6

I was always under the impression that thermal radiation is essentially infrared energy, so what's the difference between a full thermal imaging camera (e.g. an FLIR) and a standard infrared CCD chip? Surely the latter should function as the former? Yet the price difference is astronomical.

Polynomial
  • 10,562
  • 5
  • 47
  • 88
  • 1
    rawbrawb is correct. An infrared CCD detects "thermal" infrared from objects that are about the temperature of the sun. A FLIR detects infrared from objects that are about the temperature of the human body. The wavelength of these IR emissions is very different. Trying to detect emissions from body temperature objects is much harder when your lenses and detector are near the same temperature. – Joe Hass Sep 02 '13 at 02:43

1 Answers1

9

CCD's are made from Si which has a bandgap of 1.12 eV. This means that it can sense a limited amount of thermal radiation at about ~ 1 um wavelength or shorter. This is called Near IR or NIR. Thermal sensors in the meantime, sense thermally emitted radiation ~ 10 - 14 um wavelength (this is the radiation emitted by a warm body at ~ 300 kelvin). The photons associated with a thermal IR scene are 10X as long and therefore 1/10 th the energy of a NIR photon (1 um vs. 10 um). For a direct bandgap detector, Like MCT (Mercury Cadmium Telluride) these must be cryogenically cooled (77 Kelvin) or the detector will get swamped by it's own heat. There are bolometer based sensors that are less sensitive. But a FLIR is specifically a MCT detector. MCT as a detecting material is very expensive with low yields.

placeholder
  • 29,982
  • 10
  • 63
  • 110
  • I'm not sure I follow - are you telling me that thermal cameras contain cooling mechanisms that bring the sensor down to -196°C? That seems implausible considering that battery powered hand-held devices exist and don't get excessively hot. – Polynomial Sep 01 '13 at 23:09
  • An FLIR is a MCT device and must be cooled. A handheld device is a micro-bolometer with lower performance. You are conflating IR with Thermal IR (from general to specific) and FLIR with Thermal sensor (from specific to general) . Which is understandable because the company that has the largest market share is also called FLIR ... – placeholder Sep 01 '13 at 23:28
  • [FLIR](http://www.flir.com/us/) makes many kinds of sensors, including bolometer arrays that are extensively used in portable equipment. I've recently been working on one system that uses a 640x512 pixel microbolometer array. It's temperature-stabilized, but not chilled. – Dave Tweed Sep 01 '13 at 23:29
  • To top it off, different industries use different definitions for the regions of IR. Astronomy and spectrographic use Mid- IR, the Military uses thermal IR (TIR) for the same 10 -14 um band. IR in general spans a huge band of energies though. – placeholder Sep 01 '13 at 23:30
  • And FLIR started with FLIR's but now also use uBolometers. – placeholder Sep 01 '13 at 23:31
  • There are thermal cameras that use pyroelectric materials rather than MCT and do not need to be cooled. However, their resolution in temperature and space is much less than a cooled detector. – Joe Hass Sep 02 '13 at 02:45
  • "The photons associated with a thermal IR scene are 10X as long". Did't know that photon hava a size :-) – Curd Sep 02 '13 at 08:42
  • Who says that FLIR implies HgCdTe only? As I understand it, FLIR just means any infrared image sensor (not even necessarily a staring-mode focal plane array) that is not "sideways-looking" as with a pushbroom imager. Moreover, since most people are not concerned with aircraft or any other situation where pushbroom imaging would be a viable approach, I think the term is of dubious value and vague meaning when used more generally. – Oleksandr R. Feb 09 '16 at 20:31