[onlyacan]

Hello, dear openFOAM user,

For a incompressible flow, the flux phi used in convection term

fvm::div(phi, volScalarField)

is calculated in the "createPhi.H" as

linearIntepolate(U) & mesh.Sf()

Now I have a term like div( gl * U T), where gl is an extra coefficient. My current solution is create a new flux called phi_mul_gl as

phi_mul_gl = linearIntepolate(gl) * phi

and then we have

fvm::div(phi_mul_gl, volScalarField)

It works, but I don't know how this kind of term is normally handled in your case?

[Jerryfan]

Hi Can,


I think the way you use is OK. But if I were you, I might do it a little bit different.


surfaceScalarField phi_mul_gl
(
IOobject
(
"phi_mul_gl",
runTime.timeName(),
mesh,
IOobject::READ_IF_PRESENT,
IOobject::NO_WRITE
),
linearInterpolate(gl*U) & mesh.Sf()
);

Then,


fvm::div(phi_mul_gl, volScalarField)


Any way, it's the same thing. I just don't want to linearInterpolate twice.