[Azanulbizar]

Hi,

I am wondering if anyone can explain to me how does the size distribution models work and what kind of input I have to put it?

In spray modelling, instead of using Rosin-Rammler, how does one use distribution models such as fixed, general, etc etc which is located in /src/lagrangian/distributionModel

If not mistaken, I got a feeling that they are based on some Cumulative Distribution Function or some probability function. If anyone can refer to me what to look up for, I would also deeply appreciate it.

Thank you very much.

[Chrisi1984]

Hi,

with "general" you can give a volume probabilty for the droplet size in form of a table:

sizeDistribution
{
type general;
generalDistribution
{
distribution
(
( 0.000075 2.752E-03 )
( 0.000125 1.000E-02 )
( 0.000175 1.331E-02 )
( 0.000225 1.642E-02 )
( 0.000275 2.058E-02 )
( 0.000325 2.535E-02 )
( 0.000375 3.005E-02 )
( 0.000425 3.416E-02 )
( 0.000475 3.737E-02 )
( 0.000525 3.967E-02 )
( 0.000575 4.104E-02 )
( 0.000625 4.166E-02 )
( 0.000675 4.157E-02 )
( 0.000725 4.100E-02 )
( 0.000775 3.990E-02 )
( 0.000825 3.866E-02 )
( 0.000875 3.699E-02 )
( 0.000925 3.531E-02 )
( 0.000975 3.358E-02 )
( 0.001025 3.160E-02 )
( 0.001075 2.979E-02 )
( 0.001125 2.800E-02 )
( 0.001175 2.616E-02 )
( 0.001225 2.441E-02 )
( 0.001275 2.276E-02 )
( 0.001325 2.119E-02 )
( 0.001375 1.968E-02 )
( 0.001425 1.825E-02 )
( 0.001475 1.691E-02 )
( 0.001525 1.565E-02 )
( 0.001575 1.449E-02 )
( 0.001625 1.342E-02 )
( 0.001675 1.239E-02 )
( 0.001725 1.142E-02 )
( 0.001775 1.052E-02 )
( 0.001825 9.688E-03 )
( 0.001875 9.023E-03 )
( 0.001925 8.307E-03 )
( 0.001975 7.638E-03 )
( 0.002025 7.016E-03 )
( 0.002075 6.440E-03 )
( 0.002125 5.912E-03 )
( 0.002175 5.551E-03 )
( 0.002225 5.116E-03 )
( 0.002275 4.700E-03 )
( 0.002325 4.312E-03 )
( 0.002375 3.954E-03 )
( 0.002425 3.625E-03 )
( 0.002475 3.324E-03 )
( 0.002525 3.153E-03 )
( 0.002575 2.937E-03 )
( 0.002625 2.706E-03 )
( 0.002675 2.490E-03 )
( 0.002725 2.291E-03 )
( 0.002775 2.108E-03 )
( 0.002825 1.940E-03 )
( 0.002875 1.789E-03 )

);

}
}

With "uniform" you define a constant droplet size for all your droplets.

Kind regards

Chrisi

[Azanulbizar]

HI. It took me awhile. but I figured that out already. Thank you very much anyway. By the way, do you have any idea of how to plot the velocity of the parcels/particles using paraview?

When I select a plane, and use "plot over a line", I cant obtain the velocities of the parcels. Is there anyway of easily measuring the velocity of the parcels?

[Alaska1964]

Originally Posted by Chrisi1984

Hi,

with "general" you can give a volume probabilty for the droplet size in form of a table:



With "uniform" you define a constant droplet size for all your droplets.

Kind regards

Chrisi

Dear Chrisi,
Is the second column indicating the cumulative distribution (Q3(xi)) ? And the first column diameter of the particle in meters?

Thanks for your help.
kind regards,
Alaska

[chpjz0391]

Originally Posted by Alaska1964

Dear Chrisi,
Is the second column indicating the cumulative distribution (Q3(xi)) ? And the first column diameter of the particle in meters?

Thanks for your help.
kind regards,
Alaska

Hey Alaska,
I have the same question? what does the second column mean?

[Chrisi1984]

Hey together,

the second colume should be ΔQ3(xi).

Kind regards
Chrisi

[Alaska1964]

Thanks a lot Chrisi, I also think so since the summation of all data on second column is 1.

I have another question regarding the particle size distribution. Is there any model in OpenFOAM for log normal PSD?

there is a models for "normal" and "RosinRammler", but not for "lognormal". I have data with X50 X90 and geometrical size distribution (GSD). Do you have any idea how I can implement them to my spray model? I also have the data of ΔQ3(xi) for different intervals of diameter.

Cheers,

[vidyadhar]

Hello Chrisi,

Can you please let me know what are the units of value 4e-03 in the following sizeDistribution model taken from bubbleCloudProperties file.

sizeDistribution
{
type fixedValue;
fixedValueDistribution
{
value 4e-03;
}
}

I am thinking that it is in meters. Am I right.

Please let me know.

Thanks & Regards,
Vidyadhar

[minzhang]

Originally Posted by Chrisi1984

Hey together,

the second colume should be ΔQ3(xi).

Kind regards
Chrisi

Hello Chrisi,

I have two questions about the sizeDistribution.

1. The second column of generalDistribution is the number/volume/mass fraction of that particular particle/droplet size, yes?
2. You said that "With "uniform" you define a constant droplet size for all your droplets". My question is we could use "uniform" to define a uniform distribution of droplet/particle size, yes?


Thanks and best regards,
Min

[Chrisi1984]

Originally Posted by minzhang

Hello Chrisi,

I have two questions about the sizeDistribution.

1. The second column of generalDistribution is the number/volume/mass fraction of that particular particle/droplet size, yes?
2. You said that "With "uniform" you define a constant droplet size for all your droplets". My question is we could use "uniform" to define a uniform distribution of droplet/particle size, yes?


Thanks and best regards,
Min

Hello together,

Regarding the questions:
1. The first column is the diameter in m. The second column is the volume probability for the corresponding diameter.
2. Yes

Kind regards
Chrisi

[minzhang]

Originally Posted by Chrisi1984

Hello together,

Regarding the questions:
1. The first column is the diameter in m. The second column is the volume probability for the corresponding diameter.
2. Yes

Kind regards
Chrisi

Hello Chrisi,

Thank you so much for your reply!

1. What do you mean by "volume probability", is it the same as "volume fraction"?
2. With "uniform size distribution", I mean it is a distribution between minValue and maxValue, instead of "a single particle size". So I am wondering what is your meaning, the former one or the latter?
3. I want to have three particle sizes and all three corresponding volume/number fractions are the same (equal to 1/3). So I am wondering which type of size distribution (general or uniform) could be used?

Thanks again for your kind help!

Best regards,
Min

[minzhang]

Originally Posted by minzhang

Hello Chrisi,

Thank you so much for your reply!

1. What do you mean by "volume probability", is it the same as "volume fraction"?
2. With "uniform size distribution", I mean it is a distribution between minValue and maxValue, instead of "a single particle size". So I am wondering what is your meaning, the former one or the latter?
3. I want to have three particle sizes and all three corresponding volume/number fractions are the same (equal to 1/3). So I am wondering which type of size distribution (general or uniform) could be used?

Thanks again for your kind help!

Best regards,
Min

Corrections to Point 3:
3. I want to have three particle sizes and all three corresponding volume/mass fractions are the same (equal to 1/3). So I am wondering which type of size distribution (general or uniform) could be used?

My own understanding:
So I can use the "uniform" distributionModel. Because I need a uniform mass fraction and this is exactly what OpenFOAM "uniform" distributionModel gives.
Am I correct?

[minzhang]

Originally Posted by Azanulbizar

HI. It took me awhile. but I figured that out already. Thank you very much anyway. By the way, do you have any idea of how to plot the velocity of the parcels/particles using paraview?

When I select a plane, and use "plot over a line", I cant obtain the velocities of the parcels. Is there anyway of easily measuring the velocity of the parcels?

Hello Azanulbizar,

I am wondering whether you could be so kind to have a look at my question.

size distribution models in OpenFOAM 4.x

Your help would be very appreciated!

[Akimlav]

Originally Posted by minzhang

Corrections to Point 3:
3. I want to have three particle sizes and all three corresponding volume/mass fractions are the same (equal to 1/3). So I am wondering which type of size distribution (general or uniform) could be used?

My own understanding:
So I can use the "uniform" distributionModel. Because I need a uniform mass fraction and this is exactly what OpenFOAM "uniform" distributionModel gives.
Am I correct?

I have the same question

[Akimlav]

I solved problem of having 2 sizes of particles by adding one more injection model with fixedValueDistribution. And it is possible to add as much as you want.


injectionModels
{
model1
{
type patchFlowRateInjection;
massTotal 1;
parcelBasisType mass;
massFlowRate 10;
patchName inlet;
duration 0.5;
concentration 2;
parcelConcentration 1e8;
SOI 0;
sizeDistribution
{
type fixedValue;
fixedValueDistribution
{
value 1e-6;
}
}
}
model2
{
type patchFlowRateInjection;
massTotal 1;
parcelBasisType mass;
massFlowRate 10;
patchName inlet;
duration 0.5;
concentration 2;
parcelConcentration 1e8;
SOI 0;
sizeDistribution
{
type fixedValue;
fixedValueDistribution
{
value 2e-6;
}
}
}
}

[atul1018]

Hey Akimlav


I am struggling to understand the patchFlowRateInjection model, as there are not much written about it neither any tutorial available for that.



I saw that you used that model for your case, can you please explain the meaning of parameters required for the patchFlowRateInjectionModel, especially the keywords 'concentration' and 'parcelConcentration'?


Best Regards
Atul

[Akimlav]

Originally Posted by atul1018

Hey Akimlav


I am struggling to understand the patchFlowRateInjection model, as there are not much written about it neither any tutorial available for that.



I saw that you used that model for your case, can you please explain the meaning of parameters required for the patchFlowRateInjectionModel, especially the keywords 'concentration' and 'parcelConcentration'?


Best Regards
Atul

Hi!


My knowledge is limited.



patchFlowRateInjection


Don`t know more than what here is said.


I played with the numbers in order to get amount of particles that I need.


Cheers!

[Mohmed Zayed]

The following definition for the general distribution model is taken from openfoam8/src/lagrangian/distributionModels/general/general.H

Class
    Foam::distributionModels::general

Description
    A general distribution model where the distribution is specified as
    (point, value) pairs. By default the values are assumed to represent
    a probability density function, but the model also supports specifying a
    cumulative distribution function. In both cases it is assumed that the
    function is linear between the specified points.

    In both modes of operation the values are automatically normalized.

SourceFiles
    general.C

[uosilos]

Does anybody know the formulation and mathematical background of the general distribution?? Any paper or somewhere to find??


btw check my new post!!
reactingFoam - Particle Size Distribution


thanks