Relearning Electrification

The project to reopen the Airdrie to Bathgate (A2B) line in 2010 included electrification to extend the Glasgow suburban electrification network to Edinburgh via this new line. This electrification work was part of a £60 million contract to electrify 106 single track kilometres (stk) and lay 44 kilometres of track on the new line. This project, which was delivered to time and budget, was Britain’s first significant electrification since the 1994 Heathrow and Leeds North West electrification schemes.

After this long gap, A2B was to be the first of many new electrification schemes as the UK government had accepted the benefits of electrification. Between 2009 and 2012, it announced electrification of Great Western main line, North Western lines, South Wales main line, Midland main line, Electric Spine, Crossrail, Gospel Oak to Barking line and West Midlands suburban lines. In addition, the Scottish government was funding various electrification schemes. These electrification programmes totalled over 2,000stk.

The Great Western electrification programme (GWEP) started in 2010 and was to cost £1 billion. By 2016, its cost had risen to £2.8 billion and its scope was reduced. By 2017, the government had lost faith and cancelled the Midland main line, Swansea and Windermere electrification schemes. This was justified by the claim that electrification was not necessary as new bi-mode trains offer the same passenger benefits despite their diesel mode having about two thirds the power of their electric mode (issue 157, November 2017).

RIA’s cost challenge

Although electrification offers significant passenger, cost, reliability and environmental benefits, these benefits will not be realised unless the UK Government is convinced that any future electrification will cost far less than GWEP has.

The Railway Industry Association (RIA) considers that electrification remains the optimum technical solution for intensively used railways – if it can be delivered at an acceptable cost. Its technical director, David Clarke, who considers that the industry can and must deliver electrification at a lower cost, is leading RIA’s Electrification Cost Challenge, which recently produced its report. This highlights lessons from schemes in the UK, notably Scotland, and elsewhere to show that electrification can be delivered at a lower cost than GWEP.

David acknowledges that much went wrong with GWEP, but he feels that it is not helpful to assign blame as “the whole industry got it wrong” and the important thing is to recognise the problems and learn lessons. In this respect his report identified the following reasons for GWEP’s cost escalation:

• Unrealistic programme as completion date was set by delivery date for new trains determined by the Department for Transport;

• Immature estimates with little survey information or cost data from recent schemes;

• Unclear specification as Network Rail didn’t know whether the Department for Transport wanted trains to run at 125 or 140mph;

• The development of high-output electrification construction trains that had not been used before;

• Unnecessarily conservative pile design requiring piles up to 15 metres long which resulted in poor productivity with many repeat visits to individual sites;

• Competition for delivery resources, for example with North Western, Scottish and Midland main line electrification schemes all taking place at the same time;

• Introduction of new UK requirements for multiple pantograph operation at up to 140 mph (later reduced to 125mph) resulted in a new OLE design specification that was more onerous than the European Energy Technical Standard for Interoperability (ENE TSI) which was itself under revision when the project was being designed;

• In addition, the UK introduced more onerous clearance requirements than ENE TSI and it was initially perceived that the ORR expected absolute compliance rather than allowing deviation following robust risk assessment and appropriate safety measures;

• The unproven Series 1 overhead line system was developed during project delivery and was designed for 125mph multiple-pantograph operation, TSI compliance and ease of installation;

• The volume of planning permissions and consents was under estimated;

• The lack of a collaborative contracting strategy with clear objectives, shared incentives and fewer interfaces.

RIA’s electrification cost challenge report explains how lessons from the above have been learnt and implemented. Furthermore, it shows that the underlying cause of most of the above issues is the British ‘feast and famine’ approach to electrification, which meant that there was initially insufficient expertise to design, plan and deliver electrification projects on the scale of the GWEP.

This was not a problem for the much smaller Airdrie to Bathgate electrification as, in 2010, it did not have to compete for resources. In addition, it did not have the problems of unclear specification or standards changes. This perhaps explains why this electrification work was delivered to time and budget.

Team Scotland

Unlike Westminster, the Scottish Government is committed to a substantial rolling programme of electrification that, it believes, will bring significant economic, social and environmental benefits. Including A2B, it has funded a rolling programme of seven separate schemes over a ten-year period that will have electrified over 500stk once the Shotts scheme is completed in May.

The Scottish electrification experience provides useful information for RIA’s electrification study, which notes that two schemes completed in 2014, Cumbernauld and Rutherglen, delivered electrification for less than £0.75 million per stk. However, the RIA report notes that £/stk is actually quite a crude measure of performance in view of the varying amount of electrification clearance and power supply work between different schemes.

Although the Edinburgh to Glasgow main line electrification was over budget at £2 million per stk, the later Alloa and Shotts schemes, which both required significant clearance works, each cost £1.5 million per stk. The RIA report concluded: “Having a rolling programme of electrification in Scotland is benefiting from learning and experience being passed from one project to the next.” It included the following examples of good practice from the Stirling, Dunblane and Alloa electrification project:

• The separation of independent activities, even though this extends the programme, into 1) bridgeworks and other route clearance; 2) site investigation; 3) grid supplies, master feed diagram, isolation and switching design; 4) foundations and 5) OLE installation;

• Extensive ground investigation undertaken at 200-metre centres throughout the route;

• Site-specific GRIP 4 OLE design to consider site information, including clearances, to ensure accurate development of GRIP 5 detailed OLE design;

• Foundation options derived from ground investigation CAD model developed from all possible sources with 1.2-metre-cube trial holes dug at each planned location to confirm foundation setting out and design;

• Staged approach to OLE design using finalised isolation and switching design and as-built foundation positions;

• Foundations installed using MOVAX vibrating units mounted on road-rail vehicles;

• A common data model that included steelwork foundation, masts and small parts schedules, material allocation and the wiring CAD model;

• Masts installed using a road-rail vehicle-mounted manipulator, rather than a crane, with small parts steelwork pre-fixed to avoid working at height;

• To maximise wiring train productivity, particular attention was paid to special foundations to ensure that all masts would be in place for each wire-run with cantilevers and registration arms pre-registered to +/- 50mm prior to wiring;

• Extended midweek ‘rules of the route’ access negotiated so that night-time engineering access could start after the evening peak service;

• A station electrical clearance risk assessment process was developed to assess acceptable clearances for use in OLE design.

Foundations and arrestors

Amongst the various cost-saving measures included in RIA’s report, two particularly noteworthy initiatives are Network Rail’s new standard for foundation design and the use of surge arrestors to reduce clearance costs.

A major factor in GWEP’s cost escalation were obviously over-engineered foundations, up to 15 metres deep, which were the result of an analytical risk-averse design approach. The RIA report considered this to be a major factor in the programme’s poor productivity and resultant cost escalation.

Previously foundations had been designed using empirical methods derived from field tests carried out by the UIC’s Office for Research and Experiments (ORE) in the 1950s. To validate a return to this previous approach, Network Rail engaged the University of Southampton to carry out full-scale field tests to extend the ORE design methodology to 610mm-diameter circular hollow section piles over in-service loading conditions that are at the upper end of current operational experience.

The results of this research are now incorporated in Network Rail standard NR/L2/CIV/074 ‘Design and installation of overhead line foundations’. RIA’s report notes that it is encouraging that the Bedford to Corby electrification project is now installing 95 per cent of its piles using ORE design methods to achieve productivity of six piles in the available working time of 4 hours 30 minutes.

As described in issue 158 (December 2017), surge arrestors have been successfully introduced on Danish Railways to reduce bridge electrification clearances. These work by limiting any over-voltages, for example from lightning strikes. When combined with contact wire covers and an electrical insulating coating (onto an earthing plate) electrical clearances required in both wet and dry conditions are significantly reduced.

The University of Southampton was also involved in this initiative as it carried out 193kV tests under controlled conditions under Network Rail’s supervision to determine that, with this mitigation, minimum electrical clearance requirements could be reduced from 270mm to 150mm.

Just outside Cardiff Central Station, there is a low and highly skewed bridge over the railway which itself crosses a substantial culvert. To obtain the required electrical clearance, the reconstruction of this bridge had been costed at £40-£50 million and the estimate of an alternative option of track lowering and a culvert diversion was £10-15 million. Both these options would have been highly disruptive.

Instead, for a cost below £1 million, Andromeda Engineering worked with Network Rail, Siemens (surge arrestors) and GLS Coatings (insulated coating on the underside of the bridge) to provide a solution that avoided the need for these expensive and disruptive options.

Affordable electrification

GWEP has been the subject of reports by both the National Audit Office and the Public Accounts Committee that draw conclusions about programme management issues. Neither of these reports acknowledges the difficulty of ramping-up supply-chain capability for full route electrification after there having been no such scheme for twenty years.

In contrast, RIA’s electrification cost challenge report focuses on practical and technical lessons from GWEP and other projects. It shows how solutions have been implemented and gives examples of actual electrification costs throughout the UK and in mainland Europe. As a result, the report concludes that, in comparison with GWEP’s £2.8 million per stk, “all-in” electrification (excluding route enhancement and major grid connections) should normally cost between £1 and £1.5 million per stk.

The report recommends that there should be a rolling electrification programme that would maintain a core design and delivery capability and support a culture of continuous improvement. It notes that the German rolling programme of electrification, which retains learning and skills, delivers electrification at significantly lower cost than the best that is currently achieved in the UK.

Although the RIA report demonstrates that electrification can be delivered at an affordable cost, the case for electrification requires that its benefits must also be accepted. Amongst the many documents that show electrification’s benefits are Network Rail’s 2009 electrification route utilisation strategy and the Department for Transport’s 2009 Rail Electrification paper.

The DfT paper notes that electric trains are 35 per cent cheaper to operate than diesels. It also offers the small, but significant, benefit of more powerful electric trains giving a four-minute journey time saving between Cardiff and Swansea, where they must accelerate from station stop to line speed on four occasions. Yet, when this electrification scheme was cancelled, the government view was that electrification offered no time savings because this was not a high-speed route.

It is to be hoped that the UK Government accepts the strategic case for a rolling electrification programme in the same way that it has allocated £450 million to accelerate digital signalling technology deployment as a strategic policy not subject to a business case. If not, the danger is that hard won lessons will be forgotten as the historic cycle of electrification feast and famine repeats itself.