Construction Guidelines for Scanning Fabry-Perot Interferometer Kit 2

Version 2.10 (28-Jul-2018)

Copyright © 1994-2016
Sam Goldwasser
--- All Rights Reserved ---

For contact info, please see the Sci.Electronics.Repair FAQ Email Links Page.

Reproduction of this document in whole or in part is permitted if both of the following conditions are satisfied:
  1. This notice is included in its entirety at the beginning.
  2. There is no charge except to cover the costs of copying.


HENESFPI2 contains the key components to construct a confocal Scanning Fabry-Perot Interferometer (SFPI) suitable for displaying the longitudinal modes of a TEM00 red, orange, or yellow laser (633 through 594 nm). Wavelengths beyond 633 nm and slightly below 594 nm should work as well, but NOT down to 543 or 532 nm. Performance is best with narrow beam lasers like HeNes but the use of an aperture beam reduction optic should allow the modes of fat beam lasers to be displayed as well. The advantage of a confocal SFPI is that alignment with respect to the laser being tested is much less critical than with a planar-planar SFPI and back-reflections can be off-axis so that the laser is less likely to be destabilized. The benefit of HENESPFI2 is that in addition to what's in HENESFPI1, it includes parts from Thorlabs for the major mechanical assembly so that much of the scavanging for scrap sheet metal and other tid-bits is eliminated. The Thorlabs "Cage System" provides a rigid frame allowing for initial adjustment which can then be locked in place. The completed frame and other parts are shown below.

The standard HENESFPI2-1 kit includes:

HENESFPI1-1 parts:

HENESFPI2 additional parts:

Note: If you ordered a kit with mirrors other than the default 43 mm/1.7 GHz FSR, the rod length may differ and the focusing lens will not be included.

What you must provide:

The following are NOT included:

Rear Cage Plate Preparation

The CP02 is used to mount the PZT/rear mirror assembly on its front side and the photodiode and its wires, as well as the PZT wires on its rear side. The best way to do this is to drill and tap three 2-56 holes on the front side for screws to secure the PZT/rear mirror assembly and two 2-56 holes on the back side for the PD/wiring board. Take care with the photodiode as it's real easy to break off the leads. I recommend poking it through the Perf. board and immediately securing it with 5 Minute Epoxy. Then solder fine wires to the leads.

If you don't have a drill-press and are not familiar with using small drill bits and taps, it's probably best to consider some alternative. It's really easy to put holes in the wrong place at strange angles and to break drill bits and taps, which ends up making a mess because at the very least, it often impossible to extract the broken pieces. (Don't ask how I know.)

The PZT/rear mirror assembly and Perf. board can be attached to the cage plate with 5 minute Epoxy, which is soft enough that it can be removed and replaced if the need should arise.

PZT and back mirror

Prepare the aluminum spacer ring and attach the PZT disk to it. (Using the ring isn't essential but will make mounting to the Thorlabs plate slightly easier and provide a path for the PZT leads inside the plate to exit out back.)

Using a narrow file, create a notch wide and deep enough on one surface of the mounting ring for the PZT leads to pass through between the ring and mounting plate.

Use 5 Minute Epoxy to attach the PZT disk to the aluminum spacer ring around its perimeter. The active (coated) side of the PZT disk should be facing up.

The back cavity mirror must be glued to the PZT.

(The following applies to the standard 43 mm RoC/1.7 GHz mirrors. Other types may not have wedge and thus may not require this extra step.)

The 43 mm RoC mirror substrates are ground with significant "wedge" - the front and back are not parallel. Small wedge is often built in to mirrors like this but the amount of wedge here is large enough that correcting for it is desireable, especially if an adjustable mount is not used for the PZT/rear mirror. Therefore, it should be attached to the PZT at a slight angle to compensate. (This can be avoided by attaching it to the back of the PZT but I prefer the other way since being recessed, it's almost impossible to clean if that should ever be required.) Carefully remove the mirror from its protective lens tissue. Locate the thinnest part and put a mark on the side with a pencil. The currect angle can be obtained by putting a tiny piece of standard Avery sticky label under it on the thin side. :) (Very scientific, huh?) A 1x5 mm piece of label will suffice.

Center the mirror on the PZT disk taking care to orient it so the mark you added lines up with the bump. While holding it in position with a clean cotton swab, apply the tiniest amount 5 minute Epoxy in 3 locations around its perimeter. Just enough to secure it. Once the Epoxy sets up, additional adhesive can be added to increase the strength, but there really isn't much stress on this thing and too much adhesive may restrict the motion of the PZT..

Set this assembly aside covered so the mirror doesn't collect dust or fingerprints.

Front mirror

The front cavity mirror is installed in the SM1AD8 adapter. Remove one of the retaining rings. Slip one of the cavity mirrors in, coated face first. Something compliant between the back of the mirror and second ring is required to (1) prevent crunching of the mirror (which is bad form) and (2) to assure that the front face seats square. A rubber O-ring is ideal, but a piece of thin wire insulation (without the wire) can be curled up to fit. Screw the second retaining ring in so it is just snug. Don't overtighten. Be particularly careful not to touch the coated surface and not to ding either surface while tightening the ring. (Tighten from the AR-coated side - damage to it is less of a disaster!)

Remove one retaining ring (if present) from the CP02T cage plate and adjust the other one so it is just flush with the front of the CP02T.

Screw the SM1AD8 adapter in until it is 1 or 2 turns away from the retaining ring.

Focusing lens (if used)

A focusing lens with a focal point roughly in the center of the cavity (included in the standard Deluxe kit) may improve performance under some conditions but is not essential. If this isn't already installed, remove one retaining ring from the CP11, slip in the lens, and install the retaining ring just snug. If you can't find it in the package, it may already be installed. :)


Completed SFPI on Three-Screw Adjustable Base (not included)

Initial Adjustments

The confocal cavity SFPI requires that the mirrors be spaced precisely at their RoC, around 43 mm in this case. So, the resonator must have some means of fine adjustment as noted above. With the Thorlabs cage parts, coarse adjustment is provided by sliding the front mirror along the bars. Fine adjustment is done by rotating the threaded adapter in which the front mirror is mounted. The axes and orientation of the mirrors should should be coincident. Slight tilt with respect to each other isn't critical - it just shifts the center point of the spherical cavity. However, an offset may be more detrimental. Thus care in assembly with respect to where the mirror mounted on the PZT and where the PZT is mounter are important. Once the assembly is complete, it's time to do "first light" with a laser! A single longitudinal mode (single frequency) laser is best for this as it reduces any ambiguity in setting the cavity spacing, but a short normal HeNe (e.g., a JDSU 1508) red alignment laser can be used.

  1. The mirror spacing should be set as close to 43 mm as can be done with physical measurements. (I.e., a machinest's scale and Mark II eyeballs.)

  2. Attach the PZT to your ramp generator. Connect the photodiode to your scope's vertical input with resistor of a few kohms across it. (If a proper photodiode preamp is available, that's even better!)

  3. Trigger the scope externally using a sync signal from the ramp generator, or the ramp if none is available.

  4. Set up the test laser so it is aimed precisely into the center of the input mirror. (The optional lens should probably not be used at this time as it may make things more confusing.)

  5. Drive the PZT with a 20 to 30 V p-p ramp (or triangle) at 50 to 100 Hz.

  6. Observe where the intra-cavity beam is located on each mirror and adjust alignment so it is more or less centered and tight. Then check the position of the photodiode and adjust it if necessary so the trasmitted beam is centered on it. The room lights should probably be out for all this.

  7. With the scope's vertical sensitivity turned up, watch for any signal from the photodiode that is synchronized to the ramp. If the blips go negative, reverse the PD polarity. If your cavity distance and mirror alignment were perfect, the result scanning through two FSRs for a laser with 3 longitudinal modes would look similar to the photo below.

SFPI Display of Melles Griot 05-LHR-151 5 mW HeNe Laser

More likely, the peaks will be smeared out or composed of multiple small blips as in the sequence of graphics below. Or there may be nothing. Adjust the spacing of the mirrors in small increments Slowly and then then let it settle down. With any movement, the display will become quite scrambled, so be patient. If going one way makes it worse, go the other way. :) If the initial cavity spacing was within about 1 mm of being optimal, there should be only one place close by where it resolves into a beautiful display like the one above. ;-)

SFPI Display of SLM Laser as Cavity Length Approaches Optimum Starting from Too Long

(Cavity length error is approximately: +0.5 mm, +0.25 mm, +0.12 mm, +0.06 mm, +0.03 mm, 0 mm)

The entire sequence would represent a length change of a fraction of 1 mm. The amplitude of the single peak (in this SLM example) would actually increase by a larger amount than shown. Some of these diagrams are from the Toptica SFPI 100 manual, I hope they won't mind. :)

If there is no evidence of any response related to the laser, confirm that the PZT is actually beging driven - set the function generator to a few kHz and listen to the PZT! There should be a very audible tone. Check the output of your function generator and connections if there is none. And check that the photodiode responds to light.

Once it's near optimal using the sliding adjustment, tighten the set screws to secure the cage plate. Then fine tune further by rotating the SM1AD8 adapter while pressing it *away* from the locking ring or keeping the locking ring snug against the adapter. Keep in mind that since it's likely the mirror isn't perfectly centered, the alignment will also change slightly as the SM1AD8 adapter is rotated, so also peak alignment as you are doing this. (May require 3 or 4 hands.) Then tighten the locking ring to secure it.

First time users do not appreciate how precise the spacing needs to be. But it's less than 1/10th the width of a human hair - a few microns. Sliding the mount back and forth will not work.

It's also essential to avoid back-reflections into the laser, which will likely destabilize it and create chaos in the display. The alignment should be adjusted such the the reflections from the SFPI (mostly the front mirror) do NOT enter the laser's aperture. With the confocal cavity SFPI, a slight offset will not significantly affect resolution. Ideally, an optical isolator could be used but they are pricey.

Using mirrors identical to the ones in the kit, I've seen a finesse at 633 nm of 500 or more, though this depends on all the stars aligning perfectly. :) With no angular adjustment, the centerlines of the two mirrors should also be as coincident as practical to really peak the finesse. This requires either super-careful installation of the glued mirror, or fine adjustment of the position of the PZT assembly on the cage plate. In addition, down to a point, narrower beam (e.g, 1 mm) generally results in higher finesse. A wide beam can be put through an aperture (though the signal will be reduced.) Expect a finesse of several hundred at 633 nm with reasonable care. Performance at other wavelengths is not as good but it should still be usable to below 594 nm (yellow HeNe) and above 650 nm (might even be better at some longer wavelengths).

For more on SFPIs, see the section: Scanning Fabry-Perot Interferometers of "Sam's Lsaer FAQ".