inp file is okay, but when I Run CalculiX, I encounter this error: I then put ConstraintContact - ConstraintTie is more convenient but it does not allow me to use this.

I then put ElementGeometry2D for Face#1 with thickness = 12.5um, and another one for Face#2 with thickness = 20nm. Then I do mesh, define material parameter, put constraint on two edge (one ConstraintFixed, another ConstraintDisplacement) on Face#1 (because in real world, Face#1 will be stretched while Face#2 is stuck to Face#1). I then make them compound - this is the part I am not very certain, but I found this necessary because I cannot put two separate meshes for each face under the Analysis tab - CalculiX doesn't support more than one mesh. I create two faces, one at z=0 (Face#1), another at z=12.6um (Face#2). I would appreciate if you could point out where I am making mistakes. I have tried what you suggested, but I think I might be doing something wrong. It’s a matter of engineering judgment to decide whether to use shell elements or not but if the sheet is thin enough to make it difficult to obtain good solid mesh (at least 4 elements through-thickness) without significant raise in the computational effort then we go for shell elements (unless the contact must be modeled accurately since shells can’t represent its three-dimensional stress state). We often use shells for thicker sheets too, especially since many codes offer thick shell element formulation. So in this case, even a thickness of 1 mm would be enough to use shell elements. Then you will also have to assign thickness to it using FEM ElementGeometry2D.įor shell meshing usually adopted criterion is that the thickness should be less than 1/10 of the characteristic global dimension. Just model the sheet as a surface and the mesher will automatically recognize it as an object for shell meshing. Then you won’t have to worry about meshing in a through-thickness direction. For such a thin sheet, no doubt you should use shell elements.

Creating thin wall shell in gmsh