CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Gambit Meshing Problems (http://www.cfd-online.com/Forums/fluent/38012-gambit-meshing-problems.html)

Francesco B September 27, 2005 03:48

Gambit Meshing Problems
 
Hi all, I try to create a volume mesh in Gambit, this is not for me the first time I do this but...... Appear ever the same message: Inizialization failed; perturb boundary nodes and try again.... This is usually caused by problems in the face meshes. Check the skewnesses of your face meshes and make sure the face mesh sizes are not too large in areas of small gaps. What i have to do? Thanks Francesco B.

Jason September 27, 2005 08:33

Re: Gambit Meshing Problems
 
Well the first thing to do is to "check the skewnesses of your face meshes and make sure the face mesh sizes are not too large in areas of small gaps". In other words, if one of your faces has a highly skewed element, then the volume mesh would be an even higher skeweness and Gambit can't handle it. You can use the Examine Mesh tool (icon on the bottom right of the screen) or you can use the mesh check (it's literally a check mark icon in the Face Mesh commands menu) to find out the worst skewness and where it is. If it's over 0.9 then it's too high. Even over 0.85 is high, but it's possible. Remember, lower is better. If skeweness isn't the problem, then it's typically a problem with mesh size in areas of gaps. If you have two surfaces coming together, you have to make sure that the mesh size is small enough to capture the gap. Along the same line, if you have two faces coming together at an acute angle you can get this error as well. If the problem area isn't intuitive, then you might want to go back to when your geometry was real and split it into smaller volumes. This will help you isolate the problem area. Once you know where the problem is, you can work out a way to fix it.

Hope this helps, and good luck, Jason

suti September 30, 2005 04:28

Residuals for transient case
 
Hi all,

Is this true that when running fluent for a transient case, I should run each timestep until all residuals converge AND become more or less flat (ie. slope is almost zero). Can I just run each timestep until all residuals converge but not necessarily become flat? thank you.


All times are GMT -4. The time now is 21:45.