SINCE 1896, the year in which Henri Becquerel discovered that something emitted from pitchblende darkened a photographic plate, much work in experimental physics has gone into developing instruments for detecting and studying subatomic radiations. Pitchblende's mysterious emanations were soon found to be the spontaneous emission from radium of alpha, beta and gamma rays The first two rays were identified as particles of matter, the last as waves of electromagnetic energy. Among their singular properties was the ability to ionize, or electrically charge and separate into positive and negative ions, the atoms of any material through which they passed. All the instruments developed within a decade or so of Becquerel's discovery made use of these ionic charges to detect the passage of the rays that created them. Some of the instruments immediately attracted the interest of amateurs in electronics.

One of these is the simple electroscope, which nearly every radio "ham" has built at one time or another. It consists of two facing strips of gold leaf suspended in a glass bottle from a wire thrust through the cork. When an electric charge, produced by rubbing a rod of hard rubber or glass with a bit of wool or silk, is transferred to the wire, the force of electrostatic repulsion causes the vanes to fly apart. As the charge leaks off, the vanes slowly collapse. Using this instrument, Pierre and Marie Curie observed that when they brought their samples of pitchblende near it, the rate of discharge increased, and they measured the relative activity of the samples by timing the discharge.

A second instrument, which soon came into extensive use, operated on a different principle, that of fluorescence. This is the ability of many substances to give off light of a characteristic color when exposed to certain radiations. In 1911 Ernest Rutherford, the noted British physicist, made use of this principle to measure gamma rays, using an X-ray fluoroscope screen coated with zinc sulfide powder, one of the early fluorescing materials. When he placed a small sample of radium in front of this glass screen in a darkened room, he could count tiny individual flashes of light on the zinc sulfide coating, each marking the impact of a ray from the sample. Clock manufacturers later exploited the same principle in making luminous dials of a paint composed of radium salt and a powdered phosphor. Examine one of these dials in the dark under a 15-power microscope and you will observe that the light is released in sharp bursts. Each scintillation is caused by the disintegration of one atom of radium.

Many amateurs have built a version of the Rutherford instrument called a "spinthariscope." This is a small box of convenient shape fitted with a zinc sulfide screen at one end, a magnifying lens at the other and a source of alpha particles in between (scraped, perhaps, from the luminous dial of a discarded clock). The radioactive material is enclosed in a matchhead-sized envelope of black paper, supported in line with the center of the lens and screen by a common pin or a short length of wire. The inner walls of the box should be painted dead black to kill reflection. The screen may be made by dusting zinc sulfide- Du Pont's silver activated zinc sulfide, Patterson Type 1101, is highly sensitive -on the sticky side of scotch tape. The sulfide side should face the radium. When the lens is held to the eye, the observer appears to be looking into the depths of a black sky filled with shifting patterns of exploding stars. Each faint flash marks the point where an alpha particle has crashed into the screen. The number of flashes will depend on the activity of the disintegrating radium and its distance from the screen. The best distance must be found by experiment. Readers who wish a sample of radioactive material for experiments may obtain one by sending a stamped, self-addressed envelope to SCIENTIFIC AMERICAN.

With the advent of the vacuum tube, greatly improved methods of measuring radioactivity were developed-instruments that could not only count rays or particles but also sort them according to type Gases under low pressure become momentarily conductive when ionized by alpha, beta and gamma rays. A gas-filled tube fitted with a pair of electrodes will thus pass a pulse of current for each particle or ray passing through it. These pulses can be amplified and used to operate an electromechanical register. This principle led to the development of the mass spectrometer [see page 68], to the Geiger-Muller counter and, indirectly, to the new "crystal" counters now attracting much attention.

Of all subatomic detectors, however, the scintillation type remains the most versatile for research work. Linked up with a modern photomultiplier tube to do the counting, it responds easily to 100 million nuclear events per second, as against the Geiger counter's top rate of 2,000 per second. Moreover, its screen phosphors can be arranged to respond selectively to each kind of radiation, thus causing it to make qualitative distinctions between nuclear events, something a Geiger tube cannot do. With a scintillation counter even an amateur can measure cosmic rays, detect radioactive gases in the atmosphere and test the many naturally occurring radioactive minerals. In addition, he can test the fluorescent properties of gases, liquids and solids, an area in which much work remains to be done. Finally he can develop electronic circuits to meet a broad range of applications.

The photomultiplier tube and related circuitry have, in fact, revived interest in all scintillation counting. Here at last is a device more sensitive than the eye and able to watch and respond to millions of scintillations per second. It is essentially a photo-electric cell with built-in amplifier. Light falling on a sensitized electrode in the tube ejects one or more electrons. They cascade through a series of carefully positioned electrodes, each carrying a progressively higher positive charge. As these electrons knock "secondary" electrons from the intermediate electrodes, the cascade becomes an avalanche and a billion electrons reach the anode for each one ejected from the first electrode. Thereupon an amplified pulse of direct current flows from the output circuit of the tube.

In terms of number of electrons, the output pulse is huge. Yet its current when compared with that drawn by even a small Mazda lamp, is small. It can be measured directly with a cathode ray oscilloscope or other sensitive electrical instrument. But it does not have sufficient energy to drive electromechanical registers of the type used for recording nuclear events. Many pulses persist for only a few microseconds at best; hence they must be further amplified and prolonged by electronic circuits.

Circuits also serve to improve performance from the notoriously "noisy" photomultiplier tube. "Noise" appears in the tube's output as pulses which do not owe their origin to light but to electrons dislodged by heat, thermal agitation or other extraneous events. Noise pulses generally carry less energy than those from valid scintillations, and electronic circuits are designed to take advantage of this difference to sort them out. An amateur who enjoys working with electronic gear will find, even if he is a beginner, that the counter's circuits are easy to design and build. And the variety of things they can do will amaze him.

Assume that a photomultiplier tube has been mounted in a light-tight container with a phosphor screen sandwiched between the photocathode and a "window" of aluminum foil. The tube has been connected with a source of voltage prescribed by its manufacturer. The phosphor is being irradiated by a "hot" source, with the result that a mixture of noise and signal pulses is appearing at the output terminal. What kind of a circuit must be built to sift this electronic wheat from the chaff?

Within the past decade or so electronic engineers have developed special circuits for manipulating electrical pulses-the form of signal that has been exploited so successfully in radar, television and electronic computers. The new circuits can generate pulses, or clicks, at will, in almost any size, shape or frequency desired. The clicks can be added together, subtracted, sorted and reshaped. Most of the devices for doing this are easy to build.

A simple resistance-coupled amplifier usually follows the photomultiplier. The output of the tube is fed into the amplifier through a small condenser which passes the pulses but insulates the amplifier against the photomultiplier's high-anode voltage. Circuit details of such amplifiers are available in standard reference texts. They employ a few small condensers with companion resistors and one or more vacuum tubes. These can be taken from old radio sets or procured inexpensively on the war-surplus market.

To sort the now amplified noise and signal pulses, the output of the amplifier is fed into an "amplitude discriminator." This unit employs a single vacuum tube and exploits the fact that such tubes can be made to reject weak signals. If the grid of a three-element tube, or triode, is charged with an increasing negative voltage, a point is finally reached when the charge on the grid just equals the negative potential of electrons ejected from the tube's heated cathode. The force of repulsion between the two charges returns electrons to the cathode, and none passes through the grid to the tube's anode or plate. Hence, with this value of negative charge on the grid, no current flows in the plate or output circuit and the tube is said to be "cut off" by its negative "bias." When biased to cutoff, the tube becomes an amplitude discriminator. Large positive pulses superimposed in the grid neutralize a portion of its negative charge, "swing" it positive, and hence permit electrons to flow through the grid from cathode to plate, and out into the circuit as an amplified image or "triggering" pulse. Positive pulses of low amplitude, however, cannot raise the grid above cutoff and hence they have no effect on the output circuit. The designer can adjust the bias on the grid to any value desired and thus control the point at which it begins to discriminate.

Although the discriminator prevents pulses below a predetermined amplitude from flowing in the output, it passes signals above this value in a broad range of energies or sizes. The efficiency of the counter can be improved by "reshaping" these so that all are of equal voltage and current, and span equal time intervals. This is accomplished by means of a "pulse shaper," a circuit that takes pulses of random size and generates from them a uniform pulse, usually in the form of a single wave with a flat top and straight sides.

Ultimately these square waves must be counted by the electromechanical register. But these are sluggish instruments compared with pulse shapers. Those within reach of the average amateur's pocketbook are limited to about five counts per second. To operate the register, therefore, the amateur must construct one more circuit-a scaler. This unit is really an electronic dividing machine, telescoping the number of incoming pulses down to a half, a tenth, a thousandth or any other desired submultiple of the original pulses. Essentially the scaler uses the ability of a condenser to store electrical energy. Incoming pulses from the preceding circuit enter the condenser through a diode (a vacuum tube with only a cathode and a plate) which, acting as an electrical check valve, prevents their escape. As a series of pulses flows into the scaler a voltage builds up across the condenser step by step. Finally the growing charge attains a value sufficient to neutralize the negative bias on the grid of a following triode. When the cutoff point is reached, the tube begins to conduct and the condenser discharges, sending a single pulse from the tube's output. The amount of the bias can be adjusted, of course, and any submultiple of the scintillation frequency can thus be selected.

The pulses from the scaler drive the electromechanical register. Most registers employ a set of index wheels, like those that show mileage on an automobile speedometer. They are driven by a ratchet actuated by a solenoid. Each pulse advances the ratchet one notch. Various registers designed for operation from vacuum tube circuits are available on the surplus market for a dollar or so. They range in speed of operation from 50 counts per second (Central Scientific Co.) to 5 counts per second (Western Electric Company). Some can be driven reliably by a few thousandths of an ampere. Several amateurs have built their own from discarded speedometers and old electromagnetic relays.

Scintillation counters range from light portable instruments to rack-mounted jobs weighing 1,000 pounds or more. Their physical proportions depend on the amateur's resources and the use he has in mind.

Robert Detenbeck of Kenmore, N. Y., a student at the University of Rochester, who has built a couple of counters writes: "My first model included only the most essential circuits and used many modified war-surplus and radio-receiver parts. I was not sure that it would function properly and did not want to risk any more money than necessary. Hence the counter was not very compact, but no sacrifice in performance was necessary.

"Connections to all the vital points were brought out to jacks on the front panel for ease of measuring electrode and signal voltages. In order to find the optimum voltage to apply to the photomultiplier tube, the high-voltage supply was made variable. This control proved to be a more satisfactory method of varying the sensitivity of the instrument than a conventional audio-amplifier gain control, so I decided to include it in future models.

"The multiplier tube was even more sensitive than I had imagined it to be. When it was exposed to normal room illumination, the first vacuum tube of the RG amplifier following the multiplier was completely paralyzed. Any thoughts of carelessness in building a light-shield for the tube were dismissed. Light so dim that my eye could not detect it produced considerable random noise in the output.

"The first phosphor I used was a silver-activated zinc sulfide X-ray screen. For quite a while I was at a loss about what to use for window material-a material that would exclude light but pass alpha rays to the phosphor. Finally an old filter condenser furnished some aluminum foil about a thousandth of an inch thick. By careful selection I found a large piece without pinholes. The photo multiplier was mounted in a cylindrical container, the tube socket at one end and the aluminum window at the other, opposite the photocathode. On the first -test a fairly hot sample of pitchblende indicated that the zinc sulfide screen was detecting alpha rays and some beta rays with enough efficiency to override the noise background. Zinc sulfide is opaque to its own phosphorescence. Hence scintillations produced inside the zinc sulfide crystals by deeply penetrating beta and gamma rays cannot escape to act on the photocathode. Therefore the next step was to try an organic phosphor. The prospect of obtaining a large block of naphthalene, which is transparent to its own scintillations, seemed quite formidable. Finally a simple and quite satisfactory solution presented itself. I bought some naphthalene flakes of the kind used as a moth repellent. The lot contained some large, clear, thick pieces. Although the flakes were doubtless too thin for efficient gamma ray detection, a cell was made from them to replace the zinc sulfide screen. The flakes detected beta rays with good efficiency and worked as well on gamma rays as a professional Geiger-Muller counter which I borrowed to make the comparison. The problem of making a window for the organic phosphor proved easy. Because of the greater penetrating power of beta and gamma rays, two layers of aluminum foil can be used, the solid portions of each foil masking the pinholes in the other.

"The success of the first model encouraged me to construct a second one, more compact, which I could move without the aid of a wheelbarrow. A more efficient layout, purchase of a few new parts and some sacrifice of appearance achieved a miracle of condensation. In addition, the second scintillation counter included a discriminator circuit which passed the pulses from the scintillations but blocked most of the relatively weaker noise pulses. Of course it was not perfect; it lost some signals and passed some noise. It is not easy to find simple information on counting circuits, but one good book, Electronics Experimental Techniques, by William C. Elmore and Matthew Sands (McGraw-Hill Book Company) gives complete circuit data on every piece of apparatus that I needed. I also used a circuit which gave uniformly shaped output pulses, so that the average charge they placed on a condenser per second could be measured. A counting-rate meter connected to the output then gave an indication of the intensity of the radiation striking the phosphor. With this model two other phosphors-cadmium tungstate and calcium tungstate crystals-were tried. The calcium tungstate gave consistently better results.

"My current problem is that I cannot find hours enough in the day to finish all the projects the scintillation counter suggests. I intend to build a lighter model than the second. I would like to experiment with counting circuits. Two vacuum tubes enable you to add, subtract, multiply, divide and do exponentials. You can integrate and differentiate without the use of tubes, and these are only a few of the special circuits I would like to try. Finally, I want to use my counter in the field. They tell me a lot of 'hot' rock is waiting to be found."

A GROUP of amateur and professional instrument makers in San Francisco now has the distinction of having built the first Zeiss-type planetarium projector in the world outside of the famous German factory of Carl Zeiss. Until recently there were only 27 Zeiss planetariums in existence, six of them located in the U. S. These six-projectors were all built before World War II. The first was installed at Chicago in 1930, the next at Philadelphia in 1933, then Los Angeles and New York in 1935, Pittsburgh in 1939 and Chapel Hill, N. C., in 1949, the last being a secondhand instrument bought in Sweden.

When the San Francisco group started, it had only the most general information about the inner workings of a star-and-planet projector, though the Zeiss instrument had often been described- most technically in American Machinist in 1929. The modified projector built by this group is now the heart of the new Alexander F. Morrison Planetarium run by the California Academy of Sciences.

How all this came about is told in detail in the Academy's periodical, Pacific Discovery. Before the war the Academy had two men on its staff with a practical knowledge of scientific instrument making and design. One of these was G. Dallas Hanna, curator of paleontology, who had pursued instruments as a sideline, made improvements on the microscope and organized a well-equipped instrument shop. The second man was A. S. Getten, an experienced instrument maker, who maintained the institution's research equipment. During the war these two emptied a storeroom of fossils next to the shop and set up 50 workers to service U. S. Navy submarine periscopes, anti-aircraft gunsights, range finders and 6,000 binoculars. Other members of the Academy's staff also pitched in, including an ichthyologist, an aquatic biologist, a herpetologist and an operating engineer. When a serious shortage of personnel developed for the making of lenses, prisms and optical flats needed in the instruments, this department of SCIENTIFIC AMERICAN quickly put them in touch with nearby amateur telescope makers-a freight conductor, paleontologist, engineer, marine engineer and teletype specialist-who promptly became part of the Academy group. In four years this group manufactured 10,000 new optical parts for Navy instruments.

The enlarged group often talked optics, and the conversation always turned to planetariums. It was agreed that a planetarium was needed in San Francisco, but even before the war the cost of a Zeiss projector was more than $100,000, and no such money was then in sight. After the war, however, San Franciscans raised $500,000 - $350,000 of it from the Alexander F. Morrisons, the rest from citizens of high and low degree, foundations, estates and school children-to build a planetarium. The projector itself had to be built in San Francisco because Carl Zeiss was in the Eastern Zone of Germany and could not supply one, not even the detailed plans for one.

When it actually came to spending the funds on such a costly and admittedly experimental project as building a large instrument with 321 lenses and 25,000 parts, the Academy trustees were hesitant. They were hesitant until Russell W. Porter, a consultant on the project, surveyed the available equipment and skills and assured them that the Academy group could build a star projector equal to or better than the Zeiss. "That did it," says Academy Director Robert Cunningham Miller.

Hanna supervised the job, but he says that Getten, more than any other person, is responsible for the design of the projector and its driving mechanism, with intricate gear trains. It took four years to complete and cost $140,000, even though San Francisco firms gave materials and services free or at nominal price. "Everyone who worked on the instrument stayed to the end," says Hanna, "and all of us are still on speaking terms."

The drawing above shows the most obvious differences between a typical Zeiss projector and the modification designed by the California Academy. In the dumbbell-shaped Zeiss, the moon and planet projectors are nearest the center, while the heavier star projectors are on the outer end of the dumbbell. In the Academy model, these positions are reversed, greatly reducing the likelihood of vibration when the 5,000-pound instrument is first set in latitude motion, and making it a more rugged instrument all around. The Academy projector is quieter in operation than the Zeiss and is the first in which, at the flick of a switch, a tape recording will take over (if the lecturer is called away or has a frog in his throat) and automatically put the mechanism through 250 successive stunts, stopping it at the end like an automatic washing machine.

In Zeiss projectors the star images are made by shining the central light through tiny, unrealistically round holes in metal plates. Hanna and his associates improve on this by projecting the light through irregular images on glass plates, made by laying tiny, irregular particles on the plates, then putting an opaque aluminized coating over the whole surface and brushing away the particles, leaving spots of clear glass. Thirty-two such plates are needed to cover the entire sky. The plates used are the flat backs of condensing lenses, mounted behind the 32 projecting lenses of the instrument. An immense amount of work went into their development.

First Leon E. Salanave, now a planetarium lecturer in astronomy, sorted out 3,800 stars from a star catalog and computed the positions of the stars on each plate within one thousandth of an inch in two directions. Then Frances M. Greeby of the staff sorted grains of Alundum for proper size and shape under a dissecting microscope-.015-inch particles for stars of the first magnitude down to .0015-inch grains for stars of magnitude 5.80. (Twenty brighter stars were also made up in special projectors, some in color.) Another microscope with cross hairs in its eyepiece was equipped with right-angled micrometer screws so that it could be traversed and positioned accurately above the plates. Then a plate was put under the microscope and a grain of correct size pushed about on it with the extremely fine tip of a drawn glass rod until it was positioned under the cross hairs. This delicate operation was repeated 60 to 222 times per plate, each plate requiring three days of fatiguing work. The 3,800 holes thus produced, however, project as sharp images strikingly like stars as seen in the skies.

Until now the Academy instrument's moon projector has never been described. While it is the most complicated part of the apparatus-the "instrument designer's nightmare" according to Salanave-this is entirely the fault of the uncooperative, wobbly moon, whose motion it must simulate. Roger Hayward's drawings above and below make the mechanism relatively simple.

As on the Zeiss instrument, all projectors for moon and planets are double because the light from one of the halves is at times cut off by the framework needed to support the projector. Salanave writes: "In front of each projector lamp is a moon transparency. In front of these are phase shutters for changing the moon's phases. Next in the optical train come two pairs of fixed diagonal mirrors which help fold the optical train into available space, then negative lenses which increase the focal length from 13-1/2 inches to 24 inches, more diagonal mirrors, and objective lenses that project images of the moon to movable mirrors on a common axis. These mirrors must be movable, because the moon, as it travels around the earth, moves five degrees each month, alternately north and south of the sun's apparent path in the sky, the ecliptic. Motion is translated to the mirrors from a gear to a cam and through a pushrod to an arm attached to them.

"Getten, designer of the projector, thought it would be nice to project an actual photograph of the moon instead of the featureless lunar disk of the Zeiss. It was found experimentally that a moon transparency twice the correct diameter best matched the appearance of the real moon. Fred Chappell of the Lick Observatory skillfully produced, from one of the Moore-Chappell series of lunar photographs, pairs of positive prints of proper size and matched diameter, density and contrast.

"The most complicated single detail to work out proved to be the mechanism for producing smooth variation of phase each planetarium month. This problem took over Getten's thought and labors for months. A single 'cup,' like those shown in the right-hand insert in the drawing above, would produce all phases in the waning and waxing moon (starting at full) in one half-rotation and then would repeat if continued another half-rotation-unhappily with the illuminated portion on the wrong side of the terminator. The use of two transparencies, each with its cup, promised to resolve this difficulty but required switching from one projected image to the other at bright full moon. It proved that no amount of care in adjusting lamps and lenses could eliminate a very noticeable jump in the image at changeover. Getten's final solution was two pairs of oppositely moving cups in front of the two transparencies. This made it possible to effect the changeover at the 'dark' of the moon when no light comes through.

"The topmost part of the drawing above shows the phase mechanism lifted off and half turned over. Each pair of cups has two unconnected though concentric gears. It is fun to study this out. The lower cam in the middle actuates both lower gears because the gears are inescapably tied together by gear segments pivoted at the ball bearings at the top. Later, as the cams are rotated, the upper cam actuates both upper gears in an opposite direction."

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.