CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   SIMPLE algorithm and k-eps turbulence model (

zqnwpu October 14, 2002 05:17

SIMPLE algorithm and k-eps turbulence model
I'm using SIMPLE-based control volume method and K-eps model to calculate the flow fields around airfoils,but when I use much more grids or highly stretched grids near the solid surface, the computation diverges,I don't know what caused this. Can anyone help me with that? thanks!


Abhi October 14, 2002 06:35

a suggestion
keep an eye on the n+ value. See whether the code diverges when n+ is < 11.63 in one of the cells? If that is the case, then derive the value of G, epsilon & wall shear stress using linear velocity profile !

Holidays October 15, 2002 02:31

Re: a suggestion
If y+ is getting small this can get tricky indeed, as you may not be respecting your model's assumptions. In Aero applications Low-Re k-E models are much more adapted as well as Shear Stress Transport k-Omega type models where you integrate to the walls.

YANG October 15, 2002 21:46

Re: a suggestion
Dear Abhi

What's the meaning of G in your suggestion?


YANG October 15, 2002 21:55

Re: a suggestion
Hi, Abhi

What's the lower limit of n+ for k-epsilon model near the solid wall?

Thanks YANG

Holidays October 16, 2002 02:46

Re: a suggestion

You should read CFD textbook materials! Details about most classical models, their applications and restrictions can be found in the books by Versteeg and Malalasekera (1995) or Wilcox (1993) for a more exhaustive approach. These are only suggestions of reading; there exist a lot more. In these anyway you would find explanations on P and G terms (G is the production term due to body force in general) and a lot more...

YANG October 16, 2002 03:50

Re: a suggestion
Dear Holidays

Thanks a lot for you patient help.


newcomer October 16, 2002 06:00

Re: a suggestion
"G is the production term due to body force in general" - it originates when you go for Reynold's averaging. Look at the viscous dissipation.

Holidays October 16, 2002 07:02

Re: a suggestion
G is of the form Gbuoy (+Grot + Gresist) with Gbuoy which can be written as -Mu_effective/(Rho*Sigma_Rho)*g*GradRho, ie a term that accounts for buoyancy and body force effects...

Abhi October 16, 2002 08:42

Re: a suggestion
Was the question first coined by YANG ? Yang , give me some more inputs abt the way ur code is behaving. May be I can suggest something. READ Turbulent Flows by STEFAN POPE if u have not!!!!!!! A very very good book

ammar October 21, 2002 18:22

Re: SIMPLE algorithm and k-eps turbulence model
i am using the TEACH-T code which uses the SIMPLE algorithm and K-E model and i want to modify it to solve unsteady problem what type of code ur using and what do u think of TEACH-T and could the modification be easily accomplished

Milad_123 May 28, 2012 04:08

Would you please send me the TEACH-T code.
Thanks a lot.
my email :

All times are GMT -4. The time now is 10:45.