I am working on a rigid wall (for now) simulation of a pulsatile flow (as generated by a custom UDF) in a straight tube mixing with another flow injected at a different velocity (that is constant for now but will eventually be pulsatile also via a second UDF) into the same larger straight tube via a smaller concentric tube aligned axially with the larger straight tube. Both fluids have similar but different densities and different viscosities. The second fluid that is injected has a fixed concentration and I need to compute the relative concentrations of the composite mixture in each cell.
It has been suggested I use the Mixture model but It is unclear how I can specify a slip velocity between the two phases given it would need to be computed on the fly from the phase one inlet velocity from a pulsatile UDF and the second phase inlet velocity that is specified in the model as constant at the moment (and will move to an additional pulsatile UDF).
Thank you in advance for any advise.
CFD-ONLINE should provide you some information.
https://www.cfd-online.com/Forums/fluent/30048-difference-between-mixture-eulerian-multiphase.html
I would prefer the Eulerian multiphase model, it's more versatile. Mixture works well for co-flowing liquids without too much temporal variation (simple bubble columns), but considering that you are using pulsatile flows, the relative velocity between the phases will be constantly changing. I'm not convinced the mixture model will capture that sufficiently.
My experience with injecting gas into water flow is that analysing the volumetric ratios of both phases and the relative velocities of both "blobs" you would encounter completely different behaviors. In some cases dissipation of one stream into another would dominate, in other cases dynamics of blobs colliding as deforming entities may be more important, moreover, there will be some issues related to hydrodynamic instabilities of the "conjoined" flow. Given you would like to start with computer simulations, perhaps, you should begin with a regime where the dissipation dominates, and select some experimental data, where dynamics is more important, so that in tine you could enrich your model, to cope with collisions and deformation of blobs. Seems I would go for some type of a mix of models, to account for dissipation, rather than rely on the Lagrangian formula, which is certainly more up to paar with mutli-component flows. Good luck. Jefzy
I already have experimental data and I see good mixing of the fluids downstream by a distance 10 x the inlet diameter the smaller tube. It really doesn't seem as though the real data follows a "blob" mixing model. I am really trying to simulate the dilution of the injected dye media.
- Badges
- Science topic