Experimental estimation of energy dissipation in rocking masonry walls restrained by an innovative seismic dissipator (LICORD)

Abstract

This paper presents an innovative anti-seismic device for controlling the out-of-plane rocking motion of masonry walls with traditional tie-rods, called LInear COntrolled Rocking Device (LICORD). LICORD is a low-impact box connected to the extremity of the traditional tie-rod designed to mitigate rocking for medium–high intensity earthquakes. Additionally, the paper widens the knowledge about the dynamic behavior of rocking walls through the interpretation of the results of an extensive experimental campaign performed on masonry specimens composed by clay brick and cementitious mortar. Firstly, the LICORD’s single components are tested to identify their stiffness and damping properties. Secondly, free vibration tests provide actual values of coefficients of restitution on free-standing walls and walls restrained by LICORD, where the walls vary for the height to thickness ratio. For the stockier wall, the ratio of experimental/analytical coefficient of restitution varies from 88 to 98%, whereas for the slender wall, the results are less scattered, with a minimum value of 95% and a maximum value of 96%. The restrained walls are characterized by coefficients of restitution from 5 to 25% less than the values found for unrestrained walls, depending on the equivalent viscous coefficient of the shock absorbers. Moreover, LICORD demonstrated to properly absorb and damp the oscillations of the wall and control its rocking motion, strongly reducing the number of impacts and the rotation amplitudes up to 70%. Considerations about the effect of one-sided motion on the assessment of coefficient of restitution are also given. The equivalent viscous damping coefficients are observed to be on the range 4% (unrestrained wall) and 7–20% for walls restrained by LICORD.

Reduction of Housner’s Coefficient of Restitution for Masonry Walls Under One-Sided Rocking

“…Therefore, the calculation of the coefficient of restitution, function of the velocity peaks, should consider such a relevant aspect. This issue was already highlighted for diverse properties of horizontal restraints in [25] and for non-homogeneous walls in [22].…”

“…The wall is pulled out-of-plane through a tie-rod up to attain a specific displacement (50 mm, 100 mm and 150 mm) of the wall at 2.56 m from the base. A detailed description of the experimental tests and of the mechanical features of the wall can be found in [24], [25]. The only difference with the tests already performed is that the hinge is in this case formed at 0.23 m above the wall base, and not at the wall foundation.…”

“…The transverse walls have the same mechanical properties as the rocking wall. The displacement and acceleration time-histories are recorded by potentiometric displacement transducers and accelerometers respectively, positioned on the tested wall as shown in [25]. An example of the experimental outputs is reported in Figure 3, where also the velocity time histories, obtained by derivation of the displacement time-histories, are plotted.…”

“…The first and third condition refer to the experimental tests described in [22], [25], whereas the second condition regards the experimental tests illustrated here.…”

“…For the tested wall, the values of coefficient of restitution range from 88% to 98%. In [25] the tests were executed also for a slenderer wall of equal characteristics: for it, the results were less scattered, being the minimum 95% and the maximum 96%. A reference value of 95% can be then assumed for the unrestrained walls in two-sided rocking.…”