PSI - Issue 62

Annarosa Lettieri et al. / Procedia Structural Integrity 62 (2024) 789–795 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

790

2

to their long operational period with increasing loading conditions, they are prone to fatigue damage (Yen et al. , 1990). Fatigue is one of the main causes of deterioration of steel bridges that can limit the load-carrying capacity and the residual life of existing structures (Haghani et al. , 2012). Therefore, estimating the fatigue capacity of fatigue-prone details is a key task for bridge owners to plan strengthening and repair actions, guaranteeing the satisfactory performance of such structures in their service state. According to current codes (e.g., Eurocode 3 – Part 1.9 and AASHTO), the fatigue strength of steel structural details is quantified through the so-called fatigue curves (i.e., S - N curves), formerly conceptualised by Wöhler in 1860, identified using a slope coefficient and a nominal strength conventionally defined at 2 million cycles. Specifically, S N curves provide the expected number of cycles to failure ( N ) for a given demanded stress range ( Δ  ) and for a specific structural detail identified based on the potential fatigue crack location (e.g., the gross or net section in a double covered symmetrical joint). Construction standards include plain members welded and bolted joints, but riveted details are not integrated. However, some recommendations have been proposed to compensate for the absence of specific requirements. In particular, detail category 71 is recommended to verify the fatigue behaviour of any riveted structural details according to the JRC-ECCS technical report (Kühn et al. , 2008), as the 95% lower bound of experimental data of full-scale fatigue tests. Moreover, conforming to AASHTO guidelines, detail D, characterised by a slope of 3 until a stress threshold value of 48.3 MPa (similarly to detail category 71 of Eurocode), can be used. Although using the coded structural details allows for the safe-sided fatigue assessment of riveted assemblies, several studies highlighted the excessive conservative estimation obtained mainly in the long-life fatigue regime (Bruhwiler et al. , 1990; Di Battista et al. , 1998). Besides, more recent studies relying on the assessment of the remaining capacity and life of existing steel bridges demonstrated the importance of using suitable fatigue details to properly design maintenance or retrofit operations (e.g., Bertolesi et al. , 2021; Stamatopoulus et al. , 2013; Pipinato et al. , 2014; Pipinato et al. , 2009). Recently, Taras and Greiner (Taras and Greiner, 2009) discussed the inappropriate assumption that all riveted structural details behave similarly and developed a fatigue class catalogue distinguishing five riveted constructional details. The catalogue has been created as a result of the categorisation and statistical evaluation of prior fatigue test results (Greiner and Taras, 2007). In particular, a category including tension splices with symmetrical splice plate assembly has been defined and a fatigue strength of 90 MPa and slope equal to 5 has been proposed. Moreover, an additional value of fatigue strength equal to 80 MPa has been suggested in the case of the bearing pressure in the connection overcoming twice the net stress range. Although fatigue phenomena in riveted bridges have received increasing attention in the last few years, there are several knowledge gaps. Thus, correctly assessing the detail category to predict the fatigue behaviour of riveted connections is still an open issue in the research community. Within this framework, experimental fatigue test results have been statistically analysed to investigate the influence of parameters such as geometrical property and connection typology on the fatigue resistance of riveted shear splices. Experimental results have been collected from past research. At the same time, further experimental investigations have been carried out at the Strength Laboratory of the University of Salerno within activities envisaged by the Agreement between the High Council of Public Works (CSLLPP) and the ReLUIS Consortium. The work is organised as follows: Section 2 briefly describes the methodology used to build fatigue curves for the sake of clarity; Section 3 reports the experimental results considered in this study and compares the experimental fatigue curves with the curves currently used to evaluate the fatigue resistance of riveted structural details; Section 4 finally investigates the influence of some parameters such as geometrical properties and joint configuration on the fatigue strength. 2. Methodology The analysis of the fatigue experimental data is presented in the form of S - N curves. In particular, each curve has been determined by considering a linear model according to the statistical procedure defined by ASTM E739, in which the fatigue life N and stress range Δ  represent the dependent and independent variables, respectively. With these assumptions, the fatigue curves can be written as follows: log 10 = log 10 +mlog 10 ∆ (1) where log 10 A and m are the linearised curve's intercept and slope. The parameters mentioned above are determined employing the linear regression method and can be computed by applying the following relationships: