CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   OpenFOAM (https://www.cfd-online.com/Forums/openfoam/)
-   -   drag coeff to high - kOmegaSST turbulence model (https://www.cfd-online.com/Forums/openfoam/84331-drag-coeff-high-komegasst-turbulence-model.html)

 RugbyGandalf January 26, 2011 11:51

drag coeff to high - kOmegaSST turbulence model

1 Attachment(s)
Dear Users,

i have a problem with the drag coefficient calculated by the kOmegaSST turbulence model with the simpleFOAM solver :-(

The aim is to get to know about the drag coefficient of a streamlined body and compare this results with the ones got by the same body out of a wind tunnel experiment. So in fact - compare reality an CFD ;)

The coefficient i should get for cd is about 0.07, the one i got from my calculations is somewhat around 0.96...

I have calculated values for k and omega from the formulas out of the CFDWiki, but they seem to be incorrect - or a BC is the problem - i don't know :(

The eddy viscosity ratio i took was 1, but i don't know if this is correct for wind tunnel experiments...?

I will post my case files - maybe someone do have an idea...!?

Thank you very much in advance

Regards,

Martin

 RugbyGandalf February 19, 2011 10:35

1 Attachment(s)
I got the solution...

After convergence of the first calculation with the uploaded case i got to high drag coefficients...
One need to change to higher order interpolation and gradient schemes...
i will post my updated fvSolution and fvSchemes here...

 Greg Givogue February 19, 2011 14:51

What was the calculated CD?

 RugbyGandalf February 19, 2011 15:23

With the old fvS it was about 0.27 and now it is about 0.071

 Greg Givogue February 20, 2011 10:59

Nice work. How long did it take to run 5e10 iterations (5000/0.000001)?

 alex_rubel February 22, 2011 03:35

Hi, I had a equivalent problem and what I find in my case is that : when I changed
- div(phi,U) Gauss Upwind -> Gauss linearUpwindV Gauss linear; I get 40 % of errror instead of 70%
and next I changed the turbulent quantities too :
- div(phi,k) Gauss linearUpwind Gauss linear;
- div(phi,omega) Gauss linearUpwind Gauss linear;

It's still running but after 100 iterations I get 10% less of error

Alex

 All times are GMT -4. The time now is 21:29.