LED lights are going to be outdated - already?

  • Active since 1995, Hearth.com is THE place on the internet for free information and advice about wood stoves, pellet stoves and other energy saving equipment.

    We strive to provide opinions, articles, discussions and history related to Hearth Products and in a more general sense, energy issues.

    We promote the EFFICIENT, RESPONSIBLE, CLEAN and SAFE use of all fuels, whether renewable or fossil.
Status
Not open for further replies.
This is quite rational on an individual decision making basis. But what if everyone in the world followed this practice: no mercury reductions, no reduction in sulfuric and nitric acid emissions, no reduction in particulate emissions, no decrease in CO2, etc. Coal would forever reign as king and health, quality of life, and the environment would degrade until we all and other living things, cockroaches excepted, live in squalor, disease, and worse.

The need is for people to act in community for the betterment of all, and that means paying the costs for benefits not built into the economic payback formula. Economic payback alone is and always will be short-sighted until all non-economic, social, and environmental costs also are fully accounted for.


You are quite free to feel that way. I won't knock it. The trained biologist in me wont let me get warm and fuzzy over carbon dioxide. I fully support your right to spend your money the way you want to though.
 
I see our agreement as being very broad. I too consider economic return, and i suspect that every thinking person, objectively or subjectively, does the same. And fortunately we enjoy the ability to have a range of outcomes.
 
many are not lasting any longer than the old incandescents yet cost way more.
I have found this problem too. The other issue I have is that they are horrible in outside lights in the middle of winter...
 
What recycle facilities do you have for the CFLs? No program exists in my area...
 
I have found this problem too. The other issue I have is that they are horrible in outside lights in the middle of winter...
Funny thing is, the CFLs I have outside seem to work fine in the winter. I moved some of them outside when I switched the inside ones for LEDs. I was told they won't work in the cold but so far they're good. Who woulda known?

What recycle facilities do you have for the CFLs? No program exists in my area...
HD and Lowes used to but stopped. There are a few other Cdn stores that will take them and there is a bin for them at the landfill site but you can't put them in the garbage. I really thought it was bad PR for Lowes and HD to stop but I still shop there. I guess they knew most people wouldn't stop shopping.
In Ontario, you are supposed to be able to take them to Rona, Cdn. Tire and Ikea. You may not have anything other than crappy tire within a day's drive of you.
 
thats okay we'll have mercury poisoning to complain about in a few years:rolleyes:
 
In our area they get disposed of together with other hazardous waste (paints, oil etc.).
thats okay we'll have mercury poisoning to complain about in a few years:rolleyes:

Only broken CFLs release a miniscule amount of mercury. That amount is easily saved by the mercury not emitted through a coal fired power plant due to the energy savings: http://www.popularmechanics.com/home/reviews/a1733/4217864/
With the switch to LEDs that is an outdated discussion anyway.

The trained biologist in me wont let me get warm and fuzzy over carbon dioxide.

I am a bit surprised to read that casual attitude from a trained biologist. Ever heard of Liebig's law? It has already been shown that higher atmospheric carbon dioxide concentrations are not necessarily beneficial for plants: http://www.nsf.gov/news/news_summ.jsp?cntn_id=123798
Not to mention that changes in precipitation due to climate change will severely disturb agriculture.
 
Try a few thousand years... fire, candles, and oil lamps, all similarly red color temps.

Old houses look odd when lit by CFL or LED. New houses are generally very ugly, so by association...
emoji12.png
. (... a duck!)

I believe one factor that influences our perception of the color from LEDs is that generally we've chosen interior colors that looked pleasant under incandescent lights, and even the palette of choices available were based on incandescent bulbs. I think that the battle to get warmer color LEDs will fade as our paint choices evolve to look better under modern bulbs.

TE
 
Only broken CFLs release a miniscule amount of mercury.

Yeah but when they crap out they stink to high heaven. Experience with more than one dying here.
 
The article doesn't say this will replace LED lights. The article says it is a better type of LED light.

I'm anxious to get rid of all my CFLs.

I also was, based on the visibly better color quality of decent LED's. Then Philips told me to wait (well, that's not how they phrased it, but it's how I read it), because they'll be coming out with an LED this year that uses only about half the power of existing models. I'm still not convinced they actually will, but the CFL's in most of my fixtures are cheap to run in the meantime, and LED's are continuing to drop in price.

Just pulled one pin-tube and it says it uses 220ma, that's 26W and not 13W on a 120V house circuit.

Power labels are for peak power, not average power, which is what the efficiency figures use. Also, a lot of bulbs have a higher than average rated draw while warming up. From other independent sources I've seen that have measured power draw from CFL's, it's always been pretty close.
 
I love LEDs for my cold house and colder garage. The CFLs wouldn't work at all in the garage and the ones in the house when it was cold took a couple of minutes to warm up. LEDs are instant.
 
In our area they get disposed of together with other hazardous waste (paints, oil etc.).


Only broken CFLs release a miniscule amount of mercury. That amount is easily saved by the mercury not emitted through a coal fired power plant due to the energy savings: http://www.popularmechanics.com/home/reviews/a1733/4217864/
With the switch to LEDs that is an outdated discussion anyway.



I am a bit surprised to read that casual attitude from a trained biologist. Ever heard of Liebig's law? It has already been shown that higher atmospheric carbon dioxide concentrations are not necessarily beneficial for plants: http://www.nsf.gov/news/news_summ.jsp?cntn_id=123798
Not to mention that changes in precipitation due to climate change will severely disturb agriculture.


Never heard of the law, but am quite familiar with the concept, which can be put to good use in many disciplines. I agree that co2 is often not the limiting factor in plant growth. You only have to look as far as the rainforest to find a cause of the week where the soil is the limiting factor.

I'm not going to get into climate change. You have your beliefs and nothing I say or reference would sway it 1 way or another just know that politics and science rarely mix. When they try to, everything gets corrupted and it becomes hard to sort fact and fiction.
 
The science is actually quite clear and has been for 150 years (http://www.bbc.com/news/science-environment-15093234) but I agree that arguing against political beliefs has become hopeless. It still remains that we are currently burning up the actual wealth of humankind while leaving future generations a huge debt in form of water and air pollution, deserted landscapes, overflowing landfills etc. At some point we will realize that we have plenty of money, just not much what we can buy with it.
 
  • Like
Reactions: sloeffle
Graphene bulbs ARE LED bulbs. They're just a different way of producing it.

I work in semiconductor processing. I have yet to buy any of the current crop of LED bulbs available... they're all using antiquated technology that will be obsolete in the near future. LEDs have been around for a long time, but only for small circuit switches, interlocks and whatnot... usually they've been red. The idea of using them for home lighting is relatively new, and there are currently a couple big players investigating new technologies and ways of producing them cheaper. It's really all about scale. Especially when compared to memory or processor chips, the current way of producing LEDs is ridiculously slow and expensive. Most of them are still made on 50 or 150mm wafers... the rest of the industry is already 2 generations ahead, running 300mm. (200mm was in between). 450mm tools are currently being tested... those are going to be a pain in the ass... but I digress.
When flash memory was first being introduced, all chips were being plated with gold. This has since been replaced by aluminum, copper and a few other metals, all far cheaper than gold. As fabs ramp up, yield increases. As yield increases, costs per chip go way down.

TL; DR: Moore's Law... in just a few years, the cost of a LED bulb will be less than half of what it currently is, and performance will be better.
 
@drz1050, I'm a very big user of GaN HEMT transistors, which are of course made in many of the same foundries as your GaN LEDs. I'm assuming the LEDs are still on 4" wafer due to being on SiC substrate? Any plans to move GaN onto a cheaper substrate, such as Si, a'la Nitronex? SiC will never be a very cheap wafer, compared to Si, even when scaled.

I never looked into the tech, but AND had white LEDs, which I used to use for lighting in robotic vision systems, back in the early 1990's. They may have been GaN in sapphire, which is also how they made the early GaN HEMTs. Talk about, "not cheap." [emoji12]
 
grisu,being sarcastic,doesn,t seem to come out in computer human beings change when they get punched in the face human nature
 
Especially when compared to memory or processor chips, the current way of producing LEDs is ridiculously slow and expensive.

I'm not so sure about that. Even memory chips are priced in the ballpark of $1-10. Processor chips can exceed $100.

Bulk LED's sell for pennies. You can buy high power (1+ W) Cree LED's in moderate quantities from places like Digikey for as little as 50 cents each. The lamp manufacturers who are buying more than one real at a time should be getting even better pricing. An 800 lumen bulb takes 8-12 of them. There's some room for further cost reduction, but not a huge amount.

A significant part of their cost is related to the tiny drivers needed to convert effectively unlimited 110 VAC to a stable 3.3 VDC, current limited, and deal with the havoc most dimmers cause. Then there's the relatively complicated bulb bodies, and usually a non-trival amount of aluminum formed into a big heat sink that the design revolves around.

Increasing the efficacy may actually help more with cost than further reductions in the price of each LED chip. Get the heat down low enough and designers more or less stop designing their bulbs around dissipating it.

All things considered, I'd be a bit surprised if 60W-equivalent LED's ever get below $5 a bulb. They'll never reach the $0.50 a bulb price of incandescents, and probably not the $2/bulb price of CFL's. I'm really not worried about that, though. Even at $10 a bulb they're a good value.

My main interest is seeing color rendering and luminous efficacy go up. These two goals go well together, since increasing the CRI almost inherently means reducing the efficacy, so improving the latter reduces the incentive not to improve the former.

Also, another decade or so down the road, bulbs might become a minority product. With the lifespans these lights are potentially capable of, we might move towards the LED's being built in to the fixtures. I've already seen some office-style recessed troffers designed like that.
 
  • Like
Reactions: Ashful
Apologies, the foundry I work at now does not produce LEDs. I know a bit about the manufacturing of them, but do not have any first hand experience.
I currently work with ALD & PVD processes on 300mm silicon wafers.

The last I heard about changing substrates for LEDs, they were looking at GaN wafers and bare Si. Sapphire wafers are still out there in labs too. GaN wafers will have a similar cost to SiC, but the lattice construction is far better, and again, once they scale, costs will drop dramatically. They are still fighting some issues building LEDs on Si wafers, but it's coming along.
Comparing completed processor chips to LEDs.. the two are not comparable at all. For one, in each processor chip there are thousands of transistors, diodes, capacitors and resistors. Node size is at 14nm now, with 10nm on the horizon. Each wafer goes through hundreds of processing steps through the fab, while LEDs only need to travel a fraction of that amount. Per component, LEDs are ridiculously expensive to make as of now.
Improving heat loss goes hand in hand with improving technology. Remember those old computers that needed to be stored in 60 deg rooms? Some server racks/ gaming computers still require cooling, but that's only because they're constantly being overclocked... if they were running at the same speed they used to run at, there'd be no issue at all. But while megahertz used to be considered fast, now terahertz is a thing.

As the LEDs are improved, heat will go down. As heat goes down, packaging will become easier. Also, please don't compare bulk LEDs with home lighting LEDs.
Once costs drop farther, it will become very attractive for lamp/ chandelier manufacturers to integrate them into their designs.
 
As far as color, there is no such thing as a white LED. The white ones you know are either red, green & blue ones together, or a blue one covered with a phosphorous lens.
 
The last I heard about changing substrates for LEDs, they were looking at GaN wafers and bare Si. Sapphire wafers are still out there in labs too. GaN wafers will have a similar cost to SiC, but the lattice construction is far better, and again, once they scale, costs will drop dramatically. They are still fighting some issues building LEDs on Si wafers, but it's coming along.
Just to clear up some mis-information here, before anyone runs with it. There is no such thing as a GaN wafer, at least in my area of work. GaN is an epitaxial layer applied to a wafer of Sapphire, SiC, or Si. I'm sure other substrate materials exist, but these are the three used in my area of work. I do not know anyone trying to make a GaN wafer.
 
Last edited:
As far as color, there is no such thing as a white LED. The white ones you know are either red, green & blue ones together, or a blue one covered with a phosphorous lens.
If the end result is a perceived white light, then most people would call it a white LED, regardless of how it is made. If the Nobel Committee calls it white LED, that's good enough for this forum. "White LED lights emit a bright white light ... "
http://www.nobelprize.org/nobel_prizes/physics/laureates/2014/press.html
 
Apologies, the foundry I work at now does not produce LEDs.

No need to apologize. I think this is a good discussion...and I can just imagine somebody wandering onto Hearth.com looking for info on wood burning and stumbling into a discussion of the difference between gallium-nitride wafers and gallium nitride coatings on semi-conductors. :rolleyes:

Your points are fair, although I still think that hints towards a long term price minimum in the ballpark of $5 per bulb.

If the end result is a perceived white light, then most people would call it a white LED, regardless of how it is made. If the Nobel Committee calls it white LED, that's good enough for this forum. "White LED lights emit a bright white light ...

Strictly speaking, the Nobel prize was for the invention of the blue LED, but still, I see nothing misleading in the context of ordinary discussion with calling a phosphor coated or shrouded LED a white LED.
 
Bulk GaN wafers do exist. They are grown on sapphire, and then shaved off.
Here's an article talking about costs:
http://www.luxresearchinc.com/news-...ide-costs-fall-60-2020-leading-more-efficient
Okay, yes, it does exist... in a completely non-commercial form. Even your Lux Research article links to another stating GaN on GaN is a "non starter for now":

http://www.luxresearchinc.com/news-...efficient-gan-power-conversion-devices-hit-11

Again, I'm one of the larger GaN consumers in the country, being Cree's largest single customer on several of their devices. I've not come across a commercial use of GaN on GaN, in my work.
 
Yeah but when they crap out they stink to high heaven. Experience with more than one dying here.

We've gone through many CFL bulbs over the years, especially in the kitchen. Only one stunk. It was an original Philips CFL. The cheapy HomeDepot units I have had for the past 7 years have simply stopped working, no smell. We are fortunate that our local power company lets us drop off fluorescent bulbs at their office. Starting next month our county waste transfer station is also going to start recycling them.
 
Status
Not open for further replies.