CFD Online Discussion Forums - LES Boundary conditions

Hi all,

I am simulating a flow inside a room. I have simulated using RANS k-epsilon model and the results are good with experimental results.

In case of LES, smagorinsky the results are not good. I would like to know whether my boundary conditions are correct for LES simulation. Please see attached figure to see the domain.

Quote:

U:
Inlet - fixedValue - uniform (0.225 0 0)
outlet - inletOutlet
walls - slip (i have also tried noSlip)

p
Inlet - zeroGradient
Outlet - fixedValue - 0
walls - zeroGradient

nut
inlet - calculated
outlet - zeroGradient
walls - zeroGradient. (my yplus is less than 1)

In case of LES simulations, the flow is not losing any velocity and predicts very high than the experiment results.

Attachment 66338

Attachment 66339