I have a question concerning the Scalable Wallfunction Approach presented by Grotjans and Menter: "Wall functions for General Application CFD Codes".
CFX uses this wallfunction for the standard ke- and kw- turbulence-modells. This approach shifts the viscous sublayer edge towards the wall and yields (reported in the paper and in the dissertation by Grotjans, and if calculating with CFX) very reasonable and good results in a y+-range from 1 up to 500. I am always very fascinated how stable CFX runs with this approach and also tried to implement it in a small code for my own. This approach does not only limit the minimum y+-value to around 11.06, like often discussed here in the forum, but there are also differently adopted boundary conditions for the epsilon-equation (Neumann-type) and the momentum-equations compared to the standwallfunction-implementations like I could also find in OpenFoam.
Has anybody implemented the Grotjans-approach successfully, I don't get my implementation stable and the results are far away from the ones I get with the standard-wall function approach. I think there is no implementation of this approach in OpenFoam, is it?! For reformulated flux-boundary-condition in the epsilon-equation in the above mentioned paper a calibration-function is given without values, has anybody found out anything about these values?
Thank you very much for your help...
|All times are GMT -4. The time now is 18:47.|