[Naaren]

Hi,
I am working with the solver interThermalPhaseChangeFoam (https://www.sciencedirect.com/scienc...52711016300309) to simulate the growth of single bubble. When I simulate the bubble in a three dimensional domain (Cube, with bubble at center) the bubble starts to slide on the surface. It be very much helpful if someone can suggest a solution to debug this issue.


Thank you,
M Naarendharan

[Naaren]

Hi,
I am still facing the issue. Any suggestions would be very much appreciated.


Thank you,
M Naarendharan

[Tobermory]

Almost impossible to help you debug with so little information, but here's a quick suggestion - check your gravity vector. Is it pointing in the direction that you are expecting?

[Naaren]

Hi,
Thank you for your reply. I have provided the gravity vector in the negative Y direction. My study is growth of a single bubble in saturated liquid inside a cubical domain with the bottom face having a constant temperature above the saturation temperature.
What is expected: The bubble is set at the base and is expected to leave the base after growing in size.


Problem: The bubble grows in size and starts moving on the heated plane.





Image files of the problem: https://drive.google.com/drive/folde...zz?usp=sharing


Setfields, blockMesh, Boundary Conditions: https://drive.google.com/drive/folde...7y?usp=sharing




I would be happy to provide any other details.


Thanking you,
M Naarendharan

[ErenC]

First, make sure your thermophysical properties are same with benchmark problem. Also for interFoam, mesh has huge impact, if your mesh is coarse you cant capture interface accurately and in each time-step you will have worst and worst solutions.

[Naaren]

Hi Eren,
Thank you for your reply. I have given the proper thermophysical properties. Also, I validate a case in a 2D axisymmetric domain. The issue of bubble movement comes when I shift from a axisymmetric domain to a 2D (square) or 3D geometry. I have tried running for various mesh sizes of the order of microns (Usually used mesh size for capturing interface in a 2-phase simulation) as well, but it did not help in preventing the movement of the bubble. I tried changing the evaporation threshold as posted in this forum: Bubble is not increasing in size in Nucleate boiling. But even that suggestion was not able to arrest the movement of bubble.


Thanking you,
M Naarendharan

[Tobermory]

This is an interesting one. I guess that there are 3 possible reasons why the bubble might move:

1. buoyancy force (but you say it's moving normal to the gravity direction, so this can't be the case)
2. asymmetric surface tension from its contact with the lower wall (I don't see why this would be the case)
3. the background fluid is moving laterally and dragging it along

Have you plotted the pressure, velocity and density fields? Do you see movement in the background flow? What are the BCs on the side walls? Try thinking about the above, and see if you can track down the source of the movement. The pix you shared are interesting, but for anyone to truly help you, you'll need to share the actual case files, not just pix and a few random setup files. Good luck.

[Naaren]

Hi,
Thank you for your valuable suggestions. I cannot see any movement in the background flow, I checked them with the velocity and pressure contours in paraview.

I will try your other suggestions and get back to you.

I am also attaching the complete case files for this simulation: https://drive.google.com/drive/folde...PF?usp=sharing


Boundary Condition on side walls:
1.alpha: type inletOutlet;
inletValue uniform 1;
value uniform 1;


2. H: type zeroGradient;


3. p_rgh: type zeroGradient;


4. T: type inletOutlet;
inletValue uniform 443;
value uniform 443;


5. U: type zeroGradient;


"A" folder in the link is the boundary conditions file. Axis is left wall and FarStream is right wall.


Thanking you,
M Naarendharan

[Tobermory]

I had a very brief look at the case this morning, and am a bit confused. I am not familiar with this solver, but a couple of things stuck me as odd and are worth you checking out.

Your pressure field p_rgh is initialised uniformly with zero. Firstly, are you sure that the solver works with gauge pressure, and not absolute pressure?

Secondly, I would have expected to see some differences in p_rgh within the vapour bubble since the density is different to the background fluid (water). My advice for all OpenFOAM simulations is to pay very careful attention to your starting conditions.

Lastly - are you sure you want to apply zeroGradient on p_rgh for the side boundaries, and not fixedFluxPressure? Not sure on this one, but again worth looking at perhaps.

Good luck.

[Naaren]

Hi,
Firstly, thanks for your time and effort to help me.
1. Yes, the solver is developed from interFoam and is compatible with gauage pressure. I can be sure about the solver working fine with gauge pressure since, there are many tutorials which comes along with this solver and I did validation for a case.
More Info on Solver:
Solver: https://github.com/MahdiNabil/CFD-PC
Reference: https://www.sciencedirect.com/scienc...52711016300309


2. Yes, there is a difference between the values of p_rgh within and outside of the bubble. But, I didnt see any difference in background to cause the bubble to move.
I will again have a look at this. Thanks for pointing out.

3. I will change the zeroGradient to fixedFluxPressure BC in side walls and get back to you.
Once again, thanks a lot for your time and effort.


Thanking you,
M Naarendharan