[tiam]

Dear Foamers,

As I understand the wall-function nuSgsWallFunction now only lives int the foam-extend version of OpenFOAM. I am trying to compile it against the official release 2.2.0.
For some reason I get the following problem:

nuSgsWallFunctionFvPatchScalarField.C:129:9: error: ‘lookupPatchField’ was not declared in this scope
         lookupPatchField<volVectorField, vector>(UName_);
         ^
nuSgsWallFunctionFvPatchScalarField.C:129:40: error: expected primary-expression before ‘,’ token
         lookupPatchField<volVectorField, vector>(UName_);
                                        ^
nuSgsWallFunctionFvPatchScalarField.C:134:40: error: expected primary-expression before ‘,’ token
         lookupPatchField<volScalarField, scalar>(nuName_);
                                        ^
make: *** [Make/linux64GccDPOpt/nuSgsWallFunctionFvPatchScalarField.o] Error 1

Here are my "options" and "files":

nuSgsWallFunctionFvPatchScalarField.C

LIB =  $(FOAM_USER_LIBBIN)/libnuSgsWallFunctionFvPatchScalarField

EXE_INC = \
	-I$(LIB_SRC)/turbulenceModels \
	-I$(LIB_SRC)/transportModels \
	-I$(LIB_SRC)/finiteVolume/lnInclude \
	-I$(LIB_SRC)/meshTools/lnInclude

LIB_LIBS = \
	-lincompressibleTurbulenceModel \
	-lfiniteVolume \
	-lmeshTools

I digged a bit. The method that the compiler cannot find is defined in fvPatch.H, so this should be visible through -I$(LIB_SRC)/finiteVolume/lnInclude if I understand correctly. Indeed, if I append the method with Foam::fvPatch:: then it seems to find it, but gies another bug instead. But the code should probably compile without change anyway. Any thoughts?

Best,
Timofey

[wyldckat]

Greetings Timofey,

I think you need to improve your online searching skills Although I have to say it isn't very linear how to figure this out... although, the trick was to search for the compressible version "muSgs". This is further explained in the following bug reports:

• http://www.openfoam.org/mantisbt/view.php?id=874
• http://www.openfoam.org/mantisbt/view.php?id=627

Quoting Andy from the last comment on the first link:

The behaviour of the old muSgsWallFunction for LES cases can be recovered by using the mutUSpaldingWallFunction wall function.

Therefore, in theory, in your case you would need "nutUSpaldingWallFunction" instead of "nuSgsWallFunction".

Best regards,
Bruno

[tiam]

Thank you Bruno!

We actually managed to find it before seeing your answer, more or less by chance .