A series of experimental studies is conducted to examine the behavior of SN490 steel column subjected to axial load in the fire condition. This experimental work is aimed at examining the effect of the width-to-thickness ratio of flanges, the slenderness ratio of steel columns and residual stresses on the ultimate strength of an SN490 steel column at a specified temperature. From these studies, it is found that the column strength decreases with the increase of width-to-thickness ratio and slenderness ratio. Column behavior is sensitive to width-to-thickness ratios and the slenderness ratio at temperatures below 550 ∘C. However, the influence of width-to-thickness ratios and the slenderness ratios is not severe when the temperature is higher than 550 ∘C. When the temperature is above 550 ∘C, the column strength deteriorates rapidly. It is also found that the failure mode of steel columns changes from inelastic global buckling at room temperature to inelastic local buckling at elevated temperature, due to the release of residual stress in a fire. Based on the results of this study, local buckling criteria and column strength at specified temperatures are suggested.
The ultimate strength of steel columns at room temperature has been studied thoroughly in the past [1]. The aspects that affect the strength of the column, such as residual stress, initial imperfections, the width-to-thickness ratios of stiffened and un-stiffened elements, and the slenderness ratio of steel columns, have been examined extensively. As a result, the SSRC column curves have been adopted as the basis for steel column design in a number of specifications 2., 3. and 4.. However, the strength design criteria of steel columns under fire load have not been examined thoroughly.
To determine the strength of steel columns at elevated temperature, American Institute of Steel Construction (AISC) [2], European Committee for Standardization (EC3) [3] assume that column strength degrades with the combined effects of the degradation of elastic modulus and yield strength. The identical equations for calculating the buckling strength at ambient temperature are adopted to determine the strength of the column at elevated temperatures, except the degraded elastic modulus and yield strength at the corresponding temperature are used. It is assumed that the influence of column slenderness, width-to-thickness ratio and residual stress on column strength in the fire condition is the same as that at room temperature. On the other hand, Architectural Institute of Japan (AIJ) [4] assumes that at elevated temperatures column strength varies with column slenderness and the decrease of yield strength, disregarding the deterioration of the elastic modulus.
A discrepancy is also found between the design codes and those observed in the experimental works. Fig. 1 compares the column strength of A36 steel at elevated temperature between the experimental results and those determined from the codes, in which PAIJPAIJ, PAISCPAISC, and PEC3PEC3 represent the column strength determined from the codes. Pexp.Pexp. is the column strength obtained from the experimental test [5]. The column strength has been underestimated at 500 ∘C if the slenderness ratio of column is less than 52, while, column strength was overestimated more than 50% if the slenderness of the column is larger. Besides the deteriorated yielding strength and elastic modulus of steel, the strength of steel column is also affected by the factors such as the slenderness ratio (L/rL/r), width-to-thickness ratio (b/tb/t), residual stresses, and end restraints. It is not appropriate to apply the same reductions of strength for columns at elevated temperature and at room temperature. To provide an economic fire design for the steel columns without sacrificing safety, there is an urgent need to establish the basic information of the structural behavior of steel column at elevated temperatures.The fire behavior of steel columns has been investigated by various researchers experimentally and theoretically 6., 7., 8., 9. and 10.. The parameters such as the axially or rotational restraint, load ratios, slenderness ratios have been examined in these studies. However, previous researches were aimed at seeking the critical temperature of steel columns with or without restraint. The knowledge related to the column strength at elevated temperature is limited 5., 11. and 12.. In this reported study, the experimental works was carried out on 42 SN490 steel columns by considering the width-to-thickness ratio, slenderness ratio, and temperature, to evaluate the behavior of steel columns during a fire event. The effects of residual stress and boundary condition on the fire performance of steel columns were also examined.
A total of 42 SN490 steel column specimens were centrally loaded in this study to examine the effects of width–thickness ratio, slenderness ratio, residual stress and effective column length on the structural performance of steel columns in a fire event. From these experimental studies it is found that the influence of the width–thickness ratio, and slenderness ratio is significant for column behavior at temperatures below 550 ∘C. However, temperature is the dominate factor to column behavior at temperatures above 550 ∘C. Stub columns with a compact section are capable of preventing local buckling and reached a ductility of 3. It is suggested to use compact section steel columns if fire-threat is of concern. Residual stress has been greatly released in the fire event and its influence on column strength can be neglected. The failure mode of steel columns may be changed due to the release of residual stress in the fire condition. For columns with a slenderness ratio less than 80, its strength at 500 ∘C is about 70% of its nominal strength at room temperature. For temperatures of 550 ∘C and 600 ∘C, the column strength is further reduced to 60% and 40% of the strength at room temperature.
