CFD Online Discussion Forums

CFD Online Discussion Forums (
-   CFX (
-   -   Missing Information in SST Transition Model Formulation (

Josh September 13, 2010 18:42

Missing Information in SST Transition Model Formulation
Good day, all -

I'm working on a paper and simulations requiring the gamma-theta transitional SST model. In the derivation of the model in the manual just after Equation (155), it reads:

The model contains three empirical correlations. Re_theta_t is the transition onset as observed in experiments. This has been modified from Menter et al. [101] in order to improve the predictions for natural transition. It is used in Equation 151. F_length is the length of the transition zone and goes into Equation 143. Re_theta_c is the point where the model is activated in order to match both Re_theta_t and F_length; it goes into Equation 146. At present, these empirical correlations are proprietary and are not given in this manual.

My problem is with the last line: "not given in this manual". Unfortunately, they are not given in the referenced paper, either (Menter et al., 2005). They are given in the original derivation (Menter et al., 2004), but, as noted above, they've been modified "in order to improve the predictions for natural transition". I found a relatively new paper ("Correlation-Based Transition Modeling for Unstructured Parallelized Computational Fluid Dynamics Codes", Langtry and Menter, 2009) that gives the empirical correlations, but I'm unsure if these are the ones that are used in CFX.

Are the correlations in the aforementioned paper (Langtry and Menter, 2009) the ones used in CFX? If not, does anyone know where they can be accessed?


ghorrocks September 14, 2010 03:28

I am not aware of the 2009 paper you refer to so I cannot comment about what it contains. But my understanding is details like this are not published to protect CFX's IP. CFX is a commercial product and they want to make money out of it, so they give enough details so you get the general idea but not enough information so you can write you own software using their algorithm.

Josh September 14, 2010 11:49

Thanks, Glenn. I figured that was the answer, but I really didn't want to hear it.

All times are GMT -4. The time now is 20:00.