Talk:Incandescent light bulb

"The remaining energy is lost as heat"...?
Technically, it should be "most of the remaining energy is emitted as infrared radiation". This topic confuses a lot of people because absorbed infrared radiation heats up an object, but so does any absorbed electromagnetic radiation, including visible light. It is no more accurate to refer to infrared radiation as "heat" than any other energy emission which eventually gets absorbed by matter and converted to thermal energy. 2600:8806:3400:3DB:B9FB:7650:2E02:2ABC (talk) 22:32, 9 October 2019 (UTC)Nightvid


 * That's not necessarily true. First, when talking about energy we're referring to wall-plug efficiency, not luminous efficiency. In wall-plug efficiency, you count the total of the emitted radiant energy, visible, NIR, and UV. For luminous efficiency, we're only counting the visible --in proportion the eye's sensitivity to the various wavelengths in that spectrum. The luminous energy emitted by a light source is always lower than the radiant output. So, since we're talking radiant energy, in this we are counting the NIR as well.
 * What many people don't know is that glass is a poor transmitter of infrared and ultraviolet. It's great for visible and some very near IR, but that's it. At best, if the glass is quartz (fused silica) it will transmit from about 150 nm to about 2000 nm, but most other glasses have impurities that limit their range to a much narrower bandwidth, barely encompassing the visible from about 300 to 900 nm. Beyond that range the glass becomes perfectly opaque and highly absorbent to light (black). All of the energy emitted by the filament that is below or above the usable range is absorbed by the glass and turned into heat. And that is most of it, since it's blackbody radiation. (In comparison, the IR emitted by, say ... a white-hot piece of steel on a blacksmith's anvil is enough to crisp you're eyeballs and redden your skin.) A small amount of that is remitted as much lower (FIR) blackbody radiation, which is not counted, but the majority is removed from the glass due to convection of the air around the bulb. (Not to mention what is lost heating the wiring and everything else. Zaereth (talk) 23:30, 9 October 2019 (UTC)
 * Actually, soda-lime glass is very transmissive out to wavelengths of about 2700 nm (https://commons.wikimedia.org/wiki/File:Soda-lime_glass,_typical_transmission_spectrum_(2_mm_thickness).svg ). Have you ever tried focusing the light from a very bright incandescent lamp (such as a projector bulb) down with a magnifying lens? You can feel a lot of heat, which would be impossible if glass were as opaque to infrared radiations as you suggest. 70.191.56.2 (talk) 15:52, 30 January 2020 (UTC)Nightvid
 * From Far_infrared, it is greater than 15 micrometers. But otherwise, infrared, and especially Far_infrared, is commonly referred to as heat, not just because when absorbed it warms things up, but also because it is emitted by warm objects. Heat equilibrium is commonly described through conduction, convection, and radiation, the latter being pretty much Far_infrared. Gah4 (talk) 03:28, 31 January 2020 (UTC)
 * Actually, soda-lime glass is very transmissive out to wavelengths of about 2700 nm (https://commons.wikimedia.org/wiki/File:Soda-lime_glass,_typical_transmission_spectrum_(2_mm_thickness).svg ). Have you ever tried focusing the light from a very bright incandescent lamp (such as a projector bulb) down with a magnifying lens? You can feel a lot of heat, which would be impossible if glass were as opaque to infrared radiations as you suggest. 70.191.56.2 (talk) 15:52, 30 January 2020 (UTC)Nightvid
 * From Far_infrared, it is greater than 15 micrometers. But otherwise, infrared, and especially Far_infrared, is commonly referred to as heat, not just because when absorbed it warms things up, but also because it is emitted by warm objects. Heat equilibrium is commonly described through conduction, convection, and radiation, the latter being pretty much Far_infrared. Gah4 (talk) 03:28, 31 January 2020 (UTC)
 * From Far_infrared, it is greater than 15 micrometers. But otherwise, infrared, and especially Far_infrared, is commonly referred to as heat, not just because when absorbed it warms things up, but also because it is emitted by warm objects. Heat equilibrium is commonly described through conduction, convection, and radiation, the latter being pretty much Far_infrared. Gah4 (talk) 03:28, 31 January 2020 (UTC)


 * The thing is, all radiation is heat, with the possible exception of ionizing radiation, which act more like bullets. People think of IR as being heat I think because you can feel it on your face, but cannot see it. It's useful in laser cutting because the damage threshold of the optics is directly proportional to the wavelength (ie: a lens that can handle 100 joules per square centimeter per 1 picosecond pulse, at 1064 nm, will only be able to handle 50 joules at 532 nm), and the cut-off for the human cornea is somewhere around 1000 nm, minimizing the possibility of damage to the retina. The shorter the wavelength; the greater the amount of energy (heat) carried by the wave. So visible light actually warmer than IR. (In fact, you can actually assign a temperature to a photon, with X-rays being millions of degrees while microwaves are just a few degrees above absolute zero.)


 * You may notice this in the sun. Most of the IR and UV from the sun is absorbed by the atmosphere, so the warmth you feel on your face is mostly from visible light. And when you focus that light, things catch fire.


 * My mistake for not going back and checking my numbers. Unlike article space, on talk I often am going from memory, so I apologize for that. True, when talking about soda-lime plate glass. there is a large dip at 2700 nm, a peak again at 4300 nm, and then it becomes truly opaque approaching 5000. It is still highly absorbent and reflective; properties which increase as wavelength increases. And that extinction wavelength is still well within the NIR, and not even close to the blackbody peak of a tungsten filament. So you can't really say that all of the energy that is not used for visible light is emitted as IR, because it's a lot more complicated than that.


 * That said, those numbers are also for plate glass at room temperature. The glass used in bulbs, Christmas ornaments, or other thin-walled items is a type containing a high quantity of soda, a low amount of lime, and a lowered amount of manganese. In particular, it's the manganese oxide that augments the natural impurities and makes it clear, so the numbers will be slightly different for bulb glass.


 * The radiation that passes through the glass is called primary radiation. Some applications, such as lighting for thermal cameras, may actually use the secondary radiation that emanates from the glass itself, but that is much, much lower than the primary. Zaereth (talk) 21:40, 31 January 2020 (UTC)


 * Oh, and if you really break it down, of all that energy we expend making light, the only part that is not wasted is the small amount that enters the pupil. The rest just bounces around the room until it either is lost as heat or flies out the window to outer space. Zaereth (talk) 22:34, 31 January 2020 (UTC)
 * Sunlight has a lot of near-IR. If you follow the black body curves at increased frequency, the peak moves up in energy (down in wavelength) with increased temperature. By mid-afternoon, when the UV is pretty much cut off by the atmospheric (or maybe higher up) absorption, the warmth of near-IR causes many to put on unneeded sunscreen. Otherwise, near-IR is most noticeable from heat lamps and catalytic heaters, which produce little visible light, but lots of radiated warmth. Gah4 (talk) 00:40, 1 February 2020 (UTC)
 * Sunlight has a lot of near-IR. If you follow the black body curves at increased frequency, the peak moves up in energy (down in wavelength) with increased temperature. By mid-afternoon, when the UV is pretty much cut off by the atmospheric (or maybe higher up) absorption, the warmth of near-IR causes many to put on unneeded sunscreen. Otherwise, near-IR is most noticeable from heat lamps and catalytic heaters, which produce little visible light, but lots of radiated warmth. Gah4 (talk) 00:40, 1 February 2020 (UTC)


 * True, but when you compare that to the lamp with the closest output to sunlight, an open carbon-arc lamp (which is nearly all incandescent except for some powerful carbon lines in the UV), you'll notice the difference in IR output. Zaereth (talk) 01:10, 1 February 2020 (UTC)
 * I suggest the wording in the article be changed to "The remaining energy is lost mainly as infrared radiation and as heat by convection." 70.191.56.2 (talk) 20:43, 11 February 2020 (UTC)Nightvid
 * Seems a little pedantic. "Lost as heat" is an approachable and adequately accurate description. VQuakr (talk) 20:56, 11 February 2020 (UTC)
 * All of the infrared radiation ends up as heat. The visible light of course all ends up as heat too, but since production of visible light is the goal it's fair to say that the bulb produces a small amount of light, and the rest of the energy consumed ends up as heat. --Srleffler (talk) 01:34, 12 February 2020 (UTC)


 * I agree. Radiation is actually the transmission of heat, but that's really splitting hairs. All energy ends up as heat sooner or later. It's like the lowest form of energy, and, as Gah4 mentioned, all heat transmits by conduction (thermal diffusion), convection (momentum diffusion), and irradiation, and most often all three at once. Simply saying "heat" is an easy way for the general audience to get the picture.


 * Form the OP's signature, I'm guessing they are interested in the wavelengths necessary for nightvision cameras (very NIR which bulbs do emit at much greater intensities than visible light; not to be confused with thermal camera wavelengths or those glasses that shine a green laser and tell you it's nightvision), but that is still only a small fraction of the total energy. We could probably explain that better, perhaps even devote an entire section to it, but this is a very long article for a general audience, and not a class on thermodynamics, so I'd stick with the more general description. Anyone interested enough to want to know all that can easily find it with a little research. Zaereth (talk) 02:08, 12 February 2020 (UTC)


 * If I am in the minority here, I just want to point out that if we use a blackbody at 2500 K - a typical incandescent filament - over 70% of the radiation falls at wavelengths between 780 nm and 2700 nm, the point at which glass starts to become opaque. This is of course more than just the NIR - it includes some of the mid-wave infrared region as well. But most of the infrared energy is indeed in the glass transparency region. Try this tool: https://www.spectralcalc.com/blackbody_calculator/blackbody.php 2600:8806:3400:3DB:9CA7:D0F4:C496:6605 (talk) 23:33, 14 February 2020 (UTC)Nightvid
 * If you are inside, the IR that escapes the bulb will soon hit something else, like a lampshade or wall. If you are outside, some will be absorbed on its way to space, and some will completely escape. For a 2500K black body, how much escapes through all the atmospheric layers into space?  On the other hand, after not so long even the visible portion will be absorbed somewhere, including the retina of our eye, and turn into heat.  About the only case where it isn't, is a silicon solar cell which can convert down to close to the band gap of Si into electricity, but even then, with a fair amount of loss. In that case, the energy between the photon energy and the band gap is lost as heat, or what is otherwise called Non-radiative recombination. (The fancy name for heat.). Gah4 (talk) 23:51, 14 February 2020 (UTC)
 * If you are inside, the IR that escapes the bulb will soon hit something else, like a lampshade or wall. If you are outside, some will be absorbed on its way to space, and some will completely escape. For a 2500K black body, how much escapes through all the atmospheric layers into space?  On the other hand, after not so long even the visible portion will be absorbed somewhere, including the retina of our eye, and turn into heat.  About the only case where it isn't, is a silicon solar cell which can convert down to close to the band gap of Si into electricity, but even then, with a fair amount of loss. In that case, the energy between the photon energy and the band gap is lost as heat, or what is otherwise called Non-radiative recombination. (The fancy name for heat.). Gah4 (talk) 23:51, 14 February 2020 (UTC)

lava lamps, and the Easy-Bake Oven toy
The phrase: lava lamps, and the Easy-Bake Oven toy is in a sentence describing infra-red lamps, which are usually ones designed for heating use, often with red colored filters. Those applications use ordinary lamps for their heat (IR) output, not ones intended for heat output. Gah4 (talk) 19:59, 27 November 2019 (UTC)


 * Heat lamps are specially designed with a glass such as quartz that has fairly good transmittance to IR wavelengths, and are operated with low enough current per the filament so as not to produce white light. but more of an orange/reddish glow. Both lava lamps and Easy Bake ovens use standard bulbs with glass that has poor transmittance to IR. (See two sections above for more on glass transmittance.) Instead these transfer heat mainly due to convection of the air around the bulb, directly heating the air due to contact with the hot glass, which can typically reach temperatures of around 350 F. The Easy Bake actually cooks with convection much like a normal oven, and lava lamps also not only transfer heat from convection, but make that convection visible in the movement of the liquid. Zaereth (talk) 22:46, 27 November 2019 (UTC)

Semi-protected edit request on 9 March 2020
actually, tomas edison didn't invent the lightbulb, he just inproved it. He bought the patent for it from two Canadian inventors, Mathew evans and henry woodward. 207.228.78.204 (talk) 23:06, 9 March 2020 (UTC)
 * Red question icon with gradient background.svg Not done: it's not clear what changes you want to be made. Please mention the specific changes in a "change X to Y" format and provide a reliable source if appropriate.	DarthFlappy (talk) 23:12, 9 March 2020 (UTC)

Phoebus cartel limiting bulb lifespan
The article currently has the sentence: "Early bulbs had a life of up to 2500 hours, but in 1924 a cartel agreed to limit life to 1000 hours.<1>"

The article that it cites claims that this was done to increase bulb sales and is an example of planned obsolescence. The idea is that light bulbs could go up to 2500 hours, and the cartel decreased that just to increase bulb sales to scam consumers. After looking into this, I'm fairly convinced this is a misconception that's been spread and has become a well known "did you know" style factoid. In 1951, Monopolies and Restrictive Practices Commission in the United Kingdom issued a report to Parliament and noted that:

"'As regards life standards, before the Phoebus Agreement and to this day the general service filament lamp was and is designed to have, on average, a minimum life of 1,000 hours. It has often been alleged—though not in evidence to us—that the Phoebus organisation artificially made the life of a lamp short with the object of increasing the number of lamps sold. As we have explained in Chapter 9. there can be no absolutely right life for the many varying circumstances to be found among the consumers in any given country, so that any standard life must always represent a compromise between conflicting factors. B.S.I, has always adopted a single life standard for general service filament lamps, and the representatives of both B.S.I, and B.E.A., as well as most lamp manufacturers, have told us in evidence that they regard 1,000 hours as the best compromise possible at the present time, nor has an evidence been offered to us to the contrary. Accordingly we must dismiss as misconceived the allegation referred to above.' Source" Harizotoh9 (talk) 11:41, 20 April 2020 (UTC)
 * The optimal life depends on the cost of light bulbs, the cost to change one, and the cost of power. For industrial use, where the cost to change is higher, longer life but a little less efficient, by designing for 130 volts and running them at 120 volts. A 60W bulb that lasts 1000 hours uses 60kWh in its life, costing about $7.80 at the US average of $0.13/kWh. Gah4 (talk) 12:56, 20 April 2020 (UTC)
 * The optimal life depends on the cost of light bulbs, the cost to change one, and the cost of power. For industrial use, where the cost to change is higher, longer life but a little less efficient, by designing for 130 volts and running them at 120 volts. A 60W bulb that lasts 1000 hours uses 60kWh in its life, costing about $7.80 at the US average of $0.13/kWh. Gah4 (talk) 12:56, 20 April 2020 (UTC)

Immediately after WW1, Japanese lamp manufacturers entered the American and British markets. British authorities set increasingly strict specifications for efficiency (watts/HK) and service life. As ETZ reported in 1921, a test procedure for service life that was deemed suitable had been proposed the previous year. Since then, this standard has also spread to mainland Europe. So, four years later, Phoebus has not invented anything new. --Virtualiter (talk) 15:19, 30 May 2024 (UTC)

Semi-protected edit request on 21 April 2020
Change "while-light source" to "white light source" in the sentence, "a while-light source with all visible wavelengths present has a lower efficacy, around 250 lumens per watt." Jreamm (talk) 09:24, 21 April 2020 (UTC)

Semi-protected edit request on 21 April 2020
Change "becasme" to "became" in the sentence, "Lamps used for several hundred hours becasme quite fragile." Jreamm (talk) 09:46, 21 April 2020 (UTC)


 * ✅ – Thjarkur (talk) 11:31, 21 April 2020 (UTC)

glue in the cap - "Schellack"/shellac
i can only give as reference a link to the german wiki : "Als elektrischer Isolatorlack auf Wicklungen und als Kitt, der das Glühlampenglas mit dem Metallgewinde verbindet, ist er wegen der gleichen Wärmeausdehnung noch im Einsatz." see under "gegenwärtige Anwendungen" Schellack - wiki--Konfressor (talk) 08:27, 24 February 2021 (UTC)


 * Well, that's very likely so, but this is more relevant to copper wire or magnet wire, or even the shellac article than it really is for this one. That they use magnet wire to connect the electrode-stems, protruding from the bulb, to the threaded cap is no mystery. Magnet wire is ideal in most any situation where you need an extremely thin insulator with good dielectric breakdown resistance, and temperatures that may exceed the melting point of most plastics. It seems like a bit too much detail for this article, which is already over-sized. We're not an instruction manual on how to build them. Note that this does not refer to the seal between the glass and electrodes, which is done by welding to create a hermetic seal. Shellacs and glues emit volatile organic compounds (VOCs) that have a tendency to fill any vacuum, plus they burn up during the vacuuming process when the glass is heated to red-hot to purge any oxygen or water vapor clinging to the glass via adsorption. This only refers to the wires connecting the threaded cap. (Not to mention that Wikipedia is not a reliable source, so you'd need a better one.) Zaereth (talk) 08:59, 24 February 2021 (UTC)

Source for this dubious claim?
"Incandescent bulbs typically have short lifetimes compared with other types of lighting; around 1,000 hours for home light bulbs versus typically 10,000 hours for compact fluorescents and 20,000–30,000 hours for lighting LEDs."

Look, this is a straight up lie. Was this a test from a sterile lab or actual living apartments and homes? All those new high complexity expensive lamps usually die out far earlier than old primitive incandescents. It is so bad, that many people don't see value in changing into new so-called 'economical' one, because they costs far more (5-10 times more), but get broken after less than a year, while old ones work for several years without any problems. Of course, electricity wiring is a factor.
 * Presumably any test has to have some specified control conditions, though they may not be realistic for many real-world uses. Our articles do note that some of the more-modern technologies are more sensitive to environmental and other stresses than incansescents. Every bulb-package I've seen recently has an expected-lifetime on its package. LED lamp seems multiply-cited. DMacks (talk) 10:26, 7 July 2021 (UTC)


 * It's pretty well sourced in the body of the article, although it does talk about a conspiracy to limit lifetime to 1000 hours beginning back in the early days. A conspiracy that, once discovered, was apparently outlawed. The source for that is the IEEE, which is a very reliable source in the field of lighting.


 * The problem with a lot of these numbers is that they are sales brochure-type numbers and are indeed based upon ideal conditions. For example, fluorescents can have drastically varying lifetimes depending on how they are used. Counterintuitively, fluorescents last the longest when they are left on all the time; it's switching them on and off that dramatically reduces their life. When they do wear, it's caused by sputter, which is impossible to predict reliably, because the tiniest difference at the time of manufacture can make huge differences in the rate of sputter. (See: Chaos theory.) Lifetime for fluorescents can vary dramatically between different manufacturers, between different batches from the same manufacturer, or even between lamps that come off the production line one after the other. The best you can get is a ballpark range.


 * With LEDs, the technology is still too new and still changing very rapidly. Lamps that have a pleasant spectrum are still fairly new, there is a total lack of standardization, and suppliers don't really want to stock them because by tomorrow they may be obsolete, and then they'll be stuck with them. In most cases, I believe they are likely basing their info on the lifetime of the LEDs themselves, whereas in my experience the limiting factor tends to be in the drive circuitry (possibly by design). Long life is rarely in the manufacturer's best interest, so it's up to the customer (supply and demand) to dictate what they are willing to settle for. Manufacturers often push consumers in the direction they want by making older technology crappier and crappier, until nobody buys them anymore (planned obsolescence). For example, nearly all the incandescents I find on the market anymore are vacuumed rather than gas filled, which shortens their life dramatically.


 * Unfortunately, we are stuck settling for the numbers provided to us by reliable sources, and with incandescents and fluorescents, we have the benefit of hindsight and a historical overview. Zaereth (talk) 18:35, 7 July 2021 (UTC)

GLS light bulbs
GLS light bulbs redirects here. Perhaps it should say what that means... (Also: I got here trying to find out what GLS means from a listing of an LED GLS bulb, so incandescent-only is probably wrong.) 86.26.33.25 (talk) 18:30, 14 January 2024 (UTC)
 * Fixed. GLS light bulb now redirects to, which now mentions "GLS".--Srleffler (talk) 20:02, 14 January 2024 (UTC)

Are incandescent lightbulbs bad during winter
If most of the energy from these bulbs are “lost” to heat.. isn’t it true that during winter months the heat emitted (from now banned and “illegal”) aids in the necessity to supplement heat in other forms? 136.34.207.21 (talk) 05:18, 20 January 2024 (UTC)
 * An electric resistance heater, particularly one up by the ceiling, isn't a very efficient heater, either. MrOllie (talk) 05:24, 20 January 2024 (UTC)
 * If your house is heated with electric resistance heaters, usually baseboard heaters, but sometimes central heating, it is probably close enough. Coal, oil, and nuclear plants are about 33% efficient in converting heat energy to electrical energy. So, resistance heating is about 33% as efficient as home gas our oil heat. Heat pumps, depending on outside temperature, heat the house with about 3 times the electrical power used. (They don't like to call them 300% efficient.)  From primary energy (coal or oil into a power plant) to heat out of a heat pump, is about 100% efficient. Reminds me, many college dorms don't allow room heaters, but they do allow computers.  Some years ago, people were buying old(er) computers, like desk side Sun systems, which warm up the room pretty well. Computers are also about 100% efficient in converting electrical power into heat. Gah4 (talk) 23:19, 20 January 2024 (UTC)
 * It is hard to make anything that is not 100% efficient at converting electrical energy into heat. Because of entropy all energy ends up as heat, eventually. As you noted, heat pumps are usually more than 100% efficient at heating a space with electricity and burning stuff where you want the heat is usually more efficient than burning it somewhere else to produce electricity, then using the electricity to produce heat.--Srleffler (talk) 23:27, 20 January 2024 (UTC)
 * Yes, in the winter months no energy is actually wasted. It all ends up as heat. While light bulbs are around 2% efficient as light sources, they are in the end 100% efficient as heaters. Even better, much of the waste from an incandescent bulb comes in the form of infrared radiation, so the bulb heats whatever the light hits. In the summer, these bulbs are much worse than you would think since not only is ~98% of the energy they consume wasted, if you have air conditioning the A/C system consumes extra energy to remove that waste heat from the building.
 * It is hard to make anything that is not 100% efficient at converting electrical energy into heat. Because of entropy all energy ends up as heat, eventually. As you noted, heat pumps are usually more than 100% efficient at heating a space with electricity and burning stuff where you want the heat is usually more efficient than burning it somewhere else to produce electricity, then using the electricity to produce heat.--Srleffler (talk) 23:27, 20 January 2024 (UTC)
 * Yes, in the winter months no energy is actually wasted. It all ends up as heat. While light bulbs are around 2% efficient as light sources, they are in the end 100% efficient as heaters. Even better, much of the waste from an incandescent bulb comes in the form of infrared radiation, so the bulb heats whatever the light hits. In the summer, these bulbs are much worse than you would think since not only is ~98% of the energy they consume wasted, if you have air conditioning the A/C system consumes extra energy to remove that waste heat from the building.


 * Over the course of a year, incandescent light bulbs are still a net loss compared to more efficient lighting sources. It's better to have your lighting system make light and your heating system make heat.--Srleffler (talk) 23:24, 20 January 2024 (UTC)


 * I have to agree with Srleffler. Electric resistance heaters are actually the most efficient heaters around. Of course, that doesn't include losses in the generation process or electrical grid, but even so they still tend to outperform all other types of heating.


 * Heat pumps, on the other hand, are a little more tricky. They're advertised as being up 300% efficient, even though that would violate the laws of thermodynamics. In reality, that's like describing the efficiency of a fuel tanker as both the motor efficiency plus the fuel energy it's moving. That 300% efficiency is not the actual conversion efficiency of the heat pump, but the conversion efficiency plus the added heat being pumped, which is different. The conversion efficiency of a heat pump is the ration of electrical input per the amount of heat generated by the compression process. When used for heating, this compression-heat is added to the heat being pumped, so it only seems like it violates the 2nd law. But for cooling the compression-heat becomes waste, so heat pumps are less efficient at cooling than heating.


 * With incandescent lamps, a lot depends on how many windows you have, because a lot of that visible and near-IR will go right off into space if it can find a way out (unlike most IR heaters). Still, I noticed a definite increase in my heating bill when I switched my lights to LED, but in cold climates like mine (Alaska), cooling is never an issue (most people don't even have A/C). And when the Sun is up 20 hours a day in the summer, there's no need for lights. Most people don't live in such a climate, and there are better ways to compensate, like getting an electric space-heater. Zaereth (talk) 21:41, 25 January 2024 (UTC)