New Civil Engineer Civil engineering and construction news and jobs from New Civil Engineer
High Speed 2 (HS2)'s Euston station will be scaled back from 11 to 10 platforms, it has been confirmed.
In February, the Department for Transport (DfT) ordered HS2 consultants to look at scaling back the station's design to cut the construction time and save costs. Proposed design revisions included reducing the number of platforms from 11 to 10 and building the station in one stage instead of two.
In HS2 Ltd's six-monthly report to Parliament - published today - HS2 minister Andrew Stephenson said that the proposal for a smaller station will go ahead.
"At Euston, we’ve confirmed the move to a less complex, more efficient 10-platform design, which can be built in a single-stage, and can still support the full operation of the HS2 network," the report says.
"This will provide a more efficient design and delivery strategy and play a significant role in mitigating the affordability pressures recently identified. Moving to this revised HS2 Euston station design maintains the station infrastructure capacity to run 17 trains per hour, as set out in the Phase One full business case."
According to the report, the £0.4bn pressure on Euston cost estimates remains unchanged from Stephenson's last update to Parliament.
However, the report continues: "Now that the move to a smaller, less complex 10-platform single-stage delivery strategy at Euston has been confirmed (which will still support the full operation of the HS2 network), the DfT anticipates that cost pressures at Euston will be reduced as the updated station design is developed over the coming months."
In addition to this, a further £0.15bn pressure has been reported for delivering on-network works on the existing Euston network that are required to facilitate the new HS2 station.
Concerns about reducing the number of platforms have been raised by rail professionals, with one HS2 insider previously telling NCE that having one less platform gives the service less “wiggle room” should delays occur elsewhere on the line.
Rail consultant William Barter has also set out why 11 platforms are needed.
However, in its recently published Corporate Plan for 2021-2024, HS2 Ltd maintained that it was still “considering options to reduce the number of platforms at Euston from 11 to 10 while maintaining 17 trains per hour (tph) operations for the full Phase Two service”.
In the Corporate Plan, HS2 Ltd revealed that a revised concept design for the station is due to be complete by the end of this year, with a detailed design (RIBA 3) scheduled to be in place by March 2023.
Grimshaw and Arup won the Euston concept design job in 2012 but were replaced by WilkinsonEyre with WSP Parsons Brinckerhoff in February 2017. The original team was brought back for the detailed design work a year later, with more than £100M already spent on Euston station’s design.
A joint venture between Mace and Dragados (MDJV) remains in place as the station construction partner.
The HS2 platforms – for trains to the Midlands, the North and Scotland – will be 8m below ground, located in a concrete ‘box’ measuring about 90m wide by 500m long.
Enabling works at the site continue while the final designs are drawn up.
Like what you've read? To receive New Civil Engineer's daily and weekly newsletters click here.
William Barter’s earlier New Civil Engineer article has generated helpful feedback from the train time tabling community. Feedback has been overwhelmingly supportive of its main argument that HS2 has been designed for high capacity and high reliability. Euston must not be the critical bottleneck which limits the capacity of HS2.
It is theoretically possible to accommodate 17tph at Euston on 9 platforms leaving one deliberately spare – to accommodate late running or late departing trains as William advocates. However, there are three constraints which complicate matters
1. Two trains each hour (tph) need to be 8 coach trains arriving 3 minutes apart – so that 2×8 coach trains can then share one platform.
2. Ideally both sets of 8 coaches need to leave after 24 (instead of the usual 27) minutes turn around time – to vacate the platform for a 16 coach train arriving 30 minutes after the first set.
3. The coaches would need to head to the same destination by different routes, or alternate between the two northern destinations.
Thus without an effective 11th platform Euston offpeak capacity would be 15×16 coach trains and 2×8 coach trains. HS2 has probably identified 2 proposed trains which are unlikely to be more than 8 coaches and found it hard to dissuade HMTreasury that demand will be more than that.
Ten HS2 platforms is very shortsighted, hence the Euston Station developer ought to work out how HS2 can get access to one or two West Coast Main Line (WCML) platforms – once the Western Arm of HS2 has relieved the WCML. That may be possible for Classic Compatible rolling stock if some WCML platforms or approach track levels can be adjusted near Euston.
HS2 has been designed for high reliability – and if HS2’s design features do lead to high reliability we need to be capable of adding an 18th train each hour during the 2050s.
17tph has marginal reliability advantages, because it reduces the number of trains delayed behind each late running train by about a third, compared with 18tph. However, Euston & Old Oak Common, or the tunnels between, need to include passive provision capable of ultimately matching the expected 18tph reliable line capacity. Dropping the 18th path is being much too pessimistic about HS2’s design features.
I won’t go over all the arguments for 11 platforms again, as they have been clearly set out already for anyone who wants to understand them. But there are a number of highly contentious statements in the ministerial update to parliament.
First, he refers to the April 2020 Business Case service as consisting of 17 trains per hour (tph). This is true only in so far as destinations for only 17 tph from Euston are nominated, and benefits assumed from only those 17. But the service pattern diagram on page 133 of that business case clearly shows an 18th path, reserved to meet future growth. So at the very least, Mr Stephenson is abandoning this future-proofing in his quest for economies. He is entitled to do so, but not entitled to quietly forget about this path and hope no-one will notice.
It will not be possible to operate the 17tph service on 10 platforms to the same robustness as would be achieved on 11 platforms, the impact being either severely reduced levels of service reliability or frequent ad hoc terminations of late-running trains at Old Oak Common. My reasons for thinking this have been given previously; if Mr Stephenson thinks otherwise let him produce first the platform occupation plan for timetabling 17 tph on 10 platforms, then the robustness modelling that shows there is no adverse impact from loss of the service recovery capability offered by the 11th platform. It would be negligence if his decision is not supported by this evidence, so let’s see it instead of just having to put up with vague mood-music about ‘efficiency’. Again, he is entitled to say “to save money I have decided to descope the station, and I accept that something, either service levels or reliability, will be lost”; he is not entitled to lop off a platform and pretend that nothing will be lost, because that cannot be true any more than water can flow uphill.
I am afraid John Porter’s suggested changes to the station working are completely impracticable. Euston as planned with 11 platforms achieves efficiency through a standardised pattern of platform occupations supporting a sequence of movements in the station throat that avoids conflicting moves. Disrupting that pattern means either fewer trains or more operating functionality.
I accept it is better with an 11th platform, but still maintain it is theoretically possible to accommodate 17tph at Euston on 9 active platforms. I have allowed for 55 minute turnarounds for the Glasgow trains and 22 to 25 minute turnarounds on the others. 9 active platforms leaves one platform deliberately spare – to accommodate late running or late departing trains as William advocates. There now seems to be six (rather than the earlier 3) constraints recognised as complicating matters, which I am capable of meeting
1. The essential constraint to fit in an extra train, is that two trains each hour (tph) need to be 8 coach trains which fit into the SAME path 3 minutes apart and share one platform. These “2 trains” might cause timing problems elsewhere requiring tweaks to timetables.
2. If any train is more than 7 to 10 minutes late it might need to use the spare platform – or if it is an 8 coach train terminate at Old Oak (as their passengers are least likely to overwhelm the following 16 coach train).
3. Any “2 train path” needs to be deliberately chosen from standardised pattern of arrivals and departure, which has been optimised to avoid conflicts at station throats – such as those drawn up by William.
4. Ideally both selected sets of 8 coaches need to leave after 22 (instead of the usual 25) minutes turnaround time – to vacate the platform for a 16 coach train arriving 30 minutes after the first set.
5. The coaches would need to head to the same destination by different routes, or alternate between the two northern destinations.
6. Part of the timetabled recovery time might need to be moved for example to just after the “2 train path” to minimise the effect of “braking of one train knocking back on the next as they follow each other in” and the equivalent outbound problems. But being a “2 train path” points don’t need to be changed.
If those complicated constraints are met Euston offpeak capacity would be 15×16 coach trains and 2×8 coach trains, or 14×16 and 4×8 coach trains – without an effective 11th platform. HS2 has probably identified at least 2 proposed trains which are unlikely to be more than 8 coaches, satisfied the above constraints and found it hard to dissuade HMTreasury that demand will be more than that.
The easiest “platform share” I have identified would combine the planned Leeds arrival at xx48 and a Macclesfield arrival at xx51. The Leeds service could be 16 coaches if the Euston Station developer works out how HS2 can get access to one or two West Coast Main Line (WCML) 8 coach platforms – once the Western Arm of HS2 has relieved the WCML. I assume that would require 2 WCML tracks to be lowered in or approaching Euston and any throat issues resolved. Passive provision for such access is essential.
If WCML platforms for the Macclesfield service cannot be found, perhaps for only part of the day, the xx48 arrival would need to depart at xx10, immediately behind a slightly adjusted xx09 Birmingham service, using the Macclesfield rolling stock. Both services would need to be clear for the xx18 Leeds arrival to subsequently use the same platform. At those hopefully infrequent times eight coaches on both services might require neither service to call at Birmingham Interchange.
Similarly an 18th train seems to be feasible if HS2 operates a second “2 train” pair of 8 coaches to Euston each hour. That second platform sharing might require the xx25 Birmingham arrival to be the planned hourly service not calling at Birmingham Interchange, but only if WCML platforms cannot be found for the extra 18th train. There is a strong case for the 18th train to be an 8 coach Stockport & Wilmslow service arriving at xx21 – as both stations will be served by HS2 trains during Phase 2a and not all passengers will be able to transfer to planned HS2 Manchester Airport Station.
We need to remind ourselves that we are considering here potential problems 20 years in the future caused by changes which bring substantial savings in both cost and disruption at Euston.
There is an outside possibility that HS2 might have decided that the captive Manchester and Birmingham services can turnaround in 15 minutes – an EPIC assumption, which requires substantial detailed analysis as one just cannot mix and match turnaround times at Euston. Theoretically turnarounds are easiest to schedule if they are all the same or (as with the planned Glasgow’s) step down to the next possible departure.
After further discussion within the train scheduling community it is clear that my earlier ideas would work better if the proposed pairs of 8 coach trains had a standard turnaround. That requires the spare platform (needed for unusual occurrences such as an ill passenger) to be reallocated each hour. Because of that reallocation, there is likely to be a 3 minute period every hour when short “knock on delays” occur if the spare platform is in use during that 3 minute period.
HS2 is an important infrastructure project which will last for hundreds of years. I have tried to make 10 independent platforms at Euston work and have achieved several 90% solutions – which might work if lots of other things are altered. The capacity of the whole of HS2 should not depend on “mights”.
The scheduling community’s conclusion is that a self contained HS2 at Euston can handle 15×16 coach trains and 2×8 coach trains on 10 platforms with a 3 minute+ period every hour when short knock on delays after late starts etc may have to be tolerated. That maximum throughput uses HS2’s proposed paths through the Euston trackwork and points in the station “throat” and further north. It might require passenger facing crews to work Leeds – Euston -Macclesfield – Euston – Leeds 8 coach “shadow circuits”.
My earlier reply implied (but didn’t resolve) a conflict between a suggested rescheduled xx09 from Euston to Birmingham, xx11 to Leeds and xx13 to Macclesfield. Even if they could depart (and reach Old Oak) at that spacing, the last 2 would have about 1 minute longer journey times on the faster sections of HS2. In scheduling terms that early idea needs further work as it’s a fix that doesn’t quite meet the specification.
Those longer journey times wouldn’t be necessary if one doesn’t operate a service on the 18th path and the xx18 arrival from Leeds has a longer 28 minute (rather than 3 minutes shorter than standard 25 minute) turnaround. With this alternative, if the path elsewhere associated with the xx18 arrival is available, an 18th path exists but a platform is unavailable.
So to cater for the 18th path something has to be diverted to the WCML section of Euston. That need not significantly affect development costs or layout, but would require passive provision at this stage – by possibly adjusting track levels and keeping the fence line between HS2 & WCML clear of signalling equipment.
There is a strong case for the 18th train to be an 8 coach Stockport & Wilmslow service arriving at xx21 – as both stations will be served by HS2 trains during Phase 2a and not all passengers will be able to transfer to planned HS2 Manchester Airport Station. Such a Wilmslow service should be tried early and if less successful than expected, reduced in frequency to allow the 18th path to alternate to say Chester (once electrified). If successful enough the Wilmslow service could continue to my home town of Huddersfield (once electrified) – instead of its 30 minute connecting frequencies to HS2 via Manchester Piccadilly.
William Barter has shown that 18 trains per hour (tph) can be accommodated at Euston with 11 platforms and elsewhere north of Euston. That should remain the ultimate aim in 2050, but we need to find a compromise to get there. Meanwhile there is an even stronger push to save immediate cost by adopting 10 platform solutions for HS2 Euston.
William and I have set out in engineering and accountancy language some of the consequences of dropping the 11th dedicated platform. Euston is the key and it works best if it is largely self contained and the structure of the Eastern Arm services are retained/can easily be reintroduced.
Even if redesigning the Eastern Arm is proposed in the long promised Integrated Rail Plan (IRP) there is still a strong case for passive provision for access to one or two West Coast Main Line (WCML) platforms at Euston. HS2 has been designed for high capacity and high reliability everywhere. Euston must not be the critical bottleneck which limits the capacity of HS2 in the 2050s.
Most engineers want to evaluate what is being lost alongside the reportedly large cost and disruption savings at Euston.