What’s it all about, Thameslink?

The intention was to hold off from writing another article about Thameslink until the routes involved were finally decided. Our article in 2011 was rather speculative due to lack of decision making and definite information on decisions that have been made. It was felt at the time we would steer clear of an update until details were finalised which we thought would be fairly soon.

We are nearly two years further on and to some extent the situation is unchanged. A lot of this is undoubtedly due to the delay in re-tendering the Thameslink contract. Only then can we be fairly sure what is going on. Things are, however, starting to get to the point where the final scheme is being “firmed up”. Most of the attention of the recent announcement was about the decision to continue to have Wimbledon loop services going beyond Blackfriars. This and other items in short announcements, when taken together with the consultation document for the franchise and a little bit of analysis, really do start to pin down what Thameslink would be like in 2018. So perhaps now is an appropriate time to look at Thameslink again, as unlike last time we believe that in most cases we have identified the most likely terminating points, although there are still 2tph for which we just cannot find a likely destination south of the Thames.

What’s it all about when you sort it out?

What is the purpose of the Thameslink Programme? This seems like it should be an easy question to answer, yet it is not obvious. A big feature of the scheme is to improve the links between north and south of the river – hence its name – but is that the purpose of the scheme or is it the means of achieving it?

Most people would say that one of the main purposes is to increase capacity. If they mean from suburban and outer London into central London as opposed to within central London then this does not really stand up to scrutiny. Nearly all the increase in capacity is achieved by longer trains. Arguably this would have happened anyway, especially south of the river, and if not then a scheme to provide most of the longer trains would have been possible without the Thameslink Programme.

Undoubtably one of the great benefits of the Thameslink Programme is substantial station reconstruction, but we did not need to have the Thameslink Programme to achieve this. St Pancras Thameslink was rebuilt as part of the rebuilding of the main St Pancras station and is nothing to do with the project. Farringdon and Blackfriars could have been rebuilt as standalone schemes. Thameslink 2000 did not even include plans for a rebuild of London Bridge station beyond that necessary to accommodate the extra through tracks. The inspector of the first public inquiry into the scheme thought that the failure to take advantage of the situation to rebuild the station was most unsatisfactory and it is only for this reason that London Bridge got a total rebuild.

At the end of it all what makes the Thameslink Programme more than just a series of incremental railway improvements done under a common banner is the new track it will provide. That is a new viaduct between Metropolitan Junction (Borough Market) and London Bridge station, the Bermondsey dive-under and the new tunnels linking north of St Pancras Thameslink platforms and the East Coast Main Line.

At the first public inquiry those opposing the scheme argued that the new track at Borough Market was not necessary as trains could go via Elephant & Castle. The response by the promoters was that, not only was sending all trains via Elephant & Castle operationally not possible, it was the greatly enhanced link between London Bridge and Blackfriars (enabling trains to continue from London Bridge into central London) that was in essence the whole point of the scheme. Without it there would be no Thameslink Programme. As we shall see this link is not quite going to be achieved to the extent that was originally intended.

A six billion pound scheme?

Governments, when pressed to show they are supporting something, like to produce figures that are as large as possible. In the case of Thameslink, its magical £6 billion cost includes new rolling stock which would have largely been necessary anyway. It also includes two new large depots for this new rolling stock. So if we exclude rolling stock (and somewhere to store and maintain it) we are really looking at a £3.5 billion scheme. That’s still a lot of money, but a fraction of the cost of Crossrail which also has some limitations to a greater or lesser extent on the number of extra trains that can be run on three of its four branches.

A recent article in the Evening Standard refers to to the Borough Market viaduct as a “£100 million railway viaduct.” Taking into account land and compensation costs this is probably about right so the total scheme costs more than fifty times as much as the critical section around which it is based.

The Trains fiasco

With all the controversy over 4 trains per hour on the Wimbledon loop it is easy to forget the continuing fracas over the rolling stock order. Siemens was announced as preferred bidder but problems by Siemens in raising the finance in the current economic situation meant that the deal has still not been finalised. It slipped into Autumn 2012 and the final completion still seems elusive so an originally planned introduction by 2015 at the latest seems optimistic.

Apart from raising questions about why Bombardier was not offered another chance to re-tender, this has caused problems because early delivery of these trains was being relied upon for various other improvements. One very noticeable sleight of hand is that there are banners and posters at Blackfriars proudly announcing that the first 12-car trains on Thameslink are already running. There are, however, only four of these in each peak. We have the situation where station platforms at stations such as Luton have very expensively been extended, yet it looks as if they will be hardly used before 2018.

A further problem with lack of trains can been seen in the consultation document. During rebuilding works Thameslink core trains will be unable to serve London Bridge for an extended period. This is due to the fact that there will be no physical route available whilst critical junctions are worked on. According to the DfT’s consultation document the longer journey times caused by all Brighton trains being routed via Elephant & Castle means that it will be impossible to run the current level of service. This is because the number of trains required to run the diverted service will exceed the total available.

Not surprisingly the DfT consultation document doesn’t explain why new trains haven’t been delivered to take this into account. It does seem though that there will be additional disruption for passengers during the work that could have been avoided, had the order for rolling stock proceeded as originally intended.

In fact this rolling stock delay has had serious repercussions and this is made worse by the relatively recent electrification announcements. One of the justifications for some of the enhanced electrification schemes was that refurbished Thameslink stock displaced by the new Thameslink stock would be available at around the time electrification would be completed.

Southern order lots of extra railway carriages

It was then not a total surprise when in December 2011 Southern ordered £188 million of extra Electrostars when it became apparent that the trains Southern had subleased to First Capital Connect for the Thameslink Programme would not be returned in time for Southern to fulfil its capacity commitments.

In November 2012 they then announced that they had exercised their option to purchase 40 further vehicles. The exercising of their option to take up ordering forty extra carriages was arguably not surprising, but in the same announcement they confirmed the “procurement” of more rolling stock for the DfT. This was rather strange, and begged the question: just why was the DfT not ordering it themselves? Had their fingers been burnt on the Siemens Thameslink order? it is tempting to wonder whether this is the gestation of a back-up plan, one that, if the Siemens order is delayed further, means there will still ultimately be sufficient trains to run a decent but not full Thameslink service. Indeed the consultation document refers to running some extra Brighton – London Bridge services in lieu of some of the Brighton via Elephant & Castle Thameslink trains during the period of London Bridge reconstruction. Maybe these extra trains are to be used for that purpose?

Where will Thameslink Go?

It is only natural that people interested in the project, and passengers who are potentially affected, would like to know which lines will be served by Thameslink. The highly unusual thing about the project is that, despite a lot of the work having already been done and all of it planned, no-one really knows (or if they do they are certainly not making it public). Indeed the list of proposed destinations has changed an awful lot over the years. A complete surprise was the detailed table in the final version of the London and South East Route Utilisation Study published in July 2011 which we have included below. Some of the proposed destinations were unexpected and never mentioned previously. There are some variations but the wriggle room is reducing and we are close to seeing what we can expect the final version to be.

Thameslink North of The River

The complexity of the Thameslink upgrade is nearly all south of the river. The plans for north of the river have remained firm and almost unchanged for years. There are currently 16tph in peak hours running on the Midland Main Line to Bedford and Luton and also to Kentish Town. The peak period number of trains on the Midland Main Line won’t go up from today’s numbers.

What will be new from 2018 is use of the Canal Tunnels immediately north of St Pancras station and Belle Isle Junction which is on the East Coast Main Line in the Maiden Lane area. These were constructed during the rebuilding of St Pancras and approaches for HS1. By using these there will be 2tph to Peterborough, 2tph to Cambridge and 4tph to Welwyn Garden City if the table in the RUS is correct. This connection will also provide access to one of the main depots to be built at Hornsey.

It is fairly clear from the RUS that the Cambridge and Peterborough services will be a diversion of existing services, as there is no extra capacity available over Welwyn viaduct. It also appears to be the case that the Welwyn Garden City terminators will also be existing services that currently terminate at Moorgate. Diverting these trains does give opportunities for other trains to be run. It is intended to run more long distance (IEP) trains into King’s Cross though they may need to run via the Hertford Loop to avoid the two track section over the Welwyn Viaduct. It also becomes possible to run some more Hertford Loop stopping trains into Moorgate without waiting for any signalling upgrades.

If things were left unchanged there would not be sufficient capacity through Finsbury Park and for a short distance to the north. For this reason works are going on to reinstate a platform at Finsbury Park and upgrade a freight line north of the station as part of the Thameslink works. As a consequence of this work it will be possible to run extra long distance trains as a result of Thameslink freeing up capacity.

Thameslink into Kent – Or Not

The recent announcement about the Wimbledon Loop brought caused considerable reaction but within the announcement was also the statement:

At peak times, from December 2018, 16 trains per hour will approach Blackfriars from the London Bridge direction, and 8 trains per hour from the Elephant and Castle direction.

Arguably, in the short term at least, this is far more significant than a little spat over which trains continue beyond Elephant & Castle through the core section. The publication of the consultation document was the first time that it has been officially mentioned as a possibility that 16 and not 18tph would go via London Bridge in peak hours. All previous references (particularly in the consultation document) suggested “16 or 18”. Remember one of the main goals of the Thameslink Programme was to provide through trains from London Bridge to central London, and the number of these quoted has now been reduced by over 11%.

A clue to the reason for this can be found in an easily overlooked table that appeared in the final, but not the draft, version of the London and South East RUS. If you look at the total number of trains arriving at Charing Cross, Cannon Street and “Blackfriars from the Kent route” they remain identical before and after Thameslink completion. The reason for this can be found in Network Rail’s Summary Route Plan for Kent which, as previously mentioned in our Bromley Branch Line article, states than the route between Orpington and London Bridge operates at maximum capacity during peak times.

Given that everything on the “Kent” side (as opposed to the “Sussex” side) has to join the Orpington – London Bridge route at some point it is clear that the only way to run Thameslink trains from Kent is to divert existing trains. Here we have one of the same problems as the Wimbledon Loop, but worse. It would appear that part of the plan was to divert 2tph that currently run from Ashford into Cannon Street. The problem with diverting anything from Cannon Street in the peak period though is that the vast majority of people on the train will alight at Cannon Street and walk to their nearby office. Cannon Street is the terminus with by far the greatest portion of people who continue to their final destination by walking. It simply would not be worth having a consultation over this. The result would be a foregone conclusion.

It is believed that the train planners have been anxious to cut the number of trains per hour into Charing Cross from 29 to 28. The reason is probably because of the longer trains that will run there. A reduction from 29tph to 28tph at Charing Cross may not sound like much, but could make all the difference when the remaining 10-car trains are replaced by 12-car trains. The plan is for “stepping back” (where a driver does not take out the train he brought in but the next one) to be introduced but you still have the problem that the trains would take just that tiny bit longer to clear critical sets of points.

It is also believed that the number of trains to Cannon Street needed to be cut because of the problem of getting the empty trains out of Cannon Street without sending them back down the main line which is full. To explain this in detail would be an article in itself.

The problem is we have a contradiction here because the consultation document states:

The rail industry has recently been engaged in an exercise planning the timetable that could operate when the Thameslink Programme is completed. This work involves all the affected operators and Network Rail. One particular conclusion has emerged from this work: if services between Kent and the Thameslink core route run via London Bridge it becomes physically impossible, within the envisaged infrastructure available at that time, for the present level of service to continue to run into and out of London Cannon Street. In view of the high demand for Cannon Street services, we believe it may be best for services between Kent and the Thameslink core to run only via Elephant & Castle.

This is an intriguing statement and by implication the converse is true – that if you run all the Thameslink services from Kent via Elephant & Castle then the present level of service into Cannon Street can be maintained. This would also imply that another 2 or three trains per hour could serve London Bridge on top of what was previously envisaged.

The first problem we now have is that, if this statement in the consultation is true and acted upon, we then need an extra two trains per hour to enter the Thameslink tracks from the Brighton main line to make up the 16tph of Thameslink trains going through London Bridge.

The second problem is that we have is that the above statement suggests that proposed Thameslink services to Kent have not been abandoned but merely re-routed via Elephant & Castle. However we know which routes these will be because we were told definitively (supposedly) what these are in the recent announcement. One was Maidstone East and Tunbridge Wells was not mentioned. Neither was any other Kent route mentioned other than Sevenoaks, but then that always was planned to go via Elephant & Castle. So we have a bit of a contradiction here.

One alternative possibility does arise here. Instead of rerouting the Tunbridge Wells trains one could extend the Tonbridge via Redhill trains to Thameslink. The potential flaw with this is that the Tonbridge via Redhill trains are only hourly outside the peaks, but arguably that doesn’t matter because the service in question would be peak only. It must be emphasised that this option is pure speculation.

Spoilt for Choice – Or Not

Of course in the early days it was thought Thameslink was spoilt for choice for routes to take over and Dartford was often mentioned and shown on older maps. Apart from the issue of diverting passengers from where they actually want to go, it is thought a considerable problem is the 20 minute cyclical nature of SouthEastern suburban timetables which will not fit in with the 15 or 30 minute cycle of Thameslink.

In the latest consultation Dartford is only mentioned once. Even this may be a reference to the Dartford – Victoria service via Lewisham and Nunhead which had terminated at Holborn Viaduct until that station closed. This service goes over the route between Orpington and London Bridge via a bridge at St Johns but does not impinge on it directly.

What goes around goes northwards

The recent announcement finalised exactly which services will go through the Thameslink core via Elephant & Castle. Apart from the ones that go around the Wimbledon Loop there will be 2tph to Sevenoaks via Catford and Bromley South (and presumably Bat & Ball) and 2tph to Maidstone East via Bromley South. It is not clear from this whether or not the Maidstone East service is via Catford or Herne Hill. One sincerely hopes it is the former or we shall have yet more services having to cross on the flat north of Loughborough Junction station to access the through route through London.

Very slightly off topic is the issue of exactly what services will terminate at Blackfriars. We know that there will be 8tph in peak hours, presumably only eight cars long, that will terminate at the two terminal platforms at Blackfriars. That is a terminal platform occupation rate of 4tph which is on the high side but easily manageable. A comparable contrast is London Bridge where both before and after Thameslink the value is 3.3 trains per hour per platform.

One would have thought it would be easy to identify these terminating trains. After all, presumably some of them will be current Thameslink services. The problem is that the current peak hour services apart from Wimbledon Loop and Brighton services seem totally erratic and terminate all over the place. For example there is one train in the morning peak from Bedford that terminates at Kent House and one in the evening peak that terminates at Beckenham Junction.

The Catford Loop

Our best guess that most, if not all, trains that terminate at Blackfriars would would come from Kent and use the Catford loop as Herne Hill is a really awkward crossing. One disadvantage of this is that they would have to cross the main line through Blackfriars (which, of course, the Wimbledon Loop trains now have to access from the terminating lines). The advantage of using the Catford Loop is it does not need to use the very busy two track section via Penge East and Herne Hill and it is probably no slower that that route. Unfortunately, like the alternative, this is only a two track line but unlike the alternative there are no equivalent of the passing loops at Kent House. This means it will require careful timetabling to obtain a satisfactory mix of stopping, semi-fast and fast trains.

South from London Bridge

We now look at Thameslink south of London Bridge. There seems to be a policy of not diverting trains from other termini to run onto the Thameslink core. In the early days there were many options that involved diverting trains that served Victoria from places like Eastbourne and Littlehampton to London Bridge and onward to the central Thameslink core. That seems decidedly out of fashion. One reason could be the recognition of the fact that people do not like to lose their train service once it is established and they will be vociferous about it.

In many ways anything electric that runs on the first track from Norwood Junction to London Bridge is an ideal service to run through Thameslink. The trouble is that there aren’t that many of these to start with, some of the services that there are do not have a consistent and frequent enough service to fit into the Thameslink pattern and other trains rather awkwardly split at Purley, Redhill or Haywards Heath.

Caterham trains would seem to be a good choice for Thameslink, notwithstanding the fact that these trains are generally all stations trains. Tattenham Corner is slightly further out and is very lightly used off-peak as one of our commentators, a driver, pointed out and appears not to be a good utilisation of fixed eight-car rolling stock. The problem is that clearly they are running out of suitable destinations and clearly Tattenham Corner will have to do.

The above diagram shows what is high likely to be the all but one of Thameslink service south of London Bridge. Note that all are existing services except for the two Three Bridges services each hour. We said at the start that Thameslink would produce an extra four services per hour in the peaks. This will be achieved by not joining the Tattenham Corner and Caterham trains at Purley. Running each portion as a separate train will, however, only produce an extra potential twelve carriages an hour, i.e. one decent length train, as currently the combined train can already be 10 cars long. Services already run formed of a six car Caterham portion and a four car Tattenham Corner portion. Purley passengers should not prematurely rejoice in seeing the end of splitting and joining and the chaos occasionally generated in the morning rush hour when the trains won’t attach, as this procedure will almost certainly apply in future to Victoria trains to cater for the insufficient number of peak hour paths between Purley and East Croydon.

Towards a final decision

When the Thameslink franchise is finally let we should at last have a very good idea of the proposed cross-London services that will probably remain in place for a good number of years. It is clear that the complexity of the railway timetable in London means that there has been, and will continue to be, an awful lot of investigation to try to provide the best service within all the constraints. Meanwhile, before we get there, there will be an awful lot of disruption for an awful lot of people. As details of this emerge we will try to keep you posted.

The whole is greater that the sum of its parts

Going back to the original question, What’s it all about? The original Thameslink plans would appear to provide very little additional capacity for suburban and outer London as a result of the new vital links. In fact by our calculations it is 36 carriages per hour into London from the south and 32 carriages per hour from the north. It does seem though that the number of extra carriages from the south will be slightly higher if the number of services to Cannon Street is not reduced as originally proposed. Most of the substantial extra capacity will come as a result of longer trains. What we do have though is a scheme that has been considered as a whole, although in this respect probably not as well structured or managed as Crossrail. It is a complex scheme but when it comes to railways that is inevitable. What it has largely succeeded in doing is consider the issues of upgrading a railway as a whole so that we have new and upgraded track and stations and suitable rolling stock and signalling and a completely recast timetable. In just under six years time we will see if it all comes together as planned.