Hi all,

I am wondering if someone can tell me what turbulence model is best used for general automotive applications, and also race car application. I know SST is a suitable one but I have heard K-Epsilon is a good suitable one too.

Also, what are best values to set the Turbulence varibles for general/avarage solutions

Thank you in advance Russ

Hello,

Actually the SST is a mix of k-e and k-omega.

Your first question is difficult to answer without the following: What demands on accuracy do you have? And what type of flow is to be studied? And how much time is available?

Question two depends on what model is chosen. In general it's best to try the standard values incorporated in the cfd code (assuming you are using a commercial solver). If you are writing you own code then almost all basic cfd books list the values used in k-e for instance.

Here is what Launder and Spalding suggested:

Cmu 0.09 C1 1.44 C2 1.92 Sigmak 1.0 Sigmae 1.3

B.E. Launder and D.B. Spalding "The Numerical Computation of Turbulent Flows", Computer Methods in Applied Mechanics and Engineering 3, 1974, London, pp. 269-289

Best Regards

Dear Mordorgie,

Thank you for your reply, I never know that SST incorporated both turbulence models. So would you say this is the better tubulence model choice for external flow analysis of automotive models.

The demand on accurancy is just to give a reasonably accurate solution in which the results can show good understanding of the flow field. The flow is just a standard velocity flow which I have been told that a good setting for turbulence intensity for a outside flow is around 4%. The amount of time I have is resonable, but obviously I dont want to be spending 50 years on one solution!!! hehe

I am indeed using a commercial code

thanks again in advance Russ

Is the automotive industry interested large eddy simulations(LES) besides RANS modeling or LES is just still in the academic stage? Any idea how soon LES will go to the industry?