Interaction of variable axial load and shear effects in RC bridges

Historically, earthquake demands have been thought mainly to be a result of horizontal actions and the vertical component of earthquake motion has been entirely neglected or treated only as a secondary effect. The underestimation of the vertical component coupled with limitations in laboratory capabilities has resulted in deficiencies in the experimental database. The vast majority of experiments conducted on reinforced concrete piers or columns have utilized constant axial compression or no axial load whatsoever. Only a few tests have been conducted with variable axial load and fewer still with an axial loading pattern that was uncoupled from horizontal demands. Likewise, very few tests have been conducted under axial tension; while analytical study and field investigations both suggest tension is possible. All of the noted deficiencies noted have propagated into design and assessment approaches which stem mainly from the experimental findings. At a minimum, currently available methods have not been properly verified due to the lack of experimental data.;This dissertation addressed the effects of variable axial load and/or tension on reinforced concrete bridges, particularly the behavior of the piers. Two large-scale hybrid analytical-experimental simulations were used to experimentally verify the effects of vertical motion on pier loading and behavior. Two additional large scale tests were conducted to allow direct comparison of pier response under tension to pier response under compression. In the second phase of the study, 37 small-scale tests were used to extend upon the findings of the large-scale investigation and fortify the experimental database. The large number of tests made possible extensive treatment of possible axial loading patterns and effects.;The two hybrid simulations directly compared pier response under horizontal motion only to pier response under combined vertical and horizontal motion. The vertical motion was shown to cause a highly variable axial load that included extreme levels of axial compression and three cases of axial tension. More severe cracking and damage, including large increases in spiral strains, were observed in the pier tested with vertical motion.;The second pair of large-scale tests compared pier response under cyclic lateral load combined with either constant tension or constant compression. Brittle shear-axial failure was observed in the test with constant compression while a relatively ductile flexural dominant failure was observed in the pier tested with constant tension.;Five small-scale tests were designed to investigate the influence of scale and provide verification for small-scale study. Strong agreement was achieved between the two scales. Though some size dependent behavior was observed, the effects were inconsequential to the relative comparisons made amongst similarly sized specimens used in this study.;An additional 32 small-scale tests were completed to investigate the influence of the amplitude of axial loading cycles, the frequency of axial loading cycles, and sequencing or phase of axial loading peaks relative to lateral loading. Generally, behavior of the specimens including aspects of strength, stiffness, energy absorption, and failure mode were found to be highly dependent on the axial loading pattern. High levels of constant or coincident compression resulted in increases in demands that were unmatched by any increases in capacity resulting in brittle shear-axial failures. Constant or coincident tension was found to reduce peak shear loads and promote relatively ductile behavior. High frequency oscillation of axial loading was found to be more damaging than comparable constant loads. Finally, the sequencing or phasing of peak axial loads was critical to behavior and could on its own dictate failure mode. This investigation clearly demonstrated the detrimental effects of variable axial loading.