Centrifugal compressor impeller+vanless diffuser simulation problem.
 

Hi to All

I'm doing the steady state analysis of an Centrifugal compressor with vaneless diffuser. I've done the simulation with solving the entire domain in rotating reference frame and set the shroud and hub of diffuser as counter rotating wall.

Here my problem is, the results indicated that the Total Temperature and enthalpy decrease form the impeller TE Cut to the diffuser outlet, the Tt difference is above 5K. And since the Tt decrease, the polytropic efficiency of impeller &diffuser is even higher than the impeller alone. So the results is definitely inaccurate.

The turbulence model is SST and wall yplus is around 3. Inlet boundary is Pt and Tt, Outlet boundary is average static pressure(results of mass flow rate outlet still have the problem).Alternate Rotating Model is on. I have change the grid and turbulence model, but the phenomenon still exist. I'm stuck here several weeks, wish somebody can help me. Any suggestion? Thank you~

Hello, Do you use TurboGrid and CFX?
Thanks

Yes, V15.0

I have tutorials about a radial turbine..I am not sure if they will help you or not..but you can contact me on 12317@eng.asu.edu.eg to send me the details of your problem...Thanks

Please post the inlet-to-exit averaged plot of your concern.

Have you tried KE model? What turbulence model have you tried? How is the setup of heat transfer? Are you using the option of Total Energy?

Here is the Tt and T of 50% spanwise. About 0.7 of streamwise location, is the TE cut, and the Tt decrease in the diffuser.
Attachment 54940

Tried k-e and RNG k-e, still the similar results. Total Energy is used and all the wall are adiabatic.

Try (R1+S1) domains with the impeller rotating up to an immediate downstream of the trailing edge. What you simulated is "the rotating diffuser" case.