HomeElectrificationGetting electrification right

Getting electrification right

Listen to this article

In 2007, the UK Government published its white paper “Delivering a sustainable railway” which concluded that the case for network-wide electrification has yet to be made. One reason for this was that it was considered that “developments in hybrid technology, biofuels and hydrogen fuel cells will improve the carbon performance of self-powered trains.”

A joint response from Network Rail and the Association of Train Operating Companies (ATOC) pointed out that the UK is one of the few countries in the world with diesel powered high speed trains and that “using ‘diesel’ trains as ‘mini power-plants’ for traction power is inefficient and wasteful”. Soon afterwards, Network Rail published its route utilisation strategy, which made the case for electrification.

By 2009, the Government had changed its mind and announced the Great Western main line (GWML) and North West electrification programmes. At current prices (as are all prices in this article), the GWML programme was to cost £1.28 billion. It is now likely to cost £3.17 billion or £4 million per single track kilometre (stk).

Understandably, this huge cost increase caused a rethink of Government electrification policy. As a result, in July, Chris Grayling announced electrification programme cutbacks including the Midland main line scheme. He justified this decision by explaining that electrification is no longer necessary as new bi-mode trains, and the development of battery and hydrogen power, will offer the same passenger benefits.

In this way, Grayling turned the wheel of Government electrification policy full circle back to 2007.

What went wrong?

Amongst the various investigations into the GWML electrification programme was one by the Public Accounts Committee (PAC). This concluded that Network Rail failed to plan the work properly and its cost estimating was poor. It also noted that Network Rail was unable to explain why there had not been a Transport and Works Act application to avoid a piecemeal approach to the required consents that had delayed the programme.

The committee was also critical of the Department for Transport (DfT) which, it considered, was not an effective client. Furthermore, it found that the DfT had not managed the electrification programme and the associated train procurement as a whole. Thus, rather than being derived from a bottom up programme, the GWML electrification completion date was set by the delivery date for the trains that the DfT had ordered.

The National Audit Office (NAO) report concluded that Network Rail’s original plan was based on over-optimistic assumptions including underestimating the amount of bridge work and overestimating the output of its new ‘factory train’ at 18 piles per shift instead of the average of five achieved in practice.

A key factor not mentioned in these reports was that, when work started on the GWML programme, there had been negligible electrification undertaken in the previous twenty years. As a result, both Network Rail and its contractors had lost key skills and knowledge since the 1994 Heathrow electrification. In addition, there were inevitable inefficiencies associated with the rapid mobilisation of the supply chain for large-scale electrification work.

Britain’s ‘feast and famine’ approach to electrification is illustrated by the graph of UK electrification volumes delivered since 1958. This year, Britain will electrify almost 900 stk, followed by less than 200 stk the following year, after which there are no committed schemes. In contrast, Germany has been electrifying around a steady 200 stk a year for the past 40 years.

Network Rail has now changed its project planning process to provide more robust time and cost estimates (issue 156, October 2017). Mark Carne’s response to the high GWML electrification costs is that “we’ve discovered the cost of electrification of the network is very expensive” and that “in the meantime the trains are getting better”. This reflects the Government view that electrification is just too expensive and, with bi-mode trains, no longer necessary.

Electrification cost challenge

As our accompanying feature “Why electrify?” explains, a diesel-powered bi-mode can never offer the performance, environmental and reduced operating cost benefits offered by electric traction.

This is also the view of Dr Jenifer Baxter, head of engineering at the Institution of Mechanical Engineers, who, in a recent press release, stated that it is wrong for the Government to claim that the benefits of electrification can be delivered by bi-mode trains as they “do not provide the required performance or offer the most efficient or environmentally friendly solution”.

However, electrification’s benefits will not be realised unless the industry can convince the Government that it can be delivered at an acceptable cost. This is a significant challenge given that, in its report on the cancellation of electrification projects, the NAO noted that, in addition to the previously mentioned increase in GWML costs, Midland main line electrification costs had increased from £695 to £1,297 million between October 2013 and November 2014. Furthermore, electrification project delays do not inspire confidence in project delivery.

The Railway Industry Association (RIA), however, feels that the industry can and must deliver electrification at a lower cost. Hence, its “Electrification Cost Challenge” initiative is intended to demonstrate that electrification need not be so expensive. This is being led by RIA technical director David Clarke, who advises that RIA considers that electrification remains the optimum technical solution for intensively used railways if it can be delivered at an acceptable cost.

Although much has gone wrong with GWML electrification, David considers that it is not helpful to assign blame, especially as he feels that “the whole industry got it wrong”. The important thing is to recognise the problems and learn lessons as RIA’s electrification cost challenge initiative is doing. Rail Engineer has seen the initial findings of this work, which includes benchmarking and cost saving opportunities.

RIA’s benchmarking includes the electrification of 1,362 stk in Denmark over a twelve-year period and in Germany, the 225 stk Ulm to Lindau electrification, both of which are costing circa £1 million/stk. It would seem that these relatively low costs are the result of a steady rolling programme.

In Scotland, the Cumbernauld to Springburn electrification project that was delivered in 2014 cost £1.2 million/stk. The information I have from working on the 2010 Airdrie to Bathgate project shows that its electrification element was delivered at a similar cost.

The factors that RIA considers to have increased electrification costs include those identified by the PAC and NAO as well as the electrification famine prior to the 2009 GWML announcement. RIA’s cost challenge also considers cost reduction from efficient delivery including the optimisation of high output plant, foundation depths, TSI compliance, consents, bridge reconstruction, the OLE system, reduction in mast numbers and the power supply.

Although his work is still ongoing, David Clarke is convinced that RIA’s report will show that electrification can be delivered at an acceptable cost through its benchmarking and case studies.

Photo: Phil Adams.
Photo: Phil Adams.

Network Rail’s challenge statement

Network Rail also has an electrification challenge statement: “Smarter, more efficient electrification”. This is one of its challenge statements that are intended to raise industry awareness and promote research into key priority areas.

Phil Doughty, professional head of contact systems, explained to Rail Engineer: “Everything we do is being examined to see how it can be done in a more cost-effective manner.” He added that Network Rail was addressing many of the issues that RIA had identified.

He acknowledged that the GWML electrification structures and their foundations had been over-engineered and felt that one reason for this was that contracts had encouraged a risk-averse design approach. This initially resulted in foundations being designed from first principles, without considering the industry experience gained from installing OLE masts over the past seventy years – one example of knowledge lost due to the many years with no electrification. However, this has now been addressed by a new 2017 standard that incorporates previous empirical methods of foundation design.

Phil advised that Network Rail is developing automated design tools that would incorporate this previous foundation philosophy and promote lighter structures. This tool would also take account of the wider range of master-series cantilevers that are being developed. The previously limited range had sometimes resulted in the use of a heavier cantilever than required.

Although the new electrification TSI had resulted in significant costs, Phil acknowledged that this was generally due to the way the standard had been applied. He considers that improvements in Network Rail’s risk-based approach has now made it more likely that challenges to clearance requirements will be accepted. In this respect, his team were working closely with the University of Southampton on clearance requirements to structures.

This study included a series of tests done under controlled conditions to test the effectiveness of surge arresters which can limit the overhead line voltage to a level compliant with the available electrical clearance value if there is a voltage surge elsewhere, say from a lightning strike. In Cardiff, this technique, together with protective coatings, has eliminated the need to raise a bridge with an estimated cost saving of £10 million.

Photo: Phil Adams.
Photo: Phil Adams.

Team Scotland

Transport Scotland’s director of rail, Bill Reeve, recognises that electrification provides faster, lighter, greener trains that passengers like. He knows that, in most parts of the world, there is a good business case for it, but he recognises that the UK Government has lost confidence in electrification due to its high costs. In contrast, the Scottish government has had a consistent policy and programme of electrification since it took over the responsibility for Scotland’s railways from the DfT in 2005. It remains sympathetic to further electrification if it can be delivered at an affordable cost.

To determine whether this is a realistic aspiration, Bill recently arranged for David Clarke to present the current findings of RIA’s electrification cost challenge to those responsible for the delivery of electrification in Scotland, both in his team and in Network Rail. I was invited to this meeting, in part due to my involvement with the 2010 Airdrie to Bathgate project, which included electrification of 106 single track kilometres.

It was clear from this meeting that Scottish electrification generally costs less than in England. For example, the electrification element of the 2010 Airdrie to Bathgate project was well under £1.5 million/stk. On this project, the Network Rail project team had integrated the delivery of many different contracts to ensure the project was delivered on time and to provide clear accountability. Network Rail Scotland’s route delivery director, Iain McFarlane, advised that this project delivery model is used on current Scottish electrification projects and felt strongly that it works well. Everyone else in the room shared this view.

There was also a strong view of the value of a rolling electrification programme to retain knowledge and skills.

The meeting considered ways in which electrification costs could be reduced further. This reflected the findings of RIA’s electrification cost challenge and the Network Rail initiatives that Phil Doughty described. It also considered the inevitable trade-offs that arise from installing 25kV electrification on the UK’s constrained railway infrastructure. Alan Ross, director of sponsorship for Network Rail Scotland, stressed that such trade-offs had to consider whole-life cost. For example, this might mean that a ‘track lower’ could cost more overall than an initially expensive bridge lift.

These trade-offs also require consideration of the overall business case. Bill Reeve was clear that any such associated project scope discussions had to be taken in consultation with the client, Transport Scotland. This was one reason why he firmly believes that successful electrification projects require a strong client involvement.

One example of this is the Scottish high-level output specification for Control Period 6, which requires Network Rail to develop “an efficient electrification technical specification optimised for Scotland that, in support of the Investment Strategy, can deliver an efficient and affordable rolling programme of electrification.”

This was certainly a very positive meeting, at which those present demonstrated their will to make electrification work. It will, no doubt, inform future Scottish transport investment decisions and help Bill Reeve deliver his aspiration that electrification work in Scotland will restore wider confidence in electrification.

Convincing the (Westminster) Government

Those who find it depressing that the UK Government claims that less-powerful, more-expensive-to-run, polluting diesel trains are better than electric trains should be encouraged by the work being led by Messrs Clarke, Doughty and Reeve. This shows that, with current technology, it is possible to deliver electrification for no more than £1.5 million/stk and that future initiatives should further reduce this cost.

In addition, the table of currently committed electrification projects authorised since 2006 shows that Scotland has a relatively good record of on-time electrification delivery. This helps ensure projects are delivered to budget as the longer a contractor’s workforce is mobilised, the higher the cost.

There is no reason to suggest Scottish project teams and engineers are more competent than their English counterparts, so why is electrification delivery in Scotland more successful than elsewhere?

The two factors that seem to make a big difference are that Scottish electrification is a relatively small rolling programme and that Transport Scotland is a strong, informed client. The current Westminster approach is that, if electrification is too expensive, it should be abandoned, without considering why, by claiming that less-powerful bi-mode trains are as good as electrics. In Scotland the approach is let’s get around the table to fix this.

It is to be hoped that the weight of evidence, aided perhaps by exemplar projects in Scotland, will eventually convince the UK Government that electrification is a good thing. Otherwise there is unlikely to be any more English and Welsh electrification for a long time to come. As a result, hard-won lessons that have now been learnt by GWML and other electrification teams will be lost.

Bi-mode trains may not be as powerful as electric trains, but they do offer the flexibility to enable long routes to be electrified over a long period to facilitate a steady rolling programme to keep electrification costs down. If used in this way, Jo Johnson will be right in saying that bi-modes are a “great bridging technology to a low emission future” – which would be electrification.

Read more: Sustainability and the fallout from scrapped electrification plans


David Shirres BSc CEng MIMechE DEM
David Shirres BSc CEng MIMechE DEMhttp://therailengineer.com

Rolling stock, depots, Scottish and Russian railways

David Shirres joined British Rail in 1968 as a scholarship student and graduated in Mechanical Engineering from Sussex University. He has also been awarded a Diploma in Engineering Management by the Institution of Mechanical Engineers.

His roles in British Rail included Maintenance Assistant at Slade Green, Depot Engineer at Haymarket, Scottish DM&EE Training Engineer and ScotRail Safety Systems Manager.

In 1975, he took a three-year break as a volunteer to manage an irrigation project in Bangladesh.

He retired from Network Rail in 2009 after a 37-year railway career. At that time, he was working on the Airdrie to Bathgate project in a role that included the management of utilities and consents. Prior to that, his roles in the privatised railway included various quality, safety and environmental management posts.

David was appointed Editor of Rail Engineer in January 2017 and, since 2010, has written many articles for the magazine on a wide variety of topics including events in Scotland, rail innovation and Russian Railways. In 2013, the latter gave him an award for being its international journalist of the year.

He is also an active member of the IMechE’s Railway Division, having been Chair and Secretary of its Scottish Centre.


  1. As I understand the current progress, the steady learning process with the Scottish programme has refined the construction flow process, so that a run of OHL support structure installation, has a checking, and preparation process for fitting droppers and registration arms following in an integrated process rather than separated elements. This appears to be delivering several elements ahead of schedule, and with a growing pot of savings that might be available to move on to the next element of the process.

    We’ve already seen testing trains running to Falkirk Grahamston, and Stirling, and the ability to prove the aspired to timetable for a sub-40 minute Edinburgh-Glasgow is daily being tested, with the xx.00 departures being electric, units, regularly arriving 5 minutes early at Haymarket, and an awesome experience to travel up the Queen Street Tunnel with the driver notching back before reaching the summit. The swift and pragmatic move to grab some spare Class 365’s to plug the gap left when the Class 385’s had that windscreen problem should also put Scotrail in a position to roll forward on extended electrification, with a reserve of rolling stock.

    There are also several desirable infill options which can be planned as a rolling programme over several years – The Edinburgh Sub – The Anniesland via Maryhill Line (with the option then to have this as a loop route for any blockade which shuts Queen Street High Level) – The Glasgow Union Line (a starting point for a cross city link between North and South Clyde networks) The wires have also enabled the London Sleeper to terminate at Queen Street during a weekend blockade of Central.

    Each element – the Airdrie-Bathgate especially has provided a greater resilience for the 50 miles between Edinburgh & Glasgow and the loss of one route (from 4 – 5 possible permutations) has thus had minimal impact compared to the melt downs & road replacement chaos seen for the similar 50 miles between London and Brighton.

    In 1976 I started working for BR Sc Region – and the way that Scotland has such a clear ability to find ways to get things done, has continued to impress me. It would not be a great surprise to find that Bill & his team manage to find a way to get rails between Tweedbank and Longtown before long – perhaps as a way to raise the average speed of some key freight services above the current 16 mph, once they have been threaded between the passenger traffic?

  2. Whilst Network Rail are working on replacing overhead wires on the Great Eastern Main Line in & out of London Liverpool Street to & from Chelmsford with new overhead wires have been installed on the GEML in some areas in East London including between Shenfield-Romford, Romford-Ilford and Ilford-Maryland. And with new overhead wires soon to be erected at Stratford.

    And new overhead wires on the Southend Victoria line that have been installed with lots more happening on the Southend Victoria line as the current 25kv overhead wires dates back to the 1950’s that are getting old and do sag during hot weather conditions. And new overhead wires are designed to taut in all weather conditions.

  3. I completely agree with the assessment. Infill electrification to complete the GWR plans (to Swansea, Bristol Temple Meads, etc) should follow. Effective, the UK Govt should give Network Rail a guaranteed £500m a year to electrify what it can, and let it get on with it. At £1.5m per stk, this would pay for 333 stks per year, reducing the diesel miles for bimodes on the GWR until they can be eliminated. Once done, move on to the Midland Main Line, and then the lines out to the far West Country, and then infill the rest of the network over time.

  4. I wonder if the author of the article has, in his research, taken into consideration the additional costs piled onto the recent electrification projects by replacement of near and life expired assets such as bridges and upgrading of existing track layouts etc. The 4-tracking of Filton Bank and other such enhancements may also be skewing the costs onto electrification and giving the current impression that electrification is too expensive.
    As I recall, the WCML upgrade costs more than trebled due to the project having to bring the existing infrastructure up to the standard to which it should have been presented to the upgrade team and, when these legacy maintenance costs were taken out of the final bill, the increase was at a much more reasonable level. I also do not understand why remediation of abandoned bell pits is not accounted separately and possibly paid for by the government the same way as subsidence from more recent mining activities
    It would be interesting and informative to have a similar stripping out of these costs which are not directly electrification related to take a balanced view of the affordability of the various recent projects.

  5. It cannot have been cheap to enclose the GWML in a tunnel of steel. The structures look as if they were designed to carry the weight of the trains rather than the contact wires. 25kV was meant to do away with the heavy overhead structures required for the older low voltage DC systems, but these are more substantial than those used by the LNER in 1946 when it electrified the lines out of Liverpool Street.

    It is hideous too – the section between Reading and Cholsey where it passes between the Chilterns and the Berkshire Downs is a blot on the once-attractive Thames Valley landscape.

    How come that nobody noticed that there was obviously something wrong with the design?


  6. Excellent article. It should be compulsory reading for everybody at DfT and NR, now and into the future. What a sad mess.

  7. There was no clear guidance and robust policies to monitor what the contractors working for network rail were doing on GWML (actually called GWEP) there was no accountability for any work that was done or not done!, the last part of the work nearly completed is not fit for purpose and will cause nothing but delays!, even the way the materials were procured for the project was all wrong and highly expensive!!!
    The problem now is that we have fallen into the same problem as the French and put all our eggs into high speed rail and these big projects and have forgotten about the old aging infrastructure that these trains run on which is at complete breaking point due to a complete lack of investment

  8. Indeed didn’t anyone in charge notice that something was very very wrong? Plenty of onlookers did. It’s a good job that Brunel was spared the hideous sight of all those massive masts and gantries.

    Didn’t Network Rail have regular board meetings where they reviewed the progress of major projects? One can envisage such a meeting in about 2014 when the character of the GW electrification was becoming apparent: “How is the project coming along?” “Well, we are behind schedule, the infrastructure is very ugly, clearly over engineered and not entirely logical, for instance diagonal bracing is not being used on gantries, and it is way over budget.” “Splendid just carry on – money is no object.”

    The pity of it is that it so much our money that is being spent on this one project and that those who use and work on the railway will have to look at the grotesque metalwork for decades to come.

    However some of the single line supports on the Blackpool line and elsewhere are neat and almost elegant. So there is hope.


Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.