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

How to put the lens?

  1. Sep 15, 2008 #1

    cks

    User Avatar

    I'm not sure does which surface facing the object matters?

    For example,

    http://micro.magnet.fsu.edu/optics/lightandcolor/images/lensesfigure2.jpg

    for b)

    should I put the bulging surface of the convex mirror facing the object or away from the object?

    I mean are there any rules of whether to put the lens facing or away from the object?

    Thank you.
     
  2. jcsd
  3. Sep 15, 2008 #2

    Andy Resnick

    User Avatar
    Science Advisor
    Education Advisor

    The answer is that it depends on the details of the optical system. Intuitively, when using a plano-convex (or plan-concave) lens, the idea is to make sure both surfaces contribute optical power to the system to minimize spherical aberration. So if the light is collimated, put the curved side first.

    The design of optical systems looks at the combination of *all* the optical surfaces and spacings. This information is used to generate a "merit function" (some characteristics may include spot size, distortion, field of view, chromatic aberration, etc), and the design varied to minimize this function.
     
  4. Sep 15, 2008 #3

    mgb_phys

    User Avatar
    Science Advisor
    Homework Helper

    A simple rule to remember what Andy pointed out - the beam should change on both sides. So the curved face of the lens should face the parallel beam and a converging beam should face a flat face.
    This 'splititng the power' has a large effect on the image quality for simple optical systems.
     
  5. Sep 16, 2008 #4

    cks

    User Avatar

    Thanks for your answers

    I want to ask by putting the lens on the path of the laser beam, it seems the laser beam deviates a bit from the straight line.

    I have successfully made the laser beam to the fiber optics. But then, when I put the lens in front of the fiber optics, the laser no longer going in.

    When I adjust the position of lens (rotation or vertical horizontal adjustment) , the best I can get is only 1-2% without the lens.

    Can we really make the beam still traveling a straight line even if we put a lens on its path?
     
  6. Sep 16, 2008 #5

    Redbelly98

    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    Is the curved face of the lens facing the laser beam, and the flat lens face toward the fiber?

    The laser beam will be converging to a focus after the lens, and then diverging (spreading out) after the focus. So in that sense it will not be going "in a straight line" with the lens in its path.
     
  7. Sep 16, 2008 #6

    Andy Resnick

    User Avatar
    Science Advisor
    Education Advisor

    There are 5 degrees of freedom in order to align a single lens to a defined optical axis: x,y,z and tip/tilt. If your lens is displaced in x and y, even if tip/tilt are corrected, you will lose alignment. Errors in tip and tilt will result in coma, primarily, IIRC.

    There's standard techniques to optical alignment- either looking at the back-reflected spot, having a mark on the opposite wall to maintain the optical axis, having the right optical mounts. To get really tight alignment, there's special tools (autocollimators, etc).

    Coupling a laser to a fiber can be a tricky job- is the fiber single mode, multimode, polarization sensitive,etc? what is the NA of the lens- is it matched to the NA of the fiber? How about vibration isolation? Is the fiber aligned to the optical axis as well?

    When I did fiber coupling, we used essentially a microscope objective, but were able to couple in about 50% at best. I think the manufacturers can get about 80%, by hard mounting everything together.
     
  8. Sep 16, 2008 #7

    cks

    User Avatar

     
  9. Sep 16, 2008 #8

    cks

    User Avatar

    I have already successfully coupled the laser to the fiber optics. But, the power of the laser before entering the fiber optics is about 5mW, but after coming out from the fiber optics, it's about 2mW.

    Prof. asked me to make the beam size larger so that all the powers can be more focused to the tip of the fiber optics. (I don't understand is that how could making the beam size larger can make it more focussed. From what I learnt in optics, all parallel beam will be focussed to the focal point no matter what is the beam size, so a bit confused)


     
  10. Sep 17, 2008 #9

    Andy Resnick

    User Avatar
    Science Advisor
    Education Advisor

    It's not clear if you mean that you coupled 40% of the light into the fiber, or there is a substantial loss of optical power within the fiber.

    The first isn't bad at all- you may be able to squeeze a little more power into the fiber, but if you need substantially more power in the fiber, you should use a brighter laser.

    The second, there could be a lot of power entering the cladding of the fiber- either through poor coupling, poor mode matching, or through loss mechanisms like bending. In order to troubleshoot this, you need to know the specific type of fiber and laser- is the laser single (transverse) mode, for example. Is the fiber single or multimode?
     
  11. Sep 17, 2008 #10

    Redbelly98

    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    No, leave it how it is: curved face toward laser beam, flat face toward focus/fiber.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?