How does the MLX90614 Infrared Thermometer Module measure low temperatures?

ElectricAye

I'm baffled by how the MLX90614 Infrared Module is able to measure cold temperatures like -20° without making contact. Because I know nothing about such sensors, I would think that you would need to cool the MLX90614 itself down to at least the lowest temperature you want to measure, but I see no mention of this in the data sheet, etc. So am I missing something here, or can you really wire this up at room temperature and measure something that is really that cold??? or colder???

If so, how is that possible without its own black body heat overwhelming its own sensor???

thanks,
Mark

Post Edited (ElectricAye) : 11/17/2009 12:25:21 AM GMT

mctrivia

infared thermometers are usually fairly directional.

as I am sure you know all things over -273.15C give off infared radiation. The amount of radiation is proportional to it's temperature over -273.15 and will radiate out in all directions.

The infared receiver measures the amount of radiation it receives in a given period of time in a particular direction. since the radiation given off by the module itself will not be in the direction the receiver is receiving it will not be picked up.

In theory a sensitive enough receiver could measure as cold as -273.15 but in practicallity noise will always make that impossible.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
24 bit LCD Breakout Board coming soon. $21.99 has backlight driver and touch sensitive decoder.

ElectricAye

mctrivia said...
infared thermometers are usually fairly directional.

as I am sure you know all things over -273.15C give off infared radiation. The amount of radiation is proportional to it's temperature over -273.15 and will radiate out in all directions.

The infared receiver measures the amount of radiation it receives in a given period of time in a particular direction. since the radiation given off by the module itself will not be in the direction the receiver is receiving it will not be picked up.

In theory a sensitive enough receiver could measure as cold as -273.15 but in practicallity noise will always make that impossible.

Thanks, mctrivia,

I see your point about how everything gives off thermal radiation, but I think the heat transfer is related to (T₁-T₂)⁴ so it seems to me that when you try to measure things very near the same temperature as the sensor, that 4^th power would cause that term to become vanishingly small and so it would be hard to get good readings there. I started to wonder whether or not this sensor has some kind of built in cooler or heater that momentarily shifts its sensor temperature to "dither" (???) out of those awkward moments. I guess I'm just stunned that it can work with such accuracy.

John R.

The sesors are not really measuring temperature. They are measuring infrared radiation. While the sensor may be warm, it is not emitting radiation that will actually hit the sensor.

Think of this more like a light sensor. If you put a photoresisor in front of a light beam, but facing away from the source of the beam, the sensor casts a shadow. It will still read low light, even though the sensor (back side) is bathed in light.

With I/R, you are still dealing with the electromagnetic spectrum, same in most ways as light and radio frequency waves.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
John R.
Click here to see my Nomad Build Log

ElectricAye

Well, I guess one way to look at it: if the sensor is losing photons faster than it's gaining them from a particular direction, then you've got something to measure.
I'll have to get one of these and play with it sometime before I'm happy, though.

thanks,
Mark