I have simulated a frustum FB using MFiX. It starts well and run for thousands of iterations until it diverges due to the error of velocity higher than maximum allowed velocity and non-converging P0 residuals. I have checked different mesh sizes, ICs and under-relaxation factors but all of them would only change the time of divergence a bit.
Can anyone give me a hint on this case? @jeff.dietiker ic.mfx (13.9 KB)