ALE calculation suddenly stopping with no error

Questions and remarks about Code_Saturne usage
Forum rules
Please read the forum usage recommendations before posting.
Post Reply
daniele
Posts: 77
Joined: Wed Feb 01, 2017 11:42 am

ALE calculation suddenly stopping with no error

Post by daniele » Wed Jun 12, 2019 11:26 am

Hello,

I am performing an ALE simulation. The simulation stops after a few iterations (much less than the ones specified), but no error is displayed in the listing nor in the other files.
In the listing, after the last iteration, the "Final stage of the calculation" and "End of calculation" messages appear.
The values of the mesh displacements and of the fluid variables are consistent, no divergence in these values is observed.
Looking at the mesh once the simulation has stopped, I do not see any negative volume cell or other problems in the mesh that could cause the simulation to stop.
The meshing is purely hexa (ICEM).
Do you have any idea about what to check in order to understand the cause of the end of the calculation?

Many thanks in advance.
Best regards,
Daniele

Luciano Garelli
Posts: 230
Joined: Fri Dec 04, 2015 1:42 pm

Re: ALE calculation suddenly stopping with no error

Post by Luciano Garelli » Wed Jun 12, 2019 1:37 pm

Hello,

Could you post the listing and some additional information about the case?

Regards,

Luciano

daniele
Posts: 77
Joined: Wed Feb 01, 2017 11:42 am

Re: ALE calculation suddenly stopping with no error

Post by daniele » Tue Jun 18, 2019 4:30 pm

Dear Luciano,

Thank you very much for your help.
Actually, looking more in detail to the listing, I found the message "Abort due to the detection of a negative control volume.".
Do you know how to visualize negative volumes in Paraview? I cannot manage to do it.

Thank you very much in advance.
Best regards,
Daniele

Yvan Fournier
Posts: 2787
Joined: Mon Feb 20, 2012 3:25 pm

Re: ALE calculation suddenly stopping with no error

Post by Yvan Fournier » Tue Jun 25, 2019 10:10 am

Hello,

When this happens, the computation may abort before generating a dedicated visualization output, so we need to improve this.

You can re-run the computation with a slightly lower number of iterations to try to see where the volume becomes smallest (whre cells become flattened).

Also, to avoid this, using a higher mesh viscosity or a smaller time step may help. Robustness should be better with the upcoming CDO version (an option in version 6.0), but for the current stable version, ALE can be a bit "fragile" regarding mesh distortion.

Best regards,

Yvan

Post Reply