CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   Turbulence model in rotating system ? (http://www.cfd-online.com/Forums/main/4194-turbulence-model-rotating-system.html)

Li Yang December 7, 2001 07:46

Turbulence model in rotating system ?
 
Dear all,

Could anyone tell me how should I take into account the effect of a rotating frame on turbulence models (from zero equation to two-equation model) in transonic flow ?

Thank you very much in advance.

Li

Wang December 11, 2001 07:05

Re: Turbulence model in rotating system ?
 
Hi,

You should modify the model constants because these model constants are based on the plate flow for two-equation. For curved flow it is unsuitable.

So, this way can also be called as the correction of the curved surface. It is original from Launder.

Li Yang December 11, 2001 07:36

Re: Turbulence model in rotating system ?
 
Hello Wang,

Thank you very much for your reply.

I hope you know that the governing equation on a rotating frame has two different formulations: one is using relative velocity and the other using absolute velocity as variables. I would like to know what the difference the turbulence models will have when they are used in different formulations. Do you know any reference papers ?

Best regards

Li

Wang December 12, 2001 10:52

Re: Turbulence model in rotating system ?
 
Hi, Li Yang

I have no experiences on rotating case. However, I feel that your problem is not related to the turbulent model.

A rotating frame should be same as the turbulent swirl flow. That is to say, when the rotating disk is rotating, the flow will be driven by anisotropic stress, which is deviated from the two-equation turbulent model. This deviation is indepedent on co-ordinate system. So using both velocities should get same results. If you have some errors they are not from the turbulent models but others.

Abhijit Tilak December 18, 2001 20:50

Re: Turbulence model in rotating system ?
 
Well there r several approaches best ...one solve for RSM next best ASM but if ur hell bent on using two-equn turbulence model there is a model by Launder called cubic eddy viscosity model appeared in i guess int jrnl for heat & mass transfer in 1997. other by S.V.Patankar under title "coriolis modified k-e model" don't remember where & when it appeared. These flows are very ugly flows to predict & people still keep applying isotropic models to these flows.... i guess nothing's better around. Good Luck Abhijit


All times are GMT -4. The time now is 22:49.