LONG TERM differential thermal expansion was a major factor behind the collapse of terminal 2E at Charles de Gaulle airport in Paris last year, the head of the official investigation into the failure said last month.
Four people died when a 24m wide section of the terminal building collapsed at 6.57am on 23 May last year after the coldest night that month.
An interim report published in July last year revealed that composite steel/concrete wall sections had failed when struts connecting the inner concrete shell to external steel tension members punched through the 300mm thick concrete.
quiry president Jean Berthier confirmed that cracking around the strut/concrete connections caused by differential expansion and contraction of the steel and concrete was the primary cause of failure.
The interior of the structure was air conditioned, but the steel truss sat below exterior glazing in an uncontrolled environment.
The basic design is made up of two curving prefabricated composite wall units linked by insitu concrete stitches to a curved precast roof panel. This semi-tubular structure sits on insitu edge beams and columns (see diagram. ) Concrete wall sections were cast on site and the external steel truss bolted on before erection.
The collapse occurred at the point where the building was at its weakest, Berthier said. Failure probably occurred in the lower sections of the curved walls surrounding three openings created for walkways to take passengers into the terminal.
ere the reinforcement details would be different from the main building, and stress concentrations could occur.
This part of the structure collapsed, pushing the northern section off the longitudinal beam that supported it. Berthier ruled out column head failure as a cause of the collapse.
He said photographs of the debris showed there was no structural damage to the column heads on the south side, which remained intact after the collapse.
Instead, temperature induced movement progressively cracked the wall concrete until a strip fell off the inside immediately beneath the location of the anchor blocks for the external struts.
This occurred an hour and a half before the collapse.
With a reduced thickness of concrete to resist the punching shear forces, failure was only a matter of time.
Berthier also criticised the structural designers, which were led by Aeroport de Paris chief architect Paul Andreu, for attempting to produce an unconventional building without employing an independent checker.
'It was a small project but it was complicated, ' he said.
'It did not lend itself to code based calculations but only to computer modelling. On this point a second model was not produced to check the forces in the structure.' But sources in Paris questioned the effectiveness of an independent check in this case.
Punching shear failures are almost unknown in France.