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

Mixing light - frequencies

  1. Apr 24, 2007 #1

    JPC

    User Avatar

    hey

    i am wondering, whats the equation to find the result equivalent frequence of a mixture of two other different light frequences in same proportions ?
    (like if u have one red torch , and one blue torch , and u aim the light of each on the same zone)

    or is the result color not equivalent to any one frequence ?

    --------------------------------------------------

    and now as for if u have dots that reflect lets say only one very short interval of frequence. Lets say there are two sorts of these dots , and lets just asume that the first reflects a big majority of x frenquence and the other of y. on a picture u have a distributed mixture of these two dots,
    how can u calculate the result equivalent frequence ?

    or in the result color not equivalent to any one frequence ?
     
  2. jcsd
  3. Apr 24, 2007 #2
    Please do not use chat talk.
     
  4. Apr 24, 2007 #3

    ShawnD

    User Avatar
    Science Advisor

    To be honest I don't think you can mix frequencies to get a simple equation. My understanding is that you can add and subtract spectra from each other but you cannot add and subtract frequencies the way you add and subtract whole numbers (5 + 5 != 10). For example, mixing pure red light (700nm) with pure blue light (400nm) does not produce pure yellow (550nm), even though it looks like it does. You'll get an interference pattern where most of the wave appears to be 550nm, but with little bits of noise all over the wave.
    Reversing the process to go from an interference pattern to seeing the component waves is called a "Fourier Transform"
     
    Last edited: Apr 24, 2007
  5. Apr 24, 2007 #4

    jtbell

    User Avatar

    Staff: Mentor

    Most perceived colors do not correspond to single wavelengths or frequencies. The color that we perceive when a mixture of wavelengths enters the eye depends on how the rod or cone cells in the retina respond to it, and how our brains interpret the resulting signals. I don't know much about the physiology of color vision, but I do know that it's rather complex.
     
  6. Apr 24, 2007 #5

    JPC

    User Avatar

    Hum for the first case,
    theres already infinite single wavelenght that are visible
    and now infinite ways to assemble single wavelenght to make different colors

    but how many different colors our brain can make us see ? i mean, does our brain see the difference between wavelenght 550 and 550 + (1 / 10^99). Hum , must ask in the biological section of the forum

    ------------------------------------------

    but for the second case

    Doesnt it rather have to do with ourselves. If we had far more pixels in our vision, we would see images printed by our printers differently.
    Because, every sensitive cell in our eyes must make up one color. I dont know how it decides of the result color. Well for that i should ask in the biological part of the forum
     
    Last edited: Apr 24, 2007
  7. Apr 24, 2007 #6

    ShawnD

    User Avatar
    Science Advisor

    How many colors can you see? Bare minimum, at least 65536. I know that because I can tell right away when a video game is running in 16 bit instead of 32 bit color.
     
  8. Apr 24, 2007 #7

    rbj

    User Avatar

    not exactly rigorous reasoning. you haven't shown that after pruning the 4294967296 colors down to 65536, that you can differentiate every one of the 65536 colors that remain. if there are sufficiently indifferential colors remaining, some of those can be replaced with colors you had missed after the pruning. suppose you went from 4294967296 to 4294967295 colors and the color that was culled was one of them that you liked and missed, would you say that you can see 4294967296 colors? i

    t's a matter of perceptual modeling and bit allocation.
     
  9. Apr 25, 2007 #8

    NoTime

    User Avatar
    Science Advisor
    Homework Helper

    If you have red and green torches your vision might say the zone was red, orange, yellow, yellow-green or green depending on how bright each torch was.
    There is a kicker here.
    For example, if you change the surrounding area of the zone from black to white then "yellow" would change to brown.
    Nothing simple here.

    Red and blue torches would make the zone magenta or purple hues.

    The results of combining two colors of paint and the same two colors of light beams is different. So the color you see would depend on if the dots(pixles) were on a display monitor or printed on paper. The count of x dots vs the count of y dots changes the color and the color of the area surounding the dots also changes the color.
     
  10. Apr 25, 2007 #9

    NoTime

    User Avatar
    Science Advisor
    Homework Helper

    No, you won't.
    Even the bandwith of a laser, a relatively monochromatic source is somewhat more than this.
    How much change is required to see a different color probably varies considerably over the visible spectrum.
    The eye's color receptors don't overlap evenly.
     
  11. Apr 25, 2007 #10

    Claude Bile

    User Avatar
    Science Advisor

    Radiometrically speaking, e/m waves obey the principle of superposition and you cannot get any new frequencies by combining two (or more) e/m waves.

    Claude.
     
  12. Apr 25, 2007 #11

    NoTime

    User Avatar
    Science Advisor
    Homework Helper

    In free space then ok.
    AFAIK If the e/m waves encounter the appropriate medium then they can mix and generate the standard sum/difference frequencies.
    For example, mixing red light (700nm) with blue light (400nm) would produce 1100nm and 300nm.
    Neither of which is remotely in the visible range.
     
  13. Apr 26, 2007 #12

    JPC

    User Avatar

    but where u got the equation :
    frequency x
    frequency y

    result : x-y and y-x

    whats the principle , the physical explanation ?
     
  14. Apr 26, 2007 #13

    Claude Bile

    User Avatar
    Science Advisor

    NoTime refers to sum and difference frequency generation which can occur in a chi^2 nonlinear medium. Normally, in the linear regime, the polarisation of an atom and the applied electric field both oscillate at the same frequency. In the nonlinear regime, additional frequency components are introduced into the oscillation of polarisation (i.e. the oscillation of the electron about its parent nucleus) by virtue of the fact that polarisation no longer varies linearly with the applied field (this is where the nonlinear comes from).

    All media are nonlinear to some extent, but only media with certain crystalline structures will exhibit chi^2 nonlinearity. Lithium Niobate and BBO and two very commonly used chi^2 media.

    Sum and difference frequency generation requires a high-powered, coherent source (i.e. a laser). In addition, the nonlinear medium itself has to be aligned very precisely to achieve significant generation of new frequencies. Suffice to say, that unless you are a laser specialist, or engage in some form of research with lasers or optics, you are unlikely to ever witness this effect first hand, at least until laser tvs come out.

    Claude.
     
    Last edited: Apr 26, 2007
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: Mixing light - frequencies
  1. Mix polarization (Replies: 4)

  2. Mixing of light ? (Replies: 2)

  3. Element Mix (Replies: 2)

  4. Color mixing (Replies: 6)

Loading...