[tianyikillua]

Hello everyone,

I'd like to know if a dimensionedScalar (ex. T1) + volScalarField (ex. T) is still a volScalarField and the result is obtained elementwisely, that is : result(i)=T1+T(i).

I need also to use a scalar function (ex. pos) determing the positivity of a scalar. If I apply this function to a volScalarField (ex. T), can I still obtain a volScalarField and result(i)=pos(T(i)) ?

I wrote this because I'm now implementing a non-constant diffusivity in the laplacianFoam solver.

In mylaplacianFoam.C, I have

DT=(a1+b1*T)*pos(T1-T)+(a2+b2*T)*neg(T1-T)*pos(T2-T)+(a3+b3*T)*neg(T2-T);

before solving the laplacian equation.

In creatFields.H, I have

    volScalarField DT
    (
        IOobject
        (
            "DT",
            runTime.timeName(),
            mesh,
            IOobject::NO_READ,
            IOobject::AUTO_WRITE
        ),
        (a1+b1*T)*pos(T1-T)+(a2+b2*T)*neg(T1-T)*pos(T2-T)+(a3+b3*T)*neg(T2-T)
    );

where ai, bi, T1 and T2 are scalars and DT is piecewise affine in [0,T1], [T1,T2] and [T2,+Inf].

Thank you very much for your help !!

[akidess]

Yes, pos(T - X), where T is a volScalarField and X is a dimensionedScalar will return a volScalarField.