air injection in OF
 

Hello FOAMers,

I'd like to inject a secondary stream of air into a system through small holes, and wanted to use the particle injection file for this purpose (is there a better way?).

Below is the file definition. Could someone point out if it is correct? I am not sure what are all the values in the 'constantPropoerties' section, and if they are used during the computations.

I am relatively inexperienced in using the software, and would appreciate any help.

Many thanks,
amit

----
FoamFile
{
version 2.0;
format ascii;
class dictionary;
location "constant";
object coalCloud1Properties; //injecting air instead

}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

active true;

InjectionModel ConeInjection;

DragModel SphereDrag;

DispersionModel none;

PatchInteractionModel StandardWallInteraction;

HeatTransferModel RanzMarshall;

CompositionModel SingleMixtureFraction;

PhaseChangeModel none;

DevolatilisationModel none;

SurfaceReactionModel none;

PostProcessingModel none;

radiation off;

coupled true;

cellValueSourceCorrection on;

parcelTypeId 1;

constantProperties
{
rhoMin rhoMin [ 1 -3 0 0 0 ] 1e-15;
TMin TMin [ 0 0 0 1 0 ] 290;
pMin pMin [ 1 -1 2 0 0 ] 1000;
minParticleMass minParticleMass [ 1 0 0 0 0 ] 1e-15;
rho0 rho0 [ 1 -3 0 0 0 ] 1300;
T0 T0 [ 0 0 0 1 0 ] 450;
cp0 cp0 [ 0 2 -2 -1 0 ] 1187;
epsilon0 epsilon0 [ 0 0 0 0 0 ] 1;
f0 f0 [ 0 0 0 0 0 ] 0.5;
Pr Pr [ 0 0 0 0 0 ] 0.7;
Tvap Tvap [ 0 0 0 1 0 ] 400;
Tbp Tvap [ 0 0 0 1 0 ] 400;
LDevol LDevol [ 0 0 0 0 0 ] 0;
hRetentionCoeff hRetentionCoeff [ 0 0 0 0 0 ] 0.9;
constantVolume true;
}

interpolationSchemes
{
rho cell;
U cellPointFace;
mu cell;
T cell;
Cp cell;
p cell;
}

integrationSchemes
{
U Euler;
T Analytical;
}

particleForces
{
gravity on;
virtualMass off;
pressureGradient off;
}

ConeInjectionCoeffs
{
massTotal massTotal [ 1 0 0 0 0 ] 1.6e-3;
parcelBasisType mass;
SOI 0;
duration 1.0;
position (0.0 0.019 0.019);
direction (1 0 0);
volumeFlowRate constant 1.3e-3;
parcelsPerSecond 10000;
Umag constant 50.0;
thetaInner constant 0.0;
thetaOuter constant 15.0;
parcelPDF
{
pdfType RosinRammler;
RosinRammlerPDF
{
minValue 5.0e-06;
maxValue 80.0e-06;
d 50.0e-06;
n 3.5;
}
}
}

StandardWallInteractionCoeffs
{
type rebound;
e 1;
mu 0;
}

RanzMarshallCoeffs
{
BirdCorrection true;
}

SingleMixtureFractionCoeffs
{
phases
(
gas
{
O2 0.23;
N2 0.77;
}
liquid
{
// H2O 1;
}
solid
{
// ash 0.1;
// C 0.9;
}
);
YGasTot0 1.0;
YLiquidTot0 0.0;
YSolidTot0 0.0;
}

LiquidEvaporationCoeffs
{
enthalpyTransfer enthalpyDifference;

activeLiquids
(
H2O
);
}

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

I dont get it?
you want to "inject" air as particles?
Why not just interfoam

Hello Rickard,

Thanks so much for the reply.

I am running compressible, reacting flow simulations in a combustor and would now like to inject a high speed secondary stream of air into the system through very small holes (diameter ~ 1mm) located at a few select positions (~ 20 such holes), and evaluate the effect on combustion behavior.

Would the above approach work (if so, is the data entered appropriate)? What are the other possible ways?

Many thanks,
ak

Okay
But I dont understand why you want to have particles for the air
You could just define a patch, using e.g. groovy BC, or redo the original mesh using arcs to create this 1mm hole
Usually you use particles for some kind of fuel in its liquid state or similar

@RIckard: This approach could be useful in say an engine where you might not want to incur the cost of fully resolving the injector geometry, but are injecting a gaseous spray. There have been a few papers on gaseous sphere injection:

http://www.google.ca/url?sa=t&rct=j&...kbbTdg&cad=rja

I've actually been looking a similar application but haven't quite gotten it to work in a generalized case.

@ak: The file looks decent, though I'm still not sure where the constantProperties entries are used in parcel injection. If you want to spheres to eventually "evaporate" into the gaseous carrier phase, you'll need to activate one of the devolatilization models and set the parameters accordingly (depending on how soon out of the jet you want "evaporation"). It takes a little fiddling, large parameters for constantRate seem to do an OK job. This won't be as rigorous as the approach in the paper, but it may be all you need.