Hello Shayan,

I always wonder of similar questions. You can model only that, which is introduced in ANSYS. If there is no good model for your case, you cannot model. You asked very complex question, the answer is NO WAY, because in general for such problem there no precise models. The question is about some simplification of the problem. Maybe you will accept even the one existing in the ANSYS. I do not know about it.

You can solve the problem by the VOF method. The researches found in the literature start from the initial bubbles with finite sizes. Then, the deformation and coalesce of the bubbles will be simulated by Fluent.

If you intend to simulate the nucleation of the bubbles, you might have to write a UDF.

Ivan V. Kazachkov : This your opinion is by way much too absolute. There are several ways to introduce into ANSYS CFX/Fluent your own models (User Fortran, UDF's, expressions, user introduced source terms to governing transport equations, surface and volumetric source terms, etc.

For the concrete question I might tentatively agree. In ANSYS CFD there is a build-in wall boiling model, but this has been developed for flow situations in nucleate subcooled boiling (e.g. in fuel rods of a nuclear reactor), not for pool boiling. Seems like a small difference, but in fact it is quite substantially different in many aspects. In particularly because the forced convection in nucleate subcooled boiling in vertical flow arrangements (pipes, channels) is then one of the dominating factors. And because build-in closure models with their correlations are most likely not valid for pool boiling conditions.

In this nucleate subcooled boiling model there are quite a dozen of submodels, one of them is the model for bubble departure diameter and bubble departure frequency. If it would be possible to formulate your modeling approach for the pool boiling situation as the basic nucleate subcooled boiling model with only a few modifications to its submodels, than it could nevertheless be possible to implement it in this already pre-existing model framework. But this would require a detailed and thoroughly analysis of all involved submodels and relationships (by yourself).

Regards,

Dr. Th. Frank.

The issue is, that simulating wall boiling in a larger scale technical fluid domain is impossible with resolving sub-millimeter large vapor bubbles using VOF. Therefore a wall boiling model is needed, where the wall boiling phenomenon is modeled rather then resolved.

Regards,

Dr. Th. Frank.

Thank you very much for all your insightful thoughts and responses on my question. I'm a beginner in this topic and need to read and do the suggested tutorials to get the most out of your answers. Based on your discussions about the limitations of each approach I realized that I need to define my problem better.

I have conducted the experimental subcooled boiling of a low-surface-tension liquid, on a micro-finned surface (about 500 um width and height), and obtained the heat transfer performance curve. I have also performed the imaging, so I know about the average departing bubble diameter, and their frequency. So, based on what Dr. Thomas Frank described, I can proceed and select the appropriate submodels. In this way, I should be able to estimate the heat transfer performance of the surface, and hopefully validate my model against the experimental data. Is it the Eulerian approach?

However, I am also interested in observing how the bubbles behave; the way they slide over the fin surface, coalesce, and depart. To do so, I can focus on VOF model. But again, I need to be specific as Sun Kyoung Kim mentioned, if I am interested in the life cycle of the bubble from the beginning (nucleation), it is more challenging and require writing UDFs, whereas starting with two bubbles of a finite size and studying their mutual behavior is more straightforward. Using this approach, I should be able to compare the simulated movement/coalescence/departure of for example two bubbles with experimental high speed images over time. Please correct me if I'm wrong.

Thanks

Shayan Davani