EEVblog #402 – Flir E60 IR Thermal CameraPosted on December 23rd, 2012 3 comments
A mini review and some playing around with the Flir E60 IR Thermal Camera
your feedburner doesn’t seem to be working for videos anymore. Maybe you could make a post recommending people migrate to the feed you host here?
I do use IR-cameras a lot in heat transfer experiments, so here some comments on your post, to hopefully clarify a few things:
- Hanging picture:
The cam you have is a uncooled microbolometer, which basically is an array of resistors going into thermal equilibrium with the radiation falling through the optics. It has to take an offset from time to time for the Temperature calibration to work. The system has to know its internal temperature without irradiation to correct for it; usually there is a shutter that closes the optical path for taking an offset automatically every minute or so.
Microbolometers work in the long-wavelength range IR (7 to 13mum for the E60). There is only very few materials that are transparent over the full wavelength range (Zinc-Selenide to name one). It’s quite hard to build good optics for those cams, also the optics are responsible for part of the hefty price tag of the system.
- Wall “Temperature”
The hotter spots on your wall, is actually proportional to your wall surface temperature. You can see through by mean of thermal contact only: you have better thermal conductivity where the beams are than where the air is. So in summer the wall would heat up faster where the outside drywall sheet is connected to the one on the inside with a beam. I would attribute the hotter corners to heat transfer as well.
- What do you actually see:
Always keep in mind that what you actually see is radiation! Radiation from an object depends on temperature (forth power -> Boltzmann) and emissivity (shiny metal things usually have low emissivity, but high reflectivity). So even if you set the cam up to some specific emissivity, the hole picture is still just scaled in value and the temperature reading will only be correct where you happen to have guessed the emissivity correctly. Luckily most biological surfaces and most non-metallic surfaces are in the range of 0.92 to 0.99 in emissivity.
(The signal from an object is always the emission + reflection).
You can use to your advantage to improve your measurements: when measuring something hot that has a lower emissivity, choose your reflection wisely (about same temperature). The higher the temperature difference, the farer off your readings will be (fourth power increases quickly!).
Always make sure that the reflected object has uniform temperature and emissivity – best practice: just put a cardboard or thick black paper sheet in the background.
It get’s more complicated if you liked to increase the measurement accuracy in the absolute temperature reading, but that would make this already long post much much longer… In electronics +/-2C is usually good enough, also the incredibly good resolution in “temperature” allows for good hot spot identification without perfect absolute temperature reading.
During thermal experiment the results obtained has micro errors, I think this type of thermal cam can reduce it.
Leave a reply