User login


You are here

Subroutine simulation migration from HETVAL to UMATH


I'm already simulating a shape memory alloy behavior with the aid of the HETVAL subroutine, but I'd like to use UMATHT instead, since the HETVAL can't simulate correctly heat absortion(cooling down).

I've tried to add the required values for UMATHT without achieving a correct phase fraction simulation, and some times even no compilation of the .for file.

Is there any other examples that uses UMATHT besides the Stefan problem?
Any other recommendation about HETVAL to UMATHT migration?

Thanks in advance!



I don't know of any other sample UMATHT subroutines, but I believe you can introduce volumetric heat generation or absorption in UMATHT via the definition of the internal energy per unit mass, which is u.  To include volumetric terms, the proper definition for u is:

 u = u + du - q''' delta_t / rho

Note that q''' denotes a volumetric **generation** term (with units of power / volume).  The negative sign in front of q''' was a bit non-intuitive to me at first, but it is in fact correct and is due to the definition of the Jacobian for Newton's Method (more details are shown in the Abaqus Documentation for UMATHT).  I verified this definition of u for cases with heat generation, and while I don't recall doing so for heat absorption, I don't see why it would change.  Hope this helps and good luck.

Actually the problem is that I'd like to use a material property subroutine that depends in the heat transfer and not in heat generation, the subroutine works just fine with heat flux input but it doesn't in heat flux ouptut, therefore the need of another subroutine independent of heat generation like UMATHT.

Frank Richter's picture


Hello Paulo,

Steffie the Snake did not act according to her instructions.

What does Abaqus report in case of " and some times even no compilation of the .for file." ?

Did you modify the model ?

Send me
Input file together with HETVAL and other subroutines if need be
Input file together with UMATHT and other subroutines if need be
target data from measurement

Keep in mind: you may also post on
if we all fail.

Perhaps the Snake is at least willing to leak my E-mail-adress to you.




The problem is related with an overdefinition of the FLUX variable, that part is unclear.

Unfortunatelly the files are with her, so I'll have to wait until monday to get them.

Thanks for the help!


Subscribe to Comments for "Subroutine simulation migration from HETVAL to UMATH"

Recent comments

More comments


Subscribe to Syndicate