Railway Signals & Signalling

A railway's greatest danger is one train running into another — from behind, or head-on. The whole craft of signalling exists to prevent it, and its history is the story of one unchanging goal pursued with ever-better tools: never allow two trains in the same section of track at the same time.

In the beginning: almost no signalling

The earliest railways had barely any signalling. Trains were spaced out by time interval — a train was simply held for a few minutes after the one ahead had left. It was crude and dangerous: if the train in front stopped, the following driver had no warning until he saw it, often too late.

The railway policeman and the first signals

The first signals were given by hand. Railway policemen — the origin of the enginemen's nickname “bobby” for a signalman — showed flag and lamp signals from beside the line. These soon gave way to fixed signals on posts, visible from a distance and worked from a central point.

The breakthrough: the block system

The decisive advance was the absolute block system, still the foundation of British signalling. The line is divided into sections, or blocks, and the rule is simple: only one train is allowed in a block at a time. A signal box guards each end, and the boxes communicate by coded bell rings and electric block instruments. The box in rear "offers" a train; the box ahead only "accepts" it once its section is clear. No train may enter a block until the previous one has been proved to have left it.

Semaphore signals

For most of the steam era, signals were mechanical semaphore arms worked by wire from the signal box. The two essential types looked — and meant — quite different things:

• The home (stop) signal — a red arm with a white stripe and a square end. Horizontal means danger — stop; angled means clear. (Signalmen call a signal standing at danger “on” and a cleared one “off”.)
• The distant signal — a yellow arm with a black chevron and a notched fishtail end (so its very tip is a yellow V). At caution it warns that the stop signals ahead may be on; at clear the line is clear right through.

At night, an oil lamp shining through coloured glass on the arm shows the same message: red for stop, green for clear, and yellow for a distant at caution.

Upper quadrant vs lower quadrant

The big visible difference between companies was which way the arm moved to show "clear". Most railways used upper-quadrant signals, where the arm rises to clear; the Great Western kept lower-quadrant signals, where the arm drops to clear, right to the end. In both, the horizontal position always means danger:

The Great Northern also had its distinctive “somersault” signals, and bell codes and box practice varied from company to company — differences that broadly survived into the British Railways regions after 1948.

Colour-light signals

From the 1920s, electric colour-light signals increasingly replaced semaphores on busy routes — clearer in fog and darkness, and needing no mechanical wires. A modern four-aspect signal can show four running messages:

• Red — danger, stop.
• Single yellow — caution: be prepared to stop at the next signal.
• Double yellow — preliminary caution: the next signal is a single yellow, so begin easing down.
• Green — clear.

Simpler three-aspect signals omit the double yellow. Modern signals are usually worked automatically by track circuits — electrical circuits in the rails that detect a train and set the signals behind it to danger with no human intervention. The block principle is unchanged since the 1850s; only the means of detection has moved on.

Beyond the four aspects: junctions and shunting

Those four colours are the main running aspects — but you'll also see junction indicators, and small signals that control slow moves around yards and sidings:

• A junction indicator, nicknamed a “feather”, is a row of white lights set at an angle above a proceed aspect. When lit, it tells the driver the points are set for a diverging route — the angle of the lights shows roughly which way. (A signal can carry several, for several routes.)
• A position-light shunt signal shows two white lights at 45° to authorise a slow shunting move; horizontal usually means stop.
• A ground disc — the little round signal enginemen nicknamed a “dolly” — is the semaphore-era equivalent: a white disc with a red bar. Bar horizontal means stop; the disc turns so the bar tilts up to authorise the move. These are the signals that govern the shunting and siding moves the main running signals don't.

Single lines and tokens

On a single line worked in both directions the danger is a head-on collision. The classic solution is the token (or staff): a physical object — a metal tablet, staff or key — of which only one exists for each single-line section. A driver may only enter if he holds its unique token, so two trains can never be on the section at once.

Modern train protection

Signals only work if the driver sees and obeys them, so the modern railway adds automatic protection on the locomotive itself.

AWS and the “sunflower”

The Automatic Warning System (AWS), fitted widely from the 1950s, gives the driver a warning in the cab as he approaches each signal: a bell for a clear (green) signal, or a horn for anything less. The driver must press a button to acknowledge a warning — and if he doesn't, the brakes apply automatically. A round cab indicator records what happened at the last signal:

After a green it shows all black; after the driver acknowledges a caution it flips to the black-and-yellow segmented disc universally known as the “sunflower” — a constant visual reminder that he has passed a warning and must be ready to stop.

TPWS and ERTMS

• TPWS (Train Protection & Warning System) goes further than AWS, automatically braking a train that passes a signal at danger — a SPAD (Signal Passed At Danger) — or approaches one too fast.
• ERTMS / ETCS (European Rail Traffic Management System / European Train Control System) is the modern digital standard now being rolled out: it displays the driver's authority to proceed in the cab, can do away with lineside signals altogether, and continuously supervises the train's speed and position.

One principle, ever-better tools

From a policeman waving a flag to a computer supervising a train by radio, the goal has never changed in nearly two centuries: keep trains a safe distance apart, and never let two occupy the same stretch of line. Everything in the signalman's craft — and the driver's discipline in obeying every signal — serves that single end.

More in this section: The Engine Crew · How a Steam Locomotive Works