Dynamic Behavior of Multiple-Anchor Connections in Cracked Concrete

Anchors are widely used to attach structural and non-structural elements to concrete structures in nuclear, highway, and building constructions. However, the behavior of anchors and anchor connections is still not well understood, especially for anchors in cracked concrete under dynamic loading.

In an experimental research program conducted at Ferguson Structural Engineering Laboratory in The University of Texas at Austin, various configurations of anchor connections were tested. The behavior of single anchors was examined in cracked concrete, under static and dynamic loading. The behavior of near-edge, two-anchor connections under both static and dynamic loading was observed, the effect of hairpins (transverse reinforcement on the front anchor) was also assessed. Finally, the behavior of multiple-anchor connections was evaluated under earthquake-type loading in two combinations of moment and shear, with different conditions of concrete specimens, such as cracks, edges, and hairpins. Based on these test results, the behavior and design of anchor connections in cracked concrete and under dynamic loading are addressed.

In the analytic research, the tensile behavior of single anchor in concrete was modeled with axisymmetric finite elements. It was accomplished with the fixed smeared-cracking concept, using a progressive approach, which predicts the crack propagation and assigns the tensile cracking properties of concrete only to elements along the crack path. Predicted capacity and crack path were similar to those observed experimentally. The concrete plastic deformation around anchor head was not considered; therefore, the displacement behavior of anchor was not correctly predicted. A macro-model program (BDA5) provided by the University of Stuttgart was also used to predict the load-displacement behavior of multiple-anchor connections, based on single-anchor load-displacement behavior under oblique loading obtained in the earlier stage of this testing program. Its predictions compared reasonably well with the corresponding test results.