I WILL PREFACE THIS ARTICLE BY STATING THAT IRISH RAIL CLAIM TO HAVE FIXED THIS PROBLEM BY THE RETROSPECTIVE FITTING OF TRACK CIRCUIT ACTUATORS TO THE ARROW UNITS.
The ARROW rolling stock, in use throughout the Greater Dublin Area and beyond has some sort of problem with the underlying signalling system. Basically, the signalling system does not always recognise that the ARROW is there.
The computerised signalling system works by sending a low voltage current through each of the two rails. The system is divided into a number of sections, each of which will have its own track circuits. As a train enters a section, it breaks the circuit by acting as a conductor between the two circuits (one in each rail). A knowledge of Junior Cert Science will tell you that if you place a metal object across two parallel conducting wires (in this case the rails), you break the circuit. This is what the train is supposed to do. This breaking of the circuit resets the signal behind the train to red to indicate that the section ahead is occupied.
Looking at the enclosed diagram A, train 2, having entered Section 4, has set the signal behind it to red. The signal behind this, i.e., that at the end of Section 2 is orange, indicating that the section ahead is clear, but the one after that is not (because it is occupied by Train 2). The signal behind this again, i.e., that at the end of Section 1 is green, indicating that the next signal is at least orange. (Obviously if Train 2 was not there, the signal at the end of Section 3 would be at orange if Train 2 was in Section 5, in which case the signal at the end of Section 2 would be green as would that at the end of Section 1). In other words, a green signal means that at least two sections ahead are clear. There may be more.
What will happen as train 1 moves on is shown on line 2. As train 1 enters Section 2, it sets the signal at the end of Section 1 to red. As Train 1 moves into Section 3, it sets the signal at the end of Section 2 to red and, presuming that no train then enters Section 2, this will set the signal at the end of Section 1 to orange. This should be fail safe as additional factors built in would result in a signal being reset to red if, for example, one of the rails in the section ahead breaks.
Unfortunately, when you factor in the problems that the ARROWs have, as stated above, the potential for trouble exists. If, the presence of the ARROW is not recognised, you have the situation shown in diagram B. Because in the eyes of the signalling system, the stopped Arrow in Section 4 does not exist, the signal at the end of Section 3 shows green. This will mean that that at the end of Section 2 will also show green as will that at the end of Section 1. Therefore as train 1 moves through the sections, it is unaware that Section 4 is occupied. I have stopped the diagram with train 1 in Section 3 with a green signal into Section 4. The rest I leave to your imagination. Suffice to say that the chances of getting anyone to admit to this of course are nil.
There is, however one person in Ireland (other than me) who has contacts throughout Irish Rail and publishes an objective information report on Irish railways. In his edition of May 1999, he states:
P73: "For some months subsequent to this date 2700 class sets were only permitted to operate in six car apparently due to concerns about track circuit actuation".
P81: "The Arrow railcars are not trusted to operate track circuits and therefore must always be protected in the rear by a controlled signal1. As all intermediate signals between Drogheda & Dundalk are automatics any train following an Arrow railcar must wait until the entire section is clear."
There is also the issue of the Boyne Viaduct. This track was, until 1997, interlaced. This means that the two tracks ran together across the viaduct as a means of effecting single line traffic across the viaduct. Appropriate signalling is required to ensure that two trains could not be given clearance to the viaduct at the same time. With the problem with the ARROWs, there is no guarantee that this would be enforceable by a computer. It was therefore necessary to change the track to a conventional single track line, as by doing this, the points moving back and forth, depending on whether a train was going to/from the left/right hand line would provide the appropriate track circuit completeness, rather than relying on an ARROW unit to do so. (See diagram C). Line 1 shows interlaced track. The only protection to prevent a train entering the viaduct from both sides are signals. Line 2 shows the current situation, with the route set for a train travelling north. The points being set for this route can reset the signals for trains travelling south to red. Line 3 shows the current situation with the route set for a train travelling south. The points being set for this route can reset the signals for trains travelling north to red.
While this may be a safer situation overall, it would be unnecessary except for the problem with the ARROWs.
Notes:
1. Refer back to Diagrams A & B. The difference between a controlled signal and an automatic signal is that an automatic signal moves forward to green automatically as a train ahead moves from section to section. A controlled signal, in addition to assessing the state of the track circuits, must then be set to orange and green by a signalman. However, as the signalman in this case is in Connolly Station, there is no advantage to this, except that it is this signalman who gets charged with manslaughter if a fatal accident occurs, and not Irish Rail management.