Part One of the above article generated a lot of interest and a lot of comments based around the issue of driverless trains. This certainly was the issue that seemed to interest the politicians and subsequently also some of our readers. It was not, however, the issue that Mike Brown, MD of London Underground, appeared to be desperately trying to get across to the Assembly – as a number of commentators recognised.

To save you having to re-read the transcript, the short summary below provides the gist, in a very loose way, of what was said.

Q (to Mike Brown): Are you proposing to introduce driverless trains?

A (MB): We have no plans to do so with existing stock. It is meaningless to talk about driverless trains with new stock on the Bakerloo and Piccadilly lines until we actually have the finance in place to buy new trains.

[Now repeat the question or a similar question worded differently ad nauseum along with various replies that make the point that there has been no commitment to order Bakerloo or Piccadilly line trains in the future]

How Long Is An Underground Train Expected to Last?

One can see what is concerning Mike Brown. In the past the life expectancy of underground stock was generally considered to be 30-35 years. Like an old car, after a certain time it becomes too dated and too difficult to maintain and the sensible thing is to go out and buy a new one. In 2011 he saw the last of the Victoria line 1967 tube stock run after 43 years in service and very recently the last Metropolitan line 1960 A stock ran after well over 50 years in service. In the former case, it was probably not a moment too soon. In the latter, it should have been done years ago since the trains could not provide the performance really needed for the service they provided. Indeed the only saving grace was that the signalling was also so antiquated, and for many years the track so bad, that simply replacing them with new stock would have provided little real benefit.

Ultimately, Mike is no doubt aware that when it comes to delaying replacing railway stock you can “get away with it” once or twice and possibly even half a dozen times but eventually you are going to get caught out.

When can we expect to get new trains on the Bakerloo line?

London Underground’s Tube Upgrade Plan timeline extracted from a document dated February 2011. The provisional timescale for the Bakerloo and Piccadilly lines now looks very optimistic.

In London Underground’s much publicised Tube Upgrade Plan the above diagram appears. This is referring to line upgrades of which new trains is usually a vital component. Of course this timeline gives much flexibility in when the new trains would start to appear but one would expect a pre-production train or two followed by a period of evaluation and then gradual introduction of the whole fleet. What one would not expect is the first train to appear in the last year of the line upgrade.

It is then a little concerning that it has been widely reported that Boris has, according to the Evening Standard allegedly stated that:

The first driverless trains, operating with ‘train captains’ like the DLR, will be running on the Bakerloo line by 2021.

By then the 1972 stock (which is based on the 1967 stock design) running on the Bakerloo line will have been in operation for 49 years and it would be likely that some of the trains will see over fifty years in service. We don’t know if “first driverless trains” means first production trains or first prototypes. Now undoubtedly the trains could soldier on but it is only to be expected that they will get more and more difficult to maintain. Their life would exceed the 1967 stock by more than six years, and this on a line with sharp punishing curves unlike the well-designed gently-curved Victoria line which was built in an era when it was not necessary to follow the street alignment above. So even before thinks about driverless trains, Mike Brown must be getting concerned about how long he has to keep his existing trains in service.

A further concern for Mike Brown must also be that that the Mayor seems to have already decided on the nature of the replacement – namely “train captains” like the DLR. Readers have pointed out that a recent article in Modern Railways has very effectively made the point that a “train captain” must be the worse of all possible scenarios for the underground. The author of that article argues that a better option from an operating perspective would be to have someone who was there “just in case” that normally did nothing.

The problem of the Piccadilly

If that wasn’t something to keep him awake at night there is also the issue of the Piccadilly line. Current thinking is that the Piccadilly line stock is replaced after the Bakerloo. As the idea of tailor-made stock for each line is out of fashion the Bakerloo and Piccadilly lines need to have common stock, but here’s the problem:- The Piccadilly line shares track and signalling with sub-surface stock at various places so if the Sub-Surface Lines (SSL) are fully upgraded by 2018 then it would highly desirable to have new Piccadilly line trains compatible with the SSL signalling available by 2018 as well.

The chances of that happening are getting vanishingly small of course, so either the Piccadilly line trains are driven manually over shared SSL territory (in which case there are implication for running a high frequency service using automatic trains on the SSL) or the very old 1973 stock will need some modification to make it compatible with the SSL signalling.

45 year old trains, which experience severe vibration in daily life and also extremes of temperature (from winter days in the exposed suburbs to the stifling heat of the deep tubes), tend not to have the sort of electrical wiring one really wants to mess around with unnecessarily.

Why are we in this mess?

The problem largely appears to be down to PPP – the Public-Private Partnership imposed on London Underground and the Mayor by the Government of the day. One thing even the detractors of PPP generally accepted was that the advantage with PPP was that capital investment was forthcoming. So Metronet or Tube Lines would find the capital on the open market, and London Underground would only have to find the annual charge – rather like taking out a mortgage to avoid having to pay for the entire cost of a house upfront. It made sense to replace the dated trains after around 40 years, so that was what the PPP contract required. When the PPP deals collapsed though, London Underground may have gained the benefit of not being tied to some unnecessary day-to-day payments but it unfortunately also lost a reliable source of long-term secured funding.

Resignalling the Piccadilly in a sensible way

One of the many absurdities of the PPP arrangement was that the two firms involved, Metronet and Tube Lines, were free to put whatever signalling system they wanted to so long as performance targets were met – even if it was incompatible with everything around it. So Metronet, which controlled the Sub-Surface Lines, was going to put in a signalling system with little regard to what Tube Lines was later going to install on the Piccadilly. Fairly early on after the collapse this issue was tackled and it was decided that the Sub-Surface Lines and the Piccadilly Line would all be resignalled under the same contract.

Because of performance penalty payments the Infracos (Metronet and Tube lines) had every incentive to maximise capacity and reduce journey time but very little or no incentive to reduce staff costs, other than their own, since it would not be them that would see the benefit. With such decisions now back within London Underground, it can take a more holistic approach to the whole issue.

Enter the Evo Train

We can see that one of the problems with the PPP is that it discouraged risk-taking even when there were great gains to be had. One would be penalised for stepping out of line, but not necessarily rewarded for benefits achieved by doing so. When PPP was abandoned, it thus made sense to step back and consider what trains were really needed for the Bakerloo and Piccadilly, bearing in mind that these trains may be expected to be in service for up to 50 years. Sadly the opportunity to factor the Victoria Line into the equation had already been missed. As a concept, the idea of a tube train that was walk-through and either had a cab which could be later removed, or no cab at all, was thus developed. It was recognised that a lot of opportunities and issues related to this had to be looked at in detail.

The Excuse to Delay

With London Underground wanting to “get it right” and the cost of a single tube train amounting to around £10 million, the temptation for the accountants to want to put off the introduction of yet more fleets must be quite great. Remember that in the past few years Transport for London (or rather the Mayor and the treasury) have had, or will have, to fund:

  • 47 trains of 2009 stock for the Victoria line.
  • 58 trains of S8 stock for the Metropolitan line and the follow-on order of 133 S7 trains then (assuming the Croxley link is approved) a further S8 train.
  • 57 4-car Class 378 electric stock for London Overground at least some of which are likely to be extended to 5-car sets by 2014.
  • 8 class 172/0 two-car unit replacement diesel stock for the Gospel Oak-Barking line.
  • 6 new trams with an additional 4 in the future likely.

One can understand any keenness there is to put off replacing Bakerloo and Piccadilly stock for a few years.

Just Tell me what you want

The problem for Mike Brown seems to be three-fold.

Firstly, he does not want to get the blame if reliability plummets on the Bakerloo and Piccadilly through no fault of his own – something that our second piece on the Transport Committee will show he is acutely aware happened on the Metropolitan Line. Nor does he want the blame for “performance pollution” on the Sub-Surface Lines due to interference from the antiquated Piccadilly Line stock. By emphasising the issue to the Greater London Assembly has has thus covered himself by having a “I did warn you” strategy.

Secondly, he does not know what the real objectives are for the Mayor:

Is “Train Captains on the Bakerloo” really an objective or merely how the mayor envisages the true objective being carried out?

Is the real objective not to rely on skilled drivers? In which case whether or not there is a cab at the end of the train is really a side issue.

Has this whole issue become so political that if the experts decide at the end of the day that conventional drivers in conventional cabs on automatic trains really is still the best solution it will be rejected because of political idealism and a dislike of the unions?

Indeed possibly one of the most telling comments made by Mike during the Committee meeting was that he does not know what is on the Mayor’s mind. How can he possibly plan ahead if he has not been given a clear objective by his political master?

Contrast this with President Kennedy’s speech committing to “putting a man on the moon and safely returning him to earth before the decade is out”. NASA knew exactly what was expected of them. By contrast, Mike Brown and London Underground are probably not really quite sure what they are trying to achieve and don’t really have a timescale they can work to. Any military strategist will tell you that without a clear objective you are asking for trouble.

Those who have been on this planet longer than the current generation will remember the days of Horace Cutler in charge of the GLC and the consequences of decision-making then. London Transport’s management was highly criticised at that time but many would say that the problem was that the GLC interfered with day-to-day running, which it should have left to the transport experts, whilst criticising them for not making the strategic decisions and setting the objectives – the thing that the GLC itself should have focused on doing. Increasingly, there seem to be worrying parallels with the current day.

The third problem Mike Brown has is that he has a team investigating and evaluating what is possible. When we say “what is possible” we mean in terms of getting an accepted safety case and satisfying any regulatory body, as we know all of the technology is possible and has been so for several years.

Unfortunately, as it stands, this team just does not know what target date to aim for. If, for example, they had a target of first production train in service in 2020, they could plan for a pre-production train in service in 2018. That may seem a long lead-in time for those unfamiliar with rolling stock rollouts, but as the Victoria Line rollout and the Jubilee signalling conversion made abundantly clear, when it comes to trains on the Underground it is critical that you test offsite, and then test and test again.

All this would mean finalising design in 2017, based on the technology and safety cases agreed up to that date. Without a firm target delivery date, however, the project team will not know if they should be doing blue-sky thinking prior to implementation in the early 2020’s, or knuckling down to work with the technology available today and building a safety case as quickly as possible ready for putting contracts out to tender in the near future.

Getting the Right Emphasis

Ultimately, we are not trying to suggest that the issue of “driverless” trains is not an important one. Indeed as the Committee meeting – and the newspaper headlines that followed it – showed, it is probably going to be one of the most significant decisions of the next decade for the long-term future of London Underground. If you thought the drama which surrounded the introduction of the New Bus For London was excessive, then you haven’t seen anything yet. Indeed if Channel 4 are hunting for a sequel to “Made in Chelsea,” then “Made in Derby” may well end up providing a similar level of faux-outrage, angst and sweeping dramatics.

We must, however, not lose sight of the fact that decisions relating to the Deep Tube upgrades, and the rolling stock strategy therein, are critically important to the future direction of the Underground – and to its ability to run a reliable service on a daily basis. Thankfully this appears to be something that Mike Brown has recognised, as there is a lot of long term planning to be done, which may even include thinking about a near-end-of-life refurbishment for the Piccadilly line 1973 stock. This would allow investigations into what technology can do to maximise capacity and reliability, and minimise whole-life costs, to continue, with the results of that work to be incorporated into the new stock that followed. Whatever the long term plans, Mike seems fully aware that what he needs to sort out now is the finance and the timescales. For without a clear idea of both, debate over driverless trains is just that – debate.

jump to the end
There are 42 comments on this article
  1. john b says:

    Interesting piece. I think you may be taking Boris at far closer to face value than is really warranted. Political idealism is not a thing that he has – the GLC comparison isn’t really warranted. Note the complete lack of actual fights picked with the unions since he came in. His manifesto pledge was to introduce a New Routemaster, so he couldn’t really dodge that fiasco – but otherwise, he’s been content to let TfL continue on the path set by the previous administration.

    Given LU’s industrial relations history and the timescales involved, it seems vanishingly unlikely that cabless trains will be introduced as part of the next generation of rolling stock. It’s a fight which might in the long term save a small amount of money on salaries (when comparing DLR train captains to LU drivers on the ATO lines), but which provides virtually no operational benefits over ATO, in exchange for a lot of risk.

    I would be very surprised to learn that Brown’s team aren’t working on the assumption of something that looks like a 09/S-stock mash-up (cabs, Tube profile, articulated) entering service around 2020, with very limited resources dedicated to paying lip service to other options.

  2. Slugabed says:

    Thank you for this thoughtful article,which looks behind the obvious headlines and ballyhoo which passes for politics in London.
    But the article points to a greater malaise at the heart of British politics as a whole…the inability to think in a strategic,long-term manner and the consequent deferral of any major decision,hoping that someone else will do the dirty work and take any consequent flak.
    This can be seen on the Underground but also on National Rail,Power generation,Airports,the list goes on.
    This is not to say that I have an particular agenda to plug on all of these topics (though I do on some) but rather to say that I am beginning to think that ANY decision is better than none at all…

  3. Si says:

    Would revisiting the space train concept be a good idea for the Piccadilly line – certainly that is a line that needs the additional room – even resignalling the line to let there be more trains run along it wouldn’t solve the PIXC problems. I guess Crossrail will help with West London somewhat, but North London congestion on the Piccadilly will have to wait at least two decades for Crossrail 2 to provide some relief (if they go with the option that goes to Ally Pally)?

    Of course that would likely delay things on the Bakerloo and Piccadilly line train replacements further.

  4. Benedict says:

    Not entirely sure I agree with the gist of what you’re saying about the A stock or the 1973ts, PoP. The 73 stock is LUL’s most reliable fleet, and has been since 07/08. Provided whatever maintenance regeime is kept up there no reason why they shouldn’t be able to match the A stock’s mighty half century (+1) in everyday service; that would take them to 2026. As long as space can be found in or under the carriages, this longevity would surely imply that retrofitting ATO to the 73ts might prove economicaly plausible.

    A note on the A stock: It was ultimately replaced because of image, not lack of mechanical integrity. It was always going to be the case that the Mayors office and TfL would not be shown in a good light if they were running a fleet of 50+ year old trains during the olympics. The C stock fleet has always performed worse, and would always have made a better choice to replace first, especially considering that it would have enabled an inherent 16.7% increase in capacity, and, if the section to Barking had been phased to be completed earlier, would have served the Olympic venue. Not to mention that the C stock fleet is stretched to cover anyway… If the A stock’s overhauls had been kept up they would have kept going, not for ever, but for a while longer. They had the most robust body construction that the Underground has ever seen, and given the right set of circumstances could have outseen even the IOW 38ts. Such is moot though, as life has moved on now!

  5. P says:

    I’m a train driver for London Underground. Before the Olympics there were ongoing events for all LU staff where Mike Brown spoke to all staff about future plans. He used the term ‘we will never buy a train with a cab again’ the exact term as used by Boris Johnson.

    Brown didn’t seem to have a problem being clear about plans for driverless trains around the time of the London mayor elections – as I say using the exact same quote to rather crudely support one of the mayors election themes.

    LU at the time also seemed happy to support the Mayors claim that fares couldn’t be reduced – in contradiction to Livingstone’s suggestion.

    I for one think it’s very telling that it is only now Boris has been safely elected and in place that the detail (or lack of detail) about the future of LU is becoming clear, and I hope Brown’s support at the time for the mayor’s half baked scheme is something he comes to regret.

  6. john b says:

    Benedict: I’m not sure that’s really the case re the A-stock. It was always going to make sense to start the S-stock programme with trains that didn’t enter the Circle Line, simply in case they didn’t work! The remoteness and the Baker Street terminators meant that the Met was the best place for the introduction of a new class. The extra time taken to lengthen the H&Circle platforms for S7s was also a factor.

    P: Brown’s use of “we will never buy a train with a cab again” is very interesting, and I have to say I’m surprised. Thanks!

  7. Sam M says:

    “Has this whole issue become so political that if the experts decide at the end of the day that conventional drivers in conventional cabs on automatic trains really is still the best solution it will be rejected because of political idealism and a dislike of the unions?”


    But it is politically necessary for Boris to be seen to be trying to pick a fight with train drivers – everyone knows he wants to become Tory leader, and for that he needs the support of the Tory grassroots. That group of people hate almost no-one (except perhaps Abu Hamza) as much and as all-consumingly as Bob Crow and by extension all train drivers. They believe almost all of the excessive cost of the railways is explained by the cost of train drivers’ wages. They believe using unions to demand higher wages is a form of extortion/theft, and want to see train drivers punished with unemployment.

    It could be that the reason Mike Brown doesn’t want to talk about driverless trains is less that he doesn’t want to admit that they are happening in case he upsets the unions, and more (given P’s account of his previous enthusiasm for the idea above – to drivers’ faces, no less) that he doesn’t want to admit they’re not happening in case he upsets Boris.

  8. Pedantic of Purley says:

    @john b

    I was not trying to equate Boris politically with Horace Cutler but it can hardly be denied both are/were showmen. I was trying to make Slugabed’s point that you need strategic decision making at the top so that the objective can be achieved. This part of it is not party politics. You also need a clear description and dissemination of the objective so that it can be carried out. My concern is that that no-one knows what Boris’s real objective is. Driverless trains for its own sake is not a meaningful objective. Objectives are things like cut staff costs, reduce exposure to industrial action, improve reliability, increase capacity. Driverless trains may be a way of achieving one or all of those objectives.

    I accept that, unlike Horace Cutler, Boris is more a pragmatist than a political idealist but on the basis of the previous paragraph I think my comparison with Horace Cutler was entirely justified.


    I am sure that the A trains were reliable and could have continued for another twenty years or so if required. They may have been reliable but they were no longer fit for purpose. To give just one example the acceleration was hopeless which was due to its dc motors which in the modern world of traction were prehistoric. Sure you could have replaced the motors and control mechanism but would it have made any sense to do so ? One can of course argue that it is a bit daft to get new faster trains then downgrade the fast trains in the timetable but that is another issue.

    There is also the fact that when the trains were ordered the intention was that under PPP resignalling would take place shortly afterwards so that the new trains could be properly exploited. This would not have been possible with the ‘A’ stock. For the reasons given in the article that order was cancelled and as far as I am aware has not even been re-tendered let alone re-let.

    I am pleased to hear that the 1973 is now very reliable. If a rational decision is made to extend its use to 2026 I do not have a problem with that. That means that the necessary mid-life or towards-the-end-of-life refurbishments can be planned for and factored in if it can be shown that they would be cost-effective and reliability could be maintained. If I were in Mike Brown’s shoes I would not like to be in the situation where I did not know if they were going to be replaced in 2018 or 2026 because I could not take the appropriate action. One starts to agree with Slugabed that any decision is better than no decision at all.

  9. paul says:

    I’m sure that the right maintenance regime can extend the life of any set of trains for as long as anyone wants to. Isn’t the point more that this costs an ever increasing amount of money, and that new trains pay for themselves by reducing the cost of the required maintenance regime? Ditto with the signalling.

    Capacity and performance gains are helpful, but surely the reduction in maintenance cost is the biggest contributor to the cost-benefit of the upgrades?

  10. N B J C says:

    ……….Or, it could just be a case that EU railway safety legislation was pointed out to TfL, with regards to running a fully automated system, and that there is NO budget for ‘driverless trains’. In effect, they would be purchasing trains that couldn’t be run over the current infrastructure.

  11. Paul (a different one) says:

    P of P @ 1.40 pm

    This link suggests the SSR resignalling contract definitely has been relet:

  12. Pedantic of Purley says:

    But that press release from June 2011 doesn’t mention the Piccadilly line so I suspect that that was referring to the original contract that was cancelled.

  13. Greg Tingey says:

    “…it should have been done years ago since the trains could not provide the performance really needed for the service they provided it should have been done years ago since the trains could not provide the performance really needed for the service they provided”
    Correct, except for one very important point.
    The SEATING in the new trains is vile … why can’t we have proper cushions?
    Anyone who says “fire safety requirements” will be branded a liar, immediately, since coaches & buses have the same PSV requirements, and I have sat in a 4-year-old small (Under 20 seats) coach in the past month, and the seats were padded & comfortable.
    Ditto the “increase in capacity”, especially on the outer-Met “S” stock, where the increase on an outer suburban service has been achieved by removing even the uncomfortable seats – resulting in gross overcrowding on the Marylebone trains, because people DO NOT WANT to stand from Rickmansworth or further out, and take longer, because the idiots have made almost all trains all stations …Remind us, please, why was Harrow-moor Park quadrupled?
    Discuss these insanities as well, perhaps?

    “ ….assuming the Croxley link is approved)” … err I though it had just been done?
    Sorry, just re-checked, the Public Enquiry is underway, but it’s a forgone conclusion, actually.

    “Secondly, he does not know what the real objectives are for the Mayor: “
    To become Prime Minister, of course, & shaft Camoron.
    Dearie me, I thought that was obvious!

    Driverless trains ..
    Can I repeat what I said last time around, just so we all get it straight (I think?).
    In Paris can usually go to driverless, because the tunnels are usually double-track. So, as said elsewhere, you can evacuate by a “parallel” train – you can’t do that in the old deep London tubes.
    As discussed to death previously, the technology of a completely driverless train is already well-established & works.
    PROVIDED that …
    The train is in a situation that fulfills at least one of these conditions …..
    1] Above ground
    2] In a double-track tunnel
    3] In a single-track tunnel with evacuation walkway
    It also helps if you have P/f-edge doors in your UndergrounD stations – again, preferably, all of them in the U/G sections – not absolutely necessary, but really nice to have.
    This makes the introduction of truly driverless in London, a problematic issue, to say the least.
    As for having driverless in some sections, with the drivers “picking up” @ intermediate point, yes it could work in theory, but your rostering problems would be considerably larger than simply picking up a train at a depot or fixed near-depot changeover point (like 7 Sisters) – which might promptly screw your supposed reliability gain.
    Brown appears to have changed his tune, does he not?
    Perhaps (just perhaps, he IS LUL “management” after all) he has learnt the realities that we thrashed out before, which I’ve listed above? If that is the case, he should be given credit for learning.
    However, Sam M’s points are also possibly true.
    What very few people seem to realise (outside this sort of specialist discussion) is just how bad Industrial Relations seem to be on the Tube … Crow is very nasty, politically, but why is he supported? Because LUL management is excrement, frankly.
    Oh dear.

    “Maintenance regime” / “Train Life” ..
    Oh dear, what a collection of half-hearted amateurs!
    Just HOW LONG did the original “Clockwork Orange” stock work for & last?
    Go on, ‘ave a good larf!

  14. Pedantic of Purley says:


    Showing my age I remember being told, correctly at the time, that you could not have One Person Operation on true “tube” lines because it was only approved for sub-surface lines.

    I myself have made the mistake of explaining to people that we would never see DB ICE trains running through the Channel Tunnel because the safety rules prohibited this but this is now likely to happen in the next few years.

    I could go on.

    Your list of requirements may well be the current set of requirements for unattended trains – not the same as “driverless”. I had thought that these requirements were the conditions necessary for any new construction even if the train had both a driver and a guard.

    The first example I have given shows that these restrictions can change over time. After all I have never seen a motor car being preceded by a person carrying a red flag even though this was once a requirement.

    The second example shows that restrictions can be challenged and overturned if a satisfactory safety case can be made.

    There is always the possibility of derogation for existing railway systems. As we have been trying to emphasise running driverless trains is not a technical challenge. Probably the biggest challenge is to get an approved safety case which will almost certainly include challenging existing rules (both UK and EU) and demonstrating that a new set of rules can be implemented that will not create a safety risk.

    “Secondly, he does not know what the real objectives are for the Mayor: “
    To become Prime Minister, of course,

    Made me laugh.

  15. Ned says:

    Isn’t there also a requirement for all trains (including those on the Underground) to meet accessibility requirements by 2020?

  16. Ig says:

    Another good article thanks, but “he decade” is a typo….

    Now fixed. Thanks. PoP

  17. Anonymous says:

    The phasing of the line upgrades was established by LU a long time ago and formed the basis of the PPP contract requirements. The phasing was designed to try to provide a steady flow of rolling stock, signalling and track works over the first 23 years of the PPP arrangements. The work had to be phased in terms of the ability of the supply chain to deliver the work, LU to afford it and to avoid the Tube being subject to too many closures. Many of those still apply even if PPP has largely disappeared.

    It is far too simplistic to just blame everything on PPP. I appreciate it is the thinking person’s “bete noire” but it did deliver 09 stock and the S Stock contract. We also have the Jubilee Line upgraded although I accept that upgrade was torturous in its implementation and passengers were subject to far too much disruption. Blaming PPP for things like the lack of open gangway Victoria Line is just wrong as it implies only one party was at fault when that was not the case. The unwritten assumption in the article that other contractual arrangements are a “nirvana” underestimates the sheer scale and difficulty of something like Sub Surface resignalling. I will be surprised if that project glides serenely into service without any glitches.

    The SSR resignalling contract covers only small parts of the Picc Line – those parts where there is interworking with SSR trains. 73 stock will be modified and there are performance / reliability risks during and after the modification works. There is no let contract for the Picc Line upgrade – trains or signalling / control. As already mentioned the 73 stock is very reliable thanks to sterling work by Tube Lines at the Picc depots. Practices are light years ahead of those elsewhere on LU so there are significant efficiency benefits to be gained by sharing practice. The 73 stock may well need some targeted investment to keep it running but it can run on for several more years. There are risks to reliability such as an emerging age related issue across the fleet (e.g. bodywork failure) or obsolescence arising faster than expected (components become impossible to source). The much bigger problem is the condition of the signalling system and related power cables where mini upgrades are less feasible. I believe there was a potential option on the Thales contract for Picc Line signalling / ATO but I suspect that has lapsed completely now. It was a LU decision to shove the upgrade back 8 years because there was simply no money available and there was a real concern over the risk of a repeat of Jubilee Line style upgrade disruption. Northern Line was too far gone contractually for that to be stopped but the implementation concept is obviously very different.

    LU’s direction from the Mayor is effectively set by the budget and efficiency demands being made. LU will almost certainly determine the preferred method of operation and I suspect they will want to get to driverless train capability but will retain on board staff because union, stakeholder and public reaction will be whipped into a frenzy to prevent full driverless capability. The Mayor doesn’t really give a stuff provided he can make the right noises. His first term was characterised by killing off Tube Lines due to the Jubilee Line upgrade problems and then panicking about tube performance in the 12-15 months before the 2012 election. It was only by bringing in Isabel Dedring that he managed to exert some pressure and get a mini “action plan” in place to improve reliability (over and above what LU was already doing). I expect a similar level of disinterest now he has been re-elected. Boris will be happy provided he can banging the “driverless” drum every time Bob Crow gets upset or Dick Tracey feels he needs to whine on endlessly at Mayor’s Question Time / GLA Transport Commitee.

    LU will keep going with the Deep Tube programme to the point of delivering the prototype train as required by the DfT. Quite what they do with it when it is here is an interesting question. The W&C line could be equipped with platform edge doors / gates and is simple to convert to driverless. Any of the Central, Bakerloo or Picc present problems with driverless trains because of the platform / train interface issues (curved platforms, compromise height platforms etc). However I am sure people are beavering away to develop solutions or mitigations for these issues.

  18. Greg Tingey says:

    Pedant’s point: I said … “truly driverless” I didn’t say “unattended”
    Given that Vic & Central trains are ATO, effectively, the “driver” is actually a train-controller/attendant, rather than a traditional driver, isn’t she/he?
    I am well aware of changing regulatory control, but the point here is the “new” requirement for evacuation passageways makes a “driverless”, however you want to interpret that term, train less likely, rather than more; certainly in the near, say next 20 years-future.
    The prohibition agaionst DB-ICE in the CTnl wasn’t a real safety rule, it was a French political try-on, as evidenced by their screaming when the tests & announcements were performed.

    “… Practices are light years ahead of those elsewhere on LU ”
    Then why are the rest lagging behind? It’s supposed to be ONE organisation, isn’t it? Has been since 1933? Then why hasn’t Pick-a-dildo practice been extended to other parts & depots on the syatem, or is there something else going on?
    “I think we should be told” ahem.
    “Driverless” on Picc/Northern/Bakerloo ? Not in the foreseeable, for the very reasons you gave – simply the game isn’t worth the candle in terms of money, quite frankly wasted, which could be better spent on ATO-with-controller-in-cab & other intensity/signalling/stock upgrades.
    And building new tube or other sub-surface (x-rail) lines

  19. Guano says:

    There was an article in Modern Railways, a few months back, which concluded that there was little point of having train captains on London Underground. It would still be necessary to roster an attendant for every train, and it is the complexities of doing this that is the issue. A long time in the future it may be possible to have trains without attendants, which would allow the number of trains in service at a given time to be matched to the demand without undue cost; but we are a long way from there.


  20. Anonymous says:

    @ Greg – there was a lot of work underway, and I would expect is still continuing, to bring all the depots to be best practice. This is being done to bring consistency to asset management practices, improve efficiency and reduce costs. Clearly not every depot will do things exactly the same way because of different stock ages, designs, signalling kit on trains and contractual arrangements (e.g Alstom maintaining the N Line trains). You don’t maintain 72 and 09 stocks in the same way! Tube Lines and Metronet did do things differently because of the different consortium structures but before everyone blames PPP I was told some depots were still doing things in the same manner as the 70s and 80s. PPP did force the adoption of better ways of doing things and brought in investment in depot, track and station maintenance equipment & practices. You can, of course, question why practices were left unchanged but that is a wider question than saying it is an Infraco issue – they only came into being in 2002. I doubt there was leading edge best practice back in LT days no matter how much people like to portray it as a “golden era”.

  21. john b says:

    PoP: the reason I suggested things were a bit different is that from my dealings with TfL people, unlike Cutler days, the organisation (Windsor House, not necessarily the various regiments below) has a pretty strong idea of its vision for the future. So unless Bozza intervenes to actively screw things up (as on the Vanitymaster), it has a technocratic-but-decent-enough 30-year-plan.

    Greg: there is absolutely no rule in place that would mean “moving Tube trains to a DLR, manned-with-a-train-captain model would force the Tube to meet the safety requirements that are currently mandated for new-build tunnels”. That just isn’t the case. You might argue that the unions would be able to force such a shift as part of the acceptance negotiations (ie that they will only agree to staff trains running on such tracks), but that’s a very different question.

    Ned: yes-ish, but the SoS for Transport can waive them on request, so it’s not a biggie.

    Anon: ta for all that. Agreed, there wasn’t best practice on either side before the PPP; and by all accounts TL were far more competent during the PPP. Hopefully under the new TfL regime it will be easier to share best-practice where it’s relevant.

  22. Taz says:

    My question wasn’t answered before. I wonder why it might be acceptable to have an unstaffed train in new Jubilee Line tunnels east of Westminster with a walkway, but unacceptable in the smaller, older tunnels north of there. Surely passengers would not be allowed out from a stalled train between stations until sufficient assisting staff had arrived in either situation, even with a driver present today. In which case, why do the smaller tunnels prevent unstaffed train operation at some time in the future?

  23. Malcolm says:

    Just a guess, but maybe if there’s smoke or something in a stationary unstaffed train, and the control room staff see fit, they can open the doors and ask/tell the passengers to get out onto the walkway, if there is one. Whereas in a small tunnel it would not make sense to get the passengers to deploy an end-of-train ramp thingy, or even if it could be deployed remotely, there would be too much pushing and shoving to have much hope of getting them all out unhurt. But the whole thing is full of imponderable ifs and buts. If we had some eggs, we could have ham and eggs, provided we had some ham.

  24. Taz,

    I think the question wasn’t originally answered because no-one knows the answer. I think too much is made of the walkway issue because I believe that there are strict instructions that state that they are not to be used by the (non-)travelling public who are being detrained. They must always be escorted along the track.

    If I understand it correctly, Malcolm is on the right track (excuse the rather dubious pun). I suspect that the real reason for getting rid of the cab is that in an extreme situation (such as smoke in a stalled train) another train can be automatically and very quickly brought to the rear or even the front – or both simultaneously ! – of the the stalled train, the end door locks released, and passengers can very quickly be transferred to the “rescue” train and removed from the source of danger. Unfortunately the idea of removing the cabs becomes highly symbolic for both Bob Crow and the more right-wing members of Conservative party who have latched onto it. It also appeals to Boris’s desire to have a dramatic vision he can put across to his audience. Horace Cutler would have approved.

    I suspect that the critical point of the above is that the procedure to get the rescue train(s) in place can only be done quickly and safely if does not involve communicating with drivers and establishing that everyone knows exactly what they are doing and it is done automatically or centrally from the control room. It also needs a modern very flexible signalling system which ideally is bi-directional.

    The reason we keep going on about safety cases is that such a scenario is probably technically possible now and, if not, certainly will be in a few years time. That is not the big problem. The problem is to do the analysis and rigorously show that in every conceivable scenario the passenger would actually be safer and the risk to life reduced if the driver was not in charge of the train. Such risk analysis would need to take into account potential failure of critical equipment at a critical time. This may seem all far fetched but you only have to read how First Capital Connect managed to make a complete and utter shambles of a simple evacuation involving staff liason to realise the potential benefits. Also nowadays the powers that be get extremely concerned about trains full of people being stranded in hot tunnels. Any procedure that gets them out quicker and in extreme emergency can be implemented without putting any other people at risk is bound to be looked upon favourably.

    I must emphasise though before we get side-tracked too much on driverless trains that the whole point of the article is that this is all irrelevant without an order for new trains and a timescale to aim for.

  25. Greg Tingey says:

    john b
    Can you deconstruct your middle paragraph, please?
    It doesn’t seem to make sense, & I didn’t say whay you seem to think I did, or vice versa … errr …..

    ” I believe that there are strict instructions that state that they are not to be used by the (non-)travelling public who are being detrained. They must always be escorted along the track.”
    But this is a classic example of a regulation which CAN be altered, isn’t it?
    Particularly as the walkways do NOT have sleepers, rails, points & other trip hazards……

    Stalled train(s)
    But people can move through the cabs, anyway – they have done in past emergencies, most notably on 7/7/2005!
    So why do we “need” to do without cabs?

    I had not noticed the RAIB report on the Kentish Town fiasco – thanks.
    How NOT to do it in one easy lesson – almost as big a cock-up as the DLR derailment at the “triangle” in fact!
    [ ]
    Unfortunately, opening doors has to be an option – you CANNOT lock people in – if only to get more air – I would, personally, in such a situation try to open doors, but NOT get out, since the risk of crushing would be too great.

  26. Greg,

    Walkways. I tend to agree with you but I think the fear is that someone might lose their balance or be jostled by someone panicing and fall onto the track from a greater height than if they were already walking the track. I believe that the original plan was for people to use the walkway and it got changed in the light of experience so I think it would take a lot to revert it to how it originally was. But as you say that can be changed if circumstances justify it.

    Cabs. In my example I was think primarily about a totally-unstaffed train scenario or where the member of staff could not get to the rear (or front) cab to assist evacuation. Sure you could have people walking through cabs but that would restrict flow. More particularly it would restrict the size of the opening that could be available and complicate matters. People tend to get confused easily when panicing. So we don’t “need” to do without cabs but if they serve no real purpose then there is a benefit in getting rid of them.

    I suspect that ultimately there would have to some thought given to an ultimate override that allowed the doors to be opened if the train was stationary and if all remote contact was lost. Even if the train is staffed the member of staff could suddenly die of natural causes or either die or become incapacitated by an extreme event that caused the train to come to a halt.

    Air. You also mention the possible need for passengers to be able to open doors (or alternatively vents) to get outside air – presumably above and beyond what, if anything, would normally be available.

    It is just these sort of issues that would have to be thought through and there are a lot of scenarios to consider.

  27. Glenn Wallis, Assistant Secretary, Barking - Gospel Oak Line User Group says:

    A minor point. TfL did not directly fund the 8xClass 172 Turbostars for the Barking – Gospel Oak service. Although there was no requirement to include new trains in the London Rail Concession bids, MTR/Laing did include them in theirs and won. The units are leased to LOROL by Angel Trains. Otherwise, it is likely that Barking – Gospel Oak services would now be in the hands of refurbished “Centro style” 3-car Class 150 Sprinters.

  28. Mark Townend says:

    Narrow platform height walkways do seem to be less than ideal for emergency evacuation without any barrier to prevent falling. They would be unable to easily accommodate an orderly exit of a crowded train full of passengers if all the doors were opened simultaneously and a fall from the walkway would likely come into contact with the traction power rails, so these would have to be isolated in these circumstances; that, in turn would prevent passage of rescue trains.

    Regardless of whether there is a cab at the train ends a further problem with track level evacuation is the traction power again. Relocation of the power rails to the tunnel roof in the deep tubes might allow a level walkway between the running rails to be used for evacuation, without switching off the traction current; this would remove the main tripping hazard too.

  29. Anonymous says:

    @ PoP – I don’t think there is a requirement placed on LU for a revised operating methodology to be safer than the one it replaces. LU needs to demonstrate that the risk is as low as reasonably practicable (ALARP). That might get you to a position where the new method is safer but the requirement is really to be *as* safe. Oh and they are now Safety Certificates not Safety Cases. I recognise that stakeholders and the public may well make demands for everything to be safer but that is a dangerous path to tread because you remove the practicablility aspect which is what stops ludicrous amounts of money being expended for next to no benefit other than salving someone’s conscience.

    If cabs are not specified then it saves a load of equipment and cost in both capital / opex terms. Some controls obviously need to be provided for manual intervention and possibly shunting but that is just a panel at the front as already happens on different stocks / systems.

  30. Twopenny Tube says:

    Among the comments attached to an article on Dave Hill’s Guardian Blog, ‘kippers’ raises a pertinent point: “It’s odd how the people who want driverless trains also want to bring back a design of bus that requires a second crew-member who does almost nothing.”

  31. Anonymous says:

    In all the above discussions on how the Picc line signalling could be integrated with the SSL works, I’m surprised no-one has mentioned that Thales has a contract to install SelTrac on the Piccadilly line once they have finished the Northern.

    This contract was let to Thales in 2007 by Tube Lines and is still in place.

    Details here:

    Dealing with this contract is likely to be one of the reasons why the Picc signalling strategy hasn’t been resolved yet.

  32. Anonymous says:

    @ Anonymous 1211 – I would treat a press release dated from 2007 with a dose of caution. The Picc Line upgrade (PLU) was postponed by at least 5 and more likely 8 years because the costs proposed in the Tube Lines Periodic Review were unpalatable as well as other issues having got in the way such as those which are the subject of the above article. The resources that were working on the Picc Line upgrade were stepped down as far as I am aware and I would expect Tube Lines to have cancelled or heavily renegotiated the Thales contract for the Picc Line.

    Looking at the press release there is not a cat in hell’s chance of the Northern Line being delivered for 2011 (!) which means the NL contract will have been renegotiated. In similar vein the Picc Line resignalling certainly is not going to be in place for 2014 and as LU put back the PLU delivery date Tube Lines are required to respond to that change and to make sure TfL is not commercially exposed. Tube Lines will have had termination rights and Thales would not want resources being wasted when there has been a material change in circumstances. They still have work to do, over a longer than anticipated timescale, to get the NLU done. There will have had to be a negotiation about changing the NL contract and TLL would possibly have had some leverage over Thales given JLU and NLU were both late due to problems with the signalling upgrade.

    I have no confirmation as to what has actually happened about the contract but I expect TfL and Tube Lines will be starting from a different place whenever work on the PLU resumes. I would not expect that to include Thales as a confirmed signalling contractor although their experience on the Northern and Jubilee will stand them in good stead – especially if the system proves to be reliable. Not many signalling failures these days are there?

  33. Greg Tingey says:

    Anom @ 15.25
    Do “Tube Lines” still exist?
    I thought they had gone the way of Metronet a & vanished up their own orifice…..

  34. Taz says:

    Tube Lines still exists only as an TfL subsidiary, fully controlled by them. Thales Picc resignalling contract was dropped as unfunded. Sub-Surface Lines resignalling by Bombardier will now include Picc between Barons Court and Uxbridge 2016-18, also to Northfields. Whilst the S stock will operate automatically, current Picc trains will be fitted with a cab signalling display to enable the driver to operate the train as currently. Lineside signals will be removed. This does not leave much of the Picc to still be resignalled.

  35. Anonymous says:

    @ Greg – yes Tube Lines still exist as a wholly owned TfL subsidiary as Taz says. Bechtel were removed when TfL took over but Amey remain under a management contract to manage the maintenance aspects of TLL’s operations. It was telling to witness the differences between a private and a TfL owned Tube Lines. One of the reasons for retaining TLL is to act as a spur / comparitor to the “in house” former Metronet maintenance bits of LUL. As Jon Lamonte, the current MD of TLL, has been appointed as the new head of Transport for Greater Manchester I wonder if this will lead to a change in leadership / structure.

  36. JamesC says:

    It’s hard to see why people are saying that TfL did not want to look as if it was running a fleet of 50 year old trains during the Olympics. As I write this I’m sitting on a bakerloo train and to be homesteading apart from a bit of flaking paint and some dirt that could easily be delt with by a power washer, unless you look in the drivers cab it’s not that obvious to the general public that they are that old – and I doubt many people look in the cab that often.

    Clearly they will need replacing at some point, but there is no real reason why the trains could be be retro fitted with a new control system as clearly there are people around who know who all the wiring works as they are clearly being maintained still and ripping out the manual handle controls and replacing with a computer to do the same job is not completely unreasonable – it’s been done before…..

    As for signalling clearly the pic and sub surface lines should clearly have the same signaling system. Whoever came up with the idea of different systems needs shooting… What is the point of going from what was to all intents and purposes a unified system. (The electronicly controler pneumatic powered trip cock system) to a bunch of separate ato systems that don’t talk to each other, need completely separate training to use them, or maintain them.

    As for safety the ato is exactly the same as being driverless anyway as the driver in the atomic system mearly does the same as a train captain anyway by looking out the door/window to make sure it’s clear to close the doors and depart the train. The only real difference is where that person sits/stands. Some may argue that by being up the front they can loom along the platform for somebody jumping/being pushed/falling off. My argument to this would be that most people who end up on the tracks get hit anyway despite the best efforts of the driver, andwwithout wanting to put the drivers down in anyway an automated system with scanning radar/heat sensors is always going to react quicker than a human being.

  37. Greg Tingey says:

    James C
    I’m not so sure.
    They are trialling automatic “person-detection” equioment on some LC’s out in the Ely area, and they are having lots of fun & games.
    How do you distinguish between a large dog & a small child? How does rain/snow/fog/ice affect the detection equipment & how basically reliable is it, anyway?
    What do you do about the local “yoof” who try to spoof the equipment with objects on string, like the tin-can swung round fast in front of a road speed-camera?
    Have auto-detect, certainly (provided you can get it to work) IN ADDITION to CCTV, which you can then consult if you get a strange/false/alarm signal.
    Having a “driver” at the front, as opposed to anywhere else is still the best place. As a recent incident at (I think) Finchley Road demonstrated.

    Agree whole heartedly about non-compatible ATO systems.
    This has got to be stopped, RIGHT NOW.
    I would suggest electrocution via the live rail, myself!

  38. Twopenny Tube says:

    JamesC 3.11.12: “…the driver in the atomic system mearly does the same as a train captain anyway by looking out the door/window to make sure it’s clear to close the doors and depart the train.”
    If the driver departs the train, that is a sure way of rendering it ‘driverless’. In the cause of preserving intransitive verbs, such as ‘depart’, I assume this is meant to convey that the driver/train captain checks various things, before setting the controls for the journey to continue, with him/her in the train.

  39. Anonymous says:

    “But the article points to a greater malaise at the heart of British politics as a whole…the inability to think in a strategic,long-term manner and the consequent deferral of any major decision,hoping that someone else will do the dirty work and take any consequent flak.”

    Hah! From here in the US, it looks like your British politicians are masters of long-term thinking!

    We have intercity rail plans which are under development for 20 years, are fully funded, and then are cancelled with the election of a single state governor. Our Congress is infamous for not being able to think more than four years ahead. Our government budgets are routinely finished one or two months before the *end* of the fiscal year to which they apply.

    When it comes to metro lines, it’s a huge fight even to get *passive provision for a future station* in one of the densest and least-served-by-rail parts of New York City (Hell’s Kitchen).

    Honestly, when compared to our politicians, who think that flood protection is something for the next guy to worry about — even AFTER a giant flood — compared to our dysfunctional US politicians, the worst of the British politicians look like strategic, long-term thinkers who make quick decisions on difficult matters and are willing to get their hands dirty.

  40. Anonymous says:

    cabless trains are not an option on any line which operates through stations which have curved platforms.

    C.C.T.V. is essential to monitor platforms when the train is stationary as not all the platform may be visible from any point on the train. if this was to be done from off the train it would have to be done with one person per train, or platform, and it is well known that people do not concentrate well on repetitive tasks.

  41. Moleman says:

    last link, siemens sales pitch based on what they think the specification will be.

Leave a Comment

In order to make LR a pleasant place for discussion, please try to keep comments polite and, importantly, on topic! Comments that we feel do not meet these criteria, or that contain language that could cause some people trouble at work, may be moderated or deleted.

acceptable tags

* (This won't be shown, but you can link it to an avatar if you like)

Recent Articles

Friday Reading List – 17 March


As anyone looking to properly understand London’s transport needs and network knows, context, background and best-practice are important. As readers might imagine, behind the scenes here at LR Towers we thus spend a lot of time sharing links and reading

Read more ›

LR Magazine Issue Five: Overgrounded


With print copies now being prepped for dispatch to subscribers at LR Towers, London Reconnections Magazine Issue 5: Overgrounded is now available to purchase in our online store. Transport is politics, politics is transport You don’t get transport without politics.

Read more ›