W. C. Peterson, an amateur telescope maker of Pittsburgh, Pa., made his first aluminum mirror 20 years ago and has not touched

glass since he switched to metal. "In brief," he writes, "my process involves two disks of metal, one for the mirror and one that serves as a tool. One surface of the mirror blank is made concave and one surface of the tool convex so that the pair mate like a shallow ball and socket. The metal can be worked easily with a scraper and file if the experimenter does not have access to a lathe. The roughed-out blanks are ground together with successively finer grades of abrasive until their surfaces mate. Then the concave member of the pair is given a prepolish with pumice and finished like a mirror with rouge on a lap made of hard pitch.

"I have made excellent mirrors of stainless steel but advise the novice to begin with aluminum. Any of the hard bright aluminum alloys work well. They come in the form of bar stock and odd lengths can be procured occasionally from dealers in nonferrous metals' specially cut blanks are available from Henry Prescott, Main Street, Northfield, Mass. I recommend for an introductory exercise a pair of blanks in the form of disks three inches in diameter and 1/2 inch thick. The thickness must be at least a twelfth of the diameter so that the blank will not flex during the grinding operation and distort the desired curvature, but certainly it need not be thicker than an eighth of the diameter.

"I begin by drilling a carefully centered hole about 1/16 inch in diameter and 1/4 inch deep in one side of each blank as a reference center. I also make four disks of hardwood of the same diameter, about S/4 inch thick, and shellac them to seal the wood against moisture. Their use will become apparent.

"The next requirement is a pair of sheet-metal templates to serve as guides for roughing the blanks to the desired curvature. The depth of the curve determines the ratio of the diameter of the mirror to its focal length, the f number. In my experience-and most amateur telescope makers will agree-the relative aperture should not be more than f/8 or less than f/10, with f/9 as a fine compromise. In the case of a three-inch mirror, a focal length of 27 inches would represent a good choice. It is not always possible for the beginner to grind a curve that hits the specified focal length on the nose, but by aiming for 27 inches he can usually achieve a curve that ranges between 24 and 30 inches and is therefore within the accepted limits. The radius of the curvature is equal to twice the focal length. The radius of an acceptable three-inch mirror should therefore fall somewhere between 48 and 60 inches, with 54 inches the best length. I improvised a compass with which to scribe this radius: a six-foot stick with a screw at one end and an ice pick at the other. With the end of the stick screwed to the floor, the ice pick is inserted through a hole in the other end 54 inches from the screw. A three-inch-square sheet of zinc or hard brass is tacked lightly to the floor and the scriber is guided across the middle of the sheet to cut a deep groove completely across the metal. Then the sheet is flexed until it breaks along the arc, and the edges are dressed lightly with a file. The halves serve as the templates, one convex and one concave.

"To make a tool for roughing out the curve of the mirror, grind the end of a flat file to the shape of a thumbnail for use as a scraper [see Figure 2] and wrap the body of the file with electrical tape for a handle. With this tool scrape one side of the blank selected for the mirror until its curvature fits the convex template. This may sound like a job for a lathe, but the work can be done about as easily by hand. Aluminum is soft and only a small amount of metal must be removed-less than half the thickness of a dime. Then use a file to shape the other blank convex to fit the concave template. Do not strive for precision, but try to avoid deep gouges.

"The unscraped side of each blank is now cemented to one of the disks of hardwood. I use common roofing tar as cement-the kind that comes in lumps-and flow it onto the work by heating it with an electric soldering iron. A thin layer of tar is applied to the metal and the wood and the disks are simply pressed together. Seal any crack that develops between the disks by applying the hot iron. Before cementing the mirror to its wood backing I drill a hole through the wood disk large enough for a No. 6 machine screw and attach a nut to the inner face of a metal plate that is then recessed over the hole, as shown in the accompanying drawing [see illustration at left]. This provides a convenient fixture on which to mount the mirror in the telescope.

"Next I make a shallow wooden tray about six inches square and one inch deep, with three cleats screwed to the bottom 120 degrees apart and spaced to make a snug fit with the convex tool blank [see bottom illustration at left]. The tray should be attached rigidly to a firm bench that is about waist-high. Mount the tool in the tray, apply about a quarter of a teaspoon of carborundum grit to the tool and wet it with an equal amount of water. Invert the mirror over the tool and grind by pushing the mirror back and forth. The length of the strokes should be adjusted so that the mirror overhangs the tool about 1/2 to 3/4 inch at the end of each stroke. The center of the mirror should pass directly over the center of the tool. Only two grades of carborundum are used: 220 mesh and 320 mesh. The work will require less than a quarter of a pound of each grade in the case of a three-inch mirror. Normally carborundum is shipped in minimum lots of one pound, but smaller quantities can be obtained from the Edmund Scientific Company of Barrington, N.J.

"The length of the stroke is not critical, but short strokes make the curvature shallow and long ones deepen it. Rotate the mirror slowly while stroking and work it around the tool to distribute the grinding uniformly. Add water to the carborundum from time to time and replace the grit as it turns to mud and becomes ineffective. There is no hard and fast rule for adding water and replacing grit, but you develop a feel for the procedure rather quickly. Grit makes a grinding sound when it is working well, and the mirror slides over the tool with little effort. Spent grit should be wiped from the metal with a rag. (Do not flush it down the drain because it will probably clog the plumbing.)

"When the surfaces of both blanks are fully ground, flush the mirror with clean water and while it is still wet reflect an image of the sun against a wall or a screen of cardboard. Move the mirror toward and away from the screen until the sun's image is sharpest (smallest); the distance between the mirror and the screen should be between 24 and 30 inches. If it is not, check the accuracy of the templates and if necessary make up a new set and start again from the beginning. When you are satisfied that both blanks have been accurately ground with 220-mesh grit, switch to 320 and continue until all evidence of the coarser abrasive disappears.

"The next procedure may sound strange to glass-workers, although it is not new. A polishing lap is prepared of hard pitch-one that would selectively deepen the curve in the center of a mirror and result in what experienced telescope makers refer to as 'the fatal hyperbola' or 'a turned-down edge' though an extremely hard lap is rarely used for glass, it works like magic on aluminum and accounts for the ease with which beginners can make metal mirrors. One of the remaining wood disks now comes into play. Chunk tar of the roofing variety is first melted with the soldering iron and flowed over the wood to a depth of 1/8 inch. It will have little tendency to overflow. Then about a third to half as much lump rosin is melted, flowed into the tar and thoroughly blended with it. (Powdered rosin will not mix with tar. If rosin is available in powdered form, melt it and after batch cools break it into lumps.)

"Paint the surface of the warm pitch with polishing rouge that has been mixed with water to the consistency of heavy cream, place the concave face of the mirror squarely over the painted surface and swirl the mirror until the curve of the pitch conforms with that of the mirror. 'Press' the assembly by allowing it to stand and cool to room temperature. If pockets or bubbles are found in the pitch when the mirror is subsequently removed from the lap, use some of the runoff for patching the holes. Flow in just enough of the tar-rosin mixture to fill the holes. Then paint the patches with rouge and press the entire lap with the mirror as before. To test the pitch for hardness, make a firm cut across lap with a wet knife; the pitch should splinter and make a crackling sound. To soften add tar, to harden add rosin. The edge of the lap is then trimmed with the wet knife.

"The prepared lap is now charged with 320-mesh carborundum (not rouge!) and stroked with the mirror during rough grinding. I do not favor any form of circular stroke, but one must continuously rotate the mirror and more or less work around the tool in all possible orientations to preserve the element of randomness. Now to the crux of the from the center of the mirror to the edge, as judged by eye, begin to use the Foucault knife-edge test. Turn the magnifying glass to the side, align the image of the pinhole so that it almost grazes the 'knife' and, with the eye directly behind the image so that the mirror is seen as a full moon, press the knife into the light rays. When the knife is between the mirror and the image, the apparent shadow cast by its edge will move across the face of the mirror from right to left. When the blade is between the image and the eye, the shadow will cross in the opposite direction, from left to right. Manipulate the blade until it cuts the focal point of the rays. The mirror will then darken when the knife is moved and no shadow will cross the disk.

"If the curvature of the mirror is a perfect sphere, the surface will appear to be flat. If the knife blade is now moved very slightly ahead of the focal point, the surface will appear to be convex, like a ball, and if moved slightly behind, the surface will appear to be concave or cup-shaped. In the case of an f/9 mirror this is the desired test pattern; the beginner can consider himself lucky indeed if it appears early during the polishing procedure. Usually a disk will be seen that has either a pronounced bulge or a depression in the middle. Such figures corrected by altering the lap-removing pitch as required-or by changing length of the polishing stroke, or both. Strokes that result in the mirror over-hanging the lap by more than about half an inch tend to deepen the center, to correct humps or bulges. Those shorter than the normal half-inch overhang tend to bring up the center (or to depress edges). Continue to polish until the whole surface of the mirror darkens uniformly without bulges or depressions when the knife cuts the rays from a pinhole 1/16 inch in diameter at the focal point. This completes the mirror.

"Reflecting telescopes of many types have been developed during the 294 years since Isaac Newton invented the instrument, but I prefer the simple version contrived by John Hadley, an English experimenter, in the early 18th century [see illustration in Figure 5 ]. The optical assembly of my version of this instrument is supported in alignment by a heavy tube of cardboard of the kind on which rugs are rolled. It is strong, easy to cut and thick enough to take wood screws, even at the ends. I always saturate the screw holes with shellac for extra strength.

"The mirror is mounted on a disk of plywood, large enough to cap the end of the tube, by means of a machine screw that engages the nut recessed in the wood block of the mirror. Three equally spaced wood screws fasten the assembly to the lower end of the tube. The holes for the screws are equipped with rubber grommets and the axis of the mirror is aligned with the axis of the tube by adjusting the screws.

"An oblong hole can now be cut in the side of the tube near the top for admitting the eyepiece assembly. This assembly includes a small front-surface mirror for deflecting the rays to a focus just beyond the outer edge of the oblong hole. The position of the center of the hole is determined by subtracting the radius of the tube from the focal length of the completed mirror. Both the small mirror and the lenses for a variety of eyepieces [see bottom illustration] are available from the Edmund Scientific Company. "The base of the Hadley telescope resembles a three-legged milking stool 20 inches high. The top is 10 inches long and about half as wide. A single bolt in the center attaches a trunnion assembly that rotates in azimuth on the bolt. A pair of bolts extending outward from the middle of the tube constitute the elevation axis. They engage slots in the trunnion and attach to the tube through a pair of metal plates screwed to the cardboard. The tube is held in the trunnion slots by a pair of helical springs.

"Slow-motion drive in both azimuth and elevation is provided by a pair of miniature winches made of 1/4-inch rods that fit the plastic knobs of an old radio set. An outrider on the base supports the azimuth winch and its cord, which swing a lever arm fastened to the trunnion assembly. The elevation winch is built into the end of the lever arm and its cord is fastened to the end of the tube. The bottom of the tube carries a small counterweight that keeps the elevation cord under tension. By modern standards the arrangement may seem somewhat primitive. I find that it has the offsetting virtues of low cost and simplicity. During the past three decades I have built many telescopes ranging in design from a replica of Newton's tiny instrument to the highly mechanized types so popular today. None is more convenient or pleasant to use, in my opinion, than Hadley's primitive design."

Bibliography

AMATEUR TELESCOPE MAKING (BOOKS ONE, TWO AND THREE), edited by Albert G. Ingalls. Scientific American, Inc., 1926, 1937, 1953.

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.