I am currently using Fluent 5 to look at the 2-D flow over an airfoil. I am having a hard time getting good drag data, in comparison with xfoil or mses. Checking the pressure contours between the codes compare favorably, however the shear stress plots do not. The shear stress plots for Fluent are giving higher values then the other codes by a magnitude of about 2. I was using the S-A viscous model, and I am wondering how Fluent determines the transition point? Is it free transition, or do I need to fix a transition point? And if so, how do I fix a transition a point in Fluent? Any information and help would be greatly appreciated.

Rob

S-A is a fully turbulent model, i.e. the whole flow is turbulent and there can be no transition in Fluent. The full S-A equation does contain a transition term but I do not think fluent employees this term. However, I could be wrong and would appreciate the notice if you find out.

Andy

Thank you for the response Andy. I thought that might be the case with the S-A model in Fluent. Do you know if any of the viscous models in Fluent allow for transition? And if so, any help about how to specify a transition point would be appreciated.

Thanks...Rob

I don't think any of the turbulence models in Fluent are implemented with transition modeling in them. To fix a transition point you will probably have to write an UDF to turn off the turbulence model in a certain laminar region. I have never done this myself in Fluent, but I have a vague memory that this was discussed on this forum some time ago - search the archives (use the site-wide "Search" in the top right corner and limit your search to Fluent Forum).

A guy in an office down the hall from me tried something like that. As I remember from his presentation, he got poor results with it. He was looking at a 2-D turbine blade. I think he ran it laminar first to locate a transition point (I think he said that this is not really the transition point but he had to do something), then he partitioned his grid based on what he saw and turned one region into a turbulent region the other was laminar. He was most interested in entropy generation and so I suppose it was in that context that he said the resuts were poor.

Thank you everyone for your help and advise. I did as Jonas and AnMadh suggested and partitioned the grid into a laminar and turbulent zone. Thaks to that I am getting much better results all around. The Cp and Cf plots agree very well with MSES for the most part and the separartion points are pretty close also. Thank you.

Rob

[Younis saleh]

Hi everybody
please, can you help explaining the way to separate the grid to laminar and turb. ??

[Younis saleh]

hi Robe
I need your help.
how did you make the separation of the grid

[Younis saleh]

Originally Posted by Rob
;99374

Thank you everyone for your help and advise. I did as Jonas and AnMadh suggested and partitioned the grid into a laminar and turbulent zone. Thaks to that I am getting much better results all around. The Cp and Cf plots agree very well with MSES for the most part and the separartion points are pretty close also. Thank you.

Rob

hello dear Rob:
can you explain how to partition the grid into a laminar and turbulent zone