3D Printed Optics: Michelson Interferometer
Updated: Jun 13, 2019
This is the start of a series that I want to expand out. The theme is looking at optics make out of mostly 3d printed parts.
The hard part about playing with optics at home is actually not necessarily the optical components themselves. The hard part comes in mounting and aligning the optics, since everything needs to be extremely precise.
In a lab is is usually done by having big aluminium holders with fine pitch screws which tune angles very precisely to the needed amounts:
But 3d printers these days are very affordable and print quality should be more than good enough to make something usable.
So I designed my own 1 inch optics holder in fusion 360; it comes in 2 parts which have a fixed ball joint in one corner and two m5 screws to adjust the angles in the x / y plane. Two pairs of springs then hold both parts together in compression:
These could then be 3d printed, I chose pla-plus as its nice to work with and they turned out surprisingly well:
The optical element is mounted in the square front element and held in place with a nylon screw, turning each adjustment screw then changes the angle of the optic in the x / y plane. The mounts can then be attached to an optical bench (piece of plywood) by a plastic rod with a threaded insert and another 3d printed part as the base.
Here's a demo of one of my mounts steering a laser beam around:
The screws have a pitch of 0.8 mm and they are separated from the pivot by around 30 mm so some trigonometry says that each turn of the screw changes the angle of the optic by around 1.5 degrees.
Not quite as precise as the commercial stuff but pretty good. The main problem however is thermal expansion, as the plastic heats up or cools down it will deform, so you could align something and come back in a few hours to find it has drifted off. But still this should be good enough for the home gamer.
So what can these things do? How about build the most sensitive instrument ever built!!
You probably guessed from the title that this is a Michelson interferometer, the same type as that used by LIGO to detect gravitational waves.
A Michelson interferometer works by splitting a beam of light into two paths, both rays travel some distance before being reflected and recombined, if the phases match then there is constructive interference and we see a bright spot, if not then the light is directed back into the source, here's a good visualisation:
Building one from our 3d printed parts is as simple as it looks: the laser is a cheap diode job, it has a Janky beam quality but it works for a demo. The beam splitter is a partially silvered piece of glass, these are fairly cheap but if you are on a budget then a piece of glass will work to some extent. and finally its just a matter of aligning everything:
A Lens on the output shows how bad the beam quality really is. Every time the pattern inverts, that is there is a dark bit where there used to be a light bit, one of the paths has become half a wavelength longer or shorter with respect to the other. The laser has a wavelength of 650 nm, and since the moving one of the mirrors by some distance causes a path difference of twice that distance, each inversion is a mirror movement of about 160 nm.
That is incredible small! you can see that just touching the optical bench causes several inversions, but still such a tiny movement that you would struggle to measure it any other way.
In LIGO this difference is caused by gravity literally squashing time and space in between the two mirrors.
The beam quality here can be improved by placing a spacial filter in the Fourier plane of a lens to cut-off the higher frequency components:
The spacial filter here is just a piece of foil with a very small hole in it, and this would be very difficult to set this up without the super fine adjustment possible with these mounts.
I hope you've enjoyed the start of this series, it has a lot of potential. Already with just some minor additions i'm thinking; a Fourier transform spectrometer? or perhaps a laser spy microphone?