2007 Aircraft Structural IntegrityProgram ConferenceProgram Conference

The Effect of StressThe Effect of Stress Intensity Factor Models on Inspection Intervalson Inspection Intervals

Lt Col Scott FawazCenter for Aircraft Structural Life Extension

United States Air Force Academy

I n t e g r i t y - S e r v i c e - E x c e l l e n c e

y

DISTRIBUTION STATEMENT A: Approved for public release: distribution is unlimited

Acknowledgements

• Dr. Börje Andersson - Swedish Defense Research AgencyResearch Agency

• Daniel Hill – CAStLE Research Engineer• Dr. R. A. Saravanan – CAStLE Metallurgist• Jim Harter – AFGROW Lead Engineerg• Alex Litvinov – AFGROW Software Engineer

Outline• K Solutions

° Geometric & Loading Parameter SpaceGeometric & Loading Parameter Space° Verification° Validation

F ti Lif P di ti U i N K• Fatigue Life Predictions Using New KSolutions° Fatigue Life° Fatigue Life° Continuing Damage Scenario

› Phase I LifeC k Si› Crack Size

° Effect of r/t

• ConclusionsConclusions

Small differences in K SolutionsSmall differences in K Solutionsyield large cumulative differences

i f i lifin fatigue life

…and large differences in K solutions yield even a larger cumulative difference in fatigue life

Parameter SpaceK-Solutions, ≈ 1.0 million CPU Hours

• Geometry° Centrally Located Straight Shank Hole° 0.1 ≤ r/t ≤ 10.0

σbendingσbypass

› 0.1, 0.111, 0.125, 0.1428, 0.1667, 0.2, 0.25, 0.333, 0.5, 0.667, 0.75, 0.8, 1.0, 1.25, 1.333, 1.5, 1.667 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0 (r/t = 0.5, 1.0)

° Finite Width/Height Plate/h 0 0025

Pcos2θ

› r/h = 0.0025› r/b = 0.0025

• Crack Shapes° 0.1 ≤ a/c ≤ 10.0

› 0 1 0 111 0 125 0 1428 0 1667 0 2 c1

2h

2rc2› 0.1, 0.111, 0.125, 0.1428, 0.1667, 0.2, 0.25, 0.333, 0.5, 0.667, 0.75, 0.8, 1.0, 1.25, 1.333, 1.5, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0 (a/c = 0.2, 0.5, 0.8, 1.0, 2.0)

° 0.1 ≤ a/t ≤ 0.99› 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9,

2b

12

σ, , , , , , , , ,0.95, 0.99 (a/t = 0.2, 0.5, 0.8)

• Load Conditions° Tension° Bending

σo

a

σbending

g° Pin Loading (Bearing)

• 5,672,700 solutionsa2

a1

K S l tiK-SolutionVerificationVerification

Convergence: Shallow Crack1.0

/t 1 0

   σo

0.8

r/t = 1.0a/c = 0.1a/t = 0.12b = 2h = 200r = 400

2b

2h

0.6

c2r

σoa

0 2

0.4

p = 2

p = 3

0.0

0.2 p = 4

p = 5

p = 60.0

0.0 0.2 0.4 0.6 0.8 1.0

Convergence: Deep Crack1.2

p = 2

3r/t = 1.0a/c = 0 1

 

0 8

1.0p = 3

p = 4

p = 5

p = 6

a/c = 0.1a/t = 0.992b = 2h = 200r = 400

0.6

0.8 p

σo

0.4

2b

2h

0 0

0.2

c2r

σo

a

0.00.0 0.2 0.4 0.6 0.8 1.0

K S l tiK-SolutionValidationValidation

Test Specimen Configurationσ

WW

Crack plane

ta

σa1

a2

c1 c2

Marker Load Spectrum

100 cycles 100 cycles 100 cycles

10 cycles 10 cycles 10 cycles

stre

ss 2000Cycles

2000Cycles

2000Cycles repeat2000

Cycles

1 program = 8170 cycles

Fatigue Life Prediction

3.5Test 3, 7075-T651w=49.17 mm, t=6.36 mm, r=3.23 mmc1=2.76 mm, c2=1.75 mm

3

)

c1 2.76 mm, c2 1.75 mm

2.5

k Le

ngth

(mm

)

AFGROW Lef t Crack

Measured Lef t Crack

2

Cra

ck Measured Lef t Crack

AFGROW Right Crack

Measured Right Crack

1.562,000 64,500 67,000 69,500 72,000 74,500

Cycles

Crack Shape Development

Crack Shape Development

Crack Shape Development

F ti Lif P di ti U iFatigue Life Predictions Using New K Solutions

Geometry for Assessing Effect on LifeSmall Crack – Thin Sheet

W = 1.14 in, t = 0.063 in, D = 3/16 in

a 0 01 in c 0 01 in a /t 0 2

σo σbending

ai = 0.01 in, ci = 0.01 in, ai /t = 0.2

ai /ci = 1.0, r/t = 1.5

TSR = 1.0, BSR = 0.4,Small Crack – Thick Sheet

W = 4.53 in, t = 0.25 in, D = ¾ in

0 05 i 0 05 i /t 0 2 2ai = 0.05 in, ci = 0.05 in, ai /t = 0.2

ai /ci = 1.0, r/t = 1.5

TSR = 1 0 BSR = 0 4W

c12rc2

Large Crack – Thin Sheet

W = 1.14 in, t = 0.063 in, D = 3/16 in

0 05 i 0 05 i /t 0 8

TSR 1.0, BSR 0.4

σoσbending

ai = 0.05 in, ci = 0.05 in, ai /t = 0.8

ai /ci = 1.0, r/t = 1.5

TSR = 1.0, BSR = 0.4

a2 a1t

Effect on Life – Small Crack, Thin Sheet60 Two Symmetric Corner Cracks

Single Corner CrackW = 1.14 in, t = 0.063 in, D = 3/16 inai = 0.01 in, ci = 0.01 inai /t = 0.2, ai /ci = 1.0, r/t = 1.5TSR = 1 0 BSR = 0 4

40

50

(%)

TSR = 1.0, BSR = 0.4

30

20

10

0CA 10 CA 18 Falstaff 10 Falstaff 18 TWIST 10 TWIST 18 Marker 10 Marker 18

Spectrum Type and Maximum Stress Level

Effect on Life – Small Crack, Thick Sheet

45

50Two Symmetric Corner CracksSingle Corner Crack

W = 4.53 in, t = 0.25 in, D = 3/4 inai = 0.05 in, ci = 0.05 inai /t = 0.2, ai /ci = 1.0, r/t = 1.5TSR = 1.0, BSR = 0.4

35

40

(%)

TSR 1.0, BSR 0.4

20

25

30

10

15

20

0

5

CA 10 CA 18 Falstaff 10 Falstaff 18 TWIST 10 Marker 10Potential for initial inspection of damage tolerant (rogue flaw)

i l h i h i f lif-5

CA 10 CA 18 Falstaff 10 Falstaff 18 TWIST 10 Marker 10

Spectrum Type and Maximum Stress Levelnot occurring early enough in the aircraft life

Effect on Life – Large Crack, Thin Sheet300

Two Symmetric Corner CracksSingle Corner Crack

W = 1.14 in, t = 0.063 in, D = 3/16 in

200

250

(%)

ai = 0.05 in, ci = 0.05 inai /t = 0.8, ai /ci = 1.0, r/t = 1.5TSR = 1.0, BSR = 0.4

150

100

50

Potential for initial inspection of damage tolerant (rogue flaw) i l h i h i f lif

0CA 10 CA 18 Falstaff 10 Falstaff 18 TWIST 10 TWIST 18 Marker 10 Marker 18

Spectrum Type and Maximum Stress Level

not occurring early enough in the aircraft life

Geometry for Assessing Effect on Continuing Damage Scenario σg σo σbending

W = 1.14 in, t = 0.063 in, D = 3/16 in

0 05 i 0 05 ia1 = 0.05 in, c1 = 0.05 in

a2 = 0.005 in, c2 = 0.005 in

a1 /t = 0.8, a2 /t = 0.08, ai /ci = 1.0, r/t = 1.5 2a1 /t 0.8, a2 /t 0.08, ai /ci 1.0, r/t 1.5

TSR = 1.0, BSR = 0.4W

c12rc2

σo

σbending

a2 a1 t

Effect on Continuing Damage ScenarioPhase I Life

350

400W = 1.14 in, t = 0.063 in, D = 3/16 ina1 = 0.05 in, c1 = 0.05 ina2 = 0.005 in, c2 = 0.005 ina /t = 0 8 a /t = 0 08 a /c = 1 0 r/t = 1 5

300

(%)

a1 /t = 0.8, a2 /t = 0.08, ai /ci = 1.0, r/t = 1.5TSR = 1.0, BSR = 0.4

200

250

100

150

50

100

0CA 10 CA 18 Falstaff 10 Falstaff 18 TWIST 10 TWIST 18 Marker 10 Marker 18

Spectrum Type and Maximum Stress Level

Effect on Continuing Damage ScenarioPhase I Crack Lengthg

60a-Crack Tip

c-Crack TipW = 1.14 in, t = 0.063 in, D = 3/16 ina1 = 0.05 in, c1 = 0.05 ina2 = 0.005 in, c2 = 0.005 ina /t = 0 8 a /t = 0 08 a /c = 1 0 r/t = 1 5

40

50

(%)

a1 /t = 0.8, a2 /t = 0.08, ai /ci = 1.0, r/t = 1.5TSR = 1.0, BSR = 0.4

30

20

10

0CA 10 CA 18 Falstaff 18 TWIST 10 TWIST 18 Marker 10 Marker 18

Spectrum Type and Maximum Stress Level

Effect of r/t – Symmetric Corner Cracks

35

40 Symmetric Corner CracksW/D = 100, ai/ci = 1.0ai = ci = 0.05 mm"Thin Sheet" → a/t = 0.8

30F/A Thin SheetN/R Thin SheetF/A Thick Sheet

"Thick Sheet" → a/t = 0.2

20

25

Nr/t

= 3

.0

(%) N/R Thick Sheet

10

15Ni/

N

5

10

00 0.5 1 1.5 2 2.5 3

r/t

Effect of r/t – Single Corner Crack

90

100 Single Corner CrackW/D = 100, ai/ci = 1.0ai = ci = 0.05 in"Thin Sheet" → a/t = 0.8

70

80F/A Thin SheetN/R Thin SheetF/A Thick Sheet

"Thick Sheet" → a/t = 0.2

50

60

Nr/t

= 3

.0

(%) N/R Thick Sheet

30

40Ni/

N

10

20

00 0.5 1 1.5 2 2.5 3

r/t

Conclusions

• Verification° hp-version FEA + Splitting Scheme = Accurate K-

Solutions• Validation• Validation

° Fatigue life predictions are slightly conservative• 5 672 700 K solutions for unsymmetric corner5,672,700 K solutions for unsymmetric corner

cracks at a hole subject to tension, bending, bearing° Solutions available in tabular form – currently in

AFGROW› 75 – 1.5MB ASCII files

° Source code for multi-dimensional interpolation also available

Significance• Single vs. Double Cracks

° Difference always larger for single cracksy g g

• Effect on Fatigue Life° Small cracks in thin sheets: 20-50%

S ll k i thi k h t 25 45%° Small cracks in thick sheets: 25-45%° Large cracks in thin sheets: 90-300%° Continuing damage scenario: 125-350%

• Effect on Inspections° Possibility of initial inspection not early enough in aircraft life° Possibility of recurring inspections not occurring as frequently as° Possibility of recurring inspections not occurring as frequently as

required

• Effect of r/t° Significant for large cracks in thin sheets° Negligible for small cracks in thick sheets