Finite element analysis of bending capacity and bolted moment connections of cold formed steel section

Abstract

Cold formed steel sections are fabricated by pressing and folding steel plate under room temperature. Due to the cold forming effect, the cold formed steel sections usually have high yield strength from 250 to 450 N/mm2. They are widely used in building construction such as floor slabs and roof beams. The most common sections are C and Z sections with typical thicknesses between 1.2 mm and 3.0 mm; therefore, local plate buckling is always critical. Design codes such as BS 5950 and AISI predict the section capacities by summing up the capacities of individual flat plate elements of the sections. The effective width method is adopted to allow for local buckling in these elements under compression. With the advancement of cold formed steel fabrication techniques, many steel sections with more complicated profiles and higher yield strengths are available in the market. Current design codes, which are developed based on the test results of cold formed steel sections with simple section profiles, steels of low yield strength and high ductility, may not be adequate for those advanced sections. As a result, the manufacturers tend to execute full-scale tests which are time and money consuming, to determine the performance of their products. However, they cannot provide detailed structural behaviour such as stress distributions, load transfer within connection zones and implications of initial imperfections. Advanced numerical methods such as finite element analysis may readily give us understanding on the structural behaviour of cold formed steel sections In this thesis, a standardised finite element analysis procedure is proposed which is able to predict the moment capacities of cold formed steel sections. Guidelines for choosing suitable boundary conditions, type of elements, methods of load application as well as shape and size of initial parameters are also given. As shown in the calibration with test results, it is demonstrated that two dimensional shell elements may be used to model cold formed steel sections satisfactorily. In order to further simplify the finite element model, a sectional model is modelled which represents the middle portion of a long beam. The proposed length of the sectional model is 4 times the section depth of the section while the size of initial imperfection should be adopted as the measured value or assumed to be 0.2 times the section thickness if the measured values are not available. It had better to adopt true stress strain curve to complete the analysis. Bi-linear stress strain curves with peak values equal to the design strengths can be used as practising design calculation if a reasonable reduction factor, 0.67 according to the result of this thesis, is applied to the results. Otherwise, it is recommended that the true stress strain curves should be adopted in the analysis. In general, there has been a lack of knowledge on the understanding of member resistance deduction of connected members and also on the design on bolted connections. It should be noted that local concentrated stresses around the bolt holes will cause adverse effects on the capacities of connected cold formed steel sections. The uneven contributions of bolt forces are found then. Currently, the design codes only account for the load carrying capacities of bolts in the design procedure but ignore any adverse effect to the connected sections. It is also assumed that the bolt reaction centre is located at the bolt group centre and the bolt forces are equally distributed among each bolt. So, the reliability of the design methods has been questioned. Therefore, a number of finite element analyses and tests on bolted connections between cold formed steel sections are undertaken. According to the finite element analysis and the experimental tests, the results show that the moment capacities of cold formed steel section connections are only 68% to 93% of the moment capacities of the connected sections. In bolted moment connection with four bolts; the contribution of the rotational reaction forces about the first row of bolts is about 30% of the moment resistances. Moreover, the contribution of the rotational reaction forces about the second row of bolts decreases from 30% to 10% while the bolt pitch increases from 90mm to 300mm. At the same time, the contribution of the force couple between two rows of bolt is increased from 40% to 60%.