k-eps problems ?
 

i am a biginne in turbulence modelling field. now , i try to solve 1-D , turbulent channel flow problem by using low reynolds k-eps model , Implicit method. My problem is , it does not converce. And i want to know the commoa way to tread thesource term in k and eps equation.

If you have any suggestion , please post it on this page or directly contact at my E-mail address suthichock@yahoo.com

Re: k-eps problems ?
 

Dear Friend

What type of low Reynolds number model are you using? One is with zero epsilon on the wall and the other with non-zero epsilon.

If you are using the first type (e.g. Jones and Launder), most of your source of trouble lies in the additional terms in k and epsilon equations. You know what I mean? Most of the papers describe those terms as D (in k-equation) and E (in epsilon equation). The term D has a physical justification, but E is just an addition to get what was observed in the experiments (Jesus! is it turbulence MODELLING?). I think most of the pain is caused by E. The solution is that deal with D and E as you would deal with them in explicit scheme. Which means your overall scheme would be implicit but these terms will be in explicit form. Moreover another cause may be initial conditions. I sincerely wish you get the solution.

In case you are using a model with D=E=0 the source of your anxiety may be the improper initial conditions. I hope you are reading other discussions also, aren't you? You will find the answer.

Good luck

Sana

Re: k-eps problems ?
 

Dr. Sana

Thank you very much for your useful suggestions.

Suthichock