[Novel]

Hello,
I'm simulating sloshing in a system with two liquids. Sloshing is generated due to Lorentz force. The geometry is a 10x10x10 cm cube with anode and cathode at the top and the bottom. When simulating, the diffusion is higher when I increase the mesh resolution. I'm using fixed time stepping as long as my courant number is lesser than 0.1 otherwise the time step is reduced. The surface tension is set to 0.

My alpha:

    "alpha.*"
    {
        nAlphaCorr      2;
        nAlphaSubCycles 3;
        cAlpha          1;

        MULESCorr       yes;
        nLimiterIter    3;

        solver          smoothSolver;
        smoother        symGaussSeidel;
        tolerance       1e-12;
        relTol          0;
        maxIter         1000;
    }

Does anyone know why the diffusion might increase?

PS: Let me know if you need more informations.

regards,
R.

[saddy]

hey , i have been working for a long time using interfoam and interdymfoam
the problem is the fact that more the no of cells you increase, more the spurious currents appear, more smeared/diffuse interface occurs.
the solution is here : http://www.tfd.chalmers.se/~hani/kur...ankarMenon.pdf

however my suggestion is to couple level set with vof in INTERDYMFOAM
PLz let me know if you make any progress

[Novel]

Hi,
thank you for your answer do you have any reference regarding this topic?
I thought spurious currents only occurs if surface forces are calculated but as I mentioned in the question I have no surface tension.

[saddy]

the problem of interface diffusion is due to the way the volume of fluid is implemented in openfoam.
openfoam uses MULES which always produces diffused interface when interface is movving with high velocity and if small volume fraction is what we trying to track.

there are two solutions to this problem
1. volume of fluid+level set implementation
2.implementing Geometric volume of fluid, named as ISOADVECTOR its recently released in openfoam-v1706, interfoam - is modified to interflow , u can try it
its branch is also available on github, just google isoadvector

as far as your questions about references, i don't understand it
a number of researchers have implemented vof+level set and within my field of intrests , you can google "EVAPVOFHARDT" a two phase flow solver with boiling done by kunkelmann and he aslo faced this problem and solved it by coupling vof+level set with dynamic mesh
his solver "evapvofhardt is also publicly available and it doesn't show any diffusion at all.
must try it
let me know what you think