1. Introduction

A notable example of development of structural engineering following failures was that ensuing from the collapse during erection of two steel box girder bridges in 1970.

Although Victorian engineers had successfully built large tubular bridges of riveted wrought iron plating, the revival of the form of construction took place in the 1960's. This followed the development of high grade steel and welding and fabrication techniques, and appreciation of the economies in erection which could be achieved. The box form provides high torsional stiffness and strength and, in the cases of the Severn and Humber suspension bridges, facilitated the design of revolutionary bridge decks having aerodynamically efficient profiles and properties.

2. The Problems

In June 1970 a span of the Milford Haven bridge over the River Cleddau in the UK fell. The superstructure of the bridge consists of a deep continuous trapezoidal welded steel girder. There is no substantial stiffening of the section at other than the six pier supports and abutments. The collapse occurred during launching a pre-assembled deck section towards the second pier from one end, as illustrated below.

In December of the same year a span of the Westgate Bridge over the River Yarra in Australia collapsed. The bridge superstructure also consisted of a trapezoidal box. The span under construction was built in two halves, which had been jacked into position. The collapse occurred while the levels of the two halves were being adjusted.

A Committee of Inquiry into the Design and Erection of Steel Box Girder Bridges (1) concluded that the cause of the collapse of Milford Haven Bridge was the inadequacy of the design of a pier support diaphragm. It was of the opinion that BS153 (the only British steel bridge design and construction Code of Practice current in the 1960's) was inadequate for such application and that there was need to consider wider implications for the design of such bridges.

A Royal Commission was set up to investigate the failure of the Westgate Bridge (2). The causes of this were much more complex and were primarily those of adoption of a previously untried and inadequately checked method of erection and failures of site organization and communications between principal parties. These led to release of bolts in the top flange to permit alignment, to an extent that the flanges buckled.

3. The Solutions

The Merrison Committee prepared Interim Design and Workmanship Rules (IDWR). These contained comprehensive rules for the stress analysis for box girders and for the design of steel plated components in complex stress fields, with and without stiffening, and for their connections. They treated the effects of welding residual stresses and geometric imperfections linked to fabrication tolerances.

The Report of the Committee concluded with 27 recommendations, most of which have been implemented. These included wide-ranging proposals for contractual procedures including clarification of the allocation of responsibility between the Engineer and the Contractor, the independent checking of designs and erection methods, vetting of the Engineer's resources and of his and the Contractor's site personnel and appropriate handling of competitive tendering. It was recommended that the Design Rules should be used only by suitably experienced designers

4. How Practice Was Changed

The Merrison Rules laid the foundations for the production of a new British Standard, BS5400, Parts 3, 6 and 10 of which, with the additional benefit of further research, now contain comprehensive rules for steel box girders. The implementation of the recommendations led to wide-ranging changes in contractual procedures and in checking designs and construction procedures.

5. Key References

1. Inquiry into the Basis of Design and Method of Erection of Steel-Box Girder Bridges. Report of the Committee. HMSO.1973.

2. Report of the Royal Commission into the Failure of West Gate Bridge. Government of the State of Victoria. 1971.

15 May 2001