Wall modelling in the leeward face
 

Hi there,

I have embedded wall modelling in my LES code and tested that for a plane channel flow and it is working OK. The model is based on the blended linear-log law.

Now, I want to use the wall modeled-LES for a bluff body flow (square cylinder immersed in a BL). But I have some doubt about its correctness:

I say, Since there is strong wake behind the cylinder and we know for sure that we have recirculation and reverse flow, is that legitimate to use wall modelinf for that region?

On the other hand, I say the wall model just modifies the flow structure in the fist node and does not have any thing to do with flow direction. In the wake region, we have high Re flow any way?

BTW,I do not want to move toward RANS-LES and want to stay with wall function.

Do you have any idea?

Thank you