Learning Forum

[Geckoo]

Hey guys,

I want to inject liquid water in a chamber. I used a diamater number distribution provided by the manufacturer to calculate the Rosin Rammler-mean diamater and spread. As my results do not match my expectations I want to double check the diameter distributions injected.

First, I plotted a Histogram using data from trajectory samle files of two surfaces. These histograms indicate a rectangular distribution with 18 of the 20 bins around the same number of particles and the smalles diameters slightly less.

Then I calculated the number distribution using the initial values of the injection obtained by //define/injections/lp/myInjection.
Here the number distribution is extreme to the small diameters, with 1.3e+6 entries for the smallest diameter, 1.3e+3  for the second smallest and then exponentially decreasing to 3. The volume distribution has a gaussian shape but should be skewed towards the larger diameters.

As I disabled breakup and all particles pass the surface of the histogram both plots shall indicate generally the same shape, but they do not. What is the right way to determine the diameter distribution of an injector?

My properties: Inert particle type, solid cone injection, rosin-rammler distribution, Min Diam 1.07e-7, Max Diam 6.75e-5, Mean Diam. 4.36e-5, spread 3.561, Number of streams 20, number of diameters 20.

 

Thank you very much and best regards
Ruben

 

[DrAmine]

You can just open injection panel then list the injection. Fluent writes the the output into console which you copy for further inspection. I will at first di that before looking into any other things.

[Geckoo]

Thank you Amine. This is the list obtained by //define/injections/lp/myInjection resulting in a Gaussian volume distribution.
Do you have an idea why the histogram of the trajectory sample files is so different?
All the best
Ruben

[DrAmine]

Are u sampling at inlet?

[Geckoo]

No at outlet and two other planes. But I think I have found my mistake. It is probably due to the format the reports are saved and thus cutting off values above 225,000.

Best regards

Ruben

[DrAmine]

Ok.

[Geckoo]

Okay actually it just led to another problem. The histogram obtained by Results -> Plots -> Histogram look very different to the histogram obtained by the sampling files Results -> Reports -> Discrete Phase -> Histogram.

 

 

 

The extreme peak at 1.07e-7m, the minimum diamater,  in the Sampling Histogram can also be seen in the injection list file. I'd be curious to know why there is such a peak but this is a different topic. In this case, breakup is enabled.

Does anyone know why they look so different? How does Fluent calculate the histograms?
Thank you very much and best regards
Ruben