scotland (4K)
The Friends of the Far North Line
Cairdean Na Loine Tuath
the campaign group for rail north of Inverness - lobbying for improved services for the local user, tourist and freight operator

Could Centralized Traffic Control signalling create extra capacity on the FNL?

The FNL is currently signalled with a radio electronic token system. Originally installed as a cost cutting measure, it probably kept the line open when closure was a real possibility. Now the system is an obstacle to the efficient use of a much needed resource.

However the system which replaces it needs to be appropriate to the needs of the line. The FNL and the Kyle line are unique in Britain. Single track, some higher speed sections and now attempting to serve commuter, long distance and freight needs it has specific problems which need to be addressed. Solutions best suited to high density networks may not be either efficient or economic in the Highlands.

Centralized Traffic Control is usually associated with freight operations in the United States. Sometimes thought to only be used on slow, un-signalled lines (dark territory as the Americans call it) CTC is in fact often used with track circuited signals, on lines with speeds over 80 mph. and both passenger and freight operation, but is used in conjunction with track warrants where track circuiting does not exist. (Where no track circuits and signals exist the system is often called Direct Traffic Control)

In CTC a Dispatcher (a similar role to the signaler in the RETB system) authorizes a particular train to enter a section of track. Where track circuiting exists signals protect the occupied block. Track warrants are used to authorize movements on unsignalled track and the dispatcher can record these on a computer system which protects against conflicting movements. So far this may not seem much different from RETB.

Where Track Circuit section blocks exist the Dispatcher has no control over the signal indication given. Trains can only enter a section of track protected by track circuits when they have both AUTHORITY from the Dispatcher AND CLEAR SIGNALS. The track circuit can control multi aspect signals if traffic requires.

The critical difference from the token block system is that a track warrant can be issued to a particular point within a single track section, usually a milepost, when the preceding train in the same direction has cleared that point. British practice has been against procedure like this, I believe, because a Dispatcher can only rely on train crew radio reports to locate the train. American rail systems are beginning to experiment with Global Positioning Systems (GPS) to support their operations. Some simply propose to literally locate equipment which otherwise can get "lost". But there are opportunities to use GPS as part of a signaling system. It is the potential to link a CTC system with GPS train location that offers major improvements to the FNL.

GPS can locate a train within a few metres. It is not sensitive enough to tell which track a train is when tracks are adjacent but it can say where a train is on a stretch of single track line. Trains would be fitted with two unique GPS units to operate in normal service (permanently fitted in the case of Class 158 units normally assigned to the Highlands). Additional hand held units would be available at Inverness and fitted to freight and excursion trains as they entered the FNL territory. One unit would be attached to the cab and one to the rear end of the train. The GPS units would be associated with a unique train number (normally the locomotive number)

The GPS unit numbers would be programmed into the Dispatchers computer as the train came into service. The unit would continually report its position by dedicated radio link (probably backed up by a parallel report using the mobile phone network). The dispatcher would input Authority to occupy a section of track into the computer program and advise drivers of that authority.

Between each crossing loop the single track is divided into a number of GPS section blocks. These blocks would only exist within the Dispatchers computer and be marked with trackside boards to aid the driver. Block lengths would be determined by the braking distances of the normally operated trains at their authorized track speeds but would probably be 1-2 miles. The computer would display the train's location, as reported by the GPS system, on the dispatcher's computer, it would also record where Authority has been given to train. If a train is in or has been authorized to enter a GPS single track section, say, southbound, Authority can not be given to a northbound train to leave a protected crossing loop into that section.

When a train has cleared the first section of GPS monitored track the computer system would permit a following train to be Authorized. The computer would monitor whether the second train was catching the first and is so would send an automated radio warning to the second train. The existing TPWS (Train Protection and Warning System) equipment that prevents a train proceeding from a loop without a token can be adapted to provide protection. The computer would only allow a second train to proceed if the first is a sufficient distance away.

Crossing loops. To leave the crossing loop it would need authority from the Dispatcher. As funds become available, and traffic requires, crossing loops could be protected by track circuits with local signals. A train not stopping at a loop could be preauthorized to enter the track beyond the loop (the authority would be entered into the computer). The driver would only need clear signals in order to proceed. If the wrong train entered the further single track section the existing train stop equipment would operate.

Sidings. Where a train is required to leave the main line for a siding the crew would radio the dispatcher to confirm that it is clear and the Dispatcher would enter that information into the computer which would then permit Authority to be given to another train. So what are the advantages of such a system?

Firstly it allows swift occupation of the line by a second train. For example a morning train from Wick could meet an Inverness commuter train terminating at Tain. The Wick train would then run Fast to Inverness (perhaps stopping at Dingwall) but would be closely followed by the stopping train to Inverness. This would enable additional platforms for such stopping trains to be built between Inverness and Tain without needing crossing loops, improving the Invernet system. Such a second train could be a freight, making creation of freight paths more flexible. By scheduling slower trains to follow faster trains signaling conflicts are unlikely to arise. Secondly. It allows earlier access to the track for maintenance and for activities such as timber loading on the main line. Thirdly it allows signalers to advise crossing users more accurately of the location of trains to allow safe use. System Failure. In the event of a computer failure the dispatcher would revert to using Track Warrant Control by radio. Each train movement would require specific authority from the dispatcher to move from one location to another with drivers reporting when they were clear of their authority. In reality not much different from the radio token system in capacity but relying on staff following the rules. Each train would carry a spare GPS Unit for use if one of the GPS units failed. In the event of two GPS units failing special procedures would apply to report that the whole train had cleared a section.

Costs. New systems currently being tested are unlikely ever to be economic on lines such as the FNL unless installed for test purposes. The system proposed uses existing radio, mobile telephone and GPS infrastructure, and only requires development of relatively simple computer software and hardware. Costs should be relatively low, perhaps less than seven figures.

Frank Ingram