Electric Railroads (in progress)

by David S.Lawyer mailto:davylawyer1@gmail.com

Nov. 2004 (original July 2003, minor editing Oct. 2006)

1. Introduction

2. History

3. Economics and Environment

4. Technology

5. Appendix - Notes

1. Introduction

1.1 Copyright

Copyright 2003-6 by David S. Lawyer. Feel free to make copies but commercial use of it is prohibited. For example, you can't (except to an insignificant degree) combine it with advertising on the Internet. Please let me know of any errors or suggestions for improvement.

1.2 What is an Electric Railroad ?

An electric railroad is powered by electricity which is obtained from an external stationary power plant. An overhead wire over the railroad track provides electricity to power the train, using electric motors to move it. As an alternative to an overhead wire, some railroad lines get the electricity from a third rail laid along the track just above ground level.

An electric railroad uses electric locomotives in contrast to diesel or steam locomotives. It sometimes uses electric motor cars where these rail cars carrying passengers have electric motors under the floor of the cars, like light rail trains. While diesel locomotives are powered by diesel motors, they usually use an electric transmission. This is where diesel motor drives a large generator to generate electric power which is then feed to various electric motors which drive the locomotives wheels. Thus its called a diesel-electric locomotive but it's not an electric railroad.

1.3 Historical Overview

Rail electrification goes back to the last decade of the 19th century (and beyond) when streetcar lines (which were formerly pulled by horses and mules) were converted to operate by electricity. Then in the early 20th century, some mainline steam railways (where trains were pulled by steam locomotives) were electrified. After steam locomotives were replaced by the more efficient diesel-electric locomotives at mid-century in the US, there was less incentive to electrify. But some countries such as the Soviet Union (includes Russia), decided to replace diesel locomotives with electric on high traffic lines . Thus, "electric traction" replaced animal, steam and diesel traction. Sometimes, a brand new railroad line was electrified during its construction.

1.4 How Does it Work?

Electric motors drive the wheels of the train. The motors are geared directly to the wheels with no transmission. Thus while there are no gears to shift, there is usually a gear box that reduces the speed of the electric motor so as to get more torque at the wheels. Electric power is supplied by an overhead wire with the steel rails used as a ground return for the electric current. Voltages on the overhead wire range from several hundred volts to many thousands of volts. Both AC (alternating current) and DC (direct current) are used.

In Europe, AC is often at the standard frequency of 50 cycles per second (50 Hertz or 50 Hz). Frequencies for the US are 25 Hz and 60 Hz. The old German system (also used in other parts of Europe) is 16 2/3 Hz. The strange frequencies of 25 and 16 2/3 Hz will be explained later.

1.5 Why Electrify?

Today, electric railroads usually use less energy (per unit of transportation work) than diesel powered ones. By using the traction electric motors as generators, they can brake a train and recover some electric energy as a bonus benefit. This is called "regenerative braking". Electricity is clean and non-polluting at the point of use. Also, it's a lot easier to maintain an electric locomotive than a diesel-electric one (which has a diesel motor in addition to electric traction motors). Since electric power can be generated by using various types of fuel such as natural gas or coal, trains can be powered by a wide variety of fuels other than petroleum.

In comparing a diesel-electric locomotive to an all electric, one may think of the electric locomotive as like a diesel locomotive but without the diesel power plant on board to generate electricity. Electricity generation is a lot cheaper at a centralized power plant due to economies of scale. The diesel motor just isn't very economical at generating electricity since it's often operating at part loads as well as idling. And it costs to haul the diesel-generator set all over the rail lines.

Still another advantage is that one may overload an electric motor for a short period of time and get much more than the rated power out of it. Provided of course that you are connected to the electricity grid that can supply that extra power, and not getting your electricity from a diesel-generator set that can't ever supply any more power than the diesel motor can put out.

All the above listed benefits must be evaluated against the high cost of installing the overhead wires and the substations used to supply electric power to feed these wires with electricity. In some cases, there is a shortage of overhead space to string the wires, such as in tunnels and under bridges. In addition, there's the cost of maintenance of the electrical system. If electric power fails (including the railroad's power system and overhead wires) trains cease to operate. Thus the benefits of rail electrification are not always worth the cost.

2. History

2.1 1900: Electric Power Generation Inefficient

About a hundred years ago around 1900, it took about 8 times as much fuel to generate a kilowatt-hour of electricity as it does today. (see Fuel Efficiency of Travel in the 20th Century. So electric railroads were not very energy-efficient then unless they were operated by water-power-generated electricity. Another problem was that electric power just wasn't available in the quantities that would be needed by the railroads, thus forcing them to build their own power plants. Thus there was little incentive to electrify the steam railroads around 1900. While steam engines were not very efficient then either, electrification of rail lines wouldn't have saved much fuel due to the then inefficient electric power plants.

2.2 The Electric Streetcars and Interurbans Electrify

While there was little economic incentive to electrify the steam railroads in 1890, electrification was just what animal-pulled streetcars needed. Before 1890, hundreds of U.S. cities had streetcar systems with streetcars mostly pulled by horses and mules Stat_Ab_1930 Scheduled speeds were slow: only 5-6 mi/hr. mi_hr Steam engines weren't used much because most cities wouldn't permit them to operate on city streets. But cities would allow operation of clean electric streetcars. And electric operation was much cheaper and faster than using animals to pull the streetcars. The result was a boom in streetcar conversion from animal power to electric power, as well as the construction of many new electric streetcar lines. By 1900, most all streetcars were electric Stat_Ab_1930 .

With little or no competition from automobiles, streetcars expanded out beyond the city limits to the countryside and new lines were build to connect nearly towns. These were known as the interurban electric railways. The length of a single route could reach over a hundred miles. But by the 1920s, both streetcars and interurbans found themselves in fierce competition with the automobile, and the automobile was winning.

2.3 A Couple of Steam Railroads Do Major Electrification in the US

While rail electrification was a no-brainer for streetcars and interurban lines that ran on city streets, steam railroads were hesitant to electrify. But as electricity became cheaper with improved efficiency, a few of them did.

Another no-brainer for electrification was long railroad tunnels with smoke problems from steam locomotives. Electrification offered a clean solution. So a number of steam railroads electrified certain tunnels, the first being the Baltimore and Ohio in Baltimore MD in 1895. See Steam Railroads Electrify

Only two U.S. railroads bought into major electrification projects of several hundred miles: The Pennsylvania Railroad (Pennsy) primarily between New York City and Washington DC 1915-1938, and the Chicago Milwaukee St. Paul and Pacific (Milwaukee Road) in the Pacific Northwest in 1914-1920. A number of other railroads electrified short sections of line (usually under 100 miles). The result was that only 1.23% of the U.S. route mileage became electrified but much of this has since been dismantled. Per Middleton, 1974, p. 400 Steam Railroads Electrify it was 2% of track miles. This makes sense because multiple track lines (such as the Pennsy) were more likely to electrify.

The 662 miles of the Milwaukee's 3000 volt DC electrification from Seattle to Montana (with a gap) was dismantled in 1973. The Pennsy electrification of 656 route miles, primarily between Washington DC and New York City, was at 11,000 volts and 25 Hertz. Amtrak has taken over the Pennsy line from Washington DC to New York but it's still at 25 Hertz. Amtrak installed catenary of 25,000 volts at 60 Hertz from New York to Boston, converting the New Haven's old 25 Hz electrification to 60 Hz between New York and New Haven.

2.4 Electrification in Russia and the Soviet Union

As compared to the U.S., the Soviet Union (USSR) got off to a very slow start in electrification but later greatly surpassed the US. Electrification in the US reached it maximum in the 1930's which is when the USSR just started to electrify on a small scale.

In 1932 the USSR opened their first short 3000 volt DC electrified segment in Georgia (birthplace of of the Soviet dictator Stalin) on the line between the capital, Tbilisi, and the Black Sea. Grades were as high as 4.6%. The original fleet of 8 locomotives was imported from the United States and were made by General Electric (GE). The Soviets also got GE to give them construction drawings so as to enable the Soviets to construct similar locomotives. Strange as it may seem, the first Soviet locomotive to be made was not a copy, but one of Soviet design which was completed in Nov. 1932 with great fanfare. Later in the same month, the 2nd locomotive to be made in the USSR, a copy of the GE locomotive, was completed. At first, many more copies of U.S. design were made than for ones of Soviet design, since no more locomotives of Soviet design were made until 2 years later.

In 1941, prior to World War II, the USSR had electrified only 1865 route-kilometers. This was well behind the US which had electrified nearly 5000 kilometers. However, since the USSR rail network was much shorter than the US, the percentage of of its system kilometers electrified was greater than the US..

Electrification was put on hold during World War II as the western part of the Soviet Union (including Russia) was invaded by Nazi Germany. After the war, the highest priority was to rebuild the physical destruction caused by the War, so railroad electrification was further postponed for about 10 years.

In 1946, just one year after the end of World War II, the USSR ordered 20 electric locomotives from General Electric, the same U.S. corporation that supplied locomotives for the first USSR electrification. But due to the cold war, they couldn't be delivered to the USSR so they were sold elsewhere. The Milwaukee Road in the U.S. obtained 12 of them where they were nicknamed "Little Joes", "Joe" referring to Joseph Stalin, the Soviet dictator.

In the mid-1950s, the USSR decided to launch a two pronged approach to replace their obsolete fleet of steam locomotives. They would electrify the lines with high density traffic and slowly convert the rest of the lines to diesel. The result was a slow but steady introduction of both electric and diesel traction which lasted until about 1980 when their last steam locomotives were retired. In the US, steam went out about 1960, 20 years earlier than for the USSR.

But once dieselization and electrification was complete in the USSR (around 1980) they began to convert some diesel lines to electric, but the pace of electrification slowed. The result was that by 1990, over 60% of the railway freight was being hauled by electric traction. This amounted to about 30% of the freight hauled by all railroads in the world and about 80% of rail freight in the US (where rail freight held almost a 40% modal share). The USSR was hauling more rail freight than any other country in the world, and most of this was going by electric railways.

After the Soviet Union fell apart in 1991, railroad traffic in Russia, sharply declined and new major electrification projects were not undertaken.

2.5 Electrification in Western Europe

In Western Europe, petroleum was not as plentiful as it was in the United States. In addition, railroads often became owned by the government. Both of these reasons contributed to decisions favoring electrification. Today, about 40% of the lines are electrified and most of the traffic goes on these lines. But each country has a different history.

Germany and Switzerland were the leaders in European electrification. In 1911, the German State Railways standardized on a system of 15 kilovolts at 16 2/3 cycles per second. This system was also used by Austria, Norway, Sweden, and Switzerland. Note that 16 2/3 is exactly 1/3 of the standard 50 Hertz used in Europe. Thus it could be easily obtained using a motor-generator set.

The French Railroads were privately owned until just before World War II so electrification in France mainly got started after the war. France rejected the German system and instead used 1500 volt DC (which was also a poor choice). Later they adopted 25kv at 50 Hz.

It should also be noted that in Europe today, the modal share of rail for both freight and passengers has been declining. For Western Europe, the decline started to accelerate in the 1950's but for Central Europe, it started to substantially decline with the breakup of the Soviet Union.

2.6 U.S. vs Europe

Prior to World War II, The United States electrification was almost on par with other industrialized countries. While it had more miles electrified, it was a larger country with a huge rail system that would be expected to have had more miles electrified than it did. Switzerland and Germany had done relatively more while France had done less.

But in the post-World-War-II period, not only did US electrification come to a standstill, it started to decline and electrification was dismantled. In Europe, the Soviet Union, and Japan electrification resumed with renewed vigor. The result was that the US wound up by the end of the 20th century with no major electrification of freight lines. Only Amtrak, the government subsidized passenger service, had substantial electrified passenger service between Boston and Washington, DC.

Why didn't the U.S. continue electrification after World War II and why did it dismantle the relatively small amount of electrification that it already had? After World War II the U.S. railroads had lots of cash due to their high profits earned during the war and since the war didn't destroy U.S. industry, there was plenty of industrial capacity to make locomotives. During the war years, the diesel locomotive had proved itself to be economical and reliable. So the simple solution of buying diesels to replace steam was jumped at as a quick and easy way to reduce costs.

Also, the U.S. had many duplicate lines running between major city-pairs while the USSR had avoided this wasteful duplication. The result was that the low traffic on many of the US lines didn't support electrification while the higher density on the smaller Soviet network did.

Now, why was most of the U.S. electrification dismantled? One reason was that it consisted mostly of short stretches of electrification which resulted in the additional expense of changing locomotives at the junction points. Diesels could run thru these points without changing locomotives. Tunnel electrifications were often converted to diesel which produces far less smoke and fumes than steam locomotives. Some electrifications were for suburban passenger lines and these were dismantled on lines when passenger service was discontinued. Another reason was that owing to the relatively small amount of electrification in the U.S., the cost of electric locomotives was high since they were made in very low volumes. The U.S "system" of electrification was just hodge-podge and not viable in the long run unless it was to be extended (which it wasn't) or subsidized (which it eventually was).

The reason that the USSR electrified and the USA didn't is mainly due to their different economic systems: The USSR being that of long range planning of a unified system; The USA being that of many different railroad companies each going in its own direction with the primary concern of making profits now.

3. Economics and Environment

3.1 Introduction

Saving money and a cleaner environment have been the driving forces behind electrification. When rail electrification was invented in the late 1800s, the efficiency of generating electricity from coal was extremely poor and the cost was high. Yet since electric railroads were seeming clean, they were adapted for powering streetcars and for use in railway tunnels. During the first few decades of the 1900s, the efficiency of electricity generation greatly improved and one could sometimes save fuel and money by electrifying. During the last third of the 1900s, the public became more concerned about the environment, leading to public financing of electric passenger railways such as "light rail".

So the driving force behind electrification was initially environmental in the late 1800s. Low cost drove electrification thru most of the 1900s until in the late 1900s it shifted back to environmental concerns for passenger rail. We thus have the sequence of the main reasons for electrification: Environmental, Low Cost, Environmental. However, electrification was not a one way street towards greater electrification. There were also periods of de-electrification, especially the dismantling of streetcar and inter-urban lines in the US and elsewhere.

3.2 Less Air Pollution?

The first wave of electrification was that of replacing animal pulled streetcars with electric. While replacing animal traction with steam powered streetcars might have been cheaper, it would have been dirtier due to the exhaust smoke. Just look at the smoke produced by steam locomotives. Most all cities would not permit the use of polluting steam streetcars, so electricity won because it was cleaner.

Also, some railroad tunnels were electrified since a steam locomotive in a tunnel could create a lot of toxic exhaust smoke. There were a few cases of people actually being killed by inhaling such smoke.

Of course, electricity is clean only at the point of use. The generation of electricity may be very polluting to the environment at the power plant that generates the electricity. Exceptions are hydroelectric power, wind power, geo-thermal, and possibly natural gas generation. Generation by natural gas is polluting in the sense that it generates carbon dioxide, as does the burning of other fossil fuels like coal. Carbon dioxide in the atmosphere is the major cause of global warming.

Today there is no surplus of cleanly generated electric power so any additional load due to railway electrification should be presumed to come from polluting power plants. Besides being polluting, most power plants are depleting the world's reserves of fossil fuels which means less will be available for future generations. So while electric railroads are non-polluting along the tracks, they may be highly polluting at the power plants that generate the electricity for them.

3.3 Electricity Hurts Humans

Electrified railroads bring the electrically hot catenary wires into close proximity to railroad employees and passengers, etc. Electricity hurts people in two ways: 1. Direct contacts may kill people by electrocution or burn them severely, etc. 2. Exposure to the magnetic fields of AC electric currents might cause cancer.

Electric Shock and Electrocution

In the Soviet Union, which had extensive railroad electrification, about 50-60 rail employees were killed every year by electrocution. See Soviet Electrocutions. This high death rate was in part due to lack of warning signs and lack of safety education for employees. Many rail cars had hatches thru which one could climb out on the roofs of the cars, sometimes with deadly results if one contacted the high voltage catenary just above the roofs of the cars. Also work is sometimes done on the catenary under full voltage. This is safe if one knows how to do this and is extremely cautious.

In the United States, although the extent of railroad electrification is much less than in the former Soviet Union, many juveniles are killed or injured by climbing up on the roofs of railway cars. See

Rail electrification kills juveniles

Magnetic Fields

It's claimed that there may be adverse health effects, such as childhood leukemia, from AC magnetic fields. A steady DC current will not create such a field but AC will. However, even with DC, the power electronics may switch the DC current on and off and thus create sharply varying currents (ripple) on the DC catenary system which could be even more dangerous than the ordinary AC. But power electronics can also be designed to do a good job of filtering the current so as to avoid substantial AC components and thus less risks to the passengers. Filtering out the AC ripple would help reduce interference with communication and also reduce power losses in the catenary.

There has been numerous articles published about the possible hazards in of AC power lines and old home wiring. Examining all of the data leads one to conclude that there is some danger but possibly not a great deal of danger. A number of studies failed to establish that there was any problem, but they also didn't show that there wasn't a problem.

For an electric railroad with an overhand catenary, the magnetic field tends to be significantly stronger than for a power transmission line. For a power line, the magnetic field due to the current in one wire tends to be cancelled out by the currents in the other wires. As one goes away from the power lines, this effect results in the magnetic field decreasing according to the inverse square law. But for a single straight wire its magnetic field is only inversly proporional to distance one goes away from the wire. This is because there are no other wires nearby to cause cancillation.

For an electric railroad, people may find themselves located in between the two conductors (the catenary and the rails-earth) where the magnetic fields from each conductor are additive, resulting in a much larger magnetic field exposure to humans. To reduce this possible risk on AC systems an overhead wire may be strung to carry the return current that formerly flowed thru the rails. This wire carries exactly the same current as in the catenary but in the opposite direction and helps partially cancel ot the magnetic field from the catenary.

Another likely safer solution is to use DC electricity (instead of AC) that creates a constant magnetic field (similar to the earth's magnetic field). It's not thought to be harmful since the animal life has evolved on earth in the presence of the earth's magnetic field (although it's magnitude has varied over geologic time and it has even reversed direction over long periods of time).

3.4 Electric Currents Damage Metal Underground

The electric circuit for an electric railway uses the rails for the second conductor. So if no current leaked between the rails and the earth, the current in the rails would be about the same as the current in the catenary above the rails. But in reality, much current leaks out of the rails and flows thru the ground. The current tries to seek out a the paths of the low impedance (resistance for a DC system). While the dirt of the earth doesn't conduct as well as the steel rails, it's cross section is very large, resulting in low resistance and substantial currents..

Underground current will seek out metal pipes underground and current will flow thru the pipes. As the current leaves a pipe and goes into the earth, the result is corrosion of the pipe at this location. The pipe is like an anode at this location and electrons flow from the earth into the pipe (opposite of the direction of current flow by convention which considers current to be the flow of fictitious positive charges). See Electric Railroad (Russian) pp. 166-7. It seems that such corrosion in mainly a problem for DC railroads where the current constantly flow in the same direction. One can protect pipes by putting a small voltage on them so that current flows only from the earth into the pipe and not out of the pipe.

3.5 Energy-Efficiency of Electric vs. Diesel

Today in the 21st century, electric traction is usually more energy-efficient than diesel. If one compares a diesel-electric locomotive under full load pulling a train with an electric locomotive pulling the same train, there was little difference in energy-efficiency (per a Russian study in the 1980's). While the electric motors are quite efficient, the energy generation at a stationary power plant was only a little over 30% efficient. In addition there are the transmission losses of sending the electricity from the power plant to the train. But today the efficiecy of power plants has increased and made electric railroads more efficient than diesel. Even the Russian conclusion that electric was no more efficient than diesel in the late 20th century was false for the followingi reason.

But the above mentioned Russian study was not a fair comparison because operating conditions are often far removed from just moving at steady speed at full load. There are 3 reasons why electric traction was actually more energy-efficient then (and is evem more efficient today):

  1. Diesel locomotives waste energy idling their engines. For electrics, there's no such thing as idling.
  2. Diesels often operate at part loads where their engine efficiency is low
  3. Many electric railroads can recover energy by regenerative braking

Russian (Soviet) rail policy was to rapidly electrify their railroads in spite of this one study.

When making such a comparison between diesel and electric, one should consider the possible improvements to the diesel locomotive. This is done in the following section.

3.6 Improving Diesel Efficiency

Diesel engines could be shut down instead of being allowed to idle for long periods. A problem with this is the pollution caused by starting a cold diesel. Is it feasible to use heaters keep them warm when shutdown?

Diesels could be designed and operated so that they would operate more at full load. In trains with multiple locomotive units, some units could be shut down when their power isn't needed. Also, if full power isn't needed to maintain speed then one may maintain speed by accelerating (starting at cruising speed) at full power for say a minute and then coast for a minute or two to return to cruising speed. Instead of idling while coasting, it would save more energy to fully shut down the engine for a short time..

To recover some of energy normally wasted in braking and recoverable by regenerative braking, diesels could do more coasting as they approach a stopping point. This is one way to recover some kinetic energy, but it slows down the train's schedule. On grades, energy from going downhill might be stored in a flywheel storage system on the locomotive. However, it would be best if such locomotives were used as helpers to operate only in a territory with grades. In order to switch them into and out of trains rapidly, new coupler are needed for the entire North American rail system which is quite a project, but it would have many more benefits than just the above.

3.7 More sections to be added

4. Technology

4.1 The Overhead Catenary

Above the track hang bare wires which supply electricity to the train. The lowest such wire which runs almost level and supplies electricity to the train is the "contact wire" (also called "trolley wire" from streetcar days). But if this were the only wire above the train, it would sag a lot. So the contact wire is usually supported by a strong wire hung above it. This support wire sags in the shape of a catenary curve and is thus called the "catenary" or "messenger" wire. It's similar to the catenary cables of a suspension highway bridge. The vertical wires which hold up the contact wire and hang it from the messenger are just "hangers" or "droppers" (UK). All the wires mentioned above are part of the catenary system and might just be called "catenary" for short.

Sliding along the copper contact wire is a carbon strip (or brush) which is mounted on top of the pantograph which is attached to the roof of the train. The pantograph has springs which push upward on the carbon strip so that it makes contact with the contact wire.

For a picture, see: http://www.railway-technical.com/ohl001.gif


It may all sound simple but there's a lot more complexity. The contact wire needs to be positioned near the center of the track. Otherwise the carbon strip on the train's pantograph might lose contact with the contact wire. But if it were always exactly in the center, the carbon slider strip would get a groove worn in it's center. So the alignment of the contact wire is intentionally made to deviate from the center of the track and zig-zag from side to side. This is known as "stagger" or "zig-zag". This way the carbon slider wears more uniformly from the sliding wire.

Cantilever arm

To support the catenary system, the catenary cable (messenger wire) must be hung in the middle of the track. Since support poles (or masts) can only be placed at the side of the track and not in the middle of it, some kind of arm or bridge is needed to go over the track center. For multiple tracks, one way to do this is to have a heavy cable go across the tracks between the tops of two poles, one pole on each side of all the tracks. This is called a cross-span (above and perpendicular to the tracks). Another way is to have a cantilever arm go out from the top of a pole to support one catenary. The cantilever is not really a cantilever since there is usually another arm (or cable) that holds it in position, forming a triangle, with part of the pole as one leg of the triangle.

The contact wire is not exactly round but has a pair of grooves in the top part of it so that it can be clamped to the hanger wires. Thus the carbon slider on a train never hits this clamp since the clamp only covers the top half of the contact wire.

Steady arm

The system described so far only holds up the catenary (messenger) cable which in turn holds up the contact wire via hangars. But it doesn't prevent the contact wire from swaying from side to side due to winds and/or the forces applied by the carbon slider. Neither does it provide for the contact wire to follow the rails smoothly around a curve nor does it provide for zig-zag.

Thus, for the purpose of steadying the contact wire and holding it in place transversely, a horizontal rod is used which is known as a "steady arm" or "registration arm" (UK), or simply "steady". It's connected to the same poles used to support the cantilever arms. It may either be directly attached to theses poles or may be connected some way to the cantilever arms. Look down an electrified track and you'll see one steady pulling the contact wire to the right and the steady on the next pole further down the line pushing the contact wire to the left. Or conversely: left-right-left, etc. This is to put zig-zag on the contact wire. To push the contact wire away from the pole, the steady usually has two parts: a long rod that goes out across the track and another rod which pulls on the contact wire and is attached to the end of the first rod. The main part of the steady pushes while the second rod pulls.

It's like you took a small wood pole like a rake handle and attached a short rope to its end, with a hook on the end of the rope. Then reach out with the pole and hook something like a horizontal wire (say on a wire fence). You can pull on the pole to get the wire closer to you or you may push on the poll to move the wire away from you. Pushing will result in the rope pulling (since a rope can't push). This is just like the push-pull operation of a steady that "pulls" the wire away from the pole although the main part of the steady is actually pushing.

A more simple steady that just pulls the contact wire towards the pole may just consist of one rod only. Thus the steadies maintain the horizontal alignment of the contact wire. But the steady must also allow movement of the contact wire along its axis to accommodate thermal expansion and contraction of the contact wire. In the hot summer, the contact wire becomes longer and in the winter it's shorter. So the steady is hinged both at the contact wire and further back to permit such movement. The steady also has yet another hinge that permits both the contact wire and the end of the steady to lift up a little when the pantograph on the train goes under it and pushes it up. Thus the end of the steady is free to swing both from side to side to allow for thermal expansion and up and down to allow for lift as a train's pantographs pass under it.

Maintaining tension

But what about the contact wire getting longer in summer? Wouldn't this cause it to become slack and sag? This doesn't happen since the contact wire is kept under constant tension by a steel cable that pulls on it from one end thru an insulator to keep voltage off the steel cable. This steel cable goes over a pulley (at the same height as the contact wire) and then runs down to where a number of heavy weights are attached to the cable. Often two or three pulleys are used to make the tension on the contact wire greater than that of the weights. With 2 pulleys, the tension is double that of the weights. With 3 pulleys, it's quadruple.

Equalizing elasticity

In order for the carbon brushes of the pantograph to press up against the contact wire, the contact wire must be elastic in the upwards direction. No matter how much tension there is in the contact wire, there will be no force on the brushes until the wire is pushed up some. To illustrate this, try pushing sideways on a stretched rope.

If the elasticity is not uniform, the pantograph will move up higher at locations where the elasticity is high and conversely. This will result in the pantograph oscillation up and down. If it's severe enough, the carbon brushes may momentarily lose contact with the contact wire with a resulting electric arc filling the resulting air gap and damaging the contact wire.

The elasticity is greatest in the middle of a catenary span, and lowest at the pole which supports the catenary. The elasticity in the middle may be almost double the elasticity near the pole. You might ask "Isn't the steady (at the pole) hinged so that it can lift up and permit full elasticity at the pole?" Well, the steady allows the contact wire to lift but the catenary cable above it doesn't lift much since its attached to the cantilever. Why this is significant will be explained next.

In the center of the span, both the catenary cable and the contact wire lifts when a pantograph on a train passes by. In the middle of a span, when the contact wire lifts, the adjacent hangars lift too, allowing the catenary cable to lift also. And catenary has a lot of tension on it and doesn't run in a straight live, so it will lift. As it lifts, tension is kept on the hangers so that they help pull up the contact wire up. Thus the contact wire has high elasticity in the middle of a span due to the pull of the hangars. But next to the pole, the hangars are effectively almost rigidly attached to the pole and when the contact wire moves up, these hangers just become slack and don't pull up on the contact wire. Thus there's less elasticity there.

There are various ways to help cope with this problem. One way is to increase the elasticity near the pole. Or one can just do nothing and live with non-uniform elasticity. One way to increase the elasticity is to install a second catenary cable where the middle of this secondary catenary span is just at the pole. Of course, the contact wire near the pole hangs from the secondary catenary. Something like this is actually done. The secondary catenary is made short, about 1/4 of the length of a main catenary wire span (between two poles). It's not hung from poles, but is hung from the main catenary. This type of catenary system is called a "stitched catenary". See



In cold climates, a major problem is the icing of the contact wire. It can accumulate ice on it which interferes with the flow of electricity from the wire to the carbon slider (brush). The larger diameter of the wire (due to the attached ice) makes it more susceptible to be blown by the wind and it can develop oscillations caused by the wind. If it moves enough to run off the edge of the carbon strip it means arcing. At worst, it could even break the contact wire.

There are various solutions to the icing problem. One is to run high currents thru the wire in order to heat the wire and thus melt the ice. The current from operating trains is often not sufficient to melt the ice so current is put onto the catenary at one substation and removed at the next substation. Another method is a vibration device on a pantograph to shake the ice loose.

Number of wires/cables

Most electric railroads today use two wires overhead: the catenary cable and the contact wire. In the past, many streetcar and interurban lines used just a single contact wire. It was called the "trolley wire" since contact was made by a rotating wheel (or trolley) which had a groove in it like a pulley wheel. The problem with this is that the trolley wire would sag between support points. But worst of all, there would be a sudden change of slope of the trolley wire at it's points of support. If the speed was too high, the trolley might lose electrical contact at these points and arcing would damage both the trolley wire and wheel. By having closely spaced points of support, there is less of a problem. But this requires the expense of having more poles in the ground.

Another alternative to two wires is three wires. Below the catenary cable, a secondary catenary cable is suspended by hangars from the top catenary. The secondary catenary has shorter spans than the top catenary cable. Then the contact wire is suspended by frequent short hangars from this lower catenary cable. This results in less sag in the contact wire. The third wire in the catenary system also conducts current so that it does provide some benefit. However, there is additional complexity and installation cost. Such "compound" catenaries have been used on high speed passenger lines in Japan.

An alternative to a single contact wire is two contact wires spaced very close together. The pantograph carbon slider rubs on both of them.

4.2 Voltages, AC or DC

The higher the voltage, the smaller the overhead wires, since less current is needed for the same power transmitted. Power = volts x amps (assuming a perfect power factor if it's an AC system). Much higher voltage is used for AC systems than for DC systems, but this is a historical artifact. Before low-cost power electronics, it was not feasible to transform DC voltages but was feasible to do so for AC using transformers.

Electric motors are usually around 600-1500 volts. It's easy to get this voltage from an AC system by using transformers in the locomotive to transform high voltage AC from the overhead wire to the lower voltage needed to run the electric motors. Thus one could use a high voltage in the catenary. 25,000 volts AC is common, with 12,000 volts for older systems and 50,000 volts for a coal hauling operation in the US.

For DC, the original streetcars were around 600 volts. Other DC systems are 1500 volts where two motors may be connected in series to put 750 volts on each motor. 3000 volts is commonly used where two 1500 volt motors are permanently connected in series. In the former USSR a 6600 test section was in operation in the 1980s. Today, higher voltage DC is feasible due to power electronics to covert high voltage into lower voltage.

4.3 Advantages of DC

A DC system has some advantages over AC. One is that it makes more efficient use of the catenary system since the power transmitted does not vary over time. For an AC system (at perfect power factor) the power goes on and off at a rate of 100 Hz (or 120 Hz) and is equal to V x I x sin^2 (w x t) where x is the multiplication symbol, and ^ is exponentiation. w it the angular frequency (2 x pi x frequency), V is the peak voltage, and I is the peak current. The average power is just half of V x I since the average of the square of the sine is one-half. For DC we could use voltage V and current I resulting in twice as much average power (V x I).

But what about resistive energy loss in the overhead catenary system. For the AC case it's 0.5 x I^2 x R where R is the resistance. The 0.5 factor is there because the RMS current (used for resistive losses) is only 0.707 x I while I is peak current. Now the DC loss is I^2 x R which is double that of the AC case. But it's really no worse since the DC case supplies twice as much power and thus the percentage power loss due to catenary system resistance is the same in both cases. Thus we can assert that for the same percentage resistance loss, the DC system can convey twice the power.

The above is a little unfairly biased towards DC since it assumes that the peak AC voltages for AC and DC can be the same (with the same insulators). But a DC voltage may be more likely to flash over the insulator than the same peak AC voltage, since the AC peak voltage is only applied for say a millisecond or so. Even so, DC will likely still be more efficient. Electric utilities have found it economical to utilize high voltage DC for long distance power transmission.

Another advantage of DC is that there is little or no voltage induced in other circuits by the steady DC magnetic fields. The voltage induced in a wire loop is just the rate change of the magnetic field thru the loop. Since AC current is constantly alternating direction, it will induce a voltage in nearby communication circuits that are not designed to resist such induced voltages. Also, human exposure to DC magnetic fields is likely quite safe but exposure to AC magnetic fields might be hazardous. See Magnetic Fields

4.4 Regenerative Braking

For braking an electric train, it's often possible to use "regenerative braking". The electric motors, which normally drive the train's wheels, are used as generators to generate electricity which is returned to the overhead wire to be used elsewhere. Since it takes torque to turn the generator, a braking force is created. It's something like braking an auto using the engine. But for the train, the energy created is in part recycled. If it can't be recycled and put back on the contact wire, it may be wasted in resistors on board the train and turned into waste heat. If the energy is wasted, it's not "regenerative braking" but "dynamic braking".

Regenerative braking is not so easy to do. It's easy to get an electric motor to work as a generator but to get it to put the energy back on the contact line (catenary system) is a complicated (and somewhat costly) problem.

First of all, the contact line must be able to absorb the energy placed on it. Either there must be another train nearby demanding power or the substations that supply power to the catenary must be capable of having power flow in the reverse direction: from the catenary to the power grid of the country or region (50 or 60 Hz).

Another problem is that for the energy to flow down the catenary away from the train, the voltage at the braking train must be higher than the normal catenary voltage. It's similar to charging a battery: The charging voltage must be greater that the battery voltage. How does one get higher voltage from a generator?

The generator is analogous to a coil of wire rotating in a magnetic field. The voltage induced in the coil is just the rate change of magnetic flux thru the coil due to rotation, times the number of turns in the coil. That is: voltage = (number-of-turns) (webers/sec of flux change). To increase the webers/sec one may either increase the speed of rotation or the magnetic field of the generator. But usually, neither one of these actions is feasible.

Since the electric motors (being used as generators) are permanently geared to the driving wheels of the train, there is no possibility to shift gears to increase rotating speed. Also, the magnetic field is created mostly by the ferromagnetic properties of iron, and the iron is likely nearly fully magnetized so it's magnetic field can't be increased. Putting more current into the coils that excite the iron into becoming magnetized is often almost futile, because the iron is nearly magnetically saturated.

So in order to get a high enough voltage, power electronics is needed much of the time. Thus modern regenerative braking systems utilize solid-state power electronics (using IGBTs transistors) to step up the motor voltage to that of the catenary.

4.5 Protective Circuit Breakers (Locomotive)

The main motor circuits of the locomotive need to be protected against any short circuits for the same reasons it's needed in residences and business. A short circuit could draw very high currents and damage electrical equipment, both onboard the locomotive and in the external circuits feeding the locomotive. High current means high heat which could damage equipment in the substations or even the overhead wire system.

There are various types of protection. Circuit breakers can trip on too much current, just like they do it homes. There are also ground fault interrupters which are common for DC locomotives. These use a "differential relay" which operates if it detects a difference in current between that entering a motor and leaving a motor. Instead of having one for each motor, there may only be one for say every two DC motors permanently connected in series.

For an AC locomotive, there's a clever way to monitor the secondary winding of the AC power transformer. To do this, a small DC power supply may be connected between physical ground and one side of the AC power system (transformer secondary). Thus puts the AC system at a small DC voltage upon which the AC voltages are superimposed. But since the secondary AC system is isolated from ground, it should draw no current from the DC power supply. It it does, it means there is a ground short somewhere in the AC system and the current coming out of the DC power supply will trip the main circuit breaker.

We have mentioned 3 types of protection: overcurrent, differential relay (ground fault), and AC ground fault detection by DC. There is still another: overvoltage protection. It's mainly for cases of protection against lightning strikes on or near the catenary. Solid state devices which temporarily break down at high voltages can be used. They allow current to pass thru them (low resistance) when the voltage is high (such as a lightning strike) but block current (high resistance) when the voltage is low.

Now it's easy to break an AC current by just opening a switch since the electric arc created by opening the switch contacts should extinguish itself when the direction of current reverses. Not so for DC where the current always flows in the same direction. So a DC breaker needs special design to extinguish the arc. The arc tends to rise since it's hot. As it rises, insulating plates can intersect it forcing the arc to become longer in a zig-zag path. A strong blast of air can rapidly move the arc and help blow it out. The idea is to rapidly force the arc over a longer and longer path until it goes out.

Another type of circuit breaker for DC is to break the contacts in a vacuum where there is no air to become ionized to support an arc. An arc can form in a vacuum but it requires much higher voltage to sustain it. Thus DC circuit breakers can be designed, but they are more complicated and expensive than AC breakers.

4.6 More Sections to be Added

I had hoped to work on this a few hours per week (or month). But a few years have passed with almost no work done. Sorry.

5. Appendix - Notes

5.1 URLs

A Short History of High-Speed Railway in France Before the TGV Info on electrification in France.

Electric Railways Brief history of electrification in the US.

5.2 Horsecar speeds

Chicago reported a scheduled speed of almost 6 mi/hr for horse-pulled streetcars. See U. S. Department of Commerce and Labor, U. S. Bureau of the Census: Street and Electric Railways. Washington: Government Printing Office (GPO) 1905 (series title of: "Special Reports"). p. 32. Middleton claims 5-6 mi/hr in "The Time of the Trolley" by Wm. D. Middleton, Kalmbach Publishing, Milwaukee, 1967. p. 23.

5.3 Electric Railways: Statistics

U.S. Department of Commerce, Bureau of Foreign and Domestic Commerce: Statistical Abstract of the United States, US GPO, 1930, pp. 422-3, Table No. 458: Electric Railways (Summary of Operations 1890-1927). Table No. 459: Electric Railways (Mileage, Equipment, Output of Electricity, Traffic, Employees and Salaries and Wages 1907-1927).

5.4 Steam Railroads Electrify

The magazine: Trains, July 1970, was entirely devoted to the topic rail electrification. B&O is on p. 22.

When the Steam Railroads Electrified, by William D. Middleton, 1st edition (Kalmbach Books) 1974; 2nd edition (Indiana University Press) 2002. Likely the best book on the history of electrification in North America, but deficient in economic and engineering analysis. Its many photos make it more like a railroad fan book than a serious study of the rise and fall of electrification in North America. Russian locomotive order on p. 238 (1st ed.)

5.5 Books in Russian

5.6 Books in English

5.7 Articles in English

5.8 Articles in Russian