[Luuuca94]

Hello everyone,

I am doing VOF-Simulations in Ansys Fluent and I am interested in modeling condensation phase change with the Lee model (Evaporation-Condensation Model).
I am simulating condensation of water vapor on a flat surface. For this, I have defined a computational domain that is composed of a copper base-plate with a cubic fluid domain on top (see attached images). The top-surface of the cubic domain is defined as a pressure inlet while the lateral surfaces are defined as periodic walls. The complete settings are the following:

• Gravity perpendicular to surface where droplet sits
• Models: VOF (implicit body force ON), Energy ON, Laminar flow
• chosing water vapor and water liquid from fluent database (with given standard state enthalpies) and setting water as primary phase and vapor as secondary phase.
• Mass transfer mechanisms: evaporation- condensation frequency of 100'000.
• Surface tension: Wall and Jump adhesion, coeff. = 0.072 N/m (between liquid and vapor)
• Chosing copper as plate material, contact angle = 120°, 350x350x350 micrometers computational domain with the copper surface as bottom surface.
• Boundary conditions:
○ plate temperature 323.15K
○ interior fluid temp: 373.15 K
○ top-surface: pressure inlet (gauge pressure 0 bar, 373.15K inlet temp, volume fraction liquid = 0)
○ wall1-3 periodic boundary
○ wall2-4 periodic boundary
○ Operating conditions: set operating pressure at top-surface, specified operation density: 0.5542 kg/m3
• Solution methods: SIMPLE, Second Order Upwind for Energy and Momentum, Geo-reconstruct for volume fraction, default for rest.
• Convergence criteria: 10-3 for continuity, momentum and volume of fraction equation, 10-6 for energy equation
• Initialization: standard

The issue I am facing is that with this setup, the amount of water condensed on the surface is not compatible with the Wall Heat Flux measured at the bottom surface. In other words, with the wall heat flux that is measured at the bottom copper plate, the amount of water condensed should have been substantially less, considering also the difference in standard state enthalpies between water liquid and water vapor. With this setup, I get about 10-20 times the amount of water condensed that is expected when taking in account the wall heat flux at the bottom plate.
Now, If I invert the phases (vapor as primary phase and water as secondary phase) and let the rest of the setup identical, the amount of water condensed is much less and is compatible with the wall heat flux measured at the bottom copper plate. This means that the Heat Transfer Coefficient calculated by taking in account the wall heat flux is about the same as the heat transfer coefficient calculated by taking in account the amount of water condensed in the given flow time.

If anyone has a clue why inverted phases cause such a difference in the amount of condensed water it would be of great help.

I thank you in advance and best regards.