Friday 20 September 2013

Life in the Signal Box

Introduction

Railway Signalling in Britain developed over a long period and it continues to evolve. In the, as yet incomplete, series of posts 'Railway Signalling in Britain' (starting with Introduction), I described some of the equipment but I've not said much about the Signalmen themselves who made the whole system work.

The very early days of railways had something of a 'Wild West' flavour so railway companies brought ex-army men into management to impose some military discipline on the staff. But this discipline could be harsh, encouraging railwaymen, in turn, to become members of Trade Unions.

Back in the steam era, railwaymen of all grades were "a breed apart". Enginemen, in particular, had arduous working conditions but enjoyed wide respect for their skills. Signalmen were also respected for their knowledge and their ability to keep the trains moving.

In less-busy signal boxes, the work wasn't too physically tiring and there might be plenty of time for sitting around but in larger, busier signal boxes, the signalmen would be on the go throughout the shift. The majority of signal boxes were manned by one signalman at a time and locations were sometimes quite remote, so the work tended to appeal to men with independent minds who were happy with their own company (or, perhaps, were just plain 'cussed'). The safety of trains relied heavily on the integrity, concentration and training of signalmen so responsible, disciplined men were the ideal. The ideal was not always achieved and the history of signalling is a fascinating story of the piecemeal addition of ever-more-complex mechanical or electrical features to prevent signalmens' errors resulting in accidents.

Although 'mechanical' signal boxes (that is, signal boxes which rely on physical effort by the signalman to move points and semaphore signals) still survive in the United Kingdom, their use is much reduced. Power Boxes covering larger areas are now more typical, like Manchester Piccadilly Signalling Control Centre, which I described here following a visit in March 2013.

Manchester Piccadilly Signalling Control Centre.

The modernising trend is continuing towards what are called Integrated Electronic Control Centres (IECC) covering even larger areas. There's a brief description in 'Wikipedia' here. The picture below (snatched from a passing train) shows the futuristic West Midlands Signalling Centre at Saltley, Birmingham, which is progressively taking over control of the the railways in the West Midlands.

The futuristic West Midlands Signalling Centre at Saltley, Birmingham.

My Viewpoint

These notes principally relate to British Practice post-Nationalisation in the late 1950s and early 1960s. Although there was only one railway authority (British Railways Board), the system was divided into a number of 'Regions' which retained a fair amount of autonomy in some fields. In addition, those 'Regions' were the result of well over a century of amalgamations of private companies so it's perhaps unsurprising that equipment and working practices could vary significantly from place to place.

Most of the signal boxes I frequented were on the London Midland Region (and many were absorbed former London and North Western Railway establishments) so, in general, that's what I'll describe and illustrate because that's what I'm most familiar with.

In the late 1950s and 1960s I met many signalmen and, as in any walk of life, they varied a lot in knowledge and temperament. To some extent, their characters were formed by their working conditions, so I'll try to describe 'Life in the Signal Box' as I saw it.

In the post Visiting Signalboxes I talked about my interest in railway signalling and that started a whole series of posts about various mechanical signal boxes I managed to visit when I was young. There are some links in that post to articles about individual signal boxes, but I've not yet covered all the signal boxes mentioned.

Tipton Owen Street was one of the mechanical signalboxes I visited.

Unionisation

Most railway workers joined a Trade Union and many were members of the National Union of Railwaymen, usually known as the 'NUR'. In 1990 this union merged with the National Union of Seamen to form the National Union of Rail, Maritime and Transport Workers, usually known as 'RMT' (Wikipedia article here). Most footplatemen were members of the Associated Society of Locomotive Engineers and Firemen, usually known as 'ASLEF' (Wikipedia article here).

Day-to-day conditions and promotion were regulated by agreements of Byzantine complexity made between the British Railways Board and the Unions.


This blank L.M.S. 'Application for Vacancy' form E.R.O. 46075
was found inside a 1938 train register book.


Rules and Signalman's Regulations

Prior to Nationalisation, each railway had its own Rule Book. By 1950, British Rail had introduced its own Rule Book, a pocket-sized book used by all the regions. The Rule Book was reprinted in 1961, incorporating all the amendments to date.

Signalmen additionally worked to the "Green Book" or, to give its proper title 'Regulations for Train Signalling and Signalmen's General Instructions' issued in two versions in 1960. Whilst the Eastern, London Midland, North Eastern, Scottish and Southern regions managed to share one version, the Western region (never really regarded as a 'team player') needed a version of its own.

The 1961 Reprint of the 1950 British Railways Rule Book alongside the Signalmen's "Green Book" - 'Regulations for Train Signalling and Signalmen's General Instructions' (pound coin included to show size).

This was not all of the paperwork signalmen needed to be familiar with - there were also Sectional Appendices (more on Sectional Appendices can be found on Brian Robertson's site here), Local Instructions, Working Timetables and Weekly Notices (which included details of Engineering Work and Special Trains).

Classification of Signal boxes

Signalboxes were graded according to how much work was done, largely based on the number of lever movements, from Class 4 (relatively quiet) to Class 1 (busy) and Special Class (demanding). A single signalman could cope at most signal boxes but some large or busy boxes required more than one. Some boxes had a junior position which might be termed 'Telephone Boy' or 'Booking Lad' to share the work with the signalman. Strategically-important boxes sometimes had a senior position of Regulator, who would decide on the order of trains.

Hours of Opening of Signalboxes

Some signalboxes had to be open continuously if any trains were to run, such as Tipton, which controlled a busy level crossing. Other signalboxes might not be needed at night, or for parts of the weekend, such as Deepfields. There were also signalboxes that only needed to be open intermittently 'as required', such as Tipton Curve Junction. A publication showed the normal hours of opening of each signal box, but this could be changed to cater for special trains, engineering works or accidents.

Click here for larger image
Double-page from 'Hours of Opening of Signal Boxes' Book, showing the Stour Valley line.

Shift Working Patterns

This depended upon the specified opening hours but there were normally three 8-hour shifts: 6.00 a.m. to 2.00 p.m. (called 'Early Shift'), 2.00 p.m. to 10.00 p.m. ('Late Shift') and 10.00 p.m. to 6.00 a.m. ('Night Shift'). Each week, these shifts would rotate so that the man on 'Earlies' in one period would be 'Nights' in the next period. Similarly, the 'Night' man would move to 'Lates' and the 'Late' man to 'Earlies'. Since Rest Days had to be given, three men could not keep a box open continuously and a fourth man was needed to cover Rest Days. Originally, there would be three 'Regular' men who would always work the same box and their 'Rest Days' would be covered by a 'Relief' signalman who would be trained to operate a number of signal boxes, providing Rest Day 'cover' at each of them. 'Relief' signalmen would also deputise for 'Regular' signalman who were on Annual Leave or Sick. Arrangements could become quite involved.

When I was visiting signalboxes around the West Midlands in the late '50s and early '60s, the number of 'Regular' men seemed to be reducing. Modernisation of the railways was well in hand and it was clear that re-signalling schemes were going to need much smaller numbers of staff in the future. The proportion of turns covered by 'Regular' men was reducing and, particularly as older signalmen retired, 'Relief' signalmen were kept busy making up the shortfall. Signalmen were allowed to work up to 12-hour shifts and some men (particularly those with young families) were happy to accept overtime. But, of course, the writing was on the wall and the period leading up to the commissioning of Wolverhampton Power Signal Box on Monday 16th August 1965 was a sad time as many of my friends received redundancy notices.

The Block System

Most double-lines were controlled by the Absolute Block system, where signalmen communicated with adjacent signal boxes using block signalling equipment comprising electric single-stroke bells and galvanometers. There's a brief description of Block Instruments here.


L&NWR-pattern 'DN' Permissive Block Instrument, originally in use at Watery Lane Signalbox.

The single-stroke bell was controlled by a tapper key, a bit like a Morse Key. This key protruded from the block instrument case. Most of the key was metal but there was a dished knob made of a material like Ebonite which could be pressed down by a forefinger to sound the bell in the adjacent box. A spring would push the key back up, ready for the next stroke. This is a bit different from a Morse Key where there would normally be a round knob you could grasp with a thumb and two fingers. The Ebonite knobs were often noticeably worn away, having been used for the passage of many thousands of trains. Also the spring return could become rather weak, so some signalmen would use their thumb to help push the key back up. It was a good idea to avoid touching the metal part of the key - when the electric current was disconnected at the end of a stroke, the coil in the bell at the adjacent box would generate a brief 'back e.m.f.' voltage which could be many hundreds of volts and give a nasty shock. If the signalman happened to be holding a lever with the other hand, providing a more efficient 'earth', the shock was more severe.

Signalling a through train involved accepting the 'Is Line Clear?' from the box in the rear ("Taking On"), obtaining 'Line Clear' from the box in advance ("Sending On"), clearing the signals ("Pulling Off"), observing the train as it passed, sending 'Train Entering Section' ("Section") to the box in advance, ensuring a tail lamp was carried, replacing the signals, giving 'Train Out of Section' ("Knocking Out") to the signal box in rear and waiting for the box in advance to send 'Train Out of Section'. Even this repetitive process could get quite demanding at busy times when trains followed one another on short headways and, of course, trains were being handled in both directions.

The Train Register Book

A key feature of signal boxes was the Train Register Book, which I've described here. Every bell signal was recorded with its time, together with explanatory notes. Entries were supposed to be made as things happened, so as to act as an aide-memoire to a temporarily-confused signalman. Some signalmen might let a number of trains pass, remembering only, say, passing time, before bringing the train register up to date and reconstructing all the individual timings based on their familiarity with the location. The Train Register was the equivalent of the 'Black Box Recorder' for the mechanical age. In the event of an incident, the Train Registers from all the signal boxes involved would be collected (each signal box would start a new Train Register so that they could continue to log movements). The collected registers would then be subjected to forensic study to try to establish whether any irregularity has been involved. Of course, all bell signals were logged in the registers of two adjacent signal boxes so the time Box A claims he 'Sent On' a train to Box B should be the same as the time Box B enters for 'Taking On' the train.

The Box Clock

For the Train Register system to work, boxes had to have an accurate reference to the time. Every signal box had a wall clock, often mounted above the desk holding the Train Register Book. These clocks were spring-driven and required winding on a regular basis. To ensure that clocks did not become inaccurate, a Time Signal was transmitted once daily. Where signal boxes had at least one telephones, the time signal could be issued by telephone, but, where necessary, the time signal could be relayed using the block bell from an adjacent signal box with a telephone. In the West Midlands, all the boxes I visited had at least one omnibus telephone (see 'Communications' below) connecting all the signal boxes on a stretch of line together and usually extended to the District Control Office in Birmingham. The Control Office would have had an accurate clock system and at 9.00 a.m. each day it would issue the Time Signal to all signal boxes by a series of short rings on the Control Telephone circuit, so that each signal box received a distinctive 'Buzz-Buzz-Buzz-Buzz-Buzz ...'. On receipt of the Time Signal, signalmen corrected their clock as necessary and recorded in the Train Register book the necessary correction. Anything more than two or three minutes correction would initiate remedial action. Sometimes, there would be no Time Signal at 9.00 a.m. In this case, it would often be sent at 10.00 a.m. I never found out quite why this occurred.

The Lever Frame

The role of the signalman was an interesting blend of the celebral and the physical. Working the lever frame was definitely physical but it was not a matter of brute force but rather of learning the correct technique for each lever to achieve success. Some levers, particularly for points and signals nearer the box, often needed only moderate effort. Levers which operated more than one set of points or signals at a greater distance could be harder. It was sometimes better to use body weight to get a 'swing' to move a set of point blades correctly, as otherwise the point blades might not move sufficiently. But too much 'swing' could cause the blades to 'bounce' and still not 'fit-up'. Failure could result in not being able to bolt the points with the associated Facing Point Lock lever (in the case of facing points) or clear the signal for the move (because the Signal Detector was 'tied'). Distant signals, often out of sight on the end of three quarters of a mile of signal wire, could be very problematic, particularly if the cable run had become fouled by weeds or temperature change had slackened the wire. Various forms of Slack Adjuster were often provided but correct adjustment could be tricky. Where distant signals were out of sight, electrical repeaters were provided with three indications - 'ON', 'OFF' and (where the signal arm stood between valid 'ON' and 'OFF' positions) 'WRONG'. It could be very dispiriting to have used what you were confident was the correct technique only to have the repeater stubbornly show 'WRONG'.

In some signalboxes, it was considered sacrilege to operate a lever with your bare hands, and the use of a duster was mandatory. It's certainly true that perspiration from the hands can tarnish metal levers but most of the boxes I visited, even those that were kept very clean, seemed to have a more robust view.

Part of a large mechanical lever frame. This is the preserved Exeter West signalbox, preserved at Crewe Heritage Centre.

Communication Facilities

In the early days, the block signaling equipment might be the only communication with the outside world. There were bell codes to cope with various mishaps such as 'Train passed without tail lamp', 'Train running away wrong line' but there were limits and over a period telephones were introduced in most signal boxes. The pictures below show a couple of patterns of telephones widely used in signal boxes. At larger boxes, a Key and Lamp Unit might be installed, so that a number of circuits could be answered using just one handset. A Key and Lamp Unit is described here.

Some signalmen, if not otherwise engaged, would talk incessantly about every topic under the sun on the 'Box-to-Box' omnibus telephone circuit to whoever would listen, some liked to "ear-wig", listening to the wide-ranging discussions but rarely making a contribution and some used the telephones only for operational reasons.

I hope to write more about railway communications in the future.


Wall-mounting Local Battery 2-button D.C.-ringing Omnibus Telephone with separate Microphone and Receiver displayed in Shackerstone Railway Museum.


Wall-mounting Local Battery 4-button D.C.-ringing Omnibus Telephone with combined Microphone and Receiver (Hand Micro-Telephone) displayed in Shackerstone Railway Museum.

Heating Signalboxes

Signal boxes could be cold, draughty places. The ground floor of a signal box, the 'Locking Room', was provided with a long slot at ground level to allow the signal wires and point rodding to be led outside. This admitted quite a gale, which could readily find an exit onto the operating floor 'upstairs' through the numerous slots which allowed the levers in the frame to be moved. It was not uncommon to find pieces of old carpet laid across the lever slots on the operating floor, leaving only the regularly-used levers exposed. In addition, there were usually lots of glazed window units on the operating floor, many of which could be slid open, giving rise to draughts even when closed. Finally, signal boxes were located based on operational requirements and the need to give the signalman a good view of the railway he was controlling. This often meant that signal boxes were in rather bleak locations.

Heating to counter the cold was usually by means of a cast-iron solid fuel stove. A signalman re-opening a signal box after a period of closure could find it cold and damp, so getting a good fire going was normally a priority. Once lit, it would normally be kept lit until the next period of closure. Often, the stove was the only method of boiling water or cooking. The picture below shows the stove at the preserved signal box at Shackerstone.


The Signalbox Stove.

Illumination of Signalboxes

A lot of the signal boxes I visited were gas lit, although some had mains electric light. At least one (Tipton Curve Junction) was lit by 'Tilley' pressurised paraffin lamps. I was delighted to find that 'Tilley' lamps are still available - the website is here.

Toilet Facilities

Not many signal boxes had facilities connected to a main sewer. The most common arrangement was an 'Elsan' chemical toilet, often housed in a small wooden 'shed' on the operating floor level, usually reached from the landing at the top of the (external) signal box steps. The arrangement at Bloomfield Junction - a sliding door on the back wall of the signal box gave access to a separate, small landing with metal handrails leading to the usual wooden 'shed'. A modernised arrangement provided a pre-cast concrete 'hut' adjacent to the signal box. 'Elsan' are still in business but the current arrangements shown on their website seem a vast improvement on what I remember.

Furnishings

Furnishings were usually fairly basic. Generally, the floor of the box would be covered by a heavy-duty brown linoleum but this was not universal.

There was always a tall writing Desk, with the Train Register book open on the sloping top. The sloping top was hinged to reach the storage space inside which held various documents such as a copy of the Working Timetable and Weekly Notices and supplies of Single Line Working forms. It was possible to find quite old documents abandoned in this storage space. The photograph below shows a rather nice 'high chair' but I think this refinement was missing at the signal boxes I visited - normally, signalmen would stand at the desk to enter times in the register.


The Desk in the preserved Crewe Station 'A' Box.

There would usually be some sort of 'easy chair' pulled up near the stove where signalmen would spend free time. This would be arranged so that the signalman could relax in the chair and talk on the box-to-box omnibus telephone circuit. Some of these chairs were life-expired and had had various repairs carried out. I particularly remember the throne-like chair at Bloomfield Junction. The 'wings' and back had been massively extended, using odd bits of orange box wood, presumably to shield the signalman from the draughts but creating the appearance of a rather beat-up sedan chair.

There would also be a Locker for Signalmen's personal possessions. The example shown in the photograph below is a common design, divided into three lockable sections so that up to three regular signalmen could each have private storage space.

The Locker in the preserved Crewe Station 'A' Box.

Reporting Responsibilities

Boxes with more freight involvement often completed a 'Freight Train and Shunting Record Book' detailing freight trains, time spent, vehicles attached and detached. Abstracts from this would be passed to the Control Office by telephone regularly.

Page from a Freight Train and Shunting Record Book (dated 1939).

Some locations had to submit a 'Daily Freight Rolling Stock Return' to assist the Control Office with keeping track of freight vehicles (some hope - see my short post here) and ensuring that yards had sufficient empty wagons on hand for traffic to be loaded. I didn't get a copy of this comprehensive return but I did do a quick sketch.


Click here for larger image
Daily Freight Rolling Stock Return FORM A (SMALL) BR87240: Sketch showing layout of this form.


These old document systems were eventually overthrown. In the late 1960s, British Rail invested heavily in computers and a system called 'TOPS' (Total Operations Processing System) which had worked well on the Southern Pacific Railroad in America was introduced. There's a 'Wikipedia' article on TOPS here. By the 1970s 'TOPS' was also controlling freight wagons, as described here.

Supervision of Signalmen

Each Stationmaster or Yardmaster would be responsible for one or more signalboxes in his area, making regular visits to each signal box. All official visitors to a signal box, such as the Stationmaster, would make an entry in the Train Register. The Stationmaster would also make a check of entries in the train register and entries of interest (like "Clock Correct at 9.00 a.m. Time Signal") would receive a 'Tick'.

The 'Green Book' specified the exact sequence of Bell Codes to be used and complying with this was called "Working 'Up'". However, it was common for signalmen to use a somewhat simplified version called "Working 'Loose'". For instance, receipt of 'Two Beats' (Train Entering Section) was supposed to be acknowledged on the bell and the Block Indicator turned to 'Train on Line'. When "Working 'Loose'" the bell acknowledgement would be omitted.

Provided a signalman spotted the approach of the Stationmaster whilst working 'Loose', the adjacent signal boxes would be warned to start working 'Up' by sending the unoffical bell signal '1-2-1, 1-2-1' ("Gaffer Approaching"). The genuine bell signal '1-2-1' was used (where authorised) for 'Train Approaching'.

The Signalmen's Inspector could also appear unexpectedly. He was responsible for 'passing-out' new signalmen and would periodically make visits to signal boxes. He could be relied upon to have a detailed knowledge of all the signalling regulations, so signalmen could not hope to bluff this gentleman.

Book References

There are a number of books written by men who were signalmen. I've listed a couple by Adrian Vaughan:-
[1] 'Signalman's Morning' by Adrian Vaughan, published John Murray (ISBN 0-7195-3827-0).
[2] 'Signalman's Nightmare' by Adrian Vaughan, published John Murray (ISBN 0-7195-4285-5).