AC Third Rail?

[trash80]

I'm NOT suggesting it should be done, but i was just wondering is there any technical reason why third (or fourth rail) systems couldn't be AC not DC? And if not have there been any such systems in the world?

 

[Muzer]

I don't know of any technical reason why this couldn't be done. The question is, why? Once power electronics were sufficiently developed to use AC reliably on trains, you may as well up the voltage considerably to reduce losses and current. Once you do that, suddenly the clearance between the third rail and the running rails (not to mention track workers) is much too small, so you need to put the power somewhere else - hence, overhead wires.

 

[GRALISTAIR]

I'm NOT suggesting it should be done, but i was just wondering is there any technical reason why third (or fourth rail) systems couldn't be AC not DC? And if not have there been any such systems in the world?

I don't know of any technical reason why this couldn't be done. The question is, why? Once power electronics were sufficiently developed to use AC reliably on trains, you may as well up the voltage considerably to reduce losses and current. Once you do that, suddenly the clearance between the third rail and the running rails (not to mention track workers) is much too small, so you need to put the power somewhere else - hence, overhead wires.

And similarly you can use DC on overhead obviously.

 

[najaB]

I'm NOT suggesting it should be done, but i was just wondering is there any technical reason why third (or fourth rail) systems couldn't be AC not DC? And if not have there been any such systems in the world?

No reason it couldn't be done but one of the reasons to use AC is that it allows you to use much higher voltages. You definitely wouldn't want to run 25kV at ground level!

Edit: Must remember to read the thread before posting, Muzer said it much better than I did.

 

[jopsuk]

Third rail needs to be low voltage to avoid arcing to ground. To deliver the power requires a high current (as Power = Voltage x Current). High current needs low resistance per linear distance (unless you're building a heater or an incandescent light bulb). This is achieved using a big chunky rail.

However, due to the Skin Effect, high frequency (and UK grid 50Hz is high frequency) AC there's actually no point using a chunky conductor. It's only the outer centimetre or so that actually carries the current. So the conductance of that chunky rail is wasted.

This is why AC is better transmitted as high voltage (DC is as well, to be fair, and super high voltage power is now DC as power conversion is now efficient- AC took hold because traditional transformers were the only efficient way to step voltage up or down) - because you're then using relatively low current you can have thinner (or even tubular) conductors.

Electricity is weird.

 

[trash80]

Thanks for the interesting and informative replies, i was travelling on both modes of EMU on Sunday and the thought just popped in my head.

 

[KingDaveRa]

Third rail needs to be low voltage to avoid arcing to ground. To deliver the power requires a high current (as Power = Voltage x Current). High current needs low resistance per linear distance (unless you're building a heater or an incandescent light bulb). This is achieved using a big chunky rail.

However, due to the Skin Effect, high frequency (and UK grid 50Hz is high frequency) AC there's actually no point using a chunky conductor. It's only the outer centimetre or so that actually carries the current. So the conductance of that chunky rail is wasted.

This is why AC is better transmitted as high voltage (DC is as well, to be fair, and super high voltage power is now DC as power conversion is now efficient- AC took hold because traditional transformers were the only efficient way to step voltage up or down) - because you're then using relatively low current you can have thinner (or even tubular) conductors.

Electricity is weird.

This is a great explanation. Electricity never ceases to amaze (and sometimes confound) me.

I'm curious, were there any railway systems using oddball power supplies or methods of supplying it? LU springs to mind as it's different to most with it's two rail power supply, but were there others? Say using the running rails to supply power, or something?

 

[Bevan Price]

This is a great explanation. Electricity never ceases to amaze (and sometimes confound) me.

I'm curious, were there any railway systems using oddball power supplies or methods of supplying it? LU springs to mind as it's different to most with it's two rail power supply, but were there others? Say using the running rails to supply power, or something?

I would not fancy being on a train transmitting power through the axles. Should the necessary insulation fail, the entire train would become "live".

It still used overhead wiring, but Italy used to use a 3-phase electrical supply on some routes.

 

[jopsuk]

I would not fancy being on a train transmitting power through the axles. Should the necessary insulation fail, the entire train would become "live".

For most electrification (LU is an exception) the axles, wheels and track do of course form part of the electrical circuit

It still used overhead wiring, but Italy used to use a 3-phase electrical supply on some routes.

3-phase is tricky, requires multiple wires. Still a small number of mountain rack railways using it. They never worry about turning around!

 

[AM9]

This is a great explanation. Electricity never ceases to amaze (and sometimes confound) me.

I'm curious, were there any railway systems using oddball power supplies or methods of supplying it? LU springs to mind as it's different to most with it's two rail power supply, but were there others? Say using the running rails to supply power, or something?

LU uses a third (side) and fourth (centre) rail to carry the nominal 630VDC. The voltage is distributed as +420V on the third rail and -210V on the fourth. If the normal practice of full positive voltage was connected to the 3rd rail, and the return negative connection via the running rails, (as on National Rail), there would be continuous large currents passing through the (originally) cast iron and more recently steel tunnel lining segments. This would cause large circulating currents that would in time cause electrlytic corrosion of the tunnel lining.

 

[Bevan Price]

For most electrification (LU is an exception) the axles, wheels and track do of course form part of the electrical circuit

True, but the voltage/ current is much depleted at that stage, having been used to power the train.

 

[edwin_m]

The track is at the same voltage as the earth, and as the train sits on metal wheels it is at the same voltage as the track. So you won't get a shock between the train and anything that is also electrically connected to the earth, such as a station platform.

This is something of an oversimplification, especially for DC systems where there is some insulation between the rails and the earth so voltage differences can develop. Special measures are needed to prevent these becoming dangerous.

 

[4973]

True, but the voltage/ current is much depleted at that stage, having been used to power the train.

I do hope you are not suggesting that current is dissipated in the motors. While the voltage (wrt the earth return) will be reduced the current is constant through the circuit.

The fact that the earth return current per wheel is reduced is immaterial as it is only because of the number of wheels involved.


And I had to smile at the idea of 50Hz being high frequency - to me that designation starts at 100 MHz.

 

[PeterC]

Reading all the technicalities I get the feeling that the O and A level physics syliabi of my youth were a little lacking in the practicalities.

 

[najaB]

And I had to smile at the idea of 50Hz being high frequency - to me that designation starts at 100 MHz.

Indeed. It's my understanding that the skin effect at 50Hz is fairly limited is it not?

 

[D365]

Reading all the technicalities I get the feeling that the O and A level physics syliabi of my youth were a little lacking in the practicalities.

Neither of those were at all electrical-heavy for me - ironic as most of my teachers had studied as engineers...

 

[OliverS]

Indeed. It's my understanding that the skin effect at 50Hz is fairly limited is it not?

A couple of millimetres in copper as I recall. Similar in aluminium and so probably of that order in steel.

 

[AM9]

Indeed. It's my understanding that the skin effect at 50Hz is fairly limited is it not?

It does have an effect at the power levels used in traction. At 50Hz current is limited to about 10mm from the surface of a steel rail, - at 60Hz it goes down to 8mm-ish. When there's 10000A involved, rail just can't deliver.
On OLE however, a 12mm od copper cable not only has less skin effect, it is also expected to carry a few hundred amps for the equivalent traction power, and a 10-20% voltage drop is easier to handle with transfor winding options.

 

[notlob.divad]

True, but the voltage/ current is much depleted at that stage, having been used to power the train.

Please no.

The very basics of electricity.
Voltage is simply the potential difference between 2 location. Earth nominally being 0 volts. Therefore as energy is converted into another form, the voltage decreases.

Current is the flow of electricity through a conductor between 2 locations from the lower potential to the higher potential. (blame Edison) as it is the negative electrons that move.

The flow (current) across multiple branches of a circuit adds up, but is the same at any point along a single conductor and with side or a load.

In a train as has been pointed out, the total current flowing in through the pantographs/3rd rail shoes must equal the total flow through all the wheels back to earth. How it flows through the various conductors inbetween is almost immaterial as to how you transfer the energy. So long as you can supply the correct amount of energy to feed all of the components.

Finally, always remember that it is the current (flow) that does the damage. The human body will happily sit at whatever potential you decide to put it at. However you ask it to conduct the flow of electrons between 2 different potentials and it tends to dislike anything above about 30mA.

 

[4973]

As the old jingle goes "It's Volts that jolts and mils (milliamps) that kills".

Having said that voltages in excess of 25-30 kV are reputed to have a risk of burning the nervous system. I suspect it's a little exaggerated.

 

[dgl]

I suppose it's like power line operatives, they actually connect themselves and the platform they are working on to the power line, it doesn't really matter what the voltage is so long as the voltage difference between your body and the power source is 0V and there is no path to earth/a different voltage (although that voltage would probably have to be at least ~100V different to do any harm).

 

[najaB]

It does have an effect at the power levels used in traction. At 50Hz current is limited to about 10mm from the surface of a steel rail, - at 60Hz it goes down to 8mm-ish. When there's 10000A involved, rail just can't deliver.

Thanks. I suppose that's the difference between day-to-day domestic power applications where it doesn't have an effect and *real* high-power situations.

 

[Deepgreen]

I occasionally muse on the combination of, say, 25kV supply at rail level, but with the transmission cabling encased in a form of insulating tube allowing shoe contact but preventing earthing and accidental contact. All the benefits of high voltage AC without the huge infrastructure required for OHLE. Wind damage disappears and snow/ice can't form on the shrouded contact surface. Now to design the interface detail to eliminate accidental contact...!

 

[Juniper Driver]

Don't panic,they've converted some 455's to AC Motors....I had a 455/9 yesterday...Sure that was my first 455/9 AC Motored unit.

 

[Deepgreen]

Don't panic,they've converted some 455's to AC Motors....I had a 455/9 yesterday...Sure that was my first 455/9 AC Motored unit.

Indeed, but still using DC third rail for current collection.

 

[bavvo]

Just out of interest, why was 25KV settled on as the default supply voltage? Are there diminishing returns if you go any higher, or is it just impractical? Were there any experiments using higher voltages, say 30 - 50KV?

 

[GRALISTAIR]

Just out of interest, why was 25KV settled on as the default supply voltage? Are there diminishing returns if you go any higher, or is it just impractical? Were there any experiments using higher voltages, say 30 - 50KV?

You might try the sticky thread (this thread is also referenced) - loads of information - even has a link to discussion about 50 kV - plus wiki entries , general info etc.

http://www.railforums.co.uk/showthread.php?t=138487

 

[najaB]

Were there any experiments using higher voltages, say 30 - 50KV?

There are a few railways that use 50KV.

 

[GB]

Just out of interest, why was 25KV settled on as the default supply voltage? Are there diminishing returns if you go any higher, or is it just impractical? Were there any experiments using higher voltages, say 30 - 50KV?

From Wiki

The choice of 25 kV was related to the efficiency of power transmission as a function of voltage and cost, not based on a neat and tidy ratio of the supply voltage. For a given power level, a higher voltage allows for a lower current and usually better efficiency at the greater cost for high-voltage equipment. It was found that 25 kV was an optimal point, where a higher voltage would still improve efficiency but not by a significant amount in relation to the higher costs incurred by the need for greater clearance and larger insulators.

https://en.wikipedia.org/wiki/25_kV_AC_railway_electrification

 

[Taunton]

I'm curious, were there any railway systems using oddball power supplies or methods of supplying it?

This one was oddball all round

http://volkselectricrailway.co.uk/history/the-daddy-long-legs/

LU uses a third (side) and fourth (centre) rail to carry the nominal 630VDC. The voltage is distributed as +420V on the third rail and -210V on the fourth. If the normal practice of full positive voltage was connected to the 3rd rail, and the return negative connection via the running rails, (as on National Rail), there would be continuous large currents passing through the (originally) cast iron and more recently steel tunnel lining segments. This would cause large circulating currents that would in time cause electrolytic corrosion of the tunnel lining.

Related, but I've always wondered just what were the issues that afflicted the North London line 3rd rail electrification put in during the mid-1980s, only running 2-car sets, but which caused all sorts of electrical interference, including at Highbury apparently affecting the signalling and causing electrolysis on the Piccadilly and Victoria lines well below, whereas the GN suburban line there, also 3rd rail, was not doing this. It caused, after a few years of investigation, the further changeover to 25Kv overhead and the need to use dual voltage Class 313 units. It's not as if 3rd rail lines crossing the Underground are not uncommon in London, but I recall it being written that the BR engineers did not have a basic understanding of stray currents.