• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

White LED's, will they become a mainstream lighting?

Jerboy

Banned
It hasn't been too long since white LED has emerged. I was reading thisarticle just a minute ago and it predicts that white LED's can compete with modern lighting technology in about eight years from now. The home electronics of 70's and 80's actually used miniature incandescent lamps as indicator lamps and LED has almost completely displaced incandescent in indicator lamps and it is displacing backlights and other concentrated spot lights with relatively small lightoutput, but not a whole lot more. I accredit LED's for their excellent potential as indicator devices, including VCR indicators, traffic signals and such as well as very focused small scale illumination such as keylights(Photon Micro) and flashlights.

Personally I believe they make excellent accent lights and small scale spot lights, yet I doubt they'll be used to light up offices and homes. LED's appears bright, because of it's ability to focus beam in tight spot. Their total light output is rather law and efficacy is not very high.

Some of the best white LED has an efficacy of about 30 lumens to watt and a 50lumen is among the largest size available today. Considering each 50lumen units cost $5 a piece today, it will be prohibitively expensive.


Comparison based on current technology

fluorescent:

four lamp F32T8, 86CRI, 4100Kelvin lamps with published output of 2950lumens.
electronic ballast driving them at 88% the rated output.

total output: 10384 lumens
input energy: 112watts
system efficacy: 92.7 lumens/watt
lamp life~20,000hrs(until failure. lumen maintenance at 8,000hrs=95%)
cost: $3 per lamp, $20 ballast, $40 fixture=$72.00
$0.0069/lumen

white LED using combination of 50lumen modules to get equivalent output:

total output: 10384lumens
346watts
30lumens/watt
lamp life~5,000hrs(lumen maintenance at 10,000 is only 50%)
207 modules@$5 each=$1035 just for the lamps

about $0.1/lumen

Based on these data, I can conclude that LED's are inefficient, expensive, low light output per module and short lived. I don't think LED's will replace ambient lighting anytime soon. It will take ALOT to beat gas discharge technology, techincally and economically.

What's your opinion?

 
efficiency (meaning $, that is: lumens per watt, or in business thinking lumens/$) is the
ruling consideration for commercial lighting, which is why
fluorescent rules. but i question the accuracy of 5000 hr lifetime for an led. thats
24 hrs/day for 200 days, and i know leds outlasting that by a factor of 10. which is why
theyre used on traffic lights: reliability.
>>>>>another close competitor: hps (high pressure sodium)
similar to that yellowy light (low pressure sodium) in most parking lots, but whiter. efficency close to that
of fluorescent even today, even a 20% improvement would make it way more popular.
I just put in a 70w hps light on the side of the building i work at (a nursing home) and its lumens are way
above a 70w incandescent bulb. incandescant is suck: lowest lifetime, lowest efficency, 100
year old technology invented by Edison himself in the 1890s.
 
White LEDs are an interesting technology, and one that has been heavily researched - dramatic gains have been made especially in very recent times.

However, there are problems with LEDs still - colour quality is still poor, giving the strange colour shifts that fluorescent lights give (due to the very 'peaky' spectrum), colour temperature is widely variable - My new telephone uses white LEDs as illumination, and every one is obviously a different colour to the others. Older LEDs suffer badly from aging (especially in modules - this is due to inadequate thermal management predominately - a couple of weeks of a unusually hot weather can dramatically reduce the LEDs quality if they are operated during that time - this very short lifetime is probably the result of compromising design allowing overheating).

In terms of electrical efficiency, LEDs are not very impressive - they are better than most incandescent, and roughly equal to high quality tungsten-halogen. Conventional fluorescent are far better. For 'single wavelength' LEDs, efficiency is much better potentially better than fluorescent, although not as good as low-pressure sodium.

State of the art LEDs, like the lumileds 'luxeon star 5W', can produce 120 lumens from a single LED (electrical power 5 W) - with adequate thermal management, lumen maintenance is approx 80% at 50,000 hours. Such a device roughly compares in output to a 15 W 'night light' bulb. Similar performance is available in the similar brand new lumileds super-flux LEDs (approx 5 lumens for 0.2 W). As to price, LEDs aren't cheap - a 5W luxeon is about $35, and a super-flux white is about $1 (both in quantities of 1k), this price excludes to cost of heatsinking, etc.

However, LEDs are small which allows innovative fixture design, and their long lifetime allows the LEDs to be integral to the fixture (so the whole fixture is discarded when the LEDs eventually fail) - in hard to access places, this can be cost effective.

LEDs still have a long way to go before they are a mainstream lighting technology, but improvements have been fast - who would have thought it possible to use LED headlights on a car? (one was demonstrated late last year by lumileds). As to how long it will be before they become something you can buy, who knows?
 
Originally posted by: capybara
efficiency (meaning $, that is: lumens per watt, or in business thinking lumens/$) is the
ruling consideration for commercial lighting, which is why
fluorescent rules.

So far new T8 electronically ballasted fluorescent are the cheapest / lumen, very high LPW, and low lamp price and longest lumen maintenance of most white lights. (Philips Adv ult=95% maintenance@9,600hrs)

but i question the accuracy of 5000 hr lifetime for an led. thats
24 hrs/day for 200 days, and i know leds outlasting that by a factor of 10. which is why
theyre used on traffic lights: reliability.

That's not 5,000hours until failure. White LED's is a different than a single color monochromatic LED. White LED's are fluorescent lighting. It has a blue LED as a source and phosphor built into the package fluoresceces when stimulated by blue light.

Our typical white LED loses 50% of output at 5,000hours due to phoshpor degradation. This is not a concern for single color LED not using phoshpor. Also, output degradation is not a big deal in my indicator purposes, but it is deadly in ambient lighting.

>>>>>another close competitor: hps (high pressure sodium)
similar to that yellowy light (low pressure sodium) in most parking lots, but whiter.

efficency close to that
of fluorescent even today, even a 20% improvement would make it way more popular.
I just put in a 70w hps light on the side of the building i work at (a nursing home) and its lumens are way
above a 70w incandescent bulb. incandescant is suck: lowest lifetime, lowest efficency, 100
year old technology invented by Edison himself in the 1890s.

Most parking lot lights are high pressure sodium. CFL's can eventually take over HPS if lamp life and efficacy could be improved somewhat.

HPS has CRI of 21, efficacy of 80-90lumens per watt at 100W size.

100W ceramic metal halide has CRI of 93, initial efficacy of 90LPW. 80% lumen maint. at 6000hrs. The emission intensity of HID is hard to match with LED's.


 
Originally posted by: Mark R
White LEDs are an interesting technology, and one that has been heavily researched - dramatic gains have been made especially in very recent times.

Yes alot of gain has been made, but we could have hit the limit already. Much like internal combustion engine. Alot has been done to increase efficiency and reduce emission, but conversion efficiency from heat to mechanical hasn't gone much above ~25% in automotible sized engines even though it can be around ~50% in humoungous vessel sized diesel engines with piston stroke larger than our height.


However, there are problems with LEDs still - colour quality is still poor, giving the strange colour shifts that fluorescent lights give (due to the very 'peaky' spectrum), colour temperature is widely variable

Fluorescent lamps don't have bad color shifts and color quality is far better than ten years ago. The problem with LED advocates is they compare their new LED's with 1980's Cool White fluorescent lamps. Cool white lamp has a ghasty green shade and CRI of 63 rather than comparing to modern tri-phosphor lamps capable with CRI of ~85.


My new telephone uses white LEDs as illumination, and every one is obviously a different colour to the others. Older LEDs suffer badly from aging (especially in modules - this is due to inadequate thermal management predominately - a couple of weeks of a unusually hot weather can dramatically reduce the LEDs quality if they are operated during that time - this very short lifetime is probably the result of compromising design allowing overheating).

So this is one thing that still hasn't changed. Semiconductor device has an inherent thermal degradation characteristic. White LED's could very well have sudden failure from electrical surge as well.

In terms of electrical efficiency, LEDs are not very impressive - they are better than most incandescent, and roughly equal to high quality tungsten-halogen. Conventional fluorescent are far better. For 'single wavelength' LEDs, efficiency is much better potentially better than fluorescent, although not as good as low-pressure sodium.

They're about as good as automotive headlamps, which are more efficient than standard lightbulbs.

State of the art LEDs, like the lumileds 'luxeon star 5W', can produce 120 lumens from a single LED (electrical power 5 W) - with adequate thermal management, lumen maintenance is approx 80% at 50,000 hours. Such a device roughly compares in output to a 15 W 'night light' bulb. Similar performance is available in the similar brand new lumileds super-flux LEDs (approx 5 lumens for 0.2 W). As to price, LEDs aren't cheap - a 5W luxeon is about $35, and a super-flux white is about $1 (both in quantities of 1k), this price excludes to cost of heatsinking, etc.

Not gonna be practical in a loooong time. That's 24LPW base efficacy. Include the driving circuit, the system efficacy will be ~20LPW, less than 1/4 that of fluorescent technology.

Fluorescent lighting can achieve an efficacy of ~90lumens per watt including loss within ballast.


LEDs still have a long way to go before they are a mainstream lighting technology, but improvements have been fast - who would have thought it possible to use LED headlights on a car? (one was demonstrated late last year by lumileds). As to how long it will be before they become something you can buy, who knows?

Car headlights and projectors need very precisely focused light source to produce a good light pattern on the surface. Xenon-metalhalide lamps for car headlamps has an output of ~3,000lumens each in a thumb sized capsule. Could you fit 3,000lumens worth of LED's into a thumbsized capsule?



 
Yes, they will be a major source of "mainstream" lighting. But not necessarily as a primary indoor lighting source.

Smaller, more specific applications (especially where the colored light is not a problem) are ideally suited for LED lighting. The little LED keychain flashlights are a great example, and I would consider them mainstream already. Another application is emergency red "night vision safe" lighting in airplane cockpits, submarine control rooms, or even automobiles.

Now that these bright LEDs are getting cheaper (thanks primarily to great work with Indium Gallium Phosphide silicon substrates), we will see more uses for them. A quick check of an electronics parts supply catalog shows that you can get a high output 9000 millicandela red-orange (624 nanometer) 2.0V LED for $0.28 in quantities of 100 or more. Not bad. Obviously the shape of the packaging of the diode determines how well the light is actually focused.
 
Originally posted by: kylef
Yes, they will be a major source of "mainstream" lighting. But not necessarily as a primary indoor lighting source.

I meant primary area illumination lighting when I said mainstream.

Currently electric discharge lamps dominates majority of non-residential indoor lighting and over 99% of outdoor lighting.

Indoor,
low ceiling: fluorescent
high ceiling: metal halide
high ceiling warehouse: metal halide or high pressure sodium

outdoor,
parking lots: metal halides and high pressure sodium

street lights: metal halides, mercury vapor, high pressure sodium and some highway lights are low pressure sodium(180+ lumens per watt)


Residential lighting is still largely incandescent, but CFL's are kicking in.

I doubt they'll be displaced in next 25years. If they're displaced and bulbs didn't need replacement for 100,000hours, all these people in the lighting maintenance and bulb manufacturing will be out of jobs.



Smaller, more specific applications (especially where the colored light is not a problem) are ideally suited for LED lighting. The little LED keychain flashlights are a great example, and I would consider them mainstream already. Another application is emergency red "night vision safe" lighting in airplane cockpits, submarine control rooms, or even automobiles.

These are specific applcations and I wouldn't call them mainstream lighting source.


Now that these bright LEDs are getting cheaper (thanks primarily to great work with Indium Gallium Phosphide silicon substrates), we will see more uses for them. A quick check of an electronics parts supply catalog shows that you can get a high output 9000 millicandela red-orange (624 nanometer) 2.0V LED for $0.28 in quantities of 100 or more. Not bad. Obviously the shape of the packaging of the diode determines how well the light is actually focused.

Bright simply means intensity of light and is not the actual output. That's why 15degree view LED is brighter than 30 degree ones.


 
That's not 5,000hours until failure. White LED's is a different than a single color monochromatic LED. White LED's are fluorescent lighting. It has a blue LED as a source and phosphor built into the package fluoresceces when stimulated by blue light.
Couldn't you make a white LED without that problem by putting in a red, green, and blue emitter (or whatever its called) right next to each other? Then you wouldn't have to deal with the phosphor problem.
 
Couldn't you make a white LED without that problem by putting in a red, green, and blue emitter (or whatever its called) right next to each other?

This works, but means 3 devices instead of one, poorer colour quality and considerably lower efficiency (fewer lumens per watt).
 
I find it impressive how much you guys all know about lighting. But to me, it seems like you're all wrong. Lighting in people's homes is very very far down on their list of things to worry about. You can buy a 4-pack of incandescent bulbs for $2.25, and the electricity use is pretty small. If electric rates were about 3x what they are today, I can imagine lots of people switching to flourescent lighting in their homes. However, as it is now, it's simply not an issue with 98% of the population.

edit: Of course, the large-scale switching to more efficient light sources would make a large and very worthwhile difference in electricity demand, and therefore pollution and other factors. If the government or electricity companies had a large advertising/education campaign (possibly with a subsidy for those who switch... like free installation upon purchase of the fixtures or something), I can see a substantial amount of people switching. Measures like that can go well when people feel like they're contributing to a larger cause. "It's patriotic to use flourescent lighting" or something like that, while cheesy and stupid, would probably change some people's behavior if it was kept around long enough and a few high-profile changes were made (i.e. all the lights in the White House, U.S. Capitol buildings, Shaquille O'Neal's house, yadda yadda yadda...).
 
A couple years ago you could have grouped me in the "other things to worry about...I'll keep using incandescent" category. Then we changed office buildings and our new building had poorer environmental controls. With four computers running in my office, there were quite a few times where my office temperatures would rise to 85 deg or higher. If I used a 75-100W incandescent bulb in here (I don't use the overhead fluorescents because they're far too bright for an office this size) then the temps would go even higher. So I went the CFL route and I've largely been satisfied. So it wasn't the lower operating costs since electricity in central Kentucky is dirt cheap...it was the reduced heat. Longer bulb lifetime was also a factor.

CFLs still have issues though. When you replace a regular incandescent bulb, you generally don't have to worry about color temperature. If I replace a GE 60W soft white bulb with a Philips 60W soft white bulb, I can reasonably expect the light color to be the same or very close. This just isn't true with CFLs. I have CFLs that range from a warm incandescent-like Lights Of America bulb (used in my office) to various CFLs at home that range from purplish-white to bluish-white. Different brand CFLs are simply not interchangeable colorwise. Further, not all CFL manufacturers even print the color temperature on the label so you can't tell in advance.


 
I've been thinking of starting to replace my incandescents with CFLs, but I have question.... are CFLs as annoying as overhead fluorescent lighting? By "annoying", I mean the flickering on/off at 60hz. I've heard rumors/stories that this cuts brain waves, etc, and, while I don't know if I believe them or not, I do much prefer the constant light of an incandescent.

(Yes, technically, an incandescent turns on/off @ 60hz, but since the filament can't cool down fast enough, it's producing continuous light).

Also, any recommendations for brands/models? I'd like something that produces the "warm" spectrum of an incandescent bulb. Thanks in advance!
 
Originally posted by: Draknor
I've been thinking of starting to replace my incandescents with CFLs, but I have question.... are CFLs as annoying as overhead fluorescent lighting? By "annoying", I mean the flickering on/off at 60hz. I've heard rumors/stories that this cuts brain waves, etc, and, while I don't know if I believe them or not, I do much prefer the constant light of an incandescent.

(Yes, technically, an incandescent turns on/off @ 60hz, but since the filament can't cool down fast enough, it's producing continuous light).

Also, any recommendations for brands/models? I'd like something that produces the "warm" spectrum of an incandescent bulb. Thanks in advance!

Jerboy is the resident expert in lighting. (There are other knowledgable people too, but Jerboy somehow has an extreme love for it.)

Check out these threads: link1 this one talks about his setup w/ full-spectrum flourescents link3. If you use electronic ballast, you won't have flickering problems.
 
CFL's are good for several reasons. Roughly twice the lifetime of incandescents, less heat output, and less electricity consumption. the downfall is a shift in the color temp of the light (from a wamer temp, 2000K, to 3000 to 3500K, i.e. red light to blue light) radio frequency interference, and possibly flickering. Some CFL's may also have a brief startup time b4 they reach peak illumination output. And yes, they have cycles just like incandescents, although usually your eyes can't pick them up. Just look for bulbs that are close to full spectrum light, or have a warmer color temp, especially if the space your lighting has warmer colors. The best solution would be to mix it with some Incans, and some CFL's.
 
Roughly twice the lifetime of incandescents

Unfortunately. CFL's are supposed to last 6,000 to 10,000hours. That is about 6 to 10 times the life of standard A type incandescent. However, many reflector style R-x type lasts 5,000hours and often time, CFL will fail before the reflector lamp style lamps. CFL is getting alot of attention from utility companies and to appeal to consumers, their price needs to go down. You can have a CFL for ~$5 each these days, but pre-mature failure is VERY common. Some die well before incandescent.

less heat output, and less electricity consumption.

The product of increase in efficacy.

the downfall is a shift in the color temp of the light (from a wamer temp, 2000K, to 3000 to 3500K, i.e. red light to blue light)
You're talking about metal halide. CFL's or fluorescent in general has pretty constant chromaticity. Metal halide can swing 250-500K easily.

radio frequency interference and possibly flickering.
Especially so with substandard cheapies. There was a time CFL's were built with magnetic ballasts and these flickered like regular fluroescent lamps. Almost all the modern CFL's have an electronic ballast and they don't usually flicker, but substandard units has been known to flicker.



Some CFL's may also have a brief startup time b4 they reach peak illumination output.
'

This is inherent to CFL. Mercury is amalgammed to keep vapor pressure within desirable range and since mercury is absorbed into amalgam when cold, it takes a while before all the mercury is distributed evenly inside the discharge chamber.

And yes, they have cycles just like incandescents, although usually your eyes can't pick them up.

What "cycles" are you talking about?

Just look for bulbs that are close to full spectrum light, or have a warmer color temp, especially if the space your lighting has warmer colors. The best solution would be to mix it with some Incans, and some CFL's.

Full spectrum CFL's are rare and they come only in 5,000K plus.
Durotest FS CFL

 
Changing the subject slightly, my city just went and changed out almost all the traffice lights to big groups of LEDs. They look pretty cool, the color is brighter and sharper, IMHO, if one burns out no big whoop (although it seems the whole bank can go out, as this has happened at least once to my knowledge), and they use less electricity.
 
My only objection to LED traffic lights is they tend to have a narrower viewing angle than traditional traffic lights. In general this isn't a problem but there are times when you are outside that viewing angle. A couple scenarios come to mind: when you're waiting to turn right on red while the perpendicular-traffic light is green. With traditional traffic lights, it's easy to verify that the other light is still green (and hence you have at least a few seconds to start making your turn before you have to worry about oncoming traffic running you down) but it can be harder to tell with LED lights since you're sometimes outside the viewing angle depending on where the lights are located at the intersection.

 
Originally posted by: Geotpf
Changing the subject slightly, my city just went and changed out almost all the traffice lights to big groups of LEDs. They look pretty cool, the color is brighter and sharper, IMHO, if one burns out no big whoop (although it seems the whole bank can go out, as this has happened at least once to my knowledge), and they use less electricity.


Like I said, LED's are fine indicator lighting source. The original topic was about its future in mainstream as general illumination. WILL NOT happen until lumens-per-watt can exceed 85(the current efficacy of T8 fluorescent system) and cost becomes pretty close to that of fluorescent system.
 
General lighting applications are still a ways away, but specific industry applications are coming fast and furious lately. Development of high intensity green LEDs have lead to increased use in traffic signal applications (where red LED heads have had a foothold for over a decade). Now that green are available, a greater percentage of DOTs are converting signal heads over. There was a lot of resistance early on with just the red heads, not only because of the lack of green [with only red heads, the energy savings and reduced servicing vs cost ratio wasn't great], but because early designs had very poor thermal degradation [an LED bank sitting in an aluminum can during your average Southern summer could justifiably be termed one of your worst case conditions] and reduced output below acceptable norms in little more than two years. Newer designs are more rugged with greater intensity, wider viewing angles [sometimes assisted with an overall lens], and improved chromaticity matching with ITS requirements (used to be a signal engineer with North Carolina DOT, can't you tell?).

Interesting sidebar: amber LED heads never caught on because their use never met the expense. The second biggest push towards the use of LED heads (besides energy savings) was in reduction of service calls for lamp replacement. An average service call to an intersection can cost a division 100-200 smackers. Amber LED signals never caught on not because of chromaticity matching problems, but simply because amber incandescent heads, which are only a during a fraction of a typical traffic cycle, last many times longer than red and green heads, which spend a much greater percentage of time illuminated.

LEDs have made huge in-roads in EXIT light applications and are now the defacto light source for many emergency lighting products, replacing fluorescent. There are emergency products out there that have EXIT legends that are lit brilliantly and uniformly with only two or three discrete high intensity, wide angle LEDs!

High intensity LEDs are also being investigated for use in airport runway lighting applications, task lighting, and on and on...

My career has moved me into product safety certification, so I get to see new and interesting applications every day. One of the perks of the job.

Interesting thread! Keep it up!

Der Perfesser
 
(Slowly pushing this thread a little more OT 😉)

My EE friends & I thought it'd be cool to fill a ceiling with LEDs (RGB), then you'd have driver & a controller, and essentially turn your ceiling into a giant "display panel"... We figured it'd be very cool for uniform lighting (after all, it's the whole ceiling!), mood lighting, party lighting, etc. Not very practical, of course, but fun to think about 🙂

However, I hear now that they are working on OLEDs (o = organic, I think?) that would have applications very similar to this, but with much more practicality. I think that would be very cool, personally...

Just imagine the "party room" possibilities! 😀
 
they are highly efficient, and have long operating lives. some are in the 100000hr range. this means cheaper maintenance costs.

anyway, they will become mainstream when they are cheaper. lowest price i have seen is $1.35 each. some places sell them for $4 each.
 
Originally posted by: Mday
they are highly efficient, and have long operating lives. some are in the 100000hr range. this means cheaper maintenance costs.

anyway, they will become mainstream when they are cheaper. lowest price i have seen is $1.35 each. some places sell them for $4 each.

To my knowledge, white LED's are perceived as efficient, because they can produce an intense beam with such a little power. mcd is a measure of light intensity, not total luminous flux. It's the total luminous flux we're concerned about in general ambient lighting.

These 20lumen LED's exist, but they take one watt to operate and it still equates to just 20lumens per watt.

Fluorescent lamps operating on high frequency driver can achieve 85+ lumens at lamp per one input watt while maintaining CRI 85.

So in order to call LED's highly efficient they need to be at or above 85 lumens per watt including driver loss. Could you provide me with a link to white LED's with such efficiency?
 
Back
Top