Eddy-current braking: a long road to success
BRAKING: Eddy-current technology offers the potential for frictionless braking at high speeds, but despite successful experience in Germany, concerns over interference and electromagnetic compatibility have discouraged other railways from adopting the concept. Jennifer Schykowski investigates
The idea of a truly frictionless brake has great appeal, and railways have been experimenting with the use of eddy-current brakes for many years. However, only Deutsche Bahn has so far adopted the technology for commercial operation in Europe.
Eddy-current brakes seem ideally suited for high speed railways, but in practice their application for both service and emergency braking has depended heavily on external factors such as the infrastructure specifications, notably those for the signalling and train control systems and the track structure.
Nevertheless, the use of eddy-current braking has now been authorised in France, where DB's ICE3 sets are permitted to use their equipment on LGV Est. Trials are also underway in both Japan and South Korea, with a view to introducing the technology on selected high speed lines. A quiet, wear-free and odourless brake offers many potential advantages for the operator and the passenger. These characteristics make the eddy-current brake potentially appropriate for wider application, particularly since the forthcoming TSI on railway noise will demand tight adherence to lower limits.
The eddy-current brake has its origins in France, where it is sometimes known as the Frein linéaire à courants de Foucault. This commemorates Frenchman Jean Bernard Léon Foucault who discovered the underlying scientific principle in the 19th century. Foucault observed that a higher force was needed to make a vertical copper disc rotate between two magnetic poles, and at the same time the copper disc warmed up. In simple physics, the movement of a metal plate in a magnetic field induces a voltage, which in turn creates eddy currents. Thus a second magnetic field is generated in opposition to the first, and the metal plate decelerates, transforming its kinetic energy into heat. The better the conductivity and permeability of the plate, the stronger the braking force.
Whereas some inventions are rapidly adopted, others take longer to become generally accepted, and this was the case with the eddy-current brake. It was nearly a quarter-century after Foucault's death before the first patent was issued in 1892. Today the concept is applied in many different fields, such as rowing machines, motor test stands, roller coasters or free-fall tower rides in amusement parks. These are generally fitted with permanent magnets, which are used as a service brake and can generate a force of up to 1 000 kN.
This practical experience in other fields should encourage the application of eddy-current braking to railways, where it offers a number of advantages compared to conventional friction brake systems.
There is no mechanical contact between the brake and track, as the magnetic field operates across an air gap between train and rail. Thus it is wear-free and silent, requiring minimal maintenance. The braking force is independent of the coefficient of friction, ensuring high efficiency regardless of wheel-rail adhesion, for example in damp conditions or on wet leaves. This means that relatively high braking forces can be applied, which remain almost constant, even in high-speed applications. The braking force can be accurately controlled by regulation of the magnetic field. Kinetic energy from the train is absorbed by the rail and converted into heat.
Eddy currents are induced by movement in a magnetic field, which means that eddy-current braking cannot be used as a parking brake. Retardation is dependent on speed - the faster the train, the greater the braking force, subject only to the intensity of the magnetic field. This allows the brake to be finely controlled, as the magnetic field is created using electromagnets fed from an external power supply (Fig 1). The braking force of the eddy-current brakes fitted to the ICE3s is around 21 kN per unit, giving a total of 170 kN for a trainset with eight brakes running at 200 km/h.
To date, DB has equipped 67 ICE trainsets with eddy-current brakes. Four trailer cars in each eight-car set are fitted with a brake unit on each bogie, giving in total 536 brakes on 268 vehicles. By January 2008 DB could report on experience with approximately 117 million train-km of eddy-current brake operation.
The equipment has been in use for emergency braking since 2000, but only on authorised routes where the signalling and train control systems permit it. Service braking began on the Köln - Frankfurt Neubaustrecke in the summer of 2002, and since 2006, the brakes have also been used for service applications on the Nürnberg - Ingolstadt line. Both routes have LZB cab-signalling and are laid with ballastless slab track, which gives a stronger infrastructure to resist the heat in the rails generated by the braking forces.
When the trains are not operating on an approved section of line, the eddy-current brake must be deactivated by the driver. Extension of service braking to other routes is being considered, but this will require a detailed assessment of the infrastructure and other compatibility issues. The first commercial application outside Germany is on LGV Est, where the ICE3s are now certified to use their eddy-current brakes for service braking on the whole of the new line between Baudrecourt and Vaires. This is subject to a braking force limit of 105 kN, due to a perceived risk of interference with the TVM 430 train control system.
For rail applications, the braking electromagnets must be arranged in a linear, rather than rotary, alignment. Above each rail, eight magnets are fastened to an integral carrier, with an alternating sequence of north and south poles. Crossbeams connect the two magnet sets on each bogie to form a single assembly, which can be raised or lowered by a ring bellows (below).
According to manufacturer Knorr-Bremse, and DB's practical experience, the air gap between the magnets and the railhead should be between 6 and 7 mm when the brake assembly is lowered into its operating position. An easily-accessible adjustment screw is provided on each carrier, and DB depots are provided with a simple gauge to measure the air gap. In practice, the railway has found that adjustment is needed every 144 000 km.
In order to prevent inadvertent mechanical contact between the solenoid coil housings and the track, damage to the magnet, or dampness leading to corrosion, the brake housing is fitted with an energy absorbing guard plate and sealed with a synthetic resin. This has proved sufficient to avoid the worst damage and most operational disturbance.
Total weight of the ICE3 eddy-current brake assembly is 860 kg, to which must be added the weight of power cables and the control module. This relatively high weight is a result of the brakes being a late addition after the bogies had been designed; Knorr-Bremse believes that there would be a weight saving of around 15% had the eddy-current braking been integrated into the bogie from the outset.
When DB first experimented with eddy-current braking on the ICE-V test train in the 1980s, it discovered that one of the biggest obstacles to commercial application would be electromagnetic compatibility with signalling and train control systems. This could result in interference or potentially even irreparable damage to the signalling system by the electromagnetic forces generated when the eddy-current brake was activated. Even when not in use, there was a risk of physical interference between the brake and signalling equipment when the brake assembly was in the lowered position.
In total, around 70 components were identified as requiring modification before the eddy-current brakes could be applied in commercial service. Amongst these were axle-counters, level crossing actuators, point machines and hot wheelset detectors. As a result, DB Netz adopted a policy of excluding vulnerable components from its new line or infrastructure renewal projects.
Additional tests were conducted on the Köln - Frankfurt line before it opened, which identified further potential problems, including incorrect solenoid coils or pole sequence errors, and the risks associated with having too small an air gap between the lowered eddy-current brake and the track. In particular, even with the brake inactive, the magnetic field strength was only just below the maximum permitted level which could result in incorrect wheelset detection by the axle-counters. All these problems had to be addressed before the brakes could be put into operation.
Every magnet had to be examined for signalling compatibility at the Knorr-Bremse factory before it was dispatched, and DB conducted further tests on the completed brake assemblies after they had been installed in the ICE3 trainsets. Feedback from this intensive programme enabled Knorr-Bremse to develop the individual coil design in such a way that the eddy currents do not interfere with the type of axle-counter most commonly used on the German rail network.
Another issue that had to be addressed is the heating of the rails as a result of repeated brake applications, and the effect that this might have on the track structure, turnouts and other critical elements such as bridges. For operational and signalling reasons, most brake applications are made in much the same locations for each train, so the train frequency and normal headways on any given route must also be taken into consideration. If trains are braking in the same place too frequently, there is a risk that the rails will not be able to cool sufficiently between brake applications.
DB Netz is currently investigating ways in which it can monitor rail temperatures before and after the passage of each train, and considering how this might be used to manage the use of eddy-current brakes. It is also examining the critical points along each of its main lines to see which other routes can approved for eddy-current service braking.
Moving forward carefully
After the eddy-current brake had been put into regular commercial operation on the Köln - Frankfurt line, DB began to look at the costs and benefits. A detailed three-month investigation in 2002 established that the use of the eddy-current brake for service applications resulted in a substantial reduction in the wear of the trains' conventional brake disks. At present DB spends around €2m a year replacing worn brake linings on the ICE3 fleet alone, so there are obviously significant potential savings if the use of eddy-current brakes can be extended.
According to DB, the cost of the eddy-current brake can be recouped over seven years, thanks to the savings in operating and maintenance costs. Life-cycle costs over a nominal 25 years are expected to be little more than half of those for conventional disc brakes.
It seems fairly clear that the use of eddy-current brakes will be expanded in Germany over the coming years. DB has already decided that its next concrete project will be to modify the Stuttgart - Mannheim line to permit its use in ?regular service.
Clearance to use the technology elsewhere in Europe is less certain, as the same interference and compatibility issues also arise. A detailed examination of each route would be needed, and in some cases infrastructure components may have to be changed. It is perhaps instructive to note that apart from LGV Est, there has been no attempt yet to introduce eddy-current braking on any other routes in Europe's expanding high speed network. Nevertheless, regulatory provision has been made for this in the infrastructure, rolling stock and operation sections of the High Speed TSI.
Further afield, eddy-current disc brakes have been used in Japan since the 1980s, and Mitsubishi and Toshiba are now undertaking tests with eddy-current rail brakes; Hyundai-Rotem is also experimenting in South Korea. It is not clear when, or if, these projects will lead to an introduction into commercial operation, but with the rapid expansion of high speed lines in Asia there is clearly considerable potential for a wider application of frictionless braking in this market.
- DB's ICE3 fleet uses eddy-current braking in service on the Nürnberg - Ingolstadt Neubaustrecke.
- Close-up of a brake assembly, showing the electromagnetic coil and the air gap between the energy-absorbing guard plate and the rail head.
- Fig 1: Principles of linear eddy-current braking.
- An EWB 154 R linear eddy-current brake assembly, showing on the left the hexagonal screw for adjustment of the air gap.
'We are in agreement with SNCF that eddy-current brakes should be fitted to the next-but-one generation of TGV'
Dr Stefan Haas Managing Director, Knorr-Bremse GmbH