I am currently working on implementaion of a non-linear k-eps model in an in-house code (a FVM code) which was originally written for linear EVM models. I can calculate Reynolds stresses pretty well with the explicit ASM model when I am still using linear EVM. It was just a check for me to see if there was any problem with Reynolds stress calculation. It showed that there is not. But when I try to put gradients of those calculated Reynolds stresses as source terms in the momentum equations and skip eddy viscousity concept, I got ckeker-board problem and instability. I have been familiar with the problem and so I tried Obi-Peric-Scheuerer's method (AIAA J., Vol. 29, No.4, 1991, pp. 585-590), but I have still the same problem. If there is anyone out there who actually done this kind of coding work, I will be most grateful if he/she shares some of his/her experiences with me.

Hello! I have done some coding of non-linear k-epsilon models into the commercial CFD code CFX4. I implemented it into the standard k-eps treatment of CFX, separating the stress-strain relation into two parts. The first part is the standard linear term (already implemented in the code), the second part is the HOT (higher order terms, which can be quadratic or cubic or whatever). After calculating the HOT in the momentum equations, the appropriate gradients of these are added as source terms. This worked fine for me.

Anders

[idrama]

Hello Saidi!

Do you have any reference for the non-linear k-epsilon modell?

Cheers,

Claus