[HH] low-cost alternative to a thermal imaging camera

[Tom Metro]

Thermal flashlight 'paints' cold rooms with colour
http://www.newscientist.com/article/mg21328546.200

  The device comes from the Public Laboratory for Open Technology and
  Science, a non-profit group based in Cambridge, Massachusetts, that
  develops open-source tools to allow ordinary people to investigate
  environmental issues.
  ...
  Standard thermal cameras are prohibitively expensive for ordinary
  people.

Right, a Fluke Thermal Imager, for example, will set you back $2000 ~ $3000.


  In contrast, the thermal flashlight prototype costs about $40. What's
  more, it can easily be assembled by someone with no electronics
  expertise.
  ...
  The thermal flashlight is built around a single infrared thermometer.
  This scans an area of wall and picks up varying levels of radiation
  emanating from it. This temperature information is fed into a
  microprocessor, which controls a multicoloured LED light. Shine the
  flashlight against a surface and the colour shows you a real-time
  temperature reading. Areas of the wall with a cooler temperature show
  up blue, while red light shines on patches that register as warmer.

Even easier, you can buy one for $24:
http://www.amazon.com/Black-Decker-TLD100-Thermal-Detector/dp/B001LMTW2S/

I bought one of these a couple of years ago when they cost $50. It's
essentially a run-of-the-mill non-contact IR thermometer, with a slight
twist. They stick a multi-color LED on it that lights up the target area
with green, red, or blue light depending on whether the current spot you
are pointing at is warmer or colder than your starting point. (I forget
some of the details, like whether the color change is relative (I think)
or tied to a fixed absolute temperature. And whether it is continuously
variable (intensity) from red to green to blue, or if flips from one
color to the next when a threshold is tripped (product description and
my vague recollection supports the latter).)

My first thought when getting this device was how one might hack it to
emulate an expensive thermal imager. The initial thought was that you'd
stick it on a tripod, mounted to a computer controlled pan/tilt security
camera mount. Then a computer could make it raster scan a wall. How
you'd capture the color information eluded me.


  An image of the light-painted room showing exactly where heat is
  leaking can then be captured using a webcam with an online app called
  Glowdoodle or just standard time-lapse photography.

Ah! Glowdoodle (http://scripts.mit.edu/~eric_r/glowdoodle/) is the
missing piece. Looks like you need to use it in a fairly darkened room,
which also would work for an old-school solution using a 35mm analog
camera with the shutter held open.

To really emulate an IR imager the Black and Decker thermometer wouldn't
cut it. The wall would appear in only 3 color, and no varying shades.
This is perhaps where the D-I-Y thermometer is superior.

I think the B&D thermometer also has a laser pointer to help you aim (as
is common for IR thermometers), and I don't recall if you can turn the
laser off. That bright red dot would mess up your captured heat map.

One of the sales claims for the B&D was "Helps homeowners track down
power-draining drafts." Before buying I was skeptical how it could do
that. After buying it became apparent that they were stretching the
truth and not directly detecting drafts, but just temperature variations
that are often affiliated with drafts. There are tools that can actually
detect drafts, but this isn't it.

In any case, after playing around with it for a while, I lost enthusiasm
for the tool and never did really make use of it. (I was really far more
interested in detecting actual drafts than measuring insulation quality;
on a first pass for weatherization, you have a far higher impact by
stopping  air leaks than by improving insulation.) I should dust it off
and do something with it.

The real question is why hasn't some manufacturer created a low-end
thermal imager that employs this technique. There are no recent
developments that have just made it possible. All the tech involved was
pretty much available a decade ago.

For example, you create a tool a bit larger than an IR thermometer,
stick a cheap CCD camera into it, and a motorized mirror assembly in
front of the IR sensor so it can scan the field of view. Stick in a
micro to drive the scanner and composite the thermal data over the
video. Add on an LCD for aiming and image preview, plus an SD card and
USB port to get the images out of it. Presto, thermal imager for under
$300. It'd be slow. You'd have to stick it on a tripod. Pros wouldn't
want it, but you might sell enough to non-profit energy conservation
groups and early adopters to get the price under $200, at which point
you'd interest homeowners.

 -Tom