Readers of The Rail Engineer will be familiar with some of the sophisticated signalling systems in use on the national network in this country. The signal interlocking is a basic logic system that monitors every input from the signaller, checks for availability, and prevents a conflicting move being set up. Writes David Bickell
In the driver’s cab it is a different matter. Drivers are continuously reading the line ahead and adjusting the traction controls in accordance with all these visual inputs, including signal aspects, speed restrictions, platforms to call at and buffer stops at the terminus. Driving a train requires a high level of continuous concentration. Despite the dedication and professionalism of the driving fraternity, the occasional momentary lapse has the potential for serious consequences.
Back in July this year, the internet and TV news channels hosted video footage of a Class 730 train on the express route from Madrid to Ferrol lifting off the track into a catastrophic derailment on a curve at Santiago de Compostela in north-west Spain. This section of line uses the Spanish ASFA system of cab warnings supplemented with automatic braking. Signal aspect and speed data is transmitted to the train by means of balises in the track. Contemporary media reports indicated that the line speed was 50mph but the train derailed at 111mph. It was also suggested that additional infill balises may have averted the disaster.
Could this possibly happen in the UK? Sadly in the last 50 years there have been six spectacular derailments due to excessive speed resulting in a total of 19 fatalities. The responses to these and other significant accidents in this country have led to a significantly safer UK national rail network.
To monitor the variables on the line head and check that a driver is appropriately controlling the speed of the train is technically challenging. It is no surprise that technical developments to ensure that ‘driver error’ is nipped in the bud before any damage is done have lagged behind safety improvements in the signal box. Nevertheless, train cabs have, over the years, been increasingly fitted with a variety of safety systems. Invariably, the impetus to develop and implement these systems has been a recommendation from an accident involving loss of life. Such lessons of history are enshrined in the detail of RSSB Group Standards and Network Rail standards.
Automatic Warning System (AWS)
Sounds a warning in the cab when the train is approaching a signal at caution or danger. Also provides a warning approaching some speed restrictions.
A train travelling at 125 mph requires a mile and a quarter to come to rest. The big risk is that a driver missing a caution signal has no chance of stopping at the red and is heading for disaster. With the nationalisation of the railways in 1948, BR moved towards national implementation of AWS. The prototype design was ready in August 1952. Sadly, in October of that year, a three train pile-up at Harrow & Wealdstone killed 112, an accident which would have been preventable by AWS.
Roll-out took longer than expected and funding was allocated in the budget for the 1955 Modernisation Plan. The Southern Region (SR) was concerned about the value of AWS on its busy commuter routes where drivers would be running on successive double yellows and repetitively cancelling the warning.
Thus Signal Repeating AWS (SRAWS) was developed giving a visual indication in the cab of the signal just passed and next signal ahead. Unfortunately, it cost three times as much as AWS and was not proceeded with. Eventually, SR fitted standard AWS on all lines, though the 1978 London Bridge signalling centre area had to be retro fitted. AWS fitment of passenger lines was not fully completed until a few years into the 21st century.
The system consists of two magnets fitted on sleepers in the four-foot typically 180 metres on the approach to a signal. The first magnet in the direction of travel presents a permanent South pole magnetic field to the trainborne receiver. This primes the onboard equipment and changes the visual display to ‘all black’, if not already in this state.
Next comes the electro-magnet, energised only if the signal is showing green and presenting a North pole to the train. If this magnet is energised, a bell sounds in the cab (usually a ‘ping’ in modern stock) and no acknowledgment is required by the
driver. However, if the electro-magnet is de-energised, a horn sounds in the cab and the driver must press and release the cancel plunger in the cab. This causes the visual display to change, showing black and yellow spokes as a vivid reminder that the train is approaching an adverse signal aspect. The driver must control the train in accordance with the signal aspects observed. If the driver does not respond to the warning, the train brakes are automatically applied.
Standard strength magnets are coloured yellow. The AWS equipment on trains which regularly operate over DC electrified lines may be designed to be less sensitive in order to counteract the potential risk of misinterpreting magnetic fields generated by traction cables and bonds as coming from AWS track equipment. To ensure that the trainborne equipment of such trains still senses the magnetic fields of the AWS track equipment, green coloured magnets with a higher magnetic flux density are used on DC electrified lines.
AWS enhancements – speed restrictions
Following the disaster at Morpeth in 1969 (pictured below), AWS was extended to give warning of a significant step-down of permissible line speed. The Down ‘Aberdonian’ entered a 40mph curve at 84mph. There was a sense of deja vu in 1984 when the Up ‘Night Aberdonian’ took the 50mph curve at 90mph. Bungalows near the line narrowly escaped being demolished by the jack-knifed sleeping cars.
For this purpose, a warning is given in the cab for every train by means of a single permanent magnet associated with each permissible speed warning indicator. The criteria for the provision of a permissible speed indicator is that the approach speed is 60 mph or greater and the required speed reduction is one-third or more.
Similarly, temporary speed restrictions were added to the AWS portfolio after the 23:30 Euston to Glasgow sleeping car express was wrecked across the platforms at Nuneaton in 1975 after taking a 20mph temporary speed limit at 80mph. A small portable permanent magnet is installed for the duration of the TSR but has to be carefully positioned in accordance with a set of rules to ensure adequate separation of warnings with nearby signal AWS magnets.
AWS – snags
AWS has several weaknesses. Firstly it is an advisory system that does not monitor the drivers’ response to a warning. But it was never envisaged that a driver, having acknowledged the warnings leading up to a red signal, would then just drive on. Secondly, if a station stop intervenes between an AWS caution and the hazard (such as a signal at red), there is the risk that station duties cause forgetfulness. Station duties having been completed, the guard gives ‘ding-ding’ and the train starts away but the driver has not noticed that the signal is still at red. Also, it is possible to isolate the AWS and drive without protection (HST wrecked at Southall 1997).
There was a spate of serious accidents in the 1980s and beyond, caused by the ‘drive on’ and ‘ding-ding and away’ scenarios, so, in 1988, BR commenced work on a more effective system which checked that the driver was making the necessary brake application. Just a few weeks later in December, an horrific rear end collision occurred near Clapham Junction. One of the recommendations from that investigation was that BR should choose an automatic train protection system and implement it within five years. Experienced railway engineers and operators knew that five years was mission impossible. After all, it had taken half a century to complete AWS. A suitable ATP had not yet been invented and was going to be significantly more technically challenging!
Over the years AWS has doubtless reduced the number of accidents and it continues in service today as a primary driver aid.
Automatic Train Protection (ATP)
Continuously monitors the speed of the train, provides the driver with speed limit information, and sounds a warning if the driver is failing to reduce speed. If this is ignored the brakes are applied in time to stop safely.
BR introduced two trial schemes provided by different contractors to assess the technical issues and costs. The systems were fitted to HSTs on the Paddington to Bristol route and to Turbos on the Chiltern lines.
The ATP display in the driving cab is based on the speedometer. A green LED around the circumference of the speedo indicates the current speed limit. A flashing green LED shows a new speed limit ahead. A yellow LED denotes the release speed at which ATP relinquishes control. The driver must keep within the limit and reduce speed where a target speed ahead calls for a lower speed. A warning sounds if the speed is exceeded by 3 mph and the brakes are automatically applied at 6 mph or above. The trainborne receiver collects data transmitted from loops in the four foot on the approach to signals. Additional loops in between signals may be provided to give signal status updates to enable drivers to increase speed earlier if signal aspects ahead improve before the train reaches the loops at the next signal.
It became apparent that there were considerable technical difficulties in applying ATP to the national network given the age profile and variety of traction units in use. Also, a cost benefit analysis showed that, at a predicted cost per fatality prevented of £14 million, the scheme could not be justified and national application would not therefore go ahead. When this became apparent in 1994, BR and Railtrack jointly pursued a Signal Passed at Danger (SPAD) Reduction and Mitigation (SPADRAM). The two trial schemes remain in operation. There is no plan to extend ATP as the technology has been overtaken by the ETCS programme (see below).
Train Protection Warning System (TPWS)
Applies the brakes if a driver passes signal at danger, or approaches a red signal, speed limit or buffer stops too fast.
TPWS was a legacy of the SPADRAM project and came from an idea by BR Research which concluded that AWS could be enhanced by the addition of an automatic train stop and an overspeed sensor on the approach to the signal. This would reduce ATP preventable risk by about 70%, at a fraction of the cost, and in a far shorter timescale. Following the shocking Ladbroke Grove disaster in 1999 and media outburst there was considerable pressure to fast-track the project. By December 2003 all trains, over 12,000 signals, 650 buffer stops and roughly 1,000 permanent speed restrictions were fitted.
It was a significant achievement by project managers and equipment suppliers including Thales, Redifon, and Unipart Rail. Contracts were awarded by Railtrack and the UK Rolling Stock Owning companies (ROSCOs) for the supply of all the necessary trackside hardware and the provision and fleet-wide installation of trainborne equipment.
TPWS is designed to stop a train in three situations. At selected signals a train stop (TSS) will be provided at the signal and apply the brakes in the event of a SPAD. At selected signals an overspeed (OSS) facility will operate a red signal too fast. At other locations, such as on the approach to a permanent speed restriction or buffer stop, the OSS will apply the brakes in case of excessive speed.
A TSS consists of two transmitters (arming & trigger) mounted in the four foot. If the signal is at red transmitters are energised.
Receipt of both arming and trigger frequencies at the same time will result in the trainborne equipment making a brake application. An OSS consists of separately located arming and trigger transmitters set apart and at a distance before a signal as determined by line speed. The time taken by the train to pass between energised transmitters is calculated by the trainborne system which will apply the brakes in the case of excessive speed. OSS transmitters used for speed restrictions and buffer stops are always energised whereas TSS/OSS associated with
Sesigtena1ls become energised when the signal is at red and are fed from a nearby cabinet containing the signal controls. There are six track transmitter frequencies available, all within the range 64.25kHz to 66.75kHz. The specific frequencies are used in accordance with positioning rules to facilitate the interleaving and nesting of TSS & OSS on the same line.
Although TPWS will not prevent SPADs, it is designed to stop a train before reaching a point of conflict. Signals generally have a safety margin of typically 180 metres after the signal which is reserved until the train has come to a stand. This is known as the overlap. TPWS is effective at bringing trains travelling at up to 75mph to a stand before reaching danger. Additional overspeed sensors may be provided (TPWS+) to cope with higher speeds.
In 1975 a tube train failed to stop approaching the buffer stops at Moorgate station and careered into the wall at the end of the tunnel killing forty-three people. On the national system, the risk was addressed by adjusting the aspect sequence leading to a terminal platform. Originally, at many terminal stations, the signal authorising a move into the platform would display a green aspect. This was changed to maximum single yellow thereby ensuring drivers would receive an AWS caution before reaching the buffer stops. The provision of TPWS buffer stop OSS closes out the residual risk.
Driver’s Reminder Appliance (DRA)
Cuts traction power.
This was introduced into the driving cab from 1998 as a measure to try and reduce ‘ding-ding and away’ type SPADs. It is not connected to the signalling system and requires the driver to operate it in a proactive manner when a train is held at a red signal.
It consists of a switch on the driving desk which when pressed or pushed down, activates the device into ‘set’ mode. The switch displays a red light when set. This prevents the driver taking power. Reset is achieved by pulling up the switch/button.
Halts a train that passes a red signal.
Mechanical trainstops are used on lines over which London Underground trains operate and on a few other lines with a metro-style service. A trainstop is mounted near the signal and the arm is raised when the signal is at danger. When the tripcock on the train is engaged by a raised trainstop arm, a brake application is automatically initiated.
Tilt Authorisation and Speed Supervision (TASS)
TASS has been installed on some sections of the West Coast main line (WCML) and cross country routes to enable tilting trains such as Pendolino Class 390 and Voyager Class 221 to run at enhanced permissible speeds providing the tilt mechanism is functional. It also prevents tilting where clearances are restricted. TASS is a more sophisticated system than was originally provided for the tilting Advanced Passenger Train (APT) on WCML. C-APT (Control-APT) consisted of passive transponders in the four foot coded with the line speed. These were interrogated by the trainborne equipment to give the driver a digital speed limit display. Although development of the APT came to an end in the mid 1980s, some transponders are believed to still be in situ.
TASS uses Eurobalise passive transmitters mounted in the four foot which define the areas over which each train may safely tilt. Speed profiles and tilt authorisation data in ERTMS packet 44 structure is received by the trainborne equipment. The speed of the train is continuously supervised and enhanced permissible speed authorised providing the carriage tilting mechanisms are healthy. The cab ‘speed supervised’ blue light confirms tilt systems are working, balises are being read and the speed supervised according to the profile for the type of train and location. The driver may run at the enhanced permissible speeds that are displayed on lineside signs. TASS intervenes when overspeeding is detected, sounding a warning and slowing the train by 25mph before allowing the driver to continue as a non-tilting train.
European Train Control System (ETCS)
Provides cab signalling and train protection functions.
The Cambrian lines pilot scheme has previously been covered in depth in the The Rail Engineer. The national project is now moving ahead with a Hertford National Integration Facility (HNIF). Kit from several suppliers will be tested on a five mile section of the Hertford loop prior to national roll-out. A Class 313 EMU has been converted into an ETCS laboratory.
In addition to the various train protection systems described above, there are still some others used in the UK. Such systems are used on HS1, London Underground, and metro style railways/tramways in the UK. Perhaps they will form the basis of a future article in The Rail Engineer.