problem of convergence
 

Hello everybody !

I have to study the oscillations of a liquid in a U-tube, so a two-phase flow (air / water liquid).

When I use a mesh without a boundary layer, the vof model with geo reconstruct and a small time step seems to work well.

But in my case, I'm studying the influence of friction factor in transient flow, so I need to mesh the boundary layer. I've tried to decrease the under relaxation factors, even the time step but after some iterations (thousands), my simulation always diverges sharply. And I can't reduce too much my time step since the phenomenon lasts approximatively 100 s.

I hope that someone can help me with my problem.

Thanks a lot.

Jean Philippe.