Add us to your taskbar by dragging this icon RGI logo to the bottom of your screen.

Close

UK maps a rational ERTMS strategy

01 Jun 2002

INTRO: A study by 50 train control experts concludes that premature installation of ETCS Level 1 could cost up to £6bn as well as cutting network capacity. As Richard Hope reports, the preferred option is to develop Level 2 fully over the next six years, allowing most key routes to be fitted by 2015

RAILTRACK'S train control strategy was plunged into confusion on March 29 2001, when the report of the joint inquiry into train protection by Lord Cullen and Professor John Uff set a rigid timetable for fitting the European Train Control System to lines where trains exceed 160 km/h (RG 5.01 p295). They wanted this done to prevent high speed collisions like those at Southall in 1997 and Ladbroke Grove in 1999, which killed a total of 38 people.

The Uff/Cullen report said the UK's West Coast and East Coast main lines should be equipped with ETCS by 2006, and other lines by 2008. Regulations should be introduced to make this a legal requirement. They insisted that any analysis to determine the cost-effectiveness of preventing injuries and fatalities through fitting ETCS should be disregarded.

It was immediately obvious that any attempt to implement the recommendations could be immensely damaging to Britain's railways, because of the impact on funding for other purposes, and the pressure on signalling resources.

While Railtrack's resignalling and train control strategy for the West Coast Route Modernisation already included ETCS Level 1 north of Crewe and Level 2 south of Crewe, where 225 km/h operation of Virgin's Pendolinos was planned, this project is currently being reviewed. It is likely that the planned timescale for fitting Level 2, in particular, will slip.

ETCS Level 1 provides 'bolt-on' ATP by adding a balise to existing lineside signals, and is currently available. But under Level 2, cab signals rely on the transmission of vital data by a radio block centre, which draws real-time information from the interlockings at signalling centres. Level 2 thus requires other key elements of the European Rail Traffic Management System, notably GSM-R cellular radio which is not currently developed for use on busy lines in Britain (panel far right).

ERTMS meets many objectives, of which safety through ATP - the sole concern of Uff and Cullen - is but one. Interoperability, higher capacity, reduced delays, and cheaper signalling through standardisation are also important. Elimination of lineside signals at Level 2, and ultimately track circuits at Level 3, should reduce risks to those workers who now have to maintain them.

Well aware that Level 1 bolt-on ATP reduces line capacity, and that Level 2 is still a long way from being fully developed for use in Britain, the Strategic Rail Authority and Railway Safety set up the ERTMS Programme Board, on which the rail industry is fully represented, including suppliers. EPB in turn appointed an ERTMS Programme Team (EPT), which published its Final Report on April 25.

Uff/Cullen v System D

EPT identified the five systems set out in Table I. They ignored Level 3, which provides moving block, for the reason stated. System E, sometimes referred to as ETCS Level 0, is broadly equivalent to the Radio Electronic Token Block used on some rural lines in Britain; its scope is limited.

All systems provide ATP, so EPT looked first at their impact on line capacity. This is crucial, given that passenger-km is already at the highest level since 1946 and the government wants a 50% increase by 2010, along with 80% more freight tonne-km. With timetabled train-km already 15% up on five years ago, capacity has become a major issue.

Because Level 2 Systems C and D require electronic interlockings to interface with the radio block centre, implementing Uff/Cullen would have meant relying mainly on Level 1 Systems A or B. EPT says this would reduce network capacity by 12·5%.

The issue here is infill. Under System A, a driver approaching a red signal protecting a junction where capacity is critical might see it clear to green from 1 km away, yet he must continue braking almost to a stop before the balise updates the ATP. Infill helps, but it increases capital and maintenance costs along with the risk to on-track workers because wire loops laid between the rails are prone to failure, thus degrading performance.

Because the capacity issue is so crucial, and economic benefits other than safety are necessary to justify the massive investment involved in ERTMS, the option chosen by EPT for their direct comparison with Uff/ Cullen is mainly based on System D.

Under System C, continuous infill is available by radio, but drivers still have to see the lineside signal clear before they can safely accelerate. With signals in the cab instead of on the lineside, the driver can accelerate as soon as the train is authorised to enter the next block.

But the choice of System D automatically implies: