march/april 2004 IEEE power & energy magazine 63

TTHIS ARTICLE DESCRIBES THEfirst “polyphase” (more than onephase) system developed for the distri-bution of alternating current (ac)power. This two-phase system wassubsequently rendered obsolete, how-ever, by the superior three-phase sys-tem that is now universally usedthroughout the world.

The Origins ofTwo-Phase PowerToday, the large-scale generation, trans-mission, and distribution of electricpower is by means of the three-phaseac system; that is, three individual sin-gle-phase voltages and currents havinga 120˚ phase relationship to each otherand intermingled on three wires(excluding a neutral). The three-phasesystem has been adopted because itprovides for a constant rather than pul-sating power flow to motors, andbecause it is an efficient system as faras the amount of copper required perkilowatt transmitted. The theoreticalcomplexity of the three-phase system,however, delayed its complete accept-ance in the early days of electric powersystem development.

During the early 1890s, understand-ing the behavior of simple single-phaseac was enough of a challenge. It wasnot until Charles P. Steinmetz, the leg-endary General Electric scientist,developed the concept of the use of the“j” operator (unity magnitude at a 90˚phase angle) and complex numbers forac circuit calculations that the behaviorof voltages and currents in ac circuits

and machines was truly understand-able. Likewise, it was not until theintroduction of what eventually cameto be known as “symmetrical compo-nents,” during the early 20th century,that the calculation of three-phase volt-ages and currents became relativelystraightforward. This technique utilizedan “a” operator that was of unity mag-nitude at a 120˚ phase angle (–0.5 +j0.866). This operator was of signifi-cant value since, in a balanced three-phase system, the voltages and currentsare at 120˚ phase relationships to eachother.

Symmetrical components actuallyfacilitated calculations in unbalancedthree-phase circuits. They were origi-nally known as “Fortescue compo-nents” since the method was introducedin 1918 by Charles L. Fortescue of theWestinghouse Electric Corporation.Significant additional work in this areawas later contributed by Edith L.Clarke of the General Electric Compa-ny. During the late 19th century, how-ever, this calculation tool did not exist,and the fact that changes in voltage orcurrent magnitudes in one phase of athree-phase system affected the volt-ages and currents in the other two phas-es contributed to the difficulty inunderstanding three-phase circuits.

Thus, the first ventures into therealm of polyphase electric power usedonly two alternating current phasesrather than three. The two phases weregenerated with a 90˚ phase differencebetween them, and the system thatresulted was called two-phase power.

In fact, the first two-phase generatorsemployed during the early 1890s weremerely two single-phase machines cou-pled together with their rotors carefullyset relative to each other so as toachieve the required quadrature phaserelationship. Each generator, then, real-ly fed a separate two-wire, single-phasecircuit. Since the two phases were com-pletely electrically isolated from eachother, there were no interactionsbetween voltage and current magni-tudes in one phase with those quantitiesin the other phase. Therefore, from atheoretical standpoint, the two-phasesystem was more easily understoodthan was the three-phase system.

The two phases were used togetherin a four-wire system to enable theoperation of the new Tesla (or induc-tion) motor that had been developedby Nikola Tesla. In order to be self-starting, the Tesla motor requiredsome form of rotating magnetic fieldthat had to be produced by apolyphase type of supply. The two-phase system was adequate for thispurpose. The Westinghouse ElectricCorporation supplied the power plantand lighting for the Colombian Expo-sition in Chicago in 1893. Two-phasepower, produced by pairs of coupledsingle-phase generators, was usedthroughout this installation.

Two-Phase Power at Niagara FallsThe experience gained with the useof two-phase power at the ColombianExposition may have had some

Thomas J. Blalock

hist

ory

the first polyphase systema look back at two-phase power for ac distribution

ISSN 1540-7977/04/$17.00©2004 IEEE

64 IEEE power & energy magazine march/april 2004

influence on the decision by Westing-house to employ a two-phase generatordesign for the first ac powerhouse atNiagara Falls, which went into opera-tion in 1895. The generators used atNiagara Falls were of a more conven-tional design, being single machineshaving two interleaved windings ratherthan two distinct machines coupledtogether. These generators operated ata frequency of 25 cycles (25 Hz) sinceit was expected that a significant por-tion of the power produced would beused to operate rotary converters so asto obtain direct current (dc) for indus-trial uses such as aluminum produc-tion. These early rotary convertersrequired a low frequency for satisfac-tory operation.

There was obviously still a mistrustof the practicality of three-phasepower throughout the electric powerindustry at that time. For example,according to an 1896 article titled“Present Status of the Transmissionand Distribution of Electrical Energy”in the AIEE Transactions:

Where a two-phase transmis-sion with separate circuits isused, then if the separate circuits

are wound on different arma-tures, each can be regulated togive a constant voltage at thereceiving end. This is the case,for instance, in the largedynamos built by the Westing-house Company for use at theWorld’s Fair in Chicago. The dif-ficulty due to the uneven loadingof the circuits is speciallymarked in the case of the three-phase system, and it is one of theprincipal objections that havebeen urged against the employ-ment of this system for purposesof distribution.

It had already been realized, how-ever, that the three-phase configurationwas superior for transmission from thepoint of view of efficiency. Thus, spe-cial phase-changing transformers weredesigned by Charles F. Scott of West-inghouse in order to step up the two-phase generated voltage at NiagaraFalls to 11,000-V, three-phase fortransmission to Buffalo, New York.The General Electric Company wasawarded the contract to build thephase-changing transformers and sowas licensed by Westinghouse to uti-

lize the connection developed by Scottfor this purpose.

At Buffalo, some of this three-phasepower was used for rotary convertersthat supplied 110/220-V dc power forthe Edison distribution system down-town. However, some of the receivedpower was converted back into two-phase power for general lighting pur-poses in outlying areas. Motor-generator sets were used for this latterconversion because the frequency ofthe ac power was increased as well inorder to avoid undesirable flickering ofincandescent lamps. The frequencyused was actually 62.5 cycles, ratherthan 60 cycles, so as to simplify thedesign of these frequency changers.The conversion back to two-phasepower was motivated by the conviction,at that time, that satisfactory voltageregulation was more easily achieved inthe two separate phases of a two-phasesystem than in a three-phase system.

This belief in the superiority of two-phase systems with respect to voltageregulation led to the extended use oftwo-phase distribution in many locales.For example, in Cohoes, New York,(north of Albany) a 1915 hydroelectricstation was designed to generate three-phase power. However, some of thatpower was converted to two-phaseusing “Scott” type transformers inorder to supply an extensive network ofexisting two-phase feeders for lighting,rather than change those feeders tothree-phase operation.

William StanleyAdopts Two-PhaseWilliam Stanley, the man creditedwith the first practical application ofthe ac system using transformers (inGreat Barrington, Massachusetts, in1886), subsequently formed the Stan-ley Electric Manufacturing Companyin Pittsfield, Massachusetts, in 1891.Stanley adhered to the design and con-struction of two-phase generators andmotors throughout the 1890s. Thiswas only partly a result of his belief inthe superiority of the two-phase sys-tem for voltage regulation purposes.Another factor had to do with the

Original two-phase to three-phase transformers installed at Niagara Falls in1895 (photo courtesy Hall of Electrical History at the Schenectady Museum,Schenectady, New York).

march/april 2004 IEEE power & energy magazine 65

increasing development of three-phaseequipment by his major competitors,General Electric and Westinghouse,during the 1890s. Stanley’s decision tomanufacture two-phase equipmentallowed him to avoid excessive patentinfringement problems with his com-petitors. Regardless of the reasons,however, Stanley contributed to theperpetuation of the use of two-phasepower in many locations.

The Stanley Works itself generatedand utilized two-phase power. In 1907,this plant became the Pittsfield Worksof the General Electric Company, andthe two-phase power system that it hadinherited from Stanley remained in useuntil the closing of the facility in 1987.In fact, to this day, there is still one ele-vator in an old office building thereoperating with a two-phase motor.

The two-phase system in this plantwas somewhat unusual in that it was athree-wire system. One wire from eachphase was combined into what wascalled a “common” wire (not a “neu-tral”). The advantage in this was theability to use more commonly availablethree-pole circuit breakers and switches.A disadvantage, however, was that evenwith the two phases balanced, the com-mon wire carried 1.414 times the currentin the other two phase wires. Thus,economy in pulling circuits throughconduits required the use of two differ-ent sized cables. Eventually, the planthad two power distribution systems, theoriginal two-phase system and a newerthree-phase system. The two systemswere interconnected by means of phase-changing transformers. These were of adesign by Louis F. Blume of the GeneralElectric Company and utilized a wind-ing configuration differing from the“Scott” connection, presumably to avoidpatent conflicts with the WestinghouseElectric Corporation.

Since Stanley supplied equipmentto the local municipal power compa-ny, the Pittsfield Electric Company,downtown Pittsfield was also servedby a two-phase system. This, howev-er, was the more conventional four-wire type of two-phase distributionrequiring four-pole service switches.

This two-phase distribution systemremained in use until the middle ofthe last century, and vestiges of it inthe form of four-pole switches couldstill be found on the service switch-board of at least one old building inPittsfield in the early 1980s. Also,two-phase motors were still beingused to drive the elevator motor-gen-erator sets in Pittsfield’s only depart-ment store when it closed in 1988.

Other Two-PhaseInstallationsIn the village of Middle Falls, NewYork, (northeast of Albany) the NiagaraMohawk Power Corporation operated a1900 vintage, 350-kW Stanley two-phase generator in a small hydroelec-tric power station there until 1987.Another identical unit had been retiredin 1976. The output of the station wascoupled to Niagara Mohawk’s three-phase grid by means of phase-changingtransformers.

The generation of two-phase powerwas not exclusively an East Coast phe-nomenon, however. In 1898, the PacificLight and Power Company installedfour 300-kW Westinghouse two-phasegenerators in a hydroelectric stationlocated in San Gabriel Canyon, nearLos Angeles, California. This stationserved the nearby town of Azusa.

As the use of ac motors expandedduring the early 20th century, the prob-lem of providing both l15 V for light-ing and 230 V for motor use fromtwo-phase distribution systems becamesignificant. One solution was the adop-tion of a two-phase, five-wire system inwhich center taps on both phases wereconnected together to create a neutral.This, then, resulted in a “star” configu-ration (analogous to the three-phase“wye” connection) and, technically,was a four-phase system. As such, 115V (single-phase) for lighting was avail-able from any of the four phase wiresto the neutral, while 230 V (two-phase)was available for motors from the fourphase wires themselves.

In New York City, the Bronx Dis-trict of the New York Edison Compa-ny adopted this form of secondary

distribution around 1925. At that time,the Company was interested inupgrading its existing 2,400-V, two-phase primary distribution system to13,200 V, three-phase. The connectedtwo-phase motor load, however, wastoo great to consider changing thesecondary distribution system fromtwo-phase to three-phase as well, so“T”-connected (Scott) phase-chang-ing transformer banks were installedto supply a two-phase, five-wire sec-ondary distribution system.

During this era, the use of the three-phase, four-wire wye-connected distri-bution system was often considered tobe unacceptable because of the non-standard voltage (199 V) between phas-es with 115 V available from phase toneutral. Early induction motors,designed for operation at 230 V, wereless satisfactory when operated onlower voltages than are inductionmotors of today. The ability of the two-phase, five-wire distribution system tosupply the standard voltages of115/230 V was a main feature in alengthy article published in the AIEETransactions in 1925 by an engineerassociated with the Philadelphia Elec-tric Company in Pennsylvania. Thisarticle justified the continued use ofthat system.

William Stanley’s company special-ized in two-phase equipment.

The Demise ofTwo-Phase SystemsEventually, a hybrid type of three-phasedistribution system, which was knownas a three-phase, four-wire, “delta” sys-tem, came into use in certain regions ofthe United States. This system includeda center tap on one phase of a bank ofdelta-connected transformers supplying230 V. The center tap formed a neutraland, in conjunction with the two phasewires of that particular phase, was usedto supply 115/230 V services on a sin-gle-phase, three-wire basis. Motorsoperating at 230 V were supplied fromthe three phase wires of this type ofservice connection.

Buildings requiring both motor andlighting service were sometimes pro-vided with two separate services, a sin-gle-phase, three-wire service forlighting and a three-phase, three-wireservice for motors. Otherwise, a singlefour-wire service was brought into abuilding, but care had to be exercisedby electricians so as not to use the oddphase wire along with the neutral tosupply lighting loads. This odd phasewas referred to as the “high phase” or“wild phase” because considerably

more than 115 V existed between itand the neutral. This complicationassociated with the four-wire deltatype of service led to its gradual aban-donment during the latter 20th centurybecause fewer and fewer practicingelectricians were able to truly under-stand it. Also, by that time, inductionmotors had been developed that oper-ated satisfactorily on voltages lowerthan 230 V. As a result, the three-phase, wye-connected service, giving208 V between phases and 120 V fromphase to neutral, has become the stan-dard commercial type of service. Also,over the years, old two-phase primarydistribution systems were graduallyreplaced with three-phase systems. Acommon practice became the conver-sion of a 2,300-V, two-phase, four-wire distribution system into a4,000/2,300-V three-phase, four-wiresystem (with neutral).

Several clever and complex planswere devised for the temporary supplyof remaining two-phase loads from anew three-phase system, without theexpense of purchasing special phase-changing transformers. One such tech-nique took advantage of the fact thatthere is a 90˚ phase relationshipbetween one phase-to-phase voltageand the voltage from the third phase toneutral in a three-phase, four-wire sys-tem. Customers were encouraged topurchase three-phase motors, ratherthan add to their existing inventory oftwo-phase motors. Many of the oldmotors, however, lasted for quite sometime. Occasionally, a customer actuallyhad to be supplied with two services,one two-phase and one three-phase.

With rare exception today, the two-phase distribution system has become athing of the past. Its extensive usethroughout the 20th century, however,created interesting situations for electri-cal engineers accustomed to three-phase systems. Occasional oversights,resulting from the unrecognized needfor four-pole motor control contactorsdue to the existence of an old two-phasesystem, have been known to cause

havoc for electrical equipment design-ers and suppliers.

For Further ReadingJ.O. Kraehenbuehl and M.A. Faucett,Circuits and Machines in ElectricalEngineering. New York: Wiley, p. 268,1939.

Electrical Transmission and Distrib-ution Reference Book (4th ed.). EastPittsburgh, PA: Westinghouse ElectricCorporation, p. 12, 1950.

E.L. Clarke, “Determination of volt-ages and currents during unbalancedfaults,” General Electric Rev., pp.511–513, Nov. 1937.

C. Passer, The Electrical Manufac-turers (1875–1900). Cambridge, MA:Harvard, 1953.

L.B. Stillwell, “The electric trans-mission of power from Niagara Falls,”AIEE Trans., pp. 444-486, 23 Aug.1901.

“Present status of the transmissionand distribution of electrical energy,”AIEE Trans., vol. XIII, Sept. 1896.

H.G. Stott, “The distribution andconversion of received currents,” AIEETrans., pp. 125-163, 22 Mar. 1901.

B.R. Connell, “The Hydro-ElectricDevelopment of the Cohoes Companyat Cohoes, N.Y.,” General Electric Rev.,pp. 340–352, May 1915.

L.F. Blume, “Transformer connec-tions for three-phase to two-phasetransformation,” General Electric Rev.pp. 552–559, Sept. 1912.

W.A. Myers, Iron Men and CopperWires: A Centennial History of theSouthern California Edison Company.Glendale, CA: Trans-Anglo Books,1983.

“Distribution for congested areas,”Electr. World, pp. 1031-1032, 16 May1925.

P.H. Chase, “Two-phase, five-wiredistribution,” AIEE Trans., pp.737–749, June 1925.

“Changing from two-phase four-wire to three-phase four-wire distri-bution,” Electric J., pp. 214–216,June 1923.

66 IEEE power & energy magazine march/april 2004

A two-phase, four-pole service switchin a building in Pittsfield, Massachu-setts (Tom Blalock photo). p&e