Has anyone on this forum used LED desk lamp?

[hank]

We use several of these LED desk lights with 12v batteries for winter camping -- all now are filtered using a Rosco color filter to block the blue end, for eye protection. There's only one LED available that I know of that doesn't emit levels of blue light that (1) may damage the eye longerm, and (2) definitely reset the human body clock, which is not a smart thing to mess with.



Young folks -- read up on this. It's like your hearing, in normal living you slowly damage your eyes if you don't protect them, and you won't know what you've lost til it's gone.

Here's a source for simple plastic sheet filter material:
http://www.rosco.com/us/filters/roscolux.asp#Colors

We use these on our older LED and fluorescent lights:
http://www.rosco.com/images/filters/roscolux/310.gif


Look at the LEDmuseum spectra --- all but one 'white' LED he's checked, including the so-called warm white, have a big emissions spike in the 400-480 nm range.

For example:
http://ledmuseum.candlepower.us/sixth/v16.gif
With what I've learned recently about blue light risks, this one looks good:

Zexstar L5WWE1 # warm white
http://ledmuseum.candlepower.us/sixth/l5wwe1.gif

I'm putting them into my various older LED reading lights. I'd be very, very interested in buying new lights made with these.


Now --for those of you who haven't just turned your headphones up to 11 and quit reading (grin) -- here's why:


This blue light -- short wavelength, highest energy photons -- is associated with macular deterioration over time -- the commonest cause of blindness in older people, because the damage adds up over the years.

"whether blue light (400 to 500 nm), or all visible light (400 to 700 nm) was associated with AMD. ...established cases (AMD-4), but not milder cases, had significantly higher exposure to both blue and visible light over the preceding 20 years (Wilcoxon sign rank test, P = 0.027). There was no difference in exposure at younger ages. These data suggest that high levels of exposure to blue and visible light late in life may be important in causing AMD."

470-500 nm is the narrow band that affects the melatonin photoreceptors, that sets your circadian clock -- and that you want to block starting about five hours before you want to fall asleep.

http://www.springerlink.com/content/w57m0451647x5444/


Blue Light Special—Treating Circadian Rhythm Sleep Disorders
....subjects exposed to blue light had significantly ...
www.neuropsychiatryreviews.com/sep06/blue.html

" identifies 446-477 nm as the most potent wavelength region providing circadian input for regulating melatonin secretion. "
J Neurosci. 2001 Aug 15;21(16):6405-12.
Action spectrum for melatonin regulation in humans: evidence for a novel circadian photoreceptor.
Brainard GC, et al.


================================

"Environmental lighting powerfully suppresses the physiologic release of melatonin, which typically peaks in the middle of the night. This decreased melatonin production has been hypothesized to increase the risk of cancer. ...There is also fairly consistent indirect evidence from observational studies for an association between melatonin suppression, using night work as a surrogate, and breast cancer risk."
Photochem Photobiol. 2004 Apr;79(4):316-8.
Light at night and cancer risk.
Schernhammer E, Schulmeister K.

---------------------------------
Med Hypotheses. 2004;63(4):588-96.

Lighting for the human circadian clock: recent research indicates that lighting has become a public health issue.
Pauley SM.

"The hypothesis that the suppression of melatonin (MLT) by exposure to light at night (LAN) may be one reason for the higher rates of breast and colorectal cancers in the developed world deserves more attention. ... as a growing public health issue. Evidence now exists that indirectly links exposures to LAN to human breast and colorectal cancers in shift workers. ....
"Lighting fixtures should be designed to minimize interference with normal circadian rhythms in plants and animals ..... blue-light-sensitive retinal ganglion cell light receptors that control the circadian clock ...."

 

[jtr1962]

470-500 nm is the narrow band that affects the melatonin photoreceptors, that sets your circadian clock -- and that you want to block starting about five hours before you want to fall asleep.

Pretty much any white light source has some emission in this range, including incandescent. Is it the absolute intensity of the 470-500 nm portion of the spectrum or its relative value compared to the rest of the spectrum which disrupts the circadian rhythm? I suspect it's most likely the former, so the type of light source doesn't really matter so long as the intensity of the 470-500 nm portion of the spectrum is under the threshold value. Unfortunately, white LEDs have a pronounced peak in that area so to get the emissions in that band down to acceptable levels would mean an lower overall light level compared to using fluorescent lighting. A big problem though with getting rid of too much blue is that it makes the light source look ugly and give poor color rendition. Do those filters you mentioned make the LED desk lamps look very warm like incandescents or do they just remove the usual blue tinge so they look closer to pure white? Regardless of whether it might be "safer" I personally can't stand to be under any light source which makes things look yellow. I get major headaches. I've been using mostly 5000K fluorescent for home lighting for quite a number of years. It's very pleasant. It doesn't seem to disrupt my sleep patterns, either. I'm inherently a night person so I tend to live a good portion of my life under artificial light.

...There is also fairly consistent indirect evidence from observational studies for an association between melatonin suppression, using night work as a surrogate, and breast cancer risk."

I tend to discount such studies because there are too many confounding variables. Many night and especially rotating shift workers have very poor sleep patterns. Night shift workers will often sleep less because they try to get things done during the day. It's well known that not getting enough or getting poor quality sleep can cause many health problems, including cancer. I suspect this is the real cause. I would love to see a study done on day versus night shift workers where both had an equal length and quality of sleep. I doubt there would be any statistically significant difference in cancer rates.

"The hypothesis that the suppression of melatonin (MLT) by exposure to light at night (LAN) may be one reason for the higher rates of breast and colorectal cancers in the developed world deserves more attention. ... as a growing public health issue. Evidence now exists that indirectly links exposures to LAN to human breast and colorectal cancers in shift workers. ....
"Lighting fixtures should be designed to minimize interference with normal circadian rhythms in plants and animals ..... blue-light-sensitive retinal ganglion cell light receptors that control the circadian clock ...."

There is a better explanation-exposure to environmental pollution causes these cancers. In the last 50 years America and most other developed countries moved from getting around by walking, biking, subways, trolleys, and railways to using automobiles and aeroplanes. Both of these modes of transport result in high exposure to pollutants both inside and outide the vehicles. Cancer rates in general are higher in suburbs where more people spend more time in cars then in cities where many use subways. Air pollution also contaminates food supplies which in turn can result in colorectal cancer. The typical poor diet of processed foods with many chemical additive like nitrates is another cause of colorectal cancer as is the higher incidence of obesity.

There are just too many confounding variables here to even try to isolate exposure to light at night as a cause of cancers. In fact, man has been using artificial light at night for millenia. If such light caused a higher incidence of cancers then we would not be seeing the spike we're seeing now but rather the rates would have remained high and stable for the last few millenia. It would probably be more productive reducing cancer rates if we were to clean up the environment by no longer burning things for power rather than focusing on lighting.

 

[Lucien]

Hi Hank,

That's very interesting stuff. What colour light do you end up getting when using that filter with LEDs, and what kind of LEDs have you tried with it?

Also, is it easy to get hold of that stuff? I might be interested in getting hold of some down the road.

 

[hank]

I believe what Mark Twain said about getting your medical advice by reading medical books applies even more to online forums (grin). I'm not going to argue about the research, I mentioned a few examples of the literature as examples that have footnotes and links for anyone curious to read the originals.

Rosco theatrical filters are available from most any professional camera or lighting supply and online; cost about $6 for a 24" square sheet. Another supplier is called Gam; they have sleeve material to pull over fluorescent tubes also. These filters are also used (along with a UV filter) in a lot of archival libraries and art collections to reduce the blue light and UV levels.

The simplest answer is to use glasses with a bit of a tint to block the shorter blue wavelengths.

With all these, of course you notice a significant color change at first, and continue to -- but only if you keep a lot of different color temperature lights on. If you have all your lights the same, the eye and brain adjust in about fifteen minutes.

Remember two very different issues are of interest
-- lifetime cumulative exposure to high energy blue, may correlate with macular deterioration
-- melatonin/sleep is held off for the duration of and a few hours after exposure to only the narrow band around 450-470nm

Both of these CFLs have spikes in the short blue range; only the second one has a spike in the range the melatonin sensor is affected by.

http://ledmuseum.candlepower.us/seventh/sunbrite.gif
http://ledmuseum.candlepower.us/fifth/cfl1.gif

LEDs, except for the one I mentioned above, seem to all have a significant spike in the blue.
http://ledmuseum.home.att.net/specx02.htm

Astronomers have very narrow filters that cut out specific ranges precisely to get rid of problems like this -- the theatrical gels aren't that precise by any means.

I'm going to see if I can buy some of the Zexstar L5WWE1 LEDs, til something better comes along.

Applause again as always to Craig at LEDmuseum, nobody else in the world makes LED and other spectrum information easy to get.

 

[hank]

For jtr1962 -- you asked if it's the presence of blue, or the relative intensity, that matters.

For the melatonin switch, here's one reference
October 2005
Neuroendocrinology Letters
saying it's the mix -- reducing the blue vs. eliminating it completely were tested.

Excerpt, from a science news site:

======
Mark S. Rea, a biophysicist at the Rensselaer lighting center.
His team has shown that color-signaling cones can mute the ganglion cells'
impact on the biological clock, as measured by changes in the hormone
melatonin.

Secreted by the brain, melatonin not only helps trigger and maintain sleep
but also plays a role in regulating the body's internal clock (SN: 5/13/95,
p. 300). Dusk and darkness normally trigger melatonin production, whereas
bright light can suppress it.

Rea's team exposed four men to mercury-vapor lamps in two hour-long
sessions at various times between 11 p.m. and 4 a.m. In white-light
sessions, the intensity was either 450 or 1,050 lux and always included
both blue and yellow wavelengths. In other sessions, filters removed all
but 7.5 or 15 lux of blue light. The scientists monitored the volunteers'
blood-melatonin concentrations throughout the evening test periods.

The high-lux, white-light mercury lamp suppressed the nighttime melatonin
by 50 percent. So did the 15-lux blue light—despite its low intensity.
The low-lux white light didn't perform nearly as well as those or even the
7.5-lux blue light, which reduced nighttime melatonin by more than 30
percent. The researchers reported their findings in the October 2005
Neuroendocrinology Letters.

The results suggest that yellow light can blunt the body clock's response
to blue light, Figueiro says. When both blue and yellow are present, equal
intensities of the two cancel each other. Only if there's an excess of blue
will the cones signal light's presence to the biological clock.

 

[jtr1962]

The results suggest that yellow light can blunt the body clock's response to blue light, Figueiro says. When both blue and yellow are present, equal intensities of the two cancel each other. Only if there's an excess of blue will the cones signal light's presence to the biological clock.

That about answers my question, and it's good news because equal amounts of blue and yellow produce white more or less. Another paper you linked to recommended that those prone to macular degeneration avoid light with color temperature in excess of about 5000K (I assume then that means 4900K is OK even for those at risk). Great news all around because it basically means that you don't have to use only ugly yellowish incandescent type light in order to be safe. Light with a CCT in the high 4000Ks is apparently safe and also visually pleasing for interior lighting. Now we need the LED manufacturers to start producing LEDs which are intermediate between the cool whites and warm whites we have now.

 

[hank]

No. That's not right.

When you assume -- let alone recommend to people -- that because you read a caution about exposure to 5000K, you conclude and advise that 4900K has no effect and is apparently safe -- that's wishful thinking. And it's wrong.

And you can prove that's wrong by talking to anyone who does archival lighting in museums and libraries, and looking at the damage done by light and how they filter and limit light. There is no "cutoff" -- it's not binary.

The energy of photons is proportionate to the wavelength (color temperature). The bluer the photon the higher the energy (ranging on up into what we call 'ultraviolet' and then 'X-ray' photons). They're all photons, there's nothing different about them but the energy level.

A slightly lower color temperature is just an average that describes _perception_ -- it's what your eye and brain experience.

You can -- and do --- certainly have significant blue and sometimes ultraviolet spikes in any color temperature "white" light -- and you can and should look this stuff up. Why fool yourself when you can check?

LEDmuseum is a good source, maybe the only source.

Here's the LED spectrum from the
"1W Luxeon LED Desk Lamp, retail $32.50 (www.amondotech.com...)"

http://ledmuseum.candlepower.us/sixth/desk.gif

See the spike in the blue? And that's just one of many on Craig's pages that give the spectra of various kinds of "white" LEDs -- like you're talking about.

http://ledmuseum.home.att.net/specx02.htm

I don't mean to rain on your parade.

But if you're telling people something is safe, it's important not to fool
yourself. It's more important not to fool others.

Feynmann said this better than anyone else:

Here: http://www.lhup.edu/~DSIMANEK/cargocul.htm

" ... the idea is to try to give all of the information to
help others to judge the value of your contribution; not just the
information that leads to judgment in one particular direction or
another.

"The easiest way to explain this idea is to contrast it, for
example, with advertising.... the thing I'm talking about is
not just a matter of not being dishonest, it's a matter of scientific
integrity, which is another level.
...
"The first principle is that you must not fool yourself--and you are
the easiest person to fool. So you have to be very careful about
that. After you've not fooled yourself, it's easy not to fool other
scientists. You just have to be honest in a conventional way after
that. ....

.... I'm talking about a specific, extra type of integrity that is not lying,
but bending over backwards to show how you are maybe wrong,
that you ought to have when acting as a scientist. And this is our
responsibility as scientists, certainly to other scientists, and I think to laymen."

In Feynman's report on the Challenger explosion, here:
http://www.fotuva.org/feynman/challenger-appendix.html
(remember the O-ring in the ice water?) he ended his personal observations with good advice for any of us providing technology to people:

"... to be frank, honest, and informative, so that these citizens can make the wisest decisions for the use of their limited resources.

"For a successful technology, reality must take precedence over public relations, for nature cannot be fooled."

-------- Please think about it.

Otherwise we'd be lawyers. Remember, when you pronounce "lawyers" the "aw" is silent: "l...yers"

 

[jtr1962]

No. That's not right.

When you assume -- let alone recommend to people -- that because you read a caution about exposure to 5000K, you conclude and advise that 4900K has no effect and is apparently safe -- that's wishful thinking. And it's wrong.

And you can prove that's wrong by talking to anyone who does archival lighting in museums and libraries, and looking at the damage done by light and how they filter and limit light. There is no "cutoff" -- it's not binary.

Yes, I know about this-they are usually concerned about the UV since it does the most damage by far. I also know that both incandescent and fluorescent sources produce UV. Interestingly, since they're a quantum source white LEDs produce no UV at all. I also know that some but not all so-called full spectrum fluorescents purposely emit extra UV for some hocus pocus marketing reasons. I personally DO NOT recommend such lights as added UV has no additional seeing or health benefits but does in fact do things like discolor furniture and damage eyes long term.

My 4900K was indeed a conclusion based on a continuum, not on binary thinking as you seem to think. The hypothesis of the researchers was that light with higher proportions of blue may cause macular degeneration. I'm assuming then based on their recommendation (i.e. to avoid light sources with CCT over 5000K if you're in a risk group for macular degeneration) is also based on a continuum. For example, 10000K would be considered very dangerous, 8000K dangerous, 7000K somewhat dangerous, 6000K possibly dangerous, 5000K just barely safe. Of course, those are my words but usually when studies of this sort offer recommendations for safety it's based on this type of thinking. As a result, the recommendations if anything are overly conservative. And note that the recommendations only apply to those in risk groups for macular degeneration, NOT to the entire population. For most people there is no risk at all being under even 6500K, provided the UV is filtered out and it's not ridiculously intense like staring into the business end of a flashlight.

I also realize that not all lights with the same CCT are created equal. Sure, it's better to avoid a light with copious spikes in the far blue portion of the spectrum and instead opt for one with either a broader but less intense blue emission, or with a blue spike centered on a higher wavelength. And the beauty of that approach is that LEDs can be tailored to give such broad blue emissions as well as having different center wavelengths. In fact, some earlier blue LEDs based on SiC had exactly such emissions. You can also use several InGaN blue emitters in tandem which have their center wavelengths 10 nm apart in order to get a combined effect which is similar. In fact, this latter approach is quite easy since emitters are binned by wavelength.

The energy of photons is proportionate to the wavelength (color temperature). The bluer the photon the higher the energy (ranging on up into what we call 'ultraviolet' and then 'X-ray' photons). They're all photons, there's nothing different about them but the energy level.

This isn't technically correct. Once you start talking about individual wavelengths of photons you're talking about colors. Color temperature only refers to white light which by definition is a mixture of colors. You can never say that "blue" photons have a higher color temperature than "red" ones. Sure, they do have more energy. However, you can say that white light which is bluer (i.e. has a higher color temperature) has on average a higher energy level for the photons.

I also want to add that I think you're overreacting to the spike you see in the spectra of white LEDs even though you are correct to be concerned about their blue tint compared to other light sources. First off, it's at 460 to 470 nm which is well past the UV range known to be damaging to eyes. Second, if you look at the spectra of many fluorescent lights, especially the newer triphosphor ones in use for the last 20 years, you'll also see a pronounced blue spike but at an even shorter wavelength. And this blue spike is even narrower than the blue spike in white LEDs. However, people have been using these lamps for at least the last two decades with no sudden increases in macular degeneration compared to the older cool whites or incandescents. The reason is very simple-the total harmful energy one receives per unit of time from a blue spike is inversely proportional to the wavelength and directly proportional to the integral (i.e. area) of the spike. In other words, assuming the same center wavelengths a very high but very narrow spike is no more dangerous than a less high but wider one or a very broad yet still lower distribution throughout the entire blue spectrum. And while excess blue can indeed be harmful if the area under the spike is too wide you'll likely find the light too blue and too unpleasant to work under anyway. You don't see, for example, 10000K tubes being sold for interior lighting. This isn't because they are potentially harmful but rather because nobody would like them. For similar reasons 10000K blue LEDs aren't particularly pleasant to work under either but unfortunately they will remain common until LED manufacturers can improve the production process.

My conclusions here then are quite straightforward. If the white LED looks noticebly blue to you, then yes, you have cause for concern because that means the blue spike is a larger proportion of the total energy being delivered to your eye. However, if you're fortunate enough to have a white LED which does indeed look like "sunlight" white then it is no more dangerous then a fluorescent tube or a filtered incandescent of similar whiteness. In fact, the incandescent and fluorescent are probably more hazardous since their spectra most certainly do contain harmful UV wavelengths whereas the LED does not.

If I seem overly passionate here about defending my position the reason is that I don't want to see blue light as the next thing society suddenly overreacts to. It seems these things go in trends. A few researchers will find something which may harm a very small percentage of the population, and often then only when exposure is excessive. Everyone will get on the bandwagon to eliminate whatever it is from all aspects of daily life. Often, such wholesale elimination will later be found to have more harmful than beneficial effects. For example, the hysteria over carbohydrates and the resulting Atkins diet, or the one over MSG, or the one over fats, all serve as examples of such overreacting. I don't need the "blue light police", for lack of a better word, suddenly deciding that it's in my own best interest to spend my life under light largely devoid of blue wavelengths. Sadly, that's where I see studies like these ultimately heading.

It's very important to realize that many of the things which make us healthy or unhealthy are interconnected. While excess blue light in isolation may harm some people, consider that severely reducing the proportion of blue in light used for illumination may have psychological effects which are even more harmful. In other words, who cares if my eyes are healthy when I throw myself off a roof at age 50 as a result of depression stemming from years of living in a drab environment devoid of blue light? Might it not be better if maybe I keep the blue light, and perhaps I'll have some slight vision problems because of it but only if I'm fortunate enough to live to 110 since it would take that long for the cumulative effects to add up? It's a good idea to eliminate things which are quite likely to harm us severely within a normal human lifespan such as cigarette smoke. However, I question the wisdom of going after things which might only have marked effects after 200 years of exposure. Blue light seems to fall in the latter category.

 

[jtr1962]

I just saw this in the Cafe but thought it was relevant to the discussion:

Blind mice see again after cell transplants

Apparently this could lead to a cure for macular degeneration so maybe all this talk about blue light is moot.

 

[picard]

that's great idea Hotbeam. Is it possible for you to produce the lamp similar to the one in the photo?

 

[Melchior]

Time

Better hurry up, Osram nearly has their lamp ready to take over the LED desklight scene.

 

[hank]

> the incandescent and fluorescent are probably more hazardous since their spectra
> most certainly do contain harmful UV wavelengths whereas the LED does not

Some do - you can certainly see fluorescence in the beam of some of the more purplish LED lights, anything that "looks bluish" probably makes white fabric fluoresce fairly brightly. There were cautions a few years back when halogen desk lamps came into common use, because having a point source emitting quite a bit of UV so close to the eye raised concerns.

I'm glad to see some white LEDs coming out that use green rather than blue emitters.

 

[hank]

Okay, a little more research, good news and bad news. LOTS of news.

The bad news is, I'm right about LED desk lamps -- the risk adds up. Sitting longterm close to the light is the worst case condition, for which the rules are being drawn up right now.

The good news is, the lighting manufacturing and standards groups are well aware of the problems and have already classified LEDs for eye hazards.

Biggest concern: sitting close to a bright LED for a long time -- that's the worst case for eye risk.

Here's a recent document from the industry discussing exactly these issues, and how the industry and standards groups are dealing with the concern.

Nobody building hobby-grade lights is going to worry. Anyone in the industry manufacturing for retail sales is going to be aware of the rules and standards, as you would be if you were producing any other consumer lighting product.

Source and, below that, my excerpts, just for the record.

Any manufacturer will already be familiar with the industry and standards -- I wasn't til I found them this evening with some searching. It's all well developed and covered, internationally.

----- Here's the international standards group's web site. This is huge.

http://cie2.nist.gov/

COMMISSION INTERNATIONALE DE L'ECLAIRAGE
INTERNATIONAL COMMISSION ON ILLUMINATION
INTERNATIONALE BELEUCHTUNGSKOMMISSION

DIVISION 2: PHYSICAL MEASUREMENT OF LIGHT AND RADIATION
Update: Sep. 7, 2006

CIE Division 2, one of the seven Technical Divisions of the International Commission on Illumination (CIE). ...
Anybody interested to contribute to and participate in the technical work of the Division, which is carried out through its Technical Committees, is welcome .....

WHAT's NEW
* CIE 2007 Session in Beijing - abstract dealine: September 15, 2006
* Divisional and Board vote for the draft technical report of TC 2-45 "Measurement of LEDs" in progress with deadline
-----

http://cie2.nist.gov/LED_Sympo_2004/CIE_LED_symp04_prog_abstr.pdf.

CIE Expert Symposium on LED Light Sources:
- Physical Measurement and Visual and Photobiological Assessment -
June 7-8, 2004, Tokyo

------ Begin Excerpts -------
....
Class 1-limits have currently been exceeded by high-power state-of-the-art- devices, especially in the colors blue, greenish-blue and the most important white LED lamps for general lighting. (Since the following Class 1M is mostly not available, these devices would belong to Class 2 (– with similar restrictions as for laser pointers.)

... the current status is still a (mandatory) classification requirement for (even white light emitting) LED products in general.

Although posing a dominating photochemical ("blue light") hazard (!), the exempt group limits in the wavelength area from "green" to "red" are unreachable high and do currently not pose any problem for manufacturers and users in any case. However, especially the exempt limits in the "blue" area (just 2.8 cd) and also for "white" LEDs (23 cd) may have already been exceeded by high-power state-of-the-art devices.

... the worst case limits for small white light emitting LEDs are now already at about high 30 cd.

I.e. as long as the specified intensity of a phosphor converted white LED is below this limit, the actual source size needs not yet to be considered.

The very interesting white light emitting phosphor conversion LEDs for general lighting pose a dominating photochemical hazard due to their strong part of blue light emission. In this case the retinal eye hazard depends on the absorbed energy (dose) rather than the power.

Therefore, the expectable exposure duration (within one day) plays an important role for hazard assessments.

Present theoretical consideration as well as first measurements elsewhere indicate that the most simple approach is not always sufficient since the exempt limits for "blue" or (most important for general lighting) "white" LEDs may be exceeded by high-power state-of-the-art-devices.

------ end excerpts--------

 

[jtr1962]

Thanks hank for taking the time to do this research. I tend to agree that a very bright light source of any spectral composition is none too good to look at for long periods. I've also heard of the special danger with blue LEDs. The problem occurs mainly in a dark room. Because the eye response to blue light is fairly low your pupil is wide open since the brain says this light looks dim. The problem is that a blue light which appears dim can still be emitting a large amount of damaging short wavelength energy, and the wide open pupil lets quite a bit of this in. In other words, don't stare at blue LEDs for very long in a dark room. I imagine a similar but less severe problem would exist for far red LEDs as well except that red photons have less energy than blue ones. Anyway, if you must look at a blue LED head on (this is CPF so I'm sure everyone here has done so at least once ) do it in a brightly lit room where your pupil is fairly small, and don't do it for very long. There really is no analog in nature to a very intense blue light so it makes sense that the body really has no defense mechanism against it.

The second thing I take from all this is don't use a light brighter than it needs to be for the task at hand. If I'm reading then about 500 to 1000 lux will suffice. However, when I'm loading tiny surface mount parts on a pcb 1500 lux is marginal. I need 3000 lux to work effectively. However, I don't spend large portions of my life doing very close work under very bright lights so I'm not overly concerned about my cumulative exposure.

And of course don't look directly at the sun or into any artificial bright light source. I'm sure everyone here already knows that. But since this is CPF I'm equally sure everyone has done these things at least once. :devil:

When I have more time I'll go through that 101 page .pdf you linked to. A cursory examination showed it to be very, very interesting.

 

[jtr1962]

The very interesting white light emitting phosphor conversion LEDs for general lighting pose a dominating photochemical hazard due to their strong part of blue light emission. In this case the retinal eye hazard depends on the absorbed energy (dose) rather than the power.

This definitely makes a very strong case for using red/green/blue instead of blue plus phosphor in order to make white. By using rgb you can make white light with a much smaller overall blue component. Better yet, color saturation with rgb is much more intense. Due to the losses in converting blue light to yellow via phosphor I feel that long term LEDs for general lighting are headed towards rgb anyway. Another beauty of this approach is the ability to make any desired color temperature with the same fixture. That means fewer lamp types for the retailer to stock which in turn should mean lower prices.

In my albeit limited experiments with rgb I have found that I like the quality of the resulting light much better than blue plus phosphor.

 

[hank]

I'm glad the next standards meeting is in China where most of the LEDs are being made.

I don't think you need to go to mixing individual red green and blue LEDs to avoid the hazard, just use something like the LEDs I mentioned earlier -- which I guess are based on a green rather than a blue diode (3 volt).

I have no idea who actually makes these. I haven't been able to get a reply from the company selling them, which is in Romania.

http://www.zexstar.com/english/index.php?cat=11&PHPSESSID=ce9ff82a041ebe91136ffc4109f992ed#

Zexstar L5WWE1 # warm white
http://ledmuseum.candlepower.us/sixth/l5wwe1.gif

If anyone can build lamps using these, I'm very ready to buy them.

 

[Christexan]

RGB mixing is definitely more efficient, and also allows a "pure white" based on your preferences, by tweaking the sources. The problem is if using only a few emitters (the three minimum for instance), color mixing becomes a major headache (rainbow shadows, etc). Special mixing lenses are required to optimize the light, or diffusion (which reduces output) to mix the pattern better, etc, as well as the additional circuitry needed to match the LEDs... so you get better pure "white", at the expense of a lot of other headaches.
Blue/phosphor methods lose efficiency in the conversion process, but the output can be "immediately" used (direct transmission). The drawback of course is the process allows a significant portion of the "base" blue out... this can be offset by thicker phosphor layers, but that reduces the light output further through absorption. There is a reason the lumens of "warm-white" LEDs are lower than "white" LEDs, the phosphor material and thickness hurt the efficiency (the base blue LED puts out the same typical mW of output on both LEDs).

So if you put a color filter over "white" LEDs, or just use "warm white" LEDs that are low in the blue component, either method hurts efficiency, so you likely end up at the same rough results of output.

**EDIT** Just looked at those Zexstar LEDS, interesting, look forward to hearing more

 

[hank]

Noticed today on a page selling LEDs:

"... look closely at the spec sheet for either the 1 watt or the 5 watt versions ... Luxeons are classed as a class 2 lasers."

Zexstar -- yes, I really hope someone can get a reply from the provider, or find out who actually makes those. I imagine the design will show up more places eventually.

 

[hank]

Revisiting this since the science news keeps coming out.

You mentioned mood and depression, and you're right that people need blue light for alertness. It's a question of timing, since it sets the biological clock even at very low intensities.

This doctor's perspective is on mood disorders; the news is impressive about how strong the effect of timing is. He's got a wide variety of information linked to his site, including most of the recent research, I haven't found anyone else paying attention to this issue so carefully:

Yellow lenses at night for sleep: Not Such a Strange Idea
http://www.psycheducation.com/2006/08/yellow-lenses-at-night-for-sleep-not.html

 

[AndyTiedye]

Added to LED Fixed Lighting Index