Tramalot: Enhancing Croydon’s Tramlink


We recently looked at Croydon Tramlink up to around 2020. With surprisingly detailed plans available we look a bit further ahead now to the decade after, when considerable expansion is planned.

Where we left off

When we looked at the next few years, it was clear that the driver for improvement was the redevelopment of the town centre. We saw that by the time the redevelopment would be complete we would have the following frequencies on the three main routes:

  • Wimbledon – Elmers End 6tph
  • Wimbledon – Beckenham Junction 6tph
  • New Addington – Croydon – New Addington 10tph

There was also the likelihood of an additional 2tph from Elmers End to the town centre loop and back, and the likelihood that there would be additional stabling available at Elmers End. Although not desirable, it may be possible to survive for a while with temporary arrangements for stabling so, hopefully, this will not be a critical factor in any decision making until around 2020.

Not very satisfactory

The service that will be in operation when the new Croydon shopping centre opens around 2019 will really not be satisfactory for a variety of reasons. Either there won’t be any extra stabling facilities or there will be – but not used to their full capacity. We have already talked at length about the way the extra 2tph from Elmers End and around the town centre loop don’t really satisfactorily fit into any pattern. Probably most awkward of all, one is trying to integrate 10tph from East Croydon to New Addington into a pattern that involves 12tph from Wimbledon to Sandilands.

Problem identified, solution found

The next planned stage appears to be to increase the New Addington service to 12tph. This would consist of 6tph that go around the Dingwall Road loop and a further 6tph that go via the town centre loop. As we’ve discussed before, the proposed Dingwall loop will be located between East Croydon and the town centre, and its primary purpose is to enable additional trams to run without running them around the town centre loop beyond a sustainable level.

It is not entirely clear whether the 2tph to Elmers End from the town centre would remain once there are 12tph from New Addington. These have been included in the diagram below but there must be some doubt about these – not least because this would involve 26tph between East Croydon and Sandilands. The diagram below is an amalgam of TfL diagrams and, based on different documents, appears to be the current thinking.

12tph Adddington

Probable enhanced service planned for sometime after the opening of the Dingwall Road Loop

The purpose of getting 12tph to New Addington is unlikely to be about satisfying initial demand. The demand simply isn’t there – except possibly in the peaks – for such an intense service. More likely, the objective is to produce a nice regular pattern wherever possible and avoid long waits for those whose journey involves a change of trams. The 2tph to Elmers End, if they continued to run, would be “rogue” trams and not fit into any pattern.

The documents so far released suggest that the 12tph New Addington service would be introduced as soon as possible after the opening of the Dingwall Road loop and the implementation of the initial service pattern. However, 12tph to New Addington could not commence without some additional permanent stabling for the trams – more than can be accommodated at Therapia Lane combined with some temporary arrangements elsewhere.

Not doing the obvious thing

Once the 12tph to New Addington service pattern is introduced there is one obvious enhancement to be done. That is to get rid of the 2tph Elmers End service (if it has not gone already) and raise the frequency elsewhere from 6tph to 7.5tph on all routes. This, however has certain problems. For starters it is not sure that Croydon town centre could handle 22.5tph going around the loop. The other problem is that this will probably be the straw that breaks the camel’s back in infrastructure terms, and a lot of work may be required just for a marginal increase in service. The single-track Waddon flyover would probably need doubling, the Beckenham branch would require major engineering work to increase the proportion of double track and it would be questionable how one would manage 30tph in both directions between Sandilands and East Croydon. Wimbledon station should be able to handle an eight minute turnround per platform given that it used to handle a 7.5 minute turnround with a single platform.

The issue with increasing frequencies overall is that, for the most part, it would probably be unnecessary. And where it was necessary an increase of 25% would probably not be enough. It is likely to be very expensive but not deal very well with the busiest loadings which are assumed will be on the Wimbledon Branch.

Sutton Tramlink to the rescue

It is around 2020 that things may get interesting regarding tram expansion. The main objective is expected to be to improve the service on the Wimbledon branch. There would initially appear to be two insurmountable problems. The first is capacity at Wimbledon station (and associated capacity at Merton Park tram/road junction) and the second is that the Croydon town centre loop would be already running at or near capacity.

South Wimbledon and Frith Road 2

South Wimbledon to Frith Road. Not all Croydon town centre services shown.

The problem at the Wimbledon end is quite easily resolved. Diverge at Morden Road and take the route proposed by Sutton Tramlink up the A24 and terminate at South Wimbledon tube station. At the Croydon end one simply does a mirror of the Dingwall Road Loop on the western side of the town. This would involve a new tram stop with a single platform at Frith Road. The loop would enable South Wimbledon trams to serve the popular Church Street tram stop in the more downmarket area of Croydon town centre.

A South Wimbledon – Frith Road new service would not be cheap. Apart from the new route, the Waddon flyover would have to be doubled as would the track between Phipps Bridge and Morden Road. At least this would be an opportune time to do the latter, with the new extension northwards from Morden Road to South Wimbledon. Extra stabling or a new depot would be unavoidable. Nevertheless this new service, largely using existing tracks, would be far cheaper than a completely new route.

Sutton Tramlink by 2024?

If a proposed published timeline is accurate, the necessity of providing an extension to South Wimbledon will mean that advantage of the situation would be taken to build the rest of Sutton Tramlink at around the same time. For both the South Wimbledon – Frith Road and Sutton Tramlink proposal to happen at around the same time TfL is going to have to commit to some large sums of money for trams in the early part of the next decade.

A three phase strategy

Proposed timeline taken from Outline concepts for trams in South London

Service Uplift

Mentioned before but dismissed was a service uplift which involved increasing services from 6tph to 7.5tph. Once the Frith Road Loop is in operation the options available are diminishing fast. It is then that it is intended to increase the frequency. If 22.5tph around the town centre loop is not achievable there is the fallback option of sending all the New Addington trams around the Dingwall Road Loop to leave just 15tph going around the town centre loop. This would be controversial since no trams would make a complete circuit of the main town centre loop. Effectively the spur from Church Street to just short of Centrale would become redundant.

Service Uplift

Service uplift taken from Trams 2030 report

One of the uncertainties of the service uplift idea is whether or not the town centre loop can handle 22.5tph. For this reason the above diagram encompasses the possibility of either 15tph or 22.5tph operating on the town centre loop.

Longer trams

The final option in the plans so far published is to run longer trams. Whilst this might seem a simple solution to the layman it is not easy. Nor is it cheap. As a policy decision it seems that TfL are not prepared to have two trams coupled together. As a result of this, longer trams would be more viable at a time when fleet replacement would be due. When the initial trams are due for replacement there will be the opportunity to buy longer walk-through trams for Wimbledon – Elmers End and Wimbledon – Beckenham Junction routes and leave the remaining trams (including the current Stadlers) to run on the New Addington and South Wimbledon routes (possibly including the route from Sutton).

There will probably be considerable problems with longer indivisible trams at the depots. Reconfiguring a depot is usually an expensive business and extremely disruptive.

Extended Platforms

Whilst it would not be hard to extend most platforms, this is generally far more expensive than most people imagine. You can’t dig up the platform later once it is in service so you have to be sure it is free of utilities (including tram ones) before you extend. East Croydon tram stop will probably need a major rethink as it has points at both ends of the platforms. At East Croydon one also has to consider the vertical profile as the platforms lie on a gently arched bridge.

Some platforms may involve land acquisition issues. This will probably be the case with the northernmost platform at Beckenham Junction and, if Network Rail are not willing to sell, it would involve a Transport and Works Act Application. In addition there could well be difficulties at various stops if the track curves once beyond the current platform, as platforms are generally straight on Tramlink.

Loops on the Beckenham branch will need extending. There may well be power supply issues. It is also quite possible that something would have to be done to mitigate the less powerful Stadler trams having to be deployed on the New Addington branch even if it is only the need to allow extra running time in the timetable to account for the reduced power available and lower top speed. There is also one critical curve that, because of the different characteristics of the Stadler trams, may well have to be rebuilt if Stadler trams predominate on the New Addington branch.

Problems with longer trams and junctions

One of the biggest potential problems with longer trams is that they will occupy junctions in the town centre for significantly longer and so these will need to be rephased at significant cost. The tram is detected by loops in the road that pick up a signal from an antenna located at the end of the trams with only the one in proximity to the front cab being active. Intriguingly the loop not only knows that a tram has passed but also knows which tram has passed. It maybe that a new loop will need to be located beyond the exit to the junctions to take into account the longest tram on the network and report that it has cleared that junction. A possibly cheaper but less rigorous solution is to add a short timing delay if a long tram is detected by the existing loop. There is also a timing phase which kicks in if the loop failed to detect the tram clearing the loop for some reason and this will have to be adjusted for the longer trams.

For some of the junctions in the town centre you could avoid the complexity of catering for different lengths of trams (or the inefficiency of not doing so) by only having trams one length – the longer ones. This seems to be what the plan is, but whether this is by design or just an unintended consequence of the service plan is not known.

The extra time a longer tram will take to clear a crossing will cause problems for other traffic including buses. Having decided to go for long trams in the town centre it probably makes sense to make the most of each traverse of a junction and always use long trams in the town centre. That would go a long way to explain the rationale of only having 15tph through the town centre once long trams are introduced. This would be a reduction of a quarter compared to a 20tph service which is probably the current ideal maximum using the present trams. Obviously the gain in capacity depends on how long the longer trams are. If the trams are only slightly longer than currently then the sort of capacity increase being talked about simply will not be there. If, however, we assume, for example, a 60% capacity increase then 15tph of long trams would be equivalent to 24tph with the trams currently operating.

Long trams

proposed service around 2030 involving a large amount of segregation of long and short trams

Because of the problems stated above, any plan to have longer trams will have just 15tph through the town centre. All short trams will terminate by using either the Dingwall Road Loop or the Frith Road loop. Longer trams may offer considerable advantages but it definitely isn’t a “something for next to nothing” option as some people seem to think.

Beyond 2030

It is hard to imagine what could be feasible beyond 2030 and this would depend on factors such as road use. Will they get more congested or will other factors kick in and make them less attractive? There is the proposed Crystal Palace extension but that is going to require something very drastic indeed to enable the necessary frequency to be run between Sandilands and East Croydon. The tram/road junction at Addiscombe would also become a major issue.

There is the potential to do something major with Crossrail 2 and the rebuilding of Wimbledon station but it is far too early to say in what way this could potentially help.

What seems clear is that, where established, trams will go from strength to strength in London in the next decade. An existing system can take advantage of synergies in a way that a new one cannot. With no sign of any official appetite to propose a new standalone tram network it will be interesting to see how far Croydon Tramlink expands from its current established network.

Thanks to Graham Feakins and Roger Jones who have assisted with the preparation of this article. Cover photo by Sparkyscrum.

Written by Pedantic of Purley