Upgrading the Piccadilly

We last looked in detail at the state of the Piccadilly line upgrades in January, when we revealed that TfL planned to leaseback the Crossrail fleet to pay for new rolling stock. This month sees the Programmes & Investment board look at the project once again.

Item 8 on the agenda for May’s meeting covers the upgrade. Interestingly, whilst the project remains the same, it seems we have reverted to an earlier name. What had become the ‘New Tube for London’ is now ‘Deep Tube’ once again. It is a minor change, to be sure, but one that perhaps reflects the desire of the current Mayor to distance this project from his predecessor.

That this distance is necessary at all speaks to how long the Piccadilly upgrades have been brewing. The current Piccadilly fleet was first rolled out in 1975, with a forecast life of 40 years. Due for replacement in 2014, TfL were in fact forced to life-extend them instead thanks to the failure of the Public Private Partnership that was meant to deliver their upgrade.

The story since then has remained largely the same – the line needs new rolling stock, as do the other ‘Deep Tube’ lines which share a similar unit profile. The problem, however, has been the price tag. Tube fleets are bespoke and that costs money – a lot of it up front, as rolling stock manufacturers aren’t particularly inclined to build fleets they can’t sell to other clients without some cash in hand. For all it’s income, TfL has become increasingly cash-poor in recent years, partly due to the wash-up from the PPP deals and the expensive failure of original project to upgrade the signalling on the Sub-surface Lines, but also due to ongoing cuts to its budget by central government. As a result, the only real movement for a long time seemed to concern the name – with ‘Deep Tube Programme’ becoming the ‘New Tube for London’ and now, it seems, back again.

This was a situation, however, that couldn’t persist forever. The state of the Piccadilly line fleet has continued to deteriorate, as has that of the other fleets that Deep Tube is intended to eventually replace – those of the Central, Waterloo & City and Bakerloo lines. To quote TfL themselves:

A number of age-related issues have begun to occur on these trains. This is being proactively managed through an inspection and repair regime to maintain train availability. Reliability performance has declined in the last three years and is projected to deteriorate further in the final six to eight years of asset life as it becomes more challenging to sustain reliability and availability whilst addressing an increasing volume of repairs.

The leaseback

With no large pots of money on the horizon, and the extra budget pressure of further central government subsidy reduction and a fares freeze, TfL were finally forced to get creative. The leaseback of the Crossrail fleet was (or rather is) their solution.

As we wrote at the time, it is a rather clever and pragmatic one. The chances of a rolling stock leasing company (known as ROSCOs) agreeing to leaseback the Piccadilly fleet itself was always going to be low (or at least very expensive) simply because the potential for that ROSCO to shift a bespoke fleet on to another supplier is small. When the only other place your trains regularly run is the Isle of Wight (and even that now has its issues) then you’re not an attractive prospect to a ROSCO. The Crossrail (or rather Elizabeth) fleet, however, is a different prospect. The Bombardier Class 345s are eminently marketable to other train operators.

You can read more about the workings of that deal here, but the critical outcome was that it created a pot of money of sufficient size to cover the cost of a new Piccadilly line fleet, at least. Whilst the current P&I papers don’t break that cost down in detail (the appendices are not included in the public papers) the TfL Board Papers from December did give us a headline price – £1,549m has been allocated to stage 1 of the Deep Tube project, presumably representing the sum total of the leaseback arrangement and whatever earmarked capital cash TfL still had in the kitty.

What does £1.5bn buy you?

What these current papers do finally give us is an idea as to what that actually translates to in terms of physical assets. Stage 1 covers the design, manufacture and purchase of 94 new trains, along with the necessary new infrastructure and enabling works to run them. It also includes extensive improvements to Northfields and Cockfosters depots, as well as upgrades to the stabling and maintenance facilities at South Harrow. As TfL explain:

These depots were constructed in the early 1930’s and the building fabric, plant and equipment will require extensive reconstruction and renewal. This is to provide the environment and facilities required for the maintenance and repair of modern rolling stock for at least the next 40 years.

Stage 1 also covers the upgrade of high voltage traction power supplies, the DC power distribution system, platform CCTV equipment and legacy signalling systems – all of which are required before the new trains can enter service.

Project timings

That service entry is targeted at the middle of the next decade. Handily, the P&I papers give a rough timeline for the whole project across all lines, although they also accept that it is currently somewhat provisional and dependent on further project scoping.

Current Deep Tube timeline

Elsewhere in the papers, however, we do get a more specific forecast date for first-train delivery. This is August 2023, meaning that Piccadilly line passengers can likely expect to see new trains enter service in early 2024, if that date holds true. We also get some idea as to what improvements the arrival of those trains will enable – an initial push to 27tph by 2026 (A train every 2.2mins). If Stage 2 of the project is granted – new signalling – then that frequency will climb even further. The ultimate goal is clearly to match the 36tph that the Victoria currently manages to achieve.

The trains

Just what those trains might look like remains to be seen. TfL released their own vision for this rolling stock back in 2014, and much of the overall specification seems to remain the same. Most critically, in that they will carry an operator.

TfL’s Deep Tube concept from 2014

An internal view

The exact look and feel of the trains, however, will ultimately come down to the selected bidder. Five rolling stock manufacturers originally expressed an interest in the contract, but this was reduced to three after two combined to bid together and one (CAF) withdrew. The remaining bidders are Alstom, Bombardier/Hitachi JV and Siemens, with the winning bidder to be announced later this month (those wondering why the Siemens CEO is in town should look no further than this).

Competition for the contract will no doubt have been fierce, not least because the potential value of it beyond the Piccadilly line stock is huge. The total Deep Tube requirement is closer to 250 trains, making the run-on orders for the other lines a very attractive prospect indeed. It’s not just the manufacture either – this contract covers fleet maintenance too.

Piccadilly to Ealing Broadway

One of the interesting side-notes of the papers is that they also finally provide official confirmation of TfL’s plans for Ealing Broadway: that it should be taken over by the Piccadilly line. We first posited that this would happen back in 2014 and there have been a number of allusions to it from those within TfL since. Nonetheless, this does seem to be the first time the move has perhaps been made official.

The benefits of this transfer won’t accrue to the Piccadilly but to the District line. As the report highlights, it enable an uplift of 50% on District line Wimbledon trains to Tower Hill, and similar benefits to Richmond. With Crossrail 2 progressing slowly, this will be welcome news to commuters using the line.

Next steps

With the money for stage 1 in place, the immediate future for Deep Tube – at least with regards to the Piccadilly – seems finally secure. More money will be required to see the project through, but once work is underway this will get easier to support. There are few other major capital projects on the horizon for TfL, so it is hard to see how Deep Tube won’t have an element of financial priority. Whilst finding an interested ROSCO may prove hard, it is also not out of the question that TfL might not try and leaseback this fleet to fund the next stage of the project. What matters for now, though, is that Deep Tube is finally underway. We now await news of the successful rolling stock bidder.

278 comments

  1. There is an interesting point in the papers – the Picc will take over the Ealing Broadway service from the District at a future date.

    Some years ago there was a scheme being costed to install a short version of platform edge doors on the Picc platforms from Cockfosters to Heathrow. They were of a type used on the Paris Metro and can cope with curved platforms.

  2. Gah. I meant to include a comment about Ealing Broadway and forgot. I’ll update the piece now.

  3. A correction: the Class 378 units are leased to London Overground. It is the Class 345 units that are to be sold and leased back for Crossrail.

  4. If the District no longer serves EB, what happens to Chiswick Park?

  5. Makes you wonder why TfL are bothering to spend money upgrading the existing signalling to Ealing Broadway under the currently ongoing modernisation of the District line.

    Surely this will only be in for a few years before being ripped out by whatever new signalling DTUP put in. Why not just continue on legacy signalling until DTUP comes along?

  6. @James

    From the document IgnoredAmbience linked at 20.17:

    “The Ealing Broadway branch of the District line could in future be served by the Piccadilly line… In order to achieve this, Chiswick Park platforms would need to be relocated to the Richmond branch of the District line.”

  7. @James
    It looks like there might be room for a new island platform between the Piccadilly tracks at Chiswick Park.
    Another question is how the track capacity will be reallocated to allow continued access to Ealing Common depot by District trains.

  8. @Peewee
    If the Chiswick Park platforms were relocated to the Richmond branch there would be two District Line stations within 500m. I doubt that could be a viable arrangement.
    I believe it is more likely that Chiswick Park would close and Gunnersbury would be the sole District Line station in the area.
    I humbly submit that there are not many passengers who require a service between Ealing Broadway, Ealing Common or Acton Town and Chiswick Park.
    Doubtless this will now be rubbished by a denizen of the wild west.

  9. I think closing any tube station will be nearly impossible politically.

  10. Re Chiswick Park, surely the simplest (and I was going to say cheapest but given the cost of signalling and trackwork changes that would be rash…) way to serve Chiswick Park is to add in some points to the east so that Picc line trains to Ealing and Rayners Lane can call there

  11. 649 passengers a day enter at Ealing Broadway and exit at Chiswick Park. 404 exit at Acton Town; 649 who exit at Victoria would now have to change, etc. etc. – it’s all on this handy data tool.

    https://your-commute.london.gov.uk

  12. I forgot to add that the most important implication is that there is no District swap to Uxbridge, so the increase of western Piccadilly line termini to four means services to the two Heathrow destinations being even more stretched than they are now. They must be reckoning with a major migration to Crossrail, regardless of whether there is a premium on the fare.

  13. Quite a bit of interesting stuff. For once it is official and we are not relying on vague comments and press release hyperbole.

    At last we have real dates – some down to the month. Arguably, there is further slippage as we at told the first tranche of trains will be delivered from 2023 to 2026 so suggesting maybe as late as 2027 before they are fully in service. This is not a further slippage as such but we had expected this to be the full originally planned service of 33tph not 27tph.

    The confusion of is-it-33tph-or-is-it-36tph? seems to be clarified in that current plans are only for 33tph once the new signalling is in but it will be fully capable of 36tph provided the demand money is there. 36tph does seem to be the future gold standard for tube lines – except for the Waterloo & City of course which is always different.

    Taken with another paper is clear that the Piccadilly will have interim signalling which will basically be a like-for-like replacement of what we have now – but with more modern and reliable components – together with more modern supervisory electronics. This will be expected to last at least 10 years but it does mean the programme can get going and achieve something (an extra 3tph, reliability and extra capacity due to the longer trains) without relying on the more advanced replacement signalling.

    It is no surprise the trains will be supervised. For a start there is the interim signalling but also the current Mayor would not be keen to go down that route. And anyway London Underground seem to accept that the Piccadilly line is not the place to start attempting unattended operation. For one thing, implementing the first line on the Deep Tube Programme is going to be challenging enough without that additional non-essential extra challenge.

    Interesting that the Bakerloo is hot-on-the-heels of the Piccadilly upgrade. Hard to believe that the Mayor is aiming for new trains on the Bakerloo to Lewisham in 11 years when we have not yet had all the consultations let alone a public inquiry. And that is probably assuming that he can get away with a Transport and Works Act order and not need a bill through Parliament.

    Also interesting is the fact that the Central line and Waterloo & City line seem to now be due for completion in 2033 and 2034 respectively. One suspects budgetary issues have prompted this delay.

    With Northern line trains due for replacement around 2040 what’s the betting they will start augmenting the fleet in the late 2030s (to replace the ‘paused’ train order) with a further batch of these trains. And subsequently have a full replacement of the current stock with yet a further order.

    All in all though the plan of updating the existing signalling where critical components are liable to fail and having an initial tranche of trains that do not need new signalling seems quite a clever way of getting over the risks of the project becoming unaffordable halfway through.

  14. Anonymous 21:45

    Ealing Broadway to Hanger Lane Junction is not far and the supervisory system will need to know what trains are coming to optimise flow at the junction. Besides, depending on who gets the contract, it might be the same system as Four Lines Modernisation.

    NickBXN,

    Not so. For starters, incredible though it may seem, the current Piccadilly line peak service is only 6tph to T4 and 6tph to T5. So if you have 32tph on the Piccadilly line (and I would suggest that trains won’t go to Ealing Broadway until you do) you could easily have 8tph to each of T4, T5, Ealing Broadway and Uxbridge.

    The Piccadilly line currently has a maximum of 78 trains in service (out of 86 trains). In future there will be 101 trains so probably 94-95 in service. Even with the need to serve Ealing Broadway you could run a better service to Heathrow. Note also that currently Heathrow T2&3 has a train every 5 minutes to London. With airport passengers unfamiliar with the Underground and lots of luggage being carried you are probably limited to every 3 minutes at best.

    Finally, with Crossrail serving Ealing Broadway, I believe we are only looking at 4tph there – if I recall correctly. Even if it were 6tph (same as now) you could do 9tph to each of T4, T5 and Uxbridge.

  15. An Acton Town – Ealing Broadway shuttle would seem to be a possibility, if service frequencies on the other branches need improving.

    It will be interesting to see how they plan to give Chiswick Park platforms on the other side of the road and at a lower level than the existing ones, especially if it is intended to continue to use the existing (listed) station entrance.

    What happened to the proposal to use the W&C as a test bed for the new rolling stock, releasing some 1992 stock to augment the Central Line? Not to mention the idea of re[placing the 1992 stock first because it was in worse condition than the older 1973 stock?

  16. Has there been due consideration given to stabling and maintenance as part of these proposals? For instance, will District line trains still be serviced at Ealing Common Depot, and if not, where will alternative stabling and servicing be provided?

  17. Re Timbeau,

    That disappeared 3-4 years ago, the central line stock is getting retractioned and refurbished shortly (contracts awarded).

  18. @timbeau, NGH:

    On the other hand given the more relaxed approach to automation now taken and PoP’s comments about the Piccadilly not being the best place to start with it, the W&C might end up being a testbed for self-driving operation? Which would help explain why a two-station line upgrade is scheduled to take 9 years.

  19. @Ian J – There’s no doubt the W&C is a beautifully self-contained small system, perfect as a testbed for higher levels of automation. Manual driving it is reputed to be a living hell too what with changing ends every few minutes. I’ve heard in Southern / BR days it was considered very much the short straw assignment for Waterloo based main line drivers.

    As to Chiswick Park District an idea to avoid having two district station in close proximity could be to replace Gunnersbury’s existing ialand platform with a new set of side platforms on the District leg of the the Gunnersbury triangle, with two pubic street entrances. One would be sited on the north side of Chiswick High Road, not far from the existing Gunnersbury station entrance, and one on Bollo Lane, right opposite the existing Chiswick Park station entrance. Another set of side platforms would be constructed on the NLL leg of the triangle for London Overground services with access from the Chiswick High Road entrance. I’d also suggest keeping the existing platforms in use at Chiswick Park for the Ealing Broadway/Uxbridge Piccadilly trains, with the Heathrow services passing through fast in the middle. That way fewer passengers would lose any existing direct connectivity and Uxbridge would gain a direct service to Chiswick with an easy interchange for RIchmond without having to go to Hammersmith and back. Having dedicated platforms for District and LO services at Gunnersbury could help to regulate the junction more efficiently along with the common route section onwards to Richmond.

  20. @Mark Townend – I’d suggest your plans have no chance happening. They appear to involve putting a platform on a nature reserve.

    Plans for Gunnersbury are being discussed between TfL, Network Rail, LB Hounslow and Hermes, the owner of the tower block above Gunnersbury. The common theme is that each of the participants say they have no money and the others should pay for development.

    Hermes want planning permission for another tower block or two in return for releasing land.

  21. So in 10 years time we MAY have new trains on the Picc, but still looking at a 5 year+ project to upgrade the signalling. What a disaster the SSR upgrade fiasco has been.

  22. Mark says ” I’d also suggest keeping the existing platforms in use at Chiswick Park for the Ealing Broadway/Uxbridge Piccadilly trains, with the Heathrow services passing through fast in the middle.”

    Never mind “also”. This strikes me as the only action necessary at Chiswick Park resulting from the transfer of the Ealing Broadway trains to the Picc. In other words, Chiswick Park continues to be served by trains between Ealing Broadway and central London. Two crossovers required. End of.

    Any additional rebuilding of new platforms or stations in the neighbourhood would require its own separate justification, cost-benefit analysis etc, and could not be blamed on the transfer. Stations cost serious money, even above-ground ones.

  23. Chiswick Park – non-trivial westbound traffic from Chiswick, including many to Heathrow. Howls of pain might reduce if part of the plan was full-time Piccadilly stopping at Turnham Green, often mooted but still not committed?

  24. Nick BXN
    Yes, the spreading-even-thinner of the Picc over four destinations is IMHO a mistake that will come back to bite them. ( In spite of PoP’s remarkably optimistic projection of future services 😁 )
    Instead, maybe, there should have been a swap, with Acton Town – Rayners Lane ( & Uxbridge ) reverting to the District, but ….
    As it is, particularly if going to T4, the service is pretty dire right now, never mind being “divvied up” in the future.

    ^^^^^^^^^^^^^^^^^^^^^^^^^^^
    Herned
    I believe your suggestion of “extra points” to get Picc trains to serve Chiswick Park is the preferred one.

    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

    General Thought
    Are the current Picc line trains actually going to last until the replacements arrive?
    Recent trips along there suggest they might not – it’s getting to be the way the Bakerloo was about 18 months back

  25. Greg Tingey,

    But you are losing sight of the objective which is to improve the service to Wimbledon (and Richmond).

    Off-peak, some Piccadilly line trains currently terminate at Northfields. If more Piccadilly line trains to Heathrow were that desperate they would be doing it now.

    Herned has probably answered his own question. It really is probably cheaper to build new platforms rather than install new points by the time you take into account maintenance of the points and associated signalling over a 60 year period (typical amortisation timescale).

    I think the concern about the Piccadilly line trains surviving is one held by many – especially in TfL and the Mayor’s office. But let’s not overdramatise (although I know you like to). We are talking about potential extended intervals due to trains being out of service for longer periods than currently rather than the trains collapsing. If the worst comes to the worst you can usually cobble together x functional trains from y unserviceable ones (where x < y).

  26. Piccadilly line stopping at Turnham Green would be needed to preserve a connection to Ealing and several years ago TfL did say this was planned “following network upgrades” or words to that effect. I think this is covered in a previous LR article.

    FYI there is a 135m footbridge being built between Chiswick Business Park and Chiswick Park station. It has been delayed for years with haggling about money but construction has started and it is meant to be open by the end of the year.

    https://democraticservices.hounslow.gov.uk/documents/s121665/Chiswick%20Business%20Park.pdf

  27. B&T,

    I cannot absolutely confirm but I fully believe that the intention is that Piccadilly line trains will stop at Turnham Green all day, every day.

  28. Last word 2nd paragraph, should it be ‘replacement’ rather than ‘upgrade’?

    Very interesting article, thanks.

  29. @PoP
    So if you have 32tph on the Piccadilly line (and I would suggest that trains won’t go to Ealing Broadway until you do) you could easily have 8tph to each of T4, T5, Ealing Broadway and Uxbridge.

    The new trains will initially upgrade to 27tph (paragraph 4.4), with 33tph once the line has been resignalled (4.11). Paragraph 4.13 is unclear whether Ealing Broadway comes in at stage 1 or stage 2, but 32tph isn’t on the plan.

    Off-peak, a 3tph boost and diversion of Northfields trains would allow 6tph to EB (12 to Heathrow and 6 to Rayners/Uxbridge as now). Peak, however, is more difficult as Rayners Lane sees 12tph, so 27tph means just 3tph to Ealing Broadway at a time when it’s actually needed to be more. Perhaps turning 3tph in the sidings east of Acton Town can be used to maintain Rayners’ branch frequency (it is worth noting that the Piccadilly’s Western branches have a lot of stations outside London’s Travel to Work Area and so changing between branches at Acton Town is very common, so a shuttle isn’t that silly) while keeping Ealing Broadway at 6tph to London.

    I can’t imagine 8tph Piccadilly to Ealing Broadway ever happening. And 4tph would be too low. We’re probably looking at the 33tph being 15 Heathrow branch, 12 Rayners branch and 6 Ealing Broadway.

  30. @B&T But if the Picc has to stop at both Chiswick Park and Turnham Green won’t that use up a fair chunk (possibly) all of the extra 3tph capacity?

  31. SI,

    I chose 32 to keep the numbers simple. If it works for 32 it will work for 33 but be more complicated.

    I hadn’t appreciated that Rayners Lane has 12tph on the Piccadilly line in the peaks – but much fewer all the way to Uxbridge. It makes it more of a challenge but should be still doable.

    James,

    So long as you plan the timetable first and then order sufficient trains to run the timetable, stopping at more stations isn’t an issue. It is when you have a fixed number of trains and there are calls for additional stops that you have to consider if you have the rolling stock to do it without reducing the service level.

  32. Twopenny Tube,

    It depends on whether the author is referring to the entire line (which gets an upgrade) or only the trains (which get replaced).

  33. The NTfL report that Ignoredambience linked to above is pretty categorical on Turnham Green:

    Turnham Green – In conjunction with this change, Piccadilly line trains will stop at Turnham Green throughout the traffic day post-upgrade, increasing the peak tph at this station from the current 12-13tph to upwards of 40tph.

    Note also the detailed work already done on gap fillers in the context of recent discussion of Lewisham.

  34. So when will the interim signalling be introduced on the Picc? The electronics seem to be going in now, add a couple of years testing, so will that be late 2021?

  35. @PoP

    I wasn’t entirely serious when I suggested the cost of altering the signals and adding in the new crossing would be more expensive… Given that the Richmond branch tracks are the other side of a road from the existing station, it’s going to need to be a complete new station, with two side platforms judging by the site constraints, two lifts, barriers etc. and from the Croxley threads that’s going to cost at least £10m.

    A quick google gives a Network Rail cost (2011) of £48,186 per turnout, without installation costs. No idea what they are but presumably at least the same again, plus inflation and signalling works – even if that is £500k per turnout that would still leave at least £8m for maintenance.

    Also are the existing platforms listed? That would be another ongoing maintenance requirement as well as the maintenance of the new station.

  36. Herned,

    I have heard of Network Rail people claiming that a set of points can have a lifetime cost of £10 million. That is probably an extreme case but a rule of thumb is that points account for 1% of track and 20% of track maintenance costs.

    Of course they should be cheaper on the Underground but many have to be examined every 48 hours which is one reason they are anxious to get rid of them the moment they are no longer required. Even the two new sets of points on the Kennington loop have to be checked every 48 hours despite being clamped into one position all the time – except of course they have to remove the clamps to carry out the points testing.

  37. @Ian J – Thanks for spotting, definitely good news for Chiswick residents and also people on the Richmond branch trying to pick up westbound Picc trains.
    Seems that a combination of that, and the idea of turnouts for Rayners Lane / Ealing Picc services to call at the existing Chiwick Park (lowered?) platforms would meet with a lot of local support.

  38. Given the proximity of the existing station buildings at Chiswick Park to the Richmond branch tracks, it may be possible to use the existing concourse, ticket hall and barrier line, with access under Bollo Lane to an eastbound platform. Westbound would, admittedly, be more complicated.

    Picadilly stopping at Turnham Green would appear to be a sine qua non of the project, whatever happens at Chiswick Park, unless you are prepared to countenance Richmond branch passengers for Ealing, Heathrow etc, doubling back all the way to Hammersmith. An alternative might be a two-level interchange where the Overground passes under the Underground (!) near the northern corner of the Chiswick Business Park on Bollo Lane. This might be the preferred solution if the proposed Willesden – Hounslow service ever comes to pass, as it would be close enough to that line for platforms on that to be included too.

  39. Turnham Green stops for all Piccadilly services would provide a good orbital interchange between Uxbridge and Richmond. Overall, this, together with retention of existing Chiswick Park platforms to be served by Ealing Broadway/Uxbridge Piccadilly services alone, is also my hunch as the best cost/value proposition for the area. The new junction required might be offset against removal of some existing crossovers between the two lines around Hammersmith, all in the name of line (and their signalling systems’) separation. To access South Ealing depot from Turnham Green, S stock trains could continue to use their old school train stop capability and manual driving in the interim until full Piccadilly resignalling is complete under whatever system is chosen, and if platform height at Chiswick park was optimised for Piccadilly stock alone, the larger District stock could still pass these platforms out of passenger service.

  40. Further to my last comment, there may also be some additional junction simplifications possible in the Acton Town area to offset the whole life costs of the new Turnham Green junction.

  41. @PoP

    I didn’t realise it needed inspecting that often… but that is opex and not capex and given the financial situation at the moment a cheap fix now might be preferred. I know there have been lots of statements about whole-life costs etc. but that doesn’t always win

    Also the pointwork exists to make this move at Acton Town, it could be that those crossovers were removed and replaced with crossovers allowing access to Chiswick Park so no additional long-term cost.

  42. @Timbeau – I wouldn’t be surprised if they try to close Bollo Lane if the Hounslow – Willesden service happens as the service frequency across two level crossings may mean the road is hardly passable during the day anyway. My guess is that grade separation there would completely blow the project budget.

    It wasn’t clear to me from the board papers if the Picca-Dist change was part of stage 1, or if it was dependent on stage 2, which doesn’t yet have budget and timescales.

  43. As regards Acton Town, Would it be necessary to have track changes there anyway so that District Line trains accessing Ealing Common depot could cross over to Platform 4 to enter the depot from the south?

    Or would the plan be to continue to run out of service District Line trains up to Ealing Common (Or Ealing Broadway) and then reverse direction for the depot?

  44. Unclamping points to check them seems quite absurd – reminiscent of uprooting plants to inspect their roots. I could appreciate someone regularly checking that the clamps are still in place.

    But usually when you hear of something apparently ridiculous like this, it is because the story is incomplete or distorted. Professional road drivers are expected to walk round their vehicle daily, often caricatured as “to make sure the wheels are all there”. But such a check is actually beneficial for harder-to-explain reasons.

  45. @SI
    . Perhaps turning 3tph in the sidings east of Acton Town can be used to maintain Rayners’ branch frequency

    trouble with that is under current track layout you cant send any trains east on piccadilly line until train is fully berthed in sidings so that would delay the service especially if you have problem passengers not willing to get off

  46. PoP replying to my query above: “It depends on whether the author is referring to the entire line (which gets an upgrade) or only the trains (which get replaced).”
    Exactly so, the ‘them’ and ‘their’ seem to refer to the fleet/trains, likewise, the preceding sentence. Hence my question.

  47. @ALTNABREAC – there used to be a long flat diamond across to a reception line at Ealing Common depot from platfrom #3 at Acton Town, as seen here:
    https://maps.nls.uk/view/103193161#zoom=6&lat=6267&lon=9263&layers=BT
    Removed today clearly, and probably not suitable for today’s traffic levels and future aspirations. There would also have to be additional crossovers at the south end of the station to use it and allow Piccadilly trains to weave around depot movements awaiting entry. A new grade separation between Acton Town and Ealing Common stations would be good operationally, with arriving trains reversing in the spurs alongside the platforms at Ealing Common. That might be possible by dropping the Uxbridge southbound to join the Piccadilly flyunder beneath the throat fan at the south end of the depot. A turnout from the Uxbridge northbound might then cross over this line at a higher level to give access to the new reception spurs.

    I incorrectly referred to this depot as South Ealing in my earlier comment at 11:58. Can a moderator please edit this.

  48. The sale and leaseback pays for Piccadilly line stock – but it is a further hollowing out of TfL’s balance sheet.

    Silvertown Tunnel will be a PFI because the debt load is too high to allow for further borrowing.

    At what point does TfL become insolvent?

  49. @Si/NGH

    “trouble [with reversing in the sidings east of Acton Town} is under current track layout you cant send any trains east on piccadilly line until train is fully berthed in sidings”

    Not insoluble, surely – (these are sidings 21,22,23 according to Carto Metro)
    You could reverse in siding 26 to the west of the station (admittedly not so easy as the driver would have to change ends, although auto-reverse might be a feature of re-signalling), with through trains able to use platform 4 from either branch. Alternatively, the resignalling could involve a re-siting of the crossover from slow (platform 4) to fast to be east of the entrance to sidings 21-23, allowing a parallel move from platform 4 to the main line at the same time as another train is entering the sidings from platform 3.

  50. I got my platform numbers mixed up in my comment at 14:01. With reference to the historic long diamond crossing to Ealing Common depot, this started from westbound platform #2, not #3.

  51. @ Transport Insider Now In The Desert

    TfL is an official Government body backed by the UK Treasury and thus cannot technically become ‘insolvent’

    When you consider what happens to private businesses that become insolvent (i.e. they close down and stop functioning) and the critical nature things like public transport or the NHS plays in society, its pretty obvious why TfL is sill classed as an official Governmental body.

    However this does not mean there are not limits as to what a public body can do – The Treasury has set a limit as to how much TfL can borrow and this cannot be exceeded. If TfL ended up carrying too much debt then it could well be put into administration, thus maintaining day to day operations while the UK Government sorts out the problem.

  52. @Timbeau/NIG

    I’m not sure how siding 26 is of much use as it would be a triple reverse – with two reverses in platforms. Though it would avoid conflicts without the need to construct new points further east, track remodelling at Chiswick Park is listed as part of the project (Appendix 1 section vi), so I think such points will happen anyway – not least as it’s going to be the cheapest way to continue serving Chiswick Park!

    Both sets of tracks can serve both branches, however there is the problem that the sidings come off east of the points allowing inner track services to access all four platforms, creating a bottleneck. But the Chiswick Park track remodelling should sort that and make terminating a train every 20 minutes at Acton Town from the Rayners’ branch very reasonable.

    And it’s only a temporary scenario if the Piccadilly gets Ealing Broadway at stage 1. Anything north of 30tph can provide existing frequencies plus 6tph to EB.

    @PoP

    There’s quite a few peak Uxbridge trains, with 15 departures listed between 7am and 9am currently (2 of every 3 services at Rayners?). Off-peak service is a paltry 3tph (1 of every 2 services at Rayners), which I’m sure the many people who change same-platform at Rayners wish was higher.

  53. Speaking of simplifications at Acton Town, is there an operational need to have four tracks in service between A. Town and Northfields?

  54. Regarding the costs of turnouts, the cost of the material- rails, bearers, tiebars, point machines etc represents about 10% of the cost of the installation job.

  55. @ Transport Insider – As Phil says TfL can’t really go bust due to the essential nature of the services it provides. Look back to the collapse of the Metronet PPP deals – Gordon Brown immediately gave TfL £2bn to deal with the financial consequences of that collapse as TfL could not fund the demands of the banks that backed the Metronet deal. Banks *always* get their money back in deals like this.

    What is clear is that TfL are very close to the limit of sustainable borrowing and the debt levels by 2020 and lowered volume of reserves are not a nice inheritance for whoever is the next Mayor. Whether we like it or not TfL as an organisation is designed to be funded by annual fare rises and ideally ones just ahead of inflation. The current regime really is not sustainable despite all the political hoopla about increased efficiency. I find the extent of “family silver” flogging going on rather disappointing and, to be frank, there isn’t much left to sell unless you start closing down services and flogging off railway land. That’s a political non starter so fares HAVE to go up and ideally government needs to stop being ridiculous over things like funding for the road network and operational subsidies. It is utterly wrong headed that public transport fare payers are effectively funding road maintenance.

    The crucial test for whoever negotiates the next capital funding settlement with the DfT is whether TfL will receive enough capital grant to pay for Picc Line resignalling and to allow Bakerloo Line fleet replacement / line upgrade. I remain dubious about whether the Bloo extension will materialise as I think the availability of external funding just won’t be there as the economy is clearly heading for recession. DfT certainly won’t fund the whole cost of the extension.

  56. The above comments seem to forget the passengers from the Uxbridge branch of the Piccadilly don’t want additional stops like Chiswick Park. It already the slowest of the four lines from Ruislip area (Central. Chiltern, Metropolitan and Piccadilly) to Central London but easiest, with no changes to stations west of Hyde Park Corner.

    The plan in 1968 to swap the Ealing Broadway and Uxbridge branches from Piccadilly to District which reached the stage where the train indicators were started to be installed to cater for the swap was abandoned due to Public Pressure

    If TfL want to increase services to Richmond & Wimbledon then let them finance new platforms at Chiswick Park and keep all Piccadilly trains non stop between Acton Town & Hammersmith.

  57. John is right that Uxbridge Piccadilly line users will not appreciate extra stops. I suspect Turnham Green, though, is such a necessity for its interchange that it may be inevitable. (But this is unconnected with the Ealing Broadway issue).

    Stopping at Chiswick Park, however, can be limited to Ealing-Broadway-bound trains, without any new platforms having to be built, and this could be the offered compromise.

    The time delay caused by any extra stops can be set against a general speed-up which may arise due to a better performance of the new trains. (Perhaps).

  58. The argument that the Piccadilly is Uxbridge’s slowest way of getting to central London is a bit of a two-edged sword. Passengers from many other suburbs have only one route, and they are stuck with it whatever. Uxbridge users who find one of their routes too slow can always (well, sometimes) use a different one.

  59. ‘Central London’ is such a huge target as well. Its not as if everyone is travelling to Oxford Circus alone. I assume those who choose the Piccadilly over other lines from the Uxbridge area are travelling to stations in the south west of the central area or elsewhere en route. If the Met is quicker for a particular journey, even with a change, then people will use that instead. It’s not as if tickets are only valid via one route! The huge Chiswick park office development might have a growing influence on demand for inbound commuting though, especially with the new pedestrian bridge being built over the nature reserve.

  60. I’ve used the Uxbridge service via Acton and will not appreciate the longer journey times. Actually the Picadilly line may take longer to arrive to south ken than taking the met and changing at Green park and going backwards! It will just mean the met trains get busier and they are often standing only by Lane already…

    I also point out that this fast and slow things is something tfl appear to want to AVOID. We’ve already off peak fast mets go…Either stop all picc trains at chiswick or don’t. If you have fast and slow services, people WILL change trains at Acton to a faster train and cause delays by trying to board trains about to depart.

    Also, is there any possibility s stock and picc trains could swap depots between Northfields and Ealing common? The district could just use the four track all the way to northfields?

  61. Paul…….as to your last point about swapping depots……..it has been looked at and is impractical. Both Ealing and Northfields are inconvenient for S stock when the EB [Ealing Broadway] service ends.

  62. Speed of travel to “Central London” …
    Now where have we heard that before?
    I found out, last week, that if you are heading for the Eastern end of Central London, or the suburbs (Zone 3 to be precise ) that the time-differential between taking the Piccadilly Line & HeX was … 10 minutes. Because, of course, of the necessity of changing trains at Paddington & often again, later.

  63. @Paul
    People rarely changed from stopping Met trains to semi-fast/fast ones to skip two/three stops and thus two/three minutes (even though TfL recommends it). I’ll doubt they’d change trains to merely skip Chiswick Park (should those trains stop there), especially if they have a seat and would lose it – which is why few do it on the Met.

    I can’t see how Rayners-South Ken via Baker St and Green Park is going to be more popular than the direct trains. 35 minutes on the direct train – which won’t be more than 38 with two extra stops (Chiswick Park and Turnham Green), but at-best 40 via Green Park (and the more typical time, switching to a semi-fast Met train rather than being on one already, is 43 minutes. A stopping Met means 46 minutes). Add in that Green Park’s change is a lengthy walk, and it is very hard to see why people would bother putting more effort in to get there slower!

    Green Park itself is already quicker via Baker Street, but people don’t want to change cross-platform at Finchley Road when there’s direct Piccadilly trains – it’s exactly what John talks about wrt 60s proposals to swap the branches north of Ealing Common around – even if a change is easy, people don’t want to have to make it. Which is why very very few people will change at Acton Town to avoid stopping at Chiswick Park!

  64. If Uxbridge branch users are told that new trains will mean that there won’t be an annual slow-down for leafs, they may be more willing to put up with an extra stop at Chiswick Park.

  65. I was surprised to read John Bull’s article yesterday – as I wrote a treatise on the probs of the Piccadilly Line last year. This didn’t get published as I didn’t think I had the credentials to put forward such views. However I decided to update and publish it in part response to JB’s article. Wont give a link (cos of LC rules) but the post can be found on my 1londonblog.

  66. The extra stop at Turnham Green is unavoidable. There are far too many written TfL and Mayoral statements that say it will happen post upgrade. The only question now is if that means after a new train fleet arrives or after the signalling upgrade given TfL have split the work packages.

    It remains to be seen if a new train fleet is less prone to leaf-fall problems coupled with the “micro climate” on the Rayners Lane branch that causes disproportionate ice and snow. Those environmental issues may present a small challenge to the new automatic signalling and control system when / if it materialises.

    I tend to agree with those who say that people won’t change trains to save a mere 2-4 minutes to travel on a more crowded train. They’ll stay on direct trains. The real issue for me with the Picc upgrade is whether TfL can really boost the line’s frequency and keep trains moving quickly. The Picc’s tunnels are not exactly like the Victoria Line’s “racetrack” with its gentler curves and longer straight sections. If they can get a very high frequency running it will be good but it’s going to present some real challenges at central area stations *if* those trains each carry big loads.

  67. With all these talk of adding stops, does anyone know if this upgrade does anything to increase the chances of York Road being re-opened?

  68. That chance is somewhere between 0 and “Nope”!

  69. timbeau “An Acton Town – Ealing Broadway shuttle would seem to be a possibility”
    Let’s go the whole hog and reopen the shuttle to South Acton as that would give a link from the North London line to Ealing B and Crossrail 1, and hence LHR, which it doesn’t have at the moment. This would alleviate pax having to travel in to zone 1 as well as meaning no need to block a line in Acton Town for reversals.

  70. Mark Townend: A friend of mine keeps getting ‘lumbered’ with the Mill Hill East – Finchley Central shuttle which she hates for the same reason – in and out the cab every few minutes. It could possibly be another option for testing, especially as it has lighter loading that the W&C.

  71. @Alison W 15:42
    I still remember the double-ended G stock cars using the small platform (which still exists behind an advertisement hoarding) at Acton Town station, used for the South Acton shuttle.

    Unfortunately the route of the spur is built over.

    See google link here, this is the remaining abutment for the SA branch bridge over Bollo Lane: https://www.google.co.uk/maps/@51.4982605,-0.2738156,3a,75y,275.93h,87.06t/data=!3m6!1e1!3m4!1smuQxSX3pS65ULz-_xfQ9Wg!2e0!7i13312!8i6656
    Also you can just see the main District and Piccadilly line bridge beyond the entrance to Acton Works (LH side of picture).

    So better to think of how to achieve line-to-line, platform-to-platform interchange between District and/or Piccadilly, and the NLL, and the now-getting-serious ideas by West London stakeholders for a Hounslow-Chiswick-OOC-Harlesden-Neasden-Brent Cross/West Hampstead orbital.

    TfL has seen the South Acton-Brentford corridor in days past as an early stage in a larger ‘outer orbital’, and has kept an open mind on options since the OOC interchanges and OPDC developments started to get serious.

    No one is saying that a line-to-line interchange would be cheap or simple, far from it. However it would potentially create a useful West London interchange for multiple non-Central London journeys, and would therefore conform with the current Mayoral ambitions for 80% of all journeys in outer London to be on green modes (walking/cycling/public transport) by 2040.

    Indeed it is hard to see how such a policy ambition could get near any meaningful result without such interchanges being taken forwards at Chiswick and elsewhere. The Piccadilly services would need to be part of the solution, not part of the problem!

  72. Ealing Broadway to Acton Town shuttles would be easy to implement with trains reversing via the east sidings at Acton.

  73. Hardly anyone takes the Piccadilly line from stations between Uxbridge and Rayners Lane into central London proper. But there are a lot of people going to West London (e.g. Ealing, Acton, Chiswick) and some going to Hammersmith, Earls Court or Kensington. An extra stop on the Piccadilly in west London really isn’t a big deal for most of these passengers.

  74. cjw714,

    Reopening York Road was looked into some years ago and it showed almost no benefit whatsoever. Any advantage was offset by the longer journey times of others.

    The building and platforms are little more than a shell. Fit-out (including disabled-accessible lifts from street to platform level as a minimum and all relevant fire-regulations observed) would mean it would cost hundreds of millions of pounds to do it.

    Specifically addressing your question, the upgrade may even make reopening the station even more expensive depending what additional equipment needs to be installed e.g. platform doors and gap-fillers. Plus the new trains will be are 6m longer which means that probably you have to either extend the platform tunnel when it wasn’t necessary before or extend it 6m further. Not an easy or cheap job.

    So, whatever extremely low possibly there might have existed before is probably made even lower.

  75. Jonathan Roberts: Yes, I was sadly aware that the route is unavailable now (and iirc the metals on former platform 5 no longer exist) but some link between the Overground and Crossrail and/or Picc would be very effective. A South Acton-Brentford link has its merits (I used to live at Key Bridge and would have welcomed it) but it wouldn’t give a Heathrow link unless you count a bus from Hounslow.

    cjw714 / PoP: iirc the extant platforms at York Road (now on York Way, of course) aren’t long enough either, despite there being very substantial new build residential and office blocks nearby.

  76. PoP – thanks for the detailed reply and I have no doubt your final conclusion is correct. The only thing I would add is that I believe the last cost-benefit analysis was carried out in 2005, before the recent developments of the area. Of course, while the benefits may have improved, I am sure the costs will have gone up as well!

  77. Were you alluding to the previously given reason of York Road being un-viable due to lengthening journey times seeming somewhat undermined by Chiswick Park Piccadilly? One is a poor solution to an avoidable problem, the other is an avoidable solution to a poor problem. Which is which though! Both have pro’s and cons, but YR will require big bucks. Maybe one day someone will come up with a business plan, fingers crossed.

    York Road’s platforms appear to be 107.0m northbound, and 105.6m southbound.

    Incidentally, the new stock being 6m longer – this will make it the same length as a 59ts train.

  78. Ben,

    So that potentially makes it worse. Currently length of a 1973 stock train is 106.8 metres according to Mr Wiki. So platforms would be long enough for current stock but not long enough for new stock.

    I think this is almost entirely academic. With the Piccadilly line in its current form (which is unlikely to change at the eastern end) it is probably never going to be possible to justify a station at York Road simply because of the disadvantage of journey time to other passengers. Passengers from 12 stations to London, most of those stations busy ones, would have a longer journey.

    York Road, along with Down St and Brompton Road, was primarily closed in order to speed up journeys on the Piccadilly line. The cost of keeping those stations open would not have been that great in the early 1930s. York Road and Down St closed in 1932. Brompton Road did not close until 1934 when a new entrance to Knightsbridge opened but prior to that many trains passed through without stopping. The question to ask, nearly 90 years later, is ‘has anything significantly changed to alter that view?’

  79. @130

    This was discussed on District Dave. Trains entering the siding delay following trains, The existing layout doesn’t allow parallel moves from P3 to siding and P4 to fast track, and anything on the slow is stuck there until Hammersmith.

  80. PoP
    But the whole: “disadvantage of journey time to other passengers” excuse/reason/answer goes straight out of the window as totally irrelevant, the moment you insert extra stops in the Picc at the Western end, doesn’t it ( Yes/ No ?? )
    And, here we are with strong proposals for extra stops being inserted at Turnham Green & probably Chiswick Park.
    Um, err … consistency doesn’t sem to be strong point in these discussions, does it?

    IMHO: Given current & future loadings & traffic projections, the entire layout of the West end of the former District Railway branches & the Piccadilly extensions is a horrible mess & needs a complete re-think.
    From a practical engineering p.o.v. the only thing that, unfortunately (?) you have to keep is Piccadilly to Heathrow, but everything else should be up for a proper, really careful long-term evaluation.
    Remember that, at present, the Picc has 2.5 branches & the District also has 2 – if you count “Wimbleware” as a separate entity, but the peak-hour (etc) Wimbledon – Tower Hills as a “half”.
    Not easy, is it?

    Can all suggestions be carefully written, double-spaced, on one side of the paper only & then submitted to 55 Broadway 😉

    [ And, yes, I have my own ideas as to how these services should be divvied up, but “Not right now” I think … ]

  81. @Timbeau (reiterating a point I made earlier)

    Appendix 1 section (vi) mentions track remodelling at Chiswick Park.

    The installation of upgraded train arrestors and buffer faces is required at terminal locations to ensure compatibility with the new rolling stock. Detailed survey, design and enabling works will also be progressed prior to implementation of track remodelling schemes at Cockfosters, Chiswick Park, Northfields and South Harrow. This will be required to support the new Signalling and Train Control System.

    Therefore trains will almost certainly be able to swap between the fasts and slows between Acton Town and Turnham Green. This would not only allow Chiswick Park to be served by Piccadilly line trains, but also would allow conflict-free reversing moves in the sidings east of Acton Town if needed.

  82. Greg Tingey,

    To some extent a fair point.

    Chiswick Park is being moved from one line to another so is largely irrelevant.

    The difference with Turnham Green compared to York Road is that stopping Piccadilly line trains at Turnham Green is part of a bigger plan and is an necessity (probably an undesirable one) to not disadvantage some District line passengers. Any disadvantage can be offset against benefits to the Richmond and Wimbledon branches. There are no comparable gains elsewhere with re-opening York Road.

  83. I’m assuming that if Piccadilly tph increases to 26+ in the coming years, this is going to cause some station capacity issues in Central London.

    I know Holborn is planned, but won’t the Piccadilly Circus, Leicester Sq, Covent Garden triptych become hopelessly / dangerously overcrowded in years to come?

    What’s the point of adding capacity if you can’t get people on and off the trains…

  84. Greg: The other difference is that with York Road is that you’re generally slowing down passengers to Central London on their fastest route from the north/eastern Piccadilly line stops to Cockfosters (save for the option of switching to the Victoria line at Finsbury Park).

    For Chiswick Park/Turnham Green, the volume of traffic generally already goes (or will go) via a faster alternative route – Uxbridge via the Met, and Heathrow via Crossrail. So you’re not slowing down the same proportion of end destination journeys

  85. @ Ianno

    Does Hounslow not exist? Or Sudbury? While Heathrow, Uxbridge, Ruislip, etc have alternate routes that are quicker for a lot of zone 1, there’s other places that are served by the Piccadilly.

    While Turnham Green would be all trains, I can’t imagine Chiswick Park stops would be. Ideally, assuming Ealing doesn’t get transferred until stage 2, it would just be Ealing (and Northfields trains if they still exist), unless Chiswick Park needs a boost in frequency.

    Though if it is stage 1 and 27tph in the core when the handover happens, there would need to be trains terminating at Acton Town to keep frequencies at current levels, and so more trains would need to run on the slow tracks in order to give space for trains to go into the sidings. And I highly doubt they’d have trains non-stopping the platforms at Chiswick Park.

    As Malcolm pointed out upthread, the new trains would have better acceleration, so run faster. As such, stopping Chiswick Park and Turnham Green wouldn’t slow trains down vs not bringing in new trains. The same applies to York Road, but there’s a much higher financial cost to stop there, as well as the cost of increased/not reduced transferring to the Victoria line at Finsbury Park to also overcome by the benefits.

  86. Re Timbeau – Acton Town sidings. I understand what can’t be done now, but the line will be resignalled and this is an opportunity to install track suitable for the intended service. I know it’s not trivial, but probably easier than an Ealing Broadway/Ealing Common shuttle which would be inconvenient for folk making eastbound to westbound connections. I haven’t completely figured out which services would use which tracks, but assuming Uxbridge services serve Chiswick Park and serve the outer platforms, then the shuttles would slot in between Heathrow trains on the middle platforms.

  87. @130
    The paradox is that for maximum operability in the Acton Town area, it would be better for the Piccadilly Heathrows to use the outer tracks, most or all of the way from Turnham Green (west of the Richmond line junctions), and for the inner currently fast tracks to be for the ‘everything else’ services. District depot trains might also be best on the outer lines to keep out of the way of the other Piccadilly permutations (and complications). That would free up the central reversing sidings east of Acton for Ealing or Rayners shuttles, and west of Acton for new turnround intensive Piccadilly Central London services.

    The corollary is that Chiswick Park would then be best served by installing a new island platform in the centre of the formation (to be served by the non-Heathrow trains), where the former District Railway eastbound platform used to face onto the southern (DR) pair of tracks.

    Sounds like a lot of scene shifting, so also costs, to improve operability with significant gains, and others might think of even better operating layouts. So let’s just take the view that there will be a lot of optioneering between now and TfL’s finally recommended option, between Turnham Green and west of Acton Town.

  88. @ J Roberts – surely the other question that is key to all this Chiswick Park / what track uses what issue is how the SSR resignalling addresses this 4 track section. Will it be complete segregation as with the Met and Jubilee north of Finchley Road with different signalling technologies. Alternatively is LU “hoping against hope” that what *is* done for SSR will be compatible with whatever is done for the Picc Line?

    I remember a tremendous amount of work being done on the Journey Time Capability measures and targets to try to provide the correct incentives / solutions for the enormous complexity that surrounds the Acton Town / Ealing Common area plus the 4 track section. Whether LU still uses JTC in any meaningful way for signalling upgrades is perhaps a moot point. *However* the underlying technical issues about different stocks possibly with different signalling technology having to share tracks in routine *or* emergency / engineering work situations. Even with the apparent desire to take the District Line out of Ealing Broadway the issues don’t go away as District Line trains still have to reach Ealing Common depot.

  89. WW: The new signalling for the District Line will obviously have to cope with sharing some track with the Piccadilly. So it will have to play nicely with current Picc signalling. The new signalling for the Picc will then, later on, have an absolute requirement of fitting in with what will then be the established setup on the District. These are both minimum requirements – yes they may be difficult, but I don’t see that “hoping against hope” need come into it.

    I doubt if any “complete segregation” could ever be possible (even if it were desirable), because of (at least) District access to Ealing Common depot via Acton Town station, exactly as you point out.

  90. I wonder what the latest state of play is with Covent Garden station. The gross overcrowding has spread to all times (evenings, weekends). There have been various vague proposals over time, which seem to require a new surface entrance on a different site, which together with digging escalators is going to prove a grossly expensive operation. It’s apparently the third-busiest station on the Underground after 10pm at weekends.

  91. WW / Malcolm ( & others )
    Try this, just to see how complicated it is, at present, on the ground
    ( Link to Carto Metro, actually )
    You want to avoid conflicting moves, obviously, whilst serving the different end-destinations as easily as possible.
    Whatever you do, it isn’t going to be easy & I suspect that the best you will ever get is the “Least Worst” solution to all the various difficulties.

  92. Given the Crossrail connection at Ealing Broadway isn’t it likely that there will be big changes in travelling patterns to avail of faster CR journey times into London?

    Hence the thinking might be very different with passenger usage data .after December 2019?

  93. James Scantlebury

    What’s the point of adding capacity if you can’t get people on and off the trains…

    Well, exactly. To be devil’s advocate, if the stations can cope with 24tph then I suspect most could cope with 27tph and slightly longer trains. If nothing else it presents a good case for tackling frequency first. You can always close critical stations for entry and it is rare stations are too busy to allow passengers to exit. And of course, you need to replace the trains anyway so you might as well run a few more if you are able to.

    Mr Beckton,

    Tackling overcrowding at Covent Garden may be grossly expensive but, ultimately, cannot be ignored and the alternatives are probably even more expensive.

    If you ignore it, you simply push the issue onto Leicester Square station. The corollary is that, if you tackle it, you have made a decent start to reducing overcrowding at Leicester Square (and possibly Piccadilly Circus).

    I didn’t know it was the 3rd busiest tube station after 10 p.m. at weekends but could well believe it.

  94. re: signalling compatibility

    As well as Turnham Green – Ealing Common depot sharing with the District line, there’s also the Rayners Lane area (if not further west too) sharing with the Met line that will require compatibility with SSL signalling.

  95. Signalling compatibility – S stock has conventional tripcocks today for use with the trainstops on the District and elsewhere. I expect the solution will be simply to leave these in place when the trains are equipped with Seltrac. The ideal resignalling solution for the Piccadilly in the future will also be Seltrac. The section alongside the District will no doubt be able to share a lot of common infrastructure such as cabling and radio systems, and the section from Rayners Lane to Uxbridge is clearly common with the Metropolitan.

  96. @Jonathan Roberts

    “The paradox is that for maximum operability in the Acton Town area, it would be better for the Piccadilly Heathrows to use the outer tracks, most or all of the way from Turnham Green (west of the Richmond line junctions), and for the inner currently fast tracks to be for the ‘everything else’ services.”

    Why so? As the Carto Metro map linked by Greg shows, thanks to the four track section to Northfields, parallel moves are possible from Acton Town to both Ealing Common and Northfields, regardless of which train is in which platform at Acton Town. Likewise, trains arriving from both branches can arrive simultaneously, again either train can use either platform. Heathrow trains could use platforms 2 and 3, putting them in the right place to run fast to/from Hammersmith, and segregated from everything else (Rayners/Ealing/depot) using the slow lines through Chiswick Park, and calling at platforms 1 and 4.

  97. I agree that regarding Covent Garden station, pushing the problem on to Leicester Square, likewise already overloaded, is a non-starter. I think there was some hope that the Night Tube would dissipate the weekend midnight crush to an extent. Has that happened? I just wonder what the plan is.

    Looking only yesterday at the works along Cannon Street for the new Bank station entrance makes one realise just how expensive and convoluted such works become.

  98. @Timbeau

    Answering for JR with his own words:

    That would free up the central reversing sidings east of Acton for Ealing or Rayners shuttles, and west of Acton for new turnround intensive Piccadilly Central London services.

    Though I’d personally not run both trains terminating both ways at the same time – might as well run them through! 😉

    2041 crowding maps from the MTS has, with the Piccadilly upgrade, standing at 3-4/m^2 from Alperton to Acton Town – in fact the Rayners’ branch is slightly worse with the plan than without it, with standing at more than 1/m^2 from South Harrow, rather than Sudbury Hill. I’d suggest that shuttle trains aren’t in TfL’s plans, but would be a good thing to have.

  99. The discussion about Ealing Broadway and shuttle illustrates that LU has choices. I guess they’ll decide for sure a bit later.

  100. Track changes W of Turnham Green
    1. Crossings E of Chiswick Park
    It’s not possible to stop some but not all trains on a 24tph railway on 2 tracks/2 platforms without introducing the risk of delay, therefore there needs to be an ‘inner’ to ‘outer’ crossing W of Turnham Green Jn to allow Piccadilly trains onto what are currently the District tracks. Hopefully this is not controversial. Whether or not there also needs to be an ‘outer’ to ‘inner’ for District ECS [empty carriage stock] movements will depend on other decisions.

    2. Chiswick Park Station
    Either the outer tracks can be the slow lines and the centre tracks the fast which would not require a substantial rebuild of Chiswick Park station; or the inner tracks could be the slow lines and the outer tracks the fasts which would require a substantial rebuild of Chiswick Park station. Given District ECS moves will pass through Acton Town without stopping, whether they pass through platform lines at Chiswick Park seems moot.

    3. Depot access
    As it stands, access to Ealing Common depot from the East on either the inner or outer line requires a reverse at Ealing Common or some kind of multiple reverse in the sidings at Acton Town and a wrong-way move through platform 3 or 4; the latter seems prima face to be ‘way out’, the former only marginally less so, platform occupation time for these moves if the Ealing Broadway and Rayners branches is at a combined 12tph+ (plus District ECS) is going to add to the risk of delay to the Piccadilly service.
    The least-cost option looks to be adding a crossover to access the depot reversing roads to the East of Ealing Common station directly from the Down line and reversing there instead of in the platform. Aerial photos do question if there is enough space to do that, and if there is whether the necessary slow speed of approach would block the Up line for too long.
    Other options would involve trains entering the depot from the South, which given there is only one road would mean that trains could enter the depot or leave it, but not both simultaneously:
    Add a new reversing siding between the running lines North of the grade separation and a connection from the Up line from Ealing to the NW end of the lede to/from the depot. This would allow reversing in the depot entry/exit track which would slow entries to a crawl
    Connect the Southernmost of the sidings SE of Acton Town to the inner Down line and instate an outer-to-inner crossing between Turnham Green Jn and Chiswick Park. Trains for the depot could wait there for a free slot for a wrong-way move through pl 3 or 4 at Acton Town without holding up following services. That siding is long enough to allow a reasonable speed movement into it from the new connection

    However, the more I look at it, the more it appears to make sense to swap Northfields and Ealing Common depots, giving the outside tracks between Acton Town and Northfields exclusively to District ECS moves. The only interference between these moves and the Piccadilly would be trains leaving the depot crossing on the flat immediately W of Northfields station, but there’s enough space for a train to wait to exit the depot without blocking any workings inside the depot. Crossing 16tph (max?) of Up-Heathrow trains shouldn’t cause too much trouble given the trains would be accelerating onto a clear line rather than manoeuvring into a siding; there is already a grade-separated crossing of the Down Heathrow line. Heathrow Piccadilly trains would use the middle tracks all the way from Turnham Green and Ealing/Rayners trains that call at Chiswick Park would share the outer tracks with District ECS moves. There’s enough space between Acton Town and Northfields that a decent number of District trains could queue there if there was a problem accessing the depot without affecting the Piccadilly service. Piccadilly line trains could enter Ealing Common depot from Ealing Bwy or Rayners Ln the way that District trains do at the moment. Separation of services would also allow simplification of the track work around Acton Town without impacting service resilience.

  101. Moosealot
    A detailed analysis of some of the problems. Thanks.
    I thought of swapping the depots over, but – would District trains “fit” inside Northfields depot, given that it’s been used by Picc trains for quite a long time now?
    This might be a “bigger” problem than simply reverting theActon Town – Rayners Lane services to the District ….

  102. I believe the SSR resignalling will restrict the Picc/District to their own lines between Hammersmith and Acton. The current signalling will remain west of Turnham Green, as beyond Gunnersbury & Putney, with an overlaid ATO. Last plan I saw for depot workings was Turnham Green announcements but, to avoid service delays, checking train for stragglers at Chiswick Park.

  103. With plan to increase to 27tph under current signalling with only 8 more trains, can this be achieved in 2024, a 12.5% service uplift? But I think 27tph was once operated with the current fleet, but abandoned due to blocking back. Is upgrade of the current signalling envisaged, or just the new control centre due this year?

  104. @Moosealot

    Excellent summary. It is a pity the flyunder east by Northfields can’t take two tracks as then entrance and exit to the outer lines could be grade separated from Pics on the inners continuing to Heathrow.
    Further down a rabbit hole, its a shame the reason for the plan is to remove a branch from the District.

  105. The DTUP upgrade envisages a new common control room for all four lines, but with the Picc getting an interim control room this year, and cost pressures on LU, no doubt a new centre can be delayed until later. Perhaps it will be part of the Lewisham project, or maybe the Bakerloo can be squeezed into the new Picc centre and it will await Central line resignalling in the 2030’s!

  106. Surely it would make most sense to combine the Picc control room with the SSR given its interworking with the Met and District? The other DTUP lines (Central, Bakerloo and W&C) have very little to do with each other or any other line

  107. Historical note, as a reminder.
    As far as what was then called Hounslow Barracks, what is now the Piccadilly Line to Heathrow was built & operated by the (metropolitan) District Railway ( And part of the UERL group, later on ) until 13th March 1933, when “Picc” operation took over (mostly)

  108. Re James,

    That makes far too much sense! I suspect the “temporary” new control room might become permanent.

  109. Taz 21:44,

    My interpretation is that you cannot increase the frequency until you have got rid of the 1973 stock with their slower acceleration and higher dwell times due to a legacy door arrangement. So, much like the Victoria line which couldn’t really improve until the last 1967 stock train was decommissioned. So that means 2026 not 2024.

    Also, I suspect power supply arrangements will mean you need the more-energy efficient trains introduced first to give sufficient power for 27tph. I know the full upgrade includes power supply upgrades but I suspect they will be severely curtailed in phase 1.

  110. @Greg
    The loading gauge appears to be there from photos I could find on the internet. An S7 is just under 11m (~2/3 car) longer than a ’73 and aerial photos suggest they’d fit quite comfortably lengthways, too.

    @Ben
    Yes it would be nice if there was complete grade separation, but crossing one line is far easier than crossing two simultaneously: in fact I would posit that crossing a running line is less of a conflict than joining a running line, which is what happens out of most depots. I can’t say that I’m “sad” for there to be different colours on the map at Ealing Broadway. If extra capacity that becomes available on the Piccadilly means that the District service to Richmond gets better capacity and the denizens of Ealing get a slightly faster service to Earl’s Court it won’t be a disaster.

    Given the depot swap isn’t going to happen, I will stop here to avoid having a new derogatory name invented for fantasy depots.

  111. @James
    That would have been a great idea but it’s probably sailed now. I imagine TfL weren’t ready to retire the ’73 stock at the point the SSR resignalling was originally supposed to be completed but equally did not consider that their life would be long enough to be worth fitting the new gear to and/or there wasn’t enough money in the budget to do that and resignal and purchase the S stock. My guess is that procuring the signalling for the SSR and adding the option to add the same signalling to the Piccadilly 10(?) years later would have (1) fallen foul of EU procurement rules without extreme care and (2) reduced the number of interested bidders even further. Adding space at Hammersmith to slot the Piccadilly control in would have been a good idea, but having the space empty for a decade or more and possibly limiting the options — or even appearing to limit the options — when eventually procuring the Picc signalling might appear too much cost for too little gain.

    @PoP
    Given the new trains will be more efficient, even running a more intense service shouldn’t require additional power. I suspect the power supply upgrades are more to do with allowing regenerative braking; this in turn is at least as much to do with reducing heat build-up in the tunnels as it is with saving electricity (although if it can prevent the need for additional grid capacity, that represents a major cost avoided) so is probably not required until the more intense timetable come into force which — as you suggest — is only going to be possible once all the ’73s are retired. That said, it would probably be necessary to upgrade one section in order to test the regenerative capabilities of the new trains once they arrive, and if there are any efficiencies to be gained in doing the lot in series along the line then they may well get done earlier.

  112. re Moosealot,

    They will also want to upgrade from 630V to 750V to improve efficiencies especially with regenerative braking as a lot of the exisiting kit is near end of life.

    I suspect that there are number of factors preventing an increase (ignoring signalling)
    – lack of stock
    – capacity issues at certain stations e.g. Holborn
    – extra heat
    – existing stock door arrangement

    hence it might need 2/3s of the ’73 stock replacing before the balance has tipped far enough for some of the factors

  113. James, Moosealot,

    There is always the possibility that by the time the Piccadilly signalling upgrade is done, there is space available at Hammersmith Control Centre due to further technological improvements involving miniaturisation of components and more work taken off controllers so fewer desks required.

    So there is always the possibility that there will be space in 10 years time.

  114. A new grade separation between Acton Town and Ealing Common stations would be the gold standard for down direction access to Ealing Common depot operationally, with empties arriving in roads 19 and 20 on the south side and then able to shunt to any other road via headshunts at either end. That might be possible if the up Uxbridge was dropped to join the up Hounslow slow just before the flyunder beneath the yard throat fan at the south end of the depot. A turnout from the down Uxbridge might then cross over the three low level tracks via a new bridge to access the yard conflict free.

    I like the idea of releasing stragglers if neccessary at Chiswick Park en route to the depot to avoid an extended stop at Turnham Green, but it seems a good idea to ensure an empty down District can wait if neccessary on the slow between the Richmond District turnout and Piccadilly crossover so as to avoid delaying any Piccadillies turning out to Chiswick Park. That should be possible for the down direction easily, as the District junction is already close to Turnham Green while the Piccadilly connection can be added near to Chiswick Park. Its not so easy to arrange such a regulating ‘pocket’ in the up direction, but it’s less of a priority in that case as the empty Districts are entering service so won’t be getting in the way calling at Chiswick Park.

  115. Mark Townend…whilst understanding the merit of your proposal, there is unlikely to be any prospect of a flyover “just” for stabling trains. An awful lot of dead mileage to Ealing Broadway and back would have to be accumulated to cover the cost of the flyover.

  116. Why can’t returning District line trains reverse in platform 1 at Ealing Common? OK, you might need a second driver located at the station to promptly take over at the other end unless you can get auto-reverse into the depot approved. The trains won’t have passengers on it so it should be possible to do it very expediently.

  117. Hammersmith SSR control centre was built for Bombardier signalling contract, and was to include Picc Barons Court to Northfields/Uxbridge. Current Picc trains were to have cab signalling. The Thales system is more centralised, and there are big pressures on space at Hammersmith now.

    With Picc to Ealing Broadway, the line will only interwork with District depot trains, and the Met from Rayners Lane. The 4 DTUP lines have little in common now, but will share common train design and signalling systems. A new control centre will monitor these in much more detail than existing lines do, with the possibility of one day providing remote control. Hence the aim of a single control centre for them.

  118. PoP..…. I think but the crossover (Google maps) to the EB track and thence the depot is very close to the end of platform 1 and may even overlap it. I have a dim memory that trains exiting Ealing Common going west couldn’t stop at the station, but might be wrong.

    Even if they could reverse in platform 1, I certainly wouldn’t do so routinely if I’ve got three platforms a mile down the line to play with to reverse trains comparatively leisurely.

  119. PoP – Reversing in platfrom #1 would be possible but could be somewhat restrictive, as the movement into the depot could not take place at the same time as a up run-in of a Piccadilly into platfrom 2 which would thus have no overlap available.

    Another idea might be to have a ‘right turn lane’ reception track between the up and down Uxbridge lines starting just after the flyover crossing the Hounslow lines. It would be long enough for a single incoming train to wait between the running lines and would have a nice friction buffer stop overrun at its country end. An incoming train could thus wait there, without delaying following down Piccadillies, until an up Piccadilly had called at and subsequently cleared Ealing Common station and the depot junction. Then the incoming empty could cross the up Picadilly and move into one of the two spurs behind platform #2 (roads #31, #32 – one of which is out of use today but might conceivably be reinstated). A reversal in the spur could thence access any depot siding. This would require some depot junction modifications clearly including a local realignment of the up and possibly the down Uxbridge lines to make room for the centre right turning track.

  120. Taz has beaten me to it but 6 or 7 years ago (yep that long) I sat in meetings between the former SSR resignalling project and Tube Lines to discuss the implications for the 73 stock. It was certainly the plan back then to equip 73 stock to work with the SSR resignalling. Naturally enough there was a lot of concern about the impact of the work on the reliability of 73 stock – especially disturbing cabling in the trains.

    Obviously a lot has happened in that time and is no doubt still up for some debate re resignalling. I just offer the comment as a little bit of historical context.

  121. The 6 metre longer trains will fill the platforms as previous Picc line trains did, providing 5% more capacity by that alone. The Bakerloo & Central trains already fill their platforms, so such an increase is not available there, but the current W&C trains are shorter than their predecessors so longer trains may fit.

    This increase in train length, along with a more spacious train interior layout provides an additional 19% passenger capacity per train. Such an increase would be welcome on the Northern and Jubilee lines, where current trains are composed of similar length cars to the Piccadilly 1973 tube stock.

  122. Taz,

    At some platforms on the Northern line the current trains barely fit. I am pretty sure the guard ‘cut out’ the rear doors at certain stations before. I am also fairly sure there are still cases of rear doors not opening currently – but this is largely due to the platform tunnel being encroached on by false walls containing electrical equipment etc making the platforms too narrow. ATO enabled some of the previously SDO stations to have all the doors open. When they next upgrade the Northern line they really need to have a comprehensive survey to work out just how long a train is practical with suitable remedial work.

    Not so in the case of the Jubilee line. Maybe you were thinking of when they had 6-car trains – before they added a seventh in 2005. Remember the carriages are longer than their predecessors so a 7-car Jubilee line train is roughly the length of 8 cars of a more traditional train.

    The problem with longer trains on the Waterloo & City is that there is very little overrun at Bank – and there was virtually none in BR days. So there is no significant extra length to be had other than by expensive tunnelling work at the Waterloo end of the Bank platforms.

  123. If you look at TfL’s animations, etc., the passenger doors at the ends of the train are some way away from the actual end of the train. This suggests to me, that the train will be somewhat longer than the platforms, with TOps using on-board CCTV screens to monitor doors.

    Stadler is doing something similar with the new Merseyrail stock:
    http://www.bbc.co.uk/news/uk-england-merseyside-38998245

  124. @PoP: There are quite a few stations where the Northern line trains don’t open first or last doors. From memory:

    – Clapham Common
    – Clapham North
    – Moorgate
    – Charing Cross

    (There are probably a few more…)

  125. The endmost door-endmost door length is probably a more useful value for trains which do not need the driver at the platform. Unless only one double door per end carriage needs to be opened in the worst case scenario.

    The shortest platforms on the pic are seemingly at Russell Square, southbound. 105m.

    This would seem to suggest that the new trains will at least have all their doors open at every station.

    As for the Northern, if it ever were split the potential would exist to run two separate fleets. It is a pity, then (though understandable), that the rebuildings that have taken place on the Bank branch since the 1980’s have not specified 8car platforms.

  126. Does anybody know if air conditioning is still part of the spec (or was it ever?)

  127. Re Tyteen,

    Aircon has never been included however limited air cooling is the favoured option.
    Regenerative braking is the biggest game changer for lowering temperatures.

  128. @Tyteen4A03
    Air conditioning is difficult on a deep tube line because there is nowhere for the heat to go – if you make the trains cooler, the stations will get hotter. And with doors opening every couple of minutes, the heat will simply come back in again anyway.

    (not nearly as daft as trying to install aircon on an open-platform bus though………….)

  129. Really informative Programmes and Investment Committee meeting this morning with an excellent explanation of the rationale behind the current order.

    The current application for approval is very specifically for an initial tranche of trains for the Piccadilly line so the life-expired stock can be replaced and some improvement made. So it covers the 94 trains and everything else essential to introduce them (e.g. depot changes). Nothing that isn’t needed at that stage is included.

    The aim appears to be to order the minimum necessary to make a worthwhile difference in order to keep a tight rein on cash.

    A Board member asks what it committed them to and could they envisage any situation where the options for further trains weren’t taken up. The rely was they were only committed to the first order (the 94 trains). The only circumstance they could perceive of not putting in orders for the other lines is if they simply can’t afford the trains or they are dissatisfied with them for some reason.

    The Board (and General Counsel) were clearly really anxious that the sums involved weren’t revealed for (very good) reasons of commercial confidentially. In the closed session they recommended approval but with slightly changed wording – although we don’t know what changes they made.

  130. If the middle carriages of the new trains are two thirds the length of 1996 stock, which could make an eventual transition to trains with a standard door spacing easier at the existing platform edge door stations, a nine carriage train would be six metres longer than 1973 stock. Is that the plan?

  131. Re TFB,TFO,

    Sort of, the end cars are about 14.5m and the intermediate cars at 11.9m with

    Bakerloo having 7 intermediate cars
    Piccadilly having 8 intermediate cars
    Central having 9 intermediate cars
    W&C having 3 intermediate cars (possibly 4 with a new signalling system.)

  132. NGH

    Those dimensions seem a little high to me. The aim was for each car to be the same length as existing cars’ bogie centres so as not to increase centre throw. I know not what the tenders say!

  133. When trains stop with the cab in the tunnel, currently the first door opening can be beside the tunnel headwall. The new trains will allow for future platform-edge doors, which need room to open, effectively shortening available platform lengths.

  134. @Taz No reason the platform edge doors couldn’t be single leaf (double width)

  135. Timbeau

    If the train doors are width X, the platform screen doors (PSD) are X +500mm. If the PSDs are double width, they will be much heavier than the single doors. The time taken to open a given width will increase a) because of the greater distance to travel and b) because the maximum velocity will be reduced to ensure the kinetic energy of the wider door is no higher than the narrower one. This doorway would serve more customer spaces than other doors, so would become the constraint on the dwell time.

    it is in everyone’s interested to have double passenger doors on the train; that’s one of the key Deep Tube Upgrade objectives. The follow on is double PSDs operating in sync with the train doors. There’s a new BS EN on PSDs grinding its way though the consultation and publishing process that encourages just that.

    By the way, PSDs tend to be heavier than train doors, so getting synchronised operation is a challenge.

  136. Re 130,

    That was going off the number of cars TfL said few years ago which may well have changed, if you go off bogie centres @circa 10.0m with articulated stock then

    End cars at 12.0m and intermediate cars and 10.0m so:
    Bakerloo having 9 intermediate cars
    Piccadilly having 10 intermediate cars
    Central having 11 intermediate cars
    W&C having 4 intermediate cars (possibly 5 with a new signalling system.)

  137. NGH…hopefully in a couple of weeks we’ll know what the suppliers tendered.

  138. @NGH

    So, ever so roughly:
    Bakerloo – 114m
    Piccadilly – 124m
    Central – 134m
    W&C – 64/74m

    These are excellent increases across the board, the Piccadilly being well in excess of the 6m lengthening stated elsewhere.
    Is the B’loo limited by London Road depot?

  139. PoP
    The S stock will still cut out end doors at some stations.

    The Jubilee line trains are still 7m shorter than standard 8 car train the line was built for, so replacements could be 5% longer.

  140. @Taz: currently the first door opening can be beside the tunnel headwall. The new trains will allow for future platform-edge doors, which need room to open, effectively shortening available platform lengths

    Couldn’t you carve out a pocket in the tunnel for the outermost door to open into? It would only need to be as wide as the door itself, so much less excavation than extending the platform.

  141. Re Ian J,

    Often there is lots of equipment at the platform ends that is boxed in behind a false wall, in addition to platform monitors so in many case it should be relatively easy.

  142. Taz,

    I hadn’t appreciated that. I suspect the next problem would be the depots. The Victoria line trains could be slightly longer (next time) as not all the platform is used and the former Signalling Equipment Rooms located at the end of the platform tunnels are no longer used for their original purpose. However, I am told making the trains longer (next time) might still be challenging because it is often the depot that is more of a challenge than the station platforms.

    So I do wonder what the challenges would be for longer trains at depots on the Jubilee line. More on topic, I suspect that depot rebuilds (e.g. gable end at Ealing Common) are only ever rebuilt to be just long enough for the current stock.

  143. The current Victoria line trains are slightly longer than the old ones and depot pits had to be extended for the new trains.

  144. 100andthirty

    Thanks for info. For what I can recall from my two visits there is that the buffer stops are really close to the end of the shed, there is no obvious way of extending the shed and no room at the other end where there is the pointwork leading to the tunnels. And the site is very cramped.

    I may be wrong as I really got little more than a fleeting glance but it did highlight for me what I was told about the depots often potentially being the biggest problem when it comes to lengthening trains.

  145. TfL press release suggests that first operations in public service on Circle and Hammersmith & City lines takes place “this summer”.

    What is it Roger Ford says about not trusting implementation dates that are based on seasons?

  146. Re PoP,

    I think the Bloomberg entrance at Bank is the best example – 6 summers later!

  147. NTfL is an exciting project, introducing trains which may serve London to 2066, and therefore envisages later use of remote control. However, it starts with new trains, then new signalling, and then further automation as money becomes available. I just want to caution our younger followers that the current Piccadilly and Bakerloo trains were introduced in a similar vein near 50 years back, none of which came to pass. The Victoria line had just been introduced with automatic trains, although a lot of its station automation plans did not come into effect. Both the 1972 and 1973 tube stocks have ATO switch positions in their cabs which have never been used. Piccadilly line resignalling for auto-operation once appeared a strong possibility, with auto-reversal in sidings.

    Strangely it was also these trains which were once intended to introduce articulated cars with a bogie shared between adjoining cars, a prototype being developed at Acton Works. Instead, it is to be their successors that introduce this feature.

  148. Barons Court and Gloucester Rd should be skipped to speed up the western branch.

    Unecessary District Line duplication, when most Piccadilly passengers are heading to the west end or beyond and there is easy interchange if you want these stations.

    Chiswick Park, Stamford Brook, Ravenscourt Park and West Kensington are already successfully by-passed, providing fast and slow lines, boosting travel times to places like Heathrow with minimal negative impacts.

  149. P.Riddy
    Gloucester Rd is a better interchange than S Ken – at present, at any rate.
    The S Ken station rebuild may fix this…
    Leave Baron’s Court alone, please – it’s much more civilised for x-platform than Hammersmith, which I consider vile – the rebuild a few years back got almost everything wrong, from an architectural p.o.v. & doesn’t even keep the rain out (!)

  150. Yes, Baron’s Court is the insider secret…. Touch at your peril!

  151. Re Greg and SH(LR),

    Completely agree as both are needed to help reduce interchange volumes at neighbouring station that struggle to cope with high entry/exit and interchange volumes already.

    Baron Court’s should be the recommended Heathrow with luggage changing location!

  152. Even if Barons Court were to be skipped, the eastbound running speed into the tunnel is hardly express. Before the current, long-running speed restriction, it used to be the highlight of a Tube trip to run fast down the ramp in 59 stock and hit the inner single tunnel portal with a big pressure change, rattling the doors and drop vents, followed by a very intense noise of whooshing air and deep rumbling as it took the left hand bend. Just a bit of nostalgia. Perhaps it was even more intense in gate stock!

  153. I had assumed the winning bidder would have been announced by now, does anyone know if it’s been announced? Maybe they’re waiting for this whole new timetable thing to fade a bit?

  154. @ MT – definitely not announced. On the assumption that authority was granted by the Committee on 16/5 and then the mandated 10 days “stand still” period then we should have had an announcement. This leaves three possible outcomes – authority was not granted (the report to the Board on 23/5 was silent about this), there has been a challenge from an unsuccessful bidder during the “stand still” period or something else has transpired. I am not terribly surprised that we’ve heard nothing – TfL press releases are very tightly controlled by City Hall and one concerning a big train order would have to go via City Hall. We don’t get the torrent of information we used to get from TfL.

  155. Re WW,

    A considerable amount of press release writing would have to take place if the trains aren’t being made in the UK…

  156. @WW

    Thanks for the detailed insight. I’ll keep my fingers crossed.

  157. @NGH: Given the list builders: Alstom, Bombardier/Hitachi JV and Siemens, the chances of that are quite high.

    But even if they were made/assembled here, the number of foreign components would be substantial anyway…..

  158. SHLR – You’re using logic. With Widnes, Derby, Newton Aycliffe or Goole, each of the protagonists has or has announced a potential assembly facility. That’s all the politicians need to get them off the “jobs for the brits” hook. I would suspect a delay caused by a legal challenge. After all, if you’re going to be locked out of LU train orders for a generation, why not? Pure speculation, so probably due for a snip!

  159. Re 130,

    Except at least 2 of the bidders are aware and work towards total uk content so it isn’t as box ticking as it might look…

    And after all the recent record of legal challenge is good.

  160. NGH…….. working towards……isn’t that a wonderful phrase, but I’m with you on legal challenge.

  161. An article in the Standard yesterday titled “Sadiq Khan launches probe as cash-strapped TfL pays record £50m in ‘golden goodbyes'” mentions…

    “It comes at a time TfL has been so cash-strapped that it shelved plans to buy new Tube trains…”

    I hope they’re not referring to the Piccadilly Line?!

  162. MT – the point about the shelved trains is referring to the cancelled order for additional trains for the Jubilee and Northern lines………………I hope!

  163. I note that Grayling, in his Heathrow runway three announcement, indicated that there would be a further extension to the Piccadilly line. Presumably this will require an increase in the number of new trainsets too.

  164. AlisonW – “…..increase in the number of new trainsets…..”
    Is this of the yet to be announced order or their successors?

  165. @ Alison W – interesting remark from Mr G there. I’ve just looked at an updated summary document which says Heathrow Airport will confirm its surface access strategy “soon”. The document says the Picc service goes to 18tph from 2030, Liz Line to 8tph from 2040 and there is an indicative initial 4tph into the airport on each of the proposed Western and Southern Access main line rail links. The Southern link goes to 8tph from 2040. No mention of a Picc Line extension though. Presumably Mr G knows something that Heathrow isn’t prepared to say just yet.

  166. WW: “Presumably Mr G knows something that Heathrow isn’t prepared to say just yet.”

    Or, of course, he could just be making it up. (cf railways)

  167. AlisonW,

    Presumably this will require an increase in the number of new trainsets too.

    Possibly not. They might abandon transferring the Ealing Broadway branch. Except that leaves another problem on the District that you somehow ought to deal with and there is no obvious solution as to how you can otherwise increase frequency on the Richmond and Wimbledon branches*.

    * You could run a shuttle between Ealing Broadway and Acton Town as people have suggested but that might create a problem of overcrowding at Acton Town. If that shuttle used Piccadilly stock then you would still need more stock.

  168. Most but not all versions of Heathrow expansion have included a new terminal 6 (aka North Terminal or even could be Terminal 1), between the northern existing runway and the new runway. An extension of the Piccadilly would be built to serve the new terminal.

    Recent versions have dropped the new station and extension, with the terminal served by access from terminal 2 station; and even eliminated the terminal altogether. The http://www.heathrowexpansion.com website is extremely vague on the infrastructure, seems to show considerable development in the T6 area but discusses passenger changes only to T2 and T5.

    So we have to wait to find out what has been decided.

  169. @ Neil – I read a version of Heathrow’s view of required surface access improvements a number of months ago. There was next to no tube based expansion and little bus service expansion. It was as if there was no impact whatsoever / public transport could cope. That sounds beyond fanciful to me. I understand TfL and City Hall were in complete disagreement with HAL’s view. As you say we must wait and see but I suspect we won’t see much money being directed towards TfL.

    I can see the Western and Southern rail access schemes proceeding but no doubt with horrendous fare surcharges as Heathrow again funds the tunnel construction for these schemes and then rips off the public via TOCs for the best part of 25-30 years. I expect Network Rail won’t be let within 100 miles of these schemes (other than at the junctions to existing lines) as part of Grayling’s “anti NR / private is better” vendetta.

  170. @Neil “Recent versions have dropped the new station and extension, with the terminal served by access from terminal 2 station

    Terminal 5, surely?

    A new runway & terminal would presumably speed the planned removal of Terminal 4, which would simplify Piccadilly operations.

    @WW “I expect Network Rail won’t be let within 100 miles of these schemes

    If HAL pay for them (surely a pre-requisite of approval?) and for some strange reason choose Network Rail to be in charge of building the schemes, I can’t see the DfT stopping that as they aren’t the customer, plus NR get some cash from outside sources.

  171. Picc T5 extension was built with provision for further extension to possible M25 park & ride station at the time.

  172. A reminder that this thread is about the Piccadilly line. Whilst discussion on Piccadilly line access to a new terminal is acceptable, any comments solely  about Heathrow will be removed.

  173. It does seem that the separate T6, which would have required new railway links has been removed, and instead an additional terminal next to T5 will be built instead, sharing the existing stations.

    It certainly would be convenient if T4 closed, as that way all trains and tubes could be sent to T5/T6, but I can’t see any definite plans to do so. T4 is tiny in comparison, but too important to not run services to as long as it’s open

  174. T4 has always served a lot of airport workers too. Are freight operations still based in that area?

  175. At 2.28m entries/exits in 2016, T4 is not unbusy (but nor is it that busy). Quieter branches like Chesham (1.07m) and Woodford-Hainualt (intermediate stations add up to 1.56m) survive but if the airport terminal closes, then that’s the raison d’etre of the station gone.

    The cargo area isn’t that close (~1km), plus the area doesn’t lend itself to walking, so I suspect workers wouldn’t suffer much from having to get a free bus from Hatton Cross station as they likely do so from 2 stops further along the 482/490 routes at Terminal 4 currently.

    Bedfont residents might similarly use the station, but, again, distance and difficulty walking to/from the station mean that having to go to Hatton Cross is unlikely to be much of a hardship, as people are probably not walking to T4 station anyway.

    Now, the site might be redeveloped into something that could do with a tube station, but HAL’s plans seem to suggest losing the rail links to T4.

  176. The question of future use of the one-way T4 Piccadilly Line loop is intriguing.

    For resilience, there may be a good case to retain the loop as an operational going concern, if for good or bad reasons trains had to be diverted from T5. Just as the Green Park-Charing Cross line is still maintained in operational readiness. Presumably drivers would need to continue to run occasional trains to maintain route knowledge, though how that would be viable with NTfL use, is unclear.

    However a regular service on the loop could only be justified if a combination of passengers / airport workers / new travel-generative development occurred.

    So the main stimuli might be, post T4 closure, whatever new developments were authorised in the footprint of T4 (also relevant for the Crossrail station there, and would both lines be needed?), and whatever else might be worthwhile around different parts of the airport estate in order to achieve a higher public transport modal split and less vehicle pollution.

    There are two potential locations for other stations on the loop. (1) Stanwell/Air Cargo centre (largely airport workers but maybe merits too for Stanwell Village though it is technically outside the GLA area). (2) In the heart of the new ‘toastrack’ terminal zone on the west side of the loop, where there is passive provision for a station in the loop design, and an existing access to the loop tunnel.

    TfL owns the loop tunnel but HAL owns the loop T4 station – one of only two tube stations that TfL does not own! (The other is Pimlico – Crown Estate). The benefits of creating new stations could presumably be considered in any public inquiry, in due course, and advocated by any interested party. However it would appear self-evident that it is really HAL who would have to be motivated to use the loop creatively and bear the costs, rather than just have a mothballed or ‘lightly-dozing’ line.

  177. Surely there are other Tube stations not owned by TfL? Richmond and Wimbledon come to mind, and I think Barking and Upminster.

  178. timbeau,

    My thoughts almost exactly. I actually mentioned three of your four stations as examples. Jonathan explained that he really did mean ‘tube’ (as in round tunnels) stations. There is also a query on T5 – and saying it s a concrete box not a tube is stretching a point.

  179. Apparently, the order for new Piccadilly line trains has not yet been placed but it will be in the next few days. Also, a bit of a surprise, Mark Wild, head of LU, insists they will be fully air-conditioned and not just air-cooled.

  180. PoP
    If that is correct, doesn’t that mean that more heat will simply be pumped out into the tunnels & stations?
    Or will they have regenerative braking, thus reducing the total heat output?
    An interesting balance.

  181. PoP …..There was extensive debate the S stock air conditioning was ordered as to whether it was the real thing or merely “air cooling”. This led me to believe that neither term is clearly defined.

    Generally, if air is to be admitted to the saloon at a lower temperature than the saloon temperature, it needs a refrigeration process. This process will liberate some moisture from the air, so a facility needs to be provided to drain the condensate somewhere other than the passengers’ heads. Finally some heating needs to be provided to regulate the temperature, otherwise the vehicle might be come too cold. You could switch the refrigeration cycle on and off to regulate the temperature but the average refrigerant compressor doesn’t respond well to repeated switching. Thus, I think air cooling is generally a myth.

    If you want to try the effect of air cooling, experiment on a car with “air conditioning” (air cooling in this context) compared with “climate control” (true air conditioning).

  182. Greg Tingey,

    Well yes. But they must have thought about this so I presume this will be accompanied by other means of removing heat from the tunnels. Unless they only switch it on when running on the outside sections.

    Sadly, we were not enlightened further.

  183. GregT……Yes, the trains will have regenerative braking. For the same service level – e.g. the current 24 TPH – the heat reduction is enough to offset the increased energy used by the air conditioning, and possibly a bit more besides. There will be power supply enhancements and ventilation works to enable the bigger service proposed. This carefully worded argument backed with extensive temperature modelling was used to “manage” the comments of the naysayers who argued that air conditioning raised tunnel temperature.

    Out of interest to illustrate what can be done, at a conference last year, a TfL systems engineer reported that the upgraded Victoria line at 36TPH has shorter journey times and uses roughly 15% less energy than was used on the former 27TPH service with 1967 tube stock; for the whole line, not per train! This was delivered via extremely good regenerative brake, raising the maximum regenerative voltage, maintaining the acceleration rate to higher speeds, low loss conductor rail in key sections and joining all the traction sections into one.

  184. In Goole?
    Are Siemens going to open a factory there, then?
    [ I Just checked the company web-site for their UK locations & Goole is not on the list ]
    Oh yes, City AM link here, too.

    Will the same builder get the contract for the subsequent ( B’loo, Central & W&C lines ) builds, or is that a separate process?

  185. Greg Tingey, AP,

    A board member asked exactly that question (without identifying the successful bidder) at the most recent Programmes and Investment Committee meeting.

    The reply Mark Wild gave was that it was their intention that the same bidder supplies the trains for the other lines using options in the contract. The only situation he could foresee this not happening is if either TfL did not have the money for the trains (implicit was they would be forced to buy something cheaper) or that the build quality turned out to be unsatisfactory and TfL did not wish to buy any more.

  186. AP
    Thanks for that – most informative.
    Actual new trains in service, when – 2021(ish)? Or 2023, given that they have to build a new factory, first.

  187. The first orders for Siemens trains for TfL (or London Underground) ever?

    Interesting how Siemens supplied the Thameslink (and previous Desiro) orders from Germany, yet felt the need to open a UK plant, which I imagine helped win this order.

  188. Mikey C,

    Technically correct about Siemens but the sentiment is entirely wrong. Siemens supplied a lot of the cutting edge electrical equipment and power supplies for the Waterloo & City line in the 19th century. Unlike on the City & South London where they got almost everything wrong, Siemens got it right. So their involvement with London Underground and its predecessors goes back far further than you appear to think.

    If wars hadn’t got in the way then Siemens (which I think was technically a British company then) would have probably gone from strength to strength on the Underground .

  189. Siemens also provided the electrical equipment in 1914 for the original Oerlikon units on the Euston-Watford ‘new’ d.c. lines, now part of the Overground.

  190. … which then had a trip to the seaside in post-war AC trials. Err, shall we get back to the Picadilly?

  191. Absolutely right re early electrification – Siemens also provided the electrical equipment for the LBSCR overhead electrification – had it not been supplied from Germany, extension would not have been interrupted by WW1 and would have been much more extensive. I believe much of the electrification masts were in place in August 1914 but then nothing happened – which is why the Southern managed to electrify so quickly to Coulsdon in 1924 (I believe) when there were major complaints about steam suburban service levels. It is interesting to speculate (I’ll put my crayons away) what would have happened then, as the overhead electrification network would have been much bigger so decisions about third rail v overhead could have been different as well. On the otherhand, I believe the actual motor design and current /voltage was something of a deadend and much of the network worn out by the 1920’s – but the absence of supply must have been a factor.

    You are also right about ‘Siemens’ being British but the British Siemens company (what became STC?) in Greenwich concentrated on telegraph / electric cables and had been established earlier than Siemens in Germany. There were a few Siemens, all cousins etc which confuses things. There is a hotel in Milford on Sea which was the British Siemens founders home – I think he was naturalised quite earlier. Wikipedia says much more.

  192. @Simon Bass

    And part of the LBSCR network electrified by Siemens is also now part of the Overground.

  193. Siemens UK plant – will help politically with future lines, total fleet size justifies the capacity, it’s a smaller profile so doesn’t fit their existing assembly line, EU procurement rules made them a level bidder for this line and the move offers protection for different scenarios when next lines come up. And if they are really long term planners potential free trade deals globally in the promised utopia.

    Hounslow – Willesden clarification please. Is this an initial stage of the LO scheme for Hounslow – OOPD – Hendon?

    York Road – too expensive & little benefit, a second halt at the south end of the site. The KingsX developers and Camden are safeguarding reopening Maiden Lane for future rail access when build and traffic justifies the funding (est £ 8-10m).

    Chiswick Park – initially would not the existing joint Picc/District working continue on shared track between Hammersmith and Turnham Green. Change the signs from green to blue at CP. Would digital working require total segregation of lines, desirable anyway for altered platform heights and level boarding. I expected fast HRW working to continue with Uxbridge/EB ‘occupying’ the Wimbledon timetable slots on the Richmond line.

    Barons Court – needs a lick of paint. It’s a beautiful heritage station and a great alighting point for International arrivals. Also Earls Court and Gloucester Road are useful interchanges with their own significant traffic. Please support all three.

  194. One wonders whether the trains assembly might move to Widnes before a sod is turned in Goole – all depends on various competition authorities clearing the Alstom-Siemens merger.

  195. Has any detail been revealed about the design and arrangement of carriages and lengths? It’ll be most interesting to see what the thinking is as this will inform of future intentions for the other lines.

  196. @Ben The TfL announcement states Siemens are now contracted to design it (assume all the NTfL mock ups are guidelines).

    TfL has awarded Siemens Mobility Limited a contract of around £1.5bn to design and build 94 new generation Tube trains

    Also – Each new train will be six metres longer than the existing Piccadilly line trains. They will include walk-through, fully air-conditioned carriages and improved accessibility

  197. Timbeau. Either Siemens have reinvented the laws of geometry for wriggling tube cars with a gangway that are even longer than 1973 tube stock though reverse curves, or the Wikipedia prediction is tosh. I am inclined to the latter view!

  198. timbeau, 100andthirty,

    Not really helped by sometimes people quoting a number of carriages when they meant the equivalent of that number of carriages. Remember that for a long time on Crossrail the talk was of 8-car (initially) , 10-car (normally) and 12-car (if extended) trains but the actual number of carriages was never explicitly stated in the specification for the trains. Similarly for the DLR you sometimes read that the new trains will be 3-car whereas they will be one continuous train equivalent in length to a current 3-car train.

  199. The trains are only 6 metres longer in total than the 1973 stock, so six-car trains may not be out of the question, as they could in principle be very slightly narrower. However, the Wikipedia article no longer makes this “six-car” claim, until and unless someone can justify the claim and revert my edit. (It is probably not yet decided by Siemens or anyone else).

  200. Malcolm……… Even with the trains at 1973 tube stock length, gangways would be extremely challenging and interfere with provision of the single end doors. This was exhaustively explored and rejected for the cancelled 2012 tube stock. Adding an extra metre to each car would require the cars to be narrower, end and centre throws would increase as would stepping distances. And gangways would become no easier.

    PoP also makes a very valid point regarding specifications stating train length, not number of carriages. The length selected is approximately the length of the former 1959 tube stock that used to operate on the line.

  201. 100andthirty: I entirely agree that seven or eight cars are much more likely.

  202. Malcolm @20:28 Could a contractor put in a reliable price if he hasn’t fixed basic parameters like the number of bogies? Presumably Siemens knows what it is proposing?

  203. You could be right, maybe my suggestion that it may not have been decided was a bit off the wall. But if it has been decided internally, it seems that they have not yet made it public.

    What is clear is that some aspects of the trains are not yet designed, since apparently the contract includes some design work. But perhaps that is only the colour of the seat cushions.

  204. Could be they are ‘designing’ the engineering of how to manufacture the concept (to a price).

    I would expect wider doorways nearer the bogies rather than a mind-the-gap centre position.

  205. @130 16 June at 18:55 & at 20:47 – When a lad, I saw the mock-up at Acton Works for a ’73 Stock car and saw by eye it was longer than the ’59 Stock. I asked why and it was explained that instead of the 7-car trains of ’59 Stock then running on the Piccadilly, each car would be that much longer to form 6-car trains resulting in weight saving (bogies etc.) but of the same length as a complete ’59 Stock train, also in order to provide more accommodation inside for Heathrow luggage etc.

    I asked whether it would fit through those very reverse curves you mention because of the overhang and the answer was basically “go away boy, we know what we’re doing”. Well, some time later, a prototype was run through and that morning I had a ‘phone call: “It didn’t succeed!”. Not only had the sides of the prototype got bashed in but the roof where the line dipped in a vertical plane. The tunnel profiles had to be enlarged through those curves as it was too late to adjust the order as the first deliveries were being made at West Ruislip.

    From what I read, the new trains will be 6 metres longer than the present ’73 Stock. That in itself begs another question as to whether all platforms will be able to accommodate them, let alone those reverse curves in tunnel if the new cars are in the slightest bit longer than today’s stock. I’d laugh should I see the tragedy repeat itself.

  206. @Graham Feakins – the shortest platforms would appear to be at Russell Square southbound, @ 105m. If from the scant info available atm, the new trains are ~112m long, it is presumably the aim for (eventually) all doors to be berthed at all platforms (rather than first and/or rear cut out. @NGH further back in the thread has made some practical assumptions about length and bogie-centre. I suppose this is one of those instances were every dm will make a difference.

    Incidentally, from the artists impression video released in the TfL announcement, as the cars lack end doors it would appear they are shorter. Theres also a technical drawing of one used in the background of one of the scenes, but something very weird is going on with the wheelsets position, so it must be an adaptation…

  207. The NTfL feasibility study compared a proposed 10-car train with a conventional 7-car train, saving the weight of 3 bogies. The shorter each car, the less overhang on curves allowing wider trains for additional capacity, and smaller gaps at curved platforms for quicker station stops, requiring fewer planned gap-filler machines. However, additional cars to fill a platform adds to the number of bogies, weight and cost. An IRJ online report dated 28 September 2016 mentioned a planned fleet of up to 3,300 cars for the deep level lines. This suggests 140 trains of 13 cars for the Piccadilly & Bakerloo, 100 trains of 14 cars for the Central line and 10 trains of 8 cars for the Waterloo & City line, total 3,300 cars. A 13-car train would have 14 bogies, the same as the last 7-car Piccadilly line trains. To lengthen such a proposed Piccadilly line train for the Central line with one of the new cars would therefore result in a train shorter than current 1992 tube stock. Perhaps the trains will be specified to fit particular lines, rather than a standard length of car. The Victoria Line Space Train concept of 1995 proposed replacing 8-car conventional trains with possibly 10 to 13 cars, and a single pair of doors in each car. The current proposal looks to have even shorter cars, although the feasibility study shows two pairs of doors per car.

  208. Graham Feakins,

    I can’t let your repeated anecdote be left as it is as it gives a misleading impression. So, to complete the story again, the tube stock was correctly ordered by London Transport. The damage that you describe did happen. On checking the stock built it was found that five were built out-of-gauge. I believe it was only out of dynamic (not static) gauge and was down to a lack of stiffness in various parts. The manufacturer admitted it was their mistake and rather than try and fix it they agreed to pay for the tube tunnels to be eased at various locations so the trains would fit.

    In other words, London Transport did know what they were doing – it was the manufacturers who didn’t.

  209. I have written about this before but nowadays gauging of tunnels is done extremely accurately down to less than a millimetre using lasers.

    The biggest fear when putting a train down a tunnel for the first time (Northern Line Extension, Crossrail) is that some tunnel item such as a bracket has somehow slipped and fouled the gauge. They are confident the trains will fit the tunnel.

    Also, as said a number of years ago, the Piccadilly line when surveyed fouled the gauge for the new proposed trains in over 4000 places. Most of this gauge fouling was accounted for by ‘rusticles’ (iron stalactites) that could be easily be removed and most of the rest by tunnel equipment or cables that would need minor adjustment.

  210. Graham Feakins describes new 1973 Piccadilly stock not fitting through reverse curves as a tragedy. Another poster says the train builder paid for the reverse curves to be enlargened. Surely that is the solution.
    If just a couple of locations limit the size & capacity of trains, then it is location that must be re- engineered. Not rocket science as proved by the stock builder paying for enlarging & not going bust.
    The enlargement will benefit the Piccadilly line for ever.
    I am always amazed that a single tunnel on the old Southern Railway has dictated a change in 20 metre carriage width from 2.82 metres to 2.80 or less across every railway south of the Thames so all can fit the Oxted tunnel, although 95% of stock will never use it.
    The latest trains for Anglia are to be tram like in width compared to MK2 coaches of a few years ago that were a spacious 2.82 metres wide.
    Before anyone says so what. 2.82 m allows big luggage in overhead racks, 5 across seating or properly sized 4 seats in a row.
    Surely money could be found to excavate below gauge tunnels on the pic so the already cramped space does not seem like the post office railway in the next 35 year lasting stock. Every mm counts when the stock is already pint sized.

  211. According to the 2014 feasibility study, the design of the train is predicated on articulation centred on the car couplings, with bogie boxes under the seats each side of the gangway connection, so eliminating any overthrow on reverse curves. The cars in that design are significantly shorter, having 10 longitudinal seats and 2 double doors configured 2-DD-6-DD-2. I imagine this will still require quite a bit of civils work at places like Shepherds Bush eastbound (approach from west) to get the mid-sections of the cars through the s-bend in one piece.

    The study also made mention of future provision for curve gap fillers, platform doors etc., and the visualisations suggest level boarding without platform humps. This implies narrower cars, though maybe there is scope for doing the same ‘balooning’ trick as with the SSL (severely limited by Watford line). There must be considerable detail design challenges remaining for handling the implied 4-5 inches off the underfloor space, also to accommodate the cooling equipment. Much hard work to be done to meet the proposed delivery date.

  212. Nick BXN and PoP…….The planning assumption was that floor height would be at about 700mm above rail. This is similar to “classic” tube stock and 2009 tube stock. Those in between had higher floors. 700mm is still some 180mm above nominal platform height. The other planning assumption was that the platform height would be increased as part of the works to install platform edge doors and gap fillers where required.

    Lowering the floor below 700mm would give the designers considerable problems in packaging all the kit, and accommodating motor bogies under the articulation

  213. For clarification is the NTfL tube for distinct fleets with common parts and style, or interchangeable so could future carriages be swapped out between lines for example?

  214. I think this is intended to be a continuous build for four lines, so interchangeable components. The new Picc trains will again fill platforms to maximise capacity over their 40 year planned life. The Bakerloo follows the Picc, and both previously ran 7-car 1959 tube stock, so presumably both will share identical, interchangeable trains. Now the Central line requires replacement of its current 8-car trains, but the new trains will have more, shorter cars. Adding one of those to a Picc line train will result in a train which does not fill platforms, but adding two will be too long. This suggests that the Central line build will use slightly longer car lengths. However, the Central line project does not commence until 2025. The current Central line trains are to receive new motors and computers, commencing next year, to improve reliability. If this results in a step-change in their performance, there could be financial pressures to extend their life for a further ten years perhaps. Does anyone know why the W&C is expected to take as long to upgrade as the Central line?

  215. Taz….The aim was to try and use common intermediate cars and flex the end cars to deal with any length challenges.

    If bogie centres of 10.75m are used, with 13.5m end cars, then a Piccadilly line train – 10 articulated cars – would be 113m long, and a 12 car would be 134.5m. The former is nearly a metre shorter than a 1972 tube stock and the latter is about 2m longer than the 1992 tube stock. I can postulate all sorts of ways of keeping common bodies but flexing inter car gangway lengths to provide exactly the right length. The right length may or may not be exactly the same as existing trains; LU will have surveyed the lines to determine what can be accommodated.

    All that said, we need to wait to find out what Siemens actually propose

  216. With reference to the comments from various people about 73TS not fitting the tunnels as I understand it the problem was that the cars were too high in the centre when unloaded.

    This rose from two things – the steel used for the solebars was stiffer than specified. It is possible a different grade of steel was used to that in the original design. As a result the vertical curve rolled in the sole bars during manufacture was too big and was not flattened enough by the mas of the cars as they were built.

    This was compounded by the decision not to have tops on the equipment cases that reduced their weight further compounding the problem.

    I can’t remember what was done to solve the problem but one thing that can still be seen was the replacement of domed rivets in the centre of the roofs with countersunk flush rivets. This was only done on the highest cars from memory. I think you can still spot this if you stand on an overbridge and look down at the roofs.

    I don’t think much infrastructure clearance was done – the modified cars were sufficiently within gauge to fit down the Central Line when they went to Hainault.

  217. Yes Malcom,good idea!,it would be the best idea to add Chiswick Park Station to the Piccadilly line . You would need appropriate platform height new points and signalling for Piccadilly line trains to serve Chiswick Park station as at Acton Town station,you would need points as Eastbound platfrom 4 would be on the same track that the Existing Eastbound platform at Chiswick Park would be on which would be on the same platform that Piccadilly line trains from Ealing Broadway and Uxbridge would serve,which would be on the same platform that Eastbound platform at Chiswick Park would be on but for platform 3 at Acton Town would be on Piccadilly line trains from Heathrow Airport would be on as there would be a crossover points from platform 3 to platform 4 to serve the Chiswick Park Eastbound platform.For the Westbound platform at Chiswick Park Station would be on the sametrack that the Westbound platform 1 at Acton Town station would be on which would be the platform that Piccadilly line trains to Ealing Broadway and Uxbridge would serve but for platform 2 would be the platform for Piccadilly line trains to Heathrow Airport to serve, so when Piccadilly line trains depart the westbound Chiswick Park platform trains will serve the Westbound platform 1at Acton Town station for Piccadilly line trains to Ealing Broadway and Uxbridge but for platform 2 would be for trains to Heathrow Airport so after Piccadilly line trains depart westbound Chiswick Park they would serve platform 1 at Acton Town if trains were going to Ealing Broadway and Uxbridge and they would stay on the same track that both westbound Chiswick Park platform and platform 1 at Acton Town for services to Ealing Broadway and Uxbridge would be on but for Heathrow Airport would crossover to platform 2 at Acton Town after they have departed Chiswick Park westbound platform and about to arrive at Acton Town station

  218. The first of the smaller new trains for Glasgow was on show at Innotrans. Geoff Marshall has made a video going round it.

  219. Also you should add Stamford Brook and Ravenscourt Park Stations to the Piccadilly line as it would help balance pressure on the District line

  220. Re SH(LR) and TFBTBO,

    Stadler’s new Glasgow Subway stock – In engineering terms they are miniature trams designed to light rail standards / requirements rather than UK heavy rail / LU specs (e.g. no air brakes for example (Regenerative + track brake + hydraulic ). Use of the train word might not be the most suitable.

  221. Aronjit,

    I am sure neither TfL nor most other people would wish to add yet more stops to the rather slow Piccadilly line service to Heathrow.

    You talk of balancing the pressure on the District line but the S7 stock on the District line is much better at hoovering up passengers than the new Piccadilly line trains will be capable of doing. In the next few years there will be a better, more frequent service on the District line due to them finally getting Four Lines Modernisation re-signalling up and running.

    There is also the practical difficulty that Stamford Brook has no eastbound Piccadilly line platform and it would not be at all easy to add one.

  222. @ngh

    Is there any reason the new deep tube trains couldn’t also be designed/built as such?

  223. DM1……..whilst tube trains are quite petite in size their duty leads to very hard work. NGH highlighted hydraulic brakes. LU has a deep seated dislike of hydraulics on its passenger trains, but did examine the use of electric actuators for the friction brake, but the industry fed back that it wasn’t ready for such innovation.

    On the general point of light rail vs heavy rail, there continues to be blurring between the two. For example, the latest Docklands Light Railway procurement is for a train that is distinctly not light rail.

  224. Re DM1,

    The Glasgow subway is a hobbit sized version of the LU deep tube with 4′ (1219mm) track gauge, smaller diameter tunnels, much lower running speeds, low passenger numbers (total annual system usage is about the same an average Z3 NR station), much lower frequency and low duty cycle.

    Hence it within the sensible envelope for hydraulic brake where as LU and heavy rail aren’t (Heavy road vehicles all use air-brakes too so this just isn’t a rail “think”)

    Worth having a look a Paul Bigland’s photo of the new Glasgow subway stock at Innotrans to gauge it real size here:

    https://paulbigland.zenfolio.com/p192874246/hb7376a76#hb7376a76

  225. Re 130,

    Electric brake actuators…
    The design used by everyone* actually came out of a second year undergrad “team”** design project. One of my predictions was that uptake would initially be for handbrake proposes on all disc cars as that was the most significant weight saving and also solves the disc cooling issues with the no rear mini drum solutions which has so far proven to be the case as the mechanism is relatively slow acting so no ideal for sporty or ESP purposes or road vehicles (ditto WSP on rail).

    *As we were to busy to patent anything the design is used by everyone. The prototype I built is still in the parent garage!

    **the large team only had about 1.75 actual members most of the rest only turned up to the final presentation

  226. @NGH
    ” In engineering terms they are miniature trams designed to light rail standards / requirements ”

    Does that also mean they have (and need) less crash protection as, unlike a heavy-rail train, they are unlikely to come into collision with a 1,000 tonne freight train. (Londoin Underground trains are also unlikely to do so, but they do share tracks with other trains (e.g class 378s) which in turn share tracks with freight trains, and are built accordingly).

    This is one of the compromises facing the designers of tram-train schemes such as the long-delayed one in Yorkshire.

  227. Re Timbeau,

    Partially – A closed system with lower speeds are the key points. Which helps with massive weight saving over several iterative design loops.

    It could be argued that some light rail (tram ) standards aren’t high enough after [the Tramlink incident of] 2 years ago but the Glasgow stock will meet heavy rail standards in many areas

  228. @ngh/timbeau – the distinction between heavy rail/light rail/tramway is pretty artificial on any criterion. In the UK, it used to be based on speed, axle load and rail weight, elsewhere on different criteria none of which provides anything other than a legal change. technically and commercially, it is perfectly possible to sit in the same vehicle running as a tram in the streets with drive on sight, thence onto a “normal” railway – for example, in the narrow gauge extensions to the Basel tram network. The same is still – just true – of the surviving trackage from US interurbans – starting life as street operations and ending up as normal last mile freight operators without breaking step. Same in Germany where the distinction between Strassenbahn, Nebenbahn, Kleinbahn and Vollbahn was marked by a series of artificial steps which had any significance simply in the legislator’s mind. It’s just a spectrum of technical differences.

    [A good and frivolous example of all this arose recently in Switzerland on one narrow gauge “light rail” line where the need arose to replace an existing passively marked level crossing with the fully monty of barriers and lights to satisfy the legal requirements for light rail at higher operating speeds, at substantial cost. This was avoided by designating the railway as a tramway for the space of its passage over the road, and therefore requiring merely a passive sign warning of the tramway crossing. Even the Swiss saw the funny side].

  229. @Graham H: The Berninabahn was built as a tramway. I believe it still is…

  230. I seem to remember, from my days of working on it, that tram Leeds 602 had solenoid operated drum brakes for parking (and for stopping under some circumstances) was it the first?

  231. @GH & SH(LR)

    I recently observed Rhaetian Railways’ Arosabahn seeming to run quite happily through the streets of Chur without any problem. Everybody treated the train, a three car emu pulling four standard carriages, as if it was a tram with no gates, signals, horns or bells and just made space for it, even when going the wrong way around a roundabout.

  232. @SHLR – a good example of neither fish nor fowl – certainly street running in Tirano but anything but a tram when descending from Alp Gruem. The original rolling stock (at least the trailers) were clones of tram stock; the Allegras used now are scaled down mainline branchline stock.

    @James Bunting – The Chur example is better, in the sense that it has street stops- those on the Arosa line on the outskirts of Chur are exactly the same as standard tram stops, and used as such, and as you say, the train mingles with the traffic (or against it). However, I’m not sure that the line conforms to the Swiss version of BOStrab – possibly predates it.

  233. I got the idea that Crossrail delays started with PED installations. Yet new straight platforms with no existing services to maintain would be the easiest of such installations. Anyway, Picc new trains and signalling will come first. Then same for Bakerloo and Central lines. Only then were they looking at PEDs to free staff from front cab duties, with economics of that to be evaluated. Picc interworking with Met to Uxbridge will remain a problem!

  234. I think that PED’s are only intended for actual “tube” – that is below-ground – installations?
    Can anyone confirm/deny that, please?

  235. For air (draft) control yes. The Japanese use low level gates for platform safety ‘outside’ where there is no ceiling to anchor a PED.

  236. No one told the four circular above ground branches off the main metro lines in Singapore, north of the city. No PEDS & NOPO.

    I seem to remember the free circular no person operation trains in central Miami has no PEDs.

    The Lille metro when it opened in 1982 had no PEDs. No one told them for a while. Now they have PEDs.

    I told the then chief Inspecting Officer of railways, Major Oliver, about Lille no PED, NOPE operation.
    He said if it passes the Olver test they won’t be needed. He took out a very very slim pocket book. If it drives away with this stuck in the doors it won’t pass the Olver test.

  237. And Line D in Lyon, which I believe is a unique heavy-duty development of the VAL technology, is (almost) totally underground with a lightbeam safety system instead of PEDs.

  238. May I ask, is it true that Stage 2 (Resignalling) has been “paused indefinitely” (cough, cancelled) due to funding issues

    I know that between the new rolling stock (paid for by the sale and leaseback of the Crossrail trainsets) there will be a push with existing signalling to 27tph on the Piccadilly, and with the new walkthrough trains there can be expected between 20%-25% increase in capacity – which is not to be sneezed at, but surely without resignalling the value of the new trains will be lower?

    Is there any update or confirmation on this?

    I got this from a recent thread on RailForums:
    https://www.railforums.co.uk/threads/camden-town-upgrade-piccadilly-line-resignalling-axed-for-crossrail.176248/

    thanks

  239. @ Philip – I believe the official line is that TfL do not have funding for the Picc Line resignalling. They are hoping to secure funding as part of the government’s spending review process which starts this Summer. Being the cynic I am I’m very sceptical that they will get additional specific funding for any large scheme.

    Various oral and written statements in recent weeks have got Crossrail 2, Picc Line resignalling and Bakerloo Line extension all on the Spending Review “shopping list”. While that’s in line with what the Mayor wants done I don’t see London getting any extra funding. Worse I think the Mayor will be told in no uncertain terms that until he starts putting fares up, probably faster than inflation, then he won’t even be listened to. That’s my sense of the “behind closed doors” political battle that will go on. We are a long way from the days of Ken L knocking on the door of a Labour Chancellor or Boris knocking on George Osborne’s door and getting a reasonably benevolent welcome (even if through gritted teeth).

  240. Also tfl should make all eastbound Piccadilly line trains to terminate at Cockfosters Station and build even more platforms in order to improve frequency and capacity for Piccadilly line trains coming in and out of Cockfosters Station as currently during morning and evening peak hours every day Piccadilly line trains terminate short at Arnos Grove instead of Cockfosters Station. The outer platform at Arnos Grove Station gets packed with passengers trying to get to Southgate, Oakwood and Cockfosters Stations. It is only 3 more Stations until Cockfosters Station: Southgate,Oakwood and Cockfosters Stations. Well,the fact is that Piccadilly line drivers are super lazy that they dont care whether there is only 3 more Stations until Cockfosters Station ( the northern end of the Piccadilly line).I would also suggest that either we tear out the centre tracks and platforms at Arnos Grove Station and construct a longer island platforms linking eastbound platform 1 and westbound platform 2 or we can abandon Arnos Grove Station as we would tear out the platforms and station building as Arnos Grove Station will no longer exist and Piccadilly line trains will skip Arnos Grove Station.

  241. @Aronjit – You have no idea what you are talking about, and with those sorts of comments, you are not going to make any friends here.

    Firstly, I find you statement about Picc drivers being lazy to be stupid and objectionable. The drivers don’t choose where they drive to, they go where they are told. Blame TfL management if you want, but the drivers have no say in this.

    Secondly, the latest annual entry/exit figures for these stations (i.e. starting or ending at the station, not interchanging) are –

    Cockfosters – 1.93m
    Southgate – 5.59m
    Oakwood – 2.89m
    Arnos Grove – 4.61m

    So Arnos Grove is twice as busy as Cockfosters, and I reiterate, this does not include changing trains to carry on to Cockfosters. These figures were found after 5 mins of searching on the Internet, so can be easily found. Perhaps next time, a little research rather than ranting would be a good idea.

    Lastly, the trains probably turn back short because there is enough trains to run all the services to Cockfosters in the peak periods. You cannot increase capacity unless you have spare trains, and new trains take the best part of 10 years to obtain.

  242. @ARONJIT

    In addition, you should have a look at Brixton and Walthamstow Central on the Victoria Line, and Stratford on the Jubliee. Stratford has reduced the number of platforms being used (to two) and now they all operate trains using “step back”.

    By using two platforms, the arriving driver gets out of their cab and signals that the outbound train can start, and then walks the to the front of the train to take a later train. By separating the driver and train, trains can leave from the terminal like DLR trains do: the timing isn’t dependant on how fast a driver can walk.

  243. I recall the original PPP press release near 20 years back mentioned only one layout improvement for LU, an additional platform at Oakwood, presumably on the depot shunting neck to relieve pressure on Cockfosters. This has long been forgotten, probably with the improved throughput achieved with modern signalling.

  244. Arnos – Cockfosters was at one point the most frequent unplanned short trip on the system. One wonders if this is still the case.

    Will the resignalling of the SSR provide much in the way of useful spares for the Pic?

  245. @Aronjit – (RE: Arnos Grove Station turnaround): I feel I shoiuld point out that when I was living in Southgate, and commuting to Covent Garden, the only time that Cockfosters was useful to me was when the leaves damaged the tyres on the trains (see https://www.londonreconnections.com/2016/wear-and-lathing-problem-with-the-piccadilly-lines-trains/ ) and the only way to get a seat was to travel out to CockFosters, and be one of the first on a Westbound train…

    It is true howver, that TFL management do seem to have got this slightly wrong with the turnaround at AN during peak periods, it’s quite annoying having to split the journey at AN waiting for a train that will go all the way to Cockforsers – especially in the winter…

    As this is undoubtably due to capacity problems at CF, maybe they should look at increasing the turnaround capacity at Cockfosters?

    [Please spell out full station names – our standard is commentary that everyone can understand, not just enthusiasts. LBM]

  246. @Walthamstow Writer (RE: Resignalling Piccadilly Line “deferrment”)

    This is crazy… new trains paid for by a sale and leaseback of the rolling stock on the biggest and highest visibiltiy project in London for decades, with 50-60 year old signalling and infrastructure…

    So who do I complain to – it looks like the current Mayor is simply not interested apart from soundbites … and Central Government have their hands full finding a new person [Off topic snipped. LBM]

  247. In a perfect world the Picc would have seen the promised 20% uplift under PPP five years back. In current times the idea of new longer trains providing 27tph under current signalling provides 33% uplift, which is a good start. A further uplift with auto operation when funds are available will help cope as demand continues to increase.

  248. I know Philip,but you can blame tfl management but the Piccadilly line drivers have no say in this ,they do not choose where they drive to.They go where tfl management tells them to go. [Off topic material snipped. Malcolm]

  249. Someone asked why the Waterloo & City line will take as long as the entire Central line according to the above chart. I now understand that these are only outline plans, and the most economic way of converting the short W&C line is to treat it as part of the Central line works. So new cable runs, etc will be done at the same time. The whole conversion may therefore take as long on both lines. When detailed planning is finalised it may work out differently.
    Bakerloo conversion seems to have been since dropped from this programme, its lower demand being served by patched current trains. Instead, it has been grouped with the Lewisham extension plans which could justify a new fleet and common signalling throughout the Bakerloo around the same time.

  250. The current LU staff news, On the Move June 2019, has an item marking completion of the five-year repair programme on Bakerloo trains, and notes that they should now continue in service until 2035, their forecast replacement date! This leaves a big hole in the plan to roll out a new fleet on four lines (see chart above) and indicates a new target for Lewisham extension.

  251. Are they still planning to start the re-signalling on the Piccadilly this year?

    Or was that shelved/cancelled along with the Northern Line new trains?

  252. I note that Siemens new trains for the Picc … have a feature that was unfashionablesupposedly-not-good-engineering until very recently. But, then, some of the new ( Flirt ) units for Anglia also have this feature, that seems to have been raised from the dead.
    Articulation.
    After 1941, or more certainly after 1948/51, in this country, articulation of rolling-stock was, effectively, banned for over 60 years, for all sorts of “reasons” that always appeared spurious to me.
    I do hope the shade of Sir Nigel approves!

  253. Greg: The word “articulation” is used with different meanings for different forms of transport. For trains alone (not trams) it has often meant bogies shared between carriages or freight bodies (so-called Jacobs bogies). These may not have been chosen as often as you would like, but there have certainly been examples later than 1951.

    When a tram is described as “articulated”, that normally means that passengers can walk between the rigid body sections. This meaning is spreading to trains, particularly if the connection is full-width. When passengers refer to articulated tube trains, they are generally not talking about the wheels at all, but about the passenger experience of one long bendy corridor.

    Of course, in the oft-lamented Gresley quad-arts, passengers could not even walk from one compartment to another! But they did pack in a lot of seats.

  254. Malcolm: some passengers may refer to full-width walk-through trains as articulated, but that’s not what is means technically, nor what Siemens mean. Technically it means that individual carriages are not self supporting, requiring a link to their neighbour(s). Typically in railway usage that’s by means of a Jacobs bogie, but often in trams (and in buses and lorries) it’s a different sort of suspension.

    Siemens’ press release says the trains will have “an innovative articulated design that reduces the number of bogies required per full length train”, so clearly they do in fact mean articulation as generally understood.

  255. @Greg T

    Eurostar class 373s are (or should that now be “were”) largely on articulated bogies.

  256. Articulation in the UK of various configurations in addition to the Eurostar trains:
    Anglia Electric class 745 Inter-City and Stansted Express, and Bi-mode class 755
    Merseytravel class 777
    Glasgow Subway new trains
    All modern tramcars
    Newcastle railcars
    DLR existing trains (but not the new CAF ones)
    Stadler vehicles on order for TfW

    This amounts to a sizeable number of vehicles.

    As far as I’m aware Siemens have yet to declare the exact nature of the articulation on the Inspiro London trains

  257. 130
    Indeed, the number of articulated units is increasing – now.
    I remember when the “old” Eurostar units first appeared & commented that they were articuated … & how all sorts of the usual “It can’t possibly work here” ( Off Eurostar routes ) spurious tales were told, identical to all the tales as to why it was found useless after about 1950-55. It would seem that it was, indeed, fashion & that not one word of why articulation was “NOT a good idea here” was actually true.
    Curious, that.

  258. GregT…..Articulation can’t possibly work here is one of those “not invented here” things that people, industries and nations get themselves into. I guess another is that single wheelset articulated bogies with individually rotating wheels can’t work is another one: https://www.talgo.com/technological-principles#:~:text=Talgo%20is%20the%20only%20manufacturer,can%20rotate%20at%20different%20speeds.&text=For%20the%20last%20200%20years,speed%20on%20a%20straight%20track.
    I don’t believe, though, that Talgo technology is used on the Piccadilly line!

  259. Rail Engineer has an article about the new trains. The restriction to odd numbers of carriages will be a nuisance if there’s ever a roll-on order for the Jubilee; ten would match the existing platform doors closely with only the pairs needing attention, but Central-style short 11 trains are quite different.

Comments are closed.