[Vino]

Hi,

I am doing a turbulent incompressible flow simulation over cylinder at Re=10000. I am using SA model with out trip term. when i use nu_tilda/nu=3-5(recommended, but there is no limit on higher side) , I am getting my Cl & Cd results on the higher side. But when i use nu_tilda/nu=50, i am getting results matching with literature.

Now my question is "Can i use nu_tilda/nu=50 or more for the simulation?"

Thanks in Advance.!!!

[jaiswal]

hi vino

i am also working on turbulent flow simulation but for compressible flows. Can u give more details about your problem. I think u are workin in 2D, right? are u using C-grid or O-grid? what are all the boundary conditions you are applying for nutil?

[Vino]

Hi,

Yes. I am doing a 2D simulation over a cylinder using O- type grid.
At inlet & farfield, nutilda=3 nu
At wall, nutilda=0
At outlet, nutilda extrapolated from interior domain

[jaiswal]

hi again

I am solving navier-stokes equations for flow over an NACA 0012 airfoil with C-grid. I am also facing problems in getting the right solution using SA-model. I think I am not applying the right boundary conditions.

Can u explain me how you interpolate nutil at outlet from interior domain???

I am also not very clear about answer to your question.

but i think if u try one more test case say an airfoil with same BC's and still getting right answer then probably using nu_tilda/nu=50 in your case may be right.

[Vino]

I am using dummy cell type B.C , for example nutilda[i+1]=nutilda[i] at outlet