Is the end already in sight for Britain’s tilting trains?
TransPennine Express’s new trains are missing one key feature that is indispensable for other 125mph trains on the West Coast Mainline. GARETH DENNIS looks at the future of tilting trains.
A version of this article also appeared in Issue 886 (28 August 2019) of RAIL magazine.
Since 2002, tilting trains have enabled 125mph running on the curvaceous West Coast Mainline. By leaning further into curves to counteract outwards acceleration (more on the science later), these trains can reach speeds that conventional trains travelling on the line cannot.
Or at least that was the case until a few months ago.
You might not have noticed, but the first of TransPennine Express’s new Class 397 and Class 802 trains have been running up and down the West Coast Mainline (WCML) north of Preston at speeds of up to 125mph without any tilting capability at all.
This is exciting for a few reasons, but it also raises a few questions. How are these new trains managing without tilt? If they can do it, why do we need tilt in the first place? Does this mean the end for tilt on Britain’s railways?
Perhaps the simplest and most important question is whether this is good news or not.
I love seeing (or being on) tilting trains. Gliding through the Lune Gorge whilst elegantly rolling this way and that to keep speed up puts me in mind of the prototype Advanced Passenger Train (British Rail’s first commercially operating tilting train) making the same trip in the early 1980s. That glorious engineering marvel remains my favourite electric multiple unit despite its anticlimactic demise. There’s something very satisfying about a train tilting its way through tight reversing curves.
However, just like “bi-mode” trains are a compromise resulting from a lack of up-front infrastructure investment (in that case, overhead electrification), so too are tilting trains a compromise where an alignment is considered too expensive to straighten out (or bypass). Tilting trains are heavier and more complicated than their non-tilting equivalents, and so are more expensive to buy, operate and maintain. They require a series of balises to be mounted on the track, increasing infrastructure maintenance costs. They are also less roomy, which reduces the quality of the passenger experience.
On a strategic level, removing the need for tilt is very important. Given that High Speed 2’s “classic-compatible” fleet will most likely not have tilt, it proves the viability of their proposed operations beyond the limits of new infrastructure.
Until now, discussions about High Speed 2’s train service north of Manchester compared it unfavourably to the current long-distance high-speed services (LDHSSs), with suggestions that HS2 speeds and journey times would be worse once they joined the WCML. Now that TPE have shown that tilt isn’t necessary for fast timings, not only will this reduce the cost of any new fleet, but it will reduce the conflicts between the tilting and non-tilting fast trains operating through the North West and into Scotland. This avoids increased headways and the reduced capacity that conflicting speeds would result in.
But why exactly do trains tilt over in the first place? Let’s briefly look at railway curve design.
When a train goes round a curve, it exerts an outwards acceleration on the track — thanks to Newton’s second law, this is independent of train mass. Outwards acceleration is a function of train speed and curve radius (faster trains or tighter radius curves give greater outwards accelerations).
By raising the outer rail above the inner rail (applying cant, also known as crosslevel or superelevation), a component of the acceleration due to gravity results in an inwards acceleration. Inwards acceleration is a function of applied cant and track gauge (more cant or a smaller track gauge increases inwards accelerations).
Lifting the outer rail so that these accelerations are equal to each other (cancelling them out, in other words) gives us “equilibrium cant”, but to allow for variations in train speeds and to improve passenger comfort, the “applied cant” should always be lower than the equilibrium cant. The resulting difference between these two values is called “cant deficiency”, which is essentially a measure of the amount of unbalanced force through a curve.
It is important to note that limits of cant are defined by comfort and maintainability, rather than by safety. Trains have to run significantly above the design speed through a curve before derailment becomes a risk.
Engineers vary the amount of cant (and thus cant deficiency) based on the intended traffic on the railway. In the Crossrail tunnels or on HS2 where only one type of train with a very set performance will be operating, it is possible to optimise the cant alignment and carefully control lateral forces through curves.
On most of the rest of the mainline network, however, the railway has to be comfortable for trains travelling at the maximum permissible speed whilst not being hammered by stopping passenger trains or slower freight trains travelling at lower speeds. Excess cant can result in damage to track materials and an increased rate of track geometry degradation, both of which raise maintenance costs. The solution is to increase the amount of cant deficiency through curves.
On the WCML, where the speed differential between slow trains and fast trains is at its most extreme and curves are plentiful, the application of enough cant to enable 125mph for LDHSSs resulted in significant excess cant for freight trains and an unacceptable long-term maintenance liability. Hence the development and use of the tilting train, which essentially adds its own cant on top of that applied by track engineers.
Over the years, however, permanent way engineers have realised that maximising cant deficiency is actually beneficial for the operational railway, particularly in reducing the occurrence of rolling contact fatigue (gauge corner cracking, a type of rolling contact fatigue, caused the 2000 Hatfield derailment).
At the same time, the performance of stopping passenger and “slow” freight trains has increased. More electric trains and multiple units with faster acceleration, freight wagons that are capable of better ride quality at higher speeds, as well as the great leap in intermodal rather than bulk freight all contribute to a reduced speed differential between the slowest and fastest trains.
On the WCML, there’s another factor at play. Unlike the East Coast Mainline with its long straight sections and the occasional curvy bit, the West Coast Mainline has a speed profile like a sawtooth, with or without tilt in operation. Even the Pendolino isn’t particularly quick at accelerating out of curves and, in a few cases, drivers can’t make much use of the short stretches of higher permissible speeds.
This isn’t the case with the latest generation of electric multiple units. Both CAF’s Class 397 and Hitachi’s Class 802 (IET) trains have an immense rate of acceleration, and for a railway with lots of changes of speed, acceleration to make the most of the fast stretches is as important as the overall top speed.
Here’s a quick example: for various reasons, some recent Leeds-bound LDHSSs on the East Coast Mainline have been hauled by Class 90 locomotives with a top speed of 110mph rather than the Class 91 with its 125mph top speed. However, the gearing of the Class 90 means that it has a notably better acceleration than its sleeker cousin, and can thus reach 110mph more quickly. Spending longer at 110mph means that despite not reaching the full East Coast linespeed of 125mph, the Class 90-hauled trains only reach Leeds a few minutes behind their scheduled arrival time.
Whizz back to the curvier WCML, and the new trains with their better acceleration (and driver advisory systems that really make the most of their nippy performance) can get very close to the timings of the tilting Pendolinos.
TPE, CAF, Hitachi and Network Rail have undertaken extensive modelling to assess the requirements for non-tilting 125mph operation. The overhead traction equipment (not least the tension of the contact wire), signalling distances and vertical alignment (even a railway has to consider vertical accelerations to keep passengers comfortable and track materials intact) are 125mph-capable without any alteration.
Only curving forces present an issue, and having identified several test sections, CAF in particular have been running their Class 397s at the proposed speeds and ensuring that comfort in the passenger saloons isn’t impacted. Testing of this aspect of the new trains’ operation has shown that 125mph without tilt is comfortably feasible, and that the better acceleration of the units allows them to make more use of the short straight sections between curves. If all goes to plan, then the WCML north of Preston should be signed-off for 125mph without tilt in the middle of next year.
It remains to be seen if this will be extended southwards too, but given the reduction in costs that it would appear to represent and the potential benefits to passenger and freight operations, I’d be surprised if tilting capability (or the track-mounted enabling equipment) was retained anywhere beyond the life of the Class 221s and 390s.
[Edit: Given that the contract for Avanti West Coast’s new trains is for the same type of train as the TPE Class 802s, it is highly likely that this capability is going to be extended southwards.]
Within the route strategies for both Scotland and the North West, there are extensive plans for new, straighter parallel alignments to better regulate slower and faster trains — there’s no better way to improve capacity than extra steel, after all — but this is not expected to be delivered until Control Period 8 (2029–34) in readiness for HS2 Phase 2B.
In the meantime, some pragmatism from rolling stock and infrastructure engineers has gained a quick win for passengers. 125mph operations without tilt are happening as you read this issue of RAIL [edit: not passenger service though, that has to wait for the new speed signs to be put up, which looks like it won’t happen before the middle of 2020].
Was the development of tilt worthwhile? Of course. Should we mourn the loss of this clever kit? I don’t think so. In any case, the last tilting train in operation on our railway might well be departing sooner than you think.
