I'm trying to model something simple: a tank rotating (2.5 rev/seg)with water filling 75% of height at the beginning. I'm using SSG turbulence model(because I undestand that k-e is not useful in these cases), an hexaedric mesh and a standard free surface model. I'm having problems always. Or I get an error saying that the program can't find the "Effvisc" or after 15 iterations I get floating point exception-overflow (even using the upwind scheme). To fix the velocity I tried rotating the mesh and also choosing rotating walls, but I got the same errors. What could be wrong? Thanks in advance.

Hi,

RSM turbulence models are much harder to converge than the 2-equation models. Make sure you have the best quality mesh you can do, and use fine timesteps to start it off.

You might also start the simulation with k-e or SST and restart changing to SSG after the flow is settled down to make it easier for the SSG model.

Glenn Horrocks