Composite structures under ballistic impact

N.K. Naik *, P. Shrirao

Aerospace Engineering Department, Indian Institute of Technology-Bombay Powai, Mumbai 400 076, India

Abstract
In the present study, investigations on the ballistic impact behaviour of two-dimensional woven fabric composites has been
presented. Ballistic impact behaviour of plain weave E-glass/epoxy and twill weave T300 carbon/epoxy composites has been com-
pared. The analytical method presented is based on our earlier work. Different damage and energy absorbing mechanisms during
ballistic impact have been identified. These are: cone formation on the back face of the target, tensile failure of primary yarns,
deformation of secondary yarns, delamination, matrix cracking, shear plugging and friction during penetration. Analytical formu-
lation has been presented for each energy absorbing mechanism. Energy absorbed during each time interval and the corresponding
reduction in velocity of the projectile has been determined. The solution is based on the target material properties at high strain rate
and the geometry and the projectile parameters. Using the analytical formulation, ballistic limit, contact duration at ballistic limit,
surface radius of the cone formed and the radius of the damaged zone have been predicted for typical woven fabric composites.

Keywords: Ballistic impact; Woven fabric composite; Energy absorbing mechanisms; E-glass/epoxy; T300 carbon/epoxy

1. Introduction during ballistic impact are: kinetic energy absorbed by

Composite structures undergo various loading con-
ditions during the service life. For the effective utiliza-
tion of composites as protective structures, the critical
requirement is the resistance to foreign object ballistic
impact loading. For such applications, the condition for
the penetration of the projectile into the composite
target and the associated damage mechanisms should be
clearly understood. Post ballistic impact residual
strength is also an important consideration. Impact
loads can be classified into three categories: low velocity
impact, high velocity impact and hyper velocity impact.
The reason for this classification is that energy transfer
between projectile and target, energy dissipation and
damage propagation mechanisms undergo drastic
changes as the velocity of the projectile changes. One of
the possible ways of enhancing the ballistic limit is to
utilize textile composites.
Kinetic energy of the projectile when impacted into
the target is dissipated and absorbed in various ways by
the target. The main energy absorbing mechanisms
the moving cone formed on the back face of the target,
shear plugging of the projectile into the target, energy
absorbed due to tensile failure of the primary yarns,
energy absorbed due to elastic deformation of the sec-
ondary yarns, energy absorbed due to matrix cracking
and delamination and frictional energy absorbed during
penetration.
An impact event is considered to be a low velocity
impact if the contact period of impactor is longer than
the time period of the lowest vibrational mode. In low
velocity impact regime, the support conditions are cru-
cial as the stress waves generated outward from the
impact point have time to reach the edges of the struc-
tural element, causing its full-vibrational response. On
the other hand, in high velocity or ballistic impact, the
response of the structural element is governed by the
‘local’ behaviour of the material in the neighborhood of
the impacted zone, the impact response of the element
being generally independent of its support conditions.
The contact period of the impactor is much smaller than
the time period of lowest vibrational mode of the
structure. Hyper velocity impact involves projectiles
moving at extremely high velocities such that the local
target materials behave like fluids and the stress induced
by the impact is many times the material strength.
580 N.K. Naik, P. Shrirao / Composite Structures 66 (2004) 579–590

Nomenclature

a Yarn width Pd percent delaminating layers

A cross-sectional area of fibre/yarn Pm percent matrix cracking
Aql quasi-lemniscate area reduction factor rdi radius of the damage zone at time ti
Adi damaged area at time ti rpi distance covered by plastic wave till time ti
b stress wave transmission factor rti surface radius of the cone at time ti , distance
ce elastic wave velocity covered by transverse wave till time ti
cp plastic wave velocity rt1 surface radius of the cone at time t1
ct transverse wave velocity rt2 surface radius of the cone at time t2
d projectile diameter SSP shear plugging strength
da ; db damage distances ta ; tb time intervals
dci deceleration of the projectile during ith time tc contact duration of the projectile during
interval ballistic impact event
EKEi kinetic energy of the moving cone at time ti ti ith instant of time
ESPi energy absorbed by shear plugging till time ti Vf fibre volume fraction
EDi energy absorbed by deformation of second- Vi projectile velocity at time ti
ary yarns till time ti VBL ; V50 ballistic limit velocity
ETFi energy absorbed by tensile failure of primary VI ; VO incident ballistic impact velocity
yarns till time ti VR residual velocity
EDLi energy absorbed by delamination till time ti zi height/depth of the cone at time ti , distance
EMCi energy absorbed by matrix cracking till time traveled by projectile at time tiþ
ti e strain
Emt energy absorbed by matrix cracking per unit ed damage threshold strain
volume ei maximum tensile strain in primary yarns at
EFi energy absorbed by friction till time ti time ti
ETOTALi total energy absorbed by the target till time ti ep plastic strain
Fi contact force during ith time interval epy strain in primary yarns
GIIcd critical dynamic strain energy release rate in esy strain in secondary yarns
mode II e0 maximum strain in a yarn/fibre at any mo-
h target thickness ment
hl layer thickness r stress
KEpi kinetic energy of the projectile at time ti rp plastic stress
MCi mass of the cone at time ti rsy stress in secondary yarns
mp mass of the projectile q density of the target
N number of layers being shear plugged in a Dt duration of time interval that tc is divided
time interval into

The composite structural applications which involve
survivability of personnel and equipments against pene-
tration by high velocity projectiles are of much import.
Thus, it is important to have an understanding of the
penetration process of the composites by a projectile
under ballistic velocities. Considering the importance of
this subject there are many experimental studies on the
ballistic impact of composites. But the experimental
studies are specific to certain types of composite materials
under certain ballistic impact conditions. Based on these
experimental studies the ballistic impact behaviour of a
class of composite materials cannot be characterized. For
complete understanding of a class of composite materials,
analytical studies are necessary based only on mechani-
cal/fracture and geometrical properties of target and
ballistic impact parameters. But the analytical studies are
very limited. The ballistic impact performance of com-
posite laminates is dependent on the mechanical charac-
teristics of the reinforcement/matrix and the physical
characteristics of the impacting projectile and the target.
Mechanical characteristics of the target include the elastic
modulus, tensile strength, fracture strain and laminate
configuration. Projectile characteristics include mass and
shape. Thickness and size are the target parameters that
affect the ballistic impact performance. Ballistic impact is
generally a low-mass high-velocity impact caused by a
propelling source. Ballistic limit of a target is defined as
the maximum velocity of a projectile at which the com-
plete perforation takes place with zero exit velocity.
Limited number of semi-analytical and semi-numer-
ical models, which require input details from a few
experiments, are available [1–11]. But in these studies all
 N.K. Naik, P. Shrirao / Composite Structures 66 (2004) 579–590
the energy absorbing mechanisms have not been con-
sidered. Also, complete details are not provided.
The objective of the present work is to present an
analytical model for the ballistic impact analysis based
only on dynamic mechanical and fracture properties of
the target material, geometrical parameters of the target
and the projectile parameters, such as mass, velocity,
shape and size. The studies have been presented for
typical woven fabric E-glass/epoxy and T300 carbon/
epoxy composites impacted with flat ended cylindrical
projectiles. The analytical method presented can be used
for predicting the energy absorbed by different damage
and energy absorbing mechanisms, ballistic limit, con-
tact duration and the damage size. Ballistic impact
behaviour of plain weave E-glass/epoxy and twill weave
T300 carbon/epoxy composites has been compared.
Energy absorbed by various mechanisms is calculated
for a given mass and velocity of the projectile. By
knowing the energy absorbed by various mechanisms,
the total energy absorbed during impact event is calcu-
lated as the sum of the individual energies absorbed.
The energy absorbed at each time interval is calcu-
lated. Based on this, the velocity at the end of the cor-
responding time interval is calculated. This procedure is
continued for the entire contact duration. At ballistic
limit the target absorbs the entire energy of the projec-
tile. So the energy absorbed by the target is made equal
to the initial kinetic energy of the projectile. From this
the velocity of the projectile, which is the ballistic limit
of the target, is calculated.
The input to the present analytical model are: the
dynamic mechanical properties of the target, target
geometrical details and projectile shape, size, mass and
incident ballistic impact velocity. The output that can be
obtained are the ballistic limit of the target, contact
duration and damage dimensions.
2. Damage mechanisms
For the complete understanding of the ballistic im-
pact of composites, different damage and energy
absorbing mechanisms should be clearly understood.
Possible energy absorbing mechanisms are: cone for-
mation on the back face of the target, deformation of
secondary yarns, fracture of primary yarns/fibres,
delamination, matrix cracking, shear plugging and fric-
tion between the projectile and the target. For different
materials like carbon, glass or Kevlar, different mecha-
nisms can dominate. Also, the reinforcement architec-
ture can influence the energy absorbing mechanisms.
2.1. Cone formation on the back face of the target
Morye et al. [9] carried out ballistic impact tests on
woven Nylon prepreg composites. They monitored the
581
back face deformation behaviour using high speed
photography. Their study showed that a cone was
formed on the back face of the composite during the
ballistic impact event. Experimental studies by Zhu et al.
[12] on woven Kevlar/polyester laminates of varying
thickness also showed the cone formation on the back
face of the composites during ballistic impact event. The
cone formation on the back face can be explained on the
basis of transverse wave propagation during ballistic
impact.
Figs. 1 and 2 show the deformation and cone for-
mation on the back face of the composite during bal-
listic impact. Here, d is the diameter of the projectile, rti
is the surface radius of the cone formed and zi is the
distance traveled by the projectile. During the ballistic
impact event, the distance traveled by the projectile and
the depth of the cone formed are equal. Also, the
velocities of the projectile and the cone would be equal.
The yarns directly below the projectile are called
primary yarns. These yarns provide the resistive force to
the projectile penetration into the target. The remaining
yarns within the conical region are known as secondary
yarns. These yarns deform and cause some energy
absorption. The conical portion has a finite velocity
equal to that of the projectile.
For the analysis of the ballistic impact event, the
contact duration can be subdivided into certain number
of time intervals of duration Dt. The damage mechanism
and the energy absorption can be determined for each
Fig. 1. Cone formation during ballistic impact on the back face of the
composite target.
582 N.K. Naik, P. Shrirao / Composite Structures 66 (2004) 579–590
Fig. 2. Damage propagation in composite laminate during ballistic
impact event.
time interval and this effect can be taken into account
while proceeding with further time intervals. The surface
radius of the cone formed can be calculated on the basis
of transverse wave propagation. The transverse wave
velocity depends on the stress-strain curve at high strain
rates for the material.
During the second time interval the surface radius of
the cone increases from rt1 to rt2 . Surface radius rt2 is
calculated based on how much additional transverse
wave propagation has taken place during that time
interval. As the transverse wave propagation continues,
surface radius of the cone rti , cone depth zi , mass of the
cone MCi and the strain in the fibres ei increase. As the
strain within the fibre exceeds the permissible dynamic
tensile strain, the corresponding fibre fails. Increase in
rti , zi , MCi and ei continues until all the fibres are broken
or the entire energy of the projectile is absorbed by the
target. It may be noted that the strain within the primary
yarns is not constant. At any moment, it is maximum at
the point of impact and decreases as we move away from
the point of impact.
Failure of all the fibres indicates complete perforation
of the projectile into the target. At this stage, if the
projectile still has some kinetic energy, it would be
exiting with certain velocity. If the exit velocity is zero,
the corresponding incident ballistic impact velocity of
the projectile is the ballistic limit. If the projectile
velocity is zero during the contact event before all the
fibres are broken, it indicates that only the partial pen-
etration has occurred.
2.2. Energy absorbed due to deformation of secondary
yarns
Possible cone formation on the back face of the target
during ballistic impact event is shown in Fig. 1. The
primary yarns take the direct force and ultimately fail if
the strain exceeds the maximum strain limit. As a result
the complete target perforation takes place. All the
yarns apart from the primary yarns, i.e., the secondary
yarns, shown in Fig. 1 deform and absorb some energy.
The energy absorbed by secondary yarns depends on the
strain distribution within the secondary yarns. Strain in
the secondary yarns near the tip of the cone, i.e., at point
A is the same as that in the primary yarns and it de-
creases linearly to zero as we move out radially to point
B.
2.3. Energy absorbed due to tensile failure of primary
yarns
Primary yarns/fibres provide the force to resist the
penetration of the projectile into the target. The strain is
maximum in these yarns/fibres. These yarns fail under
tension when the strain of these yarns exceeds the ulti-
mate strain in tension at high strain rate. Failure of these
yarns/fibres results in absorption of some energy. Fig. 1
indicates the strain variation in primary yarns within a
layer and in through the thickness direction.
During cone formation, the tensile strain would be
maximum along the middle primary yarn, in each layer.
Within the primary yarns in a layer strain reduces
gradually, as we move away from the mid primary yarn.
This is due to the fact that the middle yarn is pushed by
complete projectile diameter, but other primary yarns
face a projectile length which is lesser than projectile
diameter.
The top layers of the laminate when impacted get
compressed apart from the cone formation, which cau-
ses the strain in the top layers to be more than that in the
bottom layers. The extent of compression is based on
the stress at the point of impact. This stress causes an
extra straining of the top layers. The variation of strain
in through the thickness direction is shown in Fig. 1.
The strain in the top layers is more than the strain in the
bottom layers. Each primary yarn has different strain.
As and when the strain of a particular yarn reaches the
dynamic tensile failure strain, failure occurs to that
particular yarn. As a result, there will be a sequential
failure of yarns starting with the yarns in the top layer,
then proceeding toward the bottom layer. It may be
noted that the strain within the primary yarn is not
constant. At any moment it is maximum at the point of
impact and decreases as we move away from the point of
impact.
2.4. Energy absorbed due to shear plugging
Cantwell and Morton [13] observed frustum-shaped
shear cut-out zone on impacting carbon fibre reinforced
angle ply laminates with steel spherical bullets. Lee and
Sun [14] and Ellis [15] reported shear plugging to be one
 N.K. Naik, P. Shrirao / Composite Structures 66 (2004) 579–590
Fig. 3. Schematic showing plug formation.
of the major damage modes on impact of angle ply
graphite/bismaleimide by blunt/flat projectiles. Fig. 3
shows schematic arrangement of plug formation during
ballistic impact. Brittle nature of carbon based com-
posites was responsible for change in damage mecha-
nism. Shear plug formation is not observed for glass
reinforced composites which have high failure strain at
high strain rates.
2.5. Energy absorbed by other mechanisms
Delamination and matrix cracking would absorb
some energy (Fig. 2). Additionally, friction between the
projectile and the target would also absorb some energy.
3. Analytical formulation for woven fabric composites
On striking a target, energy of the projectile is ab-
sorbed by various mechanisms like kinetic energy of
moving cone EKE , shear plugging ESP , deformation of
secondary yarns ED , tensile failure of primary yarns ETF ,
delamination EDL , matrix cracking EMC , friction energy
EF along with other energy dissipation mechanisms. An
analytical model is presented for the above-mentioned
mechanisms, which dominate during ballistic impact of
two-dimensional woven fabric composites [16,17].
3.1. Analytical formulation
A cone is observed on the back face of the target as
shown in Fig. 1 when struck by a projectile [9,12]. Shear
plugging on the front face and cone formation on the
back face start taking place depending on the target
material properties during the ballistic impact event.
The velocity of the cone is the same as the velocity of the
projectile. Initially, the cone has velocity equal to that of
the projectile and has zero mass. As time progresses, the
mass of the cone increases and velocity of the cone de-
creases. As the cone formation takes place, the yarns/
fibres deform and absorb some energy. The primary
yarns which provide the resistive force to the projectile
motion deform the most, thus leading to their failure.
When all the primary yarns fail the projectile exits the
583
target. Tensile failure of the yarns thus absorbs some
energy of the projectile. During the ballistic impact
event, delamination and matrix cracking take place in
the laminate area which forms the cone. The total ki-
netic energy of the projectile that is lost during ballistic
impact is the total energy that is absorbed by the target
till that time interval and is given by
ETOTALi ¼ EKEi þ ESPi þ EDi þ ETFi þ EDLi þ EMCi þ EFi
ð1Þ
An analytical model to predict the ballistic limit is
presented below. Firstly, modelling of a single yarn is
presented. The model is further extended to the woven
fabric composites. The analytical model is for the case
when the projectile impacts the target normally.
The following assumptions are made in the analytical
model presented:
(1) Projectile is perfectly rigid and remains un-deformed
during the ballistic impact.
(2) Projectile motion is uniform during penetration
within each time interval.
(3) Yarns/fibres in a layer act independently.
(4) Energy absorption due to primary yarn/fibre break-
age and deformation of the secondary yarns are
treated independently.
(5) Longitudinal and transverse wave velocities are the
same in all the layers.
3.2. Modelling of a single yarn subjected to transverse
ballistic impact
When a yarn is impacted by a projectile transversely,
longitudinal strain wavelets propagate outward along
the filament [4,18,19]. The outermost wavelet, called the
elastic wave propagates at velocity
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 
1 dr
ce ¼ ð2Þ
q de e¼0
The innermost longitudinal wavelet called the plastic
wave propagates at velocity
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 
1 dr
cp ¼ ð3Þ
q de e¼ep
Here, ep denotes the strain at which plastic region starts.
As the strain wavelets pass a given point on the yarn,
material of the filament flows inward toward the impact
point. The material in the wake of the plastic wave front
forms itself into a transverse wave, shaped like an in-
verted tent with the impact point at its vertex. In a fixed
coordinate system, the base of the tent spreads outward
with the transverse wave velocity. This is depicted in
Fig. 4. The transverse wave velocity is given by
584 N.K. Naik, P. Shrirao / Composite Structures 66 (2004) 579–590

critical strain energy release rates along warp/fill direc-

tions and the other directions. For simplicity, the cone
formed can be considered to be of circular cross-section,
rather than quasi-lemniscate. The base of the cone
spreads with a transverse wave velocity
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Z sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 ffi
ep
ð1 þ ep Þrp 1 dr
ct ¼  de ð7Þ
q 0 q de

and the plastic wave velocity is given by

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 
1 dr
cp ¼ ð8Þ
q de e¼ep

After time ti the transverse wave has traveled to a

Fig. 4. Configuration of a yarn/fibre before and after transverse
impact: (a) before impact; (b)–(d) after impact.
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Z sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 ffi
ep
ð1 þ ep Þrp 1 dr
ct ¼  de
q 0 q de
If the complete impact event is divided into a number
of small instants, then at ith instant, the time is given by
ti . By that time the transverse wave has traveled to a
distance rti and the plastic wave has traveled to a dis-
tance of rpi . The projectile has moved through a distance
zi .
Radii rti and rpi after time ti ¼ iDt is given by
X
n¼i
rti ¼ ctn Dt
n¼0
X
n¼i
rpi ¼ cpn Dt
n¼0
3.3. Modelling of woven fabric composite
Woven fabric composites consist of warp and fill
yarns, interlaced in a regular sequence. As the projectile
impacts on to the woven fabric composite, there can be
many yarns beneath the projectile. In the present anal-
ysis, all the warp and fill yarns are treated separately.
Behaviour of each yarn is analysed as explained in the
previous section. Overall behaviour of the woven fabric
composite is presented considering the combined effect
of all the individual yarns within a layer.
3.4. Strain and deceleration history: single yarn of woven
fabric composite
When a projectile impacts a woven fabric composite,
a cone with quasi-lemniscate cross-section would be
formed on the back face of the target. Quasi-lemniscate
shape is because of difference in elastic properties and
distance rti and the plastic wave has traveled to a dis-
tance of rpi . The projectile has moved through a distance
zi . Radii rti and rpi after time ti ¼ iDt are given by
X
n¼i
rti ¼ ctn Dt ð9Þ
n¼0
ð4Þ X
n¼i
rpi ¼ cpn Dt ð10Þ
n¼0
Because of stress wave attenuation, the stress in the
yarn and hence, the strain in the yarn varies with the
distance from the point of impact. Stress and in turn
strain is maximum at the point of impact and decreases
as one moves away from the point of impact. At ith time
interval, the strain at the point of impact is given by
8 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi  9
ð5Þ < ðd=2Þ þ ðrti  ðd=2ÞÞ2 þ z2i þ rpi  rti  rpi =
ei ¼
: bðrpi =aÞ  1 ;
 
ð6Þ ln b
 ð11Þ
a
Once, the strain variation in a yarn/fibre is known,
the energy absorbed by various mechanisms like kinetic
energy of the moving cone, deformation energy ab-
sorbed by secondary yarns, tensile failure of primary
yarns, delamination and matrix cracking can be calcu-
lated. The formulation for energy absorbed by various
energy absorbing mechanisms is provided in the fol-
lowing sections.
At the beginning of the first time interval of impact,
entire energy is in the form of kinetic energy of the
projectile. Later, this energy is divided into energy ab-
sorbed by various damage mechanisms and the kinetic
energy of moving cone and projectile. Considering the
energy balance at the end of ith time interval, we obtain
KEp0 ¼ KEpi þ EKEi þ ESPði1Þ þ EDði1Þ þ ETFði1Þ
þ EDLði1Þ þ EMCði1Þ þ EFði1Þ ð12Þ
Rearranging the terms in the above equation, the fol-
lowing form is obtained:
 N.K. Naik, P. Shrirao / Composite Structures 66 (2004) 579–590 585

 3.5. Kinetic energy of cone formed

1
m V2
2 p 0
 Ei1 ¼ 12 mp þ MCi Vi 2 ð13Þ
The cone formed on the back face of the target ab-
where
sorbs some energy. By the end of ith time interval the
radius of the cone formed is
Ei1 ¼ ESPði1Þ þ EDði1Þ þ ETFði1Þ þ EDLði1Þ
X
n¼i
þ EMCði1Þ þ EFði1Þ ð14Þ rti ¼ ctn Dt ð20Þ
n¼0
Each of the contributing terms to Ei is explained in Mass of the cone formed is
the succeeding sections. The terms on the right-hand
side of Eq. (14) are known at (i  1)th instant of time. MCi ¼ prti2 hq ð21Þ
From this, the velocity of the projectile at the end of ith The velocity of the cone formed is equal to Vi , the
time interval can be obtained and it is given by velocity of the projectile at the end of ith time interval.
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi So the energy of the cone formed at the end of ith time
1
mp V 2  Ei1 interval is
Vi ¼ 21  0 ð15Þ
2
mp þ MCi EKEi ¼ 12MCi Vi 2 ð22Þ

If the projectile velocity is known at the beginning

3.6. Energy absorbed due to shear plugging
and the end of ith time interval, then the deceleration of
the projectile during that time interval can be found out.
When the target material is impacted by the projec-
It is given by
tile, shear plugging stress in the material near projectile
Vi1  Vi periphery rises. As and when the shear plugging stress
dci ¼ ð16Þ exceeds shear plugging strength, shear plugging failure
Dt
occurs. As a result, plug formation takes place. This
Note that the velocity at the beginning of the ith time phenomenon is observed for carbon/epoxy composites
interval is the same as the velocity at the end of (i  1)th (Fig. 3). If at the beginning of the ith time interval, shear
time interval. Distance traveled by the projectile zi up to plugging stress exceeds shear plugging strength, then the
ith time interval is given by energy absorbed by shear plugging during that time
interval is given by the product of distance sheared,
X
n¼i
zi ¼ Dzn ð17Þ shear plugging strength and the area over which shear
n¼0 plugging stress is applied. It is given by
DESPi ¼ Nhl SSP p dh ð23Þ
1 2
Dzi ¼ Vi1 Dt  dci ðDtÞ ð18Þ where N indicates the number of layers shear plugged
2
during ith time interval and SSP denotes shear plugging
Utilizing dci , the deceleration of the projectile during strength. The energy absorbed by shear plugging by the
ith time interval, the force resisting the projectile motion end of ith time interval is given by
can be calculated. It is given by X
n¼i
ESPi ¼ DESPn ð24Þ
Fi ¼ mp dci ð19Þ n¼0

The magnitude of the force on the projectile is the same

as the magnitude of the force applied on the target, by
the projectile. Hence, this force can be used to determine
whether shear plugging takes place or not.
The above process is repeated until all the primary
yarns in the target fail, i.e., the complete perforation
takes place. The velocity at the end of time interval,
during which all the yarns are broken, is the residual
velocity of the projectile. On the other hand, if the
numerator in Eq. (15) becomes zero, then the projectile
does not penetrate the target completely with the given
initial velocity. Thus by repetition of above procedure
with various velocities so as to get complete perforation
with zero residual velocity, the ballistic limit of the tar-
get laminate can be obtained.
3.7. Energy absorbed due to deformation of secondary
yarns
The secondary yarns experience different strains
depending on their position. The yarns which are close
to the point of impact experience a strain equal to the
strain in the outermost primary yarn, whereas those
yarns which are away from the impact point experience
less strain. The strain variation is assumed to be linear
from A to B (Fig. 1). This imposes the following
boundary conditions for the variation in strain at time ti :
pffiffiffi
esy ¼ epy at r ¼ d= 2; i:e:; at point A
esy ¼ 0 at r ¼ rti ; i:e:; at point B
586 N.K. Naik, P. Shrirao / Composite Structures 66 (2004) 579–590
Here, epy is the strain in the outermost primary yarn in
that layer. The energy absorbed in the deformation of all
the secondary yarns can be obtained by the following
integration:
Z rti  Z esyi  The area undergoing delamination and matrix crack-

EDi ¼ pffiffi
rsy esy desy h
d= 2 0
 
 2pr  8r sin1 ðd=2rÞ dr ð25Þ
3.8. Energy absorbed due to tensile failure of primary
yarns
The yarns directly below the projectile, known as the
primary yarns, fail in direct tension. All the primary
yarns within one layer do not fail at one instant of time.
As and when the strain of a particular yarn reaches the
dynamic failure strain in tension, the yarn fails. It may
be noted that the length of yarns/fibres failing in tension
is twice the distance covered by the longitudinal wave.
Also, the complete length of a primary yarn is not
strained to the same extent as explained earlier. The
reason for this phenomenon is stress wave attenuation.
When the strain in yarns/fibres exceeds failure strain, it
fails and some energy is absorbed due to tensile failure.
For a yarn/fibre of cross-section area A it is given by
! EMCi ¼
Z x Z e¼e0 bx=a
ETF ¼ A rðeÞde dx ð26Þ
0 e¼0
where e0 is the ultimate strain limit. If during ith time
interval N numbers of yarns/fibres are failing, then the
right-hand side of the above expression is multiplied by N .
3.9. Energy absorbed due to delamination and matrix
cracking
Delamination and matrix cracking absorb some part of
the initial kinetic energy of the projectile. A part of the
conical area undergoes delamination and matrix cracking
(Fig. 2). The extent to which composite has delaminated
and the matrix has cracked till (i þ 1)th time instant can
be calculated on the basis of strain profile in the composite
at that time interval. From the results derived in the earlier
section it can be observed that the strain at the impact
point is the highest and it decreases as one moves away
from the point of impact. The area in which strain is more
than the damage initiation threshold strain ed , will un-
dergo damage in the form of matrix cracking and
delamination. However, complete matrix cracking may
not take place. Evidence for this phenomenon is provided
by the fact that after ballistic impact, matrix is still at-
tached to the fibres and does not separate from the rein-
forcement completely. Due to matrix cracking, the
interlaminar strength of the composite decreases. As a
result, further loading and deformation causes delami-
nation. This delamination is of mode II type. Again,
delamination may not occur at all the lamina interfaces.
Towards the end of ballistic impact event, when only a few
non-delaminated layers are left, these non-delaminated
layers are more likely to bend rather than delaminate.
ing in the conical region is of quasi-lemniscate shape,
which is taken to be Aql percent of the corresponding
circular area. During (i þ 1)th time interval the area of
delamination and matrix cracking is given by
2 2
pðrdðiþ1Þ  rdi ÞAql ð27Þ
where rdi indicates the radius up to which the damage
has propagated until ith time interval. So the respective
energies absorbed by delamination and matrix cracking
during this time interval are given by
2 2
DEDLi ¼ Pd pðrdðiþ1Þ  rdi ÞAql GIIcd ð28Þ
2
DEMCi ¼ Pm pðrdðiþ1Þ 2
 rdi ÞAql Emt h ð29Þ
The factors Pd and Pm stand for percentage delamination
and percentage matrix cracking. Energy absorbed till the
ith time interval is given by
X
n¼i
EDLi ¼ DEDLn ð30Þ
n¼1
X
n¼i
DEMCn ð31Þ
n¼1
3.10. Calculation of contact duration
The impact event starts when the projectile touches
the front face of the target. End of the event is taken to be
when all the primary yarns fail or when the velocity of
the projectile becomes zero. Total time from the start of
the impact event till the end of the event is called as
contact duration. If the end of the event occurs during
nth time interval, then the contact duration is obtained as
tc ¼ nDt ð32Þ
4. Results and discussion
Ballistic impact behaviour of woven fabric compos-
ites when impacted by a flat ended projectile has been
presented. This behaviour is based on the analytical
method presented in the previous section. Specifically,
ballistic limit, contact duration at ballistic limit and
surface radius of the cone formed at the ballistic limit
have been determined. These results have been com-
pared with the experimental results. Behaviour of plain
weave E-glass/epoxy and twill weave T300 carbon/epoxy
composite structures under ballistic impact has been
compared. Input data necessary for the analytical pre-
diction of ballistic impact behaviour in the present study
has been provided in Table 1.
 N.K. Naik, P. Shrirao / Composite Structures 66 (2004) 579–590 587

Table 1
Input parameters required for the analytical predictions of ballistic impact behaviour
Projectile details (cylindrical)
Mass 2.8 g 2.8 g
Shape Flat ended Flat ended
Diameter 5 mm 5 mm
Target details
Material Twill weave T300 carbon/epoxy Plain weave E-glass/epoxy
Vf 50% 50%
Thickness 2 mm 2 mm
Number of layers 5 6
Density 1400 kg/m3 1750 kg/m3
Other details
Stress–strain Fig. 9 Fig. 9
Time-step 1 ls 1 ls
Quasi-lemniscate area reduction factor 0.9 0.9
Stress wave transmission factor 0.95 0.825
Delamination percent 100% 100%
Matrix crack percent 100% 100%
Shear plugging strength 90 MPa –
Mode II dynamic critical strain energy release rate 800 J/m2 1000 J/m2
Matrix cracking energy 0.9 MJ/m3 0.9 MJ/m3
Fibre failure energy 6 MJ/m3 28 MJ/m3

4.1. Calculation of ballistic limit, contact duration and
surface radius of the cone formed during ballistic impact
Using the analytical method presented earlier ballistic
impact behaviour of typical plain weave E-glass/epoxy
and twill weave T300 carbon/epoxy composites has been
evaluated. The results are presented in Table 2 and Figs.
5–8. For the case of E-glass/epoxy, complete perforation
did not take place at VI ¼ 158 m/s whereas with VI ¼ 159
m/s, complete perforation took place with exit velocity
of 54 m/s. The experimentally obtained ballistic limit is
V50 ¼ 150 m/s. Analytically predicted ballistic limit is
between 158 and 159 m/s. For the case of T300 carbon/
epoxy, complete perforation did not take place with
VI ¼ 82 m/s whereas with VI ¼ 83 m/s, complete perfo-
ration took place with exit velocity of 18 m/s. The
above results are with mp ¼ 2.8 g, d ¼ 5 mm and h ¼ 2
mm.
For T300 carbon/epoxy the studies were also carried
out with mp ¼ 1:8 g, h ¼ 2 mm and d ¼ 5 mm. For this
case the experimentally obtained ballistic limit,
V50 ¼ 105 m/s. The analytically predicted ballistic limit
VBL ¼ 99 m/s (Table 2). It is generally observed that
there is a good co-relation between the experimental
results and the predicted results for both E-glass/epoxy
and T300 carbon/epoxy.
From Figs. 5 and 7 it can be seen that the ballistic
limit is much higher for E-glass/epoxy than for T300
carbon/epoxy. This is because of higher energy absorbed
by E-glass/epoxy than by T300 carbon/epoxy (Fig. 9).
Fig. 9 is for high strain rate loading. Also, shear plug-
ging strength for E-glass/epoxy is much higher than that
for T300 carbon/epoxy.
For E-glass/epoxy, energy absorbed by secondary
yarn deformation and tensile failure of yarns are the
main energy absorbing mechanisms. Shear plugging
does not take place in this case. On the other hand, for
T300 carbon/epoxy, the ultimate failure is because of
shear plugging. Energy absorbed by secondary yarn
deformation and shear plugging are the main energy
absorbing mechanisms (Figs. 6 and 8). In this case,
tensile failure of the yarns does not take place.

Table 2
Ballistic impact test results for typical plain weave E-glass/epoxy and twill weave T300 carbon/epoxy composites, d ¼ 5 mm, h ¼ 2 mm
Material Projectile Predicted ballistic Experimental Predicted damage Experimental Predicted surface
mass, mp (g) limit, VBL (m/s) ballistic limit, size, rd (mm) damage size, radius of the cone,
V50 (m/s) rd (mm) rt (mm)
Plain weave 2.8 159 150 9.6 10 35
E-glass/epoxy
Twill weave T300 1.8 99 105 – – 59
carbon/epoxy
Twill weave T300 2.8 83 – – – 61
carbon/epoxy
588 N.K. Naik, P. Shrirao / Composite Structures 66 (2004) 579–590

Projectile kinetic energy

(a) (b) Projectile kinetic energy
Cone kinetic enery
40 Cone kinetic energy
Shear plugging energy
Secondary yarn deformation energy Secodary yarn deformation energy
10 35
Delamination energy Tensile failure energy
Matrix cracking energy 30 Delamination energy
8 Matrix cracking energy
25

Energy,E(J)
6 20
Energy,E(J)

15
4
10

2 5

0
0
-5
0 100 200 300 400 500 600 700 0 20 40 60 80 100 120 140 160
Time,t (micro-sec) Time,t (micro-sec)

Fig. 5. Energy absorbed by different mechanisms during ballistic impact event, mp ¼ 2:8 g, h ¼ 2 mm, d ¼ 5 mm: (a) twill weave T300 carbon/epoxy
laminate, VI ¼ 82 m/s; (b) plain weave E-glass/epoxy laminate, VI ¼ 158 m/s.

(a) 90 1.0 100 (b) Projectile velocity

Projectile velocity
Maximum strain in to player
80 Maximum strain in top layer 0.9
Maximum strain in bottom layer
Maximum strain in bottom layer 0.8 80 Contact force
160 5 7
70 Shear plugging stress
0.7
Shear plugging stress (MPa)

140 6
60 4
0.6 60
Tensile strain (%)
Velocity, V(m/s)

120 5
50 0.5

Tensile strain (%)

3
Velocity, V(m/s)

100 4
40 0.4 40
80 2 3
0.3
30
60 2
0.2 20 1
20 40 1
0.1
0
10 0.0 0 20 0

0 -0.1 0 -1 -1
0 20 40 60 80 100 120 140 160
0 200 400 600 Time,t (micro-sec)
Time,t (micro-sec)

Fig. 6. Projectile velocity, strain, contact force and shear plugging stress variation during ballistic impact event, mp ¼ 2:8 g, h ¼ 2 mm, d ¼ 5 mm: (a)
twill weave T300 carbon/epoxy laminate, VI ¼ 82 m/s; (b) plain weave E-glass/epoxy laminate, VI ¼ 158 m/s.

(a) (b)
Projectile kinetic energy 40 Projectile kinetic energy
10 Cone kinetic energy
Cone kinetic energy
Shear plugging energy 35
Secondary yarn deformation energy
Secondary yarn deformation energy
8 30 Tensile failure energy
Delamination energy
Delamination energy
Matrix cracking energy 25
Matrix cracking energy
Energy, E (J)

Energy, E (J)

6
20

4 15

10
2
5

0
0

-5
-10 0 10 20 30 40 50 60 70 80 90 0 20 40 60 80 100 120 140 160
Time,t (micro-sec) Time,t (micro-sec)

Fig. 7. Energy absorbed by different mechanisms during ballistic impact event, mp ¼ 2:8 g, h ¼ 2 mm, d ¼ 5 mm: (a) twill weave T300 carbon/epoxy
laminate, VI ¼ 83 m/s; (b) plain weave E-glass/epoxy laminate, VI ¼ 159 m/s.
 N.K. Naik, P. Shrirao / Composite Structures 66 (2004) 579–590 589

Projectile velocity
Projectile velocity Maximum strain in top layer
(a) Maximum strain in top layer
(b)
Maximum strain in bottom layer
Maximum strain in bottom layer
160 Contact force 5 7
Shear plugging stress
90 1.0 160
140 6
0.9 4
80 140
0.8 120 5

Shear plugging stress (MPa)

Contact force, F (KN)

70 120
0.7 3

Tensile strain (%)

Velocity,V (m/s)
100

Tensile strain (%)

4
Velocity,V (m/s)

0.6 100
60
0.5 80 80 2 3
50
0.4
60 60 2
40 0.3 1
40
30
0.2 40 1
0.1 20 0
20 20 0
0.0
0
10 -0.1 0 -1 -1
-10 0 10 20 30 40 50 60 70 80 90 0 20 40 60 80 100 120 140 160
Time,t (micro-sec) Time,t (micro-sec)

Fig. 8. Projectile velocity, strain, contact force and shear plugging stress variation during ballistic impact event, mp ¼ 2:8 g, h ¼ 2 mm, d ¼ 5 mm: (a)
twill weave T300 carbon/epoxy laminate, VI ¼ 83 m/s; (b) plain weave E-glass/epoxy laminate, VI ¼ 159 m/s.

References

[1] Zhu G, Goldsmith W, Dharan CK. Penetration of laminated

Kevlar by projectiles––II. Analytical model. Int J Solids Struct
1992;29:421–36.
[2] Jenq ST, Jing HS, Chung C. Predicting the ballistic limit for plain
woven glass/epoxy composite laminate. Int J Impact Eng 1994;15:
451–64.
[3] Goldsmith W, Dharan CK, Chang H. Quasi-static and ballistic
perforation of carbon fibre laminates. Int J Solids Struct 1995;32:
89–103.
[4] Parga-Landa B, Hernandez-Olivares F. Analytical model to
predict behaviour of soft armours. Int J Impact Eng 1995;16:
455–66.
[5] Sun CT, Potti SV. A simple model to predict residual velocities of
thick composite laminates subjected to high velocity impact. Int J
Impact Eng 1996;18:339–53.
[6] Potti SV, Sun CT. Prediction of impact induced penetration and
delamination in thick composite laminates. Int J Impact Eng
Fig. 9. Stress–strain curves under tensile loading at high strain rate: (a) 1997;19:31–48.
E-glass/epoxy yarn; (b) T300 carbon/epoxy yarn. [7] Jenq ST, Kuo JT, Sheu LT. Ballistic impact response of 3-D four-
step braided glass/epoxy composites. Key Eng Mater Part 1
1998;141–143:349–66.
[8] Benloulo ISC, Sanchez-Galvez V. A new analytical model to
It is generally observed that the energy absorbed simulate impact onto ceramic/composite armors. Int J Impact Eng
by delamination and matrix cracking is of small mag- 1998;21:461–71.
nitude. [9] Morye SS, Hine PJ, Duckett RA, Carr DJ, Ward IM. Modelling
of the energy absorption by polymer composites upon ballistic
impact. Compos Sci Technonl 2000;60:2631–42.
[10] Billon HH, Robinson DJ. Models for the ballistic impact of fabric
armour. Int J Impact Eng 2001;25:411–22.
5. Conclusions [11] Wen HM. Penetration and perforation of thick FRP laminates.
Compos Sci Technol 2001;61:1163–72.
Ballistic impact behaviour of typical plain weave E- [12] Zhu G, Goldsmith W, Dharan CK. Penetration of laminated
Kevlar by projectiles––I. Experimental investigation. Int J Solids
glass/epoxy and twill weave T300 carbon/epoxy com-
Struct 1992;29:399–419.
posites has been studied. It has been observed that, for [13] Cantwell WJ, Morton J. Impact perforation of carbon fibre
identical ballistic impact conditions, ballistic limit is reinforced plastic. Compos Sci Technol 1990;38:119–41.
higher for E-glass/epoxy than for T300 carbon/epoxy. [14] Lee S-WR, Sun CT. A quasistatic penetration model for
For E-glass/epoxy, energy absorbed by secondary yarn composite laminates. J Compos Mater 1993;27:251–71.
[15] Ellis RL. Ballistic impact resistance of graphite epoxy composites
deformation and tensile failure of primary yarn are the
with shape memory alloy and extended chain polyethylene spectra
main energy absorbing mechanisms. For T300 carbon/ hybrid components. Master of Science Thesis, Mechanical Engi-
epoxy, the main energy absorbing mechanisms are the neering Department, Virginia Polytechnic Institute and State
secondary yarn deformation and shear plugging. University, December 1996.
590 N.K. Naik, P. Shrirao / Composite Structures 66 (2004) 579–590

[16] Reddy BCK, Shrirao P, Naik NK. Ballistic impact on brittle
composites: shear-plugging phenomenon. In: Batra RC, Henneke
EG, editors. Proceeding of the 14th US National Congress of
Theoretical and Applied Mechanics. Contemporary Research in
Theoretical and Applied Mechanics. Blacksburg: Virginia Tech;
2002. p. 363.
[17] Naik NK, Reddy BCK, Shrirao P. Analytical modeling of ballistic
impact behavior of woven fabric composites. In: Ko FK (Ed.),
Proceedings of TEXCOMP - 6, Drexel University, paper 6.08,
2002.
[18] Smith JC, McCrackin FL, Schiefer HF. Stress–strain relationships
in yarns subjected to rapid impact loading. Part V: Wave
propagation long textile yarns impacted transversely. Textile
Res J 1958;(April):288–302.
[19] Roylance D, Wilde A, Tocci G. Ballistic impact of textile
structures. Textile Res J 1973;(January):34–41.