CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM Pre-Processing (http://www.cfd-online.com/Forums/openfoam-pre-processing/)
-   -   wall function for twoPhaseEulerFoam (http://www.cfd-online.com/Forums/openfoam-pre-processing/96006-wall-function-twophaseeulerfoam.html)

Ofc January 10, 2012 09:20

wall function for twoPhaseEulerFoam
 
Hello Foamers,

as the title says I look for some good wall function using the twoPhaseEulerFoam solver. Right now I deal with y+ values around 0.001 to 10. I tried zeroGradient for k and epsilon, but I was not sure about the results. I tried to get higher values of y+ (above 30) for some walls and use the kqrWallFunction, but when I try to run the solver it says this wall function does not exist. Anyone knows how to add the function to the solver?

Also, since I have a small slit with very low y+ values I thought of using a laminar case there. Is there a option in OpenFOAM to use two different solvers (1 laminar 1 turbulent) for the areas or do I have to run the case turbulent first then export the values at the slit inlet and outlet and rerun the slit in another case?

Thanks for your time & help,

alberto January 11, 2012 10:01

Quote:

Originally Posted by Ofc (Post 338622)
Hello Foamers,

as the title says I look for some good wall function using the twoPhaseEulerFoam solver. Right now I deal with y+ values around 0.001 to 10. I tried zeroGradient for k and epsilon, but I was not sure about the results. I tried to get higher values of y+ (above 30) for some walls and use the kqrWallFunction, but when I try to run the solver it says this wall function does not exist. Anyone knows how to add the function to the solver?

The twoPhaseEulerFoam solver uses the same wall-functions implemented in bubbleFoam (the code is linked, that's why you do not see it into the twoPhaseEulerFoam directory. The tutorials using the turbulence model rely on wall-functions too.

Best,

Ofc January 12, 2012 10:30

Thanks Alberto,

I did look up the bubbleFOAM files now. For it seems that the solver always takes a wall function
"if (isA<wallFvPatch>(currPatch))" (from the wallFunctions.H)
I translated this line with "if a boundary is defined as a wall in the file constant/polyMesh/boundary
What is the difference if I change the wall function in the /0 folder?

alberto January 12, 2012 12:01

Quote:

Originally Posted by Ofc (Post 339040)
Thanks Alberto,

I did look up the bubbleFOAM files now. For it seems that the solver always takes a wall function
"if (isA<wallFvPatch>(currPatch))" (from the wallFunctions.H)
I translated this line with "if a boundary is defined as a wall in the file constant/polyMesh/boundary
What is the difference if I change the wall function in the /0 folder?

Yes, if you use the k-epsilon model in twoPhaseEulerFoam, you are always using the hard-coded wall-function in the code. If you specify a wall-function in the 0/ directory, it will be overridden by the solver, and it won't have any effect.

Best,


All times are GMT -4. The time now is 22:04.