Creating the Twentieth Century
Page 10
All transformers work by electromagnetic induction whose existence was demonstrated for the first time in 1831 by Michael Faraday. A loop of wire carrying AC (the primary winding) generates a fluctuating magnetic field that will induce a voltage in another loop (the secondary winding) placed in the field. In turn, the current now flowing in the secondary coil will induce a voltage in the primary loop, and both of the loops will also produce self-induction (figure 2.15). Because the total voltage induced in a loop is proportional to the total number of its turns (if the secondary has three times the number of turns of the primary, it will supply tripled voltage), and because the rate at which the energy is being transformed (W) must equal the product its voltage and its current (VA), the product of the secondary voltage and the secondary current must equal the product of the primary voltage and the primary current (leaving negligible losses aside). When Faraday wound two coils on an iron ring, rather than on a wooden one as in his initial induction experiments, he brought together the essential elements of the modern transformer.
FIGURE 2.15. Two principal designs of modern transformers made with laminated iron sheets. In the core form, on the left, the two circuits enclose separate arms of the transformer, while in the E-shaped shell form both the primary and the secondary coils envelope the central bar.
Here is the other important reason why transformers do not occupy such a prominent position in the history of the 19th-century invention: there was no protean mind behind the device, no eureka-type discovery, just gradual improvements based on Faraday’s experiments (Fleming 1901; MacLaren 1943). The need for transformers arose with the first tentative steps toward the AC system. Lucien H. Gaulard (1850-1888) and John D. Gibbs got the first patents for its realization in 1882 and displayed their transformer publicly in 1883 at the Electrical Exhibition in London. In order to add incandescent lightbulbs to an AC system that supplied arc lights connected in series (turned on and off at the same time), they used a two-coil transforming device that they called a secondary generator.
The secondary generator’s two windings were made of flat copper sheet rings inserted on cast iron wire core and insulated one from another by paper and varnish. A year later Gaulard and Gibbs staged another demonstration of this inefficient model at the international exhibition in Torino, Italy. In 1884 three Hungarian engineers employed at Ganz Works in Budapest improved the Gaulard-Gibbs device by designing two kinds of transformers for parallel connection to a generator and making them more efficient (Németh 1996). The key innovation—using closed iron cores that work much better than do the open-ended bundles—was suggested by Otto Titusz Blathy (1860–1939), the youngest engineer in the group. Shunt connection was the idea of Karoly Zipernowsky (1853–1942), head of the electrical section of Ganz Works, and the experiments were performed by Miksa Déri (1854–1938). These transformers (later called ZBD) were demonstrated for the first time in May 1885 at the Hungarian National Exhibition in Budapest, where 75 of them were used to step down the current distributed at 1.35 kV to energize 1,067 incandescent lightbulbs.
George Westinghouse purchased the rights to the Gaulard-Gibbs design (for $50,000) and took options on ZBD transformers as well, but Edison’s distrust of AC systems led him to ignore these critical developments, and transformers were the only fundamental electric device to whose development he did not make any significant contribution. In contrast, William Stanley, a young engineer working for Westinghouse, began designing improved transformers (he called them converters) in 1883, and by December 1885 he had a model that was much less expensive to make than the ZBD device and that became the prototype of devices we still use today (Stanley 1912; Hawkins 1951). In the first patented version (U.S. Patent 349,611 of September 21, 1886; three drawings on the left in figure 2.16), the soft-iron core, encircled by primary (b1) and secondary (b2) coil, was either annular or rectangular with curved angles and could consist either of a single piece of metal or wires or strips.
But, as shown in a patent drawing filed by George Westinghouse in December 1886 (U.S. Patent 366,362), soon the preferred composition of the core changed to thin plates of soft iron constructed with two rectangular openings and separated from each other, individually or in pairs, by insulating material (two drawings on the right in figure 2.16, right). From this design, it was a short step to stacked laminations that were stamped out from iron sheets in the form of letter E, which make it easy to slide prewound copper coils in place and then to lay straight iron pieces across the prongs to close the magnetic circuit; these became the most obvious marks of Stanley-type transformers (Coltman 1988; figure 2.15). Soon after the Westinghouse Electric Co. was incorporated in January of 1886, Westinghouse himself took out patents for improved designs of induction coils and transformers, including both the air-cooled and oil-insulated devices (both are still the standard practice today).
FIGURE 2.16. Drawings attached to William Stanley’s 1886 U.S. Patent 349,611 for induction coil, a prototype of modern transformers (left), and to U.S. Patent 366,362 filed by George Westinghouse in December 1886. Stanley’s device has an annular core; the later design has the core (A) made of thin insulated plates of soft iron with two rectangular openings.
Other patents for transformers were issued during the late 1880s to Elihu Thomson and Sebastian Ferranti, and in 1890 Ferranti designed the largest devices in operation for London’s Deptford station. Electricity was generated at 2.5 kV and 83.5 Hz, stepped up to 10 kV for more than 11 km transmission by underground cables, and then stepped down by transformers in the central London area to 2.4 kV required for the distribution system (Electricity Council 1973). These were three-coil devices with the central primary separated from the two secondaries by sheets of ebonite, with coils made up of copper strips separated by vulcanized fibers, covered in varnished cloth, and designed to be cooled by either air or oil; their reliability can be best judged by the fact that they remained in service for more than three decades (Dunsheath 1962).
Also in 1891 William Stanley formed his owned electric manufacturing company, whose transformers were very similar to devices that he designed previously for Westinghouse. A big increase in transformer performance was needed by 1895 in order to accommodate transmission from the Niagara Falls hydroelectric station and the specifications required by the project’s new industrial clients. Electricity generation, based on Tesla’s polyphase AC concept, was at 5 kV and 25 Hz, and three-phase transmission to Buffalo carried the current at 11 kV. General Electric built two 200-kVA transformers for the aluminum plant of the Pittsburgh Reduction Co., and a 750-kVA transformer for the Carborundum Co. (MacLaren 1943). By 1900 the largest transformers were rated at 2 MVA and could handle inputs of 50 kV. But these devices operated with considerable hysteresis and eddy current losses.
Hysteresis is the memory effect of magnetic materials that weakens the transforming capacity by delaying the magnetic response, and eddy currents are induced in metals with low resistivity. Both of these phenomena dissipate large amounts of energy, and they made the early transformer cores, made of pure iron, relatively inefficient. Eddy currents can be reduced by laminating the core, and in 1903 English metallurgist Robert A. Hadfield (1858–1940) discovered that silicon steel greatly increased its resistivity while leaving the magnetic properties largely intact. Moreover, with the same maximum flux density silicon steel has hysteresis losses 75% lower than does sheet iron, and it is not subject to the phenomenon of aging whereby the hysteresis loss can double with time when the metal is exposed to higher temperatures (> 65°C) or mechanical fatigue (Calvert 2001). All transformer cores, as well as parts of rotating machinery subject to alternating fields, are made of silicon steel, constantly saving a great deal of electricity.
More than a century after its invention, the transformer remains little more than an artfully assembled and well-cooled bundle of iron sheets and copper coils. But as with so many other pre-WWI inventions, this conservation of a fundamentally ideal design has not prevented some very impressive gai
ns in typical performance, mainly due to better production methods of silicon steel. Improved cooling and insulation were also introduced gradually beginning in the 1930s: low-power devices (up to 50 kVA) are still cooled by the natural air flow; larger ones have forced air circulation, and those above 200 kVA are usually immersed in mineral oil in a steel tank (Wildi 1981). Just before WWI, the largest transformers could work with inputs of up to 150 kV and power of 15 MW; today’s largest transformers are rated at more than 1 GVA and can accommodate currents up to 765 kV, and the best units come very close to an ideal device.
But Stanley’s (1912:573) appraisal is as correct today as it was when he confessed in 1912 to a meeting of American Institute of Electrical Engineers to
a very personal affection for a transformer. It is such a complete and simple solution for a difficult problem. It so puts to shame all mechanical attempts at regulation. It handles with such ease, certainty, and economy vast loads of energy that are instantly given to or taken from it. It is so reliable, strong, and certain. In this mingled steel and copper, extraordinary forces are so nicely balanced as to be almost unsuspected.
First Electric Motors
History of electric motors resembles that of incandescent lights because in both cases the eventual introduction of commercially viable devices was preceded by a long period of experimental designs, and because in neither case did these unsystematic effort led directly to a successful solution of an engineering challenge. Tesla’s contributions to the introduction and rapid diffusion of AC electric motors was no less important than Edison’s efforts to commercialize incandescent light. And although the two most common converters of electricity, lights and motors, fill two very different final demands and hence it is not really appropriate to rank them as to their socioeconomic importance, there is no doubt that 20th-century economic productivity, industrial and agricultural, was revolutionized even more by electric motors than it was by electric lights.
The electric motor is fundamentally a generator working in reverse, and so it is not surprising that the first attempts to use the changing electromagnetic field for motion date to the 1830s, to the years immediately following Faraday’s fundamental experiments. Given the eventual importance of motors, a great deal of historical research has been done to locate and to describe these earliest attempts as well as later developments preceding the first commercial introductions of DC motors during the 1870s and the invention of induction machines during the latter half of the 1880s (Hunter and Bryant 1991; Pohl and Muller 1984; Bailey 1911).
Certainly the most remarkable pioneering efforts were those of Thomas Davenport in Vermont in 1837 (U.S. Patent 132) and contemporaneous designs (beginning in 1834) by M. H. Jacobi in St. Petersburg. Davenport used his motors to drill iron and steel parts and to machine hardwood in his workshop, while Jacobi used his motors to power paddle wheels of a boat carrying 10–12 people on the Neva River. Both inventors had high hopes for their machines, but the cost and durability of batteries used to energize those motors led to early termination of their trials. A similar experience met Robert Davidson’s light railway car that traveled on some British railroads in 1843, and heavy motors built during early 1850s (with considerable congressional funding) by Charles Page in Massachusetts. None of these battery-powered devices could even remotely compete with steam power: in 1850 their operating cost was, even under the most favorable assumptions, nearly 25 times higher.
Viewed from this perspective, it is not surprising that during the 1850s two of England’s most prominent engineers held a very low opinion of such electric machines. Isambard K. Brunel (1806–1859), perhaps the most famous engineer, architect, and shipbuilder of his time, was against their inclusion in the Great Exhibition of 1851 as he considered them mere toys, and William Rankine (1820–1872), an outstanding student and popularizer of thermodynamics, wrote in 1859 that the true practical use of electromagnetism is not to drive machines but to make signals (Beauchamp 1997). Curiously, the first electric motor that was manufactured commercially and sold in quantity was Edison’s small device mounted on top of a stylus and driving a needle in rapid (8,000 punctures a minute) up-and-down motion. This stencil-making electric pen, energized by a bulky two-cell battery, was the first device to allow large-scale mechanical duplication of documents (Pessaroff 2002). Edison obtained a patent for the pen in August 1876, and eventually thousands of these devices were sold to American and European offices (Edison 1876).
And it was also during the late 1870s that the first practical opportunities to deploy DC motors not dependent on batteries arose with the commercial introduction of Gramme’s dynamo. This came about because of a fortuitous suggestion made in 1873 at the industrial exhibition in Vienna. Several incompatible versions of the story were published during subsequent decades, but the most authentic account comes from a letter sent in 1886 by Charles Felix, at that time the director of a sugar factory in Sermaize, to Moniteur Industriel and reproduced in full by Figuier (1888:281–282). Gramme Co. was exhibiting one dynamo powered by a gas engine and another one connected to a Voltaic pile.
During his visit of the exposition Feélix suggested to Hippolyte Fontaine, who was in charge of Gramme’s display, “Since you have the first machine that produces electricity and the second one that consumes it, why not let electricity from the first one pass directly to the second and thus dispense with the pile?” Fontaine did so, and the reversibility of machines and the means of retransforming electrical energy to mechanical energy were discovered. Electricity produced by the gas-engine-powered dynamo was led by a 100-m cable to the second dynamo, and this machine operating in reverse as a motor was used to run a small centrifugal pump.
Before the end of the decade came the first demonstration of a DC motor (at 130 V) in traction when Siemens & Halske built a short (300 m) circular narrow-gauge railway (Bahn ohne Dampf und ohne Pferde—railway without steam or horses—as an advertisement had it) at the Berlin Exhibition of 1879, with a miniature locomotive electrified from the third rail. A similar Siemens train was running in 1881 along Champs Elyséss during Paris Electrical Exposition in 1881, where Marcel Deprez displayed several motor-driven machines. As the first central electricity-generating stations were coming on line during the 1880s, such motor-driven devices began their transformation from display curiosities to commercial uses.
By 1887 the United States had 15 manufacturers of electric motors producing about 10,000 devices a year, most of them with ratings of only a small fraction of one horsepower, and many of them inefficient and unreliable (Hunter and Bryant 1991). Among the leading innovators in the United States were the Sprague Electric Railway & Motor Co. (the first one to demonstrate a powerful mine hoist in Aspen, Colorado, the precursor of electric elevators for tall buildings), the Thomson-Houston Co., and Eickemeyer Motors. Rudolf Eickemeyer’s contribution was the winding of armature coils in a form in order to mold them to the exact shape, and to insulate them thoroughly before emplacing them in the armature; both of these procedures became the industry standard (MacLaren 1943). But the company that made the greatest difference was, thanks to its rights to Tesla’s patents, Westinghouse Electric.
Nikola Tesla (1857–1943)—a largely self-educated Serbian engineer who worked before his emigration to New York in 1884 with new electric systems in Budapest and Paris (figure 2.17)—was not actually the first inventor to reveal publicly a discovery of the principle of the rotating electromagnetic field and its use in an induction motor. Galileo Ferraris (1847–1897) presented the same insight to the Royal Academy of Science in Torino on March 18, 1888, and he published his findings in April, a month before Tesla’s May 16, 1888, lecture at the American Institute of Electrical Engineers (Martin 1894; Popovié, Horvat, and Nikié 1956). But Tesla, who claimed that he got the idea for his device one afternoon in 1882 while walking in a Budapest park with a friend and reciting a stanza from Goethe’s Faust, was far ahead in developing the first practical machine (Ratzlaff and Anderson 1979; Cheney 1981; Se
ifer 1996).
FIGURE 2.17. Nikola Tesla: a portrait from the late 1880s when he worked on polyphase electric motors. Image available from Tesla Society.
He built its first models while working in Strasbourg in 1883, and after immigrating to the United States he hoped to develop the design for Edison’s company. He was hired by Edison immediately after presenting the letter of introduction he got in Paris from Charles Batchelor, Edison’s old associate, but the two incompatible men soon parted. Tesla, an outstanding scientist and a gifted mathematician, was appalled by what he felt to be Edison’s brute force approach to problem solving and believed that a bit of theory and a few calculations would have eliminated a great deal of wasteful Edisonian searching for technical solutions. In addition, Edison, with his new central electric system built on low-voltage DC, was not ready to embrace the radical ideas of an avid high-voltage AC promoter.
After he secured generous financing by investors who appreciated the potential impact of his inventions, Tesla set up his electric company in April 1887, and he proceeded to build single-phase, two-phase, and three-phase AC motors and design appropriate transformers and a distribution system. Almost immediately, he also began a spate of patent filings (totaling 40 between 1887 and 1891), and the two key patents for his polyphase motor (U.S. Patents 391,968 and 391,969, filed on October 12, 1887) were granted on May 1, 1888. After many challenges, they were finally upheld by the courts in August 1900. The first patent specification illustrates the principle of a polyphase motor’s operation (figure 2.18), and it gives clear indications of the simplicity and practicality of the device. That was Tesla’s foremost goal, not an interesting gadget but