turbulence model for ship hydrodynamics

ettax

Goodmornig.

I've tried to simulate a virtual towing tank with Star-CCM+. If I use a laminar model, solver work very well, free-surface is good and simulation converge, but forces aren't correct because I don't resolve viscous component.

If I switch to turbulent model (k-w or k-e) for computing hydrodynamics forces, solver diverge.

What can I do?? Mesh is good, y+=50 for resolve boundary layer with wall function.

Thanks for help