PSI - Issue 64

Giovanni Pietro Terrasi et al. / Procedia Structural Integrity 64 (2024) 1347–1359 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

1355

9

concentrated dry graphite lubricant. The hypothesis is that the excessive LTM draw-in on cable 2 contributed to this failure pattern. This places the LTM under a higher bi-axial contact pressure. As a result, the wires also experience higher lateral pressure and often fail in a mode like shearing-off. Another possibility is that cable 2 included two previously damaged wires causing its premature first wire failure at 1650 kN (Table 3) and the second wire failure at 1738 kN. During the dismantling of the old Kleine Emme bridge cables in 2016, several single wires were damaged. This manifested itself in local CFRP splinters approx. 1-2 cm long, which reduce the wire cross-section and represent a local notch. Such a wire has a greatly reduced tensile strength. This could explain the early first two wire failures on cable 2. The plausibility of this second hypothesis is that the averaged wire tension at failure after the subtraction of two prematurely failed wires for cable 2 is practically identical as for cable 1 (material utilization at maximum force approx. 95%). This finding led to the adoption of the comprehensive (complete) bending proof testing procedure described in chapter 2.1 when selecting the 2.2 km CFRP wires to produce the two new bridge post tensioning cables. Taking into account the intact CFRP wires, both cables achieved a material utilization of approximately 95% at failure. This is considered satisfactory for parallel-wire cables made from 25-year-old and re-used CFRP wires. In addition, failure is also good-natured. Not all wires fail at the same time, which greatly reduces the potential for damage from flying wire debris and anchor sleeve kickback. In lieu thereof, individual wires fail first, whereby the cable force can no longer be increased but the displacement can which leads to a relaxation of prestress if a post tensioning cable would fail in service. However, the failure of single wires is quasi-brittle. Regarding the sustained tensile load creep tests performed at 1300 kN before the tensile failure tests, the measured LTM draw-in of 0.87 mm (only for cable 1) would inevitably lead to prestress losses. The duration of these tests was limited to 2.75 -5.5 days, which means that no conclusive statement can be made. However, these tests were carried out at 1300 kN, i.e. at a 30% higher load than the planned prestress of the cables for the Ilfis Bridge. Creep is a non linear phenomenon. With a planned preload of 1000 kN, the creep deformation and the LTM draw-in will be significantly smaller. Monitoring of the draw-in of the Ilfis Bridge post-tensioning cables was therefore recognised to be important and implemented. The longitudinal compressive strains measured with strain gauges on the novel CFRP sleeve's surface at the abutment of the anchor were limited to -0.37% (cable 1) to -0.39% (cable 2) at failure in both tests (the sleeve compressive failure would be expected at compressive strains of -1%). Its monitoring during the creep tests at 1300 kN showed no signs of CFRP sleeve creep with stable strains of -0.28% monitored on average for cable 1 (133 test hours) and of -0.27% for cable 2 (66 test hours). 3. Ilfis Bridge strengthening with two cables made of re-used CFRP wires The Ilfis Bridge shown in Figure 8 was built in 1988 as a classic prestressed reinforced concrete bridge with a length of 44 meters. It was built as a box girder and the steel strands for prestressing run in the webs of the torsion box. Various maintenance measures were defined as part of periodic monitoring in 2021. This included, among others, urgent bearing reinforcement in the short term, i.e. strengthening of the bearings against earthquakes and flooding, as well as strengthening the bridge to meet current standards and verifications.

Fig. 8: Layout of the bridge with utility lines (red) and location of the cable sleeves with support (blue)

The structural review of the current condition showed that with updated building material parameters, updated permanent loads and road loads in accordance with SIA 269 (2011) and detailed resistance models, all verifications could be fulfilled with the exception of load model 3 for exceptional transports (type II).