CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Problems when using level set (http://www.cfd-online.com/Forums/fluent/99061-problems-when-using-level-set.html)

adsl17754 March 25, 2012 11:27

Problems when using level set
 
I tried to simulate a swirl hollow cone spray using Fluent. It was OK to use VOF method, but problem came when I tried to coupled level set with VOF. "No intersection facet found in partially filled cell" kept showing on the screen, and the program seemed not responding then...:confused: Has anyone ever met the same problem? I am a novice, Thanks for attention~

adsl17754 March 28, 2012 04:43

Maybe the meshing quality was poor
 
Hi,everyone~
I still don't know how to solve the problem, but now I am sure that the boundary conditions for VOF+Levelset are OK. So I guess it was a Meshing problem, Report Quality in Fluent showed that minimum orthogonal quality was 0.0039, was it too low for a 3D mesh? Thanks~

rk_k67 January 10, 2013 20:23

Hi,

I sometimes have a problem with running fluent in parallel for some meshes, which could be due to the trading information between domains and the fact that the phi function in level set can not be split and it would be hard to make the Phi distance function to overlap exactly on the same position where the domain is partitioned. that is where I think error comes from !!!!! ? ! ? ! Although all my meshes are fully squared, I still get the problem. Also, I just tried it differently and I got your error. Did you manage to fix your problem and get CLSVOF running !

Cheers

adsl17754 January 10, 2013 22:39

one probable solution
 
Quote:

Originally Posted by rk_k67 (Post 401294)
Hi,

I sometimes have a problem with running fluent in parallel for some meshes, which could be due to the trading information between domains and the fact that the phi function in level set can not be split and it would be hard to make the Phi distance function to overlap exactly on the same position where the domain is partitioned. that is where I think error comes from !!!!! ? ! ? ! Although all my meshes are fully squared, I still get the problem. Also, I just tried it differently and I got your error. Did you manage to fix your problem and get CLSVOF running !

Cheers

Hi rk_k67,

That's a reasonable explanation~ Parallel in FLUENT sometimes causes that problem, try to use "serial" to calculate for several iterations, then switch to "parallel", that solved my problem. Hope it works for you~

Cheers

rk_k67 January 13, 2013 18:17

Quote:

Originally Posted by adsl17754 (Post 401302)
Hi rk_k67,

That's a reasonable explanation~ Parallel in FLUENT sometimes causes that problem, try to use "serial" to calculate for several iterations, then switch to "parallel", that solved my problem. Hope it works for you~

Cheers

Thanks. yeah I do the same. I run it for a bit on serial and then I switch it to parallel and it runs for a while, but it still crashes after a little while. Have you had the same problem ?


All times are GMT -4. The time now is 22:33.