I heard of that Star-CD can simulate isolate blade row flow-field with good confidence,but behaves poor quality in simulate multi-blade row. Does anyone know whether is it the problem of user or code?

Thanks

Hi xjz,

I am not sure that I know what your question exactly is about. I use Star for performance prediction of centrifugal pumps with good results. They are normally modeled with everything including impeller, casing and outlet by means of the rotating reference-frame option (approx. 500k tet.cells). I have played a little bit around with the turbulence models and ended up with the ke-RNG. It was very hard to get the residuals down until I applied a little bit of surface roughness (Ra=0,8). Fore some reason the solver doesn't like it when I use the standard Elong 9 function for smooth walls.

Best regards Kenneth R Jensen

Regarding modelling of a centrifugal pump using Multiple rotating Frame methodology.

I am using Star-CD to model the water flow in the cooling jacket of a 4 cylinder diesel engine. The inlet junction to my model is formed by the outlet from the water pump impeller into a 'volute shaped entry junction feeding the engine block. So far i have not attempted to model the impeller.

I have understood that the MRF methodology is accurate when assuming circumferential uniformity - which holds true when the rotating region(the impeller) is physically remote from any stationary(stator) type components. This is not the case for a centrifugal pump where the 'tongue' of the volute surrounding the rotating impeller provides a profoundly circumferentially non-uniform interference of stationary and rotating region, especially true when the pump is operating off-design point.

Hence only an unsteady, (albeit periodic with blade passing) analysis can properly capture the pressure field around the volute.

....which requires a transient sliding mesh model..and is why i haven't yet had the perseverence yet to try to model the impeller! Any advice / discussion would be welcome!

If you use sliding mesh, you'll get a good result. The selection of model is also very important.

I worked on it long before.

Hai to CFD world I have a domain which has both rotating and Stationary parts. I would like know which solver ,relaxation, differencing scheme will give better convergence. I solved the same using STAR. The convergence was not good. Can you help me.