CFD Online Discussion Forums - Radiation boundary conditions for flow through boundaries in openFoam

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)

- OpenFOAM (https://www.cfd-online.com/Forums/openfoam/)

- - Radiation boundary conditions for flow through boundaries in openFoam (https://www.cfd-online.com/Forums/openfoam/183686-radiation-boundary-conditions-flow-through-boundaries-openfoam.html)

Radiation boundary conditions for flow through boundaries in openFoam

Hello all,

I am relatively new to radiation modelling, but wanted to test it in a 1D multiphase flame simulation I am running. The simulation is a 1D open tube filled with fuel and ignited at one end.

I have looked through the radiation, scattering, and absorption emission models for the gas and lagrange phases and everything seems to make sense. I have decided to use the P1 radiation model and include cloud scattering and cloud emission/absorption.

However, I am not sure which boundary conditions to use and what should be the initial conditions for G? All of the examples I can find have solid walls, but in my case I would like an infinitely thick 1D flame (no loss in z/y boundaries, which is symmetry) and open boundaries on the ends (fixed parameters at inlet end and outflow at the other end).

My "goal" is for the boundaries to have no effect on the solution and to capture radiation from the burned gas behind the flame propagating upstream into the fresh fuel/particles.

Can anyone suggest boundary/initial conditions for this type of problem?

Thanks!

Chris

Hello DustExplosion,

Have you solved your problem?
I think I'm facing with a similar problem.

I have posted a question in the following thread:

https://www.cfd-online.com/Forums/op...tml#post696511

Thank you

Regards
GB

Hi Germilly,

I did not... I ended up going in a different direction and did not include radiation in my thesis. Good luck and hopefully someone can help!

Hello DustExplosion,

Thank you for your reply.

I'm trying something, but I am not sure if it is correct.
It is probably something very stupid.
But, I'm going to post it here. Maybe some help can come.

According to the following paper (page 4), using the P1 approximation, the boundary condition for the inlet and outlet is:

$-\frac{G}{2} = \frac{1}{3\beta}\vec{n} \cdot \nabla G$

and for the wall is:

$\frac{\varepsilon_w}{2(2-\varepsilon_w)}\big( 4\sigma T^4_w-G\big) = \frac{1}{3\beta}\vec{n} \cdot \nabla G$

For the wall, I can implement the Marshak boundary condition in openFoam, as is explained in:

https://www.cfd-online.com/Forums/op...tml#post696511

According to the equations above, the boundary condition for the inlet and outlet can be obtained by doing $\varepsilon_w = 1$ and

(this is probably the stupid thing) in the boundary condition for the wall.

For the inlet and outlet, what I am doing in openFoam, is using the MarshakRadiationFixedTemperature assigning Trad = 0:

Code:

    inlet

    {

        type            MarshakRadiationFixedTemperature;

        Trad            uniform 0;

        value           uniform 0;

    }

Code:

    outlet

    {

        type            MarshakRadiationFixedTemperature;

        Trad            uniform 0;

        value           uniform 0;

    }

The case is running, but I'm not sure in what I am doing.

Regards
GB

Some hints:
The MarshakRadiationFixedTemperature bc gives you the emitted radiation of a black body at a specified temperature and emissivity. This is often a good approximation for solids. For fluids, you'll need some kind of estimation about the radiation that is entering/exiting your boundary. If you have an optically thick medium with similar temperature outside of the boundary you might be able to use zeroGradient, as there should not be much radiation going through your boundary in that case. If your fluid approximately behaves like a black body you can use one of the Marshak conditions, either for fixed temperature or for the temperature of the temperature patch.

Hello Chriss85,

Thank you for the hints.

In my case, there will be radiation entering (less important) and exiting (more important) in the inlet and outlet boundaries.

I think I cannot use zeroGradient, because the temperature in the inlet and outlet are higher than the outside temperature.

The inlet and outlet are in direct contact with the environment. Therefore, if I define the Marshak conditions, which emissivity should I use? (there is no solid surface in the inlet and outlet, like is in the wall) there is only the participating media.

If it is possible, can you please give me your opinion about doing $\varepsilon_w = 1$ and

when using MarshakRadiationFixedTemperature?

Thank you again.

Regards,
Germilly

I'm not sure how to treat one-way radiating boundaries but let me know if you find something. As I said, using the Marshak condition is likely to induce errors for non-solids, but this probably depends on what you're trying to simulate. Maybe you can increase the size of the simulated region and test if there is a negative influence of the boundary condition?