Hi

Iam new to cfd and trying to simulate the pressure recovery coefficient through a conical diffuser. Using the SST turbulence model, I am wondering what kind of mesh quality tolerance this model (and others) typically have in obtaining reasonably good results. Is a mesh quality (as evaluated using ICEM) of 0.3 generally too low

Hi,

This is strongly problem dependant. If it is a simple flow without separation a coarse low quality mesh might be OK, but if you have separations that will raise the bar and iff you have swirl (therefore likely requiring RSM turbulence models) then that raises the bar even further.

Time spent improving mesh quality is never wasted.

An ICEM mesh quality of 0.3 is OK for most situations, but that may not work for RSM models if that is required.

Glenn Horrocks

Glenn, can you check mesh quality anywhere in CFX mesher? Ive been using that for my tet meshs and wondered if you might know?

Thanks

Steve

Hi,

If you mean CFX-Mesh in workbench - no you cannot check mesh quality (other than visually viewing it and that is a pretty good check if you know what you are looking for).

The way to check a mesh is to load it into CFX-Post and calculate the mesh statistics and quality parameters.

Glenn Horrocks