Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

LED light wavelengths

  1. Sep 29, 2016 #1
    I work in a factory where we package pharmaceuticals. There are quiet a few products that are light sensitive to a certain wavelength. The lights in the older buildings are fluorescent and were have defined wavelength specification. The wavelength determined the color light/filters set in the room.

    The issue I have is that the new buildings have LED lights and they do not have wavelengths. The color is determined by temperature. I can not find a way to correlate between the two to determine what kind of lights to use to prevent photolysis. Do they correlate or is this something that needs to be tested differently?
     
  2. jcsd
  3. Sep 29, 2016 #2

    berkeman

    User Avatar

    Staff: Mentor

    Welcome to the PF.:smile:

    Can you link to the LED light datasheets? They should publish color temperature/wavelength plots...
     
  4. Sep 30, 2016 #3
  5. Sep 30, 2016 #4

    berkeman

    User Avatar

    Staff: Mentor

  6. Oct 4, 2016 #5

    rbelli1

    User Avatar
    Gold Member

    Most white led products are florescent lights. They typically emit in a deep blue wavelength and a phosphor of some formulation is used to produce lots of broad spectrum yellow light to produce the appearance (to human vision) of white light. Standard tube lights use mercury vapor to produce several wavelengths in the ultra-violet and deep violet and blue to excite the phosphors. Better bulbs (both LED and mercury vapor tubes) will use more types of phosphors to produce a light that more closely mimics black body radiation to human vision.

    Please note that this is optimized for human vision. For all other organisms your mileage will vary.

    What are the harmful wavelengths in question? See https://en.wikipedia.org/wiki/Mercury-vapor_lamp for the wavelengths emitted by mercury vapor. Much of the shorter wavelengths are absorbed by the glass envelope however significant amounts to still pass. The LED bulbs will only emit strongly in the 450nm range and longer with a strong roll off into the 800nm range. Some IR will be produced due to actual black body radiation from the waste heat generated. This will be in the very far infrared.

    BoB
     
  7. Oct 22, 2016 #6
    Hello, I'm curious about this topic, although my interest is a little off topic

    as a photographer, though I am encountering more led lighting, (I avoid it in the studio) but to shoot under led's is quite revolting, I am assuming that this is because the wavelength of an led is very precise and narrow, with little roll off. As such, the image is saturated with one color, often quite ghastly, and it becomes essentially a monochromatic image. Converting it to a black and whiter image improves the attractiveness of said image, but it's not my best choice. Channel mixing essentially doesn't work, as there is no other channel that has any data
    Any thoughts would be welcome
     
  8. Oct 22, 2016 #7

    OmCheeto

    User Avatar
    Gold Member
    2016 Award

    What camera are you using?
    Have you tried photographing a pan-chromatic reference image, in both sunlight and LED light, and comparing the two?

    ps. I'm a tad new to high tech digital photography, and pretty much at the kindergarten level of spectometry spectroscopy spectrometry the science behind light sources and their effects on how humans get cranky when their/our photographs turn out "funny".
    I suspect it has something to do with the "white balance" on our cameras.
    I recently wasted several hours of work by having mine set at "auto". :mad:

    [edit] Oops. I see you just joined the forum, yesterday.
    I would suggest, taking the plunge, and finding the "General Physics" page, and starting a new thread.
    It's not that hard. I successfully did it, several years ago, when I first arrived.
    And like me, once you find the correct slot, to insert your question, you'll find that the people at PF, are wickedly smart.
    [/edit]
     
    Last edited: Oct 22, 2016
  9. Oct 22, 2016 #8
    Thanks for reply. I'm using a Canon eosMII calibrated, shooting raw, (although that's not much different from a jpg under those circumstances)
    So, I've been shooting for 40 years plus, so I might be of help to you, maybe together we will find a solution? Yet I suspect not, as it seems to my mind an issue we cannot over come.
    It took a moment to understand your suggestion, I see it might be informative, to shoot a color chart in a full spectral range, ie sunlight, and then see what it looks like under leds? I think that it would only illuminate the patch that corresponds to the wavelength of that particular LED? I should test it.
    To be specific, the worst culprit is that magenta/purple color, my assumption is that the frequency is so narrow, it's like shooting in monochrome.
    Unlike incandescent, household lamps, old theater spots, cinema 2ks, tungsten halogen, all these lamps seem to have a much wider spectrum of color, even after a correction either in camera with a filter, or in post with a color correction. Meaning, if I shoot with my white balance set to daylight, under room lights, that's a difference of 5,000k to 3, 000k, so the image will look overall yellow/red ish, but with these lights I mentioned above, if I remove the yellow bias, (cast) I find the whole range of colors, blues through greens etc. (obviously not quite as pure as were I to shoot it with camera set to 3,000k
    So, my point is that prior to leds, there was a much broader, richer and pleasing quality to the lights. If my assumption is correct and it is because the frequency is
    extremely narrow, I cannot see a solution? ( I suspect I'm not alone in this, and that many people are pushing to change this)

    PS... going out on a limb here, it does have everything to do with white balance, yet nothing at all. When we set our white balance(avoid using anything on auto) we are setting the camera to respond to the "overall color
    temperature" IE outside, 4,500k - 6,500k indoors 2,000k-3,400k approx
    I don't see how it's possible to set a white balance to a very narrow definition, if that's how to describe it?

    pps, I see that those ghastly compact fluorescents are improving, they used to be a bilious blue, nor thay come in different temperatures, and they havea f=great spectral quality.

    Adrian
     
  10. Oct 22, 2016 #9

    rbelli1

    User Avatar
    Gold Member

    Cheap ones still are. Quality ones have not been for pretty much since inception. Look for higher CRI values to get better colors. The same applies to most LED lighting. Most of them work on the same principle as flourescent. The RGB mixing types are going to be the hardest on photography.

    BoB
     
  11. Oct 22, 2016 #10

    BillTre

    User Avatar
    Science Advisor

    I have recently taken a series of multimedia arts courses.
    LED lights are becoming more common in photography and especially video. There are several reasons of course: weight, heat production, energy efficiency.
    Pictures made using LEDs can come out fine. LEDs used in photo/video however can be expensive.
    This indicates either not so good LEDs (see @rbelli1's post (#5) for possible sources of quality differences) or white balance needs correcting.
    White balance can be dealt with with the lighting (obviously), setting white balance in the camera (at least some of them), and in variety of ways in image software (such as photoshop, lightroom).
     
  12. Oct 22, 2016 #11

    BillTre

    User Avatar
    Science Advisor

    When I recently took multimedia classes recently, we used a variety of LEDs for photography (and especially for video).
    The reasons for this include weight, heat production and energy efficiency.
    However, LEDs for photo/video can be expensive.
    I suspect your problems have to do with not great LEDs (not made for photography?;n see @rbelli1's post #5) or white balance issues (@OmCheeto's post #7).
    There are both in camera settings and software adjustments (in programs like Photoshop or Lightroom) to correct for some white balance problems.
     
  13. Oct 22, 2016 #12
    Thanks for answers. I think I should clarify. This issue pertains to stage and event lighting, not to studio at all, and I suspect the competence and knowledge of hte lighting designers and operators is a huge factor. And a;lso, I think there is a big difference of approach between "stage" and "event" lighting, oh, and this applies tio analogue or digital equally, (unless we are dealing with flicker)
    So, I need to purchase and play with some leds, it's clear they are here to stay but in the meantime,
    I'm uploading a composite image that may illustrate my point......
    The colorful image on the left was from an old style dance hall shot in 90's, (not manipulated) the two on the left are modern leds. They are not as bad as I have, and also, the have other forms of lighting present, when the light becomes a solo color from every source is a factor too,
     

    Attached Files:

  14. Oct 22, 2016 #13

    OmCheeto

    User Avatar
    Gold Member
    2016 Award

    Way to ruin my "sciencey" followup answers......

    spectrums.are.freakin.cool.png

    LED.vs.LED.2016.10.22.png

    :oldgrumpy:
     
  15. Oct 22, 2016 #14
    Wow, why ruin? that's amazing, looking at it... give me a mo
     
  16. Oct 22, 2016 #15
    Damn, that brilliant, ****, damn wow. (sorry, new, can I curse like that?
     
  17. Oct 22, 2016 #16
    you have no idea how you just made my day, week and month!!!!!!!
     
  18. Oct 22, 2016 #17
    That's exactly the problem. :)
     
  19. Oct 22, 2016 #18
    Acknowledged :)
     
  20. Oct 22, 2016 #19

    OmCheeto

    User Avatar
    Gold Member
    2016 Award

    check my [edit]

    wickedly smart, these people are.

    I think, the only reason they haven't banned me yet, is, that I remind them of a puppy; "Intellectually, as cute as a kitten, but dumber than a box of rocks."
    :redface:
     
  21. Oct 22, 2016 #20
    Sorry, where do I see edit?
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted



Similar Discussions: LED light wavelengths
  1. Body wavelength (Replies: 17)

Loading...