This study presents a design criterion developed for fatigue strengthening of a 120-year-old metallic railway bridge in Switzerland and presents a pre-stressed un-bonded reinforcement (PUR) system developed to apply the strengthening. The PUR system uses carbon fiber reinforced polymer (CFRP) plates; however, unlike conventional pre-stressed CFRP reinforcement methods, preparation of the existing metallic bridge surface is not required. This decreases the time required for on-site strengthening procedures. The principle of the constant life diagram (CLD) and two fatigue failure criteria (Johnson and Goodman) are described. Analytical formulations are developed based on the CLD method to determine the minimum CFRP pre-stress level required to prevent fatigue crack initiation. The PUR system uses an applied pre-stress force to reduce the mean stress level (and stress ratio) to shift an existing fatigue-susceptible metallic detail from the 'at risk' finite life regime to the 'safe' infinite life regime. The applied CLD method is particularly valuable when the stress history of the detail is not known and it is difficult to assess the remaining fatigue life. Moreover, it is shown that the currently adopted approach in many structural codes which emphasizes stress range as the dominant parameter influencing fatigue life are non-conservative for tension-tension stress patterns (i.e., stress ratios of 0

a b s t r a c tThe majority of fatigue strengthening studies focus on reducing propagation rates of existing cracks, ignoring the crack initiation stage. Many existing metallic bridge members however do not contain existing cracks, but rather are nearing their design fatigue life. Limited research exists on the prevention of crack initiation using carbon fiber reinforced polymer (CFRP) materials. In this paper, constant life diagrams (CLDs) are used to determine the minimum level of CFRP pre-stress required to indefinitely extend the fatigue life of existing metallic beams. It is shown that by applying a compressive force to an existing fatigue-susceptible detail using pre-stressed CFRP plates, the mean stress level can be reduced such that the detail is shifted from the 'finite life' regime to the 'infinite life' regime. The proposed fatigue strengthening approach is advantageous particularly when the stress history from the prior traffic loadings is not known. To validate the proposed method, a pre-stressed un-bonded CFRP reinforcement system is introduced and tested on four metallic beams. The proposed un-bonded CFRP system is advantageous over traditional bonded CFRP systems as it can be applied to rough or obstructed surfaces (surfaces containing rivet heads or corrosion pitting for example). Additionally, the new un-bonded CFRP system offers a fast on-site installation (no glue and surface preparation are required) and an adaptive pre-stress level. Experimental results show that strengthening using pre-stressed CFRP plates are capable of shifting the working stresses from a finite fatigue-life zone to an infinite fatigue-life zone preventing crack initiation. Although according to many structural standards, the stress range is the main parameter that affects the fatigue life of a metallic detail, the results of this study clearly show that the mean stress level also plays a significant rule in the detail fatigue life. Based on the proposed CLD approach in this paper, the combined effects of the stress range and mean stress level can be taken into account for prediction of fatigue life of metallic members.

No abstract

No abstract

No abstract