[melj]

Hello

I am working on multiphase bubble columns. It uses a non-Newtonian liquid and air. I have three questions.

1) Most of the research papers plot "bubble pdf (1/m) vs bubble diameter". How to plot the bubble pdf having the unit (1/m) on the y-axis? I have seen that there are three options in fluent to plot number density function, but none give the result in the above mentioned unit.

2) While using non-Newtonian liquids (nnl), I tried plotting the contour for viscosity of nnl, but it always shows a constant value instead of a range of values (as viscosity keeps changing for a nnl). The weird bit is that I get the proper range when I plot the viscosity for the mixture instead of nnl. Could someone please suggest possible reasons?

3) I defined the viscosity for the nnl using the non-Newtonian power law dialogue box. Which variable in fluent would reflect this range while plotting the contour of viscosity?

[vinerm]

For most of the non-Newtonian fluids, shear rate is the most important parameter. Same is true for power law. In addition to that, it could also be a function of temperature. An important aspect of power law is that it describes a family and not one type of fluid, including the Newtonian fluid. The exponential parameter, given as n in Fluent, decides which group the fluid belongs to. This being unity implies Newtonian. Larger and smaller values imply thickening and thinning fluids. Check these values in Material Panel.

Secondly, while using non-Newtonian fluid, it is expected that this fluid is the primary phase. I suppose that's already the case in your simulation since you are modeling bubble column.

You appear to be using PBM. PBM solves for number density function and not probability density function.

[melj]

Thank you vinerm.

1) I am using the shear rate dependent option. The value of exponential parameter, n, is sufficiently less than 1.

The non-Newtonian liquid is the primary phase. Yet, it seems to give the previously mentioned result. I don't understand.

Is it possible that if my range of viscosity defined using the minimum and maximum viscosity in the non-Newtonian power law dialogue box is small, it would result in a constant? Min=0.045 and Max=0.073. However, when I select the molecular viscosity for the mixture it gives this selected range.

In post processing, the dynamic viscosity (which I assume is for the mixture because the same for the non-Newtonian phase is mentioned separately) also gives the required range. Again it's for the mixture.

2) Which viscosity gives the required range. I mean effective, molecular, dynamic or turbulent. Sorry I am confused now.

3) I am using PBM. Could you please let me know the method that gives pdf?

[vinerm]

If you are using CFD-Post for post-processing, then you have to define a few things as expressions. CFD-Post starts if you use WB and click on Post-Processing tab. I would not recommend using it for multiphase cases. Check within Fluent. Small range is not an issue.

The viscosity you are referring to is the molecular viscosity. This is also known as dynamic viscosity. Turbulent viscosity on the other hand is not a fluid property, rather a flow property and depends on the flow. It varies from space to space. Effective viscosity represents sum of dynamic and turbulent viscosity. So, you should only be looking at dynamic viscosity.

PDF essentially implies the probability of some event within a certain interval, about a certain value. If I rephrase this in terms of particle size, then PDF of particle size would be probability of finding a particle within +/-, say, 1% or 2%, of 50 micron or 100 micron. Read example section at https://en.wikipedia.org/wiki/Probab...nsity_function to understand more about it. So, you have to extract number density function values from Fluent and convert those to PDF. I have never done that but should be easy.

[melj]

I couldn't plot the required contour of molecular (or dynamic) viscosity for the non-Newtonian liquid phase using Fluent either. I am not able to understand where the mistake is. I can get the answer using mixture (though it's not what I want) but not the liquid phase.

Anyways, thanks a lot vinerm.

[vinerm]

You might be using mixture model instead of Euler. In case of mixture model, a single set of momentum equations is solved. Therefore, single viscosity, density, etc. are defined using volume fractions of both phases. So, you have to use mixture to display the viscosity. Or you have to use Euler model.

[melj]

Hello vinerm

I am using Euler model.

thank you