CFD Online Discussion Forums

CFD Online Discussion Forums (
-   CFX (
-   -   Centrifugal pump analisys (

pump_passion July 2, 2009 04:14

Centrifugal pump analisys
Dear all,

is there anyone experienced in centrifugal pump simulation?

I would like to have your opinions about performing a staedy state simulation (frozen rotor algorithm) rather than a transient simulation for the prediction of Q-H curve.

What is the best way to proceed? Thanks


pump_passion July 2, 2009 11:48

In particular, i would like to know if some phonemona will be misse performing a steady state simulation rather than a transient simulation.

The simulated performance curve with a steady state simulation doesn't match the test performance curve.

I am wondering if there is something tha i do not understand and if a transient simulation would give better results.

Thank you very much.


ghorrocks July 2, 2009 18:31


The answer to that is fairly obvious - if there are transient flow features in the flow then a steady analysis cannot capture it. Also a frozen rotor approach is done at a particular angle - as the blades rotate the performance may well change significantly as it passes features in the volute.

Glenn Horrocks

pump_passion July 3, 2009 02:56

Glenn, thank you very much for your reply. I agree that there are transient flow features in the flow that a steady analysis cannot capture. Concerning the frozen rotor approach, i always run different simulation changing the impeller angular position and the head is obtained considering the average of the values at different position.

But i don't understand why ANSYS promotes the use of frozen rotor or stage in a steady state simulation when they should know that this is not the right approach to build the Head-Flow performance curve.

Thank you very much again.


sans July 3, 2009 05:09

Hi, I dont think ANSYS is biased towards frozen rotor simulations, they have left that choice to the user. Frozen rotors simulations for pumps are fairly accurate from my experience. Whether one does a steady or transient simulation for turbomachines comes down to time, resources and accuracy.

All times are GMT -4. The time now is 00:11.