The Rail Accident Investigation Branch has issued urgent safety advice for Network Rail and other infrastructure managers and the companies responsible for maintaining or inspecting longitudinal timber systems.
The advice comes as the agency investigates a September derailment of a freight train crossing a bridge in Audenshaw, Manchester.
At around 11:25 on Sept. 6, a freight train consisting of two class 66 locomotives and 24 wagons loaded with aggregate and traveling between Peak Forest and Salford derailed as it passed over the bridge. The locomotives and the leading 10 cars passed safely over the bridge, but the next nine derailed.
The last car came to a stand on the bridge. While there were no injuries, the derailment caused substantial damage to railway infrastructure and damaged some of the wagons.
The evidence available at this stage to RAIB’s investigation indicates that the derailment was caused by gauge spread within the first half of the bridge, with the track having transferred from a ballasted track to a longitudinal timber system as the train entered the bridge. On-site, RAIB recovered 13 failed LSA chair screws from the baseplates of the low-side rail.
None of these screws were marked with “HT” on their heads. Investigators said that indicates that these chair screws were probably not made of high-tensile strength material.
Early results from metallurgical analysis of these failed chair screws show that they all exhibited signs of low-cycle and high-cycle fatigue failure in bending. A number of them also exhibited a small area of unfatigued material, which eventually fractured by overload.
The plane of failure in most cases was just below the top level of the timber, typically one or two threads below the screws’ shanks.
This mode of failure of the chair screws is not easily detectable by visual inspection means. As was found from RAIB’s investigation into the derailment of a freight train at Sheffield station on Nov. 11, 2020, the upper portion of a broken screw may still offer some resistance to rotation or removal by hand, which makes detection of an impending failure by gauge widening difficult.
At the Audenshaw derailment site, RAIB noted no clear indications of significant baseplate, baseplate packing, shuffle, or timber indentation.
- Longitudinal timber systems using hardwood timbers.
- The use of PAN M6 baseplates, or similar, with only two chair screws per plate.
- The presence of non-HT chair screws, which may be more susceptible to failure by fatigue. Although breakages of the screws marked ‘HT’ have been found on the high-side rail baseplates these are yet to be fully examined.
- The quantity of packing present between baseplate and timber. This may exacerbate screw bending behaviour and subsequent fatigue failure
- A history of local gauge widening under traffic over a short period of time. This may be below the values measured by either manual (static) and automated (dynamic) means which would trigger an intervention.
- The presence of high gauge spreading forces applied to the rails due to track curvature, vehicle characteristics or a transition from ballasted to longitudinal timber supported track.
- The degree of support available from, and fixity of the longitudinal timber system to, the structure.
Be the first to comment