I'm new to the forum but I noticed that it is hard to have a focused technical discussion when the topic involves Rossi's reactors.
As I'm interested in discussing the brightness issue I asked the moderators for some help.
They have kindly suggested the concept of "technical threads":
Consider this as "Technical thread".
It starts with some assumptions and try to dig in some technical problems with bounded scope. Debate should be in the topic, and disagreement can be deep but should be kind and in-topic.
Moderators will remove/move any off topic post, like those discussing of general question, conspiracies, off-topic theories, side questions... Any aggressive, repetitive, annoying comment too, not simply insults.
For off-topic exchanges, a parallel [Debate Thread] will be created, with usual rules.
See here: [Debate Thread] Brightness of the reactor glow in the Lugano pictures and reactor temperature
Thus, for this first technical post, the subject is the use of the pictures of the glowing reactor included in the Lugano report to derive constraints on the reactor temperature.
As we are all aware, the thermal camera measurements have been widely criticized for using the wrong emissivity. Bob Higgins suggests that the real temperatures were closer to 1130 C while Thomas Clarke gives re-calculated temperatures ranging from 713 to 779 C.
The reactor appears in figure 2 where it doesn't seem to be glowing, although the heater wires within the insulating hollow alumina tubes are faintly glowing. In figures 12a and 12b it is clearly glowing.
Unfortunately the lighting conditions of figures 2, 12a and 12b are all different.
The illumination in figure 2 includes at least two point sources, as can be deduced from the shadows of the metal framing on the wall and of the tripod leg on the framing.
Figure 12a, which is taken from almost the same point of view of figure 2, lacks any obvious point sources.
Figure 12b appears to have been taken in the dark.
Of course this is an ill-posed problem and a number of assumptions have to be made, and this is why discussion on a forum is likely to be helpful.
I propose the following approach:
- (1) Make assumptions about the spectral responsivity of the three camera channels (R, G, B)
- (2) Estimate the radiance of different objects, based on reasonable assumptions on the illumination conditions and on the radiance of self-lit devices in the three bands
- (3) Average the sensed intensities in the three channels from the images over objects for which the radiance is estimated, and over the reactor areas
- (4) Deduce per-channel radiometric conversion coefficients for each image
- (5) Use radiometric conversion coefficients to convert the sensed intensities over the reactor areas to radiances
- (6) Calculate the predicted integrated reactor radiances for the three R, G and B channels at different reactor temperatures and emissivities
- (7) Deduce the range of plausible temperatures.
I have some results for (1) and have performed some work on (3) and (6), but it's work in progress. I think (2) is the most difficult and most error-prone part. Steps (5) and (7) should be routine.
The goal is to obtain a plot like the following:
(EDIT: V2 of the plot attached. Predictably, there was a silly mistake in the previous plot in the K→T conversion where I added instead of substracting, resulting in a large error.)
The biggest problem with that plot is that the dashed reference line is quite arbitrary (in addition to possible calculation errors.)
I have assumed that a 200 W blackbody 3300 K source illuminates a wall having a reflectivity of 0.7.
Also, any auxiliary information such as the following will be welcome:
- Information on the model of camera used for the Lugano report
- Information on the spectral sensitivity and particularly the near-infrared sensitivity of various commercial cameras
- Information on the room or its lighting
- Information on the backlight radiance of the PCE-830s or the scale used in fig. 1
Hopefully this thread will be a productive one and we'll settle this question; there is a chance that the outcome won't be indeterminate.