CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   k-e turbulence (https://www.cfd-online.com/Forums/fluent/36833-k-e-turbulence.html)

hung June 2, 2005 18:43

k-e turbulence
 
Hi everyone!

Now I'm modeling a turbulence flow over the spheroid by Fluent, and I used K-epsilon model. But my result ( Cf, Cp....) are quite difference to the results in the paper.

I used the K-epsilon, but I'm not sure about the coefficients in this model, for example k, epsilon....

Somebody can recommand to me, how to simulate it to get a better result. Thanks a lot!

pah

zxaar June 2, 2005 21:02

Re: k-e turbulence
 
read the discussion "Which do you prefer? 6.1 or6.2?" on this page, there what ever i have said is applicable to the case you are talking about.

hung June 3, 2005 01:09

Re: k-e turbulence
 
Thanks for your reply!

I'm doing with the Prolate spheroid 6:1 and it has angle of attach 10, 20 deg at Re=4.2*1e6. And I did with k=0.1, epsilon=0.1. But the result so bad.

Somebody, Plz recommand to me the way to simulate it. Thanks one more!

PAH

zxaar June 3, 2005 02:14

Re: k-e turbulence
 
did you read the discussion i suggested to read,

hung June 3, 2005 02:46

Re: k-e turbulence
 
Hi zxaar!

I did, that's great and detail guiding. But my problem now is Turbulence models only. Because I want to complete turbulence model at the Steaty state case first, if it's true then I go to unsteady case.

Therefore, I'm considering the Tur coefficients. Could you recommand some ideas about it to me!

Thanks a lot!

PAH


zxaar June 3, 2005 03:25

Re: k-e turbulence
 
see if you have to stik to steady state calculations of k-e type, the best results you get with k-e realisable, and if u can have yplus near to 1, use enhanced wall functions, but with k-e type models you will need prism mesh near to walls of your body. and using yplus near to 1 is your best bet.

i suppose RSM should also give good results but i haven't tried it yet. note using prism near to walls gives best pressure coefficient predictions.

hung June 3, 2005 03:46

Re: k-e turbulence
 
Thanks for your recommand!

I use the k-e model and my grid on the wall is hex, because the hybrid grid is used in my model in some blocks.

And then, at the Models is set up with k-e realisable. But in the Inlet BC, we need to set up K, epsilon. In here, I'm trouble. I just set them as I mentioned.

Thanks !

PAH

zxaar June 3, 2005 03:52

Re: k-e turbulence
 
use intensity and lengh scale instead, read what manual has to say about it, in the boundary conditions part, then judge what you want to set,

hung June 3, 2005 05:38

Re: k-e turbulence
 
I want to set the dimensionless model of flow over the spheroid, and then I use the Velocity inler is U=1 with the long axe a=1.

Do you have some ideas about dimensionless in Fluent?

Thanks a lot, Zxaaz!

PAH

zxaar June 3, 2005 06:34

Re: k-e turbulence
 
nope i never tried that


All times are GMT -4. The time now is 04:49.