The Kent Route Study (Part 1): London Bridge Metro Services


On 14 March 2017 Network Rail finally published its draft Kent Area Route Study Report. On the same day, quite deliberately, the DfT published its consultation document for the requirements of the next South East rail franchise entitled South Eastern Rail Franchise Public Consultation. Here we take a look at the Route Study report in general and more specifically at the suburban services through London Bridge operated as part of the South East rail franchise.

Not a Route Utilisation Study

This Route Study is quite different in its objectives from the Route Utilisation Studies (RUS) of old. The old RUS reports were largely aimed at people familiar with both railway terminology and practice. They were generally an exercise in looking at what could be done on a route by route basis to increase capacity, then deciding whether it was worth doing so or not. The conclusion of the early RUS reports was substantially decided by Network Rail, with input from others. It was, more or less, Network Rail who got to decide which schemes were worth proceeding with. In later RUS reports, the final decision rested with the DfT but the principle was the same.

This route study draws on various other documents, such as the those published as part of the Long Term Planning Process (such as network studies and market studies), to assess what may be desired, what can be done to achieve it, how much it might cost and what the expected BCR would be. The current practice is not to make a recommendation as to whether or not to proceed with a particular project, but instead it is suggested as a candidate for inclusion into a list of projects to be implemented in future.

The friendly format route study

What is noticeable is that the new format is designed to be much friendlier for the average railway passenger to read. That is, the people for whom the passenger railway is supposedly run. Terms like Train Operating Company (TOC) and digital railway are explained, as is a diagram giving the characteristics of different classes of train. There is also a diagram explaining how the lines are generally referred to – far from trivial when one bears in mind that the name used in passenger announcements may be quite different to that used internally by Network Rail.

Metro area with route names provided

Not so much moving the goalposts as enlarging them

One thing that is becoming more and more apparent in Network Rail’s analysis is how maximum acceptable standing capacity frequently seems to be treated as a target to get near to but not to exceed. Various categories of standing are used:

Seats available – up to 75% seats taken
Seats busy – 75-85% seats taken
Seats full – 85-100% seats taken
Standing – 0-60% standing space used
Congested standing – 60-100% standing space used
Overcapacity – allowable standing space exceeded

Passenger numbers are rising all the time, so “congested standing” risks becoming “overcapacity” in the next few years. It seems that the objective is simply to avoid overcapacity. There appears to be no sophistication, as there is with TfL metrics, to take into account the unpleasantness of the journey caused by lack of seating space or even space to stand in comfort. We seem to have entered an era in which the level of crowding on some routes may be unacceptable to passengers, but if it is within the limits of acceptability laid down by the DfT then nothing will be done about it.

This apparent acceptance of severe crowding – but not overcapacity – is in marked contrast to the draft franchise proposals where the desire to increase space for passengers is repeatedly emphasised. Somewhat ominously there is no specific mention of those passengers being seated, so perhaps the objective is simply to enable commuters simply to stand in greater comfort.

When it comes to measuring how busy trains are, more impressively, the “first instance of standing” on inward high peak trains is recorded. Hopefully, this is pragmatically assessed as there are people who will stand even if every single seat is empty. Nevertheless, this is a good metric to use as an alternative way of measuring “busy-ness”. After all, it is rather annoying to discover that the fact that you had to stand isn’t taken into consideration because, unbeknown to you, it so happened that there were other carriages that had seats available. It is also a metric that favours more trains rather than longer ones and the former is probably what most passengers ideally want.

Keeping capacity targets to a bare minimum

As always seems to be the case with their planning documents, Network Rail seems to be oblivious to the notions of either suppressed demand or deliberately providing more capacity than immediately needed to stimulate housing growth.

The apparent decision to ignore suppressed demand seems to indicate that if small improvements do occur, then we are going to get stories of selective train lengthening that fills up within weeks. This is because people will change their travel habits when they realise the train they really want to catch is now no longer full – and so it fills up again.

An example of suppressed demand

The Hayes line is reported as one of the most crowded. Indeed beyond Ladywell on an inward journey is one of only two stretches of line reported as at overcapacity. There are proposals to address this overcapacity by putting extra carriages on selective trains as and when required. Yet this is a route for which there are plenty of alternative stations available on different lines nearby.

Levels of overcrowding

In the case of the Hayes line, the plans to relieve the overcapacity do not seem to take into account passengers who actually use Catford station but would rather use Catford Bridge, which is almost adjacent to it. The reason they don’t use Catford Bridge (which is the station before Ladywell) is often because the Hayes line trains are either very crowded or full. In a similar way, passengers who know they are unable to get on at Ladywell may well decide that it is more pragmatic to walk to nearby Lewisham where there is a more frequent service and a choice of trains.

A failure to stimulate house building

More worrying than the failure to recognise suppressed demand is the failure to recognise the need to stimulate house building by creating additional passenger capacity in advance of development. Despite TfL and the Mayor emphasising its importance, there is no attempt to get ahead of the curve and stimulate house building in south-east London where opportunities exist. One of the main reasons given by TfL and the Mayor for wanting control over the metro service currently run by Southeastern was because they wanted to increase passenger capacity to attract house building. TfL claimed that Network Rail and the DfT would fail to take this into consideration. On the evidence of this, that claim was entirely justified.

This failure to stimulate housing is even more disappointing when one considers that the Network Rail press release is headlined “More housing and more train passengers“. It turns out the “more housing” part was referring specifically to supporting the projected housing growth in Ebbsfleet Garden City.

After 2024 then what?

The study considers two dates in the future and services that need to be provided by then. Of course, this does not mean we have to wait for those dates to see the plans implemented. The first date is 2024 which, presumably, is aligned with expected date of the end of the next South East rail franchise. The second date is 2044 and there appears to be no obvious reason for choosing it.

Breaking down the report

Due to the relative ease with which the report can be read, we at LR Towers feel that that document does not need a full summary to understand the salient points. Nevertheless, taking a geographic approach to its contents may help things be seen in a different perspective.

London Bridge Metro Services

Our eyes, almost inevitably, first turned to looking at what was proposed for the current Southeastern metro services into London Bridge. For clarity, the report spells out the area covered:

5.5.1. The London Bridge Metro area covers the services that operate from London Charing Cross and London Cannon Street, through London Bridge to Gillingham and Dartford via the three lines, to Hayes, to Sevenoaks via the Main Lines and the Bromley North branch

Much of what is proposed for these services was predictable. Some detail is provided that hitherto had not been available. The big surprise is, however, that other than lengthening existing trains there seems to be no strategic long-term plan.

For current Southeastern metro services through London Bridge, the only real clue as to what is proposed for 2024 onwards is Table 5.12, where, under longer trains and more trains is the comment “Further development work required”. This clearly means further development work is required to find a way of providing the capacity, rather than to determine the requirement because the projected requirement is given in a table.

In fact, more ominously, buried deep down in the text is paragraph 3.8.3 which states:

Beyond 2024 additional services would be needed on all the metro routes (except via Abbey Wood) to meet projected demand. The capacity work has confirmed that although paths are available on sections of each route, the network as a whole cannot accommodate the additional services.

No Govia via Greenwich in draft

The study, perhaps worryingly, is done on the basis of a 2018 timetable that does not include 2tph Thameslink trains going via Greenwich as proposed by GTR. Whether the failure to base the study on a timetable that takes this into account is because it is by no means certain (or even likely) these will be implemented, or whether this is simply because they haven’t got around to modifying the analysis is not clear. Network Rail promise that:

The output from the GTR consultation will be considered as part of the final Route Study.

What “considered” means is not defined.

Tell us what we don’t know

What the Route Study does when it comes to the London Bridge Metro services is more or less tell us what we already know. Significantly, it does not mention power supply problems as an issue that would impede introducing longer trains – something that definitely used to be the case. If fact (in 2.2.14) it makes it clear that this is not a problem for running selective 12-car trains. It does mention, as we already knew, that Networker trains in use on the current Southeastern metro services cannot be made compatible with Selective Door Operation (SDO) except at prohibitive cost. It does not discuss the alternative of replacement trains which is not entirely surprising as that would be very expensive and the Networker trains still have many years of life left in them.

The study reiterates the fundamental problem of providing 12-car platforms at Woolwich Dockyard. It doesn’t explain that problem though, which is that there is a tunnel at both ends of the platforms.

The study makes the point that the line on which Woolwich Dockyard is situated (the North Kent line) doesn’t really need 12-car trains. This is largely due to Crossrail at Abbey Wood providing some relief at the end of 2018. Nevertheless, because of the way trains may well traverse all three Dartford routes (as well the Hayes and Orpington lines) in the course of a day, it wouldn’t really be practical not to do the necessary work at Woolwich Dockyard unless modern SDO-compliant stock were provided.

As well as Woolwich Dockyard, the study reiterates that 12-car Networker stock cannot use platforms 4, 5 or 6 at Charing Cross.

And they do tell us something we don’t know

What is new, or at least new in a published document, is Network Rail’s definitive list of what is required for 12-car operation. On the station platform side, there is, of course, the aforementioned Woolwich Dockyard problem. Rather than be considered an insurmountable obstacle, as it has been in the past, Network Rail have simply put a guideline price of £20-50m on all of the work necessary – including Woolwich Dockyard platform lengthening. Network Rail sources suggest that it won’t be possible to open up sufficient length of tunnel(s) or eliminate completely for less than £10m. It will likely be more than that, but probably not considerably more. To put this in context, a single 12-car train costs around £15-18m.

The only other platform that still needs lengthening is Waterloo East platform B. This will probably be awkward as the station is on a high viaduct. Quite why this was originally omitted is a mystery, but it could be that mitigation measures which were acceptable in the twentieth century are no longer acceptable in the twenty-first.

Most platforms were extended in the aborted attempt to increase the service level to 12-cars in the early 1990s. Those that were cancelled when the project was abandoned due to the recession were completed a few years ago. Work was also carried out on those that were compliant in 1990 but, it seems, were no longer compliant with today’s standards. Most noticeable was New Cross platform B which had further work done on it a few years ago. So, possibly, there is something about Waterloo East platform B which meant is was acceptably long for 12-car trains in the past but isn’t now.

Also necessary, or at least highly desirable, for lengthening trains to 12-cars are a few signalling modifications. These include signalling changes to take into account platform alterations (Waterloo East platform B and one platform at Grove Park for which the signal gantry is already in place). One set of Driver Only Operation (DOO) monitors at Grove Park also needs moving. Also needed are adjustments to multiple track circuits to allow 12-car trains to stand on the Up Crayford Loop Line. Not mentioned in the report is the domino effect taking place here. When one track circuit needs altering that affects the next track circuit up the line, and so that has to be altered as well. This process has to continue until you reach one that can be altered without affecting the next one. As is often the case, what initially appears to be a simple change turns out not to be.

The often forgotten (or ignored) depot issue

Finally, on the subject of 12-car trains, the report mentions the need to reconfigure Slade Green depot for 12-car capacity. No issues are reported in doing so, which makes the residents of LR Towers at least a little bit suspicious. This is because it has previously been reported that a previous Southeastern management attempt to get the DfT to agree to the cost of running 12-car trains on a permanent post-London-Bridge-works basis faltered precisely because of the cost of providing the depot space. Unhelpfully the report states:

5.5.12. Stabling costs have been excluded from the business case in line with other Route Studies

This seems inexplicable. One only has to look at TfL proposal for improving Tube services to realise this can be a critical factor that can turn a favourable proposal into an unaffordable one.

Although the study does not really spell out the issues concerning depots, it does at least mention them. It also lists both depots and stabling locations. The TfL business plan for taking over the metro services from a future South East rail franchise never mentioned depots, yet this would have been crucial as the TfL plan seemed to indicate that a much greater number of longer trains would be introduced.

In the end, it is necessary to look at paragraph 4.8 in the draft franchise specification to find something brutally honest about this issue:

Depots are operating at, or near capacity, which means that new ones may need to be built to enable more, or longer, trains to be introduced on the network.

Bromley North Branch

We have covered the Bromley North branch previously and not much has changed since then. According to the report, there is forecast growth to the extent that it is expected that 10 cars will be needed to run the line in 2024. Before anyone gets too excited, it appears that this is the number of cars per hour, so the 2-car train shuttling between Bromley North and Grove Park every 20 minutes already account for 6 of these 10 cars.

The report mentions an aspiration to provide a service every 15 minutes on this branch. This would be much more beneficial than would first appear since, off-peak, trains would link into the cycle of trains every 15 minutes at Grove Park. Apparently, the existing franchisee, Southeastern, has proposed maintaining one train but having a driver at each end of the train to achieve the necessary fast turnaround time. If this happened then there would be some irony of the shortest trains on the metro network being the ones with the highest level of crewing. Indeed history would almost be repeating itself, as the Bromley North branch was the final service in the former BR South Eastern Division (suburban services area) to retain two crew members on the train. In the end it was the last route to go to Driver Only Operation (DOO) as the Addiscombe branch was worked with guards until it closed.

One can be cynical and suggest that Network Rail would like the idea of two drivers on trains on the branch, as this solution which would not require any action on their part. If a 2-car train running every 15 minutes were adequate (despite a projection that 10 cars, not eight would be required by 2024) then the two driver solution would, at relatively low cost, resolve the main unsatisfactory feature of the branch and avoid the need for some of the more extravagant solutions suggested.

Forecast growth of 100% for the Bromley North branch

Rather amusingly, the Bromley North branch is indicated in the report as the line with the greatest expected growth by 2044 when measured in carriages required. It shows the danger of small values leading to distorted statistics. Replacing a 2-car shuttle with a 4-car shuttle would indeed amount to 100% increase in capacity – the greatest predicted growth of any service in the entire study. Apart from the distortion due to the small numbers involved, bare statistics do not cater for the fact that any capacity increase is dependent on a corresponding increase in capacity on connecting trains at Grove Park. So, apart from anything else, there is a danger of treating the Bromley North branch in isolation.

Metropolitan Curve

At LR Towers, we are pleased to see that our favourite section of track, the Metropolitan Curve (or the Metropolitan Reversible Line to give its proper current name) gets its own mention in this study. For readers who have managed to skip over all the comments in the past about the Metropolitan Curve, here is a short summary:

The Metropolitan Curve spur is, effectively, the third side of Borough Market junction just west of London Bridge. Passenger trains rarely use it, but it was once an important way of getting empty trains out of Cannon Street in morning peak (and, to a lesser extent, into Cannon Street in the evening) whilst avoiding London Bridge. Because of the tidal flow normally present into London Bridge on the Cannon Street side (two lines in peak direction, one contra-peak direction) it is the trains running against the main flow at London Bridge that restrict the capacity of Cannon Street.

The Metropolitan Curve

The Metropolitan Curve was originally triple tracked as the nineteenth-century service pattern resulted in trains always going to Cannon Street on their inward journey then continuing via the Metropolitan curve to Charing Cross, before reversing the procedure on the outward journey. The curve was also connected with a spur to Blackfriars (and from there to the Metropolitan Widened Lines – hence its name). Despite being triple track (not two track as the Route Study states), the one track remaining is on a very tight bend. It was taken out of use as a result of the Thameslink Programme and its future was uncertain.

The Route Study includes the unambitious, but probably expedient, proposal that the Metropolitan Curve should become a siding accessible from the Cannon Street end. This means that a 12-car train could be parked there at the height of the peak. In turn, this means that the capacity of Cannon Street would be one train higher in the high peak than it would otherwise be.

Charing Cross station

Route studies are generally full of well-thought-out proposals. However, the team working on them is surprisingly small and every now and then something slips through which appears to be… well… odd. In the past, for example, we have read about extending the DLR to Hayes and a massive tunnel to supplement the Brighton line in the London built-up area before contemplating any localised surface-level improvements first.

In this report, it does seem that imagination on the subject of Charing Cross – with a suggestion that a relocated Charing Cross station be built on Hungerford Bridge over the Thames. Clearly, the success of the new Blackfriars station has influenced the study team and left them looking to repeat its success. There is also talk of closing Waterloo East station.

Given the useful interchange with Britain’s busiest terminus that Waterloo East already provides, the idea of closing it, even with another station nearby, is quite mind boggling. The benefit to passengers of relocating Blackfriars on the bridge was that it opened up the South Bank. With Waterloo East already in existence, and its closure suggested, one could hardly claim the same benefit from a relocated Charing Cross and the closure of Waterloo East.

Quite what this idea is supposed to achieve is unclear. Certainly, the study does not explain how this could be done without decommissioning, at least temporarily, the popular suspended walkways on either side, severely hampering any construction access to the bridge.

It is tempting to wonder if the proposers know of the severe restrictions in place when it comes to putting any piles in the river in this location, as the tops of the Bakerloo line tunnels are not far from the bed of the Thames here, or of the perpetual fear that there might be unexploded World War II ordnance still buried in the river bed. The likelihood of a bomb being present is remote, of course, but the consequence of disturbing one could be devastating.

One also wonders if the Route Study team are aware of the severe weight restriction on Hungerford bridge itself, which would not be conducive to added platforms. Indeed historically one of the main reasons Southern electrified was not to get rid of steam engines, but to overcome the issue of weight restrictions on Hungerford bridge. This prevented them running steam engines simultaneously on adjacent tracks. Even today the platforms that extend over the bridge are of lightweight construction. In the 1980s half of Hungerford bridge (two tracks and the siding that was on the bridge) was closed for redecking. Despite much lower passenger numbers than today, it was pretty chaotic for passengers. A repeat of something like that today would make the works at London Bridge seem like a minor inconvenience.

Hayes terminating platforms

Less baffling than the Charing Cross proposal, but still somewhat surprising, is the suggestion that in future the two platform layout at Hayes would be inadequate for future (unspecified) additional services. It would seem surprising if the existing platforms could not accommodate 4tph each. One wonders what service is envisaged for the longer term future.

In the same vein, there is, more understandably, concern over conflicting train movements at Orpington if services are going to somehow be increased.

Lewisham Station

Station capacity has not been forgotten and on the London Bridge metro lines there are proposals to improve interchange capacity at Lewisham. For many years Lewisham had interchange passageways beneath the tracks close to the junction itself (the two lines at the station form a ‘V’). These tunnels are narrow – the one leading to platform 4 especially so. When the DLR came to Lewisham, major enhancements were made to the National Rail station to facilitate interchange with it. This also provided an alternative route for interchanging passengers who need to catch their next train from a different platform. Unfortunately, the more recent facilities require a lot of walking and it is not surprising that many people prefer to use the original subways. They may be unfriendly and uninviting, but the distance involved is short.

Subway to and from platform 4 at Lewisham

At Lewisham, it would make a lot of sense if the original subways were enlarged. Alternatively, an overbridge at the same location may be possible, but Lewisham station is already high above ground level so that may not be an attractive option.

Whilst it can be argued that improving interchange at Lewisham is not really necessary, a separate proposal, not mentioned in the Study, does rather depend on this improvement.

Finally, the not-explicitly mentioned option

The report apparently goes to great trouble to hint at, but not spell out, a proposal mentioned in the TfL business case for taking over South East rail metro services. This proposal had the objective of reducing conflict at Lewisham and it was to be achieved by combining the North Kent and Sidcup loop lines to provide a service departing from Cannon Street running via Abbey Wood and returning via Sidcup to Charing Cross (and vice versa). Whilst this would reduce conflict, it would also mean that places like Plumstead and Erith (not served by semi-fast trains) may well be totally denuded of its direct Charing Cross service. The Sidcup line would lose its Cannon Street service but, off-peak at least, the alternative of all trains going to Charing Cross might be welcomed.

In addition to the above proposal, the TfL plan also included concentrating the Orpington services on Charing Cross and the Bexleyheath line services on Cannon Street.

Interestingly, when it came to reducing conflict at Lewisham, TfL also weren’t prepared to spell out what they proposed in their business case. This is, presumably, because they didn’t want to attract negative comment for an idea which would undoubtedly be controversial. Nevertheless, it was very clear from a diagram provided in that document what they proposed doing.

Diagram in the TfL business case for running South East metro rail services

Such a service would certainly reduce conflicting movement at Lewisham, which would make the existing timetable more reliable. On its own, though, it would not add capacity. It would also be much more in line with TfL’s philosophy (and the Turn South London Orange concept) of keeping the routes serving various lines simple and make the passengers change trains if necessary, rather than have a complex network with trains to all destinations and all the problems that brings.

Potential for conflicting moves – Lewisham Junction

One obvious disadvantage of the proposal is the need for more passengers to change at Lewisham – already a crowded station and one not ideally suited to interchange. Hence one reason for the keenness of both TfL and Network Rail to improve interchange facilities there.

The other potential disadvantage is less obvious. Having spent years with a scheme centred on London Bridge that does its best to segregate Charing Cross and Cannon Street flows so one does not impact on the other too much, this proposal neatly restores the dependency by specifically creating an operational route joining the two termini.

The only mention of this proposal in the Route Study is in paragraph 3.8.5, where it mentions the key challenge of conflicting train movements in the Lewisham area. Such statements are repeated in the draft franchise but, again, the proposed TfL solution is not explicitly mentioned.

It is impossible to escape the feeling that everyone knows this is a good idea but is too scared to mention it – let alone introduce it – because they know that it is going to be very unpopular with some people. The ghost of the Wimbledon Loop lives on.

Overall, a disappointment

It is very disappointing to note that back in October 2011 we were lamenting the lack of 12-car trains on the Southeastern metro services. Work on infrastructure changes necessary to introduce 12-car trains was largely completed in the 1990s. Yet here we are in 2017 looking to see if we can have a few more of these by 2024. More disappointing still, there seems to be no real solution offered to even cater for the growth predicted by Network Rail beyond 2024.

In the next part of this series, we will stay in the same geographic area and look at what the report has to say about the long established proposal to extend Crossrail from Abbey Wood to Ebbsfleet.

Like what you read? You’ll find more in our magazine

Read in-depth articles about the past, present and future of transport in London and beyond. All in a beautifully laid out print magazine that you can read at home, work or on your commute. Buy it now

Written by Pedantic of Purley