CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Turbulence model (http://www.cfd-online.com/Forums/cfx/18538-turbulence-model.html)

pop May 30, 2001 11:36

Turbulence model
 
I need help modelling a low Re flow. My fluid domain has two zones one with Re almost 3000 near the inlet and the other part is lower than 1000. So now which model to use and what are the mesh restrictions I have.

I am using CFX 5.4.1

Thanks

Pop

B Malone May 30, 2001 13:49

Re: Turbulence model
 
If the Reynolds numbers are really this low start out with a laminar model. Use the default mesh and see if it converges. Change mesh parameters to double the number of nodes and run again. Perhaps do this again. Are the solutions the same?

Try decreasing mesh size near areas where the residuals are the largest. Try to get at least 20 to 30 nodes in between all surfaces.

Check the results against any experimental data you have. Without supporting experimental data I don't trust it.

It will not hurt to try a turbulent solution. But 5.4 does not have a low Reynolds number model. You may need to look at version 4.

Vitidphan May 30, 2001 15:03

Solid-liquid flow CFX-4 or CFX-5
 
Now i have to study solid in liquid flow model. Please tell me, how do the diffence between using CFX-4 and CFX-5 for this type of model. Thank you

Jan Rusås May 30, 2001 16:31

Re: Turbulence model
 
Dear Pop,

I really do think you should investigate some of the Low Reynolds number models (in CFX4 ! if you can) But be careful with your mesh at the walls, y plus should be 1 or 2..3 depending on model. It is described in the manual under wall models.

Regards Jan


pop May 31, 2001 09:44

Re: Turbulence model
 
Dear Jan, Thanks for your reply, however I do not have CFX4!!! There is no way in CFX5.4.1? Thanks

Pop


All times are GMT -4. The time now is 14:55.