Composite M Book

COMPOSITE
MATERIALS
By:
Dr. Mark V. Bower, P. E.
Copyright 1992-2000
The University of Alabama in Huntsville
Huntsville, Alabama

ii
Composite Materials
I NTRODUCTI ON...........................................................................................................1
I ntroduction................................................................................................................1
Definition of Terms.....................................................................................................1
CONSTI TUENTS AND FABRI CATI ON TECHNI QUES............................................8
Constituents................................................................................................................8
Reinforcement Forms...............................................................................................17
Fabrication Techniques............................................................................................21
Hazards.....................................................................................................................42
LAMI NA MECHANI CS...............................................................................................48
Preliminaries............................................................................................................48
Constitutive Relations..............................................................................................53
Engineering Properties for Orthotropic Materials..................................................57
Plane Stress Orthotropic Constitutive Relation .....................................................58
Off-Axis properties of orthotropic lamina................................................................60
STRENGTH OF LAMI NA ...........................................................................................65
Tensor Polynomial Failure Criterion.......................................................................65
Quadratic Failure Criterion.....................................................................................66
R-Factor analysis......................................................................................................73
CLASSI CAL LAMI NATI ON THEORY .......................................................................75
History ......................................................................................................................75
Preliminaries............................................................................................................76
Force -- Moment Resultants.....................................................................................77
Equilibrium of a Plate Element ...............................................................................79
Displacement Field Model........................................................................................83

iii
Orthotropic Constitutive Relation...........................................................................86
The Laminated Plate Equations..............................................................................91
LAMI NATES................................................................................................................97
I ntroduction..............................................................................................................97
Alternate Expressions for Laminate Stiffnesses.....................................................97
Simplifying Assumptions on Laminate Structure..................................................99
Stress Distribution in a Laminate.........................................................................112
Laminate Failure Theories ....................................................................................113

iv
Preface
This document is a work in progress, as are most books. Every effort has
been made to ensure the accuracy of the information contained in it. That does not,
unfortunately, guarantee that every equation is without error. Having said this,
the reader is encouraged to consult other texts on composite materials. A few are
listed on this page. Further, neither the author nor The University of Alabama in
Huntsville is responsible for the application of the information contained in this
document. Good engineering practice requires the application of sound engineering
judgment.
The author acknowledges the support of The University of Alabama in
Huntsville, the Microsoft Academic Support Program, and the Dell Corporation
Academic Support Program. I acknowledge the support and assistance of my wife,
Peggy, and children, Renae, Amber, Elizabeth, and Matthew. Further, I
acknowledge the support, inspiration, and anointing of J esus Christ. Without His
help, I could not have come this far.
References:
1. Mechanics of Composite Materials, R. M. J ones, McGraw-Hill Book
Company, Washington, D. C., 1975.
2. Primer on Composite Material: Analysis, J . E. Ashton, J . C. Halpin, and
P. H. Petit, Technomic Publishing Co., I nc., Westport, CT, 1969.
3. Introduction to Composite Materials, S. W. Tsai and H. T. Hahn,
Technomic Publishing Co., I nc., Westport, CT, 1980.
4. Fundamentals of Composites Manufacturing: Materials, Methods and
Applications, A. B. Strong, Society of Manufacturing Engineers Dearborn,
MI , 1989.

Revised: 10 February, 2000 Page 1
INTRODUCTION
I NTRODUCTI ON
A composite material is defined as a material composed of two or more
constituents combined on a macroscopic scale by mechanical and chemical bonds.
Typical composite materials are composed of inclusions suspended in a matrix. The
constituents retain their identities in the composite. Normally the components can
be physically identified and there is an interface between them. Composite
materials are classified based on the shape and relative dimensions of the inclusion
and the structures. Composite materials are classified as:
Particulate
Filamentary
Laminated
I n a particulate composite, the major dimension of the inclusion is small compared
to the structural dimensions. Particulate composites may be made with small
particles, such as glass beads, or with chopped fibers. I n filamentary composites,
one dimension of the inclusion is of the same order of magnitude as the structural
dimensions. Filamentary composite materials may be made from uni-directional
tape or cloth. I n laminated composite materials, two of the major dimensions of the
inclusions are of the same order of magnitude as the structural directions.
Sandwich sections are examples of a laminated composite material.
Two additional distinctions are made in the classification of composite
materials: advanced composite materials are those composites which are made with
inclusions that have a modulus greater than that of steel (30 Mpsi, 207 GPa) and
volume fraction of inclusions greater than fifty percent, and hybrid composite
materials are those composites which are made with two or more different inclusion
materials.
DEFI NI TI ON OF TERMS
ANGLEPLY LAMINATE
Containing plies alternately oriented at plus and minus a fixed angle other
than 90 degrees to the reference direction.
COMPOSI TE MATERI ALS 2
I NTRODUCTI ON
Revised: 10 February, 2000 M. V. Bower
ANISOTROPIC
Not isotropic; exhibiting different properties when tested along axes in
different directions.
AUTOCLAVE
A pressurized heated chamber used to cure composite materials. An
autoclave is pressurized with gas, typically air or nitrogen.
BALANCED LAMINATE
A composite laminate whose lay-up is symmetrical with relation to the mid-
plane of the laminate.
BLEEDER CLOTH
A nonstructural layer of material used in manufacture of composite parts to
allow the escape of excess gas and resin during cure.
B-STAGE
An intermediate stage in the polymerization reaction of certain
thermosetting resins; the state in which most prepregs are stored and
shipped.
CAUL PLATE
A smooth metal plate used in contact with the lay-up during cure to transmit
normal pressure and to provide a smooth surface to the finished laminate.
COLLIMATED
Rendered parallel, applies to filaments.
COUPLING AGENT
That part of a sizing or finish, which is designed to provide a bonding link
between the reinforcement and the laminating resin.
CRAZING
Fine resin cracks at or under the surface of a plastic.
I NTRODUCTI ON
CROSSPLY LAMINATE
Containing lamina alternately oriented at 0 degrees and 90 degrees.
CURE
To irreversibly change the properties of a thermosetting resin by chemical
reaction, i.e., condensation, ring, closure, or addition. Cure may be
accomplished by addition of curing (cross-linking) agents, with or without
heat.
DELAMINATION
The separation of the layers of material in a laminate.
DRAPE
The ability of broadgoods to conform to an irregular shape.
ELONGATION
The amount of deformation of the fiber caused by the breaking tensile force,
expressed as the percentage of the original length.
FIBER PLACEMENT
An automated fabrication process in which the machine places fiber bundles
along predetermined paths to build up the structure.
FILAMENT
A long, continuous length of fiber, measured in yards.
FILAMENT WINDING
An automated fabrication process typically used to produce cylindrical or
spherical shape. The machine winds fiber bundles onto a mandrel that is
removed after the cure process.
FILL
Yarn running from selvage to selvage at right angles to the warp in a woven
fabric.
I NTRODUCTI ON
FILLER
A relatively inert material added to a plastic to modify its strength,
permanence, working properties, or other qualities, or to lower costs.
FINISH
A mixture of materials for treating glass fibers. I t contains a coupling agent
to improve the bond of resin to glass; and usually includes a lubricant to
prevent abrasion and a binder to promote strand integrity. With graphite or
other filaments, it may perform either or all of the above functions.
FLASH
Excess plastic material which forms at the parting line of a mold or which is
extruded from a closed mold.
GEL COAT
A quick-setting resin used in molding processes to provide an improved
surface for composites; it is the first resin applied to the mold after the mold-
release agent.
HAND LAY-UP
The process of placing and working successive plies of the reinforcing
material or resin impregnated reinforcement in position on a mold by hand.
HYBRID COMPOSITE
A composite structure composed of more than two different materials, for
example, a laminate with outer laminae of glass/epoxy and inner laminae of
graphite/epoxy.
HYDROCLAVE
Similar to an autoclave except that the chamber is pressurized using heated
water or other liquid.
INTERLAMINAR SHEAR
The shear strength at rupture in which the plane of fracture is located
between the layers of reinforcement of laminate.
I NTRODUCTI ON
ISOTROPIC
Having uniform properties in all directions. The measured properties of an
isotropic material are independent of the axis of testing.
LAMINA
A single layer or ply of material. The fundamental building block of a
laminate.
LAY-UP
A laminate that has been assembled, but not cured; or a description of the
component materials and geometry of a laminate.
NON-WOVEN FABRIC
A fabric, usually resin-impregnated, in which the reinforcements are
continuous and unidirectional; layers may be crossplied.
ORTHOTROPIC
Having three mutually perpendicular planes of elastic symmetry.
PARALLEL LAMINATE
A laminate of woven fabric in which the plies are aligned in the same
position as originally aligned in the fabric roll.
PLASTICIZER
For epoxy, a lower molecular weight material added to reduce stiffness and
brittleness; it results in a lower glass- transition temperature for the
polymer.
PULTRUSION
A fabrication process used to produce a highly collimated composite shape
(rod, bar, etc.).
POSTCURE
Additional elevated temperature cure, usually without pressure, to improve
final properties and/or complete the cure. I n certain resins, complete cure
and ultimate mechanical properties are attained only by exposure of the
cured resin to higher temperatures than those of curing.
I NTRODUCTI ON
POT LIFE
The length of time that a resin system retains viscosity low enough to be used
in processing.
PREPREG; PREIMPREGNATED
A combination of mat, fabric, non-woven material, or roving with resin,
usually in the B-stage, ready for molding.
QUASI-ISOTROPIC LAMINATE
A laminate approximating isotropy by orienting plies in several directions.
RESIN TRANSFER MOLDING (RTM)
A manufacturing process used to produce large composite structures. I n this
process, a dry lay-up is infused with resin in a molding process. May be
found in various forms such as vacuum assisted resin transfer molding
(VARTM) or Seeman Composite Resin I nfusion Molding Process (SCRI MP).
Not in wide spread use for advanced composites.
ROVING
A multiplicity of single ends of continuous filament with no applied twist
drawn together as parallel strands.
STACKING SEQUENCE
The sequence of angles and possibly materials that describes the orientation
of the individual lamina in a laminate from top to bottom, e. g.,
+45/-45/+45/-45, or 0/90/90/0, or 0/+60/-60/0/+60/-60/-60/+60/0/-60/+60/0.
Most laminates are composed of a large number of laminae, frequently in
repeated patterns, which leads to the use of shorthand notation. Using
shorthand notation the first example is written: 2[t45]. The second and third
sequences are symmetric about the mid-plane, and thus can be written:
[0/90]S and 2[0/t60]S, where the subscript S indicates symmetry.
SIZING
On glass fibers, the compounds which, when applied to filaments at forming,
provide a loose bond between the filaments, and provide various desired
handling and processing properties.
I NTRODUCTI ON
SYMMETRIC LAMINATE
A laminate that is symmetric in both geometry and material properties about
the mid-plane.
TACK
With prepreg materials, the degree of stickiness of the resin.
TAPE LAYING
An automated process in which the machine lays a composite tape, either wet
or prepreg, on predetermined paths.
TOW
A loose, untwisted bundle of filaments.
TRANSVERSELY ISOTROPIC
Having uniform properties in one plane. The measured properties of a
transversely isotropic material are independent of the axis of testing within
the plane.
UNI-DIRECTIONAL LAMINATE
A laminate with non-woven reinforcements that are all laid up in the same
direction.
WARP
The yarn running lengthwise in a woven fabric.
WET LAY-UP
A reinforced plastic which has liquid resin applied as the reinforcement is
being laid up.
CONSTITUENTS AND
FABRICATION TECHNIQUES
For the purpose of this discussion, composite materials are defined as a
marriage of two or more constituent materials on a macroscopic scale. To
understand the fabrication techniques associated with composite materials it is
important to discuss the types of constituent materials and the fabrication
techniques used to produce composite structures.
CONSTI TUENTS
MATRIX MATERIALS
Polyester Resins
Polyester is a thermoset polymer that is formed from a condensation
polymerization. Polyester has been widely used in commercial applications with
fiberglass. Applications include:
Boat hulls,
Shower stalls,
Bath tubs,
Car bodies,
Building and roof panels,
Molded furniture, and
Pipes.
Advantages for the use of polyester resin include:
Low cost (generally lowest found in composite materials) and
A wide assortment of diacids and diols can be used to give physical and
chemical properties.
Disadvantages for the use of polyester resin include:
Poor temperature capabilities,
Poor weather resistance,
Shelf life may be limited, and
Poor mechanical properties (stiffness and strength) as compared to
advanced composites.
CONSTI TUENTS AND FABRI CATI ON TECHNI QUES
Epoxy Resins
Epoxy is a thermoset polymer that forms a strong rigidly crosslinked network
of polymer chains. Epoxy has been widely used in commercial applications with
fiberglass, graphite, and aromatic fibers. Applications include:
Aircraft components,
Pressure vessels,
Rocket motor cases, and
Car bodies.
Advantages for the use of epoxy resin include:
Excellent adhesion
Excellent mechanical properties (strength and stiffness),
Excellent chemical resistance,
Excellent weather resistance,
Low shrinkage,
Good fatigue strength,
Good corrosion protection, and
Versatility in processing.
Disadvantages for the use of epoxy resin include:
Poor high temperature capabilities,
Uncured resin is toxic,
Poor handling properties (uncured), and
Relatively expensive.
Epoxies are available in multi-component and single component systems.
The cure of epoxy may be through the application of hardeners, a catalytic agent
that activates or facilitates crosslinking between the polymer chains, (a two-part
system), or through the application of heat or ultra-violet light (a one-part system).
Epoxies may be stored at freezer temperatures, which prompts long storage/shelf
life. Wide ranges of cure cycles are available.
Polyimide and Polybenzimidaole Resins
Polyimide and polybenzimidaole (PBI ) are thermoplastic polymers with
excellent high temperature (600 to 700) properties. Polyimide and PBI has been
used in commercial applications with graphite, and aromatic fibers. Applications
include:
Aircraft components.
Advantages for the use of polyimide and PBI resin include:
Excellent thermal properties, and
Good processability on conventional molding equipment.
Disadvantages for the use of polyimide and PBI resin include:
Variable resistance to solvents depending on specific compound,
Difficult synthesis process,
Difficult fabrication methods,
Resin is toxic and should be handled with great care,
Expensive raw materials, and
Very expensive (more than epoxies).
Use of Polyimide and PBI compounds is growing as the knowledge base
increases.
Phenolic Resin
Phenolic is a thermoset polymer with good high temperature properties.
Phenolic has had a long history of commercial applications as a general
unreinforced plastic and is now being used as a composite resin with graphite, and
aromatic fibers. Applications include:
Rocket nose cones and nozzles, and
Automotive applications.
Advantages for the use of phenolic resin include:
Good mechanical properties (strength and stiffness),
Good thermal properties with an ablative nature, and
Good processability.
Disadvantages for the use of phenolic resin include:
Absorbs moisture easily,
Brittle behavior, and
Relatively expensive (more than epoxies).
Carbon Matrices
Carbon matrices are produced from polymeric resins that are carefully
charred in a processed called pyrolysis. Carbon matrices may also be produced by
vapor deposition, but the process is limited to structures less that 3/16 thick.
Applications include:
Rocket nose cones and nozzles, and
Automotive applications, especially brake components.
Advantages for the use of carbon matrices include:
Very high specific heat capacity (highest known),
Good mechanical properties (stiffness and strength),
Good toughness,
Good resistance to shock,
Excellent thermal properties, and
Excellent thermal stability.
Disadvantages for the use of carbon matrices resin include:
Absorbs moisture easily,
Poor wear resistance, and
Very expensive (more than five times the cost of a phenolic ablative
composite).
Thermoplastics Matrices
Thermoplastic polymers have long chain molecules that are loosely
interconnected by weak chemical bonds and mechanical tangling. Because of the
structure of thermoplastic polymers they do not require reactive cure cycles or have
a distinct melting temperature, displaying fluid like (viscoelastic) behavior at even
room temperature. Consequently, these materials lend themselves to molding
processes. Thermoplastics include:
Polyethylene,
Nylon,
Polystyrene,
Polyester,
Polycarbonate,
Polyvinylchloride (PVC),
Acrylonitrile butadiene styrene (ABS),
Acrylic,
Polyethylene terephthalate (PET),
Polyetheretherketone (PEEK),
Polyphenylene oxide, et cetera.
Advantages for the use of thermoplastics resin include:
Large number of processing methods,
Lower fabrication times, (compared to thermosetting polymers),
Good compression strength after impact,
Good hot/wet compression strength,
Resistant to moisture absorption, and
Easy dyed or given special properties (e.g. flame retardant).
Disadvantages for the use of thermoplastic resins include:
High viscosity impairs wet-out of reinforcement,
High consolidation pressures are required, and
Mechanical, chemical, thermal, and electrical properties depend on
specific selected.
Ceramic Matrices
Organic resins are characterized by the prevasive presence of covalent bonds.
Ceramic matrices, by contrast, are characterized by the predominance of ionic
bonds, however silicon-carbide (SiC) has covalent bonding. Ceramic solids may be
crystalline, vitreous (glass-like), or mixed.
Advantages for the use of ceramic matrix include:
Dimensionally stable at high temperatures,
High chemical stability,
High thermal stability,
Applicable to extreme temperatures (2000 to 4000).
Disadvantages for the use of ceramic matrix include:
Very brittle,
Very high consolidation pressures are required, and
Very expensive to produce and maintain.
Phenolic and carbon matrice are sometimes classified as ceramic matrices.
Metal Matrices
I n comparison to organic resins and ceramic matrices, metal matrix
composites (MMCs) are characterized by the predominance of metallic bonds. I n
MMCs discontinous or continuous metal fibers are suspended in a matrix of a
differing metal (e.g. aluminum, titanium, magnesium, copper, et cetera).
Advantages for the use of metal matrix include:
Outstanding mechanical properties (stiffness and strength) for
continuous fiber MMCs,
Good wear resistance,
Thermally conductive,
Good fracture toughness for continuous fiber MMCs,
Good fatigue strength for continuous fiber MMCs,
Applicable to extreme temperatures (2000 to 4000).
Disadvantages for the use of metal matrix include:
Significant difficulties associated with the inherent non-wetability of
fibers,
Very high consolidation pressures are required, and
Very expensive to produce.
REINFORCING MATERIALS
While there is no restriction as to the material used as the reinforcing
element for modern composites there are generally three materials that are
commonly used: glass, graphite, and organic. These materials are discussed in this
section.
Glass Fibers
Glass fibers have long been used as reinforcing elements. Owens-I llinois and
Corning Glass developed a fiberglass manufacturing facility in 1937. Glass is
produced from silica sand, limestone, boric acid, and other elements. Types of glass
include:
E-glass,
S-glass (and the variation S2-glass),
C-glass, and
Quartz.
These are the four primary types of glass used in composite materials. The
type of glass is defined by the chemical composition
Advantages for the use of glass fibers include:
Applicable to wide range of geometries and sizes,
Seamless construction,
Good strength and durability,
Lower tooling costs,
I ncreased design flexibility,
Minimal maintenance, and
Corrosion resistant.
Disadvantages for the use of glass fibers include:
Mechanical properties are not as good as metals or other reinforcing
fibers.
The fiber glass production processes are shown in the following figure. Note
that the process may begin from stock (marbles) or directly from melt. The use of
stock has had better control over the properties.

Fiber glass production processes,
(a) marble process and (b) direct-melt process.
Carbon/Graphite Fibers
Carbon or graphite fibers for structural applications began production in
significant quantities in the 1950s. Graphite fibers are among the highest stiffness
and highest strength material known today. Types of graphite fibers include:
Polyacrylonitrile (PAN)-Based Fibers
Pitch-Based Fibers
Rayon-Based Fibers
Advantages for the use of graphite fibers include:
Excellent strength,
Excellent stiffness,
Excellent specific strength and stiffness, and
Disadvantages for the use of graphite fibers include:
Significantly more expensive than glass fibers, and
Brittle behavior.
Two graphite fiber production processes (PAN-based and pitch-based) are
shown in the following figure. Note that both processes use a two step
carbonization/graphitization process to convert the raw fiber into graphite.
Characteristic properties for graphite fibers from the three processes is listed in the
following table.

Graphite fiber production processes,
(a) PAN-based process and (b) Pitch-based process.

Carbon Fiber Mechanical Properties
4

PAN-based fibers
Low Modulus High Modulus
Tensile Modulus (Mpsi) 33 56
Tensile Strength (kpsi) 480 350
Elongation (%) 1.4 0.6
Density (g/cc) 1.8 1.9
Carbon Assay (%) 92-97 100
Pitch-based fibers
Low Modulus High Modulus
Tensile Modulus (Mpsi) 23 55
Tensile Strength (kpsi) 200 350
Elongation (%) 0.9 0.4
Density (g/cc) 1.9 2.0
Carbon Assay (%) 97 99
Rayon-based fibers
Tensile Modulus (Mpsi) 5.9
Tensile Strength (kpsi) 150
Elongation (%) 2.5
Density (g/cc) 1.6
Carbon Assay (%) 99
Organic Fibers
Organic fibers for structural applications were introduced for commercial
applications in 1971. Graphite fibers are among the highest stiffness and highest
strength material known today. Types of organic fibers include:
Kevlar Fibers
Nomex Fibers, and
Spectra (ultra highly oriented polyethylene) Fibers.
Advantages for the use of organic fibers include:
Very high strength,
Very high stiffness,
Very high specific strength and stiffness,
Excellent impact resistance,
High toughness, and
Disadvantages for the use of organic fibers include:
Significantly more expensive than glass fibers, and
Properties may be affected by environmental factors (e.g. ultra violet
radiation).
Specialty Reinforcements
Specialty reinforcements include:
Boron,
Silicon Carbide, and
Other.
These reinforcements were originally developed in the 1960s. Advances in
graphite and organic reinforcing materials coupled with lower costs associated with
them has impacted the growth in applications for specialty reniforcements.
Advantages for the use of boron and silicon carbide fibers include:
Very high strength, and
Very high stiffness.
Disadvantages for the use of boron and silicon carbide fibers include:
Extremely expensive.
REI NFORCEMENT FORMS
The form of the reinforcements used in composite materials spans a wide
range and has a direct impact on the mechanical properties of the structural
component. The form of the reinforcing elements also impacts the fabrication
techniques that can be used. As discussed in the first chapter, composites are
classified based on the geometry of the reinforcing element.
Fiber Terminology
Fiberous reinforcements have several specific terms used to describe the
make-up and geometry. These terms include:
Filament
Single fiber produced from a single port in the spinning process.
Diameters for common filaments (glass and graphite) range from
0.000015 inches to 0.0005 inches.
Fibers
A general term commonly used to refer to a collection of filaments.
Strand
Commonly a bundle or group of untwisted, collimated filaments. Used
interchangeably with fiber and filament.
Tow
A bundle or group of untwisted, collimated filaments usually with a
specific count.
Yarn
A twisted bundle of continuous filaments.
Roving
A number of tows or yarns collected into a parallel bundle without
twisting.
Tape
A collection of collimated (parallel) filaments usually made from tows
held together by a binder, which is typically a B-stage resin.
Woven Fabric
A two-dimensional material made by interlacing yarns or tows in
various patterns.
Braiding
A three-dimensional material made by interlacing yarns or tows in
various patterns.
Mat
A two-dimensional material made of randomly oriented chopped fibers
or swirled continuous fibers that may be held together loosely by a
binder.
Weave types
The textile industry has developed a number of different weaves that are
commonly used in applications from clothing and upholstery to composite materials.
The specific weave used in a structure may impact the drape in the fabrication
process and the mechanical properties of the structure. Typical weaves used in
composite materials include:
Plain weave
Basket weave
Crowfoot satin weave
Long-shaft satin or harness weave
Leno weave
The typical weaves are shown in the following figure with the machine directions as
indicated in the figure.
The plain weave is the simplest weave that has uniform strength in two
directions when the yarn size and count are similar in the warp and fill directions.
Plain weave fabrics are commonly used for:
flat laminates,
printed circuit boards,
narrow fabrics, and
tooling.
The basket weave is similar to plain weave except that warp yarns are woven
as one over and under tow fill yarns. The weave is less stable than plain weave.
Consequently the weave is more pliable and drape is better. Basket weave fabrics
are stronger than an equivalent weight/count plane weave fabric. Applications for
basket weave fabrics are similar to those for plain weave.
Crowfoot satin weave has improved unidirectional quality with more
strength in the fiber directions than plain weave fabrics. The crowfoot satin weave
is more pliable than plain weave fabrics and can comply to complex contours and
spherical shapes. Applications include:
Fishing rods,
Diving Boards,
Skis,
Aircraft ducts,
Channel, and
Conduit.
Long-shaft satin or harness weave has a high degree of drape and stretch in
all directions. The weave is less stable than in plain weave fabrics. Applications
include:
Aircraft housings,
Radomes,
Ducts, and
Other contoured surfaces.
Leno weave produces heavy fabrics for rapid build-up of plies. Leno weave
fabrics are used:
As inner cores of thin coatings,
Tooling, and
Repairs.
The choice of weave for a particular application will generally be a
compromise between structural and fabrication requirements. Unidrectional tape
will produce higher strength plies but are more difficult to fabricate. Drape of the
cloth can be a major consideration in structures with complex contours.

Common weaves used in composite materials.

Cloth Directions and
Name Conventions
Plain Weave

Basket Weave Crowfoot Satin

Long-shaft Satin Leno Weave
FABRI CATI ON TECHNI QUES
For the purposes of this presentation the fabrication techniques for composite
materials are arbitrarily divided into three catagories they are: manual processes,
machine processes, and mass production processes. The type of technique used in a
particular application will depend on among other things:
the number of parts to be produced,
the facilities available,
the repeatability of the parts to be produced,
the mechanical properties required in the finished parts,
the materials (resin and reinforecment) to be used, and
the size of the part.
Unlike metal manufacturing processes, composite fabrication processes can
have significant impact on part quality. Composite materials have received undo
criticism in some arenas due in part to inadequate quality control in the fabrication
processes. Consequently, it is important for the designer and analyst to understand
the composite fabrication processes and to develop an appreciation for the impact of
the fabrication processes on system behavior.
CURE PROCESSES
The majority of composite materials in production today are made with
thermoset polymeric resins. Consequently, the structure requires some kind of a
cure process to produce the final part. A generic cure cycle is shown in the
following figure.
0
100
200
300
0 50 100 150 200 250 300
t (minutes)
T

(
F
)
0
50
100
150
p

(
p
s
i
)
2/min Heating
5/min Heating
Pressure
Three hour hold at 250F
with one hour at 150psi.

Generic cure cycle with temperature and pressure required.
Vacuum Bag Processes
Whether or not a cure cycle requires additional pressure to achieve
consolidation of part the vast majority of all composites produced require use of a
vacuum bag. I n the vacuum bag process the part is covered by a release ply, next a
barrier film, then the appropriate number of bleeder ply to absorb the excess resin
from the lay-up, a breather ply to provide a flow path for trapped gasses and
volitols released during the cure cycle, and finally a bag. The vacuum bag may be a
molded bag or sheet of polymer fitted to the part. Molded bags are more expensive
but require less manual labor to install. They are commonly used in high
production applications. A generic lay-up with the various vacuum bag components
is shown in the following figure.

Typical Vacuum Bag Components.
Bagging is an important part of processing thermoset composite parts. I t has
a direct impact on part quality. I t is possible for a part, carefully laid-up, to be
scrapped due to poor bagging. I t is essential that a bag be tightly sealed and leak
free and be in perfect contact with the workpiece. A leak free bag is necessary to
achieve consolidation of the lay-up and to provide the necessary path to exhaust
evolved gasses that may be trapped in the lay-up during the fabrication process or
those that are produced by the chemical reactions in the cure cycle. Producing a
leak free bag can be challenging, but is not in general impossible. There may be
times that a bag loses its seal during the cure cycle. When this happens it is
typically unrepairable, but the part may not be lost, depending on when the leak
develops and what type of cure cycle is in use.
Perfect contact with the part must be established in the bagging processes.
Failure to establish perfect contact will produce a flawed part that may or may not
have to be scrapped due to the failure. Perfect contact appears in two points in the
bagging process, bridges and darts. These problems are shown in the following
figures. Darts are used in bag construction when flat bag material is used instead
of a molded bag. The seams produced on the finished parts due to use of darts is a
resin rich point that is not likely to adversly affect the mechanical properties of the
part. Bridges on the other hand are more than unsightly belmishes. I t is quite
common to have internal voids and delaminations in the finished workpiece in the
vicinity of bridges. Rubbing tools are used to compact the bag and remove wrinkles.
Vacuum Bag
Mold
Lay-up
Gap due to bridging

Mold
Lay-up
Vacuum Bag
Gap due to dart
Dart

Bridging in a vacuum bag process. Dart used in vacuum bag process.
Auto- and Hydroclave processes
Autoclave and hydroclave processes use additional pressure to consolidate
the laminate. Vacuum bags are used with both of these processes. The use of a
hydroclave may produce a superior cure cycle due to the improved heat transfer
from the liquid medium to the workpiece as compared to the gas used in an
autoclave. Typically the added pressures used in these processes range from 50 psi
to 200 psi. Research on the application of pressure and duration of the vacuum
held on the part for phenolic composites has indicated that part quality can be
significantly impacted by these steps.
MANUAL TECHNIQUES
Manual fabrication techniques for composite materials include manual
lay-up and manual spray-up. Of these processes, the manual techniques are
dominated by the manual lay-up process. Advantages and disadvantages for these
processes are listed in the following table.
Design considerations for manual techniques include the following:
Minimum inside radius: 0.1875 to 0.25 inches. Tighter radii are
possible but not desirable.
Minimum draft recommended: 2. Split molds can have 0.
Undercuts: should be avioded but can be made by using split or rubber
molds.
Molded in holes: lareg diameter only.
Minium practical thickness: 0.03 inches for manual lay-up, 0.06 inches
for manual spray-up.
Maximum practical thickness: unlimited total, 0.25 inches per cure.
Normal thickness variation: +0.03/-0.015 inches for manual lay-up.
t0.025 inches for manual spray-up.
Special construction possible: built-in cores, metal inserts, metal or
other edge stiffeners
Bosses: must be tapered.
Fins: special handling required.
Limiting size factor: none, other than mold size, oven size (if required)
and handling considerations.
Shape limitations: none.
Advantages and Disadvantages of Manual Fabrication Techniques
Advantages* Disadvantages*
Design flexibility Labor-intensive process
Large and complex parts can be
produced
Only one good (molded) surface is
obtained
Production rate requirements are low Low-volume production process
Minimum equipment investment is
necessary
Quality is related to the skill of the
operator
Tooling cost is low Longer cure times required
Any material that will hold its shape can
be used as a mold form
Product uniformity is difficult to
maintain with in a single part and from
part to part
Start-up lead time and cost are minimal Waste factor is high
Design changes are easily effected
Molded-in inserts and structural
reinforcements are possible
Sandwich constructions are possible
Prototyping and pre-production method
for high volume molding processes
Semi-skilled workers are needed and are
easily trained
Hazards associated with handling the
materials are higher
* Note: the horizontal alignment in the table is not intended to imply a relationship
between the points.

Finished surfaces: one (two with special tooling).
Gel-coat surface: only one smooth surface, reverse can be coated after
molding
Molded in labels:
Manual Lay-up
I n a manual lay-up process each ply of a laminate is placed by hand. The
advantage to these processes is that little in the way of equipment is required. The
disadvantages to these processes include: high degrees of variablitiy between parts,
even those produced by a single technician and generally inconsistent quality even
when performed by a highly qualified technician. As always the advantages must
be weighed against the disadvantages in a given application.
Wet Processes
I n wet lay-up processes the lamina or ply must be saturated with resin before
being laid-up in the tooling. Following saturation, excess resin must be removed to
avoid having a part that is unacceptably resin rich. The following are the basic
steps in a wet lay-up process:
1. Prepare patterns for each ply of the laminate locating all darts and folds
required to accurately follow the mold. Minimize the number of overlaps and
never superimpose overlaps. I f overlapping plies are required, keep the
overlap width to 0.75 inches, +0.25 inches/-0.0 inches. A pattern may be used
for multiple ply when the pattern is repeated in the stacking sequence.
2. Optional, prepare a lamination kit, cutting all plies to the required pattern,
marking their order in the stacking sequence with Teflon tape.
3. Coat tooling with release film or place release ply on tooling. I f Gel-coat is
desired it should be applied at this time.
4. Prepare the resin pot. Mix the resin components as required and place in a
container that is sufficiently large to lay the individual plies for the laminate.
5. Saturate the first ply of material with resin.
6. Strip excess resin from ply.
7. Place first ply on tooling with the orientation specified in the stacking
sequence, film side-up. Using a non-stick (Teflon or steel) tool press the ply
onto the tooling. Work out all air bubbles and wrinkles. Ensure that the ply
is in total contact with the tooling, working out all bridges that occur at
fillets and gaps at corners that occur at rounds. Draw excess resin from ply.
8. Remove Teflon tape marking lamina sequence. (This is very important.)
9. Saturate the next ply of material with resin.
10. Strip excess resin from the ply.
11. Apply the next ply with the orientation specified in the stacking sequence,
film side-up. Again, using a non-stick (Teflon or steel) tool press the ply onto
the lay-up. Work out all air bubbles and wrinkles. Ensure that the ply is in
total contact with the previous layer, working out all bridges that occur at
fillets and gaps at corners that occur at rounds. Draw excess resin from ply.
12. Repeat steps 8-11 until the specified stacking sequence is completed.
13. Depending on the resin system used, if the fabrication process cannot be
completed in a single operation the lay-up should be bagged with a vacuum
bag and held under a controlled environment (especially low humidity) until
the operation can resume.
14. Prepare the part for the cure process following the bagging procedure.
Manual scissors, power shears, and semi-automatic and automatic machines
may be used to cut the plies for the lay-up. When computer numerically controlled
automatic cutting machines are used, patterns are not necessary.
Prepreg Processes
1. Prepare patterns for each ply of the laminate locating all darts and folds
required to accurately follow the mold. Minimize the number of overlaps and
never superimpose overlaps. I f overlapping plies are required, keep the
overlap width to 0.75 inches, +0.25 inches/-0.0 inches. A pattern may be used
for multiple ply when the pattern is repeated in the stacking sequence. As
with the wet process, if automatic cutting machines are used patterns are not
required.
2. Coat tooling with release film or place release ply on tooling.
3. Optional, prepare a lamination kit, cutting all plies to the required pattern,
marking their order in the stacking sequence on the backing film. Return
the kit to storage as quickly as possible to maintain quality.
4. Bring the prepreg to room temperature for the lay-up process.
5. Place first ply on tooling with the orientation specified in the stacking
sequence, film side-up. Using a non-stick (Teflon or steel) tool press the ply
onto the tooling. Work out all air bubbles and wrinkles. Ensure that the ply
is in total contact with the tooling, working out all bridges that occur at
fillets and gaps at corners that occur at rounds. A hot air gun can be used to
increase ply flexibility and tack to place the ply.
6. Remove the backing film from the prepreg. (This is very important.)
7. Apply the next ply with the orientation specified in the stacking sequence,
film side-up. Again, using a non-stick (Teflon or steel) tool press the ply onto
the lay-up. Work out all air bubbles and wrinkles. Ensure that the ply is in
total contact with the previous layer, working out all bridges that occur at
fillets and gaps at corners that occur at rounds.
8. Repeat step 7 until the specified stacking sequence is completed.
9. I f the fabrication process cannot be completed in a single operation the lay-up
should be bagged with a vacuum bag and placed in cold storage under
vacuum until the operation can resume or held under a controlled
environment (especially low humidity). When lay-up resumes, if the part
was placed in cold storage, the part must be brought to room temperature
before fabrication can continue.
10. Prepare the part for the cure process following the bagging procedure.
Spray-up
I n the typical spray-up process chopped fibers, usually glass, and resin are
simultaneously sprayed onto or into an open mold. Fiber roving is fed through a
chopper and injected into a resin stream that is manually directed at the mold. The
resin system may be pre-mixed or mixed in the spray-up nozzle. After the
composite is sprayed into the mold it is hand rolled to remove air, compact the
fibers, and smooth the interior surface. Because of the nature of the process the
fibers are randomly oriented within the laminate and the behavior is transversely
isotropic. Depending on the resin system used the workpiece will be bagged and
cures in the same manner as lay-up parts.
MACHINE PROCESSES
Machine processes are typically superior to manual processes in quality,
quantity, and production time. However, they are also significantly more expensive
than manual processes to implement. Simple winding machines are in the tens of
thousands of dollars, while the most advanced, sophisticated fiber placement
machines are millions of dollars. With capital equipment costs of this magnitude
the decision to use machine processes is not a caviler decision.
At this writing there are four principal types of machines used for composite
fabrication. They are: filament winding, tape placement or tape laying, fiber
placement, and pultrusion. Of these four, only the pultrusion process incorporates
a cure cycle in the process. Parts produced by filament winding, tape laying, and
fiber placement must be bagged and cured with the specified cure cycle.
Filament Winding
Filament winding is the oldest of the machine processes. The process is
related to the turning processes used in machining operations. Filament winding is
used to produce axisymmetric structures. Parts produced by filament winding
include:
Tubes or pipes,
Cylindrical pressure vessels (Rocket motor cases), and
Spherical pressure vessels.
Fundamentally the process involves winding roving or tow around a
mandral. The winding angle ranges from nearly axial, or longitudinal (0, axial,
can be obtained in special winding operations) to hoop, or circumferential (90). I n
the winding operation dry roving is pulled through a resin bath where the roving is
saturated with resin. The excess is stripped from the roving and the roving is
drawn through a generation ring. The winding head with its generating ring
traverses the longitudinal direction of the workpiece riding on a carriage and laying
the roving on the mandral. The roving follows a helical path around the mandral,
see the figures below. As the carriage reaches the end of the workpiece it reverses
direction and lays down another layer in the opposite direction. The process
continues until the mandral is completely covered and then the machine moves to
the next ply. Because the roving does not completely cover the workpiece in a
single pass (except in the hoop direction) the roving is laid down in stripes that
alternate in direction (t). The result is something approaching a woven cloth,
similar to a plain weave fabric, except that the fiber (roving) directions are not
perpendicular. A typical filament winding machine is shown in the figure.

Fundamental components in a filament winding process.

Hoop or circumferential winding. Typical polar winding.

Multi-circuit helical winding. Note the overlap of windings.

Typical filament winding machine.
One of the most important components in successful filament winding is the
determination of the relative speed between the mandral and winding head. These
motions determine the wrapping angle and overlap of the roving. The winding
angle may be approximated from:
arctan
R
v
_

,
,
where is the winding angle, R is the radius of the workpiece, is the rotation rate
of the workpiece, and v is the longitudinal speed of the winding head. From this
equation you can observe that if the workpiece has a change in diameter along the
length the rotation rate of the workpiece and/or the speed of the carriage must be
adjusted to hold the winding angle constant. The complexity of the problem is
further complicated by polar winding at the dome of pressure vessels.
Another important component in successful filament winding is tension
control. Tension affects resin content, void content, and structural properties.
Roving tension ranges from 0.25 lbs. to 1 lb. per bundle or tow. Tension is provided
by guide eyes in line, center rotating guide eye, rotating scissor bars, drum-type
brakes (which may be electromagnetically controlled), and/or drag through the
resin bath. The first three of these tensioners are shown in a figure below.

Guide eyes in line. Center rotating guide eye Rotating Scissor bars.
Advances in filament winding technology include spherical winding
machines and multi-head winders. Spherical winding machines can produce
vessels that are cylindrical or spherical with a single opening. Multi-head winding
machines can produce vessels with a quasi-braided structure. A spherical winding
machine is shown in the following figure.
Advantages and Disadvantages for Filament Winding Processes.
Advantages Disadvantages or Limitations
Applicable to parts of widely varying
size.
Resin viscosity and pot life must be carefully
chosen and monitored.
Parts with strength in several
directions can be easily made.
Programming of the winding can be difficult.
Excellent material usage. Not all shapes can reasonably be made by
filament winding.
Forming after winding and other
techniques allow noncylindrical
shapes to be made.
Operational control of several key
parameters is important.
Flexible mandrels can be retained in
the structure to serve as liners for
tanks.
Ability to analyze (design) impaired due to
invalidation of key assumptions in
lamination theory.
Panels and fittings for reinforcement
or attachment can be easily
included during the winding process.

Parts with high pressure ratings can
be made.


A spherical winder.
Tape Lay-up
Semi-automatic and automatic tape laying machines have been developed to
reduce production times, improve consistency within parts and between parts, and
improve part quality. Tape lay-up machines use prepreg tape, unidirectional and
cloth, and are computer numerically controlled. Through appropriate programming
it is possible to eliminate the patterns use in the manual lay-up processes.
Semi-automatic and automatic tape laying machines are used to produce flat and
contoured laminates. Tape laying machines are limited in their capabilities to
surfaces with large radii of curvature. Tape laying machines cannot produce highly
geometrically complex parts.
Typical automatic tape laying machines are described by the number of
degrees of freedom (DOF) or axes of the tape laying head, for example, a head may
have three translational degrees of freedom, two rotational degrees of freedom, the
ability to start a tape, and the ability to cut the tape. This machine is described as
a seven-DOF machine. Figures below show a tape laying head, a flat automatic
tape laying machine, a coutoured laminate in a tape laying machine, and a
multi-axis tape laying machine.

Grumman tape laying head.

Flat automatic lay-up machine.

A contoured laminate on an automatic lay-up machine.

A ten DOF automatic lay-up machine.
Fiber Placement
Automatic fiber placement machines combine and extend the capability of
filament winding and tape laying machines. Fiber placement machines place
individual fiber bundles onto a mold. This is reminescent of the filament winding
process and in contrast to the laying of a tape with a tape laying machine.
However, the fiber placement machine places the fiber bundles in parallel
throughout a layer, without the overlapping in a helical winding process. Because
of the ability to place fiber bundles the fiber placement process can produce highly
geometrically complex shapes with small radii of curvature. I nternal radii are
limited by the size of the placement head, approximately 6 inches on a Viper
Placement Machine and external radii are limited to the minimum bending radius
of the fiber bundles, approximately 0.1875 to 0.25 inches. A fiber placement
machine is shown in the figure below.
Tape
Laying
Carriage

NASA Viper fiber placement machine.
Fiber placement is a comparably rapid process. Fiber placement machines
can place up to 700 inches of fiber per minute. Further, the complexity of the parts
produced by fiber placement is extrordinary. For example, the intake duct for the
F-16 is a geometrically complex part with multiple compound curves is produced by
fiber placement with relative ease.
Pultrusion
Pultrusion is an adaptation of the drawing process to composites fabrication.
This process produces long relatively narrow cross section with highly ordered and
compacted reinforcements. Cross sections produced by pultrusion range from
circular to L-channel to hat sections. The reinforcing fibers, which may be glass,
graphite, or arimid, are generally all oriented along the major direction of the
pultrusion.
I n general the process begins with dry roving that is drawn through a resin
bath and into a compaction die. From the compaction die the material is drawn into
a curing die where the excess resin is striped and the part is cured with significant
pressures. I t is possible to draw the workpiece over a bench before the final cure to
produce curved (instead of straight) sections. Automotive composite leaf springs are
an example of curved pultruded parts, a pultrusion forming operation and
automobile springs are shown in the figures. A microwave curing prepreg based
pultrusion system is shown in the figure below.

Schematic of a Pultrusion forming system.

Automobile leaf springs produced by Pultrusion forming.

Schematic of a Pultrusion system using microwave energy to cure the resin system.
MASS PRODUCTION TECHNIQUES
Manual production techniques by their very nature are limited in their
production rates. The machine processes are typically more expensive and do not
lend themselves to high production volumes. Production of composite parts for
applications such as those found in the automotive industry necessitate production
processes that have very high production rates. Molding processes are able to meet
these demands.
Molding
Molding of composite materials has its foundation in the metals casting and
forming processes. These processes include: sheet molding, bulk molding, thick
molding, and liquid molding (resin transfer molding). The advantages of these
processes include lower per part tooling costs and higher production rates.
Sheet Mol di ng
Sheet molding processes were developed in response to a request from the
automotive industry. Their desire was for a composite material process that
allowed them to use the metal bending and stamping equipment and techniques
with which they were familiar. The material used in sheet molding is called sheet
molding compound (SMC).
I n the sheet molding process chopped roving, usually glass fibers, is mixed
with resin and deposited between plastic films, usually polyethylene. The material
is then used in a stamping like process. The schematic below shows a typical SMC
machine.
Pultrusion
High material usage compared with
lay-up
Part cross-sections must generally be
uniform.
High throughput rate Problems can arise when resin or fibers
accumulate and build up at the die opening
Can give high resin contents When dies run resin rich to account for fiber
anomalies, strength is sacrificed.
Close to fiber tow properties Voids can result if dies are run with too
much opening for the fiber volume.
When quick curing systems are used,
mechanical properties are often sacrificed.

Schematic of the SMC fabrication process. Note that the flow is from left to right.
Thi ck and Bul k Mol di ng
Thick and bulk molding operations are related to the closed die or
matched-die casting and forming processes. I n these processes the bulk molding
compound (BMC) or the thick molding compound (TMC) is prepared similar to the
SMC compound, however it is mixed in buld rather than sheet. Advantages and
disadvantages of the process are listed below. The TMC process is shown in the
figure.
Matched-die Molding
Both interior and exterior surfaces
are finished.
More equipment is needed than for lay-up.
Complex shapes including ribs and
thin details are possible.
Molds and tooling are costly compared to
lay-up molds.
High production rates are possible. Transparent products are not possible with
SMC and BMC.
Labor costs are low. Molding problems (trapped water, etc.) may
cause surface imperfections such as pitting
or waviness.
Minimum trimming of parts in
needed.
Products have good mechanical
properties and close part tolerances.
Good consolidation of parts.
SMC and BMC have limited shelf-lives.

Schematic of the TMC fabrication process.
Resi n Transfer Mol di ng
Resin transfer molding is used in a number of variations with a number of
different names to identify the variations. Among these processes are: Resin
Transfer Molding (RTM), Structural Reaction I njection Molding (SRI M) Resin
I njection Molding (RI M), Vacuum-Assisted Resin I njection (VARI ), Thermal
Expansion Resin Transfer Molding (TERTM), Vacuum Assisted Resin Transfer
Molding (VARTM), and Seamans Composites Resin I njection Molding Process
(SCRI MP) to name a few. All of these processes involvethe same basic steps. The
basic steps in RTM are:
Place preform in mold.
Close mold.
I nfuse/ inject liquid resin into mold.
Cure part in mold.
Open mold.
Remove part from mold.
Clean up part.
One of the significant problems in the use of these liquid molding processes is
adequate setting of the structural preform by the liquid resin. To produce a high
quality composite part it is essential that the reinforcing fiber structures be
throughly impregnated with the resin. The resin or matrix in a composite transfers
the load from one fiber to the next. I f there is no resin present, the loads do not get
transferred from fiber to fiber, which results in an inadequate structure. Proper
infusion or impregnation of the fiber preform requires a low viscosity resin and an
extended pot life (two hours or more depending on the size of the part). I n addition
it is very important to provide a leak path for all gasses that are native to the
preform as the resin is injected. Advantages and disadvantages for the RTM
process are listed below.
Schematics for the resin transfer molding process and the resin infusion
process are shown in the following figures. Note in the RTM process that a pump is
used to force the resin into the mold and a press is used to hold the mold closed. I t
may be advantageous to use pumps to force the resin into the mold. However, high
resin flowrates may cause the preform to be dislocated from the desired position.
Resin Transfer Molding
Very large and complex shapes can
be made efficiently and
inexpensively.
The mold design is critical and requires
great skill
Production times are much shorter
than lay-up.
Properties are equivalent to matched-die
molding (assuming proper fiber wet-out) but
are not generally as good as with vacuum
bagging, filament winding, or Pultrusion.
Clamping pressure is low compared
to matched-die molding.
Control of resin uniformity is difficult.
Radii and edges tend to be resin rich.
Surface definition is superior to
lay-up.
Reinforcement movement during resin
injection is sometimes a problem.
I nserts and special reinforcements
can be added easily.
The sill level required for the
operator is low.
Many mold materials can be used.
Parts can be made with better
reproducibility that with lay-up.
Workers are not exposed to chemicals
and vapors as with lay-up.


Schematic of the Resin Transfer Molding Process.

Schematic of the Resin I nfusion Process.
Automated Spray-up
The automated spray-up processes are simply an automation of the manual
spray-up processes. I n some ways the process is related to the automated painting
processes used in industries such as the automotive industry. I n this application a
robotic arm is programmed to spray chopped reinforcement and resin into a mold.
A schematic of the automated spray-up process is shown in the following figure.
The various letters in the figure designate components of the spray-up machine.

Schematic of an automated spray-up machine.
HAZARDS
The hazards associated with the production and fabrication of composite
materials is higher than in the production of parts from conventional materials.
Some of the more exotic conventional materials, such as titanium, magnesium, and
berelium, have significant healt hazards associated with their production and/or
fabrication processes. Composite materials in general are produced from organic
chemical compounds that in their uncured state may pose significant health
hazards unless handled with great care.
HEALTH INFORMATION TERMINOLOGY
To begin the discussion we must again define some terms in the field. First
of all we must define the difference between toxicity and hazard. Toxicity is an
inherent harmful effect of a chemical. I t is a physical property of the chemical.
Hazard is controlled by exposure. Exposure to a toxic chemical required for a
hazard to exist. A chemical with Acute Toxicity has a harmful effect after single
and/or short term exposure.
Toxicity is measured in lethal doses and lethal concentrations. The Mean
Lethal Dose LD50 is expressed as a ratio in mg of chemical to kg of body weight. I t
is the amount of chemical administered by a specific route that is expected to kill
50% of a group of experimental animals. The Mean Lethal Concentration LC50 is
expressed in mg/m
3
or parts per million (ppm) in air. I t is the concentration of
chemical in air that is expected to kill 50% of a group of experimental animals.
There are for some chemicals levels below which there is no observable effect. This
level is defined as the No Observable Effect Level, NOEL.
The toxicity of a chemical is further characterized by the local effects it
causes. These effects include:
I rritation
Localized reaction resulting from either single or multiple exposures to
a physical or chemical entity at the same site.
Corrosion
Tissue destruction in such a way that normal healing is not possible.
Sensitization
Allergic reaction to a substance that develops upon repeated exposure.
Chronic Toxicity
Characterized by adverse health effects in an animal or person, which
has been caused by exposure to a substance of over a significant
portion of that animals or persons life, or by long-term effects
resulting from a single or a few doses.
Two other health terms that are used to describe the toxicity of chemicals are
used commonly today but have specific definitions in the health field. The first is
carcinogenicity. This is the ability of a substance to cause tumors. Long term
testing required to determine if a substance is carcinogenic. The results of these
tests are conclusive. The second is mutagenicity. This is the ability of substance to
cause changes in the genetic materials of cells. Short term testing can be used to
determine if a substance is mutagenic. The results of these short term tests are
speculative (non-conclusive).
To minimize the hazards associated with working with toxic substances
exposure limits are defined. Exposure limits that are defined in terms of Threshold
Limit Values (TLV). TLVs assume that the exposed population is composed of
normal, healthy adults, and does not address aggravation of pre-existing conditions
of illnesses. These limits are not fine lines between safe and dangerous
concentrations and should not be used by anyone untrained in the discipline of
industrial hygiene. Four important TLVs are:
Threshold Limit Value Time Weighted Average (TLV-TWA)
The time weighted average for a normal 8 hour workday and 40 hour
work week, to which nearly all workers may be exposed, day after day
without adverse effect.
Threshold Limit Value Short Term Exposure Limit (TLV-STEL)
The concentration to which workers can be exposed continuously for a
short period of time (15 minutes) without suffering from (1) irritation,
(2) chronic or irreversible tissue damage, or (3) narcosis of sufficient
degree to increase the likelihood of accidental injury, impair
self-rescue, or materially reduce work efficiency, and provided that the
daily TLV-TWA is not exceeded.
Threshold Limit Value Ceiling (TLV-C)
The concentration that should not be exceeded during any part of the
workday.
Permissible Exposure Limits (PEL)
PELs are legal binding airborne exposure limits, which are issued by
the Occupational Safety and Health Administration (OSHA).
Proper industrial hygiene is essential to minimize the hazards of working
with toxic substances and to ensure that the TLVs are not exceeded. Proper
industrial hygiene requires controlling the routes of exposure to the toxic
substances. Routes of contact include:
Skin and Eye Contact
At risk through skin and eye contact are: Hands, Lower Arms, and
Face. Contact with liquids, gases, vapors, or particulates should be
minimized to reduce the risk of contact.
I nhalation
I nhalation can be a significant route of exposure to toxic substances in
composite fabrication. Solvents and other volatiles may be released
from the resin systems during the manufacturing and curing of
composites. Further, dusts may be generated in the machining of
cured composite materials.
I ngestion
I ngestion is not typically not a major problem in the fabrication of
composite materials provided that there is sufficient control.
Proper industrial hygiene requires control of the processes in five areas:
Administrative, Engineering, Operations/Process, Safety, and personal.
Administrative Controls include proper: handling of materials, training, isolation of
operations, personal protective equipment, personal hygiene, warnings and labels,
housekeeping, dispensing and storage of chemicals, and emergency instructions.
The Engineering Controls include proper: plant layout, design and use of
equipment, and exhaust ventilation. Operations/Process Controls include proper:
mixing of resins (personal protective equipment as appropriate, and specific mixing
instructions -- available and followed), curing operations (use product specific cure
cycle), and handling of cured resin systems (as appropriate). Safety Controls are as
appropriate. Personal Controls include proper training of all personnel and a
commitment by all personnel to maintain a safe, hazard free workplace. This
includes a commitment by employers to effectively instruct the employees on site
hazards, warning labels, and material safety data sheets. Further, management
and employee are responsible for knowing about hazards and taking measures for
minimizing exposure.
TOXICOLOGICAL PROPERTIES OF COMPONENTS
The following section lists some of the toxicological properties of components
of composite materials. The information presented below is generic and further
specific information should be obtained regarding the specific compounds with
which you are dealing.
Epoxy Resins
Epoxy resins are always used with curing agents and commonly with a
series of other additives. Generally, more demanding handling
procedures and controls are recommended for the curing agent or other
additives. Epoxies are primary skin and mucous membrane irritants.
Some epoxies have sensitizing effects.
Hardening and Curing Agents
Aromatic Amine Hardeners
These hardeners have slight irritating effect on skin and mucous
membranes. They have been shown to cause damage to liver and
may decrease ability of blood to transport oxygen to tissues.
Exposure should be minimized or avoided.
Aliphatic and Cycloaliphatic Amine Hardeners
These hardeners are strong bases. I t is a sever irritant and is
corrosive. Exposure should be avoided.
Polyaminoamide Hardeners
This hardener produces mild irritation of skin and mucous
membranes. I t may cause sensitization. Exposure should be
minimized or avoided.
Amide Hardeners
This hardener has a slight irritant effect. Avoid inhaling dust.
Anhydride Curing Agents
This hardener is a sever eye irritant and a strong skin irritant.
Exposure should be minimized or avoided.
Polyurethane Resins
I socyanates
Most commercial isocyanates are highly toxic due to skin and
respiratory sensitization, or skin absorption and systemic toxicity.
They produce strong irritation of skin and mucous membranes of
eyes and respiratory tract. Extreme care is necessary! Good
ventilation is required!
Toluene diisocyanate (TDI )
Toluene is a mutagen. TLVs for toluene are: TLV-TWA of 0.005
ppm and TLV-STEL of 0.02 ppm. Toluene has no odor below TLV
levels. At this time there is no carcinogenic data. I t is, however,
classified as potentially carcinogenic.
Polyols
These are cure agents. At this time no particular health hazard is
indicated.
Phenolic and Amino Resins
Phenol-Formaldehyde Resins
These resins have low hazard levels. Phenol and formaldehyde
may be absorbed through skin. Good ventilation is recommended
and skin sensitization is possible.
Urea- and Melamine-Formaldehyde Resins
These resins have acute toxicity similar to the phenol-formaldehyde
resins. Skin sensitization is possible.
Bismaleimides
No extensive studies have been performed on bismaleimides at this
time. They may cause skin irritation or sensitization. Dust or vapors
may irritate eyes, nose, and throat.
Thermoplastics
Generally thermoplastics are not considered harmful to workers
health. Skin irritation is not observed and no toxic effects are known
to be associated with inhalation of dusts. Burns may present sever
hazard with thermoplastics.
Styrene Monomer
Styrene vapors can cause eye irritation. The liquid will cause eye,
skin, and mucous membrane irritation. Styrene has systemic effects
on central nervous system, liver, and kidneys have been observed. I t
is possibly carcinogenic to humans.
Reinforcing Materials
Most reinforcing materials in and of themselves are non-toxic.
However, inhalation of filler may be detrimental to health. I nhalation
may produce effects similar to asbestosis.
Carbon and Graphite Fibers
Threshold limits have been established for carbon and graphite
fibers. The limits are: (TLV-TWA) 10mg/m
3
(OSHA) and 3
fibers/cm
3
(U. S. Navy).
Aramid Fibers
The exposure limit (TLV-TWA) is set by manufacturers at 5
fibers/cm
3
. No apparent effects from inhalation are observed.
Fiber Glass
The exposure limit (TLV-TWA) for fibrous glass is 10mg/m
3
.
NI OSH recommends 3 fibers/ cm
3
. Exposure may cause
mechanical irritation of eyes, nose, and throat. I t is classified as
possible human carcinogen.
Solvents
Contact with most organic solvents causes drying and defatting of skin
and dermatitis. Some solvents are directly absorbed through intact
skin; absorption is enhanced if skin abraded or irritated. An
additional concern is the ability of a solvent to carry other substances
though skin with it.
Acetone
Acetone is a common laboratory solvent. I t was placed on the
hazardous list. However, it has been more recently removed.
The threshold limits are: a TLV-TWA of 750ppm and a TLV-
STEL of 1000ppm.
Methyl ethyl ketone (MEK)
I n addition to being a solvent, MEK is also used as an
accelerator for Gel-coat. I t causes eye, nose, and throat
irritation. The threshold limits are: a TLV-TWA of 200 ppm and
a TLV-STEL = 300 ppm.

LAMINA MECHANICS
PRELI MI NARI ES
A lamina is a flat or nearly flat thin layer of material. I n this application the
material is a composite material either tape or cloth. I n practical engineering
applications the lamina is the fundamental building block of the structure. To
understand the mechanics of laminated structures it is necessary to understand the
mechanics of the individual lamina.
NOTATION
Recall from solid mechanics that there are six unique stresses at each point
in a body. The drawing in Figure 1 shows the nine stresses acting on the faces of
an infinitesimal cube at the point in question. The stresses, , shown are given
with two indices, the first index indicates the direction of the normal to the surface
on which the stress acts and the second index indicates the direction of the stress
component. This is indicial or tensorial notation,
ij
, where the indices, i and j
range form 1 to 3.
The nine stresses shown in Figure 1 reduce to six unique stresses as a result
of the application of the principle of conservation of angular momentum and
assumptions regarding the ability of the material to support an internal couple.
The consequence is that the shear stress appear in pairs,
21 12
,
31 13
, and
32 23
, or
ji ij
.
x
1

x
3

x
2

11

12

22

23

33

13

21

31

32

Figure 1. Stresses at a point.
LAMI NA MECHANI CS
Compact or contracted notation is an alternate method for identifying the six
unique stresses at a point. I n compact notation a single index is used to identify
the individual stress. The relationship between the stress terms in tensorial
notation and compact notation is shown in Table 1.
Table 1. Relationship Between Stress Terms Expressed
I n Tensorial Notation And Compact Notation.
Tensorial Notation Compact Notation
11

1
22

2

33

3

23

4

13

5

12

6

COORDINATE SYSTEMS
I n analysis of the mechanics of a composite lamina there are four different
coordinate systems that may need to be considered. They are:
Principal Material Directions
Structural Directions
Principal Stress Directions
Principal Strain Directions
The principal material directions are defined from symmetries of the material. For
a filamentary material or cloth the first and second principal material directions
are defined by the primary fiber direction. The third direction is taken
perpendicular to the lamina. The structural directions are those that are defined by
the particular application of the composite material. The principal stress and
principal strain directions are calculated from the loading and the associated
material behavior. To distinguish between these terms the following convention
will be used:
Principal material direction stresses and strains are indicated by
numerical subscripts, i. e., 1, 2, ..., and 6, as noted above.
Structural direction stresses and strains are indicated by letter
subscripts, i. e., x, y, z, s. This notation scheme is adapted from J ones, and
Tsai and Hahn. I n this application the subscripts, x, y, and z, indicate
normal stresses or strains in the corresponding direction and s indicates
LAMI NA MECHANI CS
the in-plane, xy, shear stress or strain. The out-of-plane stresses are zero
as a result of the plane stress assumption.
Principal stress direction stresses and strains are indicated by upper-case
Roman numeral subscripts, i. e., I , I I , and I I I .
Principal strain direction stresses and strains are indicated by a
superscript, i. e.,
, and upper-case Roman numeral subscripts, i. e., I , I I ,

and I I I .
This system is relatively common within the industry. However, you should pay
careful attention to the convention used by any software package in use or when
dealing with other organizations.
TRANSFORMATION OF STRESSES FROM ONE COORDINATE SYSTEM TO
ANOTHER
I n analysis of the mechanics of a composite lamina it is necessary to
transform the stresses from one coordinate system to another. I n this process it is
important to define the sense of the coordinate rotation. A positive coordinate
rotation angle is defined as positive in the positive z-direction. Figure 2 shows the
relationship between the structural axes and the principal material directions of a
lamina. I n this figure, the positive angle, , is shown as a counterclockwise rotation
about the z-axis.
I t is easily shown and well documented in numerous solid mechanics texts
that the stresses in the structural directions are transformed to the principal
material directions by:
Figure 2. Relationship between the structural coordinate system
and the principal material directions.
x
y
x
2

x
1

x
3
=z
+
LAMI NA MECHANI CS
'
1
1
1
]
1
'
s
y
x
n m mn mn
mn m n
mn n m
2 2
2 2
2 2
6
2
1
2
2
, 1
where cos m and sin n .
STRAIN
For the analysis of most common engineering structures the deformation is
measured in terms of the change in length to the original length and is termed
strain. The tensorial definition of strain,
ij
, is:
,
_

i
j
j
i
ij
x
u
x
u
2
1
, 2
where
i
u are the displacements in the ith coordinate direction. Note that there are
only six unique strains since the strain tensor is symmetric, i. e.,
ji ij
.
Engineering strain is an alternative measure of the deformation. The definition of
engineering strain,
ij
e , is:
'
j i
x
u
x
u
j i
x
u
e
i
j
j
i
j
i
ij
. 3
The shear strains, i. e., j i e
ij
, are also commonly presented as
ij
.
As with stress, compact or contracted notation is an alternate method for
identifying the six unique strains. I n compact notation a single index is used to
identify the individual stress. The relationship between tensorial strain,
engineering strain, and engineering strain expressed in compact notation is shown
in Table 2.
LAMI NA MECHANI CS
TRANSFORMATION OF STRAINS FROM ONE COORDINATE SYSTEM TO
ANOTHER
The differences in the definition of tensorial and engineering strains, though
seemingly small, are significant. The normal strains in both definitions are equal.
However, the shear strains differ by a factor of
2
1
. The consequence of this
difference is that engineering strains are not directly transformable from one
coordinate system to another; they must first be converted to tensorial strains
before the coordinate transformation and then converted back to engineering strain.
The end result is that the transformation of the strains measured along the
structural directions into the principal material directions is
'
1
1
1
]
1
'
s
y
x
n m mn mn
mn m n
mn n m
2 2
2 2
2 2
6
2
1
2 2
, 4
where cos m and sin n , as defined previously.
PLANE STRESS ASSUMPTION
There are two assumptions that are commonly made in the analysis of solids
and structures to simplify the governing system of equations. They are:
Plane Stress, and
Table 2. Relationship Between Tensorial Strain,
Engineering Strain, And Engineering Strain Expressed
I n Compact Notation.
Tensorial Strain Engineering Strain
Engineering Strain
Expressed in Compact
Notation
11

11
e
1

22

22
e
2

33

33
e
3

23
2
1

23 23
e
4

13
2
1

13 13
e
5

12
2
1

12 12
e
6

LAMI NA MECHANI CS
Plane Strain.
The assumption of plane stresses is that the out-of-plane stresses are much much
less than the in-plane stresses. I n this application assume that the out-of-plane
direction is the z- or x
3
-direction, then 0
23 13 33
. The plane stress
assumption is commonly applied to structures that are very thin in one direction,
such as a composite lamina. The assumption of plane strain is that the out-of-plane
strains are zero, or much much less than the in-plane stresses. I f the out-of-plane
direction is the z- or x
3
-direction, then 0
23 13 33
. The plane strain
assumption is commonly applied to structures that are very thick in the out-of-
plane direction. These assumptions have similar impact on the governing system of
equations, but they are very different assumptions and they have distinctly
different meanings. The plane stress assumption is used in the analysis of
composite lamina.
CONSTI TUTI VE RELATI ONS
Hooke originally postulated a linear relationship between stress and strain,
i. e.,
E ,
where E is the constant of proportionality. This relationship has been generalized
to the full three dimensional stress state at a point. The generalized form of
Hookes Law in terms of tensorial stresses and strains is:

3
1
3
1 k l
kl ijkl ij
C ,
where
ijkl
C is the stiffness tensor. The stiffness tensor is a fourth order tensor
which represents, 3
4
or 81 independent constants. When the symmetries of the
stress and strain tensors, and assumptions about the internal strain energy are
applied these 81 independent constants are reduced to 21 and the Generalized
Hookes Law can be expressed as a matrix expression. I t is:
{ } [ ]{ } C ,
where { } is the 6 1 stress vector, { } is the 6 1 strain vector, and [ ] C is the
symmetric 6 6 stiffness matrix. I f all of the terms are shown and compact notation
is used, the general constitutive relation for any material is:
LAMI NA MECHANI CS
'
1
1
1
1
1
1
1
1
]
1
'
6
5
4
3
2
1
66 56 46 36 26 16
56 55 45 35 25 15
46 45 44 34 24 14
36 35 34 33 23 13
26 25 24 23 22 12
16 15 14 13 12 11
6
5
4
3
2
1
C C C C C C
C C C C C C
C C C C C C
C C C C C C
C C C C C C
C C C C C C
. 5
Alternately, the constitutive relation may be written in the compliance form
where strains are expressed in terms of stresses. This form is:
{ } [ ]{ } S ,
where [ ] S is the symmetric 6 6 compliance matrix, which is the inverse of the
stiffness matrix. I f all of the terms are shown and compact notation is used, the
general compliance form of the constitutive relation for any material is:
'
1
1
1
1
1
1
1
1
]
1
'
6
5
4
3
2
1
66 56 46 36 26 16
56 55 45 35 25 15
46 45 44 34 24 14
36 35 34 33 23 13
26 25 24 23 22 12
16 15 14 13 12 11
6
5
4
3
2
1
S S S S S S
S S S S S S
S S S S S S
S S S S S S
S S S S S S
S S S S S S
. 6
When observations of or assumptions about material behavior are made
conclusions may be made regarding physical planes of symmetry in the material
and consequently reduce the number of independent terms in the stiffness matrix.
ANISOTROPIC MATERIALS
Materials which display no axes of symmetry in response to mechanical loads
are defined as anisotropic, literally not isotropic. These materials have 21
independent material constants as shown in the equations above. For the analyst
this is the worst possible situation. The implications of a full stiffness matrix is
that a normal strain will produce normal and shear stresses and, conversely, that
shear strains will produce shear and normal stresses.
ORTHOTROPIC MATERIALS
Materials which display three perpendicular axes of symmetry in response to
mechanical loads are defined as orthotropic. These materials have 9 independent
material constants and the stiffness form of the constitutive relation is:
LAMI NA MECHANI CS
'
1
1
1
1
1
1
1
1
]
1
'
6
5
4
3
2
1
66
55
44
33 23 13
23 22 12
13 12 11
6
5
4
3
2
1
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0
0 0 0
0 0 0
C
C
C
C C C
C C C
C C C
7
and the compliance form of the constitutive relation is:
'
1
1
1
1
1
1
1
1
]
1
'
6
5
4
3
2
1
66
55
44
33 23 13
23 22 12
13 12 11
6
5
4
3
2
1
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0
0 0 0
0 0 0
S
S
S
S S S
S S S
S S S
. 8
For an orthotropic material it is easily shown through simple linear algebra that:
[ ]
1
1
1
1
1
1
1
1
1
1
1
1
1
1
]
1
66
55
44
2
12 22 11 11 23 13 12 22 13 23 12
11 23 13 12
2
13 33 11 33 12 23 13
22 13 23 12 33 12 23 13
2
23 33 22
1
0 0 0 0 0
0
1
0 0 0 0
0 0
1
0 0 0
0 0 0
0 0 0
0 0 0
S
S
S
S S S S S S S S S S S
C
S S S
S S S
S S S
, 9
where
13 23 12
2
12 33
2
13 22
2
23 11 33 22 11
2 S S S S S S S S S S S S + S .
This is the most common symmetry found in composite lamina, particularly
filamentary lamina. These expressions are also referred to as the on-axis
properties.
LAMI NA MECHANI CS
TRANSVERSELY ISOTROPIC MATERIALS
Materials which display an infinite number of axes of symmetry in one plane
in response to mechanical loads are defined as transversely isotropic. These
materials have 5 independent material constants and the constitutive relation is:
( )
'
1
1
1
1
1
1
1
1
]
1
'
6
5
4
3
2
1
12 11 2
1
44
44
33 13 13
13 11 12
13 12 11
6
5
4
3
2
1
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0
0 0 0
0 0 0
C C
C
C
C C C
C C C
C C C
10
( )
'
1
1
1
1
1
1
1
1
]
1
'
6
5
4
3
2
1
12 11
44
44
33 13 13
13 11 12
13 12 11
6
5
4
3
2
1
2 0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0
0 0 0
0 0 0
S S
S
S
S S S
S S S
S S S
. 11
ISOTROPIC MATERIALS
Materials which display an infinite number of axes of symmetry in response
to mechanical loads are defined as isotropic. These materials have 2 independent
material constants and the constitutive relation is:
( )
( )
( )
'
1
1
1
1
1
1
1
1
]
1
'
6
5
4
3
2
1
12 11 2
1
12 11 2
1
12 11 2
1
11 12 12
12 11 12
12 12 11
6
5
4
3
2
1
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0
0 0 0
0 0 0
C C
C C
C C
C C C
C C C
C C C
12
LAMI NA MECHANI CS
( )
( )
( )
'
1
1
1
1
1
1
1
1
]
1
'
6
5
4
3
2
1
12 11
12 11
12 11
11 12 12
12 11 12
12 12 11
6
5
4
3
2
1
2 0 0 0 0 0
0 2 0 0 0 0
0 0 2 0 0 0
0 0 0
0 0 0
0 0 0
S S
S S
S S
S S S
S S S
S S S
. 13
ENGI NEERI NG PROPERTI ES FOR ORTHOTROPI C MATERI ALS
Engineering properties are more commonly used to describe the mechanical
behavior of materials than stiffnesses or compliances. The engineering properties
that you are familiar with are Youngs modulus (also known as modulus of
elasticity, or elastic modulus), Shear modulus (also known as modulus of rigidity),
and Poissons ratio. For an orthotropic material there are three principal
directions, associated with the three directions of material symmetry. Associated
with each of these directions are a Youngs modulus,
i
E , a shear modulus,
ij
G and a
Poissons ratio,
ij
. The properties are easily shown to be related to the
compliances. The result is:
'
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
]
1
'
6
5
4
3
2
1
12
13
23
3 2
23
1
13
3
32
2 1
12
3
31
2
21
1
6
5
4
3
2
1
1
0 0 0 0 0
0
1
0 0 0 0
0 0
1
0 0 0
0 0 0
1
0 0 0
1
0 0 0
1
G
G
G
E E E
E E E
E E E
. 14
Note in this expression that there are three Youngs moduli, one in each of the
principal material directions, three shear moduli, one for each principal material
direction pairs, and six Poissons ratios. Because the compliance matrix must be
symmetric, as discussed earlier, there are only three unique Poissons ratios. The
Poissons rations are interrelated by the expression:
ji
j
i
ij
E
E
. 15
LAMI NA MECHANI CS
Through linear algebraic manipulations the stiffness matrix can be expressed
in terms of the nine engineering properties. This relationship is:
[ ]
1
1
1
1
1
1
1
1
1
1
1
1
]
1
+ +
+ +
+ +
12
13
23
2
2
1
2
12 2 1
2
2
1
13 12 23 1
2 1
23 12 13
2
2
1
13 12 23 1
3
2
1
2
13 3 1
3 2 1
23 13 3 12 2
2 1
23 12 13
3 2 1
23 13 3 12 2
3
2
2
2
23 3 2
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0
0 0 0
0 0 0
G
G
G
E E
E E
E E
E
E E
E E
E
E E
E E
E E E
E E
E E E E E
E E
E E
E E
C
C C C
C C C
C C C
, 16
where
3 2
2
1
13 23 12 2 1
2
12
2
2
2
13 3 2
2
23 3 1 3 2 1
2
E E E
E E E E E E E E E E
2
C

.
PLANE STRESS ORTHOTROPI C CONSTI TUTI VE RELATI ON
The plane stress assumption can be applied to analysis of orthotropic lamina.
When the plane stress assumption is applied the constitutive relation is simplified
from a 6 6 matrix expression to a 3 3 expression. For the compliance expression
a simple static condensation is performed to reduce the system to a system of three
equations. The plane stress compliance equation is:
'
1
1
1
]
1
'
6
2
1
66
22 12
12 11
6
2
1
0 0
0
0
S
S S
S S
. 17
Note in this expression that there is no change to the compliances, that is reduced
compliance terms are the same as those in the three dimensional compliance
expression.
Reduction of the stiffness expression is not as simple as the compliance
equation. To simplify the stiffness expression one must solve for the out-of-plane
normal strain in terms of the in-plane stresses and stiffnesses and substitute the
results back into the stiffness expression, i. e.,
LAMI NA MECHANI CS
{ } [ ]{ }
( )
0
0
5
4
33
2 23 1 13
3

+

C
C C
C
,
( )
'
1
1
1
1
1
1
1
1
]
1
'
6
33
2 23 1 13
2
1
66
55
44
33 23 13
23 22 12
13 12 11
6
2
1
0
0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0
0 0 0
0 0 0
0
0
0
C
C C
C
C
C
C C C
C C C
C C C
,
which gives:
'
1
1
1
1
1
1
]
1
'
6
2
1
66
33
13
22
33
23
12
33
23
12
33
13
11
6
2
1
0 0
0
0
C
C
C
C
C
C
C
C
C
C
C
C
C

This expression is also written as:
'
1
1
1
]
1
'
6
2
1
66
22 12
12 11
6
2
1
0 0
0
0
Q
Q Q
Q Q
, 18
where [ ] Q is the so-called reduced stiffness matrix.
REDUCED STIFFNESSES IN TERMS OF ENGINEERING PROPERTIES
The reduced stiffnesses can also be written in terms of the engineering
properties. This relationship is:
[ ]
1
1
1
1
1
1
1
]
1
12
2
12 2 1
2 1
2
12 2 1
2 1 12
2
12 2 1
2 1 12
2
12 2 1
2
1
0 0
0
0
G
E E
E E
E E
E E
E E
E E
E E
E
Q . 19
LAMI NA MECHANI CS
The reduced compliances are the same as the compliance terms.
OFF-AXI S PROPERTI ES OF ORTHOTROPI C LAMI NA
As discussed earlier there are four coordinate systems that are used in the
analysis of orthotropic lamina. The stress-strain relationship in the principal
material directions is referred to as on-axis. The behavior in the structural
directions is referred to as off-axis behavior. Shown earlier were the
transformations of stress and strain from the principal material directions to the
structural directions. Writing:
{ } [ ]{ }
xy
T
12

where { }
12
are the stresses in the principal material directions, { }
xy
are the
stresses in the structural directions, and [ ] T is the stress transformation matrix,
then we can write the inverse relation,
{ } [ ] { }
12
1

T
xy

where [ ]
1
T is the inverse of the stress transformation matrix. Noting that
{ } [ ]{ }
xy
T
12
,
where { }
12
are the strains in the principal material directions, { }
xy
are the strains
in the structural directions, and [ ] T is the strain transformation matrix, and
{ } [ ]{ }
12 12
Q ,
then we can write
{ } [ ] { } [ ] [ ]{ } [ ] [ ][ ]{ } [ ]{ }
xy xy xy
Q T Q T Q T T
1
12
1
12
1
,
or
'
1
1
1
]
1
'
s
y
x
ss ys xs
ys yy xy
xs xy xx
s
y
x
Q Q Q
Q Q Q
Q Q Q
, 20
where [ ] Q is the off-axis reduced stiffness matrix. When the matrix multiplications
are carried out the off-axis reduced stiffnesses are found to be:
LAMI NA MECHANI CS
( )
( ) ( )
( ) ( )
'
1
1
1
1
1
1
1
1
]
1
'
66
12
22
11
2 2 2 2 3 3
2 2 2 2 3 3
2
2 2 2 2 2 2 2 2
2 2 4 4 2 2 2 2
2 2 2 2 4 4
2 2 2 2 4 4
2
2
2
4
4 2
4 2
Q
Q
Q
Q
n m mn n m mn n m mn
n m mn n m mn mn n m
n m n m n m n m
n m n m n m n m
n m n m m n
n m n m n m
Q
Q
Q
Q
Q
Q
ys
xs
ss
xy
yy
xx
, 21
again, m and n are defined as before.
Through a similar process the off-axis compliance relation can be shown to be:
'
1
1
1
]
1
'
s
y
x
ss ys xs
ys yy xy
xs xy xx
s
y
x
S S S
S S S
S S S
, 22
where [ ] S is the off-axis reduced compliance matrix. When the matrix
multiplications are carried out the off-axis reduced compliances are found to be:
( )
( ) ( )
( ) ( )
'
1
1
1
1
1
1
1
1
]
1
'
66
12
22
11
2 2 2 2 3 3
2 2 2 2 3 3
2
2 2 2 2 2 2 2 2
2 2 4 4 2 2 2 2
2 2 2 2 4 4
2 2 2 2 4 4
2 2 2
2 2 2
8 4 4
2
2
S
S
S
S
n m n m n m n m
n m n m n m n m
n m n m m n
n m n m n m
S
S
S
S
S
S
ys
xs
ss
xy
yy
xx
, 23
INVARIANT PROPERTIES OF ORTHOTROPIC LAMINA
When the powers of the trigonometric functions in the transformation
equation for the stiffnesses above are converted to the multiple angle form a set of
combinations of the lamina properties are discovered. These combinations are
defined as invariant stiffness properties of the lamina. The invariants,
Q
i
U , are:
[ ]
[ ]
[ ]
66 12 22 11 3
22 11 2
66 12 22 11 1
4 2
8
1
2
1
4 2 3 3
8
1
Q Q Q Q U
Q Q U
Q Q Q Q U
Q
Q
Q
+

+ + +
24
LAMI NA MECHANI CS
[ ]
[ ]
66 12 22 11 5
66 12 22 11 4
4 2
2
1
4 6
8
1
Q Q Q Q U
Q Q Q Q U
Q
Q
+ +
+ +

and the off-axis stiffness properties are related to the invariant properties by:
'
1
1
1
1
1
1
1
1
]
1
'
Q
Q
Q
Q
Q
Q
ys
xs
ss
xy
yy
xx
U
U
U
U
U
U
Q
Q
Q
Q
Q
Q
3
2
2
1
2
1
5
4
1
1
1
4 sin - 2 sin 0
4 sin 2 sin 0
4 cos - 0
4 cos - 0
4 cos 2 cos -
4 cos 2 cos
. 25
Similarly, invariant compliance properties,
S
i
U , are found to be:
[ ]
[ ]
[ ]
[ ]
[ ]
66 12 22 11 5
66 12 22 11 4
66 12 22 11 3
22 11 2
66 12 22 11 1
2
2
1
6
8
1
2
8
1
2
1
2 3 3
8
1
S S S S U
S S S S U
S S S S U
S S U
S S S S U
S
S
S
S
S
+ +
+ +
+

+ + +
26
and the off-axis compliances are related to the invariant properties by:
1
1
4
2
5
3
cos2 cos4
-cos2 cos4
1
0 -cos4
0 -4cos4
0 sin2 2sin4
0 sin2 -2sin4
S
xx
S
yy
S
xy
S
S
ss
S
xs
ys
S
U
S
U
S
U
U
S
U
U
S
S
1
1

1

1
1 ' ; ' ;
1

1

1
1
]
. 27
OFF-AXIS ENGINEERING PROPERTIES OF ORTHOTROPIC LAMINA
J ust as with the on-axis properties, the off-axis properties can be expressed
in terms off-axis engineering properties. Associated with the structural directions
is a Youngs modulus,
y x
E E , , a shear modulus,
s
G a Poissons ratio,
xy
, and
LAMI NA MECHANI CS
coefficients of mutual influence,
y s x s s y s x , , , ,
, , , . The properties are easily shown
to be related to the compliances. The result is:
'
1
1
1
1
1
1
1
]
1
'
s
y
x
s y
y s
x
x s
s
s y
y x
xy
s
s x
x
xy
x
s
y
x
G E E
G E E
G E E
1
1
1
, ,
,
,
. 28
The coefficients of mutual influence, as defined by Lekhnitski, characterize the
stretching due to shear, first kind, and the shearing due to stretching, second kind.
Specifically, the coefficients of mutual influence of the first kind are defined as:
y x i
s
i
s i
, ,
,

and the coefficients of mutual influence of the second kind are defined as:
y x i
i
s
i s
, ,
,

.
Symmetry of the compliance matrix applies, therefore,
x
x s
s
s x
E G
, ,

.
and
y
y s
s
s y
E G
, ,

.
The off-axis engineering properties can be written in terms of the on-axis
engineering properties. These relations are:
4
2
2 2
1
12
12
4
1
1 2 1 1 1
n
E
n m
E G
m
E E
x
+
,
_

+ ,
4
2
2 2
1
12
12
4
1
1 2 1 1 1
m
E
n m
E G
n
E E
y
+
,
_

+ ,
LAMI NA MECHANI CS
( )
4 4
12
2 2
12 1
12
2 1
1 1 4 2 2
2
1
n m
G
n m
G E E E G
s
+ +
,
_
+ + ,
( )
1
]
1
,
_
+ +

2 2
12 2 1
4 4
1
12
1 1 1
n m
G E E
n m
E
E
x xy
,
1
]
1
,
_
,
_
+
2
12 1
12
2
2
12 1
12
1
,
1 2 2 1 2 2
n
G E E
m
G E E
mn E
x x s
,
1
]
1
,
_
,
_
+
2
12 1
12
2
2
12 1
12
1
,
1 2 2 1 2 2
m
G E E
n
G E E
mn E
y y s
,
where mand n are as defined before.
STRENGTH OF LAMINA
TENSOR POLYNOMI AL FAI LURE CRI TERI ON
The principle that underlies a failure criterion is that a function of material
properties and current load values can be defined that separates the domain of safe
response from the failed domain. Figure 3 shows a typical failure surface in the
principal material direction stress space. Note that the plot in the figure is a
two-dimensional slice out of the three-dimensional failure surface. Also note that
the horizontal and vertical scales are not equal. The region inside the curve is the
safe region while the region outside the surface is failed. Therefore failure is
defined to occur on the surface. The function that defines the surface is defined as
the failure criterion function. The function may be derived phenomenologically or
empirically. For metals the yield criteria proposed by Tresca and Henke and Von
Mises have theoretical basis in the mechanics of materials and are supported by
experimental evidence. The application of a failure criterion to composite materials
is much more problematic. There is minimal theoretical basis that justifies the use
of a failure criterion. However, there is ample experimental evidence that the
application of a failure criterion to composite materials.
Current State
- 50
- 40
- 30
- 20
- 10
0
10
- 600 - 500 - 400 - 300 - 200 - 100 0 100 200 300

1 1
kpsi

2 2

k
p
s
i
0. 835 f =
1. 170 R =
Figure 3. Failure Envelop in the Stress Domain.
STRENGTH OF LAMI NA
The tensor polynomial in its most general form was presented first by
Goldenblat' and Knoppov as an expansion involving the current stress state and
various strength tensors. At failure this tensor polynomial is expressed as
( ) ( ) ( ) ( ) ( ) L + + + +

op mn kl ij ijklmnop mn kl ij ijklmn kl ij ijkl ij ij
F F F F f 1 ,
where
ij
is the stress tensor expressed as a rank two tensor with indices from 1 to
3, and
ij
F ,
ijkl
F ,
ijklmn
F , and
ijklmnop
F are the strength tensors. Following the standard
notation for tensor mathematics, there is an implicit summation from 1 to 3 for each
repeated index in the expression. From this expression, when failure occurs the
polynomial expansion equals one, by definition.
Tsai and Wu presented the tensor polynomial in the form in which all
exponents are one. They showed that there is no loss in generality from the form of
Goldenblat' and Knoppov. Thus, the Tsai-Wu form of the tensor polynomial is
( ) L + + + +
op mn kl ij ijklmnop mn kl ij ijklmn kl ij ijkl ij ij
F F F F f 1 .
Alternately, using contracted or compact notation the tensor polynomial is
( ) L + + + +
l k j i ijkl k j i ijk j i ij i i
F F F F f 1 ,
where
i
is the stress tensor expressed as a rank one tensor with indices from 1 to
6, and
i
F ,
ij
F ,
ijk
F , and
ijkl
F are the strength tensors. Note in this expression the
convention used in tensor mathematics is used with the exception that the implicit
summation over repeated indices is from 1 to 6. The stresses used in this
expression may be in the principal material directions, the principal stress
directions, or some structural direction. I f stresses other that those from the
principal material directions are used, the strength tensors must be transformed
appropriately. When the tensor polynomial, ( ) f , equals one failure occurs. When
the function is less than one the lamina is not failed and conversely when the
function is greater than one the lamina is failed.
QUADRATI C FAI LURE CRI TERI ON
I n the original development of Tsai and Wu, the general tensor polynomial
failure criterion was simplified to a quadratic polynomial. I f one considers the on-
axis, or principal material direction, strength behavior of the material loaded in
plane stress, the quadratic form of the tensor polynomial is
( )
6 2 62 6 2 26 6 1 61 6 1 16 2 1 21 2 1 12
2
6 66
2
2 22
2
1 11
6 6 2 2 1 1
+ + + + +
+ + +
+ + +
F F F F F F
F F F
F F F f
,
STRENGTH OF LAMI NA
or
( )
( ) ( ) ( )
6 2 62 26 6 1 61 16 2 1 21 12
2
6 66
2
2 22
2
1 11
6 6 2 2 1 1
+ + + + +
+ + +
+ + +
F F F F F F
F F F
F F F f
.
Note that the effect of the plane stress assumption is to eliminate the terms
55 14 13 5 4 3
, , , , , F F F F F F K , thus reducing the expression to the form shown. Now
assume, without loss of generality, that the strength tensor is a symmetric tensor.
Thus,
ji ij
F F . Therefore, the failure criterion can be expressed as
( )
6 2 26 6 1 16 2 1 12
2
6 66
2
2 22
2
1 11
6 6 2 2 1 1
2 2 2 + + + + +
+ + +
F F F F F F
F F F f
.
Consider two stress states ( )
T
6 2 1
, , , and ( )
T
6 2 1
, , that are both found to
cause failure. Thus,
6 2 26 6 1 16 2 1 12
2
6 66
2
2 22
2
1 11
6 6 2 2 1 1
2 2 2
1
+ + + + +
+ + +
F F F F F F
F F F

and
6 2 26 6 1 16 2 1 12
2
6 66
2
2 22
2
1 11
6 6 2 2 1 1
2 2 2
1
+ + +
+ +
F F F F F F
F F F
.
Taking the difference between these equations one finds that
6 2 26 6 1 16 6 6
4 4 2 0 F F F ,
or
( )
6 2 26 1 16 6
2 2 2 0 F F F .
I n these load states 0
6
by definition, thus, the term in the brackets,
2 26 1 16 6
2 2 F F F , must be zero. However, the applied stresses,
2 1
, , are
independent variables, that is this expression must hold true for all stress states.
Therefore, this expression can be zero if and only if, 0
26 16 6
F F F . Consequently,
the quadratic form of the tensor polynomial failure criterion simplifies to:
( )
2 1 12
2
6 66
2
2 22
2
1 11 2 2 1 1
2 + + + + + F F F F F F f .
STRENGTH OF LAMI NA
The elements of the strength tensors can be related to the strengths
measured in standard tensile tests. These relations yield:
2 66 22 11 2 1
1
and ,
1
,
1
,
1 1
,
1 1
S
F
Y Y
F
X X
F
Y Y
F
X X
F
C T C T C T C T
,
where
T
X and
T
Y are the tensile strengths in the first and second principal
material directions, respectively,
C
X and
C
Y are the compressive strengths in the
first and second principal material directions, respectively, and S is the principal
material direction shear strength. Unfortunately, the interaction term,
12
F , cannot
be determined from a simple uniaxial tensile test. Various interaction terms are
presented in the literature. I n the Tsai-Hill failure criterion, the tensile and
compressive strengths are assumed to be equal and the interaction term is given as
2
2
1
X
. Hahn presented the interaction term for unequal tensile and compressive
strengths as
C T C T
Y Y X X 2
1
. Fundamental constraints on the form of the quadratic
tensor polynomial require that
2
12 22 11
F F F . Hahns interaction term satisfies this
requirement.
The plot shown in Figure 3 is a plot of a quadratic tensor polynomial failure
criterion. Note the characteristically ellipsoidal shape of the failure surface, this is
directly related to the quadratic form of the failure function. Further note, that the
failure surface is concave, this also is directly related to the form of the failure
function. I n three-dimensional stress space the failure surface resembles a cigar or
thin rugby ball (a football is too pointed at the ends). The function value shown in
the figure represents the criterion value at the current stress point.
QUADRATIC TENSOR POLYNOMIAL TRANSFORMATION TO OFF-AXIS FORM
Noting that a stress state can be transformed from one coordinate system to
another through the orthogonal transformation
'
1
1
1
]
1
'
s
y
x
n m mn mn
mn m n
mn n m
2 2
2 2
2 2
6
2
1
2
2
,
where cos m and sin n , the result of the tensor polynomial failure criterion
must be invariant with respect to coordinate transformation. Another way to state
this principle is: any stress state that causes failure in one coordinate system must
also cause failure in any other stress state obtained through simple coordinate
system rotation. Substituting structural direction stresses for the principal
material direction stresses in the quadratic tensor polynomial yields:
STRENGTH OF LAMI NA
( )
s y ys s x xs y x xy s ss y yy x xx
s s y y x x
F F F F F F
F F F f
+ + + + +
+ + +
2 2 2
2 2 2
,
where
'
1
1
1
]
1
'
2
1 2 2
2 2
2 2
F
F
mn mn
m n
n m
F
F
F
s
y
x

and
( )
( ) ( )
( ) ( )
'
1
1
1
1
1
1
1
1
]
1
'
66
12
22
11
2 2 2 2 3 3
2 2 2 2 3 3
2
2 2 2 2 2 2 2 2
2 2 4 4 2 2 2 2
2 2 2 2 4 4
2 2 2 2 4 4
2 2 2
2 2 2
8 4 4
2
2
F
F
F
F
n m n m n m n m
n m n m n m n m
n m n m m n
n m n m n m
F
F
F
F
F
F
ys
xs
ss
xy
yy
xx
,
As with the stiffness properties the strength tensors can be expressed in
terms of invariant properties. For the strength tensors the invariants are
[ ]
[ ]
2 1 2
2 1 1
2
1
2
1
F F u
F F u
f
f

+

[ ]
[ ]
[ ]
[ ]
[ ]
66 12 22 11 5
66 12 22 11 4
66 12 22 11 3
22 11 2
66 12 22 11 1
2
2
1
6
8
1
2
8
1
2
1
2 3 3
8
1
F F F F U
F F F F U
F F F F U
F F U
F F F F U
f
f
f
f
f
+ +
+ +
+

+ + +
.
The strength tensors in terms of the strength invariants are:
STRENGTH OF LAMI NA
'
1
1
1
]
1
'
f
f
s
y
x
u
u
F
F
F
2
1
2 sin 2 0
2 cos 1
2 cos 1

'
1
1
1
1
1
1
1
1
]
1
'
f
f
f
f
f
f
ys
xs
ss
xy
yy
xx
U
U
U
U
U
U
F
F
F
F
F
F
3
2
5
4
1
1
1
4 sin 2 - 2 sin 0
4 sin 2 2 sin 0
4 cos 4 - 0
4 cos - 0
4 cos 2 cos -
4 cos 2 cos
.
I t should be noted here that the stresses in the principal material directions are not
invariant with respect to rotation. Therefore, the failure criterion formulated in
terms of the principal material direction stresses is not necessarily invariant with
respect to coordinate rotation, a primary requirement for a failure criterion.
However, Bower and Koedam developed a proof to establish that the failure
criterion formulated in terms of the principal stresses is invariant. Therefore, the
analyst is free to use whichever formulation is most convenient, provided of course
the appropriate application of the required transformations.
STRAIN FORMULATION OF THE TENSOR POLYNOMIAL FAILURE CRITERION
Given the Tsai-Wu second order form of the tensor polynomial failure
criterion in terms of stress and the plane stress constitutive relation
j ij i
Q ,
where the
ij
Q are the so-called reduced stiffnesses, with 6 ,& 2 , 1 , j i , the criterion,
( ) g , can be expressed in terms of strain such that
( )
l k jl ik ij j ij i
Q Q F Q F g + 1
or
( )
j i ij i i
G G g + 1 ,
where
ij i j
Q F G and
jl ik ij kl
Q Q F G . Given a state of plane stress, the failure
criterion is rewritten as:
( )
2 1 12
2
6 66
2
2 22
2
1 11 2 2 1 1
2 1 + + + + + G G G G G G g .
STRENGTH OF LAMI NA
The same arguments are used in this development regarding symmetry with
respect to shear as in the preceding development. I n this expression the coefficients
are:
( )
22 2 12 1 2
12 2 11 1 1
2
66 66 66
22 12 22
2
12 22 11 12 12 11 11 12
2
22 22 22 12 12
2
12 11 22
2
12 22 12 11 12
2
11 11 11
2
2
Q F Q F G
Q F Q F G
Q F G
Q Q F Q Q Q F Q Q F G
Q F Q Q F Q F G
Q F Q Q F Q F G
+
+
+ + +
+ +
+ +
.
OFF-AXIS FORM OF THE STRAIN FORMULATION OF THE TENSOR POLYNOMIAL
FAILURE CRITERION
As with the previous development, note that a strain state can be
transformed from one coordinate system to another through the orthogonal
transformation:
'
1
1
1
]
1
'
s
y
x
n m mn mn
mn m n
mn n m
2 2
2 2
2 2
6
2
1
2 2
,
where m and n are as defined previously. Substituting structural direction strains
for the principal material direction strains in the strain form of the quadratic tensor
polynomial yields:
( )
s s y y x x
G G G G G G
G G G g
+ + + + +
+ + +
2 2 2
2 2 2
,
where
'
1
1
1
]
1
'
2
1 2 2
2 2
G
G
mn mn
m n
n m
G
G
G
s
y
x

and
STRENGTH OF LAMI NA
( )
( ) ( )
( ) ( )
'
1
1
1
1
1
1
1
1
]
1
'
66
12
22
11
2 2 2 2 3 3
2 2 2 2 3 3
2
2 2 2 2 2 2 2 2
2 2 4 4 2 2 2 2
2 2 2 2 4 4
2 2 2 2 4 4
2
2
2
4
4 2
4 2
G
G
G
G
n m n m n m n m
n m n m n m n m
n m n m m n
n m n m n m
G
G
G
G
G
G
ys
xs
ss
xy
yy
xx
,
As with the stiffness properties and strength tensors can be expressed in
terms of invariant properties. For the strength tensors the invariants are
[ ]
[ ]
2 1 2
2 1 1
2
1
2
1
G G u
G G u
g
g

+

[ ]
[ ]
[ ]
[ ]
[ ]
66 12 22 11 5
66 12 22 11 4
66 12 22 11 3
22 11 2
66 12 22 11 1
4 2
2
1
4 6
8
1
4 2
8
1
2
1
4 2 3 3
8
1
G G G G U
G G G G U
G G G G U
G G U
G G G G U
g
g
g
g
g
+ +
+ +
+

+ + +
.
The strain form of the strength tensors in terms of the invariants are:
'
1
1
1
]
1
'
g
g
s
y
x
u
u
G
G
G
2
1
2 sin 0
2 cos 1
2 cos 1

'
1
1
1
1
1
1
1
1
]
1
'
g
g
g
g
g
g
ys
xs
ss
xy
yy
xx
U
U
U
U
U
U
G
G
G
G
G
G
3
2
2
1
2
1
5
4
1
1
1
4 sin - 2 sin 0
4 sin 2 sin 0
4 cos - 0
4 cos - 0
4 cos 2 cos -
4 cos 2 cos
.
STRENGTH OF LAMI NA
R-FACTOR ANALYSI S
I n the analysis of traditional engineering materials it is common practice to
determine a factor of safety. For the traditional engineering materials one method
of determining the factor of safety is to take the ratio of the Von Mises-Henke stress
for the load to the tensile yield stress for the material. I n the analysis of composite
materials an analogous process is used, the R-factor analysis. An R-factor is not a
safety factor. An R-factor is the factor that scales the initial stress state to failure.
Strictly speaking a factor of safety would describe how far away from failure the
system is at the prescribed load. For a composite material this would require
finding the shortest distance to the failure surface from the prescribed stress state
in stress space. An R-factor scales the stress state along a radial in stress space,
which in most cases is not the shortest distance to the failure surface.
Based on the previous discussion define a new stress state, { } , that is a
multiple, R, of the initial stress state:
'
'
s
y
x
s
y
x
R .
This stress state is defined to be a stress state that causes failure. Therefore:
( )
s s y y x x
F F F F F F
F F F
f
+ + + + +
+ + +

2 2 2
1
2 2 2
,
note the above is the off-axis form of the failure function, by appropriate exchange
of indices it can also be used to show the on-axis form. I n this case the initial stress
state is known. Therefore, the unknown scale factor, R, can be found through the
solution of the quadratic equation:
( )
( )
s s y y x x
F F F F F F R
F F F R
+ + + + +
+ + + +
2 2 2
1 0
2 2 2 2
.
Fortuitously, the form of this equation is such that we are guaranteed to get a real
solution. Thus, the R-factor is:
a
a b b
R
2
4
2
2 , 1
+ t
,
where
STRENGTH OF LAMI NA
( )
( )
s s y y x x
F F F b
F F F F F F a
+ +
+ + + + + 2 2 2
2 2 2
.
Note this produces two real values. One should be positive and the other negative.
The factor of interest is the positive value. The negative value corresponds to the
factor needed to intersect the failure surface along the opposite radial in stress
space. An analogous form exists for the strain form of the failure function. I t is:
a
a b b
R
2
4
2
2 , 1
+ t
,
where
( )
( )
s s y y x x
G G G b
G G G G G G a
+ +
+ + + + + 2 2 2
2 2 2
.

CLASSICAL LAMINATION THEORY
HI STORY
The history of plate analysis began with Leonard Euler (1707-1783). I n his
approach he presented two methods: a direct method, which is an application of the
equilibrium principle; and the method of final causes, which is a variational
principle. I n this work, the plate was treated as a membrane made up of strings.
J acque Bernoulli (1759-1789) was the next to contribute to the development
of the plate theory. I n his attempt, he assumed that a plate is made up of beams.
This caused him to miss the twist term and have an incorrect stiffness constant due
to missing the Poisson stiffening. Next, Ernst Florens Fredric Chadni (1756-1827)
approached the development of a plate theory by using acoustical vibration with
sand.
I n 1809, at the direction of Napoleon, the French Academy of Science
announced a competition to develop a more accurate plate theory. October 1811
was established as the closing date for submission of the theories. The contest was
won by Mlle. Sophie Germain (1776-1831). I n this theory, she proposed that the
energy in the plate was
A A d
1 1
2
2 1
'
.
This theory was the best to date and only missed by failure to include the twist
energy.
LaGrange then proposed a plate theory in which
0 2
2
2
4
4
2 2
4
4
4
'
t
w
y
w
y x
w
x
w
k .
The French Academy of Science recognized the available plate theory was
inadequate and announced a second closure date of October 1813. No better theory
was put forward and so the Academy announced a third closure date, October 1816.
CLASSI CAL LAMI NATI ON THEORY
S. D. Poisson (1781-1840) put forth a theory in which the energy in the plate
was
A m A d
1 1 1 1
2
2
2
1
2
2 1
'
1
]
1
+
1
]
1
.
The form of the energy is correct. However, Poisson had the wrong values for A and
m.
On August 14, 1820, Navier (1785-1836) recognized that plane sections
remain plane in a plate in bending. This fundamental observation was published
in 1823. Navier was responsible for the first successful solutions for plate bending.
PRELI MI NARI ES
A laminate is an engineering structure formed by bonding two or more
laminae together. The laminae are assumed to be perfectly bonded at their
interfaces. Thus, the displacements, and hence the strains, are continuous
throughout the laminate. The laminae of the structure may have isotropic,
transversely isotropic, orthotropic, anisotropic properties, differing thicknesses, or
differing principal material direction orientations.

h/2
h/2
x y
z
Mid-plane

Figure 4. Schematic drawing of a laminated plate element showing
coordinate system and plate thickness.
Figure 4 shows a schematic drawing of a typical laminated plate element.
The thickness of the plate shown is h and the origin of the coordinate system is on
the mid-plane of the plate. The coordinate system used is a typical aerospace
coordinate system with x- and y-directions oriented in the plane of the plate and the
z-direction perpendicular to the plane in a downward direction.
FORCE -- MOMENT RESULTANTS
The internal force and moment resultants are derived from the stresses in
the plate element. These resultants are expressed in units of force per unit length
and force times length per unit length, respectively. The normal force resultant in
the x-direction on an internal face with normal in the x-direction is ( ) y x N
x
, and is
an integral through the thickness of the plate of the x-direction normal stress,
( ) z y x
x
, , , on that face. Therefore,
( ) ( )

2 /
/2 -
d , , ,
h
h
x x x
z z y x y x N N . 29
The normal force resultant in the y-direction is similarly defined as an integral of
the y-direction normal stress, ( ) z y x
y
, , . I t is
( ) ( )

2 /
/2 -
d , , ,
h
h
y y y
There is no normal force resultant in the z-direction due to the consideration of an
element that includes the total plate thickness.
The shear force resultant in the y-direction on an internal face with normal
in the x-direction is ( ) y x N
xy
, . This force resultant is defined as an integral through
the thickness of the x-y shear stress, ( ) z y x
s
, , , on that face. Therefore,
( ) ( )

2 /
/2 -
d , , ,
h
h
s xy xy
The shear force resultant in the x-direction on an internal face with normal in the
y-direction is ( ) y x N
yx
, . This force resultant is defined as an integral through the
thickness of the y-x shear stress, ( ) z y x
yx
, , , on that face. Noting that the x-y and y-x
shear stresses are equal, i.e., ( ) ( ) ( ) z y x z y x z y x
yx xy s
, , , , , , , then the x-y and y-x
shear resultant forces must also be equal, i.e., ( ) ( ) y x N y x N
yx xy
, , .
The shear force resultants in the z-direction on the internal faces with
normals in the x- and y-directions are ( ) y x N
xz
, and ( ) y x N
yz
, , respectively. They are
defined as
( ) ( )

2 /
/2 -
d , , ,
h
h
xz xz xz
z z y x y x N N 32
and
( ) ( )

2 /
/2 -
d , , ,
h
h
yz yz yz
The moment resultant in the x-direction on an internal face with normal in
the x-direction is defined to be ( ) y x M
xx
, . This resultant is found by integrating the
differential moment in the x-direction through the thickness of the plate. To
develop the differential moment in the x-direction consider the drawing shown in
Figure 5. A differential element located at (y,z) is indicated by the shaded area in
the figure. The stresses on this element are also shown in the figure. The normal
stress in the x-direction does not contribute to the moment about the x-axis while
the shear stresses ( ) z y x
xy
, , and ( ) z y x
xz
, , have moment arms z and y, respectively.
Therefore, the moment resultant ( ) y x M
xx
, is
( ) ( ) ( ) { }

2 /
/2 -
d , , , , ,
h
h
xz s xx xx
z z y x y z y x z y x M M . 34

xy

x
x
z
y
z
xz
y

Figure 5. Schematic showing the stresses on a face with normal in the x-direction.
Similarly, the moment resultant y-direction on an internal face with a normal in the
x-direction is
( ) ( )

2 /
/2 -
d , , ,
h
h
x xy xy
z z z y x y x M M 35
and the moment resultant z-direction on an internal face with a normal in the
x-direction is
( ) ( )

2 /
/2 -
d , , ,
h
h
x xz xz
z y z y x y x M M . 36
The moment resultants on an internal face with a normal in the y-direction are
developed in the same manner as above. The moment resultants on a face with
normal in the y-direction are
( ) ( )

2 /
/2 -
d , , ,
h
h
y yx yx
z z z y x y x M M , 37
( ) ( ) ( ) { }

2 /
/2 -
d , , , , ,
h
h
yz s yy yy
z z y x x z y x z y x M M , 38
and
( ) ( )

2 /
/2 -
d , , ,
h
h
y yz yz
z x z y x y x M M . 39
EQUI LI BRI UM OF A PLATE ELEMENT
Consider an infinitesimal plate element as shown in Figures 6 and 7.
Figure 6 shows the internal force resultants acting on the element along with the
externally applied force, y x p , due to a distributed load over the element. Note
that the infinitesimal element is shown in the undeformed configuration. Thus, the
various internal force resultants lie in their respective coordinate directions.
Figure 7 shows the internal moment resultants acting on the element. The
moments are indicated with two-headed arrows following the right-hand rule.
Again, the element is shown in the undeformed configuration. The forces and
moments shown in the figures are as defined previously.
Now assume that the in-plane accelerations of the plate are small when
compared to the out of plane accelerations. Applying the equilibrium conditions to
the element, summation of the forces in the x-direction yields:
( ) ( ) 0 + + + + x N N x N y N N y N
yx yx yx x x x
,
or
0 + x N y N
yx x
.
Then noting, x
x
N
N
x
x

and y
y
N
N
x
xy

, the sum of the forces in the

x-direction reduces to
0
y x
y
N
y x
x
N
yx
x
,
which further reduces to
0
y
N
x
N
yx
x
. 40
z
x y

Figure 6. Schematic drawing of an infinitesimal element showing the internal force
resultants acting on the element.

y x p
y N
x

y N
xz
x N
yx
y N
xy

x N
yz
x N
y

x
y
( ) y N N
xz xz
+
( ) x N N
yz yz
+
( ) y N N
xy xy
+
( ) x N N
yx yx
+
( ) x N N
y y
+ ( ) y N N
x x
+
The summation of forces in the y-direction reduces to
0
y
N
x
N
y xy
. 41
As drawn in Figure 6 the summation of the forces in the z-direction is
( ) ( ) ( ) ( )
2
2
, , ,
t
w
y x h y x y x t y x p x N N x N y N N y N
yz yz yz xz xz xz

+ + + + + . 42
However, this equation does not allow for the inclusion of buckling effects. To
develop the buckling terms one must apply equilibrium in the deformed
configuration. When the effect of deformation is considered, the normal force in the
x-direction will have a component in the z-direction equal to
y x
x
w
x
N
y x
x
w
N
x
x

2
2
, 43
where
x
N is the applied buckling force in the x-direction on the x-face. Similarly,
the normal force in the y-direction will have a component in the z-direction equal to
y x
y
w
y
N
y x
y
w
N
y
y

2
2
, 44
where
y
N is the applied buckling force in the y-direction on the y-face, and the
in-plane shear forces have a component in the z-direction equal to
y x
x
w
x
N
y x
y
w
y
N
y x
y x
w
N
xy yx
xy

2
2 , 45
where
xy
N is the applied buckling force in the y-direction on the x-face, and
yx
N is
the applied buckling force in the x-direction on the y-face. Then, combining the
z-direction terms in equations 43, 44, and 45, including the results from equations
40 and 41 yields an additional z-direction term, which is
y x
y
w
N
y x
w
N
x
w
N
y xy x

'
2
2 2
2
2
2 . 46
Thus, the summation of the forces in the z-direction, including effects of buckling is
obtained by combining 42 and 46. The result on simplification is
( ) ( )
2
2
2
2 2
2
2
, , , 2
t
w
h y x t y x p
y
w
N
y x
w
N
x
w
N
x
N
y
N
y xy x
xz
yz
. 47
z
x y

Figure 7. Schematic drawing of infinitesimal element showing internal moment
resultants acting on the element.
Equilibrium of a body requires that the summation of the moments about the
center of mass be equal to the rate of change of the moment of momentum.
Assuming that the rotary inertia of the plate element is small in all directions, the
sum of the moments about the center of mass must be zero. The summation of the
moments in the x-direction about the mid-point of the element yields
0
+
y
M
x
M
N
yx
xx
yz
. 48
Summation of the moments in the y-direction about the mid-point of the element
yields
0
+
x
M
y
M
N
xy yy
xz
. 49
y M
xx

y M
xz
x M
yx
y M
xy
x M
yz

x M
yy

x
y
( ) y M M
xz xz
+

( ) x M M
yz yz
+

( ) y M M
xy xy
+

( ) x M M
yx yx
+

( ) x M M
yy yy
+
( ) y M M
xx xx
+

Finally, summation of the moments in the z-direction about the mid-point of the
element yields
0 +
yx xy
yz
xz
N N
y
M
x
M
. 50
Substituting for the shear force resultants,
xz
N and
yz
N , from equations 48 and 49
into equation 47 yields
( ) ( )
2
2
2
2 2
2
2
2
2 2
2
2
, , , 2 2
t
w
h y x t y x p
y
w
N
y x
w
N
x
w
N
y
M
y x
M
x
M
y xy x
yx yy xy
, 51
which is analogous to the more familiar moment -- distributed load equation for
beams. This equation is complete for classical plate theory. I t contains four
sub-cases:
Complete -- Buckling with dynamic effects,
Dynamic -- Neglect buckling effects,
Buckling -- Neglect dynamic effects,
Static -- Neglect buckling and dynamic effects.
DI SPLACEMENT FI ELD MODEL
The next step in development of a complete plate theory for laminated plates
is the assumption of a displacement field. The Kirchoff-Love model for the
displacement of a plate element is based on observation of simple bending of a
plate. I n this model, the first assumption is
the normal strain through the plate thickness is much much less
than the normal strains in the plane of the plate,
i.e.,
y x z
>> , . The second assumption in this model is based on Navier's
observation. This assumption is
sections plane before loading remain plane after loading.
The third assumption is
plane sections initially perpendicular to the mid-plane of the plate
before loading remain perpendicular to the mid-plane after loading.
The impact of these assumptions on the deformation is shown in Figure 8.
The consequences of the second and third assumptions on the through thickness
shear strains are
0
xz
, 52
and
0
yz
. 53
Figure 9 contains a schematic of the displacement in the x and z plane of an
element of the plate in the undeformed and deformed configurations. From this
figure we determine that the displacement in the x-direction of point Q is:
( ) ( ) sin z R u Q u .

x
z
z
R
Q
z
R'
Q'
u(R)
w(R)
w(Q)
u(Q)
zsin
zcos

Figure 9. Movement of a plate element from the undeformed configuration to the
deformed position.
x-z Plane y-z Plane
Before
Deformation
After
Deformation
Before
Deformation
After
Deformation
x
z

x
z

and
y
z

y
z

Figure 8. Sections perpendicular before and after deformation.
I f we assume that the angles are small, then
( )
x
y x w

,
sin
0
and the above
expression becomes:
( ) ( )
( )
x
y x w
z y x u Q u

,
,
0
0
. 54
The tensile strain in the x-direction is:
( )
( ) ( ) ( )
2
0
2
0
, ,
x
y x w
z
x
y x u
x
Q u
Q
x

,
or
( ) ( ) ( ) y x z y x z y x
x x x
, , , ,
0 0
+ , 55
where ( ) y x
x
,
0
is the mid-plane curvature. Similarly, for the y and z plane the
displacement in the y-direction of point Q is:
( ) ( )

sin z R v Q v .
Again, applying the assumption that the angles are small, then
( )
y
y x w
sin
0
and
the above expression becomes:
( ) ( )
( )
y
y x w
z y x v Q v

,
,
0
0
. 56
The tensile strain in the y-direction is:
( )
( ) ( ) ( )
2
0
2
0
, ,
y
y x w
z
y
y x v
y
Q v
Q
y

,
or
( ) ( ) ( ) y x z y x z y x
y y y
, , , ,
0 0
+ , 57
where ( ) y x
y
,
0
is the mid-plane curvature.
Now recall that the x-y shear strain,
s
, is defined as:
x
v
y
u
s

,
( ) ( )
( )
( )
( )
,
_
+
,
_

y
y x w
z y x v
x x
y x w
z y x u
y
z y x
s
,
,
,
, , ,
0
0
0
0
,
( )
( ) ( ) ( )
y x
y x w
z
x
y x v
y
y x u
z y x
s

,
2
, ,
, ,
0
2
0 0
,
( ) ( ) ( ) y x z y x z y x
xy s s
, , , ,
0 0
, 58
where ( ) y x
xy
,
0
is the mid-plane cross curvature. Equations 55, 57, and 58, can be
combined into vector form as:
'
'
'
0
0
0
0
0
0
xy
y
x
s
y
x
s
y
x
z . 59
Applying the three dimensional orthotropic constitutive relation to equations
52 and 53 yields:
0
xz
, 60
and
0
yz
, 61
which fits with previous results. Now if one assumes that the plate is thin, then
0
z
, 62
which effectively assumes a plane stress state.
ORTHOTROPI C CONSTI TUTI VE RELATI ON
As discussed in the development of the mechanics of an orthotropic lamina,
the off-axis stress-strain relationship for an orthotropic lamina is:
'
1
1
1
]
1
'
s
y
x
ss ys xs
ys yy xy
xs xy xx
s
y
x
Q Q Q
Q Q Q
Q Q Q
, 63
where s y x j i Q
ij
& , , , , are the off-axis properties of the lamina. For a laminated
structure the mechanical properties are assumed to be constant throughout each
individual lamina. Therefore, the mechanical properties of the structure vary
through the thickness of the plate, but are constant through the thickness of the
individual lamina. Note that use of the off-axis properties does not preclude a
lamina from being oriented with principal material directions aligned with the
structural directions. The off-axis constitutive relation is used as it is more general
than the on-axis relationship.
Figure 10 shows a schematic drawing of a laminated plate with N plies, the
ply numbering system, ply coordinates, and ply thicknesses. The ply numbers
range from 1 to N, with ply 1 located on the top surface of the laminate (minimum
z-coordinate) and ply N located on the bottom surface of the laminate (maximum
z-coordinate). The ith ply has thickness t
i
, with top coordinate z
i-1
and bottom
coordinate z
i
. The total laminate thickness is h. Consequently,
2
0
h
z

and
2
h
z
N
.

1
2
i
N
N-1
x
z
-h/2 z
1
z
2
h/2
z
0
z
i
z
i+1
z
N
z
N-1
t
i
t
2
t
N-1

Figure 10. Schematic of a Laminate with N ply showing the ply numbers,
corresponding coordinates and thicknesses.
For the purposes of this development we write equation 63 with a second
subscript, k, that indicates the lamina number. This expression is:
'
1
1
1
1
1
]
1
'
k
s
k
y
k
x
k
ss
k
ys
k
xs
k
ys
k
yy
k
xy
k
xs
k
xy
k
xx
k
s
k
y
k
x
Q Q Q
Q Q Q
Q Q Q
, 64
where N k s y x j i Q
k
ij
K , 2 , 1 , & , , , , are the off-axis properties of the kth lamina and
k
ij
and
k
ij
are the associated stresses and strains. Now, considering that the
stresses in each lamina are dependent on the properties in that layer and the
properties from one layer to the next are not necessarily the same, equation 29
becomes:
( ) ( ) ( ) ( )

+ + +
N
N
z
z
N
x
z
z
x
z
z
x
h
h
x x
z z y x z z y x z z y x z z y x N
1
2
1
1
0
d , , d , , d , , d , ,
2 1
2 /
/2 -
L
( )

N
k
z
z
k
x x
k
k
z z y x N
1
1
d , , . 65
Now, substituting equations 59 and 64 into 65 yields:
'
'
N
k
z
z
k
s
k
y
k
x
k
xs
k
xy
k
xx x
k
k
z Q Q Q N
1
1
d ,
,
_
'
'
'
N
k
z
z
xy
y
x
s
y
x
k
xs
k
xy
k
xx x
k
k
z z Q Q Q N
1
0
0
0
0
0
0
1
d ,
,
_
'
'
,
_
'
'

N
k
z
z
xy
y
x
k
xs
k
xy
k
xx
N
k
z
z
s
y
x
k
xs
k
xy
k
xx x
k
k
k
k
z z Q Q Q z Q Q Q N
1
0
0
0
1
0
0
0
1 1
d d . 66
Now define the in-plane laminate stiffnesses, s y x j i A
ij
& , , , , as
( )

,
_
N
k
k k
k
xx
N
k
z
z
k
xx xx
z z Q z Q A
k
k
1
1
1
1
d
( )

,
_
N
k
k k
k
xy
N
k
z
z
k
xy xy
z z Q z Q A
k
k
1
1
1
1
d
( )

,
_
N
k
k k
k
xs
N
k
z
z
k
xs xs
z z Q z Q A
k
k
1
1
1
1
d
or in general
( )

N
k
k k
k
ij ij
z z Q A
1
1
67
and the coupling stiffnesses, s y x j i B
ij
& , , , , as
( )

,
_
N
k
k k
k
xx
N
k
z
z
k
xx xx
z z Q z Q B
k
k
1
2
1
2
1
2
1
zd
1

( )

,
_
N
k
k k
k
xy
N
k
z
z
k
xy xy
z z Q z Q B
k
k
1
2
1
2
1
2
1
zd
1

( )

,
_
N
k
k k
k
xs
N
k
z
z
k
xs xs
z z Q z Q B
k
k
1
2
1
2
1
2
1
zd
1

or in general
( )

N
k
k k
k
ij ij
z z Q B
1
2
1
2
2
1
. 68
Then N
x
becomes:
{ } { }
'
'
0
0
0
0
0
0
xy
y
x
xs xy xx
s
y
x
xs xy xx x
B B B A A A N . 69
By a similar process the relationship between the other in-plane forces and
the strains and curvatures are derived to be:
'
1
1
1
]
1
'
1
1
1
]
1
'
0
0
0
0
0
0
xy
y
x
ss ys xs
ys yy xy
xs xy xx
s
y
x
ss ys xs
ys yy xy
xs xy xx
xy
y
x
B B B
B B B
B B B
A A A
A A A
A A A
N
N
N
, 70
where
ij
A and
ij
B are as defined previously.
The relationship between the moments and the strains and curvatures is
derived by applying the same considerations of the laminate behavior used to
derive the force-strain curvature relationship to equation 35. This gives:
( ) ( ) ( ) ( )

+ + +
N
N
z
z
N
x
z
z
x
z
z
x
h
h
x xy
z z y x z z y x z z y x z z y x M
1
2
1
1
0
zd , , zd , , zd , , zd , ,
2 1
2 /
/2 -
L
( )

N
k
z
z
k
x xy
k
k
z z y x M
1
1
zd , , . 71
Now, substituting equations 59 and 64 into 71 yields:
'
'
N
k
z
z
k
s
k
y
k
x
k
xs
k
xy
k
xx xy
k
k
z Q Q Q M
1
1
zd ,
,
_
'
'
'
N
k
z
z
xy
y
x
s
y
x
k
xs
k
xy
k
xx xy
k
k
z z z Q Q Q M
1
0
0
0
2
0
0
0
1
d ,
,
_
'
'
,
_
'
'

N
k
z
z
xy
y
x
k
xs
k
xy
k
xx
N
k
z
z
s
y
x
k
xs
k
xy
k
xx xy
k
k
k
k
z z Q Q Q z Q Q Q M
1
2
0
0
0
1
0
0
0
1 1
d zd . 72
Note in the above expression the presence of the coupling stiffnesses,
s y x j i B
ij
& , , , , . Now define the out-of-plane laminate stiffnesses,
s y x j i D
ij
& , , , , as
( )

,
_
N
k
k k
k
xx
N
k
z
z
k
xx xx
z z Q z Q D
k
k
1
3
1
3
1
2
3
1
d z
1

( )

,
_
N
k
k k
k
xy
N
k
z
z
k
xy xy
z z Q z Q D
k
k
1
3
1
3
1
2
3
1
d z
1

( )

,
_
N
k
k k
k
xs
N
k
z
z
k
xs xs
z z Q z Q D
k
k
1
3
1
3
1
2
3
1
d z
1

or in general
( )

N
k
k k
k
ij ij
z z Q D
1
3
1
3
3
1
. 73
Then M
xy
becomes:
{ } { }
'
'
0
0
0
0
0
0
xy
y
x
xs xy xx
s
y
x
xs xy xx xy
D D D B B B M . 74
By a similar process the relationship between the moments and the strains
and curvatures are derived to be:
'
1
1
1
]
1
'
1
1
1
]
1
'
0
0
0
0
0
0
xy
y
x
ss ys xs
ys yy xy
xs xy xx
s
y
x
ss ys xs
ys yy xy
xs xy xx
yy
yx
xy
D D D
D D D
D D D
B B B
B B B
B B B
M
M
M
, 75
where
ij
B and
ij
D are as defined previously. A constitutive relation between the
forces and moments and the strains and curvatures is written by combining
equations 70 and 75. The constitutive relationship for a laminate is:
'
1
1
1
1
1
1
1
]
1
'
0
0
0
0
0
0
xy
y
x
s
y
x
ss ys xs
ys yy xy
xs xy xx
ss ys xs
ys yy xy
xs xy xx
ss ys xs
ys yy xy
xs xy xx
ss ys xs
ys yy xy
xs xy xx
yy
yx
xy
xy
y
x
D D D
D D D
D D D
B B B
B B B
B B B
B B B
B B B
B B B
A A A
A A A
A A A
M
M
M
N
N
N
, 76
or
[ ] [ ]
[ ] [ ]

;
'
1
]
1
'
0
0
D B
B A
M
N
. 77
Note that the A, B, D, matrix, the laminate stiffness, is a partitioned matrix that is
also symmetric.
THE LAMI NATED PLATE EQUATI ONS
The three force balance equations, equations 40, 41, and 51, are the
foundation for the governing system of equations for the displacement of the plate.
THE IN-PLANE EQUATIONS
The governing in-plane force balance equations for the laminated plate in
terms of the deflections are derived by substituting the constitutive equation into
the in-plane force balance equations; equation 76 and equations 40 and 41. Using
equation 76 or 69 to write
x
N
x
, and assuming that the laminate is homogeneous in

the x- and y-directions gives:
{ } { }
'
'
x
x
x
B B B
x
x
x
A A A
x
N
xy
y
x
xs xy xx
s
y
x
xs xy xx
x
0
0
0
0
0
0
.
Now substituting the definitions of the strains and curvatures of the mid-plane into
the equation above yields:
{ } { }
'
'
y x
w
y x
w
x
w
B B B
x
v
y x
u
y x
v
x
u
A A A
x
N
xs xy xx xs xy xx
x
2
0
3
2
0
3
3
0
3
2
0
2
0
2
0
2
2
0
2
2
. 78
Similarly, using equation 76 to write
x
N
xy
, and again assuming that the laminate

is homogeneous in the x- and y-directions gives:
{ } { }
'
'
y
y
y
B B B
y
y
y
A A A
y
N
xy
y
x
ss ys xs
s
y
x
ss ys xs
xy
0
0
0
0
0
0
.
Then substituting the definitions of the strains and curvatures of the mid-plane
into the equation above yields:
{ } { }
'
'
2
0
3
3
0
3
2
0
3
0
2
2
0
2
2
0
2
0
2
2
y x
w
y
w
y x
w
B B B
y x
v
y
u
y
v
y x
u
A A A
y
N
ss ys xs ss ys xs
xy
. 79
Substituting equations 78 and 79 into equation 40 gives:
( )
( ) 0 2 3
2
3
0
3
2
0
3
2
0
3
3
0
3
2
0
2
0
2
2
0
2
2
0
2
0
2
2
0
2
+ +
y
w
B
y x
w
B B
y x
w
B
x
w
B
y
v
A
y x
v
A A
x
v
A
y
u
A
y x
u
A
x
u
A
ys ss xy xs xx
ys ss xy xs ss xs xx
. 80
Similarly for the y-direction sum of forces
y
N
y
and
x
N
xy
are:
{ } { }
'
'
2
0
3
3
0
3
2
0
3
0
2
2
0
2
2
0
2
0
2
2
y x
w
y
w
y x
w
B B B
y x
v
y
u
y
v
y x
u
A A A
y
N
ys yy xy ys yy xy
y
. 81
and
{ } { }
'
'
y x
w
y x
w
x
w
B B B
x
v
y x
u
y x
v
x
u
A A A
x
N
ss ys xs ss ys xs
xy
2
0
3
2
0
3
3
0
3
2
0
2
0
2
0
2
2
0
2
2
. 82
Substituting equations 81 and 82 into equation 41 gives:
( )
( ) 0 3 2
2
3
0
3
2
0
3
2
0
3
3
0
3
2
0
2
0
2
2
0
2
2
0
2
0
2
2
0
2
+ +
y
w
B
y x
w
B
y x
w
B B
x
w
B
y
v
A
y x
v
A
x
v
A
y
u
A
y x
u
A A
x
u
A
yy ys ss xy xs
yy ys ss ys ss xy xs
. 83
Equations 80 and 83 are the in-plane force balance equations expressed in terms of
the displacements. They represent two simultaneous coupled partial differential
equations in
0
u ,
0
v , and
0
w . Note that if the coupling stiffnesses,
ij
B , are zero, then
the out-of-plane displacement,
0
w , is decoupled from the in-plane force balance
equations.
Note that the buckling loads,
x
N ,
y
N , and
xy
N , are constant throughout the
plate. Consequently, they do not appear in the in-plane force balance equations.
THE OUT-OF-PLANE EQUATION
The out-of-plane force balance equation for the laminated plate in terms of
the deflections is derived by substituting the constitutive equation into the out-of-
plane force balance equation; equation 76 and equation 51. Using equation 76 to
write
2
2
x
M
xy
,
2
2
y
M
yx
, and
y x
M
yy

2
, as was done previously for the forces, leads to:
{ } { }
'
'
y x
w
y x
w
x
w
D D D
x
v
y x
u
y x
v
x
u
B B B
x
M
xs xy xx xs xy xx
xy
3
0
4
2 2
0
4
4
0
4
3
0
3
2
0
3
2
0
3
3
0
3
2
2
2
, 84
{ } { }
'
'
3
0
4
4
0
4
2 2
0
4
2
0
3
3
0
3
3
0
3
2
0
3
2
2
2
y x
w
y
w
y x
w
D D D
y x
v
y
u
y
v
y x
u
B B B
y
M
ys yy xy ys yy xy
yx
, 85
and
{ } { }
'
'
2 2
0
4
3
0
4
3
0
4
2
0
3
2
0
3
2
0
3
2
0
3
2
2
y x
w
y x
w
y x
w
D D D
y x
v
y x
u
y x
v
y x
u
B B B
y x
M
ss ys xs ss ys xs
yy
. 86
Substitution of the results in equations 84, 85, and 86 into equation 51 yields the
full out-of-plane plate displacement equation with dynamic and buckling effects:
( )
( )
( )
( ) t y x p
t
w
h
y
w
N
y x
w
N
x
w
N
y
w
D
y x
w
D
y x
w
D D
y x
w
D
x
w
D
y
v
B
y x
v
B
y x
v
B B
x
v
B
y
u
B
y x
u
B B
y x
u
B
x
u
B
y xy x
yy ys ss xy xs xx
yy ys xy ss xs
ys xy ss xs xx
, , 2
4 4 2 4
3 2
2 3
2
0
2
2
0
2
0
2
2
0
2
4
0
4
3
0
4
2 2
0
4
3
0
4
4
0
4
3
0
3
2
0
3
2
0
3
3
0
3
3
0
3
2
0
3
2
0
3
3
0
3
+ +
+ +

. 87
Equations 80, 83, and 87 form the complete system of coupled simultaneous partial
differential equations that describe the displacement of the mid-plane of the
laminate with dynamic and buckling loads included.
SIMPLIFICATIONS OF THE PLATE EQUATION
As noted earlier, there are four loading cases that can be analyzed with
equations 80, 83, and 87. They are:
Complete -- Buckling with dynamic effects,
Dynamic -- Neglect buckling effects,
Buckling -- Neglect dynamic effects,
Static -- Neglect buckling and dynamic effects.
These assumptions do not directly impact equations 80 and 83. The complete
system of equations is equations 80, 83, and 87.
Dynamic Plate Equation
Applying the assumption of no buckling loads to equation 87 yields:
( )
( )
( )
( ) t y x p
t
w
h
y
w
D
y x
w
D
y x
w
D D
y x
w
D
x
w
D
y
v
B
y x
v
B
y x
v
B B
x
v
B
y
u
B
y x
u
B B
y x
u
B
x
u
B
yy ys ss xy xs xx
yy ys xy ss xs
ys xy ss xs xx
, ,
4 4 2 4
3 2
2 3
2
0
2
4
0
4
3
0
4
2 2
0
4
3
0
4
4
0
4
3
0
3
2
0
3
2
0
3
3
0
3
3
0
3
2
0
3
2
0
3
3
0
3
+ +
+ +

. 88
Buckling Plate Equation
Applying the assumption of no dynamic effects to equation 87 yields:
( )
( )
( )
( ) y x p
y
w
N
y x
w
N
x
w
N
y
w
D
y x
w
D
y x
w
D D
y x
w
D
x
w
D
y
v
B
y x
v
B
y x
v
B B
x
v
B
y
u
B
y x
u
B B
y x
u
B
x
u
B
y xy x
yy ys ss xy xs xx
yy ys xy ss xs
ys xy ss xs xx
, 2
4 4 2 4
3 2
2 3
2
0
2
0
2
2
0
2
4
0
4
3
0
4
2 2
0
4
3
0
4
4
0
4
3
0
3
2
0
3
2
0
3
3
0
3
3
0
3
2
0
3
2
0
3
3
0
3

+ +
+ +

. 89
Static Plate Equation
Applying the assumption of no buckling and no dynamic effects to equation
87 yields:
( )
( )
( ) ( ) y x p
y
w
D
y x
w
D
y x
w
D D
y x
w
D
x
w
D
y
v
B
y x
v
B
y x
v
B B
x
v
B
y
u
B
y x
u
B B
y x
u
B
x
u
B
yy ys ss xy xs xx
yy ys xy ss xs
ys xy ss xs xx
, 4 4 2 4
3 2
2 3
4
0
4
3
0
4
2 2
0
4
3
0
4
4
0
4
3
0
3
2
0
3
2
0
3
3
0
3
3
0
3
2
0
3
2
0
3
3
0
3

+ +
+ +

. 90

LAMINATES
I NTRODUCTI ON
I n the previous section the laminate stiffnesses
ij
A ,
ij
B , and
ij
D and the
governing system of equations for a laminated plate were derived. I n this section,
alternate expressions for the laminate stiffnesses are presented, the impact of
various simplifying assumptions, the distribution of stress through a laminate, and
laminate failure theories are discussed.
ALTERNATE EXPRESSI ONS FOR LAMI NATE STI FFNESSES
Figure 10 shows a schematic drawing of a laminate cross-section. I n this
figure, the laminate is shown as constructed of N laminae of thicknesses
k
t . The
laminae thicknesses are directly related to the laminae coordinates. This
relationship is:
1

k k k
z z t . 91
Recognizing this relationship, the in-plane extensional stiffnesses,
ij
A , can be
written in terms of the individual laminae thicknesses:
N
k
k
k
ij ij
t Q A
1
. 92
This expression is an alternative to that presented previously and is, at times, more
convenient to use.
The lamina centroid,
k
z , is defined as:
2
1
+
k k
k
z z
z . 93
Now using the laminae centroids and thicknesses the coupling stiffnesses,
ij
B , can
be written as:
( ) ( )( ) ( )
( )
2 2
1
2
1
1
1
1 1
1
1
1
2
1
2

+
+

k k
N
k
k k
k
ij k k
N
k
k k
k
ij
N
k
k k
k
ij ij
z z
z z Q z z z z Q z z Q B ,
or
LAMI NATES
k
N
k
k
k
ij ij
z t Q B

1
. 94
I t is arguable whether the alternate expression is simpler than the original form of
the coupling stiffnesses. The alternate form does demonstrate the relationship
between the coupling stiffnesses and the individual laminae thicknesses and their
relative location within the laminate.
The bending stiffnesses,
ij
D , can also be expressed in terms of the laminae centroids
and thicknesses. This alternate expression is:
,
_
+
N
k
k k k
k
ij ij
t z t Q D
1
3 2
12
1
. 95
Again, the relative simplicity of the alternate expression to that of the original form
of the bending stiffnesses is arguable. Nevertheless, this expression is important
because of its instructive nature. Recall from classical isotropic plate theory that
the plate stiffness, D, is
( )
2
3
1 12
Et
, where E is the modulus of elasticity, is the
Poissons ratio, and t is the plate thickness. Further recall that the plate stiffness is
analogous to the bending stiffness of a beam,
yy
EI , where
yy
I is the second area
moment of inertia about the y-axis. Next, note that the second area moment of
inertia of a rectangular beam, with height t and width b, about the centroid is
12
3
bt
.
Therefore, the beam bending stiffness is
12
3
Ebt
. Comparing the beam bending
stiffness and the isotropic plate bending stiffness, the plate bending stiffness
appears to be a beam bending stiffness per unit width and including Poissons
effects. Now remember the Parallel Axis Theorem for the second area moment of
inertia is A d I I
OC
z
C
yy
O
yy
2
+ , where point C is the centroid of the cross-section, point O
is the point where the moment is being calculated,
2
OC
z
d is the distance along the
z-axis between point O and point C, and A is the area of the cross-section. Now
examine equation 95, we observe the bending stiffness of each lamina,
12
3
k
k
ij
t Q
, and
the shifting factor of each lamina,
2
k k
k
ij
z t Q . Therefore, we conclude that the laminate
stiffness is the summation of the bending stiffnesses of the individual lamina about
the mid-plane.
LAMI NATES
SI MPLI FYI NG ASSUMPTI ONS ON LAMI NATE STRUCTURE
There are several simplifying assumptions that can be made that have a
significant impact on the force-moment strain-curvature relationship and governing
system of equations. These assumptions deal directly with the structure of the
laminate, in particular the lay-up or stacking sequence. I n this discussion we will
consider only laminated structures, i. e., N > 1, of orthotropic materials.
SYMMETRIC LAMINATES
Recall that a symmetric laminate is defined as a laminate that is symmetric
in both geometry and material properties about the mid-plane. The stiffness
equations, equations 92, 94, and 95, are expressed in terms of the number of
laminae in the laminate and the laminae reduced stiffnesses, thicknesses and
centroids. Symmetry of geometry means:
2
1
N
k t t
k
k N k

96
and
2
1
N
k z z
k
k N k

. 97
Note that equation 91 does not imply that the laminae thicknesses are constant.
Symmetry of material properties means:
2
1 , & , , ,
N
k s y x j i Q Q
k
k N
ij
k
ij

. 98
Applying the symmetry specialization to the in-plane stiffnesses,
ij
A , from
equation 92 leads to:
'
Odd
Even
N t Q
N t Q
A
N
k
k
k
ij
N
k
k
k
ij
ij
2
1
1
2
1
2
2
. 99
Applying the symmetry specialization to the coupling stiffnesses,
ij
B , from
LAMI NATES
'
+
+
Odd
Even
N z t Q z t Q
N z t Q z t Q
B
N
N
k
k k
k
ij
N
k
k k
k
ij
N
k
N
N
k
k k
k
ij k k
k
ij
ij
2
1
2
1
1
2
1
2

'
+
+
Odd
Even
N z t Q z t Q
N z t Q z t Q
B
N
k
k N k N
k N
ij
N
k
k k
k
ij
N
k
N
k
k N k N
k N
ij k k
k
ij
ij
2
1
1
2
1
1
2
1
2
1

( )
( )
'
N
N
k
k k k
k
ij
N
k
k k k
k
ij
ij
N z z t Q
N z z t Q
B 0
2
1
1
2
1
Odd
Even
. 100
The impact of the results shown in equation 100 is significant. When the structure
is a symmetric laminated plate, the in-plane problem is decoupled from the out-of-
plane problem. Therefore, equations 87, 88, 89, and 90 are independent of the in-
plane displacement and the system of equations reduces to a single partial
differential equation for the out-of-plane displacement. This simplifies the analysis
significantly.
Applying the symmetry specialization to the bending stiffnesses,
ij
D , from
'
,
_
+

,
_
Odd
Even
N t z t Q
N t z t Q
D
N
k
k k k
k
ij
N
k
k k k
k
ij
ij
2
1
1
3 2
2
1
3 2
12
1
2
12
1
2
. 101
Therefore, the laminate stiffness matrix for a symmetric laminate is:
LAMI NATES
'
1
1
1
1
1
1
1
]
1
'
0
0
0
0
0
0
0 0 0
0 0 0
0 0 0
0 0 0
0 0 0
0 0 0
xy
y
x
s
y
x
ss ys xs
ys yy xy
xs xy xx
ss ys xs
ys yy xy
xs xy xx
yy
yx
xy
xy
y
x
D D D
D D D
D D D
A A A
A A A
A A A
M
M
M
N
N
N
. 102
This makes the previous statement regarding the decoupling of the in-plane
problem and the out-of-plane even more clear. Recognize that plates of isotropic
materials, laminated or homogeneous, are necessarily symmetric. Therefore, for all
plates of isotropic materials (the traditional engineering materials of the last
century) the in-plane problem is decoupled from the out-of-plane problem. This is
one of the reasons that the majority of laminated composite structures are
symmetric laminates. Note that this does not necessarily take full advantage of the
structural capabilities of laminated composite materials.
CONSIDERATION OF THE MATERIAL PROPERTIES
Note that in the application of the symmetry specialization no further
specialization regarding the material properties have been made. Therefore,
equations 99, 100, and 101 are general for all materials. Further specializations for
the laminate stiffnesses can be made by applying restrictions to the material
properties. These restrictions or specializations are to angle ply and cross ply
laminates. From review of lamina mechanics, the off-axis stiffness properties were
expressed in terms of the invariants of the lamina properties. Now applying this to
the laminate properties, the lamina stiffness invariants,
Q
k
l
U , are:
1
]
1
+ +
1
]
1
+ +
1
]
1
+
1
]
1

1
]
1
+ + +
k k k k
Q
k
k k k k
Q
k
k k k k
Q
k
k k
Q
k
k k k k
Q
k
Q Q Q Q U
Q Q Q Q U
Q Q Q Q U
Q Q U
Q Q Q Q U
66 12 22 11 5
66 12 22 11 4
66 12 22 11 3
22 11 2
66 12 22 11 1
4 2
2
1
4 6
8
1
4 2
8
1
2
1
4 2 3 3
8
1
103
and the off-axis stiffness properties are related to the invariant properties by:
LAMI NATES
'
1
1
1
1
1
1
1
1
1
1
]
1
'
Q
k
Q
k
k k
k k
k
Q
k
k
Q
k
k k
Q
k
k k
Q
k
k
ys
k
xs
k
ss
k
xy
k
yy
k
xx
U
U
U
U
U
U
Q
Q
Q
Q
Q
Q
3
2
2
1
2
1
5
4
1
1
1
4 sin - 2 sin 0
4 sin 2 sin 0
4 cos - 0
4 cos - 0
4 cos 2 cos -
4 cos 2 cos
. 104
Using equation 104 the laminate stiffnesses can be written as:
'
,
_
'
1
1
1
1
1
1
1
1
1
1
]
1
'
N
k
Q
k
Q
k
k k
k k
k
Q
k
k
Q
k
k k
Q
k
k k
Q
k
k
ys
xs
ss
xy
yy
xx
U
U
U
U
U
U
t
A
A
A
A
A
A
1
3
2
2
1
2
1
5
4
1
1
1
4 sin - 2 sin 0
4 sin 2 sin 0
4 cos - 0
4 cos - 0
4 cos 2 cos -
4 cos 2 cos
, 105
'
,
_
'
1
1
1
1
1
1
1
1
1
1
]
1
'
N
k
Q
k
Q
k
k k
k k
k
Q
k
k
Q
k
k k
Q
k
k k
Q
k
k k
ys
xs
ss
xy
yy
xx
U
U
U
U
U
U
z t
B
B
B
B
B
B
1
3
2
2
1
2
1
5
4
1
1
1
4 sin - 2 sin 0
4 sin 2 sin 0
4 cos - 0
4 cos - 0
4 cos 2 cos -
4 cos 2 cos
, 106
and
'
,
_
'
1
1
1
1
1
1
1
1
1
1
]
1
,
_
'
N
k
Q
k
Q
k
k k
k k
k
Q
k
k
Q
k
k k
Q
k
k k
Q
k
k k k
ys
xs
ss
xy
yy
xx
U
U
U
U
U
U
t z t
D
D
D
D
D
D
1
3
2
2
1
2
1
5
4
1
1
3 2
1
4 sin - 2 sin 0
4 sin 2 sin 0
4 cos - 0
4 cos - 0
4 cos 2 cos -
4 cos 2 cos
12
1
. 107
LAMI NATES
Note that symmetry of material properties means:
2
1
N
k
k
k N k

. 108
Symmetric Cross-ply Laminates
Note in equation 104, that if the lamina angle,
k
, is 0 or an integer multiple of 90
that coupling stiffnesses between extension and shear,
k
xs
Q and
k
ys
Q , are zero. Recall
that a cross-ply laminate is defined as one in which all lamina are oriented such
that their orientation angles are either 0 or 90. Therefore, all coupling stiffnesses
between extension and shear,
k
xs
Q and
k
ys
Q , are zero. Further, 1 4 cos
k
,
k
and
1 2 cos t
k
. Now define an alternating function, , such that:
( )
'

90 1
0 1
. 109
Thus, for a symmetric cross-ply laminate the stiffnesses are:
( )
( )
'
,
_
,
_
,
_
,
_
+ +
'
0
0
1
3 5
1
3 4
1
2 3 1
1
2 3 1
Mid
Mid
Mid
Mid
N
k
Q
k
Q
k
k
N
k
Q
k
Q
k
k
N
k
k
Q
k
Q
k
Q
k
k
N
k
k
Q
k
Q
k
Q
k
k
ys
xs
ss
xy
yy
xx
U U t
U U t
U U U t
U U U t
A
A
A
A
A
A
, 110
{ } 0
'
ys
xs
ss
xy
yy
xx
B
B
B
B
B
B
, 111
as before, and
LAMI NATES
( )
( )
'
,
_
,
_
,
_
,
_
,
_
,
_
,
_
+ +
,
_
'
0
0
12
1
12
1
12
1
12
1
1
3 5
3 2
1
3 4
3 2
1
2 3 1
3 2
1
2 3 1
3 2
Mid
Mid
Mid
Mid
N
k
Q
k
Q
k
k k k
N
k
Q
k
Q
k
k k k
N
k
k
Q
k
Q
k
Q
k
k k k
N
k
k
Q
k
Q
k
Q
k
k k k
ys
xs
ss
xy
yy
xx
U U t z t
U U t z t
U U U t z t
U U U t z t
D
D
D
D
D
D
, 112
where
'
Odd
Even
N
N
N
N
N
Mid
2
1
2

Therefore, the force-moment strain-curvature relationship for a symmetric cross-ply
laminate is:
'
1
1
1
1
1
1
1
]
1
'
0
0
0
0
0
0
0 0
0
0
0 0 0
0 0 0
0 0 0
0 0 0
0 0 0
0 0 0
0 0
0
0
xy
y
x
s
y
x
ss
yy xy
xy xx
ss
yy xy
xy xx
yy
yx
xy
xy
y
x
D
D D
D D
A
A A
A A
M
M
M
N
N
N
. 113
Note that the response is further decoupled, with the normal and shear responses
mutually decoupled.
Balanced Regular Symmetric Angle-ply Laminates
For angle-ply laminates
k
, is restricted to t alternating through the laminate.
Letting be either a positive or negative angle, then
1
,
2
,
3
,

4
, et cetera, or in general ( )
+1
1
k
k
. Because the presence of multiple
angle trigonometric functions in the invariant material property expression it is
necessary to consider the off-axis material properties as expressed in equation 21,
which is now written for an individual lamina of a laminate:
LAMI NATES
( )
( ) ( )
( ) ( )

'
1
1
1
1
1
1
1
1
]
1
'
k
k
k
k
k k k k k k k k k k k k
k k k k k k k k k k k k
k k k k k k k k
k k k k k k k k
k k k k k k
k k k k k k
k
ys
k
xs
k
ss
k
xy
k
yy
k
xx
Q
Q
Q
Q
n m n m n m n m n m n m
n m n m n m n m n m n m
n m n m n m n m
n m n m n m n m
n m n m m n
n m n m n m
Q
Q
Q
Q
Q
Q
66
12
22
11
2 2 2 2 3 3
2 2 2 2 3 3
2
2 2 2 2 2 2 2 2
2 2 4 4 2 2 2 2
2 2 2 2 4 4
2 2 2 2 4 4
2
2
2
4
4 2
4 2
, 114
note in this expression that
k k
m cos and
k k
n sin . Since ( )
+1
1
k
k
, then
( ) ( ) m m
k
k

+
cos 1 cos
1
and ( ) ( ) ( ) ( ) n n
k k k
k
1 1 1
1 sin 1 1 sin
+ + +
. Then
( ) ( )
2
2
1 2
1 n n n
k
k

+
, and similarly ( ) ( )
4
4
1 4
1 n n n
k
k

+
. Therefore, equation 114
becomes
( )
( ) ( ) ( ) ( ) ( ) ( )
( ) ( ) ( ) ( ) ( ) ( )

'
1
1
1
1
1
1
1
1
]
1
'
+ + +
+
k
k
k
k
k k k k
k k k k
k
ys
k
xs
k
ss
k
xy
k
yy
k
xx
Q
Q
Q
Q
n m n m n m n m
n m n m n m n m
n m n m m n
n m n m n m
Q
Q
Q
Q
Q
Q
66
12
22
11
2 2 1 2 2 1 3 3 1
2 2 2 2 3 3 1
2
2 2 2 2 2 2 2 2
2 2 4 4 2 2 2 2
2 2 2 2 4 4
2 2 2 2 4 4
1 2 1 1 1
1 2 1 1 1
2
4
4 2
4 2
. 115
Note in the above expression that the transformation coefficients of the on-axis
lamina stiffnesses for
k
xx
Q ,
k
yy
Q ,
k
xy
Q , and
k
ss
Q are independent of the lamina number,
k. Therefore, in the calculation of the laminate stiffnesses these terms will be
distributed over the summations.
Recall that the in-plane stiffnesses for the laminate are:
LAMI NATES
'
,
_
'
'
N
k
k
ys
k
xs
k
ss
k
xy
k
yy
k
xx
k
ys
xs
ss
xy
yy
xx
Q
Q
Q
Q
Q
Q
t
A
A
A
A
A
A
1
. 116
Then substituting equation 115 gives:
( )
( ) ( ) ( ) ( ) ( ) ( )
( ) ( ) ( ) ( ) ( ) ( )

'
,
_
'
1
1
1
1
1
1
1
1
]
1
'
+ + +
+
N
k
k
k
k
k
k k k k
k k k k
k
ys
xs
ss
xy
yy
xx
Q
Q
Q
Q
n m n m n m n m
n m n m n m n m
n m n m m n
n m n m n m
t
A
A
A
A
A
A
1
66
12
22
11
2 2 1 2 2 1 3 3 1
2 2 2 2 3 3 1
2
2 2 2 2 2 2 2 2
2 2 4 4 2 2 2 2
2 2 2 2 4 4
2 2 2 2 4 4
1 2 1 1 1
1 2 1 1 1
2
4
4 2
4 2
. 117
Now recognizing that mand n are independent of lamina number, then the in-plane
laminate stiffnesses
xx
A ,
yy
A ,
xy
A , and
ss
A can be written:
( )
'
,
_
'
1
1
1
1
1
]
1
'
N
k
k
k
k
k
k
ss
xy
yy
xx
Q
Q
Q
Q
t
n m n m n m n m
n m n m n m n m
n m n m m n
n m n m n m
A
A
A
A
1
66
12
22
11
2
2 2 2 2 2 2 2 2
2 2 4 4 2 2 2 2
2 2 2 2 4 4
2 2 2 2 4 4
2
4
4 2
4 2
. 118
Now make two further specializations. First, that the laminate is regular,
this requires that all lamina have the same thickness, t. Second, that the laminate
is homogeneous, this requires that the same material is used throughout the
laminate. The consequence of the second specialization is that the on-axis material
properties are independent of lamina number. Applying these specializations to
equation 118 gives:
( )

'
1
1
1
1
1
]
1
'
66
12
22
11
2
2 2 2 2 2 2 2 2
2 2 4 4 2 2 2 2
2 2 2 2 4 4
2 2 2 2 4 4
2
4
4 2
4 2
Q
Q
Q
Q
n m n m n m n m
n m n m n m n m
n m n m m n
n m n m n m
Nt
A
A
A
A
ss
xy
yy
xx
. 119
LAMI NATES
Now consider specifically the
xs
A term:
( ) ( ) ( ) ( ) ( ) ( )
,
_
,
_
+ + +
N
k
k
k
k
k
k
k
k
k
k xs
Q n m mn Q n m mn Q mn nQ m t A
1
66
2 2
12
2 2
22
3
11
3 1
1 2 1 1 1
Again recognizing that m and n are independent of lamina number, then
xs
A
becomes:
( ) ( )
( ) ( ) ( ) ( )

+
,
_
,
_
,
_
,
_

N
k
k
k
k
N
k
k
k
k
N
k
k
k
k
N
k
k
k
k xs
Q t n m mn Q t n m mn
Q t mn Q t n m A
1
66
2 2
1
12
2 2
1
22
3
1
11
1 3
1 2 1
1 1
.
Applying the specializations of a regular balanced lay-up to
xs
A gives:
( ) ( )
( ) ( ) ( ) ( )

+
+ +
+
N
k
k
N
k
k
N
k
k
N
k
k
xs
Q t n m mn Q t n m mn
Q t mn Q t n m A
1
66
2 2
1
12
2 2
1
22
3
1
1
11
3
1 2 1
1 1
.
Note that the summations in the above are summations of integers (1) with
alternating sign. Making a third specialization, that the laminate is balanced, thus
the number of laminae in the laminate is even, Even N . Thus,
( ) ( ) 0 1 1
1 1
1

+
N
k
k
N
k
k

and therefore
0
xs
A . 120
By a similar process, for a homogeneous regular balanced symmetric angle-ply
laminate the in-plane laminate stiffness
ys
A can be shown to be:
0
xs
A . 121
Summarizing, the laminate in-plane stiffnesses for a homogeneous regular balance
symmetric angle-ply laminate are:
LAMI NATES
( )
'
1
1
1
1
1
1
1
1
]
1
'
66
12
22
11
2
2 2 2 2 2 2 2 2
2 2 4 4 2 2 2 2
2 2 2 2 4 4
2 2 2 2 4 4
0 0 0 0
0 0 0 0
2
4
4 2
4 2
Q
Q
Q
Q
n m n m n m n m
n m n m n m n m
n m n m m n
n m n m n m
Nt
A
A
A
A
A
A
ys
xs
ss
xy
yy
xx
. 122
At this point, the use of material property invariants may be applied to
equation 122, which gives:
'
1
1
1
1
1
1
1
1
]
1
'
Q
Q
Q
Q
Q
Q
ys
xs
ss
xy
yy
xx
U
U
U
U
U
U
Nt
A
A
A
A
A
A
3
2
5
4
1
1
1
0 0 0
0 0 0
4 cos - 0
4 cos - 0
4 cos 2 cos -
4 cos 2 cos
. 123
The laminate coupling stiffnesses are also zero because the laminate in
consideration is symmetric. Therefore,
'
'
0
0
0
0
0
0
ys
xs
ss
xy
yy
xx
B
B
B
B
B
B
. 124
The development of the laminate bending stiffnesses for a homogeneous
regular balance symmetric angle-ply laminate parallels the development of the in-
plane stiffnesses. For the laminate bending stiffnesses
xx
D ,
yy
D ,
xy
D , and
ss
D , the
lamina stiffnesses and mand n are independent of lamina number, then can be
written as:
( )
( )

'
,
_
'
1
1
1
1
1
]
1
,
_
'
66
12
22
11
2
2 2 2 2 2 2 2 2
2 2 4 4 2 2 2 2
2 2 2 2 4 4
2 2 2 2 4 4
2
1
2 3
2
4
4 2
4 2
2
12
Q
Q
Q
Q
n m n m n m n m
n m n m n m n m
n m n m m n
n m n m n m
z t t
N
D
D
D
D
N
k
k
ss
xy
yy
xx
. 125
Alternately, these stiffnesses can be written in terms of the invariant properties of
the material:
LAMI NATES
( )
'
'
1
1
1
1
1
]
1
,
_
'
Q
Q
Q
Q
Q
Q
N
k
k
ss
xy
yy
xx
U
U
U
U
U
U
z t t
N
D
D
D
D
3
2
5
4
1
1
2
1
2 3
1
4 cos - 0
4 cos - 0
4 cos 2 cos -
4 cos 2 cos
2
12
. 126
Now, as with the in-plane stiffness development, consider specifically the
xs
D
term. However, in this case it is easier to consider the initial form of the bending
stiffness, which is:
( ) ( ) ( ) ( ) ( ) ( ) ( )
,
_
,
_
+ + +
N
k
k
k
k
k
k
k
k
k
k k xs
Q n m mn Q n m mn Q mn nQ m z z D
1
66
2 2
12
2 2
22
3
11
3 1 3
1
3
1 2 1 1 1
3
1

Again recognizing that m and n are independent of lamina number, then
xs
D
becomes:
( )( ) ( )( )
( )
( )( )
( )
( )( )

,
_
,
_
,
_
,
_

N
k
k
k
k k
N
k
k
k
k k
N
k
k
k
k k
N
k
k
k
k k xs
Q z z
n m mn
Q z z
n m mn
Q z z
mn
Q z z
n m
D
1
66
3
1
3
2 2
1
12
3
1
3
2 2
1
22
3
1
3
3
1
11
1 3
1
3
3
1
3
2
1
3
1
3
1
3
.
Applying the specializations of a regular balanced lay-up to
xs
D gives:
( )( ) ( ) ( )( ) ( )
( )
( )( ) ( )
( )
( )( ) ( )

+
+
N
k
k
k k
N
k
k
k k
N
k
k
k k
N
k
k
k k xs
Q z z
n m mn
Q z z
n m mn
Q z z
mn
Q z z
n m
D
1
66
3
1
3
2 2
1
12
3
1
3
2 2
1
22
3
1
3
3
1
11
1 3
1
3
3
1
3
2
1
3
1
3
1
3
,
or
( )( ) ( )( ) ( )( ) ( )( )
( )
( )( ) ( )( )
( )
( )( ) ( )( )
,
_
,
_
,
_
+
,
_
+
N
k
k
k
N
k
k
k
N
k
k
k
N
k
k
k
N
k
k
k
N
k
k
k
N
k
k
k
N
k
k
k xs
z z
Q n m mn
z z
Q n m mn
z z
Q mn
z z
nQ m
D
1
3
1
1
3
66
2 2
1
3
1
1
3
12
2 2
1
3
1
1
3
22
3
1
1 3
1
1
1 3
11
3
1 1
3
2
1 1
3
1 1
3
1 1
3
.
Now examine the summations in the first bracket above,
LAMI NATES
( )( ) ( )( )
3
1
3
2
3
3
3
4
3
5
3
4
3
3
3
2
3
1
3
0
3 3
1
3
2
3
3
3
6
3
5
3
4
3
3
3
2
3
1
1
1 3
1
1
1 3
1 1

+
+ + + + +
+ + + + +

N N N N
N N N N
N
k
k
k
N
k
k
k
z z z z z z z z z z
z z z z z z z z z z z z
L
L
.
Note for a regular balanced symmetric laminate that:
, , , , ,
3 3 2 2 1 1 0 k N k N N N N
z z z z z z z z z z

K .
Therefore, the summations become:
( )( ) ( )( )
0
1 1
3
0
3
0
3
1
3
2
3
3
3
4
3
5
3
4
3
3
3
2
3
1
3
0
3
0
3
1
3
2
3
3
3
6
3
5
3
4
3
3
3
2
3
1
1
1 3
1
1
1 3

+ + + + +
+ + + + +

+
z z
z z z z z z z z z z
z z z z z z z z z z z z
N
k
k
k
N
k
k
k
L
L
.
Through a similar process, the three remaining brackets in the expression for
xs
D
are also shown to be zero. Therefore, the coupling bending stiffness between
normal and shear,
xs
D , is zero. The derivation for the other coupling bending
stiffness between normal and shear,
ys
D , is the similar, with identical results.
Consequently, the bending stiffness for a regular balanced symmetric angle-ply
laminate in terms of the lamina stiffnesses is:
( )
( )
'
,
_
'
1
1
1
1
1
1
1
1
]
1
,
_
'
66
12
22
11
2
2 2 2 2 2 2 2 2
2 2 4 4 2 2 2 2
2 2 2 2 4 4
2 2 2 2 4 4
2
1
2 3
0 0 0 0
0 0 0 0
2
4
4 2
4 2
2
12
Q
Q
Q
Q
n m n m n m n m
n m n m n m n m
n m n m m n
n m n m n m
z t t
N
D
D
D
D
D
D
N
k
k
ys
xs
ss
xy
yy
xx
, 127
or, in terms of the invariant properties,
( )
'
'
1
1
1
1
1
1
1
1
]
1
,
_
'
Q
Q
Q
Q
Q
Q
N
k
k
ys
xs
ss
xy
yy
xx
U
U
U
U
U
U
z t t
N
D
D
D
D
D
D
3
2
5
4
1
1
2
1
2 3
1
0 0 0
0 0 0
4 cos - 0
4 cos - 0
4 cos 2 cos -
4 cos 2 cos
2
12
. 128
LAMI NATES
When the results for a regular balanced symmetric angle-ply laminate are collected,
the force-moment strain-curvature relationship is:
'
1
1
1
1
1
1
1
]
1
'
0
0
0
0
0
0
0 0
0
0
0 0 0
0 0 0
0 0 0
0 0 0
0 0 0
0 0 0
0 0
0
0
xy
y
x
s
y
x
ss
yy xy
xy xx
ss
yy xy
xy xx
yy
yx
xy
xy
y
x
D
D D
D D
A
A A
A A
M
M
M
N
N
N
. 129
By way of commentary on the specializations presented above, they provide
engineering insight into the behavior of the various components of the stiffness
terms. As such, they are important engineering developments. However, in
practical application, direct calculation of the stiffnesses for the laminate in
question is less restrictive and no more complicated. The preprocessing components
of various finite element programs have the capability to calculate the laminate
stiffnesses based on the laminate structure. Alternately, spreadsheets can be used
to perform the stiffness calculations.
Figure 11. A Plot of the Stress Distribution Through a [ ]
S
45 4t T300/N5208
Laminate Under a Mixed Force-Moment Load.
Stress Distributions
- 0. 0500
- 0. 0400
- 0. 0300
- 0. 0200
- 0. 0100
0. 0000
0. 0100
0. 0200
0. 0300
0. 0400
0. 0500
- 20 - 15 - 10 - 5 0 5 10 15
(ksi)
z

(
i
n
)
sx
sy
ss
LAMI NATES
STRESS DI STRI BUTI ON I N A LAMI NATE
The Kirchoff-Love model for the displacement of a plate leads to a linear
distribution of the strain through the thickness of the plate. From a practical
viewpoint, the assumption that the laminae are perfectly bonded together at their
interfaces requires that the displacements of adjacent laminae are equal at the
interface. Consequently, the strain distribution is continuous through the
laminate. However, continuity of strain in a laminate does not imply continuity of
stress. The laminae stresses are calculated from the laminae strains and the
laminae properties. The discontinuity of the laminae properties produces the
discontinuity of the stress through a laminate.
Figure 11 shows a plot of the normal and shear stresses in a regular balanced
symmetric angle-ply laminate due to a mixed force-moment loading. Note in this
figure the significant discontinuities in all of the stresses. Between the first and
second ply, the x-direction normal stress has a 17.8 kpsi jump. Figure 12 shows a
plot of the normal and shear stresses in another angle-ply laminate due to a mixed
force-moment loading. Note in this figure the jump in the x-direction normal stress
between the first and second ply is approximately 22 kpsi. When a laminate
experiences such a large discontinuity in any of the stresses the interfacial bond is
highly stressed and may fail producing a delamination. Delamination of a laminate
Stress Distributions
- 0. 0500
- 0. 0400
- 0. 0300
- 0. 0200
- 0. 0100
0. 0000
0. 0100
0. 0200
0. 0300
0. 0400
0. 0500
- 20 - 15 - 10 - 5 0 5 10 15
(ksi)
z

(
i
n
)
sx
sy
ss
Figure 12. A Plot of the Stress Distribution Through a [ ]
S
10 4t T300/N5208
Laminate Under a Mixed Force-Moment Load.
LAMI NATES
is a potentially hazardous situation that is not, in general, addressed by laminate
failure theories.
LAMI NATE FAI LURE THEORI ES
Laminate failure analysis is based primarily on lamina failure analysis.
Laminates by their definition are inhomogeneous. Consequently, laminate failure
theories are largely empirical. Alternately, lamina failure theories, though also
empirical in origin, have a phenomenological connection. Knowing that stress
distribution in a laminate it is possible to apply lamina failure theories to the
individual lamina. This leads to two laminate failure theories, First-Ply-Failure
(FPF) and Last-Ply-Failure (LPF). Drawing an analogy to metallic material
behavior, ductile materials display two critical points in their stress-strain plot the
yield point and the ultimate load. The first-ply-failure load is analogous to the
yield point of a metal. When one ply fails, the load carrying capability of the
laminate is degraded; however, it does not generally fail catastrophically at that
load. Many laminates can continue to carry increasing loads safely after first-ply-
failure, though at reduced stiffness. The last-ply-failure load is analogous to the
ultimate load for a metal. At the last-ply-failure load, the structure fails
catastrophically.
FIRST-PLY-FAILURE
To determine the first-ply-failure load the analyst must know the laminate
stacking sequence, laminae properties, and the laminate loading. The laminate
loading may be either applied forces and moments or the resulting mid-plane
strains and curvatures. The latter is easier than the former since laminae strains
can be calculated directly from the mid-plane strains and curvatures. For the
former case, in which the forces and moments are known, the laminate stiffnesses
and the inverse (compliance) must be calculated and the mid-plane strains and
curvatures calculated in turn.
Once the mid-plane strains and curvatures are known, they are used to
calculate the strains in the individual laminae. At this point, there is some
flexibility in the process. I f the off-axis strain formulation of the quadratic failure
criterion is used, the first-ply-failure analysis goes directly into the calculation of
the R-factors. I f the on-axis strain formulation of the quadratic failure criterion is
used, the lamina strains must be transformed to the laminae principal material
directions and the failure criterion applied. I f either stress form of the quadratic
failure criterion is used, the lamina strains are used to calculate the lamina
stresses. Then, if the off-axis form of the quadratic failure criterion is used, the
analysis goes to calculation of the R-factors. The final option, using the on-axis
form of the quadratic failure criterion, requires transforming the lamina stresses to
the lamina principal material directions and then applying the failure criterion.
LAMI NATES
The first-ply-failure load is determined by scaling the original applied load
by the minimum of the R-factors calculated for all of the laminae. This process is
Laminate Stacking Sequence
Laminae Material Properties
Laminate Loading
Calculate Laminae Strains
Mid-plane Strains and
Curvatures Known?
Yes No
Calculate Laminate Compliance
Calculate Mid-plane
Strains and Curvatures
Calculate Laminate Stiffnesses
Calculate Laminae Stresses
Calculate Laminae R-Factors
Select Minimum R-Factor
Scale Laminate Load by
Minimum R-Factor

Figure 13. Flowchart for First-Ply-Failure Calculation.
LAMI NATES
shown in the flowchart presented in Figure 13. Note in this flowchart, as described
in the paragraph above, the additional steps that must be performed in the event
the mid-plane strains and curvatures are not known.
LAST-PLY-FAILURE
Last-ply-failure analysis begins with a first-ply-failure analysis. After the
first-ply-failure load is determined, the assumption is made that the failed ply (or
plies) makes no contribution to the stiffness of the laminate. Therefore, the
laminate stiffness must be recalculated. The flowchart shown in Figure 14
describes a last-ply-failure analysis. I n this process the forces and moments are
used to determine the failure load. The analysis loops through the same steps until
all lamina have been determined to have failed. Key to the process is the setting to
zero of the failed lamina stiffnesses. Note that the failed laminae are not removed
from the laminate, only the stiffnesses are set to zero. This is significant to the
laminate bending stiffness. Recall that the bending stiffnesses are dependent on
the centroidal coordinate of the laminae. I f the failed laminae were totally removed
from the laminate, the laminate stiffness calculation would not accurately reflect
the vertical spacing of the remaining laminae. Therefore, only the stiffnesses of the
failed laminae are set to zero.
As an analysis proceeds through the LPF calculations, the intermediate
values of the failure loads do not necessarily increase or decrease. Figure 15 shows
a plot of critical bending moment values versus mid-plane curvature from a last-
ply-failure analysis. Note that the curve is not monotonic. There is a generally
decreasing trend from the maximum load, which for this laminate was the initial
failure load. However, there is a region, after the initial failure, in which the
laminate sustains increasing loads.
I t is important to note that a last-ply-failure analysis presupposes that the
laminate will not fail catastrophically after the first-ply-failure load. I t is not at all
impossible in a dynamic situation for the failure to cascade through the laminate
although a last-ply-failure analysis might indicate that the laminate should not
fail. I n experimental investigation of the fracture of materials investigators
characterize the text machines used as hard and soft. This characterization has to
do with the ability of the machine to continue to apply loads to a specimen after a
maximum load is applied without catastrophic failure. I n this application, if the
structure supporting the laminate maintains a constant load (force-moment) the
laminate is more likely to fail catastrophically. Conversely, if the structure
supporting the laminate maintains a constant displacement (strain-curvature), the
laminate is less likely to fail catastrophically.
LAMI NATES

Figure 14. Flowchart for Last-Ply-Failure Calculation.
First-Ply-Failure Analysis
Set Failed Laminae Stiffnesses to Zero
Mid-plane Strains
and Curvatures
Yes No
Calculate FPF Forces and Moments
Calculate Mid-plane Strains and Curvatures
Calculate Laminate
Re-calculate Laminate Stresses
Calculate Laminae Strains
Select Minimum R-Factor
Scale Laminate Load by Minimum R-Factor
Calculate Laminae R-Factors
All Plies Failed?
Yes No
End
LAMI NATES

0
1
2
3
4
5
6
0 0.1 0.2 0.3 0.4 0.5
k
x
1/in.
M
x

(
l
b
.

i
n
.
/
i
n
.
)
Figure 15. A Plot of Applied Bending Moment Versus Mid-Plane Curvature for
an Angle-Ply Laminate From a Last-Ply-Failure Analysis.

Composite M Book

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Composite M Book

Uploaded by

Copyright:

Available Formats

COMPOSITE

, and upper-case Roman numeral subscripts, i. e., I , I I ,

, the sum of the forces in the

, and assuming that the laminate is homogeneous in

, and again assuming that the laminate

You might also like