Print 51 comment(s) - last by Astara.. on Feb 4 at 7:24 PM

The top figure showcases "LED droop", which decreases LED lighting efficiency and brightness. The Rensselaer Polytechnic Institute and Samsung have come up with a new LED, whose output profile is shown below, which greatly reduces droop.  (Source: Rensselaer Polytechnic Institute/Samsung)
New technology may help LED lighting inch closer to market

In the U.S. and elsewhere, lighting accounts for as much as a third of electricity usage.  Thus, when it comes to fossil fuels conservation, there is a large impetus to adopt more efficient lighting solutions.  Right now the most efficient lighting solution is LED lighting, which if widely adopted could save 10 percent of the power used in the U.S.  Unfortunately, LED lighting is also by far the most expensive form of lighting.  To spur research in to lower LED lighting costs, the U.S. Department of Energy (DOE) has offered a $20M USD "L Prize" for the first team to meet a rigorous set of standards.

A new breakthrough by researchers with Rensselaer Polytechnic Institute's National Science Foundation-funded Smart Lighting Engineering Research Center and Samsung Electro-Mechanics may bring LED lighting closer to affordability and help the researchers creep closer to the DOE-funded jackpot.

The new type of LED is said to be "polarization matched", but what's most important is its improved metrics.  The light offers 18 percent increase in light output and a 22 percent increase in wall-plug efficiency, which essentially measures the amount of electricity the LED converts into light.

The new kind of LED greatly diminishes a common problem with LEDs called "efficiency droop".  This phenomenon involves LEDs being most efficient when operating on low current densities, and seeing their efficiency greatly droop at higher current densities.  While all the factors have yet to be determined, electron leakage is one source of droop.  The end result of droop is that LEDs are forced to operate at lower current densities, which feature much lower brightness and efficiency in achieving light output.

Project leader E. Fred Schubert, Wellfleet Senior Constellation Professor of Future Chips at Rensselaer states, "This droop is under the spotlight since today’s high-brightness LEDs are operated at current densities far beyond where efficiency peaks.  This challenge has been a stumbling block, because reducing the current densities to values where LEDs are more efficient is unacceptable. Our new LED, however, which has a radically re-designed active region, namely a polarization-matched active region, tackles this issue and brings LEDs closer to being able to operate efficiently at high current densities." 

His team discovered that a cause of electron leakage was mismatched polarization.  The used a quantum-barrier design to help to greatly reduce the mismatch.  The conventional Gallium Indium Nitride/Gallium Nitride (GaInN/GaN) layer of the LED active region was replaced with Gallium Indium Nitride/ Gallium Indium Nitride (GaInN/GaInN).  The decrease in polarization mismatch, in turn decreased electron leakage, in turn lessening droop and delivering superior brightness and efficiency.

The gains present in lab testing were similar to theoretical models, created in computer simulations.  Professor Schubert says he expects discoveries like this one to propel solid state lighting into the mainstream, which he says will amount to vast environmental, energy, and cost benefits as well as innovations in healthcare, transportation systems, digital displays, and computer networking.

LED lighting has already found its way into many vehicles.  The list of models sporting LED headlights continues to grow, as the technology prepares to exit the high-end market and move into the mainstream auto market.

The new research is reported in this week's Applied Physics Letters.

Other researchers on the project include Rensselaer physics, Future Chips, and electrical engineering graduate students Jiuru Xu, Martin F. Schubert, and Ahmed N. Noemaun; Rensselaer Future Chips research assistant Di Zhu; Jong Kyu Kim, research assistant professor of electrical, computer, and systems engineering at Rensselaer; along with Samsung Electro-Mechanics researchers Min Ho Kim, Hun Jae Chung, Sukho Yoon, Cheolsoo Sone, and Yongjo Park.

Comments     Threshold

This article is over a month old, voting and posting comments is disabled

A few cheap LED lights available
By thornburg on 1/15/2009 2:40:26 PM , Rating: 4
Walmart sells LED "Accent" lights that fit a regular light bulb socket. They claim 40W equivalence for a 1.5W bulb. They come in "warm white" and "bright white". Rated for 30,000 hours of life. They cost $5.96.

I think you'll find, if you do the math, that it doesn't take very long for one of these to pay for itself, if you have a light that you use a lot (like a porch light you leave on all night). Even if you compare it CFL, they're still MUCH cheaper to run, and they don't contain mercury.

The problem:

They don't provide excellent illumination--the diffusion isn't good (they operate kind of half-way between a omni-directional bulb and a spot-light). They're excellent for places that would otherwise be pitch black, and you just need to see to move around (like an entry-way light), and they work pretty well if you have a flexible-neck lamp or a "can-light" that takes regular bulbs. They also make pretty good ambiant lighting for watching movies or playing games, if you don't want it totally black.

So, to all the manufacturers out there--improve this design slightly, without increasing the cost (@ 40W eq.), and offer some slightly brighter bulbs (60W eq. for sure, 75 would be nice--I don't need 100W, thank you). The brighter ones can cost more. I'd pay $8 for a 60W and $10 for a 75W...

By JediJeb on 1/15/2009 5:04:05 PM , Rating: 3
A note on the mercury content of the CFL lights. It is a very tiny amount of mercury in those bulbs, and you will probably ingest more mercury in a year drinking tap water than you would get from one of those, if you were to eat it. You would be suprised at the amount of mercury in water, tap or bottled. In our lab we do testing of drinking water, waste water and soils, and under the new permits for waste water, we had to buy distilled water to get the mercury background low enough to run the test. Even the water from our high end Deionized Water System was too high, and when I analyzed the tap water that came into the building it was 2 parts per trillion mercury. Of course the legal limit for drinking water is 200 parts per trillion, but for waste waters it is 0.5 parts per trillion, go figure.

RE: A few cheap LED lights available
By Oregonian2 on 1/15/2009 5:09:18 PM , Rating: 2
Walmart sells LED "Accent" lights that fit a regular light bulb socket. They claim 40W equivalence for a 1.5W bulb. They come in "warm white" and "bright white". Rated for 30,000 hours of life. They cost $5.96.

FWIW - In my experience the equivalence ratings on such things almost always (for me fully "always", but I'll leave an opening) creative and perhaps compared to an incandescent that's already blackened (due to tungsten evaporated off of the filament) on the inside, and perhaps spray-painted with a little black spray. In your example of "40W" equivalence where you say it has a bright-spot, the light intensity at that bright spot may be as bright as the omni-directional light coming from the bulb I described above. :-)

For me, the color of the bulb is the troubling part -- but that said we've found some fluorescents that have good efficiency and good spectrum (but undoubtedly will become unavailable due to their slightly higher prices).

RE: A few cheap LED lights available
By glennpratt on 1/16/2009 4:48:35 PM , Rating: 2
I have 60w incandescent and 60w equivalent CFLs in the same fixture in a rarely used bathroom. Sure when I first turn it on the incandescents are brighter, but after a minute the CFLs are blinding by comparison. And yes, these aren't old bulbs.

As for spectrum, every time I shop for CFLs there is a bigger selection of 'colors', I imagine it will only get better with time.

By glennpratt on 1/16/2009 4:51:02 PM , Rating: 2
Now that I read it, I think you were refering to LEDs at first, my mistake, as I have no experience there.

RE: A few cheap LED lights available
By ampman on 1/18/2009 11:29:50 PM , Rating: 2
CFL's spectrum depends on the phosphore mfg uses to turn UV into "white" light. Any phosphore has its own limited spectrum. My eyes hate CFL (vs. halogen). claims making better LED bulbs than CFLs.

RE: A few cheap LED lights available
By gstrickler on 1/15/2009 7:04:20 PM , Rating: 2
They come in "warm white" and "bright white".

The warm white may be acceptable (2900K-3200K), but the bright white (3500k-4200k) is likely to look terrible for lighting, try one before you spend much money. Incandescent bulbs are around 2500K-3200K and the better CFL bulbs are 2700K-3000K.

RE: A few cheap LED lights available
By samoak54 on 1/15/2009 11:17:43 PM , Rating: 2
Wouldn't a glazed light covering fix the problem of off-color lights? Filter the light, and the color changes based on the kind of glaze used. Problem solved.

By gstrickler on 1/16/2009 3:11:38 AM , Rating: 2
If they're "full-spectrum" LEDs, then a filter could adjust the apparent color temperature. More accurately, it can adjust the color balance, but that can effectively mimic a different color temperature. However, there will be some cost in efficiency (fewer lumens for the same power) and some extra monetary cost (for the filter).

If it's not a full-spectrum source (like florescent lamps, mercury vapor, HPS lamps, most LEDs, etc.) then filters may not help nearly as much.

By Astara on 2/4/2009 7:24:47 PM , Rating: 2
You have an interesting definition of 'better'. You say better CFL's are 2700-3000K. I see the higher end bulbs coming in nearer to 5000-5500K for daylight and true-color rendition. Sure, 2700-3000 is ok for reading, but if you want colors to look closer to daylight, anything less than 4000K looks positively like a dirty candle in my eyes. Usually the daylight spectrum bulbs cost a bit more but nearly always provide better spectrum for everyday use. I'll admit that 3000K might be more energy efficient for reading black and white text, but if you want colors to look natural, sunlight is about 5500-5750K -- and if it is a cloudy day, you are looking at whites over 6000K. Guess it all depends on what you are lighting. But I find for kitchen, garage, bathroom, -- deciding on colors for outfits or makeup or anything requiring color matching, daylight balanced are superior. But human eyes are most sensitive to the 5000K+ (5700K) range-- so bulbs tuned to 5000K (but not full-spectrum) are most efficient for readability/watt.

RE: A few cheap LED lights available
By ampman on 1/18/2009 11:19:21 PM , Rating: 2
LED light bulbs with very high color rendering index (CRI) and high luminosity are on their way. I happen to know companies in the process of making them using color mixing. They're not going to be inexpensive to acquire but they provide 100% ROI for > 15 years. The inexpensive ones use but "white" LEDs and have CRI < 75. CREE's LR6 and LR4 have CRI > 90 but are spendy and are not light bulbs. Up and coming compamies would be,, and a company in Santa Cruz, CA area that I know.

By MrPoletski on 1/20/2009 11:24:02 AM , Rating: 2
All that is needed to solve the 'harsh' light of an LED lamp is to surround it with a phosphoescent frosted shield. Much like the whited incadescent bulbs you can get that still give out light but you cant see into the filament.

ist generation automotive
By blowfish on 1/15/2009 12:18:31 AM , Rating: 2
Have you noticed one effect with some LED turn indicators? There's an instantaneous white light before you get the orange.

RE: ist generation automotive
By geokilla on 1/15/2009 12:32:06 AM , Rating: 2
I prefer white lights over orange lights. Looks better lol.

Besides, even if that does happen, the driver can barely see that white light unless they're staring into the turn signals and not at the road and his/her surroundings.

RE: ist generation automotive
By MrPoletski on 1/15/2009 4:52:10 AM , Rating: 2
they aren't lasers!

RE: ist generation automotive
By dayanth on 1/15/2009 8:44:59 AM , Rating: 5
My Eyes! The Goggles Do Nothing!

RE: ist generation automotive
By 306maxi on 1/15/2009 6:38:14 PM , Rating: 1
While I love Simpsons references I must say this one is a little wrong because when Rainier Wolfcastle said this it was because he had acid in his eyes and not because of lasers.

RE: ist generation automotive
By dayanth on 1/15/2009 7:41:37 PM , Rating: 5
Well, reference-nazi, you're right lol... But the phrase is versatile enough that it could be used as to provoke your own mental image of bright lights being shone into your eyes, causing the outburst.

RE: ist generation automotive
By dj LiTh on 1/15/2009 10:53:59 PM , Rating: 2
You sir are either a genius or a very sad sad man. I'm not going to even get into what i am for idealizing/dissing you.

RE: ist generation automotive
By MrPoletski on 1/20/2009 11:21:33 AM , Rating: 2
you watch the simpsons too much. Clearly.

By dug777 on 1/15/2009 4:42:33 AM , Rating: 2
In the U.S. and elsewhere, lighting accounts for as much as a third of electricity usage.

Certainly in WA, lighting is about 4% of residential usage.

I don't have an overall figure on hand but I would be surprised if it is was anywhere near a third of electricity usage for the grid as a whole, either in WA or anywhere else.

Would be interested to see such a figure, I could of course be very wrong :)

By AnnihilatorX on 1/15/2009 5:22:58 AM , Rating: 2
The 4% figure you used is for energy usage (including gas), not just electricity.

Street, office, car park lighting,etc I am sure consume comparably with residential.

By dug777 on 1/15/2009 7:55:40 AM , Rating: 2
Good point, but even so, your doe states that indoor and outdoor lighting was 8.8% of US household electrcity use in 2001.

With a meaningful uptake in CFLs I would be suprised if that percentage had increased since then.

By Alexstarfire on 1/15/2009 11:56:27 AM , Rating: 2
I didn't realize that only residential houses used light.... guess I'll tell Walmart and CDC to stop using their lights.

By ZmaxDP on 1/15/2009 3:08:00 PM , Rating: 4
Commercial buildings tend to run lighting all day, not in the evenings and mornings only. Also, commercial lighting is more strictly regulated by code for higher illumination levels. Typical power density for lighting in non-critical commercial office space is more than 4 times what you find in most homes. In critical applications (surgery, industrial manufacturing, etc...) it can be 4 times higher than what is placed in commercial office buildings (or more).

The other hidden factor is that all those watts are also generating heat (whatever the bulbs don't convert to light) and that heat output has to be handled by the building's HVAC systems. The typical "break even point" for a commercial building with a 72 degree internal temperature is around 50 degrees on external temperature. Anything above 50 and the building has to run it's HVAC to keep the building cool due to the heat output from lighting, electronics, equipment, and people. Below 50 and the building will be in heating mode overall. This is a general rule of thumb. Buildings with a lot of glass and poor insulation will obviously find this point it closer to the indoor temperature. Buildings that are well insulated will find it even lower. This is why optimizing a building's energy performance in Texas is a lot different that optimizing it in Washington state.

By Oregonian2 on 1/15/2009 5:01:05 PM , Rating: 2
It's also been my observation that commercial buildings are almost always using fluorescent lighting for interiors so LEDs would still be an improvement, but be less so (and at great cost -- those 40W fluorescent bulbs at industrial quantities are dirt cheap).

By dug777 on 1/15/2009 9:48:08 PM , Rating: 2
I've often wondered what sort of a person makes imbecilic and childish comments like yours, and what they get out of sounding like a retarded 8 year old with a chip on his/her shoulder. Respect from their peers, self-esteem, gold, the love of a beautiful woman?

If you'd bothered to actually read my comments you'll see that I'm clearly aware of that astonishing posibility. I even state that I'd be interested in seeing some data covering overall usage breakdown.

I am aware that commercial and some industrial buildings use a lot of lighting, but a figure of 1/3 of total usage seems very high, hence my curiousity.

By ampman on 1/18/2009 11:24:10 PM , Rating: 2
Databean mentioned 22% for lighting in their market research

Efficiency of LED lighting
By werepossum on 1/15/2009 11:35:58 AM , Rating: 2
For systems available today, LED is less efficient than high wattage HPS (high pressure sodium) and metal halide systems, as well as linear fluorescent systems. The HPS and MH in 1000W run about 130 and 110 lumens per watt respectively, or high using electronic ballasts. NO (Normal Output) T5 linear fluorescent lamps run about 100 lumens per watt with electronic ballasts. The best available high-powered white LEDs run about 80 lumens per watt, but most are far lower, 40 lpw being common. I've even seen some just last month that are 11 lpw. For comparison, the best incandescent lamps (GE Halogen + PAR lamps in narrow flood or spot distributions) run as high as 24 lpw. Bottom line - LEDs are very promising and the best option in many applications, but they are not always the best choice in lighting even when not considering cost.

RE: Efficiency of LED lighting
By McDragon on 1/15/2009 3:51:50 PM , Rating: 2
I agree, Flourescent lamps will be the choice for normal business lighting for some time still.
If absolute efficiency was the goal, Low pressure sodium is the best, reaching nearly 200 lumens/watt. The color is quite yellowish, though.

RE: Efficiency of LED lighting
By Kary on 1/15/2009 5:15:05 PM , Rating: 2
Thank you.

The "Right now the most efficient lighting solution is LED lighting" line really bothered me in that article.

Sure LEDs aren't made with mercury and so aren't poisonous to the environment (though with proper recycling the same would be somewhat true for florescent and sodium lights...) BUT that really doesn't mean they are "the most efficient lighting solution".

"The most efficient lighting solution" HAS TO GIVE OFF LIGHT MORE EFFICIENTLY THAN THE REST.

(sorry for the rant, that just really bothered me.. way to go pulling all of the relevant data werepossum)

RE: Efficiency of LED lighting
By mindless1 on 1/16/2009 6:59:00 AM , Rating: 2
It won't be long till it's true,

though right now what they have available in volume is only a little over 100 lpw.

Looking forward to LED lighting
By psychobriggsy on 1/15/2009 6:30:03 AM , Rating: 2
The day I can buy a 100W equivalent LED lightbulb, that emits a more natural light than CFLs (although I use these mostly these days), in a power envelope similar or less than CFLs, for a price less than £10, with a life span measured in decades, is the day I buy one.

Actually, the biggest benefit could be how the light is configured - it needn't be a single point source bulb, it can be a far more spread out set of LEDs achieving a far more even lighting effect. I expect that radial web-like lights that still plug into ceiling light sockets will be among the first non-bulb lights to appear. In the long term, houses may be wired with lighting circuits more suitable for LEDs (to avoid a transformer in each light) and each ceiling could have dozens of tiny light points, like downlighters but smaller.

RE: Looking forward to LED lighting
By Epsil0n00 on 1/15/2009 12:09:23 PM , Rating: 2
I think your idea of a mesh of LEDs is great! I have long wished that I could have some type of large-coverage lighting system on my ceilings instead of single light bulbs... I really like to have a lot of dispersed light in my rooms. Halogen torche lights do a fairly good job, but they get really hot and (I think) aren't that efficient. If I could have large panels of LEDs or a mesh of LEDs that would be awesome to create a even light through an entire room!

Maybe I should just build it myself!

By ThisSpaceForRent on 1/15/2009 12:55:14 PM , Rating: 2
Christmas lights, sure it looks ghetto, but you get really good lighting from them once they're spread over an entire ceiling. Actually, why aren't we using these already?

Motorcycle LED use
By RoberTx on 1/15/2009 12:09:41 PM , Rating: 2
Many motorcyclist that have switched to LEDs have found this out. LEDs bright enough to be seen in the daylight are blindingly bright at night. I was forced to make an instrument light dimmer with a photocell and transistor so I could see my LED indicators in the daylight and not blind myself at night. An LED brake light that can be easily seen in the daylight will fry retinas in the dark. Some type of ambient light sensor switch will be necessary on LED style automotive lighting.

By dani31 on 1/15/2009 2:56:20 AM , Rating: 2
I even focus it on a friend's rear from across the room and after about 40 seconds cause them to jump up grabbing their pants in pain.

Do they also sing? Or was it a broken DVD burner?

By Gzus666 on 1/15/2009 8:57:57 AM , Rating: 2
Well, these are LEDs, not laser diodes.

But, on the laser note, I am building a blue laser from an old HD-DVD 360 add on. Waiting for my flashlight and batteries to come in...seems like forever. Should be pretty powerful, got a large heatsink and everything and 300mA shouldn't be hard to hit, oh boy. Wasn't too bad on price either, about $80 for everything. Hopefully mine will be able to do a bit more than yours can since blue is a smaller wave.

By FITCamaro on 1/15/2009 10:38:43 AM , Rating: 2
Hot chick: Do you have a lighter?
Gzu: No I have a big laser in my pocket though that can light a fire.
Hot chick: *slap! and walk away*

By Gzus666 on 1/15/2009 10:50:31 AM , Rating: 2
Ha, I didn't think of that one...*ponders the possibilities and weighs the good and bad*

By phxfreddy on 1/15/2009 9:16:05 AM , Rating: 2
GeeSuS ... make sure you do not point this set up of yours towards anyones eyes. You will definitely do damage.

By Gzus666 on 1/15/2009 9:28:40 AM , Rating: 2
They will pretty much permanently damage a retina on contact at that power. One of those things you need to be VERY careful with.

By mvrx on 1/16/2009 12:04:40 AM , Rating: 2
Yes.. well aware that this could blind. I only use it very responsibly .. to burn my arses of my friends from across the room.

I do like the idea of lighting a smoke for a gal.. I probably would get slapped with that comment tho.

"Most expensive form of lighting"
By Etsp on 1/15/09, Rating: -1
By TOAOCyrus on 1/15/2009 1:07:14 AM , Rating: 3
Everyone knows exactly what he ment so why make a big stink about a simple qualifier thats not even necessary? Seriously if you really have to complain about Dailytech's writing find something real.

RE: "Most expensive form of lighting"
By Googer on 1/15/2009 1:21:01 AM , Rating: 2
I believe you are misinterpreting the meaning.

LEDs are the most expensive to acquire and implement, but not to run.

RE: "Most expensive form of lighting"
By Etsp on 1/15/2009 1:27:05 AM , Rating: 2
I was under the impression he meant overall cost... as in, the cost of obtaining it, plus the cost of running it over it's entire lifetime, as compared to other common forms of lighting...

By quiksilvr on 1/15/2009 1:54:21 AM , Rating: 2
It would be natural to assume that...if he didn't say a sentence prior that it's the most efficient way to light a place. He obviously meant the actual cost of the light bulb and not the overall benefits. You don't have to be condescending and saying burning money for light is more expensive.

By Darkskypoet on 1/15/2009 12:41:23 PM , Rating: 2
I actually doubt that it is the most expensive form of lighting. As soon as you take maintenance costs into account, especially for applications such as signage; LEDs are very quickly replacing flourescent, neon, etc. for their lower TCO.

If you start to take into account the costs of changing bulbs, and other maintenance required to keep an elevated sign in good repair; LEDs become far more cost effective. For initial upfront costs, yes LEDs are more expensive. However, depending on application, TCO varies wildly, and the markets uptake of LED lighting products in many of these commercial applications speak volumes about its cost competitive nature.

As a side note, I too would love to see LED panels replace flourescent light fixtures; however as I greatly prefer light sources in the 6000k range, they had better not simply come in the standard yellowish white light that the majority of CCFLs, and incandescent bulbs produce. Yuck.

"Paying an extra $500 for a computer in this environment -- same piece of hardware -- paying $500 more to get a logo on it? I think that's a more challenging proposition for the average person than it used to be." -- Steve Ballmer

Copyright 2016 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki