I am a student and now working with a project which is related to the water hammer in a pipe caused by hot gas bubble collapse in cool water. I tried to solve this problem by VOF method with Fluent. However, I failed to do that.

My geometry is simple. A 2-dimensional L shaped pipe (horizontal part first and then the vertical part towards upward) was established by Gambit with velocity in, pressure out and wall boundary conditions. The pipe was initially full of hot gas, and then water flows in from the horizontal Velocity In boundary.

In fluent 6.1.22, I made the following choices: 1. Solver: Segregated, 2d, Absolute, Implicit, Unsteady, 1st-order implicit; 2. Multiphase Model: Volume of Fluid, Geo-Reconstruct, Courant Number 0.25, Solve VOF Every Iteration, Implicit Body Force, Number of phases:2; 3. Turn on Energy Equation; 4. Viscous Model: k-epsilon, Standard, Standard Wall Functions, Keep the original model constants; 5. Copy water-liquid material; 6. Define air as primary phase, and water as second phase; 7. Turn on Gravity, and Specified operating density is 1.225; 8. Boundary Conditions Velocity In: Here I have a problem. I can only define the velocity magnitude for the mixture other than water. I only want that water flows in the pipe which is originally full of hot gas. Volume fraction of water is 1. Keep other parameters as original; 9. Solution: Set all under-relaxation factors is 1, Body Force Weighted Pressure, PISO Pressure-velocity coupling. 10. Initialize and iterate. Unfortunately, I cannot get any result.

My questions are: Is VOF Method applicable to my problem? If yes, what is the best way to simulate the bubble collapse in cool water with Fluent? What is wrong with my setting with respect to Gambit or Fluent?

Many thanks for help.

William Palm

In first, try without energy (and laminar ?)

BC : velocity-inlet : you define the velocity for mixture. for phase-2 you define the volume fraction 1 (water only)

After initialize (from inlet), you patch your domain : phase-2 => 0 (air only)

Display contour to see your BC (for example phase-2)

Iterate ... (with a small time step !)

If it's ok, try with energy (rho = rho(T) ?)

Steph