PSI - Issue 44

Corrado Chisari et al. / Procedia Structural Integrity 44 (2023) 1100–1107 Corrado Chisari et al./ Structural Integrity Procedia 00 (2022) 000 – 000

1102

3

The tower has a square plan with an external 7.60-m long side, mostly constant throughout the height. The base pointed barrel vault is characterised by a 4-m span with an impost at about 3.80 m and keystone at about 6.80 m, both measured from the ground level. The wall thickness measures 1.80 m. The 2 nd , 3 rd and 4 th sectors have a height equal to 5.10 m, 7.00 m, and 4.30 m, respectively. They present similar characteristics, with symmetrical openings on all sides, except for the 2 nd sector on the church-side. The in-height development is quite constant since the 4 th sector presents a plan dimension of about 7.50x7.50 m 2 . A concrete roof having a pyramid shape with an octagonal base is supported by the 5 th sector, which in turn presents a height of 4.20 m and circular elements in the 4 corners. The access to the bell tower is guaranteed from the church at the 2 nd sector level, while internal steel staircases allow for walking on the slab of the 3 rd sector, as well as on the openings of the 4 th sector. The tower is partly connected to the church at the base sector. In fact, it is placed at the right corner of the façade, thus interacting with the external longitudinal wall and the façade of the church. Conversely, the opposite base wall presents a basement which is 1-m higher than the road, constituting a boundary for the wall in the direction opposite to the church. 2.2. Description of the structural system and structural survey The predominant material characterising the bell tower under investigation is the Campanian grey tuff: a weak microporous rock usually having a compressive strength between 1 MPa and 5.5 MPa (Augenti & Parisi, 2010). Three compressive tests on tuff units extracted from the tower were performed, which returned an average compressive strength f m =2.95 MPa and an average Elastic Modulus E m =1305 MPa. Also, five penetrometer tests were performed on mortar, but three of them returned, by means of empirical correlations, very low values of the mortar compressive strength. This leads to think that the adopted mortar has very poor mechanical characteristics. In the outer leaf, it is possible to recognize also some limestone units, especially in the base sector (i.e. at the arch intrados and in the corners of the walls). Endoscopic tests showed that, in some cases, the wall presents a limestone core well interconnected with the outer tuff leaf. This was found in the wall parallel and opposite to the longitudinal wall of the church, but these constructional details were not consistently observed in the tower. A steel girder embedded in the base masonry wall was also noted in the same wall. Inside the bell tower, a reinforced concrete (RC) frame is present. In particular, columns partially included in the corners of the 2 nd sector support a 5 cm thick RC slab (floor of the 3 rd sector). The same system is reproduced in the 3 rd sector, but in this case, the supported slab (floor of the 4 th sector) consists of a perimetral walkway allowing for the positioning of the bells in the central space. The bells are sustained by a wood-steel system supported by three RC beams anchored in the masonry walls, parallel to the church longitudinal direction and without any apparent connection to the frame. As far as the overall dynamic behaviour of the bell tower is concerned, despite the central symmetry of the structure from the second level on, different overall stiffness is expected in the two main directions as an effect of the base arch opening, with a stiffer behaviour expected in the longitudinal direction of the church. Moreover, as described above, many singular situations were observed (e.g. asymmetric base boundary conditions, presence of elements of different material, etc.), which could potentially confer additional asymmetry to the dynamical behaviour of the structure. 3. Ambient vibration testing and dynamical characterisation 3.1. Experimental setup Ambient Vibration Tests (AVTs) were performed on the tower. The experimental setup consisted of 4 triaxial and 2 biaxial force-balanced accelerometers, with 2.5 V/g sensitivity and bandwidth 0-200 Hz (Fig. 2a). The accelerometers were placed at 12.6 m and 19.4 m heights, corresponding to the second floor and belfry levels, respectively (Fig. 2b). Although large literature exists about optimal sensor placement methods (e.g., Chisari et al., 2017, and Civera et al., 2021), the vertical development of bell towers and practical constraints usually governs the position of instruments. In the present case, it was not possible to reach the last level, while the first floor, strongly