Talk:Light-emitting diode/Archive 3

Colors and materials
I like the "Colors and materials" section. But some of the numbers are just a little different than what I see elsewhere.

This article implies that green LEDs require "2.18 < ΔV < 4.0". When I look through supplier catalogs, I see many green LEDs rated at "1.9 V". So are those "1.9 V green LEDs" really more of a yellowish green, or are the numbers in this article a bit off, or am I misinterpreting something?

Should I pick a LED manufacturer at random and use their part catalog as a reference to update the voltages in that table? --68.0.124.33 (talk) 23:16, 5 August 2009 (UTC)


 * I think the best way is to convert the energy of the emitted light to a voltage ( wavelength -> eV -> Voltage ), because other factors contribute to the voltage of a real LED; how much current should be passed across it, Ohmic resistance and so forth. These values should be the minimum voltages and not the maximum rating from some manufacture. Perhaps it should say "minimum voltage" in the table--Thorseth (talk) 17:15, 6 August 2009 (UTC)


 * If I do that, I get 2.2 V as a minimum voltage (for 560 nm, about the longest wavelength that is called green). So either I goofed or the simple calculation isn't adequate for commercial green LEDs, and I don't know why that would be, but I'm not an expert.  --Ccrrccrr (talk) 02:48, 7 August 2009 (UTC)


 * Ok, I just did the reverse calculation for 1.9 volts, and it comes no way near green light. I also checked two manufacture pages stating between 2.5 and 3.5 forward voltage. So I'm guessing someone made a mistake. I can't think of a process that can generate green light with only 1.9 volts available, if it does I don't think its a LED --Thorseth (talk) 08:43, 7 August 2009 (UTC)


 * Looking at low-power green LEDs (where resistive drop would be less of an issue), which usually seem to be 569 nm, I see mostly 2.1 V listed as typical, at full rated current.  That's not as far off from the simple theory as 1.9 V is, but it does violate the principal that the voltage drop needs to be at least the photon energy. --Ccrrccrr (talk) 12:16, 8 August 2009 (UTC)


 * I agree with Thorseth that, in theory, one should be able to convert back and forth between color and "minimum voltage" from first principles, and we should list that voltage in this Wikipedia article. And if it was just one row in one hurriedly-printed catalog, then I would agree that it's probably a typo and forget about it. But I'm seeing over a dozen LEDs from several manufacturers that seem to be less than the theoretical minimum.
 * Here are examples of some allegedly "green" LEDs with 1.9 Vf(typ):
 * 572 nm green LED 1.9 Vf(typ) OSRAM LG3369
 * 565 nm green LED 1.9 Vf(typ) Panasonic LNJ312G8LRA
 * 569 nm green LED 1.9 Vf(typ) ROHM SML-P11MTT86
 * I would be happy to buy a few of these LEDs and send them to Ccrrccrr or Thorseth, if there is any way that would help give every single person on the planet the sum of all human knowledge. --68.0.124.33 (talk) 01:24, 12 August 2009 (UTC)


 * Thanks for the offer to buy LEDs. But as much as I love OR, I think for WP we should instead be considering what our reliable source is for this idea, that makes so much sense to from basic principles, that the minimum voltage should be equal the photon energy in eV.  Maybe it's not that simple, and a basic textbook would set us straight.--Ccrrccrr (talk) 21:10, 12 August 2009 (UTC)


 * You're right, buying some LEDs and measuring them would be original research. And the theory as I understand it seems to conflict with this data. I'm going to update the article with what published data I have, although I would prefer a reference that explains why that is. I hesitate to use theoretically-calculated numbers printed in a physics textbook as the sole source of data for this article, if it conflicts with the above data, because as you know, "If reality disagrees with theory, reality wins." -- DifferenceBetweenTheoryAndPractice. Although I've heard sometimes perception wins over reality ;-). --68.0.124.33 (talk) 19:34, 15 August 2009 (UTC)

High power LEDs
How is this or this considered an array and not a single diode? In this datasheet it even says "Over 2,750 lumens from a single chip" and again this one says "Over 2,200 lumens from a single chip" If Haitz's Law is expressed in terms of lumens per package then it hardly seems justifiable to differentiate in this section alone. We don't say a quad-core CPU is faster (but only because there is more than one core) or this generation of CPU is faster (but only because it has more transistors) and thus we shouldn't say a quad-die LED is brighter (but only because some excuse here). There are many ways to increase the lumen output of an LED and increasing the substrate surface area is one of them. Are we going to be calculating that surface area and arbitrarily creating new categories now because one manufacturer encased 3 die in a single package? What difference does it make if one die is larger or if you are using 4 smaller die? If you are going to qualify it this way you might as well qualify statements for the whole article by placing that verbiage at the top. Don't forget the RGB white section too as those are also single-package arrays. I agree it should be mentioned that some LEDs come in a single-package array but I would never say I have anything other than a single LED regardless of the die count or substrate surface area.Thepaan (talk) 00:25, 23 October 2009 (UTC)


 * You are right. The language needs a clean up, the Krames graph on Haitz's Law, I think is meant for single diode packages. Now hundreds of diodes can be fit on a "chip", however that can still not account for the exponential growth. There is a big difference between having multiple small diodes and on big one - the problems of making good quality dies increases dramatically with the size. It would be a good idea to distinguish between on-chip/arrays and other forms of packaging.--Thorseth (talk) 08:13, 23 October 2009 (UTC)


 * What "hundreds of diodes can be fit on a 'chip'" are you speaking to? The most I've seen (in a single package) is 12, again by LedEngin. We aren't talking about processors here, we are talking about LEDs. The CPU reference was to illustrate how rediculous it is to make up new categories. Because, that is what is being done here, you people are making things up.Thepaan (talk) 08:03, 5 November 2009 (UTC)


 * We should drop the LEDEngin reference, which includes single chip only up to about 250 lm. (Their higher power ones are in fact arrays (not just multiple diodes on a chip, but multiple chips, and they happily promote that.)  But Luminus truly does have single-chip, apparently single diode emitters up to over 1000 lm.  If you read the fine print, the 2000+ lm ratings are operating at higher current than you'd want to for most applications, due to degraded life and efficiency.  So I'd recommend changing the article text to "over 1000 lm" and citing specific single-chip products like you do above, so that people don't go trying to find what you were citing and instead find an array.  That way it is indisputable.
 * It's true that we are drawing an arbitrary line between arrays and single diodes. It's silly to criticize that as being arbitrary.  We have to draw a line somewhere, unless we want to add up the total lm output all all LEDs in the world, and wherever we draw that line, one could argue that is should be drawn somewhere a little different.  We should draw the line somewhere that is conceptually simple. Single-chip sounds like as good a place as any to draw the line.  And it still allows a claim of "over 1000 lm".  Ccrrccrr (talk) 12:57, 23 October 2009 (UTC)


 * This isn't your own little fantasy world. They are "High Powered" LEDs no matter what kind of make-believe you subscribe to. As such they should be included. And, here is another one. Thepaan (talk) 07:53, 5 November 2009 (UTC)


 * I'm not sure what you are lobbying for as far a a change to the article. Our goal here is to reach consensus on what the article should say.  Is there a problem with the text there now?Ccrrccrr (talk) 12:51, 5 November 2009 (UTC)

Follow Up Previous Discussions
I notice that Losev is still credited as the inventor. This was discussed here previously, and it appeared some rearrangement was anticipated. Also there doesn't seem to be any source cited for this info. (Never heard of him myself)

The bit about a bipolar transistor emitting light under certain condition, ie. cutting the case open, biasing etc. has been removed. This was also mentioned and it appeared it was staying, but 213.162.105.199 recently edited it out saying "they tried it and it didn't work". I have also heard that this is possible to do. But OR to remove something is surely as bad /worse than trying to publish your own OR. There may be a place for this info (LED trivia?). The fact that they tried it (once or more?) surely isnt right in Wikipedia.

This is a very detailed article, and even though I worked as a Radio 'Technical' Officer, I found the intro a bit too detailed. Now for 'conventional' current versus 'true' current flow!. [Kidding] ;-) --220.101.28.25 (talk) 17:11, 24 October 2009 (UTC)


 * There is a good article by Zheludev on Losev and I put his first paper and the Soviet patent in the refs. If the bipolar transistor thing was OR then that is reason enough to remove it. If a statement is unreferenced and seems to be incorrect then I see no reason to keep it. It seems the intro has deteriorated since I last looked at it, if you could be a little more specific then I will try to make it better soon. --Thorseth (talk) 13:29, 25 October 2009 (UTC)


 * In general unsourced info can stay in an article if it's general knowledge, unlikely to be challenged. If someone's OR refutes it, that bumps it out of the category of waiving the requirement for a source.  If there's a supposedly reliable source that makes a claim, and that's in the article with that reference, and OR refutes the reliable source, that can motivate a search for other POV in other reliable sources, but the info should stay in at least until we find a reliable source that refutes it.  So I agree with keeping the bipolar light emission out without reliable sourcing.  I'm not sure it's particularly important or relevant information anyway.Ccrrccrr (talk) 14:17, 25 October 2009 (UTC)

Cost by bulb type over 10 years
This graph is meaningless drivel. The only purpose I can see for it is to promote the agenda that "LED's are better". Why do I say this? --Juckto (talk) 20:47, 19 November 2009 (UTC)
 * 1) There are no units on the left-hand axis.
 * 2) * Is it $ per lumen, $ per lamp, $ per running hour, or what? Without proper units the data is meaningless.
 * 3) * Less critical, but still relevant, is the cost measured in $US, $UK, or what? If it's not measured in any currency, you may as well just relabel the axis as "Lots - Not so much - Pretty cheap"
 * 4) The high peak for LED at the beginning indicates an intial capital cost of buying the bulb (as mentioned in the main article), which then decays into a low running cost. Every bulb type suffers a higher capital cost. Only giving it to an LED is misrepresentation.
 * 5) There is no referencing for where the information comes from. Someone explain to me why CFL suffer a cost spike every third year in their cycle?


 * The last question is probably because they need replacement every 3 years (8 hours per day ~10k hour life). But I agree that it's undocumented OR and should be removed.Ccrrccrr (talk) 02:11, 20 November 2009 (UTC)

light pollution
It is "very important" to avoid light pollution, is a biased statement that doesn't belong in the LED article. The movement that is against light, has no scientific proof to back up their crazyness. —Preceding unsigned comment added by 93.182.185.88 (talk) 14:25, 4 December 2009 (UTC)


 * Removed the "pollution subsection" (it was unencyclopedical anyway). Materialscientist (talk) 02:16, 5 December 2009 (UTC)


 * I shortened it and got rid of the "very important" bit. Backing up to the big picture, there's a growing body of information about LEDs as sources for general lighting, more and more of which is getting added here.  It will become increasingly important to have another article, e.g. LED lighting as the main place for such information.  There was a discussion above, Talk:Light-emitting_diode, that noted this problem, and suggested a solution, but I guess we are all too busy fighting vandalism and so on to get to work on it.Ccrrccrr (talk) 23:45, 5 December 2009 (UTC)


 * It's pretty important to terrestrial astronomers.  ;)
 * JWhiteheadcc (talk) 18:34, 21 February 2010 (UTC)

pronunciation guid e
Why the hell doesn't the pronunciation guide apply to "Light-Emitting Diode" instead of just its abbreviation? You would think most readers would know how to pronounce the letters "LED". —Preceding unsigned comment added by 97.118.146.126 (talk) 06:00, 26 December 2009 (UTC)


 * I don't think "Light-emitting diode" is particularly ambiguous, whereas some people confuse LED as being pronounced as a word, rather than the individual letters. This is evident from edits made to the article in the past where it's been incorrectly referred to as "a LED" rather than "an LED".  Giftiger Wunsch   [TALK]  19:38, 18 September 2010 (UTC)

Vandalism
Urm, I new to this game, thought I'd do a quick edit to remove some vandalism. It's something to the order of "GLOBAL WARNING CAN ..." I went to remove it, however, and I can't find that text anywhere in the source for the page. From where is it coming? There are a number of occurrences of it. —Preceding unsigned comment added by 69.134.179.71 (talk) 02:05, 11 February 2010 (UTC)

Noise diode
'noise diode' should redirect to here: http://en.wikipedia.org/wiki/Hardware_random_number_generator#Physical_phenomena_without_quantum-random_properties How do I setup a link and leave the displayed link text the same? (Add a title to a link)

Specifically, the part where it says "Avalanche noise generated from an avalanche diode, or Zener breakdown noise from a reverse-biased zener diode." is relevant. It doesn't even really need to be a diode. Any gate or resistor will experience some random noise. Even a spark gap oscillator has some random noise in it's output. What matters though is that the definition of a noise diode is included in that article. JWhiteheadcc (talk) 18:59, 21 February 2010 (UTC)


 * "Hardware_random_number_generator#Physical_phenomena_without_quantum-random_properties|noise diode" inside double-brackets made it work. JWhiteheadcc (talk) 05:30, 21 October 2010 (UTC)

Drawing "Parts of an LED"
Anvil being Cathode, post being Anode is not always true. The company I work for produces LED-boards showing prices at fuel stations. While the rest of the circuitry on our boards exactly stay the same every now and then our THT-workers have to place the LEDS in the opposite direction. (De- and Re-soldering isn't an option, so guess who..., as a hobbyist, is benefiting from mistakes made in the past. ;) 62.238.198.129 (talk) 21:32, 15 March 2010 (UTC)

Narrow band light sensors
The section on Narrow band light sensors seems very long for something that has apparently never been used for anything practical - if it has this should perhaps be mentioned. Also the "journal entries" seem to contain very little encyclopedic information. Searching for "mims effect" and LED on google gives 15 hits, most of which leads back to copies of this section. I question whether most of this is notable.--Thorseth (talk) 20:50, 9 April 2010 (UTC)


 * I will remove most of the content of this section if nobody objects.--Thorseth (talk) 15:13, 13 May 2010 (UTC)

Traffic signal blue-green LEDs
The color table appears to lump all wavelengths from 500 to 570 nm as Green, and does not make any distinction with the between the traditional bright lime green LEDs and the newer Blue-Green traffic signal LEDs. Generally, Green is defined by the LED industry as 540-570 nm, while 500-540 nm is usually defined as Blue-Green. The standard wavelength for "green" traffic signals in most countries is 505 nm. However, I don't have any data on voltages, so I could use a little help. ANDROS1337  04:08, 20 April 2010 (UTC)

Narrow band light sensors
I removed that section for at least the following reasons:
 * WP:UNDUE: the section was written as a tribute to Forrest Mims, not as an encyclopedic entry on LED. That material is already present in the Forrest Mims article.
 * WP:RS and WP:NOTABILITY: the section did not site any reliable reference, and I could not find such. That an LED can act as a photodetector is a trivial fact known to anyone familiar with p-n junctions. It is also obvious that any dedicated p-n junction photodetector would outperform an equivalent LED. Search on Google books for LED sensor Forrest Mims returns only books by Mims himself and their echoes. Materialscientist (talk) 05:35, 11 June 2010 (UTC)

contact lenses?
I heard there are now LED contacts, can this be verified?--Chris (クリス • フィッチ) (talk) 13:57, 5 August 2010 (UTC)

Large gif image
I have a problem with downloading a 1.4MB image (220px-P-hus_Aveny.gif) over my mobile internet connection each time I visit this page. Can it be removed? 110.23.89.159 (talk) 01:26, 18 September 2010 (UTC)


 * Yes, it serves very little purpose as it is. Ill remove it.--Thorseth (talk) 19:31, 18 September 2010 (UTC)

SMD Leds
Anyone want to upload a pic of surface mount (device) led? —Preceding unsigned comment added by 115.70.80.179 (talk) 23:43, 5 January 2011 (UTC)
 * I'll see if I can take a decent picture of one if I get time tomorrow. I don't think I have a free SMD LED, but I'm fairly sure I have a PCB or two with one on which I could photograph. I'm surprised there isn't one in the image on the article with various different types of LEDs.  Giftiger Wunsch   [TALK]  00:08, 6 January 2011 (UTC)
 * Looks like I forgot; I'm unlikely to get time to do this in the especially near future, but next time I have a spare minute and remember about it I'll see if the article still needs one and make one if it does.  Giftiger Wunsch   [TALK]  23:49, 8 January 2011 (UTC)

Efficacy at low temperatures
A Google Books snippet out of "SAE Ground Vehicle Lighting Manual" says that LED luminous output can nearly double at -40°C. However, I don't like citing a snippet as a reference. --Wtshymanski (talk) 03:25, 15 April 2011 (UTC)

Efficacy vs efficiency
We want luminous efficacy in a light-emitting diode, not just efficiency. An IR LED that was 99% efficient would have zero efficacy as a visible light source. It is correct to speak of efficacy when describing visible light sources. --Wtshymanski (talk) 13:37, 13 April 2011 (UTC)
 * As LED's have been known for decades as having high efficiency and (when emitting in the visible spectrum) luminous efficacy, and the article gives considerable detail of the progress of luminous efficacy in the development of white light LEDs, I have deleted the spurious markup that labelled the claim of high efficacy as dubious. I don't know whether it was originally inserted due to ignorance or because an earlier form of the article did have something dubious. I also reworded the comment on the high efficacy, to link to the luminous efficacy article. I removed the phrase 'as measured by its light output per unit power input', superfluous in context, but added the word 'sources' to eliminate the possibility that we refer to luminous efficacy of radiation. (LED's offer high luminous efficacy in either sense of the term, but it is the efficacy as a source that offers advantages as lighting sources.)


 * By the way, the high luminous efficacy of a LED as such is not actually suitable for comparing with, say, an incandescent light bulb. The latter, though poor in efficiency and luminous efficacy, does run directly off the mains (or battery, etc), whereas a LED (whether powered by a dc source or from the mains) usually requires a power supply that also dissipates power. So the 'input power' for relevant comparisons must be the total power drawn from the power sources (mains, battery, etc) rather than just the input to the LEDs. This issue is discussed in the luminous efficacy article, where various alternative phrases are mentioned, such as wall-plug luminous efficacy, for being more explicit that the efficacy is calculated on the entire power consumed. One of these phrases is 'lighting efficiency', which would be much nicer if only we could be sure that everyone would agree on what that means. I doubt it.

--Alkhowarizmi (talk) 10:51, 2 August 2011 (UTC)

Inaccurate Diode curve description
It states in one of the graphs for the "diode curve" that the LED starts emitting light when the "on voltage" has been exceeded, and that the typical on voltage is 2 - 3 volts. This is not completely accurate, LEDS will start lighting up at forward voltages much lower than 2 - 3 volts, the 2 - 3 volts is the typical forward voltage with a typical current flow to allow it to emit light to a reasonably observable level. If an LED has a forward voltage of around 2V and a current flow of 20mA, it will light up even for around 10 micro-amps at 1.6V, though it would only be really observable at close proximity in a dark room. Veritas Blue (talk) 06:04, 3 September 2011 (UTC)


 * I guess you’re talking about the graph under LED. I agree it’s a bit misleading; maybe it would be better to avoid talking about light because the graph only covers current and voltage. My understanding is LED light is approximately proportional to the forward current, and the forward current of any diode is roughly an exponential of the forward voltage. A particular diode’s “on” voltage depends on the order of magnitude of the forward current in this “on” mode. Vadmium (talk) 06:51, 3 September 2011 (UTC).

Semiconductor "die"
Yeah, one of the pictures showing the parts of an LED has a messed up word. It says "die" and not dye. If someone can please fix the image and re-post it I would be very greatful. — Preceding unsigned comment added by 50.84.132.155 (talk) 15:09, 28 October 2011 (UTC)
 * "Die" is all around the article and it means die (a small block), not dye (paint). Materialscientist (talk) 23:58, 28 October 2011 (UTC)

Outdated information and questional section "Energy consumption"
The section Light-emitting_diode compares a 40 Watt light bulb with a 13 Watt LED. Nowadays, a good 6 Watt LED can easily emit 600 Lumen. Of course, there are also still older, less efficient models on the market. Further, the argument that "in cold climates [...] more heating is needed" is a rather questionable argument as heating by electricity is comparably inefficient and during summer highly undesired. Overall the heating effect is so low that you couldn't sufficiently heat a room with a conventional amount of light bulbs anyway - and wouldn't want to as there are numerous more efficient methods. Especially as heating and lighting are not separately controllable this is rather an excuse against using modern technology than anything else. I've often seen slightly burnt areas around Halogen and conventional light bulbs. Last but not least, the heat of light bulbs is often a fire hazard. I've never even heard about anyone trying to use light bulbs for heating - excluding special purpose devices like lava lamps - before the EU announced to ban them and then it was usually an obvious expression of stubbornness. --93.130.51.11 (talk) 13:10, 8 July 2011 (UTC)
 * Indented line FYI - Incandescent lamps have been used for food warming, incubators, personal heater (such as valets) and for cooking, such as the Easy Bake Oven. In lighting, discussions of which is more efficient in winter has been debated, since incandescent is an efficient heater. --Danpeddle (talk) 06:05, 12 November 2011 (UTC)
 * Electricity is almost always generated by heat - in coal, natural gas or nuclear power plants. You generate heat to generate electricity to generate heat. Converting electricity into heat may be very efficient but the vice versa conversion isn't. There's always loss. That is what makes heating with electricity inefficient. For example, using the heat generated as an otherwise unneeded by-product of industrial factories and heat-pipelines is much more efficient because you virtually get your heating for free. Similar goes for solar collectors, geothermal heat, organic gas etc. --77.176.243.47 (talk) 14:49, 13 January 2012 (UTC)

Layout & polarity 3528 SMD LEDs
On November 13th 2011, I created and uploaded a schematic showing the typical layout and polarity of a 3528 SMD LED. The image was removed on December 21st without much explanation given by Thorseth, who found "removed image of dubious information value" (his comment for the edit). I think polarity information of an SMD LED is not dubious (a reference was given to the handbook of a producer of SMD LEDS). Any arguments? Spidey71 (talk) 22:27, 22 January 2012 (UTC)
 * The arguments could go something like this: The image was placed in the lead. We would like that ... "The image helps to provide a visual association for the topic, and allows readers to quickly assess if they have arrived at the right page." From the style manual Images I do not think that this images does any of this. Further more, I do not find the polarity of the LED to be obvious from the ajacent diode symbol. I am also missing a scale, to see how big the thing is. The caption (which is placed inside the image??) states that it is a "Typical 3528 SMD LED layout" ... Does this mean that there are other layouts of this type of LED? This is difficult to find out, because this particular type is not mentioned in the main text. I not saying the image can't be used, just that I have some questions my self: Does this LED have a special signifigance or is it representative of other LEDs since it belongs in the lead? Why does it need two captions? Perhaps it could be used to illustrate a section on SMD components or similar?--Thorseth (talk) 06:28, 23 January 2012 (UTC)

Is this important enough to mention?
My original edit to Light-emitting diode was reverted by User:Wtshymanski. I found the edit summary to be incorrect in its characterization of my source (I saw nothing in the source about a press release, and "not from a publicity department" would be a good thing rather than a reason to revert an edit). So I reverted the revert and was reverted again. The edit summary this time was clearer on the Wtshymanski's objection--the event wasn't important enough, which may have been justified--but it seemed to indicate Wtshymanski objected to the reliability of the source as well. The source was a reporter who had worked at the newspaper for many, many years, so I asked on Wtshymanski's talk page what made him the authority to deem this reporter an unreliable source. I confess that I violated WP:CIVIL in my wording, but I felt perfectly justified in my edit and Wtshymanski seemed unreasonable. Wtshymanski reiterated the lack of importance of the event here but indicated willingness to add the edit to Cree Inc.. I indicated a willingness to compromise and Wtshymanski blanked his/her talk page. But the edit summary for Wtshymanski's response to me was "Wikipedia is not for press releases". I don't see where I've done anything wrong, and Wtshymanski seemed happy to let me put the edit in Cree Inc..

So I didn't use a press release as a source (though the reporter might have), and the only justified objection to the edit seems to be the importance of the event. Vchimpanzee ·  talk  ·  contributions  · 20:36, 10 February 2012 (UTC)


 * Copied from WP:VPP: Vchimpanzee ·  talk  ·  contributions  · 18:47, 11 February 2012 (UTC)


 * One of the sections of our verifiability policy is titled Exceptional claims require exceptional sources and is relevant here. Your edit states that Cree's new LED "delivers twice as much light for the same price", which I would qualify as an expentional claim. Accordingly you would need multiple high quality sources to back it up. The source you give is barely one step up from a press release, the guy is obviously just repeating what he heard from the PR guys. Yoenit (talk) 00:07, 11 February 2012 (UTC)

Entangled photons
There are LEDs now that produce entangled photon pairs, might want to add that... to wit: http://optics.org/article/42841 71.139.160.159 (talk) 05:25, 8 May 2012 (UTC)

Advantages and Disadvantages sections
As they currently stand, these sections don't make sense. One cannot speak of advantages without agreeing on the reference. For instance, "efficiency" features both as an advantage and a disadvantage. Moreover, it is implicit in other bullets, such as "color". I see two ways to solve this: Either group these by topic; e.g. all efficiency related considerations could form one section. I did a first step in that direction by merging toxicity consideration into "". Another path would be to have a table that compares different light sources. That would provide a frame of reference for such considerations as efficiency, which would be much more useful for the reader than the current conflicting statements. &mdash; Sebastian 02:33, 18 February 2011 (UTC)

This digresses slightly off topic, but the "" section needs a slight revision:

"* Light quality: Most cool-white LEDs have spectra that differ significantly from a black body radiator like the sun or an incandescent light."

The very next point in the section, Area light source:, links to an article on lambertian distribution:

"The emission of a Lambertian radiator does not depend upon the amount of incident radiation, but rather from radiation originating in the emitting body itself. For example, if the sun were a Lambertian radiator, one would expect to see a constant brightness across the entire solar disc. The fact that the sun exhibits limb darkening in the visible region illustrates that it is not a Lambertian radiator. A black body is an example of a Lambertian radiator."

The link in the Light quality: section, black body, clarifies further. Stars can be roughly modeled as black bodies, but are not.



I propose a simple revision: "* Light quality: Most cool-white LEDs have spectra that differ significantly from the sun or a black body radiator like an incandescent light."

&mdash; Ptericles Ptericles (talk) 20:56, 20 May 2012 (UTC)

Index of Refraction of Silicon may be wrong.
The article states that the index of refraction of silicon is 4.24. I think that's wrong: While the index of refraction of silicon varies based on wavelength, I believe that a value of 3.55 is close to being correct. Perhaps the value of 4.24 was intended to refer to one of the many non-silicon LED combinations. (GaAs, GaAsP, GaAlAsP, etc.) Jamesdbell8 (talk) 16:04, 14 July 2012 (UTC)
 * Fixed it, thank you!--Thorseth (talk) 07:14, 15 July 2012 (UTC)

Phosphor-based LEDs Efficiency
The Phosphor-based LEDs section of the article has this sentence: "The efficiency of a typical YAG-based yellow phosphor converted white LED ranges from 3 to 5 times the efficiency of the original blue LED." I don't see how this could possibly be true, The phosphor coated LEDs loose energy from the Stokes shift, they should be less efficient. Is this statement meant to compare the use of phosphor vs a filter? Does it take into consideration how sensitive the eye is too different frequencies of light? Wingedsubmariner (talk) 01:40, 20 July 2012 (UTC)
 * The only way that makes sense if it means "efficacy" (visible light (lumens)/watt) instead of "efficiency" (total radiation, watts/watt) - a monochromatic blue light will have a low efficacy,even if it's very good at changing electric power into blue light. Using the curve in Luminous efficacy, if the lamp was blue enough, a 5-fold increase looks possible to me. Probably wants a reference or deletion.  Stokes shift is where most of the energy is lost in modern fluorescent lamps, too. --Wtshymanski (talk) 16:44, 20 July 2012 (UTC)
 * Ok, I've gone ahead and deleted that sentence, since even if it means efficacy it seems silly to compare the bare blue LED to anything, since no one would use a blue LED for lighting. Data on the relative efficiency/efficacy of the phosphor-coated LEDs to CFL/RGB LED would be useful, but I haven't found any good references. Wingedsubmariner (talk) 01:22, 21 July 2012 (UTC)

Update needed?
I read that Baltimore and other cities are replacing existing light sources in street lamps with LEDs. Is this text still accurate: "LEDs powerful enough for room lighting are relatively expensive and require more precise current and heat management than compact fluorescent lamp sources of comparable output." Thanks -- Jo3sampl (talk) 20:19, 17 August 2012 (UTC)
 * Someday, yes. But right now $1.50 CFLs are putting out the same lumens as $30 LED, so as of 3rd Q 2012 it is still a valid observation. --Wtshymanski (talk) 20:40, 17 August 2012 (UTC)
 * Ah -- thanks! -- Jo3sampl (talk) 12:29, 18 August 2012 (UTC)

Internal resistivity
No mention at all about this issue. The direct/inverse resistivity is neglectibly small/huge but when groupping lots of LEDs in paralel and/or serial circuits resistivity can grow/drop to significant values. — Preceding unsigned comment added by Srelu (talk • contribs) 03:50, 29 August 2012 (UTC)

LED Failure
What happens when an LED fails or burns out does it open or short? Just Curious — Preceding unsigned comment added by 209.165.141.254 (talk) 21:04, 20 April 2012 (UTC)
 * They can fail either way, or sometimes they jsut get really dim - either there's less light emitted or the package turns yellow or clouds over. Probably worth some text here. --Wtshymanski (talk) 15:35, 25 April 2012 (UTC)


 * In my experience (probably not good enough for Wikipedia articles :), I can only think of diodes failing on my with a short circuit, mostly due to overcurrent or overheating. This includes LEDs with way too much current, the silicon diode used as a reverse power polarity shunt on a hard disk controller (despite blowing the diode casing apart), and zener diodes. Anecdotes aside, people do write that having LEDs go open circuit is a potential problem for connecting them in series, and I think I saw LEDs advertised somewhere that had integrated zener diodes to avoid this problem (at least until the zener went open-circuit). BTW did you see the link to List of LED failure modes? It briefly mentions short circuits but not open circuits. Vadmium (talk, contribs) 12:54, 27 April 2012 (UTC).


 * In my experience, an LED (yes, an LED, not a LED) always opens when it burns out. 100% of the time.  And I have burned out dozens of them.  LEDs that burn out from long-life in a stressful application start getting dimmer, and may start flickering as well.  Their zener voltage (and thus resistance) may start increasing at this point, until the LED stops lighting at all.  At this point, the LED will be open (or nearly so); but because of the zener effect, one may be able to make it conduct (and light) again by increasing the available voltage.  This is why on graphic equalizers that have an LED on each slider, the whole group of LEDs may start flickering if just one LED goes bad (wired in series).  LEDs that are forcefully burned out (from serious overcurrent), usually burn out because the little wire going from the anode to the chip sitting on the cathode vaporizes, thus disconnecting/opening the LED.  This opening behavior in LEDs is the opposite from switching (or rectifier) diodes, which always short when they fail; however, sometimes even these will open due to the fuse effect, if the current increases greatly when the device shorts.
 * 76.6.164.233 (talk) 13:07, 24 March 2013 (UTC)

Led lights
I would like to simply know which of the available lights offer 5600K, so they can be mixed with available sunlight. Probably the ones that approximate the blue part of the spectrum, but it seems unclear. — Preceding unsigned comment added by 68.175.88.187 (talk) 14:28, 28 December 2012 (UTC)


 * Wikipedia is not a product catalog. Jeh (talk) 21:45, 26 April 2013 (UTC)

Flashing when seen on camera
Today, my little brother's friend was pulling up to the house, he was parked in the driveway, maybe calling my brother, but we saw him on the camera. On the camera, his LED headlights looked like they were flashing. I thought it was a cop car or something, but my older brother came and saw and he said, no it's our little brother's friend. He tells me that this happens sometimes on cameras with LED lights on cars. I immediately suspected that if there is such a phenomenon, then it has something to do with the LEDs pulsing vs. camera's capture-frame-rate. With a little research, I found that this is a normal thing that happens due to pulse-width modulation. This was a surprise to me because I hadn't heard of LEDs ever being pulsed - I thought they were always just on or off. I don't have any references, I'm not some kind of visual/camera expert, but this all should be pretty obvious to anyone even slightly versed in physics/science. EdwinAmi (talk) 22:20, 12 July 2013 (UTC)EdwinAmi So anyway, I'm going to be noting this phenomenon next to the mention of pulse-width-modulation in "Advantages:dimming" — Preceding unsigned comment added by EdwinAmi (talk • contribs) 22:08, 12 July 2013 (UTC)

crediting Oleg Losev…
I understand the desire to credit poor Oleg Losev (who is said to have died at 38, during the terrible Siege of Leningrad).

However, in the absence of better information I would argue against crediting Losev as an "inventor" of the LED. Losev may have been a discoverer, but not an inventor. (I don't know whether the silicon carbide LED that he developed relied on the same physical principles; perhaps another Editor can comment on whether it was a diffused junction or a heterojunction device; or whether as a point-contact device it was based on different, if related, physics. Be that as it may…)  Unless Losev's work can be shown to have indirectly given rise to adoption of LEDs in the marketplace or as a necessary element of a product, I can't accept Losev as inventor.

In other words, if his work occurred in isolation, if his ideas did not strongly influence Holonyak or others whose contributions did lead to adoption, or if his work was on a different type of device altogether, then Losev should not be listed as inventor on Wikipedia. On the other hand, I have no difficulty identifying Losev's work as that of a discoverer of electroluminescence. However, an electroluminescent device is not an LED - which must exhibit a reasonable efficiency.

Incidentally, I am well-aware that good work was done in the U.S.S.R. during the pre-WW-II era. I personally studied the papers relating to the invention of the photomultiplier; which invention was claimed by some to have been effected by one Leonid Kubetsky and then was viewed by Vladimir Zworykin during a 1933 trip to the U.S.S.R. (his native land). In that case, Kubetsky would have been inventor since his work led to the adoption of the photomultiplier. However in this case I am not aware that there is any "straight line" that can be drawn between the Losev work and that of Holonyak. If I am not correct, I welcome an authoritative response demonstrating the straight-line connection. — Preceding unsigned comment added by Jabeles (talk • contribs) 23:26, 6 March 2013 (UTC)
 * A reliable source that credits Losev for the invention is present in the article .G_PViB (talk) 01:33, 11 April 2013 (UTC)
 * So why not H. J. Round?  Clearly predates Losev, and about as much practical impact. --Wtshymanski (talk) 01:43, 11 April 2013 (UTC)
 * I'm not opposed to crediting him as well. G_PViB (talk) 01:56, 11 April 2013 (UTC)
 * Sure, and Tesla. Hey, let's put in Thales of Miletus while we're at it. That would make this a Greek invention. But seriously, the device invented at GE is the light emitting diode, and all these mad laboratory experiements are just that...not practical devices. --Wtshymanski (talk) 13:25, 11 April 2013 (UTC)
 * As long as their work is deemed important enough by a reliable source. G_PViB (talk) 17:41, 12 April 2013 (UTC)
 * I'm sorry, I think this is ridiculous. Losev discovered the phenomenon that underlies the LED's operation, but he didn't invent anything like the modern device. To say he did is like saying that William Crookes invented the cathode ray oscilloscope, complete with triggered sweep. Same for Round. I have no problem identifying Loseg and Round as independent discoverers of the phenomenon, however. Jeh (talk) 21:04, 20 April 2013 (UTC)
 * Regarding your objection - you'll need to achieve a consensus to delete the contents which warrant the categories before deleting the categories themselves. G_PViB (talk) 08:33, 21 April 2013 (UTC)
 * A critical reading of the reference given, and related material, does not support the claim of "Russian invention" or "Soviet invention". Discovery of principle, yes, but there was no discernible contribution to the line of development that led to the device we today call "Light emitting diode". A category of "Russian discoveries" would be supportable, particularly if it was on an article entitled "electroluminescence" rather than "Light emitting diode". Jeh (talk) 16:45, 26 April 2013 (UTC)
 * I'd rather go with the opinion of the Nature Photonics editors board. G_PViB (talk) 16:47, 26 April 2013 (UTC)
 * What you would "rather go with" cannot override consensus. As far as I can see you are one person trying to add the "Russian inventions" category to a number of different articles, even creating otherwise-unnecessary redirect pages just so you can slap this category on them. As to this particular question, yes, Nature has their boards of editors, but WP has its editors as well. It is not the job of WP editors to blindly copy what we find, even in sources normally as reliable as Nature. That is why competence is required here. Jeh (talk) 17:01, 26 April 2013 (UTC)
 * Perhaps if User:Jabeles didn't write his post in such a patronizing tone we could have had a more constructive debate. I argue knowing well from my area of expertise (firearms) that a swarm of editors expressing a biased opinion do not make it more verifiable. G_PViB (talk) 17:42, 26 April 2013 (UTC)
 * I see no patronizing tone in Jabeles' post at all. It is reasonable, reasoned, patronizes no one, and is well grounded in WP policy and precedent. Regardless of whether s/he intended to be patronizing or not, you do not get to excuse yourself from WP's requirements (such as WP:CIVIL) for resolving content disputes just because you don't like the tone you think was presented by someone else. Labeling others' opinions as biased does not help you either. Jeh (talk) 21:56, 26 April 2013 (UTC)
 * I'm sticking to the principial Wikipedia policies. On the other hand we have a bunch of users who got upset when they were reminded that their opinion is worth squat. Do mind that by the way of priority of policies, if infringing on your inflated ego causes you to lose sight of the content dispute it is your problem first. (Personally, I think that if we get rid of the oversensitive people it would only benefit the project.) As for User:Jabeles - his tone does represent the complex phenomenon of latent Russophobia. G_PViB (talk) 01:06, 27 April 2013 (UTC)

I tried to edit it to show Losev as discoverer, and the other 2 inventors, but the wikicode is beyond me Tabby (talk) 18:03, 28 August 2013 (UTC)

Request for Discussion of in electronic articles
Please see the corresponding discussion thread at Wikipedia talk:WikiProject Electronics. Thanks! • Sbmeirow  •  Talk  • 23:34, 15 December 2013 (UTC)

Unreliable efficiency droop source removed
I removed the following source: Rensselaer Researchers Identify Cause of LED “Efficiency Droop”, Rensselaer, Mary L. Martialay, 30 July 2013 because the it appears to be a self-published article by a publicist for the institution doing the reported research. The publication lists no sources except for the study that discovered a cause of the effect but the statements the source is used to support are the date of the effect's discovery and the magnitude of the effect. I had originally removed the content as well but have since replaced it with a request for (reliable) citations. Joja lozzo  03:49, 17 December 2013 (UTC)

watt-per-watt
The derivation of radiant efficiency in section Led reads a little bit like original research. Its not necessarily a simple task to do the spectral calculations with the data from the datasheet and a simple divide and multiply method could give wrong results for the broad sources.--Thorseth (talk) 12:21, 31 December 2013 (UTC)

Picture of Die & Bond-wire
Just thought I'd upload this here, in case someone wants to use it. It's a close-up picture taken with my digital microscope of the actual PN junction and the bondwire leading over to the cathode.

/-\ urelius ♠ &#124;) ecimus  What'sup, dog? 05:24, 2 February 2014 (UTC)

smart lighting
''Light can be used to transmit broadband data, which is already implemented in IrDA standards using infrared LEDs. Because LEDs can cycle on and off millions of times per second, they can be wireless transmitters and access points for data transport.[144] Lasers can also be modulated in this manner.''

is using light to transmit data referred to as 'smart lighting'?Jonpatterns (talk) 16:04, 10 February 2014 (UTC)

I don't believe that it is. I have done a lot of reading on Li-Fi and don't think it has been referred to as that.--Wyn.junior (talk) 00:16, 22 February 2014 (UTC)

The Health Dangers of LED Lights
While it’s well known that natural light is preferable to artificial for a variety of reasons (vitamin-D production, prevention of seasonal affective disorder), new research published in the Journal of Environmental Management shows that nighttime exposure to certain types of artificial light has an even darker side than previously understood. In particular, it suppresses the body’s ability to make melatonin, the hormone that helps regulate sleep and is celebrated for its antioxidant, mood-enhancing and cancer-fighting properties.

The main culprit is artificial light that contains the highest percentage of blue light in its full-spectrum mix. One of the top offenders in this category is the light-emitting diode (LED) bulb, which suppresses melatonin at rates five times greater than bulbs that give off warmer “orange-yellow” light, like incandescents.

LEDs have grown increasingly popular as an environmentally friendly alternative to fluorescent or incandescent bulbs. They contain no mercury, last more than 50,000 hours and use up to 80 percent less energy than traditional incandescent bulbs.

Despite these benefits, scientists encourage consumers to choose bulbs situationally, avoiding LEDs at night. Melatonin-suppressing light is “dangerous only if we expose ourselves to it during the hours when we should be in the dark, and if the exposure is sufficiently intense or long,” says physicist Fabio Falchi, of the Light Pollution Science and Technology Institute in Italy. He advises people to rely more on incandescent light after dark, especially in the bedroom. http://experiencelife.com/newsflashes/the-health-dangers-of-led-lights/ http://www.ledsmagazine.com/articles/2010/11/light-and-human-health-led-risks-highlighted.html — Preceding unsigned comment added by Nmakh (talk • contribs) 06:03, 23 April 2014 (UTC)

Another advantage: less insect deaths
LED street lamps are attracting less insects than incandescent ones. This is because they don't emit UV wavelengths. UV is the main reason why insects are attracted by lamps. (Insect traps often work with UV lamps.) The advantage is not only less insect deaths, which is certainly good for the eco system and all animals that prey on insects, it also means that LED street lamps are less polluted by dead insects and spider webs. Which is good for us because the dirt on the lamps absorbs light.--TeakHoken213.150.232.3 (talk) 13:34, 9 September 2014 (UTC)

Section deleted should be included
I wrote this section and I believe it should be included: LEDs can be used as microphones and can transmit audio over 300 meters away. Newark Liberty International Airport has 171 light fixtures being used as listening devices for security. The lights were manufactured by Sensity Systems and the company says that the potential of LEDs to collect data on the people is nearly boundless.

The user who deleted it said that any type of lighting device could be used as an audio listening device.--Wyn.junior (talk) 02:20, 22 February 2014 (UTC)


 * Please sign your post. I'll respond after that. Jeh (talk) 01:21, 22 February 2014 (UTC)


 * (Thank you.)


 * The referenced article does NOT support the editor's claim that "LEDs can be used as microphones." (Like most diodes, LEDs are only very slightly microphonic.) It does say that monitoring devices (it does not say "listening") are being built into LED-lit lighting fixtures. There is nothing about an LED lighting fixture that makes it any more suitable for concealing a monitoring device than any other sort of lighting fixture—or, for that matter, a painting hung on a wall, or a potted plant, or a trash can.


 * That some company is building light fixtures that happen to both a) use LED lamps and b) include some sort of monitoring devices (never claimed, by the way, to be sound pickup) is not germane to this article. Nor did the company rep interviewed in the video say anything about the LEDs being used to transmit the collected data (a la IrDA or similar), so that claim is unsupported too.


 * The editor misquoted the reference a second time: the company representative never said "the potential of LEDs to collect data on the people is nearly boundless", rather that "the potential for the advanced lighting is nearly boundless." And the video interview with that representative never mentioned listening at all!


 * The referenced page does link to another page with a completely unsupported claim about "one modified LED bulb that will send audio over 300 meters." Well, sure. There was a project like this in Popular Electronics decades ago. (It didn't use an LED, though, just an incandescent bulb.) The LED is still not being claimed to be the actual microphone. If it did, then as someone who has been working with both LEDs and microphones for literally decades, I would find that claim to be completely specious unless some better references were found.


 * I repeat: An LED is simply not very microphonic. If you want to build a listening device, you'd best start with an actual microphone. These can be so small as to be virtually unnoticeable: a little smaller than a typical T1-3/4 LED. You could modulate the LED and use that to send the audio somewhare, but the claim that "LEDs can be used as microphones" is not supported, and in this editor's opinion, is ridiculous on its face. Jeh (talk) 03:12, 22 February 2014 (UTC)

At the very least, a simple mention on the LED article of Newark Liberty International Airport's security system still seems appropriate.--Wyn.junior (talk) 03:55, 22 February 2014 (UTC)

It might also be appropriate to note that LEDs are not very microphonic and that these devices can be ambedded in any electronic device (if the info can be referenced). Both of those notes are very significant and appropriate.--Wyn.junior (talk) 03:57, 22 February 2014 (UTC)


 * Sorry but you need to counter the points I made, not just come back with "well I think it should." I could certainly see mentioning the EWR system in the Surveillance article. But in the video interview with the company rep there was not one mention of any way that the LED light sources contributed to the surveillance function of the fixtures... so what is the relevance here?


 * These lighting fixtures no doubt employ plastic; should we mention them in the "Plastics" article too?


 * As for the note that LEDs are not very microphonic, first things first: you need to make a case that a mention to these monitoring devices belongs here. You haven't done that. After that... there is no claim in the article you linked to nor in the video interview therein that the EWR system is picking up sound, so "LEDs are not very microphonic" would be irrelevant, even puzzling. And to anyone in the field it is also a "the sky is blue" claim, not really needing a reference. Our own article on Microphonics details the components that typically are microphonic; diodes, light-emitting or otherwise, are not among them. There really isn't any doubt or ambiguity about it. Jeh (talk) 05:20, 22 February 2014 (UTC)


 * To Wyn.junior, please do not confuse an "LED" with a "light fixture". A light fixture consists of many parts that can include LEDs, power conversion circuitry, and perhaps other things like a microphone. Be aware that sloppy editing in some articles might use the term "LED" to refer to larger assembly or fixture instead of the discrete component. Rich S 10001 (talk) 18:06, 5 October 2014 (UTC)

LED TV's
The increased lifespan of LED's in appliances should be mentioned, along with some sidelines on it. For example, TV's are now made with LED's and manufacturers claim a huge increase in life span due to the longer durability of the LED over the cathode ray tube (CRT). Appearantly though, this is all but a sales speech since they now manufacture the other components in the TV to break down much quicker. Appearantly, a comparable TV then and now is about 66% less durable, so I doubt that if you compare an old -well manufactured CRT TV- with a new LED TV, the LED TV will actually even last as long as the old CRT. See Planned_obsolescence

Mention at LED#Lifetime_and_failure KVDP (talk) 08:50, 23 July 2014 (UTC)


 * Going to need some extra-reliable sources for such claims. The CRTs in TV sets were almost never the primary point of failure or the limiting factor on lifetime. Yes, CRTs did go bad and needed to be replaced now and then, but failures in other components were far more common. Nor do we write WP articles based on what is "apparently" true (note correct spelling) And "about 66% less durable"? Where did you find that number? Are you sure it isn't 65%? Jeh (talk) 10:05, 23 July 2014 (UTC)


 * AFAIK, most consumer-grade LED TVs use organic LEDs. For them, it is just the opposite. While the typical lifespan of anorganic LEDs is magnitudes greater than that of CRTs or LCDs, the lifespan of OLED displays is considerably lower. It is typically estimated to be between 1 and 4 years, depending on various conditions. I don't have sources at hand right now, but I have seen several (including some studies) in the past. These short lifespans are not considered a problem by manufacturers, because the devices incorporating these OLEDs are considered "outdated" and "obsolete" even earlier (not my opinion). --Matthiaspaul (talk) 11:55, 23 July 2014 (UTC)


 * First, inorganic LEDs are used in place of cold-cathode-fluorescent-lamps (CCFLs) in the backlight assembly of LCD TVs. CRT TVs are not really the issue. CCFLs are reasonably efficient, effective and long-lasting as a light source, but they do contain a small amount of mercury. LEDs have challenged CCFLs because they are even more efficient, longer-lasting, low enough in price so the cost-add to consumer is not enormous, and backlight assembly designs have been well adapted to efficient use of LEDs. Second, organic LED (OLED) TVs are a more recent display type. They are emmisive displays, like CRTs and plasma TVs, where the pixellated image comes from glowing dots. LCD TVs (with either CCFL or LED backlights) are transmissive displays -- the pixellated image comes from liquid-crystal cells of variable light density, and a color-filter matrix.Rich S 10001 (talk) 18:22, 5 October 2014 (UTC)

Layout
There are too many single-sentence paragraphs and one-paragraph sections in this article. See WP:LAYOUT. Headings are for broad sections of text, not to separate each paragraph. —Designate (talk) 17:37, 11 October 2014 (UTC)

light emmiting diode
A light emitting diode (LED) is essentially a PN junction opto-semiconductor that emits a monochromatic (single color) light when operated in a forward biased direction. LEDs convert electrical energy into light energy. They are frequently used as "pilot" lights in electronic appliances to indicate whether the circuit is closed or not.

About LEDs (1/2) The most important part of a light emitting diode (LED) is the semi-conductor chip located in the center of the bulb as shown at the right. The chip has two regions separated by a junction. The p region is dominated by positive electric charges, and the n region is dominated by negative electric charges. The junction acts as a barrier to the flow of electrons between the p and the n regions. Only when sufficient voltage is applied to the semi-conductor chip, can the current flow, and the electrons cross the junction into the p region. 12:34, 26 October 2014 (UTC)117.199.213.98 (talk)

== This really is actually fascinating, That you are a quite specialist blogger. Ive joined your rss feed and sit up for looking for more of your excellent post. Also, I have shared your web site in my social networks! ==

Wow, great article.Much thanks again. Keep writing. — Preceding unsigned comment added by 188.26.155.102 (talk) 16:08, 27 November 2014 (UTC)

Cost timeline
"Until 1968, ...LEDs were extremely costly, in the order of US$200 per unit"

Prices were dropping in early 1967:

"Allied Radio's catalog No. 670 offers a GE type LED 9 light emitting diode for only $12"

Popular Electronics, January, 1967, pg 77

http://www.americanradiohistory.com/Popular-Electronics-Guide.htm

PRR (talk) 04:20, 15 December 2014 (UTC)