[Rahul Agarwal]

Dear Foamers,
Greetings!


Can anyone kindly let me know how I can construct a sharp interface while modelling a droplet using VOF interfoam (using MULES).


Even after using compression schemes, when I initialize the volume fraction using setFields, I get a diffuse interface upon visualization.


Many thanks.

[Yann]

Hello,

After running setFields you should not have any diffusion of the interface. (setFields just sets alpha to 0 or 1)

So I guess your issue might be related to visualization in paraView: uncheck the "Skip Zero Time" option when loading your case if you want to check the fields initialized with setFields on time 0. Then make sure you are plotting cell values rather than node values (see attached screenshot)

I hope this helps,
Yann

[Rahul Agarwal]

Dear Yann,


Thank you for your response. For the initial time (t =0), the interface produced is sharp.


I have one more issue concerning this - Using VOF with compression schemes and MULES, results in diffusive interface as the time progresses. However, if you see this paper - "Droplet impact on deep liquid pools: Rayleigh jet to formation of secondary droplets" (https://journals.aps.org/pre/pdf/10....RevE.92.053022), they have sharp interface obtained under similar schemes as the time progresses. Do you an idea concerning this?

[Yann]

Hello Rahul,

My first idea would be the mesh: how fine is your mesh compared to the article your linked?

AFAIK, you will not get a perfectly sharp interface, there will always have diffusion at the interface with VOF, even with compression schemes. At least with the default OpenFOAM implementation. I only had a quick look at the article you mentioned, so I don't know if the authors implemented other methods or tweaked the default parameters.

Yann

[2538sukham]

Diffusion is inherent in scalar transport eqs. If it is purely Hex cells, dynamicRefineFvMesh will refine the cells where isolines are at alpha1 = 0.5. But then, the mesh size matters. Hope this helps.