Date: Mon Nov 20 18:31:36 2006

Author: Stan Dodds

Subject: Re: Interferometer

The condition your prof is looking for is complete cancellation of two
equal-amplitude beams. That will be tricky to do, but you might be able
to improve your present setup.

As Dan Beeker points out, you may be detecting "background" light from
the room, from scattering off the optics or almost anything else.
Blocking the input beam will tell you the contributions from everything
but internal scatter. You want to minimize that by removing all other
sources the detector might see. Unfortunately you probably already did
this. Then put the beam into the apparatus and see if you can detect
any illumination on the optical surfaces of the beam splitter or
compensator plate. If so, they are scattering and you could try
replacements or cleaning ( but not the beam splitter, unless you know
how to safely clean the coating).

When the optics and background are clean, block each beam separately,
perhaps just before each mirror, and see that the intensity reaching
the detector is the same, since that's the condition for cancellation.
If not, you can try removing the compensator plate, which shouldn't be
needed for a long-coherence source anyway. Depending on which leg is
weaker, that may make the situation better or worse. You could also try
to reduce the intensity in the stronger beam, perhaps with a filter or
even by reflection from a tilted glass plate, until the cancellation is

Assuming you can get any of this to work, encourage the prof to explain
what had to be done to create the "complete" cancellation. It's useful
for students to realize that it takes specific conditions, and makes
for a pedagogically stronger demo.

Good luck.


On Nov 20, 2006, at 4:41 PM, Douglas Johnson wrote:

> Has anyone on Tap-L seen this before? I haven't. ...Doug J.
>> Cliff:
>> is he a theorist?
>> Roger
>> Cliff Bettis wrote:
>>> Hi,
>>> A faculty member wants me to set up a Michelson interferometer.
>>> That's no problem but he wants it so no light is visible at minimum
>>> and then when you block either beam you see light. So far, as close
>>> as I have come is about a 5 to ratio in intensity between max and
>>> min. I am using a green laser with no divergence added, a
>>> compensator plate , an optical table and am as near to equal paths
>>> as I think I can get. The set up is also very sensitive to breath
>>> and the heat of my hand.
>>> Does anyone have any advice? I am thinking he is asking for a lot
>>> from a classroom demo although I am willing to try.
>>> Cliff
> !DSPAM:456231c4203602815528008!