[Othman]

Hello everyone
Please, any one can give me some information about how can I setting up a two fluid model (two phase flow) in Flow-3D program (version 11.2) to represent the air-water flow behaviors in pressurized tunnel of morning glory spillway (as showed in the attached image file). By the way, alot of times I tried to running the model by readjusting the parameters (like as follows) to give the appropriate result but, unfortunately still now I don’t know what the problem in this model that doesn’t work properly. Please kindly give me some information about setting up two phase flow modeling in pressurized tunnel and which of those parameters are incorrect according to your experts.
Thanks for providing any helps and information.
Applied properties and parameters:
Number of fluids: Two fluids
Interface tracking: No sharp interface
Flow mode: Compressible
Physics
Bubble and phase change: selected two fluid phase change
Accommodation coefficient=0.01
Density evaluation: Second order monotonicity preserving approximation to density transport equation
Drift-flux
Average drop diameter=0.001 m
Drag coefficient=0.5
Richardson- Zaki coefficient multiplier =1
Minimum volume fraction of phase #1=0.1
Volume fraction of phase #2 at inversion point=1
Heat transfer : Second order monotonicity preserving
Viscosity and turbulence: selected viscous and laminar flow

[koful]

Hello Othman.

What do you want to see as a result of this simulation? If you want to simulate the dispersed air phase inside the water, you do not need two phase flow. Just air entrainment and maybe adiabatic gas region models (inside bubble and phase change) will be sufficient.

If you want to simulate bulk air volumes you should activate two fluids. Also, you need to solve it with incompressible flow and you need to add limited compressibility.

For two fluid models, to turn on the two phase velocty slip option will help convergence and stability.

These are my general comments. Feel free to ask your further questions.

[Othman]

Hello Koful,
Thanks for replied and give me your information, really I want to simulate bulk air volumes (large air bubbles) that sometimes caused interruption the water flow inside the tunnel and does not emerging out of it's outlet continuously. For this purpose, I setting up a two fluid model (as you said) but, the simulating does not continuous, the solver reduce the time step until reached min. time step and stop simulation process. If you know, please kindly tell me which of these parameters are not corrected for this model, that displayed in the attached below.

Best regards
Luqman

[koful]

Hello Othman,

The comressibility values for fluids are very high. Try 1e-9 for both. Also you can choose split lagrangian method for volume of fluid advection methods.

[Othman]

Thanks alot for your information i really appreciated it.

[elyasmin]

Hi, I am trying to do 2-phase modeling in a pipeline. I have used your set up but the time step goes too down makes the simulation so slow like 0.5 seconds in a day. Actually, in the end, it broke.
Could you please give some advices?
Thank you!

[Othman]

Hi, elyasmin
unfortunately, still i have the same problem and exhausted to setup two fluid model. However, I tried to use adiabatic bubble model and air entrainment model as another alternatives to represent bulk air volumes in the pressurized tunnel of shaft spillway but, untill now I don't get a reasonable result. If you have any idea about it please share with me, i appreciate it.
Thanks
Othman

[aliabdulsahib]

Hi Othman
I think you should do this model two fluid, free flow, with pressure solver implicit, and you must avoid the multiblock mesh

thanks

Ali Abdul-Saheb

[MSM1985]

As i have experienced :

1) 2 Fluid model is computational intensive and requires enough system resources. 2 fluid model crashing on 8 core (8 GB RAM) computer might go fine on 16 core (128 GB RAM) workstation.

2) Check for the Zmin. and Zmax. boundary conditions and both should have same fluid elevation in case of vertical drops.

3) Since flow accelerates vertically at the simulation starts, it would be good to provide a smaller initial time step in numerics, for example 0.001

[aliabdulsahib]

Hi MSM 1985

Thank you for your reply
but I have one question about your answer in item (2)
Why, the Zmin. and Zmax. B.C. the same fluid elevation.??

Thanks

Ali Abdul-Saheb

[MSM1985]

As your are standing at the same location in X axis, so same fluid elevation will be entered in Pressure boundary

[aliabdulsahib]

Hi Dear,
I have a question.
I am currently studying the formation of bubbles and air pockets in a shafts pillway.
What physics models or conditions do you suggest that I need to run in the flow 3D ?

With my best wish,