CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Autodesk Simulation CFD (
-   -   convergence criteria (

danigorgeos December 3, 2012 13:13

convergence criteria
Hi all

I'm starting with Autodesk Simulation CFD an I've some questions

Which are the convergence criteria?

I thinks it's that residuals go down E-03 like ANSYS, but not sure.

Another one. Can I change those covergence criteria?

Thanks January 4, 2013 12:13

RE: convergence criteria
The convergence criteria for Simulation CFD/CFDesign are not exactly what you would expect coming from the ANSYS/Fluent side of things. You canset and adjust convergence criteria in the 'Solution Control' interface.

1) Under the Setup tab, click Solve to bring up the Solve dialog and click Solution Control.

2a) There are several options available here. The first is intelligent solution control. This is a nice feature as it automatically adjusts analysis parameters to help control the solution and avoid divergence in unstable models. If you enable this and click the Advanced button it will bring up a dialog with several sliders. If you click edit individual it will allow you to change the residual settings for several parameters. Keep in mind these are not the same ones you would set in ANSYS/Fluent.

2b) You can also adjust the relaxation factors for various parameters in the Solution Control window. These are helpful for models which are unstable. Typically a setting of 0.5 is used, but that is not always ideal.

2c) To set the Pressure and Temperature residuals you see at the right of the Solution Control window you must be running a compressible analysis. For incompressible flow these are turned off by default.

Hope this helps.
Matt January 4, 2013 12:15

RE: convergence criteria
Also, I would direct you to the help files. These have good discussions on all of these settings. Pay close attention to the advection scheme you are using as this can affect results and stability of your analysis. Again, the help files can explain which advection scheme is appropriate for your analysis.

danigorgeos January 4, 2013 17:52

Thanks you very much Matt.

Your comment has been really helpful :) :)

All times are GMT -4. The time now is 03:53.