CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   problem with convergence (https://www.cfd-online.com/Forums/fluent/113789-problem-convergence.html)

S1m0n1 February 26, 2013 14:46

problem with convergence
 
I am having trouble getting my residuals to converge to the specified criteria. At first I set it to 1e-06 and did over 10000 iterations and it didn't converge. Then I tried 1e-04 and it didn't converge. Now I am trying 0.01 just to see if my solution will converge and still it hasn't. It seems like anything I try with any example or problem I have done in Fluent my solution never converges.

The x,y,z velocity fluctuate around 1e-2 and currently I am over 2600 iterations. Does anyone know why my solution may not be converging?

diamondx February 27, 2013 11:01

how about your mesh ???

flotus1 February 27, 2013 11:34

Without ANY information about your setup, how are we supposed to guess what causes your problem.

A CFD free user February 27, 2013 13:09

As Alex said, you can't definitely speak about convergence without have any information regarding to problem setup, but as I've experienced many times, most of the problem in starting a simulation is due to mesh quality, except that you apply proper physical properties and boundary conditions. So, my first advice is that you check the mesh quality and existence of any probable undesired wall especially in your flow path.
Hope it could be helpful

S1m0n1 February 27, 2013 15:29

Ha. Sorry I didn't provide much information. I was just looking for some general answers as why problems usually don't converge.

My mesh is 0.1m. I know this is really big but I was just starting out big and goingto work my way down.

So I guess my question is, if my mesh is too big will my solution never converge?

Thanks for everyones replies!

A CFD free user April 11, 2013 04:46

Maybe you'd better use a finer mesh, but, as far as I know you wouldn't face a dead problem as long as your simulation involves only hydrodynamic, though the accuracy simulation is low. However, mesh independency analysis reveals enough number of cells for your case but for species transport or reaction you need to use a very finer mesh.

jwillie2000 April 11, 2013 08:56

What type of problem are u looking at? The physics of the problem is important in addition to the mesh resolution you are using. Throwing more light might help us help you. Thanks! Jimmy


All times are GMT -4. The time now is 13:14.