The LED's Dark Secret (2009)
spectrum.ieee.org
spectrum.ieee.org
Counterpoint:
"Incandescent lighting will never make a come back against solid state until it can overcome the mysterious maladies known as 'the fizzle and the pop', and its voracious power consumption."
(We shouldn't forget fluorescent, by the way. And that's not solid-state; it's tubes. In residential lighting, the competition now is really only between CFL and LED.)
Methinks if CFLs were promoted by one political party instead of another as a tactic for a particular sociopolitical strategy, we'd be hearing loud denunciations of promoting a product which uses a toxic & environmentally-unfriendly gas encased in a fragile housing.
If we are politicizing light bulbs and global climate, have we gone too far?
You better save your original receipts carefully. What if you get something that is supposed to last 25,000 hours but in reality it only lasts 15,000? I can imagine getting a refund, even a partial one, will be difficult, even with the original receipt.
Firstly how do you prove it was only used 15,000 hours? It depends on how many hours it was lit per day. How do you prove that the bulb you have in your hand is the one for which the receipt was issued? A newer receipt could be used to return an old identical item.
Is the store even around where you bought it? It's 30 years later now (so at least your receipt is credible, by the date printed on it), but you have nowhere to go. You only used that bulb 10,000 hours in its deployment location, by your calculations, and it turned into toast. You have to turn to the manufacturer for a refund. (A good argument for sticking to name brand LEDs like Phillips.)
Many LED bulbs claim outrageously long lifetimes, but as of now, this has not actually been tested; no deployed LEDs are anywhere near that many hours old. Basically, it's a matter of faith. People are gambling that if they pay more for and LED, it really will last as long as it claims.
(Don't get me wrong; I use LEDs. I have eight LED flood-lights, one LED room light, and two LED range hood lights! I just wonder about the longevity. I write the installation date on every bulb and try to save the receipts.)
In any event it doesn't matter whether or not the actual life testing has been done. Product manufacturers do accelerated life testing to estimate the lifetime of the product. It's similar to asking how they know that a new car shouldn't need warranty service for 6 years when it's only just been released.
Of the bulbs purchased by my parents in those last days, so far about half don't make it to the 6 month mark. I'd say that the companies manufacturing those bulbs offloaded their throw-outs on EU consumers right then and there.
See also: the reason why poor people tend not to shop at Costco.
I tried CFLs, but this is Minnesota. And when it's -25F outside, there is essentially no light out of those things :-(
Manufacturers should definitely change their stated lifetimes if they turn out to be incorrect, but even if their stated figures turned out to be 50% too high I wouldn't be concerned. You're still getting many years of service per lamp.
Maybe the vendors are banking precisely on your thinking; that people in the future will lose their receipts, forget the dates, or consider $12 from ten years ago a trifle matter that isn't worth their time and so on, and yet that people can be conned into coughing up more cash today.
By the way, how long an LED lamp will last depends on multiple factors. It's not just the silicon itself, but all of the components in the complex power supply that is integrated into it. It could be zapped by a power surge. (Do these things incorporate TVP's?)
Notice how no electronic appliance that you can buy claims to be good for 25,000 hours of continuous use. No smartphone, no television, no home theatre receiver, no computer. Typically you get a 1 year from the date of purchase, whether it's used or not and you can pay extra for more coverage. One year gives you a little over 8000 hours if you fire it up right away and leave it on continuously.
I'm surprised they didn't want them back for failure analysis, but perhaps they have plenty already.
LED lights are now much better.
http://www.nobelprize.org/nobel_prizes/physics/laureates/201...
Thanks for the explanation. I've been learning a little about light sources; based on my limited knowledge, I wouldn't expect that the full spectrum (or spectral power distribution (SPD)) of daylight could be replicated by hacking together LEDs and phosphors, though perhaps the untrained eye would not be cognizant of the difference. Can they really pull off 'daylight'?
The other thing they have going for them is that almost every film now comes under the hand of a colorist who adjusts the elements of every scene to favor a particular color palette - ideally one that was chosen in advance of photography and that is supported by lighting, costume, and production design. A tightly constrained color palette provides thematic unity and makes a big contribution to a film's 'atmosphere,' and generally takes precedence over photographic realism. The extreme example of this is making a film in black-and-white for aesthetic reasons; although that's a bit extreme for most mainstream audiences, many popular films are almost monochromatic with a strong tint. This creates visual tension and where the plot revolves pretty clearly around heroes and villains, strong visual contrasts underline the moral contrasts of the narrative. Comedies generally have broader colors palettes and colors are more saturated, reflecting the dramatically unpredictable and chaotic interactions of the characters.
Anyway, since shooting features on digital media became practical, and high quality digital scans of film stocks became affordable, filmmakers have exerted more and more control over the color of a film Because They Can, and the increasingly minor deficienices of LEDs are that much more likely to be smoothed away in post anyway because the white balance of a scene can be so easily manipulated in the digital realm.
At the current levels needed for general lighting, droop
kicks in, and down you go, below 100 lm/W.
In other words, at their worst efficiency they still kick incandescents out of the park, and are competitive with fluorescents without the toxic chemicals. Wake me up when they solve the problem, but don't expect me to panic in the meantime.None of which put out a lot of power, but in parallel gave the light that was needed.
in 2009, they were trying to replace a lightbulb with a single led.
Only issue now is consistency in group replacements. While I'm replacing independent bulbs as they burn out, things like a vanity with 12 bulbs means dropping some $150 at once to replace all the bulbs so they will all be exactly the same model. Styling & performance is still changing rapidly enough that occasional replacement of a burnout in a group will result in inconsistent bulbs, looking sloppy.
Each theory has its champions. Theoreticians at Philipps-Universität Marburg support Auger recombination [...]. Meanwhile, Hadis Morkoç’s group at Virginia Commonwealth University seconds Schubert’s support of electron leakage [...].
Well, the consensus has progressed quite a bit since: In recent years there's been increasingly convincing evidence that Auger recombination is the dominant effect behind droop after all (see for instance [1]).
[1] http://spie.org/x108666.xml
[/Spoilers]
If 16 lm/W is 10% efficiency, why isn't 300 lm/W an impossible 187.5% efficiency?
Lumens, as a unit, are normalized to the sensitivity of human cone cells, so the same amount of radiant flux in the green part of the spectrum is “more luminous” than in the red or violet parts of the spectrum, and infrared doesn’t count at all.
As a result, I’m not sure there’s any obvious way to measure an efficiency percentage in terms of lumens/watt.