[cagkn]

Hello everyone,

I'm trying to form a 2D numerical wave tank in OF and using interFoam solver with turbulence. When you try to include both turbulence and wave action in the same model waves starts to dampen as they propagate. VOF scheme in the interFoam is the reason for that, and I'm experiencing the exact thing.

In "Multiphase and Free-Surface Flow Simulations" presentation, Mr. Paterson showed a way to overcome that problem by modifying the air convection eq. by multiplying it with phase fraction, gamma.

in UEqn.H, I've add the red line but,

MRF.correctBoundaryVelocity(U);

    fvVectorMatrix UEqn
    (
        fvm::ddt(rho, U) 
      + gamma*fvm::div(rhoPhi, U)
      + MRF.DDt(rho, U)
      + turbulence->divDevRhoReff(rho, U)
     ==
        fvOptions(rho, U)
    );

when I try to compile it, ends with this error.

/opt/OpenFOAM/OpenFOAM-v1812/src/finiteVolume/lnInclude/fvMatrix.C:2310:33: note: candidate: template<class Type> Foam::tmp<Foam::fvMatrix<Type> > Foam::operator*(const Foam::dimensioned<double>&, const Foam::tmp<Foam::fvMatrix<Type> >&)
 Foam::tmp<Foam::fvMatrix<Type>> Foam::operator*
                                 ^~~~
/opt/OpenFOAM/OpenFOAM-v1812/src/finiteVolume/lnInclude/fvMatrix.C:2310:33: note:   template argument deduction/substitution failed:
In file included from interFoamMod.C:159:0:
UEqn.H:6:33: note:   cannot convert ‘gamma’ (type ‘double(double) throw ()’) to type ‘const Foam::dimensioned<double>&’
       + gamma*fvm::div(rhoPhi, U)

What is the right way to add phase fraction value in the equation?

Sorry for the long post

Thanks in advance

Cagkan

[Mat_fr]

Hi,


Which solver did you use ?
How is gamma declared ?
This is just a compatibility problem.


Regards,
Mat

[cagkn]

Originally Posted by Mat_fr

Hi,


Which solver did you use ?
How is gamma declared ?
This is just a compatibility problem.


Regards,
Mat

Hi Mat,

I'm using openFoam v1812. The Base solver that I'm modifying is interFoam solver.

What do you mean with gamma declaration? sorry but I didn't get it

[simrego]

Originally Posted by cagkn

Hello everyone,

I'm trying to form a 2D numerical wave tank in OF and using interFoam solver with turbulence. When you try to include both turbulence and wave action in the same model waves starts to dampen as they propagate. VOF scheme in the interFoam is the reason for that, and I'm experiencing the exact thing.

In "Multiphase and Free-Surface Flow Simulations" presentation, Mr. Paterson showed a way to overcome that problem by modifying the air convection eq. by multiplying it with phase fraction, gamma.

in UEqn.H, I've add the red line but,

MRF.correctBoundaryVelocity(U);

    fvVectorMatrix UEqn
    (
        fvm::ddt(rho, U) 
      + gamma*fvm::div(rhoPhi, U)
      + MRF.DDt(rho, U)
      + turbulence->divDevRhoReff(rho, U)
     ==
        fvOptions(rho, U)
    );

when I try to compile it, ends with this error.

/opt/OpenFOAM/OpenFOAM-v1812/src/finiteVolume/lnInclude/fvMatrix.C:2310:33: note: candidate: template<class Type> Foam::tmp<Foam::fvMatrix<Type> > Foam::operator*(const Foam::dimensioned<double>&, const Foam::tmp<Foam::fvMatrix<Type> >&)
 Foam::tmp<Foam::fvMatrix<Type>> Foam::operator*
                                 ^~~~
/opt/OpenFOAM/OpenFOAM-v1812/src/finiteVolume/lnInclude/fvMatrix.C:2310:33: note:   template argument deduction/substitution failed:
In file included from interFoamMod.C:159:0:
UEqn.H:6:33: note:   cannot convert ‘gamma’ (type ‘double(double) throw ()’) to type ‘const Foam::dimensioned<double>&’
       + gamma*fvm::div(rhoPhi, U)

What is the right way to add phase fraction value in the equation?

Sorry for the long post

Thanks in advance

Cagkan

Hi!


Your gamma declaration is wrong. You have to declare gamma as a dimensioned double.
Read these lines:

UEqn.H:6:33: note:   cannot convert ‘gamma’ (type ‘double(double) throw  ()’) to type ‘const Foam::dimensioned<double>&’
       + gamma*fvm::div(rhoPhi, U)

[cagkn]

Hi,

Thank you for responding

I've read the error code, my question is how to declare gamma correctly.

I've tried "gamma()" and "gamma.value()" but couldn't managed to compile it.

[simrego]

You can find constructors for a dimensionedScalar here:
https://cpp.openfoam.org/v6/dimensio...ce.html#l00076


BUT! You said "multiplying it with phase fraction, gamma."
Then why don't you use the alpha instead? If I'm correct you can access it with alpha1 or alpha2, or maybe mixture.alpha1() and mixture.alpha2(), honestly I'm not sure which is the correct one. These are volScalarFields and I think you can use them in the equation.

[Ramzy1990]

Good Day,

You will have to declare gamma for the code to understand it. In other words, the code does not know what is gamma, or, it is defined inside the code already with another definition that is not suitable for the equation.

1- The gamma you mentioned seems to be a scalar. Hence, you want to tell the code that. This is done by declaration in side a file called creatFields ( probably) related to the solver you are using. If you open it, just copy the definition used for "rho" and use it for gamma, it will be a volScalarField. Moreover, you have to search where the "rho" is mentioned ( i.e. not only its definition as a volScalarField) and do that for gamma as well.

Note that you will have to initialize it then, i.e. use a boundary file for it like you do for velocity for e.g.

2- Another better method is to define it as a constant ( if it is indeed a constant ). You will then have to declare it as a scalar, and make it to be read from a dictionary and such.