Learning Forum

[raju.chowdhury]

Hello,

I am simulating gas jet-liquid interaction, where gas jets impinge into a liquid pool on a container base. The computational domain was meshed with a cell size <= 7 mm. I use a 2D E-E multiphase model to simulate the gas-liquid interactions. The liquid droplet size was assumed as 1e-5 m. The simulation runs well using this droplet size. However, the simulation diverges when the size changes to 1e-4 m, 5e-3 m, or 1e-2 m, whereas the simulation runs well for 1e-3 m. Except for changing the size, everything in the model setup remained the same. What could be the reason for diverging at those diameters?

 

Getting the following message for the divergence:

"Divergence detected in AMG solver: epsilon" 

[Rob]

The droplet size is used for drag so look to see what's going on in the flow. If you run single phase what does the gas velocity look like? Ie are you blasting liquid out of the pool at various droplet sizes?

[raju.chowdhury]

I want to see the effect of droplet size on wetting area on the container wall. Gas jet pattern using single phase (gas) looks okay.

[SRP]

 

 

[raju.chowdhury]

Your reply looks blank!

[Rob]

And diameter will be a function of surface shear etc and, additionally downstream droplet break up will alter the pattern. That's going to be complicated, and may not suit the Eulerian model well. How fast is the gas jet? 

[raju.chowdhury]

Mach Number ~ 0.8.

[Rob]

Which may mean you need VOF to correctly pick up the waves & sheet break up then alot of mesh. You're also likely to need a very small time step to capture all of that as VOF really won't like the velocity gradient over the free surface.