[volker]

Hey all,

currently I'm working on the comparison of heat to mass transfer in a high swirling pipe flow. And in order to achieve reasonable results the wall gradient (to calculate both the Nusselt number and Sherwood number) plays a crucial role. I experienced, that OpenFoam has quite a few problems calculating an appropriate wall gradient, i.e. OF uses a finite difference approach of first order.
To get rid of this undesirable feature, the use of a low-Reynolds model in the viscous sublayer becomes necessary (and hence a mesh refinement up to y+~1).
The drawback of choosing a low-Re for a compressible flow, such as a high swirling pipe flow with streamline curvature is that only one low-Re (LaunderSharma, which fails completely predicting the right Reynolds stress) is implemented.
Now my hope is that someone is working on a low-Re model implementation in compressible form, maybe even a full RSTM or algebraic RSTM.
Or does anyone know if there are any new models implemented in the version 1.6.x release?!?

I would be very grateful, if anybody could give me a hint.

best regards
\volker