[ChasingNeutrons]

Hi guys,

Struggling to get a solver I've made to compile and have only one somewhat obscure error message left to overcome. I am hoping that someone might be able to advise me on what I'm doing wrong. The message is:

In file included from /opt/openfoam30/src/OpenFOAM/lnInclude/dimensionedType.H:343:0,
                 from /opt/openfoam30/src/OpenFOAM/lnInclude/dimensionedScalar.H:38,
                 from /opt/openfoam30/src/OpenFOAM/lnInclude/TimeState.H:38,
                 from /opt/openfoam30/src/OpenFOAM/lnInclude/Time.H:47,
                 from /opt/openfoam30/src/finiteVolume/lnInclude/fvCFD.H:6,
                 from MOCsolver.C:32:
/opt/openfoam30/src/OpenFOAM/lnInclude/dimensionedType.C: In instantiation of ‘Foam::dimensioned<typename Foam::pTraits<arg1>::cmptType> Foam::dimensioned<Type>::component(Foam::direction) const [with Type = double; typename Foam::pTraits<arg1>::cmptType = double; Foam::direction = unsigned char]’:
/opt/openfoam30/src/OpenFOAM/lnInclude/dimensionedType.C:404:23:   required from ‘Foam::dimensioned<typename Foam::pTraits<arg1>::cmptType> Foam::dimensioned<Type>::operator[](Foam::direction) const [with Type = double; typename Foam::pTraits<arg1>::cmptType = double; Foam::direction = unsigned char]’
MOCsolver.C:86:68:   required from here
/opt/openfoam30/src/OpenFOAM/lnInclude/dimensionedType.C:287:5: error: request for member ‘component’ in ‘((const Foam::dimensioned<double>*)this)->Foam::dimensioned<double>::value_’, which is of non-class type ‘const double’
     );
     ^
/opt/openfoam30/src/OpenFOAM/lnInclude/dimensionedType.C: In member function ‘Foam::dimensioned<typename Foam::pTraits<arg1>::cmptType> Foam::dimensioned<Type>::component(Foam::direction) const [with Type = double; typename Foam::pTraits<arg1>::cmptType = double; Foam::direction = unsigned char]’:
/opt/openfoam30/src/OpenFOAM/lnInclude/dimensionedType.C:288:1: warning: control reaches end of non-void function [-Wreturn-type]
 }
 ^

This seems to flag up for the line featuring the calculation of tau in my solver:

for (int i=0; i<naz; i++)
{
    for (int j=0; j<npo; j++)
    {
        for (int n=0; n<ncells; n++)
        {
            tau[i][j][n]=Lray[celllabel[n]]*sigmaT[celllabel[n]].value()/mag(cosphi[i]*sintheta[j]);
            exp_factor[i][j][n]= 1.0+Foam::exp(tau[i][j][n]);
        }
    }
}

My suspicion is that it's to do with tau and all other variables being scalars except for sigmaT which is a dimensioned scalar (although without a dimension set) property defined in createFields.H. Also, celllabel is an array containing the indices of each cell in the order they occur for my specific routine. I had thought that using the .value() method would overcome any such problems of incompatibility but seemingly not. Hence, I'm keen on any advice which might simplify things.

Cheers for any help.

[ChasingNeutrons]

Found a work around by changing the sigmaT to a volScalarField. Now other problems to fix!