[sam1364]

Hi everyone,

I am trying to simlulate flat plate boundary layer problem in OpenFoam. the flow is turbulent (Re=5*10^6). I want to use k-epsilon model to solve the problem.
because I need to use wall function in this case, I need to have my first yPlus in the gird at a value greater than 11. So, because I need many points in the log layer region, I have used a very fine mesh right after the first point in my grid. this makes a very high aspect ratio between the first cells (y+=11) and the above cell (where fine mesh starts). as a result, the simulation will be stopped because the courant number goes high suddenly.

Do you have any suggestion how I can fix my problem?!

Thanks,

[SirWombat]

To better understand your problem, you migth wanna post an image of those cells. How did you create your mesh (ie snappyHexMesh). What about checkMesh, does ist give you an OK? Have you tried viewing your problematic cells in paraview? Are you using prism layers?

[sam1364]

I have created the mesh using gambit. OpenFoam said ok to my mesh.

Besides, I have always doubted about my numerical schemes. I would be grateful if you could have a look at my files and give any suggestions that might be helpful for me in the future.
I have included both my pisoFoam and simpleFoam numerical schemes.

[hitzhwan]

Originally Posted by sam1364

I have created the mesh using gambit. OpenFoam said ok to my mesh.

Besides, I have always doubted about my numerical schemes. I would be grateful if you could have a look at my files and give any suggestions that might be helpful for me in the future.
I have included both my pisoFoam and simpleFoam numerical schemes.

Have you solve the problem， I also met the problem？