APT - With Hindsight by Professor Alan Wickens


Introduction

Any reference to APT in the media is likely to be coupled with the phrase "ill-fated". That the project was not a commercial success is undeniable but this tends to obscure the many successful aspects of its research and development. It is in the nature of engineering that the translation of research results into application is fraught with difficulties, and this is well illustrated by the history of APT.

 

Background

The APT story started with some research carried out in the1960's. In 1962 Sydney Jones felt that there was a lack of understanding of the basic mechanics of railway vehicles - why did vehicles hunt and derail - an issue causing much mutual recrimination between the Chief Civil and Mechanical Engineers of the day. He therefore got the support of the CCE, Arthur Butland, but significantly, not the CME, to start some basic research into the dynamics of railway vehicles. The writer was fortunate to be recruited to build up a research group to carry it out. By 1964 a basic scientific understanding of the hunting problem had been acquired and, for the first time, it was possible to analyse and mathematically design vehicles that would be stable up to quite high speeds. This work was exemplified by the two axle research vehicle, HSFV-1, which was free from hunting on the roller rig at speeds up to 140 miles per hour. Incidentally, this vehicle still exists at Derby but after a long life of research work, including some crucial experiments, this significant vehicle should perhaps be in the NRM.

During the same period, the BR Passenger Business was showing that rail could compete with road and air, in spite of the negative aspects of the Beeching Report. In 1962 the accelerated East Coast Main Line service, using the Deltic locomotive, increased average speeds from 62 miles per hour to 75 miles per hour. Electrification of the West Coast Main Line to Manchester opened in 1966 with journey times reduced by between 20 and 30%. In both cases there was a positive response in passenger revenue, suggesting the famous rule that for every 1 mile/h improvement in average speed a 1% increase in revenue could be expected.

Further afield, in 1964 the Tokyo - Osaka Shin Kansen opened - another turning point in the renaissance of railways. Completely new infrastructure allowing speeds of 125 miles per hour or more could be justified by the enormous volume of traffic - of the order of 120 million passengers per year compared with roughly 6 million between London and Manchester.

With prevailing Government attitudes, and public environmental concerns, it was clear that for Britain, new tracks would be difficult to justify, and the existing extensive railway infrastructure would have to be exploited. This meant achieving higher speeds in curves. So, in 1966 a simple study on powered tilting of the car body to increase the allowable speed in curves for a given level of passenger comfort was carried out by BR Research. As passenger comfort, and not safety, was the factor limiting speed in curves, tilting had the same effect as superelevation of the track. Now tilting was not a new idea - Talgo had developed a pendular tilt system and in the US, United Aircraft were about to develop the Turbotrain. SNCF had run a pendular car in the late 1950's. Pendular tilt had serious limitations as it was slow to respond and required a large loading gauge. Powered tilt promised to overcome these problems.

 

APT - E

In 1966, it was decided to set up an Advanced Projects Group to help apply some of the results of the research work carried out at Derby. With the emerging background of high speed train activity, a research programme on a high speed passenger vehicle was proposed to Sydney Jones in November 1966. This envisaged a quite limited programme involving preliminary project investigation and supporting research of a trailer vehicle with

For experimental purposes, two trucks and a dummy body would be manufactured. It is noteworthy that, at this stage, the objectives were purely technical.

In 1968, Mike Newman joined to head up the project and design side, whilst Alastair Gilchrist headed the research and technical activities. A small team was then gathered. Mike Newman made the point that a single research vehicle would not answer many of the basic questions that were being asked about a tilting train. Were the weight estimates realistic? Could the tilt mechanism be put under the floor or was commercial space going to be used? And so on. Consequently, the scope of the project expanded to encompass research and development on an experimental train, and other test vehicles and laboratory experiments were planned. In fact we adopted a systems engineering approach, familiar in the aircraft industry, where the major design issues were identified and technical decisions made on the basis of calculation.

By February 1968 the project studies had developed to encompass a complete train. In addition to the features in the original proposal, gas turbine traction was chosen for the experimental train, as the weight targets could not be met with available diesel engines. The Rolls Royce Dart aircraft engine was a strong contender, but economics suggested a multiple installation of a cheap truck engine being developed by Leyland. The very arduous braking duty was to be met by a hydrokinetic brake.

Because of the Board's sensitivity about high speeds, Sydney Jones called the train the Advanced Passenger Train, neatly expressing the idea that the objective was not simply high speed but shorter journey times. The objectives for the train were

The objectives were now expressed in a form capable of translation into commercial terms.

In November 1965 Sydney Jones made a proposal to expand railway research. Though the then chairman, Stanley Raymond, supported the idea it was felt that the Board's financial situation prevented action but he gave approval for Sydney Jones to seek government support. Thus, during the years 1965-68, Sydney Jones and Kenneth Spring became familiar figures in the corridors of power. Seeking this support not only involved the Ministry of Transport who had a responsibility for railway research under the 1962 Act but the Government Chief Scientist, Solly Zuckermann, and various Cabinet Ministers. Success eventually came and it was decided that railway research generally would be expanded. The APT project would go ahead in late 1968, to be funded jointly by the Ministry and the Board on a 50:50 basis.

Over the next two and a half years a team was recruited (some, but not the majority, of whom were from the aircraft industry). Laboratories and test rigs were built, a test track at Old Dalby was commissioned and the APT-E train and various test vehicles designed and constructed. About 150 contractors built components and supplied parts with the final assembly at Derby. It is important to note that an important part of the project was concerned with the development of computer techniques in dynamics, aerodynamics, structures, etc. APT-E first ran on 25 July 1971 and was promptly blacked by ASLEF, the ostensible bone of contention being the single driver seat in the cab. The train was moved one night, with the help of a co-operative locomotive inspector, from the laboratory to the Locomotive Works where some needy modifications were carried out. This move precipitated a one day national strike that cost far more than the whole of the annual research budget. APT-E was blacked for over a year.

APT-E demonstrated that active tilting worked, that the tilt mechanism could all be put under the floor, that weight targets could be met and that a practicable high speed tilting train could be designed and built. It was not, nor was it ever intended to be, a practicable commercial train but a vehicle for engineering development. APT-E achieved 152.4 miles per hour on the Western Region on 10 August 1975 and demonstrated the value of tilt by running from Leicester to St.Pancras in 58 minutes 30 seconds on 30 October 1975. After an intensive test programme involving over 20,000 miles of running APT-E came to the National Railway Museum on the 11 June 1976.

 

APT-P

As with any research and development project, a major concern was how to take the project forward beyond APT-E, towards commercial application. In another industry, a new company might be set up or at least a self contained entity with all the necessary skills. This was not an option in the 1970's. With the support of the Board and the Ministry, it was possible, over a period of time, for Sydney Jones to discuss the delicate and difficult issues of responsibility with Graham Calder who had become CME in 1971. Actually, the organisation of the British railway industry, both operating and manufacturing, was not favourable to new product development or innovation. It is a tribute to Sydney Jones and Graham Calder that any way forward was found. It was decided to transfer responsibility for the prototype service trains together with the APT design group, to be led by David Boocock, and this was carried out in April 1973.

The project now had the support of the passenger business and some design work had been carried out on APT-P, to a business specification for the West Coast Main Line. This solved the problem of traction for the Leyland truck engine project had collapsed, no other gas turbine was viable and available diesel engines were too heavy. The specification was for a large train with 12 trailer cars requiring two power cars and hydrokinetic brakes to meet the highest speed envisaged of 250 km per hour. Because of the possible problems of current collection at speed if two pantographs were used, and because in Britain a 25kV line down the train was not allowed for safety reasons, the power cars were placed in the centre of the train.

There was a strong school of thought that at this point the technical risks should be reduced. A joint technical review by CMEE and Research, was carried out in early 1973, prior to the transfer of the project, having been suggested as a way of perhaps tempering the specification and reducing risk. For example, a modest reduction in maximum speed would have led to considerable simplification. This review was carried out, and the arguments were very finely balanced. The result was that the full specification emerged intact. With hindsight, this was unfortunate as the configuration was not flexible in commercial terms, and the hydrokinetic brake was to prove a technical liability. Whilst the support of the Passenger Business was vital, as of course they were the customer, it was a pity that the stability of a commercial specification was believed. It led to an inflexible configuration and increased technical risk. Commercial specifications are, inevitably, changeable according to business circumstances. Perhaps the specification of the train should not have been so influenced by the current perceived requirement, but aimed at meeting a wide range of possible requirements in a flexible way. The interaction between user requirements and emerging technology is always difficult, and leads to many mistakes in every industry. As an example, it should be remembered that the Spitfire was developed by engineers at a time when the customer, the Air Ministry, wanted a biplane.

In 1973 the Board was offered 80% of the cost of eight trains by the Ministry. An order for eight trains would have attracted vigorous participation by manufacturing industry. Unfortunately, after almost two years of deliberation the Board approved, in mid 1974, the construction of four trains. This number was subsequently cut to three trains by the Ministry in a round of public spending cuts. So, understandable financial caution had the effect of undermining the project. It meant that the prototype trains would be simply a further stage of development rather than a mainstream project, with yet another design to follow. By skimping the allocation of financial resources, the reduction in the size of the build programme gave a clear message about priorities. The cost of the APT-E programme was about £2.5 million in 1972 money - a considerable overspend over the original estimate but a modest amount for what was achieved. The cost of the APT-P programme is difficult to estimate but cannot be more than £40 million, including depots and other operational investments and probably underspent its budget. The small sums, in relation to turnover, spent on new technology by railways is in sharp contrast with some other industries. For example, between 1967-1976 Concorde cost £2 billion.

After a protracted period of design and construction, the first power car was delivered from Derby Locomotive Works, on time, in June 1977. The first trailer cars were delivered from Derby Carriage and Wagon Works, one year late, in June 1978. After another industrial dispute, proving trials of the first complete P-train started in May 1979. In December 1979 the first P-train achieved another British speed record of 161 miles per hour. The second P-train was completed in late 1979 and the third in 1980. After all these delays there was, understandably, great pressure to show that APT could carry passengers commercially. Though some technical problems emerged it was decided to put the trains in a limited service. On the 7 December 1981 a successful run was made to Glasgow in the scheduled 4 hours 15 minutes but thereafter the performance of the trains was marred by technical problems and adverse weather. There is no doubt that APT was put into the public timetable prematurely, with the result that there were embarrassing failures and the media had a field day.

Engineering development takes time and this haste was wrong for a development project. Fiat and ASEA (now ABB) took 20 years to develop tilting trains with tilt systems with the same specification as APT. Though both concerns had difficult periods with their projects both in terms of technical problems and organisational and managerial problems, their managements and customers took the long term view so vital for technological development. This long term strategic view was absent from the thinking of the Ministry and the Board at the critical time when it was needed.

The tendency at the time was to blame the technology. In reality, most of the problems were managerial, and the technical problems that existed were amenable to proper development effort. Unfortunately, during this period a major reorganisation of the CMEE Dept took place and the APT project team was disbanded in 1980, with serious effects on the project.

In 1981 the Board decided to get an outside view on the project as a whole. Ford and Dain were appointed and their report concluded that the technology was in general sound (with some reservations particularly on the braking system) but that the management of the project needed changing so that there was, once again, single minded leadership of the project. John Mitchell was appointed Project Manager and from then on rapid progress was made. By 1983, the trains were accumulating mileage satisfactorily and could have been the basis of development effort as has been carried out by Fiat and ABB.

Unfortunately, by this time the commercial requirements had changed completely. The APT was regarded as a train of a specific design, which the passenger business did not want, instead of a generic technology that could be applied in a variety of ways. And, so, much of the technology and experience was thrown away.

As a result of the persistence of Fiat and ABB the number of tilting trains now in service throughout the world is now approaching 200, in 9 countries. There are about six tilting systems available, though Fiat and ABB have most of the market at present. There is vigorous development of improved tilting systems, but not, of course, in the UK. However, at last, the benefit of tilting has finally been recognised by the recently announced orders for the WCML, cross-country and ECML services.

 

Conclusions

It is tempting to conclude from this story that APT was a disaster. It is possible to take a different view. As a research project it succeeded in generating new techniques that are being applied around the world, and there were many beneficial spin-offs.

But for the stimulus and competition of APT, HST would not have been in service as early. It might not have existed at all as its development was initiated as a low risk conventional response to APT. The research on dynamics is applied on a daily basis by train designers around the world. The original APT configuration of steerable single axle articulated train has become one of the leading edges of innovation in train design. The S-Tog commuter trains in Copenhagen have entered service, and a number of other single axle projects are underway. The ability to compute dynamic performance at the design stage, as pioneered with APT, is crucial to innovative projects of this kind. The work on aerodynamics made a vital contribution to the channel tunnel and to other high speed train projects such as the German ICE. The use of light weight Aluminium extrusions welded longitudinally has become a major manufacturing method. The Class 91 locomotive is a direct descendent of APT. The existence of APT forced Government to look at intercity transport in general and where development money was going. An interdepartmental working party reported in December 1971 on the comparative assessment of new forms of intercity transport. It compared viability of buses, APT, VTOL, STOL, and tracked hovercraft on the London-Manchester and London-Glasgow routes. The report was favourable to rail and not so to tracked hovercraft. Work on tracked hovercraft was subsequently cancelled by the Government.

The failings in the APT project were largely those of the environment in which engineering was carried out on BR. It has been said that there was a lack of strategic vision and inability to manage innovation. But in fact there was enough vision to fund research much of which was very successful. The structure of the industry, in which one organisation ran trains, specified, designed and built vehicles, supported by a manufacturing industry which was not encouraged to take technical initiative, stacked the cards against an innovative project like APT. Moreover, the engineering culture on railways had been to design, build, deliver and sort out the problems in service - a procedure appropriate to evolutionary and incremental changes to technology. APT involved a large change in technology and required sustained engineering development before application in service. The tilting technology itself was eventually made to work, and as Fiat and ABB and others have shown was well judged both in concept and specification.

© Professor Alan Wickens, formerly Director Advanced Projects, Research Dept. British Railways Board

Return to APT-P home page

 

Last Revised: 19.04.2008 13:37
by R G Latham
© 1998