Heh, CFLs are dimmable too, but if you try it every metal object in your house will probably start buzzing.
Yeah, CFL dimming is a hack. But LED dimming Just Works. (Well, for me, anyway. Some people, with some dimmer switches, report problems.)
To me, that’s almost the bigger thing with LED lighting than the efficiency gains over CFL. With CFLs, you felt Virtuous. The bulbs were ugly, the light was meh, they were slow to fully brighten, and in general they felt like something where you were sacrificing for the sake of efficiency.
LEDs don’t feel like that. LEDs are fast, responsive, pretty, and elegant; they feel high-end and premium. To use an analogy, if incan bulbs were CRTs, CFLs were DLPs and LEDs are, well, LEDs.
Maybe we need something to balance it out, you know, like halogen++, but even less efficient, and somehow even cooler.
Actually CFLs have gotten better if you get the nicer ones, but you still can’t dim them. They’re definitely better behind a lampshade.
That’s a little bit more than “vaguely expensive”.
We do, but this is a problem with the dimmer and not the bulb.
Not to be contentious, but led light is a lot more meh than CFL. People sometimes complain about the spikes in a CFL bulb’s spectral output, but at least a CFL has output across the whole spectrum. LEDs don’t. You are probably never going to see LED bulbs that are worth a damn at color rendering. efficiency is about the only thing they have going for them.
Why do you say that? CFLs work by lighting up phosphors. LEDs produce certain wavelengths directly, and then use phosphors to fill in the rest of the spectrum. There’s no reason at all to expect them to be worse than CFLs, and empirically, a lot of LEDs have excellent CRI. (And here’s a visual example, if numbers don’t satisfy you.)
Where I’ve put good LEDs, I simply don’t notice color rendering problems. (With flashlights, sure. Flashlights have terrible color rendering. But: Not the point, there.)
Not really. It’s rated for 25,000 hours. If I go to Amazon right now, I can find a 6-pack of Sylvania R30 65-watt bulbs that are rated for 2,000 hours each for $15. I’d need 12 of those to go for 24,000 hours, so that’s $30, or about a wash.
Meanwhile, the LED ones are brighter (equivalent to 75W bulbs – and they are visibly brighter than 65W ones). And they save about 50W over the incandescents, or 1250KWh over their lifetime, which is… what, $125 in energy cost? 10 cents a KWh? PLUS, I don’t need to get up on a stepladder 12 times to replace the damn thing.
So they’re obviously totally worth it amortized over time, it’s just whether you can afford the upfront cost of $30, which is for most people “vaguely expensive.”
(I’m not comparing to CFL because CFL dimming isn’t good enough. For non-dimming purposes, CFL is obviously much cheaper than LED.)
85 is not an “excellent” CRI–it’s actually relatively poor. And the spectral output in your link doesn’t look all that close to their blackbody control. Better than the CFL, though they obviously didn’t choose a very good CFL for the comparison. Par for the course with the consumer type bulbs (CFL or otherwise), but I have CFLs in my studio that are rated at 99.
The problem with LEDs is that they lose a lot of their output and efficiency when you try to correct them for higher CRI, and even then, I am not aware of any LED lamp that can match the CRI of good full spectrum fluorescent, let alone a blackbody radiator. The best ones that I have seen have CRIs in the high 80s. They also lose a lot of lumens when you try to adjust them to a lower color temperature. So if you want a lamp with a mediocre CRI that puts out a lot of very cold, blue light, then LEDs are a good choice. If you want a lower color temperature and better color rendering, you are better off sticking with CFLs.
Let me put it this way: there’s a reason why photo and film studios use fluorescent lighting and not LEDs in their color booths.
My father-in-law, who holds patents on a lot of these technologies, including the original CFL bulbs (he was a senior lighting engineer for Philips before he retired; he’s also worked with LEDs) talked my ear off a while back about why LEDs were not ready for prime time, and might not ever be. If you’d like, I’ll hit him up for his option next time I see him.
The A19 LED bulb I linked to earlier in this thread had a 93 CRI at 2700K, and 97 lm/W DOE-tested efficacy. Maybe for color-critical production work, 93 CRI isn’t good enough, but for normal usage, it seems pretty impressive. And that’s incredible efficiency at a low color temperature with that CRI, so all the strikes of LED have already been accounted for there (except directionality, which is an advantage or disadvantage, depending on application; and price).
And keep in mind, the CFLs you’re talking about that professionals use are not at all like the ones that people are buying in Home Depot for $2. GE lists the CRI of their regular, consumer CFLs at 82. So when you say “worse than CFL,” you don’t mean “worse than the CFL you’ve seen” you mean “worse than CFLs that look like incandescent bulbs.” CRI around 90 (like the Cree CR6 downlights that Home Depot sells for very reasonable prices) is better than better than any home CFL, in addition to being dimmable, more efficient, longer-lasting, etc.
Also, can you link me to those 99 CRI CFLs? Because I don’t find them in Googling; I don’t doubt you that they exist (as you are a reliable purveyor of information), but I am curious as to how expensive they are. (EDIT: Just realized you didn’t actually say CFL, you said fluorescent. Something like this $43 bulb? If so, I think that’s important for niche specialty use, but not really a relevant comparison for home users.)
Let me put it this way: there’s a reason why photo and film studios use fluorescent lighting and not LEDs in their color booths.
I’m actually shocked to learn that they don’t use expensive 5500K incandescent bulbs.
My father-in-law, who holds patents on a lot of these technologies, including the original CFL bulbs (he was a senior lighting engineer for Philips before he retired; he’s also worked with LEDs) talked my ear off a while back about why LEDs were not ready for prime time, and might not ever be. If you’d like, I’ll hit him up for his option next time I see him.
If he talked to you some years ago, he was right about LEDs then. But, not ever? They’ve come a long way, very quickly, and there’s every reason in the world to expect them to keep getting better fast. (Googling for “99 CRI LED” turned up a press release from two months ago about a new phosphor that gives 98 CRI from an LED source, though it doesn’t specify at what color temperature.)
In our kitchen remodel, we’re having Cree 12W LED recessed lights installed. They’re 2700K, and put out an impressive amount of light (something like 580 lumens apiece) – more than the 14W CFLs we’ve got in the living room and dining room. I’m thinking about retrofitting the Crees just because the light quality seems more consistent than the CFL floods.
True, and you can also get CFL’s with different spectrum profiles - I’m quite willing to pay £3 each for CFL’s with “warm” lighting, I strongly prefer them. (I mean, the cheap 50p CFL’s work, but when the bulbs last years anyway? Negligible difference.)
No, that’s actually a pretty good CRI. It’s actually only slightly less than my studio CFLs (I misspoke earlier; I have a tube lamp from Philips with a CRI of 98, but my Bluemax CFLs are actually rated at 94). I’ll admit that it’s been a couple of years since I’ve really looked into them; it does look like they have made some progress. And that’s a good color temperature for regular use (though in my studio, I prefer something more in the 4,500-5,000K range).
The output is really low, though. 200 lumens is approximately equivalent to a 20-watt incandescent. It’s efficient, but I would need a ton of those bulbs to light my studio. The CFLs that I use put out 2,100 lumens apiece–it would take ten (!!!) of those Pharox bulbs to equal the light output of just one of my CFLs, and they sell for $20 each. The 2,100-lumen BlueMax CFLs sell for $12 each.
And keep in mind, the CFLs you’re talking about that professionals use are not at all like the ones that people are buying in Home Depot for $2.
Well, no. But the same is true of that Pharox bulb, too. I don’t see the average person shelling out $20 for a 200-lumen light bulb. In order to get the output higher, though, in a form factor that will still fit into the average lamp fixture, you’d need to lose the color correction. And even then, it’s going to be a dim lamp, because while LEDs are efficient, they just aren’t all that powerful. You could make an LED lamp that put out 2,100 lumens in all directions–just add more elements. But it would be the size of a basketball.
If you want to look at the bulbs that you can find at Home Depot, then we’d need to talk about EcoSmart bulbs. 429 lumens (~30 watt incandescent), CRI of 85, and they still cost $20 apiece. Not very competitive with the CFLs, which can give you similar specs for a fraction of the price. The LEDs win on energy efficiency, but the bulbs themselves are so costly that you’ll never make up the difference through lower electricity bills.
If he talked to you some years ago, he was right about LEDs then. But, not ever? They’ve come a long way, very quickly, and there’s every reason in the world to expect them to keep getting better fast.
It was a few years ago, and like I said, I can ask him if his opinion has changed. He was pretty down on them then, though, and seemed to think that the technology had some inherent limitations that were probably unsolvable.
Sorry, I was talking about this one, which is 600 lumens.
And my point isn’t that LEDs are the best solution for everything (today); it’s that there are fundamentally no major roadblocks in the way of them becoming the best solution. They can do high CRI. They can do low (or high) color temperatures. They can put out good amounts of light. And they can be very efficient while doing those things.
Even directionality is something they’re addressing with remote phosphor designs. The data sheet for this not-yet-released Cree lightbulb has excellent dispersion, with 800 lumens at 90 CRI at 80 lm/W. Plus it’s dimmable, plus it’s instant-on with no warm-up, plus you don’t have to worry about mercury if it breaks.
Niche applications, I don’t know much about. What you need for a studio, or for an oven light, or for streetlights, or whatever. But when you look at what you need for general purpose home lighting, LED has a demonstrated ability to fill that role. And it’s doing it today at an economical price (compared to incandescent, over the long term) for downlights; and its pace of improvement is such that it’s only a matter of time before it’s practical for other lighting.
Because CFL is a mature technology, more or less. There’s probably 10-20% efficiency gains to be eked out here and there, but fundamentally, we know what it can do. White LED is young. The first practical products were only introduced a few years ago, and before that it was all nightlights and flashlights. I mean, I bought a Petzl Tikka headlamp in 2000, and it took 3 LEDs for it to give out something like 20 lumens, and the color temperature was off-the-charts blue – and that was state-of-the-art for white LED at the time, and cost something like $50. So to go from that to usable, practical, economical warm-white BR30 bulbs in the course of a decade? Huge. And things aren’t slowing down yet.
It was a few years ago, and like I said, I can ask him if his opinion has changed. He was pretty down on them then, though, and seemed to think that the technology had some inherent limitations that were probably unsolvable.
Nothing I’ve read has given me that impression, but I am curious as to what he says. My personal suspicion? If he worked on CFL designs, he probably had a bias toward seeing CFL problems as solvable, while seeing LEDs as just fundamentally flawed – like the way that asking an Intel engineer about ARM in 2005 would probably have had them rattling off the dozen reasons why ARM could never overtake them and would always be a niche player for cheap toys.
I was really excited to read about the BR30 LED bulbs earlier in the thread. I’ve tried a bunch of different CFLs in my living room, but all of them seem so much harsher than the incandescent bulbs we already have, so I’ve just left them.
I ended up replacing the bulbs in my dining room light fixture with these:
And we’ve really liked them. They have a listed CRI of 80, which I didn’t know much about until reading this thread, but we haven’t really noticed the difference.
MA subsidizes them a bit, and they’ve dropped to $15 here. Still noticeably more expensive than other bulbs, but not quite in the OMG EXPENSIVE range some other bulbs inhabit.
Okay, but this one has all the same problems, and some new ones. For one, it’s absurdly expensive. That’s $55 for the equivalent of a 45-watt bulb, which is not that bright. People might use a bulb like that in a decorative lamp, but it’s way too dim for, say, task lighting. I’d need four of them ($220!) to equal the light output of one of my $12 CFLs, and the CFL still has a better CRI (albeit not by a lot).
Here’s the problem, though–you really can’t use that bulb in a decorative lamp, because it’s also freaking huge. It’s nearly three inches in diameter. The BlueMax CFLs that I linked are two inches in diameter, and they are already a bit on the large side. They won’t fit in smaller fixtures (decorative lamps, for example), though that’s okay, since they are way too bright for those fixtures, anyway. The bulb that you linked has the output of an appliance bulb, but it’s the size of a baseball. And $55. I’m going to go out on a limb and say that practical applications for this bulb are pretty limited.
The size thing, IIRC, is one of the big LED problems that my father-in-law mentioned. Individual LEDs don’t output much light, and even though what they do put out is very energy efficient, it’s not possible to get them to output as much omnidirectional* light as people need at a size that people can actually use. So you either end up with a good size bulb that is very, very dim, or a brighter (but still pretty dim) bulb that is way too large. That linked bulb has the additional onus of being color corrected to a high CRI, which further reduces output (like, as much as 40%).
So this stuff still strikes me as very much not being ready for prime time.
*(They have more luck using LEDS in highly directional lights, like spotlights and can lights, because they can focus the output on a much smaller area).
It’s 2.375", which is definitional for an A19 bulb. It’s basically the same size as a standard incandescent bulb, and should fit into any place that the incandescent ones would fit.
The size thing, IIRC, is one of the big LED problems that my father-in-law mentioned. Individual LEDs don’t output much light
If there is one thing about LEDs that is less true over time, it is that. Haitz’s Law (an empirical law, like Moore’s, but backed up by 30 years of data) states that output of an LED package goes up by a factor of 20 every decade. So ten years ago, no way were they bright enough. Today, they’re 20 times brighter. In 2021, they’ll be 20 times brighter yet, and a package that can put out 800 lumens today will be able to put out 16,000 lumens, and for the same cost as today’s 800 lumen chip, and probably twice as efficiently, to boot.
Which means that you’ll be able to get much smaller packages that can put out a regular 60W-bulb-equivalent 800 lumens, and they’ll be MUCH cheaper.
That linked bulb has the additional onus of being color corrected to a high CRI, which further reduces output (like, as much as 40%).
That’s sort of my point, actually. With all the handicaps in the world thrown at it – a small A19 format, a warm color temperature, a high CRI – it still gives you 600 lumens at DOE-measured 97 lm/W efficiency (that efficiency is for the bulb as a whole, after all the color correction ‘n’ such, not for the bare emitter inside it).
Yes, it’s expensive. No, I wouldn’t recommend it for someone today. But if things keep working as they historically have, and the cost per lumen goes down by a factor of 20 in the next decade, then that bulb in 2021 costs $3. There’s simply no way that LEDs aren’t the future of lighting.
And if that trajectory seems unlikely to you, Haitz’s Law or no Haitz’s Law, well, here’s a trip in the wayback machine to 2007, when this contraption was a high-tech $50 LED A19. 570 lumens, CRI of 80, at 13W. Today, for $13, you can get this product, 800 lumens, CRI of 85, also at 13W.
Four years, and the price dropped down to 1/4 of what it was, while brightness improved significantly (and CRI went up as well). Or if you want to stay at constant price and lumens (that LEDnovation bulb is pretty close on both those fronts), you see massive improvements in CRI (93 instead of 80!) and efficiency (6W instead of 13W!). Essentially, in all dimensions, things are improving, quickly.
So this stuff still strikes me as very much not being ready for prime time.
As I’ve said, downlights absolutely are; they make sense, and they compare favorably to the competition. General purpose bulbs aren’t quite ready, yet. But man, look at the trajectory of change. It’s not going to be long now before they are.