|
[Sponsors] |
December 1, 2009, 11:50 |
Differents Meshes for an unique surface
|
#1 |
New Member
Tatiane H.
Join Date: Nov 2009
Posts: 4
Rep Power: 17 |
Hi, everyone!
I'm simulating an air duct in CFX. At first, the air pass through a rectangular geometry and after, the air goes down through a cylindrical geometry. I appreciate if someone could help me to do after putting " All Quad" in the rectangular part and another kind of mesh at the cylindrical part (for this, I put surface mesh and after, I extruded the surfaces). When I set the problem, the solver didn't run because there were 4 isolated surfaces. I believe the problem is in the meshes that coincide in a same surface. What can I do?! Thanks for any response! |
|
December 1, 2009, 15:46 |
|
#2 |
Member
Pratik Mehta
Join Date: Mar 2009
Posts: 32
Rep Power: 17 |
hi ,
Your method is correct before I comment, can you please tell me what meshing tool you have ?. thanks best regards Pratik |
|
December 2, 2009, 05:53 |
|
#3 |
New Member
Tatiane H.
Join Date: Nov 2009
Posts: 4
Rep Power: 17 |
Hi, Pratik!
I'm really grateful for you help. For surface mesh I have All tri, Quad Dominant, Qua w/one tri and All Quad. For volume mesh, there are Tetra/Mixed, Hexa Dominant and Cartesian. I actually don't know too what kind of mesh to put on the cylindrical part. Thanks a lot! Tatiane |
|
December 2, 2009, 07:07 |
|
#4 |
New Member
Ramandeep Singh
Join Date: Nov 2009
Posts: 20
Rep Power: 17 |
i guess u are using a way to many types of mesh.
the problem can be simply solved considering prism on the boundary surface and Hexa or Quad meshing as the volume. please read the tutorials.
__________________
regards, Raman |
|
December 2, 2009, 09:44 |
|
#5 |
Member
Pratik Mehta
Join Date: Mar 2009
Posts: 32
Rep Power: 17 |
hi,
As mention by Raman that you could just do tet + prism meshing and aviod the issues which you are falling in now . Well I would actually suggest to do the the following way if you prefer doing it differently For the rectangular geometry ,you can do hybrid mesh ( tet +prism layers ) and later you can just extrude the face to represent circular part. In this case all you need to know the extrussion lenght and the curve for attaching it. So your .tin file will only have surfaces of the rectangular geometry and so the the circular part ( which wil be extruded part ). best of luck Ciao Pratik |
|
December 2, 2009, 11:37 |
|
#6 |
New Member
Join Date: Nov 2009
Posts: 11
Rep Power: 17 |
check the quality of your mesh with your meshing tool, this will find the isolated surfaces, zoom in on them and post a snapshot of it overhere, this gives us a beter view of your mesh and your problem
|
|
December 3, 2009, 12:38 |
|
#7 |
New Member
Tatiane H.
Join Date: Nov 2009
Posts: 4
Rep Power: 17 |
First of all, thanks a lot for all the helps given!
I tried tetra/prism layer and it simulated well. Now I'll try the extrude way and see if I have success on it. I put the figure, but i don't know if it was what you wanted, Champshof, as i'm a really begginer on CFX. Best regards for everyone! |
|
December 4, 2009, 06:04 |
|
#8 |
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,871
Rep Power: 144 |
It looks like you have not merged the cylinder and square duct into the same solid. This means they will be independent fluid regions (unless you join them with a GGI but only do this when you have no other choice).
|
|
December 4, 2009, 06:16 |
|
#9 |
New Member
Join Date: Nov 2009
Posts: 11
Rep Power: 17 |
As wat ghorrock said, it looks like you have a H-grid mesh for the rectangular geometry and a O-grid mesh for the cylindrical geometry. Try to merge the O-grid mesh into the H-grid mesh, this will probably solve your problem.
A GGI is an option but I wouldn't do that because it look like an important region of your problem and it will cause a big accuracy loss (unless you refine the mesh on you interface several times) |
|
December 8, 2009, 12:36 |
|
#10 |
New Member
Tatiane H.
Join Date: Nov 2009
Posts: 4
Rep Power: 17 |
Hi, people.
I tried to merge the O-grid but the isolated volumes still exist. I tried Hexa-Dominant too, but the mesh appears all desconfigurated. So, I computed tetra/mixed and it's okay! But, I'll keep trying doing it in a different way. Thanks a lot for all helps! Best regards, Tatiane. |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
[Gmsh] Error : Self intersecting surface mesh, computing intersections & Error : Impossible | velan | OpenFOAM Meshing & Mesh Conversion | 3 | October 22, 2015 12:05 |
[Gmsh] Problem with Gmsh | nishant_hull | OpenFOAM Meshing & Mesh Conversion | 23 | August 5, 2015 03:09 |
[Gmsh] boundaries with gmshToFoam | ouafa | OpenFOAM Meshing & Mesh Conversion | 7 | May 21, 2010 13:43 |
CFX4.3 -build analysis form | Chie Min | CFX | 5 | July 13, 2001 00:19 |
Empty surface in sliding meshes. | Pierre Ricco | FLUENT | 2 | March 14, 2001 17:01 |