The paper summarises the results of three load tests on composite beams of 10m span, comprising one control composite beam (i.e. without opening) and five rectangular openings of various sizes and locations. The tests were carried out under simulated uniform load and were taken to failure of the beam at each opening. The results of the tests have been compared with existing guidance in the SCI/CIRIA publication Design for openings in the webs of composite beams. The overall model factor, representing the margin between the tests and the design method (eliminating partial safety factors and using measured properties) was in the range of 1.04 to 1.47. Reasons for this apparent conservatism are given. The control composite beam was also tested to failure without the presence of openings, and the results were compared with the method of BS 5950: Part 3. This gave a model factor of 1.04. Local instability of the web around the opening was found to be less susceptible than anticipated, and the section classification in the SCI/CIRIA design guide is redefined. Moreover, the reduction of plastic moment capacity due to the presence of axial force in the web flange sections was also found to be less severe, and a quadratic axial force moment interaction equation is adopted. The overall model factor based on the revised design method was found to be in the range of 1.04 to 1.37 with the higher safety margin corresponding to the more critical situation in which the opening is located within a zone of high shear. Consequently, a more economical and rational design method is achieved. Measured deflections compared favourably with the predicted values.
|Number of pages||7|
|Journal||Structural engineer London|
|Publication status||Published - 7 Jan 1992|
ASJC Scopus subject areas