[FlyBob91]

Hello Foamers,
I'm going deeper to the implementaton of the Spalart-Alamars turbulence model and I have a big doubt about where wall function is called.
In the kEpsilon and kOmega turbulence models the wall function is applied through

epsilon_.boundaryFieldRef().updateCoeffs();

Lne 252 in /src/TurbulenceModels/turbulenceModels/RAS/kEpsilon/kEpsilon.C, and

omega_.boundaryFieldRef().updateCoeffs();

at line 209 in /src/TurbulenceModels/turbulenceModels/RAS/kEpsilon/kEpsilon.C.

On contrary, for he SA turbulence model there is no updateCoeffs() function. Any idea where wall functon is called?

Best regards

[HPE]

Hi,

- `nuTilda` BCs are updated by nuTilda_.correctBoundaryConditions();
- There is no `nuTilda` wall function in OpenFOAM - at least in the standard variants.
- One of the derivation assumptions of the standard SA closure model is that y1+ is around 1. Therefore, by definition, wall functions are not required, and should not be used.
- Having said that some researchers have been providing various approaches to modify the standard SA model, so that it can behave as a "high-Re" number type model to reduce computational costs, e.g. A generalised wall function including compressibility and pressure-gradient terms for the Spalart–Allmaras turbulence model.

Hope it helps.

[FlyBob91]

Thank you for the reply. It makes a lot of sense. Anyway it rises a new question because the wall function is applied somewhere to nut and it also affects the value of nuTilda.
In the tutorial located in tutorial tutorials/incompressible/simpleFoam/airFoil2D the nuTilda file doesn't show any wall function (as you said) while the nut variable has a nutUSpaldingWallFunction wall function at the wall. I run two simulations with the nutUSpaldingWallFunction activated and not activated and the results for the nuTilda variable are shown in the attached files. So my question now is where the nutUSpaldingWallFunction is called.

[HPE]

Hi,

- `nutUSpaldingWallFunction` has been called by this line: this->nut_.correctBoundaryConditions();
- As you know, `nut` and `nuTilda` are different quantities.

[FlyBob91]

Thank you, now it's more clear. I was just confused by the fact that the name of the function which updates the value at the wall for nut is different than the name of the functions that update the value of the k and omega.
In summary the wall function is applied to epsilon and omega through

epsilon_.boundaryFieldRef().updateCoeffs();

omega_.boundaryFieldRef().updateCoeffs();

while at nut through

nut_.correctBoundaryConditions();

.
As last thing I wonder where the wall function is applied to the turbulent kinetic energy (k) for the models that calculate it.

Best regards and thanks for the contribution