Exposed Column Base Connections Subjected to Axial Compression and Flexure
Licensed under
Category
Published on
Abstract
This report presents results of an experimental study investigating the response of exposed column base connections subjected to axial compression and strong-axis bending. This investigation is the second phase of a broader study whose ultimate aim is the development of improved design guidelines for column base connections. A review of existing design procedures and published research reveals that current approaches to characterize the strength of exposed column base connections loaded by axial compression and flexure are not supported by adequate experimental validation. Thus, the main objectives of this study are to evaluate existing strength prediction approaches and propose refinements based on experimental verification.
The main scientific basis of this study consists of a series of seven large scale experiments on exposed column base connections subjected to a combination of axial compressive load and cyclic lateral deformations. The tests investigate the effect of various parameters on the response of the connections, including (1) the base plate thickness (2) anchor rod layout (3) presence and level of axial load (4) anchor rod strength and (5) applied loading history. The column base connection tests are complemented by ancillary tests to characterize material properties of the anchor rods, base plate, grout and concrete.
An evaluation of the experimental data relative to the existing design approaches reveals that existing strength prediction methods (such as those outlined in the Design Guide 1 published by the AISC) may be highly conservative (i.e. the average test-to-predicted ratio for connection strength is 1.86). The methods outlined in the Design Guide assume that failure of the base connection is governed by lowest base moment which activates the limit state of only one component in the connection (e.g. anchor rod capacity or yielding of the base plate in flexure). However, the experimental data suggests that the ultimate strength of the connection is governed by the formation of a plastic mechanism wherein multiple limit states are activated. To overcome this conservatism, a method for characterizing the strength of base connections is proposed. The method incorporates several favorable features from existing approaches such as a method, outlined in the Design Guide 1, which assumes a rectangular shaped bearing stress distribution. However, the improved method proposes two key refinements, which include (1) incorporation of a mechanism-based approach for determination of the ultimate connection strength and (2) consideration of alternate yield line patterns and the ultimate strength of the plate material for determination of the bending capacity of the base plate. The resulting approach is shown to characterize the strength of the base connection with significantly improved accuracy, such that on average, the test-to-predicted ratio for connection strength is 0.99, with a coefficient of variation of 6%.
All the experimental specimens show outstanding ductility (deformation capacities in excess of 6% column drift) and hysteretic energy dissipation. Only two of seven specimens show catastrophic failure, due to fracture of one anchor rod. This suggests that under seismic loading conditions, reliable inelastic action is possible in the base connection.
The report concludes by presenting strategies for incorporating the findings of this study into design guidelines and outlining on-going and future work.