Which track to pick to boost rail safety?

A number of existing train protection systems could have prevented the Paddington crash – but each has its operational drawbacks

By Mark Venables

The horrific collision at Ladbroke Grove has reawakened the clamour for an improved safety system on the UK’s rail network. The current plan recommended by the Health and Safety Executive (HSE) is to introduce the Train Protection Warning System (TPWS) to selected junctions by the end of 2003.

The site of the Ladbroke Grove accident is one one of these junctions. According to the initial HSE report, TPWS would have prevented the crash.

However, the system is not the only solution – merely the most cost-effective alternative available.

Many believe another, more costly remedy is required, namely Automatic Train Protection (ATP). First Great Western uses this on its high-speed trains, but reliability problems and trouble obtaining spares has meant that it has only been able to keep 80% of its trains equipped with an active system. ATP is also operating sucessfully on various routes in Spain, Sweden, France and Germany.

To understand the difference between ATP and TPWS, we need to look at the fundamentals of railway system signalling. Each section of track is divided into blocks and is controlled by a signal placed some distance before the start of the block. Only one train is allowed to enter any block at a time. In the most basic system, a train passing through a green signal will turn that signal red, thus warning any approaching train to stop.

This basic system, called red/green aspect, is workable for trains that operate at speeds of under 50km/h, or metro systems. For higher speeds, advance warning is required. This has led to multi-aspect signalling, whereby the block behind the train is protected by a warning light showing amber. The UK has adopted a four-aspect system, whereby there is a pre-warning, or double amber, shown prior to the amber light. This provides early warning for high-speed trains and allows local trains to run closer together under double yellows.

In any system, human error can result in signals being passed at danger and some form of warning system is needed to deal with this.

In the UK, all trains are fitted with the Automated Warning System (AWS). In this system magnets placed by the track just before a signal sound a bell in the driver’s cab when a green signal is passed. If a yellow or red signal is passed a siren sounds in the cab, a visual warning is given and the brakes are energised. The driver has three seconds to physically cancel the warning before the brakes are applied.

But with the number of signals being passed at danger increasing something more robust is required.

TPWS is essentially an add-on to AWS and provides the additional safeguard of being able to slow a train if it approaches a signal too fast.

Each signal equipped with TPWS will have pair of loops, an aiming loop and a trigger loop, placed between 200m and 450m before the signal, and another pair at the signal. The loops at the signal apply the brakes if the train passes it, regardless of any action by the driver. The pair of loops before the signal determine the speed of the train and apply the brakes if the train is approaching a red at too high a speed. But TPWS cannot guarantee to stop the train before it passes the signal if it is travelling above 100km/h.

Another limitation is that it will only be installed for signals at junctions and it does not operate oncaution signals. Nevertheless, it is estimated that 60% of cases where danger signals are passed would be prevented by TPWS.

The more sophisticated ATP system uses information transmitted by trackside beacons just before each signal to the train’s on-board computer to control the braking. ATP uses two basic items of information: the speed the train is allowed to travel in this block and the target speed it must be doing by the time it enters the next block. From this information, ATP can gradually reduce the train’s speed through each block. Its drawback is that to provide a safe braking zone, ATP leaves an empty block behind each train, thus reducing track capacity.

A further development of ATP, the distance-to-go version, allows the empty block to be dispensed with. It requires the train’s speed to be continually monitored so that the train’s on-board computer can constantly recalculate the braking distance. Extra information from the track-mounted beacons is needed to update the train’s on-board map of the line to overcome the problem of the speed monitor being misled by wheelspin, skidding or wheels’ wear.

The next stage of sophistication would be to replace the fixed beacons and change to `communication-based train control’ with moving blocks. This uses continuous radio transmissions instead of fixed beacons, plus a more sophisticated level of computer control, and has the advantage that the size of the block can be varied depending on the speed and braking capabilities of each train.

Such a system is in development for the West Coast Main Line. However, the delay in opening London Underground’s Jubilee Line extension was mainly due to problems with a proposed transmission-based signalling system, which was eventually abandoned in favour of a more conventional system.