View Full Version : Infrared vs. Heat

D Faust
05-20-2007, 08:13 AM
I know this sounds like a stupid question, but what is the difference between infrared light and heat?· A lot of times they seem to used interchangably.· Thanks in advance for clarifing this.·http://forums.parallax.com/images/smilies/confused.gif (Yes, this is related to robotics, I am attempting to sense heat.)

D Faust

Harry Brown
05-20-2007, 08:25 AM
They are the same.

05-20-2007, 09:21 AM
Here is a neat but slightly costly (but worth it) device www.robotshop.ca/home/suppliers/devantech-en/devantech-8-pixel-thermal-array-sensor.html (http://www.robotshop.ca/home/suppliers/devantech-en/devantech-8-pixel-thermal-array-sensor.html)

- Stephen

05-21-2007, 02:07 AM
This is a simplified answer to your question regarding heat and infrared.

Heat is a measure of the vibration of the molecules in a substance.· The more the molecules vibrate the higher the temperature. The energy that·is imparted to the atome within these molecules causes the emission of photons (light particles) that fall within different frequencies.· Some of these photons·can be·within the visible spectrum, such as those emitted from a lightbulb, and some are within the part of the spectrum called infrared.· Even when the temperature of a substance is not high enough for the·photons to be in the visible range,·infrared radiation is present.· As long as the object you are trying to "see" with infrared is sufficiently warmer then·the objects around it you should be able to detect it

05-21-2007, 07:52 PM
Wow Zaphod!· I wouldn't call that simplified.· Just funnin'

The way I rember it is, infrared is a specific spectrum of light invisable to the eye that·can tranfere energy in the form of heat.··Heat is·energy contained in a mass that can be measured by termperature.

05-21-2007, 08:02 PM
"Heat" is a very generic term. When people say they want to "sense heat", usually what they are really saying is "I want to detect an object that is much hotter than its surrounding environment". In a strictly accurate use of the term, you can't really "sense" heat -- instead you can measure an object's temperature, or you can detect the infrared energy a warm object emits.

"Infrared" is a much more specific term. This refers to light in the infrared spectrum -- of lower frequency than human eyes can see. An "infrared" LED emits light in a very narrow part of this spectrum.

05-22-2007, 09:47 AM
Here's my explanation. I think this is correct, but apologies for any errors...

"Heat" is commonly used to mean internal thermal energy, although my thermodynamics professor (many years ago...) would chastise anybody for using the term "heat" in any context other than "heat transfer".

Temperature determines whether a body is in thermal equilibrium with its surroundings. Internal thermal energy and temperature are related, but not the same. For example, a kilogram of water at zero degrees centigrade is substantially hotter than a kilogram of ice at zero degrees centigrade; although the two would exchange heat with another body at a given temperature at the same rate, the water contains more internal thermal energy.

Any body will transfer heat to or from its surroundings if the two are at a different temperature by means of conduction, convection and/or radiation. Radiation transfer to the surroundings of a body is in the form of electromagnetic radiation (ie light, radio waves, infrared) with a spectrum which is dependent on the temperature. Unless the body is at a very high temperature, most of the radiative energy is off the red end of the visible spectrum, that is to say infrared and longer wavelengths. If the body is at a very high temperature, the bulk of the radiation starts to creep more into the visible light spectrum causing it to glow successively red, yellow and eventually white as the temperature increases.

I think that should cover it·http://forums.parallax.com/images/smilies/smile.gif

Any body can be detected by looking for the infrared radiation it emits, which depends primarily on the temperature (and also the properties of the body itself). It's not actually dependent on the "heat" of the body (ie internal thermal energy), although one could call it the "heat transfer" from it by radiation.

To actually detect the IR naturally emitted by most bodies (as opposed to reflected from them as commonly used in object detection) you're going to need an awfully sensitive detector, though.

Hope that helps...



Additional: I just came up with an analogy that might help... but I just came up with this off the top of my head, so it comes without warranty; don't shoot me if it turns out to be flawed http://forums.parallax.com/images/smilies/smile.gif

Temperature is like voltage

Heat transfer is like current

"Heat" or internal thermal energy is like charge

Any body can be thought of as a rather funky nonlinear capacitor, whose temperature (voltage) is dependent in some way on the internal thermal energy (charge) stored in it. The transfer of·heat (current) to another body (another capacitor at some other voltage, through some resistor) depends only on the difference in the temperatures· and the resistance between them. Unless you have a heat pump (such as in a refrigerator), heat transfer only occurs from the body at the higher temperature to that at the lower temperature. If they are at the same temperature, no net heat transfer occurs no matter how much "heat" each contains. "Heat" by the way, can never be destroyed (although it can be created); it can only be transferred to another body.

Key point again: the IR you detect depends on the temperature, not the "heat", although for most purposes this may not be an important distinction...

Post Edited (Uncanday) : 5/22/2007 3:02:37 AM GMT

05-22-2007, 09:42 PM
so how would you right a code for it since iro is light and not heat yes i know this statement is not politically correct but i mean how do you do this cause if temberature effected it wouldnt your tv go crazy on a hot day or is there different ways to rig it

05-22-2007, 10:39 PM
That depends on what you're trying to do.

Your TV doesn't go crazy on a hot day because it's looking for IR modulated at 38kHz or so (the sensors used in sumo bot detectors are designed for that purpose) and because it's looking for a *very* strong signal.

The IR emitted by LEDs in a remote is many, many times stronger than the IR emitted by objects due to their temperature.


05-22-2007, 10:47 PM
science_geek (http://forums.parallax.com/member.php?u=49694),

Before you write code.·The first step would be to determine what you are trying to sense and then, what hardware you are going to use to sense it.


I think Uncanday (http://forums.parallax.com/member.php?u=49669)·said it well, the tv isn't responding to just IR.· The TV responds to a narrow part of the IR spectrum that is modulated at a specific frequency.

Post Edited (agfa) : 5/22/2007 4:45:44 PM GMT

05-23-2007, 05:42 AM
so would it be possible to use the infrared sensor included with the boe bot kit

05-23-2007, 06:20 AM
Short answer: no.

(I'm assuming that it's the reflection mode sensor also used in the sumo kits)

There are two reasons: one, it's not nearly sensitive enough. You're looking for a *tiny* amount of IR to detect temperature dependent IR emissions, and that detector is looking for a blast of IR from a high power output LED.

Two, it has an internal fiter which is looking for a 38kHz modulated signal. This is what the LED transmits, and it exists to filter out interference from other sources. Which is problematic, because those "other sources" are exactly what you're looking for...


05-23-2007, 06:36 AM
ok thanx for the answer ju =st saved me a weekend of dissapointing experiments

Loopy Byteloose
05-24-2007, 08:27 PM
The BOE bot IR sensor has a 38.5K frequency filter, but any IR photo diode can be used to recieve IR without a frequency modulation filter. Since it is lower than 5volts, you might have to use an OP AMP to increase the output to a useful level.

This is quite interesting as each term is dependent upon the body of knowledge that each term is applied to.· How about considering the utility of each term?

IR is used in a frequency dependent context to express a carrier frequency for information.

Heat is used in a power dependent context to express transfer of power or collection of power potential in a reservior.

In other words, you can use an IR sensor to communicate and you can use heat from a blow torch to melt the IR sensor. That may seem silly, but in electronics - heat is a destructive element, not useful. In fact, heat creates white noise inside transistors and tubes.·· The are Heat sensors that measure temperature from a distance; these are quite different from the IR photo diodes.

"Everything in the world is purchased by labour; and our passions are the only causes of labor." -- David·Hume (1711-76)········

···················· Tropically,····· G. Herzog [·黃鶴 ]·in Taiwan

Post Edited (Kramer) : 5/24/2007 1:39:13 PM GMT

05-25-2007, 12:35 AM
I don't think a regular photodiode will be sensitive enough to detect anything useful - passive IR motion detectors use a specialized detector with a built in low noise amplifier in order to get anything useful. Here's a link that describes one circuit in detail:


The circuit it shows is a complete self contained detector system, which is much more than you'd need. I would think something much simpler would work to translate the output of the PIR325 or something similar·into raw digital data.·If you have an ADC available, that would be even better.

My take here is that you're much more likely to get somewhere starting with something like the PIR325 described here than with a regular photodiode. I'd expect it's going to require a substantial amount of experimentation to get working, though. The best shortcut I can think of would be to buy a PIR unit for an outdoor floodlight from Home Depot and salvage the entire module...



D Faust
05-25-2007, 03:52 AM
I was going to use this to find a small·flame such as a candle. (Using two, I would like to find what side of the robot it is on.)· What is the difference between an IR photoresistor, a IR photodiode,·and an·IR photocell?· Thanks for all the input, I think that I have all the definitions of IR and heat that I could possibly need.

D Faust

Post Edited (D Faust) : 5/24/2007 8:58:15 PM GMT

Mike Green
05-25-2007, 04:06 AM
An IR photoresistor is a device that exhibits a light-dependent resistance in response to IR light.

An IR photodiode is a device that consists of a semiconductor diode that generates current in response to IR light.

An IR photocell is usually an IR photodiode optimized for power generation. Sometimes it refers to an IR photoresistor.

There are also photo-multipliers which are avalanche "vacuum" tubes with a small amount of gas in them used for detecting
very low light levels.

All of these are not necessarily limited to IR unless there are filters used to absorb non-IR light. Most of these devices
respond to a wide range of light wavelengths.

05-25-2007, 05:21 AM
I'd honestly recommend going the "home depot" route outlined above. You can buy a PIR module for <$20 and rip it apart for the necessary components. Given the description I've seen for the kind of sensors used in such devices, it appears that it is necessary to go to extreme (and expensive) lengths to get a viable signal, and I doubt that would have been the case if something very cheap like a photodiode could do the job


05-25-2007, 12:26 PM
If you need a PIR sensor go to this linkhttp://www.futurlec.com/cgi-bin/search/search.cgi they have for less than $ 2.00. is the same as motion sensor you can by at the store.
Good, luck.

John Abshier
05-25-2007, 09:10 PM
To detect a candle many of the Trinity Firefighting Contest robots use UV. Here is one sensor

05-25-2007, 09:23 PM
Well, that's $45, which is better than the $90 IR Heat sensor.

D Faust
05-26-2007, 08:13 AM
I have known about the UVtron for a little while, (and worked with it too)·but they are kind of expensive, especially when you account for driver circuit. (the bulb runs on about 300V?)· Even so, only a digital·signal is generated.· I was hoping to get away with a pair if IR and visible light phototransistor-capacitor circuits.· (Using both could eliminate actions based on false readings?)

D Faust

05-26-2007, 02:03 PM
this sounds like what you're looking for:



[edit] this looks pretty useful data also, based on a RE200B which you can find for < $2:


Post Edited (Uncanday) : 5/26/2007 7:11:16 AM GMT

06-20-2007, 04:27 AM
One more IR explanation please. IR radiation falls in two basic categories. Thermal Infrared and Reflective Infrared. Most if not all inexpensive IR detectors can't see thermal infrared light, because glass and plastics will filter out thermal infrared light. Since most of these IR detectors use glass or plastic lenses in front of them thermal IR is blocked by them. PIR sensors do detect changes in thermal IR light and will detect heat from a candle, as long as the heat source is changing by movement or heat intensity. PIR detects changes in heat. Thermal arrays work best, but are expensive, as most thermal IR sensors are. Check out the URL below for a chart showing the light wave spectrum for visible, and IR light.