code_saturne User's Forum

Hello,

Dear I am having problems with my mesh for a simulation of flow on a sphere, the case diverges. .

The number of courant reaches a value of 600 and the visualization in paraview tells me that the problem with the mesh is in the vicinity of the sphere, I have tried to follow several recommendations to solve the problem and I can not find the solution to this problem, I have made different meshes and the one that gave me the least mistakes I will attach it so that if someone wants it I can see it, the mesher that I am occupying is salome.

reviewing the listing the warning tells me this.
Criterion 1: Orthogonality:
Number of bad cells detected: 0 -> 0%

Criterion 2: Offset:
Number of bad cells detected: 7822 -> 1%

Criterion 3: Least-Squares Gradient Quality:
Number of bad cells detected: 89 -> 0%

Criterion 4: Cells Volume Ratio:
Number of bad cells detected: 0 -> 0%

Criterion 5: Guilt by Association:
Number of bad cells detected: 0 -> 0%

any contribution will be very grateful.

Greetings.

Hello,

I use you mesh and the case run without diverging. The mesh is very coarse and the check quality criteria report some issues. I think that you have to reduce the time step in order to get a Courant ~ 5 or lower.

You can read some guidelines at https://www.code-saturne.org/cms/documentation/BPG.

Best Regards,

Luciano

Thanks Luciano,

I have occupied a time step of 0.08 and of 0.01, if I decrease it the simulation time increases too much, with a time step of 0.001 the simulation time is approximately 30 days for which this solution makes me unfeasible, proble with another geometry and a higher time step where the listing does not inform about any warning and in this case the simulation time is prudent with respect to what I need to visualize.

I'm testing with GMSH to see if the problem can be solved

Greetings.

I upload the listing.

Hello,

I attach an example of the flow past an sphere. I use the geometry that you shared and I have created a new mesh. This case with a coarse mesh (aprox 115.000 cellls) took arround 13 min in a laptop to solve 500 time steps with Courant max around 7. I added the listing file in the zip.

I hope that this case can help you.

Regards,

Luciano

Thank you Luciano,

The little experience I see the account apparently, launch the case and converge, my mistake maybe was to guide me a lot by the visualization in para view and not address the problem separately.

So far everything is going well and the simulation is delivering values ​​according to expectations.

Thank you so much.