There are two contact rails in the London Underground: one with 420 V DC and one with -210 V DC.

Why not use 630 V and 0 V? Wouldn't it make one of the rails safe for people?


4 Answers 4


The reason for using an insulated return system and not an insulated feed (third rail) and return through the running rails is historical, as the government in c. 1900 required total voltage drop to be limited to 7 volts to prevent electrolysis damage by stray currents to nearby buried metal structures (a great deal of the Underground is in tunnel). A fourth rail was cheaper than heavy cables and boosters.

Although the traction supply is not directly referenced to earth, each section feeder station has potential divider resistors. The positive and negative traction rails are connected to the continuous running rail (which is bonded to earth) via high wattage high value resistors (usually of the order of 7.5 to 10 kilohms) to reference positive 420V above earth and negative 210V below earth.

This arbitrary fixing of voltage enables earth fault relays in each main section to detect positive and negative traction earth faults and indicate them at the Network Operations Centre and Line Control Rooms. An earth fault on one traction rail will drive the other rail to a full 630V positive or negative. Trains in that section are unaffected by a single earth fault, and can still run.

A modernization programme is in place to increase traction supply to 750V (+500 and -250) and lines are being converted as newer compatible rolling stock is introduced.

Where sections of line are shared with conventional surface railways which have third-rail only rolling stock, the centre rail is earthed and bonded to the running rail, and the outer rail is fed at +630 volts at present. Both types of train can run using this arrangement.

There have been a couple of cases (the Great Northern and City line 1977, and the East London line 2010) where lines formerly operated by London Underground have been transferred to National Rail ownership and operated as part of mainly above-ground rail services. These have been converted from four-rail to standard National Rail three-rail supply. Since 1900, advances in technology (e.g. insulated supports for the running rail) have removed the buried metal corrosion risk. The cost of converting the whole London Underground supply system and all the rolling stock is much too great.

  • \$\begingroup\$ With regard to the safety issue, electrocution incidents are mercifully rare, and changing the centre rail to 0V is unlikely to have any effect; for anyone straying onto the track, the greatest risk is being hit by a train. \$\endgroup\$
    – jayben
    Mar 2, 2022 at 13:08
  • \$\begingroup\$ @jayben - Indeed. I guess a 210v DC shock might be more survivable than a 630 to 750 volt one, but people shouldn't be on the track anyway. All rail staff are trained never to touch or step on a conductor rail, even if it is assumed to be 'dead'. The only exception being electrification staff, who first apply short-circuiting devices. In the UK, railways are fenced, and any trespass by the public is a criminal offence. \$\endgroup\$ Mar 2, 2022 at 13:16
  • \$\begingroup\$ Wouldn't a symmetric split (+/- 315V) be better for the fault detection purpose? (which, as I understand, is the only reason not to use a 0 / +630V split) \$\endgroup\$ Mar 4, 2022 at 11:06
  • \$\begingroup\$ @DmitryGrigoryev - the reason to split the supply was to avoid or minimise stray return currents passing through soil causing electrolytic damage/corrosion to nearby buried metal structures. The fault detection was a bonus. I believe it is the negative current which causes the damage, so that has the smaller of the splits. A surface railway (The South Eastern Railway) contemplated a scheme with 1500v split +750 and -750 volts using two outside rails, one each side of the running rails, but that never happened due to the 1922 Grouping and the Southern Railway chose +660 volts. \$\endgroup\$ Mar 4, 2022 at 12:00

Neither of the power rails are earth referenced - i.e. they are floating with respect to earth (and the running tracks). If one of them was at 0V (by this I presume you mean earth referenced) it would mean that rail was essentially inert, but the other rail would then be extremely dangerous to touch. This would not be an improvement in safety.

The reason the voltages are different is that the positive rail sits higher above the running tracks than the negative rail. This allows it to have thicker insulators, and hence it can support a higher voltage differential before breakdown.

  • 1
    \$\begingroup\$ I'm confused. If the power rails are floating with respect to earth, doesn't that make it meaningless to say that the voltages are different, since the (absolute value of the) voltage of either rail can freely swing anywhere from 0 V to 630 V? \$\endgroup\$ Mar 2, 2022 at 12:30
  • \$\begingroup\$ @TannerSwett I think they're floating with respect to earth but the difference between them is fixed. If one is at -500 V wrt to earth the other is at 130 V. \$\endgroup\$
    – Daniel
    Mar 2, 2022 at 14:01
  • \$\begingroup\$ @TannerSwett - the voltages are referenced to earth by the potential divider at each substation. \$\endgroup\$ Mar 2, 2022 at 17:57
  • \$\begingroup\$ I don't think there's much safety in a +420V rail as opposed to +630V. Touching either of them is deadly. \$\endgroup\$ Mar 3, 2022 at 10:34
  • \$\begingroup\$ The insulators supporting the current rails will withstand very much more than even the full 630 volts. The positive rail has taller insulators so it can sit higher. The insulators on the old 1200 volt third-rail system from Manchester to Bury were practically the same size as LT ones. \$\endgroup\$ Mar 4, 2022 at 10:38

Using +630V instead of +420V is indeed a possible option, and it would make the railroad equipment somewhat simpler and safer. This is what is often done with household electricity in most countries, where only one wire is "live".

It would require better insulators to cope with a higher voltage though. Perhaps such insulators were not available or too expensive when the first railway sections were built, so the voltage was split. And once these first sections have set a de-facto standard, other sections were respecting it for compatibility, even if high voltage insulators became available.

Splitting equally (+/- 315V) would have been better from electrical standpoint, but it seems that the higher height of the 420V insulators is useful in itself, as it prevents the 420V collector shoes from accidentally touching the -210V rail, which is lower.

  • \$\begingroup\$ Insulators for 630 volts were available, the reason for splitting is not to economise on insulator material, that was just a side effect. \$\endgroup\$ Mar 4, 2022 at 10:34

I believe an original reason was to be able to start the trains off 210 volts then switch to 420 volts then 630 volts as the motors sped up. Avoiding the need for such massive heat producing current limiting resistors onboard the trains.

According to an engineer working on the Underground, In the deep Underground lines, removing waste heat is a major engineering problem as ventilation is harder.

  • \$\begingroup\$ This answer is partly inaccurate. Trains are always connected to the full traction voltage 630 or 750 volts.. Limiting starting current with classic DC motors is achieved by first connecting each pair in series with resistances in series, switching out the resistances in steps, then reconnecting the motors in parallel with all resistances again in circuit, and switching out resistance until the motors have the full line voltage. Yes, this does create heat. Modern trains have three-phase AC motors with variable speed drive so heat is not such an issue. \$\endgroup\$ Mar 4, 2022 at 10:30
  • \$\begingroup\$ @MichaelHarvey Googling for information on the speed control, your comment is by far the cleanest description I've seen. Do you have a source, for further reading? \$\endgroup\$
    – Chris H
    May 17, 2022 at 14:21
  • 1
    \$\begingroup\$ @ChrisH for classic DC traction motors, here is as good a site as any mytech-info.com/2014/12/… \$\endgroup\$ May 17, 2022 at 15:04
  • \$\begingroup\$ If running on 210 V or 420 V then where was the return path? The running rails have track circuits. Switching the return to the axles sounds problematic. This answer doesn't sound right. \$\endgroup\$
    – Transistor
    Oct 13, 2022 at 17:56
  • \$\begingroup\$ @Transistor - the 'feed' and 'return' paths are always via the two conductor rails. \$\endgroup\$ Oct 31, 2022 at 9:18

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.