Optimal Design of a Composite Scarf Repair Under Uniaxial Tension Loading

T.D. Breitzman, US Air Force Research LaboratoryE.V. Iarve, University of Dayton research Institute, USG.A. Schoeppner, US Air Force Research LaboratoryB.M. Cook, US Air Force Research Laboratory

Multiscale Modeling of Composites WorkshopCleveland, OH, US

23-24 July, 2009

MotivationMotivation

• Composite Repair Needs are Increasing• Scarf Repair

– repair technology– adhesive properties– patch geometry– special requirements– patch composition

(ply-by-ply replacement?)

• Specimen preparation• Basic material properties• Experimental results• Modeling approach• Optimization• Conclusion

Outline

Large Tensile Panels – Manufacture

• Cut coupons from panels– Bond tabs (EPON 828)– Drill holes for grip interface

• Scarf portion of coupons– Scarfomatic– Manufacture vacuum table jig– Fairly repeatable

Panel Outside Dia (in) Inside Dia (in)

419T 2.74 to 2.83 1.10 to 1.12

420B 2.68 to 2.75 1.06 to 1.08

423T 2.60 to 2.69 0.98 to 1.00

419T 2.74 to 2.83 1.10 to 1.12

420B 2.68 to 2.75 1.06 to 1.08

423T 2.60 to 2.69 0.98 to 1.00

418T 2.69 to 2.79 0.94 to 0.96

419B 2.67 to 2.78 0.85 to 0.87

422B 2.8 1.1

423B 2.62 to 2.74 1.00 to 1.05

Half scarf angle = 870±0.26 (870-nominal)(pool of 40 angle data points)

Tensile Testing – Scarfed Laminate

IM6/3501-6[45/0/-45/90]s

Tool Box

B-Spline Analysis Method• Variable defect B-spline approximation

in 3-D, (p-element–particular case.• Multibasis global-local approximations

for laminated structures• General anisotropy, including

continuous point-wise variability• Cluster – BSAM “solid element”• Cluster geometries are restricted to

being parametric

1

1

q

f

qc

fcsmq3 smq2 smq1 sm -max. stress

concentration

Critical Failure Volume (CFV)Fiber Failure

Mesh Independent Damage Model

0

1000

2000

3000

4000

5000

6000

0 0.02 0.04 0.06 0.08 0.1 0.12 0.14

Displacement (in)

Lo

ad

(lb

f)

L-1 L-3 L-4 L-5 L-6 L-7 L-8 R-1 R-2

R-3 R-4 R-5 R-6 R-7 R-8 R-9

eult=1.2%

Basic material properties

Fiber direction strength

Xt=eultE11= 1.82GPaV0=0.78mm3

Weibull modulusa=40 (Wisnom (2005))

Stiffness properties

composite

0

10

20

30

40

50

60

0 0.1 0.2 0.3 0.4 0.5 0.6

ExperimentalSpline interpolation

Sh

ear

Str

es

s (

Mp

a)

Strain

0.38 ),( vmfG

FM300M

adhesive

Manufacturer data, KGR-1 instrumentation

Tensile testing2 - ¾”x4” specimens with tabs

Poisson’s ratioStrain gage comparison of x to y

Tensile Testing – Scarfed Laminate

0

0.2

0.4

0.6

0.8

1

1.2

No

rm

alized

av

era

ge

te

nsile

str

en

gth

SmallCoupons

Use

d fo

r Cal

ibra

tion

CFV WI

ScarfCoupons

IM6/3501-6; [45/0/-45/90]s

Ply-by-ply 3D analysis (BSAM)

Repaired Specimen Testing

Large virgin panels•Grip failures•Some gage section failures•Significant scatter of strength

Repaired panels•Apparent brittle failure•The patch is mainly intact•The overply delaminates

Instrumentation

Flush Repair Analysis

Patch

Adhesive

Scarf

Elastic adhesive

Scarfed panel

Nonlinear Strength Prediction

Stress in fiberdirection

Incremental Nonlinear analysis

Load increment(17.23Mpa)

Equilibration

Predict averagefiber failure load Ff

P>Ff

No

YesSTOP: Ff

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0 0.1 0.2 0.3 0.4 0.5

Av

era

ge

Fib

er

Fa

ilure

Lo

ad

Ff (

GP

a)

Applied Load P (Gpa)

Adhesive failure appears sudden anddetermines the failure of the repair

Adhesive Stress ConcentrationMotivation for Optimal Deisgn Problem

[45/0/-45/90/90/-45/0/45]

Optimal Design Problem

45

0

-45

90

90

-45

0

45

45

0

-45

90

90

-45

0

45

?

?

?

?

?

?

?

?

F=Lp(svm)

p

V

pp dvg

VgL

/1

))((1

))((

xx

Optimization solution minimizes elastic stress field in the adhesive

svm – von Mises stress

p=15

Simplex method used for optimization

Optimization Results – No overply

Case Study: Extension/Bending

Thermal only, ΔT=-150Cux=uy=uz=0 on x=±285.75mm

Thermal + Mechanical, ΔT=-150Cux=±0.05, uy=uz=0 on x=±285.75mm

Case Study: Scarf Angle

0

5

10

15

20

25

30

0 1 2 3 4 5 6 7 8

Ply by ply, 1:10.4Offset, 1:10.4Optimal, 1:10.4Ply by ply, 1:20Offset, 1:20Optimal, 1:20

von

Mis

es S

tres

s (

MP

a)

Normalized z-coordinate

Overply Repair Analysis

Scarf Geometry (2-D cut-away view)

Blue – Overply

Yellow – Adhesive

Red – Patch

Green - Adherend

Model OP1

Model OP2

Case Study: Overply Thickness

45° overply

0° overply

2t 4t 8t

2t 4t 8t

No overply

The overply is “attached” to the top of the adherend without adhesive

Model OP1

Case Study: Overply angle

0

0.5

1

1.5

2

2.5

-90 -60 -30 0 30 60 90

F.C. response to overply fiber angleat 413 MPa (no adhesive on top)

Ply by plyOffsetOptimized with overply

Fa

ilure

cri

teri

on

Overply fiber angleOverply angle is not aligned with loading direction due topeel stress in the adhesive on top

0

0.5

1

1.5

2

2.5

-90 -60 -30 0 30 60 90

F.C. response to overply fiber angleat 413 MPa (adhesive on top)

Ply by plyOffsetOptimized with overply

Fa

ilure

cri

teri

on

Overply fiber angle

Model OP1

Model OP2

Optimized patch & overply angles

Parent Ply by ply OffsetOptimized

Model OP1

Optimized Model OP2

overply -72. -45.1 5.26 -54.7

45.000 45.000 0 11.746 43.809

0.0000 0.0000 -45 -45.282 -37.435

-45.000 -45.000 90 0.11019 -121.78

90.000 90.000 90 91.267 67.234

90.000 90.000 -45 -12.123 15.001

-45.000 -45.000 0 -25.158 -54.138

0.0000 0.0000 45 48.474 44.783

45.000 45.000 45 50.417 57.455

Strength Results

No Overply With Overply

0

100

200

300

400

500

600

700

800

400 500 600 700 800

Applied LoadPly by Ply, No overplyOffset, No overplyOptimized, No overply

Pre

dic

ted

Fib

er F

ailu

re L

oad

(M

Pa)

Applied Load (MPa)

0

100

200

300

400

500

600

700

800

400 500 600 700 800

Applied LoadPly by ply, with OverplyOffset, with OverplyOptimized, with Overply

Pre

dic

ted

Fib

er F

ailu

re L

oad

(M

Pa)

Applied Load (MPa)

Strength Results

Conclusions

• Developed highly repetitive scarfing and repair procedures for IM6/3501-6 composites by using Scarfomatic device.

• Benchmarked strength prediction of scarfed and repaired panels based on small UNNOTCHED coupon tensile strength calibration– Good agreement with both scarfed and repair strength– Modeling of adhesive failure is critical for repair strength prediction

• Optimization of the patch composition to delay adhesive failure– Understood the stress transfer mechanisms with/- the overply– 20% or more effect on strength due to patch composition

• Softening of the patch– Fabric patch makeup– Stress in the adherend

• Future work– Adhesive characterization– Multidirectional loading– Testing

Case Study: Overply Thickness

0 0.2 0.4 0.6 0.8 1 1.2

No Overply1 Overply2 Overplies4 Overplies8 Overplies

Ad

he

ren

d F

ibe

r O

rie

nta

tio

n

Normalized von Mises Stress

45o

0o

90o

90o

-45o

-45o

0o

45o