Axial cyclic behaviour of RC prisms representing wall boundary zones.
| dc.contributor.author | Gokhale, Rohit | |
| dc.date.accessioned | 2025-02-23T20:43:29Z | |
| dc.date.available | 2025-02-23T20:43:29Z | |
| dc.date.issued | 2025 | |
| dc.description.abstract | Structural reinforced concrete (RC) walls effectively provide lateral load resistance in regions of medium to high seismicity owing to their relatively high in-plane stiffness. A flexure-dominated RC wall (shear-span ratio ≥ 3.0) relies on the response of its confined end regions against in-plane lateral cyclic actions. These confined end regions (also known as boundary zones) are subjected to tensile and compressive strain reversals during seismic events. A more desirable deformation capacity of a structural wall can be achieved by suppressing the premature compression failure modes in its boundary regions. Adequacy of prevailing design practice is often studied through scaled wall panel testing under in-plane lateral cyclic loading. While such wall panel experiments provide comprehensive insights into the wall behaviour, conducting parametric studies are deemed cumbersome due to the complex test setup and its resource intensive nature. Alternatively, a simplified approach of treating these end boundary zones as isolated columns (or prisms) and testing them under axial loading has been introduced in the past. Subsequently, several experimental studies have been conducted using this approach to study the parameters influencing not only the global instability but also localized compression failure modes. Therefore, the research presented in this thesis is centred around examining the various aspects associated with the idealized RC prism approach. A typical quasi-static uniaxial cyclic loading protocol often employed during prism test comprises of multiple cycles (mostly 2 to 3) at progressively increasing axial displacement amplitudes, even though an earthquake loading pattern is characterised by large number of small magnitude cycles and fewer large magnitude cycles. Inclusion of a more realistic loading history representative of earthquake demand seems warranted in the test program, however a procedure to develop such loading history for idealized wall boundary zones has not been explored adequately in the literature. This study presents a numerical procedure leading to the development of uniaxial cyclic earthquake loading protocol that comprises of realistic strain cycles representative of a chosen earthquake-type, both in terms of relationship between the tensile and compressive peaks and the corresponding cycle count at each strain range. With an increasing emphasis on performance-based design, the proposed loading protocol is structured around inelastic strain demands generated at the performance-based drift limit for structural walls. Analytical studies are conducted on a prototype wall model and the resulting axial cyclic loading protocols having near-fault and far-fault characteristics are presented. Moreover, parametric study results involving effect of some critical parameters on the cycle content are also discussed and then expressions are proposed to aid development of the loading protocol. Applicability of these proposed expressions are then scrutinized through analytical studies conducted on a different reference wall model using both, same as well as different, suites of ground motions. A review of previous experimental studies on RC prisms involving cyclic tension-compression loading suggests that most of the loading protocols differed in terms of their compression to tension strain ratios. The strain ratios were either held constant or varied with each increasing level of strain range. However, a specific study involving comparative evaluation of the inelastic demands imposed by conventional and earthquake loading protocols is missing in the literature. In this study, uniaxial cyclic tests were performed on doubly reinforced prisms idealised as the boundary elements of rectangular flexure-dominated walls to evaluate the effect of various loading histories on different compression failure modes. The uniaxial cyclic loading histories applied to these specimens included a conventional loading protocol (obtained directly from lateral cyclic loading test of a prototype wall) as well as two earthquake loading histories comprising cycle content representative of the cyclic demand expected from near-fault and far-fault earthquakes. Experiment results highlighted the low-cycle fatigue damage caused by the conventional loading protocol in comparison to the earthquake loading histories. In some of these tests, the level of compressive strains in the adopted loading protocol was reduced to study their impact on the global instability failure mode. Review of previous experimental research on idealized prisms also highlighted inconsistency in the prism height adoption. Typically, the prism height has been either considered based on the floor-to-floor unsupported height or theoretical plastic hinge length of the representative wall. Selecting an appropriate column height representative of the compression failure mode (global buckling or local instability) under consideration seems crucial to avoid experiencing failure modes different from the research objective, as observed during some of the past experiments. This study experimentally scrutinizes the role of prism slenderness (height-to-thickness) in altering the compression failure mode sequence. Unique relationships between the effective prism slenderness ratio and different compression failure modes are established using the results of this study as well as previous experimental investigations conducted by other researchers. While previous experimental studies have mainly focussed on investigating the influence of various parameters ranging from reinforcement detailing to loading history on the prism response, studies comparing the response of idealized prisms with the corresponding wall boundary zones have been rarely conducted. The underlying assumption with the prism testing concept is that the results obtained from the prism experiments are conservative and considered representative of the wall response. But a past experimental study on wall specimen and corresponding idealized boundary prism has shown that meaningful results may not be obtained without the consideration of strain gradient along the height, which was found to be a key influencing factor. This study proposes an approach to overcome this shortcoming and presented a procedure that can facilitate reliable prediction of the wall ultimate drift capacity deduced from the prism results using a material-strain limit approach by assuming constant strains over the theoretical plastic hinge length (i.e., prism height in this case). Finally, all findings from the experimental and analytical studies conducted as part of this research study are consolidated and recommendations for future research are also summarised. | |
| dc.identifier.uri | https://hdl.handle.net/10092/108114 | |
| dc.identifier.uri | https://doi.org/10.26021/15654 | |
| dc.language | English | |
| dc.language.iso | en | |
| dc.rights | All Right Reserved | |
| dc.rights.uri | https://canterbury.libguides.com/rights/theses | |
| dc.title | Axial cyclic behaviour of RC prisms representing wall boundary zones. | |
| dc.type | Theses / Dissertations | |
| thesis.degree.discipline | Earthquake Engineering | |
| thesis.degree.grantor | University of Canterbury | |
| thesis.degree.level | Doctoral | |
| thesis.degree.name | Doctor of Philosophy | |
| uc.bibnumber | in1406522 | |
| uc.college | Faculty of Engineering |