[oliveira1820]

Hello everybody!

I´m working with water temperatures between 250 and 1000 K and energy heat total sources with 1000 W m^-2 K in water which transfer heat to niobium material.

Do I need to include Thermal Radiation in my Heat Transfer model or is it despicable?

Best regards,

Pedro Oliveira

[ghorrocks]

I think you will find "insignificant" is a better word than "despicable", but I know what you mean

It is best if you work out thing like this yourself. You can get approximate estimates of radiative heat transfer from simple relations like the Stefan Boltzmann equation (https://www.engineeringtoolbox.com/r...fer-d_431.html), estimates of convection from a estimate of the heat transfer coefficient from the table at the bottom of (https://www.engineeringtoolbox.com/o...ent-d_434.html) and conduction, if it is relevant.

This will allow you to estimate what the radiative, convective and conductive energy flows and make a good decision as to whether radiation is important.

But as a rule of thumb, as radiative heat transfer goes up by the 4th power of temperature at 1000K the radiative heat transfer is going to be very important.

[oliveira1820]

Thank you so much ghorrocks, you are always very helpful.

[oliveira1820]

Actually I have 2 more questions:

1 - Do I need to activate Thermal Radiation just in the fluid that reaches 1000K and the surrounding solids or do I have to consider it to every solid and water cooling in my simulation or is this just unecessary? ( Note that the cooling water only reaches temperatures between 290 and 330 K )

2 - The Radiation Porperties (Refractive Index, Absortion and Scattering Coefficient) vary with wavelength. Can I use any wavelength I want let´s say 600 nm as reference to every material or do I have to use a especific wavelenght as reference in my simulation?

[ghorrocks]

1 - This depends on what you are trying to do, what you are modelling and what results you intend to get. You would have to describe your case in more detail and show some images.

2 - First of all, is the varying radiation properties enough to be significant? If it is significant then your only choice will be to run a grey body Monte Carlo model. These models are expensive and difficult to run accurately.

[oliveira1820]

Basically I got a water closed recipient (niobium) with extreme heat source which heats till 1000K, then I got water cooling the other surface of the recipient with 0.2l/s velocity. The cooling water passes between a aluminium diffuser and a support.

All my project has just about 30 cm.

The water cooling is in blue, the water that heats is in grey.


The purpose is studying the heat transfer between the different solid and fluid domains and how the water heats, as well the effect of the cooling water on the system.

[ghorrocks]

I presume you are interested in the temperature distribution on the solid bodies and coolant. Further I assume the water is reasonably transparent and the solid bodies are opaque. In that case you will need to define a radiation model in the fluid domain and radiation boundaries on the interfaces to the solid domains.

[oliveira1820]

Just to confirm I understood well.

I need to define the thermal radiation model on the fluid domains and only in the solid boundarys with the fluid. There is no need to define the thermal radiation in the solids itself.

Or to define the thermal radiation in the boundaries I need the thermal radiation model in the solids?

Also, if the second option is the right one wich option should I choose on boundaries:

- Conservative Interface Flux
- Opaque
- Side Dependent

[ghorrocks]

As long as the solid bodies are opaque then you do not require a radiation model in the solid bodies. This would mean you only require the radiation model in the fluid domain, and the radiation effects act on the interfaces of the fluid and solid domains.