LAST JUNE this department contained an account of what Russell Porter did to help win the war and showed a photograph of him lying on a sofa reading The Los Angeles Astonisher while his newly made polishing machine slaved directly beneath him in the cellar. The war over, he was enjoying a little well-earned rest

Now we have asked him to tell the telescope-making fraternity about that machine and here is his reply, illustrated with some of his incomparable drawings. He writes:

Long past retiring age (75 last December) I still find myself fascinated with working optical surfaces. So I obtained three 8" Pyrex blanks and went down cellar where at odd moments I tried to make them flat.

The machine used (Figure 1) embodies the Scotch yoke principle which favors compactness. Its base is a plank one foot wide and two feet long and it has a capacity up to 12" work. The 1/8 horsepower driving motor has a worm directly attached to the end of its armature shaft. To the top of the large worm wheel is attached a flat circular plate, its upper face carefully faced and lubricated. In that plate, extending from center to the edge, is a single radial slot. Locked at any desired position along the slot by means of the nut shown in the drawing, is a shouldered stud. Loose on that stud is a roller which fits the inside of the shouldered slot of the Scotch yoke shown. As the worm wheel rotates, the Scotch yoke unit is forced to reciprocate, constrained as it is by its guide rod. The other rod, the upper one, which carries the pin that pushes and pulls the work, has a rocker which confers freedom of action over any convex or concave surface.

With the adjustments set as shown, a straight-over-center (radial) stroke of about 5" is given (the slot, 3" in length, permits a 6" maximum stroke). A complete stroke is made in about two seconds.

A thumb-nut on the upper rod-the push rod-affords quick change of stroke range. If now it is desired to change to an off-center (tangential) stroke it is necessary only to loosen the little clamping lever near the base of the main casting, rotate the entire unit a little, and re-tighten the clamp. Thus the tool may be made to operate over any part of the glass.

The vertical shaft has a sprocket pinion which, by means of a sprocket chain, rotates the work table at a reasonably slow speed. An idler takes up any slack. There are no creeping belts. I have noticed no tendency to produce periodic errors on the glass being worked. The only casting required is the one embracing the Scotch yoke.

Now for the figuring of the 8" flats. I first tried the interference method described in "A.T.M." page 52. The fringes were normally observed some 10' away by introducing a sheet of window glass at 45 degrees just over the disks. The illumination is from a 2' mercury tube suspended from the ceiling, with tissue paper below it for scattering the light.

The three disks slowly approached flatness, with an error of the order of half a fringe, or a quarter wavelength. Then, as an independent check and after considerable urging by Dr. Anderson, I made an 8" concave mirror of about 10' radius of curvature. With this set-up and a newly made knife-edge stand (Figure 2) I checked the flatness as described in "A.T.M.," page 42. As is called for in the figure on that page three motions were provided, one vertical (A), one transverse (B), and one longitudinal.

At a is an enlarged view of the knife-edge window. With the pinhole image as shown, near the upper left-hand corner of the window, only a slight turn of the screws A and B will give the vertical and horizontal cutoffs. All three movements are controlled by gibs, and the screws (standard 1-1/4" diameter, 20-pitch) act against coil springs to prevent backlash by pre-loading.

The illuminant is a grain-of-wheat lamp bulb covered with a wafer of finely ground glass. There are four pinholes in the little diaphragm shown, of varying sizes, to choose from. The bulb runs on two 1/2-volt flashlight batteries. The trough shown below the window is for an eyepiece, used when studying extra-focal images. It can be swung out of the way when observing the Foucault shadows.

Comparison between the interference and concave mirror methods of testing indicates that there is about the same degree of sensitivity, and the two are mutually consistent throughout. At present all three disks are flat to 1/20 fringe, or 1/10 wavelength, which means departures from flatness of two millionths of an inch. They are not as good as those described by Selby, since they all have turned down edges of about one fringe.

Porter was asked about scale drawings of the knife-edge stand, since readers will inquire. None exist. The drawing is practically self-explanatory. Dimensions need not be identical with those of the original. Pick your own. Porter added, "Yes, made it myself, filed all the gibs."

RUMORS about inexpensive mirrors and lenses of plastics for astronomical telescopes persist. You'll soon be able, so 'tis said, to buy standard mirrors made in simple master molds and precise to 1/20 wavelength, figure perfect, for about five dollars. And, of course, after that you may feel a bit foolish working long hours over glass mirrors made individually.

Many puzzled amateurs have asked about these rumors. Are plastics optics then that good? Or are they likely soon to be that good? Let's look at the evidence.

In 1940, N.R.D.C., in final analysis the Government, awarded the Polaroid Corporation, Cambridge, Mass., a contract to do research on and develop plastics optics. Polaroid investigated 140 organic plastics and chose two as best for optics. To correspond with crown glass they selected polycyclohexyl methacrylate, or "CHM" (Nd =1.50645, v = 56.9) and to correspond with flint glass they picked polystyrene (Nd = 1.59165, v = 31D). These plastics may be poured, in molasses-like consistency, into molds made from glass masters made by conventional precision optical methods-hand work. Bake the filled molds in ovens and you have mirrors, prisms, or lenses.

The optics made from these materials weigh only half as much as glass, are homogeneous, tough, substantially free from color, haze, and strain, and are stable under temperature extremes. They are, however, soft and more easily scratched than glass.

Such optics were useful in the War, as they could be made by inexperienced help, and this released precious optical glass for other purposes. Today the largest use for plastics optics is in the projection systems of television sets. Here the Schmidt camera is used, reversed. This application affords an opportunity to compare plastics optics with glass optics.

A 12" telescope mirror, precision grade (1/10 wavelength), will resolve objects 0.38 second of arc apart, by the Dawes' formula. The corresponding 12" Schmidt with plastics optics as used in television resolves objects 2 minutes of arc apart. Ratio, 300 to 1, though for lenses, which have only one fourth as stiff a tolerance as mirrors, this ratio softens 75 to 1. (But of course, this says nothing against plastics Schmidts for television, where the 2-minute resolving power is sufficient and where increased production costs due to needless refinement would therefore be sheer waste. Simply, the 2-minute resolution approaches the 1-minute resolving power of the human eye.) In astronomical work it is pretty well established that a circle of confusion of 100 microns (1/250") diameter is adequate. The 12" plastics mirror of 12' radius, mentioned above would give a circle of confusion 1.2 mm, or about 1/20", in diameter-hopelessly large for a star image.

Thus is has proved possible to make and use telescopes and binoculars of plastics magnifying about three or four diameters. To supplant glass precision optics, better than this will have to materialize. Far, far better.

In a paper read before the Optical Society of America, Edwin II. Land, Director of Research of the Polaroid Corporation, objectively assessed the usefulness of plastics optics, making no attempt to claim, or imply, more than observable facts justified. He answered the question, can optical plastics be formed into elements sufficiently homogeneous, accurate, and stable to meet the requirements of precision optics? The following is written with that paper as its basis.

A plastics material in liquid form is poured into molds consisting of accurately ground and polished Pyrex reverses assembled with flexible tape so that the halves may approach each other during baking, and baked. The optical elements thus produced prove to be constant in refractive index within ± 0.0015, most of them being considerably better. Over a subsequent period of eight months this index may change four or five parts per million In other experiments disks 1" thick and 6" in diameter, cast between two optical flats, flexured 6 to 20 fringes after removal from the mold. In four CHM lenses of 169 cm focal length arid 12 styrene lenses of 144 cm focal length cast in reverse molds the variations in radius of curvature affected the focal powers over a range of something under 0.4 percent.

In casting prisms from flat molds there was greater trouble. The facets came out of 5 to 10 fringes from flat with angle accuracy ± 15 minutes.

In plastics optics it is as easy to cast aspherical elements such as Schmidt corrector plates good enough for television as it is to cast a sphere. Another advantage is the lightness of the optical elements. Some samples which were sent to this department seemed, by unconscious comparison with glass, almost ready to blow away; they do almost float in water. Plastics optics can be drilled, tapped, turned, shaped, planed or milled with ordinary metal-working tools. And they can be aluminized. They take a precision optical polish and are highly transparent.

In an objective lens, CHM combined with flint glass permits removal of nearly all the secondary spectrum.

Regarding plastics camera lenses: An experimental aerial camera lens of f/2.8 and 7-1/2" focal length was made with four elements of styrene and CHM and one thin element of barium crown glass. It was tested at the Mt. Wilson Observatory beside an excellent f/2.5 all-glass lens of 7.1" focal length. The performance of the two lenses was found to be practically equal for half-field angles smaller than 14 degrees but the glass was better at greater apertures. The plastics lens has negative astigmatism and curvature of field up to 15 degrees from the axis, the astigmatism rapidly increasing thereafter. The distortion curve is of unusual shape, flattening out for large field angles. Spherical aberration and its chromatic variation is smaller for the plastics objective than for glass, and the color curve is better adapted to the use of pan film with a yellow filter.

The coefficients of linear thermal expansion of CHM and styrene are 7 to 8 times that of common glasses, and their change of index with change in temperature is about 45 times as much. Protection from non-uniform temperature changes is necessary in systems containing massive plastics elements because of their low thermal conductivity, which is about one tenth that of glass. In reflecting devices such as Schmidts, temperature compensation may be had by making the mounting of material having the same thermal coefficients as the mirror.

In sum, then, the plastics optics named provide accuracy equal to the demand made by telescopes magnifying about four diameters and by camera objectives up to about 10" focal length. The prisms are far from good enough.

Plastics have a long road to travel before they come within even distant sight of existing precision optics. What of the future? Nobody knows the future.

The Society for Amateur Scientists (SAS) is a nonprofit research and educational organization dedicated to helping people enrich their lives by following their passion to take part in scientific adventures of all kinds.