I did digging awhile ago into rad collectors to figure out how they generate power - and its surprisingly simple. The equation is as follows:
Mols of Fuel * Rads * 30,000 * ChargeModifier
Which translates directly to the kw output of the collector in any given frame (so its not a direct-to-joule conversion)
What is charge modifier? Simple as well!
Its
(Fuel Temp * 1.5)/150+Fuel Temp
Which produces a flat 1 at 300 degrees Kelvin, which is a hair over room temperature.
Using this math, I can calculate that 5 rads on a collector using a standard 2-mol plasma tank should produce:
2 Mols * 5 Rads * 30,000 * 1 = 300kw per collector
So in theory, if you have 7 collectors and an L3 Singu hitting about 5 rads at the collectors, thats 2.1 MW!
Except not really. For reasons I cannot comprehend, they produce in reality a quarter of that output. Almost on the dime. I have no idea why, but radiation collectors are nearly 4 times weaker than they theoretically should be according to their System.cs file. Is there a hidden multiplier somewhere? Some secondary file modifying them? I'm not sure; but it feels like a bug.