The technology of economical local transport in the United States

The electric railway for local transport in small-town and rural areas existed in the United States between 1890 and 1940, flourishing between 1900 and 1930. Electric railways became possible after efficient rotating electrical machines were developed for the transmission of power, based on the work of Gramme and Siemens in the 1870's, and was abandoned when individual motor transport on paved common roads prevailed after World War I. The bus services that replaced the electric cars lasted little longer, and were less extensive, except for special purposes. There is now no general public transport system in the niche once occupied by electric railways. Urban electric railways and electrified steam railways are quite a different matter, and have persisted, of course. The usual term for the local electric railways is "interurban," which is, strictly, a misnomer, since their aim was not to connect cities, but to serve the countryside. In many cases street railways were associated with interurbans, and it is hard to make a clear distinction between them. Interurbans provided not only transport, but also were the earliest sources of electric power outside of large cities. In fact, the interurban railway was quite often a subsidiary of an electrical supply enterprise.

The interurban electric railway was very often built beside steam railways; in many cases one could be seen from the other from end to end. Nevertheless, the traffic was quite different. The interurban primarily offered local passenger service, and secondarily less-than-carload package services. Carload freight was usually quite minor, except for lines in special situations that served a particular industry, like cement, or a particular commodity, like coal. Many of the interurbans that survived through the Depression and beyond were either important freight carriers, or had transformed themselves into fast or suburban passenger carriers. Interurban stops were usually around a mile apart, and cars stopped on signal at each one. Packages, mail and milk churns were handled in the car vestibules. Although the parallel steam roads were often annoyed by the interurbans, the interurbans were valuable to them as connections, and probably helped much more than they hurt. Many interurbans handled a small amount of carload traffic from steam road interchanges, handling it to sidings on their lines.

Not only did the interurban stop at closely-spaced stations, it also offered an intense, usually equal-interval service. Half-hourly and hourly services from 6 or 7 am until 10 or 11 pm were usual. In these respects, it was quite different from a steam railway, whose stops were more widely spaced, and which might offer at most five or six trains a day. Fares, however, were about the same, between 2 and 3 cents per mile. This was a considerable amount in those days, so travel was not cheap.

The primary reason for the disappearance of interurbans is that their local traffic was almost completely absorbed by the individual motor car and the local delivery truck on paved roads, which gave door-to-door service under the complete control of the individual. This was true with freight traffic no less than passengers. Previously, for example, sand might be loaded into cars somewhere out on the line and brought into town, where it could be distributed by teams. A motor truck that could make the local delivery was just as able to go the 10 miles or so out to the sand pit, so that sand could move from pit to site in one movement.

On interurbans, "express" referred to less-than-carload parcels. The term "limited" described a car that did not make the rural request stops, but only served the principal stations. Such cars were put on in an attempt to attract additional traffic that actually competed with the steam carriers. Many limited cars moved at impressive speeds. Sometimes these cars were successful, but sometimes they just angered the local customers when they didn't make local stops and drove them to their autos.

In many cases, a permanent role for the interurban as a public service could have been found. However, the fixed costs represented by the investment in land, roadway, electrification and equipment, coupled with the fact that taxes, often heavy, had to be paid on all these facilities, made the enterprise unremunerative, and so there was no incentive for private capital to undertake the job. This has remained the case, and there is now no comparable public transport that is not tax-supported and heavily subsidized, including all transit even within cities, by rail or road. There is now no rural public transport at all, private or tax-supported, except for very thin bus services on main highways.

Local electric railways are now again being built in urban areas, but with a very different character and purpose. The interurbans were inexpensive to build and inexpensive to operate. The present "light rail" is expensive to build and expensive to operate, but it actually does little more than a much cheaper system would do. One might ponder the reasons why these heavy light rail systems are overbuilt and overequipped, how they are financed, and where the money goes.


The horse omnibus, operating in city streets, was the predecessor of both street railways and interurbans. Omnibuses originated in the early 19th century, and were a common feature of small as well as large cities. Two or four horses or mules pulled a vehicle of distinctive appearance fitted with seats for the passengers and open platforms at each end for the driver. The standard horse car was 16 feet long, with two axles on a 7-foot wheelbase, and weighed about 5 tons. Stops were made on request, or at fixed locations, for passengers to alight or board. A driver handled the horses, while a conductor took the fares and requested the driver to go on or to stop. If rails were laid in the streets for the cars, one horse could do the work of four. Not only was there a considerable saving in horses, but the ride was much smoother and more comfortable. The cars themselves were modified to have two fixed axles, since the wheels no longer had to steer. They had longitudinal cane seats inside, a row of windows on each side, and platforms at each end, giving them a characteristic appearance that was preserved in the earliest electric streetcars.

In the United States, it had long been the practice for rails to be laid in city streets, over which railway cars, both passenger and freight, were transferred from station to station and from station to shipper, drawn by line teams (two to four horses in tandem). In some cases, steam locomotives were not allowed as far in town as the railway station, and the cars were drawn to a suburban location for the attachment of the locomotives. In New York, for example, locomotives were not allowed beyond 42nd street, when this was still a rather rural location. This practice declined with the direct connection of different railways in major cities after 1865, but some street running was always present, even to modern times.

On a common road, about 70 lb per ton was required to draw a wagon on the level at 4 mph, with about 20 lb per ton additional for each percent gradient. A horse had a traction of about 60 lb at 4 mph. A two-ton omnibus, then, might demand 180 lb traction, allowing for a 1% gradient, which means three horses. On a railway, the same omnibus would require only about 40 lb traction on the level, so one horse could handle the car easily. If the horse slowed to 2 mph on a gradient, he could exert about 125 lb traction, so it was quite safe to use only one horse. The speed should be noted. The amount of traction that a horse can exert drops rapidly with speed, especially if continued exertion is required, so speeds above 4 mph were uneconomical in most cases. In Hollywood, horses did not have to work for 10 hours a day. Any train can easily outdistance any horse and rider, leaving them lathered and exhausted, but this would spoil the plot.

The horse- or mule-powered street railway became very popular and widespread, especially in cities. It moved at walking speed, and was easy to use. Cable systems, with a central powerhouse and transmission by wire rope, resulted in cleaner streets but not a great increase in speed. Cable cars were confined to large cities, such as Denver and San Francisco. The large investment could not be justified in less-densely populated areas. Another alternative was the "steam dummy," simply a small steam locomotive in a housing to conceal its mechanism so horses would not be terrified. Horses were usually terrifed anyway, and those who managed horses were not pleased by steam dummy lines. A typical steam dummy was an 0-4-0 with a vertical boiler, and it pulled a typical rail omnibus. Some of these lines were intended to shift freight cars in city streets and docks, not to carry passengers. There were also pneumatic railways that used a piston in an evacuated tube for traction, but these never proved practical.

The horse railway penetrated even small towns by the 1880's when a great need was felt for local transport. Rails were light, perhaps 56 pounds per yard, and narrow gauge, say 3 feet, was quite common, since these railways were promoted at the height of the narrow-gauge fever. The mules were kept in a barn where they were supplied with hay. The cars also were garaged in the barn. This was probably the source of the name "car barn" for the building in which the later electric cars were garaged and maintained. Horse railways, except for a few curiosities, were completely gone by World War I.

Developments up to 1890

Electric propulsion of railway cars was attempted as early as there were motors of any type. Thomas Davenport of Vermont displayed a miniature electric railway in 1835, and in 1838 Robert Davidson, in Scotland, carried out full-scale experiments. C. G. Page tried a full-scale reciprocating electric locomotive on the B. & O Washington Branch in 1851. All of these attempts were either of miniature railways that would not scale to full size, or unsuccessful full-size experiments. Early, inefficient motors and primary batteries were quite unequal to the task.

Efficient, powerful motors and dynamos were first designed by Gramme and Siemens in the 1870's, who recognized the importance of a steady back-emf and good magnetic circuits. Early dynamos and motors were the same machine, which could be used in either role by merely resetting the brushes that made contact with the commutator. Siemens and Halske demonstrated a small electric locomotive on a 300-metre long circular track at the Berlin Industrial Exhibition in 1879, and constructed the first commercial electric street railway in Lichterfelde in 1881. In 1883, William Siemens installed an electric tramway at Portrush, in Ireland.

Stephen D. Field, in Stockbridge, Massachusetts, and Thomas A. Edison, in Menlo Park, New Jersey made similar demonstrations of experimental electric traction as early as 1880. Charles J. van Depoele demonstrated an electric train at the Chicago State Fair in 1883. In the same year, Leo Daft experimented on the Saratoga and Mt. McGregor railway. Bentley and Knight constructed a short line in Cleveland, Ohio, and J. C. Henry established the Westport Electric Railway in Kansas City, Missouri, in 1884. In 1885, Daft equipped the Baltimore Union Passenger Railway, and van Depoele a line in Toronto. These were all short, experimental lines. By 1886, there were only 8 miles of such lines in the United States.

Siemens and Halske, and Daft, used a third rail for current supply, and sometimes a fourth rail for the return. J. C. Henry used a two-wire overhead, a system still used for trolleybuses that have no return path through rails. Bentley and Knight made contact in a conduit, a system that was later developed and used where there was an insuperable objection to overhead wires. Van Depoele used an overhead contact wire with running-rail return, however, the system that was to prevail except on rapid-transit systems in tunnels and viaducts, where third-rail could be used. Many early cars were converted from horse-railway cars. Van Depoele put the motor on the car floor, and connected it with the wheels by sprocket chains.

Frank J. Sprague entered the arena with experiments on electric traction for the New York Elevated Railway in 1886. Sprague introduced axle-hung motors beneath the car floor, the trolley pole and wheel, series-parallel switches, and the multiple-unit control with a train line. In 1888, he equipped a short line at St. Joseph, Missouri, and later the same year, significantly, the 13-mile, 20-car Union Passenger Railway of Richmond, Virginia. Sprague was the major contributor to the developing technology of the electric railway.

Sidney Howe Short (1858-1902) was a college-trained scientist from Columbus, Ohio, who became a telegraph operator at age 14. From 1880 to 1885 he was Vice President of the University of Denver and Professor of Physics and Chemistry. The Short Electric Railway Company of Denver was an experiment in serial feed using a constant-current source instead of the usual constant-voltage source. Here, the circuit had to be kept closed (Short-ed!) and was opened to put another motor on line. This arrangement did not prove to be practical. In addition to the sliding contact, Short also invented a double-reduction motor, and a gearless (axle-mounted) motor, among other things. He used conduit to begin, then turned to overhead contact. In 1889, he formed the Short Electric Company in Cleveland, Ohio.

Sprague formed the Sprague Electric Railway and Motor Company, which was taken over by the Edison General Electric Company (later General Electric). The Bentley-Knight and van Depoele patents were acquired by the Thomson-Houston company. Short invented the contact shoe with its soft metal insert, and founded the Short Electric Company of Cleveland, which manufactured traction motors. Westinghouse also entered the field, and later championed high-voltage AC electrification. In 1892 Thomson-Houston gathered the Brush Electrical Company, the Short Electric Railway Company, and Edison Electric Company under its wing and formed the General Electric Company. These were the principal companies active in the early development of electrical railways in the United States.

At the beginning of 1888, there were 13 electric railways in the U.S., with 95 motor cars and 48 miles of track. Of these, 6 were by van Depoele, 3 by Daft, including the Asbury Park line, 1 by Fisher, 1 by Short (probably the experimental series-supply Denver line), 1 by Henry, and 1 by Sprague, the St. Joseph line. In 1889, there were 805 miles and 2800 cars; in 1899, 17,685 miles and 58,569 cars; and by 1909, there were some 40,000 miles of electric railway. This was an amazingly rapid technological advance. In less than a decade after the first experiments in 1880, the technology was approaching maturity, and the standard practice was established.

Electrical Matters

As soon as the magnetic effects of an electric current were demonstrated in 1820, electricians began trying to produce motion from electricity, and soon succeeded. The first electric motor was devised about 1835, and a great future was predicted for the new power. Then, every attempt to make a practical motor that could do something useful, like run a printing press or drive a locomotive, failed, and failed badly. Scientists proved in the 1840's that electric motors were, in fact, impossible because they could not be made efficient enough to avoid burning up when doing useful work. As toys, they were fine, and examples are still sold for educational purposes, but as serious sources of power they were useless. Electric motors became something like perpetual motion. An occasional crank would come up with one, but it would fail on closer examination. It was not as difficult to produce electricity, so generators driven by steam engines and water wheels were made that supplied arc lights and electroplating establishments. The generators ran hot, but they worked and supplied enough power.

Not until the 1870's did Gramme in France and Siemens in Germany discover the beauties of back-emf. With ring (Gramme) or drum (Siemens) armatures, this permitted rotating machines with efficiencies above 90%. Now it was finally practical to transmit and subdivide electrical power. At first the machines were large, but the way to make efficient smaller machines was soon found. By the turn of the century, even fractional-horsepower motors were practical, and electric power became universal. The direct-current traction motor proved to be as well-adapted to railway traction as the steam locomotive, giving high torque at low speeds and not requiring elaborate transmission mechanisms. The alternating-current traction motor, with similar characteristics, was also known, and was applied early to railway traction in Germany. The great benefit of alternating current is the ability to transform voltages, so that a high voltage can be used for efficient transmission (at low currents), which is easily stepped down for use in motors, whose insulation cannot stand high voltages. However, these early alternating-current motors were quite large and bulky. Americans chose the smaller direct-current motors that could be mounted beneath the car floor. These motors operated on 600 V, and were started by resistances.

Before about 1915, electric mains power was available only in cities and larger towns, and sometimes only in the business districts, though it was spreading rapidly. Interurbans had to generate their own electricity, usually in coal-fired power stations. In some places, natural gas or hydropower was available. A street railway generated power at 600 V DC for direct supply to the overhead wire, and early interurbans did the same. A range of about 5 miles from the power station was the effective limit, so only a 10-mile interurban line could be powered from one station, located in the middle. Beyond this distance, voltage drops in the resistance of the feeders and trolley wire was excessive. For greater distances, additional power stations were necessary. A station might have an efficient Corliss engine, and a large compound-wound dynamo that would compensate a heavy load by raising the voltage somewhat.

Power can be efficiently transmitted over greater distances by using higher voltages. The current decreases in inverse proportion to the voltage, and the voltage drop is proportional to the current. A line voltage of about 1000 V per mile of transmission line was a rule of thumb. Changing DC voltages is difficult. Rotating voltage converters were available, but were expensive and could not accommodate voltages above about 6.6 kV. However, DC could be generated at this voltage, and stepped down to 600 V at substations along the line. This would make 100-mile lines possible with one power station and 10 substations with rotary converters.

A better way was to use AC transmission. AC voltages can be changed by transformers, which have no moving parts, require little maintenance, and are much cheaper than rotating machines. The electricity is generated by alternators at about 6.6 kV, then stepped up to, say, 26.4 kV for transmission in cities and 132 kV in rural areas. This was the method used for electrifying the country, using 60 Hz AC that was easily stepped down to the 240/120 V household supply. AC transmission dominated electricity supply from the first, leaving only relics of 110 V DC in the centers of large cities. In many cases, the interurban railway was part of an electric supply utility, which generated large amounts of electricity in central power stations.

Alternating currents never made much progress with street railways and interurbans, except for power distribution to substations. Good AC series motors were not available until after 1900, and even then required 25 Hz current, not the usual 60 Hz. This meant special power stations and transmission systems. The Denver & Interurban Railway (Denver-Boulder) adopted 11 kV, 25 Hz around 1907 for country running, but used 550 V DC in cities. These heavy cars had a pantograph for the AC, and a trolley pole for the DC. The Westinghouse traction motors could operate on either AC or DC. The use of high-voltage AC completely eliminated feeders. The current was fed to the whole 33-mile line from one intermediate point through the catenary.

The power system of a typical interuban line is shown in the diagram at the right. High-voltage AC is supplied to substations at roughly 10-mile intervals along the line. The conversion to DC could be made by (a) mercury-arc rectifiers, (b) induction- or synchronous-motor driven generators, or (c) rotary converters. Mercury-arc rectifiers were not available at first, solid-state rectifiers were not available in this period, and anyway rectifiers were not as rugged as the service demanded. Motor-generator sets were easy to use, but two machines were required, which made them not only costly, but not highly efficient (only about 90% with individual machines 95% efficient). The rotary converter was the machine of choice until quite recently. Because it is an unusual device, it will be described in some detail here for the technically interested.

A rotary converter was quite similar to a compound DC generator, with shunt and series field windings, and a rotating armature with commutator and brushes on the DC side. It also had slip rings, like an AC synchronous motor, connected to the armature windings, where polyphase AC power could be supplied. There was only one field, supplied from the DC side, and one armature winding. When it was running sychronously, that is, at a speed where the rotating field produced by the AC was stationary relative to the field windings, any DC current supplied was matched by the incoming AC current. The rotation served only to make connections via commutator and brushes to keep the DC current in the same direction. AC and DC voltages were opposite in the armature, so only a rather small current flowed in the individual windings. The slight difference current would create the torque necessary to keep the armature turning, and to keep it in synchronism. The converter was supplied from a transformer which changed the high-voltage 3-phase supply to a low-voltage 6-phase supply at a voltage somewhat less than 600 V. These machines took the heavy overloads and occasional short circuits of electric railway operation in their stride. A typical General Electric 1000 kW 8-pole rotary converter for 60 Hz ran at 900 rpm with an efficiency of 95% at full load. Such a converter weighed 20 500 lb.

Starting a rotary converter was not easy. One generally had to start it on lower-voltage transformer taps, raising the brushes to prevent sparks, and connecting the field windings so excess voltages would not be induced. Even then, it was tricky to get the converter to come up with the desired polarity. Sometimes, the DC side could be used as a motor to get the armature up to a speed where the AC could safely be connected. A storage battery or another DC source was necessary in this case. Rotary converters were more subject than synchronous motors to "hunting," or oscillatory variations in rotating speed, so they had special damping windings on the field poles. The output voltage was adjusted by changing the field excitation of a "synchronous booster" armature mounted on the same shaft with its own field windings. This booster alternator was connected in series with the AC supply. An earlier method of voltage control, the "split pole" was not as satisfactory. It was very difficult to do the obvious thing and change the transformer taps under load.

Mercury-arc rectifiers became popular towards the end of the interurban era, and were used on many heavy railway electrification projects, including suburban electrification. They had no moving parts, and were uncomplicated to start, unlike rotary converters. Much later, solid state rectifiers replaced mercury-arc and ignitron rectifiers. Solid-state electronics also made possible "chopper" control of DC traction motors, which gives essentially loss-free variable voltage supply to the motors, eliminating starting resistances and providing flexible speed control. Most modern railway electrification uses AC traction power, which is either used directly in AC traction motors, or rectified on the vehicle for use in DC traction motors.

The overhead electric supply system was called the "line." The contact or trolley wire was suspended from 18 to 22 feet above the rail; 19 or 20 feet was usual. It was made from solid hard-drawn copper, and a popular size was #0 AWG, 0.365" diameter or 106,000 CM (circular mils, the square of the diameter in thousandths of an inch), and a resistance of 0.528 ohms per mile. As higher horsepower was introduced, the increased current caused larger trolley wire to be used. The traction circuit was completed from the contact wire through the trolley pole and the motorman's platform controller to the traction motors beneath the car, and then through the wheels to the rails. The trolley pole was typically 12 feet long, with a brass wheel of 5" to 6" diameter at the top to make contact, and a spring base to press the wheel against the wire while allowing it to move side to side. The pole made an angle of about 30° with the vertical, and was raised and lowered by a cord. The light, flexible trolley pole made reasonable speeds possible without the expensive catenary support required when a pantograph is used. An alternative to the trolley wheel was the slide shoe, with a soft-metal or graphite insert. Both trolley wheels and shoes were used throughout.

The contact wire consisted of sections insulated from each other, so that trouble would be isolated. Sometimes switches were provided to connect across the insulators when necessary. The wire was fed at intervals from low-resistance feeder cables running beside the track. Switches allowed the sections to be isolated from the supply when necessary. Some money could be saved by eliminating the feeder, and letting the contact wire serve as the feeder as well. It was typical practice to serve a ten-mile length of line from a substation in the middle. If the solid contact wire was, say #0000 (4/0) gauge, about a half-inch in diameter, (the largest solid wire used) its resistance was about 0.259 ohms per mile, or 1.29 ohms in the maximum five-mile distance. If a car drew, say, 100 A, which would provide a power of about 80 hp, the voltage drop would be 129 V, bringing line voltage down by 22%. These figures give some idea of the problem, which can be solved by more copper (in feeders) or more substations, both costly alternatives. The typical nominal line voltage was usually 600 V, but it ranged from 500 V to 750 V. The higher voltages protected against voltage sags. The synchronous booster, as mentioned above, could be used to adjust the voltage when necessary. The polarity of the supply was arbitrary, when storage batteries were not used. Either the contact wire or the rails could be positive. Usually, the contact wire was positive.

Feeders were made from bundles of wires, usually with an insulating sheath. They were specified as of so many circular mils. A 1,000,000 cm feeder was made of 61 #8 wires, and had a resistance of 0.057 ohms per mile. It also represented over 8 tons of copper per mile, not a negligible expense. The stranded cable was much easier to handle than a solid conductor would have been, and was less subject to cracks and other mechanical difficulties. Incidentally, at 60°C (operating temperature), the resistivity of copper is about 1 ohm per circular-mil inch. Just take the length in inches, and divide by the circular mils, to find the resistance of a wire.

The contact wire was suspended from iron span wires attached to clamps around poles at each side of the track. The poles could be about 28 ft long, 7" diameter at the top, 9" at the bottom, set at least 6 feet in the ground slightly inclined outwards so they would be pulled vertical by the tension of the span wires. 120 ft was a common pole spacing, and this was reduced on curves. The contact wire was clamped or soldered to ears fixed in insulators attached to the span wires. Alternatively, the span wires could be insulated at their ends. Current was fed to the contact wire through a live span wire. On curves, "pull-overs" were ears that could hold the wire in position between spans. Insulating joints or splicing ears were applied at span wires. The wire had to be anchored every 500 to 1000 feet against longitudinal motion by four insulated guy wires angling out to the poles on each side. The line was normally over the center of the track, but was displaced slightly inwards on curves. The longitudinal tension in the contact wire helps to hold it in place as well as eliminating sag as far as possible.

At turnouts, "frogs" were used in the contact wire to guide the trolley wheel to the proper side. These were placed beyond the switch points, slightly to the side of divergence. When correctly placed, the trolley wheel would follow the intended direction. Similar frogs were used at crossings. The use of these frogs meant that a pantograph could not be used for current collection. If a pantograph, or similar broad sliding contact is used, the bottom of the wire must be smooth. Also, in this case the wire usually zig-zags from side to side to avoid wearing a groove in the pantograph collector. When a pantograph is used, the contact wire is usually suspended from a messenger, or catenary, wire at intervals of 10 feet or so to keep it accurately level instead of sagging between supports forty yards apart. The messenger wire offers added conducting cross-section, which may be welcome. The whole overhead structure is usually called the catenary, not just the suspension wire.

Instead of two poles and a span wire, one pole and a guyed rod can be used to suspend the contact wire. The wire can be insulated at the ears, or else the rod and guy wire can be "hot" and insulated from the pole. Another feature that was sometimes required in towns was a grounded "guard wire" suspended above the contact wire. The purpose of this wire was to keep any foreign wires falling on the overhead from touching the live trolley wire. Many such wires could be telephone wires, and the subscribers might be shocked to discover 600 V on their instruments.

Although steel has more than six times the resistivity of copper, there is enough cross-section in the rails to make a low-resistance return path for the traction current. Even so, careful and effective rail bonding is necessary. The moist earth also forms a return path, and there were some early attempts to take advantage of this by explicitly grounding the rails. These currents also took advantage of iron water and gas pipes, corroding them where the current left these adventitious conductors. More serious was the effect on some early telephone companies that used ground-return telephone circuits, and suits for damage resulted. Ground returns make exceedingly bad telephone circuits, and soon the general use of metallic circuits removed this opposition. Occasionally, ground cables were used to provide a return path in copper.

Traction motors are the key part of the electrical machinery. The Siemens dynamo used as a traction motor in 1879 was about 3 hp. The first traction motors were 2-pole machines. An early Sprague traction motor had 7.5 hp, and two of them could successfully power a 16-foot horse car conversion. Then, 15 hp motors gave better performance. These motors rotated at 350-450 rpm at ordinary speeds, and were connected with the wheels in a variety of ways. The efficiency of these motors was about 65%, so they heated easily. Two motors per car were preferred, so two-truck cars had one axle powered on each truck. These motors were connected permanently in parallel. Special trucks were made that put most of the load on one axle for maximum traction, and were called "maximum traction" trucks, unsurprisingly. They can be recognized by the unequal sizes of the wheels on the two axles. Single-reduction gearing replaced double-reduction as motors became more capable, giving higher efficiency as well as better speed. Most motors came to be mounted the Sprague way, one end on the axle, the other with a spring connection to the truck or body frame.

Later traction motors were 4-pole, or even 6-pole, with commutating poles to reduce sparking at the commutator. They might be rated at between 25 and 75 hp, with 35 and 40 hp quite usual. Two or four motors per car were applied. The efficiency of a traction motor was now about 85%, electrical input to mechanical output, at full load. The power-to-weight ratio of a motor car could not well be much less than 3 hp per ton, but 5 hp per ton would do for light and slow work. Later, cars would have up to 10 hp per ton, for higher speeds and quicker acceleration, and perhaps for pulling trailers. The motors were hung on the axles, which they drove through a single-reduction spur gear set on one end. Noses, or lugs, on the other end of the motor frame were attached to the truck frame with sprung mounts. Thus, only part of the motor weight rested directly on the axle and so was unsprung. The motors were series motors, meaning that the current passed through the stationary field windings to produce the magnetic field, then through the moving armature windings where the magnetic field produced the torque. The torque is proportional to the field and to the current, and so to the square of the current. When the armature is still, the current and torque are at a maximum. As the armature speeds up, the movement creates a back-emf that opposes the applied emf (voltage) and reduces the current. If a motor like this is breaks away from its load while connected to the line, it speeds up to destruction. As it accelerates, less current flows, which means less magnetic field, which means that it must go even faster to create the back emf necessary to keep the current down. A series traction motor has no natural speed; its speed is controlled by the load.

Traction motor ratings are determined by the rate at which heat can be dissipated. Motors of smaller rating can be cooled by fans on the armature, and this was the normal method. Larger motors, of 100 hp or more, require forced air cooling by traction motor blowers.

The current to a 25 hp traction motor under rated load was about 30 A. A single car could draw from about 60 A to 360 A under full load, depending on whether it had two 25 hp motors, or four 75 hp motors. On starting, the motors could not simply be thrown across the line, since their low resistance would allow currents even greater than those just mentioned. A resistance was placed in series to limit the current to the value considered necessary to start and accelerate the car, a value in the ranges quoted above. A somewhat higher current than the rated current was permissible when starting with a cold motor. The danger was always of overheating the motor, which would destroy the insulation. As the car accelerated, resistance was removed to keep the current up, and finally was switched out altogether for continuous running. Then, the motors could be connected or disconnected as required to maintain the desired speed. Like all electrical machines, traction motors are limited by temperature rise, which determines their power rating. Too high a temperature, especially in the armature, destroys the insulation. Inductive surges can also damage insulation by the high voltages, as when the brushes lift while carrying heavy current (flashover).

To reverse a traction motor, the polarity of either the field or the armature windings must be reversed. If you just reverse the polarity to a series motor, both field and armature are reversed, and the motor continues to turn in the same direction. Another problem is that a motor carrying a heavy armature current must have the brushes set forward in the direction of rotation to avoid damaging sparking. When this is done, the motor can only be used in this direction. The answer for traction motors was the addition of commutating poles between the field poles to create the same effect electrically. When the motor is reversed, the action of these poles reverses as well, and it is not necessary to move the brushes. Early traction motors without commutating poles could not be reversed, and the cars could move forward only.

If the car had two motors, they could be connected in series to start, since more than enough voltage was available, so the current could do double duty. When the starting resistance was out of the circuit, each motor would be on half voltage. Now the motors could be reconnected in parallel, each separately across the line, and the starting resistances brought back in to control the current. When the resistances were cut out, each motor would now be across the line separately, and could run at its maximum speed. This change in motor connections is called transition in diesel-electric locomotive parlance. If the car, running in parallel, is slowed by a heavy grade, the current will increase and the motors must be returned to series connection to avoid overheating. A motor field could be shunted by connecting a resistor across it, thereby weakening the field. This would reduce the back emf, and allow the motor to speed up (at low speeds a weakened field would allow too much armature current to flow). The switching of the starting resistances, and the transition connections, were carried out in the motorman's controller, which was a rotary drum switch, called a platform controller. At low currents, this could be done by simple contact fingers. A four-motor car usually had the motors in each truck connected permanently in parallel, and a similar transition scheme was used.

It was unsatisfactory to run the heavy currents for powerful traction motors through a platform controller. Therefore, it was arranged that the motor and resistance connections were made and broken by sturdy relays, called contactors. The relay solenoids were operated from a six-wire train line that controlled reversing, transition and starting contactor functions. The currents in these wires, about 2.5 A, were easily managed by a platform controller. The starting resistances could automatically be cut out as the motor current decreased with increasing speed. This automatic control gave smoother acceleration, and avoided circuit-breaker operation that sometimes resulted from clumsy manual control. The motorman had only to move the control handle to the starting notch, and the motors automatically came up to speed. Frank Sprague invented this system, which was called multiple-unit (MU) control, where one motorman could control several cars in a train. It was necessary on the urban rapid-transit systems where it found its main application, but many interurbans ran only single cars.

It is possible to run a DC motor as a generator. When the resulting electrical power is dissipated in resistors, the torque required to turn the motors as they deliver power can aid the braking of a car. This is called dynamic braking, and is of considerable use on diesel-electric freight locomotives. It requires contactors and controls to excite the motor fields, and to connect the resistors (starting resistors can be used) to the armature. The only economical effect of this is to save brake shoes and wheel heating, which was of little interest to the interurban. Alternatively, the electric power could be returned to the overhead, which is called regenerative braking. This also requires rather complicated controls. Regenerative braking was used on some DC-electrified heavy railways (it cannot be used on AC electrification). Dynamic or regenerative braking does not seem to have been used on interurbans.

Lightning was a great hazard both to power-station equipment and traction motors. Dynamos were directly connected to the ground returns and rails that virtually encouraged lightning. The trolley wire was always available for a lightning stroke. Actually, a direct lightning strike is always very damaging, and its effects cannot be effectively counteracted. Much more common, however, are inductive effects from a nearby strike, or even storm clouds, that produce severe earth currents. A lightning protector was always used at the trolley pole or power station, consisting of a gap that could easily break down on a high voltage pulse, and a low-impedance path to ground. Sometimes an inductance was placed in the lead to the dynamo to discourage such pulses. A metal frame for the power house was also a help. Later, all-metal cars must also have helped lightning protection. A high-voltage pulse could break down insulation, especially in low-impedance armatures, and quickly disable a traction motor.

Car Performance

It is not difficult to work out how interurban cars performed on the road from their specifications. This exercise will bring out some of the realities of operating a local service, and show what is possible. Let us assume a 30-ton car with four 40 hp traction motors, a typical large steel interurban. An efficiency of 85%, which is typical, would give 135 hp at the rail. This would produce a tractive effort of 375 x 135 hp / V pounds, where V is the speed in mph.

The resistance of the car on a level road can be estimated at 5.3 + 0.04V + 0.006V2 pounds per ton. The term proportional to V, which accounts for flexibility of the track and shocks, has been increased by about a third over the usual Davis formula to account for the lighter interurban road. The air-resistance term includes the front-and-back effects, which are important for the single car. The gradient resistance is 20 lb/ton per percent grade.

Let us assume a partial load of 2.5 tons, making the total weight 32.5 tons. This makes the resistance R = 172 + 1.3V + 0.20V2 pounds, with an additional 650 pounds for each percent of grade. The maximum adhesion at 20% (good conditions) will be 13 000 pounds. An addition of 5% is made to the mass of the car to allow for rotating mass (wheels, motor armatures) that has to be accelerated along with the car.

The first thing to be investigated is how rapidly the car can stop. Using all of the adhesion (100% braking ratio) in an emergency stop, the deceleration would be 6.4 ft/sec2, or 4.4 mph/sec. This would bring the car to rest from 50 mph in 420 ft. If the brakes require 2 sec to become effective, this adds another 147 ft, for a total of 567 ft. The single car means the brakes act quickly, and a high braking ratio can be used. A more comfortable service stop could be made at a deceleration of, say, 2 mph/sec. At this rate, a stop from 38 mph can be made in 639 ft; 54 mph, 1214 ft; 66 mph, 1778 ft; 76 mph, 2335 ft, and from 85 mph, 2889 ft (a bit over half a mile). The usual 2 sec is added in each case. Gradients can be allowed for by adding or subtracting the grade resistance to the decelerating force, as appropriate.

When the car is starting, the starting resistances are in the circuit to limit the current and the tractive effort. At a speed of 3.9 mph, the available tractive effort from the motors equals the adhesion. At a lower speed, the wheels would simply slip. If full power were applied at this point, the acceleration would be 4.4 mph/sec, the same as in the emergency stop. Since this is uncomfortably rapid acceleration, some lower value, such as 2 mph/sec, would be used (by keeping some of the starting resistance in the circuit). This value could be maintained up to about 5 mph, when full power would give 2 mph/sec. Acceleration drops rapidly as the speed increases, both because the tractive effort is inversely proportional to the speed, and because the resistance begins to increase. At 10 mph, the acceleration is 1.6 mph/sec; at 20 mph, 0.73 mph/sec; at 30 mph, 0.42 mph/sec; at 40 mph, 0.23 mph/sec, and at 50 mph, only 0.09 mph/sec. The balancing speed, when maximum tractive effort equals the resistance, is about 56 mph. On a 1% grade, a speed of about 40 mph can be sustained, and on a 2% grade, 30 mph.

If speed is plotted against distance, initial acceleration is rapid, and the speed rises to over 30 mph in the first 0.2 mile. A speed of 42 mph is reached in half a mile, and in a mile the speed would be over 50 mph. The rapid acceleration at low speeds is one of the advantages of electric traction, and a great aid if stops are frequent. Let's find out how quickly a car can run, if it stops every mile. If we assume maximum acceleration, followed by braking at 2 mph/sec, the car accelerates for about 0.8 mile, and brakes for the remaining 0.2 mile. The acceleration occupies 101 sec, and the braking 25 sec, for a total of 126 sec, say two minutes running time. If the length of the stop is one minute, this means that it takes 3 minutes to cover a mile, an average of 20 mph. The average speed of movement, however, is 30 mph, and the top speed is 50 mph. This shows how much stops slow down a schedule. On the Arkansas Valley Interurban, the Wichita-Hutchinson schedule was 2 hours for the 52 miles.

We can conclude from our figures that the average interurban car was economical and well-proportioned to its duty. It used only about 3 hp per passenger at 50 mph, which even at the current domestic rate of $0.08 per kWh is only 0.36 cents per mile for "fuel." I pay 15 times as much in my economical Toyota! The maintenance costs of one interurban car (and its track) is no doubt quite a bit less than that of several hundred automobiles, as well (one for each passenger that rides regularly).

An early estimate of the costs (1890) of running a 16-foot, two-man car gives wages at 4.5 cents per mile, power 1.35 cents, electrical maintenance of car 1.0 cent, maintenance of way about 1.0 cent each, line maintenance 0.43 cents, mechanical maintenance of car 0.2 cents, and 2.25 cents for general expenses and accidents, or a total of 11.33 cents per mile. The average pay of motormen and conductors was 18 cents an hour, or $1.44 per day. Track and line cost $12,900 per mile.

Rolling Stock

The individually powered, passenger-carrying car greatly predominated on interurbans. Most common were two-truck cars with two or four motors, resembling steam railway passenger cars. Smaller four-wheeled cars, called "single-truck" though they had no truck at all, but axles fixed in the car frame, were used for lighter duties and streetcar routes. Nonpowered cars of each type, called "trailers," were sometimes used behind power cars, but many companies used no passenger trailers at all. Cars ranged in length from 16 ft for a single truck, to 45 ft and even longer, with 32 ft or 40 ft being common. Electric railway cars were normally a little over 8 feet wide. A typical car would seat 48 or 50 passengers, and carry twice as many with standees. Early cars were all-wood, like the horse cars, with hand brakes and quite light. Steel underframes were introduced early on. After 1916, steel underframes were required for interstate service, but by that time most cars had steel underframes anyway. Later, all-steel cars became standard, as on steam railways. The weight of the cars increased with increasing use of steel, until 30 tons was not unusual. The North Eastern Oklahoma Railway had some 42' 8" steel cars seating 38, with 4 x 50 hp traction motors. The AC/DC Denver & Interurban cars of 1908 were 55' 6" steel-frame cars seating 58, with 4 x 125 hp traction motors, and weighing 60 tons. A typical street railway car was 38' long (body), 8' 3" wide, and 12' 3" high, seated 40, weighed about 24 tons, and had 4 x 40 hp traction motors. Wheels were typically 33" diameter.

The lighting circuit was usually taken directly off the trolley, with its own fuse and switch. The main current went through a main switch and fuse box. The lights could be used to detect when the overhead was energized, in case of a power failure, with the main switch open. If electric heating was used, this was on its own circuit, as was the air compressor for the brakes. Cars did not usually have storage batteries for power when the trolley was down.

The external appearance of the many types of cars that were used can easily be seen in the many photographs that are available. Unfortunately, photographs give little indication of the interior appearance and arrangement or of the electrical equipment. Cars were often painted a conservative Pullman green at first, but it was noted that the cars were not easily noticed. This was an important problem, since the interurban ran through streets, crossed many roads, and was used as a footpath. Later cars were a brighter color, with red, yellow and white frequently seen. Yellow, in fact, became typical.

There was generally no provision for providing sand to improve traction. The rotation of the trucks as the car rounded a tight curve made it difficult to provide flexible connections for sand in body-mounted boxes, and there was little room on the trucks for individual sand boxes.

Some leading manufacturers of interurban cars were J. G. Brill of Philadelphia, American Car Company and St. Louis Car Company of St. Louis, Cincinnati Car Company, and Jewett Car Company. Electrical equipment was by Westinghouse, General Electric and Reliant, among others.

Smaller cars consisted of one compartment with a motorman's station at each end. The seating could be longitudinal cane benches as familiar from streetcars, but cushioned steam-railway coach seats were more usual for the longer interurban journeys. Passengers entered by steps at each end. Larger cars might have a separate motorman's compartment, often combined with a parcels and baggage area with a wider door, a smoking compartment, and a general or ladies' compartment. Entry could be by a center door, with smoking compartment to one side and the ladies' compartment to the other, especially on the longer cars. A single-ended car had a controller at one end only, while a double-ended car had controllers at each end, so that it could be driven from either end. To reverse the car, all that was necessary was to raise the proper trolley pole (if the car had two) or to rotate a single pole by 180°, and carry the control and brake handles from one end of the car to the other. A single-ended car had to be turned on a loop or wye track at the terminus. (turntables were rare on interurbans).

An essential piece of rolling stock was the line car, used for maintenance of the overhead wire, and sometimes for general maintenance of way. One could be bought new specially built for the purpose, but often a line car was improvised from an old passenger car. A necessary addition was a roof platform, and ladder access to it, for wire maintenance. The car could pull a gondola or flat car with ballast or rails as required, and generally be of service along the line. Of course, it would carry the gangs that did the work, as well as their tools and material. A line car could have a short deck or gondola section at one end.

Few interurbans had snowplows or other equipment for dealing with snow and ice, relying on manpower alone. One useful unit sometimes found, however, was a rotary brush or sweeper to clear snow from the track. The brush was mounted at an angle so that it would brush snow away from the track and switches to one side or the other. A separate traction motor was provided to operate the brushes.

An express motor was a relatively heavy car with a wide center door for loading and unloading parcels. If specially built and not converted from a passenger car, it would have no windows on the sides except for the motorman. It could operate alone, or could pull a few standard freight cars. Mail bags were generally handled on the platforms of passenger cars.

If there was significant carload traffic, a locomotive was required. Both General Electric and Westinghouse offered two-truck locomotives, perhaps weighing 50 tons and having four 100 hp traction motors, such as the familiar Baldwin-Westinghouse Class D. These had a cab body with short inclined housings at each end, and could haul considerable trains.

The first electric cars had hand brakes only, like the horse cars that they replaced. They were mainly used to avoid running over the horse in horse-car days, or for tying down the car at a stop, and so were quite adequate. Electric cars ran much faster than horse cars, and at 40 mph a hand brake was not all that effective. The air brake was soon adopted, with a motor-driven compressor. A single car could use a straight air brake, with its rapid application and easy control. This made electric cars much nimbler than a steam train, and had an influence on the methods of operation. Where MU operation was the norm, automatic air brakes were, of course, applied. The pre-1900 cars were often all-wood, except for the trucks and other parts that had to be of metal. This made the cars light, only a few tons, and cheap to operate. Cars were from 40 to 50 ft in length, and of great variety. Some had only two axles, but double-truck cars became the rule. Couplings were unnecessary for single-car operation, and photographs show few. Where something was needed, a draw-bar or a link-and-pin coupling generally sufficed. Knuckle (MCB) couplers were rare, except on heavy lines with MU operation. Because of street-running, cars were fitted with "fenders" or "cow-catchers" that mainly prevented horses from going under the body and derailing the car. Sometimes there were merely bars on the front of the trucks, but often nothing. The horizontal bars seen across the end of the car floor are the anticlimber that prevents one car frame from slicing into another car frame, specially regrettable when the car bodies are of wood. Locomotives and cars equipped for MU operation can be distinguished by the circular receptacles for the MU connections at their ends.

Early cars were heated by stoves, then many had steam heating from coal-fed boilers in the cars. Later cars generally used the safer and less bothersome electric heating. The disadvantage of electric heating was that it went off when the power went off, making a stranded car in cold weather rather uncomfortable. Of course, the cars had electric lights, including an electric headlight. A location above the front window was most effective for a headlight, but they were usually below the front window so they were easy to reach to change the bulb. Cars often did not have those comforts of a steam train, the toilet and the water cooler. Later heavy interurbans included these features, of course. The motorman could attract attention by a foot-operated gong or bell. After air brakes were fitted, an air whistle was possible. The gong was used while running in streets, the whistle out on the line. These whistles were shrill and high-pitched, which, after all, was most effective. Inside the cars, advertising cards in a row above the windows were a common feature.

Maintenance of interurban cars, aside from the normal maintenance of railway rolling stock and air brakes, includes electrical maintenance. This consists of replacing brushes, keeping contacts clean and unpitted, and cutting down the mica between commutator segments. A high-potential test with a megger would reveal bad insulation and allow it to be corrected before it caused trouble. At long intervals, commutators must be turned on a lathe, and windings with bad insulation replaced. Most parts are obtained from the manufacturer, and need only be installed. That is to say, maintenance is slight compared to the upkeep of a steam engine or a mule, and does not require heavy, expensive equipment. This ease of upkeep is a strong recommendation for electric traction, in addition to its suitability for the job and its ready availability. Many interurban lines also had to operate and maintain a power house, with its boilers, steam engine, and alternators, but this was all indoor work. All the equipment of an electric railway is rather long-lived and reliable.

Structures and Road

Power houses and substations were essential structures. They were often built of brick, stone or concrete, and so were durable. Many remained long after their reason for existence disappeared, converted to other uses.

The car house or barn was another typical structure, usually larger than the power houses and substations, and distinguished by large doors where the cars entered. Cleaning and maintenance was carried out here, as well as the storage of rolling stock not in use. Electric cars required relatively little maintenance, and the heavy equipment necessary to service steam locomotives was not needed. Jacks could be used to raise cars so that trucks could be removed, while portable cranes lifted traction motors and other heavy items, and moved them around the area. There would be the typical small machine tools of a repair shop, such as a drill press, lathe, grinder, gear press and so on, and a special room for rewinding traction motors. A wheel lathe would be necessary to reshape wheel treads and flanges, and a small forge was useful. Inspection pits four to six feed deep where motors could be dropped were also very convenient, as was an elevated platform for inspection of the trolley and other equipment on the roof. A transfer table was often used to access parallel tracks where room was tight.

Provisions for handling passenger traffic included stations and shelters. Full stations were provided only at the larger towns on the route, sometimes only at the termini, and were rarer than stations on steam railroads. A station would have a ticket window and a place for the passengers to sit. A baggage room was sometimes supplied, where articles would be accepted for handling in the express compartments of the cars, baggage could be checked, and items stored for pickup. Some termini had substantial buildings completely devoted to the interurban traffic, which also held the company offices on the upper floor, and there might be a newsstand and shops for the passengers. In other places, the interurban station was part of a drug store or restaurant, whose employees would sell tickets and accept parcels. Street running in towns made this convenient. The steam railway station was often inconveniently at the edge of town.

The many flag stops, however, were mostly represented by shelters beside the tracks. Shelters were normally of wood or concrete, providing only a roof, three sides and benches, perhaps 8 x 10 feet in size. They were typically located where a road or street crossed the line. In rural areas, a platform was sometimes provided for milk cans and other traffic, to make it easier to load them on the car. If there was no shelter, there was at least a station sign and a gravel walkway.

The road was the same as for steam railways, except usually lighter and less extensively graded. The individually-powered cars could easily ascend short, heavy gradients that would have been very annoying on a steam road. Axle loadings were usually smaller, perhaps 15 tons at most if carload traffic was absent. If carload traffic was handled, about twice this axle loading had to be accommodated. The absence of heavy locomotives made cheaper bridges possible. The lightest rail used on interurbans was about 56 pounds per yard, but 70 pound rail was rather common, and would allow freight cars to be handled. Some companies used 85 or 90 pound rail, often secondhand from steam railways. Light rail is a false economy, so the heavier rail was a very good choice. The lighter wear on the interurban extended rail life considerably.

Rail joints were bonded with copper wire welded into plugs that were hammered into the rail web on each side of a joint. It was very important to assure a low-resistance traction current return. Not only would a poor return reduce the voltage available for traction, but current would escape and travel through the ground or iron pipes along the track, especially in towns. Sometimes the voltage drop in the return was limited by city ordinance to 10 or 15 volts to minimize this problem. Occasionally, there could even be return feeder cables where the problem was severe.

Interurban cars could negotiate curves of small radius, as at street corners, and wind their way through towns, like a street railway. Turnouts, likewise, could be as sharp as No. 4. If freight cars, or trains with normal couplings, were to be handled, these dimensions had to be eased considerably. Heavier interurbans always sought private right-of-way, or at least rather straight street-running sections.

Cars were reversed, where necessary, by using a loop or balloon track at at terminus, or a wye track. A balloon track meant that a car did not have to reverse, and was very convenient for this reason. A wye track could be equipped with spring switches so that the motorman or conductor would not have to manipulate switches, but the car would have to reverse twice. If a car could be operated from either end, then the trolley pole could be swung around, or if one was supplied for each direction, nothing special had to be provided at the end of a run.

Where short sections of double track were provided, spring switches at each end automatically directed the cars to the right tracks, and no further steps were necessary. On single-track terminal loops, the current of traffic was maintained in the same direction, and traffic might enter the loop from several directions through spring switches. Interurban turnouts were usually manually operated split switches as on steam roads, with targets to show when a switch was open. In towns, single-point switches could be used, as on street railways, but these were safe only at low speeds. The conductor of a car generally handled the switches, but the motorman had to do the job on a one-man car. Sometimes in towns the switch could be changed using a long bar without leaving the platform. This was long before the days of switches that could be electrically operated from the motorman's position. Interurbans could seldom afford a switch tender, but in busy terminals one might be assigned to direct cars in or out.


A train crew usually consisted of a motorman and a conductor. The motorman would run the car, while the conductor would collect fares. In addition, the conductor would step down to change switches, flag the car across railway crossings, and telephone the dispatcher when the car was late. As an economy measure, in later years one man did both jobs, where possible. The car would then be arranged so that the passengers would pass by the motorman in coming and going. For freight trains that did work along the line, a brakeman was necessary in addition to pass signals and handle switches. Flagging was not an obsession on interurbans. Operation had to be such as to minimize it.

Interurbans ranged from what was essentially an electrified horse car, to something almost indistinguishable from a busy steam railway. Their equipment and methods of operation covered an equally wide range. On the one end, there is operation by sight at limited speed, which is quite acceptable on double track and in city streets, and which is still commonly used with street railways. At the other limit is the full timetable and train-order operation under the Standard Code, which was used on a few heavy interurban lines. Most companies used an intermediate method, closer perhaps to the simple end, because of the regular nature of their traffic and the properties of the cars themselves.

The typical interurban had frequent traffic, with cars running every half hour or hour in each direction from 6 or 7 am until 10 or 11 pm., usually at regular intervals. The original intent in many cases was to provide double track, which would make operation by sight practicable. However, financial constraints ensured that most lines would remain single track. The regularly timed cars met at the same places at the same times after the hour, so time table operation was satisfactory. On the larger and busier lines, things were not always so regular. This meant that a dispatcher was necessary to avoid delays, as on steam lines. However, there were no operators at intervals as on a steam line to handle train orders in the usual way. In fact, a steam road would find it impossible to handle 60 or so trains a day on a single track! It was possible for the interurban to do this because of the regular-interval schedules, and the short distances between terminals, which meant that the cars could usually keep pretty well to schedule, meeting regularly at the same places at the same times. All cars ran at about the same speed, and stopped at the same places for the same lengths of time. The dispatcher had much less to do than he would in the chaos of steam railway traffic. The fundamental operating instrument of the interurban was the time table, and there was considerable optimism that it would be realistic under normal conditions.

Passing places (i.e., meeting places, sidings), were provided at intervals, where required by the time table, usually at stops, and even stretches of double track were installed at congested spots. A telephone at each passing place connected directly with the dispatcher. If a car got off schedule for some reason, its conductor telephoned the dispatcher to notify him and ask for instructions. The conductor also telephoned if an expected car did not show up on time. Unlike on a steam road, there were no operators giving OS reports, so the dispatcher had to assume all was running according to plan unless notified otherwise. Sometimes the dispatcher could turn on a light at a telephone to request that the next car there call him. Whatever the case, the dispatcher could order the delayed car to stay where it was, and the opposing car to proceed, preventing the line from being tied up waiting for delayed trains. On some lines the orders were written down and handed to the motorman, but one suspects that mostly the communications were oral, with none of the elaborate ritual of the Standard Code train order. Fortunately, the interurban existed when telephones were available, and could be used by trainmen without extra instruction. Cars carried telephones, and wires were provided so that they could be connected wherever necessary.

The direct-current traction supply was returned through the rails, so the usual track circuit could not be used for signaling. Little need was felt for the added expense of automatic block signals, so they were seldom used. One system that was used had signal lights at the ends of sections of track between stations that were operated by switches at each end controlled by the car conductors. The simplest system might have a lamp at each end to show that the track is clear, controlled by a switch at each end like the switches on flights of stairs or halls. When entering the section, the conductor would turn the light off. At the other end, he would turn it on again. A car would not enter a section when the light was off. I do not know of an actual example of this, but it is so obvious that I assume it must have been used somewhere.

More elaborate controlled signal lights might have white, green and red lights (these were the usual colors for clear, caution and stop on steam railways before 1900, and were often retained on interurbans after the general adoption of green for clear and yellow for caution). The conductor of a car entering a section would press a button corresponding to his direction that would extinguish the white light at both ends, light a red light at the other end, and a green light at this end. A motorman encountering a white light would proceed, a green light would indicate a train in the same direction ahead, and a red light would mean that an opposing car was in the section. Of course, the green light could be omitted if it was not felt necessary to facilitate following (permissive) movements. I believe the West Penn and Altoona and Logan Valley had signals somewhat like this, and probably many more companies. They largely removed the need for the constant supervison of a dispatcher.

At grade crossings with steam roads, there was often no more provision for control than at a grade crossing with a common road. Once again, however, heavy interurbans would have the crossing protected, often with interlocking and signals. In many cases, the car stopped at the crossing, and the conductor went forward to see that there was no steam road traffic. If he saw nothing, he beckoned the car forward with a flag. Note that this could easily and safely be done, because the car could move quickly and clear the crossing almost at once. This was quite different from the case of two trains at a crossing. In some cases, the steam road might restrict the speed of its trains, but this was probably usually overlooked. The steam train did not have to stop at the crossing. Accidents at grade crossings were relatively rare, though they did happen and were then usually the fault of the electric road. In some cases, electrically locked gates were provided against the interurban. When a steam train was approaching, the gates could not be unlocked.

Protection against following cars was usually not required, since cars stopped at fixed locations, and one always looked out for them there. Indeed, there are few accounts of rear collisions between interurbans. The low speeds and the quick action of the brakes must largely account for this. If a car stopped in an unusual location, the conductor always went back to leave signals, as on a steam road. However, there was no regular provision of flagmen. Several incidents are reported where two cars were following closely, the first lost contact with the wire, and the second could not stop in time to avoid collison. However, the following car was able to reduce its speed sufficiently that the collision was not serious. There must have been many cases where the following car stopped well clear.

More insight into operating methods is furnished by the following anecdote. A freight train headed into a siding to allow a following car to pass, but could not clear the main track for a following train because the siding was too short. The brakeman knew he had to flag, but could not find either fusees or torpedoes as he rummaged around. Obviously, flagging was not a frequent occurrence. The red lamp went out as he jumped down, and he fiddled with it as the car approached. Other men helped out, and they waved what lights they had about, but in the bad weather the motorman of the approaching car could not stop in time, and smashed into the car fouling the main track. There were two fatalities. What they wanted to do was to stop the car, and have it saw around the freight. The freight did not pull forward out of the other end of the siding because its motorman probably did not know he was not in the clear in time.

Some companies, in their forty-year lives, had no serious accidents, but many could regret but one or two that had fatal results to passengers. Employees, as usual, were more subject to injury and death than passengers. The bad accidents were usually head-on collisions, caused by some misunderstanding and occasionally with contributions from bad weather. What is remarkable is the rarity of such accidents, taking into consideration the informality of the operating procedures and the rarity of aids to operating safety, such as block signals. The most serious accidents occurred on electric railways that had adopted steam-railway speeds and heavy equipment, where the informal methods proved inadequate. Under favorable conditions, the nimbleness of the cars could reduce the severity of collision, or even prevent it altogether.

Speeds were relatively low. In towns, a speed of 8 mph in business districts and 20 mph in residental districts might be prescribed by ordinance. Most over-the-road averages seem to be about 25 mph, sometimes slower. The very frequent stops should be remembered, since they can drop averages greatly. Usual running speeds were probably 30 to 40 mph, explaining why informal methods of operation were safe.

In competition with horses, 40 mph is a grand speed, ten times as fast. The early interurbans had little incentive to move more rapidly, and usually moved more slowly. In competition with the motor car, however, it is nothing to point to in the advertising. It happens, however, that as long as the trolley stays on the wire, there is really no limit to the speed of the interurban. A change of gears, and a bit more horsepower, and the car can move at 70 or 80. This not only demands smoother track, but also refinements in operating methods. Most interurbans that survived to give such high-speed service had to operate like steam railways, usually with automatic block signals, and became indistinguishable from electrified steam railways.

The Sand Springs Railway operated 120 trains per day, 60 in each direction, on its 8.63 miles from Tulsa to Sand Springs, Oklahoma. The usual running time was 40 minutes, including 17 intermediate stops, giving an average of 13 mph. The maximum speed between stops can be estimated at 25 or 30 mph. The line was partly single track, partly double track, and there were probably spring switches at the ends of double track. Freight trains were instructed to stay at least 10 poles behind passenger cars. The irregular, but relatively infrequent, freight trains would be dispatched so as to not conflict with each other, while the regular passenger trains would meet at the usual places without much extra effort. This operation was, essentially, like movement within yard limits on a steam railway, but much more intense.

Freight service was usually only an afterthought, though several electric lines continued in freight service for many years after passenger service ended. LCL package service was especially appropriate, as well as milk and mail. This could be handled on the platform of a passenger car, or somtimes in a car devoted to this purpose, either specially-built or converted from a passenger car. Such a heavy express motor, with no side windows but with cargo doors, might have four motors and be able to draw several freight cars behind it. For heavier duty, General Electric, Westinghouse and others built box-cab motors with four large traction motors that were small electric locomotives, the ancestors of diesel-electric locomotives. These, with 400 hp and more, could MU if necessary and draw considerable trains. In most cases, however, interurbans merely paralleled steam roads and were not essential for freight traffic. After withdrawal of passenger services, the electric wire was no longer as advantageous, and diesel engines proved an economical substitute in many cases.

Alternatives to Electrification

It is remarkable that electrification of interurbans was as successful as it was, considering the traffic that was available, and the large investment and high fixed costs associated with electrification. Many enterprises enjoyed considerable profits in the early days. The reasons why these profits disappeared, and were replaced by deficits and receivership, have already been discussed.

There were always alternatives to expensive electrification. One, of course, was the mule, and horse railways were still being promoted when electric street railways and interurbans were burgeoning. The mule was no match for the motor age, however, and animal-powered transport soon disappeared. It was slow, old-fashioned, and left deposits in the streets.

Another was the gasoline engine, and this was a serious competitor. Internal combustion engines were almost as convenient as electricity, since they could be started and stopped when required, used convenient liquid fuel that did not have to be shoveled and did not produce ashes, and were maintained without heavy equipment. They were applied quite early to railway purposes, either with a mechanical or electrical transmission. The internal combustion engine is not well suited to traction, since it produces no output when stopped, and must be married to a transmission that corrects this defect. Electrical transmission is an obvious choice, since the engine can be kept running at an economical speed, while the electricity can produce the tractive effort.

The Strang car of 1906 is an example, which was used on the Missouri and Kansas Interurban Railway (Kansas City, MO to Olathe, KS). The car was 52' long and had two 50 hp traction motors. It consumed 0.45 gal of fuel per mile. This, and later cars, had problems with some of the hills on the 21-mile line, so by 1908 the line was electrified.

Other companies modified automobiles and trucks to run on rails, and even to pull light trailers. These were essentially light-duty expedients, on very small lines. Such equipment broke down quickly if the traffic was heavier, and never were widely used.

Gasoline-electric cars were used more widely on steam railways, but they had the serious defect of carrying a large amount of easily inflammable gasoline. Fires aggravated many accidents, so steam roads avoided internal combustion power until diesel-electric cars became available. Then, of course, such cars were widely used on steam railways.

Electricity, however, was modern, clean and efficient, and always had a good public image. Electric cars were much more reliable than the alternatives, and could support a heavy traffic with ease. In most cases, therefore, electric traction was continued as long as passenger traffic lasted, and often into the freight-only era. Finally, usually after the electrical equipment was life-expired, small diesel locomotives, such as the GE 70-tonner, replaced electrical traction, and the copper was sold for scrap.

Another form of electric railway was also tried and found wanting as early as 1893, at least for general purposes. This was the battery railway, depending on secondary batteries that could be recharged at a central station. What set off the enthusiasm for batteries was the Faure pasted cell lead-acid battery of 1880. These batteries could deliver the heavy currents required by traction motors, and at the same time gave back more than 80% of the charge given to them. This seemed to overcome the economic difficulties of primary batteries, which effectively burnt expensive zinc. With storage batteries, the cheap power of coal could be effectively transported. This seemed to offer advantages for road vehicles as well as railway vehicles, and perhaps even for lighting and heating. The batteries would be charged economically at a central location, and then distributed to the consumers.

What was found, however, was that batteries were expensive, of limited capacity, and above all heavy. A few battery car lines were tried, with batteries stored under the seats and accessible from outside the car for easy changing to replace exhausted batteries with fresh several times a day. Electric automobiles were also manufactured, and electric trucks. Batteries were excellent as stationary backup for power systems, floated across the DC mains then common, or in other services where uninterruptible power was essential and the batteries did not have to be moved, or where only brief bursts of power were required, leaving most of the time for recharging. Storage batteries even made the electric starter for internal combustion engines possible, a circumstance that wiped out the market for electric automobiles now that the starting crank and a strong arm were redundant for the motorist. Even the nickel-iron storage battery did not alter the economics by much. These things were well-known by 1890, and there has not been much advance since then. Still, hope still exists that some new system less expensive and lighter will have the same advantages. In default of this, a tank of cheap gasoline has proved the most popular alternative. It should be realized that the electric railway with overhead or third rail completely bypasses the problem of energy storage for moving vehicles, and this is one of its chief advantages.

A more modern alternative is the linear induction motor, with or without magnetic levitation. The greatest difficulty with these ideas is that they are expensive and complicated alternatives to the wheel, which is a cheap and effective way of doing the same thing. If the steel wheel on the steel rail, which gives nearly zero resistance to motion, were not available, they might show some promise. Like battery power, they work and are quite ingenious, but are uneconomic. Unlike battery power, there is little hope of overcoming their fundamental disadvantages. One could also include "monorails" in this. Monorails were suggested almost as early as railways, both suspended and supported. A line near Ballybunion in the west of Ireland gave good service for many years. However, they have stability problems, among other inconveniences for general service. They are seen in amusement parks and such places, moving quite slowly. The electric Schwebebahn has been in service in the Ruhr since 1895, and gives complete satisfaction. But there is no reason to multiply examples, since they offer no advantages except in very restricted circumstances. Yet, promoters are fascinated by them! Actually, the almost forgotten "telpherage" cable railways invented by Prof. Fleeming Jenkin in the 1880's are among the most successful of such schemes. They reached 4 to 5 mph. The interest in monorails was excited by the hope of economy, not of any better performance.


  1. O. T. Crosby and L. Bell, The Electric Railway in Theory and Practice, 2nd ed. (New York: W. J. Johnston, 1893). An excellent survey of electric railway technolgy at the beginning of the interurban era.
  2. Allison Chandler, Trolley Through The Countryside (Denver: Sage Books, 1963). This remarkable book describes interurbans in Kansas, giving a look at the typical American interurban line. There is little detailed technical information, as could be expected, but there are many anecdotes and a good appreciation of the role of the line in the communities served.
  3. W. C. Jones and N. T. Holley, The Kite Route: Story of the Denver and Interurban Railroad (Boulder, CO: Pruett Publishing, 1986).
  4. Sand Springs Railway Company, Time Table No. 2, effective March 15, 1928.

Return to Railway Index

Composed by J. B. Calvert
Created 23 January 2001
Last revised 15 March 2001