CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM Other Meshers: ICEM, Star, Ansys, Pointwise, GridPro, Ansa, ... (http://www.cfd-online.com/Forums/openfoam-meshing-other/)
-   -   ICEM tetra mesh for OpenFoam (http://www.cfd-online.com/Forums/openfoam-meshing-other/116352-icem-tetra-mesh-openfoam.html)

cfGreen April 17, 2013 12:41

ICEM tetra mesh for OpenFoam
 
Hello!
I spent several weeks now trying to setup an HVAC case using buoyantBoussinesqSimpleFoam. The mesh comes from ICEM. I got very very simple cases running but as the room geometry gets a bit more complex, I can't reach convergence. It always seems that p_rgh is the problem. I discussed this topic with other colleagues (who has no experience with OpenFoam but with other software) and I tried so many meshes of different qualities, many different boundary conditions and solver settings that I came to the conclusion that maybe OpenFoam is not capable of running unstructered meshes at all.
Is anybody out there who got OpenFoam (maybe buoyantBoussinesqSimpleFoam) running using a tetra/prism mesh from ICEM? Do you have any hints on ICEM tetra and OpenFoam? I would rather try a structured mesh but I haven't got the time to get into block meshing.

Thanks for your reply!

T.

akidess April 18, 2013 08:26

I don't have experience with that particular solver, but I have run cases with tetra and polyhedral meshes with satisfactory results.

cfGreen April 29, 2013 10:44

akidess,
thanks for your response so far. I hope somebody else can drop a short line on OpenFoam and unstructured meshes for HVAC applications. I am so desperate trying to get my case running. I am grateful for any hints I can get.

T.

Hashme October 23, 2013 17:42

I have the same problem it seems it is working fine with block mesh but not with unstructured mesh

cfGreen October 24, 2013 07:13

Hi!

In the meantime I tried to run my case with a very tidy structered mesh. Even then bouyantBoussinesqSimpleFoam didn't run. I came to the conclusion, that either my boundary conditions or somer solver preferences are wrong. Currently we use a trial version of DHCAE CastNet. It uses templates for several solvers which helped use running our case and it is capable of generating unstructered meshes. So, with all the correct settings and boundary conditions CastNet sets by default and with the unstructered meshes, I finally have managed to get bouyantBoussinesqSimpleFoam running.

Cheers,
T.


All times are GMT -4. The time now is 21:00.