[MaryBau]

Hi!

In kEpsilon.C, G is defiened as:

    volScalarField::Internal G
    (
        this->GName(),
        nut.v()*(dev(twoSymm(tgradU().v())) && tgradU().v())
    );

Why is it defined as "volScalarField::Internal" instead of just a simple "volScalarFiled"? And have something like this:

volScalarField G1= this->nut_*(dev(twoSymm(tgradU())) && tgradU());


Thanks,

Mary

[Tibo99]

I got the same question lately. Did you get an answer for your question?

Regards,

[olesen]

Originally Posted by Tibo99

I got the same question lately. Did you get an answer for your question?

The question to ask yourself: if it were a volScalarField, which is an internal field (with dimensions) and a boundary - what would you like to calculate on the boundaries?

[granzer]

This is what I have understood.
The volScalarField is a typedef of GeometricField<>. So volScalarField contains an internal field a boundaryfield etc. Now, the volScalarField::Internal is the 'type' of the internal field in the volscalarField. It's defined by typedefing of Diemnsionedfield.
For volScalarField the 'Internal' translates to:

typedef DiamensionedField<Scalar, volMesh>    Internal;

So, the code

    volScalarField::Internal G
    (
        this->GName(),
        nut.v()*(dev(twoSymm(tgradU().v())) && tgradU().v())
    );

Can be seen as

    DimensionedField<scalar, volField> G
    (
        this->GName(),
        nut.v()*(dev(twoSymm(tgradU().v())) && tgradU().v())
    );