Experimental Assessment of the Out-of-Plane Performance of Masonry Buildings Through Shaking Table Tests

Abstract

This paper presents the results of the LNEC-3D shaking table tests on two mock-ups, Brick House and Stone House, carried out in the scope of the workshop "Methods and challenges on the out-ofplane assessment of existing masonry buildings". The mock-ups have a U shape with one façade wall and two orthogonal sidewalls. The façade has a central opening and a gable on top, whereas the two sidewalls, acting as abutments, are either blind or have a window. A unidirectional seismic action, in the perpendicular direction to main wall, was applied. Out-of-plane behaviour of the façade was found, even if the response was clearly influenced by the presence of the window in one of the sidewalls, which led to significant torsion of the structure. The detailed description of the two tests and the conclusions are presented. The response of the mock-ups was evaluated based on the displacements, damage and collapse mechanisms developed as function of an increasing intensity earthquake testing protocol, in which a pre-processed strong ground motion component of the Christchurch (New Zealand) earthquake (February 21 st , 2011) was used. IntroductionThis paper focuses on the experimental study of the out-of-plane behaviour of masonry structures under seismic loading and provides a ground for the validation of the different approaches currently available for assessment of existing structures. For that purpose, tests were carried out in the LNEC-3D shake table with two mock-ups: one built using regular clay brick masonry with English bond and the other built using multi-leave granite stone masonry. The mock-ups have a U shape with one façade wall and two orthogonal sidewalls. A unidirectional seismic action perpendicular to main wall was applied. Several experts were invited to predict the out-of-plane seismic capacity prior to the disclosure of the test results, in what is commonly known as a blind prediction test.The main goal of the exercise was to promote a debate about the challenges of existing methods for the out-of-plane assessment of existing masonry buildings, leading to the definition of research needs, rather than evaluating the quantitative differences between the numerical predictions and the experimental results. Each participant was allowed to define the main control outputs according to the structural analysis method adopted, such as the collapse peak ground acceleration, ultimate displacement or energy dissipation.A qualitative assessment was carried out with the purpose of identifying the trends in the predictions of the damage patterns and collapse mechanisms, which allow highlighting the pros and cons of the different methods, discussing their applicability for engineering assessment of existing buildings and comparing the different results with blind experimental tests. To this end, a reference response spectrum and some basic mechanical properties were supplied to the participants, while a clear indication of the performance levels reached and corresponding response spectrum scaling factors had to be indicated b...

Assessment of the dynamic response of unreinforced masonry structures using a macroelement modeling approach

“…The first case study corresponds to a prototype of brick masonry subjected to shaking table tests at National Laboratory for Civil Engineering in Lisbon, Portugal . During the experimental campaign, the structure was loaded up to collapse aiming at the assessment of the out‐of‐plane response because of dynamic solicitations.…”

“…Brick masonry prototype (A) and collapse mechanism from shaking table tests: (B) main gable wall and (C) return wall with window opening [Colour figure can be viewed at wileyonlinelibrary.com]…”

“…The prototype was subjected to 1‐directional horizontal component of dynamic excitation, oriented orthogonally to the main gable wall. The tests were performed by applying a horizontal component of the 2011 Christchurch Earthquake (New Zealand) accelerogram with increasing intensities . The failure mechanism of the prototype was reached when it was subjected to a PGA of 1.27 g, and consisted on the out‐of‐plane collapse of the tympanum at the main gable wall, and the in‐plane collapse of the return wall with opening, mainly because of a rocking response (see Figure B, C).…”

Methodology for practical seismic assessment of unreinforced masonry buildings with historical value

“…Analysis is performed by using the following: (1) hand calculations; (2) nonlinear dynamic analysis of 3‐D frame models, where the walls and the lintels of the examined building are idealized as prismatic linear members, with prescribed postulated moment‐rotation response curves along each principal plane of action; (3) linear finite element dynamic analysis (FEM) based on modal superposition; or (4) discrete element models—their complexity is beyond the scope and objectives of the paper. Yet the actual seismic response of heritage URM buildings subjected to strong earthquakes (Thessaloniki 1978, Athens 1999, L' Aquila 2009, Amatrice 2016 & 17) and reports on conducted tests to experimental structures challenge the accuracy of these analysis procedures …”

“…In the absence of stiff diaphragms, masonry walls are recommended to be examined independently (eg, draft guidelines of EC8‐3 for URM structures). Even in the presence of rigid diaphragms, simulation of a URM structure as 3‐D frame model is mostly suitable for calculating the planar response of the masonry walls and not their out‐of‐plane seismic response, which is the most usual cause of damage or collapse in URM structures Furthermore, using nonlinear moment‐rotation envelope curves with a strain hardening postyielding plateau to model the behavior of masonry walls, piers, and spandrels is ill‐suited for the deep unreinforced members of a masonry structure, for which any ductility is not material‐related, but system‐related. To further complicate matters, modern seismic codes prohibit application of sophisticated analyses procedures in buildings with limited “knowledge level” of material and geometric properties, such as the URM historical buildings (EC8‐3)…”