This document contains guidelines and links to other manuals for the construction of a homodyne or heterodyne interferometer displacement measuring system based on a bare HeNe laser tube and a measurement readout based on a microcontroller. There are kits providing various levels of performance from really basic to something similar to a commercial system.
Since the lasers and measurement readouts are also available as separate kits, links will be provided to their detailed manuals. Only information not available elsewhere will be elaborated on in this document.
The heterodyne and homodyne kits consist of 4 parts:
These parts of the system can be constructed and tested separately.
For a general introduction to this technology, see LIPM: An Inexpensive Laser Interferometer-Based Precision Measurement System. This is highly recommended reading prior to construction.
There are several variations on this kit:
The typical parts are shown below:
Links:
The typical parts are shown below:
Links:
The detector in the kits for homodyne is quite primitive consisting of pair of biased photodiodes and with a Quarter WavePlate (QWP) to shift the phase of one of the signals by 90 degrees and the bandwidth is quite limited. But a better one could be constructed relatively easily.
This is similar to the homodyne system above but but with the µMD0 display.
Links:
Most of the tubes have a high divergence beam - around 8 mR. Therefore a collimating lens is included, probably in the bag with the optics. It has a focal lengh of 3-4 inches. Placing it close to the output bracket of the tube will result a beam that remains parallel enough for a short range interferometer (probably 1 or 2 feet). For longer distances, HP/Agilent beam expanders are available - ask. These result in a 3 or 4 mm beam which would remain fairly parallel over several feet.
The typical parts are shown below:
Links:
Set up the laser as it would be for stabilization EXCEPT that the polarization axes of the tube should be oriented at ±45 degrees, NOT 0/90 degrees. More on this below.
The tubes provided in the minimal homodyne kits is well behaved (so it could be used in a stabilized laser as another project!). These have two axes of polarization at right angles to one-another whose orientation is fixed for the life of the tube. The amplitude of the power in each axis varies during mode sweep. A linear polarizer could be installed at 45 degrees to the polarization axes to convert the tube in effect to be very similar to a linearly polarized laser, but at the loss of more than one half the power. A better option is to orient the tube so that the polarization axes are at 45 degrees. The PBS cube in the interferometer will than see one half the power in both polarization axes and no power will be lost. However, interestingly, the Path Length Difference (PLD) for optimum performance will be at one tube cavity length, not at 0. For the ~6 inch tube and Linear Interferometer (LI), that is around 2.75 inches; for the Plane Mirror Interferometer (PMI) it is around 1.375 inches. If a PLD of 0 is used, the behavior is, well, very interesting. ;-)
In case you're curious, the reason there isn't an equivalent "Minimal Heterodyne Kit with Unstabilized Zeeman Laser" is that such a laser would generally not produce a split/beat frequency during part of mode sweep. For homodyne, as long as there is always at least one longitudinal mode present (with no gaps due to mode hops), the interferometer will work properly subject to the constraint that the path length difference is less than a few cm.
These laser produce a narrow beam (less thatn 1 mm in diameter), but because PLD needs to be small, the diameter at the detector should be similar for both the reference and measurement beams. However, if desired, beam expander can be added. Ask.
The typical parts are shown below:
Plug the high voltage "Alden" connectors of the laser head and power supply together securely if not already attached. The power supply will probably have a DC connector so it just needs to be plugged into the included wall adapter.
Orient the laser head so that the polarization axis is at 45 degrees. It will probably be marked but if not, is easily determined with a linear polrizer.
Links: