This year marked 40 years since the commissioning of the first conventional signalling Solid State Interlocking (SSI) at Leamington Spa on 8 September 1985. To mark the occasion John Slinn and Dr Alan Cribbens organised a celebratory lunch in Leamington, which was attended by approximately 37 S&T engineers. This included many who had been instrumental in developing, implementing, and promoting SSI.
Before SSI, the large power signal boxes which had become the norm in signalling systems contained large numbers of relays produced by a labour-intensive process. It was predicted they would rise in cost and become difficult to obtain as industry moved to solid state electronics. The relay-based power signal boxes needed large equipment rooms, and it was very expensive to maintain and service the relays.
The groundbreaking processor-based, electronic SSI was a replacement for the electromechanical relay systems. It reduced the space required and wiring requirements, while improving speed and reliability through a triple-voting processor configuration. Its success in the UK led to widespread adoption globally, paving the way for the modern, Computer-Based Interlocking (CBI) systems.
First use & development
SSI was actually first used at Dingwall in 1984 in connection with Radio Electronic Token Block (RETB) signalling, but Leamington Spa was its first use to control conventional signalling. The application design was also used in 1985 for the first phase of the Docklands Light Railway in London and other uses over the years have included in South Africa, Indonesia, Hong Kong, and Australia.
Before the lunch, John Slinn gave a short history of SSI. He explained that it was most likely that some of the work carried out on British Rail’s (BR) Signal Repeating / Southern Region AWS (SRAWS) project in 1972 germinated the idea of an electronic interlocking. John was at the Leeds Electronics Exhibition in 1977 when Alan Cribbens demonstrated a triple redundant clock, which laid the foundation for a software processor-based system to replace the large relay based interlockings.
In 1980, three reports on SSI produced by a working group were published. Part A described the planned pilot scheme, Part B detailed the effect on safety, and Part C described the effects of the new technology. Even today they make interesting reading, said John.
The development of SSI was delivered by tripartite agreement between BR, GEC General Signal (now Alstom), and Westinghouse (now Siemens). The success of SSI can be attributed to the faith and far-sighted nature of a number of people in the industry. These include Ken Hodgson (BR), David Norton (Westinghouse Signals), and Tom Cunningham (GEC General Signal).
An SSI steering group was chaired by Ken Hodgson, with members David Dobbs (BR Research), David Norton, David Glyde (Westinghouse), and Jim Waller (GEC). The working group was chaired by Roger Short and comprised of Bob Barnard (GEC), Chris Brown (Westinghouse), Alan Cribbens BR Research), and David McKeown (secretary, BR). Mike Furniss (BR Research), and John Corrie (Westinghouse) assisted. Bob Barnard added that David Stratton and Robbie Bowen were also involved in the GEC SSI development.
There were a number of justifications for SSI:
- Economy: The system offered savings approaching the maximum possible in three major areas – interlocking equipment, line circuits, and panel circuits.
- Safety: The safety techniques employed provided an adequate safety level in principle and built on the experience gained from other microprocessor systems entering service.
- Technology: The system was based on a second sourced microprocessor which was a widely used industrial standard and well supported with software development aids.
Reliability: SSI utilised a two-out-of-three redundancy architecture, whereby all safety-critical functions are performed in three separate processing lanes and the results voted upon. An SSI interlocking cubicle comprises three Interlocking Processors or Multi Processor Modules (MPMs), two Panel Processors, and a Diagnostics Processor (DMPM). An SSI system can operate on two MPMs in the event of the failure of one and it does not need the DMPM to function as an interlocking, as this only drives the technician’s terminal.
Success and legacy
Due to the success of SSI, Alstom and Siemens have released products over the years (Smartlock and Westlock, respectively) which use a number of its features. In particular, both re-use the SSI data preparation language and trackside equipment. The advantage of these later versions is that the arbitrary limit of 63 Trackside Functional Modules (TFM) is raised so that a big interlocking can be handled without having to split the logic.
SSI must be considered a great success said John, as who would have dreamed that after 40 years there would be 600 interlockings with over 50,000 TFMs and associated equipment still in service around the world? He added that even if 10% of these SSIs were replaced year-on-year, there would still be 20% of the systems still in service in 2040. The average calculated MTBF of the equipment is about 2.5 years, but in practice it is delivering on average 10 years MTBF.
John concluded by thanking all those involved with the development and deployment of SSI for their help, patience, and assistance. In particular, he thanked Jim Waller, Alan Cribbens, and Bob Barnard who appeared to have never been phased by his learning experiences.
There were many stories and archive information relating to SSI shared at the lunch, and apologies and memories from people who could not be there. This included Ken Burrage, former chief S&T engineer LM Region, director of S&T engineering BRB, and chief executive of the IRSE. Ken summed up the day in his note to Alan Cribbens: “Please do express my thanks and congratulations to you, John, and the SSI team, for the magnificent achievement and the invention, development, and implementation of BR’s first electronic interlocking technology, that is still going well after 40 years.”
After the event, Alan added that it was Armand Cardani of BR who first expressed faith in SSI and supported the idea. The concept of a microprocessor-based interlocking, the design and development of the hardware and software, scheme design language, data transmission system, trackside interfaces, and the novel means of achieving safety in SSI were down to the work of his team in BR Research. Mention should also be made of Harry Ryland, who was the major contributor to the design language and wrote most of the interlocking software.
Image credit: Paul Darlington