[hcl734]

I try to simulated a 2D channel with a rib and periodic boundary condtions. But I need to keep

I am using the cyclic boundary condition to couple inflow and outflow patch.

But since pressure drops and temperature rises from rib to rib I have to modify the inlet conditions to fit.

I would like to avoid modifying the solver, because I am not really familiar with the source code.

I found an option to keep my velocity (=mass flow, incompressible, same area, dm/dt = u*A*rho) constant using fvOptions.

pressureGradientExplicitSourceCoeffs
{
fieldNames (U);
Ubar ( 1 0 0 );
gradPini gradPini [0 2 -2 0 0] 0; // initial pressure gradient
flowDir (1 0 0); // flow direction
}

But I also need to add a constant term (dq/dt)/((dm/dt)*cp*L) to compensate for the energy added by the heated walls.
How can I do this in OpenFOAM?


My idea so far is to set <<scalarExplicitSource>> for energy equation to subtract the term mentioned above, but I am not sure how to do it.

Something like:

energySource
{ 
    type            scalarExplicitSource; 
    active          true; 
    timeStart       0.2; // What are those entries for? 
    duration        2.0; 
    selectionMode   points; 
    points 
    ( 
//[all points???] 
    );                                                                                                                                                        

    scalarExplicitSourceCoeffs 
    { 
        volumeMode      absolute; 
        injectionRate 
        { 
            T         -(dq/dt)/((dm/dt)*cp*L); 
        } 
    } 
}

The idea of subtracting the source term I got from

Fully Developed Flow and Heat Transfer in Ducts Having Streamwise-Periodic Variations of Cross-Sectional Area. S. V. Patankar, C. H. Liu and E. M. Sparrow
http://heattransfer.asmedigitalcolle...icleid=1436662


Every advice is appreciated!

Best

[cbcoutinho]

Hello hcl734,

I am also working on streamwise-periodic flow with mass/heat transfer. I'm interested in implementing this in openFOAM, and it looks like you are working on this as well. Have you made any progress since you asked this question?

Cheers,
c

[hcl734]

Hello Chris,

pressureGradientExplicitSourceCoeffs works fine to set a constant mass flux.

And I implemented the heat source the way patankar suggests in the mentioned paper following this thread

http://www.cfd-online.com/Forums/ope...-transfer.html

It works for me on one grid (mass flow and temperature become constant) but fails at another which is strange and I don't understand yet.
I can share my calculations with you privately but can't publish them here, since they are part of my thesis, pn me.

Best
Hendrik