RAIB: Audenshaw freight derailment traced to LBS screws

If you stand by a railway, you expect the rails to stay the same distance apart. On 6 September 2024 in Audenshaw, Greater Manchester, that distance widened at the wrong moment. The Rail Accident Investigation Branch (RAIB) has now published its findings, setting out what failed and how we can learn from it, with the report released on 24 December 2025.

Here are the key facts you’ll want to keep in mind as you read. A freight train derailed at about 11:25, nine of its 24 loaded wagons left the track, and although no one was hurt, the line was shut for around eight weeks for repairs, according to RAIB’s summary on GOV.UK.

Let’s get the basics straight. Track “gauge” is the fixed distance between the two rails. Trains rely on that distance staying tight; if it spreads, a wheel can slip off the rail. Think of a bicycle: if the front fork opened out mid‑ride, the wheel would no longer sit safely in place.

This bridge didn’t use the usual sleepers in ballast. It used a longitudinal bearer system (LBS): long timber beams that run under each rail. The rail sits on a metal baseplate, and that baseplate is held to the timber with large screws. It’s a perfectly legitimate design used in places like bridges, but it depends on those screws holding firm.

RAIB says the derailment began when the track gauge spread because several baseplate screws had already failed through fatigue, allowing the right‑hand wheels to drop into the widening gap. In plain terms: the fixings no longer restrained the rail tightly enough to guide the wheels.

Why weren’t the warning signs picked up? Investigators found that both automated and manual inspections on this stretch weren’t reliable at spotting failing or failed screws. Routine geometry readings looked acceptable, so the system didn’t trigger extra action. This shows how a measurement can be “in spec” while a hidden risk grows.

RAIB’s analysis adds two time points worth noting. The bridge’s LBS went in during 2007. From 2015, traffic increased, which sped up screw fatigue, even though train forces stayed below Network Rail’s stated limits. In other words, the components weren’t expected to last indefinitely in this configuration.

The paper trail matters too. RAIB reports that some previous screw failures at the same spots had happened before, but local teams did not record or report them, and wider assurance didn’t correct that gap. The branch has issued eight recommendations to Network Rail, covering component assurance, management guidance, staff competence, cross‑team working between track and structures, understanding how the supporting structure affects the track, reassessing assets when traffic patterns change, improving asset records, and strengthening internal assurance of record‑keeping.

For students and educators, this is a live example of engineering risk management. Loads can be within published limits and still create fatigue over time when the configuration and service conditions shift. If we only watch the big numbers, like overall geometry, we can miss the small fasteners that quietly take the strain.

So what should an inspection regime teach us here? First, measure what matters at the level where failure begins-in this case, the fastenings. Second, bring track and bridge specialists into the same conversation. Third, treat traffic increases as a design and maintenance question, not just a timetable win. These are habits as much as rules, and they save time and money as well as keeping people safe.

Quick definitions to help you teach this case. Gauge: the fixed spacing between rails. Longitudinal bearer system (LBS): rails mounted on long timber beams with baseplates and screws, often on bridges. Fatigue: tiny cracks growing under repeated loads until the component finally fails. Dynamic track geometry: how the track measures under a passing vehicle, useful but not a substitute for checking fastenings.

What happens next? RAIB’s eight recommendations aim to lift standards for LBS design and components, improve inspection and record‑keeping, and make sure changes in traffic lead to smarter maintenance plans. If we’re going to keep freight moving and footpaths safe beneath bridges, this is the kind of practical learning we need to carry into 2026.