Instability to stress tensor coupling in /FAIL/TAB1

Alfredo Alameda
Alfredo Alameda Altair Community Member
edited July 2022 in Community Q&A

Hi,

I have one question regarding the use of /FAIL/TAB1 together with /MAT/LAW36 for advanced characterization of failure in high strength steels under a uniaxial tensile test. Right now, I am not able to make my model to account for instability effects.

This is the material definition I am currently using:

#---1----|----2----|----3----|----4----|----5----|----6----|----7----|----8----|----9----|---10----|

/MAT/LAW36/2000
Steel                                                            
#                RHO|           REF. RHO|
             0.00785
#                  E|                  v|              epmax|                et1|                et2|
             210000.                 0.3               1.E20               1.E30               2.E30
#    Nfunc|  Fsmooth|              Chard|               Fcut|                 ef|         |       VP|
         1         1                                     10.                                       1
#    Ipfun|             Fscale|  fct_IDe|               Einf|                 Ce|
         0                  1.         0
#   Ifunc1|
      1001
#             alpha1|
                  1.
#                 e1|
                  0.
#---1----|----2----|----3----|----4----|----5----|----6----|----7----|----8----|----9----|---10----|
/FAIL/TAB1/2000
         1         1                                      1.
                  1.                  1.                  2.
      3020                  1.                  1.      3019                  1.                  1.
      2008                  1.                 2.5                                   -3008         1
                            0.
#---1----|----2----|----3----|----4----|----5----|----6----|----7----|----8----|----9----|---10----|
Tables 3020 and 3019 contain the failure and instability envelopes, 2008 controls the element size regularization and 3008 is the FADEXP vs element length table. I have also tried to define instability by means of "inst_start" and a constant "FAD_EXP" parameter but it behaves similarly.

I see in the model how damage grows in the elements but this is not coupled with the stress tensor, without causing any softening. Instead, stress grows during the test by keeping the values defined by the plasticity curve until damage variable equals to 1. Then the stress drops to zero as the element gets deleted. This behavior is shown in the following graph:
image
I wonder if there exists some kind of flag for activating this coupling damage-vs-stress tensor that I am not aware of.
For your information, the RADIOSS release I am using is v2021.2



Thanks in advance and kind regards,

Alfredo Alameda

Tagged:

Answers

  • Rogerio Nakano_21179
    Rogerio Nakano_21179 New Altair Community Member
    edited July 2022

    Hi

    It may be the case you are using Ch_i_f=1, with that the solution will use FAd_Exp=-3008, and not the function 3008.

    You may try to use Ch_i_f =2 or =3 and see if have any better behavior.

     

    I hope this is helpful

    Regards

Categories