[yulu]

Hi all

I am simulating a not-too-complex problem with Fluent VoF model. The interface needs to be reasonably sharp because I'm focusing on interfacial phenomena. Because of this, I was planning to use the "width of the interface" or the idea of "how sharp the interface is" as the parameter for my mesh independence study. However I did not seem to find an accurate way to estimate/measure exactly the width of interface from the volume fraction contour. Obviously the more colour levels you use the more difficult to identify where exactly is defined as "interface". Attached figures are two situations where different volume fraction colour levels are used (10 levels vs 50 levels).


Similarly because of the colour level used, I also cannot know how many cells of meshing is occupied as the "interface".

But using denser mesh definitely makes the interface (identical view, same colour levels, same everything except mesh density) looks narrower (or sharper).

Is there any way to do this quantitatively properly?

Thank you very much!

[yulu]

These two are the interface new using a coarse mesh and a finer mesh. The volume fraction contour is using the same number of colour levels (20). So this "width of interface" has definitely decreased. How could I properly measure this in a quantitative way???

[EternalEnvy]

Quantitatively, you can use the cells number to represent the interface width. You may apply to UDF, where you can loop all the cells, then judge the cell by its C_VOF value. Interface cells should have a VOF value between 0 and 1.