La Ferté english summary

On Tuesday 13 June 2006, on the Charleville-Longuyon line, the rear carriage of iron ore train 72187 travelling at a speed of 100 km/h on track 1 from Dunkerque to Dieulouard (Meurthe et Moselle), derailed at kilometre point 190,200 in the district of La Ferté-sur-Chiers (Ardennes). The front bogie of this rear carriage (the last of a train comprising 44 carriages) mounted the outside rail upon exiting a turn with a radius of 676 metres. At kilometre point 198,700, the derailed bogie collided with a rail section that had been stored by the side of the line, knocking it onto the track and causing the rear bogie of the carriage to derail. The back of the train came to a standstill at kilometre point 200,130, having travelled 9,930 metres.
The accident only led to one slight injury (a track maintenance worker). Track 1 was badly damaged over a distance of some 10 km, and track 2 was partially covered with ballast. The electrical and signalling systems were not damaged. The damage was valued at a total of over € 3.5 million in January 2007. The "operating" consequences were significant, because both lines were affected.
As regards the track itself, no defects were detected that would require train services to be halted or traffic to slow down in the derailment zone.
The carriage (a privately-owned carriage belonging to the steel company Arcelor) had received normal maintenance and its inspection uncovered no significant anomalies in its geometry. The carriage’s load distribution appeared to be satisfactory.

As there was no "visual" evidence to explain what parameters were to blame for the derailment, a numerical analysis was commissioned from two organisations : the SNCF’s Centre for Materials Engineering (CIM or Centre d’Ingénierie du Matériel) and National Institute for Research into Transport and Safety (INRETS or Institut National de Recherche sur les Transports et leur Sécurité). Their calculations (using the VOCODYM application in the case of the CIM, and the VOCOLIN application in the case of the INRETS) were made possible by the results of the carriage’s full geometric inspection by the Etablissement de Maintenance du Matériel de Picardie (Picardie Equipment Maintenance Company).

The cause of the derailment appeared to be linked primarily to the condition of the track, even though its geometry was in line with the standard values for rail tracks. Continuous analogue measurement of the track parameters (based on the "Mauzin" values) brought to light five similar points where the line had straightened, 20 metres apart, in the zone preceding the derailment, in a transition curve at the end of a turn where a lack of a natural camber was accentuated by a raised point on the left-hand rail. Dynamic coupling took place between the rear carriage (whose coupler was not equipped with a rear fastening mechanism) and the track : periodic transverse excitations with a wavelength of 20 metres caused the carriage to oscillate transversely, and ultimately caused the wheel flange to climb up to the leftmost point of the rail (11mm). Furthermore, the rails appeared to be insufficiently greased, thus increasing the coefficient of friction between the wheel and the rail and increasing the risk of wheel flanges mounting the rail.

Three recommendations were made in the report :
- The SNCF and RFF should draw up rules for track maintenance so as to rectify these situations (correction of straightening problems according to the values quantified after detection of repetitive and periodic straightening defects at the end of bends).
- As regards the maintenance of carriages, the SNCF will need to improve the traceability of the detection of insufficient play (and of corresponding repairs) in the friction damping system of Y25 bogies.
- As regards operational aspects, the SNCF will need to improve the reactivity of its operators when faced with an emergency situation, through the use of ground-train radio and emergency procedures.

Partager la page


Sur le même sujet