Developing an UMAT

[analyst9]

Hello:
I am planning to develop an UMAT for crystal plasticity to be used in ABAQUS. Can I develop my UMAT in C++ and use with ABAQUS ? I know for the fact that in general FORTRAN UMATs are used with ABAQUS. But I am more comfortable with C++ rather than FORTRAN. Could someone please comment on this. Also, if you can please refer me to some relevant lietrature that would be great. Thank you.

 

[gwolf]

You should use FORTRAN with ABAQUS. However, in the general case, it is possible to interface FORTRAN binaries with C++ binaries. You just have to be careful about the way you pass arguments to/from ABAQUS especially w.r.t. strings and arrays.

Don't mess about with this, you're only writing equations and simple if-tests and loops down in the UMAT and any language will do for that.

If you want to be a smart-ass, go ahead and try with C but you will need to hunt around for the rules. If you've got plenty of time I'd do it just for the learning experience.

Regards,

gwolf.

 

[analyst9]

Gwolf:
Thanks for your information. I will better stick to FORTRAN as of now. May be later I will play with C++ for developing an UMAT

 

[gwolf]

I remembered a bit more about the rules with C++ subroutines: 2D arrays can come through, but not necessarily in the same row/column order, you can test it out by writing to chanel 7 (the .msg file) from within the subroutine. Can't remember the character string rules, the multi-dimensional array thing is the most significant.

Sorry, it's been too long to remember properly.

Have fun,

gwolf.