[giammy92]

Hi foamers,
i would like to know if OF calculates by itself Aref doing lref*depth = 1 in 2D case or if reads Aref that i have put in file force coefficients.
I'm asking that because i have found right force coefficients only after i have changed depth of mesh at 1, while before depth was about 1.2

[alexeym]

Hi,

To answer this question, you can take a look into $FOAM_SRC/postProcessing/functionObjects/forces/forceCoeffs/forceCoeffs.C. Constructor is:

Foam::forceCoeffs::forceCoeffs
(
...
)
:
    ...
    Aref_(0.0)
{
    read(dict);
    ...
}

I.e. Aref is set to 0 and then read from dictionary like this:

void Foam::forceCoeffs::read(const dictionary& dict)
{
    if (active_)
    {
        forces::read(dict);

        ...
        dict.lookup("Aref") >> Aref_;
    }
}

So, no, forceCoeffs function object does not calculate Aref.

[giammy92]

strange, because i have put in force coefficients dictionary Aref=lref*1 but in my mesh i hadn't 1 for depth. When i have changed depth to 1 i have had correct results

[alexeym]

Well, your 20% (changing depth from 1.2 to 1) error in coefficient values can be caused by anything but Aref: lRef, CofR, mesh, schemes, boundary conditions, convergence, etc.

[giammy92]

Now i'm repeating simulation with change of depth on mesh 2D, keeping Aref unchanged

[giammy92]

I have repeated same simulation with same bc, schemes and solution but only changing depth of mesh from 1 to 2, keeping same Aref in force coefficients dictionary, but my results are increased exctacly of a factor 2!
That means that Aref depends from depth of mesh and not from value added in dictionary, but the strange thing is that i would be expected force coefficients halved and not doubled!!

[alexeym]

Here is the code:

// lift, drag and moment
coeffs[0] = (totForce & liftDir_)/(Aref_*pDyn);
coeffs[1] = (totForce & dragDir_)/(Aref_*pDyn);
coeffs[2] = (totMoment & pitchAxis_)/(Aref_*lRef_*pDyn);

Aref and lRef are constants, pDyn is

scalar pDyn = 0.5*rhoRef_*magUInf_*magUInf_;

and depend on configuration (either rhoRef_ is constant read from configuration, or rho field read from simulation results).

totForce is

Field<vector> totForce(force_[0] + force_[1] + force_[2]);

force_ array is inherited from forces class and is calculated (depending on configuration actual calculation code varies):

// Normal force = surfaceUnitNormal*(surfaceNormal & forceDensity)
vectorField fN
(
    Sfb[patchI]/sA
   *(
	Sfb[patchI] & fD.boundaryField()[patchI]
    )
);

// Tangential force (total force minus normal fN)
vectorField fT(sA*fD.boundaryField()[patchI] - fN);

//- Porous force
vectorField fP(Md.size(), vector::zero);

As you can see normal and tangential forces depend on areas of the faces (Sfb array) of the patch, where you calculate force coefficient. And these areas depend on "mesh depth" in 2D case.

[giammy92]

Yes, i have already guessed,thank you anyway for your reply