ATP-logo

Automatic Train Protection trials 1990 onwards

In January 1990 I became involved with the track testing phase of the introduction of the ATP Pilot Schemes on BR both on the WR main line (GWML) and the Chiltern Lines (CL). This phase was to last 'about six weeks'. In the event I was involved with ATP for over 6 years!

Click on the links

bullet

GWML

bullet

Chiltern

bullet

Locomotives

bullet

Demonstrator

bullet

The fate of the National system

bullet

The future of ERTMS

Great Western Main Line

89 HST power cars were fitted with the Belgian ACEC TBL system at Landore depot, Swansea. Initially running out of Bristol St Phillips Marsh (SPM) depot the HST trials on WR eventually settled into a pattern of running between Swindon and Bristol Parkway (Stoke Gifford) on the Badminton Line. This was the first part of the route to be fully fitted with the ATP lineside equipment.

The power cars were equipped with an antenna which received information from the track-mounted beacons and loops, a tachometer to record distance travelled, a driver's display to convey information to the driver and a data entry unit to allow certain parameters to be entered into the on-board computer as required. The computer was contained in a large enclosure and was mounted in the van end of each power car and was known as the Vehicle On Board Controller (VOBC).

Track equipment was original only fitted between Swindon and Bristol Parkway but later it was extended all the way to Paddington (on the fast lines only from Reading), although Paddington  station wasn't fitted until a lot later. In the other direction it was fitted past Wootton Bassett junction to Dr Days at Bristol. It was also fitted on the Berks and Hants line from Reading to just short of Newbury but this was only fitted in parts therefore allowing further trials of the system as it ran over track with and without ATP towards Newbury. This stretch also included a couple of level crossings at Thatcham and Midgham which were also being evaluated.

The track equipment consisted of an aerial (or beacon) located in the four foot at each signal plus in-fill beacons between signals in certain locations to enable information for the train to be updated more rapidly if necessary. These were supplemented with long loops of cable in rear of certain signals e.g. at junctions, to allow even earlier updates of signal aspect changes. These cable loops could be up to 1000m in length and the principle was introduced in order to keep traffic flowing.

The HST test train usually started from Swindon but would have come down empty from Paddington - three round trips between Swindon and Bristol Parkway Yard then we would bale out back at Swindon and the HST would return ECS to London for an evening departure from Paddington. Initially we were riding shotgun during the extensive driver training running but later we were dictating the test running to evaluate the system and to trial modifications etc.

As the on-track equipment fitment progressed the test trains ran over different parts of the route. Eventually Paddington station was fitted and this included buffer-stop protection. Trials for this protection system were carried out at Newbury Racecourse station where we could set up different scenarios without interfering with the main line.

By 1998, as a result of the original Class 332 Heathrow Express units being fitted with ATP from new, the relief lines were finally equipped throughout with the system. The signals inside the Heathrow tunnels have 'double beacons' as a safeguard against failure.

I remember stopping at the Posthouse Hotel on the outskirts of Swindon and later we combined the GW and Chiltern jobs and we stayed at Aylesbury (in the company of other colleagues for a change) and drove to and from Swindon each day.

Click on the image for a bigger picture - all pictures are the author's unless otherwise credited

Paddington line up

Three ATP-fitted HST power cars at Paddington in 1992

The ACEC system consisted of an on board controller (VOBC) mounted in a large metre-square box in the van-end of the power car. An antenna was fixed to the underside of the leading bogie which received signals from beacons positioned at each signal. An axle-mounted tachometer recorded distance travelled. Data was also entered by the driver about the train type and its braking performance, although most of the data for a standard HST was already programmed in.

Driver's desk Inside the cab the standard speedometer was replaced with an ATP speedo and a data entry unit was fitted - the keypad was behind the black panel visible on the right hand side next to the telephone handset.

Author's collection

Driver's display A close-up of the speedometer reveals the LEDs around the circumference and a number of additional displays for the driver's information together with four interactive buttons
HST Data entry panel The driver's PIN number together with 'Standard' data was entered by the driver via a keypad beneath a drop down door. Any other data related to any non-standard performance issues or train make-up could also be entered here as required.
VOBC The VOBC was mounted in the van end and had inputs from the driver's display, the data entry unit, the axle-mounted tachometer and the antenna - which was mounted on the rear of the leading bogie next to the AWS receiver.

Author's collection

Lineside cabinet The ATP electronics and interface with the conventional signalling for the lineside equipment were mounted in new cabinets with an attractive stainless steel finish. This equipment processes the messages which are transmitted to the trains via the track-mounted beacons and loops. An EPROM with information related to the location of the signal, the line speed, the distance to next signal etc would be fitted within. The signal aspect would be incorporated into this data. This EPROM could be replaced with another when introducing a temporary or emergency speed restriction as required (TSR or ESR)

Author's collection

  Chiltern Lines

Meanwhile on the lines between Marylebone, Aylesbury and Bicester, the Alcatel SEL system which was fitted to the class 165 fleet was being put through its paces.

Alcatel SEL had already designed and installed the successful LZB continuous loop system in use in Germany. The system in UK was a development of this and eventually became known as SelcabŪ. This system is installed in a number of locations around Europe, notably at each end of the high speed AVE line in Spain in addition to LZB on the main route.

39 Class 165 units, both two and three car sets, were fitted with the system at Aylesbury depot as they were being delivered. One unit was allocated as a test train (it had not been fully fitted with seating) and this ran between Aylesbury and Marylebone trialling the system.

The lines between Marylebone and Princes Risborough were fitted with the lineside equipment as was the single line between Princes Risborough and Aylesbury which included Marsh Lane open level crossing. The fitting of the system to the LUL line between Neasden Junction and Amersham was delayed as it was proving difficult to arrange possessions on LUL. Eventually it became impossible hence the whole Class 165 fleet was retro-fitted with tripcocks for use over the LUL section.

In a similar manner to the HST's, the units were equipped with an antenna which received information from the track-mounted loops, a tachometer to record distance travelled, a driver's display to convey information to the driver and a data entry unit to allow the driver's PIN number and certain parameters to be entered into the on-board computer as required. The computer was more compact than the ACEC type and was mounted in the cab beneath the 'secondman's' side desk.

Track equipment was fitted between Marylebone and Princes Risborough with a few 'gaps' south of High Wycombe but later it was completed throughout utilising the redundant equipment resulting from the inability to fit the LUL section.

North of Princes Risborough it was only fitted around Bicester as there were few if any signals on the single line between the two locations, although full train protection was maintained throughout.

The SEL track equipment consisted of a loop of cable located in rear of each signal. The cables were run with one side of the loop clipped to the sleepers in the centre of the four foot and the return ran along the rail foot (on flat bottom rail) or along the outer end of the sleepers (on bullhead rail). The loops varied in length dependent upon the local requirements e.g. longer loops were fitted on the approach to junctions etc. The advantage of the SEL system was that the loops had crossovers at pre-determined distances (usually about every 50m) which served to reset any anomaly in the tachometer's distance measurement thereby ensuring accurate location approaching the signal.

As the tests and trials were prolonged the original unit used as a test train returned to York to be fitted out and we used ECS units from Marylebone out and back from the sidings at Marylebone each day. The trains ran over all parts of the line except between Princes Risborough and Bicester as it was operationally very difficult to obtain a train path between services in the days of the single line.

We stopped at the Horse & Jockey hotel in Aylesbury later moving to the Holiday Inn-Garden Court just next to the local football ground. The work in Aylesbury was eventually to last nearly seven years - not bad for a job which was originally planned to last six weeks!

I think we ate in every curry house in Aylesbury over the years.
Inside Aylesbury depot

A brand new class 165 unit sits inside the brand new Aylesbury depot. The ATP system was installed and commissioned on site by GEC Signalling on behalf of Alcatel SEL after delivery of the trains from York Works

Author's collection

Driver's desk The cab display on the Class 165 was similar to that on the HST's with green LEDs at each 5 mile/h increment around the dial.

Author's collection

Data entry unit The data entry unit was placed above the driver's desk and worked on the same principle as the HST's but there were more occasions when data was 'non-standard' e.g. two or more units coupled together.
Ready to depart for Marylebone Aylesbury Station with a class 165 about to depart on a London service 

Author's collection

Mushroom The electronics and interface with the conventional signalling for the lineside equipment was mounted in small enclosures known as Lineside Electronic Unit (LEU) or more commonly 'mushrooms' . This equipment processes the messages which are transmitted to the trains via the track-mounted loops.

Author's collection

Locomotives

Perhaps not so well known was the fitting of the ACEC TBL system to two class 56 locos on GWML and the fitting of the Alcatel SEL system to two class 47 locos on the Chiltern lines.

The class 56 locomotives underwent several tests on GWML in November 1991 and again in January 1992 being based on Westbury depot and hauling various trains of differing lengths. The class 47's as I remember only ran light engine.

56039 at Westbury

Here 56039 stands at Westbury Depot on 10th January 1992 during trials on the Western Region. The depot facilities were very basic and I remember catching flu or something similar after a freezing cold winter's day mostly spent in the pit.

56 033 at Old Dalby 56033 is pictured near Widmerpool on the Old Dalby test track on 3rd June 1992 during wheelslip tests.
56033 Another shot of 56033 at Old Dalby

ATP Demonstrator

As an aid to spreading the word about the future ATP national system, the Project decided that a means of demonstrating the merits of both systems to a larger number of people than could be fitted into an HST cab would be useful.

To this end a class 114 2-car DMU, consisting of vehicle Nos: 55929 & 54904 and built at Derby in 1956, was acquired to be converted into a demonstration unit. The original passenger DMU unit (50015 & 56015) had previously been converted into a parcels car and on general withdrawal of the class this was selected and brought to RTC Derby. The vehicles were completely stripped out and equipped with  both the ACEC and the SEL systems. The unit was also fitted with a generator set, a conference room, fully equipped kitchen as well as forward-looking cameras etc. No expense was spared on the interior refurbishment and I recollect that the unit had some very swish fittings including wall to ceiling carpets and fire-proof curtains

The unit was completed in 1991 and renumbered as  977775 & 977776 and repainted into an attractive maroon & grey livery.

Unfortunately the underframe & control equipment did not receive the same attention and the unit was dogged with problems mainly due to the unreliability of its Leyland TL11 engines. Although the unit had run on both the GWML and the Chiltern line during its early days it was notoriously unreliable.

I remember several incidents, once failing at Gerrards Cross being dumped in the Turnback Sidings and having to be rescued by a class 47 which dragged us back to Aylesbury.

On another embarrassing occasion, one of the regular high-level meetings was scheduled to be the UK and it was decoded that the 114 should be brought down to Bristol Bath Road depot and the 114 would be used to house the meeting. This was to be followed by the group being taken for a demonstration run up to Swindon and back. unfortunately the unit never left the depot as it failed to start despite the attention of the technical support staff (us) for most of the day.

One problem with operating the unit as an ATP demonstrator was that its maximum speed was only 75 mile/h whereas the 165's could do 90 mile/h and the HST's 125. Hence a large number of speed restrictions visible on the faster trains were not 'seen' and displayed by the demonstrator as its display would show only the maximum unit speed of 75 mile/h.

After the demise of the ATP Pilot Scheme project the unit found a new life with EWS as a driver trainer based at Margam, south Wales. After a few years EWS withdrew the unit and it was stored until it was purchased for preservation by the MRT Railcar Group

It is now at the Midland Railway Centre at Butterley where it is nearing complete restoration to its original condition. Look here for a picture of one car in 2003 and the latest information about its current status.

The demise of a National ATP system

Time passed and the politicians who promised that 'money was no object' regarding safety systems following the Clapham accident sat on the fence and finally announced that there would not be a national system after all.

The next stage would have seen a national system borne out of the Pilot Schemes fitted to six new schemes, Merseyrail, Wessex Electrics, Norwich-Liverpool Street, Glasgow Suburban and a couple of others which I forget.

Although the two systems were having teething troubles it was evident that funding to continue the Project would not be forthcoming and interest in ATP gradually waned towards the end of 1993.

Eventually both systems were on the point of being abandoned but the advent of Railtrack in 1994 proved their saviour and today they are fully operational. The addition of the Heathrow Express service from Paddington in 1998 and whose trains were fitted with ATP meant that the Relief Lines from Paddington to Reading were then fitted and this enhanced the effectiveness of the system.

Chiltern Railways later doubled the line from Princes Risborough to Bicester and Aynho which was then fitted with the equipment as were the new class 168 units to further enhance the system. However, ironically, due to limitations imposed by future European Railway Traffic Management System (ERTMS) requirements, the line from Aynho to Snow Hill could not be fitted and hence only has AWS and TPWS.

ERTMS

One day both ATP Pilot Schemes will be replaced by ERTMS which will include an even better ATP system but don't hold your breath as will probably not happen until the present systems are due to be replaced - around 2020.

Some initial testing was carried out at the Old Dalby test track during 2003 and the present plans are to install a pilot ERTMS scheme on the Cambrian line in 2007 but this plan and timescale is looking increasingly at risk. The opinion of some in the industry is that TPWS is good enough but the European Union has a different view.

If you would like to know more about ERTMS and train protection systems on the British railway network have a look at the ORR site here

 Top

Back to main index

Back to Old Dalby page