[Awak]

Hello everybody,

I am trying to create a mesh on a cylindrical tube equiped with a very thin valve on its wall (for FSI study), but whenever I use SHM I seem to notice some strange behaviours in the snapping process.

The valve is 0.15-0.35mm thin and has a tapered shape (please refer here for the STL file as well as other dictionaries and .obj files: Google Drive Folder )

I first disabled snapping and layers to find the appropriate refinement level for edges and surfaces. Because my valve is tapered, I cannot match exactly this geometry during this first step. This should be solved by the snapping phase. However, here is what I get when I snap (see attached screenshots).

These strange curved shapes are not present in my STL geometry (nodes should match the STL surface according to this page), so somehow the snapping is not done like it should. I don't understand what is wrongly set in my snappyHexMeshDict. Do you have any idea where this comes from and how I can solve this?

Thanks for your help,

Florian

[me3840]

Did you do a surface feature extraction on this geometry?

[Awak]

I did. The surface feature extraction was successful, but I ended up with this kind of SHM output. I solved the issue by increasing refinement and adjusting max cell number and mesh quality parameters.

[kmooney]

Make sure you turn off 'decompose polyhedra' in paraview before trying to inspect SHM meshes. This way you'll be looking at the actual cell elements and not the triangulated elements.

[Awak]

Originally Posted by kmooney

Make sure you turn off 'decompose polyhedra' in paraview before trying to inspect SHM meshes. This way you'll be looking at the actual cell elements and not the triangulated elements.

I did, I still have the same problem. The snapping step rounds the mesh instead of making it match the initial geometry's features.