[TerryNiu]

Hi friends!
I am learning CFX and there are some basic problems that confuse me.

I'm going to simulate gas-liquid two phase flow whose gas volume fraction is very high. Like annular flow. The volume fraction of gas ranges from 90% to 99%.
Which model should I choose? Eulerian-Eulerian or Lagrangian Particle Tracking?
After reading the model guide, maybe Particle Transport Modeling is appropriate? i.e. Treat the liquid phase as dropltes?

I am going to observe the velocity profile of two phase after two phase mixture flow through a vane type swirler.
The structure of it like the picture I attached.

Could someone help me, thanks

[Gert-Jan]

If the liquid volume fraction is 1%, then LPT is the way to go. But if you have annular flow with 10% liquid volume fraction, then Euler might be better.
It all depends on the question that you are trying to answer when modeling CFD. Can you tell us more?

[TerryNiu]

Originally Posted by Gert-Jan

If the liquid volume fraction is 1%, then LPT is the way to go. But if you have annular flow with 10% liquid volume fraction, then Euler might be better.
It all depends on the question that you are trying to answer when modeling CFD. Can you tell us more?

Thanks your reply.
I am going to observe the velocity profile of two phase after two phase mixture flow through a vane type swirler.
The structure of it like the picture I attached.

[Gert-Jan]

Difficult problem. Main uncertainties are with the droplets hitting the wall. Will they stick, coalescence, break up, re-entrainment, rivulets, etc. And if they re-entrain, what wil lthe droplet size be.....??

Nevertheless, if you are only interested in the velocity profile, then you could opt for LPT.
If you are interested in where the liquid ends up, then Euler might be better. Maybe it is even better to use a model that can handle both at the same time. Then mass can go from LPT to Eu, vice versa. These are available in fluent or Star-CCM+. Not in CFX.

[TerryNiu]

Originally Posted by Gert-Jan

Difficult problem. Main uncertainties are with the droplets hitting the wall. Will they stick, coalescence, break up, re-entrainment, rivulets, etc. And if they re-entrain, what wil lthe droplet size be.....??

Nevertheless, if you are only interested in the velocity profile, then you could opt for LPT.
If you are interested in where the liquid ends up, then Euler might be better. Maybe it is even better to use a model that can handle both at the same time. Then mass can go from LPT to Eu, vice versa. These are available in fluent or Star-CCM+. Not in CFX.

Thanks for your analysis and good advice, I think i should try the both model and find is there any differences.