That 60W-equivalent LED: What no one will tell you
edn.com
edn.com
Also, isn't the phrase "What you don't know, and no one will tell you" oxymoronic? I suppose "what you don't know, and only I will tell you" doesn't have the same ring to it.
Seriously though, this is pretty cool... I am fairly surprised that there hasn't been a DC standard for lighting fixtures come into play yet... Which could be much more effective in terms of cost over time, and temperature control. Given the shear number of wall warts, and other fixtures common in the home, it would make sense for a more common interface for DC (especially low watt)... USB seems to be becoming that standard, but for lighting, and safety for internal wiring a new standard is probably needed (or an old one revisited).
AC is nice because it's easy to convert to other voltages. With DC you'd need either the exact voltage - or a converter, and if you have a converter you might as well just go directly from AC.
Added bonus you get ethernet too so any devices like LED lights, cameras, etc can be programmable without wifi/zigbee/etc.
You would still need a DC/DC converter, and if you do that you might as well just go from AC.
PoE is helpful certainly, but not as a way of having universal DC in the house.
You are thinking wall-wart type applications. I'm more along the lines of infrastructure, lights and other permanently installed devices that can be designed from the ground-up to operate at 44v.
Why? What's the benefit over 110v?
edit: Here it is: http://www.youtube.com/watch?v=m5rM7QDi_5E&t=11m39s
It's a cheap trick of appearing honest by using impossible hyperbole. The meaning of the statement is so obviously not equal to its literal meaning that if the writer is challenged over the honesty of it he could plausibly defend it by saying, "It feels like this fact is not being communicated sufficiently to consumers who are not familiar with LED lighting." In other words, it's a way to manufacture compelling copy without needing any facts to back it up.
"And no one will tell you" also seals it for the future.
Besides my personal annoyance though, this kind of unbalanced and unrigorous article is just flat out dangerous to public discourse on a pretty important issue. LED bulbs have strengths and weakness, to be sure. I didn't think the heat issue was a "secret" that "nobody will tell me". It was the first thing mentioned in nearly every single article or story I've seen about LED bulbs. I've been using LED bulbs for about 4 years now (in regular fixtures, closed fixtures, outdoor fixtures in summer and winter, closed bathroom fixtures) and have not had a single issue. As others mention, even if I did there's good warranty coverage.
I'm probably more radical than most on energy issues, but I find this kind of FUD downright irresponsible.
It also is part of this annoying trend of people that really enjoy trying to poke holes in new technology. Like electrocuting elephants with evil electric distribution technology.
Not to mention that the incremental efficiency gain is negligible. Incandescent to CFL is a huge efficiency gain, so you can actually pay back the increased cost in electricity usage in a short while. A 60W equivalent CFL might consume ~13W. A similar LED bulb will be around ~9W. That extra 4W is such a marginal improvement that it would take decades to make it back in energy cost. So the only selling point is lifetime, and I've seen no indication that there is a big difference between the two -- both have similar types of failures, usually control electronics or power supply passives due to poor quality manufacturing.
Isn't that enough? I try to keep things that can kill me to a minimum.
Disclaimer: I'm sort of a chemist, so let's just say I've been around far, far worse than a few milligrams of mercury. I do have a healthy respect for safety hazards, and am very cautious, but I don't have a strong immediate inclination to avoid small amount of mercury (or DMF, or THF, or paint thinner, or...) outright as too risky.
When I read how to clean it up it sounded so scary I now stear clear of the whole room :-(
FUD? Seems like common sense to me.
If something comes with a heat sink, it's to disperse heat. If you're dispersing heat, you need airflow.
I know this. You know this. But the average consumer does not.
Consumers should be aware of this and this article accomplishes that.
Please keep any fanatical environmentalism to a minimum. Nobody is saying LEDs are bad. I own quite a few. It is, however, important to know how to use them effectively.
I'm sure that there are people out there who don't know this, but it also seems like it's approaching common knowledge.
I have a feeling this guy's overstating the problem. Perhaps because he's in the business of patenting thermal protection designs for bulb manufacturers.
Those Cree/Home Depot bulbs were brand new this spring (2013). If you've got LED bulbs that have been going for two years, they're probably something else.
http://www.cree.com/News-and-Events/Cree-News/Press-Releases...
I am trying them out in some enclosed fixtures, too early to tell how that will work out.
I have two in tulip style fixtures where the bulbs point down, they don't last long but receipts are a wonderful thing. Eventually I think the big box store I get them from will change their return rules
After putting my google-fu to use, I realized it was the plastic casing of a small light fixture in the staircase that was getting very hot and slowly melting around the incandescent bulb.
I've replaced the bulb with an LED bulb several months ago and I've checked the base of the socket, the fixture, and the bulb after hours of use, and all three are barely warm to the touch at all.
I suppose it may depend on manufacturing quality, or as the OP article mentions, airflow.
http://store.earthled.com/collections/led-light-bulbs-suitab...
Perhaps you could argue that it requires cycling because the cooling causes increased axial stress and gradually pulls the filament apart, but it is most certainly more complicated than just thermal shock. I believe that so long as you're not shattering glass, it's more about how long it's been hot so far and how many times it's been completely cooled.
Source: I was a lighting engineer at a major Independent Testing Laboratory for 5 years before going to software full time.
To get the Energy Star rating, the manufacturer must get an independent lab to measure their operation at 45C and measure the temperature on the die (per the LED manufacturer's specs), to validate that it will maintain it's life expectancy.
http://www.energystar.gov/ia/partners/manuf_res/downloads/In...
If you want to read all the nitty gritty details.
In an environment with no airflow?
6,000 hours? Where does that come from?
The L-Prize bulb claims 30K hours; most of the LED lights I've seen for sale claim lifespans in the 25K to 50K hour range. Are you saying that 6K is all we can actually expect?
To actually get just 6K hours from a bulb that cost you US$30 or more (the L-Prize bulb is $70!) would be very disappointing.
Until 2000, buying a lightbulb was an unconscious act. You go to any store, and buy a fucking lightbulb. 80% of sales were 60w. Some applications call for different bulbs.
Now, solutions for getting light out of edison sockets take up A FULL AISLE in Home Depot, with a dizzying array of choices that are difficult to evaluate.
You can buy normal lightbulbs, except that now most of them are mislabeled -- a 60W bulb may actually be 50W. 100W bulbs are impossible to find and cost $5/ea.
You can buy CFLs. Some are great, others make your living room look like a gas station bathroom in 1960. Most suck, and as a layman, I now have to know what color temperature is. And, oh yeah, many bulbs, which cost 5x-10x the original bulbs, burn out quickly in fixtures, or burn out quickly when in any orientation other than a standard lamp. (ie. put them in sideways, or mount them upside down)
Now you have LEDs, which sometimes have better light quality, but they cost $20. And while they may last longer, you need to be an engineer to evaluate the appropriate bulb for the appropriate application.
Just let me buy a bulb and pay more for electricity.
So that damn government meddling has resulted in too much choice in the free market?
The government decided that modifying the electricity demand curve was more important than me having light that meets my needs when the sun goes down.
A free market approach would allow me to make decisions based on cost and other factors. For example, my basement and garage are lit by efficient CFL and very expensive LED lighting that was installed when the house was purchased several years ago.
In my office, I find a standard incandescent bulb better than the more efficient alternatives. I value not getting a headache every evening more highly than the $8 I spend in electricity annually on those bulbs. So now I have to resort to hoarding light-bulbs.
I'm an engineer passionate about efficiency, so I tweak stuff like this. My mom isn't -- she ends up suffering with awful light.
"You can buy CFLs. Some are great, others make your living room look like a gas station bathroom in 1960."
Please, please don't. We all have a duty, with our wallets, to make CFL die. It is garbage, stopgap technology that should have been strangled in the crib. Oh and also it's toxic waste.
Buy LEDs. Buy incandescents. Buy xmas tree lights. Buy anything but CFLs.
edit: how can they be more efficient than cfls if they get hotter? how are the cooler cfls wasting energy? you can't just "lose" energy - it has to go somewhere. and normally for losses, it's heat.
A 60W-equivalent LED lamp uses 6-8W of energy. Assuming it converts roughly the same amount of energy (1.25W) into light, the remaining 4.75-6.75W needs to be converted to heat. That's about as much thermal dissipation as a low-end laptop CPU, but concentrated in a much smaller area and with a much less effective cooling system (it doesn't even have a fan). 85 degrees sounds about right.
[Edit] Still, 85 degrees is nothing compared to the extreme temperatures that a filament in a regular light bulb experiences. It's only a problem because the LED chips aren't made of carbon or tungsten like filaments are.
[1] https://en.wikipedia.org/wiki/Incandescent_light_bulb#Effica...
This isn't relevant to your argument, but that actually isn't true, that figure is about 6~8 percent.
* The "luminous efficiency" as commonly reported is just the lumen/W luminous efficacy divided by the peak value of 683lm/W. The value of 683lm/W is the peak of human eye sensitivity curve and occurs at 555nm. The only light source that can achieve 100% "efficiency" by that measure is a 100%-efficient radiator that emits monochromatic 555nm (bright green) light, the best a perfect white light source can do is about 37%. (see [1])
* That "perfect white light source" would be a 100%-efficient radiator emulating the Sun spectrum bandlimited to visible light (zero infrared, uv and no other losses).
* It's possible to make a light source that appears white to human eye and is more efficient than that by emitting mostly at wavelengths the human eye is more sensitive to, and emitting less at wavelengths it's less sensitive to. That's part of the reason why manufacturers often offer CFLs with high efficiency but poor color rendering and lower efficiency but higher color rendering. (And don't get me started on the color rendering index.)
* That definition of "efficiency" is obviously not what most people think it is when they hear that word (useful work / total work spent), some prefer to call it "luminous coefficient". But it's not all lies, it was defined that way because it's easy to compute and understand and because this is the value of interest if e.g. you're making a LCD backlight or if you're designing an interior lighting setup.
[i thought that was obvious - are you trolling?]
In terms of energy efficiency, CFLs are halfway between incandescent bulbs and LEDs. So the same comparison holds, only the differences are less extreme.
60W-equivalent CFLs use 13-15W of energy. That's approx. 1.25W into light (because the amount of light is more or less the same in all cases) and the remainder (11.75-13.75W) into heat. That's why CFL lamps get rather warm after a while. Very bright CFLs often have slots in the base for ventilation [1].
You just don't notice it as much because:
1) The heat source of CFLs is much larger than LEDs, so the average temperature is lower even though the total amount of heat is higher.
2) With all those coils and folds, CFLs have a larger surface area than most other types of lamps, so they cool easily.
[1] https://en.wikipedia.org/wiki/File:Compact_fluorescent_light...
That isn't true. LEDs are equivalent to CFLs in energy efficiency. The only LEDs that are more efficient are either lying to you (they claim the light is more directed so it "counts" as more, and they uprate the bulb), or they have worse color accuracy so you are not comparing equivalent bulbs.
> The heat source of CFLs is much larger than LEDs, so the average temperature is lower even though the total amount of heat is higher.
No. The heat source is not larger - it's in a different location, it's in the bulb part and it emits infrared light and is sent into the room. With LEDs the heat is made in the base and stays there.
> With all those coils and folds, CFLs have a larger surface area than most other types of lamps, so they cool easily.
That doesn't explain why an incandescent that makes 6 times as much heat can manage, but an LED can't. It's not the cooling of the bulb, it's the type of heat - incandescents and CFLs make their heat in the form of infrared light that is sent into the room.
Let's exclude lying vendors.
Worse color accuracy is annoying for those who care, but as far as total energy output is concerned, it shouldn't matter. A watt is a watt no matter what frequency it is radiated at.
> The heat source is not larger - it's in a different location, it's in the bulb part
Sure. And the bulb of a typical CFL is hundreds of times larger than the die of an LED.
> That doesn't explain why an incandescent that makes 6 times as much heat can manage, but an LED can't.
As I mentioned in the great-great-grandparent comment, glass tubes (CFLs) and tungsten wires (incandescent lamps) are much less sensitive to heat than tiny pieces of delicate electronics are. 85'C is child's play to glass and tungsten but deathly to silicon.
> incandescents and CFLs make their heat in the form of infrared light that is sent into the room.
The "sent into the room" part is definitely correct. LEDs overheat more easily because the heat is generated and trapped in the base. But even if you dangled a bare LED in mid-air, I doubt you could make it as cool as a CFL because the heat source is concentrated in such a small area. I'd also be surprised if those LEDs didn't generate a lot of infrared ratiation.
Ah, but if you are willing to have poor color accuracy then CFLs should be allowed that as well, and if you do that then their energy efficiency is better than an LED.
> glass tubes (CFLs) and tungsten wires (incandescent lamps) are much less sensitive to heat than tiny pieces of delicate electronics are
CFLs have switching power supplies in the base that don't like heat either. It's not just a glass tube. But I'm willing to accept that they can handle heat better than an LED emitter. (And I'm pretty sure it's just the emitter with the problem - it's certainly possible to make "delicate electronics" that can handle heat.)
the whole discussion above is concerned with the relative amounts radiated in visible and infra-red....
LEDs and CFLs are about equally efficient, but CFLs produce their heat on the bulb, and LEDs produce their heat in the base.
So CFLs (and incandescents) send their heat out into the room, which cools the bulb. LEDs keep all the heat inside which is a huge problem for them.
Personally I think bulb replacement LEDs are a dead end, and instead you should look for dedicated LEDs fixtures.
Edit to your edit: The amount of heat is actually quite small in comparison to a CFL. The temperature they're talking about is the die temperature. LED dies are tiny, so even though they can reach high temperatures, there's much less heat (waste energy) generated.
I have a 60W bulb I run through a dimmer that is in its 4th year. It is a nightlight for the kid's room so is never really turned off.
Filament resistance goes up because it is evaporating, heat goes up because the filament increases in resistance.
Higher overall resistance connected to the same voltage produces less current draw and therefore less heat. It's just Ohm's law.
The filament doesn't thin out uniformly, it tends to develop localized hotspots which leads to accelerated, localized evaporation. Eventually it breaks.
I have a 60W bulb I run through a dimmer that is in its 4th year.
You can make an incandescent bulb last an arbitrarily long time if you're willing to sacrifice efficiency by running the filament cooler. A dimmed-down 60W bulb may produce the light of a 5W nightlight bulb but use far more than 5W.
The Centennial Light Bulb has lasted a long time, not because it's never turned off, but because it's absolute shit as a light source. The filament stays cool and as a result its both terribly dim and terribly inefficient.
I didn't say heat didn't degrade incandescents, I said LEDs are much more heavily degraded when they're over their heat capacity.
The reason is due to the balance between heat generated and heat dissipated into the environment.
The bulb could convert 99% of the input electrical energy into light, and only 1% into heat. So for a 100W bulb, that would be only 1 watt of heat.
But if the heat dissipation system (i.e., the cooling system) can only dissipate 1/2 watt of heat into the environment over the same time span that the bulb is creating 1 watt of heat, the excess 1/2 watt of undissipated heat will build up over time, making the housing hot to the touch.
A CFL has a much larger volume to produce that heat, and a much larger surface area to radiate it; and even if it got very hot it wouldn't care that much. A LED is a tiny speck of semiconductor that bears the brunt of a lot of energy, and it's not rated for operating at high temperature.
It's like normal sunlight warming up the palm of your hand versus sunlight focused to a point by a palm-size loupe - it's the same total energy, but one is nice and warm whereas the other can cause serious damage.
They are not more efficient than CFLs. It is interesting that you think they are - I guess the marketing is working.
Now, you might come across some LEDs that on the surface appear to be more efficient, but actually they are misleading you. Those efficient LEDs have worse color accuracy than CFLs, so you are not comparing equivalent products.
LEDs do have room to improve, so they may get more efficient than CFLs, but they are not there yet.
The problem with LEDs, is you need a metal radiator positioned to not block the light. In a recessed fixture, the heat can get trapped.
"Ed Rodriguez [the author]...formed OptoThermal Technologies to focus on thermal management technologies related to high-power LED lighting."
It's a design issue of the bulb, sure, but especially the fixture. LEDs aren't friendly to high heat like incandescents are, and most light fixtures have been designed with complete disregard to the heat of the fixture—so long as it won't start a fire.
The CFL has a huge volume and a huge radiating area for the heat, and that area is well-connected thermally with the environment. Plus, it can operate at higher temperatures anyway.
The LED active element is a tiny chip of semiconductor material. All that heat is dumped in a tiny volume, and you can only put a heat sink on its back. Plus, the heat sink is close to the fixture and therefore necessarily hampered in its function.
It's not an easy problem to solve. If we think pie-in-the-sky solutions, perhaps a re-design of the connectors and fixtures might help - if you made a connector that was designed to take heat away from the bulb, that would help a lot; the whole fixture would become the heat sink.
Further, LED manufacturers design their directional lamps with these applications in mind. See the spiral pattern [1] on the heat sink around Lighting Science PAR bulbs? That's designed to use the thermal differential to create a nice air flow up into the fixture so that it may be used in a recessed or other partially-enclosed fixture.
1: http://www.1000bulbs.com/product/100560/LED-PAR3011HE25WW.ht...
A light bulb.
This isn't a car, which you expect to require some expertise. It's not a furnace that will be primarily handled by an HVAC specialist. It's a light bulb. Consumers have a reasonable expectation that devices like this Just Work. If the device fails to meet its promised service-level without a dozen asterisks in its usage? It's not meeting its intended purpose and it's defective.
Consumers should not have to meticulously research every small purchase. Simple, small purchases of replacement parts should not require reading a 10-page instruction manual of 6-point font, especially since you didn't get this manual until after you got the damned thing home.
By allowing manufacturers to sell these defective devices, we don't just hurt consumers, we hurt the environment with unnecessary waste of electronics, and we hurt the real, quality manufacturers who want to sell good stuff but can't because there's no way to tell consumers "all the other crap on the shelf will break in under two years and we won't" so they end up getting crushed by people who cut corners.
"Educating the consumer" is the onus of the manufacturers. It doesn't mean 10-page instruction manuals with 6-point font or hours of meticulous research [edit: this is a horrible way to educate consumers]. In this case it probably means packaging and labeling which plainly states "not for use in recessed fixtures" in a way which most anyone would understand. This could be a sticker which says just that, it could be iconography, it could be an obnoxious DVD on loop in the lighting aisle at Home Depot, or it could be all of those things.
Yes, it's a light bulb (a light bulb!) but that doesn't mean manufacturers shouldn't attempt to get consumers to use it properly.
If you want to make a special light-bulb that is only compatible with a narrowly defined set of environments, then make your own socket that is only compatible with your bulb and then sell fixtures that provide the needed environment.
If you want to use a standard socket, then you have to take the bad with the good and also deal with the kind of environments you'd find. This is the same bullcrap with devices that completely fail to charge on 0.5Watt USB power-supplies. If you can't charge off a standard USB connection, then you shouldn't use a USB port, because USB defines 0.5Watts and your device isn't USB-compatible. Legally required to support USB? Then stop pussy-footing around and actually support it instead of exclusively supporting a 1 or 2 watt perversion of the standard.
The need to educate a customer usually surfaces when someone was unable to produce a proper product. Sure, you need to warn users when your product can't be used under some circumstances, but if possible, you should aim at fulfilling the users' expectations instead.
I totally agree with what you're saying, though.
The issue at hand is that consumers want something that is impossible for manufacturers to give to them. That is, a 100% efficient lighting solution which shines light exactly when they want it, exactly how they want it, and lasts forever. Consumers have gotten used to a market where there has been almost no innovation for over a hundred years (in the driver side, fixtures are a different story). Now that there's some diversity in the market, consumers are unhappy because it's not matching expectations. The only way that will change is if manufacturers properly set expectations.
He did note that the Cree brand bulbs are actually rated for use in enclosed fixtures, but alas, they're also more expensive at the moment than these other guys.
Everything is a horribly underappreciated aspect of modern life. You'll hear the same sentence with one word changed uttered by audiophiles, vegans, electricians, foodies, urbanists, doctors, social justice advocates, janitors, seniors, juniors, teachers, genealogists, biologists, ecologists, chemists, astrophysicists, philosophers, grammarians, body-builders, trainspotters, mechanics, open-source software fans, designers, photographers, documentarians, readers, writers, gun-nuts, gun-haters, pastors, prudes, perverts and relationship councillors.
We, collectively, no longer give a fuck. We are out of fucks to give. We are way past peak-fucks and into a severe fucks crisis. We have zero fucks.
I just want to buy something and have it work. If the colour temperature is wrong and the aesthetics are therefore sub-optimal? Well that reflects the effort I put in. If it burns out in a week? It's defective, full-stop.
One wonders what the owners of gaslights thought when they bought their first simple "Just Works" light bulb.
And of course many of us were there for Grandpa Versus the Microwave Oven.
This bulb claims to be the exact replacement:
http://www.amazon.com/gp/product/B00467E3LA/
But yet this one and others that have similar lumens, temperature, and equivalent watts all don't even get close to the warmth that the halogens produced.
Fortunately, I found a cheap alternative:
http://www.amazon.com/Philips-424374-7-watt-Household-Dimmab...
It's the A19 40W that the article cautions against, and it looks slightly less sexy in the recessed can, but the color almost exactly matches the halogens I have (except when dimmed -- nothing can match the fireplace-like glow).
What's weird is that the lumens and temperature for that bulb aren't even close to what the halogens claim, but they look almost identical. Which has made me wonder if this switch to LED is way more complex than just those two measurements :P
I can't talk specifics (under NDA from a former employer), but this is a problem for which solutions exist. I don't know that manufacturers think there's much of a market there, however.
> Which has made me wonder if this switch to LED is way more complex than just those two measurements :P
It is. There are a bajillion things going on. One which you've already noticed is how dimming impacts color temperature. CRI is also a huge issue. Then there's the optics - LEDs are very directional, so they require pretty advanced lenses in order to achieve a desired lighting pattern [1]. Optics impact all kinds of things, including color temperature, and since the development cost is high you can expect cheap brands to skimp there. If you're comparing them by looking directly at the bulb, internal glare/reflection is hard to match as well.
I'd still caution against using the A19s in a recessed fixture. It's not a safety issue (that I'm aware of), but you definitely won't get the rated life out of the bulb.
Out of curiosity, did you try an LED BR30? I think you might find one from a quality manufacturer to be a bit better than the A19.
1: $20 says the lighting pattern on the Sylvania LED is smoother/more regular than the Sylvania halogen, but I'm biased since I know the guy who designed the lens.
http://www.amazon.com/Ecosmart-Powered-Recessed-Light-Dimmab...
(It's old knob & tube wiring. Electricians claim it's safe)
http://www.oldhouseweb.com/how-to-advice/knob-and-tube-wirin...
The major problem is that since knob and tube lost favor and today, demand for electricity within the home has grown considerably, and a lot of knob and tube systems were expanded improperly; this can happen with modern wiring as well. Also, the insulation issue.
But if exposed and someone can touch it, then it's dangerous.
You can make LEDs with good color accuracy but you lose efficiency making them worse than equivalent CFLs.
Then buy CFLs? They help keep toxic mercury out of the air from burning coal.
Do you seriously believe CFLs emit any mercury, or are you trolling?
[1] And not all canned tuna is Albacore. The stuff sold as "light" tuna is typically lower mercury content (don't have a source handy at the moment.)
Basically: If you want people to use less power, then increase the price or ration it. Don't ban an unrelated technology. If you want people to use less glass, haha are you kidding me, glass is made of dirt and easy to recycle.
Is there any harm caused by incandescent bulbs specifically that I have somehow missed?
The reason for why we want to have energy efficiencies is simply because we only have to produce so much more electricity to power everything else. We have not reached a point where nuclear, solar, or wind is cheaper by the MW than coal and as a result we'll continue to build those for the foreseeable future. Banning inefficient and obsolete light bulbs will prevent us from needing to burn more coal and other harmful fuels.
If we didn't have this problem with electricity generation then banning something may become less desirable.
Or you could be really direct and make a law against powering incandescent bulbs on coal power, letting everyone choose if they want to buy some renewable/nuclear energy for their wasteful bulbs.
No, but for the moment it does actually do that. And you can change what sort of light bulbs people use faster than you can change your large-scale power generation technology.
(I do agree that making environmentally-better energy cheaper is likely a more powerful way of reducing the environmental cost of electricity use than telling everyone to stop using incandescent bulbs. But "X is better than Y" is a reason for doing X, not necessarily a reason for not doing Y.)
> a law against powering incandescent bulbs on coal power
Enforcing that one would be an interesting challenge. A law that says "no selling incandescent bulbs" is rather easier to enforce.
Lots of people would agree with that or the point in general that financial penalties work better. Legislation should mandate results, not dictate methods.
The conventional wisdom here (Denmark) has been that a mixture of the two achieves best results: a high tax on energy to provide an across-the-board untargeted price incentive, plus specific targeted regulations aiming at cutting out the largest sources of inefficiency and encouraging economies of scale in more efficient technologies. So for example district heating is now mandatory if you're in an area where it's available (no furnaces in new construction), and this seems to have been quite effective in increasing its adoption, which in turn makes it easier to build efficient district-heating networks and cogenerating power plants, since there is a dense customer base.
The only cases where I don't see a CFL replacement yet are decorative clear lamps where the filament being visible is part of the aesthetic.
* There are notable gaps in the spectrum causing color distortions. CFL light mostly features a greenish or violet shade and looks "hollow". Tungsten and halogen are way better, even after 20 years of CFL development.
* Most, even the more expensive ones are not really instant-on. They reach their max brightness after some time
* As I regularly switch off lights in rooms that are empty, they tend to fail rather fast.
I think new kinds of fixtures will need to be designed, as well as new standards for DC distribution within the house (I am trying out 48V and 24V). Retrofitting existing infrastructure with LEDs makes no sense: we get wasteful, fragile and inefficient systems.
I wholeheartedly agree and would go even further and reimagine our current illumination concepts. At the moment we mostly have a few powerful point light sources. Given the longevity of LEDs we could easily design walls and furnitures with hundreds of less powerful LEDs that would produce a much more even illumination, have far less problems with head dissipation and be cheaper to produce than those high luminosity LEDs. And even more interesting, this would open the door to far more creative designs than we have now. With OLEDs you could even have glowing wallpapers: http://oled-werbedisplays.de/wp-content/uploads/2013/02/oled...
This is something I've wondered about for a while. Everything in my home aside from a few kitchen appliances is now running a AC<>DC converter. At what point does it make more sense to run DC circuits in the home? Would the gauge be a problem to run enough DC power for a desktop PC, a TV, and a ton of LED lights?
I was planning to use a power supply for every group of lights — e.g. one for the ceiling, one for under-cabinet lighting, etc. That way you don't have to run long stretches of wire with DC, and yet you don't have to build lots of tiny low-efficiency converters.
In some sense, I think you are arguing that your oven should be compatible with gasoline (cause hey, you can put gasoline in a container that will fit in an oven). Of course you aren't, I just think you are probably headed too far in that direction. I guess I'm willing to make consumers do more than make sure they fit together.
I've seen senate reports that take less time to get to the point about safety warnings. TL;DR: Many leds aren't meant to be in unventilated can fixtures and will either burn themselves up, reduce lifetime, or worse start fires. At least that's what I think I got out what was mainly a lot of talk about other types of bulbs.
Alternatively, they could be built into light fixtures designed specifically for LED bulbs, which would be less expensive overall, and would allow the cooling to be designed for the specific fixture (face up vs. face down, for example).
* They use 30-40% less power, costing you less to operate
* Cheap CFLs commonly had ballasts that died in 1-3 years despite higher life expectancy claims on the packages; LEDs are backed with 10+ year warranties
* CFLs burn out quickly if used in fixtures that are turned on and off frequently
* LEDs don't flicker/strobe like fluorescent lighting
* Virtually all LED bulbs work in dimmers, not just more expensive specialty ones
* Cheap LED bulbs don't produce bluish light like cheap CFL bulbs
That isn't true. Only LEDs with poor color accuracy are more efficient than CFLs.
> CFLs burn out quickly if used in fixtures that are turned on and off frequently
People claim that, and it makes sense, but it doesn't actually happen. Modern CFLs do just fine.
> LEDs don't flicker/strobe like fluorescent lighting
Florescent lighting does not flicker, and cheap LEDs actually flicker more than CFLs.
> Cheap LED bulbs don't produce bluish light like cheap CFL bulbs
Walmart CFLs cost $1 and have excellent color. It doesn't get cheaper than that.
It seems you made your mental list more than a decade ago and have not updated it as CFLs have gotten much much better.
Deleted comment
And my experience doesn't jive with your list. And they were not jabs [seems like you edited out that part of your response], just replies to each point. Nor did I accuse you of making it up - read again what I said. Just that simply because you have that experience doesn't mean it's like that for everyone.
I have lots of CFLs in my house and none flicker, and they all seem white to me.
I bought a cheap LED for outdoor lighting, and it's noticeably blue. Which is fine for its intended purpose, but my experience has been the opposite of yours.
You're acting like your experience is the ultimate truth and anyone has who has a different experience is attacking you or has some strange problem.
I also find LEDs to have a speckle pattern sort of like a laser. I find them very unpleasant and harsh.
Flicker - I actually have an Ikea LED lamp that has two active elements inside (2 LEDs). The supply is a simple 12 V AC transformer. I suspect the two elements are connected in a push-pull (anti-parallel) circuit, because they've started to flicker recently - if mounted push-pull and not equal in terms of brightness, that would explain it.
I thought CFLs produce more of a weird green hue - at least that's how both my eye and my cameras see it. At any rate, the main emission line of Hg is smack down the middle of the green part of the spectrum.
With incandescent bulbs I was also concerned with electric shock from broken bulb, cuts and burns... though I wonder if the LED bulb would be a fire hazard if it were to rest on something.
Yes, I purchased my LED bulbs in anger. We'll see if I've been had.
The UL has now required end of life circuits in all CFLs so this won't happen again in new CFLs. (And new is relative - this regulation came about a few years ago.)
A great example of the kind of article we should see more of on HN.