CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   ANSYS Meshing & Geometry (https://www.cfd-online.com/Forums/ansys-meshing/)
-   -   [ICEM] How to find number of elements, nodes for Tetras (https://www.cfd-online.com/Forums/ansys-meshing/74141-how-find-number-elements-nodes-tetras.html)

Karatix March 25, 2010 09:10

How to find number of elements, nodes for Tetras
 
Dear All,

currently I use ICEM and CFX Version 10.0.

I have created several automatically generated meshs around airfoils with tetras and would like to compare the accuracy of the results with the number of elements.

Here is the problem.
When I click on mesh info in ICEM I get this result.

Total elements : 201218
Total nodes : 91401

However, as I use tetras I assume that I have more nodes than elements. Does anybody know whether in ICEM the words were mixed up or my thinking is wrong?

Thank you in advance.

Best Regards,

Karatix

PSYMN March 26, 2010 14:06

ICEM CFD is right.
 
Your thinking is a little bit wrong...

That is not the number of tetras, it is the number of elements. For the same number of nodes, you will have tetras, plus shell elements (2D triangles), plus line elements (1D bars), plus point elements (0D nodes).

However, if you look at mesh info, you can also compare the number of tetras to your number of nodes.

Even then, the ratio may not be what you expect because of the way Octree Tetra divides the original Cartesian grid into 12 tetra elements per cube (plus transition regions, etc). Many other Octree algorithms divide into 8 tetras per cube, but we find 12 gives higher quality. And then if you are comparing it to DeLaunay or advancing front tetra, it just won't be similar at all.

Best regards,

Simon

Karatix March 29, 2010 12:46

Hi Simon,

thank you for your help. I found when pressing mesh info that by scrolling upwards the number of Tetra_4.

I assume and by checking with the manual Tri_3 is just the number of triangles, hence 2D and not part of the tetras.

The ratio of tetras to nodes is 4.009, hence not as I expected. Thanks.

Have a nice day,

Karatix

PSYMN March 29, 2010 15:59

Delaunay...
 
Right, to get a mesh with a ratio closer to what you would expect, try remeshing with delaunay... (this keeps the Octree surface mesh, but replaces the octree tetra with a delaunay fill)... I think I covered how to do that on our youtube site... www.youtube.com/ansysinc/

Simon

Karatix March 31, 2010 09:23

Thank you for the advice, but on version 10 I don't have this feature, but the results I get are good enough for me (apart from the stall angle).

Best Regards,

Karatix

PSYMN March 31, 2010 09:38

Smooth Transition.
 
In version 10, we just exposed it a bit differently... Look for the "smooth transition" option.

PSYMN March 31, 2010 09:40

Upgrade if you can...
 
Also 12.1 is really a lot better than 10.0... If you can upgrade, you will be glad you did.

Current customers can upgrade to the latest software for no additional charge.


All times are GMT -4. The time now is 16:02.