Local Transport Today is the authoritative, independent journal for transport decision makers. Analysis, Comment & News on Transport Policy, Planning, Finance and Delivery since 1989.

Eighteen trains an hour on HS2 – whyever not?

William Barter Potcote Towcester NN12
12 June 2020

A lot of nonsense has been spoken about the ability of a high-speed railway to handle 18 trains per hour (tph), and House of Commons transport committee chair Huw Merriman may be about to increase it (‘DfT still wants 18 trains on HS2’ LTT 29 May).

The scepticism goes back to a statement by a representative of French state operator SNCF to the House of Commons transport select committee in 2013, that their limit at that time was regarded as 12 to 13 tph. This has been parroted many times by august and unbiased bodies from the House of Lords economic affairs committee to Birmingham University, but no-one, ever, asked what SNCF thought actually set that limit. And the one thing that clearly does not set it is high speed!

A proper investigation of the issue needs to be taken in stages. Does HS2 have the technical capability to handle 18 trains per hour? If not, there need be no further argument. But if it does have the technical capability, is it realistic in practice to run at that frequency? And, if it is realistic in one hour, is it wise to do it in continuously through every hour of the operating day?

So, technical capability first. Sure, as speed increases, so does braking distance, and therefore so does train separation. The time separation between trains is then increased by a variety of other factors such as system latency and safety allowances, so the relationship between speed and minimum headway is a J-shaped curve, in effect an inverted version of the speed : flow curve for roads that readers are no doubt familiar with. 

However, both arithmetic and modelling show that with ETCS (European Train Control System) Level 2 signalling and some realistic assumptions as to block length, braking rates, etc., although the minimum headway is achieved at about 200kph, even at higher speeds the theoretical headway can be as low as 120 seconds. Although in practice local features such as overhead line neutral sections will increase this a little, the technical capability to operate 18tph at 360kph is not in doubt.

The binding constraint on train frequency proves not to be the headway at full speed on plain line. Rather, it arises in station areas, but here the issues are common to any other railway and are not specific to high speed. 

The one constraint that stands out is likely to apply to any newly-built railway, arising from the likelihood that approaches to cities will have to be tunnelled, as is the case on HS2 into London and Manchester. These long tunnels will have ventilation shafts, and the requirement arising from fire precautions is to ensure that only one train can be between each pair of shafts, or between shaft and portal, at any one time.

This is enforced by aligning boundaries of signalling blocks with the shafts and portals. However, shafts both cost money and intrude on the built-up area on the surface, so spacings have been determined by potential rescue and evacuation needs, typically at intervals of 3.3km. This is about twice the length of a standard signalling block, so increases headways, but the effect only becomes material on the approach to a station where trains are on a braking curve to stop, so that the transit time of the last shaft-to-portal section escalates. 

Even so, headways can be contained to a worst case of 150 seconds, so long as everything else about the design is perfect, such as having alternate platforms at the station so as to build the dwell time out of the headway, and turnouts approaching those platforms that do not constrain train speed any more than does the natural braking curve to the stop. And 150 seconds with 18tph equates precisely to the guideline maximum usage of 75 per cent of theoretical capacity set out in the International Union of Railways (UIC) Leaflet 406.

But that, albeit the only objective standard we have, is a guideline, so we need to be satisfied that it can plausibly apply to HS2 in practice. An 18tph timetable would have a train every three minutes, with a number of spare minutes spread around the hour to achieve clockface departures – for instance, trains leaving Euston at xx00, xx03, xx06, then xx10. So there is buffer time between each train of between 30 and 90 seconds to absorb small delays. But what about bigger ones? The key risk on a mixed use railway is when a late-running fast train gets behind a stopping train, and can only trundle behind it for the rest of the journey. However, this risk does not arise on HS2 as all trains on the 18tph core route are running at the same speed. A train presenting out of order simply fits in, with the buffer time allowing consequent delay to decay over the next four or five trains. Of more significance is what happens to it on arriving at the terminus having missed its path across the station throat, but Euston as proposed in the Phase 1 hybrid Bill could be planned to leave one platform available for this and other contingencies. For more serious delays, HS2 will have a variety of service recovery strategies that the existing railway does not. For instance, a train handed over from the conventional network seriously late could be terminated short at Old Oak Common.

As to continuous operation of 18tph rather than that being simply a peak frequency, to some extent only time will tell. If necessary, an off-peak service of a lower frequency could be defined, but surely the logical approach is not to underuse the new infrastructure because we are afraid that the old infrastructure cannot present trains to it on time. Punctuality of trains is not absolute but is the outcome of choices that can be managed. If the effect is that the conventional railway receives further investment and other management action to allow the new railway to work to the full, than that is good for everyone and not just high-speed travellers.

All in all, the technical capability of HS2 to operate 18tph is not in doubt, and the capacity limit is not set by the maximum speed. Moreover, in practice it has buffer time and recovery options to enable reliable operation in practice. I would though plead with my engineering friends and colleagues, if designing another new railway with tunnels, not simply to slap ventilation shafts down at fixed intervals, but to space them equally in terms of transit time, closing up where train speed reduces. The Victorians knew very well that is how to space block posts under manual signalling!

Strategy Programme Manager – Strategy
Hertfordshire County Council
£34,788 - £42,683 + £2,000 market forces payment p.a
Senior Investment Planning Manager
Highways England
Manchester, Birmingham, Leeds
£48,155 - 55,379.00
Senior Investment Planning Manager
Highways England
Manchester, Birmingham, Leeds
£48,155 - 55,379.00
View all Vacancies

TransportXtra is part of Landor LINKS

© 2020 TransportXtra | Landor LINKS Ltd | All Rights Reserved

Subscriptions, Magazines & Online Access Enquires
[Frequently Asked Questions]
Email: subs.ltt@landor.co.uk | Tel: +44 (0) 20 7091 7959

Shop & Accounts Enquires
Email: accounts@landor.co.uk | Tel: +44 (0) 20 7091 7855

Advertising Sales & Recruitment Enquires
Email: daniel@landor.co.uk | Tel: +44 (0) 20 7091 7861

Events & Conference Enquires
Email: conferences@landor.co.uk | Tel: +44 (0) 20 7091 7865

Press Releases & Editorial Enquires
Email: info@transportxtra.com | Tel: +44 (0) 20 7091 7875

Privacy Policy | Terms and Conditions | Advertise

Web design sussex by Brainiac Media 2020