PSI - Issue 62

Alessandro Bellini et al. / Procedia Structural Integrity 62 (2024) 315–322 A. Bellini et al. / Structural Integrity Procedia 00 (2019) 000 – 000

320

6

5. Analysis of experimental results The main outcomes of experimental tests, analyzed in terms of effects of different cutting procedures on the shape of experimental curves and accuracy of results, will be summarized and discussed in this section. 5.1. Effects of saw-cuts direction (parallel or orthogonal to the applied axial stress) Even if the final results, in terms of residual strain, should not be affected by the cutting procedure, the main effect of saw cuts direction is a local modification in the shape of the strain vs time curve. In fact, as already discussed during the presentation of the typical behavior identified during prestress release tests (Fig. 4), cutting the sample in the direction parallel to the applied axial stress usually offers negligible effects in terms of strain release, simply isolating the volume of the specimen to be extracted from the surrounding concrete. Saw-cuts along the orthogonal direction, on the contrary, are the most important ones, because they are able to interrupt the stress field acting on the specimen, causing an immediate stress release, provided that the saw-cut reaches an adequate depth. It should be noted, however, that this behavior directly derives from the stress field present in the uniaxial compression test of the column and may not be valid in general; for example, in the case of simultaneous presence of bending and shear stresses, common case in beams, each saw-cut direction (vertical and horizontal) is expected to have a more comparable effect since the stress field could be inclined with respect to the longitudinal axis. To investigate this aspect, a new experimental research, which comprises additional experimental tests on real-scale prestressed concrete beams, similar to those generally used in bridges and viaducts, is currently in progress. 5.2. Effects of saw-cuts inclination with respect to the concrete surface (45° or 90°) The inclination of saw-cuts (45° or 90°) with respect to the external surface of the column is fundamental during the extraction procedure, as it can directly affects the area of the sample which remains attached to the surrounding concrete at the end of the cutting procedure and therefore the vibrations produced at the end of the test during the final removal of the sample. Even if this is not a negligible effect on the ease of execution and on the overall success of the test, saw-cuts inclination has a secondary role on the shape of the strain vs time curve. In fact, a perpendicular (90°) inclination produces a slightly clearer effect in terms of strain release when the cutting procedure is still in progress (that is a greater local slope of the curve) but, after a short recovery time and, in any case, at the end of the test, results produced by a different inclination of the saw-cuts tend to be absolutely similar. If the difference, in terms of strain release, between two saw-cuts made as the same step in two different tests is large, this is mainly due to the possible difference in the cutting depth; this suggests that, as for the previous effects, a proper standardization of the procedure is necessary for obtaining reliable and repeatable results. 5.3. Effects of the cutting depth After defining the cutting order and the procedure for extracting the sample during prestress release tests, the cutting depth remains another parameter that can locally affects the results. In fact, if the penetration depth of the angle grinder is not sufficient, the strain release could be only partial as it involves only a very superficial part of the sample. This behavior was observed in preliminary tests performed in similar sample before the beginning of the experimental program, varying the cutting depth, but using the same cutting procedure. Results coming from these preliminary tests will not be reported here in full for reasons of space. 5.4. Effect of the intensity of applied axial force The behavior identified during prestress release tests is not strictly dependent from the applied axial stress. In fact, for the three different stress levels applied (5, 10 and 15 MPa, see Table 1), the behavior was absolutely comparable in terms of strain vs time graphs and the considerations previously expressed are valid for all the three groups of samples. The same conclusion can be obtained by considering the residual error values, with or without the effects of the differential drying shrinkage (Table 2): observing the results, a correlation between errors and intensity of the applied stress cannot be found, since the former are very similar, meaning that the procedure is reliable and independent from the prestress level present in the concrete element.

Table 2. Experimental results: residual error identified during prestress release tests. Sample type σ (MPa) Residual error without shrinkage effect ( % )

Residual error considering differential shrinkage ( % )

P5

5

8.1 7.2 8.6

6.0 6.6 6.5

P10 P15

10 15