Is LED harmful to the eyes (if one looks directly into it)?

[natalia]

I have a led flashlight (Arc-AAA, 5.5 Lumens) and my child plays often with it. I know, LEDs are not harmful to the eyes. But can it be harmful if one looks directly into it? How bright can be a flashlight, so one can look directly into it without damaging the eyes?

[Jarl]

lol... how bright can it be.... It's impossible to say "9739 lux is safe, but 9740 lux will damage your eyes" or similar (completely made up). TBH, give your child a decent light and they'll quickly learn not to shine it at their face

IMO it's fair to say that a single exposure to current single die LED's isn't going to harm your eyes. However, with the more powerful lights, such as the P7 dies and some incans, you probably can cause damage over time, just as you can by staring at the sun. 5mm LED's causing damage? No.

[Yapo]

Its the Utra violet and maybe Infra-red? that damages your eyes when staring at the sun...thats why they invented sunnies and warning labels on lasers i dont think 5 lumens of visible light from an LED would cause any permanent damage to your eyes

[AvidHiker]

I agree, a small 5mm LED should not be a problem for a child's eyes (IMO, but I'm not a doctor). The latest high power LEDs are a different story, but I think most people quickly learn not to stare directly into them.

I recall reading on wikipedia (always taken with a grain of salt of course) recently that current high power LEDs may represent a "blue light hazard", sounds possible to me and they do reference a source (actually theres a second source as well) both of which I have yet to read over.

Taken from here:
"There is increasing concern that blue LEDs and white LEDs are now capable of exceeding safe limits of the so-called blue-light hazard as defined in eye safety specifications such as ANSI/IESNA RP-27.1-05: Recommended Practice for Photobiological Safety for Lamp and Lamp Systems.[31][32] "

[Black Rose]

I bought a cheap 9-LED Dorcy on the weekend.

It's not a powerful light (3xAAA, states 40 hour runtime) but there was a warning on the packaging not to look directly into the light (i.e. corporate CYA statement).

[Alero]

If it hurts, don't do it.

[IcantC]

It usually is not a bright idea to stare into bright lights or the sun...

[Gunner12]

The light would probably run down before any damage is done. I don't think there are any LED remotely close to the sun in lumen emitted per area.

Your child would probably learn that it hurts to look into bright lights and would learn to stop looking at bright lights.

Just be careful of lasers, UV and IR LEDs.

[Jarl]

it's not so much lumens that counts, but intensity- stars emit far more lumens than your average torch (lol), but which would you rather have lighting your way?

[phreeflow]

I agree, a small 5mm LED should not be a problem for a child's eyes (IMO, but I'm not a doctor). The latest high power LEDs are a different story, but I think most people quickly learn not to stare directly into them.

I recall reading on wikipedia (always taken with a grain of salt of course) recently that current high power LEDs may represent a "blue light hazard", sounds possible to me and they do reference a source (actually theres a second source as well) both of which I have yet to read over.

Taken from here:
"There is increasing concern that blue LEDs and white LEDs are now capable of exceeding safe limits of the so-called blue-light hazard as defined in eye safety specifications such as ANSI/IESNA RP-27.1-05: Recommended Practice for Photobiological Safety for Lamp and Lamp Systems.[31][32] "

Interesting info...

[bobofish]

I doubt it would cause permanent damage at that level, but a word of warning: when I was a little kid I liked to look at the sun (was into astronomy) and often did so on trips on the freeway. It didn't hurt my eyes at all when I did it. Later in life I realized that I had damaged my eyes somewhat, with a hazy area in the middle of my field of vision, and greater than average sensitivity to light. That hazy area has only relatively recently diminished or I am not able to see it anymore. I'm only 29, but it took a good 20 years for those effects to disappear. My sensitivity to light is still there, and I usually have to wear sunglasses or squint even on a cloudy day. Just so you know, I have excellent vision, but there are definitely those problems I've mentioned.

Telling your kid the dangers of looking into the sun is one of the few things I think parents actually should coddle their children with, as it's something that can very easily scar them permanently before they realize it's bad. My parents never told me not to look at the sun, and since it didn't hurt and I was curious, I did it. No, I never put my hand on the stove or cut a limb off with sharp objects, I never was an idiot. I was however curious like all children, and didn't know better about the sun.

[Burgess]

And Natalia --


Welcome to CandlePowerForums !


_

[hank]

Remember a "white" LED has a whole lot of its output in the blue range; see the LEDmuseum spectra for those. A red one would be safer as a toy!


PubMed found this on a quick look (there's probably more to be found)
40 minutes of exposure was enough to do damage in experimental work.
Question then would be how long it takes to heal, and whether exposure spread out is as bad as continuous exposure.
http://www.ncbi.nlm.nih.gov/sites/en...indexed=google
http://www.ncbi.nlm.nih.gov/pubmed/1...d_Discovery_RA

I know there are some cases where ordinary eye doctors' eye exam lights caused blindness (when someone was left under the light while the examiner got distracted) and they're filtering those now to reduce the blue light, at least where I go.

It's not going to be painful in time to warn a kid off --- and I know as a kid I did too much staring into lights just so I could then watch the colors change in the afterimage, luckily those were low wattage incandescents!

[Crenshaw]

Best not to do that. Young kids eyes would be more suceptible to damage. You could get a photon freedom, and turn it on to its lowest level, so that it just comes on. If you can look at the LED die without pain, then it should be alright.

Or a Novatac on its lowest level shoudl be alright too.

Crenshaw

[Turbo DV8]

If it hurts, don't do it.

If it feels good, do it.

[Lite_me]

when I was a little kid I liked to look at the sun (was into astronomy) and often did so on trips on the freeway. It didn't hurt my eyes at all when I did it. Later in life I realized that I had damaged my eyes somewhat, with a hazy area in the middle of my field of vision, and greater than average sensitivity to light. That hazy area has only relatively recently diminished or I am not able to see it anymore. I'm only 29, but it took a good 20 years for those effects to disappear.

I think Manfred Mann said it best...
Mamma Always told me not to look into the eyes of the Sun!!!! But Mamma, Thats where the fun is!!!

[jtr1962]

Probably the greatest danger comes from looking at blue and especially UV LEDs in a darkened room. The danger is twofold. First off, shorter wavelengths are more damaging. Second, because the retina isn't particularly sensitive to deep blue or violet light the pupil will not contract to restrict the amount of light entering. It will remain wide open because of the darkness. At least a bright white LED will cause the pupil to contract, limiting the damage. While staring into any bright light for extended periods isn't a good idea, if you really must do it then make sure you do so in either a very brightly lit room, or outdoors in daylight. Your pupil will already be closed, and restrict the amount of light hitting the retina.

Another good idea if you're going to give children any lights to play with is to put in your mostly discharged cells (especially rechargeables which are close to needing a recharge). The nearly dead battery will inherently limit the run time, and hence exposure time. A brief flash in the eyes from a bright light isn't going to cause damage unless it's a laser, but children have a tendency to stare into lights for many minutes (I guess most of the adults here still have that tendency also ). A nearly depleted battery will make that impossible.

[HDS_Systems]

To provide a bit of perspective, the surface intensity around noon in June on a clear day is roughly 120,000 lux (Tucson, Arizona, USA - southwestern desert). This is enough to cause eye damage if you look directly at the light source - i.e., the sun. The Ra Twisty at 100 lumens is generating roughly 375,000 lux at the exit aperture - using the same surface area. Looking into the exist aperture at close range exposes you to roughly 3 time the light as looking directly at the sun.

Granted, there is no UV, but bleaching does not require UV.

Henry.

[2xTrinity]

To provide a bit of perspective, the surface intensity around noon in June on a clear day is roughly 120,000 lux (Tucson, Arizona, USA - southwestern desert). This is enough to cause eye damage if you look directly at the light source - i.e., the sun. The Ra Twisty at 100 lumens is generating roughly 375,000 lux at the exit aperture - using the same surface area. Looking into the exist aperture at close range exposes you to roughly 3 time the light as looking directly at the sun.

True if the angular diameter of the light source at that distance from your eyes is 30 arc-minutes, the same as the sun. The source size will determine how concentrated that energy will be on the retina. I could generate a certain lux value using a bunch of fluorescent tubes, or a really intense arc light, the latter would be much more damaging to look directly at. Lasers are the worst, as they produce an infinitesimally small point-image.

In the case of the flashlight, there will be basically two relelvant "light sources" -- the image of the die on the reflector (a "large" source) and the die itself (a "small" source). For the sake of my post, Ill assume we're talking about the die alone, and not the reflector (the spill/light from the die is worse at close range.... trust me )

The die will be approximately 30 arc minutes when the light is 17cm from the eye, assuming an apparent die length of 1.5mm. I know that the dome on a Cree XR-E makes the image of the die bigger than it actually is in reality, which is a square millimeter. If at 17cm distance the measured lux is 375,000 lux, the light will be effectively like a 3x brighter version of the sun. And in fact even then, the component from the reflector won't concentrate to a point as much as sunlight.

However, there is still another factor to consider -- pupil diameter. In a completely dark room, the pupil might allow in 50x as much light as in the case of someone staring at the sun on a bright sunny day. This BTW is the reason why looking at the sun during an eclipse is a particularly bad idea, as the sky is dark thus pupil is dilated more.

My suggestion is to replace the clear glass with a window that has some sort of diffusion film on it -- This will soften that harsh spill light coming from the die, and effectively make the "light source" a lot bigger/safer. Up close as a flood light, it will actually be more useful, as the light won't cast such harsh shadows.

[2xTrinity]

To provide a bit of perspective, the surface intensity around noon in June on a clear day is roughly 120,000 lux (Tucson, Arizona, USA - southwestern desert). This is enough to cause eye damage if you look directly at the light source - i.e., the sun. The Ra Twisty at 100 lumens is generating roughly 375,000 lux at the exit aperture - using the same surface area. Looking into the exist aperture at close range exposes you to roughly 3 time the light as looking directly at the sun.

True if the angular diameter of the light source at that distance from your eyes is 30 arc-minutes, the same as the sun. The source size will determine how concentrated that energy will be on the retina.

I coudl generate a certain lux value using a bunch of fluorescent tubes, or a really intense arc light, the latter would be much more damaging to look directly at.

In the case of the flashlight, there will be basically two relelvant "light sources" -- the image of the die on the reflector (a "large" source) and the die itself (a "small" source). For the sake of my post, Ill assume we're talking about the die alone, and not the reflector (the spill/light from the die is worse.... trust me )

The die will be approximately 30 arc minutes when the light is 17cm from the eye, assuming an apparent die length of 1.5mm. I know that the dome on a Cree XR-E makes the image of the die bigger than it actually is in reality, which is a square millimeter. If at 17cm distance the measured lux from the flashlight is 375,000, the light will be effectively like a 3x brighter version of the sun.

A bare LED at that distance, assuming no reflector and a 90 degree beam angle, would produce roughly ~2000 lux. However, there is still another factor to consider -- pupil diameter. In a completely dark room, the pupil could very well allow in 60x as much light as in the case of someone staring at the sun on a bright sunny day. 2000 x 60 =120,000...

Adding a reflectro could certainly increase the lux, but it's also true that the light from the reflector won't be as concentrated when it reaches your retina, so that's why I didn't factor that in.

However, I will say this -- I once built an aspheric mod which I would NEVER want to look at directly. It was uncomfortable to look at through closed eyelids. Looking at the sun with closed eyelids does not produce nearly the same discomfort. So think about that for a while


To the original poster: my suggestion is to use a light with a light diffusion film on it. Then the "light source size" will be much larger, and the light will be much safer (not as throwy, but thats the breaks).

[Tachion]

Wow… that is quite a few parameters to consider:

• Output “strength” from the light source.
• Size of the light source (LED die and reflector).
• Distance from the light source to the eyes.
• How dilated your pupils are, i.e. ambient lighting when the light is pointed at the eyes
• Exposure time.
• Which spectrum the light is in might have an affect as well.

What I’m about to write is anything but scientific and probably not what you asked for but a while back I had an experience I’d like to share. I tried looking directly into a Fenix P2D premium 100 (rebel). I wanted to check how effective a flashlight might be if I ever needed to temporarily blind a perpetrator of some sort. Luckily, I did not try it on turbo, instead I believe I selected medium (50 lumen according to the manufacturer) and I guesstimate that I looked into the light for no more than one second at a distance of 25-30 centimeters. The size of the LED in a P2D rebel looks pretty tiny… perhaps the smallest of any of the lights I have. I tried this in the basement of my apartment house. No daylight, but the ceiling lamps (60W) where on so I don’t think my pupils where more than halfway dilated.

• Output: 50 lumen
• Size: Reflector – 16 mm / LED < 2 mm (pretty small)
• Distance: 25-30 centimeters
• Pupils: Not more than halfway dilated.
• Exposure time: Less than a second
• Spectrum: Not a clue. But I'm sure someone can tell me.

It was decidedly uncomfortable and the result was in fact a bit scary. My entire center field of vision was completely “blanked” for perhaps 30 seconds! I could literally not se my hand in front of me! x2 After that it took perhaps 10-15 minutes for the “spots” to wear of my vision. My eyes seem fine today, but as someone pointed out, you never know how things will progress with time.


With children I guess you can’t really control the Distance, Exposure time or how dilated the pupils might be. What you can control is how powerful and concentrated the light is. So the idea that a few posts here already suggest, to use a low power light with a diffuser on, seems the safest way to go. I think document tape should provide some diffusion. It might be a good idea to get it on the light as seamlessly as possible and without the children watching.

But I also believe that there is no substitute for knowledge. I will definitely make sure my kids (if or when I have a few) know they should not look straight in to any kind of bright light and be careful of where they aim flashlights. Until I can tell they understand and respect the power of modern flashlights, I’d stick with giving them low power lights with diffusers on.

[MetalZone]

I agree, a small 5mm LED should not be a problem for a child's eyes (IMO, but I'm not a doctor). The latest high power LEDs are a different story, but I think most people quickly learn not to stare directly into them.

I recall reading on wikipedia (always taken with a grain of salt of course) recently that current high power LEDs may represent a "blue light hazard", sounds possible to me and they do reference a source (actually theres a second source as well) both of which I have yet to read over.

Taken from here:
"There is increasing concern that blue LEDs and white LEDs are now capable of exceeding safe limits of the so-called blue-light hazard as defined in eye safety specifications such as ANSI/IESNA RP-27.1-05: Recommended Practice for Photobiological Safety for Lamp and Lamp Systems.[31][32] "

Interesting. In fact, a little alarming to me. The research does seem to have some merit.

[hank]

An Arc AAA spectrum:
http://ledmuseum.candlepower.us/eleventh/arcd-1.gif

White LEDs all have the blue spike (that's the primary light emitted) and the broad band emitted by the phosphor after it absorbs some of the blue photons and emits most of that energy as a range of somewhat less energetic photons).

(This is why over-driven "white" LEDs get that blue-to-violet color, they're running too hot, and the hotter something is the higher energy (bluer) photons).

What we see is the average our eyes make -- that's "color temperature" and that's why you can't just rely on "warm white" as a description -- if it's a LED or a fluorescent it will still be light with a large spike of blue in it.

That scary story (thank you for the caution) -- here's your spectrum (much the same)

http://ledmuseum.candlepower.us/eighth/p2d1.gif
Spectrographic analysis of the LED in the Fenix P2D CE Flashlight.

Typical of "white" LEDs.

Blue photons carry enough energy to knock electrons off of atoms, producing ionized radicals.

Amber and red photons do not carry enough energy to knock electrons off of atoms.

http://209.85.173.104/search?q=cache...ient=firefox-a

There's your mechanism -- why the blind spot lasted a while (it might have been safer to turn off all the room lights til you recovered!) --- good reason for concern. Your next optometry exam, if they dilate your eyes, should look at your retina and tell you if they see anything cautionary. Blind spots are easy to ignore, but ought to be noticeable in exams.

"... When light hits a photoreceptor, the cell bleaches and becomes useless until it has recovered through a metabolic process called the “visual cycle.” 30 31 Absorption of blue light, however, has been shown to cause a reversal of the process in rodent models. The cell becomes unbleached and responsive again to light before it is ready. This greatly increases the potential for oxidative damage....."
http://www.mdsupport.org/library/hazard.html#blue

One reason little kids may be at more risk -- the lens of the human eye becomes more yellow over time (probably as blue light damage accumulates from ordinary daylight exposure). That blocks out blue light, as the yellow tint increases. (It may also be why older people need more light to stay alert -- it's the blue light range that keeps the brain awake!).

--
Searches for "blue light" and "hazard" will find prior posts I've made with other cites; this is not a new concern. At least one thread got moved to the 'Cafe' as inappropriate.

Check Google Scholar and Medline for current research; this is very new science and changes fast.

“Be careful reading medical books. You might die of a mis-print.” -- Mark Twain

[2xTrinity]

(This is why over-driven "white" LEDs get that blue-to-violet color, they're running too hot, and the hotter something is the higher energy (bluer) photons).

Although that's true in general, I don't think that's the reason why overdriven LEDs run "angry blue". It's not that the wavelength emitted by the die is getting shorter as much that beyond a certain temperature, the phosphor becomes less and less efficient, so less of the 450nm blue light is converted to other wavelengths. The net appearance becomes more blue.

What we see is the average our eyes make -- that's "color temperature" and that's why you can't just rely on "warm white" as a description -- if it's a LED or a fluorescent it will still be light with a large spike of blue in it.

Mercury-based sources, like HID or fluorescent, I suspect are worse -- if you look at the mercury lines, there are large emission at 405 and 430 nanometers. LEDs emit primarily at 450nm. Not only are the shorter "lines" going to be more damaging than 450 (shorter wavelength = more energetic photons). The eyes are also FAR less sensitive there, so they will be perceived as "dimmer" than a 450nm led of the same power.

I still believe color temp is going to be a reliable guide of blue light exposure though, at least comparing similar light sources to one another. If you were to compare a 450nm blue LED, and a 3500k warm white LED both radiating one watt (optical output, not input power), you'd receive orders of magnitude less blue light from the latter -- both because there is less blue light in the output, and because the warm white light will appear far brighter (eyes are most sensitive to yellow/green light), so pupil diameter will be smaller. The same effect should be noticed going from cool white to warm white, as well.

I also suspect that when talking about damage from short wavelength light, point sources are particularly bad. In nature there is a LOT of UV and blue light, but never does it come as a point source -- it is scattered/diffused in the sky. That's why I have no problem with fluorescent blacklights, but am EXTREMELY apprehensive about using UV LEDs.

http://ledmuseum.candlepower.us/eighth/p2d1.gif
Spectrographic analysis of the LED in the Fenix P2D CE Flashlight.

Typical of "white" LEDs.

Blue photons carry enough energy to knock electrons off of atoms, producing ionized radicals.

I'd be interested in seeing specific numbers as to which frequencies are ionizing, and to what extent. Ie is 450nm as bad as 430? Is 470nm still bad? 485? 500?

"... When light hits a photoreceptor, the cell bleaches and becomes useless until it has recovered through a metabolic process called the “visual cycle.” 30 31 Absorption of blue light, however, has been shown to cause a reversal of the process in rodent models. The cell becomes unbleached and responsive again to light before it is ready. This greatly increases the potential for oxidative damage....."

Interesting. This would suggest that the mode of damage from blue light is consecutive exposures on the same spot on the retina, which is much more likely when dealing with a point-source.

All this discussion reinforces my decision not to buy any more LEDs until they start to come out in the 3500k color temp range. Not only be a more useful light source lumen for lumen, it would be safer and less glaring to use around others, as well. Leave higher color temp stuff to natural light, and computer monitor backlights -- all of which are inherently diffused

[jtr1962]

I also suspect that when talking about damage from short wavelength light, point sources are particularly bad. In nature there is a LOT of UV and blue light, but never does it come as a point source -- it is scattered/diffused in the sky. That's why I have no problem with fluorescent blacklights, but am EXTREMELY apprehensive about using UV LEDs.

I agree 100% here. Fluorescents have always had very spiky spectrums, way more so than most white LEDs in fact, yet this blue light issue has never come up in the over 60 years they've been in use. However, by their nature it is impossible for them to put a very high intensity of light through your pupil. Even putting your eye right up against a T5 tube isn't going to blind you as much as staring into a 5mm superbright white LED from a foot away. Old school T12 tubes were even less intense. You could stare at them for hours if you wanted to with no discomfort. LEDs changed all that by enabling a source with a blue spike to be put into a highly focused package.

Another recommendation here to prevent eye damage in children is to retrofit wide angle LEDs. I'm referring to the top-hat kind which spread the light over 100 to 120°. Peak intensity is reduced from as much as 100 cd to usually under 1 cd. You'll still have the same number of lumens to get around if the same chip is used, but the intensity will be harmless. This lets youngsters enjoy the benefits of white LEDs. Let's face it-you could I suppose give a kid a red LED if you want to play it safe, but it has virtually no "wow" factor, and it will stink in actual flashlight use. More than likely the kid will call it his/her (insert slang term for feces which starts with a "c") light. A wide-angle white is a good compromise between safety and usability.

[hank]

> fluorescents
The blue light problem does come up with fluorescents, but much less often (see the mention of using bright white fluorescents for 'winter blues' treatment in the link above -=- something I've done myself for years now).

> wide-angle white is a good compromise
I agree.

I'm using amber/yellow LEDs for kids for camping.
They are awfully hard to find as ordinary flashlight LEDs.
They do work a lot better than the red ones for general illumination.

[Tachion]

There's your mechanism -- why the blind spot lasted a while (it might have been safer to turn off all the room lights til you recovered!) --- good reason for concern. Your next optometry exam, if they dilate your eyes, should look at your retina and tell you if they see anything cautionary. Blind spots are easy to ignore, but ought to be noticeable in exams.

Thanks for the advice about turning off all the room lights to let the eyes recover. I’ll remember to do that if anything similar ever happens again. I will most certainly never check if a flashlight might be usable as a defensive device by pointing it at my own eyes again. But you never know if you’ll get “flashed” by mistake.

In fact I never let anyone try any of my “powerful” flashlights without warning them not to look into it when they turn it on. I also instruct them not to point it at anyone else’s eyes when they turn it on (easy to do if the on/off button is placed at an unfamiliar location… “oh… here it is -> ”).

As for the optometry exam, I don’t regularly do those. So unless they can actually do something that helps, I think I’ll give that a miss and pretend my eyes are fine.

(Edit: spelling)

[natalia]

Thank you all for your answers! I think, a discharged cell will be a solution for my 2 yo daughter :-)

[Yapo]

Thank you all for your answers! I think, a discharged cell will be a solution for my 2 yo daughter :-)

I guess its never too early to be getting into flashlights

[yekim]

As usual; a simple question posted here on CPF has devolved into a scientific dissertation.

I have a mess of little kids and they all have been playing with flashlights since they could grasp things. I have not had any issues with them shining them into their own eyes or the like. I believe flashlights are a beneficial learning tool for children as they teach multiple things through their usage.

The main problem I have is that they will leave them on and forgotten in some strange place.....There are probably a $100 worth of lights mixed in with the kid's toys somewhere. Maybe more. Sometimes they leave them on and the drain causes batteries to leak.

There is also the chance that a child can tear the light apart, and this is where I think a light like the Dorcy Metal Gear light stands out because you can really wrench it together and keep a child from tearing it apart, unlike my Super 1 watt that was once found spread through the entire expanse of a bedroom, or my Surefire L2, which gets warm enough to almost cause pain if left on high.

Kids like lights, I like lights.....we have something in common..

One other consideration, I have heard that there is lead used in the manufacture of the rubber button covers of some lights. I do not have a lead test kit or anything, but I did make sure that my kids were not using lights as pacifiers, and in the case of some really cheap lights I just removed the cover.