[Albert Lee]

Dear friends,
I have used VOF method to simulate the mass/heat transfer of a two-phase system in a capsule tank. I use the Fluent 17.2. The case is, a patially filled tank (about 75% of liquid and 25% of vapor). The gravity level is zero. A temperature boundary condition was carried on the cylinderic wall of tank. I used 2D axisymmetric model. The initial state of phase distribution are shown as Fig. 1. Now, the problem I meet are as follows:
(1) Firstly, I didn't carry on the mass transfer model, in order to see the heat conducting process from the wall to interior fluid, including the vapor parts. After about 5 seconds calculation, the results show that the heat conducting abnormally into the interior fluid as shown in Fig. 2. I wonder whether this was caused by the spurious velocities came from the VOF numerical method? And how to suppress the spurious velocities?
Fig1 Model.jpgFig2 Temperature-flow field-no mass transfer.jpg

(2) Meanwhile, I added the evaporation-condensation model of FLUENT in my case, and I get the result as shown in Fig. 3, from which we can see that the temperature distribution become more chaotic. Some vaper partilces appear in the liquid region, as shown in Fig. 4. I don't know why could this happen? and how to deal with this problem?
Fig3 Temperature-flow field-mass transfer.jpgFig4 Phase distribution.jpg

In conclusion, my questions are:
(1) How to suppress the spurious velocities comes from VOF method in Fluent? Are there any methods to improve the calculation resolution of the source term of surface tension?
(2) Are these kind of nonphysical results from the 2D model? How about changing into 3D model?
(3) Is the mass transfer model appropriate for this kind of problem? What key points should pay attention to?
(4) Are there any other questions out of my consideration? Please point out!

Thank you very much! I am looking for your attention and advices!


Albert.