[Esmaeel]

Hi everyone, I am trying to study the particle size effects for a binary mixture in a fluidized beds. So the system is particle-fluid but we have three phases: particle1,particle2, and air. The problem that I am facing with multiphaseEulerFoam solver is that the air volume fraction goes below 0.38 and it becomes as low as 0.24. So, each particle volume fraction are around 0.37 for same cells and that leads to the high volume fraction for particle phase. Here is the phaseProperties dictionary. Can anyone please help me understand what is wrong with my phaseProperties which results in high solid concentration? Thanks.



/*--------------------------------*- C++ -*----------------------------------*\
========= |
\\ / F ield | OpenFOAM: The Open Source CFD Toolbox
\\ / O peration | Website: https://openfoam.org
\\ / A nd | Version: 10
\\/ M anipulation |
\*---------------------------------------------------------------------------*/
FoamFile
{
format ascii;
class dictionary;
location "constant";
object phaseProperties;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

type basicMultiphaseSystem;

phases ( particles particles2 air );

//referencePhase particles;

particles
{
type purePhaseModel;

diameterModel constant;
constantCoeffs
{
d 2e-4;
}

alphaMax 0.62;
residualAlpha 1e-6;
}


particles2
{
type purePhaseModel;

diameterModel constant;
constantCoeffs
{
d 2e-4;
}

alphaMax 0.62;
residualAlpha 1e-6;
}


air
{
type purePhaseModel;

diameterModel none;

residualAlpha 0.;
alphaMin 0.38;
}

blending
{
default
{
type continuous;
phase air;
}
}

surfaceTension
{}

interfaceCompression
{}

drag
{
particles_dispersedIn_air
{
type Ergun;
residualRe 1e-3;
}


particles2_dispersedIn_air
{
type Ergun;
residualRe 1e-3;
}

}

virtualMass
{}

heatTransfer
{
particles_dispersedIn_air
{
type RanzMarshall;
residualAlpha 1e-4;
}

particles2_dispersedIn_air
{
type RanzMarshall;
residualAlpha 1e-4;
}

}

phaseTransfer
{}

lift
{}

wallLubrication
{}

turbulentDispersion
{}

// ************************************************** *********************** //

[dscian]

kineticTheoryModel is important for particle dynamics in multiphaseEulerFoam so you need to check that. Furthermore, I don't think current solver can handle two different particle phases properly.

[Esmaeel]

Originally Posted by dscian

kineticTheoryModel is important for particle dynamics in multiphaseEulerFoam so you need to check that. Furthermore, I don't think current solver can handle two different particle phases properly.

Thanks for the reply. The current case is from fluidized bed tutorial and it uses RAS kinetickTheory. I am beginning to accept that two particle systems cannot be modeled with this solver but I might be wrong!

[demichie]

Originally Posted by Esmaeel

Thanks for the reply. The current case is from fluidized bed tutorial and it uses RAS kinetickTheory. I am beginning to accept that two particle systems cannot be modeled with this solver but I might be wrong!

The kineticTheory implementation requires major changes to be able to model multiple solid dispersed phases. In the present state it should be used only with one solid phase.

[Esmaeel]

Originally Posted by demichie

The kineticTheory implementation requires major changes to be able to model multiple solid dispersed phases. In the present state it should be used only with one solid phase.

Thanks. I was wondering if other software like ANSYS-CFX or MFIX can do such modeling of multi-particle systems with euler-euler approach?

[demichie]

Yes, they can both model multi-particle systems. If you want to do that with OpenFOAM, you have to modify the kinetic class in multiphaseEuler (the main change is the in the radial distribution functions), add a model for a multi-particle maximum packing, and add a solid-solid drag.

[Esmaeel]

Originally Posted by demichie

Yes, they can both model multi-particle systems. If you want to do that with OpenFOAM, you have to modify the kinetic class in multiphaseEuler (the main change is the in the radial distribution functions), add a model for a multi-particle maximum packing, and add a solid-solid drag.

I prefer to stick with OpenfFOAM. I would really appreciate if you could explain more on where/how the changes should be made? I have some experience with modifying solvers/libraries but this multiphaseEulerFoam is complicated/confusing!cheers