[melj]

Hello

I would like to find the drag force acting on bubbles in a bubble column.

I have tried using drag monitor in Fluent and it calculates drag force for the walls. Is the value obtained using this approach the drag force experienced by bubbles in a bubble column as well?

thank you in advance

[vinerm]

Hi

Drag force on bubble can only be predicted if the interface is being resolved, such as with VOF or Dynamic Mesh. With other multiphase models, CFD tools expect Drag coefficient to be provided by the user. If your question is about drag force based on the drag coefficient being used, then you may simply use a Custom Field-Function.

[melj]

Thank you vinerm. So i am using Eulerian multiphase model. The drag force that I calculate using the drag monitor on the wall would be calculated by Fluent using the drag coefficient being used by Fluent rite? I am not familiar with UDF in Fluent.

[vinerm]

Drag on the wall has got nothing to do with drag coefficient being used for the interphase interaction. Drag on the wall is being determined from the first principle, i.e., by using the solved velocity field.

On the other hand, drag interaction between the phases is different. Since your interest lies in predicting the drag on the bubble in the bubble column, you have to either use VOF or you have to create bubbles of various shapes as solid geometries and then monitor the drag on the bubble boundaries. If the shapes of the bubbles are known, then you do not need CFD. You can use correlations, such as Grace's or Tomiyama's. The interest, most of the time, in doing a bubble column simulation lies in predicting the gas hold-up, dispersion coefficient, etc. but not in predicting drag. That part is usually done either using DNS or experiments.

[melj]

Thank you so much. Your comments were very helpful.

[melj]

So if I knew the bubble shape, I could use the existing correlations to create a custom field function to obtain the value from CFD. Is that rite? Or did you mean using the correlations in DNS or experiments? Apologies if I have got it completely wrong.

[vinerm]

If the shape is known, you do not require CFD. Drag coefficients are usually given in terms of known parameters.

[melj]

Got it. Thank you.

[farzadmech]

Dear Friends
Has anyone used Lagrangian Particle Tracking for simulation of Bubble Plume. What Drag model is acceptable for bubbles?


Thanks,
Farzad