[kaifu]

Hi Foamers,

In two phase flow solves (like bubbleFoam and twoPhaseEulerFoam), the pressure is constructed by linearly combination of phase continuity eqn. (see page 112, Rusche phd thesis). There are two kinds of combination: mixture mass continuity and volumetric mixture continuity. In adiabatic case (like water and air), the volumetric mixture continuity is adopted because "the lighter phase is poorly conserved when the mixture mass continuity equation is used". However if we consider the subcooled nucleate boiling, since there is a phase change, we have to use the mixture mass continuity equation instead. And then the solver doesnt work. I have my own developed solver for adiabatic case (which starts from solver twoPhaseEulerFoam) which adopts the volumetric mixture continuity eqn ant it works well. However if I only change the volumetric mixture continuity eqn into mixture mass continuity eqn and nothing else is changed including the test case(2D adiabatic upward bubbly flow, refer to http://www.cfd-online.com/Forums/ope...ated-wall.html), the velocity of both phase diverged. I can see the max velocity is initially generated in vicinity of inlet. It moves from inlet to outlet, and keeps increasing during simulation. And that is a big problem. Anybody could give me suggestions or hint? I am quite sure there is a way to handle this problem since nucleate boiling model is already implemented in fluent and star-cd... Thanks very much.

// Kai