[Maveryck]

Hi everyone,

I'm simulating a flow through a channel. I impose a mean velocity which traduces in a pressure gradient calculated by openFoam.


Here are boundary and initial conditions for Pressure:

boundaryField
{
    inlet
    {
        type            cyclic;
    }

    outlet
    {
        type            cyclic;
    }

    FixedWalls
    {
        type            zeroGradient;
    }

    frontAndBack
    {
        type            zeroGradient;
    }
}

Velocity Conditions:

boundaryField
{
     inlet 
     { 
        type            cyclic;
     } 

    outlet
    {
        type            cyclic;
    }

    FixedWalls
    {
        type            noSlip;
    }

    frontAndBack
    {
        type            noSlip;
    }
}

But it is not clear to me how to determine which boundary conditions are suitable for k and epsilon in this simulation. Until now I have tried the following:

Boundary conditions for k:

boundaryField
{
    inlet 
    { 
        type            cyclic;
    } 

    outlet
    {
        type            cyclic;
    }

    FixedWalls
    {
        type            kLowReWallFunction;
        value           uniform 0.75;
    }

    frontAndBack
    {
        type            kLowReWallFunction;
        value           uniform 0.75;
    }
}

B.C. for epsilon

boundaryField
{
    inlet 
    { 
        type            cyclic;
    } 

    outlet
    {
        type            cyclic;
    }

    FixedWalls
    {
        type            epsilonWallFunction;
        value           uniform 14;
    }

    frontAndBack
    {
        type            epsilonWallFunction;
        value           uniform 14;
    }
}

I'm getting small values for nut at the walls, which means I get higher WallShearStress values than expected and thus I don't get proper uPlus-yPlus profiles (which is what I'm looking for).

Has anyone worked with similar cases? any suggestion will be wellcome

Thanks, Maveryck.

Note: y+ at the walls have values around 5-8.

[C-L]

Your boundary conditions are fine for those y+ values.
I expect this is because of the turbulence model you are using - k-epsilon is unsuitable for these flows unless you use highRe wall functions (i.e y+ levels are in the log-region).

Try v2-f if you want to use an epsilon-based model, or use an omega-based model if you don't want to use the four-equation one.

[pconen]

When you have problems finding suitable values you can try out the cfd-online tools.
https://www.cfd-online.com/Tools/turbulence.php