A 1922 article in Electrical World designated Edison as the father of the central station industry [1]. Edison’s invention of practical incandescent electric lighting in 1879 was the driving force for the success of Edison’s commercial low-voltage short-range direct current (dc) transmission systems in the United States and elsewhere, beginning in 1882. However, in 1886, high-voltage long-range alternating current (ac) transmission systems of Ganz & Co. of Budapest began to compete with Edison’s dc systems in Europe, and ac systems of the Westinghouse Electric Co. began to compete with Edison’s dc systems in the United States. Edison reacted by reaching into the toolbox of politicians.

This article provides an alternative view into the early history of electric power transmission.

In the 1888–1889 period, Edison waged the so-called “battle of the currents” against George Westinghouse in defense of his dc electric power systems by means of a public campaign to create the perception that high-voltage ac was dangerous. Nevertheless, there was a steady rise in the number of installations of Westinghouse systems during the period. Edison’s verbal assaults became irrelevant in 1889 when financiers forced the merger of his electrical companies into a single one, the Edison General Electric Company, with Henry Villard in control. An epidemic of books with details about the “battle of the currents” began in 1996 [2]–​[6]. There is obviously a sizeable market for books about a “battle” that involved Edison, Tesla, and Westinghouse. The “battle of the currents” tumbled down the rabbit hole in 2016 with the publication of a “historical” novel in which there are accusations that Edison may have attempted to have Nikola Tesla assassinated and allegations that arson engineered by Edison or by Westinghouse burned down Tesla’s laboratory [7]. The author puts in the disclaimer “nothing you’ve read here should be understood as verifiable fact,” then backpedals with “the bulk of the events depicted in this book did happen” and lists nonfiction sources and experts he consulted. There are now two planned motion pictures about the “battle of the currents,” one based on the novel, each with major directors and actors in the leading roles [8], [9].

Various histories of electrification also misrepresent the “battle of the currents” in one of two ways, omission of crucial historical developments and opinions presented as facts. A prevalent sin of omission is focus on developments in the author’s country with perfunctory mention of important developments elsewhere. An example of an opinion impersonating a fact is the following quote from a history of steam and electric power published in 2008 [10]: “As 1889 opened, Edison was clearly winning the war of the currents. Thanks to Brown, he had seized the offensive and shifted the public’s attention from the comparative efficiency of the two systems to their relative safety.” Actually, in 1890, there were 202 Edison dc central stations and 323 Westinghouse ac central stations in the United States.

In an attempt to counteract prevalent distortion, this article will cover the early history of electric power transmission, beginning with the small pre-Edison systems for street lighting by arc lamps and ending with the decline of dc transmission in the 1920s and the introduction of modern high-voltage dc transmission in the 1950s. The history of early electrification is a story of international cross fertilization and competition devoid of any significant battles.

Arc lamps were used in searchlights, a few lighthouses, some large rooms, and street lighting [11, Ch. 4]. Constant-current dc generators powered multiple arc lamps arranged in series. At first, only isolated plants powered arc lamps. The first central station in the world was built by the Brush Electric Co. of Cleveland and owned by the California Electric Light Co. of San Francisco, which began offering arc light service to subscribers in 1879 [12]. By 1886, the Brush Electric Co. had installed arc light central stations in many cities in the United States [13]. Table 1 shows the number of arc lights and incandescent lights in use in the United States in 1881–1912, combined for lights powered by isolated plants and by central stations [14], [15]. The increasing popularity of incandescent electric lighting became the driving force for the installation of dc and ac transmission systems.

Table 1 Number of Arc Lights and Incandescent Lights in the United States in 1881–1912

For the purpose of this article, a “low voltage” is arbitrarily defined as 400 V or less and a “high voltage” as 1000 V or more. Intermediate voltages are not relevant here. A “short distance” is arbitrarily defined as 6 km or less and a “long distance” as more than 6 km. Practical long-distance transmission of electric power requires a high voltage (Fig. 1 [16]), which has to be downconverted to the voltage of the end user.

Fig. 1.

40-kW (dc) copper transmission lines of 200, 400, 800, and 1600 V, all with an acceptable 2% power loss at the destination [16]. Transmission distance (blue) is in kilometers, diameter of copper wire (orange) is in centimeters, and total weight of copper (gray) is in metric tons. As the voltage increases, the calculated transmission distance increases, the diameter of the copper cable decreases, while the total weight of copper remains constant, demonstrating the great advantage of high-voltage transmission.

Show All

Practical transformers for such conversion of dc power did not exist in the 19th century. In the United States and many other countries, high-voltage long-distance transmission began with ac systems for which practical downconversion transformers were developed. However, high-voltage dc generation and transmission competed with ac systems in some European countries. The first ever demonstration of long-distance power distribution took place in fall 1882, a 57-km 2200-V dc line of 4.5-mm diameter from Miesbach to the 1882 Munich Electrical Exhibition, designed by French electrical engineer Marcel Deprez. At the destination, a dc motor powered a centrifugal water pump feeding a small artificial waterfall [17]. In the absence of dc downconverting transformers, a commercially viable long-distance dc method invented by Swiss engineer René Thury used dynamos in series at the generating station and motors in series at the receiving stations as voltage dividers [18]. Società per l’Acquedotto De Ferrari Galliera of Genoa installed the first Thury system in 1889. It sent 940 kW of 14 000-Volt dc power to Genoa from a hydroelectric station 60 km away. The most famous Thury system, from Moutiers to Lyon in France, a distance of 180 km, began transmitting 4.7 MW of 57 000-V hydroelectric dc power in 1906 [18]. This system still operated at the start of 1936 [19]. The drawbacks of Thury’s constant-current variable-voltage delivery prevented widespread adoption. The Epilogue describes modern high-voltage dc transmission systems.

On December 31, 1879, Edison demonstrated publicly the first practical incandescent electric lights and the first dc power transmission system for incandescent electric lighting (Fig. 2) [20]. The New York Herald reported from Menlo Park, NJ, USA that Edison’s laboratory and some other indoor and outdoor areas were brilliantly illuminated by a total of 53 lights [21].

Fig. 2.

Scenes from Edison’s first public demonstration of incandescent electric lighting on December 31, 1879, shown in The Daily Graphic (New York) of January 3, 1880. Image courtesy of the Library of Congress.

Show All

Edison’s Holborn Viaduct central station in London began operating on April 24, 1882 with a capacity of 2200 16-candlepower lights [22]. The Pearl Street station and distribution system of the Edison Electric Illuminating Co. of New York in lower Manhattan began operation on September 4, 1882. Within a month, it had 59 customers with 1284 lights, and on December 1, 1883, there were 513 customers with 10 297 lights. In 1887, Edison had 103 central stations in the United States, with 311 400 lamps in use [11, pp. 193–195]. It is important for this article that at the end of 1882, Società Generale Italiana di Elettricittà Sistema Edison began construction of a dc central station in Milan, which started operating in June 1883 [23].

The industrial dc motors of Frank Julian Sprague contributed significantly to the profitability of Edison’s power stations for incandescent lighting, because motors operate in daytime. At a meeting on August 12, 1886, the Association of Edison Illuminating Companies decided that, in view of the increasing interest in electric motive power, the Edison Electric Light Co. should issue a circular on the use of electric motors, giving a table of sizes of motors of the Sprague Electric Railway and Motor Company [24]. More about Edison’s central stations in the section on the “battle of the currents.”

A description of the Gaulard–Gibbs ac distribution system was published in Britain in 1883 [25]–​. Gaulard and Gibbs sold the exclusive rights to their inventions to a new company, the National Company for the Distribution of Electricity by Secondary Generators (London). Several Gaulard–Gibbs systems operated in Britain, France, Germany, and Italy in 1883–1886. Open iron core transformers were connected in series on the primary side, lights were connected in parallel to the secondary of each transformer, and the voltage at each lamp was unstable; the candlepower of each light would rise or fall as the number of lights in the system changed. The factory closed in 1887 [11, pp. 257–260].

In 1884, the experimental 3000-V Gaulard–Gibbs ac distribution system from Turin (average elevation 804 ft) to Lanzo (average elevation 1542 ft) in Italy, over a distance of 40 km, was the first one to operate publicly [11, pp. 260–265]. The installation used wire of 4-mm diameter supported on ordinary telegraph poles. The choice of a small town in the Alpine foothills as the final destination proclaimed that in the future it might be feasible to provide electricity to sparsely populated areas, something not possible profitably with Edison’s short-range dc transmission technique.

Ganz & Co. of Budapest introduced an experimental version of the ac distribution system of Károly Zipernowsky, Miksa Déri, and Ottó Titusz Bláthy (ZDB system) at the Budapest National Exposition of 1885 [27]. Unlike the Gaulard–Gibbs system, the ZDB system was commercially practical and was widely implemented [11, pp. 271–278]. The Electrical Department of Ganz & Co. hired Zipernowsky in 1878, Déri in 1882, and Bláthy in 1883 [28]. That chronology explains the order of the letters in ZDB. In 1885, Zipernowsky, Déri, and Bláthy filed patent applications for their transformers in Germany, the United States, and other countries. In 1886, they filed patent applications for the ZDB system in Austria–Hungary, Britain, France, Germany, Spain, and the United States [11, p. 276]. In the ZDB arrangement, Faraday-type (closed iron core) transformers are connected in parallel, lamps are connected in parallel, and the voltage is not significantly affected by load. Fig. 3 shows two types of early ZDB transformers [29].

Fig. 3.

Early single-phase ZDB transformers [29]. (Left) Shell-type transformer. (Right) Core-type transformer. The author added brown color to the copper windings.

Show All

By fall 1886, there were ZDB transmission systems in several European cities [30]. An 1889 brochure of Ganz & Co. listed 71 ZDB systems installed worldwide [29]. In 1892, a Ganz hydroelectric transmission system began sending 4000-V power from Tivoli (Fig. 4) to Rome, a distance of 27 km [31].

Fig. 4.

The exterior of the 4000-V hydroelectric station in Tivoli [31]. The illustration was created by Richard F. Outcault in 1892, then employed by Electrical World. In 1902, he created the popular Buster Brown comic strip series for the New York Herald, considered the first comic strip.

Show All

The Westinghouse Electric Co. was formed in January 1886 [11, pp. 291–294]. Westinghouse purchased the patent rights to the Gaulard–Gibbs system in February of that year [32]. The Gaulard–Gibbs experimental transmission from Turin to Lanzo in 1884, the experimental ZDB system at the Budapest National Exposition of 1885, and the successful introduction of commercial ZDB systems in Europe in 1886, gave Westinghouse the incentive to start developing ac transmission in the United States. The facts on the ground favored Westinghouse in his competition with Edison’s dc systems. More than 20 Westinghouse ac central stations were already in operation by the end of 1887, a respectable number when there were only 103 Edison dc stations in the United States. Edison’s “battle of the currents” could only rely on occasional much-publicized accidents and on unproven assertions about the dangers of ac.

William Stanley, Jr. started his professional career by founding a nickel-plating business, and then he worked for three electric light companies before his association with Westinghouse. In 1884 and part of 1885, Stanley did experimental work at Westinghouse’s Union Switch and Signal Co., and then moved to Great Barrington, MA, USA, where he set up a laboratory in an abandoned rubber factory at the edge of town [11, p. 294]. In spring 1886, working under contract for Westinghouse, Stanley designed and installed the first ac transmission system in the United States. He strung about 1200 m of No. 6 AWG (4.1-mm diameter) copper wire fastened to insulators nailed to trees, from the rubber factory to the center of town (Fig. 5).

Fig. 5.

(Left) Great Barrington in 1884. Pictorial map courtesy of the Norman B. Leventhal Map Center at the Boston Public Library. The author added a red arrow at the location of the abandoned rubber factory on the outskirts of town that would house Stanley’s central station in 1886. (Right) Stanley’s 1886 transformer. Press photo released by the Westinghouse Electric Co. in 1946, in the author’s collection.

Show All

Stanley’s system became operational in March 1886 [33]. He had rejected the unstable Gaulard–Gibbs design purchased by Westinghouse and invented his own single-phase dynamo, transformer (Fig. 6), and distribution system [34]. His system used shell-type transformers of a different design than the ZDB version, connected in parallel and with lights connected in parallel on the secondary, as did the ZDB system. There is not enough information in Stanley’s patents [35]–​[36] and other published material [37], [38] for the author to judge the relative merits of the Stanley and ZDB systems. Stanley’s short-distance system in Great Barrington ceased operating in June 1886. However, the experiment was successful enough to convince Westinghouse to start the construction and installation of a large number of commercial central stations and long-distance transmission systems of that same general design [38].

Fig. 6.

(Left) A two-phase distribution system in Tesla’s patent 382,282 filed December 23, 1887 and issued May 1, 1888. (Right) Earliest Westinghouse “Tesla” two-phase motor and dynamo [55].

Show All

By September 1, 1887, the Westinghouse Electric Co. had installed 24 small ac central stations in the United States and one in Canada. In 1890, more than 300 central station companies worldwide were using Westinghouse systems [11, p. 283]. The first major commercial ac power transmission in the United States took place in 1890, when the Willamette Falls Electric Co. of Portland, OR, USA, transmitted hydroelectric power from the falls of the Willamette River to Portland, a distance of 21 km. Power from 4000-V Westinghouse dynamos passed directly to the line of No. 4 B&S (5.2 mm) wire. The resulting 3300-V power at a receiving substation fed the primaries at the ends of a set of ten 330-V to 110-V reducing transformers in series [39].

British electrical engineer Sebastian Ziani de Ferranti was granted his first patent for a single-phase ac dynamo in 1882 and two patents for single-phase ac distribution in 1886 [40]–​[42]. A company formed in 1882 manufactured the first ac dynamos designed by him [43]. In January 1886, Sir Coutts Lindsay & Co., owner of the Gaulard–Gibbs central station at the Grosvenor Gallery in London, U.K., appointed him engineer, responsible for replacing the Gaulard–Gibbs system by a more practical one. The new central station, with two 2400-V Ferranti dynamos, provided electric power to a small section of London [11, p. 266–269]. Ferranti’s appointment led to lengthy litigation about the validity of his patents [43].

On August 26, 1887, Sir Coutts Lindsay and Co. reorganized and renamed itself as London Electric Supply Corporation, Limited [44]. Under Ferranti’s leadership, the new company built the large Deptford central station, which began operation in 1888, generated 10 000-V ac power, and had enough capacity to supply electricity to all parts of what is now Greater London [45].

Nikola Tesla was born in Smiljan, Austria–Hungary (now in Croatia) on June 28, 1856, local Julian calendar (July 10, 1856 in Gregorian calendar). In 1875–1877, he studied at the Polytechnic School in Graz, Austria–Hungary (now in Austria). In 1880, he briefly audited geometry and physics courses at Karl-Ferdinand University in Prague. In 1882–1884, he worked for the Edison organization in France [46]. He arrived in New York on June 6, 1884 [47]. He worked for the Edison Machine Works in New York for a few months [48]. In December 1884, he and business partners formed the Tesla Electric Light & Manufacturing Co. of Rahway, NJ, USA, maker of arc lights [49]. The arc light business was short lived.

In April 1887, Tesla and some backers formed the Tesla Electric Co. of New York, providing a laboratory for Tesla to develop his polyphase invention [50]. In October–December 1887, Tesla filed his first seven patent applications for polyphase motors, dynamos, and transmission systems, all issued on May 1, 1888 [51]. Fig. 6 shows a page from one of the patents. Two weeks after the patents were issued, Tesla described his invention at a meeting of the American Institute of Electrical Engineers [52]. Tesla sold his polyphase patents to the Westinghouse Electric Co. in July 1888 and began to work for Westinghouse in Pittsburgh. He left the Westinghouse enterprise a year later and returned to New York to begin the research that made him a celebrity [53]. Westinghouse announced two-phase induction (“Tesla”) motors and dynamos almost immediately after acquiring the Tesla patents in 1888 (Fig. 6) [54]. As expected, claims of prior invention sprouted. Sylvanus P. Thompson’s Polyphase Electric Currents, published in 1895, has an excellent chapter on early polyphase development, including descriptions of patents of various inventors [55].

AC transmission is more lossy than dc transmission, but three-phase transmission is about 20% less lossy than single-phase transmission. When transmission distances grew, efficiency became an increasingly important cost factor and three-phase transmission became the norm. The history of polyphase transmission began when visionary German electrical engineer Oskar von Miller, technical director of the 1891 Frankfort Electrical Exhibition, promoted the demonstration of an experimental three-phase transmission system at the exhibition. A hydroelectric power station in Lauffen, 175 km away, delivered 200 kW of three-phase 12 500-V power on a wire just 4 mm in diameter, using equipment provided by the Swiss firm Maschinenfabrik Oerlikon and the German firm Allgemeine Elektricitäts-Gesellschaft (AEG) [56]–​[58]. Swiss-born electrical engineer Charles Eugene Lancelot Brown of Oerlikon and Russian-born electrical engineer Michael von Dolivo-Dobrowolsky of AEG designed the system, which was a three-phase modification of Tesla’s two-phase system [59], [60]. Fig. 7 shows one of the Oerlikon dynamos at the Lauffen station [61]. Moreover, Oskar von Miller simultaneously arranged for the installation of the first commercial polyphase transmission system in the world, from Lauffen to Heilbronn, a distance of about 10 km, using the same type of Oerlikon three-phase generator and other equipment as for the Lauffen–Frankfort transmission [61].

Fig. 7.

A 200-kW 50-V three-phase Oerlikon dynamo at the Lauffen station in 1891. Transformers converted the 50-V power to 12 500 V. The switches and regulating devices were by AEG [61].

Show All

The Lauffen–Frankfort experiment was a daring and successful demonstration of three-phase transmission, which was quickly implemented commercially, as shown in Table 2 [62]. Transmission voltages increased steadily, reaching 220 000 V in 1923 [63]. Many design issues had to be resolved to make 220 000-V transmission possible [64].

Table 2 Some Early Commercial Hydroelectric AC Transmission Systems in the United States and Elsewhere [62]

The Westinghouse Electric & Manufacturing Co. [65] outbid the General Electric Co. for the contract to electrify the 1893 World’s Columbian Exposition in Chicago [11, p. 297] and installed a 1.5-MW 2000-V “Tesla” two-phase transmission system [66]. The year 1893 also marked the beginning of major commercial three-phase transmission systems worldwide. In Sweden, a three-phase 9000-V transmission system made by Allmänna Svenska Elektriska Aktiebolaget (ASEA) began sending hydroelectric power from Hallsjön to the Grängesberg iron mines, a distance of 16 km [67]. The inventions of Jonas Wenström, who was granted a patent for three-phase generators, transformers, and motors in 1890 [68], had led to the formation of ASEA. In the United States, the first commercial polyphase transmission system began operating in California in September 1893. Installed by General Electric and owned by the Redlands Electric Light & Power Co., it transmitted three-phase 2500-V hydroelectric power a distance of 12 km, initially to an ice making company that provided ice to citrus growers [69]. A three-phase clause in the specifications for the project doomed the Westinghouse proposal for a two-phase transmission system [70]. In the early part of 1893, Westinghouse had sued General Electric for infringing on Tesla patents [71] but the ligation ended amicably in 1896 when the two companies signed a patent pooling agreement [72]. By then, Westinghouse was manufacturing three-phase equipment (Table 2) [73].

Moving on, the successful implementation of the giant Niagara Falls power system, which began operating in 1896, was a major validation of three-phase transmission. The most comprehensive source of information about the project is a two-volume history of it by Edward Dean Adams, who had been president of the Cataract Construction Co., the developer of the project [73]. Westinghouse won the initial electrical contract and installed 2200-V two-phase generators and transformers that converted the 2200-V power to three-phase 11 000-V power for transmission to Buffalo, NY, USA, a distance of 42 km. General Electric provided some transformers and other equipment. Output was about 20 MW in 1898. By 1925, output was about 360 MW, approximately 2% of the total power generated by about 4000 power plants in the United States at the time [74]. As intended, industrial power users started factories in the Niagara Falls area almost immediately after the station began operating (Fig 8).

Fig. 8.

Circuits of the Niagara Falls power generation and distribution system at the start of 1897 [73].

Show All

In May 1887, an Edison engineer in New Orleans wrote a plaintive letter to the General Superintendent of the Edison Electric Light Company [75]. Excerpt: “I think it is very much to be regretted that the Edison Company have not taken more active steps to provide local companies with some means of combating Westinghouse in the matter of long distance lighting. Take this station for instance. ⋯ We could get at least 2000 to 3000 more lights in at very profitable rates, in a rich resident district from one to two miles away if we only had an alternating or converter system to supplement our regular three wire system.” The engineer had good reason to be concerned (Fig. 9) [76]–​[78].

Fig. 9.

Number of Edison dc central stations and Westinghouse ac central stations for incandescent lighting in the United States in 1887 and 1890 [2]–​[78].

Show All

Quotes from The New York Times of the 1888–1889 period describe Edison’s strategy against alternating current. March 9, 1888: “At the behest of New York State, Harold P. Brown, an electrical engineer, will conduct tests of alternating-current electricity as a method of execution. Mr. Brown said that it is almost impossible to destroy life using direct current.” The Times of December 6, 1888 had an article with the title “Surer than the rope.” The article reported that a New York State law requiring electrical execution would go into effect on January 1, 1889. The article also stated that the previous summer Brown had performed experiments on two calves at a laboratory of Thomas Edison, assisted by Edison employee A. E. Kennelly, in Edison’s presence, and that “the experiments proved that the alternating current is the most deadly force known to science.” Headline in The Times of July 24, 1889: “Testimony of the Wizard. Edison’s Belief That an Alternating Current of 1,000 Volts Would Surely Kill a Man.” The Times of August 7, 1890 reported that the first execution by electric chair did not go well.

For evidence of the misdirection in Edison’s campaign against ac transmission, we go back to 1886. In May of that year, Società Generale Italiana di Elettricittà Sistema Edison installed a ZDB ac transmission system in Milan [79], which extended the company’s electric power beyond the reach of the company’s previously installed dc system [80]. The general superintendent and chief engineer of the company was John W. Lieb, Edison’s personal representative in Milan, who had been in charge of the installation and initial operation of Edison’s Pearl Street station in New York. In 1888, the electrical utility in Palermo commented about the ZDB system installed in that city in 1887 by Società Generale Italiana di Elettricità Sistema Edison: “We are completely satisfied with the dynamos and transformers of the Zipernowsky-Déri-Bláthy system. The 22.5-kW self-exciting dynamo was put into operation on October 21, 1887 and has since continuously lighted about 520 Edison lamps in the Bellini theater, the Municipal Palace, and a number of the business shops of Palermo, without any parts having been replaced” [29].

In 1922 and 1923, Electrical World reported the implementation of large regional power networks. In Michigan, 20 hydroelectric plants and 11 steam plants joined into a 140 000-V statewide transmission system [81]. A planned expansion of the Pacific Coast Interconnected Transmission System would yield a 1640-MW 1800-mile “superpower” system from the Mexican border to Puget Sound, which would serve four states [82].

In spite of their short reach, dc central stations remained competitive in densely populated areas before the implementation of regional power grids (Table 3). However, as of 1922, ac distribution in large cities was making rapid progress and virtually all dc central station companies in the United States were hard at work on converting to ac [83].

Table 3 DC and AC Central Stations in the United States in 1902, 1907, and 1912 [15, p. 40]

Today, ac power generation and ac distribution at the user end are still the norm, but ac long-distance transmission is being challenged by high-voltage dc (HVDC) transmission [84]. This became possible when efficient high-power high-voltage ac↔ dc converters were developed and incorporated into converter stations at each end of the HVDC cable [85]. The General Electric Co. developed the first such system, an experimental one. Commissioned in 1936, it transmitted 5.25 MW of power at 30 000-V dc from Mechanicsville, NY, USA to Schenectady, NY, USA, a distance of 27 km. It operated as an element of the New York Power and Light Corp. network until it was discontinued around 1945 [85].

HVDC transmission by overhead lines or by underwater cables has a large number of advantages over high-voltage ac transmission [86]. HVDC is particularly advantageous where a transmission line must cross a large waterway, because until recently 50 km was the accepted length limit for underwater high-voltage ac transmission, up to about 100 km today with cables that have insulation of very low shunt capacitance. Beyond about 100 km, HVDC is the only technically viable option for crossing large waterways [86, p. 4]. The first modern commercial HVDC transmission line began operating in 1954, a 98-km-long 100 000-V submarine cable off the coast of Sweden, with a capacity of 20 MW [86, pp. 389–390]. Since then, many HVDC transmission lines have been installed worldwide [86, ch. 12]. For example, the Trans Bay Cable in California (Fig. 10), completed in 2010, began sending HVDC power from a 230-kV ac substation in Pittsburg, CA, USA to a 115-kV ac substation in San Francisco, CA, USA, a distance of about 90 km [87].

Fig. 10.

The Trans Bay Cable in California. The author created this graph using information in [87]..

Show All