unconditionally stable VOF calculation

chriss

I am having trouble getting a VOF simulation to converge. The problem is water sloshing in a tank.

The tank moves at some acceleration and to simulate that I change the gravity vector. After 50 to 100 time steps at 0.0025 time steps the problem suddenly looses convergence.

The gravity profile is a ramp-up and ramp down with the maximim at about 7.5m/s/s. If I run a step function with the same maximum everything works fine the the problem runs to equilibrium

So the question is, is there a way to make the fluent algorithm unconditional stable, or at least more stable than it is using PSIO and body-force weighted pressure descretization.