Rotating Frame Stationary Vaneless Diffuser?
 

I have a centrifugal impeller with a vaneless diffuser. It is simple in turbogrid to create a mesh that encompasses both impeller and diffuser with separate hub and shroud surfaces for the impeller and diffuser sections. My question is since it is one mesh, I have to solve the entire domain in the same reference frame. Is it possible or advisable to solve the entire domain in the rotating frame and set the diffuser hub and shroud surfaces as counter-rotating walls? This is not just an abstract question as I have already done many of these simulations like this and am now questioning the method.

Thanks,

Kevin

Yes, this is an acceptable method. In fact I would even say that if it is appropriate I would say it is the recommended method as it means you do not need a GGI in a critical part of the flow.

To be sure you should do a simulation with and without the frame change and check they are equivalent.

Thank you for your reply. Since I posted this yesterday I did a simulation with a split rotor and diffuser for comparison and the results look very similar. The metric that I am concerned with in this simulation was withing 0.5% which is fine for my application.