CRASHWORTHINESS OF COMPOSITE FUSELAGE STRUCTURES -Material Dynamic

Properties

K.S. RajuDepartment of Aerospace Engineering

Wichita State University

Program Monitor : Allan AbramowitzFAA William J. Hughes Technical Center

The Fifth Triennial International Aircraft Fire and Cabin Safety Research Conference, Oct 29- Nov1, 2007, Atlantic City, NJ

Background– Crashworthiness

• Maintain survivable volume• Alleviate occupant loads

– Energy Absorption• Metals – Plastic deformation• Composites – controlled failure modes

– Factors affecting energy absorption• Geometry (?)• Strain rate (?)

Energy Absorption

Pmax

EMaintain survivable volume

Energy Absorption in Composites

• Maximize energy dissipated due to formation of surfaces

– Matrix cracks, delaminations, fiber-matrix interface failure, fiber fractures

– Maximize the damage volume {multiple locations of damage}

• Minimize reduction /loss of load path

– Provision for sustained stability Stroke LS

Pmax

Pcrush

EForc

e P

∫=SL

PdE0

δ

Maximization bounded by “safe loads” that can be transmitted Maximization bounded by “stability” of the

transmitting structure/device and availability of crush space

Energy Absorption Devices

Hull D (1991) Comp. Sci Tech, 40.Bannerman & Kindervater (1984) in Structural Impact and CrashworthinessBolukbasi & Laananen (1995) Composites, 26.Carruthers, Kettle & Robinson (1998) Appl Mech Rev, 51.

-Rate effects are evident ( + or -) but in terms of stroke rate

-Strain rates not reported /measured (?)

-Strain rate not proportional to stroke rate due to complexity of localized deformation & Strain gradients

APPROACH

MATERIAL PROPERTY CHARACTERIZATION AT DIFFERENT STRAIN RATES - TENSION, COMPRESSION & SHEAR - CONSTITUTIVE LAWS WITH STRAIN RATE-EFFECTS - FAILURE MODES & STRENGTHS

STRAIN & STRAIN RATE GRADIENTS - OPEN-HOLE TENSION TESTS - BENCHMARK DATA FOR ANALYTICAL MODELS & FAILURE THEORIES

ENERGY ABSORPTION MECHANISMS - constant stroke rate tests - drop tests

SCALED FUSELAGE TUBES - drop tests

FUSELAGE + ENERGY ABSORPTION DEVICES - ASSEMBLIES

Phase-I

Phase-II & III

Phase-IV

• Phase-I– In-Plane Tensile Properties (Strength & Modulus)

• [0°]n, [±15°]ns,[±30°]ns, [± 45°]ns

– In-Plane Compressive Properties (Strength & Modulus)• [0°]n, [±15°]ns,[±30°]ns, [± 45°]ns

– In-Plane Shear Properties (Strength & Modulus)• [0°/90 °]ns

• PHASE-II– Open- Hole Tension– Pin Bearing– Interlaminar Shear– Flexure– Scaled Fuselage sections

• PHASE-III- Fracture Toughness

Tasks..

• PHASE-IV- EA device

Material Systems• Material systems selected by FAA/industry participants

– State of Kansas ( NIAR.INDUSTRY.STATE) funding • Boeing, Spirit Aerosystems, Hawker BeechCraft (Raytheon), Cessna

– FAA personnel involved• Allan Abramowitz, Curtis Davies, J. Zvanya

– Wide range of materials

• Material Systems – Newport NB321/3k70P– Newport NB321/7781 Fiberglass*– Newport NCT321/G150 Unitape

– Toray T800S/3900-2B[P2352W-19] BMS8-276 Rev-H- Unitape*– Toray T700G-12K-50C/3900-2 Plain Weave Carbon Fabric*

– Fibercote E-765/PW Carbon Fabric /Epoxy– Cytec PWC T300 3KNT Plain Weave Carbon Fabric*

• Material systems for phase-II and beyond limited to Newport & Toray systems

Tensile Properties … Quasi-Static Loading

Strength Modulus

TEST APPARATUS

• TENSION MODE TESING• Tension, shear and flexure tests

• MTS High Stroke Rate System (MTS-HSRS)– Stroke rate ~ 500 in/sec – +/- 7 inches stroke– Load capacity

• 5 kips @ rated speed• 9 kips maximum

– Load measurement• Piezoelectric load cells

– +/- 0.5kip, +/- 1kip, +/- 10kips– National Instruments PCI 6110 DAQ

• 4 Channels• 5 MHz (simultaneous sampling)• 12 bits resolution

– Test control • MTS MultipurposeTestware computer program

TEST APPARATUS

• TENSION MODE TESING• Tension, shear and flexure tests

• SLACK INDUCER MECHANISM

–Allows actuator acceleration to desired speed prior to loading the specimen

• LOW-MASS GRIPS–mechanical wedge grips –2.4 lbs –15 kip capacity

Davis, E.A., Trans. ASME, 60, 1938 Elam, C.F., Proc. Roy. Soc. Lond., A, 165, 1938 ManJoine, M.J., Trans. ASME, 66, 1944 A-21 Milkowitz, J., Trans.ASME, 69, 1947 A-21 Morrison, J.L., Engineer, Lond., 158, 1934

TENSION TEST APPARATUS

• Signal Modulation

Time Vs Load

-1500

-1000

-500

0

500

1000

1500

2000

2500

3000

3500

0.0955 0.096 0.0965 0.097 0.0975

Time

Load

Time Vs Load @Loadcell

Time Vs Load @ Gage

Tension testing – Signal Modulation

• Correction using Experimentally determined Transfer Function

Measure true force using tab gages

Fourier Spectrum

1 10 100 1000 10000 1e+05Frequency

0.0001

0.001

0.01

0.1

1

10

100

1000

10000

Am

plitu

de

10 100 1000 10000Frequqncy (Hz)

0.01

0.1

1

10

100

1000

A

TRU

E

ALO

AD

CEL

L

PIEZO CELLSTRAIN GAGE CELL

Transfer Function

L.D. Mitchell, “ Signal Processing and the FFT Analyzer – A Survey,” IJMA, Jan. 1986

Tension Testing • SPECIMEN GEOMETRY

– 2 inch gage length– 0.5 inch width– Thickness limited by loading capacity

of the testing machine• TEST RATES

– 1x10-4 in/s ( quasi-static)– 1, 10 , 100, 250 and 500 in/s– 3 specimens each

• CONSTANT STROKE RATE TESTS– Based on actuator displacement– Strain rate varies throughout the test

• Variation of strain rate is dependent on slack inducer element (s) characteristics (stiffness and mass)

90°

0°

4.50

1.25

3.25

0.5

0 0.002 0.004 0.006 0.008 0.01 0.012Measured Strain εo (in/in)

0

0.2

0.4

0.6

0.8

1

St

rain

Rat

eM

axim

um S

train

Rat

eMaterial: NB321/3K70PNOMINAL STROKE RATE (in/s)

0.00083110100250500

10-3 100 101 102 103

Nominal Actuator Stroke Rate, in/s

105

106

Failu

re S

treng

th σ

xxULT

, psi

TORAY UNITAPE

FIBER ORIENTATION [0]2

[0]2 corrected[+15/-15]s

[+15/-15]s corrected[+30/-30]2s

[+30/-30]2s corrected[+45/-45]2s

[+45/-45]2s corrected

Tension testing – Results

0 0.005 0.01 0.015 0.02 0.025

Strain ε11 (in/in)

0x100

2x105

4x105

6x105

8x105

106

Stre

ss σ

11 (p

si)

Material :TORAY T800S/3900-2B UNITAPESpecimen Type : [0]2

Nominal Stroke Rate (in/sec)0.001110100250500

• Maximum strain rate achieved ~ 250/s

• Tensile Strengths–Tension strength was observed to

increase with stroke ( strain) rate for all material systems & orientations

–Magnitude of strength increase was dependent on fiber orientation and material system

–Newport NB321/7781 fiberglass material exhibited highest strength increase amongst the materials tested

Not corrected for modulation

Effects of Combined State of Stress

• Strength increase higher for off-axis specimens

0.001 1 10 100 1000Nominal Stroke Rate, in/s

0

1

2

3

Nor

mal

ized

Fai

lure

Stre

ngth

Toray T800S/3900-2B Unitape

FIBER ORIENTATION [0]2

[+15/-15]s

[+30/-30]2s

[+45/-45]2s

IN-PLANE SHEAR STRENGTH

• NEWPORT MATERIAL SYSTEMS– Fabric reinforced systems ( 3k70P and

7781)• Shear strength increases with stroke rate• Failure mode changes at higher rates

– Unitape system• Shear strength increases up to stroke rate of

100in/s, but decrease at 250 and 500 in/s• No change in failure mode

10-3 100 101 102 103

NOMINAL STROKE RATE (in/s)

0

10

20

30

40

50

IN-P

LAN

E S

HE

AR

STR

ENG

TH (p

si)

Newport material systemsNB321/3K70 PWCFNB321/7781 FIBERGLASSNCT 321/G150 UNITAPE

K.S.Raju, S. Dandayudhapani and C.K. Thorbole, AIAA-2006-2258

Wave propagation effects

IN-PLANE SHEAR STRENGTH

• COMPARISON OF SHEAR STRENGTHS

– Fabric reinforced systems • Shear strength increases with stroke rate• Failure mode changes at higher rates

– Unitape system• Shear strength increases up to stroke rate of

250 in/s, but decreases at 500 in/s• No change in failure mode

• Corrections for modulation of load signal – Transfer function ( under progress)

10-3 100 101 102 103

NOMINAL STROKE RATE (in/s)

0

1

2

3

4

In

-Pla

ne S

hear

Stre

ngth

Qua

si-S

tatic

In-P

lane

shea

r stre

ngth

Material systems

Newport NB321/3k70 PWCFNewport NB321/7781 fiberglassNewport NCT321/G150 UnitapeToray T800S/3900-2B UnitapeToray T700G-12K-50C/3900-2 PWCFCytec PWC T300 3KNT Fibercote E-765/PW CF

Wave propagation effects

COMPRESSION TEST APPARATUS

• COMPRESSION MODE TESTING– Split- Hopkinson Pressure Bar (SHPB)

• Compressive strength– MTS servo hydraulic testing machine

• Compression at slow to medium rates

LOAD CELL

Steel Platen

Actuator

Specimen

δ

INCIDENT BAR TRANSM ITTER BARPROJECTILE

BRIDGEAM PLIFIER

OSCILLOSCOPE

BRIDGEAM PLIFIER

SPECIM EN

STRAIN GAGE STRAIN GAGE

ε (t), ε (t), ε (t)I R T

INCIDENT W AVE (I)

REFLECTED W AVE (R)TRANSM ITTED W AVE (T)

• SHPB SPECIFICATIONS– Bar diameter : 1.00 inch– Bar Material : Vascomax 350 – Bar lengths

• Incident bar : 48”• Transmitter bar : 36”

– Barrell length : 48 inches– Projectiles : 1”, 2”, 4”, 6” & 12”

• Pneumatically driven (100 psi)• Velocities ~ 2000 in/s

– PULSE SHAPING• Copper discs

– DATA ACQUISITION• Tektronix TDS 3034 Digital Oscilloscope

– SIGNAL CONDITIONING• Ectron model 778 (3 MHz bandwidth)

P.S.Follansee, Metals Handbook, vol.8, American Society for Metals, 1985K.F. Graff, Wave Motion in Elastic Solids, Dover Pub., Inc.1991

SPLIT-HOPKINSON PRESSURE BAR APPARATUS ( SHPB)

• SPECIMEN GEOMETRYRef

– Rectangular cross-section– Laminate thickness (t) ~ 0.17 to 0.25 inches– Specimen width (b) ~ 0.25 inches– Specimen height (H) ~ 0.25 inches

• SPECIMEN ALIGNMENT– Centering disc & slider ring

Ref: E. Woldesenbet & J.R. Vinson, AIAA Journal, Vol.37, Sept. 1999.

b

H

t

90°

0°

Compression Testing…Results

0 0.004 0.008 0.012 0.016 0.02Strain [in/in]

0

10000

20000

30000

40000

50000

60000

70000

80000

90000

100000

110000

120000

Stre

ss [p

si]

TR-CRU[0°]

Strain-Rate [1/s]0.00171450633763

Toray Unitape

Failure modes

0.0017 s-1 1 s-1 450 s-1 633 s-1 763 s-1

Compression Testing…Results

Toray Unitape

PIN-BEARING RESPONSE

• Material Systems– Newport NB321/7781 fiberglass– Newport NB321/3k70 PWCF– Toray T700G-12K-50C/3900-2 PWCF

• Laminate type – [45/0/45/0/45]s

• Pin diameter : 0.125 inches • Test speeds : quasi-static, 1 ,10,100, and 250 in/s

• RESULTS– Sustained loading past initial failure, decreases

at higher rates of loading– Failure mode

0 2 4 6 8 10

Time

Time to Peak load

0

500

1000

1500

2000

2500

Bea

ring

Load

(lbf

)

Material : Newport fiberglassQuasi-static1 in/s10 in/s100 in/s

PIN-BEARING RESPONSE

• RESULTS– Bearing strengths based on

peak recorded load• Hole deformation not

measured– Pin bearing strength increases

with test speed

10-3 100 101 102 103

Nominal Stroke Rate (in/s)

0

100,000

200,000

300,000

Bea

ring

Stre

ngth

(psi

)

Newport FiberglassNewport PWCFToray PWCF

FLEXURE TESTS

• LAMINATED BEAMS– Material systems

• Newport NB321/7781 fiberglass

• Newport NB321/3k70 PWCF– Layup sequence

• [0/45/45/0]

• SANDWICH BEAMS– Material systems

• Newport NB321/7781 fiberglass

• Newport NB321/3k70 PWCF– Layup sequence

• [0/45/0/45/CORE]S

• STATUS : Testing completed

SPECIMEN

LOADING ROD / ACTUATOR

LOAD CELL

0.5

1

FLEXURE TESTS – Sample results

Peak bending load Vs Normalized time for Carbon fiber beam specimens at various strokes rates.

FLEXURE TESTS – Sample Results

Peak bending load Vs Normalized time for fiberglass beam specimens at various strokes rates.

FLEXURE TESTS - RESULTSTypical strain rates achieved in the gage section of the beams

Maximum strain rate achieved in the test region ~ 100 s-1

SCALED FUSELAGE SECTION TESTS

Work in Progress

• Rate effects on delamination fracture toughness – nearing completion

• Rate effects on behavior of corrugated webs under compression– Failure modes– Energy absorption

• Signal modulation corrections for test data• Tensile characterization using SHPB ( rates ~

103 1/s)

P

P