[Yage]

Dear All,

I set the boundary condition for k/omega for walls, such as for k:

dimensions      [0 2 -2 0 0 0 0];

internalField   uniform 0.04;

boundaryField
{
    frontAndBack
    {
        type            symmetry;
    }
    inlet
    {
        type            fixedValue;
        value           $internalField;
    }
...
    walls
    {
        type            kqRWallFunction;
        value           uniform 0;
    }
}

but the values in each processor is changed after decompose:

boundaryField
{
    frontAndBack
    {
        type            symmetry;
    }
    inlet
    {
        type            fixedValue;
        value           nonuniform 0();
...
    blade
    {
        type            kqRWallFunction;
        value           uniform 0.04;
    }
...
}

Any idea about this problem?

Regards,
Yage

[wyldckat]

Quick answer: It's not a problem, it's a counter-measure to avoid shooting one's own foot.
"kqRWallFunction" is a boundary condition that is calculated by OpenFOAM itself and the provided value is for post-processing purposes.

If you were using 0, the solver would crash in the first iteration, because somewhere it would try to do a division by k=0.

[Yage]

Dear Bruna,

thanks for your reply.

But the non-zero value will be also changed to internal field value, such as I use 1 instead of 0 for k at walls

[Antimony]

Hi,

If you have specified a non-zero value for the value, then OpenFOAM would not change it to the internalField value.

My guess is that OpenFOAM replaces the value with internalField value (when the original value is zero) because it is the simplest estimate it can make based on the information it has.

Hope this helps.

Cheers,
Antimony

[Yage]

Originally Posted by Antimony

Hi,

If you have specified a non-zero value for the value, then OpenFOAM would not change it to the internalField value.

My guess is that OpenFOAM replaces the value with internalField value (when the original value is zero) because it is the simplest estimate it can make based on the information it has.

Hope this helps.

Cheers,
Antimony

Dear Antimony,

Sorry for the later reply.

My openfoam version is 2.3.0 and the omega boundary value will change to internalfield value after decompose. But both of the values are non-zero. This is very strange.

One more question:
If I refine the mesh with added layers (y+<6), so that I dont need the wallfunction. then I should use fixedValue with uniform value=1e-10 for both k and omega, is that correct?

Regards,
Yage