No new signalling yet – and no 4-minute Circle Line service either.


In my blog of 11 February I made reference to the latest proposals for the various deep tube line upgrades and, in particular, the dates by which it was intended to achieve the capacity upgrades and the proposed levels of automation. The information was part of a large body of material made available in connection with the GLA’s London 2050 infrastructure plan — a comparatively rare example of attempting to think long term as well as co-ordinating all the infrastructure requirements, of which transport is just one (though a very important one as transport infrastructure has a very long life and can take a very long time to put in place).

Well, if a week is a long time in politics, it is also a very long time in the world of public transport communications as this mass of fresh data was soon overtaken by one of the oddest press releases I have seen for a very long time (and which may be found HERE), issued on 24 March.

A Press Release for London

This is a perplexing document whose purpose was apparently to ‘confirm’ things that we already knew about and for which we had been patiently waiting for some time. £5.54 billion was being spent, it trumpeted, but TfL’s capacity to spend unimaginably large amounts of money is I think doubted by no-one and what I believe we all wanted to know is when the things we had already been promised were actually going to appear.

There were other messages in this seemingly random announcement, such as a list of things that had already happened a while ago, and an update on the improved service performance, and so on. This provided the clue that it was a press release made in advance of a significant paper being presented to the TfL Board a day or so later and presumably the Press Office wanted to get the party line out first.

The real message was that the Sub-Surface Line upgrade was going to cost a lot more than had been thought and that it was going to take much longer to deliver and slip into the 2020s. This might also have implications for the delivery deep tube upgrade, which is where I came in.

Deep Tube Upgrade

I will return to the Sub Surface Line (SSL) debacle shortly, but I thought it helpful to get an update about the deep tube programme out of the way first. This was not mentioned in the press release and logically one might expect it to have been delayed by the Sub-Surface upgrade issues. Indeed it has, quite seriously, but not, it would seem irrecoverably. Studiously ignored by the press at the time was the slippage in TfL issuing the Invitations to Tender (ITT) for the deep tube rolling stock. These were to have been issued in late February this year, but in the run up to that date it was put back to the very end of December. This curiously precise date is because a commitment was given to DfT that tenders would be issued in 2015 and this is the latest date that is possible without breaking that commitment. Shortlisted bidders who had teams standing by were not particularly pleased about this.

The delay was partly attributed to the SSL problems. Gareth Powell, the responsible LU director, stated later to the railway press: ‘there will be some minor impact to the timings of other programmes’, and ‘milestones for the NfTL have moved by approximately six months, which will have a minimal impact to our customers’. However it is now apparent that London Underground is embarking on an ‘innovative procurement’ approach that will involve the shortlisted bidders during 2015 in order to include their feedback in the final ITT. Innovative procurement is intended to use supplier expertise to reduce cost and risk before designs and tenders are formalized; that way you don’t compel specialists to produce something costly and unneeded just because the contract says so, a serious risk when specialists in the procurement team are less evident than they once were, and occasionally non-existent. It’s a good trick if you can do it, but it can only go so far as manufacturers are obviously mistrustful that sharing their expertise could result in a competitor walking off with the contract (as well as competitor information that can be used elsewhere).

It has also become apparent that the condition of the existing rolling stock was not taken into account in the initial plans. The Bakerloo stock is about 42 years old and is in a much worse state than thought; there is little option but to keep it in service for a further decade or so. The work is major, involving replacement of the floor and structural body parts and substantial repairs elsewhere. The work is likely to be done in-house and £70m has been put aside.

Meanwhile the Central Line stock is at half life and not performing particularly well. The decision has been made to give these trains a major overhaul (again in-house) and to focus on improving reliability and restoring appearance. £123m has been set aside for this, with work just about to start. This, incidentally, is in addition to further planned work during 2018-2020 which includes door overhaul, repairs to floor corrosion, in-car alterations to meet RVAR requirements and replacing the traction package with a modern ac drive system. Nothing trivial then! Although these trains are not really very good, there is no option but to keep them running and the huge amount of investment being put in (which I would argue isn’t true ‘investment’ at all – it is really exceptional maintenance) means that they should actually be good for a couple of decades.

One outcome of all this has been the decision to prioritize stock replacement by NTfL trains on the Bakerloo Line first, even though the busier Central Line has the better business case.

The latest information is:

Piccadilly Line – New signalling complete 2023, train delivery complete 2026, service upgrade to 33-36 peak trains/hour along busiest section 2026. GoA 4 capable 2028.

Bakerloo Line – New signalling complete 2025, train delivery complete 2028, service level at 27 peak trains/hour along busiest section 2028. No GoA 4 (trains will be automatic to GoA 2).

Waterloo & City – New signalling complete 2032, train delivery complete 2032, service upgrade to 25-27 peak trains/hour 2032. GoA 4 capable 2032.

Central Line – New signalling complete 2030, train delivery complete 2033, service upgrade to 33-36 peak trains/hour along busiest section 2033. GoA 4 capable 2035.

The intention to award contracts during 2016 is now unachievable but if the rest of the programme is compressed then it might still be possible to award the contract in 2017 and achieve final detailed designs by the end of 2019. One feels there was a lot of slack in the original timescale (consider that the innovative 1938 stock was conceived and deliveries begun in under four years, and that was without modern technology!).

Sub Surface Lines

Turning now to the Sub Surface Lines (SSL), it is not a happy tale. The abandonment of two attempts to resignal the lines, and the so-far drawn out and incomplete negotiations for the third attempt, means we still do not know for sure what we are getting and when. The press release revealed that the never-achievable 2018 completion date has been abandoned for a prospective completion on 2021-22 and that the estimated cost has gone up so as to be ‘more realistic’.

It is a good moment to look at how badly all this has panned out. When the PPPs got under way around 2003 the intention for the SSL to have all the upgrades complete by 2015, and completion of rolling stock deliveries in 2016 (on the District). In other words, by now we should already have had all the new signalling, and most of the consequential service improvements.

The original signalling element of the Metronet upgrade was to cost about £550m, which after Metronet’s collapse LU felt was too expensive, paying £95m to disengage from the contract and another £80m for preliminary works that were delivered and which may or may not be needed. We’ll never know if Westinghouse would have produced a working system but they were planning to use a variant on that used (successfully) on the Victoria Line and known as DTG-R (distance-to-go–radio). This was attractive in that trains’ positions would be reported by a radio system avoiding a huge amount of lineside cabling which was both expensive and vulnerable to damage and theft (there was a spate of cable thefts at that time). Westinghouse became Invensys during the period when the Victoria Line was being commissioned.

After getting rid of Westinghouse, LU looked around the world for something that already had a good track record and after a further round of tendering, a new contract was awarded to Bombardier Signalling in June 2011 for the deployment of its Cityflo 650 system, recently installed successfully in Madrid. The idea was to complete the upgrade by 2018. The price was £354m, and regarded as very good value, but of course the enabling works for the new trains had largely been done and £95m compensation paid out to the other lot, and probably a proportion of the £80m would prove redundant, so we are probably comparing £550m with about £470m, which still sounds like better value except that the contract proved undeliverable and was cancelled.

This time round, the signalling supply industry has only fielded one prequalified bidder and the rumour mill suggests the resignalling is likely to take longer than previously projected and cost nearer £500m. This may be ‘more realistic’, but allowing for the compensation paid, and the 6-7 year delay in delivery, it is a moot point whether we are any better off than the original Westinghouse contract, except the public has seen massive delays to the delivery of better services. The single bidder this time is Thales. This was the company (previously Alcatel) who had installed the Jubilee Line signalling system using their SelTrac system which, on the Jubilee Line, required very extensive railway closures – considered quite unacceptable for new installations. These will not be tolerated this time so a variation on the original theme is likely in a number of areas. The completion date now for the delivery of upgrade train frequencies is the timetable change planned for December 2022.

The press release had some bizarre stuff in it.  “Once completed most Circle line customers will see a train up to every 4 minutes instead of 10” Really? So how does that work, exactly? This is self-evidently impossible so I thought I would just check the latest service proposals, which of course showed the press release incorrect . It may be helpful to set out the current position below (but these proposals are subject to signing the contract with Thales soon and may change anyway):

Metropolitan Line

  • 28 tph north of Baker Street (12 to Uxbridge, 10 to Watford and 6 to Amersham/Chesham). (About 23 tph now).
  • 16 tph between Baker Street and Aldgate (14 now).

Hammersmith & City Line

  • 8 tph Hammersmith to Barking (6 tph now)

Circle Line

  • 8 tph Edgware Road to Hammersmith via Tower Hill (6 tph now)

District Line

  • 24 tph over main section Barking to Earls Court (with Circle Line and H&C Line this gives 32 tph Gloucester Road to Tower Hill and Aldgate East to Barking).
  • 8 tph Edgware Road to Wimbledon (and with Central London service, 16 tph Earls Court to Wimbledon).
  • 4 tph High Street Kensington to Kensington Olympia.

Several interesting points here.

  1. The specification for the Watford service states 6 tph will serve Watford Junction and the other four will operate from Croxley, from which I conclude they operate empty to and from a reversing point at or near the existing Watford Metropolitan station (unless the campaign to keep it open gets anywhere).
  2. We still have 32 tph forecast to operate between Baker Street Junction and Aldgate Junction. Obviously new signalling with headway management systems in operation can attempt to manage trains through the flat junctions at Aldgate (of which there are three), Baker Street, Edgware Road and Cromwell Road but signalling cannot achieve the impossible and in my view the weak point here is the junction at Baker Street where the sharp track curvature presently has a 15mph speed limit imposed and northbound Metropolitan trains take 50 seconds or thereabouts to clear. Near perfect train management is crucial and we should remember that the Victoria, Central and Jubilee Lines (the only ones so far to have frequencies like this) do not have flat junctions of this kind and their terminal crossings allow much higher speeds.
  3. A 32 tph service to will operate to Barking, reducing to 16 tph to Upminster. It is only feasible to reverse a proportion of this (probably half) in the bay road so smart work detraining in Platform 2 will be necessary, and something done with the recent and very aggressive 10mph speed restriction into the sidings east of the station, which appears to be track geometry related.
  4. The service pattern through Earls Court appears to call for 20 tph from Ealing/Olympia via Platform 1, and 16 from the Wimbledon branch merging in a way where 8 of the 16 from Wimbledon (for High Street) must cross the path of 16 of the 20 on the other routes that proceed to Central London. I return to my theme (blog item 1 December 2012) that during the Earls Court reconstruction, which has just started, the once in a generation opportunity wasn’t taken to alter the flyover to avoid this conflict. LU thought the business case wasn’t there whilst my view was that the developer should pay as part of the Section 106 settlement. Modern signalling should ease the pain of this and of course the inevitable pathing gaps will be times to coincide with Circles coming and going at Gloucester Road.

So that seems to be the latest on this front. There is a discourse to be made at some point about modern procurement techniques and, in particular, the consolidation of the signalling supply industry to that of a few global suppliers producing equipment for the mass market (particularly new systems in the case of Metros) rather than equipment designed specifically for very old and complex systems like LU, as would have been done once. I think that might be an interesting read and I suspect procurement rules coupled with limited choices and the perceived need to tender rather than develop a dedicated product is not helping a difficult situation.

A combination of politics and poor procurement have left frustrated passengers without what they had been promised – interesting nobody has tried to blame Boris for all this! Well not yet anyway.

More anon.

About machorne

I have always lived in London and taken a great interest in its history and ongoing development. This extended into the history of its transport services, about which I have written a number of books - I have spent most of my working life working in the industry and observing changes from within, mostly to the good, but not always so. I continue to write, and have a website with half finished stuff in it so that it is at least available, if not complete. Several new books are in hand. I have many 'works in progress' and some of these can be found on my website; the we address is
This entry was posted in London Underground and tagged , , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s