[word]

Hello,
For my work, i have to emulate the flow of air around a bicycling helmet. For this purpose, we are using SALOME to mesh the helmet, with a mesh loking as follow. Capture d’écran 2023-01-20 113019.pngWe are then using simpleFoam to simulate the flow, with a $k-\omega$ SST model for the turbulences. The initial and boundary conditions are as follow :
- U = (1 0 0) m/s constant velocity at the inlet
- p = 0 (because it is incompressible it should'nt matter), constant at 0 at the outlet

turbulentKE0.0000375;
turbulentOmega5.103103630798289;




I'm trying to reach for 5 times y+ on the first cell height when meshing the boundary limit of the hemlet.


The mesh is still very coarse,and the quality is not ideal. See the log of checkmesh attached here mesh.txt.


Finally coming to my question : when i run simpleFoam with my params, it stops abruptly with no log and no warning after computing Uz during step 1:
Time = 1

smoothSolver: Solving for Ux, Initial residual = 1, Final residual = 0.0931717124567556, No Iterations 18
smoothSolver: Solving for Uy, Initial residual = 1, Final residual = 0.0911694495182643, No Iterations 18
smoothSolver: Solving for Uz, Initial residual = 1, Final residual = 0.0955307794068746, No Iterations 22


Does anyone have a clue about this ? Is this just a mesh issue ?

[alexvaleije]

Hi,
Reading your checkMesh file, it appears that the mesh has quite some issues in the (i think) most important fields, which are aspect ratio, where you have a maximum of 1e96, and some negative volume cells.

Usually, if some elements are incorrect, the solver tries to calculate but after the velocity iterations, crashes with a floating point exception in the pressure loop. It is weird that in your report doesn't appear the error and instaead of that just collapses.

My advice would be try to run the case again with a simpler mesh near the helmet on your first try to make sure that the other aspects of your setup are correct, and then add layers of complexity. I've never meshed with SALOME so I cannot help you with that.

Let me now if your problem is solved with a simpler mesh.

Regards,
Alex

Originally Posted by word

Hello,
For my work, i have to emulate the flow of air around a bicycling helmet. For this purpose, we are using SALOME to mesh the helmet, with a mesh loking as follow. Attachment 93065We are then using simpleFoam to simulate the flow, with a $k-\omega$ SST model for the turbulences. The initial and boundary conditions are as follow :
- U = (1 0 0) m/s constant velocity at the inlet
- p = 0 (because it is incompressible it should'nt matter), constant at 0 at the outlet

turbulentKE0.0000375;
turbulentOmega5.103103630798289;




I'm trying to reach for 5 times y+ on the first cell height when meshing the boundary limit of the hemlet.


The mesh is still very coarse,and the quality is not ideal. See the log of checkmesh attached here Attachment 93066.


Finally coming to my question : when i run simpleFoam with my params, it stops abruptly with no log and no warning after computing Uz during step 1:
Time = 1

smoothSolver: Solving for Ux, Initial residual = 1, Final residual = 0.0931717124567556, No Iterations 18
smoothSolver: Solving for Uy, Initial residual = 1, Final residual = 0.0911694495182643, No Iterations 18
smoothSolver: Solving for Uz, Initial residual = 1, Final residual = 0.0955307794068746, No Iterations 22


Does anyone have a clue about this ? Is this just a mesh issue ?