Nikola Tesla Articles
Light Yesterday and Tomorrow Page 81
Westinghouse Engineer - January 1st, 1950
was more uniform. The filament material, a cellulose of nitrated cotton called tamadine, was dissolved in zinc chloride and squirted through platinum dies into alcohol to harden. In retrospect this seems a simple process, but even the final step — carbonizing the filaments in crucibles packed with carbon dust — required a masterpiece of effort to avoid warping.
At best, however, carbon-filament lamps produced only about four lumens per watt, and their burning life was irregular, averaging but a few hundred hours. Trouble was, these filaments were prone to evaporate in a high vacuum. However, the metalizing or hydrocarbon-treating process did result in a significant advance—it caused the filament to be self-regulating, because its resistance, as in a metal wire, increased with the temperature and limited the current.
In 1897 Mr. Westinghouse bought the American rights to an incandescent lamp developed by Dr. Walter Nernst, a German physicist. This lamp utilized “rare earth” oxides as illuminants, and was popular in store lighting until more convenient filament lamps appeared. With an efficiency about 50 percent greater than carbon lamps, the Nernst “glower” lamp could be burned in open air, and produced a beautiful white light, close to sunlight quality. Essentially the light source was a rod of calcium oxide, separately heated by a resistance coil (in shunt) to make it a conductor.
The Tungsten-Filament Lamp
The first tungsten lamps, introduced to the American public in 1906, were made by squirting a paste of tungsten powder and binder through fine dies to form hairpin-shaped filaments, which were then mounted in the bulb and pasted to lead-in wires. Since about ten inches of straight filament wire were required to obtain the resistance necessary for use on a 110-volt circuit, several short hairpin filaments were used in series in each lamp. Several fusions were necessary to fasten these filaments together and to attach lead-in wires, the end result being a lamp so fragile that workmen learned to expect a snapped filament if they merely jumped too heavily on the shipping-room floor. If these lamps escaped mechanical breakage of the filament, however, they burned well.
Two years after the advent of squirted-tungsten filaments, Westinghouse engineers developed the continuous-wire tungsten filament, which vastly improved shock resistance. This involved a one-piece squirted filament, which construction improved the flexibility, and lessened heat losses by elimination of a plurality of supports. The filament was squirted through a series of dies in a special atmosphere, and then sintered. This produced, for the first time commercially, a continuous tungsten lead-to-lead filament.
Although the continuous squirted-filament process was a revolutionary advance over fragile hairpin-type squirted filaments, its advantages did not remain for long. In 1911 the achievements of Dr. Coolidge of the General Electric Company made possible an even better filament. He succeeded for the first time in swaging, or hammering, and then drawing a fine tungsten wire. This drawn ductile wire at once asserted its superiority over squirted wire because of its enormously improved tensile strength, permitting lamps to operate successfully in trains, automobiles, and other places where mechanical shock is encountered.
But neither squirted nor drawn ductile tungsten wire might have been successful had it not been for a tungsten crystallization control devised in 1906 by Dr. Anton Lederer of the Westinghouse Vienna factory. Pure tungsten tends to form large crystals when burned on alternating current. These tend to “slide” by each other, or offset, along the cleavage planes, so that the effective cross-section varies, and fragility increases. But when Dr. Lederer added small amounts of thoria and other oxides to the tungsten, he found that small, non-weakening, longitudinal crystals were formed, and the wire remained strong while burning. Thus, “offsetting” was avoided.
Incandescent Lamp Nears Present Form
In 1912, Westinghouse made a mutual agreement with the General Electric Company that provided for the pooling of past and current lamp engineering, research, and machine-design information. (This agreement was terminated on August 1, 1945 by mutual consent.) The result was at once beneficial to the customer in standardization and in reduced prices.
By this time the present form of the incandescent lamp was established. In contrast with only 4 lumens per watt yielded by carbon-filament lamps in 1906, the gas-filled tungsten lamp produced some 14 lumens per watt.
Since the beginnings of gas-filled lamps in 1913, developments in tungsten incandescent lamps have been largely confined to perfections of construction and manufacture. However, they aggregate to sizable improvements in design and efficiency. Today’s 1000-watt general-service lamp, for example, produces about 22 lumens per watt. Household types have held to burning lives of 750 or 1000 hours as the most economical, since light-giving efficiency decreases rapidly if