Talk:Incandescent light bulb/Archive 1

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Archive 1 Archive 2


Off Topic Introduction

More than half of the introduction for incandescent light bulbs is talking about switching from them to CFLs. This seems completely bizzare and out-of-place. 01:21, 23 May 2007 (UTC)

Light bulbs are blackbodies...

...not spontaneous emitters. The following:

"As the electrons that travel through the filament bump into the atoms, some of the electrons in the atom may become excited. This means they temporarily boost its energy level and raise to higher orbit. When they fall back, energy is released as photons, a photon being the particle form of light."

is true, but the vast majority of the light comes from blackbody radiation, which isn't even mentioned. I guess this is what happens when you copy from 03:25, 5 July 2006 (UTC) (Andrew)

I agree - as a physics graduate, incandescent bulb spectrums are NOT shaped primarily by electron transitions in **atomic** tungsten. This is somewhat misleading and is wrong if you're trying to explain how an incandescent bulb gets its light spectrum. Black body radiation accounts for most of the spectrum's (continuous) shape and is also the reason why incandescent bulbs tend to look very natural to the human eye compared to other types of lighting - the sun's light is mostly shaped a lot by black body radiation too! Though electron transitions facilitate blackbody radiation, it's not the atomic structure of tungsten that provides the spectrum. I've reworded this. Physicists please check that I've given a correct explanation. 15:02, 8 November 2006 (UTC)

A tungsten filament isn't exactly a blackbody, either (nothing really is) - the emissivity of tungsten varies according to wavelength and temperature. Fortuitously the declining emissivity at longer wavelengths means a coiled-coil filament is a better emitter of visible light than it would be had it been a real blackbody. What we need is a cheap way to cross an incandescent light bulb with a frequency-doubling phosphor of some kind, like IR pumped lasers use, to reclaim some of the IR and fold it up to the visible part of the spectrum. --Wtshymanski 18:07, 2 November 2007 (UTC)

Material from external website?

In revision 139201 an anonymous user inserted a large block of slightly modified text from external websites ([1], [2], [3], [4]). The contributor that did this seems to have copied passsages of text from several different wesites and pasted them together with minor to moderate rewording of some sentences. This is not covered under fair use. Fortunately, others have since copyedited the text and have obscured much of the obvious parentage. More is needed however. Please keep this in mind while editing. --mav

Shouldn't we rather revert to before the infringing material was posted, and then manually apply information that subsequent posters have contributed? -- GayCom

Yuk. I hate version control. It would probably be easier to do some minor reorg, a slight rewrite and some word swapping of the offending text. Much of this text already passes the Google test because of subsequent copyediting. I just don't want to erase the contributions of all those who have edited this article since the violating text was added. --mav

I'll see if I find time to do this. We're not erasing all subsequent contributions. We're just extracting them and reapplying them to an earlier version of the page. I don't think we're on sound legal ground if we just fudge, and don't fix. -- GayCom

Vacuum versus gas filling

Hey, I'm all kinds of new here, but is there *any* substantiation to this "modern lightbulbs filled with inert gas" buisness? I know we have neon lights, but the article seems to imply that modern incandescent bulbs are not evacuated, but any good (although not safe, and not reccomended) smashing of lightbulbs proves that they are at least nearly evacuated. -j

I'm sorry, but your "smash it" test is giving you the wrong data. While it is true that some lamps are evacuated, most ordinary incandescent lamps are filled with argon (or sometimes nitrogen). This is actually essential else the vapor pressure of the hot tungsten filament would quickly boil the filament away to nothing. (A long time ago, 15W or so and smaller mains-powered lamps tended to be evacuated and larger lamps were gas-filled. I don't know if this is still true.)
Classier ordinary incandescent lamps are filled with krypton. I think this is done because the krypton conducts and/or convects less heat away from the filament, leading to higher lamp efficiency. That's why flashlights (torches) so often brag about having a "Krypton lamp!"
And finally, for quartz-halogen lamps, these are actually filled with gas at quite a high pressure. You can confirm this for yourself: the packages for bare lamps always warn that the lamps must be shielded because they may explode on failure (as the resulting arc heats the filling gas and quickly raises the pressure to the bursting point).
Finally, if you don't believe us here on Wiki, consult any manufacturer of lamps. The filling gas is often documented.
(Neon lamps, fluorescent lamps, and other low-intensity arc- or glow-discharge lamps, on the other hand, really are nearly evacuated and fluorescent lamps will certainly demonstrate this if you try your "shatter test" on them.)
Atlant 11:25, 12 May 2005 (UTC)

Thanks for the info!
The fill gas does not oppose "vapor pressure" - if atmospheric pressure opposed the vapor pressure of water, puddles would never evaporate, right? Some small lamps are still just evacuated. If you ever have cracked an incandescent lamp while it is under power, you will soon see the filament literally burning as it oxidizes to a white or yellowish powder. The old automotive sealed-beam lamps were particularly good at demonstrating this since you'd often get a small stone chip in the glass and the lamps would burn out the next time you switched them on. Oxidation is the reason for the bulb, not evaporation. The article itself says that evaporation of the filament is the ultimate failure mechanism of all incandescent lamps - independant of fill gas. I've reverted. --Wtshymanski 21:59, 12 June 2006 (UTC)

Be careful using a shatter test on 'neon' lamps. If you get a cut it can take a long time to heal due to the chemicals used to coat the inside of the glass. At least so I have heard.

Well, some "neon" lamps (the "green" ones) contain phosphor coatings and the phosphors probably aren't especially good for you, but also, some neon lamps contains trace amounts of radioactive elements to reduce and/or stabilize their breakdown voltages. I assume that the radioactive material used takes a solid form (that is, isn't gaseous such as radon or uranium hexaflouride or some such crap). But maybe we'd better discuss this over at neon lamp :-) .
Atlant 15:26, 8 September 2005 (UTC)

Plus, drops of liquid mercury and other nasty chemicals in fluorescent lamps likely turn to minute droplets and go airborne when the tubes are smashed.

I tested a 75 watt GE softwhite incandescent bulb and an offbrand bulb of the same size and found that they were at a reduced pressure of about three quarter of an atmosphere cold. They are said to contain nitrogen plus argon, and to be at a pressure which is close to atmospheric when hot. Edison 23:48, 12 March 2007 (UTC)

Davy's work

The whole paragraph about Davy's work could stand to be rewritten, preferably by someone who has consulted contemporary sources.

I added a ref for Davy demonstrating in 1802 that a battery could make a platinum strip glow. The timeline should be edited to show this, but I do not see it showing up as an editable element of the article. It seems that scores of experimenters between Davy and Edison in 1879 get put forward as the "inventor" of the lightbulb because they replicated Davy, even though their devices whether using platinum or carbon, open air or in a globe, had failings similar to those of Davy's device. Edison 17:53, 19 April 2007 (UTC)
Where is the March date coming from in [5]? Unless it is sourced I will revert to just the year. I still think this date and claim needs to be referenced to better sources, such as original publications from the earliest sources. People writing books later sometimes conflate things ((like arc versus incandescent lights), get years.names wrong, and generally perpetuate made-up "facts" from earlier popular accounts. I have found from 19th century accounts of arc lighting experiments that it was common to have slender carbon rods connected to the battery terminals, then when they were touched together to initiate the arc, they would glow white hot (incandesce) for a brief time before they wer separated and the arc initiated, so I would like to find a good account of how exactly and when exactly Davy experimented with incandescent lighting. Edison 15:07, 20 June 2007 (UTC)
Found [6] wherein the snippet viewable says Davy made platinum and other metals glow with electricity in 1802. In 1809 he used the 2000 cell battery to produce a demonstration of arc lighting. This is from the 1952 Encyclopedia Britannica. Another ref[7] is "The Electrical review" Published 1872, Vol XIV p 341, where in a snip view it says someone gave a lecture and showed on of the cells Davy used in 1802 to produce light (presumably incandescent) from electricity. A good ref appears to be [8], History of Modern Science and Mathematics By Charles Scribners Publishing, 2002, which says in 1802 Davy built a battery, the strongest of that time, with 60 plates of copper and zinc, each 6 inches square. With it he was able melt wire of iron 1/10 inch in diameter and of of platinum 1/32 inch in diameter, and to make glow and ignite sheets of gold, silver and lead. That pretty well covers making paltinum incandescent, but it does not say how lonfg it glowed before melting. "An Introduction to the Study of Electrical Engineering" By Henry Hutchinson Norris, Published 1908, J. Wiley & Sons [9] says that Davy both produced an arc and heated platinum wire white hot in 1802 (p 24), but (p.33) the glowing platinum was "soon destroyed by oxidation." Edison 21:22, 20 June 2007 (UTC)

Edison 21:22, 20 June 2007 (UTC)

I reverted to 1802 because that's what it says in Humphry Davy. If you do find out differently, then it would be nice to keep these two articles in sync (please). I also fixed the reference so that it appears next to the timeline, although it looks a bit odd. --Heron 21:27, 20 June 2007 (UTC)
Davy's electrical works and his "great battery " are described nicely in his 1812 book on chemistry,"Elements of Chemical Philosophy " viewable via Google Books [10]. The actual public demonstrations and publication of work would be a few years earlier, but his clear descriptions are clear prior art for experimenters who kept trying to perfect incandescent lights right toup to the successes of the late 1870's. He speaks of "igniting" platinum wire (p87) meaning that the wire glows and disappears when the current is applied. (It may be melted into tiny globules rather than actually oxidized). On p. 85 he refers to bringing together rods of charcoal 1/6 inch (about 4.2 mm) in diameter and 1 inch (about 25mm) long into contact, in series with the great battery of 2000 plates of copper and zinc. The charcoal became "ignited to whiteness" which I take as incandescence, before the charcoals were separated and a brilliant arc of 4 inches (about 100 mm) was produced. He also describes repeating the experiment in a vacuum, so when he made the carbons to glow white before separating them, the incandescent carbon rods in vacuum were arguably prior art so far as later light bulb inventors were concerned, by 1812. The level of vacuum was "1/4 inch of mercury" (p86) so far from perfect. On pp 84-85 he describes contemporary experiments by "Mr. Children" who had a different battery with large plates (what we would today call a "high current battery" in distinction to Davy's "high voltage battery"). The platinum wire 1/30 inch diameter (about .85 mm) and 18 inches long (7 cm) "instantly became red hot, then white hot, the brilliancy of the light was soon insupportable to the eye, and in a few seconds the metal fell fused into globules." (pp 84-85). Davy had done similar experiments making shorter platinum wire glow white hot with his own battery. The trick was not getting platinum or carbon to glow, in air, vacuum, or inert gas, the trick was getting a high enough level of incandescence, with a sufficiently white light, without going the little additional bit and causing it to melt or fuse. People like Lindsay and many others amazed their townsfolk and reporters by replicating experiments from a 1812 chemistry book and making platinum wire glow for a short while. Thomas Edison's first efforts were to try and build a regulator which would hold the platinum right at the desired temperature but interrupt the current momentarily before it reached the melting point, before he switched to a high resistance carbon filament in a high vacuum. Edison 15:07, 12 September 2007 (UTC)

incandescant platinum

The article Light bulb says in section History of the light bulb

In 1801 Sir Humphry Davy, an English chemist, made platinum strips glow by passing an electric current through them, but the strips oxidized too quickly to make a useful lamp.

I assume (not much of chemist, i) that oxidation of Pt is unlikely, and that this misinterprets Lights by Edward Willett, or something similar, which refers to

  • Davy's "fine platinum wire" and
  • to other incandescants where "the filament burned away very quickly".

I say "oxidation ... unlikely" in the belief that Davy chose Pt bcz that (and maybe Au) were the only things he could find that didn't oxidize when incandescant (and maybe bcz Au melts a lot earlier).

Here are some #s from WP:
Parameter ---------- W ----- Pt
Melting point --- 3695 K 2041.4 K
Boiling point --- 5828 K 4098 K
ρ-1 (106/m ohm) -- 18.9 09.66 (or should i be suspicious of the leading zero?)
At. wt. (amu) --- 183.84 195.078

What i understand to be the main failure mechanism in ductile-tungsten filaments is the positive feedback at the hottest points, of

  • increased thermal migration of W,
  • thinning,
  • increase of resistance-per-unit-length,
  • increased share of delivered power, and
  • increased T,

culminating in local melting across the whole cross-section, of enuf length that sag (or just instability?) lets the liquid portion flow away from its center. (Haven't noticed if that's in the article; it should be IMO.)

IMO, the presence or absence of air doesn't make a lot of difference in that kind of failure, bcz even if convection is more efficient for the first few inches than radiation alone, incandescence is required, and the operating temperatures are going to be essentially the same.

With Pt in the same geometry such failure just happens sooner; i imagine much sooner: i base that on the principle that chemical reaction rates go up very rapidly with temperature bcz they are controlled by the population in excess of the activation energy, which increases much faster than mean energy, and on my belief that thermal migration should similarly limited by an activation energy representing a surface atom breaking all but two (or if not, all but one) of its bonds, so it can swing to its next position.

The higher conductance of W (which surprises me, but whadda i know) does mean the Pt filament has twice the cross-section area for the same length, voltage and current, but that may not be what determines the geometry: with modern W, it may be economical for filaments to be made thicker and longer (for ruggedness and reduced thermal-migration failure, but with Pt the cost of filament material has to limit the feasible increase of dimensions. So i think the lower conductance is completely swamped by the melting point and perhaps dimension issues, and i am convinced the same non-oxidative failure mechanism can explain Davy's results. --Jerzy '03 Dec 15


Sublimation is another problem. -- Heron

Thx! On reflection i agree. I may have been searching for my keys under the street lamp, esp since i despair of having any hope of estimating sublimation rates, either relative between materials or relative to the solid-state thermal surface migration -- other than to note that activation energy for sublimation has to be higher than for surface diffusion. Maybe i should have taken another thermo course or four.

I think sublimation can also participate in the feedback loop, but at least initially it promotes a graceful degradation, with the bulb getting cooler (and dimmer) as the filament thins and delivered power drops.

(My first thot on reading the sublimation contrib was "Ah, then the bulb maintains a vapor pressure of the filament material, and makes a difference other than via restricting convection to a remote heat-flow mechanism." No, fool, the bulb just makes it obvious that sublimation is a second thermal-migration mechanism that you didn't even consider: thermal migration via gas phase to the sub-incancesent bulb instead of via solid phase within the filament. 'Scuse me, gotta go off and cry somewhere.) --Jerzy 19:41, 2003 Dec 16 (UTC)

Don't be too hard on yourself, Jerzy. You were right to dispute the "oxidized" claim (which I just changed).

Looking into this further, I think sublimation may be a less graceful degradation mechanism than you say, but I'm not sure. For analytical purposes, I imagined a filament divided into 100 segments in series. For a 240 V 100 W bulb (common here in Britain), I calculate the filament resistance when hot to be nominally 576 R. In my scheme that's 100 5.76 R segments in series. If, due to evaporation, one segment doubles in resistance to 11.52 R, then the total filament resistance will increase by 1% to 581.76 R. With the same supply voltage, the power will drop by nearly 1% to 99.01 W. As you said, this will cool the bulb and help to slow down the evaporation. On the other hand, think what happens to the thinned segment. The current through the filament has now dropped by 1% (from 0.4167 A to 0.4125 A), but the resistance of the unfortunate segment has doubled, so the power in that one segment has now increased by 96% (from 1 W to 1.9606 W), making it evaporate much faster! It seems that the increase in local power dissipation will strongly dominate the slight decrease in global power dissipation, causing a runaway failure, but I haven't proved that this is true for all cases.

If you have been, thanks for reading this far! -- Heron

Absolutely i'm reading you! I think we agree on what you just said; our difference is that i assumed (without justification) that evaporation/sublimation (let's just say erosion, in contrast to the reshaping of the same mass of filiment by thermal migration) was uniform; as you point out, the positive feedback effects apply either way to the extent that either phenomenon is non-uniform.
Now looking into the justification, i continue to think that the overall trend of the erosion is gradual loss of light output. However, i agree that erosion can magnify non-uniformities about as effectively as migration.

A big issue i keep coming back to, tho, is that we'd like to find phenomena that work when the non-uniformities are infinitesimal. I think your example points out the fact that power goes up faster than segment resistance, because P = VR2, which enhances the feedback, at least once the uniformities get large. But what i'm starting to worry about is that unless we're talking about irregularities that are there from the factory machining, they are likely to be swamped by the thermal noise. IMO there are some things that work against the feedback: i think infinitesimal surface roughnesses are subject to negative feedback: atoms at convex surface points are bound to fewer atoms and therefore (i think) less strongly bound; atoms at concave surface points by the same token more strongly bound. We therefore expect the background surface diffusion (the component not driven by temp differences) to favor moving atoms from convexities to concavities, and smooth out the surface. Gets more complicated as we examine more closely, i fear.

It sounds, tho, like you know no more than i abt how hard it is to oxidize Pt at those temps, & i fear our alternate failure mechanism (instead of fairly uniform erosion (or resistance increase, on the assumption that the oxide is non-conducting but non-volatile), by oxidation) is irrelevant - i.e., if there's nothing wrong with the previous editor's oxidation hypothesis!

What i haven't done is finished searching the web for statements that Davy's Pt failed by oxidation; i've been treating the suspicious text like a smoking gun, but there may be a more credible source out there that says "oxidation" explicitly.

Interesting, nonetheless. [smile] --Jerzy 03:28, 2003 Dec 17 (UTC)

You are right to say that I don't understand the chemistry of what's happening. I will leave that to someone else.

"The Invention of the Incandescent Lamp" by Globusz Publishing, among other web pages, points out (in somewhat convoluted English) that the temperature at which platinum glows is very close to its melting point, which is what makes it so unreliable. I think this is the (disappointingly simple) answer. -- Heron

I hope you will not mind if I throw my two cents in.
The failure modes of incandescent lights has interested me too. I think evaporation is an important failure mode, and you can confirm this by noting that high intensity bulbs build up a dark layer on tungsten (or whatever) on the inside of the bulb as they age. The halogen cycle helps slow down this loss. The bulb temperature must be very high though for this trick to work. This is why quartz glass must be used. Halogen lamps intended for projection use are rated for 50 hour life and are specified to operate at about 50C below the melting temperature of tungsten. The melting temperature of tungsten, I believe, is closely related to, and limits, the quoted "color temperature" of tungsten light. I believe that lamp lifetime varies with the sixth power (I forget, but the effect is big) of operating voltage, and I could look up the empirical derating formula if anyone cared.
What I really wanted to offer, though, is another observation. I believe that it is common for lamps to fail at the moment they are turned on. I have noticed that when incandescent lamps are dimmed very low using a phase control dimmer that they emit a noticable sound. I believe that tungsten lamp electrical loads draw a nominal 10x current surge when the load is first turned on, that is, the rule of thumb for sizing switch contacts, etc. is to expect this much surge. I believe this is due to the cold filament having a much lower electrical resistance initially. Careful examination of the phase control circuit waveform shows that there is an enormous, short duration current spike on each half cycle as the partially cooled filament is connected to a relatively high voltage, instead of the normal smooth sinusoidal voltage buildup. I believe that some of the corresponding noise may be from thermal distortion as the filament heats rapidly. However, I suspect, that some of the noise may be due to forces created by the rapidly changing, 10x magnetic field around the (usually) coiled filament, due to the current surge, acting on the conducor. I think this magnetic effect may also account for the tendency of the bulb to fail as it is turned on, by adding one more stress to the cold and weakened metal.

I added a source which says Davy demonstrated in 1802 that current from the powerful battery at the Royal Institution could make a fine strip of platinum glow, althoug it drew too much current and did not last long enough to be a practical lamp. The cost of a new strip of platinum each time it was lit for a few minutes was another limitation on the practicality. How can the time chart be edited to show this being in 1802? It now looks like it is before that, but I do not see the chart showing up as an editable element of the article. I seem to recall reading somewhere that Volta made a fine wire glow in 1800, but I cannot now find a reference to substantiate that. It appears that every experimenter after 1802 who made a fine platinum wire glow for a while gets credited as "inventor" of the incandescent lamp, when they did not materially improve on Davy's 1802 demonstration. Edison 17:48, 19 April 2007 (UTC)

arc lamp

Quoth the article:

In 1809 [Davy] created the first arc lamp by creating a small but blinding electrical connection between two charcoal rods connected to a battery. Demonstrated to the Royal Institution of Great Britain in 1810, the invention came to be known as the Davy lamp.

This is quite different from the Davy lamp described in the linked article (basically, a safety mechanism for an open-flame "lamp" that allows it to be used in mines where flammable gasses may be present). Is the description of the lamp here wrong, or is it not the same lamp that came to be known as the Davy lamp, or is the other article at fault, or what? The other article _feels_ more reliable to me than this one, but that's a highly subjective assessment made without any real background in the subject matter, and in any case even if the Davy lamp article is assumed correct it is not obvious how to rectify this (incandescent light bulb) article. The section probably ought to be flagged for reworking, but I'm not sure which tag best applies. For now I have used the "Not verified" flag; feel free to replace it with a more appropriate one if such exists.

"Stupid Light Bulb Tricks"

I was browsing around and came across a guy's rant [11] which also includes a lot of great info on light bulbs. I'm not sure if it's proper to include as an external link, though.

rename to just "light bulb" ?

I see that light bulb redirects to incandescent light bulb. I think it should go the other way around -- this article discusses flourescent, carbon arc, and other non-incandescent lights. So I think this article should be titled simply "light bulb". Should we simply move the article, or would it be better to split it up into a general "light bulb" article and a specific "incandescent" article ? --DavidCary 03:08, 10 Sep 2004 (UTC)

I say we should keep the name of this article as it is, and move all discussion of non-incandescent sources to other articles. I suppose that includes my table of efficiencies of various types, which would be better off in a general "electric lighting" article.
I suggest this structure:
As you can see, most of these articles exist. --Heron 08:14, 10 Sep 2004 (UTC)
I agree this page should be moved, or a separate article should be created at light bulb. - SimonP 05:34, Dec 17, 2004 (UTC)

Please note that most of the articles use the noun "lamp" rather than light. So if you're going to move this, it would probably be better to move it to "Incandescent lamp" rather than anything named "... light".

Atlant 13:01, 17 Dec 2004 (UTC)

I agree that this page should be re-named (moved to) Light bulb. --Jatkins 11:47, 1 April 2007 (UTC)
  • I disagree with calling incandescent lightbulbs simply "lightbulbs" since this same term is and will be used for compact fluorescent bulbs. This article should specify what kind of lightbulb it is about. For ceiling light fixtures with exposed bulbs, I use the extra-cost style of compact fluorescent enclosed in a softwhite A19 glass shell, identical to a 100 watt incandescent. I would refer to this compact fluorescent as a lightbulb without further specification in general conversation. In the history of electric lighting, Nikola Tesla made a 19th century light bulb which had no filament, and was not a fluorescent. The term "lightbulb" is too genericv and nonspecific, without the qualifier "Incandescent." I checked for common usage on Google. Eliminating Wikipedia and its mirrors, "Incandescent light bulb" gets 216,000 hits. "Incandescent lightbulb" gets 37,600 hits. "Incandescent lamp" gets 501,000 hits, and is the term used by Encyclopedia Britannica and Encarta. Therefore I recommend this article be called "Incandescent lamp." Edison 17:35, 2 April 2007 (UTC)

Edison screw designations

Now I'm confused. My recollection was that MES was minuature Edison screw, the standard torch (flashlight) bulb, and that LES was liliputian Edison screw, an even smaller bulb used in torches and as indicator lamps (in the days before LEDs). But the article uses MES to mean medium Edison screw, the standard light fitting. I always thought this was just ES. Also pretty sure I've come across things described as SES (small Edison screw) -- perhaps this is equivalent to the candelabra fitting mentioned? Or perhaps I'm just confused Roy Badami 18:33, 28 Mar 2005 (UTC)

We should probably eliminate the abbreviations as I suspect they vary from place to place around the world. Here in the U.S., the names of the screw bases are, in descending order:
  • Mogul (as seen on very large line-voltage lamps)
  • Medium (as seen on most light bulbs)
  • Intermediate (I think; fairly rare)
  • Candelabra (as seen on many decoractive lights)
  • Miniature (flashlights/torches/T-3ish indicator lamps)
  • Subminiature (I think; T-1ish indicator lamps)
Atlant 19:49, 28 Mar 2005 (UTC)
i'm sure theese ar least one size known as lilliput edison screw (i dunno if its the one you've mentioned as subminiture) which is smaller than miniture. i belive its what was used in christmas lights before they wen't to wire end bulbs in push in caps. Plugwash 03:34, 26 January 2006 (UTC)

Article says "the E stands for Edison, who created the screw-base lamp, and the number is screw cap diameter in eighths of an inch in the U.S." This does not compute, as that would mean an E26 is 3.25 inches in diameter (8.3cm). Thirty-seconds of an inch would make more sense. I'm going to take out the "eighths of an inch" until a better explanation is provided. ---Ransom (-- 16:53, 9 May 2006 (UTC))

And not all lamps "screw" into the socket - some lamps just push and twist. Hence my last edit. --Wtshymanski 22:02, 12 June 2006 (UTC)

It would be useful to include a bit about what Type A, Type B, etc. bulbs are. The article only discusses the screw fitting, but many manufacturers refer to Type A bulbs...


LEDs are much more efficient than the stated 20lm/w. see:

Not true - at least, not true for white LEDs. Depite the hype that pervades the industry about the future potential efficacy of LEDs used for lighting, the actual specs of real-life LED products for lighting applications is in the range of 20 - 25 lumens per watt. This is less than one third of the efficacy of compact fluorescent bulbs. You can verify this by checking the specifications of products of leading high-brightness LEDs from manufacturers such as Lumileds:

Colored LEDs may be quoted with higher efficacies because green and adjacent colors sit at the peak of sensitivity of the eye. However, this is no indicator of their usefulness for general purpose illumination. You have to be looking at the white-light specs in order to do an apples-for-apples comparison with other white light sources such as compact fluorescent.

"Light Globe"

In Australia and South Africa a light bulb is also called a light globe, but this term is not used elsewhere.

Well, my grandmother in Pennsylvania calls them light globes, so the above claim is too strong. I always figured it was a largely obsolete term from her youth, that has fallen out of favor in the U.S. Anyone else know of "light globe" being used outside of Australia and S.A.? -- Coneslayer 15:09, 2005 July 28 (UTC)

I confirm that the phrase is popular though not exclusive one in Australia. (And someone has nicely put a redirect in place.) Anyone know about new Zealand? 04:43, 28 April 2007 (UTC)


Maybe someone can offer some insight here. The spectra that I recently uploaded of a (unfiltered, clear) flashlight bulb [12] appears to have a peak emission of around 630 nm. However this [13] says the filament should therefore have a temperature of ~4600K which is clearly impossible.... Does the glowing filament deviate from a blackbody that severly as to make an approxamation like this impossible? What happened?--Deglr6328 06:39, 5 September 2005 (UTC)

Comparisons between lamp types

There seem to be several sections in this article that independently make comparisons between Incandescent bulbs and various other types of lighting. eg. Comparison of electricity cost compares with the cost of compact fluorescent, Efficiency compares with a bunch of other types, Heat makes comparisons with compact fluorescent lights, and so on. I was wondering if it'd be beneficial to have a separate article to consolidate comparisons between lighting types... or maybe it could be part of the light bulb article (mentioned above), or the lighting article.

I've worked on the light pollution article a lot recently. I started drawing a table comparing different types of lighting, but then began wondering such a list really belonged there. For one thing, it seems to be duplicating a lot of lighting information available in many of the different lighting articles, just as the comparisons in this article probably duplicate the same information elsewhere. Does anyone have any thoughts about this? Izogi 07:34, 6 September 2005 (UTC)

A comprehensive comparison, probably in table form, would be a great thing. You could make it a template so that it could be included in as many articles as it applied to.
Atlant 17:19, 6 September 2005 (UTC)
Thanks, the template sounds like a good idea. I'll put it on my list of things to do, and perhaps experiment with it over the next few days. Izogi 06:10, 7 September 2005 (UTC)

Tungsten Alloy?

I wondered if the tungsten use was an alloy or pure (mostly) tungsten? Anyone know?

I have always believed it was pure tungsten (as you want the highest-possible melting point, and the un-alloyed metal should have that), but I have no reference to back up my assumption, and I wouldn't be too surprised to find that some small amounts of alloying metals were added to improve the drawing characteristics of the pure tungsten wire. Perhaps someone else really does know :-) ?
Atlant 15:11, 8 September 2005 (UTC)
Creep (deformation) says that they use an alloy to reduce creep. njh 03:18, 31 December 2005 (UTC)


Efficacy might be a correct term, but to me it seems less specific than energy efficiency, and the linked efficacy article doesn't (currently) seem to or address the idea at all. On the contrary, it's very weighted towards psychological meanings, with small cameos for medicine and politics. It isn't anywhere near as helpful as the energy efficiency article was. The efficacy article could almost be a disambiguation page, or have some redirect text at the top to point people at energy efficiency if they want something about energy. Anyway, I'm not quite sure what the best approach is here. I'd quite like to point the links back to energy efficiency, because it seems to be more specific and it is a more widely recognised term, in my opinion. In the interests of avoiding an edit war, though, does anyone have any strong opinions on this? Izogi 22:24, 9 September 2005 (UTC)

Just because people dont understand a correct term is no reason for it to be excluded. There is a link to the term. Maybe the link needs modifying. But please be assured that the term efficacy is the correct one --Light current 07:33, 10 September 2005 (UTC)

But what was incorrect about energy efficiency in the first place? I just think it's clearer, not to mention more specific. Izogi 08:15, 10 September 2005 (UTC)

There does seem to be a difference between luminous efficacy and luminous efficiency. Try Googling for "luminous efficacy" "luminous efficiency". See this PDF from IUPAC [14] for a definition, but this is in the context of the reception, not the emission, of light. To summarise, efficacy is the lumens/watt column in our table, and efficiency is the "percent" column, which is the ratio of efficacy to the reference efficacy of 680 lm/W. Here's a tech note (PDF) from Pulnix [15] that says the same thing in simpler language. --Heron 12:29, 10 September 2005 (UTC)
If we said 'the luminous efficacy (sometimes erroneously referred to a luminous efficiency).....' would that satisfy everyone?--Light current 18:27, 10 September 2005 (UTC)
That is roughly what I put in the article a few hours ago. --Heron 21:23, 10 September 2005 (UTC)

Fine with me. Izogi 21:37, 10 September 2005 (UTC)

Yes it looks a lot more accurate now!--Light current 21:45, 10 September 2005 (UTC)

The heading "efficiency" in the table is incorrect and misleading. The percentage figures in that table express the lumens delivered by one radiated watt as a proportion of the theoretical maximum of 681. They depend only on the shape of the spectrum and they do not account for any "efficiency" concept whatsoever. To use the term "efficiency" in this context is misleading - well, actually not misleading - just plain wrong. Using the term "efficacy" is also wrong because that's a clearly defined term and this isn't it. I am not aware of any properly defined term that describes the figure shown in that column so I can't offer a solution - I only offer that "efficiency" is NOT the correct term.

You're absolutely right. I'm working on a complete rewrite of that chart. I have already gathered efficiency data for white LEDs, tungsten incandescents (clear, soft white, long life, "daylight"), thin wall BT15-style halogens, and heavy wall halogens. I'm working on long tube fluorescents and ballasts. After that, compact fluorescents, high and low pressure sodium, mecury vapor, and metal halides. I will take a look at xenon arc lamps, but I suspect they are not really used for large scale illumination. Every number in my chart has been taken from the technical datasheet of a currently shipping lighting component. Every lamp type includes specific examples from at least three manufacturers. For LEDs, I used Cree, Nichia, and LumiLeds. For incandescents, I used Philips, Osram-Sylvania, and General Electric. I should finish this up in a week or two. —Ryanrs 10:54, 16 March 2006 (UTC)
Forgot to mention, the new chart will include color temperature and CRI Ra values, when available. Osram even provides CRI R9 figures for their warm white LEDs, which is nice. —Ryanrs 10:59, 16 March 2006 (UTC)

Vandalism? "Amanda 'Mugsy' Emms"

When you go to "Incandescent light bulb" from the Search it gives you this: The invention of the light bulb is usually attributed to Amanda "Mugsy" Emms, who contributed to its development by producing a practical and viable electric lamp. However her invention was stolen by Thomas Edison who marketed the device successfully. Alexander Nikolayevich Lodygin developed an incandescent light bulb around the same time. Many others also had a hand in the development of a practical device for the production of electrically powered lighting.

But when you go to "Light Bulb" from the search it gives you this: The invention of the light bulb is usually attributed to Thomas Alva Edison, who contributed to its development by producing a practical and viable electric lamp, and marketed the device successfully. However, it is now believed that Heinrich Göbel built functional bulbs three decades earlier. Alexander Nikolayevich Lodygin developed an incandescent light bulb around the same time. Many others also had a hand in the development of a practical device for the production of electrically powered lighting.

I honestly don't know much about the lightbulb so I'm not going to correct it on a hunch (no matter how big), but I've never heard of the Amanda Emms version of the story. Also why is there two different pages that claim to redirect to the same place [Incandescent light bulb]? --Taboo Tongue 20:13, 19 November 2005 (UTC)

Woodward & Evans

It's my understanding that the Woodward and Evans light bulb did in fact work and that Thomas Edison thought that their approach was sufficiently promising to buy rights to their patents. I am not an expert in the field, however. My source is [16]. I would gladly defer to another who is more expert or who can provide a more authoritative source. On another note, what's the point of emphasizing that Woodward was not a "gentleman" (presumably a noble?) but "merely" a hotel operator. It may be true, but unless it's relevant to the history of the develpoment of the light bulb, I recommend that that fact be removed.--Nowa 00:15, 21 December 2005 (UTC)

Lots of people made wires or other substances glow with electricity,from the invention of the Voltaic pile onward, but those pre-1879 devices simply did not work in terms of operating for more than a few minutes or giving off enough light to replace gas lights, or drawing a small enough current that they could be operated from an electric system. Patenting something or having an idea is not "inventing" it. The Canadian pride sites are not reliable sources. Find a scholarly article in a peer reviewed journal, or a scholarly book from a respectable publisher, not a blog or someone's website. The patent drawing shows a big chunk of carbon, which would not have constituted a "filament". It would have drawn high current. There is no believable claim that thjey demonstrated a light bulb which was bright enough and long lasting enough to be usable, ignoring the requirement for high resistance and a small radiating surface. Nor is there evidence that their work influenced Edison or Swan. Edison 15:12, 20 March 2007 (UTC)

Improvement Drive

Thomas Edison has been nominated on WP:IDRIVE. Vote for this article and help improve it to featured status. --Fenice 14:02, 26 December 2005 (UTC)

Too many links

There are too many links, especially in the intro paragraphs, and some ("archaically"?) are not really relevant to the context. See Wikipedia's style info on this. I'm taking out links to some plain English words and irrelevent words. Benhoyt 03:10, 6 March 2006 (UTC)

11W CFL != to 60W incandescent

I have changed the comparision of electricity cost to reflect more accurately the lumen output of a CFL. 13-15w CFL's output 800-900 lumens. In the range of a 60W lamp. See for an excellent sampling of CFL wattages and lumen outputs. The existing references also support ~14w CFL =~ 60w incandescent.

Yep, ok. Looks like the CFL manufacturers are misleading the public (and hurting themselves by giving CFL a repuation for being dim!). 14W would line up with the approx 4* greater luminous efficiency listed. --njh 04:30, 20 March 2006 (UTC)

Lifetime of CFL vs incandescent

Now I'm no expert however I've been purchasing light bulbs for a while and I'm pretty sure the expected life time of a CFL one and an incandescent one is roughly the same, this is ofcourse greatly company dependent but if someone with the education could confirm and/or correct this?

Sorry, but the life of any "general service" fluorescent lamp will greatly exceed the life of a typical "general service" incandescent lamp. In round numbers, a fluorescent lamp lasts 10,000 hours whereas an incandescent lamp lasts 1000 hours.
One thing to be aware of: Each "start" of a fluorescent lamp reduces its lifetime a little (as it blasts some cathode material off the filaments of the lamp). So if the on-off cycle of a fluorescent lamp is very short, then its life won't approach that average 10K hour number I cited above. On-off cycling also has an effect on incandescent lamps, but it's much less significant, especially for lower-voltage lamps.
Another thing to be aware of is that by derating the service voltage, incandescent lamps can be made to last much longer than is normal (but their brightness and efficiency goes down as well). In North America, a typical "long life" incandescent lamp is actually a 125V or 130V lamp; when operated on 120V, it can live a very long life. The lifetime of an incandescent lamp is approximately inversely proportional to the sixteenth power of the voltage! See the article for more details.
Atlant 13:33, 31 March 2006 (UTC)

Here in Norway (and Sweden?), the normal average lifetime for an incandescent lamp is 2500 hours. This is confirmed by independent tests (text in Norwegian), and a 40 W bulb brand with such lifetime was measured to emit 396 lumen on average, which seem to compare fairly to US bulbs. Osram, the manufacturer, makes lamps in their German factory rated at 1000 hours, and claims the longer life of the lamps made in Norway is due to special regulations! I wonder why the rest of the world doesn't copy our regulations... --Kjetilho 00:40, 1 August 2006 (UTC)

From the first decade of lightbulb manufacturer, makers have known they could trade off efficiency versus lifetime. This they now sell "long life" bulbs with a 130 volt filament rating for use on 120 volt mains. The engineer's rule of thumb is that a 10% voltage variation halves or doubles filament lifetime. A bulb with a 130 volt filament used on 120 will have a warmer color tone (lower filament temperature). Voltage affects lifetime of fluorescent bulbs as well, but I do not know the tradeoff equation. There is a longer empirical equation with several terms, which suggests that a 240 volt bulb powered by 120 volts should last decades, but in practice I have seen such buylbs fail in a few years. I expect that a voltage spike from utility faults or switching or a nearby lightning jolt every couple of years may give the lie to claims of 50,000 hours for LED lights or the very long lifetimes for compact fluorescents, just as they cause the failure of other electroniuc equipment. We should guard against this article becoming a bandwagon for proposals to ban incandescents. Such coverage should not become as long as it presently is, beyond mention in the section on efficiency. The CFL bandwagon should not come to dominate this article, any more than an article about buggies talks about how great cars are. There is a separate article for CFLs. Edison 00:04, 13 March 2007 (UTC)

The section on Power (as well as comments above) provides helpful information regarding long-life (130-volt) bulbs. However, it's not clear whether the chart in the Power section ("Comparison of efficacy by wattage") assumes these bulbs are operated at 120v or 130v. Hopefully the former, but in any case the assumption should be stated in the explanatory text next to the chart.

Definition of efficiency

It makes no sense to define an efficiency percentage of a light bulb without a certain reference, since you cannot divide lumens by watts and get percents out. If you have a hypothetical light source that converts 1 W of electrical power into 1 W of light at 675 nm (typical wavelength for red laser pointers), then the energy efficiency of this light source is 100% (1 watt useful output per watt of input). However, according to the definition in the previous version (perfect monochromatic light source at 555.6 nm is 100%), the efficiency would be 2% and the other 98% are, as written in the beginning of the article, "wasted as heat". I don't think it is appropriate to call this "wasted as heat" just because the eye is a less sensitive to this wavelength than green 555.6 nm light. The appropriate way is to take the ratio useful output power (in watts) and total consumed power. One way to do is to take the part of the power which is emitted in the visible range, i.e. 420 to 700 nm (see luminosity function), in order to not count infrared and ultra-violet radiation. However, this would give too much weight to the almost-IR and almost-UV parts of the spectrum, and give an incandescent lamp a much higher efficiency than it deserves. Since the table is about efficiencies of sources for illumination, it has to evaluate light sources with a fairly broad spectrum. A light source that converts electricity into visible light in the range 420-700 nm cannot have an efficiency higher than 242 lm/W due to the fact that the eye needs other wavelength components than just 555.6 nm green in order to get the impression of a white light source. It is possible to calculate the optimum spectrum that will give you most lumens per watt while still give you daylight white rather than green. I don't know where exactly the 242 lm/W comes from (the original link pointed to an irrelevant page), but it seems quite reasonable judging from the shape of the luminosity curve, e.g. the number of lumens if you take one watt of flat-spectrum light in the range 420-700 nm.

So, to summarize: if the best possible spectrum that looks like white gives you 242 lm/W, then this should be considered the 100% efficiency point of an illumination device —Preceding unsigned comment added by hankwang (talkcontribs)

I agree with hangwang's changes to the table. This brings it in line to what my college physics book says for the efficiency of incandescent and CFL lights. Perhaps the terminology needs to be clarified, but as far as the percentages go, they should remain as Hangwang has them. Perhaps the ideal monochromatic source should be removed. Why use pure (monochromatic) green light as a reference? TimL 15:39, 16 April 2006 (UTC)

Ok I went back and read the table and the paragraph above it. The table is luminous efficiency, NOT luminous efficacy, so again Hankwang's changes are correct. The row for monchromatic source should probably be removed. TimL 15:42, 16 April 2006 (UTC)

Furthermore, the definition of Luminous efficiency is total lumimous flux over total radiant flux. So for both ideal white light source and monochromatic (555.4nm) source the correct percentage value would be 100%. I am now going to make this update. TimL 15:52, 16 April 2006 (UTC)

NB: The ratio of luminous flux to radiant flux is higher for monochromatic green light than it is for "white light". The table was correct as it was, although it was mislabeled since the column labeled "efficiency" was actually "overall luminous efficiency", as defined in the text above the table. The luminous efficiency is normalized to its highest possible value, which is the luminous efficiency of monochromatic light at the peak of the luminosity function.--Srleffler 04:00, 17 April 2006 (UTC)(composed after the text below).

Leaving aside for the moment the issue of whether Hankwang's argument is correct, his reversion was not a good edit. If you look more closely, you'll see that I made a number of other changes, including carefully going through and fixing the references, which were messed up. The reason one of the references seemed irrelevant was because edits had caused the footnote numbering not to correspond to the correct notes. (This was a problem with the old footnote system. I changed the table over to the new footnote system.) The other problem is that, independent of what definition of efficiency Hankwang feels is appropriate, the definition of luminous efficiency used in his version of the table did not agree with the definition in the article text.

Now, as to the technical issue: What needs to go in the article is not what Hankwang things is best, but what is actually used. If someone can cite a reference that defines luminous efficiency, that would be justification for changing the table. Otherwise, I propose we keep what we had. I need to revert the edits to fix the references. If you want to change the numbers please edit the table instead of reverting it back to the version with the broken references. But, please provide a reference that justifies this change in the contents of the article rather than an argument about what seems best.

Even in terms of what seems best, though, the arguments above are incorrect. A monochromatic green source has higher luminous efficiency (and efficacy) than a white light source because if you compare two sources with the same power the monochromatic source will seem much brighter. You can see better by the source with the higher efficacy. To use Hankwang's example above, it's perfectly reasonable to say that the luminous efficiency of a 675 nm source is only a few percent. Anyone who has ever compared a 1 mW red laser spot to a 1 mW green laser spot knows this. Similarly, a monochromatic green source is brighter than a white light source of the same radiant flux. It's not great for illumination because it isn't white, but such sources are used. Ever seen a green LED nightlight? They are very energy efficient, and yet you can see quite well by them. In general, Hankwang's argument seems to not understand the purpose of the luminosity function or of luminous efficiency. Hankwang also seems to confuse "efficiency" and "luminous efficiency". These are not the same thing. Efficiency is about how much light is emitted, vs. how much is wasted as heat. Overall luminous efficiency is about how much of the energy consumed produces something that is useful for vision. Red light is less useful for vision than green light. White light is less useful for vision than green light too (in the sense that the latter is brighter than the former, for equivalent levels of radiant flux.) --Srleffler 03:47, 17 April 2006 (UTC)

Good points. There are many different kinds of white! CRI) and color temperature then effect luminous efficiency. Perhaps the article should mention CRI and that standard incandescents have a perfect (100) CRI, while CFL's and even LED's do not. (and of course a green light as a CRI of 0) TimL 04:21, 17 April 2006 (UTC)
It would certainly be interesting to include both overall luminous efficiency and color rendering index in the table. They are not in general correlated with one another. I wonder if confusion between the two was what prompted Hankwang's change.--Srleffler 04:28, 17 April 2006 (UTC)

11:52, 16 April 2006 edit

The comment for this edit should have read "100% is the maximum possible value for luminous efficiency." TimL 15:54, 16 April 2006 (UTC)

I'll repeat and amplify what I said above, for others who may have missed it. The luminous efficacy of a monochromatic green source at the peak of the luminosity function is higher than that of any "white" light source (ideal or otherwise). If you don't normalize to a monochromatic source at 556 nm, you are forced to accept luminous efficiencies greater than 100%. For clarity, the table actually lists overall luminous efficiency as defined in the article. The last two entries (ideal white light and monochromatic light) are just the luminous efficiency, which is equal to the overall luminous efficiency for an ideal source that exhibits 100% conversion of the energy it consumes into electromagnetic radiation.--Srleffler 04:34, 17 April 2006 (UTC)
OK, conventional usage is a reasonable point, and in the current version the table is labeled correctly as "luminous efficiency" rather than "efficiency". But I think you will agree that the remark in the earlier article version (under "Heat") that the other 98% are wasted as heat were not appropriate. Currently it states the more reasonable 95%, but I think it is rather confusing that the numbers in the "Heat" section are different from those further down. How about an extra table column "Efficiency (usefull watts per watt) with numbers normalized to a white source?
One more thing: there appear to be several definitions of the luminosity function. The wikipedia page lists that Km=683.002 lm/W at 540 THz (555.14 nm), not 680 lm/W as in this article.
And how about moving this section to a separate page (e.g. "Luminous efficiency") that all wikipedia pages about light sources can link to?
Han-Kwang 14:19, 17 April 2006 (UTC)
Yes, I agree that if the previous version's "heat" section claimed that 98% of the energy was wasted as heat, that it was inappropriate if not incorrect. The energy not counted in luminous efficiency is certainly wasted, and of course all wasted energy eventually ends up as heat, but the statement would still be terribly misleading. Much better to say that x% of the energy is wasted as heat and an additional y% is lost because of the eye's limited spectral response.
A column giving true efficiency (watts of radiant flux per watt of power consumed) would be fine, but that doesn't seem to be what you mean. I don't know what you mean by "useful watts" per watt, here. If you're thinking of something like taking the portion of the power that falls within the visible spectrum as "useful" watts, then the result would be misleading and arbitrary. The visible spectrum does not have sharp edges, nor would it be meaningful to have a sharp cutoff between "red" and near infrared or "violet" and near ultraviolet. The eye's response tails off gradually, and it is the luminosity function that accounts for this rolloff in response. The spectrum integrated with the luminosity function is the "useful power", and it is measured in lumens, not watts because those are the units of "useful power" of light (also called luminous power).
The luminosity function is a defined, international standard. Any disagreement is probably due to errors on Wikipedia. The 680 lm/W was probably just rounded off.
Moving this table sounds like a good idea to me. I would move it, and the associated text and footnotes, to luminous efficacy. That page contains a slightly different definition of luminous efficiency, which could easily be harmonized with the definition here. Note though that there are other pages that refer people here specifically for this table. Such references would have to be sought out and fixed.
By the way, be aware of the Wikipedia policy No Original Research. We are not supposed to create new knowledge or definitions here, but to report what is being used elsewhere. This is one of the core principles that guides Wikipedia. It's sometimes helpful in discussions like this. We should not be thinking about what seems reasonable or true, but rather what we can find evidence or references for. The question is not "what would be the best way to define an 'efficiency' for light sources?", but rather "How do people (in published works elsewhere) define an 'efficiency' for light sources?".--Srleffler 00:34, 18 April 2006 (UTC)
That's why setting the ideal white-light spectrum as the benchmark for 100% efficiency seems reasonable. Since we agree that it is a reasonable statement that an incandescent lamp converts 5% of the electricity into visible light and wastes an additional 3% because of the spectral response of the eye, you would need to provide some definition how to calculate this 3%. I agree that Wikipedia is not a place for original research, but it is very commonly claimed that ILs have 5-10% efficiency, and this claim is probably based on some kind of definition. (I have a busy week ahead of me, and won't be able to spend time on this subject) Han-Kwang 09:43, 18 April 2006 (UTC)

Luminous efficacy and efficiency table

This lists glass halogen (2.3%), quartz halogen (3.5%) and tungsten-halogen (2.6%-3.6%). It seems like we have some overlap here in terminology, creating a longer table than necessary. (I also don't see how we need three rows for LED's but thats another story). Aren't glass halogen and quartz halogen two types of tungsten halogen lamps? Seems to me like we could do with just listing quartz halogen, or tungsten halogen.

About white LED's, since it is both a rapidly devloping and emerging technology, perhaps we could just have a footnote at the end of the table mentioning their possible efficiencies. TimL 04:01, 18 April 2006 (UTC)

I took out the excess halogen entry. I think the white LED entries are interesting and useful precisely because they are an emerging technology. These are commercially available for lighting, so they should still be included in the table. (I have an LED lamp on my desk, in fact.) The "prototypes" entry is useful, because it indicates the potential of this technology.--Srleffler 11:25, 18 April 2006 (UTC)

Looking at the Klipstein reference for LED prototypes I found little evidence except for a very vague press release from Cree regarding 100 lm/w led. The press release refers to 'results', not any actual working prototype. I have created a new reference that talks about a prototype LED that procuces 80 lumens per watt, but only at extremly low currents. I also saw at the Klipstein page nothing about high vs. lo0w power LED's. So I removed the Low power line. In fact the chart on the new refernce I created suggests that LED's have a higher luminous efficiacy at low power. TimL 21:58, 19 April 2006 (UTC)

Why is there a table with efficiencies of various sizes of incandescent bulbs but no reference? I have heard 5% elsewhere but now this article says 2%. Sounds bogus. The Office of Energy Efficiency of Canada [17] says "(Only 4–6 percent of the electrical energy used by an incandescent light bulb is converted into visible light. The remaining energy is lost as heat.)" A "green" website in Nova Scotia, "Clean Nova Scotia" [18] says "Because ninety-five percent of the energy used in a light bulb is radiated as heat from the filament. Only five percent of the energy used by a standard incandescent light bulb produces light." The 2% figure in the article should be replaced by 5% until someone finds a reliable source with the lower figure. There should be accurate lab results in journal articles somewhere.

Edison 14:00, 2 April 2007 (UTC)

The article at Luminous efficacy seems to substantiate quite a range of efficiencies with 1.9% being referenced for a 40W conventional incandescent and 5+% for a high-temperature incandescent. This matches well with the figures I've seen bandied about through the years; an ancient General Electric guide to light bulbs that I used to have certainly cited 2% as well. I'd suggest that we document the full range of "typical" efficiencies rather than trying to agee on a single magic number.
Atlant 16:24, 2 April 2007 (UTC)

Confusion over bamboo and cellulose coating in Edison's design

What does the following sentence mean?

Bamboo continued to be used until 1893, later enhanced by a cellulose coating, introduced around 1882 and produced until at least 1929.

Bamboo was obviously used without cellulose coating until 1882. From that year, it was coated in cellulose, but what happened between 1893 and 1929, when bamboo use had supposedly ceased, but cellulose coating had not? LX 06:05, 20 April 2006 (UTC)

The edit that seems to have injected the confusion is this one by, who was only seen for two days in December 2005. —LX 17:24, 29 April 2006 (UTC)
Since nobody has come forth with an explanation, I have removed the sentence. Feel free to insert a more coherent sentence in its stead. LX 05:50, 4 May 2006 (UTC)

Alessandro Cruto?

Could be insert also the contribution to this invention by the italian scientist Alessandro Cruto? HI

-- 10:47, 8 May 2006 (UTC)

Quartz Halogen Fingerprints

What is it about a fingerprint that destroys a quartz halogen lamp? According to, (last section, "PRECAUTIONS TO PREVENT DEVITRIFICATION OF QUARTZ ENVELOPES"), "Fused quartz is vitreous in nature but, like other glasses, can be stimulated to crystallise (devitrify) by external influences at high temperatures.... Sealed quartz cells can be used for thousands of hours without devitrification if precautions are taken to ensure that the outside surface is scrupulously clean before raising them to temperature. Any surface dirt, a water spot or a single fingerprint is a potential seed for devitrification."

The oils or fats left by the finger cause uneven heating of the surface, which puts it under stress. --Kjetilho 00:25, 1 August 2006 (UTC)

Comparison of Electricity Cost

I added a paragraph to this section, which I'll quote in its entirety here: "It should be pointed out that virtually all of the energy from light bulbs of any type is converted into heat. During cold months when a building is being heated, the heat produced by a light bulb is helping to heat the building, and this needs to be taken into account when calculating the energy cost of different types of bulbs. In a typical home using electric heat, for example, light bulbs of any type do not produce any electricity cost during cold months, since the heat produced by the bulbs simply offsets heat that would have been produced by the home's heating system."

Every discussion I've ever seen on the cost of using different types of bulbs always completely ignores the fact that light bulbs do not "use up" any electricity whatsoever, they simply turn it into heat. If the bulb is in a building which needs to be heated, as is true much of the time, then the inefficient old incandescent bulbs can be every bit as efficient as the fluorescent or LED bulbs, since the heat being produced by the bulbs is helping to heat the building. --Xyzzyplugh 13:04, 24 June 2006 (UTC)

Well, I don't know about you, but using electricity for heating would not be my first choice. Since you usually have to burn something (coal for most of the country) for the electricity, then you are already at 40% efficiency for heating. Using natural gas or heating oil is about 90-98% efficient. If not wasting the heat, then you are wasting you're money. And here in NC, heating costs far less than A/C in sticky summers, so not only is the heat wasted, it would make my A/C work that much harder. --mjtimber 18:25, 13 October 2006 (UTC)

External Links

Does the "Operation removes lightbulb from anus" news link (added with this edit: have any place in the External Links section in this article? I think not, but I thought I should ask before removing it. It seems the point of the linked story is that the guy somehow ended up with an object in his anus, and the fact that that object was a light bulb isn't terribly significant. I think the link adds no value to this article and therefore should be removed. Ra* 23:36, 14 August 2006 (UTC)


Hard to believe you can get sunburn from halogen UV - I'd say it's the heat, not the UV that burns your skin. I added the {{fact}} template, a reference would be nice if this is indeed true. Sure, filtered halogens are used as UV sources (e.g. a dentist's filler hardening light), but the percentage of UV is so small that normally, the glass in front of a halogen is for explosion, not UV protection. (Maybe the writer here confuses halogen with HMI lamps - they produce copious amounts of UV!)--Janke | Talk 10:03, 1 October 2006 (UTC)

Unfortunately, you are wrong. See:
Unfiltered, undoped halogen lamps are a significant source of UV light.
Atlant 22:14, 1 October 2006 (UTC)
Thanks! You really should put a reference to these pages in the UV section of the article, to stop others from making the same mistake I did. Greetings, --Janke | Talk 06:11, 2 October 2006 (UTC)

Left-handed lightbulbs

I heard somewhere that the New York Subway used light bulbs with a thread that is a left-handed helix instead of the standard right-handed helix, to discourage people from stealing them and using them. Does anyone have a good reference for this factoid? Perhaps it could be mentioned in the article. Itub 21:14, 2 October 2006 (UTC)

I don't know about the subways, but I've heard the same thing about 12 Volt bulbs that use the Edison base. I think we need an expert. ;-)
Atlant 22:53, 13 October 2006 (UTC)

CA, C7, C9?

I've been trying to find out the difference between CA, C7, and C9 bulbs. One, or all of them, is the Candlabra kind, but I don't know the difference. These terms may be used in the U.S. only.

I don't know about "CA", but C7 and C9 are both standard American designations that follow the usual scheme. "C" means "Candle" (flame) shaped and 7 and 9 are the size in 1/8s of an inch, so a C7 lamp is 7/8 inch in diameter while a C9 lamp is 9/8 inch in diameter.
C7 lamps used to be the standard size for indoor Christmas lights (with a "candelabra" screw base); nowadays they're mostly used only for nightlights and a few appliance lighting applications.
C9 lamps used to be the standard for outdoor Christmas lights (with an "intermediate" screw base).
Atlant 00:37, 21 October 2006 (UTC)


Do you think that the redirects at the top of the pages are maybe a bit too confusing? You know you can follow 'incandescent light bulb' to 'lamp (electrical component) to 'light fixture', then to 'stage lighting instrument' And these are all quite substantial articles. I think a disambiguation-type page would make things a lot easier to navigate...Theonlyduffman 07:19, 7 November 2006 (UTC)

the physics of the emitting of light

This article seems to me pretty lacking when it comes to the physics of incandescence..

all it says is: 'An electric current passes through a thin filament, heating it and causing it to emit light.' It really should go more in depth about the thermal motion of electrons and the fact that they the acceleration of electrons results in emission of EMR. But I'm not the one to write about it as I've only learnt this stuff in my year 12 physics course this year.

Its rather important though... the fundamental aspect of how it functions, which sets it apart from UV lamps, etc.

Tungsten being re-deposited at the hot-spots in halogen lamps a myth?

The german version of the article claims that the following statement about halogen lamps is a common myth: "A tungsten-halogen lamp creates an equilibrium reaction in which the tungsten that evaporates when giving off light is preferentially re-deposited at the hot-spots, preventing the early failure of the lamp." It claims that it is actually preferably deposited at colder spots, and the prolonged liftetime of a halogen lamps is mostly because of their higher pressure. —The preceding unsigned comment was added by (talk) 18:37, 5 December 2006 (UTC).

Atlant 16:52, 14 December 2006 (UTC)

Chlorine used in lamps?

Jobrien13057 added chlorine to the list of halogens used in tungsten-halogen lamps. I've reverted that out for now, but I'm more than willing to be proven wrong with a citation. Anyone have one?

Atlant 16:52, 14 December 2006 (UTC)

Molten, boiling filaments?

It the beginning of the article, it says: The current heats the filament to an extremely high temperature (typically 2000 to 6000 K depending on the filament type, shape, and amount of current passed through).

But I'm inclined to doubt this information, as tungsten melts at 3422K, and boils at 5555K, making me think that the filament is never "typically" heated to such a high temperature.

—The preceding unsigned comment was added by (talkcontribs).

This information was inserted by on 8 November 2006 in an otherwise-valid edit, but I'll bet they acquired some Fahrenheit/Celsius/Kelvin confusion along the way. I understand that tungsten-halogen lamps run about 50C/K below the melting point of tungsten, so feel free to be bold and fix this up!
Atlant 18:12, 18 December 2006 (UTC)

Names of the canadians who sold the patent to edison ?

I forget the names myself, but could someone add that to the history.

—The preceding unsigned comment was added by (talk) 02:53, 4 January 2007 (UTC).

edit regarding John Wellington Starr

It appears that John William Starr had the first American patent for his carbon-filament incandescent light bulb, but I can't seem to find much other information about him. —The preceding unsigned comment was added by Monstrosity (talkcontribs) 07:39, 12 January 2007 (UTC).

I think it's John Wellington Starr, and a rather sad story, died aged 21 in Birmingham England.
Edit: a little more here: Hakluyt bean 22:32, 12 January 2007 (UTC)
My mistake, thanks for the link Monstrosity 16:45, 13 January 2007 (UTC)

What are limits of link spam?

Many links to on other articles were link spam. After removing a link to on this article, I wasn't so sure it was link spam, so I reverted my link removal. I'll watch what others do and follow their lead. 22:45, 26 January 2007 (UTC)

Efficiency Trumps Health

Is there truth to the rumor that the light from incandescent light bulbs is more comfortable and less damaging to the health of human eyes, than is the light from fluorescent light bulbs? White collar workers who work daily under the light of fluorescent light bulbs report more eye strain? Are the increased medical costs and anguish worth the decrease in greenhouse emissions here? 17:18, 20 February 2007 (UTC)

Even if it were true, the point of Wikipedia talk pages is to discuss the articles, not the subject of the articles. You might want to ask again at the Sciences Reference Desk.
Atlant 17:53, 20 February 2007 (UTC)
Yes it is true that many of us can not use fluorescent lights and have to use old style incandescent light bulbs. Ask any one with epilepsy. Also many with migraine (not regular headache but the real sickness migraine with aura attacks etc) get attacks from fluorescent lights. If they decide to outlaw light bulbs in my country I guess I will have to stock up enough light bulbs to last me for the rest of my life. And I think it will cause black market smuggling and selling of light bulbs in California and Australia where light bulbs now are going to be outlawed. --David Göthberg 00:17, 22 February 2007 (UTC)

Warren De la Rue's date is wrong?

I was looking around at other websites (namely and that claim Warren De la Rue enclosed the platinum filament in the year 1820 rather than 1840 as the article says. Is Wiki wrong? Throktar 05:49, 27 February 2007 (UTC)Throktar

I noticed this same thing, after looking around I found several sites that said this. I am going to change it to 1820 for the time being, if someone finds a very reliable source that says 1840 feel free to change it back. 00:48, 28 February 2007 (UTC)

Warren de la Rue must have been something of a child prodigy, because he was born in 1815 and was only 5 years old in 1820. 1840 is the year you're looking for, guys. - Rob Carter -

Yes I noticed this too. But I was unsure if it was even the same person. When I read through Warren de la rue's page, I saw only things about achievements in Astronomy and nothing about the light bulb. I still havn't found a source that says 1840 as the date either. User:Throktar

Please provide a source which satisfies WP:ATT before claiming 1840 or 1820. Unsourced claims may be removed. The article on de la Rue in Wikipedia is not a reliable source. Harold Evans "They Made America" (Little, Brown and company, New York, 2004) says (p152)that before Edison's 1879 invention of the evacuated bulb with the carbon filament of high resistance and small radiating surface, which gave off more light than a gas light and lasted long enough to be comercially practical, and drew little enough current to be supplied from a generating station, "The idea of the bulb itself had been around for 50 years; many other inventors had tried to heat filaments to incandescence: Ira Flatow, the science reporter, gave up counting at 23." But it had not been "invented" 50 times because none of them worked or were practical or useful. On page 159 Evans says "Nobody in the world in 50 years of experiments had been able to keep an incandescent light alive for more than a few moments". Swan gave up in 1860 after experimenting for 12 years. If de la Rue or anyone else had invented a lightbulb which gave off a good light in 1840, why would the leading scientists such as Silvanus Thompson call Edison's claim an incandescent lamp could be made to work "sheer nonsense?" Sir William Preece British electrical expert claimed that "A subdivision of the electric light is an ignis fatuus" or "foolish fire. Sawyer predicted ignominious failure and denied that any light buld could be made to give off several candlepower for more than 4 hours. Edison 16:16, 20 March 2007 (UTC)
De la Rue is a fairly obscure scientist. I could find nothing specifically about his work in a moderate sized public library, and the encyclopedia articles give no specifics on light bulb work. There are blogs/websites which claim 1840 and some which claim 1820 when he was 5 years old, so these are self-evidently unreliable sources. Looks like this will require a university research library to find 19th century sources. Edison 13:59, 2 April 2007 (UTC)

Halogen lamp section should be separate article

The section on the halogen lamp should be its own article. There is already a page with that name created in 2002 as a redirect.

This would trim this lengthy article, improving readability and ease of use. A lot of relevant information from this article could also be transcribed over, creating more relevant article for those looking for information on halogen lamps.

Ng.j 15:28, 4 March 2007 (UTC)


Add  # '''Support'''  or  # '''Oppose'''  on a new line in the appropriate section followed by a brief explanation, then sign your opinion using ~~~~. Please remember that this survey is not a vote, and please provide an explanation for your recommendation.

Survey - in support of the move

  1. mild oppose -- There's too much commonality to factor out halogen lamps. -- Atlant 19:56, 2 April 2007 (UTC)
  1. support -- With a italizized embedded link to a separate article 04:48, 28 April 2007 (UTC)
  2. Oppose They are still incandescent lamps. They are not LED lamps, Fluorescent lamps, neon lamps, or any other type. It would make as much sense to break out separate articles for open-air platinum filament, carbon filament high vacuum, tungsten filament high vacuum, and tungsten filament gas filled lamps. These differ from each other about as much as halogen bulbs do from the others. Edison 06:04, 28 April 2007 (UTC)
  1. Support

Halogen contain bromine and iodine and other halogens not a vacuum in a regular incandescent bulb

  1. Support

Halogen are generally sold specifically as halogen, and function differently than regular incandescents. They are also banned many places where normal incandescents are not (for example, most college dormitories) 20:54, 15 June 2007 (UTC)


Add any additional comments:

Hey - this is my first post, so be gentle on me. I'm not quite sute how all this works. All I wanted to say was that Australia has not banned the incandescent light globe as stated in the introduction to the article. EJH69 07:38, 22 March 2007 (UTC)


Someone recently removed the brackets [ ] from the the convenience links to online sources. The practice I am familiar with is to put the brackets around the html link and then to have text outside the brackets identifying the source. With this, the weblink, which may be quite long, appears in the reference section as a colored number, and the verbal description appears as black text. Is there a convention to just have the weblink and the description of it between the <ref> and </ref> in the inline cite in the text? I reverted to the previous version pending discussion. Edison 23:54, 12 March 2007 (UTC)

Heat vs. efficiency

It is a shame that lighting efficiency is such a complex technical subject! It seems like the one thing people could most clearly agree on and understand is what percent of the energy supply turns directly into heat, so these numbers should be listed. But what ordinary folk most want to know is, what percent of the energy supply is emitted in the visible spectrum, from the light fixture as a whole. The complexities of adjusting that number to compensate for the sensitivity of the eye to various frequencies leads to numbers that should certainly be mentioned, but not as the only or even primary numbers here. (If a black box turns all of the electricity supplied into emitted orange light, to me that is 100% efficient. If another box turns all of the electricity supplied into emitted purple light, to me that is 100% efficient. If a scientist considers one to be more "efficient" than the other, because the eye sees purple better or worse than orange, that would be interesting to know, but would not change the fact that each is basically 100% efficient.)- 15:31, 4 April 2007 (UTC)

Start surge

Incandescent bulbs have a low resistance when cold, about one-tenth of the hot resistance. So, when they are turned on, they use about ten times as much power for an instant, until the filiment gets hot.- 15:31, 4 April 2007 (UTC)


Often, when an incandescent bulb is at the end of its life, it will keep working if left on and not disturbed. But after being turned off and cooling down, it will then fail immediately (sometimes with a flash) when turned on, due to the stresses of thermal shock and various changes caused by power cycling.- 15:31, 4 April 2007 (UTC)

It's actually the magnetic shock that occurs because of that 10x current surge on turn-on. Somewhere on the filament is a weak (evaporated thin) spot, and the mechanical shock caused by the electromagnetism ruptures the weak spot.
Atlant 15:38, 4 April 2007 (UTC)
Thermal stress at the thin spot may be a stronger factor than magnetic stress. There is not that strong of a magnetic field around a conductor in air with the amount of current applied to lightbulbs. Edison 18:04, 6 April 2007 (UTC)
Remember, though, at start-up time, we're talking about 10x the current so in a 120V system, so 3.3A (40W) to 8.3A (100W) flowing in a coiled coil (not a linear wire) and the wire is quite small in cross-section so not mechanically stiff. As a demo, I just tried a 60W clear lamp (in the usual US arrangement with no extra support wires for the filament) and firmly mounted so it wouldn't be vibrated by external motion. On power-up, the filament jumps several "diameters" of the coiled-coil filament and oscillates for several seconds afterwards. And, of course, tungsten is less ductile when cold so the mechanical stress is more likely to break the filament at that time.
Atlant 21:59, 6 April 2007 (UTC)
That is kind of fascinating... Some slo-mo closeup videos of bulbs failing that way at turn-on could be quite entertaining! But even such videos of good bulbs starting might be quite interesting, if there is really that much motion. Then, some college physics classes could try to construct math models, and if someone can come up with a model that matches the actual motion, we might start to have some confidence that the model was correct...- 19:31, 10 April 2007 (UTC)
I've been considering how one might capture such an image or video, but the enormous and sudden dynamic range of the brightness level between "light off" and "filament incandescing" would make it a challenging bit of photography. Maybe we need to call the Mythbusters guys ;-) ? Otherwise, maybe a really dark neutral density filter and just accept that the early few frames will be dark to black?
Atlant 22:17, 10 April 2007 (UTC)

You could take a short sequence with the correct exposure of filament off, then adjust to the correct exposure for filament on" and power it up. It would also be possible to illuminate the filament with another source (like a slide projector) and focus the light on the filament in the "dark" frames to reduce the brightness change. Some incandescent bulbs "sing" at a high frequence when on. The support locations probably help define the frequency. Edison 23:07, 10 April 2007 (UTC)

Politics Section

Has anyone any solid proof that incandescent bulbs have been banned anywhere in the world at all? The existing section lacks such evidence, and only talks about proposals and possibilites. Should the whole section be deleted or re-written? I think most if not all the proposals are political spin, and quite unreal. There are rumours that one CFL maker (Phillps in Holland) is trying to push politicians in europe to ban the bulb (!) so that it can reap vast profits from its CFL arm.Peterlewis 14:51, 6 April 2007 (UTC)

Phillips and GE would be competing with themselves, and over a 5 year period they would lose the repeat business for the incandescent bulbs that one CFL would replace, so they might make more money but there would be tradeoffs. This changeout is being treated as if it solved the world's energy problem, but we should reflect that if all incandescents had been banned and replaced overnight with CFL's in, say 2000, the continuing load growth would have eliminated all savings in a couple of years and we would still have a higher load today than in 2000. I wonder what the theft rate will be for CFLs in public places where they can be unscrewed and taken home. Actual fluorescent fixtures (without the screw socket common to incandescents and fluorescents) eliminate such easy theft. Banning screw-base fixtures in new construction as part of the building code would be a huge step forward. Edison 23:15, 10 April 2007 (UTC)
I suspect that the collapse in bulb prices has something to do with talk of "banning the bulb". Presumably all the cheap unbranded bulbs are made somewhere in the Far East (China?)and the big Western companies like GE and Philips are worried about their loss of business. Even CFLs are dropping in price, presumably for the same reason. So it is hardly worth the effort of consumers to change from incandescent to CFLs, especially as compact incandescent bulbs are now much smaller than the large CFLs. So is it worth stealng screw type or any other type from public buildings? The savings in electrcty are also smaller now wth falling electrcty prices in the UK. Peterlewis 04:31, 11 April 2007 (UTC)
From what I've read, the Australian government seems to be serious about it and has planned phasing them out in the foreseeable future (2010). I don't know if it has already been approved (signed into law, or whatever it takes there...) or it's still an initiative, though. Perhaps someone more knowledgeable about Australian politics could comment? --Itub 09:11, 11 April 2007 (UTC)
I saw a message from an Aussie on Wiki recently, which said that there is no ban on incandescents there at the moment. It seems to me that politicans are flying a kite to see what reaction occurs. Should Wikipedia simply repeat these absurd suggestions?? Peterlewis 09:53, 11 April 2007 (UTC)
Definitely we shouldn't claim that they have been "banned" without solid evidence, but we can mention that there are plans/discussions for phasing them out in several countries. There are many newspaper articles about it, and probably official government pages too, that can be cited. --Itub 11:02, 11 April 2007 (UTC)

Those alleged plans are simply spin and political hype. I have seen no substantial and well based government reports at all. All the current refs in the article are newspaper reports or based on environmental sites which have an agenda to pursue. Any government sites have very limited advice. Wki should try to keep a neutral base for its articles, and not simply push a political agenda. Peterlewis 12:41, 11 April 2007 (UTC)

This Australian government site says "Working with its state and territory counterparts, the Australian Government will gradually phase out all inefficient light bulbs and is aiming for full enforcement of new lighting standards legislation by 2009 to 2010. Special needs areas, such as medical lighting and oven lights, will be taken into consideration." Yes, we can't predict if the legislation will pass or not, but the fact is that the intent exists and is mentioned on an official site. I can't see why this shouldn't be mentioned in the article. --Itub 13:19, 11 April 2007 (UTC)
I have seen the site and its advice is very limited, and looks like the silly gov sites we often have in the UK, which are spin and not much more. It expresses hopes not promises, and Wiki should not be pushing government agendas. Governments frequently get their facts wrong, and then are forced to change policy very quickly. We have many recent examples n the UK and USA where ther govs have propagated lies, and they are frequently economical with the truth. Is ths how an encyclopedia should be run? Wiki should be providing hard facts, and CFLs have some unsavoury if not dangerous attributes. Their life is limited, and I have replaced several CFLs I used with incandescents owing to their short life and poor, harsh light emitted, and high cost. The latter are 100% efficient because the heat helps heat my home. —The preceding unsigned comment was added by Peterlewis 15:58, 11 April 2007 (UTC) Peterlewis 15:57, 11 April 2007 (UTC)Peterlewis (talkcontribs) 15:57, 11 April 2007 (UTC).
The latter are 100% efficient because the heat helps heat my home.
For someone who keeps pushing "life cycle" costs, you sure do turn your argument around when it's convenient. Electricity is not generated at anything near 100% efficiency, so even if all the electricity you're consuming becomes heat for your home, somewhere else there's a huge heat engine that's rejecting (probably) 60-65% of its input heat as waste heat as it makes your electricity. (Presuming you heat with something other than electricity,) you'd be far better off just to turn up the heat.
Atlant 16:11, 11 April 2007 (UTC)

It depends how you define efficiency. 100% of the electricity coming into my home for lighting ends up as useful energy. If you widen the system to include the power station, you will find that some power stations use the waste heat for space heating elsewhere. The Scandinavians and Russians are very good at using waste heat in CHP systems. Peterlewis 16:16, 11 April 2007 (UTC)

There is a fact, and that is that the Australian government is saying what is saying. We can report that. We cannot report what it will or will not do, of course. I suggest re-reading WP:V for the relationship between verifiability and truth as it concerns Wikipedia. --Itub 16:26, 11 April 2007 (UTC)

Government Views

I know it to be a fact that Germany after 1933 propagated the view that Jews could not be trusted. Are you seriously expecting Wiki to air such view (if Wiki existed in the 1930s)? You must be very naive if you expect citizens to believe what their governments tell them. Why should Wiki act as their mouthpiece? Wiki must stand by verifiable facts and not hide behind government statements or websites. If you must put your own POV, then you have to accept that others have a different viewpoint, and that their views need representation. Peterlewis 19:05, 11 April 2007 (UTC)

There is a huge difference between saying "this is a fact because the government says so", and merely reporting "the government says this". This has nothing to do with believing what the government says, or the government being always right. Please check Wikipedia's neutral point of view policy; the way of balancing an article is not to delete the points of view you don't agree with, but to add references to reliable sources that show the opposing viewpoint. However, saying that the government announced that they want to phase out incandescent light bulbs is not even a point of view! It is merely a statement of fact. Saying "light bulbs are evil" would be an example of a point of view. --Itub 07:49, 12 April 2007 (UTC)
I agree with your first point, but if Wiki puts gov views, then it should also put in refs to the problems. It should also point out that govs change their views frequently, and are not necessarily reliable sources, since elections may change policies when different parties take control. Wiki should also point out that statements are frequently made by politicians solely to "fly a kite" without any reference to bodies responsible for consumer safety (for example), or other public bodies. They are simply trying to attract attention to themselves, and to advance their own name. Peterlewis 09:21, 12 April 2007 (UTC)
I'm sorry, did you just compare the push to ban lightbulbs to genocide propagation? I think you did. In any case, Wikipedia is supposed to present facts. It should not point out that politicians often make false statements because a) that is unverifiable and b) unrrelated to this particular case. Furthermore, Wikipedia is not an advocate for consumer safety either. It has no duty to act as such. —The preceding unsigned comment was added by (talk) 16:00, 25 April 2007 (UTC).

Much of the introduction is about CFLs and an editors dislike for them

Why is much of the introduction to this page on Incandescent light bulbs talking about CFLs?

I realise they are mentioned in relation to proposed laws in a handful of jurisdictions to ban incandescent lights, but surely there info that is more relevant that should be placed there. Most of the last paragraph sounds to me like little more than an attack piece against those proposed laws than anything relevant to a factual article. Canderra 16:48, 30 April 2007 (UTC)

This should be split into another article about the banning of Incandescent lights. I don't see why people looking for information about the efficiency of incandescent lights (40w vs 60w) should have to look so hard to find that information. I was trying to make a decision based on the overall costs to see how much difference it would really make, and it was very hard! Benkenobi18 22:15, 8 November 2007 (UTC)
I've created a new article, Incandescent lightbulb bans, which felt like it wanted to be it's own article. All of the information there has been moved, further information on this topic can easily be added without overwhelming the incandescent light bulb article. Benkenobi18 22:58, 8 November 2007 (UTC)

4W/7W E12 candelabra nightlight

"Higher wattage bulbs tend to be more efficient than lower wattage ones. One reason for this is the fact that thicker filaments can be operated at a higher temperature, which is better for radiating visible light. Another reason is that since higher wattage bulbs would lead you to use fewer bulbs, you buy fewer bulbs and the cost of bulbs becomes less important. To optimize cost effectiveness in this case, the filaments are designed to run hotter to improve energy efficiency to reduce your electricity costs. Smaller bulbs use less electricity apiece, making the cost of the bulb more important. This is why lower wattage bulbs are often designed to last 1500 to a few thousand hours instead of 750 to 1000 hours. Designing the bulbs to last longer reduces their light output and energy efficiency." [19]

Please add lumens/watts for typical 4W/7W E12 candelabra nightlight bulbs. Explain why lower wattage bulbs are so much less efficient. Add good links to collections of real spec sheets...- 20:42, 15 May 2007 (UTC)

Split off halogen lamp?

There is a lot of content about halogen lamps - looks like it could be spun out to get this article down in size and not so rambling in content. --Wtshymanski 23:37, 18 May 2007 (UTC)

Done. Started, anyway. Some of the "science" of electic lighting could also move out of this article and into electric light. --Wtshymanski 21:56, 28 August 2007 (UTC)

output variation over service life

Are the output etc statistics given (and that appear on the labels in the stores) for a new bulb, or typical averages over the service life? Over the service life, how does the output (lumens) and efficiency (lumens per watt) vary, for ordinary 40-100W 120VAC bulbs? It seems like as the filament evaporates, the bulb would become more like a lower-wattage bulb, and have higher resistance, emit less light, and be less efficient in lm/W -- is this correct? How large is the typical change? Where is there more data about this? Eventually, if we want to compare different lighting technologies fairly, we need to be discussing typical average performance of the life of the device/product/system, not just "maximum when new" claims.- 11:20, 25 May 2007 (UTC)

  • Initial Lumens
  • Mean Lumens

"The average light output of a lamp over its rated life. Based on the shape of the lumen depreciation curve, for fluorescent and metal halide lamps, mean lumens are measured at 40% of rated lamp life. For mercury, high-pressure sodium and incandescent lamps, mean lumen ratings refer to lumens at 50% of rated lamp life (See Lumen Maintenance)."

  • Lumen Maintenance

"A measure of how well a lamp maintains its light output over time. It may be expressed numerically or as a graph of light output vs. time."

[20]- 13:17, 19 June 2007 (UTC)


(there are said to be about five billion light bulb sockets in North American households -- if we can pin down and source some such statistic, it should maybe be in the intro)

We need some photos of prism-separation of incandescent light.
These are interesting resources:

- 02:04, 26 May 2007 (UTC)

Bulb shapes, sizes, and terms

  • General Service (A)

Yellowish-white light emitted in all directions. Available in either clear or frosted. Types: General (A), Globe (G), Decorative (D) (flame, teardrop and other shapes)

  • Reflector (R)

Reflective coating inside the bulb directs light forward. Flood types (FL) spread light. Spot types (SP) concentrate the light. Reflector (R) bulbs put approximately double the amount of light (foot-candles) on the front central area as General Service (A) of same wattage.

  • Parabolic Aluminized Reflector (PAR)

Parabolic Aluminized Reflector (PAR) bulbs control light more precisely. They produce about four times the concentrated light of General Service (A), and are used in recessed and track lighting. Weatherproof casings are available for outdoor spot and flood fixtures. 120V (PAR) 16, 20, 30 and 38 bulbs: Available in numerous spot and flood beam spreads. Like all light bulbs, the number represents the diameter of the bulb in 1/8s of an inch. Therefore, a PAR 16 is 2" in diameter, a PAR 20 is 2.5" in diameter, and a PAR 38 is 4.75" in diameter.

  • Multifaceted Reflector (MR)

-- Halogen Bulbs are available in two distinct voltage types - 120V AC and 12V AC/DC. 12V MR16 and MR11 bulbs are available in multiple color temperatures.

  • High-Intensity Discharge (HID)

Types of Bulbs and Their Usage

  • HIR

"HIR" means that the bulb has a special coating that reflects infrared back onto the filament. Therefore, less heat escapes, so the filament burns hotter and more efficiently. [21]

[22] - 13:42, 19 June 2007 (UTC)

You say "Like all light bulbs, the number represents the diameter of the bulb in 1/8s of an inch" - that may be so in America but is it true in the rest of the world who use the metric system? Jim77742 09:18, 6 September 2007 (UTC)


I hope nobody objects to the reversion of the following line "During WWII, Hitler liked to burn jews and use their ashes to reduce oil usage." from of the California State University Network. Kesmet 21:33, 20 July 2007 (UTC)

Nobody would object except perhaps Revert this sort of stuff quickly, freely and without prejudice. And don't bother entering a Talk page item. Just do it! Jim77742 11:01, 6 September 2007 (UTC)

Practical light bulb

I reverted an edit which removed a reference which stated that Edison's was the first "practical" light bulb. Let's discuss what constitites an incandescent electric light, and what factors make one practical or not. The determination is not really up to an editor; it should be cited to reliable sources who say a bulb was or was not a practical incandescent light. As soon as the battery was invented, around 1800, experimenters noted that electric current of sufficient amperage would make metals such as platinum (favored for its high melting point) and the nonmetal carbon glow brightly. The arc light was demonstrated by Davy before 1810. When the thin carbon rods of the arc light were touched together to start the arc, they glowed brightly: they became literally incandescent before they were drawn apart and the aarc light was exhibited. If the conductor glowed brightly, it soon burned out, so it was a nice demo for a science class or to show a newspaper reporter, but not yet a practical light. A reporter could be impressed by a wire glowing red, but it would not be a satisfactory or practical light. If there were any oxygen left in the enclosure, it would soon burn out, and in any event the carbon from the heated rod would soon darken the glass. I say "rod" because generally early experimenters gave no thought to the need for a high resistance carbon filament, to allow powering a light from a central generating plant. They just connected a huge and expensive battery to a nearby conductor, made it glow for a few minutes, and voila. No one stood watching the demo 24 hours a day to notice that the glowing conductor only lasted perhaps one evening, like a candle wick made of platinum. Getting a patent on an incandescent electric light in the 1840's or 1850's does not constitute reducing the idea to practice. Swan and Edison in the late 1870's were the first to create incandescent lights that had any degree of practicality, as opposed to replicating the demonstration from the beginning of the 19th century that electric current makes a conductor glow for a while, until you increase the current to get near-white incandescence, soon after which it would burn out. Swan and Edison both have claims to fame in the development, but Edison accomplished a better product sooner, partly because of his greater research budget. See above under "Warren De la Rue's date is wrong?" for more references countering claims that earlier experimenters were the "true inventors." It is hard to document accomplishments of De la Rue because sometimes sources say De la Rive. Both experimented with electricity. Several experimenters were brought forward in patent litigation of the 1890's to make hard to substantiate claims of prior research (just as happened with the telephone). Edison 22:38, 5 September 2007 (UTC)

Sheep shearing

Whew. I've tried to remove some of the excess fuzz and lint from the article. Surprisingly, it only got about 4 kb shorter. I think it's better organized now - comments, please? But three hours at a time is about all I can stand. How long before it needs another clean-up? I really think the "ban the bulb" section should be sput out into a separate article if it needs to be represented at all; it's just a list of press releases at this point. We need more content - how about colored lamps? Rough-service bulbs? So-called long-life bulbs? --Wtshymanski 16:31, 20 October 2007 (UTC)

I have added some citations to GE Technical Publication TP-110. The efficiency values now in the table appear wrong, compared with this reference, and the original Web site is no longer available. --Wtshymanski 17:58, 26 October 2007 (UTC)

Moved text

Moved contents to Incandescent lamp, since this is the professional term of art for this type of lamp (and it is consistent with the names of other types of lamps). "Bulb" is a colloquial term, like "globe." I didn't know what to do about the discussion, so I left it. Agateller 12:10, 1 November 2007 (UTC)

Hold on a minute there... cat isn't under felis domesticatus. Let's talk about this before moving the article. I agree that "incandescent lamp" is the term used by many in the business, but "light bulb" is a very common term and is commonly recognized, at least on the west side of the Atlantic. --Wtshymanski 15:08, 1 November 2007 (UTC)
The Merriam Webster dictionary here in the office says "Light Bulb" and redirects to "N: Incandescent Lamp" but doesn't tag it as colloquial. Strictly speaking, a gas mantle is also an incandescent lamp so we should say "electric incandescent lamp". Googling shows about 5:1 in favor of either "light bulb" or "lightbulb" over "incandescent lamp". Seems they call them "light bulbs" in India and the UK, too, according to Google. "Light globe" gets about a quarter the hits of "incandescent lamp" and not surprisingly many of these are from Australia. Finally, ask youself - have you ever heard the joke about how many people it takes to change an electric incandescent lamp? --Wtshymanski 17:47, 1 November 2007 (UTC)

Note that pages may not be moved by cutting and pasting text from one to another. Pages must be moved with the "move" button. Nontrivial moves must be cleared first through the process described at Wikipedia:Requested moves.--Srleffler 17:49, 1 November 2007 (UTC)

International Wiki

but modern exit signs must use LEDs)

Fluorescent emergency exit signs are the standard equipment in most of the world.

Oh really? Reference? Not on my rather long-ago trips to Europe. It does say "modern". --Wtshymanski 16:38, 11 November 2007 (UTC)
Most safety codes now require halogen bulbs to be protected by a grid or grille, or by the glass and metal housing of the fixture to prevent ignition of draperies or flammable objects in contact with the lamp.

nope. With just 0.3 of the 6+ billion people on the planet, America is just 5% of the world population. Tabby 14:29, 11 November 2007 (UTC)

Well, that's sweet that you're trying to be so careful of the teeming hordes who only get to see light bulbs in the palace of the local dictator, but the United States generates considerably more than 5% of the world's electricity. Got a reference? "Safety codes" are mostly a North American/European concern, anyway. --Wtshymanski 16:38, 11 November 2007 (UTC)

Link To Other Lightbulb Types

I believe that there should be a link under the title along the lines of "Lightbulb redirects here. For Fluorescent bulb, click here. --HobbesDS (talk) 02:09, 7 January 2009 (UTC)