Govia go via Greenwich: GTR and the Thameslink Timetable


There are probably Southern commuters who would claim that nothing Govia Thameslink Railway (GTR – Southern’s parent company) said or did would surprise them. Many industry insiders would probably agree. When rumours spread rapidly about a proposal to completely revise the Thameslink timetable, however, there were various shades of disbelief.

The rumours centred on a proposed Thameslink service to Rainham (Kent) via Greenwich. Without any background context this just seemed like a barmy idea – one that not even a mad LR crayonista in a state of psychosis could come up with. The proposal, now officially published however, is about much more than that.

This is not to say that the proposals aren’t still controversial. A Thameslink timetable consultation was a complete surprise because as far as everyone else was concerned the whole thing was pretty much finalised. True, it was an open secret that the 2018 timetable didn’t actually work, but this was thought to be resolvable with a few minor tweaks or the judicious removal of the odd train. September 2016, by generally accepted industry norms, would seem to be far too late to start consulting on major timetable alterations due for implementation on Sunday 13th May 2018.

It seems incredible that the Department for Transport fully consulted on what was thought to be the definitive Thameslink 2018 timetable (give or take the aforementioned minor tweaks) and awarded the Thameslink franchise on that basis, yet we now have major changes being put forward. More remarkable still, the whole basis of having the large Thameslink franchise was so that it would be self-contained and one operator, as much as possible, would be in overall control. If accepted, these new proposals would make Thameslink so interdependent on SouthEastern that the future influence of SouthEastern on Thameslink performance would be comparable with that of Great Northern. So maybe the franchise should have been bigger still or, as some MPs argue, it has already got too big and shouldn’t be taking over every railway on which it runs a significant number of services.

We are getting ahead of ourselves, however, and so first we should step back and take a look at what the proposals contain.

There ain’t nothing quite like it

It has been said many times that Thameslink, currently run by GTR, is a franchise like no other. To begin with, it is a franchise where the DfT keeps all the revenue. The unwritten objective for the train operator is therefore not the usual one of maximising profit by getting as much revenue as possible for as little expenditure as possible. It is to keep the DfT happy so that they continue to pay the management fee.

For better or for worse, it seems that a completely unexpected – and largely unpredicted – consequence of this is that GTR is thinking more long term than other non-TfL-sponsored train operators. As a result it isn’t going to be panicked into doing something just because there is a loss of revenue – provided, of course, that what they are doing doesn’t upset the DfT.

Not content with doing one thing at a time

One would have thought that with the guards dispute on Southern and the controversial proposals for ticket office revision covering all of GTR that the company would be desperate not to get involved with another controversy. Not a bit of it! Indeed it is telling that the head of GTR, Charles Horton, is due to give a lecture next May entitled “Everything had to be done at once” Clearly the softly, softly approach is not on GTR’s agenda.

Reading the document itself, it is clear that at its heart lies a fundamental problem. The aforementioned fact that the 2018 Thameslink timetable, in its current form, won’t work. More importantly though, the consultation seems to admit that in its present form simple tweaks won’t work. As far as GTR seem concerned, it can’t be made to work. Facing up to this fundamental truth has thus led to some pretty drastic decision making.

Getting into the details

It is interesting to note that up until now all the doom-merchants have claimed that Thameslink will be a disaster because of the problems of running 24tph through the core. If the rhetoric of GTR is to be believed though, this is probably the least of their problems – a potential problem, certainly, but one that has been long recognised. The necessary measures have thus already been taken to make sure that it does not become an issue once a full service through the core is run.

So what is the bigger problem GTR anticipate? Here, those looking for something exciting and new to read about are about to be mildly disappointed. For we are back to familiar topics at LR Towers – dwell times, reactionary delay and Windmill Bridge Junction. Indeed on Southern territory this junction must be the railway operational equivalent of Mornington Crescent, because it seems the final answer to everything.

GTR have clearly looked at the fundamental reason why there is a problem and have come to the none-too-startling conclusion that it is the complex junctions in the Southern metro area. They also seem well aware that in the past few years a delay on Southern territory has tended to set off a bit of a chain reaction, meaning the overall consequential delay tends to be around three times the initial delay.

Awareness of the systemwide problems caused by complex junctions is nothing new. TfL Rail are well aware of it and for many years have argued for a more TfL-style of railway on the Southern network – i.e. fewer, more self-contained routes. Indeed much of the Turn South London Orange proposal was about sorting out junctions.

Sorting out junctions, however, is rarely an exercise without consequence. What was really necessary to solve, or at least reduce, the junction problem to manageable levels was thus an organisation willing to take the flak and actually do something about this – despite the inevitable initial unpopularity. Until now no TOC in their right mind would do such a thing. Far better to keep one’s head down, not upset the fare-paying punters and quietly collect those large delay payments from Network Rail when the cause of a delay can be attributed to them. Call them maverick, mad or brave. Call them forward thinking and full of willingness to take the initiative. Whatever your preference, GTR seem to be the first organisation willing to face up to this.

(Evenly) spaced out

On a much more positive note, and something that would likely get the approval of Sir Herbert Walker (of original Southern Railway fame), GTR have recognised that, both north and south of the river, the current timetable is a complete mess when it comes to regularly spaced trains. This is not something that can be resolved by tweaks and tends to involve tearing the timetable up and starting again. It does seem that the proposals would produce a far superior timetable in this respect and this is again something that is greatly desired by TfL, who want to see a “turn up and go” railway.

Maybe even Redhill users will be happy

The consultation states:

Frequent all-day, evenly spaced service of six trains per hour every 10 minutes is proposed between Redhill and London.

The service today from Redhill has gaps varying between 10 and 20 minutes. It is a major town with a large catchment area and it really deserves a decent service, such as the proposed train every 10 minutes in the off-peak period. It is hard to see how this could have been achieved without a major timetable rewrite.

Whilst Redhill maybe a particularly bad example of a major station with an erratic service, it is one of many examples that can be found. A solution to providing a regular interval service at most stations is not going to possible by just tinkering with the timetable we currently have.

It is not only the regularity that will be better if the proposals are accepted. GTR seem to be striving to provide a better service by increasing the frequency. It has to be said that this includes lines that will never even cover their marginal costs, such as the line to Epsom Downs (which goes up from 1 train per hour [tph] to 2tph). South of the river a very welcome proposal is 4tph off-peak on the Catford loop line – something campaigners have wanted for many years. North of the river there are similar proposed improvements, such as 2tph off-peak to Kings Lynn (once work is completed at Ely North Junction) though it has to be said that this is not a new proposal and was due to happen anyway.

Better service in the evening and at weekends

Another TfL aspiration being pursued is the extension of the normal off-peak services late into the evening and at weekends, although it is conceded in this consultation that there may have to be a reduced service on Sunday mornings to allow time for maintenance. Indeed the “be more like TfL than TfL” approach may be an attempt to fight off a TfL takeover in future so that Govia can get another bite of the cherry once the present franchise expires.

Facing reality

Refreshingly, GTR are also fully facing up to the reality of extended dwell time (for the most part). As trains get more crowded then station dwell times can get disproportionally large. It is not just the number of people wanting to get on, but how crowded the train is already, as well as such aggravating factors as a long train and a short platform. So it is a pleasure to see that reality has hit home at GTR and they now propose a one minute dwell time at busy times at stations such as South Croydon and Purley Oaks – not especially busy but the train is already crowded and the eight car platforms mean that on 10-car trains the last two carriages are not accessible.

In a similar manner the proposals also support having increased turnaround time. This would rectify a much criticised issue with previous timetables, where most industry experts regarded the turnaround times as far too tight to allow a service to be operated reliably.

Passenger numbers by Terminal and Thameslink

Dwell times are only going to get worse as demand increases. The consultation appears to confirm the fact that, despite all the current disruption, passenger numbers are rising on GTR.

Let down by Network Rail

Another reality GTR are facing up to is that some of the infrastructure needed to run the previously proposed 2018 timetable is not going to be present in time for implementation. Probably worst of all is the fact that a terminating platform to be built at Stevenage simply will not be ready by December 2018. GTR’s proposed solution is to run trains to Watton-at-Stone and provide a replacement bus service from there or Hertford North to Stevenage. This must produce one of the railway’s quietest terminating stations. Watton-at-Stone, the village where Sir Nigel Gresley died, has fewer than 1000 households and its long-closed railway station only reopened in 1982.


Another disappointment is that there will not be an eight car terminating platform at Reigate by the time the timetable comes into force. The consultation proposes the options of having 4-car trains combine with another train at Redhill or a more frequent shuttle service. The shuttle service would run up to three trains an hour which, when combined with the Great Western Reading-Redhill service, would give a surprisingly good service of 6tph between Redhill and Reigate.

Windmill Bridge Junction

More controversially, GTR believe that the originally proposed throughput of trains through Windmill Bridge Junction (just north of East Croydon) is simply not reliably workable. This would appear to be in contrast to their beliefs when pressing for a more intensive 2015 timetable. For that timetable – the notorious one that produced an unreliable evening peak from London Bridge – it has been suggested that it was at GTR’s insistence that the service into London Bridge was as intense as it was. Network Rail were arguing for slightly fewer trains – not because they thought it theoretically wouldn’t work, but because they thought it would make the service more reliable. Maybe the lesson has been learnt or perhaps it is a case of once bitten, twice shy. Either way, GTR have decided that not only must train routes be modified to minimise conflict at Windmill Bridge Junction, but the service through Windmill Bridge Junction must also be reduced slightly. This fits in well with another objective which is to reduce (or eliminate) trains terminating at East Croydon in the peak.

The problem is that once you reduce the number of trains through Windmill Junction you potentially reduce the number of trains to London. You don’t want to reduce the terminating trains at Victoria and London Bridge because it is not easy to usefully replace them with something else. That leaves Thameslink, which was proposed to have 16tph between London Bridge and East Croydon.

Terminal capacity in London issues

Here it starts getting complicated. Conventional wisdom states that you don’t want to run any Thameslink trains via London Bridge onto SouthEastern tracks because the approaches to Charing Cross and Cannon Street are as full as the termini themselves. So by running from Thameslink to SouthEastern destinations you don’t gain capacity because you lose a path into Cannon Street or Charing Cross. The critical problem seems to be Lewisham station and junction which have severely limited capacity.

New route proposals

GTR are proposing that a proposed Thameslink service to Maidstone East via Elephant & Castle in the peaks be replaced with an all day service to Maidstone East via London Bridge. GTR are convinced that there is capacity on the main line through Grove Park (not via Lewisham). The removal of this service via Elephant & Castle offers a lot of other opportunities for service via Elephant & Castle and GTR is quick to capitalise on this.

Much more controversially they propose a service to Rainham (Kent). The document on the proposals in Kent doesn’t even include the route on the detailed diagram but they do have the route to Maidstone East. It is almost as if they don’t want to draw too much attention to the proposal.

The route to Rainham can’t go via Lewisham and the line through St Johns and Grove Park is full up, so the proposal is to go via Greenwich. GTR argues that this has a lot to commend it as they are convinced that, with their service to Charing Cross permanently severed, many people on the Greenwich line want to travel on Thameslink to Blackfriars and beyond. Furthermore, they point to the rising demand from the Medway towns that is not being met.

The solution of going to Rainham via Greenwich looks clever but has been treated with deep skepticism and concern. The objections basically fall into three categories:

  • There isn’t the demand
  • Operationally it is a nightmare
  • Long term, you have reduced capacity from East Croydon to London Bridge

One could add that there are also a couple of other concerns. One is that after going to the trouble of building the Bermondsey Diveunder the number of Thameslink trains fully taking advantage of it has gone down from 18tph originally to 16tph and now to 12tph. The other great concern is its impact on the business case for the work to sort out Windmill Bridge.

The first issue with demand on any proposed Thameslink service via Greenwich is that GTR’s predictions aren’t believed. They claim they take Crossrail into account and that people will change at Abbey Wood, but one wonders if the fact that TfL fares are cheaper has been factored in. Not only that but it would also be quicker to get to Farringdon via Crossrail at Abbey Wood, so one wonders how many through passengers (beyond London Bridge) would actually be on the train. The service would appear to be too slow for passengers from further out because it stops at too many stations. Stated benefits of satisfying a future demand from the south east to the rebuilt Brent Cross shopping centre from 2021 onwards seem rather far fetched with Bluewater much closer and Stratford easily accessible from Abbey Wood via Crossrail (admittedly with a change of train at Whitechapel). Coming to that, by 2021 there may even be a brand new shopping centre near East Croydon.

Confidence in how well this option was thought out also take a dent when one reads that:

These trains are not able to call at Woolwich Dockyard due to short platforms being unable to accommodate 12 carriage trains.

This seems to either be a highly inaccurate statement or, incredibly, Thameslink class 700 trains have not been provided with Selective Door Operation – which would seem to be almost unbelievable.

The stuff of nightmares

Probably the biggest criticism is that by introducing a Thameslink route via Greenwich one introduces an operational nightmare. Still, based on past history, the mere fact that a Thameslink route is an operating nightmare seems to be no bar to insisting on its introduction (or, more strictly speaking, retention). It is almost as if the imp of the perverse first made sure that the Wimbledon loop remained part of Thameslink and then, not content with his efforts, decided to add running trains via Greenwich to the toxic mix.

It almost beggars belief that having spent millions of pounds and years of construction effort segregating the Cannon Street, Thameslink and Charing Cross flows there is now a proposal to undo some of the good work and have Thameslink services interworked with the Cannon Street services via a flat junction. Worse still, one of the critical track sections involved is only a shade longer than 12-cars. This means trains will probably slow down to a crawl – or stop altogether – before crossing over a crucial junction.

Untangling the tracks

or maybe not

Currently semi-fast trains do not run via Greenwich. There is a reason for this. It comes back to dwell time and the fact that it only takes a minute or so dwell time (in addition to time lost stopping and accelerating again) for the all-stations trains to be holding back a following semi-fast.

Dwell time on Greenwich route

Indeed a Network Rail paper highlights this very real difficulty. The issue of dwell time (or berth time) is looked at in considerable detail on the Greenwich line. In the worst case in 2014, the accumulated dwell time for stations from Abbey Wood to London Bridge via Greenwich was around 790 seconds – that’s over 13 minutes. It is not that much better off-peak. If the slow train ran every ten minutes, which is what they do on this line in the off-peak, then the “semi-fast” needs to make an awful lot of stops if it is going to not be held back by the previous slow train.

It gets worse. The ATO (automatic train operation) section of Thameslink apparently has its handover position (where trains go from being manually driven to being driven under ATO) just about at the point where the Thameslink trains would turn off for Greenwich. This has not been catered for in the complex signalling or the arrangements to switch between ATO and manual driving.

What also does not really appear to have been taken into account is that, with ATO overlay and replacement of Networker trains with modern stock, in a few years time you could probably run the extra Rainham trains into Cannon Street – no Thameslink involved. This, and possibly other factors, may well appear in the much anticipated Network Rail long term study for Kent. One hopes this document surfaces before the end of the GTR consultation in order to see what the alternatives are.

Above all, the proposed route to Greenwich produces yet another interface. It means that yet another train service has to be running well for Thameslink to work properly. It is true that Thameslink already interfaces with SouthEastern at Herne Hill, but the SouthEastern route to Victoria is relatively reliable and not problematic.

SouthEastern’s gain, Southern’s loss

A long term issue with Thameslink going via Greenwich is that it reduces long term capacity on Southern services. If you take away Thameslink routes into London Bridge from Southern territory and replace them with ones from SouthEastern territory then you have either lost that route on Southern or you need to take up a terminating platform at London Bridge to replace it. In the latter case this reduces the opportunity in future for more Southern trains to run to London Bridge terminating platforms as there will be none spare.

Likely to be of far greater concern to some is how this will affect the business case for rebuilding Windmill Bridge junction. If you can’t send the trains to London because there is no capacity, there is not much point in spending money to remove the operating constraints at Windmill Bridge Junction. There’s equally little point in eight platforms at East Croydon.

There is a counterargument which is that if the proposals show various mitigations (such as terminating at Selhurst) to deal with lack of capacity at Windmill Bridge junction then that shows that the work at Windmill Bridge is desperately needed in any case, just to improve the current situation and it will be even more vital if there is any suggestion of an improved service through East Croydon.

Probably at the back of a lot of people’s minds is the issue that once you have introduced a service and have had it established for a few years it becomes extremely difficult to withdraw it. So in ten years time it could be a case that the paths through the Thameslink core for the Rainham trains via Greenwich could be far better used for other services, but that it becomes very difficult to implement those future proposals.

A necessary risk?

It is easy to criticise the Rainham via Greenwich proposal. One suspects that GTR have taken the attitude that, even with this issue, the timetable is much more workable overall and produces a better result. They also have a fundamental problem – if not Rainham via Greenwich then where? It may not be ideal but it could be that, in their analysis, it is the least-worst option.

End of round one

The first round of consultation is due to end in December 2016. This means that GTR will have their work cut out in 2017. In the early days of the original Southern Railway during the 1920s the company introduced electrification, colour light signalling and major track layout changes in the space of a few years. Not since then has so much different activity been undertaken south of the river in such a short space of time. It looks though like GTR have an insatiable desire to be remembered as a franchise that changed the nature of the railway. Whether others will welcome that change is, of course, another matter.

Cover image by My another account

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

In Issue four we talked to Crossrail’s Chief Engineer Chris Binns about the challenges of building a new railway from scratch, and to Network Rail Chairman Sir Peter Hendy about making Britain proud of its railways once again. Buy it now

Written by Pedantic of Purley