ultra-efficient incan, 173 Lm/W

[chillinn]

A photon-recycling incandescent lighting device

Well, guess what's coming?

 

[alpg88]

theoretically possible does not mean economically feasible. will see how it plays out.

 

[chillinn]

It's not theory. It's reality. They did it (and MIT also did it back in 2018). The paper is pretty detailed and shows how they did it. The evaporation and oxidation of incandescent carbon is avoided by a structure made of carbon nanotubes that works in argon. Lifetime is as good as for LEDs. Low efficiency is solved by using a lossless optical filter that transmits visible light and reflects back infrared to prevent heat loss by radiation. It reaches maximum energy efficiency determined by human vision characteristics, unlike current LED, which are unavoidably limited to a value lower than that by losses caused by fluorescent light conversion. Design of the lossless optical filter is a difficult optimization problem due to the large number of parameters, so their solution, which is completely described in the paper, was found by a machine-learning algorithm.

reply

 

[Olumin]

Ill hold my horses until I can put it in my shopping cart.

 

[chillinn]

Well, for viable LED lighting, your horses were held for half a century (i.e. 1962 --> 2011), so it's only fair.

 

[alpg88]

actually longer,

Electroluminescence as a phenomenon was discovered in 1907 by the English experimenter H. J. Round of Marconi Labs, using a crystal of silicon carbide and a cat's-whisker detector.[9][10] Russian inventor Oleg Losev reported creation of the first LED in 1927.[11] His research was distributed in Soviet, German and British scientific journals, but no practical use was made of the discovery for several decades, in part due to the very inefficient light-producing properties of silicon carbide, the semiconductor Losev used

Light-emitting diode - Wikipedia

en.wikipedia.org

however we are not talking about new tech, this an improvement of an old tech. that only exists inside MIT lab, no different than creating a new steam powered automobile, using modern technology. i do not see it being produced on commercial scale.

 

[chillinn]

i do not see it being produced on commercial scale.

Unless you think anyone would have had enough clairvoyance to see the commercial viability of LED 60 or 115 years ago, I'm not sure it matters. But we better hope if we know what's good for us.

 

[alpg88]

I can say it with the same certainty as i would say that steam powered cars are not coming back. Leds are not standing still either, they are constantly improving.

The highest perception of brightness experienced by the human eye is generated with green light on the 555 nm wavelength. The greatest luminous efficacy which can theoretically be achieved at 555 nm is, altogether, 683 lm/W.

i'm pretty sure 170lm/w will be achieved by leds soon, if not already,

 

[chillinn]

Not likely. LED are already bumping up against theoretical efficiency limits. There's literally no room for improvement. Afa LED "constantly improving," that improvement is towards being more like incan, not with efficiency improvements, and the more like incan they are, the less efficient they are.

This isn't is a football game. LED is not your team. It shouldn't matter to you whatsoever what your light source is, so long as it is efficient and is not hurting you. LED has fallen flat on a lot of promises, but it's just as much greed as it is anything that is the problem with them. Cheap crap is cheap crap, and that is mostly the problem. But even the best LED light sources have a spectrum curve that is unnatural. If that is fixed, incan may be dead, but I haven't heard of anyone working on that, just on perceived tint, color temperature and high CRI hiding an unnatural color spectrum still weighted in the blue end. That's, so far, very bad. Again, maybe they'll fix it. Maybe they won't and incan will return in high efficiency form. We'll see.

And I don't think steam powered cars were ever really anything. Horse and buggy, otoh, never really completely went away.

 

[Olumin]

I remember reading something about a 200lm/w LED by Phillips some years ago. What happened to that?

 

[Olumin]

Yeah these ones.

Edit: Here in English.

 

[alpg88]

Not likely. LED are already bumping up against theoretical efficiency limits. There's literally no room for improvement. Afa LED "constantly improving," that improvement is towards being more like incan, not with efficiency improvements, and the more like incan they are, the less efficient they are.

This isn't is a football game. LED is not your team. It shouldn't matter to you whatsoever what your light source is, so long as it is efficient and is not hurting you. LED has fallen flat on a lot of promises, but it's just as much greed as it is anything that is the problem with them. Cheap crap is cheap crap, and that is mostly the problem. But even the best LED light sources have a spectrum curve that is unnatural. If that is fixed, incan may be dead, but I haven't heard of anyone working on that, just on perceived tint, color temperature and high CRI hiding an unnatural color spectrum still weighted in the blue end. That's, so far, very bad. Again, maybe they'll fix it. Maybe they won't and incan will return in high efficiency form. We'll see.

And I don't think steam powered cars were ever really anything. Horse and buggy, otoh, never really completely went away.

Well it turns out, 210lm/w leds already exist, and are sold in pretty much any hardware store. lol

You are right it does not matter, but as of today leds do it better than anything. imo, 99% of population do not car or even know about spectrum curve, cri, some may care about color temp, but led bulbs available in 3 tints here, soft, neutral and cool

actually steam cars really were something, you just do not know much about it, watch jay lenos garage, he has 2 in working conditions, and he tells you all about steamers.
horse buggies are only used where there is nothing else available, or due to religious practice. not a very good argument btw,

 

[chillinn]

I'm pretty certain that if there was an incan tech 100x more efficient than the most efficient LED, you guys would still choose to slowly blind yourselves.

If LED didn't suck, I probably would, too, but thus far, I haven't seen or even heard of any LED lighting, let alone an LED flashlight, that had a decent constant current driver with an emitter with tint and temperature worth the materials used to construct it. Sure, they're really really neat, and in small doses they get the job done, but except for the increase in efficiency, the cheapest, crummiest incan is head and shoulders above the most expensive LED, as far as the quality of the light is concerned.

And this only matters because that is what it is... a light. Right? I'd feel the same way about a car that was ten times more efficient but had a terrible ride, because that is the whole point, the ride, getting somewhere. I'd feel the same way about a really ultra-efficient printer that couldn't print worth a dang.

I'd like to see LED get better, but I'm not deluding myself about the current state of commercially available LED, which no matter how much we like them and prefer them to incan and even hate incan, LED is slowly blinding us and insidiously shortening our lives. That I can not get over as easy as others.

So I'll use more batteries, pay for more electricity, but at least I'll have superior light that is neither permanently blinding me nor cutting my life short.

 

[jtr1962]

Commercial LEDs are already at 235 lumens per watt. Yes, the rate of improvement is plateauing, but that's to be expected given how close we are to theoretical maximum efficiency. Spec sheet for CRI 90 version. Worst case efficiency (for 2700K and bottom of the binning range) is 172 lm/W, although typical (middle range) would be roughly 190 lm/W. Most people wouldn't be able to tell the difference between CRI 90 and the CRI 96 of the PRILD prototype. I'm sure we could make a CRI 95+ version of this LED to match the PRILD prototype, perhaps with an efficiency hit of 10% at most. That would put typical values in the low 170s worst case.

No mention of the feasibility of manufacturing here but let's consider one thing. Right now mid-power LEDs can be had for under one cent in large quantities. Even adding the driver, heatsink, and diffusers we're still managing to sell LED bulbs for under a dollar. Those numbers will only get better. Assuming this new tech can be hooked directly to line voltage, it will need to be manufactured very cheaply in order to compete with LEDs.

They mention a theoretical efficiency of 236 lm/W at 3000K and CRI 96. If we want to do an apples to apples comparison, a 3000K CRI 95 LED with a 100% efficient blue emitter would be around 210 to 230 lm/W.

What I find interesting though about this new tech is its potential simplicity. Assuming it doesn't need a driver, and can be manufactured cheaply enough, there might be some commercial viability. It seems like the filter could even be tuned so the device emits light at CCTs other than 3000K. This overcomes the primary drawback of incandescent, namely the inability to emit light at any CCT much over 3000K.

I read about the MIT version but haven't seen updates since. I'm surprised they got it to work, although that's no guarantee it'll ever be commercially viable. As others have said, it's not like LED technology is standing still.

 

[alpg88]

if there was one we'd gladly use it, we are not religiously anchored to leds as you seem to be to incan. Leds just make light better than anything else today, most efficient, cheap, simple and reliable,

that new tech uses carbon nanotubes., it is not cheap, and some medical experts ring huge bells about nanotubes, how it can negatively affect our heath, so it may not be something you'll find in commercial product,

 

[chillinn]

Funny, because that's what it seems like. The topic is incan, in the incan forums, yet most of the content is about LED. And as long as LED light has the problems that it does, re: slowly permanently blinding while taking years off lives, I must disagree regarding, "better than anything else today."

 

[alpg88]

yea good point actually, it's just that the topic is about efficiency and what is "coming", is not as efficient as current light producing devices.
but it is a good news anyway, most of great things that changed our lives were invented by accident, while working on something else, so maybe it will yield unexpected and unrelated results that will have huge effect on our future

 

[jtr1962]

I'm pretty certain that if there was an incan tech 100x more efficient than the most efficient LED, you guys would still choose to slowly blind yourselves.

If LED didn't suck, I probably would, too, but thus far, I haven't seen or even heard of any LED lighting, let alone an LED flashlight, that had a decent constant current driver with an emitter with tint and temperature worth the materials used to construct it. Sure, they're really really neat, and in small doses they get the job done, but except for the increase in efficiency, the cheapest, crummiest incan is head and shoulders above the most expensive LED, as far as the quality of the light is concerned.

And this only matters because that is what it is... a light. Right? I'd feel the same way about a car that was ten times more efficient but had a terrible ride, because that is the whole point, the ride, getting somewhere. I'd feel the same way about a really ultra-efficient printer that couldn't print worth a dang.

I'd like to see LED get better, but I'm not deluding myself about the current state of commercially available LED, which no matter how much we like them and prefer them to incan and even hate incan, LED is slowly blinding us and insidiously shortening our lives. That I can not get over as easy as others.

So I'll use more batteries, pay for more electricity, but at least I'll have superior light that is neither permanently blinding me nor cutting my life short.

This post is mostly about your subjective preference for incandescent lighting, as opposed to any objective advantage. Consider that I personally found incandescent light so lacking that I preferred to do my homework under one of those really lousy halophosphor fluorescents we had in the kitchen in the 1970s. Yes, it didn't render some colors well at all, but at least white looked closer to the type of white I saw under sunlight, not the awful yellow it appeared under incandescents. I don't care if something has a 100 CRI when the color temperature is so way off that there isn't even a white point, just shades of yellow.

LED isn't blinding us any more than any other light source. You're worried about the blue spike, go with CRI 95+ versions. The blue spike doesn't exist. At best there's a blue hump. I've little doubt future versions of LEDs will get rid of the blue spike entirely with better phosphors if science finds it to be overly harmful.

Until LEDs came along you had nothing at all if you wanted a portable light source which emits true white light similar to sunlight. Fluorescents just couldn't be scaled down to the sizes of tiny incandescent bulbs. At best you could get around the problem by painting the bulbs with a translucent blue paint. Unfortunately, that made the already low efficiency even worse, probably under 1 lm/W. Now you can get high CCT LEDs with fit in anything for pennies, plus they're over a hundred times more efficient than incandescents with a blue filter.

Your blind spot is not seeing that some people actually prefer the light LEDs emit over incandescents. I wouldn't use incandescent if it was 100 times more efficient than LED for that reason alone. I can't see well under the type of light it emits, plus I get a headache being under it for too long.

 

[chillinn]

This post is mostly about your subjective preference for incandescent lighting, as opposed to any objective advantage.

But there is an objective advantage with incan in the quality of the light itself, and in neither slowly permanently blinding due to a color spectrum with too much blue, nor shortening lives by disrupting circadian rhythms, of which studies have appeared that show that is the case with LED.

but at least white looked closer to the type of white I saw under sunlight, not the awful yellow it appeared under incandescents

Maybe initially, but the human brain will adjust the white balance pretty rapidly so that what you initially perceive as yellow will look white.

LED isn't blinding us any more than any other light source.

There is more blue light in LED than other light sources, regardless of the color temperature of the LED or what its CRI is, its color spectrum will show it has a blue spike, unlike, say, sunlight, which has blue in it, but the majority of the light from the sun is not blue light, as is the case with
LED. Studies suggest that continued exposure to blue light over time could lead to damaged retinal cells. This can cause vision problems like age-related macular degeneration.

But the disruption to circadian rhythms is likely more serious, as it has been correlated with diabetes and heart disease. Incan doesn't do that.

Your blind spot is not seeing that some people actually prefer the light LEDs emit over incandescents. I wouldn't use incandescent if it was 100 times more efficient than LED for that reason alone. I can't see well under the type of light it emits, plus I get a headache being under it for too long.

I acknowledge that (above) here:

I'm pretty certain that if there was an incan tech 100x more efficient than the most efficient LED, you guys would still choose to slowly blind yourselves.

And I have never had worse migraines than after sitting in undetectable PWM for 30 minutes.

 

[jtr1962]

But there is an objective advantage with incan in the quality of the light itself, and in neither slowly permanently blinding due to a color spectrum with too much blue, nor shortening lives by disrupting circadian rhythms, of which studies have appeared that show that is the case with LED.

The problem is that incandescent has too much red in it. Ideally, a light which can mimic sunlight exactly would be the ideal light source. As I mentioned in other threads, LED can actually come very close to that. The blue spike is more like a tiny blue hump with such LEDs. Not perfect, but still far better than any other manmade light source.

Maybe initially, but the human brain will adjust the white balance pretty rapidly so that what you initially perceive as yellow will look white.

There are limits to that which vary by person. For me the lower limit is around 3500K. Anything lower and I can sit under it for hours but it'll still look yellow.

There is more blue light in LED than other light sources, regardless of the color temperature of the LED or what its CRI is, its color spectrum will show it has a blue spike, unlike, say, sunlight, which has blue in it, but the majority of the light from the sun is not blue light, as is the case with LED. Studies suggest that continued exposure to blue light over time could lead to damaged retinal cells. This can cause vision problems like age-related macular degeneration.

Yes, but as mentioned the higher the CRI, the smaller the blue spike. I hope you know sunlight has a copious amount of UV in it. LEDs have none. UV is far more damaging than blue light.

It seems from that study that general lighting isn't the primary cause of the increase in macular degeneration. Rather, it's the much higher levels of blue light we get from the screens we stare at all day long. Not sure there's any good answers to that other than using exclusively high CRI LEDs, which have a minimal blue spike, AND turning down the brightness. Most people have their screens WAY too bright.

But the disruption to circadian rhythms is likely more serious, as it has been correlated with diabetes and heart disease. Incan doesn't do that.

Again, that's primarily due to screens on the devices we insist on using, not due to general lighting which is far lower in intensity. Common sense dictates not to stare at a screen before you go to bed.

I'd say diet and lack of exercise are much more highly correlated with diabetes and heart disease. While on the subject of disrupting circadian rhythms, why no outcry about the significant percentage of the population (i.e. night people) who are forced to fight their circadian rhythms because day people dictate the schedules the world runs on? While we're at it, two things have made this even worse over the last few decades. One is earlier school/work start times. 9 AM used to be standard for both, and even that's too early for many night people. Now I've heard of some schools opening as early as 7 AM. The other thing is daylight saving time. That pushes everything back yet another hour. The biggest stupidity is the idea to make DST permanent. I'm all for not changing the clocks anymore, but it should be permanent standard time.

And I have never had worse migraines than after sitting in undetectable PWM for 30 minutes.

PWM isn't inherent to LEDs. I personally avoid it when I design stuff, preferring instead to simply reduce drive current. Having grown up in the era of fluorescents which flicker at 120 Hz, I'm not keen on intentionally introducing any flicker into light sources, even if it's at a frequency supposedly too high for us to detect.