Some cities are taking another look at LED lighting after AMA warning
washingtonpost.com
washingtonpost.com
> unseen blue light
I have unread negative comments about this sloppy writing.
They also repeatedly use the word "intense" incorrectly. Intensity refers to candela. Not lumens or temperature.
> The original LED streetlights had temperatures of at least 4000K, which produces a bright white light with a high content of unseen blue light.
4000K is actually considered fairly mellow. Above 6000K, then things start to get blue. The earliest white LEDs were around 10000K and terrible.
> Now, however, LEDs are available with lower kelvin ratings and roughly the same energy efficiency as those with higher ratings.
And now the higher kelvin lights are even more efficient. White LEDs use phosphors to shift the blue light to longer wavelengths. This is a lossy process. Low-K tints require more phosphors and reduce efficiency. That said, 4000K looks so much nicer than 6000K that the loss is worth it.
> “As a species, we weren’t designed to see light at night,” Siminovitch said.
We've had mastery over fire for as long as we've been a species. Darkness is unnatural to humans.
Moonlight is around 4100K. So it is even more "natural" than 3000K. Should we blot out the moon next?
> it urged cities to minimize blue-rich outside lighting and recommended the use of LEDs no brighter than 3000K.
Confusing color temperature and brightness again.
And now from the AMA's press release:
> Recent large surveys found that brighter residential nighttime lighting is associated with reduced sleep times, [etc]
It seems no one has blinds in their windows. I always carry an eye-mask with me too, just in case.
The AMA press release is very low on actual content so I can't comment on most of it. They are smart enough to avoid mentioning the "blue light causes eye damage" myth for example, sticking to meaningless wishy-washy statements.
If the people raising a fuss over LEDs actually cared, they would have been campaigning against mercury-vapor and high-intensity-discharge lamps for the past several decades. A 4000K LED puts out way less blue light than either of those technologies.
I was on a hike in July during a full moon and those who were sleeping in open air (no tent or bivy bag) struggled to sleep through the night due to the moonlight.
This was in the North West of Australia, so daytime maximums were around 35C. It was too warm at night to zip your tent or bivy closed, so where possible you would try to position yourself so that you could shield yourself from the moonlight during the night.
There were quite a few afternoon nappers in the group until the moon waned a little by the second week.
At lower latitudes, a full moon over snow is stunning and surreal when you can get away from local light sources - no flashlight required. Light at night (of appropriate color temp) is quite natural.
Why not? Lets fund an orbiter to create an artificial lunar eclipse exactly on full moon.
Proper discussion of harmful emissions from LEDs should instead speak of intensities at a particular wavelength and perhaps ratios between wavelengths rather than an outdated reference designed for blackbodies.
The N-step MacAdam ellipse on a 2D CCx/CCy plane is what gets used when you dig into tint, but people love to reduce complex systems to a single number. And if you want a single number, temperature is the best we have.
I'm curious what your suggestion might look like in practice because it would be even more complex than the MacAdam ellipse.
Harm comes to our eyes when 'unnatural' non-blackbody light plays tricks with the system. Looking at an eclipse isn't harmful because of some new light that only exists during an eclipse. It's because the darkness allows our iris to open despite the presence of harmful UV light. (The same goes for vintage sunglasses that don't block UV.) So to discuss the biological safety of light one needs, at a minimum, both the average intensity to judge how open the iris will be, and some reference for how much harmful light is present. Imho that can be expressed as an intensity and a percentage. An eclipse is a low visible intensity with a high percentage of harmful light. These blue-heavy LEDs could be similarly described.
I have fond childhood memories of seeing my street bathed in the bluish-white light from these street lights, especially during those long winter nights when the ground is covered in white snow. I am quite certain that these street lights did not produce a generation of "physiologically unwell" people.
My city has now begun switching to LED lights and I think they look terrific.
But things are changing nowadays and our city also started using LEDs and by Moses, they are so much unbelievably better in all aspects! Even if they emit that "unseen blue" light, it is most definitely not worse than the insanely pervasive orange...
The moon is one (1) singular source of light with a known peak intensity, often blocked by trees or clouds, and only at its brightest one day every month. On a full moon night, the moon is high overhead, not directly in your field of view if you’re looking straight ahead.
LED streetlamps are a dozen or more point light sources, directly in the field of view, each one of them individually brighter than the moon. The 4000K LED streetlamps installed on my street last year have a very sharp spike in the short wavelength part of their spectral power distribution (come to my street and I’ll lend you some diffraction grating glasses to see the spectrum for yourself). The LED lamps are also incredibly bright, I’d say 2–4x the luminance of the prior streetlamps or the sodium lamps on nearby streets.
Because the new lamps are brighter, cities save money by using fewer of them, spaced more widely. This causes the light to be less even than past streetlamps, meaning some parts of the street are intensely illuminated and other parts are in shadows. The lamps are shielded at the top, but they shine out to the side at a very wide angle, meaning they still cause a ton of glare from a block away or more. When originally installed they had poor shielding in the direction of people’s second-floor bedroom windows. They’ve since fixed that problem and only people on the first floor still have light shining directly into their windows.
Seeing these and objects they illuminate in your peripheral vision causes the eyes to become bright-adapted (i.e. entirely knocks out night vision) making objects in shadow appear darker than under other lighting, and the whole scene to have very high contrast. (It takes something like a half hour to become dark-adapted once again.) The glare is extremely distracting. Overall they make the street feel like an industrial warehouse.
The light is not like fire, or past streetlamps.
Beyond that, the new lamps are not more energy efficient than modern sodium lamps. Indeed, if you make them more than twice as bright, they use substantially more energy than well made sodium lamps.
When I walk my dog at night I’ve taken to wearing orange safety glasses (basically plastic filters which block out the short-wavelength part of the spectrum). That I can see noticeably better after blocking half the light put out by these streetlamps indicates that there is some problem with their current design. (With the orange glasses, my rod vision continues to function properly.)
LED streetlamps should be at ~2700K CCT or below, with almost no light in the blue part of the spectrum. They should be no brighter than past streetlamps. They should be placed close together and not too high up, and shielded to mostly direct their light downward rather than sideways.
Under such circumstances they would be fine. (LEDs aren’t an inherently bad technology. They have the advantage that they last longer without needing to be replaced. This saves money for city maintenance departments.) Unfortunately current models on the market don’t meet those criteria.
Same story for car headlamps. The current LED headlamps are a safety risk (blinding pedestrians and other drivers with the glare). LED technology could be used to make fine car headlamps; it’s the ones currently on the market that are terrible.
You are confusing color temperature with how the lights are installed. If your streetlights were 4000K and spaced properly you wouldn't have an issue with them. The penny-pinching distribution (high lumens, wide angle, distantly spaced) of lights chosen by your city is at fault. That layout will create maximum glare and is the most miserable layout imaginable since distant lights will always be shining directly in your eyes.
I have spent 10 years studying color science. I know what color temperature is. I’m just explaining why the moon is a poor analogy. I have an issue with bright short-wavelength light sources at night, regardless of lamp configuration. It knocks out night vision.
There is a wide literature (dozens if not hundreds of research papers) on this topic. It’s not exactly some deep mystery.
The lamp configuration just makes things worse (and the use of fewer, brighter lamps provides much of the economic incentive to install the new lamps, along with federal subsidies).
That is a problem that LEDs could enable (by making very bright lamps available) but is not really a problem of LEDs but a problem of using any bright light source wrongly.
My understanding is that the use of lower-energy and more durable light sources provide most of the incentive to install new lamps. I.e. electricity bill is lower and you need to change the lamps less frequently which reduces maintenance cost.
This makes the installation cost more attractive to the city.
The electricity bill is (potentially) lower if you replace very old lamps with LEDs at the same brightness.
The best sodium lamps are quite efficient though, so LEDs have only marginal advantage compared to replacing the old lamps with newer sodium lamps. If you make the LEDs dramatically brighter, then they actually end up using more electricity.
I haven’t seen any especially convincing numbers about an LED lamp project that saved on electricity bills. I’m not an expert in the subject though, and I haven’t looked too deeply into digging up data.
As others have noted, we sleep poorly under strong moonlight.
I live in a very rural wooded area, and I can leave my blinds open except when the moon is at its brightest -- I sleep very poorly if I don't use blackout curtains.
Perhaps the fittest humans were the ones who stayed the most awake and aware of nocturnal predators during the brightest nights of the month?
But you make some good points. I support your recommendations.
http://press.endocrine.org/doi/abs/10.1210/endo-119-5-2201?j...
http://jap.physiology.org/content/110/3/619.short
Fires and the moon emit very little light at this frequency.
*Future generation of LEDs might shift from indigo (445nm) to violet to improve color rendition, but cost will go way up and efficiency will go way down so I doubt it will be common. And some stores are already adding UV LEDs to their lighting in order to make florescents really pop.
Edit: and a reference, the datasheet for the latest and greatest XHP50 LED: http://www.cree.com/~/media/Files/Cree/LED-Components-and-Mo...
Spectral charts are on page 11. Peak emission is at 445nm. Beyond 415nm there is nothing.
Cheaper LEDs will use a longer wavelength of blue, typically around 475nm.
So, "blue" whatever. If you are keeping my yard dark so I don't have to pull the shades just to block out the light, well, win!
Of course, other areas that still feel the need to light up every square centimeter because security, YMMV.
I'll add that I find the current LED streetlight colors better than the orange sodium vapor "malaise" that we've grown so accustomed to.
http://darksky.org/wp-content/uploads/bsk-pdf-manager/AMA_Re...
A representative section.
"It is estimated that a “white” LED lamp is at least 5 times more powerful in influencing circadian physiology than a high pressure sodium light based on melatonin suppression. Recent large surveys found that brighter residential nighttime lighting is associated with reduced sleep time, dissatisfaction with sleep quality, nighttime awakenings, excessive sleepiness, impaired daytime functioning, and obesity. Thus, white LED street lighting patterns also could contribute to the risk of chronic disease in the populations of cities in which they have been installed. Measurements at street level from white LED street lamps are needed to more accurately assess the potential circadian impact of evening/nighttime exposure to these lights."
[1] https://www.goodreads.com/book/show/350025.Disenchanted_Nigh...
- Install the ones that keep people awake along highways and suchlike.
- Install lights that don't disturb sleep so much in neighborhoods.They should be banned in favor of lower-color-temperature lights.
As phrased, it also sounds really far-fetched since cataracts are not a disease of the retina. Maybe that's a problem with the phrasing and not the claim, though.
I didn't read this entire article, but it seems to be discussing the general mechanism of light damage. There was no reference to LEDs. I'd want to see something that gave reason to believe that the blue in LEDs poses a threat.
Edit:Adding LED to the search results in many spurious sources in 1st two pages, with a single NIH study. Again, more reading required: http://ehp.niehs.nih.gov/122-a81/
This particular topic falls into the same category as 'phone radiation & brain cancer' news, where claims are made, proven, disproven and so on. If either proves true enough, we'll all be in the same boat together, I guess.
So I'm no expert but I don't think this is true. All wavelengths of light exist naturally in some amount.
> The energy then dissipates in your eye and causes damage.
Energy "dissipating" in the eye is how you see.
high intensity UV light most certainly causes both corneal and retinal damage, both acute and as an ongoing effect over time.
i have never seen nor can i find any reference to visible blue light. i think* that heavily phosphored leds with high CRI and low color temp do emit quite heavily above visible and translate it down. so maybe there is some leakage?
The fact that exposure to extreme amounts of light can damage the eye is not evidence for the claim that LEDs at typical brightness damage the eye.
As the article indicates, they are less effective at disrupting sleep than a LCD monitor or smartphone.
If we're banning an outdoor light for being less disruptive than your phone, should we also mandate Flux style warm color phones at night by law?
EDIT: It's not just me: https://forum.justgetflux.com/topic/2591/exactly-why-do-you-...
And yes, it is reasonable to do so. For one, everyone's got to share the same streetlights. For two, if these lights really are harmful to sleep, then the cost of using lights that are engineered to not produce certain wavelengths is probably minuscule compared to the net public benefit, even if you only want to limit the criteria being considered to things that are easily assigned a cash value.
Mandating Flux on phones, on the other hand, is a very different proposition in every way but the most superficial of similarities. People can make their own decisions about their phone's color scheme, and those decisions don't really affect anyone else in any direct way.
Where 4 lane / artery / highways get brighter, whiter (bluer) lights, while 2 lane feeders and smaller roads have warmer yellow lighting.
It produces an interesting nighttime aerial effect as the color indicates the nature of the street.
I can't find any source to corroborate so I'd love if someone with more worldly travel experience could corroborate.
I first became aware of the phenomenon playing Cities: Skylines, a European developed city builder game which features the multi-colored street lights in European skins.
There are 9 grades of lighting, ranging from that required on a motorway interchange, through urban motorways, rural motorways, major roads, etc, to residential areas.
Here's the relevant standard: http://vili.pmmf.hu/~mate/oktatas/VilTech_levelez%F5/EN13201... (BS = British Standard, but EN = European Norm, so the British Standard implements the European Standard. [I think that's how it works.])
This is the code of practise for Britain: http://persona.uk.com/orga46newark/ha_docs/disposit_docs/dd1...
The "Manual for Streets" has an overview, from p120: https://www.gov.uk/government/uploads/system/uploads/attachm...
(Similar documents should exist for every EU country, but the British ones are in English and easiest to find with Google.)
I recommend that. It is a great lamp colour temperature and way to wake up.
The flip side of lights that disrupt melatonin is that they can be good for you when you first wake up if natural light is not that strong in the morning where you are (or you aren't exposed to much of it when you first wake up).
Due to an effect called metamerism, our brains can perceive the same apparent color from sources with different underlying spectral power distributions (power vs. wavelength).
So in a sense, the color temperature of a black body does not have much underlying connection with the spectral power distribution of LED lamps.
With that substantial caveat, the lower the correlated color temperature of the bulbs that you purchase, the less of the overall spectral power is from the diode, and the more from the phosphors.
If there ends up being a link to blue light exposure for circadian effects then bulbs with lower CCT that have more emission from the phosphors are the safer bet.
Edit: I should have written "without blue light in spectrum eyes work harder". Basically any deviation from the spectrum of Sun have cosequences for eyes performance long term
http://www.eyedolatryblog.com/2014/02/myopia-may-be-reversib...
https://www.google.com/search?q=is+it+hard+to+focus+on+blue+...
http://www.eyedolatryblog.com/2014/02/myopia-may-be-reversib...
This is off the cuff, but it comes down to circadian rhythms being influenced by the color of light. Before artficial light, the sunrise and sunset have a softer light with a lot more red wavelength in it. This would correspond more closely with the color of light from an incandescent, like 2500-3000 Kelvin. The mid day light has a lot more blue wavelength in it relatively, where it is much more of a White light, I think daylight is like 15,000K. The high pressure sodium, HPS, lights have an orange light in the 2,200K range. The majority of LED street lights being installed are 4000-5000K, so they have a lot more of the blue light. Essentially the blue light tricks your brain into thinking it's mid day and messes with your sleep patterns.
The first generations of LED street lights could really only effectively make the 4000-5000K light. With new generations coming out every 3 months or so though they have been able to create models that can generate the 2500-3000K warm red light, but they're not quite as energy efficienct as the 4000K lamps, maybe 10% or something. The 2500K is actually better for the Dark Sky movements as well and a lot of cities are moving to specify these warmer temperature for those reasons.
There are a whole bunch of other things going on as well. The road way lighting standards were designed around the light patterns of a HPS lamp and haven't been updated to reflect the light patterns possible with LEDs, hence a lot of LED projects have too bright of streets.
For the sleep pattern issues, when you're installing LED lights in your home, you want to put the warm lamps, 2500-3000K, lamps in bedrooms and such. Then for living spaces or offices you'd want to put the 4000K lamps to help keep you awake during the day. There are also lamps/fixtures available that will automatically tune the color output of the light to be in sync with the outdoors to help keep circadian rhythms in sync.
This company has been pushing the tunable lamps hard. http://www.planled.com/archives/portfolio-item/entraining-ci...
http://humancentriclighting.org/ This site appears to be fairly dead, but they were more of an independent group promoting it the last year or two.
No, it is 5500K or 6500K depending on which metric you use.
That is right, which is why I have a nice project to use a heliostat with optical fibre to use as daylighting for a house interior.
The colour temperature shall have a nice warm glow in early morning and evening, which shall aid sleep preparation. I'll publish a log if I believe the project is successful.
Set it and forget it is the best smart house technology.
http://darksky.org/light-pollution/lighting-crime-and-safety...
It's got nothing to do with intensity, which would be impossible to achieve, for all practical purposes. The article misused the word here. Rather, it's about color temperature.
I think this problem is overblown, because I've not heard similar complaints leveled against mercury-vapour lamps. I've grown up around them nearly all my life.
Though to your color temp point, if we could get lower color temperature that'd be great. IIRC red LEDs are much more efficient than full spectrum anyway.
'Disturbing sleep' means they alter your hormone production & release, which is something that begins to affect you over weeks and months.
High color temperature lights disturb sleep- without promoting wakefullnes.
As a former, quasi-astronomer, nothing at night is more impressive than a star lit sky. My first night in Colorado was unbelievable.
http://www.flagstaffdarkskies.org/for-wonks/lamp-spectrum-li... says low pressure sodium lamps significantly improve star viewing compared w/ LED due to the narrower spectrum.
Downvote science, bitches.
Moving is a comically inappropriate response.
My bedroom is at the back of the house and feels like a cave, as is common for NYC bedrooms, so it's not the light in my eyes that is the issue. Having a delayed melatonin release due to the light is more my concern.
Edit: it's confusing that a cool blue light is a high temperature, and a warm orange light is a low temperature.
[1] When people switched from coal fireplaces to gas heat a common complaint was that it didn't smell like home any more.
Yeah about that. Seems we have a ways to go still wrt nocturnal blue light usage and health advocacy.
Big point #2 (an actual advantage now) is reduced maintenance cost because the lamps (theoretically) last longer between replacements.
The energy efficiency claims were always a red herring.