Tall building fires are rare events.
Until last month none had occurred in a classic concrete framed tower block, in recent years at least. So the detailed investigation into the effects of the blaze that destroyed Madrid's Torre Windsor will be eagerly awaited by fire engineering specialists all over the world.
'This was a big fire, and it will have implications not just for Spanish building codes but everywhere, ' says Jeremy Hodge, director of research at standards approval body BRE Certification.
'It certainly seems to have confirmed that 'strong floors' at regular intervals can minimise the risk of progressive collapse.' 'We've all had a gut feeling that concrete would behave well in these circumstances, but it's never been proven, ' adds Arup Fire's Susan Lamont.
'Because it's easy to get the specified fire resistance with a concrete frame there has been no incentive for the concrete industry to research whole building response. This building has stood up, but it will be fascinating to see how much spalling and damage the concrete has sustained.' Last October the upper 22 floors of the 56 storey Torres Parque Central office tower in Caracas, Venezuela, burnt for more than 17 hours but did not collapse. This was a steel framed structure, protected with a spray applied mineral wool and cement based material applied during construction in the late 1970s.
But Lamont says any fire protection would become irrelevant once the fire had burned long enough.
'Nothing can insulate the steel indefinitely from the high temperatures of a major fire. Fire protection just slows down the rate at which the steel heats up, ' she adds.
'The good performance of the Caracas tower was ultimately down to the continuity of the joints in the structural frame.' Lamont is due to speak at NCE's Fire Engineering Conference in London on 12 April and will challenge the finite element analysis (FEA) method used in the official report to model the collapse of the World Trade Center.
The National Institute for Science & Technology's (NIST) final 10,000 page report and recommendations has still not been published. Original publication date of a draft for comment was October last year; latest rumours suggest it could be delayed even as far as the summer.
An interim report detailing NIST's conclusions on the collapse mechanisms for both towers was published in December last year (NCE 21 December 2004).
'NIST's FEA model does not include a thermomechanical analysis. This models the forces in the structure generated by the thermal expansion of the steel, ' says Lamont.
'Instead, the NIST team seems to have made assumptions on what forces would be involved. We think this has produced an inaccurate collapse scenario.' A spokesman for NIST defended its FEA model, saying that NIST 'believes the physics of the FEA model are correct and the significant phenomena relevant to determining the probable collapse sequence have been adequately captured.' He adds: 'NIST is committed to putting accuracy, quality and completeness ahead of schedule, taking whatever time is required to do the job right.' Dave Parker