Seismic behavior of precast CFST double-column piers with various connection types

To investigate suitable connection details for precast CFST double-column piers, three specimens were subjected to low cyclic loading tests. The primary research variables considered were the connection types both in column-foundation and column-beam joints, namely cast-in-place (CIP) connections, socket connections, and U-shaped anchor rebar (URA) connections. Detailed evaluations were conducted on the seismic behavior of these specimens, including failure modes, strain responses, load-carrying capacity, ductility, stiffness, and energy dissipation. The experimental results indicate that the specimen with socket connections (with socket depths of 2.05d and 1.6d in foundations and beams, respectively, where d represents the outer diameter of the steel tubes) can exhibit seismic behavior comparable to that of the specimen with CIP connections. Owing to the opening of column-beam joints, the load-carrying capacity of specimens featuring URA connections in column-beam joints and socket connections (with a socket depth of 2.05 times the diameter, denoted as 2.05d) in column-foundation joints may be significantly lower than that of specimens with either all-CIP or socket connections. However, a superior deformation capacity could be achieved. Additionally, based on ABAQUS software, finite element (FE) models were developed for parameter analysis. The analytical findings suggest that, in contrast to CFST single-column piers, where an optimal socket depth of 1.0d or even 0.625d suffices in some studies, the optimal socket depth for CFST double-column piers with multiple sockets should not be less than 1.2d. This is crucial to ensure an equivalent hysteresis performance to CIP connections. Furthermore, the strength of the socket concrete (ranging from C40 to C80) has little impact on the seismic performance. •Three specimens with different connection types were tested under cyclic loads.•Specimen with socket connections exhibit similar behavior to cast-in-situ specimen.•Optimal socket depth of double-column piers should exceed 1.0d (steel tube diameter).•Finite element models were created to predict the hysteretic behavior of specimens.