NEARLY 200 YEARS AGO William Watson, an English physician who loved to tinker in fields in which he was an amateur, tackled a fundamental question that was bothering all the great minds in the newly discovered field of electricity. An electrified body, it was observed, gradually loses its charge, no matter how well it is insulated. How does the charge escape? Is it carried off by the surrounding air? Watson made an experiment which he hoped would settle this question. He sealed two wires in a glass bulb, leaving a gap between them, and applied a voltage across the gap. Then he began to evacuate the air from the bulb with a pump. If the leakage of charge across the gap declined as the air was removed, he would have demonstrated that air could indeed carry electrical charge. But to his dismay, the leakage actually increased as air was evacuated from the bulb. What was even more mystifying, when he had removed most of the air the bulb suddenly became filled with a pink glow that danced and shimmered like the Northern Lights.

Watson, not realizing that a gas becomes a better conductor of electricity as its pressure is lowered, considered his experiment a failure. He named his bulb the "aurora tube" and bequeathed it to the world as a curiosity. Actually it is doubtful that any phenomenon found in nature has been more fruitful than Watson's discovery of the glow discharge. It gave rise to multitudes of experiments by thousands of physicists. Modifications of Watson's tube-cathode-ray tubes- led Wilhelm Roentgen to the discovery of X-rays, J. J. Thomson to the discovery of electrons and John Fleming to the invention of the radio tube. More recent versions of the tube have become familiar in the shape of neon signs, Geiger counters, fluorescent lamps, voltage regulators, counting tubes for digital computers and many other gadgets.

Forrest H. Frantz of Mississippi State College, a physicist who has made a specialty of investigating varieties of these tubes, observes: "We can discern no end to these remarkable diggings. The more we work them, the more pay dirt we strike. Within the last 10 years one branch, which has to do with the leaky insulators called semiconductors, yielded the transistor, a device which may well open a whole new era of technology. Another promising line of work going on now has to do with the way electrons are emitted by electrodes shaped into exquisitely fine points [see "A New Microscope," by Erwin W. Muller; SCIENTIFIC AMERICAN, May, 1952]. Another branch is the micro-gap-a pair of closely spaced electrodes which behave in unexpected ways when a voltage is applied across them. An amateur who has not checked into electrical discharges, particularly glow discharges, is missing out on a chance for some fascinating experiments.

"One of the first to make real headway in the field of glow discharges was the French physicist J. P. Gassiot. By introducing traces of individual gases into the aurora tube, he produced glows of various characteristic colors. Neon gave an orange glow; hydrogen, crimson; mercury vapor, violet, and so on. Gassiot's work was one of the stepping stones toward modern spectroscopy.

"The great popularizer of the glow discharge was an amateur-the German Heinrich Geissler. By profession he was an instrument maker, but his name is remembered for his work on discharge tubes. He made them in such profusion and in so many intriguing shapes and forms that they are known as 'Geissler tubes.' It seems ironic that he should be known as the father of neon advertising signs while the name of Masson, their discoverer, goes almost unmentioned.

"By 1869 it was clear that rarefied gas conducts better than gas under pressure for the reason that molecules have greater freedom of movement in a partial vacuum. But if the last molecule could be removed from a tube, the conduction should, it was reasoned, drop to zero. Johann W. Hittorf of Germany built a really good air pump in order to confront this theory with a conclusive test. Although the pump did not achieve the primary objective, the by-product of the experiment assured Hittorf a lasting place in the history of science. As the pumping progressed past where the characteristic glow appeared, Hittorf observed near one of the electrodes a dark region which grew larger as the pressure dropped. He also noted a strange fluorescence that lighted the walls of the tube behind the positive electrode. He suspected this fluorescence was excited by rays emitted by the negative electrode. To test this, he sealed a small cross of lead inside the tube between the cathode and anode. It cast a shadow. As we now know, it was blocking a stream of electrons coming from the cathode. Here was the primitive beginning of television!

"The behavior of electrical discharges through a gas is influenced by many factors which can be adjusted to produce a dazzling variety of effects. The most important variables are: the pressure of the gas in which the discharge occurs, the type of gas, the size of the gap between the electrodes in the tube, and the details of the electrical circuit, the cathode material, the shape of the electrodes and the voltage across the gap.

"Not all these elements conveniently controlled in a home-built apparatus. Probably the easiest to play with is the pressure. To duplicate the effects pictured here [see Figure 1] you will need a 12-inch glass tube with electrodes sealed in, a fair vacuum pump and a high-voltage source. The tube can be bought at a scientific supply house. The pump which may be a mechanical one of the piston type, is not difficult to build, but the piston and valves must be fitted closely and should be sealed with a good grade of stopcock grease. The necessary voltage (about 15,000 volts) can be obtained from an induction coil as shown. A direct-current machine would be preferable, because the pattern of the glow is determined by the direction in which the current flows, but such machines are expensive. For casual observations an induction coil will do. It produces an alternating current, but with a much stronger surge in one direction than the other. The strong voltage pulses come fast enough so that the discharge will appear steady although it is actually intermittent.

"It is interesting to apply a constant voltage to the electrodes of the tube and observe the results as the pressure is progressively lowered.

"If the experiment is performed in a fairly dark room, you will see the glow of the corona discharge around the electrodes, when the voltage is applied. Then, as the pressure drops, occasional spark streamers crackle across the entire gap. These are like extremely thin lightning flashes. When the pressure gets down to about a hundredth of an atmosphere, the streamers appear continuously and in great numbers. Shortly thereafter the sparks give way to a silent glow-the pink of the aurora, sometimes called the Geissler discharge. With further lowering of the pressure the pink glow pulls away from the cathode, leaving a bright bluish glow next to the cathode and a dark space (known as the Faraday dark space) around that. As you continue to pump, both the 'negative' (bluish) glow and the Faraday dark space expand at the expense of the pinkish region that fills the rest of the tube. Next the negative glow separates from the cathode, having a second dark space (the Crookes dark space) and an orange-colored luminous sheath (the 'cathode glow') around the cathode. Simultaneously the pinkish glow breaks up into a series of separate vertical stripes. At still lower pressure the Crookes dark space grows until it pervades the entire tube. To a casual observer the cathode glow seems to make physical contact with the electrode. Actually it floats a fraction of an inch above it. In between is still another dark space called the Aston dark space. At the other end of the tube a similar 'anode' dark space separates the anode from the anode glow, which takes the form of a bright, thin striation at the extreme end of the positive column.

"The various effects, as they appear, give an approximate idea of the pressure of the gas in the tube. The noisy spark streamers change into the silent pink glow when about 90 per cent of the air has been removed. The pressure is then 5 just enough to support a column of mercury 80 millimeters high. (At sea level atmospheric pressure, air supports 760 mm. of mercury). At a pressure of nine mm. the brighter glow appears at the cathode and anode. Current through the tube increases sharply at this point, as you can observe by placing an ammeter in series in the circuit. The positive column appears at about five mm., along with the negative glow. Striations begin to break up the column at one mm. The glass walls of the tube fluoresce at one-half mm. under the bombardment of electrons from the cathode. The Crookes dark space appears at one-quarter mm. and begins to cut into the positive column seriously at one-tenth mm. The negative glow vanishes at one-fiftieth mm. and the anode glow at one-ninetieth mm. At a pressure of one 500th of a mm. X-ray emission begins, causing the glass walls to fluoresce again, this time with a distinctive color that depends upon the chemical composition of the glass. Often it is a greenish-blue. Below a thousandth of a mm., when all but a millionth part of the air has been exhausted, the electrical resistance of the gap increases sharply, and you must use an ammeter of high sensitivity to measure the minute trickle of current that continues to flow. A piston pump should be able to reach all but the lowest of these pressures.

"If you do not have facilities for building the pump and generator, you can still experiment with glow discharges by setting up micro-gaps [see the lower part of Figure 2 ]. They exhibit some of the effects observed in large tubes-which implies that they may replace these tubes in some electronic circuits. They can also be used as microphones, as instruments for investigating aerodynamic effects and in other applications which offhand might appear beyond a device of such primitive simplicity.

"Micro-gaps operate on relatively low voltages, which can be derived from transformers of the type used in radio receiving sets. The circuit for an adequate power supply is shown in the upper part of the drawing below. The output can reach about 900 volts, hence the experimenter should exercise appropriate caution when working with it. The voltage is varied by means of two rheostats connected in series with the primary circuit of the power circuit of the power transformer.

"An enclosed gap will sustain a glow discharge across as much as a quarter of an inch of separation if the pressure is made low enough. For the pump you use a converted automobile-tire pump. You simply reverse the piston washer and give it a thick coat of stopcock grease. The pump then sucks in air when the handle is pulled up. One sharp stroke of the pump handle will exhaust the gap sufficiently for most observations.

The vacuum is maintained by clamping the hose immediately. All the characteristic effects of the larger glow discharges will appear in the little gap, but in reverse order as air leaks slowly into the enclosure.

"The best material to use for the electrodes is platinum wire, but copper or will do. Both electrodes may be pointed, or the cathode may be spherical and the anode pointed. This makes it easier to align the electrodes. You can make a spherical anode by connecting two short lengths of fine platinum wire to the power supply, bringing the tips together and then separating them.

The arc that forms will melt the tip of one of the wires and surface tension will draw the molten metal into a bead. To limit the arc current, a resistor should be placed in series with one of the wires. For a platinum wire five thousandths of an inch in diameter the resistance should be 60,000 ohms. Thicker wire requires a smaller resistance.

"The resistor also serves as an automatic voltage and current regulator in the completed apparatus. You need a higher voltage to start a glow than to keep it going. At the instant the switch is turned Oil, before a spark has jumped the gap, there is no current. The full 900 volts of the power supply are exerted across the gap. When the gap is ruptured by the spark and electricity flows, some of the available pressure is used up pushing the current through the resistor. The voltage is now divided so that there is about a 600-volt drop through the resistor and a 300-volt drop across the gap.

"The upper limit of current through the gap can be adjusted by varying the resistance. If it is 60,000 ohms, the maximum will be about 10 milliamperes. For values of 600,000 ohms and six megohms, the upper limits are approximately one milliampere and a tenth of a milliampere respectively. (You may find that radio supply dealers do not stock resistors in these values. If not, the generally available values of 56,000, 560,000 and 5.6 million ohms, in the two-watt size, are satisfactory.)

"Open gaps of the type shown in the lower right portion of the drawing in Figure 2 have not received as much attention as those designed for operation at reduced pressure. Their electrode spacings cannot be more than about five thousandths of an inch. Unless they are studied under a microscope, many details of the glow will be missed. My son's toy microscope in the 50X range gives satisfactory results It shows, in addition to other details, the overheating of the cathode when the current is too high and bright flashes as bits of metal are sputtered off the cathode by positive ion bombardment.

"By adding a vacuum tube voltmeter or a sufficiently sensitive ammeter to your apparatus, you can determine the current flowing in the gap circuit. This will show the remarkable way in which the resistance of the gap varies through a wide range of applied voltage. The voltmeter measures current indirectly. You hook it across the resistor and measure the voltage drop. Dividing this figure by the value of the resistor gives the current through it, and, since the two are in series, through the gap also. It is more convenient, of course, to measure the current directly with an ammeter of appropriate range.

"An ammeter capable of measuring the billionth-ampere range will show that current actually starts to flow whenever a voltage, however small, is applied across the gap. This current consists of electrons dislodged from the metal of the electrodes by the action of light and also of electrons knocked off atoms in the air by cosmic rays and natural radioactivity. Normally these negative particles merely bounce around in the gap at random. When voltage is applied across the electrodes, however, they are pushed toward the anode, and the movement constitutes a current. As the voltage is increased, more of them join the parade and the current grows. A point is finally reached when all that are released reach the anode; further increases in voltage then fail to produce corresponding increases in current. If the gap is now shielded from light, the current will drop sharply. It does not fall to zero, however, because cosmic rays and radioactivity continue to create ions. Thus in the beginning the conductivity of the gap depends on the action of light and other external radiation. It is not self-sustaining.

"As the voltage is increased further, a point is reached when the accelerated electrons acquire sufficient energy to dislodge other electrons from the gas molecules by collision. These new electrons and resulting positive ions now join the current, are accelerated and in turn acquire ionizing energy. Thus the current increases exponentially. This phenomenon is called gas amplification. Although the current represented by the original photoelectrons has been increased enormously, the conductivity of the gap is not yet self-sustaining. When the light is shut off, it will drop to near zero. This is the region of the corona discharge.

"A slight further increase in voltage will cause the current to increase faster than the exponential rate. The gap will then show negative resistance, which means that, as the current increase voltage drop across the gap decreases. Now current will continue to flow when the light is shut off; the discharge has become self-sustaining. Between the self-sustaining (dark current) region and the self-sustaining (glow) region is the region of spark discharges. These are not necessarily accompanied by the bright flash and sharp click commonly associated with sparks.

"After the self-sustaining current has started, the resistance of the gap continues to drop for a time as the voltage increases. Eventually it reverses again, increasing positively. The two points of reversal mark the boundary of the glow discharge region. A still higher voltage drives the resistance into a third and final reversal. Again the resistance value is negative. Now you have reached the region of the electric arc. The current and temperature of the gap increase enormously as the arc grows and the electrodes erode rapidly.

"The currents associated with these various types of discharge are of the order of billionths of an ampere for the dark current region, millionths of an ampere for the corona, thousandths of an ampere for the glow and amperes for the arc. This is only a rough classification. You may, for example, get an arc discharge with a current of only one-hundredth ampere if you make the gap spacing small enough.

"A plot of the current-voltage relationship across a micro-gap of five thousandths of an inch in air will have the general shape of the curve in the chart in Figure 3. The resistances to be used for the various current ranges are shown. Observe that in the A region an increase in current causes a decrease in voltage thus indicating negative resistance. This property is exhibited by only a few devices other than the discharge gap notably the vacuum tube and the transistor. It is of immense practical interest.

"One of the many jobs that a negative resistance device can do is to switch a circuit on and off automatically at high speed. For this purpose it is used as a 'relaxation oscillator.' A diagram for one that employs a micro-gap as the active circuit element appears as the top diagram in Figure 4. In this hookup the discharge across the gap is automatically turned on and off at a steady rate

"The frequency of the oscillator may be varied by changing the value of either the capacitor, which is connected across the gap, or the resistance. When the product of the value of the capacitor is multiplied by the resistance is large, the frequency will be low; when their product is low, the frequency will be high. The resistance must be kept fairly large or the device may refuse to oscillate.

"The B portion of the characteristic curve is interesting because the voltage remains essentially constant over a relatively wide range of current. This property may be applied for voltage regulation. Gas-filled tubes are widely used in electronic apparatus for this purpose, the glow being sustained by only a few milliamperes. I have built a circuit around one of these tubes that holds the output essentially constant at 150 volts while the input varies from 275 to 500 volts!

"In the C region of the characteristic curve the gap voltage increases as the current increases. Under the microscope you can see no change in the area cathode covered by the blue glow, but the intensity of the glow increases. This is a region of positive resistance, and variations in the pressure of the gas will cause comparable variations in the intensity of the current. This property suggests that the gap should be useful for translating rapid changes in air pressure into changes in current. It might, for example, convert sound waves into electrical impulses.

"In 1922 Phillips Thomas, a research engineer for the Westinghouse Electric Corporation, conceived the idea of putting a glow-discharge gap to work as a microphone at radio station KDKA. Having no diaphragm, it would avoid some of the resonance effects that impair the performance of conventional microphones. Thomas' microphone was used in broadcasts from KDKA for a time, but it was subject to high-frequency oscillations, noise and the disintegration of electrodes. Today, when we can take the engineering of radar and rockets in stride, these problems scarcely seem insurmountable.

"Last summer I did considerable work on adapting a glow discharge gap to detect turbulence in the air. I was trying to use it as the sensing element in a turbulence meter in our aerophysics laboratory.

"One of the major difficulties with micro-gaps is that their electrodes disintegrate rapidly. However, August Raspet, who heads our department, felt that the problem could be minimized by keeping the operating current very low, thus reducing the bombardment of the cathode. There were also other problems to be considered-oscillation, noise, maintaining the glow discharge and making the equipment small enough for airborne operation. We finally developed an experimental instrument which largely overcame these problems. It gave a reasonably stable measurement of turbulence for a period of 40 minutes-a demonstration that the micro-gap has promise of becoming a practical tool of aeronautical research workers.

"An amateur may observe the sensitivity of his micro-gap to air movements by connecting it into the circuit configuration shown at the bottom of the group of diagrams in Figure 4. Blowing on the gap even lightly causes a deflection of the meter

"The circuits presented here have been simplified to make their construction as easy as possible for beginners. The simplification was achieved, however, at the expense of refinements which can compensate for some of the shortcomings of glow-discharge gaps. An amateur can easily discover clues to the cure of these shortcomings by consulting reference texts. In the course of his experiments he may even hit upon an idea that will help to advance the utility of the glow discharge."

Bibliography

FUNDAMENTALS OF DISCHARGE TUBE CIRCUITS. V. J. Francis. Methuen & Co., Ltd., 1948.

GASEOUS CONDUCTORS: THEORY AND ENGINEERING APPLICATIONS. James Dillon Cobine. McGraw-Hill Book Company, Inc., 1941.

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/