Hello,

I am solving statics equations for a thick disk (or short cylinder) with a round hole in the center. The load is applied evenly around the hole, and parallel to the cylinder axis. I am solving for displacement {ux, uy, uz}, z is of course along the axis. Doesn't seem too difficult.

The solution that I get shows that displacement varies as I go around the whole (at constant radius). It seems as if there is Exp[2 I [Phi]]] factor. Intuitively, there should not be such factor - the geometry is symmetrical, force is symmetrical, boundary conditions are symmetrical.

The only thing that I have been able to identify as a possible cause is the meshing. When I display the "Wireframe" of the mesh, it shows what looks like a square-ish pattern around the (round) center hole.

In the attached image I can clearly see a square around the hole, where the mesh is less dense (or is it something else), and I think this may be producing unphysical artifacts in my solution. This particular mesh is denser than I actually use, so the square is more visible.

So, my question is, how can I generate a mesh without such spurious effects? I could not find anything definitive in documentation. And is this really what causes the problem? Thanks in advance and happy new year.

Yuri.