Hello I am simulating the flow in a hydrocyclone. I am using RSM with a single water phase. I have tried to take into account all the recomandations appearing in the Fluent doc; developed profiles at the inlet, first computation with the k-e model until convergence, and low under-relaxation factors. I have also checked that Y* is consistent with my choice of standard wall function (between 10 and 47). The probleme is that I am unable to obtain convergence. The residuals are oscillating between 10e-2 and 10e-5. I can send to anyone a picture on the residuals behaviour. Thanks.

they are often very difficult to converge. Read documentation about convergece strategies for RSM. It's often usefull do some iteration with k-epsilon and after use RSM. (in a case I've done also 50 initial iteration with euler equations)