Automatic Warning System
Following a Signal Passed at Danger (SPAD) accident in poor
visibility at Harrow and Wealdstone in 1952 when 112 persons were
killed, British Railways decided to deploy their Automatic
Warning System (AWS) over the whole network to provide train
drivers with an in-cab warning of the indication of the next
signal. This was a non-contact version of a system originally
used on the Great Western Railway. Following a long development
and approval programme, widespread installation started in 1956.
This system is still in use today.
AWS uses a pair of magnets, located in the four-foot about 200
yards (184m) in rear of the signal, to provide an indication of
the status of the signal, while a receiver on the leading vehicle
provides the input for the cab indications. A sound indication
and a visual reminder system provide a warning for the driver
that a stop or restricted speed signal is ahead of the train and
a sound only indication is provided for a signal showing a green
aspect. The driver must acknowledge the restrictive signal
warning within 3s or the train brakes will apply automatically.
AWS is provided on 98% of the UK main line railway network.
The use of AWS was extended to protect certain permanent speed
restrictions after an accident at Morpeth in 1969, which was
caused by a driver failing to reduce the train speed for a speed
restriction. AWS was installed at speed restriction locations
where the train approach speed is over 60 mph or the reduction in
speed required is more than one third of the approach speed. This is mandated in a railway group standard.
After an accident at Nuneaton in 1975, the use of AWS was
extended further to include temporary speed restrictions. It is
now also used for emergency speed restrictions.
Although AWS has been partially effective in reducing train
movement accidents, it has not eliminated SPAD or over-speed
errors entirely. By design, it is only intended to provide an
alert and a reminder of a restrictive signal aspect or speed
restriction. As long as the alert is acknowledged, the driver
may continue to drive the train at any speed. A number of
accidents occurred where drivers forgot the restrictive aspect
warning – despite the visual reminder in the shape of the
so-called ‘sunflower’ device.
AWS may be considered to a limited extent as a fail-to-safety
system since the main trigger element is a permanent magnet.
Failure of the electro magnet results in a warning indication to
the driver. However, removal of the permanent magnet from the
track is not detected.
Reminder Appliance (DRA)
The Driver’s Reminder Appliance (DRA) was introduced
from 1998 to assist with the prevention of SPADs, particularly at
station starting signals. It is not really a train protection
device in the narrow sense of the terminology. The DRA device
consists of a push button in the driver’s cab. The driver
is required to operate this whenever the train is detained at a
red signal. The button disables traction and prevents the driver
from restarting the train until he has reset the button. The
primary purpose is to prevent trains starting against a red
signal when inadvertently given the "right away" by a conductor
or station staff. The effectiveness of this system is a matter
of debate because its operation may become
‘automated’ as part of the train starting
Train Protection and
Warning System (TPWS)
To overcome the limitations of AWS, an enforcement system was
developed for the British railway system, known as TPWS (Train
Protection and Warning System). This is designed to enforce
observance of restricted speed requirements and signal stops by
imposing a full brake application when over-speed is detected or
when a train is being driven past a stop signal. TPWS was first
tried on a section of the Thameslink route in 1996 and was then
installed across most of the UK network between March 2000 and
For each signal or speed restriction equipped with TPWS, two
pairs of electrical loops are placed between the rails. One pair
is placed at the signal itself, the other pair some 200 to 450
metres on the approach side of the signal. Each pair consists
of, firstly, an arming loop and secondly, a trigger loop. The
loops are activated if the signal is showing a stop aspect.
The pair of approach loops, first met by the train 400 to 200
metres before the signal, are set between 4 and 36 metres apart.
When the train passes over the arming loop, an on-board timer is
switched on to detect the time elapsed while the train passes the
distance between the arming loop and the trigger loop. This time
period provides a speed test. If the test indicates that the
train is travelling too fast, a full brake application will be
initiated. In case the train passes the speed test successfully
at the first pair of loops but then fails to stop at the signal,
the second set of loops at the signal will cause a brake
application. At the signal, both loops are adjacent so that, if
a train passes over them, the elapsed time will be so short that
the brake application is initiated at any but crawling speed.
TPWS does not provide the enforcement range of a full ATP
system. If the train’s approach speed is too fast, TPWS
will apply a full brake but the train may still overrun the
signal if the speed is high enough. There is usually an
"overlap" of 200 yards (183 metres) between the signal and any
potential obstacle (train or points) in the block that it is
protecting so there will be a much reduced risk of collision. With a possible total distance of 2000 feet (about 600 m) between
the brake initiation and the block entrance, trains "hitting" the
first loops at up to 120 km/h (75 mph) can be stopped within a
distance sufficient to avoid a collision.
In an attempt to reduce the SPAD risks at certain high-speed
locations, an add-on to TPWS, called TPWS+ is provided at certain
signals where train speeds are above 100 mph. An additional loop
pair is set about 770m in rear of the signal in order to provide
the braking distance for a train "tripped" at 100 mph. Therefore, in these locations the signal approach has two speed
traps. There are about 500 sites chosen on a risk assessment
A further variation of TPWS, designed to be compliant with
European requirements (q.v. below) and known as TPWS-E, was
tried on a section of the GWML but it was not proceeded with
further in order to allow rapid deployment of the already
approved TPWS equipment.
TPWS is also provided at many (about 3000) Permanent Speed
Restrictions (PSRs) to ensure that a train does not pass through
the section of line at too high a speed. TPWS has also been
provided at terminal platforms to ensure the train speed is
reduced to 10 mph on the approach to the stops. This has had the
effect of reducing capacity at some terminals because of the time
taken for trains to clear the routes over the throat into the
terminus. This type implementation may also encourage drivers to
re-motor when travelling along a platform.
In spite of the limitations of TPWS, it is suggested in
published data that 60% of potential accidents due to SPADs can
be prevented by the installation of TPWS at critical locations.
This has been achieved, it is said, at 10% of the installation
costs of a full ATP system. However, this financial target was
not achieved because of a decision to monitor the status of the
TPWS beacons. With the lack of any in-built failure warning
capability for TPWS, it was decided to link installations to the
signal in rear, so that, in the event of a TPWS failure, the
signal would display a red aspect. This addition to the original
specification significantly increased costs.
TPWS does not replace the existing AWS system. AWS is
retained, so that drivers still receive the warnings advising of
adverse signal indications. The TPWS equipment was designed to
interface with the existing on-board AWS equipment on trains or
to replace it so that it could be fitted quickly.
Token Block (RETB)
In rural areas of the UK, where long sections of single line
require token block operation, a system for centralised control,
using modern computer technology, was adopted. It is known as
Radio Electronic Token Block (RETB).
Each train operating over the single line is equipped with a
special speech and data radio transmitter/receiver with a unique
identity. At the start of the single line, the driver stops and
calls the control centre for authority to enter the section. If
the line is clear, the signalman in the control centre transmits
a coded "electronic token" data message which is received by the
train and then shows the authority for that section on a cab
display. The driver will then call for confirmation that he can
enter the section. Once in the single line section, he will
advise the control centre that he has cleared the loop track. A
clearance marker board is specially provided to help him. When
he has reached the end of the single line section, the driver
calls the control centre again and offers to give up the token. After a
"handshake" procedure by the control centre, he sends the
token back by radio data transmission to release the section.
The signaller is provided with a computer system that
allocates the coded tokens to each section and prevents more than
one token being issued for an occupied section. It also receives
the tokens sent back by each train as it reaches the end of the
single line section.
At the exits of the single line sections, the points are
permanently set in the direction of normal running and are
"trailable" for trains entering the section, i.e. they allow a
train to pass through at reduced speed using the wheel flanges to
move the point (switch) blades aside reset to the normal
A "Distant Board" complete with AWS ramp, warns the driver
that he must slow down for the movement over the points leaving
the single line. The Points Indicator shows the position of the
points. A "Stop Board" at the end of the passing loop warns the
driver to stop and ask for permission to enter the next single
line section. Stop board locations are provided with TPWS loops
that are linked by radio to the signal controller. A "Loop
Clear" board indicates to the driver when the rear of the train
is clear of the points.
Sicherungssystem (Indusi, Tyne and Wear Metro, Germany and
Indusi is a German designed main line railway warning and
supervision system used on the Tyne & Wear Metro. It is also
standard in Germany and Austria. A track-mounted inductive
transponder is used to transmit signal warning and speed limit
codes to the train. The transponder is mounted on the sleeper
ends just outside the four-foot, unlike most other systems where
the transmitter is mounted between the rails.
The approach to a danger signal is protected by a transponder
that indicates a maximum speed and causes emergency braking if a
preset level is exceeded. On the main line version, an adverse
distant signal indication must be acknowledged by the driver to
prevent an emergency brake application.
The system is used in Germany for lines with a maximum speed
up to 160 km/h and in Austria for line speeds up to 120 km/h. In
the more recent electronic version, it includes speed supervision
to a braking curve. It is not fully designed to vital
Warning System (CAWS, Ireland)
Some sections of the main line routes in the Republic of
Ireland and the whole of the line between Dublin and Cork are
equipped with coded track circuits that provide in-cab signal
indications. The system is known as the Continuous Automatic
Warning System (CAWS).
The in-cab signal indications repeat line-side indications and
are accompanied by an alarm buzzer when there is a change to a
more restrictive aspect. The driver is required to acknowledge
the alarm within 8s to prevent an irrevocable automatic emergency
brake application. After the operation of the emergency brake,
there is a two-minute delay before the system can be reset and
the train is allowed to proceed.
The system is not vital in that the driver can acknowledge a
restrictive signal warning and can then allow the train to
proceed without reducing speed.
(Trip-Cocks, London Underground)
LUL uses mechanical train stops combined with fixed blocks and
individually calculated signal overlaps to provide train
protection on most of its lines. The system prevents collisions
by providing an individually calculated full speed braking
distance beyond every stop signal so that a train "tripped" by
the train stop will come to a stand without infringing a
restricted block. Trains are restricted to 10 mph for 3 minutes
after being tripped to enforce driving on sight at caution speed.
This is known as SCAT (Speed Control After Tripping).
None of the systems mentioned is used for continuous speed
supervision and all of them can be isolated in the cab and the
train can be driven at normal speeds regardless of signal
aspects. Most of the systems require a positive action to issue
a warning or restrictive data. However, TASS displays some of
the behaviour of a true ATP system in that it can detect missing
As mentioned before, in the case of TPWS, the transmitters at
a location are linked to the signal in rear so that this signal
will show a red aspect in the event of TPWS failure at the next
signal. This is because passing trains cannot detect failures of
the track-mounted equipment.
Whilst the described systems above all provide some protection
against collisions and over-speed derailments, none provide the
full and vital protection that is available from modern Automatic
Train Protection systems.
Operation with Train Stops (Glasgow Subway)
The Glasgow Subway has fixed blocks divided by stations. Each
block stretches from station starting signal to station starting
signal. There are no intermediate signals except at the depot
connection. A recently replaced ATO system uses Siemens
equipment with track-mounted beacons and on-broad processors that
control the train driving and braking functions. Each station is
provided with 2 approach beacons for the ATO profile requirements
and a third start beacon in the platform that is linked to the
starting signal and provides authority to proceed to the next
station. The system software is designed to SIL Level 2.
Train protection is provided by contactless train stops
provided at each signal. The equipment was supplied by SAGEM. The track mounted device consists of a permanent magnet
supplemented by additional induction coils to indicated a proceed
signal. Absence of the induction signal will trigger a train
If a train fails to respond to the ATO commands (or a train in
manual mode passes a signal at danger) the on-board receiver will
trigger an irrevocable emergency brake. The driver is required
to operate a reset button to restart the train and speed is
limited to 25 km/h until another start beacon is passed in the
Control (ATO and ATP, London Underground)
On the Victoria and Central Lines, full Automatic Train
Protection is provided by two different Automatic Train Control
systems that also include Automatic Train Operation. ATP failure
enforces manual driving with speed of movement reduced to 10 mph.
The system on the Victoria line was introduced in 1967 and has
been partially upgraded twice.
Both systems use continuous data transmission through coded
track circuits but they are each unique to the line upon which
they operate. The Victoria Line system uses four codes while the
newer Central Line system uses 13 codes.
British Rail installed two Automatic Train Protection systems
with full speed supervision for trial purposes in the early
1990s, one on the Great Western Main Line (by ACEC Belgium
– now Alstom) and one on Chiltern Railways (Selcab by
Alcatel) between Marylebone and Aynho Junction. Both are
intermittent systems with infill loops, added to allow early
release of the braking demand and its supervision when signal
The information transmitted to the train consists of signal
aspect, routing, applicable speed restrictions, the distance to
the next signal and gradients. Drivers are shown the permitted
train speeds by LEDs displayed around the circumference of the
cab speedometer. Green LEDs show the target speed while yellow
LEDs show the permitted release speed. Information on the number
of signals to the next red and speed restrictions is also
The drivers set up the systems using train data input unit in
the cabs that interface with the vehicle computers. The systems
generate three speed curves, one for movement authority, a
warning curve and an intervention curve. Each is separated by 3
mph. If the train exceeds the warning curve speed, the driver
gets an audio/visual warning. If the speed reaches the
intervention curves, the brakes are applied.
Although the systems were introduced as a trial they are still
operational and, since the Southall accident, it has been the
policy for both train-operating companies that a train will not
be allowed to enter service unless the ATP system is
and Speed Supervision (TASS)
TASS (Tilt Authorisation and Speed Supervision) has been
introduced on the West Coast Main Line (WCML) in order to allow
tilting train to operate safely within the somewhat restricted UK
railway infrastructure gauge. The primary purpose of TASS is to
ensure that a train is prevented from tilting where clearances
between adjacent trains or between trains and infrastructure are
restricted. TASS also imposes line speed limits for equipped
trains depending on whether or not the tilting system is
The TASS system is installed on the Virgin Pendolino Class
390and Super Voyager Class 221 fleets and is designed to European
Rail Traffic Management System (ERTMS) standards. Data is
transmitted to the train by track mounted "Eurobalises" and
collected by an antenna mounted under the leading vehicle. Speed
limits are different for the two classes of tilting trains.
The speed limits for tilting trains are displayed on line-side
signs alongside the signage for non-tilting trains. As the train
passes over the first TASS balise, the driver is shown an
indication light to verify the operating status of the system.
Each TASS balise transmits to the train-borne equipment the
position of the next balise, thus ensuring a continuous
‘daisy-chain’ of supervision. A further indication
shows when tilt is enabled. The driver is responsible for
driving within the correct speed throughout the trip. AWS and
TPWS are provided since TASS does not sense or transmit signal
Over-speed is indicated by an audio-visual alarm and, if speed
is not reduced, the train brakes are applied automatically. The
brakes are released and the alarm can be reset when the train
speed is reduced to the correct level. Where clearances are
restricted, the train automatically stops tilting but the speed
is still monitored by TASS. The train speed must be reduced by
25 mph, that is, conventional operation, if tilt fails or a
balise is not detected.
The Docklands Light Railway (DLR) uses an ATP system with full
continuous speed supervision known as Seltrac, supplied by
Alcatel of Canada. Seltrac is a transmission-based, automatic
train control (ATC) system, combining both automatic train
protection (ATP) and automatic train operation (ATO). The duplex
transmission system is via a continuous track mounted cable, with
loops crossovers at 25m intervals forming train position
validators. Effectively, the system provides pseudo moving block
capability thanks to very short ‘virtual’ blocks.
Train detection and inductive data transmission between track and
train are effected by means of the cable. A fixed block back-up
system uses axle counters for train detection.
Trains are operated fully automatically without driver
intervention but a train captain is provided on every train and
can operate the train at reduced speed in an emergency. A
Vehicle On Board Computer (VOBC), linked to the transmission
system, controls the on-board vital and driving functions. All
trains are controlled by Vehicle Control Computers (VCCs) from a
central control facility.
The system has a good safety record but it requires continuous
track cables and uses a back-up axle counter train detection
system. It is therefore relatively expensive to install and
maintain. As with any system requiring extensive track-mounted
cabling, there is the risk of damage to the cables during track
maintenance activities. This type of system can only be
considered for a metro-type operation with a high service
Voie-Machine 430 (TVM 430)
Phase 1 of the Channel Tunnel Rail Link (CTRL) is equipped
with the French continuous transmission ATP system known as TVM
430. This is also the system used in the Channel Tunnel and the
system that will be used on Phase 2 (except for the station area
at St. Pancras International. TVM 430 ("track to train
transmission") is a cab signalling system used on the more recent
TGV lines, developed from the earlier TVM 300 system by the
French company CSEE.
With TVM 430, the line is divided into fixed blocks from 500
to 1500m long. The length of the block depends on the location,
civil track speed limit and the train capacity required in the
area. Line-side marker boards indicate block boundaries
visually. Each block carries a speed code that is injected into
the rails as part of the track circuit current and transmitted to
the train as it progresses through the block. There are five
standard codes representing speed limits between 0 km/h and 300
km/h. The codes are generated according to the condition of the
route ahead, that is, the distance to the next
‘obstacle’. In the case of a failure, the driver can
"drive on sight" up to a maximum speed of 35 km/h.
The driver is presented with the codes in the cab display with
the target speed at the end of the current block and the target
speed at the end of the following block. The target speed is the
speed at which the train should exit the current block and enter
In older versions of TVM, the target speed indication for the
driver was updated only at every block boundary, resulting in a
stepped speed profile. With TVM 430 the train has a continuously
varying target speed through calculations by the on-board
computer, giving a much more realistic speed profile for the
driver to follow.
Eurostar trains are provided with a system of network codes in
order to allow the train to comply with varying line speed limits
over different routes. On lines where the maximum speed limit is
300 km/h (186 mph), a different network code is used from that
used on the section through the Channel Tunnel, where the speed
limit is 160 km/h (100 mph).
TVM is a safe, reliable and well-proven system but it relies
on track circuit based continuous transmission technology and is
therefore expensive to install and maintain.