Early next year Westinghouse Brakes will be launching the EP2002 distributed braking system for passenger vehicles, built around the company's new RBX4 valve. Operational trials on the München Metro have followed an extensive factory testing programme

Peter Johnson
Marketing & Sales Director Westinghouse Brakes (UK) Ltd

COMBINING mechanical and electronic components in a highly reliable and physically compact package, the innovative EP2002 braking system will be launched next year by Westinghouse Brakes, now part of the Knorr-Bremse Group. Building on the successful platform created by the EP98 family, it has been in design since 1996.

EP2002 is targeted at new-build passenger trains. It is intended to provide vehicle builders and operators with increased equipment functionality and lower installation costs. The primary aim was to improve performance and safety, whilst reducing the cost of the braking system.

An essential feature is the distributed functionality, with equipment mounted at the point of control and components linked by an electronic databus. Our objective was to produce a standard valve that could be adjusted to each customer's requirements, removing the need for application-specific components. Customers have been consulted at every stage, to ensure the valve has been designed to meet their requirements.

Distributed control

Westinghouse Brakes believes that distributed control is the future of train braking, and has been at the forefront of developing distributed control architecture. We see the key advantage of distributed control as enabling low-cost valves to be positioned at the point of control, connected together by a Local Area Network.

In a typical pneumatic brake system, this would allow control of service and emergency braking on individual bogies, with wheelslide protection provided on each axle. Individual valves can be controlled on a single vehicle, group of vehicles, or on a whole-train basis.

Distributed control offers several advantages. The first is a reduction in size. Combining the mechanical components, electronics and software into a single unit enables electronic control of functions that were previously mechanical, and allows unnecessary interfaces to be removed.

A second benefit is the lower total installed cost. Underframe piping represents a significant part of the build cost for pneumatic braking systems. Localising pneumatic valves reduces the length of piping needed.

The third benefit is the improved behaviour in the event of a failure. Grouping control on a per-bogie basis increases the number of autonomous cells, and so the effect of a single-point failure is reduced. Duplication of key elements such as databus links and brake management control units ensures minimum brake loss following a single-point failure, enhancing the overall safety of the brake system.

Future trends

Five technical trends have helped the development of EP2002. These are advances in mechatronics, databus control, safety traceability and safety integrity levels, as well as improvements in material design life.

Mechanical parts, electronic components and software have been supplied as physically separate units. To reduce packaging and interface costs, the recent trend has been to embed the electronics and software within the mechanical assembly. This is known as mechatronics, as seen in the RBX4 valve used for EP2002.

Databus development provides the ability to distribute 'smart' valves around a vehicle and transmit the control and diagnostic information. Reliable and relatively low cost systems have been available for some time. Brake management systems now in service incorporate features such as brake demand, friction/dynamic blending and cross-car blending, all controlled via a databus.

Safety traceability is also important. The increasing reliance on software for safety-related functions has caused concern within the industry. Recent changes to the Euro Norm standards reflect this concern, and specifically require demonstration of safety throughout the design process. Experience has shown that it is problematic, if not impossible, to apply the EN safety process retrospectively to an existing design, as claiming service experience in lieu of traceability is not permitted.

The improved definition of EN standard Safety Integrity Levels has had a significant effect on the cost of introducing software into safety-related functions. It has also changed the way that systems are partitioned, such that flexibility can be maintained within SIL0 systems whilst SIL3 or SIL4 can be achieved in high-integrity systems.

Improvements in materials technology are allowing careful selection and extensive testing to extend service life and time between overhauls. Maintaining required performance over the full service life of a vehicle requires detailed knowledge of the materials involved and their long-term characteristics. Incompatible materials can cause premature failures. As an example, the chemical reaction between certain synthetic oils and rubber components can cause degradation of performance or failure of the interface.

Smart valve

Central to the EP2002 architecture is the RBX4 smart valve. This merges full electronic management with a pneumatic control valve, giving a number of significant advantages. These include service and emergency brake controls, wheelslide protection, diagnostic performance monitoring and vehicle interfaces.

With the addition of an electronic card the smart valve is transformed into a 'gateway' valve. This enables whole-train brake management, adding further interfaces to related systems. The gateway valve also provides a brake management function, blending friction braking, dynamic braking produced by the traction motors and regenerative braking controlled by the traction package. The gateway valve has the ability to control auxiliary equipment, such as sanding, and has communications ports linking into the vehicle databus.

Further integration has been achieved by using the same pneumatic valve for service brake control and wheelslide protection.

Product testing

Safety was at the heart of the EP2002's six-year design and development programme, to support our objective of achieving a stand-alone product safety case. Westinghouse Brakes has invested heavily in testing facilities at its Chippenham plant, where the RBX4 has undergone extensive validation. The test criteria used are well beyond any normal operating profile, to ensure high service reliability.

Component testing was a key element in the programme, to ensure customer acceptance of the innovative technology in the new product and to prove its suitability for the harsh environment of train underframes. The valves have undergone highly accelerated life testing, combined environmental and reliability testing, ambient temperature life testing, type testing, and generic product construction safety testing.

Cyclic testing was based on metro trainset and multiple-unit applications. The purpose of the ambient life testing was to observe potential wear-dependent failure modes, allowing analysis and correction through design change, prior to the RBX4 entering commercial service.

Climatic testing looked at performance over an operating temperature range of -40°C to 55°C, to make the product suitable for use worldwide. Vibration tests representing the full service life cycle were performed.

Radiated emission and susceptibility testing was also carried out to ensure compliant operation in the electrically 'noisy' rail environment.

The RBX4 Valve has also been extensively validated on our wheelslide protection system evaluation rig. This tests performance across a range of brake actuator and adhesion profiles. We also used a specially-built five-car rig to test the complete brake control and management system. The rig can simulate train system interfaces including dynamic braking, databus control and brake management. By reducing the time required for on-train commissioning, this allows more extensive testing of the critical interfaces.

The first EP2002 system to enter passenger service has been fitted to a München U-Bahn trainset. The module incorporates two RBX4 valves providing independent control of cylinder pressures. An extended period of testing should provide operational data and confirm the performance of the RBX4 smart valve and gateway valve in the field.

  • Fig 1. Basic principles of Distributed Brake Control compared to the more traditional centralised control
  • CAPTION: Key to the EP2002 architecture is the RBX4 smart valve, which combines electronic management and pneumatic controls in a single modular housing