TAGGED: fluent-mesh
-
-
March 4, 2024 at 11:31 pmGengyan LiuSubscriber
I have this moving boundary in axysimmetrical FLUENT simulation. However, regardless what I do, the region after the dynamic mesh remeshing becomes a lot bigger than the original mesh. Here are my settings, how do i fix it? Also setting the max length to a much smaller value did not fix the issue rather would remesh everything else into that size, which slows down computation significantly. Are there any other ways to get fine mesh specifically in between that gap between the moving boundary and the wall without using adaptive meshing? Adaptive meshing does not work in my case because I will have to remove the moving boundary later on and adaptive cells wont allow that.Â
This is the original mesh and the unmoved boundaryThe below is the mesh after the boundary movement, as you can see, the post dynamic mesh gap between the moving boundary and the wall has a lot less fine mesh than everywhere else.
I am using diffusion smoothing, remeshing, max skew 0.6, max length is that finest mesh size, min length is half of that.Â
Any help would be appreciated.Â
-
March 5, 2024 at 4:07 pmFedericoAnsys Employee
Hello,Â
what remeshing method are you using?
-
March 6, 2024 at 2:59 amGengyan LiuSubscriber
-
March 6, 2024 at 1:00 pmFedericoAnsys Employee
Can you try with Unified Remeshing? This has been shown to be robust in preserving mesh distribution.
-
March 7, 2024 at 1:10 pmFedericoAnsys Employee
Can you show the Mesh Scale Info, before initializing your case?
-
-
March 7, 2024 at 12:29 amGengyan LiuSubscriber
Whenever I try unified remeshing, the application just exits when I click on Calculate. No error or whatsoever. It just first tells the generic info whenever one runs a calculation, saving .gz or .cas files and how many cores I have etc. Then just exits.Â
-
March 7, 2024 at 1:09 pmFedericoAnsys Employee
That would be something worth troubleshooting too. What version of Fluent are you using? And what types of cells are you using in the remeshing zones? It looks like Tetrahedral from your screenshots but are there others? e.g. do you have inflation layers?
-
-
March 29, 2024 at 6:32 amGengyan LiuSubscriber
I am using 22 v1, and I am using triangle. The simulation is axiosymmetric. I do not have inflation layers. Also, I have realized that after 1 iteration, it just says divergence in AMG solver. I have tried to lower the mesh density to around 20% of what is visible here, then it runs perfectly fine. I am using 5e-9 second time steps. I know that the wall does not have enough time to move more than 1 cell length, so no negative cell volume problem. Should I just update ansys to the 24 version? I am worried about backwards compatibility if something happens.Â
-
- The topic ‘How do I define a specific dynamic mesh size?’ is closed to new replies.
-
791
-
427
-
265
-
201
-
180
© 2024 Copyright ANSYS, Inc. All rights reserved.