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

Mirrors facing eachother

  1. Mar 7, 2006 #1


    User Avatar
    Science Advisor

    If we have two mirrors, each one facing the other (producing what i'll describe as recursive reflection) then, supposing we could zoom into one of the mirrors as much as we'd like, then at what point would the image become unrecognizable. Would there be a microscopical-sized reflection of the other mirror? How small can/does it get?
  2. jcsd
  3. Mar 9, 2006 #2
    Depending on how we choose to look at light, it would be as small as the wavelenght of light
  4. Mar 11, 2006 #3
    I disagree.
    There is no way that a single spot as small as a wavelenghth of light could contain an entire picture of its reflective ancestory under that condition.
    Most likely, there would be a serious degredation of resoultion or field of view, or both.
  5. Mar 12, 2006 #4


    User Avatar
    Gold Member

    Uh, you'd wouildn't get recursion if you zoomed in. You'd get a close-up image of the device, camera or head of whom/what-ever is doing the viewing.

    So there's only two ways this could be set up:
    -if your view angle was oblique, but that means eventually, your light path will slip off the side of the mirror and you'll simply be viewing a really zoomed in image of the laboratory wall.
    - you put the reflective surface *in front* of your viewer. Which means you use a semi-silvered mirror, which means every bounce of the light loses brightness, which means after a few bounces, there's no light left to get through.

    I am not sure your experiment is possible even in principle.
    Last edited: Mar 12, 2006
  6. Mar 12, 2006 #5
    If put two mirrors parallel (at about .5 meter) you have a Fabry-Perot optical resonator. If the distance is less (about .5 cm) you may have a filter.

    This device has its own proper frequencies so if one of them is not all reflecting (such as 99% of reflecting) the radiation only emerges from the cave if its frequency is equal (approx.) to one of the proper frequencies of the resonator.

    The common use is in gas laser such as Ar-Ne or CO2. You put an amplification media inside the cave, and if the maximum of gain is very close to one of the proper frecuencies of the cave (in other case the radiation will not be amplified) then, the radiation emerges amplified and you have a cool laser€ :D
  7. Mar 14, 2006 #6
    At a separation near the wavelength of the initial light, I believe that significant tunneling of higher wavelengths and lower amplitude through the mirror (one dimensional box) would occur. I guess that the image resolution of this tunneled radiation to that of the initial image would vary approximately as the squared ratio between initial light frequency and tunneled light frequency.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook