N88-22402

IMPACT DAMAGE IN COMPOSITE LAMINATES

Joseph E. GradyStructural Mechanics Branch

NASA Lewis Research Center

ABSTRACT

Damage tolerance requirements have become an important consideration in the

design and fabrication of composite structural components for modern aircraft.

The ability of a component to contain a flaw of a given size without serious

loss of its structural integrity is of prime concern. Composite laminates are

particularly susceptible to damage caused by transverse impact loading.

The ongoing research program described herein is aimed, therefore, at develop-

ing experimental and analytical methods that can be used to assess damage tol-

erance capabilities in composite structures subjected to impulsive loading.

The objective of this presentation is to outline some significant results of

this work and the methodology used to obtain them, including

(I) Identification of the mechanisms that cause delamination damage initi-

ation and growth under impact loading (Grady and Sun, 1986)

(2) Calculation of the dynamic delamination "fracture toughness" in com-

posite laminates (Sun and Grady, 1987)

(3) Computational simulation of dynamic response and damage initiation in

composite laminates (Grady, Chamis, and Aiello, 1987)

(4) Measurement of the effect that impact damage has on the structural

integrity of a laminate (Grady and Chamis, 1988)

2-235

https://ntrs.nasa.gov/search.jsp?R=19880013018 2018-06-15T19:23:14+00:00Z

LEWIS BALLISTIC IMPACT RESEARCH

The unique ballistic impact test facility at the Lewis Research Center is used

for performing a variety of instrumented impact tests. In addition to multiple

channel high-speed digital data acquisition capabilities, a series of high-

speed cameras can be used to record the impact event photographically. The

impact cannon itself is able to accommodate impactors ranging in size from 1/2

to 6 inches in diameter. It has been used to simulate the impact of birds on

engine components and is currently being used to investigate dynamic crack

propagation in composite materials.

Post-impact inspections of damaged specimens can be conducted with any of the

available supporting experimental facilities, which include:

• Ultrasonic C-scan and x-ray inspection facilities, which are used to

detect the type and location of impact damage

• A vibration testing facility used to measure natural frequencies and mode

shapes of damaged specimens

• Static and fatigue load frames to measure the reduction in strength,

stiffness, and durability of test specimens caused by impact

OVERHEAD CAMERA

REMOTETV CAMERA

E "i_ SPECIMEN

FLOODLIGHTS ..... _"......... I_

IMPACT SPECIMEN BALLISTIC IMPACT FACILITY

CD-88-31801

2-236

DYNAMIC FRACTURE TOUGHNESS IN COMPOSITES

A combined experimental and analytical approach was used (Grady and Sun, 1986, 1988) to predict both the dynamic response measured in a simulated engine blade and the dynamic delamination fracture toughness of the composite mate- rial. Impact tests were conducted on cantilevered graphite/epoxy laminates, and a two-dimensional finite element model was developed to computationally simulate the post-impact dynamic response. Correlations between calculated crack-tip strain energy release rates and high-speed photographs of crack initiation and propagation were used to estimate the dynamic fracture toughness of the composite material.

0

4 1 - . I

COMPUTATIONAL SIMULATION

--- RIGHT CRACK TIP 20 - LEFT CRACK TIP - o EXPERIMENT

ENERGY BENDING RELEASE STRAIN, RATE,

P , in. in. Ib

I I

in. G, - 10

- i n2

0 100 200 TIME, ps TIME, p~s

STRAIN RESPONSE AT IMPACT POINT ENERGY RELEASE RATE DUE TO IMPACT CD-88-91802

2-237

DYNAMIC DELAMINATION BUCKLING: COMPUTATIONAL SIMULATION

A unique dynamic delamination buckling and delamination propagation analysis capability has been developed and incorporated into the MSC/NASTRAN finite ele- ment analysis computer program. This capability consists of (1) a modifica- tion of the direct time integration solution sequence which provides a new analysis algorithm that can be used to predict delamination buckling in a lami- nate subjected to dynamic loading, and ( 2 ) a new method of modeling the compos- ite laminate using plate-bending elements and multipoint constraints. these modifications, NASTRAN is used to predict both impact-induced buckling in composite laminates with initial delaminations and the strain energy release rate due to extension of the delamination (Grady et al., 1987). It is shown that delaminations near the outer surface of a laminate are susceptible to local buckling and buckling-induced delamination propagation when the laminate is subjected to transverse impact loading. The capability now exists to pre- dict the time at which the onset of dynamic delamination buckling occurs, the dynamic buckling mode shape, and the dynamic delamination strain energy release rate.

With

HIGH-SPEED PHOTOGRAPHS (22 000 FRAMESISEC)

135 Bsec AFTER IMPACT

180 psec: ONSET OF BUCKLING

225 psec: PROPAGATION

270 psec: CRACK ARREST

COMPUTATIONAL SI M U LATl ON kssa..

FIRST BUCKLING MODE AT 190 psec

$

aA 6 EIGENVALUE #1

ONSET OF LOCAL DELAMINATION BUCKLING7

f /

TIME, psec DYNAMIC BUCKLING ANALYSIS

CD-88-31803

ORICINM PAGE IS OF POOR QUALITY

STIFFNESSLOSSDUETO IMPACTDAMAGE

A series of additional post-impact tests were conducted on the composite impactspecimens to measure the effect of impact damageon the structural integrity ofthe laminates. A decrease in stiffness, such as that caused by impact damage,results in a corresponding decrease in the natural frequencies of a damagedcomposite structure. To quantify this effect for the case of impact-induceddelamination, the first four natural frequencies of a series of cantileveredcomposite specimenswith initial embeddeddelaminations were measuredbeforeand after impact. The measured reduction in the natural frequencies of thefirst four bending modes for varying amounts of impact damageare comparedwiththe results of a two-dimensional plane strain finite element simulation of thedamagedspecimens in the figures below. The difference between the two curvesrepresents the extent to which damageother than delamination has occurred inthe laminates.

1.0

1.0 m

0.5 B

0.5

MODE 1

i 't ] 'i

---41"--- FINITE ELEMENTEXPERIMENT

MODE 2I I I II I I I

MODE 3

I I I2 4 6

MODE 4

I I I I8 0 2 4 6

DELAMINATED AREA, in.2

CD-88-31804

2-239

DETECTINGIMPACTDAMAGE

Composite laminates are particularly susceptible to damagecaused by trans-verse impact. Ply debonding, or delamination, is the most serious type ofimpact damage. Onemethod of detecting delamination in composites is by non-destructive ultrasonic C-Scan. Shownbelow are C-Scans of composite laminatesthat have been impacted at various energy levels, causing different amounts anddistributions of damage. The corresponding effects of this damageon the natu-ral frequencies of vibration are shownbelow. The loss of stiffness caused bydelamination is most apparent in the reduction of natural frequencies for thehigher modesof vibration. Since impact damagesuch as delamination is oftenundetectable visually, either of these methods can serve as a practical meansof field-testing a composite structure to determine if significant amounts ofinternal damageexist.

m !

V = 5232in.Is 5520 5976

ULTRASONICC-SCANSOF IMPACTEDLAMINATES

NATURALFREQUENCY,

kHz

10 [-- BENDING

8 _ MODE

6

1 2 3 4

DELAMINATIONLENGTH, a, in.

CD-88-31805

2-240

EFFECT OF DELAMINATION ON BUCKLING LOADS

Instrumented composite impact specimens were tested in compression after being

impacted at velocities up to i000 ft/s with a soft rubber projectile. Longitu-

dinal strain measurements at three gage locations were used to identify the

buckling loads. Small amounts of impact-induced delamination damage were found

to cause a significant drop in the critical buckling loads, as shown in the

last figure. Compressive longitudinal stress near the delaminated area can

cause a secondary "local" buckling of the delamination and an increased dropin the effective load-carrying capability of the laminate.

GRAPHITE/EPOXY20 PLIESm

_GAGE 3/

/

F---_L---', \ _ _ _--F/

GAGE1J _GAGE 2

5000BENDING 2500¢,_'al'l=_ • i ira

O I RMII1, i

in.

_m. 0

-250O0

-- GAGE1.... GAGE2-,---- GAGE3

I P'il "\ I150 200 250

RIIr.KI INil

LOAD,Ib

LOAD,Ib

TENSILE LOCAL

150

-% "xj

100 --COMPRESSIVEj

_LOCAL STRESS5O

I I I0 1 2 3

DELAMINATIONLENGTH, in.

C D-88-31806

2-241

EFFECTOF IMPACTDAMAGEONPOST-BUCKLINGDISPLACEMENTS

Impact damagewhich is not detectable by visual inspection may cause a signifi-cant loss in the structural integrity of a composite laminate. This is illus-trated below, using post-impact compression tests of damagedgraphite/epoxylaminates. Experimentally measuredpost-buckling configurations at severalload levels show that internal damagecaused by transverse impact can cause alarge "direction-dependent" stiffness loss in composite laminates.

Under uniform compression loading, the laminate buckles in the direction ofthe prior impact load, as shownon the left. The measureddisplacement shapeshows that the effective bending stiffness is significantly reduced from thatof the original undamagedlaminate. If lateral restraints are used to supportthe laminate in this direction, buckling occurs in the opposite direction, andthe laminate showsa muchhigher post-buckllng stiffness at all load levels.

"_12

9

6

3

[ I I0.75 0.50 0.25

DISPLACEMENT,in.

LOAD, FIb

O 150[] 175

POSITION, in. 15 F

/_F 12 --

TI '6300 in./sec

15 in. "_---©6

I J0 0.25 0.50

DISPLACEMENT,in. •

CD-88-31807

2-242

REFERENCES

Awa, V.S., 1983, "Effect of Specimen Size on the Buckling Behavior of Lam-

inated Composites Subjected to Low-Velocity Impact." Compression Testing

of Homogeneous Materials and Composites, ASTM STP 808, Richard Cait and

Ralph Papirno, eds., American Society for Testing and Materials,pp. 140-154.

Grady, J.E., and Chamis, C.C., 1988, "Effect of Impact Damage on Structural

Response of Composite Laminates." To be presented at the Fourth Japan-

U.S. Conference on Composite Materials, Washington, D.C.

Grady, J.E., Chamis, C.C., and Aiello, R.A., 1987, "Dynamic Delamination Buck-

ling in Composite Laminates Under Impact Loading: Computational Simula-

tion." Composite Materials: Fatigue and Fracture, ASTM STP, P.O. Lagace,

ed., American Society for Testing and Materials. Also published as NASATM 100192.

Grady, J.E., and DePaola, K.J., 1987, "Measurement of Impact-lnduced Delamina-

tion Buckling in Composite Laminates." Proceedings of the 1987 SEM Fall

Conference on Dynamic Failure, Society for Experimental Mechanics.

Grady, J.E., and Sun, C.T., 1986, "Dynamic Delamination Crack Propagation in a

Graphite/Epoxy Laminate." Composite Materials: Fatigue and Fracture,

ASTM STP 907, H.T. Hahn, ed., American Society for Testing and Materials,

Philadelphia, pp. 5-31.

Joshi, S.P., and Sun, C.T., 1987, "Impact-lnduced Fracture in a Quasi-lsotropic

Laminate." Journal of Composites Technology and Research, Vol. 9, No. 2,

pp. 40-46.

Sierakowski, R.L., Ross, C.A., and Malvern, L.E., 1981, "Studies on the Frac-

ture Mechanisms in Partially Penetrated Filament Reinforced Laminate

Fi_Les. rlnai Report DAaG2_-7_-G-O007, U. _. Army Kesearcn u_llce.

Sun, C.T., and Grady, J.E., 1988, "Dynamic Delamination Fracture Toughness of

a Graphite/Epoxy Laminate Under Impact." To appear in ASTM Journal of Com-

posites Technology and Research.

Whitcomb, J.D., and Shivakumar, K.N., 1987, "Strain-Energy Release Rate Analy-sis of a Laminate with a Post-Buckled Delamination." NASA TM 89091.

2-243