What are the causes of electrification projects taking ever longer, and how could this be rectified

[mike57]

The WCML electrification from Weaver Junc to Glasgow was delivered in just over from 4 years for about 220 miles of railway from announcement to completion. Current projects such as parts TPE electrification seem to be progressing at glacial pace by comparision.

What can be done to speed them up? What has changed? Costs have also increased massively even taking into account inflation.

The original project cost for 1970 WCML work was apparently £25,000,000 which adjusted for inflation equates to around £500,000,000 in 2022 terms. Depending on the source the GWML electrification, which was a similar scale seems to have come in around £3bn, thats a 6 fold increase in real terms.

Was the BR approach more efficent, is the current obsession with sub-contacting everything making an environment where the companies involved get bogged down with the contractual arrangements rather than actually delivering the work.

The underlying technology is similar, and other technology advances should speed things up.

Where is the extra money going, assuming my 6 fold figure is roughly correct, the 'value' of the resulting infrastructure doesn't seem to be 6 times better. Or did BR hide the cost of projects in order to get approval?

So what would we have to do to deliver large scale electrification at inflation adjusted 1970 costs and similar timescales.

 

[GRALISTAIR]

Multi fold- I will give one reason only at the moment.

Health and Safety has quite rightly massively improved. You can't run wiring trains like they used to.

 

[zwk500]

Another key issue is lack of properly defined scope from the outset, or a rapidly shifting scope.

Something that's often mentioned in passing is also difficulties in acquiring and retaining knowledge. It's not just about project teams being disbanded or shuffled about, but also the best engineers are in high demand and getting them involved with a project can require beating out competing offers for their services.

 

[quantinghome]

BR had in-house capability to get stuff done. It brought in contractors to do the work, but design was done internally and institutional knowledge was maintained.

 

[Nick Ashwell]

BR had in-house capability to get stuff done. It brought in contractors to do the work, but design was done internally and institutional knowledge was maintained.

THIS!

Not just a railway thing either, a lot of businesses have moved to outsourcing work and it leads to viscous spiralling costs whereas retaining staff and running projects internally nets knowledge and low costs, both long and short term!

 

[HSTEd]

Just a few:

1. Trusting magical plant equipment that has never been used on a real electrification scheme to deliver unprecedented improvements in productivity and cost. See GWRM.
2. Massive overspeccing of installed equipment for PR objectives that are never going to be tenable in actual service, like 140mph, again see GWRM.
3. Massive overspeccing of equipment to allow marginal savings in other parts of the railway operation - like allowing two pantographs up on a formation at high speed, solely so you can save vehicles by ordering 5-car trainsets (running doubled) instead of 9 car ones. Again see GWRM.
4. Railway industry-governmental complex abandoning standards set in Britain for "proper" ones from "proper" railways in the EU - see electrifaction clearances on EGIP and GWRM etc etc etc.

Lots of problems really.

 

[The Planner]

Have a look at some of the videos of how it was done. Wires over the border on the BFI player for example and see if we could do it that way now.

 

[swt_passenger]

Multi fold- I will give one reason only at the moment.

Health and Safety has quite rightly massively improved. You can run wiring trains like they used to.

I think you meant “cannot”?

 

[tbtc]

The original project cost for 1970 WCML work was apparently £25,000,000 which adjusted for inflation equates to around £500,000,000 in 2022 terms. Depending on the source the GWML electrification, which was a similar scale seems to have come in around £3bn, thats a 6 fold increase in real terms.

Was the BR approach more efficent

did BR hide the cost of projects in order to get approval?

I don’t claim to know much about other industries but my understanding is that many/most construction projects will have gone up by a lot more than general inflation; I’d be interested to know how building a mile of new motorway today compared to the costs of the initial programme over fifty years ago, maybe that would be a better benchmark than RPI (as there will have been broadly similar pressures from improved safety standards / increased costs of materials etc)?

But, since you allude to it, BR were able to budget a bit different to today. The £25m would have been pretty much the cost of the materials/ staff, rather than things like having to compensate the local TOC for the disruption caused by months of closures/ disruptions/ diversions (a lot of which are then passed onto passengers, given the modern expectation that delayed journeys be compensated for).

Maybe there were also some ‘clever’ bits of accounting, where things like remodelling junctions were done from the ongoing operational budgets before the electrification purse was opened, from some stories I’ve heard.

Look at how Crossrail going to Reading became “free” once GWML electrification was being paid for from a different pot… look at how the wires won’t reach Oxford until someone else pays for its track layout to be remodelled

I’m not criticising BR, they operated at the time they did with the constraints at the time (no “passenger charters” to worry about), but any public/private railway today would have to be accounted for so that true costs were shown and nothing hidden away. It’s just a whole different environment, and I don’t think we can blame privatisation/ nationalisation for it

Re some of the other points here, just as a lot of enthusiasts seem to only use/see public transport that “Carries fresh air” OR is “overcrowded” (but is apparently seldom in between!), the way we talk of projects can be exaggerated… was it done “on the cheap” or “gold plated”… there’s rarely a middle ground

Plus, an in-house team of electrification specialists sounds brilliant but it’s dependent on a government that provides stable long term funding for such things. As usual, the root cause of the problem seems to be one of Westminster’s willingness to give the railway a secure commitment than anything else (however it’s much easier to have this debate by resorting to arguments about Health’n’Safety/ EU/ privatisation etc!)

 

[GRALISTAIR]

I think you meant “cannot”?

Indeed. Thankyou. Now corrected.

 

[Bald Rick]

I don’t claim to know much about other industries but my understanding is that many/most construction projects will have gone up by a lot more than general inflation; I’d be interested to know how building a mile of new motorway today compared to the costs of the initial programme over fifty years ago, maybe that would be a better benchmark than RPI (as there will have been broadly similar pressures from improved safety standards / increased costs of materials etc)?

Correct.

The closest we have had to a new motorway recently is the Cambridge - Huntingdon A14 project; 12 miles of new motorway standard road plus 9 miles of upgraded dual carriageway, mostly in open countryside, a snip at £1.5bn. If we pretend it was all brand new dual 3 lane motorway (it wasn’t, as above) that’s £71m per mile. Best case is that is May 2020 prices, although I suspect it may well be an earlier price basis (And therefore be more).

But how much did they used to cost?

Hansard reports that in 1974 an average mile of rural dual 3 lane motorway cost £1.6m.

https://hansard.parliament.uk/Commons/1974-04-10/debates/30ef05c0-0dc1-419f-aa85-7c944503b4ad/MotorwayCosts

Mr. Ralph Howell
Share this specific contribution
asked the Secretary of State for the Environment what is the total average cost of building a mile of motorway.
Mr. Carmichael
Share this specific contribution
The construction costs vary widely depending on the nature of the terrain through which the motorway passes and on whether the location is urban or rural. The rural motorway costs per mile for construction completed in 1973 at mid-1973 prices were from £657,000 to £1,400,000 with an average of £940,000 for dual two-lane and from £657,000 to £3,600,000 with an average of £1,600,000 for dual three-lane.Urban motorway costs are usually considerably higher and with smaller samples no meaningful averages can be given. These construction costs do not include land costs, the variation in which is so great according to geographical location and adjacent land use that any statement of range and average cost would not be meaningful.

RPI from mid 1973 to May 2020 is roughly a 12 fold increase in prices. So that £1.6m/ mile should now be £19.2m / mile. Compare to the £71m it actually costs (best case).

But, since you allude to it, BR were able to budget a bit different to today. The £25m would have been pretty much the cost of the materials/ staff, rather than things like having to compensate the local TOC for the disruption caused by months of closures/ disruptions/ diversions (a lot of which are then passed onto passengers, given the modern expectation that delayed journeys be compensated for).

Correct. It also wouldn’t have included the cost of any engineering trains or plant (they were overheads), or ancillary staff (overheads), or even, for many projects, design (overheads). Little in the way of cross charging different teams, it was all budgeted elsewhere.

(Pun not intended about overhead costs for overhead line.)

Also some things simply weren’t done then. Environmental assessments. ‘Stakeholder’ consultation and communication. ’Health & Safety’, to put it mildly (It wasn’t enacted until 1974).

Maybe there were also some ‘clever’ bits of accounting, where things like remodelling junctions were done from the ongoing operational budgets before the electrification purse was opened, from some stories I’ve heard.

Correct, three right answers!

 

[snowball]

To go back a bit further than Bald Rick's Hansard quote, the core stretch of the M6 (junctions 13-29) opened in 1962-3 and cost about two-thirds of a million pounds per mile.

 

[6Gman]

Have a look at some of the videos of how it was done. Wires over the border on the BFI player for example and see if we could do it that way now.

I once saw an album of photographs of the Crewe - Manchester electrification works. The most arresting image was the demolition of the old London Road station - two guys on the roof, prising out coping stones with crowbars and letting them drop to a small fenced off area below!

 

[Energy]

’Health & Safety’, to put it mildly (It wasn’t enacted until 1974).

WCML electrification wasn't done safely (see video)...

Look at how Crossrail going to Reading became “free” once GWML electrification was being paid for from a different pot… look at how the wires won’t reach Oxford until someone else pays for its track layout to be remodelled

It was originally going to go to Maidenhead till electrification and station remodelling was done by someone else.

3. Massive overspeccing of equipment to allow marginal savings in other parts of the railway operation - like allowing two pantographs up on a formation at high speed, solely so you can save vehicles by ordering 5-car trainsets (running doubled) instead of 9 car ones. Again see GWRM.

The class 387s? Remember the original splitting plan for the 800s to serve more destinations (if I remember correctly) was later killed of due to GW ATP increasing the time it takes to split (the timetable wasn't workable anyway I think). 802s should have been all or mostly 9 cars though.

 

[Snow1964]

BR basically charged cost of the equipment and getting it installed, plus any direct works such as bridge reconstruction

The old style accounting didn’t include diverting trains and paying compensation to operators (passengers or freight).

Basically they wired it as is, any improvements to junctions or signalling were not electrification costs, but route modernisation.

The other great difference is they kept experienced teams, so if they went to plant a mast and found a cable or pipe they redesigned it on the spot. Screw a bracket to a retaining wall they found a way (and carried some spare parts with lots of pre drilled holes). The modern way is to employ contractors, and if there is a problem, pause the job, sort it, go back later. Slows it down, costs more.

 

[HSTEd]

The class 387s? Remember the original splitting plan for the 800s to serve more destinations (if I remember correctly) was later killed of due to GW ATP increasing the time it takes to split (the timetable wasn't workable anyway I think). 802s should have been all or mostly 9 cars though.

I was referring to the work to allow multiple pantographs up at 125/140mph, which was done to allow 5-car 80x orders instead of 9-cars, otherwise they would not have been able to run doubled at full speed.

 

[JamesT]

I was referring to the work to allow multiple pantographs up at 125/140mph, which was done to allow 5-car 80x orders instead of 9-cars, otherwise they would not have been able to run doubled at full speed.

But if you don't allow for 2x5 80x, how do you serve the requirements? Balloon out the rolling stock order so everything is 9-car? Retain some 5-car in the order but then have to juggle diagramming so the 5-car never ends up on a peak service?

 

[Energy]

I was referring to the work to allow multiple pantographs up at 125/140mph, which was done to allow 5-car 80x orders instead of 9-cars, otherwise they would not have been able to run doubled at full speed.

I know you're talking about the IETs, is there that much difference in the OHLE for 110mph 12 car 387s and 125/140mph IETs?

 

[HSTEd]

But if you don't allow for 2x5 80x, how do you serve the requirements? Balloon out the rolling stock order so everything is 9-car? Retain some 5-car in the order but then have to juggle diagramming so the 5-car never ends up on a peak service?

The cost of the additional rolling stock to just run a uniform fleet of 9-cars would be a tiny fraction of the cost of doing what they actually did.
The operational savings from running them are not that large once you include the extra running from having ten cars for the peak capacity instead of nine with the same number of seats - it would all be peanuts next to the money blown on the extra OLE capability.

Uniform fleets is how the GWML long distance service was always run historically after all..... HSTs for all

 

[JamesT]

The cost of the additional rolling stock to just run a uniform fleet of 9-cars would be a tiny fraction of the cost of doing what they actually did.
The operational savings from running them are not that large once you include the extra running from having ten cars for the peak capacity instead of nine with the same number of seats - it would all be peanuts next to the money blown on the extra OLE capability.

Uniform fleets is how the GWML long distance service was always run historically after all..... HSTs for all

I’m not sure the maths are that clearcut. GWR have 58 5-car 80x. Presumably you wouldn’t need quite as many as you’re not running doubled up services, but there are plenty that currently run singly so we can’t just halve it. Let’s say 45, roughly three quarters. 4 extra cars at £2.8m each comes to £504m.
My understanding is that the massive increase in cost of the GWML electrification is mostly down to the poor planning and implementation of the electrification, rather than an improvement in spec to handle two pantographs at 140mph. (The NAO report mentions the requirement was ‘downgraded’ to 125mph, so they must have started at 140mph). So let’s compare with the original cost of the electrification from https://www.nao.org.uk/report/modernising-the-great-western-railway/ - £736m. Would you really be able to save two thirds of that cost by specifying only 125mph single pantograph operation? And that would be good enough for 110mph multiple pantograph operation for the 387s and have all the other reliability improvements over previous electrification projects? (E.g. the ECML knitting that falls down if you sneeze on it).

 

[D365]

The other great difference is they kept experienced teams, so if they went to plant a mast and found a cable or pipe they redesigned it on the spot. Screw a bracket to a retaining wall they found a way (and carried some spare parts with lots of pre drilled holes). The modern way is to employ contractors, and if there is a problem, pause the job, sort it, go back later. Slows it down, costs more.

In theory, that should mean technical drawings/schematics are updated correctly. Not that it would happen in practice, I’ll bet.

 

[Royston Vasey]

I once saw an album of photographs of the Crewe - Manchester electrification works. The most arresting image was the demolition of the old London Road station - two guys on the roof, prising out coping stones with crowbars and letting them drop to a small fenced off area below!

Fenced off? Luxury! What more do you want?!

 

[Magdalia]

A couple of points that I think have not been mentioned yet.

One is that date of approval of schemes in different eras are not comparing like with like. When schemes such as Weaver Junction-Glasgow and GN suburban were approved, a lot of the detailed planning had already been done, sometimes many years earlier.

Another is that, as part of the less rigorous approach to Health and Safety, some schemes were able to do wiring work between morning and evening peaks on weekdays. Much of Bishops Stortford-Cambridge was wired in this way, with the engineers working on one line and the trains running, with single line working, on the other line. Much of Ely-Kings Lynn was also done between morning and evening peaks on weekdays, but with a substitute bus service.

 

[yorksrob]

Single line working is something that needs to be brought back.

As I've said before, it can't be beyond the wit of man to design a wiring train with a raisable protective wall to shield workers from the opposite track.

 

[The Planner]

Single line working is something that needs to be brought back.

As I've said before, it can't be beyond the wit of man to design a wiring train with a raisable protective wall to shield workers from the opposite track.

They can do SLW, but it has to be risk assessed and clearly anything running has to be diesel if its a renewal or similar. It also boils down to what service you can run on that SLW and how you deal with the passenger numbers cramming on to a reduced service.

 

[Bald Rick]

As I've said before, it can't be beyond the wit of man to design a wiring train with a raisable protective wall to shield workers from the opposite track.

Not much use for demolishing bridges, replacing signalling, change platform canopies, and all the other things that make up electrification. The wiring activity is but a small percentage of the whole project.

 

[yorksrob]

They can do SLW, but it has to be risk assessed and clearly anything running has to be diesel if its a renewal or similar. It also boils down to what service you can run on that SLW and how you deal with the passenger numbers cramming on to a reduced service.

Indeed. In the case of trans-pennine, you'll be pushing everyone onto the Calder valley anyway, so anything you can run wrong line will add capacity.

Not much use for demolishing bridges, replacing signalling, change platform canopies, and all the other things that make up electrification. The wiring activity is but a small percentage of the whole project.

Going back to the older UK clearance specifications would help with the other bits and bobs

 

[The Planner]

Indeed. In the case of trans-pennine, you'll be pushing everyone onto the Calder valley anyway, so anything you can run wrong line will add capacity.

If you haven't got a natural diversionary route, it still doesn't solve it. Didcot to Oxford being a prime example.

 

[HSTEd]

My understanding is that the massive increase in cost of the GWML electrification is mostly down to the poor planning and implementation of the electrification, rather than an improvement in spec to handle two pantographs at 140mph. (The NAO report mentions the requirement was ‘downgraded’ to 125mph, so they must have started at 140mph). So let’s compare with the original cost of the electrification from https://www.nao.org.uk/report/modernising-the-great-western-railway/ - £736m.

It also came from the fact that the original estimate was totally unrealistic.
This was the era when Network Rail was claiming that renewal of third rail systems was more expensive than a new 25kV installation after all.

 

[mike57]

At the risk of getting shot down in flames...

If routes like the current TPE Leeds - Manchester route (mainly 75mph max, with scope to increase to maybe 90mph but not much further), how much cheaper and quicker would a side or bottom contact protected 3rd rail at say 1200v DC be to install. Multi voltage stock is now commonplace, the DC buses and motors tend to work at these sort of voltages anyway, no bridges to raise, no catenary to thread through tunnels, station canopies to alter, a whole lot of work is no longer required. Gapping can be addressed with a small supercaptitor power pack. 3rd rail can do 90-100mph, none of the existing route is good for mare than that. Installation could very simple taking advantage of short nightime periods. Containerised feeder stations. Might not be the gold atandard, but I am sure you could progress a lot quicker, at far less cost. After all the L&Y did side contact 1200v DC over 100 years ago.

Judging by current progress trains between Leeds and Manchester will still be diesel in 2040. We need to find a way to deliver cost effective improvements. And remember once you have done Standedge route you really need to do Calder Valley

If you ever get a new 'high speed route' between Leeds and Manchester that will be 25kV OHL but I think given the challenges and costs currently its pie in the sky. and if it ever does get built the 3rd rail would probably be due for renewal by then, so maybe revisit if its worth going to 25kV OHL, but by that point it would be local services only and the need may not be there to upgrade, just replace like for like, as the current barriers will still be there.