The Silent Guardian: Check Rail (Guard Rail) Explained

The frog nose of a railway crossing — the pointed steel tip at which two rails intersect at a diamond crossing or diverging turnout — is the most mechanically vulnerable point in the entire switch and crossing assembly. A wheel passing over it at 160 km/h exerts a dynamic impact load of several hundred kilonewtons at the precise moment when the wheel’s support transitions from one rail to the other across the 50–80 mm gap at the nose tip. If the wheel arrives at the nose at the wrong lateral position — even 2–3 mm further inward than specified — its flange will strike the nose face rather than the rail head, generating a massive shock load that will progressively deform the nose, damage the wheel flange, and in the worst case deflect the wheelset onto the wrong rail — a derailment in the switch at high speed.

The check rail positioned opposite the frog, on the inner side of the track, exists to prevent precisely this scenario. By contacting the back of the wheel flange and physically restraining the wheelset from moving inward by more than a defined amount, it guarantees that the wheel approaches the frog nose at exactly the correct lateral position, regardless of any lateral forces acting on the wheelset from the curve, the switch blade geometry, or the vehicle dynamics. The check rail never touches the wheel tread. It never carries vertical load. It does only one thing: it ensures the wheel is in exactly the right place at the moment it crosses the frog nose. That single function has prevented an incalculable number of crossing derailments in 180 years of railway operation.

What Is a Check Rail?

A check rail is a secondary rail, installed parallel to and inside (gauge-side of) a running rail, whose sole function is to control the lateral position of a wheelset by contacting the back of the wheel flange. It is not a running rail — wheels do not normally roll on its head surface. It is a restraint device: it provides a physical boundary to inward wheelset movement that supplements or replaces the guidance provided by the opposite running rail’s gauge face.

The check rail contacts the back of the wheel flange — the inner face of the flange, between the two wheels on the axle. This is fundamentally different from the guidance provided by the running rail’s gauge face, which contacts the front (outer face) of the flange. The check rail acts from the inside; the running rail acts from the outside; together they constrain the wheelset to a defined lateral envelope.

The Two Primary Applications: Crossings and Curves

Application 1: Crossings in Switches and Crossings Assemblies

The crossing (frog) is the component of a turnout or diamond crossing where the two rail lines intersect. At a crossing, the running rail must be interrupted to allow the flanges of wheels on the crossing route to pass — creating a gap of 50–80 mm (the “flangeway”) through which the flange passes while the wheel tread transfers from the wing rail to the crossing nose rail.

During this transfer, the wheel is briefly unsupported at the flangeway gap — the wheel tread is bridging the gap on its rolling surface, while the flange passes through the opening. If the wheel arrives at the gap at the correct lateral position, the load transfer is smooth and the nose receives only normal rolling load. If the wheel arrives too far inward — even slightly — the flange strikes the nose tip rather than the rail head, creating a severe impact load.

The check rail on the opposite side of the track from the crossing nose addresses this by pushing the wheelset outward (away from the nose) as it approaches the gap. The check rail contacts the back of the inner wheel flange, physically preventing the wheelset from moving inward beyond the specified check gauge limit. The wheel is held in the correct lateral position throughout the crossing passage.

Application 2: Sharp Curves

On curves with radius below approximately 150–200 metres (tight curves typical of metro systems, tram networks, and industrial railways), the centrifugal force and flange-rail interaction forces push the outer wheel flange hard against the outer rail gauge face. In extreme cases, the combined vertical and lateral forces on the outer wheel can cause the wheel flange to climb the gauge face of the outer rail — the wheel literally rides up and over the top of the rail head, causing derailment by “wheel climb.”

A check rail on the inner rail of the curve addresses this by restraining the inner wheel from moving outward (toward the curve centre). Since the inner and outer wheels are connected on the same rigid axle, restraining the inner wheel from moving inward (relative to the outer wheel) prevents the outer wheel from moving outward and climbing the outer rail. The check rail does not prevent all lateral force on the outer rail — it reduces it by sharing the restraint load between inner and outer rails simultaneously.

Check Gauge: The Critical Dimension

Check gauge is the key dimensional parameter that determines whether a check rail performs its guidance function correctly. It is measured as the distance between:

• The running edge (gauge face) of the crossing nose (or outer running rail on a curve), and
• The acting face (gauge face) of the check rail

This dimension is measured at rail head level, perpendicular to the track centre line.

Check gauge relationship:

Check gauge = Back-to-back wheel distance + flangeway clearance

Standard gauge (1435 mm) typical values:
Running gauge (track gauge): 1435 mm (+3/-0 mm tolerance)
Back-to-back wheel distance: 1360 mm (EN 13715)
Check gauge: typically 1391–1395 mm
Flangeway (nose gap): 41–45 mm

If check gauge is too wide (check rail too far from the crossing nose), the wheel can move inward far enough to strike the nose before the check rail catches the back flange — the check rail fails its protective function. If check gauge is too narrow (check rail too close to the crossing nose), the wheelset is over-constrained and the check rail contacts the wheel on every pass regardless of lateral position — causing rapid wear of both the check rail face and the back of the wheel flange, and potentially generating derailment forces in the opposite direction.

Check Rail vs Running Rail: The Full Comparison

Guard Rails on Bridges: A Separate Safety Function

The term “guard rail” is used for check-rail-like structures in a second, distinct context: on railway bridges and viaducts, where a derailed wheel must not fall through the structure or cause the vehicle to overturn off the edge. These bridge guard rails (also called derailment guards or safety rails) are positioned alongside the running rail at a distance that retains the wheel of a derailed vehicle on the bridge deck — preventing it from falling between the girders or tumbling off the side of the bridge.

Bridge guard rails perform a structural containment function, not a guidance function. They are not in contact with wheel flanges during normal operation — they only engage when a wheel has left the rail. Key design requirements differ from check rails in crossings:

• Vertical position: Bridge guard rails are positioned so that a derailed wheel cannot drop below the level at which it would be contained by the guard rail — typically within 50–75 mm of the running rail head level.
• Horizontal position: Positioned close enough to the running rail that a derailed wheel, moving laterally, is arrested before it reaches the edge of the bridge structure.
• Structural strength: Must withstand the impact of a derailed wheelset at the speed and load of the worst-case derailment scenario on that structure.
• Continuity at expansion joints: Bridge guard rails must accommodate bridge deck thermal movement at expansion joints without creating a gap that a derailed wheel could catch.

Check Rail Applications by Location Type

Maintenance of Check Rails: What Goes Wrong

Check rails are maintenance-intensive components because their functional dimension — check gauge — changes as the acting face of the check rail wears under repeated flange back contact. Even a few millimetres of acting face wear can push the check gauge beyond tolerance, causing the check rail to fail its protective function before the wear is visually obvious.

Key maintenance activities for check rails:

• Check gauge measurement: Measured at each maintenance inspection — typically monthly on high-traffic S&C. Any deviation beyond ±1 mm from the specified value requires either check rail adjustment or replacement. Check gauge is measured with a dedicated gauge — not a standard track gauge — because it requires measuring from the nose running edge to the check rail acting face simultaneously.
• Flangeway clearance: The gap between the running rail gauge face and the check rail acting face must be maintained within specification to ensure correct wheel passage without jamming. Dirt, ballast, and corrosion products accumulate in the flangeway and must be cleared.
• Acting face wear: The check rail acting face wears laterally under wheel flange back contact. When wear reduces the acting face below the minimum specified depth, the check rail must be replaced. Head-hardened grades (R350HT or higher) extend the replacement interval on high-contact locations.
• Height relative to running rail: The check rail must be maintained at the correct height — slightly below the running rail head — to ensure flange back contact occurs without inadvertent tread contact. Differential settlement of the check rail fastening system can change this height relationship.

Editor’s Analysis

Frequently Asked Questions

Q: Why is the check rail slightly lower than the running rail head?

The check rail is set 10–20 mm lower than the running rail head specifically to ensure that the wheel tread does not contact the check rail head surface during normal passage. If the check rail were at the same height as the running rail, wheels passing through the check-railed section might roll on both the running rail and the check rail simultaneously — creating a four-point contact condition that would introduce uncontrolled lateral forces and potential binding. By setting the check rail lower, only the wheel flange back contacts the check rail acting face, while the wheel tread continues to roll on the running rail head exclusively. The height difference (10–20 mm depending on the specification) is calculated to provide the required clearance between the check rail head and the wheel tread at the minimum wheel diameter within the service tolerance range, while still positioning the acting face at the correct height to contact the flange back at the designed back-to-back wheel dimension.

Q: What happens if the check gauge is too wide — what does a “nose strike” look like?

A nose strike occurs when the check gauge is too wide (beyond specification tolerance) and a wheel approaches the crossing nose with insufficient outward restraint — the back flange is not caught by the check rail soon enough, and the wheelset moves inward until the front face of the flange contacts the crossing nose tip rather than the wing rail head. The impact is an abrupt lateral shock load concentrated at the nose tip — the steel-on-steel impact of a wheel flange face hitting the pointed crossing nose at line speed. The consequences depend on severity: a light nose strike at low speed may produce a sharp “clang” audible to trackside observers and minor deformation of the nose tip; a heavy nose strike at high speed can batter the nose tip into a mushroomed shape that catches subsequent wheels, generate severe dynamic forces on the vehicle bogie, or in extreme cases deflect the wheel onto the wrong rail — derailment at the crossing. Repeated nose strikes progressively deform the nose tip and eventually require nose replacement or crossing renewal. The diagnostic signature of nose strikes in track recording data is a characteristic sharp spike in the lateral force channel at the crossing location — something that track recording vehicles and wayside acoustic monitoring can detect, enabling proactive maintenance response before visible damage accumulates.

Q: On what type of curves do check rails become necessary, and what determines the threshold?

The necessity for check rails on curves depends on the derailment risk — specifically whether the combination of curve radius, speed, cant deficiency, and vehicle characteristics creates a risk of wheel climb on the outer rail. The wheel climb derailment mechanism requires the ratio of lateral (Y) to vertical (Q) force at the outer wheel contact to exceed approximately 0.8–1.2 (the Nadal limit), which depends on the flange angle, friction coefficient, and contact geometry. On larger radius curves (above 500–600 m on mainlines), the lateral forces are manageable within normal wheel-rail contact without check rails. As radius decreases below approximately 200 m, the Y/Q ratio approaches the Nadal limit for loaded vehicles at any practical speed, and check rails become necessary to restrain the inner wheel and reduce the net lateral force on the outer rail’s gauge face. The practical threshold varies between network standards: UK Network Rail specifies check rails on curves below 200 m radius; Deutsche Bahn below 190 m; Paris Metro on all curves below 150 m. Tram and light rail networks, with their very tight urban curves (50–100 m radius), almost universally fit check rails on curved sections as a standard installation.

Q: How does a check rail affect noise on sharp curves?

Check rails actually reduce the dominant noise source on sharp curves — the flange squeal generated by the outer wheel flange grinding against the outer rail gauge face as the wheelset negotiates the tight curve. On a curve without a check rail, the outer wheel flange is forced hard against the outer rail gauge face under centrifugal and steering forces, creating a high-frequency stick-slip oscillation that generates the characteristic high-pitched squeal audible from sharp curves on metro and tram networks. When a check rail is fitted on the inner rail, it shares the lateral restraint between the inner and outer rails — reducing the contact force on the outer rail gauge face. Less force on the outer rail gauge face means less stick-slip excitation and reduced squeal. Additionally, rail lubrication is often applied to check rail acting faces and running rail gauge faces on tight curves specifically to reduce friction in the flange-rail contact zone — both for noise reduction and to reduce the wear rate on the check rail acting face and wheel flange back. Some metro networks use dedicated flange lubrication systems on their sharpest curves, applying a controlled quantity of lubricant to the check rail acting face as each train passes.

Q: Are check rails required at all crossings, or only at certain types?

Check rails are required at virtually all fixed crossing noses in the track — both at turnout crossings (where one route diverges from another) and at diamond crossings (where two routes cross at an angle). The only common exception is the swing nose (or moveable nose) crossing — a modern crossing design where the nose itself moves, like a switch blade, to provide a continuous rail surface for both routes rather than a fixed gap. In a swing nose crossing, the nose swings to one side for the straight route and to the other for the turnout route, eliminating the flangeway gap and the need for a check rail on that crossing. Swing nose crossings are more expensive than fixed nose (common) crossings but eliminate the nose strike risk and reduce the maintenance intensity of the crossing significantly. They are increasingly specified on high-speed and high-traffic crossings where the conventional check rail solution creates high inspection and replacement costs. For common crossings (fixed nose), check rails remain mandatory and must be maintained to tolerance as described above.