Hello,

I am running a large case in segregated solver and keep getting a warning saying "turbulent viscosity limited in xxx number of cells". This number grows pretty constantly for each iteratrion step until i get a floating point exception. The flow field was initialized from a previous run with a slightly different geometry. So there might be cells with impropriate initialization. I tried adjusting the relaxation factors but it just slowed the divergence down a bit. What could cause this problem and how could I solve it?

Thanks

Troubleshoot by plotting turbulent viscosity and see where the extreme values lie. If it's at the location of modified geometry, then try clearing the solution and re-calculating.

Are you ramping CFL #? It might be already ramped up after the 1st simulation, so reset it to a lower value after replacing geometry.