[aeroujj]

I am doing a project in which I want to model flow in a device with a centrifugal pump fitted to cause all the flow. The device also has openings to the atmosphere. I have to basically model the effect of the pump on the fluid and not the insides of the pump. I am a newbie to fluent but I have simulated for the same project in Autodesk Simulation CFD also. I came to fluent because my institute has a licence for it and also fluent has axisymmetric swirl model available(mine is this case). In Sim CFD I used to define the pump domain MATERIAL as a centrifugal pump. Also I only had to give mass flow rate input. In fluent there is no such thing as defining the material as "centrifugal pump" so I suppressed the geometry and applied velocity inlet BCs to the corresponding edges. But this causes the continuity residuals to stay large(round off errors in velocity inlet input). Is there a better way to do this? Will there large continuity residuals really cause that much trouble or have negligible effect? Also my model runs fine with k-omega SST model but diverges for the reynolds stress model(fluent manual suggests RSM for high rpm swirl flows), so is it due to this velocity inlet thing?

Please help