PSI - Issue 44

Salvatore Pagnotta et al. / Procedia Structural Integrity 44 (2023) 1909–1916 Author name / Structural Integrity Procedia 00 (2022) 000–000

1915

7

4. Experimental test on the beam-column joint endowed with friction connection On the basis of the results reported above, it can be stated that the best-performing friction material is the thermal sprayed aluminum, since it provided the highest friction coefficient with a remarkable stability, only once the surface is already worn by a preliminary test. Assuming a friction coefficient of 0.6, the design of a specimen of the beam column joint described in Section 2 were carried out. The specimen is constituted by an external beam-column joint (Fig. 8 (a)), in which the column is a HEB300 steel profile 3.3 m long and a HSTC beam 1.55 m long. The T stub was made from a HEB300 profile in order to simplify the construction process. The connection between T stub and column was made with four M24 bolts class 10.9, while that between T stub and HSTCB was made with eight M20 bolts class 10.9. To reduce the components constituting the dissipative device, the friction pads were not considered, and the coating of thermal sprayed aluminum was made on the steel angles (Fig. 8 (b)). The connection between steel angles and vertical central plate was constituted by five M20 bolts class 10.9, while that between steel angles and column was made with four M24 bolts class 10.9. An axial load of 100 kN was applied on the column to simulate the effect of dead load. The displacement-controlled test was carried out by applying on the beam, at a distance from the center of rotation of l = 1.31 m, a displacement history defined on the basis of the suggestions of ACI 374.2R-13 (2013). The amplitudes of the cycles were determined by using the yield rotation (equal to 2.5 mrad) multiplied by an increasing coefficient. Five amplitudes were considered, having maximum displacement of ± 2 mm, ± 5 mm, ± 10 mm, ± 20 mm and ± 40 mm. Two cycles were carried out for each amplitude. The bolts of the friction device were torqued to achieve a preload of 17 kN, so that the yielding moment of the connection can be calculated as M y =  n b n s F pc z = 38 kNm, where  is the friction coefficient, n b the number of bolts (5), n s the number of surfaces (2), F pc the preload on each bolt (17 kN), and z the internal lever arm of the connection. The expected sliding force, equal to 29 kN, is obtained dividing M y by l. The force-displacement curve obtained during the cyclic test is shown in Fig. 9. Generally speaking, the preliminary results showed that the beam-column joint provided hysteresis loops which are wide and stable, and no damage nor cracking were observed on the specimen. The hysteresis loops are characterized by a higher value of sliding force in the case of sagging moment and a lower value in the case of hogging moment. The sliding force tended to reduce during the test, due to the same phenomena affecting the behavior of the linear dissipative device, namely loss of preload force acting on the bolts and wearing of the surfaces in contact. Similarly to the results of test 1 on the linear dissipative device, the curve is characterized by sections in which the device slid constantly and sections where it showed the “stick and slip” phenomenon, although less pronounced. The variation of force due to the latter phenomenon, being not significant, did not affect noticeably the overall response of the connection, differently from what was seen during the tests on the linear dissipative device.

Fig. 8. (a) specimen of the innovative beam-column joint endowed with friction dissipative device; (b) detail of the friction dissipative device.