by clicking the arrows at the side of the page, or by using the toolbar.
by clicking anywhere on the page.
by dragging the page around when zoomed in.
by clicking anywhere on the page when zoomed in.
web sites or send emails by clicking on hyperlinks.
Email this page to a friend
Search this issue
Index - jump to page or section
Archive - view past issues
Concrete In Australia : June 2008
CONSTRUCTION SAFETY What went wrong in the West Gate Bridge tragedy by Ben Cosson This article highlights the importance of clear and timely communications that must exist between asset owners, designers, contractors and construction workers to ensure successful and safe project outcomes are met. 35 men were killed outright or as a result of injuries sustained from the collapse of a 112m span of the West Gate Bridge on the 15th October, 1970. When span 10-11 failed approximately 2000t of steel and concrete came crashing down onto the muddy banks of the Yarra, taking workers, their machinery, tools and sheds with them. A Royal Commission was established to investigate the failure of the bridge and is the basis for the discussion that follows (1). The commission found that the sources of the failure predominantly developed from two main causes, namely: • failure of the designers to give a proper and careful regard to the process of structural design along with failing to adequately check the safety of the erection proposals put forward by the original contractors • contractors adopting a previously untried method of erection that required greater care than what was exercised. The design and subsequent failings Prestressed concrete approach viaducts on either bank, supported on slender concrete columns of 67m spans, led up to the bridge itself. The design was a cable-stayed box girder with the following characteristic elements: • The cable stays and supporting towers were in a single vertical plane on the longitudinal centreline of the bridge. • A trapezoidal box section with two internal vertical webs was chosen for the girder. • Low alloy high yield steel was used for much of the box girder. • Careful control of steel quality and construction techniques was employed to avoid brittle failures in the steel. • Panels of stiffened plating were produced in the fabrication shop. • Panels were sub-assembled into units, called boxes. Each box was 16.0m long by 12.7m wide. Two boxes comprised the full cross-section of the box girder, 25.5m wide at the upper flange. • High strength friction grip bolts were used for all connections other than those made in the fabrication shop. • Reinforced concrete, used for the road deck, was made to work in a composite manner with the upper fl ange plate by the use of stud type shear connectors welded to the upper flange plate. • Cantilever brackets were used from the main box girder to extend the overall width of the road, in this case to 37.2m overall. This was a complex design. Apart from using a box girder, this component was continuous over many supports, having Wreckage of the collapse from ground level VPRS 24/P3 Inquest Deposition Files, unit 120. a section divided by internal webs and multiple cells and also including sloping outer webs to give a trapezoidal section. Although a structural analysis had been undertaken, at that time a relatively low level of research and knowledge into the behaviour of large box girder bridges had been carried out. The actual calculations and analysis carried out at the tender stage was completed in three months and by the designer’s admission was “very quick going”. Inadequate consideration at the time of tender design contributed to the substantial amount of the subsequent troubles that took place throughout the project. Gross simplifi cations were made during tender design, including one case where the fl exural rigidity of span 10-11 had been assumed constant throughout its length, when in actual fact there existed large differences of rigidity from box to box. After the failure the designer acknowledged the unsatisfactory program used during the tender design. With this knowledge the designer should have immediately acted to warn the successful contractor responsible for the analysis of the stresses during the erection stage and proceeded without delay in making a proper design check. The commission found that it appeared neither of these steps were followed. It was common practice at the time for British major bridge designers to leave the analysis of the stresses during the erection stages to the contractor. In order to ensure erection calculations were in accordance with the engineers’ design criteria it was essential for the contractors to receive a set of the designer’s bridge design calculations. Despite a provision in the contract stating this requirement, along with repeated requests by the contractor to receive the design calculations, the designer resolutely refused to hand over any calculations. This early lack Concrete in Australia Vol 34 No 2 23