-1

I have a question about electrified railways that seems obviously unsafe, but first you have to know how the "normal" third rail works.

Third rail power comes in over a middle rail and is returned via the other normal rails. That sounds plenty dangerous to me. So if you stand with one foot on the normal rail, and one foot on the third rail, your body completes the circuit and you get shocked. These DC third rail systems use 600 to 1500 DC so i think you will likely die if you step on the rails that way.

So I am wondering, if that is an acceptable risk for all third rail systems, why not just use the two normal rails and don't have a third rail at all?

Note there are already tons of systems for electrically isolating one rail from the others, whether its normal rails or third rail, as well as from the ties.

Note about steel and conductive efficiency. You can have a strip of aluminum on the side of your rail, or underneath it. A lot of third rail systems already do this. Also makes it safer by not being directly on the top.

Note about block signalling systems. This uses electric circuit completed by wheels and bogeys to signal which blocks of track are occupied. However, like i said about third rail, the power is return via the normal rails. So the normal rails already have power current going thru them, which apparently does not interfere with the block signalling system, therefore I don't see how a normal two-rail power system would interfere with it either.

I feel like I'm missing something obvious. If that is the case I'm sorry but I cannot fathom a reason for not using the two normal rails as power rails. To me it seems basically 99.99% the same thing as a third rail on the ground, so why do that? Why not just use both normal rails?

DrZ214
  • 1,029
  • 10
  • 19
  • The main problem is the solid conducting axle of every bit of rolling stock made since before Stephenson's Rocket; it will short out your tracks - it's not like a model train set where wheels are insulated from the axle. How are you going to cope with diamond crossings too? No, totally unfeasible and more dangerous. – Andy aka Dec 25 '21 at 08:56
  • @Andyaka I keep telling everyone, this is why i mentioned an aluminum strip on the side of the rail, contacted by a shoe not the wheels. But i cant force ppl to read the whole post and/or carefully and i can't stop ppl from downvoting either if they think it's a dumb question due to missing information. – DrZ214 Dec 25 '21 at 21:58

2 Answers2

4

According to Wikipedia, that scheme (which is also used on model trains) was initially used, but various practical issues like insulating axles (which have to be mechanically strong) and too-conductive sleepers made it more desirable to introduce the third rail:

Running rails for power supply The first idea for feeding electricity to a train from an external source was by using both rails on which a train runs, whereby each rail is a conductor for each polarity, and is insulated by the sleepers. This method is used by most scale model trains, however it does not work so well for large trains as the sleepers are not good insulators. Furthermore, the electric connection requires insulated wheels or insulated axles, but most insulation materials have poor mechanical properties compared with metals used for this purpose, leading to a less stable train vehicle. Nevertheless, it was sometimes used at the beginning of the development of electric trains. The oldest electric railway in Britain, the Volk's Railway in Brighton, England was originally electrified at 50 volts DC using this system (it is now a three rail system). Other railway systems that used it were the Gross-Lichterfelde Tramway and the Ungerer Tramway.

Spehro Pefhany
  • 376,485
  • 21
  • 320
  • 842
  • Thanks for this info. But I still question further because, if the power is on an inner aluminum strip, then the wheels themselves will never contact it and therefore not have to be insulated. Shoes can contact the strip like in 3rd rail. – DrZ214 Dec 24 '21 at 22:16
  • 1
    Interestingly, the problem of insulating sleepers and axles was also a factor in making _model_ railways with a 'third rail'. https://en.wikipedia.org/wiki/Third_rail_(model_rail) – Bruce Abbott Dec 24 '21 at 22:17
  • @DrZ214 So you're talking about mounting the third rail to the side of one of the load-bearing rails? Might make designing (rail) switches a bit more challenging. – Spehro Pefhany Dec 24 '21 at 22:24
  • @SpehroPefhany Completely disagree. You can have a small gap in power and still cross over the gap at speed perfectly fine. Tram lines do that with overhead wires all the time, no need for continuous connection when crossing that overhead "switch" (It's not a switch because it doesn't need to be). For rails the same thing applies but you can still connect the gaps with a lower wire that crosses under the 20 cm or so of switching rail. You just don't have to pick up power from that gap. – DrZ214 Dec 24 '21 at 22:28
  • Okay, they obviously made the conscious decision to add a (non-trivial) third rail structure. I can imagine it's possible that with advances in materials it may be more practical to avoid the third rail than it was back then. – Spehro Pefhany Dec 24 '21 at 22:30
  • Also third rail generally has been abandoned on modern high speed rail systems in favor of overhead wires and pantographs, despite the corridors being much more controlled than slow rail. – Spehro Pefhany Dec 24 '21 at 23:42
1

Note there are already tons of systems for electrically isolating one rail from the others, whether its normal rails or third rail, as well as from the ties.

Not at traction voltages. The running rails typically sit on sleepers in ballast and the impedance between the rails is in the order of ohms. Insulated joints are possible for track relay (train detection) circuits but not for traction voltages.

Note about steel and conductive efficiency. You can have a strip of aluminum on the side of your rail, or underneath it.

They're generally not required as the rail cross-sectional area is very high so resistance is low.

A lot of third rail systems already do this. Also makes it safer by not being directly on the top.

If aluminium was on top it would wear and deform with the weight of the train.

Note about block signalling systems. This uses electric circuit completed by wheels and bogeys to signal which blocks of track are occupied.

Not quite. The wheels and axles short-circuit the relay circuit so that the relay drops out. Train routes can only be set when all the required track circuits are energised thus proving them clear (rather than proving them occupied).

So the normal rails already have power current going thru them, which apparently does not interfere with the block signalling system, therefore I don't see how a normal two-rail power system would interfere with it either.

Your scheme relies on the axles short-circuiting the rails for signaling purposes but somehow not short-circuiting for traction purposes. You haven't explained how this can be achieved.

I feel like I'm missing something obvious. If that is the case I'm sorry but I cannot fathom a reason for not using the two normal rails as power rails. To me it seems basically 99.99% the same thing as a third rail on the ground, so why do that? Why not just use both normal rails?

When both running rails are used as ground returns they are at a voltage very close to ground potential and problems with leakage to earth are less of a problem. (Current returning through the ground can cause rapid corrosion of metal pipes (gas and water) running alongside the railway.) Third-rail systems stand the conducting rail off the ground on insulators - usually ceramic style.

Another problem is at points / switches / junctions. The rails cross over. Even with insulated joints there would be a massive short circuit as a wheel bridged the gap momentarily.

Further complications

  • To prevent interference between traction current and signaling track circuits very often AC track circuits are used with DC traction systems and DC track circuits with AC systems.
  • Some third-rail systems use one continuous rail for the traction return and use insulated joints in the other to form the track circuits and reverse the polarity on each consecutive track circuit so that a joint failure won't give a wrong-side failure.
  • London Underground uses four rails: two running rails with track circuits, an outer rail for DC supply and centre rail for return.
  • Other DC traction systems use insulators in both rails and require impedance bonds to allow the DC current past the joints.
  • Railway wheels are solidly attached to the axles. The axle and wheels are rigid and both wheels turn at the same speed. Differential on curves is accomplished by the slight conical profile on the wheels causing the outer wheel to run on a larger diameter than the inner wheel. It would not be reasonably practical to insulate the wheels.
Transistor
  • 168,990
  • 12
  • 186
  • 385
  • A lot of points here i have no answer to. But a few i think i do. Insulating wheels should not be necessary. If it's an aluminum strip somewhere, that can be contacted by a shoe. Points/switches/junctions, you don't need to pick up power there. 1 or 2 seconds without power will not be noticed, the train will cruise on thru. Also, your 3rd quote and response...im not sure if you didnt notice the word "not" or are simply agreeing with me there. I said "already do this", and by "this" i meant a side strip or understrip of aluminum, so it is **not** on top and won't wear out that way. – DrZ214 Dec 24 '21 at 23:01
  • However i cannot think of any way around all the other problems you mentioned so I'm accepting this answer. But, guess what, it prompts another question lol, an even more fundamental idea of why not just have a single 3rd rail and no need to return power thru the other rails. You can still complete a circuit that way. Im going to open that new q very soon. – DrZ214 Dec 24 '21 at 23:02
  • Edit, never mind, i got my answer from this post: https://electronics.stackexchange.com/questions/448188/why-are-overhead-electric-wires-allowed-to-have-higher-voltages-than-third-rails and also a little reasoning about it. Bad idea for multiple trains on a network and ofc branching paths. – DrZ214 Dec 24 '21 at 23:22
  • "*Points/switches/junctions, you don't need to pick up power there ...*" No but you do need to insulate. Try drawing a points layout and show where the insulators would be and how you would avoid the short circuit I have described. – Transistor Dec 24 '21 at 23:45