[Msure]

Originally Posted by atul1018

Hello


-In the original reference experimental setup of Fessler & Eaton, it is clearly written that the setup is held vertically (See 2.1, first paragraph first sentence: https://citeseerx.ist.psu.edu/viewdo...=rep1&type=pdf)


-I also had downloaded the presentation from that seminar but in my slides the gravity is showed to set in +x direction (see attachement).


-Yes, you should compare the particle velocity profiles at measurement locations.


Best Regards
Atul Jaiswal

Hello,

I think you are right about the direction of gravity, then I tried all kinds of boundary conditions, meshes, and dispersion model. They didn't work. The results are the same as yours - particles not right but the fluid velocity is right compared with the experiment. Now I will try another experiment which is particle dynamical movement in a ventilated room to validate the DEM model of OpenFOAM, to see if the phenomenon we got in OpenFOAM in the backward-case is individual one. Thanks.

Regards,
Shuo

[atul1018]

Hello Shuo


So far, i couldn't find any problem in my case but unfortunately particle dispersion is not occurring as expected.


Anyway good luck for new case and keep me updated about your progress.


Best Regards
Atul

[Msure]

Originally Posted by atul1018

Hello Shuo


So far, i couldn't find any problem in my case but unfortunately particle dispersion is not occurring as expected.


Anyway good luck for new case and keep me updated about your progress.


Best Regards
Atul

Hello Atul,

When I tested another case, the results are still not very good. What I compared is the particle concentration in a ventilated room. I doubt I calculated the concentration wrong. I calculated it in this way:

template<class CloudType>
inline const Foam::tmp<Foam::volScalarField>
Foam::KinematicCloud<CloudType>::concentration() const
{
tmp<volScalarField> tconcentration
(
new volScalarField
(
IOobject
(
this->name() + ":concentration",
this->db().time().timeName(),
this->db(),
IOobject::NO_READ,
IOobject::NO_WRITE,
false
),
mesh_,
dimensionedScalar(dimless, Zero)
//extrapolatedCalculatedFvPatchScalarField::typeName
)
);


scalarField& concentration = tconcentration.ref().primitiveFieldRef();

for (const parcelType& p : *this)
{
const label celli = p.cell();

concentration[celli] += p.nParticle()*p.mass();
}

concentration /= mesh_.V();
//concentration.primitiveFieldRef() /= 1;//mesh_.V();
//concentration.correctBoundaryConditions();

return tconcentration;
}

and also a similar way to calculate the particle velocity in that backward facing case. So how did you calculate particle velocity in the y-direction in backward case. I want to know if my statistical approach for particles are wrong.
Thanks!

Regard,
Shuo

[atul1018]

Hello


As i don't know the details about your ventilation case, I cant really say about your code regarding concentration calculation.



Regarding BFS case, I extracted particle ID, velocity and position at certain time (say 0.07 sec after insertion of particles). The extracted data are evaluated at slice around each measurement location (x/H=2,5,7,9,12) keeping the measurement location at center of slice. Now you have particle distribution along y direction in each slice. Now you can plot particle velocity profile for the particle in that slice.


I extracted particle data exactly the same way done by Greifzu et.al. You should read the paper of Greifzu et.al.: https://www.tandfonline.com/doi/pdf/...0.2015.1104266



Best Regards
Atul

[YANGLU1021]

Hi, atul1018. Did you solve the problem? I have met the same problem recently. I also found that the particles in the Y-axis direction do not settle down. If you have found the solution, could you please share your method? Thanks a lot.

[YANGLU1021]

Originally Posted by atul1018

Hello


As i don't know the details about your ventilation case, I cant really say about your code regarding concentration calculation.



Regarding BFS case, I extracted particle ID, velocity and position at certain time (say 0.07 sec after insertion of particles). The extracted data are evaluated at slice around each measurement location (x/H=2,5,7,9,12) keeping the measurement location at center of slice. Now you have particle distribution along y direction in each slice. Now you can plot particle velocity profile for the particle in that slice.


I extracted particle data exactly the same way done by Greifzu et.al. You should read the paper of Greifzu et.al.: https://www.tandfonline.com/doi/pdf/...0.2015.1104266



Best Regards
Atul

Hi, atul1018. Did you solve the problem? I have met the same problem recently. I also found that the particles in the Y-axis direction do not settle down. If you have found the solution, could you please share your method? Thanks a lot.

[atul1018]

Hey YANGLU1021

Did you solve the problem? I have met the same problem recently. I also found that the particles in the Y-axis direction do not settle down. If you have found the solution, could you please share your method? Thanks a lot.

-I tried with all I could but I could not able to get better results for particles. Let me know if you make any progress.

[YANGLU1021]

Originally Posted by atul1018

Hey YANGLU1021

-I tried with all I could but I could not able to get better results for particles. Let me know if you make any progress.

Thanks for your reply. I will try some ways to see if this situation can be improved. I will tell you if I get any progress. Thanks and have a good day!