The thread about streetcar bulbs prompts me to offer a question about trolley wire power. As I understand, streetcar (and trolley bus) power is DC current. But DC is very poorly transmitted over distances, meaning that the trolley wire would quickly lose sufficient voltage to power the vehicles as distance increases from the source. I seem to recall seeing heavy wires on poles along the streetcar/bus routes that a friend explained to me were "feeder lines." These lines, he said, carried AC current which transmits much more efficiently so voltage loss is minimized over distances. From these feeder lines would be connections to the overhead wires at intervals, through rectifiers to change the current to DC. This would keep the DC voltage on the trolley line much more consistent and sufficient to power the vehicles. Am I anywhere near being correct in all this?? If so, would this scheme also be the case for today's light rail transit, and even elevateds, subways, and railroads (but with appropriately different voltages, electrical conductors, etc.)?

There's sumptin' missin' from that story. The heavy cable(s) on the pole line are indeed likely "feeders". But, feeders are at the same 600VDC potential as the contact wire; they just provide more copper to carry more power.

In photos of classic interurbans, there is usually a crossarm higher on the pole, carrying three smaller wires on big insulators. That's a high voltage AC transmission line, that runs from the power station (either the railroad's own generating station or commercial power) to the railroad's substations, spaced every three to ten miles, depending on what standards the line was built to. The Substations are the "rectifiers" in the system... back in those days the only viable way to convert AC to DC was mechanically, with a rotary converter... essentially a motor-generator with common windings. Feeders then ran from the substations out along the trolley wire, with taps every several poles, to keep the trolley voltage up.

The severe transmission losses with DC is one reason why as technology became more advanced, interurbans went to first 1200VDC, then 1500VDC. even 2400VDC was tried, not successfully. Each doubling of the voltage doubled the distance between required sub-stations.

The voltage didn't stop at 2400 VDC. Both the Lackawanna and Milwaukee had very good success with electrification at 3000 VDC.

Oh, now I get it. Thank you for that clarification, Dennis, and the info about even higher voltages from "EDM." (Incidentally, I was amazed some years ago to come across the Black Mesa and Lake Powell Railroad in Arizona that operates on 50,000 volts!)

3000 volts is probably the maximum for DC railway operation. 11-12,000 and 25,000 are the most common AC voltages. Trolley voltage from 550 to 750 VDC has been standard for streetcars, rapid transit and most light-rail operations for over a hundred years. This subject came up recently in a discussion with someone who is very knowledgeable on matters of electric railway voltages. He told me about how the engineers who designed the BART system thought they were saving money by going to 1000 volts on the third rail--fewer substations needed. But when it came time to replace wiring in the cars, they found that standard electric wire (such as that used in diesel locomotives and normal subway systems) is only good for 750 volts, and the next standard size, which is probably rated for 1500 or 2400 volts, is too thick to fit in the conduits. So they had to order a large enough supply of 1000 volt wire to (probably) last a lifetime.

But that's 50,000V AC. High voltage AC railway electrification technology developed several decades after the initial street railway electrification, mainly in response to the problems of transmitting adequate power for mainline railroad work. I'm not going to claim that I know very much about high voltage electrification, but I think it's safe to visualize them as having transmission voltage on the trolley wire, and a portable substation in each locomotive.

Denver's new electric rail systems use 750 VDC for light rail and 25,000VAC for the commuter rail. The most common voltages today.

Ira Schreiber

DC was what powered the preferred type of traction motor. Transformers, however, are AC-only devices. Thus the conundrum facing the designers of any electrification back in the day: Use the type of power that lets you use the better type of motor but requires the voltage to be lower because you cannot step it down in a transformer (and thus need closely spaced substations and and for really heavy current might require third rail instead of trolley wire) ... OR use AC which can be at much higher voltage, stepped down with a transformer aboard the locomotive or car but have to use motors which are similar to DC motors (brush / commutator) but have some inferior characteristics.

One further good thing about AC is that the step-down transformer can be designed with multiple secondary taps to produce a selection of voltages and that selection becomes the basis for controlling the locomotive, instead of the variable resistance grids and series / series-parallel / parallel switching on DC equipment.

Also worth pointing out is the need for low frequency AC when using it with large motors which is why most here used 25 Hz. In Europe, much of it was at 16 2/3 Hz.

Despite whatever the drawbacks are to the AC brush/commutator traction motors, the PRR used them on all their locomotives until the 1950's and even the famous GG1 was so powered.

Some railroads like GN combined the advantages of high voltage AC distribution with DC traction motors. They did this by having a motor-generator set aboard the locomotive. The AC voltage was stepped down with a transformer, fed to the motor and the DC fed to the traction motors. Controlling the generator via its excitation was part of the control system.

One might reasonably ask what was gained if they still needed one of those "less desirable" AC motors to turn the generator. But this would have been a different type of AC motor entirely: an induction motor. These are great motors. They are simple and robust and power much of the equipment of the world. They just don't do variable speeds easily. Generally they are single speed machines either synchronizing to the AC frequency or chasing it.

That makes them ill-suited to be used directly as traction motors. That didn't prevent their limited use, for example, in the original GN old Cascade Tunnel electrification. As traction motors there is a way to connect them to produce a choice of 2 speeds but this is still very limiting.

Beginning in the 1950's it became possible to use AC with DC motors by way of rectifiers. The first of these used mercury vapor devices popularly known by the GE version called Ignitron. They required complex circuitry but did enable conversion of AC to DC in quantity. A few years down the road silicon diode rectifiers became available in sizes suitable for locomotive use. And later, thyristors came along that tranformed AC into DC but in a way that enabled control over what portion (time) of the AC waveform was being allowed to pass. This enabled controlling the locomotive in way similar to an electronic light dimmer.

Since then, all sorts of advancements have been made for both AC and DC systems and the latest use those simple and robust AC induction motors as traction motors, supplying them with variable frequency AC from computer-controlled inverters. Those things start with DC so if the power distribution is AC, it's rectified into DC first, creating an AC-DC-AC path.

For reasons that I won't pretend to completely understand, while there seems to be insufficient advantage to installing low frequency systems on new installations which is why places like New Haven to Boston use the normal 60 Hz commercial power frequency, apparently there is sufficient advantage to 25 Hz to keep using it where it currently exists such as the majority of the rest of the Northeast Corridor. (Modern locomotives and cars can accept either.) The rotary converters are mostly being replaced by electronic ones. Even that is open to debate as European manufacturer of rotaries pointed out their advantages at the IEEE/ASME Joint Railroad Conference in 1997. Supposedly they are still being installed new in Europe.

Steve

As I recall, 750 volt is more common today because of efficiencies in dropping voltages using transformers, and then converting with a rectifier.

Historically, AC was converted to 600v DC with a rotary converter, which consumed energy in the conversion process. The inefficient system was improved with automatic substations which would start up as draw down occurred on the line. A wacky electronic guru named Bion J. Arnold thought this up - I imagine him a bit like Dr. Emmett "Doc" Brown of Back to the Future - but perhaps he was not that out there. After creating plans for Chicago's streetcar subway system and cross-town "L"s, he took over management of the Elgin and Belvidere interurban. The first automatic substation was in a little town called Union IL.

250 Volts was used for industrial switching trolley locomotives.

1200 VDC was another rather frequent interurban and electric railroad power supply.

One thing to keep in mind was that DC traction motors were perfected long before AC motors. What is understood as the first successful electric street railway system was in Richmond, Virginia, in 1888 with DC power.

AC traction motors did not become available until after 1900. By that time many hundreds of different systems using DC were in operation. AC motors were not compatible with existing systems so companies had no incentive to change.

Early AC traction motors could not be used with 60 Hz power because of inductance in motor windings. Therefore lower frequency AC was required. 60 Hz motors were many years in the future.

In 1908 General Electric was so anxious to get into the AC traction business that it contracted to supply equipment to the Washington, Baltimore, and Annapolis interurban with a money back guarantee. If the AC equipment was not satisfactory, GE would take it back and provide DC equipment for payment of the difference between the AC bid and a DC bid mde at the same time.

The GE built AC equipment was not sataisfactory so GE supplied the WB&A with DC equipment under warranty in 1910.

AC cars required a heavy transformer on board so they weighed a lot more than comparable DC cars. They had many mechanical problems. They could not operate on the DC conduit power in Washington so the terminal was on the outskirts of the city.

There is no difference between the losses for AC and DC, the difference is that AC can easily be stepped up and down in voltage with a simple transformer, whereas DC is much harder (100 years agao VERY hard and cost prohibitive). The loss is a factor of current, wire resistance(gauge and type of metal used), and length. More current is more loss, more resistance is more loss, more length is more loss. In the end we care about watts, 100 watt light bulb is as bright at any voltage. We can trade voltage with current and keep the watts the same. Many of us have 200 amp 240vac service to our house, go outside and see how thick the cable is going into the house. It has to be sized to handle 200 amps. Many of us, as an example, have 7200VAC 'primary' voltage stepped down with a transformer to 240VAC, when we are using 200 amps there is only 6.7 amps going into the transformer! Before the substation, let's say your town is fed by a 250,000VAC transmission line, you only represent less then 0.2 amps at that voltage. The same thickness/material of wire can handle a certain currant at ANY voltage, all we have to do is improve the insulation-the wire going into your house has a rubber jacket that is 'safe' to touch (please don't do this, the rubber is often old and brittle), the high voltage line is hung from several feet of ceramic insulator, separated from other conductors by several feet, and separated from the nearest expected 'human' type exposure by a good distance too.

DC as said, cannot easily be changed in voltage. So when it was in use, the power plant was making voltage at 120 volts for your house, and it stayed 120 volts the whole time. To keep the losses acceptable, the wire has to be a -LOT- thicker. To feed a subdivision with AC at high voltage would probably be a 3/8ths thick wire. To feed it at low voltage and let everyone be able to do the same things as before, that wire would have to be pure copper as thick as your thigh..or worse.

Now that conversion due to electronics isn't so hard, there are some places in the world were there are long distance high voltage DC transmission lines.

Robert

Actually, 2,400 volts Direct Current (in Michigan) turned out to be at the practical limit for 3rd rail that wouldn't arc and short to the car or ground, so for 3,000 Volts Direct Current, the Delaware, Lackawanna & Western went back to overhead wire, but with more complicated catenary. The extra wires somewhat reduced the need for feed wires. Black Mesa & Lake Powell's 50,000 Volts Alternating Current was fed from one end of the line, with no extra supply elsewhere. The voltage drop at the other end of the line still left more than 25,000 Volts available, enough to run the train. Amtrak's 25,000 Volt system to Boston has relatively frequent substations to prevent too much voltage drop.
Modern rotary converters can "coast" during momentary electrical supply disruptions that might damage electronic replacements. Here, the expensive old fashioned technology has a reliability advantage under certain conditions.

Here in Fort Collins, Colorado the overhead power is derived by passing 480 volt 60 HZ 3 Phase AC in a floating delta configuration through a full wave 3 phase bridge rectifier using 12 1500 PIV 250 Amp silicon diodes. The output comes out to about 680 volts DC. We use selenium stacks for transient suppression. So far it has proved to be a bullet proof and reliable setup.

Here is a little side story. The three plus mile transbay tunnel is the biggest gap of substations on BART. The third rail in the tubes are highly clad to increase conductivity. Loma Priata knocked one at the San Francisco end of the tubes but the Oakland one remained operational. Or maybe it was visa versa. BART was able to limp the trains in the tunnel out on the remaining one. No adjacent substations on BART are fed from the same portion of the net.

Michigan Railways was the line I was thinking of when I said above that 2400VDC wasn't successful. IIRC the line only ran a year or so before revising the voltage to 1200. I've seen some of the porcelain insulators that line used for trolley rail chairs... they look like high tension insulators. On the CTA, we just used wood.