Problems with VOF solver

Benjamin

Hi everybody,

I am trying to model the waves of a yacht sailing upright using the VOF multiphase model and, as explained in the tutorials, I used the geo-reconstruct scheme as it is said that it has a better accuracy in predicting the shape of the waves. I struggled for several weeks in both 2 and 3d to get a solution to converge (the turbulent viscosity ratio was too high). Varying the mesh density and type did not change anything. Only a laminar solver would converge. Finally, I tried the two other schemes (implicit and Euler explicit) and did not had any trouble to get my results to converge with any turbulence shceme. Does anybody know why changing the scheme of the VOF affects the turbulence viscosity? How do you guys do to model correctly a free surface? What accuracy can you expect on the waves pattern? Do you use Gambit or some other more used friendly mesher?

Thanks very much,

Ben