[s-rod]

Hi,

I've been working on the simulation of a simple vessel with an inlet and two outlets (one on top and the other on bottom). The gas-liquid mixture gets into this vessel and it is split in two streams by a tee. Every stream enters a pipe that acts as a cyclonic separator.

I ran tests in the laboratory and I found that the systems reaches a quasi-steady state when the liquid gets a determined level, and I have no liquid int top outlet nor gas in bottom outlet.

I need to get a cfd model that describes this behavior. I generated a coarse mesh and started to run simulations with the mixture model and the k-omega SST model, steady state solver, the gas being the primary phase. But the model was highly divergent. I healed the geometry in some areas where was not good and the system did not diverge anymore, however the solution I am getting has a lot of gas in the bottom outlet.

This is my first complete multiphase simulation and I've tried a lot of things, for example, i refined the mesh, I changed to K-epsilon, I ran the problem with no liquid at all in the inlet and then tried to incrementally add the liquid in the inlet stream, but I always find that the solution has a lot of gas in the bottom outlet or the system diverges.

My questions for you are:
1. Is there any tips and tricks manual o best practices text in multiphase situations that can help me?
2. What turbulence model should I be using? Do you agree with me on K-omega SST.
3. Most important: Do you know any tricks I can use to improve the solutions I am obtaining?

I append a raw image of the process (please don't be rude on my bad-drawing skills XD)

I really appreciate your help and thank you in advance,

PD: I also tried changing relaxation factors, and eulerian model. Nothing worked.

[carlaoc19]

Turbulence: RSM
multiphase model: VOF