Evaluation of the behavior and ultimate capacity of unbonded monostrand-anchorage systems under concentric and eccentric inelastic cyclic loading

Highlights

• A total of 108 monostrand anchorages with different sizes and types were tested.

• Several cases of cyclic amplitude, frequency and eccentricity were applied.

• Measurement system captured the ultimate deformation reliably.

• Early fractured of only one or few wires inside anchorages limited the capacity.

• An analytical model of the anchorage behavior is proposed.

Abstract

Unbonded post-tensioning (PT) monostrands have traditionally been used in buildings to sustain monotonic loads on members subjected mainly to gravity loads. As a result, most of the technical information on commercially available unbonded PT anchorage systems focused only on applications for gravity-loaded members. Despite the recent experimental research efforts on the behavior of PT anchorages under earthquake-simulated demands, this information has not been well quantified in design documentation. Consequently, further research is needed on unbonded PT anchorages for their use in members subjected to seismic loads. This paper presents the results of a comprehensive experimental evaluation on the ultimate capacity of monostrand anchorages subjected to high-amplitude concentric and eccentric cyclic loads. In addition, this paper addresses the influence of anchorage type, loading patterns and strand size on the ultimate deformation capacity of these anchorages. Specimens consisted of monostrands assembled with anchorages at both ends. Two types of anchorages were tested. Moreover, seven-wire, uncoated, low-relaxation PT monostrands in two sizes were considered. Several cyclic loading conditions were applied, in order to evaluate the behavior of anchorages under different scenarios. The ultimate capacity of the specimens was dominated by a premature fracture of one or few wires inside the wedges and two types of wire fractures were recognized: (1) with little reduction in the wire cross-sectional area and (2) with a notable reduction in the wire cross-sectional area prior the fracture. It was also observed that increased number of wire fractures correlated with larger strain capacity in some specimens. In addition, it was observed that eccentric loads reduced the strain capacity by about 18% for eccentricities of 6% in some cases. The fracture of the specimens occurred at relatively small strains, with values as low as 1.4%. Therefore, strains in unbonded post-tensioned strands should be limited to about 1.0% when designing for seismic loads. Furthermore, an analytical model was proposed to evaluate the influence of anchorages in unbonded post-tensioning precast structures.

Introduction

In the past few decades, the earthquake engineering community’s attention has been drawn to assessing and controlling the damage produced in buildings after major earthquake events, with the purpose of mitigating economic losses and providing resilience. The motivation for this new approach was the occurrence of significant damage in code-compliant structures and the subsequent economic losses involved in repairing that damage after the seismic event [1]. For instance, it has been reported in recent earthquakes that code-compliant reinforced concrete (RC) buildings sustained major damage, leading to significant post-earthquake economic losses and business downtime [2], [3], [4]. Therefore, reducing residual damages and improving the post-earthquake structural performance remains an important challenge.

Self-centering rocking structures have been proposed to reduce the residual damage (damaged-controlled structures) and to improve the post-earthquake performance of RC buildings. One such type of structural systems was introduced in the mid-1990s, which consisted of precast concrete (PCa) elements connected by unbonded post-tensioning tendons [5]. Although different configurations have been proposed over the years, all of them share three basic components (Fig. 1): (1) precast structural elements not rigidly connected to allow gap opening, (2) a restoring force mechanism (unbonded post-tensioning tendons), and (3) additional energy-dissipation mechanism. These self-centering rocking systems proved to be very efficient in mitigating residual deformations and controlling the damage of RC structures in large-scale experiments [5], [6], [7], [8]. The superior performance shown experimentally later prompted the application of self-centering rocking mechanisms in the seismic-resistance design of structures with other materials such as timber [9], masonry [10] and steel [11].

Although some other means to provide self-centering capacity have been proposed, such as shape memory alloys [12], [13] or fiber-based strands, the most common method used in RC building structures relies on unbonded post-tensioning steel strands. In unbonded post-tensioned structures, the strands transmit the tensioning forces to the main structure by means of anchorage systems inserted in the concrete. In consequence, the anchorage systems are essential to ensuring the restoring force and the safety of the overall structure. Unbonded post-tensioning (PT) monostrands have traditionally been used in buildings and bridges to sustain monotonic gravity service loads [14]. As a result, most of the technical information on commercially available unbonded PT anchorages focuses only on applications for gravity-loaded members [15]. Moreover, their design approach was based on research that has been conducted to evaluate PT stresses associated with the service and ultimate capacity of members subjected only to gravity loads [14], [16], [17]. When applying these conventional PT anchorages to members subjected to lateral loads a different condition is introduced, since strands are expected to be subjected to short-duration repeated stress variations during a seismic event. For instance, it has been reported in recent shaking table tests of unbonded post-tensioned precast concrete buildings that PT tendons underwent several cycles of large-amplitude elongations at moderate frequencies [7], [8], [18].

This paper describes the details and results of a comprehensive experimental program conducted to evaluate the behavior and ultimate capacity of unbonded monostrand PT anchorages subjected to cyclic loads. In addition, this paper discusses the influence of cyclic loading parameters on the deformation capacity of strand-anchorage systems. The development of an analytical model for the response of monostrand-anchorage systems is also discussed.