[batta31]

Hi guys, I've got a question:
I would like to set up a simulation with a "laminar" turbulence model, but with non uniform values of nu over the domain. Is that possible in openFoam?

Thanks
Simone

[kmooney]

Do you mean that you want a non-uniform Newtonian viscosity field?

[batta31]

Yes, actually I want to use the distribution of nut of a different simulation and apply it to my new case, which has to run as a laminar case but with non uniform viscosity.

[kmooney]

Originally Posted by batta31

Yes, actually I want to use the distribution of nut of a different simulation and apply it to my new case, which has to run as a laminar case but with non uniform viscosity.

If you can create a volScalarField out of your old nut distrubution perhaps like this:

volScalarField myNuField = nut;

You should be able to drop it into the diffusion term in most any solver as such:

        fvVectorMatrix UEqn
        (
            fvm::ddt(U)
          + fvm::div(phi, U)
          - fvm::laplacian(myNuField, U)
        );

Cheers!

[batta31]

That's a very clever idea! Thank you very much kmooney, I'll try as soon as I can. If it doesn't work I'll come back.

Cheers!

[batta31]

Hi again,
I come back to this thread because I've a doubt..

Since I have a non-uniform field of viscosity, besides the term suggested,

fvm::laplacian(myNuField, U)

shouldn't I insert even a term like:

(fvc::grad(U) & fvc::grad(myNuField)


To better understand my reasoning, think you have to derive the NS equation but with non-uniform spatial viscosity..if you then explode the viscous term,

div(nu*grad(U))

you should obtain both the terms above, right?

Cheers
Simone