[RobertHB]

Dear Foamers,

I'm doing laminar flow simulations in the Stokes regime (Re << 1). Currently my rectangular box chaped domain has smooth walls treated with fixedValue boundary conditions.

For a next set of simulations i want to test the influence of various wall roughness factors on the flow. I know there is the nutURoughWallFunction treat nut at the walls but this requires the use of a turbulence model. Given the nature of my flow the use of a turbulence model just to calculate nut is of not much use as my turbulence variables range around e-10. But this seems to be the only way to implement some rough wall treatment.

Do you consider this to be the right approach? Are there different methods for implementing wallroughness in laminar flows in OpenFoam?

Cheers
Robert

[wildfire230]

I wonder if there is a way to add a random "kick" normal to the surface to each of the grid points with a distribution about your roughness of interest?


By the way, what solver are you using for the Stokes regime? simpleFoam? I am actually working on a related problem using a slightly modified simpleFoam where I simply deleted the convection term in the equation. I'm a little unclear about boundary conditions because zeroGradient seems definitely inappropriate for the pressure, but I'm not really sure what's better.


Best regards

[RobertHB]

Originally Posted by wildfire230

By the way, what solver are you using for the Stokes regime? simpleFoam? I am actually working on a related problem using a slightly modified simpleFoam where I simply deleted the convection term in the equation. I'm a little unclear about boundary conditions because zeroGradient seems definitely inappropriate for the pressure, but I'm not really sure what's better.

I'm using simpleFoam with the turbulence model set to laminar. Haven't thought to much about BC's because i do not care for the pressure field. And the velocity field matches perfectly with the analytical solution.