PSI - Issue 64

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

1352

6

prestress of 1000 kN for the bridge. After unloading, the final step of each experiment was to test the cable sample to failure in order to determine the tensile strength of the system and to analyse the failure mode. The samples tested at Empa are summarised in Table 2. In order to prove the robustness of the novel CFRP sleeves the same two filament wound CFRP sleeves were used for both cable samples, i.e. the end sleeves were separated from the failed CFRP wires of the tested cable number 1, cleaned and reused for the second combined creep-tensile test on cable 2. In cable 2 four polyethylene organisers were added to define the position of 37 wires precisely and to reduce differences in length when casting the anchors. The graphite-powder-based lubricant between the inner sleeve surface and the LTM was different in the two cable samples. A spray based on graphite powder mixed in a propane butane gasmix was used in both cases, in the first cable test the graphite spray was used after mixing it with a release oil. This change in lubricant had an appreciable influence on the LTM-draw in measured in the tests of cable 2. Tensile creep testing followed by quasistatic tensile failure testing was performed on Empa's 30 MN tensile testing machine ELS 1043 (Figure 5) built by T. Erismann in 1984 (Erismann, 1984).

Table 2. Tested full-scale 37-wire CFRP cable samples for sustained tensile creep followed by tensile testing to failure (5.2 m each)

Cable number Cable 1

Duration of initial tensile creep test at 1300 kN

CFRP sleeve no 1

CFRP sleeve no 2

Organizer

Lubricant

133 hours

SN03

SN04

none

combined graphite powder in oil

Cable 2

66 hours

SN03 (reused, from cable 1)

SN04 (reused, from cable 1)

4 along free length

dry graphite powder in high doses

On both parallel wire cable samples (37 reused CFRP wires each) resistive strain gauges of type HBM 1-LY41 6/350 were bonded on two peripheral wires at midspan. For cable 2 an additional strain gauge of the same type was bonded on the central CFRP wire at midspan. In order to monitor the strains during loading on CFRP sleeves, sleeve number SN03 was equipped with two strain gauges of type HBM 1-LY41-20/120 measuring in cable (wire) direction which were positioned 180° offset. The distance of the strain gauge edge to the end of the steel supporting shell was 70 mm. No strain gauges were applied on the second CFRP anchor sleeve SN04.

Fig. 5: Cable 2 installed in ELS 1043 tensile testing machine of Empa

During loading at creep load level of 1300 kN and during the following tensile creep experiments the draw-in of the LTM into the CFRP sleeve was monitored by laser gauges. The laser gauges were removed in the quasi static tests to failure of the cables in order not to damage them. The draw-in measurement was performed with two laser gauges of type Baumer OADM 20I6460/S14F and Baumer OADM 12u 6460/S35A in cable 1, monitoring the relative displacement of the sleeve end and the LTM end-face (punctiform measurement) A line laser sensor of type MICRO-EPSILON scanControl 2660-50 was implemented in cable 2 to monitor the LTM and CFRP cable bundle draw-in during loading and creep test, which was capable of precisely tracing the