Part 5
This crude arrangement was changed in the latter part of 1880 to a screw shell and a ring for the base terminals, wood being used for insulation. The socket was correspondingly changed. This was a very bulky affair, so the base was changed to a cone-shaped ring and a screw shell for terminals. Wood was used for insulation, which a short time after was changed to plaster of Paris as this was also used to fasten the base to the bulb. It was soon found that the tension created between the two terminals of the base when the lamp was firmly screwed in the socket often caused the plaster base to pull apart, so the shape of the base was again changed early in 1881, to the form in use today.
An improved method of connecting the ends of the filament to the leading-in wires was adopted early in 1881. Formerly this was accomplished by a delicate clamp having a bolt and nut. The improvement consisted of copper plating the filament to the leading-in wire.
In the early part of the year 1881 the lamps were made “eight to the horsepower.” Each lamp, therefore, consumed a little less than 100 watts, and was designed to give 16 candlepower in a horizontal direction. The average candlepower (spherical) in all directions was about 77 per cent of this, hence as the modern term “lumen” is 12.57 spherical candlepower, these lamps had an initial efficiency of about 1.7 lumens per watt. The lamps blackened considerably during their life so that just before they burned out their candlepower was less than half that when new. Thus their mean efficiency throughout life was about 1.1 l-p-w (lumens per watt). These figures are interesting in comparison with the modern 100-watt gas-filled tungsten-filament lamp which has an initial efficiency of 12.9, and a mean efficiency of 11.8, l-p-w. In other words the equivalent (wattage) size of modern lamp gives over seven times when new, and eleven times on the average, as much light for the same energy consumption as Edison’s first commercial lamp. In the latter part of 1881 the efficiency was changed to “ten lamps per horsepower,” equivalent to 2¼ l-p-w initially. Two sizes of lamps were made: 16 cp for use on 110-volt circuits and 8 cp for use either direct on 55 volts or two in series on 110-volt circuits.
EDISON’S THREE-WIRE SYSTEM
The distance at which current can be economically delivered at 110 volts pressure is limited, as will be seen from a study of Ohm’s law. The loss of power in the distributing wires is proportional to the square of the current flowing. If the voltage be doubled, the amount of current is halved, for a given amount of electric power delivered, so that the size of the distributing wires can then be reduced to one-quarter for a given loss in them. At that time (1881) it was impossible to make 220-volt lamps, and though they are now available, their use is uneconomical, as their efficiency is much poorer than that of 110-volt incandescent lamps.
Edison invented a distributing system that had two 110-volt circuits, with one wire called the neutral, common to both circuits so that the pressure on the two outside wires was 220 volts. The neutral wire had only to be large enough to carry the difference between the currents flowing in the two circuits. As the load could be so arranged that it would be approximately equal at all times on both circuits, the neutral wire could be relatively small in size. Thus the three-wire system resulted in a saving of 60 per cent in copper over the two-wire system or, for the same amount of copper, the distance that current could be delivered was more than doubled.
DEVELOPMENT OF THE ALTERNATING CURRENT CONSTANT POTENTIAL SYSTEM
The distance that current can be economically distributed, as has been shown, depends upon the voltage used. If, therefore, current could be sent out at a high voltage and the pressure brought down to that desired at the various points to which it is distributed, such distribution could cover a much greater area. Lucien Gaulard was a French inventor and was backed by an Englishman named John D. Gibbs. About 1882 they patented a series alternating-current system of distribution. They had invented what is now called a transformer which consisted of two separate coils of wire mounted on an iron core. All the primary coils were connected in series, which, when current went through them, induced a current in the secondary coils. Lamps were connected in multiple on each of the secondary coils. An American patent was applied for on the transformer, but was refused on the basis that “more current cannot be taken from it than is put in.” While this is true if the word energy were used, the transformer can supply a greater current at a lower voltage (or vice versa) than is put in, the ratio being in proportion to the relative number of turns in the primary and secondary coils. The transformer was treated with ridicule and Gaulard died under distressing circumstances.
Information regarding the transformer came to the attention of William Stanley, an American, in the latter part of 1885. He made an intensive study of the scheme, and developed a transformer in which the primary coil was connected in multiple on a constant potential alternating-current high-voltage system. From the secondary coil a lower constant voltage was obtained. An experimental installation was made at Great Barrington, Mass., in the early part of 1886, the first commercial installation being made in Buffalo, New York, in the latter part of the year. This scheme enabled current to be economically distributed to much greater distances. The voltage of the high-tension circuit has been gradually increased as the art has progressed from about a thousand volts to over two hundred thousand volts pressure in a recent installation in California, where electric power is transmitted over two hundred miles.
INCANDESCENT LAMP DEVELOPMENTS, 1884–1894
In 1884 the ring of plaster around the top of the base was omitted; in 1886 an improvement was made by pasting the filament to the leading-in wires with a carbon paste instead of the electro-plating method; and in 1888 the length of the base was increased so that it had more threads. Several concerns started making incandescent lamps, the filaments being made by carbonizing various substances. “Parchmentized” thread consisted of ordinary thread passed through sulphuric acid. “Tamadine” was cellulose in the sheet form, punched out in the shape of the filament. Squirted cellulose in the form of a thread was also used. This was made by dissolving absorbent cotton in zinc chloride, the resulting syrup being squirted through a die into alcohol which hardened the thread thus formed. This thread was washed in water, dried in the air and then cut to proper length and carbonized.
The filament was improved by coating it with graphite. One method, adopted about 1888, was to dip it in a hydrocarbon liquid before carbonizing. Another, more generally adopted in 1893 was a process originally invented by Sawyer, one of the Americans who had attempted to “sub-divide the electric light” in 1878–79. This process consisted of passing current through a carbonized filament in an atmosphere of hydrocarbon vapor. The hot filament decomposed the vapor, depositing graphite on the filament. The graphite coated filament improved it so it could operate at 3½ lumens per watt (initial efficiency). Lamps of 20, 24, 32 and 50 candlepower were developed for 110-volt circuits. Lamps in various sizes from 12 to 36 cp were made for use on storage batteries having various numbers of cells and giving a voltage of from 20 to 40 volts. Miniature lamps of from ½ to 2 cp for use on dry batteries of from 2½ to 5½ volts, and 3 to 6 cp on 5½ to 12 volts, were also made. These could also be connected in series on 110 volts for festoons. Very small lamps of ½ cp of 2 to 4 volts for use in dentistry and surgery were made available. These miniature lamps had no bases, wires being used to connect them to the circuit.
Lamps for 220-volt circuits were developed as this voltage was desirable for power purposes, electric motors being used, and a few lamps were needed on such circuits. They are less efficient and more expensive than 110-volt lamps, their use being justified however only when it is uneconomical to have a separate 110-volt circuit for lighting. The lamps were made in sizes from 16 to 50 candlepower.
Electric street railway systems used a voltage in the neighborhood of 550, and lamps were designed to burn five in series on this voltage. These lamps were different from the standard 110-volt lamps although they were made for about this voltage. As they were burned in series, the lamps were selected to operate at a definite current instead of at a definite voltage, so that the lamps when burned in series would operate at the proper temperature to give proper life results. Such lamps would therefore vary considerably in individual volts, and hence would not give good service if burned on 110-volt circuits. The candelabra screw base and socket and the miniature screw base and socket were later developed. Ornamental candelabra base lamps were made for use direct on 110 volts, smaller sizes being operated in series on this voltage. The former gave about 10 cp, the latter in various sizes from 4 to 8 cp. The miniature screw base lamps were for low volt lighting.
The various manufacturers of lamps in nearly every instance made bases that were very different from one another. No less than fourteen different standard bases and sockets came into commercial use. These were known as, Brush-Swan, Edison, Edi-Swan (double contact), Edi-Swan (single contact), Fort Wayne Jenny, Hawkeye, Indianapolis Jenny, Loomis, Mather or Perkins, Schaeffer or National, Siemens & Halske, Thomson-Houston, United States and Westinghouse. In addition there were later larger sized bases made for use on series circuits. These were called the Bernstein, Heisler, Large Edison, Municipal Bernstein, Municipal Edison, Thomson-Houston (alternating circuit) and Thomson-Houston (arc circuit). Some of these bases disappeared from use and in 1900 the proportion in the United States was about 70 per cent Edison, 15 per cent Westinghouse, 10 per cent Thomson-Houston and 5 per cent for all the others remaining. A campaign was started to standardize the Edison base, adapters being sold at cost for the Westinghouse and Thomson-Houston sockets so that Edison base lamps could be used. In a few years the desired results were obtained so that now there are no other sockets in the United States but the Edison screw type for standard lighting service. This applies also to all other countries in the world except England where the bayonet form of base and socket is still popular.
THE EDISON “MUNICIPAL” STREET LIGHTING SYSTEM
The arc lamp could not practically be made in a unit smaller than the so-called “1200 candlepower” (6.6 ampere) or “half” size, which really gave about 350 spherical candlepower. A demand therefore arose for a small street lighting unit, and Edison designed his “Municipal” street lighting system to fill this requirement. His experience in the making of dynamos enabled him to make a direct current bipolar constant potential machine that would deliver 1000 volts which later was increased to 1200 volts. They were first made in two sizes having an output of 12 and 30 amperes respectively. Incandescent lamps were made for 3 amperes in several sizes from 16 to 50 candlepower. These lamps were burned in series on the 1200-volt direct current system. Thus the 12-ampere machine had a capacity for four series circuits, each taking 3 amperes, the series circuits being connected in multiple across the 1200 volts. The number of lamps on each series circuit depended upon their size, as the voltage of each lamp was different for each size, being about 1½ volts per cp.
A popular size was the 32-candlepower unit, which therefore required about 45 volts and hence at 3 amperes consumed about 135 watts. Allowing 5 per cent loss in the wires of each circuit, there was therefore 1140 of the 1200 volts left for the lamps. Hence about 25 32-candlepower or 50 16-candlepower lamps could be put on each series circuit. Different sizes of lamps could also be put on the same circuit, the number depending upon the aggregate voltage of the lamps.
A device was put in the base of each lamp to short circuit the lamp when it burned out so as to prevent all the other lamps on that circuit from going out. This device consisted of a piece of wire put inside the lamp bulb between the two ends of the filament. Connected to this wire was a very thin wire inside the base which held a piece of metal compressed against a spring. The spring was connected to one terminal of the base. Should the lamp burn out, current would jump from the filament to the wire in the bulb, and the current then flowed through the thin wire to the other terminal of the base. The thin wire was melted by the current, and the spring pushed the piece of metal up short circuiting the terminals of the base. This scheme was later simplified by omitting the wire, spring, etc., and substituting a piece of metal which was prevented from short circuiting the terminals of the base by a thin piece of paper. When the lamp burned out the entire 1200 volts was impressed across this piece of paper, puncturing it and so short circuiting the base terminals. Should one or more lamps go out on a circuit, the increase in current above the normal 3 amperes was prevented by an adjustable resistance, or an extra lot of lamps which could be turned on one at a time, connected to each circuit and located in the power station under the control of the operator. This system disappeared from use with the advent of the constant current transformer.
THE SHUNT BOX SYSTEM FOR SERIES INCANDESCENT LAMPS
Soon after the commercial development of the alternating current constant potential system, a scheme was developed to permit the use of lamps in series on such circuits without the necessity for short circuiting a lamp should it burn out. A reactance, called a “shunt box” and consisting of a coil of wire wound on an iron core, was connected across each lamp. The shunt box consumed but little current while the lamp was burning. Should one lamp go out, the entire current would flow through its shunt box and so maintain the current approximately constant. It had the difficulty, however, that if several lamps went out, the current would be materially increased tending to burn out the remaining lamps on the circuit. This system also disappeared from use with the development of the constant current transformer.
THE ENCLOSED ARC LAMP
Up to 1893 the carbons of an arc lamp operated in the open air, so that they were rapidly consumed, lasting from eight to sixteen hours depending on their length and thickness. Louis B. Marks, an American, found that by placing a tight fitting globe about the arc, the life of the carbons was increased ten to twelve times. This was due to the restricted amount of oxygen of the air in the presence of the hot carbon tips and thus retarded their consumption. The amount of light was somewhat decreased, but this was more than offset by the lesser expense of trimming which also justified the use of more expensive better quality carbons. Satisfactory operation required that the arc voltage be increased to about 80 volts.
This lamp rapidly displaced the series open arc. An enclosed arc lamp for use on 110-volt constant potential circuits was also developed. A resistance was put in series with the arc for use on 110-volt direct current circuits, to act as a ballast in order to prevent the arc from taking too much current and also to use up the difference between the arc voltage (80) and the line voltage (110). On alternating current, a reactance was used in place of the resistance.
The efficiencies in lumens per watt of these arcs (with clear glassware), all of which have now disappeared from the market, were about as follows:
6.6 ampere 510 watt direct current (D.C.) series arc, 8¼ l-p-w. 5.0 ampere 550 watt direct current multiple (110-volt) arc, 4½ l-p-w. 7.5 ampere 540 watt alternating current (A.C.) multiple (110-volt) arc, 4¼ l-p-w.
The reason for the big difference in efficiency between the series and multiple direct-current arc is that in the latter a large amount of electrical energy (watts) is lost in the ballast resistance. While there is a considerable difference between the inherent efficiencies of the D. C. and A. C. arcs themselves, this difference is reduced in the multiple D. C. and A. C. arc lamps as more watts are lost in the resistance ballast of the multiple D. C. lamp than are lost in the reactance ballast of the multiple A. C. lamp.
This reactance gives the A. C. lamp what is called a “power-factor.” The product of the amperes (7.5) times the volts (110) does not give the true wattage (540) of the lamp, so that the actual volt-amperes (825) has to be multiplied by a power factor, in this case about 65 per cent, to obtain the actual power (watts) consumed. The reason is that the instantaneous varying values of the alternating current and pressure, if multiplied and averaged throughout the complete alternating cycle, do not equal the average amperes (measured by an ammeter) multiplied by the average voltage (measured by a volt-meter). That is, the maximum value of the current flowing (amperes) does not occur at the same instant that the maximum pressure (voltage) is on the circuit.
THE FLAME ARC LAMP
About 1844 Bunsen investigated the effect of introducing various chemicals in the carbon arc. Nothing was done, however, until Bremer, a German, experimented with various salts impregnated in the carbon electrodes. In 1898 he produced the so-called flame arc, which consisted of carbons impregnated with calcium fluoride. This gave a brilliant yellow light most of which came from the arc flame, and practically none from the carbon tips. The arc operated in the open air and produced smoke which condensed into a white powder.
The two carbons were inclined downward at about a 30-degree angle with each other, and were of small diameter but long, 18 to 30 inches, having a life of about 12 to 15 hours. The tips of the carbons projected through an inverted earthenware cup, called the “economizer,” the white powder condensing on this and acting not only as an excellent reflector but making a dead air space above the arc. The arc was maintained at the tips of the carbons by an electro-magnet whose magnetic field “blew” the arc down.
Two flame arc lamps were burned in series on 110-volt circuits. They consumed 550 watts each, giving an efficiency of about 35 lumens per watt on direct current. On alternating current the efficiency was about 30 l-p-w. By use of barium salts impregnated in the carbons, a white light was obtained, giving an efficiency of about 18 l-p-w on direct current and about 15½ on alternating current. These figures cover lamps equipped with clear glassware. Using strontium salts in the carbons, a red light was obtained at a considerably lower efficiency, such arcs on account of their color being used only to a limited extent for advertising purposes.