Seismic Shear Strength of Bridge Piers

The initial section highlights the inherent difficulty in designing structures, particularly bridge piers, to resist shear failure under seismic conditions. Reinforced concrete members are commonly subjected to axial load, flexure, shear, and torsion. Since shear failures are non-ductile and catastrophic, modern earthquake-resistant design methods aim to prevent them. However, the non-linear properties of composite materials and geometric discontinuities from cracking make a purely analytical approach to optimal design extremely complex. Current practice often employs over-design for shear using capacity design principles to ensure a ductile flexural failure mechanism, even though this can lead to inefficient material use. The unpredictable nature of earthquake attacks and the possibility of increased flexural strength beyond design values further complicate the challenge of completely avoiding shear failure. The study emphasizes that the random nature of seismic loading differs significantly from monotonic shear, underlining the need for improved understanding of the behavior of spirally reinforced columns under these conditions.

This section discusses the prevailing design philosophy for earthquake-resistant bridges. The economic reality dictates that complete resistance to the strongest possible earthquake is infeasible. The focus is on designing bridges to withstand minor to moderate earthquakes without damage and to resist strong earthquakes without collapse, though some structural and non-structural damage might occur. Ideally, damage should be readily visible and accessible for repair to ensure some level of continued functionality. The text points out that designing for plastic hinges in the superstructure during an earthquake is both impractical and undesirable due to the combined influence of factors such as impact, temperature, creep, and shrinkage. Plastic hinging in foundation structures should also be minimized to reduce damage in less accessible areas. As a result, bridge piers are typically designed to yield under strong seismic events, dissipating energy through plastic hinges and limiting seismic load input to the structure. This approach allows for less severe loading requirements compared to elastic response designs.

This section justifies the research by emphasizing the lack of sufficient data on spirally reinforced circular columns subjected to repeated reversed cyclic loading. While the behavior of reinforced concrete members under shear has been extensively researched, relatively little attention has been given to spirally reinforced circular columns under combined loading conditions and cyclic loading. This is particularly crucial for earthquake-resistant design, especially considering the prevalence of squat columns in certain regions like Japan, where short columns are often used in various buildings. In bridge piers, low aspect ratios frequently result from high seismic design coefficients demanding large-diameter columns. The existing research gap makes this study essential to enhancing the understanding of shear performance of spirally reinforced circular bridge piers under seismic events. The study aims to address this gap to help improve the design process for these critical structural components. Prior research, while extensive on other types of concrete members, did not provide the necessary information for the specific focus of the current study.

This section reviews existing research on shear behavior in reinforced concrete members. Past studies focused on identifying basic behavior of various shear-resisting mechanisms, improving both qualitative descriptions and quantitative evaluations. While significant progress was made, much of the work focused on monotonic loading, which limits its direct applicability to seismic situations. Code provisions often exhibit conservatism when extrapolating these results to seismic conditions. Two main theoretical approaches to modeling shear behavior are discussed: the additive principle (attributing excess strength to concrete mechanisms like aggregate interlock and dowel action) and modifying the 45-degree truss model to incorporate both forces and deformations within the failure region. This section concludes that the variable truss model, such as that employed in the Diagonal Compression Field Theory, and plastic analysis may be superior to the traditional 45-degree truss analogy for modeling the complex behavior of reinforced concrete members under seismic loading. The focus has been mainly on modeling stiffness loss from shear deformation, while the equally important characteristics of strength and ductility have received less attention.

The experimental program involved the testing of 25 circular bridge pier models. These models were specifically designed so that shear strength or ductility would be the limiting factor in their performance. The design considered several key parameters, including spiral reinforcement content, column aspect ratio, and axial load. Details of the column units, including longitudinal reinforcement specifics (with variations in Units 2, 14, 15, and 25, using different grades of steel and numbers of bars), are referenced but not explicitly detailed here. The support beams were cast in sets of three (except for Unit 25, which lacked transverse reinforcement), using plywood molds. The circular form of the columns was achieved through the use of flexible sheet metal molds. The fabrication of the reinforcing cages involved creating the support beam cage first, then tying the vertical column bars to it, ensuring adequate anchorage with 90-degree hooks embedded in the beams and welded to an annular steel plate at the top. Sufficient spirals were used to hold the column bars securely.

The vertical piers were subjected to quasi-static horizontal loading at a fixed height above the base. Loads were incrementally increased in steps, and the loading cycles varied between pre-determined peak displacement ductility levels. The primary method of monitoring performance was observing the load-deflection hysteresis response. The testing procedure involved an initial phase of five load-controlled cycles to about 75% of the ideal flexural strength to establish the yield displacement. Subsequently, displacement control was used to reach higher displacement ductility levels (µ). The standard loading sequence was initiated with five cycles to 75% of the ideal flexural strength to determine the yield displacement. For subsequent cycles, loading proceeded from positive to negative peak values without intermediate stops. The testing was concluded when significant strength and stiffness degradation was observed. The methodology allowed the researchers to study the response of the columns in detail and how they degraded under repeated loading cycles.

Strain gauges were strategically positioned to measure concrete strains, initially following a likely 45-degree crack pattern, later adjusted to a 35-degree inclination based on observations. The data acquisition system comprised a 200-channel Solartron Analogue Scanner, Data Transfer Unit, Digital Voltmeter, and Facit Printer (or Tape Puncher). Initially, readings were printed directly, but later, punched paper tapes were used for faster data reduction using a computer program and the CANDEPACK facility at the university. The system allowed for detailed measurements of strain and other relevant parameters. The initial placement of strain gauges followed an assumed 45-degree crack path, but this was adjusted after the first six units were tested. The final data acquisition was done via a computer system allowing for faster processing of the collected data from the numerous gauges that monitored several aspects of the columns' response to the various loading conditions.

The results of the 25 statically tested column units are analyzed collectively to correlate the influence of various parameters with overall performance. Column performance is assessed in terms of achieved shear strength and displacement ductility, hysteresis behavior, and energy dissipation characteristics. Other quantities such as spiral strains and forces, curvature distributions, and concrete strains are also considered. The results are compared with code-specified shear strengths and predictions from plastic theory. It was observed that except for Unit 9, all column units lacked sufficient margin of ideal shear strength over flexural strength to satisfy capacity design requirements. This principle mandates a shear strength at least 1.22 times the flexural strength for Grade 380 steel. The study found that even Unit 9 would be considered inadequate under strict application of this principle. The final failure mode was strongly influenced by the amount of spiral reinforcement, and various failure modes and their associated characteristics are discussed in detail.

A key finding is the significant influence of spiral reinforcement content on column performance. Increased spiral reinforcement content led to higher maximum loads achieved, delaying the onset of shear strength decay to higher displacement ductility levels. The rate of shear strength decay was initially faster with less spiral reinforcement, tapering off at higher ductility levels. Columns with lower aspect ratios tended to exhibit lower load-carrying capacity compared to those with higher aspect ratios, within the same group. However, the rate of strength decay did not vary significantly with changes in aspect ratio. Columns with limited or no ductility showed a relatively constant drop in strength between the first and fifth cycles at different ductility levels, while those with moderate ductility experienced more drastic drops. The presence of axial load further accelerated strength decay.

Energy dissipation is analyzed as a crucial performance aspect under seismic loading. The area enclosed by hysteresis loops represents the energy dissipated through inelastic deformation. Shear-dominant members, exhibiting pinching hysteresis loops, showed reduced energy dissipation. The study investigated the influence of spiral reinforcement content on energy dissipation, finding that the greatest differences were attributed not just to the variation in reinforcement but also to the resulting changes in failure mode. Columns with moderate ductility demonstrated stable energy dissipation, whereas those failing prematurely in shear showed considerably lower energy dissipation. The relative strength index (Vif/V) was identified as a crucial factor influencing energy dissipation performance. Columns experiencing spiral fracture exhibited significantly lower cumulative energy dissipation until the first fracture occurred compared to columns with moderate ductility.

Concrete strains at incipient crushing and spalling were estimated through linear interpolation of strain gauge readings. These strains were found to be highest at the location of maximum bending moment. The timing of crushing and spalling was not precisely determined in terms of ductility level. The curvature distribution showed a significant influence of shear deformation at higher ductility levels (µ=6), with curvature potentially reversing near the loading end. The results highlight the substantial influence of shear effects on concrete behavior and the importance of considering them in the design process. Comparisons with the New Zealand Concrete Design Code (NZS 3101) and other code approaches revealed differences in predicted shear strengths. The experimental results are compared with existing code provisions, such as the New Zealand Concrete Design Code, and discrepancies are observed, highlighting the need for improvement and refinement in the existing design procedures.

This section details the derivation of design equations for computing various strength and ductility values for circular reinforced concrete columns under seismic loading. The proposed design approach enables the assessment of shear and flexural strengths, as well as displacement ductility capacity. By calculating the displacement ductility capacity, the failure mode of the column can be predicted. Results from other research are compared with values predicted by this proposed approach, and a brief comparison is made with the New Zealand Concrete Design Code. The chapter culminates in a step-by-step explanation of an integral flexure/shear ductile design procedure. This new design methodology aims to provide a more accurate and reliable way to predict the behavior and failure modes of spirally reinforced columns under seismic conditions, addressing limitations of existing methods.

The proposed design equations are compared with results from other research and provisions of the New Zealand Concrete Design Code. This comparison serves to validate the proposed method and highlight its advantages and limitations. Specific details on the comparison methodology are not provided in this summary but are noted to be included in the full text. The study's findings are compared to existing design codes, including the New Zealand Concrete Design Code, to identify the strengths and weaknesses of both approaches. Areas of agreement and disagreement between the proposed approach and existing codes are noted, suggesting potential improvements to existing methods and informing future research directions. The comparisons provide critical context for assessing the accuracy and practicality of the new design procedure.