Problem with a filter surface.
 

Hi,
I am trying to calculate the streamwise flow field in a channel where the upper wall is solid and the lower is porous. The normal profile of the axial velocity matches the experimental PIV results at the solid wall but it is too smoothed at the porous wall (velocity gradient too small at wall). Permeability used for the simulation was according to experimental measurements. I did used initially k epsilon model and later SST with y+=1 but none of them resolved the problem near to the filter surface. Any clue?
Regards,
Eduardo

Do you have a boundary layer forming on the porous material? I suspect you probably do but it would be quite a different one to normal boundary layers. This means traditional wall functions would not handle this very well. Also I don't think CFX applies wall functions to porous surfaces anyway.

You might have to use an LES approach to directly model the wall boundary layer at the porous surface.