You are on page 1of 378


PDA-W® Pile Driving Analyzer Software


IMPORTANT - READ CAREFULLY: This End-User License Agreement "EULA") is a legal agreement between you (either an individual or a
single entity) and Pile Dynamics, Inc. (PDI) for the PDI SOFTWARE PRODUCT identified above, which includes computer software and
associated media and printed materials, and may include "online" or electronic documentation ("SOFTWARE PRODUCT" or
"SOFTWARE"). By installing, copying, or otherwise using the SOFTWARE PRODUCT, you agree to be bound by the terms of this EULA. If
you do not agree to the terms of this EULA, do not install, copy, or use the SOFTWARE PRODUCT.


The SOFTWARE PRODUCT is protected by copyright laws and international copyright treaties, as well as other intellectual property laws
and treaties. The SOFTWARE PRODUCT is licensed, not sold.

1. GRANT OF LICENSE. This EULA grants you the following non-exclusive rights:

- Software Product. You may install and use the enclosed SOFTWARE PRODUCT on your computer to analyze foundation piles
or similar structural elements under the action of a hammer impact.

- You must acquire and dedicate a license for the SOFTWARE PRODUCT for each computer on which the SOFTWARE
PRODUCT is used or to which it is distributed. A license for the SOFTWARE PRODUCT may not be shared or used
concurrently on different computers.

- You may make a single back-up copy of the SOFTWARE PRODUCT. You may use the back-up copy solely for archival


- Separation of Components. The SOFTWARE PRODUCT is licensed as a single product. Its component parts may not be
separated for use on more than one computer.

- Rental. You may not rent or lease the SOFTWARE PRODUCT to other parties.

- Software Transfer. You may permanently transfer all of your rights under this EULA only as part of a sale or transfer of the
SOFTWARE PRODUCT, provided you retain no copies, you transfer all of the SOFTWARE PRODUCT (including all
component parts, the media and printed materials, any upgrades, this EULA and, if applicable, the Certificate(s) of Authenticity,
AND the recipient agrees to the terms of this EULA. If the SOFTWARE PRODUCT is an upgrade, any transfer must include all
prior versions of the SOFTWARE PRODUCT.

- Limitations on Reverse Engineering, Decompilation, and Disassembly. You may not reverse engineer, decompile, or
disassemble the SOFTWARE PRODUCT,except and only to the extent that such activity is expressly permitted by applicable
law, notwithstanding this limitation.

- Termination. Without prejudice to any other rights, PDI may terminate this EULA if you fail to comply with the terms and
conditions of this EULA. In such event, you must destroy all copies of the SOFTWARE PRODUCT and all of its component

3. UPGRADES. If the SOFTWARE PRODUCT is an upgrade from an earlier version of the SOFTWARE PRODUCT, whether from Pile
Dynamics, Inc. or another supplier, you may use or transfer the SOFTWARE PRODUCT only in conjunction with that upgraded product,
unless you destroy the upgraded product. If the SOFTWARE PRODUCT is an upgrade of a Pile Dynamics, Inc. product, you now may
use that upgraded product only in accordance with this EULA. If the SOFTWARE PRODUCT is an upgrade of a component of a
package of software programs which you licensed as a single product, the SOFTWARE PACKAGE may be used and transferred only
as part of that single product package and may not be separated for use on more than one computer.

4. OEM COPYRIGHT. All title and copyrights in and to the SOFTWARE PRODUCT (including but not limited to any images, photographs,
animations, video, audio, music and text incorporated into the SOFTWARE PRODUCT), the accompanying printed materials, and any
copies of the SOFTWARE PRODUCT, are owned by Pile Dynamics, Inc. The SOFTWARE PRODUCT is protected by copyright laws
and international treaty provisions. You may not copy the printed materials accompanying the SOFTWARE PRODUCT.

5. DUAL-MEDIA SOFTWARE. You may receive the SOFTWARE PRODUCT in more than one medium. Regardless of the type or size of
medium you receive, you may use only one medium that is appropriate for your single computer. You may not use or install the other
medium on another computer. You may not loan, rent, lease, or otherwise transfer the other medium to another use, except as part of
the permanent transfer (as provided above) of the SOFTWARE PRODUCT.

6. OEM PRODUCT SUPPORT. Product support for the SOFTWARE PRODUCT is provided by Pile Dynamics, Inc.

7. TECHNICAL SUITABILITY. The Wave Equation Approach is an aid in the analysis of impact driven piles. Results may differ from
actual conditions depending on the realism of the user's input parameters and the adequacy of the general wave equation approach. It
is strongly recommended that users verify their analysis results in the field by static and/or dynamic measurements.

8. PROGRAM DEFICIENCIES. Program deficiencies which would severely limit the use of the SOFTWARE PRODUCT will be corrected
and the corrected version will be made available to the licensee at no cost for a period of one year following the receipt of the
SOFTWARE PRODUCT by the user. Such updates would be announced by way of Newsletter sent to the licensee. It is the licensee's
responsibility to order such a corrected version.

9. NO WARRANTIES. The software product is provided "as is" without warranty of any kind.

10. LIMITATIONS OF DAMAGES. To the maximum extent permitted by applicable law, in no event shall PDI be liable for any damages
arising out of the use of or inability to use this PDI product.
PDI Policy on Hardware and Software Transfer

1) Any stand alone software license (e.g. CAPWAP, GRLWEAP, PIT-W professional version) is non
transferable. Program keys (either hardware or software keys) are also non transferable. PDI will not alter
the software owner's file unless in cases of company mergers, acquisitions or company name changes.
PDI reserves the right to request proof of the reason for the name change.

2) Hardware is transferable along with its operational or basic software (e.g. PAK with its installed PDA-W,
PIT with PIT-W standard version),subject to the following conditions:

2.1) Contact PDI for payment of transfer fee:

2.2) The new owner must receive training by Pile Dynamics. Exceptions will be made only to those
previously trained in the use of the equipment by Pile Dynamics or one of its authorized representatives.
Training requirements
and fees are:
PDA: 3 days
CHA: 1 day
PIT: 1 day
Training at Pile Dynamics' offices: $800/day
Training at the buyer's location: $900/day including travel time, plus Travel and living expenses for one

Once PDI receives the transfer fee and trains the new owner, PDI will perform a complimentary checkup
and calibration on the equipment, and will change the equipment owner's file. The equipment will then
display and print the new owner's name. In addition, the new owner will be considered a registered user.
Only registered users receive technical support, notices
of upgrades and invitations to PDI Users Days. The cost of shipping the
equipment for checkup and calibration, as well as the cost of any necessary repair as identified by Pile
Dynamics during the checkup, is the responsibility of the new owner.

3) As a rule Pile Dynamics buys back equipment only as a trade in for newer models of the same
equipment. From time to time PDI may decide, at its own discretion, to purchase used equipment in other
situations. Such occasional purchases should not imply a change in PDI's no buy back policy.
PDA-W Manual of Operation

Table of Contents


1 General Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1

1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
1.2 Safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
1.3 A Word About Use of PDA-W . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
1.4 Getting Started . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4
1.4.1 Menu Bar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4
1.4.2 Status Bar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4
1.4.3 Customizing the Toolbars . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4
1.4.4 Units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4
1.4.5 Data Entry and Two-Letter Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4
1.4.6 Attaching Comments to Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-7
1.4.7 Helpful Hints for Customizing Windows for your PAK (or PC) for PDA-W . 1-8
1.5 Data Acquisition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-9
1.5.1 Hardware Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-9
1.5.2 Setting up to Acquire New Data (PAK and Remode PAX/PAL only) . . . . . 1-9
1.5.3 Basic Operation Summary for Collecting Data . . . . . . . . . . . . . . . . . . . . . 1-14
1.5.4 Accepting New Data: PAK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-15 Accepting New Data: PAX 1-16
1.5.5 Accepting New Data: PAL-R Connected to the PDA-W Program . . . . . . 1-16
1.5.6 Offset Test (PAK and Remote PAX/PAL only) . . . . . . . . . . . . . . . . . . . . . 1-20
1.5.7 Calibration Test (PAK and Remote PAX/PAL only) . . . . . . . . . . . . . . . . . 1-21
1.5.8 Changing Sensors or Sensor Status During a Test (PAK and Remote
PAX/PAL only) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-21
1.5.9 Recording the Penetration Depth LP (PAK and Remote PAX/PAL only) . 1-21
1.5.10 Determination of Wave Speed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-22 During Driving . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-22 By Calculation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-22 By Wave Up Inspection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-22
1.5.11 Elastic Modulus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-23
1.5.12 Pile Impedance - “EA/C” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-23
1.5.13 Composite Piles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-23
1.5.14 Damping Constant JC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-24
1.6 Saving Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-25
1.6.1 Deleting Blows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-26
1.6.2 Reducing Data Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-26
1.7 Replaying Data Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-27
1.7.1 Display of Multiple Data Files Simultaneously . . . . . . . . . . . . . . . . . . . . . 1-28
1.7.2 Changing Project/Pile Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-28
1.7.3 Area Calculator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-29
1.7.4 Data Adjustments for Velocity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-29
1.7.5 Changing Calibrations (or Replay Factors) in Existing Files . . . . . . . . . . 1-31

February 2009 Pile Driving Analyzer Manual, PDA-W i


1.7.6 Variable Wavespeed WC Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-32

1.7.7 Entering a Drive Log to Document the Penetration Depth LP . . . . . . . . . 1-33
1.7.8 Incrementing the BN or LP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-34
1.7.9 BN Filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-34
1.7.10 Merging Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-34
1.7.11 Changing Hammer Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-35
1.7.12 Function Keys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-35
1.8 Customizing the Graph Display . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-35
1.9 Damping Parameter (JC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-37
1.10 Warnings and Limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-37
1.10.1 Data Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-39
1.11 CAPWAP® Preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-40
1.12 Plotting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-40
1.12.1 Copy to Clipboard . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-41
1.13 BETA (Integrity/Damage Evaluation) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-41
1.14 General Data Interpretation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-42
1.14.1 Resistance Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-43
1.14.2 Capacity Monitor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-43
1.14.3 Tension Envelope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-43
1.14.4 Output Quantities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-44
1.14.5 Saving Quantity (SQ) Results for PDAPLOT or EXCEL or Printing . . . . 1-44
1.14.6 Short Recommended List of Useful Result Quantities . . . . . . . . . . . . . . . 1-45
1.14.7 More Complete List of Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-45
1.14.8 Capacity Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-46
1.14.9 Capacity Evaluation Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-47
1.15 Your Responsibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-48
1.15.1 PDA Certification Examination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-50
1.16 Help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-50

2 General Interpretation and Application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1

2.1 Suggested Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1
2.1.1 Dynamic Load Testing - Capacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1
2.1.2 Pile Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2
2.1.3 Pile Stresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2
2.1.4 Pile Integrity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2
2.1.5 Hammer Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2
2.2 PDA Output Quantities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-7
2.3 PDA Capacity Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-9
2.3.1 Damping Factor Methods: RSP (i.e., RS1, RS2, RSM, RP#) . . . . . . . . . . 2-9
2.3.2 Maximum Resistance Method: RMX (RX#) . . . . . . . . . . . . . . . . . . . . . . . 2-10
2.3.3 Automatic Method; RA2, RAU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-10
2.3.4 Unloading Method; RSU (RU#) 2-10
2.3.5 Energy Formula Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-11
2.3.6 A Quick Review of Capacity Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-11
2.3.7 Brief Advice in Application of Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-12

ii Pile Driving Analyzer Manual, PDA-W February 2009


2.4 Proportionality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-13

2.5 Testing Composite Piles or Drilled Shafts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-15
2.6 PDA Testing on Steel Followers (or “dollies”) for Concrete Piles . . . . . . . . . . . . . . 2-17

Appendix A: Case-Goble Method Derivations & Theory Guide

Appendix B: PDIPLOT Users Manual
Appendix C: Examples for PDA
Appendix D: Helpful Hints for Field Testing and Data Interpretation Using the Pile Driving
Appendix E: Dynamic Testing of Pile Foundations During Construction
Pile Evaluation by Dynamic Testing During Restrike
Appendix F: Static Pile Load-Movement from Dynamic Measurements
CAPWAP Correlation Studies
Appendix G: High-Strain Dynamic Testing of Drilled Shafts and Cast-in-Place Piles
Formalized Procedure for Quality Assessment of Cast-in-Place Shafts Using
Sonic Pulse Echo Methods
Appendix H: Testing Methods of Driven Piles
Computer-Based Wave Equation Analysis of Pile Driveability
Appendix I: Guide Specification for Dynamic Pile Testing
Specifications & Instructions for High-Strain Dynamic Testing of Drilled and Cast-
in-Place Shafts
Sample Specification for Pile Integrity Testing Using P.I.T.™
Capacity Evaluation by the Pile Driving Analyzer®

February 2009 Pile Driving Analyzer Manual, PDA-W iii

iv Pile Driving Analyzer Manual, PDA-W February 2009
Pile Driving Analyzer
Pile Dynamics, Inc.

PDA-W Manual of Operation

manual version: February 2009

(corresponds to PDA-W version 2008.098.046 and higher)



TEL: USA-216-831-6131
FAX: USA-216-831-0916

For use with PDA models PAK, PAL and PAX


1.1 Introduction

The Pile Driving Analyzer ® (PDA) from Pile Dynamics is a very useful tool for measuring and
determining the effects of impacts on a pile. The impact is often applied by the pile driving
hammer on a driven pile, but may also be due to the impact of a large drop weight applied to
a bored or augered pile, or drilled shaft. The PDA monitors acceleration and strain sensors
which are quickly attached to the pile by bolts, and process these signals after each hammer
blow during driving or restrike. The signals are digitized by the PDA, results are computed, and
the data array of the signals for a blow is stored. The PDA-W program either is instrumental
in controlling the data acquisition, or can reprocess existing files. The data may be interpreted
for pile bearing capacity, compression stresses induced at top and bottom, tension stresses
along the shaft, energy transferred to the pile or shaft, and pile integrity.

February 2009 Pile Driving Analyzer Manual, PDA-W 1-1

1.2 Safety


working conditions. THINK and BE PREPARED, especially with power (we suggest you USE
12 VOLT D.C. battery for the PDA to minimize risk, and battery powered drills whenever
possible to prepare the pile), and for ATTACHING SENSORS to PILE (we suggest that the PILE
DRIVING CREW CLIMB THE LEADS and ATTACH TRANSDUCERS, or wait till pile top is near
ground to reduce climb). Remember that YOUR SAFETY IS YOUR FIRST PRIORITY; avoid
dangerous tasks or situations.

The new PAX model of PDA has self-contained battery power for up to 8 hours operation.
Further the use of wireless data transmission with the PAX allows the PDA operator to
attach sensors with pile on the ground; “sensor protectors” then cover the sensors and
transmitter, protecting them during the pile lifting process, and eliminating the need to
climb the leads to attach the sensors. This therefore not only improves safety, but also
speeds up the testing process.

1.3 A Word About Use of PDA-W

Any W01 or X01 file created by any PDI Pile Driving Analyzer (model: PAK, PAL, PAX or even
GCPC) can be reanalyzed by the PDA-W program. There may be some files that the origin is
uncertain and thus cannot be interpreted properly and the PDA-W program will refuse to open
them since the calibration is uncertain. The PDA-W program will also read SPT Analyzer files,
although some functions will not work.

There are four different PDA systems currently available from Pile Dynamics (PDI) which can
acquire new data. They are the PAK, two versions of the PAL, and the PAX. All use the same
PDA-W program for data analysis. The PAK and PAL-R also use PDA-W to acquire the data.

PAK - Our traditional PDA, the PAK is based on standard PC technology with large hard disk,
and a built-in floppy drive, CR read/writer, or USB port for data transfer. It will operate up to 30
minutes on internal battery power (the battery operates 8/5 times longer if operating the PDA-W
program due to power management in Windows instead of the DOS PAK program). The PAK
simultaneously acquires up to 4 channels of strain and 4 channels of acceleration (2
piezoelectric and 2 piezoresistive). This feature is particularly helpful when 4 channel strain
measurement is desired/required for drilled shafts, sheet piles or spiral weld pipes, or where
measurements are required at two different locations along the shaft. Data is collected and
interpreted on site by the engineer. PDA-W data storage is limited only by free disk space.

PAX -This small unit is battery powered for up to 8 hours operation (with 12 DC input connector
to car battery if extended operation is needed). The touchscreen serves as a user friendly
interface with simple and intuitive menus to guide the user (see the separate manual for PAX
operation). The PAX is a full Windows PC (current preference is XP operating system), with
USB and network ports for data transfer. The (optional) wireless data transmission eliminates
the main cables (a potential source of problems), and speeds the full length monitoring of piles.
The unit operates conventionally with the test engineer on site, or in a “remote” mode with PAX
on site and the engineer in his office to maximize the testing efficiency and minimize testing
costs. The “remote” testing is accomplished through broadband internet connection (see the
PAX manual for instructions). Remote testing allows the test engineer to test multiple piles per
day from different job sites, and then allows immediate processing of the data (no delay due
to travel return to the office) using CAPWAP ®, PDIPLOT and PDI-CURVES so that the
reporting of results is accomplished with minimum delay in time.

1-2 Pile Driving Analyzer Manual, PDA-W February 2009

PAL - This very small unit acquires 2 channels each of strain and acceleration. (Special
adaptor cables are available if 4 channels of strain are required.) The PAL microprocessor
operates a full 8 hours on built-in batteries. The PAL is available in two modes, PAL-L (Local)
and PAL-R (Remote); data from both PAL versions is compatible with PDA-W on a laptop
computer (purchased separately or already owned by the PDA engineer). The PAL is now
obsolete and no longer available for purchase (being replaced by the more versatile PAX).

1. For the PAL-L, the on-site engineer controls all operations. During data collection or review,
the force and velocity data are displayed on the screen. Four results are displayed (e.g.
CSX, TSX, EMX, and RMX) from among thirteen user choices. Data from about 1,600 blows
are stored on removable 40 MB memory cards (the user selects the frequency SX of blows
to be saved; SX1 for restrikes, or SX5 for driving). Larger memory cards, or multiple memory
cards make data storage unlimited. Data must then be “converted” into W01 format for PDA-
W and this conversion process takes about one second per blow (about half an hour for a
full 40 MB memory card). This system is best suited for a small number of impact
blows. Use on larger projects requiring long driving sequences is better given to the PAK,
PAX or the PAL-R.

2. The flexible PAL-R can either (A) acquire and process data by attaching directly to an on-
site laptop PC, or (B) acquire and transmit data by cell phone to the office for true remote
processing, or ( C) simply acquire data in a stand-alone mode with processing to be
performed at a later time. Data from approximately 15,000 blows can be stored on a 40 MB
memory card (note that is almost ten times more than for the PAL-L: data is stored on the
memory card in R01 format which does not need extra time consuming “conversion” and is
directly readable by the PDA-W program).

A. The PAL-R operating in the “on-site mode” is similar to the PDA model PAK. The
engineer takes both the PAL and his laptop PC to the site. Data is transmitted from the
PAL to the PC operating the PDA-W program through a serial cable operating at very
high data transmission rates allowing operation to 60 blows per minute in real time. All
blows can be captured up to speeds of about 100 blows per minute, and stored files with
all blows can be transferred to the PC after the test is completed.

B. In the “remote mode”, the engineer is not on site. Instead, a trained technician or pile
crew attaches the sensors to the test pile (the crew often currently attaches the
sensors). After sensors are connected, the on-site PAL is connected by a cell phone
modem to the expert PDA engineer’s office PC running PDA-W. Turning the PAL on
starts the PAL data transmission. The expert PDA engineer immediately takes over and
controls data acquisition and interpretation. The PDA-W program can optionally
transmit results back to the PAL-R for display to the field personnel. There are
significant cost and time savings. Since all travel is eliminated, office data analysis can
be started immediately after data collection, and decisions can be applied sooner to
benefit the project. Testing can be scheduled at the convenience of the contractor, and
results are generally available sooner with the PAL-R than for other PDA models.

C. In the “stand-alone mode”, the field technician or piling crew attaches the sensors to the
pile. The on-site PAL-R then collects and saves the data to the memory card. The data
can be later sent by standard modem, or the PAL-R or just the memory card can be
taken to the office PC running PDA-W. No results are calculated or displayed anywhere
until the data is eventually transferred to the PDA-W program either locally or via
telephone link.

February 2009 Pile Driving Analyzer Manual, PDA-W 1-3

1.4 Getting Started

The PDA-W software program should be installed on the PDA (hardware model PAK) or on a
personal computer (for PDA hardware model PAX or PAL or for general reprocessing of existing
PDA data files like the previous DOS version “PDAPC”). If installed on a personal computer,
the PDA-W program requires a hardware key (dongle). It is recommended to place the program
and accompanying files in a folder called “PDAWIN”. Using the installation disks (or CD) will
guide you though a proper installation. Help is always available using the F1 function key.

1.4.1 Menu Bar

At the top of the screen, several standard Windows FILE functions such as printing, opening
or closing files are included, as well as EDIT, VIEW, OPTIONS, etc. The Menu Bar of a blank
PDA-W screen (no data file) has less options than when a data file is present.

1.4.2 Status Bar

The Status Bar near the bottom of the screen contains the Save Location (SL) and blow number
(BN) information. The SL value displays the number of blows saved in the data file. For any
data file, you may proceed directly to any of these blows by typing SLxx where xx is the
sequence number (e.g. SL32 will go to the 32nd record in the data file). The Status Bar also
shows the version number of the program when in reprocessing or the ACCEPT/STANDBY
status when collecting data (Accept/Standby is controlled by the F2 function key). Farther
to the right is the user input echo area where the keyboard entries are shown (with a short
“description” if a two-letter command is issued).

1.4.3 Customizing the Toolbars

With VIEW/TOOLBARS you define which toolbars are active. In general, all toolbars should
be active (exception may be the UNITS toolbar if you work exclusively in a single units system).
Toolbars can be moved (via “drag and drop” method) to any location on the screen. Some
could be placed horizontal under the Menu Bar, and others along the left or right edge of the
screen (when acquiring data with a PAK, a special data acquisition toolbar will automatically
appear along the right edge). You should also have the STATUS bar active (checked).

1.4.4 Units

PDA-W will operate in either English (E - kip-sec-ft), Metric (M - tonnes-sec-m), SI (SI - kN-sec-
m), or MKS (MKS - N-sec-m) units. Click the appropriate units on the units toolbar or select
OPTION/UNITS to confirm your selection. All data collected will then have these units. The
units can be changed to another system after data collection by changing the units type. The
units last used will be remembered for the next time you use PDA-W.

If you always work in one units type, you can disable the Units Toolbar (VIEW/TOOLBARS)
to save screen space and avoid accidentally changing to different units.

1.4.5 Data Entry and Two-Letter Commands

Many functions are performed by clicking on an ICON or on making choices from the Menu Bar
and communicating through a dialog box. Communication in this way is standard Windows
operation and really needs no further special explanation. The various functions are described
below. To facilitate program operation some “shortcuts” have been implemented using a brief
input command sequence initiated by direct keyboard entry. These special commands are
often initiated by a “two letter command sequence” sometimes followed by a numerical or
character string entry (although the vast majority of such commands are usually two letters,

1-4 Pile Driving Analyzer Manual, PDA-W February 2009

some commands have three or rarely four letters). The user entry is displayed on the lower
right area of the PDA-W status bar.

The general input is a two letter command followed by input (value or character string) followed
by a "return". The symbol 5 will designate this "return". In some cases, further input is not
required and therefore the two letter command is simply followed by the "5". For example,
typing "AR21.65" will change the area (AR) to 21.6, or typing "TS5" changes the time scale.

Helpful Hint: If you know the variable in one units system, but are working in another, then
appending a units letter (E, M or S) will change the input appropriately. For example, if you
are operating in the English units system, LE17.7M will change LE of 17.7 m into 58 ft.

In a few cases a letter or multi-letter input may follow the initial two letter command and is in
turn followed by the normal 5 return. For example, typing "DPR5" will change the current
graphics display directly to "resistance as a function of time" from whatever is currently
displayed (default is force and velocity). Typing "Q1RMX5" will change the first print quantity
Q1 to the RMX Case Method capacity computation.

For a hypothetical function "NAme" the two letter command is NA. The following command
sequences are allowable:





where ii is any integer ranging from -32000 to 32000, nnn.nn is any real number and can be
entered with or without the decimal (if no decimal is given, then the input is assumed to be the
whole number with a fraction of zero; (400 and 400.00 are equivalent), and lll (i.e., "labels") are
alphanumeric character strings.

Some functions rotate or "toggle" through a predefined sequence, eventually returning to the
starting value. The two letter function designation is then the only required input. For example,
The "TS" command (F8 function key) toggles sequentially through four different time scales.
The F5, F6 and F7 function keys toggle through several popular display options.

After many two letter command inputs, a simple return will execute the same function again.

A brief list of these two letter commands includes the following, and includes in many cases an
alternate method of achieving the same result. No argument implies the function just produces
the desired result directly. After entry of the command and argument (if appropriate), the
command is executed with an <Enter> on the keyboard.

February 2009 Pile Driving Analyzer Manual, PDA-W 1-5

Two Letter Commands (selected list only)

Command Argument Function Alternate Method

OP string OPerator name New (= New Blank Doc.)
PJ string ProJect name New
PN string Pile Name New
PD string Description New
LE value LEngth New
AR value ARea New
AR ARea Calculator
EM value Elastic Modulus New
SP value SPecific weight New
WS value Wave Speed New
WR value Weight of Ram New
WH value Weight of Helmet
JC value Case damping New, J+ J- ICONS
Q1 (also Q2... Q9) string Quantity (result) Menu/View
LP value Length of New, End, BC Switch
LI value add-on Length New
FR value sampling FRequency New
MB value Max Blow rate (BPM) New
VT value Velocity Time shift
DL value DeLay Ins, Del
RF1 (RF2, RV1, value Replay factors AF ICON
FS value Force Scale ICON
VS value Velocity Scale ICON
DS value Displacement Scale ICON
ES value Energy Scale ICON
TS Time Scale T> T< ICONS, [F8]
TB value Time Beginning T- T+ ICONS

1-6 Pile Driving Analyzer Manual, PDA-W February 2009

Command Argument Function Alternate Method
DP? (e.g. DPS) DisPlay mode right click graph,
[F5, F6, F7]
RA Replay Automatic RA ICON
SL value Replay GoTo PgUp, PgDn, RA ICON
Storage Location
CL value Compression Limit
LS value Length to Splice
OF OFfset ICON (balance check)
CT Cal Test ICON
MT Manual Trigger not recommended for
PC string enter comments
UN value UNits ICON

A listing of available two-letter commands is also accessible from the PDA-W program by
selecting View/”KeyBoard Commands” from the Menu Bar. This will bring up your default
browser to display a list with explanations.

1.4.6 Attaching Comments to Data

The user can make a “comment” either during data acquisition or during reprocessing using the
PC two-letter command (with PAX, use the NOTES button). During data acquisition, a
comment entered will be attached to the next blow. If the comment is entered after blows have
stopped, it will be attached to the last blow. During reprocessing, comments are attached to
the current blow being displayed. Comments are displayed in the graph screen in the lower right
corner of the upper data window of PDA-W. Comments are read by the PDIPLOT program for
reporting results. Blows with comments will be saved and never deleted (if you want to delete
a blow with a comment, you must first delete the comment and then you can delete the blow

Comments can be entered in either of two methods:

I Dialog Box type “PC” and then Enter. In the dialog box you may enter and edit a
message. Since only one comment is allowed per blow, you can use
the dialog box option to edit an existing comment

II Direct Entry type the message immediately following the PC (e.g. PCthis is a message)
followed by a Enter when the message is complete (This example message
then appears as this is a message). This is the fastest way to enter
messages in PDA-W or on the PAK if no messages exist already for a blow
(if a message exists, the previous message will be deleted and the new
message applied in its place. Only one message per blow is allowed. Use
the message editor dialog box described above to edit an existing message).

February 2009 Pile Driving Analyzer Manual, PDA-W 1-7

1.4.7 Helpful Hints for Customizing Windows for your PAK (or PC) for PDA-W

Microsoft has decided that by default, file extensions are not shown to users - and this is on a
file system based on extensions! While this seems strange, that's the way it is. With this logic,
a pile with filename “A3-RESTRIKE.W01" appears simply as “A3-RESTRIKE”. If you rename
a file with a hidden extension it will then have two extensions - e.g. renaming
“A3-RESTRIKE[.W01]”, where [.W01] is hidden, to “A3-RESTRIKE.W01" will result in a file
named “A3-RESTRIKE.W01[.W01]”. With the second extension then hidden, this file then
appears to be “A3-RESTRIKE.W01". Notice that this method of handling file names could lead
to some interesting situations. A file named “A3-RESTRIKE.TXT.W01" would be a valid W01
file as far as the PDA is concerned, but would appear to the user as “A3-RESTRIKE.TXT”.
However, to be a true W01 file, it must be created by the PDA-W program. Fortunately there
is a way to enable file extensions in Windows and avoid confusion (and we strongly suggest you
modify your default for your PC and PAK):

- Click on the 'Start Button' and follow the first pop-up menu to 'Settings'
- Follow the second pop-up menu to 'Control Panel' and click on 'Control Panel'
- From the Control Panel, select the 'View' menu bar item, then 'Folder Options'
- Click on the 'View' tab
- Locate the check box that says 'Hide file extensions for known file types'
- Make sure this box is not checked, then press the ‘Apply’ button
- Close Folder Options and Control Panel

After disabling the hidden file extensions, you should be able to detect any files that were
inadvertently renamed with two extensions. These file can be renamed with a single extension
or they can be left as is.

A second helpful feature improves the performance of the mouse or cursor for the PAK
hardware (PAX and PAL each use the touchscreen as a mouse; the PAX can also use an
external mouse connected via the USB port as a backup).

The built-in “mouse” on the PAK is a sealed mini joystick. Experience has shown the user
quickly can adapt to the PAK joystick with only a short adjustment period (with the tip of your
finger placed on the top of the red mouse, slight pressure moves the cursor slowly while larger
pressure moves it quickly). To customize the speed of movement and help locate the
cursor (if the cursor is not readily apparent, it can be found easily by activating the “Locate
Cursor” option; this particular option is available on some PAK units but may not be available
on your PC), do the following:

- Click on the 'Start Button' and follow the first pop-up menu to 'Settings'
- Follow the second pop-up menu to 'Control Panel' and click on 'Control Panel'
- Click on ‘Mouse’
- To set the speed, click on ‘Pointer Speed’ tab
- We suggest ‘Fast’ for Speed, ‘None’ for Acceleration, and ‘Long’ for trail length
- To set the cursor locator (PAK only), click on ‘Pointer Options’ tab
- Place a ‘check’ in the box for ‘Locate Cursor’ (search activated with CTRL and SHIFT)
- Close VersaPoint Control Panel and Control Panel

A third helpful feature is to change the default color of the Display for the PAK (if monochrome
type). The PDA-W screen should normally be a light color, particularly for the PAK in the field.
If it is a dark color, do the following:

- Click on the 'Start Button' and follow the first pop-up menu to 'Settings'
- Follow the second pop-up menu to 'Control Panel' and click on 'Control Panel'
- Click on ‘Display’

1-8 Pile Driving Analyzer Manual, PDA-W February 2009

- Click ‘Appearance’ tab
- Click on ‘Scheme’ to open the selection menu
- Click on ‘High Contrast Black’
- Click on ‘Apply’ and then close Display Properties and Control Panel

1.5 Data Acquisition

1.5.1 Hardware Selection

After installing the software and starting the program, the hardware must be selected by SETUP
on the Menu Bar. Next select SERIAL ANALOG/DIGITAL (SAD) and choose the hardware
option you desire (INTERNAL for PAK, REMOTE for PAL or PAX, or VOID for reprocessing
existing files on a standard PC).

Note that most of the right hand side special function keypad (on old PAK units) does
not work with the PDA-W program (newer PAK models do not have this right side extra
keypad). Everything for the PDA-W program is usually done using the main keyboard
(QWERTY), function keys, and mouse (same tools that are available on your PC version).

1.5.2 Setting Up to Acquire New Data (PAK and Remote PAX/PAL only)

After selecting the Hardware (PAK or REMOTE_PAX/PAL), you must open a new file in which
the data will be placed. Select FILE/NEW from the Menu Bar, or click on the NEW icon (in the
PRINT/EDIT/HELP toolbar which is accessed by VIEW/TOOLBARS). A “WO1 New File Setup”
dialog box will appear. This box has several “pages” of information to be completed. The user
should answer each question.

A. Project Page - this generally contains descriptive information such as names. The Pile
Name (PN) also will double as the name of the file in which the data is kept. The save
frequency (SX) controls how frequently blows will be retained during data collection
(generally SX can be set to the value 1 {one}). The value of SX can be changed at any
time during data collection with the two-letter command SX (e.g. SX4 will change to save
every 4th blow), although there is little reason not to save every blow (SX1) since with
normal archiving and clearing of the large hard disk, memory is not an issue. The result
SQ frequency is normally set to zero since the data can be processed later (the
PDIPLOT program does not require these files; SQ files are only for old DOS programs -
users still in DOS should upgrade to Windows software). The Length Increment (LI) is the
difference value added to the Length of Penetration (LP) value each time the remote blow
count switch is activated or the END key is pressed during data acquisition. (The LP and
LI functions help correlate the pile progress with the visual pile driving record; the pile
driving blow count log can be entered during after processing of data).

After the new entry has been made, from the main data screen the names can be changed
directly. Project Name [PJ] - allows user to change the project name by typing
"PJprojectname5" where “projectname” can be any alpha numeric character string
including spaces and other special characters. Pile Name [PN] - labels all following blows
for identification (default filename for data files for CAPWAP and PDAPLOT) by typing
"PNpilename5". Printed Description [PD] - is used to attach further information to the
data, usually the Pile Driving hammer and/or Pile type used (e.g., Delmag D16,
HP14x89). Activate by typing "PDdescription5" (where “description” is any alpha numeric
label). Operator [OP] - documents the PDA operator collecting or analyzing the data. The
PJ, PN, PD and OP two letter commands (followed by corrected text) can be used after
data is collected to modify names entered on the Project Page.

February 2009 Pile Driving Analyzer Manual, PDA-W 1-9

B. Pile Properties - defines the pile material properties of cross section area at the sensors
(AR), the Length (LE - it is extremely important that this is length from the sensors
to the pile bottom), and material properties (modulus EM (E), weight density SP (ρg),
and wavespeed WS ( c ) ); these three quantities have a known relationship of E = ρc2 and
when any one parameter is changed at least one other parameter will be changed
automatically by the program to keep these parameters in proper relation to each other).
There is an “Area Calc” button to help you calculate the correct pile area (see section
1.7.3). The Area Calculator automatically appears when accessing this tab. These
parameters can later be adjusted by the two-letter commands (e.g. AR400 changes the
Area to 400).

Special Note: For NON-UNIFORM PILES, you should assume a uniform pile with EM and
AR at gages, ignore Case Method results, and model the non-uniform pile with CAPWAP.

The Specific Weight Density (SP) of the pile material in kips/ft3 (English), Ton/m3
(Metric), or kN/m3 (SI). For steel, SP should be .492 kips/ft3, 7.85 Ton/m3 or 77.3 kN/m3.
For normal concrete, values of 0.150 kips/ft3, 2.45 Ton/m3 or 24.0 kN/m3 are common.
For tests on augercast piles, the density of grout is approximately 10% lower than the
density of concrete. For timber, the density is variable and MUST be measured for each

The Wave Speed (WS) in the pile in either ft/sec (English) or in meters/sec (Metric or SI).
For steel piles the wavespeed is approximately 16,807 ft/sec or 5,123 m/sec. For
concrete or timber the wavespeed is variable and should be determined for each pile on
an individual basis; values between 10,000 and 15,000 ft/sec or 3,000 to 4,500 m/sec
are common.

The Elastic Modulus (EM) of the pile in ksi (English), Ton/cm2 (Metric) or MPa (SI). The
modulus "EM" will be recomputed from WS and SP if either of these change. If the user
changes EM, the PDA-W program will recompute the WS.

For an elastic pile, the wavespeed, modulus and density are related by
EM = ρc2 = (SP/g) (WS)2
where ρ is the MASS DENSITY, c the wavespeed, and g is the gravitational constant (32.2
ft/sec2 or 9.81 m/sec2) and dimension conversions are required to obtain the modulus in
the correct units.

Variable Steel (precise) Concrete (typical) Timber (typical)

30,000 ksi 5,000* ksi 2000* ksi
EM 2,100 T/cm2 400* T/cm2 160* T/cm2
210,000 MPa 40,000* MPa 16,000* MPa
SP .492 k/ft3 .150 k/ft3 .060* k/ft3
7.85 T/m3 2.45 T/m3 1.0* T/m3
77.3 kN/m3 24.0 kN/m3 10.0* kN/m3
WS,WC 16,800 ft/sec 12,400* ft/sec 12,400* ft/sec
5,120 m/sec 4,000* m/sec 3,960* m/sec

* - Result highly variable; MUST be measured for each pile.

The PDA-W program assures the elastic equality for EM, WS and SP. If any of the three
variables are changed, the corresponding complimentary variable is automatically
changed (changing EM changes WS; Changing WS or SP changes EM). For example if

1-10 Pile Driving Analyzer Manual, PDA-W February 2009

WS is measured at 4000 m/sec and the density is assumed to be 2.45 Ton/m3, then the
elastic modulus will be computed automatically to be 400 Ton/cm2.

LS - The Length to Splice can be entered as a two letter command (e.g. LS23 will place
a marker at length 23 from the sensor location) and a vertical line will appear at this depth
(below sensors) on the graph (at time 2*LS / WC after the initial rise marker). This line
can be deleted by entering LS with no value. While this LS cannot be entered as a new
page, it is mentioned here as it relates to the pile length LE and can be useful for
segmental piles.

C. Hammer - the hammer can be selected from LIST (by clicking the appropriate hammer)
or by CUSTOM (user defines the hammer type, energy rating, ram weight, and description
of hammer type like diesel, air, hydraulic et al). The list is identical to the list in the
GRLWEAP program.

There is also a “Max Blows/Minute” entry field. The value entered should correspond
to slightly higher than the maximum operating rate for the hammer. This will avoid
allowing the PDA to trigger on false blows due to hammer bounces, particularly for
hydraulic hammers or when single blow drop hammers are used. The MB will be
limited to 200 blows per minute (BPM) at a sampling rate of 20 KHz or 10 KHz, to 133
BPM at 6.66 KHz, 109 BPM at 5 KHz, 85 BPM at 4 KHz, and 63 BPM at 3.33 KHz. The
MB rate can be later changed with the MB two letter command function followed by the
value (in blows per minute). The BN Filter function (section 7.9) can be used if MB
was improperly set during data acquisition to eliminate “bounce blows” if MB was set
too high for drop hammers, or to renumber the blows if MB was set too low for double
acting hammers (although it will not restore missing blows).

WR - The Weight of Ram is entered only for further hammer performance analysis or if
the ram velocity at impact VRI and/or cushion stiffness (KCP) are to be calculated, or for
the F=ma analysis (measuring the force by instrumenting the ram with an accelerometer).
Input the ram weight in the same force units as contained in modulus of elasticity (e.g.
kips for modulus in ksi) of as contained in the FMX output.

WH - The Weight of Helmet (Drive Cap) in force units consistent with the force output
is required only for hammer cushion analysis (FCP and KCP quantities) and is applicable
only to air, steam, hydraulic, or drop hammers on steel piles. WH can only be input or
changed in the main data collection screen (but is mentioned only here as it is a related
topic). The WH is also important when performing the F=ma data analysis (measuring the
force by instrumenting the ram with an accelerometer).

D. F/V Sensors - User selects accelerometer type, calibration units, trigger channel (Trig),
digitizing frequency, active sensors (check boxes), and enters the calibration of the
sensors (by direct entry or by the “SELECT” feature which retains calibrations by sensor
serial number).

The SELECT function has edit capability for adding, changing or deleting sensors from
a data file; to add a sensor, click the NEW button and input the sensor ID and its
calibration, and its type if the sensor is an accelerometer. To edit a sensor, click on EDIT
and change the value. To delete a sensor click DELETE). The SELECT feature is
recommended since the sensor calibrations are then stored with your PDA.

Based on experimental results, the calibration accuracy of the PDI calibrations for
both strain and acceleration sensors is approximately 2% or less. At low
temperatures (about 32 degrees F or 0 degrees C), the PE acceleration calibration
in g/v should be decreased by about 2% according to the manufacturer.

February 2009 Pile Driving Analyzer Manual, PDA-W 1-11

If not using “smart sensors” and due to non-compensated hookup, the strain
sensitivity when using PR accelerometers is changed slightly by the cable length
(correction is not applicable for strains when using PE accelerometers as it is a
compensated hookup). The calibration factor should be increased by about 1.5%
for each 100 ft (30 m) of main cable. This calibration is compensated for when
using smart sensors and the correction does not then apply.

Calibration adjustment factors are small and can be applied after data collection
using the replay factors in the AF Icon dialog box. (See section 1.7.5 Changing
Calibrations (or Replay Factors) in Existing Files)

PAX units with wireless operation have “smart sensor” ability. Smart sensors
remember their own serial number and calibration constant and tell the PAX this
information automatically, so it cannot be wrong, and thus eliminates common
mistakes. Changing sensors does however require the user to request new
calibrations (see PAX manual and the CALIB button).

Frequency - the sampling rate per channel for the data. Values of 20,000*, 10,000, 6667,
5000, 4000 and 3333 only are acceptable for the PAK. (For PAX data, the PAX-L sets the
frequency while the PAX-R has the PDA-W in control of this selection.) To change directly
to a specific frequency, simply specify that frequency (i.e., "FR50005" specifies a
frequency of 5,000. We recommend a 5,000 Hz frequency for most PAK data
acquisition (using the PAK’s 1023 integer sample size results in a total time record
of 204.8 msec). If the last part of the record has no real activity ( e.g. is "straight" or
“flat”), then a higher FR value could be chosen (e.g. 6667 or 10,000). For very long piles,
the velocity takes longer to stabilize and a lower frequency (longer time period) is
required. The 20,000 Hz frequency may be useful for testing short SPT rods. The table
gives the total record duration in milliseconds.

Frequency Milliseconds
Samples/sec for PAK data
Hz 10-3 sec
20,000 51.2

10,000 102.4

6,667 153.6

5,000 204.8

4,000 256.0

3,333 307.2

The Sampling frequency for the PAX may result in longer times since the sample record length
might be either 1024 samples, or 2048 samples, or 4096 samples. A selection of a 10,000
sampling frequency with a 2048 record length is common for the PAX, giving a total sample
time of 204.8 msec. The greater sampling rate gives greater precision on pile length (or
damage at the pile toe).

Accelerometer Type - The PAK (or PAX) has two piezoelectric and two piezoelectric
acceleration channels and a total of four strain channels. The accelerometer type is
selected by clicking the accelerometer type between piezoelectric [A1, A2, F1, and F2]

1-12 Pile Driving Analyzer Manual, PDA-W February 2009

or piezoresistive [A3, A4, F3, and F4] accelerometer (or both types in 8 channel
operation). If using piezoelectric, the main cable attaches to the input plug on the
right of the PAK or PAX (right as defined as looking at the screen from the front).
If using piezoresistive, attach to the left input plug. In 8 channel operation, both
cables are attached. For 4 channel PAX or PAL units (where the accelerometer type is
chosen in advance at time of manufacturer), the cable connection is obvious. Most now
prefer the “twist-lock” 1/4 turn connectors as superior. There are older cable/sensor
systems with a “treaded” connection, and an “adaptor cable” is available on request). PR
accelerometers are the generally preferred PDI recommendation as overall they give
better data quality in steel-to-steel impact situations; PE accelerometers are usually
sufficient for cushioned concrete pile testing.

Piezoelectric accelerometers (long time PDI standard) generally work well for concrete
and timber piles, and for steel piles driven by hammers with a hammer cushion. For
steel on steel impacts, Piezoelectric Accelerometers on aluminum blocks give
acceptable results in many, but not all, situations. In the most severe situations, PE
accelerometers may fail to perform satisfactorily. Often, these accelerometers do not
work for SPT testing.

Piezoresistive accelerometers are intended for high acceleration applications such as

steel piles driven by hammers without cushions, SPT soil samplers or other steel to
steel impacts. Also data quality is generally better for diesel hammer applications.
The piezoresistive accelerometers will function well in lower acceleration applications.

Warning: It is crucial that you have the correct accelerometer type selected for the
sensors you plan to use! The sensors MUST also be connected to the correct input
connector for the PAK (there is only one sensor type and one connector for PAX4
or PAL PDA units). For the PAK, piezoelectric sensors (PE) should be attached to
the right hand side connector, and piezoresistive accelerometers (PR) should be
connected to the left side connector - side defined as you are viewing the display
screen). If you need to change sensor type during a test, you MUST close the
current file and open a new file (and change the type). The only real limitation then
is that files of different accelerometer types cannot be “merged” later into a single
data file. The PAX-4 or PAL operates on either one accelerometer type or the other
depending on how the internal hardware was initially built and ordered. The
“welcome screen” informs the user of the accelerometer type required. This
accelerometer type must be used with the PAX-4 or PAL or it will not work.

Vibratory Hammers - Note, selecting Vibratory data disables the balancing features.
Pressing the END key or the blow switch will acquire one set of data (may be several
cycles of data but all will be stored in one SL save location). There are some special
quantity results.

E. Geophone Sensors - an inactive feature at this time.

Upon completion of all pages described above click on OK to continue. PDA-W then displays
the working graph screen. The PAK has then prepared a file to accept blows. For the remote
PAX or PAL-R, going through this process is needed to set the sensor calibrations, project
descriptions, pile properties, etc. The PDA-W will default to the last settings used when the
remote PAX or PAL-R begins data transmission. The program will also return to these settings
when a NEW pile is requested.

February 2009 Pile Driving Analyzer Manual, PDA-W 1-13

1.5.3 Basic Operation Summary for Collecting Data

A. PREPARE PILE by drilling holes about 2 diameters below top (2 holes per strain
transducer, 1 hole per accelerometer, 1 hole to support cable - cable may be supported
together with an accelerometer but not together with a strain sensor). Use sets of gages
diametrically placed about neutral axis to cancel bending. [all transducers use 1/4 inch
(6mm) bolts for quick attachment]

For concrete piles: drill first hole, install anchor firmly, attach drill template, and then
drill second hole for strain transducer; install concrete anchors (USA: use 3/8 inch
drill bit for 3/8-20 anchors for the 1/4-20 inch bolts) with a setting tool making sure
they are firmly attached (attach sensor and tighten a bolt; anchor should not be
loose). If the anchor is not firmly seated, the bolt cannot be tightened, so the
anchor will slip and cause bad data. It is imperative that the anchor be firmly

For steel pipes: drill and tap holes (USA: use a 7/32 inch bit and tap for 1/4-20

For H piles: use clearance holes in center of web. (USA: use 5/16 inch bit).

B. If you are using a PAK or Remote PAX/PAL, make a New page which creates a new file
for the next pile (see also section 1.5.2 Setting up to Acquire New Data). Enter data for
project, pile name (used for file storage), ARea, LEngth (below sensors to pile bottom),
SP (specific weight), and WS (wave speed) or EM modulus. Enter Hammer information
(custom or from list). Enter transducer calibrations (directly or SELECT from list). Note:
if you are using the PAX in local mode, please refer to the PAX model for instructions on
how to start a new pile.

C. CHECK TRANSDUCERS by attaching strain and acceleration transducers to "connection

cable" and into long 19 pin "main cable" which is attached to the PDA (for PAK, PE
accelerometer attach on right; PR accelerometers attach to left). Sequentially set the
trigger channel to each sensor and tap each sensor to check for response (this quick test
assures that the sensor works and saves potentially large delays later). Check the Offset
(section 1.5.6 Offset Test) for strain transducer balance [-5v to +4v acceptable].
Disconnect the main cable from PDA while attaching transducers.

If PAX wireless system is used, attach sensors to the transmitters, turn on the
transmitters and establish connection with the PAX. Tap sensors, and change trigger
channel to assure all sensors are working. Checking the CAL TEST pulse (that all
signals are active) and BALANCE offset may be sufficient if same sensors are always
used and they performed satisfactorily on previous pile test. See the PAX manual for
further details.

D. ATTACH TRANSDUCERS TO PILE after pile is in vertical position in leads (we suggest
till pile top is near ground to reduce climb. An even better suggestion is to use the new
PAX wireless data acquisition system and attach sensors on ground prior to lifting the
pile). Tighten all bolts (loose bolts will produce poor data and damage the sensors).
Check Offset (1.5.6 Offset Test) strain balance again.

Note that the main cable should be supported so that no extra weight is placed on the
sensor connectors and sensor cables. It should be oriented to the open side of the
leads so that as the pile is driven it will freely follow the pile top down the ground and
not become tight and break.

1-14 Pile Driving Analyzer Manual, PDA-W February 2009

Sensors can be attached on the ground prior to lifting the pile into the leads. For H
piles, a small hole can be cut into the web to allow the cables to pass through without
going around the flanges. The flanges themselves protect the sensors sufficiently well.

For steel pipe, timber and concrete piles, PDI has designed “sensor protectors”
to cover and protect the sensors during the lifting process and many find them
very helpful to saving time in attachment for testing. Because you attach the
sensors, you know they are right. This also saves considerable time by installing the
sensors prior to lifting the pile, thus not having to install the sensors in the air up in the
leads after lifting the pile, and improves safety. The PAX wireless system (lacking any
connection cables between sensors or the PAX) is fully protected by these “sensor
protectors” and is the standard operation for PAX wireless.

E. PAK and Remote PAX/PAL only - Select Quantity Results (Q1, Q2.., Q9) to calculate.
Set "SX" for frequency of blows to save. Generally for the PDA-W program usually SX
is always set to one (1) since storage is generally unlimited. After acquisition and final
processing for any job, the amount of data to be permanently saved can be quickly
reduced (1.6.2 Compressing Data Files).

F. PAK and Remote PAX/PAL only - To acquire data (1.5.4 Accepting New Data), put PDA
in ACCEPT mode using F2 function key, type “MT” (Manual Trigger) to assure PDA is in
active state (MT should be used only with PAK; use CT for PAX or PAL remote tests), start
hammer and take data. After test, press F2 to place PDA in STANDBY and remove
transducers. For remote PAX or PAL-R, press Collect Data and follow instructions and
at end of data collection press “Done”.

G. Save data file with FILES/SAVE menu selection (1.6 Saving Data). For old PAK units
with floppy disk only, up to 172 blows/file will fit on a 1.4 MB floppy disk in normal data
storage mode. New PAK units download the data for permanent storage either though the
USB port or the Ethernet port, which is fast and can take unlimited data size. The file size
can be compressed in two ways:

with the W- ICON (1.6.2 Compressing Data Files; as a minimum, save several
consecutive blows at end of drive; or several at begin of restrike. If the file has too
many blows for a floppy disk, set the “save frequency” to eliminate some blows during
long sequences). The number of blows to be saved and the estimated file size will be
shown to guide the user in this selection. Transfer the data to floppy or CD writer or
USB or Ethernet for new PAK units, or from memory card (PAL) with Explorer for
permanent saving of data.

The data itself can be subjected to a “zero loss data compression routine” (like a ZIP
file) by accessing FILE/OPTIONS and then checking the “Compress W01 Files”
selection. This is highly recommended since it reduces file size with no loss of
data. In this way, about 3 times the number of blows will fit onto a single floppy disk
(e.g. about 500 blows per 1.4 MB floppy disk), and take less space on hard disks or
USB memory sticks (or whatever your choice for more permanent storage of data. The
exact number of blows depends on the smoothness of the data itself.

1.5.4 Accepting New Data: PAK

To acquire data the PAK must be in accept mode (see status bar indicator). To change from
STANDBY to ACCEPT either press the F2 function key, or the ACCEPT/STANDBY ICON. In
ACCEPT mode, first type “CT” or “MT” (Manual Trigger) to assure the PAK will accept
new data. (The MT function should be avoided for the PAL). Then the hammer may be started

February 2009 Pile Driving Analyzer Manual, PDA-W 1-15

and impacts applied. The frequency of saving results is controlled by the SX command. There
is not much reason to save anything less than every blow initially (SX1). Extra data can be
eliminated later (1.6.2 Reducing Data Files).

Changing pile variables (e.g. AR, LE, WS, etc.) will ask if the change it to apply to all data
(including data already collected) or only to new data after the change request (subsequent
blows only).

The overall wavespeed can be changed blow by blow using the left and right arrows. These
values will be retained in the data (blow by blow) for reprocessing (see section 1.7.6 for further
information). Accepting New Data: PAX

A separate manual specifically for the PAX is available to describe all modes of data acquisition
by the PAX: conventional on-site with cables, on-site with wireless, or remote.

1.5.5 Accepting New Data: PAL-R Connected to the PDA-W Program

Turn the PAL-R on. After the welcome screen, the PAL-R soon automatically goes to the Main
Menu (do not touch the screen during the PDI logo welcome screen is active). There are a
few simple steps that need to be followed.

A. The PDA-W program must be in the Remote Mode (1.5.1 Hardware Selection). The
baud rate for the PC must be set using the “PAK-ette I/O Monitor”. Maximize the PAK-ette
and then select SETUP in the dialog box, and then select SERIAL PORT, and finally
choose BAUD and make your selection (usually 9600 for cell phones, and up to 115,200
for direct connect: 57,600 is recommended for most direct connect applications).
Minimize the PAK-ette window after selection is complete.

The SETUP menu selection also allows you to set the AUDIO options. You can make
the PDA-W program produce various sounds when various actions are encountered.
To activate the sounds you also must have a sound card and must activate the
SADCOM sounds in the Control Panel (Sounds).

Most of the other PAK-ette functions probably need no adjustment. (Many are for
diagnostic use).

B. The PAL-R Main Menu SETTINGS are used to enter a Project name, change Units type,
Keyboard definition (QWERTY or ABCDEF; depends on familiarity of the user with
standard keyboards and typing skills), or allow user to customize his standard messages
(press on any field and edit the message). The PDA-W program uses its own default
settings (values from last use) for a new file (sensor calibrations, pile area AR, modulus
EM, density SP, wavespeed WS, etc) as outlined in Section 1.5.2 Setting up to Acquire
New Data. The PDA-W data can be edited later if any parameter is in error and needs
correcting (e.g. AR, WS, LE etc or even the pile name PN). Data is always collected by
the PAL-R at 5,000 Hz.

C. Initially the PAL-R is OFFLINE, meaning it is not connected by direct cable to a PC or to

a cell phone modem system. To establish connection to the cell phone, connect the serial
output cable from the data capable cell phone to the RS232 connector on the PAL-R.
Turn on the cell phone (batteries should be charged; some phones allow simultaneous
connection to a charger or car 12 volt DC auxiliary power jack and to data connection
cables while some do not). Many phones require connection at specific baud rate to the
PAL-L (19200 is a popular choice as of year 2002; check with your phone supplier for this

1-16 Pile Driving Analyzer Manual, PDA-W February 2009

information). Change the PAL-R baud rate in SETTINGS to match this required phone
transmission rate (this step allows the phone to digitally communicate with the PAL).

D. Press CONNECT to bring up connection choices (phone numbers and baud rates).

E. In the CONNECT sub-menu, press SELECT for the desired choice. Note that to achieve
a connection that the selection must match the Baud rate for the data capable phone
above. Once a connection has been made, the Main Menu reappears and ONLINE
should be observed when a direct connection to a PC running PDA-W via either a cable
or a cell phone has been confirmed by the PAL.

Connection selections are altered by pressing CHANGE.

1. A ‘Connection Name’ can be entered (e.g. ‘PC’ if direct connect, or ‘Office/modem’ if

a cell phone connection is to be made.

2. If you want a cell phone communication, next enter the phone number to dial. This
phone should be connected to a modem which is connected to a PC running PDA-W.
If a phone number is entered, an ACCESS CODE is then requested to assure digital
transmission (experience suggests this is not necessary and maybe not even
desirable, and leaving the access code blank is perhaps best. For USA and Airtouch
our local carrier suggests *3282, although we caution against this. Consult your local
cell phone carrier). We also suggest a power booster for your antenna (contact PDI
for further details). If you are having trouble connecting, make sure the PAL is outside
the vehicle, or move the vehicle slightly to another location to improve reception.
There are times when the cell towers are busy, and simply retrying the connection (dial
again) is necessary.

3. If you want to connect directly to a laptop PC in the field running PDA-W do not enter
any phone number (blank entry defines as a “direct connect”and assumes PAL-R will
be connected by a short RS232 serial cable to the serial port of a laptop in the field;
note it is now difficult to find laptops with serial ports).

4. The Baud Rate is selected next from among several choices (usually 9600 or 19200
for cell phones in USA; faster baud rates up to 115,200 can be used for direct
connect). See the PAK-ette Monitor for available baud rates. The baud rate of the
PAL-R must match that for the data capable cell phone. Press “EXIT” to continue.

5. Last entry is the INIT STRING (which may be left blank).

F. Messages can be sent to the PDA-W “Message Exchange” by pressing SEND MESG.
Make selection of a preprogrammed message, or create a “custom text”. EXIT from the
message monitor when you finish sending messages. Messages sent are confirmed if
successful. Note the cell phone may be used offline to communicate either before or after
the test, or (preferred) a second cell phone can be used during the test to keep in contact.
The Message Exchange feature allows limited communication which may often prove to
be beneficial and sufficient. On most projects, a second cell phone to maintain voice
communication is preferred if possible and available.

The message from the PAL-R may even be heard verbally if you have an authorized
speech module (e.g. “Text-to-Speech”)

February 2009 Pile Driving Analyzer Manual, PDA-W 1-17

G. To acquire new data, press COLLECT. A brief menu for limited entry/selection is shown.

º Enter the Pile Name to be tested (Press the + increment, or the - to decrement the
current name, or press the words PILE NAME to make a new entry),

º Enter the pile length (from sensors to pile bottom),

º Enter the Drive Type,

º NORMAL - this Drive Type option is used during a test during driving which will
send the next current blow available and may skip some blows during sending
so you are left with a sample of all data but are seeing the most recent blow
available (although it should be noted that ALL blows are stored on the memory
card and are thus available for later retrieval), or

º RESTRIKE - This Drive Type option is for a restrike to send every blow for a
limited number of blows although sending will fall behind data collection. Use
this option only if you plan on a limited number of blows for the entire test.

º Enter the “Starting LP” - this documents the current length pile penetration below the
reference elevation. If you do not plan to use the remote blow count entry switch,
we recommend you leave this blank (zero).

º Enter the “LP Increment” - if the blow count switch is to be used, enter the value for
depth increment to be recorded (e.g. one foot or 0.25 m). Press the blow count switch
to increment LP by the “LP increment”. If you do not plan to use the blow switch (or
if data is not a restrike), then leave this value zero.

º Enter the “LP Alt-Increment - if another increment may be chosen for final control (e.g.
one inch or 25 mm)

º Press COLLECT to begin data collection.

H. When entering “Collect”, the PAL-R first balances the sensors. WAITING FOR INPUT
message is observed (this may take a few seconds; balancing requires that the sensors
be connected prior to starting this COLLECT request. The “gage offsets” are shown
for the two sensors prior to acquiring data. If the sensors do not stabilize to within +4 to
-5 volts, then the sensor should be changed or the attachment adjusted so that it does fall
within this range.

I. The PAL-R goes to ACCEPT (lower left corner). Set the PDA-W program to ACCEPT
using the F2 function key. When an impact is detected, the PAL-R displays data and
sends the data to the PDA-W program (either through the direct connect cable or via cell

Basically the PDA-W program is in charge at this point in selection of the trigger
channel and the saving frequency (SX). We would suggest that you always consider
SX1 as the appropriate selection frequency (save as many blows as you can; you can
always reduce the file size later).

To replace sensors or deactivate, press ACCEPT and the PAL-R will then go to
STANDBY (or press F2 function key from PDA-W). This prevents false blows from
being acquired. Make sure you return to ACCEPT when ready to get new data.

1-18 Pile Driving Analyzer Manual, PDA-W February 2009

We suggest that after connect and balancing, that a “Calibration Test” be made
to assure sensors are working correctly. Press the CT icon on the PDA-W
program (or type “CT” on the keyboard and Enter). A square and triangle shape
should be observed (if the “triangle is “distorted”, press CT again until a triangle
is observed). The triangle (velocity) is only available for PE type accelerometers.

J. Various information is shown at the top of the PAL-R including Pile Name (PN), Project
(PJ), Blow Number (BLOW), Blows Per Minute (BPM), and LP and LI. The data
connection and sending status is shown in the box just above the SEND MESG control.
Messages can be sent during data collection using SEND MESG (Messages sent by the
PDA-W program will be displayed until acknowledged; messages que). The type of data
is controlled by the SHOW box (VF is preferred as it shows the average V and average
F; both are unscaled and may not show proportionality). The vertical scale can be
magnified by pressing the V ZOOM box, and the horizontal scale can be selected by the
H ZOOM box. The trigger channel can be selected for Standalone mode with the TRIG
box (the PDA-W program controls the PAL-R if there is a connection for real time

K. The quality of the cell phone line affects the time to send a blow. With a good connection,
data will be sent relatively quickly. With a marginal connection, errors can be generated.
If no phone connection can be achieved, try to place the phone outside the vehicle,
or move the vehicle to another nearby location to avoid interference, or simply try
again (cell phone system may be temporarily overloaded with other calls). If an error
is made, the PDA-W program asks the PAL-R to resend the information; the PDA-W
program and the PAL-R will continue to request the same data until no error was made
and thus assuring good data is acquired. The PAKETTE monitor shows the data
transmission activity in its upper window (minimize the size to show only this
activity). If cell phone service is poor, then it may even drop the call phone connection;
if the connection is lost, then the PAL-R will automatically attempt to call again to re-
establish connection and resume sending data. Using a power booster for the antenna
will aid transmission in weak signal areas (contact PDI for details).

L. When the data set is complete (pile driven, or restrike finished), press “DONE?”. This
ends data collection for that file and returns to the Main Menu.

During data collection at slow baud rates and for faster hammers even though the
PAL-R may collect every blow, it may not be fast enough to send every blow (in
NORMAL mode, the PAL-R sends the next available blow so you are always as current
as possible). At typical cell phone 9,600 baud rate, it takes about 5 seconds to send
one blow of data.

M. In the Main Menu, the FILES option in PDA-W allows deletion of all stored files, or
selection of specific files to be sent to the PDA-W program (PAL-R must be connected to
PDA-W for successful transmission).

º To select a desired file, use the (page up, page down) or (line up, line down) to
highlight the desired file (pile name). The date of data collection and approximate file
size are displayed (number of blows can be approximated by dividing file size by about

º press SELECT FILE.

º press SEND to transfer the data to PDA-W, or DELETE to erase the selected file, or
UNSELECT to return to the file manager.

February 2009 Pile Driving Analyzer Manual, PDA-W 1-19

º When all file operations are completed, press EXIT to return to the Main Menu.

N. If the PAL-R is in its MAIN MENU, then the PDA-W operator can select FILES and then
PAL DIRECTORY to review data available. After selecting a desired file, it can be
downloaded in entirety or partial files can be requested for transmission (e.g. last 10
blows at end of drive from a long sequence) or some frequency of data (e.g. every 10th
blow). This is a very useful feature.

O. When finished, press DISCONNECT to go OFFLINE (hangs up the cell phone; turn the
cell phone off when finished to avoid excess connect charges). It is suggested that you
disconnect the phone after each pile and reconnect just prior to the next pile driving to
minimize your cell phone hook up time and charges.

P. If there is a problem with connection, data can be collected in the STANDALONE mode
in the COLLECT sub-menu and transferred at a later time to the PDA-W program on a PC
using the file manager in (L or M) above, or using the PCMCIA card directly inserted into
a PC with (P) below.

Alternately, data collected in the standalone mode or the “normal mode” with real-time
transmission can be sent (or resent) after connecting the PAL to a cell phone to the
office PC running PDA-W in an “office controlled” manner. The PAL-R must be in
the main menu. The office engineer the selects FILE / PAL-R DIRECTORY. A list of
the available files in the PAL-R memory card is the shown. The engineer can the
select a specific number of blows at the beginning and/or end of the data file and a
representative sample frequency in between (e.g. 3 selects every third blow only) to
be sent to him. In this way the amount of data can be reduced while still retaining the
most important data from the begin of restrike or the end of drive. This is particularly
helpful to retrieve the last 10 or 20 blows at the end of driving, or all early restrike
Q. Collected data is saved on the memory card in “R01" or “R02" files in a compressed
format. After data collection is completed, the PAL-R can be turned off and the card can
be removed and then inserted into a computer PCMCIA slot. (Files downloaded from
the card to a PC can be emailed as another alternative). This R01 or R02 data can
then be read directly from the memory card files by the PDA-W program. To accomplish
this, select FILE/”Open R01 File” and select the appropriate file from the memory card or
from your hard disk if you have transferred the R01 or R02 data there. This will bring up
a “new page” and that other ”missing” information must be entered (e.g. like pile area AR
and sensor calibrations).

R. ”Results” from the various PDA-W computations (CSX, TSX, EMX, BTA and Capacity) can
be transmitted back to the PAL-R for display in the field by clicking the SETUP/”Remote
PAL Results” from the Menu Bar. Then check the results you want displayed in the field.
If you want to hide results from the field personnel, then make sure the checkboxes are
blank. It would be a good idea to always transmit the “warnings” and “velocity scale” to
alert the user to potential data quality problems and to display the data proportionally.

1.5.6 Offset Test (PAK and Remote PAX/PAL only)

Prior to acquiring data, you should first check the strain sensor OFFSET using the OFFSET
ICON or selecting SETUP/”BALANCE CHECK (OFFSET)” from the Menu Bar (“Offsets” are
usually displayed in “volts” and values from -5.0 to +4.0 are generally safe; if outside this range
you may want to loosen that sensor and adjust (stretching or compressing the sensor) prior to
data collection to prevent “clipped signals” in the A/D process which make the data useless).
Negative values indicate the sensor is currently in a tension condition; positive values indicate

1-20 Pile Driving Analyzer Manual, PDA-W February 2009

the sensor is in a compression condition. The main cause of bad offsets is either non-flat
mounting surfaces or incorrect spacings between the holes (3 inch or 76 mm). It is
imperative that the hole spacing be correct for the strain sensors to function properly.
Use a drilling template on concrete piles.

1.5.7 Calibration Test (PAK and Remote PAX/PAL only)

A Calibration Test should be performed by clicking the CALIBRATION TEST ICON (or typing
CT) after the sensors are attached to the pile but prior to any test. This will confirm that the
PDA hardware is in calibration and that the sensors should work when the impact is applied.

1.5.8 Changing Sensors or Sensor Status During a Test (PAK and Remote PAX/PAL only)

If a sensor fails during data collection, the hammer should be stopped and the sensor replaced.
This also requires changing the calibration entry for that sensor. This is accomplished by
clicking on the sensor ICON for that changed sensor (lower right edge of PDA-W screen on the
Sensor Status Bar).

This Sensor Status Bar also has check boxes to define active sensors which could be used to
turn off one accelerometer (this is often acceptable since the two velocity records are usually
very similar for good data; it is almost never acceptable to turn off one strain since bending is
often present so a defective strain sensor should be replaced; only exception is perhaps for H-
piles when the sensor is a attached to the center of the web, but even then bending must be
small or errors will be generated by using only a single strain sensor).

The trigger channel can be changed by the radio button selection near the center of this Bar.
This sensor must be active to cause the PDA to acquire data. Selecting an accelerometer is
often the preferable choice (since bending can be uncertain).

1.5.9 Recording the Penetration Depth LP (PAK and Remote PAX/PAL only)

The PDA data for piles monitored during installation can be documented as a function of pile
depth below some reference using the LP value (“Length of Penetration”). The initial value of
LP at the beginning of driving is entered on the New Page setup dialog box under the Project
information (or can be added later with the “LPvalue” command). The Length Increment (LI),
which is the unit depth the blows will be counted for (e.g. one foot or 0.25 m, etc.) should also
be entered.

The “blow count switch” should be connected to the PDA. It has a “click button” which should
then be pressed each time a “unit penetration” has been achieved during driving (e.g. once
each foot or once each 0.25 m). The same person visually logging the pile and recording the
drive log can be assigned to press this blow count switch once for each time he records an
entry to his drive log. The current LP value will then be incremented by LI.

LP can also be incremented by pressing the END key rather than the blow count switch.

LP The Length Penetrated in ft or meters. An input of "LP245" would set the current
penetration of the pile into the soil at 24 (ft or meters).

LI The Length Increment of penetration is added to the current LP value each time the END
key or blow switch is depressed. The default value is 1.0; if any other increment is used,
the LI value is shown. For example, for penetration recorded in inches, the LI value
should be set to "LI1/125" since this represents 1/12 of a ft.

February 2009 Pile Driving Analyzer Manual, PDA-W 1-21

The "LI" entry sets up the "alternate LI value" (i.e., LI.25 displays LI0.25). Entry of
“LI7” will switch back to LI1.0, thus LI can be toggled back and forth between fine and
coarse increments.

1.5.10 Determination of Wave Speed

Pile wave speed WS (and modulus EM) must be correct for correct evaluation of the data,
especially for capacity determination. For steel, wave speed is 16,807 ft/sec (5123 m/sec).
The wave speed for concrete and timber must be determined for each pile. Choosing the wrong
wave speed results in force, stress, energy and capacity calculation errors. Thus proper
wavespeed determination is critical to successful dynamic pile testing. During Driving

If wave up indicates some tension reflection (local "valley" in Wave Up at 2L/C), wave speed
determination is possible. Records during easy driving (e.g. low blow counts, or high set per
blow) are best because the tension return from the pile toe is most obvious. Investigate the WD
and WU screen display. Use the correct length below gages (LE). The first rise time marker
is automatically positioned at time A; shift the second dashed rise time marker (using left and
right arrows) to the time B (the beginning of the wave up valley at 2L/C). Time marker A is
program selected and cannot be changed. If you would like to select another rise time, then
shift the T1 marker using the INS and DEL keys. (Restore the T1 marker to the peak using DL0
command.) The PDA calculates the computed wave speed WC; enter this value into WS.

Note that the peak markers T1 and T2 do not necessarily align with the initial peak and
the reflected “valley” when the resistance is moderate, or if the pile has minor tension
cracking. The T1 and T2 markers should be aligned with the initial peak and the
corresponding “valley” if (and only if) there is a very easy driving condition (large set
per blow) and the Wave Up shape (and duration) at 2L/c mirrors the Wave Down
shape (and duration) at the initial peak. By Calculation

If the sensors are attached to the pile at a location far above ground location, and if
the pile is uniform, then if the pile should be perfectly proportional, compare VT1
times EA/c with the FT1. Take the ratio and change the WS by this ratio and try again.
This method may work for hydraulic, air or drop hammers, but will not work well for
diesel hammers due to the long gas precompression phase of the diesel cycle. By Wave Up Inspection

In any case, including for diesel hammers, the wave up curve should be smooth
through the peak velocity input time for uniform piles. There should be no steps or
jumps through the period of time from the initial rise to the peak input. If the wave up

1-22 Pile Driving Analyzer Manual, PDA-W February 2009

is too low at this time (local valley), then the WS is probably too low. If the wave up
is too high (local increase), then WS is probably too high. Phase shifts should be
minimized (eliminated) using VT prior to assessing the wave up curve. (See section
2.5 of printed manual for further information and illustrating figures).

1.5.11 Elastic Modulus

The PDA calculates elastic modulus EM automatically from the wave speed WS and specific
weight SP. The elastic modulus for steel is (30,000 ksi, 210,000 MPa, 2100 tonne/cm2) and for
concrete is typically around (5,000 ksi 35,000 MPa,350 tonne/cm2), although the exact value
can be different for different concrete strengths. WS, SP and EM are always related as:

EM = (SP/g) (WS)2

where SP is (0.150 kips/ft3, 24 kN/m3, 2.45 tonnes/m3) and g is (32.2 ft/s2, 9.81 m/sec2) for
concrete. The density of grout for augercast or CFA piles is typically 10% lower than the
density for concrete. Timber density should be measured from a small sample.

Example: concrete pile WS = 13,260 ft/s = 4041 m/sec).

EM = [0.150 kips/ft3(1326O ft/sec2]/[(12 in/ft)2(32.2 ft/s2] = 5700 KSI English

EM = [24 kN/m3(4041 m/sec)2]/[9.81 m/sec2] = 39,950 MPa SI
EM = [2.45 tonnes/m3(4041 m/sec)2]/[(100 cm/m)2(9.81 m/sec2) = 407.8 tonnes/cm2 Metric

1.5.12 Pile Impedance - "EA/C"

is in cm2 or in2. The PDA then calculates the pile impedance from:

Z = EA/c (kN-sec/m, tonne-sec/m or kip-sec/ft) = EM * AR / WS

Note: The "tonne" force unit (Metric; do not confuse with English "ton") represents 1000 kg.

1.5.13 Composite Piles

For a steel pipe pile filled with concrete, initial values can be computed from the weighted
average unit density SP and the weighted average modulus (using some initial assumed
concrete modulus).

SP = (ARs * SPs + ARc * SPc) / (ARs + ARc)

EM = (ARs * EMs + ARc * EMc) / (ARs + Arc)

where the subscripts s and c refer to steel and concrete respectively.

Entering these weighted averages will yield an initial estimated wavespeed (estimated because
the modulus was estimated). If the testing suggests a different wavespeed (as in section, enter the suggested wavespeed and the modulus will be adjusted.

Note: for composite concrete filled steel pipes, the concrete should be filled to the top of the
steel, and even slightly crowned, to assure good bond between the concrete and steel during
impact. Use a plywood pile top cushion to protect the top concrete surface during impact.

February 2009 Pile Driving Analyzer Manual, PDA-W 1-23

1.5.14 Damping Constant JC

Damping increases with finer soil grains. Generally, damping is greatest at the pile toe;
however, the soil along the shaft may control when friction is high. Based usually on the soil
at the pile toe, the following are given as general guidance.

Recommended Case damping constant "JC" values for the RMX methods are:
0.40 to 0.50 for clean sands
0.50 to 0.70 for silty sands
0.60 to 0.80 for silts
0.70 to 0.90 for silty clays
0.90 or higher for clays
The RMX method is preferred. RMX is particularly useful when moderate to high soil quakes
are expected or observed. The RX7 method is equivalent to RMX with a damping factor of 0.7.
Caution is given for low blow counts (high set per blow) to be conservative as low blow counts
are indicative of low capacity. It would be helpful to reduce the hammer energy to obtain a
higher blow count (smaller set per blow). Many also compare results with the RA2 method
(which is independent of JC).

Recommended Case damping constant "JC" values for the RSP methods are:
0.10 to 0.15 for clean sands
0.15 to 0.25 for silty sands
0.25 to 0.40 for silts
0.40 to 0.70 for silty clays
0.70 or higher for clays

Generally, the RSP methods are not often used because there are better methods available.
RSP sensitivity to JC increases for finer soils or at low blow counts. For long piles where the
velocity goes negative before 2L/c, the unloading methods (RSU) may be appropriate and
these RSP damping factors are then appropriate for the RSU methods also.

Unless grain size analysis is available, visual inspection of the soil may be misleading.
A lower prediction results by selecting a higher JC. A soil plasticity index(P.I.) above 5
may imply larger JC values.

The Case Method in the field using a damping factor JC allows a capacity estimate. In all
cases, we highly recommend CAPWAP ® “signal matching” analysis of the data as a
better way to estimate pile capacity. CAPWAP is a rigorous numerical analysis which models
the pile and soil behavior. CAPWAP also produces a simulated static load test curve. After the
CAPWAP analysis, a JC value can be chosen to estimate the CAPWAP result (or a static load
test failure load if the pile has been tested statically). It is important to realize that careful
CAPWAP analysis is standard practice for the high strain dynamic pile testing method when
assessing pile capacity.

It is important to understand that the capacity predicted by dynamic testing is the capacity
at the time of testing. Usually, pile capacity is low at the end of driving, and may
increase substantially after driving with time due to “set up”. Set up is usually larger and
also takes longer to develop for finer grained soils. In some cases (e.g. weathered shale
or dense saturated silts at the pile toe), capacity may reduce (or “relax”) after driving is
completed. Subsequent testing during a “restrike” after installation is generally preferred
to take full advantage of capacity gains, or protect against capacity losses. The wait time
required for best results in a restrike varies with the soil type and project scheduling

1-24 Pile Driving Analyzer Manual, PDA-W February 2009

At high blow counts (typically greater than about 10 blows per inch; or set per blow less
than 2.5 mm), dynamic testing may underpredict the total capacity since full soil
mobilization is not achieved (particularly at the pile toe in coarse grained soils). A larger
hammer or more energy is then suggested if higher capacity must be proven.

The screen can display resistance (RTL with JC = 0.0 and RSP with JC) vs time. The portion
of RTL similar to the velocity is probably due to damping. Changing JC until the RSP curve has
a "flat" appearance (eliminating velocity dependent forces) may help determine the static load.
Checking resistance versus time is recommended. Note "2L/C" must be correct. This
method works best for low friction piles (high percentage of end bearing).

1.6 Saving Data

As the signal of each transducer (two or four strain, and two or four acceleration) is digitized,
the data is also stored as separate signals (i.e., F1, F2, V1, V2 - note that velocities are stored
for PAK and PAL rather than accelerations; the PAX stores strain and acceleration signals
directly, and not velocity).

Saving in W01 Format - After testing is complete, or datas is reprocessed, data is permanently
saved to file by the FILE/SAVE Menu selection or by clicking the SAVE ICON. The “W01" file
created (file name will be generated from the PILE NAME with a W01 extension) can be placed
in any folder. You might want to keep all piles tested in a “jobsite folder” that you create for
each project (the PAX creates a job folder based on the PROJECT name). The W01 file from
a PAK or PAL contains 1024 data points each for force and velocity, with the pile information
(area, length, name etc...). Data from a PAX contains either 1024, 2048 or 4096 data points
each of force and acceleration. The PDA-W program will read any data created by a PAX, PAK
or PAL.

See also section 1.6.2 Reducing W01 Data Files (this causes further reduction in file size, but
at the expense of removing blows from the file). If changes are made to an existing data file
(active sensors, calibrations, LP driving logs, reduced file size by eliminating excessive blows,
CAPWAP adjustments, added comments, changed names or hammer information, etc), the
data file must be saved again to remember the changes made. Saving the reprocessed data
in a new file name will retain the original data file (although usually the reprocessed data is
saved to the original file name.

A new “folder can be created prior to the test using Windows Explorer (right click in any
area and then with NEW, select FOLDER, and then name the folder preferably with the
project name). You can create a folder even after the data is acquired by FILE/SAVE from
the Menu Bar which brings up the SAVE AS dialog box. Right click in the large “window”
showing the files, and select NEW, select FOLDER, and rename the folder.

February 2009 Pile Driving Analyzer Manual, PDA-W 1-25

If power is interrupted before the data file has been saved, a provision in the PDA-W
program provides for a “backup file”. If an unsaved backup file exists when you restart the
program, a prompt asks if you want to save the file. This is your one and only
opportunity to recover lost data.

Data can be saved in a no-loss compressed format resulting in reducing the storage
space required by a factor of 2 to 3 times compared with old X01 format files. The actual
saving depends on the shape of the data. To save in this compressed format, select
FILE/OPTIONS/”Compress W01 Files” and make sure this option is “checked”. This is
highly recommended.

Saving in X01 Format - You may convert W01 files into the old DOS format X01 files by
clicking FILES/”Convert to X01 File”. This allows backward compatibility to the PDAPC
program. However, the W01 files are superior to X01 files and there is little need to perform this
task. X01 files are also considerably larger than W01 files. More blows fit into one floppy using
the W01 format than in the X01 format (so you can save a larger sample of data). There are
other advantages to W01 file (e.g. hammer definition and storing information such as the PC
comments, drive log, et al). The PDIPLOT program (Windows program) is considerably more
powerful and easier to use than the old PDAPLOT program (DOS program) to summarize the

1.6.1 Deleting Blows

Individual blows of “bad data” may be deleted from any W01 file by displaying the blow to be
deleted and then selecting EDIT/”DELETE CURRENT RECORD” from the Menu Bar (Ctrl+Del)

1.6.2 Reducing Data Files

There is no limit to the number of blows in a W01 file. The only limit is the available hard disk
space. However, to keep a file with hundreds or thousands of blows creates very large files
which are difficult to transfer to another PC or store for archive purposes. In a large sequence
of blows, often one blow is similar to the next and changes occur rather gradually. Thus, a
“sample” of the data is often sufficient.

To produce a “sample” which automatically reduces the file size, click the “W- ICON” or
EDIT/”SQUEEZE W01 FILE” from the Menu Bar. The “W01FILE ADJUST” dialog box appears.
Of course, some data is more important than other data. For example, the first blows of a
restrike are more important than later blows, or the end of driving blows are more important
than the early blows. Using the entry boxes, you can select the starting save location (SL
value), ending location (SL value), number of blows to save consecutively at beginning and end,
and the frequency of blows in between. Increasing the “save frequency” value reduces the
total blows saved. Based on the user’s selections, the number of blows saved and approximate
file size are displayed. When the file size is sufficiently small, click the OK button to finish and
the file will be shortened by removing some blows. Prior to taking this step, it is
recommended to check the BN Filter (section 7.9) to correct any problems first, since BN
Filter only works properly when ALL blows are present.

The blow numbers can be renumbered based on the value assigned to the first blow
in the dialog box; subsequent blow numbers will be adjusted by the difference between
the first blows current and new entry blow numbers.

A. Reducing data files removes some blows. The bi-colored bar at the bottom of the dialog
box shows the effect of the reduction. White portions are retained in the reduced files
while shaded portions will be removed. Hovering over this bar with the mouse will reveal

1-26 Pile Driving Analyzer Manual, PDA-W February 2009

the Save Location sequence number (SL) and blow number for the selected portion of the

B. (Once a file is reduced, some blows will be lost permanently which is usually no problem.
Eventually you run out of room if you do not clean up your hard disk. If you want to create
multiple files from the same original data, such as a sequence of blows in the middle of
the testing, copy and re-label the file using a different name with Windows prior to
reducing the original file).

C. (It should also be obvious that eventually you should transfer data from your hard disk to
a more permanent storage medium such as floppy disk, CD, zip disk, USB memory
device, a network server etc using Explorer, etc. You must maintain sufficient free
space to be able to acquire new data.). Once saved elsewhere, original data can be
removed from the PDA (PAX, PAK or PAL) to free up space for the next data acquisition.

D. It is always recommended to archive files to another mass storage device (e.g. zip disk,
floppy, DVD or CD). The W- ICON can compress files with many blows during driving to
a more manageable size for permanent storage.

1.7 Replaying Data Files

Data files that have been saved in either W01 or X01 format can be reprocessed. Click on the
OPEN ICON and there you can select the folder and file type (W01 or X01), search for the
pile/file and then OPEN it. Alternately, access FILE on the Menu Bar. In this FILE menu, you
can select to OPEN either W01 files or X01 files or even R01 files (from PAL memory card).

A. Once opened, any blow can be accessed by the SLxx command (e.g. SL23 will go to the
23rd blow in the file; this is not necessarily BN 23). Save Location [SL] shows the number
of blows in the current file and the location in this sequence of the current displayed blow.
The SL information (on the lower status bar) is followed by the blow numbers (BN). The
Replay toolbar has special ICONS that will go directly to the first or last blows in the file.

B. The data can be replayed sequentially blow by blow with PgUp and PgDn.

C. The data can be replayed forward (increasing BN) by typing the two-letter RA command,
or clicking the RA (forward) ICON or pressing the F12 function key. The data can be
replayed in reverse (decreasing BN) by the RA (backward) ICON. The replay speed can
be adjusted from the OPTIONS/REPLAY on the Menu Bar and then selecting either FAST
or SLOW. The ESC key (or the RA stop icon) will stop this automatic replay.

There are some FILE options that can be user specified. From the FILE/OPTIONS selection
the following can be selected by “checking” the topic:

1. Allow Multiple Documents - Lets you have more than one window with each window
having data from a different data set (1.7.1 Display of Multiple Data Files

2. Autoscale Document on Load - this feature only affects reprocessing of existing data files.
It adjusts the force scale when reading a pile to the anticipated optimum scale. Probably
you may always want to leave this on if you test different pile types. If however you
are on a certain site and are always adjusting the force scale for reprocessing to a
specified value then turning this feature off may save you some time. Additional
instructions are found in Section 1.8 Customizing the Graph Display.

February 2009 Pile Driving Analyzer Manual, PDA-W 1-27

3. Compress W01 Files - This feature results in smaller file sizes for the same amount of
data. It is a no loss compression. It is logical to always use this feature.

1.7.1 Display of Multiple Data Files Simultaneously

More than one set of data can be viewed at the same time. This would allow for example the
end of drive data to be directly compared with the restrike data of the same pile, or results of
one pile compared with those of a second pile. To activate this helpful feature, from the FILE
menu select OPTIONS and check the ‘Allow Multiple Documents’ feature. Then with multiple
documents opened, select WINDOWS/’TILE HORIZONTAL’ from the menu bar. You can then
click on any of the open documents to make that the active window. You can shrink or expand
any window by a drag and drop technique on any boundary.

1.7.2 Changing Project/Pile Parameters

The original data in a file may be in error. For example the length could be wrong, or even the
pile name could be in error. You can change these global parameters (which generally apply
to all blows of the current file) with the two-letter input commands (e.g. AR256 would change
the area to 256). The new parameters will be used then for further processing as long as this
file is open. These parameters can be permanently retained only by saving the file again.

Some files may have variable pile lengths, or other changed variables. This might occur either
during data collection when a pile is spliced, or perhaps by merging files together (e.g. merging
the second section of a pile after splice to the end of the driving of the first section). In that
case, changing the LE will affect only the data matching the current LE for that blow.

The overall wavespeed can be changed for each blow independently (1.7.6 Variable WC

The “precision” of any variable (e.g. AR, LE) is selected in EDIT/”OUTPUT QUANTITIES” from
the Menu Bar. Clicking on the appropriate “Class Name” shows a list of variables affected by
that selection. The user can then change the precision (number of significant digits to the right
of the decimal point). The quantity dimension names are also changeable. The names and
precision can be RESET to the PDI default values from the menu bar “Reset” included at the
top left of this dialog box.

The following parameters set in the New Page input above can later be changed with the two-
letter input command method (entry of the two letters followed by either numerical value or an
alpha-numeric label).

Documentation related commands:

PN Pile name
PJ Project name
PD Description of pile or hammer
PC Print a comment in result file (each blow can have one separate comment)
OP Operator

Q1..9 Quantity location of result (follow with 3 letter result name)

Pile properties related commands:

AR Area (Type "AR" then "Enter" to get “Area Calculator”, a helpful feature)
LE Length below sensors to pile bottom
SP Specific weight
WS WaveSpeed (used to calculate EM; use "SW" to ESTIMATE)
WC Wavespeed Calculated (WC used only for 2L/C TIME; WC < WS)

1-28 Pile Driving Analyzer Manual, PDA-W February 2009

EM Elastic Modulus ("EM" is automatically calculated from SP and WS, or WS from
EM and SP)
PDA assumes uniform pile; use "AR, SP & EM" at sensors

CL Compression stress Limit - puts horizontal line on force graph at this stress level
LS Length to Splice (referenced to sensors) - puts vertical line on graph at this time

JC Case Damping Factor

LP Length of Penetration (M or FT)

LI Length Increment (M or FT)
(use BLOW-SWITCH or "END" to increment LP by LI)

Hammer performance related:

ER Energy Rating (used for ETR = EMX/ER)
WR Weight of Ram (used for VRI, KCP)
WH Weight of Helmet (used for FCP, KCP)
MB Max "Blows / Minute" (used to prevent triggering on "BOUNCE" blows of drop

1.7.3 Area Calculator

A special function has been provided to simplify determination of this important parameter. To
access this function during data replay, click the AR ICON, type AR (with no numeric value),
or click the Area Calculator button in the NEW setup. Select the basic pile type (Pipe, Square,
H, Octagon, Triangle, Hexagon, Monotube, Sheet). The H pile and Monotube selections will
provide a simple lookup table. The other selections will display a shape. User enters the
overall section Width and Void Diameter (for no void, this Void Diameter should be entered as
Zero), and the resulting area is displayed. Click OK to accept the result. If the Area Calculator
is interrupted before completion (by either OK or Cancel), it may automatically be minimized
to the task bar; click on Area on the task bar to restore the function. The area calculator also
calculates the effective circumference (CI) and bottom area (BA) to be used by CAPWAP (for
H piles, the circumference and bottom area assume a fully plugged section). The user can
override the area or circumference by direct entry.

The Area Calculator automatically is activated when a new data set is being created.

1.7.4 Data Adjustments for Velocity

A few parameters influence the data (only adjusts the velocity since there is no easy “logic” to
adjust the strain results). Good quality data should not require significant adjustment. If the
data is inconsistent or unreasonable, the sensors or their attachment to the piles may be faulty.

A. VT - Velocity Time - the velocity curve can be shifted relative to the force in time. Enter
the floating point value of the time shift desired (the shift can be partial time increments -
positive/right or negative/left; e.g. VT-1.2 shifts the velocity 1.2 samples negative/left).
This adjustment is commonly needed to reduce apparent phase shifts in the data
and is recommended good practice, especially when performing CAPWAP.

B. VE - The Velocity at the End of the record is adjusted to zero. The curve is "pivoted"
starting at the VA time (often point 100 or 200) of the 1024 point data array to the VE time
(data points are assigned values 0 to 1023; usually point 1023 is used for VE (or 2047 or
4095 for longer sample length PAX data). Values larger than the the maximum number
of data samples - 1023, 2047 or 4095 - are ignored). The VE time can be selected earlier
with this command. If the VE value is changed, a time marker (negative only) will show

February 2009 Pile Driving Analyzer Manual, PDA-W 1-29

the VE time where velocity is defined as zero (and then usually the rest of the record is
ignored for interpretation). Use the two letter command to change (e.g. VE900 changes
the adjustment time to data sample 900). This may be useful for lower sampling
frequencies (longer data acquisition time) if the velocity curve then tends to “drift” or curve
when it obviously should be flat, or if a second impact occurs at the normal end of the
record. This feature is also helpful if the “assembly drop” occurs at the end of the record
and thus the velocity is non zero at that time.

C. VA - This sets the beginning time for Velocity Adjustment. The velocity at the record end
may not be zero (due to imperfect signals and signal conditioning) as it really must be.
The PDA adds a triangular velocity correction starting at the VA point such that the final
velocity is zero. The default time is 200 (or 100) time samples for PAK or PAL data, or
larger values for PAX when higher sample lengths are selected (the normal point is the
time where the data is triggered). The location is shown by a negative marker on the time
scale in the GR screen (color of these lines can be changed in the graph colors option;
white line on white background will hide the line).

D. Precompression adjustment for diesel hammers - If hammer type is “diesel” then

adjustment of the pretrigger data (during the precompression of the gasses) is availablefor
the velocity collected by PAK or PAL. The actual adjustment is made from the
EDIT/”HAMMER PROPERTIES” on the Menu Bar. The value can be entered as from 0
to 0.5 g’s. “Reset PreComp” reverts to the default value of 0.3 g’s. No adjustment is
needed for PAX data, and no adjustment is allowed.

E. FF - applies a smoothing to the curves (boxcar filter with an input number of samples -
e.g. FF3 takes the average of three running samples). The maximum FF value is
frequency (FR) dependent to prevent gross distortion of the data. FF filters both force and
velocity data. Note that for the PAK and PAL, the data is automatically smoothed at
frequencies less than FR 20,000 by the ratio of frequencies (e.g. at FR 5,000, the data is
smoothed by a factor of 4 { = 20,000 / 5,000 } ). For the PAX, the true sampling rate is
40,000 Hz, so lesser selected sampling rates are similarly filtered (e.g. 10,000 already has
an average of 4 samples). The value for FF is normally one (no extra filtering) which
is the PDI recommendation. New data will always default to 1. If FF is changed, it will
be stored with the file and the replay will use the selected stored value. The Velocity can
be filtered at a different rate using VF, although this is not recommended.

F. DL - The Delay Time shifts both time indicators (T1 and T2) by DL time samples. Inputs
ranging from -200 to +700 are allowed. Force, velocity and displacement (FT1, VT1, DT1,
FT2, VT2) or waves (WD1, WU2) at specific times (T1 and T2) can be obtained using DL.
A delay which increases the RSP capacity probably indicates a large soil quake or sharp
impact (TMX value gives the DL required for RSP to give the maximum capacity RMX).

Note: INS or DEL Inserts or Deletes extra (integer) time increments onto the DL
value, effectively shifting the T1 time marker placement. The
above parameters are global meaning they will apply to the entire
data file. The T1 time affects the plots of RT and RS versus time
starting point. Using DL0 will then reset the time markers to the

G. PAX wireless data from two independent channels must be “aligned” in time. Extra data
is captured by the PAX wireless so that one set of signals can be shifted in time relative
to the other if necessary. The PAX does this during data acquisition. This alignment can
be reviewed or adjusted using EDIT then RADIO_DATA_ALIGNMENT. In this window,
choose the MODE (auto or select - select gives more control of which sensor combination
is to be preferred; this is helpful in case of one bad accelerometer so that you can try to

1-30 Pile Driving Analyzer Manual, PDA-W February 2009

align on some sensor combination that excludes that bad accelerometer). NORMALIZE
and DIFFERENTIATE are highly recommended to be checked, and the METHOD (1D-A)
is strongly suggested. Pressing ALIGN (CURRENT blow, or ALL) with analyze the current
blow or all blows (each blow is independently evaluated). The user can use the slide bars
under the METHOD selection to individually adjust alignment (although this is almost
never needed with the 1D-A method). The alignment is shown in the lower left data
display. The SHIFT needed to align data is shown in the right table display, along with the
TRIGGER channels for the wireless sensors and the alignment method. The SHIFT is
usually very small if accelerometers are used to align, or if the hammer is not a diesel.
The SHIFT value may be significant if strains are selected for either trigger channel. User
can scroll through the data using the up or down arrow, or PgUp or PgDn keys.

The following adjustment may be applied to each blow individually.

H. CW adjustments - The velocity of any individual blow may be changed to add acceleration
over selected portions of the velocity curve resulting in a match to a user input SET (SET
is the final net penetration per blow which should match the visual observation of net set
per blow; input SET will change the “blow count” or input of “blow count” will update the
SET). The SET value entered will be displayed on the displacement versus time graph
(e.g. DPFW) so the user can assess the impact of any adjustment. The goal is for the
user to make the displacement graph as a function of time to be as reasonable and
realistic as practical. Normally this should result in a flat tail at the end of the curve that
matches the SET per blow.

The “Defaults” will insert preset times (T1, T2....) and preset acceleration adjustments
(A12, A34 and AC). Acceleration adjustment A12 is added between times T1 and T2.
Acceleration adjustment A34 is added between times T3 and T4. Acceleration
adjustment AC is added from time TC to the end of the record. Changing A12 or A34
will automatically adjust AC so that the final displacement (DFN) matches the user
input SET. All adjustments can be deleted by clicking the “Remove” button.

The CAPWAP manual may give further information on this process.

1.7.5 Changing Calibrations (or Replay Factors) in Existing Files

If an error was made in entry of calibration (e.g. the wrong sensor was entered), it can be
corrected by clicking the AF ICON, or click on VIEW then SENSORS. After clicking SELECT
of the subject sensor, you can change the sensor ID or the calibration (select from the list; click
on ID to alphabetize the list. You can also create NEW sensors for the list or EDIT existing

Alternately the data can be input directly into the AF dialog box. Then you must choose which
blows the change is to apply (all, current to end, first to current, or current only) and then click
the APPLY button (you will be asked to confirm this major change).To save changes for future
reprocessing, you must save the file again.

The Replay Factors can be similarly changed. Replay Factors adjust the magnitude of the
sensor output. In general, there should be no reason to do this for good data and proper
entry of parameters. Thus, a Replay Factor of 1.00 is standard and is desirable. Minor
adjustments up to 2% or maybe 3% are sometimes used (e.g. factors 0.98 to 1.02), and
are permitted since that is the basic sensor calibration accuracy. Changing to any other
larger factor may not be justified. (One exception is when an accelerometer is not
axially aligned with the pile and its signal is then reduced by the cosine of the angle.
The signal can be increased to restore the correct magnitude by entry of the inverse

February 2009 Pile Driving Analyzer Manual, PDA-W 1-31

of the cosine into the replay factor.) Replay factors are shown on the plots from
PDA-W in brackets after the calibration constants for each sensor.

1.7.6 Variable Wavespeed WC Values

Review of section 1.5.10 on determination of wavespeed is highly recommended for those

testing concrete or timber piles.

There may be occasions where the overall wavespeed for the entire pile length is not consistent
with the modulus and density at the transducer location and the user has been therefore given
the flexibility to input separate values.

Although direct input is not allowed, the Wave speed Calculated (WC) (from the input
length LE and the T1 and T2 time indicator marks) is easily adjusted by the left and right
arrow cursor keys. The WC value should usually be equal to or less than WS. For
example, the overall wavespeed of concrete piles may be slowed due to minor cracking;
WC significantly faster than elastic solution wavespeed WS should NEVER be used
for uniform driven piles. WC may differ slightly from WS due to the discrete sampling
frequency. WC is used only for the 2L/c computation and does NOT affect the relationship
between WS, SP and EM.

WC might be faster than WS for multi-section spliced concrete piles where the top section
has lower strength and lower wavespeed than a previously driven segment, or where the
sensors are attached to the concrete in a composite pile with a concrete top section with
a protruding long steel H pile at the bottom.

In the case of concrete piles, the overall wavespeed may vary progressively (gradually get
lower) during the driving of one pile due to minor tension cracking or joint related phenomena.
In this case the user should use the rise-to-rise method to determine the overall wavespeed
(WC) used in the Case Method capacity computations. In practice, WC wavespeed is almost
always highest at the beginning of each data set, therefore determine the highest WC and make
sure it is entered for the first blow (and WS set to this value). (WC is also usually either equal
to or lower than WS for single section uniform concrete piles.) To adjust subsequent blows,
select OPTIONS/”Calculated Wavespeed” from the Menu Bar. There are three choices:

A. Constant for Pile - means any change in wavespeed caused by left or right arrow keys will
change effective wavespeed WC of every blow.

B. Blow by Blow Edit - each blow can be independently adjusted by the left and right arrow
keys. This option will be rarely needed for data entry except for perhaps files with only
a very limited number of blows. This method is used after the “Blow by Blow Auto Edit”
to keep the variable WC without further changes.

C. Blow by Blow Auto Edit - in this mode, the user should start at the first blow of the data set
with the assumed correct WC at the beginning of the data set (and correct WS). Each
time PgUp is pressed, the program goes to the next blow and assigns that blow the same
WC as the previous blow. The user continues with PgUp until an adjustment is needed
in WC; the adjustment is made on any blow with the left and right arrow keys prior to
proceeding to the next blow with PgUp. The user continues through the entire data set
with PgUp making adjustment when necessary as the WC gradually slows (or in rare
cases increases). Note that PgDn does not change the WC value. When the WC has
been properly adjusted for every blow, the file should be saved to retain these
values for future use. For files with variable wavespeed, after the WC has been
adjusted for all blows in this “auto” more, the wavespeed calculation method

1-32 Pile Driving Analyzer Manual, PDA-W February 2009

should be changed to the “blow by blow edit” mode to prevent further accidental

While not exactly a WC function, the LS function can be used to help locate splices in a jointed
concrete pile (if a tension reflection from the splice can be observed). The LS and WC
functions can then be used to perhaps help determine WC for specific pile sections. LS - The
Length to Splice can be entered and a vertical line will appear at this depth (below sensors)
on the graph (at time 2*LS/WC after the initial rise marker). This line can be deleted by entering
LS with no value.

1.7.7 Entering a Drive Log to Document the Penetration Depth LP

For cases where the length of penetration was not entered during pile installation (1.5.9
Recording the Penetration Depth LP), this information can be conveniently added later. This
is accomplished with the EDIT/”Drive Log” option on the Menu Bar which brings up a dialog box.

A. The first step is to enter the “Final Depth LP” and “Final LI” value in the lower left. If the
LI value is consistent throughout the driving no further entry is required and pressing the
“Generate List” button completes this process.
If the LI value changes at one point (e.g. from initial 1 foot to final 1 inch), then the
initial LI (“Transition LI”) and the “Transition Depth” where the LI changes from the
initial value to the final value also need to be entered. Note that if there is no value,
the value should be entered as zero (0) and not left blank.

You can insert lines, delete or edit any depth values. It is suggested that if you do to
click the “Show Hidden” box to view all information. Inserted lines are helpful when
there is a temporary stop and an uneven depth increment.

B. Step A should then generate a full list of automatically decreasing “Depths” in the left most
column in the spreadsheet of the dialog box.

C. Now enter the “Blow Count” from the driving log in decreasing penetration order (blow
count for last penetration depth entered first). Continue until the list is completely entered.

Note that the cell information for “Blow Number”, “Blows/”depth unit” (e.g. bl/ft), and
“Set/Blow” are then automatically calculated and entered purely from the blow count
and depth entries.

If an error was made in data entry then the blow count record will be shifted up or down
from the depth column. To correct this if (when) it occurs, place the highlight on the
depth where the error occurs and click. Then click on the “Insert Blow Count” or
“Remove Blow Count” buttons to shift the data entries at and below that location up or
down relative to the depth column. If inserting, then enter the new value in the blank
cell generated.

D. The ground reference elevation can be entered and the pile bottom elevation (EL) is then
calculated. If the pile is not driven vertically, also enter the horizontal and vertical values
for the inclination (e.g. 1 horizontal for 6 vertical), and the angle of installation from
vertical will be computed and the bottom elevations adjusted appropriately.

E. When completely finished and satisfied, click “Apply” and then “OK” to accept your entries.
The LP values and “Set” for each blow number will then be adjusted to the corresponding
values as per the table created above.

February 2009 Pile Driving Analyzer Manual, PDA-W 1-33

F. These entries can be retained permanently with your data if you then save the file again.
Requesting the drive log again will bring the previous entries into view for editing.

If you press “Cancel” (or ESC), nothing will be retained from your effort.

1.7.8 Incrementing the BN or LP

Clicking EDIT / INCREMENT (BN/LP) will request a dialog box allowing you to edit the blow
numbers (BN) or penetration depth (LP) by some user entered value for all blows or some
subset of blows. If for, example, LP was incremented in the field data acquisition but missed
for one increment (so that increment has about twice as many blows as it should have), then
go to the blow where the increment change was probably missed and request the Increment
LP function, enter the value of one length increment, and answer to apply only to blows from
current to end of file. If the reference was changed, you can adjust all LP values by answering
apply to entire file.

1.7.9 BN Filter

Clicking EDIT /BN FILTER will give a dialog box allowing you to edit systematic blow rate and
thus blow counting errors. A graph shows the blows per minute (BPM) versus BN for all blows
(first blow in file at bottom; placing the cursor on the data point will reveal the BN and the BPM
for that blow). If the BPM is too high due to a bounce blow, entry of a reasonable MAXIMUM
BPM less than the value in error will cause that blow to be eliminated from the data set (blow
will be eliminated and the BN values resequenced). If a blow was missed during data collection
such that the BPM for that one blow is about half the BPM for surrounding blows (or 1/3, or 1/4),
enter reasonable values for both allowable MAXIMUM and MINIMUM BPM (you can click on
the graph itself or enter in the dialog value entry boxes) and if double or triple or quadruple the
low BPM result then places the thus computed BPM within these limits, then the BPM is
corrected and the BN adjusted to indicate a blow(s) was missed (skip one, or more, BN’s and
add one or more to all subsequent BN’s in the file). When the limits are considered correct,
press APPLY to activate the corrections. This function may only work correctly on the
original data where each and every blow was saved (e.g. the BN are sequential with no
skips). Thus, this function should be run BEFORE deleting any blows either individually or
using the EDIT/SQUEEZE function.

In case the MB (max blow rate) is set too low (e.g. MB set to 60 for a hammer running at 90
BPM), only every other blow will be acquired and will be labeled sequentially with an incorrect
BPM (e.g. 45 instead of 90). For this example, to correct the BPM and BN values, entering a
Maximum BPM of about 100, and a Minimum BPM of about 80, should correct the data to
restore the right BPM and BN. However, as soon as it is noticed, the MB value should be set
to a speed slightly higher than the anticipated maximum blow rate.

1.7.10 Merging Files

If a file is open, another file can be appended to it using the EDIT / MERGE W01 FILE function.
This is useful if a pile is spliced, or if a restrike is desired to be appended to the driving blows.
The length can be different for each file (as it would be for spliced piles), but the area, density,
modulus and wavespeed should be identical (or you should not merge the files). You MUST
use the same sensor type for each test to be able to Merge files. After merging, the user
may store the merged files in the standard save methods, and it is best to rename the new file
so the original data is maintained. This function requires careful thought and correct application
by the user.

1-34 Pile Driving Analyzer Manual, PDA-W February 2009

1.7.11 Changing Hammer Properties

If the hammer was improperly entered during data acquisition (or not entered and now desired
to be entered), use EDIT / HAMMER PROPERTIES to access a dialog box. Clicking LIST will
bring up the entire database of hammers to allow selection.

1.7.12 Function Keys

The function keys (F1 through F12) perform various fast requests to aid data processing or
even data collection (these function keys help speed operation and the user benefits by
remembering their existence). The following guide summarizes these helpful keys:

F1 Help
F2 Accept/Standby toggle
F3 no action
F4 no action
F5 Display DPS or DPFV
F6 Display DPFW or DPFR
F7 Display DPF or DPW
F8 TS (time scale toggle)
F9 no action
F10 no action
F11 no action
F12 RA (replay automatic)

1.8 Customizing the Graph Display

The window sizes for the graphs and the (left side) text window can be adjusted by the drag
and drop method.

• The data type on the graph display on the right side can be changed by:

right clicking on this graph area, or selecting VIEW/”GRAPH DISPLAY” from the Menu
Bar. Several selections become apparent and can be selected by clicking on the
appropriate choice, or

entering the display code by direct keyboard entry (e.g. typing DPFV will display the
signals from the Individual Sensors - all F’s and all V’s), or

quickly changed with the F5 (DPF or DPW), F6 (DPFW and DPFR), or F7 (DPS and
DPFV) function keys (as outlined on the PAK keyboard).

DPF Standard force and velocity [F7]

DPW Wave down and wave up [F7]
DPR Resistance with and without damping correction
DPE Energy and displacement (also DPE2 if performing SPT testing)
DPFW Force, velocity, and wave up, displacement [F6]
[F6; highly recommended]
DPS Average force, force1, force2 [F5]
DPV Average velocity, velocity1 and velocity2
DPFR Force, velocity and resistance RS and RT [F6]
DPD Individual velocity and displacement (V1, V2, D1, D2)
DPFV All 4 individual signals (F1, F2, V1, V2) [F5]
DPM Momentum
DAS Acceleration and Strain (DASF adds FFT)

February 2009 Pile Driving Analyzer Manual, PDA-W 1-35

8 channel operation results:
DPV Display all four velocities
DPD Display all four displacements
DPS Display all four strains (forces)

• The time scale on the graph display can be changed by pressing the F8 function key, or
entering the TS command on the keyboard (optionally followed by a value; e.g. 100 for
100% of time scale), or the clicking the TIME SCALE ICONS (T< and T>). If the scale is
not the full scale, then the starting time can be changed with the TBvalue keyboard input
(e.g. TB12 will start at 12 milliseconds) or clicking the ICONS (T- and T+).

• The graph vertical scales can be changed clicking the scale ICONS for Force, Energy,
Displacement, Acceleration or Strain as needed. The main Force scale can be reset with
the AUTOSCALE ICON (scale selected to give generally acceptable and normal force
signals). Some vertical scales (force FS, velocity VS, energy ES, and displacement DS)
can be changed by two letter commands (e.g. FS1500 will change the force scale to
1500). Using these two letter commands can get around scale changes that seeming
refuse to change using the ICON method. Changing the Force scale will normally
automatically change the Velocity scale (and vice versa) since the data will always be
displayed proportionally by the EA/c impedance factor.

If the Velocity scale is desired to be independent of the force scale, click OPTIONS
and then VELOCITY SCALE, and select INDEPENDENT. This may be desired when
using top transducers, or when the pile is highly nonuniform at the sensor location.

• The colors used can be customized by clicking the GRAPH COLORS ICON on the
custom toolbar, or selecting OPTIONS/”GRAPH COLORS” from the Menu Bar. A dialog
box then allows changing line types (e.g. solid, dash, dot), line colors, and background
color. Click APPLY to accept the new color selections and OK to exit the box. Dark
colored lines on a white background are recommended for optimum viewing.

• Cursor Keys: Allow special data or time shifting to occur.

Right or Left Controls the T2 time placement for the Case Method 2L/c time period for
determination of capacity. Changing T2 computes a new wavespeed WC
but will not affect the impedance EA/c controlled by WS. For fast T2 time
placement use [Ctrl] with Right and Left cursor keys.

INS or DEL Inserts or Deletes extra time increments onto the DL value, effectively
shifting the T1 time marker placement. T2 will always be 2L/c after T1.
This feature is sometimes helpful when the auto selection of the “rise time”
is not ideal.

• The left text area displays the Inputs (PJ, PN..., AR, LE....) and “Output Quantities” (EMX,
CSX, RMX, BTA etc.). The Output Quantities can be user selected by entering the two-
letter prefix followed by the desired quantity (e.g. Q4EMX will place the energy EMX in the
4th result quantity. A list of Quantities is shown by clicking the “Q?” icon. The Quantities
can be changed from the list/dialog box..

• By right clicking in the left text area, the user can redefine the text window to for example
a listing of the force values (or velocity, etc) in engineering units as a function of sample
number (there are 1024 samples per blow for PAK or PAL data, and up to 4096 points for
PAX data). The values or samples of the individual forces are followed by the average
force. By left clicking on the right graphics area at some desired time in the data (put the
mouse arrow on the point in the data that you are interested in and left click), the

1-36 Pile Driving Analyzer Manual, PDA-W February 2009

numerical values are then highlighted for that time in the left text window. (The lower
selections are the “raw data” in A/D values and are perhaps not very useful to you -
although the “Strain Sensors” shows the actual A/D values which for normal balancing of
PAK should begin at about values near 2048 (look also at the Quantity SO1, SO2, SO3
and SO4 fo rthis initial offset value). If the balance value is quite different, or changing
dramatically from blow to blow, then there may be some pile yielding or sensor slipping,
so this particular feature can be used as a data quality check tool).

• The font type, color and size in the left text area can be adjusted by right clicking on the
text area and selecting FONT. (‘Ariel’ is a good choice for font type selection).

1.9 Damping Parameter (JC)

The JC factor applies only to the basic Case-Goble capacity computations RMX, RSP, and
RSU. To change the damping factor type JCvalue (e.g. JC0.45 will make JC equal to 0.45).
Alternately clicking the J- or J+ ICONS will decrease or increase the current JC in 0.05
increments. This may be helpful when viewing resistance on the graph (DPFR or DPR).

• The capacity methods can be selected for certain damping factors by the quantity
selection. For example RX5 is RMX with JC of 0.5. Using these specific quantities (e.g.
RX4, RX6, Rx8...) rather than the general RMX gives perhaps a more clear indication of
the method and further allows the user to select more than one damping factor to view the
sensitivity (e.g. select both RX4 and RX7). Similarly, RP5 and RU5 are RSP and RSU
respectively with a damping factor of 0.5.

• Shaft Friction resistance (SFR) and End Bearing (EBR) reflect the damping constant JC,
while SF5 and EB5, for example, reflect the damping factor 0.5.

See also section 1.5.14 Damping Constant JC

1.10 Warnings and Limits

Immediately above the graph and below the Menu Bar are twelve data boxes that check data
quality and quantity results. If anything is found to be unusual, the box is highlighted (reverse

• The left six boxes check data quality such as signal clipping (a major disaster making the
data unusable and data should be promptly discarded as “unusable”), proportionality,
strain ratio (bending), velocity ratio, and velocity/displacement at end of record. (see
1.10.1 Data Quality).

• The right six boxes check the results for conformance to project specifications. This
includes, driving stresses, energy transfer, pile damage and bearing capacity.

• The “limits” for many of these warnings can be user adjusted by clicking the “φ ICON” or
selecting EDIT/”DATA QUALITY PARAMETERS” from the Menu Bar. These limits on
many computed quantity results (stress and energy) can be set to user specified values
which relate to the site limitations. For example, for a steel pile, the compression stress
limit (CSX) is often set to 90% of the steel yield strength (the compression limit for bottom
stress CSB might be lower than the top compression stress limit due to nonuniform
contact with sloping bedrock), the tension limit (TSX) is set to the tension strength of the
pile, and the energy limits are set to values which give the usual range of energy typically
expected for the hammer type being used.

February 2009 Pile Driving Analyzer Manual, PDA-W 1-37

Recommended Driving Stress Limits:
Stress Type SI Unit Limit English Unit
Steel Compression 0.90 Fy 0.90 Fy
Steel Tension (may be limited by weld) 0.90 Fy 0.90 Fy
Prestressed Concrete Compression (top) 0.85 f’c - fpe 0.85 f’c - fpe
Prestressed Concrete Tension fpe + 0.25 (f’c)0.5 fpe + 3.0 (f’c)0.5
Reg. Reinforced Concrete Compression 0.85 f’c 0.85 f’c
Reg. Reinforced Concrete Tension 0.70 Fy (As/Ac) 0.70 Fy (As/Ac)
Timber 3σa 3σa

Fy Steel yield strength (in MPa or psi)
f’c Concrete 28 day strength (in MPa or psi)
fpe Effective prestress (in MPa or psi)
As Steel reinforcement area
Ac Concrete area
σa Allowable static timber stress (in MPa or psi)

Note that in English units, the result limit should be converted to ksi for entry into the limit field.

CL - The Compression Limit stress (in stress units) can be displayed as a horizontal line on
the force display using the CL command (e.g. in English units, CL32 will place a dotted
horizontal line on the force graph at the value of 32 ksi; in SI units CL32 places the
horizontal line at 32 MPa). This can be used to visually relate the forces to the pile
material strength limits.

Typical Energy Limits:

Hammer Type Steel Piles Concrete Piles
Diesel 25 to 50% 17 to 40%
Single acting Air 40 to 65% 30 to 55%
Double acting Air 22 to 45% 20 to 45%
Hydraulic, or >75% of >50% of
Drop (free) hammer hammer monitor
Drop (cable) 35 to 60% 25 to 50%

In the above table, the energy percentages are all compared with the hammer rating. For
hydraulic hammers, results should be compared with the reading of the hammer output
(caution: in the USA many hydraulic hammers primarily from Europe are reading energy in kJ
units rather than kip-ft so the apparent result looks too large and may exceed the hammer rating
in English units). The ranges for drop hammers are guesses at this time since a large database
does not exist.

1-38 Pile Driving Analyzer Manual, PDA-W February 2009

• The limits on “End Velocity” should be tolerant of some error and values of +1.0 m/s and
-1.0 m/s are recommended as reasonable. The limits on “End Displacement” should be
set to about +300 mm and -10 mm to assure the displacement is within a reasonable
range. These ranges are to have the PDA help assess the quality of the sensors.
Increasing the Pre-Compression Adjustment value (found in EDIT / Hammer Properties)
may increase and improve the End Displacement (if negative) for hard driving cases.
Using slightly longer precompression time (e.g. 35 msec) for the PAX may also assist
improving the velocity for diesel hammers.

• The user can select either USER INPUT or CALCULATED for the “Target Capacity”
comparison with a user selected capacity method.

“User Input” is appropriate for restrikes where the soil strength has stabilized and
therefore long term capacity has been achieved.

“Calculated” is perhaps appropriate for end of drive testing when the soil strength is
reduced due ot the driving process; after driving, the capacity is expected to change
over time. The Required Ultimate Capacity and Estimated Shaft Resistance (perhaps
from a static analysis, or previous tests including restrike information) are entered
along with estimated shaft gain and toe loss potentials. A “Target Capacity” at end of
driving is then calculated. If the input assumptions (shaft resistance, shaft gain and
toe loss) are correct, then the Target Capacity will become the Required Ultimate
Capacity after sufficient time. Thus you may choose to drive the pile to the Target
Capacity at the end of drive (often saves cost since most piles have capacity gains
with time). You should confirm the long term capacity by restrike after an
appropriate wait period, or by static testing, if you employ this technique.

• When all values have been set, then click on APPLY.

1.10.1 Data Quality

Quality data is important, since if data is erroneous, then any analysis or result is meaningless
or wrong. Transducers must be checked and properly calibrated (ASTM D-4945 suggests
calibration of all components every two years, or if repaired). Old sensors (acceleration or
strain) should be inspected and recalibrated periodically; recent accelerometers have improved
quality, and new accelerometers are available for even steel-to-steel impacts imparted by
hammers with no cushion or SPT rigs.

Data acquired during either driving or restrike should be PROPORTIONAL, CONSISTENT, and
REASONABLE. Generally good data will have reasonable PROPORTIONALITY between the
force and velocity at the first peak (for uniform piles), and will be RELATIVELY CONSISTENT
from blow to blow (data shapes and results); both F and V data should start and end at zero.
Results (especially displacement as time function) should also be REASONABLE: the final
displacement (DFN) should match the observed set per blow (this might need some minor user
adjustment for CAPWAP blows; see 1.7.4.H and 1.11), and the energy transferred should be
within normal percentages of the hammer rating. It is highly recommended that you let the PDA
inspect the data quality; heed all warnings and strive to eliminate any source of poor quality

Proportionality can often be judged better by inspecting the Wave Up curve. For good
“proportional” data, there should be no sudden step jumps in the Wave Up curve at the
time of the first peak. “Phase shifts” between force and velocity data (which can create
local disturbances in Wave Up) can and should be minimized using the VT function.

February 2009 Pile Driving Analyzer Manual, PDA-W 1-39

Poor quality data could be due to 1) bad sensors, 2) bad attachment of sensors to the pile, 3)
excessive bending in pile due to nonuniform impact, 4) bad cables, 5) bad connection cable,
or 6) nonuniform pile or defect. Check sensor attachment, or change sensors and all cables,
or test another pile to correct the problem. Most data problems are usually due to faults in the
sensors, attachment to pile, and/or the cables. If data quality is poor, then do NOT submit

1.11 CAPWAP® Preparation

The CAPWAP adjust CW ICON prepares the data for CAPWAP analysis.

• For the new Windows CAPWAP program, the PDA-W program makes the CAPWAP data
adjustments as described in the CAPWAP manual (A12, A34, AC etc.). The data
adjustments to produce a reasonable displacement versus time function with the correct
set per blow are further described in section 1.7.4 Data Adjustments for Velocity. The
data adjustments made will be included in this file.

• The entry of LP (depth or length of penetration into the soil), Circumference and Bottom
Area (please watch dimensions!) are input parameters only needed in the CAPWAP
analysis and should be completed for reference (Circumference and Bottom Area may be
automatically generated by the area calculator function - section 1.7.3 Area Calculator).

• The Blow Count and/or Set should be entered. “Set” is the observed net final permanent
penetration per blow, and can be compared with DFN.

• Clicking DEFAULTS will set the times (T1, T2,... TC, etc.) and adjust the acceleration
adjustments (A12, A34, AC) so that the final set DFN is equal to the Set per blow. The
Set can be adjusted or any of the acceleration (except AC) and time parameters can be
adjusted and the program will automatically adjust AC to achieve the entered Set. If the
displacement curve still needs adjustment (after DEFAULT input), the general
recommendation is to first change A34 value.

• All velocity adjustments can be removed by clicking the REMOVE button.

1.12 Plotting

Two options exist. Click either the PL ICON or the HP ICON. The screen will change and
display (a portion of) the print format output. Clicking the printer icon or selecting either PRINT
or PRINT PREVIEW from the FILE on the Menu Bar will allow displaying a preview of the entire
print page (for the monochrome PAK display the PL result may be reverse color from the actual
printout), and subsequently printing of the current blow. The data plotted will be similar to the
data displayed in the GR screen. You can change the data while in either the PL or the HP
screen by the same commands as in the GR screen (amplitude scale, time scale, the blow
displayed with PgUp or PgDn, or other input command). Return to the Graph screen by clicking
the GR ICON.

In the HP mode, right clicking on the screen will bring up a dialog box allowing additional
plotting selections. Several plots can be made with [ ] braces showing a single axis group
based on combinations of the following:

1-40 Pile Driving Analyzer Manual, PDA-W February 2009

[F,V] Force, Velocity
[WD,WU] Wave Down, Wave Up
[RS, RT] Resistance Static (includes JC), Resistance Total (JC=0)
[E,D] Energy, Displacement
[F1, F2] Force 1, Force 2 (i.e., individual Force curves)
[V1, V2] Velocity 1, Velocity 2 (i.e., individual Velocity curves)

1.12.1 Copy to Clipboard

Data can be exported to spreadsheets using the EDIT/”Copy to Clipboard” option on the menu
bar. The data itself (rather than graphics image under normal PrtSc copy) is then available on
the clipboard. It can then be pasted into any spreadsheet for further computation and
specialized plots.

The COPY ICON (or PrtSc key) copies an image of the screen to the clipboard.

1.13 BETA (Integrity / Damage Evaluation)

Normally, the wave up is a monotonically increasing function during the first 2L/c after impact
due to the shaft resistance which causes upward traveling compression waves. Damage along
the shaft of a pile returns an upward traveling tension wave. This tension (negative) from
damage is superimposed on the compression (positive) due to soil resistance reflections
causing a local decrease in the wave up function. The PDA inspects the wave up curve for
local minimums and when present, signals that the pile is damaged. The warning is in the
“warnings and limits” area just under the Menu Bar and above the graphical data in the GR
screen. It also appears as a vertical line on the GR screen which shows the Beta factor (BTA)
and damage location (LTD). (The vertical line type and color can be user defined in the Graph
Colors option.) The length to damage (LTD) evaluation can be enhanced using the LS function
(for example, if a pile is spliced at 35 below sensors, then enter the length to the splice with
LS35; the LS and LTD lines can then be compared visually to see if the damage is near the
splice, as is often the case). LS - The Length to Splice can be entered and a vertical line will
appear at this depth (below sensors) on the graph (at time 2*LS/WC after the initial rise
marker). This line can be deleted by entering LS with no value.

Nominally the BTA factor represents the percentage of pile cross section compared with the full
cross section. However, short local defects may be under estimated by the method. Soil
resistances above the damage complicate the issue although a crude attempt is made to
compensate for resistance. A subjective rating was developed to estimate the extent of
damage based on the BTA value. However, piles with damage should always be taken
seriously and investigated. It may be possible that “apparent damage” is caused by bending
stresses, or poor quality data (e.g. noise on the signals). A false indication can also be
caused by a “phase shift” between force and velocity. To investigate and correct for
phase shifts you might try using the VT function to eliminate the shift (e.g. VT0.7 to shift
velocity to the right or VT -0.7 to shift to the left). Real damage should cause consistent
readings from blow to blow.

Defects near the bottom of the pile may be caused by the wrong wavespeed or the wrong pile
length, so correct entry of these values is important. For concrete piles, inspect the earliest
easy driving blows to determine the real WS. Using higher sampling frequencies (particularly
in the PAX with larger record sizes) may give earlier warning for toe damage to steel piles.
Compare the later blows with earlier blows and look for sequentially earlier return of the tension
wave to detect damage.

Large shaft friction on long piles causes the velocity to become negative prior to 2L/c; if this
condition exists, the BTA computation may incorrectly indicate damage. Look for sharp

February 2009 Pile Driving Analyzer Manual, PDA-W 1-41

decrease in wave up to confirm damage. Gradual decreases in these early unloading cases
may only result form the early unloading and not necessarily damage.

If damage is detected, the engineer should always review the result to determine if the reading
is true (or potentially false). This review is best done by inspecting the wave up curve in the
first 2L/c for sharp local decreases (smooth gradual changes may be due to soil resistance,
early unloading, or sweeping piles or simply data quality; sweeping piles may be acceptable or
deficient depending on the degree of sweep) .

BETA Value (BTA) Description

100% Uniform Pile
80 to 99% Slight Damage
60 to 80% Damage
< 60% Pile Broken
(generally pile is then rejected)

It should be noted that in the above table, categories are suggested. However the difference
between a pile with BTA of 81 and another pile with BTA of 79 is only minor. The rating scale
is really a continuous function with no definite boundaries.

Pile with larger damages (BTA values certainly less than 80) should be assessed for their
suitability. Defects far down the shaft may in some cases be not a problem if the pile is a
friction pile and has sufficient resistance above the damage. “End bearing” at the damage is
generally unreliable since the top and bottom may be poorly aligned, or the reinforcing steel
may deteriorate with time. Capacity estimates for damaged piles should be avoided.
Broken piles should be assigned zero capacity and should be replaced.

Further inspection of Beta can be made by clicking the Beta ICON (β) on the custom toolbar.
The limits can be adjusted (by minor amounts) and the time periods shifted. The DEFAULT
setting can be restored by the clicking Defaults. The graphic display can plot several user
selected parameters. Note that there is a second integrity evaluation method called BTB (also
available as a quantity output). BTA computation looks purely at the magnitudes of the wave
up, while BTB also considers the width of the “defect”. Generally the two results are somewhat

A graphical representation of the size and location can be displayed (and positioned on the
screen by drag and drop) by clicking the “View Beta Damage” ICON (pile shape).

Piles with severe damage may be unreliable for long term conditions so no capacity estimate
should be given.

1.14 General Data Interpretation

Result interpretation should be by a professional engineer who knows wave theory and pile
driving. The PDA investigates driving stresses, pile integrity (damage), hammer performance,
and bearing capacity. For best results in capacity, test the pile during restrike with a 7 day wait
to allow for strength changes with time. The Geotechnical engineer should review capacity
results for uplift, settlement, and downdrag.

STRESSES: PDA finds average max Compression Stress (CSX) at sensors (using assumed
modulus); max Compression stress for any strain sensor is CSI. A visual "Compression Limit"

1-42 Pile Driving Analyzer Manual, PDA-W February 2009

is input with "CL". Calculated Compression stress estimate at the toe (for uniform piles) is CSB;
local non-uniform contact stress on rock is NOT considered (e.g. pile toe may fail structurally
due to local contact stress well below the pile material yield strength). Tension stress (TSN for
upward in 1st 2L/c; TSX (for tension down also) below sensor (uniform piles only) can damage
concrete piles (keep below prestress if possible).

DAMAGE: PDA inspects WAVE UP for damage (should INCREASE MONOTONICALLY during
first 2L/c). Damage is usually a SHARP REDUCTION in WAVE UP. For nonuniform pile, PDA
may find "damage". ALWAYS visually check data (between rise time markers). Bad bending,
wrong length or wavespeed, splices, large phase shifts (can correct this with VT function) or
"noise" (try FF) may also indicate damage. Generally: BTA>90 is OK; BTA<60 is REJECT. For
other BTA, use judgment. Do not give capacity of damaged piles!

HAMMER PERFORMANCE: compare computed EMX and rated hammer energy. Enter
hammer energy rating ER for ETR (energy transfer ratio) EMX/ER. Can calculate stroke STK
of OPEN END DIESELS ONLY and max hammer cushion force FCP (requires WH) and
stiffness KCP (requires both WR and WH) for non-diesel hammers on steel piles only.

CAPACITY: This is often the most difficult result to interpret. PDA estimates the capacity AT
THE TIME OF TESTING only; often it is a reduced strength during driving. We suggest testing
during RESTRIKE for best long term capacity and comparison with static tests. Ideally the
restrike wait period is similar to wait for static test; often 7 to 14 days is suggested. SET PER
BLOW should be minimum 3 mm (less than 100 bl/ft) to assure full capacity mobilization.
CAPWAP analysis is always recommended in all cases of capacity determination. PDA
assumes pile is uniform; if nonuniform, you MUST use CAPWAP. Geotechnical engineer
should review results, and recommendations considering uplift, settlement, negative friction.

1.14.1 Resistance Distribution

A crude graphical approximation to the location of resistance along the shaft, and breakdown
between shaft and toe, is given with the VIEW / RU DISTRIBUTION function. The damping
factor can be changed. The scale of the graph is dynamic and normalized to the largest unit
resistance (so results may initially appear large even for small shaft resistance for the initial
blows of driving, but when shaft resistance increases at the end of driving, and you return to the
initial blows, then the distribution for the initial blows will look small). The crude division
between total shaft friction and end bearing can be made from the Quantities SFR and EBR (or
SF5 and EB5 for JC of 0.5 etc).

1.14.2 Capacity Monitor

A table for the “measured Hiley formula” (also known as “Energy Approach”) showing capacity
as a function of blow count (English units) or as a function of set per blow (Metric or SI units)
is accessed by the menu VIEW / CAPACITY MONITOR. This formula has been researched by
Paikowsky and recommendations are to use it only with end of drive data. Further the factor
“k” must be set to values less than 1.0 (values between 0.75 and 0.5 are common; logic would
suggest lower values for cohesive soils and/or for lower blow counts). In use, first find the blow
count or set per blow for the current blow, and the value of corresponding Q is the desired
result; all other values of Q are meaningless. This method is provided for convenience only;
PDI does not endorse the use of this method.

1.14.3 Tension Envelope

A visual depiction of the tension as a function of pile length can be viewed by clicking on EDIT
/ Tension Envelope. The scale is relative and automatically selected to the maximum result
for each data set. The maximum tension (TSX) and the tension (TLS) at the LS location are

February 2009 Pile Driving Analyzer Manual, PDA-W 1-43

displayed. TLS at location LS is useful in assessing the tension at a specific splice location.
The TLS parameter can also be selected as an Output Quantity.

1.14.4 Output Quantities

• The user can select 9 results from a long list of available computations to be displayed as
9 quantities Q1 through Q9. A list of all possible quantities is available by selecting
VIEW/”QUANTITY LIST” from the Menu Bar, or clicking on the “Q?” ICON. This list
contains a brief description of the computation result.

• The “precision” of any variable or quantity can be modified by selecting EDIT/”OUTPUT

QUANTITIES” from the Menu Bar. Clicking on the appropriate “Class Name” shows a list
of variables affected by that selection. The user can then change the precision (number
of significant digits to the right of the decimal point). The title names are also changeable.
The names and precision can be RESET to the PDI default values from the menu bar
“Reset” included inside this dialog box.

1.14.5 Saving Quantity (SQ) Results for PDAPLOT or EXCEL or Printing

Printing or plotting results with either PDIPLOT, PDAPLOT or EXCEL is a highly

recommended practice for transmission of PDA results to the client. When a W01 data file
exists, results can be prepared for report presentation of results in summary format (tabular
and/or plotted).

The preferred method is to use the PDIPLOT program. The PDIPLOT program reads the
W01 data files and automatically assembles data for presentation using the 9 quantities last
selected by PDA-W (last close of the PDA-W program). The PDA-W program should be closed
prior to starting PDIPLOT for most efficient reading of data for large files. PDIPLOT is a
Windows program; the user is encouraged to read the PDIPLOT manual (press F1 function key
to bring the help manual up in your browser).

Data can also be saved by selecting FILE/”SAVE SQ FILE (DOS)” or FILE/”SAVE SQ FILE
(WIN)”. The WIN SQ data file is in “engineering units” and can be directly read by spreadsheets
for other uses. The DOS SQ file has factors applied and is only useful with the DOS program
PDAPLOT (older version of PDIPLOT). PDAPLOT users are encouraged to update to the new
better PDIPLOT program which is also easier to use.

For compatibility with the DOS program PDAPLOT, select FILE/”SAVE SQ FILE (DOS)”:
The quantity results and comments are included with each blow, and headings are given
as required for the PDAPLOT program. Results are saved in an ASCII text file (*.q01...).
Results have decimal point shifting which will be handled by the PDAPLOT program.

For compatibility with Windows application EXCEL select FILE/”SAVE SQ FILE (WIN)”:
The date/time, blow number, penetration (and elevation if entered in the Drive log),
quantity results, data quality indication, and comments are included with each blow, and
headings are given to allow further processing in spreadsheet. Information is made to an
ASCII file (*.txt) which is Tab delimited. Results are included with real values at the same
precision as the PDA program had specified. The file name and folder location can be
changed prior to clicking SAVE.

Results obtained by PDA-W can be printed by saving the results in an SQ Windows

file. The file can then be copied to some memory medium and given to clients
requesting a copy of field PDA output (previous practice in some parts of the world
required a printed output). Since field printout is considered obsolete practice, giving

1-44 Pile Driving Analyzer Manual, PDA-W February 2009

this file to the client allows the client access to all filed data and he can then print the
data himself, or import it into a spreadsheet for his verification purposes.

Note: the saving of DOS SQ results is obsoleted by the new PDIPLOT program for
Windows (which replaces the PDAPLOT program for DOS). The PDIPLOT program reads
the W01 files and stores the results in a table that can be both printed by range or by
average, or plotted versus blow number or penetration. The nine results (Q1 through Q9)
last used by the PDA-W program will be the values inserted into this table.

1.14.6 Short Recommended List of Useful Result Quantities


Shaft Resistance: SFR - function of JC, or SF4, SF5, etc.
End Bearing: EBR - function of JC, or EB4, EB5, etc.

1.14.7 More Complete List of Results

Enter selection with Q# COMMAND (i.e. "Q3CSX" sets CSX for quantity 3)

TLS * TENSION STRESS at distance LS below sensors
CSX * Max average axial COMPRESSION STRESS at gage (FMX/AREA)

Hammer performance:
EMX * ENERGY TRANSFERRED to pile-(most important measure)
EF2 - ENERGY from F2 (SPT - ASTM D4633; RAT - Length Ratio)
E2F - ENERGY from F2 (From first 2L/c computation only)
E2E - ENERGY from FV (From first 2L/c computation only)
VRI - Ram Impact Velocity (Steel piles only with ASHD Hammers; WR,WH)
FCP - FORCE in Hammer Cushion
KCP - Hammer Cushion stiffness

February 2009 Pile Driving Analyzer Manual, PDA-W 1-45

Damage / Integrity: (user must always manually inspect "WU" to confirm PDA result)
LTD - LENGTH TO DAMAGE (below sensors) (also BT2 & LT2)

CTN - Max TENSION FORCE at or below sensors (1ST 2L/C only)
FT1 - FORCE at marker TIME ONE (FT2 at TIME TWO)

VMX - Max VELOCITY at sensors
VT1 - VELOCITY at marker TIME ONE (VT2 - at TIME TWO)

DMX * Max DISPLACEMENT at sensors
DFN * DISPLACEMENT AT END of data record
DT1 - DISPLACEMENT at marker T1

WU2 - WAVE UP @T2;

1.14.8 Capacity Methods

Change (Q1..Q9) using two letter command e.g. "Q3RMX" (Q3 will be RMX)

RSP - Original "Case Method" capacity result (depends on JC).

JC guide: 0.1 - 0.3 SAND; 0.3 - 0.5 SILT; 0.5 - 1.0 CLAY
RP# - RSP with J=.# to get RSP with second J; i.e. RP4 is RSP(J=.4)

RMX - Maximum "Case Method" capacity searches RSP at different T1 times for MAX result,
(depends on JC; should probably NEVER use JC<0.4, unless with static correlation)
NOT INTENDED for piles with bottom in CLAY: JC>0.8 UNUSUAL
RMX(JC=.5) is often a good initial choice (confirm with CAPWAP)
RX# - RMX with J=.# to get RMX with second J; i.e. RX6 is RMX(J=.6)

RSU - RSP(JC) for high friction cases (early unloading with negative velocity prior to 2L/c).
RU# - RSU with J=.# to get RSU with second J; RU5 is RSU(J=.5)

RA2 - Automatic Method (independent of JC) used for piles with LOW or MODERATE
this is a good first choice if no other information is available (confirm with CAPWAP)

RAU - AUTOMATIC METHOD (independent of JC) used if shaft resistance is very low
(Basically end bearing piles with little or no shaft friction).

1-46 Pile Driving Analyzer Manual, PDA-W February 2009

SFT - SHAFT FRICTION TOTAL - no correction for damping
(SFx is SFR with JC = 0.x; e.g. SF5 uses JC = 0.5)


(EBR has a crude damping correction based on JC: Ebx is EBR with JC = 0.x)

For capacity of NON-UNIFORM PILES, use CAPWAP to determine capacity, NOT Case
- all results to be reviewed & approved by Geotechnical Engineer -
Should compare Case Method result with static load test to determine proper damping factor,
or at least Case Method result with CAPWAP to confirm which method and damping factor to

1.14.9 Capacity Evaluation Considerations

For capacity evaluation by the PDA, we ALWAYS recommend using CAPWAP ® analysis to
check the PDA result, so really common “standard practice” is to use PDA and CAPWAP
together which increases reliability compared to PDA alone. For capacity testing, there are a
few simple conditions to get help you get the best result.

1. The PDA test should be a RESTRIKE after a wait following installation. The wait period
should be longer for finer grained soils and ideally will be comparable to the wait used for
static tests. For silts and clays (and even fine sands) the wait time is often 7 days or
more. Since you can do a restrike test with the PDA at any time, if there are several tests
with different wait times then they can be often plotted as a straight line on a log-time
scale (if the increase is due to pore pressure effects) to estimate capacity at other times.
Capacity usually increases, and this is called “set-up”. Set-up can be caused by pore
pressures, but also due to pile wobble and arching, and also aging (or chemical
cementing) and these might be linear with time (up to some limiting time). Selecting an
early “high energy blow” is recommended for analysis. Relaxation (or capacity decrease)
has been observed in limited cases of end bearing in dense saturated silts or piles driven
into some weathered shales; testing after at most 7 days for relaxation should be sufficient
(and then select an early “high energy” blow for analysis).

2. The set per blow must be at least 3 mm per blow (blow count less than 100 blows per foot)
to assure that full soil resistance is mobilized (otherwise if set per blow is less or blow
count higher then the test has only mobilized part of the total soil resistance, and the PDA
result will be only a lower bound estimate as it indicates only the resistance activated).
This may often happen as the pile is driven to a low set (high blow count) and then due
to capacity increase with time (setup), the set per blow will be even smaller (blow count
higher) during restrike due to the capacity increase with time. Possible suggestions

a) This may require in some cases a bigger hammer (larger ram weight) or a higher
drop height. Usually a weight of 1 to 2 percent of the desired ultimate test load is
sufficient. It should be mentioned that using higher weights (like 5% of the ultimate
test load) allows you to satisfy the requirements for Rapid Load Testing.

b) In some cases (very low set per blow - high blow count during restrike), the end
of drive end bearing can be added to restrike shaft resistance to compensate for
perhaps not activating the full capacity of the pile and to project a higher total.

February 2009 Pile Driving Analyzer Manual, PDA-W 1-47

c) In some cases like closed end pile piles, it is possible to increase the pile
impedance by filling the steel shell with concrete before the restrike. The increased
impedance of pile then causes a higher force input which in turn can overcome higher
soil resistances.

3. For proper correlation of static and dynamic tests, the load test must be run to failure (and
the PDA test must cause good permanent pile set per blow as described in point 2 above).
PDA usually correlates best with the Davisson limit load method (often regarded as one
of the more conservative evaluation methods). If the static load test is a rapid plunging
failure, then all failure load interpretations will be similar. If there is continual increase in
load with increase in displacement, then the Chin method result, for example, would be
much higher than the Davisson limit.

If Davisson is not the method used in evaluation, then a correlation between Davisson
and the other method can be established, but the correlation must account for end
bearing differences (pile type and soil type at pile bottom) to estimate the other load
method result from either the PDA or Davisson result.

The safety factor should relate to the failure definition (a Chin definition of failure from
a static load test should require a higher safety factor for the same probability of failure
than a Davisson load definition or correlating PDA/CAPWAP result), and to the number
of tests. For a given ultimate capacity, a lower safety factor can be assigned if a larger
percentage of piles are tested (or alternatively stated, a higher allowable load can be
used) since some of the uncertainty is removed by this more extensive testing. Many
codes recognize this truth (e.g. Eurocode, AASHTO 2009, Australian AS2159, et al).

In summary, PDA test should be a restrike with a similar wait time after pile installation, both
static test and PDA test must cause soil failure (PDA test must achieve a good set per blow),
and static test interpretation method should be by Davisson. More testing should result in a
more efficient and thus less costly design.

1.15 Your Responsibility

The PDA/PAK is a very powerful tool when used properly to assess the entire pile driving
process. The minimum suggested education for a person to operate the PDA is a FOUR YEAR
ENGINEERING DEGREE with preferably a civil or geotechnical specialty. Final result
interpretation should only be by a LICENSED PROFESSIONAL ENGINEER who understands
wave propagation theory, pile design, pile driving, and preferably has an understanding of the
geotechnical aspects of the project; a complete training program is highly recommended.
Training materials include powerpoint presentations and example data. Of course, reading the
PDA-W manual and therefore knowing its contents is essential.

While it may appear that PDA operation is relatively simple and that anyone can do it, actually
each job presents unique challenges and often recommendations or other data interpretations
should be made on site. This is particularly true for special test programs or for the first
dynamic test piles driven on site to verify the preliminary driving criteria; in these cases, there
are also often many "visiting dignitaries", and decisions made at this crucial time often can
influence the overall foundation cost. Assigning this responsibility to a technician, or an
engineer with no background in dynamic pile testing or PDA, is not standard or good practice,
and should be totally avoided as it increases liability. A geotechnical engineer should review
results for uplift, settlement, downdrag, etc.

It is expected that you, the PDA operator, will become or are familiar with all aspects of data
acquisition and analysis to assure that correct results and interpretation be made. This implies

1-48 Pile Driving Analyzer Manual, PDA-W February 2009

understanding the theory, principles, applications, and limitations as they apply to your
situation, since interpretation and application of your results are ENTIRELY YOUR
RESPONSIBILITY. You are encouraged to seek advice and review of your work from those
with more experience and more expertise.

It should be noted that the dynamic testing estimates for the pile capacity indicate the
mobilized pile capacity at the time of testing. At very high blow counts (low set per blow),
dynamic test methods tend to produce lower bound capacity estimates as not all resistance
(particularly at and near the toe) is fully activated. At refusal blow counts in restrike,
“superposition” of restrike shaft resistance with end of drive end bearing may be beneficial.

Static pile capacity from dynamic method calculations provide an estimate of the axial pile
capacity. Increases and decreases in the pile capacity with time typically occur (soil
setup/relaxation). Therefore, dynamic testing during restrike tests usually yield a better
indication of long term pile capacity than a test at the end of pile driving. The capacity
of a pile at the time of driving may often be less than the long term pile capacity particularly for
piles driven in fine grained soils (clays, silts and even fine sands). During pile driving, excess
positive pore pressures are often generated. These pore pressures reduce the effective stress
acting on the pile thereby reducing the soil resistance to pile penetration, and thus the pile
capacity at the time of driving. As these pore pressures dissipate, the soil resistance acting on
the pile increases as does the axial pile capacity. This phenomena is routinely called soil

Relaxation (capacity reduction with time) has been observed for piles driven into weathered
shale, and may take several days to fully develop. Pile capacity estimates based upon initial
driving or short term restrike tests can significantly overpredict long term pile capacity.
Therefore, piles driven into shale should be tested after a minimum one week wait either
statically or dynamically (with particular emphasis than on the first few blows). Relaxation has
also been observed for displacement piles driven into dense saturated silts or fine sands due
to a negative pore pressure effect at the pile toe. Again, restrike tests should be used, with
great emphasis on early blows; often a wait period of one or two days is satisfactory but
depends on the permeability of the soil.

Larger diameter open ended pipe piles (or H-piles which do not bear on rock) may behave
differently under dynamic and static loading conditions.

Numerous other factors are usually considered in pile foundation design. Some of these
considerations include additional pile loading from downdrag or negative skin friction, soil setup
and relaxation effects, cyclic loading performance, lateral and uplift loading requirements,
effective stress changes (due to changes in water table, excavations, fills or other changes in
overburden), settlement from underlying weaker layers and pile group effects. These factors
have not been evaluated by GRL and have not been considered in the interpretation of the
dynamic testing results. The foundation designer should determine if any of these
considerations are applicable to this project and the foundation design, and if they are to then
adjust the dynamic test results accordingly.

Comparison of unit friction results from CAPWAP with expected soil strengths (or upper limiting
values) should be performed. The resistance distribution should be evaluated, and adjusted
if needed, particularly near the pile toe. In fact, CAPWAP is considered an essential part of
good PDA practice.

It should be noted that there is a considerable body of literature available on the Pile
Dynamics’s website ( in the TECHNICAL LIBRARY. Specifications are given for
insertion into job specs (and can be edited to fit the specific client or project), and many
REFERENCE PAPERS are given on many topics. It is strongly suggested the practicing PDA

February 2009 Pile Driving Analyzer Manual, PDA-W 1-49

tester should read many of these papers, both to gain understanding. The papers may also be
of benefit when discussing issues with his client. All technical information is available free of
charge when downloading.

Pile Dynamics accepts no liability whatsoever of any kind for use and application by others. If
questions (measurement or result analysis) arise, or you need help, it is therefore then YOUR
RESPONSIBILITY to seek advice or consulting review from PDI or others through peer review

1.15.1 PDA Certification Examination

There is a growing need for verification of ability to properly collect and interpret PDA data. An
independent company from PDI has produced an examination which assesses the PDA
operator’s skills. This company is “Foundation QA pty ltd” based in Melbourne Australia.
Information about the company, the examination, and goals can be found on their website
( It is anticipated that specifying agencies and government authorities
will require PDA users to become certified as part of their overall quality assurance programs,
and in fact several USA State Departments of Transportantion now require certification prior
to working on their projects. The exam is now supported by PDCA (Pile Driving Contractors
Association) in North America. PDI strongly endorses PDA operators to become certified
through this process.

1.16 Help

Help is available by pressing F1 function key or clicking Help on the Menu Bar. If your help
does not work, you might need to upgrade your browser, or check that the files (Help.htm and
Index.htm) exist in the PDAWIN directory (or location where PDA-W exists).

1-50 Pile Driving Analyzer Manual, PDA-W February 2009


2.1 Suggested Applications

The PDA provides a variety of useful information for pile driving operations. An interpretation
overview is given below and in Appendix E ("Helpful Hints ..."). A derivation of the methods is given
in Appendix A. Training can be arranged directly with Pile Dynamics. PDA workshops are given on
a regular basis, and can be tracked under EVENTS on the PDI website ( .

2.1.1 Dynamic Load Testing - Capacity

The Pile Driving Analyzer® (PDA) was developed initially for determining static pile capacity.
Correlations with static load tests have shown generally good agreement and in many cases the PDA
has routinely reduced (or replaced) static testing. Use of the PDA for capacity evaluation requires
that results be carefully analyzed and compared with usually CAPWAP® or perhaps limited static
tests. The capacity computation can be used as follows.

• Design - On large jobs, it can be very cost effective to test for capacity as a function of pile
penetration (and elapsed time, see below). Higher design capacities for the same required driving
resistance (set per blow or blows per inch) can be used to reduce the number of piles or the pile
penetration reduced for the same capacity. Both can result in substantial cost savings. Magnitude
and distribution of soil resistance (skin friction and end bearing) can be estimated.

• Setup and Relaxation Effects - By testing at the end of driving and again after a wait period has
elapsed, soil strength changes as they affect pile capacity can be easily monitored. It is always
recommended to test after an appropriate waiting period so that time dependent soil effects
are included in assessing the service load capabilities of the pile. If setup is substantial, then
either fewer piles or significantly shorter piles can be considered, resulting in substantial savings;
savings would be many times the cost of the testing.

• Supplement Static Tests - For large capacity piles, offshore installations or very small projects,
where static load tests are avoided due to time or cost constraints, the PDA provides valuable
information which should be used with other engineering tools such as static geotechnical methods
or wave equation analysis. Where multiple static tests are initially requested, the PDA can replace
some tests, often at considerable time and cost savings. The PDA can also help determine the
best penetration or driving criteria for piles to be statically tested. Modern design and installation
codes recognize the importance of dynamic pile testing. In the United States, important codes
referencing PDA testing include ASTMD4945, ASCE, PDCA, AASHTO (and individual State DOT
codes), IBC, and the US Army Corps of Engineers. Some codes allow reduced safety factors (or
increased resistance factors for LRFD codes) for additional testing.

• Construction Control (Quality Assurance) - Testing selected production piles spot checks capacity
over the entire site. On larger sites, testing one day every two weeks, or 5 to 10 percent of all piles
distributed over the site provides measurable quality control. Piles with unusual blow counts or
penetrations during initial driving should be dynamically tested, perhaps during restrike. Special
circumstances, project specification, or code may require more piles tested (up to 25%, and even
up to 100% of all piles).

February 2009 Pile Driving Analyzer Manual, PDA-W 2-1

2.1.2 Pile Selection

Dynamic pile driving stresses and capacity for different pile types on a particular job site can be
investigated. On large construction sites, cost savings can be realized if a dynamic test determines
the best pile and driving system to produce the optimal bearing capacity and depth.

2.1.3 Pile Stresses

Measurement of maximum compression force (FMX) and stress (CSX) indicates if damage to the pile
top is likely. The PDA can quantify the effect of reduced throttle settings or air pressures, stroke or
drop heights, changes of the helmet or cushions on the stresses induced at the pile top. The PDA
can determine maximum bending stress (in the plane of transducer attachment) in either individual
strain transducer (CSI) to aid in hammer pile alignment. The compression force (CFB) or stress
(CSB) at the pile bottom can determine if toe damage is a possibility, although local stress
concentrations at toe due to contact with sloping rock cannot be evaluated and might be substantial.

Concrete piles are particularly susceptible to damage from tensile stresses during driving. If damage
is being observed during driving, maximum tension forces (CTN) or stress (TSN) at any location along
pile shaft below the transducers for first 2L/c only due to upward tension from low resistance in first
cycle of wave reflections (or CTX and TSX to include less scientific estimate of tension including
downward tension after 2L/c) can be evaluated as a function of cushion-cap-helmet changes from
standard force-velocity measurements.

2.1.4 Pile Integrity

All pile types (even solid section concrete, H or timber) can be investigated for early tension wave
reflections caused by pile damage, or by bad splices. Estimates of the extent (BTA) and location
(LTD) can be made. BT2 and LT2 give a secondary damage indication and location if it exists. Both
BTA and BT2 are shown graphically on the data display (if they exist). Note there is also an alternate
BTB function. (see section 1.13 or Appendix A for more details on damage).

2.1.5 Hammer Performance

Pile driving resistance (blow count or set/blow) depends on soil resistance, impact stresses and
energy transferred to the pile. Refusal blow counts can be caused by either high soil resistances or
poor hammer performance. The PDA computes energy transferred from the integral of the product
of force times velocity (equivalent to the work done on the pile). The maximum energy EMX can be
compared with the hammer's rated or potential (Wh) energy to determine a TRANSFER RATIO
(which is an indication of the efficiency of hammer driving system). ETR compares EMX with the
manufacturer’s rating, while ETH compared with Wh for open end diesels. The maximum transferred
energy (EMX) is typically 60 to 20 percent of the manufacturer's rated energy (ER). Lower energy
transfer ratios usually indicate a hammer in need of repair or driving system in need of modification.
Typical hammer performance can be assessed through Figures 2.1.1 through 2.1.4, showing
histograms of transfer ratios for differing pile type/hammer combinations. These figures relate energy
transfer ratios at the end of driving. Compare measured with predicted performance from wave
equation analysis. Effects of changes in capblock, helmet or cushions and throttle settings are easily
observed. Preignition of diesel hammers can be detected.

The PDA computes the ram impact velocity (VRI) which can be used to obtain the ram kinetic
energy (0.5M*VRI2) for air, steam, hydraulic, or drop hammers. Comparison of the kinetic energy
with the rated energy (to give actual hammer efficiency) or with the transferred energy EMX may help
determine where the energy losses are occurring. The effectiveness of different hammers can be

2-2 Pile Driving Analyzer Manual, PDA-W February 2009

compared. For external combustion (air/steam/hydraulic or drop) hammers on steel piles, the force
in the hammer cushion (FCP) and the hammer cushion stiffness (KCP) can be computed.

PDA energy measurements of EMX are also applicable to SPT soil samplers and dynamic
penetrometers according to the ASTM D4633 Standard (and also as mentioned in ASTM D6066)
where the term EFV is used (EMX and EFV give exactly the same result). Due to high accelerations,
SPT subsections are instrumented with glued on foil gages rather than using bolt-on strain
transducers (contact PDI for details). SPT energy measurement allows evaluation of a normalized
N-value (called “N60") to compensate for variations in SPT device efficiencies to improve upon soil
strength estimates from SPT N values. N60 can be computed from the measured N, the measured
energy transfer (EMX), and the theoretical potential energy Wh for the SPT ram (Wh) from the

N60 = N* EMX / (0.6 Wh)

It should be noted that an older version of ASTM D4633 was a measurement standard that
had been withdrawn. It considered the normal proportionality between force and velocity
and therefore required only measurement of force and obtained energy from the integral of
the force squared (divided by impedance). The result of this computation are given by the
PDA in the quantity EF2. The method also required several “correction factors”, particularly
for short rods. These correction factors are NOT contained in EF2 but must be applied
separately. However, when rods are non-uniform, this proportionality assumption is in error
and the results are also misleading. Errors were also potentially serious for the joint
masses, and particularly if the joints were loose causing early tension reflections. The time
ratio of first tension return compared with the theoretical 2L/c is shown by the quantity RAT,
which should be between 90 and 120 % for a valid test. To avoid the complexity, and
possible errors from this EF2 method, and considering that the correct method of energy
evaluation integrating the product of force times velocity is contained in the EMX method,
the EMX method is now the defacto standard in ASTM D4633 and in use by many test
agencies today and is the PDI endorsed method of SPT evaluation.

For Open End Diesel Hammers only, the ram stroke (STK) may be computed from blows per minute
(BPM) from the equation

h[ft] = 4.01 (60/BPM)2 - 0.3 (English) = 1.22 (60/BPM)2 - 0.1 (Metric or SI)

The potential energy for open end diesel hammers can be computed from this stroke (STK or h)
times the hammer’s ram weight W.

February 2009 Pile Driving Analyzer Manual, PDA-W 2-3

Figure 2.1.1: Hammer Performance of Diesel Hammer

2-4 Pile Driving Analyzer Manual, PDA-W February 2009

Figur e
2.1.2: Hammer Performance of Steam/Air Hammer on Steel Piles

February 2009 Pile Driving Analyzer Manual, PDA-W 2-5

N = 55; MEDIAN = 35.5%












0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%



N = 34; MEDIAN = 29.4%












0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%


Figure 2.1.3: Hammer Performance of Steam/Air Hammer on Concrete/Timber Piles

2-6 Pile Driving Analyzer Manual, PDA-W February 2009

2.2 PDA Output Quantities

The PDA computes the following quantities. The first letter generally indicates the variable type, i.e.,
F force, D displacement, E energy, R resistance, T time, V velocity. See also Section 2.3 and
Appendix A. Click on the “Q?” icon for a complete list. Bold and underline are more important.

FMX the maximum measured compression uniform force at the transducer location.

CSX compression stress maximum (FMX/Area) 6assumes uniform over section>

MEX maximum uniform strain (CSX/EM) {MEI = CSI/EM}

CSI compression stress maximum of individual transducer with higher value.

VMX the maximum downward velocity at the transducer location.

AMX the maximum acceleration at the transducer location.

EMX maximum energy transmitted past the transducers.

DMX maximum downward displacement at the transducer location.

DFN the final downward displacement at the transducer location.

FT1 force at "impact" or "time 1" (first time marker). {also FT2 - time 2}

VT1 the velocity value at time 1. {also VT2 - time 2}

FVP force versus velocity proportionality at first peak. ( FT1 / {VT1 * EA/C} )

DT1 the displacement at time 1.

WD1 the force of the downward traveling wave at time 1; [FT1 + (VT1 * EA/c)]/2.

WU2 the force of the upward traveling wave at time 2; [FT2 - (VT2 * EA/c)]/2.

CTN maximum computed tension force below the transducers (superimposes max upward tension
with min downward compression in first 2L/C).

CTX maximum of CTN or maximum downward tension after 2L/C.

TSX tension stress maximum (CTX/Area) 6assumes uniform pile> {TSN = CTN/AR}.

CFB computed compression force at the bottom of the pile.

CSB compression stress at bottom of pile (CFB/ARea ; assumes pile is uniform; note that non-
uniform contact stresses with rock are NOT considered)

BPM hammer operating rate in blows per minute.

STK stroke of ram (for open-end diesel hammers only)

February 2009 Pile Driving Analyzer Manual, PDA-W 2-7

BTA pile integrity factor for pile damage analysis. {also BT2 for second damage}

LTD length to damage (below sensors). {also LT2}

SFT skin friction total - not reduced for damping.

SFR skin friction - reduced crudely for damping correction. (SF5 would be SFR with JC of 0.5)

EBR end bearing - corrected for damping (EB6 would be EBR with JC of 0.6)

RSP Case Method static resistance using damping (JC) at fixed time 1. (Time 1 can be changed
with DL or PK). (RTL is RSP with JC = 0)
RP# is RSP with JC = 0.# (i.e., RP5 is RSP with JC = 0.5)

RMX maximum Case Method resistance (using JC) occurs at time TMX during blow.
RX# is RMX with JC = 0.# (i.e., RX6 is RMX with JC = 0.6)

WDX wave down at time TMX; for adjusting RMX for different damping factor.

DBX computed maximum toe displacement at TMX for end bearing piles.

RSU the Case Method resistance using damping corrected for early unloading, i.e., velocity
upward (negative) before the 2L/C time. Primarily for piles with high friction.
RU# is RSU with JC = 0.# (i.e., RU7 is RSU with JC = 0.7)

RAU Case Method static resistance automatically selected and independent of damping JC.
Theoretically correct for piles without skin friction (i.e., end bearing piles).

RA2 Case Method static resistance automatically selected (independent of JC) for piles with
moderate skin friction and end bearing of any magnitude.

RLT energy formula upper limit capacity at refusal blow counts only. {QBC and QUT are formula
result for set from LP blow count and USR input, respectively}

USR "User" input observation value ("USRXXX7")

TRP time from rise to peak ("rise time")

MF0 maximum ram momentum at impact from force impulse. {MWO - wave down}

VRI velocity of ram at impact from momentum and ram mass.

FCP maximum force in hammer cushion for steel piles. (FMX + MH * acc)

KCP stiffness of hammer cushion for ASH hammers on steel piles.

EF2 energy from integral of force squared (obsolete ASTM D4633). {EV2 - from vel squared}RAT
is length ratio to 2L/C for ASTM D4633. (E2F is value of EF2 at time 2L/c after begin of
initial rise. E2E is value of EMX at time 2L/c. New ASTM D4633 just requires measurement
of EMX (EFV).

2-8 Pile Driving Analyzer Manual, PDA-W February 2009

2.3 PDA Capacity Methods

Note: Review Appendix A (for derivations) and Appendix E Helpful Hints (for applications).

2.3.1 Damping Factor Methods: RSP (i.e., RS1, RS2, RSM, RP#)

The derivation of capacity requires that the pile is a uniform pile (linearly elastic with constant cross
section along the length). The PDA computes the total capacity (static plus damping) from the
following Case Method equation:

RTL = [F(T1) + F(T1+2L/C)]/2 + [v(T1) - v(T1+2L/C)] * [EA/c]/2

RTL = [FT1 + FT2]/2 + [(VT1 - VT2) EA/c]/2

where T1 = TP + DELAY and TP is the time of the first relative maximum velocity peak (defined as
impact) and DELAY (DL) is a time required in soils capable of large deformations (large toe quakes)
before achieving full resistance.

The pile bottom velocity, Vb, is estimated from the pile top velocity from:

Vb = VT1+ [FT1- RTL]/(EA/c)

Assuming that damping is proportional to bottom velocity Vb, the static capacity is then:

RSP = RTL - JC(EA/c)Vb

where JC is the Case damping constant (nondimensionalized by the pile impedance EA/c). After
rearranging, the following expressions can be also obtained:

RSP = RTL - JC (VT1* EA/c + FT1- RTL) = RTL - JC [2(WD1) - RTL]

RSP = (1-JC)(FT1 + VT1 * EA/c)/2 + (1+JC) (FT2 - VT2 * EA/c)/2

RSP = (1-JC)(WD1) + (1+JC)(WU2)

To obtain the static capacity, set "JC" as appropriate for the soil. Damping constants, JC, are
empirically derived from correlation with static load tests (see Section 2.4 for recommendations).
RP# can provide a second capacity with a different J than JC to investigate sensitivity (i.e., RP4 is
RSP with JC = 0.4). This "original" method using damping factors can be sensitive to JC, particularly
at low driving resistance. For piles with significant end bearing and large quakes, a "delay" was often
needed to improve correlation. The Maximum Method (Section 2.3.2) automatically searches the
entire record and is now preferred to this original method.

Note: If resistance exceeds 400 blows per meter or 10 blows per inch, the Case Method may
underestimate the full capacity. The hammer must mobilize the full soil strength by
producing a permanent pile set per blow in excess of 2.5 mm (O.1 inch).

February 2009 Pile Driving Analyzer Manual, PDA-W 2-9

2.3.2 Maximum Resistance Method; RMX (RX#)

The maximum resistance RMX is useful for displacement piles with large end bearing where short
rise time impacts or large soil quakes do not activate the toe resistance by the time the peak stress
wave first reaches the pile toe. For RMX, increase the value JC by 0.2 over the regular JC values
by soil classification (section 2.4). JC should also be at least 0.4. RMX may be correlated with
static load tests or CAPWAP to determine the proper JC value. With a fixed 2L/C, time T1 is varied
from Tp to Tp + 30 msec to search for RMX. The time of RMX relative to the first peak (TMX) varies
with JC. WDX is wave down at TMX. DBX estimates the max toe displacement for low friction piles.

RX# can be used for J = 0.# (e.g. RX6 is RMX with JC = 0.6). RMX with JC of 0.5 (e.g. RX5) is
often a reasonable solution, unless CAPWAP or "local experience" suggests a higher JC value.

Note: RMX in silts and clays (low end bearing) with high damping factors should be used with
caution, and RMX for piles in soft soils (low end bearing) with low blow count (high set per
blow) during driving should be handled with great caution since overpredictions can be
serious (be cautious in selecting the blow for analysis if capacity decreases rapidly blow by
blow in a restrike in such a case; it would be advisable to have a correlating static load test).
Further, RMX resistance may only materialize at unacceptably large displacements.

2.3.3 Automatic Method; RA2, RAU

The method RAU automatically computes capacity (without selecting a damping JC) for the first time
where the computed pile bottom velocity (Vb in Section 2.3.1) is zero.

Vb = VT1 + (FT1 - RTL) / (EA/c) = 0

When velocity is zero, damping is zero and the resistance is only static. For piles with zero skin
friction, RAU is independent of JC, is an exact solution and is recommended for piles with very
little skin friction. Increasing friction reduces the RAU result.

The RA2 method is for piles having end bearing plus moderate skin friction and is also unaffected
by JC. Although in general agreement with CAPWAP or static tests, RA2 is not always conservative.
The application of the RA2 method is encouraged; it should be compared with other capacity
methods such as RMX (JC = 0.5). If large discrepancies are noted, then further confirmation by
CAPWAP or load tests is necessary. Actually, CAPWAP is routine standard practice in modern
dynamic testing.

2.3.4 Unloading Method; RSU (RU#)

For the above methods, the displacements monotonically increase (positive velocity for the first 2L/C
period) and thus all resistance is mobilized simultaneously. If the measured velocity is negative
before 2L/C, then the upper portion of the pile resistance unloads before the bottom resistance is
mobilized. The simultaneously occurring capacities are then less than the total resistance. Half the
total friction in the upper unloading zone is added as a correction (K). (RU# is RSU of JC = 0.#.)

K = [F(T3) - V(T3)(EA/c) - F(TP)+ V(TP)(EA/c)]/2

where T3 = (2TP + 2L/C - T0) and T0 is the time of zero velocity (before 2L/C).

RSU = RTL + K - JC(FT1 + VT1 * EA/c - RTL - K)

2-10 Pile Driving Analyzer Manual, PDA-W February 2009

2.3.5 Energy Formula Methods

If energy and displacement are measured at end of driving, capacity can be estimated from

QUT = QBC = 2 * EMX/(DMX + SET)

where SET is the final set per blow from DFN (or USR) for QUT or SET is calculated from the last
blow count for QBC. This method should only be used with "end of drive" data (never with restrike).
A reduction factor (k) should be applied for best results. This method is given in the “Capacity
Monitor” which is accessed from the VIEW menu. A minimum estimate (QUN) for very easy driving
may be obtained by assuming the SET equals DMX, and applying a factor k of 0.5. Comparing this
minimum value with the RX# and taking the larger of the two results in the RQ# method

e.g. RQ5 = Max (RX5, QUN)

2.3.6 A Quick Review of Capacity Results

These capacity solutions assume uniform pile profiles. Piles which are non-uniform should be
analyzed with CAPWAP ®. Capacity is the most difficult determination for the engineer using
dynamic testing methods. The engineer should report capacity only if data quality is good
(particularly true for RA2). Selection of JC can be critical for RMX (or RSP), especially in easy
driving (dynamic methods are better when blow counts exceed 2 or 3 blows per inch - when set per
blow is less than 6 to 12 mm per blow). The time 2L/c must be accurate. A more extensive
discussion is given in Appendix E.

Dynamic testing methods indicate the mobilized pile capacity at the time of testing. At high blow
counts (small pile sets), the capacity mobilized may be less than the true ultimate capacity As in a
static test, if ultimate resistance is higher than the load applied, then a small net set will be observed
and the test only confirms a lower bound solution as not all resistance (particularly near the pile toe)
is fully activated. Dynamic tests have better correlation at blow counts less than 10 blows per
inch (2.5 mm or more set per blow) where the set per blow is sufficient to mobilize soil resistance.

Correlation with static tests is encouraged. Comparison with CAPWAP (a “signal matching” method)
is another alternative, and most dynamic testing uses CAPWAP to confirm the results, and many
codes require a signal matching analysis as part of the procedure. CAPWAP is a rigorous numerical
analysis technique to determine the complete soil model (total resistance and distribution, damping
and quake values) in an iterative procedure. Although CAPWAP is available for PCs, many PDA
users obtain a consulting service for confirmation by a non-partial source. Use of other engineering
tools are also encouraged such as geotechnical static analysis, wave equation, static tests, and
previous experience on similar local job sites.

Static pile capacity from dynamic method calculations provides an estimate of the axial pile capacity.
Increases and decreases in the pile capacity with time typically occur (soil setup/relaxation).
Therefore, dynamic testing during restrike tests usually yield a better indication of long term
pile capacity than a test at the end of pile driving. The capacity of a pile at the time of driving may
often be less than the long term pile capacity particularly for piles driven in fine grained soils (clays,
silts and even fine sands). During pile driving, excess positive pore water pressures are often
generated. These pore pressures reduce the soil effective stress thereby reducing the resistance
to pile penetration, and thus the capacity during driving. As the pore pressures dissipate, the soil
resistance increases as does the pile capacity. This phenomena of capacity increase with time is
called “setup”. For piles in highly sensitive soils, the blow count may be very low (high set per blow)
during driving and caution is suggested in dynamic testing; high resistance on the first restrike blow

February 2009 Pile Driving Analyzer Manual, PDA-W 2-11

may decrease rapidly with subsequent blows. It is then better to be conservative and select a later
blow until a correlating static load test can confirm the higher predictions.

Relaxation (capacity reduction with time) has been observed for piles driven into weathered shale,
and may take several days or weeks to fully develop. To assure that serious capacity reductions do
not occur, piles driven into shale should be tested after a minimum one week wait either statically or
dynamically (with particular emphasis on the first few blows). Relaxation has also been observed for
displacement piles driven into dense saturated silts or fine sands due to a negative pore pressure
effect at the pile toe. Often a wait period of one or two days in silts is satisfactory but depends on the
soil permeability. In any relaxation case, restrike tests should be used, with great emphasis on early
blows. Being conservative suggests analyzing an early blow, even if the hammer performance
is low. It is therefore important in analyzing relaxation cases that the hammer work up to its
capability in as few blows as possible, particularly when blow counts are high (low set per
blow) so that the full capacity is mobilized on the early blows. Additional blows will show
continued capacity increase back to the end of drive conditions, but this increase does not reflect long
term service conditions.

The PDA calculates the pile capacity at the time of testing. Restrike tests are necessary to
determine soil setup or relaxation effects; appropriate waiting times must be considered as
soil strength changes with time. To compare end of drive PDA capacity with a static test
performed a week later invites a poor correlation. Even in sandy soils substantial strength
increases are often observed, particularly in sands with high fine grain content. Equally important
(and fortunately less frequent) is when a reduction of capacity with time occurs (known examples;
piles driven into weathered shales, siltstones, claystones or displacement piles driven into dense
saturated silts or fine sands.) Always RESTRIKE some piles to assess soil strength Gains or
losses (setup or relaxation with time). Restrike periods comparable to wait periods for static
testing produce the best correlation (and calculation of JC) with service loads or load tests.

Capacity results should be reviewed by the appropriate geotechnical engineer. Some

geotechnical concerns in pile foundation design include additional downdrag pile loading from
negative skin friction, soil setup and relaxation effects, cyclic loading performance, lateral and uplift
loading requirements, effective stress changes (due to changes in water table, excavations, fills or
other changes in overburden), settlement from underlying weaker layers, pile group effects and the
difference between dynamic and static loading conditions. Open ended pipe piles, or H-piles which
do not bear on rock, may behave differently under dynamic and static conditions.

2.3.7 Brief Advice in Application of Results

The following brief messages assist the PDA user in interpreting and guide him/her in suggested
procedures in achieving a successful test.

Capacity for remolded soil or changed pore pressure at time of test. Expect shaft resistance
increases (set-up) with wait time after drive, especially in clay, silt, or even fine sand. Temporary
decrease during drive is also due to lateral motions; later capacity increases as soil squeezes in
contact with pile. Resistance can decrease at toe (relax) in soft shale or dense silts/sands. These
changes are then included in the soil capacity evaluated in a restrike test after an appropriate
waiting time.

Dynamic "restrike test" should be made (ideally) at least 7 days after end of drive to confirm strength
changes with time. Capacity increase causes higher restrike blow count for similar hammer energy
so for restrike may need a bigger hammer (determine with WEAP wave equation), higher stroke,

2-12 Pile Driving Analyzer Manual, PDA-W February 2009

less pile cushion to activate full ultimate load (blow count must be less than 400 bl/meter). Hammer
should be "warmed up" (on another pile) so it works at good efficiency for first blow since capacity
can change quickly during restrike blows in sensitive soils or in relaxation cases. Best result is
generally for the earliest "high energy" blow.

Capacity for service loads should require restrike, followed by CAPWAP analysis and/or static test
(two restrikes may allow log time projection). Results must be reviewed/approved by geotechnical
engineer to consider capacity, settlement, negative friction (downdrag), time effect, scour, and uplift
loadings (compression capacity is computed) to avoid serious long term effects. Improper use or
application of results may cause serious problems

Tests during driving are valuable for hammer performance and driving stress. For concrete piles,
testing from the early first blows reviews tension stresses due to easy driving; if stress is high,
reducing stroke or increasing/changing pile cushion should reduce stresses. Both stresses and
hammer performance can confirm wave equation input and establish best driving procedure.

To confirm the full ultimate capacity, pile must achieve a minimum permanent set for each blow of
at least 2.5 mm (400 blows/meter). Measure the pile set per blow very carefully, and note it can
change quickly. If high blow count is expected, use larger hammer, higher ram stroke, or less pile
cushion (thin and/or used cushion) to increase energy and force in pile.

2.4 Proportionality

The force and velocity are proportional by the impedance Z for one directional waves in uniform
elastic piles in the absence of upward reflected waves. Thus for long piles with sensors a significant
distance above ground, force and velocity (times impedance) should appear similar on the PDA data
screens. The force may be higher than the velocity due to early reflections from soil resistance (if the
sensors are near the ground level, or if the impact is smooth from a thick plywood cushion for a
concrete pile) or the pile toe, due to increases in the pile impedance below the sensors, or due to
diesel hammer precompression of the gasses. The force at the first peak may be lower if the pile
impedance reduces as in a tapered timber pile.

In the figure at left, the force

is larger than the velocity
(times impedance) at the first
peak. But the force is also
larger before the initial rise by
a similar amount. Thus the
WaveUp (F-Zv)/2 is “smooth”
through the impact zone (from
initial rise to first peak) and
has no “steps”, and thus this
data is “proportional”.

Proportionality concepts can

be used to assess the
wavespeed since the strain is
proportional to the velocity
divided by the wavespeed. In
the following figures, the first
figure shows the velocity is
higher than the force at the

February 2009 Pile Driving Analyzer Manual, PDA-W 2-13

first peak. The assumed wavespeed WS is too low (and we can also see this in the delayed return
time). If we could not see the 2L/c return time, and IF the soil resistance can be assumed to be
negligible near the pile top (often true due to very soft soils, or even water in “near-shore” situations),
then we can try different WS values to improve the proportionality at the first peak. In the second
figure a higher WS causes the proportionality to be good (and the 2L/c return time is confirms the WS
selection) and perhaps more importantly the wave up to be smooth through the impact zone..

Generally, WC is set to the same value as WS, but may be slightly different than WS due to the
discrete sampling rate (WC = 2*LE / {observed return time}). WS might be different from section to
section of a spliced concrete pile due to different section strengths (age of curing, mix, etc.), and the
WS value should reflect the concrete properties where the sensors are attached (so as to achieve
proportionality when appropriate). Once established for any section, WC can only be progressively

2-14 Pile Driving Analyzer Manual, PDA-W February 2009

slower due the application of more blows as the section experiences microcracks, tension cracks, or
joint connection problems allowing “gaps”. WC is adjusted by the left and right arrow keys to match
the return time (rise to rise method preferred) and accounts for different section properties and
changes in overall wave return due to cracks and gaps, but the basic material wavespeed WS is a
section property and remains unchanged.

2.5 Testing Composite Piles or Drilled Shafts

Testing of drilled shafts (also called bored piles) and augercast piles with drop weights and a Pile
Driving Analyzer® is common in many areas of the world. In referring to such “cast-in-situ pile testing”
(meaning for drilled shafts and augercast piles), there are items which might not be obvious. We
currently recommend to use four strain transducers (one transducer every 90 degrees around the pile
circumference) for drilled shafts and augercast piles to get a better average strain, and this helps for
shafts which can be so nonuniform (especially true for non-cased shafts). If F1 was bad during your
test, you still would have had at least the F3 and F4 opposite pair (should turn F2 off also if it was
opposite to the bad F1 signal). The PAX or PAK can acquire 8 data channels (4 strain plus 4 acc;
actually 2 acc is generally enough as they are usually the same). The number of sensors and type
of accelerometers is selected in the PAK using the F/V Sensors tab of the NEWFILE SETUP (see the
PAX manual for data collection specifics). You need different connection cables for the left
Piezoresistive accelerometer main input plug if you do not already have them (the strain channels
are wired differently so normal Piezoelectric connection cables for right side will NOT work in PAK's
or PAX’s left side); the long main cables are identical regardless of accelerometer type.

Further, note that if the pile has a composite section, it must be considered in the analysis. For
example consider a cased drilled shaft with casing size of 830 mm diameter and 12 mm wall. The
steel area is about 300 cm2; concrete area is 5000 cm2. The total Force in the pile is (strain) x EA.
The analysis of such piles only makes sense if the strain in the steel is IDENTICAL to the strain in
the concrete (this suggests that the ram impact the concrete in compression, through the plywood
cushion, and the concrete expands laterally due to Poisson's ratio effect and increases the bond
between steel and concrete; if you instead hit the steel, then the steel expands and tries to pull away
from the concrete destroying the bond so we do NOT recommend hitting only the steel). Since the
steel modulus is about 5 times higher than that of concrete, the EA of the steel is about 6,300,000
[kN] (300 cm2 x 210,000 MPa) relative to 20,000,000 [kN] (5000 x 40,000 MPa) for the concrete, so
to ignore the steel (which in this example is about 1/4 of entire EA of pile) could be a relatively large
error. Due to relatively high percentage of steel, the average density exceeds normal concrete
density; compute the average density from:

rhoavg = [rho x area]steel x [rho x area]concrete / areatotal. (1)

Similarly you can estimate the average modulus E using the steel modulus and an estimated
concrete modulus.
Eavg = [E x area]steel x [E x area]concrete / areatotal. (2)

The PDA-W program then calculates the estimated wavespeed WS from

WSavg = [Eavg / rhoavg]0.5 (3)

Note that WS for piles with high percent steel will be higher than for piles with concrete alone; the
higher the percentage of steel, the higher the WS (approaches WS of steel as %steel approaches
100%). If you can possibly observe the overall wave speed WStrue from 2L/c reflection (if it exists and
is obvious), then you can adjust the final modulus E = rhoavg x [WStrue]2.

February 2009 Pile Driving Analyzer Manual, PDA-W 2-15

For a thick shell (e.g. 830 x 12 mm diameter shell), it is better to drill and tap the steel shell only
rather than cutting "windows" in the shell and bolting to the concrete. The windows could be a rather
significant portion of the steel (reducing the EA of the steel), creating a "nonuniform" pile, making a
"stress concentration" point, and cutting the steel probably does not help the concrete quality (due
to heat from torch). Again remember the basic assumption is that the strain in the steel MUST equal
the strain in the concrete for such composite piles if any analysis is to be meaningful, so attaching
to the steel should be at least equivalent if not preferred. You can always cut windows later if you
don't like what you see from attaching to steel.

Obviously the dynamic test should be performed after a wait time sufficient to gain good concrete
strength (there is evidence that the soil also needs time to recover its strength). The transducers
should be attached to the pile a minimum of one diameter below the shaft top or any other non-
uniformity changes, and preferably two diameters below. For large diameters this creates a hardship.
Since many shaft tops are flush with the ground, a considerable excavation is required and the one
diameter location is then more attractive (definitely use 4 strain transducers). The force can also
be measured from the mass times the deceleration of the ram, and this may significantly
reduce the effort for testing drilled shafts. This requires other considerations and the PDA-W
program can do all the calculations necessary. Contact PDI for further details of this “F=ma” method.

PDI has also developed “top transducers” for measuring force. Contact PDI for details.

As a good alternative when using strain transducers, the test shaft can be "built up" above ground
surface. This may be particularly attractive if there are protruding rebars which can be hidden and
protected in the build-up section. A steel casing provides both a form as well as strengthens the shaft
top. Choose a transducer location which is on the actual shaft; attaching near (just above or just
below) cross section changes (e.g. in casing thickness or concrete strength) can complicate testing
and analysis; the larger the nonuniformity the farther below the change you should locate the sensors
to avoid "end effects or local bending effects” or “dead corners”. Ideally, the shaft at and below the
measuring location will be uniform in all respects, so a small excavation if the shaft has been later
extended above ground surface is generally recommended so sensors are attached to original pile
concrete. Obviously, casting the main shaft and the extension at the same time with similar concrete
is then beneficial (lower portion of shell can be removed to create a smooth surface for sensor
attachment). The entire built up section can be removed after the test has been completed (rebars

It helps to know the shaft’s total volume of concrete; compare it with the theoretical volume to guide
in making a nonuniform pile model during CAPWAP® (This may be particularly helpful and easy if the
volume per unit depth is known from automated volume monitoring equipment such as the PDI “Pile
Installation Recorder for Augercast piles (PIR-A)). All test results should be subjected to CAPWAP
analysis where both soil and shaft impedances versus length can be adjusted. It may be helpful
(some cases required) to use the "radiation damping model" in CAPWAP to achieve reasonable
capacity solutions. It should be noted that the density of typical grout used for augercast piles is
about 10% lower than for normal concrete. In many cases, the top concrete for shafts drilled under
slurry or grout from an augercast may have relatively lower strength and thus lower WS at the sensor
location than for the overall pile average due to some material contaminations. Therefore the
apparent shaft impedances below the top may be higher than for the top due to both concrete volume
variations and material strength variations.

The shaft top to be tested is cushioned by a few layers of plywood placed directly on the concrete
(perhaps 50 mm if the pile top is reasonably level and smooth; more if there are variations). The
diameter of the plywood cushion should be about 80 (to at most 90) percent of the shaft diameter to
centralize the blow. A steel striker plate of 50 mm to 100 mm thickness is sometimes placed above

2-16 Pile Driving Analyzer Manual, PDA-W February 2009

the plywood and with a diameter equal to the cushion diameter. This striker plate is often NOT
necessary and in fact would complicate the F=ma measurement of force on the ram (and is
discouraged for F=ma applications).

The ram for testing should be guided so the pile top and ram bottom are parallel during impact to
make as uniform a contact stress as possible during impact. A short section of pile leads, or an
external guide tube or cage, or internal guide shaft are three alternatives. The ram weight required
is generally at least one percent of the ultimate load to be tested; for shafts with a high end bearing
a minimum of 1.5 to 2 percent of the ultimate load is more desirable. Higher ram weights are
generally always acceptable, but are obviously more difficult to lift so there is a tradeoff. In fact, ram
weights of 5% or more of the ultimate test load can be used to generate Rapid Load Tests and meet
the specification for these tests. Larger rams (e.g. 2% of ult capacity) are REQUIRED as shaft
diameter and required loads increase. The ram is ideally a simple drop weight which can be raised
to a variable height of drop. Rams using both steel and concrete have been constructed (PDI would
be happy to show you a few possible ideas to get you started). The largest known ram weight used
to date is about 80 tonnes (800 kN); there are many 20 tonnes (200 kN) rams. Rams can be made
in segments which can be assembled (e.g. bolted together) at the job site; segmental weights can
be helpful in transporting the weights, or in lifting).

The test is usually started with a relatively low drop height (say 1.5 ft or 500 mm). The permanent
set for each and every blow should be carefully measured. It is best if the ram is a complete free fall
(e.g. not connected to crane by a cable and winch). Free release mechanisms of several types have
been used in many different countries, including mechanical trips, hydraulic release devices, and
cutting a support cable (using hydraulic cutters). The uniformity of stresses can be checked, and
adjustments to cushion thickness or ram/shaft alignment made before application of subsequent
blows. Subsequent blows can be made from a higher drop height if stresses are acceptable. Higher
drop heights need not be made if the set per blow was large (say more than 3 mm per blow) and the
ultimate capacity therefore then fully mobilized, or if the client is satisfied that a certain minimum load
has been achieved (this perhaps depends on the safety factor desired for the test).

This brief description gives you a generic overview of how to approach the testing this particular pile.
Some alterations or adaptions may be required.

2.6 PDA Testing on Steel Followers (or “dollies”) for Concrete Piles

Concrete piles are sometimes driven to final depth using steel followers. In many cases the pile top
is driven below water or below the ground surface using followers. The need for testing concrete
piles using steel followers is then obvious. In many cases it can be a very good idea, and in some
cases it may be the only alternative to get data. (The follower's actual effect can be found by testing
simultaneously on pile and follower with PAK or PAX, and by testing a similar pile without the

Testing concrete piles using steel followers requires extra considerations and extra care. At issue
primarily is first how to take the data, and second how to analyze the data. If the sensors are
attached to the pile, then it does not matter if there is a follower and the data collection and analysis
of the pile data is normal. If the sensors are attached to the steel follower, the following is the
suggested procedure.

1) Ideally, the follower and pile have the same impedance. If this is not the case then the pile must
be considered non-uniform. The sensors should be placed on a uniform section of the follower
to avoid end effects at either top or bottom (ideally at least one diameter from any nonuniformity).

February 2009 Pile Driving Analyzer Manual, PDA-W 2-17

2) The cushion between the follower and pile should be kept to a minimum (like one sheet of
plywood, or better still a sheet of lead); if a thick plywood cushion is needed to protect the concrete
pile, then it should be primarily placed above the follower. If the cushion between the pile and
follower allows for significant movement, then the pile will be non-uniform.

3) Data should be taken with inputs EM, WS, and SP of the material to which the sensors are
attached. (e.g. If the sensors are attached to a steel follower, then steel properties are input.)

4) IF the pile and follower impedance are identical and IF there is minimal cushion, ONLY THEN can
you properly use the Case Method and assume a “uniform pile” (implying “uniform impedance”).
Input an equivalent WC such that the total time is identical to the sum of the individual portions....

2*LEtotal / WC = 2*{LEsteel / WSsteel + LEconcrete / WSconcrete}

or in other words...

WC = LEtotal / {LEsteel / WSsteel + LEconcrete / WSconcrete}

5) Even IF the pile and follower impedance are identical and IF there is minimal cushion, follower and
pile alignment must be very good to avoid a gap between the two. If data proportionality is then
good for the first 2L/c, then you can analyze as a uniform pile. Stresses in the pile and even
capacity might be acceptable provided there is NOT a tension reflection at 2L/c (Wave up at 2L/c
must be in compression and force must be greater than zero).

6) Even IF the pile and follower impedance are identical and IF there is minimal cushion and IF
follower and pile alignment is very good (no gap) SO THAT data proportionality is then good for
the first 2L/c, then you can analyze as a uniform pile and can probably rely on BTA. HOWEVER,
if the BTA is very small such that it generates a real tension (negative or zero force BEFORE 2L/c,
then a gap has opened and the BTA may be less than shown by the PDA.

Obviously there are many conditions (IF's) in the above (points 4, 5 and 6). The usual case when
sensors are attached to the follower will have the follower/pile system being NON-UNIFORM
(different impedances). In that case we MUST analyze the data with CAPWAP and the following
rules are then followed.

a) The pile must be modeled as nonuniform (steel, plywood, and concrete with their respective
lengths, modulus, density, and area).

b) The first pile segment ABSOLUTELY MUST be completely uniform in CAPWAP and have
impedance therefore equal to EM*AR/WS (all of the steel). This can be checked in the model
output. If the length of steel is less than one full segment, then you MUST adjust (increase) the
number of pile segments (and thus soil segments by same ratio) until the first segment is uniform
steel with impedance of EM*AR/WS, or alternately change the steel length to the section change
to equal the first segment (this change of length will cause a relatively small error, especially
compared to the potentially major error caused by the first segment not being uniform).

c) You can adjust the plywood cushion effect on the element impedance in the ZG screen for the
element that has this transition.

d) The Case Method capacity is wrong since the pile is non-uniform. Further the 2L/c time is in error.
Capacity estimates from the PDA-W should therefore be disregarded. If a capacity estimate is

2-18 Pile Driving Analyzer Manual, PDA-W February 2009

required, then the data must be analyzed by CAPWAP. IF you follow these rules (and IF there is
no tension), THEN the CAPWAP capacity will be acceptable.

e) BTA from the PDA and Case Method for a NON-UNIFORM pile are meaningless. The BTA of the
pile could be investigated also using the ZG fast impedance changes. However, BTA will never
be as reliable from data on a follower compared with data taken with sensors directly on the pile.
If the test on follower (and CAPWAP) reveals some potential BTA problem, and the BTA question
is absolutely crucial to know with certainty, then there should be a re-test with the sensors placed
on the pile (if this is at all possible).

f) If BTA is the prime question to be answered, then you should make effort to place the sensors on
the pile so that you have the best quality data to analyze.

To process data from sensors on a steel follower driving a concrete pile with CAPWAP which have
incorrect field input data (e.g. incorrect input of EM and AR of the concrete pile when the sensors
were on steel) requires corrective action. Simply input the correct values of EM and AR for steel and
reprocess the data. Then model the pile correctly as non-uniform in CAPWAP as steel (location of
sensors), plywood, and concrete as described above.

February 2009 Pile Driving Analyzer Manual, PDA-W 2-19

2-20 Pile Driving Analyzer Manual, PDA-W February 2009
of the



The Case Method

Wave Mechanics, Theory and Derivations,
Examples and Practice Problems

© 2012 Pile Dynamics, Inc.

30725 Aurora Road
Cleveland, OH 44139
Examples and Practice Problems


In order to understand how the PDA calculates certain quantities from pile top force and velocity
measurements it is necessary to understand the underlying theory. The best way to study the
underlying theory is with the treatment found in Timoshenko’s Theory of Elasticity (note
references can all be found on which is a summary of closed form solutions and
examples developed by various mathematicians in the 19th century. These closed form solutions
have been applied to the Case Method measurements. The collection of formulas and
equations developed for the purpose of calculating soil resistance, pile stresses, hammer
performance parameters, pile integrity factors and other quantities are all part of the Case
Method which was developed during the late 1960s and 1970s both at Case Western Reserve
University and Pile Dynamics. Besides looking at the papers and books referenced in this
description, ample references contained in are recommended reading for the
PDA user. Furthermore, the user should be familiar with ASTM D4945, latest edition.

The following derivations of wave speed and proportionality are not strictly correct in a
mathematical sense. They should be understood as an illustration of the basic wave
propagation process and should provide the reader with a “feel” as to what is happening in a
pile when it is struck by a rigid mass. All formulas of the Case Method were derived on the
assumption of a uniform (constant area, elastic modulus and mass density), linearly elastic rod
whose length is much greater than its diameter or width. We may sometimes violate these
requirements in actual piling situations and we then should try to evaluate how large an error
may result.

This document uses a variety of symbols and 2-letter codes for various material and other pile
properties. In derivations we will represent with Greek letters certain material properties while
for actual problem solving we may show the PDA 2-letter codes. The following are traditional
and PDA symbols frequently encountered in this document.

Name Common PDA SI Units US Units

Symbol Symbol
Elastic Modulus E EM MPa Ksi
Specific Weight SP kN / m3 lbs / ft3
Wave Speed c WS m/s ft / s
Cross sectional ARea A AR cm2 inch2
Mass density ρ - kg (N s2 / m) kips s2 / ft
A variety of subscripted symbols are used to represent the various dynamic quantities and in the
mathematical formulations; however for certain values of these curves, the PDA uses 3 letter
acronyms for output description. Important quantities are listed in the following table (additional
quantities can be found in the PDA’s “Q?” listing.

Quantity description Representation in Related PDA Related PDA output

(units) Equations Quantities acronyms
Force (kN, kips) F(t) Max. Force, FMX,
Force at time 1, 2 FT1, FT2
Acceleration (g’s) a(t) Max. Acceleration AMX
Velocity (m/s, ft/s) v(t) Max. velocity VMX,
Velocity at time 1, 2 VT1, VT2
Strain (10-6) ε(t) Max. Strain MEX
Stress (MPa, ksi) σ(t) Max. Measured CSX
Compressive Stress
Wave-down (kN, kips) Fd(t), Fd1 Force in Wave-down WD1
at time 1
Wave-up (kN, kips) Fu(t), Fu2 Force in Wave-up at WU2
time 2
Displacement (mm, u(t) Max. Displacement, DMX,
inch) Displacement at end DFN
Transferred Energy Et Max. transferred EMX
(kJ, ft-kips) energy
Transferred Energy ηt Transferred Energy ETR
Ratio (1) Ratio (or efficiency)
1. The Wave Speed

Consider a rod whose particles are at rest at time t, just before an impact occurs. Suppose then
that the rod is suddenly loaded by a force, F. A short time, t, later, the impact force has
compressed a portion of the pile top having length L.

Since L has been compressed within a time t, we consider the speed with which the pile top
has been compressed the wave speed
c = L /t (1.1)
Because of the compression, a point, A, has moved a distance, u. The displacement, u,
being the result of compressing the rod with the impact force F over a distance L can be
computed from rod cross-sectional area, A, and elastic modulus, E, as
u = F L / EA (1.2)
The velocity of the point A pile particle, actually its change of velocity due to force F, is called
the particle velocity v. It can be calculated from the deformation u divided by the time
increment during which it occurs.
v = u /t (1.3a)
Combining Eq. 1.2 and 1.3 the change of particle velocity can be calculated from
v = (F / EA)(L / t)
And remembering Eq. 1.1 we obtain
v = F c / (EA) (1.3b)
Since this velocity was achieved during time period t, we can also calculate the acceleration of
our particle at point A.
a = v / t (1.4a)

a = F c / (EA t) (1.4b)

Using Newton’s Second Law, which is
F=ma (1.5a)
and knowing that the accelerated mass at point A is equal to the product of the mass density of
the pile material, ρ, the cross sectional area, A, and the compressed pile length, L (m = ρAL),
the force can now be written as
F = (ρ A L) F c / (EA t) (1.5b)
After cancelling the A and F terms and remembering that L/t is the wave speed c, we obtain
c2 = E/ ρ (1.6)
Thus, we have found that the wave speed, c, is the square root of the product of mass density
and elastic modulus and , therefore, depends only on the pile material properties and not, for
example, the frequency of the applied force (admittedly though this is only true for our
simplifying assumptions of a very slender, elastic rod).

In summary, let us remember that

(a) the “Wave Speed” is the speed with which a compression (or tension) wave (or zone)
moves along a rod.
(b) the “Particle Velocity” is the speed with which a particle in a rod moves as a wave
passes by.

Example Problem 1: Wave Speed

Calculate the wave speed for concrete with a dynamic elastic modulus of 35,000 MPa (5,000
ksi) and unit weight γ = 24 kN/m3 (150 lb/ft3). Repeat the calculation for timber (E = 12,000 MPa
or 1,800 ksi and γ = 8 kN/m3 or 50 lb/ft3) and steel (E = 210,000 MPa or 30,000 ksi and γ = 77
kN/m3 or 492 lb/ft3).

2. Proportionality and Pile Impedance

Let us again consider the uniform elastic rod and a stress wave traveling along its length at
wave speed, c. If the force at the wave front is F, we noted in the previous section, Eq. 1.3b,
that the change of particle velocity and v = u/t can be expressed as a relationship between
the force, stress or strain and the particle velocity in a stress wave:
v = F c/EA (force) (2.1a)
v = f c/E (stress) (2.1b)
v=εc (strain) (2.1c)
These relationships express a proportionality between the particle velocity v and either applied
force or stress or strain. The proportionality factors are composed of pile material properties A,
E and c and/or ρ. While we normally use velocity, v, in these expressions, it is important to
remember (a) that the force really caused an increase of velocity (if the velocity was not zero
before impact) and (b) that this proportionality only holds if no effects other than one wave
traveling in a given direction is present.

The inverse of the proportionality constant, c/EA, is

Z=EA/c (2.2a)
which is also called the “pile impedance”. This term implies that rod offers a resistance to
(impedes) the change in velocity. In fact, the impedance (which has the units of force divided by
velocity) is that force which changes the pile particle velocity by 1 m/s (ft/s) Note the following
alternate forms of impedance. For example, by replacing E in Eq. 2.2a with c2ρ (Eq. 1.6) we
Z=ρcA (2.2b)
or after replacing ρ A with the pile mass per unit length M/L
Z=Mc/L (2.2c)

Example Problem 2: Impedance

Compute the impedance of a pile for a 27.5 cm (11 inch) square concrete pile of (30 m (100 ft)
length using concrete properties as in the wave speed example. Do the computations for all
three equations 2.2 a, b and c.

3. Basic Wave Mechanics
3.1 The wave equation

The foregoing considerations can be put in a stricter mathematical form (from

Hooke’s and Newton’s Laws) leading to the one-dimensional wave equation:
ρ (∂2u/dt2) = E (∂2u/dx2) (3.1)
where u is the rod displacement at time t and location x and where the left and right
hand partial derivatives are the acceleration and strain in the rod, respectively. This
equation is referred to as the linear one-dimensional wave equation which has a
general solution
u = f(x-ct) + g(x+ct) (3.2)
which implies that a displacement pattern in the rod may consist of two components,
g and f. Note that the f displacement pattern will have the same argument if, for
increasing times t + t, the x-coordinate increases by c t; similarly the g pattern will
have the same argument if, for increasing times t + t, the x-coordinate decreases by
c t.
f(x – ct) = f(x+ct – c(t +t)) (3.3a)
g(x + ct)= g(x-ct + c(t +t)) (3.3b)
Thus, the g and f displacement patterns have
merely shifted downward (positively) and
upward (negatively) along the pile as time
increases. They shift at a speed c as seen
before. We will, therefore call the two
traveling displacement patterns a downward
wave and an upward wave.

Since the particle velocity, v, and the

acceleration, a, are time derivatives of the
displacement, the velocity and acceleration
patterns are also downward and upward
traveling waves. Similarly, since the strain,
stress and force can be derived from the
displacement pattern by derivative with
respect to x, these three quantities also do
not change pattern as the shift upward or downwards along the pile. The solution to the wave
equation shows also that the total particle displacement, and therefore all of its derivatives, is
the sum of the displacements in the upward and the downward wave. Thus,
Displacement: u = u d + uu (3.4a)
Velocity: v = vd + vu (3.4b)
Acceleration: a = a d + au (3.4c)
Strain: ε = εd + εu (3.4d)
Stress: σ = σd + σu (3.4e)
Force: F = Fd + Fu (3.4f)

If we apply these findings to piles during impact, then we may get the following situation
(assuming no soil resistance).

Remember that within the initial downward input wave, there are compressive forces, causing
proportional downward directed particle velocities. Let us designate the forces and velocities in
the downward wave with the subscript “d” and write the proportionality condition as:
Fd = Z vd (3.5)

After a time L/c (L is the pile length), the impact wave caused by the pile driving hammer arrives
at the pile bottom where it is reflected. An example for a wave induced by a pile driving hammer

is shown in the above figure. We will study what happens at the time of wave reflection a little

As we will see in more detail, an upward traveling tension wave has a downward directed
particle velocity (like the downward traveling compressive wave), which means that on a free
pile bottom, the velocity (and thus the displacement and acceleration) doubles while the forces
cancel each other.

3.2 Upward and downward traveling waves

We now define a sign convention:

 Compressive forces, stresses, strains are positive
 Tension forces, stresses, strains are negative
 Downward directed particle velocities, displacements, accelerations are positive
 Upward directed velocities, displacements, accelerations are negative.

Consider an impact against the bottom of the pile. It will generate an upward traveling
compressive wave with upward directed (negative) particle velocities. Thus for upward traveling
waves the proportionality condition includes a minus sign.

Upward traveling waves, therefore, have a particle velocity that is negative (upward) for positive
(compression) forces and positive (downward) for negative (tension) forces. Thus, for upward
traveling waves
Fu = -Z vu (3.6)
The total force, F, and velocity, v, measured at any location is the total force and total velocity at
the measurement point and, as we have seen in the general solution to the basic wave
equation, they are the result of superposition of the forces and velocities in the downward and
upward traveling waves.
F = Fd + F u (3.7)
v = vd + vu (3.8)
If the velocities are converted to forces by multiplication with the impedance Z, Eq. 3.8 becomes
Zv = Fd – Fu
which can be combined with Eq. 3.7 to solve for the forces (and thus also velocities) in the
upward and downward traveling waves.
Fd = ½ (F + Zv) (3.9a)
Fu = ½ (F - Zv) (3.10a)
By proportionality we also find that
vd = ½ (F/Z + v) (3.9a)
vu = ½ (-F/Z + v) (3.10b)

In other words, if we measure the force, F, and the velocity, v, at a point of the pile, then the
force in the downward traveling wave at that point can be determined from the average of force,
F, and velocity times impedance, Zv. Similarly, the force in the upward traveling wave can be
determined from one half of the difference between force, F, and velocity times impedance, Zv.

3.3 The Classical reflection model

The type and magnitude of the reflection depends on the type of resistance at the pile bottom.
Let us first consider the simple case of a free pile bottom. If the compressive wave arrives at a
free pile bottom an imbalance exists since the wave has no pile mass to accelerate and no pile
material to strain; therefore a reflection occurs. Because the pile end is free, the force at that
point must be zero. The classical way to look at what happens at the free end of a pile when the
compressive wave arrives is described in the following figure. On the left we see the
compressive wave moving downward in the real pile. At the same time a wave is assumed to
travel upwards in a virtual pile. The two waves will arrive at the same time at the real pile
bottom. In order to satisfy the condition of no force at the pile bottom, the upward traveling wave
has to be a tension wave which moves the particles down ward. So after the reflection is
finished, there is an upward traveling tension wave in the real pile which has downward directed
particle velocities.
Putting these considerations in equation form, if the pile bottom is free (in other words, if there is
no resistance force acting at the bottom and the resistance R = 0) from superposition we obtain
F d + Fu = 0 (3.11)
Therefore, the force in the upward traveling wave is equal and opposite the downward traveling
incident wave.
Fu = -Fd (3.12a)
The associated velocities are
vu = -Fu/Z = Fd/Z = vd (3.12b)
And therefore
v = vu + vd = 2 Fd/Z (3.12c)
In other words the velocity at the bottom will be twice the velocity in the downward (or upward)

If we now consider a pile encountering a rigid pile bottom support, then the pile bottom condition
is one of zero motion (velocity, displacement, acceleration). Thus when the compressive wave
arrives at the bottom, the reflection wave has to have an upward directed (negative) particle
velocity (so that the velocities cancel). The proportionality condition for the upward traveling
requires a negative sign and we therefore get an upward traveling compressive force (positive)

Example Problem 3: Wave-down and Wave-up Values

Given force and velocity at the pile top of a square prestressed, precast concrete pile (see figure
below), what is the magnitude of both the downward and upward traveling wave forces at both
times, t1 and t2.

(Note, force and velocity values were rounded and cannot be exactly scaled in the figure).

4. Soil Resistance Assessment

4.1 Resistance Waves

Suppose that an impact wave has reached a point along the pile which is located a distance x
below the top. The impact wave reaches that point at a time x/c after the impact. The soil
responds to the pile’s sudden downward motion, caused by the impact wave, with a sudden
upward directed resistance force R. This shaft resistance force R is a concentrated passive
force representing the unit resistance times the pile perimeter times a certain length increment
Δx. Note that R is a passive force, i.e., it acts against the direction of motion and only while the
pile is moving (residual stresses are ignored at this point).

The suddenly applied force R creates upwards and downwards traveling waves above and
below. The two waves add their force and velocity effects to the impact wave (superposition).
The two resistance waves each have a magnitude R/2. To satisfy equilibrium and continuity, the
upward wave is in compression and the downward wave in tension. Both waves therefore have
an upward directed particle velocity satisfying the continuity condition at x (the pile does not tear
apart). The forces in the waves together balance R: the compressive wave pushes downward
above the resistance force application; the tensile waves pulls downward underneath the force

Again, the forces and particle velocities in the upward and downward resistance waves are
Fdr = - ½ R (4.1a)
Fur = ½ R (4.1b)
vdr = -½ R/Z (4.1c)
vur = -½ R/Z (4.1d)
which means that the forces are compression and tension to balance the resistance force and
the particle velocities are directed upward (negative) in either wave to maintain continuity.

The end bearing, Rb, is a force applied at the pile toe and therefore generates only a single,
upward traveling compression wave with upward directed particle motions. Since the end
bearing is only activated by the impact wave at time L/c, its effect will be felt at the pile top only
a time 2L/c after impact.

4.2 Shaft Resistance from Force – Velocity Difference

Of course, we can divide the pile in many sections, each having a concentrated shaft resistance
force, however, in the following we will only consider one shaft resistance force Ri located at x
as representative of all shaft resistance forces. The upward traveling compressive shaft
resistance wave caused by Ri reaches the pile top at time t = 2x/c after the impact. The tensile
resistance wave reaches the pile bottom together with the impact wave at time t = L/c where it is
reflected in compression while the impact wave is reflected in tension. Both the original tension
wave from the shaft resistance waves, now compressive, and the impact wave, now tensile, are
joined by the end bearing compressive wave and all three waves then travel upward to the top
where they arrive at time t = 2L/c. This process is illustrated in the Depth-Time (x-t) plot below.
Note that compressive and tensile waves are represented by solid and dashed arrows,
respectively, and that the waves due to impact, shaft resistance and end bearing are
distinguished with white, aqua and red colors.

If we assume a fixed pile top (velocity is prescribed), then forces in the upwards traveling
resistance wave have to be met by a downward traveling compressive wave so that there is no
change in velocity at the pile top. Therefore, the pile top force will suddenly increase by a
magnitude Ri/Z, relative to the pile top velocity times impedance (vZ), before time 2L/c. (Note
that we could also have assumed a free top in which case the forces would have to cancel and
the velocities would double leading to a sudden negative velocity change at the pile top of
magnitude –Ri/Z relative to the pile top force (F).) In any case, upon arrival at x/c, the upward
traveling compressive shaft resistance wave causes a separation of the pile top force and
velocity curves by an amount Ri.

Actually the foregoing consideration is also valid even if the measurements are not made at the
pile top. Consideration of the upward compressive resistance wave of magnitude ½Ri, having

an upward particle velocity equal to –½Ri/Z gives a total difference between the force and
proportional velocity of Ri = Ri/2 - (-½Ri/Z)Z. Therefore it is not an assumption or requirement of
the Case Method that measurements be taken at the pile top. Since we are measuring both F
and v we can separate upward from downward waves at the point of measurements. In fact,
measurement at the very top would contain undesirable local contact stresses, so we generally
measure at least two pile diameters below the top (preferably one circumference).

Example Problem 4: Shaft Resistance from Force and Velocity times Impedance

Study the following graph and notes.

(a) Determine the apparent shaft resistance force, Ri, acting between points A and B. Calculate
Ri as a percentage of the maximum impact force.
(b) Question: Is Ri the total shaft resistance?
(c) Question: Is Ri a static resistance force?

4.3 Resistance from Wave Up

While it is instructive to work with F and vZ, it is even more helpful to work with the force in the
upward wave, Fu. This is because the Wave-up does not include impact waves or other
downward wave effects which distract from what we want to see: the effect of soil and pile end
on the top measurements.

We have seen in Section 3.2 that the forces in the upward traveling and downward traveling
waves (in the following we will just refer to Wave-up and Wave-down to refer to these forces)
can be calculated from the measured force and velocity with the following two simple formulas.
Fd = ½ (F + vZ) (3.9a)
Fu = ½ (F - vZ) (3.10a)
In other words the force in the Wave-up is ½ the difference between F and vZ which in turn is ½
the shaft resistance according to what we learned in Section 4.2. We, therefore, can state that
Ri|B-A = 2 (FuB – FuA) (4.2)
In words: the shaft resistance acting on the pile between points A and B is equal to twice the
Wave-up force at time tB minus the Wave-up force at time tA.

The following graph shows the transformation of the measurements to the wave forces in the
typical PDA display. The graph incudes scale (or rather full scale range) information [measured
force, F, force in Wave-up, WU, Wave-down, WD, (all forces in kN), measured velocity v (m/s),
total display time, TS (ms) and Start of display from the beginning of the record, TB (ms)] and
the active sensors A3, A4, F3, F4.

This is a record where the soil resistance is really low; in fact F – vZ and, therefore the shaft
resistance is practically zero just before the return of the impact wave. At time 2L/c the velocity
sharply increases and the force decreases. At that time the wave up curve, being one half of the
difference between F and vZ, goes negative, indicating that Wave-up is then a tension force.
Before that Wave-up is practically zero, again, indicating a very low shaft resistance.

4.4 Calculating the Soil Resistance from Wave-up and Wave-down

Let us designate as time t1 the time when the impact wave passes by the sensor location and as
time t2 = t1 + 2L/c when the toe reflected impact wave returns to the sensor location. Thus, at
time t1 we have an impact wave of magnitude Fd1 traveling downward towards the pile toe. If the
resistance force Ri acts constant throughout the time x/c < t < L/c, then at time t2 = t1 + 2L/c the
upward traveling wave contains the effects of

(1) the impact wave after reflection at the pile toe where it became an upward
traveling tension wave of magnitude –Fd1
(2) the directly upwards traveling compressive wave from the shaft resistance,
magnitude ½ Ri
(3) the initially downward traveling tension resistance wave, now traveling upward in
compression after reflection at the bottom, magnitude ½ Ri
(4) the compressive wave caused by the end bearing, magnitude Rb

Combining all upwards waves at time t2 we obtain in the order (1) through (4) for the Wave-up at
time t2:
Fu2 = -Fd1 + ½ Ri + ½ Ri + Rb (4.3a)

The second and third term on the right hand side of Eq. 4.3a represent the total shaft
resistance; adding to it the end bearing makes up for the total resistance Rtl. Thus, the
combination of all upward traveling waves contains the resistance and the bottom reflected
(negative) impact wave of time t1. We can, therefore, rewrite Eq. 4.3a as:
Rtl = Fd1 + Fu2 (4.3b)
Eq. 4.3 can also be expressed in terms of measured forces and velocities at time t1 and t2 as:
Rtl = ½ [F(t1) + F(t2)] + ½ Z [v(t1) – v(t2)] (4.4)
Rtl is the total resistance encountered during a complete passage of the wave between time t1
and t2, i.e., during a time period of 2L/c. There are differences between this total resistance and
the ultimate static capacity of the pile and various considerations are necessary to calculate Rs.
(a) Elimination of soil damping.
(b) Proper choice of time t1 such that Rs is fully mobilized when F and v samples are
(c) Correction for an Rs that decreases between t1 and t2 because of early pile
rebound or unloading indicated by a negative velocity before 2L./c.
(d) Time dependent soil strength changes (setup or relaxation). Since the dynamic
methods give the resistance at the time of testing, it is always recommended to
test piles both at the end of driving for an assessment of the strength of the
remolded soil and by restriking after a waiting period for the determination of the
long-term ultimate capacity. It should not be surprising that the capacity at the
end of driving is not equal to the long term pile capacity after an extended waiting
period. The waiting period has to be appropriate for the type of soil at the test
(e) The pile penetration under the hammer blow. The pile must experience a
permanent set (in general we recommend at least 2.5 mm or 0.1”) during the
testing for a full mobilization of the soil resistance. If no (or very little) permanent
set is achieved then the indicated capacity relates to the mobilized value only
which may be less than the pile’s ultimate capacity. This condition is roughly
analogous to a static proof test not run to failure because of a limitation of the
test setup. The pile set should also not be too large (say more than 12 mm)
under the test blow or dynamic effects in the soil could lead to calculated
capacities which are greater than the ultimate pile capacity.
Considerations (d) and (e) are self-explanatory. The first three considerations will now
be investigated in more detail.

Example Problem 5: Calculating total Resistance

In Example Problem 3 determine the total resistance Rtl

(a) from Wave-down and Wave-up and
(b) from the corresponding individual force and velocity values.
use the data points identified in the Example Problem 3, i.e., with time 1 at the first major peak.

4.5 Calculation and consideration of soil damping

Damping is associated with the pile velocity and a case can be made that the major soil
damping occurs at the pile tip. We can obtain the pile toe velocity from consideration of the
arrival and reflection of the impact wave, the Ri waves and from the Rb wave.
vb, = (2 Fd1 – 2( ½Ri ) – Rb) / Z (4.5)
Again the R terms amount to the total resistance and we, therefore obtain under consideration
of Eq. 4.3b:
vb = (2Fd1 – Rtl)/Z (4.6a)
vb = (Fd1 - Fu2)/Z (4.6b)
Knowing the pile toe velocity, the damping component of the total resistance force, Rd, may be
estimated usinv a simple linear damping model as
Rd = J vb (4.7)
The viscous damping factor has units of N/m/s or kips/ft/s. This is a quantity which is rather
difficult to work with. For simplification we non-dimensionalize it by division with the pile
impedance Z; we call the new constant the Case damping factor, Jc.
Jc = J / Z (4.8)
Multiplying the toe velocity (Eq. 4.6b) with the Case damping factor leads to the estimated
damping resistance:
Rd = Jc (Fd1 – Fu2) (4.9)

The total resistance is the sum of the static and damping resistance. The static resistance can
be expected to be the ultimate static resistance, Ru, if the pile has been penetrating into the soil
permanently under the hammer blow We then can calculate the ultimate capacity of the pile
RU = Rtl - Rd (4.10)
and therefore

RU = (Fd1 + Fu2) – Jc(Fd1 – Fu2) (4.11a)

RU = (1 - Jc) Fd1 + (1 + Jc) Fu2 (4.11b)

The Jc damping constant primarily relates to the soil grain size near the pile tip or the major
bearing layer and can be back calculated from Eq. 4.11b if measurements have been taken on
the pile and its ultimate capacity, RU, is known from either a static test run to failure or from
CAPWAP. In that case Jc is the only unknown in Eq. 4.11b.

Example Problem 6: Calculating Ru for t1 at First Peak Velocity

(a) Using the expressions for Wave-down and Wave-up in terms of the measured force and
velocity at times t1 and t2, rewrite Eq. 4.11b in terms of the measured force and velocity.
(b) In Example Problems 3 and 5, for times t1 and t2 identified, calculate the toe velocity
and, assuming a Case Damping factor Jc = 0.2, calculate the damping force and
determine the static capacity by subtracting the damping force from the total resistance.
(c): Discuss the RU result obtained. How sensitive is it to the damping factor Jc (for example,
calculate RU also for Jc =0.3)? Why would the static resistance be so sensitive?

4.6 Selection of time t1 and the RMX method

Static soil resistance is mobilized and increases with pile displacement. To mobilize the ultimate
resistance requires that the pile moves under the hammer blow deeply enough to activate the
resistance both along pile shaft and at the toe. The required maximum displacement can be
quite large for large displacement piles. Fortunately, the hammer impact generally produces a
relatively large temporary displacement, even if the pile has only a small permanent set. In
general, therefore, we expect the ultimate capacity to be mobilized, if the pile set is greater than
2.5 mm or 0.1 inches. That would require that the maximum (temporary) pile set was large
enough to cause soil failure. The next figure shows a force, velocity, wave-up and displacement
record. Note that the displacement reaches a DMX value (at the sensor location) of 25 mm or 1
inch before it rebounds settling at a final permanent set of 2 mm or 0.08 inches.

It is informative to look at the above record more closely. The Wave-up is near zero until at the
second solid time line (2L/c after the first major force and velocity peak) where the Wave-up
sharply increases, corresponding to an increase in force and a decrease in velocity. This

compressive wave up is caused by a high toe resistance while shaft resistance in this case is
nearly zero. Indeed, this pile was driven to rock, encountering high stress both at the bottom
and the top (to be discussed below).

The following figure shows six curves. On top are the measured force, F, and velocity, vZ.
Below F and vZ we see the Wave-up, Fu (WU) and displacement, u (D), curves and below that
the total and static resistance curves, Rtl (RT) and Rs(RS). Rs has been calculated with a
damping factor Jc = 0.6. Also marked on this graph are certain important points: maximum
force, FMX, maximum velocity, VMX, maximum displacement, DMX, Wave-up at time t2, WU2,
static resistance at the first Peak (t1) RP6 and maximum static resistance RX6 (also called RMX
for Jc = 0.6).

The displacement reaches a maximum of about 15 mm (0.6 inches) shortly after time t2 (2L/c
after the first major peak velocity). Not shown in this graph is that the displacement will
eventually decrease to a final value DFN = 2.5 mm or 0.1 inches. The resistance curves in the
bottom set of curves were calculated by evaluating Eq. 4.11b for each point in time beginning at
the first major force and velocity peak. The resulting Rtl and Rs values were then plotted at the
associated time t1. As mentioned above, in the figure below, the static resistance curve was
calculated for a damping factor Jc = 0.6. The difference between the static and the dynamic
curve is the dynamic resistance Rd. The static curve increases from an RP6 value of 1510 kN
(340 kips) reaching a maximum value RX6 = 3050 kN (690 kips). The maximum Rtl is 5200 kN
(1180 kips). The highest damping force exists at the time t1 where the Rtl value is 4810 kN (1080
kips) and the damping force is therefore at that time 4810 – 1510 = 3300 kN (1080 - 340 = 740
kips). The damping force decreases while the static resistance increases which is due to the
fact that the velocity decreases while the displacement still increases. It is obvious from this
example, that the RMX method is more reasonable for this pile of 610 mm or 24 inch width.

Three observations are important and support the conclusion that the RP method should not
normally be used. It is a method easily understood and evaluated in hand calculations (and
therefore used in our example problems), but is not generally used in practical applications. Also
please note the following:

1. Damping factors have to be chosen differently for the RPi and the RXi methods. In the
present case RP3 (RP with Jc = 0.3) and RX6 (RX with Jc = 0.6) would yield
approximately the same results. The literature still shows damping factors for the RPi
method. In most instances these values would be too low for the RXi method.
2. The sensitivity of the results to an improper damping factor choice is much greater for
the RPi than the RXi method. As a demonstration the table below shows the the Case
Method results for damping factors of 0.5, 0.6 and 0.7. Obviously, the RPi capacities are
much more sensitive to damping than RXi values (about 35% vs 7% per each 0.1
change of Jc), because of the higher velocities at time t1.
3. The RPi values tend to be too low for large displacement piles because the resistance
would not be fully mobilized at t1.


Table of various Case Method results for the above example case (note the data was taken
from PDA Example 1).
Method  J = 0.5 J = 0.6 J = 0.7
RPi (kN) 2060 1510 960
RPi (kips) 460 340 220
RXi (kN) 3290 3050 2920
RXi (kips) 740 690 660

RAU (kN) - 2630 -

RAU (kips) - 640 -
RA2 (kN) - 2850 -
RA2 (kips) - 590 -

4.7 Other methods of interest: RAU, RA2

If the toe velocity (Eq. 4.6) becomes zero some time after impact, then according to the Case
Method definition, the damping resistance Rd is also zero. This implies that the any resistance
which is present at this time is static and therefore Independent of a damping constant. This
solution occurs when

vb = 0
or after substituting for the toe velocity we find that Fd1 = Fu2 at the time when the bottom
velocity is zero. Therefore, calling the associated capacity RAU we can write the following
RAU = Fd1 + Fu2 (4.12)
With the condition that Fd1 = Fd2. We call this capacity value RAU, because it is automatically
static and no damping factor has to be chosen. Graphically it can be seen in the above
resistance versus time curves when Rtl and Rs(t) are for the first time equal. Since this equation
assumes resistance to be at the pile toe, it generally will work well if there is little skin friction.
One of its applications is also for early, easy driving cases. However, the RAU method may
give unrealistically low results in harder driving where large distributed skin friction is present;
the result will be conservative, i.e. a lower bound solution.

It would be convenient to obtain an estimate of capacity without having to guess a damping

factor even in cases of friction piles. For this reason, the RA2 method was formulated which is
more generally applicable than RAU. However, both RAU and RA2 methods frequently
underpredict. Since 2011 it is therefore preferred to run the iCAP analyses during data collection
and in that way determine the appropriate damping factor during testing.

In the above example, the RA2 method gives a capacity prediction of 2850 kN (640 kips) and is,
therefore, in reasonably good agreement with the RX6 method while RAU with 2630 kN (590
kips) is somewhat low (see also the below table).

4.8 The Unloading Correction Method, RSU

The Case Method of capacity prediction “measures” the soil resistance acting at the same time
all along the pile. If the energy is sufficient to move the whole pile at the same time downwards
when the resistance reaches ultimate, this method leads to satisfactory results. For piles which
have a deep penetration relative to the impact induced wave length, the Case Method may
underpredict if a substantial amount of the total soil resistance is distributed along the shaft and
if, during hard driving, the pile top already rebounds before the resistance is fully activated along
the bottom part of the pile. When the pile top velocity becomes negative before the stress wave
returns at time 2L/c, the pile top is moving upward and some of the skin friction near the top
begins to unload.

For the RPi Method an approximate correction can be calculated in the manner demonstrated in
the figure below. Note that this correction is only reasonable if the pile top velocity becomes
negative prior to t2 = t1 + 2L/c. Also, t1 must be chosen at the first major velocity peak.
 Determine the difference time, tU, between the time that the pile top velocity becomes
zero and the wave return time t2 (The time, tU, multiplied by the wave speed, c, and
divided by 2 represents the length of pile, Lu, over which unloading has likely occurred.)
 Measure the resistance, ΔRUN, that may have unloaded by taking the Wave-up value at
time t1 + tU. (note that this is only one half of the resistance at t1 + tU; the assumptoin is
here that not all resistance has fully unloaded.

 Add RU to RTL which leads to the corrected RTLU.
 Determine the toe with RTLU taking the place of RTL in Eq. 4.6.
 Apply the proper damping factor (to be verified by CAPWAP).

In the figure below (PDA Example data 21c), the bottom graph shows again the Rtl and Rs
curves. Both decrease at a rather steep slope immediately after time t1. This is typical for
unloading cases where the energy provided by the hammer is just not sufficient to maintain a
downward pile motion for a sufficiently long time for complete, simultaneous resistance
activation. This immediate decrease of the resistance curves also means that RPi and RXi are

In this example, RTL is 4480 kN (1010 kips) and RTLU is 5240 kN (1180 kips) which means that
the unloading correction, ΔRUN, was 760 kN (170 kips). Assuming a damping factor Jc = 0.3
(relatively low damping factors are used for the RPi Method) we obtain RP3 = 3550 kN (800
kips) and RU3 = 4540 kN (1020 kips). Note that, compared to RTL, the increased RTLC causes
the toe velocity and therefore the damping resistance to decrease.

4.9 Total and static shaft resistance (skin friction)

We have seen in Sections 4.2 and 4.3 that the upwards traveling shaft resistance waves create
a difference between F and vZ or an increasing Wave-up before time 2L/c. The question is
now, how we can get a closed form estimate of both the total shaft resistance and the total static
shaft resistance acting on the pile. Consider the figure below. It shows F and vZ on top and Fd
and Fu below. Indicated are also times t1 (first major velocity peak and t2 (t1 + 2L/c). A black
heavy horizontal bar between the top and bottom graph, beginning at t1 and ending at t2 is a
schematic of the pile with its top at t1 and its toe at t2.

The A-time line indicates where the Wave-up curve is still zero. Assuming that any shaft
resistance acting at the top (actually at the sensor location) would be activated at time t1, we can
say that from the top to the point A along the pile, no (or not much) shaft resistance acts. At
Point B a small amount of shaft resistance has its effect and from this point on the Wave-up
curve increases somewhat linearly to point C. The difference FuC – FuB is ½ of the total (static
plus damping) shaft resistance acting between point B and C.

The problem is now, that we do not know the shaft resistance acting between C and the peak
pile toe reflection, because of the reflected impact wave which creates a valley in the Wave-up
curve. We solve this problem by going back from point C a distance tcL and extrapolating linearly
to point L. The point thus determined defines ½ SFT, providing an estimate of one half of the
total (damping plus static) shaft resistance.

The next question is how we can figure what the static shaft resistance is. We solve this
problem in an approximate manner by reducing SFT proportionally to the RMX resistance.
Thus, the reduced shaft resistance is calculated as
SFR = SFJ = SFT (RXJ / RX0) (4.13)
and the associated end bearing is
EBJ = RXJ-SFJ (4.14)
(e.g., for Jc=0.5: SF5 = SFT(RX5/RX0) and EB5 = RX5 – SF5).

Note that in the above derivation of end bearing it is assumed that the shaft resistance will be
activated at time t1, however the end bearing (and therefore the maximum capacity value) will
take more displacement and thus a longer time for complete activation. For that reason, the sum
EBR + SFR does in general not equal the RPJ result. Note also that this method can only yield
a reasonable static shaft resistance estimate for uniform piles without a major unloading
problem which would be apparent by the Wave-up curve decreasing before 2L/c.

Example Problem 7; Estimates of Shaft Resistance and End Bearing

With the measurements shown below taken on a uniform 450 mm (18”) square prestressed
concrete pile, calculate (for capacity determination use time t1 and Jc = 0.5):

The cross sectional area A (AR) The wave speed c (WS)

The elastic modulus E (EM) The pile impedance Z
Velocity at time t1 v1 (VT1)
Wave-down at time t1 Fd1 (WD1) Wave-up at time t2 Fu2 (WU2)
The total resistance at time t1 RTL The static resistance at t1 RP5
The total shaft resistance at t1 SFT The static shaft resistance at t1 SFR
The estimated end bearing at t1

 The concrete specific weight is γ (SP) = 23.6 kN/m3 (0.15 k/ft3)
 Since the maximum resistance does not occur at t1, the EBR value which relates to the
RMX method will be different from your estimate for t1

Would the RAU method be appropriate?
Would this be a case benefitting from the unloading correction?

4.10 Energy Approach capacities QUS, QUT

This method of capacity calculation from F and v measurements is not an original Case Method
approach, but has been described by others, for example Paikowsky. The Case Method looks at
individual force and velocity values and determines resistance from a force equilibrium point of
view. The Energy Approach is based on the conservation of energy as it was done for years in
dynamic formulas; in contrast, however, this Energy Approach uses measured values for energy
and energy losses (pile rebound) instead of estimates.

Consider the following figure. It shows a simplified plot of elasto-plastic resistance, R, vs. pile
displacement, u. Beginning at point “0” this simple plot suggests that the resistance increases
linearly with displacement until point “1” where the displacement reaches the quake value and
the ultimate resistance, Ru. Beyond that point the resistance does not increase while the
displacement increases further to point “2” where the maximum displacement is reached.
Beyond that point 3, the pile rebounds with the resistance decreasing linearly at a slope as
defined by the quake.

At point “3” where the maximum displacement is reached, the soil resistance has done a
maximum amount of work, after that energy is given back to the hammer. This amount of
energy is equivalent to the area under the force-displacement curve or
Emax = Ru (umax - ½ q) (4.15a)
And substituting for q
Emax = Ru (umax – ½ [umax – ufin]) (4.15b)

Ru = 2Emax / (umax + ufin) (4.16)

As we will see in Chapter 7, the maximum energy transferred to the pile, called EMX, can be
calculated from force and velocity records. Furthermore, the measured maximum pile top
displacement is DMX. So, if we knew the ufin value we could readily evaluate Eq. 4.16. PDA-W
solves this problem by accepting a set per blow value, SET, as an input. Alternatively it can use
the final displacement, DFN, from double integration of the measured acceleration. The
corresponding results calculated are then
The program also calculates an RQJ value which is either QUT or RXJ, whichever is greater.

The problem with this approach is that either QUS or QUT are really dynamic resistance values
(not static) and that they therefore tend to be non-conservative. Also using pile top quantities is
not strictly correct if we consider an energy balance for the soil. And finally, the soil does not
offer a total concentrated resistance force which is elasto-plastic. Considering the resistance ats
distributed along the pile and considering it consisting of a static and a damping component is
definitely a more realistic approach. A thorough study has not been made, but it appears that for

end-of drive situations, the result is about 40% higher than CAPWAP and for restrikes it is, on
the average, about twice as high as CAPWAP.

5. Stress Calculations

Pile damage can be the result of poor hammer alignment (high local contact or bending
stresses), obstructions in the ground which cause the pile to be bent or subjected to a non-
uniform toe resistance (high local contact or bending stresses). The most common cause for
pile damage is, however, an overstressing due to high hammer impact forces which can
generate excessive compression stresses at the pile top or bottom (high end bearing) or also
high tension stresses somewhere along the length of the piles. Concrete piles are particularly
vulnerable to excessive tension stresses.

5.1 Pile top (sensor location) stresses

The PDA measurements of strain on two or even four sides of a pile, multiplied by the elastic
modulus, yield stresses at the measurement location. The single highest stress at a transducer
location is called CSI by the PDA. If the transducer happens to be in the plane of highest
bending then CSI is a good indicator of bending stresses at the sensor cross section. It is
calculated simply as:
CSI = EM max[εi] (5.1)
where max[εi] is the highest strain measured by anyone of the two or four strain transducers.

Of course, other cross sections may have different bending stresses. Static bending, e.g., due
to an inappropriate guiding of the pile, cannot be detected by the PDA. Thus, while CSI may be
helpful to judge the hammer-pile alignment, particularly when 4 strain sensors are used, the
PDA cannot provide a thorough bending assessment neither at the top nor anywhere else along
the pile. Important is also the average stress at the sensor location, CSX, because it is what is
normally compared with the allowable driving stresses. CSX can be calculated from the average
of the strain readings as follows:
CSX = ½(ε1 + ε2) EM (5.2a)
for two strain transducer applications or
CSX = ¼(ε1 + ε2 + ε3 + ε4) EM (5.2b)
for four strain transducers.

Also, please note that the stresses above the sensor location cannot be easily calculated from

5.2 Pile toe stresses

Suppose a pile is driven to a very hard layer. As we have seen, in that case it is theoretically
possible that the pile will experience a pile bottom force which is twice the impact force (or twice
the force in the Wave-down). In general, since not even a hard rock is absolutely rigid, such
high end bearing force cannot materialize, however, it is definitely possible that the pile toe
stresses exceed those at the top. The total toe resistance force is calculated by the PDA
considering the maximum total (static plus damping) resistance minus the effect of the shaft
resistance. This force is called CFB; approximately it is equal to

CFB = RX0 – cb SFT (5.3a)
where cb is an adjustment factor which since 2011 is chosen by the PDA as 0.5 for
conservatism. The corresponding stress is
CSB.= CFB/AR (5.3b)
The following figure shows an F and vZ record for a pile with little shaft resistance and high end
bearing (note the strong increase of force at time 2L/c). At the top the maximum stress,
averaged over the cross section (233 MPa or 33.8 ksi) happens not at impact, but when the
wave returns from the pile toe. Evaluating this record for pile toe stresses according to Eq. (5.3a
and b) yields a pile to stress of 264 MPa or 38.2 ksi or about 13% more than the top stress. The
individual strain records also indicated bending of 5% above the average stress at the peak
stress level; at impact, where the stresses were lower, the bending effect was more

5.3 Pile tension stresses caused by Wave-up

For concrete piles tension stresses must be known, but at the sensor location, tension stresses
are usually not very high, because of the reflection taking place at the pile top where tension
has to be zero. We therefore have to calculate the tension stresses at points below the location
of the sensors. We can do this by remembering that the force at a point is the sum of the forces
in Wave-down plus the force in the Wave-up. In easy driving we normally see a tension wave
traveling upwards arriving at the pile top around the time 2L/c. In hard driving we sometimes see
a downward traveling tension wave.

For the easy driving case, consider the following figure. It shows both Wave-down (blue-dash)
and Wave-up (black, solid). Wave-up becomes strongly negative at time 2L/c. Thus a tension

wave travels up from the pile bottom due to the reflection of the impact wave. Let us call t=0 as
the time at which the maximum impact force is apparent. As shown in the L-t plot underneath
the record, we can calculate the force at any level x as the sum of the downward wave
emanating from the top at time t=2x/c after the time of impact plus the upward wave arriving at
the top at time t=2L/c. If we chose t=0 such that the upwards traveling wave, Fu2, represents the
highest tension and if we choose x such that Fd3 is the lowest downward traveling compression,
then Fx will be the highest tension force in the pile.

The following figure shows how we can expand on this concept to determine the tension stress
envelope caused by the recorded event. (The PDA calculates the tension envelope in the same
 Determine the point of minimum Wave-up and determine minFu.
 At the time of minimum Wave-up we draw a heavy bar backward in time for 2L/c and call
the beginning “the pile top” and the end point “the pile toe”. The reason is that a
downward compression wave observed just before the time of minimum Wave-up will
have a tension reducing effect very near the pile top.
 In the Wave-down plot, draw a horizontal line at a distance of minFu above the zero line
from “pile top” to “pile toe”
 Where the Wave-down is less than minFu, the difference between the horizontal line and
the Wave-down curve is the net tension force along the pile.
 Determine the point of minimum Wave-down, minFd, which happens to be the point
where the maximum net tension occurs. You can calculate the distance below the pile
top where the tension is maximum from the relative distance of the point of minFd from
the point of minFu.

Again, the maximum net computed tension (CTN) occurs when the downward compression
stress is a minimum (time t3) and can be found mathematically by
CTN = minFu + minFd (5.4a)
The associated stress is
CSN = CTN/AR (5.4b)
Obviously, this method is only correct for uniform, undamaged piles.

Example Problem 8: Tension Stress Calculation from Wave-up

In the following Wave-down and Wave-up record, determine the following values:
(a) Minimum (maximum tension) Wave-up as a percentage of maximum Wave-down;
(b) Minimum Wave-down as a percentage of maximum Wave-down;
(c) Relative distance from the pile top where maximum net tension occurs.

5.4 Pile tension stresses caused by Wave-down

At the end of Section 3.3 we have considered a situation of a rigid support at the bottom of a
pile and we have seen that this situation causes a compressive wave up reflection at the pile
toe. This upward traveling wave will have an upward directed particle velocity. When this wave
arrives at the pile top it can encounter either a downward moving ram which causes a
compressive downward wave or, particularly if the ram is very light and has lost its momentum,
something close to a free pile top which then causes a downward tension reflection. In general,
when driving is very hard (which means the hammer does not have enough momentum to keep
the pile in compression), damaging tension waves can happen in the hard driving case. The F,
vZ records in such cases have a large negative velocity at a time when the force is relatively
low. Obviously, with Fd = ½(F + vZ), the negative velocity will make the Wave-down a tension
wave. The figure below shows the example of a large negative velocity not offset by a positive
force and, therefore, a negative Wave-up curve. The PDA determines the minimum Wave-down
value and then searches for a trailing minimum compressive Wave-up; adding it to the
maximum tensile Wave-down yields the maximum net tension force, CTX. The associated
stress TSX is calculated by division with the cross sectional area. Again this only works for
uniform piles. Note, however, that the PDA outputs TSX as always the greater of the tension
from upward or downward waves.

6. Damage Detection

For a uniform pile, an upward traveling tension wave should be observed only after reflection
from the pile tip and should therefore come at time 2L/c. If an upward tension wave is observed
prior to 2L/c, it must be due to a reduction in pile impedance, i.e., either a reduced cross
sectional area or a reduced elastic modulus or a reduced mass density all of which should be
considered either a damage or defect in a uniform pile.

Consider the figure below showing on the left hand side a schematic pile which has impedance
Z1 on top and which has a reduced impedance, Z2, below x. As shown on the right hand side of
the figure, because of the impedance reduction the impact wave Fd1 will be partially reflected at
x sending a reflection wave Fu1 upwards which arrives at the top at time 2x/c. It will be apparent
at the pile top and can then be evaluated. Additionally, a different wave Fd2 will continue to travel
to the pile toe. Consider the wave forces acting at section A and B, i.e. just above and below the
impedance reduction; because they have to be in equilibrium we have:
Fd,1 + Fu,1 = Fd,2 (6.1)
Similarly, the velocities in these waves have to be equal on both sides or else we would not
have continuity (the pile would separate):
vd1 + vu1 = vd2 (6.2a)
Considering proportionality we can replace the velocities in Eq. 6.2 by the corresponding wave
Fd1/Z1 – Fu1/Z1 = Fd2/Z2 (6.2b)
We are now defining the integrity factor
ß = Z 2 / Z1 (6.3)
which is 1 for Z2 = Z1 (undamaged) and zero for a completed damaged pile. Now after division
with Z2, Eq. 6.2b can be expressed as
Fd2 = ß (Fd1 - Fu1) (6.2c)
and combining E. 6.1 and 6.2c to eliminate Fd2 we find that
ß = (Fd1 + Fu1 ) / (Fd1 – Fu1) (6.4)

Let us assume that Fu1 is an upward traveling wave which is tensile and one half of the
magnitude of the impact wave. In that case ß = (1-0.3)/(1+0.3) = 0.7/1.3 = 0.54. Thus in that
case Z2 would be slightly more than ½ of the pile top impedance.

Reality is actually more complicated because the wave Fd1, by the time it has reached the point
A has already lost some intensity due to the resistance acting over the distance x. For that
reason then PDA-W program uses a modified equation which considers the effect of soil

ß = (Fd1 – 1.5Rx + Fu1 ) / (Fd1 – 0.5Rx – Fu1) (6.5)

Although not strictly correct (the 1.5 factor on the Rx term in the numerator is higher than
supported by wave theory), Eq. 6.5 gives reasonable and conservative results in most cases.
Several reasons can be given, among them:
 Theoretically an extrapolation of the shaft resistance to the point of highest tensile wave
(Fu1) should be made (particularly, where damage is gradually worsening with depth).
 Soil damping reduces the downward wave more than indicated by the Rx value, because
of its temporary (velocity dependent) nature.
 Structural/pile material damping reduces impact and reflection waves, particularly in
concrete piles with microcracks.
 Soil resistance at the bottom of the damage (in extreme cases like an end bearing)
reduces the reflection.
 Compressive wave reflections from an impedance increase at the end of the damage (if
the damaged portion is shorter than the wave length).
 Resistance from connecting steel strands or other damaged pile sections at the bottom
of the damage.
While these and other effects add resistance or resistance like effects and therefore reduce the
apparent damage reflection, the PDA generally calculates a ß-value where high shaft
resistance exists above the damage that is low or conservative.

In order to provide a uniform damage assessment, the following table shows a classification
scale which has been proposed (Rausche et al., 1979), a paper which is also the source of Eq.

Damage Assessment based on ß-values

ß Suggested Pile Condition
1.0 Uniform
0.8 – 1.0 Slightly Damaged
0.6 – 0.8 Damaged
< 0.6 Broken

Of course, such damage assessment cannot be directly applied to a crack, broken weld, bent
steel pipe or many other damage situations which do not conform to the basic assumption in the
derivation: ß is a measure of the remaining impedance or cross-section. However, if the
impedance ratio becomes 0.6 or less, it is unlikely that the pile can be fully functional. Also, a
distinct pile toe reflections is then rarely observed. In all cases it is suggested to try modeling
the supposedly damaged pile with CAPWAP to confirm the findings by the simpler ß-method.

Finally a word should be said about the determination of the location of the damage. The PDA
calculates the time 2x/c from the onset of wave down at impact to the beginning of the apparent
tensile reflection, Fu1. This time multiplied by ½ of the wave speed is the best estimate of the
beginning of the damage. Obviously, an incorrect wave speed will produce an incorrect length x.
In fact, if the wave speed had been assumed too low, the PDA might indicate damage just
above the pile toe. If and only if damage can be definitely ruled out, then the wave speed can be
be determined by increasing it until the PDA’s damage indication disappears.

The following example record was taken on a 356 mm (14 inch) square prestressed concrete
pile. Before damage occurred, given the length of LE=19.5 m (63 ft) below gages and a clear
toe reflection, the wave speed was determined to be WS=4,040 m/s (13,250 ft/s). The record’s
force scale was adjusted so that Fd1 = 100% (actually 2600 kN, but we can do the calculation
non-dimensionally). Clearly the Wave-up record shows a tensile reflection beginning at a time
which is 75% of 2L/c. Thus damage is apparent at a depth of 14.5 m (47.6 ft). The Wave-up
value just before the tensile reflection is 4.4% of full scale (Rx is therefore 8.8%). The tensile
reflection Fu1 is -30%. Introducing these values in Eq. 6.5 leads to
ß = [100 – 1.5 (8.8) + (-30)] / [100 - 4.4 –(-30)] = 45%
indicating a broken pile. Note that ignoring the relatively minor resistance effect would have
yielded a ß of 54% (Eq. 6.4).

Example Problem 9: Damage Assessment
For the following record of a 35.7 m (117 ft) long (below sensors) steel pile, calculate the depth
of damage and its severity. What could be the reason for such a clear damage reflection in a
steel pile?

7. Hammer Performance

The energy transferred to the pile, Et(t), can be found from the work done on the pile
Et(t) = ∫ F(t) du = ∫ F(t) v(t) dt (7.1)
which we can obtain if we integrate the product of force F and velocity v over time. The
maximum value is the maximum transferred energy EMX.
EMX = max.[Et(t)] (7.2)
It is important to realize that only this transferred energy, EMX, is capable of actually doing
work, rather than the hammer’s rated energy, Er (called ER by the PDA). The transferred energy
only allows the hammer’s performance to be judged, but only in a statistical manner, by its
energy transfer ratio (or transfer efficiency, ηt) which is defined as

ηt = ETR = EMX/ER (7.3)

Additional definitions of interest are the impact velocity which for a given stroke h is

vi = (2 g h)1/2 (7.4)

with g being the gravitational acceleration. Given the ram mass, mR, he kinetic energy is

Ek = ½ mR vi2 (7.5)

If we measure the impact velocity then we can calculate the actual impact velocity and from it
the actual hammer efficiency, ηH, as the ratio of the measured divided by the rated energy.
Note: the hammer efficiency expresses losses in the hammer, occurring prior to impact. The
transfer ratio (or efficiency) expresses energy losses occurring in hammer, driving system
(cushions and helmet) and at the pile top surface.

While Radar (PDI’s Hammer Performance Analyzer, HPA) or other device can measure the
impact velocity, it is also possible under certain circumstances (and it may be simpler) to
calculate it from the F and v records. The evaluation of the records would require applying the
principles of impulse and conservation of linear momentum. The impulse can be calculated from
either the measured force as
MFO = ∫ F(t) dt (7.6)
The integral should be evaluated from time 0 (when the force at the pile top begins to increase)
until the time when the ram velocity again becomes zero. Since we don’t know that time we
have to assume that the ram velocity becomes zero together with the pile top velocity. This
assumption precludes the application of the impulse-momentum relationship for concrete piles
with soft cushioning. Also, since diesel hammers have energy added during the impulse
evaluation period, this hammer type does not lend it to this method either.

A force impulse can also be calculated from the Wave-down curve:

MWO = ∫ Fd(t) dt (7.7)

This integration should go from time 0 until the time when the Wave-down becomes zero.
Equating the impulse to the momentum of the ram, which is equal to ram mass times impact
veloity (mR vi) we can calculate the ram impact velocity as either
vi = MFO / mR (7.8a)
vi = MWO / mR (7.8b)
This ram impact velocity can be used to obtain the kinetic energy
Ek = ½ mR vi2 (7.9)
which can be compared with the rated energy, ER, to obtain the hammer efficiency
ηH = Ek / ER (7.10)
Also, comparing the kinetic energy with the maximum transferred energy EMX will demonstrate
the effectiveness of the rest of the driving systems.

In summary, two energy ratios are important and must be distinguished:

The hammer efficiency: ηH = EK/ER
The transfer ratio (efficiency): ηt = ETR = Et / ER = EMX/ER
For open end diesel hammers, it is also important to check the hammer stroke. Given the time
between hammer blows, T, and assuming that the ram travels freely (no friction or other losses
of energy) the time for the ram fall (or ram rise) is equal to T/2. If the velocity increases linearly
due to the gravitational acceleration, g, then
vi = g T/2 (7.11a)
and combining with Eq. 7.4 to cancel vi, we obtain
(2 g h)1/2 = g (T/2)
h = (g / 8) T2 (7.11b)
Since a diesel hammer loses some of its ram velocity due to the precompression of the gases,
based on field tests and wave equation simulations we found that h would be more correctly
calculated after subtracting a loss term, hL = 0.1 m or 0.3 ft. Thus the Saximeter formula is
h = (g / 8) T2 - hL (7.11c)
For example, if the time between two hammer blows is 1.5 s, then h is 2.66 m (8.75 ft).

Example Problem 10: Hammer Performance

1. Calculate the stroke of a diesel hammer if the hammer runs at 38 blows/minute.
2. Consider a hammer with a 10 kip ram with a rated stroke of 3.25 ft and an observed impulse
of 4.00 k-sec (MFO) and EMX of 18.0 k-ft.
What is the hammer’s:

a. Rated energy,
b. Rated impact velocity,
c. Actual impact velocity,
d. Kinetic energy,
e. Hammer efficiency,
f. Transfer ratio (efficiency)?


Example Problem 1: Wave Speed, c

   E   γ  Wave Speed 
   MPa  kN/m3  m/s 
Concrete  35,000  24  3,782 
Timber  12,000  8  3,836 
Steel  210,000  77  5,172 
   E  γ  Wave Speed 
   ksi  lb/ft3  ft/s 
Concrete  5,000  150  12,446 
Timber  1,800  50  12,934 
Steel  30,000  492  16,833 

Example Problem 2: Impedance, Z

   E   Gamma  Speed  A  Z=EA/c  Mp 
   MPa  kN/m^3  m/s  cm^2  kN/(m/s)  Mg 
Concrete  35,000   24 3782  756.25 700 5.55 
Timber  12,000   8 3836  756.25 237 1.85 
Steel  210,000   77 5172  756.25 3070 17.81 
    E    Gamma  Speed  A  Z=EA/c  Mp 
    ksi   lb/ft^3  ft/s  inch^2  kips/(ft/s) kips/ft/s^2 
Concrete  5,000   150 12446  121 48.6 0.392 
Timber  1,800   50 12934  121 16.8 0.131 
Steel  30,000   492 16833  121 215.6 1.285 

Example Problem 3: Wave-down and Wave-up Values

EM‐MPa  42,000   EM‐ksi  6,000

SP‐kN/m3  24   SP‐lb/ft3  150
AR‐cm2  3,600   AR‐inch2  576
WS‐m/s  4,143   WS‐ft/s  13612

Z ‐kN/(m/s)  3,649   Z ‐kips/(ft/s)  254
Z*v(t1) ‐ kN  3,649   Z*v(t1) ‐ kips  838
Z*v(t2) ‐ kN  3,649   Z*v(t2)‐ kips  838
F(t1)‐kN  4,000   F(t1)‐kips  900
F(t2)‐kN  ‐200   F(t2)‐kips  ‐50
Fd1=0.5[F(t1)+Zv(t1)] ‐kN  3,825   Fd1=0.5[F(t1)+Zv(t1)] ‐kips  869
Fd2=0.5[F(t2)+Zv(t2)] ‐kN  1,725   Fd2=0.5[F(t1)+Zv(t1)] ‐kips  394
Fu1=0.5[F(t1)‐Zv(t1)] ‐kN  175   Fu1=0.5[F(t1)‐Zv(t1)] ‐kips  31
Fu2=0.5[F(t2)‐Zv(t2)] ‐kN  ‐1,925   Fu2=0.5[F(t2)‐Zv(t2)] ‐kips  ‐444

Example Problem 4: Shaft Resistance from Force and Velocity times Impedance

The resistance force between point A and B amounts to approximately 47% of the impact force
and that includes both static and dynamic resistance components. There could be additional
resistance on the shaft below point B, but the magnitude is not obvious from the record. There is
little or no shaft resistance acting above point A.

Example Problem 5: Calculating total Resistance

In Example Problem 3 determine the total resistance Rtl
(a) from Wave-down and Wave-up and
(b) from the corresponding individual force and velocity values.
use the data points identified in the Example Problem 3, i.e., with time 1 at the first major peak.

Z*v(t1) ‐ kN        3,649    Z*v(t1) ‐ kips         838 

Z*v(t2) ‐ kN        3,649   Z*v(t2)‐ kips          838 
F(t1)‐kN  4000   F(t1)‐kips  900
F(t2)‐kN  ‐200   F(t2)‐kips  ‐50
Fd1‐kN  3825   Fd1‐kips  869
Fd2‐kN  1725   Fd2‐kips  394
Fu1‐kN  175   Fu1‐kips  31
Fu2‐kN  ‐1925   Fu2‐kips  ‐444
v(t2))Z ‐kN  1900    v(t2))Z ‐kips  425
(b)RTL= Fd1 + Fu2 ‐kN  1,900     (b)RTL= Fd1 + Fu2 ‐kips  425 

Example Problem 6: Calculating Ru for t1 at First Peak Velocity

(a) Using the expressions for Wave-down and Wave-up in terms of the measured force and
velocity at times t1 and t2, rewrite Eq. 4.11b in terms of the measured force and velocity.


RU = (1 – Jc) Jd1 + (1 + Jc) Fu2
RU = (1 – Jc){½[F(t1) + v(t1)Z]} + (1+Jc){½[F(t2) – v(t2)Z]}
RU = ½ {F(t1) + v(t1)Z +F(t2) – v(t2)Z – Jc[F(t1) + v(t1)Z - F(t2) + v(t2)Z]}

(b) In Example Problems 3 and 5, for times t1 and t2 identified, calculate the toe velocity
and, assuming a Case Damping factor Jc = 0.2, calculate the damping force. Then
determine the static capacity by subtracting the damping force from the total resistance.

Z ‐kN/(m/s)  3649   Z ‐kips/(ft/s)  254 
Z*v(t1) ‐ kN  3,649    Z*v(t1) ‐ kips  838  
Z*v(t2) ‐ kN  3,649    Z*v(t2)‐ kips  838  
F(t1)‐kN  4000   F(t1)‐kips  900 
F(t2)‐kN  ‐200   F(t2)‐kips  ‐50 
Fd1‐kN  3825   Fd1‐kips  869 
Fu2‐kN  ‐1925   Fu2‐kips  ‐444 
(a)RTL‐kN  1900    (a)RTL‐kips  425 
vb = (2Fd1‐RTL)/Z ‐ vb = (2Fd1‐RTL)/Z  ‐
m/s  1.58   ft/s  5.17 
vb Z ‐ kN  5749   vbZ ‐ kips  1313 
Rd = 0.2 vb Z ‐kN  1150   Rd = 0.2 vbZ ‐kips  263 
Ru = RTL ‐ Rd ‐kN  750   Ru = RTL ‐ Rd ‐kips  162 

(c) Discuss the RU result obtained. How sensitive is it to the damping factor Jc (for example,
calculate RU also for Jc =0.3)? Why would the static resistance be so sensitive?

Answer: Increasing the damping factor from 0.2 to 0.3 would increase the damping
resistance from 1150 to 1725 kN and therefore reduce RU to 175 kN. A further increase
of Jc would make the RU negative (note that the Case Method will not allow negative
resistance forces and just set the result to zero.)
This high RU sensitivity to damping can be attributed to the high velocity return at 2L/c,
being as high as the impact velocity and, therefore a relatively low RTL (less than ½ of

Example Problem 7; Estimates of Shaft Resistance and End Bearing

With the measurements shown below taken on a uniform 450 mm (18”) square prestressed
concrete pile, calculate (for capacity determination use time t1 and Jc = 0.5):

The cross sectional area A (AR) The wave speed c (WS)

The elastic modulus E (EM) The pile impedance Z

Velocity at time t1 v1 (VT1)
Wave-down at time t1 Fd1 (WD1) Wave-up at time t2 Fu2 (WU2)
The total resistance at time t1 RTL The static resistance at t1 RP5
The total shaft resistance at t1 SFT The static shaft resistance at t1 SFR
The estimated end bearing at t1

Would the RAU method be appropriate? Answer: No, too much shaft resistance.

Would this be a case benefitting from the unloading correction? Answer: Possibly,
because the velocity becomes negative prior to 2L/c

Example Problem 8: Tension Stress Calculation

In the following Wave-down and Wave-up record, determine the following values:
(a) Minimum (maximum tension) Wave-up as a percentage of maximum Wave-down;
minFu = -45%
(b) Minimum Wave-down as a percentage of maximum Wave-down;
minFd = 0%
(c) Relative distance from the pile top where maximum net tension occurs.
x = 23% of LE
Based on (a) and (b), the maximum tension in the pile CTN = 45 – 0 = 45% of the maximum
wave down.

Example problem 9: Damage Assessment

For the following record of a 38m (125 ft) long (35.7m or 117 ft below sensors) steel pile,
calculate the depth of damage and its severity. What could be the reason for such a clear
damage reflection in a steel pile?

The resistance wave is FuR = 2.5% and Rx is therefore 5%

The damage tensile wave is Fu1 = -51%
Thus ß = [100 – 1.5 (5) + (-51)] / [100 – 2.5 – (-51)] = 41.5 / 148.5 = 28%
The damage is located at 48% of LE (i.e., 17.2m or 56.2 ft) below the sensors or 19.5 (64.2 ft)
below pile top.

For steel piles, this could be an indication of a broken weld or a sharp bend in the pile. Of
course, if the damage happened on a hard layer at a depth corresponding to the damage
length, this could also be a collapsed (accordion type damage) pile bottom. In any case, the ß
value would not be a true indication of Z2.

Example Problem 10: Hammer Performance

1. Calculate the stroke of a diesel hammer if the hammer runs at 38 blows/minute.

Answer: h = (9.81 / 8)(60 / 38)2 – 0.1 = 2.96 m
h = (32.17 / 8)(60/38)2 – 0.3 = 9.7 ft
2. Consider a hammer with a weight of 44.5 kN (10 kips) ram with a rated stroke of 1.00 m
(3.30) ft and an observed impulse of 17.8 kN-s (4.00 kips-s) (MFO) and EMX of 24.4 kJ
(18.0 ft-kips). What is the hammer’s:
a. Rated energy (ER) ER = 44.5 * 1 = 44.5 kJ
ER = 10.0 * 3.30 = 33.0 ft-kips
b. Rated impact velocity vir = (2*9.81 1.0)1/2 = 4.42 m/s
vir = (2*32.17*3.30)1/2 = 14.6 ft/s
c. Actual impact velocity (VRI) vi = 17.8 / (44.5/9.81) = 3.92 m/s
vi = 4.00 / (10.0/32.17) = 12.8 ft/s
d. Kinetic energy Ek = ½ (44.5/9.81) 3.922 = 34.9 kJ
Ek = ½ (10/32.17) 12.82 = 25.5 ft-kips
e. Hammer efficiency ηh = 34.9 / 44.5 = 25.5 / 33.0 = 78%
f. Transfer ratio (efficiency) (ETR) ηt = = 24.4/44.5 = 18.0 / 33.0 = 55%


PDIPLOT Manual of Operation
Table of Contents

1. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

2. Using Help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

3. Program Start-up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

4. Start-up screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

5. International Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

6. Size of the graph screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

7. Opening a PDA File . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

8. The Graph Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

Series Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Format Graph . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Graph Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

9. The Table Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

Keyboard commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Blow Count . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Filter Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Selecting a range of rows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Seeing a blow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Refreshing the Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Entering Drive Log . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Column menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Go to Maximum and Go to Minimum . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Go to Value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Filter Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Penetration editing functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Modify LP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Show Elevation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Edit USR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Recalculate ETR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Modify BLC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Deleting and renumbering blows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Entering Test Date . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Printing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

10. The Print Preview Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

10.1 Average Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

October 2010 PDIPLOT Manual 1-22

Depth increment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Blow increment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Line increment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
10.2 Range Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
10.3 Statistics Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
10.4 Features common to all modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Copy to clipboard . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

11. Automatic checking of Penetration, Blow Number and Data Consistency . . . . . . 18

12. Appending (merging) files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

13. Time Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

14. Saving Graphs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

15. Reading Saved Graphs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

16. Program Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

Units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Labels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

17. Page Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

18. Printing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

19. Opening several documents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

20. Optimizing program performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

October 2010 PDIPLOT Manual 2-22

Pile Dynamics, Inc.

PDIPLOT version: 2010.2.72 - October, 2010
30725 Aurora Road
TEL: USA-216-831-6131
FAX: USA-216-831-0916

1. Overview
The PDIPLOT program will directly read data from PDA W01 or X01 files and present them in
graphical and/or tabular form. The program allows the presentation in a variety of different ways.
The use of this program helps satisfy the reporting requirements of ASTM D4945, and simplifies
and clarifies presentation of results to the client.

PDIPLOT has two modes of operation: Full and Courtesy Mode. Full Mode requires that the PDA-W
program is installed and running in your computer. In this mode, PDIPLOT will automatically open
the PDA program and communicate with it, sending and receiving commands and data. In Courtesy
Mode the program will only be able to read and edit files generated by a copy of PDIPLOT running
in Full Mode (files with extension .PIL).

• Although PDIPLOT will work with any version of the PDA-W program, the use of version
2005.096 or later is highly recommended for greater flexibility and better performance.
• Data collected using the SPT Analyzer will only be read by PDIPLOT versions later than
2008.2.66 together with PDA-W versions later than 2008.098.038 (with SPT Analyzer Data
all capacity-related quantities will always show zero). Trying to open a W01 file generated
by the SPT Analyzer with earlier versions of PDIPLOT/PDA-W will cause an error message.
2. Using Help
The Help feature is invoked by clicking on "Help ÿ Program Help" when available, or by pressing
F1 anywhere on the program. The help file is shown using your default Internet browser. It is
suggested that you do not close the browser after using help, but minimize it instead. This will make
future calls to help faster. NOTE: You can ignore any message saying that the browser will restrict
running scripts or ActiveX controls.

3. Program Start-up
When the program starts, it will first compare the system date with the expiration date. If there is
less than ninety days for the expiration date, a message will be issued and the program will run
normally. The program will not run after the expiration date.


October 2010 PDIPLOT Manual 3-22

1. The full version number and expiration date can be seen by clicking on "Helpÿ
About". Please contact Pile Dynamics, Inc. if you need to update your version.
2. Attempting to read a PDA file containing blows with date stamps later than the
expiration date will cause the program to stop working. Should this accidentally
happen, please contact Pile Dynamics, Inc. for a repair patch.

Next, PDIPLOT will try to read the owner name from file PDIPLOT.OWN, whose default location
is \Program Files\PDIPLOT. If the file is not found, or if PDIPLOT is unable to read it, the user is
asked to choose another file.

1. This file contains the owner name that will be imprinted on all outputs. This file is
encrypted; attempting to edit it will cause PDIPLOT to stop working. To modify the
owner name, please contact Pile Dynamics, Inc.
2. It is possible to change the owner file location or folder after the program is loaded,
by clicking on "Edit ÿ Program Options ÿ Files". The change will take effect next
time the program is loaded, when a new check on the file validity will be executed.

Finally, PDIPLOT will try to open the PDA-Win program. If for any reason the attempt is not
successful, a message box will be shown with three options: "Abort", "Retry" or "Ignore".

1. If the PDA-W program is available but the PDA-W hardware key is not inserted, the
user can click "Retry" after correcting the situation.
2. If the PDA-W program is not available, the user can still run PDIPLOT in Courtesy
Mode, by clicking on "Ignore". In Courtesy Mode, PDIPLOT will not be able to read
data from PDA files (W01 or X01). It will only be able to read and edit “PIL” files
created by another copy of PDIPLOT running in Full Mode (e.g. files with extension
.PIL). This allows you to share the program (and your owner file) with your clients
so they can further process the data you create for them.
3. Clicking on "Abort" will close the program.

NOTE: Once the program is set to Courtesy Mode, it will always open in this mode, unless the
PDA-W program becomes available again. In this case, it will automatically switch to Full Mode.

4. Start-up screen
After the program loads, a blank graph called "Untitled" appears on the screen. The basic layout
can be changed by clicking on "Edit ÿ Layout" at the menu bar, or right clicking anywhere on the

The settings that can be changed through this menu are:

1. Axes Style (Solid, Dash, Dot, etc.)

2. Axes Color
3. Tick Marks Color
4. Numbers Color (only possible after a graph has been loaded)
5. Grids style (Solid, Dash, Dot, etc.)
6. Grids Color
7. Y title color
8. Main title color
9. Heading color

October 2010 PDIPLOT Manual 4-22

1. Some line styles are not correctly printed on some kinds of printers. For example,
dashed and dotted lines will show as solid lines on some color ink jet printers.
2. Setting any color to white will make the corresponding item not to show on the
screen and printer. For this reason, the program will not accept changing the axes
colors to white. As to the other items, this fact can be used to make that item not
3. The graph toolbar can be removed by clicking on "View ÿ Toolbar". This Toolbar
is very useful and saves time.

5. International Issues
The way periods (dots) and commas are printed on numbers, as well as date and time formats are
adjusted according to the Windows "Control Panel" settings. For entering numbers in the program
only periods will be used as decimal separators. However, the program will automatically convert
commas to periods.

6. Size of the graph screen

The "View" menu has two settings: "Screen Size" and "Full Size".

When set to "Screen Size", the whole graph size will be changed to fit the computer screen, when
the program is run with its default size setting (which will cover the whole work area of the
computer). This is automatically adjusted, regardless of screen resolution or size. If the program
window is resized, scroll bars will appear as needed. Please note that in this mode of operation the
relation of the axes sizes to the letters and numbers sizes depends on the particular screen size
and resolution of the computer, so the image will not appear exactly as it will be printed. Line size
proportions are kept as much as possible.

When set to "Full Size", the graph will be shown the same size as it will be printed. The graph is
always sized so that it is centered on the paper free area (total paper minus margin area).
Therefore, depending on the screen resolution, the active desktop area and the graph size, it is
possible for the "Full Size" view to be bigger or smaller than the "Screen Size" view. The graph size
is a function of the printer, paper size and margins as set in "Page Setup" . Note: Make sure that
all desktop toolbars are set to "Always on top" (right click on the toolbar for a pop-up menu),
otherwise part of the graph will appear below the toolbar.

7. Opening a PDA File

After clicking "File ÿ Open PDA File", or clicking on the "Open PDA File" Toolbar button (or
pressing Ctrl+F), then a File Open window will appear. The user can choose the type of file –
possible file types are "*.W01" or "*.X01".

NOTE: PDIPLOT will remember the folder of the last opened file; also, the names of the six
last files opened will be available in the "File" menu – click on any file name to open it again.

PDIPLOT will open the PDA file using the PDA-W program, which will read each blow on the
selected file, calculate the quantities last used by PDA-W for Q1 to Q9 and send the results back
to PDIPLOT. The quantities calculated and transferred are those used the last time the PDA-Win
program was closed. Any Q1 to Q9 quantities that are not defined (as in the case of a recently
installed PDA-W) will default to "VMX", and a warning message will be issued. Also, a warning
message will be issued if penetrations or blow numbers are not continuously increasing, or any
other inconsistency is detected. Please refer to 11. Automatic checking of Penetration, Blow
Number , Blow Count and Data Consistency for further explanations.

October 2010 PDIPLOT Manual 5-22

NOTE: If the quantities determined are not the ones desired for PDIPLOT, you must close
PDIPLOT, open PDA-W and select the proper quantities Q1 to Q9, close PDA-W and then
reopen PDIPLOT.

After opening the file, PDIPLOT will show either a new Graph Window or a new Table Window. You
can select which window appears first upon opening a new file, by clicking on "Edit ÿ Program
Options ÿ Operation" and selecting the "Operation" tab. Those two windows, the Graph and the
Table, are the main program windows, and will be described in more details in the next chapters.

To move from the Graph Window to the Table Window, click on the “Data Table” Toolbar button,
or on the menu item "Edit ÿ Data Table", or type the letter "T" (with or without Alt, Shift or Ctrl).

To move from the Table Window to the Graph Window, click on the “smiling face” Toolbar button,
or on "File ÿ Exit to Graph", or type the letter "G" (with or without Alt, Ctrl or Shift).

8. The Graph Window

This window shows the graph generated with data read from the Table Window. The image shown
on the screen is very similar to what will be sent to the printer. Next is a description of the Graph
Window features and controls:

1. Right clicking the mouse on any axis will open the Series Window, which can also
be accessed by clicking on "Edit ÿ Series", or by clicking the “Series” Toolbar
button. This window has seven tabs, six for the X axes and one for the Y axis.

a. For the Y axis, the user can adjust the following:

i. Variable - Default is Blow Number, can be changed to Penetration
or Elevation if this information is available.
ii. Precision, or number of digits after the decimal point – automatically
set to number of decimal places of value per division, it can later be
changed by user.
iii. Number of divisions (horizontal grid lines), value per division,
minimum scale and maximum scale. The number of divisions can be
changed between two and 15. To change any of the values, click on
the corresponding cell. Pressing "Enter" after the value has been
typed allows the user to see the effect of the change on the other

The values shown on the grey cells are for information only, and cannot be
changed by the user. They are (left to right): Minimum value in the series,
Maximum value in the series and Number of values that are out of range.
Scale Maximum and Minimum are originally set by PDIPLOT so that no
value is out of range, and the value per division is in most cases an integer.
Setting the scale maximum below the maximum value or the scale minimum
above the minimum value will make some points (blows) to be out of range.
The number of blows in this condition is shown in bold on the "O/Range"
cell. All blows with numbers or penetrations outside the scale range will not
be shown on the graph. The corresponding blow number or penetrations will
be shown in red on the Table Window.

b. For any of the X Axes, the user can make the following adjustments:
i. Line style and line color. There is also an option for "extra thick on
printer", if a noncontinuous line is chosen. When this option is
checked, the corresponding line will be printed with double
thickness. Screen display will not be changed. It is also possible to

October 2010 PDIPLOT Manual 6-22

set the current line style and line color as defaults for the current
ii. Variable (quantity Q1 to Q9) assigned to the axis. Default will have
Q1 and Q2 assigned to the left axis group, Q3 and Q4 to the middle
axis, and Q5 and Q6 to the right axis group, allowing six plots per
page. Note that there is an "<Empty>" option. If chosen, the
corresponding axis will contain no graph, and the settings that are no
longer needed will become invisible. Note: The scale of a given
quantity will be recalculated if that quantity is moved to a graph with
different number of vertical divisions, except if the scale for that
particular quantity was previously manually changed by the user.
iii. Precision, or number of digits after the decimal point – automatically
set to number of decimal places of value per division (as set by your
selection in PDA-W), can later be changed by user.
iv. Value per division, minimum scale, maximum scale. This works
similarly as for the Y axis, except that now the number of divisions
(corresponding to the vertical grid lines) can only be changed in the
"Edit ÿ Setup ÿ Gap/Vert. divisions" menu, or by clicking on the
"Format Graph" Toolbar button and choosing the "Gap/Vert.
divisions" tab.

c. These commands appear below the tabs:

i. The "Save Settings" (red) button will save the following information
for all axes: Variable assigned to each axis, Graph Color, Graph
Style, Extra Thick, Precision, Scale minimum and Scale maximum,
Number of horizontal grid lines, Number of vertical grid lines, Color
or B&W graph. Pressing "Recall Settings" will set all those
parameters to the saved values. IMPORTANT: 1) Precision, Scale
Maximum and Scale Minimum are saved for all variables, even those
that are not being exhibited; 2) If "Penetration" or "Elevation" is set
for the Y axis, and a file that does not have penetrations or
elevations information is loaded, PDIPLOT will show Blow Numbers
in the Y axis, and the Blow Numbers scale will be changed to the
saved blow number scale; 3) If "Blow count" is set for any of the X
axes and the loaded file does not have this information, PDIPLOT
will show Q4 instead; 4) If Variables are changed on the PDA-Win
program, or if a PDIPLOT (PIL) file is read with different variables,
results are unpredictable.
ii. The "Recall Settings" (green) button will recall all settings
previously saved with the "Save Settings" button.
iii. The "Average" adjustment acts as a low pass filter, and is used to
smoothen the curves. PDIPLOT calculates the default value so that
about 200 points are plotted. WARNING: If "Average" is set too high,
important information may be suppressed from the graph. This
adjustment can also be accessed through the "Edit ÿ Average"
iv. The "Filter" button calls the Filter Window described in the Table

2. These adjustments can be made by clicking the "Format Graph" Toolbar button,
or on the "Edit ÿ Setup" menu item:

a. On the "Title/description" tab

i. Title font Size: changes font size of Main Title and PD Description.

October 2010 PDIPLOT Manual 7-22

ii. Title font Bold: when checked, main title is shown in bold characters.
iii. Title font Color: when clicked, allows changing the colors of the title
iv. Add description to title: if checked, will add the text of the
"description" field (PD) to the graph title; this field originally contains
the PD information from the PDA-W, or PIL file, but can be edited in
the Table Window.
v. Same Line: PD description will be added to the same line as the title
vi. New Line: PD description will be shown on a new line below the title
vii. Bold: when checked, PD description is shown in bold characters.

b. On the "Comments" tab (allows adjustment of sinuous lines on blow

numbers or penetrations corresponding to "Graph" comments)
i. Style: controls line width and appearance
ii. Color (same color will be used for color or black and white graphs)
iii. Undulation Amplitude: adjusts amplitude of sinuous line. There are
five choices: 0 - flat line; "extra small" - about 0.5 mm peak to peak;
"small" - about 1 mm peak to peak; "medium" - about 2 mm peak to
peak and "large" - about 3 mm peak to peak
iv. Undulation Frequency: adjusts frequency of sinuous line. There are
five choices 10, 20, 30, 50 or 70. Setting represents the number of
segments per graph, or twice the number of cycles per graph.

c. On the "Font/color" tab

i. Preferred font: will change the font of all the letters and numbers in
the graph
ii. Default Graph in “color” or “black and white”: colors can later be
changed using the individual commands. Default style is also
changed. On black and white graphs, the two curves on each graph
box are best distinguished by different line thickness.

d. On the "Gap/Vert. divisions" tab

i. Number of vertical divisions for each graph (Left, Middle or Right).
ii. When Gaps are activated, all curves will show a discontinuity if the
difference between two consecutive blow numbers is greater than
the value shown. This value can be adjusted in steps equal to the
average blow number frequency (calculated as {Last BN-First
BN}/Total Number of Records). Please note that setting this value
too high is equivalent to disabling gaps.

3. All Graph Comments (see also Table Window-Comments) will be shown on the
lower part of the Graph. (Only comments that are “selected” will be graphed - user
can change selection for each comment in the Table mode). A sinuous line will cut
across the corresponding blow number or penetration. Comments are identified
(keyed) by a number at the right of the line.

4. Clicking on "File÷Page Setup" will open the Page Setup Window, which allows the
user to change the printer, paper size and margins. Graphs will only be printed in
Landscape orientation.

5. Clicking on "File÷Print" or typing Ctrl+P will open the Print Dialog Box.

6. Clicking on "Table÷Time Summary" or on the “clock” Toolbar button will open the
Time Summary Window.

October 2010 PDIPLOT Manual 8-22

9. The Table Window
Its upper left part, below the Toolbar, contains changeable titles read from the file, like Project
Name, Description and Pile Name. These titles can be freely edited by the user.

The upper right part contains other data read from the first blow contained on the file, like
cross-section area, elastic modulus, etc. The operator name (OP) and the date and time of the first
and last blows are also shown. In order to change those values (except date and time stamps,
which cannot be changed), it is necessary to open the PDA-Win program with the "See blow", make
the change there, and "refresh" the data.

NOTES: The information contained in the upper right part of the Table Window, like
cross-section area, operator name, dates, etc., can be "dragged and dropped" to the boxes
containing changeable titles like Project Name, etc. It is also possible to "drag and drop"
data from other Windows applications, like word processors, into those boxes.

The lower part contains a “spreadsheet” with values obtained from the PDA-W program: one
column for the blow numbers, one column for the penetrations (LP), one column for Blow Counts,
nine other columns for Quantity Values (Q1 to Q9), and the last column for “Comments”.

NOTE: The Comments shown are those saved in the W01 data file for the data displayed.

To move around the spreadsheet, use the mouse, the keyboard arrows or the following keyboard

Home: go to first column (blow number)

End: go to last column (comment)
Ctrl + Home: go to first column of first row
Ctrl + End: go to last column of last row
PgDn: move one screen down
PgUp: move one screen up
Ctrl + Pg Dn: go to last row
Ctrl + PgUp: go to first row

NOTE: If the above commands are not working, click anywhere on the spreadsheet.

It is possible to change the width of the columns of the spreadsheet. Passing the mouse over the
header row will change the cursor to a horizontal arrow. When this happens, moving the mouse
right or left with the left button pressed changes the column width.

On the extreme right of the toolbar there is a "Units" drop-down menu. The program will first load
using either the unit system last used on the PDA, or the Preferred Unit System as set in "Tools ÿ
Program Options ÿ Units" on the Graph Window. To change the unit system (not possible when
reading PIL graphs saved by PDIPLOT), click on the down arrow and then on the name of the
desired unit system. Or use "Tab" to highlight the unit (or click on it), then press "E" for English, "M"
for Metric or "S" for SI. The data will be reloaded from the PDA-W program, and all scales and filter
settings will be reset. The quantities unit names (for instance "inch" or "mm") are taken from the
PDA-W settings. Please refer to the PDA-W manual to change those settings. NOTE: Upon normal
termination of the PDIPLOT program, the PDA-Win unit system will return to its original setting.

To the left of the "Units" drop-down menu window there is a "Blow Count" drop-down window. This
is used to change the way the Blow Count information is presented. In English units, the default
presentation is "blows per foot" (blows/ft). In Metric or SI units, default presentation is "blows per
meter" (blows/m). The other choices are “blows per 6 inches” (blows/.5 ft - specially useful for SPT),
"blows per inches" (blows/in), "inches per ten blows" (in/10 blows), "blows per 30 cm" (blows/.30

October 2010 PDIPLOT Manual 9-22

m), "blows per 25 cm" (blows/.25 m), “blows per 15 cm” (blows/0.15 m - specially useful for SPT),
"blows per 10 cm" (blows/.10 m) and "millimeters per 10 blows" (mm/10 blows). All options are
available, regardless of the unit system chosen.

The total number of blows is shown below the Operator name. This number will change as blows
are excluded or included.

Accessed by the “Filter” Toolbar button, or “Column ÷ Filter”, the Filter Window allows the user
to set a lower and/or an upper limit for the corresponding variable (quantity to be filtered is selected
by the tab). Values that fall outside the selected range will be excluded, that is, the blow will not be
shown on the graph and tables. The whole row will be crossed out, and the variable values that are
outside the range will be shown in red, so the user can see which variable is forcing the line to be
excluded. The number of excluded blows due to that filter setting will be shown in the "O/Limit" cell,
in bold. Blow counts will not be recalculated.

Selected rows can also be excluded or included using the "-" or "+" Toolbar buttons, respectively.
This can also be accomplished by clicking on "Edit÷Exclude Line" or "Edit÷Include Line". This
action will override exclusions made by the filter function described above. However, values outside
the filter range will still be shown in red.

There are three ways of selecting a range of rows:

1. Click on any column of the first row of the range, then move the mouse to the last
row of the range while pressing the left mouse button ("dragging").
2. Click on the first row of the range, then scroll the spreadsheet until the last row of
the range appears, and left click the mouse on any column of that last row while
holding the "Shift" keyboard key.
3. Click on the first row of the range, then hold the "Shift" key while moving the cursor
by pressing the down or up arrow on the keyboard.

The selected range goes from the first highlighted row to the last highlighted row, regardless of how
many columns are selected. To "unlock" the range, click on any cell, or select a whole column.

Seeing a blow: If there is only one row selected, it is possible to open the PDA-W program on that
blow by clicking on the “camera” Toolbar button, or clicking "Edit÷See Blow". The user can not only
investigate the blow, but also make changes like in wave speed or Jc factor. After making the
changes, it is necessary to refresh the data. After using the PDA-W program, close it for example
by clicking on the "X" icon (it is not recommended that you minimize the program. PDIPLOT reads
new data much faster if the PDA-W is not an open item on the Taskbar).

WARNING: Sometimes when using PDIPLOT the PDA-W program will change its "Serial
Analog/Digital (SAD)" setting. If you are using PDIPLOT on a PAK, make sure to check this
setting before starting to acquire data.

Refreshing the Table: Clicking on "Edit÷Refresh", pressing the F5 key, or clicking on the refresh
Toolbar button will force all data to be read again from the PDA-W program. All changes made on
pile parameters, deleted blows, etc. will reflect on the new table. Changes in unit system or in
Q1..Q9 require that you leave PDIPLOT first.

Entering Drive Log: The Blow Count Editor can be accessed from the Table Window, by clicking
on the “Enter Drive Log” Toolbar button, by clicking on "Edit÷Enter Drive Log" or by pressing
Control + D. It is not available if PDIPLOT senses that a drive log had been entered on PDA-W.

October 2010 PDIPLOT Manual 10-22

On the left part of the Editor there is a table showing all blow numbers, from the one immediately
before the first one on the file, to the last one on the file. Blow numbers that do not exist on the file
are shown with green letters. The penetrations corresponding to blows where PDIPLOT detected
a penetration change are shown on the right column. Those penetrations are also shown on the
drop-down box at the bottom of the "Go to" region. This drop-down box, together with the buttons
at the top, allows the user to browse through the table. Penetrations can be moved up or down
using the corresponding buttons on the "Move" region. It is also possible to enter a penetration on
a given blow number, by double clicking on the corresponding cell. Existing penetrations can be
edited or deleted also by double clicking on its cell.

The "Increment Penetration" button at the top of the table allows the user to add a fixed value to
all penetrations on the table, to the current (cursor) penetration, to all penetrations from the
beginning to the current one, or to all penetrations from the current to the last one.

The above commands allow the user to make small corrections to penetrations already entered
using the blow switch in the field, for example. For entering a new drive log (or for more extensive
corrections on existing ones), the commands on the "Auto" field should be used. Those commands
allow the user to enter a drive log starting from (a) the beginning of the file, (b) from the end of the
file, or (c) from any intermediate penetration (up or down).

1. To enter a drive log from the beginning of the file (often recommended for restrike
testing), choose "from beginning" then click on the down arrow. The cursor will move
to the "Initial penetration" box. Accept the existing value, or change it to a new one.
Then click on "Next" or press Enter. The cursor moves to the "Penetration
increment" box. Enter an increment (default is one length unit). Click on "Next" or
press Enter, and the cursor will move to the "Blow Increment" box. Enter the number
of blows for the first increment. Press "Next" or Enter, and the values are accepted;
the cursor stays at the "Blow increment" box, so that a new value can be entered for
the next penetration increment (if the value is the same as before, just click on
"Next" or type Enter). If the penetration increment is not the same as before, click
on the corresponding box and change it, then click on "Next" or press Enter. Inside
the frame at the bottom of the table the user can verify the number of blows from the
last increment to the current one, and from the current one to end. This procedure
should be followed until the last blow is reached. At this point the "Accept" button is
enabled, and clicking on it or pressing Enter will transfer the penetrations to the
Table Window.

2. To enter a drive log from the end of the file (normally recommended for pile initial
driving installations with many blows), choose "from end" then click on the up arrow.
The cursor will move to the "Final penetration" box. Accept the existing value, or
change it to a new one. Then click on "Next" or press Enter. The cursor moves to
the "Penetration increment" box. Enter an increment (default is one length unit).
Click on "Next" or press Enter, and the cursor will move to the "Blow Increment" box.
Enter the number of blows for the last increment. Press "Next" or Enter, and the
values are accepted; the cursor stays at the "Blow increment" box, so that a new
value can be entered for the next penetration increment (if the value is the same as
before, just click on "Next" or type Enter). If the penetration increment is not the
same as before, click on the corresponding box and change it, then click on "Next"
or press Enter. Inside the frame at the bottom of the table the user can verify the
number of blows from the last increment to the current one, and from the current
one to the start of the file. This procedure should be followed until the first blow is
reached. At this point the "Accept" button is enabled, and clicking on it or pressing
Enter will transfer the penetrations to the Table Window.

3. To enter a drive log from any penetration (suggested use is for a file merged from

October 2010 PDIPLOT Manual 11-22

end of driving and restrike where you position at the first blow of restrike), after
positioning the table to the selected BN and depth, choose "from current", then click
on the up or down arrow. The procedure is similar to item 1 above for the down
arrow, or item 2 above for the up arrow, except that now it is not possible to change
the "Current penetration" value.

After pressing "Accept", PDIPLOT will check if there is a valid initial and final penetration, and if all
the penetrations are continuously increasing. If this is not the case, a warning will be issued, and
PDIPLOT will try to move the cursor to the penetration causing the problem. If no inconsistences
are detected, PDIPLOT will transfer the penetrations and blow counts to the Table Window,
interpolating the penetrations between blow count changes.

Comments: Comments are shown on the rightmost column. Green letters on the cells in the first
and second columns indicate rows containing comments. Comments can be added, removed or
edited. This can be done by clicking on the “ABC” Toolbar button, or pressing "Enter" or clicking on
a highlighted comment cell. This will bring up the "Comments" window. This window also contains
information like blow number, penetration, date and time stamp. Any one of those can be copied
to the comment box by mouse drag and drop. Date and time can be added to the comment by
clicking on the corresponding buttons. A drop-down menu allows the user to define a comment as
"Table" (will be shown only on table printout), "Graph" (will be shown only on graph printout), "Both"
(Table and Graph), or "None" (will be shown only on the Table spreadsheet, and will not be printed).
A prefix [T], [G], [B] or [N] is added to the comment, to indicate its type. Prefixes are for information
only and will not be printed or graphed.

Several additional comment functions are available on the "Edit" menu, or by right clicking when
a comment item is highlighted, or by pressing the shortcut keys mentioned below:

1. Repeatedly pressing F4 will toggle the comment type, from [T] (Table) to [G] (Graph)
to [B] (Both) to [N] (None) – only available when a single comment item is
2. Pressing F3 will initiate a text search. All existing comments are searched for the
text entered by the user. If the text is not found below the current comment, the user
will be asked whether to cancel search or restart from beginning of table. The
search is not case sensitive.
3. Pressing F12 will move the cursor to the next available comment.
4. Clicking on "Change All Column Types" allows the user to change all comment
types at once.

NOTE: PDIPLOT will try to read comments that were recorded with the blow using
the PC command of the PDA-W program. The reading of comments will only work
with PDA-W version 2001.85 or later. If for any reason PDIPLOT is not able to read
the comments, the user will be warned, and program execution will proceed
normally. If PDA-W is later upgraded, the reading of comments can be reactivated
on Setup. Those comments will be of the "Table" type, by default. This can be later
changed as described above.

Clicking the mouse on the header row of any column highlights the whole column. When this
happens, the "Column menu" will pop-up. This menu is also directly accessible on the menu bar.
Menu options depend on the selected column:

1. Go to Maximum and Go to Minimum value on any column. Excluded rows are

always ignored. These functions are also available as Toolbar buttons (up and down

October 2010 PDIPLOT Manual 12-22

2. Go to Value – blow number or penetration, according to the selected column. If a
nonexistent blow count or penetration is chosen, the cursor will go to the next larger
value available. These functions are also available as Toolbar buttons (“#” button).
When there is any inconsistent penetration and/or blow number, an additional button
will be enabled. When this button is clicked, the first inconsistent blow number or
penetration is shown. Clicking on "OK" will move the cursor to that location

3. Open the Filter Window (Also available as a Toolbar button - arrowed parallel lines).

4. Penetration editing functions: those functions are only available when the cursor
is on the LP column.
a. Choosing "Modify LP" on the Column Menu, or clicking on the "LP" Toolbar
button (available with the cursor anywhere in the table), will bring up the
"Modify LP's" Window. This window allows the user to:
i. Change the LP unit. For example, it is possible to show penetrations
in feet while using SI units, or to show penetrations in meters while
using English units. If the "Recalculate LP's" option is checked, then
all penetrations are converted to the new unit system. For example,
if penetrations are in feet and both "Change LP unit to m" and
"Recalculate LP's" are checked, clicking on "Accept" will make all
penetrations to be multiplied by 0.3048 to convert feet displayed into
meters. If "Recalculate LP's" is not checked, then penetrations are
not changed, and only the unit label is changed accordingly.
ii. Add (or subtract) a constant value to all penetrations.
b. Choosing "Show Elevation" will make all penetration values to be changed
to the corresponding elevations. Default values and precision are taken from
the PDA quantity "EL". After "Show Elevation" is pressed, a new option "Edit
Elevation" is available. Clicking on it, or simply clicking or pressing Enter on
an elevation value, will bring up the "Edit Elevation" window. This window
allows entering or changing the ground elevation and the inclination angle
for battered piles. It is also possible to change the Elevation precision
(number of digits after the decimal point). Changing back to Penetrations
mode is possible by choosing "Show Penetration" on the Column menu.

NOTE: If elevations were not entered on the PDA-W using the Drive
Log Editor, "EL" values will be all zero. In this case, PDIPLOT will not
show Elevations as an option for the Y axis on the Graph window,
and it will not print Elevations on the tables. In those cases it is
necessary to open the Edit Elevation at least once before elevation
values can be actually used (if the elevation window is opened and
no values are entered, a zero ground elevation and a vertical pile is

5. Edit USR: only available on USR column (if selected as a PDA-W Quantity), brings
up the Edit USR window. This window also appears when clicking or pressing Enter
on any USR value. Typical uses of the USR (user observation) quantity are
recording of drop heights of free fall hammers, or recording of kinetic energies of
hydraulic hammers. The user can change the quantity unit and description. A new
unit and/or description can be set as default, instead of the values set on the PDA-W
program, or instead of a previously set default value, by checking the appropriate
boxes. A new value can be entered for all blows, for the current blow (first blow if the
whole column is highlighted), for all blows from beginning to current, or all blows
from current to end. If the "current blow" option is chosen, clicking on the up or down
arrow (or pressing the corresponding keyboard keys) will accept the value, and

October 2010 PDIPLOT Manual 13-22

move the cursor to the previous or next blow, respectively. This allows for a faster
entering of different values for each blow – for example, enter the first value, click
on the down arrow, enter next value, and so on.

6. Recalculate ETR: only available if ETR (energy transfer ratio), EMX (maximum
energy) and USR (user observation) are shown. It brings up the "Recalculate ETR"
window, which can also be called by clicking or pressing Enter on any ETR value.
There are two options:
a. Calculate ETH - Energy Transfer-Drop Height – this option is typically used
on tests using drop hammers with variable heights. Make sure drop heights
were entered on the USR column. Enter a Hammer Weight (HW) in Force
Units (kip, ton or kN); ETH is calculated as 100 x EMX/(HW x USR). NOTE:
"ETH - Energy Transfer Ratio - Stroke" is calculated by version 2002.092 or
later of the PDA-W program as 100 x EMX/(WR x STK), and cannot be
edited on PDIPLOT.
b. Calculate ETK - Energy Transfer-Kinetic -- this option is useful for example
with hydraulic hammers. Make sure Kinetic Energy (in k-ft, tn-m or kJ) was
entered on the USR column; ETK is calculated as 100 x EMX/USR.

7. Modify BLC: (only available if penetrations are continually increasing) The "Modify
BLC" window comes up if a blow count item is double clicked, or by right clicking the
mouse after a range of rows had been selected. It allows editing of blow count
values, without changing penetration values. Blow count values can be changed
individually (Current only) or in groups: Selection (if a range had been selected
before right clicking), All, Beginning to current, Current to end, Blow number range
and Penetration range. If "Current only" is chosen, clicking on the up or down arrow
(or pressing the corresponding keyboard keys) will accept the value, and move the
cursor to the previous or next blow, respectively. This allows for a faster entering of
different values for each blow – for example, enter the first value, click on the down
arrow, enter next value, and so on.

WARNING: this function should only be used for small adjustments of

automatically calculated blow count values.

Deleting and renumbering blows: If editing a W01 file, it is possible to completely remove a blow.
To do that click on the blow, press See blow, delete it on the PDA-W Window (by pressing Ctrl +
Del, then Yes), then Refresh the Table. If you have PDA-W version 2001.085.011 or later, it is
possible to renumber the blows and/or change penetrations by choosing Edit÷Increment (BN/LP)
on the PDA Window. It is also possible for example to remove "bounce" blows with the Edit÷BN
Filter function. After Refreshing the data on PDIPLOT, all changes made on the PDA Window will
be transferred to the PDIPLOT Table. Please note that all scales and filter settings will be reset
after Refreshing.

Entering Test Date: When a W01 file is created on PDA-Win by reading raw data from the PAL-R
(R01 file read from the memory card), sometimes the date and time information is lost. When this
happens, PDIPLOT will read 00:00:00 as date and time for all records, and the user will be warned
that editing and printing of time summary will be disabled. Only in this case it is possible to enter
a test date, by clicking on "Edit÷Enter Test Date". On the "Enter Test Date" Window, click on the
correct day on the calendar, then click on "Accept" (or just press "Enter"). To move from one month
to the other, click on the right and left arrows at the top of the calendar.

October 2010 PDIPLOT Manual 14-22

Printing: Printing can be done for the entire file or only a selection of blows. Clicking on "File÷Print
(Average, Range or Statistics)" or on the print button the user will be asked to choose between
Average, Range or Statistics mode. PDIPLOT will always remember the last choice made. This will
bring up the Print Preview Window, which is described next.

10. The Print Preview Window

This window shows a preview of the table output, before it is sent to the printer. It also allows the
user to customize the printout. According to the user choice on the Table Window, the Print Preview
Window will open in one of three modes:

10.1 Average Mode

Only averages, standard deviations, max and min values are printed for each set of blows.
Individual values are not printed. There are three possibilities in this mode, according to the
selection made in the Print Options Window (accessible by clicking on the "Options"
(wrench) Toolbar button):
a. Depth increment - this option is only available if the "blow count" column is not
empty, and if penetrations are continually increasing. It will print the average,
standard deviation, etc. for all blows between depth increments. There are two ways
of choosing the depth increments:
i. By entering a "Transition Length Increment" (Transition LI), a "Transition
Depth" and a "Final Length Increment" (Final LI). The depth increment will
be equal to Transition LI from beginning of file to Transition Depth, and equal
to Final LI from Transition Depth to the end of file. PDIPLOT tries to set the
depth increments so that the intervals always end on a multiple of Transition
LI. For example, if the first LP on a file is 1.38 and Transition LI is set to
0.25, the first interval will go from 1.38 to 1.50, the second interval will go
from 1.50 to 1.75, and so on. NOTE: PDIPLOT tries to read the values of
Transition LI, Transition Depth and Final LI from the PDA-W program. If blow
counts were not entered on the PDA-W program, Transition Depth and Final
Length Increment will both default to one (meter or foot).
ii. When blow counts were entered in the field using the blow count switch (or
the "End" key), or using the PDIPLOT Blow Count Editor, a "Use Drive Log
Data" box will be available. When this box is checked, the depth intervals will
correspond exactly to the depths where the blow count switch was pressed,
or to the depths entered on the PDIPLOT Blow Count Editor. This option is
also available when blow counts were entered using PDA-Win program
version 2003.094.011 or later. In this case, the depth intervals will
correspond to the values on the "depth" column of PDA-Win's "Drive log"

b. Blow increment - prints averages, etc. for each group of blows within a certain blow
increment. PDIPLOT will calculate the default blow increment based on the size of
the file, for a reasonable final number of pages. The user can later change the blow
increment on the Print Options Window. Please note that processing time depends
on the total number of pages.

c. Line increment - prints averages, etc. for each group of recorded blows within a
certain line increment, regardless of their actual blow numbers. PDIPLOT will
calculate the default line increment based on the size of the file, for a reasonable
final total number of pages. The user can later change the line increment on the
Print Options Window. Please note that processing time depends on the total
number of pages.

When printing averages, an additional frame appears in the lower left part of the Print

October 2010 PDIPLOT Manual 15-22

Options Window. This frame has two check boxes:
a. The leftmost check box selects whether or not PDIPLOT will add a blank space
between increments.
b. The rightmost check box selects whether overall statistics for all non-excluded blows
will be included at the end of the increments. Values shown on the overall statistics
will be the same as those chosen for the increments (possible values are Average,
Standard Deviation, Maximum, Blow number of Maximum value, Minimum, Blow
number of minimum value).

10.2 Range Mode

Will print the variables for each blow; the average, standard deviation, max and min values
for the whole range can be included at the end of the last page, according to the options
checked in the "general" section of the Print Options Window (accessible by clicking on the
"Options" button). PDIPLOT can print every blow, every second blow, etc., according to the
Line Print Frequency setting.

If Line Print Frequency is equal or greater than two, the "Advanced" button is enabled.
Clicking on this button will open a sub window that allows the user to:
a. Choose if the line print frequency will apply to all lines (blows), or if a certain number
of lines either at the beginning or at the end of the record will be shown without
skipping any line.
b. Select whether all blows will be analyzed for average, standard deviation, etc., or
if only the printed blows will be analyzed (by means of the check box "Include all
blows in statistics").

The default line print frequency setting is calculated based on the size of the file, for a
reasonable final number of pages. Please note that processing time depends on the total
number of pages.

10.3 Statistics Mode

Will print only the Average, Standard Deviation, etc. for a range or ranges of blows,
according to the options checked in the "Statistics for" section of the Print Options Window,
accessible by clicking on the "Options" (wrench) button. The options are:
a. Entire File (default).
b. A Selected Range of blows. A selection can be made before or after the Print
Options Window is opened. To make or change a selection after the Print Options
Window has been opened, click on the "Select Range" button. The main table will
be shown, allowing the user to make or change the selection. Clicking on the "Back"
button will close the main table and return to the Print Options Window, where
further selections can be made.
c. All blows from the first to a specified number (first XX records).
d. All blows from a specified number to the last one (last XX records).

Statistics will be shown in the following order, according to the selections made: Entire file,
First XX blows, Selected range, Last XX blows.

10.4 Features common to all modes

At the top of the Print Preview window there are Toolbar buttons whose functions are, from
left to right (equivalent menu options are in parenthesis):
a. Send the output to the printer ("Print Table" on the Print menu). See "Printing"
b. Print both Table and Graph ("Print All" on the Print menu). The user can choose
whether the graph is printed before or after the table (PDIPLOT will remember the
last choice). NOTE: this option is particularly useful when using Adobe® Acrobat®,
since it will generate one single PDF file for both table and graph.

October 2010 PDIPLOT Manual 16-22

c. Adjust page settings (only available through "Page Setup" on the Print menu). See
"Page Setup"
d. Change the print options ("Options" on the Edit menu, or Control+O). There are a
number of options that are common to all modes:
i. At the top of the Print Options Window there are nine drop-down selection
boxes. They allow the user to select which Quantity variable is shown on
each column. This allows you to rearrange the order of Quantities for
ii. Below the drop-down selection boxes there is a row of check boxes. They
allow the user to eliminate columns. This is done starting from the rightmost
column, and proceeding with the columns to the left of the eliminated
column. Reactivation of columns is done from left to right. NOTE: PDIPLOT
does not check if there are repeated columns; it is possible, for instance, to
choose the same Quantity variable on all nine columns.
iii. Underneath the check boxes there is another row of drop-down boxes used
to change the precision, that is, the number of digits after the decimal point
of each variable. Default precisions are those used by the PDA-W program
last time it was closed. Please refer to the PDA-W manual for instructions on
how to change those default precisions.
iv. Below the precision boxes the user can choose whether to print the entire
file or a selection. The selection of a range of rows to be printed must be
made in the Table Window, before calling the Print Preview Window.
v. The check boxes in the "general" area allow the user to include or remove
the following items:
(1) Average of the values (with AVG). NOTE: in average mode, if no
item is selected in the "general" area, PDIPLOT will still print the
average of the blows
(2) Standard deviation (with STD)
(3) Maximum value (with MAX)
(4) The number of the blow where the maximum value occurred for the
first time (with MAX Blow#)
(5) Minimum value (with MIN)
(6) The number of the blow where the minimum value occurred for the
first time (with MIN Blow#)
(7) Pile properties: area, length, wave speed, specific weight, elastic
modulus (Print Pile Properties)
(8) Time Summary (Print time summary) - will list the driving periods and
stopped periods, as detected by PDIPLOT (see Time Summary
Window ). Will also list total driving time, total stop time and total test
(9) Comments (Print comments) - only "Table" [T] comments will be
shown. See Table Window - Comments.
(10) The user can also select if a list of the quantities printed, with their
corresponding explanations, is printed at the top of the first page, at
the bottom of the last page or not at all.
vi. The "Save Settings" (red) button will save the following information for all
columns: Variable assigned to each column, Active columns, Show Depth,
Precision of numbers on each column, with AVG, with STD, with MAX, with
MAX Blow#, with MIN, with MIN Blow#, Print Pile Properties, Print time
summary, Print comments, Placement of quantity descriptions, Line print
frequency and Depth or Line Increment.
vii. The "Recall Settings" (green) button will recall all previously saved settings.
A warning message will come up showing the names and order of the
variables that were being used on the PDA-W program when the "Save
Settings" button was last pressed. If the current variables are not the same,

October 2010 PDIPLOT Manual 17-22

or if they are not on the same order, the recall settings operation may not
bring the expected results.
e. Change the font name and size of the whole printout
f. Show the first page, previous page, next page or last page
g. Close the Print Preview Window and go back to the Table Window ("Close" on the
Print menu, or Control+E)

Copy to clipboard – part or all the text can be copied to the clipboard and transferred to
other programs like Spreadsheets, Word Processors, etc.
a. When the cursor changes to an "I", text within a single line can be selected by left
clicking and dragging of the mouse.
b. It is also possible to select the whole text, by clicking on "Edit÷Select All" or typing
c. Once a text is selected, copying to the clipboard can be accomplished by clicking
on "Edit÷Copy Selection", or by right clicking of the mouse and then clicking on
"copy", or by typing Control+C.
d. It is also possible to copy only the statistics at the end of a Print Average or Print
Range table, by clicking on "Edit÷Copy÷Statistics".
e. It is not possible to copy the graph screen to the clipboard.

11. Automatic checking of Penetration, Blow Number, Blow Count and Data Consistency
When PDIPLOT reads data from the PDA-Win program, it checks if there are penetration changes,
if the penetrations are continually increasing, if all blow numbers are greater than zero, and if blow
numbers are continually increasing. Also, PDIPLOT will detect any change in LE (Length below
gages), AR (Cross-section area), SP (Specific Weight), EM (Elastic Modulus), JC (CASE damping
factor) and WC (Overall wave speed).

If penetration changes exist and are continually increasing, PDIPLOT will calculate the blow
count, and interpolate the penetrations between changes. This is done according to the
blow numbers and penetrations read from the file. If the penetration of the last blow is not
different from the one before, PDIPLOT will extrapolate the final penetration, based on the
last blow count available. This penetration is probably wrong, so the user is warned of the
fact. In order to change this last penetration, click on any column of the last blow, click on
"See Blow" , change the penetration on the PDA-W program and then refresh the data.

If penetrations are not continually increasing, (a) the program will issue a warning message,
indicating the first inconsistent penetration, (b) Blow Counts will not be calculated, and (c)
it will only be possible to draw plots against blow number. In this case, the Penetration
column on the Table Window will show the penetrations as they were read from the PDA

If any blow number is not greater than zero, or if blow numbers are not continually
increasing, the program will issue a warning message, indicating the first inconsistent blow
number, and it will be impossible to enter Blow Counts using the Blow Count Editor.
WARNING: Please have in mind that in those cases calculated blow counts and
penetrations are in error.

If the time stamps on subsequent blows are not continuously increasing, the program will
show a warning message, and no time summary will be available in this case.

Whenever a change in LE, AR, SP, EM, JC or WC is detected, it will be listed in a warning
message. Also, a Table Comment showing the new value will be shown on the row where
the change was detected (this comment will not be saved in the W01 or X01 file). If a
comment already existed on that row, the new comment will be appended. Note: the auto

October 2010 PDIPLOT Manual 18-22

comment feature can be switched on or off for each individual quantity by clicking on Edit÷
Program Options ÷Operation; the warning message cannot be switched off.

When there is any inconsistent penetration and/or blow number, an additional button in the "Go to
value" option of the Column menu will be enabled. When this button is clicked, the first inconsistent
blow number or penetration is shown. Clicking on "OK" will move the cursor to that location.

IMPORTANT: PDIPLOT will not save files with inconsistent penetrations or blow counts. If
you intend to save the file at the end of the work it is therefore important that all inconsistent
penetrations and/or blow counts are corrected before saving. To accomplish this, the
recommended method is: read the PDA file with inconsistent penetration(s) or blow
number(s); after table is loaded click on the depth or BN column, click on "Go To Number",
click on "Find first inconsistent penetration (BN)", open PDA (by clicking on the camera
icon), correct data, refresh. Repeat this process until no further errors are detected.

Besides reading the penetrations, PDIPLOT also reads the Blow Count information sent by the
PDA-Win program. If this information is available (that is, if all blow counts sent by PDA-W are not
zero), the program will use that information, and will not calculate the blow counts and the
interpolated penetrations. This is the case if a driving log has been entered through the "Edit÷Drive
Log" option of the PDA-W program, or if the blow counts for all blows have been entered through
the "Capwap Adjust" option. Changing the blow count presentation (for example to blows/inch,
blows/.25 m, etc.) is still possible in those cases, as PDIPLOT will make the necessary

12. Appending (merging) files

Appending or merging PDA files can be accomplished by selecting “FileÿAppend File” or by typing
Control+F on the Table Window.

NOTE: The preferred method for appending or merging files is by using the "EditÿMerge
W01 file" function of later versions of the PDA-Win program. The PDIPLOT Append File
function should only be used in case files with different sensor arrangements must be
merged, for example, one file containing records using two strain transducers and two
accelerometers with another file containing records from the same pile, but using four strain
transducers and four accelerometers.

13. Time Summary Window

This window is accessible from the "Table÷Time Summary" menu of the Graph Window, or from
the "Edit÷Time Summary" menu on the Table Window, or by clicking on the “clock” Toolbar button
on either Table or Graph windows. It shows a table containing a list of the main time events
detected by PDIPLOT. Besides the "Start" and "End" events corresponding to the first and last
blows, PDIPLOT will detect a "Stop" event whenever the time interval between a given blow and
the next recorded blow is greater than the "Minimum Stop Time" shown below the table. The blow
immediately following a "Stop" event will correspond to a "Start" event. By changing the "Minimum
Stop Time" (in minutes), events will be added or removed from the table.

For each event, the table contains columns showing, from left to right: Event type (Start, Stop, End),
Blow Number, Penetration, Date and time, Suggested comment and Existing comment. Suggested
comments are automatically generated by PDIPLOT, according to settings which can be changed
using the "Customize" button described later. Existing comments are copied from the Table
Window. Individual suggested comments can be changed into existing comments by double clicking
on them, or pressing "Enter" while the suggested comment is highlighted, or clicking on the "Single"
button on the "Accept Comment" field below the table. If there is already a comment for that blow,
the user will be asked if it should be replaced. Clicking on "Yes" will allow the user to edit the
comment before accepting it. Clicking on the "All" button of the "Accept Comment" field below the

October 2010 PDIPLOT Manual 19-22

table will accept all suggested comments. All existing comments will be replaced, so the user will
first be asked to confirm the operation.

Clicking on the "Customize" button will bring up the "Customize Time Summary" window. This
window allows the user to change the default text for the different kinds of events: Start (of test),
(Intermediate) Stop, Restart and End (of test). A blank default text means no comment will be
suggested for this kind of event. Clicking on the "Add Date", "Add Time" or "Add Interval" buttons
will place a key text in the cursor location, which will be replaced by the corresponding information.
A drop-down box at the right of each custom text allows the user to choose the kind of comment
each event will default to. Options are: [T]Table - Comment will only be shown on Table printout;
[G]Graph - Comment will only be shown on Graph printout; [B]Both - Comment will be shown on
both Table and Graph printouts; [N]None - Comment will not be shown on any printout, it will only
appear on the Table Window. NOTE: Time labels can be customized on "Tools÷Options÷Labels".
See item 16. Program Options.

Existing comments can be edited from the Time Summary Window, by double clicking on them, or
pressing "Enter" when a comment is highlighted.

14. Saving Graphs

Saving of graphs for future use is available on the Graph Window and on the Table Window, by
clicking "File->Save As" or File->Save ("Ctrl + S"). The default file name is the Pile Name, with
extension .PIL (e.g. TP-1.PIL). The first time a file is saved, only "Save As" is activated (the user
will be asked to confirm file name). After first saving, pressing "Save" (or "Ctrl + S") will save without
asking for a file name. Clicking the Save Toolbar button on the Table Window will always invoke
the "Save As" function.

15. Reading Saved Graphs

On the Graph Window, clicking "File÷Open PDIPLOT File" or typing "Ctrl + O" will allow the user
to read data previously saved with the "Save" Function. After a ".PIL" file is chosen, PDIPLOT will
open a new graph with information read from the existing PIL file. Please note that it is not possible
to change the unit system of stored graphs. Also the "See Blow" and "Refresh" functions are
disabled. All other functions are available.

NOTE: PDIPLOT will not open PIL files containing any blow with penetration or blow
number lower than the one before. Files with zero or negative blow numbers will also be
rejected. Please refer to "A few words about Penetrations, Blow Numbers and Blow Count"
for further explanations.

16. Program Options

Program options can be changed via the "Edit÷Program Options" menu of the Graph Window. It
has tabs for different groups of adjustments. All settings are saved and used when opening the next
PDA file, or next time PDIPLOT is started. The adjustments that can be made are:

a. Under "Units"
i. Preferred Unit System (English, Metric or SI)
ii. Open PDA File with Preferred Unit System (as above) or with PDA Unit
System (as used the last time PDA-Win was closed)

October 2010 PDIPLOT Manual 20-22

b. Under "Files"
i. Help File - File called when help is invoked (default is file Plothelp.htm,
usually located in the "Program Files-PDIPLOT" folder).

WARNING: changing the file to an incorrect folder or file name will

cause the help feature to stop working

ii. Owner File - File containing the owner name for printouts. Normally file
"PDIPLOT.OWN" is located in the "Program Files-PDIPLOT" folder.

WARNING: This file is encrypted. Trying to edit it will cause the

program to stop working

c. Under "Labels"
i. The words used to represent time intervals like day(s), hour(s), etc. on time
summaries can be changed here.

d. Under "Operation"
i. Don't read comments: when checked, PDIPLOT will not attempt to read
comments entered on the PDA-W program with the PC command. If this box
is not checked and PDIPLOT is not able to read the comments, the box will
be automatically changed to checked.
ii. Start with Graph or Table: sets the starting window after a file is loaded
iii. Auto comments for changes in: when checked, a comment will automatically
be added on blows where a change in the corresponding quantity has been

17. Page Setup

Clicking on "File÷ Page Setup" from the Graph Window or from the Print Preview Window will bring
up a dialog box where the user can select the printer, the paper size and the margins. Margins will
be entered in the Preferred Unit System chosen on "Setup÷ Units". After any change, the graph
or table is redrawn on the screen. A "Sample" picture shows the aspect of the paper with the
margins drawn as blue lines. Please note that the size indicated in the caption of the "Sample"
frame reflects the total paper area used by the particular printer chosen. This size is slightly smaller
than the total paper size. Changing the printer will change the useful paper size, and will slightly
change the size of the picture on the screen. Also note that only four types of paper are allowed:
Letter, Legal, A4 and A5, and the orientation is always "Landscape" for Graph and "Portrait" for

If after any change in paper size or margins the resulting graph size in any direction is smaller than
about 100 mm (4"), the user will be warned and the change will not be accepted. However, even
if the margins are accepted, the user has to check if the numbers and letters are not overlapping.
In this case, smaller margins or a larger paper should be chosen.

18. Printing
Clicking on "File ÷ Print" (or typing Ctrl+P) on the Graph Window or on the Print Preview Window
will bring up the print dialog box. This allows the user to choose the number of copies (maximum
is 100) or, for Table printout, print all pages, only the current page or a range of pages. Clicking on
"Print" (or pressing "Enter") will send the data to the printer. Clicking on "Don't Print" (or pressing
"Esc") will leave the dialog without printing. NOTE: The additional copies will be generated by
PDIPLOT, so it does not depend on a particular printer driver allowing multiple copies or not.

October 2010 PDIPLOT Manual 21-22

19. Opening several documents
Several documents can be opened in one session of PDIPLOT, by repeating the "Open PDA File"
or "Open PDIPLOT File" procedure. To change from one document to another, click on "Windows".
The Graphs can be shown Cascaded, Tiled Horizontally or Tiled Vertically.

20. Optimizing program performance

As the number of records becomes larger, the time it takes for PDIPLOT to perform operations like
calculating averages for drawing the six curves, or checking through the entire table when a filter
setting is changed can become noticeable. Of course, the actual time will depend on the computer
processing speed and memory. However, there are a few measures that can speed up the use of
the program, without losing any information:

a. Avoid reading big W01 or X01 files. Use the PDA-W "Squeeze W01 file" option first.
With this option, it is possible to save some records in the beginning, some at the
end, and skip some blows in between, so that no important information is lost. Have
in mind that usually graphics do not look very good if more than 200 points are
shown. That is why by default PDIPLOT sets the "Average" number so that about
200 points are shown. Tip: If you want to keep the original W01 or X01 file with all
records, save the reduced file for PDIPLOT under a different name.
b. After reading a W01 or X01 file on PDIPLOT, save the graph as a PIL file. It is much
faster for PDIPLOT to read PIL files than W01 or X01 files. If you need to combine
several files, generate a PIL file as soon as all W01 files have been read.
c. Avoid "Refreshing" the data unless required (refreshing the data repeats reading the
data again by PDA-W).
d. Avoid using the filter function to eliminate only a few too low or too high values. Use
"Go to Max" or "Go to Min", then exclude the corresponding line. Repeat the
operation for eliminating the next lowest or highest value.
e. Make sure that the PDA-W program is closed before opening PDIPLOT. While using
PDIPLOT, close any instance of the PDA-W program that appears on the task bar.
f. Do not open the same W01 or X01 file twice on one PDIPLOT session.
g. Before going back to PDIPLOT, make sure to close the PDA-W program after it is
opened with the "See" button.

October 2010 PDIPLOT Manual 22-22



PDA Example Data
for PDA-W Operation

compiled August 1999 (March 2000 edits) by:

Pile Dynamics, Inc.

4535 Renaissance Parkway
Cleveland Ohio 44128 USA

tel: 216-831-6131
fax: 216-831-0916

(C) Copyright 1999, 2000, by Pile Dynamics, Inc.

This is intended to give a brief look into the power of the PDA in solving problems, and
to alert the user to assure good data quality. Different examples teach different ideas.
A basic knowledge of PDA operation is assumed. The PDA-W program is the basis of
this discussion. If you have not yet read the PDA-W manual, read it, and particularly
read chapter 1 on operating commands. Basically these commands are two letter
inputs; some commands cause action to occur while others require a numerical input
value or a character string (title or short name).

You should know how to input data such as pile length (below gages to pile toe; type
"LE30" to enter a length of 30) and cross sectional area (type "AR100" if area is 100,
or type just "AR" and an area calculator will assist you). LE and AR are located in upper
left of the screen, along with the pile properties Modulus EM, density SP, and wave
speed WS. For concrete and timber piles, the density SP must be entered (density may
vary from one timber pile to another so should be measured for each pile). The wave
speed WS for concrete and timber is also variable and must be measured (or
estimated). The modulus of elasticity EM is computed automatically from SP and WS.

To obtain real data you also have to enter the calibrations (A1, A2, F1, and F2) and
select the accelerometer type and perhaps the trigger channel. Let us say that strain
transducer 1876 has a PDI calibration of 106 and is attached to F1. You can enter
this in the F/V Sensors page of New input, or by the AF Icon if reprocessing old data.
Either way if a calibration has already been entered it will be recalled, or prompt you
to enter the calibration for future reference. This is a good way to always have your
calibrations with you in the field. The other sensor inputs work similarly. If you have
sensors attached, you might tap them while in Accept (Accept-Standby is selected
by the [F2] function key. Check the Offset (type OF while in Standby or click the
balance check Icon). Try the Calibration Test (type CT while in Accept, or click the
CT Icon). Read about these features in the PDA manual. Since, we will primarily
concentrate on reviewing the example data provided, we will spend no further effort
on transducers and data acquisition and instead concentrate on data review and
interpretation while in Standby (use [F2] function key to get into Standby).

Of course, to follow these data examples, you will first have to load them on your PDA.
Since your SL location is arbitrary, this discussion will use the large blow sequence
number to identify specific blows. (If you have a new PAK, the example data may
already be loaded into the C:\EXAMPLES directory)

HELP available at all times by pressing the [F1] function key (which is the most
important portion of the PDA Manual).

This text and data files contains several examples. You should go through Example 1
in great detail, as it describes many of the commonly used features and data
interpretation methods. The other examples may demonstrate certain items, or other
features, and you might want to review them. Again your understanding of Example
1 is very important. Have fun.

Example 1 (Bent 17-2)

Purpose: This data set has many features, such as hammer performance, stress
issues, capacity including setup and data quality. We will spend extra
effort here to fully describe many features which we will automatically
assume you know when you try other data sets. Please go through this
data set in detail.

Concrete pile (16 x 16 inch; 40.6 x 40.6 cm), prestressed

Hammer: D36.23
Soils were silts (as I recall, but really doesn't matter for these discussions)
Desired ultimate load was (500 kips; 227 T; 2230 kN)

1 For a concrete pile we are interested in compression and tension stresses. We are
almost always interested in capacity and hammer energy. For a diesel hammer, the
stroke is of interest. Therefore select appropriate "Quantity" results which include:

CSX maximum compression stress at measured location averaged for section

CSI maximum compression stress of either strain transducer
TSX maximum tension stress computed for any location below sensors
EMX maximum energy transferred to pile
FMX maximum force in pile (CSX * Area)
STK ram stroke calculated by Saximeter formula for open end diesel hammers)
RA2 ultimate capacity estimate (independent of damping assumptions)
RMX ultimate capacity estimate based on JC damping factor)
DFN final displacement at end of the blow (set)

We can accomplish this individually, for example, by entering Q3TSX to put TSX
into the third quantity Q3. To see all the output quantities and assign them to
output values, go to the View menu and select “Quantity List.” The buttons with Q
values will be grayed out, but will turn to Q3=TSX, etc. when the TSX output quantity
is selected. These selections can be made for all nine output quantities.

2 Go to the first blow of the file (actually it is BN 10). You can get to the first blow
by clicking on the button with two vertical lines and an arrowhead pointing left (||<)or
type SL1. Type RI to recall the input parameters (LE, AR, etc) for this pile.

Note that there are a few warning messages on the screen. Look at the individual
signals; type DPFV or right click on graphics area. Right clicking brings up a list
of possible graph displays. Note that this selection of outputs can also be shown
by going to the View menu, then to the “Graph Display.” From this menu, choose
“Individual Force/Velocity.” Experiment on your own with a few of the choices and
fid your favorite - DPFW is my personal favorite for routine use. This DPFV
displays all both strain (force), and both velocity signals. We see that the two
velocities have somewhat similar shapes, while the forces are different at the first
peak by more than the allowed ratio (3/4 or 4/3).

3 Scroll through the data. Use PgUp or the mouse wheel (if you have one) to get to
the next blow. Notice that only the F1/F2 warning box remains.

Actually, you will notice for all blows except the first that the V1/V2 warning never
shows and the two velocities are similar (one exception is BN 911/1022 [SL56]
where one velocity is obviously bad). The strains are sometimes similar (see BN
1032/1143 [SL75]; note the F1/F2 warning box is absent) and sometimes quite
different (see BN 490/601 [SL31]; note the F1/F2 warning box is present). Thus
when you see this warning, you should probably look at the individual strains using
DPFV. For BN 490/601 [SL31], you could also compare results of CSI (3.89 ksi,
26.9 MPa) versus CSX (2.86 ksi, 19.3 MPa) and see that the stress of one
sensor is significantly higher (CSI) than for the average (CSX). You might
want to set a Compression Limit (CL) of say 4 ksi or 28 MPa as a visual reminder
of what you consider a limiting stress. Type CL4, or CL28.

4 You should at some relatively early blow confirm the wave speed (find BN 70/181
[SL10] with the PgUp or PgDn; we suggest this blow as it is an early blow with good
magnitude and sharp impact.). You could now return to force and velocity (type
DPF). Perhaps expand the time scale (click the “T>” or “T<“ buttons on the
toolbar, or hit the [F8] function key). Although the force scale is automatically
selected and the scale (FS1500 or FS6000) is displayed just in the upper left of the
force plot, you could change the force scale (click the toolbar button with the
curve and red upward pointing arrow or type FS1200 or FS5000).

We prefer to use the rise-to-rise method, so you need to look at wave up and wave
down (hit the DPF/DPW special key, or press [F7], or type DPW). There are
two sets of time lines describing 2L/c. The full height dotted lines line up with the
first peak (and 2L/c later), assuming the delay DL is zero. The half height dashed
lines should line up with the start of the initial rise of the downward traveling
compression wave. We will concentrate on these dashed markers. In our case,
note that the second, or right, dashed time marker is positioned about where the
wave up curve begins to fall (this is the rise of the upward traveling tension wave).

You could adjust this marker with the left or right cursor arrows (try this and
observe also that the value WC changes; WC is located just below WS in upper
left). If you move the marker too far to the right using the right arrow, a new
warning box(BTA:) is highlighted, suggesting that there may be damage (finds a
significant relative decrease in wave up prior to the right dashed marker; a tension
wave before the end of 2L/c which must be due to an impedance reduction or
probably in this case the pile bottom).

5 The PDA has a routine which can try to find a suggested wave speed (type SW
for Select Wave speed). While it may get you close, the routine is not yet perfect
so you must review and final adjust with the left and right arrows. If you change WC,

you should re-enter WS to match this WC, and note that EM (Elastic Modulus) will
change with WS to meet the requirement that E = ρc2. For now the original WS is OK
(type RI to recall the original parameters, optionally change FS again as it was reset
by RI).

6 We suggest that the display be returned to force and velocity (right click on the
plots and choose or type DPF, or DPFW; we prefer DPFW). Note that the
compression stresses for this blow are within the CL limits. The tension stress TSX
is 0.58 ksi (3.9 MPa) which is below the designed prestress of 0.8 ksi (5.4 MPa).

7 The stroke STK for BN 70/181 [SL10] is 6.18 ft (1.883 m) and the energy is 17.09 kip-
ft (23.2 kN-m). What is the energy rating of the D36-23 hammer? Let's find out. Go
to the Edit menu, then select “Hammer Properties” or press Ctrl-H. In the box
that appears, click the “List” box. The box that appears is sorted by ID number.
Because DELMAG is first in the ID, D36-23 is close and can be found by scrolling
down. Click on the D36-23 line to select the hammer then click “OK.” You will be
taken to the original dialog box. Note the new inputs in the fields for the D36-23
(Ram Weight = 7.93 kips, Energy = 88.50 kip-ft). Now click “OK” to enter this
hammer selection.

Note that the measured energy for this blow is considerably less than the rating
(about 19%). You could calculate this using the Windows calculator by entering
(17.09/88.5<Enter> or 23.2/119.99<Enter>). The PDA also has an energy transfer
ratio function ETR, and so you might want to type Q9ETR and then this calculation
will be performed for every blow automatically (see the lower right corner of the
quantity result box and ETR is calculated and displayed as 0.16). Typical ETR for
open end diesel hammers is about 25% at end of driving (this blow is not yet end
of driving, so the 19%, although low, is nothing to be alarmed about, yet...).

8 You could scroll through the data using the PgUp key, but there is another way. The
PDA can automatically play the data at about one blow per second (gives you some
time to think) with the RA function (type “RA”), or go to the Options menu and
select Replay and Forward (note Slow is checked in the Replay submenu).
This can also by done by clicking the “RA” toolbar button with the green arrow
pointing right. The replay then will continue until either the end of the data set or
until you stop with the ESC key. Look at the ETR value (currently .18) then Be
ready to click the circled RA with a line through it button or hit ESC when ETR
suddenly drops (it will occur at BN 190/301 [SL16]; go back using PgDn to this blow
if you miss it). Note that the stroke reduced to only 5.57 ft (1.685 m) for BN 190/301
[SL16]. Further, the blow is much smoother, and the force amplitude is greatly
reduced. The hammer then sputtered through many blows (perhaps in response to
a specific soil layer?); at times perhaps the ram did not even hit the impact block as
there is no rapid change of slope at the first dashed time marker. The ETR of
approximately 10% is very low.

Note also that during this time, the capacity changes very little, the compression
stresses are reduced, and the tension stress goes to zero. The tension is zero
because the compression input is small relative to the capacity so no net tension
is obtained. Look at DPW; the downward compression is larger in the first 2L/c
than the upward tension at 2L/c, so there is no net tension.

9 Play through the rest of these blows with DPW (use either PgUp or RA) and stop on
the first blow when the ETR increases again (BN 390/501 [SL26]; ETR = 0.22). Note
that now the stroke has increased again (6.45 ft; 1.966 m), and therefore so have the
stresses; compression stresses are acceptable, although there is still bending, and
the CSI is almost 50% larger than CSX! The tension stress TSX is now about equal
to the prestress. Looking at DPW, note that the upward tension at 2L/c is greater
than the downward compression at the time of the second dashed time marker so
there is a net tension, thus the computed value of TSX.

The "Damage" warning box is displayed for this blow. It suggests a BTA of 79,
which is relatively minor. Also note F1/F2 is very low (0.38), which suggests
bending. In this case there is no splice. Look at either DPFV or DPS. Since
bending is severe, and since BTA is relatively high (89), probably the pile is not
damaged. We generally look at a series of blows; if all blows indicate damage,
then damage is possible (or you need to look at the wave speed), but if most blows
do not indicate damage, then probably the pile is undamaged.

10 If you are still on BN 390/501 [SL26], play forward through the data. At BN 450/561
[SL29] notice the very high tension TSX which is about 25% higher than the
prestress value. Continue replay; stop at BN 850/961 [SL49]. You should have
noticed that the capacity gradually increased to 455 or 409 kips (2022 or 1818
kN), but is still less than the desired ultimate, but the pile is almost fully embedded
in the soil. Now what?

During this time, the F1/F2 Bending warning disappears at BN 550/661 [SL34]
(you might want to look at DPFV through the entire sequence of blows?) as the
alignment improves or the cushion becomes uniformly compressed. Actually, you
can replay the data several times. Do this replay (maybe replay more than once)
and observe the energy, observe the capacity and observe the stresses. Note
also that the PDA determined tension stress decreases as the capacity
increases; this is typical. Also note that as the capacity increases, the relative
"velocity increase" then decreases at 2L/c (second dotted time marker).
Concrete piles should be monitored in early easy driving when resistance is low
if you are to determine what the maximum tension stresses are during driving.

11 You are now on BN 850/961 [SL49] (or go there). If you adjust the wave speed
manually (left or right arrows), you will observe that the wave speed is now slightly
slower than originally used. In fact, I would suggest a value of WC 12743 ft/s (3954
m/s). This decrease was probably caused by a very minor tension cracking of the
pile due to the high tension stresses. The cracking is not serious. WC is allowed to

be less than WS, since the delay is due to cracking. WS (and hence EM) should not
be changed. (You might want to review the wave speed discussion earlier in points
4 and 5 for more practice). This wave speed adjustment makes only a very minor
difference in the present case, but in more severe cases the result could be
substantially adjusted; know why and when to do this. We would suggest you look
at the wave speed discussion in the manual.

12 If you are now on BN 850/961 [SL49], please note the current storage location (the
first SL value; the second SL value is the maximum number of blows stored in the
file. Look at DPFW. Note that the force and velocity (upper plot) are nearly the
same (proportional) at the first peak, but that the velocity decreases faster than
the force so the curves progressively separate; this is the result of the soil's
passive resistance on the shaft (remember the pile is now almost fully embedded).
In the lower plot, the wave up has a gradual increase {WU = (F-ZV)/2}, where Z is
the impedance EA/c.

Let us compare with the earlier blows; go to the first blow of this example data set.
Click the “Go to Beginning” button (||<) on the toolbar or type SL1. Note that the
force and velocity are very similar for the first 2L/c between the two half height dashed
time markers. The earliest blows were with the pile shaft almost completely above
ground and thus no shaft resistance is expected. Therefore the force and velocity are
similar, and the wave up is almost flat. Play through the data (click the RA button or
type RA) and observe the effect of the shaft resistance building up as the pile is driven.
Stop at BN 890/1001 [SL51].

13 The information in the upper left text box reveals data for BN 890/1001 [SL51] was
taken on 04-Sep-91 at 3:37 p.m. As the pile was fully embedded, the capacity was
less than desired, and it was time for the afternoon crew break so we recommended
to wait and restrike the pile. At 4:11 p.m. (34 minutes later), driving was resumed.
Hit PgUp to go to the next blow after the wait. Observe the differences in the force
and velocity curves, and in the wave up. Toggle back and forth, several times if
necessary between BN 890/1001 [SL51] and BN 895/1006 [SL52] (PgDn and PgUp)
to closely observe the changes. The force and velocity difference increases, and the
wave up increases, due to the increased shaft resistance. Note also that the velocity
reflection at 2L/c (second full height dotted time marker) is greatly reduced after the

This time effect is called "set-up", and is a capacity increase due to time after
driving due to pore pressure effects and perhaps lateral earth pressure effects
(during driving the pile may move laterally due to whipping or even Poisson ratio
effects which reduce the normal forces on the pile shaft, and during the wait the
earth pressures then seek to increase the normal and thus frictional force on the
shaft). Note that the capacity (RA2 552 kips; 2454 kN, and RMX 594 kips; 2643
kN) are above the desired ultimate. In fact, if we had waited overnight, and
performed a restrike the next day we probably would observe even more capacity
increase. In some soils such as clay where porosity is low the wait period must be

substantial (a week or more) to have the full resistance gain, while in other cases
such as coarse sands the gain comes quickly. The soil grain size (sand, silt, clay)
effects the pore pressure change stabilization. If capacity is an issue, you will
almost always want to test some piles in restrike to assess set-up effects; in
fact, most tests for capacity are on restrike tests. You should be aware that
some soils also have rapacity reduction with time. Read the capacity discussion
in the PDA manual Appendix paper HELPFUL HINTS...

The pile could be stopped here, except perhaps for other considerations such
as settlement or minimum scour depth (pile was for a bridge over a river), and
test piles are for establishing order lengths for piles to be driven later; so, to
check for weak layer below, the pile was driven further. Play through the data
(type RA). The compression stress is near the CL compression limit.
Tensions are very near zero. The capacity increases slightly, the stroke is
about 8 ft (2.44 m), and the energy transfer ratio ETR is 0.25 to 0.30 (slightly
above average).

14 You might at this time want to return to the first blow of the data set and run
through the entire example again. In the field (or later during a reprocess of the data
as we have just done), you might want to create a "result file" which contains the
nine computation results for each blow in ASCII form; this data can be processed by
the PDAPLOT program to print and plot a summary of your test with minimal effort.
To create this file, click the File menu and select “Save SQ File.” The default file
name will be Ex-1.001. Select a new name and directory, if desired. Clicking “OK”
and saving the SQ file will cause the program to run through the entire saved data
set to make a record of each output quantity, the blow number and the time of the

15 During actual data acquisition while pile driving, or during data replay, you can insert
comments into the SQ file using the PC command (follow the PC by a short message
and then press <Enter> to execute). This is very helpful for recording field
observations and is highly suggested for your normal practice.

16 You will also need to save the data from your field test to a permanent file. To save
the data, click the save toolbar button, or go to File then click Save or press
CTRL-S. Again, choose a storage location and a file name. Note the PDA default
extension is .W01. You might want to reduce the total number of blows to reduce
excess storage (use the W- Icon).

17 You probably want to save selected blows for CAPWAP. Go to the desired blow to
save (perhaps in this case either BN 895/1006 [SL51] just after the wait, or BN
1060/1171 [SL79] at the very end (you might want to save both blows). Go to File
then click “Save CAPWAP Record...” or click the CW Icon. The next dialog box
has parameters that must be filled. Answer the questions which for our pile and
this blow, the Length Penetration is 71 ft (21.6 m), the blow count is about 84 blows
per foot (275 blows per meter; as determined by the visual observer (the set per

blow will be calculated from the blow count; enter either blow count or set and the
complimentary value will be calculated), DFN is taken from the PDA program for the
blow, the Pile Circumference is 4.667 ft (1.6 m), and the Bottom Area is 256 in2 or
1.777 ft2 (0.16 m2). When all the information is entered, click “Save.”

18 Good practice also plots force and velocity measurements for selected blows of
interest (e.g. end of drive, beginning of restrike..). Select the force and time scales,
then click the “pl” button on the right toolbar. Examine the output, then go to
File and Print, or click the printer button on the toolbar or press CTRL-P (only do this
now IF you have the right output device attached). The Windows Print dialog box
will appear, and the output can be printed.

19 Eventually, you will need to police your hard disk and transfer to floppy disks data
to save. This can be done in the Windows Explorer or similar programs.

20 If you have not yet read the PDA manual, now would be a good time to do this. The
reference information contained in the Appendix will help with both theory, and
practical application (particularly the HELPFUL HINTS...). Good luck.

Example 2 (Clark 28)

Purpose: This data set demonstrates the driving of a H pile to hard rock.
Compression stresses at the top and bottom (toe, tip) may be critical.

Steel H pile (14 x 89) Area 26 in2 (167.7 cm2)

Hammer: MKT DE 70B
Soils: really doesn't matter for these discussions, but toe (tip) will be on hard limestone
Desired ultimate load: unknown

1 For a steel pile we are interested in compression stresses (Tension stresses are
generally of no concern unless a splice is deficient). For piles driven to rock, the
bottom compression stress may be of interest. We are almost always interested
in capacity and hammer energy. For a diesel hammer, the stroke is of interest.
Therefore select appropriate "Quantity" results might include:

CSX maximum compression stress at measured location averaged for section

CSI maximum compression stress of either strain transducer
CSB maximum compression stress computed for pile bottom (e.g. Q3CSB)
EMX maximum energy transferred to pile (could also use EFV for more resolution)
FMX maximum force in pile (CSX * Area)
STK ram stroke calculated by Saximeter formula for open end diesel hammers)
RA2 ultimate capacity estimate (independent of damping assumptions)
RMX ultimate capacity estimate based on JC damping factor)
DFN final displacement at end of the blow (set)

These quantities can be assigned by going to the View menu and selecting
“Quantity List,” or by typing Q1CSX, for example.

2 Go to the first blow of the file (actually it is BN 75) by clicking the button (||<) on
the left side toolbar or type SL1. Type RI to recall the input parameters (LE, AR, etc)
for this pile.

3 View the data with the DPFW option. Note that there is a tension wave up (WU).
The capacity for this blow is only about one third the FMX maximum force value.
The compression force at the pile bottom (CSB) is very small.

4 Replay the data; type RA or click the “RA” toolbar button with the green arrow
pointing right. Note that toward the end of the drive (last blow is BN 629 [SL77]),
the stroke increases (energy EMX increases as stroke increases), the wave up
becomes a compression reflection, the bottom compression stress (CSB) becomes
significantly larger than the top stress, and the capacity is now significantly larger
than FMX. You might want to try to CAPWAP a blow from the end of drive.

5 While the data was initially taken with RMX and damping constant JC of 0.7; this
damping constant was just an initial guess. Perhaps we should look at a variety

of Case Method capacity predictions. We can look at several methods at one
time. Set Q4RX4, which is the RMX method with a JC of 0.4 (see F1 Help
discussion). Also set Q5RX6 and Q6RX8, which are the RMX method with JC of 0.6
and 0.8 respectively. Looking at different damping factors allows you to assess the
sensitivity to your assumption. Set Q7RA2, which is the RA2 method independent
of any JC.

6 Display DPFR and note that the dashed RS curve is zero at the first time line. Since
it is unlikely that the pile has zero resistance, this capacity is unreasonable. You
might try a couple different values of JC and watch the effect on the RS curve. One
method of selecting JC is to produce a flat RS curve. Replay the data using several
JC values. Up until the blows when this pile hits rock, a JC of 0.4 makes a
reasonably flat curve after the initial first long dotted time marker.

7 For the early blows RX4 seems probably slightly too high. RX6 and RA2 are in
general agreement. For the later blows (after pile hits rock), the RA2 method seems

8 For all cases where capacity is marginal or an issue, we do strongly recommend a

CAPWAP analysis. In this case, since the data is from driving, a late blow (such as
any blow BN 625 [SL73] or higher, except BN 627 [SL75] which has low energy) is
the best blow to analyze by CAPWAP. CAPWAP will also compute the stresses in
the pile, so that the pile bottom stress can be further reviewed.

9 The capacity which is indicated is the capacity at the time of testing (end of driving).
It is likely, or at least possible, that the shaft resistance will increase after driving due
to set-up. Usually there would also be a restrike after some wait time (length of wait
depends on soil type) and a CAPWAP would be made on the restrike data as well.

Example 3 (Composite pile - Janesville)

Purpose: This example demonstrates testing a concrete filled pipe pile. The
question to answer is how to handle the data input of pile description. The
example also shows decreasing resistance as the soil is remolded during

Steel pipe 10.75 inch O.D. with a wall of 0.3 inches (27.305 cm O.D. with wall 0.762 cm)
Steel area 9.82 in2 (63.35 cm2); concrete area 85.767 in2 (553.3 cm2)
Pile was filled with concrete after driving but before PDA testing.

Hammer: Vulcan 08 (single acting air hammer)

Soils: fine grain soils with significant clay content.
Desired ultimate load: unknown

1 For all piles we are interested in compression stresses. We are almost always
interested in capacity and hammer energy. For a non-diesel hammer, the stroke
cannot be computed but blows per minute might be more interesting. Therefore
select appropriate "Quantity" results might include:

CSX maximum compression stress at measured location averaged for section

CSI maximum compression stress of either strain transducer
MEI maximum individual microstrain
EMX maximum energy transferred to pile (could also use EFV for more resolution)
FMX maximum force in pile (CSX * Area)
BPM blows per minute
RA2 ultimate capacity estimate (independent of damping assumptions)
RMX ultimate capacity estimate based on JC damping factor)
DFN final displacement at end of the blow (set)

2 Go to the first blow of the file (actually it is BN 2) [SL1]. You can get to the first
blow clicking the (||<) button or typing SL1. Type RI to recall the input parameters
(LE, AR, etc) for this pile.

3 The first thing we can do is to find the average density. Using the equation:

SP = {A1 * S1 + A2 * S2}/{A1 + A2}

where A1 is the steel area (9.82 in2), S1 is the steel density (0.492 k/ft3), A2 is the
concrete area (85.767 in2), and S2 is the concrete density (0.150 k/ft3). Please work
through the above computations to confirm the average density for this pile (0.195
k/ft3). Type the SP value.

4 Using the same basic formula, find the estimated average modulus (use steel
modulus 30,000 ksi; concrete modulus 5,000 ksi or 34,576 MPa). The answer is
7,970 ksi {51,056 MPa} (but you really should confirm this). For an assumed

modulus of 4,000 ksi {27,661 MPa} for concrete, the calculated average modulus is
7,025 ksi {48,579 MPa}. Enter either modulus value; perhaps your selection of
concrete modulus will be influenced by your knowledge of the concrete strength.
Note that upon entering SP and EM, that WS will automatically change to conform
to E = ρc2.

5 Something to think about: The force is the product of strain times modulus times
area. The steel EA is about 30% of the total EA (depending on your assumption of
concrete modulus). Thus, if the steel and concrete have identical strain, then the
steel carries about 30% of the total force; this is not negligible and cannot be
ignored. The strains in both steel and concrete must be identical or any
analysis is meaningless. Think about this. Thus it is important to consider both
steel and concrete and neither can be neglected. I would suggest you drill and tap
holes into the steel for the sensor attachment; the pipe should be completely filled
with concrete and the hammer should basically strike the concrete through a
plywood pile top cushion. This procedure is further described in the "Helpful Hints"
paper in the PDA manual Appendix. Read it if you have to do this type of test.

6 Why did we do step 3 and 4 above? Because the wave speed must match the
modulus and density. We need a good estimate of the wave speed for the test.
After you do a test, you will know the wave speed. Find the wave speed from the
measured reflection from the toe (perhaps 12,600 ft/sec for this case) and enter it.
Then enter EM to calculate the corresponding new modulus (6,677 ksi).

If you take data with one assumed wave speed, and then find out that wave
speed is wrong, changing to the correct WS will correct the original false

7 View the data with the DPFW option. The capacity is slightly lower than FMX, and
there is indication of the pile toe in the data for BN 2 [SL1] but the indication is weak.
The toe reflection becomes more clear with every additional blow as the tension
reflection from the toe becomes larger. This tension increases because the
resistance is decreasing as the soil is remolded and pore pressures increase during
the restrike. Such capacity reduction is typical for restrike tests when piles are in
clay. In fact, piles driven in clay are known to have reduced capacity and that time
is required to attain their final static capacity; this is why load tests are allowed only
after a sufficient wait period. The best correlations are achieved when the wait after
driving has been sufficiently long and an early blow is analyzed (choose an early
restrike blow for CAPWAP). While the capacity decreases during the test, it
should again recover with additional wait time and the long term service capacity
should again be most similar to the first early blows.

8 Observe how the wave up decreases during the first 2L/c from the first blows to the
last blows. Wave up reflects the shaft resistance. A linear ramp for wave up would
indicate a uniform resistance distribution along the shaft. You could also look at the

capacity (particularly RMX) as a function of number of blows during the restrike.
You could also look at the shaft resistance indicators (SFT and SFR) for each blow.

9 You will also observe that the final displacement DFN increases rapidly from blow
2 (0.14 inch per blow; 3.4 mm) till the end (0.47 inches per blow; 11.9 mm). It is
important in many cases to observe the blow count for several successive small
increments (blows per inch or blows per 25 mm) rather than averaged over a whole
foot (or over say 250 mm) if the restrike continues into a long redrive of the pile.

10 There is quite a bit of bending (CSI is significantly higher than CSX; you could look
at DPFV and you will have to change the force scale FS to a larger value) for all
these blows; the hammer pile alignment during restrike is more difficult to maintain
so this is quite often true. You might note that the maximum individual strain is
about 1400 microstrain (MEI), which is also a bit of a worry; can the pile survive?
Fortunately the average strain (MEX) is only about 1000 microstrain which most
materials can handle. Perhaps on other piles to be tested on site, the stroke could
be lowered or the amount of plywood pile top cushion increased.

11 Replay the data several times looking particularly at aspects of capacity decrease
as the restrike continues and their effect on the data; type RA or click the “RA”
toolbar button with the green arrow pointing right.

Example 4 (Stella 22)

Purpose: This data demonstrates a steel pile which suffers damage near the pile
toe (and at a splice).

Steel H pile (10 x 42) Area 12.4 in2 (80 cm2)

Hammer: Delmag D16-32
Soils: really doesn't matter for these discussions
Desired ultimate load: unknown

1 For a steel pile we are interested in compression stresses (Tension stresses are
generally of no concern unless a splice is deficient). For piles driven to rock, the
bottom compression stress may be of interest. We are almost always interested
in capacity and hammer energy. For a diesel hammer, the stroke is of interest.
Therefore select appropriate "Quantity" results might include:

CSX maximum compression stress at measured location averaged for section

CSI maximum compression stress of either strain transducer
CSB maximum compression stress computed for pile bottom
EMX maximum energy transferred to pile (could also use EFV for more resolution)
FMX maximum force in pile (CSX * Area)
STK ram stroke calculated by Saximeter formula for open end diesel hammers)
RA2 ultimate capacity estimate (independent of damping assumptions)
RMX ultimate capacity estimate based on JC damping factor)
DFN final displacement at end of the blow (set)

2 Go to the first blow of the file (actually it is BN 5). You can get to the first blow
clicking the (||<) button or typing SL1. Type RI to recall the input parameters (LE,
AR, etc) for this pile.

3 View the data with the DPFW option. Note that there is a tension wave up (WU).
This first blow indicates some possible "damage" near the pile top. Look at the next
two blows (PgUp twice). Now there are several warnings for BN 15 [SL3] indicating
suspect data quality. Why? The best way to further investigate is to look at the
data. Type DPF to separate the curves into forces on the top axis and velocities on
the lower axis. We see that the two force signals are practically identical while the
two velocity signals are quite different.

4 Usually the velocity signals are very similar when the signals are of good quality.
Thus we can see that one of the accelerometers must be bad. If you continue
through the data you see that by BN 75 [SL10] the data is even more dissimilar. The
V2 velocity bears resemblance to the force data, while the V1 data seems unusual.
Probably the A1 accelerometer was not tightened on the pile properly and has
slipped with each blow and become progressively more loose. Turn off the V1

signal by clicking the AF button on the toolbar, removing the check in front of
A1 and clicking “Apply”.

It is OK to use only one accelerometer. Proper use of the PDA however almost
always requires use of two strain transducers to properly account for bending. If
it had been determined that one of the strain transducers was bad, then you
should stop the test and tighten the strain transducer, or replace it, or replace the
cable system.

5 Return to the first blow, type DPFW, and play through the data using RA. You
should also expand the time scale (press F8 function key or type TS). You will note
that the capacity increases and then decreases as the pile goes through various
"layers" and the to reflection at 2L/c changes intensity. Stop at BN 116 [SL37]
(when you see the first damage warning).

6 The damage indicator is very near the bottom (the long extra line near the 2L/c time).
The indicated LTD (length to damage is 124.48 ft; 37.94 m; which is close to the
length below sensors LE of 127 ft (38.7 m). With continue blows, note that the
damage gets progressively worse (lower BTA values) and progresses up the pile
(shorter LTD). You might want to make BTA and LTD as two of your nine computed
quantities. Stop at BN 227 [SL64] (review this sequence by PgDn and PgUp). This
damage is damage occurring near the toe (no pile toe protection was used).

7 Continue on to BN 237 [SL66]. Note that now there are two damages indicated. The
larger damage (BTA 43 is at LTD 120 ft), while the other damage (BT2 91 is at LT2
85 ft) is less severe. This second location corresponds to a splice. Continue to the
end of this data set (type RA or click the “RA” toolbar button with the green arrow
pointing right).

8 Go to the data set continuation (Example 4B). At this point, you may have noticed
that the previous pile EX-4A was finished at 7:23 PM on December 21. It was then
redriven as EX-4B the following day at 1 PM. The bad accelerometer problem was
fixed, so you can turn both accelerometers on (and look for A12 in the upper right
of the graph display box). Play the data through to BN 103 [SL34]. We see that the
"Damage" lines appear at the splice location and just prior to the toe; the wave up
decreases at the second D line, which is prior to the right half height dashed 2L/c
marker, indicating a tension reflection from before the pile bottom and this is
possible only if the toe is damaged. For steel piles, we know the wave speed very
accurately, so we are certain the pile toe is damaged.

Example 5 (closed end pipe to till - RTA-517)

Purpose: This data set demonstrates the driving of a steel pipe pile from soft soils into
a dense glacial till.

Steel pipe pile (14 inch O.D. x 0.312 inch wall) Area 13.4 in2 (86.5 cm2)
Hammer: Vulcan 506 single acting air hammer.
Soils: soft silt over dense glacial till.
Desired ultimate load: unknown

1 For a steel pile we are interested in compression stresses (Tension stresses are
generally of no concern unless a splice is deficient). We are almost always
interested in capacity and hammer energy. Therefore select appropriate "Quantity"
results might include:

CSX maximum compression stress at measured location averaged for section

CSI maximum compression stress of either strain transducer
CSB maximum compression stress computed for pile bottom
EMX maximum energy transferred to pile (could also use EFV for more resolution)
FMX maximum force in pile (CSX * Area)
BPM blows per minute
RA2 ultimate capacity estimate (independent of damping assumptions)
RMX ultimate capacity estimate based on JC damping factor)
DFN final displacement at end of the blow (set)

2 Go to the first blow of the file (actually it is BN 1). You can get to the first blow
clicking the (||<) button or typing SL1. Type RI to recall the input parameters (LE,
AR, etc) for this pile.

3 View the data with the DPFW option. Note that there is a tension wave up (WU)
and that the capacity is much lower than the input force so the final set per blow is
large (low blow count).

4 Play through the data (RA). There is a late impact of the hammer assembly which
comes earlier and earlier as the data progresses; force and velocity increase
proportionately. Starting at BN 526 [SL67], the capacity begins a relatively rapid
increase as the pile toe encounters the till layer. Note that the displacement at the
end of the blow decreases rapidly and becomes near zero at the end of the data set
(refusal driving).

Example 6 (Seattle)

Purpose: This data set shows the driving and damage of a prestressed concrete pile
due to high driving stresses.

Concrete octagonal pile (24 inch O.D. with 15" void) Area 300 in2 (1935.5 cm2)
Hammer: Kobe K45 single acting diesel hammer.
Soils: silts and clays.
Desired ultimate load: unknown

This pile was partially installed when driving was interrupted by a time curfew. This data
is a next day redrive and includes setup which gradually reduce as driving continues.

1 This data was taken in the late 1970's with an analog tape recorder. The average
force and the average velocity were recorded; individual signals were not recorded.
Therefore, there is only a single channel of force and a single channel of velocity.
Click the AF button to change to F1 and A1 only. Actually this has already been
done, but you can confirm it with AF or look at the active sensors in the upper right
of the graphics area (A1 F1).

2 For a concrete pile we are interested in compression stresses and tension stresses.
We are almost always interested in capacity and hammer energy.

3 Determine the area for the pile (Press AR <Enter>). For an octagonal pile click the
“Octagon” tab. Enter the diameter (24 inch; 61 cm) and then the void diameter (15
inch; 38.1 cm).

4 Find the hammer (K45) using the Hammer Properties dialog box (Edit menu,
Hammer Properties). Look at stroke (STK), maximum transferred energy (EMX or
EFV), and energy transfer ratio (ETR = EMX/rating).

5 Look at driving stresses (CSX and TSX).

6 Replay the data (RA command). The first blows have the highest stroke and highest
transfer energy and then decrease slightly and remain stable for the duration of
testing. Compression stresses generally decrease slightly with the stroke. Note that
as driving continues that the capacity decreases, and the tension stresses
increase. Maximum tension stresses (BN 369) [SL68] up to 1.23 ksi (8.6 MPa) are
similar to the prestress level. The blow count was 240 blows/ft (800 bl/m); high
tensions at high blow count results from a large toe quake (confirm by CAPWAP).

7 The pile breaks suddenly during BN 408 [SL82], and is completely broken at BN
409 [SL83]. Use the LS command to find the damage location at 56 ft or 17.1
m(type LS56 or LS17.1). If you then look backward through the data (PgDn), you
will notice a slight local reduction in wave up (WU) for BN 407 [SL81] (and even
earlier blows back to about BN 369) [SL68] at what eventually becomes the break

location (56 ft; 17.1 m). The pile broke due to the large number of high stress blows;
there is indication in BN 408 [SL82] of two break points which could indicate the pile
had a sweep along the axis and flexure or bending was also involved. The
continued driving of a slightly damaged pile may result in a complete break.

8 You might want to replay the data set and watch the wave up curve (display DPFW)
between the two half height dashed time markers. This pile primarily has shaft
resistance and relatively little end bearing as would be expected for a pile in clay.
The wave up curve reduces from BN 2 to the end of the data set as expected due
to remolding and loss of set up during the restrike. The early BN 2 is probably most
indicative of the long term capacity.

9 The shaft friction is estimated from either SFT or SFR (display both
computations). Look in the PDA manual for a further description of the difference.
Note that SFR is dependent upon the damping constant JC.

10 While the data was initially taken with RMX and damping constant JC of 0.55; this
damping constant seems too low for a clay soil. Perhaps we should look at a
variety of Case Method capacity predictions. We can look at several methods
at one time. Set Q3RP9, which is the original method with a JC of 0.9 (see F1 Help
discussion). Also set Q4RX5 and Q5RX9, which are the RMX method with JC of 0.5
and 0.9 respectively. Set Q6RA2, which is the RA2 method independent of any JC.

11 It appears the RP9 method is too low. Display DPFR and note that the dashed RS
curve is zero at the first time line. Since it is unlikely that the pile has zero
resistance at refusal blow count, this capacity is unreasonable. You might try a
couple different values of JC and watch the effect on the RS curve. For piles with
significant shaft resistance and little end bearing, the method of selecting JC to
produce a flat RS curve does not really work very well.

12 For the early blows RX5 seems probably too high. RX5, RX9 and RA2 are in
general agreement for the later blows (after about BN 21 [SL10] the damping factor
seems not to make much difference).

13 SFT is higher than RX9 so is likely overpredicting the shaft resistance for some of
the very early blows. The shaft resistance SFR seems reasonable, or perhaps low,
but the exact shaft resistance is always better determined by CAPWAP.

14 For all cases where capacity is marginal or an issue, we strongly recommend a

CAPWAP analysis. In this case, it would be most appropriate for an early blow
before the set up capacity gains are lost due to continued driving. Probably BN 2 is
the best blow to analyze by CAPWAP.

Example 7 (Hopkinson bar)

Purpose: This data demonstrates wave propagation theory in a free pile. The total
pile length is 43.6 ft (13.3 m).

1 The "pile" or "target rod" is a small diameter steel pipe which is instrumented at four
locations. The top measurement location is 41.6 ft or 12.7 m (from sensors A3 and
F1 to the bottom). The next location is 39.6 ft or 12.1 m (A4 and F2) to the bottom.
The next location is 23.25 ft or 7.1 m (A1 and F3) to the bottom. The fourth location
is the bottom itself (A2 only; strain at the free end is always zero).

2 Use the “AF” button to select various pairs of data. Set the appropriate length
for each data set. View the F and V data; also view the wave down and wave up
(use F7 function key) to see the downward compression wave and the upward
tension wave.

3 You might also determine the length of the hammer (same type pipe but with a
shorter length; answer is 18 ft or 5.5 m). This is easiest done with BN 5 [SL5] where
the digitizing frequency was the highest (20,000 Hz).

4 You could think about the wave propagation effects, and amplitudes. For example,
the bottom velocity is twice the top impact velocity. Do you know why? What is the
impact velocity of the short hammer rod for BN 5 [SL5] where the velocity at impact
(VT1) of the target rod at A3 location was 3.2 ft/sec (0.97 m/s)? (Answer is 6.4 ft/sec
or 1.94 m/s, but do you know why?).

Answer: Assume the hammer rod has a down velocity of 2 due to free fall of the
hammer; this requires a zero stress state in the hammer. In order to
satisfy continuity, both the hammer and the target rod must have the
same velocity. Also the forces must satisfy equilibrium. To satisfy
both equilibrium and continuity, compression stress waves will
propagate up the hammer and down the target rod. the particle velocity
of these waves will be 1. The up velocity of 1 in the stress wave in the
hammer superimposes on the pre-impact down velocity of 2, resulting
in a net down velocity of 1, which then matches the down velocity of 1
in the target and this satisfies continuity. You can solve these
simultaneous equations if you want to try.

Example 8 (Korea)

Purpose: This pile exhibits some possible relaxation (capacity reduction) with time
after driving. This is a restrike test.

Steel pipe pile driven open end (16 inch or 409 mm O.D. with 0.35 inch or 9 mm wall)
Area 17.4 in2 (112 cm2)
Hammer: Drop hammer with unknown ram weight and variable drop height.
Soils: silts
Desired ultimate load: unknown

1 The first few blows inclusive of BN 5 [SL4] are of low energy and the set per blow
DFN is essentially zero. Thus these capacities are probably lower bound since
the full soil resistance was not activated. Ideally, in relaxation cases it is
desirable to have high energy blows right at the beginning, but this was unfortunately
not done.

2 For blow 6 (BN 6 [SL5]) the energy was about double and there is measurable set
per blow of 0.37 inch (9.4 mm). The capacity indicated is slightly over 500 kips (over
2250 kN).

3 For BN 7 [which has been eliminated from the data set], the energy was very small
and the BPM is very high. This is a "second hit" or "rebound blow". The hammer
strikes in BN 6 and then bounces and strikes again in BN 7. If the Max Blow Rate
had been set to say 100 BPM during data collection, this "false blow" would
not have been acquired. The Max Blow Rate can be set during PDA startup, or
using the MB command (e.g. MB100<Enter>).

4 Blows 8, 9 and 10 [SL6,7,8] show increasing energy and increasing capacity. BN

11 [SL9] has similar capacity to BN 10 but has a lower energy. Continued blows
show a gradual increase in capacity (to over 600 kips or over 2660 kN by the end of
the data set) with similar energy input. These capacity gains may result from
negative pore pressures at the pile toe caused by dilation of the soil during the short
time. Relaxation in shales may come from reduction of normal pressures after

It is possible that the pile is gaining capacity with depth and that this is not

5 Ideally to determine if this is a relaxation case, a higher energy should be

applied as early as possible, preferably by the second blow. Then if the
capacity increases very rapidly as the energy holds constant, then relaxation
is present. Thus we are looking for an early "high energy" blow. The service
capacity would be more like this early blow; the capacity gains with increasing
blows will likely be lost again after the test as soil pressures equalize.

Example 9 (Spiral welded pipe)

Purpose: This example includes two spiral welded pipes. Stresses are nonuniform
(requiring 4 strain measurements) on one pile and uniform on the other
(normal 2 strain acquisition would be fine).

Steel spiral weld pipe Area 65.2 in2 (420.6 cm2)

Hammer: IHC S70 double acting hydraulic
Soils: unknown
Desired ultimate load: unknown

For spiral weld pipes, you should try to drill as far away from the welds as
possible as this is a potential nonuniformity (in some piles the walls do not perfectly
align on either side of the weld).

1 For Example 9A (pile TP7), the strains are nonuniform. Look at DPS for any blow
other than BN 1, BN 10, or maybe BN 12 (for these 3 blows only [SL1, 10, 12] the
F3 and F4 are similar).

2 Look at the average force (DPF mode) from either "pair" of diagonally opposite
strains for most blows (say F1 and F2 only) and then look at the other "pair" (F3 and
F4) and you will note that they are quite different. Turn strain transducers off or
on with the “AF” button. Data for either single pair looks odd compared with the

3 Compare this with the same type of analysis for Example 9B (pile TP5) where the
stresses for all four are similar (DPS) and the average force is relatively unaffected
by strain pair selection, and appears smooth and reasonable compared with the

4 It is clear that for Example 9A, that if all four strain transducers are used the average
force is then similar to the average force for Example 9B (a sister pile). So we may
conclude that taking 4 strain readings helps improve data quality in difficult
cases such as spiral welded pipe piles. Four strain transducers also usually
is a good idea for large diameter drilled shafts.

Example 10 (problem pile 16C)

Purpose: This 14 blow restrike shows a spliced mechanically jointed pile with a gap
at the joint. There are also a data quality problem. Several piles on site
had suspected problems due to pile excavation and/or pile heave.

Regularly reinforced concrete (15.75 inch or 40 cm section), Area: 248 in2 or 1600 cm2
Hammer: BSP 7 ton hydraulic Soils: unknown; Desired ultimate load: unknown

1 First note there is a data quality problem. The "Proportionality box" [F/V warning] is
ON and this is serious. The "V1/V2 warning is also displayed. The best way is to
inspect the data first with the DPFV command. The two velocities are quite
different which is very unusual; generally they are nearly identical even in high
bending situations. This indicates that one or the other is probably in error.

2 We then see that the two strains for BN 1 are quite different due to bending
(coincidentally the strains are nearly the same at the first time marker so that the
F1/F2 box is OFF, although the stresses CSI and CSX are clearly different.

3 Returning to the velocity difference, it is now best to display DPV which shows the
individual velocities along with the average velocity and average force. From this
display, the V1 curve looks low while the V2 curve is reasonable. Turn off the
A1 accelerometer with “AF” button and the two warning boxes (Proportionality
and V1/V2) disappear. The data is now proportional and looks reasonable. The V1
accelerometer was not oriented axially along the pile axis. We can correct the
amplitude by entering the AF ICON, and then applying a “Reply Factor” of about 3.3
to ALL blows in this data set and activating the accelerometer again since it now
looks identical. This factor represents the inverse of the cosine of the alignment
error angle. Now we can proceed with the analysis of the pile.

4 With DPFW displayed, BN 1 shows a clear major damage at about 82 ft or 25 m (use

the LS command to locate). There is no observable pile toe reflection at 151.6
ft or 46.2 m. A small defect at another splice location (37.7 ft or 11.5 m) can also be
seen (perhaps change the force scale to half or less of the default value to enlarge
the data and see the minor defect more clearly.

5 BN 2 still shows the damage, but the size of the major damage reflection at 82 ft or
25 m is reduced. BN 3 and BN 4 are perhaps similar. BN 5 is a higher energy blow
and the pile toe reflection is now quite apparent. Continued blows show less of
the defect at 82 ft o 25 m, and by the last blow BN 14 this defect is almost absent.

6 There probably was a large initial gap which was sequentially closed during
subsequent blows. The gap could be caused during excavation. There is still cause
for concern if the pile is subject to tension or lateral loading. In compression, the
long term condition is unproven. Perhaps the pile should be rejected.

Example 11 (Clipdata)

Purpose: This example includes three pipe piles. Stresses are nonuniform and data
quality is a major problem.

Steel pipe Area 10.1 in2 (65 cm2)

Hammer: Delmag D30-32 single acting diesel
Soils: unknown
Desired ultimate load: unknown

1 The data for these piles has both the F1/F2 warning box and the "signal may be
clipped" box. For the display DPF or DPFW, while the initial force and velocity data
for BN 1 is reasonable, playing the data for the first 6 blows of Example 11A shows
quite different average force versus average velocity for each blow. For blow 4
[SL4], the Proportionality warning [F/V warning] is given.

2 Since it seems to be a force problem, type DPS to display the individual strains.
You may have to hit the HOME key also so that you see the average force and
velocity in the upper half of the graph and the individual F1 and F2 in the lower half
of the graph.

3 Clearly the F1 signal is very high and has an unusual flat spot at the peak. The PAK
has a wide balancing range, but the offset (OF) was probably too high and thus
when the dynamic data is superimposed the peak is then out of range for the A/D
convertor. Perhaps the sensor may have worked better if the OF was less than
5 v as suggested by the PDA manual.

4 The maximum stress for this sensor was very high: at least 55 ksi (380 MPa),
perhaps above the yield strength of the pipe. Playing through the data (RA), the
F1 and F2 data remain quite different due to bending, but the signal clipping
disappears for many blows in the middle of the data set. If any blows are to be
analyzed, then select one of these where the "Signal may be Clipped" box is absent.

5 The other piles (Example 11B and 11C; example 11C is particularly dramatic) from
the same job site show similar problems.

Example 12 (PSC Toe Damage)

Purpose: This data set shows the driving and damage of a prestressed concrete pile
due to driving obstructions or driving stresses above the concrete strength.
Also the wave speed needs adjustment.

Square concrete pile (18 inch or 45.7 cm O.D.) Area 324 in 2 (2090 cm2)
Hammer: unknown air hammer
Soils: unknown.
Desired ultimate load: unknown

1 For BN 1 [SL1], display DPFW. Note that the wave up fall is delayed slightly
compared with the second short dashed time marker. Use the right arrow to adjust
the wave speed to align the right short dashed time marker with the fall of the
wave up (the first dashed marker is aligned with the start of the wave down (or force
and velocity initial rise). We suggest the WC value (displayed just below the WS
value) should then be about 12,600 ft/s (3841 m/s).

2 Type WS12600 or WS3841. Note the automatic change in EM.

3 The original strain was multiplied by the original WS (13,253 or 4039) to produce a
proportional scale. Since the determined wave speed is different, the PDA-W
program automatically adjusts for the changed WS assumption in the strain data.
4 The capacity is very low for the first blows. The wave up is flat between the dashed
time markers indicating zero shaft resistance. The compression stresses CSX and
CSI are low. The tension stress TSX is acceptable considering that it is slightly less
than the prestress. Play the data (RA) through BN 289 [SL27].

5 Beginning at BN 289 [SL27] through BN 389 [SL37] note that the capacity increases
rapidly by a factor of 8 (from about 100 kips to about 800 kips). The tension wave
up at 2L/c changes into a compression reflection.

6 At BN 399 [SL38], the tension wave suddenly reappears and the capacity reduces
by a factor of 3 (to 270 kips). By BN 409 [SL39], it appears that the 2L/c should be
slightly less and by BN 419 [SL40] and then BN 429 [SL41] the wave up fall occurs
earlier still. Since WS or the wave speed c did not increase, then L or LE must have
decreased. Using the LS function we can determine a new shorter pile length of 42
ft or 12.8 m (type LS42 or LS12.8 for BN 429 [SL41] and observe that LS lines up
with the begin of the wave up fall).

7 Toward the end of the data, the length again becomes shorter still (LS of 39 ft or
11.9 m).

Example 13 (MBA Drilled shaft)

Purpose: To demonstrate data taken on a drilled shaft.

Shaft: 16 inch or 61 cm OD.

Hammer: Drop hammer with variable height.

1 A series of blows of increasing drop height was applied. First a relatively low drop
was given to check overall integrity of the shaft, alignment of the hammer, and stress
levels. Display DAS and look at all four strain signals (four strain signal data is
recommended for drilled shaft testing). Bending is minor in the F1/F2 axis, but
relatively large in F3/F4. The alignment in the F3/F4 axis was adjusted and the
difference was reduced for subsequent blows.

2 The Proportionality box is ON. It is ON primarily because of the plywood cushion

used to protect the shaft top causes a slow rise time. Display DPFW and the wave
up curve is smooth through the impact zone; this indicates good data so the
Proportionality warning can be in this case safely ignored.

3 Replay the data. The subsequent higher drops create ever higher stresses, larger
final displacements and higher capacities. We suggest you may want to CAPWAP
this data, perhaps using the Multiple Blow Analysis (MBA) or Radiation Damping

Example 14 (Fail2)

Purpose: This data set demonstrates testing SPT hammers. Two different
penetration depths are given (25 and 30 ft depths as included in the title).

SPT AW rod (Area 1.1 in2 (7.1 cm2)

Hammer: Safety hammer.

1 SPT testing may be performed with a PDA. For data quality reasons, we do
recommend special instrumented drill rod sections with foil gages glued directly
to the rod rather than strain transducers (contact PDI). If you want to test AW rods,
then you need an instrumented AW section. If you want to test NW rods, you need
an instrumented NW rod. The instrumented section must match the rod type you
plan to test, otherwise undesired reflections are generated which corrupt the data.

2 For this test there was only one active strain bridge glued to the instrumented
section. Turn off F3 and use only F4 with the AF button. The F1/F2 warning
disappears (ignore the "always use 2 strain" message). For uniform rods and good
data, the force and velocity should be more or less identical for the first 2L/c as in
the present case.

3 Expand the time scale with the TS command or the F8 function key. The force rises
slightly before the velocity. Shift the velocity negative (left) by one time increment
with the VT-1 command.

4 The rated energy of a safety hammer is 0.35 kip-ft (.475 kN-m). Enter this rating in
the Hammer Properties dialog box (Edit, then Hammer Properties). You can name
the hammer “Safety” if you desire, but Rated Energy = 0.35 kip-ft. The maximum
transferred energy (EMX) has only two significant digits. A third may be desired,
and can be added by going to the Edit Menu and selecting “Output Quantities.” Click
“Energy in the Output Class list, then select “EMX” from the OQ list. Finally, change
the precision using the arrows in the Precision box. Note that the ETR is about 50%
(relatively low efficiency for SPT).

5 You may also calculate EF2 which is the direct integral of force squared (no
correction factors applied as per ASTM D 4633; you may apply the factors to this
result if you so choose). The time ratio as per ASTM D 4633 is given by the quantity

6 As you will note EF2 is slightly higher than EFV, partly due to the connectors and
partly due to the rise time. For parallel wall pipe and uniform rod size throughout,
EFV and EF2 will be generally similar. For "upset" wall pipe, the force in the rod is
reduced and velocity increased relative to each other because of the reduction in rod
section causing a tension reflection. For upset walled pipe, the EF2 computation is
then in error theoretically and only the EFV method is reliable.

Example 15 (Data Spikes)

Purpose: Data quality is a major problem

Steel H pile Area 15.5 in2 (100 cm2)

Hammer: Ice 42S single acting diesel
Soils: unknown
Desired ultimate load: unknown

1 This data has numerous problems which cause numerous warnings. First it is noted
that there are 4 strain and 4 acceleration channels. Replay the data (RA) and note
that the data with all sensors active is poor.

2 Investigate the strain data first. Display DPFV and review the data. It appears there
may be at least one and maybe two usable strain signals. Try to find the good
signals by turning off the bad units with the AF button. Except for blows BN 1 and
BN 2, the F1 and F3 look reasonable and very similar to each other.

3 The velocity data (DPV) is very consistent for all blows.

4 Display DPFW or DPF and replay the data. Data is now acceptable for further use.

Example 16 (Till->Rock)

Purpose: This example includes a pipe pile driven to rock.

Steel closed end pipe (16 inch or 40.65 cm OD) Area 24.3 in2 (157.1 cm2)
Hammer: Kobe K22 single acting diesel
Soils: unknown
Desired ultimate load: unknown

1 For a steel pile driven to rock, the bottom compression stress may become important
so select CSB quantity for display.

2 Find the energy rating for this hammer (ER) using the Hammer Properties Table.

3 Note that the CSI is higher than CSX and the F1/F2 box is ON. Look at DPS.
Replay the data (RA button). The driving was interrupted at BN 544 [SL55] (stop at
12:10 and restart at 12:16) to adjust alignment. Note that after BN 546 [SL56], the
individual strains are then very similar. This is important to have good hammer pile
alignment for high stress conditions like driving to rock.

4 Return to the first blow of the data set. Observe that the capacity is less than FMX,
and that there is a tension reflection at 2L/c. As driving continues the FMX
increases as the stroke STK also increases in response to the capacity increase.
Eventually the capacity increases to above the maximum force.

5 Note that for blows on rock that the bottom stress CSB is higher than the top stress
CSX. If the bottom hits the rock unevenly (it is NOT likely that the pile bottom and
the rock surface are parallel), then there will be additional stress concentrations at
the bottom. It is important to be careful not to overdrive steel piles to rock. A high
yield strength of the steel should be specified whenever possible. Pipe piles should
be visually inspected prior to concreting.

Example 17 (Cl-Sa > Limestone)

Purpose: This example includes a pipe pile driven to rock. Some hammer variability.

Steel closed end pipe (10.75 inch or 27.3 cm OD by 0.25 inch or .64 cm wall)
Area 8.3 in2 (53.3 cm2)
Hammer: Vulcan 06 single acting air
Soils: weak clayey silty sand over limestone @ 72'
Desired ultimate load: unknown (GRL job 963079)

1 For a steel pile driven to rock, the bottom compression stress may become important
so select CSB quantity for display.

2 Find the energy rating for this hammer (ER).

3 For the first blow of the data set [SL1], observe that the capacity is much less than
FMX, and that there is a large tension reflection at 2L/c. Actually, from display DPW
we see that the wave up is almost a mirror image of the wave down 2L/c earlier (this
is common for piles with extremely small resistance). The set per blow is large.

4 Replay the data (RA). Note the rapid change between BN 73 [SL33] and BN 76
[SL36] as the pile encounters rock. The capacity increases dramatically, and the
tension reflection changes to a compression reflection. The set per blow decreases
from 1.4 inch per blow to essentially refusal (for some unknown reason the data
quality of BN 76 [SL36] and 77 [SL37] is suspect - display DPFW or DPF - so we
suggest making the comparison with BN 78 [SL38] where data quality is good).

5 The capacity RMX increases to above the maximum force FMX once the pile bottom
hits rock. Note that for blows on rock that the bottom stress CSB is higher than the
top stress CSX. If the bottom hits the rock unevenly (it is NOT likely that the pile
bottom and the rock surface are parallel), then there will be additional stress
concentrations at the bottom. It is important to be careful not to overdrive steel
piles to rock. A high yield strength of the steel should be specified whenever
possible. Pipe piles should be visually inspected prior to concreting.

6 It is interesting to note that prior to rock (BN 70 to 73 [SL30 to SL33]) the force at
impact (FT1) was about 195 kips (870 kN). After encountering rock (BN 78 [SL38]),
the impact force (FT1) was about the same while the force at 2L/c had increased
to 250 kips or 1110 kN. The highest pile top stress was due to the upward
compression wave superimposing on the incoming stress wave from the hammer
impact. The top stress was now about equal to the usual allowable yield strength
of mild steel (90% of F'y). It is very good that the hammer pile alignment is good and
that CSI and CSX are similar. Otherwise, the pile top could be damaged by high
local contact or bending stresses.

Pile driving on damaged pile tops should not be allowed as top damage acts as a
spring and absorbs useful energy and causes a false increase in the blow count.

7 It is interesting to observe the hammer performance for this pile. The hammer was
run at reduced pressure for the early blows and the BPM rate was lower (about 45
BPM), and stroke was also relatively low. The energy transfer ratio ETR is less than
50% for the early blows. The stress at impact is about 20 ksi (140 MPa).

8 After the beginning blows, the supplied air pressure and volume was increased (by
BN 47 [SL7]) and the BPM increased to about 55 blows per minute (BPM) and the
transferred energy ratio ETR increased to over 55% (EMX about 11 kip-ft or 15 kJ).

9 It is interesting that after the pile bottom hit rock (BN 78 [SL38] and later), the energy
transferred became quite variable from blow to blow. While the larger blows had
similar energy transfer, there were many smaller blows with considerably less
energy transferred. It appears there was great difficulty in keeping the hammer
running at a consistently high performance level; often in these cases, the hammer
tries to lift off the pile (called "racking") and the pressure/volume must be adjusted.

10 Air hammers are not constant stroke but depend upon both pressure and resistance.
The air pressure for raising the ram is delivered only for about the first half of the
stroke. The valve is then turned off and ram coasts upward against gravity, reaching
full stroke before dropping due to gravity. Before impact the pressure valve is
activated to lift the ram for the next stroke. The hammer cushion thickness and
valve timing must be correct for proper air hammer operation.

Example 18 (2 Breaks)

Purpose: This data set shows the driving and damage of a prestressed concrete pile

Square concrete pile (14 inch or 35.6 cm O.D.) Area 196 in2 (1265 cm2)
Hammer: ICE 640 double acting diesel hammer; 4 inch (10 cm) plywood pile cushion
Soils: soft overburden over a hard layer at about 50 to 55 ft depth
Desired ultimate load: unknown (GRL job 963086)

1 The pile begins with easy driving. The Warning Box for early blows with little friction,
and particularly driven by diesel hammers, can be ignored if the wave up between
the two short dashed time markers is basically flat (display DPFW). This is
particularly true when the BTA value is near or above 90% as is the case here.

If the dashed time markers are not apparent, they may be activated by the
changing the time marker color by clicking on the “Color” button.

2 Replay the data (RA). For blows up to BN 305 [SL38], the driving is normal. The
stresses CSX and TSX are low through about BN 281 [SL35], and then the capacity
and CSX stress both gradually increase to BN 337 [SL42]. Although the F1/F2
warning is not present, the CSI indicates considerable bending (display DPS -
separate the curves with the HOME key if necessary).

3 From BN 345 [SL43] through BN 369 [SL46], the BTA value rapidly decreases to
51%, indicating substantial damage. The damage is located about 60 ft or 18.3 m
below the sensors. You might use LS60 or LS18.3 to locate this damage. Perhaps
the pile broke due to excessive bending as the pile was driven into the hard layer
which held the pile bottom and prevented its further lateral movement.

4 From BN 473 [SL59] through BN 505 [SL63], a second damage occurs and the BTA
decreases to 47%. This damage is located at 50 ft or 15.4 m. Use LS50 or LS15.4
to locate.

5 Note that once a pile becomes severely damaged, the capacity calculation with
the original length is in error as the warning message clearly says. Using the
shorter length to damage may give a better temporary capacity estimate, but
we suggest that the pile be rejected as its long term serviceability cannot be

Example 19 (CEP with Splices)

Purpose: This example includes a pipe pile with several splices.

Steel closed end pipe (20 inch or 50.8 cm OD) Area 23.1 in2 (149.2 cm2)
Hammer: Conmaco 100E5 single acting air
Soils: unknown
Desired ultimate load: unknown

1 Pile is spliced several times. Up to BN 12/363 [SL13], the total pile length is 50 ft or
15.25 m. The length below sensors to the bottom is 47 ft or 14.3 m.

2 At BN 18/18 [SL14], the length is increased by 50 ft or another 15.25 m, making the

length below sensors 97 ft or 29.6 m. Use RI (or the F11 function key) to recall the
new length, and note the two hour delay in the date and time.

3 At BN 4/369 [SL30], the length is increased by 50 ft or another 15.25 m, making the

length below sensors 147 ft or 44.8 m. Use RI and note the delay in the date record.

4 At BN 16/761 [SL59] the length is increased by 50 ft or another 15.25 m, making the

length below sensors 197 ft or 60.0 m. Use RI.

Example 20 (Top Yield)

Purpose: This example pipe pile is driven with eccentric impacts causing pile top
damage. Shaft resistance and damping constant sensitivity are

Steel closed end pipe (12.75 inch or 32.4 cm OD by 0.25 inch or .64 cm wall)
Area 9.8 in2 (63.2 cm2)
Hammer: Delmag D12-32 single acting diesel
Soils: clays, pile fully embedded; sensors at ground surface
Desired ultimate load: unknown (GRL job 973006)

1 The F1/F2 and V1/V2 warning boxes are shown so display DPFV. There is a large
force difference. Hammer pile alignment is poor.

2 The CSI value is well above CSX and CSI is near or above the yield stress for most
pipe piles. For BN 7 [SL1], the F3 curve does not return to near zero. It appears
likely that the pile has suffered some yielding and thus a permanent strain. BN 8
[SL2] also has a permanent strain offset at the end of the blow.

3 From BN 9 [SL3] through BN 16 [SL10], notice how the two strain signals become
more similar with each blow. The yielding for these blows is probably right at the pile
top as the end strains return to near zero. The pile top is probably now bulging or
mushrooming at the top. Such pile top damage evens out the impact creating its
own uniform stress condition.

The velocities also become more similar. It appears that the initial velocity
difference was due to pile yielding and not to calibration or attachment problems.

4 It is wise to not attach too close to pile tops. Yielding the pile could cause
permanent damage to your strain transducers.

5 Later blows (BN 15 [SL9] and higher) suggest a small strain offset at the end of each
blow. Since the PDA automatically balances the strain between each blow (strain
is zero at begin of each blow as you can verify), the accumulation of many blows
each with a small permanent strain could eventually lead to a large total cumulative
strain and thus strain transducer damage. It is suggested that restrikes be limited
to a few blows to prevent damage. Restrikes of more than a few blows should be
undertaken only if hammer pile alignment is good (bending is confirmed small by
F1/F2) and there is some need for more blows.

6 You might want to replay the data set and watch the wave up curve (display DPFW).
This pile primarily has shaft resistance and relatively little end bearing as would be
expected for a pile in clay. The wave up curve reduces slightly from BN 8 to BN 31
[SL2 to SL25] as expected due to remolding and loss of set up during the restrike.
The early BN 8 [SL2] is probably most indicative of the long term capacity.

7 The shaft friction is estimated from either SFT or SFR (display both
computations). Look at the F1 Help under "Results-Capacity" for a further
description of the difference. Note that SFR is dependent upon the damping
constant JC.

8 While the data was initially taken with RMX and damping constant JC of 0.5; this
damping constant seems too low for a clay soil. Perhaps we should look at a
variety of Case Method capacity predictions. We can look at several methods
at one time. Set Q3RP9, which is the original method with a JC of 0.9 (see F1 Help
discussion). Also set Q4RX5 and Q5RX9, which are the RMX method with JC of 0.5
and 0.9 respectively. Set Q6RA2, which is the RA2 method independent of any JC.

9 It appears the RX5 method is too high. RP9, RX9 and RA2 are in general
agreement. SFT is higher than RX9 so likely is overpredicting the shaft resistance
while SFR seems reasonable.

10 For all cases where capacity is marginal or an issue, we do strongly recommend a

CAPWAP analysis. In this case, it would be most appropriate for an early blow
before the set up capacity gains are lost due to continued driving. However, we
should not analyze the really early blows with obvious yielding (BN 7 and 8 do not
return the force to zero at end of record). Perhaps BN 10 [SL4] is a satisfactory blow
to analyze by CAPWAP.

Example 21 (Setup)

Purpose: This data set shows the driving of a prestressed concrete pile, and
subsequent capacity increases due to setup demonstrated by restrikes
after 1 day and 75 days. Also the wave speed needs adjustment.

Square concrete pile (18 inch or 45.7 cm O.D.) Area 324 in2 (2090 cm2)
Hammer: Kobe K35
Soils: top 20' below reference in water; then in clayey sand soils
Desired ultimate load: unknown (GRL job 933058)

1 For BN 781 [SL1], display DPFW. Note that the wave up fall is delayed slightly
compared with the second short dashed time marker. Use the right arrow to adjust
the wave speed to align the right short dashed time marker with the fall of the
wave up (the first dashed marker is aligned with the start of the wave down (or force
and velocity initial rise). We suggest the WC value (displayed just below the WS
value) should then be about 11,500 ft/s (3506 m/s).

2 Type WS11500 or WS3506. Again, note the automatic recalculation of EM.

3 The original strain was multiplied by the original WS (12,200 or 3720) to produce a
proportional scale. Since the determined wave speed is different, The program
PDA-W has already now adjusted for the incorrect WS field assumption in the strain

4 The data from EX-21A is from only the end of driving (BN 781 [SL1] at penetration
85 ft below reference to BN 855 [SL16] at penetration 94 ft). The stroke STK is low,
and the blow count is very low (large set per blow - display DPFW). The Case
method capacity is also low, probably well below the design load. The capacity is
also quite sensitive to the selected damping constant (try Q4RX4, Q5RX6, Q6RX8
and RA2). Probably only CAPWAP can determine what the capacity is at the end
of driving.

5 Because of the low stroke, the compression driving stress CSX is also quite low
compared with the pile compressive strength. The CSI value is generally only
slightly higher than CSX indicating good alignment (at least in the axis of

6 Tension stress TSX is relatively high at 0.80 to 0.89 ksi (5.5 to 6.3 MPa). In fact,
although the prestress level is unknown, these stresses are likely to be
approximately the same as the prestress. This pile might have even seen higher
tension stresses in earlier blows. Notice that the width of the wave up curve for the
tension wave (or the velocity) at 2L/c is wider than the input wave (force or velocity
at impact), and the shape is different. This is a sign of minor tension cracking.

7 Data from EX-21B is from a restrike one day after driving (the reference elevation
changed due to tide). Note, you still need to use the new wave speed (11,500 or
3506) and new modulus.

8 There is a dramatic capacity increase (at least double) from the end of driving
to the one day restrike. The shaft resistance is also increased (compare SFT from
EOD to BOR; End of Drive to Begin of Restrike). The capacity and shaft resistance
decrease slightly during the test.

9 Data from EX-21C is from a restrike 75 days after driving. Note, you still need to use
the new wave speed (11,500 or 3506) and new modulus.

10 There is a further capacity increase (at least double) from the one day restrike
to the 75 day restrike. The shaft resistance is also increased (compare SFT). The
capacity and shaft resistance for the first few blows are relatively low. The later
blows appear to give more reasonable results for capacity. The shaft resistance is
reflected in the wave up curve. The higher the wave up between the two half height
dashed time markers, the higher the shaft resistance. The wave up is generally
proportional to the integral of the resistance so also reflects the location of the
resistance along the shaft. The PDA manual has further details.

11 The RA2 method appears to be unreasonably low. Since the velocity is negative
substantially before 2L/c, the upper soils may be unloading early and you could
consider the RSU capacity method (see manual for further details.) Of course,
CAPWAP should be performed to confirm any capacity evaluation.

Example 22 (Marine)

Purpose: This data set shows the driving of a prestressed concrete pile, and
subsequent capacity increases due to setup demonstrated by restrikes
after both a very brief stop and after 5 days. The pile is slightly nonuniform.

Octagonal concrete pile (20 inch or 50.8 cm O.D.) Top 8' has six 2" OD Dowel bar holes
Area at gages 312 in2 (2013 cm2). Solid section has area of 331 in2.
Hammer: Delmag D46-23
Pile: F'c of 8 ksi (55 MPa) and a prestress of 1.2 ksi (8.4 MPa)
Soils: top 20' below reference in water; then in clayey sand soils
Desired ultimate load: 600 kips (2675 kN) compression and 250 kips (1115 kN) uplift

1 For BN 1, display DPFW. The capacity is very low. The compression stresses CSX
and CSI are reasonable. The tension stress TSX is acceptable, although for BN 4
[SL2], the tension is about equal to the prestress, and the tension exceeds the
prestress for BN 25 [SL5] through BN 40 [SL6] (1.34 ksi; 9.4 MPa) and 1.41 ksi (9.9)
for BN 100 [SL10].

2 Replay the data (RA) and stop at BN 160 [SL14]. The stroke STK is low, and the
blow count is very low (large set per blow - display DPFW). The Case method
capacity is also low, probably well below the design load. The capacity is also quite
sensitive to the selected damping constant (try Q4RX4, Q5RX6, Q6RX8 and RA2).
Probably only CAPWAP can determine what the capacity is during driving.

3 Because of the low stroke, the compression driving stress CSX is also quite low
compared with the pile compressive strength. The CSI value is generally only
slightly higher than CSX indicating good alignment (at least in the axis of

4 There was a short stop time (about 7 minutes) between BN 160 [SL14] and BN 205
[SL15] (not every blow is recorded in this data set). Note the very strong capacity
increase; probably the RX4 method has too low a damping constant.

5 Observe that because of the large capacity increase, the tension is reduced to
practically zero. With continued blows note that as the stroke increases (probably
the fuel throttle setting was increased), the compression stresses increase
accordingly. CSX reaches 4.45 ksi or 30.8 MPa near end of driving. The allowable
compression stress is 85% of F'c minus the prestress or 5.6 ksi or 38.8 MPa.

6 Since capacity remains similar from BN 205 [SL15] to the end of data, the tension
increases (up to 1.47 ksi or 10.4 MPa) due to the larger input compression wave.

7 The capacity at the end of initial drive (BN 831 [SL32]) is only at best half the
desired ultimate capacity. We will stop and wait for capacity to increase, and
test later during restrike.

8 The force and velocity are non-proportional at the peak. This is due partly to the
diesel hammer, but also due to the nonuniform section 8 ft below the sensors.
Assume the pile is uniform during data collection and then model nonuniformity with
CAPWAP to find capacity.

9 There is a dramatic capacity increase (at least double) from the end of drive (July
5 EX-22a) to the five day restrike (July 10 EX-22b). You can compare both
simultaneously in PDA-W by using the Windows/”Tile Horizontal” feature of the
menu bar. The shaft resistance is increased (compare SFT).

The capacity and shaft resistance for the first blow are relatively low. Diesel
hammers generally give a low result for the first blow since the stroke is limited to
the trip stroke. The later blows have a higher stroke and give more reasonable
results for capacity (we would suggest BN 2 or BN 3 [SL2 or SL3] as the most
representative for service load conditions).

10 The RA2 method appears to be unreasonably low. Since the velocity is negative
before 2L/c, the upper soils may be unloading early and the RSU capacity method
may be applicable.

11 The shaft resistance is reflected in the wave up curve. The higher the wave up
between the two half height dashed time markers, the higher the shaft resistance.
The wave up is generally proportional to the integral of the shaft resistance so
also reflects the resistance location along the shaft. The PDA manual has further

12 It would possibly appear that both compressive and uplift capacity of this pile
exceeds the required resistances. Possibly further wait time would result in further
capacity increases. CAPWAP should of course be performed to evaluate

13 Display DPFW. Change the time scale to view the entire record with the “TS<“
button or by pressing F8. Go to BN 15 [SL10] of EX-22B and note that the final
displacement is very negative (DFN or look at displacement graph on the full time
scale). The PDA normally adjusts the velocity record to zero at the very end of the
data using the VA and VE parameters (see manual). Normally VE is 1024 (the last
point in the data). You can select a different ending point for the velocity by
changing the VE value (e.g. VE570 would make the 570th point the location of zero
velocity). Note also that there is a negative only solid marker showing the location
of the new VE and that VE will be in bold if VE becomes less than 500. Values
around either 800 or 580 give more reasonable displacement adjustments.

14 If the data is basically zero for the last half of the record, you could change the
sampling frequency to a higher rate. A higher rate at a fixed 1024 point sample
results in a shorter time. Do not automatically make this change without first
inspecting the data. (Note: you can only change sampling frequency for new data;
the sample frequency for existing data like this cannot be changed.)

Example 23 (Fill - Basalt)

Purpose: This example includes an H pipe pile driven to hard rock. There were also
errors made in entry of sensor calibrations. The hammer stroke was
reduced to control stresses.

H pile 12x63 Area 18.4 in2 (118.7 cm2) normal A36 steel.
Hammer: Bermingham B3505 single acting diesel
Soils: Fill and Saprolyte over Basalt bedrock.
During pile driving the following Blow counts 7,6,9,10,42,41 BPF then 10 BPI were
observed and recorded. After a short break another 10 blows were applied.
Desired ultimate load: unknown

1 For a steel pile driven to rock, the bottom compression stress may become important
so select CSB quantity for display.

2 Find the energy rating for this hammer (ER).

3 The first blow is the start up blow from the trip height drop. For diesels, the second
and later blows usually have higher stroke. Go to BN 2 [SL2].

4 For BN 2 [SL2], observe that the capacity is much less than FMX, and that there is
a large tension reflection at 2L/c. Actually, from display DPW we see that the wave
up is almost a mirror image of the wave down 2L/c earlier (this is common for piles
with extremely small resistance). The wave up is not flat between the two short
dashed time markers, suggesting data quality problems. The set per blow is large.

5 The V1/V2 warning box is displayed. If you replay the data (RA), you note that this
warning is present for each and every blow. (The bending/damage warning is also
present for most blows).

6 Display DPFV (separate the curves if necessary with HOME). Note that the two
strains are similar (they are mounted back to back in the center of the web), but the
velocities are very different. Further neither is in proportionality with the strain
(one is too high and the other too low), so neither velocity is correct as is and
simply turning one or the other OFF will not solve the problem. An error was made
in entering the calibrations for the two accelerometers.

7 Using the AF Icon, find the entered calibrations (A3 of 435 and A4 of 285). Actually
the calibrations should be reversed (A3 should be 285 and A4 should be 435). Thus
enter the correct calibrations and apply them to all blows of the data set.
Calibrations for strain and piezoelectric accelerometers are scaled directly. The
PDA handles calibrations for piezoresistive accelerometers as an inverse function.

8 Note that now the two velocity signals match first each other and also the force
proportionality and the wave up is much more flat between the two dashed time

markers as it should be for a pile with little penetration and little shaft resistance.
Display DPFW. This example demonstrates a good reason to try to use transducers
with similar calibrations so that if this transpose error is made then the effect is

9 The rise for this hammer is very rapid since it is a steel to steel impact (no hammer
cushion). The "Bending/Damage" warning is displayed near the first peak (D-line).
Expand the time scale (F8 function key or TS command) to its most expanded scale
(or type TS10) and note that the force rises before the velocity. We can shift the
velocity in time with the VT command (try VT-1). The rise is then almost identical
and the "warning" disappears.

10 Replay the data (RA). The first 32 blows are in soft material (see blow counts given
above with soil description. The pile then enters a dense soil above the bedrock for
the next 2 ft or 80 blows and the capacity increases to about 450 to 500 kips (2000
to 2300 kN).

11 Note that before BN 93 [SL49] the stroke was high and stress CSX was about 90%
of the steel yield strength of A36 material (and CSI was slightly above). As the pile
was approaching rock, the stroke was then reduced for subsequent blows to reduce
stresses. At about BN 113 [SL59], the CSX again reached 90% of yield and close
to 100% of yield strength by BN 117 [SL61], and exceeded the nominal yield for BN
121 [SL63], at which point the hammer was stopped.

12 The bottom stress CSB is about equal to the nominal yield strength for BN 121
[SL63]. The pile bottom is now on rock and the set per blow was 0.1 inch (10 BPI)
or 0.25 cm. Note that the wave up is now a compression reflection. Following a 4
minute stop to review results, a few more blows were applied with a higher stroke.
Stresses were very high for these blows. You must limit the number of blows at such
high stresses or damage may occur.

13 If the bottom hits the rock unevenly (it is NOT likely that the pile bottom and the rock
surface are parallel), then there will be additional stress concentrations at the
bottom. It is important to be careful not to overdrive steel piles to rock. A high
yield strength of the steel should be specified whenever possible. If great care
is not taken, then H piles should be equipped with pile points or rock shoes to
prevent bottom damage.

Example 24 (Mech. Splice)

Purpose: This data set shows the driving and damage of a prestressed concrete
pile. Setup during the splice time is observed. One accelerometer
seems to be better than the other.

Octagonal concrete pile (16.5 inch or 41.9 cm O.D.) Area 225 in2 (1452 cm2)
Bottom section 70 ft (21.3 m). Mechanical splice with top section 68 ft (20.7 m).
F'c 6 ksi (41.5 MPa); Prestress 1.03 ksi (7.1 MPa)
Hammer: Delmag D46-23 single acting diesel hammer
Soils: unknown
Desired ultimate load: unknown

1 The pile begins with easy driving for EX-24A (driving the bottom section). The
Warning Box for early blows with little friction, and particularly driven by diesel
hammers, can be ignored if the wave up between the two short dashed time markers
is basically flat (display DPFW). This is particularly true when the BTA value is near
or above 90% as is the case here.

If the dashed time markers are not apparent, they may be activated by the TM
(time marker) command.

2 There is a spike in the F1 data for BN 33 [SL5] (look at the full time scale). This is
probably due to a cable problem. If this were consistent, then the cable or
transducer or both should be replaced prior to further testing.

3 The capacity at the end of the first section is perhaps about 320 kips (1430 kN) for
BN 165 [SL18]. Note that tension comes from a downward tension after 2L/c and
not during first 2/Lc as TSN is always zero.

4 The pile was mechanically spliced. The splice took a little over one hour. Driving
resumed (EX-24B) and BN 181 [SL1] and 184 [SL2] indicate a capacity of 417 kips
(1860 kN) with RX6 so there is evidence of some setup on site.

5 Replay the EX-24B data (RA). Note that the F1/F2, the D1 and the Damage warning
boxes are apparent for some blows. The F1/F2 is not considered serious since the
compression stresses are relatively low and CSX and CSI are not vastly different.

6 The D1 box indicates that the A1 accelerometer may not be performing as well as
the A2 accelerometer for EX-24B. Display DPD and replay the data. The final
displacement for A1 is negative for low impact blows or hard driving, and for such
a well cushioned blow should not be.

Perhaps the accelerometer was not axially aligned during attachment

(usually the pile driving crew attaches the sensors to the pile). You could

turn A1 off using the AF Icon (or increase the magnitude by a replay factor
by one over the cosine of the angle alignment error).

7 Display DPFW. Go to the beginning of the EX-24B data and replay the data (RA).
The Damage indicator first appears at BN 306 [SL18]. The suggested damage
location is 57.2 ft (17.4 m). Type LS57.2 or LS17.4. If you change the force scale
to investigate (FS300 or FS2000) and see more detail in the wave up curve, you
may conclude that the damage is really at 62 ft or 18.9 m; type this LS value for
any blow from BN 600 to BN 692). Note that the total pile length is 138 ft (42.1 m).
The sensors are 4 ft (1.2 m) below the top. Therefore the splice is 64 ft (19.5 m)
below the sensors. As it happens, there was a visual crack in the pile about 2 ft (0.6
m) above the splice before it penetrated the ground location. This problem is not the
mechanical joint itself, but either the rebar detail or casting of the pile.

Note: prestressed piles are not prestressed at the ends due to needed
development length, so need regular dowel bars to transmit the tension across the
joint zone.

8 Up until BN 692 [SL44], the damage had slowly progressed and the BTA value had
gradually reduced to 81%. The crack then quickly deteriorated further and the BTA
value dropped to 78 for BN 696 [SL45] and 65 for BN 700 [SL46]. The PDA
engineer tried to alert the crew when the rapid deterioration began, but a few blows
transpired before the hammer could be shut off. The pile in this condition would be
probably rejected.

9 A "discussion" ensued where the contractor did not believe the PDA engineer and
insisted on continuing the driving of the pile. After a 30 minute wait, driving
continued (BN 702 to BN 713 [SL47 to SL58]), and the defect became even worse
(the BTA value shows a low of 50). Note that the final displacement was much
larger for these blows (DPFW or quantity DFN) than for the blows prior to damage.
The contractor reluctantly agreed the pile was indeed broken and driving was
terminated. Note the defect location has over last few blows moved up the pile to
about 58 ft or 17.7 m (instead of original splice location of 64 ft).

Example 25 (Multi-break)

Purpose: This data set shows the driving and damage of a prestressed concrete
pile. which breaks several times.

Square concrete pile (12 inch or 30.5 cm O.D.) Area 144 in2 (929 cm2)
Hammer: Delmag D22-23 single acting diesel hammer
Soils: unknown
Desired ultimate load: unknown

1 Replay through some of the early blows. The tension stress for BN 33 [SL7] is
relatively large. Display DPFW, or perhaps better still display DPW. The wave up
begins to fall at about the second dashed time marker indicating no problem with pile

2 Look at BN 43 through 73. Note that the wave up begins to fall earlier. It seems that
there is considerable bending in the pile (F1/F2 warning, CSI is much larger than
CSX; you might want to temporarily look at DPS before returning to DPW).

3 By BN 83, the sharp wave up reduction begins prior to the second dashed time
marker, indicating a reduced stiffness or damage. The damage is located at 77.8
ft or 23.7 m. Enter this value with LS. This is the first damage.

4 Continue driving and by BN 121 (or certainly BN 126 ) another early reduction is
beginning in wave up. Between BN 152 and BN 157 the damage is completed and
the length to this damage is only 70 ft or 21.4 m. Mark this with LS. This is the
second damage.

5 Almost immediately a new damage is located in BN 162 (complete break by BN

226). This location is at 63 ft or 19.2 m. Mark with LS. This is the third damage.

6 By BN 306, the pile has broken a fourth time at LS 59 ft or 17.9 m. By BN 366, the
pile has broken a fifth time at LS 50 ft or 15.4 m. By BN 501, the pile has broken a
sixth time at LS 38 ft or 11.5 m. By BN 676, the pile has broken a seventh time at
LS 29 ft or 8.9 m. By BN 726, the pile has broken another (8th) time at LS 21 ft or
6.4 m.

7 This pile has broken sequentially eight times. The capacity computation is
meaningless for broken piles. Compression stresses remained high throughout; it
would have been better to drive the pile at a reduced stroke to lower the stresses.
It is possible that the there were defects in concrete quality or in handling. The set
per blow remained large and was not representative of other piles on site.
Eventually the pile was abandoned.

Example 26 (Setup)

Purpose: This data set shows the driving of a prestressed concrete pile, and
subsequent capacity increases due to setup demonstrated by restrikes
after 2 days and 6 days.

Square concrete pile (18 inch or 45.7 cm O.D.) Area 324 in2 (2090 cm2)
F'c 6 ksi (41.5 MPa); prestress 0.86 ksi (6 MPa)
Hammer: Bermingham B3505 (12" plywood cushion)
Soils: silty clay and clayey sand
Desired ultimate load: 560 kips or 2500 kN (GRL job 962041, pile C3)
Blow counts: 37 blows/ft at EOD; 24 blows/inch for 2 day restrike; 47 blows per inch at
6 day restrike.

1 Replay the data for EX-26A. This is data for the initial installation of the pile. The
stroke (STK) and compression stresses are low for much of the driving. Tension
stresses are also low compared with the prestress.

2 The bending becomes rather large by about BN 627 [SL39] and continues to be
quite large till almost the end of the data (Compare CSI with CSX, and look at the
F1/F2 warning indicator). The early damage warnings are not considered serious
because driving is easy for a diesel hammer, bending is considered partly
responsible, and the BTA values are generally not very low and are not consistent
blow to blow.

3 Toward the end of EX-26A, the tension stresses increased to about 0.80 ksi or 5.7
MPa. Set per blow was relatively large.

4 Computed capacity was at most half of the desired ultimate capacity. There is little
shaft resistance since the wave up is nearly flat for the first 2L/c.

5 After a 2 day wait, the pile was tested during restrike (EX-26B). The estimated
capacity is quite dependent on the selected damping constant JC. RX8 yields about
580 kips (2600 kN) which is approximately equal to the required ultimate. Other
lower damping constants indicate higher capacity exceeding the requirements. The
RA2 method seems low for the first blows and then gradually increases to match the
RX8 method by the end of the data set. The shaft resistance is considerably higher
than at the end of driving as shown by the increasing wave up during the first 2L/c.

6 After a 6 day wait, the pile was tested again during restrike (EX-26C). The estimated
capacity is quite dependent on the selected damping constant JC. RX8 yields about
700 kips (3100 kN) which is well in excess of the required ultimate capacity. Other
lower damping constants indicate higher capacity exceeding the requirements. The
RA2 method seems low for the first blows and then gradually increases to match the
RX8 method by the end of the data set. The shaft resistance is considerably higher
than at the end of driving as shown by the increasing wave up during the first 2L/c.

7 Because the indicated capacity is more than required, consideration was given to
optimizing the pile length. As in any case where capacity is an issue, CAPWAP
analysis of a representative blow from the beginning of the last restrike would be
appropriate to more precisely evaluate capacity and determine shaft resistance
distribution. Knowing resistance distribution allowed the engineers on this project
to shorten other piles on site by 20 to 30 ft (6 to 9 m) each and thus create
substantial savings for this foundation. This was a static test pile.

8 The compression stresses for both restrikes were relatively large. The
recommended compression stress limit would be 85% of F'c minus the prestress or
4.24 ksi or about 29.1 MPa. The compression stresses exceed this limit by a small
amount. It would then be very important that the impact be applied as uniformly as
possible so that bending and local contact stresses are minimized so that possibility
of damage is reduced.

Example 27 (Preignition)

Purpose: This data set shows the driving of a prestressed concrete pile where the
diesel hammer preignition adversely affected the pile driving installation.
Some minor data adjustments improve the data.

Octagonal concrete pile (24 inch or 61 cm O.D.) Area 477 in2 (3077 cm2)
Driving stress limits: CSX 5.05 ksi (34.9 MPa); TSX 0.85 ksi (5.9 MPa)
Hammer: MKT DE 110 single acting diesel
Soils: 30' sand, brief fat clay layer over bearing in very dense fine to coarse sand
Test pile had been placed in 30" casing with 20' clear depth. Below the casing, a
slightly undersized hole (22") was augered to within 2' of minimum tip elevation and
stabilized by slurry. Pile was driven from 21 ft to 39 ft on Feb 4, and driven an extra
3.5 ft on Feb 5. The pile bottom finished about 1' above minimum tip elevation.
Desired ultimate load: 650 kips or 2900 kN (GRL job 972006, pile 2)
Blow counts: 20 blows for 3/8" (equivalent to 640 blows/ft) at end of drive (EOD)
20 blows/inch for begin of redrive (BOR)
20 blows for 1/2" (equivalent to 480 blows/ft) at end of redrive (EOR)

1 Calculate the area of the octagonal pile. Type AR then select the “Octagon” tab.
Enter 24 in the “Width” field and 0 (zero) in the “Diameter of Void” field.

2 Enter the List in the Hammer Properties dialog box (CTRL-H) and find MKT DE 110.
Select EMX and ETR as result quantities.

3 Replay the data (RA). Note that proportionality box is displayed for a few of the
early blows when resistance is low and driving is easy. During this time the
compression stresses are very low and the precompression of the gases is a large
percentage of the total force input. The wave up (DPFW or DPW) is a delayed
mirror image of the downward compression, so the warning can safely be ignored.
Tension stresses were at most 0.60 ksi (4.2 MPa) in easy driving so are
considerably less than the allowable limit.

4 Between BN 250 and BN 310 [SL9 to 11], the capacity and stroke STK increased
quickly and the tension reduced, probably as the pile bottom encountered the bottom
of the predrilled hole. The set per blow decreased to essentially refusal conditions.

5 Driving was then similar until BN 970 [SL33]. The stroke was generally about 7 to
7.5 ft (2.1 to 2.3 m). Energy transferred was generally low compared to the
manufacturer's rating. ETR was typically 12 to 16%. After BN 970 [SL33] the
hammer was stopped for about 30 minutes to replace the pile cushion.

6 Driving then continued with only a gradual and minor increase in indicated capacity.
Capacity at the end of drive was about 800 kips (3600 kN). Up to about BN 1558
[SL65], the stroke was about 7.8 to 8.1 ft (2.4 to 2.5 m) and the ETR was typically
14 to 17%.

7 After BN 1558 [SL65], the hammer was stopped for 12 minutes for hammer
lubrication. The stroke and energy then gradually declined to the end of driving at
BN 1735 [SL86]. Final stroke was about 7.5 ft and ETR was 14%. The hammer was
stopped at this point, even though the pile was still 4 ft above minimum tip elevation
since it was the end of the work day.

8 At the beginning of the following morning, pile driving continued on the same pile
and data is provided as EX-27B. It is readily apparent that the stroke, energy, and
driving stresses have greatly increased. The transfer ratio ETR is as high as 29%
for some blows.

9 This energy transfer ratio for the redrive (ETR 29% for BN 2 [SL1] of EX-27B) is
about double the transfer at the end of driving (ETR 15% for BN 1735 [SL86] of EX-
27A). The hammer at end of driving at the end of the day had already driven many
piles that day. Even though the air temperature was about freezing (32oF or 0oC),
the hammer was very hot since many blows were required to drive each pile. If fuel
is injected into the hammer and the flash point of the fuel is below the hammer
temperature, then the fuel will burn even prior to impact. When the fuel burns
prematurely, this is called "preignition". Preignition causes an increase in the gas
pressures inside the combustion chamber. The ram as it continues to fall is now
resisted by a higher than normal pressure and therefore requires more energy to
compress the gasses prior to impact. Thus some of the available kinetic energy is
used only to compress gas, and is therefore not available to transfer to the pile to do
useful work. Preignition at the end of drive caused a lower energy transfer. On
redrive, the hammer was still cold, so preignition did not occur and energy transfer
was normal.

10 To review the preignition, let us change the force scale to FS1000 for English or to
FS5000 for SI. Look at the force at the time of the first half height dashed time
marker. The force is much higher for the end of drive due to preignition than the
force from the begin of restrike when the hammer was cold and no preignition
occurred. Since the force is really the gas pressure times the cylinder area, the
pressure at end of drive is higher due to preignition as explained in point 9 above.
Different blows have different amounts of preignition (the preignition is particularly
strong for BN 1685 of EX-27A [SL79] for example).

11 During the restrike (EX-27B), the pile was driven another 3.5 ft (1 m). This driving
took almost 1000 blows and caused the hammer to heat up again; preignition may
be in the early stages at this time. The transfer ratio ETR reduced to 19 to 20%;
stresses and stroke also reduced.

12 Tension stress TSX increases quickly for the redrive EX-27B to 0.75 ksi (5.1 MPa)
by BN 18 [SL7]. Compression stress CSX reaches 3.71 ksi (25.6 MPa) for BN 46
[SL14]. For BN 46, the tension is now about identical to the limiting tension stress
(0.85 ksi, 5.9 MPa), even though the pile is at refusal blow counts. This is due to

most of the resistance being at the pile bottom and the presence of a large toe

13 Capacity at end of initial drive (EX-27A) was about 800 kips (3600 kN). Capacity at
the begin of redrive (EX-27B BN 2) increased to 900 kips (4000 kN). Both were at
refusal blow counts so the full resistance may not have been activated and thus both
capacity results may be lower bound estimates. In general, in refusal driving, as
input energy and impact force increase, capacity also increases. The capacity is
well above the required 650 kips (2900 kN). CAPWAP analysis is still recommended
to finalize the result and determine dynamic parameters such as damping and

14 Return to the beginning of the redrive EX-27B. Display DPFW and view the entire
time record (use TS or just type TS100). Observe that the final displacement at the
end of the blow is slightly negative. The data is being adjusted such that the velocity
at the end of the data is zero. The final velocity point defined as zero is VE. VE is
by default 1024 (the VE value is shown just below the data and above the 9
computed quantity results). The velocity is adjusted beginning at point VA until VE.
The default VA value is 200 (displayed near the VE display). The VA point 200
roughly corresponds to the time of the data trigger. Prior to the trigger, the PDA
does some small self balancing. After the trigger the self balancing is turned off. It
is after the trigger point that most of the adjustment is therefore needed, and thus
after point VA200. If we instead try VA of 100 or 50, then the final displacement is
perhaps more reasonable for this data case. Type VA100 or VA50. Replay the data
from both driving and redriving with this new adjustment.

While the final displacement is more reasonable with the minor VA adjustment, the
interesting results of stresses, energy and pile capacity essentially do not change.

Example 28 (Relaxation)

Purpose: This example includes an H pile driven to weathered shale. There was
significant relaxation (capacity loss) with wait after installation.

H pile 12x74 Area 21.8 in2 (140.6 cm2) normal A36 steel.
Pile was driven with tip protection, but this was probably not needed in soft rock.
Hammer: MKT DA35C single acting diesel
Soils: Clayey silt over weathered shale - claystone
During driving the pile blow counts reached 21 BPI at EOD. Following a 5 day wait,
redrive began with 10 BPI and gradually increased to 21 BPI at end of redrive.
Desired ultimate load: 300 kips (1340 kN).

1 Initial pile driving (EX-28A) proceeds from easy driving to hard driving. This gradual
transition occurs as the pile penetrated the weathered weak rock. Capacity at end
of driving reaches over 600 kips (2700 kN) at refusal blow counts of 21 blows per
inch (21 blows per 25 mm). Pile stresses CSX and CSB are modest. The toe
reflection was compressive.

2 After a 5 day wait. The same pile was tested during a restrike (EX-28B). Please
note that the length input during this test was incorrect. The length below sensors
(LE) should still be 27.5 ft (8.4 m), as can clearly be noted in the toe reflection.

3 The first restrike blows unfortunately had low energy. The capacity for the first two
blows was approximately equal to the desired ultimate capacity. The first high
energy blow (BN 3 [SL3]) had a capacity of only about 400 kips (1780 kN). It is not
clear whether the capacity is 400 kips (1780 kN) or less; ideally the very first blow
would have had high energy. It is very important to have early high energy blows
when relaxation is suspected. There is a definite tensile toe reflection for the early
blows. The blow count reduced to only 10 blows per inch (10 blows per 25 mm) at
the begin of restrike, thus confirming the capacity reduction from end of driving.

4 As restrike continued, the capacity gradually increased as did the blow count. At the
end of the restrike, the conditions were comparable to those recorded at the end of
drive. Although capacity has increased again temporarily, the capacity will reduce
again with additional wait time.

5 It is important to realize that this was only a 5 day restrike test. It is possible that
additional capacity loss may occur with additional wait time. To achieve adequate
long term service capacity, the pile must therefore be driven to significantly higher
capacity at the end of drive so that after losses the residual capacity is satisfactory.
When driving in soft rock, some require driving to refusal, and then even an
additional 200 or more blows (provided that stresses are acceptable). This
overdriving may result in higher end of drive capacity, so that when relaxation losses
occur, the final residual capacity is adequate for project needs.

Example 29 (Wavespeed Change, Setup, Cracking)

Purpose: This example involves a 40 cm square concrete pile. It demonstrates that

• a lower wavespeed than assumed is required (WS11,000 ft/sec); this improves

the proportionality as well as the 2L/c time.

• there is loss of setup in shaft resistance on this 700+ blow redrive (pile driven
from 14.8 m to 33 m)

• there is some minor tension cracking (the WU at 2L/c initially reflects the WD
as in SL20, and then later is much “wider” as in SL40.

• The tension crack eventually shows up as a minor damage (starting SL31).

Example 30 (H-pile splice failure)

Purpose: This example includes an H pile which has a splice failure. Pile is
14x89 H section. Hammer is Delmag D30-32. Pile has two 55 ft sections
so length from sensors to splice is LS52 ft. Soil is land fill over sand.

Example 31 (SPT example)

Purpose: This example includes three SPT files. Shows variability between
hammer types.

• Ex-31A has a normal safety hammer on an AW rod at depth of 22.5 ft. The first
blow has an open joint, but all subsequent blows are OK. ETR typically 65%

• Ex-31B is the same safety hammer but at 37.5 ft depth. The joint starts closed
but is often open (loose) causing an early zero crossing if using EF2. ETR
typically 65%

• Ex-31C is a donut hammer on a spooling winch at the same site and at depth
20 ft. ETR typically only 40%

Example 32 (Loose Strain Sensor; High Stresses)

Purpose: This example includes a small pipe pile.

• One strain sensor was not firmly attached so after the first blow it is useless.
Fortunately turning it off seems to give reasonable force data (perhaps bending
is minimal). In the usual case it is never allowed to use one strain sensor.

• After turning the bad sensor off (look at DPFV and then turn it off with the AF
Icon), the compressive stresses are very high (about 60 ksi; 420 Mpa);
fortunately this is reject oil pipe with very high yield strength.

• The data for SL2 is still not right due to final force offset. Toward the end of the
data set one velocity is not as reliable and so you might also prefer to use one
accelerometer (AF Icon).

Example 33 (Timber Pile)

Purpose: This example includes a timber pile with very high wave speed.

• Ex-33A is a free pile wavespeed created by using a more sensitive PIT

accelerometer on a pile laying horizontally on the ground and struck with a
sledgehammer. Using the correct pile length (21.33 m; 70 ft), and adjusting the
WC (by left and right arrows), we find that the true wavespeed in this timber pile
(from South America rain forests) is about 5200 m/sec (17060 ft/sec). This is
a faster wavespeed than steel.

• EX-33B is data from real pile driving on the same project. The original data with
the original low wavespeed has poor proportionality. If the high wavespeed is
used, the data looks reasonable. Note WS from rise to rise is more difficult to
evaluate than for peak to peak method (OK to use peak method here due to low
resistance and also timber piles do not have cracks).

Example 34 (Spliced Concrete Pile)

Purpose: This example includes a 40 cm square concrete pile with no-tension splices.
Use the LS19.4 and LS39.4 to locate the splices. The “damages” roughly
correspond to these locations and indicate a “gap” has formed between the
pile sections. Top section is 19.4 m, and setting LS19.4m and then adjusting
WS suggests a value of WS 3,800 m/s. Making this change gives good
proportionality. The gap seems relatively large at the bottom splice. This
gap closes with more applied blows.

Example 35 (Pipe - Toe Damage)

Purpose: This example includes a steel pipe.

• There are some data quality issues as evidenced by some data spikes for the
later blows in the F1 channel (look at DPFV to confirm). This is probably
caused by a failing sensor or failing cable (probably the cable since it occurs
late in the blow rather than near the peak accelerations).

• The pile toe is damaged during the driving. Compare BN175 [SL30] before
damage with later blows. Already by BN691 [SL73] damage clearly exists, and
by the end of driving BN820 [SL109] it is even more clear in that the tensile
reflection begins too early. The PDA shows damage (BTA71 for SL109).

Example 36 (Spliced Pile with Damage at the Splices)

Purpose: This example includes a spliced concrete pile. Each section is 12 m long and
sensors are attached one meter below the pile top.

• EX-36A gives the driving of the first section (no real problems except perhaps some
tension cracking in the section during easy driving. Later blows appear OK.

• EX-36B has the second section driving. The pile shows damage at the splice (LS11)
at BN213 [SN22], and has a major break at BN345 [SL36]. The pile driving continued
anyway to BN466 [SL49]. Pile splice design seems defective.

• Even though the pile was damaged, it was spliced again. EX-36C contains the third
section driving (the LE should be 35 m, not 37 m as the file suggests; change LE35).
The first splice is then at 23 m, and the second splice is at 11 m below the pile sensors.
The reflection from the 23 m splice is seen immediately, and the reflection from the
upper 11 m splice becomes apparent by BN 70 [SL4].

• Even though the pile was damaged, it was spliced again. EX-36D contains the fourth
section driving (LE47). The first splice is at 35 m, the second splice is then at 23 m,
and the third splice is at 11 m below the pile sensors. The reflection from the 23 m and
35 m splices are seen immediately, and the reflection from the upper 11 m splice
becomes apparent almost immediately and grows to a substantial reflection by BN 502
[SL27]. It is surprising that a long pile (47 m; 154 ft) has such large set per blow at end
of drive (20 mm/blow; 0.8 inch/blow). So, is the pile still really intact for all four
sections, or is the effective pile length less? While there appears to be some evidence
that there still are 4 sections in line, it is not conclusive. Further, if this pile is to resist
tension, then the obvious defective splices will not allow tension from the lower

• EX-36R contains the 5 day restrike. There is significant increase in shaft resistance,
but the defective splices are still observable.

Example 37 (Data Problems With 4 Strains on Drilled Shaft)

Purpose: This example shows several problems including poor attachment of sensors
for first blow. Problems for later blows could also relate to insufficiently
tightened sensors.

Example 38 (Yielding Pipe With Quick Setup)

Purpose: This example shows driving a pile at low resistance. Following a short wait
the capacity substantially increases. Unfortunately the alignment is not good
and pile top yielding is then present, clipping some data and making
selection of the CW blow more difficult.

Example 39 (Yielding Pipe With Quick Setup)

Purpose: This example shows driving a nonuniform pipe pile. Can be used to
demonstrate nonuniform pile modeling for CAPWAP.

Example 40 (Relaxation in Fine Sands and Silt of Pipe Pile)

Purpose: This example shows comparison of high energy end of drive data at 5 blows
per inch, with lower energy restrike (one day later) with 6 blows per inch.
(Shows variability of hydraulic hammers). Thus even though blow count went
up, the capacity went down due to relaxation in these fine saturated soils.
Continued re-driving shows capacity builds back up to EOD case as negative
pore pressures are made.

Example 41 ()

Purpose: This example shows 12.75 inch x 0.375 pipe piles driven with a diesel hammer
into fine grained soils. Many different features are observed in several different
data sets.
41a Test shows substantial bending for the first blows. The bending is large for the first
59 blows. The stresses are therefore large (CSI) compared to normal pipe pile
material strength, and the first blow shows material yielding (and maybe also for
BN2). Pile driving is briefly stopped twice to adjust and improve the alignment and
thus reduce the bending stresses. Capacity may have increased due to setup
during first interruption of about 3 minutes (but no noticed difference for second

41b Test shows a loose strain transducer attachment through BN 130. After a seven
minute stop driving continued (with bending being large). A 16 minute stop was
made after BN 270 to adjust alignment.

41c Test shows initial calibration pulse with very large cyclic noise. After correcting main
problem (improve power supply, or replace cables), another calibration pulse
shows great improvement. However, A3 still has minor noise on the signal and
could be turned off to further improve the signal quality. Stop after BN 101 to
improve alignment (which works for awhile but gradually gets worse again).

41d Test shows yielding of the pile on this brief restrike. Stresses from strains are very
high. The energy transfer (only meaningful when pile is not yielding; e.g. BN 8 and
later) is very large compared to the hammer rating.

41e Test shows capacity is large for first blow in data set (BN3) and decreases rapidly
with continued blows. Shows importance of saving every blow initially. Capacity
then increases at end of this redrive from 12 to 17.5 m. Bending is relatively small
for this pile, showing good alignment. Energy transfer is better than average,
particularly for the first blow which may have residual extra fuel and thus
generating a high stroke.

Example 42 (collapsing pipe pile)

Purpose: This example shows a pipe pile with collapsing pile wall (confirmed by physical
inspection). The pile has a conical point (previous flat plat bottomed piles had
collapsed so this was attempting to avoid the problem). There is a data quality
problem starting after BN 262 and continuing through BN 1462 (turn A4 off for
these blows by two step process using selective ranges form AF Icon. E.g BN
262 to end, followed by BN 1462 to end). The pile integrity seems acceptable
through about BN 1873. At BN 2023 there is clear damage indicated at 85 ft
below sensors, and that gradually rises to a location 62 ft below sensors
(confirmed by visual inspection and dropping a tape to obstruction). It is
interesting that the pile toe can still be seen in the record.

Example 43 (Changing wavespeed WC)

Purpose: This example shows a wavespeed change during the driving of a prestressed
pile. Early blows have relatively little shaft resistance (pile mostly in air).
Continued driving shows significant increase in shaft resistance. However
the main item of interest is the pile wavespeed. The wavespeed WS is
clearly about 13,200 ft/s (4,025 m/s) for early blows. This also produces a
reasonable proportionality. However, as driving progresses, the effective
overall wavespeed (WC) slows down by more than 10% as minor cracking
develops and finishes at 11,667 ft/s (3,556 m/s). The WS is a material property
and should not be changed (and thus is higher than WC). To obtain a variable
wavespeed for different blows as driving progresses, first start at the first blow
(SL1) and press OPTIONS/”Calculated Wavespeed” and then select “blow by
blow auto edit” and an {auto} indicator will appear after the WC indicating you
are now in an automatic mode. In this mode, you can adjust the wavespeed for
any blow by using the left and right arrow keys to adjust the 2L/c time. When
you then press PgUp to go to the next blow, the previous wavespeed will
automatically be applied to the next blow. (If you go back with PgDn, the
wavespeed is unchanged). The first slower wavespeed is needed about BN
409 (press the right arrow once on this blow) to get WC 12,833 ft/s (3,912 m/s).
At BN 509, slow the wavespeed WC again with a right arrow to 12,623 ft/s
(3,848 m/s). Note high tension stresses for the next 1000 blows. Following an
11 minute stop at BN 1569, the WC value changes rather dramatically to
12,222 ft/s (3,725 m/s) for BN 1579 and there is evidence of setup during this
brief stop, and tension stresses decline to practically zero. The Wc further
slows by BN 1849, and again at BN 1979, and to the final 11,667 ft/s (3,556
m/s) at BN 1949. When this is completed to avoid further changes, the file
must be changed to the “blow by blow edit” mode (and an {edit} will then follow
the WC). In the “edit” mode changes by left or right arrow keys affect only that
single blow so it is safer to leave such changing wavespeed files in this mode.
If you CAPWAP a late blow with WC < WS, you will need to use a slower
overall wavespeed value (accessed by the PM pile model Icon).




For Field Testing and Data Interpretation



Garland Likins

Pile Dynamics, Inc.

Cleveland, Ohio 44128 USA

Tel 216-831-6131
Fax 216-831-0916

PDA-W Manual

©20099 Pile Dynamics, Inc.

Table of Contents

INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

THE DIGITAL PILE DRIVING ANALYZER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

TRANSDUCER PERFORMANCE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

TRANSDUCER CALIBRATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

TRANSDUCER ATTACHMENT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

DATA QUALITY CHECKING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

WAVE SPEED DETERMINATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

MODULUS OF ELASTICITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

DATA RECORDING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

DATA INTERPRETATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

STRESSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

PILE DAMAGE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

HAMMER PERFORMANCE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

CAPACITY DETERMINATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

STATIC TESTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

SPECIAL APPLICATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

RECENT PDA PROGRAM FEATURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

TESTING DRILLED SHAFTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

SUGGESTIONS FOR A SUCCESSFUL TEST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

TECHNICAL NOTES: TEFLON PILE TRAINING AID . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26


Dynamic pile testing has become truly a routine procedure no longer confined to researchers.
It has been proven cost effective and reliable, and as a result many contractors, consultants
and government agencies have acquired the necessary equipment to service their own needs;
others prefer to obtain the testing as a service from those offering such testing services.

Microprocessors have revolutionized our everyday lives from hand calculators to microwave
ovens. Computers used to be large, unfriendly, and relatively inaccessible. However, since
the advent of personal computers, computers have steadily become smaller, faster, and more
powerful such that they are now virtually everywhere and used in all facets of instruction and
engineering. They have been of great benefit to the civil engineer in solving complex analysis
problems. For example, wave equation analysis of pile driving was perhaps the first non-
military application of computers to the civil engineering field. The PC has made the PDA a
user friendly device. The new Windows PDA-W program is a further step in this direction. The
touchscreen operation of the PAX makes operation simple and intuitive.


The first analog dynamic pile test hardware was built at Case Western Reserve University in
about 1968 under the supervision of Dr. G.G. Goble. In 1972, Pile Dynamics (PDI) introduced
its first commercial device (Model DA, then called the "Pile Capacity Computer"). The first use
of the name "Pile Driving Analyzer" (often simplified to PDA) came in 1974 with introduction of
PDI’s Model EA. In 1981, PDI delivered its first fully digital PDA (the “blue box” Model GA)
based on the Motorola 68000 microprocessor (delivered before the Apple MacIntosh based on
the same microprocessor became available). The GA was updated to GB, then GC, then GCXS
in about 2 year intervals. With emerging PC technology, PDI developed a PC version of our
PDA in 1989 (called the "GCPC") and subsequently the (PAK) in 1991.

The PAK has many advantages over the previous "blue box" model (i.e., Model GCXS with 90
blow memory storage). The PAK is fully self-contained and currently has a PC processor
(operating on a Windows XP platform). (Older PAKs can be upgraded by contacting PDI.) The
PAK during operation displays the data curves on the VGA graphics screen and/or optional
external VGA monitor. Various PAK models have standard serial RS232, parallel, USB and/or
Ethernet ports. The PAK has a standard built-in membrane keyboard with function keys labeled
with the PAK functions and an extra keypad for some frequently needed commands. This is
more rugged and dust resistant than the standard detachable keyboard. The PAK accepts a
standard detachable keyboard for convenience in office use.

The PAK is light weight and of small size (140 x 280 x 380 mm) which is only 30 percent of the
"blue PDA" volume! The PAK looks like a "briefcase" and the soft carry case allows us to hand
carry the PAK through airports (PAK weighs only about 9kg - 20 pounds) and check as baggage
for air flights only one transit case with cables and transducers, etc. The PAK operates on
either 90 to 250 volt AC power in the office, or a 12 volt DC car battery in the field.

The PAK signal conditioning now contains channels for 4 strain transducers, 2 piezoelectric
accelerometers (for decades our standard accelerometer), 2 piezoresistive type
accelerometers, and four integrators (to get velocity). Four channels each of strain and velocity
allow us now to simultaneously acquire four strain and four velocity signals where extra backup

on the pile is requested (such as offshore oil projects), or for drilled shafts (has improved data
quality), or if two locations on the same pile need be tested (e.g. above and below a splice).

In 1999, PDI introduced the PDA model “PAL”. This unit is 8 x 7 x 4 inches (210 x 180 x 110
mm) in size or approximately 25% of the PAK volume, and weighs only about 7 pounds (about
3 kg); most of the weight is batteries which allow the PAL to operate up to 8 hours. The PAL-R
(“R” represents “Remote”) allows the PDA to be operated on site by an engineer in his office
using cell phone communication. Of course, the PAL-R can also be operated on site by an
engineer on site. The PAL-R communicates either on site or remotely to the PDA-W program
operating on any standard laptop personal computer.

The PAL-R has 2 strain channels and 2 acceleration channels (user can specify either
piezoelectric or piezoresistive type). If four strain sensors are needed, then two sensors can
be “combined” with a special cable into on strain input. A second pair can be combined into a
second strain input. Thus four sensors can be applied by averaging pairs.

The PAX was first introduced in 2007. It is a Windows XP device, and also runs 8 hours on
internal batteries. The touchscreen makes operation simple. The graphics display is bright and
easily viewed in all lighting conditions. The unit is available in either 4 channel or 8 channel
versions. Wireless capability (eliminating the main cables) was added in 2008. The PAX is
equipped with programs to operate locally on-site, or remotely via a fast broadband internet

Both the PAK and the PAL use the PDA-W program for data acquisition and processing (the
PAX uses PDA-W for post processing of its data). The PDA-W program is a Windows program
replacing the DOS program previously operating the original PAK units. The Windows PDA-W
program has all the advantages of Windows programs (ease of use, common setups, etc.).
The PDA-W program guides the user into entry of information trough dialog boxes. The pile
properties are displayed (with dimensions). The PDA will automatically adjust the modulus,
wavespeed, and specific weight values so that they are always in proper theoretical
relationship. The program was designed with software control functions given through
keyboard entry or clicking Menu Bar selections or ICONS. The program makes frequent use
of dialog boxes which guide the user in a logical way to correct input.

The calculated values for nine selected parameters are displayed and can be automatically
summarized as required by ASTM D-4945 by the PDIPLOT program. Using the PDA-CURVES
program plots of various blows can be combined into one page.

There is considerable "HELP" available. In fact, the main operating manual is available as Help
electronically on the screen. PDA-W even evaluates the data quality and display "warnings"
for the new user if something appears to be wrong. The software lets the user enter limits on
output results to automatically have compared with testing results; violations are highlighted.

PAK or PAX can store every blow for every pile tested. This is limited only by the available hard
disk free space (previously the DOS version could only temporarily store 3440 blows). The user
can then select "after the fact" which blows (or even all blows) to save for permanent history.
Another advantage of this archive storage is that all individual transducer records are retained
and can therefore be replayed with only one accelerometer or to investigate bending. It also
means that the replay of data will always match exactly the original data.

Because everything is controlled by keyboard, the training process is simplified. Extra time can
be spent on theory and practical applications. It makes introduction to new PDA operators less
time consuming. Eliminating the expensive tape recorders also reduces the overall cost.


Without acquiring quality data, results are suspect. First, confirm the transducers are
functioning properly. Test the transducers as described in the PDA manual; replace or repair
defective units. Also regularly check the cables and connections (wiggling the wire near the
connector while observing the strain "offset" values). Signals with sharp "spikes" or signals that
quickly ramp off either positive or negative during testing are often caused by "intermittent
connections"; this is most often due to fatigue of the main cable at the point of attachment to
the pile during driving. If problems are detected, first check if all connectors are tight then if the
problem persists replace the main cable or connection cable. If the problem is only on one
acceleration channel, turn off that accelerometer and continue taking data; replace that
accelerometer for the next pile. Set the trigger to the appropriate transducer as the cables and
connections are checked as above. Sufficient spare cables and connectors should be available
on site to replace defective units as required. The wireless PAX system eliminates the main
cable and (relatively infrequent) connection cable problems, and thus promises to improve
overall data quality.


The field settings of ARea, LEngth, Density (SP), modulus (EM), wavespeed (WS), damping
factor (JC), and delay (DL) do not affect the data recorded (which is essentially raw strain and
velocity, or acceleration for the PAX) and can be corrected in reprocess (the field results may
be wrong!). The calibration settings for F1-F2-A1-A2 MUST be correct, requiring good
transducer calibrations. Smart Senosros with the PAX system eliminate the possibility of
incorrect entry. WS value (c) is the wave speed at the sensors which can be different than the
average (WC) used in the overall wave return (2L/c) due to cracks, etc. The user should verify
the wave speed is acceptable. The "Units" (SI, Metric, English) must be properly chosen. If the
modulus EM will be computed to match the Density SP and wave speed WS (EM = SP * WS2).

The two transducer systems (strain and acceleration) produce "proportional" data when both
are working properly and are properly calibrated. If the strain (force) and velocity are not
proportional, then strain or accelerometers should be recalibrated. In our studies, calibration
accuracy of any old individual strain transducer is within about 5 percent. In late 1992, we
developed a strain calibration which gives repeatability to within 1% (calibrations are traceable
to NIST - USA national standards). We recommend recalibrating all older strain transducers
(ASTM D4945 requires calibration every two years). Strain transducers can be recalibrated by
PDI, or calibration systems are available from PDI for purchase and use by our clients.

No data is abundantly available to verify the accuracy of the accelerometer calibration, although
the accelerometer manufacturers generally privately quote 5 to 10 percent accuracy figures.
The manufacturer of the PE accelerometers quotes an accuracy of 1.6%. NIST quotes a shock
calibration accuracy of only 5%. PDI developed in 1993 its own system for calibrating
accelerometers based on a Hopkinson's Bar; PDI estimates its calibration accuracy is about
2% (more accurate than NIST), and PDI results are in good agreement with the piezoelectric
accelerometer manufacturer. Calibrations by PDI are traceable to NIST since 1996. PDI
studies of old accelerometers suggests original PCB calibrations could be low by up to 15

percent. We recommend recalibrating any accelerometer acquired before 1993.
Equipment (or calibration service) is available to satisfy ASTM recalibration requirement of
every two years.

To check the PAK's calibration, the PAK "Cal Test" function should be used, on a periodic
basis, to compare values with known inputs described in the PDA manual; this will detect
problems with the PDA itself. Using this Cal Test feature also helps determine if the transducer
is functional. An external calibration box was developed to check PAK or PAL and can be
verified by independent laboratory accreditation agencies (who require ISO compliance).


The transducers should be firmly attached to steel piles as the high acceleration, could
cause slippage, resulting in inconsistent and therefore erroneous data. If the strain transducer
is not secured properly, damage to the transducer could also result. For concrete piles, over
tightening the transducers may result in pulling of the anchors instead; it is therefore very
important to firmly seat the anchors. It is always very important to have both strain signals
working since bending usually makes the strain signals on opposite sides quite different; if a
strain transducer fails partway through a test, it should be replaced immediately with a good
working unit. The transducers should be attached diagonally opposite the pile neutral axis so
that bending is canceled. It should be noted that it is really strain which is being measured. It
is further assumed that the pile is of a linear elastic material. In high strain situations, the pile
material may go into the plastic range and/or pile top damage may result. This could result in
unrealistic and inconsistent forces,and therfore erroneous energies and capacities in these
cases. Since the acceleration signals are very similar, generally one good unit is all that is
required for a successful test (unless bending is excessive).

The transducers also should (if at all possible) be attached at least 2 and preferably 3 or more
diameters below the pile top to avoid end effects and local contact stresses. In general, the
farther from the pile top the gages are attached, the better quality the data becomes (the only
difference this makes to the results is that the maximum energy EMX is reduced due to the
energy required to compress the pile above the transducers). Also avoid attaching near cross
section changes, welds (horizontal, axial, or spiral), splices, stiffeners or other nonuniformities.
It should be noted that "telltale" pipes cause complications which must be properly accounted
for. For regularly reinforced concrete piles (i.e., not prestressed) which are common in Europe,
the transducers should always be near the pile top to avoid including cracks between the
attachment points which could induce serious errors. For drilled shafts, using 4 strain
transducers is strongly recommended for either the PAK or PAL as it produces better data
(especially if attaching 2 diameters below the top is impractical).

Since 1994, we use piezoelectric shear accelerometers for all pile types (steel, concrete,
timber) and get better data on concrete piles than with the older type compression piezoelectric
accelerometers, particularly when accelerations are higher due to compressed cushions (we
no longer suggest using “Low G” or “Hi G” accelerometers). Shear accelerometers are
mounted on aluminum blocks for even high shock steel-on-steel applications (provided a printer
or other peripheral device is not attached to the PAK creating ground loops), or on plastic
blocks for older "blue PDA" users. These PE accelerometers are marginal for SPT safety
hammers, but unacceptable for most auto SPT hammers.

Piezoresistive type accelerometers are also capable of good data on steel to steel impacts from
uncushioned hammers or SPT test situations. Piezoresistive accelerometers can be used for
all pile types and all hammer types and in all cases gives data of at least the quality of the older
piezoelectric type accelerometer. Signal conditioning and software in the PAK allow the user
to operate with either accelerometer type. The PAL operates with only one type accelerometer
(type user selectable).


To confirm quality data, put the TS timescale to full and check for consistency; for similar
hammer blows, the data should look similar. Check each accelerometer individually (DPFV).
If one accelerometer is more consistent blow to blow, it is better to use only that accelerometer
than average a bad signal. To evaluate velocity quality, inspect the displacement (DPFW,
DPD, or DPE command on PDA-W). Better data has correct final displacement (compared with
observed set per blow); questionable data or negative (upward) final set will require more data
adjustment in CAPWAP and may overpredict capacity and underestimate transferred energy.
New shear piezoelectric or piezoresistive accelerometers give better data. In the field, begin
with all transducers active. By watching the data display, the engineer can select the trans-
ducers. If velocity is unstable from blow to blow, switch to A1 or A2 to find one accelerometer
which is functioning properly. Most of my time during the hammer operation is spent observing
the PDA graphic screen to view data quality (primary emphasis, since without good data any
analysis is meaningless) and assess bending, tension, large quake cases, friction distribution,
damage, wave speed, capacity methods, etc.

The user of a PAX can select a combination of sampling frequency and record size to achieve
about 100 to 200 msec sampling time in normal cases (200 msec is recommended total sample
time, and 10,000 Hz sample frequency with a 2K record length is a good selection). The current
default digitizing frequency is 5,000 Hz for the PAK for a total sample time of 200 msec (actually
the PAK samples always at 20,000 Hz, and then averages 4 samples to create an effective
5,000 Hz record; the averaging has a minor “boxcar filtering” effect). The user should change
to a higher sampling frequency (and shorter total sample time) only after careful inspection of
the entire velocity record; when it is assured that the velocity record is flat in the later portion
of the record, then and only then can the user select a higher frequency (FR function request)
for digitizing (results in a shorter time record). A longer time (lower frequency) may be required
for very long piles (e.g. offshore oil applications).

It is very important to check for proportionality of force to velocity for the major input rise
(peaks don't necessarily have to match exactly in amplitude). The relative change of the force
rise to the velocity rise is important. It is often easier to see this effect in the WU curve (use F7
function key or type DPW or DPFW to view Wave Up and show difference between force and
velocity, or press the F6 function key and WU will then be in the lower data window); WU should
be “smooth” through the impact and be free from obvious "step functions". For diesel hammers,
the slow compression build-up results in the force being higher than the velocity prior to impact;
this is acceptable as the slow pressure build-up time is greater than 2L/c and thus causes soil
resistance reflections. For slow rise times and impedance increases just below the transducer
location or gages near the ground and high friction in the upper soil layers, the force can
exceed the velocity at the first peak. For impedance decreases just below the sensors, the
velocity can exceed the force. The sensors also should NEVER be attached near (either just
above or below or worse still, straddling) an impedance or cross section area change as the
stress path can rapidly change through this area. If you have a choice, select a location on the

pile at least one diameter below any cross section change and with as long a uniform section
below that location as the place to attach your transducers. You may decide there is a local
phase shift between the force and velocity rise; such phase shifts may be reduced by the VT
time shift function (e.g. "VT-1" will shift the velocity signal one data point to the left; non-integer
shifts in VT are permitted).

Results should be reasonable. The energy transfer should be in line with energy transfers
of similar hammer pile combinations, the displacement curve should be reasonable,
compression stresses should probably be similar from pile to pile. If some result looks
suspicious, it is up to you to at least investigate and see if something is in error and needs

The PDA-W has available "expert advice" to inspect data quality. We strongly suggest you
review these highlighted “warnings” displayed in boxes above the graphs: proportionality,
excessive bending, clipping of strain signals, and accelerometers - ratio and integrated results.


In the PDA-W the force and velocity (multiplied by impedance) are displayed at the same force
scale so that proportionality between velocity and force can be easily evaluated. Some
computations (such as capacity and tension stresses) are influenced by wave speed input; if
wavespeed is wrong then capacity is inaccurate. Further, the proportionality factor (EA/c
or ρcA) is proportional to the wave speed; correct determination of wave speed is
essential for a good test.

For steel the wave speed (WS) is a known constant (16,810 ft/sec; 5122 m/sec). However, for
concrete (or timber) wave speed must be measured as it can be quite variable (need to also
measure density SP of each timber pile to get modulus EM = SP * WS2). The following figure
shows the normal way to get the wave speed. The method (see figure) uses wave down (WD)
and wave up (WU) derived from measured F and V, preferably in early easier driving. When
the resistance is low, the wave up has a similar shape to the wave down, except it is inverted
and delayed by 2L/c (usually true for uncracked prestressed piles but not necessarily for
cracked reinforced piles). The wave speed should NOT be taken from the peak WD to the peak
(actually valley) WU times (times 1 to 2, and the length below sensors). Because as the
resistance increases, the peak (or valley) return time (shown as time 3) will come sooner than
time 2 and using a peak-to-peak method will lead to an erroneous (high) wave speed since time
1 to 3 is too short for the true 2 L/c. For regularly reinforced concrete piles, the peak valley is
often delayed due to hairline cracks leading to an erroneously slow wave speed. Fortunately,
the "rise-to-rise" method avoids both problems as shown in the figure and therefore we
strongly recommend this "rise-to-rise" method. The times 4 to 5 as shown are constant and
represent the true 2L/c time regardless of the resistance. If wave speed is assumed during the
test and later determined to be different, corrections can and must be made. A discussion of
these corrections is in Technical Note to this paper.

Special Note for regularly reinforced concrete piles: Since wave speed is difficult to
measure (and may appear variable due to cracks), generally a predetermined wave speed c
(WS) is often used (may vary depending on supplier due to mix design, aggregates, etc.) and
a separate speed WC used only for the 2L/c time; WC (includes time delays due to cracks) must
always be equal to or lower than WS, the true speed in an uncracked section.


The Modulus of Elasticity (E or EM) converts the measured strain (ε) to stress (σ), and then to
force (F) by multiplying by pile cross section area (F = EAε). The modulus is computed from
the wave speed c (WS) and mass density ρ (=SP/g) by

E = ρc2
EM = (SP/g) * WS2

For steel the specific weight (SP = ρg) is 0.492 k/ft3 (7.85 T/m3, 77.3 kN/m3) and for concrete
the typical value assumed is 0.152 k/ft3 (2.45 T/m3, 24.0 kN/m3). For timber the density must
be determined for every pile (typical values 0.058 k/ft3, 0.93 T/m3, 9.1 kN/m3). The modulus
for steel is a constant (30,000 ksi, 2100 T/cm2, 210,000 Mpa). For concrete or timber, the
modulus varies as a function of wave speed WS and is calculated automatically for the SP and
WS values assigned.


To acquire new data, the PDA must be in the ACCEPT mode. Using PDA-W, data storage is
in memory. After data is collected, it is saved to a file (file size and number of blows limited only
by the free hard disk space; thus you can usually save every blow SX1). If the PDA-W program
is interrupted before the file is save, a backup provision will alert the user the next time the
program is started that there is a “backup” file and asks if you want to save it (the answer is
almost always “yes”). All four sensor signals are saved independently. The advantage of
digital recording is that data collected and saved in the field is exactly the same as during

For the PAK, the user can now easily store computed results of (e.g. EMX, CSX, RMX, RA2,
BTA, BPM, etc.) in an electronic (ASCII) file using the SQ command in the FILE menu bar
selection. The SQ file generated can then be read by the PC program PDAPLOT which can (a)
plot the results versus blow number or depth, and (b) print the results or averages (and
optionally standard deviations, maximum minimum for full ASTM D-4945 compliance) as
function of blow number or depth. PDAPLOT further summarized start and stop times for
productivity analysis. Comments inserted during data acquisition with the PC command are
retained in the file and printed with associated blow number; the comments (and/or short
interruptions in driving) can also be located on the plot in the new 1996 PDAPLOT version.
Although it is generally recommended to use the remote blow switch during data acquisition to
document the penetration of the pile, the new 1996 PDAPLOT has an improved capability to
enter the driving log. The PDAPLOT improves reporting to a professional quality result and
even speeds report preparation; several improvements were made in late 1995 and early 1996
which greatly simplify program use.

Proper data storage and presentation is improved with minimal effort and (as required by ASTM
D4945-96) is necessary for complete documentation (professional liability) reasons.
Representative force and velocity data for each pile tested should be plotted and included in
your report to conform with ASTM D4945-96.



We should always know the maximum stresses (or forces) so that the driving stresses can
be determined and kept between limiting values. The maximum force FMX is the compression
force at the transducers and needs little explanation; the force could be slightly higher just
above a point of high shaft resistance along the shaft, but this increase would be modest and
not likely to cause pile damage compared with pile top damage from local contact stresses or
bending stresses due to non perfect alignment. For piles with cross sectional area changes
along the length, wave equation studies should be made to find a stress amplification factor.

The input wave is transmitted to the toe and, for piles with little friction, if a stiff end bearing is
present, a compression wave will be reflected if this resistance is large relative to the input
force. This can potentially result in a doubling of the stress at the pile toe in "fixed condition
cases". The PDA can Compute Force at the Bottom (CFB) in these cases where blow counts
are also likely to be high. Since the bearing pressure is rarely uniform along the bottom, stress
concentrations are likely. The limiting force (stress) at the bottom should be lower that the limit
for the pile top. Toe reinforcement often helps prevent toe damage.

The tension force is generally of interest for concrete piles and usually only during easy
driving although for large quake soils, high tension stresses can still be a problem even for
refusal driving. CTN only considers the maximum net tension from the first returning from the
toe; CTX also considers the downward tension wave late in the blow.

Using the input pile area (AR), PPD-W calculates the maximum compressive stress (average
FMX/AR = CSX; or max of either individual strain CSI) and, for concrete piles the tension stress
(CTX/AR = TSX, or CTN/AR = TSN). The stress at the pile bottom (CSB) can be estimated from
(CFB/AR). These quantities (CSX, CSI, TSN, TSX, CSB) can be computed and eventually
summarized numerically and for later plotting by PDIPLOT. We recommend that driving
stresses be included in the results requested for computation and display.

High driving stresses are a leading cause of pile damage. If either compression (top or bottom)
or tension driving stresses are to high, the cushion could be increased or the hammer energy
reduced (lower stroke). It is recommended to limit compression driving stresses (CSX) to:

Steel (compression) 0.9 Fy

Concrete (compression) 0.85 f'c - prestress
Timber (compression) 3 x (allowable design stress)

Tension stresses for prestressed piles are often limited to 3(f'c)0.5 + prestress (English units).

Bending stresses can superimpose on the axial stresses and create critical situations. The
leading causes for bending are poor hammer pile alignment and non square pile tops. Every
effort should be made to keep the bending as measured in the difference between F1 and F2
measured strains at a minimum. Another cause of local top damage is poorly fitting helmets
(too small, too big, or non flat impact surfaces).


Pile damage is generally noted by a BTA value less than 100; BTA is loosely defined as the
reduced pile cross section (really impedance reduction) divided by the full cross section. If BTA
is less than 100, a third time marker is shown. The QBTA or "quick beta" function always does
a preliminary scan for damage even if BTA is not selected. The damage search was
extensively improved in 1995. BTA can now find up to two damages (BTA and BT2). It is
suggested that the user consider and investigate the possibility of local bending as a possible
alternative if damage is near the pile top.

If damage is indicated by the PDA, the operator should immediately note its suggested location
(LTD - Length To Damage, as measured from below the sensors) from the third (middle) time
marker, and probably stop the hammer to prevent further possible damage. The first question
to ask is if the pile is non-uniform, or if a splice detail is causing a false indication. Earlier blows
can be reviewed to inspect if the problem area is getting worse. If the problem is very near the
pile bottom, perhaps the indicated length or wave speed is slightly in error (check LE or WS);
again check earlier blows or pile records. If stresses are high, try to change the driving system
(more cushion or lower stroke) to reduce harmful stresses for further production piles. If the
pile is a closed end pipe, drop a tape inside to measure the length or a light lowered (if
sunshine, reflect from a mirror) will allow visual inspections. Replay the forces and check if
bending is severe in the records which could cause a minor disturbance in the data and
therefore a false indication. Is the "damage" caused by a splice or other pile non-uniformity?
Is the "damage" caused by the data quality? Can it be "eliminated" with FF or VT adjustments?

In reality, the PDA operator should always check the wave up curve himself; a monotonic
increase for 2L/c after the initial rise indicates no damage while a reduction in wave up (relative
increase of velocity to force) before 2L/c after the initial rise is likely due to an impedance
reduction, crack or gap. If the reduction is sharp or dramatic, there is likely some damage.
If the pile is long and has considerable distributed friction, then the velocity may become
negative before 2L/c. In some cases the wave up curve can then reduce before 2L/c due to
unloading effects and may fool the PDA. The user should visually inspect the wave-up record.

NOTE: Sometimes damage detected by the PDA during pile driving cannot be seen during
restrike because of a) soft cushion on restrike, b) due to setup and a higher friction, and c) also
possibly because cracks in concrete "heal" with time after driving. The high friction essentially
prevents motion at the damage location and therefore no reflection waves are generated which
would be indicative of the damage. In order to "see" the damage during restrike, a much
sharper impact (less thickness or more stiff cushion; or higher drop if high friction) would be
required, however, that may also damage the pile top so care and caution are advised.
Sometimes the restrike tests are not sufficient to evaluate a pile for integrity, particularly if the
soil set-up is strong. For that reason, in Sweden for example (where they drive reinforced,
jointed concrete piles through clay) pile integrity tests by the high strain method are best
conducted at the end of driving. Bearing capacity tests, of course, must be performed after
a setup period.


Hammer performance should always be evaluated. The most important indicator is EMX,
the maximum energy transferred into the pile. Comparison of EMX with the potential energy
Wh (better still, with the manufacturer's rating) is generally all that is necessary. Considerable
effort was made to compile transferred energy at the end of drive for various combinations

of hammer types and pile types; statistical results are presented in the PDA Manual. Energy
at the end of driving is the only easily comparable value since soil resistance and hence blow
count are reasonable; for several reasons (such as alignment) hammer evaluation during
restrike is more difficult. If more resolution is needed click EDIT/”OUTPUT QUANTITIES”
and change the precision. The PDA can also test SPT rigs as per ASTM D-4633.

If a hammer is not working well, the transferred energy will be relatively low and it will take
longer to drive the pile so productivity suffers. If the hammer is not performing well and a blow
count or set per blow is used as the pile acceptance criteria, then the pile could be accepted
prematurely at a lesser pile penetration and the actual pile capacity could be dangerously too
low and foundation failures could result. It is generally to everyone's advantage to have a well
performing hammer so the project gets completed as efficiently and quickly as possible, and
just as importantly that the penetration is sufficient and the capacity is adequate.

The hammer operating rate in blows per minute (BPM) can be determined by the PDA, up to
the speed limit of the PDA. If the PDA misses a blow during acquisition, the BPM value of the
PDA will be half the actual BPM. For single acting (open end) diesel hammers only, the stroke
(STK) can be calculated from the BPM using principals of gravity and empirical correlations

"STK" = H[m] = 1.22 (60/BPM)2 - 0.1

"STK" = H[ft] = 4.01 (60/BPM)2 - 0.3

Further PDA analysis is possible by looking at the momentum calculations (see PDA manual;
MF0 is generally preferred to MW0). If the ram weight (WR) is input, the maximum ram velocity
at impact VRI (for ECH or ASH hammers only) can be calculated from MF0 ( VRI = MF0/WR)
and to compute the kinetic energy of the ram (.5 [WR/g] VRI2) to investigate if losses are
primarily in the hammer or the lower driving assembly. These momentum equations can
sometimes lead to unusual answers and results should be viewed with caution and accepted
only if reasonable.

A new output quantity is the time from rise to peak ("TRP") which is a measure of rise time (data
samples actually) which in turn depends on cushion stiffness. The force FCP and the stiffness
KCP in the hammer cushion can be computed if both WR and the helmet weight (WH) are input.
These FCP and KCP values are only correct for (restricted to) air/steam/hydraulic or drop
hammers on steel piles.

PDI offers two other devices for hammer performance testing. The Hammer Performance
AnalyzerTM (HPA) uses Doppler radar technology to obtain the ram impact velocity which is
particularly useful for air/steam/hydraulic/drop hammers or for SPT energy determination. The
SaximeterTM counts blows, blow rate, and can calculate the stroke for every blow of an open
end diesel hammer; results (averages) can be stored in memory for an automatic report
presentation (data down loaded to PC in ASCII and printed with SAXPRINT program).


This is the biggest challenge facing the engineer, and potentially the most costly if he makes
a mistake. However it is also the most rewarding when the PDA can reduce expensive static
testing or determine that length and cost of a pile foundation can be reduced. Unfortunately,
it is impossible to give guidelines that apply to all situations. You should read all applicable
sections of the PDA manual.

It should be emphasized that the 2L/c time period and wave speed WS be correctly
determined and entered, or else large errors (particularly those with clearly distinguishable
reflections at 2L/c) may result; the rise to rise method is required (see PDA manual section on
wave speed). Wave speed influences both 2L/c and also the elastic modulus to convert strain
to force; a 10% error in wave speed results in a 20% error in modulus and/or force. The
modulus is also linearly dependent upon the pile material density. While density is known for
steel or concrete (density of grout for augercast piles is 10% less than density for concrete), the
density of timber piles can widely vary and must be measured for each pile to assure accuracy.

All PDA Case Method capacity results assume that the pile is uniform (cross section and
modulus versus length). If the pile is not uniform, the PDA capacity result may not be
reliable. The CAPWAP program can accurately model non-uniform piles, and should be
used for capacity determination for all non-uniform pile cases.

The RSP methods use empirical damping factors JC determined from soil type (implies the soil
is properly identified by grain size, and that the soil at the boring is similar across the site). The
empirical study included data primarily from restrike (or end of driving in sands) with moderate
blow counts. Unfortunately, the RSP method is sensitive at low blow counts; small JC changes
can make large capacity changes. For large soil quakes, the full toe resistance may not be fully
active at 2L/c unless a time delay is used. For concrete piles with 2 peaks (from non-uniform
pile cushion compression), selecting the second peak (use a delay DL) usually gives a better
solution. View RT-RS curves and adjust JC until a "flat" curve is obtained; however as shaft
friction increases, this technique becomes less reliable. For shaft friction, the SFT computation
makes no allowance for damping (SFR has a crude correction).

The RMX method searches for the maximum resistance during the entire blow and thus
overcomes some of the limitations of the RSP methods for small or large blow count, or high
quake situations. Many find a JC of 0.7 for RMX (or RX7) gives a good first estimate. Although
temptation exists, do not use damping factors less than 0.4 with this RMX method without
substantial proof from CAPWAP or a static test that a lower damping factor correlates well.
Higher damping factors may be necessary, but a study indicates JC factors above 0.8 are
usually not needed. For friction piles in clay (where high J factors are normally appropriate),
the full resistance should be active during the first 2L/c cycle anyway (RSP = RMX). Sensitivity
to the damping factor can be studied by viewing multiple results (e.g. RX5 for JC of 0.5 and RX8
for JC of 0.8 etc).

For uniform piles with zero shaft resistance, the RAU method is theoretically the perfect method
as all theories are correct and the method is independent of a damping constant. It makes no
difference if it is easy or refusal driving; the key is that the force and velocity must be
proportional for the entire first 2L/c (implies good data) and the 2 L/c must be correctly chosen.

The method RA2 has shown considerable promise in determining the ultimate load even for
piles with little to moderate shaft friction and this method also does not require the selection of
a damping factor. Results are generally in good (not necessarily great) agreement with results
from CAPWAP and therefore the method deserves at least a casual consideration on every
project. if RA2 differs from the damping factor methods (RX7), then investigate further. If the
pile is driving through a layered soil, the RA2 method has the additional advantage that the
damping factor does not need adjustment. Again, 2L/c must be chosen correctly.

For longer piles with high friction distributed along the shaft, the velocity may become negative
prior to 2L/c and the upper soil layers begin unloading even prior to the loading of the lower soil

layers. Most all methods then underestimate capacity and the unloading method RSU may be
beneficial (RSU attempts to determine how much friction has unloaded and adds it back into
the equations as a correction factor). RSU uses the JC damping factor (RU7 is RSU with JC
= 0.7). However, CAPWAP (or static test) should be performed as soon as possible to verify
the correct procedures.

Note: A "dynamic formula" method QUT (QUT = 2*EMX/(DMX + SET) can be calculated which
uses EMX and DMX measurements in a "modified Hiley formula" (where SET is calculated from
the blow count as the permanent penetration per blow.) Although we do not recommend
dynamic formulas, this one at least contains some actual measured values, and may indicate
the potential benefit to be gained at the very high blow counts. RLT is the limiting value of QUT
for zero set per blow (note that RLT is not the same as RTL which is RSP with a Jc of zero).
We do not recommend reliance upon QUT but the PDA can compute QUT for user defined SET
(entered as USR), or QBC from blow count switch information. Using the VIEW then
CAPACITY MONITOR shows the variation.

On practically all of our projects, we perform at least one CAPWAP to confirm our field
methods; for larger projects, we generally perform CAPWAP for about 30 to 50 percent of
tested piles, although usually only for the data at the end of drive and/or begin of restrike. We
always recommend some CAPWAP analysis for confirmation. Inspecting resistance
distribution, unit friction values, and end bearing determined by CAPWAP (for restrikes and end
of drive) and comparing with soil boring and static analysis calculations often results in better
recommendations of total capacity, optimum driving criteria or pile length. Further discussion
of CAPWAP is beyond the scope of this paper except that confirmation by this independent
analysis can detect unusual conditions not noted simply by PDA testing. With the PAK or PAX
PDA and their capability to run CAPWAP (or take you laptop along to field), a preliminary
CAPWAP analysis should be performed on site; major surprises after performing the field work
are then kept to a minimum.

If the static load test is not run to failure, then only a lower bound "proof load" can be given;
similarly if the blow count is very high (small set per blow, typically less than 3mm/blow or
greater than 10 blows per inch) then the dynamic test's ultimate capacity may be low.
However, it is currently not possible to determine just how low (if any) the capacity may be at
these small sets per blow (depends greatly on soil type and end bearing percentage). Ideally
a larger hammer, higher drop height, or less cushion, could be used to increase the impact
force and energy transferred to produce a higher set per blow and mobilize more capacity. In
refusal driving, it may be more important to increase the impact force than to increase the
energy transferred. The soil resistance which can be overcome is theoretically at most twice
the impact force; practically this limiting ratio is usually more like 1.2 (1.0 to 1.5; site and pile
type dependent) due to damping and quakes.

Dynamic testing estimates for the pile capacity indicate the mobilized pile capacity at the
time of testing (i.e. end of drive EOD, or begin of restrike BOR). Increases and decreases in
the pile capacity with time typically occur (soil setup/relaxation). Therefore, dynamic testing
during restrike tests usually yield a better indication of long term pile capacity than a test
at the end of pile driving. Pile capacity during driving is often less than long term pile
capacity. Capacity can temporarily be reduced during driving by lateral pile motion. For piles
driven in fine grained soils (clays, silts and even fine sands), soil is remolded and excess
positive pore pressures are often generated during driving. These pore pressures reduce the
effective stress acting on the pile thereby reducing the soil resistance to pile penetration, and
thus the pile capacity at the time of driving. As these pore pressures dissipate, the soil
resistance acting on the pile increases as does the axial pile capacity. This capacity increase

is routinely called soil setup. The full setup effect often requires several days to even weeks
to obtain a good correlation with the long term pile capacity. A one day restrike is not a
sufficient wait period for many soils. and a comparison of end of drive dynamic tests with static
tests performed weeks later will undoubtedly produce disappointing results. Many engineers
fail to understand this and compare end of drive PDA results (or even one day restrikes) with
static tests run two weeks later, with obviously poor correlation. They improperly blame the
PDA instead of the capacity change is due to set-up strength changes. This is a common
problem we hear. Time dependent soil strength changes should be clearly explained and
discussed with the client.

Because of soil strength changes during driving, restrike tests are therefore considered very
important, especially if good correlation is to be achieved between the dynamic test and a
static load test. The optimum waiting period for the restrike varies depending on the pile type
and soil type; some report that testing (static and/or dynamic) reveals that capacities are still
increasing after more than a year, while in other soils no observable setup is ever noticed.
Knowledge of the soil type, whether the soil is saturated, and local experience can be used as
guides for selecting the wait period before restrike tests. As a general rule, a longer wait time
before restrike is better. In many cases recommendations of a seven day wait (similar to static
test recommendations) result in near full setup. Setup effects in silts and clays, or soils with
relatively high fine content, since pore pressure effects are then governing the behavior) often
are linear with log time (since pore pressures decrease linearly with log time; results of the shaft
portion of the resistance can be plotted as a straight line on a capacity versus log time graph).
Skov and Denver (1988) have proposed the following

R2/R1 = 1 + A log10 (T2/T1)

where R1 and R2 are the capacities at the times (T1 and T2) of interest of the first and second
restrike events respectively relative to the end of driving. The constant "A" is determined by
experience or by measurement. If two restrikes (with less than refusal blow counts) are
available the constant "A" can be calculated. Bullock has shown that early restrikes of 15
minutes and one or two hours can be helpful (plotting the EOD test at one minute while defining
unity time at one day is often reasonable; there are then three log cycles to one day, while a 10
day restrike is only one additional log cycle).

In some soils the gain may be more linear with time. For example in sands, driving may create
an oversized hole due to lateral pile motions (e.g. pile whipping) that can cause arching effects.
In the case of closed end pipes with slightly oversized end plates, the end plate may act as a
“cookie cutter” in the soil, weakening the shaft friction temporarily. In these cases, the normal
earth pressures may slowly reestablish themselves, and thus cause additional shaft resistance
as the normal forces increase with time. Some soils are cemented, and take time to reestablish
the chemical bonding.

Particularly in the case of high shaft friction in clay, silt or even fine sands plus high end bearing
in sand, the full friction may only be available after a setup period and seen during restrike,
while the end bearing is perhaps best determined from the end of driving (assuming there is no
relaxation). If and only if the restrike blow count is very high (e.g. refusal), a summation of
these two cases (by CAPWAP solutions) may in some instances be possible and give a better
indication of the total capacity available during service conditions. It is only in cases of
extremely high blow count restrikes (low set per blow) that this need be considered. At low to
moderate blow counts (set greater than 3 mm/blow, blow count less than 100 bl/ft), this
superposition of results is definitely not advisable nor necessary as the full resistance has
already been mobilized. Because of possible relaxation (see following paragraphs), adding

EOD end bearing to restrike friction (without confirming by static tests) is not always a good
idea since it could result in an overprediction.

Although less common (fortunately), some soils exhibit relaxation or a capacity reduction
as a function of time. Some suspected cases are in reality due to hammer performance being
poor at the end of driving causing relatively high blow counts, and the hammer performing much
better during restrike or redrive resulting in relatively lower blow counts; the PDA can easily
identify these cases by looking at the hammer performance indicator EMX.

There are cases of real relaxation and the PDA User should be aware that a potential capacity
loss can result in foundation failure and the ensuing legal problems. Three cases of relaxation
quickly are noted. First is the case of piles driven into some shales. The relaxation potential
is generally greater in weathered shale and may take several days to fully develop. These
shales have been identified in many parts of North America and probably also exist in other
parts of the world; losses in capacity of up to half the end of drive capacity have been observed!
Capacity estimates based upon initial driving or short term restrike tests can significantly
overpredict long term pile capacity. Therefore, piles driven into shale should be tested after a
minimum one week wait either statically or dynamically (with emphasis on the first few blows).
The second case is for displacement piles driven into very dense soils (often silts) with little
shaft friction; the high end bearing at the end of driving reduces with time, the soil beneath the
toe springs back since the friction is unable to keep the toe in a compressed condition (residual
stresses also reduce). In a third situation, negative pore water pressure during driving into
dense saturated silts or fine sand causes artificially high effective stresses and end bearing
during driving. With time, the pore pressures return, and effective stresses reduce causing the
decrease in end bearing. Again, restrike tests should be used, with great emphasis on early
blows; often a wait period of one or two days is satisfactory but depends on the permeability
of the soil.

For any relaxation case, some later restrike test (preferably with PDA testing) should be
performed with a very careful measurement of the set for the first few blows which can then be
compared with the end of drive blow count; look for the PDA capacity of the first high energy
blow. Capacity often increases rapidly blow by blow and if a "later blow" is chosen, the
capacity may overpredict the true static condition. Therefore, in relaxation cases choose the
earliest blow with reasonable energy to produce an acceptably large set per blow (look at FMX,
EMX and capacity versus blow number for trends, ratios and other telling identifiers).

Soils are difficult. The variety and variability of soils encountered horizontally across the site
or vertically along the pile shaft causes complications. Clean coarse grained sands are
generally well suited to dynamic capacity analysis, even at the end of driving, since capacity
changes with time are usually minimal; however, the end bearing of larger diameter
displacement piles may be underpredicted at higher blow counts (which is a large problem area
for piles driven in clean sands; another problem is change in water table and effective stresses
between time of testing and the service condition, particularly seasonal variations or if the site
has been temporarily dewater for construction; the geotechnical engineer should review these
changes, as well as settlement concerns for the piles, and pile groups, as well as scour and
other concerns when adapting the dynamic testing results into his design and installation
criteria). Similarly, testing of piles driven to a hard competent rock is often quite

In between clean sands and pure clays are a wide range of other soils. In the majority of all
cases, good capacity results can usually be obtained by following two simple rules.

1) Make sure resistance is fully activated.
2) Restrike the pile after a wait period.

Further discussion of these two simple rules follows:

1) Make sure resistance is fully activated. The hammer energy should be sufficient to
produce a pile set per blow greater than about 3 mm/blow (less than 10 blows per inch).

2) Restrike the pile after a wait period appropriate for the soils on site. As the soil
pressures equalize around the pile (larger hole during driving due to lateral motions or
over size pile shoe), as pore pressures decrease, as the soil structural changes after
being remolded by the driving process, and other strength changes (setup increases or
relaxation decreases) are best evaluated by restrike testing.

In a small percentage of cases (estimated at less than 5 percent of all sites) capacity
determination is unsuccessful if the dynamic test is performed on restrike (after appropriate
wait) and the blow count is reasonable. CAPWAP confirmation is always recommended as
it better determines soil behavior and identifies unusual soil conditions; when conditionals are
unusual, static tests should be recommended. Larger diameter open ended pipe piles (or H-
piles which do not bear on rock) may behave differently under dynamic (no plug effect) and
static loading conditions (plug develops), and caution when testing these piles is suggested.

In cases where the pile is driven to high blow count with significant end bearing and then large
additional shaft friction from setup occurs, it may be beneficial to stop one pile just above the
bearing layer (low tip resistance) such that on restrike the full friction may still be mobilized.

One suggestion to reduce testing costs is to use a "sister pile" approach. First, have several
piles installed before arriving on site; this also assures that the contractor has worked through
problems often occurring while setting up his pile driving hammer at the beginning of a project.
Then, a few days into the project (or after a weekend), you can test a couple piles during driving
and several piles previously installed to similar criteria all in the same day. This gives the
maximum information with the minimum testing time for small budget projects. Of course, on
many projects the specifications require testing during driving of all test piles, often for hammer
performance or driving stress information, and then again during a restrike after a specified
wait. (The remote PDA operation may be useful here to minimize the additional trips to the
jobsite to contain costs).

On larger projects, periodic testing during production will identify site variations and hammer
consistency or identify malfunctions; the recommended testing frequency is generally at least
one test day every two weeks. Testing relatively early in the project detects problems early;
solutions can then be devised to correct the problem. The revised driving procedures are then
applied to the production piling thus avoiding having to perform major corrective action on a
large number of already driven production piles which is by far more expensive than the cost
of the dynamic testing.

The PDA is obviously a very powerful analysis tool when properly applied. From the preceding
it is shown that capacity is a complex problem with many features contributing to the testing
success (or failure). Obviously good quality measurements are required; if the data is poor then
any analysis is suspect. Organizations with PDA's should make every effort to make
measurements on all of their in-house pile projects as it will detect most common problems, and
reduce liability. Projects without dynamic testing rely on numerous assumptions and
therefore generally attract legal problems and litigation. It is much better to avoid court

battles through proper documentation (PDA measurements) and solving problems early
in the project.

Numerous other factors are usually considered in pile foundation design. Some of these
considerations include additional pile loading from downdrag or negative skin friction, soil setup
and relaxation effects, cyclic loading performance, lateral and uplift loading requirements,
effective stress changes (due to changes in water table, excavations, fills or other changes in
overburden), settlement from underlying weaker layers and pile group effects. These factors
have not been evaluated by the PDA and have not been considered in the interpretation of the
dynamic testing results. The foundation designer should determine if any of these
considerations are applicable to his project and the foundation design.


Sometimes we hear people say the correlation of dynamic and static tests is bad; they assume
the dynamic test is wrong and the static test is right. However, many static tests have
problems. Many tests are performed by people who infrequently do static tests when required
(they are less than enthusiastic) or by technicians who don't understand the test or realize
something is wrong.

Many tests lack a good, linear, recently calibrated load cell, resulting in serious errors. I
recall a project where our prediction was low (compared with jack pressure loads). Further
investigation showed the last jack calibration was 20 years old! Fortunately, the engineer
insisted the jack be recalibrated and the new jack calibration confirmed that our dynamic test
result was correct. Reading jack pressures only can overestimate the true capacity (by up to
30%), particularly if spherical heads are not used to account for non axial loadings. The latest
ASTM D1143 fortunately requires load cells (for ultimate loads of 100 tons or more) and
spherical bearing plates.

I have observed errors in displacement measurements. Reference beams should be

anchored far from both the test and reaction piles to avoid ground movement distortions;
unfortunately, long reference beams may experience other problems like temperature effects.
The engineer should review the stiffness of the static test to eliminate obvious errors,
particularly for multiple tests at the same site. As a simple example, I saw a test recently that
was loaded to over 1500 tons, yet had zero displacement; application of load without some
displacement indicates error as no soil is perfectly elastic.

I read where the reaction method caused a great difference. On the same pile, a test with
dead weight gave about twice the load on the same pile using reaction piles. The reason was
attributed to changed effective stresses in the soil due to the anchors. If effective stresses
change, due to overburden or water table changes, then the pile capacity will also change.
Occasionally the soil conditions are altered due to drilling or jetting or driving or vibrating
reaction piles (particularly if the later reaction piles are at a close spacing and extend below the
toe of the test pile.

Another concern is failure load evaluation. Have you heard "we loaded the pile to xxx tons"
which seems very high, and then find the xxx tons was at a displacement of 6 inches (150 mm)
or more! Such displacements exceed the usual serviceability limits of most structures. There
are different methods for evaluating failure loads: slope methods, stiffness methods,
extrapolated ultimate load, some percent of shaft diameter, etc. If the test is essentially linear
and then quickly plunges, all methods give similar answers. But if the curve continuously takes
more load at increasing displacement, then a wide range of answers can result.

Davisson's method can be easily described mathematically (although what is the modulus for
concrete or the stiffness of a concrete filled steel pipe?). It is not intended for uplift tests but
rather for compressive "quick test loads" (such as CRP or ASTM Quick) and which should not
cause excessive creep loading or other serviceability problems at the working loads. The
Davisson limit is NOT programmed into the PDA or CAPWAP, but correlations are often better
if the static test is evaluated at a reasonable displacement as defined by Davisson. The
dynamic test result (given as a single value) should consider time dependency, reasonable
set.... and a tolerance of 5 to 10% of the impact force is more realistic. To blindly compare one
number from a dynamic test with one number from a static test, without reviewing all test
information, test details, sequence of installation and all data available, may lead to
disappointing results.


Most piles tested are one material and uniform area (such as H piles, steel piles, or prismatic
concrete) along their entire length. In these cases, attachment of transducers directly to the
pile and the subsequent analysis is straightforward (described in the manual and/or this paper).

There are situations where the pile is not completely uniform, but testing is still desirable. The
first such case is perhaps a concrete filled steel pipe. This is maybe uniform along the length
but made of two materials. Should we attach the transducers to the steel or cut "windows" in
the steel and attach to the concrete? I shall try to explain what we do and why. The basic
premise is that the strain in concrete and strain in steel are equal. If this is not true, then any
reasonable analysis is not possible. If the pile is properly constructed with good quality
concrete filling the steel and hit the concrete uniformly through a plywood cushion, then all
should be satisfactory.

If the concrete is under filled with respect to top of steel, a uniform stress is unlikely and
probably the bond between steel and concrete will be broken when the hammer strikes only the
steel shell. In this case, we recommend cutting off the excess steel at the top and then hitting
the concrete, since the concrete impedance is usually many times the steel impedance (we
generally prefer to hit the concrete through a pile cushion). Compute an average densit and
average modulus:

[average SP = {(SP*AR)steel + (SP*AR)con}/ARtotal]


average EM= {(EM*AR)steel + (EM*AR)con}/ARtotal],

compute the wavespeed (WS=[EM/SP]1/2) and use the total ARea (concrete plus steel). Then
confirm the wave speed WS from the PDA test (look for toe reflection from rise to rise method)
and calculate the average EMavg = SPavg * WS2 .

If the steel impedance is very low (at least an order of magnitude less) compared with the
concrete impedance, you could remove a "window" in the steel large enough for transducers
to be attached to the concrete. As a word of caution, the concrete should be filled to the top
of the pipe; hit the concrete through a wood pile top cushion (thin or compressed for restrike
tests). Compressing the concrete increases the bond between steel and concrete due to
Poisson's effects; hitting the steel first decreases the bond and generally gives poor

If the steel impedance is relatively large, then cutting "windows" could make a stress
concentration in the concrete which is not desirable, making the concrete strain too large.
Remember we are measuring strain. If the piles are manufactured under good quality control,
it may then be better drill and tap the steel and to attach to the steel and not cut windows. If
proportionality is then good, the bond is probably OK, and data acceptable.

A second special application is a pile which is non-uniform along its length,. such as a steel
pipe of variable wall thickness, a composite pile (a concrete section with H pile projection at the
bottom), a steel follower driving a concrete pile, or a tapered section (monotube or timber pile).
As a general rule, transducers should be placed some distance away from sudden cross section
changes. Stress path changes, "end effects", or two dimensional (out of plane) strain adversely
affect the data near the change. For the non-uniform pipe pile, placing the transducers at least
one half diameter below the change is preferred, although this is not always possible. For steel
pipe piles (particularly spiral welded pipes) place the strain transducers as far from the welds
as is practically possible (4 strains for spiral weld improves data quality).

Attaching transducers to a steel follower when driving or testing concrete piles has several
advantages. Because the steel modulus is known, the strain to force conversion is better
defined. Further, the follower has to be prepared only once and then numerous piles can be
tested without individually attaching transducers to each pile. A few simple precautions are
necessary for a successful test. 1) The impedance (EA/c) of the follower should be as close
as possible to the impedance of the pile. 2) Do not put transducers very close to the bottom
of the follower; leave at least a half diameter distance. Also stay away from other stress
concentrations such as stiffeners. 3) The location of the pile cushion is extremely
important. The amount of cushion material between the follower and pile should be kept to an
absolute minimum as large relative movement between the follower and pile will make analysis
very difficult; the main cushion normally used at the pile top should instead be placed between
the hammer and the top of the follower and will have the same protecting effect on the pile.

For this follower/concrete test, or other composite or tapered piles, there is a question of how
to analyze the data. Since the Case Method is not intended for non-uniform piles, it is
probably best if one assumes the pile is uniform (EM, WS, SP values at transducer location
apply to entire pile) and then analyze the data later by CAPWAP (the entire first element in
CAPWAP must also be uniform so either locate transducers at least one meter above non-
uniformity, or increase the number of pile segments in CAPWAP, or simply tell CAPWAP that
the length to the first change is at least one meter, even if it is not). The maximum force and
energy will be correct, capacity by the Case Method in the PAK will be unreliable, but capacity
should be estimated by CAPWAP anyway.

Underwater testing has been accomplished using special transducers. A complete waterproof
system has been used to depths of several hundred meters (offshore oil platforms), or if the
transducers are to remain submerged for long time periods. The depth is really limited by the
water depth at the site, although special custom "underwater main cables" are required in this
application. These main cables are generally relatively easily retrieved (and cables are most
expense part of the system), although the transducers are generally sacrificial for each pile as
the cost to retrieve is higher than the cost of the transducers.

A simpler (less costly) underwater test system is available when the transducers are near the
water surface (say within about 30 meters or 100 ft of water depth), and no long duration

submerged period is anticipated. Specially prepared transducers with individual long
transducer cables allow connection into the system connection cables at a point above water;
all connectors must be kept above water. Again the transducers may in many cases be less
expensive than the cost to retrieve them.


If you have an older program version, there are many improvements available to you with the
newer versions. It is the responsibility of the PDA User to request these program upgrades
from PDI and install the new PC program to keep current. Changing your PDA-W program is
as simple as uploading a new file. Check your program version. If your version is older than
one year, please contact us for an update for your PDA.

Please also submit additional suggestions (changes, improvements, etc.) to Pile Dynamics for
our consideration; many features (improvements) have resulted from your valued suggestions.
The PDA is equipped with several features which make automated data processing possible.
The professional PDA user should read the manuals (or other available material such as
technical papers on the PDI website in the technical library) and adapt this
technology into his practice.


Testing drilled shafts with drop weights and a PDA is common in many areas of the world, but
we offer this extra advice. As the pile diameter increases try to use four strain transducers (one
transducer every 90 degrees around the circumference) to get a better average strain which
seems to help for nonuniform shafts (especially true for non-cased shafts). If F1 was bad
during your test (you should turn F2 off, if it was opposite F1), you still have had 3 others
including the F3 and F4 "opposite pair". Particularly for very large diameters, F1 and F2 might
be at only 90 degrees, and F3 and F4 also at 90 degees. It is imperative to know which are
the opposite strain pairs (e.g. F1 and F2, or F1 and F3, or F1 and F4) in case one is
determined to be bad and its diagonally opposite mate also turned off. Using the "AT"
command, the PAK can do 8 channel (4 strain plus 4 acc; actually 2 acc is generally enough
as the velocities are all about the same). You need special connection cables for the left
Piezoresistive accelerometer 19 pin input plug if you do not already have them (the strain
channels are wired differently so normal piezoelectric connection cables for right side will NOT
work in left side of the PAK); the 19 pin main cables are identical. You could optionally get
Piezoresistive accelerometers so you could measure four accelerations, or do specialty tests
such as instrumenting the same pile at two locations, or testing steel-on-steel impacts like SPT.

Further, note that if a composite section exists, it must be correctly considered. For example
consider a cased drilled shaft with casing size of 830 mm diameter and 12 mm wall. The steel
area is about 300 cm2; concrete area is 5000 cm2. The pile Force is (strain) x EA, so since the
steel modulus is significantly more than the concrete modulus, the FORCE contributions of the
steel and of the concrete are actually quite similar in this case, so ignoring the steel would
result in a serious mistake. The analysis of such composite piles only makes sense if the strain
in the steel and the concrete are IDENTICAL. (the ram should impact the concrete in
compression through the plywood cushion, and the concrete expands horizontally due to

Poisson's ratio to increase the steel to concrete bond; if you instead hit the steel, then the steel
expands and pulls away from the concrete destroying the bond so we do NOT recommend this).

In our example, since the steel modulus is about 5 times higher than that of concrete, the EA
of the steel is about 6,300,000 [kN] (300 x 210,000 MPa) relative to 20,000,000 [kN] (5000 x
40,000 MPa) for the concrete, so cannot ignore the steel which is about 1/4 of entire EA of pile).
Due to relatively high percentage of steel, compute average density from

[rho x area]steel + [rho x area]concrete / areatotal.

Finally, get E = rho x WS2. Note that WS for piles with a high percent of steel will be higher than
for piles with concrete alone; the higher the percentage of steel, the higher the WS. You can
observe overall wave speed from the 2L/c reflection.

For a thick shell (e.g. 830 x 12 mm diameter shell) it is better to drill and tap the steel shell
rather than cutting "windows" in the shell and bolting to the concrete. The windows can be a
significant portion of the steel (reducing the EA), creating a "nonuniform" pile and a "stress
concentration", and the heat of torching the steel does not help the concrete quality. Again
remember the basic assumption is the strain in the steel MUST equal the strain in the concrete
for composite piles if any analysis is meaningful, so attaching to the steel should be equivalent.
You can always cut windows later if you don't like what you see from attaching to steel.

The dynamic test should be performed after time to allow sufficient concrete strength. Attach
transducers to the pile a very minimum of one diameter below the top of the shaft. Preferably
this attachment would be two diameters below the top, but for large diameters this can create
a hardship. Since many shafts are constructed flush with the ground surface, a considerable
excavation could be required and the one diameter location is then more attractive; we suggest
the four strain transducer attachment system for those with a PAK as a compromise.

As a good alternative, the shaft to be tested could be "built up" above the ground surface. A
steel casing provides the form as well as strengthening the shaft top to withstand the impact.
It is good to choose a transducer location which is on the actual shaft; attaching near (just
above or just below) cross section changes, including casing and high strength concrete, can
complicate the testing and analysis; the larger the nonuniformity the farther below the change
you should locate the transducers to avoid "end effects" and assure uniform stress distribution.
Ideally, the shaft at and below the measuring location will be uniform in all respects, so a small
excavation even if the shaft has been extended above the ground surface is generally
recommended. The built up section can be removed after the test has been completed. If the
concrete for the buildup section is cast at same time as the shaft (that is our ideal
recommendation), then the concrete in the buildup will have the same properties as the shaft;
removing the lower portion of the steel casing and attaching the strains to the smooth concrete
simplifies shaft preparation.

More recent alternatives would include instrumenting the ram (assumed to be single piece, or
at most two pieces) with an accelerometer and using F=ma to measure the “force” (as special
versionm of PDA-W is then required for this option), or inserting a real force measuring “top
transducer”. In both cases the accelerometer measurement is then made on the pile pile head
near the pile top, perhaps one half diameter below the top. Contact PDI for details about either
of these options.

The ram to be used for testing should be guided to assure the pile top and ram bottom are
parallel during impact to make as uniform an impact as possible. A short section of pile leads,
or an external guide tube or cage are two alternatives. The ram weight required is generally
at least one percent of the ultimate load to be tested; for shafts with a high amount of end
bearing a minimum of 1.5 to 2 percent of the ultimate load is considered more desirable. And
if you get up to about 5% of the test requirement, then the dynamic test also satisfies the
requirements for a rapid load test. Higher ram weights are generally always acceptable, but
become more difficult to achieve as shaft diameter and required loads increase. The ram is
ideally a simple drop weight which can be raised to a variable height of drop. Rams using both
steel and concrete have been constructed. Ram weights up to 77tonnes (770 kN) are known.
Rams can be made in segments which are assembled (e.g. bolted together) at the job site;
segmental weights help in transporting, or in lifting the weights prior to the free fall release
described below.

The shaft top to be tested is cushioned by a few layers of plywood placed directly on the
concrete. The diameter of the plywood cushion should be about 80 (to at most 90) percent of
the shaft diameter to centralize the blow. A steel striker plate of 50 mm to 100 mm thickness
is sometimes placed above the plywood and a diameter equal to the cushion diameter. The
striker plate should be avoided if the F=ma option is selected.

The test should be started usually with a relatively low drop height (say 1.5 ft or 500 mm). The
permanent set for each and every blow should be carefully measured. It is best if the ram is
a complete free fall (e.g. not connected to crane by cable). Free release mechanisms of
several types have been used in many different countries. The ram can be supported from a
frame by a cable link. Cutting the cable then allows a free drop. Perhaps the better way is to
use hydraulic clamps to hold, and then release, the ram.

The uniformity of stresses can be checked, and adjustments to cushion thickness or ram/shaft
alignment made before application of more blows. Subsequent blows can have a higher drop
height if stresses are acceptable, and if the ultimate capacity was not reached because the set
per blow was small (say less than 3 mm per blow). Higher drop heights are not needed if the
client is satisfied that a certain minimum load was achieved, and this perhaps depends on the
safety factor desired for the test.

It will be helpful in the analysis to know the total volume of concrete used in the shaft; you can
compare with the theoretical volume and may help model the nonuniform pile during CAPWAP.
In many cases, apparent shaft impedances are higher than for the theoretical concrete volume
alone. All test results should be subjected to a CAPWAP analysis, and it may be helpful (some
cases required) to use the "radiation damping model" to achieve reasonable capacity solutions.
CAPWAP now has a multiple blow analysis which can be applied to drilled shafts.

This brief overview is intended to give you an overview of how to approach the testing of this
particular pile type. Since each site and each contractor has different capabilities and
restrictions, it is impossible to cover all possible alternatives in this technical note. We do offer
advice in reply to your specific needs; contact Pile Dynamics if you need further assistance
getting started in this field of testing.


Many factors enter into a successful test. Good communication and cooperation between
all parties (engineer, contractor, owner, and PDA test engineer) is the foundation. By
appropriate advance planning, a good test program can be "designed" which should answer all
the main questions for the engineer. Depending on project size, soil type and site variability,
a test could be performed in one site visit or may need several days and second visits for
restrike tests. Ideally sufficient information will be gathered but the engineering effort must be
balanced so that the contractor is not overly burdened and the owner's cost is kept reasonable.
Good communication is essential if the PDA engineer is to arrive at the test site at the proper
time. “Remote testing” is another alternate to cost containment.

It is observed that often those who succeed are those who want to succeed. Some have
used the expression "technology champion" for this engineer. If the engineer is enthusiastic
and has the desire, he will usually find a way to make it work; he will apply the methods in the
correct way and interpret results consistently and with reason (making effort to account for
capacity changes with time or assuring blow counts are less than refusal). If the engineer is
reluctant and only performs the test because someone else ordered it, then chances of a good
result are then considerably diminished. It boils down to having a good attitude.

It can further be said that those who are successful PDA users take the time to understand the
principles and theory of dynamic wave travel. They read manuals and understand proper
application. They make an effort to perform the right test procedure and perform it at the proper
time in the project. There is now in recent years a growing concern about quality of the testing.
As a result, a “PDA certification examination” is now offered by Foundation QA
( and administered in the USA by PDCA ( This
exam rates the ability of the tester in several areas, with ratings of Basic, Intermediate,
Advanced, Master and Expert. Many authorities now require certification for the tester to be
qualified to preform testing services, and some organizations require higher ratings (e.g.
Advanced or above). Pile Dynamics endorses this certification process.

PDA testing is not just to be applied as a “cook book” manner to fulfill the project specifications.
There is a reason for the test request. Failing to understand the goals of the test will often
lead to disappointing results and lack of future credibility.

Prior to going to the job, the engineer should inventory his equipment. Currently the latest PDA
model (PAX) is battery powered, and the wireless transmitters are also battery powered. It
should be obvious that the PAX and wireless transmitter batteries should be fully charged prior
to use. If a transducer is found to give unreliable data, it should be repaired or replaced. (If you
have bad data from a transducer on the previous job, at the end of that project do something
about it then since chances are it will not work any better the next time). I recall about 30
years ago when on a site, I experienced transducer problems and although I replaced
everything eventually I had no more fresh sensors to try and was told by my client to "keep
smiling" when the owner came around; after the job was finished, however, I still had to write
the report. With bad data, report writing is difficult (impossible?) and if you cannot write a
report, you may not get paid! Remember that if the data is bad, the test is worthless, so make
every effort (exchange sensors and cables) to achieve good data. Also points out that you
need to have sufficient spare sensors and cables. Going to the site with only the bare
minimum number of sensors and cables is an invitation to disaster. At best you can call

and have extra shipping cost to get overnight replacements (which adds cost and delays the

The engineer must make every effort to arrive at the job site with everything he needs (PDA,
cables, transducers, tools, bolts, safety equipment ....). He may rely on the contractor to supply
some items such as a tack-hammer or wrench, but communication is then very important.

The equipment should be assembled and checked upon arrival at the site; if a problem is found
or something missing, you then have a chance to fix the problem or find replacements. Attach
the transducers and verify their performance by tapping them. If the transducers don't work
then, chances are they won't work attached to the pile either. This has saved much
embarrassment and hours of contractor time. I can assure you that the contractor does not
appreciate you sorting out your "problems" while on his time clock. Quality data should be our
primary goal.

Prepare the pile in advance. Drill the pile on the ground and consider your safety. It should be
noted that currently battery powered drills are available and preferred, and make you
independent of on-site power generators and extension cords. A GRL engineer was once
asked to stand directly under and drill a large concrete pile while it was lifted only with the crane
(no cribbage to support pile). Think what would happen if the brakes on the crane failed and
crushed that engineer! Fortunately, our engineer properly refused. Lifting and setting the pile
in position is perhaps the most dangerous activity on the piling site; I have lost track of how
many injuries (and even deaths) I have heard about. Stay clear of objects which can
potentially fall on you. I vividly recall a massive piece of steel breaking off a hammer and
landing exactly where I had been standing only seconds before. Our engineers have had
several close encounters with headache balls that have fallen (and in some cases smashed our
equipment cases). A little common sense should allow you to return home injury free at the end
of the day.

When asked to drill the pile after it has been lifted and the pile top is then say 50 ft (15 m) up
in the air, we now refuse to "drill in the leads". We suggest just waiting and test the next pile
which can then be properly prepared on the ground (or use a proper lift basket if it must be
tested). Chances are if you think something is dangerous, OSHA considers it a violation (e.g.
it will cost you money) and could result in injury. Safety is your first priority. THINK at all times.

We also suggest that you show a member of the piling crew how to bolt your transducers
to the pile and let him do all the climbing (if necessary). With the new wireless transmitters,
the piles can be prepared on the ground, sensors attached by the engineer prior to lifting the
pile, strain offsets checked (and adjusted if needed), and then sensors covered by rigid foam
“sensor protectors”. Then the pile with protected sensors can be lifted into place with almost
complete confidence they will survive and also perform properly (because the engineer himself
verified the attachment). For larger projects, or where time is critical, the contractor personnel
can be instructed how to drill and prepare the pile, even in advance of arriving on site. This
allows the testing engineer to concentrate on getting good data and its interpretation rather than
spending time preparing the pile for the test. Contractor personnel are very skilled and
generally can do pile preparation easily.

Make sure you get as much site information as possible. What is the hammer? Hammer
cushion? Pile top cushion for concrete piles? Obtain copies of appropriate soil borings and site

plan. Get copies of the driving logs (blow count records). Collect load test data if available.
Know the dates of installation for each restrike pile tested, and also for the static test. It is
much easier to get information before you leave the job site. The most important site parameter
is the set per blow ("blow count") often counted as "blows per ft, blows per inch, or blows per
meter,....set per 10 blows,..." This is perhaps your assurance that the capacity is fully
mobilized. It is also a value used in the CAPWAP process. For restrikes, draw a reference line
on the pile, drive the pile 5, 10 or 20 blows, draw a second line using the same reference, and
finally measure the total permanent pile movement. For restrikes, do not simply get the blow
count for the entire first foot of restrike (note that 20 blows or even less is usually sufficient for
a restrike); try to get blows per inch for each inch, or total set for a limited number of blows.
Assess the accuracy of the set per blow. Is the blow count changing at the end of drive? If a
restrike, is the first blow with low energy? Or is the capacity changing quickly? How reliable
was the displacement measured? How accurately were the "inches" drawn (or were they
"guesstimated" as the pile was being driven)?

Measure the pile yourself (length and cross section area) and see if that agrees with the section
you are told as the force (and capacity and energy ...) are all directly proportional to the pile
cross sectional area. Determine the length below gages accurately as this is essential
information also. For concrete or timber piles, determine the wave speed. For timber piles, get
the density of each pile tested by actual weight/volume measurements. Calculate the elastic
modulus ( PDA-W does it for you); check and see if this value is reasonable. Enter all
information or PDA settings.

Be careful that data storage is sufficient. Save every blow (either for restrike or during driving,
since the storage on the PDA large hard disk is not restrictive). We strongly recommend you
use the Windows based PDIPLOT and PDI -Curves programs to summarize your data and
satisfy ASTM D4945 reporting requirement. PDIPLOT eliminates the need to save results in
the SQ file.

Make sure you get all the PDA data needed (end of drives and restrikes). If possible (and if you
have CAPWAP) do at least a preliminary analysis of some data to assure yourself there are no
"surprises" later. You should have a reasonable number of CAPWAPs for every job; we
suggest that as a rule-of-thumb that there be about 2 to 5 CAPWAPs for every 10 piles tested.
CAPWAP important data sets such as restrikes or end of drive, after a splice, or possible
bearing layers.

Ideally, everyone know how to proceed before you leave the site. Again, good communication
is the key here. Special precautions or instructions for production piles must be clearly stated.
The report should be written as soon as possible. You remember site details while they are
still fresh in your mind. The project needs your report so they can make crucial decisions (if you
annot get the report early, at least give summarized results and verbal communication so the
project is not delayed) The ASTM D4945 specification lists many items which should be
included in this report. Site details are often described for the permanent record. All CAPWAP
or PDA results should be clearly and concisely presented.

Technical Note: Teflon Pile Training Aid

The PDA can be demonstrated without having a full scale pile test. This simplifies training.
Attach the strain and acceleration sensors to a TEFLON rod (like the PIT demonstration rods).
For example, PDI has put sensors on a teflon rod about 90 mm below the top of a rod which is
400 mm in total length. The length is not very important (longer helps in entering pile length,
but shorter like 400 mm is easy to carry and store; Teflon deforms plastically if not supported
during storage on a flat surface for its entire length. Teflon rods are also relatively expensive
and big ones are heavy).

The diameter of the PDI teflon test rod is about 32 mm. Because of this small diameter, our
smaller strain sensors (51 mm or 2 inch hole spacing, instead of the normal 76 mm or 3 inch
hole spacing) were attached with a clearance hole and long bolts and nuts (you cannot really
use tapped holes). Probably the normal strain transducers will work for larger diameter teflon
rods. Accelerometers are attached in a similar way.

You can enter the correct area for the Teflon rod. The approximate density SP for Teflon is
21.7 kN/m3 (2.17 T/m3 or 0.135 K/ft3). The wavespeed WS for Teflon is about 800 m/s (2625

You can use a standard PIT hammer to apply the impact. You need to take some care in
applying the impact as bending can be significant (can view this on PDA-W with the DPFV
display option). The bottom end can be free if just hand held, or fixed if it is placed on the floor
to model different conditions and this may help the user to understand a little wave theory.

This information can be of value to you in demonstrating your PDA to a client, or for training
new engineers. Please contact PDI if you are interested in having PDI send you small diameter
Teflon rods. Of course, you could obtain the Teflon rods yourself. If the rod diameter is small,
then you probably need the small mini strain transducers (available from PDI).























Dynamic Pile Testing
Dynamic pile testing has become routine practice for many practicing foundation engineers. Many
major codes (National and International), specifications, publications from deep foundation trade
organizations, proceedings from International conferences, and civil engineering
handbooks/textbooks are listed below which attest to the widespread usage of dynamic pile testing
and wave equation analysis. The need for dynamic pile testing results from the increased need for
quality control due to today's increased foundation loads.

Cost Impact:

The cost of implementing dynamic pile testing is negligible in many parts of the United States since
both wave equation analysis and dynamic pile testing are already in common usage. Use of these
modern methods has probably resulted in fewer foundation failures, reduced claims because
measurements have replaced guesses and assumptions, and improved quality at reduced overall
testing costs (dynamic tests can be applied to a higher percentage of piles since they are significantly
less expensive than static load tests). For locations where wave equation and dynamic testing have
not yet been implemented, these modern techniques will assist in raising allowable loads, leading to
more economic foundations, and offer substantial cost savings compared with current practice.


While it would be impossible to list all sources, the following several of major sources which
prominently reference wave equation analysis and dynamic pile testing.

a) ASTM D4945-96 "Standard Test Method for High Strain Testing of Piles"
This standard consensus document D4945 describes a proper dynamic pile test in a similar
way that ASTM 1143 describes a proper static load test. Several code and standard
organizations have adopted the basic outline of this Standard in their own efforts, including
AASHTO (see below).

b) American Society of Civil Engineers "Standard Guidelines for the Design and Installation of Pile
Foundations", ASCE 20-96. (1997)
To quote from the Abstract from this consensus process document, "This Standard provides
a guideline for an engineering approach to the design and subsequent installation of pile
foundations. The purpose is to furnish a rational basis for this process, taking into account
published model building codes and general standards of practice." Dynamic testing and
wave equation analysis are featured.

c) US Army Corps of Engineers "Design of Pile Foundations", Engineering Manual (EM 110-2-2906).
To quote from the Abstract, ".. this book is based on the present state of technology for pile-
soil-structure-foundation interaction behavior." This manual represents current
recommended practice for the US Army Corps of Engineers and prominently features
dynamic testing and analysis.

d) U.S. Department of Transportation, Federal Highway Administration "Design and Construction of

Driven Pile Foundations." Publication No. FHWA-HI-96-033, two volumes, (1996).
This manual (over 1000 pages) is used for workshops being presented through FHWA to
numerous State departments of transportation. Volume I covers design. Volume II, covering
installation and inspection, reflects FHWA recommendations and devotes over 120 pages
to wave equation analysis and dynamic pile testing. Copies are available from PDCA and
the National Highway Institute. Numerous State departments of transportation have adopted
these provisions in their State codes.

e) American Association of State Highway and Transportation Officials- "AASHTO", "Standard

Method of Test for High Strain Dynamic Testing of Piles." AASHTO Designation T 298-93 (1993)
A consensus standard developed by several State highway engineers active in dynamic pile
testing. This standard reflects the acceptance of dynamic testing. Numerous states have
their own codes specifying wave equation and dynamic pile testing, but are too numerous
to list here (DFI has compiled all State specifications into a massive two volume reference).

f) American Association of State Highway and Transportation Officials- "AASHTO", "Standard

Specifications for Highway Bridges", sixteenth edition 1996.
Contains provisions for dynamic testing and wave equation analysis. Economic incentives
included show reduced safety factors when both wave equation and dynamic measurements
are made in conjunction with static tests.

g) Deep Foundations Institute "Inspector's Manual for Driven Pile Foundations." Second Edition,
Copyright 1997
Deep Foundations Institute, the most widely known trade organization in the United States,
furnish literature reflecting current practice as compiled in a consensus process. A recent
Manual from DFI presents current practice and recognizes the dynamic testing and analysis.

h) Pile Driving Contractor's Association (PDCA) "Design Specifications for Driven Bearing Piles."
(June 4, 1999)
This active organization has prepared their own model piling Specification through a
consensus review process. This current draft reflects the use and acceptance (including
economic incentive and lowered safety factors) for performing wave equation and dynamic
pile testing.

i) Proceedings from the 5th International Conference on the Application of Stress Wave Theory to
Piles, Orlando 1996.
This conference had a 70 member International organizing committee and is held every four
years. The 1200 page proceedings contain 102 papers by authors from 24 countries on 6
continents and summarizes worldwide dynamic pile testing.

j) "Canadian Foundation Engineering Manual", 3rd Edition (1992)

This manual makes extensive reference to dynamic pile testing and wave equation. Codes
from several individual countries in Europe, Asia, and South America already include
dynamic pile testing.

k) "Standard Handbook for Civil Engineers", fourth edition, by Frederick Merritt et al, Copyright 1996.
Pile section includes dynamic monitoring and wave equation analysis. Most recent
textbooks also reference dynamic methods of analysis and testing.

l) “International Building Code”, to become effective April 2000.

Section on pile testing 1807.2.8.3 references testing by ASTM D1143 (static) or D4945
(dynamic) methods. This code will become the official building code for the United States and
is supported by BOCA, SBC and ICBO.