[Ema40]

Hello.

I am simulating an open channel flow with VOF.
The free surface is not well defined and the transition between water and air occurs through a large (not physical) transition.

There is a way to reduce this transition and to have a free surface better defined? I know that I can reduce the mesh size. Are there other methods?

Thank you

[silent2608]

Mesh size and interface reconstruction scheme are your only options as far as I know (compressive is ok, geo-reconstruct is very good but very slow)

[Ema40]

Thank you, but in Fluent only these schemes for VOF are available for the Volume fraction calculation:

- first order
-second order
- quick
-compressive
- modified hric
-bgm

What do you suggest to me?
Thank you!

[silent2608]

mesh: for vof use mapped quad mesh in 2d (equally sized rectangles)

uniform mesh size or gradual cell jumps!!!! skewed mesh will result in utter destruction


to avoid instant crash through momentum divergence set VOF model to explicit, 1e-06 volume fraction cutoff, courant 0,25 in model window (defaults)
implicit scheme is faster but you need to know with which settings you get realistic results, so I will propose conservative settings


PISO
Green Gauss Node Based
Body Force Weighted
Compressive (geo reconstruct better but more computing cost)

Transient: first order implicit (vof explicit - only first order available)

timestep: variable and courant <= 1


explicit vof + compressive + bounded second order is almost as good (and faster) but you need to know about your flow (timestep) or you will face divergence problems and unphysical behaviour


getting physical / good results requires understanding of vof model, interface reconstruction etc. pp.
take these settings and adjust after reading more on the subject, proper setup is not trivial

[Ema40]

Thank you very much!!
Regards