Laser interferometers use light interference to accurately measure distance.  Usually you need a bunch of expensive lab gear, but here is a way you can build one with stuff from your junk drawer.  I started working on it as I getting ready to go to a lecture at Fermilab on gravity waves.  They plan to use interferometers to detect space distortions caused by binary neutron stars (details).  I thought it would be fun to try to make my own.  I was curious, not only to see it work, but to see if I could build one with stuff lying around the house.  The answer is yes.  Of course, the notion of "stuff lying around the house" takes on a different meaning when it's my house we're talking about.  I had, for example, a front-surface mirror and some high-quality glass disks in my junk drawer, but I suspect a regular mirror and ordinary glass would also have worked.

I had tried the double-slit experiment (details) a few days ago, and that worked right away.  I cut two slits close together in a piece of paper with an razor knife and directed a laser pointer through the slits onto another piece of paper.  The interference patterns (breaks in the red line) appeared immediately.  I found I really only need one slit, since the light refracts and reflects off the edges of the paper on both sides of the slit.

An interferometer uses coherent light (such as the beam from a laser pointer) to show how light waves can cancel. The beam is split into two beams (usually using a partially-silvered mirror) and then recombined. Where the light waves are in-phase, they add and the image is brighter. Where the light are out-of-phase, they subtract and the image is darker. The result is usually a series of light and dark lines that show where the light undergoes constructive and destructive interference.

The interferometer was a lot harder to build and it took a couple of tries.  I didn't have any beam splitters, so I used glass disks - UV filters from my camera lenses.  The glass should be very flat and clear - no bottles or drinking glasses.  I suspect you could use microscope slides, window glass, or even pieces of broken glass you find on the garage floor with adequate results.  If you shine light at a farily shallow angle, almost exactly half will reflect and half will penetrate through the glass.  I oriented the laser beam at a shallow angle to get a strong reflection off the glass surface.  I used a lens I took out of a cheap disk camera to spread the beam at the end.  Almost any lens should do, though if the lens is weak (like one from a pair of eyeglasses), you may need to move the target (a piece of paper) back a few feet to get a larger pattern.  The only piece you might need to buy is a laser pointer - usually under $5 on ebay.

Figure 1 (to the right) shows the laser pointer (top), lens (left) and two glass discs.  Figure 2 (below) is a schematic representation of the experiment.

Here is the procedure:

1.  Anchor the laser pointer (I used a pipe clamp) to shine horizontally through the first glass.  I taped the switch so it would stay lit.

2.  Adjust the angle of the first glass so that the light is split into two beams of equal strength.  If the reflected beam is weaker, reduce the angle between the glass and the beam.

3.  Reflect Beam 2 with a mirror so that it exactly intersects Beam 1 at the surface of the second glass.  Use a front-surface mirror if you have one.  A regular mirror will probably create a second reflection, so use whichever reflection is strongest.

4.  Adjust the angle of the second glass so that the two beams combine into a single beam.

Note that part of Beam 2 will go directly through the second glass and combine with part of Beam 1 that reflects off the second glass.  You can see these stray beams shining on one of the aluminum blocks at the top left of Figure 3.  We can ignore these extra beams, shown in the figure as a pink line.

5.  Put a lens in the path of the combined beams so it spreads the beam into a viewable target on a piece of paper.  The image should be at least 1/2" across for easy viewing.

A photograph of the setup is shown in Figure 3 to the right.  In the photo, the beam is split into two by the front glass disk.  The left beam travels through the glass to the back disk.  The right beam is reflected onto the back disk by a mirror.  The beams converge on the back glass disk, travel through a small spreading lens (you can barely see it through the back disk as a dark area between two aluminum blocks and resting on top of two floppy disks) and then spreads into a red pattern on the envelope in the top left corner.  I aligned all the pieces vertically with cardboard shims.

I also took close-ups of the projected beam.  The first image is a photograph of Beam 1 only (I blocked the right beam), the second photo is the projected right beam only (blocking the left), and the third image shows the interference pattern created by both beams together.  The interference patterns appear as thin diagonal black lines .  We see it is working because when I block either beam, the black lines disappear, as shown in the first two close-ups.  Cool, huh?  My roommates were teasing me because I was so excited that it actually worked.

My results are not as good as if I did it the right way (details), but I didn't have to buy anything, so this is an even more interesting result.  My interferometer has a sensitivity of about 1/10,000,000 (one part in 10 million). If I touch the mirror with a feather, the microscopic motion visibly shifts the interference lines.  Unfortunately, gravity waves (if they exist - no one has measured them yet) only warp space by about 1/1,000,000,000,000,000,000,000, so I won't be conducting any Nobel Prize winning experiments with this setup.

If I were to try this again, I would use two mirrors, as shown in the illustration below. In this configuration, each beam would go through one glass and be reflected off one mirror and one glass. This symmetry would make it less important that the reflected and transmitted beams be equal in intensity. It would also make the two path lengths more equal, reducing the risk of uneven phase rotation.

I would also like to try building a regular Michelson Interferometer, but use a small mirror instead of a beam splitter, and shine the laser beam on the mirror's edge. This would allow half of the laser beam to reflect off the mirror and the other half to go straight past the edge. I would probably need a front-surface mirror for this. I suspect a section cut from disk from inside a computer hard drive would make an acceptable mirror. If anyone tries this method, I would like to hear about it.

If you have questions, you can contact me at davidlynnthomson (AT) yahoo dot com (I write my e-mail address like that so the spam spiders don't find it).

Click here to go back to my home page.

Hits since January 25, 2005:
Free Hit Counter