PSI - Issue 64

Aeneas Paul et al. / Procedia Structural Integrity 64 (2024) 1287–1294 Aeneas Paul et al. / Structural Integrity Procedia 00 (2019) 000 – 000

1288

2

1. Introduction Many bridges are approaching the end of their theoretical lifespan and are in the meantime exposed to loading scenarios, that exceed initial design assumptions by far (Sanio et al. (2018)). Often this results in significant shortcomings in their bearing capacities, e.g. for fatigue and shear (Fischer et al. (2014)). Especially for prestressed concrete bridges, stress corrosion cracking (Atrens and Ramamurthy (2013)) is a relevant failure mechanism, which can lead to tendon breaks. At early stages this damage mechanism often runs unrecognized in the background, since initially no cracking of the concrete happens (Schacht et al. (2019)). Then, monitoring plays a central role to detect damage early (Mischo et al. (2022)). Current methods for detecting tendon breaks are a) acoustic emission analysis (Käding et al. (2022)), b) electromagnetic measurements (Scheel (2006)) and c) coda wave interferometry (Sträter et al. (2023)). However, all have their drawbacks, such as being a) strongly influenced by environmental conditions, b) not suitable for permanent monitoring or c) still in the early stages of development. This contribution presents an alternative approach, that utilizes the strain fingerprint of tendons re-anchoring in the surrounding injection mortar or concrete in case of a break (Sanio et al. (2021)), (Sanio and Mark (2020)). This changes the local strain field in the concrete nearby and even propagates to the surface where it can be detected using distributed fiber-optical sensors (DFOS, see (Fischer et al. (2019)), (Konertz et al. (2019)), (Speck et al. (2019))). Based on experiments, the impact of the re-anchoring process on the concrete surface strain is investigated. Therefore, in two beams tendons are cut while the strength and thus the re-anchoring length were varied. Associated pull-out tests served to quantify the bond behavior. The experimental results from DFOS are compared with Changes in strains and stresses in case of a tendon break highly depend on the tendons’ ability to re -anchor in the surrounding material, typically concrete or injection mortar. Full re-anchoring means the full re-introduction of the initial prestressing force P 0 of the separated tendon over the transfer length l pt into the concrete. While the far-away ends of the separated tendon are assumed to remain anchored in their original positions, the near ends re-anchor, as visualized in Fig. 1a. To illustrate the re-anchoring principle, the longitudinal stresses from an FE simulation with linear coupling between tendon and concrete are superimposed herein. Re-introduction of the prestressing force P 0 released from the tendon to the concrete happens through the bond and with respect to the bond stress τ b . The force quantity transferred to the concrete P ( x ) at a certain location x is obtained from eq. (1) as the double integral of τ b over the lateral surface area of the concrete-tendon interface A b . ( ) = ∬ ( ) (1) With constant circumference, the bond stress just varies over the length 0 ≤ x ≤ l pt . In general, τ b can be decomposed into three components: The first is a constant base part, which results from friction and adhesion. The second part is strain-dependent and reflects the Hoyer effect, that captures the tendon ’ s effort, to reshape into its originally larger, unstressed cross-section (Hoyer (1939)), causing wedging in the surrounding material. This strain-dependent part is most noticeable at the break point (Hegger and Bertram (2010)). The third part reflects the tendon ’ s effort to shorten and reshape to its initial length, resulting in a slip-dependent bond stress due to the different displacements of tendon and concrete. It is also maximal at the break point. Thus, τ b is maximal at the break point and decreases over the transfer length l pt . At this length only the base part remains active and a constant prestressing force is reached (Sanio and Mark (2020)). According to eq. (1), P ( x ) is zero at the break point. With the distance, P ( x ) increases until P ( l pt ) = P 0 is reached at the end of the transfer length . Moreover, the varying stress field at the interface of tendon and concrete alters the strain field in the surrounding concrete, too. The local loss of longitudinal prestress is superimposed with the wedging of the Hoyer effect and yields a typical strain field or fingerprint. numerical ones from linear-elastic finite element (FE) simulations. 2. Fiber-optical measurement system to detect tendon breaks 2.1. Basic principle of a tendon break