Hi

I'm trying to implement the LES model in a code that uses immersed boundaries in a cartesian grid. How would you treat the walls?

It is difficult to use dynamic procedures since there is now obvious way to do averaging.

Using "simply" Van Driest: how do I determine the distance to a wall?

What other method could I use?

Thanks for your help!

Katrien

LES does not provide solutions for near wall region. There are only 2 ways to treat near wall region.

1) High resolution near wall boundary, which could prove costly 2) Use the analytical wall function to determine y+ and then u+ at the first grid point away from the wall. You can find this in any standard CFD text.

Indeed, I was planning to go for option 2. How do you determine the distance from the wall with immersed boundaries? Any reference?

In order to determine the distance from the wall you should use any geometry textbook. Just drop a perpendicular from the point on wall surface and then determinte point of intersection. Knowing coordintes of two points you can easy calculate this distance. What about 3D space imagination?

I would like to advise you to apply improved Smagorinsky model which does not requare knowing of this distance. Look

By E. Leveque, F.Toschi, L.Shao and J.-P. Bertoglio, Shear-improved Smagorinsky model for large-eddy simulation of wall-bounded turbulent flows.// J.Fluid Mech (2007) vol 570, pp 491-502.

This model is much more simpler then classical dynamic model. I used it for LES channel calculations and got very good results. Even for Re_tau = 1000...5000.

dear alexey,

nice to see that the SISM is already used in practical applications. interested in details of your LES using SISM

best, Emmanuel Lévêque