Railway track is the fixed guiding and load-bearing structure on which trains run. In its most familiar form, it consists of two parallel steel rails secured to transverse supports and carried on a prepared foundation. The term can describe an individual running line or the complete track system, including junctions, crossings, fastening assemblies, and supporting layers. Track is a central component of a railway, but it is distinct from the wider railway system of vehicles, signalling, stations, bridges, and operating arrangements.
Track performs several functions simultaneously: it supports concentrated wheel loads, guides vehicles through curves and junctions, transmits braking and traction forces, and maintains a sufficiently smooth running surface. On many railways, the rails also carry electrical return currents or form part of train-detection circuits. Its engineering therefore combines structural design, metallurgy, soil mechanics, vehicle dynamics, surveying, and maintenance planning. The essential challenge is to preserve precise geometry while accommodating repeated loading, temperature changes, water, and gradual material deterioration.
Modern track ranges from relatively light industrial lines to heavy-haul mineral railways and precision-built high-speed rail. Although their component sizes, construction methods, and maintenance regimes differ, all depend on a continuous load path from the wheel–rail contact to the ground and on reliable control of rail position.
Historical development
Guided routes for wheeled vehicles predate modern railways, but the direct ancestors of contemporary track were the wooden wagonways used in European mining districts during the early modern period. Wooden rails reduced rolling resistance compared with ordinary roads and helped guide heavily loaded wagons. By the eighteenth century, iron plates were increasingly applied to wooden rails to reduce wear. These arrangements were suited to modest speeds and loads but required frequent attention to damaged running surfaces and weak supports.
Two competing forms of iron track emerged. Plateways used L-shaped plates whose raised edges guided wheels without flanges. Edge railways instead used relatively narrow rails with the guiding flanges carried on the wheels. Edge rails ultimately became dominant because their running surfaces could be kept clearer of debris and their profiles developed into efficient structural sections. Early cast-iron rails were brittle, and short rail lengths created numerous joints that were vulnerable to impact and movement.
A major advance was John Birkinshaw’s 1820 patent for rolled wrought-iron rails. Longer, tougher rails were better suited to the growing loads imposed by the steam locomotive. The Stockton and Darlington Railway, opened in 1825, and the Liverpool and Manchester Railway, opened in 1830, helped establish locomotive-hauled railways as practical transport systems. Their development also demonstrated that reliable operation required an integrated track structure rather than rails alone: sleepers, drainage, formation, fastening, and alignment all affected performance.
During the nineteenth century, several rail profiles competed. Bullhead rail became characteristic of British practice, with the rail held in cast-iron chairs attached to sleepers. Flat-bottom rail, associated with Charles Vignoles in Britain and developed in related forms elsewhere, could be attached more directly to its supports. Improvements in steelmaking during the second half of the century made steel rails increasingly economical. Steel provided substantially greater resistance to wear and repeated loading than the wrought iron it displaced.
Twentieth-century development concentrated on heavier rail sections, improved rail steels, elastic fastenings, concrete sleepers, mechanized maintenance, and the elimination of frequent rail joints. Continuous welded rail spread widely after the Second World War. High-speed passenger operation and very heavy freight traffic subsequently encouraged different refinements: exceptionally accurate geometry and smooth transitions for the former, and high load capacity, durable foundations, and wear-resistant components for the latter. Older and newer technologies nevertheless coexist, since track is usually renewed progressively rather than replaced across an entire network at once.
Gauge, alignment, and track geometry
Track gauge is the distance between the inner faces of the two running rails, measured at a specified depth below their top surfaces. Standard gauge is 1,435 mm. Other major systems include 1,520 mm gauge across much of the former Soviet network, 1,524 mm in Finland, 1,668 mm Iberian gauge, and 1,676 mm gauge on much of the Indian subcontinent. Metre-gauge and 1,067 mm networks are also extensive. Gauge determines vehicle compatibility, but it does not by itself determine speed, axle load, or carrying capacity.
Gauge differences can create a break of gauge, requiring passengers or freight to transfer, vehicles to exchange bogies, or suitably equipped wheelsets to change gauge. Dual-gauge track accommodates more than one gauge by using three or, in some arrangements, four rails. The closely spaced rails complicate fastenings and junctions. Loading gauge is a separate concept: it defines the permitted vehicle envelope, while structure clearances ensure that vehicles can pass bridges, platforms, tunnels, and adjacent trains.
Horizontal alignment comprises straight track, circular curves, and transition curves. A transition curve introduces curvature gradually, avoiding an abrupt change in lateral acceleration. The clothoid is widely used because its curvature increases linearly with distance, although other transition forms also occur. Vertical alignment consists of gradients joined by vertical curves. These are designed around traction, braking, passenger comfort, vehicle clearances, and the risk of excessive changes in coupler or wheel loading.
On a curve, the outer rail is commonly raised above the inner rail. This elevation is called cant or superelevation. In a simplified equilibrium model,
where is the height difference between the rails, is an appropriate effective transverse spacing, is speed, is gravitational acceleration, and is curve radius. Actual design applies railway-specific conventions and limits. Cant deficiency describes operation faster than the equilibrium speed for the installed cant; cant excess occurs at lower speeds. Mixed passenger and freight routes require compromises because trains may traverse the same curve at very different speeds.
Track geometry also includes longitudinal level, lateral alignment, cross level, and twist. Twist is the change in cross level over a specified distance and is particularly important because it can unload individual wheels. Measurement wavelengths matter: a short local defect and a long undulation of equal amplitude can excite different vehicle motions. Consequently, geometry standards specify not merely permissible deviations, but also measurement methods, reference lengths, operating speeds, and intervention thresholds. Accurate gauge alone cannot establish that a track is fit for service.
Rails, sleepers, and fastening systems
A conventional rail has a head that contacts the wheel, a vertical web, and a foot that distributes load to the support. Its section works as a beam, spreading a wheel load along several sleepers rather than applying it at a single point. Rail size is commonly designated by mass per unit length. Modern main-line sections frequently fall in the approximate range of 50–68 kg/m, although lighter and heavier sections serve particular duties. Section shape, steel grade, support spacing, and foundation stiffness are all relevant; mass alone does not determine performance.
Rail steels are generally high-carbon steels selected for hardness, strength, and resistance to wear and contact damage. Head-hardening processes increase hardness in the part most exposed to repeated wheel contact. Material selection involves trade-offs: a harder rail may resist wear effectively, but its suitability also depends on toughness, residual stress, weld quality, and the prevailing wheel–rail conditions. Curved heavy-haul track may need different grades from lightly trafficked straight track.
Sleepers, also called ties, maintain gauge and distribute rail forces into the underlying structure. Timber sleepers are relatively light, resilient, and adaptable at complicated junctions. Their durability depends on species, preservative treatment, moisture exposure, and mechanical condition. Concrete sleepers, commonly prestressed, provide consistent dimensions and substantial mass, and are widely used on modern main lines. Steel sleepers and composite sleepers serve selected applications, including routes where weight, material availability, or resistance to biological deterioration matters.
Sleeper spacing affects rail bending, support loads, and the distribution of forces into ballast. A spacing near 0.6 m is common on many main lines, but actual values vary with design and traffic. Concrete sleepers also require adequate support beneath their intended bearing zones. Poorly supported sleepers can rock or strike the ballast under passing wheels, causing deterioration even when the sleeper itself remains structurally sound.
Fastenings hold the rail in position while allowing controlled elastic movement. Modern systems often combine resilient steel clips, shoulders, insulating components, and rail pads. The pad reduces local contact stresses and influences vertical stiffness and vibration transmission. Fastenings resist rail uplift, lateral movement, and longitudinal forces from braking, traction, and temperature changes. On timber track, baseplates spread loads and provide attachment points for spikes or screws; older bullhead systems use chairs and wedges known as keys.
The rails are usually inclined inward slightly to suit the wheel tread geometry. Inclinations such as 1 in 20 or 1 in 40 occur in different systems, but compatibility depends on the combined wheel and rail profiles. Electrical insulation may also be built into the fastening assembly, particularly where train detection or traction-current control requires separation between rails and supporting structures.
Ballast, slab track, and the ground beneath
Traditional track rests on ballast, a layer of coarse, angular crushed stone. Ballast distributes sleeper loads, provides drainage, and resists movement through particle interlock and friction. It also permits geometry correction: lifting and tamping equipment can reposition the track and rearrange stone beneath the sleepers. Suitable ballast must resist crushing and abrasion while maintaining a particle shape and grading that support drainage and stability.
The ballast shoulder outside the sleeper ends contributes strongly to lateral resistance, while stone between sleepers helps resist longitudinal movement. Beneath the ballast, a sub-ballast or prepared protective layer may separate coarse stone from the underlying soil, reduce stress concentrations, and limit the upward migration of fine material. The earthwork surface supporting these layers is commonly called the formation; the natural or improved soil beneath it forms the subgrade.
Water is a major cause of track deterioration. Saturated fine-grained soils may lose stiffness, while repeated loading can pump water and fines upward into the ballast. Contaminated ballast drains less effectively and may deform more rapidly. Ditches, cross-drains, underdrains, and properly graded surfaces are therefore structural necessities rather than auxiliary features. Frost-susceptible ground introduces additional problems: freezing can cause heave, and thawing can leave a temporarily weakened foundation.
Slab track replaces the conventional ballast bed with a concrete or asphalt-based supporting structure. Rails may be mounted on individual blocks, continuous slabs, or other engineered support arrangements. Elasticity is supplied through pads, fastening systems, or resilient layers rather than through ballast alone. Slab systems are widely used in tunnels, urban railways, and some high-speed networks, where restricted access or the need for stable geometry makes reduced routine adjustment attractive.
Ballasted and slab track have different whole-life characteristics. Ballasted track is comparatively easy to adjust after settlement and can often be renewed in stages. Slab track can preserve geometry with less frequent intervention, but requires a well-controlled foundation and can be difficult to correct after substantial differential settlement. Repair after derailment or structural damage may also demand more specialized work. Neither system is universally superior; tunnel conditions, ground behavior, traffic, maintenance access, and construction cost influence the choice.
Changes in support stiffness require particular attention. At a bridge abutment or the boundary between slab and ballasted track, a sudden stiffness change can produce increased dynamic loads and uneven settlement. Transition zones may use modified sleeper spacing, additional rails, improved ground, or graduated support stiffness. Their purpose is to distribute deformation smoothly rather than merely strengthen the immediate boundary.
Wheel–rail interaction and dynamic loading
The wheel and rail meet over a small contact patch whose dimensions depend on load, profile, and material properties. Contact stresses are consequently much higher than the average stresses in the rail or sleeper. Hertzian contact theory provides a useful idealized treatment of elastic contact, although real railway contacts can involve plastic deformation, surface roughness, contamination, and non-elliptical contact patches. The rail must withstand millions of such loading cycles over its service life.
A conventional wheelset has two wheels fixed to a common axle. Their tread shapes allow the effective rolling radii to change as the wheelset moves laterally. On a curve, a difference in rolling radius helps the outer wheel travel farther than the inner wheel even though both rotate at the same angular speed. Real steering is more complicated than this geometric description: suspension forces, wheel and rail wear, friction, axle spacing, and flange contact all influence the path.
Small relative motions within the contact patch generate creep forces, which provide traction, braking, and steering. These forces are limited by available adhesion. Water, oil, leaf contamination, and other surface conditions can alter adhesion substantially. Low adhesion may impair braking or cause wheel slip, whereas excessive friction at selected contacts can increase wear and lateral forces. Railways therefore distinguish between managing friction at the running surface and lubricating the gauge face of curves.
At sufficiently high speeds, some wheelset and suspension combinations develop hunting oscillation, an unstable lateral and yawing motion. Its onset depends on effective conicity, suspension characteristics, vehicle mass distribution, and other factors. Track irregularities can excite vehicle motions even below instability thresholds. This is why vehicle approval and track standards must be considered together: a geometry defect harmless at one speed or for one vehicle type may produce unacceptable forces in another case.
Repeated tangential and normal loading can cause rolling contact fatigue, including surface cracks that may grow into more serious defects. Wear and fatigue interact: material removal can eliminate shallow cracks, but excessive wear changes the profile and reduces rail life. Wheel flats and out-of-round wheels impose impact loads, while rail corrugation produces periodic excitation that can increase noise, vibration, and component damage.
Track design therefore cannot be based only on static axle load. Dynamic amplification depends on speed, wheel condition, rail irregularities, and support stiffness. Bridges and other structures may introduce additional resonance considerations. Controlled rail grinding, wheel reprofiling, suitable suspension design, and well-maintained geometry act on different parts of this coupled mechanical system.
Rail joints, welding, and thermal forces
Traditional jointed track connects rail ends with bolted joint bars, also called fishplates. A small end gap accommodates temperature-related length changes. The joint nevertheless interrupts the rail’s bending stiffness and running surface. Passing wheels produce repeated impact, which can loosen bolts, batter rail ends, and promote settlement beneath the joint. Joint maintenance was historically a large part of track work, and the characteristic rhythmic sound of older trains arose partly from wheels crossing these discontinuities.
Continuous welded rail joins individual rails into long lengths, greatly reducing routine joints. Flash-butt welding heats and forges aligned rail ends together under controlled pressure. Aluminothermic welding uses a chemical reaction to produce molten steel in a refractory mould around the rail ends and is widely employed for field connections. Both methods require control of alignment, weld geometry, material quality, and finishing. A weld can become a significant defect source if its metallurgy or running-surface profile is inadequate.
A freely expanding rail would change length approximately according to
where is the coefficient of thermal expansion. For steel, a representative value is about . A freely moving kilometre-long rail would therefore change length by roughly 0.6 m for a 50 °C temperature change. In welded track, however, fastenings, sleepers, and ballast largely restrain this movement. The temperature change appears primarily as longitudinal force.
For an ideal, fully restrained rail, the thermal stress magnitude is approximately
where is Young’s modulus and is measured from the rail’s stress-free temperature. Heating produces compression; cooling produces tension. The stress-free, or neutral, temperature is therefore a key track-management parameter. Installation and destressing procedures establish an intended condition, although later rail movement or maintenance can alter the actual state.
Compressive forces can contribute to track buckling when lateral resistance or alignment is inadequate. Risk is influenced by ballast condition, curvature, fastening performance, temperature, and recent disturbance of the track. Tensile forces in cold conditions can contribute to rail or weld breaks, especially where defects are present. Rail temperature, which can differ considerably from air temperature under sunlight, is the relevant thermal quantity.
Expansion devices remain necessary in selected locations, particularly where track interacts with long bridges or other structures that undergo significant movement. Designing these interfaces requires consideration of both rail forces and structural movement. Continuous welded rail does not eliminate thermal expansion; it changes how the track accommodates its effects.
Turnouts, crossings, and route connections
A turnout, also called a set of points or a switch in different usages, allows a train to move from one track to another. Its principal components are movable switch rails, fixed stock rails, connecting rails, and a crossing assembly. The switch rails are shaped to fit closely against the stock rails, creating a continuous guiding path into the selected route. Their position must be established and secured before a train passes.
At the crossing, also called a frog, the wheel paths intersect. A conventional fixed crossing includes gaps that permit wheel flanges to pass through. These gaps interrupt support near the wheel’s running path and create demanding impact and contact conditions. Check rails help guide wheelsets so that flanges follow the correct path past the crossing nose. Their spacing and relationship to wheel dimensions are safety-critical.
For higher-speed or more heavily loaded applications, movable-point or swing-nose crossings can provide a more continuous running surface. Such arrangements reduce the discontinuity at the crossing but require additional moving components, actuation, locking, and detection. Turnout speed is governed by the diverging alignment, transition design, and dynamic constraints, not simply by whether a movable crossing is installed. High-speed junctions may require very long, shallow-angle layouts.
A crossover uses two turnouts to connect adjacent tracks. A diamond crossing allows two routes to intersect without necessarily permitting movement between them. Single and double slips combine crossing and switching functions within a compact area. These arrangements can save space in station approaches, but their concentrated components and complex load paths increase inspection and maintenance demands. Track designers balance land constraints against reliability, permitted speed, and operational flexibility.
Points may be operated manually, electrically, hydraulically, or pneumatically. Interlocking coordinates them with signals and route protection so that conflicting movements cannot be authorized under the designed operating logic. Detection establishes whether the switch rails are in the required position; locking prevents unintended movement during passage. Physical track geometry and signalling logic must agree, but they remain distinct engineering systems.
Turnouts are especially exposed to wear, debris, snow, ice, and impact. Switch heaters are used in cold climates to reduce obstruction by ice and snow, although they do not remove every winter hazard. Long bearers support several rails across the junction, and local stiffness can vary sharply. Inspection therefore includes switch-tip condition, crossing wear, gauge, fastening security, bearer support, and the correct functioning of operating and detection equipment.
Construction, inspection, and renewal
Track construction begins with surveying and preparation of earthworks, drainage, and formation. The required accuracy depends on route speed and system type, but errors in the underlying works can be expensive to correct after rails are installed. On ballasted lines, sleepers and rails may be placed individually or in prefabricated panels. Track-laying trains can deliver and assemble components continuously along a prepared formation. Slab-track construction places greater emphasis on accurate positioning of support units, concrete quality, and curing.
After initial placement, the track is aligned, levelled, and brought to its specified geometry. Tamping machines lift the track and insert vibrating tools into the ballast on either side of sleepers, compacting stone beneath them. Ballast regulators shape the bed and shoulders. Dynamic stabilization equipment applies controlled vibration and loading to improve the resistance of recently disturbed ballast. Subsequent traffic also consolidates the structure, so initial construction and later geometry correction must be coordinated.
Inspection addresses both geometry and component integrity. Track-recording vehicles measure gauge, alignment, level, cross level, and related parameters. Inertial measuring systems, optical methods, and satellite positioning can support localization and comparison over time. Dedicated sensors and instruments can inspect rail profiles, fastenings, sleepers, and clearances. Measurement systems require calibration and interpretation: an apparent change may reflect a measurement difference rather than physical deterioration.
Internal rail defects are commonly sought by ultrasonic testing. Surface and near-surface flaws can be investigated with methods such as eddy-current testing and magnetic techniques, while visual inspection remains important for visible damage and contextual conditions. Not every technique detects every defect, and some locations, including welds and complex crossing components, require specialized procedures. Rail-break detection through signalling equipment is not a substitute for preventive rail inspection.
Rail grinding removes controlled amounts of material to restore profile and manage surface defects. Milling provides another material-removal method, particularly where substantial correction is required. Ballast cleaning screens out fine material and restores a more suitable grading, whereas undercutting removes material from beneath the track. Renewal may replace rails, sleepers, fastenings, ballast, or several of these together. The appropriate intervention depends on the deterioration mechanism rather than age alone.
Maintenance planning combines inspection results with traffic, speed, axle load, component history, and available access time. Machine-learning methods and statistical models can help identify recurring defect patterns, while digital twins may organize geometry, asset condition, and operational data. Their usefulness depends on representative measurements and reliable engineering interpretation. Some defects develop rapidly or are difficult to observe, so predictive models operate alongside defined inspection requirements and immediate-action limits.
Specialized track and interfaces with other systems
Heavy-haul railways emphasize resistance to high axle loads and repeated tonnage. They commonly use substantial rail sections, durable sleepers, strong fastening systems, and carefully engineered foundations. Tight curves can be especially demanding because lateral forces, wear, and rolling contact fatigue coincide. High-speed passenger track places exceptional demands on geometry, transition design, aerodynamic clearances, and consistency of support. These priorities overlap, but a design optimized for one traffic pattern is not automatically optimal for another.
Urban railways present different constraints. In rapid transit tunnels, restricted access and intensive service can favor slab track and durable fastening arrangements. In tramways, grooved rail provides a flangeway within a paved street surface. The surrounding pavement must accommodate traffic loads, drainage, electrical requirements, and access for rail replacement. Embedded track also requires careful control of movement between the rail and adjacent road materials.
Some steep railways supplement wheel–rail adhesion with a toothed rack engaged by a pinion on the vehicle. Such rack railways require precise positioning of the rack relative to the running rails. Other guided systems depart more radically from conventional practice: monorails use a single principal guideway, and magnetic-levitation systems guide and support vehicles without ordinary wheel contact during normal running. Their guideways perform analogous functions but are not conventional two-rail track.
Electric traction creates important track interfaces. Running rails often provide the return path for traction current, while some systems use additional conductor rails. Electrical bonding must maintain intended continuity, and insulated joints may separate sections where required. Track circuits use the rails as part of an electrical system for detecting train occupancy. Other railways use axle counters, which detect passing wheelsets rather than relying on electrical shunting through the rails. These choices affect insulation and bonding arrangements.
In direct-current systems, leakage from the intended return path can cause stray-current corrosion in buried metallic structures. Track insulation, return-circuit design, and monitoring help control this interaction. Noise and ground-borne vibration are other significant interfaces, particularly near buildings. Resilient fastenings, under-sleeper pads, ballast mats, and floating-slab systems can reduce transmission over selected frequency ranges, but their effectiveness depends on the combined vehicle, track, and ground dynamics.
Environmental exposure can determine the form of the track structure. Flooding may wash out ballast or undermine embankments; extreme heat can increase compressive rail forces; unstable slopes can displace the formation. Coastal conditions may accelerate corrosion, while windblown sand can obstruct moving components and contaminate ballast. Designs account for these mechanisms through drainage capacity, erosion protection, materials selection, earthwork stabilization, and inspection of vulnerable locations.