Test Procedure

The specimens tested were subjected to two different stages (see scheme in Fig. 1): phase 1 or pre-cracking and phase 2 or long term loading test which was extended up to the unloading and recovery. In the first phase the specimens were tested in a closed-loop servo-hydraulic system according to a four-point bending with a 450 mm span and 25 mm notch, using crack mouth opening displacement (CMOD) as the control signal. In addition to the clip gauge placed at the bottom of the beam, the crack-opening displacement (wf) was monitored through a linear variation differential transducer (LVDT) fixed on a lateral side of the specimen 12 mm above the bottom surface of the beam. The process was interrupted at different wf levels (wp = 0.25 — 1.50 — 2.50). Table 4 shows the wf of each of the beams tested.

The pre-cracked beams were then unloaded and immediately after placed in a creep frame. The design of the test set up for the long term flexural test were based on the previous work by Bast et al. [7] and basically identical to that used by [4, 8, 9]. The steel frame used allows to simultaneously perform the long term test in various specimens set in a column. In this case, three specimens were loaded in each column, under a 4-point bending test configuration using steel rollers as a free supports (see Fig. 2). The lateral LVDT used to measure the wf during the first stage (which always remained settled to the beam) was used for the control of the crack evolution.

00

о

Table 3 Characterization of the FRC at 28 days

P. Pujadas et al.

Fig. 1 Diagram of the complete creep test procedure

Table 4 Variation in environmental conditions over the test days

The humidity and temperature in the S1 and S2 conditions of the experimental program were different. Table 4 shows the variation in environmental conditions over the test duration.

In the S1 the specimens were tested in a climate-controlled room under relatively constant conditions of humidity and temperature (Table 4). The S2 beams were exposed to ambient conditions with alternating humidity and temperature (as observed in Table 4). The theoretical load levels considered in the study were between 50 and 60 % of the load F_p, registered when the pre-cracking stage was stopped (see Table 5).

Table 5 Summary of loading conditions and pre-crack-width

Fig. 2 Creep frames during the specimens loading a S1 and b S2