Battery electric and battery hybrid trains

There has been much publicity over the last few months about the use of batteries as a propulsion energy source on trains. For example, in the UK, the first modern battery powered passenger train entered service on the extension from Kirkby to Headbolt Lane on 7 October 2023; Rail Operations Group’s Class 93 locomotives will feature a battery pack as well as diesel power and 25kV operation from the OLE; and Hitachi is testing a tri-mode conversion of a TransPennine Express Inter-City Express train adding a battery pack instead of a diesel generator on one car.

In addition, LNER has ordered 10 tri-mode (electric-diesel-battery) inter-city trains, the forthcoming Piccadilly line trains have a battery ‘get to the next station’ facility and, in Birmingham, some of the CAF trams have batteries to serve the extension to Edgbaston Village. All of these are hybrid trains/trams and there will be more to come. In Germany, Siemens has launched a fleet of battery hybrid trains (see later).

A recent Railway Industry Association (RIA) paper, which David Shirres reviews elsewhere in this issue, envisages a lot more bi-mode operation, and these trains will generally charge their batteries whilst running under OLE, although sometimes short OLE sections may be provided at, say, reversing points, to charge batteries during a train’s layover.

The report also sees a role for battery-only trains for some duties. For these applications, fast charging is all important and a system solution is required. The reason for this was amply illustrated when Simon Green, Great Western Railway’s (GWR) engineering director, addressed the Chartered Institution of Railway Operators, and Dave Horton, GWR’s chief mechanical engineer (Battery Train Fast Charge Project), addressed the Institution of Mechanical Engineers’ Innovation in Traction seminar (covered in full elsewhere in this issue). They both discussed the system issues that have been addressed in the development and proving work for the West Ealing to Greenford fast charge trial.

Background

In Rail Engineer 193 (Nov/Dec 2021), David Shirres reported on the three trains demonstrated at COP26. As well as the two hydrogen powered trains (which were not ready to be driven by hydrogen) on show, a Vivarail battery-powered conversion of an old London Underground D stock train carried invited guests on return trips to Barrhead. This had been the first electric (London Underground!) train to cross the Forth Bridge under its own power. This led to a proposal to demonstrate the train and its fast-charging technology on the West Ealing to Greenford branch line in West London.

Fast forward to the end of 2022, and, sadly, Vivarail entered administration. The DfT was still keen to see the trial continue and granted approval for GWR to purchase the fast charge intellectual property rights together with the prototype train, the three Class 230 DEMU 2-car sets made redundant from the Bedford to Bletchley line, and the remaining D stock cars in store at Long Marston. Getting a new system like this into service is far from trivial, but GWR approached the home straight when, on 18 March 2024, it announced it had started trials with the train and its fast charger on the Greenford branch.

Journey so far

No article about this project is complete without mention of the late Adrian Shooter who was the inspiration behind the project to upcycle ex-London Underground D stock which David Shirres described in Issue 186 (Sept/Oct 2020). His was the inspiration for the current battery configuration and the notion of the fast charge system.

The first incarnation was exported to the USA for ‘Pop Up Metro’ demonstrations. The idea was that batteries would replace the diesel generators on the train, feeding the existing Struckton Traction Control Unit (TCU) and acting as a sink for regenerative energy. Next was a train for a demonstration at COP26 in 2021, where the Vivarail team repurposed the original 230001 prototype into a three-car battery train with a modern interior.

It had been intended that two battery modules would replace the two diesel generators, but Adrian wanted three batteries, leading to the need to relocate the main air reservoir, as well as needing to find space for a new Auxiliary Battery Box, a new ‘Fast Charge Box’, three large DC-DC converters, and a lot of heavy duty cabling. The main battery modules contain lithium ion cells with a thermal management system to control the heat build-up during charging/discharging, together with creating optimum battery conditions in cold weather. Each module weighs 1.7 tonnes and is connected to its own DC-DC converter, the purpose of which is to step down the battery voltage (nominal 666V) to 48V for the battery thermal management system.

Part of the challenge was to deliver a means of charging the train during a normal terminus layover without overloading what might be a rural electricity supply. In simple terms, three elements are required. Clearly two of them are an electrical supply and a means to connect that to the train. The third ingredient is what is known as a Fast Charge Battery Bank (FCBB). The FCBB, in this case a 430kWh, 760V battery bank, is trickle charged from the electricity supply which might be a 400V, 63A three-phase connection. The fast charge aspect is that the battery can charge the train at 760Vdc at a rate of 1.8MW. We’ll get to why it’s not that simple later!

The train-to-shore connection is via carbody-mounted shoes connecting to current rails placed between the running rails. Clearly, there are all sorts of protections required to ensure that the rails are not live until the train is stopped and the shoes are connected and are proved to be isolated before departure, as shown in the adjacent diagram. Each retractable shoe has a carbon-copper pick up capable of carrying 1,000A and the conductor rails are aluminium with a stainless steel contact surface. In the four foot, one central positive rail is flanked on either side by a negative rail, so there are three rails in total. The shoes on the train are arranged such that the positive shoe is always in the middle and always makes contact with the central positive rail, regardless of the train’s orientation, hence there is no risk of polarities getting mixed up.

The shoe-contact mechanism was subject to extensive tests including a carefully developed test to simulate ice and snow. The following day it snowed for real!

As discussed earlier, the range on a fully charged battery depends on a lot of factors such as stopping pattern, load, gradients, and speed. The trial train has travelled over 110km under its own power. Initial testing has already shown that driving style has an influence on range.

This is a factor that is often discussed on metro systems but is all the more important when the energy source has limited capacity. In practice (as with metros) for a given run time, the best technique is to accelerate up to speed as quickly as possible and then coast as much as possible. In the case of the Class 230, which has AC induction traction motors, coasting is beneficial because it minimises energisation losses in the motors. With permanent magnet motors, such as used in most electric cars, this would not be such a problem.

Getting the train approved to run has been challenging as both the train and the charging system required approval, leading to an unusually complex process compared with what is normally in a train operator’s scope, involving both Safety Significant and Safety non-Significant (Common Safety Method terms) changes for both train and infrastructure. Where a Safety Significant change is involved the change proposer can, in addition to hazard identification/risk assessment, justify the safety of the change by reference to compliance with standards, by cross acceptance from another demonstrably safe system in use elsewhere or by explicit risk estimation. If there is a choice, the first is usually the most straightforward and the latter the least.

In the case of the fast-charging system, both fixed and train-based elements are all novel meaning explicit risk estimation was required as part of the process to demonstrate that risks were reduced so far as is reasonably practicable. In addition, there was a requirement to demonstrate that the train is compatible with the infrastructure over which it will run and with any other trains on the lines concerned (e.g., EMI) and the fast-charging equipment on the infrastructure had to be demonstrated to be compatible with all the trains that might run over it. Both track and train elements had to be demonstrated to be compatible.

The future

GWR has a relatively large number of self-contained branch lines. Simulations have been carried out by an engineering undergraduate from the University of Birmingham on work placement with GWR. Simulations for all the London/Thames Valley branches, based on a concept train with the second-generation TCU fitted to the SWR Class 484, showed that operation on all these branches is feasible. From this, availability of electricity supplies was evaluated on the conservative assumption that no connection could be made to the Network Rail 25kV supply.

This work showed that while there were bulk electricity supply points nearby, some were showing as ‘constrained’, meaning that there is a risk that the supply will need upgrading leading to a longer time required to obtain the supply.

Connection to the electricity grid is therefore a major constraint and even for the trial, it has taken two years to get a connection at West Ealing.

For future lines there are still 12 useable D-stock driving motor cars and 43 trailer cars in storage, plus the three ex-Bedford to Bletchley line units which could potentially be converted, or the system could be applied to other fleets, although some of the benefits of the comparatively light D stock cars (energy required from the grid, recharge time, etc.) might be lost.

Siemens BEMU enters service in Germany

The first Siemens Mireo trains entered service in 2020. These articulated electric multiple units are of a lightweight welded integral monocoque construction and have energy efficient silicon carbide power electronics. Currently, there are almost 1,000 Mireo EMU cars in service in two, three, or four-car sets.

At Innotrans in 2022, Siemens presented one of 27 Mireo Plus B battery electric multiple unit (BEMU) trains that had been ordered by Landesanstalt Schienenfahrzeuge Baden-Württemberg (SFBW) 2020. Under this contract, Siemens will maintain these units for 30 years with 100% availability. This is said to be possible due to the data analytics of the cloud-based Siemens Mobility Railigent X Suite predictive maintenance system.

After a lengthy period of testing and development, the first four two-car Mireo plus B BEMUs entered service on 8 April. These units have 120 seats, are 46.6 metres long and have a maximum axle load of less than 20 tonnes.

They operate at a maximum speed of 140km/hr although they are designed to operate at up to 160km/hr. Their traction power in both electric and battery mode is 1.7MW which gives the Mireo plus B an acceleration of 1.1m/s2 in both modes.

The capacity of the battery packs, one per car, can be varied between 400kWh and 600kWh depending on the duty required. Lithium titanium oxide (LTO) battery technology is used. For this application using 500kWh battery packs, the unit has a battery-only range of 120km with the battery unlikely to be more than 70% discharged after this distance. Batteries are charged from the 15kV 16 2/3Hz overhead line, regenerative braking and two charging stations at Achern and Biberach stations. Based on the thousands of kilometres testing to date, Siemens estimates that the batteries will last for at least 15 years.

The Mireo’s heating uses an energy efficient heat pump. Its air conditioning system uses propane as a refrigerant which is a sustainable refrigerant with thermodynamic properties that ensure a high efficiency.

Other orders for the Mireo plus B will see 31 units on the German East Brandenburg network and seven units in Denmark’s Midtjylland region entering service in 2024 and 2025.

When announcing the introduction of the trains into service, Albrecht Neumann, Siemens Mobility’s CEO for rolling stock, said that Siemens considers that the Mireo Plus B is now a mature technology.

What next?

The Siemens introduction has shown that an existing platform can be adapted to include a BEMU into the product mix. Equally, the GWR project shows what’s possible when adapting an existing train, although it might be challenging to interface the batteries to traction packages never designed to charge batteries. For new builds, all this will be considered but, if converting an existing train, engineers will need to assess the impact of adding several tonnes of batteries on its structural integrity, suspension, braking, and wheel/rail interface as well as traction system integration.

Rail Engineer has already heard about proposals to provide BEMUs on some of the diesel powered lines in the 750Vdc area in the south of England where, perhaps, some comparatively short sections of third rail might be provided to top up the battery charge during the journey. Using this process, Waterloo to Exeter, East Grinsted to Uckfield, and Ashford to Hastings are possible. In Merseyside, there is talk of the BEMU versions of the Class 777 units being capable of running as far as Wrexham, providing plenty of opportunity to extend the reach of the Merseyside electrics.

It seems likely that a very large proportion of the rolling stock expected to be ordered over the next few years – see ‘Unplanned Rolling Stock Procurement’, Rail Engineer 206 (Jan/Feb 2024) – will be BEMUs or battery diesel hybrids.

This article was based on presentations by Simon Green, GWR engineering director, and Dave Horton, GWR chief mechanical engineer (Battery Train Fast Charge Project), and a Siemens press conference when the Mireo plus B entered service.

Lead image credit: Malcolm Dobell