Fatigue in Riveted Railway Bridges - A study of the fatigue performance of riveted stringers and stringer-to-floor-beam connections

Doktorsavhandling, 2002

This thesis is mainly concerned with two fatigue-related problems in riveted railway bridges. The fatigue life of riveted bridge girders is investigated using a series of fatigue tests performed on full-scale riveted stringers taken from an old riveted railway bridge. The remaining fatigue life of the tested stringers was found to be substantial and the fatigue damage which had accumulated in these old bridge members during the service life of the bridge was negligible. The test results indicate that the fatigue design curve for detail category C = 71 in EC3 provides a safe estimate of the fatigue strength of riveted bridge girders. An examination of the results obtained in previous tests performed on similar riveted members and connections supports this conclusion. The tested specimens displayed excellent fatigue and fracture performance. In particular, the inherent structural redundancy of these built-up members contributed to a relatively slow and ductile fracture scenario. Fatigue cracks were temporarily arrested when passing from one component in the built-up member to another. The behaviour of riveted double-angle stringer-to-floor-beam connections has been studied experimentally and using FE-analysis. These connections have displayed unsatisfactory fatigue performance in several riveted bridges and a relatively large number of fatigue-damage cases have been reported for these connections in Sweden and abroad. Finite-element analysis is used to examine the type of secondary forces acting on stringer-to-floor-beam connections in open-deck riveted bridges and to study the local response of different components in these connections. A series of full-scale fatigue tests has also been conducted on three riveted bridge parts to examine the fatigue performance of the double angle connections. The main mechanism behind fatigue cracking in riveted stringer-to-floor-beam connections is induced deformation, which is repeated cyclically. Interaction between different load-carrying systems in the bridge and the deformation of stringer-ends associated with bending subject the connection angles to out-of-plane distortion, the effects of which were initially not taken into account in the design. The presence of fatigue damage in riveted stringer-to-floor-beam connections does not immediately diminish their load-carrying capacity. As it is induced by applied deformation, the development of fatigue damage in these connections is associated with a reduction in connection stiffness and the rate of damage development is consequently reduced. In the case of the tested bridge parts, this led to a somewhat slow damage rate and fatigue cracks in the connection angles were eventually arrested, before they substantially reduced the load-carrying capacity of the connections. On the basis of the results obtained from the laboratory tests and a study of the damage cases reported for riveted-stringer-to-floor-beam connections, general guidelines relating to the inspection, maintenance and repair of riveted stringer-to-floor-beam connections are also given.