I would like to collimate a light beam stemming from a single-mode LED source. As the source is pretty big (a few millimeters), collimation with a single lens gives a bad result, i.e. a quite nonparallel beam.

I was wondering if a fiber optic taper could be used to create a smaller secondary source, for instance with the following setup:

enter image description here

My questions are:

  • Would that work ?
  • If yes, what are the limitations ? For instance with taping ratio, or the transmission rate.

If you are aware of some other way to reduce the size of a single-mode light source, I'd be happy to hear your suggestions.

  • $\begingroup$ Some standard things done with tapered fibers, but not an answer to your question. When attempting clean-up an optical source, one loses power. The question is always how clean do you need? and how much power do you lose? $\endgroup$ – Peter Diehr Mar 27 '16 at 22:21
  • 1
    $\begingroup$ As @PeterDiehr points out, you can't do it without losing power. A more common trick is to use a lens to focus the source, and then put a pinhole in the focal plane. Obviously, you throw away a lot of light. $\endgroup$ – garyp Mar 28 '16 at 3:41
  • 1
    $\begingroup$ Take a look at Etendue: en.wikipedia.org/wiki/Etendue. In general the answer to your question will be negative. Why use an LED, anyway? Lasers are very cheap, these days, and they do exactly what you want. $\endgroup$ – CuriousOne Mar 28 '16 at 3:52
  • $\begingroup$ @CuriousOne: Thanks for your comment. Unfortunately I need to use a given wavelength (488nm) as the final aim is to illuminate a fluorescent protein with a sharp absorption spectrum. And at this wavelength, I can't find any laser diode at less than 2k$. Then, you are right that lasers are directly collimated, but I want to send the beam on an array of micro-mirors so I need a rather large beam (approx 25mm or 1"). I could try to enlarge a laser beam but I expect a speckle, which I would like to avoid to ensure the homogeneity of the illumination. $\endgroup$ – Ratbert Mar 28 '16 at 11:45
  • 1
    $\begingroup$ Were you to use a laser, you'd still have to focus it onto a pinhole to clean up the spatial frequencies (to remove speckle). Have you considered scanning your (reduced size) collimated beam over the mirror array? But in any case, I can't quite see why you need a collimated beam if all you're doing is illuminating biologics. Can you provide more info about your experiment? $\endgroup$ – Carl Witthoft Mar 28 '16 at 11:56

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.