[syoon]

I'm running oodles to simulate a backward facing step.
Running with laminar inlet boundary condiiton (no fluctuation) is okay.

Now, I'm trying to use turbulentInlet to impose white noise at the inlet.
Then, from oodles/pitzdaily, I can use the following if U=10 m/s.

inlet
{
type turbulentInlet
referenceField uniform (10 0 0);
fluctuationScale (0.02 0.01 0.01);
value uniform (10 0 0);
}

My problem is my referenceField is not uniform.
I do have non-uniform experimental data.
I wonder how I can incorporate my experimental data in the above.

If I don't use "turbulentInlet", I can incorporate my experimental data as:

inlet
{
type timeVaryingMappedFixedValue:
setAverage off;
}

and impose my experimental data in /constant/boundaryData/inlet/0


Could anyone kindly let me know how I can apply the experimental data using turbulentInlet?

Regards,
sungho

[cedric_duprat]

Hi sungho,

What do you mean by non-uniform data ?
I think that the inlet of the backward facing step could be a "simple" periodic channel flow.

in that way, check the directMappedPatch.
If you want more paper on this boundary condition, you can check Eugene de Villier Thesis.

The problem if you want to use your experimental data is that you will interpolate these data in you computational mesh and I think you will loose your fluctuations.

I hope it'll help you

Cedric

[syoon]

Thank you, Cedric.

I meant 'experimental data' by non-uniform data.
I agree that a "simple" peridodic channel flow would be the best inlet boundary condition.
Would you let me know where I can find Eugene de Villier thesis?

Regards,
Sungho

[cedric_duprat]

Hi Sungho,

Maybe you can have a look at the turbulenceSIG on the wiki to find information. http://openfoamwiki.net/index.php/Si...Implementation

Don't hesitate to add comments or informations on the wiki.

Cedric

[syoon]

Thank you, Cedric.
I downloaded the thesis and I'm currently exploring directMappedPatch.
Thank you for the information.