Internal face to boundary condition ?

[jgerardin]

I use a displacement of internal face in order to avoid negative volume on first cell. With moving wall (by fixed velocity), only the boundary condition moves and so at a moment, the boundary condition moves enough to be at the same place of the first internal face (which can create negative volume if boundary condition moves behind the first internal face).
As my cells are originally orthogonal and keep the same orientation for the moving face, I obtain then a new orthogonal cell with different volume. So I seems to keep good quality.

I'll give you more feedback later!

[jgerardin]

I've got some new questions about my problem.

I simulate the degradation of a solid (yes, it's strange to use a CFD code to simulate a solid, but I have the habit of Code_Saturne). So, in my solid, there is no velocity. Till I don't modify the mesh, I have no velocity and no pressure inside my solid. But when I start moving the mesh, a velocity and a pressure appear inside my solid. I try to put high viscosity (1e+20) but it remains a very small velocity (something like 1e-9).
Is there a way to not calculate the flow, or clipping the value of velocity/pressure to 0?

[Yvan Fournier]

Hello,

You can clip all you want in cs_user_extra_operations.f90, but it might not be sufficient.

The same, you can set iconv(iu), iconv(iv), iconv(iw) to 0 to deactivate convection for velocity (and idiff or idifft to deactivate diffusion or turbulent diffusion), but I am not sure this will work, as velocity-pressure coupling may still occur, and the fact you are using ALE might lead to other terms being accounted for automatically.

So you might want to try either one of these options, but I'm not sure of the results (I would try deactivating convection first). If none work, we'll try to think of something else.

Regards,

Yvan

[jgerardin]

Thank you for your answer, I'll try that and give you a feedback.