1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Debug help: optical laser interference setup fails to work

  1. Mar 29, 2015 #1
    I’m looking to demonstrate destructive optical interference using a laser pointer. I’m splitting the beam with a prism and projecting the beams onto one point on the wall. Somehow I’m failing to get any effect, however.

    Setup:
    1. I’m using a static laser pointer, a prism and a handheld mirror. The laser is at 0cm, the prism is at 7cm, the mirror is at 3cm (between the laser and prism out of the way of the original beam), the wall is at 15cm.

    2. I’m pointing the laser dot through one side and using the internal reflection to project the dot through another prism face onto a wall.

    3. I’m using the part of the light that reflects back from the prism and reflecting it into the dot on the wall using a mirror. The reflected beam is weaker than the one from the internal reflection, but it has good focus and decent intensity.

    4. I’m observing the projected laser pattern when the two dots meet. I don’t have an expectation to see an unchanging cancellation or interference pattern, but I do expect to observe changes, fringe patterns, flickering or vibration in the projected dot when the two beam projections meet.

    Reasons I don’t expect to see an unchanging pattern are:
    A) I’m using a handheld mirror and my hands aren’t capable of statically holding a mirror to tens of nanometers :) That said, I should still be able to observe some kind of changes - flickering or optical artifacts.

    B) One beam is slightly weaker (the one that reflects back). It may be tens of percent or many tens of percent less intense, but not by order(s) of magnitudes. Witnessing some sort of artifacts seems reasonable.

    C) The beams are at a slight angle - one edge of the projection has had to travel longer. However seeing artifacts again should be a reasonable expectation.

    Equipment:
    i. Red laser pointer

    The laser was purchased from http://www.amazon.co.uk/gp/product/B0051223CU?psc=1&redirect=true&ref_=oh_aui_detailpage_o01_s00. The pointer emits a red dot that (viewed very closely) seems to have a sparkling effect with lots of tiny-tiny dots. Its focus point is about 4mm, the focus at three meters is the same (or no more than 1-2mm larger), the focus at 25m is small as well - somewhere between the original 4mm and definitely not larger than a few cm. The label warns of eye exposure, declares compliance with 21 CFR and says the power is less than 1mW. The website says the wavelength is 650nm. It seems to be a genuine laser pointer. When doing the hair experiment and moving it in into the beam, I do see some sort of patterns indicative of interference.

    ii. Glass prism

    The glass prism is from http://www.amazon.co.uk/gp/product/B00BNSU7FQ?psc=1&redirect=true&ref_=oh_aui_detailpage_o01_s00. It reflects from the inner surfaces and reflects back to the original source.

    iii. The mirror is http://www.amazon.co.uk/gp/product/B003EB57MI?psc=1&redirect=true&ref_=oh_aui_detailpage_o01_s00 (acrylic safety mirror). It’s not of the highest optical quality, but it does reflect well (the loss is less than from splitting the beam using the prism).

    Attachments:
    Images:
    • labels on laser
    • view into the pointy end of the laser, the diode lens being visible
    • components
    • focus point size at 3m
    Video:
    • experiment with me combining the beams and nothing happening
     

    Attached Files:

    Last edited by a moderator: May 7, 2017
  2. jcsd
  3. Mar 29, 2015 #2
    I was thinking maybe it's a contrast issue ( and ) and tried to make sure my pathlengths are similar for the two beams. I wasn't able to get any improvements.

    I may need to play around with the contrast from pathlength issues some more, but that doesn't seem to be a promising avenue.

    Also, I'm taking it for granted that laser light is coherent out of these diodes. I'm wondering whether I should somehow verify that assumption as well and whether I can check it without special equipment somehow. Also thinking about whether my experiment about reflecting the dots together is valid or whether the beams need to combine with each other at right angles (I don't see why they would if I'm looking to just get a flickering effect as a proof of concept for now - that seems to be a prerequisite to get nice interference shapes.

    I may need to rethink handholding the mirror and maybe get a better optical splitter unless I get any other new ideas.
     
  4. Mar 29, 2015 #3

    blue_leaf77

    User Avatar
    Science Advisor
    Homework Helper

    So you expected to see fringes, or what? The problem might be because the fringe separation is either to small or to big. What about placing a diverging lens several cm just before the wall? And making the mirror stand still is also recommended.
     
  5. Mar 30, 2015 #4
    I did expect to see a change in the laser dot. Mostly I expected fringes that flicker due to small vibrations in my hand (fringes along the lines of (along the lines of https://upload.wikimedia.org/wikipe...son_Interferometer_Red_Laser_Interference.jpg or https://web2.ph.utexas.edu/~phy-demo/demo-gifs/6d40_10a.jpg).

    When waiting tens of seconds, there moments be times where hand microvibrations settle enough to change less than 0.03mm per second, which would correspond to 50x the wavelength and therefore <50Hz vibrations. That should be perceptible to the eye in at least some form.

    I can use a lens to magnify the picture, but given that I'm seeing other artifacts in the dot pretty well, I doubt I'll see more when magnifying the lens. Since I don't merge the beams until they hit the projection screen, it'll also be difficult to merge the beams at the lens.
     
  6. Mar 30, 2015 #5
    Just off the top of my head, I think the bandwidth of the laser pointer is too great.
    destructive interference means the waves are 180 degrees out of phase,
    I don't think the laser diode photons are in phase enough for you to see much effect.
    Try setting up some feedback, like pointing the beam straight into a mirror.
     
  7. Mar 30, 2015 #6
    Feedback? What should pointing it into the mirror do? Apologies, can you elaborate?
     
  8. Mar 30, 2015 #7
    The laser diode has a very short cavity length, and supports many optical modes.
    Interference would be from one mode interfering with itself.
    Pointing it into a mirror could isolate off whatever mode stands in the new,cavity length.
     
  9. Mar 30, 2015 #8

    Andy Resnick

    User Avatar
    Science Advisor
    Education Advisor
    2016 Award

    That will never work. You need to first spatially filter the beam to increase the transverse coherence length.
     
  10. Mar 31, 2015 #9

    blue_leaf77

    User Avatar
    Science Advisor
    Homework Helper

    Actually the shorter the cavity length, the less number of modes that can be accomodated by the semiconductor's gain profile.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Debug help: optical laser interference setup fails to work
  1. Laser interference (Replies: 2)

Loading...