft) length are readily
processed through the system of Figure 5.
This system can handle part widths up to 3.5 m
(12 ft) and part thicknesses up to 0.9 m (3 ft). Even
though liquid penetrant is not reclaimed, the spray Figure 3. View of load bars and hangers at part
system uses about one eighth of the volume of liquid loading area.
penetrant used in open tank systems for an equal
number of test parts and minimizes potential
contaminations. The system can handle tremendous
numbers of test parts. Two inspectors can operate the
system and perform many times the volume of
viewing that can be handled through open tank
systems. The system contains its own motorized
conveyor system, thus eliminating overhead crane
systems and reducing ceiling height and loading area
requirements.
Parts are moved through the system on load bars
hung from a motorized overhead monorail conveyor.
Parts are mounted on specially designed hangers that
are hung, as applicable, on the overhead load bars
(Figure 3). Parts hangers are designed for a minimum
of liquid penetrant entrapment areas to preclude
bleedout of liquid penetrant on parts after washing. Figure 4. Overall view of chemical cleaning system. DC-10® vertical
The conveyor is equipped with a variable speed
stabilizer spars measuring 5.4 m (18 ft) long and 2.7 m (9 ft) wide are
control for speeds from 5 to 43 mm/s (1 to
8.5 ft/min). Normal conveyor speed is 1.5 m/min ready to proceed through system.
(5 ft/min).
Adjacent and parallel to the liquid penetrant
system is an automated spray chemical cleaning
system (Figure 4) that operates as an integral part of
the liquid penetrant system by precleaning and
postcleaning parts. It consists of two tandem alkaline
cleaning modules, followed by rinse, deoxidizer,
rinse, protective coating, hot water rinse, deionized
rinse, and hot air knife modules. The conversion
coating application module is only operated after
post cleaning, as applicable. Turntable conveyor
systems with pneumatic lowering platforms
(lowerators) are located on both ends of the system
to permit the feeding of parts from cleaning to liquid
penetrant or reverse and the loading, unloading, and
temporary storing of parts, whichever is needed.
During loading of parts, electrical jumpers are
connected from load bar suspension straps to

LIQUID PENETRANT TESTING 6.13

 penetrant system of Figures 5 and 6 is set at
Figure 5. Overall view of liquid penetrant spray 25 mm/s (1 in./s) or 1.5 m/min (5 ft/min). This
system viewed from liquid penetrant spray speed permits the liquid penetrant dwell time on
application module end. the material to be within specification.
The liquid penetrant reciprocator air spray
pressure gage is set at 125 to 140 kPa (18 to
20 lbf/in.2). This provides a light uniform coating of
liquid penetrant on the parts and covers cavity areas
adequately. Two liquid penetrant reciprocators
operate up and down vertically for a span of 3.6 m
(12 ft) inside the spray module (Figure 6). They can
each be at full vertical span or half span (top half)
depending on the volume of parts on the load bar.
Photoelectric load sensors activate the liquid
penetrant spray nozzles as the end of the load bar
enters the spray module. A curtained observation
area with near ultraviolet radiation is located at the
exit end of the liquid penetrant application module
for the examination of parts for adequate liquid
penetrant coverage. Observation platforms and six
banks of tubular ultraviolet lamps are provided for
this examination.
The water wash module is equipped with several
Figure 6. Liquid penetrant spray applicators banks of water nozzles that provide a mild water
mounted in vertical conveyance system inside spray. Nozzles are adjustable from the outside for
liquid penetrant spray module. different types of parts. The wash water pressure is
set at a gentle 140 kPa (20 lbf/in.2) to prevent
overwashing of parts. Parts exit the water wash
module through forceful air knives that remove
much of the surface wetness. The drier consists of
two hot air knives mounted vertically inside the
module for a 3.5 m (12 ft) span. The heaters are set
to produce a maximum 66 °C (150 °F) temperature
inside the module. Forceful blowers enhance
adequate drying of parts. Between the drier and the
inspection module is a storage station where several
load bars may be stored while awaiting entrance into
the viewing booth. Audible and visible signals are
emitted inside the viewing booth when load bars are
exiting the drier to alert inspectors that load bars
may be moved transversely (on overhead tracks),
clearing the way for additional load bars.
Parts on the conveyor are moved with long
hooked poles into the viewing booth (Figure 7). The
Figure 7. Outside view of ultraviolet radiation inspection module from load bar is moved into place on one of the two
part exit end. pneumatic lowering platforms directly above two
parallel pits. Parts are lowered into the pits gradually
as viewing with near ultraviolet radiation is
performed longitudinally along the load bar. Cracked
test coupons are examined for system performance
before viewing the parts. Following viewing, load
bars carrying acceptable parts are routed back
through the cleaning process to remove any liquid
penetrant or developer residue. If required, a
protective coating is applied during this cycle.

Second Automated Liquid Penetrant

Testing System for Aircraft
Manufacturing
Figures 8 and 9 show another example of an
automated liquid penetrant testing system installed
in an aircraft manufacturing company plant.
Aluminum extrusions and sheets up to 2.4 m (8 ft)
wide and 8.5 m (28 ft) long can be handled by this
fully automated fluorescent penetrant testing system.
A computer controls the speed of the conveyor,
automatically turns each station on and off and, if a

6.14 AEROSPACE NONDESTRUCTIVE TESTING

 Figure 8. Schematic diagram of automated conveyorized liquid penetrant system layout. Cleaned parts are located on conveyor;
loaded parts travel through liquid penetrant spray booth; drain dwell time is consumed in travel to rinse station. Following rinse,
parts are dried in drier and nonaqueous developer is sprayed on parts in developer spray booth. Developing time is consumed in
travel to ultraviolet radiation inspection booth; white light inspection follows before parts unloading.

Ultraviolet radiation inspection

White light inspection

Unload area Load area
Developer
Dryer

Rinse

Drain dwell

Liquid penetrant

malfunction occurs, shuts down the whole system.

The seven processing steps for this system are the Figure 9. Area for liquid penetrant testing
following: (1) preclean and dry test parts, (2) apply aviation components includes dip tank operation
high sensitivity water washable fluorescent liquid (left) and automated spray system (right).
penetrant, (3) allow dwell time, (4) rinse with water,
(5) dry, (6) apply nonaqueous wet developer, (7) view
under ultraviolet radiation, and (8) clean.
This automated liquid penetrant testing system
has helped the cost, schedule, and quality assurance
of aerospace parts.

Automation Example Summary

Automation of the liquid penetrant process is
often a very effective way to establish excellent
process control, minimize wasted penetrant
materials, and improve inspection times and results.
The key to any automated system is the
establishment of a carefully timed process that
addresses all the factors of a manual system but
eliminates or minimizes human error. Automated
processes can be used for large and small parts. The
only limitation for large parts is the size of the means to amplify, record, and interpret the signal
facility and equipment needed. For small parts, a produced during the scan.
limitation is the ability to properly secure or rack the
parts during handling in an automated process.
Efforts to automate penetrant processes should
Automation of Liquid Penetrant
consider the timing of the different steps in Interpretation
performing a manual process. Each step must have Automated interpretation of liquid penetrant
adequate time to allow the step to be completed testing indications is a useful and logical extension
within the guidelines of the governing technical of pattern recognition technology. Attempts have
data. been made to develop systems based on camera
Penetrant and developer applications in an images and laser scanning (ASNT 1999, 249-243).
automated process can be enhanced by electrostatic Several systems have been implemented but have
or atomized spraying of test surfaces. Thisminimizes not been economically beneficial. Although
wasted product and usually maintains a cleaner automated readout has been demonstrated to
process environment. approach human capabilities in specific applications,
In addition to automating the application and the pattern recognition capability of the human
removal process for penetrant inspections, the visual operator exceeds that of automated readout systems.
scanning of the completed process is also possible. In addition, the false call rate of an automated
Although human inspectors are usually the norm for system may exceed that of a human operator.
penetrant inspection, scanning systems can be Therefore human intervention is necessary for
validated to improve process control and inspection critical applications.
results. Scanning systems must consist of three basic
functional aspects: a source of ultraviolet radiation,
a photometer to detect fluorescent indications, and a

LIQUID PENETRANT TESTING 6.15

References
ASNT. 1999. Nondestructive Testing Handbook 2: Liquid Penetrant
Testing, 3rd edition. Columbus, OH: American Society for
Bibliography
Nondestructive Testing. Cartz, L. 1995. Nondestructive Testing, 127-140. Materials Park, OH:
Larson, B. 2002. DOT/FAA/AR-01/95. Study of the Factors Affecting the ASM International.
Sensitivity of Liquid Penetrant Inspections: Review of Literature Iddings, F.A., and P.J. Shull. 2002. ―Liquid Penetrant.‖ Nondestructive
Published from 1970 to 1998. Washington, DC: U.S. Department of Evaluation Theory, Techniques and Applications. New York, NY:
Transportation, Federal Aviation Administration. Marcel Dekker, 17-60.
SAE. 1995. SAE AMS 5608D, Cobalt Alloy, Corrosion and Heat
Resistant, Sheet Strip and Plate 40Co-22Cr-22Ni-14.5W-0.07La
Solution Heat Treated. Warrendale, PA: Society of Automotive
Engineers.
USDOD. 1999. MIL-STD-1823, Non-Destructive Evaluation System
Reliability Assessment. Washington, DC: United States Department
of Defense.

6.16 AEROSPACE NONDESTRUCTIVE TESTING

 X
7
CHAPTER
CHAPTER

Magneti
c
Particle Contents

Testing Part 1. Magnetic Particle

Testing Principles, 7.2
Part 2. Magnetic Particle
Testing Standards and
Specifications, 7.10
Part 3. Magnetic Particle
Testing Applications, 7.13
References, 7.17

Contributors
Charles W. Eick
Lisa Brasche
Thomas C. Dreher
Steven M. Younker
7.1
 Magnetic Particle Testing Principles
PART1

In magnetic particle testing, the magnetic Diamagnetic materials have a negative susceptibility
properties of ferromagnetic materials may be used to under a magnetic field and thus weaken the
detect discontinuities in ferrous aerospace magnetic field. In paramagnetic materials, there are
components. When a magnetic field is induced into a some unpaired electrons. Paramagnetic materials
component, discontinuities or geometry changes in have positive susceptibility and can be slightly
the object will cause interruptions in the continuity attracted in magnetic fields but do not retain
of the field. Finely divided magnetic particles applied magnetism. Ferromagnetic materials have a greater
to the test object become trapped and accumulate at number of unpaired electrons and thus a net
interruptions in the magnetic field on or near the magnetic moment. In ferromagnetic materials, the
surface. These particle accumulations, also called electron moments can be aligned in domains. When
indications, can then be seen. Depending on the these domains align with a magnetic field, strong
desired sensitivity, magnetic particles may be either magnetic properties result. In the unmagnetized
colored for detection in light or coated with a state, the domain orientations are randomized, but in
fluorescent dye for detection under ultraviolet the magnetized state they are aligned and can be
radiation. Except for applications where surface retained. The relationship between magnetization
conditions interfere with particle migration across (magnetic dipole moment per unit volume) and
the surface of a test object, fluorescent magnetic magnetic field intensity is:
particle testing is generally the method of choice for
testing of machine finished aerospace components.
Figure 1 illustrates a magnetic leakage field (1) M   mH
produced at a discontinuity. As the magnetic field
passes through the test object, the disruption caused where M is the magnetization in amperes per meter
by the crack forms opposing magnetic poles and (A/m), m is the magnetic susceptibility, and H is the
causes part of the field to leave the test object, magnetic field intensity in amperes per meter. The
forming a magnetic leakage field between the two bold lettering indicates that these are vector
sides. When magnetic particles are applied to the quantities, meaning that these values have a
area, the magnetic leakage traps the particles and directional component.
forms a bridge across the crack. Magnetic induction is related by:

(2) B  0 H  M 
Magnetic Materials 
Magnetic fields arise from the motion of where B is magnetic flux density in units of tesla (T)
electrons. Magnetic materials also contain electrons and 0 is the permeability of free space (0 = 4 
and are affected by magnetic fields. 10-7 N/A2). These equations are often recombined
The magnetic properties of a material is into the forms:
determined by how the field interacts with the
material. Materials can be classified as diamagnetic,
paramagnetic, or ferromagnetic. In diamagnetic (3) B  0 1   m H  0r H
materials, the electrons are paired and have no net
magnetic moment: they exert no magnetic force. where r is the relative permeability. Permeability is
the ability of a material to have a magnetic field
created within it. High relative permeability and
Figure 1. Magnetic field lines in test object containing crack. high magnetic susceptibility are related (r = 1+ m)
and allow materials to be inspected with magnetic
Magnetic particles particle testing. Table 1 lists some magnetic
Magnetic field lines
characteristics of common materials.
When placed in a magnetic field of increasing
S N intensity, the domains in a ferromagnetic material
will become more and more aligned until saturation.
As the field is reduced, magnetization will reduce but
does not return to zero when the field is at zero. A
hysteresis loop (Figure 2) shows this effect. For
material not previously magnetized, the curve starts
Crack at 0,0 on the B and H axes. Increasing H (magnetic
field intensity) has a positive effect on B (magnetic
flux density) until the saturation point p1 is reached

7.2 AEROSPACE NONDESTRUCTIVE TESTING

 Table 1. Material magnetic properties.

Material Magnetic Property Relative Permeability

Diamagnetic gold 0.999964

silver 0.99998
copper 0.999991
lead 0.999983

vacuum 1.0
Paramagnetic air 1.00000036

aluminum 1.000021
Ferromagnetic steel (0.9 , carbon) 100.0
cobalt (99 , annealed) 250.0
nickel (99 , annealed) 600.0
iron (98.5 , cold rolled) 2000.0
Iron (99.8 , annealed) 6000.0

Figure 2. Hysteresis loop showing effect of magnetic field intensity H on magnetic flux density B: (a) virgin curve of hysteresis
loop; (b) hysteresis loop showing residual magnetism; (c) hysteresis loop showing coercive force; (d) hysteresis loop showing
reverse saturation point; (e) hysteresis loop showing reverse residual magnetism; (f) complete hysteresis loop.
(a) (d)
B B
P1 P1

Zero flux density and zero Br

magnetic strength
–Hc H
H
O Reverse O
magnetization
saturation point

P2
(b) (e) B
B
P1 P1
Residual Br Br
magnetism
–Hc H
H O
O
Reverse
magnetization Residual magnetism
point –Br

P2

(c) B (f) B
P1 P1
Br Br

–Hc H –Hc H
O O Hc
Coercive Reverse
force residual point
–B r

Legend
B = magnetic flux density P2
H = magnetic field intensity
O = origin beforemagnetization
P = saturation point

MAGNETIC PARTICLE TESTING 7.3

 (Figure 2a). When H is returned to zero, residual direction whereas longitudinal fields are used to
magnetism will exist as shown in Figure 2b. The detect circumferential discontinuities. Note that the
point at which the field intensity is zero but strongest leakage field is produced when the
magnetism remains is the retentivity point with discontinuity causing the leakage is perpendicular to
residual magnetism Br. As the magnetic field is the field direction. If the discontinuity is at an
reversed and increased in intensity (in the negative orientation other than perpendicular to the field, the
direction), the magnetism will go to zero at the point indication will be weaker.
of coercivity as shown in Figure 2c. Increasing the Circular fields can be created by passing an
negative field intensity will result in saturation of electrical current directly through a test object. This
the object in the opposite direction (polarity) at point is typically accomplished by clamping the test object
P2 (Figure 2d). Figures 2e and 2f show the reverse between contact heads on a wet, horizontal ―mag‖
residual magnetism and full hysteresis loop. machine, but contact can also be made through
Understanding the hysteresis loop for a material clamps or prods. It should be noted that passing
can be helpful for many reasons. One of these is that current directly through the test object comes with
it is generally necessary to demagnetize an object the risk of arc burning the test piece. This risk is
following completion of a magnetic particle significantly increased with prods. For this reason,
examination. Demagnetization can be done by prods are rarely approved for machine finished
reversing the magnetic field to the coercivity point. aerospace components.
Although not commonly done, it is also possible to Circular fields are often induced in hollow or ring
demagnetize a material by heating it above its curie shaped test objects or test objects having through
temperature. The curie temperature is the point holes by placing a conductor through the opening
above which a material can no longer retain its and passing a current through this central conductor.
magnetic property and becomes paramagnetic with Conductors can be constructed from solid bars (often
no alignment of magnetic domains. Demagnetization copper or aluminum) so that they can be clamped
is important because residual magnetism can between the contact heads on a wet horizontal
negatively affect subsequent manufacturing magnetic unit; wires or cables may also be used. The
operations such as chip forming and electron beam diameter of these conductors must be large enough
welding. Residual magnetism can also result in to carry current high enough for magnetic particle
excessive wear in the service of components such as testing. The magnetic field created around this
bearings because residual magnetism can attract conductor is shared with the test piece. Although
metallic shavings. often referred to as a central conductor, the term
central is somewhat misleading: these conductors
can be positioned so that they are centered or
Magnetization Fields concentric with the test piece but can also be offset
In magnetic particle testing, there are two or eccentric. Magnetizing currents must be adjusted
primary directions of the magnetic fields: circular depending on whether conductors are concentric or
and longitudinal. Numerous methods can be used to offset. A conductor might be offset because its
generate each of these fields in the component being diameter is too wide to fit in the test piece. Most
examined. Circular fields permit the detection of specifications require multiple magnetizations when
discontinuities oriented in the axial (longitudinal) a conductor is offset. One big advantage of central
conduction is that this noncontacting technique
dramatically reduces the risk of arc burning. It is
Figure 3. Circular field induction magnetic particle testing with generally considered good practice to insulate these
magnetic particle bench unit: (a) current through single test object; conductors to further reduce the risk of arc burning
and to prevent metal-to-metal contact. Figure 3
(b) current through multiple test objects.
illustrates the creation of circular magnetic fields in
(a) test objects.
In some cases, conductors can be used to create
circular fields in test objects having L or U shaped
sections (such as flap tracks), but caution should be
exercised in these instances because distortion of the
field can occur: magnetizing currents may need to
be increased because of the lack of a closed loop for
the field to follow. To prevent arc burning, some
maintenance facilities have placed test objects
Electric Magnetic field
immediately next to a conductor to induce circular
current
fields in solid test objects that would typically be
magnetized with direct current. Although this
(b) Cracks proximity can result in some portion of the test piece
having a field with a circular or circumferential
component, this technique does not produce a truly
circular or encircling field. Because only a sector of
the larger field encircling the conductor is shared,
areas on the test object will not be properly
inspected. This practice is generally not
Magnetic field recommended unless approved by the cognizant
Magnetizing current engineering organization.

7.4 AEROSPACE NONDESTRUCTIVE TESTING

 Longitudinal fields can be induced by encircling the toroidal field, current flow is produced by
the test object with a coil and applying a current to positioning the component in a rapidly collapsing
the coil. The current flowing in the coil produces a or varying magnetic field. This method is generally
donut shaped magnetic field around the coil. When a limited to ring shaped test objects such as bearing
ferromagnetic test object is placed into this field, the races, gears, and test objects having
field seeks the path of least resistance and length-to-diameter (L/D) ratios less than three. A
preferentially flows though the test object, producing toroidal field is typically produced by positioning a
a longitudinal (axially oriented) field. highly permeable, laminated core along the central
The diameter of the coil in relationship to the axis of a coil and then placing the ring shaped test
diameter of the test object is referred to as the fill object so that the laminated core extends axially
factor. This fill factor can have a significant effect on through its bore. The core must be magnetically
the intensity of the magnetic field produced at any isolated from the test object and electrically isolated
given amperage. As a result, when changes are made from the magnetic particle test equipment. As the
to the coil diameter, techniques should be reverified. magnetic field rapidly expands or contracts, a
Figure 4 illustrates a longitudinal field being induced current is induced in the ring shaped test object and
in a test object. creates a secondary circular magnetic field
Coils on wet horizontal magnetic particle testing (Figure 5).
units come in a variety of sizes. Coils can also be A variation of the toroidal method that has been
formed from lengths of insulated cable. This is often successfully applied for detection of circumferential
referred to as the cable wrap method. Some
manufacturers also make portable alternating current
coils for field applications.
Figure 4. Longitudinal field induction during magnetic particle testing.
Unlike circular fields that are generally contained
within the test piece, longitudinal fields leave and Current
enter the test object, forming two distinct poles at
these locations. These poles provide a means for
measuring the residual magnetic field in a test
object. For this reason, longitudinal magnetization is
typically performed as the final step before
demagnetization. One disadvantage associated with Bath
the formation of these poles is that the locations
where the field enters and exits the test object Magnetic
cannot be effectively examined. field Transverse cracks
Many variations of circular and longitudinal
fields occur with wet horizontal magnetic particle
testing equipment. A variation of a circular field,
often referred to as a toroidal field, can be
produced by inducing a current in the test object. In

Figure 5. Toroidal magnetization of test piece.

Primary magnetic field

Toroidal magnetic field

Magnetizing coil Induced current

Discontinuities in part

Primary current

Laminated iron core

MAGNETIC PARTICLE TESTING 7.5

 discontinuities on disk or ring shaped test objects
Figure 6. Flat coil magnetization of test object: (a) axial view; relies on the use of a flat coil placed next to the test
(b) cross section. object. This results in the formation of a radial field
in the test object and is similar to magnetic
induction achieved by placing a conductor adjacent
(a) to a test object. This field is perhaps more akin to a
longitudinal field in that poles are produced where
the field enters and exits the outside diameter and
inside diameter of the test object. As with
longitudinal fields, the surfaces where the field
enters and exits the test object are generally not
inspectable. Figure 6 illustrates flat coil
magnetization.

Field Intensity
Establishing a technique that produces the proper
magnetic field intensity throughout all areas of
interest in the test object undergoing testing is one
of the most challenging and important aspects of a
reliable magnetic particle test. The field intensity
must be strong enough to generate leakage fields
(b) needed to form indications at discontinuities of
Magnetic field interest but not so strong so as to attract excess
particles that can mask small indications. These
fields are generally established experimentally, by
using test objects with either known or artificial
discontinuities such as quantitative quality
indicators or by using a flux density measuring
device such as a hall effect tesla meter, which can
measure the magnetic field in air at the surface. A
tangential field intensity of 3 to 6 mT (30 to 60 G) is
generally recognized as a useful range for magnetic

Figure 7. Quantitative quality indicator response at varying current: (a) 200 ampere turns; (b) 1900 ampere turns;
(c) 1900 ampere turns residual; (d) 3750 ampere turns; (e) 5500 ampere turns.
(a) (b) (c)

(d) (e)

7.6 AEROSPACE NONDESTRUCTIVE TESTING

particle testing. However, indications can be be found in ASTM E 1444 (ASTM 2005). Although it
produced on quantitative quality indicators at values is generally agreed that these formulas typically
below 1.5 mT (15 G). Figure 7 illustrates a series of result in overmagnetization of the test object
longitudinal magnetizing shots of a quantitative (depending on the permeability of the material being
quality indicator shim affixed to the surface of a examined), the formulas can still serve for
high strength steel bar. Calculations indicated that establishing a starting current value or as a cross
5625 ampere turns are needed for the test. Figure 8 check during technique development. Any of these
illustrates the indications produced by a tools in isolation can lead to improper magnetization
lab-generated fatigue crack in the same bar. Figure 9 if used incorrectly. Quantitative quality indicators
illustrates details of the same crack. must be in intimate contact with the surface of the
Before the introduction of quantitative quality test object to perform properly. Tesla meters are
indicators and hall effect tesla meters, formulas were directionally sensitive and particularly difficult to
often used to calculate current values necessary to use for measuring longitudinal fields on complex
achieve adequate field intensity. These formulas can geometries.

Figure 8. Response from 0.9 mm (0.036 in.) crack at varying current: (a) 200 ampere turns; (b) 1900 ampere turns;
(c) 1900 ampere turns residual; (d) 3750 ampere turns; (e) 5500 ampere turns.

(a) (b) (c)

(d) (e)

Figure 9. Lab generated fatigue crack: (a) fluorescent particle inspection response; (b) magnetic particle response at
3750 ampere turns; (c) fractured surface.

(a) (b) (c)

Length = 0.914 mm
(0.0360 in.)

Length = 0.919 mm Length = 1.016 mm

(0.0362 in.) (0.0400 in.) Length = 0.333 mm (0.0131 in.)

MAGNETIC PARTICLE TESTING 7.7

 Figure 10. Current waveform: (a) at 500 A single-phase direct current,
Current Waveform
illustrating varying output over time; (b) expanded view of Another important consideration in establishing a
nonsinusoidal waveform. proper magnetic particle testing technique is the type
of current used to generate the magnetic field.
(a) Magnetizing currents can either be alternating
current or direct current. Direct current has the
advantage of being able to detect subsurface
100.0 mV per division

discontinuities whereas alternating current has the

advantage of producing a more uniform field over
test objects having complex geometries, because of a
phenomenon known as skin effect. Another
advantage of alternating current is increased particle
mobility from constant cycling of the waveform
from positive to negative. Direct current can be
generated from a battery or capacitor, but those
60.00 ms per division sources are seldom used in aerospace applications.
Most direct current for magnetic particle testing is
(b) produced by rectifying an alternating current. There
are three principal variations of this rectified
alternating current: half-wave direct current,
100.0 mV per division

full-wave direct current, and three-phase full-wave

direct current. Half-wave direct current is sometimes
referred to as pulsed direct current and combines the
benefits of increased particle mobility provided by
alternating current with the subsurface capability of
direct current. Three-phase full-wave direct current
is generally considered to provide the best
subsurface detection capability, but this point has
10.00 ms per division
been debated.
In addition to the basic waveform, another factor
often overlooked but given more attention in the
twenty-first century is the specific shape of the
Figure 11. Current waveform: (a) at 1 kA alternating current showing waveform. This includes both the shape of the pulse
120 ms rise time; (b) expanded view of nonsinusoidal waveform. (that is, the amplitude of each of the individual
(a) cycles as well as the rate at which the current rises
and falls) and the shape of each individual cycle
(sinusoidal or sawtooth). Both of these factors can
200.0 mV per division

influence the relationship between the indicated

current value and the intensity of the magnetizing
field. The factors are dictated by how the current is
controlled by the various equipment manufacturers.
Personnel developing and performing testing must
be conscious of this because applying the same
current values on two different pieces of equipment
may produce different results: techniques should be
60.00 ms per division verified on each piece of equipment used. Figures 10
to 12 illustrate the variability of waveforms
(b) produced by typical wet horizontal mag machines.
200.0 mV per division

Magnetic Particles
Magnetic particles are very small ferrous particles
with high permeability and low retentivity. They can
be either visible particles or particles with a
fluorescent dye coating. The visible particles come in
several different colors to help contrast on the test
surface to make discontinuities more visible. To
detect indications with fluorescent particles, the test
10.00 ms per division object is viewed in a dark area under ultraviolet
radiation, sometimes called black light. For
applications requiring the detection of very small
indications, fluorescent particles are used because a

7.8 AEROSPACE NONDESTRUCTIVE TESTING

bright fluorescent indication on a dark background
is easily seen. Although visible particles have the Figure 12. Current waveform: (a) at 5 kA alternating current;
advantage of not requiring special lighting (b) expanded view showing nearly sinusoidal behavior near maximum
conditions, they are rarely used in aerospace output.
applications because of their lesser sensitivity. In
general, dry particles are used for inspection of tools, (a)
of material handling equipment, and of low stress
welds where larger discontinuity sizes are allowed.

100.0 mV per division

Magnetic particles can also be used in wet
applications, where the particles are suspended in an
oil or water bath. The wet method may use either
fluorescent or visible particles, depending on the test
object. Because of the requirement for detecting very
small discontinuities, production aerospace
components are usually inspected with fluorescent
wet particles on a bench machine. Inservice testing is
generally performed with wet fluorescent particles, 70.00 ms per division
often provided in aerosolized containers, in
combination with electromagnetic yokes, portable (b)
coils, and power packs using cables and clamps
having braided copper contacts.

100.0 mV per division

Particle size and shape play a role in determining
how the particles will react under a magnetic field.
In general, the sensitivity of dry visible particles is
inverse to their size, small particles being more
sensitive than large particles. However, a very high
concentration of small particles is needed to produce
an indication of a large discontinuity, but a high
concentration of smaller particles could mask smaller
indications. The ideal situation for dry particles,
therefore, is to have a balanced mixture of larger 10.00 ms per division
and smaller particles. Wet visible particles are
usually small because it is difficult to hold large
the bench machine and letting it run for at least
particles in suspension.
30 min. It is good to brush the sides of the tank to
return precipitated particles to the suspension. Doing
so will help prevent adding particles unnecessarily,
Magnetic Particle Wet Bath which could make the particle concentration too
Intensity high. After 30 min., a clean 100 mL pear shaped
centrifuge tube is filled to the 100 mL mark,
The particle concentration of the magnetic demagnetized by moving the tube through a
particle wet bath is very important to the success of demagnetizer, and then placed in a vibration free
the test. If the concentration of particles in place for the particles to settle. The amount of time
suspension is too low, particles may be too few to the bath is allowed to settle depends on the bath
form an indication and small discontinuities may be liquid or vehicle. If the liquid is a petroleum
missed. If the particle concentration is too high, distillate, a minimum of 60 min. is normally
excess particles may mask the indication and make it prescribed. If it is a water bath, 30 min. is generally
hard to see. A high particle concentration may also long enough. After the settling time, the volume of
create a confusing fluorescent background over the settled particles is read and compared to the
test object surface, decreasing the necessary contrast recommended levels. The operator may need to add
between the surface and the indication itself. particles or a suspension vehicle to reach the correct
Maintenance of the suspension bath is important. As wet bath concentration.
particles are flowed over the test object, many are When the particle volume is read, the amount of
lost to dragout (that is, they cling to test objects loose foreign material (contamination) should be
removed from the bath) over time, requiring the noted. Excess foreign material (30 percent or more of
addition of particles to preserve the correct the particle volume) will require adjustment of the
concentration. Because the particles are of varying bath or perhaps replacement. The particle
sizes, this dragout effect may also reduce the concentration check is performed under an
proportions of particles in the suspension, lowering ultraviolet lamp. The fluorescence of the particles
testing sensitivity. For this reason, the bath should makes it easy to see both particle volume and
be replaced at regular intervals. amount of contamination. Contamination can be
Required bath concentration may be according to seen as a dark, nonfluorescent band or clumps in the
the manufacturer‘s recommendation or may be particle precipitate. Another check is performed at
defined in the applicable standard. For most this time to measure the fluorescence of the liquid
applications, 0.1 to 0.4 mL per 100 mL is acceptable. vehicle. Excess fluorescence requires adjustment of
The bath concentration is determined by energizing the bath or its replacement.

MAGNETIC PARTICLE TESTING 7.9

 Magnetic Particle Testing Standards
PART2 and Specifications

Standards Calibration and System

Magnetic particle testing is performed in
accordance with an appropriate standard. Table 2
Performance of Equipment
lists the relevant industry specifications from Magnetic particle equipment must be calibrated
SAE International and ASTM International for regularly to maintain proper efficiency. Items
magnetic particle testing. Note that testing standards needing calibration on stationary equipment are
may imply but do not dictate acceptance criteria. ammeter accuracy, timer control, and quick break
Acceptance criteria are usually agreed on or defined circuit. Yokes are calibrated using a so-called dead
by the customer. These standards also include weight test. All control instrumentation must be
calibration requirements for control equipment. calibrated on a periodic basis. The time period(s) for
calibration of these items can be found in
ASTM E 1444, Standard Practice for Magnetic

Table 2. Relevant industry specifications for magnetic particle testing.

Specification
Title Purpose
Number
ASTM E 125-97 Standard Reference Photographsfor Magnetic Providesphotographsoftypes and degreesofseverity for
Indications inFerrous Castings. discontinuities that occur in steel and other ferrous castings as
indicated by the dry powder magnetic particle method.

ASTM E 709 Guides for Magnetic Particle Examination. Describes techniques for both wet and dry magneticparticle
testing.

ASTM E 1444 Practices for Magnetic Particle Examination. Establishes minimum requirements for magnetic particle testing.

SAE AS 5282 Tool Steel Ring for Magnetic Particle Inspection. Establishes requirements for the manufacture and certification of
tool steel rings for magnetic particle testing. These rings are used
typically for determining the sensitivity of magnetic particles in
accordance with AMS 3040 through 3046 and for magnetic
particle systemperformancechecks. Therings are evaluated with
full-wave rectified alternating current through a copper
conductor bar.

SAE AS 5371 Reference Standards, Notched Shims for Establishes requirements for the manufacture of reference
Magnetic Particle Inspection. standards containing artificial discontinuities (notches) for use in
magnetic particle testing. Shims are applied to the surface of the
test object to establish field direction and to verify that sufficient
field has been produced.

SAE AS 4792 Water Conditioning Agents for Aqueous Describes water conditioning agents used to facilitate aqueous
Magnetic Particle Inspection. wet method magnetic particle testing. Such conditioning agents,
in powder or liquid form, provide suitable corrosion protection,
wetting, and particle dispersal when mixed in water for
application of magnetic particles on the surface of an object for
magnetic particle testing, as described in AMS 3042, AMS 3044,
AMS 2640, MIL-STD-1949, and others. Water conditioning agents
can consist of varying combinations of such components as
dispersants, surfactants, corrosion inhibitors, and antifoaming
agents. Individual components may be added to the bath of a
system to develop specific properties.

7.10 AEROSPACE NONDESTRUCTIVE TESTING

Particle Inspection (ASTM 2005). These control reliable testing near each end of the test object
instruments include the following: meters for the where leakage fields can otherwise cause excessive
ambient and visible testing light and ultraviolet background noise. The quick break feature generates
radiation; master shot timer used to calibrate the an induced current by the rapid collapse of the
machine shot timer; master shunt and ammeter used magnetic field, forcing the field out the end of the
to calibrate the machine ammeter; hall effect probes test objects.
and tesla meters used for determining magnetic field A check may also be performed for an internal
intensity; and field meters used to check residual short. The headstock is extended with nothing (or
magnetic fields. something nonconductive) between the heads to
Modern light meters are usually equipped with complete the circuit. A piece of wood may be used
two sensors, one for ultraviolet radiation and one for for this purpose. The machine is then energized, and
visible light. With the area of interest illuminated the ammeter is observed. Any ammeter reading
with the ultraviolet radiation or visible light, the above zero may indicate an internal short.
appropriate sensor is placed in the location required The particle suspension concentration check as
by the standard, usually where the test object is to be described above should be performed at the
processed. Minimum required intensity for frequency given in the applicable standard.
ultraviolet lamps is typically 1000 µW/cm²to A system performance check of magnetic particle
1200 µW/cm²measured at 38 cm (15 in.) from the equipment operation should be accomplished daily.
sensor. Values significantly higher than this are There are two basic methods to do a system
generally achieved in practice because standard hand performance check. Whichever method is used, it
held ultraviolet lamps equipped with 100 W mercury should be documented in a procedure and in a daily
vapor bulbs generally emit 2500 µW/cm²to 4000 system performance check record. The first method is
µW/cm²when the bulbs are new, and the distance to use a test object with known discontinuities or a
between the surface of the test object and the lamp is test object fabricated with artificial discontinuities. If
often 15 to 20 cm (6 to 8 in.). Because of the inverse the known discontinuities can be visually identified,
square law, surfaces frequently need to be examined then the equipment is considered to be performing
with irradiation at 10 to 20 mW/cm² . Ambient light within acceptable limits. Another common method is
in the testing booth is usually a maximum of 22 lx to use a ketos ring or an AS5282 ring (shown in
(2 ftc) and may require measurement with the ASTM E 1444, Appendix X1) with side drilled holes
ultraviolet lamps on, depending on the specification. simulating discontinuities at various depths.
Some manufacturers require 5.4 lx (0.5 ftc) Figure 13 is a diagram of a test ring. The AS5282
maximum, and keeping visible light levels to the ring is preferred for aerospace applications because
lowest practical level is generally considered good of the consistency of its response due to the
practice. Evaluation of indications under white light
typically requires visible light intensity values of
1.1 klx (100 ftc) at the testing surface. Note that the Figure 13. Magnetic particle test ring.
inspector must adapt his or her vision to light
conditions before testing.
The shot timer is usually checked using a D
calibrated master shot timer. The probe is clamped 12
between the head and tailstocks, the timing device is
set for the required pulse length, and then the 11
machine is energized. The reading on the shot timer
is then compared to the pulse length set on the
10
machine. 1
In a similar way, the master shunt and ammeter
are used to calibrate the machine ammeter. The 9 2
shunt is clamped between the head and tailstocks,
the machine intensity is set to the required value(s), 3
and the machine is energized. The master ammeter 8
4
reading is then compared to that of the machine 7
ammeter. 6 5 19 mm (0.75 in.)
If a bench machine is equipped with a quick
break circuit, the circuit can be checked using a 32 mm
quick break fixture. The quick break fixture usually (1.25 in.)
consists of a special circuit box equipped with a red
light emitting diode that lights when the coil is 125 mm (5.0 in.)
energized and the quick break is functioning
correctly. The box is placed into the coil, and the
machine is energized at 1000 A minimum. The red
light emitting diode must light each time a shot is 22 mm
given. The quick break feature is normally required (0.87 in.)
only on three-phase full-wave rectified and direct
current powered equipment. It is intended to ensure

MAGNETIC PARTICLE TESTING 7.11

 SAE International specification requirements for its
manufacture. The ring is placed on a 25 mm (1 in.)
Electrical Safety
central conductor, and the conductor is clamped A major safety concern when using magnetic
between the heads of the machine. The machine is particle testing in stationary units is the electrical
then energized at various amperages to induce grounding of the equipment. All stationary
magnetic fields of increasing intensity in the ring. At equipment must have approved grounding before the
each current level, magnetic particles are flowed equipment is used. In addition, a shock resistant
over the ring, and the ring circumferential surface is rubber mat should be placed in front of the machine
observed to determine the number of indications to provide additional safety for the operator. At all
visible. The required sensitivity level on the ring may times, the operator must comply with the
be found in the applicable magnetic particle testing manufacturer‘s safety precautions. Although a bench
standard. machine may able to supply thousands of amperes of
current, for safety reasons voltage is restricted to
32 V or less. Typical bench machines use voltages of
12 to 18 V. Power packs for portable use may range
up to 30 V.

7.12 AEROSPACE NONDESTRUCTIVE TESTING

 Magnetic Particle Testing
PART3
Applications

The uses of magnetic particle testing techniques requirements that must be complied with during
in aerospace are varied. Some typical applications inservice, overhaul, and rebuild processes. Figures 14
are aircraft wheel bolts, aircraft brake disks and and 15 show some typical applications using wet
rotors, aircraft brake keys, landing gear axles, horizontal magnetic particle testing equipment.
landing gear components and attach points, aircraft Figure 14 shows an encircling coil that induces a
wing connection bolts, engine mounting points, longitudinal field to detect circumferentially oriented
engine mounting bolts, bearing races, engine cracks. The sample can be placed anywhere in the
shafting and gears, and helicopter rotating coil but is usually placed close to the inside diameter
components. so that lower amperage may be used. Figure 15
Among the advantages of magnetic particle shows direct conduction through the test object for
testing are portability and ease of use. In addition to circular magnetization. In this case, the field lines
its wide use in the production of aerospace are circular on the surface of the test object and are
components, it also has important applications for used for the detection of longitudinally oriented
inservice aircraft of all kinds. Because it can be cracks.
performed using an oil vehicle, corrosion prone
components may be safely inspected without water.
Magnetic particle testing can also reveal Figure 15. Magnetic particle testing: (a) head shot; (b) central
near-surface discontinuities that cannot be seen by conductor through test object.
fluorescent penetrant. The testing process is also
(a)
much faster than for fluorescent penetrant testing.
Many maintenance manuals for aircraft systems
components have magnetic particle testing

Figure 14. Aviation magnetic particle testing

using wet horizontal units: (a) using encircling
coil; (b) clamping test object between
conduction heads.
(a)

(b)

(b)

MAGNETIC PARTICLE TESTING 7.13

 Issues holes in the end of a flap track. Magnetization from
inside the hole provides for reliable detection of
As shown in Table 3, magnetic particle testing radial cracks around the full circumference of the
has advantages and limitations when used for the hole. (A quantitative quality indicator can be seen on
detection of surface and near-surface discontinuities the top surface.)
in ferromagnetic materials. With proper use as a
surface testing technology, it is has been found to be
one of the most efficient methods in terms of cost,
reliability, and sensitivity. Table 4 lists some issues Figure 17. Circular magnetization with insulated
that can arise from incorrect operations. central conductor: (a) central conductor inserted
in flap track holes; (b) magnetization of central
conductor.
Small Test Objects
Figure 16 illustrates clamps and cable blocks that
(a)
can be used to create a circular field when the
contact areas of the test object are small and
therefore vulnerable to arc burning.

Flap Track
Figure 17 illustrates an insulated central
conductor used to create a circular field around the

Figure 16. Circular magnetization using clamps and cable block:

(a) clamps with block; (b) clamps connected to flap track to magnetize it.
(a) (b)
(b)

Table 3. Advantages and limitations of magnetic particle testing.

Test detects surface and near surface anomalies in ferromagnetic materials, that is, materials that can be
magnetized.
Test detects both production discontinuities (such as seams, laps, and quenching and grinding cracks) and inservice
Advantages fatigue cracks.
Test is sensitive to small cracks and discontinuities.
Test is rapid and low in cost.
Indications are produced directly on the surface.

Component must be ferromagnetic.

Intensity of the magnetic field must be adequate for the test object being inspected.
Properlighttype and intensity are necessary.
Demagnetization is generally required.
Limitations Alignment of magnetic field and discontinuity is critical.
Relatively smooth surface is required.
Coatings may adversely affect sensitivity.
Cleaning is necessary before and after testing.
Complex test object geometries are difficult to test.

7.14 AEROSPACE NONDESTRUCTIVE TESTING

Table 4. What can go wrong with magnetic particle testing.

Problem Cause Solution

Bath concentration too low. Insufficient particle concentration due to Add particles.
magnetic dragout.

Bath concentration too high. Evaporation or dragout of vehicle. Add suspension vehicle.

Bath contaminated. Improper precleaning of test objects. Filter contamination out of bath or change
bath. Properly preclean test objects.

System fails performance check. Preventive maintenance not compliant. Follow preventive and calibration requirements
per applicable standard.

Wrong procedures followed. Perform maintenance on equipment as required.

Wrong standard used.

Malfunction in equipment.

Ultraviolet radiation intensity too low. Low operating voltage. Check for correctoperatingvoltage.

Weak bulb. Change bulb.

Dirty lens. Clean lens.

Incorrect measurement procedure. Follow measurement procedure.

Measuring equipment faulty or out of Verify measuring equipment operation and

calibration. calibration.

No magnetic field detected in or on test No flux leakage from test object. Test object not ferrous.
object.
Wrong field applied (circular versus
longitudinal).

Insufficient magnetizing force applied.

Field intensity indicator inoperative.

Magnetic field in test object too strong. Overapplication of magnetizing force applied. Demagnetize and applycorrectmagnetizing
force.

Electric burn or arc marks on completed Testobject notproperlysecured in contact Ensure full contact with headstock and tailstock
test objects. heads. pads.

Prods used on aircraft parts. Do not use prods on aircraft parts.

Magnetic particle indications weak. Insufficient magnetic field. Increase magnetic field intensity.

Insufficient particle concentration. Check wet bath concentration.

Wrong field direction applied. Apply correct field direction for direction of
suspected discontinuity.

Excess particles mask ends of test Excessmagnetic field intensity. Reduce magnetic field intensity.
object.
Excess wet bath concentration. Check wet bath concentration.

Test object will not become Wrong field applied. Usecorrect field (direct currentversus
demagnetized. alternating current).

Insufficient field intensity applied. Use stronger magnetic field to demagnetize.

Consider movingtestobject to east/west

direction.

MAGNETIC PARTICLE TESTING 7.15

 Flap Carriage Landing Gear Wheel Flange
Figure 18 illustrates the use of a special fixture Figure 19 illustrates an electromagnetic yoke
that allows a flap carriage to be clamped between being used to generate a longitudinal field in a
the heads to generate a circular field. This circular landing gear wheel flange to look for cracks
field was necessary both because of the offset radiating from the brake attach holes. This test could
between the spindle and the roller flanges and have also been accomplished using a portable power
because of the small contacting surfaces at the ends pack and central conductor through the holes.
of the roller flanges. The test object can be held in However, the cracks of interest extend inward from
this position also during longitudinal magnetization the hole toward the axle.
using the coil, but higher amperages would be
required because the test object would be near the
center of the coil. Notice that the complex shape of Figure 19. Longitudinal magnetization using
the test object called for a large number of electromagnetic yoke.
quantitative quality indicators: six are visible as
small gray patches on the flange surface.

Figure 18. Longitudinal and circular magnetization: (a) setup with

fixtures; (b) closeup of tailstock.

(a)

(b)

7.16 AEROSPACE NONDESTRUCTIVE TESTING

References
ASNT. 2008. Nondestructive Testing Handbook 8: Magnetic Testing,
3rd edition. Columbus, OH: American Society for Nondestructive Bibliography
Testing. Callister, W.D., Jr. 1997. Materials Science and Engineering. New York,
ASTM. 2005. ASTM E 1444, Standard Practice for Magnetic Particle NY: John Wiley and Sons.
Testing. [Supersedes MIL-STD-1949A.] West Conshohocken, PA: Halmshaw, R. 1991. Non-Destructive Testing, 2nd edition. London,
ASTM International. United Kingdom: Edward Arnold.
Lindgren, A. 1989. ―Magnetic Particle Testing,‖ Metals Handbook 17,
9th edition: Nondestructive Evaluation and Quality Control.
Materials Park, OH: ASM International.
Lindgren, A., P.[J.] Shull, K. Joseph, and D.[J.] Hagemaier. 2002.
―Magnetic Particle.‖ Nondestructive Evaluation: Theory, Techniques,
and Applications. New York, NY: Marcel Dekker.

MAGNETIC PARTICLE TESTING 7.17

 X
8
CHAPTER

Electromagne
tic Testing
Contents
Part 1. Principles of
Electromagnetic Testing, 8.2
Part 2. Eddy Current Coil
Techniques, 8.4
Part 3. Nonconventional
Electromagnetic Methods, 8.18
Part 4. Relevant Standards and
Specifications, 8.25
Part 5. Aerospace Applications
of Eddy Current Testing, 8.27
References, 8.48

Contributors
Flynn Spears David A. Raulerson
John C. Brausch Jeffrey G. Thompson
Richard H. Bossi Andrew P. Washabaugh
David [Link]
8.1
PART1
Principles of
Electromagnetic Testing

Electromagnetic testing is an important and primary field, an electrical current will be induced in
widely used method within the broad field of the test object. These currents, known as eddy
nondestructive testing. Electromagnetic test methods currents, are generated in a conductor by changing
employ various submethods, or techniques: eddy magnetic fields. Faraday discovered that the
current testing, remote field testing, flux leakage movement or oscillation of a magnet was reduced or
testing, magneto-optic testing, and magnetoresistive rapidly damped when a nonmagnetic conductive
testing. Eddy current testing is one of the most disk was placed near the magnet. Örsted observed
widely used methods for aerospace materials because that, when a compass is placed near an electrical
of its application to inservice inspection for current, the position of the compass needle moves
aluminum structures. Definitions of terms used in perpendicular to the direction of the current. Örsted
eddy current and electromagnetic testing are further discovered the presence of a magnetic field
included in the glossary. around a current carrying conductor and observed
that equal and opposite currents flowing in adjacent
conductors cancelled this magnetic effect.
Figure 1 shows a simple model of the relationship
Electromagnetic Theory of primary and induced (eddy) currents. Conductor A
From 1775 to the mid-1900s, several scientific represents a section of a test coil. Conductor B
experiments investigated the many facets of represents a section of the object under test. When a
electromagnetism. These scientific experimenters — primary current is applied to the test coil, a primary
Coulomb, Ampere, Arago, Örsted, Maxwell, and field is developed about conductor A. Bringing
Kelvin — all investigated and cataloged most of what conductor B into the primary magnetic field induces
we know today about magnetism and a secondary current, an eddy current, in conductor
electromagnetism. Electromagnetic induction was B. The induced current produces an electromagnetic
not observed and explained before the 19th century. field that opposes the primary field and is directly
James Clerk Maxwell summarized the first 50 years proportional to the magnitude of the secondary
of this development in his Treatise on Electricity and (eddy) current. Changes in conductor B such as
Magnetism (Maxwell 1891). Faraday discovered the conductivity, permeability, or geometry will cause
process of electromagnetic induction. When an the eddy current to change. When the eddy currents
electrically conductive test object is placed in the change in conductor B, the intensity of the
electromagnetic field about conductor B will also
change. These changes or variations from the
secondary electromagnetic field affect the current in
Figure 1. Alternating current in wire results in magnetic field that conductor A. The detection of these changes can be
induces opposite alternating current in adjacent wire. displayed in some form and used to characterize an
object under test.
By inducing eddy currents into a conductive
material and electronically monitoring the field
Conductor B behavior, the material properties or characteristics
(test object)
can be assessed. With newer technology, specially
wound sensor coils can be fabricated with printed
circuit techniques in an array configuration. The
iP sensor coil is energized with an alternating current
power source to create a changing magnetic field
around the coil. This action of the magnetic field
from the coil in a nearby conductor is called
induction. The ability to induce eddy currents is a
ΦE ΦP function of the material conductivity, permeability,
and geometry of the wire windings and test piece.
Both conventional and unconventional eddy
iE
Conductor A current techniques rely on Faraday‘s law of
(test coil) electromagnetic induction:

Legend
E B
(1)  
iP = primary current t
ΦP = primary magnetic field
iE = secondary (eddy) current In differential form, this law describes the
ΦE = secondary magneticfield connection between the magnetic field vector B and

8.2 AEROSPACE NONDESTRUCTIVE TESTING

the electric field vector E at points of
three-dimensional space at time t (second), including Figure 2. Illustration of induced currents from coil above test object
points inside electrical conductors. In particular, this (ASNT 2004, 7).
Primary
law shows that a time varying magnetic field B Coil in eddy electromagnetic field
(produced by a moving permanent magnet, the field current probe
from a coil of wire carrying a changing current or
some other source) in the vicinity of any electrical
conductor having conductivity  will induce a time Direction of
primary current
varying electric field E and thus a time varying eddy
current density J at an arbitrary point near the
Induced field
surface of the conductor:

(2) J = E
Direction of
eddy currents
To gain a general understanding of the relevant
Conducting specimen
relationships and concepts, consider the familiar case Eddy current intensity
of a time varying magnetic field B produced by an decreases with increasing depth
external alternating current in a coil of wire. By
Lenz‘s law, the direction of the current density vector
J at any point in the conductor, and hence the
direction of the eddy currents, is always opposed to Figure 3. Eddy currents induced in conductive plate by magnetic field
the change in the direction of the external electric B produced by wire coil carrying alternating current: (a) view of eddy
currents that produced magnetic field B in the first current distribution in plate; (b) view of eddy current penetration into
place. This opposition is illustrated in Figures 2 and plate (ASNT 2004, 160).
3, where a coil carrying an alternating current near a
conductive plate has induced eddy currents in the (a)
plate. The magnetic fields in Figure 3 tend to be
excluded from the conductive plate, especially at Coil current
high frequencies. The magnitude of the eddy currents B
diminishes as the depth increases. Eddy currents
The ability of a conductor to induce a voltage in
itself is called self-inductance or simply inductance, B B
and its unit of measure is the henry. The inductance B
Conductor B
of a coil depends on the number of turns or
windings, the diameter of the wire, and the core
material around which the wire is wound. Wire by (b)
itself will have a certain amount of inductance.
Exploring the field around a very long conductor
with a compass needle reveals that the lines of the
magnetic field are circles with their centers on the Eddy
wire (Figure 4). A convenient way to visualize the Coil current currents
direction of the field about a wire is known as the
B B Skin depth
―right hand rule‖: if a right hand grasps a
conducting rod or wire whose current is traveling in
the direction that the thumb is pointing, then the
magnetic flux lines are traveling around the
conductor in the direction that the four fingers are Figure 4. Magnetic field around coil inducing eddy currents into
pointing. The magnetic lines of force, or flux, are surface.
always at right angles to the direction of flow of the
electric current. The flux density of the magnetic
field is greatest at the surface of the wire, or coil.
Therefore, flux density of the total longitudinal
magnetic field will be greatest at the inside of the Alternating current
coil.

Direction of coilfield
Directionofeddy
current field
Eddy currents

Discontinuity

ELECTROMAGNETIC TESTING 8.3

 Eddy Current Coil Techniques
PART2

Basic Eddy Current Testing For electrical circuits, with sinusoidal voltages
and currents, the complex impedances cause phase
Introduction shifts between the voltages and currents. The
amplitudes and phase angles can be represented
Eddy current techniques for nondestructive graphically on a vector plot and are related to the
testing are based on the principles of complex impedances analyzed with an impedance
electromagnetic induction. When a coil is excited by plane plot. Figure 5 shows two sinusoidal waves
an alternating current and is placed near a 90 degrees out of phase. The blue waveform has an
conductive test object, the impedance of the coil amplitude of 1.2 and leads the red waveform by
changes. This change is associated with the fact that 90 degrees whereas the red waveform has an
the primary field set up by the eddy current coil amplitude of 0.7 and lags the blue waveform.
induces eddy currents within the electrically A vector plot of the two waveforms is shown in
conductive object. For basic discontinuity detection, Figure 6.
a discontinuity in the test object causes
redistribution and reduction of the eddy currents
within the test object and consequently changes the
excitation coil impedance. The changes in the Figure 6. Vector plot of Figure 5. Angle
excitation coil impedance are used to infer
represents difference in phase anglebetween
information about the discontinuity.
two waveforms.
90 degrees
Impedance Plane Analysis
The coil impedance is comprised of the coil 2
inductance due to coil size, shape, number of turns,
and the resistance of the wire used to make the coil.
The inductance causes the coil impedance to be 1
complex (not just a resistance) and therefore requires 180 degrees 0 degrees
vector techniques to analyze the impedance and
impedance changes. Impedance plane analysis is a
vector graphical technique used to analyze electrical –2 –1 1 2
circuits and plot complex impedances. Impedance
–1
plane analysis is typically used to show circuit
impedance variations due to varying frequency. For
eddy current testing, it is also used to represent other –2
influences on impedance such as discontinuities,
material properties, and liftoff of the coil from the
surface being tested. This analysis approach applies 270 degrees
to eddy current testing with sinusoidal waveforms.

Figure 5. Sinusoidal waveforms out of phase by 90 degrees.

1.5

1.0
Amplitude

0.5

–0.5

–1.0
720 900 1080
–1.5
0 180 360 (degrees)540
Angle representing time

8.4 AEROSPACE NONDESTRUCTIVE TESTING

 Eddy current probes are made up of coils and demodulation. Quadrature demodulation involves
cables with complex electrical impedances. The multiplying or mixing an in-phase signal (cosine)
complex impedance has a real (resistive) and an and a 90 degree out-of-phase signal (sine) with the
imaginary (reactive) component. With resistive signal from the eddy current coil. This is typically
components, like a resistor, the current and voltage accomplished with the probe drive signal (in-phase)
are in phase. In reactive components, like inductors and a phase shifted version of the probe drive signal
(coils) and capacitors (cables), the voltage and (Figure 9). The mixing process results in sum and
current are out of phase by ±90 degrees, respectively. difference frequencies: the baseband frequency and
The probe and cable are made up of resistance, twice the drive frequency. After mixing, the low pass
capacitance, and inductance, which form an filtering removes the higher frequency (2 drive)
electrical circuit. The actual electrical schematic is component, resulting in two low frequency signals
complicated, but Figure 7 represents the dominant
contributions to the impedances. At normal
operating frequencies for typical inspections, the Figure 7. Simple electrical schematic for eddy current coil.
cable capacitance has a minimal effect on the total
impedance and therefore will be neglected for the
rest of this discussion. The focus will be on the coil
resistance and inductance only.
Complex impedances are often represented in
rectangular form, readily plotted with R
two-dimensional cartesian coordinates. This form is
referred to as an impedance plane when analyzing Cc
impedances and is commonly used to analyze eddy
current probe responses. The resistive component is
L
plotted on the positive horizontal axis, and the
reactive component is plotted on the positive vertical
axis. The reactive component is due to the
inductance and is a function of frequency: Legend
Cc = capacitance of cable
L = inductance
R = resistance of coil wire
(3) L  2fL  XL

where f is frequency, XL is the inductive reactive

component of impedance, and  is the angular
frequency. Figure 8. Impedance plane for eddy current coil: (a) vector
representation; (b) point representation (upper right quadrant).
Figure 8a shows an impedance plane with real
component R and inductive reactive component XL
plotted along the real and imaginary axes, (a)
respectively. The series combination of R and XL is X+
the total impedance. The total impedance Z is the
vector sum of R and XL. Figure 8b shows only the
Zm
impedance plane‘s upper right quadrant, pertinent
for eddy current testing analysis. L1
Variations in material electrical properties,
discontinuities, and geometry will affect the strength
and distribution of induced eddy currents. Changes 
in the eddy current flow will result in changes in the R– R+
inducing coil‘s current. These changes can be R1
expressed as a change in the coil‘s electrical
impedance, so changes in material properties can be
associated with specific changes in the impedance of
either the excitation coil or sensing/pickup coils.
Modern eddy current instruments measure
voltages proportional to the impedance changes and
display them on an oscilloscopic display referred to X–
as an impedance plane display. They operate at
frequencies ranging from about 10 Hz to 40 MHz. (b)
X+
The frequency is selected for the requirements of
each application. The probe drive voltage is assumed
to be sinusoidal, but there are pulse driven
instruments too. Z
The impedance plane display is generated by the XL
electronic processing of the eddy current signal.
Changes in coil impedance result in changes in the
amplitude and/or phase of the coil voltage. The
amplitude and phase changes are extracted from the R
eddy current signal by a process called quadrature R+

ELECTROMAGNETIC TESTING 8.5

 Figure 9. Simple schematic of analog eddy current instrument input circuitry.

Driver
Oscillator
Ground
Ground

Y Low passfilter
Receiver 1 X
Receiver 2 Real channel

90-degree
phase shift

Y
Low pass filter
X
Imaginary channel

proportional to the real and reactive components of

Figure 10. Impedance plane plot (ASNT 2004, 472). coil impedance. These are the displayed eddy current
signals.
Figure 10 shows an impedance plane plot for
three materials; aluminum, copper, and steel.
C Typically, impedance plane data are shown in a
Steel normalized format where the impedance in air is
normalized to a reactance value of 1. In Figure 10,
the nonmagnetic materials, copper and aluminum,
are below the air line and the magnetic material,
steel, is above the air line. This difference occurs
µ because the high relative magnetic permeability of
Liftoff magnetic materials causes an increase in reactance.
For the nonmagnetic materials, the inductive
Normalized reactance X (relative scale)

reactance of the coil is reduced because of the eddy

Pn
currents. Trajectories from the metal points vary
significantly based on various conditions such as
coating and thickness variations. Note that liftoff of
Magnetic
Air the coil from any of the materials leads to the same
Nonmagnetic air point.
The impedance of a coil can change when it is
placed next to or in direct contact with an
n electrically conductive or ferromagnetic part.
Liftoff Figure 10 shows the impedance change relative to
S
C multiple conditions which include varying liftoff,
varying conductivity and varying permeability,
T thickness, cracks, plating, and layer spacing.
Figure 11 shows how the impedance of a coil
changes as it comes in contact with
Pa Aluminum nonferromagnetic or ferromagnetic conductive
specimens with and without discontinuities. In
Pc Figure 11a, the nonferromagnetic conductive
Copper material causes the complex impedance to change in
the direction toward the air impedance. This is
because the eddy currents are being disrupted by the
Normalizedresistance R (relative scale) discontinuity, thus reducing the magnetic field
associated with the eddy currents. And this causes
the coil magnetic field to increase towards the air
Legend value. In Figure 11b, the ferromagnetic conductive
C = crack S = spacing between aluminum layers material causes the coil magnetic flux to increase
Pa = plating aluminum on copper T = thinning in aluminum because the material magnetic permeability has a
Pc = plating copper on aluminum µ = permeability
greater effect than the material conductivity. The
Pn = plating nonmagnetic n = conductivity nonmagnetic materials discontinuity, in this case, causes the complex

8.6 AEROSPACE NONDESTRUCTIVE TESTING

impedance to increase more than the case without
the discontinuity because the effect of the Figure 11. Impedance plane plots effects: (a) nonferromagnetic
conductivity is being reduced by the discontinuity. specimens; (b) ferromagnetic specimens (ASNT 2004, 51).
In both cases, the discontinuity‘s effect on the (a)
impedance is related to the distance between points
2 and 3 in Figure 11.

Reactance X (relative scale)

1

Filters
Eddy current instruments typically have the
2
ability to adjust internal filters to suppress noise and
unwanted signal responses. The two types of filters
3
commonly employed are the high pass filter and the
low pass filter. In some cases, band pass filters are
also available, combining both low and high pass
filters. By using these filters, the signal response can Resistance R (relative scale)
be restricted to a specific range of frequencies, while
suppressing frequencies above and below this range,
resulting in less signal distortion and noise from (b)
nonrelevant sources as shown in Figure 12. The high

Reactance X (relative scale)

pass filter allows high frequencies to pass, or 2
suppresses low frequencies which are produced by
slow or longer periodicity changes such as those
indicating hole ovality. The low pass filter allows 3
low frequencies to pass, or rejects high frequency
noise generated by the probe or instrument. 1
The filter frequencies referred to here are related
to the frequency response of the signal generated by
the coil as it passes over a discontinuity at a given
speed. It is essentially a measure of the time Resistance R (relative scale)
(duration) that the coil is sensing a fatigue crack and
registering a response. Often referred to as the peak Legend
rotational frequency, this discontinuity frequency is 1. Coil in air.
measured in hertz (Hz) and is independent of, and 2. Coilover specimenwithdiscontinuity.
should not be confused with, the test, or oscillator, 3. Coil over specimen without discontinuity.
frequencies typically measured in kilohertz (kHz) or
megahertz (MHz).

Figure 12. Illustration of high pass filter and low pass filter selection
to allow selected frequencies to pass.
Factors That Can Affect Eddy
Current Testing Passed
If the test system or material properties (such as frequencies
conductivity, magnetic permeability, or geometry)
Increasing signal
attenuation (dB)

high pass filter

are changed, then the eddy current flow induced in a low pass filter
Frequencies
Frequencies

rejected by
rejected by

Indication
frequency

conductive material can also change. This, in turn,

changes the coil‘s impedance. It is a cascade event.
As one part of the picture changes, everything else
shifts in response. The signal that is detected during
an inspection might be created by some combination
of material changes. The following sections review 0 100 200 300 400 500 600 700 800 900 1000
how various factors can affect the response of an Frequency (Hz)
eddy current test system.

Conductivity amount of time. Each material from carbon to

copper has a unique conductivity value. The unit of
A material‘s conductivity, the reciprocal of its conductance is the siemens (S), the reciprocal of an
resistivity, is an electrical property that provides a ohm (Ω), but conductivity is commonly expressed as
measure of how well electrons travel through a a percentage of the International Annealed Copper
material. The ability of electrons to move in a given Standard (IACS). This standard is based on a specific
conductor is affected by its alloy content, grade of high purity copper, defined as having an
temperature, heat treatment (hardness), and stress. electrical conductivity of 100 percent and a value of
Conductivity can further be defined as a measure of 5.8  107 S/m at 20 °C. Other materials are defined
the ability of electrons to flow through the atomic as having a percentage of this standard. Table 1 lists
lattice of a material. The higher the conductivity conductivities of materials with different chemical
value, the greater will be the number of electrons compositions. The resistivity values are also given,
that can pass through the material in a given where 1 IACS is 1.72  10–6 Ω·m.

ELECTROMAGNETIC TESTING 8.7

Table 1. Electrical resistivity and conductivity of selected metals (ASNT 2004, 53).

Conductivity Resistivity
Metal
MS/m (percent IACSa) Ω·m (µΩ·cm)

Aluminum, pure 35.38 (61.00) 2.83  10–8 (2.83)

Aluminum (99.99 percent) 37.67 (64.94) 2.65  10–8 (2.65)
Antimony 2.55 (4.40) 3.92  10–7 (39.18)
Bronze, commercial annealed 25.52 (44.00) 3.92  10–8 (3.92)
Cadmium 14.62 (25.20) 6.84  10–8 (6.84)
Calcium 28.25 (48.70) 3.54  10–8 (3.54)
Chromium 5.10 (8.80) 1.96  10–7 (19.59)
Cobalt 16.01 (27.60) 6.25  10–8 (6.25)
Copper 58.00 (100.00) 1.72  10–8 (1.72)
Gold 40.60 (70.00) 2.46  10–8 (2.46)
Iron, pure 10.44 (18.00) 9.58  10–8 (9.58)
Iron ingot (99.9 percent) 9.05 (15.60) 1.11  10–7 (11.05)
Magnesium, pure 22.39 (38.60) 4.47  10–8 (4.47)
Molybdenum 19.14 (33.00) 5.22  10–8 (5.22)
Nickel 14.62 (25.20) 6.84  10–8 (6.84)
Selenium 8.35 (14.40) 1.20  10–7 (11.97)
Silver, tin solder 9.63 (16.60) 1.04  10–7 (10.39)
Steel, high alloy 1.68 (2.90) 5.94  10–7 (59.45)
Tin, pure 8.70 (15.00) 1.15  10–7 (11.49)
Tin foil 2.44 (4.20) 4.10  10–7 (41.05)
Tungsten 18.21 (31.40) 5.49  10–8 (5.49)
Zinc, commercial rolled 16.24 (28.00) 6.16  10–8 (6.16)
a. International Annealed Copper Standard.

Other factors affecting conductivity are Eddy currents provide a means for accurately
temperature, heat treatment, grain size, hardness, and measuring the comparative conductivity of materials.
residual stresses. A change in the temperature of the The flow of eddy currents is directly related to the
test object will change the electrical conductivity of material‘s conductivity. If the conductivity increases,
the material of that object. In metals, as the the flow of eddy currents increases, and conversely
temperature is increased, the conductivity is if the conductivity decreases, the flow of eddy
decreased. Carbons and carbon compounds have currents decreases. Figure 13 shows the locus of
negative temperature coefficients; therefore, their points on an impedance plane display of a coil for
conductivity increases as temperature is increased. the change from air to contact with nonmagnetic
Heat treatment also affects electrical conductivity by materials having differing conductivity values. In
redistributing elements in the material. Depending each case, the conductivity values are in units of
on the materials and the degree of heat treatment, megasiemens per meter with percent International
conductivity can either increase or decrease as a Annealed Copper Standard in parentheses. The
result of heat treatment. Stresses in a material due to impedance of a test coil varies with the conductivity
cold working produce lattice distortion or dislocation of a nearby material. The coil‘s inductive reactance
entanglements. This mechanical process changes the is plotted on the Y axis, and coil resistance is plotted
grain structure and hardness of the material, thereby on the X axis. In these figures, the zero (0)
changing its electrical conductivity. Hardness in ―age conductivity point, or air point, occurs when the
hardenable‖ aluminum alloys decreases the electrical coil‘s magnetic field is not influenced by anything
conductivity of the alloy. For example, a brinell but air.
hardness of 60 is represented by a conductivity of
13.3 MS/m (23 IACS) whereas a brinell hardness
of 100 for the same alloy would have a conductivity
of
11.0 MS/m (19 IACS).
8.8 AEROSPACE NONDESTRUCTIVE TESTING
Magnetic Permeability field H increases, the concentration of the lines of
force becomes greater. The number of lines in a
The magnetic permeability provides a measure of given unit area is called the flux density B. The
the ease with which the magnetic domains of a magnetic permeability  is the ratio of the flux
material align themselves with an externally applied density B through a material to the applied field
magnetic field. This can be understood by magnetic field strength H. Materials with high
representing a magnetic field as lines of force that magnetic permeability tend to concentrate the
form a closed path. As the strength of the magnetic magnetic flux. This flux density can limit the depth

Figure 13. Conductivity loci of points on impedance plane display: (a) conductivity curve in siemens (%
IACS); (b) conductivity and liftoff loci (ASNT 2004, 54, 325).

(a) Air, 0 (0) Conductivity

Reactance X (relative scale)

1.2 (2)

2.9 (5)

5.8 (10)

58 (100)

Resistance R (relative scale)

(b)

0 IACS (air) Conductivity locus

Graphite, 0.1 MS/m (0.2 IACS)

 Titanium alloy, 0.6 MS/m (1.0 IACS)

Inductance X (relative scale)

Stainless steel, 1.6 MS/m (2.7 IACS)

Liftoff loci Bronze, 5.8 MS/m (10 IACS)

Aluminum alloy, 20.3 MS/m (35 IACS)


Copper, 58 MS/m (100 IACS)

Resistance R (ratio)

Legend
IACS = International Annealed Copper Standard
 = angle of separation between liftoff curve and conductivity curve

ELECTROMAGNETIC TESTING 8.9

 of penetration of the field being applied for variations. To remove these permeability effects
examining a test object. Because of the effect of during an inspection, a direct current coil can be
permeability on the applied field, for ferromagnetic used with a regular alternating current test coil. The
materials other electromagnetic testing methods such direct current coil establishes a magnetic field that
as magnetic bias probes, remote field testing, and causes the flux density in the magnetic material to
flux leakage (applicable only for large volume remain saturated. Under this condition, the
discontinuities) may be suitable alternatives to permeability is constant and the only flux changes
standard eddy current testing of ferromagnetic within the article will be from the test coil.
materials. As an example, consider a magnetic steel bar
Permeability is further defined as a measure of placed in an encircling coil. As the coil current is
the ease with which a material‘s atoms can be increased, the magnetic field of the coil will increase.
aligned, or the ease with which it can establish lines The magnetic flux within the steel will increase
of force. Materials are rated in a scale that compares rapidly at first and then will tend to level off as the
them with each other. Air is assigned a relative steel approaches magnetic saturation. When
permeability of 1. A basic determination of increases in the magnetic force produce little or no
permeability  is obtained: change on the flux within the steel bar, the magnetic
domains in the material are aligned and the bar is
Number of flux lines magnetically saturated. When ferromagnetic
produced with material as core materials are saturated, the permeability becomes
(4)   Number of flux lines constant. In a practical sense, this is accomplished
produced with air as core using a coil that is driven by a direct current source
or a large permanent magnet. In practice, it is not
Ferromagnetic metals and alloys, including possible to do this for typical eddy current aerospace
nickel, iron, and cobalt, tend to concentrate inspections in the field.
magnetic flux lines. Ferromagnetic materials or
sintered ionic compounds are also useful in
concentrating magnetic flux lines. Many other Operating Frequency
materials used in aerospace applications, including Although all electromagnetic methods are
aluminum, titanium, and carbon/graphite fiber governed by Maxwell‘s equations, the distinctive
composites, are nonferromagnetic and have a nature of each method stems from differences in
relative permeability of 1. excitation frequencies, the nature of the probes used,
Eddy current testing can be applied to both and the signal analysis techniques for characterizing
conducting and magnetic materials. For example, for the state of the test object. Eddy current testing
discontinuity detection (with possible sizing) in operates in the middle range of frequencies (50 Hz to
nonferromagnetic materials, the responses to both 10 MHz), compared to magnetic flux leakage testing
near surface and far surface discontinuities are (low frequencies, near 0 Hz) and microwave testing
displayed as changes in the local, effective (high frequencies, in excess of 100 MHz).
conductivity of the material. In ferromagnetic
materials, the permeability leads to a shallow depth
of penetration and often limits the inspection to Skin Effect/Depth of Penetration
surface or near-surface discontinuities. Electromagnetic tests in many applications are
Inspection of ferromagnetic materials calls for most sensitive to test object variables near the test
special probes. In these materials, permeability is coil, because of skin effect. Skin effect is a result of
generally not constant but depends on the magnetic mutual interaction of eddy currents with operating
field acting on the material. In addition, permeability frequency, test material conductivity, and magnetic
variations within the material itself may mask the permeability. These three variables define the skin
signals produced by discontinuities or conductive depth or the standard depth of penetration  at
which the eddy current strength or intensity is
reduced to 37 percent of its surface value. The
Figure 14. Curve of intensity versus depth of penetration, where  is standard depth of penetration is given by:
standard depth of penetration.
1
100 (5)  =
f 0r

where f is the coil operating frequency (Hz), r is the

relative permeability of the material being tested, o
is permeability of free space (4π  10-7 H/m), and 
Intensity (percent)

is conductivity (S/m). Figure 14 shows a depth of

penetration curve of intensity versus depth in units
of standard depth of penetration.
37
The skin effect, the concentration of eddy
currents in the test object material nearest the test
coil, becomes more evident as the frequency,
13 conductivity, or magnetic permeability is increased.
5 Because the relative magnetic permeability is 1.0 for
1 2 3 nonmagnetic materials and the conductivity of a
particular material is typically known, the depth of
Depth of penetration penetration is controlled by the operating frequency.

8.10 AEROSPACE NONDESTRUCTIVE TESTING

 Table 2. Effect of test variables on surface intensity and depth of penetration.

Factor Effect on Surface Intensity Effect on Depth of Penetration

Increase Decrease Increase Decrease

increase √ √
Conductivity decrease √ √
increase √ √
Magnetic permeability decrease √ √
increase √ √
Frequency decrease √ √
increase √
√
Magnetic coupling decrease
√ √
increase
√ √
Coil current decrease
√ √
increase
Temperature √ √
decrease √ √

The depth of penetration value is important for eddy

current testing because it can be easily adjusted for a Figure 15. Eddy current standard depth of penetration for materials of
given test. Table 2 shows the effect of various eddy various conductivity at different frequencies (ASNT 2004, 481).
current factors on the surface intensity and depth of 100 (4.0)
penetration. Figure 15 shows the eddy current Titanium
Standard depth of penetration, mm (in.)

standard depth of penetration for various alloy

conductivity materials at different frequencies. It is Unified Numbering System
10 (0.4) S30400 austenitic chromium
important to note that the depth of penetration for nickel stainless steel
an eddy current test is also affected by the coil
dimensions, especially at low frequencies.

1 (0.04) Unified Numbering

Geometry System A97075 wrought
Geometric parameters — including liftoff, coil aluminum alloy
dimensions, conductive layer thickness, object size, 0.1 (0.004)
and object shape — affect how well the coil‘s energy
can be coupled into the material being tested and High alloy steel
can limit the quality and reliability of an eddy
current inspection. Geometric variations can create
0.1 1 10 100 1000
signals that are several times greater than the signals
Frequency (kHz)
of interest from discontinuities. However, special
techniques or probe designs can limit the effect of
geometry as a variable during the test. to the edges of test objects can be reduced by
Eddy current testing is normally defined as a incorporating a magnetic shield around the test coil
near-surface method. The distribution of eddy or by reducing the test coil diameter. Edge effect is a
currents in a test object varies exponentially with term most applicable to the inspection of sheets or
depth from the surface. The current density in the plates with a probe coil. Edge effect is also evident in
test object is most dense near the test coil. The depth testing lap seams close to installed fastener holes
of penetration effect is discussed in the previous and blade slot edges on rotors.
section.

Edge Effects
The electromagnetic field produced by a test coil
extends in all directions from the coil. The coil‘s field
precedes the coil by some distance determined by the
coil parameters (operating frequency and test coil
characteristics). As the coil approaches the edge of a
test object, eddy currents become distorted by the
edge signal. This is known as edge effect. Response

ELECTROMAGNETIC TESTING 8.11

 Liftoff edge effect occurs because part of the magnetic field
is outside the test object.
Electromagnetic coupling between the test coil
and test object is very important in an eddy current
test. The distance between the test coil and test Conventional Coil Design
object is called liftoff; the test coil response is very The eddy current coil can come in a variety of
sensitive to this spacing because the fields are designs. For aerospace applications, the most
strongest near the coil. One way to minimize the common coil is a small wound coil around a ferrite
liftoff effect is in the design of the probe that houses core and is used for high frequency eddy current
the test coil: for instance, the test coil can be spring surface inspection. Depending on its size, the coil
loaded to maintain contact on a flat surface. Another can be used along edges and in corners, around
way to minimize liftoff is with a reflection style coil fasteners or even in bolt holes. Coils are often built
fabricated to minimize direct coupling between the with a ferromagnetic shield around them, which
driver and the receiver. restricts the field and lets the coil be used closer to
Figure 16 shows the impedance plane response an edge, corner, or fastener. Encircling coils and
when liftoff increases between the probe and the test internal coils can be used to inspect tubing.
object surface. As liftoff increases, the loci of points
approach the inductance for air. The upper portion of
the impedance plane is the magnetic domain where Absolute Single Coils
responses occur from ferromagnetic materials. The The absolute single-coil configuration uses a
lower portion is the domain where responses are single coil to induce the eddy current and detect
obtained from nonmagnetic materials. Note the changes due to material or geometry. This
nonlinear (logarithmic) changes among the liftoff configuration is considered an absolute coil
loci for equal increments of spacing. When the arrangement in that it tests the specimen without
spacing is moved near the edge of the test object, an any comparison to either another portion of the test
specimen or to a known reference specimen.
Absolute coils are wound with sizes that depend on
Figure 16. Impedance plane display loci of points as liftoff value the frequency needed for depth of penetration and
changes: (a) liftoff loci; (b) edge effect loci (ASNT 2004, 324). physical sizes. High frequency coils for surface crack
detection can be small.
(a) Unified Numbering System
G43400 steel alloy
Liftoff Split Core Coils
0.13 mm (0.005 in.) Split core coils use two separate receiving coils
wound in opposite directions on a split cylinder of
Inductance X (relative scale)

0.25 mm (0.010 in.) ferrite whose cross section resembles two D shapes,
one D being the mirror image of the other. The two
0.38 mm (0.015 in.)
0.51 mm (0.020 in.)
Ds are back-to-back so that their cross section looks
Air Magnetic
1.02 mm (0.040 in.)
like a bifurcated circle. These D shaped receiver coils
Nonmagnetic are typically bound together with a drive coil wound
0.76 mm (0.030 in.)
0.51 mm (0.020 in.)
around them both. A differential coil is used to
exclude environmental noise such as temperature
0.25 mm (0.010 in.) variations and bulk material properties. A
Liftoff differential coil can also cancel surface, edge, and
Unified Numbering System interface signals between layers in open hole
A97075 wrought aluminum alloy inspections. In this configuration, the probe is
classified as a differential reflection probe. A split
Resistance R (ratio) core coil pair without the drive coil may also be used
in a differential bridge configuration.
(b) To illustrate effects of some probe parameters
Unified Numbering System
G4340 steel alloy such as coil orientation, coil size, and coil tilt,
images can be collected by indexing a probe across a
Liftoff A small cross section (for example, 150 μm, or 0.006
in.) of wire as shown in Figures 17. The footprint
Inductance X (relative scale)

assessment can also provide insight into a probe‘s

working sensitivity. Results have shown a correlation
between a reduced probe footprint and a reduction
Air Magnetic in the probe‘s response to larger fatigue cracks.
Nonmagnetic Many standard techniques for aircraft engines use a
split D shaped coil with the coil axis running normal
to the test object as shown in Figure 18.
Liftoff Split core probes are generally specified
according to coil diameter. Small D shaped coilswith
0.51 mm (0.020 in.) diameter cores have been used.
Unified Numbering System
A97075 wrought aluminum alloy Coils with 1.27 mm (0.050 in.) and 1.52 mm
(0.060 in.) diameter cores are some of the larger
D shaped coils used for the detection of relatively
Resistance R (ratio) small discontinuities. The smaller diameter coils

8.12 AEROSPACE NONDESTRUCTIVE TESTING

provide a greater ratio of signal-to-noise but require frequency no longer depends on the coil diameter,
a greater number of indexes to cover the same and the coil can be larger, so that the field of view is
amount of test area. as wide as the contact area between the coil and the
test surface. Because part noise is integrated over the
contact area of the coil, the practical size of the wide
Wide Field Probes field probe is determined by the required
The term wide field probes typically refers to discontinuity detection size and the anticipated noise
probes with circumferentially wound coils whose of typical test surfaces.
axes are parallel to the test surface to achieve a Wide field probes can be further categorized as
wider field of view. Two receiving coils are abutted bobbin coils, shaped coils, and flex coils. Bobbin
(so that a circular end of one cylindrical coil is next coils generally refer to coils whose circumferential
to the end of the other) such that their currents flow windings are circular or nearly circular (Figure 18b).
in opposing directions. These two receiving coils are Bobbin coils have cores of air, ferrite, annealed steel,
often supplemented with a separate drive coil in the or other materials. Shaped coils are a derivative from
same probe. When the probe is applied with the axis the bobbin coils in that the coils are shaped to fit a
parallel to the test surface, the discontinuity distinct geometry.

Figure 17. Axial alignment of probe having split D shaped coil: (a) tilt perpendicular to split; (b) acceptable alignment;
(c) tilt in split axis (ASNT 2004, 489).

(a) (b) (c)

Index position, mm (10–3 in.)

Index position, mm (10–3 in.)

Index position, mm (10–3 in.)

1.9(75) 1.9(75) 1.9(75)

1.3(50) 1.3(50) 1.3(50)

0.6(25) 0.6(25) 0.6(25)

0 2.5 5.1 0 2.5 5.1 0 2.5 5.1

(100) (200) (100) (200) (100) (200)
Scan position, mm (in.–3) Scan position, mm (in.–3) Scan position, mm (in.–3)

480 480
Amplitude (mV)

Amplitude (mV)

Amplitude (mV)

160
320 320

160 80 160
0 0 0
0 0 0 0.6 (25)
5.1 (200) 0.6 (25) 5.1 (200) 0.6 (25) 5.1 (200)
2.5 (100) 1.3 (50) 2.5 (100) 1.3 (50) 2.5 (100) 1.3 (50)
0 0 0
1.9 (75) 1.9 (75) 1.9 (75)

Scan position, Index position, Scan position, Index position, Scan position, Index position,
mm (10–3 in.) mm (10–3 in.) mm (10–3 in.) mm (10–3 in.) mm (10–3 in.) mm (10–3 in.)

Figure 18. Eddy current probes: (a) split, D shaped reflection coil; (b) wide field bobbin coils; (c) wide field flex coil
(ASNT 2004, 490).

(a) (b) (c)

ELECTROMAGNETIC TESTING 8.13

 Flex coils differ from bobbin and shaped coils in
that each coil is housed in a flexible protective jacket
Multiple-Frequency Testing
that lets the coil be pressed into a geometry to Single-frequency instruments apply one selected
conform to the surface as shown in Figure 18c. The frequency to the probe. This frequency may be
main advantage of the wide field probe is the permanently set, as in a 60 kHz conductivity meter
increased throughput provided by a single-pass test. with a dedicated coil to discriminate liftoff and
However, the increased throughput comes at a cost. conductivity. Single-frequency instruments
There are several disadvantages for the wide field intended for general purposes allow selection of the
coils. First, signal response greatly diminishes at a frequency to fit the application and can drive a
nonoptimal coil orientation. However, because the wide range of probes. Multiple-frequency
required discontinuity orientation is typically ideal instruments are normally set up to generate from
for a wide field coil, orientation sensitivity is two to 32 different frequencies selectable by the
generally not an issue. Second, rolloff is typically operator. Multiple-frequency testing is
greater where discontinuity signal amplitude exceeds accomplished through three different techniques:
the threshold of rejectable indications. Although this (1) time domain multiplexing, (2) frequency domain
rolloff is of greater concern than rolloff of the multiplexing, and (3) pulsed frequency testing.
D shaped reflection probe, the amplitudes of larger Multiplexing refers to sequentially interleaving
discontinuity sizes are generally much greater than multiple frequencies over a single channel to the test
the threshold required for the test. Although coil. Instruments using time domain multiplexing
discontinuities beyond the rolloff range will be and frequency domain multiplexing generate and
reported at a reduced size, they will still exceed the switch among discrete frequencies. Each technique
rejection threshold and therefore will need to be has its advantages and disadvantages. Some
addressed before part acceptance. Third, a wide field instruments can multiplex in the frequency domain,
cannot identify the circumferential location of an in the time domain, or in both concurrently. With
indication. Circumferential location can be useful in time domain multiplexing, multiple frequencies are
locating a discontinuity, but the axial location of the generated sequentially as shown in Figure 19.
discontinuity is usually enough. If knowing the In digital systems, a number of frequency time
circumferential position of a discontinuity is critical, slots are selected. Analog systems usually have a
an overlapping test with a coil can help. fixed number of time slots. The coil is energized at
one frequency for a predetermined period of time.
The frequency generator is then switched to a
Orthogonal Field Probes different frequency in the second time slot for
The term crosswound probe applies to a pair of another specific period of time. This process
coils oriented at ninety degrees to each other continues until all selected frequencies have been
(orthogonal) and crossing one another on the side applied to the coil. The process then starts again with
oriented toward the test surface. These probes the first frequency. Timing of each specific frequency
produce a field pattern with four orthogonal lobes is critical to accurately process and display the
that intersect the test surface in a pattern resembling information from each individual time slot on the
a four-leaf clover. There are a variety of proprietary output device. With frequency domain multiplexing,
coil designs used to produce this orthogonal field multiple frequencies are applied to a coil at the same
pattern and improve the performance of the simple time, so this method of multiplexing is commonly
crosswound coil pair. The main feature of these called simultaneous injection. In this scenario, all of
probes is a signal minimized by liftoff from the test the selected frequencies are applied to the coil
surface. continuously. Figures 19 is a time domain view of
Orthogonal field probes may be operated in either time and frequency domain multiplexing,
reflection or differential bridge modes and share the respectively. The recurring pattern in Figure 20 is the
characteristic that rotation of the probe by ninety result of the simultaneous mixing of the four input
degrees produces a mirrored image of a discontinuity frequencies as opposed to the sequential mixing in
signal on an impedance plane display. This the time domain multiplexing of Figure 19.
characteristic field lets a carefully designed probe The advantages of time domain multiplexing are
cancel out effects such as liftoff and permeability that it permits maximum power to be applied to the
and conductivity variations that occur over an area probe and that it permits maximum use of the
larger than the probe. The rotatable probe is very dynamic range of an instrument at each discrete
sensitive to cracking indications without noise due frequency. Drive and gain stages can be optimized
to coatings, surface roughness, heat affected zones, for each time slot. Optimizing helps analyze a wide
or material variations. range of frequencies on the same probe, where the
Because orthogonal field probes subtract out response at certain frequencies may be much greater
variations in conductivity, some types may be used than at others. A large number of frequencies can be
for inspecting a wide variety of materials for surface used for a single test with instruments capable of
or near-surface cracking at one particular frequency. 32 time slots, each addressable to a different
These probes may also be used over a very wide frequency. The test frequency applied during an
frequency range. Although historically used as weld individual time slot in a well designed system is
test probes, these latter characteristics have led to monotonic and free of spurious signals. Although
the greater use of orthogonal field probes as general hardware is minimized, timing processes become
purpose crack detection probes where one particular critical and adequate settling time must be provided
probe can have a wide variety of applications. when switching time slots.
Multiple-frequency eddy current instruments can
be used to detect indications such as corrosion in

8.14 AEROSPACE NONDESTRUCTIVE TESTING

multilayer aircraft structures, to sort for improper
material presence in structures, and to minimize Figure 19. Time domain multiplexed waveform
probe wobble in confined spaces or on varying part (ASNT 2004, 176).
surfaces. Each condition to be suppressed must
produce significant impedance changes for one
frequency and less significant changes for other A B C D
frequencies. An example would be using a

Amplitude (relative scale)

dual-frequency inspection for subsurface corrosion
while compensating for air gap between the layers. A
low frequency would be selected that would allow
sufficient penetration to detect the corrosion. Air gap
responses would also be present from this frequency.
Using a higher frequency would respond to the air
gap but not have sufficient penetration to respond to
the corrosion. The second signal admixed should
have a phase distinctly different from the signal of
interest at each frequency. The analysis of these
signals can become extremely complex, and most
conventional multiple-frequency testing is limited to Time (relative scale)
dual-frequency testing. This technique is also used in Legend
tube testing to discriminate between outside surface A. 400 kHz.
discontinuities and structural signal responses from B. 200 kHz.
within the tubing. C. 100 kHz.
D. 20 kHz.

Pulsed Eddy Current Figure 20. Frequency domain multiplexed

Pulsed eddy current instruments apply a brief waveform, for simultaneous application of
pulse to a transducer at a periodic rate. The source is frequencies at 100, 200, and 400 Hz, as well as
pulsed to generate a broadband frequency whose 20 kHz (ASNT 2004, 177).
spectrum is a function of pulse width. A pulsed eddy
current instrument generates multiple short pulses at
a controlled rate. The frequency spectrum for the
series of short pulses is harmonic in the frequency
domain. The harmonics are an integer multiple of
Amplitude (relative scale)

the pulse rate, the magnitude of the spectrum being

determined by the pulse shape. Pulsed eddy current
instruments usually display a time based result and
allow gating of the signal at selectable time intervals.
The distortion of the eddy currents affects the pulse
shape. Time gate signals can be analyzed to
discriminate discontinuity types and sizes. The main
advantage of pulsed eddy current over conventional
single-frequency eddy current is its broadband
frequency.

Time (relative scale)

Eddy Current Sensor Arrays
An array of eddy current coils is one means of
increasing the area and speed of inspection. eddy current arrays offer advantages over single-
Increasing throughput comes at a cost because eddy sensor testing.
current arrays require more channels. Although Consideration must be given to the test area of
multiplexing can significantly reduce the costs of interest, probe configuration, inspection speed,
parallel processing of multiple signals, the costs will instrument signal processing ability, and
still be greater than those of a system with a single discontinuity orientation to the probe. Eddy current
probe. In addition to the increased equipment costs, arrays provide a convenient means for scanning an
probe costs increase with more sensors. Printed area. Surface and subsurface cracks can be
circuit eddy current coil arrays can reduce sensor visualized, which digitally records the test surface,
costs and provide flexible coil arrays. These printed helps data interpretation, and obviates raster
circuit coils have fewer numbers of turns or scanning with a single coil indexing finely enough
windings than conventional coils, which reduces the to catch short cracks. Eddy current array probes
nominal electrical signal and requires different come in a variety of shapes, sizes, and number of
impedance measurement instrumentation. In elements. Figure 21 shows an eddy current array
addition to the sensor production costs, there are probe with 32 pairs of driver and pickup coil pairs in
also costs to normalize coils and ensure their proper two rows. The coils are 4.4 mm (0.175 in.) in
sensitivity. If these costs and problems are addressed, diameter. The two rows are offset from each other by
half the diameter of the coils.

ELECTROMAGNETIC TESTING 8.15

 Indications can be analyzed conventionally on an
Figure 21. Photograph of eddy current array impedance plane by placing cursors over the defect
sensor: (a) top; (b) face with two rows of indication. An array inspection increases both the
driver/pick-up coil pairs. inspection speed and reliability.
(a)

Signal Processing
Signal processing techniques include parametric
modeling, pattern recognition, frequency
discrimination, digital filtering, and phase
discrimination. These techniques can greatly improve
the signal-to-noise response for detection of features
and for converting raw measurements into
meaningful electrical and geometric properties.
1. Parametric modeling involves adjusting slope,
amplitude, direct current offset, phase, or other
parameters to obtain a best fit to some
(b) predetermined model in order to significantly
reduce edge variations.
2. Imaging and pattern recognition can reduce
edge effects.
3. Frequency discrimination can be as simple as
setting filters to eliminate unwanted noise, but
filters cannot discriminate between an indication
and noise.
4. Digital filters supplement instrument filters and
improve the ratio of signal to noise.
5. Phase discrimination can be used to extract
discontinuity signals if there is enough phase
separation between the discontinuity signal and
the unwanted signals.
An aerospace application that makes efficient use Regardless of the signal processing in a given
of eddy current coil arrays is the detection of test, it is important that test parameters, such as
subsurface cracks at the edge of a doubler. With a probability of detection data, are equivalent to those
conventional single coil instrument, it is difficult to used to collect the reference data. Without
discriminate the presence of the subsurface doubler equivalent test parameters, it is difficult to ensure
edge from a crack. Slight variations in either that test thresholds and discontinuity sizing are
thickness or conductivity can cause signals that are accurate.
difficult to interpret. The inspection configuration is
shown in Figure 22. The inspection is performed
from the exterior side of the skin to detect cracks on
the far side, adjacent to the subsurface doubler. Remote Field Testing
Results of a scan are shown in Figure 23. The Remote field eddy current testing is popular
fastener indications are shown along with multiple because it can inspect regions not only near the
small cracks at the edge of the doubler. The doubler probe but also throughout the thickness of the
edge can be identified by a change in color. The material. Remote field eddy current testing operates
indication turns red if the discontinuity signal is at relatively low frequencies that penetrate the entire
more than 50 percent of the reference notch signal. layer. In the basic remote field probe, there is one

Figure 22. Aircraft doubler inspection.

A
Fastener
A Exterior skin

Crack location

Doubler
View A-A

8.16 AEROSPACE NONDESTRUCTIVE TESTING

exciter coil and one detector coil. The two coils are
separated by a distance characteristic of remote field Figure 23. Eddy current array image identifying cracks along edge of
testing. If the exciter and detector were placed close doubler.
together, the detector would measure only the field
generated by the exciter in its vicinity. In that case,
the setup would essentially be a standard eddy
current test setup in send and receive mode. For the
remote field test shown in Figure 24, the separation
is significant: the detector coil senses variation in Fastener
the remote field. The exciter coil is energized with
alternating current at frequencies from 50 to
1000 Hz for ferrous metals. Higher frequencies are
used for nonferrous materials.
The exciter coil typically carries a current of 0.1
to 1.0 A, the limitation being probe temperature and, Crack
for some probes, magnetic saturation of the probe
core. The detector signal depends directly on the
exciter current and frequency. Lowering the
frequency and increasing the exciter current yields
larger penetration depths. Test parameters (such as
frequency, drive voltage, and sample rate) are chosen
by taking into consideration factors such as probe
pull speed, discontinuity sensitivity requirements,
magnetic permeability, tube conductivity, and tube
wall thickness. The detector coil electronics must be
very sensitive and capable of high gain for sensing
changes in the remote field.

Figure 24. Electronic components of remote field test system (ASNT 2004, 209).

Detector
coil Phase and amplitude detector Computer

Reference signal

Exciter
Power amplifier Oscillator
coil

ELECTROMAGNETIC TESTING 8.17

 Nonconventional
PART3 Electromagnetic Methods

Meandering Winding permeability, proximity, or layer thickness). Also,

measurement procedures require less standardization
Magnetometer and calibration than with conventional methods.
This approach has been used in a variety of
Meandering winding magnetometry is a magnetic aerospace applications, including detection of hidden
field based eddy current sensing technique that uses and surface breaking cracks, imaging of corrosion,
novel windings and modeling to examine composite inspection, and quality assessment for
conducting or magnetic materials. The sensors and roller burnishing, friction stir welds, and coatings.
sensor arrays are fabricated using printed circuit
techniques to enable precise control of winding
dimensions, and the meandering winding Meandering Winding Magnetometer
magnetometers are designed to enable modeling Geometry
from basic physical principles. This modeling lets
multivariate inverse techniques determine absolute The meandering winding magnetometer is an
property values (electrical conductivity, magnetic inductive, eddy current based sensor that is designed
specifically for the nondestructive characterization
of material properties near the surface (Goldfine
Figure 25. Meandering winding magnetometer is inductive, eddy 1993). As shown in Figure 25, the original sensor
current based sensor designed for nondestructive characterization of design had a meandering primary winding for
material properties in near surface region: (a) basic sensor geometry; creating a spatially periodic magnetic field when
driven by an electrical current. Secondary windings
(b) typical depth of penetration plot of nonmagnetic low conductivity
(sense elements) are on opposite sides of the primary
(1.0 MS/m, or 1.72% IACS) material. for sensing the response. More recent designs of
(a)
meandering winding magnetometers and arrays have
Z Primary ID improved on the meandering drive construct
Y
winding  (Schlicker 2004; Schlicker 2006). Printed circuit
O microfabrication techniques produce identical
X sensors. Fabricating the windings on thin and
Material
under test VS2 flexible substrates produces conformablesensors.
H The simple layout for the windings permits operation
over a wide frequency range, typically from 1 kHz to
VS1
40 MHz.
Secondary When interrogating a conductive material, in an
windings aircraft engine disk slot or bolt hole for example, the
current in the drive produces a time varying
ID
magnetic field that induces eddy currents in the
(b) material under test. These induced eddy currents
3
100 within the metal follow the same path as the linear
2  = 16.7 mm
Effective penetration depth (10–3 in.)

7 drive segments. In other words, the eddy current

Effective penetration depth (m)

5 pattern induced in the material under test looks like

10–3
3
a reflected image of the drive winding geometry.
7  = 3.6 mm When a crack, corrosion damage, an inclusion,
5 2 surface roughness, or an internal geometric feature
3
alters the flow of these eddy currents, the inductive
For nonmagnetic material of 10 sensing coils sense an absolute magnetic field altered
2 conductivity 1 MS/m (1.72 7 near the feature. Absolute inductive sensing coils,
IACS) unlike differential ones, let models based on physical
5
10–4 principles analyze the data. For example, the goal
7 3 might be to measure the sensor proximity to the
5 2 surface (liftoff) at each sensing element and the
104 105 106 107 electrical conductivity of the material. Local
Frequency (Hz) variations in conductivity might then indicate
Legend
damage.
For eddy current sensors at high frequencies, the
H = magnetic field intensity X, Y, Z = dimensional axes
induced eddy currents are confined to a thin layer
ID = drive current  = wavelength (due to the skin effect) near the surface whereas at
V = secondary winding voltage
low frequencies this layer penetrates deeper into the

8.18 AEROSPACE NONDESTRUCTIVE TESTING

material under test where it is limited by sensor
geometry. For meandering winding magnetometers Figure 26. Examples of meandering winding magnetometer array
and arrays, the depth of penetration of the magnetic sensor constructs: (a) linear arrays for surface mounted sensor
field into the material under test at lower frequencies applications; (b) scanning array with two rows of sensing elements.
is limited to a fraction of the drive winding spatial
(a)
wavelength  (Figure 25a). Thus, for the same drive
current frequency, the magnetic fields from a larger
spatial wavelength sensor will penetrate farther into
the material under test than a shorter spatial
wavelength sensor. This depth is shown in the plot of
Figure 25b for wavelengths of 3.6 mm and 16.7 mm
for relatively low frequencies, such as under 1 MHz
for titanium or nickel alloys. Over 10 MHz, the
spatial wavelength does not significantly affect the
depth of penetration of the fields.
The layout and connections to the secondary
windings can be adapted to different measurement
applications to improve sensitivity to the material
properties of interest. In Figure 25a, the secondary (b) Scan path width
windings are connected together to create a single Sensing 24 mm (0.947in.)
elements
sense element over a relatively large area. This type
of sensor is well suited to applications where the
average material condition is of interest, such as
coating characterization. As shown in Figure 26a,
connections can be made to individual sense 
elements to produce a linear array of sense elements.
This type of array is suitable for surface mounted
applications where multiple locations are monitored
simultaneously, such as fatigue damage monitoring
to assess crack growth rates. These arrays can be Drive
mounted in difficult-to-access locations with the
leads brought out to a more convenient location to
facilitate eddy current measurements without
disassembly of the surrounding structure.
Alternatively, Figure 26b shows the secondary
windings subdivided into smaller elements in the Figure 27. Three conductivity/liftoff measurement grids shown on same
X direction. This type of array is useful for creating plot: (a) real and imaginary part of transinductance; (b) magnitude and
property images when scanning the array in the phase of transinductance.
Y direction over regions with a complex geometry.
Imaginary part of transductance (nH)

During scanning, data are taken at each sensing (a) 2

Air point
element as it traverses a part to produce an image of 20 MHz
each unknown property of interest. These images 0 S B A
reveal spatial variations of measured properties and H
material conditions. In each format, the –2 H
O B O
transimpedance (sensing element voltage divided by Liftoff B
–4 S
drive current) is measured independently for each
H
sensing element.
–6 S 0.2 MHz
The meandering winding magnetometer was 2 MHz
O
designed to be accurately modeled with rapid Conductivity
computations and to require dramatically reduced –8
0 2 4 6 8 10 12 14 16
calibration and standardization. In some situations,
before measuring a test object‘s absolute electrical Real part of transductance (nH)
conductivity, the instrument can be calibrated
against air without reference standards. For coating (b) 20
Phase of transductance (degrees)

applications, and when the material being examined Air point

can be represented with one or more material layers, 0 A
coated reference standards are not required but may H
B O
verify instrument operation. This is described, for –20 Liftoff
B H S
example, in ASTM E 2338 (ASTM 2006).
–40 O 0.2 MHz

Measurement Grid Methods –60

S
2 MHz
The meandering winding magnetometer response Conductivity
is converted into material or geometric properties 20 MHz
–80 0 2 4 6 8 10 12 14 16
using databases of precomputed sensor responses.
These sensor responses are typically computed from Magnitude of transductance (nH)
models using knowledge about the sensor geometry
and properties of the material under test. In two

ELECTROMAGNETIC TESTING 8.19

 Table 3. Some meandering winding magnetometer applications.

Item Comment

Crack detection Instrument will detect cracks like conventional eddy current and can sense distributed microcracks from fatigue by
regional property change.

Engine Flexibility and conformability of meandering winding magnetometer arrays make them suited to examination of
components inservice engine components such as disk slots and blade dovetail regions.

Bolt holes Flexible and conformable eddy current arrays can be inserted into bolt holes to inspect for cracks and other
damage.

Hidden corrosion In many cases, corrosion is hidden beneath coatings or other material layers. The meandering winding
magnetometer approach and multivariate inverse methods can account for presence and properties of surface
layers to assess corrosion.

Coating Sensors have been used extensively for coating characterization. For insulating coatings, the liftoff measurement
characterization itself provides direct measure of coating thickness. For metallic or magnetic coatings, multiple excitation
frequencies are used to infer coating properties, such as thickness, electrical conductivity, and magnetic
permeability, as long as electromagnetic properties of coating are distinct from those of substrate.

Damage mapping Meandering winding magnetometer arrays can be mounted at material surfaces to monitor damage as it occurs or
can be scanned over surfaces to reveal damage. Periodical scans can show evolution of damage.

dimensions, the database can be visualized as a values using an inverse method (a rapid table
measurement grid. As shown in Figure 27, the lookup). In general, solving the inverse problem for
measurement grids map the real and imaginary such a nonlinear response is difficult and calls for
portions (or magnitude and phase) of the iterative techniques. However, these techniques are
transimpedance into unknown properties of interest. relatively slow. Instead, an inverse technique
Typically, the grids relate two measured parameters searches the database (a measurement grid) and then
to two unknowns, such as the electrical conductivity interpolates in grid cells so that property data can be
and liftoff or magnetic permeability and liftoff. imaged almost instantly. This image processing
Figure 27 shows the directions of increasing speed is especially important when there are tens of
electrical conductivity and liftoff. The precomputed thousands of data points.
databases clearly capture the sensor‘s nonlinear
response to variations in material and geometric
properties. For more than three unknowns, higher Measurement Systems
dimensional versions of these databases are used. A complete measurement system includes a
In practice, the response of each sense element is laptop computer for performing the multivariate
converted into material property and geometric inverse methods, impedance measurement
instrumentation, probe electronics, and the sensor or
sensor array. A layer of foam, a bladder, or a balloon
Figure 28. Eddy currents induced in conductive plate by magnetic field is typically placed behind the sensor to let the sensor
(not shown) from flat, electrically conductive foil (ASNT 2004, 163). conform to the surface geometry of the test object.
For automated systems, actuation is incorporated
Primary leads into the equipment.
Because the meandering winding magnetometer
array sensing elements are single-turn coils, to
obtain reasonable signal levels at low frequencies, it
Secondary leads is necessary to apply more current than typically for
conventional eddy current sensors. The applied
current can be 1 A in some cases. However, at the
high frequencies used for surface breaking
discontinuities in engine components (5 MHz to
Conducting foil 32 MHz), there is plenty of signal without requiring
such high drive currents, even for a single-turn coil.

Representative Applications
The meandering winding magnetometer has been
used in many aerospace applications. Table 3 lists
Induced eddy currents some examples.
in conducting plate

8.20 AEROSPACE NONDESTRUCTIVE TESTING

Magneto-Optic Imagers Figure 29. Arrangements for production of magneto-optic images:
Magneto-optic imaging is a real time eddy (a) reflection optical arrangement; (b, c) sensor magnetization and
current imaging technology that relies on the resultant image at two different times (ASNT 2004, 161).
magneto-optic effect. This technology has been used
to image cracks and other discontinuities in (a) Light source
Viewer
electrical conductors such as aging aluminum
airframes. Polarizer

Analyzer
Sheet Eddy Current Generation
The magneto-optic imager uses sheet current Circular coil
Magneto-optic sensor
induction (Figure 28). Lenz‘s law ensures that the
direction of the eddy currents is opposed to the
change in the direction of the currents in the foil Sample
Crack Reflector
(solid and dotted lines represent currents roughly
180 degrees out of phase). As in the case of the coil
of Figure 2, eddy currents diminish as the depth (b) Magnetization Image
increases. Note that the eddy currents under the
footprint of the foil tend to be very uniform.
Moreover, these eddy currents have return paths
outside the footprint of the foil, unlike a coil where
+ +
the return paths are in a circular region mostly
within the coil‘s footprint. This technique of
inducing eddy currents is crucial for the operation of
magneto-optic eddy current imaging devices.

(c) Magnetization
Magneto-Optic Displays Image

Figure 29 illustrates a reflection optical

arrangement for producing images of the magnetic
fields H (static or dynamic) associated with
discontinuities such as cracks. The magnetization M
of the magneto-optic sensor with polarized light and
an analyzer (another polarizer) produces these
images. It happens that the magneto-optic sensors
have only two states of magnetization (that is, +M or
–M) as shown by the arrows representing M in
Figure 29b and 29c. When viewed in polarized light,
one of these two states of magnetization (+M in this Figure 30. Magneto-optic sensor must be modulated by bias magnetic
case) can be made to appear dark by rotating the field to provide magneto-optic image of discontinuity (open hole) by
analyzer; –M will then appear light by comparison. using linear sheet eddy current induction in reflection optical
Hence, a magneto-optic sensor is essentially a
arrangement (ASNT 2004, 162).
two-state (dark and light) imaging device. Any
magnetic field H, including a magnetic field Magnetic bias field
associated with eddy current distortions caused by (from bias coil)
perpendicular flux
density B mT (G)

discontinuities, can with such an optical 0.3 (3)

Nominal

arrangement produce magneto-optic images that 0

closely resemble the actual discontinuities. Although –0.3 (–3)
the present discussion emphasizes nonmagnetic
electrical conductors like aluminum, it should be A A
clear from Figure 29 that magneto-optic displays
Magneto-opticsensorandimage
could, in principle, be used to form images of surface
breaking cracks in magnetic materials such as steel.
In this case, either static or slowly varying magnetic
fields magnetize the material and the magneto-optic Back reflecting surface
displays the magnetic flux leakage from the crack.
The magneto-optic stack of imager layers is shown Sheet currents in copper
in Figure 30. induction foil

Magneto-Optic Indications A A
Figure 31 illustrates idealized and actual
magneto-optic eddy current images for two different Work piece
techniques of eddy current excitation. Magneto-optic Induced eddy
currents Discontinuity
eddy current images are always accompanied by a
background of serpentine magnetic domains that
form the images. This background cannot be

ELECTROMAGNETIC TESTING 8.21

 completely eliminated, but adjusting the bias
Figure 31. Magneto-optic eddy current images magnetic field can minimize background domains.
of rivet with crack extending to right: (a) linear The eddy current frequency in the images of
excitation mode; (b) rotating excitation mode Figure 31 is 50 kHz. Figure 32 shows the images of
(ASNT 2004, 163). discontinuities on a standard using a rotating field.
In this case, the sheet current is made to change
directions across the area of interest in a rotating
(a)
fashion so that current will intersect the crack from
all directions. The standard had two layers of 1 mm
(0.04 in.) thick aluminum (Unified Numbering
System A82024, temper 3) with four rivets and a
second layer hole. Rivet A did not contain an
anomaly. Rivet B contained a first layer electric
Idealized image discharge machined (EDM) notch of 45 degrees and
2.5 mm (0.10 in.) length. Rivet C contained a
Actual image horizontal electric discharge machined notch 1.8 mm
(0.07 in.) long. Rivet D contained a second layer
(b) electric discharge machined notch, 5.0 mm (0.20 in.)
long. Site E had a 9.55 mm (0.376 in.) diameter hole
in the second layer to mimic material loss by
corrosion. The excitation frequency was 100 kHz for
rivets A, B, and C, but 10 kHz for the subsurface
discontinuities of rivet D and site E.

Idealized image
Magneto-Optic Image Interpretation
Actual image Although the length of surface breaking cracks
can be estimated roughly from magneto-optic eddy
current images, the images are not good for
determining crack width. For example, the apparent
Figure 32. Magneto-optic eddy current images crack width in Figures 31 and 32 is considerably
made with rotating eddy current excitation: wider than the actual crack width because these are
(a) discontinuity free rivet A at 100 kHz; images of the magnetic fields near the crack that are
produced by the flow of eddy currents around the
(b) rivet B at 100 kHz; (c) rivet C at 100 kHz; crack. These eddy currents are significant only
(d) rivet D at 10 kHz; (e) site E at 10 kHz (ASNT within about one standard depth of penetration from
2004, 164). the crack as measured parallel to the surface of the
(a) (d) test object. A rough rule of thumb for surface
breaking cracks is that the apparent width of the
crack is roughly twice the standard depth of
penetration , so there is no way to estimate actual
crack width. The actual crack length is roughly given
by the apparent crack length, as seen on the image,
minus the standard depth of penetration . This
approximation works best at high frequencies and
low power where the rivet edge is discernable and
readily located.
Using the foregoing rule of thumb, the higher the
(b) (e) eddy current frequency, the smaller the standard
depth of penetration  and the narrower the
apparent crack width w = 2. The magneto-optic
eddy current images in Figure 31 were made at an
eddy current frequency of 50 kHz with a standard
depth of penetration  = 544 μm (0.021 in.) in a
wrought aluminum alloy (Unified Numbering System
A97075, temper 6, or 7075-T6) but images are
possible at frequencies up to 200 kHz with a
standard depth of penetration  = 305 μm (0.012 in.)
(c) in this same material. Hence, the apparent crack
width in aluminum at 200 kHz will be w = 533 μm
(0.021 in.) whereas at 50 kHz the apparent crack
width will be twice as large, w = 1009 μm (0.043 in.).
At still lower frequencies, the apparent crack width
is wider still. At 10 kHz in the wrought aluminum
alloy, for example, the standard depth of penetration
 = 1170 µm (0.046 in.) and the apparent crack width
w = 2.34 mm (0.092 in.).

8.22 AEROSPACE NONDESTRUCTIVE TESTING

 Figure 33. Experiment setup with conventional Figure 34. Magnetoresistive sensor array with sheet current excitation
eddy current testing. foil.
Giant magnetoresistive
sensor One-dimensional array of
Excitation coil magnetoresistive sensors
Sheet current

Sample

Specimen

To make magneto-optic eddy current images that magnetic field is sensed by the giant
resemble actual discontinuities as closely as possible, magnetoresistive sensor. Alternately, a pulse
the highest possible eddy current frequency that still excitation current can be applied to the induction
permits detection of the discontinuity should always foil, and an array of giant magnetoresistive sensors
be used. This is true for both surface and subsurface can be used to measure the transient magnetic field
discontinuities. Discontinuities more than one at different locations simultaneously. The output
standard depth of penetration below the surface of a signals from the giant magnetoresistive sensor array
material can sometimes be difficult to detect unless contain time and space information. The information
the eddy current magnitude is great enough. can be used to estimate the defect depth and location.
Accordingly, high power settings are invariably used Because the magnetoresistive sensors measure
to detect subsurface discontinuities or corrosion at magnetic field directly, they can work at lower
depths of two or more standard depths of frequencies and detect features at greater depth than
penetration from the surface of materials such as can conventional eddy current testing.
aging aluminum airframes. Typically, the highest The magnetoresistive sensor array can be
possible power level is selected, and the highest combined with a mobile automated scanner system
workable eddy current frequency is the one that (Figure 35). This system can be used to scan over a
achieves the best possible discontinuity resolution. wing structure to create an image as shown in
Some designs of magneto-optic eddy current Figure 36. The relative sensitivity of
imaging devices permit eddy current magnitude to
be much greater than normal. This improvement
permits discontinuities at three or four standard
depths of penetration below the test surface to be Figure 35. Photographs of magnetoresistive
detected, depending on the size and nature of the sensor array system combined with mobile
discontinuity. automated scanner (MAUS) system:
(a) equipment setup; (b) magnetoresistivesensor
head on aircraft wing.
Magnetoresistive Sensing (a)
In the twenty-first century, sensors that can
measure the field because of induced currents have
been developed. Of these, the giant magnetoresistive
probes appear to be the most promising (Arvin 1998;
Daughton 1994). These probes offer exceptional
levels of sensitivity, small size, and low cost. A
simple giant magnetoresistive configuration is
shown in Figure 33, where the excitation coil
encircles the sensor (Na 2005). The magnetic field
generated by the excitation coil induces currents in
the specimen. The induced currents are distorted if a
crack is encountered. The giant magnetoresistive (b)
sensor picks up perturbations in the fields associated
with the induced eddy currents.
Figure 34 shows an alternative approach to using
the giant magnetoresistive sensor where the
sinusoidal excitation current is applied to an
induction foil. An array of giant magnetoresistive
sensors can be used to pick up the normal
component of magnetic flux density generated by
the induced currents. In the absence of a
discontinuity, the magnetic flux tangential to the test
surface results in a null signal. Anomalies in the
specimen result in a normal component of the
magnetic flux density. The normal component of the

ELECTROMAGNETIC TESTING 8.23

 magnetoresistive sensors compared to eddy currents signal-to-noise ratios for subsurface discontinuity
are shown in Table 4, based on data from an Air detection than conventional eddy current probes for
Force program (Bossi 2008). The magnetoresistive cracks around fasteners.
sensors were found to provide greater

Figure 36. Magnetoresistive sensor scan of wing splice.

Left wing Notches in standard
Skin splice

Rear Front
spar spar

Magnetized fasteners Skin splice

Table 4. Signal-to-noise comparison of eddy current and magnetoresistive sensors for notches below
4.06 mm (0.16 in.) of aluminum.

Notch Size, mm (in.) Eddy Current Magnetoresistive Sensor

5.1 (0.20) 1.5 9.6

5.5 (0.22) 1.5 14.0

6.4 (0.25) 1.5 18.0

7.6 (0.30) 3.0 45.0

8.24 AEROSPACE NONDESTRUCTIVE TESTING

 Relevant Standards and
PART4
Specifications

Specifications and standards define the specifications exist in three basic areas: equipment,
requirements that goods or services must meet. processes, and personnel.
Standards for equipment include criteria that address ASTM International and other organizations
probes, artificial discontinuities, and test results. publish standards for test techniques. Some other
Reference standards are work pieces with well standards are for quality assurance procedures and
defined physical characteristics, which may contain are not specific to a test method or even to testing in
artificial discontinuities, for instrument general. The United States Department of Defense
standardization and calibration as well as test has replaced most military specifications and
procedure verification. standards with industry consensus specifications and
Specifications represent instructions specific to a standards. A source for nondestructive testing
contract or work place. The language of a standards is the Annual Book of ASTM Standards.
specification may be incorporated into a standard Table 5 lists some standards relevant to eddy
through a review process. Standards and current testing.

Table 5. Electromagnetic testing standards published by ASTM International and other organizations.

Organization Standards

ASTM International (formerly American ASTM B 244, Standard test Method for Measurement of Thickness of Anodic Coatings on
Society for Testing and Materials) Aluminumand of Other Nonconductive Coatings on Nonmagnetic Basis Metalswith Eddy
Current Instruments
ASTM E 376, Standard Practice for Measuring Coating Thicknesses by Magnetic-Fieldor
Eddy Current (Electromagnetic) Examination Methods
ASTM E 703, Standard Practice for Electromagnetic(Eddy Current) Sorting of Nonferrous
Metals
ASTM E 1004, Standard Test Method for Determining Electrical Conductivity Usingthe
Electromagnetic (Eddy Current) Method
ASTME 1316,Standard Terminology forNondestructiveExaminations:Section C,
Electromagnetic Testing
ASTM E 1629, Standard Practice for Determining the Impedance of Absolute Eddy Current
Probes
ASTM E 2338, Standard Practice for Characterization of Coatings Using Conformable Eddy
CurrentSensorswithoutCoating ReferenceStandards

American National Standards Institute ANSI B3.1, Rolling Element Bearings— AircraftEngine, Engine Gearbox, and Accessory
Applications – Eddy Current Inspection

American Society for Nondestructive ANSI/ASNT CP-189, ASNT Standard for Qualification and Certification of Nondestructive
Testing Testing Personnel (2011)
ANSI/ASNT Standard CP-105, ASNT Standard Topical Outlines for Qualification of
Nondestructive Testing Personnel (2011)
ASNT Recommended Practice No. SNT-TC-1A, Personnel Qualification and Certification in
Nondestructive Testing (2011)

Canadian General Standards Board CAN/CGSB 48.14-M86, Advanced Manual for Eddy Current Test Method (1986); R 1997
CAN/CGSB 48.9712, Qualification and Certification of Nondestructive Testing Personnel

Chinese National Standards Z 8005100, General Rules for Eddy Current Testing

European Association of Aerospace AECMA PREN 2002-20, Aerospace Series. Test Methods for Metallic Materials: Eddy Current
Industries Testing of Circular Cross-Section Tubes

European Committee for Standardization EN 12084, Non-Destructive Testing– Eddy Current Testing– General Principles and
Guidelines

continued on next pagELECTROMAGNETIC TESTING 8.25

 Table 5. Electromagnetic testing standards published by ASTM International and other organizations (continued).

Organization Standards

International Organization for ISO 9712, Nondestructive Testing – Qualification and Certification of Personnel
Standardization

Japanese Standards Association JIS Z 2314, Test Methods for Performance Characteristics of Eddy Current Testing
Instruments

SAE International (previously Society of SAE ARP 891A, Determination of Aluminum Alloy Tempersthrough Electrical Conductivity
Automotive Engineers) Measurement (Eddy Current) (R 1988)
SAE ARP 4402, Eddy Current Inspection of Open Fastener Holes in Aluminum Aircraft
Structure
SAE ARP 4787, Eddy Current Inspection of Circular Holes in Nonferrous Metallic Aircraft
Engine Hardware

United States Department of Defense MIL-P-85585, Probes, Eddy Current, Unshielded, Single Coil, Absolute
MIL-STD-1537B, Electrical Conductivity Test for Verification of Heat Treatmentof
Aluminum Alloys, Eddy Current Method

8.26 AEROSPACE NONDESTRUCTIVE TESTING

 Aerospace Applications of PART5
Eddy Current Testing

Advantages and Limitations Table 6 summarizes the principles, advantages,

limitations, and applications of the eddy current
of Electromagnetic Testing method.
Like other nondestructive methods, eddy current
Modern eddy current and other electromagnetic tests provide measurements of material properties
test techniques offer a low cost means for high and dimensions and detection of discontinuities.
speed, large scale testing of metallic materials. The
Eddy current testing provides a nearly instantaneous
method can use small probes for access to readout or image result of the process in test.
difficult-to-reach areas. Eddy current testing is one Consequently, the method can be used in high speed
of the most widely used methods in the aerospace production lines to test bar stock, tubes, sheets,
industry. The applications include the following: plates, welds, and other symmetric parts. These parts
1. Eddy current testing is used to measure the pass through test coils or are scanned by moving test
electrical conductivity of metals. Conductivity is probes.
measured in sorting (identifying or separating Small, portable eddy current instruments provide
materials by composition or structure); in a simple and rapid means for testing materials.
evaluation of overaging or heat damage to Manual quality testing by individual operators, as
aluminum structure; and in identifying and well as mechanized test systems, can sort mixed lots
controlling heat treatment conditions. of materials to track degradation of materials and
2. Eddy current testing measures metallic and equipment in service, and to verify process quality.
nonconducting coating thickness, where the Limitations of eddy current tests are a direct
coating and base material have significantly consequence of the specific nature of the test and of
different electrical or magnetic properties. the response of electrically conductive test materials
Nonconducting coating thickness is measured in to the externally applied, time varying magnetic
noncontact measurement of thickness of fields used to excite eddy current flow. In general,
metallic foils, sheets, plates, tube walls, and eddy current tests apply only to test materials with
machined parts with one-sided access. significant electrical conductivity, such as metals,
3. Eddy current testing detects material alloys, and composites with conducting layers or
discontinuities transverse to the induced eddy conducting reinforcing fibers. They can be used,
current field — discontinuities such as cracks, however, to measure thicknesses of nonconducting
seams, laps, score marks, plug cuts, drilled holes, layers on the surface of conductive metallic
and other holes and laminations at cut edges of materials by the liftoff effect, in which the coating
sheet or plate. Discontinuities detected include separates the test probe from the conductive material
fatigue cracks on surfaces and in bolt holes in by the thickness of the nonconductive coating or
aircraft structures; low frequency inspection of sheet material.
aircraft structures for the detection of subsurface Eddy current tests provide maximum test
cracks; and aircraft engine inspection. sensitivity for the surface and near-surface layers of
4. Eddy current testing is used to detect exfoliation the test material next to the source of excitation. In
corrosion, metal thinning, and metal spacing in some cases, it may be difficult or impossible to
aluminum aircraft structures. penetrate to the center of thick specimens because of
5. Eddy current testing detects alpha case skin effect and attenuation of the electromagnetic
hardening on titanium alloys and detection of field at a certain depth below the surface. Eddy
titanium in aluminum brazed titanium current tests tend to be insensitive to laminar
honeycomb. discontinuities parallel to the induced eddy currents.
6. Eddy current testing determines depths of case Eddy current tests do respond, however, to
hardening of steels and some ferrous alloys. discontinuities transverse to the flow of eddy
7. Eddy current testing helps timing or locating of currents within test materials, where these
the motions of hidden parts of mechanisms, discontinuities interrupt, lengthen, or distort the
counting metallic objects on conveyor lines, or current flow paths.
detecting metallic missiles in flight.
8. Eddy current testing precisely measures
dimensional symmetric machined or ground and
polished metallic parts, such as bearing and What Can Go Wrong
bearing races, small mechanism components and Eddy current testing is a primary test technique
others. for detecting cracks and corrosion of aircraft
structures and engine hardware during maintenance
overhaul. These inspections are performed in
accordance with procedures developed by
theELECTROMAGNETIC TESTING 8.27
 manufacturer. Such tests can detect fatigue cracks can be replaced to restore the craft to airworthiness.
resulting from cyclic loading during flight, takeoff, Fracture mechanics and nondestructive testing
or landing. Eddy current testing can also be used to programs have consistently shown that eddy current
detect stress corrosion cracks to indicate the extent tests are reliable in detecting cracks that would be
of corrosion damage or to identify aircraft missed by radiography or that require careful
components damaged by fire: damaged components cleaning or coating removal to be detected by liquid

Table 6. Eddy current method summary.

Aspect Comments

Physical Principle

Induced electric field lines (10 Hz to Eddy currents are sensitive to conductivity, permeability, and geometry. Single or
10 MHz) are perturbed by discontinuities multiple eddy current coils and sensors are used to induce fields and detect changes.
creating magnetic fields that affect the
inductivereactance ofthetesting probe.

Applications

Detects anomalies including surface and Choose eddy current for conductivematerial inspections for cracks, corrosion, damage,
subsurface cracks, corrosion, inclusions, coatings, or changes that are surface or near surface in nature.
and damage in materials.
Depth ofpenetration for detectiondependson frequency and coil size. Lower frequency
Measures thickness. and larger coilsareused for deeperpenetration.

Measures material electrical conductivity Resolution is affected by frequency, higher frequency withsmallercoilsincreases
for alloy sorting and heat damage effects. resolution.

Advantages

Very fast, low cost method for surface and Noncontact technology does not require couplant.
subsurface discontinuity detection.
Coil designs and magnetic sensors can be developed to address variations with the eddy
Coils can be designed for many current method.
configurations including tubing.

Limitations

Must be conductive material. Magnetic materials affect the sensitivity.

Depth of penetration for detection Liftoff from surface and surface roughness variations affect readings.
depends on frequency and conductivity.

Test Object Inspection Issue

Fiber reinforced polymer (FRP) composite Eddy current testing is not commonly applied to composites but can detect changes in
laminate compositecharacteristics whenmatrix or fibers are conductive.

Castings Cracks, laps, seams, hottears, shrinkage, coldshuts, near surfaceinclusions.

Forgings Cracks, inclusions.

Machined parts Cracks, inclusions.

Fastened structure Cracks, corrosion.

Welded joints Cracks, lack of fusion, porosity.

Coatings Thickness, degradation.

Engines, motors, and components Cracks, material sorting, heat treat, case depth.

Subsystems Cracks.

Inservice Fatigue cracks, corrosion, material thinning.

8.28 AEROSPACE NONDESTRUCTIVE TESTING

Table 7. What can go wrong with eddy current testing.

Issue Comment Options to Check or Consider

Operating on thin material Sorting material or testing Increase test frequency to decrease the depth of penetration.
thin skin material.

Testing low conductivity materials For crack detection on Increase thetest frequency to decrease thedepthofpenetration
titanium alloy, for example. andincrease surfacecracksensitivity.

Checking for heat damage Large or small area Spot probe (absolute) to map or size area; differential coil to
conductivity change. minimize signal response at boundaries; cross axis to test without
having torenull.

Thickermaterial ordoublelayer Looking for indications Lower the test frequency to increase the depth of penetration.
structure below the surface or into Coil size may need to be increased for the frequency of operation.
second layer. Consider magnetoresistive sensors.

Part geometry changes Contoured or convex Smaller diameter probe, pencil, right angle, conformal probe,
surface. special design with guide shoe.

Large surface area Where discontinuity size Large diameter probeor array probeconfiguration can ensure
may not be an issue or 100percent testcoverage.
where the complete surface
must be tested without
robotic scanner.

Thick paint layer or nonconductive Increased paint layers Use a larger probe coil.
coating over test surface create more liftoff from the
parent surface material.

Noise and discontinuity signal Noise and discontinuity Employ filtering to improve signal-to-noise ratio or reduce the
signal are not clearly effects of surface or geometric effects on the discontinuity signal.
defined.

Multiple-layer bolt hole inspection Discontinuity signals Use filter or differential coil.
masked by interface signal
response.

No signal movement after instrument System appears frozen. Check for correct mode of test (differential, absolute, reference).
is nulled

Loss of signal Signal goes off screen. Check the conductivity of the base material for the test frequency
being using.

Signal spike Balance or null signal is Check probe end for exposed coil. Check connector cable ends for
spiking or fuzzy. broken wire. Check that test mode is correct. Check for board
component failure.

Drift of balance signal After nulling the Rebalance or null in a discontinuity free section of the standard
instrument and probe, or test surface. If there is still signal movement, an internal
signal may drift when componentmay be goingtoward failure.
probe is on test object.

Poor mix signal Poor frequency selection Capture a clean signal of interest for mixing out. Select proper
may result in a residual frequency ratios. Rotate probe to minimize noise or resultant
signal larger than initial residual horizontal with indications going up.
signal.

Poor near-surface signal resolution Indications close to surface If approved, change frequency to improve phase response of
have no separation in discontinuity signals. Adjust the X and Y (gain or scale) of the
phase. instrument to improve separation.

Signal moves off screen when testing Field spread from test coil Shielded coil minimizes field spread and allows close testing at
near edge or material change as in causes edge effects. edgesor fasteners. Usedifferential coil.
fastener.

ELECTROMAGNETIC TESTING 8.29

 Table 7. What can go wrong with eddy current testing (continued).

Issue Comment Options to Check or Consider

Erratic signal from bolt hole Signal appears to be oblong Ensure that probe is in complete contact with bolt hole’s inside
or off screen. surface by filling in machined separation in probe head.
Separation can be filled with piece of folded paper. If probe is
adjustable, adjust it to inside diameter of bolt hole.

Signal moves off screen while Poor access ortest object Maintain constant probe to surface contact with guide, shoe,
scanning. geometryprevents spring loading, or mixing. Complete liftoff cannot be mixed out,
probe-to-part contact. but probewobblecan be.

Surface noise while scanning or Signal appears noisy or to If approved, lowertest frequency topenetrate farther into
testing have wide baseline. material. If surface is rough, then place nonconductive material
(tetrafluoroethylene tape or terephthalate film) between coil and
test surface to minimize noise.

penetrant or magnetic particle testing. In addition, Conductivity Reference Standards

these studies also reveal that very small cracks can
be detected by high frequency eddy current tests. As Nondestructive test reference standards should be
a consequence, aircraft structural and engine used and should have a measurable physical
inspections must use eddy current testing to ensure characteristic related to the critical properties of
product reliability. Fortunately, this demand for eddy interest. For standardization of eddy current
current testing has generated research and conductivity, the number of reference standards may
development by aircraft and jet engine depend on the test purpose and the accuracy
manufacturers, airlines, the federal government, and required. The conductivity range of the reference
eddy current equipment manufacturers. Development standards used for standardization must be within
has resulted in advances in instruments, probes, and the range of the instrument and must cover the
other aspects of the technology. conductivities to be measured. The reference
A number of factors that can influence eddy standards should have the same change in resistivity
current results. Table 7 lists issues that can go wrong with temperature as the test material. It is desirable
and what can be done to improve or correct these for the conductivity difference between the low end
difficulties. reference standard and the high end reference
standard to be between 10 and 25 IACS. When a
standardization curve is established for a general
purpose instrument, the number of reference
Conductivity Measurements standards must be enough to develop a smooth
Electrical conductivity is commonly measured to continuous curve over the range of interest.
identify or separate materials as well as evaluating Physical reference standards for conductivity may
heat treatment conditions, such as overaging or heat be obtained from the National Institute of Standards
damage to aluminum structures. Eddy current and Technology (NIST), vendors, or manufacturers.
conductivity meters usually differ with respect to For convenience of transportation and storage,
operating frequency, liftoff compensation, conductivity standards are usually kept relatively
temperature compensation, sensitivity, probe size, or small. Reference standards must have sufficient size
means of presenting test results. Most conductivity to prevent edge effects or thickness from having a
instruments can provide results in siemens per meter, bearing on conductivity readings. These
as a percentage of the International Annealed Copper requirements can be satisfied by requiring a length
Standard ( IACS), or as customized displays of and width to 25 mm (1 in.) greater than the probe
signal amplitude or phase angle. Whatever type of diameter and the thickness greater than 3.5 times the
instrument is used, reference standards must standard depth of penetration at the test instrument
standardize or verify operation of the instrument, frequency. Conductivity reference standards should
permitting quantitative and qualitative judgments on be flat, have a smooth surface, and be free of
all future test results. coatings. Standards, sometimes called secondary
Laboratory evaluations are an important step in reference standards, should be accurate within
establishing eddy current test procedures. A suitable ±0.3 MS/m (±0.5 IACS) of the nominal value. The
laboratory evaluation typically involves analysis of secondary reference standards are calibrated from
test object composition, heat treating times and primary standards maintained in an oil bath and at a
temperatures, and mechanical processing. Laboratory controlled temperature.
evaluations establish a conductivity value for an The reference standards are identified by the
alloy. material‘s conductivity, the reciprocal of its
Conductivity is also discussed above, under coil resistivity. When testing aluminum alloys, it is
techniques. necessary to maintain two or more aluminum alloy
reference standards. Many reference standards,
particularly those of aluminum alloys, are subject to

8.30 AEROSPACE NONDESTRUCTIVE TESTING

metallurgical changes if exposed to temperatures of
65 °C (150 °F) or greater. Surfaces of reference Figure 37. Effect of impurities on conductivity of copper
standards can also corrode if exposed to moisture or (ASNT 2004, 329).
other hostile environments. Rough handling can 60(103)
damage reference standards and cause errors in Cadmium
Zinc
conductivity readings. For these reasons, standards
Silver

Conductivity, MS/m ( IACS)

should be transported and stored in dry, clean, 55 (95) Nickel
protected areas not subject to excessive Selenium
temperatures. Aluminum
50 (86) Tin
Manganese
Alloying Elements Effect on
Conductivity
45 (78)
If a pure metal were tested in the annealed or Arsenic
unstrained condition, a single conductivity value Silicon
would be obtained, such as the IACS Phosphorus Iron Cobalt
40 (69)
value arbitrarily assigned to pure annealed 0.02 0.04 0.06 0.08 0.1
0
copper.
However, most structural metals are alloys consisting Impurity (percent)
of a base metal with several alloying elements added Legend
to obtain specific properties. The addition of alloying IACS = percentage of International Annealed Copper Standard
elements to a pure metal alters its conductivity.
Usually, the alloying element provides centers of
interference to electron flow. This reduces the
alloyed metal‘s conductivity below that of the pure Figure 38. Effect of additive on conductivity of aluminum
metal. (ASNT 2004, 330).
Although the conductivity of pure copper is
58.0 MS/m (100 IACS), care must be exercised 38(66)
Ni
when testing copper, because its conductivity is Cr Si
influenced by small additions of impurities or 37(64) Cu Zn
alloying elements. Conductivity is also affected by Ti Ag Fe
Mn
Conductivity, MS/m ( IACS)

the metal‘s temperature at the time and place of eddy

current testing. Touching the reference standard with 36(62)
the operator‘s finger at the point of the test area can
change its temperature and apparent conductivity. 35(60)
The specific resistivity of standard annealed copper V
changes by 0.4 per 1 K (∆T = 1 °C, or 1.8 °F) of Mg
temperature change in tests near room temperature, 34(59)
20 °C (68 °F). 0 0.1 0.2 0.3 0.4 0.5 0.6
The conductivity of a metal is decreased as Impurity (percent)
increasing amounts of alloying elements are added. Legend
Even small amounts of foreign atoms can greatly IACS = percentage of International Annealed Copper Standard
reduce conductivity. Some alloying elements have a Ag = silver
Cr = chromium
much greater effect on conductivity than others. Cu = copper
Atoms that differ the most from the base metal in Fe = iron
size and electron distribution cause the greatest Mg =manganese
decrease in conductivity. Lattice distortion by the Mn = magnesium
Ni = nickel
alloying atoms (and particles of different chemical Si = silicon
composition) inhibits the flow of electrons through Ti= titanium
the lattice. Because of variations in chemical V = vanadium
composition resulting from the tolerances in alloy Zn = zinc
additions, a conductivity range rather than a specific
conductivity value is reported for each alloy.
An aluminum alloy has the highest conductivity
Figure 37 shows the decrease in conductivity of
and lowest strength in its fully annealed temper.
copper caused by various impurities. Figure 38
After quenching from the solution heat treatment,
shows the influence of metallic additives on the
strength is increased and conductivity is decreased.
conductivity of aluminum.
Many aluminum alloys are unstable after solution
heat treatment, even if held at room temperature.
Conductivity of Aluminum Alloys During this time, a certain amount of atom
migration takes place to initiate the formation of
Conductivity measurement is often applied to
submicroscopic particles. This process, called natural
aluminum alloys because of their extensive use in
aging, increases the strength of the alloy and either
the aerospace industry and because of the wide
has no effect on conductivity or causes a slight
variation in the electrical conductivity and
decrease. Table 8 shows the ranges of electrical
mechanical properties between different alloys and
conductivity for aluminum allo
heat treatments. Specific conductivity ranges have
been established for each alloy and temper.
ELECTROMAGNETIC TESTING 8.31
 Table 8. Ranges of electrical conductivity for aluminum alloys (ASNT 2004, 332).

Alloy and Tempera Electrical Conductivity

Unified Numbering System Minimum Maximum

(UNS) Aluminum Association (AA) (% IACSb) (% IACSb)
MS/m MS/m

UNS A91100 AA 1100 33.1 (57.0) 36.0 (62.0)

UNS A92014, untempered AA 2014-0 28.1 (48.5) 29.9 (51.5)
UNS A92014, temper 3XX AA 2014-T3XX 18.3 (31.5) 20.3 (35.0)
UNS A92014, temper 4XX AA 2014-T4XX 18.3 (31.5) 20.3 (35.0)
UNS A92014, temper 6XX AA 2014-T6XX 21.5 (37.0) 24.1 (41.5)
UNS A92019, untempered AA 2219-0 24.9 (43.0) 26.7 (46.0)
UNS A92019, temper 3XX AA 2219-T3XX 15.7 (27.0) 18.0 (31.0)
UNS A92019, temper 62X AA 2219-T62X 18.0 (31.0) 20.6 (35.5)
UNS A92019, temper 8XX AA 2219-T8XX 18.0 (31.0) 20.6 (35.5)
UNS A92024, untempered AA 2024-0 26.4 (45.5) 29.0 (50.0)
UNS A92024, temper 3XX AA 2024-T3XX 16.2 (28.0) 19.1 (33.0)
UNS A92024, temper 4XX AA 2024-T4XX 16.5 (28.5) 18.9 (32.5)
UNS A92024, temper 6XX AA 2024-T6XX 20.3 (35.0) 23.8 (41.0)
UNS A92024, temper 8XX AA 2024-T8XX 20.9 (36.0) 24.7 (42.5)
UNS A93003 AA 3003 25.8 (44.5) 29.0 (50.0)
UNS A96061, untempered AA 6061-0 27.3 (47.0) 29.6 (51.0)
UNS A96061, temper 4XX AA 6061-T4XX 20.6 (35.5) 24.1 (41.5)
UNS A96061, temper 6XX AA 6061-T6XX 23.2 (40.0) 26.1 (45.0)
UNS A97075, untempered AA 7075-0 25.5 (44.0) 27.8 (48.0)
UNS A97075, temper 6XX AA 7075-T6XX 17.4 (30.0) 20.3 (35.0)
UNS A97075, temper 73X AA 7075-T73X 22.0 (38.0) 24.7 (42.5)
UNS A97075, temper 76X AA 7075-T76X 20.9 (36.0) 22.6 (39.0)
UNS A97178, untempered AA 7178-0 24.9 (43.0) 27.3 (47.0)
UNS A97178, temper 6XX AA 7178-T6XX 16.8 (29.0) 19.7 (34.0)
UNS A97178, temper 76 AA 7178-T76 20.3 (35.0) 22.6 (39.0)
a. Xs represent numerals unspecified in this list.
b. Percentage of International Annealed Copper Standard.

Heat Treatment and In annealing, the metal is heated enough to

remove the effects of cold working by
Mechanical Stresses recrystallization and grain growth. The annealing
process reduces obstacles to electron flow such as
The conductivity of a particular alloy is mobile dislocations and so improves the conductivity
influenced by heat treatment and by stresses of a metal. Increased annealing times favor more
introduced in the metal during fabrication. Within complete diffusion, greater coalescence, growth of
their permissible tolerance, the combined effects of larger grains, and associated increases in
alloying elements, heat treatment, and fabrication conductivity. The distribution and stresses
stresses establish a conductivity range instead of a established by the substitution of alloying atoms for
specific conductivity value for any metal alloy and those of the base metal reduce the conductivity of
heat treatment condition. Changing the number and the metal. The greater the number of solute atoms of
distribution of dislocations, alloying atoms, and a specific alloying material, the greater the reduction
particles of different compositions can alter the in conductivity.
properties of metals. These changes can be Heat damage from fires can affect the heat
accomplished through various types of heat treatment condition of a metal. For aluminum
treatment. The three principal types of heat aircraft involved in fires, eddy current testing is used
treatment are annealing, solution heat treatment, to determine whether the aluminum has been
and precipitation heat treatment along with artificial affected and needs replacement.
aging.

8.32 AEROSPACE NONDESTRUCTIVE TESTING

Factors Affecting Conductivity Tests penetration. With fixed frequency instruments,
not much can be done, except for stacking
Any variation from the condition for which eddy multiple materials that must be measured to
current conductivity instruments are standardized eliminate the influence of thickness variations
may have a significant effect on the accuracy of on the test results. However, if variable
their measurements. A standard measurement frequency instruments are available, then the
procedure will typically review the effects of these operating frequency can be altered to match the
interferences. Several test condition variables have effective depth of penetration value and
been studied to determine the effect on eliminate errors caused by thickness variations.
measurements of true base material conductivities: 6. A change in the temperature of the test object
(1) surface curvature, (2) edge effects, (3) material will change the electrical conductivity of its
thickness, (4) aluminum cladding thickness, (5) liftoff material. In metals, as the temperature is
or nonconductive coating thickness, and (6) increased, the conductivity is decreased.
temperature Nonmetallic composites have negative
1. Eddy current tests with rigid probes on surfaces temperature coefficients, so their conductivity
having small radii of curvature are avoided. A increases with temperature.
curved surface can change the effective probe Cladding with commercially pure, high
coil liftoff for a rigid probe along with the true conductivity aluminum on aluminum alloy base
material mass subjected to eddy current materials (which have lower electrical conductivities
influence. Eddy current laboratory evaluations than pure aluminum) causes a greater positive
for establishing conductivities should be made deviation in apparent conductivity as the cladding
on test objects with flat surfaces. For convex thickness is increased. In typical cases, the cladding
surfaces, a V block or modified probe body adds only about 5 percent to the total sheet thickness
holder maintains distance between the probe but results in an apparent conductivity change of 2.3
face and the curved surface. A thick flat to 4.1 MS/m (4 to 7 percent IACS). Eddy current
specimen can be machined to various radii or conductivity testing (at 60 kHz) for acceptance of
curvatures to make curved surface standards. aluminum clad material is usually restricted to
Alternatively, flexible sensor designs can material less than 2.3 mm (0.09 in.) in total thickness.
conform to the geometry of the surface and Testing of thicker materials requires removal of
overcome many limitations of surface curvature. most of the aluminum cladding layer by spot facing
2. Edge effects can be demonstrated by moving the a small area and taking a reading there. This reading
eddy current probe toward the edge of a test is taken on a small percentage of parts in the lot. A
sample and observing the change in the correction factor is determined between the spot
instrument display or readout. Test samples must faced areas and those where the cladding has not
be large enough to prevent edge interference of been removed. The remaining parts are checked
the probe‘s field and may be shielded to focus or without spot facing, and the correction factor is
limit the induced field. If parts having a narrow applied to the readings. After correction factors have
width must be tested, a centering jig or holder been determined, a table can be developed and used
should be used to maintain the probe on center. so that spot facing is no longer needed.
Edge distance curves are used to apply
correction factors to conductivity readings on
production parts.
3. Material thickness will affect eddy current test Coating Thickness
results when thickness less than the depth of
penetration. Techniques that automatically Measurements
measure the material thickness and electrical In the aerospace industry, numerous metal parts
conductivity at the same time can compensate are coated or plated to obtain special surface
for this effect. properties such as corrosion resistance, wear
4. The skin effect causes the currents to be resistance, thermal insulation, or improved
concentrated near the surface next to the appearance. These coatings are applied by various
excitation coil. This effect increases with means, including electrodeposition, hot dipping,
increasing test operating frequency, test object cladding, and spraying. For controlling the thickness
electrical conductivity, and test object magnetic of the deposit and for testing, reliability and rapid
permeability. The currents decrease measurements of coating are necessary. There are
exponentially or almost exponentially with many methods available including magnetic particle,
depth, depending on the test object shape and mechanical, optical, radiographic, ultrasonic, and
thickness. In addition, the phase angle of the electromagnetic testing. Each of these methods can
current lags increasingly as the depth increases. be applied depending on the damage mechanism,
The phase angle lag, measured in radians, uses discontinuity, and structure under test.
the phase angle of the current density at the There are four types of coatings and base
surface as a reference angle. materials that lend themselves to eddy current testing.
5. Whatever the test material, the standard depth The operating procedure used with the test
of eddy current penetration is a function of instrument is determined by the specific combination
material conductivity, operating frequency, and of materials. These material combinations are
magnetic permeability. Because all of these classified asfollows:
parameters are in the denominator of the depth 1. Metal foil and sheet or metal coating can be
of penetration equations, an increase in any of measured on a nonconductive base material
these values will result in a shallower depth of such as metal film on glass, ceramics, or

ELECTROMAGNETIC TESTING 8.33

 plastics. Thickness measurement of metal layers situation, in which the cladding is the poorer
is covered in the next section. conductor of the two metals, pertains to lead
2. Metal cladding with a higher conductivity than coatings on copper or to nickel coatings on
the base metal can be measured, such as copper, aluminum. The conductive coating thickness may be
zinc or cadmium on steel or pure aluminum on experimentally determined for a particular
aluminum alloy. combination of coating metal, eddy current probe,
3. Metal cladding with a lower conductivity than and test frequency by increasing the cladding
the base metal can be measured, such as nickel thickness in steps until a given increase in thickness
on aluminum. does not change the indicated meter reading
4. Nonconductive coatings can be measured on a (Figure 39).
metallic base material, such as anodic film or
paint on aluminum or other organic coatings on
metals. Nonconductive Coating Thickness
The measurement of nonconductive coating
thicknesses on metallic base materials is basically a
Conductive Coating Thickness liftoff measurement. The measurement is applicable
Cladding thickness is measured in two types of to nonconductive coatings such as anodic coatings,
situations where the conductivities of two metal and paint or plastic on a conductive base. Normally,
layers are very different. The first situation, in which magnetic attraction gages are used to measure
the cladding is a better conductor than the base nonmagnetic coatings (conductive or nonconductive)
material, pertains to copper, zinc, or cadmium on steel, and an eddy current test is used to measure
coatings on steel base materials. The second nonconductive coatings on nonmagnetic base
materials. The spacing from probe to test object is
referred to as liftoff.
Figure 39. Relative instrument readings for various thicknesses of Figure 40 shows the impedance plane response
copper plate on brass base (ASNT 2004, 460). when liftoff is increased at 100 kHz. The upper
portion of the impedance plane is the magnetic
domain with responses from ferromagnetic materials.
The lower portion is the domain where responses are
obtained from an aluminum alloy tested. Note the
nonlinear changes along the liftoff locus for equal
increments of spacing. Commercial manufacturers of
eddy current equipment produce instrumentation
Copper Copper specifically for nonconductive coating measurement.
These instruments provide greater accuracy in
measuring liftoff than do general purpose instruments.

Brass Brass Brass Brass

Eddy current standard

Layer Thickness Measurements
depth of penetration For some applications it is important to be able to
determine the thickness of a layer. Layers could be
metallic layers or components as well as spacing
Figure 40. Impedance graph showing liftoff effect at 100 kHz. Liftoff (air gaps) between solid layers. Many of these
is spacing of probe to part in units of 25 μm (0.001 in.) (ASNT inspections are for quality control during
2004, 461). manufacture.
Unified Numbering
System G43400 nickel
chrome molybdenum
Metal Thickness Measurements
alloy steel Tests for measuring metal thickness are generally
used on chemically milled sheet stock, thin walled
5 tubing, metal foil bonded to nonmetallic materials,
and parts that may be thinned from corrosion. For
Reactance X (relative scale)

thickness measurement, ultrasonic testing is more

10 common and often more accurate than eddy current
15 testing.
Air 20 Magnetic For eddy current thickness tests with
40 Nonmagnetic conventional coil designs, at least three specimens
30 with known metal thicknesses are typically needed
20 as reference standards to standardize the equipment.
One reference standard represents the minimum
10
acceptable thickness, the second represents the
maximum acceptable thickness, and the third is from
Unified Numbering the middle of the test range. In some cases, a bare
System A97075 wrought
aluminum alloy metal surface or specimen is necessary. All reference
standards must have the same conductivity,
Resistance R (relative scale)

8.34 AEROSPACE NONDESTRUCTIVE TESTING

permeability, substrate thickness, and geometry as
the test object. Figure 41. Eddy current impedance plane testing of bonded honeycomb
Some eddy current instrumentation can panel in wing tip (ASNT 2004, 475).
simultaneously determine the conductivity and Probe
thickness of the metal part, along with the liftoff.
These measurements are typically accomplished with Chemically milled step
multiple frequency measurements. The instrument
does not have the same requirements for the Scan
reference standards and may only require a single 1.6 mm 0.45 to 0.70 mm
standard having similar, but not necessarily the (0.06 in.) (0.018 to 0.028 in.)
same, conductivity as the test object material.
As an example, consider the localized thinning of
a chemically milled step in a bonded honeycomb
wing tip. A test must both detect and measure the
Honeycomb
thickness at the discrepant area of the Unified
Numbering System A97075, temper 6, wrought
aluminum alloy skin. The skin thickness ranges from
0.45 to 0.70 mm (0.018 to 0.028 in.) in the Figure 42. Eddy current impedance plane results from inspection of
chemically milled area. Reference standards 0.41, wing tip: (a) results from normal chemically milled step; (b) results
0.46 and 0.51 mm (0.016, 0.018 and 0.020 in.) thick from undercut chemically milled step (ASNT 2004, 476).
are used to calibrate the eddy current instrument at
20 kHz. The liftoff is rotated on the screen display by
a phase adjustment so that it is displayed in the (a) Scan
horizontal direction. The probes are placed at the
inboard or thin part of the bonded panel and
scanned outboard across the chemically milled step A B C
as shown in Figure 41. When the test object had a
normal chemically milled step, the screen display

Reactance X (relative scale)

trace shown in Figure 42a is obtained. When the part
0.46 mm (0.018 in.)
has local undercut in the chemically milled step,
then the trace shown in Figure 42b is obtained. A B
By marking the screen display with the minimum
0.46 mm (0.018 in.) thickness during standardization,
the depth of the undercut could be determined.
Numerous test objects were evaluated, and test
objects showing less than minimum thickness were C
rejected and sectioned and the eddy current results
were thereby verified. Resistance R (relative scale)

Metal Spacing (b) Scan

There are times when a gap separates two metal
sheets. The gap may be filled with a nonconductive
adhesive layer or a nonmetallic shim or may be A B C
purposely produced as a fixed dimension. If it is
Reactance X (relative scale)

desired to measure the gap or spacing, an operating

frequency must be chosen so that eddy currents will
be generated in the second (subsurface) layer. The 0.46 mm (0.018in.)
frequency chosen should produce eddy currents that A B
penetrate both metals to a depth of 3. When the
gap is zero (both metals touching), the combination
appears infinitely thick and produces a vector point
on the conductivity locus.
As the gap increases, the inductive reactance C
increases and the phasors produce a locus of points
(in the vertical direction) that cross the conductivity Resistance R (relative scale)
locus at some point. As the gap increases, the locus
of points will also intersect the thinning locus at a
point that represents the thickness of the upper metal
specimen as shown in Figure 43. At this point, the the problem with depth of penetration. If both
gap has reached a value where eddy currents are members are equally thick but one has a lower
being generated only in the upper layer. Hence, any conductivity, the probe should be placed on the low
further increase in the gap between the two metal conductivity specimen.
specimens will not cause a further change in As an example of a gap spacing inspection,
impedance. If a larger gap needs to be measured, the consider engine nose cowl inlet ducts. During
operating frequency must be reduced. fabrication of these ducts, problems have been
Gap or spacing measurements are best made with encountered in maintaining a gap between the outer
the probe located over the thin member; this reduces aluminum skin and inner titanium skin for the

ELECTROMAGNETIC TESTING 8.35

 purpose of hot air deicing. An eddy current fabricated from a plate of 1.5 mm (0.06 in.) thick
technique was developed and used to evaluate the Unified Numbering System A92219 wrought
gap at various locations around the nose cowl. aluminum alloy (Figure 44) to represent the outer
Results indicated that a consistent spacing was not skin of the cowl. The 0.5 mm (0.02 in.) thick
being maintained and that the gap spacing was commercially pure titanium shims represent the
below minimum tolerance in some areas of the duct. internal skin of the cowl. The nonconductive acrylic
Because of this, aluminum button spacers were step wedge represents the air gap or space between
installed between the aluminum and titanium skins the inner and outer skins. The instrument was
to make the gap consistent. Before taking gap standardized by placing the probe on the low
thickness measurements, a reference standard was conductivity (titanium) side of the standard at the
zero and 2.5 mm (0.10 in.) gap positions. The
frequency and instrument settings varied until a
Figure 43. Probe coil impedance curve showing effect of metal spacing maximum spread was obtained on the screen display.
between two flat, parallel, metallic conductive sheets (ASNT
2004, 462).
Corrosion Inspection
Low frequency eddy current techniques have been
Spacing used successfully for detecting hidden corrosion
Air
Reactance X (relative scale)

(Figure 45). The areas most prone to corrosion are

Metal spacing those that collect moisture from leakage or
(maximum gap) condensation. Once the corrosion starts, it becomes
intergranular and causes continuous exfoliation until
Conductivity Thinning locus
locus the corrosion propagates all the way through the
member. If the corrosion products collect between
Liftoff locus faying surfaces, a blister may form and cause
Metal spacing (zero gap) interface separation or spacing.
Conductivity locus Thinning or spacing effects can be detected and
(infinite thickness) measured by low frequency eddy current testing.
Low frequency eddy current techniques have
Resistance R (relative scale)
detected corrosion in jet aircraft. An operating
frequency of 4.5 kHz produced a standard depth of
penetration of 1.8 mm (0.07 in.), enough to
guarantee penetration into the second layer or
member. A shielded 9 mm (0.35 in.) diameter probe
Figure 44. Eddy current reference standard for spacing measurement: was designed to minimize edge effects. Reference
(a) setup; (b) from top; (c) from side (ASNT 2004, 475). standards to simulate spacing or thinning were
produced by end milling flat bottom holes in an
(a) Eddy current probe
Coil aluminum skin. A spacing of 0.5 mm (0.02 in.)
produced a full scale vertical movement of the flying
Titanium
Aluminum
Figure 45. Typical corrosion inspection
applications for low frequency eddy current
(b) 38 mm 19.0 mm (0.75 in.) testing: (a) interstitial corrosion; (b) thinning
(1.5 in.) minimum (ASNT 2004, 482).
(a)
50 mm
(2.0 in.) Probe

(c)
1.0 mm
(0.04 in.)
1.5 mm
(0.06 in.) Corrosion
0.5 mm 2.0 mm
(0.02 in.) (0.08 in.) (b)
0.02 mm 2.5 mm
(typical) (0.10 in.) Plastic step wedge
Probe

0 gap
1.5 mm (0.06 in.) Unified
Numbering System A92219
wrought aluminum alloy Corrosion

8.36 AEROSPACE NONDESTRUCTIVE TESTING

dot, and the 0.13 mm (0.005 in.) gap caused a An aluminum thickness taper gage (Figure 46a) of
proportionally smaller movement. Gaps greater than alloy similar to the aluminum skin material is used
0.5 mm (0.02 in.) were considered serious corrosion, to calibrate the eddy current instrument for metal
and gaps between 0.25 mm (0.010 in.) and 0.51 mm thinning. The probe is placed on the taper gage at
(0.020 in.) were considered suspect. the 1.3 mm (0.05 in.) thickness, and the response
Moisture entrapment between an aluminum shown in Figure 46b is obtained. The probe is then
fuselage belly skin and a titanium doubler causes slowly scanned to the thin end of the gage and is
corrosion thinning in the skin. To detect the lifted off at 0.2 mm (0.01 in.) thickness, yielding the
corrosion and determine metal thinning, the response shown in Figure 46c. Rescanning the taper
inspector can be given a sketch showing the location gage and lifting off the probe at various thickness
and outline of the internal doubler. The skin material increments (Figure 46d) provides the thickness
and temper can be identified along with the calibrations. The instrument is now calibrated for the
electrical conductivity. test. The operator looks for corrosion thinning while
the aluminum external skin is scanned over the area
of the internal doubler (Figure 47a). If there is no
corrosion, the response shown in Figure 47b will be
Figure 46. Screen display at 20 kHz for 1.3 mm obtained. Severe corrosion will yield a response
(0.05 in.) thick skins: (a) aluminum taper gage; similar to that shown in Figure 47c. The depth of the
(b) taper gage signal at 1.3 mm (0.05 in.) corrosion is determined by using the taper gage to
measure the depth from maximum amplitude
thickness; (c) taper gage signal at 0.2 mm
response as shown in Figure 47d. In this case, the
(0.01 in.) thickness; (d) completed thickness corrosion thinning is 0.63 mm (0.025 in.), 50 percent
standardization (ASNT 2004, 474). of the skin thickness. Cracks may be associated with
severe corrosion and, if they exist, may produce the
(a)
response shown in Figure 47e.
Material and temper 38 mm (1.5 in.)
identification
0.13 mm
(0.005 in.)
minimum
1.9 mm
(0.075 in.)
Crack Detection
Surface Crack Detection
178 mm
(7.0 in.)minimum
Eddy current testing is used to find cracks in
aircraft structures during inservice maintenance
checks. Impedance plane analysis provides a better
(b) Thickness, mm (in.) understanding of crack response loci with other
4 variables on the impedance plane. The value of this
approach is evident in Figure 48a where the crack
Reactance X (V)

3 response locus is 90 degrees out of phase with the

liftoff locus for Unified Numbering System
2
Air
G43400 nickel chrome molybdenum alloy steel.
1 Because of the wide separation angle between these
1.3 (0 .05) two loci, cracks in steel are fairly easy to detect. The
Liftoff
0 lower amplitudes and phase angles for electric
0 4 8 12 16 20 discharge machined notches in the steel sample are
Resistance R (V) also shown. Figure 48b shows that for aluminum
alloys, the electric discharge machined notch and
(c) Thickness, mm (in.) crack response lie between the liftoff and
4 conductivity loci. In this figure, the operating
Liftoff 0.2 (0.01)
frequency was 200 kHz. Here the smaller notch
Reactance X (V)

3
response almost parallels the liftoff response, but the
2 angle of separation and amplitude both increase as
the notch depth increases. A fairly good response is
1 Air obtained from a deep crack in a similar aluminum
1.3 (0 .05)
specimen. It can be seen in Figure 48b that the crack
0 response is almost parallel to the conductivity locus.
0 4 8 12 16 20 This characteristic is true for all nonmagnetic and
Resistance R (V)
magnetic materials.
(d) Thickness, mm (in.)
Skin cracks, which initiated at fastener holes in a
4 fuselage splice (Figure 49a), grew by fatigue in a
0.2 (0.01) circumferential direction. Flush head aluminum
Reactance X (V)

3 rivets joined the fuselage skin to the frames and

0.5 (0.02) longerons. Detection of the surface cracks was
2 possible with standard eddy current crack depth
0.8 (0.03)
1 1.0 (0.04) meters. These instruments required scanning between
Air
1.3 (0.05) each row of fasteners, which was time consuming.
0 To speed up the test, a procedure (operating at 5 kHz)
0 4 8 12 16 20 was developed for phase analysis screen display
Resistance R (V) instrumentation. With this type of equipment, the

ELECTROMAGNETIC TESTING 8.37

 Figure 47. Impedance plane displays for exfoliation corrosion metal thinning: (a) diagram; (b) display for
no corrosion; (c) display for severe corrosion; (d) method for determining depth of corrosion; (e) surface
crack response in corroded area (ASNT 2004, 474).

(a) Probe Probe

0.63 mm Corrosion 1.3 mm
(0.025 in.) Aluminum skin (0.05 in.)

Titanium doubler Crack

(b) (d)
0.63 mm

(arbitrary unit)
(arbitrary unit)

Reactance X
Reactance X

(0.025 in.)

1.3 mm
(0.05 in.) Air
Air 1.3 mm
Liftoff (0.05 in.)
Resistance R Resistance R
(arbitrary unit) (arbitrary unit)

(c) (e)

(arbitrary unit)
(arbitrary unit)

Reactance X
Corrosion
Reactance X

response

Air Air 1.3 mm

1.3 mm (0.05 in.)
(0.05 in.)
Resistance R Resistance R
(arbitrary unit) (arbitrary unit)

operator can scan in a direct circumferential line

Figure 48. Notches and cracks on impedance plane display: (a) electric over the installed fasteners (Figure 49a). Figure 49b
discharge machined notches and crack in steel; (b) electric discharge shows the response obtained by scanning over the
machined notches and crack in aluminum (ASNT 2004, 477). rivets with no cracks in the skin. In Figure 49c, the
crack response is in the same direction but has a
(a) much greater impedance, resulting in a clearly
Crack discernible increased amplitude signal.
Reactance X (relative scale)

Liftoff
Notch of 0.63 mm (0.025 in.)
Steel Subsurface Crack Detection
0.5 mm (0.02 in.)
0.25 mm (0.01 in.)
Subsurface crack detection can be performed in
aircraft structures using low frequency eddy current
Magnetic techniques. Figure 50 shows how cracks through the
Air thickness of second and third layer aluminum
Nonmagnetic
structures appear on the impedance plane at 1 kHz.
The surface crack response in the first layer runs
almost parallel to the liftoff locus. The second layer
Resistance R (relative scale) crack runs parallel to the conductivity locus, and the
third layer crack response is to the right of the
conductivity locus. The phase lag causes the phase
17.4 MS/m
(b) (30 percent shift of the crack signals as the crack depth below
Air Liftoff Deep the surface increases.
International
Reactance X (relative scale)

Magnetic crack
Annealed Figure 51 shows two example probe
Nonmagnetic Copper
Conductivity  configurations for inspection of subsurface cracks
Standard)
around fasteners. Probes are typically scanned over
Notch of 0.5 mm (0.02 in.)
0.35 mm (0.015 in.)
the surface, placed between the fasteners, or
0.15 mm (0.006 in.) positioned so that they encircle the fastener. As an
Liftoff example, low frequency eddy current equipment
30
23.2 MS/m operating at 500 Hz has been used to detect 1.0 mm
40 (40 percent (0.04 in.) long subsurface cracks in aluminum
International structure through aluminum layers 5 mm (0.2 in.)
Annealed
Copper thick. This test technique has been effective for
Resistance R (relative scale) Standard) detecting cracks in second layer structure and has
detected cracks missed by radiography. Although the

8.38 AEROSPACE NONDESTRUCTIVE TESTING

 Figure 49. Cracks in first layer initiating at aluminum fastener holes: (a) looking inboard, left side
shown, right side opposite; (b) scan over rivets with no crack in skin at 5 kHz using 9.6 mm (0.38 in.)
diameter probe; (c) scan over rivets with crack in skin (ASNT 2004, 478).

(a)
Forward

Typical scan
direction for Crack
all fasteners (typical)

(b) (c)
Crack
(relative scale)
Reactance X

(relative scale)
Reactance X
Rivets Rivets

Air Aluminum skin Aluminum skin

Air

Resistivity R (relative scale) Resistivity R (relative scale)

low frequency, 500 Hz eddy current technique is

effective, it has several limitations. First, it cannot be Figure 50. Direction of surface and subsurface crack indications in
used to test through outer layers thicker than 6.3 mm aluminum on impedance plane: (a) test setup at 1 kHz; (b) impedance
(0.25 in.) unless lower frequencies are used. plane with enlargement of circled area (ASNT 2004, 480).
Secondly, it requires multiple positions around the
circumference of a hole for crack detection unless an (a)
array is used. Coil
Larger probes prohibit testing between fasteners
separated by a distance less than the spot probe
diameter, or the wall thickness of a ring probe. On
an aluminum structure, 15 mm (0.6 in.) thick, 1 1.0 mm (0.04 in.) surface
containing steel fasteners, a 100 Hz encircling probe 2 1.25 mm (0.05 in.) subsurface
can be used. The alternating current coils produce an 3 3.2 mm (0.125 in.) subsurface
electromagnetic field that penetrates both the
aluminum and steel fastener. In addition, the steel Cracks
fastener acts as a core and tends to concentrate the
electromagnetic field. Cracks 13 mm (0.5 in.) long Air
(b)
were detected through outer layers of aluminum Nonmagnetic
Reactance X (relative scale)

8.9 mm (0.35 in.) thick, and longer cracks may be

detected through even thicker outer layers. The Conductivity 
encircling probe can be used where the fastener head Liftoff Liftoff
1
spacing is 6.3 mm (0.25 in.) or greater. A
disadvantage of the encircling probe is the absolute 23
measurement system. Changes in fastener length,
fastener permeability, aluminum thickness, edge
17.4 MS/m (30 IACS)
margin, nonconductive shims and conductivity may
cause signal variations that interfere with the test.
Crack detection sensitivity for laboratory Resistance R (relative scale)
conditions are shown in Figure 52. Crack detection
in the field is impaired by signal noise from
temperature variations, by separations between the
layers, and by variations in probe placement. Field
experience with a 500 Hz probe confirms that it can
reliably detect 13 mm (0.5 in.) long cracks under a
5 mm (0.2 in.) thick upper layer and 6.3 mm
(0.25 in.) long cracks under a 2.5 mm (0.10 in.) thick
upper layer. Hundreds of such test procedures have
been developed for crack detection in various
aircraft components and structures.

ELECTROMAGNETIC TESTING 8.39

 Figure 51. Typical applications for low frequency subsurface crack
Subsurface Cracks in Window Belt
detection: (a) second layer cracking; (b) cracking in fastener hole Splice
(ASNT 2004, 482). Cracks have been detected in the first row of
(a) rivets above the longitudinal belt splices of some
aircraft windows. The cracks initiated at fastener
holes in the internal (second layer) skin and grew in
Probe a longitudinal direction as shown in Figure 53. The
test may be performed by scanning between
fasteners with a 10 mm (0.4 in.) probe at 2 kHz
(Figure 54a). The crack response is also shown in
Figure 54. To speed up the test, a 15 mm (0.6 in.)
diameter probe was scanned along and across the
Crack aluminum fasteners at 1 kHz (Figure 54a). The crack
response is shown in Figure 55. Both test techniques
produce easily interpreted crack responses on a
(b) Encircling probe screendisplay.
Fatigue cracks caused by pressurization cycles
have been detected in the window belt panels on an
airplane with many hours in service (Figure 56). An
eddy current test was performed with a 13 mm
(0.5 in.) diameter probe at low frequency, from
500 Hz to 1 kHz. The reference standard, simulating
Crack the window belt splice joint panel, contained a
25 mm (1.0 in.) notch, electric discharge machined
where the thickness changed. A nonmetallic straight
Figure 52. Instrument response at 500 Hz related to crack length and edge was used to guide the center of the probe
upper layer thickness in typical aircraft structure (ASNT 2004, 482). directly over the center of the splice. For instruments
equipped with meters, the liftoff and phase angles
100 are set so that liftoff drives the meter up scale and so
Crack 6.3 mm Crack 13 mm (0.5in.)
(0.25 in.) that passing the probe over the electric discharge
80
machined notch drives the meter down scale. For
Crack 3.2 mm
(percent of full scale)
Instrument response

instruments equipped with impedance plane displays,

(0.125 in.)
60 the phase rotation is null balanced to produce a
crack response like that shown in Figure 54. When a
crack signal is obtained with low frequency eddy
40 current tests on the outside of the aircraft,
verification requires removal of the interior seats,
lining, and insulation to conduct a visual or high
20
frequency eddy current check of the internal surface
of the window belt panel.
0 0 1.25 2.5 3.8 5.1 6.3 7.6 8.9
(0.05) (0.10) (0.15) (0.2) (0.25) (0.3) (0.35)
Upper layer thickness, mm (in.) Bolt Hole Inspection
Small cracks in material next to fastener holes
may go undetected until the cracks grow to a size
Figure 53. Cracks in second layer initiating at aluminum fastener that allows detection at the surface not covered by
holes: (a) outboard side; (b) inboard side (ASNT 2004, 483). the fastener head. Unfortunately, in many highly
Up stressed areas, a crack of this size may continue to
(a) Forward Eddy current probe grow to failure at a rapid rate. For these critical
locations, the fastener may be removed and eddy
current testing performed in the open hole. This bolt
hole eddy current inspection is widely used
throughout the aircraft industry, both civilian and
military, for detection of fatigue cracks within holes
from which structural fasteners have been removed.
Test areafull length Automated bolt hole inspection provides reliable and
Window belt splice gap of windowbelt repeatable test results, detecting cracks as small as
0.13 mm (0.005 in.).
(b)

Manual Bolt Hole Probe

Figure 57 illustrates a manual eddy current bolt
hole probe. The probe coil axis is perpendicular to
Crack the material next to the hole. An adjustable collar
Window belt splice gap positions and locates the coil a desired distance
Crack inside the hole. Usually, testing is started with the

8.40 AEROSPACE NONDESTRUCTIVE TESTING

Figure 54. Detection of crack in second layer by scanning between fasteners with 10 mm (0.4 in.) probe at 2 kHz: (a) scanning
procedure; (b) no crack response; (c) crack response (ASNT 2004, 483).

(a) (b) (c)

Forward Up
Crack

Aluminum 1.3 mm
fasteners

(relative scale)
(0.05 in.)

Reactance X
(relative scale)
Reactance X
Scan 0.6 mm Air Aluminum Air
direction (0.025 in.) Aluminum

2.2 mm Liftoff Liftoff

(0.085 in.)

Crack (third layer) Resistance R (relative scale) Resistance R (relative scale)

Figure 55. Detection of crack in second layer Figure 57. Eddy current hole probe (ASNT 2004, 468).
by scanning over fasteners with 15 mm (0.6 in.)
probe at 1 kHz: (a) no crack response; (b) crack Coil assembly
response (ASNT 2004, 484).
Coil wires sealed in body probe with epoxy cement
(a)
(relative scale)
Reactance X

Liftoff Aluminum Probe body

Adjustable
collar Insulating sleeve
Set screw Coilwire soldered to microdot
Fasteners terminal
Seal connector and shield in
probe body with epoxy cement
Resistance R (relative scale) Microdot connector
Coil wire soldered
(b) Crack response Shield
(relative scale)

coil just below the test material surface, and the

Reactance X

Liftoff Aluminum probe is manually rotated 360 degrees. After each

rotation, the probe is advanced axially in increments
to ensure adequate overlap and coverage. The
spherical end of the probe is often split in the middle
so that a small rubber or plastic V wedge may be
Resistance R (relative scale) inserted to make the probe snug in oversized holes.
These wedges are rarely used for probes less than
6.3 mm (0.25 in.) in diameter. During bolt hole
testing, either an absolute or differential technique is
Figure 56. Window belt crack location: used. The absolute technique typically uses one coil
(a) area; (b) close view (ASNT 2004, 484). on a ferrite core; measurements are made while the
material is in direct contact with the probe coil. The
(a) Test areas differential technique uses two coils and compares
full length measurements obtained from the uncracked
A of window reference material under one coil with the test (and
belt
possibly cracked) material under the other coil. A
bridge circuit is unbalanced if one coil is over sound
A material and the second coil is over a discontinuity.
Figure 58 shows crack responses typical of bolt hole
probes inserted into an aluminum sample. A
single-loop response from an absolute bolt holes
(b) Section A-A probe is shown in Figure 58a, and a double-loop
Typical crack area response from a differential probe is shown in
top and bottom of Figure 58b.
window belt panel Skin

Sealant Reference Standards for Bolt Hole

Crack
Window Testing
belt
Reference standards aid in calibrating or
adjusting instrument controls and are used to ensure

ELECTROMAGNETIC TESTING 8.41

 that cracks will be detected with a predetermined reference standard works well if the notches are less
sensitivity, based on depth, length, and location. than 0.13 mm (0.005 in.) wide.
Figure 59 shows the locations where cracks may be
expected in bolt holes. Reference standards are
fabricated by simulating cracks with electric Automatic Bolt Hole Testing
discharge machined radial slots of various depths When numerous bolt holes need to be tested, it is
and lengths as shown in Figure 60. This type of often desirable to eliminate manual scanning and
use a device that automatically rotates the probe in
the hole. Eddy current equipment manufacturers
Figure 58. Normalized impedance graph showing crack responses for have developed such systems. For shallow holes and
absolute and differential bolt hole probes: (a) absolute probe crack tube ends, a system has been designed in which the
rotating probe is attached to the end of a short
response at 100 kHz; (b) differential probe crack response at 500 kHz
spindle.
(ASNT 2004, 468). Fully automatic eddy current discontinuity
Crack detection systems can detect and record
(a) discontinuities on conductive metal surfaces and in
structural fastener holes, reducing the errors of
Scan conventional hand scanning. The scanner‘s speed
can be varied from 0 to 0.25 Hz (0 to 150 rotations
Reactance X (relative scale)

per minute), providing simultaneous linear and

Coil rotational or spiral motion. Either filtered or
unfiltered signals from the scanner can be displayed
on a chart recording or oscilloscope. These systems
can detect cracks as small as 0.25 mm (0.010 in.)
long by 0.13 mm (0.005 in.) deep in aluminum. Scan
depths are adjustable from 6.3 to 3.8 mm (0.25 to
1.50 in.).
Manually operated rotating eddy current test
Air
Aluminum
systems are available for rapidly detecting small
cracks in aircraft bolt holes. They use small, light,
hand held scanners and various sizes of bolt hole
Resistance R (relative scale) probes. These instruments are simplified, easy to use,
and portable. They generally operate from 100 kHz
Crack to 6 MHz and use differential probes rotating 600 to
(b)
3000 times per minute (10 to 50 rotations per
Scan second). The resolution of the differential scanning
coil system follows the physical laws for general
eddy current testing; when the probe moves on an
Reactance X (relative scale)

orthogonal path over a crack, the crack depth is the

Coils principal factor affecting the amplitude. The savings
in time compared to the hand held technique is
about 80 percent.
Aluminum
In principle, all metals can be tested, including
titanium, aluminum, and steel. The detection
sensitivity depends on the roughness of the surface
and begins at about 0.20 mm (0.008 in.) depth.
Modern eddy current instruments have two basic
displays for bolt hole eddy current signals: (1) the
impedance plane display, representing the change in
total impendence, the combination of resistance and
Resistance R (relative scale) inductive reactance in terms of phase and amplitude
(Figure 61a); (2) the sweep display, where the vertical

Figure 59. Typical location of cracks adjacent to fastener holes: (a) at top; (b) at middle of hole; (c) at bottom; (d) on beveled
surface (ASNT 2004, 469).
(a) (b) (c) (d)
Top view

Sectioned
side view Crack

8.42 AEROSPACE NONDESTRUCTIVE TESTING

amplitude of the signal represents the vertical
component of the signal displayed in the impedance Figure 60. Standardization block for bolt hole and surface probe tests
plane display while the horizontal axis represents the (ASNT 2004, 470).
rotational position of the rotating probe with one
sweep per revolution (Figure 61b). Calibration slots
for hole probe
Although the inspection is fairly straightforward,
the reliability of the inspection is highly dependent
on the equipment setup parameters and signal
interpretation. Parameters such as drive frequency
selection, filtering, and phase angle adjustment are
critical. Improper selection of drive frequency, phase
adjustment, or filters can introduce excessive
inspection noise as a result of being too sensitive to Calibration slots
signals from nonrelevant surface features or may for surface probe
result in suppression of relevant responses, resulting
in significant discontinuities going undetected.

Figure 61. Conventional coil response displays: (a) impedance plane

Probe Drive Frequency display, phase analysis of impedance change in terms of amplitude and
Eddy current fields are confined to the surfaces of phase; (b) sweep display, signal amplitude versus probe rotational
conductive materials as a result of the skin effect. position.
Selection of the oscillating probe drive frequency
establishes the depth that the eddy current field (a)
penetrates into the surface of a given material.
Frequency selection is critical for maximizing the
signal-to-noise ratio. Too low a frequency will result

Reactance X (relative scale)

in reduced sensitivity to small surface cracks. Too
high a frequency will elevate noise from nonrelevant
surface blemishes, scratches, and burrs. As a general
rule of thumb, the frequency should be selected to
establish one standard depth of penetration equal to
a depth of interest.

Rotation Frequency
The peak rotational frequency can be determined
if the coil size and coil scan speed are known. For a
bolt hole eddy current test, the surface speed can be
approximated with the rotational speed of the Resistance R (relative scale)
scanner, in hertz (revolutions per second):
(b)
Peak rotational Surface speed
(6) =
frequency Effective coil diameter
where:
Amplitude (relative scale)

(7)
Surface 
speed  Scanner  π  Hole
Hz diameter

For a shielded coil in high frequency applications,

the effective coil diameter Deff can be approximated
as the coil outside diameter (OD). For unshielded
coils, the effective coil diameter is approximately the
coil diameter plus four times the standard depth of
penetration ;

(8)
Shielded coil Deff  Coil outside diameter
Probe rotational position (relative scale)
(9)
Unshielded coil Deff  Coil outside   ranges of frequencies to pass above and below this
4
diameter
frequency. Because filter design varies with
Therefore, the peak rotational frequency for instrument model and effective coil size can vary
rotary bolt hole inspection can be estimated for any with probe types, filter ranges must be verified
combination of the hole diameter, rotational speed, experimentally by evaluating the amplitude and
and effective coil diameter in Figure 62. symmetry of the signal response for particular
Although these examples identify the defect rotational speeds and hole diameters. An approach
threshold frequency, filters must be selected to allow for simplifying filter selection is to establish a fixed
range of filter settings for a range of hole diameters

ELECTROMAGNETIC TESTING 8.43

 and for a given scanner rotational speed and and an asymmetric signal from a crack about the
effective coil diameter (Figure 63; Table 9). baseline. Also note that improper filter selection
An example of the effect of proper and improper results in excessive sensitivity to noise from
filtering is illustrated in Figure 64. Improper filtering superficial hole damage. Furthermore, the instrument
can distort both the reference discontinuity response gain needed to achieve the required sensitivity from
the reference discontinuity can be much greater for
the unoptimized filter selection than for the
optimized filter selection, indicating a partial
Figure 62. Peak rotational frequency versus hole diameter for various
suppression of the signal of interest. This example
rotational speeds and hole. Effective coil diameter is 1.5875 mm clearly illustrates that proper filter adjustments can
(0.0625 in.). effectively reduce noise to the point where noise
would likely not be confused with a crack signal and
where the symmetry of the discontinuity response is
1000
dramatically improved.

800 Phase Angle

Frequency (Hz)

Apart from proper filter adjustment, further

improvements in signal-to-noise and signal
600 evaluation can be achieved by rotating the phase to
maximize discontinuity response of interest while
minimizing liftoff from hole ovality or surface
400 blemishes. In this approach, sufficient separation
must exist between the liftoff and discontinuity
response of interest. The separation is governed by
200
the material electrical properties (conductivity and
2.54 5.08 7.62 10.16 12.7 15.2 17.8 permeability) and probe oscillating frequency.
(0.1) (0.2) (0.3) (0.4) (0.5) (0.6) (0.7) Enough phase angle separation can be achieved for
Legend Hole diameter, mm (in.) many common aerospace metals, particularly
0
= 1500 rotations per minute aluminum and nickel based super alloys.
= 1250 rotations per minute To achieve separation with most surface scan
= 1000 rotations per minute impedance plane instruments, the liftoff is generated
= 750 rotations per minute
by repeatedly lifting the probe from the reference
= 500 rotations per minute
standard and rotating the phase angle to get a

Figure 63. Peak rotational frequency versus hole diameter for 1.5875 mm (0.0625 in.) diameter coil at
25 Hz (1500 rotations per minute). Red line represents high pass filter, and blue line represents low pass
filter settings adjusted for hole diameter with specific instrument and probes.

1600

1400

1200
Frequency (Hz)

1000

800

600

400

200 2.54 3.81 5.08 6.35 7.62 8.89 10.16 11.43 12.70 13.9 15.24 16.51
(0.1) (0.15) (0.2) (0.25) (0.3) (0.35) (0.4) (0.45) (0.5) (0.55) (0.6) (0.65)

Legend 0 Hole diameter, mm (in.)

= low pass filter
= peak rotational frequency
= high pass filter

8.44 AEROSPACE NONDESTRUCTIVE TESTING

horizontal liftoff. Vertical gain is then adjusted to Signal Interpretation: Impedance
achieve the required sensitivity in the vertical
direction. The same liftoff response can be applied to Plane versus Sweep Displays
rotary bolt hole inspection by rotating the probe As previously demonstrated, nonrelevant noise
against the flat, unflawed surface of a reference can be minimized by appropriate filter selection and
standard. This technique suppresses unwanted noise phase adjustment. When these parameters are
from superficial scratches in the sweep display, properly established, relevant crack indications will
anomalies that could not be eliminated by filtering exhibit a phase response clearly distinguishable from
alone (Figure 64b). Note: instrument gain required to the liftoff line as shown in the impedance plane
achieve the minimum required reference response display of Figure 64c. Indications that exhibit a
may be much higher with than without phase horizontal or off-phase response are often the result
adjustment; however, the overall signal-to-noise of hole ovality or superficial hole damage.
ratio is much improved (Figure 64c). In the sweep display, the response from
superficial hole damage can be misinterpreted as a

Figure 64. Effect of filter settings on signal response of reference standard notch and coupon containing
fatigue crack: (a) without proper filter or phase adjustment; (b) with filter adjustment only; (c) with filter
and phase adjustment.
Reference Standard Fatiguecrackcoupon

(a) Instrument setting

Phase: 144°
Gain: 65.9 decibels
LPF: 500 Hz
HPF: 200 Hz
RPM: 1500

Scratch Crac k signal Liftoff (noise)

(b) Instrument setting
Phase: 144°
Gain: 61.2 decibels
LPF: 1500 Hz
HPF: 500 Hz
RPM: 1500

(c) Instrument setting

Phase: 27°
Gain: 70.6 decibels
LPF: 1500 Hz
HPF: 500 Hz
RPM: 1500

Crack si gnal

Sweep display Sweep display Impedance plane display

Table 9. Peak rotational frequency versus hole diameter.

Filter Hole Diameter

7.9375 to 5.555625 mm 5.55625 to 7.9375 mm 7.9375 to 11.1125 mm 11.1125 to 10.05 mm

(5/16 to 7/32 in.) (7/32 to 5/16 in.) (5/16 to 7/16 in.) (7/16 to 3/4 in.)

Low pass 500 Hz 500 Hz 700 Hz 1500 Hz

High pass 150 Hz 200 Hz 300 Hz 500 Hz

ELECTROMAGNETIC TESTING 8.45

 crack indication without the additional information whereas fluorescent liquid penetrant testing is often
of an impedance plane presentation. By evaluating enough for discontinuities over 1.8 mm (0.07 in.)
both the impedance plane and sweep displays, long. The required discontinuity sizes, orientations,
relevant crack indications can be clearly test areas, and surface conditions dictate the test
distinguished from nonrelevant surface conditions. technique. Parameters such as coil type, test
These distinctions are based on the periodicity and frequency, scanning, indexing, and filtering are
amplitude of the response in the sweep presentation chosen to optimize the ratio of signal to noise for
and on phase and amplitude in the impedance plane. each test.
Additional complications arise when foreign Under manufacturing and maintenance, eddy
material is embedded either on the bore surface or at current tests can be further subdivided according to
faying surfaces of multilayered structures — metal geometry — standard and complex.
shavings from fasteners or drill bits. If the 1. Standard tests typically include test object
contamination is a metallic particle with very configurations that lack significant geometry
different electrical conductivity or permeability than signals or other significant noise. Examples of
the test object, the contamination will usually standard geometries such as continuous surfaces,
exhibit a distinct phase response clearly bolt holes, and scallops are shown in Figure 65.
distinguishable from cracks. However, contamination Surface tests are typically free of strong
with electrical properties similar to those of the test geometry signals for as-manufactured
object can exhibit responses (sweep and phase) components, but test zones can sometimes run
indistinguishable from those of cracks. To eliminate close enough to bolt holes or other features to
these conditions may call for further cleaning, pick up a geometry signal. Surface tests can also
honing, or reaming of the hole. require thresholds so small that additional work
is needed to differentiate discontinuity
responses, especially for inservice components
Examination of Engine that can exhibit fretting damage unless the
probe automatically adjusts for liftoff.
Components Although bolt holes and scallops have
Eddy current testing is a well established significant edge geometries, coil configuration
technique for the detection of surface discontinuities and scan techniques can typically be selected so
in conductive test parts. The United States Air Force that no appreciable edge signal is present during
and other organizations have used automated eddy the tests. However, significant geometry signals
current testing to ensure that critical engine can occur if a bolt hole probe is not accurately
components are free of rejectable surface centered or if scallops cannot be followed
discontinuities. precisely. These geometry signals can render the
Eddy current tests of military aircraft engines can test of such features impractical. Lack of
be categorized as manufacturing, depot or field tests. sophisticated probe positioning can permit
When aircraft engine components are initially unwanted geometry signals. Bolt hole and
produced, eddy current tests of the manufactured scallop geometries with very low test
parts are often required by specification to ensure requirements force standard geometries to use
that the new engine components are free of complex test techniques to extract very small
undesirable material properties or machining discontinuity signals out of the background
discontinuities. Inservice parts similarly require noise.
depot tests at predetermined intervals to ensure that 2. Complex geometries include broach or dovetail
parts are free of small discontinuities in critical, slots, oil drain holes, shaped holes, antirotation
predetermined zones. Manufacturing tests and depot windows, knife edges, air foils, and other
tests are performed as preventive screening. geometries that produce large, unwanted edge
Structural integrity requirements are based on signals as shown in Figure 65. These geometries
fracture mechanics analyses using mission critical require special techniques, sophisticated signal
parameters for each engine component. For a processing, or both to suppress the large,
helicopter engine, conservative service life limits and unwanted geometry response, unless flexible
critical discontinuity sizes are determined by using array probes are used. The development of tests
minimum material properties and other worst case for complex geometries typically requires
operating conditions. Aerodynamic, heat transfer, extremely accurate probe centering to obtain a
stress, and life analysis models are used along with very consistent edge response combined with a
the criticality of a failure to determine test zones and signal removal or differencing technique. These
discontinuity requirements for critical engine edge responses can easily be one hundred times
components. The aviation community recognizes greater than the desired discontinuity response.
that eddy current tests can reliably detect small Often, differencing techniques alone are
discontinuities and locate critical discontinuities. insufficient for suppressing strong geometry
Eddy current tests are typically required for signals, so signal processing techniques are
discontinuities smaller than 1.8 mm (0.07 in.) long required.

8.46 AEROSPACE NONDESTRUCTIVE TESTING

 The testing of critical components of aircraft
Figure 65. Discontinuity locations for fighter engines requires careful attention to ensure that test
engines: (a) corner of cut for balance or drain parameters are carefully set to meet test
slot; (b) corner of balance flange scallop; requirements. Automated equipment is essential to
(c) disk rim; (d) air holes for oil drain, fastener precisely position and scan sensors on geometries
or cooling; (e) corner of antirotation window; where slight inaccuracies in coil positioning can
(f) through knife edge seal; (g) blade interior produce indications that lead to rejection of an
(ASNT 2004, 488). acceptable part. To further improve ratios of signal
to noise with very stringent discontinuity
(a) requirements, sophisticated signal processing
techniques are often applied. Great care is also taken
to minimize system variations. Although certain
techniques are used to increase test throughput, the
driving factors for scheduled tests of critical engine
component tests are adherence to the discontinuity
(b) requirements and test zones specified by the
manufacturer.

(c)

Live rim or dovetail

Blade slot
Corner discontinuity
in disk rim

(d) Corner
discontinuity

Surface discontinuity

(e)

(f)

(g)
Web

Bore

ELECTROMAGNETIC TESTING 8.47

References
Arvin, W.F., and R. Rempt. 1998. ―Detection of Deep Flaws in DOD. 2002. MIL-HDBK-1783B, Engine Structural Integrity Program
Aluminum Structure with Magnetoresistive Sensors.‖ Review of (ENSIP). Arlington, VA: United States Department of Defense.
Progress in Quantitative Nondestructive Evaluation 17A [San Diego, ECT. 2003. Report ECT R8418-R1, Electrical Conductivity of Materials.
CA, July-August 1997]. New York, NY: Plenum, 1039-1042. Virginia Beach, VA: Eddy Current Technology.
ASNT. 2004. Nondestructive Testing Handbook 5: Electromagnetic Fitzpatrick, G.L., D.K. Thome, R.J. Skaugset, E.Y.C. Shih, and
Testing, third edition. Columbus, OH: American Society for W.C.L. Shih. 1993a. ―Magneto-Optic/Eddy Current Imaging of
Nondestructive Testing. Aging Aircraft: A New NDE Technique.‖ Materials Evaluation
ASTM. 2006. ASTM E 2338, Standard Practice for Characterization of 51(12): 1402-1407.
Coatings Using Conformable Eddy Current Sensors without Coating Fitzpatrick, G.L., D.K. Thome, R.J. Skaugset, E.Y.C. Shih, and
Reference Standards 03.03. West Conshohocken, PA: ASTM W.C.L. Shih. 1993b. ―Novel Eddy Current Field Modulation of
International. Magneto-Optic Garnet Films for Real-Time Imaging of Fatigue
Bossi, R., B. Koltenbah, L. Udpa, Y. Dengy, L. Xin, Z. Zeng, and S. Udpa. Cracks and Hidden Corrosion.‖ SPIE Proceedings 2001,
2008. AFRL-RX-WP-TR-20084363. Final Report for Contract Nondestructive Inspection of Aging Aircraft. Bellingham, WA:
FA8650-06-C-5210, Advanced Nondestructive Evaluation Sensor International Society for Optical Engineering, 210-222.
Modeling for Multi-Site Inspection. Wright-Patterson Air Force Goldfine, N., and J. Melcher. 1991. United States Patent 5 453 689,
Base, OH: Air Force Research Laboratory (October). Magnetometer Having Periodic Winding Structure and Material
Daughton, J., J. Brown, E. Chen, R. Beech, A. Pohm, and W. Kude. 1994. Property Estimator.
―Magnetic Field Sensors Using GMR Multilayer.‖ IEEE Transactions Goldfine, N., V. Zilberstein, A. Washabaugh, D. Schlicker, I. Shay, and
on Magnetics 30(6): 4608-4610. D. Grundy. 2003. ―Eddy Current Sensor Networks for Aircraft
Goldfine, N. 1993. “Magnetometers for Improved Materials Fatigue Monitoring.‖ Materials Evaluation 61(7): 852-859.
Characterization in Aerospace Applications,” Materials Evaluation Goldfine, N., Y. Sheiretov, D. Schlicker, and A. Washabaugh. 2008.
51(3): 396-405. ―Rapid, Nonlinear ‗System‘ Identification for NDT, Using Sensor
Maxwell, J.C. 1891. A Treatise on Electricity and Magnetism 2, Response Databases.‖ Materials Evaluation 66(7): 747-755.
3rd edition. New, York, NY: Dover Press (1954). Hagemaier, D.J., and A.P. Steinberg. 1982. ―Low-Frequency Eddy
Na, J.K., and M.A. Franklin. 2005. ―Detection of Subsurface Flaws in Current Inspection of Aircraft Structure.‖ Materials Evaluation
Metals with GMR Sensors.‖ Review of Progress in Quantitative 40(2): 206-210.
Nondestructive Evaluation [Golden, CO, July 2004] 760(1): Hagemaier, D.J. 1983. ―Eddy Current Impedance Plane Analysis.‖
1600-1607. Melville, NY: American Institute of Physics. Materials Evaluation 41(2): 588.
Schlicker, D., N. Goldfine, A. Washabaugh, K. Walrath, I. Shay, Hagemaier, D.J. 1986. ―Applications of Eddy Current Testing to
D. Grundy, M. Windoloski. 2004. United States Patent 6 784 662 Airframes.‖ Nondestructive Testing Handbook 4: Electromagnetic
B2, Eddy Current Sensor Arrays Having Drive Windings with Testing, 2nd edition. Columbus, OH: American Society for
Extended Portions. Nondestructive Testing, 369-421.
Schlicker, D., N. Goldfine, A. Washabaugh, K. Walrath, I. Shay, Hagemaier, D.J., M. Collingwood, and K. Nguyen. 1992. ―Eddy Current
D. Grundy, M. Windoloski. 2006. United States Patent 7 049 811 Standards — Cracks Versus Notches.‖ Materials Evaluation 50(10):
B2, Test Circuit Having Parallel Drive Segments and a Plurality of 1225-1231.
Sense Elements. LaCivita, K.J., J.C. Brausch, and D.C. Laufersweiler. 2006. ―Improved
Bolt Hole Eddy Current Rotational Filtering Procedures and Phase-
Analysis Based Interpretation to Reduce False Calls.‖ In-Service
Inspection of Aircraft Topical Conference, Wichita, KS. Columbus,
Bibliography OH: American Society for Nondestructive Testing, 9-13.
Ansley, G., and R.M. Neufeld. 1976. ―Application and Development Libby, H.L. 1971. Introduction to Electromagnetic Nondestructive Test
Work in Low-Frequency Eddy Current — An Update.‖ Presented at Methods, 1st edition. New York, NY: Wiley.
Air Transport Association Nondestructive Testing Forum, Tulsa, OK. McFarlan, T.W. 1977. ―Low-Frequency Eddy Current Test for Detection
ASM. 1992. ―Eddy Current Inspection.‖ ASM Handbook 17: of Corrosion beneath Aircraft Skin Surfaces.‖ Presented at Air
Nondestructive Inspection and Quality Control, 9th edition. Transport Association Nondestructive Testing Forum, Hartford, CT.
Materials Park, OH: ASM International, 164-194. McMaster, R.C. 1986a. ―The Origins of Electromagnetic Testing.‖
ASNT. 1983. ASNT Level III Study Guide: Eddy Current Testing Method. Materials Evaluation 43(8): 946-956.
Columbus, OH: American Society for Nondestructive Testing. McMaster, R.C. 1986b. ―Introduction to Electromagnetic Testing.‖
Benson, C.W. 2001. ―‗Footprint Assessment of Eddy Current Probes.‖ Nondestructive Testing Handbook 4: Electromagnetic Testing, 2nd
Review of Progress in Quantitative Nondestructive Evaluation edition. Columbus, OH: American Society for Nondestructive
[Ames, IA, July 2000] 20B. Melville, NY: American Institute of Testing, 2-12.
Physics, 1995-2000. McMaster, R.C., and S.S. Udpa. 1986. ―Basic Concepts and Theory of
Berens, A.P. 1992. ―NDE Reliability Data Analysis.‖ ASM Handbook 17, Eddy Current Testing.‖ Nondestructive Testing Handbook 4:
9th edition: Nondestructive Evaluation and Quality Control. Electromagnetic Testing, 2nd edition. Columbus, OH: American
Materials Park, OH: ASM International, 689-701. Society for Nondestructive Testing, 25-51.
Caraway, J., and C. Benson. 2000. ―Automated Eddy Current Inspections McMaster, R.C., and S.S. Udpa. 1999. ―Eddy Current Nondestructive
for Turbine Engine Component Life Extensions.‖ RTO Meeting Evaluation.‖ Wiley Encyclopedia of Electrical and Electronics
Proceedings 34: Paper 30, 1-4. Brussels, Belgium: North Atlantic Engineering. New York, NY: Wiley.
Treaty Organization, Research and Technology Organization. Neufeld R.M. 1976. ―Detection of Cracks in Second Layer Structure with
Concordia, M., P. Ballard, L.S. Price, J. Stawarz, R. Stone, B. Arispe, and Low-Frequency Eddy Current.‖ 36th National Fall Conference
T.J. Braun. 1998. ―Extended Field ECI Applications Progress in Air [Houston, TX]. Columbus, OH: American Society for Nondestructive
Force Depot Level Jet Engine Inspections.‖ ASNT Fall Conference Testing, 46-65.
and Quality Testing Show Paper Summaries [Nashville, TN]. Pellicer, J. 1983. ―Sliding Probe Eddy Current System for Improved
Columbus, OH: American Society for Nondestructive Testing, 166- Fastener Hole Inspection.‖ Presented at 1983 Air Transport
171. Association Nondestructive Testing Forum, Kansas City, MO.

8.48 AEROSPACE NONDESTRUCTIVE TESTING

Rogel, A.P., and J.J. Scalese. 1983. ―Automatic Eddy Current Bolt-Hole Zilberstein, V., I. Shay, R. Lyons, N. Goldfine, T. Malow, and R. Reiche.
Scanning System.‖ Materials Evaluation 41(7): 839-843. 2003. ―Validation of Multi-Frequency Eddy Current MWM Sensors
Staveley Sensors. 1999. ―The Use of Filters for Eddy Current Testing.‖ and MWM-Arrays for Coating Production Quality and
Staveley Sensors Bulletin. Issue SB991.1 (April). Kennewick, WA: Refurbishment Assessment.‖ ASME/IGTI Turbo Expo [Atlanta, GA].
Staveley Sensors. New York, NY: ASME International.
Stepinski, T., and N. Mszi. 1993. ―Conjugate Spectrum Filters for Eddy
Current Signal Processing.‖ Materials Evaluation 51(7): 839-844.

ELECTROMAGNETIC TESTING 8.49

 X
9
CHAPTER

Ultrasoni
c
Testing
Contents
Part 1. Principles of Ultrasonic
Testing, 9.2
Part 2. Basic Ultrasonic
Techniques, 9.14
Part 3. Specialized or Emerging
Techniques, 9.21
Part 4. Relevant Standards and
Specifications, 9.28
Part 5. Application Examples, 9.30
References, 9.37

Contributors
Richard H. Bossi Byron B. Brenden Gary E. Georgeson Doron Kishoni Victoria A. Kramb Jocelyn
Lisa Brasche Barry A. Fetzer [Link]
9.1
E y
r
i G
c .
T
A h
. o
L m
i p
n s
d o
g n
r
e G
n a
r
W y
i
l L
l .
i W
a o
m r
k
P m
. a
M n
o
t
z
e
r

J
e
f
f
r
e
 Principles of Ultrasonic Testing
PART1

Ultrasonic testing plays a critical role in the they are not detectable with a normal incidence
production and inservice inspection of aerospace beam, wedges are then used to give the transmitted
structures. It is applied to metallic and composite beam an oblique direction. Ultrasonic beams that
parts and structures by using a wide range of enter a test object at an angle oblique to the surface
techniques, frequencies, and waveform types. The work well for detecting cracks perpendicular to the
applications range fromroutine to unique. surface: the beam interacts with the corner trap
Predominately, the ultrasonic methods arechecking created and reflects back to the transducer.
for discontinuities such as cracks, corrosion, Ultrasonic testing is based on the transmission,
delaminations, porosity, and inclusions. Ultrasonic reflection, refraction, scattering, and mode changes
testing may also be used for determining of mechanical waves in materials. The wave is a
dimensional measurements and material properties small mechanical disturbance that transfers energy
(Vary 1980; ASNT 1998). Table 1 lists aerospace through a medium (solid, liquid, or gas). For
materials and their inspection issues. The table aerospace applications, the physics of ultrasound
indicates where ultrasonic testing is applied for plays a very critical role in the inspection of many
material inspection. Eddy current, radiographic, materials because the waves‘ mechanical motion is
penetrant, magnetic particle, and other quality related to the material properties of density and
assurance methods maybemore appropriate for modulus (Shull 2002, 91). In aerospace applications,
particular inspection issues. A normal incidence ultrasonic waves cover a wide range of frequencies
ultrasonic beam is best for detecting planar above the audible range but are most commonly
discontinuities perpendicular to its path. When applied in the 1 to 10MHz range.
discontinuities have an oblique orientation such that

Table 1. Aerospace material inspection issues.

Material Inspection Issues Comments on Ultrasonic Test Application

All types cracks, voids, inclusions, Ultrasonic testing requires alignment of ultrasonic beam to test
thickness, coatings subject: either normal to surface of part or oriented at precise
angles to entry surface. Normality is found by peaking signal
response from entry surface or from alignment of ultrasonic
transducers. Curved surfaces may require surface following so
beam orientation can be maintained throughout inspection area.
Radii require concentric alignment of probes to radius in order to
remain normal to surface. For cracks, angle beams may be used to
find crackreflection based onorientation.

Fiber reinforced polymer composite consolidation, porosity, Through-transmission and pulse echo are primary techniques for
laminate/glass epoxy inclusions, fiber-to-resin acceptance of composite laminate materials using automated
ratio, delaminations, scanning with water coupled piezoelectric transducers. Depending
wrinkles, surface layers, on material, configuration, thickness, and sensitivity requirements,
curved surfaces, radii, many ultrasonic techniques are applicable: hand held, resonance,
noodles laser, air coupled, spectroscopic, roller probe, and others. All
testingrequirescorrelation withacceptancestandards. Generally,
consolidation and porosity are monitored by acoustic attenuation
using through-transmission testing, pulse echo amplitude from
back wall, or pulse echo with reflector plate. Delaminations, voids,
and some inclusions are detectable with through-transmission
testing. Pulse echo reflection is more sensitive to inclusions
depending on materials. Wrinkles may be detectable with
ultrasonic testing using pulse echo B-scan imaging at relatively
high frequency (over 3 MHz). Surface coatings can be monitored
with high frequency (for example, 20 MHz) pulse echo or
resonance ultrasonic testing. Pitch catch configuration may be
used across radii to check for quality of noodles in root of
T shaped joints.

9.2 AEROSPACE NONDESTRUCTIVE TESTING

Table 1. Aerospace material inspection issues (continued).

Material Inspection Issues Comments on Ultrasonic Test Application

Foam core composite cracking, voids, density, Through-transmission testing is most common inspection with
bonding to skin, inclusions, standard water squirter systems nominally at 1 MHz, looking for
fluid ingress, skin quality, widerangeofdiscontinuities. Ultrasound maybe air coupled.
skin porosity Resonance and low frequency vibration techniquesmay be used
where access to both sides is limited.

Honeycomb core structure bonding of core to skin, Through-transmission testing is most common technique with
crushed/damaged core, filled standard water squirter systems at 1 MHz, looking for core-to-skin
core, inclusions, skin delaminations, disbonds, and damaged core. Ultrasound may be air
quality/porosity coupled. Resonance andlow frequency vibrationtechniquesmay be
used where access to both sides is limited.

Carbon/carbon consolidation, dry plies, Ultrasonic testing is used to detect delaminations and porosity.
porosity, delamination, Inspection concerns exist about means of coupling to
wrinkles carbon-to-carbon surface.

Castings cracks, voids/porosity, Ultrasonic testingpulse echo normal beam, angle beam, and phased
inclusions, shrinkage, weld array inspection are used for discontinuity location and detection.
repairs, dimensional Grain size effects cause noise that affects detail sensitivity.
tolerances Complex geometry may limit coverage.

Forgings cracks, inclusions, grain size, Ultrasonic testing pulse echo inspection of billets checks for
residual stress inclusions, porosity, and voids. Angle beam ultrasonic testingis used
for inspection for cracks. Complex geometry maylimitcoverage.

Machined parts cracks, residual stress, Pulse echo angle beam inspection and phased arrays are used for
dimensional tolerance, crack detection. Internal dimensions can be checked with high
repairs frequency ultrasonic testing.

Fastened structure cracks, corrosion, alloy type Pulse echo and phased array ultrasonic tests with angle beams are
used for cracks around fasteners. Normal beam ultrasonic testing is
used for corrosion detection by measuring loss of material in top
layer.

Welded joints voids, porosity, lack of Normal or angle beam ultrasonic testing is used for cracks, voids,
fusion, lack of penetration, inclusions, lack of fusion, and lack of penetration in welds. Phased
undercut, shrinkage, cracks, arrays can be used for beam steering in both pulse echo and pitch
slag, inclusions, residual catch modes along welds.
stress

Bonded joint assemblies disbonds, voids, degradation, Ultrasonic testingnormal to bond interfacedetects disbondsand
bondline thickness voids. Spectroscopy can be sensitive to subtle interface changes
correlated to bond quality in some studies. Resonance and low
frequency vibration techniques detect stiffness changes.

Coatings paint thickness, conductive High frequency pulse echo ultrasonic testing can measure paint
layers, thermal coatings, thickness. Thermal coatings, insulation, and low observables usually
insulation, low observable require low frequency in through-transmission mode if possible or
coatings resonance ultrasonic testing.

Subsystems cracks, residual stress, Pulse echo angle and normal ultrasonic testing are useful for crack
surface condition detection.

Inservice and/or damaged structure impact damage, heat Ultrasonic testing is useful for composite impact damage and
damage, moisture ingress, disbond/delaminations with normal beams. Moisture ingress may be
fatigue cracks, corrosion, detected by ultrasonic testing because of changes in wave speeds
lightning strike, and attenuation. Fatigue cracks may be detectable with normal or
disbonds/delaminations angle beam ultrasonic testing. Corrosion is detectable by changes in
thicknessdetected by ultrasonic testing for accessible layers.
Resonance and low frequency vibration techniques are commonly
used to detect inservice damage in core structures.

ULTRASONIC TESTING 9.3

 The two mostcommon ultrasonic techniques are
through-transmission and pulse echo ultrasonic
Basic Wave Characteristics
testing (Figure 1). The through-transmission The ultrasonic wave propagates in a medium as a
technique measures attenuation of sound through a physical displacement of material atoms from their
material from one transducer to another in order to equilibrium position. Figure 3 shows a simple
determine the uniformity of the material or to find sinusoidal wave depiction for an ultrasonic wave
discontinuities. Pulse echo ultrasound reflects from where the vertical axis represents the level of
features such as cracks, delaminations, inclusions, displacement or amplitude A and the horizontal axis
layers, or geometric features. Less common is the represents time t or distance x. Time and distance are
pitch catch configuration (Figure 2), where two related by the velocity v of the wave where v = x/t.
transducers are on the same side of the test object. The wavelength  is the distance between peaks of
There are many variations to the application of the sine wave and is related to the frequency by
ultrasound based on the frequency, waveforms, and f = v/. As thefrequency increases, thewavelength
transducers or sensors that can be used. becomes smaller. The wavelength is also the period P
of the wave when distance is converted to time on
the horizontal axis. Figure 3a represents a wave
Figure 1. Through-transmission and pulse echo ultrasonic testing whose particle motion is shown in time. The wave in
showing effect of discontinuity on wave penetration in material. Figure 3b is shifted in time such that its particle
displacement shows differing amplitude, the wave
Through-transmission actually moving in a different direction from the
ultrasound wave in Figure 3a at each point in time. This is a
Through-transmission phase shift . This phase difference and the relative
ultrasound
particle motion are important when multiple waves
Pulse echo Pulse echo
interact in a material.
Transducer The material displacement and the direction of
the wave may vary representing different modes of
Early
Test sample reflection the transmission. Figure 4 shows the difference
between longitudinal (compression) waves and
transverse (shear) waves: longitudinal waves displace
material particles in the direction of wave
Little or no transmission propagation whereas transverse waves displace
particlesperpendicularly to the wavedirection. These
Transducer transverse waves may also be polarized to have a
particular direction of transverse motion. Because of
the difference in the particle motion, the velocity of
the waves willbe different. The wavelengths will be
different as a function of the frequency.
Figure 2. Pitch catch configuration for two transducers. Table 2 lists the velocity and wavelength of
ultrasound in common materials. Note that
Transmitting transducer Receiving transducer
transverse wave velocity is not given forwater or air
Discontinuity because liquids and gases cannot support transverse
wave particle motion. The materials shown in
Table 2 are linear elastic materials, so velocity is not
a function of frequency in them. For these materials,
the modulus is related to acoustic velocity in
homogeneous and isotropic materials by:

Figure 3. Sinusoidal plane waveforms showing wavelength, period, and velocity: (a) wave; (b) wave at different phase
(ASNT 2007, 36).
(a) (b)

P or  P or 
1.0 1.0

0.5 0.5 A
Amplitude A
Amplitude A

t t
0 or 0 or
2 4 6 8 10 12 14 x 2 4 6 8 10 12 14 x

0.5 0.5

1.0 1.0
Time Time
Legend
A = amplitude of sine wave
P = period, used with time axis t
 = phase shift
 = wavelength, used with distance axis x

9.4 AEROSPACE NONDESTRUCTIVE TESTING

(1) E  v2
(2) E  v2 1   1  2
where E is the elastic modulus (N/m2), v [the letter 1 
vee] is longitudinal velocity (m/s), and  is density In Equation 2,  [the greek nu] is Poisson‘s ratio, the
(kg/m3). The unit balance is achieved because 1 kg ratio of transverse contraction to longitudinal
mass is equal to N·s2·m–1. The more precise extension:
expression for longitudinal waves is:
(3)   – transverse
longitudinal
Figure 4. Directions of particle vibration:
(a) longitudinal wave, also called compression or where  is the symbol for strain. Poisson‘sratio 
has a value between 0 and 0.5 for solid isotropic
pressure wave; (b) transverse wave, also called
materials. Aluminum for example has a value of
shear wave (ASNT 2007, 36). around 0.33. For the transverse waves, the
transverse modulus G (N/m2) is given by:
(a)
2
(4) G  v trans
Direction of propagation Direction of particle motion

Wave propagation in thin linear elastic materials

(b) such as plates, shells, and sheets will exhibit wave
velocities that are a function of the frequency. A
wave traveling in these media will become
Directionof dispersive, which means that, as the wave progresses,
propagation the different frequency components in an original
signal will begin to separate in time because each
Direction ofparticlemotion frequency travels at a different speed.

Table 2. Velocity and wavelength of ultrasound in common materials.

Acoustic
Density impedance Velocity Wavelength (mm)
Material –3
(g·cm ) Wave (mm/µs) 500 kHz 1 MHz 2.25 MHz 3.5 MHz 5 MHz 10 MHz 15 MHz 20 MHz
(g·cm–2·s)

Air 1.20  10–3 0.400 Longitudinal 0.33 0.66 0.33 0.15 0.094 0.066 0.033 0.022 0.017
Water 1.0 0.400 Longitduinal 1.48 2.96 1.48 0.66 0.423 0.296 0.148 0.099 0.074
Acrylic 1.15 3.11E  105 Longitudinal 2.70 5.40 2.70 1.20 0.771 0.540 0.270 0.180 0.135
Transverse 1.10 2.20 1.10 0.49 0.314 0.220 0.110 0.073 0.055
Graphite/ 1.55 4.65  105 Longitudinal 3.00 6.00 3.00 1.33 0.857 0.600 0.300 0.200 0.150
epoxy
Aluminum 2.71 1.71E  106 Longitudinal 6.30 12.60 6.30 2.80 1.800 1.260 0.630 0.420 0.315
Transverse 3.10 6.20 3.10 1.38 0.886 0.620 0.310 0.207 0.155
Magnesium 1.72 9.98E  105 Longitudinal 5.80 11.60 5.80 2.58 1.657 1.160 0.580 0.387 0.290
Transverse 2.30 4.60 2.30 1.02 0.657 0.460 0.230 0.153 0.115
Titanium 4.50 2.73E  106 Longitudinal 6.07 12.14 6.07 2.70 1.734 1.214 0.607 0.405 0.304
Transverse 3.10 6.20 3.10 1.38 0.886 0.620 0.310 0.207 0.155
Steel, mild 7.80 4.60E  106 Longitudinal 5.90 11.80 5.90 2.62 1.686 1.180 0.590 0.393 0.295
Transverse 3.20 6.40 3.20 1.42 0.914 0.640 0.320 0.213 0.160
Steel, 7.83 4.54E  106 Longitudinal 5.80 11.60 5.80 2.58 1.657 1.160 0.580 0.387 0.290
stainless Transverse 3.10 6.20 3.10 1.38 0.886 0.620 0.310 0.207 0.155
Nickel 8.88 5.00E  106 Longitudinal 5.63 11.26 5.63 2.50 1.609 1.126 0.563 0.375 0.282
Transverse 2.96 5.92 2.96 1.32 0.846 0.592 0.296 0.197 0.148
Nickel 8.59 5.00E  106 Longitudinal 5.82 11.64 5.82 2.59 1.663 1.164 0.582 0.388 0.291
chromium Transverse 3.02 6.04 3.02 1.34 0.863 0.604 0.302 0.201 0.151
alloy

ULTRASONIC TESTING 9.5

 change by about a factor of 2, and a change of
Ultrasonic Transmission and 20 dB in amplitude represents a change by a factor
Reflection of 10. Table 3 shows the attenuation fraction and the
dB value that corresponds to it. Measured in
The transmission and reflection of ultrasound in decibels, the neper is 8.7 dB. Figure 5 is a notional
media and across interfaces is fundamental toits diagram showing the effect of frequency on the
application in nondestructive testing. Within a attenuation of ultrasound. When the frequency is
materialand atinterfaces, the ultrasound maintains low, the wavelength may be larger than small
continuity ofparticle velocity, acoustic pressure, and features in a material such as the crystalline grains
phase. These conditions, particularly at boundaries, in metals, so attenuation is low. As the frequency
determine the amplitude and direction of the increases, wavelength becomes smaller and these
transmitted and reflected waves. ultrasound-to-grain interactions will increase
The velocity and wavelength ofultrasound play attenuation due to scatter.
important roles in the selection of ultrasonic The ability of ultrasound to detect features during
techniques and applications. As the ultrasonic inspection is a function of the changes in acoustic
frequency increases, the wavelength becomes shorter impedance at interfaces. This forms the basis for the
and smaller features can be detected. But as the vast majority ofultrasonic inspection techniques
frequency increases, the smaller wavelength will used for discontinuity detection. Transmission and
increase attenuation due to scatter of theultrasound reflectionofultrasonicpressureacrossinterfaces are
and the inspection depth becomes limited. given by:
Attenuation of the sound pressure amplitude of a
plane wave (in two dimensions) can be represented: At Z2
(9) T 
A0  Z2  Z1
A 
(5) A ead and
0
where A is the end pressure amplitude, Ao is the Rt Z2  Z1
initial pressure amplitude,  is the attenuation (10) R  R  Z Z
coefficient (in nepers per unit distance), and d is the 0 2 1
distance traveled. For travel over a distance d, the where At is transmitted pressure amplitude, R is the
attenuation is: reflection coefficient (ratio of reflected Rt to initial
A R0 signal amplitude), T is the transmission
0
(6)  d  ln coefficient (ratio of transmitted to initial signal
A amplitude), and Z1 and Z2 are the acoustic
In practice it is useful to use a logarithmic impedances of materials 1 and 2 at an interface
measurement to express reduction (attenuation) or (Krautkrämer 1990, 15-16). Acoustic impedance is
increase (gain) of a signal. The decibel (dB) unit, the product of the acoustic velocity and density of
named after Alexander Graham Bell, is based on a the medium through which the ultrasonic signal is
logarithmic ratio: propagating. Table 4 lists interface transmission and
I reflection coefficients and their corresponding
(7) 1 dB  10 log10 decibel values for several possible aerospace
I0 inspection interfaces. The table shows that, because
of the large change in acoustic impedance at an air
where I0 and I are the initial and final intensity or interface, it is difficult to couple sound into air. The
power. Intensity or power is proportional to the
high impedance mismatch between air and most
square of the potential V (volts) or pressure
materials to be inspected results in low transmission
amplitude A. This gives the conversion:
A coefficients.
fraction of theThis, in turn,
signal allows to
intensity only
beatransmitted
small
(8) 10 log A2  20 log
I  10 log10 2 10
across the interface. Therefore, immersion or contact
A0 A0 ultrasonic testing is preferred unless a noncontact
10 I
0
where Ao is initial signal amplitude and A is the and dry condition is strictly specified. The velocity
final signal amplitude. of material particles at the interface are equal, so
The dB measure of attenuation is useful for the
range of values that occur in ultrasonics. A change
of 6 dB in the amplitude of a signal represents a

Figure 5. Notional attenuation of ultrasound

with frequency.
–15
Attenuation (dB/cm)

–10

–5

0
0 1 2 3 4 5
Frequency (MHz)
 0.001 60
0.0001 80
0.00001 100

9.6 AEROSPACE NONDESTRUCTIVE TESTING

the pressure is higher in a high density material for detection depending on the relative material
next to a low density material. The transmission acoustic impedances. The equation for the reflection
coefficient minus the reflection coefficient equals 1. coefficient can result in a negative number. This
A negative reflection is a phase inversion, which means that the waveform is phase reversed (upside
cancels excess pressure of the incoming wave for down) at theinterface. This effect can beseen in the
interfaces of high to low density. acoustic waveform and can be useful for the
Cracks are an interface between the materialand interpretation of interfaces. Figure 6 shows an
air and thus provide a large amplitude reflection, example, using simulation software, of the change in
making them easy to detect in pulse echo or waveform for different interfaces for an acrylic block
through-transmission testing. Inclusions are scanned in water at 5 MHz. Notice how differently
embedded material whose interface with the base the phase ofthereflected waveform is changed bya
material may or may not have sufficient reflection crack (Figure 6c), versus a steel inclusion (Figure 6d).

Table 4. Interface transmission and reflection coefficients.

Transmission Reflection Transmission Reflection

Interface Coefficient Coefficient Amplitude Amplitude
2Z2/(Z2 + Z1) (Z2 – Z1)/(Z2 + Z1) (dB) (dB)

Lead zirconate titanate to air 0.00002 –0.99998 –92.69 0.00

Air to lead zirconate titanate 1.99998 0.99998 6.02 0.00
Lead zirconate titanate to water 0.08227 –0.91773 –21.70 –0.75
Water to lead zirconate titanate 1.91773 0.91773 5.66 –0.75
Graphite epoxy to air 0.00017 –0.99983 –75.29 0.00
Air to graphite epoxy 1.99983 0.99983 6.02 0.00
Water to graphite epoxy 1.51713 0.51713 3.62 –5.73
Graphite epoxy to water 0.48287 –0.51713 –6.32 –5.73
Acrylic to graphite epoxy 1.20000 0.20000 1.58 –13.98
Graphite epoxy to acrylic 0.80000 –0.20000 –1.94 –13.98
Aluminum to air 0.00005 –0.99995 –86.58 0.00
Air to aluminum 1.99995 0.99995 6.02 0.00
Water to aluminum 1.84046 0.84046 5.30 –1.51
Aluminum to water 0.15954 –0.84046 –15.94 –1.51
Water to titanium 1.89715 0.89715 5.56 –0.94
Titanium to water 0.10285 –0.89715 –19.76 –0.94
Water to steel 1.93766 0.93766 5.75 –0.56
Steel to water 0.06234 –0.93766 –24.10 –0.56

Figure 6. Simulated 5 MHz signals as function of interface materials in acrylic sample for water immersion test:
(a) configuration; (b) A-scan away from discontinuity; (c) A-scan of crack; (d) A-scan of steel inclusion.
(a) (c) 0.8
0.6
Amplitude (relative units)

Transducer 0.4
0.2
0
Discontinuity (air
–0.2
or steel insert)
–0.4 Insert signal
–0.6
–0.8
6 8 10 12 14
Time (µs)
(b) 0.8 (d) 0.8
0.6 0.6
Amplitude (relative units)

Amplitude (relative units)

0.4 0.4
0.2 0.2
0 0
–0.2 –0.2
–0.4 Front face –0.4
of acrylic Back face
–0.6 –0.6 Insert signal
of acrylic
–0.8 –0.8
6 8 10 12 14 6 8 10 12 14
Time (µs) Time (µs)

ULTRASONIC TESTING 9.7

 Often there is interest in the transmission or 1990). Figure 7 shows the effect of 5 MHz
reflection of ultrasound for thin layers or gaps. A ultrasound on a brass inclusion in a composite
crack or delamination, for example, is a thin gap laminate. At certain frequencies, the reflection is
embedded in a material. Coatings are also thin layers reduced and the ultrasonic transmission increases.
of material. When the ultrasonic wavelength is
longer than the gap or layer, multiple reflections
occur at each interface and the reflection and
transmission pressure depends on the phase position Ultrasonic Refraction
of the wave. Equations for transmission T and Ultrasound will refract (change direction) when
reflection R are given (Krautkrämer 1990, 19) as: passing from one medium to another based on the
relative velocity within the two materials as
1 governed by Snell‘slaw:
(11) T 
Z 2 22d  sin1
1 0.25 1  2  sin   

v1
 Z2 Z1  (13) 
and sin2  v2

  where v1 and v2 are the acoustic velocities in the two

2d 
0.25  Z1  2
sin2
   materials, 1 is the incident angle in material 1, and
 Z2 Z1 
2
 2 is the refracted angle in material 2.
(12) R     Figure 8 shows the refracted beams in a test
 2  
 1  0.25 Z Z  sin2  2d  material 2 from an angulated beam in material 1.
1  2    The interaction of the particle motion at the interface
  Z2 Z1  
will also give rise to mode conversions. As the angle

where d is the gap thickness, Z1 and Z2 are acoustic of incidence is increased from the normal to the
impedances forthe material and the thin layer or surface, both longitudinal and transverse angle
gap respectively, and  is wavelength (Krautkrämer beams are created in the part. When the incident
angle is increased such that the longitudinal wave is
refracted at 90 degrees to normal, the first critical
Figure 7. Simulated 5 MHz transmission and angle is reached and longitudinal waves will no
longer exist in the part. Further increase of the
reflection signals of brass insert in composite as
incident angle will reach the second critical angle,
Relative transmittance reflectance

function of brass thickness. where the refracted transverse wave is refracted at

1.0 90 degrees to normal and will not exist in the part.
Figure 9 shows the modeling of the waves and the
0.8 complications that arise frommultiplereflections
and refractions. In Figure 9a, thetransducer is at
0.6 Transmission 9.6 degrees and producesbothlongitudinal and
transverse waves with mode conversions at the
0.4 interfaces. In Figure 9b, with the transducer at
19 degrees tothe surface normal, transverse wave
0.2 Reflection inspection is performed. Note that water does not
support transverse waves. Because of the multiple
0 wave complication for angle beam inspections, it is
0 0.2 0.4 0.6 0.8 0.0 1.2 1.4 typical to use a wedge with an angle greater than the
Insert thickness (mm) first critical angle so that only transverse (shear)

Figure 8. Refraction of ultrasound transmitting from one material to another: (a) angle of incidence
producing longitudinal wave; (b) first critical angle, no more longitudinal wave.

(a) (b)

First critical angle

1 Reflected wave Reflected wave

1
Material 1 Material 1
Material 2 Material 2
2t
2t
2l Longitudinal wave Transverse wave

Transverse wave
Legend
1 = angle of incidence
2l = angle of refraction for longitudinalwave
2t = refraction for transverse wave

9.8 AEROSPACE NONDESTRUCTIVE TESTING

waves are generated (Figure 10). Table 5 lists critical
angles. The velocity of the transverse wave is slower
Diffraction
than the longitudinal wave, so its shorter wavelength Ultrasonic waves, like electromagnetic waves,
improves resolution. Table 6 lists the incident angle diffract at the edges of objects. When ultrasonic
for creating 30, 45, or 60 degree angle beams of waves interact with an aperture, the sound field will
either longitudinal or transverse waves for diverge as shown in Figure 11. The edges of cracks
immersion water coupling or contact coupling. are sources of spherical diffraction waves that can be
detected if they have sufficient intensity at the
detector.
Figure 9. Example of ultrasonic beams in
aluminum plate from ultrasound: (a) in water at
9.6 degree incident angle; (b) at 19 degree Transducer Sources and
incident angle.
Detection of Ultrasound
(a) Ultrasound is most commonly generated and
Transducer at 9.6° detected bytransducers, sensors that convert
electrical energy into mechanical motion and also

Water
coupling X
Figure 10. Wedge for transverse wave angle beam inspection in
Aluminum aluminum plate.
plate Z
(a) Transducer on
polystyrene wedge for
45°transverse wave in
Transverse wave aluminum plate X Notch inplate
in aluminum
Z
Longitudinal wave Longitudinal
in aluminum wave in water

(b) Longitudinal
wave in
water
Transducerat19° Reflected wave (b)

Corner trap reflection from no tch

Amplitude (scalable units)

0.05
Water
coupling X
0
Aluminum
plate Internal
Z
–0.05 reflections in
wedge
Transverse wave in
aluminum 20 40 60 80
Time (µs)

Table 5. Critical refraction angles.

V1L V2L V2T First Critical Second Critical
Material 1 Material 2
(mm/µs) (mm/µs) (mm/µs) Angle Angle

Water acrylic 1.48 2.7 1.1 33.2° not applicable

aluminum 1.48 6.3 3.1 13.6° 28.5°
titanium 1.48 6.07 2.4 14.1° 38.1°
stainless steel 1.48 5.8 2.3 14.8° 40.1°
high temperature nickel 1.48 5.82 3.02 14.7° 29.3°
chromium alloy

Acrylic aluminum 2.7 6.3 2.5 25.4° not applicable

titanium 2.7 6.07 2.4 26.4° not applicable
stainless steel 2.7 5.8 2.3 27.7° not applicable
high temperature nickel 2.7 5.82 3.02 27.6° 63.4°
chromium alloy

ULTRASONIC TESTING 9.9

 Table 6. Incident angles (1) for refracted beams. See Figure 8.

Longitudinal L Transverse T
Material 1 Material 2
30° 45° 60° 30° 45° 60°

Water acrylic 15.9° 22.8° 28.3° 42.3° 72.1° not applicable

aluminum 6.7° 9.6° 11.7° 17.2° 24.7° 30.8°
titanium 7.0° 9.9° 12.2° 13.8° 19.7° 24.4°
stainless steel 7.3° 10.4° 12.8° 18.8° 27.1° 33.9°
high temperature nickel 7.3° 10.4° 12.7° 14.2° 20.3° 25.1°
chromium alloy

Acrylic aluminum 12.4° 17.6° 21.8° 25.8° 38.0° 49.0°

titanium 12.9° 18.3° 22.7° 34.2° 52.7° 77.0°
stainless steel 13.5° 19.2° 23.8° 35.9° 56.1° not applicable
high temperature nickel 13.4° 19.1° 23.7° 26.6° 39.2° 50.7°
chromium alloy

Several characteristics of these transducers are

Figure 11. Diffraction of acoustic waves at aperture.
listed in Table 7. The coupling coefficient is a
measure ofthe conversion of electrical energy into
Barrier
acoustic energy and should be high. The dielectric
constant should match the electrical impedance of
the detector electronics. Table 8 contains general
information about the transducer material types. The
Diffracted spherical transducers come in a variety of sizes and shapes to
wavefront meet the operational needs. Most often the issues for
the operator are the selection of the transducer
frequency, bandwidth, diameter, and focal length.
These issues are briefly discussed below.
The generation of ultrasound is a function of the
Plain wave transducer design and the electrical excitation. The
excitation is normally apulse of electrical energy or
a tone burst of a specific frequency. With spikes or
square wave pulses, the transducer will respond at
Aperture the fundamental frequency for the piezoelectric
material at thethickness excited but will contain a
frequency spectrum depending on the electrical pulse
width and the manufacture of the transducer.
Transducer manufacturers use piezoelectric materials
that suppliers have cut to the thickness that will
optimize performance at a design frequency. For the
mechanical motion into electrical energy. Table 7 most generalized applications, the transducers are
lists sensor materials common in ultrasonic test manufactured at 1, 2.25, 3.5, 5, 10, and 15 MHz that
systems: lead zirconate titanate (PZT), lead adequately cover a range of aerospace ultrasonic
metaniobate (LMN), lithium niobate (LiNbO3), requirements.
1–3 composites, and polyvinylidene fluoride (PVDF). Transducers that have very little frequency range
The first three are ceramic. Polyvinylidene fluoride is are called narrowband transducers, usually made
a flexible polymer that has much lower density than lightly backed sothat they essentially ring at their
the ceramics. Polyvinylidene fluoride is used in fundamental frequency. These transducers usually
broadband and flexible arrays. The ceramic have a high Q factor, a high conversion efficiency of
transducers are generally more powerful and electrical pulse energy into acoustic energy. Highly
sensitive. damped transducers have a backing applied to the
The 1-3 composite transducers are ceramic rods piezoelectric material during assembly that reduces
embedded in a composite matrix that can improve resonant ringing. Damped transducers willbe able to
the range of the transducer and imaging quality. The generate a broader energy spectrum than
composite transducer is formed bydicing the surface narrowband transducers but usually have a lower
of the ceramic and then filling the interstices with Q factor because of damping. Although it might be
epoxy material. The selection of the dicing thought the high Q factor would be more valuable,
spacing/size and fill materials optimizes the in practice the broadband transducer with a short
performance of the transducer. This matrix duration electrical spike pulse is more common. Its
construction provides a means to adjust the broad spectrum and short duration pulse result in
dielectric constant in order tooptimize pulse form verysensitive depth timing. The narrowband
and amplitude (Splitt 1998).

9.10 AEROSPACE NONDESTRUCTIVE TESTING

Table 7. Ultrasonic transducer materials: lead zirconate titanate (PZT), lead metaniobate (LMN), lithium niobate (LiNbO3),
composite, and polyvinylidene fluoride (PVDF).

Acoustic Frequency
Transducer Density Impedance Constant Coupling Dielectric
Material g/cm3 Coefficient Constant Mechanical Q
g·cm–2·s (rayl) (Hz·m)

7.8 3.05 × 107 1980 0.49 1700 75

PZT-5A
PZT-5H 7.8 3.1 × 107 2030 0.52 3000 70
PZT-4 7.4 3.1 × 107 2030 0.41 1300 80
LMN K81 6.1 1.9 × 107 1525 0.3 300 15
LMN K83 4.5 2.45 × 107 2740 0.39 180 300
LMN K85 5.7 1.85 × 107 1675 0.35 750 15
1-3 composite 4.2 to 5.2 8 × 106 to 15 × 106 [Link] 1500 0.6 to 0.75 900 to 2000 2 to 30
LiNbO3 36-Y 4.64 3.2 × 107 3680 — — 10,000
PVDF 1.8 2.7 × 106 1500 0.14 10 5

Table 8. Ultrasonic transducer general information.

Type Comment

Modified lead zirconate titanate (PZT) PZT5A PZT5A is widely used for generalpurposes; choice for 1-3 composites, hydrophones,
accelerometers, level sensors, acoustic emission, pressure, flow, nondestructive
testing, medical, knock, sonar, igniters.

PZT5H has high coupling and dielectric constants; good for 1-3 composites, arrays,
PZT5H
actuators, sensitive receivers, and line hydrophone applications.

PZT4 PZT4 is highly resistant to depoling under severe mechanical stress and electric
drive; low dielectric losses at high electric fields; high power acoustic radiating
transducers, for ultrasonic cleaning, welding and sonar, highvoltage generators,
medical therapy, and high intensity focused ultrasound.

K81 has very stable parameters under time, temperature, and pressure variations
Modified lead metaniobate (LMN) K81 and has low acoustic impedance, low aging, and Q factor. Used in discontinuity
detectors, thicknessgages, accelerometers, high frequency hydrophones and to
measure acoustic emissions, pressure, knock, flow, level, and well logging under
high pressure and temperatures up to 300 °C.

K83 has low dielectric constant coupled with high frequency constant, resulting in
lowercapacitance for higherfrequency driving, and has low acoustic impedance.
K83
K85 has higher signal-to-noise ratio with low Q factor and low acoustic
impedance.
K85
Piezo composites have a very low mechanical Q factor, low acoustic impedance,
Piezo composites and high electromechanical coupling factor — ideal for discontinuity detectors,
thickness gages, and medical imaging.

PVDF is flexible and acoustically well matched to composite materials. It can be

formed into shaped transducers, providing special focal characteristics. It is useful
Polyvinylidene fluoride (PVDF) in applications requiring very thin piezo elements, where ceramics are too fragile
and cannot be shaped to desired geometries. Center frequencies of 3 to 10 MHz
and –6 dB fractional bandwidths exceeding 100 pecent are typical with PVdF
transducers. It is possible to achieve uniformity of ±1 dB between elements of a
multielement PVdF transducer array.

Very high Q factor transducer material with potential for high temperature
(>500° C) applications.

Lithium niobate LiNbO3 36-Y

ULTRASONIC TESTING 9.11
 transducer will have ringing effects that result in a It is possible to focus ultrasound using lenses.
longer ultrasonic pulse, making the interpretation of Lens material is selected according to the ultrasonic
signal echo times difficult. Figure 12 shows the velocity in the material for refraction. Figure 13
difference between a broadband and narrowband shows how focusing can help resolve features. Both
transducer for separating features in a sample. The steps are detected with an unfocused broad beam
narrowband transducer signal (Figure 12c) has more (Figure 13a) at thetransducer location; afocused
acoustic power, but the sensitivity to separate out beam (Figure 13b) allows finer details to be detected.
features is lost relative to the broadband transducer The transducer generation of ultrasonic waves is
signal (Figure 12b). The broadband signal is able to subject to near- and far-field effects on the
resolve much finer details. uniformity of the beam. Very close to the transducer

Figure 12. Simulated 5 MHz signals as function of transducer bandwith: (a) configuration; (b) broadband, 5 MHz, 80 percent
bandwidth; (c) narrowband, 5 MHz, 20 percent bandwidth.

(a) Transducer

Ultrasound

Step 3
Step 4

(b) (c)
Amplitude (relative units)

Amplitude (relative units)

0.2 0.2
Front surface of
stepb lock
0.1 0.1

0 0

–0.1 –0.1
Step 4
–0.2 Step 3 –0.2

5 10 15 20 25 5 10 15 20 25
Time (µs) Time (µs)

Figure 13. Simulated 5 MHz signals: (a) unfocused broad beam; (b) focused transducer beam.

(a) (b)

F ontsurfa ce of
0.2 0.2
s ep block
Amplitude (relative units)

Amplitude (relative units)

0.1 0.1

0 0

–0.1 –0.1
Step 4
Step 4
–0.2 –0.2
Step 3

5 10 15 20 25 5 10 15 20 25
Time (µs) Time (µs)

Transducer Transducer
Ultrasound
Ultrasound

Step 4 Step 3
Step 4

9.12 AEROSPACE NONDESTRUCTIVE TESTING

theultrasonic waves arejustforming in themedia
and are subject to constructive and destructive
Other Waveforms
interference that can affect the uniformity of the In addition to longitudinal and transverse waves,
pressure waves. As the distance increases from the there are other modes of ultrasound that are
transducer to the far field, the ultrasonic beam sometimes used in the aerospace industry (Table 9).
becomes more uniform but spreads out with Many of the modes are generated by conversion
distance. The diameter and the frequency of the from other modes at interfaces because of beam
transducer govern the length of the near field and orientation, refraction indices, or other geometric
the beam spread in the far field. These effects are characteristics. The wave characteristics can be used
shown in Figure 14. Inspection in the near field can to enhance certain inspections in particular
be problematic because of the zones of low and high geometries.
intensity that can misrepresent reflection from
features. The near field is defined as:
Figure 14. Near and far field effects of transducer.
d2 2
(14) N  Transducer
4
N
where d is the diameter of the transducer, N is the 
near field distance, and  is the wavelength. The
angle of divergence of the beam beyond the near d
field is given:

(15) sin  1.2
d Legend
d = diameter of transducer
N = near field distance
where  is the half angle of the spread shown in  = half angle of spread
Figure 14.

Table 9. Waveform types.

Wave Name Characteristic Generation Applications

Longitudinal (or pressure or Particle displacements are in from transducer planar inspection by attenuation
compressional) direction of wave propagation. measurement or reflection from
features

Transverse (shear) Particle displacements are by first refraction angle at angle beam inspection in parts;
perpendicular to direction of interfaces or by special shorter wavelength than
wave propagation. transverse transducers longitudinal for better resolution
or sensitivity

Lamb (or guided) Waves travel in thin (relative to conversion of compressional or emerging method for in-plane
wavelength) medium. transverse waves in thin media inspections for discontinuities
over extended distances, and
material properties

Surface (or rayleigh) Wave travels on surface. second refraction index on surface breaking cracks and
surface residual stress measurements

Creeping (or creep) Wave is within material but near low angle of entry between near near-surface crack detection
surface. first or second refraction index

ULTRASONIC TESTING 9.13

 Basic Ultrasonic Techniques
PART2

Standard Pulse Echo and Access

The access to the part, the region of interest,
Through-Transmission Testing orientation of the discontinuity, or the material
The pulse echo and through-transmission character of interest often determine whether pulse
ultrasonic test techniques are used at frequencies echo, through-transmission, pitch catch, or other
appropriate for the thickness and attenuation techniques will be used. For structures where only
characteristics of the materials being inspected. The one side is accessible, pulse echo and pitch catch
sensitivity to fine detail is a function of wavelength approaches are normally required. For structures that
and active beam size (diameter). The major issues for canbeaccessed frombothsides,
aerospace applications are the access tothe part, through-transmission testing is preferred if the
coupling of the ultrasound, speed of inspection, and geometry is not too complicated.
sensitivity. Pulse echo inspection has the advantage
of being able to locate in-depth features or
anomalies from one side of the part. The Coupling and Scanners
through-transmission technique has the advantages Coupling of the conventional ultrasound
of more power, no near-surface resolution issues, transducer to the object under inspection is generally
and calibration for acoustic attenuation necessary in the inspection schemes listed in
measurement. Table 10 lists common Table 10. For hand held operations, common
implementations of ultrasound for hand held and coupling materials are water, cleaning fluids,
mechanical scanners.

Table 10. Common means of ultrasonic insonification.

Implementation Method Typical Forms Comments

Hand held contact pulse echo hand scanning Place single transducer at locations or scan over areas of interest
and observe A-scan waveform display. Water or gel is used for
coupling.

through-transmission hand scanning Hold two transducers, one on each side of part, and check for
transmission. Alignment can be difficult.

pulse echo surface scanning Hand heldprobescan be connected to positionencodersand

moved by hand to create linear B-scan or area map C-scan data.

linear array scanning Arrays of transducers can be moved by hand to scan over part.
E-scans (similar to B-scans) are created at each location using
electronic scanning of the array elements.

phased array Array transducers can be placed at region of interest, and array
scanned in phase to create sector (S) formatted B-scans

Mechanical scanners water tank immersion scanning Water tankshave greatest versatility for scanning schemes,
(through-transmission testing), using one or many transducers and orientation for
pulse echo, pitch catch) through-transmission, pulseechoorpitchcatch techniques. Part
must fit in water tank.

surface riding dribbler or bubbler Mechanical systems can replace hand held motion and scan over
(pulse echo) regionsofinterest, creating widevarietyof scans. Transducer
rides on surface in special shoe or captured water column to
maintain orientation and proper coupling.

water squirtersystem (through- Water pressurecan beused to create nonturbulentstream

transmission and pulse echo testing) through nozzle that also contains transducer. Ultrasound is sent
through water column and allows inspection of very large parts.

9.14 AEROSPACE NONDESTRUCTIVE TESTING

alcohol, and gels. For mechanical scanners, the flexibility for inspection of parts that can be
common coupling means are immersion in water, placed in a water tank. Figure 16 shows a surface
surface riding water dribblers, or water squirter riding transducer on a portable scanner system.
systems. Figure 15 shows animmersion system that The portable scanner allows flexibility for field
has two transducer arms that can be moved about a inspection. Figure 17 shows a multiple-axis gantry
part for many combinations of inspection at various system that can handle a complex object for water
angles. Immersion systems offer considerable squirter coupling. Such a system can be designed
to handle very large components. The squirter
can be used in the pulse echo as well as
Figure 15. Immersion scanner. through-transmission test mode.
Scan speeds depend on the data spacing of the
inspection. Mechanical scanners can run at speeds
over 0.5 m/s (20 in./s) and then index at a selected
test data spacing. The data acquisition spacing
should be about a third of the minimum
discontinuity size detection requirement or smaller.
For faster coverage of large areas, multiple
transducers or transducer linear arrays are used. The
array technology is commonly mounted on the
surface following dribbler scanners, either portable
or gantry mounted. The index stepping can then be
an entire array width of about 100 to 200 mm
(several inches or more); data acquisition stepping
(that is, indexing) remains small and results in faster
area scanning while maintaining required detection.

Data Display: A-, B-, and C-Scans

Figure 16. Portable scanner. The displays of ultrasonic data are commonly
called A-, B-, and C-scans. Table 11 explains the
scan types. The A-scan is a plot of amplitude versus
time of the signal measured with a transducer at
one position. The B-scan is a consecutive series of
A-scan waveform values displayed as an image, with
one axis as time and the other axis as distance from
the transducer. Each grayscale level in the image
represents the amplitude of a corresponding A-scan.
The B-scan is likened to a cross section of the test
object. A C-scan provides a planar view — for
example, of a plate lying flat and viewed from
above. The two axes on a C-scan are used to index
locations in the scanned area. The bitmapped image
is related to a parameter taken from the A-scan at
each point, usually the maximum signal amplitude
within a specified gate in the time domain.

Figure 17. Gantry scanner inspection of large

aircraft component.

ULTRASONIC TESTING 9.15

 Table 11. Common ultrasonic testing displays..

Display Scan Type Description Comments

A- scan plot of ultrasonic signal voltage Waveform signal that registers ultrasonic amplitude as function of
versus time time, representing propagation time for transmission and
reflection in test material. Depth in part can be inferred from
propagation time.

B- scan image of time versusposition with The B-scan is image display of array of A-scans. Transducer is
gray scale in image of ultrasonic moved in one direction across test object. Forphased array
signal level transducers, B-scan format can be sector formed by changing
angle of beam from transducer and is called an S-scan.

C- scan image of x and y positions on part; C-scan is normal output of through-transmission scanning where
gray scale of image is value selected received ultrasound signal is typically magnitude of ultrasound
from A-scan at that location signal at position on test object. For pulse echo images, C-scan
value can be amplitude in signal at particular time (amplitude
C-scan) or time that particular amplitude first occurs within gated
region of A-scan (time-of-flight C-scan).

Figure 18. Photograph of composite flat bottom holes used for demonstration of A-, B-, and C-Scan
display.

0 50 100 150 200 250

Scale (mm)

Figure 19. A- and B-scans: (a) A-scan plot of amplitude versus time; (b) B-scan image of time versus
reflector depth. The B-scan presents multiple A-scans; color or gray shade indicates amplitude.
(a) (b)
Front
Internal Back surface
Front surface feature surface
echo echo echo
60
Transducer output (relative units)

40
Time

20
0
–20 Back
surface
–40
–60
0 2 4 6 8 10
1.2
2.5
3.8
5.0
6.2
7.5
8.8

Time (relative units)

Incremented points of interrogation
(arbitrary scale)

9.16 AEROSPACE NONDESTRUCTIVE TESTING

 Figure 18 is a photograph of a flat bottom hole
sample in a 4.3 mm (0.170 in.) thick composite Figure 20. A-scan and amplitude C-scan: (a) A-scans can be interpreted
panel. The flat bottom hole values tell the thickness for amplitude or gated amplitude peak; (b) C-scan plot of amplitude as
ofmaterial at hole locations when ultrasonically function of position.
inspected from the opposite side of the panel.
Figure 19 shows the relationship between A- and Internal Back
(a)
B-scans. The B-scan in Figure 19 is taken over the Front surface feature surface
toprowofflat bottom holes in Figure 18. The echo echo echo Value
vertical time axis in the B-scan has the front surface 60 extracted
at the top and back surface at the bottom. The
deepest cut holes have thus caused a reflection 40

Transducer output (units)

nearer the top surface. The shallowest cut hole shows Electronic
20 gate
a reflection very close to the back surface echo. This
standard was constructed to test the detectability of 0
a pulse echo ultrasonic test system fora single ply
near the back surface. Figures 20 and 21 show how –20
the pulse echo generated A scans can be interpreted
in a C-scan. –40
The C-scan is the two-dimensional planar display
–60
of a value extracted from a series of A-scans taken 0 2 4 6 8 10
over the same area. Using a gating system, either the
peak amplitude in the gate orthe first crossing of a Time (relative units)
threshold in the gate is often used for the display (b) 250 mm
value. The C-scan of Figure 20 shows the peak
amplitude; Figure 21 shows either the first crossing
of a threshold of the time of flight.

100 mm

Figure 21. A-scan and time-of-flight C-scans: (a) A-scan gated for
transit time signal; (b) signal’s time-of-flight indexed to shade of gray,
indicating reflector depth in C-scan.

(a) Back surface

Front surface echo extracted
echo
Internal feature for peak in gate
60 echo extracted
for first in gate
40
Transducer output (units)

20

–20

–40

–60
0 2 4 6 8 10
Time (relative units)
(b)
250 mm

100 mm

ULTRASONIC TESTING 9.17

 Figure 22. Simple resolution sensitivity for scanning: transducer beam size relative to feature size affects sensitivity. In this
drawing, discontinuities are delaminations.

0.6
0.4
0.2
0 D
–0.2 Transducer
–0.4
–0.6
6 8 10 12 14 16
Time (µs)

Discontinuity
Discontinuity size = 2 D Discontinuity size = 1 D size = 0.5 D Discontinuity size = 0.25D
Amplitude (relative units)

0.6 0.6 0.6 0.6

0.4 0.4 0.4 0.4
0.2 0.2 0.2 0.2
0 0 0 0
–0.2 –0.2 –0.2 –0.2
–0.4 –0.4 –0.4 –0.4
–0.6 –0.6 –0.6 –0.6
6 8 10 12 14 6 8 10 12 14 6 8 10 12 14 6 8 10 12 14
Time (µs) Time (µs) Time (µs) Time (µs)

against a reference standard. For longitudinal pulse

Figure 23. Reference standard incorporating thickness steps and flat echo testing, a flat bottom hole is a common
bottom holes. reference standard. Figure 23 is an example of a
standard design that incorporated thickness variation
using a step block configuration and some flat
bottom holes. When using angle beams and corner
trap detection methods for discontinuity detection,
the sensitivity is based on the signal amplitude
compared to a standard. Discontinuities whose
reflection intensity is different from the standard
may be detected with less sensitivity. However,
accept/reject criteria for discontinuities are normally
based on the reflection amplitude in the standard,
usually a machined notch selected for a very
conservative signal. There are numerous blocks
commercially available for calibration of ultrasound
inspection. Figure 24 shows several types of metallic
Sensitivity calibration blocks (ASNT 2007, 197). Figure 24cis
the International Institute of Welding (IIW) block
Sensitivity to discontinuities is a function of the that is very common. For many aerospace
frequency and bandwidth of the ultrasound. inspections, specific reference standards are used
Sensitivity is also a function of the orientation, beam (ASNT 2007, 499).
size, and scan step resolution. When the beam is
normal to the discontinuity and the beam size is
smaller than the discontinuity, the sensitivity can be
estimated from the signal change as the beam is Resonance
scanned over the discontinuity zone, provided no Resonance testing is an alternative ultrasonic
other features in the component interfere with wave technique of interpretation of the ultrasonic
propagation. Figure 22 demonstrates this effect. transducer response. Resonance testing is based on
When the beam is larger than the discontinuity, the establishing a standing wave in the material under
sensitivity is based on the amplitude of the signal test(ASNT 1991, 376). The standing wave occurs
assuming that no other features affect measurement. when the effective thickness of the material is equal
The amplitude based sensitivity must be calibrated to an integral number of half wavelengths.

9.18 AEROSPACE NONDESTRUCTIVE TESTING

Resonance testing is usually applied to either Figure 25 compares a 5 MHz pulse echo
thickness measurement or bond testing. For time-of-flight C-scan of impact damage with the
thickness measurements, the frequency is normally resonance scan‘s Xand Youtputs, separately plotted.
swept over a range of frequencies where the The top image of Figure 25 uses a colorized scale to
wavelengths are smaller than the object thickness. represent the time-of-flight (depth) in the part of the
When the resonant frequency is found, the material signal reflection. The resonance scan is at 237 kHz,
thickness can be gaged according to multiples of its much lowerthan thefrequencies typically used in
corresponding wavelength. the pulse echo technique. Depending on the material
In bond testing, a tone burst at a particular and its thickness, either the X or the Y plots may
frequency is typically used to establish the effective have greater sensitivity to particular material
mechanical impedance of the test object, the material changes. In Figure 25, the scans are taken on the
serving as a termination load for the piezoelectric impacted side of a composite foam core (25 mm, or
transducer. The transducer‘s electrical impedance at 1 in., thick) sandwich structure; Figure 26, however,
the selected frequency can be displayed as an shows resonance inspection of the damage from the
impedance plane plot. As the transducer is scanned far side of the sandwich structure using 152 kHz.
over the test object, changes in mechanical This far side inspection through the foam could not
impedance, such as disbonds or composite impact be accomplished with pulse echo ultrasonic testing
damage, change the electrical impedance output of because of the attenuation of the foam core. The low
the transducer. When used in a scanning mode, it is frequency application ofresonance testing can be
customary to display the X and Y values of the very effective on attenuating and complicated
impedance plane outputas C-scanimage plots. structures, particularly bonded structures.

Figure 24. Ultrasonic test calibration blocks: Figure 25. Comparison of ultrasonic images of
(a) angle beam calibration block; (b) ASME basic impact damage in composite test object:
calibration block; (c) IIW block (ASNT 2007, (a) pulse echo; (b) resonance X plot at 237 kHz;
197). (c) resonance Y plot at 237 kHz.
(a)
9.5 mm 25 mm (a)
Reference
point (0.37 in.) (1.0in.)

75 mm 4 3 2 1
(3.0 in.)
25 mm 63 mm (b)
radius
(1.0 in.) (2.5 in.)
radius
3 mm 4 3 2 1
(0.125 in.) (c)
0.75 mm
(0.03 in.)

(b) 4 3 2 1

Figure 26. Far side test of foam filled sandwich

Side drilled 3t minimum
structure using 152 kHz resonance to detect
t/2 hole (typical)
discontinuities 3 and 4 of test object in
≥38 mm (1.5 in.) Figure 25: (a) X plot; (b) Y plot.
150 mm (6.0 in.) minimum (a)

3t/4
t/4

Notches (optional) t
t/2
4 3

(c)

Side drilled holes 2 mm (b)

(0.08 in.)
1.5 mm
(0.06 in.) Curved notch
25 mm (1 in.)
radius 100 mm
50 mm
(2.0 in.) (4.0 in.)
radius 4 3

ULTRASONIC TESTING 9.19

 Figure 27 shows a diagram of the linear array
Phased Array operation of electronic (E) scanning and sectorial (S)
Ultrasonic phased arraysuse multipleultrasonic scanning. Groups of elements are used together with
elements and electronic timing delays togenerate timing delays to cause the beam to have a particular
and receive ultrasound. Beams are created by shape. The performance of characteristics of phased
constructive or destructive interference through the array systems can be assessed using ASTM
superposition of the waveforms based on the timing procedures (ASTM E 2491 2008). In electronic
ofthe pulsing oneach transducer. Phased arrays scanning (Figure 27a), a small number of transducers
offer advantages over conventional single probe are grouped for sending a wave and then the group is
ultrasonic testing because a phased array beam can shifted along the array to create a scan. In sectorial
be electronically scanned and the beams created by scanning (Figure 27b), the array of elements is timed
the superimposed waveforms can be steered and to cause a series of beams to be directed from a range
provide focus. The electronic scanning (E-scan) of angles slightly different from each other. Changing
permits rapid coverage of test objects. Typically, thetiming causes the elements tofire in turn along
phased array scanning will be an order of magnitude the array and interrogate a single point on the test
faster than single probe scanning. Beam steering, object in what is called a sector scan. Resolution can
called sectorial scanning (S-scan), is used for be controlled by the group spacing. By using a large
mapping components at selected angles. S-scanning number of elements and appropriate timingofthe
is useful for inspections where only a minimal elements, the superposition of the waves from each
footprint is possible. The S-scan display is similar to element can cause a focusing effect (Figure 27c). The
a B-scan but is an imageofthe signal in time versus focusing can bedynamically changed tocreate an
the angle. Electronic timing can be used to effective continuous depth focus.
superimpose thewaveforms from each element to Phased array implementation requires an
optimize the beam shape at the discontinuity for the ultrasonic instrument that can appropriately control
greatest detection sensitivity. Because the phased the timing of signals and record the echo
arrays can direct the ultrasound beam without information. The probe design, however, may need
moving thetransducer, they can generate B-scan to be specifically constructed for some applications,
formatted images without mechanical motion. and the phased array probes are considerably more
Phased array ultrasound is similar in principle to expensive than conventional ultrasonic transducers.
radar and sonar. The arrays can be linear or The predominant advantages of phased arrays are
two-dimensional. Linear arrays are more common. the speed, flexibility, and angle scanning into
The individual elements in the array are wired locations of interest.
independently for pulsing and time shifting.

Figure 27. Phased array scanning: (a) E-scan; (b) S-scan; (c) superposition focusing (ASNT 2007, 91).

(a) (b) (c)

(relative scale)

Focal law 1
Delay

Focal law 5
Linear array

Beam spot N
Acoustic field 1 Acoustic field 5 N
Beam spot 1

l

9.20 AEROSPACE NONDESTRUCTIVE TESTING

 Specialized or Emerging Techniques PART3

Air Coupled Ultrasonic Testing Figure 28. Air coupled through-transmission testing of aircraft
The coupling of ultrasound into a test object is component at 50 kHz .
normally performed through a liquid or gel. This
Air coupled transducer Hand held through-transmission scanning yoke
requires that the couplant contact the surface. In
some cases, it is preferred that the surface not be
contacted; in other cases, it is impractical to apply
couplant. It is possible to couple ultrasound through
the air but not without difficulty. Table 4 lists the
transmission and reflection coefficient for ultrasound
at interfaces. For an interface of lead zirconate
titanate (a piezoelectric element) to air, the
transmitted amplitude has a 92 dB loss. After the
sound is transmitted into a part and then exits, the
total signal loss can be on the order of 160 dB.
Additionally, the attenuation of ultrasound in air can
be significant as the frequency increases, to over
100dB/m at 1 MHz. Toovercome attenuation, air
coupled systems have large, high power transducers
with high dynamic range receiving systems and
operate at lower frequencies (50 kHz to 1 MHz) than
conventional ultrasound. Air coupled ultrasonic
testing is performed in either through-transmission
or pitch catch modes because the tone burst is high Figure 29. Air coupled, pitch catch, ultrasonic
powered. The front surface reflection from the object test of fillet weld radius at 250 kHz.
would dominate the signal if the transducer were
used in pulse echo mode. The pitch catch mode uses
two transducers, but they might not be directly
aligned as in through-transmission testing.
Advances are being made in air coupled
transducers to improve performance. Impedance
matching layers are designed to reduce the losses at
the transducer-to-air interface (ASNT 2007, 131-132;
Bhardwaj 2009). Other air coupled transducer types,
not relying on lead zirconate titanate elements, have
emerged and become more common (Song 2006).
Materials with lower acoustic impedance such as
sandwich panels with foam or honeycomb core
construction are good candidates for air coupled
ultrasonic testing. Figure 28 shows a low(50 kHz)
hand held air coupled inspection of an aircraft
control surface in the field. These are typically
honeycomb structures. A large yoke holds the
transducer and scans around the part while the
operator observes the through-transmission signal
on the display. Care must be taken around the edges
of parts with air coupled ultrasonic testing because Figure 30. Air coupled, through-transmission
the signal can leak. Figure 29 shows a pitch catch testing of bonded laminate panel at 400 kHz.
application of air coupled ultrasonic testing at about
250 kHz, and Figure 30 shows a 400 kHz air coupled
C-scan inspection of a bonded composite laminate.
The technique is adequate for gross discontinuities
such as disbonds or voids but is less sensitive to
porosity or inclusions than is water coupled
ultrasonic testing.
Void/disbond
in adhesive

ULTRASONIC TESTING 9.21

 Electromagnetic Acoustic Laser Ultrasonic Testing
Laser techniques are applicable to the generation
Transducers (EMATs) and detection of ultrasound (Monchalin 2004; ASNT
Ultrasound can be coupled without liquids by 2007, 107-114).The laser energy input into the test
electromagnetic acoustic transducers. This technique materialcreates internal mechanical waves. Surface
is applicable to conductive ormagnetic materials displacement from mechanical waves in the material
because it uses lorentz forces to create and detect can likewise be detected by laser methods. The
ultrasound. Lorentz force is due to the interaction of primary advantage of laser methods is that no
eddy currents and magnetic fields where aforce is mechanical contact is made with the object. Table 12
created by a changing electric current in a magnetic lists other advantages and disadvantages of laser
field. A transducer that contains a coil winding and ultrasonic testing.
a bias magnet can create forces in a nearby The absorption of the laser optical energy
conductor and detect its motion. The quality of the produces thermal heating that can generate useful
transducer depends on the intensity of the magnetic elastic waves by expansion. The pulse width of the
field and currentdensity available. laser generation beam will affect the frequency of
Figure 31 shows cross sectional views of practical the ultrasound. At low pulse energies, the
electromagnetic acoustic transducer designs. Notice predominant ultrasound generation mechanism is
how the electromagnetic acoustic transducer the localized thermal expansions known as the
construction can create a number of particular wave thermoelastic regime. At high pulse energies,
motions byorienting the lorentz force throughthe ablation can occur. Care should be taken to avoid
relative eddy current and magnetic field power levels that melt or evaporate surface material
configurations. (surface ablation). Figure 32 diagrams three modes of
Electromagnetic acoustic transducers areuseful laser interaction on a steel surface where the energy
because they can operate in vacuums, at high is absorbed. The thermoelastic effect on a free
temperature, at high speed, or in moving tests. Their surface is shown in Figure 32a and results in a
abilities include self alignment, phased array conical beam (Figure 33a). To generate a forward
configurations, and excitation of horizontally wave normal to the surface of Figure 33b, a
polarized transverse waves for the measurement of transparent constraining layer as shown in
stress orfortestsof anisotropic weldments. Their Figure 32b may be used. Ablation will also generate
major drawback is that they areless efficient than a normal beam but damages the surface. To increase
piezoelectric transducers. Careful modeling is needed absorption and avoid ablation damage, test surfaces
to design electromagnetic acoustic transducers that may be painted black. Graphite fiber reinforced
perform well for a particular application.

Figure 31. Cross sections of practical electromagnetic acoustic transducer configurations: (a) spiral coil exciting radially polarized
transverse wave propagating normal to surface; (b) tangential field electromagnetic acoustic transducer for exciting plane
polarized longitudinal waves propagating normal to surface; (c) normal field transducer for exciting plane polarized transverse
waves propagating normal to surface; (d) meander coil transducer for exciting oblique longitudinal or vertically polarized
transverse waves, rayleigh waves, or guided modes of plates; (e) periodic permanent magnet for exciting obliquely propagating
horizontally polarized transverse waves or guided horizontally polarized transverse modes of plates (ASNT 2007, 116).

(a) (b) (c)

S S N

N N S
N S

(d) (e)
S

S N S N S N S N
N N S N S N S N S

9.22 AEROSPACE NONDESTRUCTIVE TESTING

Table 12. Advantages and limitations of laser ultrasonic testing.

Laser techniques are noncontacting.

Inremote applications, unlikeothernoncontactultrasonic methods, laser techniquescan operate from substantial
standoff distances.
Broadband signal contains many frequencies.
It is easy to scan contoured parts. Ultrasound beam is generated to propagate normal to test surface.
Scan rate is limited only by ultrasound generation, laser repetition rate, and ultrasound transit times.
Advantages Laser techniquescan interrogate parts withlimited access.
Laser techniquescan performinspectionsinhostileenvironmentswherehumanpresence is notpossible.
Laser techniquescan inspectmovingparts.
Laser techniques can easily generate surface waves and plate waves.

Surface must be suitable to absorb laser energy and generate ultrasound without damage; ablation layer may be
required.
Too much power may damage some surface types.
Broadband signal withlimited strengthat particular frequencies.
Signalquality may be degradedrelative to standard transducers.
Laser techniques can be large and expensive and require laser safe rooms for protection.
Limitations Off normal scanning is usually limited to about 45 degrees. Then part must be repositioned and/or reoriented to
laser beam.
For high power lasers, pulse repetition rate may limit scan rates.

Figure 32. Types of laser generated ultrasound: Figure 33. Radiation energy distribution patterns
(a) thermoelastic or free surface; (b) constrained of laser generated ultrasound: (a) free surface
surface; (c) ablated surface (ASNT 2012, 382). thermal expansion; (b) constrained thermal
expansion or ablation (ASNT 2012, 382).

(a)
Air Steel
(a)

1.5 rad
(90 deg)
Laser beam Heated zone
Laser
impinges
here
1 rad (60deg)
0 rad (0deg)
(b)

Laser beam Heated zone

(b)
Transparent 1.5 rad
bonded plate (90 deg)

(c)

Laser beam Heated zone 0 rad (0deg)

Ablated
material

ULTRASONIC TESTING 9.23

 polymer composites represent a good application for
laser ultrasonic testing because the surface has a thin
Spectroscopy
transparent epoxy layer with black fibers below, In an earlier section, it was pointed outthatthe
constraining heating and creating longitudinal ultrasound can be either broadband or narrowband
ultrasound normal tothe surface (ASM 2001). For depending on whether the piezoelectric transducer is
contoured composite parts, the laser can be scanned damped or undamped and that the frequency is
easily, without undue concern for mechanical affected by the thickness of the element. The
alignment normal to the surface. Angles of up to frequency content in the transmitted and received
about 45 degrees from normal have been signals can be analyzed using spectroscopic
successfully inspected using laser ultrasonic techniques to aid interpretation of the ultrasound. If
scanning. With laser pulse repetition rates of 400 Hz a broadband transducer is the pulser, the signal in
and higher, inspection speed canbe greater than the beam will contain a range of frequencies. Figure
6 m2/h (64 ft2/h). 35 shows the configuration of an ultrasonic
Laser interferometry is the commonly used spectroscopy system (Fitting 1981).
detection technique for laser ultrasonic testing. This The spectrum at the receiving transducer will be
requires a separate laser aimed at the location where altered from the initial spectrum depending on the
the acoustic energy is expected to be at the surface. medium, the material traversed. Layered materials or
The generation and detection laser combinations can inclusions in particular can show spectral shifts
operate in the pulse echo, through-transmission, or indicative ofmaterial condition. Small changes in
pitch catch configurations. Laser ultrasonic testing of thickness, such as corrosion and roughness, are
a complex ply drop part is compared to a detectable with spectroscopic analysis. Spectroscopy
conventional C-scan in Figure 34. The test object is a witha broadband source suffers from bandwidth
composite ply drop standard containing multiple limitations and amplitude at each frequency. A
areas of ply drops and numerous inserts. Similar sequence of narrowband tone burst frequencies or a
sensitivities were realized with similar scan times of chirp (tone burst with changing frequency) can
the two systems. improve the signal-to-noise ratio in spectral analysis
(Tucker 1997). One technique of ultrasonic
spectroscopy is to use angle beams for quantitative
inspection of adhesive bonds (Adler 1999). By
Figure 34. Ply drop standard ultrasonic test: combining longitudinal and transverse wave beam
(a) photograph; (b) laser ultrasonic C-scan; spectroscopy, it is possible to extract material
(c) conventional ultrasonic C-scan. properties associated with adhesive bonds
(Wang 1991).
(a)

Surface Waves
Surface waves are constrained to propagate along
the surface of a solid or liquid. Most of their energy
is concentrated in a relatively small region about
one wavelength deep below the surface. The most
common surface waves are called rayleigh waves,
where the wave travels in a solid medium with an air
or vacuum boundary (Shull 2002, 115). The rayleigh
wave velocity can be estimated by the expression:

(16)v 0.87  1.12 

R  vT
1  
(b) where vR is the rayleigh velocity,  (the Greek nu) is
Poisson‘s ratio, and vT is the transverse (shear)
velocity.
The particle motion is elliptical, like the motion of
a buoy on water as a wave passes. Surface waves
can be generated by the use of a wedge transducer at
the second critical angle where the mode converted
transverse wave is at 90 degrees. Surface waves are
slower than transverse waves for a given material,
typically in the range of 0.87 to 0.96 vT. They can
(c) propagate relatively long distances because the
energy propagates mostly near the surface and is not
spread throughout the bulk of the material. Surface
waves are useful in aerospace applications for crack
and surface blemish detection (Halabe 1999). They
can be sent around curved surfaces to detect surface
cracks in difficult to reach locations (Lavrentyev
1999).

9.24 AEROSPACE NONDESTRUCTIVE TESTING

Guided/Lamb Waves Figure 35. Ultrasonic spectroscopy system (ASNT 2007, 156).
Guided or lamb waves are waves that travel
within a structural boundary. The theory of lamb
waves was originated by Horace Lamb to describe
the characteristics ofwaves propagating in plates Transmitter
(Lamb 1917). Most structures are natural waveguides
when thewavelengths arelarge enough tointeract
with the upper and lower boundaries of the
structure. To get some idea of how guided waves are Transmitting transducer Spectra
developed in a waveguide, imagine many bulk waves
bouncing back and forth inside a waveguide with Test object
mode conversions between longitudinal and Magnitude
transverse constantly takingplaceateachboundary.
The resulting superimposed waveform traveling Receiving transducer
along the waveguide is the sum of all these waves.
The traveling wave velocities are a function of
frequency because the various wavelengths that
Phase
correspond to the frequencies have different ratios
relative to the thickness of the waveguide. The Amplifier
Analysis
dependence of the velocities on the frequency is system
known as dispersion. Dispersion distorts the wave
shape as the wave propagates. This is different from
bulk waves where the velocity is independent of the
frequency. The guided wave appears complex
because it contains a range of frequencies traveling
at different velocities. Changing the spacing between
the source and detector will change the waveform. Figure 36. Guided wave motions in plate: (a) symmetric;
Changes in the material condition of the waveguide (b) antisymmetric (ASNT 2007, 100).
between the source and detector will also change the
waveform.
These waves are also referred to as structural
waves because in a thin element relative to the
wavelength, the wave engages the entire piece in (a)
symmetric or antisymmetric surface motion.
Figure 36 shows the representation of these modes.
Figure 37 then shows an example of a dispersion (b)
curve for guided waves in a 3 mm (1/8 in.) thick
aluminum plate where the phase velocity is plotted
against the frequency. Phase velocity depends on
frequency. The figure includes the transverse
horizontal mode that also propagates inthe structure
but is not dispersive (Redwood 1960; Rose 1999).
The principal benefits of guided waves are the
ability to inspect over long distances, the ability to
inspect hidden structure, and inspection speed
because large area coverage reduces scanning. Figure 37. Example of dispersion curves in guided wave modes in
Greater sensitivity than conventional normal beam 3.18mm (0.125 in.) thick aluminum plate, showing symmetric (S0)
ultrasonic testing can be obtained, even with low and asymmetric (A0) lamb modes as well as transverse horizontal (SH0)
frequency guided wavetesting techniques. However, mode.
guided waves also present several difficulties: their
generation, data extraction, and data interpretation. 100
Guided wave energy can be induced into a
waveguide by various techniques. The challenge is to
excite a particular mode at a specific frequency. 80
Normal beam probes can be used. Angle beam
Phase velocity (mm/µs)

sensors can also be used to impart beams that lead to 60 S0

desired kinds of guided waves in a plate. A comb
transducer (a number of different elements at a
specific spacing) can be used to pump ultrasonic 40
energy into a plate, causing wave propagation of a SH0
certain wavelength in the waveguide (Rose 1999).
Laser ultrasonic testing is also an efficient means of 20
A0
generating guided waves.
0
0 0.2 0.4 0.6 0.8 1.0
Frequency (MHz)

ULTRASONIC TESTING 9.25

 A variety of different problems can be tackled in For nonlinear behavior, Hooke‘s law is expanded
the aircraft industry using guided waves (Rose 2000, to include higher order terms. The nonlinear
1080-1086). Aircraft skins are well suited to guided characteristics ofthematerial can be shown to be
wave testing. Figure 38 shows setup for guided related to the harmonic frequency characteristics by
waves to travel across a lap splice. Ultrasonic energy the parameter  that can be measured
is passed from the transmitter to the receiver through experimentally. It is given by the expression:
the lap splice bond. Integrity of the bond line can be
evaluated in this manner by the quality of the signal. (18) 8A2
  A12k2x

Nonlinear Acoustics where A2 is the amplitude of the second harmonic

frequency, A1 is the amplitude of the fundamental
Homogenous, isotropic materials are generally frequency, k is the wavenumber of the fundamental
considered to have linear elastic behavior according frequency, and x is sample thickness (Yost 1999,
to Hooke‘s law (ASNT 2012, 2007): 2067). This expression says that the ratio of the
harmonic to the fundamental frequency squared
(17)   E  provides a measure of the nonlinearity.
 Nonlinear measurements are the subject of
where E is Young‘s modulus, or material stiffness, research for fatigue and microcracking of materials.
The measurement techniques are subject to a number
 is strain, and σ is stress. Acoustic waves of caveats to obtain useful results (Cantrell 2001,
transmitted in this material will retain their shape Shui 2008, Bermes 2008).
and frequency. However, materials are not uniform
and will contain some degree of heterogeneities that
give rise to nonlinear acoustic behavior. In
particular, microcracking and fatigue result in Real Time Ultrasonic Testing
nonlinear acoustic effects. During nonlinear
transmission, the waveform is distorted because the Imaging
compression of the waveform will transmit slightly Ultrasound can beobserved in real time(that is
differently than the rarefaction portion. When this while it is recorded) byusing several technologies.
happens, harmonics of the fundamental frequency One technique uses a piezoelectric material deposited
are generated. Figure 39 shows this graphically. The on a silicon readout chip. This semiconductor
detection of the harmonics can be used to measure technology takes images at 30 frames per second.
the degree of nonlinear behavior in the material. The system uses a transducer to excite a large area
with ultrasound at 1 to 10 MHz. The reflected
ultrasonic energy is focused by an acoustic lens onto
Figure 38. Guided wave testing of lap splice (ASNT 2007, 105). the acousticsensitive detectorarraysiliconchip.
Typically, the array has about 120  120 pixel
elements. The picture generated represents the
Transmitter acoustic pattern from the illuminated object.
Receiver Figure 40 shows an example of an ultrasonic
imaging camera. The camera has a water filled lens
cavity, a rubber coupling interface, and a liquid
1 crystal display. It can be scanned over the surface to
be inspected. A portable computer is used to control
2
the operational featuresand displays the image.
Figure 40b shows the image pattern obtained over a
composite material directly on the unit liquid crystal
display. Discontinuities in the composite are readily
imaged. The technology is particularly effective for
Figure 39. Nonlinear effect on acoustic propagation results in distorted rapid field inspection of potentially damaged areas.
waveform with harmonic frequency f. Another device for real time ultrasonic imaging is
an acousto-optic sensor (ASNT 1995, 278-284). The
acousto-optic sensor is based on a mesophase
material whose optical characteristics are changed by
Distorted Output composed the ultrasound pressure on the surface. The real time
output signal of attenuated f0
Input signal f0 imaging of the ultrasound field on the sensor is
Material detected by a camera focused on the sensor. Sensors
sample have been readily made in sizes up to 150  150 mm
(6  6 in.); larger sizes are possible.

Acoustic Holography
+ Harmonic 2f0 Acoustical holography is atechnique used to
Transmitter Receiver form an optical image of an ultrasonic field. It is
useful in nondestructive testing because of its
excellent lateral resolution, its ability to focus deep

9.26 AEROSPACE NONDESTRUCTIVE TESTING

within a test volume, and its speed of data television system forms an image of the test object
collection. Two types of acoustical holography using the energy diffracted into the first order. Thus,
equipment have been used for nondestructive testing the liquid surface converts an ultrasonic field pattern
applications. In one system, a hologram is formed by into an optical field pattern that can be read by a
scanning a focused transducer over a plane. In the closed circuit television system.
other system, a liquid surface is used as an ultrasonic Images formed in 100 μs are produced 60 times
detector. per second. This characteristic of the liquid surface
allows examination of up to 20 m2/h (215 ft2/h). At
any instant, a 75  75 mm (3  3 in.) field in the
Scanned Focused Transducer object is seen. Using 5 MHz ultrasound, the field
Holography displays 5625 resolvable picture elements
The scanned focused technique has been used to instantaneously. Thus testing speed and resolution
map voids and inclusions in thick walled metal are the strong features of this system.
components and is useful for sizing (Brenden 1974, The system can be calibrated to provide a map of
Holt 1977). The transducer is usually focused on the the absolute attenuation of the object. Video pixel
surface of the piece being examined and is used as brightness is used as an indicator of attenuation.
the ultrasonic source and receiver. A complete scan More information can be found in the literature
provides a holographic record. The digital (ASNT 2007, 300-302; Hildebrand 1974).
holographic data can then be processed by computer
to focus at any depth within the test volume. This
focusing capability makes possible the measurement
Figure 40. Real time ultrasonic camera system: (a) with portable
of the size and shape of any discontinuity within the
computer; (b) liquid crystal display on search unit.
test volume. Holograms can be made using either
longitudinal or transverse waves.
Measurement accuracy depends upon the depth of
(a)
the discontinuity, the size of the scan plane, and the
wavelength of the ultrasound in the test material.
The uncertainty ΔB of lateral dimensional
measurement is given by:

(19) r
 
B 2L
where L is the size of the scan plane, r is the depth,
and  is the wavelength.

Liquid Surface Holography

The second type of acoustical holography uses a
liquid surface as an ultrasonic detector. A liquid
surface holography system has been successfully
used for the inspection of composite aircraft
components. These systems provide excellent lateral
resolution and greatly increase the speed of
inspection. Video image processing equipment is
included in the system primarily to frame average
for noise reduction in real time. Attenuation
(b)
quantification and discontinuity identification can
be automated.
Holographic images of ultrasonic fields produced
by liquid surface detectors give realistic displays of
relative ultrasonic intensity levels. An object beam
transducer generates a tone burst of 3 or 5 MHz
ultrasound, 50 to 100 μs in duration. This wave
packet propagates through the test object, and
ultrasonic lenses act on the transmitted ultrasound to
project an image of the object on the liquid surface.
A second beam of ultrasound, generated by a
reference beam transducer, is mixed with the object
beam at the liquid surface to form an interference
pattern that shapes the liquid surface into a grating.
The amplitude of the grating is proportional to the
product of the object beam and reference beam
pressure amplitudes. When the liquid surface grating
is illuminated by coherent infrared energy from a
laser diode, some of the energy is diffracted. Where
the grating amplitude is large, more infrared energy
is diffracted; where it is small, less is diffracted. A

ULTRASONIC TESTING 9.27

PART4
Relevant Standards and
Specifications

Ultrasonic inspection inthe aerospaceindustry Additionally, there are a number of industrial

follows a number of specifications. Each company specifications available from industry groups that
may have both internal specifications for itself and may be applied. Table 13 lists a number of the
specifications that it requires of its suppliers. relevant ultrasonic test specifications.

Table 13. Relevant industry specifications for ultrasonic inspection.

Specification
Title Purpose
Number

SAE AMS 2628 Ultrasonic Inspection, Titanium and Provides details for ultrasonic inspection of wrought titanium and
Titanium Alloy Bar and Billet titanium alloy products over 12.7 mm (0.5 in.) in diameter. Procedure is
used typically for locating internal discontinuities, such as cracks, voids,
spongy areas, and other structural discontinuities, which may or may not
be exposed to surface. Most inspection is by longitudinal, pulse echo
immersion with some customers calling for transverse inspection as well.
Specification includes provision for zoned inspection, revision
incorporated as result of Federal Aviation Administration funded research
results.
UltrasonicInspection ofProductOver0.5” Provides procedures for pulse echo ultrasonic inspection of flat,
Thick rectangular, round, cylindrical, andcontouredproducts having thickness
or cross sectional dimension greater than 12.7 mm (0.50 in.), using either
SAE AMS 2630B contactorimmersionmethods, and usinglongitudinal wave ortransverse
wave modes or combinations of the two, as necessary. This specification
may apply to testing finished machined parts provided parts meet basic
testability requirements, such as size, contour, metallurgical structure,
and thickness. This procedure has been used typically for locating and
defining internal discontinuities such as cracks, voids, laminations, and
other structural discontinuities that may or may not be exposed to
surface.
Ultrasonic Inspection of Thin Wall Metal Provides procedures for ultrasonic inspection of thin wall metal tubing of
Tubing titanium, titanium alloy, and corrosion and heat resistant steels and
alloys having nominal outside diameter over 4.762 mm (0.1875 in.) with
outside diameter to wall thickness ratio of 8 or greater and wall
thickness variation not exceeding ±10 of nominal. This process has
been used typically for locating internal discontinuities, such as cracks,
SAE AMS 2634 voids, seams, and other discontinuities, which may or may not be
exposed to surface.
Ultrasonic Thickness Testing Provides general instructions for accomplishing ultrasonic thickness
measurements. Measurements can be made from one side of material
when access to opposite side is restricted. Thisrecommendedpractice is
intended for, but not limited to, use at maintenance and overhaul
facilities to inspect aerospace structures and hardware for material loss
orremainingthickness after reworkor fabrication processes.
Standard Practice for Ultrasonic Pulse- Describes method for ultrasonic examination of materials by pulse echo
Echo Straight-Beam Examination by technique using straight beam longitudinal waves introduced by direct
SAE ARP 2654 Contact Method contact of search unit with part. This technique can be used in inservice
applications and on more limited basis for production inspection for
detection of embedded discontinuities.
StandardPracticefor Fabricating and Provides procedure for fabricating aluminum alloy ultrasonic standard
Checking Aluminum Alloy Ultrasonic reference blocks that can be used for checking performance of ultrasonic
Reference Blocks testing equipment and controlling ultrasonic tests of aluminum alloy
products using pulsed longitudinal waves introduced by either contact or
ASTM E 114-10 immersion.

9.28 AEROSPACE NONDESTRUCTIVE TESTING

Table 13. Relevant industry specifications for ultrasonic inspection (continued).

Specification
Title Purpose
Number

ASTM E 164-13 Standard Practice for Ultrasonic Contact Covers techniques for ultrasonic A-scan examination of specific weld
Examination of Weldments configurations joining wrought ferrous or aluminum alloy materials to
detect weld discontinuities.
ASTM E 214-05 Standard Practice for Immersed Describes ultrasonic examination procedure for detection of
Ultrasonic Examinationbythe Reflection discontinuities in materials using instruments that transmit and receive
Method Using Pulsed Longitudinal Waves pulsed longitudinal ultrasonic waves in immersion. This is process
typically used for detection of embedded discontinuities in billets and
forgings.
ASTM E 317-11 Standard Practice for Evaluating Describes procedures for evaluating following performance characteristics
Performance Characteristics of Ultrasonic ofultrasonic pulseecho examination instruments and systems: horizontal
Pulse-Echo Testing Instruments and limit and linearity; vertical limit and linearity; resolution — entry surface
Systems without the Use of Electronic and far surface; sensitivity and noise; accuracy of calibrated gain
Measurement Instruments controls.
ASTM E 428-08 Standard Practice for Fabrication and Provides procedure for fabrication and control of metal alloy reference
Control of Steel Reference Blocks Used in blocks used to verify performance of ultrasonic instrumentation and
Ultrasonic Examination transducers. Similar approaches areused for other alloys moreprevalent
in aviation including titanium and nickel.
ASTM E 494-10 Standard Practice for Measuring Providestest procedure for measuring ultrasonic velocity in materials
Ultrasonic Velocity in Materials with conventional ultrasonic pulse echo discontinuity detection
equipment. Thismethod is often used to determineskinthickness and
detection losses due to corrosion.
ASTM E 587-10 Standard Practice for Ultrasonic Angle Describes method for ultrasonic examination of materials by pulse echo
Beam Examination by the Contact technique using angular incidence. Approach is often used for detection
Method of cracks on backside of part.
ASTM E 664M-10 Standard Practice for the Measurement Describes procedure for measuring attenuation in materials or
ofApparent AttenuationofLongitudinal components with flat, parallel surfaces using conventional ultrasonic
Ultrasonic Waves by Immersion Method equipment. Important for ensuring adequate inspectability throughout
part's volume.
ASTM E 797M-10 Standard Practice for Measuring Provides test procedure for measuring ultrasonic velocity in materials
Thickness by Manual Ultrasonic Pulse- usingcontact pulse echo method. Often used to determine skinthickness
Echo Contact Method and to detect losses due to corrosion.
ASTM E 1001-11 Standard Practice for Detection and Describes procedures for ultrasonic examination ofbulkmaterials orparts
Evaluation of Discontinuities by the by transmitting pulsed, longitudinal waves through liquid couplant into
Immersed Pulse-Echo Ultrasonic Method material and observing reflected waves for indication of presence of
Using Longitudinal Waves discontinuities. This pulse echo, immersion method is often used for
inspection of billets, forgings and structural elements for detection of
embedded discontinuities, voiding, and porosity.
ASTM E 1065M-14 Standard Guide for Evaluating Describes measurement procedures for evaluating transducers used in
Characteristics of Ultrasonic Search Units ultrasonic inspection.

ASTM E 1901-13 Standard Guide for Detection and Describes procedures for contactultrasonic examination ofbulkmaterials
Evaluation of Discontinuities by Contact or parts by transmitting pulsed ultrasonic waves into material and
Pulse Echo Straight Beam Ultrasonic observing indications from reflected waves using pulse echo. Technique
Methods can be used in inservice applications and on limited basis for production
inspection for detection of embedded discontinuities.
ASTM E 2491-13 Standard Guide for Evaluating Describes procedures for evaluating some characteristics of phased array
Performance Characteristics of ultrasonic test instruments and systems.
Phased-Array Ultrasonic Testing
Instruments and System

ULTRASONIC TESTING 9.29

 Application Examples
PART5

Advantages and Limitations What Can Go Wrong

Ultrasonic testing ofaerospace materials, likeall Ultrasonic testing is the method of choice in
nondestructive test methods, has advantages and aerospace for composite material inspection and is
limitations. Table14 lists the major issues with used in a number of metallic inspections as well. It is
ultrasonic testing. useful to note what can sometimes go wrong with
ultrasonic testingsothat theproper application is
performed. Table 15 lists the major issues with
ultrasonic testing.

Table 14. Advantages and limitations of ultrasonic testing.

Advantages of Ultrasonic Testing

Safe Benign test without risks encountered with ionizing radiation methods.

Mechanical measurement Physics of ultrasound are based on mechanical waves related to basic material properties.

Simple to complex Can be performed with simple hand held low cost equipment.

Technology can be scaled to more difficult inspection and geometries.

Automatable Systems can be adapted to be fully automated with image display outputs.

Speed Can be adapted to provide fast coverage of large areas by using high speed scanners or arrays of
transducers.

Limitations of Ultrasonic Testing

Coupling Must normally be coupled to surface of part to be inspected.

Coupling methods can sometimes be detrimental or prohibitive to inspection.

Does not normally transmit across gaps in parts so inspection only covers up to material gaps.

Interpretation Ultrasound waves come in multiple modes that are converted at interfaces based on materials and angles.

Interpretation of returning or transmitted signals can be difficult if modes and beam paths are not
understood.

Standards Standards of aerospace inspection should be representative of component being testing, and these can be
difficult and/or expensive to obtain.

9.30 AEROSPACE NONDESTRUCTIVE TESTING

 requires immersion techniques that place the focal
Turbine Engine inspection point of the transducer close to the region being
Aerospace turbine engine disk materials and interrogated. Multiple fixed focus transducers may
components are generally inspected with eddy interrogate a specific depth zone within the billet.
current but may also undergo ultrasonic immersion Transducers used for billetinspections mayalso have
techniques. High resolution, high sensitivity tworadii of curvature to account forthe diameter of
inspections are performed using 10 MHz transducers. the part as well as the desired depth of inspection.
Some inspections, especially at large depths, may Phased array probes can also reproduce the focusing
require the use of 5 MHz transducers. Although effect of several different fixed focus transducers. A
actual embedded discontinuities in aerospace time corrected gain or depth amplitude correction is
materials vary widely in composition and physical used to maintain a constant sensitivity level
description, sensitivity has traditionally been throughout the depth zone. Sensitivity is usually
referenced to machined targets, such as flat bottom determined by referencing the focused beam to the
holes or side drilled holes. Some reference standards amplitude from a flat bottom hole target at the depth
offer spherical targets of different sizes and of interest. Calibration of a transducer may require
reflectivity. Spherical targets in reference standards multipleflat bottom hole targets at various depths
are beneficial because they can be made to more throughout the inspection zone to ensure a constant
closely represent the reflectivity range expected for sensitivity (Kramb 2004a).
embedded discontinuities and can be detected from
multiple directions in the specimen (Stubbs 2005b).
Forging Inspections
Billet Inspection Forging inspections are performed on a part
shape roughly the size and shape of the engine
In the first stage of production, the raw material component but before its final machining. This is
may be in the form of a large billet or pancake. called a sonic shape because it is suitable for
Large bulk materials such as these are interrogated ultrasonic testing. Sonic shapes allowtests through
primarily by using longitudinal ultrasonic beams at relatively flat surfaces, minimizing geometry effects
normal incidence. Full coverage for these inspections

Table 15. What can go wrong with ultrasonic testing.

Issue Comment Options to Check or Consider

Lack of penetration Material attenuation is too high. Decrease frequency, increase ultrasound power, use larger
transducers, use tone burst power, or consider resonance technique.
Lack of signal Coupling is poor. Beam is lost, not reflected Check for air bubbles; check surface roughness; change couplant,
back to transducer. material, or method; check material velocities and possible beam
paths; or add wetting agent to water to improve coupling efficiency.
Lack of sensitivity Frequency is too low, so pulse is too broad. Increase frequency, narrow pulse, increase transducer damping,
switch to transverse waves, or reduce data spacing.
Lack of resolution Beam is too large. Use smaller transducer, use focused transducer, or reduce data
spacing.
Poor resolution Contact transducer signal at surface has Add delay line to transducer to change impedance match at front
near front surface too much ringing to detect near surface surface, or increase damping to reduceringing.
discontinuities.
Speed Data take too long to acquire. Use array of transducers, increase data spacing, or use larger
transducer.
Access Transducer access ormaterialcontact is Use reflector plate on back side, use reflector to bounce ultrasound
difficult. beam, usecontacttechniques, usespecially designed transducers,
flood object to transmit through, or send beam along part rather
than across.

ULTRASONIC TESTING 9.31

 as shown in Figure 41. Similar to billet inspections, Serviced Engine Inspections
forging inspections primarily use longitudinal
beams. However, to follow the contours of the sonic Inspection of serviced engine components
shape and provide full coverage, some longitudinal requires full coverage of the interrogated region.
beams are refracted at small incident angles. In a Microstructural changes that may have occurred in
welldesigned scanplan, these refracted longitudinal service or final machining of the part may introduce
inspections will account for the loss in sensitivity as additional noise into the ultrasonic response.
the transducer approaches a corner, as well as Machining marks, as well as complex geometries in
provide additional sensitivity for embedded the finished part, must also be accounted for in
discontinuities misaligned with respect to the design of inspections for serviced engine
inspection surface. Ultrasonic coverage for forgings components. In addition to the challenges inherent
requires a near surface inspection zone beginning as in inspecting the finished product, the detection
close as 1.52 mm (0.060 in.) below the part surface. sensitivity requirements for serviced components
These requirements for high near-surface sensitivity may be higher than that originally required forthe
restrict the number of transducers to those that can new production material. Longitudinal wave
satisfy this requirement. inspections are used over most part geometries, so
Also similar to the billet inspections, subsurface that processing discontinuities that may have been
focusing is used in forging inspections, with multiple missed, or modified by service loads, may be
zones within the part used to provide full depth detected. To satisfy thelarge range of coverage and
coverage and sensitivity. Inspection sensitivity is sensitivity requirements within complicated
usually referenced to flat bottom hole targets of a geometries, serviced engine component inspections
certain size at a particular depth. An added may include refracted transverse in addition to
complication to forging inspections is the presence longitudinal inspections (Figure 42). Refracted
of flow lines created during forging. Toreduce the transverse beams may be directed circumferentially
effects of grain noise, grain orientations are around the part to detectdiscontinuities producedby
preferred that produce locally high noise regions service loads. Transverse beams directed axially
(requiring smaller depth inspection zones) and along the bore inside surface provide additional
narrower beams (Margetan 2002, Kramb 2004b). detection capability for embedded discontinuities
Partsurface curvature can result in an increase or hidden in the high noise regions of the bore. Near
decrease in sensitivity relative to a flat surface. In surface sensitivity is provided by high angle
contrast to billet inspections where part curvature is transverse beamsdirected circumferentially around
accounted for by using a curved transducer, forging the bore and web regions. These high angle
inspections typically account for surface geometry transverse beams are sensitive to surface
through gain correction. Curvature correction factors discontinuities and surface condition, sometimes
that vary withradius of curvature and depth in the detecting machining marks and scratches on the
test object can be applied to adjust sensitivity fora surface. Ultimately, the combination of multiple look
flat surface. Other correction factors can also be angles and modes results in an inspection that
applied to the calibration gains to adjust sensitivity provides nearly full coverage of the interrogated
for material attenuation or for variations in the volume (Figure 43).
calibration target response. In contrast to forging and billet inspections,
calibration and sensitivity for serviced engine
components are usually referenced to side drilled
Figure 41. Schematic drawing of sonic shape holes because they can be used to set up both
cross section with contour of final machined longitudinal and transverse beams (Klaassen 2004).
turbine disk cross section embedded within sonic Most serviced engine component inspections use
shape. Note envelope between finished part and surface focusing techniques. However, phased array
sonic shape may be as small as 1.5 mm (0.06 in.). technology has been implemented into some turbine
engine disk inspections (Kramb 2005a; Kramb
2005b). The phased array ultrasonic technique also
providesinspectioncapability that canbeautomated
with subsurface focusing techniques.

Turbine Engine Disks

Turbine engine disks are typically inspected in
production facilities, where repeatability is
important. Despite careful design, inspection
personnel may affect repeatability more than the
inspection system does. Many systems automate data
acquisition, storage, and display, but the operator
controls positioning, transducers, discontinuity
Sonic shape
cross section evaluation, and gate positioning. Operator variability
is difficult to quantify and control. In high volume
and operator intensive production inspections,
automation has been shown to improve system
Final turbine disk
cross section reliability and repeatability.
Phased array technology lends itself to inspection
automation. The individual element response

9.32 AEROSPACE NONDESTRUCTIVE TESTING

provides information regarding the alignment of the
probe, while focused beams can be used to steer and Figure 42. Multiple mode inspection requirements for serviced turbine
focus the beam at the region of interest. One system engine component.
that incorporates inspection automation with phased Mechanical motion or beam steering
array technology is the fully automated ultrasonic follows contours of curved surface
inspection system developed under the United States
Air Force turbine engine sustainment initiative
program (Stubbs 2005a, 87; Stubbs 2005c, 346-353).
The engine test system incorporates a six-axis
robotic manipulator and phased array ultrasonics to Longitudinal, circumferential,
perform immersion ultrasonic inspections of rotating and radial/axial refracted Turbine disk bore
transverse bore inside
turbine engine components. Figure 44 shows a diameter inspections
picture of the engine test system. Designed for depot
level inspections requiring fully automated testing, Axis of rotation for generating
the engine test system has demonstrated highly transverse beam
repeatable inspection capability under a variety of
conditions.
Longitudinal and circumferential
transverse beam inspections

Ultrasonic Inspection of
Aerospace Structures
Ultrasonic testing is used extensively for the Figure 43. Serviced engine component coverage maps: (a) longitudinal;
detection of cracks, corrosion, delaminations, and (b) circumferential 45° transverse; (c) circumferential 60° transverse;
disbonds in aerospace structures. Given the size of (d) radial-to-axial 45° transverse; (e) all coverage.
many aircraft structures, large scale automated
scanning systems have been developed toinspect (a) (b) (c) (d) (e)
various aircraft structural components. Figures 15
through 17 show automated inspection systems used
in the aerospace industry. Aerospace inspection
involves both inspections at manufacture and
periodic inspections in the field. Many ultrasonic test
designs such as squirter systems require the
disassembly ofthe component from an aircraft for
inspection. Over theyears, mobile ultrasonic test
equipment has been developed and used in depot
maintenance environments for on-wing inspection
ofcracks and corrosion.

Manufacturing Inspection Figure 44. Turbine engine sustainment initiative

Ultrasonic testing is commonly used for the (TESI) fully automated ultrasonic inspection
inspectionofstock material, suchasaluminumsheet system.
or steel forgings, at the manufacturer‘s facility. High
speed arrays of transducers are typically used to keep
pace with production. For composite inspection,
ultrasonic testing is used to validate the
consolidation of the material following the cure
cycle. Ultrasonic beam theory and system
technology, discussed above, are used to focus
energy and provide coverage based on the geometric
configuration of the aerospace components. The
major concerns for composites are the detection of
porosity, foreign material inclusions, disbonds, and
delaminations. The ultrasound system will be
qualified for the inspection based upon the ability to

ULTRASONIC TESTING 9.33

 detect the discontinuity criteria defined bystructural Boththrough-transmission and pulseecho testscan
requirements and builtinto test standards. Figure 45 be performed simultaneously. Similar systems can be
shows amultiple-axis squirter ultrasonic testing designed for very high scanning rates by the
system for the inspection of complex contoured addition of transducer arrays and surface following.
aircraft structures. Such systems can align tothe Typical production data spacing will be set at one
surface at the proper angle for the desired inspection. third the minimum discontinuity size to be detected
by the specification. This means that, for a 6 mm
(0.25 in.) discontinuity, data spacing will be set at
2 mm (0.080 in.). Single transducer systems can
Figure 45. Multiple-axis ultrasonic system for
operate at about 1.9 m2/h (20 ft2/h) of coverage,
inspection of composite structures having depending onpartcomplexity. Arraybased systems
complex contours. can testfaster than 9 m2/h(100 ft2/h), and some
systems canoperate atover 90 m2/h (1000 ft2/h).

Cracks in Multilayer Structures

Lap and splice joints are multiple-layer structures
found in most aircraft where access is available from
a single side of the multiple-layer stack. Cracks
around fastener holes also require inspection.
Because conventional bolt hole eddy current
techniques require the removal of the fastener,
ultrasonic techniques cost significantly less and
minimize maintenance induced damage. An angle
beam transverse wave technique looks forfatigue
cracks around fastener holes. Corner reflectionof the
incident wave between the crack face and layer
surface is the primary component of the crack signal
response, but tip diffraction and the scattering of
Figure 46. View of vertical leg inspection site with adjacent multilayer creep waves around the hole also play a role. This
technique was successfully developed and deployed
(horizontal) joints for lower wing skin spar cap.
for second layer crack detection of the Lockheed
C-141 Starlifter® lower inner wing spanwise splice,
ensuring reliable operation through the end of the
Vertical leg fleet life (Andrew 1996). Alternatively, phased array
Location
of web techniques can also be used to sweep aregion
Potential crack surrounding a bolt hole and detect cracks if present.
Fasteners locations (inred)
(gray regions)
Cracks in Vertical Risers
Interior
Vertical risers and stiffening ribs in aircraft
structures can be found in a variety of forms, one
Ultrasonic signal sent in at angle to propagate example being shown in Figure 46. In most cases,
down vertical leg to fastener holes costly teardowns or entry by the operator into the
Exterior
wing is required for access to these fastener sites
using conventional ultrasonic testing. Often these
measures can result in maintenance induced damage
or unacceptable variation in operator performance
given the difficult work conditions. Oneimportant
Transducer
Wing skin example is the weep hole, placed in the risers of
inner wing panels in C-141aircraft to allow fuel to
beproperlydistributed during flight. These sites were
initiation sites of fatigue crack growth, with cracks
Figure 47. Angle beam pitch catch technique. in both the near and far (shadow) regionofthe hole.
Angle beam transverse wave test techniques are
required to detect fatigue cracks at such locations
Fastener X Far crack with limited accessibility of the transducer from the
site inriser outerwing. A diagram ofthisinspection problem is
shown in Figure 47. To detect the presence of a
Near fatigue crack in thefar crack location, ameans was
crack developed to generate and measure circumferential
Y creep waves around the weep hole with two
transducers in pulse echo and pitch catch modes
(Nagy 1994). Subsequent work developed and
Pitch transducer
45° transverse wedge validated an automated ultrasonic procedure
demonstrating the capability of finding cracks
around holes with limited access under depot
Catch transducer maintenance conditions (Aldrin 2001).

9.34 AEROSPACE NONDESTRUCTIVE TESTING

 Figure 48. Lap splice geometry showing typical scribe locations.

Up

Upper skin Inboard

Upper bonded
doubler

Sealant
A-A
A
Lower skin

Typical scribe location A-A

View from outside

Many conventional ultrasonic test techniques where the sealant was removed. Typically, an
exist for aircraft components. These techniques additional clearance of 2.5 mm (0.10 in.) must be
typically use portable hand held ultrasonic examined circumferentially along the length of the
discontinuity detectors and have been used fuselage. Toensure an adequate inspection, the
successfully since 1980. However, as many airframes transverse wave transducer must be indexed
are being used beyond their original life expectancy, circumferentially as it moves along the length of the
inspection requirements are becoming more fuselage to ensure that the narrow transverse wave
complex, requiring the development of new, more beam is aligned with the crack enough to cause a
sophisticated inspection techniques to enable these sufficient echo. This precaution is reliable but
inspections to be performed with minimal aircraft tedious and time consuming.
disassembly and related maintenance actions to Analternate method uses a 16-elementultrasonic
access and prepare the region to be inspected. An phased array. Rather than using a fixed mechanical
example of inspection of such a structure is the wedge to create a sound beam with a single refracted
lower forward spar cap in the center wing of the angle in the part, multiple groups of the 16 elements
C-130 Hercules® (Lindgren 2005). are energized with time delays, enabling the sound
beam to be electronically refracted, focused, or
swept, based on the ―focal laws‖ generated by the
Phased Array Ultrasonic Testing for computer in the phased array instrument and the
Scribe Line Crack at Lap Splices input from the inspector. Figure 49 shows a phased
Paint is routinely removed from airplane skins array instrument and ultrasonic transducer ona
witha variety oftechniques. However thefillet seal reference standard.
on the edge of a fuselage lap splice joint must be
carefully removed by mechanical means. A plastic
Figure 49. Phased array instrument and
scraper is recommended; sharper objects have been
improperly used toremove thefillet seal, and often ultrasonic transducer on reference standard.
the upper skin at the lap splice is used as a guide for
the tool. At the edge of the lower skin, the tool can
cause a ―scribe line‖ that can serve as a crack
initiation point. An initial inspection is required to
examine the fuselage skin panels to visually identify
panels that contain scribe marks. Periodic ultrasonic
inspections are required to ensure that the scribesdo
not grow into cracks. Figure 48 shows the
configuration of the aircraft skin and scribe line.
Transverse wave ultrasound can be used to
generate a sensitive wave in the skin to detect
echoes reflected from cracks that can grow from the
scribe mark. Because the scribe marks caused by the
tools can occur away from the edge of the upper
skin when the sealant is removed, theinspection
must be performed on the surface of the lower skin

ULTRASONIC TESTING 9.35

 For the scribe line application, sweeping the
Figure 50. Phased array ultrasonic scan geometry: (a) A-scan; sound beam over a range of angles obviates a
(b) sector scan. circumferential scan, because the transducer sweeps
the beam automatically. A traditional A-scan
(a) (amplitude versus time) display is provided, and the
inspector can adjust the angle displayed, but the
primary display monitored by the inspector is the
sector scan (Figure 50). The sector scan displays
angle versus time-of-flight information. The
amplitude of the return signal is given a color that is
displayed on the sector scan. A two-color pallet can
be used in which a crack signal that exceeds the
reject threshold causes a binary change of color that
(b) simplifies the signal interpretation. The inspector is
required only to move the transducer array
longitudinally along the fuselage using the skin lap
40.0 as a probe guide. The array increases the inspection
reliability while improving the scan speed.
Ultrasonic phased arraysarebecoming more
common for inservice airplane inspection
applications. Large multiple-element arrays are used
to examine large structures for disbonds. The large
arrays enable a wide swath of structure to be
examined at one time with multiple longitudinal
wave transducers. The technique is very reliable for
composite panel inspections and for inspections of
composite repairs.

9.36 AEROSPACE NONDESTRUCTIVE TESTING

References
Adler, L., S. Rokhlin, C. Mattei, G. Blaho, and Q. Xie. 1999. ―Angle Kramb, V.A. 2004b. ―Use of Phased Array Ultrasonics in Aerospace
Beam Ultrasonic Spectroscopy System for Quantitative Inspection Engine Component Inspections: Transition From Conventional
of Adhesive Bonds,‖ Review of Progress in Quantitative Transducers.‖ 16th WCNDT 2004 — World Conference on NDT
Nondestructive Evaluation 18B [Snowbird, UT, July 1998]. New [Montréal, Canada, August-September 2004].
York, NY: Plenum, 1553-1559. Kramb, V.A. 2005a. ―Coverage and Sensitivity Considerations for
Aldrin, J., J.D. Achenbach, G. Andrew, C. P‘an, B. Grills, R.T. Mullis, Complex Geometries Using Phased Array Ultrasonics.‖ ASNT Fall
F.W. Spencer, and M. Golis. 2001. ―Case Study for the Conference and Quality Testing Show 2005: Paper Summaries 87.
Implementation of an Automated Ultrasonic Technique to Detect Columbus OH: American Society for Nondestructive Testing.
Fatigue Cracks in Aircraft Weep Holes.‖ Materials Evaluation Kramb, V.A. 2005b. ―Use of Phased Array Ultrasonics in Aerospace
59(11): 1313-1319. Engine Component Inspections.‖ In Proceedings of the 8th Joint
Andrew, G., T. MacInnis, and R.T. Mullis. 1996. ―Second-Layer DoD/FAA/NASA Aging Aircraft Conference [Palm Springs CA,
Ultrasonic Inspection of C-141 Splice Joints.‖ Nondestructive 2005].
Evaluation of Aging Aircraft, Airports, and Aerospace Hardware. Krautkrämer, J., and H. Krautkrämer. 1990. Ultrasonic Testing of
SPIE Proceedings 2945: 436-443. Bellingham, WA: SPIE. Materials, 4th edition. Berlin, Germany: Springer Verlag.
ASM. 2001. ―Quality Assurance,‖ ASM Handbook 21: Composites. Lamb, H. 1917. ―On Waves in an Elastic Plate.‖ Proceedings of the Royal
Materials Park, OH: ASM International. Society of London A 93, 114-128.
ASNT. 1991. Nondestructive Testing Handbook 7: Ultrasonic Testing, Lavrentyev, A. 1999. ―NDE of Metal Surface Breaking Cracks under
2nd edition. Columbus, OH: American Society for Nondestructive Adhesive Coating.‖ Review of Progress in Quantitative
Testing. Nondestructive Evaluation 18B [Snowbird, UT, July 1998]. New
ASNT. 1995. Nondestructive Testing Handbook 9: Special Nondestructive York, NY: Plenum, 1757-1763.
Testing Methods, 2nd edition. Columbus, OH: American Society for Lindgren, E., J. Abel, M. Concordia, T. MacInnis, J. Mandeville,
Nondestructive Testing, 278-284. J.C. Aldrin, P. Christiansen, D. Fritz, T. Mullis, F. Spencer, and
ASNT. 1998. Topics on Nondestructive Evaluation Series 1: Sensing for R. Waldbusser. 2005. ―PoD Results and Deployment of the
Materials Characterization, Processing and Manufacturing. Inspection for the Vertical Legof the C-130 Center Wing
Columbus, OH: American Society for Nondestructive Testing. Beam/Spar Cap.‖ Proceedings of the 8th Joint DoD/FAA/NASA
ASNT. 2007. Nondestructive Testing Handbook 7: Ultrasonic Testing, Aging Aircraft Conference [Palm Springs CA, 2005]. Springfield,
3rd edition. Columbus, OH: American Society for Nondestructive VA: National Technical Information Service, for the Federal
Testing. Aviation Administration.
ASNT. 2012. Nondestructive Testing Handbook 10: Nondestructive Margetan, F.J., M. Gigliotti, L. Brasche, and W. Leach. 2002.
Testing Overview, 3rd edition. Columbus, OH: American Society for DOT/FAA/AR-02/114 Final Report, Fundamental Studies: Inspection
Nondestructive Testing. Properties for Engine Titanium Alloys. Springfield, VA: National
ASTM E 2491. 2008. Standard Guide for Evaluating Performance Technical Information Service, for the Federal Aviation
Characteristics of Phased-Array Ultrasonic Testing Instruments and Administration.
Systems. West Conshohocken, PA: ASTM International. Monchalin, J.P. 2004. ―Laser-Ultrasonics: From the Laboratory to
Bermes, C., J.Y. Kim, J. Qu, and L.J. Jacobs. 2008. ―Nonlinear Lamb Industry,‖ Review of Progress in Quantitative Nondestructive
Waves for the Detection of Material Nonlinearity.‖ Mechanical Evaluation 23A: 3-31. New York, NY: Plenum.
Systems and Signal Processing 22(3): 638-646. Nagy, P.B., M. Blodgett, and M. Golis. 1994. ―Weep Hole Inspection by
Bhardwaj, M. 2009. ―Phenomenally High Transduction Air/Gas Circumferential Creeping Waves.‖ NDT&E International 27(3):
Transducers for Practical Non-Contact Ultrasonic Applications.‖ 131-142. Oxford, United Kingdom: Elsevier.
Review of Progress in Quantitative Nondestructive Evaluation 28A: Redwood, M. 1960. Mechanical Waveguides: The Propagation of
920-927. New York, NY: Plenum. Acoustic and Ultrasonic Waves in Fluids and Solids with
Brenden, B.B., and H.D. Collins. 1974. ―Acoustical Holography with Boundaries. New York, NY: Pergamon.
Scanned Hologram Systems.‖ Holographic Nondestructive Testing. Rose, J.L. 1999. Ultrasonic Waves in Solid Media. London, United
New York, NY: Academic Press, 405-428. Kingdom: Cambridge University Press.
Cantrell, J.H., and W.T. Yost. 2001. ―Nonlinear Ultrasonic Rose, J.L., and L.E. Soley. 2000. ―Ultrasonic Guided Waves for the
Characterization of Fatigue Microstructures.‖ International Journal Detection of Anomalies in Aircraft Components.‖ Materials
of Fatigue 23 (supplement 1): 487-490. Amsterdam, Netherlands: Evaluation 59(10): 1080-1086.
Elsevier. Shui, G., J.Y. Kim, J. Qu, Y.S. Wang, and L.J. Jacobs. 2008. ―A New
Fitting, D., and L. Adler. 1981. ―Ultrasonic Spectral Analysis for Technique for Measuring the Acoustic Nonlinearity of Materials
Nondestructive Evaluation.‖ New York, NY: Springer. Using Rayleigh Waves.‖ NDT&EInternational 41(5): 326-329.
Halabe, U., and R. Franklin. 1999. ―Fatigue Crack Detection in Metallic Shull, P.J., and B.R. Tittmann. 2002. ―Ultrasonics.‖ Nondestructive
Members Using Spectral Analysis of Ultrasonic Rayleigh Waves,‖ Evaluation Theory Techniques and Applications. New York, NY:
Review of Progress in Quantitative Nondestructive Evaluation 18B Marcel Dekker.
[Snowbird, UT, July 1998]. New York, NY: Plenum, 1771-1777. Song, J., S. Holland, and D. Chimenti. 2006. ―A Spherically Focused
Hildebrand, B.P., and B.B. Brenden. 1974. An Introduction to Acoustical Capacitive-Film Air-Coupled Transducer.‖ Review of Progress in
Holography, Rosetta edition. New York, NY: Plenum. Quantitative Nondestructive Evaluation 25B: 908-915. New York,
Holt, A.E., and W.E. Lawrie. 1977. ―Ultrasonic Characterization of NY: Plenum.
Defects.‖ Acoustical Holography 7: 599-609. Splitt, G. 1998. ―Piezocomposite Transducers — A Milestone for
Klaassen, R. 2004. ―Calibration Approaches for Phased Array Ultrasonic Testing.‖ Insight 40(7): 760.
Ultrasound.‖ Proceedings of the Workshop on Phased Array Stubbs, D.A. 2005a. ―Overview of the USAF Automated Ultrasonic
Ultrasonics for Aerospace Applications. Dayton, OH: University of Inspection System for Detecting Internal Defects in Aging Turbine
Dayton. Engine Components.‖ Proceedings of the 8th Joint DoD/FAA/NASA
Kramb, V.[A.] 2004a. ―Defect Detection and Classification in Aerospace Aging Aircraft Conference [Palm Springs CA, 2005]. Springfield,
Materials Using Phased Array Ultrasonic.‖ 16th WCNDT 2004 — VA:National Technical Information Service, forthe Federal
World Conference on NDT [Montréal, Canada, August-September Aviation Administration.
2004].

ULTRASONIC TESTING 9.37

Stubbs, D.A. 2005b. ―Probability of Detection for Embedded Defects: Vary, A. 1980. ―Ultrasonic Measurement of Material Properties.‖
Needs for Ultrasonic Inspection of Aerospace Turbine Engine Research Techniques in Nondestructive Testing 4: 159-204. London,
Components.‖ Review of Progress in Quantitative Nondestructive United Kingdom: Academic Press.
Evaluation 19B [Golden, CO, July 2004]. Melville, NY: American Wang, W., and S.I. Rokhlin. 1991. ―Evaluation of Interfacial Properties
Institute of Physics, 1909-1916. in Adhesive Joints of Aluminum Alloys Using Angle-Beam
Stubbs, D.A., R. Cook, D. Erdahl, I. Fiscus, D. Gasper, J. Hoeffel, Ultrasonic Spectroscopy.‖ Journal of Adhesion Science and
W. Hoppe, V. Kramb, S. Kulhman, R. Martin, R. Olding, D. Petricola, Technology. 5(8): 647-666.
N. Powar, and J. Sebastian. 2005c. ―An Automated Ultrasonic Yost, W.T., and J.H. Cantrell. 1993. ―The Effects of Artificial Aging of
System for Inspection of Aircraft Turbine Engine Components.‖ Aluminum 2024 on Its Nonlinearity Parameter.‖ Review of Progress
Insight 47(3): 346-353. in Quantitative Nondestructive Evaluation 12B [La Jolla, CA, July
Tucker, J.R. 1997. ―Ultrasonic Spectroscopy for Corrosion Detection.‖ 1992]. New York, NY: Plenum, 2067-2073.
Review of Progress in Quantitative Nondestructive Evaluation 16B
[Brunswick, ME, July-August 1996]. New York, NY: Plenum,
1437-1441.

9.38 AEROSPACE NONDESTRUCTIVE TESTING

 X
10 CHAPTER
CHAPTER

Radiologic
Testing
Contents
Part 1. Radiologic Test Principles, 10.2
Part 2. Radiologic Test
Techniques, 10.10
Part 3. Factors Affecting Image
Quality, 10.19
Part 4. Sensitivity Measurement, 10.29
Part 5. Radiologic Test Interpretation,
10.31
Part 6. Specialized and Emerging
Radiation Techniques, 10.33
Part 7. Radiologic Testing
Standards, 10.36
Part 8. Applications of Radiologic
Testing, 10.37
References, 10.43

Contributors
Richard H. Bossi
Clifford Bueno
Trey Gordon
10.1
Timothy E. Kinsella
 Radiologic Test Principles
PART1

Introduction radiation is directly related to the square of the

distance from the source:
Radiologic testing, or radiographic testing (RT), is  2
an important nondestructive test method, unique in d1
its ability to see through objects. Radiologic testing (1) I 2  I1 
differs from most other methods because of its d2 
freedom of orientation and noncontact operation.
The basic principle of radiologic testing with photons intensitIy1 is
where atthe d . at distance d1 and I2 is the
intensity
distance
2
(X and gamma) is the attenuation of photon
Therefore, compared to a distance d from a
transmission through material according to the source, the radiation intensity drops to 25 percent
density D times the thickness T, sometimes referred when the distance is doubled (2d), 11 percent when
to as D  T. This effect makes features or tripled (3d), and 1 percent when the distance is 10
discontinuities that alter the D  T readily detectable. times longer (10d).
Radiologic testing with other radiation sources, such The attenuation of X- and gamma radiation by
as neutrons, follow similar operational principles as their interaction with matter results from a series of
X-rays but with attenuation characteristics particular single events. During each such event, a photon is
to that specific radiation‘s interaction with matter. removed from the beam after undergoing an
Radiologic testing can be performed in a wide interaction with an atomic nucleus or an orbital
variety of configurations. The most common is a electron. The primary interactions that occur at the
combination of source, object, and detector typical energies used for aerospace radiologic testing
separated in space. A radiation beam passes from the are photoelectric absorption, compton scattering, and
source, to the object, then through the object, where pair production. The probability for absorption or
it is altered. The transmitted beam is then detected scattering for any particular radiation type and
by an imager. The effect of radiologic testing is to energy with a specific element is referred to as the
take a three-dimensional object and image it in two cross section. There are three forms in which
dimensions with the image pixel representing the attenuation coefficients are expressed: atomic cross
density or thickness variations in the object. Thus, section, whose unit is the barn (1 b = 10–28 m2); mass
with radiologic testing, the information in a attenuation coefficient (cm2/g); and linear
three-dimensional object is compressed into a attenuation coefficient (cm–1). The linear attenuation
two-dimensional display. coefficient is the probability per unit path length
that a photon will be removed from the beam. The
linear attenuation coefficient is usually represented
Physics of X-Ray Attenuation by the symbol  (mu) and can be determined from
the mass attenuation coefficient by multiplying the
Attenuation of radiation is the decrease in mass attenuation coefficient by the density of the
intensity caused by distance, by passage through a material. The total linear attenuation coefficient is
material, or both. The decrease in the intensity of given by the expression:

Table 1. Radiation attenuation mechanisms (Berger 1975).

Attenuation Mechanism Comment

Photoelectric effect In the photoelectric effect, a photon is absorbed by an atom that in turn ejects an electron from the
atomic shell. The photoelectric effect absorption is greater by Z 4 where Z is the atomic number.

Compton scatter Compton scatter is an effect where an X-ray photon scatters inelastically off an electron, loses part of its
energy, and changes direction. The larger the angle of scatter, the more energy lost by the X-ray photon.
The klein-nishina formula (Berger 1975) can be used to compute the cross section of the scattering
process as a function of X-ray energy and determine the predominant angle of scattering.

Pair production Pair production is the creation of an elementary particle and its antiparticle from an X-ray or gamma ray
photon. Pair production occurs only at high photon energies (>1.02 MeV), where the photon near a
nucleus will create an electron and positron pair. The inverse also occurs. Called electron position
annihilation, it isusedinpositronemissiontomographymedical scanners (Ter-Pogossian 1975;
Bossi 1988). In this case, an electron and positron interact (positron annihilation) and produce two
gamma rays (of about 511 keV) travelling in opposite directions.

10.2 AEROSPACE NONDESTRUCTIVE TESTING

(2)   ph  c  pp where c is the mass attenuation coefficient of
the compound, n is mass attenuation coefficient
of each element of the compound, and Wn is the
where c is the attenuation coefficient for compton mass fraction of each element. Multiplying by the
scatter, ph is the attenuation coefficient for the density of the compound will result in the linear
photoelectric effect, and pp is the attenuation attenuation coefficient that can be used to calculate
coefficient for pair production. transmission and attenuation.
Table 1 explains the attenuation mechanisms. The
photoelectric effect dominates at low energy,
compton scattering at intermediate energies, and pair
production at high energies. At the energy level Table 2. K edge energy levels for common elements.
close to the binding energy of an electron shell is a
sharp increase in absorption called an edge. The K Element Atomic Number (Z ) K edge (keV)
shell energy for atoms is an important effect in
radiologic testing of high atomic number elements.
Table 2 shows K shell edge values for several Aluminum 13 1.56
elements. During X-ray production, these K shell Titanium 22 4.97
edges create characteristic X-rays in large quantities Iron 26 7.12
that affect the X-ray spectrum. Other shell levels Nickel 28 8.35
exist but at low energies that usually do not affect Copper 29 9.00
radiologic testing. Figure 1 shows the attenuation of Tungsten 74 69.6
photons as a function of energy for tungsten, Gold 79 80.9
demonstrating the relative contributions of Lead 82 87.9
photoelectric effect, compton scattering, pair
production, and K shell edge.
The attenuation of penetrating photon radiation
is exponential. The intensity I(E) is a function of the Figure 1. Attenuation of X-rays by tungsten
X-ray energy E transmitted through an absorber (ASNT 2002, 646).
(shield):
10000
E t
(3) I E   I0 E e
Linear attenuation coefficient (cm–1)

1000

where I0(E) is the initial intensity, t is the absorber 100

thickness, and (E) is the linear absorption
coefficient. The specific attenuation coefficient is a 10
function of the energies in the beam spectrum. For
practical purposes, an effective energy that 1.0
represents an average of the X-ray spectrum is used
to calculate most estimates of transmission. 0.1
Figure 2 shows an example of how an object will
0.01
change the transmitted intensity of an X-ray beam.
Figure 3 shows the linear attenuation coefficients for
0.001
common aerospace materials as a function of energy.
0.01 0.1 1 10 100
The attenuation coefficient (E) is commonly
obtained from tables of attenuation coefficients for Energy (MeV)
Legend
elements, such as those in Radiographic Testing = pairproduction
(ASNT 2002, 611-652) and elsewhere. The tables = photoelectric
often report the mass attenuation coefficient  the = scattering
linear attenuation coefficient  divided by density . = total
This results in a compound unit of area to mass,
such as square centimeters per gram (cm2/g). The
tables may also report the attenuation cross section
Figure 2. Change in attenuation for change in material volume.
cs, which is area per atom:
 A Photons with intensity I 0
t Transmitted intensity I
(4) cs  
 Na
I = I0e–µt
where A is the atomic weight (grams per mole) and
Na is Avogadro‘s number (6.02  1023 atoms per
mole).
The mass attenuation coefficient allows for the I = I0e–µ(t–x)
calculation of attenuation in compounds:
       
(5)       W1    W2  …    Wn
 c  1  2  n
Void of thickness x Material with X-ray linear
attenuation coefficientµ

RADIOLOGIC TESTING 10.3

 Manipulation of the attenuation equation will multiplicative correction factor is used. This
allow straightforward calculation of absorber correction is referred to as the buildup factor B.
thickness:
(7) I  I0 Be ts
I0
(6) ln  t
I The buildup factor(s) for specific materials and
photon energies may be found in the literature
This attenuation equation is based on narrow beam (Shleien 1998) and can be approximated:
measurements, which assume that the only radiation
reaching the detector has been scattered through an B  1  t
angle of less than 0.01 steradian — principally the (8)
direct primary radiation. In actual practice, broad
beams let much more scatter reach the detector. The buildup factor depends on the atomic number Z
To correct the attenuation equation for the broad of the absorber and the energy of the initial photon.
beam conditions of radiation shielding, a simple A convenient practical measure of radiation
attenuation is the half-value layer (HVL). The
half-value layer of any specific material is that
Figure 3. Attenuation coefficient as function of energy. thickness that will reduce the radiation intensity to
half of its initial value. The half-value layer is
10000 related to the linear absorption coefficient:

(9) HVL  0.693

1000 2  
ln

100
Similarly, a tenth-value layer (TVL) is that
Attenuation coefficient (cm–1)

thickness of material that will reduce the radiation

intensity to one tenth of its initial value. The
10 half-value layer and tenth-value layer are related as
follows:

1.0 (10) 3.33 HVL  1 TVL

0.1
Therefore, if the half-value layer for a particular
material were 50 mm (2 in.) the corresponding
tenth-value layer would be 167.6 mm (6.66 in).
Table 3 lists some half-value layer thicknesses in
0.01
0.01 1.00
centimeters for some common materials (HEW 1970;
0.10
ASNT 1985).
Energy (MeV) Table 4 indicates approximate radiation energies
Legend compatible with various industrial objects typically
= Aluminum
= Magnesium inspected with radiologic testing.
= Carbon It should be noted that, as the radiation energy
= Titanium increases, the differences between absorbing
= Iron materials become less pronounced than at lower
energies. Because of the photoelectric absorption, the
atomic number of an absorber has a large effect on

Table 3. Half-value layers in centimeters as function of energy.

Density
Material 40 keV 60 keV 100 keV 200 keV 500 keV 1 MeV 3 MeV
g/cm3
Water 1.0 2.60 3.36 4.05 5.06 7.16 9.80 17.5
Acrylic 1.05 2.55 3.10 3.64 4.49 6.35 8.67 15.5
Carbon 2.22 1.52 1.77 2.05 2.54 3.87 4.90 9.52
Aluminum 2.7 0.453 0.917 1.50 2.10 3.29 4.19 7.25
Titanium 4.54 0.069 0.202 0.559 1.16 1.87 2.60 4.36
Iron 7.96 0.0240 0.0726 0.235 0.596 1.04 1.45 2.40
Nickel 8.85 0.0169 0.0522 0.175 0.496 0.901 1.28 2.09
Copper 8.9 0.0161 0.0493 0.169 0.499 0.934 1.32 2.17
Lead 18.7 0.0058 0.0172 0.0649 0.0649 0.402 0.869 1.44
Tungsten 11.3 0.0046 0.0138 0.0085 0.0496 0.283 0.566 0.904
10.4 AEROSPACE NONDESTRUCTIVE TESTING
radiation absorption at energies of 100 kV or less. At
high energies in the 1 MeV range, the material Figure 4. X-ray tube configuration (ASNT 2002, 60).
density becomes the major controlling factor in
determining radiation absorption. A 10 percent Target angle
Cathode
change in radiation energy has a very definite effect Anode target Electron filament
at low energies. In the megaelectronvolt energy beam
ranges, this same change in transmission energy can size
hardly be detected.

Radiologic X-Ray Sources

Bremsstrahlung X-Ray Sources: Tubes
and Linear Accelerators
There are two types of bremsstrahlung X-ray Effective focal size
sources used in aerospace: X-ray tubes (photon Actual focal size
energies up to 450 keV) and electron linear
accelerators (photon energies 1 to 10 MeV). Both
sources generate a continuous bremsstrahlung X-ray
spectrum with a maximum energy determined by the
kilovoltage peak (kVp) energy of the electrons
Figure 5. X-ray tube spectra at 150 and 450 kVp, at 20 degrees target
accelerated into a solid target, typically tungsten.
For an X-ray tube, by far the more common of angle, and with 1 mm aluminum filter.
the two sources used in aerospace, the electrons are
generated in a compact vacuum tube and accelerated 3.5
between the cathode (filament) and anode (target) by
a high voltage potential difference. The deceleration 3.0
of the electron when it strikes the target creates the
bremsstrahlung (―braking radiation‖) X-rays. The 2.5
X-rays will be generated with an energy spectrum.
Number of photons  108

Figure 4 shows a conventional X-ray tube 2.0

configuration where the electrons are accelerated
from a cathode emitter to the target anode. The
1.5
elements in the figure are contained in the vacuum
tube and the X-rays will exit through a window of
material, normally beryllium. The beryllium window 1.0
is a low Z material that allows low energy X-rays to 450 kVp
pass. Often filters of aluminum or copper will be 0.5
added in front of the window to harden the beam by 150 kVp
removing some of the low energy content. Figure 5 0
shows two example spectra, 150 kVp and 450 kVp, 0 50 100 150 200 250 300 350 400 450
from an X-ray tube with a 1 mm aluminum filter
Energy (keV)
window.

Table 4. Approximate radiation energies compatible with various absorbers.

Radiation Source Aluminum or Other Light Metals Steel

(kVp) mm (in.) mm (in.)
2 to 25 0.025 to 0.275 (0.001 to 0.11) 0.025 to 0.25 (0.001 to 0.01)
25 to 50 2.5 to 19 (0.1 to 0.75) 0.25 to 6.35 (0.01 to 0.125)
50 to 150 13 to 75 (0.5 to 3.0) 6.35 to 19 (0.125 to 0.75)
100 to 250 50 to 200 (2.0 to 8.0) 6.35 to 44.5 (0.125 to 1.75)
150 to 400 75 to 300 (3.0 to 12.0) 9.5 to 75 (0.375 to 3)
Iridium-192 not used not used 16 to 100 (0.625 to 4)
Cesium-137 not used not used 19 to 100 (0.75 to 4)
1 MeV not used not used 38 to 125 (1.5 to 5)
Cobalt-60 not used not used 38 to 175 (1.5 to 7)
8 to 12 MeV not used not used not used (3.12)
24 MeV not used not used not used (3.18)

RADIOLOGIC TESTING 10.5

 For an electron linear accelerator (linac) source, radiologic testing. They must have a long half life,
the electrons are accelerated down a guide by means provide large photon intensity from a small amount
of radio frequency (RF) voltages and into the target. of material, be simple to manufacture, and be cost
Figure 6 shows the layout of a linear accelerator. The effective. The isotopes in Table 5 are most often used
radio frequency waveguide design is based on the in radiologic testing and cover the relevant energy
frequency chosen and energy to be generated. The range required for imaging. The main drawback for
electrons are emitted from a cathode in the radio isotope sources for aerospace applications is the low
frequency waveguide. The output necessarily comes output levels for a small material size. The output of
in short pulses based on the operating frequency of the source is described by the specific activity, the
the waveguide, typically on the order of 4 µs pulses decay rate per unit mass. The SI quantity related to
at 400 pulses per second. The generation of the the phenomenon of natural and artificial
X-ray from the bremsstrahlung interaction is radioactivity is the becquerel (Bq), and that of
forward rather than to the side as in the X-ray tube. ―specific activity‖ is becquerel per kilogram (Bq/kg).
The target is therefore thinner and the self-shielding If the source becomes too large physically, the
makes for a harder (higher energy shift) spectrum of effective focal spot increases, which leads to image
X-rays. unsharpness and loss of resolution.

Gamma Sources (Isotopes) X-Ray Source Comparison for

Gamma ray sources are a reliable alternative to Aerospace
an X-ray tube, and even though many different The relative advantages and limitations of the
isotopes exist, only a few are appropriate for X-ray sources are listed in Table 6.

Figure 6. Linear accelerator configuration (ASNT 2002, 70).

Radio
Radio frequency source frequency Laser
system

Pulse transformer Accelerator

Electron
gun Collimator

Table 5. Characteristics of isotope sources commonly used for radiologic testing (ASNT 2002; Sentinel 2009; Eastman Kodak 1980).

Practical Specific Activity

Isotope Energy Half Life
TBq/g (Ci/g)
Thulium-170 52 to 84 keV 127 days 37 (1000)
Ytterbium-169 63 to 308 keV 32 days —
Selenium-75 97 to 401 keV 120 days —
Iridium-192 340 keV mean 74 days 13 (350)
(140 to 1200 keV)
Cesium-137 660 keV 33 years 0.925 (25)
Cobalt-60 1170 to 1330 keV 5.27 years 1.85 (50)

10.6 AEROSPACE NONDESTRUCTIVE TESTING

Focal Spot Size Microfocus Tubes
An important property of X-ray tubes that is Microfocus X-ray machines with focal spots as
critical to aerospace applications is the effective small as 1 m (0.001 mm) can provide enhanced
focal spot size that generates the image. Inside the discontinuity detection with resolution greater than
X-ray tube, the area on the target being bombarded obtainable with conventional radiologic equipment.
by accelerated and focused electrons from the heated The source size contributes to the radiologic quality,
filament that emits X-rays is known as the actual influences the geometry of the inspection, and sets
focal spot size. To reduce the apparent size of the radiologic image definition and resolution limits.
focal spot, the exit window from the X-ray tube for Figure 7 shows a comparison of standard focus with
the beam is positioned at a small angle as shown in microfocus. The geometry shows how with a
Figure 4. From the position of the X-ray microfocus target the image of a feature can be
film/detector, the focal spot area appears as the enlarged with little or no blurring (unsharpness).
projection of the actual focal spot on the detector Radiologic testing with very small X-ray sources
plane but reduced in size by the tangent of the angle offers the advantages of sharper images (less
of the target. This projection is referred to as the
effective focal spot and is smaller than the actual
spot size. This minimizes the X-ray source size for Figure 7. Magnification comparison: (a) standard radiologic test
imaging. Because of this angle, the X-ray beam geometry; (b) geometry with microfocus spot.
intensity will drop sharply toward the heel side
(lower left side of anode target in Figure 4). The (a)
intensity drop is known as the heel effect. Test object Blur (geometric
Focal spot sizes usually increase with increasing unsharpness)
kilovoltage rating to prevent the melting of the
target material. Typical focal spot sizes are in the 0.5 Featureimagesize
to 2 mm range for common X-ray tubes. Table 7
shows the industrial standards defining the ways
focal spot sizes are measured. Smaller focal spots Feature
have resulted in microfocus X-ray systems with Detector
plane
targets in the 0.001 to 0.1 mm range and nanofocus
with submicrometer sizes. Microfocus sources may
be required for applications requiring large (b)
geometric magnification or nonfilm techniques to Test object
compensate for resolution lower than film. In
gamma radiologic testing, the pellet of radioactive
material functions as the focal spot. The relatively Featureimagesize
large size of the pellets causes a large effective focal
spot. Typically, the isotope sources have larger focal
spot sizes than X-ray tubes and are not widely used Detector
in aerospace applications. plane

Table 6. Advantages and limitations of X-ray sources.

Type Advantages Limitations

X-ray tube Photon energy range is ideal for most aerospace 450 keV max energy not sufficient for large structures.
applications. Focal spots may be too large for all applications.
Reliable. Continuous spectrum limits contrast control.
Easy to operate. Limited X-ray photon output due to target heating.
Flexible — multiple power/focal spot/energy/ beam angle
configurations are available.
Some models are portable.

Linear accelerator Excellent test object penetration due to high output and Expensive and large.
high energies (MeV). Facilities requiresignificant shielding.
Continuous spectrum limits contrast control.

Isotopes Veryportable and require no external power. No on/off switch— always energized.
Can be positioned inside hard to reach structures. Large focal spots.
Low output that decreases over time because of half life
of isotope.
Strict control by regulatory agencies.

RADIOLOGIC TESTING 10.7

 blurring) and higher magnification than with of the setup, and that which reaches the imager
conventional radiologic sources. The importance of attenuates in air.
microfocus X-ray sources becomes obvious when Two difficulties with microfocus radiologic
one considers the effect that geometric unsharpness testing are the limitation of the X-ray power and the
has on the radiologic image of small but critical small field of view for the image. To use a very small
discontinuities. Geometric unsharpness, discussed focus point on the target for the X-ray source, the
elsewhere in this chapter, blurs the discontinuity electron intensity for generating the X-rays must be
image and reduces contrast. A microfocus source limited to avoid damage to the target. Microfocus
decreases geometric unsharpness and allows X-ray tubes run at amperages hundreds of times
effective projection radiologic testing. lower than conventional tubes. Also, in projection of
Projection radiologic testing (image the image, the field of view is necessarily reduced
magnification) occurs when the specimen is moved such that many images are needed to cover large test
away from the image plane (film or digital detector). objects. For this reason, microfocus radiologic
Conventional industrial techniques generally use testing is usually performed with digital detector
little or no direct image magnification because the arrays and mechanical manipulators so that many
image unsharpness and potential contrast loss images can be taken quickly.
caused by the large focal spot of the X-ray source
and film graininess or screen mottle compromise the
sensitivity to small details. However, the application
of microfocus X-ray sources to difficult or critical Exposure Factor
industrial inspection problems has shown that much The exposure factor is a quantity that combines
more detail can be imaged by projection milliamperage (X-ray) or source strength (gamma
magnification. Significantly reducing the size of the rays), time, and distance. The exposure factor is the
focal spot in an X-ray tube provides a number of milliamperage times time divided by the distance
advantages. When the radiologic object is positioned squared from an X-ray source at a given kilovoltage.
close to the source and away from the detector plane, Radiologic techniques are sometimes specified in
an enlarged primary image is obtained on the X-ray terms either of kilovoltages and exposure factor or of
imaging system. This large image facilitates the radioactive isotope and exposure factor. Exposure
detection of fine structural detail. It also increases factor can prove to be a very useful parameter. It is
the size of the original detail sometimes in the necessary merely to multiply the exposure factor by
background image noise produced in the imager. the square of the distance to find, say, the
The fine detail makes detection and interpretation milliampere minutes or millicurie hours required. For
easier for the radiographer and reduces fatigue. an X-ray tube, alteration in the filament current (mA)
Geometrical enlargement also increases the produces a direct change in the quantity of radiation
signal-to-noise ratio, which improves image emitted but has no effect on radiation energy.
contrast. By moving the radiologic object away from Furthermore, filament current (mA) and time are
the film, most of the forward scattered radiation usually interchangeable. Many X-ray techniques are
(noise) misses the detector because of the geometry defined by the milliampere second (mA·s) or
milliampere minutes (mA·min) required at a given

Table 7. Standards for focal spot measurement.

Size Range Organization Title

>0.3 mm ASTM International ASTM E 1165, Standard Test Method for Measurement of Focal Spots in
<500 keV Industrial X-Ray Tubesby Pinhole Imaging

>0.2 mm European Committee for Standardization EN 12543-2, Characteristics of Focal Spots in Industrial X-Ray Systems
<200 keV for Use in Non-destructive Testing: Part 2, Pinhole Scanning Method

>0.2 mm European Committee for Standardization EN 12543-3, Characteristics of Focal Spots in Industrial X-Ray Systems
<200 keV for Use in Non-destructive Testing: Part 3, Slit Camera Radiologic Method

>0.2 mm European Committee for Standardization EN 12543-4, Characteristics of Focal Spots in Industrial X-Ray Systems
<500 keV for Use in Non-destructive Testing: Part 4, Edge Method

<0.3 mm Deutsches Institut für Normung (German DIN EN 12543-5, Characteristics of Focal Spots in Industrial X-Ray
<225keV Standards Institute), for the European Systems for Use in Non-destructive Testing: Part 5, Measurement of the
Committee for Standardization Effective Focal Spot Size of Mini and Microfocus X-Ray Tubes

>0.1 mm International Electrotechnical Commission IEC 60336, Medical X-Ray Equipment— X-Ray Tube Assemblies for
<200keV Medical Diagnosis, Characteristic of Focal Spots

10.8 AEROSPACE NONDESTRUCTIVE TESTING

kilovoltage and distance to give the right exposure. It
The product of milliamperage and time is constant
(11) D  log
for the same photographic effect. This is known as Io
the reciprocity law and is valid for X-ray and
gamma exposures over the range of radiation where D is film density, Io is intensity of incident
intensities and exposure times in normal industrial light, and It is intensity of transmitted light when
viewing radiographic film. Typically, radiographs
radiologic testing. The operator may set the X-ray
with a film density greater than 2.0 in the region of
exposure time based on the available current
interest have suitable sensitivity for the inspection.
limitations of the X-ray tube at a particular
Film densities greater than 4.0 are too dark to
kilovoltage setting. Reciprocity law failure can occur
interpret.
at very high dose rates, very long exposures, or with
Alternately, digital radiologic testing is becoming
unusual detector combinations.
more common for aerospace applications. A digital
radiograph or digital image is produced when the
same X-ray transmission process is used with a
Radiologic Detection and digital radiologic testing system such as a digital
detector array or an image intensifier system whose
Imaging output is digitized with a charge coupled device
A radiograph is an image on a recording medium camera. In the twenty-first century, the aerospace
produced by the penetrating radiation passing industry has been converting from film or analog
through the material or object being tested. Figure 8 radioscopic systems to digital testing by using digital
shows a typical radiologic configuration. The detector arrays and computed image processing and
source-to-object distance, object-to-detector storage. A particular advantage of digital over film
distance, and source-to-detector distance are radiologic testing is the elimination of film
important configuration values that affect the processing chemistry and its environmental costs.
exposure times and quality of radiographs. The The digital approach also has advantages in
image at the detector location is a projected map of production feedback and image distribution.
the D  T effect of the penetrating radiation
transmitted through the object.
Radiologic film as the detector records this Figure 8. Radiologic configuration of source, object, and detector.
permanent image. When the film is exposed to X- or
gamma rays, an invisible change called a latent Detector
image is produced in the film emulsion. The areas so Test object
exposed become dark when the film is immersed in a Collimator
developing solution, the degree of darkening X-ray beam
depending upon the amount of the exposure. After
development, the film is rinsed, preferably in a
special bath, to stop the development. The film is
then put into a fixing bath, which dissolves the
unexposed radiation sensitive salt in the emulsion.
The film is washed to remove the fixer and is dried X-ray source
so that it may be handled, interpreted, and filed. The Source-to-object distance
levels of darkening on the film are measured by the Object-to-detector
distance
relationship of X-ray exposure and film density. This
relationship is expressed with a tool called the
characteristic curve (sometimes called the H and
D curve, named for photographic scientists F. Hurter
and V.C. Driffield):

RADIOLOGIC TESTING 10.9

 Radiologic Test Techniques
PART2

Film Radiologic testing features or discontinuities and is discussed elsewhere

in this chapter.
Film radiologic testing entails the generation of With standard film radiologic testing, the X-ray
an X-ray film image when the film or film casette is film may be selected with variables of resolution,
placed directly behind an object and most often in speed, and latitude.
contact with the object. Unlike film projection using
a microfocus source, standard radiologic testing uses 1. Resolution varies because of the silver halide
a geometric magnification factor of 1 whereas particle size, the basis for the imaging. Particle
projection radiologic testing may use a geometric sizes are typically in the range of 0.1 to 10 µm.
magnification from 1.5 to 50. After exposure to photon energy, the particles
In standard radiologic testing, an image with are activated so that they can be developed into
acceptable spatial resolution will be generated for silver grains by the processing chemicals and
most X-ray tubes regardless of focal spot size. The cause film darkening. Larger particles will allow
limiting factor for spatial resolution in the image the latent image to form more quickly and thus
will be the film type and/or the front fluorescent or provide a faster exposure. Smaller grain size will
lead screen that might be used. This method permits require longer exposure but with finer detail
an increase in X-ray output that can be obtained resolution. Figure 9 shows the characteristic
from large focus X-ray tubes (>0.5 mm) or linear curve for some typical X-ray films that plots the
accelerators, which typically have 1 or 2 mm focal film density as a function of the log of the
spot sizes. This may be needed to penetrate large relative exposure. The characteristic curve is
aerospace castings or structures. Geometric useful for calculating exposures needed to adjust
magnification causes an increase in the image film density. The curve generally shows an
unsharpness. Unsharpness degrades the image exponential increase in film density with
quality and ability to detect and quantify small increasing exposure. This characteristic increases
sensitivity to small changes in photon
transmission with darker films.
2. The curves also are used to show the speed of
Figure 9. Characteristic curve for several radiologic film types the film, that is, the amount of radiation needed
(ASNT 2002, 167). with a particular film density. In Figure 9, Film
X is faster than Film Y.
3. The latitude of a radiologic film refers to the
4.0 range of radiation exposure that will result in a
particular range of film density. In Figure 9,
Film Z has wider latitude than Films X and Y.
3.5 Because of the exponential response of the film,
the latitude can be small relative to the range of
transmitted intensities in a complex object. For
3.0 these cases, film radiologic testing will often
require multiple exposures or multiple film loads
(Film X and Film Y) to cover the necessary
2.5 exposure range for suitable inspection of the
object.
Density D

In many cases, radiologic film is used with front

2.0
and sometimes back fluorescent or lead screens. The
screens reduce the exposure time, improving test
speed. They may also reduce low energy scatter
1.5
radiation noise in the image. The fluorescent screens
convert X-ray photons to light, which then exposes
1.0
the film. Typical fluorescent screen materials are
Film Z cadmium tungstate, lanthanum bromide, and
Film X
Film Y gadolinium oxysulphide. The fluorescent screens
0.5
necessarily reduce image resolution due to the grain
size of the fluorescent material. Lead screens convert
X-rays to an electron emission that enhances the
0
film exposure. The high atomic number of the
0 0.5 1.0 1.5 2.0 2.5 3.0 fluorescent or lead screens increases the absorption
of X-rays significantly over the absorption of film
Log relative exposure by itself and explains why they are important for
higher energy (>100 kVp) X-rayexposures.

10.10 AEROSPACE NONDESTRUCTIVE TESTING

Computed Radiologic Testing photomultiplier tube. The analog signal from the
photomultiplier tube is digitized according to a
(Photostimulable Phosphors) lookup table and then turned into a digital
radiograph for display on a computer monitor.
Computed radiologic testing uses a In computed radiologic testing, the actual pixel
photostimulated luminescence phosphor system that size in the digital image (hence the spatial resolution
forms a latent image (similar to film) on a storage of the image) is primarily determined by the
phosphor image plate when exposed to X-rays. scanning laser focal spot size. This is different than
Figure 10 shows in block form an overview of how a film, where the resolution is determined largely by
computed radiologic testing image is generated and the properties of grains in the film and not by the
digitized. The image plate is read out electronically film processor. However, the image plate properties
using a red laser beam in a computed radiologic and composition do affect spatial resolution, and
testing scanner. When stimulated by the red laser manufacturers sell plates that may have different
light, the image plate converts the deposited energy specified resolution. The flexible image plate is a
into blue light, which is collected by a multilayer structure where the phosphor layer is
made of BaFBr:Eu2+. In addition to the phosphor
layer, other layers of material in the plate serve as
Figure 10. Generation of computed radiologic protective surface layers, support material, and light
digital image. shielding/backing layers. The plates come in various
sizes that can fit in rigid casettes or be cut to fit
difficult aerospace structures (like the inside of an
Step 1 engine or airframe).
Image plane is exposed to ionizing radiation; latent With film radiologic testing, the image density is
image is created.
related to the amount of incident radiation in a
logarithmic fashion. In a computed radiologic testing
imaging system, the image density is linearly related
Step 2 to exposure. The phosphor grains in the image plate
Image plane is scanned with red stimulus light; blue store the absorbed incident radiation energy in the
phosphor light is generated from image plane.
form of trapped electrons. Because the phosphor
material has a very large but limited number of
available traps, the number of trapped electrons
Step 3 increases linearly with exposure until all traps have
Blue photostimulable luminescence is collected by been filled.
optics and channeled to photomultiplier tube. The latent image by the electron trapping process
is relatively stable and can be stored for days, the
time depending on the particular screen
manufacturer. Readout of the latent image can be
Step 4 accomplished by laser stimulation at an appropriate
Photomultiplier converts luminescence to analog wavelength, resulting in the emission of visible light
electrical signals in proportion to (luminescence) from the phosphor grains; this
luminescence collected.
process is referred to as photostimulated
luminescence. Again, since the luminescence is
correlated to the latent image, it has the intensity
Step 5 pattern of the original radiation image. In this way,
Analog electrical signals are amplified, filtered, the storage phosphor acts as an energy transducer,
passed through a proportional analog-to-digital converting the radiation pattern to a visible light
converter, and synchronized to pixels correctly pattern. Because the processes are completely
spaced within binary matrix.
reversible, the storage phosphor can undergo a vast
number of cycles without noticeable degradation.
That is, once the initial latent image has been read,
Step 6 any residual latent image can be subsequently erased
Computer algorithms assign each pixel of binary and the imaging screen reused. The number of reuses
data proportional gray scale level via lookup tables
to create original gray scale data matrix (data file). is limited by mechanical wear and tear rather than
by degradation of the phosphor.
Because the inherent resolution of film is greater
than the resolution of the image being captured,
Step 7 radiographers (once geometry and other sources of
Viewing software transforms original grayscale data unsharpness are considered) generally do not
file into original image; original image can be describe film resolution in absolute dimensional
output to monitor or printer; digital process units and almost exclusively focus on contrast alone.
algorithms may be introduced; image is
dispositioned. In the case of phosphor imaging, a physical
resolution limit is encountered that can limit
discontinuity detection regardless of contrast. For
example, consider an image quality indicator
Step 8 (discussed elsewhere) having a hole of a diameter
Originating processedimagewith electronicrecords
are archived to optical or magnetic media.

RADIOLOGIC TESTING 10.11

 twice the thickness (2T) of the image quality characteristics (such as different latitudes), a suitable
indicator, selected to be 2 percent of a 3 mm object. step wedge is also useful in establishing the system
This equates to a 120 µm hole diameter. A phosphor dynamic range during system evaluation, setup, or
system with a pixel size larger than 120 µm will not routine inspection.
be able to resolve the 2T hole even though there is Another aspect of phosphor, differing somewhat
sufficient contrast sensitivity for the 2 percent from film, is its greater sensitivity to scatter
penetrameter. Though the system can detect small radiation. Related to this sensitivity are lead screens,
changes in object contrast or density, it cannot either for intensification or for shielding. In general,
resolve features smaller than its laser spot size. thicker lead or other metal foil can be used to
Therefore, consideration must be given to actual achieve maximum scatter reduction and
pixel size or resolving capability of the system in intensification than is normally needed with film,
combination with the scan speed in order to select especially at energies above 150 keV and for gamma
the correct system operating parameters. sources. Experimentation is required to determine
Even though film density and phosphor density the best combination for particular applications
are different, it is possible to use them the same way. (Hytec 2001).
For example, an acceptable film radiograph may be
defined as having a density of 3.0 in the area of
interest and with minimum and maximum densities
of 2.5 and 3.5. This density is based both on the film Video Radioscopy
properties and on the characteristics of the film Radioscopic testing, or radioscopy, is a technique
viewer or light box, such as the maximum light whereby an image is produced electronically rather
intensity. Similar guidelines can be produced for than on film. There is typically little lag time
phosphor images as well, although the numbers will between the item‘s exposure and the resulting image.
likely be different. If given a choice between a Radioscopy has also been called real time
lighter phosphor image and a darker one, the darker radiography. In most instances, the electronic image
one is preferable, as with film. One key benefit of results from the radiation passing through the object
phosphor over film is that much higher densities can being inspected and interacting with a screen of
easily be captured and displayed whereas film material that fluoresces or gives off light when the
densities are limited by the brightness of the light interaction occurs. The light may be directly
box. captured by a video camera system with lens
Because the storage phosphor process is linearly coupling or through an image intensifier tube. X-ray
related to the amount of incident radiation over a image intensifiers are commercial products designed
much greater practical range than with film, it for real time radiologic test applications (Dolan
exhibits a much wider latitude than film. In many 2002). Some digital detector arrays discussed in the
cases, this correlation can permit either shorter next section are also constructed to operate as real
exposures or a much wider range of thicknesses to time radiologic testing detectors. The image formed
be imaged in a single exposure. This unique is ―positive‖: brighter areas show where higher levels
characteristic curve also makes for much more of transmitted radiation reached the screen. This
forgiving exposures. Because of the steep slope and image is the opposite of the negative image
narrow linear band of film characteristic curves, a produced in film radiologic testing.
relatively slight error in exposure will cause part of The significant differences between film
the image to be in the low contrast area of the curve radiologic testing and video radioscopy testing are
and require another exposure to be made with the speed at which an inspection can be
corrections. With phosphor images, on the other accomplished, the quality of the image, the
hand, considerable variation in exposure is possible equipment used and its related cost, and the means
while maintaining the same contrast. of image analysis and storage. An advantage of
While there are fundamental differences between radioscopic testing is inspection speed. With video
film and phosphor radiologic testing, many of the radioscopy, the image is available almost
practical technique principles can be applied to both. instantaneously as the radiation passes through the
For example, the standard rules regarding the test object. Lag time is less than 1 s and not a
geometric factors in radiologic testing also apply to practical consideration. This feature allows for
both phosphor and film imaging: generally prefer greatly increased inspection speed. For example,
viewing a dark image to viewing a light image; lengths of a welded joint 1 m (3 ft) long can be
select a particular density in the area of interest to inspected in less than 60 s. With sample positioning
judge of proper technique; and use image quality equipment, the test object can be moved — that is,
indicators to help indicate image quality. However, rotated or tilted — so that the inspection area can be
additional parameters normally unmeasured in film shifted and an entire test object inspected in minutes
radiologic testing can be important in phosphor or seconds.
imaging. For example, in the case of phosphor, it has System resolution depends on factors that include
been shown that resolution can impact a wide X-ray source focal spot size, magnification, and
variety of implementation issues, including imaging system performance. Imaging system
throughput and cost. Thus, resolution test patterns performance depends mainly on the quality of the
can be used to evaluate or demonstrate the image intensifier, camera, frame grabber board,
resolution of a system. Similarly, because various monitor, or other components used. The camera bit
phosphor imaging systems exhibit different response depth or dynamic range is important for contrast

10.12 AEROSPACE NONDESTRUCTIVE TESTING

quality in the images. With the large volumes of data
that can be taken with radioscopic testing, data Figure 11. Block diagram of typical digital X-ray imaging system.
storage techniques are also an important
consideration.
In general, radioscopic test systems cannot
resolve a discontinuity as small as film can,
because of limitations of the image intensifier, X-ray to
X-ray source Object electron
fluorescent screen, or camera. However, because convertor
magnification and sample positioning are easy, (direct/indirect)
systems resolution can be significantly improved
with microfocus X-ray tubes. Although with
magnification, the field of view for each inspection
zone will be small, the ability to manipulate the test
object while observing it provides adequate coverage Digital image Conversion to Readout
promptly. The quality of the manipulation equipment display digital image electronics
for video radioscopic systems is as important as the
X-ray imaging characteristics. Radioscopic test
systems are available in a number of configurations
and sizes ranging from tabletop models to units that
fill a large room.

Figure 11 shows a block diagram of a typical

Digital Detector Arrays digital X-ray imaging system. The digital detector
array performs the third, fourth, and fifth steps of
Digital radiologic imaging relies on a digital the flowchart.
detector array to acquire static X-ray images (like A digital detector array consists of a primary
film or computed radiologic testing) and real time X-ray or gamma ray detection medium followed by
imaging (like radioscopic systems). A digital detector a pixelized read structure, with various means of
array is an electronic device that converts ionizing digitizing and transferring the image data. For the
or penetrating radiation into a discrete array of primary X-ray conversion material, there are
analog signals. The signals are subsequently luminescent materials such as scintillators or
digitized and transferred to a computer for display as phosphors, and photoconductive materials also
a digital image corresponding to the radiation known as direct converter semiconductors. For
pattern imparted on the input region of the device. readout structures, the technology consists of charge
These devices can range in speed from many minutes coupled devices, complementary metal oxide silicon
per image to many images per second, up to and in based detectors, amorphous silicon thin film
excess of radioscopic rates (usually 30 frames per transistor diode readout structures, and linear or area
second). crystalline silicon pixel diode structures. Following
Digital detector arrays can be used to enhance the capture of the X-rays and conversion for analog
productivity as well as the quality of nondestructive readout, the signal is typically amplified and
testing. They are used as a diagnostic tool in the digitized. The following steps describe in more detail
manufacturing process, for inline testing on the options for the X-ray capture, converted image
production lines, and inservice as hand held or capture, and digitization steps. An understanding of
scanned devices for field aerospace applications. these is important when selecting a digital detector
They can be applied in industrial computed array.
tomography systems and as part of large robotic
scanning systems for imaging of large or complex
structures including weldments and aerospace
castings. Because of the digital nature of the data, a Step 1: X-Ray Capture
variety of new applications and techniques has X-ray capture in digital detector arrays may take
emerged, enabling quantitative inspection and place with either scintillators or semiconductors.
automatic discontinuity recognition. Digital detector
arrays can also be used to detect various forms of
electromagnetic radiation, or particles, including X- Ray Capture with Scintillators
gamma rays, X-rays, neutrons, or other forms of Scintillators are materials that convert X-ray or
penetrating radiation. gamma ray photons into visible light photons, which
The digital image by its nature will provide are then converted to a digital signal using
numerical results important for engineering and technologies such as amorphous silicon arrays,
manufacturing. There is no standard system to charge coupled devices, or complementary metal
address all applications. There are almost as many oxide silicon devices together with an
choices of detectors as there are ways to configure analog-to-digital converter. Because there are
the overall system. Economics, speed, quality, and various stages of conversion involved in recording
the impact on the overall processes are all keys to the digital image, it is important to retain
designing or selecting a system. information during conversion in the scintillator.

RADIOLOGIC TESTING 10.13

 The properties desirable for ideal scintillators are as the amorphous silicon approach offers excellent
follows: collection efficiency for large areas in a thin,
• high stopping power for X-rays obtained by compact, robust package. However, pixel pitch is
dense materials (avoiding signal loss from typically 100 µm or larger.
scattering in scintillator); Charge coupled devices are small light sensors
• high efficiency of light yield, conversion of with high pixel densities. They use a transparent
X-radiation to light; polysilicon gate structure for reading out the device
• matched emission spectrum of the scintillator to and, because of their high pixel fill factor, are very
the spectral sensitivity of the light collection efficient in collecting the light produced from the
device; phosphor material. A larger field of view can be
• low afterglow during and after termination of obtained with charge coupled devices through a lens
the X-ray irradiation; or a fiber optic transfer device to view a phosphor or
• stable output during long or intense exposure to scintillator screen, but the lens approach has poor
radiation; light collection. Fiber optic image plates have
• temperature independence of light output; and significantly improved light collection efficiency but
• stable mechanical and chemical properties. are expensive and not amenable to large fields of
view. For small fields of view, the directly coupled
These properties permit high efficiency, stability, charge coupled device provides high spatial
and robust operation, yielding ideal imaging resolution and high light collection efficiency.
performance. Two common scintillators are terbium Complementary metal oxide silicon read
activated gadolinium oxisulfide (Gd2O2S:Tb) and structures are based on complementary metal oxide
cesium iodide (CsI). semiconductors — circuits popular for
microprocessors, memories, and application specific
X-Ray Capture with Semiconductors integrated circuits. Complementary metal oxide
silicon technology, leveraging the multibillion dollar
(Photoconductors) semiconductor industry, enables low cost production
A photoconductive material converts X-rays to of pixelized devices. Like charge coupled devices,
electron hole pairs that then get separated by the they are formed with crystalline silicon, but the read
internal bias of the device as defined by the material structure is individually addressed. Unlike charge
properties, such as the manufactured charge coupled devices, where charge is actually transferred
imbalance in the semiconductor material. As with across active pixel regions, complementary metal
scintillating materials, another electronic element is oxide silicon technology has individually addressed
needed to capture the signal produced, such as an pixels. Complementary metal oxide silicon image
electrode structure with pixelization, possibly with sensors draw less power than but produce more
additional added electron bias on one electrode to electronic noise than charge coupled devices. Like
separate the electron hole pairs. But unlike a charge coupled devices, they can couple to various
scintillating material, there is a lower likelihood that scintillators, either directly, or by lens or fiber optics.
the charges produced will have as much lateral Thin silicon transistors have been made
spread as experienced optically in luminescent commercially available in large, flat panel sensors.
materials. Amorphous silicon, through large area silicon
Also, because the photoconductive material deposition and etching, solves the size constraints of
converts the X-ray signal directly into electron hole charge coupled devices and complementary metal
pairs, there is greater conversion efficiency than with oxide silicon devices. Because the phosphor or
the production of light. For X-ray applications, photoconductor layer is typically deposited or
photoconductive materials such as amorphous coupled directly onto the silicon, optical or electron
selenium, cadmium telluride, and mercury iodide transfer is efficient. However, the readout circuitry in
have been used because of their high atomic these devices requires a large pixel space to
numbers and the ability to manufacture these accommodate the thin film transistor, data lines, and
materials into a monolithic structure. Other scan (gate) lines required for operation, thus limiting
photoconductive materials are available or may how small a pixel this device will permit. The
become available in the future. It should be noted amorphous silicon read structure is composed of
that, although light is not generated from these over a million pixels that include photodiodes. The
materials, lag and burn-in can occur because of diode has a sensitivity that peaks in the middle of
subtle effects of sweeping the charge out of the the visible spectrum where a number of good
semiconductor. phosphors emit. The electric charges generated
within every pixel of the photodiode are read by the
active matrix of thin film transistors in place. The
thin film transistor matrix, essentially a matrix of
Step 2: Capture of Converted switches, is scanned progressively. At the end of
Image each data line is a charge integrating amplifier,
which converts the charge packet to a voltage,
The choice of capture technology depends on the followed by a programmable gain stage and an
application. For small fields of view, the directly analog-to-digital converter, which converts the
coupled charge coupled device or complementary voltage to a digital number transferred serially to a
metal oxide silicon approach will provide very high computer, where data are formed into an N  M
spatial resolution (with 10 µm pixels) and efficient pixel image (where M is number of rows and N is
light collection. For large field-of-view applications, number of columns).

10.14 AEROSPACE NONDESTRUCTIVE TESTING

Step 3: Selection of Digital Figure 12. Digital radiologic image quality factors.
Read Structure
Finally, once the image has been captured,
digitization techniques typically convert the analog Digital radiographic image quality
signal into discrete pixel values. For digital detector
arrays, the digitization is typically at 8 bits (256 gray
values), 12 bits (4096 gray values), 14 bits (16 384
gray levels), or 16 bits (65 536 gray values). The Detector properties Technique parameters
higher the bit depth, the more finely the signal is
sampled. The digitization does not necessarily define
the gray level range of the digital detector array. The
useful range of performance is defined by the ability Noise versus dose Energizing current
of the read device to capture a signal in a linear
relation to the signal generated by the primary
conversion device. A wide linear range warrants a Basic spatial resolution Exposure time
high bit depth digitizer. It should be noted that, if
digitization is not high enough to cover the
information content from the read device, Number of frames
Detection efficiency averaged
digitization noise might result. This can be
manifested as posterization: discrete bands of gray
levels in the image. Conversely, if the selected Detector lag Focal spotsize
digitization is significantly higher than the range of
the readout device, then the added sampling may not
improve performance. Second, if the digitization is Magnification and
Internal scatter geometry
completed well beyond the linear range of the read
structure, these added gray levels will not be usable.
For example, 16 bits of digitization do not Bad pixel performance Detector calibration
necessarily indicate 65 536 levels of linear
responsivity.
The useful range of a detector is frequently
defined as the maximum usable level, without
saturation, in relation to the noise floor of the digital
detector array, where again no useful differentiation suggests that providers of digital detector arrays
can be extracted from the data. This range is offer a spider diagram that summarizes the
sometimes referred to as the detector dynamic range. performance of a detector using a numerical scheme
listed in the standard that highlights the strengths or
weaknesses of a digital detector array. The purchaser
Digital Detector Array Selection for can easily review such diagrams and decide what is
most important for the application.
Aerospace Applications
An important prerequisite for a good digital
X-ray detector system is for it to control the entire
imaging chain and to reflect its capability in the Computed Tomography
final image. The technology of image capture, the The term tomography comes from the Greek,
representation of images as digital data, their tomos, meaning a ―cut‖ or ―slice.‖ Computed
processing, the enhancing of data for a specific tomography (CT) uses measurements of X-ray
image display, and the nature of the display transmission from many angles to present the image
technology each contribute to this capability. For in cross sectional maps (tomograms). Computed
image interpretation, the quality of images from the tomography can provide quantitative information
detector is important for the choice of the detector about the density, constituents, or dimensions of the
and system specifications. component imaged. The completed tomographic
The dominant contributions to a digital radiologic image is digital with an image intensity value
image, hence the final image quality, come from two assigned to each pixel of the image. The pixel is
sources: (1) the inherent property of a detector and actually a voxel (volume element) because it
(2) the radiologic technique itself. Some contributing represents the two-dimensional cross section plus a
factors affecting the radiologic image quality are third dimension (depth) defined by the slice
shown in Figure 12. thickness. The medical field uses hounsfield notation
A standardized methodology has been established for the image data:
for evaluating the inherent detector properties as
listed in the previous paragraph and may be found  – w
in ASTM E 2597, section 10.7 (ASTM 2012b). This (12) H  1000
document provides procedures for evaluating and w
recording digital detector array properties by
where H is the hounsfield number (named in honor
providers so that a purchaser may compare devices
of Godfrey Hounsfield, who in 1979 won a Nobel
with standardized criteria. The ASTM standard

RADIOLOGIC TESTING 10.15

 Figure 13. Computed tomographic system configurations for Table 8. Hounsfield units.
generations of scanners: (a) narrow beam, single detector, object
translation and rotation for successive scans; (b) fan beam, multiple Material Hounsfield Value
detectors, object translation and rotation for successive scans; (c) fan
beam, multiple detectors, object rotation; (d) revolving source, ring of
detectors (ASNT 2002, 319). Air –1000
Water 0
(a) Object Acrylic 110
Carbon 580
Source Bone 1000
Detector
Aluminum 1900
Iron 24 000

prize for the development of computed tomography),

 is the measured attenuation number, and w is the
attenuation number for water. Using this scale,
Object Detector water takes on the value of zero, vacuum (or air) is
(b) –1000, and bone is 1000. A change of one integer in
the hounsfield scale represents a 0.1 percent change
in attenuation value. Table 8 shows a numerical
scale for hounsfield units.
Source Because of the large range in numbers for
industrial materials, hounsfield units are not usually
applied in industrial computed tomography.
However, medical scanners can be used for industrial
inspection and, therefore, these units may appear in
aerospace applications.
Computed tomography uses manipulating systems
that obtain X-ray projection from all orientations
about the object. This can be performed in a variety
of ways. Figure 13 shows the generations of
scanners that have been developed for medical and
(c) Detector industrial applications.
1. The first generation scanner of Figure 13a used
a narrow beam with a single detector. The test
Source object was scanned across the beam, then
rotated and scanned back. Although very good
data could be obtained, the process was very
slow.
2. Figure 13b shows the second generation scanner
that uses an X-ray fan beam and multiple
detectors. The object is translated across the
beam and then rotated by the fan angle. This
second generation configuration is the most
common used for large industrial objects.
3. Figure 13c shows the third generation scanner.
(d) The object is of a size that will fit within the fan
beam. In the configuration, the object may be
Detector ring tomographically scanned by rotation — a faster
process than translation and rotation. The object
size can be larger than the fan if it is offset
Object from the center so that at least slightly more
than half the object fits within the fan. As the
object rotates, the portions outside the fan will
rotate through the X-ray beam and can be
Source reconstructed. The rotate configuration is used
for smaller industrial objects.
4. Figure 13d shows the fourth generation of
medical scanner, which uses a ring of detectors
and a rotating tube. This technique provides
very high throughput scanning and can be
applied to some aerospace products, such as
turbine rotor blades.

10.16 AEROSPACE NONDESTRUCTIVE TESTING

 The systems in Figure 13 all take a single slice Computed tomography can also be important in
through the object. Following a scan at one position, manufacturing and process development by
the object must be moved in the axial direction with providing feature and anomaly location,
a step and a scan repeated for full coverage of the configuration control, and the direct measurement of
object. Figure 14 is an example of a turbine blade dimensions for engineering acceptance. Ultimately,
inspection where a digital radiograph and a computed tomography can allow the acceptance of a
tomogram are shown. The radiograph does not show product on the basis of quantitative measurements
the internal cross sectional configuration of the test and engineering criteria. Such an engineering
object because of its complex geometry. The analysis rather than a qualitative inspection standard
tomogram provides the detailed cross section to can reduce scrap and increase component reliability.
determine dimensional tolerances and inspect for Computed tomography provides a means of
anomalies. The tomogram slice shows the wall acquiring geometry (reverse engineering) rather than
thickness of the casting and discontinuities in the digital modeling in three-dimensional computer
cast material if present at the slice plane. aided design. Computed tomography is also effective
Figure 15 shows another computed tomographic during product failure analysis.
configuration that uses a cone beam X-ray field and
an area detector that can provide full volume
computed tomographic data sets very rapidly. The
reconstruction process is complicated and time Figure 14. Test images of aircraft turbine blade: (a) digital radiograph;
consuming, but its speed and data volume can be (b) computed tomography cross section (ASNT 2002, 326).
particularly advantageous. This technique is efficient
for small and medium sized objects. The scanning (a)
manipulation uses the third generation approach;
even when the test object extends outside the cone
beam, however, the reconstructions will contain
useful imaging of regions inside the cone beam. (b)
When computed tomographic data are taken on the
5 mm (0.2 in.)
complete object by using a volumetric data
acquisition and reconstruction, a variety of image
plane displays or surface reconstructions can be
performed on the data as shown in Figure 16 (Hytec
2001). This capability is particularly useful for
engineering development and failure analysis of
electronics or materials.
The basic resolution of a computed tomography
system is determined by the effective beam width of
the X-ray beam in the object. The effective beam is a
function of the source and detector dimensions and
the position of the object with respect to them. The
vertical resolution of the slice volume is determined
by the thickness of the collimation apertures. The
contrast sensitivities of computed tomography
systems are inherently high, often in the range of 0.1
to 1.0 percent. The signal-to-noise ratio improves Figure 15. Conical beam computed tomographic system for volumetric
with increases in beam width, number of views, computed tomographic imaging.
beam intensity, and interrogation time.
Because of the high signal-to-noise ratio in any
voxel, computed tomography can detect features
below the resolution limit of the image. For features
that are larger than a single voxel, the contrast
sensitivity improves by the square root of the
number of pixels making up the feature. For a
feature smaller than a pixel, the apparent density is
averaged over the image pixel and, therefore, the
signal for that image voxel is reduced. This is called
a partial volume effect. Although the signal is Conical X-ray beam
reduced by the partial volume effect, the feature may
still be detected. This point is significant in X-ray source Object
computed tomography: often large image voxels
may detect small features without resolving them.
The partial volume effect is also useful for locating Detector array
edges for dimensional measurements with accuracies
to a tenth of the pixel size.

RADIOLOGIC TESTING 10.17

 Figure 16. Volumetric computed tomographic image processing of battery: (a) digital radiograph; (b) horizontal slices;
(c) vertical slices; (d) surface rendering (ASNT 2002, 327).

(a) (b) (c) (d)

Computed tomographic systems are slow and tomography has been developed for several very
expensive. For routine production quality, the specific circumstances: (1) expensive finished
application of computed tomography depends on the products, such as launch vehicles and ballistic
relation between the object value, computed missiles; (2) critical items in a larger assembly, such
tomographic scanning cost, and the cost of as rocket nozzles or turbine blades; and (3)
alternatives. The more complex and costly an information vital for process control, such as
assembly, the more likely that computed tomography dimensional control and reverse engineering.
can be cost effective. With few exceptions, computed

10.18 AEROSPACE NONDESTRUCTIVE TESTING

 Factors Affecting Image Quality PART3

Image quality, or sensitivity to detail, in for steel, aluminum, and magnesium at 150 kVp. In
radiologic testing is described by the two factors general, a 1 percent change of thickness will produce
shown in Figure 17: contrast and resolution. sufficient density change on film to be visible when
Contrast is composed of the inherent subject contrast viewed on most metal subjects. But with magnesium
and the image system contrast. The inherent subject and lighter materials, it can be difficult to record a
contrast is a function of the penetrating radiation 2 percent thickness change; in thick sections, as the
type, radiation energy, and its attenuation as a curves indicate, less change in transmission for a
function of the material and thickness of the test change in thickness takes place. Object contrast is a
object. The contrast is also affected by the imaging limiting factor in light metals and materials with
system response to the transmitted radiation and both low density and atomic number:
dynamic range over which it is operated. The
resolution is composed of the inherent imaging (13) I  I0 et
system resolution of pixel or grain size for detectors
and the geometric unsharpness. The geometric
unsharpness is affected by the source spot size, where I is the transmitted beam intensity, I0 is the
object size, and the relative distances between the incident radiation beam intensity, t is the material
source, object, and detector. The geometric setup of thickness, and  is the attenuation coefficient.
the X-ray exposure can influence contrast through The subject contrast C will be given by a change
the scatter radiation field. Orientation will also affect in transmitted intensity for a change in thickness:
the ability to detect details. I
(14) C   t
I
Radiologic Contrast
The subject contrast for a given thickness change is
Sensitivity improved by increasing the attenuation coefficient.
This decrease is accomplished by decreasing the
Contrast in a radiograph is the difference in the
X-ray energy.
resultant density produced for a given change of
X-ray or gamma ray absorption. It is affected by
many factors, some of which must be compromised.
The radiologic contrast sensitivity combines the Table 9. Relative penetration of 200 kVp X-rays.
subject contrast and the contrast in the imaging
system. The choice of X-ray equipment is important: Thickness
Material Density (g/cm2)
the higher the kilovoltage selected for the X-ray tube, cm (in.)
the shorter the effective wave length of the generated
radiation and the greater the penetrating power. Lead 11.4 0.16 (1/16)
Table 9 shows typical thicknesses of materials that Copper 8.5 1.2 (0.5)
can be usefully penetrated with 200 kVp X-rays. Steel 7.4 1.9 (0.75)
If the penetrating power of the radiation is great,
each increment of thickness in the object will absorb Titanium 4.1 2.5 (1.0)
proportionally less than if the penetrating power Aluminum 2.7 10.0 (4.0)
were lower. On the other hand, if kilovoltage is low, Magnesium 2 12.5 (5.0)
less radiation will be transmitted through the object.
Each small change in absorption due to thickness of
material will then cause a relatively large change in
transmission. Thus, the lower the voltage, the greater Figure 17. Factors that affect radiologic image quality.
the radiologic contrast. Therefore, kilovoltage may
be lowered to perform an inspection but not be
Radiologic image quality
increased above the level specified without
engineering approval.
Subject or object contrast must also be considered
by the radiographer. At X-ray voltages from 30 kVp
to 5 MVp, aluminum has a lower absorption per unit Contrast Resolution
thickness than steel. Therefore, it takes a greater
thickness change of aluminum to cause a given
change in X-ray transmission than with steel. Hence,
it follows that aluminum has less object contrast
Subject Image Inherent Geometric
than steel. Figure 18 shows graphically the change in
thickness versus the change in transmitted radiation
RADIOLOGIC TESTING 10.19
 For materials of approximately uniform thickness intensities, then lower contrast will be necessary to
where the range of transmitted X-ray intensities is record details throughout the image, probably with
small, the technique producing high contrast will less sensitivity. In cases where an extreme range of
show all portions of the area of interest with an intensities is transmitted, it is necessary to expand
increased radiologic sensitivity. If, however, the the dynamic range of the image system. For film,
imaged test object transmits a wide range of X-ray dynamic range may be obtained by loading the film
holder with two high contrast films of different
speeds. The kilovoltage and exposure are so chosen
that the thick portions of the object are satisfactorily
Figure 18. Radiation transmission versus thickness for steel, aluminum, recorded on the faster film and the thin portions on
and magnesium at 150 kVp. the slower film. Digital radiologic testing systems
usually provide a relatively high dynamic range such
100
that one exposure is sufficient for a significant range
of material thicknesses.
The radiation energy chosen must be compatible
with the attenuation in the object. For low
attenuating objects, low energy radiation produces
final radiologic images with good contrast.
Conversely, for inspection of thick, highly
attenuating objects, the radiation must have
sufficient penetrative capability to produce an image
10
within a reasonable period of time. For high contrast,
96 to 99 percent of the incident radiation should be
Percent transmission

attenuated by the subject. Increasing kilovoltage

reduces contrast because the quantity of radiation at
any given energy increases and because the
proportion of radiation with a short wavelength
(high energy) increases disproportionately. High
Steel Aluminum Magnesium energy radiation can penetrate the object more
readily and thus reduce contrast. Figure 19 shows
1.0
the effect on the final image of low or high contrast.
The right image in Figure 19 shows that, for a given
subject, a doubling of kilovoltage can increase
transmitted radiation 15 to 30 times. This example
shows the disproportionate effect a small kilovoltage
change can have on a particular inspection.

–1.0
1 2 3 4 5 6
Spatial Resolution
The resolving capability of detail in a radiograph
Thickness of material is a function of the resolution of the imaging system
and the geometry of the X-ray inspection setup.
Geometrical relationships affect image sharpness and
Figure 19. Radiologic images of honeycomb showing effect of doubling help control distortion. Also, artifacts in image
kilovoltage on contrast: (a) 15 kVp; (b) 30 kVp. processing affect resolution.

Magnification
To resolve small features in aerospace castings or
welds, geometric magnification may be used in
X-ray imaging either with film, digital radiologic
testing, real time radiologic testing, or computed
radiologic testing. When the radiologic specimen, for
a fixed source-to-detector plane distance, is moved
closer to the source, its image becomes magnified.
The resulting size of the radiologic image varies with
its position relative to the source and the detector.
The factor of the change in size is known as
geometrical image magnification:

M SOD  ODD
(15)  SOD
where ODD is the object-to-detector distance and
SOD is the source-to-object distance. This geometric
effect is shown in Figure 20.

d
 Magnification improves resolution and reveals
details. Unfortunately, as the magnification increases, Figure 20. Geometric magnification.
edge definition may become worse due to geometric
unsharpness caused by a large X-ray focal spot.
Therefore, the focal spot size must be kept small M = (ODD +SOD)/SOD
such that the ensuing geometric unsharpness is kept Detector
lower than any unsharpness contribution from the
image detector. The resolution of the X-ray film or Object
detector/image plate determines the size of the focal Feature
spot below which unsharpness is negligible. magnified on
In some cases, especially digital radiologic testing image plane to
Focal show finedetail
with digital detector arrays, microfocus X-ray may spot f
be required. These sources can easily enlargefeatures
by factors of 20 or more for either film or nonfilm
techniques.

Geometric Image Unsharpness Source-to-object

distance Object-to-detector distance
(Geometry and Source Size)
The term image unsharpness recognizes that there
is always some blurring in the image. Specifically,
geometric image unsharpness Ug is due to the finite
size of the source of radiation and the geometric
magnification. Figure 21 shows how geometric
unsharpness is formed. The distance on the detector Figure 21. Geometric unsharpness.
over which an edge of a feature is spread is known
as the penumbral shadow or the geometrical Ug = ODD  f/SOD Detector
unsharpness. The equation to determine unsharpness
is:
Object
f
(16) Ug = ODD 
SOD Focal spot Ug (geometric
size f unsharpness)
where f is the diameter of the focal spot at its
widest, ODD is the object-to-detector distance, and
SOD is the source-to-object distance.
Because the X-ray source always has a finite size,
geometric unsharpness or image blur will always
occur. The magnitude of geometrical unsharpness is
directly proportional to the X-ray source size. The
only way to reduce the image blur for a fixed Object-to-detector Source-to-object
radiologic setup is to use a machine with a smaller distance distance
focal spot.

Total Image Unsharpness (Including because of the finite focal spot size. With the
Detector Contribution) relatively high resolution capability of film, many
inspections may be performed with no
In addition to geometric unsharpness, total magnification, so only the film‘s resolution
unsharpness of the radiologic image is also affected characteristics need to be considered. If more
by the characteristics of the X-ray film, digital magnification is needed than is allowed, a smaller
detector array, image plate (in computed radiologic focal spot size or another X-ray tube may be
testing), or radioscopic imager. Typically, whichever selected.
unsharpness contribution is greater will control the For nonfilm radiologic techniques (computed
total image unsharpness. In standards for film and radiologic testing and radioscopy), managing the
nonfilm radiologic media are detailed requirements total image unsharpness is especially important as
for the maximum geometric unsharpness as a the detector contribution is often significant and
function of inspected object thickness (ASTM must be considered in addition to the geometric
E 1742, 2012a). These limits are extended in digital unsharpness. In addition, to measure the sizes of
detector arrays and computed radiologic testing to small pores accurately in aerospace structures, one
apply to the total image unsharpness, incorporating must ensure that there are enough image pixels
the effects of the image medium. Historically, this within the indication. This resolution will routinely
was first an issue with florescent screens in film require up to 5 magnification, so a calculation of
casettes, which resulted in unsharpness contributions the geometrical and total unsharpness will be
because of the larger detector (ASTM 1999). important.
Typically, for film, the total unsharpness In any given situation, the geometric unsharpness
requirements only limit the maximum geometrical that can be tolerated sets the lower limit for the
magnification that can be used in a radiologic test adjustable parameters, such as focal spot size,

RADIOLOGIC TESTING 10.21

 source-to-object distance, and object-to-detector
distance. Further demands on sharpness reduce Figure 22. Possible geometric distortions:
intensity. The unsharpness is inversely proportional (a) optimum geometric fidelity condition;
to the source-to-object distance whereas intensity is (b) effect of poor positioning ratio with large
inversely proportional to the square of this distance. penumbral shadow; (c) condition improved by
Thus, the tradeoff of intensity for sharpness is not increasing source-to-object distance; (d) same
equitable. Nonetheless, this uneven exchange is condition achieved by decreasing part thickness
necessary in many cases because it is very important or distance from object to film; (e, f) effect of
to achieve good geometric definition. Shadow geometric misalignment (ASNT 2002, 143).
formation must be watched to ensure satisfactory
sharpness and low image distortion. Distortion
cannot be entirely eliminated: some parts of the test
object will be farther from the film than other parts, (a) (d)
and radiation from any source cannot be made S
ideally parallel. Images will always be imperfect.

Image Distortion S
The central ray of the X-ray source should be
aligned perpendicular to the part being
radiologically tested, and the detector plane should
be placed parallel to the test object. This positioning
projects the image of the test object in the true shape O
of the object. Any deviation from these relative O
positions of source, object, and detector will produce I I
images with some degree of distortion as shown in
Figure 22. Alignment is particularly critical for crack
detection. Because discontinuities revealed in
radiologic images are usually identified by shape,
images free of distortion are very important in
radiologic interpretation. Where complex structures (b) (e) S
are encountered in aircraft inspection, it is often
impossible to locate the test objects in the most
desirable position, and sometimes inspection is
facilitated by planned distortions. Interpretation of S
distorted images is not impossible, but the film
reader needs to visualize mentally the geometry of
exposure and substitute visualization of the distorted
image with the projection of the image by the O
radiation source. This ability requires practice and O
experience.
I I
Placement of Detector
After the detector (film or nonfilm) has been
chosen, the image position in relation to the test
object must be considered. In production radiologic (c) S (f)
testing of small test objects, this is a simple matter of
S
laying the test objects on the imager. With complex
structures, locate the imager so that the plane of the
area of interest and the detector are perpendicular to
the beam to prevent distortion in the final image.
In positioning X-ray film, care should be used to
prevent sharp bends in the film or applying pressures
to the film holder that can produce pressure marks or
crimp marks (artifacts) on the final image. In
radiologic testing of curved surfaces, the source and
film should be positioned, if possible, to take best O O
advantage of the inverse square law and to prevent
as much distortion as possible. Flexible film holders I I
should be used to place the film or photostimulable
phosphors as near as possible to the test surface.

Legend
S = source
O = test object
I = image plane

10.22 AEROSPACE NONDESTRUCTIVE TESTING

Source-to-Detector Distance contributing to image information. Exposure due to
scatter is usually referred to as fog and substantially
The sharpest image would be formed by having a reduces the image contrast. Scatter radiation can
source-to-detector distance (SDD) so great that the have three different sources as shown in Figure 23.
rays would be parallel at the image plane. However, One source of scatter can be the area around the test
because radiation intensity or quantity is diminished or other objects that may be in the radiation beam.
in relationship to the inverse square of the distance, This is usually referred to as reflective scatter. A
the radiation quantity available to expose the second source is scatter radiation from objects
detector would be very small, and exposure times behind the detector. This is usually referred to as
would become impractically high. Therefore, in the backscatter radiation. The third source of scatter
production of the radiologic image, economics and radiation is the test object itself. This scatter can
practicability must be considered. It is recommended obliterate the object‘s edges and is called
that the longest practical source-to-detector distance undercutting. The amount of scatter radiation is
be used for critical exposures to improve image affected by the radiation energy and the atomic
sharpness. If the source-to-detector distance is number of the element doing the scattering. Low
changed, the following formula can be used to atomic number materials scatter radiation more than
correct the exposure. Because an increase in distance materials with high atomic numbers do.
causes a decrease in beam intensity only the
intensity is to be changed, not the kilovoltage, when
correcting for changes in source-to-detector Masking and Filters
distance D. The formula is: Several techniques can be used to reduce the
T2 T1 amount of scatter. Radiologic cones or masks can be
D 2  D 2
(17) made of lead or other highly absorbing materials to
2 1
confine radiation to the area to be exposed. Lead in
many different forms can be placed behind the
where D1 is the original distance (SFD), D2 is the X-ray film and test object to reduce excessive
new distance (SFD), T1 is the original exposure backscatter. Lead foils, about 80 to 100 µm (a few
(MA·s), and T2 is the new exposure (MA·s). thousands of an inch) thick, may be placed between
For example, a technique calls for exposing a test the test object and the X-ray detector to filter some
object at 91 cm (36 in.) using 300 MA·s. However, of the scattered radiation before it reaches the film.
the tube head needs to be moved to make a 1.20 m The foils absorb the lower energy scattered radiation
(48 in.) source-to-detector distance. What would the while letting higher energy radiation form the image.
new exposure be? Substituting the values into the Filters in this position will reduce subject contrast. In
above equation gives T2  1296 = (300)  (2304), or some cases, the scatter can be so great that special
T2 = (300  2304)/1296. The equation solves for the techniques must be applied. Masking the test object
new exposure, T2 = 533 MA·s. is often required because of large variation in
thickness; absorption differences that lead to
Summary: Rules for Radiologic excessive scatter in one region affect image quality
in another.
Geometry
Five general rules can be stated that promote
quality assurance from geometry in standard Figure 23. Sources of scatter radiation.
radiologic testing:
1. The X-rays should proceed from as small a focal
spot as other considerations will allow.
2. The distance between the source and the object
should be as great as practical.
3. The imager should be as close as possible to the
test object unless magnification is necessary for
resolution. Primary radiation
4. The central beam should be as nearly
Wall

perpendicular to the film as possible.

5. As far as the shape will allow, planes within the
specimen plane of interest should be parallel
with the imager.
Reflected
scatter
Forward scatter
Scattered Radiation inside object

Sources of Scatter
Whenever X-rays interact with material, one or Backscatter
more of the following will occur: absorption,
scattering, and penetration. In industrial radiologic
testing, scattered radiation can present a problem
because it can expose the imager without Floor

RADIOLOGIC TESTING 10.23

 Grids radiologic testing have different causes of image
artifacts than film does and can be a challenge to
The control of scatter requires common sense and deal with for aerospace applications. For example,
ingenuity. A concrete, wood, or composite floor will inspection of large castings can lead to the computed
generate enough back scattered radiation to fog a radiologic system operating in a dust filled
film. Film holders should always be laid on or environment, causing scratched plates or specks in
backed with a sheet of 3 mm (0.125 in.) thick lead. the image, and for digital radiologic testing, it can
The backing should be as large as the primary lead to the system being operated in a noncontrolled
radiation field. This thickness of lead is enough for temperature environment that may make bad pixels
radiation generated up to 300 kVp, but where unstable. The following sections address common
fluorescent screens are in use, 6 mm (0.25 in.) should film artifacts, various computed radiologic test
be used. The potter bucky grid is a device especially artifacts, and detail on bad pixel management for
constructed to absorb object scatter radiation (Curry digital detector arrays, which is key for aerospace
1990). Effectively, this grid is like a venetian window applications.
shade; it consists of strips of lead and strips of
material comparatively transparent to radiation. The
strips of lead absorb object scatter radiation at Film Artifacts
angles other than the direct beam. To prevent the The Nondestructive Testing Handbook 4:
lead strips from being revealed in the image, the grid Radiographic Testing (ASNT 2002) gives a very
is moved during exposure so that the image of the thorough review of common film artifacts and false
lead strips is actually distributed over the entire indications that must be understood to accurately
image but does not show detail. These potter bucky interpret film radiographs. Summarized in Table 10
grids are used usually for radiologic testing of low are artifacts that are critical to identify for aerospace
atomic number materials where scatter is a big applications.
problem, especially in medicine.

Computed Radiologic Artifacts

Diffraction
Various test artifacts can affect the final
In crystalline materials, X-ray diffraction can interpretation of the computed radiologic image.
scatter the X-ray beam coherently in preferred Many are specific to discontinuities in the image
directions. For aerospace applications, this is a plate or to mechanical issues with the scanner
common problem with titanium radiologic testing. hardware. By recognizing the cause of the artifact,
The coherently scattered effect can appear as an an inspector can either obtain a new image plate or
indication in the image. The easiest way to address the issues with the scanner. Typical
determine that the source of the indication is computed radiologic artifact types are as follows:
diffraction is to reorient the X-ray source relative to
the test object and repeat the exposure. The 1. Light specks or other light patches may occur
indication will remain at the same point on the test where the image plate has become contaminated
object if it is a real indication. If it is a diffraction with dust or dirt particles (an image plate issue).
effect, it will move or vanish. 2. Image detail may be lost where image plates are
worn or degraded (an image plate issue).
3. Residual images (sometimes called ghost images)
may result from incomplete erasures or damage
Artifacts to the image plate (an image plate issue).
Film images, digital radiologic images produced 4. Gray streaks or patterns of light and dark may
by computed radiologic testing, real time radiologic result from scanner slippage, laser slippage, or
test imagers, and digital detector arrays — all should other mechanical scanner maintenance issues (a
be free from artifacts that might mask or be scanner issue).
confused with discontinuities in the test material. 5. Pressure marks (dark areas on a lighter
However, both digital detector arrays and computed background as described in ASTM 1999) usually

Table 10. Film artifacts critical for aerospace.

Artifacts Cause Aerospace Relevance

Scratches Radiologic film emulsion is quite sensitive and may be scratched by Can be mistaken for cracks.
abrasive material, fingernails, and rough handling during loading or
unloading.
Spotting If fixer solution comes into contact before development, light areas or Can be mistaken for voids in welds, or foreign
spots will appear. If drops of developer or water inadvertently touch the material in honeycomb structures.
film before placement in the developer, dark spots can result.
Fog Fog is an overall, small density increase caused when unexposed film is Reduces contrast sensitivity to low contrast
exposed to some chemicals, low levels of radiation, high humidity, or features such as shims or other thin materials.
darkroom light leaks.

10.24 AEROSPACE NONDESTRUCTIVE TESTING

 result from compression damage of the image organization and for managing new bad pixels after
plate (a scanner issue). the digital detector array is in service.
As can be gathered, many of the image artifacts Single and even some cluster bad pixels can have
result from image plate detector handling, wear and their pixel value restored to an approximate value by
tear of computed radiologic system components, and interpolating that pixel value using the surrounding
electronic overprocessing. They are undesirable and pixel values. A worst case scenario is a true
usually indicate poor housekeeping or other discontinuity obscured by a bad pixel and its
operational issues. However, these types of artifacts neighbors. If a discontinuity size is the size of a pixel
are not prejudicial unless they hide, mask, or can be or smaller, then that pixel pitch shall not be used for
confused with any indication of interest in the that inspection without geometric magnification.
examination. It is not recommended that any artifact Because there is always some blur, on a
or other digital condition be removed from any pixel-by-pixel basis, any discontinuity information
image through digital manipulation or digital in the bad pixel‘s area gets spilled over into
restoration. Only original digital images shall be neighboring pixels. This effectively makes a potential
submitted to qualified personnel for review. discontinuity larger, thus easier to see if a
discontinuity happens to be in that area.
Interpolation on bad pixels does not impact the
Real Time Radiologic Artifacts performance of neighboring good pixels; it simply
Real time imaging systems are subject to artifacts restores the bad pixel‘s estimated pixel value.
and distortions because of flaws in the Therefore, the interpolation process does not hide a
X-ray-to-light conversion phosphor, the discontinuity but in fact may accentuate one because
photocathode, electron optics, and scintillator output it restores the signal to that pixel, showing features
in the image intensifier or camera system. Artifacts at a reasonable estimation of their true size.
can also form due to overexposure damage over time. Another scenario is that of a large,
These artifacts will display themselves similarly to noncorrectable cluster (called a cluster kernel pixel)
the bad digital pixel management described below of bad pixels that might be the same size as the
when a still image is captured from the system. The discontinuity indication. Noncorrectable pixels are
lens systems used in radioscopic imagers are also usually clustered pixels that do not have enough
subject to vignetting which can distort the image. good neighboring pixels to fully restore information.
Usually with radioscopy, the object is in motion such Consequently, these clusters remain bad and will
that the features of interest move across the screen. likely remain either completely dark or completely
The artifact remains at a constant location on the white, depending on their nature during the service
screen, but the features are moving and are thus seen life of the digital detector array. This is a situation
over many locations in the imaging system. where a bad cluster might hide all or a part of an
indication. The greater the number of these bad
clusters in a digital detector array, the higher the risk
Digital Radiologic Bad Pixel of missing a discontinuity because of an overlap of
Management the cluster with an indication. The best way to
manage this is for the user, in coordination with the
Any pixel of a digital detector array that has a responsible engineering organization or Level III, to
performance outside the specification range is called select a digital detector array with a specified limited
a bad pixel. Commercially available digital detector number of these bad clusters in the region of the
arrays usually have bad pixels that must be managed detector used for interpretation. If a group of cluster
to take advantages of these arrays‘ superior image kernel pixels are outside the region where
quality. Part of this management is the need to interpretation is done, then those cluster kernel
develop requirements on (1) allowable bad pixel sizes pixels might be acceptable in practice.
and (2) total number of bad pixels, both of which are Alternatively, if the technique allows geometric
based on the application. Also prescribed is a magnification to reduce the effective size of cluster
retesting for pixel characterization because bad kernel pixels so that they do not interfere with
pixels can change during image acquisition and interpretation, this might allow the use of the digital
processing. detector array. Performing this magnification
Bad pixels are classified into several categories: compensation should not alter other properties such
isolated bad pixels, a cluster of bad pixels, or a line as geometric unsharpness, properties that might
of bad pixels. Clusters are further divided into impair the inspection.
relevant and irrelevant clusters. Irrelevant clusters In either scenario, it is important that the user be
are not correctible; their cluster kernel pixel lacks at provided with a means to track the number and
least five good neighboring pixels. A complete location of all bad pixels, including cluster kernel
definition of the different types of bad pixels is pixels. This allows a ready reference to differentiate
found in the manufacturer qualification standard bad pixels from genuine indications. The risk of false
ASTM E 2597-07e1, Standard Practice for positives due to these uncorrectable clusters is
Manufacturing Characterization of Digital Detector usually low, as the user will have a record of where
Arrays. the bad pixels and cluster kernel pixels are located.
Most digital detector arrays have some bad pixels. As stated above, the manufacturer delivers a bad
Techniques to identify bad pixels by the pixel map with every digital detector array, so it is
manufacturer are found in ASTM E 2597-e1. As of easy to compare the map to the image to determine
2013, standards have been in development by ASTM if the anomaly is an indication or a bad pixel or
International for identifying bad pixels at the user cluster. If so agreed between the user and the
manufacturer, the bad cluster can be marked as such,

RADIOLOGIC TESTING 10.25

 either by color or other means. It should be noted image. The radiograph was then modified by using
that, once an uncorrectable cluster is identified, the the randomly created bad pixel map and corrected
region and its surrounding two-pixel wide perimeter using a bad pixel correction algorithm. Figure 24
is not to be used for interpretation, unless also shows the modified image. The bad pixel
magnification techniques effectively reduce the corrected image looks like the original image and
cluster size. does not interfere with salient features.
Solitary bad pixels and even correctible clusters Because solitary bad pixels and small correctable
will not create false information in the radiograph bad clusters do not necessarily impact interpretation,
after a correction for bad pixels. As an example, these pixels can be interpolated. Most manufacturers
Figure 24 shows a simulated radiologic image of a will provide this capability in the acquisition or
20 mm iron plate, with several 0.2 mm (1 percent) analysis software; by agreement the contracting
shims placed on it. There are holes in each shim, parties (the cognizant engineering and user
with diameters of 0.4, 0.8, 1.2, 1.6, 2.0, 2.2, and organizations) use interpolation for the application
2.4 mm. The pixel pitch used here is 0.2 mm. Bad at hand. It is important that a map of the bad pixels
pixels were randomly created using a computer in a given digital detector array be provided. The
program but with controlled numbers. Cluster user organization can use this map as a reference to
formations were also allowed and embedded in the confirm that an anomaly is in fact a bad pixel. The
same map can also be used to track the formation of
new bad pixels, or the development of bad clusters,
including uncorrectable and cluster kernel pixels.
Figure 24. Bad pixel corruption of signal and recovery using correction. Provided the digital detector array bad pixels and/or
This example has 1 percent of digital detector array pixels as bad pixels. clusters are significantly smaller than the
discontinuities of interest, the operator may choose
Original image Bad pixel image not to interpolate individual bad pixels and small
clusters, as this would not impact interpretation.
Regardless of whether interpolation is done, each
bad pixel is identified through tests recommended in
ASTM E 2597-e1 and flagged as a bad pixel recorded
to a bad pixel map or image.
Digital detector array manufacturers, with the aid
of ASTM E 2597-07e1, are publishing bad pixel
results for different models of digital detector arrays.
This is the average prevalence and range of the
different types of bad pixels as listed in ASTM
E 2597-07e1 for any given model. In most
circumstances, an individual instrument will operate
within the range of bad pixels (clusters and lines).
An important aspect of managing bad pixels is to
Modified image Corrected image select the digital detector array considering these
statistics and, in particular, the prevalence of cluster
kernel pixels. This is one of the factors among all of
the detector properties that need to be considered in
detector selection. The selection of a make and
model from a manufacturer must also include an
evaluation of the bad pixel data of that model.
As with specification of other properties, the
responsible engineering organization sets the
discontinuity requirements required for the
inspection. The nondestructive test engineer in the
user organization must consider the aspect ratio of
the discontinuity, whether a crack, porosity, or
inclusion.

Figure 25. Number of effective pixels to cover discontinuity based on feature contrast as well as noise in digital detector array.

• High signal-to-noise ratio Effective number of pixels covering • Medium signal-to-noise

digitaldetectorarray longest dimension of discontinuity ratio digital detectorarray
• Highcontrast objects • Lower contrast objects

1 pixel 2 to 3 pixels 4 to 6 pixels >6 pixels

High risk, not Moderate risk, acceptable with high Low risk Best practice, if
recommended signal-to-noise ratio and large contrast available

10.26 AEROSPACE NONDESTRUCTIVE TESTING

 Figure 25 summarizes pixel coverage alternatives. is a very short crack or a crack between two adjacent
Discontinuity pixel coverage affects the bad pixel fasteners, but even here the edge changes or
requirements in relation to the effective size of the becomes jagged under magnification.
discontinuity. Pixel coverage can be adjusted by Obtaining parallelism between the X-ray beam
geometric magnification or by modifying the pitch. and the crack plane is difficult. Cracks do not always
For a given size and shape of discontinuity, the user initiate at the expected origin and often are not
develops a desired spatial resolution and perpendicular to the test surface. When the X-ray
unsharpness that has a corresponding pixel pitch. beam passes through a crack at any angle other than
For detection and for managing bad pixels, the directly along the crack plane, both the width of the
longest dimension of a discontinuity should be crack and the intersect angle determine the density
covered by at least three pixels. Single-pixel change and indication contrast. As the angle
coverage of the longest dimension of a discontinuity between the X-ray beam and crack plane increases,
is not recommended for purposes of detection. An both film density change and contrast decrease. The
indication may look like a bad pixel and be missed. film indication becomes broad and more diffuse until
The pixel coverage of the array is also affected by it vanishes into the background.
the array‘s signal-to-noise ratio and the Detection of cracks depends on crack width,
discontinuity‘s contrast, size, and aspect ratio. depth, total metal thickness, and angle of intersection.
When only the intersection angle varies, crack
detection becomes a matter of statistics or
Orientation probability. Table 11 reflects the probability of
detecting a crack at various intersect angles. The
Radiologic testing reliably detects cracks if
stringent criteria are met. It is very easy to produce
an apparently high quality radiograph that does not Table 11. Correlation of beam divergence with
show an existing crack or with a crack indication so
crack detectability.
faint it can barely be seen. The resolution of cracks
depends on total density change and the contrast of Beam to Crack Probability of Crack
an indication against the background of the object.
The human eye can detect density changes of 0.02 in
Angle Detection
the characteristic curve, but to detect cracks, a (degree) (percent)
density change of 0.05 in the characteristic curve is
more practical. 0 96
There are several factors that produce density 3 89
changes on X-ray film. The primary factor in the 6 82
case of crack detection is a change in thickness or 9 75
mass between the crack and test object. A general
15 48
rule is that the crack must be at least 2 percent of
the test object thickness if it is to produce a readable 21 30
indication. This rule has variables that influence film 27 23
density changes, and in some cases as little as a 45 4
1 percent thickness change will produce a visible
indication. In other instances, a crack exceeding
5 percent of the test object thickness may not
produce a readable density change. Regardless of Figure 26. Density changes due to different crack widths and
total density change across an indication, if contrast intersection angles: (a) narrow versus wide crack indication; (b) similar
is not high, crack indications can be missed. width and depth cracks at two different angles of X-ray beam to crack
When an X-ray tube focal spot is centered plane.
directly over a crack whose depth is parallel to the
beam (X-ray beam and crack plane coincide), the (a)
film density change will be a function of the ratio of
crack depth to the metal thickness. Indications of D D
narrow cracks with parallel sides will appear as fine
lines with high contrast. The lines are dark on film
and white in digital radiologic images. Wide cracks
with sloping sides will result in broader indications
Indication width
of lower contrast. Figure 26 illustrates the changes in
film density between cracks of varying widths when 30°
(b)
the X-ray tube is centered over the crack origin. The 60°
stress on a test object will affect crack width. For
example, compressive stress in the lower wing D
surface of an aircraft on the ground tends to reduce D
crack width. The compressive stress is due to the
weight of the structure, engines, ordnance pylons,
and so forth. Jacking the aircraft to place the lower
wing surface in neutral stress or in tension is
Indication width
frequently done to enhance detection of small cracks.
Cracks and their indications tend to curve or deviate Legend
from a straight line. An apparent exception D = effective depth or maximum density change

RADIOLOGIC TESTING 10.27

 table indicates that the probability of detecting a be detected even outside the 16 cm (6.3 in.) radius.
crack with an intersect angle of 9 degrees is This calculation emphasizes the need for information
75 percent, or three out of four. Conversely, the on crack location and orientation before developing
chances of missing a crack with a 9 degree intersect an X-ray procedure, plus the requirement for
angle is 25 percent, or one out of four. When accurate tube head alignment during equipment
developing X-ray procedures to be used for detection setup.
of cracks, the maximum angle of intersection is
5 degrees, which corresponds to an 85 percent
probability of detection. The preferred limit is
5 degrees, which corresponds to a 90 percent Figure 27. Radiation cone radii at various
detection probability. intersect angles and focus-to-detector distances.
A 2.5 or 5 degree intersect angle X-ray beam will
Focal spot
not project over the surface of a 36  43 cm (14 
17 in.) film plate at normal focus-to-film distances.
The entire film will be exposed, but only a small
cone of radiation will be within the desired or
allowable intersect angle limits. Figure 27 reflects the
radiation cone coverage at various intersect angles Intersect angle
and focus-to-detector distances or source-to-object
distances. When developing procedures, Table 12 can Focus-to-detector
distance
be used to determine the necessary ratio of
focus-to-detector distance to source-to-object
distance. For example, a focus-to-detector distance
of 1.8 m (72 in.) is required to be within the 5 degree
intersect angle limit, 16 cm (6.3 in.) on either side of Cone of
the aiming point. Cracks occurring farther than radiation
16 cm (6.3 in.) from the aiming point will produce
indications with reduced film density change and
contrast, and there is a greater chance of not Aiming point Radius
detecting them. Cracks with radial orientation toward
the aiming point (or center of the circular beam) can

Table 12. Radiation cone radii at various intersection angles and focus-to-detector distances.

Focus to Detector Distance Radius of Cone, cm (in.)

cm (in.) 2.5 degrees 5.0 degrees 7.5 degrees 10.0 degrees
90 (36) 3.99 (1.57) 8.00 (3.15) 12.04 (4.74) 16.13 (6.35)
120 (48) 5.33 (2.10) 10.67 (4.20) 16.05 (6.32) 21.49 (8.46)
150 (60) 6.66 (2.62) 13.33 (5.25) 20.07 (7.90) 26.87 (10.58)
180 (72) 7.98 (3.14) 16.00 (6.30) 24.08 (9.48) 32.26 (12.70)
210 (84) 9.32 (3.67) 18.69 (7.35) 28.09 (11.06) 37.59 (14.8)

10.28 AEROSPACE NONDESTRUCTIVE TESTING

 Sensitivity Measurement PART4

Image Quality Indicators Detail Sensitivity

There is a need to be able to quantitatively define Detail sensitivity of the radiologic image is
how sensitive a radiologic image is. The devices that revealed by the capability of visualizing the image
achieve this are image quality indicators (IQIs), also quality indicator holes. When the 2 percent image
known as penetrameters. Table 13 summarizes many quality indicator is used on the test object, it is
of the common standard image quality indicators. usually required that the 2T image quality indicator
They provide an indication of what the image reader hole be visible, where T is the thickness of the image
can expect to see in a film or digital image of the quality indicator. If the 2T hole can be seen, the
actual test object. The image quality indicator is image is said to have 2 percent radiologic sensitivity.
often specified in the film technique and is placed in The film reader can then assume the capability of
every exposure. If the image quality indicator is not seeing any discontinuity that represents a 2 percent
identified in the technique, consult with the dimensional change of the object total thickness. The
appropriate NDT Level III for instructions. Image 1T hole does not represent 1 percent image
quality indicators are not generally required for sensitivity because the thickness of the image quality
inspection for the presence of debris, for proper indicator has not been reduced to 1 percent of the
assembly of components, for dissimilar material test object thickness. Calculations reveal that
inspection, for honeycomb structures, or for the visibility of the 1T hole in a 2 percent image quality
presence or absence of material. Figure 28 shows a indicator actually indicates 1.4 percent image
radiologic image of a penetrameter used in sensitivity. Resolution of the holes is a combined
radiologic testing. measure of image sharpness and contrast and is thus
a measure of image quality.
The regular and expected outline of a hole is
more readily seen than a crack line. The image
Contrast Sensitivity quality indicator should not be placed over an area
The image quality indicator material thickness is of interest, because the image quality indicator or
added to the thickness of the test object. This lead identification numbers may hide discontinuities.
increase in thickness causes more radiation to be In some cases, the image quality indicator cannot be
attenuated, and the image quality indicator outline is placed on the actual test specimen. In these instances,
seen on the final image as a less dense area in the it is acceptable to place the image quality indicator
imager. Due to the additional radiation absorption, on a separate block of the same material and of the
this change serves as a measure of image contrast. same thickness as the specimen. It should be
The human eye is normally used as a detector in remembered that the purpose of the image quality
reading film radiologic images. The eye responds to indicator is to reveal the image quality to the film
differences in the quantity of light being transmitted reader. Therefore, it should be placed in the least
through the film due to the density differences. It is advantageous position. However, the density should
usually assumed that, under practical industrial film not vary more than +30 percent, or -15 percent,
inspection conditions, the eye can detect film density from the area of interest.
differences of 0.02 or more in the characteristic The plaque image quality indicators suffer from a
curve, which corresponds to a light transmission number of disadvantages, the most serious of which
difference of 4.72 percent. Because density is the minimum thickness of 13 µm (0.005 in.).
differences of 0.02 are barely discernible, good MIL-STD-00453 provides additional information on
practice is to strive for a film density difference of image quality indicators. Effective radiologic
0.08 to ensure good visualization of discontinuities. inspection requires techniques that have optimum
For digital imaging displays, the image contrast is geometry, imager choice, contrast, and density.
converted to digital numerical values within the
dynamic range of the imaging system. In an 8, 10,
12 or 16 bit imaging system the dynamic range is Figure 28. Radiologic image of image quality indicator.
1 to 256, 1 to 1024, 1 to 4096 and 1 to 65 000,
respectively. The image display can stretch the
contrast to make it visible to the human viewer
provided that the contrast signal is greater that the
noise. Normally the contrast should be at 2.5 times
the noise to be readily detected.

RADIOLOGIC TESTING 10.29

 Table 13. Table of common standard image quality indicators.

Standard Type Comments

ASTM E 747 wire imagequality Measures bothspatial resolutionandcontrast

indicator sensitivity.
For use with 0.67 to 15 cm (0.25 to 6 in.) thick
materials
Six wire thicknesses.
Wires of titanium, aluminum, magnesium, steel,
bronze, nickel alloy, or copper.
Identify the smallest wire visible.
Typically correlated to hole type sensitivity using table
in standard.

ASTM E 1025 hole type image Measures bothspatial resolutionandcontrast

quality indicator sensitivity.
Material thickness of 1 or 2 percent.
Three hole sizes.
Titanium, aluminum, magnesium, steel, bronze, nickel
alloys, copper.
Identify the smallest hole visible.
Sensitivity quantified by percent thickness and
smallest hole visible, such as 2-2T.
These image quality indicators are specified with
minimum thickness and hole diameter, so they may
not actually be true 2 percent thickness on very thin
materials.

ASTM E 1647 contrast sensitivity Measures only contrast sensitivity.

gage Four levels of contrast: 1, 2, 3, and 4 percent.
Made from same material and thickness as test object.
Should be used with E 2002 or EN 462-5 duplex wire
image quality indicator to determine image quality
level.
Difficult to make for thin materials.

Measures only spatial resolution by using wire pairs.

EN 462-5, or duplexwireimage Thirteen wire pairs (platinum/tungsten) embedded in
ASTM E 2002 quality indicator plastic.
Determine degreeofresolution by distinguishingwire
pairs.
Expensive.

Similar to ASTM E 1025.

ASTM E 1742 hole type image Based on MIL-STD-453.
quality indicator

Similar to ASTM E 1025 except it has true 2 percent

True 2 Pratt & Whitney™ thickness (even for thin materials).
image quality TAMimagequality These image quality indicators use a minimum hole
indicator indicator diameter and so do not measure true spatial
performance on thin materials.
Almost like foil, the true 2 percent image quality
indicator is fragile because thin.

10.30 AEROSPACE NONDESTRUCTIVE TESTING

.
 Radiologic Test Interpretation PART5

General Film Interpretation

To obtain the greatest value from radiologic The inspector reading radiographs should be
testing, characteristics of the image must be acquainted with the exposure technique used,
understood and properly applied. It is possible to material radiographed, conditions of processing, and
make false inferences from radiologic testing, the geometry of the exposure setup. In this way, the
judgments that could result in improper disposition inspector can judge more accurately the images
of the test object. It is the duty of the radiographer produced and interpret the discontinuities more
to continually guard against this possibility. accurately. To determine whether the test object is
Interpretation is affected by process characteristics rejectable or acceptable, the inspector will generally
reflected in the image. To effectively evaluate consult with the structural or design engineer unless
images, it is essential for the interpreter to be standards have been established. To properly apply
familiar with the test objects, their application, information obtained through radiologic testing, the
dimensions, and material. The interpreter should be material inspected must also be accurately identified
thoroughly familiar with the technique that was used with respect to the object radiographed. Lead letters
to produce the image, how the image was processed, or numbers are radiographed along with the object
the codes and standards that apply, the usage of the in order to provide identification. This identification
test object assembly, and the acceptance criteria. must be recorded and keyed to the object.
The wide variety of test objects and of their For film radiologic interpretation, it is important
fabrication complicates radiologic interpretation. that film be reviewed in a proper darkroom with a
Radiologic testing is conducted to ensure that a proper intensity light box viewer. The film reviewer
material or test object has the required integrity to must be suitably adapted to the dark in order to see
reliably perform the function for which it was indications (ASTM E 94, 2010). Practitioners of film
designed. The effects of discontinuities or radiologic testing know the importance of dark
manufacturing deviations must be correlated with adaption for proper film interpretation.
the function of the test object. Specifications spell
out the discontinuities that may be considered
detrimental to the function of the test object and the
acceptable magnitudes of the discontinuities. Digital Image Interpretation
It is the duty of the image interpreter to recognize Digital images from computed radiologic testing,
the various discontinuities, assess their magnitudes, real time radiologic testing, or digital detector arrays
and relate them to the specification required. The offer additional interpretation opportunities beyond
responsibility and capability of the radiologic the issues on film interpretation. A number of factors
interpreter cannot be overemphasized. Often many can adversely affect image quality. Image processing
human lives and investments of millions of dollars and enhancement can be applied to digital images to
depend on the judgment of the radiologic interpreter. improve data usefulness. Techniques include
Any information that can be of assistance in making convolution edge detection, mathematics, filters,
a judgment concerning discontinuities should be trend removal, and image analysis. Table 14 lists
fully used. digital image processing and interpretation
Interpretation of the shadow images visible in the techniques provided through software.
image is an acquired skill, and there is no substitute
for experience. Experience aids the image interpreter
in recognizing discontinuities and in identifying
where they can be expected to occur in a particular
component or structure. Mistakes in radiologic
interpretation are most often a result of misreading
artifacts. There are a number of density patterns that
resemble welding and casting discontinuities that are
often unjustified causes for rejection.

RADIOLOGIC TESTING 10.31

 Table 14. Digital image processing techniques.

Technique Comment

Enhancement programs Histogramequalization redistributesintensitiesofimageoverentire rangeofpossible intensities (usually

make information more 256 graylevels).
visible. Unsharp masking subtracts smoothed image from original image to emphasize intensity changes.

Convolution programs are Highpass filter emphasizes regions with rapid intensity changes.
typically 3 x 3 masks Lowpass filter smoothesimages, blursregionswithrapidchanges.
operating on pixel
neighborhoods.

Math process programs Add images process adds two images together, pixel by pixel.
perform variety of functions. Subtract images process subtracts second image from first image, pixel by pixel.
Exponential process or logarithm process raises e to power of pixel intensity or takes log of pixel intensity,
nonlinearly accentuating or diminishingintensityvariation overtheimage.
Scaler add, subtract, multiply, or divide process applies the same constant values as specified by the user
to all pixels, one at a time. The process scales pixel intensities uniformly or nonuniformly.
Dilation morphological operationexpandsbrightregionsofimage.
Erosionmorphological operationshrinksbrightregionsofimage.

Noise filters decrease noise Adaptive smoothing filter sets pixel intensity to value somewhere between original value and mean value
by diminishing statistical corrected by degree of noisiness. Good for decreasing statistical, especially single-dependent noise.
deviations. Median filter sets pixel intensity equal to median intensity of pixels in neighborhood and is excellent for
eliminating intensity spikes.
Sigma filter sets pixel intensity equal to mean of intensities in neighborhood within two of the mean.
Good filter for signal independentnoise.

Trend removal programs Row-to-column fit fits image intensity along row or column by polynomial and subtracts fit from data.
remove intensity trends This trend removal technique chooses row or column according to direction that has the least abrupt
varying slowly over the changes.
image.

Edge detection programs The first difference subtracts intensities of adjacent pixels and emphasizes noise as well as desired
sharpen intensity transition changes.
regions. The sobel operator weighs inner pixels twice as heavily as corner values. This three-by-three mask
calculates intensity differences.
Morphological edge detection finds the difference between dilated (expanded) and eroded (shrunken)
version of image.

Image analysis programs Gray scale mapping alters mapping of intensity of pixels in file to intensity displayed on computer screen.
extract information from Line trace plots intensity versus position for horizontal, vertical, or arbitrary direction and lists intensity
image. versus pixel location from any point along slice.
Histogramrepresentsnumberofpixelsofgiven intensityvalueas area proportional to that number.
Histograms are typically plotted as bar charts.
Image extraction extracts portion or all of image and creates new image of area selected.
Images statistics program calculates maximum, minimum, average, standard deviation, variance, median,
and mean square intensities of image data.

10.32 AEROSPACE NONDESTRUCTIVE TESTING

d
 Specialized and Emerging PART6
Radiation Techniques

Laminography approximate formula, and (3) the double marker

approximate formula. The reverse geometry X-ray
Computed tomography allows imaging of object source described below can be readily used in stereo
slices with great detail, but with the restriction that radiologic testing by positioning two detectors at the
the objects have to be penetrated from the full appropriate separation distance.
angular region about the object (that is, from
360 degrees of radial directions). Because of high
absorption or limited access, this is not always
possible, as in the case of flat components such as Reverse Geometry Radiologic
multilayer printed circuit boards or welding seams in
large components. Laminography is an alternative Testing
technique that can be used to focus at a particular The main difference between conventional X-ray
layer or depth in an object to improve the detection and reverse geometry radiologic testing is in the
sensitivity to features or discontinuities otherwise X-ray source and receiver (Figure 30). Conventional
hidden by intervening structure in a single radiologic testing uses a point source, whereas the
radiograph. Figure 29 shows the principle of reverse geometry X-ray technique uses a scanned
laminography where multiple views are taken of an source near the specimen. Further, these techniques
object. By proper positioning of the imaging, a use a point detector, as opposed to an area detector
particular plane of interest can be brought into focus such as X-ray film. A large portion of the X-ray
while other planes are blurred. Laminography can be scatter (noise) bypasses the small, distant point
performed by using relative mechanical motion of sensors, thus improving the signal-to-noise ratio.
the X-ray source, the detector, and the object in such Reverse geometry X-ray has several possible
a way that a single plane stays in focus. Such advantages over conventional radiologic testing.
systems have been used medically but have not been First, the X-ray detector can be miniaturized and
particularly advantageous in industry. easily positioned inside a complex structure (such as
a wing), enabling images of each surface of the
structure to be obtained separately. Second, multiple
detectors enable simultaneous data from several
Stereo Radiologic Testing
Stereo radiologic testing is a radiologic technique
using two separate radiologic images with a source Figure 29. Digital laminographic method where 1 to n images are taken
shift parallel to the test surface, or imaging plane. with various orientations of source objects and detector, allowing
The movement of the source between exposures is
multiple focus planes in test object to be reconstructed from data set
about the same as the distance between the human
eyes. When the processed images are viewed, the (ASNT 2002, 305).
right eye sees one image and the left eye sees the
other. The brain combines the images, giving the
impression of a three-dimensional image. View 1 View 2 View n
The two main purposes of stereo radiologic Source
imaging are to provide depth information and to aid
in the interpretation of complex radiologic images.
Parallax is used to determine the depth of an
indication. These techniques are based on the
principle that, from two exposures made with
Feature
different positions of the X-ray tube, the depth of
the discontinuity is computed from the shift of the
shadow of the discontinuity.
The shadows of objects closest to the source will
have the largest shadow projection. An object close Example plane of reconstruction
to the background does not appear to change
position whereas an object farther from the Object
background appears to shift a moderate amount. The
Detector
amount of left or right movement of the projected
shadows is directly proportional to the distance of Image 1 feature location
the object from the imager. A similar triangulation Image 2 feature location
relationship is the basis for most of the calculations Image n feature location
from parallax. Radiologic parallax techniques use Reconstructed image aligns
three variations of the similar triangles relationship: features at plane reconstruction
(1) the rigid formula, (2) the single marker

RADIOLOGIC TESTING 10.33

 perspectives without moving the structure or the small details. The speed and resolution of backscatter
measurement system. Data from several perspectives imaging is controlled by the scatter intensity from
provide a means for locating discontinuities and the volume of interest. To improve quality, the
enhancing separation of features at the surface from radiation intensity needs to be high and the scan
those inside the structure. Finally, the amount of times slow. For more overview inspections, the
secondary scattered radiation contributing to the systems can be run faster at reduced resolution and
noise in the radiograph is reduced. However, the sensitivity.
X-ray source is subject to limitations in X-ray Backscatter radiologic testing can be performed
energy, reliability, operations, and cost. in many combinations of collimated or uncollimated
source and detector configurations (Bossi 1988). A
typical X-ray backscatter technique is the flying spot
approach: indexing a pencil beam of radiation over
Backscatter Radiologic an object and mapping the measured backscatter
Testing signal to the beam position (Shedlock 2005). The
image formed is the volume backscatter. Adding
Backscatter radiologic testing involves the single- collimation to the detectors makes it possible to
sided collection of scattered radiation rather than the measure the backscatter intensity at a point for
transmitted radiation to form an image. In radiologic depth profiling. By combining depth and area
testing, only a small fraction, <5 percent, of the imaging, three-dimensional tomographic information
initial photons will be used to create the image. Most is obtained (Black 1995).
of the radiation is scattered. The backscatter X-ray backscatter imaging has proven useful in
radiation can be detected and used to identify checking of fuses and detonators. The fuses are made
subsurface material or geometric variations. of organic materials and show little contrast to the
Backscatter imaging can be implemented by a heavy cases in transmission radiologic testing. Being
variety of techniques for depth profiling and lighter organic materials, however, fuses scatter
imaging (ASNT 2002, 379-401). Several applications radiation well and absorb little. Aircraft corrosion
of backscatter radiologic testing include aircraft within lap joints causes swelling between the rivets,
foreign object detection, corrosion in aircraft leading to a phenomenon called pillowing.
structures, and inspection foam insulation such as in Exfoliation corrosion causes the metal itself to swell
the Space Shuttle exterior tank. It has also been used and in extreme cases to burst. These effects are
for ordnance inspection (Costello 1980). Backscatter detectable by backscatter imaging. Figure 31 is an
X-ray imaging has been successful for cargo, example of a backscatter image using a flying spot
vehicle, and baggage inspection. X-ray backscatter is technique, a one-sided approach looking behind the
often slower and uses more expensive equipment aircraft skin for internal systems assembly and
than other nondestructive test methods sensitive to checking for foreign objects (Shedlock 2010).

Figure 30. Reverse geometry X-ray technique measures features

without mechanical motion. Flash Radiologic Testing
Flash radiologic testing is a special type of
radiologic testing used to produce a single stop
Detector motion image or a series of sequential images of
Raster scanning
X-ray tube high speed dynamic events. The stop motion
requirement places an upper limit on the exposure
Scatter time. This time limit in turn depends on the velocity
of the test object. Flash radiologic testing can be
used for some types of high speed process
Detector
monitoring, ballistics, and explosive studies
Electron (ASNT 1976).
beam

Neutron Radiologic Testing

Primary
beam Neutron radiologic testing is similar in principle
to X-ray radiologic testing, and the information
supplied is complementary in nature (Berger 1975).
Object The interactions of X-rays and neutrons with matter
Detector are fundamentally different, however, forming the
Slices basis of many unique applications for neutron
imaging. X-rays interact with the electron cloud
surrounding the nucleus of an atom, but neutrons
interact with the nucleus itself. In general, as the
Computer atomic number increases, X-ray attenuation
increases for lower X-ray energies and/or higher
density materials. There is no such rule, however, for
the attenuation of thermal neutrons. Some light
elements (such as hydrogen and boron) have high
thermal neutron attenuation coefficients whereas

10.34 AEROSPACE NONDESTRUCTIVE TESTING

.
some heavier elements (such as lead, copper, and
iron) have smaller attenuation coefficients Figure 31. Backscatter image of fuselage mockup pane.
(Figure 32). This allows for some unique applications
such as imaging water level in a metal container or
detection of corrosion (heavy in oxygen and
hydrogen) in a metal assembly. Materials containing
hydrogen or boron can be delineated from other
elements in many cases with neutron radiologic
testing where X-rays are inadequate. It is thus
possible to expose a specimen (such as an explosive
or fuel, or even water, sealed in a metallic container)
to a beam of thermal neutrons and to project an
image having excellent resolution and contrast
distinguishing between the charge material and its
container. The same applies to plastics used as liners
or spacers in metallic aerospace joints. Neutron
radiologic testing has been used for the detection of
water in honeycomb structure inspection. Although Figure 32. Neutron attenuation in some materials.
it is excellent for detection, it is more expensive to
perform than alternatives. Explosive bolt inspections
for spacecraft and military aircraft are cases where Lead
the inspection is worth the expense. A complete
diagram comparing X-ray and neutron attenuation Copper
for all elements is shown in Figure 33.
Neutron radiologic testing has generally been Iron
limited by the operational costs of the source from
the standpoints of both safety and adequate source Polyethylene
strength. The neutron generating source can be a
reactor, the target of an accelerator, or a neutron
emitting isotope. The test specimen is placed in a Water
collimated beam of neutrons in a geometry like that
for X-rays. The neutrons are usually thermalized and Oil
interact with attenuation characteristics of Figure 32.
But other energies, such as fast, epithermal or cold 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
neutrons, can be used for some particular Attenuation coefficient (cm2/g)
penetration or contrast needs. The transmitted
neutrons are collected at the imaging plane. The
image is created by exposing high resolution film by
means of a neutron to light or the electron
Figure 33. Neutron and X-ray attenuation of elements.
conversion process. Gadolinium metal or gadolinium
oxysulphide phosphors are examples of typical
neutron converter screens. Gadolinium has a 1000
particularly high neutron absorption cross section.
Mass attenuation coefficient (cm2/g)

The converted signal can be imaged by film, other

100
radiologic imagers, or digital detector arrays.
10

1.0

0.1

0.01
0 10 20 30 40 50 60 70 80 90

Element number (Z )

Legend
= X-ray 100 keV
= X-ray 200 keV
= thermal neutrons

RADIOLOGIC TESTING 10.35

PART7 Radiologic Testing Standards

General standards are prepared by an engineering weldment generally acceptable for normal use.
society, company, or government agency as a guide Radiologic standards prepared by ASTM
in determining if castings and weldments are sound. International are approved for use. Table 15 lists
These standards are based on experience and ASTM standards dealing with radioscopic and
engineering judgment to provide a casting and radiologic testing of castings and weldments.

Table 15. Radiologic testing standards.

Category Standard

General ASTM E 747-04(2010), Standard Test Method for Controlling Quality of Radiographic Examination
Using Wire Penetrameters.
ASTM E 748-02(2008), Standard Practices for Thermal Neutron Radiography of Materials.
ASTM E 1000-98(2009), Standard Guide for Radioscopy.
ASTM E 1025-11, Standard Practice for Design, Manufacture, and Material Grouping
Classification of Hole-Type Image Quality Indictors (IQI) Used for Radiography.
ASTM E 1254-08, Standard Guide for the Storage of Radiographs and Unexposed Industrial
Radiographic Film.
ASTM E 1316-13c, Standard Terminology for Nondestructive Examinations.
ASTM E 1390-12, Standard Specification for Illuminators Used for Viewing Industrial
Radiographs.
ASTM E 1416, Standard Test Method for Radioscopic Examination of Weldments.
ASTM E 1441-11, Standard Guide for Computed Tomography (CT) Imaging.
Reference Radiographs ASTM E 155-10, Standard Reference Radiographs for Inspection of Aluminum and Magnesium
Castings.
ASTM E 192-13, Standard Reference Radiographs of Investment Steel Castings for Aerospace
Applications.
ASTM E 1320-10, Standard Reference Radiographs for Titanium Castings.
Digital Reference Images ASTM E 2422-11, Standard Digital Reference Images for Inspection of Aluminum Castings.
ASTM E 2660-11, Standard Digital Reference Images for Investment Steel Casting for Aerospace
Applications.
ASTM E 2669-11, Standard Digital Reference Images for Titanium Castings.
Computed Radiologic Testing ASTM E 2007-10, Standard Guide for Computed Radiography.
ASTM E 2033-99(2013), Standard Practice for Computed Radiology (Photostimulable Luminescence
Method).
ASTM E 2445-99(2009), Standard Practice for Qualification and Long-Term Stability of Computed
Radiology Systems.
ASTM E 2446-05(2010), Standard Practice for Classification of Computed Radiology Systems.
Digital Radiography with ASTM E 2597-07e1, Standard Practice for the Manufacturing Characterization of Digital Detector
Digital Detector Arrays Arrays.
ASTM E 2698-10, Standard Practice for Radiological Examination Using Digital Detector Arrays.
ASTM E 2736-10, Standard Guide for Digital Detector Array Radiology.
ASTM E 2737-10, Standard Practice for Digital Detector Array Performance Evaluation and Long-
Term Stability.

10.36 AEROSPACE NONDESTRUCTIVE TESTING

.
 Applications of Radiologic Testing PART8

Applications of radiologic testing to the prevalent in light alloy castings, consist of tiny voids
aerospace industry are many and various. Radiologic scattered through part or all of a casting. Gas
techniques can be applied to research, development, porosity and shrinkage porosity in aluminum alloys
manufacturing, and inservice environments. Only a are examples of dispersed discontinuities. In
few examples are discussed below. radiologic images of sections more than 13 mm
(0.5 in.) thick, it is difficult to distinguish images
corresponding to the individual voids. Instead,
Advantages and Limitations dispersed discontinuities may appear on film
deceptively as mottling, dark streaks or other
Radiologic testing has advantages and limitations irregularities.
as listed in Table 16. In general, castings are irregular in shape and
vary in thickness, so technique and equipment
should be able to penetrate the thicknesses and
What Can Go Wrong material of the test object with adequate latitude and
Table 17 lists various problems in obtaining sensitivity.
useful radiologic images and suggests solutions to Correct radiologic procedure selects the lowest
apply. voltage that will do the job in a reasonable exposure
time. Where many castings are examined, a
convenient technique is to establish a reasonable
Castings exposure time and select the voltage required for the
particular thickness to be tested. Good practice
The forming of metal to various shapes by normally requires that exposures be longer than 60 s.
pouring molten ferrous and nonferrous alloys into When castings with great differences in thickness
molds accounts for much of the critical components must be tested in one exposure, an increase in
of aircraft. Most castings that require X-radiologic voltage will provide wider latitude, as well as shorter
testing are made of light metals such as aluminum exposure time, but reduce contrast. If other factors
and magnesium alloys. Difficulties inherent in this remain constant, the most desirable combinations of
manufacturing process plague the foundry. Because voltage and exposure time for a specific test object
a metal occupies a larger volume when molten or are governed largely by sensitivity.
hot than at room temperature, rapid cooling can
cause voids, shrinkage cavities, hot cracks or metal
rupture. The molten metal also traps considerable
gases from the air. These may result in tiny, regularly
Welds
shaped bubbles in the solid casting. Some molten Radiologic testing is used in aerospace for welded
metals, such as aluminum, accumulate gas on the products. The most efficient systems use robots and
surface. This may be trapped in the casting if digital imagers. The system in Figure 34 is used for
adequate precautions are not taken to prevent welded ducts in commercial aircraft. The
trapping the gas in the mold. In addition, mold
materials may wash from the walls of the mold into
the casting, forming inclusions that reduce the Figure 34. Automated digital radiologic test cell for weld ducts.
casting‘s strength.
Radiologic testing is ideally suited to the Digital detector
inspection of castings because the most common X-ray
Ducting pipe with welds sources
casting discontinuities are three-dimensional and so
almost independent of inspection angle. Exceptions
include fine cracks, cold shuts, unfused chills, and
chaplets. To reveal these, the radiation must be at or
near the same parallel plane as the discontinuity.
Hairline surface cracks, such as those made by
grinding, are seldom revealed by radiologic testing.
It is possible in most cases to identify the
radiologic images of discontinuities typical in
castings. This information is valuable to the foundry Manipulator
in developing procedures necessary to meet a
standard of quality. Although the discontinuities
commonly encountered in aluminum and
magnesium castings are similar to those in ferrous
metals, a group of irregularities called dispersed
discontinuities often occur. These anomalies,

RADIOLOGIC TESTING 10.37

 Table 16. Radiologic testing advantages and limitations.

Technique Advantages Limitations

Film Is proven technology. Filmexposuretimes and processingturnaround arelong.

Provides permanent record. Requires extensive experience and training of personnel.
Can be used to examine many different shapes and Filmimagedegradesovertime.
sizes. Film storage and retrieval are inefficient.
Is very sensitive, if discontinuity reduces cross section Difficult to merge image with other inspection data.
thickness enough. Silver in film and in processing chemicals is hazardous
Permits characterization of discontinuity. waste.
Is volumetric. Limited use of information.

Eliminates film and chemicals. Complete screen erasure is necessary.

Reduced exposure doses. Computed radiologic plate reader may not be portable.
Increases information capture. Image plates will have scratch induced artifacts over
Phosphor imaging Reusable screens. lifetime.
(computed Screens can be cut for many different sizes/shapes.
radiologic testing)
Eliminates film and chemicals. Detectors are expensive
Reduces exposure doses. Must compensate for limited spatial resolution by
Increases information capture. geometric magnification and possibly microfocus
Best image quality of digital methods. source.
Digital imaging Operating costs lower than with film. Detectors require special handling (compared to film).
(digital detector Very high signal-to-noise ratio. Detectors are rigid.
arrays) Large dynamic range covers large range of object
thicknesses.

Less image blur. Low power output of source.

High image magnification. Extended image acquisition times.

Rapid testing. Low brightness level.

Imaging of dynamic events. Operator dependent visual acuity.
Robotic manipulation provides freedom to view details
Microfocus
or move to other orientations.

Volumetric feature detection. Expensive.

Radioscopic
Three-dimensional digital data. Large system not suitable to inservice inspection.
Effective measurement tool for engineering and Conventional computed tomography requires taking
manufacturing development. X-ray measurements from all directions.

Provides depth information. Background intensity reduces the contrast resolution.

Computed Provides images of object slices. Mechanical scanning is complicated.
tomography Images sharply only one slice at a time.

Images from one side. Limited spatial resolution.

Can image a volume rather than a plane. Limited depth penetration.
Can detect the difference between filled and unfilled Overly sensitive to materials of low atomic number.
Laminography voids.
Can perform limited chemical analysis.

Provides time resolution or stop motion. Image contrast is more limited than in conventional
radiologic testing.
Backscatter Low intensity per pulse needs fast timing of sensor and
radiologic testing screen, reducing sharpness and noise.
Large focal spot limits sharpness.

Provides depth information. Not always possible to have the sensor in contact with
Aids interpretation of complex images. test object.
Certainorientation and discontinuitygeometriescan
Flash radiologic cause measurement errors.
testing
Higher absorption by hydrogen and someother Requires conversion screens.
elements allows unique applications. Requires nuclear reactor, special accelerator, or
radioactive isotope.

Stereo radiologic
testing
10.38 AEROSPACE NONDESTRUCTIVE TESTING
 manipulator positions the complex duct shape in the traditional film radiologic testing. With the
proper orientation for radiologic testing, using automated system, the inspection time is about 30 to
geometric magnification (about 4.5) to gain 60 s per weld at 100 percent coverage and 1-1T
resolution sensitivity. The system can be image sensitivity. The automated digital radiologic
programmed to go through a number of prescribed test system improves product quality through rapid
steps for radiologic views that optimize the feedback to the welder. For the product shown in
inspection. Because of the complex shape, the Figure 34, the rejection rate was reduced fivefold
inspection would be very time consuming using through process feedback.

Table 17. What can go wrong with radiologic testing.

Issues Options

Insufficient resolution Use smaller focal spot.

Increase source-to-object distance.
Decrease object-to-image distance.
Select finer grain film or higher resolution imager.
Use microfocus X-ray tube and magnification.
Check whether contrast limitations are affecting theresolution, usecollimation, and reduce
scatter.
Reorienttest object to optimize discontinuity location image.
Use edge enhancement.

Insufficient contrast Reduce X-ray energy.

Increase numberofphotonsused to create image.
Use highercontrast film.
Use imager with less noise.
Useimagerwithgreater dynamic range.
Reduce scatter.
Perform inspection away from other structure or walls.
Use filtration on X-ray beam or between object and imager.
Collimate beam to region of interest.
Increase object-to-detector distance.
Change orientation.
Increase integration or averaging fordigitalsystems.
Use contrast enhancement.
Use computed tomography.

Insufficient latitude Select high latitude film.

Use multiple film loads.
Use high dynamic range digital imagers.
Increase the X-ray energy.

Exposure takes too long Select faster film or digital imager.

Increase X-ray current.
Increase X-ray energy.
Reduce source-to-detector distance.
Use radioscopic or digital imaging system.
Reduce filtration.

Artifacts, nonuniformity Check film processing chemistry.

Check darkroom light.
Clean surfaceofcassette ordetector.
Check X-rayspotuniformity.
Check X-ray tube heel and toe effects over the field of view.
Reorient test object and reshoot.
Check digital detector array bad pixel file.
Check digital detector array calibration table.
Check scatter radiation sources.

Can’t determine discontinuity location Use stereo radiologic testing, laminography, or computed tomography.

Access to only one side or limited Try X-ray backscatter.

access Try reverse geometry X-ray.

RADIOLOGIC TESTING 10.39

 Honeycomb Figure 35. Honeycomb inspection for F-15
A typical condition that occurs in honeycomb tactical fighter aircraft: (a) configuration;
structures is the accumulation of water in the core. (b) system under control surface; (c) image of
Entrapped water freezes and expands at high honeycomb.
altitudes. The expansion distorts the cells and can
break the bonds between the core and the face sheets.
When this condition exists, the face sheet can vibrate, (a)
causing adjacent bonds to fail and that bond failure
to propagate. Entrapped water also causes corrosion
of both face sheet and core material in metallic
honeycomb structures. As a maintenance inspection,
radiologic testing is conducted to evaluate core
damage and water content. Entrapped water in
honeycomb cells usually appears as a smooth,
consistent area with light image density, not grainy
or porous.
1. Epoxy in honeycomb cells appears grainy,
heterogeneous. If the cell is not spotty and (b)
completely filled, the epoxy will be around each
cell‘s periphery.
2. Radiologic inspection for moisture detection can
be made with the honeycomb core cell walls in
either the vertical or horizontal plane. The
preferred method is with the core cell walls in
the horizontal plane because core cells that are
partly filled with moisture are more readily
identifiable (less easily confused with solid
adhesive).
3. If practical, cells partially filled with water can
be confirmed by repeating the radiologic (c)
procedure from the opposite direction.
4. Radiologic exposures indicating filled core cells
are not always conclusive for moisture detection
and should be confirmed by other means if
possible.

Honeycomb Case History

All flight control structures, including the vertical
and horizontal stabilizers on the United States Air
Force F-15 aircraft fleet are inspected for moisture
entrapment in the honeycomb cells for flight safety
purposes. One location that has performed these tests following film review. Continuous operation with the
is the Warner Robins Air Logistics Center, Warner digital system did not require this setup.
Robins, GA. At one time, the inspectors had to climb Furthermore, if other exposures, other angles, or
onto the aircraft, place films, and then position an stimuli such as vibrations to accentuate the water
X-ray source below the structure, followed by entrapment were wanted, the interrogations could be
commanding an exposure safely from a shielded completed live by using the sensitive detector array.
control room. Ergonomic and safety improvements were achieved
A scanning system has been installed with a with this system by eliminating hundreds of trips per
digital radiologic detector for moisture detection. airplane up and down ladders by inspectors who
Figure 35 shows the scanning systems and one might fall. Lastly was a 95 percent reduction in film
image of minute levels of moisture. and chemicals. Cost was reduced by the medium
In the system, a continuously scanning itself, chemicals, management of the waste streams,
manipulator moves an X-ray source and the digital and archiving of the film.
detector operating at 30 frames per second over the Because hydrogen and boron atoms greatly
monolithic structures with articulation to optimize attenuate thermal neutrons, neutron radiographic
the transmission through the cells without testing reveals adhesives, corrosion, or moisture in
overlaying the cell walls. The application has been composite structures. In Figure 36, specimens have
economical. Compared to prior film techniques, there an aluminum honeycomb core with adhesively
was a 90 percent reduction in inspection cycle time. bonded skins of graphite epoxy, boron epoxy, and
Much of this inspection cycle time was realized fiberglass.
because the results are both instantaneous and
continuous. Film required each setup (source
position) to remain in place for subsequent
exposures if needed until that area was cleared

10.40 AEROSPACE NONDESTRUCTIVE TESTING

Corrosion used for composites details. The ability of computed
tomographic images to show internal material
Corrosion may occur in aircraft materials, which variations is particularly advantageous. Figure 38
reduces strength and expedites possible failure. This shows a computed tomographic image of a
deterioration of the metal may be due to electrolytic composite J stiffener, where the variations in the
action, moisture, chemicals, or gases that attack the
metals; intergranular action due to improper heat
treatment at the time of manufacture; or other Figure 36. Two images of honeycomb structure: (a) using X-rays;
factors. This condition usually occurs on internal (b) using neutrons.
surfaces of such components as tubular supports or
housings. Because corrosion represents a change of
material and occurs in all directions, it is easily
(a) (b)
detected by a proper radiologic exposure. If
corrosion has proceeded enough to provide an
indication, the support is weakened and may fail.
1 1

Cracks
Cracks and cracklike discontinuities are found in
diverse components or structures and are very 2 2
dangerous. They are particularly dangerous when
structures are subjected to vibration or fatigue
loading because cracklike discontinuities propagate.

75 mm (3.0 in.)
Cracklike discontinuities appear in a radiograph as
very straight and sharply outlined dark or black lines
and sometimes as diffused, jagged lines. In some 3 3
cases, they have a dendritic (treelike) pattern. Scatter
radiation from the sides of a crack can amplify the
crack indication in a test image. This is the most
125 mm (5.0in.)
difficult service failure to detect by radiologic testing
Legend
because these crack separations are usually not 1. Graphite.
associated with other detectable conditions that give 2. Boron.
clues to their presence. A cracklike discontinuity 3. Glass reinforced polymer.
oriented at any angle other than 90 degrees to the
X-ray film and not parallel with the X-ray beam
offers very little difference to the radiation
transmission and may not be visible in the radiologic Figure 37. Turbine blade X-ray images: (a) image of full blade;
image. Radiologic testing can only be depended (b) zoom image of crack (ASNT 2002, 406).
upon to reveal cracklike discontinuities aligned
within about 7 degrees of the X-ray beam. This (a)
depends upon the thickness and width of the crack.
Normally, cracks easily detectable by X-ray are
visible to the naked eye. Radiologic testing may be
used to determine the extent of cracks or other
conditions that might have been detected by visual,
magnetic particle, or liquid penetrant methods. In
castings, cracklike discontinuities can be from
shrinkage, hot tears, cold shuts, or other mechanisms
of the casting process. In weldments, cracks may be
longitudinal or transverse. Lack of weld penetration
produces a cracklike discontinuity. The forging
process can introduce cracks, laps, and seams that
appear cracklike in radiologic images.
An example of crack detection with high (b)
sensitivity is the application of a microfocus X-ray
and zoom imaging on a jet engine turbine blade
(Figure 37). The entire blade (Figure 37a) displays no
obvious discontinuities, but on close examination at
about 12 magnification (Figure 37b), a small crack
at the trailing edge of the blade is visible.
Crack

Composites
Radiologic testing is applied to composite
structures for detection of variations in resin and
fiber content and also when composites are joined
with other materials. Computed tomography can be

RADIOLOGIC TESTING 10.41

 Figure 38. Computed tomographic image of Geometry
graphite epoxy woven J stiffener showing ply Radiologic testing is an excellent means to locate
condition and consolidation (ASNT 2002, 326). and evaluate foreign objects. Foreign objects may be
free rivets, bolts, or other objects that could be
detrimental to the function of the component or
assembly.
Radiologic testing has found wide use in the
evaluation of various assemblies to determine status
or condition.
Radiologic inspections, after completion of repair,
ensure quality of workmanship. On occasion,
components are misassembled. In some areas, it is
not possible to check dimensions by gage or visual
means. Radiologic testing may be used if precautions
ensure geometry and internal spacing, such as gaps.

13 mm (0.5 in.)

Electronics
Radiologic testing of electronic boards is best
accomplished by using projection microfocus
radiologic testing with an electron focal spot smaller
than 0.1 mm (0.004 in). A useful technique with
microfocus testing is zooming, dynamically
positioning the object with a manipulator between
the X-ray tube and the detector. Magnification is
achieved when the object is moved away from the
detector and toward the X-ray tube. Figure 39
illustrates a single integrated circuit that was
consolidation and the ply layups can be evaluated, initially situated for low magnification of 2. The
particularly at the T junctions. Computed integrated circuit was then zoomed toward the X-ray
tomography can image wavy or crimped fibers after tube through 10, 50, and 250 magnification. The
curing; they are difficult to detect with other higher the magnification, the more detail one can see
methods. on the video monitor. Zooming is not possible with
conventional X-ray tubes because of the penumbral
effect of larger focal spots.
Wear
Rivets and bolts may wear the skin, spar, and Figure 39. Microfocus radiologic testing of
frame holes so that there is not a correct fit in the electronics using X-rays (ASNT 2002,
holes for adequate strength in joints or attachments 561 and 405).
of a wing section. This wear can be due to continued
flexing of components or due to stress from
turbulent weather or an adverse landing. This
condition may also result in radial cracks from bolt
holes. Bolt hole failure is extremely difficult to detect
by radiologic testing: any angle of exposure results
in superimposition of the bolt or nut over the crack.
Loose bolts and rivets have been detected
satisfactorily, but elongation of rivet holes caused by
bearing failure or sheared rivets should not be
confused with elongation of holes from drilling. If
fatigue has damaged a riveted joint, half moon
indications would all be on the same side of the rivet
and the rivets in the joint should show similar
indications of failure. Intermittent indications would
normally be considered within the tolerance of their
fabrication.

10.42 AEROSPACE NONDESTRUCTIVE TESTING

References
ASNT. 1976. Proceedings of the Flash Radiography Symposium [Houston, Dolan, K.W., and J.J. Haskins. 2002. Ch. 10, ―Radioscopy‖: Part 4,
TX]. Columbus, OH: American Society for Nondestructive Testing. ―Imaging Systems.‖ Nondestructive Testing Handbook 4:
ASNT. 1985. Nondestructive Testing Handbook 3: Radiography and Radiographic Testing, 3rd edition. Columbus, OH: American Society
Radiation Testing, 2nd edition. Columbus, OH: American Society for Nondestructive Testing, 265-274.
for Nondestructive Testing. Eastman Kodak. 1980. Radiography in Modern Industry, 4th edition.
ASNT. 2002. Nondestructive Testing Handbook 4: Radiographic Testing, Rochester, NY: Eastman Kodak Company.
3rd edition. Columbus, OH: American Society for Nondestructive HEW. 1970. Radiological Health Handbook. Washington, DC: United
Testing. States Government Printing Office, for United States Department of
ASTM. 1999. E 1000-98, Standard Guide for Radioscopy. West Health, Education and Welfare.
Conshohocken, PA: ASTM International. Hytec. 2001. Technical Report HT 107990-0003, Small Flashlight
ASTM. 2010. E 94, Standard Practice for Radiographic Examination. Results. Los Alamos, NM: Hytec Incorporated (February).
West Conshohocken, PA: ASTM International. Lawson, L.R. 2002. Ch. 14, ―Backscatter Imaging.‖ Nondestructive
ASTM. 2012a. E 1742M-12, Standard Practice for Radiographic Testing Handbook 4: Radiographic Testing, 3rd edition. Columbus,
Examination. West Conshohocken, PA: ASTM International. OH: American Society for Nondestructive Testing, 380-401.
ASTM. 2012b. E 2597-07e1, Standard Practice for the Manufacturing Sentinel. 2009. Product Catalog. Burlington, MA: QSA Global.
Characterization of Digital Detector Arrays. West Conshohocken, Shedlock, D., B. Addicott, E. Dugan, and A. Jacobs. 2005. ―Optimization
PA: ASTM International. of an RSD X-Ray Backscatter System for Detecting Defects in the
Berger, H. 1975. Practical Applications of Neutron Radiography and Space Shuttle External Tank Thermal Foam Insulation.‖ SPIE
Gaging. West Conshohocken, PA: ASTM International. Proceedings 5923. Penetrating Radiation Systems and Applications
Black, G. 1995. Final Report, WL-TR-93-4015, Advanced Development of VII. Bellingham, WA: SPIE.
Backscatter Imaging Tomography. Wright Patterson Air Force Base, Shedlock, D., T. Edwards, and C. Toh. 2011. ―X-Ray Backscatter Imaging
OH: Air Force Wright Laboratory. for Aerospace Applications.‖ Review of Progress in Quantitative
Bossi, R., K. Friddell, and J. Nelson. 1988. ―Backscatter X-Ray Imaging,‖ Nondestructive Evaluation 30 [San Diego, July, 2010]. New York,
Materials Evaluation 46 (11): 1462-1467. NY: Plenum.
Costello, D., J. Dickerson, J. John, M. Marlowe, L. Parks, C. Preskitt, Shleien, B., L. Slaback, and B. Birky. 1998. Handbook of Health Physics
N. Rubinstein, D. Smith, A. Trip, and H. Weber. 1980. ―Automated and Radiological Health. Baltimore, MD: Lippencott Williams and
Inspection Device for Explosive Charge in Shell — AIDECS‖ Wilkins.
[abstract]. Materials Evaluation 38(9): conference program, 23. Ter-Pogossian, M., M. Phelps, E. Hoffman, and N.A. Mullani. 1975. ―A
Curry, T., J. Dowdy, and R. Murry. 1990. Christensen’s Physics of Positron-Emission Transaxial Tomograph for Nuclear Imaging
Diagnostic Radiology, 4th edition. Philadelphia, PA: Lippincott (PETT)‖. Radiology 114 (1): 89-98. Oak Brook, IL: Radiological
Williams and Wilkins. Society of North America.

RADIOLOGIC TESTING 10.43

 X
11CHAPTER

Thermographic
Testing

Contents
Part 1. Thermographic
Principles, 11.2
Part 2. Thermographic
Instrumentation, 11.7
Part 3. Interpretation and
Analysis of Thermographic
Results, 11.11
Part 4. Emerging Thermographic
Techniques, 11.14
Part 5. Thermographic Test
Standards, 11.16
Part 6. Application Examples, 11.17
Part 7. References, 11.21

Contributors
Steven M. Shepard
Xavier P. V. Maldague
11.1
 Thermographic Principles
PART1

Thermography offers a means of either Analysis

single-sided orthrough-transmission noncontact
inspection of flat or curved areas on metal, The subsurface state of a test object is inferred
composite, polymer, or ceramic structures (ASNT fromthe infrared surface temperature map as the test
2001; Ibarra-Castanedo 2007; Milne 1985; Vavilov object responds to thermal stimulus. Large or severe
1999). Typical aerospace applications include discontinuities may be viewed directly in the
detection ofwater in honeycomb core sandwich infrared image sequence. However, signal or image
structures, location of foreign objects in composite processing is typically applied to detect small or
layups, detection of impact damage and more subtle discontinuities, or to perform
delamination in composites, and measurement of quantitative analysis.
coatingand wall thickness in turbine blades. In response to the applied excitation, the test
Although there are many approaches to object returns to equilibrium as thermal energy
thermographic nondestructive testing, all its generated at the surface diffuses into the bulk.
techniques involve atleastthree distinct operations: Thermal diffusion can be obstructed by thermal
stimulation, temperature measurement, and analysis property variations due to either designed structure
(Figure 1). (such as a wall, coating layer, spar, doubler, or core)
or discontinuities (such as delaminations, inclusions,
porosity, or water) that affect the local surface‘s
Stimulation return to equilibrium, indicated by anomalous
A test object is heated (or cooled) by means that surface temperature, compared to a nearby region
include light, forced air, direct contact, without anomalies. Figure 2 demonstrates how the
electromagnetic induction, or solar heating. The surface temperature is affected by an obstruction to
technique of stimulation is based on the material the diffusion of heat from the surface. In simple
composition of the test object and anticipated cases, large subsurfacediscontinuities significantly
discontinuity type, the degree of precision/accuracy affect the flow of heat, causing anomalous cooling.
required, and practical staging considerations (for These discontinuities may appear as obvious hot (or
example, available physical footprint and electrical cold) spots in the infrared image sequence, spots
power, temperature limits in the workspace, surface easily identifiable by the operator as the test object
condition, and geometry). responds to excitation. However, dedicated signal
processing is often used to improve signal-to-noise
performance to detect smaller or more subtle
Temperature Measurement discontinuities or to perform quantitative analysis or
An infrared camera provides a dynamic thermal automated discontinuity detection.
image of the test surface as it responds to excitation. It is important to recognize that the basic
The amplitude of each pixel in the infrared image thermographic test process described above involves
indicates the local surface temperature. at least two stages of energy conversion essential to

Figure 1. Basic steps in thermography.

Step Technique Application
· Solar heating Qualitative:
· Ambient heating trapped water, ice
Passive
Stimulate
Active · Optical:
flash, step, modulated Delamination
· Direct contact Impact damage
· Forced convection Foreign objects
· Electromagnetic induction Fluid detection

Measure surface
Infrared camera
temperature change
Qualitative:
Direct viewing of infrared image
water, delamination
Raw
Analyze Quantitative:
Processed · Thermographic signal reconstruction thickness/depth
· Lock-in thermal diffusivity
· Principal component analysis adhesion
· Pulse phase automated detection
material characterization

11.2 AEROSPACE NONDESTRUCTIVE TESTING

atem Q 1 2   nL222t 
chenainqiu
negfsuwlh
reesruelat. pUanrltiikceulnaornfdoerm
stroufcetn
iveergteysitsing 

injected into a test object and is measured after it
passes through thevolume (for example, radiography
(3) T  x,t  
CL 
 n1
e  



orultrasonic testing), thermography is anindirect

process. Excitation energy (such as light or hot air) in which C, L, , and  are the heat capacity, slab
applied to a test object surface is absorbed raising thickness, thermal diffusivity, and density,
the surface temperature. Subsequent cooling due to respectively. Figures 3 and 4 show the plot of
diffusion of thermal energy into the test object Equation 3 on linear and logarithmic grids. The
volume results in changes in the infrared radiation logarithmic plot is initially a straight line with
emanating from the test surface. These energy a –0.5 slope, indicating unobstructed
conversion steps occur in even the simplest one-dimensional diffusion, followed by a horizontal
thermographic application. To a large extent, the line (slope = 0), indicating the termination of the
practice of thermography involves optimizing these diffusion process. Extensions of these two straight
conversion processes for a particular application.
Excitation sources and infrared cameras must be
chosen so that internal diffusion is both generated Figure 2. Infrared radiation emanates from test surface as injected
and measured effectively. thermal energy diffuses into test object bulk. Anomalous radiation
pattern occurs because of internal discontinuity.

Heat Conduction Infrared radiation

Anomalous cooling
In most approaches to thermographic
nondestructive testing, a region of the test surface in
the infrared camera field of view is heated uniformly. Test
The resulting diffusion of heat from the surface into object
the testobjectbulk is described byaone-dimensional Surface heating Thermal diffusion Discontinuity Obstructed diffusion
heat equation. Timing of excitation may be an
instantaneous pulse, a step, or a modulated sequence.
However, thefundamental nature ofthe diffusion
process, and the interaction of injected heat with Figure 3. Comparison of surface temperature response for 2 mm
internal structure, remains unchanged, regardless of (0.08 in.) slab and infinitely thick pulse heated carbon fiber reinforced
the timing of the excitation. The basic thermographic polymer.
nondestructive testing heat conduction process is 1600
illustrated by the behavior of a solid test object
instantaneously heated on one face by a uniform 1400
pulse. In the case of an infinitely thick, discontinuity
1200
free test object, the temperature response T of a
Temperature (mK)

singlepoint on the surface, relative to its temperature 1000

before excitation, is given by:
800
Q
(1) T t 
e  600
t
where e is the thermal effusivity of the material, Q is 400
the energy flux at the surface at time t = 0, and t is 200 2 mm thick carbon fiber reinforced polymer
time [Carslaw 1986]. A linearized version of Ideal carbon fiber reinforced polymer
Equation 1 is obtained by taking the natural 0
logarithm of both sides, as shown in Equation 2: 0 5 000 10 000 15 000 20 000 25 000 30000
(2)   
log T
Q       Time (ms)
log   0.5 log  log t
 e 
Figure 4. Logarithmic plot of surface temperature response for 2 mm
(0.08 in.) thick pulse heated carbon fiber reinforced polymer slab
Equation 2 is a straight line with a slope equal
to –0.5. Internal discontinuities or structure in the shows that back wall of slab causes departure from ideal straight line
test object may affect the one-dimensional flow of behavior at time t * = 3.1s.
heat and cause deviations from this linear behavior. t* = 3.15
In very broad terms, the practice of thermography 1000 2 mm carbon fiber
could be described as creating one-dimensional heat reinforced polymer
Temperature (mK)

flow in a test object, measuring the surface 100

temperature, and identifying deviations from the
slope = – 0.5
expected linear behavior.
The most basic deviation from ideal behavior 10 25 mm carbon fiber
occurs in the adiabatic slab, where no heat passes reinforced polymer
through the walls of the test object, and the diffusion
process is effectively terminated when heat reaches 1
the opposite surface. In this case, the solution to the 1 10 100 1000 10 000 100 000
heat equation is expressed as an infinite series: Time (ms)

THERMOGRAPHIC TESTING 11.3

 lines intersect at a time t*, related to the thickness High thermal effusivity materials, such as metals,
ofthe slab: do not exhibit large temperature changes in
response to excitation and, consequently, offer
2
relatively low intensity infrared signals that may be
(4) t* L masked by noise or background radiation during the
 
thermographic measurement. Polymer composites
have relatively low thermal effusivities and typically
Once t* is measured, Equation 4 can be used to provide strong infrared signals far above the level of
calculate either the thickness of the slab or its background infrared radiation and instrumentation
thermal diffusivity, if the other variable is known
noise.
(Maldague 1993).
Thermal diffusivity and effusivity trend similarly
The linearity and slope characteristics of the
in most materials. Metals have high diffusivities (fast
logarithmic temperature-versus-time plot shown in
heat propagation) and effusivities (resistance to
Figure 4 are independent of the slab material.
temperaturechange). Conversely, materialssuchas
However, the scale of theplot depends on the
polymers or glass have low diffusivities and
thermal diffusivity and thermal effusivity of the test
effusivities, corresponding to slow response and
object. Thermal diffusivity  is theratio ofthermal
large signal amplitude as shown in Figure 5. For
conductivity  to heat capacity C:
identical heating conditions, carbon fiber reinforced
 plastic with lower  and e has higher initial and
(5)   final temperatures and longer t* than the steel test
C object.
In the nonadiabatic case, where heat passes
The thermal diffusivity of a material determines t*, throughthe back wallofthe slab to deeper layers,
the time required for heat to propagate through a the surface temperature response becomes more
given thickness (Equation 4). In high thermal complex. Figure 6 shows the effect of changing the
diffusivity materials (such as copper or aluminum), back wall thermal conductivity on the plot of
heat propagates rapidly, to the extent that temperature versus time. Although t* still indicates
thermographic nondestructive testing may require the depth of the first back wall interface, the degree
high speed infrared cameras and excitation sources. to which the subsequent behavior deviates from
In low thermal diffusivity materials (such as glass or linearity varies, based on the ratio of the thermal
plastic), heat propagates slowly and the effusivities of the two layers (E = e2/e1). When E is
thermographic test process may be too long. smaller than one, the second layer impedes the flow
Polymer composites generally have intermediate of heat and retards the cooling of the front surface
thermal diffusivities well suited to thermographic after time t*. Conversely, when E is greater than one,
nondestructive testing with standard frame rate heat flow through the interface is enhanced, and
(60 Hz) infrared cameras and conventional sources. surface cooling after t* is accelerated. In either case,
Table 1 lists some thermal properties of common detectability of the back wall interface is enhanced
aerospace materials. as the contrast between layer effusivities increases.
Thermal effusivity e is the square root of the
product of thermal conductivity and heat capacity,
indicating the thermal inertia of a material; that is, Discontinuity Detection
its resistance to temperature increase as heat is
applied (Almond and Patel 1996): The surface temperature responds to a subsurface
discontinuity according to both the geometry of the
discontinuity and the contrast between the thermal
(6) e  C properties of the discontinuity and those of the

Table 1. Thermophysical properties of selected materials.

Thermal Conductivity Density Heat Capacity Thermal Diffusivity Thermal Effusivity

Material κ ρ C α e
[W/mK] [kg/m3] [J/kg·K] [–(m2/s) × 10–6] [Ws1/2/m2K]
Air 0.024 1.2 1006 19.9 5
Aluminum 237 2702 903 97.1 24 047
Carbon fiber 7 1600 1200 3.65 3666
Carbon fiber 0.8 1600 1200 4.2 1239
Fluoropolymer (Teflon® ) 0.25 2170 1050 0.11 755
Glass fiber 0.38 1900 1200 0.17 931
Glass fiber 0.3 1900 1200 0.13 827
Stainless steel 14.6 8000 480 3.68 7641
Steel 46 7900 440 13.2 12 645
Water 0.60 1000 4181 0.14 1550

11.4 AEROSPACE NONDESTRUCTIVE TESTING

surrounding host material. The key geometric For a nonadiabatic discontinuity, the
properties of the discontinuity are expressed by its thermophysical properties of the discontinuity
aspect ratio: material relative to those of the surrounding matrix
also affect detectability. As in the case of the
two-layer system described above, detectability of a
(7) Discontinuity diameter discontinuity is strongly influenced bythe ratio of
Aspect ratio  Discontinuity depth discontinuity and host thermal effusivities. Detection
is difficult in cases where the host and discontinuity
For the case of an adiabatic discontinuity (no heat properties are similar, and easier where they are
transfer through the discontinuity), heat from the different, as expressed by the effusivity contrast:
test surface reaches the discontinuity and is
completely obstructed. As the region surrounding
the discontinuity cools, heat trapped by the
discontinuity will flow laterally across the Figure 5. Pulse response of 3 mm thick steel slabs
discontinuity toward these cooler outlying areas. ( = 1.37  10–5 m2/s, e = 1.4  10–4 J/[m2·s0.5 K]) and carbon fiber
Thus, a point at the center of a discontinuity with a ( = 4.17  10–7 m2/s, e = 1.24 10–3 J/[m2·s0.5 K]).
large aspect ratio will initially behave identically to
a semi-infinite two-layer system. The influence of 10 000
geometry on conduction becomes more pronounced
over time and also near the edge of the
discontinuity. However, as the indication fades,
geometric effects become more pronounced, T (0.033 s) = 1.5 K
eventually reaching a point where the discontinuity
is undetectable. These considerations are 1000 Carbon fiber

Temperature (mK)
summarized in Equation 8, a rule of thumb reinforced polymer
expressing the geometric limit for detectability.
t* = 6.831 s
Discontinuity diameter T (0.033 s) = 0.132 K
(8)  1 T (33 s) = 0.104 K
Discontinuity depth 100 Steel
T (33 s) = 0.053 K
In practice, this ideal geometric limit of detectability
t* = 0.231 s
is difficult to achieve. It depends on factors that
include camera optics and sensitivity, amount of
excitation energy, and signal and image processing. 10
Equation 8 is useful, however, in that it describes an 10 100 1 000 10 000 100 000
ideal, best case scenario for detectability.
Discontinuities that do not meet the rule of thumb Time (ms)
criterion are not likely to be detectable, even under
the most favorable circumstances.

Figure 6. Temperature versus time plot for 1.5 mm thick layer of carbon fiber on various substrates.
Deviation from linearity increases with contrast between layer effusivities.

10 000

slope = –0.5
1000
Temperature (mK)

1.5 mm carbon fiber

reinforced polymer
100 Acrylic E = 0.4
Fluorocarbon resin E = 0.6
Carbon fiber reinforced polymer E = 1
Water E = 1.3

10 Stainless E = 6.13
Steel E = 11.3
Aluminum E = 17.0

1 t*
10 100 1000 10 000 100 000

Time (ms)

THERMOGRAPHIC TESTING 11.5

 criterion. However, care must be taken in making
these measurements, as the apparent size in the
ediscontinuity  ehost display may not represent the true size of the
(9) Effusivity contrast 
ediscontinuity  ehost feature. The following considerations apply:
1. For a directly viewed image, the size of the
For low effusivity contrast situations, high discontinuity in the apparent image of the video
performance cameras and careful experimental display is likely to be larger than the actual size,
configuration may be required. An aspect ratio because of the lateral flow of heat trapped by
greater than one and a high effusivity contrast are the discontinuity. For a rectangular
both necessary for discontinuity detection (that is, discontinuity, the true size of the principal axes
both geometric and material conditions must be is more closely represented by the full width at
met). Strong material contrast does not compensate half maximum of the temperature (Almond
for low aspect ratio, nor vice versa. As material 1994).
contrast becomes weaker, the effective aspect ratio 2. Discontinuity size is more accurately represented
limit becomes larger: in practice, the minimum by the apparent image after through-pixel
detectable aspect ratio increases in inverse processing, such as thermographic signal
proportion to the effusivity contrast. reconstruction derivatives or lock-in phase.
3. Accuracy of discontinuity size measurement is
reduced asthediscontinuity lateraldimension
Discontinuity Sizing approaches the detection limit, or as signal
strength decreases.
Thermographic nondestructive testing offers the 4. Discontinuity size measurements acquired earlier
ability to image a subsurface discontinuity and in the acquisition sequence are generally more
measure its dimensions in the resultant image, as accurate than those acquired later, as lateral
required by many aerospace inspections based on diffusion effects become more pronounced over
lateral discontinuity dimensions as an accept/reject time.

11.6 AEROSPACE NONDESTRUCTIVE TESTING

 Thermographic Instrumentation PART2

Equation 11 is the Stefan-Boltzmann law. The

Infrared Cameras and Remote emissivity is a characteristic of the emitter. Equation
Temperature Sensing 11 describes an ideal black body ( = 1) that absorbs
and reradiates all incident radiation. However, a
The infrared camera is the enabling technology of gray body ( < 1) is an inefficient absorber that may
modern thermography: it allows remote alsoreflect or transmit radiation.
measurement ofthe surface temperature ofatest Measurement of absolute temperature requires
object. First introduced in the 1960s, the camera calibration of the infrared camera for a given
provides a dynamic image of the test surface where emissivity. However, absolute temperature
the amplitude of each pixel in the image indicates measurement is not essential to mostthermographic
radiant emission and can be correlated tothelocal test applications, where only the relative temperature
surface temperature. In thermographic testing, the change with respect to preexcitation temperature or
internal condition of a thermally stimulated test adiscontinuity freereference is required.
object is inferred from the surface temperature
response measured by aninfrared camera. Although
it is possible to perform thermographic
nondestructive testing without an infrared camera Camera Technology
(for example, early systems used temperature Modern infrared cameras use solid state focal
sensitive paints and tapes to measure surface plane arrays: a lens assembly focuses incident
temperature, and liquid crystal sheets are used for radiation on a tiny rectangular array of detectors on
some applications), the twenty-first century infrared an integrated circuit chip. Commercially available
camera offers dynamic, whole field temperature cameras offer numerous options and enhancements;
measurement that surpasses its predecessors in their sizes range from palm sized microbolometers to
sensitivity, resolution, and ease of use. large, high speed laboratory instruments. The
Details of infrared camera engineering and the cameras are designed tominimize stray radiation
underlying solid state detector technology are widely from the internal components and typically provide
documented elsewhere and not discussed in detail a continuous video display or encoded digital image
here (Dereniak 1996). However, an understanding of stream at rates of 30 Hz or higher. Cameras typically
the principles of camera operation and accurate operate in either the 2 to 5 µm or 8 to 14 µm
temperature sensing are essential because they let a spectral range and may operate at room temperature
practitioner optimize the inspection configuration orrequire cryogenic cooling by an internal cooling
and prevent error. Infrared cameras used in engine, depending on the detector.
thermographic nondestructive testing are sensitive to Detectors can be broadly classified as either
electromagnetic radiation with wavelengths from quantum detectors, which directly detect incident
2 to 14 µm. The spectral radiance N of an object at photons, or microbolometers, which detect changes
temperature T (kelvin) is given by Planck‘s radiation in thermophysical properties of a small mass.
law: Quantum detectors, such as indium antimonide or
2hc2 mercury cadmium telluride, are significantly more
(10) N   5  hc  sensitive than and have faster response than
 exp  1 microbolometers, but they require cryogenic cooling
T  and are more expensive to manufacture.
Microbolometers operate in the 8 to 14 µm spectral
in units of watts per square meter per steradian per range at room temperature without cryogenic
micrometer, W·m–2·sr–1·µm–1, where c is the speed of cooling. Although less expensive than quantum
light, h is Planck‘s constant,  is Boltzmann‘s detectors, the transient signal-to-noise performance
constant, and  is the wavelength of the emitted for an individual pixel is lower than that of a
radiation. quantum detector. The manufacturer‘s specifications
Equation 10 expresses the amount of energy that for sensitivity and noise are often calculated for
a body at temperature T emits at a particular steady state performance, and not the transient
wavelength . Integrating Equation 10 over all situation encountered in most thermographic test
wavelengths yields the radiant emission I that is applications.
radiated from the test surface and is detected by an Camera selection for thermographic testing
ideal infrared camera: requires that the camera has the sensitivity, speed,
and spatial resolution for a particular application, in
(11) I  T 4 addition to the broader considerations of cost and
size. Many camerasoffercalibratedradiometric
where  is the infrared emissivity (which ranges temperature measurement, which provides data in
between zero and one) and  is a constant. units of temperature. However, many thermographic
nondestructive testing applications are based on

THERMOGRAPHIC TESTING 11.7

 temperaturecontrast, and do not requireradiometric Optics
output. Additional features (for example, on-board
digital storage, real time data communication, or Infrared cameras use lenses to focus radiation
coregistered visible light cameras) are available but from the target onto the focal plane. The optics often
not essential for most thermographic nondestructive comprise complex multilens assemblies. They canbe
testing applications. Key camera performance effectively described by their focal ratio and working
parameters for thermographic testing are these: array (lens-to-surface) distance. The focal ratio (also called
size, frame rate, noise equivalent temperature f number) is the ratio of focal length to the diameter
difference, and focal ratio (also called f number). of the aperture).

Focal length
Array Size (12) f ratio 
Entrance aperture diameter
Focal plane arrays are typically designed with a
4:3 aspect ratio to accommodate display devices. A The flux of radiation through the lens increases with
baseline array size of 320 wide by 256 high is entrance aperture diameter. In thermographic
typical of commercially available devices, and higher nondestructive testing applications, where signal
density arrays (640 ×480 or greater) have become strength is often limited, a lens with a lower focal
more common. The array size and optics determine ratio is often preferred because it provides a stonger
the size of the field of view at the inspection plane. input signal.
Forthermographic testing, the size ofa single pixel
projected onto the target limits the detectable
discontinuity size. For most applications, it is
recommended that at least nine pixels overlap the Excitation Techniques
minimum detectable discontinuity area. There are several means for thermal excitation in
thermographic nondestructive testing: radiative
Frame Rate heating (such as light, infrared, microwave, or
electromagnetic induction), mechanical stimulation
The frame rate at which the detector measures (sonic/ultrasonic, cyclic stress, convection, a direct
temperature determines the camera‘s ability to contact with hot/cold source), as well as chemical
measure fast transient events. Within each frame (exothermic reaction of a binary adhesive) and
period, the detector acquires data during the electrical (joule heating) stimulation. The excitation
integration time. For uncooled microbolometer may be applied to the surface of the test object (light
cameras, the integration may occur over the entire or direct contact excitation) or applied internally
frame period, while quantum detectors may use only (induction, sonic, or chemical heating).
a small fraction of the frame period to collect Thermography may be performed as a single-sided
photons. Some cameras allow the user to adjust or dual-sided access technique, depending on
integration time, to capture transient events, orto whether the excitation source and camera are on the
compensate for excessively high or low infrared same or opposite sides of the test object. As a
signal strength. Although acquiring every nth frame practical matter, single-sided testing is often
from any camera can simulate a low frame rate, a preferred, because both sides of complex structures
high frame rate is limited by the detector and may not be accessible. Figure 7 shows some common
readout electronics and by the detector integration techniques of thermal excitation.
time. For thermographic nondestructive testing, the
frame rate should be selected to match the time scale
of the material and discontinuity type of interest. Passive Heating: Solar and Ambient
Examples are provided below. Qualitative inservice inspection of aircraft for
trapped water can be performed using only an
Noise Equivalent Temperature infrared camera and noninstrumental excitation (for
example, solar heating or cooling during flight).
Difference Although these approaches are slower and less
The sensitivity of an infrared camera can be precise than instrumental excitation, they are easy to
described as the smallest temperature difference that perform and permit rapid inspection of large areas.
can be reliably detected. However, like any electronic Solar heating involves letting the aircraft remain
instrument, the camera is inherently susceptible to outdoors in direct sunlight and then moving it into a
noise. Noise is generated by a number of sources and hangar or shaded area. The aircraft is observed with
determines a sensitivity limit expressed as an an infrared camera for several minutes while the
apparent temperature difference. The manufacturer‘s skin cools. Because they provide a conduction path
specification of noise equivalent temperature for the heated skin, internal structures (such as ribs,
difference is typically based on measurement of a stringers, or doublers) appear in the infrared image
temperature stabilized target, averaged over many ascoolfeatures relative to the unsupported skin.
frames. Consequently, the resulting noise equivalent Similarly, water in contact with the skin also appears
temperature difference is sometimes impractical for relatively cool and can usually be visually
thermographic testing, where temperature often distinguished from structural features. Analysis of
changes rapidly and where the instantaneous noise solar heating results is typically limited to direct
equivalent temperature difference is usually viewing of the infrared camera image by an
unavailable but more desirable than the average operator. The technique is used to detect large
measurement. amounts of water and is usually used as a screening
technique to be followed up by a secondary

11.8 AEROSPACE NONDESTRUCTIVE TESTING

inspection or destructive interrogation of the quantification, and automation are required, are
indicated region. often well served by flash excitation. Flash
An alternative approach to noninstrumental thermography is currently the most widely
excitation exploits the fact that aircraft are exposed implemented form of thermographic testing. For
to low temperatures during flight and gradually most applications, commercial flash thermography
return to ambient ground temperatures after landing. systems can test about 0.1 m2 (1 ft2) in less than 10 s
The aircraft proceeds to a hangar immediately after and can be fully automated.
landing and is observed with an infrared camera as
it returns to ambient temperature. Ice that has
formed during flight will liquefy and return to
Forced Air
ambient temperature more slowly than the rest of the Hand held heat guns offer a simple and
aircraft. Areas of the aircraft in contact with ice will inexpensive way to heat a test surface. The devices
appear anomalously colder than the rest of the are typically either high temperature industrial heat
aircraft. guns (maximum about 644 K = 371°C = 700°F) or
hair dryers (maximum 339 K= 66 °C = 150 F). Cool
air may also be used for excitation, but care must be
Optical Techniques taken to avoid condensation or deposition of residue
For both aerospace manufacturing and on the part surface. In either case, the inspector
maintenance, the most widely used thermographic moves the device over thearea tobeinspected and
excitation techniques are optical. Using visible light observes the response with an infrared camera.
to heat the surface, optical excitation offers Results may be viewed directly in the infrared
noncontact coverage of a larger area than, say, camera output; for additional sensitivity, time
ultrasonic testing(Cargill 2006). Opticalexcitationis domain signal processing may beapplied.
particularly attractive for large structures. In most Forced air excitation may be used to detect
cases, optical excitation systems do not require trapped water, delamination, or impact damage in
critical alignment and positioning, so a large scale composites or to evaluate repair patches. When using
thermographic system is less complex and less forced air, one must keep in mind that the heat gun
expensive than a system where precise multiple-axis is a significant source of infrared radiation, which
alignment and positioning must be maintained. may be reflected from the test surface. The reflected
Optical excitation is usually performed using radiation may dominate the infrared image, resulting
either xenon flash lamps or halogen heat lamps. The in an inaccurate temperature measurement. To
excitation may be delivered through pulses, in steps, minimize these effects, the heat gun should be
in a modulated (on/off or sinusoidal) sequence, or by moved far from the infrared camera optical axis as
amoving line source scanned over thetest surface soon as heating has been completed. If industrial
(Maldague 1993). In comparing these techniques, it heat guns are used, care must also be taken to
is easy to lose sight of the fact that although they prevent heat damage to the test surface.
involve different hardware and, often, different
approaches to signal processing, the fundamental Direct Contact
thermal diffusion at an internal obstruction of
incident heat flow is identical for all of them. For applications such as trapped water in
The means of optical excitation depends on the honeycomb structures or inspection of thick or low
nature of the anomaly to be detected (size and diffusivity structures (for example, concrete or glass
material of the anticipated discontinuity and fiber composites), the thermal excitation may be a
surrounding matrix), the scale of the application (test hot or cold object placed in direct contact with the
time and area to be covered), and whether test surface. Because the infrared camera‘s sight line
automation and/or quantification will be required. to the test surface is blocked during the period that
For example, very large scale applications, where the thermal excitation device is applied and
severe anomalies are likely but not requiring removed, this approach is not appropriate for
quantification or automation, may be better served detecting near-surface features or other applications
with a step, continuous, or modulated source. where thermal interactions are likely to be rapid (for
However, production applications, where speed, example, detection ofcorrosion inaluminum skins).

Figure 7. Common thermographic excitation techniques: (a) light; (b) forced air; (c) direct contact.

(a) v (b) (c)

Hot

Flash Scan

Cold

Step Modulate

THERMOGRAPHIC TESTING 11.9

 However, the technique is widely used to detect repeated until the entire area of interest has been
trapped water in composite control structures. interrogated. As an alternative to start-stop
Typically, a heat blanket is placed on the surface for inspection of discrete areas, it is also possible to
several minutes and then removed. In an alternate inspect an extended area by continuously moving a
configuration, the component is removed from the heat source and camera in tandem relative to the test
aircraft and placed in an oven fora specified period object. The heat source (for example, a halogen
of time. After heating, the component or test area is lamp) typically heats a line perpendicular to the
observed for several minutes. Because ofthe high motion of the source. The amount of heat applied to
heat capacity of water, areas containing water tend the test object is a function of the source-to-surface
to remain cooler and persist longer than dry areas in distance and the source velocity, as well as the
the image. source intensity. Continuous scanning systems have
The direct contact technique is relatively easy to been successfully deployed for large scale
perform, allows forinspection ofalarge area, and inspections such as the detection of trapped water in
provides a simple means of controlling how much helicopter rotor blades and ofwall thinning in steel
heat is injected (by varying the heater temperature boiler tubes. Components with cylindrical symmetry
and the duration of contact). Sources are typically have also been inspected continuously, using a
electronically controlledresistive heaters; however, turntable to rotate the part while the heater and
simple sources such as hot water bottles, bricks, or camera remain stationary.
sandbags can be used in the field. Interpretation of scanned data is typically
qualitative. In the most basic implementation, the
operator views the output of the infrared camera
Electromagnetic Heating during the scanning as a region of the surface passes
Techniques such as induction or microwave under the heater. Anomalous hot or cold spots are
heating can be applied tomaterials with sufficient treated as discontinuity indications and interrogated
metallic content (for example, an aluminum with a secondary method. Additional sensitivity and
honeycomb core supporting a glass fiber laminate). precision can be obtained by a combination of
Electromagnetic excitation can be used to heat the careful hardware configuration and signal
interior rather than thesurface ofatestobject (the processing. If the precise camera and source velocity
penetration depth of the heating being inversely and lens-to-surface distance are known, it is possible
proportional to the frequency of the electromagnetic torearrange the pixels ofthe acquired sequence so
radiation). Although electromagnetic techniques can that it appears to be stationary. Once this
penetrate deeply into a test object, the surface rearrangement has been performed, the sequence can
temperature signal associated witha subsurface be processed using thermographic signal
anomaly is still subject to the limitations imposed by reconstruction or other time domain techniques, as if
diffusion (Lehtiniemi 1994; Osiander 1995). it were stationary and uniformly heated.
In continuous scanning, the velocity of the scan
is critical. The scan must be slow enough so that the
Continuous Scanning field of view still covers the surface when the
In the schemes presented above, a discrete region response of the deepest features of interest occurs.
of the test object contained within the infrared However, the scan must be fast enough to capture
camera field of view is excited and analyzed. the transient response of shallower features too. In
Inspection of areas larger than the camera field of some cases, separate scans for near-surface and
view requires that the excitation and analysis be deeper features may be required.

11.10 AEROSPACE NONDESTRUCTIVE TESTING

 Interpretation and Analysis PART3
of Thermographic Results

7. Care must be takento ensure that theinfrared

Direct Viewing of Image image represents radiation emitted from the test
Contrast surface and not reflected off that surface. The
apparatus, the operator or other personnel in the
Forlarge discontinuities (aspect ratio >>1) with area, or other radiant sources (such as reflected
strong effusivitycontrasts, qualitative interpretation sunlight or space heaters) may cause false
of inspection results may be possible by viewing the indications.
contrast between the discontinuity and background
in the infrared image sequence. The typical responses
are either a hot spot, indicating an area of retarded
cooling consistent with a discontinuity having E > 1, Signal Processing of Image
or a cold spot, indicating accelerated cooling
consistent with a discontinuity having E < 1. Sequences
Direct viewing is intuitive and easy to implement, Signal processing has become an essential
using apparatus as simple asan infrared camera and element ofthermographic nondestructive testing
a heat gun. As a result, it is popular in maintenance practice because it offers significant advantages over
applications such as the detection of trapped water direct viewing of unprocessed thermograms in
in honeycomb core or validation of patch repairs. repeatability, sensitivity, and quantitative
However, there are several implicit assumptions and measurement. Signal processing routinely identifies
potential pitfalls in interpreting inspection data features undetectable by direct viewing. The power
directly. and convenience of personal computers coupled with
1. Direct viewing is not sensitive to low aspect the availability of real time digital data from many
ratio discontinuities. Discontinuities with aspect commercial infrared cameras have enabled the
ratios smaller than 6 may be difficult to detect development of dedicated signal processing
this way. algorithms that can be performed in a few seconds,
2. Anomalous indications may appear blurry due immediately after signal acquisition.
to lateral diffusion.
3. Interpretation is based on visual recognition of Thermographic Signal Reconstruction
contrast between a discrete discontinuity and a
discontinuity free background. This presumes The thermographic signal reconstruction
that a discontinuity free background exists, and technique is a widely used processing approach in
that the discontinuity is smaller than the region aerospace applications. Unlike direct viewing, it
being inspected. A discontinuity larger than the treats each pixel independently and does not
inspection region and a similarly sized presume the presence of a discontinuity or an intact
discontinuity free region may appear to be region in the field of view (Shepard 2003). It is
identical. particularly useful in situations where high
4. Direct viewing is sensitive to surface sensitivity or automation is required and has been
temperature variations because of nonuniform adapted to pulsed, step, and modulated excitation.
heating or emissivity variations on the test The technique builds on the simplicity of the
surface. These variations may lead to false behavior of the logarithmic temperature-versus-time
positive indications. plot (Figure 8) and expands on the concept of
5. The entire excitation response sequence must be finding deviations from linearity. In thermographic
viewed, because a single image taken from the signal reconstruction, the logarithmic temperature
sequence does not represent an interrogation of response of each pixel is fit toa simple equation (for
the entire test object volume. The duration of example, a low order polynomial) by using a least
the sequence must be long enough to represent squares fit algorithm. The resulting equation is a
the passage of heat from the surface through the replica of the original data but is free of temporal
volume of interest. noise, enabling further mathematical operations
6. It is not necessarily true that a subsurface without adding noise.
feature that appears later in the directly viewed The derivatives of the logarithmic surface
sequence is deeper than one that appears earlier. temperature, calculated with respect to logarithmic
The timing of the maximum contrast signal that time, are extremely useful in identifying and
the eye perceives depends on the product of amplifying deviations from normal thermal
depth and diameter and on the ratio of thermal diffusion. Figure 8 illustrates a logarithmic surface
effusivities, so that under certain circumstances, temperature plot of steel slab with the first and
a small deep feature could appear earlier than a second derivatives. For an infinitely thick,
larger feature that is closer to the surface. discontinuity free test object, the first and second
derivatives have constant values of –0.5 and 0,

THERMOGRAPHIC TESTING 11.11

Figure 8. Logarithmic plots of steel slab: (a) temperature versus time; (b) first derivative; (c) second derivative.

(a) Temperature change (K) 4 (b) (c) 0.6

–0.1

Second derivative
3

First derivative
0.4
2
–0.3
1 0.2

0 –0.5 0
0 5 10 0 5 10 0 5 10
Time [ln(ms)] Time [ln(ms)] Time [ln(ms)]

respectively — independently of the test material or impeded. Conversely, a negative second

the amount of energy applied to that surface. The derivative indicates E < 1; that is, cooling is
presence of a subsurface interface perturbs the accelerated by the interface. (b) A bipolar second
one-dimensional condition and causesthe derivative indicates a spatially discrete
derivatives to change in a predictable way. discontinuity. The initial peak represents the
Analysis of the logarithmic derivatives of each perturbation of the incident one-dimensional
pixel yields several pieces of information unavailable heat flow. The subsequent peak indicates lateral
with contrast based techniques. heat flow away from the discontinuity (or
1. The first and second derivatives both occur toward it, where E < 1).
earlier in the sequence than the directly viewed Thermographic signal reconstruction processing is
contrast maximum. The second derivative peak extremely effective for retrieving small or subtle
occurs just as the directly viewed contrast discontinuities. However, care must be taken to
begins to appear. As a result, there is ensure that the reconstructed replica is an accurate
significantly less blurring and higher feature representation of the original signal. The acquisition
definition in the derivative images. duration should be approximately three times longer
2. The time at which the second derivative than the time t* associated with the deepest interface
maximum occurs depends only on depth, not tobe detected. There is no benefit to acquiring data
discontinuity diameter. The maximum of the after this time, when one-dimensional diffusion has
second derivative corresponds to ln(t*), as terminated. If possible, the acquisition sampling rate
described in Equation 2. Thus, it is possible to (that is, the frame rate of the infrared camera) should
determine the depth or thermal diffusivity of a be fast enough to acquire a significant number of
feature from a single pixel, by measuring the frames before the time t* associated with the features
time at which the second derivative peak occurs. of interest.
3. The overall shape of the derivatives provides Figure 9 shows an example of comparing the raw
useful information about the localsubsurface and first derivative signals (at three points in time)
state of the test object. (a) A positive second for a graphite epoxy skin over aluminum core.
derivative indicates an interface where E > 1; Discontinuities at the skin-to-core interface are
that is, the one-dimensional flow of heat is significantly enhanced in the first derivative images.

Figure 9. Images of five-ply graphite epoxy laminate on aluminum honeycomb core: (a) raw images;
(b) first derivative thermographic signal reconstruction.
(a)

0.3 s 0.64 s 1.28 s

(b)
 Lock-in Thermography manner with respect to the source. However, as the
thickness (or discontinuity depth) approaches , the
When a test surface is heated with a modulated, thermal diffusion process is affected by the
periodic source, the surface response is a discontinuity in thermal effusivity and an anomalous
combination of a constant temperature increase and surface temperature response occurs. In lock-in
periodic deviations above and below the constant thermography, the magnitude and phase of the
contribution. In lock-in thermography, the periodic surface temperature are calculated for every pixel
response is isolated and analyzed in the frequency with respect to the excitation signal, and magnitude
domain to assess the subsurface state of the test and phase images are evaluated by the operator. The
object (Busse 1994). The underlying theory is based magnitude image is susceptible to variations in
on an observation by Alexander Graham Bell in surface emissivity and heating uniformity, as well as
1880 that led to the development of the ―thermal background reflection. However, the phase signal is
wave‖ concept by Rosencwaig (Bell 1880; relatively immune to these factors and extremely
Rosencwaig 1980). As the heating period increases, sensitive to variations in the periodic signal response.
the oscillating heat component applied to the test The relationship between magnitude and phase
surface can penetrate more deeply in each cycle, as images is analogous to the relationship between the
described by a highly attenuated thermal wave with direct temperature image and its logarithmic
penetration depth equivalent to the thermal diffusion derivatives in thetime domain. Figure 111 1 shows the
length : effect of the excitation frequency on indications in a
2 graphite epoxyskinover aluminumcore testobject.
(13)   The lock-in approach has been successful with
cp optical sources (such as halogen lamps) for inspection

where cP is specific heat,  is thermal conductivity, of composite structures. The modulation frequency is
 is density, and  is modulation frequency. selected for the thermal diffusivity of the test object
Figure 10 shows a schematic of how the modulation and the thickness or depth to be inspected. For
frequency and thermal diffusion length affect the aerospace composite applications, these frequencies
thermal distribution in a test object as related to the are typically in the range of 0.01 to 0.05 Hz
test object thickness. The phase is most sensitive to (acquisition period 20 to 100 s). Lock-in
a subsurface feature when thermal diffusion signal-to-noise performance and artifact rejection
approaches the depth of the feature. improvesasthesquare root of thenumber of
For a test object where thickness << , the surface acquired cycles. Aminimum of three cycles is
temperature response follows the source closely recommended. As a result, the total acquisition time
whereas, incaseswhere thickness >> , the deposited forlock-in processing is relatively long when
heat diffuses into the test object without interference compared to t*, thetimerequired to detect an
and the surface temperature behaves in a predictable interface using pulse excitation.

Figure 10. Lock-in thermography: (a) schematic; (b) phase effect of ratio thickness to thermal diffusion length.

(a) (b) 10
Lamp
Normalized phase (degrees)

–10

–20
1 2 3
–30

–40
µ 1
Sample L
–50
0 0.5 1.0 1.5 2.0 2.5 3.0
L >> µ L << µ L~µ
1 2 3 Thermal thickness (L /µ )

Figure 11. Lock-in phaseimages of five-ply carbon fiber reinforced polymer laminate on aluminum honeycomb withincreasing
excitation frequency f : (a) f = 0.01 Hz; (b) f = 0.04 Hz; (c) f = 0.1 Hz; (d) f = 0.5 Hz.
(a) (b) (c) (d)

THERMOGRAPHIC TESTING 11.13

PART4 Emerging Thermographic Techniques

Pulse Phase Thermography Principal Component Analysis

Pulse phase thermography is a processing In principal component analysis, an entire flash
technique that applies frequency domain processing thermography data sequence is represented as an
normally associated with lock-in thermography to M ×N matrix (M > N), where rows and columns
data from pulse heated test objects (Maldague and represent temperature at a given pixel location and
Marinetti 1996). In pulse phase thermography, a time, respectively. The matrix is diagonalized using
fourier transform is performed on the temperature singular value decomposition, and the resulting
decay data for each pixel of a pulse heated test columns are a set of orthogonal functions that
object, enabling the computation of magnitude and represent the content of the data set (Rajic 2002).
phase images as in lock-in thermography. The phase The columns occur in order of decreasing strength,
images, although analogous to the lock-in and typically, the first few are the principal
thermography phase images, are limited in ―thermal components, which can be used to generate an
wave bandwidth‖ and corresponding penetration accurate replica of the data (Cramer 2011).
depth range by acquisition and camera parameters. In many cases, the principal components can
Specifically, the high frequency (and near-surface effectively distinguish between surface features,
detection) limit is determined by the camera frame nonuniform heating effects, and various classes of
rate whereas the low frequency (deep feature subsurface features. Under controlled conditions, the
detection) limit is a consequence ofthe duration of technique has beenused for automatic discontinuity
the acquisition. The low frequency limit must be recognition. However, it is based on image statistics
carefully considered in performing pulse phase rather than physical behavior of the test object and
thermography, because the acquisition duration must can be significantly affected by optics, camera noise,
be long enough to ensure that the low frequencies of reflectionartifacts, and background objects that may
interest have been sampled. Conventional flash be included in the field of view.
thermography acquisition durations will be on the
order of 2t* of the deepest interface whereas the
duration of the modulation frequency required to
generate a thermal wave at the same interface will Vibrothermography
be 2t*. However, during the longer pulse phase Unlike most thermographic test techniques where
thermography acquisition period, the test object thermal excitation is applied uniformly tothetest
continues to cool and the decaying signal becomes surface, in vibrothermography (also known as sonic
increasingly susceptible to noise and reflection of thermography, acoustic thermography, and
background radiation from the test surface. These thermosonics), high energy acoustic excitation is
problems have been mitigated to some degree by applied to a point on the test surface. Although a
thermographic signal reconstruction processing discontinuity free solid produces little or no thermal
before fast fourier transform. response to sonic excitation, heat is generated at the

Figure 12. Crack detection with vibrothermography: (a) infrared camera detects sonic heating at crack;
(b) thermogram of crack in steel forging.

(a) Thermal energy (b)

Infrared
Crack camera

Infrared
Test radiation
object

Acoustic horn
Sonic
energy Coupling medium

11
faces of a closed crack by the resultant rubbing and modulated sequence. In most cases, crack indications
slapping. The result is selective excitation of cracks are easily identifiable against the null background,
that creates local heating at the test surface near the so that extensive signal processing is not required.
crack (Figure 12). Successful approaches using both single-frequency
The technique was reported in the late 1970s and swept frequency excitation have been
(Henneke 1979) but has been the subject of renewed demonstrated.
interest early in the twenty-first century (Favro Although the principle of vibrothermography is
2000). High speed infrared cameras and synchronous simpleand straightforward, anumberofissues must
signal processing have made it possible to detect be addressed before the technique can be
extremely small cracks almost instantly. successfully implemented. At present, these are often
Vibrothermography is particularly effective for cast determined empirically, through trial and error, and
or forged metal parts (for example, turbine blades or are the subject of many research studies.
vanes) where no signal is likely to occur in a 1. An appropriate insertion point for the sonic
discontinuity free test object. excitation must be identified.
A basic vibrothermography configuration consists 2. A coupling medium (for example, paper, tape, or
of a low frequency, high energy ultrasonic horn (0.5 foil) may be required to minimize the possibility
to 3 kW, 10 to 50 kHz), an infrared camera, and a of damage to the test surface.
fixture to control and maintain test object-to-horn 3. The test object must be held in a fixture that
contact (Figure 12). The key parameters for holds it rigid and does not provide a leakage
vibrothermography are the excitation frequency and path for the acoustic energy.
amplitude, the (static) force applied to the test object 4. A means of validating the coupling of sonic
by the horn, and the duration of the excitation. Once energy into the test object must be identified.
determined, these parameters need to be precisely 5. Although closed cracks can generate strong
reproduced for each test object. The excitation may signals, large open cracks may generate signals
be applied as a short pulse, a longer step, or a only at the tips.

THERMOGRAPHIC TESTING 11.15

PART5 Thermographic Test Standards

Until recently, national and international involving the space shuttle Discovery in 2003. The
standards for thermography have been only document describes a process that one would use to
indirectly relevant to aerospace nondestructive determine whether a prospective application is a
testing, dealing primarily with either the calibration feasible candidate for flash thermography and, if so,
and characterization of infrared camera performance how equipment and parameters canbedetermined. It
orforgeneral preventive maintenance. However, is based on simple analysis of the logarithmic
increasing acceptance of thermographic temperature-versus-timebehavior ofselected points
nondestructive testing has led to the development of on the surface of a test object assumed to have a
ASTM E 2582, Standard Practice for Infrared Flash back wall that limits the volume of interest.
Thermography of Composite Panels and Repair In commercial and military sectors,
Patches Used in Aerospace Applications in 2007. The nondestructive test procedures and standards are
initiative behind ASTM E 2582-07 came from the developedbyairframe manufacturers, applied by the
National Aeronautics and Space Administration, original equipment manufacturer and its suppliers in
where thermography played a key role in the Return manufacturing and applied by end users and service
to Flight program after the catastrophic accident providers after the aircraft has been deployed.

11.16 AEROSPACE NONDESTRUCTIVE TESTING

.
 Application Examples PART6

organizations forthe detection of trapped water in

Advantages and Limitations honeycomb sandwich structures (used in rotor blades
Like all nondestructive test methods, and fixed wing rudders), elevators, ailerons, and
thermographic testing of aerospace materials has other control structures. Various forms of excitation
both advantages and limitations (Table 2). Some of are used, including heat guns and lamps, flash
these limitations are practical and can be mitigated excitation, direct contact, and passive heating.
with equipment or procedures. For example, highly Thermographic detection of water is based on the
reflective surfaces can be prepared with a low uniquely high heat capacity of water relative to
emissivitycoating. However, limitations that arise as other materials in aircraft construction. Water
a consequence of heat diffusion, such as the absorbs incident heat but changes temperature far
minimum detectable aspect ratio, are fundamental less than composite or metallic materials. As a result,
and cannot be circumvented. surfaces in contact with trapped water appear to be
cooler than surrounding dry areas.
The success of a trapped water inspection will
What Can Go Wrong vary, depending on several factors: contact between
water and the inspection surface, extent to which the
Implementation of thermography can be simple honeycomb cell is filled, number of adjacent filled
or complex, depending on the application. It is cells, core material (for example, aluminum,
useful to note what can sometimes go wrong, so that fiberglass, or foam), presence of excess adhesive or
corrective measures can be taken. Table 3 lists the sealant, and skin thickness.
major issues. In severe cases of water entrapment (for example,
if several adjacent cells are nearly full and water is
in contact with the test surface), the inspection is
Trapped Water in Composite relatively simple and can be conducted using a heat
gun or lamp while the technician inspects the
Sandwich Structures surface directly with an infrared camera. Cool spots
Thermographic nondestructive testing has been associated with trapped water are readily
implemented by numerous military and commercial identifiable: they persist long after the heat source

Table 2. Thermography advantages and limitations.

Provides area inspection.

Is noncontact.
Displays results as image.
Advantages Can be applied from single side or through-transmission.
Can be used with active heat sources or passively.
Can be automated.
Can measure depth, size, and thermal diffusivity.

Discontinuities with aspect ratio < 1 are undetectable (fundamental).

Discontinuities with effusivity similar to host are undetectable (fundamental).
Sensitivity decreases with depth (fundamental).
Limitations Reflective surface may generate artifacts.
Inspection time depends on material properties.
Thick components may require large excitation.
Quantification may require specialized cameras and sources.

THERMOGRAPHIC TESTING 11.17

 Table 3. What can go wrong with thermography.

Issue Comment

Unable to detect deep feature Additional excitation or alternate source may be required to maintain signal strength above
noise.
Acquisition period should be at least 2t * of deepest feature.
Discontinuity aspect ratio must be >1.
Feature thermal effusivity is similar to matrix.

Unable to detect near-surface feature Frame rate is too low; high speed camera may be required.
Excitation is too long and masks feature.
Surface features or material variations mask feature.
Optics must project sufficient (> 9) pixels onto anomalous region.

Unable to detect subsurface features Feature thermal effusivity is similar to matrix.

Feature aspect ratio is too small, <~1.
Surface contaminant (water, grease) may mask response.
Feature is masked by internal clutter (such as fibers or sealant).
Acquisition time is too brief.
Optical resolution is insufficient for small features.
Sample may be infrared transparent: paint or infrared opaque coating is required.
Excitation technique is inappropriate for test object.
Radiation or convection response may be dominant.
Lateral heat flow to/from adjacent areas may dominate result.

Artifacts appear in infrared image. Surface is reflecting radiation from excitation source or background.
Image darkens farther from center (vignetting).
Remove radiators or shield the optical axis.
Prepare surface with paint or tape to reduce reflectivity.
Perform external nonuniformity correction to include lens.

disassembling the part from the aircraft and heating

Figure 13. Infrared image of helicopter rotor it in an oven. After the source is removed, the
blade during heating by hot air gun. surface temperature is monitored for a time (about
15 min) significantly longer than the laminate t*
(about 1 s). During observation, wet areas return to
equilibrium much more slowly than surrounding dry
areas. This approach is widely used with heat blanket
excitation for inspection of composite control
structures on commercial aircraft.
With additional instrumentation, the inspection
can be performed more quickly and with validation
of the presence of water. For example, using
thermographic signal reconstruction derivative
analysis of flash thermography data, a partially filled
single cell can be detected and confirmed in less
than 20 s (Figure 14). However, the coverage area of
a single flash thermography shot is on the order of
0.1 m2 (1 ft2), compared to approximately 1 m2
(10 ft2) for the heat blanket, so that many such shots
has been removed. Figure 13 shows an example of are required for a single control structure.
finding water in honeycomb by using a hot air gun
and an infrared camera on a rotor blade.
The inspection becomes more challenging for Quality Assurance in
smaller amounts (both cell fill volume and number
of cells) of water; for water not in direct contact with
the inspection surface; and where the presence of
Composite Manufacturing
In the process of manufacturing composite
excess adhesive or sealant also absorbs incident heat,
structures, straypieces ofpolymer or paper backing
which generates a false ―cold‖ indication that
material may adhere to a prepreg sheet and become
resembles water. In such cases, qualitative inspection
inadvertently embedded in the layup. Once the part
may be performed by heating the surface to a
is cured, such foreign objects compromise the
temperature substantially higher than ambient
strength and integrity of the structure. Detection
through direct contact (heat blanket) or by

11.18 AEROSPACE NONDESTRUCTIVE TESTING

C
before deployment of the finished part is essential. After the part is cured, subsurface delaminations
Thermography may be performed on either the green that obstruct the flow of heat from the surface will
(uncured) part or after the part is cured. be detectable in most cases using conventional
In the green state, the low density of the prepreg thermographic processing techniques. However,
material limits the depth penetration of the applied anomalies such as foreign objects or kissing bonds
heat. Active thermography with optical excitation (mechanical contact without adhesion) only partially
may be effectively performed on the first four to six obstruct the flow of heat. These more subtle
layers of the layup. In practice, this thickness often discontinuities do not generate a large surface
coincides with the compression (―debulk‖) process temperature response to incident heat and are often
that is performed after the addition of every few ply undetectable using conventional contrast based
layers to remove wrinkles and air gaps. In the green processing. Instead, they slightly retard the transit
state, small voids, delaminations, or foreign objects time required for heat to propagate from front to
may be masked by the fiber structure, which is quite back surfaces of the test object. Transit time is most
apparent in the thermography result. Signal quickly and accurately measured using the second
processing may be required, in addition to derivative peak detected with flash thermography.
two-dimensional image processing to minimize fiber Figure 15 illustrates thermographic testing of a
clutter. composite engine cowling. The result of the

Figure 14. Flash thermography of composite aircraft spoiler with trapped water: (a) raw thermal image
acquired 11 s after flash shows dark spots due to water and excess adhesive; (b) thermographic signal
reconstruction first derivative image identifies water entrapment sites.
(a) (b)

Figure 15. Thermographic testing of composite jet engine cowling: (a) inspection station;
(b) thermographic signal reconstruction (first derivative) of 24 shots.

(a) (b)

THERMOGRAPHIC TESTING 11.19

inspection is a thermographic signal reconstruction fiber reinforced plastic panel. In this case, the data
first derivative image, which shows significant have been processed to display the t* parameter
internal details of the cowling. Figure 16 is an extracted from the temperature-versus-time curve
illustration of foreign object detection in a carbon for each pixel in the image.

Figure 16. Unprocessed infrared image of polymer inserts in carbon fiber reinforced plastic panel: (a) eleven
seconds after flash heating; (b) thermographic signal reconstruction image showing t* of each pixel.

(a) (b)
References
Almond, D.P., and S.K. Lau. 1994. ―Defect Sizing by Transient Ibarra-Castanedo, C., M. Genest, P. Servais, X.[P.V.] Maldague,
Thermography. I: An Analytical Treatment.‖ Journal of and A. Bendada. 2007. ―Qualitative and Quantitative
Physics D: Applied Physics 27:1063-1069. Assessment of Aerospace Structures by Pulsed
Almond, D.P., and P. Patel. 1996. Photothermal Science and Thermography.‖ NDT&E International 22 (2-3): 199-215.
Techniques. London, United Kingdom: Chapman and Hall. Lehtiniemi, R., and J. Hartikainen. 1994. ―An Application of
ASNT. 2001. Nondestructive Testing Handbook 3: Infrared and Induction Heating for Fast Thermal Nondestructive
Thermal Testing, 3rd edition. Columbus, OH: American Evaluation.‖ Review of Scientific Instruments
Society for Nondestructive Testing. 65(6): 2099-2101.
Bell, A.G. 1880. ―On the Production and Reproduction of Sound Maldague, [P.V.] 1993. Nondestructive Evaluation of Materials
by Light: The Photophone.‖ Proceedings of the American by Infrared Thermography. London, United Kingdom:
Association for Advancement of Science 29 (October): Springer.
115-136. Maldague, X.P.[V.], and S. Marinetti. 1996. ―Pulse Phase
Busse, G. 1994. ―Nondestructive Evaluation of Polymer Infrared Thermography.‖ Journal of Applied Physics
Materials.‖ NDT&E International 27(5): 253-262. 79(5): 2694-2698.
Cargill, J.S., J.A. Pecina, L. Gintert, and D. Skelton. 2006. Milne, J.M., and W.N. Reynolds. 1985. ―Nondestructive
―Evaluation of NDT Technologies for Organic Matrix Evaluation of Composites and Other Materials by Thermal
Composite Testing.‖ Materials Evaluation 64(9): 857-864. Pulse Video Thermography.‖ Thermosense VII. Proceedings
Carslaw, H.S., and J.C. Jaeger. 1986. Conduction of Heat in of SPIE 520. Bellingham, Washington: SPIE, 119.
Solids, 2nd edition. New York, NY: Oxford University. Osiander, R., J.W.M. Spicer, and J.C. Murphy. 1995. ―Thermal
Cramer, K.E., and W.P. Winfree. 2011. ―Fixed Eigenvector Nondestructive Evaluation Using Microwave Sources.‖
Analysis of Thermographic NDE Data.‖ Thermosense: Materials Evaluation 53(8): 942-948.
Thermal Infrared Applications 33 [Orlando, FL]. Rajic, N. 2002. ―Principal Component Thermography for Flaw
Proceedings of SPIE 8013. Bellingham, WA: SPIE. Contrast Enhancement and Flaw Depth Characterization in
Dereniak, E.L., and G.D. Boreman. 1996. Infrared Detectors and Composite Structures.‖ Composite Structures 58: 521-528.
Systems. 1996. New York, NY: Wiley. Rosencwaig, A. 1980. Photoacoustics and Photoacoustic
Favro, L.D., X. Han, Z. Ouyang, G. Sun, H. Sui, and Spectroscopy. New York, NY: Wiley.
R.L. Thomas. ―Infrared Imaging of Defects Heated by a Shepard, S.M., J.R. Lhota, B.A. Rubadeux, D. Wang, and
Sonic Pulse.‖ Review of Scientific Instruments [Link]. 2003.―Reconstruction and Enhancement of
71(6): 2418-2421. Active Thermographic Image Sequences.‖ Optical
Henneke, E.G., K.L. Reifsnider, and W.W. Stinchcomb. 1979. Engineering 42(5): 1337-1342.
―Thermography, an NDI Method for Damage Detection.‖ Vavilov V.P., and S. Marinetti. 1999. ―Pulsed Phase
Journal of Metals 31(9): 11-15. Thermography and Fourier Analysis Thermal Tomography.‖
Russian Journal of Nondestructive Testing [translated from
Defektoskopiya] 35(2): 134-145.

RADIOLOGIC TESTING 11.21

 X
12 CHAPTER

Shearograp
CHAPTER

hic and
Holographi
c Testing Contents
Part 1. Introduction, 12.2
Part 2. Shearography Operation, 12.10
Part 3. Relevant Standards, 12.16
Part 4. Applications, 12.19
Part 5. Laser and Acoustic Excitation
Safety, 12.29
References, 12.32

Contributor
John W. Newman
12.1
 Introduction
PART1

Shearographic and holographic interferometry (Johnson 2006), was developed for nondestructive
nondestructive testing use laser based imaging test applications.
interferometers to detect, measure, and analyze both The electronic image shearing interferometer was
surface and subsurface anomalies in materials or pioneered in the early 1980s by three researchers,
structures by imaging submicroscopic changes to a John Butters at Loughborough University in the
test surface when an appropriate stress is applied United Kingdom, S. Nakadata in Japan, and M. Hung
(ASNT 2009). Although laser light is a at Oakland University in the United States (Hung
nonpenetrating radiation, shearography and 1982, Nakadata 1988). The commercial development
holography can inspect aerospace structures for of the shearographic camera as a tool for
discontinuities such as impact damage, disbonds, nondestructive testing led to the delivery of a
delaminations, inclusions, near-surface porosity, production shearographic test system in 1987 for a
wrinkled fibers, fiber bridging, and cracks. The major aircraft production program. A portable
shearographic and holographic test techniques are shearographic system was introduced in 1989 to fill
mature and effective solutions for a wide range of a need for fast, large area field inspection of aircraft
aerospace nondestructive test applications including honeycomb structures for bird and hail strikes.
composite aircraft panels, control surfaces, metal Portable vacuum and thermal shearographic systems
honeycomb or foam core panels with metal or are commonly used for nonvisible impact damage,
composite face sheets, elastomer or cork bonds, disbond, and delamination detection on composites
composite overwrap pressure vessels, and solid and for the evaluation of composite repairs. Large
composite laminates. shearographic systems are also installed and used.
Key benefits of shearography and holography Since commercial introduction in 1987,
include full field coverage, high throughput, shearographic technology has steadily advanced,
noncontacting interface, and noncontamination. In with the development of reliable, low cost, solid state
addition, shearography can perform nearly in real lasers; high resolution 12-bit charge coupled device
time, that is, can provide images while it scans. Also, cameras; and low cost, high speed computers. A
shearographic cameras do not need to move along milestone has been the development of phase
contours precisely and so can test a structure at an stepping shearography and the resulting high
offset angle. Shearographic and holographic images definition images, greatly simplifying test data
are relatively easy to interpret. Operator training interpretation.
programs through Level III are commercially
available, and shearography is included in the
civilian aviation standard NAS 410 (AIA 2008) and
SNT-TC-1A (ASNT 2011).
Shearographic Testing
As with all nondestructive test technologies, the A shearographic test system consists of a laser
strengths and limitations of the shearographic light source, a shearing image interferometer, an
method must be understood. It is necessary to image processing computer, display monitor, and a
qualify applications through verification tests, to means to provide controlled and repeatable stress to
develop written procedures, and to train and certify the test object. The shearographic optical system is
operators. Probability of detection has been analyzed what is referred to as a common path imaging
rigorously for a number of important shearographic interferometer, offering two significant advantages
applications. Once qualified, shearographic over holography. First, shearographic cameras are
techniques can be very cost effective and achieve more resistant to environmental motion and can
production throughputs from 2 to 110 m2/h (0.4 to operate without the vibration isolation tables
20 ft2/min), depending on the technique and the required for holographic systems. Second,
application. shearographic cameras create images showing the
first derivative of the out-of-plane deformation of
the test surface in response to a change in load.
Unlike holographic fringes, shearography is
Development relatively insensitive to test object bending or
The principles of holography were described in deformation due to the applied stress and the fact
the late 1940s (Gabor 1949). After the development that the reference beam is from the object itself;
of the laser in the 1960s, a light source with shearography, however is still highly sensitive to
sufficient power and coherence became available for local deformation caused by a discontinuity.
practical applications of holography (Leith 1964). Shearographic cameras are sensitive to changes
During the 1960s and 1970s, holographic in the distance from the object surface to the camera.
interferometry, based on work by K.A. Stetson In practice, Z axis surface deformations may be as
small as 2 to 20 nm depending on the environmental

12.2 AEROSPACE NONDESTRUCTIVE TESTING

 Figure 1. Schematic diagram of digital shearographic camera and thermal stress system.

Shearographic image
calibration device/data Laser: narrow line,
variable diffusion beam Test part (honeycomb
panel shown)
Charge coupled
device camera

System and imaging software

Phase
stepper
P1
Phase stepper
controller
P2
Phaseshiftmirror Beam
splitter
Monitor Twoaxis
images and data tilt mirror

Test part stress

controller and
Disbond skin
sensors (vacuum,
to core
thermal, or
vibration)

Test part stress device (thermal shown)

noise. Large test objects can be inspected with a few The separation distance and direction, or shear
images using a large field of view or many images vector, between the paired points P1 and P2 on the
with a smaller field of view that may be test surface determines the camera sensitivity to
automatically stitched together. The field of view for surface deformation. The light intensity detected by
a shearographic camera depends on the maximum each pixel in the charge coupled device camera is
allowable discontinuity size, camera resolution, laser determined by the complex interference of the light
illumination power, the ability to uniformly apply a wave from these two points on the target. Figure 2
stress change, and background noise. shows how the coherent, single-frequency light from
Figure 1 shows a shearographic schematic adjacent points on the part is combined in each pixel
diagram that includes the laser and optical elements in the shearographic camera. The random phase
for test object illumination and an imaging shearing difference  results from the random surface
optical system consisting of a beam splitter with a roughness on a diffusely reflecting test surface.
two-axis tilting mirror, a second mirror with a lead Stressing the part causes a relative phase shift ∆
zirconium titanate phase stepper, and the charge between light from well bonded homogeneous
coupled device camera. The laser light is expanded material and light from the surface above
through lenses to illuminate the test area on the discontinuities or heterogeneous material. In two
panel. Light from point P1 is reflected from the panel dimensions, this is expressed as ∆(x,y).
surface, where it is well bonded to the core. Light The phase stepper in the optical system applies a
from point P2 is reflected from the surface above a π/2 phase step at video frame rates (typically
skin-to-core disbond. If the panel is stressed with a 30 frames per second) to one leg of the shearing
small temperature change or a partial vacuum, the interferometer to allow the calculation of the phase
panel face sheet above the disbond will deform map and subsequent quantitative determination of
out-of-plane towards the shearographic camera. This the deformation derivatives between two stress
shorter distance traveled by light from point P2 states. As the applied load on the test object is
causes a phase shift with respect to the light from changed, two sets of phase stepped images are
point P1. Light from both P1 and P2 are combined by captured, and the phase calculation is performed for
the shearing interferometer at a single pixel in the each pixel over the image, by using the following
charge coupled device array. equation for the four-step approach:

SHEAROGRAPHIC AND HOLOGRAPHIC TESTING 12.3

  shift and strain change.
Figure 2. Signal intensity changes with phase
 
 Strain state 1    +  Strain state 2
 
 
 
 
 
 
 
 
 
Legend  
=P1 bonded
= P2 disbonded  
= sum of light intensity captured by charge coupled device pixel
 
 
–1  I6 x , y  Stress State 2
 x, y   tan I x , y    I x, y
8 7
 I5 x , y  
(1) –1 I2  x , y    I
 tan I4 x, y  I x, y   I3 x , y (6)
I
5  x, y
 x , y
 1   I   x, y  cos  x, y     x, y 


where I1 through I8 are eight sequentially phase 

stepped, captured images, described by the following
equations of stress states: 
I6 x , y   I   x, y 
 
Stress State 1 (7)
I x , ycos x , y     x, y   
 2 
(2) I1 x ,y   I  x , y  Ix , y cos  x , y


I2  x, y   I   x, y  I7  x, y   I  x , y 

 I  x, y  cos  x, y   
(8)
 I   x, y  cos    x, y     x, y   
(3)
 2 


I3  x, y   I  x , y  I8  x, y   I  x , y 
(4)
 I x , y  cos  x, y    (9) 3
   I x , ycos  x , y     x, y   

 2 

I4  x, y   I  x , y   where I  is bias intensity, I  is modulation intensity,
∆ is ratio test object deformation from applied load,
I x, y  cos  x , y   3
(5)
 and  is the random phase variable from reflection
2 
 of laser light from the test object.
The result of the phase calculation yields the
wrapped phase map in Figure 3a. This is an example
of the deformation derivative for a deformed flat
metal plate with a changing point load at the center.
The deformation is of a 200 × 200 mm (8 × 8 in.)

12.4 AEROSPACE NONDESTRUCTIVE TESTING

Cop
Figure 3. Shearographic images: (a) phase Figure 4. Shearogram of 610  360 mm (24 in.  14 in.) honeycomb
map of deformation derivative; panel showing cells, disbonds, embedded tubes, and metal inserts.
(b) unwrapped phase map; (c) integrated
unwrapped phase map; (d) three-dimensional
representation of platedeformation.
(a)

metal plate with a 100 mm (4 in.) diameter reduced

(b) thickness in the center subjected to a changing load.
Figure 3b shows the unwrapped phase map. Figure
3c shows the result of integrating the phase map,
and Figure 3d shows a three-dimensional
representation of the metal plate deformation.
Shearographic images, properly calibrated, contain
dimensional data that can be extracted by knowing
the image scale on the test surface and the shear
vector. These data, combined with the phase
information, allow the numerical calculation of a
test object‘s actual deformation under applied load.
Combined with the stress data, discontinuity
deformation rates as a function of applied stress
change can be calculated. This technique has been
used, for example, to grade impact damage in
(c) honeycomb and differentiate between crushed core
and disbonds.
In practice, real time phase maps and unwrapping
can be performed at video frame rates, and
quantitative measurement of the Z axis deformation
can be computed. Shearographic images often
provide exquisite detail of surface damage,
subsurface discontinuities, and structural features.
Discontinuities can be located, and their dimensions
and areas can be measured. Shearographic testing
can use a variety of stressing techniques on the test
object, including heating or cooling, applying
vacuum or pressure, mechanical bending, projecting
acoustic (audible or ultrasonic) signals, and
(d)
microwave radiation.
Figure 4 shows a vacuum stress shearogram of a
600 × 350 mm (24 × 14 in.) section of a spacecraft
metal honeycomb panel with 0.6 mm (0.025 in.)
thick face sheets. The cells, disbonds, and voids are
detected. Two shearograms were made in 14 s with a
pressure reduction of only 3.4 kPa (0.5 lb/in.2). The
metal face sheet over the honeycomb cells develops
microscopic dimples as the air pressure in the
shearographic test chamber is reduced. Features and
discontinuities distort the pressure pattern. One
discontinuity is a disbond on top of a coolant tube,
undetectable with through-transmission ultrasonic
nondestructive testing because of the tube beneath
it. Shearography, in many cases, can detect
anomalies in areas with complex geometry or
conditions.

SHEAROGRAPHIC AND HOLOGRAPHIC TESTING 12.5

 Figure 5. Schematic diagram for holographic interferometry exciting vibration in bonded test object.

Holography image Test part

calibration device/data vibration
excitation
Variable beam splitter
Laser: single frequency,
variable diffusion beam
Deformed material above
disbondat surface
System and imaging software

Charge coupled Fixed beam

device camera splitter
Monitor
images and data Disbond at
substrate
Test part stress
controller and
sensors
vibration

Test part exciter fixture

Holographic Testing the setup is more difficult and time consuming for
holography than shearography. Holography is more
Holography cameras combine laser light from the susceptible to vibration because of the long reference
target with laser light from the laser itself to detect beam, so the implementation of holography for
test object deformation. Unlike shearographic nondestructive testing is limited.
cameras that generate the first derivative of the
Z axis deformation, holographic interferometer
images show the total Z axis displacement, which
includes part rotation, translation, and deformation. Stress Loading
Figure 5 shows the configuration for a holographic Both shearographic and holographic testing
system with a vibration excitation system for the require a load change to be applied to the test object
inspection of bonded structures. The object is to reveal discontinuities. These changes are typically
illuminated with laser light while a reference beam is very low, compared to normal operating conditions
created with a variable ratio beam splitter to directly of the test piece. For example, composite overwrap
illuminate the charge coupled device detector. When pressure vessels are typically pressurized to
the reflected light from the test object is combined 0.1 percent of nominal working pressure or less
with the reference beam, an interference pattern is to detect cracks, impact damage, and fiber bridging.
created. As the object is stressed, the interference Table 1 lists loading techniques that are commonly
patterns are changed, and these changes can reflect applied, including thermal, partial vacuum, pressure,
underlying discontinuities in a structure. By taking and vibration.
two exposures of the hologram, one at the baseline Figure 6 shows two examples of shearographic
stress and a second at a test stress level, an stress mechanisms. Two 1 kW thermal lamps, shown
interference pattern containing contours for any in Figure 6a on the top of the shearographic system
surface displacement changes can be seen on the tripod, apply infrared energy onto a region of an
image of the object. aircraft radome. When the heat lamp is turned on,
Some of the most successful applications of the surface temperature of the radome increases and
holography have been with vibration excitation. The the surface expands. Features or discontinuities in or
vibration, generally in the 20 to 60 kHz range, is near the surface will cause a surface displacement
applied to the test piece during a single exposure. detected by the shearographic camera. Figure 6b
Discontinuity indications appear dark or black in the shows a large vacuum chamber that can handle large
final image. Without the added shear vector or aerospace sandwich core panels and control surfaces.
overlapped images characteristic of shearography, The shear camera is mounted on an X/Y scan gantry
the holographic data are simple to interpret. inside a vacuum test chamber that can handle large
However, from the practical aspect of application, aircraft sandwich core panels and control surfaces.

12.6 AEROSPACE NONDESTRUCTIVE TESTING

Table 1. Shearographic testing stressing methods for aerospace structures.

Technique Units Range Applications

Thermal   ( °F) 0.5 to 48 K or °C Laminate panels.

(1 to 30 °F ) increase Impact damage, delamination, wrinkled fibers, porosity,
inclusions, embedded foreign material, repairs.

Sandwich panel honeycomb, foam cores.

Impact damage, skin-to-core disbond,
damaged core, foam-to-foam disbonds,
metal core–to-skin disbonds, repairs,
kissing disbonds, skin delamination, water.

Resin transfer molded composites.

Resin lean areas, porosity, damage.

Engine stators, vanes, composite fan blades.

erosionstripbond, voids, resinlean areas, damage, foreign
objects.

Steel, aluminum, ceramics, composites.

Surface breaking or near-surface breaking cracks.

Composite overwrap pressure vessels with metal liners.

Disbonds attheliner-to-composite bond, fiber bridging.

Partial  kPa (lb/in.2 ) -0.14 to -49 kPa Elastomers.

Vacuum (-0.02 to -7.0 lb/in.2) Coatings, rubber and plasticvoids, disbonds, tires, solid
differential rocket motor liners, rubber-to-substrate bond,
cork-to-substrate bond.

Sandwich panels–to-honeycomb, foam cores.

Impact damage, voids, disbonds, radomes, aircraft control
surfaces, flaps, air brakes, helicopter blades, turbine engine
ducts, laminated wood structures.

Composite overwrap pressure vessels.

Composite overwrap pressure vessel fiber bridging, liner
disbonds.

Pressure  kPa (lb/in.2 ) 0.07 to 35 000+ kPa Composite overwrap pressure vessels and composite rocket
(0.01 to 5000+ lb/in.2) motors.
differential Impact damage, composite cracks, broken fibers, fiber
bridging, porosity.

Pressure vessels and heat transfer structures.

Metal pressure tanks, liquid propellant rocket engine exit
cones, thrust ramps, piping, space vented core honeycomb.

Vibration Air coupled 0.5 to 20 kHz 90 to 125 dB Foam rocket thermal protection systems.
Mechanically coupled Damage, disbonds, delamination, cracks.
4.0 to 250 kHz
Lightweight honeycomb.
Spacecraft solar panels, solar cell bond.

Metal honeycomb.
Metal honeycomb turbine engine fan cases, fan blade
erosionstripbond, metal-to-metal bondedpanels and
honeycomb.

Metalbrazedbondedandplasma sprayedengine
compressor seals.
Disbonds.

SHEAROGRAPHIC AND HOLOGRAPHIC TESTING 12.7

 Figure 6. Stress loading: (a) shearographic camera with two 1 kW thermal lamps testing aircraft radome; (b) large low vacuum
test chamber for aerospace parts.
(a) Thermal lamps (b)

Gantry scanner

Low vacuum
chamber wall
Shearographic
camera

Shearographic camera

The air pressure inside this chamber can be reduced under higher ambient pressure, exerts an internal
from ambient to –35 kPa (–5 lb/in.2) in less than 15 s, pressure on the skins. Features or discontinuities in
although typical vacuum test pressures for aircraft the skin cause differential surface displacements that
sandwich panels are in the range of –0.7 to –10 kPa stand out from the surrounding surface. This
(–0.1 to –1.5 lb./in.2). Under an external pressureless differential is detected with the shearographic
than 6.9 kPa (lb./in.2), the air in the core, sealed camera.

Figure 7. Measurement of shearographic indications.

20 mm
(0.79 in.)

12 mm
(0.47 in.)

13 mm
(0.51 in.)

12.8 AEROSPACE NONDESTRUCTIVE TESTING

Image Interpretation and Figure 8. Signal-to-noise ratio measurement of signal strength:
Analysis (a) acoustic shearogram; (b) plot from line scan.
Shearographic and holographic images provide (a)
direct measurement of discontinuity dimensions,
signal-to-noise ratios, Z axis deformation, and, for
shearography specifically, deformation rate as a
function of load change. Figure 7 shows dimensional
measurements on a shearographic image based on
the shear vector calibration data and the image scale
data. The cosine of the shear vector angle, multiplied
by the magnitude of the shear vector, is subtracted
from the measurement to obtain the actual size.
Precision measurements can be made only when an Line scan
indication is in the center of the camera field of
view. Away from the center, the test object surface
may be at an angle to the camera in one or two
axes. As with all full field optical systems, the cosine
errors for all axes must be summed to correct the
final indication measurements. In practice, the
operator may move the discontinuity to the center of
the field of view and retest, thereby allowing the
precise measurement of the indication.
The signal-to-noise measurement in the image is
very useful for both determining the optimum test
parameters and for quantitative measurements of the
signal strength for grading discontinuity indications.
The signal-to-noise ratio is based on measurements
of the pixel gray level in the area adjacent to the
discontinuity indication and the peak or average (b)
value across the indication. Figure 8 shows an
acoustic stress shearographic image of foam
insulation on a rocket cryogenic fuel tank. The Signal level
discontinuity signal-to-noise ratio changes with the
test parameters such as the shear vector, stress
change levels, stress technique, and excitation
amplitude. Once the basic shearographic technique is
selected, the signal-to-noise ratio can be used to
optimize the signal strength by adjusting the test Signal-to-noise ratio Noise floor
parameters.

SHEAROGRAPHIC AND HOLOGRAPHIC TESTING 12.9

 Shearography Operation
PART2

The operation of a shearographic test system 900 mm (36 × 36 in.), but the sensitivity to
involves a series of important system setup steps discontinuities changes considerably as the field of
(Table 2) in a prescribed sequence to achieve view is increased. To obtain the maximum
acceptable test results. Additionally, reference throughput for in-process shearographic inspection,
standards with discontinuities must be tested at the the maximum field of view is desired.
setup configuration values to validate the sensitivity The maximum field of view for a shearographic
levels. Published consensus standards are discussed camera is determined by the following variables:
below. (1) the shear camera charge coupled device pixel
count in horizontal and vertical axes, Px and Py;
(2) dimensions X and Y of the largest allowable
Table 2. Shearographic test setup and process steps. discontinuity (Dx and Dy) for a specific structure and
application; (3) minimum pixel count Ix and Iy
Step Setup required to define a discontinuity indication;
(4) optical resolution of the shearographic system;
and (5) environmental factors affecting the
1 Fixturing the test object. indication‘s signal-to-noise ratio.
2 Determination of field of view. The maximum allowable discontinuity size
3 Development of test scan plan. (Dx, Dy) is determined through engineering fracture
4 Focusing of camera. mechanics techniques, stress analysis of the
5 Calibration of shearographic camera. structure, or historical failure analysis data.
Maximum allowable discontinuity sizes typical in
6 Selection of test technique, of shear vector, and of stressing
nondestructive test specifications generally include
equipment.
an additional safety factor. The quantity and
proximity of discontinuities larger than the
maximum allowable discontinuity size must be
Fixturing of Test Object recorded for analysis and disposition. To maximize
Test object mechanical stability is important inspection throughput, the maximum field of view
during shearographic testing. A test object that must be determined that allows the operator to
moves during stressing or data acquisition may detect the indications equal to or larger than the
cause the image to decorrelate, requiring retest. Test specified maximum allowable discontinuity.
objects should be securely positioned against a
mechanically stable backstop or support fixture. Any Definition of Discontinuity Indication
debris should be removed from the surface of the
fixture. Thin or lightweight test objects may vibrate Nondestructive testing often requires an operator
because of vacuum chamber noise or ambient to identify discontinuities on a monitor. A low
machinery nearby. Shims, clamps, tape, or foam pads threshold for an indication in pixels is needed. A
may be used to secure the test object and dampen single pixel in an entire computer monitor screen
vibration. A secured test object appears black when cannot be detected by an operator and used to define
the reference images are captured. a discontinuity. Camera pixel sensitivity is not
During testing in a vacuum chamber, the test uniform, and all charge coupled device detectors
object must never come into contact with any part of have missing pixels or pixels with variable electrical
the chamber walls or frames. Even small movements response to incoming light. Multiple pixels together
of several tenths of a millimeter (one or two having a measurable signal-to-noise ratio can be
hundredths of an inch) can degrade image quality. used to define a minimum discontinuity indication.
Vacuum loads on chamber walls can reach several Pixel counts of 5 × 5 pixels to 12 × 12 pixels, with a
thousand kilograms (many tons) for even a modest signal-to-noise ratio greater than 1:1, have been
sized chamber, and the steel frames and panels may used to define an indication. Figure 9 shows
deform significantly. Test objects that touch these probability-of-detection curves for three idealized
components move and decorrelate the shearographic operators evaluating indications having pixel sizes
image. of 3 × 3, 5 × 5, 7 × 7, and 10 × 10. All operators
detected the indication at the 10 × 10 size more than
95 percent of the time. For a typical monitor image,
with 1350 × 1024 pixels, these results indicate a
Field of the View discontinuity measuring about 0.74 percent of the
Shearographic testing applied to aircraft and screen width × 1 percent of the screen height can be
spacecraft structures can use a field of view varying detected by operators 95 percent of the time or more.
from 100 × 100 mm (4 × 4 in.) to more than 900 ×

12.10 AEROSPACE NONDESTRUCTIVE TESTING

Image Scale and Field of View Figure 9. Probability of discontinuity detection plotted as
With a minimum discontinuity indication defined function of discontinuity pixel dimensions for three idealized
as (Ix, Iy) = 10 × 10 pixels, the required image scale operators (red, green, blue).
or resolution for any shearographic camera can be
determined. For the maximum usable field of view, 100
the dimensions of the largest allowable discontinuity 95
(Dx, Dy) must allow detection better than 95 percent
of the time. Hence Dx/10 and Dy/10 can be set equal

Probability of detection (percent)

to the image scale. For example, with a maximum
allowable discontinuity of 9 mm, the image scale Is
is 0.9 mm per pixel. With the image scale defined for
a specific application, the field of view is then
simply the image scale times the shear camera
charge coupled device pixel count in the X and 50
Y directions, Px and Py. For a shearographic camera
with a charge coupled device resolution of 1350 ×
1024 pixels, the field of view to image a 9 mm
discontinuity 95 percent of the time is therefore:

Horizontal 0.9 mm
(10) field of view = 1350 pixels
pixel
= 1215 mm 0
3 3 55 77 10  10
Vertical 0.9 mm Discontinuity indication area(squarepixels)
= 1024 pixels
(11) field of view pixel
= 921.6 mm
Other Issues Affecting Field of View
Total field  Shearographic testing requires uniform stressing
(12)
of view = 1215 mm 921.6 mm of the test article over the entire field of view for
each shearogram. Variations in acoustic stressing or
= 1.12 m2 heating during thermal stressing can lead to variable
discontinuity sensitivity. It is important to apply load
The above calculation is idealized. In the real changes during shearographic testing in a repeatable
world, additional factors may degrade image quality manner uniformly over the entire field of view. A
and reduce the probability of detection. Figure 10 nondestructive test reference standard panel with
shows an example of field of view selection. Four representative discontinuities sized to be larger and
discontinuity indications are evident in Figure 10a smaller than the maximum allowable discontinuity
with a field of view of 1200 × 900 mm. The smallest size should be tested in the center and corners of the
indication (arrow) is 50 × 12 pixels. At this scale, field of view to ensure uniform sensitivity. A weaker
however, size and area measurements are inaccurate. indication, with a reduced measured signal-to-noise
A close image, such as Figure 10b with a field of ratio at locations in the corners or along the sides,
view of 450 × 450 mm, allows measurements with requires a reduction in the image scale and
2.5× the accuracy. corresponding field of view.

Figure 10. Shearographic images of aircraft laminate panel with different fields of view: (a) 1200 mm 
900 mm on left; (b) 450 mm  450 mm on right. (Image on left has been cropped.)
(b)

(a)

SHEAROGRAPHIC AND HOLOGRAPHIC TESTING 12.11

 Figure 11. Shearographic image of large 3.5 m  0.9 m (11.4 ft  3 ft) aluminum honeycomb aircraft
flap tested in 18 shearographic tests using vacuum stress.

Size of the test object may also be a factor in

determining the field of view. If maximum usable
field of view leads to the object filling only a small Figure 12. Thermal shearogram for section of
part of the final image, the field of view should be large aircraft radome: (a) eight sequential tests
reduced so the image of the object fills the field of from fixed position; (b) calibrated tests of
view. This produces better overall imaging in the discontinuity indications at angle normal to
same test time. surface.
(a)
Scan Plan
Test targets larger than a single field of view
image require multiple images to cover the surface
to be inspected. Multiple images must have overlap
to ensure that no region is missed. Typically, a
10 percent overlap in both directions is
recommended. Manual hand held shearographic
systems can use a taped string as guidance for
camera placement on large panels. Large gantry
systems usually have a graphical interface to allow
rapid scan programming on large parts.
Figure 11 shows a large aluminum honeycomb
aircraft flap scanned in 18 shearographic tests using
vacuum stressing. The image field of view overlap is
set at 10 percent, but the software places the images
in sequential cascade from the left edge. More than
60 disbonds are detected in the image set. Other
large test objects with compound curves such as
radomes can be scanned in large sections, with the
radome being rotated 60°per section, allowing the
viewing angle to the target to remain within 40°of
normal. The images are not calibrated due to the (b)
varying geometry of the test object within the field
of view, but if discontinuities are detected, close-up
calibrated shearograms can be made to measure and
locate the discontinuities on the part (Figure 12).

Focusing of Shear Camera 60 mm (2.34 in.)

Once the field of view is set, the shearographic
camera must be focused on the target area. Unless
the camera has auto focus, manual focusing is
required to obtain optimal image quality. Focusing 90 mm (3.58in.)
the shearographic camera is more difficult in laser
light than white light due to the granular nature of
the speckles. The shearographic camera laser spot
projector can be used to place light targets on the
object for focusing, or a piece of tape may be placed.
The camera should be focused with the maximum
lens iris opening to minimize depth of field. Because
lens focus depends on wavelength, shuttering the
laser to focus in room light may make images out of
focus. After the camera is focused, the iris should be
reduced to the normal setting for operation.

12.12 AEROSPACE NONDESTRUCTIVE TESTING

Calibration Figure 13. Shear vector angle convention: (a) +45° vector;
Feature and discontinuity indications in a (b) –60° shear vector.
shearographic image can be measured and the
out-of-plane deformations can be determined if the (a)
+45° shear image
system is calibrated for the image scale and the shear
vector. While shearographic cameras are available
with either a fixed or variable shear vector, each test
setup for a given image magnification and shear
vector requires calibration. Shearographic image 0°shear
calibration consists of two measurements. First, the image
image scale is measured in pixels per unit distance at
the surface of the test panel. Second, the shear
vector consists of the angle and measured amount of
separation between the two images of the test area
formed by the shearing interferometer at the part +Y
surface. The shear vector in a shearographic test –Y
+45° shear vector
procedure is expressed as the amount of image
separation and the angle, by using the shear angle (b)
convention shown in the Figure 13 diagram. For
example, the shear vector is written as SV = 6.4 mm
(0.25 in.) at ±45°. Starting with the shear camera
adjusted for a 0°shear condition, the sheared image –60°shear
is moved to the right (+X) or up/down, never image
adjusted in the direction of –X. For a +45°shear 0°shear
image
vector, the image is moved in the +X and
+Y direction. For ~60°substituted symbol for an
approximately equal shear vector, the image is +Y
adjusted in the +X and –Y directions. This –Y
convention (ASTM 2007) allows determination of
deformation in the Z axis direction from the
unwrapped phase map. In addition, adherence to this –60° shearvector
convention allows the operator to use the phase of
an indication to distinguish between a repair and
damage to a structure.
Manual calibration of a shearographic test setup Figure 14. Shear vector: (a) calibration card; (b) image scale dots
is accomplished before the start of any test by through shear camera.
placing a shear vector calibration card on the test
surface with the shearographic camera displaying the (a)
live image. Figure 14a shows a shear vector
calibration card with two sets of marks or dots with Image scale (pixels per in.)
known separation sized so that all are visible at the
desired image magnification. The two dots more
widely separated are used for the image scale while +45°
the two dots closer together, and possibly at an angle,
are used to set the shear vector. The shear vector
selected is based on test objects for the material,
thicknesses, and part geometry. The image scale dots 0°
should be at least 50 mm (2 in.) apart. Shear vector
With the calibration card in place, the shear vector is
adjusted so that the left shear vector dot is placed on
top of the right shear vector dot. The operator will (b)
then see three shear vector dots on the screen. The
image scale is set into the shearographic image
processor using the system software shear camera Image scale (pixels per in.)
calibration tool to measure the distance between the
image scale dots. With the shear vector applied, four
paired image scale dots will be seen on the monitor +45°
(Figure 14b). The image scale is determined by
measuring the distance between the left or right sets
of dots. The shear vector is set by adjusting the shear
camera tilt mirror to move the left dot in the moving 0°
sheared image onto the right dot in the stationary
Shear vector
image. Once the shear vector is set, three dots will be
seen, the center dot consisting of two dots
overlaying each other.

SHEAROGRAPHIC AND HOLOGRAPHIC TESTING 12.13

 Shearographic cameras must be recalibrated each
Figure 15. Shearograms of 160 mm (6.3 in.) diameter composite time the shear vector or the image magnification is
overwrap pressure vessel: (a) technique with pressure stress shows fiber changed unless the camera is fitted with encoders
bridging discontinuity only; (b) thermal shearography reveals both fiber and software for automatic resetting. Shearographic
bridging and square fluorocarbon resin insert. cameras are also available that can calibrate an
image automatically by projecting light patterns
(a) with known dimensions onto the plane of the test
object. The system software then calculates the
image scale and sets the shear vector that the
operator selects.

Selection of Stress Technique and of

Shear Vector
The selection of the shearographic stress
technique for a particular structure is highly
dependent on the discontinuity type, geometry of the
structure, and material properties. One stress
technique might reveal anomalies whereas another
might not. Table 1 lists stress techniques of thermal,
partial vacuum, pressure, vibration, and their areas
of application.
Figure 15 shows the difference in sensitivity of
using pressure stressing compared with thermal
stressing shearography on a carbon fiber composite
overwrap pressure vessel with an aluminum liner. In
Figure 15a, the pressure stress shows the fiber
(b) bridging discontinuity only. In Figure 15b, the
thermal stress shows the fiber bridging and a
fluorocarbon resin insert between the 9 mm (0.35 in.)
thick composite and the aluminum tank liner.
Pressurizing the composite overwrap pressure vessels
increases the hoop strain on the aluminum liner and
increases the compressive force on the fluorocarbon
resin insert between the liner and the composite
overwrap material. Pressure stress yields little strain
change at the site of the fluorocarbon resin insert
but a strong change for the fiber bridging
discontinuity. On the other hand, heating the surface
of the composite material causes thermal expansion
at a higher rate and time than for the deeper
aluminum liner. The fluorocarbon resin has a
different thermal diffusivity and thermal expansion
coefficient from the surrounding structure. The result
is a local high strain state and a highly detectable
shearographic response.

12.14 AEROSPACE NONDESTRUCTIVE TESTING

 Representative nondestructive test standards are
critical for selecting the best stressing technique. Figure 16. Shearographic testing of aircraft slat: (a) conventional
Selection may involve multiple tests with different image; (b) shearographic image with thermal stress; (c) shearographic
techniques to capture all possible discontinuities. In image with vacuum stress.
the example of a commercial aircraft slat shown in
Figure 16a, shearographic tests were run with (a)
thermal stress (Figure 16b) and vacuum stress
(Figure 16c) in a large test chamber. Because the
leading and trailing edges lie horizontal in the
image, a shear vector oriented parallel to the edges
was selected with a shear distance of 6.3 mm
(0.25 in.) at an angle of 0 degrees. The results for
thermal and vacuum shearography are substantially
the same except for the signal intensity for a few
discontinuities. In particular, a disbond indication at
the top left edge is substantially stronger with
thermal rather than vacuum stress. This disbond had
a small open vent to the atmosphere, limiting the
deformation caused by a pressure difference using
vacuum stress. The thermal stress image also reveals
detailed indications inside the largest indication.
Despite the weaker but still very detectable
indication seen in the thermal shearogram near the
trailing edge, compared to the vacuum response for
that discontinuity, thermal shearography overall
provides a better signal response for the indications (b)
shown in this test object. Thermal shearography has
the advantage of being performed either in a shop
environment or on the aircraft without removing the
slat, if clearance allows access to both sides of the
test object.

(c)

SHEAROGRAPHIC AND HOLOGRAPHIC TESTING 12.15

 Relevant Standards
PART3

Industry Specifications and then to design nondestructive test standards

accordingly. A nondestructive test standard built
Standards with fluorocarbon resin inserts, ideal for simulating
Shearographic and holographic testing in the foreign material in a composite laminate panel for
aerospace industry must be performed to desired ultrasonic testing, may bond during the panel cure.
specifications. Each company may have its own Although ultrasonic testing detects the signal change
internal specifications and specifications it requires in through-transmission or a reflection from the
of its suppliers. Additionally, there are a number of fluorocarbon resin due to the impedance change, this
industrial specifications available from industry standard may not represent a disbond for vacuum
groups that may be applied. Table 3 lists a number of stress shearography, which detects only actual
reference specifications, standards, and guidelines disbonding at the inclusion boundary rather than the
for shearography and holography. inclusion itself. Weakly bonded foreign material will
not be detected if the elastic bond strength is greater
than the applied pressure drop on the test surface,
usually only in the 3.5 to 13.8 kPa (0.5 to 2.0 lb/in.2)
Shearographic and range. Thermal stress shearography will detect the
Holographic Testing Standards insert if the coefficient of thermal expansion Ct
differs from that of the composite laminate. The
In shearography and holography, as with all thermally induced stresses during shearographic
nondestructive test methods, proper test standards testing of the composite test object are on the order
are important for the development of a procedure. of 70 to 3450 kPa (10 to 500 lb/in.2) or more
They should be used to determine the correct depending upon the material Ct and the temperature
operation of a system before and after a production change applied to the part, usually in the 1 to 11 K
test or run, as determined by written practice. It is (2 to 20 °F) range. The mismatch in thermal
important to understand the various discontinuity expansion between the materials is often enough to
types and characteristics in a given structure and

Table 3. Industry standard documents.

Agency Documents

Aerospace Industries Association AIA NAS 410 Rev 3, NAS Certification and Qualification of Nondestructive Test Personnel
American National Standards ANSI, Z 136.1-2000, Safe Use of Lasers
Institute
American Society for Nondestructive ASNT Recommended Practice No. SNT-TC-1A, Personnel Qualification and Certification in
Testing Nondestructive Testing
ANSI/ASNT CP-189, Standard for Qualification and Certification of Nondestructive Testing
Personnel
ASTM International ASTM C 274, Terminology for Structural Sandwich Constructions
ASTM D 3878, Terminology for Composite Materials
ASTM E 1316, Terminology for Nondestructive Testing, Section K Shearography and Holography
ASTM E 2581-07, Standard Practice for Shearography of Polymer Matrix Composites, Sandwich
Core Materials and Filament-Wound Pressure Vessels in Aerospace Applications
ASTM F 1364-03, Standards Practice for Use of a Calibrated Device to Demonstrate the Inspection
CapabilityofanInterferometricLaser Imaging NondestructiveTireInspection System
Code of Federal Regulations (United 21 CFR 1040.10, Laser Products
States)
21 CFR 1040.11, Specific Purpose Laser Products
29 CFR 1910.95, Occupational Noise Exposure
European Committee for BS EN 60825-1(2001-08), ASNT Safety of Laser Products, Part 1 – Equipment Classification,
Standardization Requirements, and User’s Guide
Laser Institute of America Laser safety guides and laser safety training courses.

12.16 AEROSPACE NONDESTRUCTIVE TESTING

 Figure 17. Carbon fiber laminate panel (24 plies) with two fluorocarbon resin inserts tested: (a) vacuum
shearography; (b) thermal shearography.
(a) (b)

allow detection of the insert. Figure 17 shows a

carbon laminate panel where vacuum shearography Figure 18. Shearogram of 152  203 mm
detects disbonds only whereas thermal shearography (6  8 in.) composite sandwich panel with
shows both fluorocarbon resin inserts and disbonds. eight-ply carbon fiber face sheets and flame
Shearographic reference standards with artificial resistant aramid core.
discontinuities are needed for calibration, inspector
training, and consistent implementation of
acceptance criteria. They should be built with the
same materials and processes used to manufacture
the actual part. If paints or coatings, for example,
are expected at the time of inspection, these should
be applied to the standards as well. A reliable
technique to manufacture shearographic standards is
to use a section or cutoff piece of an actual panel
with holes drilled through the far side face sheet,
through the core, and finished with an end mill.
Additional face sheet material can be bonded over
the far side holes. Figure 18 shows a shearographic
image of a standard where an end mill was used to
simulate disbonds at different depths in the panel.
The center row of milled flat bottom holes represent
face sheet-to-adhesive failure; the bottom row
represents adhesive-to-core failure. A matrix of
holes may have different diameters and depths that
can help simulate core-to-adhesive disbonds and
adhesive-to-face sheet disbonds.

SHEAROGRAPHIC AND HOLOGRAPHIC TESTING 12.17

 Figure 19. Nondestructive test standard panel with five-ply carbon fiber face sheets and flame resistant
aramid honeycomb manufactured with machined core and fluorocarbon resin inserts containing natural
discontinuity.

Nondestructive test discontinuity standards are elevated cure temperatures may cause a locally
also designed for shearography using stepped face higher residual tensile stress that tends to prevent
sheet thickness with an array of insert diameters and deformation of the material above the insert during
insert material types representative of potential shearographic inspection. Cores should be machined
foreign material in the composite layup clean room. at the site of the insert. Figure 19 shows a
Such panels are usually designed in a matrix. It is discontinuity standard manufactured with carbon
recommended that the width of each face sheet fiber face sheets and flame resistant aramid
thickness step be greater than 4× the discontinuity honeycomb. The inserts start at 6.4 mm (0.25 in.) in
diameter and that programmed discontinuities be size. All of the inserts, as well as a very small natural
separated by a distance of at least 4×their diameter. disbond identified by the arrow, are detected with
Another consideration in manufacturing thermal shearography. The small shear vector of
nondestructive test discontinuity standards is the 2.5 mm (0.1 in.) at 0 degrees, provides excellent
development of residual compressive stresses that detail and skin-to-core disbond sensitivity. The
tend to constrain Z axis deformation during design objective should be to create an artificial
shearographic testing if the inserts are between the discontinuity with known characteristics (diameter,
face sheets and core materials. During the autoclave area, depth) that reacts the same way as real
cure cycle, the materials are highly compressed, and discontinuities with the same characteristics.
any protrusion from the surface of the core at the

12.18
 Applications PART4

Advantages and Limitations to factors that inhibit its use. Table 6 lists some of
the issues that can arise and what can be done to
The advantages and limitations of shearographic address them.
and holographic testing are listed in Tables 4 and 5,
respectively. Within the scope of these limitations,
shearographic and holographic interferometry
techniques have been developed and proven for Shearographic Testing
numerous aerospace nondestructive test applications. Systems Application Examples
Shearographic test systems are for either portable
or fixed production applications. Portable
What Can Go Wrong? shearographic systems can be tripod mounted or
Shearographic testing has advantages for many configured for on-vehicle field inspection. Fixed
aerospace applications, but the technique is subject production systems may use test chambers to allow

Table 4. Advantages and limitations of shearography.

Advantages Limitations

Shearographic techniques are real time and full field It is sometimes difficult to obtain reflected laser light from test object with
inspections. sufficient power and uniformity to generate useful image. Black parts have poor
reflection, limiting inspection to small areas in order to produce image quality
Shearographic techniques are noncontact, good enough for evaluation of indications. In such cases, coating such as dye
noncontaminating, nonwetting. penetrant developermay be used to enhance reflectivity oftestsurface.
Can be applied during manufacturing process to Highly curved parts, such as filament wound tanks, may have glare area of high
detectdiscontinutiesasearly aspossible. laser light intensity that can cause image saturation requiring high dynamic
Single-side inspection is possible. range cameras.

Automated acoustic stressshearographic systems for Camera and test part should remain mechanically stable. Motion between
foam insulation on launch vehicle fuel tanks typically camera and object during data collection may degrade image and data quality or
operate with throughputs up to 112 m2/h make shearographic images unusable. Test surface and shearographic camera
(1200 ft2/h). must be in same position in space within about 0.003 to 0.3 mm (0.0001 to
0.01 in.) along X and Y axes and within 0.3 to 2.5 mm (0.01 and 0.1 in.) along
Dual-channel systems for helicopter blades, Z axis.
honeycomb, or foam core structures and laminate
panels can operate with high throughput of 74 m2/h Sensitivity of shearography to discontinuities depends on many factors including
(800 ft2/h). part geometry, discontinuity area, discontinuity type (embedded foreign material,
void, or delamination), material stiffness (or modulus), and depth from surface.
Hand held portable vacuum shearographic In general, as discontinuity depth increases, minimum detectable size also
instruments for on-aircraft honeycomb applications increases. Evaluation of shearography for given application must take into
can inspect up to 14 m2/h (150 ft2/h). consideration all factors affecting system, discontinuity detection throughout
full range of depth and location, and operator issues such as field of view that
Shearographic techniques inspect structures with
affect probability ofdiscontinuitydetection.
wide range of materials and geometries that might be
challenging for conventional nondestructive testing Shearographic image consists of two offset interferograms of test part surface,
techniques. so image of discontinuity indications appears larger in direction of shear vector
axis by amount equal to magnitude. Shearographic image of complex shapes or
Shearographic techniques do notrequireexact part
at edges of panel can complicate image analysis.
contour following. Test parts can be inspected at
angles of 45° or more, allowing inspections of tight Shearography depends on test part deformation to reveal subsurface
and otherwise inaccessible areas. discontinuities. Elastic materials such as composite laminates, sandwich panels,
plywood, metal, and compositematerials usuallyhave critical discontinuity sizes
Scan gantry motion precision, number of motion axes
large enough to be detected with shearographic techniques. Brittle materials
required, complexity and time for test part scan
and materials with very small coefficients of thermal expansion (Ct), such as
programming, and cost are lower than for some other
glass, silicone nitrides, and ceramics, however, usually have very small critical
nondestructive test methods. discontinuity sizes and are not candidates for shearographic testing.

SHEAROGRAPHIC AND HOLOGRAPHIC TESTING 12.19

 vacuum stress and scan gantries to inspect large Portable Shearographic Systems
panels or structures. As with all laser devices,
exposure of the operator to laser emissions must be Portable shearographic systems are designed for a
controlled and shearographic instruments and wide range of applications. Some include built-in
systems must comply with state and federal laws stressing mechanisms using heat, vacuum, or
regarding radiation health. ultrasonic signals; others require an external means
of applying stress during the nondestructive test. All

Table 5. Advantages and limitations of holographic interferometry.

Advantages Limitations

Holographic testing offers excellent nondestructive Holographic image interferometer is highly sensitive to environmental
testing capabilities for small parts from 3 mm to vibration, and, except for veryrareexceptions, mustbeperformedonvibration
1500 mm (0.1 in. to 60 in.). isolation tables or structures.

Primary holographic test application is for aircraft Holographic imagesshow totalpartdeformation under applied load. Resulting
engine abradable compressor seals and resin transfer indication may be overcome by panel bending or twisting as with thermal
moldedcompositesusingvibrationstressing. Method stresstechniques. Therefore, holography is limited to stressingtechniquesthat
has generallyreplaced ultrasonic testing, offers timely minimize panel deformation so as not to obscure local deformations caused by
throughput, and can test complex geometries that internal discontinuities.
prohibit ultrasonic testing.
Holographic bond inspection works best when more compliant material is
Digital holography is widely used for aerospace bondedwith adhesive, diffusion, ormetal braze bond to stiffer, less compliant
electronicsapplications and is specified under substrate. Sensitivity is reducedwhenviewingmorecompliant surface.
MIL STD 883 for hermetic seal inspection of electronic Detected discontinuity diameter-to-depth ratios in elastic materials, such as
devices. composites and metals, typically ranges from 1:1 to 1:3.

Table 6. What can go wrong with shearography.

Issue Corrective Action

Image decorrelates during stress Stabilize part. Eliminate noise sources causing sympathetic vibration of part.

Shearographic image fails to unwrap Data are collected while stress on part is changing too rapidly; eliminate
camera ortestpart motion.

Low light level reflected from part Open iris. Reduce laser diffusion. Move camera closer to part.

Excessive light from part Reduce iris. Increase laser diffusion. Increase distance from camera to part. Tilt
camera away from plane of part surface.

Testpart moving, poorimage quality Stabilizepart. Reduce shear vector. Change shearvector direction to be
perpendicular to part motion. Use laser doppler trigger.

Excessive air current noise in image Reduce shear vector. Use small fan to create turbulent air flow in shear camera
light and imagepath.

Uneven part illumination due to variation of part Slew laser bean elevation or azimuth with respect to camera view to optimize
reflectivity or geometry testpart illumination.

Honeycomb core panel disbond indications disappear Disbond or void is vented with leak path to test chamber. Inspect panel with
during vacuum shearography (in real time view) rapid pressure change using reverse mode.

Thin laminate panels warp during thermal stress If possible, apply heat to both near and far side surfaces simultaneously with
causing low discontinuity indication signal-to-noise approximately same heat flux to eliminate panel warpage.
ratio

Discontinuity indications become weaker away from Insufficient coverage of field of view with part stress technique; reduce field of
center of field of view with either thermal or air view or enhance stressing equipment.
coupled acoustic drive stress

12.20 AEROSPACE NONDESTRUCTIVE TESTING

Cnd
portable systems, however, include a laser attaches to flat or highly curved surfaces, even
illumination source, the real time image processing upside down. The system operates from the built-in
computer, and a means to support the camera either laptop computer or small touchscreen on the
by tripod or a means to vacuum attach to the surface inspection unit. Figure 21 shows the application of a
of the part or vehicle being inspected. Figure 20 portable system to an aircraft rudder, a system
shows a compact design for composite impact designed to detect voids and disbonds at both the
damage inspection where the inspection head near side and far side skin-to-core bond lines.

Figure 20. Portable thermal stress shearographic system: (a) shearographic instruments; (b) shearographic system attached with
vacuum to aircraft radome.

(a) (b)

Figure 21. Aircraft rudder inspection: (a) portable vacuum shearographic system; (b) shearographic image of disbond.

(a) (b)

SHEAROGRAPHIC AND HOLOGRAPHIC TESTING 12.21

 Production Shearographic Systems Figure 22 shows a shearographic system used to
inspect the space shuttle external tank foam
For production applications, shearographic insulation for damage, voids, and disbonds. The tank
systems can be mounted in X/Y scan gantries and measures 47 m (154 ft) in length and 8.4 m (27.6 ft)
can operate at throughputs typically in the range of in diameter. The foam insulation was inspected for
10 to 50 m2/h (100 to 500 ft2/h) per camera. First discontinuities to reduce the possibility of foam
introduced in 1987 in a major aircraft production debris striking the fragile thermal protection system
program, gantry mounted shearographic systems tiles or the carbon-carbon leading edge of the wings
share many operational features with ultrasonic test on the orbiter, the latter being the cause of the
C-scan systems: teach/learn part scan programming, Columbia disaster. Figure 23 shows the
electronic imaging of the entire part, image analysis, shearographic inspection of a large composite
discontinuity measurement tools, and automated aircraft structure‘s composite honeycomb fuselage
operation. Because shearography is noncontact, the section.
gantries do not need to follow test object contour
precisely.

Figure 22. Shearographic tests: (a) rocket fuel tank; (b) space shuttle external tank sprayed foam insulation disbond and damage
visible after air coupled acoustic stress.

(a) (b)

Figure 23. Shearographic tests: (a) composite sandwich aircraft fuselage being placed in test chamber;
(b) shearogram of fuselage nondestructive test standard showing 12.7 mm (0.5 in.) disbond indications.
(a) (b)

11.4 mm
(0.45in.)
 Both metal and composite helicopter blades are scanned in less than 15 min. Figure 25 shows the
easily tested in production with either thermal or results of a simultaneous two sided vacuum
vacuum shearographic techniques. Figure 24 shows a shearographic scan of a 3.7 m (12 ft) section of a
production shearographic system with dual helicopter blade scanned in 7 min. The stitched
shearographic cameras and gantries in a 10 m (32 ft) shearograms display images down to the individual
test chamber. The pressure reduction during cells and allow measurement and location of all
inspection cycles between ambient and 2 kPa indications.
(0.3 lb/in.2) below ambient. Helicopter blades are

Figure 24. Dual-channel shearographic test system inspection: (a) control console; (b) test chamber with two 10 m (32 ft) long
robotic gantries and two 1 W shearographic cameras.

(a) (b)

Figure 25. Vacuum shearographic test of honeycomb helicopter blade.

SHEAROGRAPHIC AND HOLOGRAPHIC TESTING 12.23

Table 7. Material diffusivity.

Material Material Diffusivity (m2/s)

Pyrolytic graphite, parallel to layers 4.2 × 10−6

Pyrolytic graphite, normal to layers 3.6 × 10−6
Aluminum 8.4 × 10−5
Aluminum 6061-T6 alloy 6.4 × 10−5
Steel, carbon (1 ) 1.2 × 10−5
Steel, stainless 304A 4.2 × 10−6

Stress Loading Examples before or after heating is generally determined by

experiment. However, for isotropic materials and
The following are examples of shearography with structures with thermal diffusivity  below
various stress techniques applied to aerospace 25 m² /s × 10−6, thermal shearography using the
components. reference-before-heating mode generally yields good
results. Conversely, materials with thermal diffusivity
Thermal Shearography greater than 25 m2/s × 10−6 tend to give best results
in the reference-after-heating mode. Table 7 lists
Thermal shearography has been applied thermal diffusivity  for several materials. Thermal
successfully to a wide range of structures and diffusivity combines the key parameters of thermal
material types including solid laminates, metal or conductivity, density, and specific heat capacity that
composite sandwich panels (limited to near side determine thermal flow through the material from
inspections only), and composite overwrap pressure the surface:
vessels, and even for crack detection in metal and
ceramics. Unlike thermography, which images k
(13)  =
surface temperature changes (or derivatives) over cp
time, thermal shearography uses a temperature
change to reveal local changes in the coefficient of
thermal expansion Ct, changes due to an anomaly. w reaclp is specific he(aW
thheerm t /c[amp·
aKci]t)y, a(nJ/d[kgi·
sKd]e)n,skitiys
Thermal shearography is unaffected by test object conductivity
(kg/m³).
emissivity and is equally able to inspect bare
For anisotropic materials and structures, such as
aluminum or painted surfaces without any
sandwich panels, the thermal shearography response
preparation except to remove grease or dew. For
can be highly complicated by bonding different
materials with a positive Ct, which includes most
materials, thickness, fiber types, and orientations.
materials such as carbon laminate, aluminum, steel,
and fiberglass, the test surface is heated. For Composite laminates are highly anisotropic: thermal
materials with a negative Ct, such as aramid, the test diffusivity out of plane measures typically
object must be tested by cooling the surface. 3.5 m² /s × 10−6 whereas in-plane thermal diffusivity
Thermal stress is typically accomplished using can reach 1300 m² /s × 10−6.
radiant light with a high infrared component. These Several useful examples of thermal shearography
1 kW quartz lamps, with adjustable reflectors are detection of sandwich structure disbonds,
mounted to the shearographic camera, can heat any detection of solid composite laminate wrinkles and
target in the field of view at distances up to 3 m impact damage, and composite laminate repair
(10 ft). Thermal shearography performed using hot inspection. Figure 26 shows thermal shearography
air guns to heat the part is discouraged but can for disbond detection in sandwich panels with
produce adequate results in materials with a low carbon fiber face sheets and aluminum core. The
thermal diffusivity if the user is very careful to heat panel is 203 × 305 mm (8 × 12 in.) tested in the
the inspected area evenly and if the field of view is reference-after-heating mode. The test panel
small: areas heated first begin to cool while other contains a number of intended discontinuities.
areas are being heated. Manual heating of the test Figure 27 shows wrinkled fibers in a thick 20-ply
object with a hot air gun is intuitive, is informed by solid carbon fiber laminate structure that is 350 ×
experience, and is difficult to quantify. For materials 350 mm (14 ×14 in.). This inspection was performed
in reference-before-heating mode. Figure 28 shows
with a negative Ct, cooling the test object can be
accomplished using compressed carbon dioxide or a an impact damage thermal shearographic image of
vortex cold air gun, but with the same limitations. nonvisible damage. The field of view is 125 ×
The reference images may be captured either 125 mm (5 × 5 in.) and shows the impact site,
before or after heating. Thermal stress parameters broken fibers, and the underlying delamination.
may either be time based or be controlled by real Figure 29 shows a thermal shearogram of a
time measurements of the test surface temperature composite repair. Repairs can contain porosity,
change during heating. The selection of reference disbonds, and foreign material detectable with
shearography.
Partial Vacuum Shearography to equilibrium with pressure changes, letting the
surface deformation relax and the shearographic
Partial vacuum shearography requires a test indications disappear.
chamber that can reduce the inside air pressure. For The Z axis displacement for the partial vacuum
portable testing, a vacuum window may be used stressed disbond in an isotropic material may be
consisting of a polycarbonate polymer plate with calculated (Prescott 1961):
edge seals and a variable, controllable vacuum
blower. Often a small shear camera is built into a  4 
PD
portable inspection head. Partial vacuum stressing W = k 3 
(14)
reduces the outside ambient air pressure on the test  EH 
object. Internal disbonds or delaminations in the
material contain small quantities of air, and when
the air pressure on the part surface is reduced, the where D is diameter, E is elastic modulus, H is the
test surface deforms at the anomaly. Critical for depth, k is thermal conductivity (W/[m·K]), P is
partial vacuum shearography is ensuring there are uniform partial vacuum pressure change, and W is
no leaks, so that the entrapped gas in the displacement of a circular disbond. The Z axis
discontinuity will rapidly vent within the timeframe displacement increases with the fourth power of the
of the data acquisition. Vented discontinuities come disbond diameter D. For discontinuities at the same

Figure 26. Thermal shearography of carbon Figure 28. Thermal shearographic image of
fiber skin with aluminum honeycomb, in nonvisible impact to solid carbon laminate wing
reference-after-heating mode. panel.

Figure 27. Thermal shearography of wrinkled Figure 29. Thermal shearographic image of
fibers in 20-ply thick solid laminate with disbonds in 200 mm (8 in.) diameter composite
reference-before-heating mode. repair on aircraft vertical stabilizer.

SHEAROGRAPHIC AND HOLOGRAPHIC TESTING 12.25

Figure 30. Shearographic image of end of 160 mm (6.3 in.) diameter Pressure Shearography
composite overwrap pressure vessel, showing fiber bridging at Pressure shearography has provided good results
transition to barrel. on a wide variety of pressure containing structures
including composite overwrap pressure vessel
composite drive shafts, tubular struts, and solid
propellant rocket motors. Pressure shearography
uses a gas to pressurize the test object from 0.01 to
5 percent of working pressure. Anomalies that affect
the hoop strain or create strain concentrations
generate indications on the surface. Test articles are
usually placed on a support frame allowing part
rotation. Pressure shearography is performed by
capturing reference images at a bias pressure,
increasing the pressure to a test level, and then
capturing the stressed data. The vessel can be vented
to prepare for the next test. Shearographic data can
be taken both during pressurization and venting, if
the phase of the data taken during venting is
inverted to match that taken during pressurization.
Flight hardware, such as composite overwrap
pressure vessels, may require specific cleanliness
precautions, such as using ultrapure gaseous
nitrogen with oil free valves, fittings, and piping.
Each area on the tank is tested sequentially with the
composite overwrap pressure vessel rotated 30°to
45°after each section is completed. The end domes
are also inspected at a 45°angle. Figure 30 shows a
shearographic image of a 160 mm (6.3 in.) diameter
depth, doubling the diameter increases the composite overwrap pressure vessel tested at
displacement 16×. As a result, small changes in the 34.5 kPa (5 lb/in.2 differential). The field of view on
discontinuity diameter can have a large effect on the tank is only as large as can be fully illuminated
the resulting displacement, which roughly with the shear camera laser. Controlling the test
corresponds to the signal strength and the pressure requires valves to pressurize and vent the
signal-to-noise ratio. To double the displacement for test object. Commercial pressure shearographic
a given size disbond D1 at the same depth, a systems are available that allow rapid, preset
disbond D2 would have to be only 18.9 percent pressurization and venting of the composite
larger: overwrap pressure vessel and a set minimum bias
pressure to increase test throughput and preload
composite overwrap pressure vessels. Typical
(15) D2 = D1 42 = 1.189 D1 distance between the shear camera and the
composite overwrap pressure vessel is 914 to
1143 mm (36 to 45 in.).
The displacement also decreases with the third Pressure shearography has also been used to
power of the depth H. The displacement is 8×smaller image porosity and poor consolidation of composite
for a discontinuity of the same size and at twice the tubes and pressure vessels. The tensile loads on the
depth. This variation does not take into fibers in the hoop and longitudinal directions on the
consideration that if the disbond is in a solid tank create strain concentrations at the site of
laminate close to the far side of the panel surface, porosity, voids, and poor consolidation. Figure 31
most of the deformation occurs on the far side, away shows three shearographic images of 250 mm (10 in.)
from the shearographic camera. In practice, solid × 40°field of view stitched together to create a
laminates can be effectively inspected with 250 mm (10 in.) wide × 120°view of a carbon fiber
shearography halfway through their thickness before wound composite tube. The part was pressure loaded
near side signal strength diminishes even faster than between 690 to 760 kPa (100 to 110 lb/in.2
the 1/H3 factor suggests. For solid laminates, double differential). The 690 kPa (100 lb/in.2) bias pressure
side inspection yields the best results and captures put the fibers in tension, and the shearographic data
disbonds through the full thickness on the part. were collected with the 69 kPa (10 lb/in.2) increase.
Honeycomb panels behave differently because the
large volume of entrapped air in the cells (nonvented
core) presses on both the near and far side face Vibration Shearography
sheets equally. Both near and far side disbonds are Vibration shearography uses either an air coupled
often detected depending on the stiffness of the core or a mechanically coupled sound source to induce
material. vibrations into the test object in the frequency
Partial vacuum is highly effective for face ranges for the resonance modes of anomalies. Two
sheet-to-core disbonds and delamination in the face types of vibration waveforms have been used with
sheet. Images of partial vacuum shearography have success: (1) sweep frequency and (2) bandwidth
been used throughout this chapter.
 Figure 31. Pressure shearographic image of Figure 32. Perforated face sheet aluminum honeycomb shearography:
composite tube showing porosity, voids, and (a) conventional photograph; (b) shearographic image with natural
poor consolidation. disbonds indicated by arrows and four milled flat bottom holes from
far side, measuring 13 and 20 mm (0.5 and 0.8 in.) in diameter.

(a)

limited white noise. An air coupled acoustic drive

should be considered for lightweight structures or
insulation materials, such as honeycomb with thin
face sheets and foam insulation used for launch (b)
vehicle cryogenic fuel tanks. A mechanically coupled
acoustic drive should be considered for
metal-to-metal bonding, metal honeycomb, and
structures with perforated face sheets. Figure 32
shows an aircraft engine inlet duct with perforated
face sheets and honeycomb core. Figure 32a is the
conventional photograph, and Figure 32b is the
shearographic image showing natural disbonds and
flat bottom holes. Vibration shearography is fully
real time: the system operates continuously during
data acquisition as the camera scans the test object.
Mechanically coupled acoustic drive testing is
performed using either a clamped-on or vacuum
attached vibration exciter. Air coupled acoustic drive
testing is performed using a speaker with a
directional horn. Power levels for air coupled
acoustic drives typically are from 250 to 2000 W
with sound levels sometimes as high as 125 dB. Both
air coupled acoustic drives and mechanically coupled
acoustic drives can produce sound levels above
limits for safe exposure. The images of disbonds
appear at frequencies above most of the structural
resonance modes.

SHEAROGRAPHIC AND HOLOGRAPHIC TESTING 12.27

Figure 33. Holographic test of turbine engine compressor seal.

0° 60° 120°

180° 240° 300°

Vibration Holography has been used with excellent results on metal braze
bonded parts, such as the aircraft turbine engine
Vibration holography is best used for smaller test compressor seal. Figure 33 shows an engine seal test
objects, with test surfaces up to 0.6 to 0.9 m (2 to imaged with bandwidth limited white noise
3 ft) and with complex shapes easily placed onto a excitation between 28 kHz and 80 kHz. These images
vibration isolated table in a holographic system. reveal both programmed disbonds and numerous
Holographic cameras do not use a sheared image and production discontinuities. The black indications are
provide clear, precise images of the test object. The disbonds between the brazed metal seal material and
test piece is generally hydraulically clamped to the the substrate steel shroud. The indications in section
vibration exciter. With piezoelectric exciters capable 0°to 75°are programmed disbonds; others are
of frequencies up to 200 kHz, vibration holography manufacturing discontinuities.
 Laser and Acoustic PART5
Excitation Safety

Laser Safety Class 2 laser systems are generally considered

safe under most working conditions because the
Laser shearographic and holographic test systems blink reflex (aversion response time) of the human
use laser light to illuminate the surface of a test eye prevents damage in the event of an accidental
article being inspected. The laser provides a exposure. Caution should be taken, however, with
convenient source of monochromatic coherent light regard to the intentional long term direct viewing of
that makes the implementation of shearographic and this laser light and concentrating the power of the
holographic testing possible. With the exception of laser light with positive magnification optics, such as
extremely low powered laser systems, virtually all a telescope or magnifying glass.
laser products pose some form of hazard. The most Class 3 laser systems are typically divided into
common hazards associated with lasers come from two categories: low power (CDRH 3a/IEC 3R) and
the direct exposure of the eyes and skin to the laser high power (CDRH/IEC 3b) systems. Precautions
light itself. Within the United States, laser systems required for the low powered systems are very
are classified in accordance with the regulations set similar to those associated with Class 2 lasers but
forth by the Center for Devices and Radiological with an increase in allowable power output from
Health in the Food and Drug Administration. 1 mW (Class 2) to 5 mW (CDRH 3a/IEC 3R). Care
In the United States, the Center for Devices and again must be given to intentional direct viewing of
Radiological Health has lent its initials (CDRH) to the unexpanded laser light and to magnification
laser classes. In the twenty-first century, both it and optics that increase the power of available light.
the American National Standards Institute are Class 3b lasers pose a unique category, ranging
moving toward adoption of international consensus from relatively safe lasers with outputs slightly
standards. In the European community, laser system above the 5 mW (CDRH 3a/IEC 3R) threshold to
standards are overseen primarily by the International relatively dangerous lasers with outputs up to
Electrotechnical Commission (IEC) and associated 500 mW. Class 3b lasers can easily produce burns to
national bodies, such as the British Standards both eyes and skin as power levels increase from
Institution. Additional federal, state, and local about 50 to 100 mW. The Class 3b classification
regulations may also apply to further regulate the applies to both visible and invisible lasers, thus
use of a laser product for a given application. Many increasing the potential risk of accidental exposure.
of these secondary regulations are based on Class 4 lasers and laser systems are considered
classification data provided by the American hazardous for both eye and skin exposure.
National Standards Institute (CDRH 2013; ANSI Additional hazards include fire and the production of
2007;IEC 2011). airborne contaminants, such as ozone. In addition to
the increased hazards, use of Class 4 laser systems is
generally very restricted and requires medical
General Laser Classification surveillance for operators whose eyes can be
Lasers and laser systems are divided into four exposed.
general classes (Classes 1 through 4) corresponding
to levels of increasing hazard. Each of these classes
may in turn have additional subclasses (for example, System Classification and
1M, 2M, 3a, 3R) that further define the general Equipment Selection
hazards associated with a given laser product. Laser
The classification of a laser system is based on
classification is based on wavelength, output power,
the type and level of exposure an operator would be
and whether the laser has been designed to operate
exposed to during the normal operation and
in a continuous mode or pulsed mode. Pulsed lasers
maintenance of the laser system. Based on these
by design release their stored energy in pulses that
conditions, it is very possible that laser systems
typically last well under 0.25 s; femtosecond (10–15 s)
classified as Class 1, Class 2, or Class 3a may contain
pulses can easily be achieved with the proper laser
Class 3b or Class 4 lasers.
system. Most shearographic and holographic test
From a practical standpoint, laser systems for
applications use continuous lasers.
shearographic and holographic test systems should
Class 1 lasers or laser systems are generally
be classified so as to provide the most usable system
considered safe because their power is low or
for the operator with the least restrictions. To
because the laser system has been interlocked in
minimize hazards and restrictions, systems with
such a way as to prohibit operator exposure to laser
classifications of Class 1, 2 and 3a (IEC 3R) are often
emissions.
preferred to those with Class 3b and Class 4
classifications.

SHEAROGRAPHIC AND HOLOGRAPHIC TESTING 12.29

Table 8. Common sense rules for laser safety.

Number Rule Comment

1 Never stare directly into operating laser system or at Intentionalextendedviewingofbothdirect and mirrorlike
bright mirrorlike reflections produced by laser light reflections can cause injury orblindness.
reflected from metallic or other highly reflected
objects.
2 Avoidunnecessary eyeexposuretobothdirect and When possible, close shutter of laser emissions or turn off laser
reflected laser emissions. power when working near front of laser system and access to
laser light is not required.
3 Do not leave laser systems powered and unattended, Turn off laser power and, whenever possible, remove laser
or with personnel unfamiliar with basic laser safety interlock key from Class 3b and Class 4 laser systems to prevent
procedures. unauthorized access to operating laser system.
4 Maintainlaseremissionswithincontrolledworking Be aware of all stray laser emissions and ensure that they do not
area. pose hazard to people nearby.
5 Warn bystanders or observers about presence of Never stare directly into operating laser system or at bright
laser emissions and possible hazards. mirrorlikereflections produced by laser light reflected from
metallic orotherhighlyreflected objects.
6 Do not use viewing optics such as telescopes to These devices can increase concentration of laser light being
view light from laser system. viewed.
7 Do notdisassemble, override, orotherwisemodify Classification of laser system is based on operator access during
safety interlocks and sensors for any shearographic normal operation and maintenance of laser system. Modifying
or holographic laser system. system optics, interlocks, or enclosures may invalidate
classification of laser system.

Shearography and holographic systems classified

as Class 1 and 2 laser systems generally do not
Acoustic Stress Safety
require any special safety consideration beyond a Hazardous levels of acoustic noise are not
basic understanding of the safe use of lasers. Under directly associated with shearographic testing but
normal working conditions, Class 3a laser systems may result from stressing techniques during its
extend allowable output emissions of the laser application. Sound levels of 130 dB are currently
system by 5× those of Class 2 laser systems without being used for a number of shearographic
additional restrictions beyond a more in-depth applications using acoustic or mechanical vibration
knowledge of safe laser operation. stressing. In acoustic stressing, large compression
Class 3b and Class 4 laser systems should drivers with focused horns vibrate the test object. In
generally be avoided for all but laboratory or well the United States, noise exposure regulations for
controlled environments because of operating industry are defined by the Occupational Safety and
restrictions and the need for additional medical Health Administration as documented within
surveillance. 29 CFR 1910.95. According to 29 CFR 1910.95, noise
When working with any laser system, a few exposure to sound levels above 85 dB must be
common sense rules of laser safety go a long way mitigated either through environmental controls or
toward establishing a safe working environment. with personnel protective devices, such as earplugs
or muffs. Long term exposure to sound levels above
85 dB has been shown to produce gradual hearing
Recommended Common Sense loss.
Rules for Laser Safety Sound pressure levels referenced by this standard
can be readily measured using inexpensive sound
Table 8 lists several common sense rules that level meters available through many audio and
apply to laser safety. Bypassing interlocks or electronic supply houses. The measurements are
modifying system enclosures may allow access to made using an A-weighted/slow response setting.
higher levels of laser illumination than normally Limitations as to the permissible time over which an
accessible for a certified laser product. Normal individual can be exposed to increasing levels of
eyeglasses are not considered viewing optics: they noise are defined by a United States code (OSHA
merely correct the natural vision of the human eye 2014, Table G-16) and range from 8 h at 90 dB A to
and do not increase the concentration of the light
being viewed. 15 min at 115 dBA (dBA meaning decibels in a curve
A, for environmental sound). Additionally, no
exposure to sound intensities greater than 140 dB
must be permitted.
 Because of variations in the application of 1. Be conscious of both operator and bystander
acoustic stressing, a worst case exposure exposure levels. If personnel other than those
corresponding to the maximum output of the performing an inspection are present, ear
acoustic driver over the expected work period (up to protection should be made available.
8 h per day) should be assumed. Noise protection 2. Warning signs should be posted outside the
devices should be selected so as to bring personnel danger area to warn individuals entering the test
exposure levels to no more than 90 dB (preferably area of possible high intensity noise exposure.
85 dB) over the course of an 8 h workday. General 3. Always use the minimum required noise level
noise recommendations for acoustic and mechanical for the inspection being performed.
vibration stressing include the following: 4. Always assume that the noise source is
potentially active unless it has been rendered
safe (preferably by removal of power).

SHEAROGRAPHIC AND HOLOGRAPHIC TESTING 12.31

.
References
AIA. 2008. AIA NAS 410, National Aerospace Standard Gabor, D. 1949. ―Microscopy by Reconstructed Wavefronts.‖
Certification and Qualification of Nondestructive Test Proceedings of the Royal Society 161: 454.
Personnel, Revision 3. Arlington, VA: Aerospace Industries Hung, Y.Y. 1982. ―Shearography: A New Optical Method for
Association of America. Strain Measurement and Nondestructive Testing.‖ Optical
ASNT. 1995. Section 3, ―Laser Based Nondestructive Testing Engineering 21: 391-395.
Methods.‖ Nondestructive Testing Handbook 9, 2nd edition: IEC 60825. 2011. IEC 60825, Safety of Laser Products. Geneva,
Special Nondestructive Testing Methods. Columbus, OH: Switzerland: International Electrotechnical Commission.
American Society for Nondestructive Testing, 103-157. Johnston, S. 2006. Holographic Visions. Oxford: Oxford
ASNT. 2009. Newman, J.W. Chapter 13, ―Laser Testing: University Press, 194.
Holography and Shearography.‖ Nondestructive Testing Leith, E.N., and J. Upatnieks. 1964. ―Wavefront Reconstruction
Handbook 10, 3rd edition: Nondestructive Testing Overview. with Diffused Illumination and Three-Dimensional Objects.‖
Columbus, OH: American Society for Nondestructive Journal of the Optical Society of America 54(11):
Testing, 473-490. 1295-1301.
ASNT SNT-TC-1A. 2011. ASNT Recommended Practice Nakadata, S. 1988. ―Phase Detection of Equidistant Fringes for
No. SNT-TC-1A, Personnel Qualification and Certification in Highly Sensitive Optical Sensing.‖ Journal of the Optical
Nondestructive Testing. Columbus, OH: American Society for Society of America A5(8): 1265-1269.
Nondestructive Testing. OSHA. 1999. OSHA Technical Manual (OTM), Section 3:
ANSI Z136.1. 2007. ANSI Z136.1, American National Standard Chapter 6, Laser Hazards. Washington, DC: United States
for Safe Use of Lasers. Washington, DC; New York, NY: Department of Labor, Occupational Safety and Health
American National Standards Institute. Administration [OSHA]; United States Government Printing
ASTM E 2581-07. 2007. Standard Practice for Shearography of Office.
Polymer Matrix Composites, Sandwich Core Materials and OSHA. 2014. 29 CFR 1910.95, ―Occupational Noise Exposure.‖
Filament-Wound Pressure Vessels in Aerospace OSHA Hazard Communication Standard. [Code of Federal
Applications. West Conshohocken, PA: ASTM International. Regulations: Title 29, Labor.] Washington, DC: United States
CDRH. 2013. 21 CFR, Chapter 1, Subchapter J, Part 1040.10, Department of Labor, Occupational Safety and Health
―Laser Products.‖ Washington, DC: United States Food and Administration [OSHA]; United States Government Printing
Drug Administration, Department of Health and Human Office.
Services, Center for Devices and Radiological health; United Prescott, J. 1961. Applied Elasticity. New York, NY: Dover, 402.
States Government Printing Office.

12.32 AEROSPACE NONDESTRUd

 X
13 CCHHAAPPT TEERR

Acoustic
Emission
Testing
Contents
Part 1. Acoustic Emission
Principles, 13.2
Part 2. Acoustic Emission Test
Techniques, 13.7
Part 3. Acoustic Emission Standards
and Specifications, 13.9
Part 4. Acoustic Emission Test
Applications, 13.11
References, 13.17

Contributors
Michele D. Dorfman
Valery F. Godinez-Azcuaga
Eric v.K. Hill
James L. Walker II

13.1
PART1 Acoustic Emission Principles

Acoustic emission is typically defined as the structural health monitoring. However, these
transient elastic waves generated by the rapid release attempts were unsuccessful largely because of
of energy from sources within a stressed material. instrumentation limitations. Extraneous signals from
Some of the most common sources of acoustic background noise and the equipment‘s overall
emission in metals are crack growth, rubbing of limited signal processing capabilities forced
fracture surfaces against each other, plastic terminations of in-flight monitoring programs by
deformation, and fracture of intermetallic aircraft manufacturers and the United States Air
precipitates (Pollock 1989). In composite materials, Force (Carlyle 1989). Since then, advances in
detectable emissions produced by matrix cracking, computer instrumentation have opened new
delaminations, and fiber breakage can be monitored possibilities for acoustic emission test applications,
by acoustic emission testing (Kouvarakos 1996). For not only in aircraft structural health monitoring but
acoustic emission to be produced, the structure or also in areas such as proof testing of composite
material being tested needs to be subjected to fuselage sections and aircraft composite pressure
external stresses, which in turn increase sufficiently vessels, as well as in-flight monitoring of rotating
the local stress around the discontinuities to a level components.
that causes the discontinuity to release energy in the Continual increases in computer capability and in
form of mechanical waves. The external stresses are data storage capacity — together with
typically produced by tension, compression, bending, miniaturization of electronics and sensors, plus
pressurization, temperature loading, or even newer and more powerful data analysis — have
electrical discharges. Stress dependency enables created a mature environment for acoustic emission
acoustic emission testing to be used as a real time testing to be widely applicable in the aerospace
global monitoring technique that can yield valuable industry.
information about the behavior of the specimen or
structure during service. This stress dependency is a
disadvantage in that a load must be applied to the
structure under test — not a simple feat when dealing
Data Acquisition and
with large aerospace structures such as aircrafts, Processing
rocket motor cases, and satellites. However, this In a typical test, acoustic emission transducers are
requirement can really be an advantage because it is mounted on the test structure and subjected to
not necessary to remove a specimen from its external load. These transducers, typically
working environment to test it. piezoelectric, convert the transient elastic stress
Acoustic emission testing works as a preventive waves generated by discontinuity growth activity
monitoring technique because a discontinuity or a into electrical voltage signals. The signals are
damaged area will produce many detectable amplified and sent to the acoustic emission data
emissions, long before the structure actually fails. acquisition system, where they are digitized,
Unlike more traditional nondestructive test methods analyzed, and stored. Typically, field programmable
— such as radiologic, electromagnetic, and ultrasonic gate arrays extract signal features from digital
testing — acoustic emission testing does not actively signals and analyze the features in real time, that is,
send out a probing signal into the material and then as the signals are acquired. These acoustic emission
listen for a return signal; rather, acoustic emission features are presented graphically to the operator in
testing detects the stress waves that are generated by the form of distribution, correlation, or time plots,
discontinuities within the structure. In that sense, which show the evolution of the acoustic emission
acoustic emission testing complements these active signals as the loading conditions change. A typical
nondestructive test methods by identifying specific acoustic emission test setup is shown in Figure 1.
areas of interest for follow up. It can be used as a During an acoustic emission test, a mechanical
global nondestructive test method. This capability is stress wave generated by the release of energy due to
one of the reasons for the success of acoustic discontinuity growth is detected by a piezoelectric
emission test applications such as monitoring of sensor, which converts the mechanical stress wave
pressure vessels in the petrochemical industry or into an electrical voltage signal. The electrical
detection of active corrosion in storage tanks. voltage signal is then passed through a preamplifier
Acoustic emission testing provides global results in and a frequency filter. The preamplifier provides a
real time, which can include the position of an gain from 20 to 60 dB and can include a high pass
acoustic emission source and its intensity, whereas or band pass filter.
intensive techniques provide specific information on Traditionally, the frequency range of an acoustic
discontinuity sizing. emission signal is from 20 to 500 kHz, although in
Acoustic emission testing was used in the certain civil infrastructure applications it could be as
aerospace industry in the mid 1970s for aircraft

13.2 AEROSPACE NONDESTRUCTIVE TESTING

low as 15 kHz and in other laboratory experiments
as high as 800 kHz. In aerospace applications, the Figure 1. Typical acoustic emission test setup.
most common frequency band goes from 100 to
300 kHz, encompassing the 150 kHz resonant
frequency of the most commonly used sensor
(Hill 1992). This frequency range eliminates low
frequency background noise and high frequency Acoustic emission
noise caused by electromagnetic interference but Preamplifier instrument
Acoustic
also limits the range of acoustic emission signals emission
that can be detected. Elimination of extraneous noise signal
is crucial to the success of any acoustic emission test
application and must be thoroughly investigated to • Detection
ensure the quality of data collected. Acoustic • Measurement
The amplified and filtered electrical signal is then emission sensor • Recording
• Interpretation
fed into the acoustic emission data acquisition
• Evaluation
system, where it is amplified again, digitized, and
processed by the field programmable gate array. Acoustic
Depending upon the type of application, the signal emission
features extracted by the field programmable gate source
array can be run through preestablished algorithms
for real time alarming, or the whole digitized signal
Data display
and its features can be stored for posttest analysis.
The signal features are typically displayed on the
Stimulus Stimulus
instrument screen in the form of two- or
three-dimensional history plots, correlation plots, or
distribution histograms. The data can also be
displayed in numerical tables.

Acoustic Emission Features Figure 2. Acoustic emission waveform and parameters.

To determine the sensitivity of the acoustic
emission test, an amplitude threshold is commonly
set. This threshold serves as a dividing line, Rise time
determining which signals will be recorded and
which will be neglected on the basis of their peak Energy
amplitudes. The threshold is useful for filtering out Threshold
low amplitude signals such as background noises.
Setting the threshold amplitude is important because Amplitude
too low a threshold will flood the data acquisition
system with meaningless data. Once a signal crosses
the threshold, features are extracted from the signal.
A typical acoustic emission signal, or hit, can be
represented as a complex trace of damped, sinusoidal
voltage versus time. Figure 2 shows the six most
commonly employed acoustic emission
quantification parameters of amplitude, duration, Duration
counts, rise time, counts to peak, and energy that
may be extracted from the signal. 1 2 3 4 5 6 7 8 Counts
The primary parameter is the amplitude of the
waveform signal. This value represents the maximum
amplitude of the waveform at its peak. The
amplitude is recorded as a voltage and is measured waveform is a function of both amplitude and
in the logarithmic scale of decibels because of the duration. Acoustic emission energy is defined as the
large dynamic range of the signals involved: from area under the rectified waveform or the measured
microvolts to volts. In the analysis of a structure, an area under the rectified signal envelope. Counts are a
amplitude threshold is commonly set and serves as a function of the duration and are equivalent to the
dividing line, determining which signals will be number of times the waveform passes the threshold.
recorded and which will be neglected on the basis of Counts-to-peak represents the number of counts
their peak amplitudes. The threshold is useful for from the time the waveform passes the threshold to
filtering out low amplitude signals such as the point where the amplitude reaches a peak, and
background noises. Setting the threshold amplitude the rise time is the duration of time it takes the
is important; too low of a threshold will flood the waveform to do so. These six acoustic emission
data acquisition system with meaningless data. quantification parameters can be used to
The duration of a waveform is measured from the differentiate between different types of source
point the signal first passes the threshold to the mechanisms.
point where it finally falls below the threshold and is
typically measured in microseconds. The energy of a

ACOUSTIC EMISSION TESTING 13.3

 Table 1. Description of acoustic emission features.

Feature Description Units

Hit based features

Time of hit Time at which signal crosses the threshold. microseconds (µs)
Amplitude Maximum or minimum signal excursion during hit. Reference is decibels (dBAE)
1 µV before preamplifier equals 0 dBAE.
Duration Time difference between first and last threshold crossings. microseconds (µs)
Acoustic emission counts Number of times signal crosses threshold during duration. counts
Measured area under rectified signal Area under envelope of rectified acoustic emission signal divided 10 µV·s per count
envelope (MARSE), or energy counts by number of counts.
Rise time Time interval between first threshold crossing and signal peak. microseconds (µs)
Counts to peak Number of counts between first threshold crossing and peak counts
amplitude.
Absolute energy (also available as Integral of squared voltage signal divided by resistance reference attojoules, 10-18 J (aJ)
time driven feature) over duration of waveform.

Time driven features

Average signal level (also available as hit Signal amplitude averaged during duration of hit. decibels (dBAE)
based feature)
Root mean square (also available as hit Root mean square voltage of acoustic emission signal. volts (V)
based feature)

Combination features
Average frequency Mean acoustic emission counts per cycle. kilohertz (kHz)
Initiation frequency Ratio of counts (to peak) to rise time at cracking. kilohertz (kHz)
Reverberation frequency Acoustic emission counts to peak per cycle. kilohertz (kHz)

Frequency features
Frequency centroid Center of mass of signal power spectrum. kilohertz (kHz)
Peak frequency Frequency at which power spectrum reaches its maximum kilohertz (kHz)
amplitude.
Partial power Percentage of total energy in power spectrum contained in percent ( )
predetermined frequency band.

State-of-the-art acoustic emission test systems

can extract four types of acoustic emission features:
Acoustic Emission Output
(1) hit based features, (2) time driven features, The typical output of an acoustic emission test
(3) combination features, and (4) frequency features. system is a plot of the hit information in graphic
Moreover, these systems can record parameters form. The principal plots are the hits-versus-time
recorded by other sensors, such as load, temperature, plots, the energy-versus-time plots, and the
pressure, strain, displacement, and rotation speed. amplitude histogram of the hits. These may either be
These external parameters are important for plotted for each transducer or be summed together.
successful acoustic emission test applications Figure 3 shows an example output of events versus
because they correlate with acoustic emission time for a sample being load tested. As the load
features and represent, directly or indirectly, the increases with time, the number of events (hits
stress state of the structure. Table 1 shows a list of analyzed as signals from the test zone) increases
acoustic emission features and their descriptions. until failure. Tracking the increased event count
makes possible a prediction of failure onset.

13.4 AEROSPACE NONDESTRUCTIVE TESTING

 Figure 3. Typical output of acoustic emission test showing events versus time.

16

4000
14
3500
12 Load
3000
Number of events  1000

10
2500

Load (lbs)
8 Events
2000
6
1500

4
1000

2 500

0 0
0 5 10 15 20 25 30 35 40 45 50 55 60 65
Time (s)

When multiple transducers are used, the timing of

hits can be analyzed to infer source location. The Figure 4. Example of plotting location of events when six sensors are
emission event sends a signal in all directions that spaced on 16 m  3 m composite spar section under test.
may be detected by multiple transducers. Figure 4
shows an example of plotting the location of events
when six sensors are spaced on a 16 × 3 m
composite spar section under test. The vertical axis
in Figure 4 is the energy in the event.

3000
2400

1800
Energy

1200

600

0 3.0
2.5
4
2.0
8 1.5
1.0
12
0.5
16 0

ACOUSTIC EMISSION TESTING 13.5

 In practice, it is observed that once a given load effect, each acoustic emission signal may occur only
has been applied and the resulting acoustic emission once. Therefore, acoustic emission tests have a now
has ceased, additional acoustic emissions will not or never quality. In this sense, acoustic emission
occur until that stress level is exceeded, even if the testing is at disadvantage when compared to
load is removed completely and then reapplied. This methods that can be repeated by different operators
behavior is known as the kaiser effect (Kaiser 1953). or with different instruments, without affecting the
Figure 5 demonstrates a basic plot of acoustic structure or the discontinuity.
emission test history showing cumulative energy At points D, E, and F, another effect called the
versus load. As the load is increased from points A felicity effect is shown. At point F, the applied load
to B, the acoustic emissions increase. From B to C, is high enough to cause significant emissions even
the emissions do not increase and do not increase though the previous maximum load at D was not
again until B — the kaiser effect. A second reached. This phenomenon is known as the felicity
occurrence of the kaiser effect is shown between C´ effect, and the felicity ratio is the ratio of load at
and B.́ Load levels that have been previously which emission restarts relative to the previous load.
exerted on a material do not produce acoustic A felicity ratio equal to 1 is an indication that
emission activity. In other words, discontinuities damage has not been created, whereas a felicity ratio
created in a material do not expand or move until lower than 1 indicates the presence of damage.
that former stress is exceeded. Because of the kaiser

Figure 5. Example of emission versus load plot to show kaiser and felicity effects.

2.0  105
Cumulative acoustic emission energy (energy counts)

1.5  105

E F Felicity ratio = 0.986

1.0  105

D
5.0  104

C B Felicity ratio = 1.00

C
B Felicity ratio = 1.06
0 A
10 20 30 40 50 60 70 80 90 100
Ultimate load (percentage)

13.6 AEROSPACE NONDESTRUCTIVE TESTING

 Acoustic Emission Test Techniques PART2

Acoustic emission techniques are typically used

to evaluate materials and structures for failure. The
Time Based Feature
test techniques involve the loading of the structure, Techniques
sensor configuration, and data interpretation. The Another category of acoustic emission test
structure loading depends on the type of structure techniques is based on time based features collected
and its operational characteristic. The quantity of at established time intervals regardless of the
sensors and their placement depend on the fidelity presence of acoustic emission test hits. These
desired, accessibility, and practical considerations of techniques use average signal level or root mean
channel cost and weight. Sensors are usually square signal values, typically with one or more
mounted by gluing to the structure, more often than external parameters to determine when some well
not with hot glue. characterized process is out of its normal behavior.
Applications such as leak detection and valve failure
are good examples of these techniques.
Loading
The structure under acoustic emission testing
requires some form of stress loading for the failure Pattern Recognition Methods
mechanisms to emit the mechanical waves to be With the availability of significant computer
sensed. In aircraft structural tests, stress loading is capabilities at relatively low cost, numerical
accomplished on loading frames. The acoustic techniques such as pattern recognition are becoming
emission test data acquisition system usually more common in nondestructive testing for
includes a channel for the input of the mechanical analyzing complex problems or automating the
load values. For aircraft in service with on-board evaluation (Emel 1992; Chen 1988). The analysis of
sensors, the flight loads are the loading mechanism. acoustic emission features is only one approach to
For fuselage structure tests, the load may be the pattern recognition. Other approaches start with
pressurization cycles. These may or may not be waveforms.
measured by on-board strain sensors. In other In most practical test applications as well as
structures, such as pressure vessels, the loading is commercial software implementations, pattern
pressurization. recognition is treated as a classification process that
assigns an input to a category called a class. By
restricting the problem in this way, general
Hit Based Method techniques have been developed that satisfy the
A common acoustic emission test technique is the needs for test data analysis, interpretation, and
detection of hits, whether defined in terms of evaluation. Within this framework, patterns in the
amplitude, energy, counts, or duration. Table 2 same class exhibit similar properties and are called
(Carlyle 1989) shows how the signals detected may ―similar.‖ Patterns in different classes exhibit
be sorted as acoustic emission test parameters and different properties and are called ―dissimilar.‖ The
how those parameters are related to fiber reinforced definition of class properties for the pattern
plastics failure mechanisms. Hill (Hill 1992) has recognition model is a key issue whereas measures of
characterized the three primary failure modes in similarity and comparison depend on the complexity
composites materials using the magnitudes of of the problem and the information available. Data
amplitude, duration, counts, rise time, counts to peak, are classified by two kinds of pattern recognition:
and energy associated with each acoustic emission supervised and unsupervised.
hit in fiberglass/epoxy pressurevessels.

Table 2. Acoustic emission test parameters and associated failure mechanisms in fiberglass/epoxy pressure vessels.

Acoustic Emission Transverse Matrix Longitudinal Matrix

Delaminations Fiber Breaks
Parameter Cracking Cracking

Amplitude low medium high low to medium

Energy low medium high very high
Counts low high high medium to high
Duration short long long short to medium

ACOUSTIC EMISSION TESTING 13.7

 1. Supervised pattern recognition involves a similarity. In the absence of prior knowledge for
learning process where each new, unknown set recognition, as is often the case in acoustic
of data is processed and classified into known emission testing, pattern recognition techniques
classes by comparing its features to a database are used unsupervised. In such cases, the
or by applying rules derived from the learning number of classes or categories must be
process. The classifier design proceeds by estimated and data must group meaningfully as
learning from examples, and the designer has a training set during classification.
previous knowledge about the number of classes What makes pattern recognition problems
as well as a set of known examples (called a difficult is that there can be a large degree of
training set) to be used in the classifier design. variability of inputs that belong in the same class,
Different supervised algorithms might be used, relative to the differences between patterns in
depending on the complexity of the problem, different classes. Some data are not really separable.
starting from the simplest minimum distance Users should understand that pattern recognition is
classifier to complex neural networks (Batchelor simply a tool for evaluating the problem and
1974; Tou 1974; Pao 1989). designing the classifier. However, measurement setup
2. Unsupervised pattern recognition is the process and data quality are always key issues. Also, in the
by which objects are classified into general case of unsupervised pattern recognition, the
groups according to their similarity. This process problem is not uniquely defined and multiple
does not require any previous knowledge or solutions should be expected. Choosing the best
database. Objects are classified into groups by solution is then up to the inspector through expert
comparing their features and deciding on their knowledge.

13.8 AEROSPACE NONDESTRUCTIVE TESTING

 Acoustic Emission Standards
and Specifications PART3

As with any other nondestructive test method, qualification. There are no acoustic emission test
standards and specifications in acoustic emission standards or specifications specifically for the
testing are used to define the requirement that a aerospace industry. However, some of those issued
product or service must meet. The three basic areas within other industries are used. Table 3 lists a
where acoustic emission standards and specifications number of these standards and specifications along
exist are: (1) standards for equipment, including with their publishers.
transducers; (2) processes; and (3) personnel

Table 3. Acoustic emission testing standards.

Organization Standards

ASME International ASME Boilerand PressureVessel Code: Section V, Nondestructive Examination. Article 13, Continuous
Acoustic Emission Monitoring

ASTM International E 650-97, Standard Guide for Mounting Piezoelectric Acoustic Emission Sensors
E 750-98, Standard Practice for Characterizing Acoustic Emission Instrumentation
E 751, Standard Practice for Acoustic Emission Monitoring during Resistance Spot-Welding
E 976, Standard Guide for Determining the Reproducibility of Acoustic Emission Sensor Response
E 1067-96, Standard Practice for Acoustic Emission Examination of Fiberglass Reinforced Plastic Resin(FRP)
Tanks/Vessels
E 1002, Standard Test Method for Leaks Using Ultrasonics
E 1106-86(1997), Standard Method for Primary Calibration of Acoustic Emission Sensors
E 1118-95, Standard Practice for Acoustic Emission Examination of Reinforced Thermosetting Resin Pipe
(RTRP)
E 1139-97, Standard Practice for Continuous Monitoring of Acoustic Emission from Metal Pressure
Boundaries
E 1316-00, Standard Terminology for Nondestructive Examinations
E 1419-00, Standard Test Methodfor Examination of Seamless, Gas-Filled, Pressure Vessels Using Acoustic
Emission
E 1781-98, Standard Practice for Secondary Calibration of Acoustic Emission Sensors
E 1932-97, Standard Guide for Acoustic Emission Examination of Small Parts
E 2075-00, Standard Practice for Verifying the Consistency of AE-Sensor Response Using an Acrylic Rod
E 2076-00, Standard Test Methodfor Examination of Fiberglass Reinforced Plastic Fan Blades Using Acoustic
Emission
E 2191, Test Methodfor Examination of Gas-Filled Filament-Wound Composite Pressure Vessels Using
Acoustic Emission

American Society for ANSI/ASNT CP-189, Standard for Qualification and Certification of Nondestructive Testing Personnel
Nondestructive Testing ASNT Recommended Practice No. SNT-TC-1A, PersonnelQualificationand Certification in Nondestructive
Testing

continued on next page

ACOUSTIC EMISSION TESTING 13.9

Table 3. Acoustic emission testing standards (continued).

Organization Standards

European Committee for DIN EN 14584, Non-Destructive Testing — Acoustic Emission — Examination of Metallic PressureEquipment
Standardization during Proof Testing; PlanarLocation of AE Sources
EN 1330-9, Non-Destructive Testing — Terminology — Part 9, Terms Used in Acoustic Emission Testing
EN 13477-1, Non-Destructive Testing — Acoustic Emission — Equipment Characterization — Part 1,
Equipment Description
EN 13477-2, Non-Destructive Testing — Acoustic Emission — Equipment Characterization — Part 2,
Verification of Operating Characteristics
EN 13554, Non-Destructive Testing – Acoustic Emission – General Principles

International Organization ISO 12713, Non-Destructive Testing — Acoustic Emission Inspection — Primary Calibration of Transducers
for Standardization ISO 12714, Non-Destructive Testing— Acoustic Emission Inspection — Secondary Calibration of Acoustic
Emission Sensors
ISO 12716, Non-Destructive Testing — Acoustic Emission Inspection – Vocabulary

Japanese Institute for JIS Z 2342, Methods for Acoustic Testing of Pressure Vesselsduring Pressure Testsand Classification of Test
Standardization Results

Japanese Society for JSNDI NDIS 2106-79, Evaluation of performance Characteristics of Acoustic Emission Testing Equipment
Nondestructive Inspection JSNDI NDIS 2109-91, Methods for Absolute calibration of Acoustic Emission Transducers by Reciprocity
Technique
 Acoustic Emission Test Applications PART4

Advantages/Limitations What Can Go Wrong?

Table 4 lists several of the advantages and Acoustic emission measurements involve some
limitations of acoustic emission testing for aerospace skill in the setup and interpretation of the signals.
nondestructive testing. Table 5 lists some of the issues that can arise and
what can be done to address them.

Table 4. Advantages and limitations of acoustic emission.

Advantages Comments

Acoustic emission testing is dynamic test Acoustic emission testing provides response to discontinuity growth under imposed
method. structural stress. Static discontinuities will not generate acoustic emission signals.

Acoustic emission testing can detect and Because only limited access is required and the structure acts as the conduit for signal,
evaluate discontinuities throughout entire discontinuitiesinaccessible to othernondestructivetestmethodsmay be detectedwith
structure during single test. acoustic emission testing.

Acousticemissiontestingrequiresonly Acoustic emission sensors can be mounted far from critical discontinuity zone.
limited access.

Structures inspected during service require Acoustic emission testing is performed during system operation or as part of test program .
little or no downtime.

Limitations

Acoustic emission testing requires that Acoustic emission testing is most often performed as part of planned mechanical testing.
system be in stress.

Noises from both structure and Sensor placement, frequency filtering, and time gating can be used to reduce the effects of
environment require interpretation. noise.

Monitoring must be continuous. Acoustic emission test system must be turned on during stressing to detect hits that
represent damage. If acoustic emission test system data acquisition is interrupted or not
functioning during any loading, then emissions that would help predict failure may be
missed.

Number and placement ofsensors affects Sensor sensitivity to small features is affected by number of sensors, their distance from
sensitivity. location ofinterest, theinterveningmaterial types, and sensordampening.

Sensors add weight and complication to For inservice aerospace application, test system will add weight and operational cost.
structure.

Acoustic emission testing requires Acousticemissiontestsignal measurement sensitivity and interpretationdependonsystem

experience to set detection thresholds. parameters such as gain and threshold. Gain too high or threshold too low will saturate
the data base; gain too low or threshold too high may miss critical features.

ACOUSTIC EMISSION TESTING 13.11

Table 5. What can go wrong with acoustic emission testing.

Issue Comment Options to Check or Consider

System sensitivity Thresholds too low can make system too sensitive to Determining of optimal threshold avoids excessively
background noise; thresholds too high can make limited sensitivity. Finding this setting may require one
systeminsensitive to emissionsofinterest. selectionof adequate sensortype andfrequency range,
as well as pencil lead break studies in test material.

Sensor sensitivity Poor sensitivity can be caused by insufficient Check amount and quality of couplant. Confirm enough
coupling between sensor and material, by lack of contact area between sensor and material, and check
contact, and by preamplifier orsensor failure. whether preamps and sensors are in correct working
order.

Sensor placement Lack of coverage can be caused by inadequate Attenuation measurements using pencil lead breaks
separation between the sensors. must be performed to space sensors optimally.

Noise External noise sources — such as mechanical joints, Before initiating of test, check for potential noise
hydraulic actuators, electromagnetic interference sources and use dampening materials wherever
from motors or switches, and others — can possible. Choose appropriate highpass filters in system
contaminate or prevent acquisition ofgood acoustic or preamps.
emission data.

External parametrics Veryimportant for correlation ofstress with In any test, have access to external parametric
acoustic emission. measurement that allows monitoringofload, directlyor
indirectly.

Loading Load rate can affect test quality and of amount of Plan ahead for loading schedules depending on failure
data generated. Moreover, because of kaiser effect, loads previously observed, on design parameters, or on
testcannot be repeated if overload occurs. service conditions.

Source location Sourcelocation depends highlyon threshold of Verify that wave speed usedcorresponds to mostlikely
detection and on wave speed. wave mode to break threshold.

Examples Figure 6. Halon fire extinguisher bottle in test

oven.
Pressure Bottles
The acoustic emission test of pressure bottles is a
useful example, in particular halon fire extinguisher
bottles (Beattie 1998). Halon 1301
(bromotrifluoromethane, CF3Br) is a halide gas used
to suppress fire. By increasing the pressure in the Oven
bottle and plotting acoustic emissions, an evaluation
of the bottle quality can be performed. The acoustic Bottle
emission tester for airline halon 1301 (CF3Br) fire
extinguisher bottles was developed at Sandia
National Laboratory under the sponsorship of the
Federal Aviation Administration and the Air
Transport Association. The bottles are constructed of
an austenitic, high manganese, stainless steel alloy
(Unified Numbering System S21900). The means of produced by the same growing discontinuity will
pressurization for these bottles is heating in an have their sources all located in the same region on
industrial oven up to a surface temperature of 66 °C a specimen. Therefore, one should search for a
(150 °F). Figure 6 shows a bottle being placed in an spatial clustering of the emissions. In this program,
oven for testing. as each event is located, it is checked to determine
The acoustic emission halon bottle test system whether it is a member of a cluster of events. Two
uses six channels. The software program controls the concepts are used in setting the failure criteria for
loading, collects and analyses the acoustic emission the halon bottle tester. First, the size of a
data, and evaluates the results, giving a pass or fail discontinuity is roughly proportional to the number
grade to the bottle at the end of the test. The of emission bursts produced. Second, the severity of
analysis is not based on the individual acoustic a discontinuity is proportional to the rate of increase
emissions but on the located acoustic emission of the number of emission bursts in the cluster with
events. A basic assumption is that acoustic emissions an increase in load. A simple method is used to
determine when the rate of bursts versus load is very high amplitude acoustic emission events were
increasing as a function of load. observed. These may correspond to ―pop-in,‖ a
A halon bottle recertification system was widely observed phenomenon associated with crack
designed for testing halon 1301 fire extinguisher initiation. Based on these observations, three feature
spheres measuring 125 to 400 mm (5 to 16 in.) in sets were developed: (1) the average time domain
diameter. The system was put into operation at the absolute signal amplitudes in successive 100 µs wide
beginning of the 21st century and has tested bands; (2) the fraction of energy between 100 and
thousands of halon bottles. The greatest financial 230 kHz, which was a minimum in the transducer
advantage resulting from the new test procedure is response function, the fraction between 230 and
in decreased bottle handling. After removal from the 300 kHz, and the fractions in successive 100 kHz
aircraft, the halon bottle is recertified immediately wide frequency bands up to 800 kHz; and
and returned to the same aircraft. Significant cost (3) standard acoustic emission parameters, such as
savings are gained from a reduction of inventory amplitude and rise time. Correlation plots between
and related recordkeeping and from the fact that the each of these features and the signal amplitudes and
halon need not be removed from the container. other parameters were then visually examined and
Depending on the size of the bottle, the actual test evaluated. Custom software can rank features by
procedure takes an average of 30 min. their ability to separate data classes and identify
significant features. The set of average signal
amplitudes and spectral features above 400 kHz were
Drivetrain Damage Classification discarded as they contained little useful information.
Like other active mechanical systems, a drivetrain Because of similarities in behavior, the remaining
produces noise during normal operation. Acoustic two bands above 230 kHz were consolidated into a
emission testing can be used on such systems to single feature. The final set of spectral features,
detect and analyze faults (Friesel 1999). Figure 7 is a therefore, consisted of the fraction Fl of total signal
photograph of the aft end of the intermediate energy between 100 and 230 kHz and the fraction F2
gearbox of the aft drivetrain of a BV-107™ of total signal energy between 230 and 400 kHz.
helicopter. An acoustic emission transducer with a
short waveguide bolted to the gearbox casing can be
seen near the center of the photograph. Two front
Figure 7. Photograph of intermediate gearbox
end techniques are used to reduce the mechanical
noise in the sensor. The first is to use a resonant shows acoustic emission transducer.
transducer coupled with a high pass filter. Acoustic
emission from damage can be detected at high
frequencies, and a transducer with a resonant peak
above about 100 kHz will preferentially amplify
these high frequency components. Subsequent high
pass filtering then suppresses low frequency noise.
The second technique for noise elimination is an
amplitude threshold that eliminates low amplitude
signals. The threshold is typically either set at a fixed
voltage level or allowed to float at some fixed level
relative to a continuously updated noise
measurement.
Finding a correlation between acoustic emission
signals and some independent measure of damage
type allows a training set to be created for classifier
development. The subsequent successful
development of a classifier relies on a uniqueness
principle: under given conditions, acoustic emission
from a given source must contain recognizable
characteristics unique to that source. Rather involved
processing may be necessary to identify these
characteristics, or in some cases the entire acoustic
emission waveforms from a given damage source Figure 8. Photograph of crack in pinion gear.
may be effectively identical and clearly distinct from
sources of noise (Friesel 1986; Melton 1982).
Figure 8 shows the growth of a crack from a
notched tooth of a pinion gear. The first stage of
data analysis consisted of visually reviewing
waveforms and spectra from the most significant test
intervals. Signals observed early in the test interval
appeared to be of lower amplitude and showed a
greater fraction of the total energy at lower
frequencies. Signals occurring later in the test, when
the crack presumably had advanced appreciably,
appeared to show both higher amplitudes and a
greater fraction of the energy at higher frequencies.
At the start of this second region, a small number of

ACOUSTIC EMISSION TESTING 13.13

 without reconfiguration or retraining. The system
Figure 9. Fraction of energy in frequency band 230 to 400 kHz versus requires that an appropriate set of baseline data be
amplitude for electric discharge machined notch/crack growth data. obtained. It is necessary to evaluate the system
performance on other components and drivetrains,
particularly in flight. This is necessary to verify the
38 observations reported here, to establish confidence in
the capabilities of the fault detection system, and to
identify components that cannot be effectively
monitored with this system. Additional study also
should be carried out to identify ideal sensor
Feature F2 (percent)

mounting locations and establish gain criteria.

Structural Testing
Acoustic emission testing is frequently used in
structural testing and provides useful information on
the development and location of crack growth. An
example is monitoring of notched full scale
honeycomb sandwich composite curved fuselage
8 panels during loading for detecting notch tip damage
42 76
initiation and evaluating damage severity in such
Amplitude (dB) structures (Leone 2008). The results show that
damage initiation at the tips of the notches and
damage progression along the panel could be
detected and located. These acoustic emission test
Figure 9 is a correlation plot of the fraction of F2 results were correlated with the deformation and
versus amplitude for signals recorded during the strain fields measured through strain
interval when crack growth was observed. The event photogrammetry at these notches throughout loading.
numbers illustrated in this figure are those of This correlation aids in interpreting the acoustic
randomly selected data points and indicate the emission test results. The fretting among the newly
sequence of detection of the signals. These numbers created fracture surfaces generated a large number of
show that the amplitude and fractional energy of low intensity acoustic emission test signals, but the
later events progress upward and to the right, high intensity signals at high load levels provided a
suggesting that crack severity is correlated with this good measure for anticipating incipient fracture.
progression. A similar plot of F1 versus amplitude Also, the acoustic emission test results located
showed an unsurprising progression downward and internal disbonding caused during panel fabrication.
to the right. The dense cluster of data in the lower The large number of low intensity acoustic emission
left of Figure 9 should at least roughly correspond to test signals generated from the disbonded regions
the data from previous test intervals when a large was associated with fretting among the disbonded
electric discharge machined notch was present and surfaces.
there was no crack growth. Composite honeycomb sandwich fuselage panels
Using a novel classifier, which detects damage, have been used successfully in general aviation and
provides an indication of damage severity, and commuting aircraft. Compared with conventional
identifies progressive damage, an acoustic emission metallic based structures, the composite honeycomb
based health and service monitoring system for sandwich panel exhibits significant weight savings
helicopter drivetrains has been developed. The and higher in-plane strength, stiffness, bending
system was shown to be effective for detecting rigidity, and stability. A well established concern
seeded fault damage and crack growth in SH-60™ when using these advanced structures, however, is
drivetrain components using data obtained from a their propensity to contain internal nonvisual
ground test facility. The classifier depends primarily damage. Such damage is normally a result of
on only two features and does not require extensive fabrication anomalies, such as core-to–face sheet
knowledge or analysis of the mechanical dynamics disbonding or impact damage caused by tool drop,
of the components. Also, the classifier performed runway debris, or bird impact. Such impact damage
effectively on dissimilar drivetrain components consists primarily of internal delamination, often

Table 6. Test matrix: damage scenarios, applied loads, and types of acoustic emission sensor.

Initial Damage Maximum Pressure Maximum Hoop Maximum Longitudinal

Panel No.
Scenario kPa (lb/in.2) Load, kPa (lb/in.2) Load, kPa (lb/in.2)

CP-5 250 mm × 13 mm (10 in. × 0.5 in.) 93.1 (13.5) 6900 (1000) 690 (100)
longitudinal notch
CP-6 250 mm × 123 mm (10 in. × 0.5 in.) 125.5 (18.2) 9300 (1350) 9300 (1350)
notch inclined at 45°

13.14 AEROSPACE NONDESTRUCTIVE TESTING

d
 the longitudinal direction, and in panel CP-6, the
Figure 10. General view of fullscale aircraft
notch was machined at 45°to the longitudinal
structural test evaluations and research test direction. The panels were subjected to various
facility. combinations of internal pressure, hoop loading, and
longitudinal loading.
The testing of the composite curved panels was
conducted at a facility of the Federal Aviation
Administration, Atlantic City International Airport,
NJ (Figure 10). The test facility subjected full scale
curved fuselage panels to combined loadings of
internal pressure, hoop, longitudinal, and shear
loads, representative of those experienced by an
aircraft under operating conditions. The system was
able to use either air or water as the pressurizing
medium; water was used in these composite panel
tests. Hoop and longitudinal edge loads were applied
to the panel by 14 hoop and eight longitudinal,
water actuated lever arms. Loads were applied to the
panels via 28 holes, each measuring 13 mm (0.5 in.)
diameter, along each hoop edge and via 16 holes
along each longitudinal edge.
The acoustic emission testing instrumentation
combined with matrix cracking and fiber fracture, configuration is shown in Figure 11. To determine
and local core crushing, as well as core–to–face the optimal sensor type for current and future tests,
sheet disbonding. Significant efforts were made since preliminary testing with three different sensors was
the 1970s to develop nondestructive test methods conducted during the test program of the six
(for example, ultrasonic and radiologic testing) for composite sandwich panels. The sensors operated in
various applications and requirements. These are the frequency range of 35 to 100 kHz, with a
mostly focused on detecting and mapping volumetric resonant frequency of 55 kHz and a 40 dB integrated
damage. A point of interest in damage detection in preamplifier. A group of eight sensors was placed at
composites is the ability to detect incipient damage 45°intervals, forming a 56 mm (22 in.) diameter
at the tip of an existing anomaly and to monitor its circle, with the notch at the center.
growth in real time. Figure 12 shows the acoustic emission event
Table 6 shows a test matrix of two panels that history as a function of loading and time. Each dot
contained a 250 mm (10 in.) long notch. The in the figure indicates a three-hit event. An acoustic
dimensions of the panels were 3.175 m (125 in.) long emission signal was recorded as an acoustic emission
and 1.85 m (73 in.) wide with an internal radius of event if it hit three nearby sensors within a
1.88 m (74 in.). The face sheets were made of prescribed time interval. An excellent correlation
0.216 mm (0.0085 in.) thick, plain weave, carbon exists between the notch tip damage progression,
fabric prepreg. The layup in the test section of the observed with strain photogrammetry, and the
panel was [45/0/45]. The honeycomb core was made corresponding acoustic emissions accumulated
of aramid, 19.1 mm (0.75 in.) thick in the central test during loading of the two panels. Two-dimensional
section. In panel CP-5, the notch was machined in location plots for both panels showed that damage

Figure 11. Acoustic emission system and sensor layout.

Y position versus X position
(all channels)
Data
acquisition
30
1
1 8 20
8 2
10
2 7
0 3
3 6 7
–10
4 5 6 4
–20
5
–30
–20 0 20

Notched composite panel Data analysis

ACOUSTIC EMISSION TESTING 13.15

 Figure 12. Load history and accumulation of acoustic emission events and their amplitudes generated
during loading of panels: (a) panel CP-5; (b) panel CP-6.

(a) 10.3 (1500) 110

100
8.3 (1200)

Acoustic emission event amplitude (decibels)

90
Longitudinal
Load, MPa (lb/in.) load
6.2 (900)
80

Hoop load
70
4.1 (600)

60

2.1 (300)
50

0 40
0 1000 2000 3000 4000
Time (s)

(b) 10.3 (1500) 110

100
8.3 (1200)

Acoustic emission event amplitude (decibels)

90
Load, MPa (lb/in.)

6.2 (900)
80

70
4.1 (600)

60

2.1 (300)
50

0 40
0 1000 2000 3000 4000
Time (s)

events corresponded to acoustic emission signals photogrammetry measurements. Further, it has been
that hit at least five neighboring sensors. Such shown that by eliminating the fretting emission and
five-hit events are, by definition, of relatively higher other unwanted noise, the location and damage
intensities. The pattern of the event clustered near progression could be clearly and confidently
the notch tips of the two specimens is similar to the determined. This was accomplished by recording the
actual damage progression measured from the strain high intensity events. However, because unwanted
photogrammetry data and the final damaged region emission generates emission of an intensity range
at the notch tip after fracture. similar to that generated by matrix dominated
The results of this study show that acoustic damage, it was concluded that the high intensity
emission testing can detect and locate damage events be filtered by accounting for acoustic
initiation and monitor its progression. Good emission signals that are strong enough to hit five
qualitative correlation was obtained between the neighboring sensors, that is, by accounting only for
acoustic emission results and the strain the five-hit events.

13.16 AEROSPACE NONDESTRUCTI

References
Batchelor, B.G. 1974. Practical Approach to Pattern Hill, Eric v.K. 1992. ―Predicting Burst Pressures in Filament
Classification. New York, NY: Plenum. Wound Composite Pressure Vessels Using Acoustic Emission
Beattie, A.G. 1998. Report DOT/FAA/AR-97/9, An Acoustic Data.‖ Materials Evaluation 50(12): 1439-1445.
Emission Test for Aircraft Halon 1301 Fire Extinguisher Kaiser, J. 1953. ―Conclusions and Results from Sound
Bottles. Washington, DC: Federal Aviation Administration Measurement in the Tensile Stressing of Metals.‖ Archiv für
(April). das Eisenhutten Wessen 24(1-2): 43-45.
Carlyle, J.M., and S. Vahaviolos. 1989. ―Inspecting Aging Kouvarakos, M., and E.v.K. Hill. 1996. ―Isolating Tensile Failure
Aircraft.‖ Quality [Magazine] 28(11). Mechanisms in Fiberglass/Epoxy from Acoustic Emission
Chen, C.H. 1988. ―High Resolution Spectral Analysis NDE Signal Parameters.‖ Materials Evaluation 54(9): 1025-1031.
Techniques for Flaw Characterization, Prediction and Leone, F.A., Jr., O. Ozevin, V. Godinez, B. Mosinyi, J.G.
Discrimination.‖ NATO ASI Series F44, Signal Processing Bakuckas Jr., J. Awerbuch, A. Lau, and T.-M. Tan. 2008.
and Pattern Recognition in Nondestructive Evaluation of ―Acoustic Emission Analysis of Full-Scale Honeycomb
Materials. Amsterdam, Netherlands: Springer, 155-173. Sandwich Composite Curved Fuselage Panels.‖ SPIE Smart
Emel, E., and E. Kannatey-Asibu. 1988. ―Tool Failure Structures and Materials and Nondestructive Evaluation and
Monitoring in Turning by Pattern Recognition Analysis of Health Monitoring 2008 [San Diego, CA]. Bellingham, WA:
AE Signals.‖ Journal of Engineering for Industry — SPIE(March).
Transactions of the ASME 110: 137-145. New York, NY: Melton, R.B. 1982. ―Classification of NDE Waveforms with
ASME International. Autoregressive Models.‖ Journal of Acoustic Emission
Friesel, M.A., and P. H. Hutton. 1986. ―Acoustic Emission Source 1(4): 266.
Identification Using Long Waveguide Sensors.‖ NDT Pao, Y. 1989. Adaptive Pattern Recognition and Neural
International 9(3 [June]): 203. Networks. New York, NY: Addison-Wesley.
Friesel, M.A. 1999. Contract DAAJ02-96-C-0038, SBIR Phase II Pollock, A.A. 1989. ―Acoustic Emission Inspection.‖ Metals
Final Report, ―Application of Acoustic Sensors for Handbook 17, 9th edition Materials Park, OH: ASM
Helicopter Monitoring.‖ Princeton, NJ: Physical Acoustics International, 278-294.
Corporation (April). Tou, J.T., and R.C. Gonzalez. 1974. Pattern Recognition
Principles. Reading, MA: Addison-Wesley.

ACOUSTIC EMISSION TESTING 13.17

 X
14 CHAPTER
CHAPTER

Bond
Testing

Contents
Part 1. Introduction, 14.2
Part 2. Bond Testing Methods, 14.5
References, 14.10

Contributors
14.1
David K. Hsu
Richard H. Bossi
Dennis P. Roach
 Introduction
PART1

Bonded structures are ubiquitous in aerospace Bond inspection during service is concerned with
applications; their inspection, maintenance, and the detection of damage in the form of disbonds
repair are important to the aerospace industry between adhesively bonded layers of delaminations
(Armstrong 2005). Bonded structures can take a in composite laminates. In the case of sandwich
variety of forms, including adhesively bonded structures, disbonds between the face sheet and core
metallic or composite joints and sandwich structures are of interest. These disbonds almost invariably lead
with various face sheets and core materials. to a reduction in the stiffness of the structure,
Honeycomb and foam sandwich structures are especially the contact stiffness on the surface. This
widely used on aerospace structures, especially as change in stiffness can be detected by a number of
light control surfaces such as ailerons, spoilers, nondestructive testing techniques. Two main causes
rudders, and trailing edge flaps. Bonds are being for delamination and disbond damage are low
applied in increasingly significant (primary and velocity impacts occurring during maintenance, such
secondary) structures for design, weight, and cost in as those due to dropped tools, and impact damage
manufacturing, and for repairs. As the application of caused by hail, runway debris, and bird strikes.
bonding for critical applications increases, the Depending on the face sheet thickness and stiffness,
quality of the bonding to ensure bond strength impact damage of honeycomb sandwich structures
increases in importance. may not leave visible indications on the surface
Bond testing during manufacture is concerned (referred to as barely visible impact damage). One of
with proper assembly of the adherends and adhesive, the nondestructive testing goals is therefore to
the geometric fitup, and bondline thickness. Basic characterize the severity of the discontinuities or
nondestructive testing is concerned with detecting damage in terms of size, shape, location, and
feature discontinuities such as voids, thinning, and severity. In applying the nondestructive test
unbonded regions. Advanced techniques that methods, accessibility of the inspection area plays a
measure modulus, stiffness, or nonlinear key role. Certain techniques, such as
characteristics of the interface are also of interest. through-transmission air coupled ultrasonic testing,
However, nondestructive testing methods do not can only be applied when two-sided access is
directly measure the adhesive strength in bonded available. For thick honeycomb sandwiches with
joints. The strength of a bond is measured in the only one-sided access, the detection of far side
plastic regime of material behavior and, therefore, is disbonds then becomes more challenging.
not a parameter that could be nondestructively Tables 1 and 2 list some of the typical
measured directly in the elastic regime. nondestructive testing and alternative approaches for
Nondestructive test methods can measure assessing bond quality that are applied to bonded
characteristics that under certain conditions can joints. The principles and the application of those
indicate a potential loss of strength, or a weakened techniques are noted along with issues that should
bond relative to a strong bond. Correlation of elastic be taken into consideration. In general, using more
(nondestructive) properties has not been consistently sophisticated techniques increases the sensitivity to
demonstrated for all forms of weak bonds that might finer feature characteristics at a cost of time and
be encountered in manufacture. Therefore, associated effort to acquire data and extract information.
tests and measurements may be required to assist
nondestructive testing as part of the process control
of bonding operations.

14.2 AEROSPACE NONDESTRUCTIVE TESTING

Table 1. Acoustic techniques and applications for bond line evaluation.

Technique Principles and Application Issues

Tap test Sound emitted from mechanical tapping on Subjectivebutgoodfordisbonddetectioninthin

structure is evaluated for changes. Detects structure, particularlyhoneycombwiththinskin
disbonds by change in acoustic dampening or (lessthanfiveplies).
mechanical stiffness ofstructure.
Bond testers (1 to 100 kHz); sonic and Low frequency pitch catch technique for Goodfordisbondanddamagedetection,including
mechanical impedance analysis sending and receiving pressure waves. Velocity, far side honeycomb structure. Frequency
mechanical impedance, and thickness affect optimization can enhance sensitivity.
measurement.
Resonance ultrasonic testing Low and mid frequency ultrasound using Good for disbond and damage detection,
(100 to 400 kHz) continuous waves or wave trains. Senses particularly in bonded metallic or solid laminate
changes in mechanical impedance by structures.
interactionoftestobject withtransducer.

Acousto-ultrasonic testing Mid frequency ultrasonic pitchcatch technique Stress wave factor has similarity to acoustic
with signal interpreted to extract stress wave emission analysis.
factor correlated to bond properties. Sensitivity
to weak bondshasbeenclaimed.

Pulse echo or through-transmission Detects voids and disbonds in adhesive joints Generally regarded as main technique for
ultrasonic testing (1 to 5 MHz typical) by changes in acoustic impedance at inspectionbecauseitprovesadhesivecontactwith
interfaces. Can be used for bondline thickness adherends. But interfaces can be acoustically
measurement and for porosity measurement in coupled without adhesion — “kissing”bonds.
bonded repairs.

Air coupled ultrasonic testing Low frequency, usually 50 to 400 kHz. Detects Used for inservice inspection of control surfaces.
delamination of skin-to-core bonds.

Transverse wave ultrasonic testing Usestransverse acoustic vibration to increase More difficult to implement than longitudinal
sensitivity to bondline features relative to (pressure) wave pulse echo ultrasonic testing.
conventional pulse echo or
through-transmission ultrasonic testing.
Ultrasonic spectroscopy Changes in spectral content of ultrasound can More difficult to implement than conventional
(1 to 5 MHz typical) be sensitive to changes in adhesive and ultrasonic testing. Consistentreliabilityhasnever
interface. Some data correlation to certain been shown. Variations in thickness will be
types of weak joints has been noted, detected.
particularly with degraded bonds.

Angle beam ultrasonic spectroscopy Extracts propertiesofbondedinterfacesuch as Requiressophisticateddataextraction, subjectto

(1 to 5 MHz typical) thickness, density, and modulus. Correlates to characteristicsoflaminateandadhesivematerials
changes in manufacturing parameters. Ismore and to theirshapeandvolume, all of whichmake
sensitive to those changes than conventional interpretation difficult.
ultrasonic testingis.

Guided wave ultrasonic testing Guided waves that travel in adherend or Difficulty in launching and receiving of wave into
adhesive are affected by subtle changes in bondline makes it difficult to distinguish subtle
boundary conditions. difference at weak bond.

Nonlinear acoustics This approach relies on detecting harmonic Difficult to implement and obtain reliable signals.
(500 kHz to 10 MHz) features in mixed acoustic signals where
kissingor weak bondtransfers energy across
joint differently in compression or tension
motion of acoustic waves. Nonlinear
measurement of velocity change is also
possible by testingsample underload.

Acoustic emission testing Senses emission of acoustic signals during [Link]

loading. Weak bonds typically have higherlevel be repeated.
and rate of emission of signals relative to
strong joints.

BOND TESTING 14.3

 Table 2. Various methods and applications for bond line evaluation.

Method Principles and Application Issues

Infrared thermography Infrared waves detect changes in heat transfer Detects disbonds and delaminations.
capability oftestobject. Weak bonddetection
would require some change in thermal
conductivityofweak versusstrong adhesive
joint.
Vibrothermography Vibration induces heatingat strained Most effective on cracklike interfaces.
interfaces.
Dielectric property measurement Measures changes in adhesive dielectric Applicable to environmental degradation changes
properties. that could weaken bonded interface.
Radiologic testing (X-ray) Detects volumetricfeatures. Can detect missing adhesive and adhesive
thickness variations.
Shearographic and holographic testing Stress (thermal or mechanical) is applied to Loading must affect interface, so shearographic
part and displaces surface. Disbonds and weak testing and holographic testing are best on thin
bonds are indicated by change in stiffness skins.
from differences in surface motion relative to
strong bonds.

High power ultrasound testing (not High power vibrations from low frequency Energyleveltestingthatwillnotdamagestrong
nondestructive testing) (20 kHz) ultrasound horn are used to stress bondscanbe difficultto [Link]
bonds. Weak bonds fail at lower levels of stress failure is notwellunderstood.
than strong bonds.
Stress waves (not nondestructive Localized dynamic proof test that sends tensile Method is subject to sizeandconfigurationof
testing) loading stress waves through interface. Waves bondedjoint.
of calibrated strength can find weak bonds by
creating small disbond.
Proof testing (not nondestructive Loads part enough to indicate joint will be safe Expensive to implement on full scaleparts. Can be
testing) in service. performed on extracted plugs.

Witness coupons (not nondestructive Sample is fabricated at same time as structural Test is not on part in service.
testing) bond by using identical materials and
processes or is cut from finished bond and
then mechanically tested to prove bond
strength.

14.4 AEROSPACE NONDESTRUCTIVE TESTING

 Bond Testing Methods PART2

The nondestructive test techniques for testing where k is the stiffness constant of the surface, m is
bonded structures may be divided into three the mass, and  is the time of contact. The above
categories: (1) mechanical techniques of testing; equation is based on a simple harmonic vibration
(2) sonic and ultrasonic techniques using low assumption for the tapped surface and provides a
frequencies (Cawley 1990), high frequency bond quick way to determine its local spring constant k
testing techniques, and conventional ultrasound independent of the tapper mass and largely
including specialized data acquisition and analysis independent of the tap force used. Physically, when
(Drewry 2009); and (3) emerging nondestructive test a surface is tapped with a different force, the
techniques with bond testing capability. The amplitude of the vibration changes, but the time of
mechanical techniques include the manual contact , related to the period of frequency of the
(qualitative) tap test, the instrumented (quantitative)
tap test, and portable load displacement devices such
as the elasticity laminate checker. The low frequency Figure 1. Force-versus-time diagrams of honeycomb sandwich panel:
techniques refer to several established test (a) undamaged region; (b) damaged region. Vertical axis shows
techniques that do not require a coupling fluid or gel
increments of 0.5 relative amplitude units; horizontal axis, of 100 µs.
between the transducer and the test surface; these
include mechanical impedance analysis, the (a)
membrane resonance technique, and the pitch catch
technique. In addition to the low frequency ―sonic‖
techniques, ultrasonic testing using the low end of
the frequency spectrum (typically less than 1 MHz) is
sometimes used on bonded structures as well. A
number of additional nondestructive test techniques
Force, 0.5 per division

have emerged and proven to be effective in the

inspection of bonded structures; these include
shearography (Newman 2005), thermography and
thermal wave imaging (Crane 2001), air coupled
ultrasound (Peters 2004), and stress wave testing
(Bossi 2002). Stress wave testing is one of a few
techniques that can potentially provide a measure of
bond strength, although strictly speaking it is not
nondestructive but rather a localized load test.

Tap Testing ∆t 0.00 µs 1/∆t ∞

Manual tap testing using a coin or other metal Time, 100 µs per division
piece is the most widely practiced inspection for
bonded structures and composites. Despite its (b)
qualitative nature, it remains the most convenient
and cost effective method for inspecting bonds.
Audible tap testing is based on discernment of the
frequency content of sound versus unsound
materials when subjected to a light tapping.
Force, 0.5 per division

Structures with reduced stiffness produce a duller

―thud,‖ with the human ear detecting the increased
low frequency content, while stiffer structures create
a higher, ―ringing‖ note. The physical basis of the
instrumented tap test is that the time of contact
between the tapper and the surface, that is, the width
of the force-versus-time curve, is a function of the
stiffness constant k and the mass m of the tapper
(Hsu 2000):

m
(1)   
k ∆t 0.00 µs 1/∆t ∞
Time, 100 µs per division

BOND TESTING 14.5

 vibration, remains approximately the same. For hearing based, manual tap tests of bonded
Figure 1a shows the time of contact response on an structures on aircraft, manufacturers recommend
undamaged part of a honeycomb panel tapped by various convenient tapper sizes and masses in their
an accelerometer. The larger amplitudes correspond service manuals. One difficulty with the qualitative
to taps of greater force. Similarly, Figure 1b shows manual tap test is that benign substructures such as
the time of contact response of an impact damaged ply overlap, core splice, ribs, and spars of a part can
portion of the panel as tapped by the same often lead to confusing audio responses that can
accelerometer. The damage in the honeycomb core hamper the making of a call. Over the years, a
causes a reduction of the contact stiffness k and number of instrumented tap test devices have been
hence leads to a longer time  of contact. The developed (Mitsui 2014, Georgeson 1996). To take
reason that  is largely independent of the tap force advantage of the visual judgment of the inspector, a
and vibration amplitude is the same as that of a computer aided tap tester was developed for
pendulum swing whose period is approximately producing tap test images based on the contact time
independent of the amplitude for small oscillations.  and stiffness k. Figure 2 shows three instrumented
It has been demonstrated using aircraft composite tap test devices, and Figure 3 shows a tap test image
parts that the stiffness k deduced from a tap test is produced by the computer aided tap tester on a
in agreement with the k determined in mechanical carbon fiber reinforced polymer honeycomb panel
load displacement tests (Peters 2001). containing engineered discontinuities. In an
extensive test program for composites that involved
both the conventional and emerging nondestructive
test techniques, the computer aided tap tester was
Figure 2. Three examples of instrumented tap test devices: found to have an 85 percent probability of detection
(a) hand held automated tapping device; (b) rapid damage detection for discontinuities of 38 mm (1.5 in.) in diameter
device; (c) computer aided tap tester. (Roach 2003). When a manual or instrumented tap
test is used in the inspection of honeycomb panels
(a) (either carbon fiber reinforced polymer or graphite
fiber reinforced polymer), the limit for face sheet
thickness is about eight or nine plies for good
results.

Mechanical Impedance Analysis

The mechanical impedance of a structure, defined as
the ratio of the applied force F to the resultant
velocity v, is a measure of the test object‘s resistance
to its own motion. Like the tap test, the mechanical
impedance technique also exploits the reduced
stiffness, hence reduction in the resistance to motion,
of a structure containing damage. For example,
(b) when the same force is applied to the face sheet of a
sandwich over a region of fractured core, the
resultant velocity will be higher and the mechanical
impedance will be lower than for a region of no
damage. Like electrical impedance, mechanical
impedance is also frequency dependent. In
commercial mechanical impedance instruments, the
probe consists of two piezoelectric crystals with a
driver positioned behind the receiver within the
same holder. The driver converts electrical energy
into sonic vibrations, and the receiver, being in
direct contact with the test surface, converts the
modified vibrations into electrical signals for
(c) processing by the instrument. In general, a
discontinuity will produce a signal with an
amplitude proportional to its stiffness with a
possible phase change. The displayed information
can be either an impedance plane (flying dot), a
meter deflection, or a horizontal bar graph. Alarm
thresholds can be used to provide audible or visual
warnings.

Low Frequency Bond Testing

Low frequency bond testing refers to bond testers
that operate below 100 kHz and are generally called
sonic bond testers. Sonic bond testers typically are
dry coupled: they do not require liquid couplant and
operate in the audible or nearly audible frequency

14.6 AEROSPACE NONDESTRUCTIVE TESTIN

 Figure 3. Tap test image of carbon fiber reinforced polymer honeycomb sandwich panel containing
artificial discontinuities.

19 (0.75) Contact time (µs)

38 (1.5)
57 (2.25) 607 to 648
76 (3.0)
95 (3.75) 566 to 607
114 (4.5)
133 (5.25) 525 to 566
152 (6.0)
171 (6.75) 484 to 525

mm (in.)
191 (7.5)
210 (8.25) 443 to 484
228 (9.0)
248 (9.75) 402 to 443
267 (10.5)
286 (11.25) 361 to 402
305 (12.0)
324 (12.75) 320 to 361
343 (13.5)
362 (14.25) 279 to 320
381 (15.0)
400 (15.75) 238 to 279
419 (16.5)
(0.25)
(1.25)
(2.25)
(3.25)
(4.25)
(5.25)
(6.25)
(7.25)
(8.25)
(9.25)
(10.25)
(11.25)

(13.25)
(14.25)
(15.25)
(12.25)

(16.25)
6
32
57
83

260
286

337
362
387
413
311

mm (in.)

range. Different techniques for transmitting and

receiving energy have been developed for low Figure 4. Two portable bond testers. Each
frequency bond test applications. Each technique design offers choice of techniques: mechanical
introduces a pressure wave into the specimen and impedance, high frequency (resonance), and
then detects the transmitted or reflected wave. pitch catch modes.
The pitch catch impulse test technique uses a
dual-element, point contact, dry coupled, low
frequency sonic probe. One element transmits
acoustic waves into the test part and a separate
element receives the sound. The sound propagates in
a complex wave mode across the test piece between
the probe tips. The return signals are processed and
the difference between the effects of good and bad
areas of the part along the sound path are analyzed
and compared. A complex wave front is generated
internally in the material as a result of characteristic
velocity, acoustical impedance, and thickness. The
time and amount of received energy is affected by
the changes in material properties, such as thickness,
disbonds, and discontinuities. The instrument
processes the received impulse and displays the
received information on a phase and amplitude
meter. Figure 5. Computer aided instrument for low
Figure 4 shows portable bond test instruments frequency bond testing.
being used on a test standard. Figure 5 shows a
computer aided instrument. This instrument operates
with a chirp between 3 kHz and 25 kHz, after which
an optimum frequency may be chosen from the
spectral response (Smith 2010). Scanning at the
selected optimum results in the best accuracy for
detection and sizing of discontinuities.

High Frequency Bond Testing

High frequency bond testing, also referred to as
resonance testing, uses interrogating frequencies of
25 to 500 kHz. It is similar in application to contact
ultrasonic testing in that a transducer with a hard

BOND TESTING 14.7

 wear face is acoustically coupled to the item under Shearography is implemented by comparing two
inspection using a liquid couplant. High frequency interference patterns on a detector plane, typically
bond testing uses special narrow band transducers, ―before‖ and ―after‖ an object is stressed. If the
which, when coupled to the item under test, produce stressing causes motion, and subsequent out-of-
a continuous sound field in the material. The test plane deformations, thereby changing the optical
material, in turn, provides a mass loading on the path, then the speckle patterns differ. These images
transducer, thereby increasing the transducer can be compared by subtraction or other algorithms
bandwidth as well as changing the transducer‘s to obtain an image of the object with fringe patterns
resonant frequency. Anomalies (such as disbonds) or superimposed. These fringe patterns can then be used
changes in material thickness result in changes to to identify the presence, size, and depth of disbonds
the transducer loading that cause changes in in a structure. Holographic techniques may also be
transducer resonance. These changes are used, although they are normally more difficult to
subsequently detected as differences in phase and implement than shearography (Heslehurst 2009).
amplitude of the electronic detection circuits. When disbonds or weak bonds (where the modulus is
Acoustic impedance changes can be thought of as a altered) are present, the surface over the poor bond
variation in the ability to transmit sound. With the region is able to displace a greater distance relative
probe and the material under test in contact, the to the surface over the strong bond locations.
changes in the material‘s acoustic impedance cause a Shearography or holography can be usefully applied
corresponding change in the electrical impedance of to detect these greater displacement regions.
the transducer, and these electrical impedance Shearography is particularly effective on thin skin
changes are monitored by the instrument. where it is easier for the surface to displace under
load.
Ultrasonic Techniques
Ultrasonic testing, discussed in its own chapter, is Thermography
a very common method for bond evaluation in both Thermography or infrared testing, discussed in
its basic form and in advanced techniques. the chapter on thermographic testing, is a
Through-transmission and pulse echo ultrasonic nondestructive test method that uses thermal
techniques are suitable to detect disbonds in many gradients to analyze the physical characteristics of a
cases. Through-transmission ultrasonic technique is structure. Because of the lateral thermal flow in
commonly used for sandwich structures. Nonlinear materials, thermography is useful for thin structures,
and spectroscopy techniques have tested for the particularly for skin over honeycomb core
bonds with the particular intent of extracting greater inspection. Thermography can be very useful for
information about the bond condition. Parameters detecting disbonds over metallic core and in thin
such as modulus, density, thickness, and resonant (<2.5 to 5 mm [0.1 to 0.2 in.] thick) top layer
frequency can be useful in the characterization of structures. When disbonds occur, the flow of heat
the consistency of the bonding process. Ultrasonic into the structure is altered and the surface
techniques such as spectral analysis, angle beams, temperature is different over the disbond relative to
transverse waves, and guided waves can be used to the bonded regions. For skin over honeycomb core,
assist in bond characterization. however, this disbond detection is problematic.
Air coupled inspection is an accepted technique Thermography is a wide area test method relatively
for inservice inspection of control surfaces for easy to implement in cases where the test materials
skin-to-core disbonds. The low frequency (50 to are suitable.
400 kHz) is effective despite the significant reflection
loss at an air-to-solid interface due to the acoustic
impedance difference between air and material. The Weak Bond Determination
advances in transducer technology and electronics No standard nondestructive test method can
are making air coupled ultrasonic testing a practical provide a value for the bond strength in a joint. A
technique for bonded structures in the field. There number of studies have tried to correlate a measured
are two types of air coupled transducers: parameter to strength (Alers 1977; Rokhlin 1981;
piezoceramic (disk or composite) and capacitive. Hagemaier 1985; Rokhlin 1986; Pilarski 1988;
Most of the field applications use piezoceramic type. Adams 1988; Yuhas 1988; Wegman 1989; Dickstein
Air coupled ultrasonic nondestructive testing 1989; Grady 1989; Cawley 1993; Scala 1995). These
implemented in the through-transmission mode with studies examined parameters such as bulk modulus,
transducers mounted on a yoke for aircraft shear modulus, nonlinear coefficients, density,
components affording two-sided access is discussed porosity, acoustic attenuation or transmission,
in the chapter on ultrasonic testing. and/or some combination of these characteristics. In
certain cases of bond degradation, a correlation can
be made. However, the problem remains unresolved
Shearography for the detection of the general weak bond condition
Shearography, discussed in another chapter, is a where the loss of strength may occur due to effects
wide area interferometric imaging technique that can such as contamination or out of process conditions.
detect micrometer sized displacements in the surface Fatiguing a bond by ultrasonic vibration has been
of a structure. Shearography equipment monitors the tested as a potential technique (Couchman 1979).
surface of a structure for any changes in the surface Proof testing and witness coupon testing remain
strain field. Stressing the material in the appropriate the standard methods for validating strength. Proof
way ensures that the subsurface anomalies are testing can be very expensive on full scale
manifested on the surface of the structure. structures. The full scale proof test is normally

14.8 AEROSPACE NONDESTRUCTIVE TESTING

required to be at 100 percent of limit load. An Stress wave loading and ultrasonic fatiguing are
alternative to a full scale test would be localized alternative approaches that have been successful for
stress testing at selected locations in the bonded localized testing of a bond joint (Bossi 2002; Bossi
joint. The localized test relies on the assumption that 2004; Arrigoni 2008; Bossi 2009). Stress wave
if a bond is strong at one location, the process is testing uses strong mechanical waves to create a
validated for strength at nearby locations, similar to tension load on the bond. If the bond is weak, the
witness coupons but performed on the structure that joint will fail at the test point. This failure can be
is actually used. The extent or range to which the detected by sensors in the test instrument or by a
relationship of a test at one location satisfies nearby follow-up nondestructive test technique, such as
locations depends on the variability of any specific ultrasonic testing. Although stress wave techniques
bonding process. In some cases, a hole is cut into the are not nondestructive, for nominal strength bonds
structure to perform this test. Plugs can then be tested at a load level below failure, the bond is not
tested in place or extracted for testing on a machine. damaged and the structure is ready for service.
The tested region must then be repaired.

BOND TESTING 14.9

References
Adams, R.D., and P. Cawley. 1988. ―A Review of Defect Types and Hsu, D.K., D.J. Barnard, J.J. Peters, and V. Dayal. 2000. ―Physical Basis
Nondestructive Testing Techniques for Composites and Bonded of Tap Test as a Quantitative Imaging Tool for Composite Structures
Joints.‖ NDT International 21(4): 208-222. on Aircraft.‖ AIP Conference Proceedings 509. Review of Progress
Alers, G.A., P.L. Flynn, and M.J. Buckley. 1977. ―Ultrasonic Techniques in Quantitative Nondestructive Evaluation 19B [Montreal, July
for Measuring the Strength of Adhesive Bonds.‖ Materials 1999]. Melville, NY: American Institute of Physics, 1857-1864.
Evaluation 55(4): 77-84. Mitsui. 2014. WP632 — Tapping Exfoliation Detector. Accessed January.
Armstrong, K.B., L.G. Bevan, and W.F. Cole, II. 2005. Care and Repair of [Link] Tokyo, Japan: Mitsui Engineering and
Advanced Composites, 2nd edition. Warrendale, PA: Society of Shipbuilding. New York, NY: Bittech.
Automotive Engineers. Newman, J.W. 2005. ―Holographic and Shearographic Applications in
Arrigoni, M., S.E. Kruger, A. Bouin, D. Levesque, M. Lord, and J.-P. Aerospace Manufacturing.‖ Materials Evaluation 63(7): 746-750.
Monchalin. 2008. ―The Use of Laser Doppler Interferometry Based Peters, J., J. Zachary, A. Nielsen, and D.K. Hsu. 2001. ―Comparison of
on a Fabry-Perot Etalon for Shock Adhesion Test Applied to Local Stiffness of Composite Honeycomb Sandwich Structures
Adhesively Bonded Materials.‖ 1st International Symposium on Measured by Tap Test and Mechanical Test.‖ Review of Progress in
Laser Ultrasonics: Science, Technology and Applications [Montreal, Quantitative Nondestructive Evaluation 20B [Ames, IA, July 2000].
Canada, July]. Boucherville, Quebec: National Research Council Melville, NY: American Institute of Physics, 1031-1038.
Canada, NRC Publications Archive. Peters, J.J., D.J. Barnard, and D.K. Hsu. 2004. ―Development of a
Bossi, R.H., K. Housen, and W. Shepherd. 2002. ―Using Shock Loads to Fieldable Air-Coupled Ultrasonic Inspection System.‖ Review of
Measure Bond Joint Strength.‖ Materials Evaluation 60(11): 1333- Progress in Quantitative Nondestructive Evaluation 23B [Green Bay,
1338. WI, July-August 2003]. Melville, NY: American Institute of Physics,
Bossi, R.H., K. Housen and W. Shepherd. 2004. ―Application of Stress 1368-1375.
Waves to Bond Inspection.‖ Paper 49-174, presented at SAMPE Pilarski, A., and J.L. Rose. 1988. ―Ultrasonic Oblique Incidence for
Spring Conference, Seattle, WA (May). Covina, CA: Society for the Improved Sensitivity in Interface Weakness Determination.‖ NDT
Advancement of Material and Process Engineering. International 1(4): 241-246.
Bossi, R.H., K. Housen, C. Walters, and D. Sokol. 2009. ―Laser Bond Pilarski, A., J.L. Rose, and K. Balasubramaniam. 1990. ―The Angular
Testing.‖ Materials Evaluation 67(7): 819-827. and Frequency Characteristics of Reflectivity from a Solid Layer
Cawley, P. 1990. ―Low Frequency NDT Techniques for the Detection of Embedded between Two Solids with Imperfect Boundary
Disbonds and Delaminations.‖ British Journal of NDT 32(9): Conditions.‖ Journal of the Acoustical Society of America 87(2):
454-461. 532-542.
Cawley, P., T. Pialucha, and M. Lowe. 1993. ―A Comparison of Different Roach, D. 2003. ―Improving In-Service Inspection of Composite
Methods for the Detection of a Weak Adhesive/Adherend Interface Structure: It‘s a Game of CATT and MAUS.‖ Paper presented at 7th
in Bonded Joints.‖ Review of Progress in Quantitative NDE 12 DoD/FAA/NASA Conference on Aging Aircraft (New Orleans, LA.
[La Jolla, CA, July 1992]. New York: Plenum Press, 1531-1538. September).
Couchman, J.C., B.G. W. Yee, and F.H. Chang. 1979. ―Adhesive Bond Rokhlin, S.I., M. Hefets, and M. Rosen. 1981. ―An Ultrasonic
Strength Classifier.‖ Materials Evaluation 37(5): 48-50. Interface-Wave Method for Predicting the Strength of Adhesive
Crane, R.L., T. Astarita, H. Berger, G. Cardone, G. Carlomagno, T. Jones, Bonds.‖ Journal of Applied Physics 52(4): 2847-2851.
M. Lansing, S. Russell, J. Walker, G. Workman. ―Aerospace Rokhlin, S.I., and D. Marom. 1986. ―Study of Adhesive Bonds Using
Applications of Infrared and Thermal Testing.‖ 2001. Nondestructive Low Frequency Obliquely Incident Ultrasonic Waves.‖ Journal of
Testing Handbook 3: Infrared and Thermal Testing, 3rd edition the Acoustical Society of America 80(2): 585+.
Columbus, OH: American Society for Nondestructive Testing, Scala, C.M., and P.A. Doyle. 1995. ―Ultrasonic Leaky Interface Waves
489-526. for Composite-Metal Adhesive Bond Characterization.‖ Journal of
Dickstein, P., Y. Segal, E. Segal, and A.N. Sinclair. 1989. ―Statistical Nondestructive Evaluation 14(2): 49-59.
Pattern Recognition Techniques: A Sample Problem of Ultrasonic Smith, N. R, J.R. Kollgaard, and L. Dickinson. 2010. ―Improved Flaw
Determination of Interfacial Weakness in Adhesive Joints.‖ Journal Detection in Honeycomb Sandwich Structure Using Pitch-Catch
of Nondestructive Evaluation 8(1): 27-35. (Low Frequency) Bondtesters.‖ Paper 55-190, presented at SAMPE
Drewry, M.A., R.A. Smith, A.P. Phang, D. Yan, P. Wilcox, and D. Roach. Spring Conference, Seattle, WA (May). Covina, CA: Society for the
2009. ―Ultrasonic Techniques for Detection of Weak Adhesion.‖ Advancement of Material and Process Engineering.
Materials Evaluation 67(9): 1048-1057. Wegman, R., and J.R. Mitchell. 1989. ―Evaluation of Adhesive Bond
Georgeson, G., S. Lea, and J. Hansen. 1996. ―Electronic Tap Hammer for Strength by Nondestructive Testing: Advanced Materials: The Big
Composite Damage Assessment.‖ SPIE Proceedings 2945, Payoff.‖ Proceedings of the 21st International SAMPE Technical
Nondestructive Evaluation of Aging Aircraft, Airports, and Conference [Atlantic City, NJ, September]. Covina, CA: Society for
Aerospace Hardware. Bellingham, WA: SPIE, 328-338. the Advancement of Material and Process Engineering, 196-209.
Grady, D.E., and M.E. Kipp. 1989. Structural Failure. New York: Yuhas, D.E., B.P. Dolgin, C.L. Vorres, H. Nguyen, and A. Schriver. 1988.
Wiley, 1-40. ―Ultrasonic Methods for Characterization of Interfacial Adhesion in
Hagemaier, D. 1985. ―Nondestructive Inspection.‖ Adhesive Bonding of Spectra Composites.‖ Interfaces in Polymer, Ceramic and Metal
Aluminum Alloys. New York: Marcel Dekker, 337-423. Matrix Composites: Proceedings of the Second International
Heslehurst, R. 2009. ―Optical NDT of Adhesively Bonded Joints.‖ Conference on Composite Interfaces [Cleveland, OH, June]. New
Materials Evaluation 67(7): 837-842. York: Elsevier, 595-609.

14.10 AEROSPACE NONDESTRUCTIVE TESTING

 X
15 CHAPTER
CHAPTER

In-Situ Testsfor
Structur
al
Health Contents
Part 1, Principles, 15.2

Monitori
Part 2, Methods, 15.4
Part 3, Applications, 15.5

ng
References, 15.14

Contributors Gary E. Georgeson

Richard H. Bossi
John C. Duke
Michael D. Fogarty 15.1
 ras

N Andrew P. Washabaugh

e
i
l

J
.
G
o
l
d
f
i
n
e

G
l
e
n
n

M
.
L
i
g
h
t
E
.
I
.
M
a
d
a
 Principles
PART1

Health monitoring of systems and structures is service or disassembled as during maintenance. The
of interest for management of technically complex implementation of in-situ nondestructive testing
assets, including condition based maintenance, structural health monitoring is subject to a number
prognosis, maintenance scheduling, and early of caveats for value (cost and weight), reliability, and
warning of imminent failure. Nondestructive sensitivity.
testing plays a critical role in the structural risk The principal difference between continuous
management of aerospace systems. Maintenance and monitoring and periodic inspection is that, in the
structural engineering performs periodic inservice latter, data are collected atregular intervals. This
inspections during the life of a structure to ensure its period is much briefer than the propagation to
continued safe operation. Sensors applied on board failure of any degradation mechanism expected in
the aerospace vehicle or system are commonly the service of the structure or system. Inservice
referred to as structural health monitoring (SHM) nondestructive testing is performed periodically over
devices and provide information to help assess and regions ofa structure toobtain a detailed check on
manage the status of the structure. The most the condition, most commonly to detect cracks of
common structural health monitoring systems subcritical size and to monitor their onset and
provide information about the structure, such as growth. Structural health monitoring approaches are
monitoring of loads in certain locations and of flight often conceived to continuously monitor a large area
dynamics. There is continued interest in structural of an aircraft with sensitivity. In the more near term
health monitoring information to detect cracking, and practical applications, the sensors are directed at
delaminations, disbonding, degradation, or other particular locations or ―hot spots.‖ The sensors are
damage that may occur in service. This regime of checked ona periodic basis as in-situ nondestructive
structural health monitoring is effectively in-situ testing. For structural health monitoring concepts
nondestructive testing that can detect and measure that involve global sensor evaluation of a structure,
damage (Stolz 2009). In situ is a Latin phrase the sensors needtobe several orders ofmagnitude
meaning ―in place‖ — in-situ inspections do not more sensitive than a local ―hot spot‖ sensor to
require that the aerospace system be removed from assess critical discontinuities from far away and to

Table 1. Nondestructive test sensors for in-situ structural health monitoring.

Physical Principles Mode Scale Purpose Sensors

Vibration passive structure global damage accelerometer

Strain passive in local component or environmental and mechanical resistance strain gages, fiber
structure fatigue optic
Acoustic emission passive in component or structure damage incident piezoelectric
Ultrasonic guided waves active structure damage piezoelectric, magnetostrictive,
electromagnetic acoustic,
others
Ultrasonic bulk wave active local structure crack detection and growth piezoelectric pulse echo or pitch
scattering catch
Electromagnetic active local structure crack detection and growth eddy current
monitoring
Leak active/passive local structure crack detection, service life, vacuum check
degradation
Temperature passive component environmental and chemical thermocouple, fiber optic
degradation
Chemistry active/passive component chemical degradation fiber optic
Moisture passive structure environmental degradation fiber optic

15.2 AEROSPACE NONDESTRUCTIVE TESTING

provide sufficient warning for removal from service measurements such as strain, temperature, and
(Duke 2004). For nondestructive test personnel, chemistry canprovideusefulinformationconcerning
in-situ nondestructive test techniques for structural the service condition of a structure. In the structural
health monitoring require an awareness of the health management plan, this additional information
structural discontinuity detection requirements and informs inspectors of potential structural damage.
the validation of the discontinuity detection Vibration analysis and strain monitoring are
sensitivity (Buynak 2009). potential approaches to assess conditions of an
In-situ nondestructive testing structural health aircraft through measurements at selected points.
monitoring sensors and systems are based on Acoustic emission detects changes in a structure near
physical principles of mechanical motion, the sensor. Ultrasonic and electromagnetic
electromagnetics, optics, and acoustics. Most techniques rely on nondestructive test principles for
nondestructive test methods offer techniques that active inspection at selected locations. Some cracks
can be used for in-situ nondestructive testing. canbedetected withleak testtechniques.
Table 1 lists physical measurements commonly Temperature, chemistry, and moisture sensing can be
considered. In-situ nondestructive test principles go useful for assessing degradation and for managing
beyond the conventionally applied nondestructive systems.
test methods for crack detection. Other

IN-SITU TESTS FOR STRUCTURAL HEALTH MONITORING 15.3

 Methods
PART2

In-situ nondestructive testing methods for inservice aircraft crack monitoring with mixed
structural health monitoring have been developed for success on such aircraft as the F-111™, VC-10™,
a variety of scenarios. The common methods are KC-135™, F-14™, and Cessna® aircraft (Carlyle 1999;
listed in Table 2. Demeski 1997). Chapter 13 discusses acoustic
Resistance and fiber optic (Guemes 2009) strain emission in more detail.
gages are useful for fleet surveillance, mission Ultrasonic sensors, mostly piezoelectric, are
planning, and condition based maintenance. Strain considered for a variety of structural health
gages are installed in military aircraft to monitor monitoring applications. The most common are
loads atvarious locations. However, because dataare localized inspection ofregions where repairs have
not required for missionreadiness, sensors might not been made or damage is expected. Some applications
bereplaced when they fail. The United States Navy of consideration are to use guided wave modes over
uses a strain based fatigue analysis approach for considerable distances to evaluate a structure (Guao
highly stressed low cycles-to-failure components 2007).
designed using a safe life approach (Iyyer 2010). The Electromagnetic sensors, particularly flexible
limitation of such an approach is that the remaining circuit eddy current, can be mounted on or near
life may be underestimated by as much as a factor of structures to continuously monitor locations of
10 because of material variation (Christodoulou interest, such as the growth of metal cracks below
2004). patch repairs.
Fiber optic gages have not been used in aerospace Comparative vacuum monitoring is an unusual
applications as much as other civil structures method of monitoring cracks in a structure. It uses
because of the bulkiness of the interrogation flexible plastic channels sealed against the structure
optoelectronic units (Volanthen 2009). and attached to a vacuum system. When a crack
Acoustic emission sensing is a common method occurs in the structure under the channels, the
for monitoring structures during structural tests. vacuum is lostand indicates a discontinuity.
Acoustic emission testing has been tested for

Table 2. Common nondestructive test methods for in-situ structural health monitoring.

Method Application Comments

Resistance strain gaging Measures load in structure. Data help to determine fatigue life in structure and to schedule
needed maintenance.

Fiber optic strain gaging Measures load in structure. Embedded in structure.

Acoustic emission testing Listens for impact damage. Spacecraft applications.

Ultrasonic testing Detects discontinuities. In-situsensors and periodicevaluation.

Electromagnetic testing Detects cracks and fatigue. In-situ sensors for continuousorperiodic evaluation.

Comparative vacuum monitoring Detect cracks and crack growth. Patches over regions of potential crack initiation and growth.

15.4 AEROSPACE NONDESTRUCTIVE TESTING

 Applications PART3

of failure (due to cracking, disbond, or delamination)

Advantages and Limitations by using the load displacement curve produced
Implementation of structural health monitoring during mechanical testing. Local strain gages can
for aerospace applications has been limitedbya provide information regarding where failure is
number of factors. As the structural health occurring if they are close enough, but they suffer
monitoring sensing and data interpretation from significant limitations in identifying specific
technology improves, these limiting factors will be details. Often, test engineers cannot verify where
overcome. Some advantages and limitations of damage is initiating or growing. In-situ
structural health monitoring arelisted in Table 3. nondestructive testing can address this issue.
Along with traditional mechanical testing, a
nondestructive test technician may be required to
Table 3. Advantages and limitations of structural periodically inspect locations of the structure under
health monitoring. test. In such cases, testing has to be stopped — that
is, the load is removed from the structure under test.
This load removal lets the nondestructive test
technician safely hold an ultrasonic transducer in
Advantages place to inspect the unit. However, for safety, an
Continuous monitoring operator is not allowed to hold a transducer to a
structure during loading: the load must be removed
Instantornear real timeinterpretation before the inspection. This process may leave a high
Global or local coverage demand test machine idle until nondestructive
testing is performed to check for growth of a crack,
delamination, or disbond. Moreover, the precision of
the dimensional measurement is limited by the
Limitations nondestructive test technique and by the technician‘s
Added weight to structure ability to discern small changes. In the twenty-first
century, improvements have been made in structural
Sensor operational reliability testing with the application of in-situ nondestructive
Sensor lifetime and repair testing. These systems increase the sensitivity,
Sensitivity to critical discontinuity criteria accuracy, and repeatability of measurements and
information that lead to detecting the onset of
Probability of detection reliability failure. Real time nondestructive testingofstructural
components throughout a testing regime can be
done using mounting techniques and associated
methods. The output is captured by a data
Structural Testing acquisition system and recorded for direct
An important application of in-situ comparison to the loading condition through
nondestructive testing is the structural testingof complete failure.
aerospace composites. The study of laminated,
bonded, and cocured composite interface behavior Sensor Configurations and Mounting
under load is important to the proper design and use
of these structures. As part of the evaluation, Although ultrasonic testing, using longitudinal or
structural analysis codes are used to predict locations lamb waves, is the most common method for in-situ
and load levels for failure initiation. Manufacturers nondestructive testing in composites, other methods,
of composites and of structures incorporating such as eddy current or microwave measurements,
composites use various mechanical test instruments are not excluded. Pulse echo, through-transmission,
to stress representative specimens to failure in order and various pitch catchultrasonic techniques canbe
to validate and refine the structural performance usedwithcarefulmounting ofthetransducers. There
codes. These instruments range from small, are, in fact, a wide variety of sensor attachment
laboratory fixtures that produce a controlled static methodologies. Several effective mounts and
load to full scale multiple element systems designed configurations have been developed and patented,
to test an aircraft wing-to-body response to a variety which support in-situ nondestructive testing for
of loadingconditions. composite structural testing.
Accurately defining the onset of failure is In general, a mounting means must keep the
extremely valuable for structural analysis codes. sensor fully attached and coupled to the surface of
However, it is usually not possible to locate initiation the structure throughout the test. In many cases, the
in-situ mounts must maintain contact and coupling

IN-SITU TESTS FOR STRUCTURAL HEALTH MONITORING 15.5

 under high bending or torsional strains. The specific
Figure 1. Vacuum attached water coupled
in-situ mount types are selected to suit the particular
mount for focused ultrasonic pulse echo structural test, inspection location, and needed test
monitoring of disbond initiation at lap joint. data. Some examples of mounting configurations are
shown in Figures 1 to 3.

In-Situ Nondestructive Test

Systems
Depending on test requirements, there is a wide
range in the number of sensors mounted to support a
structural test. Multiplexers that can handle many
inputs are used for sequentially interrogating the
sensors at a predetermined repetition rate and are
either part of the ultrasonic test system or connected
toit. Subscale lab tests may require only one ortwo
sensor mounts and a low channel multiplexer. Full
scale structural testsoftenrequire 30, 60, ormore
channels and two or more banks of multiplexers
linked together. Of course, cabling between sensors
Figure 2. Spider mount for ultrasonic testing: and multiplexers is required, and routing options for
(a) top view; (b) bottom view. Elastic band large numbers of cables need to be considered
holds mounting cap down on ultrasonic carefully.
transducer screwed into center ring containing A data acquisition system that includes a
ultrasonic gel couplant. nondestructive testdevice, suchasanultrasonic test
instrument, is used to collect and store the sensor
(a) data. The sensor data are monitored in real time with
a technician watching updated plots on a general
user interface or a digital signal that indicates a
preselected condition. Figure 4 shows a simple setup.
The benefits of the in-situ nondestructive test
method can be summarized as follows.
1. Structural response can be directly compared to
part loading regime.
2. When high damage sensitivity is available, it
detects small discontinuities and damage
initiation or growth.
3. Monitoring of multiple locations during static
(b) and fatigue cycling is possible.
4. In-situ inspection does not require periodic
stoppages for inspection.
5. Automatic signal output does not require
constant operator observation.
6. In-situ inspection can be used to automatically
stop static and fatigue testing when damage is
detected.
7. Cost ofoperation is low, evenover extended
time periods.

Figure 3. Magnetically coupled ultrasonic pitch catch mount allows composite joint to be monitored for
internal initiation of growth during testing: (a) transducers are angled into joint and couple ultrasound
into it using material that conforms to corner; (b) magnets couple to one another and hold in-situ mount
firmly against joint.

(a) (b)

15.6 AEROSPACE NONDESTRUCTIVE TESTING

Crack Monitoring Figure 4. Eight-channel multiplexer is connected to ultrasonic test
Damage tolerant design concepts discussed in the instrument, cycling through each transducer during test. When preset
chapter on aerospace structures, as allowed by MIL- alarm is triggered, red light (shown) indicates disbond.
STD-1530C (DOD 2005), provide a mechanism to
determine the inspection thresholds and intervals for
crack detection on aircraft. Periodic inspection
intervals are planned based on the nondestructive
inspection capability for crack detection such that
any crack below the inspection sensitivity crack size
threshold would not growto critical length in fewer
than two inspection periods (Fahr 2014; Goranson
1993; Grandt 2004).
To aid such periodic inspections, in-situ structural
health monitoring methods have been suggested:
placing sensors at strategic locations on an aircraft to
reduce the cost of periodic inspections (Roach 2009).
Such systems have been demonstrated but not
implemented as of May [Link] other structural
health monitoring concepts, monitoring of cracking
has beensuggested to detectcrackinitiationand then
providing an alert so that action can be initiated to
remove the system or structure from service. This
monitoring regime requires either extensive coverage Figure 5. In-situ structural health monitoring of F-16 fighter for
of the structure or prior knowledge of where cracks cracks under patch repair: (a) patch in wheel well; (b) periodic system
are likely. Acoustic emission and vibration interrogation.
monitoring are two methods that could provide
extensive coverage. Acoustic emission with sparse
sensor spacing may detect crack initiation at (a)
relatively large distances, provided that a true crack
is distinguished from noise. Structural vibration
monitoring requires that the allowable crack size be
large enough to affect the global structure vibration
response. In-situ nondestructive testsystems target
specific regions for crack detection. These approaches
are topics for research as of May 2014.
An in-situ crack monitoring system was
demonstrated in the United States Air Force program
for hot spots (Haugse 2005; Malkin 2007).In this
case, an identified crack was repaired by the
application of a composite overlay patch. The goal of
the demonstration program was to show that an in-
situ sensor could be used to monitor that no further
crack growth would occur under the patch. A
periodic check of the sensors could be performed to
show that the repair was working and that the cracks
below the repair were not growing. Figure 5 shows
the location in the wheel well of an F-16 where a
repair patch was bonded over the cracked skin and
sensors werebonded onto the patch. To the patch was (b)
bonded a piezoelectric based sensor array that uses
guided waves to check for cracks from multiple send
and receive positions. After a specified number of
flights, the sensor circuitry can be checked
periodically to detect crack growth under the patch.
The ability to detect crack growth under repairs has
been demonstrated, but such a system has not been
implemented as of May 2014.

Aircraft Damage Monitoring

Guided ultrasonic waves, discussed in the chapter
on ultrasonic testing, often propagate in different
wave modes and orders and are known to be
dispersive — that is, their velocity depends on
frequency. Guided waves are generated using any
number of transduction methods such as piezoelectric
transducers or laser ultrasound. They propagate

IN-SITU TESTS FOR STRUCTURAL HEALTH MONITORING 15.7

 Figure 6. Magnetostrictive sensor monitoring data compared to reference data.

Reference waveform - reference waveform Plate end

Initial pulse
After placing a 51 percent thickness deep notch
13 mm (0.5 in.) long waveform - reference waveform

Front patch Back patch

Amplitude (0.1 volt/division)

25 mm (1 in.) long waveform - reference waveform

38 mm (1.5 in.) long waveform - reference waveform

51 mm (2.0 in.) long waveform - reference waveform

–150 0 150 300 450 600 750 900 1050

(–0.5) (0.5) (1.0) (1.5) (2.0) (2.5) (3.0) (3.5)
Distance, mm (ft)

through structures that are thinrelative totheir place is acceptable because the sensor is relatively
wavelengths and can be transmitted over relatively inexpensive.
long distances. Like other ultrasonic waves, guided
waves detect cracks when they interact with the
crack interface and reflect back to the sensor. Damage Detection
Typically, a tone burst signal at a desired frequency The technology can also be used to monitor
or a signal swept over a frequency range is used. cracks or bonded structures, suchasadhesively
bonded doublers. Figure 6 shows data for sawcut
notches from 0 to 50 mm (2 in.) long under the
Magnetostrictive Sensor Guided Wave 100 mm (4 in.) wide bonded doubler. These
Example monitoring data show that notches as small as
The magnetostrictive sensor is a guided wave 13 mm (0.5 in.) are detected.
device developed originally for pipe inspection and The technology has been demonstrated for
adapted later for aircraft. The sensor consists of a detecting fatigue cracking in longerons on an A-10
thin magnetostrictive strip material (usually an iron fatigue test specimen (Figure 7). Magnetostrictive
cobalt alloy) about 25 to 31mm (1.0 to 1.25 in.) long sensor data has been collected at a number of these
by 13 mm (0.5 in.) wide and 125 µm (0.005 in.) sites over a large range of fatigue cycles. The
thick, bonded to the part under inspection with an magnetostrictive sensor monitoring data indicated
excitation coil about the same size and about 250 sites where crack initiation and growth were
µm(0.01in.) thick placed over the iron cobalt alloy. observed at about 75 to 125 mm (3 to 5 in.) from the
The sensor is configured to apply a time varying magnetostrictive sensor. These indications were
magnetic field to the material under test and pick up verified witheddy current tests.
magnetic induction changes in the material caused The magnetostrictive sensor technology has also
by the guided wave. For aircraft applications, where been tested in the repair of an aluminum bulkhead
thetestrange is relatively short and discontinuity of the F-16 fighter to monitor the bond line of the
sizes to be detected are small, the test frequency doubler. The sensor is actually in the high velocity
ranges from about 200 kHz to 1 MHz. Detection is air stream, and an exciting probe is periodically
achieved through the reverse process. Monitoring is placed on the bonded iron cobalt alloy strip material
accomplished byleaving the magnetostrictive sensor to excite the guided waves through the bond. The
bonded in place. Data are collected and processed application is illustrated in Figure 8. The technique
periodically. Waveform processing can be simple has been tested on aircraft but not validated as of
because the sensor is fixed in position. Leaving it in May 2014.

15.8 AEROSPACE NONDESTRUCTIVE TESTING

 Figure 7. Magnetostrictive sensor locations on A-10 fatigue test specimen.

FS 395 FS 480

WS 23
10 1 15
12 14
17 2

18 3
9 8 4 16
11 WS 23

FS 130 FS 265 FS 380 FS 430 FS 541

7
13 5
WS 100
6

Region of air force installed wing lower cover repair

FS 405 FS 468
Magnetostrictive sensor locations
= wing lower cover
= fuselage upper longeron Magnetostrictivesensorsthathavefounddiscontinuities during
= fuselage upper auxiliary longeron fatigue cycling

Wing Leading Edge Figure 8. Magnetostrictive sensor strip on F-16 fighter. Periodically,
The loss of the space shuttle Columbia as a result when aircraft is on ground, exciting coil is placed over iron cobalt
of impact damage from foam debris during ascent sensor material bonded over repairs, and monitoring data are collected.
led the National Aeronautics and Space
Administration to investigate the feasibility of
on-boardimpactdetectiontechnologies. Duringthe
accident investigation, it was demonstrated that
catastrophic foamimpacts to thereinforced
carbon-on-carbon wing leading edge could be
detected withaccelerometers as wellasacoustic
emission sensors.
Based on this successful demonstration,
additional testing was performed during the shuttle
―return to flight‖ initiative to establish a database of
sensor system responses to a variety of impacts. The
impacting materials included foam, ice, and ablators
at ascent velocities as well as simulated
hypervelocity micrometeoroids and orbital debris. On
the basis of the availability of flight qualified
instrumentation, accelerometers were selected for
implementation of a system to detect impacts on the
leading edge of wings. Although the higher
frequency range of acoustic emission sensors offered
agoodratio of signal to noise, the development and
flight qualification of acoustic emission sensors and powereddata acquisitionand processing units with
dataacquisition hardware would have beenschedule wireless transmitters were also mounted inside the
and cost prohibitive. An array of 66 accelerometers wing cavity to record data from the accelerometers
during launch, ascent, and orbit.
was installed on the leading edge spar of each wing
in all of the space shuttles (Figure 9). There were Each data acquisition unit monitored three
roughly three sensors per panel spaced at three of accelerometers and one temperature sensor. Data
from the temperature sensors helped predict
the four corners where the panels were mounted on
remaining battery life in orbit within the cold wing
the spar. A much higher sensor density than required
provided redundancy in the case of sensor or data cavity, which was not environmentally controlled.
acquisition failure, as well as more data for analysis Because of limited telemetry bandwidth, the raw data
werefirstpreprocessed in the dataacquisition units
to detect and localize impact signals. Battery

IN-SITU TESTS FOR STRUCTURAL HEALTH MONITORING 15.9

 Figure 9. Shuttle wing leading edge impact detection system.

Accelerometer
Wing leading
edge with Reinforced
22 reinforced carbon-to-carbon
carbon-carbon acquisition panel
panels

Wireless
data
acquisition

Thermal sensor

to calculate rootmean square acceleration values Spektr module. Despite an exhaustive, sophisticated
and detect peaks that may represent impact signals. effort to locate the leak in the Spektr module once it
Summary files of root mean square acceleration peak was depressed, none of the available techniques were
values were wirelessly transmitted to an on-board successful. This experience has shown that the best
laptop computer and then relayed to the mission time to locate a leak is while the leakage is active,
control center at Johnson Space Center, Houston, because theinternal energy ofthe escaping gas is a
Texas. The summary peak information was analyzed significant source of physical phenomena to pinpoint
to represent impact events. For suspect impact a leak.
events, raw time history data windows were then There are many possible detection ideas that can
downloaded for further analysis. be envisioned for locating a leak, but in space
Results from this system were used along with applications, mostconcepts suffer fromapplication
data from video and radar during the mission to limitations and many practical considerations. It
specify locations on the wing leading edge for should be noted that most space structures are very
focused inspections with the orbiter boom sensing densely packed with equipment and supplies so it is
system. The system was successfully used on all most likely that these materials will be between an
shuttle flights after the Columbia accident. A number astronaut and a leak in the pressure shell. Hence,
of probable small impact events both during launch visual or thermal cameras that can see damage or
and ascent, as well as on orbit were detected, thermal cold spots indicating a leak are impractical
although none were damaging. because the visual field of the pressure shell is
limited. A 2007 NASA report reviewed many ofthe
commercially available systems that might help
Spacecraft Leakage locate leaks (Wilson 2008).
Leaks through the pressure wall generate acoustic
In manned spacecraft, leaks of atmosphere to and ultrasonic signals, both in the air and in the
vacuum are a hazard due to micrometeoroid orbital structure (Wilson 2008, Richards 2013). It should be
debris along with failures of seals and gaskets, pointed out that the signal from a leak sounds like
structural failures, and human error. A collision of a noise and does not present a definable phase signal
Russian resupply cargo spacecraft with the Mir space like a signal from an acoustic emission event or from
station‘s Spektr module on 25 June 1997 was due to micrometeoroid orbital debris impact, so common
human error, causing damage to the solar panels and location algorithms from acoustic emission are
creating a major leak in the Spektr module. The invalid forleak location. Currentultrasonic based
result was a depressurization and costly disuse of the

15.10 AEROSPACE NONDESTRUCTIVE TESTING

C
systems already exist for indicating leaks in
Figure 10. Wireless, self-contained distributed
chemical and petroleum industries (ADEC 1999;
Wilson 2008). Ultrasonic clamp-on flow meters that impact detection system module with four
use mass or volume balance techniques can be sensors attached.
attached to pipelines to detect small leaks. Because a
pipe acts as a waveguide, the signals can travel far
from the source. Another commercial system is based
on airborne ultrasound, where themicrophone is
directional. The operator has only to wave the
microphone around looking for the largest signal. In
a space vehicle, the problem requires a more
sophisticated solution: an integrated vehicle health
monitoring system automatically locates a leak in a
structure. Leak location in principlerequiresanarray
of sensors and algorithms to process the ultrasonic
signals.
In the case of airborne signals, they are highly
attenuated above 50 kHz in air but can travel
reasonable distance below 40 kHz (Lawrence 1982).
One concept for leak monitoring on the International
Space Station is based on arrays of directional Figure 11. Sensor array: (a) closeup of lead
microphones. Such asystem is simple inconcept, but zirconium titanate element mounted to printed
there are concerns that many of the structural circuit; (b) output from cross correlation
impediments within a space vehicle might make such showing direction to noise source with predicted
a system ineffective. This concept is challenging (red) and actual (white) vectors shown.
because the airborne sound might travel a tortuous,
reflected path in reaching the microphones, making
detected directions erroneous. (a)
In a space vehicle, much of the pressure shell is a
thin metal wall, so structure borne ultrasonic signals
travel as a first order flexural mode plate wave. That
type of wave can range in frequency from audible
frequencies up to the megahertz range. Ultrasound
about 5 m from a small leak has detected frequencies
as high as 400 kHz in structures like those in the
International Space Station. One benefit of this
method is that such a broad bandwidth would let the
system be optimized for the quietest frequencies of
the background noise. If the transmission
characteristics of the structure were known, then it
would be possible from amplitude data to infer noise
source location with a small set of sensors
(Tian 2009). One such sensor forimpact detection
consists of a small, low power, wireless, high speed,
four-channel digitizer (Figure 10)(Champaigne 2005;
Champaigne 2007). The system features a (b)
commercially available device wirelessly connected
0.25
to a computer.
Vertical spatial frequency (mm–1)

0.2
An alternative system (Holland 2005; Holland 0.15
2006) involves cross correlating one element with 0.1
each of the other elements in a 64-element array 0.05
(Figure 11). This process can be done serially on a 0
noise signal to map the direction of the noise. This –0.05
process does depend on having enough signal –0.1
attenuation to suppress sensitivity to reflections, –0.15
–0.2
which travel a long distance. From that map, vectors
–0.25
pointing to various noise sources can be generated.
–0.2 –0.1 0 0.1 2.0
Horizontal spatial frequency (mm–1)
Fatigue and Stress
Eddy current sensors and sensor arrays can be
permanently or temporarily placed near, on, or
within a structure at selected locations to monitor configurations that form an array in a plane parallel
damage and usage variables to support life to a test surface. One such sensor is a meandering
management decisions. Arrays are sensor constructs winding magnetometer described in the chapter on
with a common drive (primary winding) and electromagnetic testing. When used for damage
multiple one-dimensional or two-dimensional sensor detection and monitoring, embedded or surface

IN-SITU TESTS FOR STRUCTURAL HEALTH MONITORING 15.11

 Figure 12. Meandering winding magnetometer array: (a) mounted to monitor fatigue cracks at bolt hole
on aircraft structure; (b) fastener geometry with sensor location.

Smart washer
(a) (b) meandering winding
magnetometer rosette

Crack

mounted eddy current sensors perform interrogation

Figure 13. Four-directional magnetic stress gages: (a) nine gages similar to that of conventional eddy current sensors.
mounted to preexisting assembly installed around unmodified main These sensors can serve as smart washers and
rotor shaft; (b) array for main rotor shaft torque measurement. linear arrays for nondestructive testing in locations
that would otherwise require disassembly. Mounted
under a fastener collar, nut, or bushing retainer, a
(a) smart washer with a circular meandering winding
magnetometer array can detect cracks before they
propagate fromunder afastener. Only athin and
conformable sensor flies on the vehicle. Inspections
are performed by plugging in a portable data logger
to a connector at an easy access location. Because
eddy current sensors do not need to be on when the
crack initiates, no electronics need to fly on the
vehicle.
Practical fatigue monitoring using surface
mounted eddy current sensor arrays has been
available since the mid-1990s. Hundreds of coupon
tests and many demonstrations have been completed
(b) (Goldfine 1996; Goldfine 2003). The method is based
Location of meandering on the realization that fatigue cracks, during the
winding magnetometer
arrays early or advanced stages, affect the material‘s
electromagnetic properties, such as electrical
conductivity and magnetic permeability. Full scale
fatigue monitoring of critical structural members
Sensors was demonstrated in 1998 on P-3 aircraft (Goldfine
2001) in difficult-to-access locations. A rosette
configuration with a meandering winding
magnetometer is practical for some in-situ structural
health monitoring. Figure 12a shows an example of
how such sensors are placed under a washer.
Alternatively, the sensors can become a part of a
―smart‖ washer (Figure 12b).
four-directional For noncontact monitoring of dynamic stress, a
magnetic stress gages, network of eddy current based, four-directional
expanded view magnetic stress gages are installed on a preexisting
assembly around a rotor shaft (Denenberg 2008).
Figure 13 shows the configuration. The four
meandering winding magnetometer sensors are

15.12 AEROSPACE NONDESTRUCTIVE TESTING

C
stacked, with axes of sensitivity in four different
Figure 14. Four-directional magnetic stress gage measured torque

Four-directional magnetic stress gage estimate (N·m)

directions: 0°, 90°, 45°, and –45°. The layout,
orientation, secondary element size, and other versus applied torque.
geometrical properties are designed to make the
sensing elements of one sensor insensitive to the 5 000
magnetic fields of the other three sensors. This
insensitivity permits measurement of directional
4 000
properties, such as conductivity or permeability, in
four directions simultaneously.
Although each sensor has a different liftoff, the 3 000
differences are known and can be compensated for
by precomputed databases of sensor responses. The
installationhas three sensor strips, eachwiththree 2 000
four-directional magnetic stress gages covering a
third of the shaft‘s circumference. In each
four-directional magnetic stress gage, the magnetic 1 000
permeability and thus the stress are measured. The
stresses at all nine locations can then be used to 0
determine the torque, pitch, and lift loads. Figure 14 0 1 000 2 000 3 000 4 000 5000
shows a nearly linear relationship between the
hysteresis corrected magnetometer measurements of Applied dynamic torque (N·m)
permeability, the stresses in the rotor shaft, and the
resulting stresses between magnetometer estimated
torque and applied torque.

IN-SITU TESTS FOR STRUCTURAL HEALTH MONITORING 15.13

References
ADEC. 1999. ―Technical Review of Leak Detection Goranson, U.G. 1993. ―Damage Tolerance Facts and Fiction.‖
Technologies 1.‖ Anchorage, Alaska: Alaska Department ICAF 17, 1993 — Durability and Structural Integrity of
of Environmental Conservation (October), 31. Airframes [Proceedings of 17th International Committee on
Buynak, C.F., J.C. Brausch, and E.A. Lindgren. 2009. ―Lessons Aeronautical Fatigue Symposium, Stockholm, Sweden,
Learned from the Deployment of Inspection Procedures for June]. Warrington, UK: EMAS Publishing, for the
Complex Aircraft Structural Components.‖ Structural Health International Committee on Aeronautical Fatigue (ICAF).
Monitoring 2009 2. Lancaster PA: DEStech Publications, Grandt, A. 2004. Fundamentals of Structural Integrity, Damage
2181-2187. Tolerant Design and Nondestructive Evaluation. Hoboken,
Carlyle, J.M., H.L. Bodine, S.S. Henley, R.L. Dawes, R. Demeski, NJ: Wiley,20-22.
and E.v.K. Hill. 1999. ―Practical AE Methodology for Use on Guao, H., and J. Rose. 2007. ―Multifeature Optimization of
Aircraft.‖ ASTM STP 1353, Acoustic Emission: Standards Guided Wave Modes for Structural Health Monitoring of
and Technology Update. West Conshohocken, PA: ASTM Composites.‖ Materials Evaluation 65(10): 1035-1041.
International, 191-208. Guemes, A., and B. Soller. 2009. ―Optical Fiber Distributed
Champaigne, K.D., and J. Sumners. 2005. ―Wireless Impact and Sensing: Physical Principles and Applications.‖ Structural
Leak Detection and Location Systems for the ISS and Health Monitoring 2009 1. Lancaster PA, DEStech
Shuttle Wing Leading Edge.‖ IEEE Aerospace Conference Publications, 14-28.
(March). Haugse, E. 2005. USAF/AFMC AFRL Final Report, Contract
Champaigne, K.D., and J. Sumners. 2007. ―Low-Power F33615-00-D-3052. ―Air Vehicle Technology Integration
Electronics for Distributed Impact Detection and Program (AVTIP) Delivery Order 28: Structural Health
Piezoelectric Sensor Applications.‖ IEEE Aerospace Monitoring for Bonded Repairs‖ (March).
Conference (April). Holland, S.D., J.[H.] Song, D.E. Chimenti, and R. Roberts. 2005.
Christodoulou, L., and J.M. Larsen. 2004. ―Using Materials ―Leak Detection in Spacecraft Using a 64-Element
Prognosis to Maximize the Utilization Potential of Multiplexed Passive Array to Monitor Structure-Borne
Mechanically Complex Systems.‖ Journal of Materials Noise.‖ AIP Conference Proceedings 820, Quantitative
(March): 15-19. Nondestructive Evaluation [Brunswick, ME, July].
Demeski, R.J. 1997. Manuscript, Cessna T303 S/N T30300260, Warrendale, PA: American Institute of Physics, 885-891.
Installation and Design Substantiation for In-Flight Safe Holland, S. D., R. Roberts, D.E. Chimenti, and J.H. Song. 2006.
Life Monitoring Specimen. Daytona Beach, FL: ―An Ultrasonic Array Sensor for Spacecraft Leak Direction
Embry-Riddle Aeronautical University. Finding.‖ Ultrasonics 45 (1-4): 121-126.
Denenberg, S., T. Dunford, N.[J.] Goldfine, D. Grundy, C. Parker, Iyyer, N., and N. Phan. 2010. ―Durability Issues and
M. Robuck, D. Schlicker, Y. Sheiretov, and V. Silberstein. Management of Aging P-3C Aircraft.‖ 11th International
2008. ―Magnetic Stress Gages for Torque and Load Conference on Fracture [Turin, Italy, March 2005]. Red
Monitoring in Rotorcraft.‖ 64th Annual Forum [Montreal, Hook, NY: Curran Associates, 2078-2083.
Canada, April-May]. Alexandria, VA: American Helicopter Lawrence, B.D., and J.A. Simmons. 1982. ―Measurements of
Society. Atmospheric Attenuation at Ultrasonic Frequencies and the
DOD. 2005. MIL-STD-1530C, Aircraft Structural Integrity Significance for Echolocation by Bats.‖ Journal of the
Program. Arlington, VA: United States Department of Acoustic Society of America 71(3): 585-590.
Defense (DOD). Malkin, M. 2007. Final Report, Contract 699032L, Structural
Duke, J. 2004. ―Structural Health Monitoring: Before and After Health Monitoring for Bonded Repairs: F-16 Application.
Crack Birth.‖ Materials Evaluation 62(1): 53-55. Seattle, WA: Boeing Phantom Works.
Fahr, A. 2014. Aeronautical Applications of Non-Destructive Malkin, M., M. Leonard, M. Derriso, and E.D. Haugse. 2007.
Testing. Lancaster, PA: DEStech Publications, 11-13. ―Hot Spot Monitoring: Development of a Framework for
Goldfine, N.J., and D. Clark. 1996. ―Introduction to the SHM System Design.‖ Structural Health Monitoring 2007:
Meandering Winding Magnetometer (MWM) and the Grid Quantification, Validation, and Implementation 1
Measurement Approach.‖ SPIE Proceedings 2944, NDE [Proceedings of the 6th International Workshop, Palo Alto,
Techniques for Aging Infrastructure and Manufacturing CA, September]. Lancaster, PA: DEStech Publications, 786–
(December), 186-192. 792.
Goldfine, N.J., V.A. Zilberstein, D.E. Schlicker, Y. Sheiretov, Malkin, M., X.P. Qing, M. Leonard, and M. Derriso. 2007.
K. Walrath, [Link], and D. Van Otterloo. 2001. ―Structural Health Monitoring for Bonded Structural
―Surface Mounted Periodic Field Current Sensors for Repairs: F-16 Application.‖ Proceedings of the Third
Structural Health Monitoring.‖ SPIE Proceedings 4335, European Workshop: Structural Health Monitoring 2006
Advanced Nondestructive Evaluation for Structural and [Seattle, WA, July]. Lancaster PA: DEStech Publications.
Biological Health Monitoring (July), 20-34. 167-175.
Goldfine, N.J., V. Zilberstein, A.J. Washabaugh, D. Schlicker, Richards, W.L., E. Madaras, W.H. Prosser, and G. Studor. 2013.
I. Shay, and D. Grundy. 2003. ―Eddy Current Sensor ―NASA Applications of Structural Health Monitoring
Networks for Aircraft Fatigue Monitoring.‖ Materials Technology.‖ Presented at 9th International Workshop on
Evaluation 61(7): 852-859. Structural Health Monitoring [Stanford, CA, September].
Roach, D., and J. Pinsonnault. 2009. ―Use of Mountable Sensors
to Address Periodic Inspections for Cracks on Regional
 Aircraft.‖ Structural Health Monitoring 2009 1. Lancaster, Volanthen, M., and R. Caesley. 2009. ―Compact, Lightweight
PA: DEStech Publications, 175-182. Optical Fibre Bragg Grating Instrumentation Designed
Stolz, C., and M. Neumair. 2009. ―Structural Health Monitoring, Specifically for Aerospace Applications.‖ Structural Health
In-Service Experience, Benefit and Way Ahead.‖ Structural Monitoring 2009 1. Lancaster PA: DEStech Publications,
Health Monitoring 1. Lancaster PA, DEStech Publications, 126-133.
59-66. Wilson, W.C., N.C. Coffey, and E.I. Madaras. 2008. Technical
Tian, J., and E.I. Madaras. 2009. Technical Memorandum NASA Memorandum NASA TM-2008-215347, Leak Detection and
TM-2009-215948,An Amplitude-Based Estimation Method Location Technology Assessment for Aerospace
for International Space Station (ISS) Leak Detection and Applications. Hanover, MD: National Aeronautics and Space
Localization Using Acoustic Sensor Networks. Springfield, Administration, Center for AeroSpace Information
VA: National Technical Information Service (November). (September).

IN-SITU TESTS FOR STRUCTURAL HEALTH MONITORING 15.15

 16 CHAPTER

Leak
Testing
Contents
Part 1. Introduction, 16.2
Part 2. Leak Test Techniques and
Applications, 16.3
References, 16.12

Contributor
Kane M. Mordaunt
 Introduction
PART1

As with other nondestructive test methods in the

aerospace industry, leak testing is sometimes treated
Allowable Leakage
as a necessary evil or overlooked completely until Requirements
production processes reveal how necessary it is to Not all leak test techniques offer the same level of
inspect aircraft structures and systems before sensitivity when applied to aircraft structures and
operating. There are several significant advantages systems. To execute a leak test correctly, allowable
of leak testing structure and systems before leakage requirements must be clearly defined so that
admitting fluid. an appropriate test technique can be applied. In
1. Leak testing with a dry tracer gas rather than many cases, more than one technique may be
the wet operating fluid saves time when a leak available, and the end user may have a choice. In
is found. Draining, ventilating, and disposing of some cases, it may be necessary to investigate and
the operating fluid would be costly in time and derive allowable leakage rates in accordance with
money. engineering directives. For instance, a design
2. Tracer gas leak testing can be engineered to engineer knows that the aircraft fuel tanks must not
satisfy performance requirements about the life leak fuel. However, it is a mistake to state that no
of the product in a compressed time frame. For leakage is allowed because there is no such thing as
example, the aircraft coolant system may be a structure that will never leak. Given enough time
guaranteed to lose so little coolant that it will and money, one could develop more sensitive leak
only need to be topped off once every two test techniques, but in reality it would require
years. The leak test performed during aircraft infinite resources to meet a requirement that states,
production cannot take two years to verify that ―zero leakage allowed.‖
the system does indeed perform as promised. Defining a realistic and productive allowable
However, using a tracer gas leak test, it is leakage requirement entails a process of education
possible to validate such a requirement in and negotiation between the responsible NDT Level
about 1 h. III and the customer engineering personnel. For
3. Tracer gas leak testing is usually much safer example, after taking time to define terms, the Level
than testing with the operating fluid. It would III and the design engineer can agree on a realistic,
be very dangerous to construct an aircraft fuel allowable leakage requirement, ―no visible leakage.‖
tank and then fill it with fuel in a production No visible leakage can be translated into a number,
shop unequipped to deal with such flammability measured, and used as a go/no-go threshold. For fuel
hazards. The same is true for an airplane tanks, ―no visible leakage‖ really means that fuel
hydraulic system designed to operate at evaporates as fast as it leaks at typical operating
34.5 MPa (5000 lbf/in.2). One defective conditions. Laboratory testing corroborates with
connection is all that is needed to result in a
industry peers who agree that the ―no visible fuel
disastrous spill that could damage the airplane
leakage‖ requirement for aircraft fuel tanks is
and even injure employees. For these reasons
and others, dry, inert, low pressure tracer gas equivalent to about 10-5 Pa·m3·s–1 (10–4 std cm3/s)
leak testing is a good idea before admitting fluid tracer gas leakage. This number can be used to select
into aircraft structures and systems. leak test techniques exhibiting this sensitivity.
Just as for fuel tanks, all other aircraft fluid
transport systems go through a similar exercise to
define a usable allowable leakage rate that meets the
requirements of that system for the life of the
aircraft.

16.2 AEROSPACE NONDESTRUCTIVE TESTING

 Leak Test Techniques and PART2
Applications

Table 1 summarizes commonly used leak test be considered to further narrow the choices. The goal
techniques for aerospace applications. A table such is to provide the most productive and least costly
as this can help narrow the field of leak test test for the application while satisfying engineering
techniques for a given application. For instance, to requirements. The techniques in Table 1 are
meet a 10–5 Pa·m3·s–1 (10–4 std cm3/s) allowable discussed below to provide a better understanding of
leakage requirement for fuel tanks, the table shows their capabilities and limitations. In this discussion,
at least eight techniques to choose from. After the gross leakage includes leak rates above about
field is narrowed, other factors such as safety, cost, 10-3 Pa·m3·s–3 (10–2 std cm3/s); fine leakage is defined
test duration, facility requirements, and others may as anything smaller.

Table 1. Sensitivities of leak test techniques in aerospace industry.

Pa·m3·s–1
(std cm3/s)
Technique
102 10 1 10–1 10–2 10–3 10–4 10–5 10–6 10–7 10–8 10–9 10–10 10–11 10–12 10–13
(103) (102) (10) (1) (10–1) (10–2) (10–3) (10–4) (10–5) (10–6) (10–7) (10–8) (10–9) (10–10) (10–11) (10–12)

Ammonia/phenolphthalein

Bubble, film, ideal

conditions

Bubble, foam

Bubble, water

Bubble, immersion

Acoustic

Pressure change

Vacuum change

Fluorescent penetrant

Helium mass
spectrometer, sniffer
Helium mass
spectrometer, evacuation
Helium quartz membrane,
benchtop

Helium quartz membrane,

portable
Helium, thermal
conductivity

LEAK TESTING 16.3

 technique. When the leaking area is approximated
Figure 1. Ammonia/phenolphthalein reaction
this way, a bubble solution (see below) may be used
indicates leak in aircraft fuel tank. to pinpoint the leak precisely so it may be repaired.
However, before a leak can be repaired, the point of
internal leakage must also be identified. Techniques
for chasing leaks in fuel tanks are covered below, by
using air and soap bubbles and by using helium leak
testing.
After the ammonia bubble test, the
phenolphthalein indicator paint is rinsed with tap
water and the structure is allowed to dry naturally or
with high pressure air. Assembly may then resume.
The advantages and limitations of the ammonia/
phenolphthalein technique are listed in Table 2.
Ammonia/phenolphthalein leak testing is more
expensive and more hazardous than other leak test
techniques. However, the technique is still used as it
has been since the 1960s.

Bubble Leak Test

Ammonia, Air, and Phenolphthalein One of the simplest leak tests to apply is the
Leak Tests bubble technique. Typically, the test part is
pressurized to a specified pressure with air, and the
Anhydrous ammonia diluted in air to 5 to inspector then applies soap bubble solution as either
10 percent by volume is an effective tracer gas a thin bubble free film of liquid or thick foam over
mixture for leak testing aircraft fuel tanks or similar the site to be checked. Figure 2 shows application of
structures. It is rare to see ammonia used to leak test a bubble free thin film of soap solution to identify
other systems, such as tubes, because the ammonia small leaks. Figure 3 shows the use of foam to
tracer gas medium is corrosive and toxic. To use the identify large or gross leaks. If the part is small
technique, the ammonia/air mixture is injected into enough, it may be immersed in water, jet fuel, or
the tank at a specified pressure, typically 34 kPa other fluid, unless corrosion or other concerns
(5 lb/in.2 gage) or so and allowed to dwell for about prevail. Immersion bubble testing (Figure 4) is
1 h to obtain well mixed tracer medium in the tank. typically the most sensitive means of bubble testing
There are ways to shorten dwell time by using the but impractical for large parts.
wing fuel vents as distribution manifolds or by using Soap bubble solution may also be used in
a passive mixer, such as an air amplifier using the vacuum box applications where the test part is first
coanda effect, which mixes gases while the test coated with either thin film or foam bubble solution
cavity is being pressurized. and where a vacuum box or cup is then placed over
Once the ammonia tracer medium is mixed in the the area (Figure 5). Vacuum is slowly drawn while
cavity, a special paint containing phenolphthalein the surface is closely observed for bubbles. It is
acid/base indicator is sprayed on the external important to watch closely and initially apply
surfaces of the fuel tank. After the indicator paint vacuum slowly in case of a gross leak that can
has dwelled for a few minutes, a visual test identifies instantly blow away the bubble solution before
potential leaks. Where there is leakage, the ammonia, forming a bubble and before it can be observed. The
a strong alkaline, reacts with the phenolphthalein advantages and limitations of the soap bubble
and turns violet or purple. Figure 1 shows an aircraft technique of leak testing are listed in Table 3.
fuel tank where a leak has been detected with this

Table 2. Advantages and limitations of ammonia/phenolphthalein leak testing.

Locating leaks is easy and fast because of visual nature of test. Potential leaks are quickly narrowed down.
Advantages Sensitivity of test is perfect for aircraft fuel tanks. Fuel tank leakage is allowed as long as it cannot be seen, and
technique meets this requirement.

Anhydrous ammonia is extremely toxic to humans and explosive above about 12 percent by volume, so
extraordinary measures are required to control ammonia/air mixtures, to containpressurizedgases, and to ventilate
work areas in eventof spills. Evacuation proceduresmust also be implemented.
Ammonia is extremely corrosive, so no materials vulnerable to corrosion may be installed in fuel tank structure
before leak testing.
Limitations Becauseofhazardous nature of ammonia, tracer gases mustoften bescrubbedbefore releasing to atmosphere,
creating extra expense.
Phenolphthalein indicator paint is solvent based and volatile, so it must be applied in paint booth or controlled area
to contain vapors.
Rinse water must be processed through waste water systems.

16.4 AEROSPACE NONDESTRUCTIVE TESTING

Figure 2. Thin film liquid soap bubble test. Acoustic Leak Test
Like the bubble test, acoustic leak testing is
straightforward and relatively inexpensive. Typically,
the test object is pressurized or evacuated, and the
inspector then uses an acoustic leak detector to listen
for leakage that manifests typically in the high
frequency or ultrasonic range, so it is sometimes
called ultrasonic leak testing. Figure 6 shows an
acoustic leak detector in its carrying case. Pointed at
a leak source, the acoustic detector picks up the
sound of air rushing through the leak as it flows

Figure 5. Bubble test with vacuum box or cup.

Figure 3. Thick foam soap bubble test.

Foam

Figure 6. Acoustic leak detector.

Figure 4. Immersion bubble test.

Table 3. Advantages and limitations of bubble testing.

Locating leaks is easy and fast due to visual nature of test.

Sensitivity meets many allowable leakage requirements in aerospace applications.
Technique can be used in pressure tests and tests with vacuum box or cup.
Advantages
Retesting is easy if bubble solution dries.
Bubble tests are effective for identifying areas that need follow-up by quantitative techniques or for pinpointing
apertures for repair.

Technique is messy, may present slip hazard, and must be cleaned inside fuel tanks to prevent contamination.
Longer dwell times are needed to catch fine leaks and may not be practical for some production applications.
Gross leaks can easily be missed unless inspector is skilled in using film and foam to check suspect areas. Checking
Limitations can take more time to do correctly.
Checking large areas decreases sensitivity of soap bubble leak test due to production pressures and tedium that
causes inspectors to hurry.
Leaks may be missed unless lighting and visual access are ideal.

LEAK TESTING 16.5

 from or into the test object. Figure 7 shows a entire fluid system under test. This overall leak rate
vacuum bag applied to an aircraft passenger door for can be useful to determine if a system is tight
an ultrasonic leak test. Table 4 lists the advantages enough to put fluid in; if not, another leak test
and limitations of the technique. Acoustic leak technique must be used to pinpoint the leak source
testing has significant limitations in locating fine in the event of a pressure change leak test failure.
leaks but can help the inspector to quickly pinpoint The required accuracy of the pressure change leak
gross leaks larger than about 0.001 Pa·m3·s–1 test may need more sophisticated instrumentation
(0.01 std cm3/s). and increase the cost. The simplest pressure change
leak test requires (1) pressure regulator with valve,
(2) stopwatch, and (3) pressure connection to the
Pressure or Vacuum Change system under test. Figure 8 shows a schematic
configuration of these elements. The most accurate
Leak Test pressure change leak test requires, in addition to the
Pressure or vacuum change leak testing is previous equipment, (4) a temperature sensor, such
probably the most common leak test technique in the as a thermocouple, and (5) a barometric pressure
aerospace industry. It is simple to implement, is sensor. Temperature and pressure sensors are
inexpensive, and under the right conditions can be required for cases where the pressure change leak
fast and accurate. It is the only leak test technique test is performed over an extended time during
that provides an overall cumulative leak rate of the which ambient conditions may change. Changes in
ambient conditions can mask or amplify the leak
rate. The advantages and limitations of pressure or
Figure 7. Vacuum bag for leak testing of aircraft vacuum leak testing are listed in Table 5.
passenger door with acoustic leak detector.

Pressure Change to Define Allowable

Leakage Rate
Designing a pressure or vacuum change leak test
requires a defined allowable leakage rate, as
discussed above. Caution must be exercised to check
that the allowable leakage requirement is calculated
in accordance with the total volume V (cm3) of the
system under test, especially if the system design
changes, say, from one aircraft model to the next.
The allowable time rate P/t (Pa/s) of pressure
change is calculated with Equation 1, where P is
pressure, t is time, and V is volume. Q (Pa·m3·s–1 or
std cm3/s) is the overall allowable leakage rate from
the system:
P Q
(1) 
t V
The design engineer must provide a value for Q but
typically provides a value for P/t instead. It is
prudent for the Level III inspector, if provided a
value for P/t, to back calculate the allowable
leakage rate Q and confirm with the design engineer
that the rate is in accordance with design
requirements.

Table 4. Advantages and limitations of acoustic leak test technique.

Locating leaks is easy and fast because of audible nature of test.

Techniquecan beusedinpressure and vacuumtests.
Advantages
Technique can provide fast scans for gross leakage from distance.
Equipment is inexpensive and easy to use.

Sensitivity limits technique to gross leaks of about 10–3 Pa·m3·s–1 (10–2 std cm3/s) or larger.
Limitations False positive leak indications can be caused by leaks from known sources such as plant compressed air.
Leak indications may be maskedin noisyenvironmentswhereairoperatedtools are used.

16.6 AEROSPACE NONDESTRUCTIVE TESTING

 Equation 1 provides a numerical value for P/t
assuming Q and V are known. In this case, it is Figure 8. Pressure change leak test apparatus. To accomplish vacuum
necessary to iterate to define the allowable pressure change leak test, replace pressure regulator with vacuum source and
change P (atm) and the test duration t (s). This is use vacuum gage to measure vacuum in test object.
done by choosing either P or t and then calculating Pressure gage
the remaining unknown. Based on these values, the
operator must determine if the test is feasible, that is, Valve
that a desired pressure change can be accurately Compressed
air
measured in the allotted time with the available
equipment and instrumentation. Also, the operator Test piece
needs to confirm that the test duration is reasonable Pressure regulator
Safety
for the production shop. If the test is not feasible or shield
acceptable, another value for t or P must be
selected until a feasible set of test parameters is Timer
derived. Once an acceptable pressure change leak
test is defined, the operator must determine if
temperature or pressure compensation is necessary to
ensure accuracy. (3) Pabs  Pg  Patm
As a rule of thumb, if the duration of the
pressure/vacuum change leak test is 900 s (15 min) Here, Pg is the gage pressure reading and Patm is the
or longer or if the production environment is subject atmospheric or barometric pressure. At sea level,
to even minor temperature changes during the test, Patm = 101 kPa = 14.7 lbf/in.2.
temperature compensation may be necessary. If the
duration of the pressure/vacuum change leak test is Pressure Change with Temperature
1800 s (30 min) or longer and if the production
environment changes rapidly with the weather, Correction Only
pressure compensation may be necessary. The Where the internal system temperature changes
equations below define how to calculate time rate of and the external atmospheric pressure are constant
pressure change for the conditions noted. during the leak test, the time rate of pressure or
vacuum change is calculated according to Equation 4:
Pressure Change without Temperature P P T1 
or Pressure Correction 1  2 

Equation 2 represents the simplest time rate of (4) P  T2 

 t
pressure change leak test where only the pressure t
inside the test part is measured with time. It is
assumed that the internal system temperature T and
Here,
absolutT1 and
tempT2 are the (K
atures or R),
initial and calculated
final internal
e er
external atmospheric pressure are constant during according to Equation 5 or 6:
the leak test.
P (5) T K  T C  273
P1  P2
(2) 
t t where T is temperature in celsius (C) or kelvin (K) or:
where P1 and P2 are the initial and final internal
absolute pressures and t is the elapsed test time (6) T R  T F  459.67
(min). Absolute pressure Pabs (kPa or lbf/in.2) is
calculated according to Equation 3: where T is temperature in fahrenheit (F) or rankine
(R).

Table 5. Advantages and limitations of pressure or vacuum leak testing.

Thetechniqueprovidesoverallleakrate for systemundertest.

Advantages
Thetechniqueisinexpensive and easy.

The technique cannot locate leaks if the test fails. Another test technique must then be used.
Sensitivity is limited to about 10–3 Pa·m3·s–1 (10–2 std cm3/s) and greater in aerospace applications, so this
techniquewillnotsatisfy many allowable leakage requirements.
Pressure change tests of large volumes, such as large fuel tanks, require long test times and accurate temperature
Limitations and pressure sensors.
Mathematical corrections can be confusing to test personnel.
Design engineers often carry over pressure change requirements from one aircraft model to another without
allowing for different volume tested. The Level III inspector has to monitor this process closely to provide accurate
requirements and calculations.

LEAK TESTING 16.7

 Pressure Change with Temperature Helium Leak Test
and Pressure Correction Helium leak detection technology has advanced
When the internal system temperature changes significantly. The development of turbomolecular
and the external atmospheric pressure also changes pumps has made liquid nitrogen cryogenic pumps
during the leak test, the time rate of pressure or used in helium mass spectrometers obsolete. This
vacuum change is calculated according to Equation 7: advance helped helium leak testing become standard
in the aerospace industry and made it feasible for
P P   P P  T1 daily use in the production shop. Furthermore,
P turbomolecular helium mass spectrometers do not
(7)  1a 1g 2a 2g require advanced technical skills to operate. Helium
T2
t t detection technology was pushed even further by
Here, P1g and P2g are the initial and final internal diffusion technology whereby a heated quartz
gage pressures (kPa or lbf/in.2) and P1a and P2a are membrane acts as a filter allowing only atoms the
the initial and final atmospheric pressures (kPa or size of helium and smaller (such as hydrogen) to
lbf/in.2). pass through to the sensor. Fortunately, the
background concentration of hydrogen in the
atmosphere is insignificant and any signals are
Tracer Dye Leak Test safely assumed to be those of helium. Diffusion
helium detectors are small, light, robust, durable,
Tracer dye leak testing (generally using liquid sensitive, and economical — making them well suited
penetrant as a tracer fluid) is uncommon in for aerospace applications. They are limited to
aerospace industries. Table 6 lists the advantages and helium sniffing applications, but fortunately most
limitations of the tracer dye leak test. It is generally aerospace applications of leak testing are well suited
limited to troubleshooting to help trace leaks in for sniffing.
aircraft fuel tanks where other means, especially Multiple helium leak detection techniques are
soap bubbles, have failed and more expensive used in aerospace: (1) sniffer probe, (2) tracer probe,
technology, such as helium sniffing, is not available. (3) accumulation, (4) bell jar, and (5) hood. Although
Given alternatives, the inspector would put tracer all of these techniques may be used on demand, this
dye at the bottom of the list. The disadvantages of discussion will be limited to sniffer probe, tracer
dye penetrant leak testing, as noted below, make it probe, and accumulation helium leak detection
difficult to work with, but a skilled technician could techniques, all of which can be performed with a
learn to use it effectively. sniffing leak detector. These seem to be the most
Dye penetrant leak testing works by creating a common techniques in aerospace production. Table 7
pressure differential across the leak path by using lists the advantages and limitations of helium leak
either vacuum or pressure and applying a fluorescing testing.
liquid dye material to the high pressure side. The dye
material may be held over the leak path using a
small reservoir constructed of small vacuum cups or Sniffer Probe and Accumulation
plastic bagging (Figures 5 and 7). With pressure Helium Leak Test
differential applied, the dye then seeps through the
leak path, and the inspector uses an ultraviolet lamp Most helium leak test applications in aerospace
to illuminate the low pressure side in search of a use a sniffer probe: the part in test is pressurized and
fluorescent contrast. Once an indication fluoresces, escaping helium is drawn into the leak detector
the leak path is generally identified and may be through a sniffer. Figure 9 shows a photograph of a
repaired. helium leak test with the operator scanning the
sniffer probe over the part. Many aircraft parts, such
as integral wing fuel tanks, are relatively large,

Table 6. Advantages and limitations of tracer dye leak testing.

Advantage Technique has good sensitivity, to about 10–6 Pa·m3·s–1 (10–5 std cm3/s).

Tracer dye qualified for aircraft use is relatively expensive and not commonly available.
Because of technique's infrequency, users are inexperienced. Trial and error is generally involved.
Dye penetrant materials are contaminants to aircraft systems such as fuel tanks. Technique is notrecommended for
use inside aircraft tubing systems. All dye material must be removed from aircraft unless engineering authorizes
deviation. Cleanup istime consuming.
Limitations Common application of dye penetrant leak testing is for chasing leaks in fuel tanks, where external leak's location
is known but internal one is not. Often, shops will apply reservoir of dye material over external leak, then evacuate
fuel tank to draw dye through. Technique can be hit and miss, as often there is no way to observe internal leak
directly. Therefore, inspectors have to guess at soak time needed to draw dye through leak path. Soak time is often
under- or overestimated, so either no dye flows in or so much flows in that it spreads over wide area inside tank
and prevents location of leak. Tank must then be cleaned and test restarted. Each time test is repeated, tank must
be plugged, evacuated, reopened, inspected, and cleaned.

16.8 AEROSPACE NONDESTRUCTIVE TESTING

making it necessary to maximize the scan rate. The
Figure 9. Sniffer probe helium leak test with
faster the part can be scanned, the sooner production
assembly can resume. However, faster scanning is vectors to show air current direction and center
typically done at the expense of sensitivity and arrow indicating direction of operator motion.
accuracy, so optimizing the process involves careful
balance of both requirements. To increase the scan
rate and maintain accuracy, modified leak detectors
equipped with fast flow pumps and extended sniffer
lines make it possible to scan a large aircraft wing
surface in about 1 h.
Another factor in testing large parts like aircraft
wing fuel tanks is false positive leak indications. To
pressurize a wing fuel tank, many temporary plugs
must be used, and each one might leak. Helium from
a leaking plug may contaminate the test area and
cause false positives. To reduce helium
contamination of the test area, it is best to repair
such plugs. It is also helpful to establish a prevailing
air current using fans or other ventilation to remove
lingering helium. As long as the sniffer probe is Helium leak testing of tubing systems is
scanned in physical contact over the test surface, straightforward — with one precaution. A plugged
leakage can be detected reliably even with moderate tube system traps air, so when helium is introduced,
air movement. A helpful tip is to work by moving the air will compress to the tube end opposite of the
downwind so that in the event of a leak, the helium injection point. Because of the tight confines
inspector is not frustrated by frequent leak of the tube, gases do not mix well. In practice, it
indications caused when the air current carries a could take days or longer for gases to mix, so it is
helium bubble to the sniffer probe from a distance. necessary to either evacuate air before helium
In situations where increased sensitivity is injection or to bleed it off during or after helium
required, say, for a very small leak, the accumulation pressurization. To demonstrate the importance of
technique is useful. For most applications, it simply purging air out of the tube, consider the example in
involves bagging the suspect leak area while the test Figure 10. Here, a 30 m (100 ft) section of tubing is
part is under helium pressure, waiting until helium pressurized from the left side to 690 kPa (100 lb/in.2
accumulates inside the bag and then puncturing the gage) helium at sea level, where the barometric
bag with the sniffer probe to take a sample. This may pressure is 101 kPa (14.7 lb/in.2 absolute). Using the
provide several orders of magnitude improved simple math shown in the figure, about 4 m (13 ft)
sensitivity, but it is slow and thus appropriate for of the tube will not contain helium. The section of
infrequent troubleshooting. A helpful tip is to use the tube without helium would not be properly
plastic bagging material impermeable to helium, tested.
such as nylon plastic film typically used in
composite layup.

Table 7. Advantages and limitations of helium leak testing.

Technique has good sensitivity, to about 10–8 Pa·m3·s–1 (10–7 std cm3/s).
Technique has fast scan rates, up to 150 mm/s (6 in./s) for customized detectors.
Helium is nonhazardous gas, safe for breathing when mixed with 21 oxygen.
Inert gas will not damage exposed materials.
Technique is available worldwide.
Advantages
Technique is relatively inexpensive.
Helium is clean and easy to dispose of.
Dry gas permits concurrent processes.
Lean enabling technology— heliumleak testingcan be done anywhere, anytime.
Technique is becoming more established as standard in aerospace industry.

Technique requires long learning curve to become proficient.

Limitations
False positives can be frustrating, so test is only as good as setup.

LEAK TESTING 16.9

Figure 10. Fraction of air in helium leak test of unpurged tube.
How much of an unpurged 30 m (100 ft) tube is exposed to 690 kPa (100 lb/in.2 gage) helium?

Helium, 690 kPa (100 lb/in.2 gage), 26 m (87 ft) Air, 100 kPa (14.7 lb/in.2 absolute),
4 m (13 ft)

Tube, 30 m (100 ft)

Figure 11. Tracer probe helium leak test configuration.

Vacuum Helium sprayer

exhaust Helium leak
detector

Helium bubble
Helium leak detector applied externally
sniffer probe

Carrier air
Inlet
Leaking fitting
Tube withfittings
Vacuum pump Plug Plug

Tracer Probe Helium Leak Test Chasing Aircraft Fuel Tank

Some aircraft parts cannot be pressurized because
of structural or component limitations. These test
Leaks
objects must instead be evacuated to complete a leak Troubleshooting aircraft fuel tank leaks can be
test. These situations require a tracer probe type leak one of the most time consuming, expensive, and
test where helium is sprayed externally on possible frustrating aspects of aircraft maintenance, whether
leaks and drawn through leak paths by a vacuum for production or inservice applications. Once a leak
pump. Figure 11 shows the helium leak detector has been found on the exterior of a tank, it is
being used to sniff the exhaust port of the vacuum necessary to find the interior aperture of the leak.
pump. Tracer probe leak testing can work effectively, The interior fuel tank leak locations are most often
but long runs of tubing sometimes require long chased by applying air pressure to the exterior leak
dwell times from when helium is sprayed externally location, then using soap bubbles inside the fuel
on the fitting to when the leak detector senses tank to search for leak indications. The soap bubble,
helium leakage. Tracer probe leak testing requires air back blow technique works well where leaks
careful engineering design to introduce enough propagate directly through the structure. However,
laminar flowing carrier gas (air) to transport helium the technique is more temperamental with longer
to the leak detector without diluting it below the leak paths. Leaks have been known to run 6 m (20 ft)
detection limit. Dwell times for each fitting must also from the interior to the exterior exit point. Long path
be calculated so the inspector can determine when to leaks are nearly impossible to find using soap
expect a reading and when to move to the next
fitting.
Some aircraft parts cannot be pressurized in a
Figure 12. Helium leak test with vacuum box.
practical manner, but some test parts may be
configured to hold helium tracer gas at 0 Pa by
displacing the air inside and then sealing the part. In
such a case, it is then possible to use a variation of
the tracer probe leak test where a vacuum box or cup
is used to draw helium through leaks (Figure 12).
Like the tracer probe leak test above, the exhaust of
the vacuum box is sniffed for helium. The dwell time
for this test is very brief. In practice, the sensitivity
is adequate for most aircraft requirements, and the
scan rate can be relatively fast. Larger and smaller
vacuum boxes are available or can be custom
designed to fit the specific need.
bubbles. Instead, where challenging, and even not so Figure 13 shows the external location of a leak
challenging fuel tank leaks are observed, a very where a plastic pressure chamber has been sealed
effective technique for chasing them is to use helium over the leak and then punctured by a gas nozzle to
back blow and a leak detector inside the fuel tank to inject helium. Figure 14 shows the use of a leak
search for leak indications. A very sensitive leak detector in the tank to locate the interior leak source.
detector can help find even the smallest indications. Once located by the detector, bubbles may be used to
Once the internal leak is located, it may be repaired help pinpoint the source of the leak if desired.
and retested using the same helium back blow Table 8 summarizes some advantages and limitations
technique. of using bubble testing versus helium leak testing for
chasing aircraft fuel tank leaks.

Figure 13. Helium back blow pressure chamber Figure 14. Intrinsically safe helium leak detector
with injection nozzle inserted. probe chases leak in airplane fuel tank.

Table 8. Bubble testing and helium tracer techniques for location of aircraft fuel tank leaks.

Consideration Bubble Testing (Bubble Fluid and Air) Helium Tracer Technique

Personnel Bubble testing requires two people. Onepersoncan conduct heliumtracer test.
It may take some time to become proficient with technique.
Entering fuel tanks with electrical equipment must be done in
accordance with local fire and safety regulations.
Leak path Leak chasing is effective for shortleak paths Helium tracer testing is effective for chasing long and short leak
directly through fuel tank. paths.
Bubble test effectiveness decreases with longer
leak paths.
Leak location Exact exterior leak location must be known to It is easy to locate multiple leak sources. Apply tape over each
effectively apply air flow. interior leak source and continue to inspect area for other leak
sources.
Access may make it difficult to direct air over Exact exterior leak point is notrequired: pressurechamber may
exact exterior leak location. cover 25 cm2 (several square inches) of area.
Pressurization Overpressurizing of leak path can further Leaks are typically found with less than 14 kPa (2 lb/in.2) helium
damage seals. in exteriorpressurechamber, thusensuring no additional seal
damage.
Greater pressurescan be applied withnozzle and additional
person to apply heliumdirectlyoverexteriorleak.
Exteriorpressurechamber must bewell built to hold pressures for
extended times. Use approved cleaning solvents and cleaning
techniques.
Repair Bubble testing cannot be used to verify repairs Repairs can be verified using helium back flow, obviating
verification of fuel tanks. configuration of fuel tank for pressure test.
Verification of fuel tankrepair requires
pressurizingoftank.

LEAK TESTING 16.11

References

Bibliography Hablanian, M.H. 1997. High-Vacuum Technology: A Practical

Guide, second edition. New York, NY: Marcel Dekker.
Agilent. 2014. Leak Detection. Accessed 4 February. Mix, P.E. 1987. Introduction to Nondestructive Testing:
[Link] A Training Guide. New York, NY: Wiley.
[Link]. Lexington, MA: Agilent Technologies [Varian Oerlikon. 2014. Accessed 4 February. [Link]
Vacuum Products]. leyboldvacuum/e-catalogue. Pfäffikon, Switzerland:
ASNT. 1998. Nondestructive Testing Handbook 1: Leak Testing, Oerlikon Leybold Vacuum.
third edition. Columbus, OH: American Society for Varian. 1992. Basic Vacuum Practice, third edition. Lexington,
Nondestructive Testing. MA: Varian Vacuum Technology.
Brown, T.L., and H.E. Lemay, Jr. 1985. Chemistry: The Central
Science. [Upper Saddle River,] NJ: Prentice-Hall.
 X
17 CHAPTER

Other
Methods

Contents
Part 1. Barkhausen Effect, 17.2
Part 2. Microwave Testing, 17.5
Part 3. Dielectrometry:
Capacitance Sensing, 17.12
Part 4. X-Ray Diffraction, 17.15
Part 5. X-Ray Fluorescence
Spectroscopy, 17.18
Part 6. Fourier Transform Infrared
Testing, 17.20
Part 7. Beta Backscatter, 17.22
Part 8. Resonance and Vibration
Testing, 17.24
References, 17.27

Contributors
Richard H. Bossi Paul J. Lomax Jeffrey A. Umbach
Trey Gordon Alan J. Rein Andrew P. Washabaugh
Leanne [Link] Karl F. Schmidt Gregory J. Weaver
Mark P. Lessard John A. Seelenbinder
Chester Lo Surendra Singh
17.1
 Barkhausen Effect
PART1

The barkhausen effect consists of sudden changes can be detected as voltage pulses in an inductive coil
in the magnetic domains of ferromagnetic materials either encircling the material or placed in close
subject to a time varying magnetic field, resulting in proximity to the material surface. The detected
electrical impulses that can be detected with a signals, called barkhausen noise after Heinrich
sensing coil. The barkhausen effect is one of the Barkhausen (Barkhausen 1919), are generated by
electromagnetic techniques used for nondestructive abrupt motion of magnetic domain walls inside the
testing of ferromagnetic materials (mainly steel) in material when they break away from the pinning
various industries including aerospace, pipeline, sites (grain boundaries, dislocations, or inclusions)
automotive, power plant, and steelmaking. The once the local magnetic field is large enough to
phenomenon was discovered early in the twentieth overcome the pinning force. Barkhausen noise
century (Barkhausen 1919) but remained ascientific depends onthe stress state and microstructure of
curiosity for decades because it seemed to have no materials, and the technique has found applications
practical use. The revival of interest in the in evaluating residual stresses and microstructural
barkhausen effect stems from research observing changes provided that the competing factors can be
distinct variations in barkhausen signals with controlled. The technique is primarily sensitive to the
applied and residual stresses (Leep 1967; Pasley surface or near-surface conditions of materials, as
1970). Since then, the barkhausen effect has beenthe the signals generated deep inside thematerial are
subject of extensive research focusing on its increasingly attenuated at high frequencies due to
measurements of residual stresses, evaluation of eddy current shielding. Typical measurement depths
surface treatment (such as grinding, shot peening, vary between 0.01 and 2 mm depending on the
and case hardening), and material degradation such conductivity, permeability, and the frequency range
as fatigue and radiation damages. For complete ofsignal processing and analysis.
reviews and in-depth discussion of the barkhausen
effect, readers are referred to the literature (Tiitto
1996; Matzkanin 1979; Dobmann 1998). Instrumentation
A basic configuration for barkhausen noise
measurements is shown in Figure 2. A surface sensor
Physical Principles probe consists of a C core electromagnet for applying
When a magnetic material is subjected to a time a time varying magnetizing field to a test sample,
varying magnetic field, its magnetization changes in and an induction coil, either with or without a ferrite
small, discontinuous steps (Figure 1). These changes core, to detect a barkhausen signal. Special sensor
probes are commercially available for inspecting
components with various complex geometries such
as camshafts, inner tube surface, and bearings. The
Figure 1. Magnetizing curve of ferromagnetic detected signals are amplified and filtered to improve
materials showing discrete changes in magnetic the signal-to-noise ratio using a band pass filter
flux density and emissions of barkhausen noise typically from about a few kilohertz to hundreds of
signals. kilohertz (Tiitto 1996; Matzkanin 1979; Dobmann
1998). Several measurement parameters, including
B
maximum signal amplitude, root meansquare value,
total signal power, and number of barkhausen
pulses, are extracted from the filtered signal to
evaluate the material conditions of interest
(especially residual stresses) based on calibrations
(Tiitto 1996; Matzkanin 1979). Commercial
H barkhausen measurement systems are available for
Coercive assessment of residual stress and near-surface
field
conditions of steel components with typical
repeatability of 2 (Tiitto 1996). The barkhausen
Barkhausen noise measurement is unitless and must be calibrated
signal against some baseline or reference that establishes
the residual stress value for acceptance. The
accept/reject threshold is usually established
empiricallywithtestcoupons, other testssuchas
X-ray diffraction, and sometimes mechanical load
tests.

17.2 AEROSPACE NONDESTRUCTIVE TESTING

 Figure 2. Typical setup for measuring barkhausen noise using surface sensor probe.

Barkhausen noise signal

High pass filter Low pass filter

(3 kHz) Amplifier (250 kHz)

Excitation coil
Current source

Signal processing and

display of result or
computer for further
analysis and data storage
C core Induction coil
magnet on ferrite core

Test piece

Basic analog barkhausen systems often lack the

flexibility of optimizing the test conditions and Figure 3. Stress dependence of barkhausen noise when magnetizing
provide only a single measurement parameter (the field is parallel or perpendicular.
root mean square value) of the detected signals.
Progress has been made through digital instruments
with embedded computers (AST 2014) that provide Barkhausen
users with control of magnetizing field frequency noise signal
and strength, and of the analyzer frequency pass Parallel
band to select the range of measurement depths. The
instruments also generate an array of measurement Perpendicular
parameters allowing multivariate analysis of their
correlations with the material properties. Automated
barkhausen imaging systems have been developed
for inspecting spatial variations in material
properties such as residual stress distribution
(Altpeter 1998; Lo 2003).

Applications of Barkhausen Tension

Effect Stress

Stress Measurements
One of the principal industrial applications of
barkhausen noise is evaluation of residual stresses calibration of barkhausen noise for biaxial stress
when microstructural variables are relatively measurements has also been demonstrated (AST
constant. The technique operates bythe fact that 1990; Buttle 1990). The useful range of barkhausen
stresses affect domain wall movement and hence the stress measurements in general extends to about
barkhausen signals due to the magnetoelastic 50 percent of the yield strength with a resolution of
interaction (Jiles 1995). For materials having a 30 MPa (4350 lb/in.2) (Buttle 1990). The barkhausen
positive magnetostriction such as steels, barkhausen effect has been applied to measure stress in turbine
signals increase under uniaxial tensile stress along diffuser cases, compressor blades, compressor disks
the magnetizing field direction but decrease under (Barton 1974), weld seams (AST 1990; Tiitto 1996,
compressive stress. If the magnetic field is normal to Buttle 2000), pipes (Jagadish 1989; Crouch 2004),
the stress, the reverse is true. Figure 3 shows how the camshafts (Fix 1990), gear teeth(Ceurter 2002), and
barkhausen noise responds to tension or compression ballbearings (Matzkanin 1979; Ogden 1991;
stressand inparallelorperpendicular fields. The Desvaux 2004).

OTHER METHODS 17.3

 Surface Treatment steels, the general trend is that barkhausen signals
increase as martensite recovers with increasing
Barkhausen noise has been applied to evaluate tempering time or temperature (Buttle 1987).
steel components subjected to surface treatment Crystallographic texture is present in most rolled
including shot peening, hardening, and grinding steel sheets and plates. This produces anisotropy in
(Tiitto 1996; Matzkanin 1979). The barkhausen magnetic properties and can be readily detected by
signal typically decreases as surface compression barkhausen noise (Tiitto 1996; Dhar 2001).
increases from shot peening; after the material loses Neutron embrittlement in nuclear pressure vessel
its surface integrity from overpeening, the steels has been monitored through barkhausen noise
barkhausen signal intensity rises (Tiitto 1996). The [27, 28] (Altpeter 2003; Buttle 1987a; Buttle 1987b).
feasibility of measuring residual stress depth profiles The signal level generally decreases with increased
in shot-peened steels byadjusting analysis frequency neutron flow because of irradiation damage in the
ranges has been demonstrated through comparison form of precipitates or discontinuity clusters.
with stress measurements using X-ray diffraction Barkhausen noise measurements have been shown to
(Jiles 1993). The use of barkhausen noise for have potential for detection of fatigue damage in
evaluating carburized and induction hardened steels, steels (Tiitto 1996; Altpeter 2003; Moorthy 1999).
including those in aerospace bearings and landing
gears, has beenstudied (Bach 1988). Grinding burns
that affect hardness and stress make barkhausen
noise deviate from that obtained from a properly Summary
ground surface (Tiitto 1996; Fix 1990; Ceurter 2002; Barkhausen noise measurement is a viable
SAE 1991). Caution should be taken when nondestructive testing technique for residual stress
interpreting inspectionresults fromsurface modified measurements and for detection of surface and
components: barkhausen signals are sensitive to near-surface microstructural changes in materials
several factors including microstructure and stresses. caused by processing or service. Commercial systems
Multiple barkhausen measurementparameters or a based on the barkhausen effect are available for
combination of different magnetic measurement industrial applications, and the technique has
techniques (such as incremental permeability) could recently evolved into imaging systems for mapping
provide a means of characterizing the depth profiles surface conditions of components. Interpretation of
ofresidual stress and microstructures in surface inspection results remains a difficult task, especially
treated components (Dobmann 1998). when several different material parameters are
varying simultaneously. Interpretation presents a
considerable challenge to the technique and justifies
Microstructure further studies to understand the mechanism of the
With suitable calibration, barkhausen noise can barkhausen effect at a fundamental level. Of
be used for evaluating microstructural damage from particular interest to the engineering community is
heat treatment, processing, or service. The data correlation to establish acceptable, negligible,
microstructural factors that affect barkhausen noise, repairable, and reject inspection criteria. Currently,
including grain boundaries, dislocations, inclusions, data correlation between hardness/stress conditions
precipitates, or secondary phase, can be broadly and resulting barkhausen inspection readings are not
described by the material hardness. In general, fully accepted and used. These data are essential for
barkhausen noise intensity decreases with increasing quality product and for minimizing of replacement
hardness (Tiitto 1996; Dobmann 1998). For heat and refurbishing costs.
treated components such as tempered martensitic

17.4 AEROSPACE NONDESTRUCTIVE TESTING

 Microwave Testing PART2

Electromagnetic energy provides the means for

nondestructive testing in a number of methods — Figure 4. Electromagnetic spectrum.
such as visual, eddy current, radiologic, and
thermographic testing — as discussed in earlier
chapters. Afurther part of the electromagnetic Frequency (Hz) Wavelength (m)
spectrum for nondestructive testing applications is
between radar and infrared (Figure 4). Microwaves
lie in a range of several hundreds of megahertz 106 3  102
(MHz) to several hundreds of gigahertz (GHz).
Table 1 lists some microwave frequency bands (IEEE Radiowaves
2009). Millimeter waves lie in the band between 108 3  100
3  1010to 3  1011Hz (30 to 300 GHz) and having
wavelengths between 10 and 1 mm. The Radar
submillimeter range is referred to as the terahertz 1010 Microwaves 3  10–2
(THz) regime, between 3  1011and 3  1012 Hz Millimeter waves
(300 GHz to 3 THz), where wavelengths are between Terahertz 3  10–4
1 and 0.1 mm. In the 5 to 50 GHz range, microwaves 1012
have photon energy in the range of 2.0  10–5eV to Infrared
2  10–4 eV. Because electromagnetic energy in these 1014 3  10–6
Visible
frequency ranges does not penetrate electrically
conductive materials, microwavesareusedprimarily Ultraviolet
for volumetric examination and for bulk 1016 3  10–8
characteristics of dielectric materials. They can also
beused for surface and near-surface measurements
3  10–10
on conductive materials. These are nonionizing 1018 X-rays
energy levels, substantially below the energy level of
molecular bonds. Microwave testing is therefore 3  10–12
relatively safe and can be performed without 1020
contacting the test object. The method can be
applied to complex geometries (Kharkovsky 2007;
Ramuhalli 2004; Schmidt 2013; Zoughi 2000).

Table 1. Microwave frequency bands.

Band Wavelength Range Frequency Range

Designation (mm) (GHz)

L 300 to 150 1 to 2
S 150 to 75 2 to 4
C 75 to 37.5 4 to 8
X 37.5 to 25 8 to 12
Ku 25 to 16.7 12 to 18
K 16.7 to 11.1 18 to 27
Ka 11.1 to 7.5 27 to 40
V 7.5 to 4 40 to 75
W 4 to 2.73 75 to 110
mm 10 to 1.0 (“millimeter wave”) 30 to 300
THz 1.0 to 0.1 300 to 3000

OTHER METHODS 17.5

 used in microwave interferometry. In this
Microwave Principles configuration, the electromagnetic field at each
Microwaves, millimeter waves, and terahertz detector consists of both the emitted and reflected
radiation operate at frequencies where the energy.
interaction with matter is best described in terms of In the high frequency microwave and terahertz
waves (or fields) that penetrate, reflect, or refract. regimes, with the shorter wavelengths of a fewmm
The electromagnetic wave will be changed in or less, time based spectroscopy can be useful. In this
amplitude or phase as it passes through or is technique, the reflected signal is detected as a
scattered by any object it interacts with. The basic function of time, creating a waveform similar to
penetration of this radiation is based on the same ultrasonic pulse echo. An advantage of this approach
depth of penetration equation discussed in the eddy with high frequency electromagnetic waves is that
current chapter: the reflection from features, such as disbonds, does
1 not completely block signals from features behind as
(1)   happens with ultrasound. The reflection intensity is
 f r 0  based on the dielectric constant. However, a metallic
where f is frequency (Hz), r is relative permeability, inclusion would completely block features behind it.
0 is permeability of free space (4 ×10-7 H/m), and
 is conductivity (S/m). In the case of pure
aluminum where the conductivity is 35 MS and the
relative permeability is 1, at a frequency of 3 GHz,
Microwave Interferometry
Microwave interferometry requires access to one
the depth of penetration would be only 1.5 m,
side of a part. When microwave energy washes over
demonstrating how microwaves are unable to
a part, some energy is reflected and some is
penetrate metals. Free electrons in conductive
transmitted at every interface of changing dielectric
material move under the microwave electromagnetic
constant in the field of the transmitter. Displaying
field, blocking transmission and creating reflection
at the surface of the conductor. the reflected energy combined with the transmitted
The mechanisms of electromagnetic wave
interaction with matter are dipole interactions and
ionic conduction. Dipole interactions occur when Figure 5. Microwave measurement setups:
polar molecules are aligned and oscillate withthe (a) bistatic transmission; (b) bistatic scattering;
electrical field. Ionic conduction occurs when (c) bistatic reflection; (d) monostatic microwave
charged species distributed in the material are interferometry.
coupled with the oscillating electrical field. The
decaying rotational or torsional energy of molecules (a) Material
is dissipated by interaction with adjacent molecules system
as heat in the material. As illustrated with Isolator
microwave heating of water, this effect can be
Source Detector
substantial. But the chemistry of the material is
unaffected except through thermal input. Horn
antenna
Microwaves and millimeter waves are typically
generated by an antenna oscillator and commonly
operate in a resonance cavity of tuned geometry. The (b)
instruments that generate radiation at these Material
wavelengths are not very large or expensive. For the system
higher frequency terahertz radiation, however, Isolator
ultrafast femtosecond laser systems add considerable Source Detector
cost. The penetration of terahertz radiation is
sensitive to moisture, and the spectral sensitivity of Horn
antenna
chemicals in the 0.5 to 5 THz range makethe terahertz
regimeofinterestfor somespecialapplications.

Microwave Imaging (c) Material

system
Nondestructive testing applications can be Circulator
performed with microwaves in the near field orfar Isolator
Source
field and use reflection, transmission, or scattering.
The electromagnetic energy is conducted in a Horn
waveguide or horn, and propagated from the open Detector
antenna
ended waveguide. The waveguide aperture geometry
defines antenna pattern energy distribution of the (d) Material
transmitted signal. The waveguide can be scanned system
over the test part to create an image. The systems Detectors
(two or more)
may be monostatic, using a single source and
receiver antenna, or bistatic, using a separate source Source
and receiver. Figure 5 shows bistatic configurations Horn
for reflection, transmission, and scattering antenna
measurements and the monostatic configuration

17.6 AEROSPACE NONDESTRUCTIVE TESTING

electromagnetic field creates a microwave Z dimension associated with maximum rate of
interference pattern. Reflections from the features change of the electromagnetic field in the Z
interfere withtheinitial transmission signal. The dimension. This is achieved by moving the emitter
interference signal is measured at position (and receiver) in the Z direction, orby selecting a
increments over the scanned area to create an image different receiver position, or by changing the
of features in the test object (Schmidt 2010). frequency of the emitted field. This facilitates
The combination of dielectric constants of discrimination of laminar features.
nonconducting materials used in aerospace
applications ranges from under about 2.0 for
polytetrafluoroethylene (Teflon®) toover 13.0 for Applications
engineered ceramic materials. At the typical
application frequency range of 10to 35 GHz, the Microwave, millimeter wave, and terahertz
wavelengths  in the materials range from under radiations are particularly applicable to dielectric
2.3 mm (0.09 in.) to over 20.5 mm (0.81 in.). The material inspection and characterization of foams,
minimum detectable feature can be much smaller plastics, and ceramics. Coatings and coating
than the wavelength. The electromagnetic fields of thickness are of particular interest in aerospace
the emitted energy and the reflected energy are components, as is moisture measurement. Table 2
combined at the receivers. The magnitude of the lists some aspects of microwave testing. The
phase difference between the emitted and reflected equipment expense is not particularly high except in
electromagnetic fieldsdetermines the signal voltage. the terahertz range, where generation and detection
The rate of change of the interference signal require sophisticated technology.
value defines the contrast of the image. Thus, the
contrast value of a feature is optimized at a

Table 2. Applications of microwave, millimeter wave, and terahertz testing.

Inspection of foam coatings for voids and delamination.

Inspection of dielectrics such as ceramics and plastics, including fiber reinforced plastics, ceramic matrix
composites, and foams for voids, inclusions, delaminations, inclusions, and disbonds.
Thickness and density mapping.
Coating characterization and thickness.
Applications Bond line condition.
Corrosion under paint and coatings.
Moisture detection and measurement.
Cure state in dielectrics.
Surface crack detection in metallic materials.

Noncontact.
No coupling medium.
One-sided.
Can image discontinuities behind discontinuities.
Advantages Can image part volume through voids and structure.
Can be used in scanning modes or imaging modes.
Compact equipment.
Environmentally safe.
Spectroscopy.

Penetration in or through conductive materials.

Limitations Limited sensitivity as function of frequency, positioning, and technique.
Limited but emergent number of specifications and standards for implementation.

OTHER METHODS 17.7

 Fiber Reinforced Resins and marcelling. The circular indications are also artificial
discontinuities, including delaminations and foreign
Honeycomb Structure objects. Comparisons of images at different phase
The image in Figure 6 is a fiberglass test panel angle (Z axis distance to reference plane) can be
containing intentional marcelling (wrinkling), used to establish relative depths of features and of
laminar features, and foreign objects in a laminate Z axis dimension of features.
structure of organic resin with fiberglass Radomes are fiberglass, designed to allow radar
reinforcement and honeycomb structure. The linear transmissions. In manufacture and service, they are
indications oriented diagonally and vertically are subject to damage that affects performance.
Microwave inspection applies very well to radomes
(Figure 7). Figure 8 demonstrates the clear detection
Figure 6. Microwave interference scan image of fiberglass panel with of artificial discontinuities at two microwave
marcelling, disbonds, and foreign objects. frequencies.

Measurement of Density or Thickness

17
16 Microwave interference scanning is effective for
15 density/thickness measurement by correlation of the
14 phase shift of the interference standing wave to the
13 effective thickness of the sample material. This phase
12 shift is caused by the change in distance, as
Y (relative scale)

11
10 measured in wavelengths, for the emitted and
9 reflected signal to transit the material. The
8 relationship of the phase vectors in two receiving
7 channels physically separated inphase is calculated
6 to determine the density/thickness at each point
5
4 measured. Density/thickness (depth) can be directly
3 extracted from the resulting interference signal.
2 Figure 9 shows a scan image for a glass fiber
1 reinforced epoxy resin wedge, in which the channel
0 phase relationship has been calibrated to present
0 2 4 6 8 10 12 14 16
X (relative scale)

Figure 8. Millimeter wave images of aircraft

Figure 7. Aircraft radome test panel with artificial discontinuities: radome test panel at two different frequencies:
(a) photograph; (b) diagram. (a) 33 GHz; (b) 73 GHz (Ravuri 2008).

(a) (a)

(b) c
a
a

a
(b)
a a a a a a a a
a

b b b
a
a

c
a
a

Legend
a = 50 mm (2.0 in.)
b = 100 mm (4.0in.)
a

c = 450 mm (18 in.)

17.8 AEROSPACE NONDESTRUCTIVE TESTING

 Thickness, mm (in.)
Figure 9. Calibrated thickness measurement of glass fiber reinforced epoxy resin wedge.

38 (1.5)
25 (1.0)
13 (0.5)
0
250 (10)

200 (8)

150 (6) ≥40 mm (1.6 in.)

35 mm (1.4 in.)
30 mm (1.2 in.)
100 (4) 25 mm (1.0 in.)
20 mm (0.8 in.)
15 mm (0.6 in.)
50 (2) 10 mm (0.4 in.)
125 5 mm (0.2 in.)
(5.0) 100
75 ≤0
(4.0)
(3.0) 50 0
(2.0)

direct measurement of thickness. The wedge is reflectors. Figure 10 shows an example of microwave imaging at 24 GHz
tapered from 0 to 44 mm over its 280 mm length. inspection of a multilayer composite panel injected with one 1.0 mL and four
The varied thickness supports a calibrated range of 0.01 mL water droplets.
8 to 29 mm. For material of uniform thickness, this
calibration can be assigned to measure density Foam
directly.
The insulation for the space shuttle external tank was foam, sprayed on. The
catastrophic failure of space shuttle Columbia was attributed to pieces of the
Moisture Detection foam dislodging during launch and causing damage tothewing. The low
Microwave interference scanning is very density and thickness of the foam made conventional ultrasonic and
responsive to moisture in materials under inspection. radiologic testing of little use. Microwaves offer a useful way to penetrate the
The dielectric constant for water depends on both foam and have sensitivity to internal discontinuities (Kharkovsky 2005). Figure
frequency and temperature but is much higher than 11 shows an example of microwave inspection ofatest article ofthe space shuttle
the dielectric constant of most dielectric materials, foam using a 100 GHz monostatic small horn antenna. The microwave sensor is
particularly the dielectric constants of organic and scanned over the foam to check
ceramic composites in aerospace applications. Thus, forinternal voids and discontinuities. The scan
highly distributed moisture is detected as a change in image is sensitive to the structural features of the test
bulk material dielectric constant; very small article and shows the intentional inserts at the top of
quantities of moisture are detected as anomalies in the image consisting of three rubber inserts
the material, and water droplets are detected as
 Figure 10. Image of water permeation in multilayer three-dimensional
composite panel: (a) cross section; (b) 24 GHz (image).
OT
(a)
HER

ME 1.5 mm (1/16 in.)

air gap 6.4 mm (0.25 in.) acrylic
THO
DS 17.9

Water insertion 13 mm (0.5in.) layer

points open cell foam

(b) 0.1 ml water

(typical by four places)

1.0 ml water
0 (forming aring)
50 (2)
100 (4)

150 (6)
50 (2)
200 (8) 100 (4)
150 (6)
200 (8)
and three voids. The dashed box in the image was of images, optimized at different planes, during a single
an unintentional void during manufacture of the test scan of the part, and allows slight shifts of the
standard. Figure 12 shows a similar image using optimized plane in addition to the phase separation
focusing and polarization to improve the image of the receiver channels. Figure 13 shows scan
quality. images of a ceramic composite sample with artificial
discontinuities at slightly different frequencies. The
interrogating frequencies are (left toright): 33.91,
Engineered Ceramics 33.99, and 34.12 GHz. The fabric structure at the
Microwave interferometry is effective for optimized plane is readily apparent. Delaminations
detection of delaminations and other features in in the optimized plane appear dark gray, and
engineered ceramic materials, such as ceramic superimposed nearer features appear light gray.
matrix composites, which present challenges to Persistence through frequency shifted images
examination by other methods because of their indicates extent of depth. The images are thus
complex heterogeneous material. Manipulation of optimized at different depths, demonstrating the
the emitted frequency and sampling of the combined depth differentiation of multiple, overlapping
field effectively moves the optimized plane within laminar discontinuities and defining isolated features
the material. This facilitates gathering multiple (Schmidt 2014).

Internal Voids and Structure

Figure 11. Image of foam insulation with horn
antenna at 105 mm standoff (Kharkovsky 2005). Microwave energy propagates through voids and
similar features in parts. This enables imaging of
Rubber inserts Voids parts having complex structure and limited access.
Figure 14 shows a demonstration ona composite
structure with carbon fiber reinforced plastic face
sheets and an aluminum honeycomb core. The
sample has a single-ply face sheetonthenear side
and four plies on the far side. There are three
segments where adhesive is removed from the bond
line between the aluminum honeycomb core and the
far side face sheet. The far side face sheet has three
changes in thickness, associated with composite ply
drops, as shown. The image shows variations in
microwave intensity related to the material changes
in the testobject (Schmidt 2013).

Corrosion under Paint

Measurement of paint thickness and corrosion
under paint are useful implementations of
microwave technology. Figure 15 shows an example
of corrosion under a painted surface that can be
readily imaged with microwaves.

Figure 12. 100 GHz images of sprayed-on foam panel using lens focused millimeter wave imaging system:
(a) perpendicular polarization; (b) parallel polarization (Zoughi 2005; Zoughi 2006).
(a) (b)
 Figure 13. Ceramic matrix composite sample with artificial discontinuities interrogated at three frequencies: (a) 33.91 GHz;
(b) 33.99 GHz; (c) 34.12 GHz. Dimensions are in millimeters (Schmidt 2014).
(a) (b) (c)

315 315 315

300 300 300
285 285 285
270 270 270
255 255 255
240 ≥–0.016 240 ≥0.837 240 ≥1.531
225 –0.138 225 0.738 225 1.441
210 –0.261 210 0.639 210 1.352
195 –0.384 195 0.541 195 1.262
180 –0.506 180 0.442 180 1.173
165 –0.629 165 0.343 165 1.083
150 –0.752 150 0.245 150 0.994
135 135 0.146 135 0.904
–0.875
120 120 0.047 120 0.814
–0.997
105 105 –0.051 105
–1.120 0.725
90 90 –0.150 90
–1.243 0.635
75 75 75
60 60 60
45 45 45
30 30 30
15 15 15
0 0 0
5 25 45 65 85 105 125145 5 25 45 65 85 105 125145 5 25 45 65 85 105 125145

Figure 14. Scan image and diagram of sample Figure 15. Corroded panel (dashed area):
with carbon fiber reinforced resin face sheets (a) painted with thin layer of paint; (b) 24 GHz
and aluminum honeycomb core. image showing corroded area along with
Adhesive indication of corrosion pitting (Kharkovsky
gaps Ply
drops 2006).

(a)
6.3 mm

8.3 mm

12.7 mm
(b)

This freehand 24 GHz scan image presents far

side face sheet thickness in gray scale, dark for
thickest, and presents absence of far side bond line
adhesive as a light shift in gray scale due to the
difference in material chemistry.
The electromagnetic fieldpropagates through the
carbon fiber reinforced resin because the carbon
fibers are not connected electrically by their
dispersion in the resin matrix and may also be
isolated by coating. At this frequency, the
wavelength in air is 12.43 mm, compatible with
propagation through honeycomb cell structure
approximately 6.25 mmwide.

OTHER METHODS 17.11

PART3
Dielectrometry: Capacitance
Sensing

Dielectric and capacitance sensing techniques are

used to examine and characterize insulating or
Sensors
weakly conducting dielectric materials. Sensor Sensors used for capacitance sensing are called
electrode structures apply an electric field to the test dielectrometers. Dielectrometers are used to measure
object and respond to changes in the electric field the properties of many liquid and solid dielectrics,
related to the dielectric properties of the tested composites, and surface coatings. Representative
material. The effective impedance or admittance applications include cure state monitoring of
between the electrodes result from the complex polymer composites, adhesives, and epoxies;
dielectric permittivity and geometrical properties of detection of discontinuities in glass fiber epoxy
the material under examination, including the composites; moisture diffusion monitoring; and
presence of any air gaps between the sensor porosity measurements of ceramics, such as coatings
electrodes and the examined material. Often, the that provide a thermal barrier at high temperatures
measurements areconverted intophysical geometric (Little 2010).
or dielectric properties, such as the dielectric A typical dielectrometry system includes a sensor
permittivity, electrical conductivity, and loss tangent, tip, a probe, and measurement instrumentation. The
which are then correlated to other properties of probe electronics provide amplification and
interest, such as the cure state, moisture content, buffering forthe sensor signal. The electronics in
concentration of impurities and additives, coating turn are connected to instrumentation for measuring
thickness, density, and aging status. This use of the terminal impedance or admittance of the sensor
electric fields is analogous to the use of magnetic electrodes. Most systems are self-contained and
fieldswitheddycurrent sensors and magnetometers. provide front panel access for instrument control
Different electrode structures and multiple excitation and data display; some also use a computer, which
frequencies allow spatial and temporal variations of provides greater flexibility for data display,
the dielectric and material properties to be processing, and storage. Buffers or signal amplifiers
determined. These sensors and associated in the probe electronics let the sensor be at a
instrumentation are suitable for system monitoring distance from the instrumentation. This separation is
and diagnostics and can be used for optimization of useful for manufacturing settings where it is
designand performance characteristics (ASNT 2004; convenient to have all of the control systems and
Von Hippel 1954; Auld 1993; Schull 1990). A diagnostic information in a central location.
summary of the advantages and limitations of The goal of dielectrometry is to infer physical and
dielectrometry is provided in Table 3. electrical characteristics of a test object from its

Table 3. Advantages and limitations of capacitance sensing.

Method is sensitive to dielectric properties that indicate material condition.

Method is sensitive to curestate, layer thickness, moisturecontent, porosity, delamination, and
cracks.
Advantages Measurements may be contacting or not.
Portable equipment can give immediate results.
Flexible, single-sided sensors can inspect complex sizes and shapes.
Recording devices allow permanent records.

Method is useful only for insulating or weakly conducting materials.

Depth of penetration is limited for single sided sensors.
Limitations Through-thicknessmeasurementsrequire access to bothsidesoftestobject.
Edgeeffects can influence inspectionresults.
Conducting coatings or surface layers can limit inspectability.
admittance orimpedance measured at electrical many materials, the effective properties are
terminals. This test is easiest using parallel plate dispersive, so the effective properties depend on the
electrodes for materials with homogeneous electrical excitation frequency. Dispersiveness can be
properties. When a voltage is placed across the attributed to heterogeneous material properties, such
electrodes as shown in Figure 16, the electric field is as particles embedded in a matrix or multiple layers
spatially uniform in the central region between of different material properties, or can be attributed
electrodes and points in a single direction to multiple physical processes, such as multiple
perpendicular to the electrode surface. For a conduction mechanisms. Different techniques or
homogeneous material of permittivity  and ohmic models are then used to determine the properties of
conductivity  (inverse of resistivity ), the interest. As examples for multiple layered materials,
capacitance C and conductance G (inverse of equivalent circuits and expressions for the terminal
resistance R) canbeexpressed: response that account for the properties of each layer
are used. For two-phase composite materials, the
A effective dielectric properties of the composite can be
(2) C  related tothe dielectric properties and geometry of
d
the constituent materials.
The effects ofmaterial heterogeneity can also be
1 A displayed graphically. Plots of the imaginary part of
(3) C  
R d the impedance, admittance, or permittivity versus the
real part are called cole-cole plots. When cole-cole
where A is the electrode area and d is the gap plots are semicircular, they provide a quick
thickness between the electrodes. Material dielectrics verification that the material is homogeneous and
have permittivities that scale with the permittivity of frequency nondispersive. If the dielectric properties
free space, 0 = 8.854 pF/m. When excited by a of permittivity and conductivity are spatially
sinusoidal voltage in time, with frequency f and dependent or dependent on frequency, then the cole-
angular frequency  = 2f, the corresponding cole plots will deviate from the semicircular shape.
equivalent admittance Y (inverse of impedance Z) In many circumstances, only one side of the test
can be expressed: material is accessible or the spatial variations of the
1 material properties may be of interest. In these
(4) Y   G  jC situations, it is generally impractical to use a parallel
plate sensor: one-sided measurement is required.
Z
Single-sided sensor schemes can be visualized as
where j = √(–1). parallel plate capacitors that have been opened so
The dielectric material properties can also be that the sensor responds to the material properties
expressed in terms of the complex permittivity *: through changes in the fringing electric fields
between the electrodes. Guard electrodes can also
(5) *     j  limit unintentional coupling of the fields to the sense
electrode. These single-sided electrodes are suited to
r 0 
inspections of materials having complex geometry or
where r is the material‘s relative permittivity. curvature and for surface mounting or embedding
Note that typical excitation frequencies range
fromless than 1 mHzforinsulating materials up to
several megahertz for semiconducting materials. For Figure 16. Parallel plate electrode sensor: (a) basic sensor; (b) sensor
parallel plate electrodes, the sensor geometry is with guard electrodes.
accurately captured by the ratio A/d of the area to
thegap when the electrode widths are much larger
than the gap so that the fringing fields at the (a) I
electrode sides can be neglected. Placing guard
electrodes around the sense electrode with their
voltage the same as the sense electrode voltage, such
as using an operational amplifier voltage follower V0 cos (t)
circuit, helps minimize the effects of the fringing Sample
fields (Figure 16b). A guard electrode can also be
placed behind the sense electrode to further reduce Fringing fields
extraneous coupling from the fringing fields. Related Electrodes
physical properties such as moisture content, density, (b)
porosity, and impurities are then inferred from the I
dielectric properties. Empirical measurements can Drive
generally map values of the physical variable to
values of the material permittivity and conductivity.
Insome cases, quantitative modelscanbeused. V0 cos (t)
For simple systems, measurements at asingle
excitation frequency can be used to determine both Guard electrode driven at
the permittivity and conductivity of the material. Sensing potential of sensing electrode
These can reflect absolute dielectric properties for the electrode
material or effective properties if other effects such Guard electrode driven at
as material layer thicknesses are not accounted for in potential of sensing electrode
the conversion from the sensor impedance data. In

OTHER METHODS 17.13

 for monitoring of material condition during aging or The depth of penetration of an electric field in
processing. For example, this type of dielectric test material depends on the geometry of the
sensor can be used to monitor the cure state of electrode structures for the single-sided sensors. For
materials such as sealants, paints, and composites. interdigitated electrodes, the spatially periodic
The single-sided sensors can take a variety of variation of electric potential along the surface (in
design. In one format, the fringing electric field the Y direction) produces an electric field that
between two or more circular electrodes is used to penetrates into the test object (in the Z direction).
examine the material. In another format, a circular The electric potential at the surface obeys Laplace‘s
drive electrode is encircled by one or more sense and equation and can be represented as a fourier series of
guard electrode rings. For these circular sensors, the the spatial modes. This series analysis shows that the
nominal capacitance between the drive and sense electric field for each spatial mode at a given
electrodes is relatively small so that large voltages frequency decays exponentially (in the Z direction)
are required. with a penetration distance proportional to the
An alternative single-sided electrode format is an spatial wavelength of the periodic electrodes
interdigitated electrode structure (Figure 17a). The (Figure 17b). Sensors with longer spatial
interdigitated electrodes provide a relatively large wavelengths have larger penetration depths and
capacitance between the drive and sense electrodes respond to material properties far from the sensor
within a given sensor footprint. The spatial dielectric interface, whereas smaller wavelength
periodicity of the interdigitated electrodes is denoted sensors respond to properties primarily near the
by the wavelength . One set of electrode fingers is interface. Multiple wavelength sensors provide
driven by a sinusoidally time varying signal vD with spatial profile information about the test material
known amplitude and frequency while the second set and permit simultaneous measurement of the
of interdigitated fingers is capacitively loaded and dielectric properties and thickness of air gaps that
floats to a sensed voltage vS or is virtually grounded may be present between the test material and the
with a terminal current iS. The sensed signal sensor. These multiple-wavelength sensors can use
amplitude and phase with respect to the driven multiple sets of interdigitated, spatially periodic
voltage depend on the complex permittivity of the electrodes attached to a common substrate as in
adjacent dielectric. With these structures, the Figure 17b or colocated designs that interweave the
inversion between the measured sensor terminal electrodes for a multiple-wavelength measurement
variables and the dielectric properties is more within a single-sensor footprint. Segmenting of the
complicated than with the parallel plate structures, electrodes into arrays also permits wide area imaging
but iterative and model based techniques are being of dielectricproperties.
used.

Figure 17. Single-sided sensor design with interdigitated electrode: (a) interdigitated dielectrometry sensor in one sided contact
with material under test; (b) electrode spatial wavelength limiting electric field intensity depth of penetration into test material.

(a) (b)
 Sensor 1 Sensor 2
2 d2 ≅  2 ÷ 3
1
Test material

d1 ≅  1 ÷ 3

X Y
Electric fieldlines

+ Legend
VS + d = electric field penetration depth
– V VD= sinusoidally varying time signal
VS = sensedvoltage
–D X, Y, Z = directional coordinates
= wavelength of spatially periodic electrodes

17.14 AEROSPACE NONDESTRUCTIVE TESTING

 X-Ray Diffraction PART4

Since the discovery of the diffraction of X-rays aluminum, iron, and associated alloys. An example
from a crystal in 1912, the X-ray diffraction method of a diffraction pattern from a sample of
has found various applications in the measurement polycrystalline aluminum is shown in Figure 19. The
of properties of crystalline material. In the aerospace angles at which the peaks occur and their relative
industry, X-ray diffraction is used widely in process amplitudes are a fingerprint for the crystalline
control and development and has become a staple in structure(s) of the material.
only limited manufacturing applications. For the example in Figure 19, only a narrow
Applications for X-ray diffraction include residual section of the generated X-ray spectrum was used.
stress measurements, phase analysis, orientation Typical energy spectra are shown in Figure 20 for a
analysis for single crystals, and preferred orientation molybdenum X-ray source. It is apparent from the
analysis for polycrystalline material. spectra that characteristic energy lines such as K
Diffraction can occur for any type of can be superimposed on the bremsstrahlung
electromagnetic radiation provided that the radiation curve and those lines can serve as nearly
appropriate periodic spacing in a subject material is monochromatic X-rays.
present. The electromagnetic waves that reflect from
the periodic spacing and then constructively
interfere are said to diffract. Figure 18 shows the
diagram for the diffraction principle and the Figure 19. X-ray diffraction pattern from copper target X-ray tube,
derivation of the bragg equation: showing angle of diffraction for particular planes in polycrystalline
sample of aluminum.
(6) n  2 d sin 

15 (111)
which describes the angle of the diffracted X-rays
Intensity (counts)

forparticular plane spacing andaparticular

10
wavelength, where d is the lattice spacing, n is the
order number of the diffraction pattern,  is the (200)
5
X-ray wavelength, and  is the angle of diffraction. (220) (311)
For X-rays with wavelengths in therange of (222)
0.229 nm (chromium X-rays) to 0.071 nm 30 40 50 60 70 80 90
(molybdenum X-rays), diffraction occurs from
Two theta (degree)
crystals with lattice spacing on the order of 0.50 to
0.05 nm.

Energy Figure 20. Energy spectra for molybdenum

X-ray tube.
Aerospace applications of X-ray diffraction
generally use X-ray generators in the range of 20 to
40 keV, appropriate for investigating the crystalline
structure of metals such as nickel, titanium, 6
X-ray intensity (relative units)

Figure 18. Diagram of bragg diffraction. 4

X-rays (wavelength) 3
Ray 2 n = ab + bc = 2d sin 
Ray 1 2

1

0
a c 0 10 20 30
Lattice
d spacing
Wavelength (nm)

OTHER METHODS 17.15

 the phases. Also, residual stress measurements on
Coverage: Beam Shape and anisotropic material can be misleading because the
Penetration part has to be tilted with respect to the beam during
Typically, X-ray diffraction systems are used to the measurement, and the modulus may vary with
investigate small volumes of material with a single orientation.
measurement. The divergent beam that leaves the
X-ray tube is often masked to illuminate areas with
dimensions of 2.5 mm (0.1 in.) and with an effective Detection Means
penetration of about 13 µm (5 ×10–4 in.). Hence, for The designs of diffraction equipment are various,
evaluation of polycrystalline material, fine grains each adapted for the measurements made. Figure 21
provide opportunities to diffract the X-rays and shows a basic diffractometer where the specimen to
provide representative measurements. If the material be evaluated is placed at the center and rotates
is heterogeneous or anisotropic, however, diffraction throughthe angle theta  while the detector moves
results can be misleading. Such a case might occur if on the arc that traces 2. Sweeping through angles
material with relatively large grains is used for a of 2 from approximately 10 to 150 degrees provides
retained austenite measurement where the enough diffraction peaks to fully identify crystalline
amplitudes of diffraction peaks are directly related to structures and measure the dimensions of the
crystals. An example of an application wouldbefor
material identification where crystal types such as
Figure 21. Typical X-ray diffractometer used for body centered cubic, face centered cubic, hexagonal,
material identification or residual stress orother crystalline structures are known forgiven
measurements. materials. Full diffraction patterns are most often
used for alloy development rather than
nondestructive testing.
For other applications, such as retainedaustenite
measurements, only a few of the diffraction peaks
are required to measure the relative amounts of
austenite (a body centered cubic structure) and
ferrite (a face centered cubic structure). For such
cases, a detector that receives diffracted X-rays over
a limited range at one moment can collect data as
quickly as possible. Such linear array detectors
improve productivity significantly. Retained
austenite measurements are most often used to
evaluate heat treatment for process control rather
than as a nondestructive testing tool for every
article.
Another type of measurement that uses only one
diffraction peak is the measurement of residual
stress. A portable residual stress measurement system
is shown in Figure 22. This system can be used in
the field on an aircraft structure in service. For such
measurements, a linear array detector can be placed
at the appropriate 2 angle to collect the reflection
from one full diffraction peak. Figure 23 shows a
closeup of a titanium cantilever being measured in

Figure 22. Portable X-ray diffraction system for aerospace applications.

Figure 23. Experimental configuration for X-ray
diffraction to measure residual stress in titanium
cantilever. Red arrows indicate X-ray beam path.

Detector Detector
such a system with two linear detectors on either
Figure 24. Photograph of laue pattern of single
side of the X-ray output port. The detector/X-ray
tube is then rotated to a series of positions/angles crystal.
relative to the surface of the part, and the detector
records the new locations of the diffraction peaks,
recording small changes in the lattice spacing that
define strain in the direction of sample rotation.
Stress is calculated from the measured strain.
Residual stress is generally used as a process control
tool to evaluate the results of peening operations,
machining methods, or other processes where
residual stress may either reinforce or degrade apart.
One application that uses X-ray diffraction for
nondestructive testing of every article is crystal
orientation evaluation for single crystal turbine
blades. With the laue method, X-rays are diffracted
from a small region of the root or the tip of the
blade, and a pattern of diffraction spots defines the
orientation of the crystal. An example of a back
reflection laue pattern is shown in Figure 24, where
the diffraction pattern is recorded on film. In the
back reflection method, the film is held in a cassette
and the X-ray beam passes through the center of the
cassette, diffracts from the lattice planes of the
sample, and forms a pattern on the film. The pattern
of diffraction spots is known, and the rotation of the
pattern corresponds directly totherotationofthe
crystal with respect to the coordinate system of the
blade. For most production facilities, electronic
detectors have replaced film.

OTHER METHODS 17.17

PART5 X-Ray Fluorescence Spectroscopy

Energy dispersive X-ray fluorescence schematically how X-ray fluorescence spectroscopy

spectroscopy is an analytical method in which the is applied. Low energy X-raysorgamma rays froma
emissions characteristic of particular elements are source excite characteristic X-rays of elements in the
measured to indicate their prevalence in a test test sample, usually an unknown metal. The energy
object‘s material (ASTM 2013). The method can be of radiation from the excitation source must be high
used in the laboratory or in the field to identify enough to ionize the atoms in the sample by ejecting
materials. Modern handheld instruments are electrons from the innermost shells. When the
available for rapid alloy assessment in the shop electron from a higher shell fills the vacancy left by
or field (Niton 2010). the ejected electron on a lower shell, an X-ray
photon is emitted whose energy is characteristic and
unique for the atom from which it originated. These
Physical Principles characteristic X-rays are registered by a detector that
generates an X-ray spectrum of the sample to
X-rayfluorescence spectroscopy works onthe measure the X-rays‘ energies and intensities. By
principle of excitation of a sample at one or more analyzing such spectra, the instrument can quantify
energies and then measuring theemitted response of elements in the sample (ASNT 2010b).
the materialunder test. Figure 25 shows Table 4 lists examples of X-ray fluorescence
energy lines fora selection of elements. Figure 26
shows a schematic of a possible spectrum for an
Figure 25. X-ray fluorescence spectroscopy method. aluminum alloy. Computer analysis of the spectrum
can identify the elements and their concentration.
Energy source This information can be provided directly to the
Emitted radiation
operator, or a computer can identify the alloy of the
detector test object.
Emitted radiation

Instrumentation
The X-ray fluorescence method has been made
Gamma rays or X-rays
portable by development of handheld instruments.
An example is shown in Figure 27. These units can
Unknown metal
use an X-ray source ora gamma isotope. The X-ray
sources are usually in the range of 35 to 40 keV.
Sealed radioactive sources such as cadmium-109,
iron-55 or americium-241 are also available. The
energy of the source must be higher than the
fluorescent radiation desired to detect elements of
interest. For heavy elements, a high energy source is
required. For light element detection, the emission
Figure 26. Schematic of X-ray fluorescence spectrum for
2024 aluminum (UNS A92024).
Figure 27. Handheld X-ray fluorescence device.

Al
Intensity

Cu
Mg
Mn
0 2 4 6 8 10
Energy (keV)
 Table 4. Energy lines for X-ray fluorescence from selected elements.

Atomic Number Element Line Wavelength (nm) Energy (keV)

3 lithium (Li) K 2.8 .0543

4 beryllium (Be) K 11.4 0.1085
5 boron (B) K 6.76 0.1833
6 carbon (C) K 4.47 0.277
12 magnesium (Mg) K1,2 0.9890 1.254
13 aluminum (Al) K1 0.8339 1.487
14 silicon (Si) K1 0.7126 1.740
22 titanium (Ti) K1 0.2749 4.511
23 vanadium (V) K1 0.2504 4.952
24 chromium (Cr) K1 0.2290 5.415
25 manganese (Mn) K1 0.2102 5.899
26 iron (Fe) K1 0.1936 6.404
27 cobalt (Co) K1 0.1789 6.930
28 nickel (Ni) K1 0.1658 7.478
29 copper (Cu) K1 0.1541 8.048
30 zinc (Zn) K1 0.1435 8.639
40 zirconium (Zr) K1 0.07859 15.78
41 niobium (Nb) K1 0.07462 16.62
42 molybdenum (Mo) K1 0.07094 17.48
47 silver (Ag) K1 0.05594 22.16
48 cadmium (Cd) K1 0.05350 23.17
50 tin (Sn) L1 , K1 0.3600, 0.04906 3.444, 25.27
73 tantalum (Ta) L1 , K1 0.1522, 0.02155 8.146, 57.53
74 tungsten (W) L1 , K1 0.1476, 0.02090 8.398, 59.32
82 lead (Pb) L1 , K1 0.1175, 0.01654 10.55, 74.97

energy may be very low. Therefore, a very sensitive

detection system must be used to evaluate light Figure 28. X-ray fluorescence output of elemental materials to
elements. The handheld instruments can be adjusted differentiate alloys by magnesium content.
and optimized for light or heavy alloy sorting.

Nondestructive Testing
Applications
In the aerospace industry, the X-ray fluorescence
systems are used for alloy testing. A particular
example is the testing of fasteners (Niton 2013).
Improper fasteners present a risk to the structural
integrity of an aircraft. Using an X-ray fluorescence
system, fastener alloy content can be easily checked
as shown in Figure 28. In this case, the right-hand
image shows the desired magnesium content in the
aluminum alloy.

OTHER METHODS 17.19

PART6 Fourier Transform Infrared Testing

The fourier transform infrared (FTIR) method materials. With the increasing use of fiber reinforced
performs spectral analysis sensitive to the chemical plastics in aerospace applications, the need for the
constituents at the surface of a substrate. The method in the field has grown and portable systems
method uses infrared radiation and measures the have been developed (Figure 29) (Arnaud 2011).
frequencies of absorbance or reflectance to Applications of these instruments in aerospace
characterize the chemical signatures of organic include the detection of contamination before
materials. Fourier transform infrared instruments are bonding; degradation of polymers due to heat,
used in both the laboratory and the field. Modern chemical, or environmental damage; and coating
portable instruments are bringing fourier transform thickness (Seelenbinder 2007; Seelenbinder 2009).
infrared applications to the aerospace factory and
inservice inspection needs such as characterizing
surface contamination, heat damage, and resin Applications
thickness.

Heat Damage
Physical Principles Fourier transform infrared spectroscopy can be
The fourier transform infrared method consists of used to measure incipient heat damage in the resin
illuminating a surface with infrared radiation, component of a composite material. As an organic
detecting the reflected infrared radiation, and based material, a composite is susceptible to thermal
converting the signals into a frequency spectrum. damage if temperatures exceed a threshold. In
The fourier transform infrared technique works on aircraft, this can occur for a number of different
the principle of absorption and reflectance of reasons, but the overall effect is thermally induced
radiation in the near infrared range and infrared molecular changes in the resin component of the
range of the electromagnetic spectrum. The infrared composite. Infrared spectroscopy readily detects
region is generally divided into the near, mid, and several different mechanisms forthese molecular
far regions. Table 5 lists the wavelengths and changes. In an aerobic process (at the composite
energies of electromagnetic waves around the surface), thermal damage is evidenced by the
infrared region. The infrared light excites vibrations formation of carbonyl absorbance bands (carbon
in molecular bonds that cause absorption at specific oxygen double bonds) in the 1700 to 1750 cm–1
wavelengths. These absorption bands then become a region of the mid infrared spectrum. In subsurface
method to identify particular chemical species. The anaerobic processes, there are subtle changes
spectral content is a function of the molecular throughout the infrared spectrum as key bonds in the
characteristics of the surface being inspected and resin backbone are broken or otherwise changed.
how those chemical species absorbthe illumination. These molecular changes can lead to physical
The mid and near infrared ranges are useful for changes and weakening ofthe composite structure,
aerospace materials such as organic resins and resulting in cracks and delamination. Particularly
composites. challenging is the detection of incipient thermal
damage where damage may not be visible or where
damage may extend beyond the area of visible
Instrumentation damage.
Diffuse reflectance fourier transform infrared
Fourier transform infrared systems in the spectroscopy is used to detect and track the
twentieth century were laboratory devices used to molecular changes in composites induced by thermal
characterize themolecularchemistry ofpolymer overexposure. Predictions can be made as to the

Table 5. Infrared spectral regions.

Region Wavelength (µm) Wave Number (cm–1) Frequency (THz)

Near infrared 0.7 to 2.5 14 000 to 4000 420 to 120

Mid infrared 2.5 to 25 4000 to 400 120 to 12
Far infrared 25 to 1000 400 to 10 12 to 0.3
amount of thermal exposure a composite has been composite surfaces requires an elliptical lens
subjected to through calibration models developed assembly to gather radiation reflected diffusely; due
for the resin system of interest. Calibration models to overlapping absorbance from the substrate resin,
are developed by correlating spectra of known limits of detection on composite surfaces are
samples to the level of thermal exposure in those typically higher than on metallic surfaces
prepared samples through partial least squares or (Seelenbinder 2011).
another multivariate calibration technique. Handheld
fourier transform infrared instruments are available
precalibrated for resin systems; their performance is Polymer Coatings
verified against high and low reference standards. Polymer coatings onmetal surfaces must meet
The test is useful for determining the depth and specifications for the metal component to perform in
breadth of thermal damage to help sanding and service. Specular reflectance infrared spectroscopy is
patching for repairs. The recent availability of used to determine the thickness of a polymer coating
handheld fouriertransform infrared spectrometers on a metal surface as well as profile the uniformity
enables these measurements to be made ofthecoating onthe surface. Measuring the extent
nondestructively on large objects such as aircraft in of curing of a polymer film on metal surfaces is
support of aircraft manufacturing and maintenance another application for the specular reflectance
operations (Higgins 2014). fourier transform infrared technique. A handheld
instrument permits nondestructive measurements of
large areas of coated metal. By using an onboard
Surface Contamination and spectral database, the identity of the polymer on a
Treatment coated metal surface is ascertained to ensure that the
Release agents used in the process of forming correct compound was applied. Pretreatment of the
composite parts typically must be carefully removed metal surface is often necessary for good polymer
by plasma treatment to ensure that bonding is bonding. The changes that mayoccur in a polymer
effective. If the plasma treatment is ineffective, the coating as a function of use, ambient conditions, and
surface will retain some of the release agent and degradation are tracked nondestructively by a
may not be properly activated for optimized handheld fourier transform infrared instrument
bonding. If the plasma treatment is too aggressive, independently from the size, location, or orientation
the residual surface contaminants will be removed of the test object (Seelenbinder 2010).
but the underlying composite may be thermally
damaged. In metals, the presence ofcontaminants on
surfaces can affect bonding processes as well, and
primers and surface coatings will not adhere Figure 29. Handheld fourier transform infrared instruments:
properly. Diffuse reflectance mid infrared (a) models; (b) application.
spectroscopy can track the effectiveness of plasma
treatment of composite surfaces to ensure that the (a)
residual release agent is removed and surfaces are
properly activated for optimizing bonding (Rein
2014).
The presence of unwanted organic surface
contaminants can affect the manufacturing and
repair of both composite and metal objects. For
composites, the presence of residual contaminants
such as silicone or hydrocarbon oil, which may be a
result of manufacturing processes or from airborne
mists, can affect composite-to-composite bonds and
weaken bonds. Surface contamination can be
measured by using handheld fourier transform (b)
infrared spectroscopy. Calibration curves are
developed by measurement of standards containing
known amounts of contamination; typically
quantities ofcontamination are expressed asan area
concentration, such as micrograms per square
centimeter (µg/cm2). Surface contaminants, such as
silicone, have unique infrared frequencies; the
intensity of the absorbance at those frequencies
directly correlates to the amount of the contaminant
present. Uncoated metal substrates allow forthe
most sensitive measurements; a grazing angle
sample interface can obtain limits of detection
approaching 0.2 µg/cm2. Direct measurement on

OTHER METHODS 17.21

PART7 Beta Backscatter

Coating thickness must be inspected in the backscatter is anadditional method that can be
aerospace industry, and various methods are called on for coating inspection.
available to measure it (Lomax 2014). Eddy current Developed in thetwentieth century for printed
liftoff and high frequency ultrasound are useful for circuit board plating measurement, the beta
checking coatings, depending on the coating and backscatter method has beenlargely replaced in that
substrate materials and on the thickness range. For applicationby X-rayfluorescent techniques because
example, eddy current is excellent for paint they can be used without contacting the test surface.
thickness over aluminum, and ultrasound is useful The beta backscatter technique, however, remains
for coatings with significant acoustic impedance useful for several aerospace applications such as
change between coating and substrate. Beta aluminide coatings on turbine blades and stealth
coatings (Justice 2008).

Table 6. Beta emitters useful for beta backscatter method.

Physical Principles
The beta particle is an energetic electron emitted
Source Energy Half Life (Years) from an atomic nucleus during radioactive decay. If
a stream of beta particles is directed at a sample
surface, they will interact with thematerial and be
C-14 (carbon) 160 keV 5751 scattered in different directions. A beta particle is
Pm-147 (promethium) 220 keV 2.8 said to be backscattered if it leaves an object from
the same side as the one it has entered. The
Tl-204 (thallium) 770 keV 3.0
penetration depth of beta particles into a material is
Ra (radium) 1170 MeV 19.4 a function of the electron density of the material and
Sr-90 (strontium) 2270 MeV 28.0 the energy of the beta particle, which depends on the
Ru-106 (ruthenium) 3540 MeV 1.0 type of isotope. Table 6 lists some common beta
particle emitters that can be used for coating
inspection (ASTM 2009). In addition, the ability of a
material to backscatter a beta particle also depends
on its electron density or atomic number. This ability
is represented by the backscatter coefficient and is
Figure 30. Plot of backscatter coefficient versus atomic number defined as the ratio of backscattered beta particles to
(ASNT 1985). impinging beta particles. The backscatter coefficient
as a function of atomic number is shown in
Figure 30. As the atomic number increases, the
Pb backscatter coefficient increases and results in a
higher measuredcountrate. As the coatingthickness
0.6 Sn W increases, the number of backscattered beta particles
0.4
Ag increases up to a maximum, at which the material is
Au considered to have saturation thickness. The
Backscatter coefficient

Cu
Fe saturation thickness will vary depending on the
Rh energy of the incoming beta particles; as the energy
0.2
Ni increases, the saturation thickness increases and
Cr
thicker coatings can be measured.
0.1 When coated samples are measured, two extremes
Si in count rate can be observed. The count rate X0is
Al
for the uncoated base material, and the count rate Xs
is for the coating at saturation thickness. For a
0.05 C
coating less than saturation thickness, the measured
count rate X lies between X0 and Xs, the actual value
Be
depending on the object thickness. The measurement
of these count rates is the basis of the thickness
measurement. In practice, a normalized count rate is
5 10 20 40 60 80 used instead of the absolute count rate. The
normalized count rate Xn will vary between 0 and 1
Atomic number Z and is given by Equation 7.
 XX0 Figure 31. Schematic of beta backscatter system.
(7) X n 
XsX0

Because the count rate depends on the atomic

number of the test object, this measurement
technique can be affected by compositional changes
in the test material, which can lead to experimental
error. Conversely, the backscatter method can be
used to determine alloy compositions. Detector

Instrumentation
Figure 31 is diagram of a beta backscatter system. Beta source
Backscattered beta
The probe system is placed in contact with the
coating, and only beta particles that pass through
the probe aperture will interact with the coating and Aperture
base material. The backscattered beta particles are Coating
measured with a geiger müller tube, and the signal
strength is proportional to the scattering property of Substrate
the coating and substrate. The count rates are also
influenced bythe size ofthe probe aperture. As the
aperture decreases in diameter, fewer electrons can
return to the counter and the absolute difference in
X0 and Xs becomes smaller, thus increasing the
relative error of measurement. For this reason, it is
advisable to use the largest possible aperture for a
particular set of conditions. as aluminum on nickel, chromium on aluminum, and
tinon nickel. The method can also measure paint,
plastic, enamel, or ceramic coatings over metal.
Beta backscatter instruments require calibration
Applications standards and are commonly used as laboratory
Beta backscatter instruments can measure tools, but a portable system weighing less than
coatings where the substrate and coating have a 0.9 kg (2 lb) has been developed for field use.
difference in atomic number of approximately 4 — Because these units use radioactive sources, their
for example, zinc coatings on iron. Other metal storage and handling are regulated in state and
coatings on metal substrates are also possible, such federal jurisdictions.

OTHER METHODS 17.23

PART8 Resonance and Vibration Testing

Resonance acoustic testing and vibration testing

are techniques that effectively listen to a part under
Physical Principles
a particular test or during operation and use the Parts or structures will have natural resonance
detected signature of the received signal to modes of vibration occurring at specific frequencies
determine the condition. The resonance acoustic that are afunction ofmaterial properties, such as
measurements are oftenreferred to asmodalanalysis elasticmodulus, density, and thepart‘sphysical size
of the natural resonant frequencies of a part. The test and geometry. When a part is vibrated, either in a
is often a mechanical tap of some type on the sample continuous mode, orby an impulse, the modes of
at atest stand or could be an actuator placed onthe vibration will come into play and the part will ring
part. When the part is damaged, such as by cracking, atitsresonant frequencies. Transducers are used to
the natural frequencies may shift sufficiently to be detect the surface motions or transmitted acoustic
detected. The sensing mechanisms range from vibrations in air. The signature from the transducer
noncontacting acoustic microphones, to contact placed to capture each signal characterizes it. For
sonic or ultrasonic and noncontacting surface impulse resonance systems, typically a fast fourier
displacement measurements. transform is performed on the data to create a
frequency domain spectrum. The signal is digitized
with a sampling rate based on the highest frequency
of interest, Fmax. The nyquist criterion requires the
Figure 32. Fast fourier frequency spectrum from vibration test sampling to be twice Fmax. Figure 32 is an example
(ASNT 2010a, 461). of a frequency spectrum from an impulse vibration
test. For continuous sweptsine wavesystems, data
can be discretely collected across a broadband
160 frequency range, generally ranging from a few
140 hundred Hz to many MHz. Figure 33 illustrates a
continuous swept sine frequency spectrum. The
120
frequency spectrum can be analyzed to assess the
Amplitude (mG)

100 characteristics of the part of under test. Typically,

80
many frequency constituents will need to be
assessed.
60

40

20 Instrumentation
0
There are three main schemes for resonance
sensors and instrumentation, according to
0 200 400 600 800 1000 1200
application needs and corresponding techniques:
Frequency (Hz) (1) impulse resonance vibration testing, (2) process
compensated resonance testing, and (3) scanning
vibrometry.

Figure 33. Continuous swept sine wave frequency spectrum from resonance test.
Amplitude (relative depiction

179.950 240.185
Frequency (kHz)
 1. A basic configuration for resonance vibration Resonance Ultrasound Spectroscopy
analysis is to place a sensor on or near the part.
In some cases, an acoustic microphone placed Resonance ultrasound spectroscopy measures
near the part is sufficient. In other cases, one or resonant frequency to determine nondestructively
more sensors may be mounted to the structure. the elastic moduli of carefully prepared samples of
2. Process compensated resonance testing uses two specific geometries. Resonance ultrasound
or more piezoelectric transducers in contact with spectroscopy can provide an accurate means of
the part. One transducer is used to drive the part quantifying an elastic modulus by comparing models
in a swept sine pattern, and the other transducer with measurements (Migliori 2005).
or transducers receive the resonance vibration of First, models are generated to predict resonant
the part. Process compensated resonance test frequencies based on input sample dimensions and
data are not processed via fast fourier transform. initial estimates of the elastic moduli. The modeled
The hardware combining transceiver and frequencies are compared to measured frequencies,
transducer operates from 500 Hz to 12 MHz and the model is iteratively adjusted to fit the data.
(Figure 34). Some applications of process The result is a set of nondestructively measured
compensated resonance testing are discussed material properties for the sample part. Sample parts
below. can be spheres, right cylinders, or parallelepipeds
3. Vibration can be measured with noncontacting made of materials having isotropic, cubic,
laser vibrometers. Surface displacement is hexagonal, tetragonal, ororthotropic crystalline
measured by interferometric principles using a geometry. Generally, the samples have dimensions
laser source (Kilpatrick 2008). The laser on the order of millimeters.
vibrometer may be operated in a fixed position
or be scanned over a surface to capture the Process Compensated Resonance
modal surface vibrations of a structure. It can
also be used to visualize wave propagation in a Testing
structure. Process compensated resonance test systems
combine proprietary software algorithms that detect
structural changes in components and compensate
for normal manufacturing parameter variations that
Nondestructive Test may obscure out-of-specification conditions. Process
Applications compensated resonance testing can be applied in
many different ways, depending on the application.
Process compensated resonance test software is used
Impulse Resonance Vibration Testing to develop sorting modules that evaluate frequency
responses oftested components. Withthe sorting
Impulse resonance vibration testing is basically modules, the software provides rapid, accurate, and
experimental modal analysis simplified for operator independent pass/fail assessments. Such
application to high volume production assessments canincrease production yield, optimize
manufacturing and quality control testing. The part life, and significantly reduce component failures
generic, step-by-step procedure follows (Stultz 2014). in the field (Jauriqui 2010).
1. Excite the part with a known and repeatable Population Characterization, Outlier Screening,
force input. This force is typically generated by and Process Monitoring. One of the most frequent
a controlled impact or actuator providing applications of process compensated resonance
broadband or sinusoidal energy over the testing is for population characterization and outlier
frequency range of analysis. screening. Precise resonance measurements can
2. Measure the structural response of the part to characterize the consistency of a component
the applied input force using a dynamic sensor population, as a result either of the manufacturing
such as a microphone or accelerometer process or service history. The technique accurately
(vibration pickup) and a swift analog-to-digital
converter with antialiasing filters.
3. Process the acquired time data with a fast fourier Figure 34. Low volume hardware for process
transform for analysis in the frequency domain. compensated resonance testing, including signal
4. Analyze the consistency of the frequency generator/processor, control computer, and
spectrum from part to part by comparing each
testing fixture (right) with temperature display.
to a spectral template created from known good
(structurally sound) parts. Mechanical
resonances are indicated as peaks in the
frequency spectrum of the response. Good parts
have consistent spectral signatures (that is, the
mechanical resonances are the same among
parts whereas anomalous parts have irregular
spectral signatures. Generally, these templates
are set to evaluate the consistency of the
frequency and amplitude of ten or fewer peaks.
Any deviation from the specified range of peak
frequency or amplitude constitutes a structurally
significant difference and provides a
quantitative and objective criterion for rejection.

OTHER METHODS 17.25

 measures the structural similarity of a component to Targeted Discontinuity Detection. Process
a reference population of good parts. The acceptable compensated resonance testing can also be used for
variation is defined by the reference set of good targeted anomaly detection, which accurately
parts, and can thereafter be used to determine if a measures the structural similarity of a component to
test part falls within the allowed tolerances. The a known population of good and bad parts. In this
tolerance limits can be adjusted to accept predicted test, the technique‘s genetic pattern recognition
levels of process variation. The test yields a simple algorithm identifies resonance patterns defining
pass/fail result and a statistical score indicating the both the good and bad part populations and assigns
test part‘s position in the good population. a probability that a tested part belongs to either
Resonance measurements can also be used for group. The test provides a simple pass/fail decision
process monitoring in service. When a component is and a statistical score indicating the part rank in
first designed, approved, and tested, resonance can both good and bad populations.
be measured to describe the structure‘s response Figure 35 illustrates an application of process
characteristics for comparison with future compensated resonance test technology to anomaly
components. Statistical process control analysis can detection in aircraft brake parts (Hunter 2010) The
be used with resonance data and additional figure compares spectra of good parts to spectra of
processing information to assess process capability, anomalous parts. The consistency of resonance peak
to discover manufacturing contributions to positions between good parts is very high, but in
component variation, and to ensure that products fit damaged parts the peak positions and patterns can
their specifications (Piotrowski 2008; Schwarz 2005). be significantly disrupted.
Life Monitoring. Life monitoring accurately tracks
changes in a part‘s resonance signature throughout
its lifetime. Process compensated resonance test
spectra are extremely sensitive to changes within
Figure 35. Process compensated resonance test spectra of good and the same component. Spectra can be collected from
bad components. a given part and compared each time it visits the
shop. Part variation due to repair does not confuse
4 5 6 part comparisons: the part is always compared only
Good parts to itself. Resonance changes detected are due to
4 5 6 — consistent changes in the part‘s structural integrity and can be
repeatable used to evaluate the accumulation of fatigue,
spectra damage, or other life limiting conditions. This
4 5 6 evaluation offers opportunities for life extension for
parts currently constrained by runtime limitations.
4 5 6 Part life monitoring can provide either a simple
pass/fail result or a more in-depth review of
resonance data.
5 6
Figure 36 shows a part life monitoring study
performed on UNS A96061 temper 6 (6061-T6)
Gouged parts
4 56
— significant aluminum sample coupons. The figure shows spectra
peak shifts for the same part as it accumulated mechanical
fatigue cycles. The spectrum shifts to the left as
4 5 6
fatigue accumulates, with accelerated shifting and
pattern changes at the onset of cracking.

Figure 36. Fatigue crack detection in study monitoring part life.

New

10 000 cycles

20 000 cycles

30 000 cycles

35 000 cycles

40 000 cycles

42 500 cycles

Cracking at 45 000 cycles

References
Altpeter, I., G. Dobmann, S. Fabender, J. Hoffmann, J. Johnson, Buttle, D.J., E.A. Little, C.B. Scruby, G.A.D. Briggs, and J.P.
N. Meyenndorf, and W. Nichtl-Pecher. 1998. ―Detection of Jakubovics. 1987a. ―A Study of Neutron Irradiation Damage
Residual Stresses and Modular Growth in Thin of -Iron with Magnetoacoustic and Barkhausen Emission.‖
Ferromagnetic Layers with Barkhausen and Acoustic Proceedings of the Royal Society of London, Series A:
Microscopy.‖ Nondestructive Characterization of Materials 8 Mathematical, Physical and Engineering Sciences 414(1846):
[Proceedings of the Eighth International Symposium, 221-236. doi: 10.1098/rspa.1987.0141.
Boulder, CO, June 1997]. New York, NY: Springer, 677. Buttle, D.J., C.B. Scruby, G.A.D. Briggs, and J.P. Jakubovics.
Altpeter, I., and G. Dobmann. 2003. ―NDE of Material 1987b. ―The Measurement of Stress in Steels of Varying
Degradation by Embrittlement and Fatigue.‖ Review of Microstructure by Magnetoacoustic and Barkhausen
Progress in Quantitative Nondestructive Evaluation 22A Emission.‖ Proceedings of the Royal Society of London,
[Bellingham, WA, July 2002]. AIP Conference Proceedings Series A: Mathematical, Physical and Engineering Sciences
657. Melville, NY: American Institute of Physics, 15-21. 414 (1846): 469-97.
[Link]/10.1063/1.1570115. Buttle, D.J., [Link], C.B. Scruby, and R.A. Langman. 1990.
Arnaud, C.H. 2011. ―Handheld IR in the Hangar.‖ Chemical and ―Comparison of Three Magnetic Techniques for Biaxial
Engineering News 89 (34), 43-45. Stress Measurement.‖ Review of Progress in Quantitative
ASNT. 1985. ―Beta Backscatter.‖ Nondestructive Testing Nondestructive Evaluation 9. New York, NY: Plenum Press,
Handbook 3: Radiography and Radiation Gaging, 3rd 1879-1885.
edition. Columbus, OH: American Society for Buttle, D.J., W. Dalzell, M.T. Hutchings, and A.J. Allen. 2000.
Nondestructive Testing, 700-703. ―Residual Stress in Butt Weldments of 50D Steel Measured
ASNT. 2004. Nondestructive Testing Handbook 5: by Neutron Diffraction and Magnetic Techniques.‖ ICRS-6,
Electromagnetic Testing, 3rd edition. Columbus, OH: Sixth International Conference on Residual Stresses 2
American Society for Nondestructive Testing, 346-349. [Oxford, United Kingdom, July]. Grantham, Lincolnshire,
ASNT. 2010b. ―X-Ray Fluorescence Spectrometry.‖ United Kingdom: Institute of Materials, Minerals and
Nondestructive Testing Handbook 10: Nondestructive Testing Mining, 923-931.
Overview, 3rd edition. Columbus, OH: American Society for Ceurter, J.S., C. Smith, and R. Ott. 2002. ―The Barkhausen Noise
Nondestructive Testing, 495. Inspection Method for Detecting Grinding Damage in
ASNT. 2010a. Starry, K.A. ―Vibration Analysis.‖ Nondestructive Gears.‖ Gear Technology (November/December): 26-31.
Testing Handbook 10: Nondestructive Testing Overview, 3rd Crouch, A.E., and T. Beuker. 2004. ―In-Line Stress Measurement
edition. Columbus, OH: American Society for by the Continuous Barkhausen Method.‖ Proceedings of
Nondestructive Testing, 457-471. International Pipeline Conference 2004 [Calgary, Alberta,
AST. 1990. European Patent 372 030, ―Biaxial Stress Canada]. New York, NY: American Society of Mechanical
Determination Method for Ferromagnetic Materials Engineers, 997-1003. IPC2004-0233/doi: 10.1115/IPC2004-
Generating Calibration Data from Test Piece, Measuring and 0233.
Calibrating Barkhausen Noise Level‖ (June 1990). Desvaux, S., M. Duquennoy, J. Gualandri, and M. Ourak. 2004.
Pittsburgh, PA: American Stress Technologies. ―The Evaluation of Surface Residual Stress in Aeronautic
doi:10.1016/0963-8695(91)90693-W. Bearings Using the Barkhausen Noise Effect.‖ NDT&E
AST. 2014. Barkhausen Noise Analysis. Pittsburgh, PA: International 37(1): 9-17. doi: 10.1016/S0963-
American Stress Technologies. Accessed 15 May 2014, 8695(03)00046-X.
[Link] Dhar, A., L. Clapham, and D.L. Atherton. 2001. ―Influence of
ASTM. 2009. B567-98, Standard Test Method for Measurement Uniaxial Plastic Deformation on Magnetic Barkhausen Noise
of Coating Thickness by the Beta Backscatter Method. West in Steel.‖ NDT&E International 34(8): 507-514.
Conshohocken, PA: ASTM International. doi:10.1016/S0963-8695(00)00079-7.
ASTM. 2013. ASTM E 572, Standard Test Method for Analysis Dobmann, G., I. Altpeter, R. Becker, R. Kern, U. Laub, and W.A.
of Stainless and Alloy Steels by Wavelength Dispersive Theiner. 1998. ―Barkhausen Noise Measurements and
X-Ray Fluorescence Spectrometry. West Conshohocken, PA: Related Measurements in Ferromagnetic Materials.‖ Topics
ASTM International. in Nondestructive Evaluation 1: Sensing for Materials
Auld, B.A., A.V. Clark, S.R. Schaps, and P.R. Heyliger. 1993. Characterization, Processing and Manufacturing. Columbus,
―Capacitive Probe Array Measurements and Limitations.‖ OH: American Society for Nondestructive Testing, 233-251.
Review of Progress in Quantitative Nondestructive Fix, R.M., K. Tiitto, and S. Tiitto. 1990. ―Automated Control of
Evaluation 12A [La Jolla, CA, July 1992]. New York, NY: Camshaft Grinding Process by Barkhausen Noise.‖ Materials
Plenum, 1063-1070. Evaluation 48(7): 904-908.
Bach, G., K. Goebbels, and W.A. Theiner. 1988. Higgins, F. 2014. Application Note, Non-Destructive Evaluation
―Characterization of Hardening Depth by Barkhausen Noise of Composite Thermal Damage with Agilent’s New Handheld
Measurement.‖ Materials Evaluation 46(11): 1576-1580. 4300 FTIR. Danbury, CT: Agilent Technologies.
Barkhausen, H. 1919. ―Zwei mit Hilfe der Neuen Verstarker IEEE. 2009. IEEE 521-2002, IEEE Standard Letter Designations
Entdeckte Erscheinungen [Two Phenomena Revealed with for Radar-Frequency Bands. New York, NY: IEEE Aerospace
Help of New Amplifiers].‖ Physik Zeitschrift 20: 401-403. and Electronic Systems Society.
Barton, J.R., and F.N. Kusenberger. 1974. ―Residual Stresses in Jagadish, C., L. Clapham, and D.L. Atherton. 1989. ―The
Gas Turbine Engine Components from Barkhausen Noise Influence of Stress on Surface Barkhausen Noise Generation
Analysis.‖ Journal of Engineering for Power (October): in Pipeline Steels.‖ IEEE Transactions on Magnetics 25(5):
349-357. 3452-3454.

OTHER METHODS 17.27

Jauriqui, L. 2010. ―A More Comprehensive NDE: PCRT for Niton. 2013. Quality Assurance of Critical Fasteners: Thermo
Ceramic Components.‖ Ceramic Engineeringand Science Scientific Portable XRF Analyzers Enable Rapid Elemental
Proceedings 31: Mechanical Properties and Performance of Analysis. Billerica, MA: Thermo Scientific.
Engineering Ceramics and Composites 5. Hoboken, NJ: Ogden, W.P. 1991. ―Detection of Subsurface Tensile Stress in an
Wiley, 145-156. doi:10.1002/9780470944127. Aircraft Engine Main Shaft Bearing Using Barkhausen
Jiles, D.C., R. Kern, and W.A. Theiner. 1993. ―Evaluation of Noise.‖ Applications of Residual Stress Technology,
Surface Modifications in High Strength Steel.‖ Conference Proceedings [Indianapolis, IN, May]. Materials
Nondestructive Testing and Evaluation 10(6): 317-331. Park, OH: ASM International, 161-167.
Jiles, D.C. 1995. ―Theory of Magnetomechanical Effect.‖ Journal Pasley, R.L. 1970. ―Barkhausen Effect: An Indication of Stress.‖
of Physics D: Applied Physics 28:1537. Materials Evaluation 28(7): 157-161.
Justice, M. 2008. ―Beta Backscatter for Coating Thickness Piotrowski, D., L. Hunter, and T. Sloan. 2008. ―Process
Measurement: Still Viable After All These Years.‖ Metal Compensated Resonance Testing of JT8D-219 1st Stage
Finishing 106(10): 25, 28-30. Blades.‖ Presented at Air Transport Association NDT Forum
Kharkovsky, S., F. Hepburn, J. Walker, and R. Zoughi. 2005. 2008 [September 2008]. Accessed 23 May 2014,
―Nondestructive Testing of the Space Shuttle External Tank [Link]
Foam Insulation Using Near Field and Focused Millimeter PCRT_of_JT8D-T1_Blades.pdf.
Wave Techniques.‖ Materials Evaluation 63(5): 516-522. Ramuhalli, P.,N. Qaddoumi, and R. Botsko. 2004. Chapter 12,
Kharkovsky, S., and R. Zoughi. 2006. ―Millimeter Wave ―Microwave Testing.‖ Nondestructive Testing Handbook 5:
Nondestructive Evaluation of Corrosion under Paint in Steel Electromagnetic Testing, 3rd edition. Columbus, OH:
Structures.‖ Review of Progress in Quantitative American Society for Nondestructive Testing, 286-320.
Nondestructive Evaluation 25B [Brunswick, ME, August Ravuri, M., M. Abou-Khousa, S. Kharkovsky, R. Zoughi, and R.
2005]. AIP Conference Proceedings 820. Melville, NY: Austin. 2008. ―Microwave and Millimeter Wave Near-Field
American Institute of Physics, 1277-1283. Methods for Evaluation of Radome Composites.‖ Review of
doi:10.1063/1.2184671. Progress in Quantitative Nondestructive Evaluation 27B
Kharkovsky, S., and R. Zoughi. 2007. ―Microwave and [Golden, CO, July 2007]. AIP Conference Proceedings 975.
Millimeter Wave Nondestructive Testing and Evaluation: Melville, NY: American Institute of Physics, 976-981.
Overview and Recent Advances.‖ Instrumentationand doi:10.1063/1.2902772.
Measurement Magazine, IEEE 10(2): 26-38. Rein, A., and P.L. Tang. 2014. ―Analysis of Plasma Treated
doi:/10.1109/MIM.2007.364985. Carbon Fiber Reinforced Polymer (CFRP) Composites by
Kilpatrick, J., and V. Markov. 2008. ―Matrix Laser Vibrometer Portable Fourier Transform Infrared Spectroscopy.‖ JEC
for Transient Modal Imaging and Rapid Nondestructive Composites Magazine 87: 124-127.
Testing.‖ SPIE Proceedings 7098, Eighth International SAE. 1991. Aerospace Recommended Practice (ARP) 4462,
Conference on Vibration Measurements by Laser Techniques: Barkhausen Noise Inspection for Detecting Grinding Burns
Advances and Applications [June]. Bellingham, WA: SPIE. in High Strength Steel Parts. Warrendale, PA: SAE
doi:10.1117/12.802929. International.
Leep, R.W. 1967. ―The Barkhausen Effect and Its Application in Schmidt, K.F., J.R. Little, Jr., [Link], and [Link]. 2010.
Nondestructive Testing.‖ Physics and Nondestructive ―Optimizing a Portable Microwave Interference Scanning
Testing: Proceedings of the Symposium on Physics and System for Nondestructive Testing of Multi-Layered
Nondestructive Testing [Dayton, OH, September-October Dielectric Materials.‖ Review of Progress in Quantitative
1964]. New York, NY: Gordon and Breach, 439-453. Nondestructive Evaluation 29B [Kingston, RI, July 2009].
Little, J. 2010. ―Development of Scanning Microwave AIP 1211. Melville, NY: American Institute of Physics, 1249-
Technology for Ceramics in Extreme Environments.‖ 1256.
Advances in Science and Technology 65: 153-162. Schmidt, K.F., J.R. Little, Jr., and W.A. Ellingson.‖ 2013.
Lo, C.C.H., J.A. Paulsen, and D.C. Jiles. 2003. ―Development of a ―Microwave Interferometry Inspection of Dielectric
Magnetic NDE Imaging System Using Magnetoresistive Materials.‖ ASNT Annual Conference [Las Vegas, NV,
Devices.‖ Review of Progress in Quantitative Nondestructive November]. Columbus, OH: American Society for
Evaluation 22A. Melville, NY: American Institute of Physics, Nondestructive Testing, 193.
931-938. Schmidt, K.F. 2014. ―Quantitative NDE of CMCs by Microwave
Lomax, P.[J.] 2014. ―Trends in Coating and Thickness Interferometry, Validation.‖ Presented at 38th Annual
Measurement.‖ Quality Digest Magazine (May). Conference on Composites, Materials, and Structures [Cocoa
Matzkanin, GA., R.E. Beissner, and C.M. Teller. 1979. Report Beach, FL, January]. Washington, DC: Advanced Ceramics
NTIAC-79-2, ―The Barkhausen Effect and Its Applications to Association.
Nondestructive Evaluation.‖ San Antonio, TX: Southwest Schull, P.J., [Link], P.R. Heyliger, J.C. Moulder, and B.A.
Research Institute. Auld. 1990. ―Characterization of Capacitive Array for NDE
Migliori, A., and J.D. Maynard. 2005. ―Implementation of a Applications.‖ Research in Nondestructive Evaluation 2(1):
Modern Resonant Ultrasound Spectroscopy System for the 11-27.
Measurement of the Elastic Moduli of Small Solid Schwarz, J., J. Saxton, and L. Jauriqui. 2005. ―Process
Specimens.‖ Review of Scientific Instruments 76(12): Compensated Resonant Testing in Manufacturing Process
121301. doi:10.1063/1.2140494. Control.‖ Material Evaluation 63(7): 736-739.
Moorthy, V., B.K. Choudhary, S. Vaidyanathan, T. Jayakumar, Seelenbinder, J. 2007. Application Note, Handheld FTIR
K. Bhanu Sankara Rao, and B. Raj. 1999. ―An Assessment of Spectroscopy Applications Using the Agilent ExoScan 4100.
Low Cycle Fatigue Damage Using Magnetic Barkhausen Danbury, CT: Agilent Technologies.
Emission in 9Cr–1Mo Ferritic Steel.‖ International Journal Seelenbinder, J. 2009. Application Note, Composite Heat
of Fatigue 21(3): 263-269. Damage Measurement Using the Handheld Agilent 4100
doi:10.1016/S0142-1123(98)00079-6. ExoScan FTIR. Danbury, CT: Agilent Technologies.
Niton. 2010. Positive Material Identification (PMI). Billerica, Seelenbinder, J. 2010. Application Note, Epoxy Primer Thickness
MA: Thermo Scientific. on Aluminum Measured with the Handheld Agilent 4100
ExoScan FTIR. Danbury, CT: Agilent Technologies.
Seelenbinder, J. [Link] Note, Detection of Trace Zoughi, R., S. Kharkovsky, and F. Hepburn. 2005. NASA
Contamination on Metal Surfaces Using the Handheld Technical Report 20050215575, Microwave and Millimeter
Agilent 4100 ExoScan FTIR. Danbury, CT: Agilent Wave Testing for the Inspection of the Space Shuttle Spray
Technologies. on Foam Insulation (SOFI) and the Acreage Heat Tiles.
Stultz, G.R., R.W. Bono, and M.I. Schiefer. Manuscript, Huntsville, AL: National Aeronautics and Space
Fundamentals of Resonant Acoustic Method NDT. Administration, Marshall Space Flight Center.
Cincinnati, OH: The Modal Shop. Accessed May 2014, Zoughi, R., S. Kharkovsky, and F. Hepburn. 2006. ―Microwave
[Link]. and Millimeter Wave Testing for the Inspection of the Space
Tiitto, S. 1996. Handbook of Measurement of Residual Stresses Shuttle Spray on Foam Insulation (SOFI) and the Acreage
8: Magnetic Methods. Lilburn, GA: Society for Experimental Heat Tiles.‖ Review of Progress in Quantitative
Mechanics, 179-224. Nondestructive Evaluation 25B [Brunswick, ME, August
Von Hippel, A. 1954. Dielectric Materials and Applications. 2005]. AIP Conference Proceedings 820. Melville, NY:
Boston, MA: Artech House. American Institute of Physics, 439-446.
Zoughi, R. 2000. Microwave Non-Destructive Testing and
Evaluation. London, United Kingdom: Kluwer Academic
Publishers.

OTHER METHODS 17.29

 18CHAPTER

Formulas

Contents
Visual Testing, 18.2
Penetrant Testing, 18.2
Magnetic Particle Testing, 18.2
Electromagnetic Testing, 18.3
Ultrasonic Testing, 18.3
Radiologic Testing, 18.4
Thermographic Testing, 18.5
X-Ray Diffraction, 18.6
Visual Testing Magnification
Di Si
Wavelength and Frequency (7) M  
D0 S0
Wavelength  and frequency  of light are
related
by its velocity v: where D0 is object distance from optical
axis, Di is image distance from optical axis,
M is the

(1)v   magnification, S0 is object size, and Si is image

size.

where  is wavelength, v (italic Roman vee) is

velocity of light waves in the medium, and 
(Greek nu) is frequency.
Penetrant Testing
The forces involved in capillary action are
Energy given:
The energy of visible radiation is
expressed: 2 
(8) P 
R
(2) E  h where P is the capillary pressure, R is the
where E is energy,  (Greek nu) is frequency, radius of curvature of the liquid surface, and
and h  is the surface tension of the liquid.
is Planck’s constant:

(3) h  6.63  1034 J  s Magnetic Particle Testing

–15
 4.14  10 eV s
Magnetic Field Intensity
Index of Refraction Magnetic field intensity is
expressed:
The index of diffraction is described:
vv (9) M  m H
(4) n 
vm where M is magnetization in A/m (amperes
where n is index (degrees) of refraction, vm per meter), H is magnetic field strength, and
is the velocity in a medium, and vv is the m is magnetic susceptibility.
velocity in a vacuum.

Magnetic Induction
Angle of Refraction The magnetic induction is described:
The actual angle of refraction is
determined by a
mathematical relation called Snell’s law: (10) B = 0  H  M 
(5) n0 sin 0  n1 sin 1 where B is magnetic flux density in tesla (T) and
µ0
2 permeability of free space (µ0 = 4 ×
is the
10-7
where n0 is the index of refraction in the f i o
initial medium, n1 is the index of refraction in where f is focal length, i is
the new medium, 0 is the initial angle of image distance, and o is
incidence, and 1 is the angle of refraction.
object distance.

Lens Law
The focal length or focal distance of a lens
is the distance from the principal plane to
the focal plane.
1 1 1 18.2 AEROSPACE NONDESTRUCTIVE TESTING
(6)  
N/A ). These equations are often recombined:

(11) B = 0 1  m H
=  0 r H

where r is the relative permeability. The permeability is the

ability of a material to have a
magnetic field created within it. High relative permeability and
high magnetic susceptibility are related [r = (1+ m)] and allow
materials to be inspected with magnetic particle testing.
Electromagnetic Testing (18) E  v2

Electromagnetic Induction where E is the elastic modulus (N/m2), v is

Electromagnetic induction is velocity (m/s), and  is density (kg/m3). The
described: unit of force is the newton, where 1 N = kg·m–
1·s–2.
B A more precise expression for
(12)   E = –
longitudinal waves is as follows:
2
E = v 1  
t 1 – 2
(19) 
1 –  
where B is the magnetic field vector, E is the electric  
field vector, and t is time.
The time varying eddy current density J is
given:
where Poisson’s ratio  (Greek nu) is the ratio of
transverse contraction –trans to longitudinal
(13) J  E extension long, where  is strain:  = –trans/ long
.
Poisson’s ratio  has a value between 0 and 0.5 for
where  is the material conductivity. solid materials. Aluminum for example has a
value of around 0.33. For the transverse
Voltage Induced in Coil waves, the shear modulus G (in pascals) is
Electric potential is generated in a given:
coil: 2
 
(20) G vtrans

(14)V

– where (italic Roman vee) is transverse velocity.
tran
Nd v s
dt
where N is the number of turns in the coil, V Attenuation
is the voltage induced, and  is the magnetic The attenuation due to scatter and
flux linking the coil. absorption of the sound pressure of a plane
wave can be represented:
–d
Depth of Penetration (21) p  p 0e

The standard depth of penetration (in where d is the distance traveled, p is the end
meters) is given: pressure, p0 is the initial pressure,  is the
attenuation coefficient. The pressure is the
1
(15)   square of
 f r 0  the amplitude (A) or magnitude of the

signal seen on the ultrasound display.
where  is the depth of penetration, f is the In practice, it is useful to use a logarithmic
coil operating frequency (Hz), r is the relative measurement to express reduction
permeability, 0 is the permeability of free (attenuation) or increase (gain) of a signal. The
space (4 decibel (dB) unit named after Alexander
× 10-7 H/m), and  is the conductivity (S/m). Graham Bell is defined:
 A2   A 
(22) 10 log10  2   20 log10  
Ultrasonic Testing  A 0  A 0 

where A is the final signal amplitude and A0
Wavelength is initial signal magnitude.
The wavelength  is related to the velocity
v
(italic Roman vee) and frequency f: Transmission and Reflection
The transmission T at an interface is
described:
v
(16)  
f
(23) T  2 Z2
Acoustic Impedance Z2  Z1
Acoustic impedance Z is the product of
And its reflection R is described:
density 
and velocity v: Z2 – Z1
  v (24) R 
(17) Z Z  Z
2 1

where Z1 and Z2 are the acoustic impedances of

Modulus materials 1 and 2 at an interface.
Modulus is related to acoustic velocity in
homogeneous and isotropic materials:

FORMULAS 18.3
Snell’s Law Half Value Layer
The refraction of ultrasound when passing A convenient practical measure of radiation
from one medium to another is based on the attenuation is the half-value layer (HVL). The
relative velocity within each material as HVL of any specific material is that thickness
governed by Snell’s law: that will reduce the radiation intensity to one-
sin   v half its initial value. The half-value layer is
(25) 1  1
related to the linear attenuation coefficient:
sin2  v2 
2
 ln 0.693
where v1 and v2 are the velocities in material (31) HVL   
1 and material 2, respectively, 1 is the
incident angle in
material 1, and  2 is the refracted angle in Similarly, a tenth-value layer (TVL) is that
material 2. thickness
of material that will reduce the radiation
intensity to one-tenth its initial value. HVL
Transducer Near Field and Far Field and TVL are related as follows:
The near field of a transducer is defined:
d2 – 2 d2 (32) 3.32 HVL  1 TVL
 
(26) N
4 4 

Attenuation Coefficients
where d is the diameter of the transducer, N is
the The attenuation coefficient may be given
near field distance, and  is the wavelength.  as linear attenuation coefficient (cm-1), mass
is usually much smaller than d. The angle attenuation coefficient (cm2/gm), or cross
ofdivergence of the beam beyond the near section (cm2):
field is given:
  
(27) sin   1.2 
    
(33)
d 
where  is the half angle of the beam spread. where (/) is the mass attenuation coefficient
and
 is density (g/cm3). The attenuation as a cross
Radiologic Testing section (cs) is area per atom:

cs   A
Radiation Attenuation with Distance (34) 
 Na
Radiation attenuation with distance is
given by the inverse square formula: where A is the atomic weight (g/mole) and NA
is
Avogadro’s number (6.02 × 1023 atoms/mole).
(28) I1 d12  I2 d2
2
The mass attenuation coefficient allows for
the calculation of attenuation in compounds
by the expression:
where I1 is the intensity at distance d1 and I2 is the      W      
  1  W  … 
intensity d2.      
(35)
 2 Wn
c 1 2
n

Radiation Attenuation by Material

where (/)c is the mass attenuation
The attenuation of penetrating photon coefficient of the compound, (/)n is the
radiation mass attenuation coefficient of each
is exponential. The intensity I(E) is a function element of the compound, and
of the X-ray energy transmitted through an Wn is the weight fraction of each element.
absorber (shield) and can be expressed:
Buildup Factors
(29) I1 E   –Et The buildup factor B for broad beam
I0 E e
radiation attenuation is expressed:
where I0(E) is the initial intensity, t is the A useful manipulation of the
absorber thickness, and (E) is the linear attenuation equation will allow
attenuation coefficient.
 straightforward calculation of the absorber 
thickness: (36) I  I0 Be–t
 I 
and can be approximated:

(37) B  1 t
(30) ln  0   t
I  The buildup factor depends on the atomic
number Z of the absorber and on the energy
of the initial photon.

18.4 AEROSPACE NONDESTRUCTIVE TESTING

 Film Density Exposure with Distance Correction
The levels of darkening on the film due Distance can be corrected to optimize
to the X-ray exposure are measured in exposure:
units of H and D
(named for Hurter and Driffield): T2 2  T1 2
(44) 
 It  D2 D1
(38) D  log
I0 

where D is the density in H and D units, I0 2

D 2
is the intensity of incident light, and It is the  D2 
T2  T1 T1 2
intensity of light.
transmitted
(45)
=
 D 1 D12
Computed Tomography Units
where D1 is original distance from source to
The medical field uses the hounsfield sensor (SFD), D2 is new distance from source
notation for image data: to sensor (SFD), T1 is original exposure (mA·s),
 – w and T2 is new exposure (mA·s).
(39) H  1000 
w
Thermographic Testing
where H is the hounsfield number,  is the
measured
attenuation number and w is the
attenuation
number for water. In this scale, water takes
on the Temperature Response
value of zero, vacuum (or air) is –1000, and The temperature response T on the surface of a
bone is 1000. uniformly thick object is given:
Q
(46) T t  
Subject Contrast e t
Radiation attenuates through material:

(40)I  I0 e–t where e is the thermal effusivity of the material,

Q is the energy flux at the surface at time t = 0,
where I is the transmitted beam intensity, I0 and t is time.
is the incident radiation beam intensity, t is
the material
thickness, and  is the attenuation Thermal Effusivity
coefficient. The subject contrast C will be The thermal effusivity e is expressed:
given by a change in transmitted intensity
for a change in thickness.
Differentiating and solving the equation
(47) e  C
above will give:



I where C the specific heat capacity ((J/(kgK)),  is the
(41) C 
I
 –  t thermal conductivity (W/(mK)), and  is the density
(kg/m3).

Magnification
Thermal Diffusivity
The geometric magnification factor is Thermal diffusivity  is expressed:

given: 

SOD + ODD
(42)   (48)  
M
SOD C
f
where SOD is the source-to-object distance
and ODD is the distance from the object to
the detector.

Geometric Unsharpness
Unsharpness can be calculated:
 density (kg/m3).
where C the specific heat
capacity ((J/(kgK)),  is the
thermal conductivity Heat Propagation Time
(W/(mK)), and  is the The time t* required for heat to propagate
through a given thickness is expressed:
(43) Ug  ODD  SOD

t*  L2 a
(49)
where f is the maximum dimension of the 
focal spot, ODD is the distance from the
object to the
detector, and SOD is the distance from the
source to
the object. where L is the thickness and  is the
thermal diffusivity.

FORMULAS 18.5
 Capacitance and Conductance X-Ray Diffraction
For a homogeneous material of
permittivity  and ohmic conductivity  Bragg Equation
(inverse of resistivity ), the capacitance C
The bragg equation describes the angle of
and conductance G (inverse of resistance R)
the diffracted X-rays for particular plane
can be expressed:
spacing and a particular wavelength:
A
(50) C  (52) n    2  d  sin
d

1 where d is the lattice spacing, n is the order
(51) G  A
 d number of the diffraction pattern,  is the
R
angle of diffraction, and  is the X-ray
wavelength.

where A is the electrode area and d is the gap

thickness between the electrodes. Material
dielectrics have permittivities that scal–e1 with the
permittivity of free space, 0 = 8.854 pF·m .

18.6 AEROSPACE NONDESTRUCTIVE TESTING

 19CHAPTER

Glossary

Contents
Part 1, Introduction, 19.2
Part 2, Terms, 19.3
References, 19.21

Contributor
Matthew J. Golis
 Introduction
PART1

This glossary uses terms found in the third working to written contracts or procedures should
edition of the Nondestructive Testing Handbook consult definitions referenced in standards when
(ASNT 1998-2012). The definitions have been appropriate. For example, persons who work in
selected and in many cases modified to satisfy the accordance with standards published by ASTM
practical needs of the aerospace community. International are encouraged to refer to definitions
The definitions in this ASNT Industry Handbook in those standards (ASTM 2014).
volume should not be referenced for tests performed Some definitions are related to physical processes
according to standards or specifications or in or quantities and are clarified in the literature
fulfillment of contracts. Standards writing bodies (CRC 2014; Mordfin 2002).
take great pains to ensure that their standards are This glossary is provided for instructional
definitive in wording and technical accuracy. People purposes. No other use is intended.

19.2 AEROSPACE NONDESTRUCTIVE TESTING

 Terms PART2

A acoustic emission (AE): Transient elastic waves

resulting from local internal microdisplacements
A-scan: One-dimensional display of ultrasonic in a material. By extension, the term also
signal amplitude as function of time or depth in describes the testing and measurement using this
test object. phenomenon.
absolute coil: Coil that responds to the acoustic emission activity: Number of bursts (or
electromagnetic properties of that region of the events, if the appropriate conditions are fulfilled)
test part within the magnetic field of the coil, detected during a test or part of a test.
without comparison to the response of a second acoustic emission count: Number of times the
coil at a different location on the same or signal amplitude exceeds the preset reference
similar material. threshold. Sometimes called ringdown counts.
absorbed dose: Amount of energy imparted to acoustic emission rate: Number of times the
matter by an ionizing event per unit mass of amplitude has exceeded the threshold in a
irradiated material at the place of interest. specified unit of time.
Absorbed dose is expressed in gray (Gy) or rad. acoustic impedance: Material property defined as
absorption: Phenomenon in which radiant energy the product of sound velocity and density of the
enters a material and stays there rather than material. The relative transmission and reflection
being transmitted, reflected, or refracted. The at an interface are governed in part by the
energy converts to another form, as when acoustic impedances of the materials on each
X-ray photons become electrons or sound side of the interface.
waves create heat. activation: Process by which neutrons bombard
absorption coefficient: Fractional decrease in stable atoms and make them radioactive.
intensity of transmitted energy due to activity: Degree of radioactivity of a particular
absorption processes per unit of thickness isotope. Activity is expressed as the number of
of absorbing material. atoms disintegrating per unit of time. Measured
absorptivity (absorptance): In thermography, in becquerels.
proportion (as a fraction of 1) of the radiant acuity: See neural acuity; vision acuity.
energy impinging on a material‘s surface that is adaptive thresholding: Threshold value varying
absorbed into the material. For a blackbody, this with inconstant background gray level.
is unity (1.0). alpha particle: Positively charged helium ion
accelerator: (1) Device that accelerates charged emitted by certain radioactive materials. It is
particles to high energies. Examples are X-ray made up of two neutrons and two protons;
tubes, linear accelerators, and betatrons. hence, it is identical with the nucleus of a
(2) Linear accelerator. helium atom.
acceptance criterion: Benchmark against which test alternating current magnetization: Magnetization
results are to be compared for purposes of by a magnetic field generated when alternating
establishing the functional acceptability of a current is flowing.
part or system being examined. ambient light: Light in the environment as opposed
acceptance standard: Specimen, similar to the to illumination provided by a visual test system.
product to be tested, containing natural or ambient operating range: Range of ambient
artificial discontinuities that are well defined temperatures over which an instrument is
and similar in size or extent to the maximum designed to operate within published
acceptable in the product. Examples include performance specifications.
plates with notches, holes, or grown cracks of ampere (A): SI unit of electric current.
known dimensions. ampere per meter (A/m): SI compound unit for
accommodation: Of the eye, adjustment of the lens‘ magnetic field intensity. The measurement
focusing power by changing the thickness and 1 A/m, for example, describes a current of 1 A
curvature of the lens by the action of tiny flowing through a coil of 1 m diameter.
muscles attached to the lens. Accommodation amplitude response: Property of a test system
facilitates the viewing of objects near and far. whereby the amplitude of the detected signal is
accuracy: Degree of conformity of a particular measured without regard to phase. Many simple
measurement to a standard or true value. meter-style instruments use this type of readout.

GLOSSARY 19.3
 amplitude, echo: Vertical height of a received signal B
on an A-scan, measured from base to peak for a
video presentation or from peak to peak for a B-scan: Data presentation technique typically
radio frequency presentation. applied to pulse echo techniques of ultrasonic
angle beam: Ultrasound beam traveling at an acute testing. It produces a two-dimensional view of a
angle into a medium. The angle of incidence or cross sectional plane through the test object. The
angle of refraction is measured from the normal horizontal sweep is proportional to the distance
to the entry surface. along the test object, and the vertical sweep is
angle beam testing: Technique of ultrasonic testing proportional to depth, showing the front and
in which transmission of ultrasound is at an back surfaces and discontinuities between.
acute angle to the entry surface. background noise: Extraneous signals caused by
angle beam transducer: Transducer that transmits random signal sources within or exterior to the
or receives ultrasonic energy at an acute angle ultrasonic testing system, including the test
to the surface. This may be done to achieve material. It has electrical, mechanical, or
special effects such as setting up transverse or chemical origins. Sometimes called grass or
surface waves by mode conversion at an hash.
interface. back reflection: Signal received from the far
angle of incidence: Included angle between the boundary or back surface of a test object.
beam axis of the incident wave and the normal backscatter: (1) Interaction of radiation with matter
to the surface at the point of incidence. such that the direction of travel after scattering
angle of reflection: Included angle between the is over 90 degrees and often close to
beam axis of the reflected wave and the normal 180 degrees to the original direction of travel.
to the reflecting surface at the point of (2) In transmission radiologic testing, interaction
reflection. of radiation with matter behind the image plane
angle of refraction: Angle between the beam axis such that scattered radiation returns to the
of a refracted wave and the normal to the image plane, often adding fog and noise that
refracting interface. interfere with production of an image of the
anisotropy: Condition in which properties of a specimen. (3) Of scatter imaging, interaction of
medium (velocity, for example) depend on incident radiation with a specimen that scatters
direction in the medium. the radiation through large angles frequently
anode: (1) In radiologic testing, the positive greater than 90 degrees to the original direction
electrode of a cathode ray tube that generates of travel. Such radiation is used to form an
ionizing radiation. (2) Positively charged image or to measure a parameter of the
terminal, which may corrode electrochemically specimen, usually through digital techniques.
during production of electric current. Compare backscatter imaging: In radiologic testing, a family
cathode. of radioscopic techniques that use backscatter.
anomaly: Variation from normal material or product backscattering, infrared: Reflection of thermal
quality. energy, whether generated by the ground and
antinode: Point in a standing wave where certain reflecting off clouds or unwanted front surface
characteristics of the wave field have maximum reflections from transparent objects.
amplitude. band pass filter: Frequency filter that has a single
array: Group of sensors arranged to locate signal transmission band between two cutoff
sources, to expedite scanning, or to acquire frequencies, neither of the cutoff frequencies
signals whose excitation is phased in its timing. being zero or infinity.
array transducer: Transducer made up of several bandwidth: Difference between the cutoff
piezoelectric elements individually connected so frequencies of a bandpass filter.
that the signals they transmit or receive may be baseline: (1) Horizontal trace across the A-scan
treated separately or combined as desired. See display. It represents time and is generally
also phased array. related to material distance or thickness.
artifact: In nondestructive testing, an indication that (2) Standard measurement, average
may be interpreted erroneously as a measurement, prior measurement, or other
discontinuity. criterion for comparison and evaluation in
artificial source: Point where elastic waves are quality control.
created to simulate an acoustic emission event. beam: Defined stream of radiation particles all
The term also denotes a device used to create traveling in parallel paths.
the waves. beam quality: Penetrating energy of a radiation
attenuation: Decrease in transmitted energy beam.
intensity over distance. The loss may be due to beam spread: Divergence from a beam of radiation
absorption, scattering, reflection, leakage, beam in which all particles are traveling in parallel
divergence, or other material effects. paths. Widening of the sound beam as it travels
attenuation coefficient: Fractional decrease in through a medium. Specifically, the solid angle
transmitted intensity per unit of material that contains the main lobe of the beam in the
thickness. The attenuation coefficient usually far field.
includes scattering and absorption effects of the becquerel (Bq): SI unit for measurement of
host material. radioactivity, equivalent to one disintegration
per second. Replaces curie.
beta particle: Electron or positron emitted from a
nucleus during decay.

19.4 AEROSPACE NONDESTRUCTIVE TESTING

beta ray: Radiation stream consisting of beta camera: Device that contains a sealed radiation
particles. source, where the source or shielding can be
bonds, kissing: Bonded conditions wherein typical moved so that the source becomes unshielded
test methods indicate a material continuity (to make a radiologic exposure) or shielded (for
through the thickness of the bond whereas in safe storage).
fact there is no significant physical strength to candela: Base unit of measure in SI for measuring
the bond. luminous intensity. The luminous intensity in a
borescope: Industrial endoscope; a periscope or given direction of a source that emits
telescope using mirrors, prisms, lenses, optic monochromatic radiation of frequency 5.40 ×
fibers, or video signals to transmit images from 1014 Hz and that has a radiant intensity in that
inaccessible interiors for visual testing. direction of 1.4641 mW/sr. Symbolized cd.
Borescopes take their name from bores because Formerly known as candle.
borescopes were originally used in machined casette cast structure: Internal physical structure of
apertures and holes such as gun bores. Some a casting evidenced by shape, orientation of
borescopes are flexible; some are rigid. grains, and segregation of impurities.
boundary echo: Reflection of an ultrasonic wave casette, film: Lightproof container for holding
from an interface. radiologic film in position during the radiologic
Bq: Becquerel. test‘s exposure. The cassette may be rigid or
brehmsstrahlung: Electromagnetic radiation flexible and may contain intensifying screens,
produced when electrons‘ path and kinetic filter screens, both, or neither.
energy brings them close to the positive fields of cassette: See casette.
atomic nuclei — as when, for example, electrons cathode: (1) Negatively charged terminal in an
strike a target provided for this purpose. The arrangement that produces current by chemical
electrons slow down, giving up kinetic energy as reactions. Compare anode. (2) In radiologic
X-radiation. testing, the negative electrode of an X-ray tube,
brittleness: Characteristic of a material that leads to the electrode from which electrons are emitted.
crack propagation without appreciable plastic cathode ray: Stream of electrons emitted by a
deformation. heated filament and projected in a more or less
broad band: Having a relatively wide frequency confined beam under the influence of a
bandwidth. Used to describe pulses that display magnetic or electric field.
a wide frequency spectrum and receivers capable celsius (centigrade): A relative scale of temperature
of amplifying them. related to the kelvin scale (0 °C = 273.12 K; 1 °C
burr: Raised or turned over edge occurring on a = 1 K). Temperature scale based on 273 K (0 °C
machined part and resulting from cutting, = +32 °F) as the freezing point of water and
punching, or grinding. 373 K (100 °C = 212 °F) as the boiling point of
burst: Signal whose oscillations have a rapid water at standard atmospheric pressure.
increase in amplitude from an initial reference centigrade: Celsius.
level (generally that of the background noise), central conductor: Metal bar passed through a hole
followed by a decrease (generally more gradual in a test object, used for creating a circular
than the initial rise) to a value close to the magnetic field within the object.
initial level. CGS system: Obsolete system of measurement units
burst duration: Interval between the first and last based on the centimeter, gram, and second.
time a specified threshold is exceeded by a Superseded by SI.
particular burst. characteristic curve: Curve that expresses film
burst emission: Qualitative term denoting acoustic density as function of log relative exposure.
emission when bursts are observed. Opposite of These curves are useful in determining exposure
continuous emission. correction factors and in defining the gamma
burst rise time: Time interval between the first characteristics of the film.
threshold crossing and the maximum amplitude charge coupled device (CCD): Solid state optical
of the burst. sensor used in imaging systems. Incoming
radiation induces electrical charges stored in
semiconductor structures for reading.
C circular magnetization: Magnetization in an object
C-scan: Presentation technique applied to acoustic resulting from current passed longitudinally
data and displaying an image of through the object itself or through an inserted
two-dimensional test object with scaled grays or central conductor.
colors representing the ultrasonic signals. The circumferential coil: See encircling coil.
amplitude represented in each pixel may be a closure: Process by which a person cognitively
pulse echo, through-transmission, or pitch catch completes patterns or shapes that are
value calculated from each A-scan datum. incompletely perceived.
calcium tungstate: Fluorescent chemical compound coherent radiation: Radiation at the same energy
that emits visible blue violet light when and phase.
irradiated by X-rays or gamma rays. coherent scatter: Form of scatter where no energy is
calibration, instrument: Adjustment of instrument lost.
readings to a known reference standard. Used to coil: One or more loops of a conducting material; a
ensure instruments are set up in a consistent single coil may be an exciter and induce
manner from one inspection to the next. currents in the material, or it may be a detector
or both simultaneously.

GLOSSARY 19.5
 coil spacing: In eddy current testing, the axial conductivity, thermal (k): Material property
distance between two encircling or inside coils defining the relative capability to carry heat by
of a differential or remote field test system. conduction in a static temperature gradient.
cold cathode ionization gage: Pressure measuring Conductivity varies slightly with temperature in
gage for low pressures, in the range of 13.3 mPa solids and liquids and with temperature and
to below 0.13 nPa (10–4 torr to below 10–12 torr), pressure in gases. It is high for metals (copper
that works by measuring a discharge current has a k of 380 W·m–1·K–1) and low for gases and
associated with the ionization of gas by porous materials (concrete has a k of
electrons confined in a magnetic field. Also 1.0 W·m–1·K–1). Compare conduction, thermal.
called a philips discharge gage or penninggage. contact head: Electrode assembly used to clamp and
collimator: Device for restricting the size, shape, support an object to facilitate passage of electric
and direction of the irradiating beam, thereby current through the object for circular
limiting beam spread and its consequences. magnetization.
color: Vision sensation by means of which humans contact technique: Testing technique in which the
distinguish light of differing hue (predominant transducer face makes direct contact with the
wavelengths), saturation (degree to which those test object through a thin film of couplant.
radiations predominate over others), and contact transducer: Transducer used in the contact
lightness. technique.
comparator coils: In electromagnetic testing, two or continuous emission: Qualitative term applied to
more coils electrically connected in series ongoing acoustic emission in which the bursts
opposition and arranged so that there is no or pulses are not individually discernible — for
mutual induction (coupling) between them. Any example, emission from a leak.
electromagnetic condition that is not common to continuous wave: Wave of constant amplitude and
the test specimen and the standard will produce frequency.
an imbalance in the system and thereby yield an contrast: (1) Difference in color or brightness
indication. Compare differential coils. between a test indication and its background or
compensator blocks: Material added to regions of a surroundings. (2) In film radiologic testing, the
test object to flatten its radiologic test image. difference in film blackening or density
complex plane diagram: Graphical presentation of observed from one area to another and resulting
complex quantities where the real and from various radiation intensities transmitted by
imaginary components are represented along the the object.
horizontal and vertical axes, respectively. Types contrast, subject: Ratio of radiation intensities
of complex plane diagrams include impedance transmitted by selected portions of object being
plane and voltage plane diagrams. radiographed.
composite materials: Engineered materials formed control cable: Cable connected to isotopic radiologic
through layering of physical elements aimed at test source and used to move the source in and
meeting specific mechanical and chemical out of the exposure device.
performance criteria. Examples include corner effect: Strong reflection obtained when an
honeycomb panels, boron epoxy, and graphite ultrasonic beam is directed toward the
epoxy sheeting. intersection of two or three intersecting surfaces.
compressional wave: Longitudinal wave. corrosion: Loss or degradation of metal caused by
compton scatter: Reduction of energy of incident chemical reaction and appearing as pitting and
photon by its interaction with an electron. Part loss of material. Galvanic corrosion is caused by
of the photon energy is transferred to the electrical contact between dissimilar metals and
electron, giving it kinetic energy, and the by entrapped fluids that permit the electronic
remaining photon is redirected with reduced interchange of ions between adjoining metal
energy. surfaces. Mechanisms of corrosion include
computed tomography: Technique by which crevice corrosion and stress corrosion in the
radiation passing through an object is displayed form of cracking.
as one slice or layer of that object at a time. couplant: Substance used between the transducer
Image data are processed in three dimensions. and the contacting surface to permit or improve
conductance (G): Transmission of electric current transmission of ultrasonic energy into or from
through a material. Measured in siemens (S). the test object.
Inversely related to resistance R (ohms). coupled: (1) Of two electric circuits, having an
conduction, thermal: Heat transfer occurring when impedance in common so that a current in one
more energetic particles collide with — and thus causes a voltage in the other. (2) Of two coils,
impart some of their heat energy to — adjacent sharing parts of their magnetic flux paths.
less energetic (slower moving) particles. This coupling: Percentage of magnetic flux from a
action is passed on from one atom (or free primary circuit that links a secondary circuit;
electron) to the next in the direction of cooler effectiveness of a coil in inducing eddy currents
regions. Thus, heat always flows from a warmer in the test object.
to a cooler region. Compare conductivity, coupling coefficient: Fraction of magnetic flux from
thermal; thermal diffusion. one circuit (test coil) that threads a second
conductivity, electrical (): Ability of material to circuit (test object); the ratio of impedance of
transmit electric current. Measured in siemens the coupling to the square root of the product of
per meter. Inversely related to resistivity . the total impedances of similar elements in the
two meshes.

19.6 AEROSPACE NONDESTRUCTIVE TESTING

crack: (1) Break, fissure, or rupture, sometimes V dead zone: Interval following the initial pulse at the
shaped and relatively narrow and deep. surface of a test object to the nearest inspectable
Discontinuity that has a relatively large cross depth. Any interval following a reflected signal
section in one direction and a small or where additional signals cannot be detected.
negligible cross section when viewed in a decay curve: Graph showing radioactive intensity as
direction perpendicular to the first. a function of time for an isotope. Decay curves
(2) Propagating discontinuity caused by applied are used in determining exposure times in
stresses such as mechanical flexing (fatigue radiologic testing.
crack). Difficult to detect unaided because of decibel (dB): Logarithmic unit for expressing
fineness of line and pattern (may have a radial relative signal power, such as the loudness of a
or latticed appearance). sound, in proportion to the intensity of a
crack, fatigue: Progressive crack that usually is reference signal. One tenth of a bel. Decibel in
initiated at a free surface such as a fastener hole signal amplitude is twice that in signal power.
and is caused by the repeated mechanical One decibel equals ten times the base ten
loading and unloading of the object. logarithm of the ratio of two powers.
critical angle: Incident angle of the ultrasound defect: Discontinuity whose size, shape, orientation,
beam where the refracted beam is parallel to the or location make it detrimental to the useful
surface and above which a specific mode of service of its host object or which exceeds the
refracted energy no longer exists. accept/reject criteria of an applicable
cross talk: Unwanted signal leakage (acoustical or specification. Some discontinuities do not
electrical) across an intended barrier, such as exceed an accept/reject criterion and are
leakage between the transmitting and receiving therefore not defects. Compare crack;
elements of a dual transducer. Also called cross discontinuity; indication.
noise and cross coupling. definition: Description of linear demarcation
crushed core: Mechanically deformed inner core of sensitivity or the detail sharpness of object
honeycomb composite components. outline in a radiologic test image. It is a
crystal mosaic: Multiple crystals mounted in the function of screen type, exposure geometry,
same surface on one holder and connected so as radiation energy, and characteristics sensor
to cause all to vibrate as one unit. (such as film).
cumulative bursts: Number of bursts detected from deformation: Change of shape under load.
the beginning of a test. delamination: Separation of plies in composite
cumulative characteristic distribution: In acoustic layered materials.
emission signal processing, a display of the delay line: Material (liquid or solid) placed in front
number of times a characteristic meets a of a transducer to cause a time delay between
preselected criterion. the initial pulse and the front surface reflection.
curie (Ci): Unit for measurement of the quantity of depth of field: The range of distance over which an
radioactivity, corresponding originally to imaging system gives satisfactory definition
radiation from atomic disintegrations from 1 g when its lens is in the best focus for a specific
of radium; replaced by becquerel (Bq) in SI, distance. Compare depth of focus; field of view.
where 1 Ci = 3.7 × 1010 Bq. depth of focus: Focal zone. Distance a sensor may
current flow technique: Magnetizing by passing be moved from a lens system and still produce a
current through an object using prods or contact sharp image.
heads. The current may be alternating current or depth of penetration: See skin effect and standard
rectified alternating current. depth of penetration.
current induction technique: Magnetization in detail: In radiologic testing, the degree of sharpness
which a circulating current is induced in a ring of outline of an image or the clear definition of
component by a fluctuating magnetic field. an object or discontinuity in the object. See also
cutoff frequency: Upper or lower spectral response definition.
of a filter or amplifier, at which the response is detector coil: See sensing coil.
a specified amount less (usually 3 or 6 dB) than detector, X-ray: Sensor, X-ray.
the maximum response. developer: In film radiologic testing, a chemical
solution that reduces exposed silver halide
crystals to metallic silver.
D dewaxing: Removing the expendable wax pattern
damping capacity: Measure of the ability of a from an investment mold by heat or solvent.
material to dissipate mechanical energy. differential coils: Two or more physically adjacent
damping material: Highly absorbent material used and mutually coupled coils connected in series
to cause rapid decay of vibration. opposition such that an imbalance between
damping, transducer: Material bonded to the back them, causing a signal, will be produced only
of the piezoelectric element of a transducer to when the electromagnetic conditions are
limit the duration of vibrations. different in the regions beneath the coils. In
damping, ultrasonic: Decrease or decay of contrast, comparator coils are not adjacent or
ultrasonic wave amplitude controlled by the mutually coupled. Compare comparator coils.
instrument or transducer.
damping: (1) Limiting the duration or decreasing
the amplitude of vibrations, as when damping a
transducer element. (2) Deliberate introduction
of energy absorbers to reduce vibrations.

GLOSSARY 19.7
 diffraction: (1) In acoustics, the reflection of a wave divergence: Term sometimes used to describe the
front when passing the edge of an ultrasonically spreading of ultrasonic waves beyond the near
opaque object. (2) In radiation technologies, a field. It is a function of transducer diameter and
special case of scatter, where coherently wavelength in the medium. See beam spread.
scattered photons undergo interference or dose: See absorbed dose.
reinforcement, resulting in patterns indicative of dose rate: Radiation dose delivered per a specified
the scattering medium. See also X-ray unit of time and measured, for instance, in
diffraction. sieverts per minute. See also absorbed dose.
diffuse reflection: Scattered, incoherent reflections dosimeter: Device that measures radiation dose,
from rough surfaces. such as a film badge or ionization chamber.
direct current: Electric current flowing continually double-crystal technique: See pitch catch
in one direction without variation in amplitude technique.
through a conductor. See also full-wave rectified dual transducer: See send/receive transducer.
direct current and half-wave direct current. dynamic range: Ratio of maximum to minimum
direct current field: Active magnetic field produced reflective areas that can be distinguished on the
by direct current flowing in a conductor or coil. display at a constant gain setting.
direct viewing: (1) Viewing of a test object in the
viewer‘s immediate presence. The term direct
viewing is used in the fields of robotics and E
surveillance to distinguish conventional from echo: Reflected acoustic energy or signal indicating
remote viewing. (2) Viewing of a test object such energy.
during which the light image is not mediated eddy current: Electrical current induced in a
through a system of two or more lenses (as in a conductor by a time varying magnetic field.
borescope) or transduced through an electronic eddy current testing: Nondestructive test technique
signal (as with a charge coupled camera). The in which eddy current flow is induced in the test
term direct viewing is used in some object. Changes in the flow caused by variations
specifications to mean viewing possibly with a in the specimen are reflected into a nearby coil,
mirror or magnifier but not with a borescope. coils, hall effect device, or other magnetic flux
Compare indirect viewing; remote viewing. sensor for subsequent analysis by suitable
disbonds: Separation of bonded surfaces in instrumentation and techniques.
structures such as skins from honeycomb core edge effect: In electromagnetic testing, the
structures, patches from aluminum structures disturbance of the magnetic field and eddy
and similar adjoining laminated sheet-like currents because of the proximity of an abrupt
materials. change in geometry, such as an edge of the test
discontinuity: Interruption in the physical structure object. Sometimes called end effect. The effect
or configuration of a test object. After generally results in the masking of
nondestructive testing, a discontinuity indication discontinuities within the affected region.
may be interpreted to be a defect. Compare effective depth of penetration: In electromagnetic
defect; indication. testing, the minimum depth beyond which a test
discontinuity, artificial: Reference discontinuity system can no longer practically detect a further
such as hole, indentation, crack, groove, or increase in specimen thickness. Usually equal to
notch introduced into a reference standard to about three standard depths of penetration.
provide accurately reproducible indications for effective focal spot: Size and geometry of focal spot
determining test sensitivity levels. after target interaction. Viewed from along the
discontinuity characterization: The process of primary beam central axis at the target, the
quantifying the size, shape, orientation, location, effective focal spot would appear nearly square
growth, or other properties of a discontinuity and smaller than the actual focal spot area
based on test data. covered by the electron stream.
discontinuity resolution: Property of a test system effective penetration: In a material, the maximum
that enables the separation of nearby indications depth at which a test signal can reveal
in a test specimen. discontinuities.
dispersion: In acoustics, variation of wave phase electrical noise: Extraneous signals caused by
with frequency. external sources or electrical interferences
dispersive medium: Medium in which the within an ultrasonic instrument. A component
propagation velocity depends on the wave of background noise.
frequency. electromagnet: Ferromagnetic core surrounded by a
distal: In a manipulative or interrogating system, of coil of wire that temporarily becomes a magnet
or pertaining to the end opposite from the when an electric current flows through the wire.
eyepiece and farthest from the person using the electromagnetic acoustic transducer (EMAT):
system. Objective; tip. Transmitting transducer based on the force
distance amplitude correction: Compensation of exerted on a current flowing in a magnetic field.
gain as a function of time for difference in A receiving transducer that detects the current
amplitude of reflections from equal reflectors at produced by moving a conductor in a magnetic
different sound travel distances. Refers also to field.
compensation by electronic means such as swept
gain, time corrected gain, time variable gain and
sensitivity time control.

19.8 AEROSPACE NONDESTRUCTIVE TESTING

electromagnetic testing (ET): Nondestructive test felicity ratio: Measurement of the felicity effect.
method for materials, including magnetic Defined as the ratio between (1) the applied load
materials, that uses electromagnetic currents, (or pressure) at which acoustic emission
either alternating or direct current, to yield reappears during the next application of loading
information about the test object. In and (2) the previous maximum applied load.
nondestructive testing, the term electromagnetic ferrite: Any of several magnetic substances that
testing is not applied to methods such as consist of an iron oxide combined with one or
radiologic and thermographic testing, in which more metals (such as manganese, nickel, or zinc)
the interrogating energy is a form of having high magnetic permeability and high
electromagnetic radiation. electrical resistivity.
electronvolt (eV): Kinetic energy acquired by an ferromagnetic material: Material such as iron,
electron in passing through a potential nickel, or cobalt whose relative permeability by
difference of 1 V in vacuum; 1 eV = ~1.60 J. magnetizing force is much greater than unity. A
The electronvolt is commonly used to express ferromagnetic material is strongly affected by
the energy of X-rays. magnetism and exhibits hysteresis.
EMAT: See electromagnetic acoustic transducer. fiber optics: Technology of efficient transmission of
encircling coil: In electromagnetic testing, a coil or light through transparent fibers such as glass,
coil assembly that surrounds the test object. quartz, and plastic by means of total internal
Such a coil is also called an annular coil, reflection.
circumferential coil, or feed-through coil. It is field of view (FOV): Range or area where things can
commonly used for testing wire, tubing, and bar be seen through an imaging system, lens, or
materials as well as small components in aperture. Angular subtense (expressed in angular
production. degrees or radians per side if rectangular or in
end effect: Alternative term used for edge effect angular degrees or radians if circular) over
when testing bar and tube shaped materials. which an instrument will integrate all incoming
eV: Electronvolt. radiant energy. In a radiation thermometer, the
evaluation: Review, following interpretation of field of view is the target spot size; in a scanner
indications, to determine whether they meet or imager, the field of view is the scan angle,
specified acceptance criteria. picture size, or total field of view. Compare
excitation coil: Coil that carries the excitation depth of field.
current. Also called primary coil or winding. See field of vision: Range or area where things can be
sensing coil. perceived organoleptically at a point in time,
exfoliation: Corrosion that progresses approximately assuming the eye to be immobile.
parallel to the outer surface of the metal, film badge: Package of photographic film worn as a
causing layers of the metal to be elevated by the badge by radiologic test personnel (and by
formation of corrosion product. workers in the nuclear industry) to measure
expanded sweep: Short duration horizontal sweep exposure to ionizing radiation. Absorbed dose
positioned to allow close examination of a can be calculated by degree of film darkening
signal. caused by irradiation.
exposure factor: In X-ray radiologic testing, the film holder: See casette, film.
quantity that combines source intensity film speed: Relative exposure required to attain a
(milliampere), time (usually minute), and specified film density for a particular film.
distance. It is the product of milliamperage and filter: (1) Data processing component or function
time divided by distance squared and determines that excludes a selected kind of signal or part of
the degree of film density. a signal. (2) Network or device that passes
electromagnetic wave energy over a described
range of frequencies and attenuates energy at all
F other frequencies. In radiologic testing, the
far field: Zone beyond the near field in front of a thickness of absorbing material placed in a
plane transducer in which signal amplitude primary radiation beam to selectively remove
decreases monotonically in proportion to longer wavelength radiation, thereby adjusting
distance from the transducer. Also called the the quality of the image.
fraunhofer zone. fire damage: Reduction in strength caused by
fatigue fracture: Progressive fracture of a material subsequent heating to alloys of aluminum and
that begins at a discontinuity and increases detectable using eddy current detection of
under repeated cycles of stress. The phenomenon changes in electrical conductivity.
leading to fracture under repeated or fluctuating fixing: Procedure used in film processing that
stresses having a maximum value less than the removes undeveloped silver salts in the emulsion
tensile strength of the material. from the surface of the film, leaving only the
feature extraction: From an enhanced image, developed black silver of the image on the film.
derivation of some feature values, usually flat bottom hole: Type of reflector commonly used
parameters for distinguishing objects in the in reference standards. The end (bottom) surface
image. of the hole is the reflector.
felicity effect: Appearance of significant acoustic flaw: Unintentional anomaly. Compare defect;
emission at a load (or pressure) level below the discontinuity.
previous maximum applied.

GLOSSARY 19.9
focal point: Point at which the instrument optics ghost: False indication arising from certain
image the radiation detector at the target plane. combinations of pulse repetition frequency and
In a radiation thermometer, this is where the time base frequency. See also wrap around.
spot size is the smallest. In a scanner or imager, gradient: Slope of characteristic curve for specified
this is where the instantaneous field of view is film density.
smallest. graininess: Film characteristic that results from
focus: Position of a viewed object and a lens system improper film processing and that consists of
relative to one another to offer a distinct image the grouping or clumping together of many
of the object as seen through the lens system. small silver grains into masses visible to the
See accommodation and depth of field. naked eye or with slight magnification.
focus, principal plane of: Single plane actually in grains: (1) Solid particle or crystal of metal. As
focus in a photographic scene. molten metal solidifies grains grow and lattices
focused beam: Sound beam that converges to a intersect, forming irregular grain boundaries.
cross section smaller than that of the element. (2) Individual crystals that make up the
focused transducer: Transducer thatproduces a crystalline structure of metal.
focused sound beam. grass: See background noise.
fog: In film radiologic testing, increase of film gray (Gy): SI unit for measurement of the dose of
density caused by sources other than from the radiation absorbed per unit mass at a specified
intended primary beam exposure. Heat, location. Replaces the rad where rad denotes
humidity, pressure, and scatter radiation can all radiation absorbed dose, not radian. 1 Gy =
cause fogging of the film. 1 J/kg = 100 rad.
foil: Metal in sheet form less than 0.15 mm gray level: Integer number representing the
(0.006 in.) thick. brightness or darkness of a pixel or, as a
foreign materials: Inclusions that may be sand, composite value, of an image comprised of
slag, oxide, dross metal, or any dissimilar pixels.
material in the material being examined. In guide tube: Cable connected to isotopic radiation
radiologic test film, foreign materials may source and used to move the source in and out
appear as isolated, irregular, or elongated of the exposure device.
variations of film density not corresponding to Gy: Gray.
variations in thickness of material or to cavities.
FOV: Field of view.
frequency: Number of complete wave cycles passing H
a given point per second or the number of halation: Spreading of light around a bright image
vibrations per second. on a fluorescent screen or developed film.
frequency, fundamental: In resonancetesting, the half-wave direct current: Single-phase alternating
frequency at which the wavelength is twice the current half-wave rectified to produce a
thickness of the test material. pulsating unidirectional current. Also called
frequency, pulse repetition: Number of pulses per half-wave current.
second. hall detector: Semiconductor element that produces
frequency, response: Amplification (gain) of a an output electromotive force proportional to
receiver over a range of frequencies. the product of the magnetic field intensity and a
frequency, test: In ultrasonic testing, the nominal biasing current.
wave frequency in a given test. hardness: Resistance of metal to plastic
front surface: First surface of the test object deformation, usually by indentation. However,
encountered by an ultrasonic beam. the term may also refer to stiffness or temper or
full-wave rectified direct current: Single-phase or to resistance to scratching, abrasion, or cutting.
three-phase alternating current rectified to harmonic: Vibration frequency that is an integral
produce a unidirectional current. The rectified multiple of the fundamental frequency.
current contains ripple. heat affected zone: Base metal that was not melted
during brazing, cutting, or welding but whose
microstructure and physical properties were
G altered by the heat.
gamma rays: High energy, short wavelength hertz (Hz): Measurement unit of frequency,
electromagnetic radiation emitted by the nucleus equivalent to one cycle per second.
of a radioactive isotope. Energies of gamma rays horizontal linearity: Measure of proportionality
are usually between 0.01 and 10 MeV. X-rays between positions of indications on the
also occur in this energy range but are of horizontal trace and the physical positions of
nonnuclear origin. their corresponding reflectors.
gate: (1) Electronic device for selecting signals in a hysteresis: Apparent lagging of the magnetic effect
segment of the trace on an A-scan display. when the magnetizing force acting on a
(2) The interval along the baseline that is ferromagnetic body is changed; phenomenon
monitored. exhibited by a magnetic system wherein its state
gauss (G): Traditional unit of magnetic flux density, is influenced by its previous history.
replaced in SI by tesla (T). 1 G = 0.1 mT. hysteresis loop: Curve showing flux density B
gauss meter: Gage that measures magnetic flux plotted as a function of magnetizing force H as
density in gauss (or tesla). magnetizing force is increased to the saturation
geometric unsharpness: See unsharpness, point in both negative and positive directions
geometric. sequentially. The curve forms a characteristically
shaped loop.
I indication: Nondestructive test response that
requires interpretation to determine its
image: Visual representation of a test object or relevance. These include such things as meter
scene. deflections, shadows on radiographs, blips on
image enhancement: Any of a variety of image screens, or localized discolorations on surfaces.
processing steps, used singly or in combination See also defect; discontinuity; indication, false;
to improve the detectability of objects in an indication, nonrelevant.
image. indication, discontinuity: Visible evidence of a
image orthicon: Television tube that uses the material discontinuity. Subsequent interpretation
photoemission method. Compare vidicon tube. is required to determine the significance of an
image processing: Actions applied singly or in indication.
combination to an image, in particular the indication, false: Test indication that could be
measurement and alteration of image features interpreted as originating from a discontinuity
by computer. where no discontinuity exists. Compare defect;
image quality indicator: Strip of material of the ghost; indication, nonrelevant. False indications
same composition as that of the test material, are an economic liability for inspection because
representing a percentage of object thickness they must be investigated.
and provided with a combination of steps, holes, indication, nonrelevant: Indication due to
slots, or series of wires. When placed in the path misapplied or improper testing. May also be an
of the rays, its image provides a check on the indication caused by an actual discontinuity that
radiologic test technique. Also called does not affect the usability of the test object (a
penetrameter. change of section, for instance).
image segmentation: Process in which the image is indication, relevant: Indication from a discontinuity
partitioned into regions, each homogeneous. (as opposed to a nonrelevant indication)
immersion technique: Testtechniqueinwhichthe requiring evaluation by a qualified inspector,
test object and the transducer are submerged in typically with reference to an acceptance
a liquid (usually water) that acts as the coupling standard, by virtue of the discontinuity‘s size,
medium. The transducer is not usually in shape, orientation, or location.
contact with the test object. indirect viewing: Visual test during which the light
impact damage: A condition found in composite image is mediated through a system of two or
layered structures wherein internal disbonding more lenses (as in a borescope) or transduced
occurs with little external sign of its existence. through an electronic signal (as with a charge
impedance (Z ): Opposition that a circuit presents to coupled camera). Compare direct viewing.
the flow of an alternating current. Often inductance (L): The ratio of the magnetic flux to the
expressed for a coil as being comprised of a current causing it; more accurately, the
resistance R plus an inductive reactance XL. self-inductance of the circuit.
impedance analysis: In electromagnetic testing, an inductive reactance (X ): Reaction of the inductor
L
analytical technique that consists of correlating to the changing value of alternating current.
changes in the amplitude, phase, quadrature Inductive reactance is measured in ohms.
components, or all of these of a complex test Numerically it is equal to the product of the
signal voltage to the condition of the test coil‘s inductance L, its excitation frequency f,
specimen. and 2π. It is often written as L, where angular
impedance, acoustic: See acoustic impedance. frequency  = 2πf.
impedance diagram, normalized: Diagram in which inductor: Passive electrical device employed in
the impedance of the probe in air is a reference circuits for its property of inductance. With or
value to which impedance values in other without a ferromagnetic core, it can be a coil
conditions are compared. Usually the plotted that impedes the flow of alternating current.
data are (1) the measured reactance divided by infrared: Beyond infrared, referring to radiation
the reactance of the coil in air versus (2) the with frequency lower than, and wavelength
measured resistance less the resistance in air greater than, that of the color red. See infrared
divided by the coil reactance in air. radiation.
impedance plane diagram: Graphical representation infrared radiation: Radiant energy beyond the color
of the locus of points indicating the variations red, of wavelengths from the red visible
in the impedance of a test coil as a function of a (0.75 µm) to about 300 µm, between the visible
parameter, such as conductivity or liftoff. and microwave regions of the electromagnetic
in-motion radiologic testing: Technique in which spectrum.
either the object being radiographed or the infrared thermography: Imaging of a temperature
source of radiation is in motion during the field through the emitted infrared radiation. See
exposure. infrared radiation. Compare thermography.
incident radiation: Primary radiation striking an inherent discontinuities: Discontinuities produced
object at closest point. in the material at the time it is formed (for
example, in metal, during solidification from the
molten state).

GLOSSARY 19.11
initial permeability: Slope of the induction curve at K
zero magnetizing force as the test specimen
begins to be magnetized from a demagnetized kaiser effect: Absence of detectable acoustic
condition (slope at the origin of the B,H curve emission until the previous maximum applied
before hysteresis isobserved). stress level has been exceeded.
initial pulse: Pulse applied to excite the transducer. kelvin: Absolute temperature scale related to the
It is the first indication on the screen if the celsius (or centigrade) relative scale. The kelvin
sweep is undelayed. Also called the main bang. unit is equal to 1 °C; 0 kelvin = -273.16 °C; the
May refer to an electrical pulse or an acoustic degree sign and the word degree are not used in
pulse. describing kelvin temperatures.
inside diameter coil: Coil or coil assembly used for
electromagnetic testing by insertion into the test L
piece, as with an inside-surface probe for
tubing. Sometimes called bobbin coil or inserted lamb wave: Type of ultrasonic wave propagation in
coil. which the wave is guided between two parallel
inspection reliability: Level of confidence, surfaces of the test object. Mode and velocity
expressed explicitly in statistical terms or depend on the product of the test frequency and
implied in noncritical cases, that is believed to the thickness. Also called plate wave.
exist while performing nondestructive tests. leakage flux: (1) Magnetic flux of the coil that does
Sometimes expressed in terms of probability of not link with the test object. (2) Magnetic flux
detection (POD), percentages of hits, misses, or that leaves a saturated or nearly saturated
false calls. specimen at a discontinuity.
interface: Physical boundary between two adjacent lens: Transparent object that refracts light passing
media. through it in order to focus the light.
interface triggering: Triggering the sweep and liftoff: Distance between the probe coil and the test
auxiliary functions from an interface echo object. In an electromagnetic test system output,
occurring after the initial pulse. Also called the effect observed due to a change in coupling
interfacesynchronization. between a test object and a probe whenever the
interlaced scanning: Process whereby the picture distance between them is varied.
appearing on a video screen is divided into two light: Visible radiation; radiant energy that can
parts. Interlaced scanning reduces flicker by excite the retina and produce a visual sensation;
increasing the electron beam‘s downward rate of visible portion of the electromagnetic spectrum,
travel so that every other line is sent. When the from about 380 to 800 nm.
bottom is reached, the beam is returned to the light metal: Low density metal such as aluminum,
top and the alternate lines are sent. The odd and magnesium, titanium, or beryllium.
even line scans are each transmitted at 1/60 s, linearity, amplitude: Constant proportionality
totaling 1/30 s per frame and retaining the between the signal input to the receiver and the
standard rate of 30 frames per second. The eye‘s amplitude of the signal appearing on the display
persistence of vision allows the odd and even of the ultrasonic instrument or on an auxiliary
lines to appear as a single image without flicker. display. Also called vertical linearity.
International Annealed Copper Standard (IACS): line pair: Pair of adjacent, parallel lines used to
Conductivity measurement system in which the evaluate the resolution of an imaging system.
conductivity of annealed, unalloyed copper is location plot: Spatial representation of acoustic
arbitrarily rated at 100 percent and in which the emission sources computed using an array of
conductivities of other materials are expressed transducers.
as percentages of this standard. longitudinal wave: Acoustic wave in which particle
interpretation: Determination of the cause and motion in the material is parallel to the
relevance of test indications and their direction of wave propagation. Also called
significance to the evaluation of the test object. compressional wave.
inverse square law: Physical law stating that, from loss of back reflection: Absence or significant
a point source of radiation, the intensity of reduction of an indication from the back surface
energy is inversely proportional to the square of of the test object.
the distance from the origin. low pass filtering: Linear combination of pixel
ionizing radiation: Form of radiation that can values to smoothen abrupt transitions in a
displace orbital electrons from atoms. Types digital image. Also called smoothing.
include X-rays, gamma rays, and particles such lumen (lm): Luminous flux per steradian from a
as neutrons, electrons, and alpha particles. source whose luminous intensity is 1 cd.
IQI: Image quality indicator. luminance: Ratio of a surface‘s luminous intensity
irradiance: Power of electromagnetic radiant energy in a given direction to a unit of projected area.
incident on or radiated from the surface of a Measured in candela per square meter.
given unit area. Compare radiance. luminosity: Luminous efficiency of radiant energy.
isotropy: Condition in which significant medium lux (lx): SI unit of illuminance, equal to one lumen
properties (sound speed, for example) are the per square meter (1 lx = 1 lm/m2).
same in all directions.
M maintenance philosophy: Theorganizedapproach
to keeping operational systems in satisfactory
machine vision: Automated system function of condition through use of fleet management and
acquiring, processing, and analyzing images to prescribed inspection practices. ―Fail safe
evaluate a test object or to provide information design,‖ ―retirement for cause,‖ ―condition based
for human interpretation. A typical system maintenance,‖ and ―structural health monitoring‖
consists of a light source, a video camera, a are different protocols for ensuring the integrity
video digitizer, a computer, and an image of critical test articles.
display. manipulator: In immersion testing, a device for
magnetic field: Distribution of a vector quantity angular orientation of the transducer and for
that is a measure of an exerted magnetic force. scanning motion in three axes.
May be used with both magnetic flux density B mask: (1) Square matrix of n × n with different
and magnetizing force H. The flux lines of a values that serves as a filter in image
typical magnetic field traverse the component in processing. (2) In radiologic testing, a cover with
a direction essentially parallel with its an aperture to view a specific area; mask plate.
longitudinal axis. (3) In radiologic testing, a selective radiation
magnetic field intensity (H): Strength of a filter.
magnetic field at a specific point. Measured in match plate: Device used in a high intensity
ampere per meter. illuminator to limit the light to a specific area,
magnetic flux density (B): Normal magnetic flux typically less than the size of the film
per unit area, measured in tesla (T). radiograph. See mask.
magnetic flux leakage field: Magnetic field that matte: Tending to diffuse light rather than reflect it;
leaves or enters the surface of an object. not shiny. The term matte is generally applied to
Excursion of magnetic lines of force from the smooth surfaces or coatings.
surface of a test specimen. It is the basis for the mechanical properties: Strength, hardness,
electromagnetic test technique for the detection toughness, elasticity, plasticity, brittleness,
and analysis of a surface discontinuity or near ductility, and malleability are mechanical
surface discontinuity using the flux that leaves a properties used as measures of how metals
magnetically saturated, or nearly saturated, test behave under a load (stress).
object at a discontinuity. microwave testing: Nondestructive testing method
magnetic flux meter: Electronic device for that uses, for its probing energy, electromagnetic
measuring magnetic flux leakage. Included are radiation at radio frequencies — from 0.3 to
items such as gauss meters and hall effect field 300 GHz, with wavelengths from 1 mm to 1 m.
meters. MKS system: Obsolete system of measurement units
magnetic particle testing: Nondestructive test based on the meter, kilogram, and second.
method using magnetic leakage fields and Superseded by SI.
indication materials to disclose surface and mode conversion: Change of ultrasonic wave
near-surface discontinuities. propagation mode upon reflection or refraction
magnetic permeability: Ratio of magnetic induction at an interface.
to magnetizing force. This relationship is either mode converted signal: Unintended signal from
(1) absolute permeability µ, in general the mode conversion of primary test angle, due to
quotient of magnetic induction B divided by the interaction with component geometry such as
magnetizing force H, or (2) relative permeability the signals after a back wall signal in a long
µr (or specific permeability), the magnetic narrow bar.
permeability of a material in comparison to that
of free space. For nonferromagnetic materials, mode of vibration: Manner in which an acoustic
the relative permeability is equal to unity. For wave is propagated, as characterized by the
particle motion in the wave (longitudinal,
ferromagnetic materials, the relative
transverse, lamb, or surface).
permeability varies from some specificmaterial‘s
modulus of elasticity: Ratio between stress and
nonmagnetized initial value through a peak
strain in a material deformed within its linear
value and then drops to unity under magnetic
elastic range.
saturation conditions due to increases in the
mottle: In radiologic test images, nonuniform
applied magnetizing force.
density where it should be uniform, resulting
magnetic saturation: That degree of magnetization
from scattered radiation, secondary radiation,
where a further increase in magnetizing force
forward scatter, and film irregularities. Often
produces no significant increase in magnetic
confused with graininess.
flux density in an object. In this region, the
MT: Magnetic particle testing.
permeability of the material is the same as that
multifrequency: Twoormorefrequenciesapplied
found in air or nonmagnetic materials.
sequentially or simultaneously to the test coil.
magnitude: Absolute value of a complex quantity
multifrequency technique: Use of the response of a
(number) without reference to the phase of the
test specimen to more than one frequency,
quantity.
usually to separate effects that would be
main bang: See initial pulse. indistinguishable at a single frequency.
multiple back reflections: Repetitive echoes from
the far boundary of the test object.

GLOSSARY 19.13
multiple-echo technique: Technique wherethickness nondestructive inspection (NDI): Alternativeterm
is measured between multiple back reflections, for nondestructive testing used in the aviation
minimizing error from coatings or from changes industry, particularly associated with military
in temperature or contact pressure. inservice maintenance.
mutual inductance: Property of two electrical nondestructive testing (NDT): Determination of the
circuits whereby a voltage is induced in one physical condition of an object without affecting
circuit by a change of current in the other that object‘s ability to fulfill its intended
circuit. function. Nondestructive test methods typically
use an appropriate form of energy to determine
material properties or to indicate the presence of
N material discontinuities (surface, internal, or
narrow band: Relative term denoting a restricted concealed). Sometimes called nondestructive
range of frequency response. evaluation, nondestructive examination, or
NDE: (1) Nondestructive evaluation. nondestructiveinspection.
(2) Nondestructive examination. See nonferromagnetic material: Material not
nondestructive testing. magnetizable and essentially not affected by
NDI: Nondestructive inspection. See nondestructive magnetic fields. Examples include aluminum,
testing. brass, austenitic stainless steel, and all
NDT: See nondestructive testing. nonmetallics with a relative permeability of
near field: Distance immediately in front of a plane unity.
transducer in which the ultrasonic beam exhibits nonrelevant indication: See indication, nonrelevant.
complex and changing wavefronts. Also called normal incidence: Condition in which the axis of
the fresnel field or fresnel zone. the ultrasonic beam is perpendicular to the entry
near ultraviolet radiation: Ultraviolet radiation with surface of the test object; that is, where the
wavelengths ranging from about 320 to about angle of incidence is zero.
400 nm. Formerly called black light. null: To adjust a bridge circuit so that the test
neural acuity: Ability of the eye and brain together sample and reference arms produce equal and
to discriminate patterns from background. opposite currents through the detector.
Discrimination is influenced by knowledge of
the target pattern, by the scanning technique,
and by an indication‘s relationship of figure to O
ground. ohm (Ω): Measurement unit of electrical impedance,
neutron: Uncharged elementary particle with mass both resistance and reactance.
nearly equal to that of the proton. optimum frequency: In electromagnetic testing, that
neutron fluence: Integrated exposure (product of frequency that provides the largest signal-to-
current and time) of neutrons per unit area. noise ratio obtainable for the detection of an
neutron flux: Neutron current; quantity of neutrons individual material property.
passing through a unit area per unit time. orientation: Angular relationship of a surface,
neutron radiologic testing: Radiologic testing using plane, discontinuity, or axis to a reference plane
a neutron beam. or surface.
neutron radioscopy: Radioscopy using a neutron
beam.
nodal points: In angle beam testing, the locations of P
reflections at opposite surfaces as a wave pancake coil: Probe coil whose axis is normal to the
progresses along a test object. surface of the test material and whose length is
noise: In electromagnetic testing, any nonrelevant not larger than its radius.
signals that tend to interfere with normal parallax: Apparent difference in position of an
reception or processing of a discontinuity signal imaged point according to two differently
of interest. The origin may be an electric or positioned sensors.
acoustic source, nondetrimental discontinuities, parameter distribution: Display of the number of
or abrupt changes in the acoustic properties of times an acoustic emission parameter falls
the test material. See also signal-to-noise ratio. between the values x and x + x as a function of
noncontact transducer: In ultrasonic testing, a x. Typical parameters are amplitude, rise time,
sensor designed for wave propagation through and duration.
gas. particle motion: Movement of particles of material
nondestructive characterization (NDC): Subelement during wave propagation.
of nondestructive testing concerned with the pencil break source: Artificial source using the
description of material properties and their fracture of a brittle graphite or equivalent
behavior within components and systems. An cylinder in a suitable fitting to simulate an
example includes the use of eddy current acoustic emission signal. Also called hsu-nielson
techniques to identify the presence of fire source.
damage in aluminum alloys. penetrameter: See image quality indicator.
nondestructive evaluation (NDE): Another term for period: Value of the minimum duration after which
nondestructive testing. In research and academic the same characteristics of a periodic waveform
communities, the word evaluation is often or a periodic feature repeat.
preferred because it emphasizes interpretation by phantom: Reference standard used to verify the
knowledgeable personnel. performance of diagnostic ultrasonic systems.
phase analysis: Analytical technique that primary reference response level: Ultrasonic
discriminates between variables in a part response from the basic reference reflector at the
undergoing electromagnetic testing by the specified sound path distance, electronically
different phase angle and amplitude changes adjusted to a specified percentage of full screen
that these conditions produce in the test signal. height.
See also phase detection. probability of detection (POD): Measure of
phase angle: Angular equivalent of the time inspection reliability based on the statistical
displacement between corresponding points on performance of detection during a controlled
two sine waves of the same frequency. A phase study using a collection of precise test targets
shift is a change in the phase relationship providing indications of interest for a given test
between two alternating quantities of the same scenario. Statistical statement from a test
frequency. procedure indicating how likely a given
phase detection: Derivation of a signal whose discontinuity length may be reliably found.
amplitude is a function of the phase angle probe: See sensor; transducer.
between two alternating currents, one of which probe index: Point on a transverse wave or surface
is used as a reference. A phase sensitive system wave transducer through which the emergent
is one whose output signal depends on the beam axis passes. See also point of incidence.
phase relationship between the voltage returned process control: Application of quality control
from a pickup or sensing coil and a reference principles to the management of a repeated
voltage. process.
phase velocity: Velocity of a single-frequency pulse: Transient electrical or ultrasonic signal that
continuous wave. has a rapid increase in amplitude to its
phased array: In ultrasonic testing, a phased array maximum value, followed by an immediate
is a mosaic of transducer elements in which the return.
timing of the elements‘ excitation can be pulse echo technique: Ultrasonic test technique in
individually controlled to produce certain which discontinuities are revealed by echoes
desired effects, such as steering or focusing the from the transmitted pulses.
beam. pulse frequency: In ultrasonic testing, the number
photoelectric effect: Emission of free electronsfrom of pulses generated or transmitted per unit of
a surface bombarded by sufficiently energetic time (usually seconds). Also called repetition
photons. Such emissions may be used in an rate.
illuminance meter, calibrated in lux. Interaction pulse length: Measure of pulse duration expressed
of photons with atoms in which the full energy in time per pulse or in number of cycles per unit
of the photon is absorbed by an orbital electron, of time.
removing the electron from the atom. pulse technique: Multifrequency technique in which
photoemission: Method by which an image orthicon a broadband excitation such as an impulse is
television camera tube produces an electrical used. Either the frequency components are
image, in which a photosensitive surface emits extracted and analyzed or the interpretation is
electrons when light reflected from a viewed based directly on characteristics of the time
object is focused on that surface. Compare domain waveform.
photoconduction. pulse tuning: Control of pulse frequency to optimize
photon: Particle of electromagnetic radiation. system response.
photoreceptor: Photon sensor. Examples include
film and electronic detector elements.
physical properties: Nonmechanical properties such Q
as density, electrical conductivity, magnetic Q, or quality factor, of a coil: Quality factor of a
permeability, thermal diffusivity, dielectric coil, as the ratio of inductive reactance to
constant, and thermal expansion. resistance.
picture element: See pixel. quadrature: Relation between two periodic
piezoelectric effect: Ability of certain materials to functions when the phase difference between
convert electrical energy (voltage) into them is 90 degrees.
mechanical energy (stress) and vice versa. quality: Ability of a process or product to meet
pitch catch technique: Ultrasonic test technique that specifications or to meet the expectations of its
uses two transducers, one transmitting and the users.
other receiving on the same or opposite surface. quality assurance: Administrative actions that
Also called double-crystal technique or specify, enforce, and verify a quality program.
two-transducer technique. quality control: Physical and administrative actions
plane wave: See longitudinal wave. required to ensure compliance with the quality
plate wave: See lamb wave. assurance program. Quality control may include
point of incidence: Point at which the axis of the nondestructive testing in manufacturing or
sound beam leaves the wedge of an angle beam service.
transducer and enters the test object. See also
probe index.
primary radiation: Radiation emitting directly from
the target of an X-ray tube or from a
radioactive source.

GLOSSARY 19.15
R radioscopy: Radiographic testing technique in which
gamma rays, X-rays, or neutrons are used to
rad: (1) SI symbol for radian. (2) Radiation absorbed produce an image on a video or screen display
dose; unit of absorbed dose of ionizing as opposed to a latent image on a film. The test
radiation. One rad is equal to the absorption of object or interrogating optics may move in real
10–5 J (100 erg) of radiation energy per gram of time to present a moving radiologic test image.
matter. Replaced by the gray (Gy). range: In ultrasonic testing, the maximum path
radian (rad): Measurement unit of plane angle length displayed. See also sweep length.
subtending, in a circle, an arc equal in length to rarefaction: Thinning or separation of particles in a
the radius. propagating medium due to the decompression
radiance: Radiant flux per unit solid angle and per phase of a longitudinal ultrasonic cycle.
unit projected area of the source. Measured in Opposite of compression. A compressional wave
watts per square meter steradian. Compare is composed of alternating compressions and
irradiance. rarefactions.
radiant energy: Energy emitting as electromagnetic rayleigh wave: Ultrasonic wave that propagates
waves. Also called radiation. along the surface of a test object. The particle
radiant flux: Radiant energy‘s rate of flow, motion is elliptical in a plane perpendicular to
measured in watts. the surface, decreasing rapidly with depth below
radiant intensity: Electromagneticenergy emitted the surface. The effective depth of penetration is
per unit time per unit solid angle. considered to be about one wavelength. Also
radiant power: Total radiant energy emitted per unit called surface wave.
time. receiver: (1) Section of the ultrasonic instrument
radiation safety officer: Individual supervising that amplifies echoes returning from the test
program to provide radiation protection. The object. (2) Transducer that picks up the echoes.
representative appointed by the licensee for reference coil: In electromagnetic testing, the
liaison with the applicable regulatory agency. section of the coil assembly that excites or
radio frequency display: Presentation of unrectified detects the electromagnetic field in the reference
signals. See also video presentation. standard of a comparative system.
radio operating characteristic (ROC): A measure of reference standard: A material or object for which
inspection performance that uses both the relevant chemical and physical
probability of detection and probability of false characteristics are known and measurable, used
calls as its major variables. as a comparison for, or standardization of
radiographer: Person that performs, supervises, and equipment or instruments used for,
is responsible for industrial radiologic test nondestructive testing. A simulated test article
operations. with artificial discontinuities used for
radiologic test interpretation: Determination of establishing and periodically checking required
the cause and significance of discontinuities test sensitivity settings.
indicated on a radiologic image. reference threshold: Preset voltage level that has to
radiologic test screens: Thin sheets used to intensify be exceeded before an acoustic emission signal
the effect of radiation on films. The screens can is detected and processed. This threshold may be
be made of a fluorescent material or a metal adjustable, fixed, or floating. See also threshold
such as lead. Metallic screens absorb secondary level.
and scattered radiation, which helps to improve reflection probe: Coil system that uses both an
image quality. excitation and a detection or sensing coil on the
radiologic testing: Use of penetrating radiant same side of the sample.
energy in the form of X-rays, gamma rays, or reflectivity: Ability of a surface to reflect
neutrons for nondestructive testing of objects to electromagnetic radiation, expressed as ratio 
provide images of the objects‘ interiors. Also of the intensity of the total energy reflected
called radiography; radiologic testing. from a surface to total radiation on that surface.
radiography: See radiologic testing. For a perfect mirror, reflectivity  approaches
radiologic testing (RT): Use of penetrating radiant 1.0; for a blackbody, the reflectivity is zero.
energy in the form of X-rays, gamma rays, or refracted beam: Beam transmitted in the second
neutrons for nondestructive testing of objects to medium when an ultrasonic beam is incident at
provide images of the objects‘ interiors. Also an acute angle on the interface between two
called radiography; radiographic testing. media having different sound speeds.
radiology: (1) That branch of medicine which uses refraction: Change in direction of an acoustic wave
ionizing radiation for diagnosis and therapy. as the ultrasonic beam passes from one medium
(2) Science of electromagnetic radiation, into another having different acoustic speeds. A
particularly ionizing radiation. change in both direction and mode occurs at
radiometer: Device used to measure irradiance or acute angles of incidence. At small angles of
radiant energy of specified frequencies. Different incidence, the original mode and a converted
radiometers exist for different frequencies. In mode may exist simultaneously in the second
nondestructive testing, radiometers are used to medium.
measure UV-A output, or leaked visible light, in refractive index: Ratio of the speed of the incident
microwatt per square centimeter (µW/cm2). See wave to that of a refracted wave. It is known as
also irradiance. Compare photometer. the refractive index of the second medium with
radiometric photometer: Radiometer for measuring respect to the first.
radiant power over a variety of wavelengths.
reject: Minimize or eliminate low amplitude signals roentgen (R): Unit for measurement of radiation
(such as electrical or material noise) so that intensity; amount of radiation that will generate
other signals may be further amplified. This one electrostatic unit in 1 cm3 of air at standard
control can reduce vertical linearity. Also called atmospheric conditions. The roentgen (R) has
suppression. been replaced by an SI compound unit, coulomb
rejection level: Level above or below which a signal per kilogram (C/kg).
is an indication of a rejectable discontinuity. roof angle: In a dual-element delay line transducer,
rem: Roentgen equivalent man. A unit of absorbed the tilt angle by which the transducer elements
ionizing radiation in biological matter. See of the delay line are oriented to direct the beams
sievert. of the two elements to intersect at a specified
remote viewing: (1) Term introduced in the late zone in the medium.
twentieth century to denote visual testing root mean square (RMS): Statistical measure of the
mediated through a system of two or more magnitude of a varying quantity, such as current.
lenses (as in a borescope) or transduced through Square root of the mean square of a set of
an electronic signal (as with a charge coupled measures, usually a time series.
camera). Indirect viewing. (2) In telemetry and RT: Radiographic testing; radiologic testing.
robotics, the technology and visual display of
scenes not in the viewer‘s immediate presence.
repetition rate: See pulse frequency. S
reserve vision acuity: Ability of an individual to saturation: Condition in which high amplitude
maintain vision acuity under poor viewing signals on a display screen do not increase with
conditions. A visual system with 20/20 near increased gain and appear flattened.
vision acuity under degraded viewing conditions scalar: Quantity completely specified by a single
has considerable reserve vision acuity compared number and unit. Examples include weight and
to that of an individual with 20/70 near vision speed.
acuity. scanning: Movement of the transducer over the
resistance, electrical (R): Opposition to transmission surface of the test object in a controlled manner
of electric current through a material; ratio of so as to achieve complete coverage. May be
voltage to current. Measured in ohms (Ω). either a contact or immersion technique.
Inversely related to conductance. scattering: Reflection of ultrasonic waves by small
resistivity (): Ability of a material to resist electric discontinuities or surface irregularities.
current. Measured in ohm meter (Ω·m), the scintillation: Emission of light of specific
resistance of a cube made of the material whose frequencies after the absorption of
dimensions are 1 m on each side. Inversely electromagnetic radiation, such as X-rays or
related to electrical conductivity  (siemens per gamma rays.
meter) scintillation detector: Radiation measuring device
resolution: Aspect of image quality pertaining to a based on a scintillating material.
system‘s ability to reproduce objects, often search unit: See transducer.
measured by resolving a pair of adjacent objects selectivity: Characteristic of a test system, a
or parallel lines. See also line pair; resolving measure of the extent to which an instrument
power. can differentiate between the desired signal and
resolving power: Ability of detection systems to disturbances of other frequencies or phases.
separate two points or lines in time or distance. send/receive transducer: Transducer consisting of
Resolving power depends on the angle of vision two piezoelectric elements mounted side by side
and the distance of the sensor from the test separated by an acoustic barrier. One element
surface. Compare resolution. transmits; the other receives.
resonance: Condition in which the frequency of a sensing coil: Coil that detects changes in the flow
forcing vibration (ultrasonic wave) is the same of eddy currents induced by an excitation coil.
as the natural vibration frequency of the In simple probes, the sensing and excitation
propagation body (test object), possibly resulting coils are usually one and the same.
in large amplitude vibrations. The resonance sensitivity: Measure of a sensor‘s ability to detect
principle is used for determining acoustic speed, small signals. See resolution.
object thickness, or presence of laminar sensitization: Condition of exposed silver halide
discontinuities. emulsion in radiologic film before development.
resonant frequency: Frequency at which a body sensor, X-ray: In radiologic testing, a device or
vibrates freely after being set in motion by some material that changes with and provides
outside force. evidence of contact with ionizing radiation.
retina: In the eye, the tissue that senses light. Examples include X-ray film, X-ray sensitive
ringing signals: (1) Closely spaced multiple signals phosphors, and electronic devices such as linear
caused by multiple reflections in a thin material. detector arrays. See detector, X-ray.
(2) Signals caused by continued vibration of a shadow: Region in a test object that cannot be
transducer. reached by ultrasonic energy traveling in a
ringing technique: Test technique for bonded given direction. Shadows are caused by
structures in which unbonds are indicated by geometry or the presence of intervening large
increased amplitude of ringing signals. discontinuities.
ringing time: Time that the mechanical vibrations shear wave: See transverse wave.
of a transducer continue after the electrical
pulse has stopped.

GLOSSARY 19.17
shielding: Material or object used to reduce spectroradiometry: Measurement of electromagnetic
intensity of or exposure to penetrating radiation radiant power and spectral emittance, used
or external interference. particularly to examine colors and to measure
shoe: Device used to adapt a straight beam the spectral emittance of light sources.
transducer for use in a specific type of testing, spectroscopy: Spectrophotometry or
including angle beam or surface wave tests and spectroradiometry in which the spectrum, rather
tests on curved surfaces. See also wedge. than being analyzed only by a processing unit,
SI (International System of Units): Measurement is presented as a digital signal for computer
system in which the following seven units are analysis or in a visible form to the operator for
basic: meter, mole, kilogram, second, ampere, organoleptic examination.
kelvin, and candela. spectrum: (1) Amplitude distribution of frequencies
siemens per meter (S/m): SI unit of conductivity. in a signal. (2) Representation of radiant energy
sievert: SI unit for measurement of exposure to in adjacent bands of hues in sequence according
ionizing radiation, replacing rem. 1 Sv = 1 J/kg to the energy‘s wavelengths or frequencies. A
= 100 rem. rainbow is a well known example of a visible
signal: Physical quantity, such as voltage, that spectrum.
contains relevant information. speed of light: Speed of all radiant energy,
signal processing: (1) Acquisition, storage, analysis, including light, is 2.997925 × 108 m/s in
alteration, and output of digital data through a vacuum (approximately 186 000 mi/s). In all
computer. (2) In infrared and thermal testing, transparent materials the speed is less and varies
manipulation of temperature signal or image with the material‘s index of refraction, which
data to enhance or control a process. Examples itself varies with wavelength.
for infrared radiation thermometers are peak spherical wave: Wave in which points of the same
hold, valley hold, averaging, and sample and phase lie on surfaces of concentric spheres.
hold. Examples for scanners and imagers are Often associated with point sources of sound.
usually referred to as image processing and spurious echo: General term denoting any ill
include qualitative characterization, quantitative defined indication that cannot be associated
characterization, alignment, isotherm with a discontinuity or boundary at the location
enhancement, image subtraction, image displayed. Also called parasitic echo.
averaging, and image filtering. squirter system: An inspection apparatus composed
signal-to-noise ratio: Ratio of signal values of two or more immersion transducers
(responses that contain relevant information) to acoustically connected through a stream of
baseline noise values (responses that contain flowing water. Test parts are positioned between
nonrelevant information). See noise. the two transducers and are thus mostly
skin effect: Phenomenon wherein the depth of inspected with the through-transmission
penetration of electrical currents into a approach.
conductor decreases as the frequency of the standard: (1) Physical object usually containing an
current is increased. At very high frequencies, artificial discontinuity and used for comparison
the current flow is restricted to an extremely or calibration, such as a calibration block.
thin outer layer of the conductor. See standard (2) Concept or practice that has been established
depth of penetration. by authority, custom, or agreement to serve as a
skip distance: In angle beam tests of plate or pipe, model rule in the measurement of quality or the
the distance from the sound entry point to the establishment of a practice or procedure.
exit point on the same surface after reflection (3) Document to control and govern practices in
from the back surface. Also called V path. an industry or application, applied on a national
Snell’s law: Physical law that defines the or international basis and usually produced by
relationship between the angle of incidence and consensus.
the angles of reflection and refraction. standard atmospheric conditions: Atmospheric
source, acoustic emission: (1) Unique mechanism pressure of 101.325 kPa (14.6959 lbf/in.2).
that generates acoustic emission. (1) Place of an Temperature of 293.15 K (20 °C or 68 °F). The
acoustic emission event. density of dry air at these conditions is
source location: Determination of the location of an 1.2041 kg/m3 (0.07517 lbf/ft3).
acoustic emission source from arrival times by standard depth of penetration: In electromagnetic
using multiple transducers and triangulation testing, the depth at which the magnetic field
schemes. intensity or intensity of induced eddy currents
source: Machine or material from which ionizing has decreased to 37 percent of its surface value.
radiation emanates. The square of the depth of penetration is
specification: Set of instructions or standards inversely proportional to the frequency of the
invoked by a specific customer to govern the signal, the conductivity of the material, and the
results or performance of a specific set of tasks permeability of the material. See also skin effect.
or products. standing wave: Wave in which the energy flux is
spectral power distribution: Radiant power per unit zero at all points. Such waves result from the
wavelength as a function of wavelength. Also interaction of similar waves traveling in
known as spectral energy distribution, spectral opposite directions as when reflected waves
density, and spectral distribution. meet advancing waves. A particular case is that
spectrophotometry: Measurement of the luminance of waves in a body whose thickness is an
or illuminance produced by electromagnetic integral multiple of half-wavelengths, as in
radiation as a function of wavelength. resonance testing.
Stefan-Boltzmann law: Relationship governingthe thermography: Imaging or viewing of an object or
wavelength independent rate of emission of process through sensing of heat emitted by it.
radiant energy per unit area. The law relates the The temperature patterns on the material surface
total radiation intensity to the fourth power of produce corresponding radiation patterns. Thus,
absolute temperature and emissivity of the heat flow by both conduction and radiation may
material surface. For example, intensity (heat be observed and used to locate material
flow) from a copper block at 100 °C (212 °F) is discontinuities. Most often, thermography is
300 W/m2 (95 BTU·ft–2·h–1). (Stefan-Boltzmann based on sensing of infrared radiation. See also
constant for photon emission = 1.52041 × infrared thermography.
1015 photon·s–1·m–2·K–2.) threshold level: Setting of an instrument that causes
stepped wedge: Reference object, with steps of it to register only those changes in response
various thicknesses in the range of tested parts‘ greater or less than a specified magnitude.
thicknesses, for the radiologic testing of parts through-transmission technique: Test technique in
having thickness variations or complex which ultrasonic energy is transmitted through
geometries. The stepped wedge must be made of the test object and received by a second
material radiologically similar to that of the transducer on the opposite side. Changes in
radiologic test object and may include received signal amplitude are taken as
penetrametric features (such as calibrated holes) indications of variations in material continuity.
in any or all steps. time of flight: Time for an acoustic wave to travel
stereo imaging: Imaging technique involving the between two points. For example, the time
capture and display of two images of the same required for a pulse to travel from the
object from different angles. Binocular viewing transmitter to the receiver via diffraction at a
simultaneously of the two images simulates a discontinuity edge or along the surface of the
three-dimensional viewing. test object.
straight beam: Ultrasonic longitudinal wave tone burst: Wave train consisting of several cycles
traveling normal to the test surface. of the same frequency.
stress corrosion cracking (SCC): A form of transducer: (1) Device that converts mechanical
corrosion caused within ductile metals by the energy to electrical output or vice versa.
presence of tensile stresses and a corrosive (2) Piezoelectric device that converts attributes
environment. For aluminum, the presence of of the stress-versus-strain field of an acoustic
chlorides, even in minute quantities, can lead to wave into an electrical signal of voltage versus
rapid crack growth even if the stress is below a time. Sensor; probe.
typical crack growth stress level. transducer, differential: Piezoelectric twin-element
surface wave: See rayleigh wave. or dual-pole transducer.
survey meter: Portable instrument that measures transducer element: In an ultrasonic transducer, the
rate of exposure dose or ionizing radiation piezoelectric crystal to be coupled to the test
intensity. surface. Also called the crystal.
Sv: Sievert. transducer, flat response: Transducer whose
sweep: Uniform and repeated movement of a spot frequency response has no resonance or
across the display screen to form the horizontal characteristic response within its specified
baseline. Also called time base. frequency band. The ratio between the upper
sweep delay: (1) Delay in time of starting the sweep and lower limits of the frequency band are
after the initial pulse. (2) Control for adjusting typically not less than 500 kHz.
the time. Also called time delay. transducer, resonant: Transducer that uses the
sweep length: Length of time or distance mechanical amplification due to a resonant
represented by the horizontal baseline on an frequency (or several close resonant frequencies)
A-scan. to give high sensitivity in a narrow band,
typically 10 percent of the principal resonant
frequency at the –3 dB points.
T transducer, wide band: Transducer whose response
tesla (T): SI unit of measurement for magnetic flux to surface displacements is relatively flat over a
density. 1 T = 1 Wb/m2 = 10 000 G. broad frequency range.
thermal diffusion: Process by which thermal energy transmission angle: Incident angle of a transmitted
is transferred from hot or cold regions and ultrasonic beam. It is zero degrees when the
finally is spread out. See also conduction, beam is perpendicular (normal) to the test
thermal. surface.
thermogram: Thermal map or image of a target transmission technique: See through-transmission
where the gray tones or color hues correspond technique.
to the distribution of infrared thermal radiant transmitter: (1) Transducer that emits ultrasonic
energy over the surface of the target (qualitative energy. (2) Electrical circuits that generate the
thermogram). When correctly processed and signals emitted by the transducer.
corrected, a thermogram graphically represents transverse vertical (polarized) wave: Transverse
surface temperature distribution (quantitative wave in which the plane of vibration is normal
thermogram). to the incidence surface.
transverse wave: Type of wave in which the
particle motion is perpendicular to the direction
of propagation. Also called shear wave.

GLOSSARY 19.19
transverse wave transducer: Transducer that W
generates transverse waves in a test object. Also
called a shear wave transducer. water column: Tube filled with water and attached
two-transducer technique: See pitch catch to the front of a transducer to couple an
technique. ultrasonic beam to a test object. A delay line
ultrasonic: Of or relating to acoustic vibration between the initial pulse and the front surface
frequencies greater than about 20 kHz. signal. Also serves as a coupling device. See also
ultrasonic absorption: Damping or dissipation of delay line.
ultrasonic waves as they pass through a water entrapment: A condition wherein water has
medium. See also attenuation coefficient. entered into the hollow spaces making up the
ultrasonic spectroscopy: Analysis of the frequency core of a honeycomb structure.
content of an acoustic wave. Generally water jet: Unsupported stream of water carrying
performed mathematically using a fast fourier ultrasonic signals between the transducer and
transform. the test object surface. Also called a squirter or
ultrasonic spectrum: Usually thefrequency range water column.
from 20 kHz to 50 MHz, which may extend water path: In immersion testing or with a water
much higher in special applications. column, the distance from the receiving
ultrasonic testing: Method of nondestructive transducer‘s face to the test object‘s front
testing, using acoustic waves at inaudibly high surface.
frequencies as the interrogating energy. wavefront: In a wave disturbance, the locus of
Unified Numbering System (UNS): Alphanumeric points having the same phase.
system for identifying alloys according to a waveguide: Device to transmit elastic energy from a
registry maintained by ASTM International and test object to a remote transducer — for
SAE International. example, a wire joined at one end to a test
unsharpness, geometric: Fuzziness or lack of object and at the other end to a transducer.
definition in a radiologic image resulting from wave interference: Production of a series of
the source size, object-to-sensor distance, and maxima and minima of sound pressure as a
the source-to-object distance. consequence of the superposition of waves
UT: Abbreviation for the ultrasonic method of having different phases. Often used to describe
nondestructive testing. near and far field effects.
wavelength: Distance between repeating values of a
wave — for example, the distance from one peak
V to the next peak on a sine wave. Wavelength is
V path: See skip distance. a fundamental descriptor when discussing wave
vector quantity: Any physical quantity whose behavior, system sensitivity, and diffraction
specification involves both magnitude and effects.
direction. Examples include magnetic and wave train: Series of waves or groups of waves
electric fields, electrical impedance, and velocity. passing along the same course at regular
vertical limit: Maximum useful readable level of intervals.
vertical indication on an A-scan. wear face: Protective material on the face of a
vertical linearity: See linearity, amplitude. transducer to prevent wear of the piezoelectric
video: Pertaining to the transmission and display of element.
moving images in an electronic format that can wedge: Device used to direct ultrasonic energy into
be displayed on a monitor or screen. a test object at an acute angle. See also shoe.
video presentation: Displaypresentation of moving wheel transducer: Device that couples ultrasonic
images in which radiofrequency signals have energy to a test object through the rolling
been rectified and usually filtered. contact area of a wheel containing a liquid and
vidicon tube: Television tube that uses the one or more transducers.
photoconduction method. Compare image wobble: In electromagnetic testing, an effect that
orthicon. produces variations in an output signal of a test
visible light: Radiant energy in the 400 to 700 nm system caused by probe rocking on a surface or
wavelength range. coil displacement within an encircling coil.
vision acuity: Ability to distinguish fine details Often used to set variable phase displays to a
visually at a given distance. Quantitatively, it is horizontal orientation as a means for
the reciprocal of the minimum angular standardizing signal presentations on vector
separation in minutes of two lines of width displays.
subtending one minute of arc when the lines are wrap around: Display of misleading ultrasonic
resolvable as separate. reflections from a previously transmitted pulse
visual testing: Method of nondestructive testing because of excessive pulse repetition frequency.
using electromagnetic radiation at visible See also ghost. Largely corrected with modern
frequencies. digital instruments.
volt (V): Measurement unit of electric potential.
X X-ray fluorescence (XRF): Radiologic testing technique used
for material characterization, based on wavelengths of
X-ray: Penetrating electromagnetic radiation emitted when the fluorescence from object irradiated by X-rays.
inner orbital electrons of an atom are excited and release XRD: X-ray diffraction.
energy. Radiation is nonisotopic in origin and is generated XRF: X-ray fluorescence.
by bombarding a metallic target with high speed charged
particles, usually electrons.
X-ray diffraction (XRD): Radiologic testing technique used for
material characterization, based on change in scattering of
X-radiation as a result of interaction with test material. See
also diffraction.
References
ASNT. 1998. Nondestructive Testing Handbook 1: Leak Testing, ASNT. 2008. Nondestructive Testing Handbook 8: Magnetic
third edition. Columbus, OH: American Society for Testing, third edition. Columbus, OH: American Society for
Nondestructive Testing. Nondestructive Testing.
ASNT. 1999. Nondestructive Testing Handbook 2: Liquid ASNT. 2010. Nondestructive Testing Handbook 9: Visual Testing,
Penetrant Testing, third edition. Columbus, OH: American third edition. Columbus, OH: American Society for
Society for Nondestructive Testing. Nondestructive Testing.
ASNT. 2001. Nondestructive Testing Handbook 3: Infrared and ASNT. 2012. Nondestructive Testing Handbook 10:
Thermal Testing, third edition. Columbus, OH: American Nondestructive Testing Overview, 3rd edition. Columbus, OH:
Society for Nondestructive Testing. American Society for Nondestructive Testing.
ASNT. 2002. Nondestructive Testing Handbook 4: Radiographic ASTM. 2007. ASTM E 1316, Standard Terminology for
Testing, third edition. Columbus, OH: American Society for Nondestructive Examinations. West Conshohocken, PA:
Nondestructive Testing. ASTM International. doi:10.1520/E1316.
ASNT. 2004. Nondestructive Testing Handbook 5: CRC. 2014. CRC Handbook of Chemistry and Physics,
Electromagnetic Testing, third edition. Columbus, OH: 95th edition. Boca Raton, FL: CRC Press.
American Society for Nondestructive Testing. Mordfin, L. 2002. Handbook of Reference Data for
ASNT. 2005. Nondestructive Testing Handbook 6: Acoustic Nondestructive Testing. West Conshohocken, PA: ASTM
Emission Testing, third edition. Columbus, OH: American International.
Society for Nondestructive Testing.
ASNT. 2007. Nondestructive Testing Handbook 7: Ultrasonic
Testing, third edition. Columbus, OH: American Society for
Nondestructive Testing.
Index

A allowable damage limit, 2.31

alloy composition
aramid (aromatic polyamide) fibers, 2.25, 2.27
arc burn
AAWG (Airworthiness Assurance Working Group), 2.22 beta backscatter method, 17.23 caused by magnetic field induction, 7.4,
abradable seals, 4.8 X-ray fluorescence spectroscopy, 17.18–17.19 7.14, 7.15
abrasive wear, 2.12 Aloha Airlines Flight 243, 2.21–2.22 on turbine blades, 5.20
absolute coils, for eddy current testing, 8.12, 8.18
aluminum Ares I, 2.5
of bolt holes, 8.41 bolt holes, 8.41, 8.42 Ares V, 2.5
acceptance/rejection criteria, development of, 1.6, 2.9 bonded joint assemblies, 4.7 artifacts. See radiologic testing, artifacts in
acoustic emission testing, 13.1–13.16 castings, 4.5, 4.6 ASIP (Aircraft Structural Integrity Program), 2.15,
advantages and limitations, 13.11 conductivity, 8.9 2.17, 3.3
applications, 13.11–13.16 eddy current testing, 8.2, 8.27, 8.32 aspect ratio
for bond evaluation, 13.14–13.15, 14.3 fastened structures, 4.6, 8.37–8.40 of infrared camera detector array, 11.8
cross reference table, 4.12, 4.18 fire damaged, 8.32 of thermographically detectable discontinuity, 11.5,
principles, 13.2–13.6 forgings, 4.6 11.6, 11.11
standards and specifications, 13.9–13.10 impedance plane plot, 8.6 attenuation cross section, 10.3, 18.4
techniques, 13.7–13.8 machined parts, 4.6 attenuation of ultrasonic waves, 9.6, 18.3
what can go wrong, 13.12 magnetoresistive sensing, 8.24 attenuation of X-rays, 10.2–10.5, 18.4. Seealso linear
acoustic excitation, safety in use of, 12.30–12.31 part life monitoring, 17.26 attenuation coefficient
acoustic holography, 9.26–9.27 plate, guided waves in, 9.25 automatic eddycurrenttesting, 3.9
acoustic impedance, 9.5, 9.6, 9.7, 9.8, 14.8 sheet, ultrasonic inspection, 9.33 bolt hole inspection, 8.40, 8.42–8.43
formula for, 18.3 stress-to-strain curve, 2.10 engine components, 3.9, 8.46, 8.47
of ultrasonic transducers, 9.11 thickness taper gage, 8.37 avionics, 2.2, 2.3. See also electronics
acoustic leak testing, 16.5–16.6 welded joints, 4.7
of spacecraft, 15.10–15.11 of World War II aircraft, 1.5
acoustic thermography, 11.14–11.15
acoustic waves. See ultrasonic waves
X-ray diffraction, 17.15
X-ray transmission, 10.19, 10.20
B
acousto-optic sensor, 9.26 aluminum alloys B-47 Stratojet, 2.14–2.15
adhesive bonds, 1.4. Seealso bond testing; backscatter radiologic testing, 10.34, 10.35, 10.38
castings, radiologic testing, 10.37
bonded assemblies eddy current testing, 8.11, 8.12, 8.35, 8.44 bad pixels, 10.25–10.27
adhesive wear, 2.12 fatigue resistance, 2.14, 2.16 barely visible impact damage
adiabatic behavior, 11.3–11.4 on composites, 2.30, 2.31
magneto-optic imaging, 8.22
adiabatic discontinuity, 11.5 natural aging, 8.31 on honeycomb sandwich structures, 14.2
admittance, 17.13 viewing distance for, 5.8
representative properties, 2.11
aerospace nondestructive testing, 1.3–1.4. See also stronger 2000 series, 2.14, 2.16 barkhausen noise measurements, 4.12, 4.19, 17.2–17.4
nondestructive testing (NDT) beta backscatter, 17.22–17.23
X-ray fluorescence spectroscopy, 17.18, 17.19
aerospace structure, 2.3 billet, ultrasonic inspection, 9.31
aluminum alloys, conductivity
aerospace systems, 2.2, 2.3 black light. See ultraviolet lamps; ultraviolet radiation
with cladding of pure aluminum, 8.33
aerospace vehicles, 2.2, 2.3–2.5 blending, of turbine blades, 5.20
eddy current measurement of, 8.30–8.32, 8.33
aging aircraft factors affecting, 8.8 blown core, 2.33
Aloha Airlines incident and, 2.22 aluminum honeycomb. See also honeycomb core; bobbin coils, 8.13
eddy current testing, 8.27 honeycomb structures Boeing 737, 2.2
regulatory responses, 2.22 with carbon fiber reinforced resin face Aloha Airlines Flight 243, 2.21–2.22
subsurface discontinuities or corrosion, 8.23 Boeing F-18, 2.4
sheets, 17.10
updated damage tolerance plan, 2.21 Boeing RS68 liquid rocket engine, 2.7
microwave testing, 17.10
widespread fatigue damage, 2.21–2.23 radiographic testing, 10.40 bolt holes, eddy current testing, 8.40–8.46. See also
air coupled ultrasonic testing, 9.21, 14.3, 14.5, 14.8 shearography, 12.12, 12.25, 12.27 fastener holes
aircraft, 2.3–2.4. See also aging aircraft; military aircraft thermography, 11.12, 11.13 automated, 8.40, 8.42–8.43
Aircraft Structural Integrity Program (ASIP), 2.15, ammonia/phenolphthalein leak testing, 16.4 erratic signal, 8.30
2.17, 3.3 amorphous silicon image capture devices, 10.13, 10.14 meandering winding magnetometer, 8.20, 15.12
aircraft structure, 2.3 multiple-layer, 8.29
analytical design life, 2.17, 2.19
coordinate systems for, 2.4 radiologic testing, 10.42
annealing, 8.32
typical components, 2.2 anomalies. See also discontinuities ultrasonic testing, 9.34
airworthiness, approaches to, 2.19–2.20 aircraft design and, 2.8, 2.16 boltedstructures, 4.6. Seealso fastenedstructures
Airworthiness Assurance Working Group (AAWG), 2.22 bolts, free, as foreignobject, 10.42
in composite materials and structures,
airworthiness certification, 2.3, 2.13 2.28, 2.34 bombers, 2.3, 2.14. See also military aircraft
bonded assemblies, 2.34–2.35 B-47 Stratojet, 2.14–2.15
F-111, 2.15, 2.17, 2.25, 3.3, 15.4

20.1
bond strength, 14.2, 14.5, 14.8–14.9 castings, radiologic testing, 1.5, 10.37 composites, 2.24–2.31. Seealso impactdamage, to
surface contamination and, 17.21 cracks in, 10.41 composite components
bondtesting, 1.4, 14.1–14.9. Seealso delaminations; reference images, 10.36 anisotropy of, 2.25
disbonds ceramic matrix composites, microwave interferometry, anomalies and damage, types of, 2.28, 2.34
categories of techniques, 14.5 17.10 in bonded assemblies, 2.32–2.35, 4.7
descriptions of selected techniques, 14.5–14.9 ceramics, 4.5 ceramic matrix, microwave inspection, 17.10
inspection issues, 4.7 porosity, dielectrometric measurement, 17.12 certification of structures, 2.28–2.30
introduction to, 14.2 certification of aircraft. See airworthiness certification; definition of, 2.24–2.25
magnetostrictive monitoring, 15.8 composites: certification of structures design development with, 2.29–2.30
methods of, 4.7 CFRP. See carbonfiberreinforcedpolymer(CFRP) development for aerospace application, 2.24
tables of techniques and applications, 14.3–14.4 characteristic curve dielectrometry, 17.12, 17.13, 17.14
ultrasonic, 4.7, 9.3, 9.19, 9.21, 9.24, 14.3, 14.7–14.8 crack detection and, 10.27 durability, damage resistance, and damage tolerance
bonded assemblies, 2.32–2.35. See also bond strength; film, 10.9, 10.10 of structures, 2.30–2.31
bond testing; honeycomb structures; laminates, phosphor, 10.12 eddy current testing, 8.28
composite; sandwich structures charge coupled devices, 10.9, 10.13, 10.14. See also fastened structures, 4.6
advantages and limitations, 2.32 digital detector arrays fatigue properties, 2.27
certification for airworthiness, 2.34–2.35 chromatic aberration, 5.5 fluid exposures, 2.26–2.27, 2.28, 2.29, 17.9
common anomalies, 2.34 chromatic contrast, 5.5–5.6 fourier transform infrared, 17.20–17.21
repair schemes for, 2.35 Civil Air Regulations (CARs), 2.13 heat damage, 17.20–17.21
borescope applications civil aviation, categories of, 2.3 holographic testing, 12.20
airframe, 5.16–5.20 cladding, metal. See also coatings honeycomb sandwich fuselage panels, 13.14–13.16
engine, 5.20–5.21 conductivity measurement and, 8.33 inspection issues, 4.3–4.4
borescopes, 5.8–5.11 thickness measurement, 8.34 matrix materials for, 2.24, 2.25
what can go wrong, 5.16 cleaning mechanical properties, 2.26–2.27
boron epoxy skin, 10.40, 10.41 of bolt hole, 8.46 modified failure modes, 2.10
boron fibers, 2.25 for magnetic particle testing, 7.14, 7.15 overview of materials, 2.24–2.26
bragg equation, 17.15, 18.6 for penetrant testing, 6.6–6.7, 6.11, 6.13, 6.15 penetrants causing harm to, 6.9
bremsstrahlung X-ray sources, 10.5–10.6 for visual testing, 5.8, 5.16, 5.17 reinforcement fibers for, 2.25–2.26
bubble leaktesting, 16.4–16.5, 16.10, 16.11 for visual testing of composites, 5.21 repair of, shearogram, 12.24, 12.25
buckling, 2.11 cluster kernel pixels, 10.25, 10.26 representative properties, 2.11
building block approach to certification coating thickness measurements resin transfer molded, holography, 12.20
of bonded structures, 2.34–2.35 beta backscatter method, 17.22–17.23 stiffened structures using, 2.27
of composite structures, 2.29–2.30 eddycurrentmethod, 8.27, 8.29, 8.33–8.34, 17.22 structural health monitoring of, 15.5–15.6
buildup factor, 10.4, 18.4 fourier transform infrared method, 17.20, 17.21 surface contamination of, 17.21
butt line, 2.4 overview of methods, 17.22 temperature effects on, 2.26, 2.29, 2.30
coatings. See also cladding, metal; plating temperature effects on conductivity, 8.33
characterization of, 8.20 testing methods, 4.3–4.4
C corrosion hidden under, 8.20
dielectrometry of, 17.12, 17.14
visual testing, 5.21–5.23
composites, radiographic testing, 10.41–10.42
C-5 aircraft, 3.3 impedance plane plot and, 8.6 with neutrons, 10.40, 10.41
C-130 Hercules, spar cap, ultrasonic inspection, 9.35 inspection issues, 1.4, 4.8 composites, shearographic testing
cable wrap method, 7.5 magnetic particle testing and, 7.14 overwrap pressure vessel, 12.6, 12.7, 12.14, 12.16,
candela (cd), 5.5 meandering winding magnetometer and, 8.19, 8.20 12.24, 12.26
capacitance polymers on metal, 17.21 specifications and standards, 12.16–12.18
of cable, 8.5 surface contamination and, 17.21 stressing methods, 12.7, 12.24–12.27
of parallel plate sensor, 17.13, 18.6 testing methods, 4.8 composites, thermographic testing, 11.4, 11.9, 11.10
capacitance sensing, 17.12–17.14 ultrasonic testing, 9.3, 9.8, 17.22 ASTM standard, 11.16
capillary action, 6.3, 6.6, 18.2 cobonded structure quality assurance in manufacturing, 11.18–11.20
carbon fiber, 2.25–2.26 laminate, 2.32 trappedwaterinsandwichstructures, 11.9–11.10,
carbon fiber face sheets, 12.24, 12.25 stiffening, 2.27–2.28 11.17–11.18, 11.19
carbon fiber reinforced polymer (CFRP), 2.26. See also cocured structure composites, ultrasonic testing, 9.2–9.3
fiber reinforced polymer (FRP) composite laminate closed cell foam core, 2.34 air coupled, 9.21
on aluminum honeycomb, microwave testing, 17.10 laminate, 2.32 of brass inclusion, 9.8
on aluminum honeycomb, thermography, 11.13 stiffening, 2.27–2.28 complex contours and, 9.34
honeycomb panel, tap testing, 14.6 coercivity, 7.3, 7.4 laser scanning, 9.24
shearographic testing, 12.24, 12.25 cole-cole plots, 17.13 liquid surface holography, 9.27
thermal response, 11.3, 11.4, 11.5 color, 5.5, 5.6, 5.7 as method of choice, 9.30
thermographic foreign object detection, 11.20 commercial aircraft, safety factors in design, 2.8 phased array, 9.36
visual testing, 5.21–5.23 commercial scheduled air transport, 2.3 to validate consolidation, 9.33
carbon/carbon laminates, ultrasonic testing, 9.3 comparative vacuum monitoring, 15.4 compound lenses, 5.5
carbon/epoxy laminates, 2.26, 2.27 complementary metal oxide silicon technology, compression failure, 2.11
castings 10.13,10.14 compton scattering, 10.2, 10.3
eddy current testing, 8.28 complex geometries, 1.4 computed radiologic testing, 10.11–10.12
inspection issues, 4.5 Composite Materials Handbook 17, 2.28–2.29 advantages and limitations, 10.38
penetrant testing materials, 6.6 composite overwrap pressure vessel, 12.6, 12.7, 12.14, artifacts in, 10.24–10.25
testing methods, 4.5 12.16, 12.24, 12.26 image processing and interpretation, 10.31–10.32
thick sections of, 1.4 composite ultrasonic transducers, 9.10, 9.11 standards for, 10.36
ultrasonic testing, 9.3
vibrothermography, 11.15

20.2 AEROSPACE NONDESTRUCTIVE TESTING

computed tomography, 10.15–10.18 corrosion, radiologic testing, 10.41 creep, 2.11
advantages and limitations, 10.38 backscatter imaging, 10.34 creeping (creep) waves, 9.13, 9.34
of composite structures, 10.41–10.42 with neutrons, 10.35, 10.40 critical angles of ultrasonic wave, 9.8–9.9
hounsfield units in, 10.15–10.16, 18.5 corrosion, ultrasonic testing, 9.33 surface waves and, 9.24
of solidrocket motor, 2.7 spectroscopic, 9.24 critical damage threshold, 2.30, 2.31
standard for, 10.36 costs critical design review, 2.9
conductance, 17.13, 18.6 design life and, 2.21 cross reference tables, 4.1–4.19
conductive layers, 4.8 inspection intervals and, 1.3–1.4 introduction to, 4.2
conductivity of structural weight, 1.3 materials, structures, and processes, 4.3–4.9
defined, 8.7 of toughened resin material system, 2.30 testing methods, 4.10–4.19
in dielectrometry, 17.13 value of nondestructive testing, 1.5, 2.8 cross section, for X-ray attenuation, 10.3, 18.4
factors affecting, 8.7–8.8 couplingcoefficient, of ultrasonictransducers, 9.10, 9.11 crosswound eddy current probe, 8.14
conductivity, andeddycurrenttesting, 8.6–8.8 crack growth. Seealso fracture mechanics cure state monitoring, with dielectrometry, 17.12, 17.14
with low conductivity materials, 8.29 acoustic emissions from, 13.2, 13.13–13.14, 15.7 curie temperature, 7.4
with meandering winding magnetometer, 8.18, damage tolerant design and, 2.18–2.20, current waveform, in magnetic particle testing, 7.8, 7.9
8.19, 8.20 3.2–3.3, 15.7 curved surfaces. See also contoured surfaces
orthogonal field probes and, 8.14 fail safe approach and, 2.18 conductivity measurements and, 8.33
skin depth and, 8.10–8.11 meandering winding magnetometer and, 8.19 radiologic testing of, 10.22
conductivity measurement, eddy current method, 8.8, safe life design and, 2.17 ultrasonic inspection sensitivity and, 9.32
8.27, 8.30–8.33 steel in F-111 accident and, 2.15 ultrasonic surface waves on, 9.24
alloying elements affecting, 8.31, 8.32 crack initiation
of aluminum alloys, 8.30–8.32, 8.33 acoustic emissions from, 13.13
factors affecting, 8.33
heat treatment and, 8.32
in-situ monitoring for, 15.4, 15.7
part life monitoring for, 17.26
D
mechanical stresses and, 8.32 safe life design and, 2.17, 2.18 D x T effect, 10.2, 10.9
reference standards for, 8.30–8.31 by thermal damage to composite, 17.20 damage mapping, with meandering winding
thickness measurement combined with, 8.35 crack size. See also discontinuity size magnetometer, 8.20
conductivity meters, 8.14, 8.30 advances in fracture mechanics and, 1.6 damage resistance
contact transducers, ultrasonic, 1.5 damage tolerant design and, 2.18–2.19 of composite parts, 2.30
contoured composite parts, ultrasonic testing, 9.34 high frequency eddy current tests and, 8.30 defined, 2.30
laser techniques, 9.24 in-situ monitoring of, 15.7 damage tolerance, 1.6
contoured surfaces. See also curved surfaces magnetic particle testing and, 7.14 of composite parts, 2.30–2.31
eddy current testing, 8.29 magneto-optic imaging and, 8.22 defined, 2.30
effect on testing, 1.4 nondestructive testing signal magnitude and, 3.4 damage tolerant design, 2.18–2.23
contrast, in microwave interferometry, 17.7 penetrant sensitivity and, 6.5 aging aircraft and, 2.21–2.22
contrast, radiologic, 10.19–10.20 probability of detection and, 3.4 fail safe approach and, 2.21
crack detection and, 10.27, 10.28 cracks, acoustic emissiontesting, 13.7, 13.13–13.14, 13.15 institution of modern programs for, 2.20–2.21
formula for, 18.5 cracks, eddy current testing, 8.27–8.28, 8.30. Seealso preexisting damage and, 2.18–2.19
image quality indicator and, 10.29 fatigue cracks, eddy current testing probability of detection and, 3.2–3.3
in inspection of castings, 10.37 corrosion and, 8.37 safe life design and, 2.17
options for improving, 10.39 impedance plane analysis, 8.6 slow crack growth approach and, 2.19–2.20, 2.21
scatter and, 10.23 orthogonal field probes, 8.14 structural health monitoring and, 15.7
contrast, visual, 5.5–5.6, 5.7 subsurface crack detection, 8.38–8.40 three damage categories in, 2.30–2.31
illumination and, 5.16 subsurface cracks in window belt splice, 8.40 U.S. Air Force requirements, 2.15, 2.17, 2.20–2.21
viewing angle and, 5.16 subsurface doubler crack, coil arrays, 8.16 damaged structures
contrast sensitivity gage, 10.30 surface crack detection, 3.9, 8.37–8.38 inspection issues, 4.9
coordinate systems, for aircraft structures, 2.4 testing methods, 4.9
cracks, fatigue. See fatigue cracks
copper cracks, liquid penetrant methods. See liquid De Havilland Comet, 2.14
conductivity, effect of impurities, 8.31 penetrant testing decibel (dB), 9.6, 18.3
impedance plane plot, 8.6 cracks, magnetic particle testing, 7.2, 7.14 deformation. See strain
corrosion. Seealso stresscorrosioncracking cracks, magneto-optic imaging, 8.21–8.23 delaminations. See also bond testing; disbonds
diffracted light technique, 5.11–5.12 cracks, magnetoresistive sensing, 8.23, 8.24 acoustic emissions testing, 13.2, 13.7, 13.14–13.15
forms of, 2.12 cracks, meandering winding magnetometer, 8.19, 8.20 in engineered ceramics, 17.10
hidden, 8.20, 8.36 cracks, radiologic detection, 10.41 fourier transform infrared method, 17.20
in honeycomb structures, 10.40 image distortion and, 10.22 microwave testing, 17.7, 17.8, 17.10
landing gear pitting corrosion, 5.17–5.18 orientation and, 10.27–10.28 partial vacuum shearography, 12.26
magneto-optic imaging, 8.22, 8.23 cracks, ultrasonic testing, 9.7, 9.8, 9.33, 9.34–9.36 thermographic testing, 11.9, 11.19
meandering winding magnetometer testing, 8.20 around fastener holes, 9.34 ultrasonic testing, 9.8, 9.18, 9.33
microwave testing, 17.10–17.11 Delta IV rocket, liquid fuel engine for, 2.7
diffraction and, 9.9
moisture in developers and, 6.9 at lap and splice joints, 9.34, 9.35–9.36 demagnetization, 7.4, 7.5, 7.14, 7.15
overview, 2.12 nonlinearity in, 9.26 density of image. See film density
under paint, 17.10–17.11 with surface waves, 9.24 densityofmaterials
thinning by, 2.12, 8.36–8.37 in vertical risers, 9.34 acoustic properties and, 9.5
widespread fatigue damage and, 2.22 cracks, vibrothermographic detection, 11.14–11.15 microwave interference measurement, 17.8–17.9
corrosion, eddy current testing, 8.27–8.28, 8.36– cracks, visual inspection dents. See also impact damage
8.37, 8.38 in airframe, 5.16–5.17 optical profiling of, 5.12–5.13
multiple-frequency, 8.14–8.15 in composites, 5.21–5.23

INDEX 20.3
Departmentof Defense (DOD), 2.3. Seealso military discontinuities. See anomalies; corrosion; crack entries; eddy current testing, aerospace applications, 8.27–8.47.
aircraft; United States Air Force (USAF) delaminations; disbonds; impact damage, to composite See also cracks, eddy current testing; discontinuity
Composite Materials Handbook 17, 2.28–2.29 components; inclusions; porosity; probability of size, in eddy current testing; eddy current testing;
Joint Services Specification Guide (JSSG-2006), 2.15, detection (POD); voids fatiguecracks, eddycurrenttesting; impedance plane
2.20–2.21 discontinuity size. See also crack size analysis
depth of field, 5.8 growth over time, 3.3 advantages and limitations, 8.27, 8.28
depthof penetration, ofeddycurrent, 8.10–8.11, 8.28, inspection intervals and, 1.3–1.4, 3.3 automatic, 3.9, 8.40, 8.42–8.43, 8.46, 8.47
8.29, 8.33 in magneto-optic imaging, 8.22–8.23 bolt hole inspection, 8.29, 8.30, 8.40–8.46
formula for, 18.3 probabilityofdetectionand, 2.8 coating thickness measurements, 8.27, 8.29,
frequency selection and, 8.43 in shearographic testing, 12.9, 12.10–12.11 8.33–8.34, 17.22
in magneto-optic imaging, 8.22–8.23 structural design philosophy and, 2.16 conductivity measurements, 8.8, 8.27,
with meandering winding magnetometer, 8.18–8.19 in thermographic testing, 11.6, 11.8, 11.11 8.30–8.33, 8.35
depth of penetration, of electric field in in ultrasonic testing, 9.18, 9.34 corrosion inspection, 8.14–8.15, 8.27–8.28,
dielectrometry, 17.14 weight restrictions and, 1.3, 1.4 8.36–8.37, 8.38
depth of penetration, of microwaves, 17.6 discontinuity size, in eddy current testing discontinuities detected, 8.27
depth of surface anomalies, visual measurement, equivalent reference data and, 8.16 engines and engine components, 3.9, 8.12, 8.27,
5.12–5.13, 5.16 pulsed current, 8.15 8.35–8.36, 8.46–8.47
design life, 2.21 required method if <1.8 mm long, 8.46 jet engine inspections of opportunity, 2.21
design of aircraft. See also damage tolerant design with split core coils, 8.12–8.13 layer thickness measurements, 8.34–8.36
phases of, 2.8–2.9 with wide field probes, 8.13–8.14 magnetoresistive sensing compared to, 8.24
philosophies of, 2.16–2.17 dispersed discontinuities, in light alloy castings, 10.37 for structural health monitoring, 15.4, 15.5
systems design, 2.9, 2.10 dispersion of waves, 9.25 summary of, 8.27, 8.28
destructive testing, in design process, 2.9 DOD. See Department of Defense eddy current testing, nonconventional, 8.18–8.24
developers, in penetrant process, 6.3–6.4, 6.6 doubler magneto-optic imagers, 8.21–8.23
application of, 6.7 magnetostrictive monitoring, 15.8 magnetoresistive sensing, 8.23–8.24
in automated system, 6.13, 6.15 subsurface cracks at edge, 8.16 meandering winding magnetometer, 8.18–8.20,
corrosion promoted by residue of, 6.9 titanium, corrosion thinning and, 8.37 15.11–15.13
dwell time, 6.8 drivetrain damage classification, 13.13–13.14 edge effects, in eddy current testing, 8.11, 8.12,
what can go wrong, 6.12 dual-frequency eddy current testing, 8.15 8.29, 8.33
diamagnetic materials, 7.2, 7.3 durability, of composite parts, 2.30 of engine components, 8.46
dielectric constant dwell time edge-of-light technique, 5.11–5.12
microwave interferometry and, 17.6–17.7, 17.9 developer, 6.8 effusivity contrast, 11.5–11.6, 11.11
moisture detection and, 17.9 penetrant, 6.4, 6.7, 6.11, 6.13, 6.14, 6.15 elastic deformation, 2.10
of ultrasonic transducers, 9.10, 9.11 dynamic range of imaging system, 10.12, 10.15, 10.19, elastic modulus, 2.10–2.11
dielectric material properties, 17.13 10.20, 10.29 acoustic velocity and, 9.5, 18.3
dielectrometry, 17.12–17.14 measured by resonant ultrasound
differential bridge mode, 8.12, 8.14 spectroscopy, 17.25
for bolt holes, 8.41
differential reflection probe, 8.12, 8.14
E electrical conductivity. See conductivity
electromagnetic acoustic transducers, 9.22
diffracted light technique, 5.11–5.12 E glass (electrical glass), 2.25 electromagneticcoupling, betweencoilandtestobject,
diffraction economics. See costs 8.11, 8.12
of ultrasonic waves, 9.9 eddy current, 8.2–8.3 electromagnetic induction, 8.2–8.3, 8.4, 18.3
of X-rays in crystalline materials, 10.24, 17.15-17.17 eddycurrent testing, 8.4–8.17. Seealso eddycurrent electromagnetic spectrum, 5.3
X-ray, method, 17.15-17.17 testing, aerospace applications; frequency, of eddy electromagnetic testing, 8.1–8.47. See also barkhausen
diffuse reflection, 5.3–5.4 current probe; impedance plane analysis; liftoff, in noise measurements; eddy current testing; microwave
digital detector arrays, 10.13–10.15 eddy current testing testing
advantages and limitations, 10.38 basic physics of, 8.2–8.3, 8.4, 8.28 formulas for, 18.3
bad pixel management, 10.25–10.27 coil arrays, 8.15–8.16 nonconventional methods, 8.18–8.24
in honeycomb inspection, 10.40 coil designs, 8.12–8.14 principles, 8.2–8.3
image processing and interpretation, 10.31–10.32 cross reference table, 4.11, 4.13 standards and specifications, 8.25–8.26
standards for manufacturing and use, 10.36 depth of penetration, 8.10–8.11, 8.28, 8.29, 8.33, for structural health monitoring, 15.4
in weld inspection, 10.37, 10.39 8.43, 18.3 electromagnetic theory, 8.2–8.3
digital filtering, 8.16 discontinuities detected, 8.27 electron linear accelerators, 10.5–10.6, 10.7
digital image correlation photogrammetry, 5.13 edge effects, 8.11, 8.12, 8.29, 8.33, 8.46 electronic scanning (E-scan), 9.20
digital radiograph factors affecting response, 8.7–8.12 electronics. See also avionics
image quality, 10.15 filters in, 8.7, 8.29, 8.43–8.44 digital holography of, 12.20
ofturbine blade, 10.17 frequencyrangeof, 8.10 inspection issues, 4.9
digital reference images, 10.36 geometric parameters in, 8.11–8.12 radiologic testing of, 10.42
digital video imagers, 5.8 historical development of, 1.5 testing methods, 4.9
diopter, 5.5 introduction to, 8.4 electrostatic liquid penetrant spray technique, 6.11,
disbonds, 14.2. Seealso bondtesting; bonded multiple-frequency, 8.14–8.15 6.13, 6.15
assemblies; delaminations nonconventional, 8.18–8.24 emulsification, 6.3, 6.4
acousticemissiontesting, 13.14–13.15, 14.3 pulsed, 8.14, 8.15 excessive time of, 6.12
Aloha Airlines incident and, 2.22 remote field, 8.16–8.17 emulsifier, 6.5
holographic testing, 12.28 signal processing in, 8.16 endurance limit, 2.16
shearographic testing, 12.5, 12.12, 12.15, 12.16– standards for, 8.25–8.26 energy of light, 5.3, 18.2
12.18, 12.21, 12.22, 12.24–12.26, 12.27 what can go wrong, 8.29–8.30 engineering manufacturing development, 2.9
ultrasonic testing, 9.19, 9.21, 9.33, 9.36

20.4 AEROSPACE NONDESTRUCTIVE TESTING

engines, 2.6–2.7. See also jet engines fatigue damage, barkhausen noise measurements, 17.4 fixed wing aircraft, 2.3
inspection issues, 4.8 fatigue life, 2.17, 2.22–2.23 flap carriage, magnetic particle testing, 7.16
safe life design, 2.21 fatigue loading, 2.9, 2.10, 2.14 flap track, magnetic particle testing, 7.14
testing methods, 4.8 fatigue of inspector, 3.9 flash radiologic testing, 10.34, 10.38
visual testing, 5.20–5.21 eye fatigue, 5.6 flash thermography, 11.9
engines, eddy current testing, 8.27, 8.46–8.47 posture and, 5.16 ASTM standard, 11.16
with meandering winding magnetometer, 8.20 fatigue resistance principal component analysis in, 11.14
nose cowl inlet ducts, gap evaluation, 8.35–8.36 of 2000 series aluminum alloys, 2.14 signal reconstruction in, 11.18, 11.19
split core coil for, 8.12 measurement of, 2.16 flat coil magnetization, 7.6
environment of inspector, 3.8–3.9 fatigue testing, in aircraft development, 2.9 flex coils, 8.13–8.14
erosion, and widespread fatigue damage, 2.22–2.23 for damage tolerant design, 2.18 FlexCoreTM, 2.33
exfoliation corrosion Federal Aviation Administration (FAA) flexible borescopes, 5.10
eddycurrent testingfor, 8.27, 8.36–8.37, 8.38 aging aircraft and, 2.22 flight readiness review, 2.9
X-ray backscatter imaging of, 10.34 airworthiness certification from, 2.3 flight testing, 2.9
export approval, 2.3 airworthiness standards, 2.13 fluorescent liquid penetrant testing, 6.2, 6.4. See also
exposure factor, 10.8–10.9 certification for composite structure, 2.29 liquidpenetranttesting; ultraviolet lamps; ultraviolet
change in source-to-detector distance, 10.23, 18.5 damage tolerant design and, 2.20, 2.21 (UV-A) radiation
historical background, 1.5, 2.13 historical development of, 1.5
probability of detection estimation and, 3.3 penetrant materials for, 6.5
F uninspectable structures and, 2.20
widespread fatigue damage and, 2.22
fluorescent magnetic particle testing, 7.2, 7.8–7.9 See
also ultraviolet lamps; ultraviolet (UV-A) radiation
F-15 fighter aircraft, 10.40 Federal Aviation Regulations (FARs), 2.13 fluorocarbonresininserts, 12.14, 12.16, 12.17, 12.18
F-111 fighter bomber, 2.15, 2.17, 2.25, 3.3, 15.4 damage tolerance and, 2.20 flying spottechnique, 10.34
f number (focal ratio), 11.8 fatigue effects and, 2.14 foam composite
FAA. See Federal Aviation Administration(FAA) felicity effect, 13.6 inspection issues, 4.4
fail safe design approach, 2.18, 2.20, 2.21 ferromagneticmaterials, 7.2–7.4. Seealso barkhausen testing methods, 4.4
failure modes, structural, 2.10–2.12 noise measurements; magnetic particle testing; steel foam core sandwich structures, 2.34, 14.2
false call rates, 3.5–3.6 ferromagnetic materials, eddy current testing inspection issues, 4.4, 4.7
Faraday’s law of induction, 8.2–8.3 alternatives to, 8.10 testing methods, 4.4, 4.7
FARs. See Federal Aviation Regulations (FARs) impedance of coil and, 8.6–8.7 ultrasonic testing, 9.3, 9.19, 9.21
fastened structures liftoff in, 8.12 foam insulation
inspection issues, 4.6 limited to surface or near surface, 8.10 microwave inspection, 17.9–17.10
magnetoresistive sensing, 8.24 ferrous materials, fatigue curve for, 2.16 radiologic testing, 10.34
testing methods, 4.6 fiber reinforced polymer (FRP) composite laminate. See rocket fuel tank, 12.9
ultrasonic testing, 9.3 also carbon fiber reinforced polymer (CFRP); shearographic testing, 12.9, 12.22
fastener holes, eddy current testing. Seealso bolt holes composites; graphite fiber in epoxy polymer foaming adhesives, 2.33
subsurface cracks, 8.24, 8.38–8.39 eddy current testing and, 8.28 focal length, 5.4–5.5, 18.2
surface cracks in fuselage, 8.37–8.38 inspection issues, 4.3 focal ratio (f number), 11.8
fastener holes, ultrasonic inspection, 9.34 testing methods, 4.3 focal spot size, in radiologic testing, 10.7–10.8, 10.10
fasteners. Seealso rivets ultrasonic testing, 9.2 geometric unsharpness and, 10.21
alloy content testing, 17.19 fiberglass, 2.25. See also glass/epoxy composite as small as considerations allow, 10.23
hydrogen embrittlement of, 2.12 on aluminum honeycomb core, 10.40, 10.41 fog, 10.23, 10.24
fatigue, 2.11 material characteristics, 2.11 foreign objects. See also inclusions
aircraft design and, 2.11, 2.13, 2.14–2.15, 2.16, 2.17 microwave inspection of panel, 17.8 microwave testing, 17.8
nonlinear acoustic effects of, 9.26 fiberglass/epoxy pressure vessels, acoustic radiologic detection, 10.34, 10.42
widespread damage due to, 2.21–2.23 emissions, 13.7 shearographic detection, 12.16, 12.18
fatigue cracks. See also crack entries fibers, reinforcement, for composites, 2.25–2.26 thermography, in composite structures, 11.18–11.20
in aircraft wing spar cap, 5.19–5.20 fiberscopes, 5.10 forging lap, in F-111 accident, 2.15
in aircraft wing under panel, 5.16–5.17 field of view, 5.8 forgings
corrosion leading to, 2.12 fighter/attack aircraft, 2.3, 2.14. See also F-15 fighter eddy current testing, 8.28
detected in part life monitoring, 17.26 aircraft; F-111 fighter bomber inspection issues, 4.6
in F-111 accident, 2.15 fill factor, in magnetic particle testing, 7.5 penetrant testing materials for, 6.6
samples forprobabilityof detectionstudy, 3.11 filleting, adhesive, on sandwich structures, 2.33 radiologic crack detection, 10.41
at stress risers, 2.12 film artifacts, 10.24, 10.31 testing methods, 4.6
widespread fatigue damage due to, 2.21–2.22 film density, 10.9, 10.10, 18.5. Seealso characteristic thick sections of, 1.4
fatigue cracks, eddy current testing, 8.7, 8.27. Seealso curve ultrasonic testing, 9.3, 9.31–9.32, 9.33
cracks, eddy current testing contrast sensitivity and, 10.29 vibrothermography, 11.14, 11.15
boltholeswithfastenerremoved, 8.40–8.46 crack detection and, 10.27, 10.28 fourier transform infrared testing, 17.20–17.21
liquid penetrant testing, 6.2, 6.11 film radiologic testing, 10.9, 10.10 fracture mechanics, 1.6, 2.18, 3.3
magnetic particle testing, 7.7, 7.14 advantages and limitations, 10.38 military aircraft engines and, 8.46
during maintenance overhaul, 8.28 dark adaptation for, 10.31 reliability of eddy current testing and, 8.28
with meandering winding magnetometer, 8.19, interpretation, 10.31 fractures
8.20, 15.12 filtering, digital, 8.16 fatigue induced, 2.11
probe footprint and, 8.12 filters, in eddy current testing, 8.7, 8.29, 8.43–8.44 stress corrosion fracture, of landing gear truck
ultrasonic testing, 15.8 finish systems, and widespread fatigue damage, beam, 5.17–5.18
window belt splice, 8.40 2.22–2.23 frequency, in magneto-optic imaging, 8.22–8.23
fatigue curve, 2.16 fire damage, eddy current testing for, 8.28, 8.32
of carbon/epoxy, 2.27

INDEX 20.5
frequency, of eddy current probe, 8.10–8.11 grinding burns, barkhausen noise analysis of, 17.4 microwave testing, 17.8
for bolt hole inspection, 8.43 groundtesting, 2.9 moisture issues with, 2.33, 10.35, 10.40, 11.9,
low frequency, for corrosion inspection, 8.36–8.37 guided (lamb) waves, 9.13, 9.22, 9.25–9.26 11.11, 11.17–11.18
low frequency, subsurface crack detection, in structural health monitoring, 15.4, 15.7–15.8 radiologic testing, 10.20, 10.40, 10.41
8.38–8.39 shearographic testing, 12.5, 12.22, 12.23
with multiple-frequency instruments, 8.14–8.15 tap testing, 14.5–14–6, 14.7
with orthogonal field probe, 8.14
frequency, of electromagnetic radiation, 5.3, 18.2
H testing methods, 4.4
thermographic testing, 11.9, 11.10, 11.11, 11.12,
frequency, of ultrasonic waves, 9.2, 9.4, 9.5, 9.6 H and D curve. See characteristic curve 11.13, 11.17–11.18
frequency discrimination, 8.16 half-value layer (HVL), 10.4, 18.4 ultrasonic testing, 9.3, 9.21
frequency domain multiplexing, 8.14 hall effect tesla meters, 7.6, 7.7, 7.11 wing tip, thickness measurement, 8.35
fretting wear, 2.12 halon fire extinguisher bottles, 13.12–13.13 Hooke’s law, 9.26
FRP. See fiber reinforced polymer (FRP) composite hardened steels, barkhausen noise analysis, 17.4 hounsfield units, 10.15–10.16, 18.5
laminate hardness, barkhausen noise intensity and, 17.4 human factors, 3.5, 3.8–3.10
FTIR. See fourier transform infrared testing heat capacity, 11.3, 11.4 HVL (half-value layer), 10.4, 18.4
fuel tanks, aircraft. See also leak testing heat conduction, 11.3–11.4 hydrogen embrittlement, 2.12
chasing leaks in, 16.10–16.11 heatdamage, to resinofcomposite, 17.20–17.21 by chemical removal of finish, 2.23
“no visible leakage” requirement, 16.2 heatpropagationtime, 18.5 high strength materials and, 2.12, 2.23
full scale development, 2.9 heat treatment. See also temperature hysteresis loop, 7.2–7.4
fuselage station, 2.4 of alloys, 8.32
fuses, X-ray backscatter imaging of, 10.34 barkhausen signals and, 17.4
conductivity and, 8.8, 8.27, 8.32
future usefulness, 1.2
microstructural damage from, 17.4 I
retained austenite measurements and, 17.16 IACS (International Annealed Copper Standard), 8.7–8.8,
types of, 8.32 8.30, 8.31
G heel effect, 10.7 ice, thermographic detection of, 11.9
gamma radiation, 10.2. See also radiologic testing helicopter, as general aviation aircraft, 2.3 illuminance (lux), 5.5, 5.6, 5.7
digital detectors for, 10.13 helicopter blades illumination, 5.5–5.7, 5.8
gamma ray sources, 10.6 production shearography, 12.23 image intensifiers, X-ray, 10.12, 10.13
exposure factor with, 10.8–10.9 water trapped in, 11.10, 11.18 image processing techniques, radiologic, 10.31–10.32
focal spot size, 10.7 helicoptercomponents, safelifedesign, 2.17 image quality
imaging with, 10.9 helicopter drivetrain, acoustic emission testing, apparent, 5.16
in phosphor imaging, 10.12 13.13, 13.14 digital radiologic, 10.15
gaps helicopter engines, fracture mechanics analysis, 8.46 radiologic, 10.19–10.28
between metal layers, eddy current measurement, helium leak test, 16.8–16.11 image quality indicators, radiologic, 10.29–10.30, 10.36
8.35–8.36 hexagonal core, 2.33. See also honeycomb core image unsharpness, 10.21–10.22, 18.5
transmission or reflection of ultrasound, 9.8 high frequency bond testing, 14.7–14.8. See also immersed transducers, ultrasonic, 9.5
gas turbine engines, 2.6–2.7. See also jet engines resonance ultrasonic testing impact damage, to composite components, 2.30–2.31
general aviation, 2.3 high pass filter, in eddy current testing, 8.7, 8.44 acoustic emissions testing, 13.14–13.15
internal combustion engines in, 2.6 high power ultrasound testing, 14.4 shearographic testing, 12.5, 12.24, 12.25
geometric image unsharpness, 10.21, 18.5. See also high strength steels thermographic testing, 11.9
penumbral shadow of F-111 fighter bomber, 2.15 ultrasonic resonance testing, 9.19
geometric magnification, 10.20–10.21. See also hydrogen embrittlement of, 2.12 visual testing, 2.30, 5.22, 5.23
magnification of landing gear components, 2.17 impact damage, to honeycomb sandwich structures, 14.2
bad pixels and, 10.25, 10.26 representative properties, 2.11 impact damage, to space shuttle wing, 15.9–15.10
in electronics testing, 10.42 holographic testing impact damage, visual testing
formula for, 18.5 advantages and limitations, 12.20 composite components, 2.30, 5.22, 5.23
in weld inspection, 10.39 applications, 12.2, 12.28 diffracted light technique, 5.11–5.12
giant magnetoresistive sensors, 8.23–8.24 of bonds, 14.4, 14.8 viewing distance, 5.8
glare, 5.16 cross reference table, 4.11, 4.17 impedance, acoustic. See acoustic impedance
glass fibers, 2.25. See also fiberglass image interpretation and analysis, 12.9 impedance analysis, mechanical, 14.5, 14.6
glass/epoxy composite introduction to, 12.2, 12.6 impedance of dielectric, 17.13
inspection issues, 4.3 laser safety in, 12.29–12.30 impedance plane analysis, 8.4–8.7
microwave thickness measurement, 17.8–17.9 specifications and standards, 12.16–12.18 bolt hole cracks, 8.42–8.43, 8.45–8.46
pressure vessels, acoustic emissions, 13.7 holography, acoustic, 9.26–9.27 bondedhoneycombwingtipinspection,8.35
testing methods, 4.3 honeycomb core, 2.33–2.34 conductivity loci in, 8.8, 8.9
gliders, 2.3, 2.6 closure configurations for, 2.34 discontinuities in, 8.4, 8.6–8.7
graphite fiber, 2.26 common anomalies in, 2.34 liftoff loci in, 8.9, 8.12, 8.34
graphite fiber inepoxy polymer. Seealso fiber reinforced thin gage, 2.34 withorthogonalfieldprobe, 8.14
polymer (FRP) composite laminate honeycomb structures, 14.2. See also aluminum spacing between metal sheets, 8.35, 8.36
on aluminum honeycomb core, 10.40, 10.41, 11.12 honeycomb; sandwich structures subsurface crack detection, 8.38
inspection issues, 4.3 acoustic emission testing, 13.14–13.16 surface crack detection, 8.37–8.38
laser ultrasonic testing, 9.22, 9.24 air coupled ultrasonic testing, 9.21 impulseresonancevibrationtesting, 17.24, 17.25
material characteristics, 2.11 composite fuselage panels, 13.14–13.16 inclusions, barkhausen noise and, 17.4
testing methods, 4.3 impact damage, 14.2
grid, potter bucky, 10.24 inspection issues, 4.4

20.6 AEROSPACE NONDESTRUCTIVE TESTING

inclusions, ultrasonic detection, 9.7, 9.8. See also level 4 penetrant used for, 6.5 leak testing techniques, 16.3
foreign objects for missile propulsion, 2.7 acoustic, 15.10–15.11, 16.5–16.6
acoustic holography, 9.27 testing methods, 4.8 ammonia/phenolphthalein, 16.4
in composites, 9.33 Joint Services Specification Guide (JSSG-2006), 2.15, bubble testing, 16.4–16.5, 16.10, 16.11
spectroscopic, 9.24 2.20–2.21 helium, 16.8–16.11
index of refraction, 5.4, 18.2 pressure or vacuum change, 16.6–16.8
wavelength and, 5.5 tracer dye, 16.8
inductance
for air, 8.12
K lenses, 5.4–5.5, 18.2
life monitoring, 17.26
of a coil, 8.3, 8.5 K edge energy levels, 10.3 liftoff, in eddy current testing
induction, electromagnetic, 8.2–8.3, 8.4 kaiser effect, 13.6 coating thickness and, 8.20, 8.27, 8.34
inductive reactance, 8.5, 8.28 KDS(knowndiscontinuitystandards) panels, 6.10, 6.12 defined, 8.12
inelastic deformation, 2.10 ketos ring, 7.11–7.12 effect on readings, 8.11, 8.12, 8.28
infinite life design, 2.16 Kevlar, 2.25, 2.27 in impedance plane analysis, 8.4, 8.6, 8.9, 8.12
infrared cameras, 11.4, 11.7–11.8 kilovoltage peak (kVp) energy, 10.5 with meandering winding magnetometer, 8.19, 8.20
infrared emissivity, 11.7 measurement of, 8.35
infrared spectral regions, 17.20 with orthogonal field probes, 8.14
infrared testing. See thermographic testing L light, 5.3–5.7
in-situ testing. See structural health monitoring light meters, 5.7
lamb waves. See guided (lamb) waves
inspection intervals in magnetic particle testing, 7.11
laminate bonds, 2.32
adjusted, for engine, 5.21 light sources, 5.5, 5.6
laminates, composite. See also delaminations; fiber
crack monitoring and, 15.7 lighting for magnetic particle testing, 7.11
reinforced polymer (FRP) composite laminate
damage tolerance programs and, 2.19, 2.31 lighting for visual testing, 5.5–5.7, 5.8
air coupled ultrasonic testing, 9.21
discontinuity size and, 1.3–1.4, 3.3 of composites, 5.21, 5.22, 5.23
shearographic testing, 12.16–12.17
with fail safe design, 2.20, 2.21 what can go wrong, 5.16
laminography, 10.33, 10.38
inspection starting point, 2.22 lighting materials, 5.5, 5.6
landing gears
inspection systems, and probability of detection, 3.5 limit of validity, of model maintenance program, 2.22
barkhausen noise analysis, 17.4
inspections. See also inspection intervals; nondestructive linear accelerators, 10.5–10.6, 10.7, 10.10
inspection issues, 4.9
testing (NDT) linear array ultrasonic testing, 9.14, 9.15
pitting corrosion, 5.17–5.18
of composite components, 2.30, 2.31 linear attenuation coefficient, 10.2–10.4, 18.4
safe life design, 2.17
design process and, 2.8, 2.9 hounsfield units and, 10.16
testing methods, 4.9
human factors affecting, 3.5, 3.8–3.10 object contrast and, 10.19
wheel flange, magnetic particle testing, 7.16
nondestructive (NDI), 2.8 linear elastic materials, 9.4–9.5
lap joints
inspections, inservice liquid penetrant testing, 6.1–6.15. See also developers,
backscatter imaging of corrosion, 10.34
eddy current, 8.37 in penetrant process
ultrasonic inspection, 9.34
inspection issues, 4.9 advantages and limitations, 6.11
lap splices
scheduled and unscheduled, 2.36 applications, 6.11–6.15
guided wave testing, 9.26
testing methods, 4.9 automated, in aircraft manufacturing, 6.11, 6.13–
scribe line crack, 9.35–9.36
ultrasonic testing in, 9.3 6.15
laser interferometry, 9.24. See also holographic testing;
instructions, written, 3.8, 3.10 automated interpretation of, 6.15
shearographic testing
insulation, 4.8. See also foam insulation cross reference table, 4.10, 4.11
laser profiler method, 5.12
integrally cocured structure, 2.32 introduction to, 6.2
laser safety, 12.29–12.30
interferometry. See holographic testing; laser material options for aerospace applications, 6.6
laser ultrasonic testing, 9.22–9.24
interferometry; microwave interferometry; portable process, 6.5, 6.6, 6.7
with guided waves, 9.25
shearographic testing principles, 6.3–6.4
laser vibrometers, 17.25
internal combustion engines, 2.6 rebleed procedure in, 6.8
lateral dimensional measurement, with acoustic
International Annealed Copper Standard (IACS), 8.7–8.8, of rotating engine components, 6.5
holography, 9.27
8.30, 8.31 safety in, 6.9
latitude, radiologic
International Institute of Welding(IIW), ultrasound standards and specifications, 6.10
of film, 10.10
calibrationblock, 9.18, 9.19 techniques, 6.5–6.8
in inspection of castings, 10.37
International Space Station, 2.4, 2.36 visible (dye) method, 6.2, 6.6, 6.12
options for improving, 10.39
International Traffic in Arms Regulations, 2.3 visual testing step in, 6.5–6.6
of phosphor, 10.12
interpersonal issues, 3.9 what can go wrong, 6.11–6.12
laue pattern, 17.17
inverse square formula, 18.4 liquid rocket engines, 2.7
layer thickness, eddy current measurements, 8.34–8.36.
isotopes for radiologic testing, 10.6, 10.7 liquid surface holography, 9.27
See also thickness
loads, in design process, 2.9, 2.14, 2.16
layered materials, ultrasonic spectroscopy, 9.24
Lockheed C-141 Starlifter, wing cracks, 9.34
lead foils, 10.23
J lead screens
lock-in thermography, 11.13
longitudinal (compression) ultrasonic waves, 9.4–
jetaircraft, corrosion, eddycurrenttestingfor, 8.36 for film imaging, 10.10
9.5, 9.13
jetengine disks, automatic robotic inspection, 3.9 for phosphor imaging, 10.12
polarized, 9.22
jetengines, 2.6–2.7. Seealso engines; turbine blades; lead sheet, 10.24
refracted, 9.8–9.9
turbine engine disks leak testing, 16.1–16.11
spectroscopy with, 9.24
damage tolerance approach, 2.21 advantages of, 16.2
lorentz force, 9.22
high reliability required of, 1.4 aircraft fuel tanks, 16.2, 16.10–16.11
loupe, 5.8
inspection issues, 4.8 allowable leakage rate defined for, 16.2
low frequency bond testing, 14.3, 14.5, 14.6–14.7
inspections of opportunity, 2.21 spacecraft, 15.10–15.11

INDEX 20.7
low pass filter, in eddy current testing, 8.7, 8.44 management, 3.9 moisture effects. See also water, thermographic
lumens (lm), 5.5 marcelling (wrinkling), 17.8 detection of
luminance, 5.5, 5.6 mass attenuation coefficient, 10.2, 10.3, 18.4 on composite materials, 2.26–2.27, 2.29, 2.30,
luminance contrast, 5.5, 5.6 for neutrons, 10.34–10.35 10.40, 17.9
luminance ratios, 5.6 material characterization, nondestructive, 1.2 corrosion inspection and, 8.36–8.37
luminous flux, 5.5, 5.6 materials with foam core, 2.34
luminous intensity, 5.5, 5.6 continual new development of, 1.6 with honeycomb core, 2.33, 10.35, 10.40, 11.9,
lux (lx), 5.5, 5.6, 5.7 cross reference tables, 4.3–4.9 11.11, 11.17–11.18
representative characteristics of, 2.11 motors. Seealso engines; rocketmotors
thermophysical properties, 11.4 of subsystems, 4.9
M ultrasonic test applications, 9.2–9.3
matrix materials, of composites, 2.24, 2.25
multilayer joints, ultrasonic inspection for cracks, 9.34
multiplexing
machined parts meandering winding magnetometer, 8.18–8.20 in eddy current testing, 8.14, 8.15
inspection issues, 4.6 for structural health monitoring, 15.11–15.13 in structural health monitoring, 15.6
testing methods, 4.6 mechanical impedance analysis, 14.5, 14.6
ultrasonic testing, 9.3 method tables, 4.10–4.19
machining cracks, liquid penetrant testing for, 6.2, 6.11
magnesium
microbolometers, 11.7, 11.8
microcracking, nonlinear acoustic effects of, 9.26
N
castings, 4.5 microfocus radiologic testing National Aeronautics and Space Administration (NASA).
machined parts, 4.6 advantages and limitations, 10.38 See also space shuttles
for World War II aircraft, 1.5 of electronics, 10.42 damage tolerance approach, 3.3
X-ray transmission, 10.19, 10.20 turbine blade crack detection, 10.41 published guidelines, 3.5
magnesium alloys, castings, radiologic testing, 10.37 microfocus X-ray tubes, 10.7–10.8, 10.10, 10.13 rocket launch vehicles, 2.5
magnetic coupling, between coil and test object, microscope photography, 5.22 National Transportation SafetyBoard(NTSB),2.13
8.11, 8.12 microscopes, 5.8 Aloha Airlines Flight 243 and, 2.22
magnetic field microstrain, 2.10 NDC. See nondestructive characterization
eddy currents and, 8.2–8.3 microstructural damage, barkhausen noise analysis NDE. See nondestructive evaluation; nondestructive
in object for magnetic particle testing, 7.4–7.8 of, 17.4 examination
magnetic field intensity, 7.2–7.4, 7.6–7.7, 18.2 microwave imaging, 17.6 NDI. See nondestructive inspection
magnetic flux density, 7.2–7.4, 18.2 microwave interferometry, 17.6–17.7, 17.8–17.9, 17.10 NDT. See nondestructive testing
magnetic leakage field, 7.2, 7.4, 7.6, 7.11 microwave spectroscopy, time based, 17.6 neper, 9.6
magnetic materials, 7.2–7.4. See also magnetic microwave testing, 17.5–17.11 neutron radiography, 10.34–10.35
permeability advantages and limitations, 17.7 advantages and limitations, 10.38
magnetic particle testing, 7.1–7.16 applications, 17.7–17.11 of composite structures, 10.40, 10.41
advantages and limitations, 7.13, 7.14 cross reference table, 4.12, 4.18 standard for, 10.36
applications, 7.13–7.16 materials suitable for, 17.5 nickel. See also ferromagnetic materials
calibration of equipment, 7.10–7.11 measurement setups, 17.6 nickel alloy castings, 4.5
cross reference table, 4.11, 4.12 nickel forgings, 4.6
microwaves
current waveform in, 7.8, 7.9 frequency bands of, 17.5 noise. Seealso acoustic emission testing; barkhausen
daily system performance check, 7.11–7.12, 7.15 generation of, 17.6 noise measurements
electrical safety in, 7.12 interactions with matter, 17.6 in eddy current testing, 8.29, 8.30, 8.44–8.45
formulas for, 18.2 receiver operating characteristic and, 3.5–3.6
MIL-HDBK-1823, 3.3–3.4, 3.5, 3.7, 3.11
historical development of, 1.5 militaryaircraft. Seealso bombers; Departmentof noise equivalent temperature difference, 11.8
magnetization fields in, 7.4–7.6 Defense(DOD); United States Air Force(USAF) nonadiabatic behavior, 11.4
principles, 7.2–7.9 categories of, 2.3 nonadiabatic discontinuity, 11.5–11.6
quantitative quality indicators for, 7.7 engine components, eddy current testing, 8.46–8.47 nondestructive characterization (NDC), 2.8, 19.14
of small objects, 7.14 F-15 fighter, 10.40 nondestructive evaluation (NDE), 2.8, 19.14
standards and specifications, 7.10–7.12 fatigue effects for design, 2.14–2.15 nondestructive examination (NDE), 2.8, 19.14
water conditioning agents for, 7.10 missions of, 2.3 nondestructive inspection (NDI), 2.8, 19.14
what can go wrong, 7.15 safety factors in design, 2.8 nondestructive testing (NDT). See also inspections;
magnetic particles, 7.8–7.9 standardization of testing, 1.5, 1.6 reliabilityofnondestructivetests; standards for
concentration in wet bath, 7.9 nondestructive testing
for transport, 2.3, 2.14
magnetic permeability millimeter waves, 17.5, 17.6 defined, 1.2, 2.8, 19.14
basic physics of, 7.2, 7.3, 8.9–8.10 applications of, 17.7, 17.8 design process for aircraft and, 2.8–2.9
eddycurrenttestingand, 8.9–8.11, 8.14 MIL-STD series of directives, 1.5 explosion in capabilities of, 1.6
impedance plane plot and, 8.6 MIL-STD-1530, 2.15, 3.3, 15.7 historical development of, 1.5–1.6
magnetic particle testing and, 7.8 importance to aerospace industry, 1.3–1.4
mirrors, 5.8
meandering winding magnetometer and, 8.20 missiles, 2.2, 2.3, 2.5, 2.13 issues in, for aerospace industry, 1.3–1.4
relative, 7.1, 7.3, 8.10, 18.2 missions professional organizations, 1.5–1.6
magnetic susceptibility, 7.2, 18.2 synonymous with nondestructive inspection, 2.8
aircraft, 2.3
magnetization, 7.2, 18.2 defining for design process, 2.8 tables of methods, 4.10–4.19
magneto-optic imagers, 8.21–8.23 spacecraft, 2.4 Nondestructive Testing Handbook, first edition, 1.5, 2.8
magnetoresistive sensing, 8.23–8.24 modal analysis, 17.24, 17.25 nonlinear acoustics, 9.26
magnetostrictive sensors, 15.8, 15.9 modelassistedprobabilityofdetection, 3.11 for bond evaluation, 14.3
magnification, 5.5, 5.8, 18.2. See also geometric moire measurement, 5.13 nonmetals. See also composites
magnification moisture detection, by microwave interferometry, 17.9 inspection issues, 4.5
of composite materials, 5.21, 5.22, 5.23 moisture diffusion monitoring, with dielectrometry, 17.12 testing methods, 4.5
magnifiers, 5.8 nuclear pressure vessels, neutron embrittlement in, 17.4

20.8 AEROSPACE NONDESTRUCTIVE TESTING

O piezoelectric materials, 9.10, 9.11. See also
transducers, ultrasonic
projection radiologic testing, 10.8, 10.10, 10.42. See also
microfocus radiologic testing
object-to-detector distance, 10.9, 10.22, 10.23 pillowing, 10.34 proof testing, 1.2, 14.4, 14.8–14.9
optical methods, 5.8–5.15 piston engines, 2.6, 4.8 propellant
optical principles, 5.3–5.7 pitch catch ultrasonic testing, 9.4, 9.14 inspection issues, 4.5
optical profiling instruments, 5.12–5.13 air coupled, 9.21 solid, 2.7
orthogonal field probes, 8.14 fatigue cracks at weep hole, 9.34 testing methods, 4.5
overextended core (OX-core), 2.33 in-situ mount for, 15.6 propeller, 2.6
oxygen systems, penetrants for inspection of, 6.9 laser technique, 9.24 propulsionsystems, 1.4, 2.6–2.7. Seealso engines;
pitting corrosion, of landing gear truck beam, 5.17–5.18 rocket motors
Planck’sradiationlaw, 11.7 creep in, 2.11
P plastic deformation, 2.10–2.11
plastics. See also composite materials
pulse echo ultrasonic testing, 9.4, 9.14–9.18
for disbonds, 14.3, 14.8
paint, 1.4, 4.8. See also coatings
inspection issues, 4.5 laser technique, 9.24
corrosion under, 17.10–17.11 penetrant materials causing harm to, 6.9 pulse phase thermography, 11.14
cure state monitoring, 17.14
testing methods, 4.5 pulsed eddy current testing, 8.14, 8.15
eddy current testing and, 8.29
welded joints, 4.7
thickness over metal, 17.22, 17.23 plating. See also cladding, metal; coatings
pair production, 10.2, 10.3
parallax techniques, radiologic, 10.33
hydrogen embrittlement and, 2.12
impedance plane plot and, 8.6
Q
paramagnetic materials, 7.2, 7.3, 7.4 quadrature demodulation, 8.5
POD. See probability of detection (POD)
parametric modeling, 8.16 Poisson’s ratio, 9.5, 9.24, 18.3 quasi-isotropic layup, 2.26, 2.27
partial vacuum shearography, 12.9, 12.25–12.26
polarized transverse waves, 9.4, 9.22
partial volume effect, 10.17
polymerresins, 2.24. Seealso fiberreinforcedpolymer
patch repairs
monitoring crack growth under, 15.4, 15.7
(FRP) composite laminate R
porosity. See also voids
thermographic validation of, 11.9, 11.11, 11.16 radiant energy, 5.5, 5.6
porosity, in aluminum alloys, 10.37
pattern recognition methods radiographic testing. See radiologic testing
porosity, in composite materials, 2.29
acoustic emission features and, 13.7–13.8, 13.13– radiography. See radiologic testing
pressure shearography of, 12.26, 12.27
13.14 radioisotopes, 10.6, 10.7
ultrasonic detection of, 9.33
in eddy current testing, 8.16 radiologic testing, 10.1–10.42. Seealso computed
postemulsifiable penetrant materials, 6.5
liquid penetrant interpretation and, 6.15 tomography; film density; microfocus radiologic
postemulsifiable process, 6.3, 6.4, 6.5
overview, 13.7–13.8 testing; projection radiologictesting
posture of observer, 5.16
in process compensated resonance testing, 17.26 advantages and limitations, 10.38
potter bucky grid, 10.24
peak rotational frequency, 8.7, 8.43–8.44 advent of, 1.5
prepreg, 2.26
peening operations, residual stress applications, 10.37–10.42
pressure bottles, acoustic emission testing, 13.12–13.13
barkhausen noise and, 17.4 artifacts in, 10.24–10.27, 10.31, 10.39
pressure change leak test, 16.6–16.8
X-ray diffraction and, 17.17 basic physics of, 10.2–10.5
pressure shearography, 12.9, 12.26
penetrameters, 10.29–10.30 for bond evaluation, 14.4
pressure vessels
penetrant testing. See liquid penetrant testing computed, 10.11–10.12, 10.24–10.25, 10.31–10.32,
composite overwrap, 12.6, 12.7, 12.14, 12.16,
penumbral shadow, 10.21, 10.22, 10.42 10.36, 10.38
12.24, 12.26
performance and proof tests, 1.2. Seealso proof testing cross reference table, 4.11, 4.15
fiberglass/epoxy, acoustic emissions, 13.7
permeability. See magnetic permeability digital, 10.9
nuclear, neutron embrittlement in, 17.4
permittivity, 17.13, 17.14, 18.6 energies compatible with various absorbers,
primary design and fabrication, 2.9
phase discrimination, 8.16 10.4–10.5
primary structure, 2.3
phase of ultrasonic wave, 9.4 exposure factor, 10.8–10.9, 10.23
principal component analysis, 11.14
changed at interface, 9.7 film, 10.9, 10.10, 10.31, 10.38
probability of detection (POD)
continuous at interface, 9.6 focal spot size, 10.7–10.8, 10.10, 10.21, 10.23
crack size and, 3.4
with thin layer or gap, 9.8 formulas for, 18.4–18.5
for damage tolerant design, 2.18, 2.19
phase stepping shearography, 12.2, 12.3–12.4 geometric rules for promoting quality, 10.23
estimation of, 3.5
phased array ultrasonic testing, 9.14, 9.20 imagequality, factorsaffecting, 10.19–10.28
introduction to, 3.2–3.4
for bolt hole cracks, 9.34 imagingprinciplesfor, 10.9
luminance ratios in detection area and, 5.6
with electromagnetic acoustic transducers, 9.22 interpretation, 10.31–10.32
mistakes in understanding of, 3.7
for scribe line crack at lap splices, 9.35–9.36 introduction to, 10.2
model assisted approach to, 3.11
for turbine engine disk inspections, 9.32–9.33 microfocus X-raytubes, 10.7–10.8, 10.10, 10.13
origin of concept, 1.6
phosphors, photostimulable, 10.11–10.12. See also orientation of crack and, 10.27–10.28
required in aerospace industry, 2.8
computed radiologic testing principles of, 10.2–10.9
threshold for, 3.3
photoconductive materials, 10.14 projection magnification in, 10.8
probability of detection curve, 3.2
photoelectric effect, 10.2, 10.3, 10.4–10.5 real time, 10.12–10.13, 10.25, 10.31–10.32
estimating, 3.3–3.4
photogrammetry, 5.13 reference images, 10.36
pixels in discontinuity indication and, 12.10
correlated with acoustic emissions data, 13.14, 13.16 scattered radiation in, 10.23–10.24
sample set for, 3.11
photography, digital, 5.8 sensitivity measurement, 10.29–10.30
process compensated resonance testing, 17.24, 17.25–
photometers, 5.7 specialized and emerging techniques, 10.33–10.35
17.26
photometry, 5.5, 5.7 standards for, 10.36
profiling instruments, optical, 5.12–5.13
photon, 5.3, 10.2 techniques, 10.10–10.18
profit, 2.8
photostimulated luminescence, 10.11 what can go wrong, 10.39
X-ray sources for, 10.5–10.8

INDEX 20.9
radiometry, 5.5 removingagents, forliquidpenetrants, 6.3 sandwichstructures, 2.32–2.34, 14.2. Seealso bond
radioscopic testing, 10.12–10.13 solvent removers, 6.5, 6.6, 6.12 testing; bondedassemblies; foamcore sandwich
advantages and limitations, 10.38 residual stresses, 3.7. See also stress structures; honeycomb structures
artifacts in, 10.25 barkhausen noise measurements and, 17.2–17.4 air coupled ultrasonic testing, 9.21
standards for, 10.36 conductivity and, 8.8 in shearographicvacuumtestchamber, 12.6, 12.8
radomes X-ray diffraction measurements of, 17.15, satellite, 2.5
burns to, 5.22 17.16–17.17 Saturn V, 2.5
glass fibers in, 2.25 resins, polymer, 2.24. See also fiber reinforced polymer scanned focused transducer holography, 9.27
microwave inspection, 17.8 (FRP) composite laminate scanning vibrometry, 17.24, 17.25
shearographic testing, 12.6, 12.8, 12.12, 12.21 resistive (real) component of impedance, 8.5, 8.6 scattered radiation, 10.23–10.24
rayleigh (surface) waves, 9.13, 9.22, 9.24 resistivity, 8.7–8.8, 8.30, 8.31 in backscatter radiologic testing, 10.34, 10.35
reactive component of impedance, 8.5, 8.6. See also resolution of radiographic image, 10.39 in reverse geometry radiologic testing, 10.33–10.34
impedance plane analysis resolution of visual image from sides of crack, 10.41
readily detectable impact damage, 2.30 for borescopic inspection of engine, 5.21 scintillators, X-ray capture with, 10.13–10.14
real time radiologic imaging, 10.12–10.13 reference standard for, 5.11, 5.16 scratches, optical profiling of, 5.12–5.13
artifacts in, 10.25 resonance ultrasonic testing, 9.18–9.19, 14.3, 14.7–14.8 scribe line crack at lap splices, 9.35–9.36
image processing and interpretation, 10.31–10.32 resonance vibration testing, 17.24–17.26 secondary bonding, 2.32, 2.34
real time ultrasonic imaging, 9.26 resonant ultrasound spectroscopy, 17.25 secondary structure, 2.3
receiver operating characteristic (ROC), 3.5–3.6 retained austenite measurements, 17.16 sectorial scanning (S-scan), 9.20
reciprocity law, 10.9 retirement for cause systems, 3.9 self-inductance. See inductance
reference standards reverse geometry radiologic testing, 10.33–10.34 semiconductors, X-ray capture with, 10.14
defined, 8.25 right hand rule, 8.3 serviced engine component inspections, 9.32, 9.33
design and fabrication of, 2.9 rigid borescopes, 5.9 shaped coils, 8.13
liquid penetrant testing, 6.10, 6.12 rivet holes, radiologic inspection, 10.42 shear loading
magnetic particle testing, 7.11–7.12 rivetedstructures,4.6. Seealso fastenedstructures of adhesive bonds, 2.32
meandering winding magnetometer, 8.19 rivets. Seealso fasteners compression failure due to, 2.11
radiologic testing, 10.36 free, as foreign object, 10.42 shear modulus, 18.3
shearographic testing, 12.10, 12.11, 12.16–12.18 magneto-optic imaging, 8.22 shear ultrasonic waves. See transverse (shear)
reference standards, eddy current testing ROC (receiver operating characteristic), 3.5–3.6 ultrasonic waves
bolt hole inspection, 8.41–8.42, 8.45 rocket fuel tank, foam insulation, 12.9 shearographic testing, 12.1–12.27
conductivity, 8.30–8.31 rocket motors, 2.7 acoustic excitation safety in, 12.30–12.31
corrosion, 8.36 composites for, 2.24 advantages and limitations, 12.19
filter settings, 8.44, 8.45 high reliability required of, 1.4 applications, 12.2, 12.7, 12.19–12.28
gap thickness, 8.36 inspection issues, 4.8 of bonded structures, 14.4, 14.5, 14.8
metal thickness, 8.34–8.35 testing methods, 4.8 calibration, 12.13–12.14
reference standards, ultrasonic testing, 9.18 rockets, 2.2, 2.3, 2.5 cross reference table, 4.11, 4.17
composite ply drop part, 9.24 evolution of design requirements, 2.13 discontinuity indication in, 12.10–12.11
flat bottom holes, 9.16–9.18, 9.31, 9.32 rogue flaw, 2.8, 3.3 field of view, 12.10–12.12
side drilled holes, 9.19, 9.31, 9.32 rolled stock, 4.6 fixturing of test object, 12.10
turbine engine components, 9.31 rotor shaft, magnetic stress gages, 15.12–15.13 focusing of camera, 12.12
reference standards, visual testing rotorwing aircraft, 2.3. See also helicopter image interpretation and analysis, 12.9
apparent image quality and, 5.16 evolution of design requirements, 2.13 introduction to, 12.2–12.5
spatial frequency resolution target, 5.11 rubber, 4.5 laser safety in, 12.29–12.30
reflection, optical rudder flange cracks, 5.17 operation, 12.10–12.15
glare and, 5.16 portable systems, 12.20–12.21
principles of, 5.3–5.4 production systems, 12.22–12.23
reflection, ultrasonic, 9.6–9.8
reflection coefficient, ultrasonic, 9.6, 9.7
S selection of stress technique, 12.7, 12.14
specifications and standards, 12.16–12.18
refraction S glass (structural glass), 2.25 stress loading examples, 12.24–12.27
of light, 5.3, 5.4–5.5, 18.2 SAE AS5282 ring, 7.10, 7.11–7.12 stress loading for, 12.6–12.8, 12.11, 12.14
of ultrasound, 9.8–9.9 safe life design, 2.14, 2.17 what can go wrong, 12.20
regulations, evolution of, 2.13. Seealso Federal Aviation uninspectable structure and, 2.20 shot peening. See peening operations, residual stress
Administration (FAA) safe life extension program, 2.17 signal processing, in eddy current testing, 8.16
reinforcement fibers for composites, 2.25–2.26 safety, of inspector of engine components, 8.46, 8.47
reinforcement of structures, 1.4 with acoustic excitation, 12.30–12.31 signal processing, in thermographic testing, 11.2, 11.11–
relative permeability, 7.2, 7.3, 8.10 with lasers, 12.29–12.30 11.13
reliabilityof nondestructive tests, 1.6, 3.1–3.11. Seealso in liquid penetrant testing, 6.9 signal reconstruction, thermographic, 11.11–11.12, 11.18,
probability of detection (POD) in magnetic particle testing, 7.12 11.19–11.20
human factors and, 3.5, 3.8–3.10 safety factors simultaneous injection, 8.14
introduction to, 3.2–3.4 compared to otherindustries, 1.3, 2.8 skin effect, 8.10–8.11, 8.27, 8.33
mistakes in estimation of, 3.7 in primarydesignprocess, 2.9 eddy current probe frequency and, 8.43
modelassistedapproachto, 3.11 in safe life design, 2.17 with meandering winding magnetometer, 8.18–8.19
performance measures in practice, 3.5–3.6 weight of structures and, 1.3, 2.8 skins
summary of, 3.11 sampling, 1.2 of bonded assemblies, 2.32
reliability of product, 2.8 guided wave testing of, 9.26
remote field eddy current testing, 8.16–8.17 slow crack growth approach, 2.19–2.20, 2.21
smart washers, 15.12 static loads, 2.9 surfacefinish, widespreadfatiguedamageand,
SN curve (fatigue curve), 2.16 failure modes and, 2.10 2.22–2.23
Snell’s law structural design and, 2.16 surface strain measurement, 5.13
for light, 5.4, 18.2 statistical inferences approach, 1.6 surface tension, 6.3
for ultrasound, 9.8, 18.4 stealth coatings, 17.22 surface treated components, barkhausen noise
solid propellant, 2.7 steel. Seealso ferromagnetic materials; high strength analysis, 17.4
solvent cleaning, 6.6 steels; stainless steel surface (rayleigh) waves, 9.13, 9.22, 9.24
solvent removable penetrant, 6.5 barkhausen noise measurements, 17.2–17.4 syntactic core, 2.34
solvent removers, 6.5, 6.6 castings, 4.5 systems, aircraft, 2.3
overapplication of, 6.12 eddy current testing, 8.11, 8.12 design of, 2.9, 2.10
sonic bond testers, 14.5, 14.6–14.7 fastened structures, 4.6 inspection issues, 4.9
sonic shapes, 9.31–9.32 forgings, 4.6 testing methods, 4.9
sonic thermography, 11.14–11.15 high temperature, 2.11 ultrasonic testing, 9.3
source-to-detector distance, 10.23 hydrogen embrittlement, 2.12
source-to-object distance, 10.9, 10.23 impedance plane plot, 8.6
space shuttles, 2.5, 2.36
damage tolerance approach for, 3.3
machined parts, 4.6
representative properties, 2.11
T
flash thermography, 11.16 retained austenite measurements, 17.16 tap testing, 14.3, 14.5–14.6
foam insulation, 10.34, 12.22, 17.9–17.10 thermal response, 11.4, 11.5 temperature. See also heat treatment
impact damage to wing leading edge, 15.9–15.10 welded joints, 4.7 characteristics of materials and, 1.4
Space Station, International, 2.4, 2.36 X-ray transmission, 10.19, 10.20 conductivity and, 8.8, 8.33
spacecraft, 2.2, 2.3, 2.4, 2.5 stereo radiologic testing, 10.33, 10.38 creep and, 2.11
atmosphere leakage, 15.10–15.11 stereo viewing, in fiberscopes, 5.16 eddycurrent depthof penetrationand, 8.11
evolution of design requirements, 2.13 stiffening of structures, 1.4 tensile strength, 2.11
visible penetrant testing, 6.2 composite structures, 2.27 tenth-value layer (TVL), 10.4
spacing, eddy current measurement stiffening ribs, ultrasonic inspection for cracks, 9.34 terahertz radiation, 17.5, 17.6
corrosion induced, 8.36–8.37 stiffness, material, 2.11, 9.26 applications of, 17.7
between metal layers, 8.35–8.36 stiffness of sandwich structures thermal coatings, 4.8
spar cap cracks, 5.19–5.20, 9.34, 9.35 disbonds and, 14.2 thermal conductivity, 11.4
spatial frequency resolution targets, 5.11 mechanical testing techniques and, 14.6 thermal diffusivity, 11.3, 11.4, 11.13, 18.5
spatial resolution, radiographic, 10.20–10.23 strain, 2.10, 9.26 for several materials, 12.24
image quality indicators and, 10.30 strain gages, 15.4, 15.5 shearography and, 12.24
specific activity, 10.6 strain measurement thermal effusivity, 11.3, 11.4, 11.5–11.6, 11.11, 11.12,
specific modulus, 2.26 moire method, 5.13 11.13, 18.5
specific strength, 2.26 photogrammetry, 5.13 thermal shearography, 12.9, 12.24, 12.25
specifications. Seealso standardsfor stress, 2.10, 9.26. Seealso residualstresses thermoelastic effect, 9.22, 9.23
nondestructive testing surface strain measurement and, 5.13 thermographic testing, 11.1–11.20
basic concept of, 8.25 ultrasonic measurement of, 9.22 acoustic, 11.14–11.15
developed in design phase, 2.9 stress corrosion cracking, 2.22 advantages and limitations, 11.17
spectral distribution of light, 5.5, 5.6 eddy current testing for, 8.28 application examples, 11.2, 11.17–11.20
specular reflection, 5.3 originating at a pit, 5.18 artifacts in, 11.18
spherical aberration, 5.5 stress corrosion fracture, of landing gear truckbeam, of bonded structures, 14.4, 14.5, 14.8
splice joints, ultrasonic inspection, 9.34 5.17–5.18 cross reference table, 4.11, 4.16
split core coils, for eddy current testing, 8.12–8.13 stress gages, magnetic, 15.12–15.13 discontinuity detectability, 11.4–11.6, 11.8
stainless steel stress risers, 2.12 discontinuity sizing, 11.6
eddy current testing, 8.11 stress wave testing, 14.4, 14.5, 14.9 emerging techniques, 11.14–11.15
for World War II aircraft, 1.5 stress-to-strain curve, 2.10 formulas for, 18.5–18.6
standard depth of penetration, of eddy current, 8.10– linear, 9.4 instrumentation, 11.7–11.10
8.11, 8.33 structural failure modes, 2.10–2.12 interpretation and analysis, 11.11–11.13
in magneto-optic imaging, 8.22–8.23 structural health monitoring, 15.1–15.13 principles of, 11.2–11.6
standards for nondestructive testing, 1.5–1.6, 8.25. with acoustic emission testing, 13.2, 15.4, 15.7 standards, 11.16
See also reference standards; specifications advantages and limitations, 15.5 what can go wrong, 11.18
acoustic emission testing, 13.9–13.10 of composites, 15.5–15.6 thermophysical properties of materials, 11.4
eddy current testing, 8.25–8.26 for crack monitoring, 15.4, 15.7, 15.8, 15.12 thermoplastics, 2.24
focal spot measurement, 10.8 of fatigue and stress, 15.11–15.13 thermosets, 2.24
holographic testing, 12.16 with guided ultrasonic waves, 15.4, 15.7–15.8 thick components, 1.4
image quality indicators, 10.30 methods for, 15.4 eddy current testing, 8.29
liquid penetrant testing, 6.10 principles of, 15.2–15.3 thickness
magnetic particle testing, 7.10 of space shuttle wing, 15.9–15.10 compression failure and, 2.11
radiologic testing, 10.36 of spacecraft leakage, 15.10–15.11 corrosion and, 2.12
shearographic testing, 12.16 systems for, 15.6 thickness, eddy current measurement
thermographic testing, 11.16 structural modification point, 2.22 of coatings, 8.27, 8.29
ultrasonic inspection, 9.28–9.29 structural waves, 9.25. Seealso guided (lamb) waves conductivity and, 8.33
visual testing, 5.13–5.15 subcontractors, and design process, 2.8, 2.9 impedance of coil and, 8.6
static life designs, 2.16 subsystems. See systems, aircraft of layers, 8.34–8.36
surface contamination, FTIR method, 17.21 of metals, 8.34–8.35

INDEX 20.11
thickness, microwave interference microfocus X-ray crack detection, 10.41 ultrasonic waves
measurement, single crystal, X-ray diffraction, 17.17 attenuation, 9.6, 18.3
17.8–17.9 vibrothermography, 11.15 basic characteristics, 9.4–9.5
thickness, ultrasonic measurement, 9.19 visual testing, 5.20 diffraction, 9.9
spectroscopic, 9.24 turbine engine compressor seals, holography, focused beam of, 9.12, 9.20
thin layers, ultrasound in, 9.8 12.20, 12.28 modes of, 9.13
thin structures turbine engine disks refraction, 9.8–9.9
eddy current testing, 8.29 barkhausen noise measurements, 17.3 transmission and reflection, 9.6–9.8, 18.3
required sensitivity for testing of, 1.4 eddy current inspection, 3.9 ultraviolet lamps
safety factors and, 2.8 ultrasonic inspection, 9.31–9.33 in automated system, 6.14
thinning by corrosion, 2.12 turbine engine sustainment initiative, 9.33 daily checks of, 6.8
eddy current measurement, 8.36–8.37 turbofan engines, 2.6–2.7 intensity measurement of, 7.11
through-transmission ultrasonic testing, 9.4, turbojet engines, 2.6–2.7 safety issues with, 6.9, 6.10
9.14–9.18 turboprop engines, 2.6–2.7 what can go wrong, 6.12, 7.15
air coupled, 9.21 TVL (tenth-value layer), 10.4 ultraviolet light. See ultraviolet lamps;
for disbonds, 14.3, 14.8 type 1 penetrant, 6.5 ultraviolet (UV-A) radiation
laser technique, 9.24 type 2 penetrant, 6.5 ultraviolet (UV-A) radiation
time based spectroscopy, microwave, 17.6 in liquid penetrant testing, 6.3, 6.4, 6.5,
time domain multiplexing, 8.14 6.8, 6.9, 6.10, 6.12, 6.14
titanium
bonded joint assemblies, 4.7
U in magnetic particle testing, 7.2, 7.8, 7.11,
7.15
castings, 4.5 ultrasonic fatiguing, 14.9
uninspectable structure, 2.20
ofdoubler, corrosionthinningand, 8.37 ultrasonic imaging, real time, 9.26
United States Air Force (USAF). See also
eddycurrent testing, 8.10, 8.11, 8.27 ultrasonic leak testing, 16.5–16.6
Department of Defense (DOD); military
fastened structures, 4.6 of spacecraft, 15.10–15.11 aircraft
forgings, 4.6 ultrasonic spectroscopy, 9.24
Aircraft Structural Integrity Program
machined parts, 4.6 for bond evaluation, 9.24, 14.3
(ASIP), 2.15, 2.17, 3.3
welded joints, 4.7 ultrasonic testing, 9.1–9.36. See also phased
automated turbine engine inspection,
X-ray diffraction by, 10.24, 17.16–17.17 array ultrasonic testing; transducers,
9.33
titanium alloys ultrasonic
damage tolerant design and, 2.20–2.21, 3.2–
eddy current testing, 8.29 A- , B-, and C-scans, 9.15–9.17
3.4
hydrogen embrittlement of, 2.12 advantages and limitations, 9.30
unmanned air vehicles, 2.3
representative properties, 2.11 air coupled, 9.21, 14.3, 14.5, 14.8
unsharpness of image, 10.21–10.22, 18.5
toroidal magnetic field, 7.5–7.6 application examples, 9.30–9.36
toughened resin material systems, 2.30 automated inspection systems, 9.32–9.34
basic methods, 9.14–9.20
tracer dye leak test, 16.8
training of inspector, 3.8 beam alignment to test object, 9.2 V
ofbondedstructures, 4.7, 9.3, 9.19, 9.21, vacuum monitoring, comparative, 15.4
transducers, ultrasonic, 9.9–9.13
9.24, vacuum shearography. See partial vacuum
broadband, 9.24
with comb configuration, 9.25 14.3, 14.8 shearography
calibration blocks for, 9.18, 9.19 vertical risers, ultrasonic inspection, 9.34
composite, 9.10, 9.11
coupling of transducer to object, 9.14– vibration holography, 12.28
coupling to object, 9.14–9.15
9.15 vibrationmonitoring, forcrackdetection, 15.7
damped, 9.10
cross reference table, 4.11, 4.14 vibration shearography, 12.9, 12.26–12.27
electromagnetic, 9.22
data display, 9.15–9.17 acoustic safety in, 12.30–12.31
narrowband, 9.10, 9.12, 9.24
formulas for, 18.3–18.4 vibration testing, 17.24–17.26
near and far field effects of, 9.12–9.13,
hand held forms of, 9.14 vibrothermography, 11.14–11.15
18.4
historical development of, 1.5 of bonded structures, 14.4
Q factor, 9.10, 9.11
immersion techniques, 9.15, 9.31, 9.33 video borescopes, 5.10–5.11
transimpedance, 8.19, 8.20
transmission, ultrasonic, 9.6–9.8 introduction to, 9.2, 9.4 for engine inspection, 5.21
major issues for aerospace, 9.14 referencestandardforresolutionof, 5.16
transmission coefficient, ultrasonic, 9.6, 9.7
mechanical scanners, 9.14, 9.15 video imagers, digital, 5.8
transverse (shear) ultrasonic waves, 9.4, 9.5,
9.13 principles of, 9.2–9.13 video probes. See video borescopes
resonance testing, 9.18–9.19, 14.3, 14.7– video radioscopy, 10.12–10.13
polarized, 9.4, 9.22
14.8 viewing angle, 5.7, 5.16
refracted, 9.8–9.9
scan speeds, 9.15 viewing distance, 5.8
spectroscopy with, 9.24
sensitivity to discontinuities, 9.18 vigilance of inspector, 3.9
tubetesting,eddycurrenttechnique, 8.15
specialized or emerging techniques, 9.21– visible (dye) penetrant testing, 6.2
tungsten, attenuation of X-rays, 10.3
turbine blades. Seealso jet engines 9.27 developer for, 6.6
aluminide coatings on, 17.22 standards for, 9.28–9.29 not used before fluorescent penetrants,
for structural health monitoring, 15.4, 6.12
arc burn on, 5.20
15.5, 15.6, 15.7–15.8 penetrant materials for, 6.5
barkhausen noise measurements, 17.3
of turbine engine disks, 9.31–9.33 vision acuity, 5.7
blending of, 5.20
what can go wrong, 9.30, 9.31
computed tomography cross section,
10.17
digital radiograph, 10.17
visual testing, 5.1–5.23. See also impact water line, 2.4 wing spar cap cracks, 5.19–5.20, 9.34, 9.35
damage, water washable penetrants, 6.5 wing splice
visual testing waveforms, ultrasonic, 9.13 magnetoresistive sensing, 8.23–8.24
advantages and limitations, 5.2 wavelength ultrasonic testing, 9.34
applications, 5.2, 5.16–5.23 of electromagnetic radiation, 5.3 wing station, 2.4
cross reference table, 4.10 of light, 5.6, 18.2 wing tip, eddy current thickness measurement,
formulas for, 18.2 of ultrasonic waves, 9.4–9.5, 9.6 8.35
introduction, 5.2 wear, 2.12 witness coupon testing, 14.4, 14.8, 14.9
optical methods, 5.8–5.15 radiologic inspection for, 10.42 World War II
optical principles for, 5.3–5.7 widespread fatigue damage and, 2.22– aircraft design considerations, 2.14
preparation for, 5.8 2.23 nondestructive testing in, 1.5
standards for, 5.13–5.15 weep hole, 9.34 wrinkling (marcelling), 17.8
as step in penetrant testing, 6.5–6.6 weight
what can go wrong, 5.16 aluminum alloys and, 2.16
voids. See also porosity
acoustic holography, 9.27
composite materials and, 2.26, 2.30
discontinuity size and, 1.3, 1.4
X
air coupled ultrasonic testing, 9.21 safety factors and, 1.3, 2.8 X-ray diffraction, 17.15–17.17, 18.6
in composites, 5.21, 5.22, 5.23 welds as problem in radiologic testing, 10.24
microwave propagation through, 17.10 anisotropic, 9.22 X-ray fluorescence spectroscopy, 17.18–17.19
shearographic testing, 12.5, 12.26, 12.27 barkhausen noise measurements, 17.3 X-ray image intensifiers, 10.12, 10.13
crater cracks in, 6.8 X-ray sources, 10.5–10.8
eddy current testing, 8.14 X-ray tubes, 10.5, 10.7–10.8
X-rays. See also radiologic testing
W inspection issues, 4.7
in liquid fuel engine components, 2.7 attenuation of, 10.2–10.5, 18.4
walk-around inspection, viewing distance for, magnetic particle testing, 7.9 penetration of various materials, 10.19
5.8 penetrant testing materials for, 6.6
water, thermographic detection of radiologic testing, 10.36, 10.37, 10.39,
with ambient heating of ice, 11.9
in composite sandwich structures, 11.9–
10.41 Y
testing methods, 4.7
11.10, 11.17–11.18, 11.19 yield strength, 2.11
ultrasonic testing, 9.3, 9.21, 9.22
by continuous scanning, 11.10 yield stress, 2.10
ultrasoundcalibrationblock, 9.18, 9.19
with direct thermal contact, 11.9–11.10 Young’s modulus, 2.10–2.11, 9.26
visiblepenetranttesting, 6.2
direct viewing of image, 11.11 WestarTM satellite, 2.5
with flash thermography, 11.18, 11.19 wet horizontal mag machines, 7.4, 7.5, 7.8,
in helicopter rotor blades, 11.10, 11.18 7.13 Z
in honeycombstructures, 2.33, 11.9,11.11, wide field eddy current probes, 8.13–8.14 zoom imaging
11.17–11.18 widespread fatigue damage, 2.21–2.23 cameras, 5.5
with solar heating, 11.8–11.9 windowbeltsplice, subsurface cracks, 8.40 microfocus radiologic testing, 10.41,
water conditioning agents, for magnetic wing cracks 10.42
particle under panel, 5.16–5.17 visual test magnification, 5.8
testing, 7.10 radiologic testing, 10.27
water effects. See moisture effects ultrasonic testing, 9.34

INDEX 20.13
Figure Sources

For providing illustrations to this book, ASNT Chapter 8. Electromagnetic Testing

thanks the following contributors, as well as others
not listed here. Parties wishing to republish any of Figures 1, 14 —Laser Technology (F. Spears),
these illustrations should first get the permission of Seattle, WA.
ASNT and the contributor. Figures 5-9 — D.A. Raulerson, West Palm Beach,
Florida.
Figures 12, 61-64 — J. Brausch, Wright-Patterson Air
Chapter 2. Aerospace Structures Force Base, OH.
Figure 9 — American Institute of Aeronautics and Figures 25-27 — Jentek Sensors (A.P. Washabaugh),
Astronautics (B.K. Wood), Reston, VA. Chula Vista, CA.
Figure 11 — British Airways Heritage Center and
Waterside Headquarters, London, United Kingdom. Chapter 10. Radiologic Testing
Figure 31 — Hexcel, Stamford, CT.
Figures 10, 11 — ASTM International (C. Bueno,
T. Gordon), West Conshohocken, PA.
Chapter 5. Visual Testing Figure 39 — FeinFocus USA, Stamford, CT.
Figure 8 — Pratt and Whitney, Hartford, CT.
Figure 10-11 — General Electric, Lewistown, PA. Chapter 13. Acoustic Emission Testing
Figure 15 — Micro-Epsilon, Ortenburg, Germany.
Figure 17 — Trilion Quality Systems (J. Tyson), West Figures 1, 4-5, 7-9 — Physical Acoustics
Conshohocken, PA. (V. Godinez-Azcuaga), Princeton, NJ.
Figures 10-12 — SPIE (V. Godinez-Azcuaga),
Bellingham, WA.
Chapter 6. Liquid Penetrant Testing
Figure 1 — Delta Airlines (L. Brasche), Atlanta, GA. Chapter 17. Other Methods
Figures 8, 12, 15 — American Institute of Physics,
Chapter 7. Magnetic Particle Testing College Park, MD.
Figure 7-9 — Rolls-Royce, London, United Kingdom Figures 27-28 — Thermo Fischer Scientific
Figure 14-15 — Delta Airlines (L. Brasche), Atlanta, GA. (M.P. Lessard), Tewksbury, MA.
Figure 29 — Agilent Technologies (A.J. Rein,
J.A. Seelenbinder), Danbury, CT.

20.15
 NDT
Aerospace

American Society for Nondestructive Testing