[Christoph96]

Hi,

I am currently working on a two-phase system with interFoam. It is a water flow over a cavity filled with air (Re = 25). The walls of the cavity are hydrophilic (where the water should be) and hydrophobic to avoid a water inrush into the cavity. So far, this configuration works fairly good for phase fraction, but the velocity field causes problems. The cell direct under the interphase at the wall has a velocity that is much to high (about factor 100) That is why this cell disturbs the complete velocity field in the surrounding.

If I replace the dynamic contact angle by a zeroGradient the velocity field is fine, but even with constant contact angle the problem occurs.

I have already tried different discretization methods of the transient term (Euler, Crank Nicolson), changes in the solving process of alpha, extreme low Courant numbers (Co = 0.01) and changes in grid resolution. Unfortunately, nothing led to an acceptable result.

If I increase the Reynolds number (20000) and use a turbulent model, the problems do not occur. This might be due to higher inertial forces which dominate the numerical errors.

Does anybody have an idea to avoid these artefacts?

Kind regards,
Christoph

Interface of the case represented by the white line (above water, below air):
interface.png

BlockMeshDict:
blockMeshDict.txt

Alpha:
alpha.water.txt

fvSchemes:
fvSchemes.txt

fvSolution:
fvSolution.txt

[axjjienn]

Can you provide some additional information about your case? I am not sure I understand what you are trying to simulate, as water will rush into the cavity under action from gravity.