[IronLyon]

Hi! I have some questions about the material radiation properties, as showed in the pic. below
radiation properties.PNG
these parameters should be changed when you use the radiation option, right?


For my model, it is the cooling process for fluid fluid (forced convection of helium gas to cool down liquid metal CuNi). As showed below
Figure002.png
as I can image, the metal itself radiates (high temperature 1773K), should I choose the radiation option in the metal domain, or should I choose the radiation option in the gas domain or both? (Because I found when choose the radiation option only in metal domain, it seems nothing happened, or we can say the result didn't change.

And when I choose the radiation in the gas domain it was significant different, namely the cooling result was 100K lower. And a friend also told me when using the radiation option, accually it should be choosed in the gas domain).

I'd like to know why? Or accually what's the correct way to choose radiation option (in which domain or both)? Thank you!

[Opaque]

Not clear how you are doing this, so here a few questions:

1 - Do you have separate domains for gas and liquid? If yes, you are using beta features, correct?
2 - Thermal radiation modeling is a whole field on its own. Do you understand the details of what is being modeled? If not, you will be going blind into a bunch of results with no background to interpret them.
3 - Properties are just that for which there may be some thermodynamic modeling or property modeling independent of the transport model, say kinetic theory or else. If they change with the transport modeling assumption, I guess they are not properties anymore.
4 - modeling details are set in the domain panel
5 - in general most metals are highly absorbing materials, and thermal radiation barely travels through it; therefore, it is not modeled. Gases are a different story since many of them are fairly transparent but not water vapour, carbon dioxide and a few others (search for the behavior of the gas of interest).
6 - if there are high temperatures, you will see the radiation modeling will definitely change your results. If the gas participates (absorbs/emits), its temperature distribution will change as well.

Hope the above helps,

[evcelica]

Including Radiation inside the solid domain means just that, Inside, not just the surface. You have to include radiation in the gas domain, because that is where the rays are traveling. They travel through the fluid, from solid surfaces to other solid surfaces. If the gas absorbs/transmits radiation as well, then that is where the absorption coefficient would be needed, as it would be a participating media. for helium, in a small space, I'm guessing it will be pretty transparent compared to the surfaces, and just surface to surface would be sufficient. This would be done by activating radiation in the gas, and specifying the radiation properties at the fluid solid interfaces. Diffuse fraction, grey model or bands, emissivity, as a start for the simplest assumptions.

[IronLyon]

Originally Posted by Opaque

Not clear how you are doing this, so here a few questions:

1 - Do you have separate domains for gas and liquid? If yes, you are using beta features, correct?
2 - Thermal radiation modeling is a whole field on its own. Do you understand the details of what is being modeled? If not, you will be going blind into a bunch of results with no background to interpret them.
3 - Properties are just that for which there may be some thermodynamic modeling or property modeling independent of the transport model, say kinetic theory or else. If they change with the transport modeling assumption, I guess they are not properties anymore.
4 - modeling details are set in the domain panel
5 - in general most metals are highly absorbing materials, and thermal radiation barely travels through it; therefore, it is not modeled. Gases are a different story since many of them are fairly transparent but not water vapour, carbon dioxide and a few others (search for the behavior of the gas of interest).
6 - if there are high temperatures, you will see the radiation modeling will definitely change your results. If the gas participates (absorbs/emits), its temperature distribution will change as well.

Hope the above helps,

Thank you for the reply, it did help me somehow. I just make some supplements of my question.
1- yes, I have two separated domains, one chamber for gas, one ball for metal. Both of them are fluid，and I have used the beta features (A error therefore appears when I choose the radiation option, it says no constant physical...cause I used the beta features).
2- what I want to model is the cooling process of the melt metal, the heat transfer happens mainly in two ways(fluid fluid):
one is the forced convection (no gravity so there is no free convection) or directly saying "gas conduction", the other one is the radiation of the metal (1500 K).
So that's my motivation to use the radiation option.
Actually I know the "roles" the "gas conduction" and the "radiation" play in the cooling process, I means like the power for conduction is 40W and 20W for radiation (according to the paper for our experiment before). But I don't know the alloy material properties of radiation which we used (hard to find even online), so before I just used the default value, like refractive index 1, absorption co. 1/m. So I think the result was no sense in such condition.
3- like mentioned in 2, I means when I use gas helium, so I should define its radiation property parameters, like refractive index and absorption co. instead of using the default value.
5- I understand what you mean is that only in the gas domain the radiation option should be choosed, not in the metal doamin cause of thermal radiation barely travels through it, right?
Thanks a lot.

[IronLyon]

Originally Posted by evcelica

Including Radiation inside the solid domain means just that, Inside, not just the surface. You have to include radiation in the gas domain, because that is where the rays are traveling. They travel through the fluid, from solid surfaces to other solid surfaces. If the gas absorbs/transmits radiation as well, then that is where the absorption coefficient would be needed, as it would be a participating media. for helium, in a small space, I'm guessing it will be pretty transparent compared to the surfaces, and just surface to surface would be sufficient. This would be done by activating radiation in the gas, and specifying the radiation properties at the fluid solid interfaces. Diffuse fraction, grey model or bands, emissivity, as a start for the simplest assumptions.

Thank you. Like the guide book said, for purely transparent cases only the Monte Carlo and Discrete Transfer model should be used. So I will choose one of them for the helium gas.
In the material properties, like for helium, the refractive index is 1.000034912, so the default value 1 is suitable, right? And the absorption coefficient should be found from the hand book, depending on the wave length and temperature, right? And I found the equation from ansys guide book below
1.png (formula 1)
Thermal radiation power can be calculated from
2.png
Spectral intensity can be calculated by P divide steradian [W⋅sr−1⋅m−1], I understand the spectral intensity is namely the radiation intensity, right? But it should be the I or I0 according to the formula 1? Then how to get another radiation intensity (I means the I and I0)?

[Opaque]

For clarification, since vocabulary is essential when describing complex physics to others: there is no such fluid-fluid heat transfer unless both materials are together in the same container w/o separation by any surface, i.e. multiphase flow within the same domain.

You can start by assuming He fully transparent. You still have to decide if you want to model spectral details of the gas, or a crude approximation as Gray. Either way, He will not absorb uniformly across the spectrum, and it will have transparent bands in the infra-red. Therefore, the calculation will give a bound for your model.