Michelson Interferometer Build From LEGO(R) Bricks

About: Open source hardware for photonics!

A low-cost, full functional Michelson interferometer with nanometer precision is build on a honey-comb optical breadboard, all constructed from LEGO(R) bricks. The installation demonstrates a creative approach of open hardware for young researchers in the field of optics and photonics.

Attached is the LEGO(R) Digital Designer file, that contains the complete construction of the Michelson Interferometer.

Step 1: Setting Up the Optical Breadboard

The base of any of our optical setups is the optical breadboard.

FUNCTION:

The optical breadbroad is designed to fulfill a variety of features (that you may not recognize from the

beginning..)

  • it offers a multitude of attachment points for your optomechanical components
  • the attachment points are arranged periodically in two dimensions; thus the positioning of optomechanical components becomes rather simple
  • it features a huge mechanical stability (that can not be said from the original LEGO(R) baseplates, themselves, that are highly flexible and definetely not suited for optical setups)
  • it reduces significantly external vibrations and mechanical motions; thus, it is optimal for the setup of interferometers

HOW IT WORKS

The interieur of the optical breadboard is constructed having the honeycom structure in mind. This is common for professional breadboards and optical tables (compare with tables from newport.com or thorlabs.com or tmc.com).

CHALLENGE OF CONSTRUCTION

LEGO(R) bricks are designed to build parallelly or orthogonally on each other; for the breadboard, particularly its honeycomb strucutre, it is inevitably necessary to build with a mutual angle of 60 degree between two layers. At the same time, due to the dimensions of an optical breadboard (0.4 x 0.8 m^2), a multitude of bricks is required, that therefore shouldnt either be too expensive or unavailable from the brick shop.

We have choosen standard LEGO(R) bricks for the construction and found a layer-to-layer-concept that was capable to fulfil the demands: 1st row is parallel to the long side of the baseplate, 2nd row is rotated by +60 degree, 3rd row again is parallel to the long side of the baseplate, 4th row is rotated by -60 degree. The 5th row, again is parallel to the long side of the baseplate - then the top plates can be simply added.

Such concept requires at least 5 layers to end up with a complete honeycomb structure; larger thicknesses can be realized by adding at least two more layers (and manifolds of it).

We recommend sizes of the breadboard with a side ratiio of 1:2. Also, it is quite simple to start with (at least) two LEGO(R) baseplates, as it is demonstrated here in the video.

Step 2: Setting Up the Optomechanical Components

This video shows the construction of any single optomechanical component of the Michelson interferometer.

The heart of the setup is the mirror mount, that obeys two.axis adjustment with high precision and mechanical stability.

Step 3: Adjustment of the Michelson Interferometer

Share

    Recommendations

    • Sweet Treats Challenge

      Sweet Treats Challenge
    • Build a Tool Contest

      Build a Tool Contest
    • Faux-Real Contest

      Faux-Real Contest

    13 Discussions

    0
    None
    jbumstead

    Question 4 months ago

    This is fantastic work! I have been trying to figure out how to make cheap kinematic mirror mounts, and so I was very excited to find your brilliant lego design!

    I have a few questions. Why are the lenses set up in this way: one between the beamsplitter and one of the mirrors, and one between the screen and beamsplitter? What are the focal lengths of the lenses? Would you get the same result if you placed a concave lens before the beamsplitter?

    Thanks!

    9 answers
    0
    None
    myphotonicsjbumstead

    Answer 4 months ago

    Hey, thanks for your compliments!

    The lenses have two functions:

    1) Increase the interference pattern.

    2) Create rings instead of lines in the pattern.

    We take two lenses in this setup to seperate these functions. So you can decide on your own if you want to have lines or rings in the interference pattern. The focal lengths are f2=99,6mm in the partial beam and f1=26,5mm for the lense before the screen.

    But you also can take one lens before the beamsplitter with a focal length f=60-100mm, depending on which lengths your partial beams have. In my estimation, you can also use a concave lens but you can get problems with the size of the optics (beamsplitter, mirrors) if you have a lens with a short focal length. Otherwise with a high focal length you need a high distance to the screen for a wide pattern.

    In both setups (one-lens, two-lens) you can manipulate the number of rings and the size of the middle ring in your interference pattern. In the one-lens setup it's depending on the difference of the lengths from the two partial beams. In the two-lens setup it's depending on the distance to the mirror.

    Best regards Felix

    0
    None
    jbumsteadmyphotonics

    Reply 4 months ago

    Hi Felix, Thanks for your clear explanation! I can see how the lenses can make the beam more divergent (and larger) at your screen. But can you explain more about how different lenses placed in different positions result in different patterns? I think (but I am not sure yet) that two Gaussian beams displaced axially will result in a target diffraction pattern, and that two Gaussian beams displaced laterally will result in lines for the diffraction pattern (similar to the double slit experiment). Do you know if my logic is right? If so, how do the lenses switch between these two cases?

    Kind regards, Jon

    0
    None
    myphotonicsjbumstead

    Reply 4 months ago

    Hey Jon,
    to create lines you need just one lens for increasing the pattern. You can see it in the adjustment Video at 1:25min. In this case you overlay two plane waves. A laterally displacement of the beams create the lines. The larger the shift the thinner the lines are. A axially displacement of the mirror has no effect.
    In the two-lens setup you generate one spherical wave and overlay it with a plane wave. The optical path differences become larger moving outwardly, so the rings grow up.
    An axially displacement of the lens manipulate the divergence of the partial beam.
    It is similar in the one-lens setup with rings: Here you overlay two divergent beams with different divergences. Because of this difference you get higher optical path differences moving outwardly in the pattern, so the rings grow up. You can see it in the figure.
    With the distance from mirror to beam splitter you manipulate the divergence of each partial beam. With different distances you get different divergences so that you get rings. The red patterns shows this effect. The higher the distance the bigger the number of rings.

    Best regards
    Felix

    IMG_4276.JPGIMG_4273.JPGIMG_4275.JPGIMG_4274.JPGIMG_4277.JPGIMG_4278.JPGkonzentrische Ringe.png
    0
    None
    jbumsteadmyphotonics

    Reply 4 months ago

    Thanks for taking the time to give me such a thorough explanation. I have learned a lot from your demo and this conversation.
    I have attached a couple drawings to help in understanding the setup with the lenses. My only question is that in the one-lens set up, you said that the beams are like planar waves. I think my drawing is off because after the laser beams go through the lens, they get focused at the same distance (the focal length of the lens) and then start diverging until they hit the screen. It looks like they are more like spherical waves. Am I mixed up again?
    The drawing with the different diverging wavefronts really helped. I tried to simulate this problem. Attached are two gifs with generally results. One is the interference pattern that results when beams are laterally shifted. One is the interference pattern when beams are axially shifted (i.e. wavefronts have different divergence like your picture).

    IMG_9280.JPGIMG_9281.JPGNoneNone
    0
    None
    AdrianaM63jbumstead

    Reply 2 months ago

    Hi Jon,

    how do you get this multicoloured pattern?

    Regards,
    Adriana

    0
    None
    jbumsteadAdrianaM63

    Reply 2 months ago

    The multicolor pattern is a false color scale generated by a simulation I ran. Yellow corresponds to high intensity (bright red in real life when using a red laser) and blue corresponds to low intensity (darkness in real life).

    0
    None
    myphotonicsjbumstead

    Reply 4 months ago

    Hi Jo, you're welcome!
    Yes you are right: From the lens to the screen you overlay two divergent beams with the same distance to the focus. So the spherical wavefronts have the same radius.
    I meant that you overlay two planar waves in the beamsplitter, as you have already sketched in your draft.
    I can't open your (probably animated) gifs. I just saw the same picture two times.
    Much regards
    Felix

    Unbenannt.JPG
    0
    None
    jbumsteadmyphotonics

    Reply 4 months ago

    Excellent. I think I have a grasp on what is going on now thanks to your explanations! I can't upload gifs here, so I can send them another way if you are interested.
    Kind regards,
    Jon

    0
    None
    ChrisMentrek

    2 years ago

    Erstaunlich!

    What is used as the beamsplitter / Strahlentelier? Is it simply a piece of glass?

    Thanks for posting this! I want to build one!

    0
    None
    dmwatkins

    3 years ago

    This. Is. Awesome.

    0
    None
    tomatoskins

    3 years ago on Introduction

    This looks really cool! Thanks for sharing and welcome to the community!