CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Siemens (
-   -   Unresolved & Problem Cells of Solid Type (

Harry August 1, 2005 12:57

Unresolved & Problem Cells of Solid Type
when I use the es-aero tool to generate volume meshes for the extern flow around a car surface and the run check option to check the final meshes, there are always some unresolved cells and problem cells of solid type. and if I change their type to fluid and press the run check button again, the numbers of unresolved cells and problem cells will be zero. How come? and does anyone how to fix them properly? thanks in advance.

Paulh August 2, 2005 15:11

Re: Unresolved & Problem Cells of Solid Type
As I said in your last email on this subject, just ignore them. If you have meshed your vehicle appropriately; correct cell size, good refinement scheme, using a sub-surface of good quality, the bad cells will probably not effect the solution. Besides, fixing all of the problems will take a very long time.

The bad cells have been labeled bad because they probably won't run. Moreover, when you write the .geom file with the button that's inside ES-Aero, additional bad cells will be found and turned to solid. Bad cpmatches will also be on the chopping block. These checks and the subsequent model modifications are increasing your chances of getting a model that will run.

That said, there are a few things to do that will decrease your 'bad' cell count.

Get the latest versions of ammbatch and surfacewrapper (if you are wrapping your geometry). Depending on how old your versions are, you could see significant improvement in mesh quality.

Check your sub-surface for problems, visually as well as with the surface mesh checker. Issues in the sub-surface can/will manifest themselves in the trimmed mesh.

Choose appropriate mesh/template generation parameters. Use a good default surface cell size. Define surface and volume refinements that will capture the more detailed portions of your geometry. If the default refinement scheme is placing transitions in areas of high curvature, expand the refinement. And finally, don't use too many lines to control hard edges in the trimming.

I hope this helps.


All times are GMT -4. The time now is 02:22.