THE COLOR photographs of the Great Nebula in Orion on the opposite page were made to demonstrate the performance of a remarkably simple and relatively inexpensive camera of the cryogenic type that was invented three years ago by Bill Williams of 2 Heather Lane, Mahwah, N.J. 07430. At the time Williams was 17 years old. He made the photographs on a muggy night near Key West. To make the top photograph on the opposite page he filled a compartment of the camera with dry ice that lowered the temperature of the film to approximately-75 degrees Celsius. He exposed the cold film to the nebula for 15 minutes. The bottom photograph was made soon thereafter with the same kind of film in the same camera but with an exposure of 20 minutes at a temperature of 25 degrees C. Notwithstanding the longer exposure, the higher temperature prevented the photosensitive emulsion from recording more than a fraction of the information that was caught in less time by the colder film.

The basic technique of photographing dim celestial objects by chilling photosensitive emulsions was developed for astronomical photography by Arthur A. Hoag of the Kitt Peak National Observatory. The method is simple in principle but somewhat complex in application. The emulsion can be cooled to the required temperature simply by pressing the film against the side of a metal box that is tightly packed with pea-size lumps of dry ice. The trouble is that the cold film causes moisture in the air to condense as fog that fills the interior of the camera.

The problem is solved by exhausting air from the interior of the camera with a vacuum pump. Now, however, the camera must be placed inside an airtight vessel reinforced sufficiently to resist the crushing weight of the atmosphere. The external controls of the camera and some parts of the vessel tend to collect frost, but this problem can be solved by recourse to electric heating units. The resulting hot spots give rise to new complications that lead to still other links in a growing chain of compensating gadgetry that suggests the design of a modern automobile.

The cost of such cameras exceeded the budget of most amateur astronomers A few constructed somewhat less complex cameras by substituting liquid nitrogen as the refrigerant. The liquid was evaporated in an insulated boiler by an electric heating unit. Cold, dry gas from the vessel swirled past the film and swept the moisture-laden air from the assembly (see "The Amateur Scientist," SCIENTIFIC AMERICAN, August, 1969). This scheme did not catch on widely because liquid nitrogen is not readily available in most communities. Moreover, it is expensive, difficult to store and hazardous to manipulate in the dark.

That was the state of the art when Williams became interested in astronomy. He explains as follows how he solved the problem of making deep-sky photographs with limited funds.

"Like most beginners who develop an interest in astronomy, I tired rather quickly of looking at the moon and a dozen or so other 'easy' objects. After leafing through magazines that display spectacular deep-sky photographs, I decided to convert my telescope into a camera. The technical literature, however, was discouraging. It disclosed that the best astronomical photographs had been made by freezing the film in specially designed cameras that utilize vacuum chambers, actuating valves, O rings and liquefied gases.

"I could not help wondering why the film had to be so cold. The answer involved a property of photographic emulsions that I had not known about. Everyone who makes snapshots is familiar with the inverse relation between the speed of the camera shutter and the size of the lens opening For a given amount of light to strike the photosensitive emulsion either the shutter speed is fast and the lens opening small or the shutter speed is slow and the lens opening large. The predictable response of the emulsion to this relation is termed the reciprocity rule.

"I did not know that the reciprocity rule fails when the light becomes so dim that exposures must exceed more than two or three minutes. I had always assumed that the speed at which film is rated by the manufacturer is valid under all circumstances. For example, high-speed Ektachrome film is assigned a speed rating of ASA 160. I was astonished to learn that this rating applies only to exposures of about .01 second. The rating does not change substantially even for exposures of a few minutes, but when the shutter must be kept open for two hours, the speed of high-speed Ektachrome falls to ASA 5! Tri-X emulsion, which is rated at ASA 400 for making snapshots, falls below the rating of Plus-X (ASA 125) in the case of time exposures that exceed 30 minutes. I also learned that some slow films, such as Eastman Kodak's spectroscopic emulsions, resist reciprocity failure and that emulsions of this kind are relatively unaffected by refrigeration.

"The latent photographic image that is created when photons dissociate silver halide molecules of the emulsion is not necessarily permanent. Given sufficient time, the ions tend to drift back together again at about the same rate at which they are formed. Thereafter no net gain in exposure results no matter how long light falls on the film.

"Lowering the temperature of the emulsion tends to suppress the recombination of the ions and hence to suppress reciprocity failure. Although cooling lowers the intrinsic sensitivity of the film, it disproportionately suppresses reciprocity failure. The result is a substantial net gain in the speed of refrigerated film for exposures lasting more than about a minute.

"The net gain in speed also improves the resolution of deep-sky photographs by reducing the effects of guiding errors, which arise from the necessity of keeping the telescope pointed exactly at the celestial object as it apparently moves across the sky. The shorter exposure also reduces the interval during which the image jiggles as a consequence of atmospheric instability. It is much more likely that the atmosphere will be steady and that the observer can keep the cross hairs of the telescope centered on a star for 10 minutes rather than for an hour. Incidentally, I have observed that an accurately guided six-inch telescope makes far better photographs than a poorly guided 20-inch instrument.

"Cooling greatly reduces the required exposure intervals in the case of both black-and-white and color emulsions. It also improves the latitude of both kinds of film, meaning that it enables the emulsion to register dim features of the image without overexposing bright areas. In addition cooling restores the color balance that is ordinarily distorted by making long exposures in dim light. Color film consists of several layers of emulsion, each with a unique reciprocity characteristic. Cooling suppresses reciprocity failure disproportionately in the less sensitive layers of emulsion with the result that all colors register faithfully even in dim light. The effect is strikingly apparent in the first of my two photographs of the Orion nebula, which reproduces blues and greens that are not evident in the photograph made at ambient temperature.

"When I set out to make a conventional cold camera, I enlisted the help of a friend, Scott Usher. We tried to simplify the construction in a number of ways without much success until we observed that frost does not form on photographic film that is sandwiched tightly between flat surfaces chilled by subliming dry ice. The literature had given us the incorrect impression that the film had to be in a vacuum. Clearly all we needed was a transparent barrier of low heat conductivity. We got the idea of making a heat insulator in the form of a vacuum cell with opposing windows. We intended to press the film tightly against the external surface of one of the windows by a flat plate chilled to the required-75 degrees C.

"Our vacuum cell proved to be faulty. While we were considering how to correct the design, it occurred to us that transparent heat insulators need not be based on a vacuum. Why not fill the cell with alcohol or a comparable liquid that freezes at a very low temperature?

"It also occurred to us that a transparent solid of low heat conductivity would doubtless work as well as a liquid. For example, a cylindrical plug of glass with flat, polished ends might do the job. The film could be pressed tightly against one end of the cylinder by a metallic container of dry ice. Light would be admitted through the other end of the cylinder.

"The idea seemed promising. How long should the cylinder be so that frost would not form on one end until the other end had been cooled for a period of time greater than the longest exposure of interest? By experimenting we found that a plug about two inches long did not frost over for an hour.

"Next we learned that a clear plastic, such as methyl methacrylate (Plexiglas), can be substituted for glass if the ends of the cylinder are made reasonably flat, parallel and polished. We also wondered if a thick window would degrade the quality of the image. We knew that similarly thick plugs of glass have been used to correct spherical aberration of lenses of large aperture but short focal length. The focal ratio of our telescope was about f/8. For this reason we felt that any possible overcorrection of spherical aberration would be trivial.

"We also investigated the question of how much light might be absorbed by two inches of plastic. Normally about 4 percent is lost by reflection at each surface. Our experiments disclosed that methyl methacrylate transmits approximately 92 percent of white light. Glass transmits slightly less light, is a some- what better heat insulator and is more scratch-resistant than plastic. It is also more difficult to cut and polish.

"In its final form our camera consist of three principal subassemblies. W made the body of a thick-walled tube of opaque plastic, one end of which is externally threaded. The threads engage mating internal threads of a tubular bracket that attaches the camera to the telescope. Images of celestial objects are focused by screwing the body into or out of the mounting bracket. From the . unthreaded end the internal wall of the body tube was bored to larger diameter to accept the plastic cylinder. The enlarged bore terminates in a shoulder somewhat more than halfway through the tube. The plastic cylinder rests against the shoulder [see illustration at right]. The outer surface of the plastic cylinder extends to within about 3/8 inch of the unthreaded end of the body tube.

"A saw kerf that extends approximately three-quarters of the way through the body tube at the level of the outer end of the plastic cylinder admits 35millimeter film to the camera. The film is pressed against the plastic by the flat bottom of a cylindrical box that contains lumps of dry ice Pressure between the lumps of ice and the bottom of the box is maintained by a helical spring acting against a sliding disk that functions as the lid of the box. The box assembly is clamped to the housing by a retaining ring. Pressure between the bottom of the box and the film is maintained by stud bolts, which act through a pair of helical springs against the retaining ring.

"I use three interchangeable plastic cylinders. One is for focusing the telescope, an operation that subjects the plastic to abrasion and invites scratches. The other plugs, which are handled carefully to avoid scratches, are used alternately for making photographs. Frost forms on the cold plug immediately after it is removed from the camera. The plug warms and dries while the next exposure is made with the alternate plug.

"I have worked with both cut and roll film. Initially I made a circular cutter- a small, sharp blade attached to a cylinder of plastic that turned on an axial shaft. The bottom of the shaft carried a rubber foot. Placing the film on a flat support, I set the rubber foot in the center of the desired disk of film, lowered the cylinder on the shaft so that the cutting edge of the knife made contact with the plastic and then rotated the cylinder to cut out the disk.

"Recently I devised a simple technique for using roll film. I push the leader of a new roll of 35-millimeter film into an empty cassette, which acts as a light tight storage compartment. Exposures are made on the strip between the cassettes. The film has to be removed from the camera after each exposure to exchange the plastic plugs and refocus and reset the telescope. During this time I put the cold, brittle film and cassettes in a light-tight box to warm and dry. I mark one cassette with a piece of adhesive tape so that I can identify it by touch in the dark.

"The cold, brittle film between the cassettes must be handled gently until it warms. Frost collects on the emulsion and wets the film when it melts. Part of the moisture can be wiped off without damaging the emulsion by using conventional darkroom techniques. Exposed portions of the film are pushed into the storage cassette after they have warmed and dried. The film can be transferred between the light-tight storage box and the camera on moonless nights without risk of exposure if one stays away from man-made sources of light.

"To make a photograph I line up the telescope, adjust the clock drive and focus an object, such as a bright star, in the plane that will be occupied by the film. The focusing can be done in either of two ways. A conventional eyepiece can be fitted with a short tubelike spacer. The end of the spacer lies in the focal plane of the image. To focus the camera place the end of the spacer on the surface of the plastic plug and screw the body of the camera in or out of the mounting bracket to the point at which the image of the star is smallest.

"The camera can be focused more accurately by the method used most frequently by professional astronomers. Place a dab of India ink, a razor blade or any similar sharp opaque edge on the surface of the plastic plug. Manipulate the telescope to the position at which the image of a bright star is bisected by the opaque edge.

"When the eye is placed close to the opaque edge, the observer sees a portion of a disk of light that represents the objective mirror. Part of the disk appears to be cut off by a dense, straight shadow. If the telescope is now moved so that the straightedge cuts deeper into the image of the star, the shadow will move either in the same direction as the telescope or in the opposite direction, depending on whether the plane of the straightedge lies inside or outside the focal plane of the telescope.

"Screw the camera into or out of the bracket to the point at which the entire disk of the objective mirror appears as a uniform pattern of light and shade that changes form but not position when the straightedge cuts deeper or less deep into the image. The camera is then exactly focused. Telescope makers will recognize this focusing technique as a variant of the familiar Foucault knife-edge test. Exposures are then made by opening a flap shutter [see illustration at right].

"I have patented the camera and granted an exclusive license for its manufacture and sale to Celestron Pacific of Torrance, Calif., but permission is hereby granted to amateurs to make a duplicate camera for their own private use. I have now worked with the camera for three years in California, Vermont, Florida and New Jersey. It is light, portable and easy to use. It enables anyone who owns a telescope that can be kept pointed toward a selected star to make deep-sky photographs of professional quality."

AMATEUR astronomers have had great fun with paraboloidal mirrors of metallized glass during the half-century since Albert G. Ingalls first explained in these columns how to make a simple reflecting telescope. Now two Californians, Alex McEachern of Manhattan Beach and Paul Boon of Garden Grove, call attention to the fun that can be had by making roughly paraboloidal structures of metallized cardboard. A big cardboard paraboloid is easy to make with very small focal ratios: f/.25 and less. It can concentrate an impressive amount of energy from a large area, but its relatively crude curvature is incapable of reflecting the intercepted rays to a sharp focus. Even so, it can be made to pick up ordinary speech sounds from distances of up to 100 yards and the songs of certain birds from more than five times that distance.

A cardboard paraboloid only four or five feet in diameter, covered with aluminum foil, can easily gather enough infrared rays from the sun to broil (and even burn) a hamburger within a couple of minutes. An aluminized paraboloid of this size is ideal for running a hot-air engine that operates on the Stirling cycle. When a big aluminized paraboloid is fitted with dipole antennas, it can pick up strong signals from otherwise weak radio sources such as distant ultrahigh frequency television stations and broadcasts of the earth's pattern of cloud cover and surface temperature from weather satellites.

"In the interest of simplifying the design of the paraboloid," write McEachern and Boon, "we resort to two approximations. First, we assume that a continuous curve can be formed by joining together a series of short straight lines. Second, we assume that a dish-shaped curve can be made by joining the edges of a series of approximately triangular sections.

"To design the parabola decide on the value of a constant that we call an increment and abbreviate with the letter I. The increment should approximate the size of the target you have in mind-three inches, say, for a hamburger, two inches for a microphone or for the heated surface of a hot-air engine. Next decide on the radius R of the parabola-say 27 inches for a fairly large structure. Now decide how far from the bottom of the dish at its middle you want to position the target. We designate this distance d. If you want to broil a hamburger at a point 12 inches from the bottom of the dish, d equals 12. Finally, decide how many sections you want in the paraboloid. The accuracy of the curvature increases with the number of sections, but so does the difficulty of making the paraboloid. A disk of 12 sections is sufficient for concentrating solar energy.

"To calculate the shape of the required sections draw a chart or a table [see illustration at left]. In the example the increment is three inches and the radius 27 inches; the paraboloid will have 12 sections. Number consecutively the column headed 'Row number.' The corresponding numbers for the column headed x are the product of the increment multiplied by the number of the row (3, 6, 9, 12 and so on in the example). Continue to fill in the x column until the product of the increment and the number of the row exceeds the sum of the radius plus the increment (in this example 27 plus 3). The algebraic equation for a parabola states that y is equal to the product of a constant a multiplied by the square of x. It turns out that the constant a is equal to the reciprocal of four times the distance to the bottom of the dish, or 12 inches in this example. The constant a is therefore equal to . Therefore .

"For the first row y is equal to .0208 X 3 X 3 = .187, for the second row .0208 X 6 X 6 = .749 and so on. Entries in the column headed y are equal to the value of y in the succeeding row. For example, the value of y in row No. 1 is .75, which appears as the entry for y in row No. 2. The value for the next column, which is headed z, involves a bit of arithmetic. First multiply 1 by itself. In this example 12 = 3 X 3 = 9. Next subtract the value of y in each row from the value of y in that row. Add the difference to the square of I. For example, in row No. 1, y-y = .75 - .19 = .56. Add this difference to the square of I: 9 + .56 = 9.56. Finally, the equation at the lower right of the table states that the desired value of z is equal to the square root of this sum: (9.56) = 3.09. Similarly, the value for column No. 2 is equal to [9 + (1.64 - 75)] = (9.89) = 3.14, and so on.

"The first entry in the column headed V is equal to the first entry in the preceding column. In this example the value is 3.09. Succeeding entries, however, are equal to the sum of z in the row plus the value of V in the preceding row. For example, the value of V for row No. 2 is equal to the sum of 3.14 (the value of z from row No. 2) plus 3.09 (the value of V from row No. 1 of V), or 6.23. A similar calculation shows that the value of V for row No. 3 is equal to 3.27 + 6.23 = 9.50.

"Values of the last column, headed D, must also be found by doing a bit of arithmetic. For each row subtract from the value in that row of V the value in that row of x, multiply the difference by 3.1416 and divide the product by the number of sectors in the paraboloid. In this example the divisor is 12. The value of D for the first row is equal to (3.09 - 3) X 3.1416/12 = .02. The largest value in the column headed V specifies in inches the length of the sides of an isosceles triangle from which the sectors of the paraboloid will be cut. In this example that length (from row No. 9) is 33.55 inches.

"One final calculation must be made to determine the length of the base of the triangle. The base is equal to twice the length of a side multiplied by the trigonometric sine of an angle that is equal in degrees to 180 divided by the number of sections. In this example the base is equal to 2 X 33.55 X sin 180/12; 180/12 = 15 degrees. Trigonometric tables list the sine of 15 degrees as .259. Therefore the base of the triangle is 67.10 x .259 = 17.37 inches.

"Draw the isosceles triangle full scale, with sides of 33.55 inches and a base of 17.37 inches, on a sheet of cardboard. Beginning at the apex, mark off a series of intervals along both sides of the triangle that are equal in inches to the entries in column V. For instance, beginning at the apex make marks at 3.09 inches, 6.23 inches, 9.50 inches and so on to the base of the triangle. Draw straight lines across the triangle to connect equivalent pairs of intervals.

"Next, beginning at the first interval (3.09 inches), measure inward along the straight line from both of its ends a distance equal in inches to the entry in column 1). For example, measure inward from both ends of the line that is 3.09 inches from the apex a distance of .02 inch. Place a prominent dot on the line at this distance from each side of the triangle [see illustration, left]. Proceed to the next line (at 6.23 inches) and place a similar dot at distances of .06 inch from each side of the triangle. After dots have been placed on all lines interconnect all dots on each side by broken lines. Finally, cut off the sides of the triangle along the broken lines. Use the resulting pattern for cutting 11 identical pieces of cardboard.

"The edges of all 12 triangular pieces of cardboard, when they are joined with adhesive tape or an equivalent material, automatically bend into the desired paraboloid. We leave to the experimenter the fun of designing a suitable framework for supporting the structure."

Bibliography

EXPERIENCES WITH COOI.ED COLOR EMULSIONS Arthur A. Hoag in Sky and Telescope, Vol. 28, No. 6, pages 332-334; December, 1964.

THE THEORY OF THE PHOTOGRAPHIC PROCESS. C. E. Kenneth Mees and T. H. James. Macmillan Company, 1966.

OUTER SPACE PHOTOGRAPHY FOR THE AMATEUR. H. E. Paul. Chilton Book Company, 1967.

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.