What is a Track Circuit?

On 1 April 1872, the Pennsylvania Railroad placed a section of the New Jersey Railroad under an experimental new system devised by William Robinson — a telegraph engineer who had noticed that a telegraph line interrupted by a grounded conductor could be used to detect the presence of that conductor. Robinson applied the same logic to railway track: use the rails as conductors, use the train’s axles as the grounding element, and use an electromagnetic relay at the end of the section to signal whether a train was present. The test was successful. Within two decades, track circuits had spread across the American rail network. Within half a century they were the global standard for train detection.

A century and a half later, in an era of GPS positioning, fibre-optic sensing, and Communications-Based Train Control, track circuits remain the primary train detection technology on the majority of the world’s mainline railways. Understanding why requires understanding both their elegant simplicity and their critical role in the defence-in-depth structure of railway safety.

How a Track Circuit Works: The Closed Loop Principle

A track circuit divides the track into discrete sections (blocks) by placing electrically insulating joints in the rails at each end of each section. A battery or transformer supplies a low voltage (typically 1.5–12 V DC or an audio-frequency AC signal) to one rail at one end of the section. The current flows along that rail to the other end, passes through the relay coil, and returns via the other rail.

The key safety property — and the reason the track circuit has remained the dominant train detection technology for 150 years — is that every failure mode produces a danger indication. A broken rail, a severed cable, a flat battery, a corroded connection: all cause the relay to drop and the signal to show red. The system can only show “clear” when everything is working correctly and the section is genuinely unoccupied. This is the definition of fail-safe design.

Types of Track Circuits

1. DC Track Circuits (Classic)

The original Robinson design uses a DC voltage — battery or rectified mains supply. DC track circuits are simple, reliable, and inexpensive, but have several limitations on modern electrified railways:

• Traction return current interference: On DC-electrified railways (750 V, 1,500 V), traction return current flows through the running rails and can mask or simulate track circuit signals, causing false occupancy or false clear indications.
• Insulated rail joints: DC circuits require insulating joints in both rails at each section boundary — physically cutting the electrical continuity with non-conductive material. On continuously welded rail (CWR) networks, drilling and inserting glued insulating joints into every rail join is expensive and creates potential points of mechanical weakness.
• Rail resistance sensitivity: DC circuits are sensitive to rail-to-rail resistance — rusty rail surfaces, leaf contamination, and wet conditions can raise the shunt resistance of a train’s axle above the pickup threshold, causing the train to become invisible to the circuit.

2. Audio-Frequency (AF) Track Circuits

Audio-frequency track circuits replace the DC supply with an AC signal at a specific frequency, typically in the range 50–100 Hz (e.g., 83.3 Hz, 91.6 Hz, 100 Hz — standardised frequencies used by Siemens, Alstom, and Thales track circuit equipment). The relay is replaced by a tuned receiver that only responds to the specific transmitter frequency. This approach provides several advantages over DC circuits:

• No insulated rail joints required: Because the receiver only responds to a specific frequency, sections can be separated using frequency coding rather than physical insulation — adjacent sections use different frequencies. This eliminates the need for insulated joints in CWR track.
• Immune to DC traction return: The AC signal at the circuit frequency is distinguishable from DC traction return currents, enabling reliable operation on DC-electrified railways.
• Multiple section encoding: The transmitted signal can carry additional encoded information — track status, speed restrictions — that is received by the train’s onboard equipment and displayed to the driver. This is the basis of cab signalling systems that use the track circuit as a data channel to the train.

3. Coded Track Circuits (Jointless)

Jointless coded track circuits — used extensively by Alstom (UM71/TVM), Siemens (FTGS/GRS), and Thales (SACEM/STAIL) — are the modern standard for electrified mainlines and high-speed lines. They transmit a coded AC signal into one rail and receive it from the other at each end of the section, with the section length defined by the transmission/reception characteristics of the signal rather than physical rail joints.

Coded circuits can transmit speed code information to trains equipped with track circuit receivers — the code transmitted in the rails ahead tells the train’s onboard computer what speed it is permitted to travel. This is the data transmission mechanism used by TVM 430 on French TGV lines and by many North American cab signalling systems: the track circuit is simultaneously a train detection device and a data channel for speed information.

Track Circuit vs Axle Counter: Full Comparison

The Shunt Resistance Problem

The most operationally significant failure mode of track circuits is shunt resistance failure — the condition where a train’s axle does not create a sufficiently low electrical resistance between the two rails to drop the relay, causing the train to be invisible to the signalling system. This is called a “high shunt” or “poor shunt” condition.

The required shunt resistance for a track circuit to correctly detect a train is typically less than 0.5 ohms across the axle. Factors that increase shunt resistance include:

• Rust and oxide on the rail surface: Steel rail oxidises rapidly on lightly used track, forming a non-conductive iron oxide layer. The first train to use a section after a period of disuse may not be detected by the track circuit until its wheels mechanically abrade the rust from the rail head.
• Leaf contamination: Compressed leaf film (black rail) is a significant electrical insulator. On leaf-contaminated track in autumn, shunting failures are a documented cause of signalling system failures.
• Sand from sanders: Sand deposited between the wheel and rail by the train’s sanding system is electrically non-conductive and can prevent adequate shunting — a direct conflict between WSP/sanding needs and track circuit reliability.
• Wheel tread contamination: Oil, grease, or traction gel on wheel treads increases shunt resistance.

High-shunt conditions are a safety-critical failure — the train exists but the signal system believes the section is clear, potentially allowing a conflicting movement to be authorised. This is why track circuit maintenance includes regular inspection of rail surface condition and why some railways use dedicated rail-cleaning trains ahead of service in poor weather conditions.

How Track Circuits Enable Block Signalling

Track circuits are the enabling technology for automatic block signalling — the system that keeps trains separated by ensuring no two trains occupy the same block section simultaneously. The block is the fundamental unit of railway capacity management:

• When a train enters a block section, the track circuit drops the signal at the entrance to red.
• The signal one block behind drops to yellow (caution — next signal is red).
• Two blocks behind drops to yellow or double yellow depending on the system.
• As the train leaves the block, the circuit re-energises and the signal sequence clears behind it.

The minimum headway between trains on a track circuit-based fixed block system is determined by the block length (longer blocks = more separation = lower capacity) and the approach signalling sequence (how many blocks of warning are given before the red signal). On a 1,500-metre block with a 3-aspect signal system, minimum headway is typically 3–4 minutes. Moving block systems — enabled by CBTC or ETCS Level 3 — eliminate fixed blocks entirely, using continuous train position data to calculate dynamic separation, enabling headways of 90 seconds or below.

Track Circuits and the ETCS Transition

As European railways migrate to ETCS Level 2 and Level 3, the role of track circuits is changing. ETCS Level 2 uses track circuits (or axle counters) for train detection and broken rail detection but eliminates lineside signals — the movement authority is transmitted to the train by radio rather than displayed on trackside signals. Track circuits remain essential for their train detection and broken rail detection function but are no longer the primary authority communication mechanism.

ETCS Level 3 — the future specification — would eliminate track circuits entirely, relying on train-reported position for train detection. This introduces a critical gap: broken rail detection. If track circuits are removed and a rail breaks, ETCS Level 3 provides no infrastructure-based mechanism to detect the break before a train reaches it. This is the fundamental technical obstacle to ETCS Level 3 deployment, and it is unresolved. Several research programmes — including the use of DAS fibre-optic sensing as a broken rail detection supplement — are addressing it, but as of 2026, no ETCS Level 3 deployment in revenue service has been completed on a mainline railway.

Editor’s Analysis

Frequently Asked Questions

Q: Why do signals turn red when power fails — shouldn’t they stay green if nothing is wrong?

The fail-safe design of track circuits is intentional and fundamental to railway safety. The relay requires continuous electrical current to remain energised and show a clear indication. When power fails — or when any component in the circuit fails, or when the rail breaks — the relay loses power and drops to its de-energised position, which drives the signal to danger. This means a failure is indistinguishable from an occupied section, and trains are stopped. The alternative — a signal that fails to green — would be catastrophically dangerous, potentially allowing trains into sections where another train is already present, or where a rail is broken. “Fail to safe” (fail to danger for train operations) is the core principle of all safety-critical railway signalling design: the consequence of a component failure should always be more restrictive movement authority, not less.

Q: How long can a track circuit section be?

Section length is limited by the electrical resistance of the rails and the sensitivity of the relay or receiver. On a DC track circuit, the maximum practical length is approximately 500–1,000 metres — beyond that, the rail resistance reduces the circuit current to a level where reliable relay operation cannot be guaranteed, particularly when the rail is wet (which increases leakage to earth). Audio-frequency track circuits can operate over longer sections — typically 1,000–3,000 metres — because the AC signal can be transmitted at higher voltage and is less affected by rail-to-earth leakage resistance. Very long sections also reduce signalling system capacity, since the entire section must be shown as occupied whenever any part of it contains a train. High-capacity lines use shorter sections (sometimes as short as 50–100 metres in station areas) to maximise the number of trains that can be managed simultaneously.

Q: Can a track circuit detect the number of trains in a section?

No — a standard track circuit is a binary device. It can only indicate whether the section is occupied (relay dropped) or clear (relay energised). It cannot distinguish between one train and two trains, or between a train and a broken rail in the same section. This is its fundamental limitation compared to continuous position reporting systems like CBTC or ETCS: the signalling system knows that something is in a section but not precisely what or where. This is why fixed block signalling (based on track circuits) requires conservative headways — the entire block length must be kept clear behind a train, even if the train is only at the far end of the block.

Q: What is an insulated rail joint and why is it needed for track circuits?

An insulated rail joint (IRJ) is a junction between two rail sections where the electrical continuity of the rail is deliberately interrupted by a non-conductive material — typically a fibreglass or nylon composite insert bonded with epoxy adhesive. IRJs are installed at each end of every DC track circuit section in both rails, creating the electrical boundaries of each detection block. Without IRJs, the track circuit current would flow beyond the intended section, making it impossible to determine which section contains a train. IRJs are mechanically weaker than continuously welded rail and require regular inspection and replacement. The development of audio-frequency jointless track circuits — which use frequency coding rather than physical insulation to define section boundaries — was motivated in part by the high maintenance cost and mechanical implications of IRJs on high-speed and high-tonnage lines.

Q: Why can’t axle counters fully replace track circuits?

The single capability that axle counters cannot replicate is broken rail detection. A track circuit detects a broken rail because the fracture opens the electrical circuit — the current can no longer flow from one end of the section to the other, and the relay drops to danger. An axle counter counts axles at the section boundaries but has no knowledge of the rail condition between the boundaries. If a rail fractures between two axle counter heads when no train is in the section, the axle counter will report the section as clear — which it is, in terms of train occupancy — but a train travelling into the section will encounter a broken rail at full speed with no prior warning. For this reason, network operators that have converted to axle counters on routes with high broken rail risk (typically high-tonnage freight routes in cold climates) must compensate with increased physical rail inspection frequency or supplementary broken rail detection technology. Track circuits on such routes remain essential precisely for their broken rail detection capability.