Learning Forum

[Oleg Makarov]

I have a small question (may be stupid?). I am modeling material MAT_024 with a yield stress equal to 5.5 MPa. I want to simulate material brittle fracture using the SPG method and I tried two methods for the definition of material failure criterion:

1. First method is to define IDAM=1 and very small plastic failure strain (say e_f = 0.0001). In that case failure occurs but it seems that bonds between particles are not fully broken. A situation occurs where plastic deformations continue to grow even after the bond is broken. At the same time, the material in the failed zone continues to behave as if it were bonded. Why does this happen? As I understood the SPG mechanism should completely break the bond when reaching the fracture criterion (in this case e_f=0.0001) and a crack should form in the material.
2. The second method is to define MAT_ADD_EROSION (in addition to MAT_024) with criteria of maximum equivalent stress (SIGVM = 5.5 MPa). In that case, I use IDAM = "Programm controlled" in ANSYS Workbench (as I understand it is IDAM=0). The result is a very brittle fracture with no material further deformation and particles are clearly disconnected from each other after failure. Maybe it is appropriate but I can not understand why there is such a big difference with the First method if the SPG bond break mechanism remains the same?
3. Also, I cannot display the fracture zone in LS-PrePost in the Second method since equivalent stress decreases after a fracture. In the First method I used plastic strain bounds (e_p > 0.0001) to change the color of particles with failed bonds. But in 2nd case I can not visualize the zone of destruction, since equivalent stress can't be the criteria for coloring because of decreasing the value after fracture. What can I do in that case? 

Thank you in advance!

[Ram Gopisetti]

hi , IDAM controls the mechanism of damage where as MAT_ADD_DAMAGE controls the criteria for the material on when damages forms. for MAT_24 with IDAM 1 has to be used and it is coherent to plastic strain defined in that material card and you dont need to define the FS vale again in the section cards. Moreover, MAT_24 is not good for brittle materials, i hope your research should answer it , there are material models like *MAT_BRITTLE_DAMAGE,*MAT_JOHNSON_HOLMQUIST_CERAMICS and concreat related material for handellign brittle materials. you can explore show deleated elemnts in the view option and see if it helps , 

cheers, ram 

[Oleg Makarov]

Sorry, but I don't understand you at all. In this post, I am asking about MAT_ADD_EROSION, but not about MAT_ADD_DAMAGE or any damage at all.

"IDAM controls the mechanism of damage where as MAT_ADD_DAMAGE controls the criteria for the material" - as I understand IDAM controls failure criteria too by parameters FS and STRETCH. Different IDAM values control the failure quantity that will be assessed (plastic strain (IDAM=1), shear strain (IDAM=3), etc). Therefore I don't need MAT_ADD_EROSION or another cards to simulate material failure for the case MAT_24 + IDAM=1.

"for MAT_24 with IDAM 1 has to be used and it is coherent to plastic strain defined in that material card and you dont need to define the FS vale again in the section cards" - why?? In MAT_024 we define only the yield stress vs. plastic strain (hardening rule) and therefore we have to set the failure criterion inside SECTION_SOLID_SPG, right? I use ANSYS 2023R1 (LS-Dyna R13.1), maybe it is important...

"Moreover, MAT_24 is not good for brittle materials" - this is my main question - why I can not model brittle material if I set very very small value of critical plastic strain (parameter FS)? Since this value is very small the fracture should be the same as if I define the critical stress in MAT_ADD_EROSION.

BUT when I tried this I found that material crushing looks very different for these two options. I am wondering because SPG bond breakage follows the same rule - FS > FS_critical & STRETCH > STRETCH_critical.