Durability & Damage Tolerance Testing and Analysis Protocols for Composite Structures Life Factor, Load-Enhancement Factors, and Fatigue Life

Transcription

1 Durability & Damage Tolerance Testing and Analysis Protocols for Composite Structures Life Factor, Load-Enhancement Factors, and Fatigue Life Waruna P. Seneviratne, PhD John S. Tomblin, PhD National Institute t for Aviation Research Wichita, KS FAA Joint Advanced Materials and Structures (JAMS) 5th Annual Technical Review Meeting July 21-22, 2009 Wichita, Kansas

2 FAA Sponsored Project Information National Institute for Aviation Research John Tomblin, PhD (Executive Director) Waruna Seneviratne, PhD (Research Scientist) Federal Aviation Administration Curtis Davies Program Manager FAA William J. Hughes Technical Center, NJ Larry Ilcewicz, PhD FAA Chief Scientific and Technical Advisor for Composite Materials FAA/Seattle Aircraft Cert. Office Peter Shyprykevich Consultant (Ret. FAA). Workshops for Composite Damage Tolerance & Maintenance 2009 FAA/CACRC/EASA Tokyo, Japan 2008 AIRBUS Toulouse, France 2008 CMH-17: Cocoa Beach, FL and Ottawa, Canada 2007 FAA/CACRC/EASA - Amsterdam, Netherlands 2006 FAA Workshop - Chicago, IL

3 Research Program Objectives Primary Objective Develop a probabilistic approach to synthesize life factor, load factor and damage in composites to determine fatigue life of a damage tolerant aircraft Secondary Objectives Extend the current certification approach to explore extremely improbable bl high h energy impact threats, t i.e. damages that t reduce residual strength of aircraft to limit load capability Investigate realistic service damage scenarios Inspection & repair procedures suitable for field practice Incorporating certain design changes into full-scale substantiation without the burden of additional time-consuming and costly tests

4 Scatter Analysis of Composite Static Scatter Fatigue Scatter Life Factor Load Enhancement Factors

5 Load-Enhancement Factor Approach Increase applied loads in fatigue tests so that the same level of reliability can be achieved with a shorter test duration N F α L + 1 Γ α L 1 α L ln( p) 2 χ γ ( 2n ) 2n Load (Scatter) Factor α R + 1 Γ α λ R LF = ln( R) 2 χ γ (2n) 2n λ 1 α Load Enhancement Factor (LEF) α L αr α L + 1 Γ α L LEF( N ) = α L ln( R) N χ 2 γ (2n ) 2n R 1 α R

6 Material Databases LEF» AS4/E7K8 (457/17)» T700/#2510 PW (240/7)» 7781/#2510 8HS (204/7) Laminate Data» T700/#2510 UNI (853/47)» T700/#2510 PW (863/48)» T700/E765 UNI (834/47)» T300/E765 PW (722/48)» AS4C/MTM45 UNI (1151/86)» AS4C/MTM45 8HS (1083/78) Adhesive Fatigue (390 spec./12 data sets)» Loctite Paste» PTM&W paste (2 bondline thicknesses)» EA 9696 film Adhesive Effects of Defects» T700/#2510 PW & EA9394 PFS (70/6)» T700/ PW & EA9394 (SLS) (20)» T800/ UNI & EA9394 (SLS) (20)» 7781/NB321 8HS & EA9394 (SLS) (20)

7 Sample S-N Curves (contd..) OH (ult) % of /#2510 FG Ult. (OH) OHC ksi OHT ksi R = 0 R = 5 R = -0.2 R = Cycles

8 Fatigue Scatter Analysis Techniques Fatigue Life Weibull Shape Param meter Individual Weibull ni Joint Weibull α xij ln( xij ) M n i α α xij j= 1 fi 1 1 M M j= j= i= 1 i 1 = n ln( xij ) = 0 n fi Sedeckyj Equivalent Strength Model 1 S 300 S σ r σ e = σ a + N f C ( 1) 250 a σ 200 Sendeckyj (w/ static) Sendeckyj (w/o static) Individual Weibull Joint Weibull Data Pooling Techniques Stress (ksi) NADC Fatigue Scatter Analysis α I > α J > α S Equivalent Static Strength Data Cycles, N AS4/E7K8 PW 0.0 OH (R=-1) OHC (R=5) OHT (R=0) OH (R=-0.2) CAI- BVID (R=5) CAI - VID (R=5) DNC (R=-1) DNC (R=-0.2) OH (R=-1) TAI - BVID (R=0) TAI - VID (R=0) Flex (R=0) OH (R=-1) CAI - BVID (R=5)# CAI - VID (R=5)# CAI - LID (R=5)# CAI - VID (R=5) 10/80/10 0/100/0 HRH 10 Sandwich 25/50/25 40/20/40

9 Adhesive Fatigue Scatter Aerodynamic (side) loads Sendeckyj (w/ static) Sendeckyj (w/o static) Shear flow Vertical load Disbond between fuselage halves Fatigue Life Shape Pa arameter Individual Weibull Joint Weibull CTD RTD RTW CTD RTD RTW CTD RTD RTW CTD RTD RTW Loctite EA9696 PTM&W (0.06") PTM&W (0.16")

10 2 nd Generation Weibull Shape Parameters αsendeckyj (w/static) αsendeckyj (w/o static) αindividual Weibull (w/o static) αjointweibull (w/o static) Pooled - Composite MLE α β α Modal RRX α β α Modal RRY α β α Modal Pooled - Composite+Adhesive i MLE α β α Modal RRX α β α Modal RRY α β α Modal Static Scatter Factor Fatigue Scatter Factor NF # of Lives (N) NAVY NIAR Maximum Likelihood Estimation NAVY NIAR NAVY NIAR AS4/E7K8 - C AS4/E7K8 - C+A T700/#2510 -C T700/#2510 -C+A 1.30 AS4 ~ Individual (C) AS4 ~ Individual (C+A) AS4 ~ Sendeckyj (C) /#2510 -C 7781/#2510 -C+A AS4 ~ Sendeckyj (C+A) AS4 ~ Sendeckyj (C) +Individual (A) LEF LEF Test Duration, N Test Duration, N

11 Test Spectrum Generation Life factor Load-enhancement factor Truncation levels Environmental factors % of CAI/TA (ult) AS4/E7K8 PW 10/80/10 -- Ult. (CAI) 20 plies (500) ksi 0/100/0 -- Ult (TAI) 20 plies (500) ksi 20 plies (1500) ksi Cycles

12 Application of LEF/N F Hybrid Structural Substantiation Metals: severe flight loads result in crack-growth retardation Composites: severe flight loads significantly contribute to flaw growth in composite structures and reduce the fatigue life Exceedances per 1000 flight hours High frequency, low loads Truncation level(s) (a) Combined load-life test (b) Combined load-life spectrum Spread the high load cycles throughout the spectrum (may require crack growth analysis for hybrid structures) Must preserve the stress ratios Low frequency, high compressive loads Clipping level for compressive loads (metals) Clipping level for tensile loads (metals) Low frequency, high tensile loads Compressive Tensile Load level

13 Load-Life-Damage Hybrid Approach LLD Overview Load-Life Shift Application of LLD

14 1.E Stress (psi) LLD Hybrid Approach Complete Durability phase Durability phase Initial LEF based on data scatter High energy impact on test article Durability & Damage Tolerance Phase Details of damaged area & critical load conditions New LEF based on DaDT Element data scatter 4.0 ARAMIS out-of-plane displacement 3.0 Scaled elements 1.0 Damage Length/Height (in) Number of Cycles DaDT element tests Width Height Scatter analysis of DaDT element tests Number of Cycles

15 Load-Life Shift 2 = 1 N Total N 1 Example calculation of desired Test Duration: T T N R No Damage (LEF=1.033) LID (LEF=1.014) Desired Test Desired Test N R 2 N 1 Load-Life Shift T N 1 R N 1 AS4-PW BVID VID T N 2 R N 2 LID 1.16 LEF Load Test Duration (N) Life Damage

16 Composite Test Issues Progressive Failure Flow Growth Fatigue & Damage Tolerance

17 Damage-Tolerance Element Tests Scatter analysis or flaw growth threshold Scaling Primary load path (LC) Load redistribution (SC) Flaw-growth measurements Compliance change Stable or critical growth Loading mode Stress ratio SL3 (60%) Load Load Pmax ΔP C1 C1 C2 C2 Pmin +Δε Strain εmin εmax Strain

18 Effects of Damage on α Damage Tolerance Element Tests Data scatter associated with final failure is conservative or representative of scatter at onset of damage propagation Life Shape Parameter BVID VID LID Unimpacted 0.0 With Static Data Without Static Data Individual Joint Sendeckyj Analysis Weibull Analysis ress (psi) Str AS4/E7K8 PW - 25/50/25 CAI (R = 5) BVID, VID, & LID Cycles (n)

19 Full-Scale Validation Static Damage Tolerance ST001, ST002, ST003 Static Tests (CAT1) NDI Element Full-scale Impact Tests BDLL NRLL CAT2 & CAT3 ST001(R) Static Test (CAT2) ST004 Fatigue Test (CAT2) ST004(R) Fatigue Test (CAT3 2) ST005 Static Test (CAT3) ST006 Fatigue Test (CAT3) Repair Durability Load- Life Shift LLD Hybrid Approach

20 CAT2 Aft Spar (FWS 45) ST004 DaDT (Impact Damage) INBD Top skin FWD INBD INDB Aft spar web FWS 44 AFT 1.5 DLT P O N 0 0 N M 0 0 M L ### ### L K ### ### ### ### K J ### ### ### ### ### ### J I ### ### ### ### I H ### ### ### ### H G ### ### ### ### 0 G F ### ### ### 0 F E ### ### ### E D D C C B B A ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### 0 A a ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### 0 a b b REAR SPAR c c INBD d ### d e ### ### ### e f ### ### ### f g ### ### ### ### ### g h ### ### ### ### ### h i ### ### ### ### ### i j ### ### ### ### ### j k ### ### ### k l ### ### l m ### ### m n ### ### n o o p p P O Inspections after 1.5 DLT REAR SPAR OUTBD FWS 44 FWD

21 CAT3 Front Spar (FWS 65) CAT3 INBD ST005 Static FWD FWD INDB Damages to front spar (web + flanges) INDB INBD AFT Front spar web Secondary web INDB FWD CAT3 INBD FWD INDB Skin delamination Skin delamination Bottom-aft flange disbond Front Web INDB ST006 DaDT Bottom-aft flange disbond AFT UP

22 Static Summary ST icrostrain. Compressive M Strain Gages ARAMIS 0 A2 A3 R7A R8A A6 A8 A4 A5 A7 Gage Number Test Article Wing Damage Category Onset of Damage Propagation - Local Load (lbf) First Sign of Fracture - Global Fracture ST001 Right CAT unloaded ST002 Left CAT ST003 Right CAT ST001(R) Right CAT

23 CAT 2 Residual Strength Damage progression along aft spar (top skin) of ST004 (CAT2 damage) during residual strength test after 2-DLT cyclic test with LEF Large Damage growth across CAT impact damage occur just pass ultimate load (NRUL)

24 CAT 3 Residual Strength FWS (in) NRLL micros train Leading edge buckling A1 A3 A5 A7 A12 A2 A4 A6 A8 Skin delamination along aft spar (starting from root A4, A5, and A7) (microstrain) Axial Strain microstrain) Axial Strain ( % 60.0% 70.0% 75.0% 80.0% 85.0% 90.0% 92.0% 94.0% 95.5% Load Redistribution FWS (in) NRLL 50.0% 60.0% 70.0% 75.0% 80.0% 85.0% 90.0% 92.0% 94.0% 95.5% Damage Propagation Fracture across CAT 3 damage just before limit load during residual strength after ½ DLT fatigue test Load (lbf)

25 Summary Load Enhancement Factor 2 nd Generation Weibull shape parameters Static Strength: Fatigue Life: Address evolution/maturity of material systems, manufacturing processes, test techniques, etc. Reduced test matrix and Shared database concept Integrate design specific details gained from coupon and subelement tests into the LEF approach Layup, loading modes/r ratios, Environments,.. Bonded joints, interlaminar shear, sandwich,.. Realistic analysis approach for scatter Appropriate analysis techniques for diverse design details User-friendly automated procedures Notch effects on scatter for damage tolerance testing Adhesive scatter is a concern (reliability!!!) Application of LEF Hybrid structures LEF NAVY NIAR Test Duration, N Number of Cycles 1.E Stress (psi)

26 Summary Load-Life-Damage Approach Incorporation of damage into scatter analysis Investigate large VID damage Scaling Detectability Load-Life Shift Investigate different categories of damages/repairs in the same full-scale test article damage Design change substantiation, i.e. gross weight increase LEF during certification vs. improved LEF Life extension or determination of retirement life Damage Threats and Inspections Load Probability of threats/occurrences Probability of detectability Mitigate risks of unintentional failure INDB FWD Load Inspection intervals using CFU model (cost and reliability) Life Strategic placement of health monitoring equipments Progressive damage analysis (NLFEA) or scaled component tests Damage

27 Questions/Notes