Skip to main content

How to align an interferometer

Task-oriented. Assumes the interferometer is already built — see Your first interference pattern if not.

Alignment is the step that defeats most first-timers. The fringes are real and waiting; they only appear once the two beams overlap almost perfectly. This guide is a tight recipe for getting there, plus fixes for the usual ways it goes wrong.

The core procedure

  1. Remove the lens. Align with bare beams first — it's much easier to judge two sharp dots than two fuzzy discs.
  2. Turn on the laser. Find the two dots on the screen (one per mirror arm).
  3. Coarse overlap. Using the adjustable (kinematic) mirror's two screws, walk one dot onto the other. Adjust one screw at a time so you always know which direction you're moving.
  4. Fine overlap. Get the dots sitting exactly on top of each other. The closer they are, the better the fringes.
  5. Reinsert the lens. Each dot spreads into a disc. Where they overlap, fringes appear.
  6. Tune the fringes. Tiny screw adjustments turn straight fringes into centred concentric rings. Aim to bring the centre of the rings to the middle of the screen.
🖼️ Image placeholder — align-dots-to-rings.jpg

Show: A 3-panel progression: two separate dots → dots overlapping → ring pattern centred. The visual "before/after" of alignment.

Adjust one screw at a time

The single biggest beginner mistake is turning both screws at once and getting lost. The two screws move the dot along two independent directions (think "left–right" and "up–down"). Move one until that axis is right, then the other. Make small turns — a few degrees of screw moves the dot a surprisingly long way.

Fixes for common alignment problems

I can only ever see one dot

One beam is missing the screen or the beam splitter. Check that both mirrors are actually catching their beam and reflecting it back. Re-seat the mirror cube that seems dark.

The dots overlap but no fringes appear

  • Re-check the lens is in and roughly centred on the beam. No lens, no visible fringes.
  • The overlap may look good but isn't quite there. Nudge the screws in tiny increments — fringes often appear suddenly at the last moment.
  • The path lengths may differ by more than the laser's coherence length. Try to make the two arms roughly equal in length.

Fringes appear but won't sit still / they swim constantly

This is vibration or air currents, not misalignment.

  • Put the setup on a heavier, more stable surface.
  • Stop touching the table; wait a few seconds after each adjustment.
  • Shield it from draughts and breath. Don't lean on the table while looking.

Fringes are there but very faint / low contrast

  • Reduce stray room light — dim the lights or shade the screen.
  • Check the two beams are similar brightness; a badly off-centre beam splitter sends very unequal light into the two arms.
Stability beats everything

A perfectly aligned interferometer on a wobbly table is useless; a slightly imperfect one on a solid bench is a joy. Sort out mechanical stability first, then chase the fringes.

Setting up a camera instead of the screen

To record fringes (e.g. for fringe-counting measurements):

  1. Get a good pattern on the screen first.
  2. Replace the screen with the camera cube; fix it with base plates so nothing shifts.
  3. Open the camera software and lower the exposure until the fringes are crisp rather than blown-out white.
  4. Re-centre with tiny mirror-screw tweaks if the pattern drifted when you swapped.
🖼️ Image placeholder — align-camera-fringes.png

Show: The interference pattern as seen in the camera software, well-exposed and centred.