Theoretical question: Could an incandescent light bulb be used as a temperature sensor?

Question

Let's assume a plain old incandescent light bulb.

It runs on a fixed (assuming an ideal supply grid) voltage and will regulate the current going through it as its resistance raises with the temperature until it reaches some kind of equilibrium.

Then we could assume the external temperature would affect the bulb, and its filament, which would in turn affect the current going through the bulb, to keep things in balance.

Therefore, I question if we could measure the outside temperature by measuring the current going through the bulb?

This is a theoretical question that piqued my curiosity.

A hot wire has been used to measure air flow, but they don’t have a large glass protective shield around them. — Solar Mike, Jul 19 '20 at 19:09
@SolarMike, yes, this is commonly done on cars to measure the air mass going through the intake — Thomas, Jul 19 '20 at 19:12
Tangentially related, but you can use a lightbulb to detect radio waves as well! https://en.wikipedia.org/wiki/Hot-wire_barretter — Hearth, Jul 19 '20 at 19:39

score 3 · Accepted Answer · answered Jul 19 '20 at 19:05

3

Since it is a theoretical question, and no one can prove a negative, I'm going to say yes. You could determine the ambient temperature by observing changes in an incandescent light bulb's current.

I apologize for not knowing the source, but a wise person once said "Everything is a temperature sensor but some are better than others". When the question is not theoretical but practical we will choose one of the better sensors, and an incandescent bulb is not one of those.

answered Jul 19 '20 at 19:05

Elliot Alderson

31,192
5
29
67

1

A search for that exact phrase brings up [this question](https://electronics.stackexchange.com/q/386691/11683) as the top hit. – Dave Tweed Jul 19 '20 at 20:21

score 2 · Answer 2 · answered Jul 19 '20 at 21:00

The best way would be to break the glass envelope and use the filament as an RTD. Tungsten has a positive temperature coefficient of 0.45%/K, similar to copper. Under those conditions the filament would run just a bit above ambient temperature.

It isn't essential to remove the envelope but it would help couple temperature changes to the filament. Some lamps have a vacuum inside, some a gas mixture.

You can do what you are suggesting (in fact most things that can be measured measure temperature in one way or another, from noise in a resistor to gain of a transistor to capacitance of a ceramic capacitor) but it will probably be measuring the convective cooling of the envelope and radiative cooling of the filament more than the temperature of the surrounding air itself, so it will be quite noisy and unstable (a small breeze will result in much more heat flow required to maintain the filament temperature). Also, radiation losses (quite significant at the temperature a filament runs at) are not necessarily related to whatever is near to the bulb, more what is visible from the filament.

Theoretical question: Could an incandescent light bulb be used as a temperature sensor?

2 Answers2