CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Gambit size function query (http://www.cfd-online.com/Forums/fluent/45166-gambit-size-function-query.html)

Jon June 26, 2007 07:10

Gambit size function query
 
I am struggling with an annoying gambit problem. I have assigned a size function to a curved surface and a boundary inlfation layer to the same surface. I then try to use a tet based mesh on the surface. The size function seems random and applies it to parts of the face and not to others.

When I try and visualise the size function it starts at a much higher number than that I set, which I have checked again since. Is there an over ride in the defaults that could be causing this or is a it a known bug??

Any ideas would be appreciated.


nick June 26, 2007 09:54

Re: Gambit size function query
 
First, I would attach the b.l. mesh to the surface, pre-meshed with hex cells, then I would mesh the flow domain using the option "Meshed S.F. on B.L. cap". You may need to divide the domain in smaller volumes. Visualize the b.l. mesh only after you set, at least, the spacing of the mesh for the volume that contains the surface with the b.l. mesh.

Jon June 27, 2007 05:29

Re: Gambit size function query
 
It is only a 2D issue at the moment. I have done as you suggested but I am still getting errors. In places along the curve the size function is obeyed and in others the size function is ignored when I mesh the face. It's a shame I cannot post a picture otherwise it would make things a lot simpler to describe!

ritmat June 27, 2007 05:39

Re: Gambit size function query
 
maybe you should mesh the edge first ignoring size function, than boudary layers ignoring size function and then use your size function to mesh your face

Jon June 27, 2007 06:20

Re: Gambit size function query
 
I have tried that also. The issue is with the first cells away from the boundary layers. I have even tried creating a small face over the egde to try that but the same thing occurs. I have stuck the photo on the web below. If you click on it it enlarges. Hopefully one of you kind people can help me solve this. You can see where I have tried the artifitial face over the boundary layer, but that failed also.

Thanks in advance.

http://jonpaton.blogspot. com/

ritmat June 27, 2007 07:02

Re: Gambit size function query
 
looks strange on picture. You have to be sure that there are no size function or other mesh on the others faces which can interact with this face. And then i will try to change parameter of the size function like the growth rate

nick June 27, 2007 10:33

Re: Gambit size function query
 
it seems that the size function works well.it is fine close to the surface and coarse far from it. the picture it's not clear, but I notice that the spacing on the surface is very small, probably too much.you need to reduce the grow rate of the size function or increase the spacing on the surface to get a more uniform mesh.

marcolovatto December 21, 2011 05:02

I am having exactly the same problem
 
Hello folks,

I am struggling with exactly the same problem. In fact, I have two faces and two edges, where each edge is part of only one of the faces. For both, exactly the same boundary layer and size function parameters are applied. For one face, the tri mesh is fine, for the other, the annoying problem occurs.

I still haven't tried to generate the mesh of a face without having set the size function for the other face. Anyway, this topic is 4,5 years old and I write hoping someone have already solved this problem.

Thanks for any help.

Marco


All times are GMT -4. The time now is 16:47.