Fatigue assessment of two identical steel truss bridges

It is not a simple task to assess the fatigue capacity of steel bridges in order to prolong their lifespan. The traffic load history for existing structures is rarely known in detail, constraints within the static system can often be difficult to estimate as well as the choice of the fatigue detail category. Too conservative assumptions may result in replacement of existing bridges, since minor overestimations of stresses have significant effects on the lifespan of the structure, due to the log-log relationship of the well-established S-N curves. One way of reducing uncertainties, related to loads and load-effects as well as to the dynamic effects, is to perform measurements on existing structures exposed to live loads. This paper presents the fatigue evaluation of derived fatigue stress-cycles based on rainflow counting (RFC) on measured data from two different, but identically built, steel bridges in Sweden. The Rautasjokk Bridge located along the Iron Ore line (Malmbanan) were found to be exposed to more and higher stress cycles compared to the Åby Bridge, located along the Main line (Stambanan). For the Rautasjokk Bridge, the trains loaded with iron ore were responsible for about 80% of the fatigue damage on the stringer beams. Presented in this paper is also a comparison of theoretical methods for fatigue assessment, prescribed in the Swedish bridge assessment code. Furthermore, suggestions for improvement of theoretical assessment methods are also presented.