[Jack001]

Hi all,

Sometimes seeing is believing. In my constant/turbulenceProperties one can choose to set "turbulence" on or off. I would expect this to have the same effect as if the other keyword in the same file- "simulationType" is set to laminar. But I want to be sure of this. So for example, in simpleFoam, in the UEqn.H, the term turbulence->divDevReff(U) seems to be where the turbulence model comes into play (right!?)

But there is no logic telling the solver whether to use the turbulence model or not, depending on what we specified in constant/turbulenceProperties! How does the solver know if we turned turbulence on or off? I was hoping maybe I could put in some print statements at the very least to see if the turbulence model is on or off! Maybe I am missing something regarding what the (class!?) "turbulence" is. But I could really use someone to point me in the right direction on this.

[agustinvo]

Hi

As you say, when you turn off the turbulence, it just do not execute correct, so the different variables (k, epsilon, nut ...) remain as they were in the previous iteration.

The solver knows , it seems it looks for it each time step. It makes sense when, for instance, you want first a simulation without resolving turbulence, and then you switch it on to start to solve your turbulence variables.

I hope I was clear!

[Jack001]

Hi agustinvo,

Thanks for your reply. I believe you but I wanted to understand exactly what happens in the code! So I did the following:

In rhoPimpleFoam.C I edited to include a print statement

if (pimple.turbCorr())
{
    Info << "Turbulence correction is on!" << endl;
    turbulence->correct();
}

but this print happens even if turbulence is turned off! So there must be somewhere else where the solver is told not to solve the turbulence model! Any ideas? Also in the simpleFoam.C solver there isn't even an if statement so where does the solver take into consideration the on/off?

[akidess]

The distinction is not made in the solver but in the turbulence library. If you use a RAS model and set turbulence to off, you in fact use the laminar "turbulence" model.

[Jack001]

Ok great! Could you point me in the direction to which file I should look at to see where this happens? I am using the LES smagorinsky model, so I am looking in

$WM_PROJECT_DIR/src/TurbulenceModels/turbulenceModels/LES/Smagorinsky/Smagorinsky.C

but to be honest I am overwhelmed c++ - I find it incredibly hard to follow the flow of information, how data is passed between objects etc. Searching through this code has prompted another question- Where in the code does the solver select the turbulence model? It must happen somewhere inside the class turbulence right? If so where can I find the details on this?

[agustinvo]

Hi

inside the correct function, you have

RASModel::correct();
if (!turbulence_)
{
return;
}

So if you have turbulence off, turbulence_=0, !turbulence_=1, and it goes out of the correct function.

[Jack001]

OK thanks thats what I was looking for!