incorrect to use nut wall functions for k-epsilon?
 

Hello,

Is nut used only by Spalart Allmaras turbulence model or can it be specified in wall functions for other models as well?

I tried to run k-omega SST turbulence model with the following wall functions specified on the walls:

omega { type omegaWallFunction; value uniform 1; }
nut { type nutSpalartAllmarasWallFunction; value uniform 1; }

Specifying k instead gives similar results:

omega { type omegaWallFunction; value uniform 1; }
k {type kqRWallFunction; value uniform 0.75;}


When I use realizable k-epsilon I get very different results depending on wheather k or nut is specified by the wall.
Is it correct to use wall functions for nut for k-omega SST, but incorrect for K-epsilon models?

Regards