metalresearcher - 28-10-2010 at 03:06
Not exactly chemistry related but temperature measuring equipment for the lab.
I got a contactless IR thermometer from ebay which works fine and I have deliberately selected the one with the highest range (550oC), The device
works fine so that is not the problem.
But all these devices are limited to 350-550oC, so for over 550oC I have to stick to my K or S thermocouple.
The hotter the stronger IR so the device needs to be *less* sensitive at high temps.
Wht do these devices have a low limit ?
[Edited on 2010-10-28 by metalresearcher]
Wizzard - 28-10-2010 at 06:07
Perhaps the sensitivity needed for the lower temperatures makes it so a certain type of sensor must be used, which get flooded above a certain
temperature.
not_important - 28-10-2010 at 09:42
Most work on the relative emissivity at one or two IR wavelengths, which haves cures based on the black body curve at a given temperature (see
Wikipedia for black body and Planck's law). The pea wavelength shifts with temperature, and the total radiation is related to the 4th power of the
temperature. Also real surfaces typically are not black bodies but rather grey bodies, and with emissivity that may vary with wavelength. Net result
is that it's difficult to make a single device that is accurate over very wide temperature ranges, as already said the increased power can overwhelm
the sensors intended for lower temperatures, and they're not the right match for the curves of widely differing ranges. Also there's not a lot of
demand for such - industrial user generally have a fairly constrained temperature range for a given process.
Non-contacting temperature measurement of temperatures that have noticeable emissions in the visible are often done using a pyrometer (again see
Wiki).
bquirky - 28-10-2010 at 10:04
I have a feeling that over 500deg C a regular but well characterised RGB digital camera could be made to function as a basic temprature probe.
or you chould get a grating and make a spectrometer to do it 'properly' 
metalresearcher - 28-10-2010 at 13:15
A camera as a temperature probe ?
How do you do that ?
not_important - 28-10-2010 at 13:41
See the topics I gave - the emission curve spectra vary with temperature, contactless measuring systems can measure the difference in light (IR or is
depending on temp) at two wavelengths and determine the temperature of the surface being 'looked' at - given several assumptions.
So in theory a colour camera could tell temperature be the hue of the light, the brightness on two channels, although the sensitivities of the
channels to wavelengths other than those intended - IR and UV - might defeat the proposal.
12AX7 - 28-10-2010 at 22:43
My natural color camera is calibrated to within 10% of the absolute temperature.
Tim
not_important - 29-10-2010 at 09:37
Is that the colour of the illumination? If so it may fail for lower temperatures; a candle flame has a colour temperature of 1800 to 1900 K ordinary
incandescents 2700 to 3300 K, studio/photofloods around 3400, xenon arc is roughly 4100 (it's not a black body fairly flat through the visible with
some big spikes in the near IR), daylight and typical flash 5500 to 6100, overcast daylight around 7000, northern sky light 9000 to 10000. All of
these are for near solar spectrum curves or for sources that act similarly on typical films films, cooler sources may be outside the calibrated range.
metalresearcher - 29-10-2010 at 10:43
That is true but consumer grade (and even professional ones like Canon EOS-1 or Nikon D4) cameras have a color temperature setting starting at 2700K
(2400oC) so unless measuring temperatures in an electric arc or graphitizing Acheson furnace, it is useless for normal lab use.
BTW I just purchased from ebay a Sestos temperature controller . I tested it with a type S thermocouple and the temerature readout worked fine !
Here the user manual:
http://www.sestos-hk.com/english/download/D1S-EN.pdf