Application of high-frequency mechanical impact treatment for fatigue strength improvement of new and existing bridges

Abstract: This thesis investigates the application of High-Frequency Mechanical Impact (HFMI) treatment for fatigue strength improvement of weldments in existing and new bridges. In the former case, the welds have already been subjected to fatigue loading and accumulated damage before treatment. A fatigue testing program is set up, comprising welded specimens subjected to fatigue loading before HFMI treatment to investigate the efficiency of HFMI treatment on existing structures. Moreover, additional fatigue test results are collected from the literature and analyzed. HFMI treatment is found to be very efficient in extending the fatigue lives of existing structures regardless of the accumulated fatigue damage prior to treatment, given that any surface cracks, if exist, have not grown more than 2.25 mm in depth. For practical applications, HFMI treatment is only recommended if the critical details are verified to be free of any surface cracks. Remelting the surface with a tungsten electrode before HFMI treatment is another solution which has rarely been studied on existing structures. Therefore, several experimental investigations are conducted including fatigue testing, measurement of residual stress, hardness testing and scanning the welds topography to study the effect of combining these two post-weld treatment techniques. The combined treatment is found to be efficient as it induces higher and deeper compressive residual stress and local hardening. These aspects are all considered in numerical simulations conducted to investigate the fatigue behaviour of new and existing weldments treated using this combination. The results verify the superiority of the combined treatment to both individual treatments (TIG & HFMI). Nonetheless, because of the complexity associated with TIG remelting, the combination is only suggested for existing structures containing shallow fatigue cracks which can be fused by a tungsten electrode. One major hindrance to applying HFMI treatment on weldments in steel bridges is the lack of design rules and recommendations such as consideration of stress ratio (mean stress) and overloads.  Therefore, a correction factor (λHFMI) to account for the mean stress effect is derived. This factor is used to magnify the design stress range for fatigue verification of HFMI-treated welded details existing in road and railway bridges. λHFMI is calibrated using measured traffic data that includes millions of vehicles and hundreds of trains. In addition, the characteristic load combination associated with the serviceability limit state is found to be the most appropriate for verifying the maximum stresses in road bridges. Based on the work conducted in this thesis, a complete methodology is proposed for the design and assessment of HFMI-treated welded details in new and existing steel bridges. Finally, the effect of corrosion on the performance of HFMI-treated weldments is studied by analyzing collected test results. Despite the observed reduction in fatigue endurance of HFMI-treated details due to the removal of top layers improved by residual stresses, the obtained fatigue lives are still longer than the design lives assigned for new welded details even in extreme corrosion conditions. However, corrosion protection and removal of sharp HFMI groove edges via light grinding are still necessary to reduce the susceptibility of weldments to corrosion.