GW Modernisation Project - the mistakes and the future?'

[JonasB]

To my mind that is the consequence of going for a 140mph-capable catenary with a MASSIVE wire tension factor. At least it'll last!

Do they really have to be that massive? They look really big compared to other high speed lines.

The LGV Est européenne, with a maximum speed of 320 km/h:

 

[edwin_m]

Do they really have to be that massive? They look really big compared to other high speed lines.

The LGV Est européenne, with a maximum speed of 320 km/h:

The point was originally made with respect to West Drayton, where there are four tracks. It's a little misleading to compare a two-track OLE, where in practically every case each track cantilevers from its own post, with a multi-track system that requires booms or portals. Perhaps someone could post a similar view of OLE on one of the two-track sections of GWML?

 

[LNW-GW Joint]

The LGV OHLE seems quite similar to our WCML/ECML 2-track sections.
The masts are quite slim, and so is the registration gear.
It has an AT feed added on both sides, but this is no higher than the basic masts.
We seem to feel the need to raise it much higher, adding to the visual footprint.

Pictures I've seen on the 2-track section to Newbury show that same thick and high masts as on the 4-track sections, mostly using twin-track cantilevers.

Here's a picture I found on Wiki of the OHLE on HS1 at Ebbsfleet - one of the few multiple track sections.
It's more cluttered than the LGV, but still less than GW Series 1.
I guess it is all rated for multiple-pantograph operation at at least 140mph.

 

[Domh245]

I may be misremembering, but I seem to recall reading that the LGVs (and HS1) use a lower wire tension than GW (and hence less chunky equipment), but that the pantograph uplift force is higher than that on conventional UK pantographs, which is how they solve the issue of keeping contact with multiple pantographs at high speed (and even then, rather dubiously if you watch a Eurostar sparking it's way towards the Chunnel if you are driving along the M20). I am not sure why we are doing it the other way around (lower pantograph uplift but higher tension) but I suspect that there'll be a good reason for it.

 

[dviner]

The LGV OHLE seems quite similar to our WCML/ECML 2-track sections.
The masts are quite slim, and so is the registration gear.
It has an AT feed added on both sides, but this is no higher than the basic masts.
We seem to feel the need to raise it much higher, adding to the visual footprint.

From what's been quoted in other threads, on the GWML electrification the AT feeds are positioned so that, in the case of them falling, they would earth out on the structures. On the LGV OLE, the AT feed may end up on the ground, or waving around in the wind.

 

[Olaf]

Indeed. I don't like one aspect of the Select Committee system, which is the tendency for backbenchers, or indeed witnesses, to try and get their moment of glory, as they know that is the way to get on the box for the 6 o'clock news.

That is the nature of the beast in all walks of life, however the system is designed to allow members to express their view while also arriving at a general consensus in the final outcome.

Still, I am glad that the late Norman St John-Stevas brought in the SCs. They have been useful in the rail scrutiny area. I think the problem with their reports, is that they don't themselves have to act on any recommendation and prove they were right. In this case their report could be used as an excuse to chicken out of further electrification and we need the latter, at least to have an opportunity to get away from fossil fuels. Arguing about BCRs deflects from that higher goal IMO.

The reports are there to provide informed analysis, allowing Parliament to take whatever action it sees fit. I think the PAC looks at execution, not Government policy so it does not have the remit to instruct the Government, which rightly must have the final say.

 

[JonasB]

I may be misremembering, but I seem to recall reading that the LGVs (and HS1) use a lower wire tension than GW (and hence less chunky equipment), but that the pantograph uplift force is higher than that on conventional UK pantographs, which is how they solve the issue of keeping contact with multiple pantographs at high speed (and even then, rather dubiously if you watch a Eurostar sparking it's way towards the Chunnel if you are driving along the M20). I am not sure why we are doing it the other way around (lower pantograph uplift but higher tension) but I suspect that there'll be a good reason for it.

If you are correct, that might explain the difference. Do you know what the wire tension used in the UK is? I haven't found the LGV wire tension, but Swedish main lines use between 7 and 15 kN, depending on speed.

 

[snowball]

According to the table at the very end of this book, written by a contributor to this forum, the tension in Series 1 (the design used on the Great Western line, that has been accused of overdesign) is 16.5 kN in the contact wire and 13 kN in the catenary wire.

In Series 2, currently being used on the lower-speed electrifications in Scotland and NW England, it's 11-15 kN in the contact wire and 11-12kN in the catenary wire, depending on the subtype. I don't know what subtype is used where, except that I think "Master Series" refers to a new compromise between Series 1 and Series 2.

Most catenary installed from the 1970s until recently was Mark 3, 3a, 3b, ..., in which the tension in the contact wire seems to be between 9 and 10 kN, an in the catenary wire between 8.5 and 11 kN.

Curiously I don't see any entries in that table relating to the ECML, except for the first line of Mark 3d which says "various locations on ECML". I always thought the ECML was Mark 3b.

 

[HSTEd]

ECML is a mix of various technologies - indeed I believe they had composite aluminium-steel caternary wire for a wile.

 

[LNW-GW Joint]

Curiously I don't see any entries in that table relating to the ECML, except for the first line of Mark 3d which says "various locations on ECML". I always thought the ECML was Mark 3b.

The text (9.2.4) describes Mk3d as being a 2000s upgrade design for the ECML (ie the work planned to life extend and remove headspans at key places).
I'm pretty sure the original was 3b (north of Hitchin); 3a would be WCML north of Weaver Jn, and King's Cross-Hitchin-Royston.
The rollout table seems to be missing a few entries, one of which is Newcastle-Drem on the ECML.

 

[edwin_m]

ECML is a mix of various technologies - indeed I believe they had composite aluminium-steel caternary wire for a wile.

According to March Modern Railways it still has, but it is being gradually replaced.

 

[HowardGWR]

Anyway, what's wrong with 'future proofing'? There won't be an HSL from Bristol to London any time soon, so what's wrong with making the line capable of higher speeds and ensuring robust performance?

 

[Class 170101]

Anyway, what's wrong with 'future proofing'? There won't be an HSL from Bristol to London any time soon, so what's wrong with making the line capable of higher speeds and ensuring robust performance?

A political lack of understanding of the costs involved.

I imagine the whole life costs of Series 1 and 2 will be significantly less than the Mark III design installed on the ECML once the number of de-wirements and un-planned renewals are taken into account but todays politicans cannot see this, being a here today and gone tomorrow type of character.

 

[edwin_m]

Presumably when the scheme was approved it was specified that it would be compatible with 140mph running, multiple pantographs etc. So that should have been taken into account in the costings and the fact it was signed off would suggest the people who did the signing were content with the specification.

The problem is that actually delivering to the specification has cost way more than was expected. And by the time that became clear de-scoping the specification would just have caused extra delay and possibly cost more than it saved.

 

[coppercapped]

Presumably when the scheme was approved it was specified that it would be compatible with 140mph running, multiple pantographs etc. So that should have been taken into account in the costings and the fact it was signed off would suggest the people who did the signing were content with the specification.

The problem is that actually delivering to the specification has cost way more than was expected. And by the time that became clear de-scoping the specification would just have caused extra delay and possibly cost more than it saved.

I suggest that it was the other way around. The costings were done to one specification and the build was to another set of specifications.

There was a letter published in Modern Railways some months ago and a similar one in, IIRC, in Rail from one of the people involved in the original costings. (I'm afraid I don't have my back copies to hand so I can't give the references).

The gist was that the costings for the Great Western electrification were based on the Mark3b electrification equipment used on the East Coast but allowing for changes and improvements which were known about at the time for 125mph and single pantograph operation. These estimations were done between 2006 and 2008 and these were the numbers used for Adonis' announcement in 2009.

The 140mph running and multi-pantograph operation came later as did the balls-up with the electrical clearance standards.

The National Audit Office's report from November 2016 has this to say:

In 2007 the Department decided to buy high-speed diesel trains for the route under its Intercity Express Programme, to replace ageing trains and increase capacity. This changed in 2009, when the Department announced that the line would be electrified and that it would buy a combination of electric and bi-mode high-speed trains. The Department’s objectives could only be met by working with Network Rail and the train operator, to complete the electrification and other infrastructure works and introduce new trains. The Department did not produce a business case bringing together all the elements of what became the Great Western Route Modernisation industry programme until March 2015. This was more than two years after ordering the trains and over a year after Network Rail began work to electrify the route.

My italics.

 

[HowardGWR]

Perhaps worth remembering that the NAO now calculates the BCR for the GWMP (that's the whole lot, to Penzance including electrification) at 1.6 instead of 2.4. Even if it fell below 1.5, it would still be value for money, but expressed as 'low' value for money. At the moment in stands at 'medium to high'. These BCRs do not put a price on the societal and environmental cost of emissions, nor do they take account of the lesser ability to cope with the sheer quantity of expected growth.

It seems to me that pursuing the project,as envisaged, in total, is well worth it.

 

[LNW-GW Joint]

Perhaps worth remembering that the NAO now calculates the BCR for the GWMP (that's the whole lot, to Penzance including electrification) at 1.6 instead of 2.4. Even if it fell below 1.5, it would still be value for money, but expressed as 'low' value for money. At the moment in stands at 'medium to high'. These BCRs do not put a price on the societal and environmental cost of emissions, nor do they take account of the lesser ability to cope with the sheer quantity of expected growth.
It seems to me that pursuing the project,as envisaged, in total, is well worth it.

The more important impact is the denial of funds for other worthy projects (spent on finishing GW), the worsening of the BCR for all electrification projects, and the loss of confidence in NR's capability to deliver.

NR's solution, which is to complete the design much more thoroughly before starting work on the ground, is a recipe for projects not being started in the first place.
Would Adonis (or more importantly the Treasury) have approved the project if he had known the true cost and timescales?

 

[QueensCurve]

I passed some massive OHL masts today near West Drayton.
They needed 4 piles each to support them.
The masts were those needing a supporting anchor.
On older lines, the anchor is a wire, but on GW it is a large fixed rod.
Both mast and anchor were set on separate metal platforms, each needing 2 piles for support, so 4 all told.
It looks massively over-engineered.
Just saying...

It is massively overengineered and we will pay the price with less future electrification.

 

[QueensCurve]

To my mind that is the consequence of going for a 140mph-capable catenary with a MASSIVE wire tension factor. At least it'll last!

And yet French High speed lines engineered for 186mph+ seem to get by with slimmer structures and more elegant OHLE. Where have we gone wrong?

 

[Deepgreen]

And yet French High speed lines engineered for 186mph+ seem to get by with slimmer structures and more elegant OHLE. Where have we gone wrong?

My thoughts exactly. Have we simply drowned in a sea of muddled requirements and standards, leading to a cover-all solution?

 

[coppercapped]

The more important impact is the denial of funds for other worthy projects (spent on finishing GW), the worsening of the BCR for all electrification projects, and the loss of confidence in NR's capability to deliver.

NR's solution, which is to complete the design much more thoroughly before starting work on the ground, is a recipe for projects not being started in the first place.
Would Adonis (or more importantly the Treasury) have approved the project if he had known the true cost and timescales?

The point about changes in the specification causing an increase in costs was understood by all parties.

At the time NR could borrow money from the financial markets and this was added to its debt. At the time of the Hatfield accident NR was permitted by the Government of the day to borrow the funds it needed rather than have the Treasury supply the money - this was because the Chancellor, Gordon Brown, did not want the Government's indebtedness 'on balance sheet' to increase because of national and international financial issues so he preferred an 'off balance sheet' solution. NR could borrow at very low interest rates because the Government had agreed to guarantee the debt.

Borrowing was institutionalised by the Office for Rail Regulation which adopted a concept called the 'Regulated Asset Base' which was the ratio of borrowings to the nominal asset value of Network Rail. As long as this ratio did not exceed a given value everything was hunky-dory. (Except for the ever increase part of the Network Grant from the Government which went into servicing this debt pile).

So - up the the time in 2014 when Network Rail's debt became part of the National Debt, NR could borrow happily to cover any cost overruns. This meant if the initial estimates were wrong - or the spec. changed - it didn't matter, NR simply borrowed the money it needed. The DfT knew about it, the ORR connived in it and the Government set up the arrangement in the first place.

The reclassification of the debt closed this source of cash - and this is now why the over-engineering and the cost overruns are now causing such pain.

It's not simply due to incompetence on NR's part, it was working within the limits of the structure set for it. If anything the roots of the problems lie with Gordon Brown's decision to permit NR to borrow on the open market and the institutional structures set up to support it. This meant that NR floated away from the real world of costs and returns as more money could easily be borrowed - with the effect that checks on over-engineering, gold plating, or arguing about inappropriate standards, were not considered necessary.

The question you pose is only partially relevant. The original estimates were probably not far out - if the electrification had been done on that basis. The point is - it wasn't and if NR had been limited to that amount of money the mission creep, the acceptance of inappropriate standards and all the rest would most likely not have happened. The question is to my mind, moot but equally if the calculations had shown that electrification would have been too expensive for the traffic carried then not electrifying would be the right answer.

Incidentally - let this be an awful warning for those that still consider that spending on HS2 and on NR comes out of two independent pots.

It doesn't.

 

[HowardGWR]

I agree that MML electrification has a better BCR on paper, than that of GW, but it always had so. Thus I assume that to begin with GW was a decision based on other merits. It should also be remembered that the pace of investment, originally set, was so high that MML and TPE / Northern would have been all well on the way to completion. Well, we have had our 'pause' and our moment of reflection, and I don't see how anything has changed. The railways need massive investment in all their projects and that is something our government just has to grasp. All other western EU countries are able to make great strides with their HSL projects because their present lines are already electrified and modernised. We have to do both at the same time - but where are the resources to come from?

Keeping to GW modernisation, if you look at the Portishead branch, there is already evidence of gross incompetence on that one (see recent report), but we should still do it.

 

[SamYeager]

Keeping to GW modernisation, if you look at the Portishead branch, there is already evidence of gross incompetence on that one (see recent report), but we should still do it.

Is that the one where estimated costs have risen from £58m to £175m?

 

[edwin_m]

I agree that MML electrification has a better BCR on paper, than that of GW, but it always had so. Thus I assume that to begin with GW was a decision based on other merits.

I think rolling stock was a big part of the reason to go for GW first.

With the HSTs nearly life expired, a new long-distance fleet was needed for GWR. Ordering new diesels would effectively postpone electrification for several decades. Also electrification would release most of the London commuter DMU fleet, to start a cascade resulting in more DMUs for the North where despite other electrification schemes passenger growth was outstripping fleet capacity. At the time it was thought this would avoid the need to buy any more commuter/regional DMUs for the foreseeable future, so a proposed order was cancelled.

It didn't quite work out like that. The long-distance fleet has ended up as all bi-mode, and Northern has ordered new DMUs. However a significant cascade is still planned, including an influx of older units for Northern. Delays to it have caused a number of other problems.

By contrast the MML electrification would not release a single regional/commuter DMU if today's service pattern continued, and even adjusting local services would only release a handful. Some HSTs would be replaced but the Meridian units providing the bulk of the service would have no obvious place to go to.

In principle this isn't the government's problem as the leasing company bears the cost if they can't find a user for their trains. However effectively engineering the off-leasing of a significant fleet of expensive units would leave the government open to legal challenge or at very least ensure that the risk of it happening again was priced into the rental fees for future fees.

 

[Senex]

By contrast the MML electrification would not release a single regional/commuter DMU if today's service pattern continued, and even adjusting local services would only release a handful. Some HSTs would be replaced but the Meridian units providing the bulk of the service would have no obvious place to go to.

But wouldn't all the rolling stock questions have formed a part of the benefit/cost analysis, with the MML still shewing a far better case?

Nevertheless, I agree that the stock question does look to have played a significant role in the choice of GW to go first, but how far was it also a political decision, with Devolved Welsh Labour being seen as far more important and influential than South Yorkshire Labour in the eyes of the Brown government?

Still, the East Midlands can happily put up with those ghastly 222s for a good few more years yet.

 

[LNW-GW Joint]

Network Rail's 2009 Electrification RUS hedged its bets on the BCR for GW.
It gave figures of "2+" to "positive" depending on the operational costs of the IEP fleet (which were not known at the time).
It was also based on Stockley-Maidenhead being funded by Crossrail.

The MML BCR was supposed to be "infinite" because over a 60-year period the cost of doing the electrification was less than the cost of not doing it.

Also remember this was at a time of sky-high oil prices, favouring the electrification BCR.

In the examples given in the text, TP's BCR was shown as 1.6, Manchester-Blackpool/Windermere was 0.7, and XC was an amazing 5.1 (assuming Colton Jn-York funded by TP).

 

[edwin_m]

But wouldn't all the rolling stock questions have formed a part of the benefit/cost analysis, with the MML still shewing a far better case?

The study quoted by LNW-GW Joint looked at each route individually, with some dependencies such as the assumption that TP was done before XC. It didn't try to ascribe any benefit to re-use of rolling stock on other routes, and it would have been virtually impossible to do so because it would have had to assume that the routes in question weren't being electrified as well and this would have resulted in far more options to be evaluated.

A wider business case could have been made for a specific package of electrifications and cascades, and this would have pushed up the benefits of GWML but made no difference to MML electrification where there is no cascade benefit to speak of. I don't know if anything like this was ever looked at.

 

[fflint]

How wonderful to see a page full of well researched information and reasoned argument. Hopefully this could set a standard for all threads.

 

[QueensCurve]

It's not simply due to incompetence on NR's part, it was working within the limits of the structure set for it. If anything the roots of the problems lie with Gordon Brown's decision to permit NR to borrow on the open market and the institutional structures set up to support it. This meant that NR floated away from the real world of costs and returns as more money could easily be borrowed - with the effect that checks on over-engineering, gold plating, or arguing about inappropriate standards, were not considered necessary.

Perhaps it is a little unfair to lay the blame with Gordon Brown.

Looking at the historical context, British Rail was constrained by the External Finance Limit which prevented borrowing in the markets to fund infrastructure enhancement. BR were forbidden from making infrastructure investments without the Government's permission. This frustrated electrification and other enhancements that might have been welcome.

The use of borrowing against the regulatory Asset Base by NR provided a method - albeit one described as "unsustainable" for funding investment in the infrastructure. During this period a number of substantial improvements were completed.

Now NR's debt is back in the Public Sector Borrowing requirement there is already the dead hand of the treasury putting the brakes on improvement.

 

[QueensCurve]

This may not be the right thread, but I have just caught up with More GOBLin closures as wires delayedin the current Modern Railways. This is another indictment of Network Rail's competence regarding electrification:-

Network Rail says the delay is due to incorrect designs to a number of the structures carrying the overhead lines meaning they couldn't be installed at planned locations. Late delivery of materials and structures has also been blamed for causing further delays.

Now that electrification is back under way, it is to say the least depressing that it is not being delivered effectively.:cry::cry::cry: