Recent Enhancements to the GISSMO Failure Model in LS-DYNA · Recent Enhancements to the GISSMO...

Dynamore GmbHIndustriestraße 2

70565 Stuttgarthttp://www.dynamore.de

Recent Enhancements to the GISSMO Failure Model in LS-DYNA

André Haufe1, Markus Feucht2, Frieder Neukamm2, Paul DuBois3

1 DYNAmore GmbH, 70565 Stuttgart, Germany2 Daimler AG, EP/SPB, 71059 Sindelfingen, Germany3 Consultant, 63065 Offenbach, Germany

2European LS-DYNA Conference 2011 - Strasbourg

Weight Composites

High strength steel

Light alloys

Polymers

Safety requirements

Cost effectiveness

New materials

Design to the point

New power train

technology

Technological challenges in the automotive industry

3European LS-DYNA Conference 2011 - Strasbourg

Technological challenges in the automotive industry

Damage

σ σ

m axε

E

Failure

=fail true

E

σ σ

Anisotropyc

a

b

η

ε( )eE

σ σ

σ y

Fracture growthDebonding

Weight Composites

High strength steel

Light alloys

Polymers

Safety requirements

Cost effectiveness

New materials

Design to the point

New power train

technology

Plasticity

4European LS-DYNA Conference 2011 - Strasbourg

???

Forming simulation

Closing the process chain

� v. Mises or Gurson model

� Strain rate dependency

� Isotropic hardening

� Damage evolution

� Failure models

(damage variable necessary!!)

IIσ

Iσ

IIIσ

� Hill based models

� Anisotropiy of yield surface

� Kinematic/Isotropic hardening

� Failure by FLD

(post-processing)

� No computation of damage

IIσ

IσIIIσ

IIσ

Iσ

IIIσ

Crash simulation

5European LS-DYNA Conference 2011 - Strasbourg

Different ways to realize a consistent modeling

One Material Model for Forming and Crash Simulation

� Requirements for Forming

Simulations: Anisotropy, Exact

Description of Yield Locus,

Kinematic Hardening, etc.

� Requirements for Crash

Simulation: Dynamic Material

Behavior, Failure Prediction,

Energy Absorption, Robust

Formulation

� Leads to very complex model

Modular Concept for the Description of Plasticity and Failure

� Plasticity and Failure Model are

treated separately

� Existing Material Models are kept

unaltered

� Consistent modeling through the

use of one damage model for

forming and crash simulation

*MAT_ADD….(damage)

6European LS-DYNA Conference 2011 - Strasbourg

Produceability to Serviceability

Gurson

Barlat Gurson

Forming simulation Crash simulation

000,0,,, ftplεσ

fplεσ ,

Mapping

tpl ,,εσ

background

Fo

rtra

n-P

rog

ram

� Anisotropy� Yield locus

� Damage� Dynamic effects

Schmeing, Haufe & Feucht [2007]

Neukamm, Feucht & Haufe [2007]

Incompatible Models:Isochoric plastic behavior

7European LS-DYNA Conference 2011 - Strasbourg

Produceability to Serviceability: Modular Concept

Damage model

Material model Material model00,

, tplε

plεσ ,

Mapping

tpl ,ε

D D

Damage model

plεσ ,

Forming simulation Crash simulation

D D

Modular Concept:

•Proven material models for both disciplines are retained

•Use of one continuous damage model for both

8European LS-DYNA Conference 2011 - Strasbourg

GISSMO

Barlat Mises00,0

,, tplεσ

plεσ ,

Mapping

tpl ,,εσ

Fo

rtra

n-P

rog

ram

D DGISSMO

plεσ ,

Forming simulation Crash simulation

Ebelsheiser, Feucht & Neukamm [2008]

Neukamm, Feucht, DuBois & Haufe [2008-2010]

Produceability to Serviceability: Modular ConceptCurrent status in 971R5

9European LS-DYNA Conference 2011 - Strasbourg

J. Lemaitre, A Continuous Damage

Mechanics Model for Ductile Fracture

Overall Section Area containing micro-defects

Reduced (“effective“) Section Area

SS <ˆS S

SSD

ˆ−=

Measure of

Damage

Reduction of effective cross-section leads to reduction of tangential stiffness

� Phenomenological description( )D−= 1

*σσ

GISSMO – a short descriptionEffective stress concept (similiar to MAT_81/224 etc.)

10European LS-DYNA Conference 2011 - Strasbourg

Gurson

Mises

FormingCrash

GISSMO

Damage Evolution

Damage overestimated for linear damage accumulation

Failure Curve

Neukamm, Feucht, DuBois & Haufe [2008-2010]

GISSMO - a short description Ductile damage and failure

triaxiality

11European LS-DYNA Conference 2011 - Strasbourg

Gurson

Mises

Forming

Crash

Evolution of Instability Material Instability

Material Instability

( )v

n

locv

Fn

F εε

∆=∆− 11

,

Tensile test specimen DIN

EN 12001

GISSMO – a short description Engineering approach for instability failure n

t

1

2ψ

triaxiality

Neukamm, Feucht, DuBois & Haufe [2008-2010]

12European LS-DYNA Conference 2011 - Strasbourg

GISSMO – a short descriptionInherent mesh-size dependency of results in the post-critical region

0,0

0,2

0,4

0,0 0,1 0,2 0,3 0,4 0,5

Engineering Strain

En

gin

ee

rin

g S

tre

ss

Experiment

0,5mm

1mm

2,5mm

Simulations of tensile test specimen with different mesh sizes

Regularization of mesh-size dependency

element size

Influence of damage in postcritical region

13European LS-DYNA Conference 2011 - Strasbourg

DMGTYP: Flag for coupling (Lemaitre)

( )D−= 1* σσ

DCRIT, FADEXP: Post-critical behavior

0

0

True Strain

Tru

e S

tress

GISSMO dmgtyp2

MAT_024

0

0True Strain

Tru

e S

tress

m=2

m=5

m=8

−

−−=

FADEXP

CRIT

CRIT

D

DD

11

* σσ

GISSMO – a short description Generalized Incremental Stress State dependent damage MOdel

14European LS-DYNA Conference 2011 - Strasbourg

To be considered:

8 Specimen geometries

5 Discretisations

GISSMO Identification of damage parameters: Range of experiments and simulations

15European LS-DYNA Conference 2011 - Strasbourg

0,0

0,2

0,4

0,6

0,8

1,0

1,2

1,4

1,6

1,8

2,0

-0,1 0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7

Element size 1mm

GISSMO Equivalent plastic strain vs. triaxiality

fε

triaxiality

16European LS-DYNA Conference 2011 - Strasbourg

Flachzugproben DIN EN 10002

0,00

0,10

0,20

0,30

0,40

0,50

0,60

0,00 0,10 0,20 0,30 0,40

Mini-Flachzugproben ungekerbt

0,00

0,10

0,20

0,30

0,40

0,50

0,60

0,0 0,2 0,4 0,6 0,8

Mini-Flachzugproben Kerbradius 1mm

0,0

0,1

0,2

0,3

0,4

0,5

0,6

-0,10 0,10 0,30 0,50

Arcan

0

2

4

6

8

10

12

0,0 0,5 1,0 1,5

Scherzugproben Kerbradius 1mm, 0°

0,00

0,10

0,20

0,30

0,40

0,50

0,60

0,00 0,05 0,10 0,15 0,20

Scherzugproben Kerbradius 1mm, 15°

0,00

0,10

0,20

0,30

0,40

0,50

0,00 0,05 0,10 0,15 0,20

Versuch

GISSMO

Gurson

constant (v. Mises)

Small tensile test specimen Notched tensile specimen, notch radius 1mm Shear test, inclined 15

Tensile specimen DIN EN 12001 Shear test, straight

GISSMO vs. Gurson vs. 24/81 Comparison of experiments and simulations

17European LS-DYNA Conference 2011 - Strasbourg

GISSMO� Failure Strain constant� Fracture energy constant� Identification of Damage Parameters

is more straight-forward

Gurson� Resultant Failure Strain constant� Failure energy depending on el. size� Identification of damage parameters

is difficult

Gurson vs. GISSMO – “regularized” Regularization of element size dependency

18European LS-DYNA Conference 2011 - Strasbourg

Example: tension rod

damage

instability

GISSMO input

19European LS-DYNA Conference 2011 - Strasbourg

Example: Arcan shear test

instabilitydamage

damage

triaxiality

20European LS-DYNA Conference 2011 - Strasbourg

� Constant failure criterion � Linear damage accumulation� Failure not predicted correctly

� GISSMO-Criterion� Linear accumulation

of damage� Possibly overestimated

damage

� GISSMO-Criterion� Nonlinear damage

accumulation� Rupture predicted correctly

GISSMO Deep-draw simulation of cross-die using GISSMO

21European LS-DYNA Conference 2011 - Strasbourg

Forming simulation:*MAT_36 (Barlat 89)*MAT_ADD_EROSION

(GISSMO)

Crash Simulation:*MAT_24 (Mises)*MAT_ADD_EROSION

(GISSMO)

Plast. strains Thickness distribution Damage

Mapping

Process chain with GISSMO

22European LS-DYNA Conference 2011 - Strasbourg

GISSMOFailure criterion extd. for 3D solids

Lode

Parameter-1

1

0

-0.5

0.5

0.50

-1-0.5

1Triaxialität

Bru

chdehnung

Triaxialität

Bru

chdeh

nung

-0.5 0 0.5 11

� For shells (2D with the assumption of plane stress ) triaxility

and Lode angle depend on each other.

� fracture strain is a function of the triaxiality

� For Solids (3D) both the Lode angle and triaxiality are

independent

� fracture strain is a function of triaxiality and Lode angle

Shells (2D) Solids (3D)

Baseran [2010]

η

ηη

ξξ

23European LS-DYNA Conference 2011 - Strasbourg

Parameter definition

1

3

m

vM vM

Iση

σ σ= =

3

3

27

2vM

Jξ

σ=

3 1 2 3J s s s=mit

[Source: Wierzbicki et al.]

Stress domain in

sheet metal forming

GISSMOFailure criterion extd. for 3D solids

1 60orξ θ= − = − °

0 0orξ θ= = °

1 60orξ θ= = °

24European LS-DYNA Conference 2011 - Strasbourg

Parameter definition

1

3

m

vM vM

Iση

σ σ= =

3

3

27

2vM

Jξ

σ=

3 1 2 3J s s s=mit

[Source: Wierzbicki et al.]

Stress domain in

sheet metal forming

ξ

η

Xue

Hutchinson

Gurson std.

Xue

Hutchinson

Gurson std.

η

fε

[Experimental data by Wierzbicki et al.]

GISSMOFailure criterion extd. for 3D solids

25European LS-DYNA Conference 2011 - Strasbourg

Summary

� Features of GISSMO:

� The use of existing material models and respective parameters

� The constitutive model and damage formulation are treated separately

� Allows for the calculation of pre-damage for forming and crashworthiness

simulations

� Characterization of materials requires a variety of tests

� Automatic method for identification of parameters is to be developed

� Offers features for a comprehensive treatment of damage

in forming simulations

� Available in LS-DYNA V9.71 R5

� Verification und validation of concepts are under way

26European LS-DYNA Conference 2011 - Strasbourg

END