Upgrading the Piccadilly: Calling Time on Mind the Gap?


In recent months on London Reconnections we have looked at the future of most of the tube lines, with articles on the Metropolitan Line, the other Sub-Surface Railway (SSR) Lines and the deep tube lines. Some lines, however, deserve a look in more detail than others – and one of those is the Piccadilly.

We omitted the Piccadilly Line from the general look at deep level tube lines because it was worthy of at least one article on its own. In fact it will actually get the best part of two, because it is important to look at the various important developments without, as much as possible, the distraction of the debate on driverless trains. Here we look at the facts, and where possible leave the politics and educated guesses of whether the trains will actually have any staff on them to a follow-up article.

Early in 2013 we looked in great detail at the Northern Line. We warned that although we were pretty confident we could paint an accurate general picture of what is expected to happen, we could not be correct in every detail. This is not least because even if we are correct as to current thoughts and policy, it does not necessarily mean that these will be present in the final product. The same caveat applies to this article.

New Tube for London the Piccadilly Line

The Piccadilly Line Upgrade is not just another upgrade. Mindful of the fact that tube trains tend to stay in service for at least 35 years – and this length of time is getting longer – the decision was taken on the Piccadilly to really plan for the future. This covers every aspect of an underground line not just the trains. Effectively, the Piccadilly Line Upgrade is the intended output of a massive forward-thinking research & development project currently called “New Tube for London” (NTfL). As we saw in a previous article, this is supposed to be about a new tube specification for the Piccadilly, Bakerloo, Waterloo & City and Central Lines. Funding, however, is currently only available for the Piccadilly Line, so despite the grand title it is perhaps best to think of it as the Piccadilly Line Upgrade, with the other deep lines to perhaps follow at some point in the future.

It’s NOT just about driverless trains

Because the Piccadilly Line Upgrade is a forward looking project it is really driven by one factor – and although both the media and various political figures persist in thinking otherwise, that factor is emphatically not driverless trains. The feature which underpins just about everything about this project is energy. This is either in the form of minimising the electrical energy required to run the trains or getting rid of excess heat energy. Obviously these two facets are not unrelated.

That said, the driverless trains issue is important to the designers and the decision has been taken at an early stage to design the upgrade so that the trains are capable of being run in service without any member of staff on board. This is generally referred to as Unattended Train Operation (UTO). The fact, however, that the trains are capable of being operated in UTO mode does not mean that they will.

The Platform Train Interface

On modern transport networks once a system is designed to be UTO-capable then a mandatory requirement almost always now follows – the network or line in question should have platform-edge doors at all stations, including the above ground ones. Furthermore platform levels must be aligned with the floor level of the trains. This is certainly the situation in Paris with Ligne 1 and one suspects that some of our more astute readers are rapidly grasping the significant consequences that this means for the Piccadilly Line.

As if that were not challenging enough, a further requirement is that there should be no significant gap between the train and the platform. This is obviously going to be quite a challenge on a tube system famous for its “mind the gap” announcements. One where in a great number of cases simply straightening existing platforms is just not feasible.

Before we look at the trains themselves and the technology needed to support them, therefore, we need to look at some of the fundamental questions that will need to be addressed about the destinations of the future Piccadilly Line. For, as previously noted, those familiar with the line will have quickly seen that if we have platform edge doors there are going to be considerable problems at Ealing Common station (where the platform is shared with the District Line S7 stock) and the Rayners Lane – Uxbridge stretch of line, which is shared with Metropolitan Line S8 stock.

The Rayners Lane – Uxbridge Problem

Uxbridge Rayners Lane Prior

Originally, the problem at the extreme end of the shared line between Rayners Lane and Uxbridge was going to be resolved by the simple expedient of terminating all Piccadilly Line trains at Rayners Lane. When the disadvantage to passengers was quantified, however, it was realised that this would produce considerable disbenefits in terms of longer journey times. The extra time involved was not that great per passenger, but there were an awful lot of them.

Many people originally believed that terminating at Rayners Lane would be inevitable due to disability legislation and the need for step-free access onto the train. In fact derogation was successfully sought when the S stock on the Metropolitan Line was introduced, so there is no reason to believe that it could not be obtained for the Piccadilly Line stock. The question then was thus whether to pursue trains being UTO-capable, despite the disadvantage to passengers, or to abandon it.

There was, however, another problem. Great care was being taken with the new Piccadilly Line trains to provide appropriate doors at the ideal spacing – but as a consequence there was no way that these would be in the same position as S stock doors. With the doors on the two types of rolling stock destined to be out of alignment, platform edge doors effectively became a non-starter and consequently running trains on this joint section would seem to be totally incompatible with UTO.

The current thinking thus appears to be that the Piccadilly Line should be UTO-capable except for Rayners Lane to Uxbridge. If the Piccadilly Line is run as a UTO line, then for these stations it is expected that a DLR style train captain will need to be on board to supervise boarding and alighting. If UTO were to happen on the rest of the line it is not clear whether the train captain will board at Rayners Lane (westbound) or South Harrow (the previous stop), or whether this would simply mean that one was present for the entire journey (although this would obviously require a greater number of staff to achieve).

Uxbridge Rayners Lane Post

The Ealing Common Problem

Further south there is the issue of shared track with the District Line through Ealing Common station. This has been an operating nuisance for many years. In fact in 1949 there were proposals to lay additional tracks and rebuild Ealing Common station in order to provide segregated lines for the District and Piccadilly Lines between Acton Town and Hanger Lane Junction – the junction just north of Ealing Common station where District Line trains turn off for Ealing Broadway. This and a number of other schemes were rejected by the government of the day – most likely because there was not the money available, rather than because it was not worth doing in 1949.

If it were just a case of sharing the track then sharing with the District Line would not, on its own, have been that great an issue. Nevertheless, as lines get more trains per hour the opportunities for sharing track without delaying or disrupting services diminish. At the same time the consequences of delays become more significant and disruptive.

One solution would be to finally implement the 1949 proposal to separate the two lines, but it is probably much more difficult to do that today with increased urban development. Alternatively one could require each platform at Ealing Common to be permanently staffed with a train dispatcher who has to give authority for the doors to close and further authority for the train to depart. Neither of these two options is, however, currently the plan. Instead a radical solution is proposed – one that a first sight appears to verge on madness. Bear with us, however, for with further thought it is a solution that can be seen to make a lot of sense.

The proposal is that the Ealing Broadway branch of the District Line be transferred to the Piccadilly Line.

There are various reasons why this may seem, initially, to be a bad idea. Most notably:

  • The District Line will have only just been resignalled and now it looks like the branch to Ealing Broadway will need resignalling again.
  • Some journeys such as Richmond to Ealing Broadway would involve going all the way to Hammersmith to change.
  • Chiswick Park station would need to be added to the Piccadilly Line, which would mean platform height alterations and new points so the Piccadilly Line would call there. This would not be a popular move with most station users who would probably see their daily journey involving a change back onto the District Line at Hammersmith or Barons Court.
  • Ealing Common depot is the depot for the District Line on the west side of London. If the Piccadilly Line goes to Ealing Broadway instead, then the District Line depot will not be on the line it serves.
  • The Piccadilly Line serves Heathrow. One would not want additional branches on the Piccadilly Line if it meant lowering the proportion of trains that serve this busy transport hub.
  • The Piccadilly Line effectively already has three ultimate destinations in the west (Heathow Terminal 5, Heathrow terminal 4 and Uxbridge). A fourth would make operating the line even more difficult.

The counter arguments, however, put this proposal in a different light:

  • Given that this is a long term plan there is no need to re-signal the Ealing Broadway branch twice. For example, knowing the long term plan, one could continue to have District Line trains driven manually between Acton Town and Ealing Broadway until the Piccadilly takes over.
  • Subsequent to this strategy being adopted it now actually looks more and more likely like that either the same signalling system will be used for both lines or at least there will be a higher level of compatibility making this less of an issue.
  • The problem of longer journeys involving a change at Hammersmith and doubling back could be eliminated by stopping all Piccadilly Line trains at Turnham Green. This would also eliminate the current situation when Piccadilly Line trains only stop at Turnham Green at certain times.
  • The platforms for Chiswick Park could be relocated on the Richmond branch. Furthermore, by diverting all former District Line Ealing Broadway trains to Richmond the service from Chiswick Park would be much better than previously with double the number of trains.
  • The fact that the depot isn’t on the line doesn’t really matter. In west London many drivers book on and off at Earls Court so many shifts would be unaffected. There would be some “dead running” but this cannot be avoided.
  • One can only send a limited number of trains to Heathrow. Due to passengers typically being unfamiliar with the tube, possibly jetlagged, and also having a lot of luggage you are never going to be able to run that many trains to Heathrow anyway as the dwell times have to be longer. Even if the Piccadilly only runs a maximum 32tph there are plenty of trains to serve Uxbridge and Ealing Broadway.
  • In 2023 the concession for Heathrow Express expires. It is not currently clear what will happen then, but presumably the mayor would be free to run more Crossrail services to Heathrow – and without charging a fares premium. If that were to happen, and given that there is little chance of any significant capacity expansion at Heathrow in the short to medium term, then it could reasonably be expected that the number of passengers on the Heathrow branch of the Piccadilly Line would actually go down.
  • The number of trains required to serve Ealing Broadway will not be that great if passengers transfer to Crossrail, so it wouldn’t take many trains to provide an adequate service.
  • The fact that the branches of the Piccadilly line are imbalanced is irrelevant. If you run 32tph through the central area then you need the facility to turn round 32tph on the western branches regardless of what is happening on the east side of the line. You are not going to need more than 8tph for Uxbridge and they need to dovetail in with the 16tph on the Metropolitan Line to Uxbridge. There is a limit to what you can run to Heathrow so in practice you can also serve Ealing Broadway without other branches suffering.
Ealing Broadway with Piccadilly Line

The Underground Map in showing Ealing Broadway and surrounding stations in 2024?

The main advantages of Piccadilly Line trains serving Ealing Broadway instead of the District Line are:

  • It makes the Piccadilly Line completely independent of the District Line for normal daily working.
  • The District Line can run 16tph to Richmond in the peak period instead of the currently inadequate 8tph.

Turnham Green

Turnham Green has always had a strange and quirky history of being partially served by the Piccadilly Line. For many years this happened early in the morning and late at night when the District Line was running with extended intervals and on Sundays. With a much more frequent Sunday service on all lines (except the Waterloo & City of course) the all day service on Sunday was dropped.

As we have already seen though, the upgrade of the Piccadilly Line raises a couple of issues regarding stopping at Turnham Green. The first is that if the District Line no longer goes to Ealing Broadway then it would make some sense for Piccadilly Line trains to stop at Turnham Green as the first (or last) station served by both lines. This would avoid longer journey times for journeys such Richmond to Acton Town as well as reduce others such as Heathrow to Richmond. The latter could probably more sensibly be done by bus to Feltham and then main line train but for those relying on the Underground Map or with bulky luggage the Underground will always be the more obvious option.

The other issue at Turnham Green is platform edge doors. These are extremely expensive to install, so realistically you either have to abandon stops at Turnham Green or you have to make installing the platform edge doors worthwhile by having trains call there all day. It is likely significant that London Underground recently launched a consultation for views on additional Piccadilly Line trains stopping at Turnham Green. Whilst this was partially done because of local pressure to provide a better service at that station and the timing was probably not what London Underground would have liked (far too early), the idea in principle probably did not go down too badly at LU headquarters. It has the feel of similar consultations on direct services to Chesham or withdrawing the District Line to Olympia on weekdays – all part of the process of getting the pieces to fit into the final scheme.

BorisWatch spotted a local report that claims that the mayor has agreed that Piccadilly Line trains will stop all day “when the Piccadilly line is upgraded” which seems to support all the above, although this should likely be read as actually meaning when the Piccadilly Line goes to Ealing Broadway. In reality this would not be expected before both Crossrail provides additional relief for Heathrow traffic and all the old 1973 stock trains, with their less optimal performance capability, are withdrawn. On that basis 2023 is an optimistic estimate for this happening.

Sharing track with the District Line

Turnham Green is not the only station where there is some interaction between District and Piccadilly Lines. The Piccadilly Line currently has rarely used platforms at Ravenscourt Park (both directions) and Stamford Brook (westbound only). It would seem that that Piccadilly Line trains will no longer be able to call at these platforms under any circumstances, there being no economic way of them being upgraded. There is also currently the option for the Piccadilly Line and the District Line to share each others tracks between Acton Town and Hammersmith. There would seem no reason why this should not continue in future between Turnham Green and Hammersmith providing suitable junctions were installed at Turnham Green. The only restriction would be that Piccadilly Line trains would be unable to stop at any platform not equipped with platform edge doors.

Piccadilly Line – the trains

As we mentioned at the top of this article, traditionally railway carriages have lasted for between 35 and 40 years before needing replacing. Recently on the Underground we have seen this extend to 45 or 50 years due to a better quality build. At the same time a deep-level tube train has hardly changed at all in outward appearance other than the colour of the finish since the 1938 stock. A Victoria Line 2009 stock looks much the same. The doors are in the same position and are approximately the same width. The bogie and carriage layout is the same. The passenger air conditioning system is the same in both stocks – none at all. It has long been recognised that nowadays what you specify at the outset is probably what you are going to be stuck with for the next fifty years.


A Siemens promotional video of their potential offering for the Piccadilly Line. Because of the tight specification if another supplier won the contract the train would probably be very similar. This is all just window-dressing. The interesting and critical parts of the new train will not be visible.

As stated at the outset, there were two related factors that really drove the need for a radical rethink of what a tube train should be. Contrary to most people’s immediate thoughts neither of these are the need to operate without a driver. The critical factors are heat and energy.

When the tube was built no-one gave a thought to the extraction of heat. Quite the opposite. The tube was quite cool but at least not directly exposed to the outside weather and there was thus no real reason to consider it. Even as air conditioning became technically feasible in mainstream life, few people expected to encounter it in the UK and certainly not on public transport. In any case heat on the Underground really wasn’t a problem. There wasn’t much heat created with trains not as frequent as now – certainly not off-peak which was most of the day – and fewer passengers added their own body heat. The newly-built tube was also bored through cool subsoil and thus came with its own cooling system built in. The trouble is, after a hundred years use, that cool subsoil had now warmed up and is now helping keep the heat trapped in rather than removing it.

A multitude of factors thus mean that the need to extract heat can no longer be ignored.

Energy has also become critical. The obvious reason is cost. There would not be much point in spending large sums of money eliminating drivers if energy costs were higher and potentially easier to bring down anyway. For London Underground it was more than that though. There was also the ethical issue of minimising greenhouse gases. One can ethically source one’s electricity but that only stops someone else having theirs ethically sourced. Finally, removing excess heat from people and trains on the Underground is hard – so why not try to avoid producing the heat in the first place? Essentially any energy that isn’t recycled through regeneration will ultimately end up as heat and require yet more energy to extract that heat outside the tunnels.

One of the biggest contributors to the energy requirements of a train is weight, and a lot of the unladen weight of a multiple unit train is thus in the bogies. So a good way to start saving energy is to cut down on the number of bogies in the train. This has a double benefit because in the current traditional tube train there is absolutely no place available for the air-conditioning units – as clearly an air-conditioning unit is never going to fit in the roof of a deep tube train. Cut down the number of bogies though and it gives you some space under the floor. Of course it is not enough just to get the heat out of the train. It then needs to be got out of the tunnel and on the Piccadilly Line it is intended that around a dozen stations on the line are going to have big projects to remove the heat from the tunnels at that location.

The original idea was to have a bogie at each end of the middle carriage, with all the other carriages having a bogie located at the end furthest from the middle. The current thinking though appears to be that the bogie should be shared between the two adjacent carriages. Each end of the train would have a full bogie of course, so the number of bogies is one greater than the number of carriages rather than twice the number of carriages – a good saving.

This is a considerable saving, although it is worth noting that the carriages will be shorter at 10.7 metres long rather than the current 17.5m. The carriages need to be short so they can be wider as they will be less restricted by sharp curves. As is now swiftly becoming standard across the TfL network, the carriages on the Piccadilly will be walk-through and each carriage will have two wide double doors. It will be interesting to see what develops with the doors. If you make them wider that makes them heavier as well as causing them to take longer to shut (unless you close them quicker). Expect to see novel use of modern lightweight composite materials that we would tend to associate with aircraft not tube trains – especially on the doors.


Again, Siemens design is largely conceptual, but it gives some idea as to what we’re likely to see in terms of door arrangements and internal layout. Photo courtesy of plcd


Inside the Siemens mockup. Again, it gives an idea as to what we are likely to see. Photo courtesy of plcd

It will not just be the trains will be more energy efficient. All aspects of power supply will be looked at and critically examined. Energy loss in the conductor rails will be reduced by increasing the nominal voltage from 630V to a more standard 750V. By the time this happens this will not be new on the Underground, as by then the Sub-surface Railway ought to be operating at 750V and of course both the District and the Bakerloo work at 750V when running over Network Rail controlled tracks. Going from 630V to 750V might not seem that big a deal, but the transmission loss is inversely proportional to the square of the voltage – so it is more significant than may first appear.

On the Piccadilly Line there will, of course, be regenerative braking. But here it is taken a stage further. On other lines it is necessary for a train to be on that particular line and wanting to draw power to be able to take advantage of it. On the Piccadilly Line the necessary changes will not only be made to the substations but also to the substation feeders so that pretty well any train on any line can reuse the electricity.

Also on the power side it is known that there is an aspiration for the train to have sufficiently powerful batteries to get the train into a station if traction current is lost. This would considerably help in emergency situations where it is necessary to cut the power to deal with an incident (e.g. person under a train). Unfortunately due to the way the infrastructure is designed is it usually necessary to remove the power over a much larger area meaning trains are trapped and people are forced to walk through tunnels. Clearly the procedures would need rewriting as it is currently presumed that if there is no traction current then a train cannot move.

In 2013 the idea of battery powered tube trains seems a bit optimistic given the present energy density of batteries. It is true that the civil engineer has battery-powered locos on the Underground and has done so for many years but it would be impractical to carry the batteries found in those locos in a tube train just for the rare occasion when they are needed. What needs to be remembered though is that battery technology is advancing all the time and installing a sufficiently powerful lightweight battery in 2020 may just be feasible.

Calling time on “Mind the Gap”?

Another critical component of the Piccadilly Line upgrade is the need to do something about the gap between the platform and the train if the platform is not straight or very nearly straight. The proposed solution is to have some form of retractable spacer at critical locations.

This will be no small challenge. Dwell times are critical so this would have to be fast-acting. It would also have to be very reliable. Failure to engage would be a serious inconvenience, possibly causing trains to be unable to stop at that station. Failure to disengage would in fact be even worse and probably lead to the all services through the relevant station being suspended whilst the problem is fixed. This is probably the issue that the project team are most nervous of. Although it is a tried and tested system to a certain extent, such spacers are not commonly retroactively fitted to lines and it is only really Ligne 1 in Paris that has such a system fitted to an underground line built over 100 years ago. Having a retractable spacer clearly adds signficantly to the complexity as it has to be interlocked into either the signalling system or the platform-edge door control mechanism.

When will we get the new trains?

The intention is that the first of the new trains would appear in 2021. Given the radical nature of the project this appears to be a very ambitious timetable. It is supposed to be complete by 2024 and there are expected to be over a hundred new trains. They will have to be delivered with a cab because it is felt that it is unacceptable to not have a “driver” at the front if there are no platform edge doors – although for some reason this is not thought to be necessary for Rayners Lane – Uxbridge. Only when the last of the 1973 stock is withdrawn can the platform-edge doors be fitted and it is only when all the platform-edge doors are fitted that the cabs can be removed (or rather converted into a passenger area).

What is not known is what impact the retendering of the signalling contract for the sub-surface railway will have on all this. And of course no-one knows for sure how well the 1973 stock trains currently on the Piccadilly Line will survive old age. The project team therefore have the twin-challenges of getting this radical scheme right first time and avoiding any further delay. Given the concerns of delivering the SSR upgrade on time one will imagine that their progress will be scrutinised very carefully indeed.

Written by Pedantic of Purley