LS Dyna

LS Dyna

Topics relate to LS-DYNA, Autodyn, Explicit STR and more

User Defined Material with EOS

    • WooMin Jang

      Hi,i'm struggling with using user defined material with eos..

      User Defined material (Fluid) goes well only with existing ueos 21v, which is gruneisen EOS (Material code i defined was verified through some examples.)

      I conclude problem is related to common variable 'pc(i)' which is pressure-cutoff.

      pc(i) variable doesn't work with another built-in eos, but only work with ueos because pc(i) defiend only in ueos

      I have questions.

      1) Is there any pressure-cutoff common variable to link with built-in ls-dyna eos??

      2) Is there any common variables to call EoS-calculated Pressure in library?

      3) or are there other methods to use the umat code with eos ?? 

      I attached part of the code.

      Any opinion is okay. Thanks in advance.



    • Ushnish Basu
      Ansys Employee

      The pressure cutoff value comes from the material, e.g. MAT_NULL has a PC parameter which is passed on to the EOS

      You can see it's use in the standard umat44v (note the common block where it is stored)

      If you want to use a user material with a built-in EOS, you can combine them in the PART card

      Make sure to set the NHV for the material to be 4 larger than what you actually need, and start your material history variables (in hsv or hsvs) only from the 5th position. The first 4 positions are reserved for EOS variables such as internal energy, bulk viscosity and current volume. 

      The EOS calculations are done after the user material routine is called, so there's not direct way to get the pressures calculated in that step. However, you can get the pressure for the previous step from the stress tensor at the beginning of the step.


      • WooMin Jang



        But I wonder that in umat44v, there only exists one code line to use umat with eos.

        I also did that, But it didn’t work with any built-in eos in simple newtonian fluid code i defined.

        But, in nonsense, it works very well with user-defined eos, where there is pc(i) for pressure-cutoff.

        what would be the cause of this difference?? (Is ‘c$omp threadprivate (/eosdloc/)’ essential for setting pc?)

        Already, i said i suppose pc(i) will be the cause… and what could be the solution??

        I’m very confused…

        P.S. what are 4 history variables stored exactly in order ??


        Thank you for your sincere help.





    • Ushnish Basu
      Ansys Employee

      The first three history variables are the ones I mentioned, in that order. The 4th one depends on the particular EOS. Unfortunately, I am not able to provide any further detailed help that is specific to your situation


      • WooMin Jang

        OK anyway, thanks for your advice.

Viewing 2 reply threads
  • You must be logged in to reply to this topic.