[Gimlas]

Dear forum users,

I would like to ask you about a simulation of water condensation from humid air. I have created a geometry of vertical pipe of diameter 0.5 m and height 2m. It was meshed using nonstructural mesh. An inflation layer consisting of 30 sublayers was created where the thickness of the first layer was equal 500um.
Fluent settings:
- gravity on
- energy on
- multiphase VoF (2 eulerian phases)
- standard k-e model with enhanced wall treatment
- species on

I have created mixture template consisting of air and water vapor as the primary phase (properties: incompressible ideal gas, sutherland eq. for viscosity, the rest of parameters were constant). The second phase was water (liquid) of constant properties. The phases interactions were set as follows:
-mass transfer - evaporation-condensation between phase 2 and phase 1 species H2O vapor: constants - evaporation 0.1, condensation 0.9. Condensation temperature 373.15 K
- surface tension coefficient between phases is constant 0.07 N/m

Boundary conditions:
- the primary phase is flowing from the bottom of the pipe to the top
- inlet pressure 5 Pa (which gives 3,6 m/s, Y+<5)
- H2O vapor mass fraction 0.2
- inlet temperature 380 K
- secondary phase volume fraction at inlet 0
- outlet gauge pressure 0 Pa
- wall constant temperature 283 K

All equation were set for second order. Default Courant number 0.25. Time step 0.001 s. Number of time steps 1000. Iteration per time step 50.

The simulation run smoothly. At each time step the convergence is achieved at residuals level 1e-06.

AND NOW THE PROBLEM

Although the contour graph clearly shows that at the cold wall of the tube the mass transfer appears with constant reaching 2e-02 kg/m3s (as the phenomenon defined above it has negative value), there is no secondary phase in the domain. Its volume fraction is equal to 0. The net mass flux is approximately equal to 0. This means that there are no water droplets left in the pipe. They did not leave the domain because the mass fluxes of liquid water at the boundaries are equal to 0. What am I doing wrong? Why there are no liquid droplets or film creation on the pipe wall?

Thank you for your help.

[Gimlas]

another question related to this topic:
According to Ansys help:
26.2.8.*Including Mass Transfer Effects
evaporation-condensation
enables you to apply the evaporation-condensation model as the mass transfer mechanism. This model is available with the mixture and Eulerian multiphase models.

Does this mean that evaporation-condensation model cannot be used with VoF model?

I've made some calculation using mixture model and the vapor condensated at the cold wall of the tube. But the velocieties of both phases are amost identical (upward) which seem to be unlikely.