CROSS Safety Report
Partial roof collapse at shopping centre
This report is over 2 years old
There was a recent partial roof collapse at a shopping centre which was, says the reporter, built in the early 1970s in a form which may have been used extensively throughout the UK during this period and possibly beyond.
Key Learning Outcomes
For civil and structural design engineers:
Connections and fixings can often be the weak link in structures and attention to detail is required
An attribute of ‘safety’ is to assure that the design is not disproportionately vulnerable to minor error
Manufacturers' instructions provide helpful guidance on fixings. The Construction Fixings Association (CFA) website and CIRIA publication C778 Management of safety-critical fixings are also useful references.
For building owners/managers:
The specification of any inspection and maintenance requirements for façade fixings should be recorded in the operation and maintenance manual
The fixings may require a combination of regular visual inspections with full inspections at appropriate intervals
The fixings should be inspected by a person who is competent to inspect them
Find out more about the Full Report
The Full Report below has been submitted to CROSS and describes the reporter’s experience. The text has been edited for clarity and to ensure anonymity and confidentiality by removing any identifiable details. If you would like to know more about our secure reporting process or submit a report yourself, please visit the reporting to CROSS-UK page.
There was a recent partial roof collapse at a Shopping Centre which was, says the reporter, built in the early 1970s in a form which may have been used extensively throughout the UK during this period and possibly beyond. Luckily, due to the timing in the early hours of the morning, this roof collapse did not claim any loss of life but has raised considerable concerns.
The failure mechanism was ascertained from site visits, but the underlying cause is not known, and the report was submitted in the hope that it will be useful in the understanding of the mode of failure and in the identification of similar connections which may occur in other buildings. The building is of reinforced concrete construction, with columns supporting a grid of beams at roof level, which support the flat roof construction over shop units and a mall. The flat roof had a mineral felt weathering, on insulation board, on profiled metal decking. The metal decking spanned between steel lattice trusses, which themselves spanned between the reinforced concrete beams at roof level.
The lattice girders were connected to the high level reinforced concrete (RC) beams via a top flange bearing plate which was bolted to an angle cleat (bolts ‘B’) which was in turn bolted to the RC beams by bolts ‘A’ . The arrangement is shown in the figures below and the diagrammatic section. It was clear that some of the bolts used to secure the angle cleats to the RC beams (bolts ‘A’) had failed in tension. They appeared to have been pulled out of the concrete. A closer inspection of some bolts which were found in the debris showed that the bolts which had failed (‘Drop-In Wedge Anchors’ — manufacturer unknown) had done so within the shield of the anchor, within the embedded depth of the bolts. The threaded steel shields were later found to be no more than 1.6mm in thickness.
The reporter did not see any evidence of corrosion but was surprised at how thin the walls of the shells were. It appeared that it is most likely that the failure occurred as a chain reaction when a single bolt, or group of bolts at a connection failed (4 No bolts ‘A’ per connection). It was clear to see that the failure started in one corner, where the bolt shells had failed in tension. But further along, where the support cleats were still in position on the RC beam, the fixing bolts ‘B’ between the lattice girders and the cleats had failed in shear as the roof collapsed. The collapse followed a significant period of heavy rainfall.
The fall of the roof towards rainwater outlets (which would dictate the ‘deep end’ for any ponding water) should have accumulated more loading and possibly cause failure at that end if the outlets were blocked or flow restricted. However, this was not the case and failure occurred at the ‘shallow end’, the opposite end to the outlets. Remedial works to enhance the loading capacity of all of these connections were necessary.
Submit a report
Your report will make a difference. It will help to create positive change and improve safety.
Our secure and confidential safety reporting system gives professionals the opportunity to share their experiences to help others.
No feedback has yet been published for this page.
Expert Panel Comments
Expert Panels comment on the reports we receive. They use their experience to help you understand what can be learned from the reports. If you would like to know more, please visit the CROSS-UK Expert Panels page.
From the limited information available it is difficult to identify the cause of this failure which apparently originated at one fixing point and resulted in a progressive collapse. What was the force that caused them to fail? The bolts pulling out may be a symptom rather than the cause and the following comments are not an explanation of what happened here but reminders of the issues that can affect a fixing such as this.
The bolts attach the angle to the concrete so carry shear but also tension due to the eccentricity of the truss bearing on the angle. If the roof had ponding water and the beam was deflecting the point of bearing would move towards the edge of the angle increasing the tension significantly, failure could then have been by combined shear and tension. Normally it is assumed that bolts ‘A’ are in shear only but if shimmed out too much they move into shear + bending and that makes a big difference.
Bolts ‘A’ could also be levered out to some degree depending on whether the cleat is ‘stiff’ enough. Given there is only one line of bolts at B, there will be moment due to the eccentricity of load between B and the concrete face. These are secondary forces that may need to be considered.
Another possible cause could be corrosion of the 1.6mm steel shield although the reporter did not notice any evidence of this. Sequential temperature changes over the years or a sudden change in temperature might have had an effect. It would be helpful if anyone else has experienced similar failures and has an explanation as to why the fixings have lasted for 30 years without obvious problems.