BY THE TIME THEY HAVE made two telescopes most amateur telescope makers develop some secondary interest in optics. While continuing to use their telescopes, they indulge in the side hobbies of making eyepieces, optical flats or any of the wide variety of optical instruments-such as the spectrograph.

One amateur spectrograph maker who uses his spectrograph both for fun and in his work is Robert C. Fairall of Victoria, B. C., a chemist-metallurgist connected with a foundry. Fairall says that he "felt an urge to see whether a grating spectrograph could be made from materials found in the woodshed, scrap box and local stores. At your suggestion," he writes, "I send rough sketches of the spectrograph I made. This instrument gives 'everyday' scientific results. Arc spark, glow tube and solar spectra can be recorded-stellar spectra also if a large enough telescope is available-on motion-picture film in strips about 20 inches long."

From Fairall's shop sketches Roger Hayward has drawn the illustrations on these pages. The lower part of the one below depicts the spectrograph in its kite-shaped, fiber-board case, which is about four feet long. The instrument's essential parts and working principle are shown in the upper part of the same illustration, which shows that this conspicuous enveloping case is not the spectrograph itself but only a shell for the auxiliary purposes of holding the essential parts in fixed relationship and excluding light.

The substance to be analyzed is inserted in the electric arc between the carbons, and the arc is struck. The light passes through the narrow slit and reaches a diffraction grating at the rear, which disperses it into a spectrum along the curved negative film. Qualitative analysis is made by comparing the photographed spectral lines with spectral data $or known substances. The analysis may be made quantitative by estimating the brightness of the lines.

Today spectrograph analysis is replacing the more familiar chemical-analysis because it is quicker and easier, may be done with smaller samples, gives a permanent record and reveals all the metals that are present instead of only the ones that are looked for. Similarly the grating type of spectrograph is crowding back the more conventional prism type, largely because diffraction gratings, which were long scarce because so difficult to make, are becoming more available, while quartz suitable for making prisms is becoming more and more scarce.

In the prism type of spectrograph the light is dispersed to form a spectrum by refraction. In the grating type, dispersion is accomplished by diffraction from a periodic structure consisting of thousands of narrow, closely spaced, equidistant, straight parallel grooves. The working principle of the diffraction grating is explained in most elementary optical works, for example in The Principles of Optics, by A. C. Hardy and F. H. Perrin (McGraw-Hill Book Company, New York). In a coating of aluminum .0001-inch thick, deposited on Pyrex by thermal evaporation, the grooves of the grating are ruled with the fine edge of an artificially shaped diamond, slowly drawn across the surface by a ruling engine, or ruling machine.

Since even small gratings thus painfully made-the engine may work a week on one-cost at least $200, thin casts from these originals in collodion mounted on glass are often substituted. These copies cost roughly 10 per cent as much as original gratings; their usefulness, except on exacting work, is correspondingly greater. They are the "poor man's" grating.

RETURNING to the illustration and following the sequence from the substance under analysis to thc spectrogram, the first component of the assembly is the arc light. In the Fairall spectrograph this consists of a pair of spectrographic carbons-ordinary ones are not pure enough-obtainable from laboratory supply houses or from the National Carbon Company, Cleveland, Ohio. These may be supported in any kind of rough home-made stand. The stand has been omitted from the drawing for the sake of clarity. It may consist of a wooden base with simple adjustments for height, inserted in which is a vertical wooden upright carrying two horizontal arms of tin with ends folded around the carbons. These arms are slightly springy, permitting the upper carbon to be pushed into momentary contact with the lower to start the arc, after which smoked glasses should be worn to protect the eyes from ultraviolet radiation.

Traditionally arc spectroscopy calls for five to 15 amperes at 220 volts direct current, used for 10 seconds or more at a time. If this were a real necessity its effect would be to discourage probably 99 per cent of amateur spectroscopists at the outset. Closer investigation reveals that it is one more case of "what would be nice if"-that is, if cost were not an issue, as in some professional laboratories. Instead, 10 amperes at the household 110 volts alternating current can be made to work satisfactorily. To pre" vent blowing of fuses a resistance may be inserted in series with the arc; for example, one or two electric flatirons. Nichrome-wound heating elements may also be used. Amateurs who can find an ancient direct-current dynamo capable of about one kilowatt at voltages between 50 and 220 will enjoy a few advantages, such as the added fun of experimenting with spectra made by flashed arcs between iron, copper, aluminum, brass, lead, graphite and other electrodes that will not maintain an arc with alternating current.

A little of the substance to be analyzed is ground to a powder and then inserted in a drilled hole or vertical saw-cut in the end of the lower carbon, where it is vaporized in the intense arc; or else a corner of the sample may be inserted directly in the arc. By using a neon-sign transformer with a condenser bridged across its secondary, spark spectra may be obtained. Great caution should be observed because of the danger of serious shock.

Not shown in the drawing is a condensing lens between the arc light and the slit. This focuses the light powerfully on the slit, rendering the spectral lines much more brilliant; its effect is to place the arc in the slit.

The purpose of the vertical sliding diaphragm in front of the slit is to shift the spectrum up or down on the film, permitting comparison spectra of other substances to be recorded beside the one already there. The slit must always be meticulously adjusted parallel to the lines of- the grating. Its opening may be perhaps two or three thousandths of an inch.

SINCE the grating is ruled on a concave spherical surface, it acts as its own collimator and requires no lenses as in prism spectrographs. The grating used by Fairall is a replica of an A. A. Michelson original with 25,000 lines per inch. It has a radius of curvature of 1,060 millimeters, approximately 41.75 inches, and was obtained for $16 from the Central Scientific Company of Chicago.

As will be seen by comparing the upper and lower parts of the drawing, the essential parts of the spectrograph are arranged on a "Rowland circle." In l883 the experimental physicist Henry A. Rowland of Johns Hopkins University announced his discovery that if a slit and concave grating are arranged on a circle forming a radius half the radius of curvature of the grating, the entire spectrum would be in focus exactly on and around that circle. In the Fairall spectrograph the slit, grating and film remain fixed as the Paschen-Runge type of mounting.

Amateur telescope mirror makers will quickly recognize in a part of the drawing a geometrical identity with the Foucault test setup. The knife-edge would be at the point labeled "zeroth order." The curve of the grating is not coincident with the Rowland circle since it has twice that circle's radius.

Though only one spectral order is used in the Fairall spectrograph, gratings throw several pairs of lateral spectra or higher orders on either side of the central image at "zeroth order." These partly overlap, but the first order is the brightest. Why do gratings throw several spectra? In fact, why do they cast spectra? These questions may send one to an optical treatise for the explanation. The answer in brief is interference, retardation and reinforcement, but the demonstration cannot be presented here. A fair share of the fun in constructing optical apparatus is the insight into optics-an exact and therefore satisfying science- that develops as the scientifically curious worker does a little research reading. A hint, but only a hint, of what is meant by "orders" of grating spectra is given when one looks across a field planted in checkerboard pattern. The median row might then be called the zeroth order; second, third, and even fourth diagonal orders may be seen extending at widening angles on either side of it.

"Prospective constructors of this spectrograph," Fairall writes, "will no doubt make their own modifications of it, such as perhaps to use the remaining orders, substituting a semicircular case." The slit could also be refined to afford automatically parallel motion of the jaws, supplanting the present method of adjustment by gentle tapping with the fingernail. An efficient slit is an interesting challenge to the painstaking mechanic, as is explained in Amateur Telescope Making, page 248. The job looks simple, may prove so-and may not. A dark slide could be added to the film carrier. Adjusting facilities, so improved as to be always strain-free, could be provided for the grating. The grating must be so placed vertically that the spectrum will fall in the center of the film when the shutter is opposite the center of the slit. Narrower film may be used. The zeroth order can be isolated by a light-baffle to prevent its light from being scattered over the film and reducing contrast.

A useful and simple addition is a lens or telescope eyepiece of about one-inch focal length attached to the side of a long stick near one end of it. The other end of the stick is pivoted at the center of the Rowland circle. As the stick is swung in an arc across a spectrum, prominent lines may be identified visually by their color.

How the spectrograph is used in chemical analysis remains to be explained. This is done by measuring certain lines in the spectrum of the unknown substance in relation to lines in known substances, or by direct comparison of the spectra themselves. Fairall constructed a simple spectrum-measuring engine in which the film is wrapped around a drum attached to a large worm wheel rotated gradually by a worm that is turned by a hand wheel. The worm wheel is divided to read in 10-Angstrom units, and the device contains a projector with fine cross-line so that parts of the spectrum under close study can be projected on a screen beside a comparison spectrum or a comparison scale. Another method is to photograph the spectrum of a known material, such as iron, on the spectrogram next to the spectrum of the unknown and to see whether known lines in the latter can be correlated with lines in the former, not only by position or wavelength, but by intensity.

But all this is the ultimate in spectroscopy, and is largely mechanical. There is a less routine, more romantic side to it. When the novice is first given a spectrogram he is aware that it is more than Q meaningless row of lines. He knows that each line is characteristic of a molecule or atom that emits it, and that there are precise methods of identifying the corresponding compounds or elements from some 300,000 different spectral lines in extensive tables of standard wavelengths. Yet the worker soon comes to recognize spectra directly by their "faces"-the pattern of the lines. The once meaningless maze of lines resolves itself, just as in a strange city the streets, the buildings and people's faces gradually impress their individuality on a newcomer.

And there is joy in the realization that one has come to understand. What the physician sees in an X-ray photograph that the layman misses, or what he sees in your skin color, vigor, nervous condition and the like even before you have crossed his consulting office to a chair; what the native American sees when watching a baseball or football game that the Tierra del Fuegian misses: the spectroscopist even while still a novice may see in that strip of apparently fortuitous lines, groupings of lines, and spacings-a spectrogram. When you see your old friend Henry approaching, you do not examine him with a caliper and scale and fingerprint equipment, finally saying, "Why, this is Henry!" Similarly you come to know the spectra of many metals at a glance.

The new Practical Spectroscopy of G. R. Harrison, C. F. Lord and F. R Loofbourow (Prentice-Hall, Inc., New York), all of the Massachusetts Institute of Technology, contains a practical chapter on the precise identification of spectral lines and working tables of the main lines encountered in spectroscopy. Other books contain extensive charts and atlases of spectra.

THERE IS a vast literature about spectroscopy, but most of it is too technical for the amateur. Works like the one just named, also the Chemical Spectroscopy of Walter Brode (John Wiley and Sons, New York) and the Experimental Spectroscopy of Ralph Sawyer (Prentice-Hall), the three basic textbook treatises that are today in print, are partly elementary but partly for the physicist, to this extent they are not ideally suited for the amateur, who has long been forced to do the best he could with what was available. A book has just been published, however, that appears to be the long-sought guide for the amateur spectroscopist. This is the Manual of Spectroscopy, by Theodore A. Cutting (Chemical Publishing Company, Brooklyn, N. Y. ). This book starts at the level of the average tyro, with the assumption that he will make his own spectrograph; instructions for making prism and grating types are included. At the start he need know practically nothing about spectroscopy. It steers him with direct statements of elementary facts that the authors of treatises for physicists take for granted. By stages it reaches well into advanced practical spectroscopy, dealing with ore, mineral, alloy and inorganic chemical analyses. It includes tabular data on characteristic lines of the elements and an extensive table-chart showing the wavelength spacing of the lines. Such a book, written as it is by a spectroscopist and not by an assembler of potboilers, has been overdue for at least 25 years. It is a book that the amateur can digest without having to piece together fragments that do not fit and without straining at inference to fill the gaps.

ONCE MORE this department has been asked by one of its readers to compile and publish a complete list of possible sources of scratches of the kind that appear mysteriously on optical surfaces while they are being polished. Perennially SOS appeals arrive from "glass pushers" who have reached their wits' end in striving to trace the source of these scratches, and who imagine that diagnosis by mail can show them the cause of their trouble.

No possible list could include all the ordinary and extraordinary sources of scratches, many of which are bizarre. In one instance a dozen beginners, working together in the cellar shop of an advanced amateur and storing their materials between times on a large table, were perplexed for weeks by scratches. They had overlooked the household cat which, put in the cellar at bedtime, prowled the night through over the gritty floor and then over their work.

If half of the scratches may be traced to specific sources, another half will disappear with experience in ways that may never be noticed. The worker, though not too conscious of his new ways, has nevertheless come of age optically.

Suppliers and Organizations

The Society for Amateur Scientists
5600 Post Road, #114-341
East Greenwich, RI 02818
Phone: 1-877-527-0382 voice/fax

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

At Surplus Shed, you'll find optical components such as lenses, prisms, mirrors, beamsplitters, achromats, optical flats, lens and mirror blanks, and unique optical pieces. In addition, there are borescopes, boresights, microscopes, telescopes, aerial cameras, filters, electronic test equipment, and other optical and electronic stuff. All available at a fraction of the original cost.

SURPLUS SHED
407 U.S. Route 222
Blandon, PA 19510 USA
Phone/fax : 610-926-9226
Phone/fax toll free: 877-7SURPLUS (877-778-7758)
E-Mail: surplushed@aol.com
Web Site: http://www.SurplusShed.com