I use the eulerian multiphase model to solve water/air mixing flow. Water and air flow to a big volume from two inlets(water-inlet and air-inlet). The outlet is at the bottom of the big volume. I take two steps to solve this problem. The first step,I didn't select the volume fraction in solution controls panel. When I acquired the converged solutions in the first step, I select volume fraction to iterate. But after some iterations, there are reversed flow in faces of outlet and give the message:"divergence detected in AMG solverressure correction". Thanks for all references!

- Use transient Eulerian Model. - You *MUST* select volume fraction simultaneously to solve the case. - If you get divergence, try to reduce the time step size as well as the pressure under-relaxation factors (specially of pressure).

Dear Anshul

I am also facing same type of problem only. If you can get the doubt cleared, please get back to me. But the same problem works well with vof model.

Regards Karthick

What I have figured out is that VOF model requires really small time-step (at times 1/100th of the Eulerian model) which makes the whole simulation really slow. Hence, I prefer Eulerian model although, sure, you do have to feed in the droplet size. 100micron is normally the decent size to start (actually max size). You can reduce it to 10micron but don't go below that. It takes a little time to get the fair enough droplet size. However, still it's worth it since you might do the whole job in less than 1/10th of the time required for VOF model..

Dear Anshul

You are right as per the vof is concerned. But sometimes we have to give phase-phase interaction also which is available only in eulerian model. I cannot move over to VOF model.

Regards Karthick