[alfiogn]

Hi all,


I'm trying to perform langrangian particle tracking on a simple case.


Basically my flow domain is made by a hollow cylinder where I impose a flow rate at the inlet and a rotation at the internal wall. So my flow rotates and advances in axial direction.
Moreover the flow is steady state.



I want to perform a completely passive tracking, so where particles follow streamlines.


I expect particles to remain at their initial radial coordinate and to move towards the outlet.
This does not happen at all. I upload two images: the first is the initial state of the particles and the second is the state at the first time step.


Moreover, I initialize the particle velocity with their velocity position in the flow.


Thanks in advance!
Giorgio








PS: this is my kinematicCloudProperties dictionary:

/*--------------------------------*- C++ -*----------------------------------*\
  =========                 |
  \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox
   \\    /   O peration     | Website:  https://openfoam.org
    \\  /    A nd           | Version:  7
     \\/     M anipulation  |
\*---------------------------------------------------------------------------*/
FoamFile
{
    version     2.0;
    format      ascii;
    class       dictionary;
    location    "constant";
    object      particleProperties;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

solution
{
    active          true;
    coupled         false;
    transient       yes;
    cellValueSourceCorrection on;
    calcFrequency   1;
    maxCo           1;
    maxTrackTime    10000;

    interpolationSchemes
    {
        rho             cell;
        U               cellPoint;
        mu              cell;
        DUcDt           cellPoint;
        "(.*)"          cell;
    }

    integrationSchemes
    {
        U       Euler;
    }

    sourceTerms
    {
        schemes
        {
            U semiImplicit 1;
        }
    }
}

subModels
{

    radiation                   off;
    dispersionModel             none;
    collisionModel              none;
    heatTransferModel           none;
    surfaceFilmModel            none;
    stochasticCollisionModel    none;
    patchInteractionModel       localInteraction;

    particleForces
    {
        pressureGradient {}
    }

    injBase
    {
        type                initManualInjection;
        parcelBasisType     fixed;
        massTotal           0;
        nParticle           1;
        U0                  (0 0 0);
        SOI                 0;

        sizeDistribution
        {
            type        fixedValue;
            fixedValueDistribution
            {
                value   1e-13;
            }
        }
    }

    injectionModels
    {
        Cluster0
        {
            $injBase
            positionsFile       "Cluster0Positions";
        }
    }

    localInteractionCoeffs
    {
        patches
        (
            INLET
            {
                type escape;
            }

            OUTLET
            {
                type escape;
            }

            INTERNAL
            {
                type stick;
                e    0.97;
                mu   0.03;
            }

            EXTERNAL
            {
                type stick;
                e    0.97;
                mu   0.03;
            }
        );
    }

}


constantProperties
{
    rho0    1;
}

cloudFunctions
{}


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

[alfiogn]

update:


I manage to do it by setting the sphereDrag force instead of the pressure gradient.
In this way the force tends to generate a Cauchy problem that results in:

as the size of the parcel goes to zero (c.f. sizeDistribution dictionary).






Giorgio