[Prosper]

Hi All,

I am trying to solve a multiphase problem in which a spherical (3D) drop deforms in a surrounding fluid under certain body&interface forces. I am using interDyMFoam to solve this problem.

To get an accurate interface shape, I let the program ran for a few steps and the interface looks OK to me. (see attached)

However, the problem is that I got none-zero velocity magnitude on the interface, even though no external forces are present. (as shown in attached plot) I don't know why this is the case. I think the interface shape is smooth enough not to induce interface tension driven velocities.

Another problem I found is that, when I run the simulation with certain body force and check the interface deformation after some time, I found that the volume of the sphere seem to be shrinking. I wonder if this is caused by the interface not being accurately captured, or some other errors?

I appreciate any comments and suggestions. Thank you!

-P

[Prosper]

Just want to mention that the volume change is not really large, but my calculations require much higher accuracy.

[mihaipruna]

VOF methods suffer from interface smearing.

Check out this paper and follow the references for the interface sharpening model.
http://www.djs.si/proc/nene2011/pdf/802.pdf

[ybapat]

Hello ,

If you are using surface tension these velocities can appear as a result of parasitic current. You can read about it, as it is well reported phenomenon with VOF model.

Regards,
-Yogesh

[micpage18]

Hi,
I had a problem similar than yours. It was with a very slow velocity at the interface and spurious currents were appearing after a some seconds. I succeed to resolve my problem by changing default schemes to linear schemes.

But, it'll probably not be sufficient for your case. Recently, they add a face-based momentum equation to help with this problem. You can give a try with the last development version via GIT.

https://github.com/OpenFOAM/OpenFOAM...777bb10a3e1dda

Good luck

[Prosper]

Originally Posted by micpage18

Hi,
I had a problem similar than yours. It was with a very slow velocity at the interface and spurious currents were appearing after a some seconds. I succeed to resolve my problem by changing default schemes to linear schemes.

But, it'll probably not be sufficient for your case. Recently, they add a face-based momentum equation to help with this problem. You can give a try with the last development version via GIT.

https://github.com/OpenFOAM/OpenFOAM...777bb10a3e1dda

Good luck

Hi Michael,

Thanks for your reply!

I will try Weller's new method, it looks promising. I would also like to try your ways of using the linear schemes. I think most of the schemes I am using now are different forms of linear schemes . Here is my fvSchemes file (same as default):

FoamFile
{
    version     2.0;
    format      ascii;
    class       dictionary;
    location    "system";
    object      fvSchemes;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

ddtSchemes
{
    default         Euler;
}

gradSchemes
{
    default         Gauss linear;
}

divSchemes
{
    div(rhoPhi,U)   Gauss upwind;
    div(phi,alpha)  Gauss vanLeer;
    div(phirb,alpha) Gauss linear;
    div((muEff*dev(T(grad(U))))) Gauss linear;
}

laplacianSchemes
{
    default         Gauss linear corrected;
}

interpolationSchemes
{
    default         linear;
}

snGradSchemes
{
    default         corrected;
}

fluxRequired
{
    default         no;
    p_rgh;
    pcorr;
}

Looks like the only one that is not linear is 'snGradSchemes'. Can you tell me if you made any other changes? Thanks!

-P

[Prosper]

Originally Posted by ybapat

Hello ,

If you are using surface tension these velocities can appear as a result of parasitic current. You can read about it, as it is well reported phenomenon with VOF model.

Regards,
-Yogesh

Hi Yogesh,

Thanks for your reply!

I did some search and found that this is has been a problem in many simulations... Looks like I need some special treatment for the interface.

Best,
-P

[Prosper]

Originally Posted by mihaipruna

VOF methods suffer from interface smearing.

Check out this paper and follow the references for the interface sharpening model.
http://www.djs.si/proc/nene2011/pdf/802.pdf

Hi Mihai,
Thanks for your link!
The two references they cited showed very good results of volume conservation. I may need to do some work to implement that. Thanks!

-P

[micpage18]

Hi Prosper,

You can try theses change

div(rhoPhi,U) Gauss linear;
div(phi,alpha) Gauss vanLeer01;
div(phirb,alpha) Gauss interfaceCompression;

Best regards,

[Prosper]

Originally Posted by micpage18

Hi Prosper,

You can try theses change

div(rhoPhi,U) Gauss linear;
div(phi,alpha) Gauss vanLeer01;
div(phirb,alpha) Gauss interfaceCompression;

Best regards,

Hi Michael,

Thanks for your help!

I tried this and found that the interface velocity has a smaller magnitude than before. But I would still get a nonsmooth interface shape after some time.

Also, I still observed the total volume of the sphere expanding after certain time. Even where the velocity is almost zero, the interface could still be moving. Don't know exactly why. I'll try other ways that can possibly preserve the volume. Thanks.

-P