[matejmuller]

Hello!

I'm getting strange u+ values by using LES with Smagorinsky (see attachment).

I'm modeling open channel flow, my y+ values are under 1 (around 0.6), for the wall boundary condition i'm using:

nuSgs --> nutUSpaldingWallFunction
U --> fixedValue uniform (0 0 0)
p_rgh --> zeroGradient
alpha.water --> zeroGradient

The mesh refinement towards the wall is linear with a step around 1.2.

It seems like the u+ values follow a linear law in the viscous and logaritmic region.

With a coarser mesh, where y+ is greater then 30, the results are OK (the u+ values follow the logaritmic law).

Has anyone experienced the same issues? In my knowledge one can use y+=1 values in LES models because of the vanDriest damping. Is the openfoam LES model limited only for y+ values in the logaritmic region (this is y+>30).

Regards, Matej

[sheaker]

Hello.
I'm not sure but as I know y+ should be from 10 to 300.
CFL should be < 1.

[matejmuller]

Hi!

Thanks for the reply... Yes, it looks like it is only suitable for larger y+ values, regardless the van Driest damping...or the damping is to big, considering that the u+ values behave linear also in the logaritmic region?...i'm not sure

Maybe my LESproperties are not well defined? I've used the properties from the tutorial with delta set to vanDriest.

regards, matej