Rob, too much latent heat making the solution unstable, could it also be in case when increasing physical time of the simulation?

Using EWF for physical time of 50s, solution converged and residulas looked stable for both time step sizes of 0.005 and 0.001 seconds. Results were pretty much the same.

When increasig physical time to 150s with 0.005s for time step size, residuals are fluctuating at the end and I am looking to see if that would make the results unaccountable?