Chapter 13: New Optical Alignment Tools

In Chapter 12, I commented again on classical optical instruments since there was interest in an earlier discussion. This got me thinking about what had changed in optical technology since the period ending about 1950 when there was a rather canonical set of classical optical metrology tools. There has been a huge technological change since then, yet little has been written about coherently integrating the changed pieces of technology. That is what I am going to try to do in this Chapter.

Virtually all optical systems have three major components: a light source, some optics that modify the light from the source, and a detector. I will address the changes in these three areas separately to give some order to the discussion and then summarize the consequences of the changes in designing new optical metrology instruments.

Light sources

In the years before the laser, that is, before 1960, man-made sources were either incandescent or some sort of electrical arc. Incandescent sources were physically large and not very bright but fairly stable in terms of position and intensity. Arc sources were brighter or could concentrate their brightness into a smaller size but were much harder to control in terms of the location of the brightness and its intensity level. Even when the HeNe laser came along and solved the brute intensity issue, workers went to great lengths to create point sources by using a microscope objective and a pinhole to “clean up” the source.

Today, I can create a bright, to-the-point-of-eye safety concerns, essentially perfect spherical wavefront with a single-mode optical fiber patch cord and a battery-operated 30 mW LED source used for testing optical fibers for continuity that sells for $20. (See, for example https://www.amazon.com/AKLTM-Visual-Fault-Locator-Tester) The only practical downside to this source is that it is too bright for some applications, and the intensity is not adjustable. There are commercially available laser diode sources that are perfect for illuminating a single-mode fiber, are fully adjustable from pW to mW, and are available in various wavelengths from NUV to NIR. These low-power, adjustable sources typically cost 50-100x the battery-operated ones, but you have control and stability.

Optical elements

Between the source and detector are optics that modify the light before it reaches the detector. Various optical elements could go here, but I will stick to lenses and mirrors. About the time of the invention of the laser, you could begin to buy lenses out of a catalog, with Edmund Optics leading the way. This made it much easier for someone to try out a particular combination of lenses on an optical bench to prototype a new system without waiting for custom (and expensive) optics to experiment with.

Another development starting about the same time was computer-aided lens design. This meant there was more variety in optical systems that could be assembled without much expense and delay because designing new optical systems was easier. This development led to a commercial market for impressive photographic camera lenses. At the same time, governments realized it was much easier to spy on everyone with sophisticated telescopes in planes and satellites.

This, in turn, led to testing optics interferometrically to make better-quality optics than previously. Once testing was easy, there was a push toward deterministic polishing methods, such as the MRF technique developed by QED Technologies. This meant that the lenses listed in catalogs were sufficiently high quality that diffraction-limited systems could be laid out on a tabletop optical bench. The only obstacle to diffraction-limited performance was the precision of the optics’ alignment.

Detectors

Before about 1970, most images were captured on photographic film, a reliable and inexpensive method of capturing and storing massive amounts of data. However, if your camera loaded with film was in space and you wanted to see the pictures, there was a problem. The other demand came from photographers with expensive lenses who wanted to see the results of their pictures sooner rather than later. This led to the development of electronic cameras, but because they used the newest technology, they were pricey. If you had a spy satellite or a large terrestrial telescope, you could afford an electronic camera, but they were not affordable for the general public before about 2000.

For scientific use, electronic cameras are incredibly useful. An easily affordable megapixel camera with a shutter speed range of microseconds to seconds and additional gain was readily available by 2005. Such a camera coupled with microscope optics was capable of diffraction-limited imaging with a 10 x objective over a 1 mm square field of view with instant image availability for further processing. One could argue that photographic film has a higher data storage density than an electronic camera. However, in conjunction with its associated computer, the electronic camera is so much more flexible as a data interface that it is the obvious choice for data capture.

The whole is greater than the sum of the parts

When you consider an autocollimating or autostigmatic optical instrument that incorporates a light source, optics to project and receive light from the source, and a camera to capture and store the image, you now have much more to work with than they did in the day of the classic instruments. Between the adjustability of the light source and the camera you have a useful range of intensity variation of something on the order of 1010. The data cube in one megapixel, 16-bit image is in the same order.

During initial alignment to get reflected light into the instrument, you can turn the laser to its maximum 1 mW level for a Class 1 laser device and see the light under many ambient lab lighting conditions, so you don’t have to darken the lab to find the light. Once the light is back in your instrument, you don’t have to work your head into an odd angle to view in an eyepiece; you can comfortably observe the image on a monitor. Finding a zero setting or finding coincidence is no longer subjective; you get a repeatable, objective number on a monitor. You don’t have to write down your measurement result; you tap a key to store the result or the whole image on a computer that controls the instrument.

Implication of technological changes on the mechanical design of classic optical instruments.

If you knew nothing about the design considerations of classical optical metrology instruments, think of autocollimators as an example, but had today’s technological resources, would you design an autocollimator to look like a classic example? I argue, no. Would the new instrument have a large aperture? No. You get plenty of light back into a small aperture with a laser source.

Would you use a round, precision-ground barrel as a mechanical datum to aid in setting the crosshair to the center of the aperture? No, you can set zero by pointing the instrument at a small retroreflector. Would you use an incandescent light source that draws 10 or more Watts or a laser diode that draws 10 mW? Would you use the new technology for light sources and cameras but then attach it to a custom controller for readout? No, you would attach your device to a generic computer via a USB cable and put the custom controller and readout into the software on the computer.

I conclude that with the new technology, classical optical instruments can retain their same function and accuracy without requiring the older instruments’ mass, size, external features, and electrical demands. No, the new instruments will not look the same as the old or be massive, yet they will perform the same function as well or better than the old. The hurdle is to set aside the picture of a classic autocollimator and ask where is the modern functional equivalent. Optical metrology instruments using the new technology are now catalog or online store items.

The old and the new with greater functionality using today’s technology