I am modeling in Abaqus a metal plate subjected to tension (CT), with an existing crack surface, using XFEM technique. I need to perform cyclic analysis so I need to assign hysteretic response to the material. To do that, I need to implement damage initiation and evolution according to hysteretic energy criterion. In Abaqus, this latter is not applicable if damage is already defined by traction separation laws (in my case max principal stress). I defined a surface criterion to the crack according to Paris Law. Is it a problem if I am using XFEM and the damage is defined only by hysteresis energy criterion, without implementing tension separation laws criterion for damage?
Dear Danilo
I Should mention that Xfem is a numerical procedure for solving an equation in non-continuum media.It isn't a fracture propagation criterion. Every crack Should have a criterion to propagate. In Abaqus, cohesive zone and Xfem based cohesive zone are methods for simulating cracks. As a result of my previous sentences is that you can use the Abaqus defaults or define UDMGINI subroutine for damage initiation on your own criterion, but I don't see any subroutine for damage evolution. I recommend you to use your own code rather than using Abaqus.
Best Regards
Dear Danilo
I Should mention that Xfem is a numerical procedure for solving an equation in non-continuum media.It isn't a fracture propagation criterion. Every crack Should have a criterion to propagate. In Abaqus, cohesive zone and Xfem based cohesive zone are methods for simulating cracks. As a result of my previous sentences is that you can use the Abaqus defaults or define UDMGINI subroutine for damage initiation on your own criterion, but I don't see any subroutine for damage evolution. I recommend you to use your own code rather than using Abaqus.
Best Regards
Dear Farzad,
I defined a fracture criterion assigned to the crack surface (XFEM) according to Paris Law (I supplied c1, c2, c3, c4, Gtresh/Gc, Gpl/Gc, ...). Do you think this way is not working? For damage initiation I assigned traction separation law and for evolution an Energy criterion. The only fatigue response parameters have been assigned to the crack surface.
What is your opinion about this way to model fatigue cracking?
In the following the model script related to this assignment
*Material, name="Al 7075-T6_nEW"
*Damage Initiation, criterion=MAXPS
5.72e+08,
*Damage Evolution, type=ENERGY, mixed mode behavior=POWER LAW, power=1.
42200.,42200.,42200.
*Density
2810.,
*Elastic
7.19e+10, 0.33
*Plastic
5.3e+08, 0.
5.7e+08, 0.0005
6e+08, 0.002
6.2e+08, 0.01
**
** INTERACTION PROPERTIES
**
*Surface Interaction, name=IntProp-1
1.,
*Surface behavior
*fracture criterion, type=fatigue, mixed mode behavior=power
0.5,-0.1,5e-7,1.15,0.1,0.9,42200,42200
42200,1,1,1
Do you suggest this program? Do you guess ABAQUS is not working well for this kind of simulations?
- Badges
- Science topic