CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Low-Reynolds-k-epsilon model (http://www.cfd-online.com/Forums/fluent/28424-low-reynolds-k-epsilon-model.html)

Chris April 9, 2001 12:24

Low-Reynolds-k-epsilon model
 
Hi

I'm doing a project on wall bounded flow using Fluent 5. I would just like to know one thing, Does the standard k-epsilon model with the full low-Re-model as the near wall treatment work well for Low Reynolds number? B'se i have tried everything possible but i'm not getting good results with the Low-Re-ke model. Infact fluent has also suggested using using the RNG models for low Reynolds number and i'm getting better reults with that. Can u explain why, if u can. I'm particularly comparing these models by plotting the y+ vs U+ values comparing with exepriment and what the standard Low Re-ke model by Lanuder and Sharma is giving me, is a good profile in the viscous sublayer region and when it gets in the buffer and fully turbulent region, U+ values reduces and y+ increases. So i'm thinking the problem might be with U_tau, therefore going to the wall shear stress.

Can u help me figure out what i might be doing wrong?

Thanks

John C. Chien April 9, 2001 13:44

Re: Low-Reynolds-k-epsilon model
 
(1)."Can u help me figure out what i might be doing wrong?". ??? (2). It's better to say "Can u help me figure out what Fluent might be doing wrong?" (3). As a matter of fact, you should ask "Can Fluent give you the accurate solution?" long before you start using the code. (4). You will have to ask the vendor's support engineer, if you are interested in getting good answers. (we don't know what is in the code.)

chris April 10, 2001 09:34

Re: Low-Reynolds-k-epsilon model
 
Thanks for your correction, John but how can I contact the vendor's surppot engineers i'm really interested in getting good answers. It's part of my project acessing the k-epsilon model types provided by fluent.

chris

Scott W April 10, 2001 13:29

Re: Low-Reynolds-k-epsilon model
 
I will help guide you to your answer, but I'm not an expert in turbulence. The Fluent training manual (I assume either you or whoever licensed Fluent went to the training to get the training manual that Fluent 'neglected' to send its customers) says the ke model is only applicable for fully turbulent flows (ie. low Re numbers are not too good). In addition, the training manual states that the standard wall functions or nonequilibrium wall functions are poor for low-Re effects; however the two-layer zonal model is supposed to be apoplicable to low-Re flows.

chris April 11, 2001 14:50

Re: Low-Reynolds-k-epsilon model
 
Thanks for your reply

Your are right, the wall function and nonequilibrium wall function perform pooly for low Reynolds effect. The two-layer zonal model works well but I Would like to know if it solving for the two equation model for kinetic energy and dissipation?

Fluent also allows you to define the Low-Re-model by Launder and Sharma for the near wall treatment. Can you tell me more about low-Re-model in Fluent?

Thanks Chris

Xiao Hu April 13, 2001 17:11

Re: Low-Reynolds-k-epsilon model
 
For the two layer zonal model, the whole domain is divided into two regions, i.e. turbulent core region and viscosity-affected region. For the turbulent core region, high-Re k-e model is used (you chose this model). For the viscosity-affected region, one equation model of Wolfstein is employed.


All times are GMT -4. The time now is 16:02.