CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM (http://www.cfd-online.com/Forums/openfoam/)
-   -   stitchMesh and snappyHexMesh (http://www.cfd-online.com/Forums/openfoam/71280-stitchmesh-snappyhexmesh.html)

gdbaldw December 23, 2009 03:09

stitchMesh and snappyHexMesh
 
Is there a robust method for stitching a snappyHexMesh to an underlying Hex Mesh? I presented at Milan in 2008 a robust open source software technique (Blender/CalculiX/Python) for creating a high quality pure Hex Mesh boundary layer grid for any compound/complex surface. At that same OpenFOAM Workshop snappyHexMesh was introduced. What I have recently attempted with some success is to generate an STL that precisely matches the outer surface of the Hex boundary layer, use snappyHexMesh against this STL, and then mergeMeshes/stichMesh the snappyHexMesh to the Hex boundary layer. This method almost always succeeds, however the mesh quality at the stich often has errors. I'd like to learn how to tune both snappyHexMesh and stichMesh to be robust for this approach. I'm using OF-1.6.

My motivation is that the boundary layer mesh precisely matches the detailed features of the underlying body, yet with a very thick boundary layer these features are masked and create an outer surface that is well suited to snappyHexMesh. I've generally found that both patches at the stitch need to have similar face density. However, I have yet to discover a robust recipe for acceptable mesh quality at the stitched patches, and would rather not spend more time on this if it is a fruitless exercise. If this could be made robust, the lack of feature support in snappyHexMesh would no longer be a barrier to its use.

Input and Advice Welcomed,
Doug


All times are GMT -4. The time now is 18:23.