In the 2011 Tohoku earthquake and the 2016 Kumamoto earthquake, some buildings were demolished because of damages to precast concrete piles near the pile head. In Japan, superstructures are designed using working stress design for intermediate earthquakes and a lateral load carrying capacity check for severe earthquakes. However, substructures are not required to be designed for severe earthquakes and the structural performance of piles in terms of moment capacity, shear capacity, and ductility after yielding of members is not clear. For better seismic performance, steel-encased precast concrete piles (hereafter, SC piles) are often used at the pile head in low and mid-rise buildings in Japan. Extensive experimental works have been conducted recently to study the flexural performance of SC piles i.e. moment capacity and curvature. However, the study of numerical model which includes the effect of steel buckling is not investigated when SC piles have high axial force. In this paper, the numerical analysis by using multi-spring model (hereafter, MS model) was conducted to reproduce the experimental results of five SC piles with axial force ratio, 0~0.35. It was also clarified the drift angle at ultimate condition by using the simplified formula based on concrete ultimate strain, neutral axis depth and plastic hinge length.
Lateral load – drift angle relations were well simulated with the numerical model by MS model until ultimate capacity. In analysis, ultimate capacity tended to reach when concrete crushing happened. The modelling of steel buckling was important to reproduce the hysteresis loop with high accuracy. The ratio of experimental and numerical maximum shear force was 1.02~1.08 except for the positive direction of specimen SC6. In the drift angle at 95% times maximum capacity defined as ultimate capacity, analytical results had good agreement with experimental results. The equation on AIJ guidelines was also shown to compare with ultimate capacity in the tests. In all specimens except for the specimen SC7, the calculation underestimated experimental values and this equation was found that it is not able to apply for SC7, which is small steel thickness.
In this study, the revised simplified equation was also shown to reproduce the ultimate capacity. The revised equation was simplified and consisted of concrete ultimate strain, neutral axis depth and plastic hinge length. Concrete ultimate strain and plastic hinge length were the same value with MS model. Neutral axis depth was calculated by the regression equation considered the effect of axial force ratio. As the results, the computation of ultimate drift angle was able to reproduce experimental results. However, this computation tended to overestimated the test result of SC6, which is high axial force ratio, 0.35.
