🚀 go-pugleaf

In article <CnEBL0.5Gx@cryton.demon.co.uk> pcats@cryton.demon.co.uk (Patrick Arnold) writes:
>: If they're silicon CCDs, which I believe all of them are, forget it.
>: Thermal-IR photons simply are not energetic enough to register on them.
>
>Sure they are. The only problem is that you need fairly high temperatures. I have
>a small monochrome CCD camera that will easily produce a picture of a hot 
>soldering iron, etc...

Picky, picky, picky...

(For those not up on this stuff:  as the object gets hotter, its emissions
not only increase in intensity, but also shift toward shorter wavelengths.
Eventually, it begins to glow visibly, as witness ordinary incandescent
light bulbs.  Well before that, it's putting out a fair bit of short-
wavelength IR, and thus is easily seen with sensors that are useless for
the more usual run of "thermal" IR.)

>As an aside, I have played with using the PIR detectors intended for intruder
>alarms as sensors in a slow-scan imaging system, with fair results. The main
>drawback is quite short range, ie a few meters or less.

Hmm, yes, I hadn't thought of those.  Those *are* honest-to-God thermal IR
sensors.  Unfortunately, at room temperature their S/N ratio is very poor;
this is why the thermal-IR imagers are cooled.
-- 
Critics have long said "NASA specializes| Henry Spencer @ U of Toronto Zoology
in pork"; now that's White House policy.|  henry@zoo.toronto.edu  utzoo!henry