Design and Construction of the IRT: Electrical Engineering (Kimmelman)

Design and Construction of the IRT: Electrical Engineering

Barbara Kimmelman
Survey Number HAER NY-122, pp. 283-264

Historic American Engineering Record
National Park Service
Department of the Interior
Washington, DC. 20240

The records in HAER were created for the U.S. Government and are considered to be in the public domain. It is understood that access to this material rests on the condition that should any of it be used in any form or by any means, the author of such material and the Historic American Engineering Record of the Heritage Conservation and Recreation Service at all times be given proper credit. For information on HAER, visit Built In America: Historic American Buildings Survey and the Historic American Engineering Record, 1933-Present, Library of Congress American Memory Project.

Introduction

[page 284] The New York subway was planned as a rapid transit road. The Commissioners and engineers responsible for its construction meant something quite specific by rapid transit. They meant, first, high speeds. When Alexander E. Orr, President of the Board of Rapid Transit Railroad Commissioners, was asked in 1896 to define rapid transit, he replied that his aim was to provide New York with speeds of 20-25 mph in the business district and 40-45 mph uptown.¹ One purpose of the road would be to "bring the extreme limits of the City into closer relations,"² and only much higher speeds than those offered by existing transit could realize this goal.

A corollary of the need for high speeds was a right-of-way for the exclusive use of the rapid transit road.³ Speed was both dangerous and impossible in the congested business areas, except on a road above or below the street which avoided the hazards and delays of surface operation.

In order to relieve congestion, high-speed cars had to run at frequent intervals. But the speed and frequency of traffic presented problems of passenger safety and comfort, which had to be solved if the road were to be operated successfully.

New York's Rapid Transit Commissioners considered the problem of an appropriate motive power in relation to these requirements - high speed, independent right-of-way, heavy traffic, passenger safety, and comfort. Numerous types of motive power, principally steam, cable, electricity, and compressed air, were at this time either in use or proposed for rapid transit systems throughout the world. Each offered technical advantages, each had its own peculiar drawbacks.⁴ The Commissioners concluded that their requirements demanded electricity. Electric traction was chosen for the New York subway not on the basis of isolated technical considerations, but as an integral part of the projected system of rapid transit as defined by the Commission.

This approach was not widely evident in the decade 1890-1900 when New York's succession of Rapid Transit Commissions gave shape to the future subway system. Rapid transit was seen by some as any type of organized transportation within or around an urban area. In 1891 the Chief Engineer of the Boston Rapid Transit Committee toured European cities to examine rapid transit systems. His report included electric street railways of various types, as well as cable and steam roads. The distinction between surface and rapid transit, crucial to the New York Commissioners, was not important.⁵

Electricity as a motive power often involved the introduction of electricity on existing transit originally powered by different means. This process of electrification undermined the "systems approach" to [page 285] motive power choice, stressing instead a commercially competitive comparison of the technical merits of idealized systems.⁶ A common way to classify various transit schemes was by reference to motive power. In 1899 Cassier's Magazine devoted an issue to the electric railway. Articles written by experts in their respective fields considered questions ranging from the development of specific types of equipment to analysis of entire systems.⁷ Overhead trolleys, underground conduits, surface, elevated, and tunneled urban roads, as well as interurban and heavy trunk lines, were discussed. These systems answered different needs, and required different modifications of electrical and auxiliary technologies. The common denominator was electric traction.

The Board of Rapid Transit Railroad Commissioners did not view their problem in this manner. They were not concerned with transforming an existing road to electric power. All engineering aspects of the enterprise were designed "from scratch", giving the Board ample opportunity to realize its systems approach. The fact that in 1891 a tunneled road, rather than an elevated, was under consideration for the new system immediately narrowed options as to motive power. The Board was adamant that steam should not be used on the road, so that the New York subway be spared the hazards and discomfort of smoke-filled tunnels. Electricity and compressed air were the only options considered seriously.⁸ A consulting engineer reported to the Commission in 1891 that the use of electricity would do much to dispel the riding public's prejudice against tunnel transit, and noted recent precedents for this type of system.⁹ European transit systems which might serve as models for New York justified a trip abroad by the Commission's Chief Engineer, William Barclay Parsons, in 1894.¹⁰ Unlike that of his Boston counterpart three years earlier, his report included no reference to surface electric railways. He examined only elevated and tunneled roads, an emphasis suited to the Commission's conception of rapid transit. Electric, steam, and cable operation were all represented.¹¹ Parsons believed the designers of New York's future rapid transit road would learn more, including what not to do, from steam or cable operated tunnel systems than from electric trolley or conduit street practice.

Parsons was distressed by conditions in the tunnel roads powered by steam. He found the air in the tunnels of London s Metropolitan and Metropolitan District Railways "extremely offensive" because of exhaust gases, although the recirculation of steam by condensing locomotives kept the air drier than would otherwise have been expected. The Glasgow City and District Railway, begun in 1888, and the Paris tunnel of the Chemin de Fer de Sceaux, constructed in 1894, were both to be steam operated. Parsons was concerned with the great attention to ventilation demanded by both roads. Parsons suggested another motive power might have been chosen by the Paris company, except that the tunnel was to serve a branch of a steam line and uniformity of service was desired.¹²

[page 286] Parsons was most interested in electric operation, whether on elevated or tunneled lines. He discussed the operation and the power plants of the City and South London Railway, which had electric locomotives; of the Liverpool Overhead Railway, with motor cars linked in two-car trains with current supplied by steel conduit laid between the rails; of the Waterloo and City Railway, under construction in 1894, the specifics of its electric operation still undecided.¹³ Parsons also took note of two electrically powered American roads, the elevated Intramural Railway built for the Chicago Columbian Exposition, and the seven-mile Baltimore Belt Tunnel of the Baltimore and Ohio Railroad.¹⁴ He persistently found the "important feature," "the most interesting part," "the great feature" of these roads to be the electric operation.¹⁵

Parsons did not embrace electric traction without qualification, nor did he attempt a simplistic evaluation of electric operation. "It cannot be assumed," he wrote, "that electricity has some mysterious properties which render it vastly superior and more economical than steam as a motive power. This idea is fallacious in the extreme."¹⁶ He recognized that "with all things being equal" steam would be more economical than electricity; steam drives a locomotive directly, while a dynamo installation, at great first cost, also required coal and water, plus provided opportunity for power loss in the electric equipment in addition to losses in the boiler and engine.

Parsons knew all things were not equal. Locomotives required the best grades of coal, and conditions on board prevented the most efficient operation of the boiler. A stationary power plant allowed the introduction of equipment, such as steam condensers and fuel water heaters, which greatly increased boiler efficiency. Parsons compared three electric roads, the Liverpool Overhead, the City and South London, and the Chicago Intramural, to New York's steam roads, the Manhattan Company and the Brooklyn Rapid Transit elevated lines. The three electric roads showed savings in coal consumed and also in coal cost per ton, since poorer grades of fuel could be burned efficiently in the stationary plants. Parsons considered that this "very striking economy in favor of electricity" more than compensated for the greater power losses inherent in the electrical system. Parsons also noted that cost of locomotive repairs per ton per mile were less on the City and South London than on the Manhattan elevated.¹⁷ Although the report simply presented findings without recommendations, Parsons was won over to electric operation. He told the New York press upon his return that the projected subway, with heavy trains running frequently required electric motive power.¹⁸

In December 1894, a few months after Parsons presented his report, the Board of Rapid Transit Commissioners submitted his plans and estimates for the new road to a committee of experts comprised by Abram S. Hewitt, Octave Chanute, Thomas C. Clarke, William H. Burr, and Charles Sooysmith. The committee members recommended electricity as motive power. They expected electric motors, with their quicker acceleration, to attain higher overall speeds than steam locomotives, thus increasing traffic capacity.¹⁹

[page 287] Consideration of numerous factors, including motor performance, boiler efficiency, maintenance costs, and tunnel atmosphere were involved in the Commission's evaluation of motive power.²⁰ Although other options were not ruled out, electricity remained the unofficial choice. The official decision was not made until after the contract for the road was awarded to John B. McDonald in 1900. Although the contract did not specify motive power, it restricted the contractor to forms not requiring combustion. McDonald soon petitioned to use electricity rather than compressed air, and the Board approved the choice.²¹

During the mid-l890's the relative merits and applicability of steam versus electricity were widely debated, whether for heavier trunk lines or for street railways.²² The systems approach of Parsons and the Commission made this debate irrelevant to the subway system. Streetcar and trunk line roads differed widely from each other and from underground lines in purpose and operating conditions. Electricity might indeed have been the appropriate motive power for each, but for different reasons, and its application to each type of road required solution of different problems.²³

The system of motive power and auxiliary technology worked out by the subway engineers was innovative primarily in that it incorporated within one system techniques and practices from other established systems. The electrical system included a single central generating station producing 3-phase, 25-cycle current at 11,000 volts; distribution to sub-stations where voltage was stepped down and current was transformed to direct current; then distribution via cables to a third rail. Power from the third rail was conducted to the car motor by a collecting shoe.

Subway practice therefore took advantage of advances in electric light and power generation and transmission, and very recent practices on other elevated and tunnel roads. The frequency of traffic, and the complexity of track arrangements at certain locations demanded, for greater safety, a system of automatic block signalling and interlocking switching which borrowed heavily from standard steam railroad practice. Application of this system to electric roads gave the subway engineers their most interesting opportunity for innovation. Safe operation also required rolling stock of sturdy framing and fireproof construction. The design of all-steel cars for the subway was the innovative response to this problem. The discussion below, and the technical sections which follow it, will attempt to demonstrate the limited technical contributions available to the subway from electric street railway practice, and the extent to which subway engineers drew upon the experience and extant steam trunk lines and recently constructed rapid transit roads.

The advances in electric traction technology between 1885 and 1900 attracted the attention of rapid transit experts. Uncertainty and controversy surrounded the commercial introduction of electric traction on street and elevated railways in New York. And street railway practice had only limited applicability to rapid transit roads.

[page 288] New York's surface and elevated lines were the first to experiment with electric traction. The modern successes of pioneers like Charles J. Van DePeople, Leo Daft and Stephen Field, did not immediately win the support of New York's traction men and financiers.²⁴ Before the application of electricity for traction would be commercially feasible, difficult problems required solutions. These included the development of a satisfactory dynamo, transmission of current to the motors from the power source, motor wear-and-tear under the severe conditions of varying speeds and start and start-stop motion, and appropriate mounting and gearing of motors.

The early inventors sought answers, but not until Frank J. Sprague's 1888 installation in Richmond, Virginia, was a commercially successful system devised which incorporated solutions to all these problems.²⁵ Sprague's equipment included single reduction gearing, and a combination of spring and axle support for the motor which was later widely adopted.²⁶ Current was transmitted to the motor via overhead trolley and poles.

Sprague's Richmond installation marked a turning point in attitudes toward electric traction. The 1888 meetings of the American Street Railway Association devoted proceedings to the problem of street railway motive power, and hailed electricity as the motive power of the future.²⁷ But as late as 1894, electric traction was still in New York's future. Other cities and towns had "electrified" eagerly; by 1891 over 200 electric railways had been installed in the United States.²⁸ Amid this rapidly expanding industry, New York's hesitation is noteworthy.

In 1894 the directors of the Metropolitan Street Railway Company decided to install electric service on its Lenox Avenue Line.²⁹ The company viewed the line as experimental; the conduit was constructed to accommodate cable if the electrical service was discovered to be inadequate. By spring of 1895, the General Electric underground trolley system was installed and tested.³⁰ Noting the economies of electric operation and feeling that passengers were pleased with the service, the company planned extensive conversion of much of its system to electricity by early 1897. It was expected that by the end of that year over 40 miles of the Metropolitan's tracks would be served by electricity.³¹

The status of electric traction in New York was at this time by no means certain. The success of the Lenox road did not prevent the Metropolitan company from considering compressed air as well as electricity for its 1897 improvements. At the completion of the electric installation, Metropolitan had horses, cable, and electricity powering its cars.³² And the Manhattan Railway Company continued to wait, content with steam-powered locomotives on the elevated lines.

Despite the company's recognition of the superiority of electricity, the persistence of multi-powered operation on the street railways underlines the ways in which street railway and rapid transit service had diverged. The [page 289] electric cars of the Metropolitan company operated within a system on which cable and horses served, if not optimally, at least acceptably from the company's standpoint. No trains were operated; all cars ran as single units. While an electric car could move faster than a horse on unobstructed track, the congestion of New York s streets erased this advantage in areas where fast service was most desired.³³ Rapid transit systems with exclusive rights-of-way avoided this problem. On these systems, cars were often hooked together as trains, pulled by an electric locomotive analogous to the steam locomotive, operating at speeds and conditions unfamiliar to the electric street railway.³⁴

In 1897 a crucial innovation in electric traction brought the two classes of electric transportation closer together in one important area: motors. Frank J. Sprague demonstrated his system of multiple unit motor control on the Chicago els, enabling a train to operate electrically without a locomotive, using motors mounted on each car synchronized by controller circuits under command of a motor controller.³⁵ Advances in motor design would now be equally applicable to individual street cars or rapid transit trains. In 1897 William Fransioli, engineer of the Metropolitan Street Railway Company, traveled to Chicago to evaluate the motor performance on the els.³⁶ His trip underscores the importance of multiple unit control in allowing an exchange of technology between street and elevated practice.

Multiple unit control spurred the adoption of electric traction for rapid transit. The equal distribution of weight on the driving wheels throughout the entire train, in contrast to the concentration of weight on the locomotive, made this system suitable for use on elevated structures. Interest finally awakened in New York. Shortly after the initial success in Chicago, the President of the Brooklyn Union Elevated Railroad advertised for bids to transform his lines to electric power, and he expressed a preference for Sprague's multiple unit operation.³⁷

There were numerous methods of conducting electricity to car motors. Street railways relied on overhead trolleys and underground conduits of various designs. During the last two years of the nineteenth century the elevated railways, unfettered by crowded street conditions, began to adopt third rail conduction. Upon his return from the General Electric plant of Schenectady in 1897, Chief Engineer Cornell of the Brooklyn Elevated was favorably impressed by the third rail he had observed in conjunction with Sprague's new multiple unit system. He felt it would be appropriate for the Brooklyn Bridge crossing; a year later Brooklyn Rapid Transit chose a third rail for its Brooklyn Bridge franchise.³⁸ Third rails were installed on Brooklyn and Boston els, and were used on interurban electric roads as well.³⁹

The flurry of activity in electric rapid transit after 1897 finally aroused the interest of the management of the Manhattan Railway Company. Since the early 1890's, the Westinghouse and General Electric companies had sought in vain to win the important Manhattan contract.⁴¹ Before 1897 the company's hesitance was perhaps justified on technical grounds. But [page 290] once multiple unit control and third rail conduction had been tested and found satisfactory in Chicago, Boston, Brooklyn, and other major cities, technical arguments against electricity lost much of their force.⁴²

In June 1898 George J. Gould, President of the Manhattan Railway Company, at last agreed that electric operation would increase the company's earning power, and stem the flood of accusations levelled by New Yorkers against the slow speeds, dirt, smoke, and general nuisance of his road. By early 1899, Gould had not made up his mind concerning the relative merits of compressed air and electricity. In November of that year, however, he finally awarded to Westinghouse the big contract for supplying the electrical generators and equipment for power house and substations. General Electric later received the contract for the motor equipment of the cars.⁴³

The Manhattan company insisted that its earlier actions (or, rather, inaction) reflected due caution, and wariness of untried methods. As soon as others had demonstrated the feasibility of such methods, the Manhattan elevated was quick to follow.⁴⁴ Contemporaries suspected the motives of the Manhattan directorship. The Times said that observers in Philadelphia attributed Gould's actions to the expectation of subway construction in the near future, and that he hoped his improvements would weaken support for the projected rapid transit tunnel. This, coupled with a drop both in revenues and the number of riders which the els had experienced since 1896, could indeed have pushed Gould toward modernization.⁴⁵

The electrification of the Manhattan Railway Company between 1899 and 1902 was important to the rapid transit subway. The Manhattan's conversion paralleled the early years of subway construction. During the planning of the subway system, the Manhattan elevated road combined the most up-to-date practice with close proximity to the subway engineers. Parsons traveled to Europe twice after the tunnel construction began, and he and his staff visited other American cities,⁴⁶ but the Manhattan line provided them with a working model of a rapid transit system in their own backyard.

The role of the Manhattan as a model for the subway was more than a technical artifact to be observed and copied. The Manhattan lines served as a practical working model for the subway. There were similarities in the generating and distribution systems, because the same men drew from their immediate experience with the els and applied it to the subway.⁴⁷ The electrification of the elevated drew prominent electrical engineers to New York. Some taught the subway engineers; E. P. Bryan, with a background in steam railroading, relied heavily on Alfred Skitt of the Manhattan company for advice on electrical matters.⁴⁸ Lewis B. Stillwell, coming to the Manhattan company direct from his work with high-tension alternating current for the Niagara Power Company, was consultant to the elevated on the design and installation of electrical equipment. In 1902, he became electrical director of the Interborough Rapid Transit Subway Construction Company. He therefore superintended the electrical work for both the elevated and tunneled roads.⁴⁹

[page 291] Stillwell also brought part of his staff from one company to the other. H. N. Latey worked under Stillwell in the electrical department of the Interborough company, and from 1898 to 1901 he had served as assistant on electrification of the Manhattan elevated. W. C. Phelps, assistant engineer in the Mechanical Department of the Subway Construction Company in charge of structural design of the main and sub power stations, had worked on preparations for the Manhattan electrification.⁵⁰

As the tunnel neared completion, the Manhattan lines, New York's closest approximation to the subway's electrical technology, were used to test rolling stock and motors.⁵¹ The identity between the systems was of course not complete. For example, the Manhattan did not adopt the switching and signaling system desired by the subway engineers, and it was the Boston els which helped with this problem.⁵² But the similarities between New York's elevated and subway systems were an important factor in New York's rapid transit situation. No technical incongruence stood in the way of consolidation and integration of the two systems in 1903, when the Interborough Rapid Transit Company took over the operation of the Manhattan line.⁵³

If the electrification of the Manhattan elevated lines brought the most modern electrical equipment to the attention of subway engineers, the Board of Rapid Transit Railroad Commissioners' attitude assured that the subway would take advantage of the most advanced developments in electric technology. The original contract, as noted above, contained no precise power specifications. The Commissioners felt that "development and progress in this field is so rapid that it has always seemed to the Board the proper course to defer decision in respect to all details to the last moment, fixed by the necessity of beginning the operation of trains by a certain time."⁵⁴ This remarkable approach, which recognized that "the whole subject of electric traction is comparatively new," allowed the subway engineers to take full advantage of the changes and improvements worked out by the manufacturing companies competing for the traction market.

The relationship between the electrical manufacturing companies and the New York traction companies is an interesting one.⁵⁵ It has already been noted that manufacturers solicited the attention of the traction interests to the merits of their various machines and systems. The problem of equipping roads as large as those in New York spurred technological progress through interaction between the transit and manufacturing companies.

The electrification of New York's surface roads, els, and subway demanded installations of unusually significant size and nature. The Metropolitan Street Railway plant was described in 1901 a the largest polyphase alternating current railway plant in operation.⁵⁷ The plant of the Manhattan Railway Company, under construction that same year, was expected to be "by far the largest steam-driven electric generating plant in the world." The 59th Street power plant of the Interborough company was larger still.⁵⁸

[page 292] The contracts awarded to the manufacturers were correspondingly large. When General Electric received the order for the Manhattan's motor equipment, it was the largest such contract ever let in the United States.⁵⁹ Allis-Chalmers considered its elevated and subway contracts so important that the company issued a pamphlet advertising its role in equipping of the new systems.⁶⁰ The contracts between the Interborough and the Union Switch and Signal Company were again the largest of the kind awarded to that time.⁶¹

The manufacturers responded to the needs and demands of the transit system with modifications and innovation. Westinghouse required a specially constructed steel boring machine for the manufacture of the Manhattan company's generating equipment, and both General Electric and Westinghouse modified standard motors to meet the specifications of the Interborough engineers.⁶² Most significant were the signaling system and rolling stock for the Interborough lines; each involved the commercial introduction of large scale innovative designs.

More than one manufacturing company was at work on the problem of applying automatic block signals on electric roads; the Boston rapid transit system had found a solution slightly different from New York's. The Pennsylvania Railroad was interested in the all-steel rolling stock designed for the Interborough.⁶³ (See Signalling System and Rolling Stock sections, below.) This suggests that the design of large urban and interurban transit systems revealed certain key problems requiring quick and commercially feasible solutions. The interaction of the operating and manufacturing companies in solving these difficulties adds to the importance of the IRT s technical history.

The Main Power Station

[page 305] The choice of electricity as the motive power for the New York subway system represented the culmination of one series of careful decisions and the initiation of another. Parsons' early support of electricity did not extend to a preference for a particular type of generating and transmission system.¹ Electric street and railway power houses varied in size, arrangement, and equipment; direct or alternating current could be produced, each requiring different methods of current distribution.² Designing a suitable system for the subway demanded consideration of the conditions and requirements of rapid transit service.

Many of the direct current railway power houses were products of the 1880's and early 1890's, when many small electric lines were springing up, often in connection with direct current lighting companies.³ As the small railways extended their tracks, transmission of current over greater distances was required. The situation of the predecessors of the Brooklyn Rapid Transit Company is typical. Engineers found that voltage drop was great along the direct current feeders to the far limits of the lines, and voltage boosters were installed at various power stations to assure that sufficient current reached the track. However, direct current transmission over any great distance, even at the increased voltage, involved a tremendous investment in copper.⁴ As the system expanded and load increased, the direct current generating stations revealed themselves to be costly and inefficient.⁵ With respect to power house equipment, then, the requirements of any extensive urban transportation system, whether surface, elevated, or tunneled, were becoming quite similar by the turn of the century. In each case the goal was the same: to produce large amounts of power, and to distribute it to the often far-flung limits of the systems' track. This similarity was ultimately reflected in the standardization of railway power house practice. Alternating current, first commercially introduced in stations transmitting current long distances for power purposes,⁶ became the choice of the large urban railway systems. The Brooklyn company had completely converted from direct current to alternating current by 1904. The power houses of the Third Avenue Railroad Company, the Metropolitan Street Railway Company, and the Manhattan Railway Company, all completed by 1904, produced alternating current at high voltage for transmission to substations for reduction of voltage and conversion to direct current for traction. All were strikingly similar in design.⁷ In this area more than in any other, the designers of rapid transit systems could look to the larger street railway companies as models of standard practice, and all could gain from the recent experience of large alternating current power stations.

Lewis B. Stillwell, the engineer responsible for the design and installation of the Interborough's electric power equipment, was one of the avant-garde involved in developing alternating current for use in urban railways. Before he took on the subway assignment, he was electrical director of the Niagara Falls Power Company, and consulting engineer to [page 306] the Manhattan Railway Company during electrification of its lines. The Manhattan Company had obtained the services of W. E. Baker, who had supervised the electric installation of Chicago's Metropolitan West Side "el". However, the company particularly wanted Stillwell because of his experience with large high tension systems, acquired at Niagara, and earlier with the Westinghouse Company,⁸ which he applied in the design of the high voltage switching equipment at the Manhattan station.⁹

Stillwell was even better prepared for his work on the Interborough system. He could draw on his Manhattan Railway experience, where he was applying high voltage alternating current technology to an electric traction system. He worked on both projects at once, accepting an appointment as consulting electrical engineer to the subway in 1900. He wrote to August Belmont that "my other engagements will aid, rather than interfere with, my work for your Company," an assertion borne out by the marked similarity between the elevated and subway power houses, both in system design and in type and make of equipment.¹¹

As part of his work for the Manhattan Railway Company, Stillwell prepared a report analyzing the advantages and disadvantages of various proposals for the power house and distribution system. His recommendations were adopted for both the elevated and the subway, and since no analogous report seems to have been done exclusively for the subway designers, examination of this report is justified.¹²

Stillwell considered nine alternative plans for delivering power, via a third rail, to the Manhattan's car motors. The plans varied essentially in the number of power houses proposed, and whether direct, alternating, or both types of current should be generated. Stillwell rejected both types of combination plant, one with the same generators capable of producing either alternating or direct current, the other with separate generators producing the different currents, because of the great complexity of the apparatus, the high cost of installation, and the dearth of previous experience with the equipment. Four small direct current power houses were undesirable, primarily because of increased fire risk, smoke nuisance, and complexity of operation.

Stillwell also rejected two other proposals for direct current generation, each calling for two power houses. One, a 2-wire system with powerful boosters, was out-dated and impractical, similar to the system just abandoned by the Brooklyn Rapid Transit Company. An even stronger objection to this, and to the three-wire proposal, involved the use of the track and elevated structure as the neutral conductor for the system. When large differences in load existed between different parts of the line, the great difference in electrical potential along the structure might prove damaging and dangerous.

Stillwell considered plans for two alternating current powerhouses, one with three-wire and one with two-wire distribution. He rejected the latter plan because of the great cost for copper conductors required by its distribution system. A powerful argument against both was the greater initial cost for construction [page 307] and equipment if two power houses rather than one were erected. Stillwell's goals were simplicity and economy of operation, and he believed they would be met by generation from a single large plant.

Indeed, Stillwell vetoed multiple generating stations in all the above proposals. In addition to the greater complexity of this arrangement, the difficulty of finding suitable sites within the city, coupled with the high cost of real estate, argued against multiple generating facilities. The plans he seriously considered called for a single power house. The problem was to determine the most suitable type of current.

Because of the great distance from power house to track, a direct current station required rotary converters at the central plant to produce alternating current for transmission, then a series of substations for conversion back to low-voltage direct current for the third rail. Stillwell saw advantages to this plan; its initial cost was barely more than that of an alternating current station, and most objections to the other plans did not apply here. Starting with alternating current at the generators eliminated the need for power house converters. Current at high voltage was sent from the generators directly to the substations. For Stillwell, the set of rotary converters required in the direct current station represented an additional possibility of malfunction. The increase in the amount of machinery increased cost and, more important, made operation more complex which he knew was unnecessary in an alternating current station. He also believed a direct current plant limited the system's easy ability to expand capacity, as extra converters would be required for additional power to be delivered at a distance. The simpler alternating current plant would be more cheaply and easily expanded.¹³

A single large alternating current plant required purchase of only one major site and reduced the amount of costly copper needed for the distribution system, and numerous small substations feeding individual sections of track could keep voltage differences along the line at a minimum. Stillwell recommended the construction of a single central alternating current station, and the Manhattan Railway Directors agreed. The station, located at 74th Street and the East River, began operation in 1901.¹⁴

John B. McDonald, the subway contractor, was an early advocate of electricity for subways. But he did not decide on a particular system until late in 1901. By then, Stillwell had been a consultant to the Subway Construction Company for over a year.¹⁵ His work for the Manhattan elevated, as well as his earlier experience with high voltage alternating current generation, must have convinced him to adopt an almost identical system for the Interborough.¹⁶

Stillwell's careful analysis of the faults and merits of the available technologies was for the most part, as applicable to the subway as to the elevated road.¹⁷ In each case, power was to be provided to a geographically extensive system of track from a central location in midtown Manhattan. All the arguments concerning the technical advantages of alternating current generation under such conditions were of equal force for the subway. As for the number of powerhouses to be constructed, the New York real estate market similarly restricted the options of both companies.

[page 308] One reason the Manhattan Railway Company had settled on a single large generating station was the difficulty of finding an appropriate second site on the West Side of Manhattan.¹⁸ A steam-powered generating plant required proximity to transportation facilities for coal delivery, an abundant supply of fresh water for the boilers (in New York taken from the City mains), as well as cooling water for steam condensing apparatus. It was also desirable to have the power plant close to the center of its distribution area. Riverside locations were ideal in terms of coal delivery (by barge) and access to the river for condensing water. However, when the subway contractor turned attention to this problem in 1901, the choice East River sites were already taken, by the Edison Waterside plant at 40th Street, the Manhattan Railway station at 74th Street and the Metropolitan Street Railway plant at 96th Street. A site much further downtown would at that time have been far from the distribution center of the road then under construction.

The contractor and Rapid Transit Commissioners were forced to consider several less ideal sites. Parsons strongly favored a midtown location.¹⁹ He opposed the suggestion of a downtown location at 9th Street; he and E. P. Bryan also opposed Long Island City, a location apparently favored by McDonald. Locating the station way downtown, or removing it from Manhattan Island completely, would require a greater investment in electrical conduit from the power house to all substations. Late in 1901, McDonald finally purchased a block at 58th-59th Streets, between 11th and 12th Avenues, previously occupied by Switt Company slaughterhouse and refrigeration plants. The site cost more than others under consideration, but expense was secondary to promotional as well as technical considerations. A power station the city could "take pride in" demanded a central location.²¹ (For discussion of the architectural design of the power house, see Architectural Report.)

John Van Vleck, the mechanical engineer who designed the power station, announced at the time of purchase that he had not formulated final plans.²² However, Stillwell had worked out the electrical system to his satisfaction, and the Manhattan Railway power house was a useful model for the boiler plant as well.²³ By the end of 1901, the major subcontracts for the power house equipment had been awarded to Allis-Chalmers for the steam engines, to Babcock and Wilcox for boilers, and to Westinghouse for the generators and exciters.²⁴

Excavation for the power plant began in April 1902.²⁵ August Belmont saw that the early completion of the station was necessary in order to maintain the projected timetable for the opening of the subway. In March 1902 E. P. Bryan saw to the progress of work at the Westinghouse and Allis-Chalmers shops. He reported that all the subway work was going smoothly, and that four engines and generators might be installed by autumn 1903. Parsons kept watch on the construction, and in January 1904 gave the first tour of the plant to members of the American Society of Civil Engineers.²⁶

The power house was originally 540 feet long, with its west wall closed by a bulkhead to allow for later expansion. During 1902, McDonald received Contract 2 for the Brooklyn extension. He met the anticipated increase in power requirements by enlarging the main power station. By January 1905, the work of extending the building to its completed length of 694 feet was underway.

[page 309] Despite a five-month delay due to labor strikes, four of the projected eleven 5000-kilowatt main generating units were in place, each supplied a battery of six boilers, when the subway began operation in October 1904. These units had for several weeks provided current during rush hours for the elevated lines, demonstrating their fitness for service. By December, two more generating units were installed, the plant housed 42 boilers and complete installation was anticipated shortly.²⁸

* * *

The structural design of the 59th Street power station was the responsibility of William C. Phelps, who had done similar work on the Manhattan Railway power house between 1899-1901.²⁹ The structure was divided into two essentially separate buildings, the southern half was to house the boiler plant, the northern half the steam engines and electrical equipment. This was a standard design for large steam-operated plants.³⁰ Each "room" extended the full length of the building. Galleries along the north wall of the generating room supported electrical switches and control board, and galleries along the south wall supported the auxiliary steam piping. The main northern gallery also housed equipment for a repair and machine shop.³¹

John Van Vleck designed the boiler plant of the power house according to a unit plan which, when the station was completed, divided the plant into six independent functional sections. Each unit contained two batteries of six boilers each, feeding two steam engines in the generating room. Power-operated valves disconnected each boiler/engine unit from the main system; any number and combination of units could be operated at a time. For each unit, there were also two condensers, one for each battery of boilers, and likewise two boiler-feed pumps, two smoke-flue systems with economizers, and two complements of auxiliary apparatus. The twelve boilers were symmetrically arranged around one of the six chimneys. Five of the units were identical; the sixth broke the symmetry with a steam turbine plant, installed instead of reciprocating steam engines to power the generator for lighting the subway tunnels.³²

The Interborough Company was not the first to adopt such a unit plan in its power house design. The Metropolitan Street Railway plant was arranged to allow for the separate operation of units, even though in practice all machines were operated together. The Manhattan Railway plant was designed for unit operation on a smaller scale than the Interborough; each chimney served two units, each which contained an engine supplied by four boilers.³³ The Interborough scheme was unique in that Van Vleck expanded and developed the idea of unit design by arranging the coal bunkers above the boilers, which were divided by the chimneys into seven separate units. Spontaneous coal combustion could be localized easily, and the plan also allowed for storage of differing grades of coal. The system of steam piping from boiler to engine was also exceptional; the identical steam piping in each unit, according to Van Vleck, gave "a piping system of maximum simplicity, which can be controlled in the event of difficulty with a degree of certainty not possible with a more complicated system."³⁴ This simplicity, coupled [page 310] with the possibility of independent operation, lent great flexibility to the operation of boilers and engines. In addition, a section of steam piping, or an economizer unit, might be repaired without closing adjacent sections.

Adoption of a unit design reflected a desire for simpler, more elegant operation, ease of repairs, and flexible use of coal and steam. The latter consideration was key. The cost of coal represented the single greatest operating expense of a steam-powered generating station, and any modification of which resulted in its more efficient use, represented substantial savings. Such modifications could be introduced into the coal circuit, feed water and steam circuit, or condensing water circuit, effecting cuts in the cost of electric power production even before the current passed the switchboards.³⁵ Adoption of the unit plan, and other modifications introduced by Van Vleck, reflected the desire to produce power as cheaply as possible through the efficient burning of coal and economical use of steam.³⁶

The 59th Street power house received its coal via Hudson River barges.³⁷ The coal was unloaded at a 700 foot long, 60 foot wide pier, specially built by the City's Department of Docks and Ferries. Coal was weighed and crushed in an electrically operated hoisting tower at the pier, and deposited on motor-driven 30-inch underground coal conveyors for delivery to the power house. Elevating conveyors at the west end of the plant carried coal 110 feet, where 20-inch horizontal conveyors distributed coal to the bunkers. To guard against the accumulation of coal at important junctions, each conveyor in the system ran 10 feet per minute faster than the conveyor that supplied it.³⁹ Automatic, self-reversing trippers along the conveyors ensured even distribution to the bunkers.

The independence of the seven bunkers allowed the company to deliver different grades of coal to the different bunkers. A system of distributing conveyors arranged beneath the coal bunkers allowed the plant operators the options of delivering coal from a bunker direct to the hopper beneath it, or of delivering coal to any or all other hoppers along the belt conveyors. High grade coal from one bunker could be delivered during heavy-load periods; when power demands were less, low grade coal from another bunker, or combinations of grades from different bunkers, could be fed to all the boilers. This flexible use of coal, mixing and matching various grades to suit operating conditions, reduced operating costs in two ways: the initial purchase price was reduced, since cheaper grades could be included and stored separately; matching grades of coal to the demands for power meant more efficient consumption of the coal stockpile.

Beneath the boilers, ash hoppers delivered their load to ash cars, pulled by storage battery-powered locomotives back beneath 12th Avenue to the pier for unloading into barges. The distinguishing feature of the entire coal/ash system was its completely automatic operation.⁴⁰

[page 311] The plant's projected capacity called for 72 Babcock and Wilcox sectional water tube boilers, identical in design to those installed in the Metropolitan and Manhattan Railway plants.⁴¹ Automatic coal handling ended at some of the boiler grates, which were hand-fired from a platform erected between the two rows of boilers. Others were provided with Roney automatic stokers. Within two years, however, the company installed Roney stokers for each boiler, hoping to economize on fuel and labor costs.⁴²

City mains provided all the feed water, since the use of jet condensers rather than surface condensers precluded the recycling of steam for boiler feed. Feed water was heated partly in its storage reservoirs by water discharged from the condensers' hot wells. Further heating was accomplished by economizers, placed in a level above the boilers.⁴³ Hot boiler flue gases could either enter a chimney directly, or at the discretion of the plant operator, pass first into the economizer apparatus. Here, heat exchange between the gases and the enclosed feed water sufficiently raised the water's temperature prior to entering the boiler tubes.

The arrangement of the boilers, economizers, and chimneys was the most original and significant feature of the boiler plant. Van Vleck's structural design, complemented by the function of equipment, enhanced both the safety and efficiency of plant operation. Floor space was efficiently used. The giant Custodis radial chimneys did not pass through to the building foundations. They were supported instead by steel columns, their bases raised well above the boiler room floor level.⁴⁴ Space normally required for the chimney bases was therefore available for other uses. At the Manhattan Railway power house, standard practice was followed in placing the boilers on two levels, one above the other.⁴⁵ At 59th Street Van Vleck used his extra floor space to place all his boilers in two long rows on the main operating floor. The absence of a second boiler level allowed for a higher, well-lit boiler room. Ventilation into the floor above, in addition to the row of windows, helped reduce temperature extremes and the dangers of escaping steam.⁴⁶ The boilers were also installed higher above the floor than was standard, providing for a correspondingly higher combustion chamber with either hand or automatic stoking.⁴⁷ Van Vleck set the economizer units, customarily installed directly beside the boilers, on the upper floor. The removal of economizers and flue connections to another level further reduced the chances of operational disturbances on the boiler floor.

Placing the economizers above the boilers also widened the boiler room while reducing the total width necessary for the installation. Again floor space was freed for new uses. Van Vleck used this space for the steam piping, enclosed in a side gallery between the boiler and generator rooms. By setting the piping apart, with its controlling valves power-operated by men outside the actual area, Van Vleck decreased the danger and nuisance of leaking steam entering either the boiler or generator areas.

[page 312] A "distinctly new and interesting" steam piping arrangement allowed great flexibility in the application of steam to the engines.⁴⁸ Each group of six boilers fed a steam main which divided upon entering the pipe gallery. From here, steam could follow one of two paths, depending on the valve configuration. Steam could enter two 14-inch mains leading to receivers in the basement, which fed the high-pressure cylinders of the engine, or it could enter a manifold, a system of 12-inch pipes connecting the steam mains of all the boiler groups. With the valves to the manifold shut, each boiler/engine group could be operated completely independently, an especially valuable feature during repair work. With the manifold valves open, the 12-inch pipes acted as an equalizing steam header, distributing steam from all the boilers to all the engines.

A boiler could be disconnected from its corresponding engine, feeding only into the manifolds; likewise an engine could be powered by steam from any boiler. This system ingeniously combined the advantages of the equalizing steam header with those of the unit plan, thereby permitting manipulation of the boiler/engine connections to suit different operating conditions. In this sense the steam piping system was analogous to the coal distribution system beneath the bunkers.

Steam passed from the piping system to the engines in the generating room. The 12,000-horsepower Allis-Chalmers reciprocating engines were "twins", consisting of two compound engines connected by means of a crank to either end of a single main shaft. Each component compound engine consisted of a horizontal high pressure cylinder which emptied steam into a vertical low-pressure cylinder. Both cylinders attached to the single crank at either end of the main shaft. These two cranks were set at different angles, an arrangement providing greater uniformity in the main shaft rotation.⁴⁹

Allis-Chalmers had first installed this type of engine three years earlier at a Manhattan Railway plant.⁵⁰ Small modifications and improvements had since been made, the most important of which was the substitution of poppet valves for Corliss valves in the high pressure cylinders. The Manhattan engine design had been criticized because Corliss valves were considered inadequate if super-heated steam were employed.⁵¹ The use of the more suitable poppet valves at 59th Street was an important improvement,

The Interborough directors and engineers had considered steam turbines before deciding in 1901 upon the reciprocating engines. They found that conventional turbines did not have the capacity required for the subway power house. Brown, Boveri and Company in Switzerland were the only firm that constructed suitable 3500-kilowatt units. The IRT Company made the conservative choice of reciprocating engines for the traction system, but chose three Westinghouse turbine-generators for the smaller subway and station lighting circuit. Almost immediately, however, the management recognized that advances in turbine technology demanded that any expansion of generating capacity be turbine-generator installations.⁵²

[page 313] Two Alberger jet condensers served each steam engine.⁵³ Condensing water from the Hudson River entered an oval intake tunnel at a river wall beneath the power house pier, extending under 12th Avenue through solid rock to the eastern 11th Avenue end of the plant. Water was filtered through a series of fine screens behind a coarser steel grillage at the entrance to the intake tunnel. A horse-shoe shaped conduit, built on top of the oval tunnel, served as a discharge tunnel.⁵⁴ At the center of the pier, two timber conduits carried the hot discharge water to either side of the intake screens, to prevent its mixing with the cool water heading for the condensers.

A jet of water entered the condensing chamber through a spray cone. Steam, leaving the low pressure cylinder of the engine, entered the opposite end of the chamber and was condensed by direct contact with the cool river water. The Alberger condensers were of the barometric type, containing a tail pipe with a barometric water column. This column allowed discharge water to flow from the tail pipe into the hot well against atmospheric pressure, keeping air out and thus preserving the vacuum until the next engine cycle.

The use of jet condensers, in which condensate mixed with the spray of cooling water and flowed out through the hot wells, prevented the recycling of steam for boiler feed. Recycling was possible with the use of surface condensers, in which cooling water enclosed in small metal pipes circulated through the condensing chamber. Contact with the cool metal condensed the exhaust steam. In this way condensate was kept separate from the cooling water, and could re-enter the boiler as feed. Where pure water supply was limited or expensive, the recycling of exhaust steam was an important economy.

The Interborough Company, which purchased its feed water from the City of New York, might have saved money with the use of surface condensers. Of the three other large railway power plants in the city, however, only one, the Metropolitan, used surface condensers. The Third Avenue Railroad and the Manhattan Railway plants had jet condensers. The objection to surface condensers was the presence in the condensate of lubricating oil from the engines, The water could not be sent to the boilers unless the oil was removed; a difficult and expensive task given the methods then in use. Both the Third Avenue and the Manhattan engineers expected to install surface condensers when a satisfactory method of oil removal was developed, but until then, purchasing clean water from the city was considered more economical.⁵⁵

These same considerations motivated the choice of jet condensers for the main engines of the Interborough plant. But as the Westinghouse turbines of the lighting installation required no oil, surface condensers were more economical. An Alberger counter-current surface condenser served each turbine unit. The cooling water circulated in tubes from the top downward, while the steam entered the chamber at the base. The water of condensation, leaving the chamber at the base, was therefore heated by contact with the entering steam on its way to the feed tanks, eliminating the need for additional heating equipment.⁵⁶

[page 314] Westinghouse supplied the generating equipment for the power house.⁵⁷ Nine of the projected eleven 5000-kilowatt alternating current generators were in place by October 1904. The generators were directly connected to the steam engines, with the hub of the revolving field forced onto the main shaft between the two component compound engines of each engine unit. The fly-wheel effect of the revolving field, turning at 75 rotations per minute, helped maintain a high uniformity of rotation without an auxiliary fly-wheel.⁵⁸ Five 250-kilowatt direct current generators provided 250-volt exciting current for the revolving fields. Three were driven by direct connection to induction motors, the others by 400-horsepower marine-type steam engines. Current from the exciter plant could also be switched into the circuits supplying the motors for the station's auxiliary machinery.

The generators produced 3-phase, 25-cycle alternating current at 11,000 volts. The armature windings embodied a new design, with the ends of U-shaped copper coil conductors slipped through the armature slots and soldered together to form closed coils. Otherwise the generators were virtually identical in size and design with those installed by Westinghouse in the Manhattan Railway plant.⁵⁹

Stillwell and the electrical engineers chose the 5000-kilowatt generator because a large unit was desired which could still be directly connected to the engine shaft using only two bearings. Larger units required more bearings for direct connection, which the engineers deemed inadvisable because of greater opportunity for malfunction. Smaller units did not suit the rapid load changes characteristic of railway service, which required sudden increases or decreases in both the morning or early evening. Plant operation was simpler if one or two large machines were brought on or off the line to effect the desired change rather than cutting in or out many small units. The 5,000-kilowatt unit therefore represented the best size for the plant.⁶⁰

The Westinghouse turbo-generator installation divided the line of alternators at the center of the operating room. The turbines were each directly connected to a 1250-kilowatt alternator. The total rated capacity of the station, including both the engine and the turbine plants, was 80,000 horsepower. The engineers expected actual effective operation at 100,000 horsepower, and an additional 30,000 horsepower was proposed for the western extension.

Current traveled from the generators through the switchboards for distribution to the substation. The high tension switches were on the main gallery, at the operating floor level along the northern wall. All switches for the 11,000-volt current were operated under oil. The control operator could send current to one of two complete sets of bus bars in brick compartments on the mezzanine floor below the circuit breakers, by closing the appropriate selector switch. The operator overlooked the main generating floor from the switchboard gallery above the circuit breakers, which contained General Electric's generator and feeder control boards, exciter boards, and [page 315] control panels for the lighting plant and auxiliary apparatus. Current flowed from the main to auxiliary bus bars and from there to the feeder circuits for the substations. Each circuit was controlled by a type-H oil switch, operated by an electric motor which opened and closed the switch by means of great springs. The operator worked the switch by hand from the control board, but the switch was designed to open automatically in case of overload, as were the alternator switches to the control boards. A time attachment could set the overload relays to open at a pre-determined time, from 3 to 5 seconds, after the overload current began. In this way current would automatically be prevented from flowing through the main switchboard, or along the feeders to the substation, on any malfunctioning circuit. The use of the oil switches, automatic relays, and the arrangement of the control apparatus with respect to the operating floor, underscored the careful attention given to switching control where high voltage was used.

Through the plant, the adoption of the unit plan, combined with Van Vleck's unique arrangement of equipment, brought simplicity and flexibility to plant operation. Along each unit, power flowed smoothly and directly; coal downtakes led to the single boiler level; steam passed from six boilers arranged symmetrically with respect to the engines; along the single line of engines and generators, steam flowed in from the south and electric current flowed out to the north toward the main switchboards. Plants with two tiers of boilers, or multiple lines of engines, could not approach the simplicity of piping and wiring, so important for ease of repairs and operation, achieved at the 59th Street power house.

* * *

After 1904 the Interborough company modified and expanded the plant in response to load growth and the desire to increase the efficiency, and hence the economy of plant operation.⁶¹ Soon after operations began, the company set up a laboratory for coal analysts at the unloading dock. Coal was sampled as it left the barge, evaluated according to company specifications, and a bonus or penalty was awarded to the supplier for an especially good or poor quality. The coal laboratory ensured that the plant furnaces received coal most suited to plant conditions, increasing plant efficiency.

The original sixty boilers were supplied with hand-fired grates. When the installation was completed to 72 boilers late in 1904, the new units had Roney mechanical stokers, and, as noted above, all units had mechanical stokers within two years after operation began. By 1907, expanded transit demands required the plant increase its capacity. Company engineers determined that the installation of additional stokers would provide a 50% gain in steaming capacity per boiler and 18 boilers received additional equipment. This represented a pioneering innovation in boiler practice, [page 316] by which increased steaming capacity, with more steam produced per pound of coal, was obtained from a fixed area of heating surface. Economies of space and of fuel utilization were both realized.

In 1909-1910 the company took advantage of the additional capacity by installing five 7500-kilowatt Curtis-type General Electric vertical turbo-generators supplied from the low pressure cylinders of the first five reciprocating units. Surface condensers were installed for each turbine. By taking steam from the engines at close to atmospheric pressure, the turbines increased generating capacity by 15,000 kilowatts without requiring a corresponding increase in the heat of the steam, which constructed a yet more efficient use of coal. The engine/turbine combination was more efficient than any steam turbine then available.⁶²

In 1917 the Interborough extended its track. The company had steaming capacity since 1913 by gradually replacing 42 single and double Roney stokers with 7-retort Taylor underfeed stokers with a greater coal burning capacity.⁶³ After 1917 the substitution was completed, and with the use of induced draft the boilers developed 250% of the original rating. In 1917, the company installed three General Electric horizontal 35,000-kilowatt turbo-generators, the largest and most fuel efficient of this type then available. In conjunction with this installation the company provided 30 boilers with superheaters; steam temperature could reach 150 degrees Fahrenheit before entering the engine. The turbines amply increased generating capacity. In 1924, increased steam requirements were met by four additional boilers with underfeed stokers.

Switching and control equipment required modification as the power capacity of the station increased. Modernization in 1915 effectively prepared the way for the 1917 expansion. At this time a central control service, covering the operation of the 59th Street and 74th Street stations and the substations, was set up at 59th Street. Control operators integrated the service of the two large stations to the substations, and kept in contact with the transportation divisions dispatching power in accordance with existing load and equipment conditions.

From 1924 until New York City took over the subway in 1940, no significant additions or changes were made in the 59th Street plant. After 1940, the city made studies in order to plan modernization of the power equipment. In 1954 a J. G. White Engineering Company report termed the plant "an engineering museum piece," an "exhibit of primary pioneering in applied engineering as of 50 years ago."⁶⁴ Although the city had finally initiated a modernization program, only one new high pressure boiler/generating unit was installed between 1946 and 1952. Five of the original reciprocating engines were still in operation, in conjunction with old low pressure boilers. Much of the plant equipment, the old boilers, the old switchgear, and the coal and ash handling facilities, were dangerous and inefficient.

[page 317] In 1959, the City accepted the offer of the Consolidated Edison Company to take over control and operation of the rapid transit power plants.⁶⁵ Con Ed immediately launched a modernization program for the stations. The 59th Street plant was soon completely overhauled. Of the original equipment, little remains. The exhibits of the "engineering museum" were gradually discarded in favor of more modern, efficient equipment.

Electrical Equipment, Main Power Station, 59th Street

Detailed specifications and description of selected equipment of the original installation.

Chimneys: 5 (later 6) Alphonse Custodis radial chimneys, 162 ft. high from base (230 ft. above street level), 15 ft. diameter at top, weight 1200 tons, supported by steel girders and columns.

Boilers: ultimate installation, 72 Babcock and Wilcox sectional water tube boilers, 21 water tube sections, 14 tubes high, 6008 sq. ft. heating surface each, designed for working pressure 225 lbs. steam. For 36 boilers, hand-fired Gibson grates provided, gate area 8 ft. deep, 2.5 ft. wide, blowers and air ducts beneath boilers (1 blower per 3 boilers, Sturtevant blower driven by direct-connection to a 2-crank 7.5x13x6.5 inch upright compound steam engine.

Steam Engines: Allis-Chalmers twin compound engines, nominal capacity 8000-hp, actual capacity 12,000-hp at 75 rpm and steam at 175 lbs. at the throttle; shafts of hollow forged open hearth steel; cranks of fantail type, cast steel, crank pins of nickel steel, diameter of shaft at hub of revolving element 37 1/16 inches, main bearings 34 inches.

Alternator Generators: Westinghouse, 5000-kw delivering current at 11,000 volts, 263 amps, 3-phase rotating field, stationary armature, height 42 feet, weight 389,000 lbs; speed, with 40 field poles, at 75 rpm, giving 3000 alternations/minute, or 25 cycles per second; at 75 rpm flywheel capacity. 32 ft. high, 332,000 lb. field not less than 37,000 lbs. Armature stationary, exterior to field, consists of laminated ring supported by cast iron frame, U-shaped copper bars fit partially closed slots, (4 bars/slot)

Exciters: 5 250-kw direct current Westinghouse dynamos, delivering current of 1000 amps at 250 volts, speed 150 rpm; 3 exciters driven by induction motor, 2 by non-condensing, vertical quarter-crank compound steam engines, built by Westinghouse, Church, Kerr, and Co., max capacity 600-hp.

Turbine-Generator Lighting Plant: 3 Westinghouse steam turbines, multiple expansion parallel flow type, each consisting of two turbines operating in tandem; estimated output per unit 1700-hp with a steam pressure of 175 lbs. at throttle and 27 inch vacuum in exhaust pipe; guaranteed satisfactory operation with steam superheated up to 450 degrees F.⁶⁶

Chronology of Alterations at the 59th Street Power House made by Consolidated Edison since 1960.

From compilation prepared from Annual Reports and employee magazines by Ms. Dorothy Ellison, Consolidated Edison Co., 4 Irving Place, New York, New York.

1959. Con Ed acquires plant, embarks on smoke control program at all three transit plants.

1960. Old low-pressure boilers shut down, reducing smoke emissions from stacks. Installation of modern high-pressure boilers. Interconnections established between 59th Street and other transit and Con Ed plants. Labor force for plant operation reduced from 1,200 to less than 700. Topping turbines installed (also at 74th Street).

1962. Additional low-pressure boilers replaced by high-pressure units. 22,000-kw topping turbine installed.

1966. Two new boilers and 35,000-kw turbo-generator installed. Single 500-foot stack replaced western-most four 240-foot stacks (preliminary to gas turbine installation)

1968. Plant completely converted from coal to oil and gas fuel.

1969. $53 million spent in modernizing the three transit plants in previous decade. Consolidated Edison uses steam from the transit plants to supply steam system.⁶⁷

The Power Substations

[page 329] Electricity generated at the 59th Street power house for use in the subway went first to the eight original sub-stations, most of which are still in operation today.¹ Here power was altered, processed, packaged, and put into a form appropriate for the job it had to do in the tunnels. Here also were air compressors, supplying the electro-pneumatic signaling and switching installations.

Transmitting high voltage alternating current from a main power house to sub-stations for conversion to lower voltage direct current suitable for the car motors was a relatively new practice in railway work.² It took advantage of recent advances in alternating current technology, particularly the development of a satisfactory 25-cycle rotary converter, first introduced by Westinghouse in its Niagara installation just a few years before.³ The use of alternating current, requiring less copper conductor and resulting in lower transmission cost, allowed the economical and efficient transmission of power to sub-stations spaced approximately two miles apart along the subway route.

Because suitable real estate was difficult to find in the built-up downtown areas, contractor McDonald suggested that some of the sub-stations be placed underground. In February 1901 he requested the aid of the Rapid Transit Commission in acquiring the right to excavate under public lands at City Hall Park, Union Square, and Longacre (Times) Square. McDonald's contract made him responsible for the purchase of all lands for power facilities and he hoped to cut down his expenses by using city rather than private property. After consulting its lawyers, the Board decided that it lacked authority to grant this request. McDonald had to build his sub-stations above ground.⁴

It was desirable to have the distribution distance to the subway as short as possible after conversion to direct current at the sub-stations. In the downtown areas McDonald obtained sites no more than one-half block from the route. In the far less crowded up-town locations, the Simpson Street and the Hillside Avenue sub-stations were nearly adjacent to the track.⁵

Two adjoining city lots, each 25x100 feet had to be purchased to house sub-station equipment. The resulting 50 foot width allowed installation of eight to ten rotary converters with their sets of transformers. In Sub-station #13 on West 53rd Street, foundations were laid for ten rotaries; the remaining seven were built to receive eight rotaries.⁶

Foundations for eight to ten rotary converters was a provision for the future. The original 1901 Westinghouse contract called for only 26, 1,500-kilowatt rotary converters, or four to five per sub-station.⁷ In 1909 Westinghouse responded to a second call, this time for 3,000-kilowatt units. In the plans for the 1916-1918 general system expansion, additional contracts to both Westinghouse and General Electric provided 4,000-kilowatt rotaries, some of which replaced the older 1,500-kilowatt machines.⁸ During expansion, [page 330] Sub-station 11 at Park Place was demolished, and its replacement, a half block from the original site was equipped with 4,000-kilowatt units. In 1923 additional 4,000-kilowatt General Electric and Westinghouse units were installed.⁹

The remaining seven of the original eight IRT sub-stations are still standing. Number 19 on West 132nd Street is no longer in use and its equipment has been removed. The others still operate daily [in 1978] with equipment from the earliest installations. [The last of the rotary converter equipment has since been retired.]

* * *

At the 59th Street power station 11,000-volt, 3-phase, 25-cycle alternating current destined for the sub-stations passed along single conductor cables to the main switchboards. From here high tension feeder cables, with three strands of copper conductor, each carrying one phase of the current, extended through vitrified clay ducts from 11th Avenue under 58th Street to the subway structure at Broadway. Paper insulation separated the three conductors. The cables were 000 Bond S gauge, sheathed with lead. Insulating rubber placed between the sheaths protected them from electrolysis.

The side walls of the subway tunnel contained 64 ducts on either side of the subway, stacked 32 ducts high and 2 ducts wide. Through these ducts the alternating current cables extended through the tunnel to the sub-stations. At each passenger station the side walls receded sharply. Following this path with the cables would have unnecessarily increased transmission distance. The subway engineers, therefore, routed the cables beneath the station platform, effectively turning them on their side; beneath the platform the ducts were 32 wide and 2 high. At the end of each platform the cables re-entered the side wall, adopting their original configuration.¹⁰

At each sub-station the cables ran from the subway to a manhole, or vault, at the front of each sub-station building. They followed tiled ducts to the rear of the stations, which carried them directly under their proper oil switch, where each of the three conductor strands attached to one of the three fixed terminals of the switch. The oil switches included a motor-operated reverse current relay between the incoming cable and the bus bars, which opened the switch in case of short circuit. This, coupled with the action of the overload time delay relays at the alternating current feeder switches at 59th Street would disconnect the cable from the power house and sub-station during any disturbance on the line. This protected adjacent cables from possible damage, and allowed much of the station operation to continue unaffected.¹¹

The pathway of current through the sub-station was protected at every crucial point with similar oil circuit breakers. From the incoming alternating current circuit breaker, current passed through a disconnect switch to the 11,000-volt high tension bus. As noted above, the connections at the switches maintained three distinct conductors for each phase of the current; each conductor carried current, through another disconnect switch and oil circuit breaker, to one of a bank of three single-phase transformers. The voltage of each phase was therefore separately reduced. Current at 550-600 volts passed through another switch to the [page 332] three-phase rotary, which converted the alternating current to direct current and sent it through a switch/breaker set to the direct current bus.¹². From here a feeder cable carried the 600-volt propulsion current to the third rail.¹³

The arrangement of equipment was identical in the eight original sub-stations.¹⁴ On the main floor were the rotary converters, arranged in two parallel rows along the building's length. Placed between each rotary and the nearest side wall was its bank of three transformers.¹⁵ A raised gallery at the rear of the building supported the control and switch boards and the direct current feeder oil circuit breakers.

The high tension alternating current cables passed from their entrance vault along a basement wall to the high tension breakers, located on the main floor beneath the gallery. These were operated by heavy springs wound by an electric motor. The high tension bus, carried in brick compartments, cut the building along its width on the main floor. Cables extended from the bus to the alternating current panel board in the gallery.

From here, conductors returned to the main floor to bring current to the voltage transformers. These machines were air cooled.¹⁶ In the basement, an air chamber extended beneath each longitudinal row of transformers. A motor-driven blower at the head of each chamber filled it with air slightly above atmospheric pressure. This pressure pushed cool air upward through passages to the transformer buses. Air flowed through the coils and out the top of each transformer unit.¹⁷

Each rotary converter stood in its own hard-wood frame. The frame was not bolted to the Portland cement foundations; the rotary weight was expected to hold the unit in place. The rotaries were the heaviest equipment of the sub-station. Two hand-operated cranes, at the front of each sub-station on the main floor, were provided for the rotary installation and service.¹⁸

The rotary converters could be started in one of two ways. By switching the conductors of the direct current side of a rotary into a direct circuit to the direct current bus, a station attendant could start the rotary as a direct current motor. A special starting control panel in the gallery contained the appropriate switches. When the panel instruments indicated that the rotary had reached a speed synchronous with its transformers, the attendant threw the switch connecting the transformers to the alternating current side of the rotary.

If the sub-station had been entirely shut down and the direct current bus were dead, the rotaries could be started from a motor starting set. Each set consisted of a 3-phase induction motor directly -connected to a direct current generator, [page 333] mounted together on a common base. The motor had enough capacity to start all the rotaries in the sub-station, although once one was started and put on-line, the rest could be started from the direct current bus.¹⁹

The direct current switch boards extended across the width of the gallery. The oil circuit breakers for the direct current feeders were located in individual brick compartments facing the direct current switch boards, with each breaker directly opposite the hand-operated switch which closed its particular circuit. Company engineers believed this arrangement was first introduced in the IRT sub-stations. The isolation of each direct current feeder breaker protected the others from damage should a breaker open automatically due to a short circuit.²⁰

The General Electric Company provided the instrument panel mounted on iron columns above the switches and control boards. The three feet between the instruments and the bench board allowed the plant attendants to view the operating floor while working at the controls. The direct current feeder cables carried current from the gallery to the basement and extended through ducts to the subway structures.²¹

The sub-stations were responsible for much more than provision of the propulsion current to the third rail. High voltage alternating current from the 59th Street turbo-generator installation, intended for tunnel and station lighting, was routed first through voltage transformers in the sub-stations. The reduced current did not pass through a rotary converter; it went directly to a second transformer within the tunnel, which further reduced the voltage to the level required for the lighting and auxiliary power circuits. In addition, all power for the electro-pneumatic signaling and switching installations originated at the sub-stations. Motor-generators provided alternating current for the signals' track relay circuit; Ingersoll-Sergeant air compressors supplied the pneumatic cylinders which controlled the movements of switches and signals.²² See Signaling, Switching and Safety section.

Although the appearance of the original IRT sub-stations was neat and symmetrical, the current pathway was somewhat confused and indirect. Cables carried electricity from front to back, from level to level. Engineers were beginning to recognize, however, that at least for small installations, the advantages of a raised control gallery were outweighed by those of placing all equipment on one level. An energy pathway as direct as possible from the incoming high tension feeders to the outgoing direct current cables minimized confusing connections and simplified station operation. The newer sub-stations built during the 1916-1918 system expansion abandoned the old design; and all transformers, rotary converters; and switch and control panels stood on the main level.²³ The original sub-stations were the only ones built by the IRT with raised switch board galleries.

* * *

[page 334] The original sub-stations have changed remarkably little as the New York subway system changed and expanded. As noted above, the need to increase the stations' current-handling capacity resulted in two additional rotary converter installations, with the necessary transformers. Each time, the capacity of the new rotaries was higher, 3,000-kilowatts in 1909 and 4,000-kilowatts in 1916 and 1923. However, since provision had been made for additional installations, and because the 4,000-kilowatt units were actually smaller in size than the older 1,500-kilowatt machines, the new equipment fit neatly into the sub-stations without interfering with the original design.

The IRT did make some changes in the sub-stations to accommodate expansion. The sub-station at Park Place was replaced, relocated, and equipped in 1917, as already noted, with higher capacity equipment. The same year the west back wall of sub-station 12 was extended through the next lot, making space available for offices and allowing extension of the gallery.²⁴ In 1919, substation 13, at 53rd Street, which was built with greater capacity than the others, received a new switch board. In its downtown location, it was expected to carry much of the increased load of the system.

Although the Consolidated Edison Company took control of the 59th Street power station in 1959, the Transit Authority retained control of the power sub-stations. The Transit Authority had by this time determined that mercury-arc rectifier units, without troublesome moving parts and more efficient at light loads than rotary converters, would take on all expansions in load. Today solid-state rectifiers are expected to replace the old rotary units, and the original sub-stations will be phased out of the system.²⁶ Control panels for the new units are presently located on the galleries of the sub-stations.

The downtown sub-stations handle not only IRT lines, but lines which were originally pant of the BMT system. They generally operate 24 hours a day. The uptown West Side Manhattan stations still serve only the original Contract I IRT route. The sub-stations at 143rd Street and at Hillside Avenue contain only equipment from the original 1901 Westinghouse contract; the early system expansions did not affect the load on the far-flung stretches of track. The plant attendants operate these stations much as their predecessors did in 1904. The rotaries are manually cleaned and serviced. Only the blinking lights on the recently installed mimic boards and rectifier control panels reflect the subsequent modernizations of the system.

These uptown sub-stations are in use only during rush hours and other peak load periods. They will be the first of the original sub-stations to be abandoned and dismantled.²⁷ As long as they are used they will continue as "operating museums", the most important equipment remaining from the original IRT installation.

Electrical Equipment, System Sub-Stations

Rotary Converters: 1500-kilowatt, 3-phase Westinghouse Electric and Manufacturing Company. D. C. EMF 625, D. C. Amps 2400, Alts. 3000, 250 rpm, 25 cycles, weight 130,000 lbs. Rotary field of low fly-wheel capacity, giving converter great synchronizing power; armature is slotted drum type, with core of laminated steel, armature winding of parallel type; commutator bars of hard drawn copper, insulated by mica, brushes of carbon.

Voltage Transformers: Westinghouse air blast transformer, 550-kw, 411 volts, 300 Alts., at full load. Air for cooling delivered at 1 ounce per square inch, 2000 cubic feet of air required per minute.

Motor-Generator Starting Sets: (A) Westinghouse Type C induction motor, 3-phase, wound for 3000 Alts., 300-400 volts, 4 poles. (B) Westinghouse 4-pole direct current generator, 2-bearing type, directly coupled to motor. Motor and generator mounted on common bed plate.²⁸

Sub-station Locations

* From H. F. Parshall and H. N. Hobart, Electric Railway Engineering, New York, 1907, p. 271.

** From Commissioners' Report 1906, p. 246.

Rotary Converter Installations, by Sub-Station

* No rotary converters installed at this foundation

** The discrepancy between the number of rotaries installed by 1906 and the total in the original installation is accounted for by the late completion date of these sub-stations. The track which they served, from 145th to 242nd Streets, was completed between 1904-1906, and put in operation after the subway's official opening in October 1904.

*** From Commissioners' Report, 1906, p. 246.

**** Book of Drawings: Sub-stations, Special, Book no. 1, "Synchronous Converters - IRT Sub-stations," 21 November 1955. Book held at Power Department, Transit Authority, IND Division Sub-station, 126 West 53rd Street, New York.

The Right-of-Way, Third Rail, and Rolling Stock

[page 341] John McDonald's contract contained the following specifications for rolling stock: cars were to be numerous enough to provide for at least three-car trains on local tracks at two minute intervals and four-car trains on express tracks at five minute intervals; car construction allowing quick loading and discharge of passengers; attractive appearance, a minimum seating of 48 persons; and thorough ventilation.¹ The contract included no specifications for motive power equipment, since both electricity and compressed air were being considered. The Commissioners later explained that they had purposely set aside decisions on electrical equipment of rolling stock because of the relative youth of electric traction. Rapid technical development characterized the field and in the Board's opinion, to commit itself too early would preclude the installation of the best and most up-to-date equipment. The contracts for electrical equipment of cars were not let until 1903,² leaving the subway engineers time to evaluate the suitability of different types of equipment for tunnel service.

Two major problems to be solved; a suitable means of current conduction to the car motors had to be developed as did the efficient and safe use of current by the motors. The limited head room in the subway, plus low clearance on curves, also had to be considered. The Commissioners' goal of high capacity and speed with frequent stops, further limited choices. Finally the engineers wanted rolling stock that would not break or wear down easily and that would be virtually fireproof in the event of collision.³

* * *

The limited head room in the tunnels prevented the use of overhead trolleys as current collectors.⁴ Third rail systems were already successfully operating in Chicago, Boston, Brooklyn, and on the Manhattan elevated lines, as well as on interurban roads.⁵ McDonald's electrical consultants urged him in 1900 to adopt the third rail for direct current conduction.⁶

Additional factors influenced the choice of the direct current third rail. The use of alternating current had been briefly considered, but was rejected for two reasons. Work on alternating current motors for traction had been discouraging; and those available in 1900 were relatively untried and less successful than direct current motors.⁷ In addition, the Manhattan Railway Company adopted the third rail in its 1899-1900 electrification. For the subway directors, the Manhattan system was in this case more than simply a technical model. Adoption of a three-phase alternating current system would have prevented interchangeable operation with the elevated lines, which the subway officials considered particularly desirable. This was primarily a commercial and practical consideration,⁸ the importance of which became evident when the two roads came under joint management in 1903.

In 1901, acknowledging the Boston elevated road as their technical model,⁹ the engineers decided for a third rail. A third rail system adapted to the conditions of subway service was, however, difficult to design. The engineers had to determine how best to arrange the space between the third rail, the rolling stock and fixed structures, so as to allow clearance while efficiently using the limited space in the tunnels. Also needed were a suitable collecting shoe and adequate protection of the 600-volt third rail.¹⁰

[page 342] Protection of the third rail was a controversial subject. Injuries to workers along the elevated route, directly or indirectly related to the third rail, underscored the need for adequate protection in the cramped tunnels.¹¹ The final decision on protection was made in 1904, late in the subway construction timetable. Parsons, Stillwell, and electrical consultant Cary T. Hutchinson considered several methods. Parsons wanted to cut down on the use of wood in the tunnels, and suggested a concrete protection of a crossed-T form, which could serve as a footwalk for workers. Hutchinson supported this proposal. Stillwell favored a wooden covering, which he felt would also form an adequate footwalk.¹² He was impressed with the success of the third rail system of the Wilkes-Barre and Hazelton Railway, the first commercially significant installation of a protected third rail. The rail guard was a 2-inch thick pine plank above the rail allowing clearance for an over-running contact shoe.¹³

The Interborough ultimately adopted Stillwell's design of a wood plank protection.¹⁴ A 2-inch thick, 10-inch wide plank covered the rail, supported 2 5/8 inches above the rail by a timber beam, running parallel to the rail, to which the protecting plank was bolted. This left open only the side toward the running rails. The horizontal board extended beyond the edge of the rail making accidental contact with the rail difficult.¹⁵

The third rail itself, of rolled steel 4 5/8 inches high and of equal width at the base, weighed 75 pounds per yard. It contained low percentages of carbon and manganese to increase rail conductivity. Resistance was nonetheless eight times that of an equal section of copper. Granite insulating blocks, spaced nine feet apart; supported the rail, and were cemented to an iron pedestal bolted in turn to the wooden track ties. Copper bonds bridged the joints between the 60-foot lengths of rail. Stillwell chose the Mayer and Englund Company's "protected" bonds, which he had earlier installed on the Manhattan elevated. Four bonds spanned each gap, two riveted on either side, and two on the base of each rail.¹⁶

A third rail ran along each set of running rails for the length of the subway. Carbon circuit breakers connected the single-conductor, paper-insulated and lead-covered direct current feeder cables to the rail. Each track breaker was controlled electro-pneumatically from the sub-station supplying it. At the sub-station, lamps indicated the open or closed status of the breaker.

The track section between two sub-stations was divided at mid-distance, each half served by the nearer sub-station. Each station was therefore located at the mid-point of its service area.¹⁷

A flexibly arranged system of emergency switches allowed current to be cut from an entire section of the line in dire emergencies; or from a single track within a section without affecting adjacent tracks, if localized repair work were needed. The switches, located in the ticket booths and conduit manholes along the route, opened circuit breakers at the sub-station terminals of the direct current feeders.¹⁸

[page 344] Current traveled from the third rail to the car equipment by means of over-running contact shoes mounted on the motor trucks. While some railways, including the Manhattan elevated, relied on the weight of the over-running shoe to maintain effective contact with the rail, use of a protected rail demanded a slimmer, lighter shoe to fit between the plank and the conductor. The Interborough hinged the shoe to the truck and used springs to maintain proper downward pressure on the rail. Two slim prongs were made weak relative to the rest of the shoe casting. If the shoe met with an obstruction, it would break at the prongs rather than at the car supports, preventing damage to the car wiring system. Current traveled through the wiring system to the car motors.¹⁹

The choice of motor equipment under the car was determined less by physical constraints of the tunnels than by operating conditions. Although local subway service was to be similar to that of Manhattan elevated in terms of the number and spacing of stations, IRT trains were to be run less often. This demanded a more efficient acceleration, and therefore careful regulation and control of current to the motors.²⁰ The Interborough chose the General Electric-Sprague system of multiple-unit motor control. The system was in use on the Manhattan lines, but the subway installation introduced some innovative modifications.

The introduction of multiple-unit control in 1897 had been crucial to the growth of electric rapid transit systems.²¹ The system, as defined by its inventor, Frank J. Sprague, involved, "a plural control of a plurality of controllers, by which a number of units can be assembled into a train, each unit being absolutely complete without any dependence upon or relation to any other except so far as relates to control of the several main controllers."²²

Each car's equipment included the current collectors, propelling motors, and main controllers directed by the master controller from a single point on the train. Each car unit was equipped to meet only its own requirements. Sprague stressed this aspect, recognizing that the greatest possible car speed between stops was attained by the vehicle with the greatest percentage of its weight directly on its driving wheels. A locomotive represented a concentration of enough power to drag the weight of the cars behind it. With multiple-unit control, all cars of the train had the same characteristics of load capacity, motor equipment, and rate of acceleration. The multiple-unit system enhanced the advantages which the adaptable, high acceleration electric motor brought to rapid transit, surpassing the locomotive in speed and acceleration, and the single motor car in traffic capacity.²³

Both the Westinghouse Company and General Electric produced multiple-unit control systems. The Westinghouse equipment was electro-pneumatic; the power motivating the individual motor control switches was compressed air, regulated by valves in turn operated by electric circuits. The General Electric-Sprague system installed by the Interborough was entirely electrical; the motor switches were operated by a system of circuits and solenoids.²⁴

[page 345] In the standard G. E. system, the motorman controlled speed and acceleration by varying the current in the solenoids via a complex of parallel-series circuit connections between the train motors and a large resistance, composed of several small resistances which could be cut out of the circuit one by one by contactor switches. After an initial series connection of the two car motors with the car resistance, speed was increased by gradually cutting resistance out of the circuit. With all resistance out, the motors operated in series at full voltage. By switching the two motors to a parallel connection, each in series with a resistance, speed could be further increased by again decreasing the resistance. Highest speed was attained with the motors at full voltage in a parallel connection. Electro-magnets, controlling the switches for each contactor, were in turn activated by one of two small drum controllers placed at either end of the car. A master controller, replaced by the motorman, provided simultaneous control of all the train motors via a continuous low-voltage train-line extending from car to car.²⁵

The system, the standard G. E. type M control, was considered the most advanced of its day when installed on the Manhattan elevated lines in 1900.²⁶ The Interborough installation incorporated improvements over the type M control An innovative design of the master controller allowed automatic acceleration at a predetermined current while maintaining the option of manual control at lower amperage. If the motorman threw the control handle to the full-voltage position, current-limiting relays prevented excess current from entering the motors at the early stages of acceleration, only gradually allowing greater and greater current to bring the train to full speed.²⁷

Additional changes and improvements included the division of each car's resistances into two groups, each used exclusively with one of the car's two motors. A contactor switch short-circuited only one resistance rather than several. This arrangement allowed larger cross sections of the resistance grids. On the subway cars two or more contactors were mounted on the same base, instead of individually as on the Manhattan installation, saving space and avoiding the need for heavy wires to interconnect two contactors. The G. E. company also considered the subway's contactor short circuit blow-outs more efficient than those it supplied to the elevated, since they combined reduction in weight with a higher current-carrying capacity.²⁸

General Electric and Westinghouse both received contracts for motors. Before selecting specific equipment, Stillwell planned and supervised a series of tests, similar to those done during the elevated electrification. The subway tests, performed in the factories, on the G. E. experimental track in Schenectady, and on a half-mile stretch of the Manhattan els, were considered by the Street Railway Journal to be "... the most extensive and exacting that have ever been made in electric railway practice." Stillwell's tests were subsumed under 43 headings, and included examination of speed/voltage ratios, heating and cooling curves, insulation, power consumption, controllers, and performance of commutators at different amperages.²⁹

The G. E. and Westinghouse motors ultimately supplied to the Interborough differed slightly in some respects. The G. H. motor, with gear and gear case, weighed 5,900 lbs.; the Westinghouse, 5,750 lbs. The G. E. gear reduction ratio [page 346] was 19 to 63; the Westinghouse, 20 to 63. The G. E. magnet frame was unsplit; the Westinghouse was divided in half. Both, however, were specially designed to meet the rigorous specifications developed by Stillwell. The results were high capacity motors (desired speeds of express trains required performance at 325 amps and 570 volts), which could be mounted in the cramped quarters (50 inches from wheel to wheel) under the cars.³⁰ Although the Interborough employed double-truck car mounting, both car motors were installed on a single truck, the other truck being a trailer. The Interborough used 5-car trains for local service, each with three motor cars and two trailer cars, on which neither truck suspended motors. Express service used 8-car trains, with five motor cars and three trailers. The same motors and gearing were employed for both classes of service.³¹ In this way the Interborough obtained the most flexible use of its motor equipment.

* * *

Beneath every electric car was a maze of air piping and electrical circuits. Severe structural damage to the cars could result in damage to the electrical equipment, short-circuiting, and possibly fire. In a subway it was essential to minimize the risk of fire and smoke. Stillwell was convinced that wiring practice had not kept pace with advances in motor construction and control systems. He claimed that standard wiring in electric cars deteriorated after a few years of service. The design and insulation of the electrical wiring beneath the cars was therefore a key aspect of the company's goal of fire-proof car construction.³²

The Interborough introduced two types of cars into its tunnels during the first year of operation. The original contracts were for composite wood and steel cars. By February 1904, the company had ordered the innovative fire-resistant all-steel cars.³³ The designs of both cars incorporated important structural features and careful insulation of electrical wiring.

The wiring beneath the car included the propulsion circuit, heating and lighting circuits, and the motor control circuits. The first three circuits were self-contained for each car to car. Only the low voltage motor control circuits had a train length component. Also extending the length of the train was the train line pipe for the Westinghouse air brakes, the other multiple-unit system aboard the trains.³⁴

All wiring on the original shipment of composite cars was suspended beneath the cars, outside the main car bodies. The interior flooring was doubled, and enclosed a layer of asbestos rolled felt. A 1/4-inch asbestos board called "transite" sheathed the underside of the pine flooring. Both steel and asbestos separated the motor trucks from the car bodies. The exterior of the steel-framed wood-slat cars was sheathed with copper.

Wire insulation consisted of a layer of paper, succeeded by layers of rubber, weather proofed cotton, and asbestos. Only the heating and lighting circuits entered the car bodies proper; at the forward end of each car the [page 347] control circuit wires passed to a switchboard in the motorman's cab. Where wires passed through the car floor, steel chutes lined with asbestos protected the wires from mechanical injury and sealed out dust and dirt. Asbestos-lined boxes enclosed all junctions and fuses and prevented damage to wire insulation from vapors rising from the fuses.³⁵

Louis Duncan and Cary T. Hutchinson, the electrical consultants to the Rapid Transit Commission, reviewed the specifications and drawings for the composite cars early in 1904. They concluded that, due to the care taken with the wiring, the cars were "an improvement upon any car built before." The thorough insulation between the electrical systems and the car body ensured little damage to the latter in the event of an electrical fire beneath the car. However, if during a bad collision the copper sheathing came away from the car, the likelihood of splintering wood making contact with damaged wires rendered the cars as flammable as most others.³⁶

The all-steel cars contained some wood in the door posts, and a small amount was used for the interior finish around the windows. All wood was treated with a fire-proofing compound. The sides were steel sheathed, and the floor was of corrugated iron overlaid with a fire-proof 1.5-inch thick board called "monolith", placed directly on the steel under-framing, Asbestos boards also lined the ceiling and sides of the car.³⁷

Because not only the side framing but flooring was of steel, supporting the piping and wires presented new problems. A wood floor would have been strong and workable and allowed flexibility in the support arrangement, but the desirability of completely fire-proof construction precluded this. It was impossible to carry the wires within the car body for the same reason. All standard methods of support involved one of these alternatives. As the innovator in all-steel car construction, the IRT was forced to abandon standard practice in wiring support design.³⁸

The Interborough engineers adopted the practice, used in the interior wiring of buildings, of enclosing the wires in grounded metal conduits, which were either clamped to the steel underframing or passed through openings in the plates and beams of the car support structures.³⁹ The system combined the advantages of durability, strength, low maintenance cost, protection of the wiring from mechanical damage, and immediate grounding of a short circuit. A disadvantage was the increased difficulty of guarding against abrasion of insulation where wires entered or left the pipes. The insulation was the same as on the composite cars. The amount of insulation represented an attempt by the engineers to provide adequate protection while using a minimum of rubber and woven materials which would produce much smoke in case of an electrical fire.

The reduction of insulating materials depended on the design and location of the wiring system. By arranging the contactors in a box at the center of the car, with resistances in two rows on either side, the lengths of the leads, and therefore the amount of insulation minimized. Bus lines from the contact shoes were directly attached to the main propulsion line in enclosed fuses; rather than running the length of the cars. The connections from the propulsion line to the switchboard was also short. Both arrangements further reduced the length of wiring.

[page 348] The conduit pipes, of the "Loricated" design produced by the Armorite Conduit Company of Pittsburgh, were wrought iron with both interior and exterior surfaces covered with hard enamel. The enamel protected against rust and decreased the chance of abrasion to the insulation as wires were drawn into place. The use of S-shaped and elbow-bent piping lowered the number of times a wire needed to enter or leave a conduit.⁴⁰ Because of the importance of the piping, the Interborough set up a special shop in a repair shed of the elevated division. Workmen prepared templates for the various bends required, and produced a stock of interchangeable parts, greatly easing the installation process and cutting labor cost.⁴¹

On June 1, 1906, two empty trains, composed of all-steel and composite cars, collided on a storage track between 103rd and 110th Street. Although the wood portions of both types burned, filling the tunnels with the smoke, the all-steel cars fared far better than the composite, which were almost completely destroyed. Copper sheathing clearly did not effectively fireproof the cars.

As a result of the accident, the IRT company decided on steel cars. Loathe to waste its initial investment, however, the company used composite cars with the all-steel cars for several years thereafter. The value of all-steel cars impressed entrepreneurs in interurban railroading. The Pennsylvania Railroad had originally provided George Gibbs with the facilities to perfect his designs. After the subway accident, William Gibbs McAdoo, President of the New York and New Jersey Tunnel Company, decided that only all-steel cars would run in his tunnels.⁴² The Interborough innovation in car design proved an important contribution not only to urban rapid transit, but to interurban electric railroading.

Inventory of Rolling Stock

General Electric Motors: type 69, 200-hp. capacity; magnet frame unsplit; fixed coils wound with flat copper ribbon, insulated with mica and a specially prepared fire-proof and water-proof fabric; armature, series slotted drum barrel winding, copper bar conductors separately insulated, core of iron-clad type; commutator segments of hard drawn copper; gears of high grade cast steel, pinion of forged steel; gear case bolted directly to motor frame (this feature first introduced on Manhattan elevated installation); mounting, nose suspension.

Westinghouse Motors: no. 86; similar to Westinghouse "50-C" motor, 200-hp. capacity; magnet frame of cast steel split in two; field coils of copper strands wound on edge, insulated with mica and asbestos; armature, 20 inches in diameter, 1930 lbs.; slotted drum type, of sheet steel and cast iron, insulated by mica; commutator segments of rolled and hard drawn copper; gears of solid cast steel with cut teeth, pinion of forged steel with cut teeth; gear case of malleable iron.

Trucks: built by the Baldwin Locomotive Works, Phila., to designs and specifications of Engineers Gibbs and Thompson, Interborough Rapid Transit Company, embracing the Motor Car Builders Standards, incorporating latest in both steam and electric railway practice; special design of truck bolster and spring plank created space necessary for motors; motor trucks arranged for nose suspension; wheels and axles from Standard Steel works, wheels with 5 5/8 inch steel tires; height of both motor trucks and trailer trucks 30 inches, wheel base of motor trucks 6 feet, 8 inches, of trailer truck 5 feet 6 inches; diameter of motor truck axle at center 6.5 inches, at wheel seat 7 3/4 inches, trailer axle diameter at center 4 3/4 inches, at wheel seat 5 3/4 inches.

Air Brakes: Westinghouse Air Brake Company provided piping, valves, brake cylinders; shoes of "Diamond S" type supplied by American Brake Shoe and Foundry Company; Christensen Engineering Company, Mil., provided air compressors, of Christensen No. 2 independent type, motor driven, and Christensen automatic multiple Unit governor

Cab Equipment (Steel Cars): master controller and energizing switch, circuit breaker resetting switch, marker, cab heater switch.

Vestibule Switchboard: multiple-unit control cut-out switch, main motor/trolley connecting switch, lighting, heating air compressor and governor apparatus.⁴³

Signaling, Switching, and Safety

[page 356] The safe and efficient control of train movement greatly concerned the designers of the IRT. Subway engineers planned and built express tracks, alongside and parallel to the local tracks, to accommodate the high speed service. At the express stations, at the 96th Street division into the uptown east and west side lines, and at the City Hall station, complex track arrangements demanded proper coordination of train movements. High speeds, frequent service, and complex track layouts required a reliable system of automatic signaling and switching.¹

Street railways were not models for the subway signal system. High speed was not a condition of street-level service, whatever the motive power, and the cars were directed by the traffic regulations and conditions of all street traffic. No independent signal system was necessary.

The subway engineers thus turned elsewhere. The system ultimately installed was described by contemporaries as using "old and well-tried methods" and also applying "some entirely new principles."² The mystery combined innovative design with techniques and practices drawn from various classes of high speed railway service.

Steam railroad practice served as the general model for electric railway signaling systems. The conditions of subway service were identical to those on steam railroad trunk lines. An author in the Railroad Gazette noted that "train service on express tracks will be similar to that on steam roads in that the trains will run at speeds of from 35-45 miles an hour... making no stops for long distances."³ Similar conditions of service resulted in the adoption of similar methods of traffic control.

In 1900 heavy traffic on the interurban railroads was something new.⁴ English railroads handled a high rate of traffic much earlier; accordingly block signaling was introduced there in 1841.⁵ Block signaling is the division of the length of the track into blocks or sections; a signal at the entrance to each section indicates the presence or absence of a train within that section. Block signaling technology quickly developed from semaphore signal arms, hand-operated from the engine cab or by a signal man in communication via telegraph with the farther end of the block, to various methods of powered operation. Hydraulic power, water power, glycerine, or alcohol was standard for these systems.

American railroads began to adopt block signaling in the 1870's, improving on British practice by having the arm dropped rather than raised to the horizontal danger position. In this way a failure in the signal system would put all signals at danger, decreasing the likelihood of collisions. But because American roads had less traffic than the British, because hydraulic systems were unreliable, and legislatures hesitated to impose controls on private railroads, the adoption of block signaling in the United States was limited.⁶

[page 357] American automatic block signaling received a boost when in 1881 George Westinghouse formed the Union Switch and Signal Company. Westinghouse applied his experience with the air brake to the development of electro-pneumatic automatic signal. A low voltage direct current track circuit operated the valves of a compressed air system, which in turn controlled the position of the signal. Electricity traveled by wire to one rail at the head of the block, then back down the track to the signal end, where the current crossed to the other rail through the relays of the signal circuit, and returned to the head of the block. Each sectio