[vinz]

Dear snappyHexMesh users,

I've been playing with snappy for a few days and I am wondering something. Is it beter to start snappy with a very fine mesh as a base ans only do 1 or 2 level of refinement around my surface, or to start with a coarse mesh and then impose 6or 8 level of refinement in the snappyHexMeshDict?

To give an example, let say we have a mesh at level 1 in blockMesh, the two options are:
- use this mesh and chose a uniform refinement of 4 around the surface in snappy.
- refine blockMesh mesh by 4 in each direction and then use snappy without refinement, or with a refinement level of 1.

Not easy to explain, but i hope you'll understand.
If somebody have an idea, that would be nice.

Regards,

Vincent

[Axel_T]

Hi Vincent,

the only important thing is the maximum and minimum size of our cells.

When you want to set a refinement with a ratio of 1:8 towards the surface, and the cell size at the surface schould be, lets say 2mm, the initial cell size would be 16mm and the surfaceRefinement-level in sappy would be then 3. (2³=8)

Since very small cells make your solution slower and more unstable, and a too coarse mesh delivers not enough information, this is a important point to think about.

If you are starting with a very coarse mesh from blockMesh, snappy would need perhaps more time for the refinement iterations. But I'm not shure - perhaps you could do some experiments and post the results here.