Issue 35

P. Bernardi et al, Frattura ed Integrità Strutturale, 35 (2016) 98-107; DOI: 10.3221/IGF-ESIS.35.12 103 to its effectiveness, being a duplicate of the well-known Bresler and Scordelis beam tests [20], always regarded as benchmark data. The choice of the experimental program undertaken by Podgorniak-Stanik [18] has instead been related to the availability of several experimental data monitored during test execution, mainly concerning the crack pattern evolution with increasing loads. Description of experimental tests [17, 18] The attention has been initially focused on three beams without stirrups, named OA1, OA2, OA3, tested by Vecchio and Shim [17]. These specimens had the same rectangular cross section - 305 mm wide and 552 mm deep - and a net span respectively equal to 3660 mm, 4570 mm and 6400 mm, corresponding to an increasing amount of tension reinforcement, heavy enough to make the beams critical in shear. The main geometrical details of the considered specimens and their reinforcement arrangement are summarized in Fig. 2a. The three beams were characterized by a progressively increasing concrete compressive strength f c (which was equal to 22.6, 25.9 and 43.5 MPa for specimen OA1, OA2 and OA3, respectively). The main characteristics of the adopted reinforcement, both in terms of geometrical details and steel properties, can be found in [17], to which reference is made. 305 552 OA1 OA2 OA3 64 64 M30 M30 M25 M30 M30 50 85 85 85 85 85 300 500 M25 M20 M25 M20 M10 50 M15 300 250 BN50 BN50D BN25 BN25D 25 40 40 40 40 40 M15 #3 (a) (b) M25 M25 Sample Span (mm) L (mm) Longitudinal Reinforcement OA1 3660 4100 2 M25 b , 2 M30 OA2 4570 5010 2 M25 a , 3 M30 OA3 6400 6840 2 M25 b , 4 M30 Sample Span (mm) L (mm) Longitudinal Reinforcement BN25 1352 1502 3 M15 BN25D 1352 1502 3 M15, 10 #3 BN50 2700 3000 2 M20, 1 M25 BN50D 2700 3000 2 M20, 1 M25, 10M10 Figure 2 : Geometric dimensions (in mm) and reinforcement arrangement of the analyzed beams: (a) OA [17] and (b) BN series [18]. All the tests were performed under loading control, with a central point load, until the approaching of the ultimate stage, when the procedure was switched to displacement control so to allow the evaluation of the post-peak behavior. As already mentioned, the purpose of this experimental program was to recreate, as much as possible, the Bresler and Scordelis tests [20], in terms of geometrical dimensions, reinforcement details, material strengths and loading . Compared to these latter, the beams tested by Vecchio and Shim [17] exhibited indeed a very similar behavior, with only few minor differences; as a consequence, only the specimens described in [17] have been considered in the FE analyses reported herein. In addition to this series, other four beams (originally named BN25, BN25D, BN50, BN50D) belonging to the extensive experimental program carried out by Podgorniak-Stanik [18], have been numerically analyzed. Series 25 and 50 differed from each other in terms of transverse cross-section dimensions, respectively equal to 300 mm x 250 mm and 300 mm x 500 mm, and in terms of net span, which was nearly doubled for the second one (Fig. 2b). Moreover, the two series were characterized by almost the same tension reinforcement ratio, which was equal on average to 0.85% for beams BN25 and BN50, and to 1.21% for beams BN25D and BN50D. These two last specimens contained indeed additional small longitudinal bars distributed along the web. More details about the geometric dimensions and reinforcement arrangement of the examined beams can be found in Fig. 2b. All the beams were cast by using concrete with a cylindrical compressive strength equal to 37 MPa. Mechanical properties of reinforcing steel and rebar details can be found in [18]. All the tests were performed under loading control, by applying a central point load in several steps. At the end of each step, the load was lowered to approximately 90% of its current peak value and then increased again. During all test execution, crack pattern evolution and crack width were monitored in detail.