STRIP THE ASTRONOMICAL TELESCOPE of its clock drive, film magazine, spectrograph and related accessories I you put it in a class with a blind man's cane. Like the cane, it informs you that something is out in front. Shorn of appendages, the telescope tells you next to nothing about the size, temperature, density, composition or other physical facts of the bodies which populate space. Not more than 20 celestial objects, other than comets, appear through the eyes as interesting patterns of light and shade. Only one, the moon, displays any richness of surface detail. All other bodies look much as they do to the naked eye. There is a greater profusion of stars, but as a spectacle the night sky remains substantially unchanged.

That is why the experience of building a telescope leaves some amateurs with the feeling of having been cheated.

A few turns at the eyepiece apparently exhaust the novelty of the show, and they turn to other avocations.

Other amateurs, like Walter J. Semerau of Kenmore, N.Y., are not so easily discouraged. They pursue their hobby until they arrive at the boundless realm of astrophysics. Here they may observe the explosion of a star, the slow rotation of a galaxy, the flaming prominences of the sun and many other events in the drama of the heavens.

Semerau soon decided, however, that he tad to have a larger telescope equipped with devices to gather more information than his eye could detect. Accordingly he went to work on a 12 1/2 inch Newtonian reflector, complete with film magazine and four-inch astrographic camera. Both were assembled on a heavy mounting with an electric drive, calibrated setting-circles and slow-motion adjustments. He could now not only probe more deeply into space but also do such things as determine the distance of a nearby star by measuring its change in position as the earth moves around the sun. To put it another way, he had made his "cane" longer and increased his control of it. When the sensitivity of modern photographic emulsions are taken into account, Semerau's new instruments were almost on a par with those in the world's best observatories 50 years ago.

During these 50 years, as Cecilia Payne-Gaposchkin of the Harvard College Observatory has pointed out, we have gained most of our knowledge of the physics of the universe. Most of this knowledge has come through the development of ingenious accessories for the telescope which sort out the complex waves radiated by celestial objects.

Having built the monochromator, Semerau felt he was ready to attempt one of the most demanding jobs in optics: the making of a spectrograph. Directly or indirectly the spectrograph can function as a yardstick, speedometer, tachometer balance, thermometer and chemical laboratory all in one. In addition, it enables the observer to study all kinds of magnetic and electrical effects.

In principle the instrument is relatively simple. Light falls on an optical element which separates its constituent wavelengths or colors in a fan-shaped array; the longest waves occupying one edge of the fan and the shortest the other. The element responsible for separation may be either a prism or a diffraction grating: a surface ruled with many straight and evenly spaced lines. The spectrograph is improved by equipping it with a system of lenses (or a concave mirror) to concentrate the light, and with an aperture in the form of a . thin slit. When the dispersed rays of white light are brought to focus on a screen, such as a piece of white cardboard, the slit appears as a series of multiple images so closely spaced that a continuous ribbon of color is formed which runs the gamut of the rainbow.

The characteristic lines of a substance need not always appear at the same position in the spectrum. When a source of light is moving toward the observer, for example, its waves are shortened- the Doppler effect so frequently mentioned in this issue. In consequence the spectral lines of atoms moving toward the observer are shifted toward the blue of the spectrum. The lines of atoms moving away are shifted toward the red. Velocity can thus be measured by observing the spectral shift.

When an atom is ionized, i.e., electrically charged, it can be influenced by a magnetic field. Its spectral lines may then be split: the phenomenon known as the Zeeman effect. Intense electrical fields similarly leave their mark on the spectrum.

These and other variations in normal spectra provide the astrophysicist with most of his clues to the nature of stars, nebulae, galaxies and the large-scale features of the universe. The amateur can hardly hope to compete with these observations, particularly those of faint objects. However, with well-built equipment he can come to grips with a rich variety of effects in the nearer and brighter ones.

"If you are willing to settle for the sun," writes Semerau, "you shuck off a lot of labor. A three-inch objective lens, or mirror of similar size, will give you all the light you need. The rest is easy. Many amateurs have stayed away from spectroscopes because most conventional designs call for lathes and other facilities beyond reach of the basement workshop, and many are too heavy or unwieldy for backyard use.

"About four years ago I chanced on a design that seemed to fill the bill. My employer, the Linde Air Products Company, a division of the Union Carbide and Carbon Corporation, needed a special spectroscope for industrial research and could not find a commercial instrument that met their specifications. The Bausch & Lomb Optical Company finally located a design that looked promising. As things worked out, it was adopted and is now on the market. My instrument, shown mounted in Figure 2, is a copy of that design.

"The concept was proposed by H. Ebert just before the turn of the century. The instrument is of the high dispersion, stigmatic type and employs a plane diffraction grating. As conceived by Ebert, the design was at least 50 years ahead of its time. In his day plane gratings were ruled on speculum metal, an alloy of 68 per cent copper and 32 per cent tin which is subject to tarnishing. This fact alone made the idea impractical. Ebert also specified a spherical mirror for collimating and imaging the light. Prior to 1900 mirrors were also made of speculum metal. It was possible but not practical to repolish the mirror but neither possible nor practical to refinish the finely ruled grating. Consequently a brilliant idea lay fallow, waiting for someone to develop a method of depositing a thin film of metal onto glass that would reflect light effectively and resist tarnishing. Then John Strong, now director of the Laboratory of Astrophysics and Physical Meteorology at the Johns Hopkins University, perfected a method of depositing a thin film of aluminum on glass.

"The process opened the way for many new developments in the field of optics. One of these is the production of high-precision reflectance gratings ruled on aluminized glass. Prior to being coated the glass is ground and polished to a plane that does not depart from flatness by more than a 100,000th of an inch. The metallic film is then ruled with a series of straight, parallel saw-tooth grooves-as many as 30,000 per inch. The spacing between the rulings is uniform to within a few millionths of an inch; the angle of the saw-tooth walls, -the so-called 'blaze angle,' is held similarly constant. The ruling operation is without question one of the most exact mechanical processes known, and accounts for the high cost and limited production of gratings.

"In consequence few spectrographs were designed around gratings until recently. About five years ago, however, Bausch and Lomb introduced the 'certified precision grating.' These are casts taken from an original grating. It is misleading to describe them as replicas, because the term suggests the numerous satisfactory reproductions which have appeared in the past. The Bausch and Lomb casts perform astonishingly well at moderate temperatures and will not tarnish in a normal laboratory atmosphere. The grooves are as straight and evenly spaced as those of the original. The blaze angle can be readily controlled to concentrate the spectral energy into any desired region of the spectrum, making the gratings nearly as efficient for spectroscopic work as the glass prisms more commonly used in commercial instruments Certified precision gratings sell at about a tenth the price of originals; they cost from $100 to $1,800, depending upon the size of the ruled area and the density of the rulings. Replicas of lesser quality, but entirely adequate for amateur use, can be purchased from laboratory supply houses for approximately $5 to $25.

"The remaining parts of the Ebert spectrograph-mirror, cell, tube, slit and film holder-should cost no more than an eight-inch Newtonian reflector. Depending on where you buy the materials, the entire rig should come to less than $100. By begging materials from all my friends, and keeping an eye on the Linde scrap pile, mine cost far less.

"There is nothing sacred about the design of the main tube and related mechanical parts. You can make the tube of plywood or go in for fancy aluminum castings, depending upon your pleasure and your fiscal policy. If the instrument is to be mounted alongside the telescope, however, weight becomes an important factor. The prime requirement is sufficient rigidity and strength to hold the optical elements in precise alignment. If the spectrograph is to be used for laboratory work such as the analysis of minerals, sheet steel may be used to good advantage. For astronomical work you are faced with the problem of balancing rigidity and lightness. Duralumin is a good compromise in many respects. Iron has long been a favored material for the structural parts of laboratory spectrographs because its coefficient of expansion closely approaches that of glass. When mirrors are made of Pyrex, an especially tough cast iron known as meehanite has been used to counteract the effects of temperature variation.

"The optical elements of my instrument are supported by a tube with a length of 45 inches and an inside diameter of 8 1/4 inches [see drawing in Figure 3]. The walls of the tube are a sixteenth of an inch thick. The eight-inch spherical mirror has a focal length of 45 3/8 inches. The grating is two inches square; it is ruled with 15,000 lines per inch. The long face of the saw-tooth groove is slanted about 20 degrees to the plane of the grating. The width of each groove is 5,000 Angstrom units, or about 20 millionths of an inch. Such a grating will strongly reflect waves with a length of 10,000 A., which are in the infrared region. The grating is said to be 'blazed' for 10,000 A. A grating of this blazing will also reflect waves of 5,000 A., though less strongly. These waves give rise to 'second-order' spectra which lie in the center of the visible region: the green. In addition, some third-order spectra occur; their wavelength is about 3,300 A. Waves of this length lie in the ultraviolet region.

The angle at which light is reflected from the grating depends upon the length of its waves. The long waves are bent more than the short ones; hence the long and short waves are dispersed. A grating blazed for 10,000 A. will disperse a 14.5-A. segment of the first-order spectrum over a millimeter. My instrument thus spreads a 2,200-A. band of the spectrum on a six-inch strip of film.

"The film holder of my spectroscope is designed for rolls of 35-millimeter film. Light is admitted to the holder through a rectangular port six inches long and four tenths of an inch wide. By moving the holder across the port, it is possible to make three narrow exposures on one strip. This is a convenience in arriving at the proper exposure. The exposure time is estimated on the basis of past experience for one portion of the film; the interval is then bracketed by doubling the exposure for the second portion and halving it for the third.

"The most difficult part of the spectrograph to make is the yoke which supports the grating. Much depends on how well this part functions. It must permit the grating to be rotated through 45 degrees to each side, and provide adjustments for aligning the grating with respect to the mirror. The ruled surface must be located precisely on the center line of the yoke axis, preferably with provision for tilting within the yoke so that the rulings can be made to parallel the axis. In my arrangement this adjustment is provided by two screws which act against opposing springs, as shown in the drawing in Figure 4. The pressure necessary to keep the grating in the parallel position is provided by four springs located behind it. Two leaf springs, one above the other, hold the grating in place. The assembly is supported by an end plate from which a shaft extends. The shaft turns in a pair of tapered roller-bearings which, together with their housing, were formerly part of an automobile water-pump. A flange at the outer end of the housing serves as the fixture for attaching the yoke assembly to the main tube. It is fastened in place by two sets of three screws each, the members of each set spaced over 120 degrees around the flange. One set passes through oversized holes in the flange and engages threads in the tube. These act as pull-downs. The other set engages threads in the flange and presses against the tube, providing push-up. Adjusting the two sets makes it possible to align the yoke axis with respect to the tube.

"The shaft of the yoke is driven by a single thread, 36-tooth worm gear that carries a dial graduated in one-degree steps. The worm engaging the gear also bears a dial, graduated in 100 parts, each representing a tenth of a degree. The arrangement is satisfactory for positioning spectra on the ground glass or film but is inadequate for determining wavelengths.

"All plane gratings should be illuminated with parallel rays. Hence the entrance slit and photographic plate must both lie in the focal plane of the mirror. Small departures from this ideal may be compensated by moving the mirror slightly up or down the tube.

"The spectral lines of the Ebert spectrograph are vertical only near the zero order and tilt increasingly as the grating is rotated to bring the higher orders under observation. The tilting may be compensated by rotating the entrance slit in the opposite direction while viewing the lines on a ground glass or through the eyepiece. The effect is aggravated in instruments of short focal length.

"The cell supporting the mirror, and its essential adjustments, are identical with those of conventional reflecting telescopes. If no cell is provided and the adjustment screws bear directly on the mirror-which invites a chipped back-then no more than three screws, spaced 120 degrees apart, should be used. This is particularly important if the screws are opposed by compression springs; more than three will almost certainly result in a twisted mirror.

"The film magazine is equipped with a 48-pitch rack and pinion, purposely adjusted to a tight mesh so each tooth can be felt as it comes into engagement. It is this arrangement that makes it possible to move the film along the exposure port and make three exposures on each strip of film. Lateral spacing during the racking operation is determined by counting the meshes. Although the magazine accommodates standard cassettes for 35-mm. film, it is not equipped with a device for counting exposures. I merely count the number of turns of the film spool and record them in a notebook.

"The back of the magazine is provided with a removable cover so that a ground glass may be inserted as desired. It also takes a 35-mm. camera, a convenience when interest is confined to a narrow region of the spectrum such as the H and K lines of calcium or the alpha line of hydrogen. The back may be changed over to an eyepiece fixture which may be slid along the full six inches of spectrum. This arrangement provides for a visual check prior to making an exposure; it is especially helpful to the beginner.

"Care must be taken in illuminating the slit. If the spectrograph has a focal ratio of f/20 (the focal length of mirror divided by the effective diameter of grating), the cone of incoming rays 11 should also approximate f/20 and the axis of the cone should parallel the axis of the mirror. The slit acts much like the aperture of a pinhole camera. Consequently, if the rays of the illuminating cone converge at a greater angle than the focal ratio of the system, say f/10, they will fill an area in the plane of the grating considerably larger than the area of the rulings. Light thus scattered will result in fogged film and reduced contrast. Misalignment of the incoming rays will have the same effect, though perhaps it is less pronounced. Baffles or diaphragms spaced every three or four inches through the full length of the tube will greatly reduce the effects of stray light, such as that which enters the slit at a skew angle and bounces off the back of the grating onto the film. The diaphragms must be carefully designed, however, or they may vignette the film.

"The components are assembled as shown in the drawing in Figure 4. The initial adjustments and alignment of the optical elements can be made on a workbench. An electric arc using carbons enriched with iron, or a strong spark discharge between iron electrodes, makes a convenient source of light for testing. The emission spectra of iron have been determined with great precision, and the wavelengths of hundreds of lines extending far into the ultraviolet and infrared (from 294 to 26,000 A.) are tabulated in standard reference texts. Beginners may prefer a mercury arc or glow lamp because these sources demand less attention during operation and emit fewer spectral lines which are, in consequence, easier to identify. The tabulations, whether of iron or mercury, are useful for assessing the initial performance of the instrument and invaluable for calibrating comparison spectra during its subsequent use.

"Recently I have been concentrating on the spectroscopic study of sunspots. To make a spectrogram of a sunspot you align the telescope so that the image of the sun falls on the entrance slit. The objective lens of my telescope yields an image considerably larger than the slit. The image is maneuvered, by means of the telescope's slow-motion controls, until a selected sunspot is centered on the slit, a trick easily mastered with a little practice. The spectrum is then examined by means of either the eyepiece or the ground glass. The spot is seen as a narrow streak which extends from one end of the spectrum to the other. The adjustments, including the width of the entrance slit, are then touched up so the lines appear with maximum sharpness.

"Successive spectral orders are brought into view by rotating the grating through higher angles. The upper spectrum at the top shows the first order. The one beneath is made in the second order. Note that although fewer lines per inch appear in the second order, there is no gain in resolution. Shifting the grating for the detection of a higher order is analogous to substituting eyepieces of higher power in a telescope. You get a bigger but proportionately fuzzier picture The film magazine is substituted for the eyepiece and three exposures made in both the first and the second order. In many cases the range of intensity between the faintest and brightest lines exceeds the capacity of the film to register contrast. Three exposures, one estimated for the mid-range intensity and the other two timed respectively at half and twice this value, will usually span the full range.

"Gases in the vicinity of a sunspot often appear to be in a state of violent turbulence. At any instant some atoms are rushing toward the observer and others away. The spectral lines show proportionate displacement from their normal positions in the spectrum-the Doppler effect-and register as a bulge in the central part of the line occupied by the sunspot. This explains the dark streak extending through the center of the spectra reproduced in Figure 1.

"A portion of this same spectrum, photographed in the fourth order and enlarged photographically, appears in Figure 5. It includes the H and K lines of calcium, at wavelengths of 2,933 and 3,960 A. respectively. Observe that a segment in the center of each of these two lines-the segment representing the sunspot-is split. The light streak occupying the area within the split section is referred to as 'emission over absorption' and, in this instance, indicates the presence of incandescent calcium at an altitude of about 100,000 miles above a region of cooler matter in the spot. Had the glowing calcium been lower, its emission would have been absorbed by the. intervening solar atmosphere and it would have photographed as a dark absorption line. My interpretation of this spectrogram is that a solar prominence,. carrying incandescent calcium from the sun's interior, arched up and over the sunspot. We are looking down on top of it. Reconstructing such events from evidence buried in the myriad lines of spectra is an endless challenge and one of the hobby's many fascinations."

Bibliography

EXPERIMENTAL SPECTROSCOPY. Ralph A. Sawyer. Prentice-Hall, Inc., 1944.

Suppliers and Organizations

The Society for Amateur Scientists
5600 Post Road, #114-341
East Greenwich, RI 02818
Phone: 1-401-823-7800

Internet: http://www.sas.org/