[nono91]

Hi guys
I’m performing CFD analysis of regenerative pump and my result of pressure that the pump at given flow rate can make is app. half of that what it according experiment should be.
I tried to perform the simulation using both Fluent and CFX, Frozen rotor method, turbulence model K-epsilon (lots of people according various articles are using successfully this one) also tried RSM, but result are almost the same, half value of experiment. I would like to try resolve boundary layer but I’m limited with number of elements and I’m afraid that the number of elements would grow greatly….
Mesh is hexahedral and was at beginning 1M elements, now I’m calculating with 6M elements but doesn’t seem to help…
Does anyone have done similar simulation of regenerative pump and have had similar issue? Or does anyone has any advice what could I change? The flow is highly swirling so RSM does make sense but no improvement achieved….
THX

[JBeilke]

The first thing is, that you have to run it transient with rotating mesh.

You can find some examples of similar (side chanel) machines here:

www.beilke-cfd.de

[nono91]

Yeah, I know that could be the way but as I've read couple of research papers almost no one have performed it transient but steady-state and according them the agreement between experiment and simulation was very good. That's the reason why I was asking if I'm possibly doing something wrong
Anyway thanks for the link I'll look at it

Edit: The papers from you seems to be very useful, Thanks a lot!

[zts818]

Hello. I'm facing similar issue you had when running CFD for a regenerative pump. Did you solve the problem?