Get point coorditnate in umat for anisotropic problem

[sanshi2000]

Hello. I am using Gasser-Holzapfel-Ogden model through user material subroutin. In this model, fiber direction varies and depending on the points computed, so the Cauchy stress and elasticity modulus outputed by umat subroutine are also related to these points. Any suggestion on how to get the coordinates of points, at where umatln.f is running, in umatl.f subrotuine? Thanks a lot. 

[FEAP_Admin]

The only way to do this is to add the coordinates to the subroutine arguments up the call chain to the point where where modlfd( ) or modlsd( ) is called [for example, in  fld2d1.f].  The coordinates of the Gauss points are available in all the elements so they just need to be passed down.  The changes are easy but need to be done consistently in all the elements.   

[Prof. R.L. Taylor]

There is a common block named 'elcoor.h'  (see files in /include) that contains two arrays: xref(3) and xcur(3).  Many of the solid elements compute values for the xref(ndm) (these are the reference coordinates of quadrature
points) during formation of element arrays (e.g., see strn3d.f or stre3m.f in /elements/solid3d).  Adding this common block to your constitutive module should make the coordinate available.

[Schandra]

Dear Feap community,

Is there a possibility to access xl() coordinates in a umatl. I am printing out xl() for my umat and they are not the way I was expecting them to be in comparison to an existing user-element/material subroutine.

Thanks,
Chandra

[K.Li]

The xl()  array has the nodal coordinates of the element, is this what you are expecting?

[Schandra]

Dear Mr. Li,

Thanks for your reply.
When I print out the xl() coordinates from my umat, somehow I am unable to get the expected values.
As an example I took a unit cube to print out the coordinate values from my umat. On the contrary, I am able to see the correct coordinates of the element in Ofile.

Thanks,
Chandra

[FEAP_Admin]

Did you read message #2 in this thread?!  XL( ) is not passed to the material models.

[Schandra]

Dear FEAP_Admin,

I have misread it initially, please excuse me. Now it is clear to me.
Thanks,
Chandra

[JStorm]

note, that there is a bug in some elements which yields to wrong values in xcur:
http://feap.berkeley.edu/forum/index.php?topic=896.0