So, I would like to share my new findings for to the "lightbulb" question that we discussed briefly before the pollution entered the thread. This is related to the Lugano test, but maybe replicators could also find it interesting. Maybe this is obvious for someone trained in the field but it news to me.
Question: Can an internal heat source in e.g. an alumina tube produce more energy than what is actually estimated by using a thermal camera?
I said earlier that a light bulb was an example of such a device, and given that alumina is rather transparent in the visible and in the near infared region it could very well be so for an alumina tube also. Actually I got the opportunity to test a FLIR heat camera on Friday evening, and I used it to look at a halogene larg size (r=3cm) light bulb. My impression was that you could see nothing of the filament inside the lamp on the display of the camera because the glass of the light bulb was not transparent for the wavelengths the camera used. Instead what you could see was the surface temperature of the light bulb and the temperature ranged from 80C to 100C (this was a lamp had been on for a long time).
To make a crude (over) estimation of the power radiated from the lamp we can assume a sphere with r=3cm, T=373K, sigma=5.67e-8, emissivity=1,
Radiated power = 4*pi*0.03^2*(373)^4*5.67e-8W = 12.4W
The lamp had a specified power of 46W
The conclusion is that it is indeed possible to underestimate the internally generated power if there is a partially transmittive shell and a heat camera is used to estimate the radiated power.
It makes me wonder even more why this device configuration and measurement method was chosen for the Lugano test.