Concrete columns subjected to biaxial loading

• Main Author: Tow, Jun Yi.
• Other Authors: Tan Kang Hai
• Format: Final Year Project
• Language: English
• Published: 2012
• Subjects: DRNTU::Engineering::Civil engineering::Structures and design
• Online Access: http://hdl.handle.net/10356/48950
• Institution: Nanyang Technological University

Biaxially-loaded columns usually locate at the corners of a building. With the introduction of high strength concrete, design of columns now involves smaller cross-sectional area and longer length. This renders a column to be classified as slender. The relevancy of the current design codes to design biaxially-loaded slender columns becomes an important consideration. In the experimental programme presented in this report, three columns were tested to failure to obtain the failure load, deflections and strain measurements. The experimental results would be compared to the design codes used in Singapore. This includes the newly implemented Eurocode 2 (EC2) as well as the British Standard (BS 8110) which was withdrawn from use since March 2010. A numerical analysis was performed using a finite element program to support the experimental results. Experimental results indicated that the assumptions made in the codes are reasonable and valid, i.e., plane strain across an inclined neutral axis was observed and the concrete limiting strain was verified to be around 0.0035. The biaxial failure load for the column specimen Ca-2-40-40 was recorded at 1370 kN. BS 8110 classifies the test column to be a short column. On the contrary, EC2 classifies the test column to be slender. Theoretical failure loads were calculated using the design methods described in the respective codes with the exclusion of the respective safety factors. Analysis results showed that the tested column should be classified as slender. By considering it as a short column, BS 8110 over-estimated the column capacity by 43 percent. Whereas, EC2 using method of nominal curvature (MNC) under-estimated the column capacity by 14 percent. A number of key parameters of calculation specified in both codes were also investigated. Finite element analysis software SAFIR was used to analyze the tested column. The analysis results were closed to the experimental results. This indicated that the results produced by SAFIR were reliable with respect to the design of biaxially loaded slender columns at ambient condition. Based on the specific test conditions in this study, EC2 (MNC) was more conservative than BS 8110 and provided better estimation to the actual biaxial failure load. In addition, designs based on BS 8110 were deemed to be unsafe in this study. More experiments should be conducted to fully investigate the design of biaxial loaded slender column conforming to BS 8110 as well as EC2.