It is nearly four years since Rail Engineer reviewed the state of signalling within Network Rail and much has changed in the intervening period. The organisation is now different and a more pragmatic view is being taken on the deployment of new technology.
A recent discussion with Kevin Robertshaw (programme director for signalling projects) explored progress on all the main elements for the future of main line signalling delivery in the UK, which have resulted in the much greater confidence that now exists for successful outcomes to the plans being put forward.
Organisation
A recent development has been the introduction by Network Rail of an Infrastructure Projects Engineering Group. Heading this up is Helen Samuels, the Engineering Director for Infrastructure Projects (IP), who has responsibility for level 2 assurance and the Integrated Design Group for each engineering discipline. There is a drive for engineering to become focussed on railway systems rather than allowing the individual disciplines to dictate the technology and methodology to be deployed in decision- making isolation.
Work is now delivered through four regional directors with only two national delivery groups remaining – track under Steve Featherstone and signalling under Kevin Robertshaw.
Devolution has empowered the route managing directors, who are accountable for assessing asset and customer needs, to spend their infrastructure budgets in the most effective way. Route control of output specifications, taking into account the important local knowledge factor, is leading to a reduced need for change in support of improved project delivery. Work is delivered internally within the routes or contracted to outside suppliers via the Infrastructure Projects group as appropriate.
These changes to the organisation should lead to spending the available money in a more effective way and the associated reduction in unit costs. Of course, in the period of change, there is always a risk of changing project scope and an associated increase in costs.
Railway Operating Centres
The projected ROCs are making good progress. Intended to eventually replace all existing signal boxes and signalling centres, they are located at Gillingham (East Kent), Three Bridges (South of England), Basingstoke, Romford (Anglia), Didcot (Thames Valley), Cardiff (South Wales), Derby (East Midlands), Rugby (WCML), Manchester (North West), York (ECML), Edinburgh (East of Scotland), Cowlairs (West of Scotland).
The number 11 is not set in stone and debate continues as to whether the existing signalling centre at Saltley, covering the West Midlands, will become a ROC. Similarly, the RETB centres in Scotland at Inverness and Banavie are likely to remain as will the ETCS control room at Machynlleth. The C2C route from London Fenchurch St to Southend, currently controlled from Upminster, will be transferred to Romford ROC utilising full traffic management pod working (with control desks arranged in a cluster) but with the interlocking equipment remaining in situ.
All ROCs are now operational with Derby (East Midlands Control Centre) close to reaching its full capacity and potential. More importantly, the ROCs are delivering what was intended, the integration of signalling and operational management. With all parties sitting in the same premises, decision making and mutual co-operation is significantly improved, all helping to achieve optimum performance.
The long-term objective of including electrical control rooms inside ROCs has yet to be commenced, but the intention is for this to happen at York when the existing signalling control centre for the York area of the ECML moves across. Derby and Rugby were the earliest ROCs and lessons learned from subsequent ones may mean some retrospective changes in layout and operation at these two places.
Traffic management and C-DAS
The early vision of TMS (traffic management systems) being rolled out across the network in quick succession has not been realised. The ‘beauty parade’ of three suppliers, whilst useful, revealed that integrating these into existing signalling and operations technology in both ROCs and existing power boxes would not be as straightforward as first thought.
Currently, only three applications are being progressed – at Romford, Cardiff and Three Bridges, the latter for the Thameslink central core. The Romford system, being provided by Thales, is about to become operational in two stages: the first stage will be an isolated system to support the operators with decision making at Upminster; the second stage will be a fully integrated TMS located at Romford TOC, thereby closing Upminster as an operating location. A TMS pod simulator has been in Romford as a functional training facility that fully replicates the Upminster area of control. Once proven, the system will be extended to other areas of the Anglia route commensurate with signalling upgrades.
At Cardiff, the TMS, also being provided by Thales, will be an isolated standalone system that will produce real time train graphs and other tools for the signallers to see and react to in the way of a decision support. The effectiveness of this approach will be monitored as it could mean a quicker deployment elsewhere if of value.
For Thameslink, managing the arriving trains in an optimum sequence from the various lines north and south of the central core will be a challenge. Everything will be fine if the timetable is running exactly to plan, but operators know that the opportunity for delayed arrivals is considerable. Hence the need for TMS, with Hitachi having the contract to deliver a system which will assist the signallers at Three Bridges to minimise the disruptive effects of out-of-course running.
These three pilot schemes will be analysed closely and will hopefully lead to a rapid roll out once the success criteria are established.
In parallel, many TOCs are providing DAS (Driver Advisory System) to achieve more efficient train running and fuel minimisation. Already seen as effective, the concept cannot reach full potential until every train can take account of other train movements and this will only occur when C-DAS (Connected DAS) is available. Whilst some connectivity to existing train describer systems can be achieved, the full potential is unlikely to be realised until a linkage with TMS systems is widely available.
Both TMS and C-DAS will need to progress in parallel if the capacity gains from optimised train pathing and regulation are to be achieved.
Modular signalling
The concept of modular signalling to achieve a more cost-effective approach for secondary lines was emerging around four years ago. The idea of building a complete signalling system in a factory from standard components, and then shipping to site as a tested entity, appeared to be very attractive and would lead to lower costs. The number of foreseen modules was quite small; signal posts that could be lowered, minimum number of lineside phones, measured cable lengths with plug couplers, standard electronic interlockings with common data sets, level crossings to a standard design and common power modules all made technical and economic sense. No deviation from this principle would be allowed.
In practice, it has not worked out quite like that. Every rail route is different and it was soon realised that alterations to a standardised design would be needed to fit the particular circumstances. This, together with the tendency for a typical development cycle to identify additional requirements, often results in a change of project scope.
It is, however, recognised that elements of the modular signalling principle have real value and will be included in new future signalling projects as a standard offering. Put simply, with devolution to the route managing directors, the choice will be theirs. At one extreme, they may adopt a ‘standard signalling offering’ concept to the maximum possible extent with minimum technology and low cost or, at the opposite extremity, they could go for a signalling system at a higher cost to cater for every operational circumstance that is likely to occur. It will be interesting to see what transpires.
Vehicular level crossings
These remain the biggest safety risk on the railway and big efforts are being made to improve that. Whenever a road level crossing comes up for renewal, a hierarchy of options kicks in. These are:
» Close the crossing;
» Provide a bridge or underpass as an alternative;
» Use a four-barrier crossing, automated with obstacle detection (OD);
» Provide a four-barrier crossing controlled and monitored by CCTV;
» Install an ABCL (Auto Barrier crossing Controlled Locally).
Note that AHB (Auto Half Barrier) and AOCL (Automatic Open crossing Controlled Locally) do not exist in this hierarchy as these are seen as having the highest safety risk. AHBs are also expensive to provide and maintain from a whole life cost perspective. Where they exist currently, they may be replaced on a like for like basis if no other practical alternative exists but new AHB sites are highly unlikely to be provided.
The OD crossings are now finding favour as expertise is established around the country, building familiarity with the design, application and operational use of the technology. Early reliability problems with the LiDAR (Light Detection and Ranging) equipment have been largely resolved and confidence is now at a high level that this is the future wherever level crossings have to be perpetuated. It is recognised that it will need to be horses for courses and, on higher speed lines, the emphasis will be on getting rid of crossings in their entirety.
ERTMS / ETCS
Rolling out ETCS systems remains a challenge. Whilst the overall project has now been taken over by the Digital Railway group, designated as the System Authority and which will do all the planning and generic design. The IP Signalling team will still be delivering the associated infrastructure and conventional schemes in ‘ERTMS ready’ form.
The recent report in issue 147 (January 2017) of an interview with Digital Railway head David Waboso revealed that eight potential routes are identified as candidates for ETCS deployment in the foreseeable future. These are all in need of re-signalling, which the IP team would normally have as its responsibility. The two groups will therefore work together to ensure that any work undertaken before the ETCS design criteria is finalised will be ERTMS compatible.
Two examples of this are already in being. Something has to happen on the Norwich to Yarmouth/Lowestoft route and, as mentioned in January, a conventional signalling scheme is being developed which will eventually convert to ETCS at a later date.
More significant is the re-signalling of Kings Cross station, where the signalling and existing track layout is over forty years old, based around a GEC geographical interlocking with a Henry Williams NX panel, double slips point work, bespoke iron work, all of which is difficult to maintain reliably. A scheme is thus being developed to renew all of this by March 2020 with conventional signalling out toward Wood Green.
The southern end of the ECML is an early candidate for ETCS, but this will not be ready in time for the new track layout, which involves re-opening the third Gasworks tunnel so that train routing into platforms does not have to happen in the immediate station throat area. Clearly, eventual conversion to ERTMS operation will form part of the route’s future plans.
Two committed ERTMS projects are underway. One is the Thameslink central core including the ATO (Automatic Train Control) overlay, and it is worth recording that a successful trial of two trains running simultaneously under ATO conditions took place recently.
The other is the Paddington to Airport Junction route where ETCS will provide a train protection system for Crossrail trains of equal safety status to the GW ATP system which is life expired.
Resources and finance
Network Rail has worked hard to increase the number of signal engineers, both internally and externally with its suppliers. This has been very successful, with signalling design and testing resource levels at an all-time high leading to a much more confident situation when compared to four years ago.
The pressure on Network Rail finances is well publicised and this may well cause changes in scope for future signalling plans. However, all projects underway in CP5 will be seen through to completion, so the short-term outlook is secure. For the longer term, the Government is actively promoting ways of increasing rail capacity and ERTMS/ETCS is seen as a major contributor to this. It is therefore likely that digital signalling schemes will have a high priority within rail investment plans.
In summary, the way forward is clearer and more realistic. It is evident that the IP Signalling and Digital Railway teams need to develop a very close working relationship as this will be key to success.
Written by Clive Kessell