CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM Verification & Validation (http://www.cfd-online.com/Forums/openfoam-verification-validation/)
-   -   Jump in epsilon values near the wall : low re k-epsilon model (http://www.cfd-online.com/Forums/openfoam-verification-validation/112285-jump-epsilon-values-near-wall-low-re-k-epsilon-model.html)

malaboss January 25, 2013 07:09

Jump in epsilon values near the wall : low re k-epsilon model
 
2 Attachment(s)
Hi FOAMers !
I am now studying an airfoil profile which small chord makes it look like a cylinder.
The thickness of the airfoil is about several cm and the velocity at the inlet is 10m/s.
I considered my fluid as incompressible due to low velocities.
k at the inlet is 1.5 ( 10% of turbulence)
epsilon at the inlet is (19,2, I choosed the highest mixing length that I considered to be about half of the airfoil thickness)

First, I tried to use a (high reynolds) k epsilon turbulence model with a wall function. It gave me some weird as epsilon near the wall was really high : several tens of thousands.
I saw in the following tutorial /tutorials/incompressible/simpleFoam/pitzDaily similar values.
Is it OK to have such great values near the wall for epsilon ?


Then, I decided to use a low Re kepsilon model launderSharma, and by initializing k an epsilon near to zero close to the wall (as it is done in the incompressible/boundaryFoam/boundaryLaunderSharma tutorial) I find a high jump for my espilon values as you can see in the pictures :
the first one is the espilonfield and the second one is the plotting of epsilon values in the radial direction.
I already checked y+ values and i the average is lower than 1.

I guess I should not have that result. Do you know why i get this and how to fix it ?:confused:
If you need it, I can provide more information about my dictionnaries.

Thanks to all for your help

maHein February 1, 2013 17:36

High values for dissipation near the wall are normal. Just have a look at the validation simulation of a flow ofer a flat plate performed by NASA http://www.grc.nasa.gov/WWW/wind/val.../fpturb01.html. You could just convert your data to dimensionless epsilon+ and compare it to the validation results.


All times are GMT -4. The time now is 23:50.