[Jasper Z]

Dear all,

I am wondering how should we specify U0 in injectionModels? I am simulating the Diaz bubble column using DPMFoam. I have varied U0 from 0 to 100 m/s with other parameters unchanged, but it seems that this doesn't affect the results much. My code is as follows:

injectionModels
    {
        model1
        {
            type            patchInjection;//manualInjection;
	    parcelsPerInjector	1;
	    patch	    INLET;
            massTotal       0;
            parcelBasisType fixed;
            nParticle       1;
            SOI             0;
	    parcelsPerSecond	293.354;//0.0024*0.04*0.2/(4/3*PI*0.002^3)
            U0              (0 100 0);
            sizeDistribution
            {
                type        fixedValue;
                fixedValueDistribution
                {
		    	value	0.005;
                }
            }
 	    flowRateProfile	constant 1.92e-5;
	    duration	200;
        }

Can anyone give a hint on what does U0 do and how to specify U0?

[Jasper Z]

Originally Posted by Jasper Z

Dear all,

I am wondering how should we specify U0 in injectionModels? I am simulating the Diaz bubble column using DPMFoam. I have varied U0 from 0 to 100 m/s with other parameters unchanged, but it seems that this doesn't affect the results much. My code is as follows:

injectionModels
    {
        model1
        {
            type            patchInjection;//manualInjection;
	    parcelsPerInjector	1;
	    patch	    INLET;
            massTotal       0;
            parcelBasisType fixed;
            nParticle       1;
            SOI             0;
	    parcelsPerSecond	293.354;//0.0024*0.04*0.2/(4/3*PI*0.002^3)
            U0              (0 100 0);
            sizeDistribution
            {
                type        fixedValue;
                fixedValueDistribution
                {
		    	value	0.005;
                }
            }
 	    flowRateProfile	constant 1.92e-5;
	    duration	200;
        }

Can anyone give a hint on what does U0 do and how to specify U0?

The water velocity results for U0=1m/s and 100m/s are shown in the attachment, and the results are pretty similar.

[joshwilliams]

Originally Posted by Jasper Z

Dear all,

I am wondering how should we specify U0 in injectionModels? I am simulating the Diaz bubble column using DPMFoam. I have varied U0 from 0 to 100 m/s with other parameters unchanged, but it seems that this doesn't affect the results much. My code is as follows:

injectionModels
    {
        model1
        {
            type            patchInjection;//manualInjection;
        parcelsPerInjector    1;
        patch        INLET;
            massTotal       0;
            parcelBasisType fixed;
            nParticle       1;
            SOI             0;
        parcelsPerSecond    293.354;//0.0024*0.04*0.2/(4/3*PI*0.002^3)
            U0              (0 100 0);
            sizeDistribution
            {
                type        fixedValue;
                fixedValueDistribution
                {
                value    0.005;
                }
            }
         flowRateProfile    constant 1.92e-5;
        duration    200;
         }

Can anyone give a hint on what does U0 do and how to specify U0?

The way you are defining it appears correct. I would guess that the issue is elsewhere, perhaps in the forces you define. For example, what drag model do you use? What is particle stokes number? Maybe the particles quickly relax to the local fluid velocity (stokes number<<1). In one case for my own work, I have set inlet velocity to 30 m/s in the same way you did. The particles generally quickly adapted to the fluid velocity within a few steps so it did not really influence simulation results either.

[Jasper Z]

Originally Posted by joshwilliams

The way you are defining it appears correct. I would guess that the issue is elsewhere, perhaps in the forces you define. For example, what drag model do you use? What is particle stokes number? Maybe the particles quickly relax to the local fluid velocity (stokes number<<1). In one case for my own work, I have set inlet velocity to 30 m/s in the same way you did. The particles generally quickly adapted to the fluid velocity within a few steps so it did not really influence simulation results either.

Dear Josh,

Yes you are right, the reason maybe from the drag force exerted on the bubbles. I am right now trying to use patchFlowRateInjection, but I am not sure about the definition of parcelConcentration. Is it the injected particle number in one second?

Much thanks!
Jasper

 injectionModels
    {
        model1
        {
           type             patchFlowRateInjection;
            phi alphaPhi.water;
            patch            INLET;
            duration         1;
            SOI              0;
        parcelBasisType  fixed;
            massTotal        0;
            nParticle        1;
            concentration 1;
            parcelConcentration 1800000;
            
            sizeDistribution
            {
                type        fixedValue;
                fixedValueDistribution
                {
                    value   0.0003;
                }
            }

        }

[joshwilliams]

Hi Jasper,


From looking at PatchFlowRateInjection.H, the variable is the parcel per unit volume flow rate. So it should be number of particles-seconds per metre cubed (?, they say it has units n/m3 so I am unsure)

        //- Parcels to introduce per unit volume flow rate m3 [n/m3]
        const scalar parcelConcentration_;

[Jasper Z]

Originally Posted by joshwilliams

Hi Jasper,


From looking at PatchFlowRateInjection.H, the variable is the parcel per unit volume flow rate. So it should be number of particles-seconds per metre cubed (?, they say it has units n/m3 so I am unsure)

        //- Parcels to introduce per unit volume flow rate m3 [n/m3]
        const scalar parcelConcentration_;

Hi Josh,

I also looked at the PatchFlowRateInjection.H. If my understanding is correct, does that mean if I have a flow rate of 0.001m3/s, and the parcel number is 10 per second under this flow rate; then the parcelConcentration should be 10*1000=10000?

Kind regards,
Jasper