Applicability of low-yield-strength steel for ductility improvement of steel bridge piers

Abstract

An experimental investigation on the use of low-yield-strength (LYS) steel to improve the ductility capacity of box-shaped steel bridge piers is presented in this paper. Five specimens in which the lowest cross section was fabricated with LYS steel plates were tested under cyclic lateral loads. Sectional configurations and the thickness of LYS steel plates of the specimens were taken as the main test variables. An additional specimen that does not have LYS steel plates was also tested as a benchmark specimen. The test results revealed that the specimens with stiffened LYS steel segment having proper thickness have greater ductility and energy absorption capacity. The stiffened specimens with much thinner plates are vulnerable to sudden brittle failure whereas those with much thicker plates are undesirable due to the excessive strength. The unstiffened specimens with thick LYS steel plates have enough strength but the ductility is not sufficient. In some cases, the increase in the member strength due to the great cyclic strain-hardening characteristic of LYS steel could be partly balanced by a local buckling deformation.

Introduction

It was evident from past earthquakes such as the 1995 Hyogoken-Nanbu earthquake in Japan that most of the box-shaped steel bridge piers have undergone an excessive lateral deformation due to local buckling of component plates. These columns eventually failed as a result of loss of strength and ductility. Therefore, sufficient ultimate strength and ductility capacity should be provided for steel bridge piers in order to withstand severe earthquakes. Consequently, the Japanese seismic design code was revised in 2002 [1] considering the lessons learned from the 1995 Hyogoken-Nanbu earthquake.

Concrete-filling, horizontal diaphragms with shorter intervals, more longitudinal stiffeners, double skin tubes, seismic shear walls, inner cruciform plates, corner plates, and most recently tapered plates are among the several techniques that had been introduced in the past to improve the ultimate strength and ductility of steel columns [2], [3], [4], [5], [6], [7], [8], [9]. Concrete-filled steel tubes are popular all over the world for their excellent strength and ductility characteristics. Researches on concrete-filled steel tubes are widely available (e.g., [2], [3]). Particularly, partially concrete-filled steel tubes are widely used in highway bridge piers in Japan [2]. This technique is simple and effective but the main disadvantage is that the concrete infilling increases the axial load on the foundation. The repair work after an earthquake would also be difficult due to the filled-in concrete. Setting diaphragms in short distance intervals and the use of heavily stiffened sections are also very useful in preventing early occurrence of local buckling [4]. However, the normal strength steel material is incapable of accepting large accumulated plastic strain resulting from repeated loads. On the use of double skin tubes, Zhao and Grzebieta [5] have carried out a series of bending and compression tests on concrete filled double skin tubes (CFDSTs) that have outer tubes with width-to-thickness ratio ($b/t$ ratio) ranging from 16.7 to 25 and inner tube with the $b/t$ ratio of 20. They have found an increase in energy absorption capacity and ductility in CFDSTs compared to empty single skin tubes. Takaku et al. [6] experimentally investigated the strength and ductility performance of steel piers with internal multi-cell panels made from low-yield-strength (LYS) steel. The main finding of their test was that the columns with LYS steel panels are capable of absorbing more energy than the mild steel panels. In a method introduced by Yamao et al. [7], inner cruciform plates have been used to improve the seismic resisting capacity of steel piers. These experimental and theoretical investigations have revealed that a better performance than that of ordinary piers can be achieved in terms of ductility and energy absorption capacity when inner cruciform plates are used in box section piers. As per the large-scale cyclic loading tests conducted by Terayama et al. [8], the introduction of corner plates increases the ductility of stiffened steel columns because it prevents the brittle fracture of the corner between flange and web plates. In a recent work by Takaku et al. [9], the use of tapered plates in steel bridge piers has been introduced as an effective and economical means to improve ductility. Tapered plates allow a better fitting of the bending moment distribution resulting from the applied loading, and a larger spread of the plate yielding. Design recommendations have also been proposed for bridge piers with tapered plates based on analytical and experimental studies.

Another possible way to improve the deformation capacity of box section steel bridge piers is to use LYS steel plates at the lowest panel. The clear difference between the normal steel and the LYS steel is that the latter has a considerably low yield strength, usually around 100 MPa. The LYS steel can be stretched much more compared to the normal strength steel, and has tensile strength around 2 to 3 times of its yield strength. It also exhibits excellent strain hardening characteristics when subjected to cyclic loads. Researches on the application of LYS steel in building structures in the forms of hysteretic dampers and shear walls can be widely found (e.g., [10], [11], [12], [13], [14], [15]). The main advantage of the use of LYS steel in hysteretic dampers is their effectiveness under smaller vibrations. Dampers with normal steel do not function effectively under small vibrations because the steel is in the elastic range [10], [11]. Nakashima [12] tested several shear panels made of LYS steel to examine the hysteretic behaviour. The relationship between the peak shear force and the number of cycles clearly demonstrated strain hardening during cycles having the same drift angle as well as increasing drift angles. Also, the comparisons between shear force versus drift angle relationships of shear panels made of low-yield steel and SS400 (nominal yield stress, ${\sigma }_y=235\mathrm{MPa}$) showed more significant strain hardening in low-yield steel panels than in conventional mild steel. A simple model to simulate the hysteretic behaviour of such shear panels has also been proposed by Nakashima et al. [13]. The application of LYS steel shear panels in moment resisting steel frames together with seismic response characteristics and a design procedure are available in Matteis et al. [14], [15].

In this study, the fundamental behaviour of a steel column fabricated using stiffened or unstiffened LYS steel plates at the lowest cross section was experimentally investigated. The main aim of this study was to check the ductility improvement of steel bridge piers. Six specimens were tested under repeated horizontal load and constant axial load. One of the specimens served to be the benchmark specimen that consists of only normal strength steel plates of grade SM490 (nominal yield stress, ${\sigma }_y=314\mathrm{MPa}$). The other five specimens consisted of normal steel of grade SM490 and LYS steel plates with yield strength of around 100 MPa. Test results were analyzed to find out the ductility, ultimate strength and energy absorption characteristics, emphasizing the effects of plate thickness and sectional configuration.