CFD Online Discussion Forums

CFD Online Discussion Forums (
-   ANSYS Meshing & Geometry (
-   -   [ICEM] Multiple edges (

asal February 10, 2013 08:19

Multiple edges & Delaunay Violation in ICEM
Hello everyone.

I have generated a kind of Hybrid mesh for a geometry. I generate two mesh Hexa and tetra , then I merge them with the Merge Meshes ==> Method ==> Merge volume meshes.
I have faced with two main problem.
I got some Delaunay violation elements in tetra part. How can solve this problem? this is an example of these element types:

2nd is in the connected edge I got some Multiple edges elements.
the questions is how can I solve this problem to avoid getting multiple edges? I try to merge these two surface meshes, but everything collapsed!!
I have followed the tutorial which is provided in ANSYS portal: "Mesh Merging in a Hybrid Tube"

But here also we got Multiple Edge error:
See this figure:

anyway, What kind of probable problem occur during simulation with Multiple edges? and how can I solve these problems? this is also an example of multiple edges in my mesh:
See figure:

asal February 15, 2013 03:44

No Idea?!!

PSYMN March 22, 2013 11:10

The dealunay violation is only an issue if you plan to run a delaunay mesher later... If you already have a tetra mesh, then you can ignore that one.

Multiple edges are simply edges with more than 2 surface elements connected. In this case, you have the side walls and the interface elements. This is listed under "possible problems" and not "errors". Your solver probably won't mind the multiple edges at all. However, if you want flow to pass thru the interface between the hexas and tetras you have two choices... Either set up the interface correctly in the solver to allow that to happen (this varies with the solver), or just delete the shells in the interface part before you export to the solver.

If you decide to remove the shells in the interface part, make sure you also put the volume elements on either side into the same part or you will get an uncovered faces error (and that will cause a problem with your solver).

All times are GMT -4. The time now is 21:05.