CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   ANSYS Meshing & Geometry (http://www.cfd-online.com/Forums/ansys-meshing/)
-   -   ICEM CFD Scripting (http://www.cfd-online.com/Forums/ansys-meshing/91532-icem-cfd-scripting.html)

Prakash.Paudel August 14, 2011 15:14

ICEM CFD Scripting
 
Can someone with ICEM CFD scripting experience give me some insight on whether Tetra or Hexa method is better/easier to script or it does not really matter ??

stuart23 August 16, 2011 06:59

Tetra is much easier to script because you do not have to assign anything to geometry. If you are trying to build a complex structured mesh around an object using ICEM scripting, you will probably have to reference individual surfaces, lines or points. Unless you are manually building the geometry in ICEM, the names of the geometry objects are created at the import of the CAD model, and are impossible to predict. Therefore you will find it very difficult trying to assign to geometry if you don't know what it is called. The only exception would be if you can auto assign the blocks, however you can not verify that the blocking has been assigned properly unless you open up the mesh and inspect it.

Have you looked into hexacore meshing? It has SOME of the solver benefits of hexa mesh, but with the ease of use of unstructured meshing.... Just a suggestion.

Stu

PSYMN August 17, 2011 15:53

my two cents...
 
I agree with stuart23. Both use the same basic scripting language, but hexa requires you to be more specific. Tetra can be done globally (although you could get very specific about specific curve or surface parameters if you wanted to be... That said, many users have been very successful with ICEM CFD Hexa scripting and it may even be more commonly done than Tetra scripting.

If you do want to go with Hybrid meshing; instead of HexCore, I would suggest trying Octree tetra converted (12 to 1) to Hexas... I prefer the way it is all conformal and handles size transitions better. You could also try the new TGrid Cutcell method available in ANSYS Meshing.


All times are GMT -4. The time now is 19:20.