[manxu]

Hi,

I am using a VOF model to simulate water droplet motion by applying a body force. I am using :
- VOF Explicit + CICSAM + PRESTO
- Body force = gravity. (using the implicit body force formulation increases the time for convergence. SO I switched off the formulation, though I have gravity)
- Specified operating density = 0 (since air and water are present)

Mesh size is 2microns and time step is 1e-8. Even if I have 200 iterations per time step, it goes through all the 200 iterations per time step without reaching the absolute convergence (fluent default) point.

Is this OK as long as it doesn't diverge or should I increase the no. of iterations per time step ?

P.S: I tried reducing the time step to 1e-9, still no convergence. (no divergence either)

Thanks

[Zaktatir]

What about the boundary condition! and The reference of your pressure

[manxu]

All the boundaries are pressure-outlets with only one wall at the bottom. The reference pressure is 101325Pa at 0,0.

[Zaktatir]

your mesh is very fine and if spurious velocities are great then you won t mange to avoid them. By the way what about the surface tension force!

THE CSS is phyically more conservative and may enhance stability.

Do you want to make a DNS very resolved simulation i find your mesh very fin? How big is the droplet?

I simulated a couple of weeks ago for some folks the spreading and the impact of droplet on the wall and it was high converged

[manxu]

The droplet's radius is about 50microns. Yes surface tension force is quite important but the body force is larger than surface tension forces.

What is CSS and do u think that the parameters/discretization i used are correct ?

[beer]

Hi

I find it often helpful to use the selective AMG solver as well as ILU and BICGSTAB instead of the standard smoothing and solving of the differential equations. I had a few interesting convergence improvements in multiphysics in the past.

Best Regards