PSI - Issue 78

Laura Ragni et al. / Procedia Structural Integrity 78 (2026) 2094–2101

2095

while seismic responses are mainly governed by short-term viscosity, long-term viscosity becomes significant during slow loading, leading to stress relaxation and displacement loss. Recent studies, including those on unbonded fiber reinforced elastomeric isolators (Dall’Asta and Ragni 2006, Tubaldi et al. 2017, Di Domenico et al 2023), confirm that HDRBs show limited sensitivity to velocity changes within seismic velocity ranges. However, in real-scale push and release tests on base-isolated buildings, the dynamic response will only reflect seismic behaviour if the pushing phase is fast enough. If the pushing is too slow, often due to limited hydraulic capacity in field testing as well as the high forces involved (as all bearings are tested together), the long-term viscous effects dominate, altering the observed response (Braga and Laterza 2004, Athanasiou 2025, Dall’Asta et al. 2022). To address these complexities, the paper presents a detailed analysis of a recent experimental campaign on the CHIP building, a two-story steel braced frame structure with a hybrid isolation system combining HDRBs and Low Friction Sliding Bearings (LFSBs). The building’s design and isolation system are documented in prior studies (Dall’Asta et al. 2020), while this paper synthesized the viscous behaviour observed during testing, described in detail in (Dall’Asta et al. 2022). In particular, both in-field tests and laboratory qualification tests (type tests) of the HDRBs, allowing for separate evaluation of short- and long-term viscous effects, are described. Then, a simple linear viscoelastic model is used to interpret the results. This model, consisting of a Maxwell element in parallel with a Kelvin element, was chosen for two main reasons: ( i ) to demonstrate that deviations from expected seismic behaviour are primarily due to slow-rate viscosity, and ( ii ) to provide a practical, easily calibrated tool using both field and lab data. More in details, the model was calibrated for a single HDRB using the type test data, then the model has been extended to simulate the entire isolation system, including the friction from the LFSBs. It is shown in the paper that the model successfully captures the viscous displacement loss during slow loading and the subsequent free oscillation around a non-zero displacement. It also accurately predicts the final residual displacement once long-term viscosity effects dissipate, confirming the system’s recentring capability (provided the low friction of LFSBs). It is worth to note that at seismic velocities, the Maxwell element behaves like a spring, and the combined stiffness of the Maxwell and Kelvin elements matches the nominal dynamic stiffness used in seismic design standards. Thus, the model aligns with standard equivalent linear models for seismic analysis while also capturing low-velocity behaviour. In the last part of the paper the model was in fact used to simulate faster loading ramps, leading to the expected free vibration under the initial position without final displacements. 2. In-situ experimental tests on the ChIP building The tested structure is the ChIP (Chemistry Interdisciplinary Project) building, a new research center at the University of Camerino, housing chemistry and physics laboratories. Funded by the Italian Department of Civil Protection (DPC) following the 2016–2017 Central Italy earthquakes. The superstructure is a steel braced frame with pinned joints, based on a 7.2 m × 7.2 m modular grid, comprising seven modules in each direction and a 1.9 m cantilevered perimeter (Fig. 1). The substructure includes reinforced concrete foundations at varying depths and rigid columns connecting them to the isolation level. The seismic isolation system consists of 28 High Damping Rubber Bearings (HDRBs) with soft rubber (600 mm diameter, 184 mm rubber height) placed around the perimeter, and 36 Low Friction Sliding Bearings (LFSBs) in the center to support higher vertical loads (Fig. 2).

Level 2 (roof)

Level 1

Level 0 above isolation Level 0 below isolation

Level -1 Level -2

Push-and-release device

Isolation plane

Fig. 1. Section of the CHIP building

The system was designed for a nominal isolation period of 3.60 seconds. To perform in-field tests, a custom push and-release device was integrated into the building design. It features a quadrilateral articulated steel frame with a