Post on 03-Jan-2016
description
Bringing Science to Life
Impact Modeling of Random Carbon Fiber Composites
PI: Srdan Simunovic
Haeng-Ki Lee, J. Michael Starbuck,
Pakal Rahulkumar, Raymond G. Boeman
Oak Ridge National Laboratoryhttp://www-explorer.ornl.gov/composites
Bringing Science to Life
Summary for FY1999
Modeling1. Developed a number of material models based on micro-mechanics that include
fiber debonding, matrix micro-cracking and fiber failure.
2. Implemented material models into Dyna3d and performed simulations for simple geometries
3. Documented developments in journal papers, conference proceedings and technical report
Experiments1. Developed new fixture design for characterization of controlled crash
failure of composite materials
2. Tested several composite materials materials for fixture evaluation and development
3. Documented developments in project reports and conference proceedings
Bringing Science to Life
ModelingConstitutive Modeling for Impact
Simulation of Random Fiber Composites
Haeng-Ki Lee, Pakal Rahulkumar,
Srdan Simunovic
Bringing Science to Life
Modeling - Summary
ObjectiveTo develop analytical and numerical tools that efficiently predict the
behavior of carbon-fiber based random composites in vehicular crashworthiness simulations
AccomplishmentsMaterial constitutive models have been developed based on
micromechanical formulation and combination of micro- and macro-mechanical damage criteria
Developed models incorporate probabilistic micromechanics for modeling of evolutionary damage in composite
Models are implemented into finite element code DYNA3D to simulate crashworthiness of composites
Bringing Science to Life
Micromechanics and Equivalence Principle
Heterogeneous
composite
Equivalent homogeneous
material
o or o o or o
I IIFiber
Matrix
Micromechanics based modeling
Heterogeneous composite is replaced by equivalent homogeneous material using equivalent principles based on homogenization
Bringing Science to Life
Mechanical Response of Composite
Low crack density increases apparent yield point
strain
stre
ss (M
Pa)
0 0.0025 0.005 0.0075 0.01 0.0125 0.0150
20
40
60
80
100
120
high crack density without debondinglow crack density without debondinghigh crack density with debondinglow crack density with debonding
Composite with high initial volume fraction of fibers shows quick damage evolution
Bringing Science to Life
Experimental Comparison
Mechanical response Evolution of damaged fibers in material
Bringing Science to Life
Preliminary Impact Simulations
Simulation is used for identification of model characteristics for solving boundary
value problems
Impact force
Node 21
Damage index during impact
Displacement in the z-direction at node 21
Bringing Science to Life
Initial Tube Impact Simulations
Simulation is used for analysis of material model features in tube crush test. The model does not incorporate material failure (FY2000 task)
Effective stress during impact Damage index during impact
Bringing Science to Life
Modeling of Delamination in Composites
• Many possible fracture mechanisms across interfaces
– van der Waals forces– Adhesion bond breaking– void formation and
coalescence• Fracture mechanisms could be
rate-independent or rate-dependent
Mathematical Model
Cohesive Zone
Actual MaterialCohesive Zone
Bringing Science to Life
Cohesive Zone Modeling of Delamination
• Cohesive zone models implemented as cohesive elements are used simulate various fracture mechanisms
• Cohesive zone tractions account for the energy dissipation during crack formation and propagation during delamination failures
• Composite bulk material behavior modeled by regular finite elements and can be elastic, elastic-plastic, viscoelastic, etc.
• Easy to handle automatic delamination nucleation, crack path selection, and multiple failure locations
Bringing Science to Life
Delamination - Technical Issues• Separation of composite bulk and interfacial contributions to measured
global fracture energy• Effect of bulk material behavior, i.e. viscoelasticity, plasticity, and
composite ply orientations on measured global fracture energies• Insights into 3-D effects at free edges/corners that cause failure
initiation• Mechanical performance issues - first cracking strength, stiffness
degradation, post-failure strength, impact energy, under quasi-static and dynamic conditions
• Structural scaling effects of geometry on accompanying energy absorption and failure modes
Bringing Science to Life
Example: Compressive Shear Test of Automotive Windshield Experiments - Simulation
(a) Initial configuration
(b) Crack initiation
(c) Stable crack growth
(d) Instability associated with peak stress
a)
1 m m
b )
1 m m
c)
1 m m
d )
1 m m
Experimental and simulated stress-
strain curves
0
2
4
6
8
10
12
Str
ess (
MP
a)
Strain rate s-1
0 1 2 3
Strain
0.00240.02400.24002.4000fits
Simulation
Bringing Science to Life
a = 0.25h a = 0.72h
a = 1.67hinstability
a = 20h separation
Compressive Shear Test of Automotive Windshield
• Successful in modeling the stable and unstable regimes of crack growth
Bringing Science to Life
Delamination Modeling - Summary
• Family of cohesive elements will be used to model degradation of ply-wise material structural integrity
• Separates constitutive description of fracture and bulk material properties
• Suitable for various materials and solution procedures• Promising for use in multiple failure situations• Need to validate cohesive zone modeling approach by
comparison with experiments from quasi-static and dynamic failure tests
• Refine and develop cohesive zone models for fracture mechanisms accompanying failures in composites
Bringing Science to Life
ExperimentsEnergy Absorbing Mechanisms in Composite
Strips under Progressive Crushing
J. Michael Starbuck, Raymond G. Boeman
Srdan Simunovic
Bringing Science to Life
Objectives of Experimental Work
• Develop an experimental setup for discerning the deformation behavior and damage mechanisms that occur during progressive crushing of composite materials.
• Provide experimental data to support the development of analytical material damage models.
• Isolate damage modes associated with frond formation in composite tubes by testing plate geometries and investigate the effects of:– plate width, plate thickness, strain rate, contact profile shape,
contact profile constraint
Bringing Science to Life
ORNL Fixture
• Modified version of existing test fixture used for crush testing of composite plates
• Observable crush zone• Long crush length: approximately 50
mm• Interchangable contact profile• Out-of-plane buckling prevented by roller
supports• Frictionless roller for contact profile
constraint• Frictional forces can be quantified (TBD)• Adaptable for dynamic testing
Bringing Science to Life
Roller Constraint Effect
• Fixture allows for activation of different damage mechanisms– Delamination
– Fiber micro-buckling
– Fiber kink bands
– Fiber fracture
a) Tight constraint b) Loose constraint c) No constraint
Bringing Science to Life
Typical Load-Displacement Traces
Baydur Glass Fiber CSM
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
0 10 20 30 40 50 60 70Displacement (mm)
Load
(N
)
Specimen Width - 50 mmProfile Radius - 6.4 mmLoad Rate - 5 mm/min
Crush Initiator - 45o ChamferConstraint - None
Akzo 556 Triaxial Braid Panel # 10-13
0
500
1000
1500
2000
2500
3000
3500
4000
4500
0 10 20 30 40 50 60 70Displacement (mm)
Load
(N
)
Specimen Width - 38 mmProfile Radius - 6.4 mmLoad Rate - 5mm/min
Crush Initiator - 45o Chamfer
No Constraint
Loose Constraint
Akzo Prepreg #602 Cross Ply Panel # CP1
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
0 10 20 30 40 50 60Displacement (mm)
Load
(N
)
Specimen Width - 50mmProfile Radius - 6.4 mmLoad Rates - 5mm/min
Crush Initiator - 45o Chamfer
Tight Constraint
Loose Constraint
No Constraint
Graphite Cross-PlyFailure mode: Delamination
Graphite BraidFailure mode: Fiber Buckling
Glass CSMFailure mode: Fracture
Bringing Science to Life
Experiments - Summary• Designed unique test fixture for conducting progressive crush tests on
representative automotive composite materials
• Validated and demonstrated test methodology for using a roller constraint to activate/deactivate certain damage mechanisms
• The effects of constraint condition, profile shape, specimen width, and loading rate will be investigated for chopped carbon fiber/epoxy laminates
• Alternative energy absorbing mechanisms will be studied by considering alternative specimen geometries
• Modifications to the test fixture design will be explored for conducting dynamic tests
Bringing Science to Life
Plans for FY2000
Modeling1. Develop continuum mechanics based material models that will incorporate
results from micro-mechanical models
2. Develop continuum mechanics based model for handling post-peak response in the material (model will address ACC’s concern about ubiquitous "m"-s)
3. Develop delamination model based on cohesive theory approach (model will be used to handle degradation of laminate integrity)
Experiments1. Perform tests on P4A materials and on other carbon fiber composites if
necessary.
2. Analyze different failure modes
3. Develop procedures for triggering different failure modes and mapping loading and constraint space conditions