Jan@xxxxxxxxxxxxxx writes:
<< A detector that is at 20 degrees C simply doesn't have
enough energy differential to detect human skin that is only
at 30 degrees C.
>>
Although I tend to agree with much of Jan's post in ***connection with
imaging film ** I disagree with the quoted section. You can go and buy an ear
thermometer at your local drug store for $50 which does almost exactly what
is described without any cooling. The sensor head will be at say 22C ambient
and sensing the ear canal at 35-37C.
They typically rely on an IR lens and a thermopile (array of a few
hundred micro thermocouples). The skin in the ear canal (ideally the tympanic
membrane) approximates a black body radiator and the sensor head maintains an
approximately isothermal radiant environment except where the IR enters
through the lens. The thermopile generates a voltage dependent on the
difference of radiant heat from the enclosure on one side of the thermopile
versus that entering from the lens going to the other side of the pile. To
get absolute temperature the head needs two additional sensors (typically
thermistors) to measure enclosure temperature and thermopile temperature.
The reason it is harder to do uncooled measurements is that the energy
transfer depends on the fourth powers of temperature difference. So an
uncooled sensor has very much smaller signals to work with since the
temperature difference is small. In an idealised test environment the
consumer ear sensors can achieve accuracies of around 0.3C. Clinically they
are not so accurate for a variety of reasons.
Somebody asked about heat seeking missiles and these have huge temperature
differences to work with (1000's C) but need much fast responding sensors,
the consumer thermopiles are pretty slow. The earliest ear thermometers used
pyroelectric sensors and a mechanical chopping scheme. Some newer mems based
sensors (micro bolometers etc) are just becoming available which may be used
in these consumer IR applications in future.
Regards,
Tim Hughes
>>Hi100@xxxxxxx<<
< This message was delivered via the Olympus Mailing List >
< For questions, mailto:owner-olympus@xxxxxxxxxxxxxxx >
< Web Page: http://Zuiko.sls.bc.ca/swright/olympuslist.html >
|