The Great Lightbulb Conspiracy

HughPickens.com writes: Markus Krajewski reports that today, with many countries phasing out incandescent lighting in favor of more-efficient and pricier LEDs, it's worth revisiting the history of the Phoebus cartel — not simply as a quirky anecdote from the annals of technology, but as a cautionary tale about the strange and unexpected pitfalls that can arise when a new technology vanquishes an old one. Prior to the Phoebus cartel's formation in 1924, household light bulbs typically burned for a total of 1,500 to 2,500 hours; cartel members agreed to shorten that life span to a standard 1,000 hours.

Each factory regularly sent lightbulb samples to the cartel's central laboratory in Switzerland for verification. If any factory submitted bulbs lasting longer or shorter than the regulated life span for its type, the factory was obliged to pay a fine. Though long gone, the Phoebus cartel still casts a shadow today because it reduced competition in the light bulb industry for almost twenty years, and has been accused of preventing technological advances that would have produced longer-lasting light bulbs. Will history repeat itself as the lighting industry is now going through its most tumultuous period of technological change since the invention of the incandescent bulb?

"Consumers are expected to pay more money for bulbs that are up to 10 times as efficient and that are touted to last a fantastically long time—up to 50,000 hours in the case of LED lights. In normal usage, these lamps will last so long that their owners will probably sell the house they're in before having to change the bulbs," writes Krajewski. "Whether or not these pricier bulbs will actually last that long is still an open question, and not one that the average consumer is likely to investigate." There are already reports of CFLs and LED lamps burning out long before their rated lifetimes are reached. "Such incidents may well have resulted from nothing more sinister than careless manufacturing. But there is no denying that these far more technologically sophisticated products offer tempting opportunities for the inclusion of purposefully engineered life-shortening defects.""

Comments

"the Phoebus cartel still casts a shadow today"

[Anonymous Coward]

the Phoebus cartel still casts a shadow today because it reduced competition in the light bulb industry

I see what you did there!

Re:"the Phoebus cartel still casts a shadow today"

[Anonymous Coward]

Joking aside, I'd argue that it did the opposite. By removing the durability as one aspect of differentiability, the cartel commoditized the light bulb. Not being able to compete on quality or features generally leaves the manufacturers no choice but to compete on price. Another aspect of the bulb lifetime is that it's easy to make a bulb last longer: You just dim it a little. But by doing that, you reduce the already bad efficiency. For a bulb which consumes many times its item cost in electricity, increasing the lifetime by lowering the efficiency drives up the total cost. That's unless there's a significant cost associated with changing the light bulb, then longer lasting bulbs can make economic sense, and - surprise - you can buy longer lasting (slightly dimmer) bulbs, but the regular bulbs still last 1000 hours.

Regarding the durability of CFLs and LED bulbs: These types of light sources have embedded electronics that age faster at high temperatures, unlike incandescent light bulbs. To get the rated lifetime out of these modern lights, use them in well ventilated fixtures which keep the heat away from the socket. In my experience, the lifetime of CFLs follows the usual bathtub curve: Some duds die in the first weeks or months, almost none die between a year and well beyond the rated lifetime, and then at some point the failure rate goes up again. None of the LEDs in this house have failed so far (after close to three years since installation), so I have no reason to expect that they won't last the rated lifetime. All CFLs and LEDs have by far recovered the investment cost in saved electricity.

Re:"the Phoebus cartel still casts a shadow today"

[Mashiki]

None of the LEDs in this house have failed so far (after close to three years since installation), so I have no reason to expect that they won't last the rated lifetime.

LED's I've yet to have a problem with. CFL's, I've had nothing but problems with, ranging anything from massive flicker bad enough to cause migraines to them going up in smoke in a matter of months even in your standard lamp base. It seems to me that manufactures the first couple of years after CFL's became common started cutting costs by reducing the quality of the components themselves. Leaving you with a good glass fixture, and cheap ass electronics. Most of the failures I've seen after pulling them apart fail on resistors or capacitors. Lot of the people saying "the caps are over heating" to me, in all the cases where I've seen a capacitor fail, it's followed the same path as the "bad cap" scandal that hit PC motherboard makers in the early 00's. That is, fake caps.

I dunno about LEDs, but CFLs don't last

[rossdee]

I dunno about LEDs, but CFLs don't last as long as they are sposed to.
Of course they still save money in power costs

Re:I dunno about LEDs, but CFLs don't last

[Bomarc]

I refuse to get any any more LED light bulbs... every one that I've purchased - from multiple companies - has burned out prematurely. NOT WORTH THE COST. And CF are dangerous. (If one breaks, you need to open the windows and leave the room for 1/2 hr.) Further - no viable light bulb replacements will work with dimmer switches (Which my house has many).

Re:I dunno about LEDs, but CFLs don't last

[Anonymous Coward]

Hey it's your wallet..

I do totally agree on the CF - i had some break but, that was actually not the glass tube with the poisonous mercury vapour, but the electronics inside it with -going by the smell- even more toxic ingredients. One actually 'exploded' leaving the bulb hanging on some wires. I do not know yet if LED lights are any safer in that regard as they require some, albeit less, electronics.

For dimming, you might look at the better LED lights as some of them are dimmable - in Europe they are quite common, and if dimmable, explicitly say so.

Re:I dunno about LEDs, but CFLs don't last

[gewalker]

I was unscrewing one of the twisty CFL's that died after probably 1000 hours use. It basically exploded in my face, about the top 15% of the bulb was small shards though the rest was intact. Yes, we got screwed by the government forcing them on us before they were ready for prime time, just like the water saving toilets that don't flush unless you cycle them a few times.

Re:I dunno about LEDs, but CFLs don't last

[Technician]

I write the date on all my bulbs. Failed bulbs are never replaced with the same brand. The theory goes that short life bulbs will be circulated out of service and long life bulbs will remain.

Note to manufactures, to get on my bad boy list, have high premature deaths. To get on my recommended list, be the last man standing in my testing.

Failures fall in two modes. Lumen maintenance and failure. Most LED's dim over their lifetime. I bought a 3 pack of lower wattage "candelaubra lamps and used them in bathrooms as nightlights. I noticed they were quite dim after about 7 months. Used the 3rd bulb as a comparison as I used only two at the time. I photographed the result with a digital camera on manual settings so all exposures were taken with the same setting and posted the result online. You don't want your short life bulbs mentioned by name in a poor review.

My general observations are older bulbs had higher failure rates than the current line as the technology improved. LED's are an absolute must in locations with occasional use such as bathrooms, but often leave much to be desired where they are on 24/7 or 8-12 hours a day. A CFL in a seldom switched location will often have better lumen maintenance than an LED.

Note on the package on LED's, they are most often rated for only 3 Hours a day. For now use them in hallways, the garrage,storage areas, and bathrooms, I am having some great performance on some newer bulbs in the living room, but it is too early to call, but it is looking promising.

Re: I dunno about LEDs, but CFLs don't last

[AvitarX]

Every led I've purchased (I only buy energy star , because they are actually testing them for light quality ,unlike with CFLs ,so that may make a difference ) , has said it was dimmable , to the point I thought it was intrinsic in the technology .

They're also the only bulbs that have any real life in my old house . My porch light lasts 2-3 weeks incandesant . I'm 4 months on the led I put in it , on convenience alone it's worth the price .

Re: I dunno about LEDs, but CFLs don't last

[RabidReindeer]

Every led I've purchased (I only buy energy star , because they are actually testing them for light quality ,unlike with CFLs ,so that may make a difference ) , has said it was dimmable , to the point I thought it was intrinsic in the technology .

They're also the only bulbs that have any real life in my old house . My porch light lasts 2-3 weeks incandesant . I'm 4 months on the led I put in it , on convenience alone it's worth the price .

You sound like a neighborhood in a town not far from me.

Apparently the houses close to the power substation routinely got something like 20% overvoltage. The lights burned bright, but they burned out fast.

In theory, LED bulbs would actually be a better bet, since they're current-driven, not voltage-driven and the voltage is stepped down. So as long as the LED voltage was less than what it took to fry the LEDs, they might burn a bit brighter, but that's about all.

No, LEDs aren't inherently dimmable, since, like I said, they're current-driven and running on reduced voltages internally. Furthermore, a lot of dimmers are not rheostatic, but instead work by modulating the "pulse width" (for lack of the proper term) of the A/C waveform. An incandescent bulb has thermal mass, so that results in dimmer light (and color change thanks to lower temperature). A flourescent light will generally not be amused, and an LED light is only likely to work if it has some sort of way to convert the pulse width to relative current level (such as via capacitors).

I have some dimmable LED lights and they work great, but it was unusual seeing the same lamp color as the light dimmed. And the light curve for the LEDs doesn't match what resistive lamps do, so there is noticable non-linearity in the brightness relative to the dimmer setting.

Re: I dunno about LEDs, but CFLs don't last

[Ol Olsoc]

The important issue for either CFL's or LED's is the little power supply at their base. And that's where the problems can arise.

If cheap parts are used, the lamp life can be really short.

That being said, some of the stories are a tad apocryphal. I've replaced all my tungsten bulbs with CFL's, and I had one failure over maybe ten years. And that was started by a faulty lamp base.

I haven't had any of the RFI problems reported by some either. If the power supply is not designed properly, it can emit RF, which can interfere with radio reception.

But all of these seemingly horrid issues are mostly via the internet, and everyone I know who uses them hasn't had the issues. So i suspect a lot of this is apocryphal internet stories, so I put them in the same category as solar panels self destruct the moment their warranty goes out, and The Tesla is going to burst into flames, so buy a gasoline powered vehicle which won't, nonsense. Now what I would really really like is a separate line for my LED lamps that will not require the little power supply. Then the lighting would last just about forever, and I could take that off my bucket list of maintenance items.

Re:I dunno about LEDs, but CFLs don't last

[Wycliffe]

I don't know why Americans don't have a dual flush system on their toilets like all of Europe does.

Because most americans would always use the "big flush" anyways. Water is too cheap in the USA and water saving devices are
worthless in this context. If you need to save water, the best way to actually save water is to make the consumer actually want to
save water. Many places in the USA, water saving is a joke as there is plenty of water. In the places where there actually is a
water shortage problem then they need to ration it per person and charge higher and higher tiers for people that use more than the
average. Basically, there needs to be a luxury tax on excess water consumption. In most place toilets aren't the problem but
rather it's all the swimming pools, lawn sprinklers, and industrial uses.

Re:I dunno about LEDs, but CFLs don't last

[TheRaven64]

Sounds like something's wrong with your wiring. I switched to using CFLs over a decade ago and the shortest lifetime I've had on one is 6 years (I think - one might have gone after 4). The first time I moved house, I brought a load with me, but they'd become so cheap that I didn't bother the last time. The only incandescents I've had last longer than a year are ones that are rarely used. I worked out that - back when they were expensive - that after 3 months of operation they'd saved me more in electricity than the cost of an equivalent incandescent, so they've been a pretty good investment.

Re:I dunno about LEDs, but CFLs don't last

[bigwheel]

I've read that the Energy Star people admit they screwed up with CFL, but are determined to not make the same mistake with LED

In order to get the Energy Star label, a CFL bulb has to meet certain efficiency requirements. But the rating says nothing about longevity. In theory, fluorescent bulbs should last a long time. But the built-in electronics are the usual source of failure. This is particularly the case with ceiling lights and other bulbs where the electronics are on the top, and often in an area where they do not get much cooling. So, the cheap - or more importantly *Crappy* - bulbs can carry the same certification as the good ones. So, CFL got a bad name, which is also fail for the Energy Star folks.

With LED bulbs, the Energy Star people wanted to make sure that they don't make the same mistake. So, in order to get the label, a bulb has to meet the efficiency standards, plus demonstrate that they can handle the run-length requirements. And there are many different requirements, depending on the type of bulb and its intended usage. In order to get the Energy Star label, they are tested for something like 9 months.

So, the moral of the story is that if you buy an LED that carries the Energy Star label, it should not fail prematurely. But the down-side is that LED technology continues to improve, with the most recent chips putting out something like 250 Lumens/Watt. An agile manufacturer might be able to quickly get this technology to market with an excellent new bulb. But it cannot carry the Energy Star label until it has been through rigorous testing, which takes nearly a year.

Re:I dunno about LEDs, but CFLs don't last

[PopeRatzo]

I've got an LED bulb that's been burning since 2007. I'm pretty sure that's more than 50,000 hours.

And, I've got LEDs on dimmers.

I'm sorry that your bad experience with LED lighting has caused you to go back to gas lamps, but really, they save a lot of energy and work really well. And I love not having to get up on the step stool several times a year to replace recessed bulbs in the ceiling.

Re:I dunno about LEDs, but CFLs don't last

[seven of five]

And CF are dangerous. (If one breaks, you need to open the windows and leave the room for 1/2 hr.)

I know mercury vapor hysteria is totally in vogue, but really? Just put the pieces in a ziplock bag and take it to the local hardware store.

Re:I dunno about LEDs, but CFLs don't last

[Immerman]

Were you grabbing it by the bulb? They typically specifically warn you not to do that on the package. Instead you're supposed to only screw/unscrew them by grabbing the plastic base. Basically that funky glass coil has *horrible* mechanical properties and would have to be radically thicker to be able to withstand the sorts of torque that your average incandescent bulb can handle with no problems.

Re:I dunno about LEDs, but CFLs don't last

[gstoddart]

When I did the arithmetic quite a few years ago, the break even point for CFLs was about 3 months.

Really? Is this for a 24x7 light?

Because I find it hard to believe in 3 months the extra few dollars/bulb is offset in energy savings.

Re:I dunno about LEDs, but CFLs don't last

[jeffmeden]

Out with it, where are these scathing reviews of yours?

Re:I dunno about LEDs, but CFLs don't last

[Smidge204]

I've been using the twirly type CLFs in a ceiling fan "glass ball" light for years (upside-down and enclosed, expressly against the manufacturer's warnings to not use them inverted in enclosed fixtures!).

In fact I've gotten into the habit of dating them with a sharpie before I install: Nov 2011. Since this is in my bedroom it's used for several hours a day, every day. Coming up on 10,000 hours, which is the rated life of the bulb, despite the warranty-voiding installation.

That said, the early generations of CFLs were absolute shit. Don't let that turn you off on the tech, and a few extra bucks for buy a decent brand is worth it.
=Smidge=

Re:I dunno about LEDs, but CFLs don't last

[TWX]

Toilet flushing at my house happens generally between four to eight times a day. The amount of water that we spend on the landscaping dramatically exceeds that. The amount of money I'd spend on new commodes would probably exceed a decades' water savings, and the current commodes use very inexpensive, easy to source parts for what little repair they need, which to date has been a new chain and a new flapper valve. These toilets were manufactured in the late seventies and will probably continue to work for another 30+ years.

And that's not even accounting for the one in the basement, that deposits into a sewage lift pump, which needs a certain amount of liquid before it pumps the blackwater from the basement out to the municipal sewer. I'd have to flush a toilet two or three times to engage the lift pump, which wouldn't save anything.

Re:I dunno about LEDs, but CFLs don't last

[Pope]

LMAO "wasted energy", they use 10x less energy than incandescent for the same rated lumen output.

Re: I dunno about LEDs, but CFLs don't last

[Balthisar]

>but it output is configured in current mode

It's not even that difficult. You probably know exactly how they work, and are only struggling for words.

"Constant current" is the same as "constant voltage" if the load is static. If an LED needs 100mA and the voltage (as you accurately described) is constant, there's no "current mode" regulation needed. Just a known resistance.

For others, LEDs are definitely current devices. Remember: current isn't *put*; it's *drawn*. If the conductor is big enough (e.g., no resistor), then regardless of the voltage, LEDs will suck up all the juice they can, glow brightly for a short amount of time, and then die. So with a known voltage, put a resistance in series, and you have a stable LED semiconductor.

A good switching power supply will produce a stable output voltage regardless of the input voltage (within specs, that is). Ergo failure of LEDs due to overcurrent situations is most likely the result of crappy switching power supplies.

Re:I dunno about LEDs, but CFLs don't last

[gstoddart]

You know, if the argument is that I save money on my electric bill AND do something for the environment, great.

If the argument becomes "gee, you'll probably spend more money over the long run, and you're doing the electric company a favor by subsidizing their reductions" ... well, then I'm less convinced. Especially since most electric companies just seem hell bent on gouging the consumer.

And, I never said anything about a hummer being better than a hybrid.

I'm saying if you pay a $10+K premium for the car, and after 5 years or so spend another $5+K on new batteries AND you don't get the fuel economy promised to you ... then overall, you're paying more for less, and other than reducing local emissions haven't really done much.

I like the idea of hybrids which actually save the consumer money and reduce emissions. I'm not entirely certain it achieves that.

If in the end it's just a shell game where we're more out of pocket Well, the polluting less becomes a negative incentive, unless you have the luxury of saying you're willing and able to do that.

Because, really, nobody is factoring that into cost of living equations, and most of us are already squeezed as it is.

Re:I dunno about LEDs, but CFLs don't last

[djrobxx]

Further - no viable light bulb replacements will work with dimmer switches (Which my house has many).

Although I'm sure some must exist, I haven't seen an LED bulb that's NOT dimmable, even the newest cheapest ones. Their dimmability is one of their their best advantages over CFL. Dimmable CFLs do exist, but they don't work very well; they are usually quite bright at the lowest dim setting.

I put in around 20 recessed CREE (EcoSmart-branded) LEDs almost 4 years ago. All of them are on dimmer switches. None have burned out yet. I've since put in lots of others both indoors and out. I did have one landscape LED fail - it started blinking like a turn signal a few weeks after I installed it. I also had one screw-in bulb that failed within a week. In both cases I was able to exchange them at the store since they were early failures.

Re:I dunno about LEDs, but CFLs don't last

[Barsteward]

the best way is to use grey water from the showers and hand basins to flush toilets, it should be a crime to use drinking water to flush a toilet

Re:I dunno about LEDs, but CFLs don't last

[Immerman]

I hear you there. Typically since I know I'm working outside design tolerances I'll wrap the bulb in a plastic bag before installing/removing it from a socket where I have to grab it by the bulb, to reduce the mess if there's an implosion. I also make sure to be very gentle compared to if I were installing an incandescent bulb, and that sometimes means having to screw/unscrew an incandescent bulb into the socket a few times first to rub off any corrosion or burrs, and/or do the same to the CFL in a more accessible socket.

Re:I dunno about LEDs, but CFLs don't last

[Waffle Iron]

I was looking at LED replacement bulbs at the hardware store the other day ($20 each). I am suspect as to their efficiency.

Get a Kill-A-Watt meter and test the power consumption of LEDs youself. All the ones I've checked have used just about exactly what it says on the package.

They have large heat sinks on the which get very hot. That is wasted energy.

They have heat sinks because the LEDs need to stay very cool to work properly. Incandecent bulbs don't use heat sinks because they need to heat up to thousands of degrees just to get a small fraction of the photons they emit into the visible range. Now which do you think is wasting more energy?

There is no way to pack an efficient transformer into such a small space.

I doubt that any CFL or LED on the market is using a plain 60Hz transformer. They're using switching power supplies, which can be very efficient. That's becuase they crank the frequency up to a range where a small transformer *is* efficient.

Houses need wired seperately with a lower voltage appropriate for powering LED lights.

You'd still need a switching power supply to match the low voltage to the exact needs and wiring pattern of the particular LEDs. That's why most every PC have a separate power supply on the motherboard just inches away from the main power supply to convert 5VDC to whatever the processor needs.

Not to mention the power loss of low-voltage wires. If you put 100W of LED lights (about 6 bulbs) in a room at the end of a 50-foot run at 5V, you'd be pulling 20 amps. If you used 14AWG wire, at 0.25 ohms for the 100 foot round trip, you'd have a 5V voltage drop just from the resistance of the wire. You would also be violating code, which would require you to install a dedicated 12AWG circuit just to power 100W. That's obviously completely unworkable.

In summary, all of your uninformed "gut feel" opinions on these technical issues are unsurprisingly wrong.

Re:I dunno about LEDs, but CFLs don't last

[TheRaven64]

Let's do it again. The last set of CFLs I bought cost 50p each (the ones before that were 30p, but that was a special offer). According to my electricity bill, I pay a tiny fraction of a penny under 15p/kWh. That means that the bulb costs slightly more than 3kWh of electricity. It's a 15W bulb replacing a 60W one, so that's a 45W saving. Assuming that the incandescent is free, then it takes 75 hours of operation for it to save money. At 3 hours a day, that's 25 days. When I first did the arithmetic, the CFLs were 3-4 times more expensive, so it worked out at 3 months.

This is why I find the resistance to CFLs so hard to understand. It's saved me quite a bit of money over the last decade.

Re:I dunno about LEDs, but CFLs don't last

[CastrTroy]

In my city, they started campaigning for people to save water. The result? People saved so much water that they ended up having to raise the rates because they weren't making enough to run the water system. The system basically has a fixed cost to run, regardless of how much water goes through it.In the end, we cut our water usage in half (averaged over the city), but we now pay twice as much for our water.

Re:I dunno about LEDs, but CFLs don't last

[liquid_schwartz]

I have worked in the solid state lighting industry. You typically hear 50,000 hours but there are a few points that must be met to achieve that:: - The LED must not exceed a certain temperature (usually 135C at the die though that can vary a bit between manufacturers) - The LED must not exceed a rated level of current flow - The 50k hours rating allows for a dimming over time of ~20% though again that varies between different manufacturers The most common failure points for LEDs are heat, over driving the LEDs, poor regulation circuits resulting in over current, and using too cheap of components which fail before the full lifetime is met. A lot of claims are made regarding both brightness and longevity that don't hold up, especially if it's an off brand from China. Those will always claim 50k to 100k hours of operation at 100% (or more) of the LEDs full rated light.

Re:I dunno about LEDs, but CFLs don't last

[liquid_schwartz]

I dislike waste and generally try and conserve but it always seems like a scam when you save yet end up with the same bill. It's the environmental version of no child left behind where not everyone can be above average. If we all save water our bills won't all go down we'll just pay more for less. I've seen this scam multiple times with water mostly though also with electricity. Since water demand has limited elasticity I expect someone to pull an Enron on water supplies soon, especially if we get another corporate administration.

Re:I dunno about LEDs, but CFLs don't last

[turp182]

Toto's do (start at about $300, but well worth it). They have a two stage flush, and 15/30 second refills depending on the stage. The only thing they suck at is... heavy dark matter on the back of the bowl. That's what the $1,000+ versions are pressurized...

During a large party it's nice to have a toilet that can move 4 people per minute through the bathroom line.

Great toilets, made in America. They get crazy expensive on the high end (they go all Japanese with digital controls and extra spray stuff, warm air, etc.).

Re:I dunno about LEDs, but CFLs don't last

[Khyber]

"Because I find it hard to believe in 3 months the extra few dollars/bulb is offset in energy savings."

Welcome to California, where we pay more for power than just about anywhere else in the USA. And it's not even a flat rate per kWh.

Re:I dunno about LEDs, but CFLs don't last

[Rei]

Where's your controlled, statistically significant comparative study data? Or are we supposed to go on an anecdote? Because we do have lots of data - for example, here Consumer Reports talks about their testing results.

Mercury? Every bulb CF tested contained less than 5mg. Let's go with 4mg as our figure (even though some are under 2%). 17-44% of said mercury will vaporize if you leave it sitting around for 8 hours. Let's say you clean it up and 10% gets into your air, which is probably a gross overestimate. What percent of that will you breathe and have actually get incorporated into your body? Probably in the low single digits, but lets be pessimistic and say 20%. So 80 micrograms. The mercury of a mere 1 1/2 cans of tuna.

But wait, there's more. The mercury in CFLs is "inorganic" (metallic, unbound) mercury, while mercury found in food is almost exclusively "organic" (methyl and dimethyl mercury). "Organic" mercury, being much more bioavailable, has many times worse health consequences per microgram.

The short of it? Don't stand in a closet and smash dozens of CFLs and then fan them while hovering over them and breathing deeply for a day or so. Otherwise, you're fine.

Re:I dunno about LEDs, but CFLs don't last

[Bing Tsher E]

I'm not sure what's more repelling about your comments in this topic. Your incorrect explanations, or the way you flame other people for their incorrect understanding.

  Suffice it to say you don't know shit about what you're talking about. For instance, the line drops on a two meter cable run of DC or AC are going to be the same. (We're talking 60 hertz, right? There will be significant drop on a 400Mhz voltage if the line isn't impedance matched.)

I'd say "go take a class about electronics" but judging from the tone you take here you'd spend the whole time arguing with the instructor thus fucking over the other class attendees. So just sit in your darkness. But shut up.

Re: I dunno about LEDs, but CFLs don't last

[ericloewe]

An LED would have little problem running from a PWM dimmer, as long as the eletronics around it don't blow from it.

The diode itself couldn't care less about the voltage waveform, as long as it stays away from the breakdown voltages.

Re:I dunno about LEDs, but CFLs don't last

[nblender]

For dimming, you might look at the better LED lights as some of them are dimmable - in Europe they are quite common, and if dimmable, explicitly say so.

The Cree brand bulbs I've been buying are pretty terrific and they are nicely dimmable. But i've noticed a problem with the dimming. I presume they dim by doing some sort of PWM to change the duty cycle and thereby result in some dimming... With the Leviton slide dimmers I've been buying, and set to full brightness (slider all the way up) the Cree bulbs have a flicker with a period of about 6 seconds. As in, every 6 seconds, the light will turn off for approximately 50-70ms. I have one circuit that has a couple of sockets. I put a Cree bulb in one socket, and a Phillips bulb in the other socket. At full, the Cree still exhibits the problem while the Phillips does not. However, the Phillips doesn't dim correctly. Whereas the Cree will dim in a nice linear sort of fashion, the Phillips will dim about 20 percent for the first portion of the slider, and then will maintain that brightness until the slider gets sufficiently far down and then the Phillips just turns off...

I have a Sylvania that just doesn't dim at all.

I think that covers all three of the major manufacturers. So far the Cree is the best except for the iritating 'blink' at full brightness... All of my Crees exhibit the same symptom on different dimmer switches throughout the house. If I put two Cree's in two sockets on one dimmer, they both blink, but at slightly different periods.

"We're not there yet".

Re: I dunno about LEDs, but CFLs don't last

[jeffmeden]

You don't PWM an AC waveform. You PWM a DC output (as it has no waveform, it peaks and stays peaked.)

PWM is the most convenient term (the OP even stated such) and given a fixed frequency, you can think of a thyristor dimmer (the kind in question) as strictly a PWM tool (since each position of the dial corresponds to a specific on-time:off-time ratio), the only difference being that the duty cycle does not correlate 1:1 with the output power (you have to integrate since you are clipping part of a sine).

Re: I dunno about LEDs, but CFLs don't last

[DamnOregonian]

You don't PWM an AC waveform. You PWM a DC output (as it has no waveform, it peaks and stays peaked.)

You do in fact PWM an AC waveform for dimming (less fires that way). You're cutting apart a sine wave, so the result is less obvious than simple DC duty-cycle, but it is PWM nonetheless.

No, PWM dimming is caused by a PNP transistor circuit, not a capacitor.

Or this fantastic invention called a thyristor. You have probably encountered one of them in your average wall dimmer.

Re:I dunno about LEDs, but CFLs don't last

[Nimey]

Innumeracy and a reflexive resistance to "the government telling me what to do".

Re:I dunno about LEDs, but CFLs don't last

[CastrTroy]

Water service kind of works a lot like internet service when you look at it. The main costs are in providing service, and how much you use doesn't really affect the overall cost of receiving that service. I lived in a town once that had fixed water bills for all houses. I think they saved a lot of money by not having to have someone drive around checking all the water meters every month, or even paying for installation and upkeep of water meters. This was in the days before the internet though, so you'd actually have to send someone around to check the water meter if you wanted to bill them.

Re: I dunno about LEDs, but CFLs don't last

[serbanp]

No LED-based lighting fixture uses a voltage regulator and a series resistor to limit the current through the LED.

The LED is driven by a current regulator; yes, it usually has a small sense resistor to measure and adjust the driving current, but that's something totally different from a resistor used to limit the current.

Remember, the 'D' in LED stands for (semiconductor) "Diode", which means that in forward conduction it has an exponential relationship between current and voltage. The only way to control the brightness is to control the current, you can't rely on the equivalent "resistance" (i.e. Vd/Id).

The premature failure of the LED bulb is caused exclusively by the embedded electronics degrading at high temperature. Since the LED is still dumping a lot of heat when working, the heat must be dissipated (hence puny LEDs and bulky, machined Al heatsinks) or the electronics get fried.

Guess what the drive to reduce the manufacturing cost will do to the quality of the LED bulbs?

Re: I dunno about LEDs, but CFLs don't last

[RabidReindeer]

Unless you have a very big heavy-duty resistor (rheostat) in your wall dimmer, it's not directly controlling the AC line. You have a variable resistor modulating the duty cycle of the A/C by biasing the control line to a thyristor - typically an SCR or triac. An SCR can modulate one size of the power sine, a triac is 2 SCRs back-to-back, so it can modulate both sides.

A typical circuit has the resistor connected to a capacitor across the power legs of the circuit. Do you know what a capacitor is? The junction of the resistor and capacitor typically connects to a diac so that the trigger will be essentially digital rather than analog. When the diac fires, it triggers the gate of the 3-terminal thyristor. By adjusting the resistance, hence the charge time of the capacitor, the point in the A/C sine wave where the thyristor fires and cuts off the power can be moved earlier or later in the cycle, thus regulating the total power output at the expense of making the sine wave a dirty mess. This is why dimmers often cause radio hash. But they're fairly energy-efficient.

Yes, I'm being a snot. You see, I really DO know what I'm talking about. I've not only built them, I've repaired a few as well.

A rheostat would be the blunt-instrument approach and not only is bulky, but also gets hot, since it's a resistor carrying a lot of wattage.

Re: I dunno about LEDs, but CFLs don't last

[RabidReindeer]

There's 2 ways to "dim" an LED. One is to reduce the input current. The other is to run it via pulsed DC instead of of continuous current. It's not dimmer - in fact, you can run higher current so it could be brighter. But just like motion pictures and TV, the percieved effect of this strobing is a brighter or dimmer light.

Light bulbs dim by taking advantage of the fact that the filament doesn't immediately go black when the current is removed. So by switching on and off rapidly, you get an adjustment of the average filament temperature.

LEDs, of course, don't do that. Although white LEDs are typically UV downshifted in spectrum via phosphors, and phosphors do have residual glow so they'd be less obviously strobing. My overhead LED bulbs do fade slowly when I turn out the lights.

If you're modulating an LED via power line duty cycle (PWM), it may or may not work since the AC is being converted to DC and if the convertor has a capacitive or inductive aspect, it will smooth out the resulting DC voltage, which doesn't help, since LEDs are being driven by the current. They need a clean on/off strobe or current regulation to dim (or "dim") effectively.

Re:I dunno about LEDs, but CFLs don't last

[russotto]

Chances are your dimmers and bulbs just aren't compatible. There are at least four kinds of dimmers out there -- those labeled for incandescent only, those labeled for magnetic ballast, those labeled for electronic ballast, and universal.

These labels are typically wrong, of course (nothing's ever easy).

An incandescent-only dimmer is a leading-edge (forward) phase cut dimmer. It works by turning on the power partway into the sine wave, cutting off the rising edge. It requires only two wires (hot and load), and obtains power for its own use by using the low-resistance path through an incandescent filament when it is off. Generally works poorly if at all with an LED fixture.

An electronic ballast dimmer is a trailing-edge (reverse) phase cut dimmer, and works by turning the power OFF partway into the sine wave. It requires a neutral wire as well as hot and load. Originally intended for low-voltage halogen fixtures using an electronic ballast.

A magnetic ballast dimmer is also a leading-edge (forward) phase cut dimmer, but requires three wires. Originally intended for low-voltage halogen fixtures using a magnetic ballast.

A universal dimmer is a two-wire forward phase cut dimmer that is supposed to work well with both types of ballast, but in practice just sucks.

Your LED lights will likely dim properly with either an electronic ballast dimmer, or a magnetic ballast dimmer (even though the LED certainly uses an electronic ballast), but not both, and will work poorly or not at all with the other types. And of course if you have multiple brands of LED they could require different types.

Re:I dunno about LEDs, but CFLs don't last

[russotto]

Like most complaints about the government that I see on Slashdot, this never happened. They set energy efficiency standards for lightbulbs, that's it. Some companies decided to meet those requirements with CFLs, some with LEDs, some with high efficiency incandescents.

There are a very few incandescents which meet the current interim standards. There are none which meet the final standards, which were in fact chosen knowing that. Claiming it was just a matter of setting efficiency standards rather than banning the incandescent bulb is sophistry.

Its not the CFL/LED

[Charliemopps]

the CFL/LEDs last forever... but most don't operate off of 120/240 volts. So there are transformers in the base that ramp the voltage up. The transformers do NOT last 50k hours. That's what burns out.

Re:Its not the CFL/LED

[itzly]

CFL have electrodes at the end of the tube, which will fail over time. You can tell by the fact that the ends of the tube start to get black.

I'm sold on LED bulbs...

[PortHaven]

I've tried 3 brands.

1. One cheap generic (long center square stack with LEDs), this model failed quickly and had poor light.
2. Discount sale LED bulb at Home Depot, with a soft rubber like coating on the exterior. Poor non-uniform light, returned.
3. 3 pack of LED bulbs from Costco. 60 watt replacement. These bulbs have about 50% more lumens than the CFLs. I've had one failure out of about 20+ bulbs. The light is in fact brighter and more uniform than even my incandescent bulbs. At $9.99 for 3 when on sale at Costco. I've decided that ALL my bulbs will be replaced by these LEDs.

(Oh and the one failed bulb was replaced by the the manufacturer.)

Re:I'm sold on LED bulbs...

[s_p_oneil]

I didn't find my favorite brand at Costco, but I second this. I also recommend people try a few different kinds in different light fixtures in their house before deciding. I have a few bulbs that are a real pain in the ass to change (e.g. so high over my staircase or outside that I need a pole to change it), and those were the first to be switched (in the hopes that I never have to change them again).

The initial cost can be high, but if you only buy 1 or 2 a month, the cost is spread out, and your electricity bills should go down gradually over time. As a gradually switch over, I also find myself needing to buy less and less light bulbs over time. The CFL's I was buying were burning out way too fast (faster than incandescent it seemed). I don't know when the first LED's I bought will fail, but so far every last one I bought is still doing fine. A few brands/types I tried give off weak lighting, so I stuck them in fixtures that have 3 bulbs to lessen the impact.

Re: I'm sold on LED bulbs...

[Richy_T]

On the downside, you did miss a great opportunity to play out a Monty Python sketch.

TL;DR

[Anonymous Coward]

So at the end of this whole pointless post you get to your thesis: "there is no denying that these far more technologically sophisticated products offer tempting opportunities for the inclusion of purposefully engineered life-shortening defects."

Thefuck? Cheap shit, made in countries with little to no regard to quality standards, is going to break down faster than you want it to. Not faster than it should, just faster than you want it to. Buy your bulbs from someone who knows what ISO means and you might have better luck. There is no conspiracy here, fluorescent bulbs have been used for decades in the commercial lighting business, and LEDs have been used for over a decade, with no ill effect. Pay what it's worth for a decent lamp and you will get the appropriate longevity. Pay .99 for a sealed fluorescent coil, electronic balllast, and enclosure or 3.99 for a 10W power supply, LED array, heatsink, and enclosure and you will get shitty performance.

Re:TL;DR

[itzly]

There's no guarantee that the bulb that is sold for a high price isn't actually the same piece of cheap shit inside, just sold for a greater profit.

Can we trust the numbers

[MyLongNickName]

At 50,000 rated hours, almost no light in my house should burn out in my lifetime. Yet, my experience with CFLs is that they don't last nearly as long as the advertised life. So the issue becomes a question of whether we can trust the numbers that the manufacturers put out.

Does anyone have a link to a reputable 3rd party investigating the true lifespan of CFL vs LED? By brand? By usage pattern? I'd invest in the bulbs if I were reasonable sure that I would get my money's worth.

My Compact Flurorscents die

[gurps_npc]

way too early.

I want a required "Good till" date printed on them, that guarantees they last at least X days, just like soda.

Yeah, most of them will last a lot longer than the printed date, because chances are you won't buy them and install them on the day they make them.

But still, if a curly bulb is supposed to last 5 years, and it dies one year after you install it, there should be an easy way to get a refund.

CFLs are supposed to last longer?

[Ronin Developer]

NOT in my house and NOT with the expected life expectancy listed on the packing! Of course, due to power fluctuations (we still have a 100A feed vs 200A and overhead wires), we constantly have bulbs burning out. Yes, major portions of the house wiring have been redone.

If they had surge protection in the bulbs, they would probably last a lot longer and I would get my money's worth due to the cost vs power savings (7W equivalent to 75W incandescent). My kids leave lights on all day...so it makes a big difference over time.

We just put in a "sunlight" white LED bulb in the kitchen to replace a CFL. Holy crap is that bright yet energy efficient!

To be really efficient

[rossdee]

Get a LED headlamp and only light up the area of the room that you are looking at.
You'll only need 3 or 4 watts that way.

Completely converted house to LED, 3 have died.

[bigattichouse]

So, the "return" process is iffy. I didn't have my receipt when one died and I took it back to Lowe's for an exchange of the same model (Phillips).. they said they couldn't be sure it was under warranty, I told them it was supposed to last 10 years, and they had only been selling them for a few months. They begrudgingly swapped it out.

Anyway, the other 2 bulbs, I decided to pull them apart. I dug out the silicone potting, and found the failure was in a large capacitor, visibly bulging. I haven't had time to replace the bit - but I'm pretty sure that's all that blew on it. Tested the individual LEDs and they are fine.

So both failures were due to purchasing the cheapest possible components, specifically a "largish" (like 0.3uF 200v) capacitor. My guess is that there was a larger cap that would handle the load, but they needed to reduce the size. Initiating the failure was probably one or more line spikes.

Re:Completely converted house to LED, 3 have died.

[Ogive17]

Many Bothans died to bring you LED light.

Re:Completely converted house to LED, 3 have died.

[Khyber]

200V cap? You want minimum 250V for a 120V line.

Costco bulbs almost too bright

[swb]

I've bought about a half-dozen of the Costco bulbs and so far no problems. You're right, they are bright, almost too bright!

The ones I've used the most have been replacements for the typical 65 watt recessed can lights. I have a mix of Philips Halogena 45 watt (reduced power halogen, "same" output as a 65 watt), normal incandescent and CFL. The Costco LEDs are by far brighter than any of the others, in some cases they seem almost too bright.

They all seem to have decent dimming performance, too, although I mostly use a dimmer on my bedside light. Not super linear in dimming, but acceptable for the most part.

Re:Costco bulbs almost too bright

[Maury Markowitz]

I noticed this too. I bought Ikea and Cree "60 watt" rated bulbs, and they were much brighter than the 60W Sylvanias and Philips I took out of the fixture. A LOT brighter, I took a photo:

http://maurysrandomproductreviews.wordpress.com/2013/12/05/leds-cree-and-ikea-ledare-60w-replacements/

I had a three-lamp fixture in the kitchen and after replacing the bulbs the room was much brighter overall. Too bright at first, but we quickly got used to it.

This idiocy again

[Anonymous Coward]

Nobody would have profited from longer lasting lightbulbs.

In 1000 hours, a 100W lightbulb costs an order of magnitude more electricity than the puchase price. You can easily increase the lifetime of such a bulb to 10k hours, simply be reducing the operation temperature by 20% or so.

Of course, this halves lumens/W, to to get the same brightness, you need 200W of power - which means you pay twice as much over those 10k hours as if you have bought 10 100W bulbs to last that time.

High power lamps, for example in flashlight, used to be specified to operation times below 100h, because this allowed them to almost double the battery runtime...

MPAA, RIAA, TelCos, ISPs, Airlines, ...

[LOGINS SUC]

While this may be an intellectually interesting story, I hardly think we need to consider a 100 year old defunct cartel. I'm far more worried about modern cartels, consider those in the title and there are many other besides - investment banks, teacher federations, De Beers... We need only glance outside our own personal bubbles to recognize massive manipulation starting with advantageous legislation perpetuating inefficient business models and see consumers are exploited from all directions by cartels.

bathtub curve applies

[rapiddescent]

I replaced 50x GU10 50W bulbs for 3W LED equivalents (no longer available) that were more expensive (slightly warmer light). Here in Scotland, energy prices are more than the USA - so the initial investment of 50 bulbs cost 20x as much as the GU10's burt due to the lower wattage (3W vs 50W) would pay back in 2 years (which they have) from lower overall electricity prices.

However, we've had a lot of failures. So far over 10% of the 50 have failed - usually blowing the main house fuse when they went. So the porblem at the moment is there is no way to assess the failure rate for LED household bulbs. This is having quite an impact on the payback period for the bulbs. .

LEDs

[Geoffrey.landis]

The Department of Energy had a pretty rigorous test regimen set up for testing LED bulbs.
http://www.lightingprize.org/6...
What is needed is a good (and trustworthy) rating agency to test and qualify the bulbs.

But, of course, everybody wants to buy the cheapest ones, not ones tested to long lifetime.

For what it's worth, I have about 60 LED bulbs in my house, from about fifteen maufacturers. So far, four have failed.

Further - no viable light bulb replacements will work with dimmer switches (Which my house has many).

That was true five years ago-- these days it seems all of then are rated to work with dimmers . I have some Philips LED bulbs on a dimmer in the dining room-- they work fine.

Re:LEDs

[Jawnn]

The Department of Energy had a pretty rigorous test regimen set up for testing LED bulbs. http://www.lightingprize.org/6... What is needed is a good (and trustworthy) rating agency to test and qualify the bulbs.

Yeah, like the FDA...
[rimshot]
Jeez, I crack myself up.

Do longevity tests account for crappy power?

[GGardner]

CFLs, of all brands, have not lasted nearly as long as advertised at my house. I don't think I've had any last more than a year. However, the power at my house is terrible -- lights flicker and dim several times a day, and I completely lose power several times a year. All the computers are on UPSes, but it would be prohibitive to put all the lights on one. Old fashioned, incandescent light bulbs seem much more robust than at least CFLs, and I'm not too excited to test LEDs. So, do any of these lab tests which promise CFLs and LED that last for year test with real-world power sources?

Re:Do longevity tests account for crappy power?

[PhoenixFlare]

Are you implying that people who don't have awful-quality power aren't running with "real-world" power sources? What you're describing isn't normal by almost any stretch of the imagination.

Personally i've been using CFLs in all our lamps for close to 10 years, I think maybe 1 or 2 have failed out of about 9 or so during that period.

LEDs need much closer tolerances

[gelfling]

LEDs are FAR FAR more sensitive to bad manufacturing tolerances which is why cheaper, relatively speaking, units are such shit. On the other hand a quality unit from Cree is probably going to last as long as it should. I would also stay away from dim-able bulbs since they require more circuitry. Same with dim-able CFLs. The transformers are fragile enough as it is.

The Government also ruined my washer and dryer

[zerofoo]

I recently tossed a set of 4 year old Bosch HE front-loading washers and dryers. The washer was full of mold and the dryer needed 2-3 cycles to adequately dry clothes.

I performed all the preventative steps periodically running a hot cycle, running bleach and cleaning agents through the machine, leaving the door open after every use, cleaning out the dryer vent twice per year...etc.

None of it mattered. The service tech says these problems are common to all HE machines. They simply do not use enough water at a high enough temperature to adequately flush the machine. They dryers are also garbage as the manufacturers are forced to use small burners and short cycle times to meet energy efficiency requirements.

My mom's 25 year old Kenmores washed and dried clothes without complaint for many years - now we are saving the planet by putting crappy appliances in a landfill every few years.

I finally opted for a non-HE washer and dryer (Speed Queen if anyone cares). They are old-school commercial-duty devices. No WiFi, no touch screens, no weather or twitter feed - simply clean clothes in half the time.

Yup they use more water and gas, but I wouldn't have it any other way.

I was told that by 2016 or 2017 these units will no longer be made thanks to more efficiency regs. It's madness.

How does this relate to CFLs and LED bulbs - well - I like my LED bulbs - my CFLs were almost all garbage that lasted a year or two. Most CFLs that I bought didn't last anywhere near their rated 7 year life.

It would have been better to simply give people economic incentives to buy the more efficient bulbs instead of being forced to buy the bulbs.

Re:The Government also ruined my washer and dryer

[andyring]

I worry/wonder about this too. We have a basic no-frills washer and dryer. I bought the washer probably 6-7 years ago at BestBuy, a Maytag model I think. Dryer was given to my wife and I (used) about 4 years ago. They do the job well, although I've had to disassemble the washer a couple times when something tiny (wife's hair tie, or little kid sock) somehow got through the internal plumbing and stuck in the water pump. But it's no big deal to do it.

My wife really wants a set of those fancy schmancy shiny front loaders that run about $1,000 each. But I've heard too many horror stories about them dying like yours, or expensive electronics dying, etc.

And then I hear my grandmother (bless her soul) has had literally the same washer and, I think, dryer since about the 1940s. She died this spring but my grandfather is still alive and at the same house so it's still getting used. They raised 11 kids too!

I don't care how "efficient" an appliance is. If you can run it for 70+ years without trouble, that's the ultimate in efficiency.

LEDs should be date stamped

[swb]

I think a lot of people who have over the years tried to use more energy efficient bulbs found that their actual lifespan was all over the map. I'm sure this has led to a lot of people being turned off and going back to incandescents.

When they decided to phase out incandescents they should have made bulb makers date stamp the bulb with a "good until" date AND mandate that any bulb burned out before this date is eligible for a free, over-the-counter replacement.

This would have greatly improved consumer confidence and forced manufacturers to be either more realistic about lifespans or not skimp on components.

What I've found odd about CFLs is that they seem to fail strangely with no discernable pattern. I've gotten some to last in extreme places (outdoors, through subzero winters) and had several fail in places you think they wouldn't, indoor lamps with good ventilation.

Life-shortening defects

[RDW]

But there is no denying that these far more technologically sophisticated products offer tempting opportunities for the inclusion of purposefully engineered life-shortening defects.

Like, for example, a $1000 phone with a battery the user can't change...

your washer has a cycle time?

[YesIAmAScript]

This is crazy. Just because the service tech told you something doesn't make it true.

I have an HE washer/dryer that predates yours. I got it them 10 years ago and they're still going strong. It was the Maytag Neptune, which was the first HE washer on the US market. There was a flaw in the door latch on the first year or two model but I was lucky to avoid that, mine is from just after that.

The washer works fine, although it is nice if you leave the door open for a day once in a while to dry it out in there otherwise, since the door is sealed, any moisture left in the drum after a cycle just sits there until next time you use it. It doesn't have anything to do with hot water, hot water only stays hot for a short time and hot water doesn't kill mildew anyway, if it did you wouldn't need to scrub or bleach the grout in your shower! Later models from Samsung and LG don't have this problem.

The dryer doesn't even have cycle times. It just runs until the clothes are dry. It does this using a dryness sensor, the same type which has been around since 1980 or so. If you do run it on a timed cycle, you can adjust the time it runs in one minute increments. So I have no idea what your tech was telling you about mandating short cycle times or burners that aren't hot enough.

Re:your washer has a cycle time?

[Anonymous Coward]

This is crazy. Just because the service tech told you something doesn't make it true.

Your rebuttal to a service tech, whose sole job is to know every detail about HE washers and driers, is to bring up a single instance of anecdotal evidence that is not even about the same brand. Talk about crazy.

incandescents will be available for awhile yet

[roc97007]

"Rough Service" incandescents, designed for outdoor hard to reach places in harsh conditions, where CFLs are not appropriate and LEDs have not yet made inroads, are still available, cost about $2 apiece, have a rated lifespan of 10,000 hours, and are not affected by the ban on incandescents. Just sayin'...

How long the bulbs burn for is irrelevant...

[Anonymous Coward]

the important thing is how many people of a particular type that it takes to change one.

It's cold where I am

[tipert]

I don't mind the excess heat from my incandescent bulbs. It saves on power for the baseboard heaters.

LED early failure is rather common

[millertym]

I'm a huge LED fanboy. I've been buying only LEDs for the last 3-4 years. The early ones especially have not held up. I believe I only have 1 still functioning out of the 4 I bought in my first full room conversion. The newer ones - GE model 100w equivalent I typically get from Sam's Club - have been a lot better. I've only had 1 fail so far out of the 10 I have put into service. In any case, it's not a far fetched claim that a significant percentage of these lights last no where close to as long as they claim they will. A battle hardened customer service call should be able to replace those that fail early.

Re:Move away from the 120V screw-based sockets?

[russotto]

I'm surprised that more homes aren't equipped with house wide 12/24 volt DC, that would go a LONG way towards decreasing the costs/efficiency issues with LED's and many home electronics.

Um, no. Lower voltage means greater losses within the house, given the same wire thickness. Doing the step-down as close as possible to the use means higher efficiency.

A potential class action suit

[zerofoo]

Evidently enough people had similar complaints across many brands - enough to warrant a potential class action lawsuit:

http://www.slate.com/articles/...

The court decided against hearing the case, but there were over 23,000 complaints for one manufacturer. That tells me there is something wrong with the design of many HE machines.