Learning Forum

[Cathleen]

Hi All, 

 

I'm a relative novice to fluent. I'm running into issues trying to model evaporation from a liquid pool into a vapor+gas space. 

I am using the multiphase(VOF) model with two phases (water vapor and water liquid). I have mass transfer via evaporation/condensation enabled between the two phases. 

It is transient and energy model is on. 

 

The geometry is a skinny rectangle (.1mm x 2mm) mesh is square (0.02mm). BC is a temperature at the bottom wall, sides are symmetry. 

 

I separate the liquid and vapor phases via cell registers and patch to start with liquid on one half of the geometry and vapor on the other half. 

 

When I run this, the residuals bounce around (do not seem to be converging) and the liquid seems to diffuse into the vapor-space. I would expect to see the liquid interface recede as the water evaporates - not diffuse. I do not have gravity on (but the results are the same regardless). Any ideas what I am missing here? Thanks!
 

 

[shitizsehgal]

I think if the fluid is stationary, you don't need k-epsilon model. Could you please provide some more information about the problem that you are trying to simulate( A paper or some document) ? I can try to help.

[Cathleen]

Hi, I've tried to reply twice, does not seem to be posting. I'm going to try and post the documentation as an image instead of a PDF. 

The fluid is stationary. Thanks for the feedback on the k-epsilon model, I will take a look at that. The overarching problem is a falling menisicus and I am trying to model (track) the interface between the liquid and vapor layers. I've reduced the problem to a large scale, two-phase stationary problem. It *should* show the interface receding as the fluid evaporates but instead the liquid water moves away from the heat source (no gravity). I've attached some slides that show the model in more detail. Thanks for your help.

[Rob]

From a computational point of view, why would vapour evolve at the free surface and not at the heat source? Surely the heat source is the hotest part of the domain? 

[Cathleen]

Hi Robert, I'm not intending on modeling boiling. Perhaps I should have the wall and initial condition at the same temperature. I'm trying to isolate the evaporation from the free surface so ideally the interface should drop as the liquid evaporates. You're right, with the way I have it set up with the surface at 200C & the initial at 100C, I should see vapor generating at the wall and bubbling up, but I don't see that either. 

[Cathleen]

Just to follow up, I tried it with the initial conditions and wall BC set to 90C and it does the same thing (interface is all dispersed and the liquid level doesn't "drop").

[Rob]

OK, slightly more complicated. VOF doesn't have a limiter (yet) to have phase change solely at the interface so you'd need to tag those cells and go from there. That may be changing so read the release notes in the next couple of versions. 

[Cathleen]

Hi Robert, when you say tag those cells do you mean to specify that evaporation will only occur around a specific group of cells? If so, how would I update it as the interface moves out of those cells? Thanks!