In the 1st PDF you said:
"Sounds are additive, but not in a simple linear way. Two 30 dB noise sources results in 33 dB. Four of the same noises makes 36 dB. Eight 30 dB noise sources makes 39 dB, and 16 makes 42 dB: Each doubling of identical noise sources results in a 3 dB increase in noise."
Anyway, they're additive. Wouldn't be this a way to let you measure
very low level noise by
adding more sources ? Say, you have a cheaper sound meter that can go no lower than 35dB. You have no way to verify the claim that some computer case or fan is really 28 dB. But if you pile up 8 of them, that would be 28+9=37, well within the range of the sound meter ...
I know, it's super-cheesy and very expensive for one person to get all those, but for a determined retailer (or definitely for the manufacturer) that has many of those items that need to be tested laying around, this could be better than nothing ...
I was looking for sound meters to go very low and I noticed indeed that the closer you come to the magic 20dB level, the more expensive they get, even to ~3K$+. Yay!!
Then, of course, you would need a very quiet room. But with the cheesy-method(tm), you bring the noise-of-interest above the background noise level, then measure it.
Touch my polygon again, and I'll P-P-K you. -Konoko