Modelling coated material

I'd say that a multi-layered shell formulation would be the logical and easiest way to go about it. I am however, not 100% is you can use an isotropic material with DEFI_COQU_MULT - that might not be a problem for you though. I really haven't experimented much with composite shells.
To find some cases that use multi-layered shells, try this:

find /opt/aster/STA10.2/astest/*.comm | xargs grep -l DEFI_COQU_MULT

Documentation here:
www.code-aster.org/V2/doc/v9/man_u/u4/u4.42.03.pdf

Regards,

Claus

Hi

what I would do in your case is to mesh the bulk material with solids (hexa 8 node) and then use the surface mesh (quad 4) of this solid to defin a thin DKT shell represeting the coating.

Reusing the surface mesh of the solid, you will have all shell nodes shared with the bulk solid so you don't need to play with kinematic constraints to link the two parts.

To do that, in Salome, you need to create one group of solid elements for the bulk material and one group of surface elem. for the coating. Then in Eficas, in AFFE_MODEL you assign 3D modelling to the group of solids and DKT to the group of surface elements.

After that, you just go on defining the materials and shell properties as usual (AFFE_CARA_ELEM).

It should work well normally

JMB wrote:

Claus wrote:

find /opt/aster/STA10.2/astest/*.comm | xargs grep -l DEFI_COQU_MULT

Actually I would suggest just:
[code:1]
grep -l "DEFI_COQU_MULT" /opt/aster/STA10.2/astest/*.comm
[/code:1]

The 'find' is really not needed, and the xargs not required since the pipe "|" is not being used.

Regards, JMB

You're right, it was a quick copy/paste from a forum post.

To the original poster: Let us know which method works best and gives the best approximation.