25kV AC vs 750V DC

[Class377/5]

Almost every DC compatible unit delivered here in the past decade or so has had the power artificially limited on the DC as it would overload the traction supply system. Compare a 377/5 at full acceleration on the Brighton main line vs the MML.

One way to describle the difference is on DC the unit feels like its being dragged and on AC it floats.

 

[edwin_m]

Another benefit of AC is the relative simplicity of voltage conversion of an AC network (by transformers).

Another benefit of AC power supplies is that they follow naturally from the supply grid power (which is AC), without the need for capacitive conversion to DC.

Indeed, until the recent advent of power electronics at reasonable prices there was now way to change the voltage of a high power DC supply without losing most of it as heat. The supply to the Southern is taken from the grid at relatively few supply points, then fed as AC (typically 33kV) down trackside cables to the substations where they are transformed down and rectified to DC (not sure where "capacitive conversion" comes in to it though).

With an AC system the rectifiers are on the train instead of the substation. Traditionally the motors needed DC, and although modern electrics have AC motors these need variable frequency so the supply is converted to DC and power electronics used to create a suitable AC for the motors.

 

[matchmaker]

I'm fairly sure that anything past Weymouth should be grey on that.

And I'm not sure that Orkney, Shetland and the Western Isles have much 25kv electrification

 

[starrymarkb]

Over the years I've lost shoes twice. Once between Victoria and Clapham (they just fell off, why? well errr) and between London Rd and Mouslecombe, rock fall ripped off two. I've also been on trains that have splutted and jerked along due to ice on the rail. Plus delays die to electrical supply issues. Oh and a naughty shoe on another train ripped up about 2 miles of rail from their pots.

From what I've read train speeds etc are even resticted by power through the over head. So high speed trains running under 1500dc (NL) and 3000dc (B) hace much lower speeds and total power. Even German ICE's and French TGV's have a lower top speed on the German 15kv AC lines than the 25kv lines on the LGV's in F,B,NL etc.

Speaking of ICEs, the currents drawn when an ICE 3 is under DC is too high for a single pantograph, so a 8 car set will have 2 pans raised.

Which makes me wonder if a 16 coach 374 will have four pans up under Belgian/Dutch DC (I know they will have two up under AC)

 

[JB_B]

But apart from the capital cost, running cost, safety and performance issues, what has 25kV ever done for us?

Part of your perception is down to how stuff is reported, part is perhaps you not being aware.

I'm sure that's correct - particularly true for reactionary delay when you'll almost always be told specifically if the wires are down - rather than just "broken down train". It's also impossible for passengers to gauge the additional failures (relative to OHLE) in areas not immediately obviously related to the 3rd rail itself (e.g signalling, more line-side equipment to go wrong, more complex point-work, etc).

Possibly the apparent relative physical fragility of OHLE (to the uninformed eye) also contributes to the perception (illusion?) that 3rd rail is more reliable.

 

[edwin_m]

One of the big problems with conversion of third rail to DC is the complexity and expense of the changeover points due to the different earth strategies. Third rail you want everything going back down the rails to reduce electrolysis from the DC. 25kv would be dangerous unless earth strongly bonded. I'm not really sure why we couldn't put rectifiers In all stock, switch third rail to AC bit by bit, then strongly bonded to earth, then wires put in as a way of phasing out DC though.

Another solution could be if the battery electrostar could be applied as a mod to the existing fleet, we could keep the DC and AC systems apart by a few miles and use batteries in the gap.

Any UK EMUs built in the last couple of decades can switch between DC and AC supply in a few seconds, or have the mountings for the missing AC or DC equipment to be attached easily. So DC to AC conversion is relatively easy, at least where the signalling equipment is dual immune. Once a particular section had all its OLE installed, provided the signalling is dual-immune changeover could be done during a weekend shutdown which would include connection of the necessary electrical bonding.

Much of the Southern uses both trains and signalling that are relatively easily converted. However the inner areas in particular are more difficult as the 455s for example can't easily be converted and the 1970s/80s signalling would need extensive immunisation or replacement. Hence any large scale conversion programme would most likely start away from London and work inwards, not reaching the inner areas until the problem trains and signalling are life-expired.

 

[hassaanhc]

I'm sure that's correct - particularly true for reactionary delay when you'll almost always be told specifically if the wires are down - rather than just "broken down train". It's also impossible for passengers to gauge the additional failures (relative to OHLE) in areas not immediately obviously related to the 3rd rail itself (e.g signalling, more line-side equipment to go wrong, more complex point-work, etc).

Possibly the apparent relative physical fragility of OHLE (to the uninformed eye) also contributes to the perception (illusion?) that 3rd rail is more reliable.

That is also true, and isn't it only the ECML which has a lot of issues with fragile wiring? How often do overhead power failures happen? South West Trains have had several already in 2014, most recent was a couple of days ago on some lines near Waterloo, two 750v cables combusted days apart at Twickenham, one at Wimbledon, plus a few failures in the Bournemouth/Portsmouth areas. Plus one evening a gapped train at Clapham Junction across more than one line and the rescue unit got gapped too

 

[Jonny]

Not sure what you're driving at here. There are places where there is both 3rd rail and OHLE on the same lines and yet there aren't any issues.

The issue is with DC electrification, there needs to be a great deal of bonding to deal with stray currents that may cause electrolytic corrosion. This is not such an issue for AC as the repeatedly reversing current has the effect of reversing or preventing any electrolytic reactions.

I understand there are specific engineering solutions that enable such issues to be overcome, however they tend to be "difficult" (as in expensive).

 

[JB_B]

No. It's because you have forgotten all those winter mornings when the entire network south of the Thames was paralysed due to snow and ice while everything north of the Thames that ran off the OHLE ran normally. The same when parts of the south of the UK suffered from flooding.

I'm only traveling once or twice a week and that's usually off-peak so I probably miss the worst of the icing delays. I do remember 3rd Dec 2010 as a particularly bad one - IIRC the only train out of Brighton that morning was the sole FGW service.

I can imagine that 3rd rail operation could create additional problems when lines are flooded - can you quantify the extra constraints? I can recall seeing the other line flooded to railhead level but still in use (near Haywards Heath) - would you be going much deeper anyway?

 

[swt_passenger]

The issue is with DC electrification, there needs to be a great deal of bonding to deal with stray currents that may cause electrolytic corrosion. This is not such an issue for AC as the repeatedly reversing current has the effect of reversing or preventing any electrolytic reactions.

I understand there are specific engineering solutions that enable such issues to be overcome, however they tend to be "difficult" (as in expensive).

Another point is that the latest technology fitted in the Thameslink dual voltage area (the example we regularly discuss) may have been designed to provide for the utmost levels of electrical safety and corrosion protection, and would therefore form a precedent for any future fits. So the fact that some existing dual voltage changeover sections are relatively simple doesn't necessarily mean that any new installations won't be as complex as Thameslink's system.

I guess we'll see in due course when they get round to the Basingstoke area works...

 

[Lou92]

OK, I have a blonde question to ask: if the overhead wire is the live wire, and the neutral wire is the track, then wouldn't that make the track as 'hot' as the overhead line? I mean, aren't they switching back and forth from live to neutral 50 times per second? I understand that DC power lines can be monopoles, but AC requires 2 wires to provide the full cycle.

 

[swt_passenger]

On a very straightforward AC system (say the output from a simple portable generator) the two lines would alternate, positive and negative with respect to each other each cycle at 50 Hz, but in a practical system the 'live wire' alternates between positive and negative values, and the neutral remains at zero.

Using a household analogy, the live wire varies between +230V and -230V, and the neutral remains at nominally zero volts, very close to earth potential. The AC railway is basically the same, but at a much higher nominal voltage.

 

[DaveNewcastle]

. . . . if the overhead wire is the live wire, and the neutral wire is the track, then wouldn't that make the track as 'hot' as the overhead line?
. . . AC requires 2 wires to provide the full cycle.

Yes, AC (as well as DC) requires a minimum of 2 conductors between the power source and the power load, and yes, they 'alternate' between each other in terms of which conductor is positive and which is negative, but (and this is the crucial bit) all of this is simply a description of the circuit in isolation. There is no risk to a person from such a circuit until they make contact with 2 parts of the circuit which are at different voltages. In practice, any one part of the circuit can, by design or accident, be connected to earth without changing the behaviour of the circuit. In some circuits, one of the 2 conductors will actually be replaced by an earth connection. When this is done by design, then as much current will be returned via the earth as is delivered by the 'hot' conductor. But there is negligible hazard as the risk to life would require contact between the high current carrying earth and some other surface which isn't at the same voltage, and as the earth or railway track are good conductors, then they will all be at the same potential (or voltage).

I hesitate, however, so say your proposition is wrong, as there is acually a hazard in the case of high voltage systems such as power distribution and overhead power on the railways, as there is always going to be some finite resistance along any conductor (including the earth), and as a consequence, there will be a voltage drop along the conductor. In extreme cases of high currents and high voltages, then we should expect that any part of the circuit may be at a voltage relative to earth, and it will be where return currents pass through long lengths of rail, cable and bondings.
It is the bonding of exposed return conductors to earth at various points that minimises the risk to persons.

Importantly, is the arrangement of dealing with faults by using the earth as a high capacity retun path to the source which will force a fuse or circuit breaker to detect the excessive current and disconnect the circuit. By design, this means that even 'earthed' surfaces should be allowed to reach high voltages briefly, without hazard to persons, which is one reason why even return circuit conductors which are referred to earth will be installed with insulation. (But there are a few different earthing arrangements in electrical supplies, with different risks and benefits).

So while railway track will not be as 'hot' as the 'live' conductor, it should always be treated as if it might be at a dangerous voltage.

 

[LNW-GW Joint]

I believe the autotransformer systems being used for new AC electrification further improves the business case for AC (lower installation and running costs).
This is currently fitted on WCML South and is being installed on the new NW wiring project.
You need fewer (expensive) grid feeder points and there is less interference.

 

[Lou92]

Swt and Dave - many thanks for taking the time to answer my question! X

 

[WatcherZero]

I must admit the autotransformer technology has always puzzled me, something about using a 50kv supply which is then stepped down into two 25kv feeds?

 

[swt_passenger]

I must admit the autotransformer technology has always puzzled me, something about using a 50kv supply which is then stepped down into two 25kv feeds?

There's a reasonably simplified description of the auto transformer system on pages 20/21 of the electrification RUS, which might help - it isn't too technical:

http://www.networkrail.co.uk/browse...ation strategy/networkrus_electrification.pdf

I reckon you can think of it as a doubled up transmission system where the trains sit between one side of the 50 kV supply and a centre-tapped connection which is connected to the rails. If you compare it with the drawing of a conventional system, it's clear that the main difference is that the rail's connection is to a centre tap off the supply transformer secondary, rather than one end of the conventional transformer secondary.

Another view point is to think of each autotransformer winding as a 2:1 reduction transformer...

Hope that helps...

 

[DaveNewcastle]

I must admit the autotransformer technology has always puzzled me, something about using a 50kv supply which is then stepped down into two 25kv feeds?

Not so much "stepped down" as a 50kV supply with a centre tap (which creates two 25kV supplies from the 3 conductors, which appear as 2 supplies out of phase with each other plus a return or zero).

The purpose is to provide additional current to the principal conductor - the overhead live wire - when it is needed most, that is during high current demands. We know that any load will tend to pull the supply voltage down, though in reality it will also pull the zero towards the supply by the same process; we just don't think of the zero moving, as we use that as a reference. But in a twin 25kV circuit, any load which pulls the zero towards the supply voltage, is also pulling it away from the other, out-of-phase, supply, with the apparent effect of increasing the supply voltage on the out-of-phase line, which will be of use in supplying additional power elsewhere.
Connections via a transformer are made at periodic distances between the out-of-phase supply and the main supply, and during times when the principal supply is under load (and therefore under voltage), the transformer provides additional current into the principal supply from the higher voltage . These are simply wired to transfer energy from the out-of-phase conductor to the principal conductor. The circuit might even appear pointless utill you consider the effect of load current on the 2 supply voltages.

If you're familiar with the return paths of currents in a 3-phase supply, and how the return current from one phase can return via the loads on the other phases (and therefore the neutral conductor does not need to be 3 times the gauge of the line conductors) then you're more than half way there already!

EDIT: Ah! I see that swt_passenger has replied already - hopefully our two replies are consistent and complimentary! Just like two 25kV power supplies!

 

[Hophead]

I'll chuck in another drawback of 3rd-rail electrification (as currently applied): current isolation. If there's someone on the tracks who shouldn't be there or a fatality, the electricity needs to be switched off. Not only does this stop everything (that's electric) in the vicinity, it usually extends into a greater area. Last week's London Road (or was it Moulsecoomb) incident had the power switched off as far away as Preston Park, causing delays on the main line.

I've read that Network Rail were looking at additional isolation in the London Bridge area, at least, following excessive delays over a wide area - don't know if anything's come of this.

If your electrification's 5m up in the air you don't, at least, need to push the big red 'OFF' button when a trespasser goes for a walk along the line.

 

[Bald Rick]

I'll chuck in another drawback of 3rd-rail electrification (as currently applied): current isolation. If there's someone on the tracks who shouldn't be there or a fatality, the electricity needs to be switched off. Not only does this stop everything (that's electric) in the vicinity, it usually extends into a greater area. Last week's London Road (or was it Moulsecoomb) incident had the power switched off as far away as Preston Park, causing delays on the main line.

I've read that Network Rail were looking at additional isolation in the London Bridge area, at least, following excessive delays over a wide area - don't know if anything's come of this.

If your electrification's 5m up in the air you don't, at least, need to push the big red 'OFF' button when a trespasser goes for a walk along the line.

This happens pretty much every day somewhere on the DC network. Yesterday's was at Dover.

There is a project in CP5 to deliver 'quicker faster' isolations, which in effect means remote operation of the hook switches that divide the relevant 3rd rail sections up into smaller chunks, particularly at junctions. Also included are remotely operated negative short circuiting devices to eliminate the risk from residual current.

The London Bridge area will get them as and when sections are remodelled, so the first batch should be in use early next year.

 

[noddingdonkey]

One of the modern reasons for favouring 25kv (though I believe doubts over the future of DC go right back to the 50's) was several incidents in recent years of under supply when a 10 (or 12, I forget which) carriage train went past greedily sucking up all the juice causing other shorter trains on the line to stall from under voltage.

But that happened the first time they sent a 91 to Skipton - every 308 between Shipley and Skipton failed.

 

[starrymarkb]

But that happened the first time they sent a 91 to Skipton - every 308 between Shipley and Skipton failed.

Though that was a long section with just one low powered feed (ie fine for the usual units)

 

[tsr]

This happens pretty much every day somewhere on the DC network. Yesterday's was at Dover.

And yet there is relatively little said here about the major problems on the North London line as well as via Tottenham Hale due to yesterday's OHLE supply issues. These stranded around 6 trains at the peak of the incident, I believe, including a 378 stuck between stations, on board which the air handling systems failed almost immediately, resulting in multiple passengers feeling unwell. So it goes to show that sometimes, despite trespassing causing more delays to 3rd rail electrified lines, it is impossible to avoid delays on electrified routes for many other reasons.

Had the supply systems for the NLL service had greater redundancy, I suppose that may well have been of assistance. From my personal experience, other than emergency isolations and blankets of ice, electrical supply infrastructure problems on 3rd rail routes generally seem to limit themselves to fairly small areas, meaning that routes may only be closed for a fairly short distance, once said hook switches are dealt with. Of course, if there are too few crossovers (a major bugbear of mine), this makes less of a difference, because starting & terminating short is not as readily made an option.

 

[edwin_m]

A lot of the problems we have with OLE are likely to be to do with skimping on design, construction or maintenance. Other countries that do things properly seem to have far fewer issues - a lesson that Network Rail has hopefully now learned.

There have been several winters recently when icing has affected large parts of the third rail network at the same time.

 

[jopsuk]

Crossovers may help when there is disruption but they do require far more maintenance than plain line and can thus cause disruption!

 

[Taunton]

Crossovers may help when there is disruption but they do require far more maintenance than plain line and can thus cause disruption!

I have never understood how, a generation ago and more, we could have huge numbers of crossovers and points, all seemingly reliable (I can never recall a single points failure in all the upward of 100 point ends at Taunton station in the days of mechanical signalling) and straightforward to install. Nowadays they seem to cost £1 million each, are hugely complex to install, are provided too sparingly, and the few that are left suffer from regular "points failures" like never before.

 

[Nym]

Probably due to them being far more (and sometimes needlessly) complex.

Good old fashioned points system (Such as air or straight mechanical) with mechanical locking and a simple 10 terminal PL&D 4' box is a lot simpler than a modern Shorelock system, and this is a case of there simply being a lot more to go wrong.

Compare a set of Chairlocks or Clamplocks to Shorelocks and their oceans apart, there's even more complex than the old M63s that they replace. (And in my opinion, a slightly worse / cheaper design).

Or perhaps it could be to do with more parts of the points being locked against the signalling system?

Most of the shorelock failures I know of though are more to do with the points machine failing in some way rather than the switchblades or locking and detection.

 

[edwin_m]

Probably had a lot to do with mainly manual operation and a large number of staff available nearby for maintenance and to attend to problems quickly. Modern points also tend to have significantly higher turnout speeds, which takes advantage of train performance to reduce journey times and junction occupancy. This means they are much larger with more complex mechanisms for control and detection.

 

[Class 170101]

And yet there is relatively little said here about the major problems on the North London line as well as via Tottenham Hale due to yesterday's OHLE supply issues. These stranded around 6 trains at the peak of the incident, I believe, including a 378 stuck between stations, on board which the air handling systems failed almost immediately, resulting in multiple passengers feeling unwell. So it goes to show that sometimes, despite trespassing causing more delays to 3rd rail electrified lines, it is impossible to avoid delays on electrified routes for many other reasons.

Had the supply systems for the NLL service had greater redundancy, I suppose that may well have been of assistance.

Is there a feeder station on the NLL in this area? The nearest ones in this area I can think of are Northumberland Park, Bow (Junction) and Rye House, all on West Anglia / Great Eastern.

Additionally in times gone by parts of the Great Eastern, ECML out of Kings Cross were only electrified originally at 6.25kv ac as there were concerns about the clearances of wiring with 25kv ac.

 

[yorksrob]

I've also been on trains that have splutted and jerked along due to ice on the rail.

Although that's not exclusive to the third rail.

I was once travelling on a Eurostar that kept stopping and starting all the way between Leeds and Donny due to ice on the wire, where it terminated.

(the event is etched on my memory because it made me miss the first leg of the Sussex Slammer ).