[vhfpmr (OP)]

If I point this at a piece of clear plastic (or glass) it measures the temperature of the plastic, not the temperature of the view through it. So why does it measure the temperature of whatever object I point it at, and not just the temperature of the window on the sensor?

[alancalverd]

A really good question!

[chris]

Alan, are you beaten?

This is a terrific question!

[paul cotter]

I can't answer offhand but I do know one aspect: the accuracy of the reading depends on the emissivity of the surface in question.

[Bored chemist]

Try pointing it at something warm, like a radiator, with a piece of (cool)  clear polythene between the thermometer and the radiator.

[chris]

Is it something as simple as the device uses a lens to focus the IR from whatever source?

[alancalverd]

Good thinking, BC and Chris!

Infrared themometers are usually classed as pyrometers, so only register temperatures higher than the instrument itself. However non-contact clinical thermometers are often specified to work in ambients up to 45 deg C so apparently can reliably measure temperatures below that of the sensor assembly.

Given a blank sheet, I'd start with a semiconductor Peltier device and some focussed infrared optics, so that the sensor is strongly optically coupled to the target and we measure the power required to bring it down to a reference temperature below ambient. This makes sense for a large (forehead or pizza oven) target but the medical gizmos read tympanum temperature, where the target is commensurate  with the device window. I've used a similar technique to measure nanowatts of optical and ionising radiation but the device cost thousands of pounds, not fifty!

Yes, definitely a good question.

[evan_au]

Infrared Thermometer: how does it measure the target temperature?

Using infra-red, I guess...

What matters is what part of the IR spectrum it uses, and whether it uses one wavelength or multiple
- it does claim to work over a wide range of emissivity (0.1 to 1), so I assume it is more than 1 wavelength?
- It appears to be intended to be used through 14 inches of air, which is pretty transparent to infra-red of most wavelengths.
- However, plastic is a complex web of long organic molecules, which probably have many resonances throughout the IR spectrum, so you would measure the temperature of the plastic?
- Glass has a simpler SiO₂ molecular unit, in a semi-liquid state; it is transparent to some parts of the IR spectrum, so it may be able to measure through glass?

PS: I'm not entirely sure what this means:

Not for Human: The Lasergrip is intended for use on inanimate objects only. Measuring the temperature of a human or animal is unsafe

I assume they mean "Unsafe to a human"? (although if you put it too close to a dog, it might be unsafe for the IR thermometer?)
- If you tried to fit this device in the ear canal, it would do some damage to a human or animal!
- But as it is optimised for a wide range of temperatures (-50C to +610C), I assume it will not be very good at distinguishing normal body temperature (37C) from mild fever (38C)?
- So you should not use it for medical diagnosis?

[alancalverd]

The gadget illustrated in the OP is definitely a laser-guided pyrometer, very handy for spot measurements over a wide range of temperatures (I have one for a pizza oven) but you generally don't want to point a non-surgical laser at a patient's face and it is probably insufficiently accurate for clinical use anyway.

https://www.braunhealthcare.com/uk_en/thermometer/thermoscan-7/ (other types are available) is an example of a clinical ear thermometer that boasts a "pre-warmed tip" and thus can generate an internal reference temperature.

[chris]

I still haven't got my head around the explanations being offered here yet.

Can someone please summarise, in really simple terms, what we think is the answer to this question?

Cheers

[alancalverd]

After a bit more searching I learned that the trick is, as you suspected, to use a lens to focus the target radiation onto the sensor, usually a thermopile. A thermistor provides an independent absolute measurement of the "cold" junction temperature - a "virtual reference point". 

There is some effect from intervening absorbent material such as dust or water vapor. However the lens will be in thermal equilibrium with the sensor so its contribution is negligible.

[chris]

Thanks Alan - and woohoo, I've finally begun to think a bit more like a physicist in my old age!

[chris]

PS - I think this is a brilliant question and I'm very grateful to @vhfpmr for raising it. Can't believe I'd never considered this!

[Bored chemist]

plastic is a complex web of long organic molecules, which probably have many resonances throughout the IR spectrum, so you would measure the temperature of the plastic?

Not all plastics are created equal
Polyethene is transparent to quite a lot of the relevant IR spectrum.



PE spectrum.png (22.17 kB . 489x368 - viewed 468 times)
It's almost as if they chose it deliberately to make the lens of the thermometer from.

[chris]

Just trying to understand the graph, @Bored chemist - so where would be IR wavelengths sit relative to those absorbency peaks for the polythene bonds?

[Bored chemist]

Just trying to understand the graph, @Bored chemist - so where would be IR wavelengths sit relative to those absorbency peaks for the polythene bonds?

All those wavelengths are IR.
I realise that cm^-1 is probably a "cursed unit" (It's proportional to energy, which is sometimes useful in spectroscopy)
500 cm^-1 is 20 microns
3500 cm^-1 is 2.857 ?m
Most of the range of the IR is not absorbed by HDPE

[chris]

Thanks. Now I understand!