Life Time Evaluation of Orthotropic Steel Bridge Decks

The fatigue performance of orthotropic steel bridge decks depends on the magnitude and number of stress cycles applied, and the deck details. The stress induced in a deck plate will be reduced by the presence of a stiff surfacing material. This paper describes how the likely fatigue lifetime of an orthotropic steel bridge deck with asphalt surfacings could be estimated using probabilistic assessment methodology. In particular, the daily variations in temperature, which affect the asphalt surfacings, are considered. An example of application of the methodology to a realistic bridge is described. Introduction Orthotropic steel bridge decks have been selected in the design of long span steel bridges because of their favourable characteristics of their high longitudinal stiffness, light weight and short installation time. Two basic types of longitudinal ribs are used; open ribs, and closed ribs of a trapezoidal or rounded cross section. The closed ribs are more commonly used than the open ribs because open ribs are less torsionally rigid and require more welding. Fatigue cracks, however, have often developed at welded connections in orthotropic steel bridges with closed ribs due to the high number of high magnitude axle loads. Many relevant investigations have been conducted internationally (e.g. Wolchuk 1990; Beamish et al. 2006; Jong 2006; Miki 2006; Sugioka et al. 2007). Six locations of fatigue cracks have been identified (Kolstein 2007): (1) at the welded joint between the deck plate and the longitudinal rib between the crossbeams; (2) at the welded joint between the deck plate and the longitudinal rib at the crossbeam; (3) at the longitudinal rib splice welded joint; (4) at the welded joint between the longitudinal rib and the crossbeam; (5) at the welded joint between vertical stiffener and deck plate; and (6) at the butt welded joint in the No r us e w ith ou t A uth or' s p erm iss ion OBC-08-21 Koichi Sugioka 2 deck plate. The rib-to-deck joints (crack type 1 and 2) have four types of potential fatigue cracks as shown in Figure 1: (A) propagating from the weld root through the deck plate; (B) propagating from the weld root through the weld throat; (C) propagating from the weld toe in the deck plate through the deck plate; (D) propagating from the weld toe in the longitudinal rib through the longitudinal rib. The rib-to-deck joints are generally prone to fatigue failures due to out-of-plane bending moments in the deck plates and longitudinal ribs generated by axle loads. Out-of-plane bending moments result in high local flexural stresses in the deck plates and longitudinal ribs since thicknesses of the deck plates and rib walls are relatively thin. Figure 1 Fatigue cracks at rib-to-deck joint. Fatigue is important for many steel bridge structures, but there are numerous uncertainties in definitions when failure occurs. Current conventional design of steel bridges and fatigue analysis is usually based on S-N curves and the Palmgren–Miner hypothesis. In reality, however, the S-N curves are only available for a limited number of structural details, stated by classification tables of design codes. The conditions governing fatigue cracking are generally the structural geometry, material characteristics, and loadings. These conditions are difficult to assess accurately. Thus, an appropriate analysis of fatigue is required to treat the problem in a probabilistic manner. Temperature effects on asphalt surfacing are also not generally considered in conventional design. The proposed approach to assess fatigue lifetime to cracking has similarities to the Pacific Earthquake Engineering Research (PEER) performance-based earthquake engineering (PBEE) methodology. An example application of the methodology to the rib-to-deck welded joints of a realistic bridge using Monte Carlo simulation with one thousand samples is described to predict the remaining lifetime considering hourly temperature effects under each condition. Deck plate Longitudinal rib Rib-to-deck weld A C