Well, in the context of the subject, both could be true, but it is signal sighting that is the basis of this article. Positioning and aligning signals so that train drivers can read and interpret them has been an important activity ever since signalling was invented in the early days of railways. To the uninitiated, this might seem a rather trivial task, but it is not a straightforward process and there are many factors, both physical and human, that have to be considered and agreed to achieve a good result that is acceptable to everyone.
Within the UK, RSSB (formerly the Rail Safety and Standards Board) has been working for some time on a new rail industry standard (RIS) to achieve both nationwide consistency and a better understanding of signal sighting for the main line railway.
Before 1999, regional practices, developed over many years, were still being used for signal sighting and subsequent positioning of signals. A Group Standard did exist, but it primarily set out the basic requirements without any of the detail on why they were needed or how they could be met.
On 5 October 1999, two trains collided head on at Ladbroke Grove, on the approaches to Paddington, causing the death of 31 people and injuring more than 500. In the ensuing enquiry, one factor that emerged was the adverse effect of gantry-mounted signals and possible driver confusion for trains leaving the terminus. It became clear that better documentation and instruction was needed and a revised Group Standard (GK RT8037 Signal Positioning and Visibility) was published in 2001. This stabilised the situation and gave the necessary direction to the people carrying out signal sighting assessments.
Since then, the UK has implemented the Railway (Safety) Directive, Common Safety Method Regulations. These changes to legislation, together with ongoing changes to the structure of the UK rail industry, have highlighted the need to review and update signal sighting assessment requirements to take account of a further 15 years’ experience and understanding of the issues involved. The update will include an explanation of the need and rationale of each requirement along with guidance on how to meet these in terms of technical details, what needs to be assessed and who is responsible for the assessment.
Back to basics
So what is needed in the process of sighting (and indeed the siting of) signals? In simple terms, the goal is to confirm that drivers can reliably read and correctly interpret the displayed signal aspects and indications taking account of the train service being operated.
This might seem easy, but a number of people and organisations need to agree the suitability of each and every signal, indicator and sign. These include the signal engineer who provides and maintains the trackside equipment, the railway operators who plan the train services and the train companies who provide the drivers and operate the trains. These three are the nucleus of a Signal Sighting Committee (SSC), which brings together the necessary competence and experience to make the assessment. Cost, safety and engineering practicalities all have to be taken into account as well as future-proofing for the planning of new services and introduction of new rolling stock.
Key to reaching a good result is a good assessment plan that everyone agrees with and can support with the necessary resources.
In order to set down the requirements for signal sighting, it is necessary to understand the needs of the ‘end user’. This is the driver who has to, in sequence, read the signal aspect and indications, interpret their applicability, decide if they apply to the train being driven, interpret their meaning, decide the action to be taken and then do it.
Studies indicate that, on average, the time to assimilate all this for a simple signal is around seven seconds, taking into account conditions such as weather and day/night, although experienced drivers can do this more quickly. The actual time needed is assessed for each signal as the operational context is different for each location. From this a Required Readable Distance (RRD) is derived.
In many cases, the signal sighting is relatively easy to agree but there are instances where the RRD is difficult to achieve for a variety of reasons and this makes the signal sighting decision more difficult. So how does it all work in practice?
Positioning of new signals
When a resignalling scheme takes place, or when new signals have to be provided, the schematic plans show the signal layout that is needed for the train service to be operated, influenced by line speed, types of traffic, gradients, rolling stock characteristics and braking profiles, and other elements. The signal sighting assessment confirms that the proposed positions of signals and structures are compatible with the train service.
Factors that can influence signal sighting are:
- Post or gantry position, left hand side, right hand side;
- Existence of other infrastructure such as electrification stanchions, bridges, level crossings and stations;
- Day and night conditions;
- Impact of weather conditions, for instance direct sunlight;
- Drivers’ cabs and viewing angles;
- Risk of over reading (seeing the next signal beyond the intended one);
- Intrusive local conditions including streetlights and traffic lights;
- Multiple signals in a driver’s field of vision,such as on a gantry;
- Type of signal: colour light (bulb or LED),semaphore, ground/shunt signal, banner repeater, route/stencil indicator, call on indication, trackside sign;
- Curvature of track;
- Tunnel signals and light/dark contrasts.
The SSC takes all of these into account during the assessment and reaches a decision on the optimum sighting arrangement that can reasonably be achieved. This might be a recommendation that a signal is positioned in a different location to that originally envisaged.
Ideally, the signal aspect should be positioned and aligned so that it appears close to the centre of the driver’s normal line of vision and the most prominent display should be the most restrictive (usually the red stop aspect).
Impact of change to existing signals
More difficult to assess is what happens when unplanned changes, both within the rail infrastructure and adjacent to the railway, take place. These can ‘creep’ in without any real recognition of the effect on signal sighting.
This was very much the case at Ladbroke Grove, where the signals had originally been sighted (and sited) to suit an all-diesel railway. The provision of electrification for the Heathrow Express service significantly altered the signal sight lines and worsened the readability of multiple signals on a gantry just outside Paddington.
Other things that can adversely affect signal sighting include:
- Changes to the environment such as new buildings, roads, street lighting;
- Growth of vegetation;
- Changes to track geometry/position;
- Changes to track layouts – provision of double track, bi-directional working;
- Introduction of new trains with different driving positions or braking curves;
- Revised line speeds;
- New stations or revised station layouts;
- Changes to level crossing type or operation.
There will be others, and most will hopefully be captured in the change process whereby signal sighting can be re-assessed. There is always the risk, however, that a change will occur gradually, so it is important that people remain vigilant and report emerging problems before anything untoward happens.
Means of sighting
So how is signal sighting actually achieved? In former times, it meant the SSC going out to site with a mock-up of the signal and holding it in the position intended so that all could make a judgement. As indicated previously, most positions were obvious but the difficult ones could cause a lot of options to be tested with, invariably, a compromise being reached.
Nowadays, with video wizardry, a more sophisticated method is available. Where a new signal is to be positioned on an existing and unchanged piece of track, signal positions can be superimposed upon a video showing the view from the cab. If, however, it is a new piece of railway or a substantially changed layout, then a simulated picture of the route is produced upon which the signals are placed. The level of certainty will depend on the sophistication of the video and the accuracy of the images shown.
Where there is doubt, site visits once the infrastructure is in place might be necessary. It is for the SSC to decide what is needed to support a good assessment decision. At the end of the day, the drivers will experience the results of signal sighting ‘in the flesh’.
A complementary article describing one innovation in the technology of signal sighting appears in this edition of Rail Engineer.
Mitigation, safeguards and risk assessments
As hinted, the positioning of some signals can never be ideal and a compromise has to be reached. If the sighting of a signal is known to be poor, then a banner repeater indicator is sometimes provided on the approach to the signal. This enables the driver to know whether the as yet ‘unseen’ signal is showing a ‘proceed’ or ‘stop’ aspect and therefore whether or not to continue braking.
Originally, banner indicators (effectively a black arm on a white background that moves through 45° when the main signal aspect is proceed – rather akin to a semaphore arm) only showed ON or OFF indications, but recently ‘green’ banner indicators have been implemented on some routes that offer a three-state indication, thus giving the driver better information on the actual aspect being displayed on the main signal.
Like all other lineside signals and indications, banner repeater indicators are also subject to signal sighting assessment.
Signal sighting is one of many assessments that may be necessary before a change to the railway is commissioned. Others include the risk of signal overrun, permissive working and signal layout driveability.
Drivers failing to obey signals or misreading them has been a problem for many years, with the term SPAD (Signal Passed at Danger) becoming a familiar word in the English language. Automatic Train Protection (ATP) systems exist to mitigate the impact of a SPAD and range from the least sophisticated like AWS, through to TPWS with a degree of speed measurement and control, and up to ETCS with constant train supervision. Where a signal sighting assessment identifies that a signal has poor readability but it is not practical to improve it, other risk mitigations are available.
Signalling policy in the UK is always to build in an overlap for main line running signals that provides a safeguard in case the driver fails to stop in time, whatever the reason. Poor adhesion conditions in the leaf fall season can be a real risk, as can adverse weather such as dense fog. However, despite the additional protection that an overlap provides, it should not be used as a means of easing the sighting requirements.
Into the future
As ETCS is gradually introduced with in-cab movement authorities replacing lineside signals, it could be presumed that signal sighting will become a thing of the past. This is not entirely true as even ETCS requires marker boards to define the places where trains may be required to stop, and these will need sighting criteria to apply as before. Ground and shunt signals are likely to remain in station areas and depots where sighting will continue to be important – many minor train collisions occur in these places.
Hence the need for improved documentation. Following a period of consultation, and with participation from a cross industry team involving Network Rail, RDG and the ORR, Rail Industry Standard RIS-0737-CCS was published by RSSB in June. It is very comprehensive and gives both detail and guidance on the multitude of situations that are likely to arise.
RSSB members are developing their own briefing materials to support introduction of this standard into their organisations. The RIS is more a reference guide rather than something that should be read cover to cover, allowing a reader to pick out an area of interest and learn how it should be done.
Thanks to Richard Barrow, lead CCS Engineer at RSSB, for both initiating and facilitating the article.
Written by Clive Kessell
This article was first published in September 2016.