[sega]

Hello!

I have read a two-phase flow tutorial from the university of Göteborg:
http://www.tfd.chalmers.se/~hani/kurser/OS_CFD_2007/HassanHemida/Hassan_Hemida_V OF.pdf

Its a short but very useful description of the interFoam solver.

But I have two questions:

1) About transportProperties: Citation: "It gives also some coefficients for two power laws used for the interpolation for the gamma function"

How crucial are these coefficients, and at what place are they used?
I can remember that I sometimes deleted these two entries.

2) About the implementation of the gamma transport equation. Citation "In OpenFOAM, the necessary compression of the surface is achieved by introducing an extra artificial compression term into the VOF equation (3) as follow: [...]
where Ur is a velocity field suitable to compress the interface.

Now I want to know something about this velocity field suitable to compress the interface.

How does it look like and how is it calculated?

I had a quick look into gammaEqn.H and found something about gammarSchemes, but I think this is something else ...

Greetings from Germany. S.

[holger_marschall]

Hi Sebastian,

for question (2) and further details to the compression velocity have a look to Henrik Rusche's Ph.D. thesis:

Rusche, H. Computational fluid dynamics of dispersed two-phase flows at high phase fractions Imperial College of Science, Technology & Medicine, Department of Mechanical Engineering, 2002.

Basically the compression velocity is oriented normal to the phase interface by the normalized vector ~grad(gamma).
In first instance the velocity should have a magnitude in the order of the local velocity U. However, in order to overcome problems at stagnation points (etc.) one have to limit this with the maximum velocity in the whole flow domain. In this way the compression velocity never gets zero and ensures a sharp interface. Furthermore (if you have a look on how it is implemented) the compression term is limited to both boarders of the volumetric phase fraction (gamma->0 and gamma->1) being bounded (and conservative).

All in all the (artificial) interface compression is done by a term that was introduced for numerical reasons (in oder to counteract the numerical diffusion) but does not bias the VoF-solution!

best regards
Holger

[nimasam]

hi dear foamer
some thing about interface make me messed up, i use a modified version of interFoam (interFoam with source term something like whats done interphaseChangeFoam) but my interface is highly diffusive (look fig), whats the problem? how can i solve it? should i increase Calpha? or maybe change interface compression?

[Saideep]

Hi guys,

I understand how the compression term comes into play in the alpha advection equation.

But in the fvSchemes part for the compression term we have an option to provide "interfaceCompression".
This scheme is from my knowledge somehow related to the "interfaceProperties/interfaceCompression" code which has the following:

    scalar limiter
    (
        const scalar cdWeight,
        const scalar faceFlux,
        const scalar phiP,
        const scalar phiN,
        const vector&,
        const scalar
    ) const
    {
        // Quadratic compression scheme
        //return min(max(4*min(phiP*(1 - phiP), phiN*(1 - phiN)), 0), 1);

        // Quartic compression scheme
        return
            min(max(
            1 - max(sqr(1 - 4*phiP*(1 - phiP)), sqr(1 - 4*phiN*(1 - phiN))),
            0), 1);
    }

Could anyone help me out where is this function being called and how is cAlpha related here?

Thanks,
Saideep

[JIM SU]

Hi guys and InterFoam users;

Although I use and receive the correct results with this solver, I don't know where does this code impose the necessary interface boundary conditions for velocity and shear stress (u1=u2 & tau1=tau2) ?

thanks a lot
best regards.