1 Get all your modeling sorted then add radiation the solve time with radiation will be longer. Yes radiation can be significant some times depending on the a and e values. Make calculations by hand to estimate where the heat is going - assume the camera is a simple cube. You will have to calculate radiation and convection at the same time which you can not do directly. An iterative solution is needed. Your understanding of the problem will be far greater with preliminary calculations (don't just jump into CFD).2. Radiation to the environment should be used (and internal surface to surface) - but what is the temperature of the environment? If this is a simulation of the camera in the real world that the surrounding environment is set ~10C less then the ambient. In this case 40C (use this value until you find a better one the fits your application). If the simulation is used to compare the camera in a heat chamber then the walls of the heat chamber will be equal to the ambient air temperature.
3. e = 0.8 until you find a better value for each surface.
4. Natural convection. Make sure your computational domain is large enough to capture the natural convection. So the domain may need to be 1500 mm above and below the camera. Experiment withe this size and increase / reduce over time to see how it affects results. In CFD the domain foundry velocity is 0.000000 In the real world there is always some air movement from wind, hvac, movement of the object and people... So it is not unreasonable it set some air velocity in Initial conditions (say 0.05).
The camera is 30 x 30 x ??. What is the other dimension?
First of all let my thank you. The informations and tips really helped me. I did some calculation and experimented with different settings in CFD.
The best results I recived with internal and environment radiation with e=0.8. (like you said)
I set the air velocity to 0.1 m/s (when the environment is "real world")
In our heat chamber there is a higher but unknown air velocity.
The camera is 30x30x40 mm