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 : December 2008
same collapse strength which may or may not be sufficient to prevent disproportionate collapse. The columns in higher stories directly above the lost column will have near-zero axial forces although they may have moments associated with benefi cial Vierendeel action in the fl oors. The notion of lower fl oors “hanging off” higher structure, as clearly happened at the World Trade Center, demands some nontypical stronger elements at higher levels. In the case of WTC, these elements were the “hat trusses” designed to support communication masts. Belt-truss type buildings Tall offi ce buildings over, say, 30 stories typically have intermediate plant-rooms at about 30 story spacings. These plant rooms may occupy the full footprint of the building and be about double the height of typical stories. They do not need windows but they do need air intake ducts and exhausts. For “tube-in-tube” buildings, it has been a common practice to use such stories to provide concrete walls or steel trusses as “belt-trusses” linking all of the lateral-load elements of the building. This reduces wind load defl ections and is sometimes successful in inducing reversed curvature of the building at higher levels under wind. Such intermediate/top belt trusses would also provide nontypical stronger elements at higher levels and, in cost terms, they would help pay for themselves by stiffening the wind response. TheWorld Trade Center also had “sky lobbies” for the lift system. Perhaps sky-lobbies can also be used to accommodate belt trusses. Guidelines for detailing for lost columns Thinking about fl exural load-path analysis also provides some guidelines for lost support detailing which are pretty much the same as for earthquake “ductile detailing”: For concrete buildings one could provide: • beam bottom rebars through columns of area not less than, say, 33% of top rebars; fully developed and lapped with midspan bottom rebars • some minimal beam top rebars across the whole span • anchor slab bottom rebars into beams or lap them across load-bearing walls • closer minimum beam and column ties for confinement of concrete, for restraint of compression rebars after spalling of concrete cover and for shear strength • full tension splices for all column splices. For steel buildings one could provide: • column splices with suffi cient tensile strength to support one or several fl oors hanging above a lost column (this could be an issue with field-bolted splices) • beam splices and connections should have some moment strength. The actual sagging moment strength of simple (hinged) connections needs to be realistically assessed. Do the Ronan Point rules provide a basis for US code writers? There is a huge difference between the collapse at Ronan Point and that at the World Trade Center. “NIST does not believe that buildings should be designed for aircraft impact.” (NIST 2005, p 216). One has some doubt that tall buildings can or should be designed to withstand an attack comparable to that at World Trade Center particularly with the wider aircraft now coming into use. Arguably, it should be the responsibility of the aviation industry to keep terrorists/ hijackers out of cockpits and, perhaps to segregate pilots in cockpits in such a way as to prevent entry of terrorists in fl ight. This would, at worst, transform terrorists into hijackers; still able to threaten the plane, crew, and passengers but not tall buildings and their occupants. Perhaps the Oklahoma City attack is a more reasonable precedent for code writers. Given that, then the Ronan Point rules, particularly those dealing with notional load-path analysis are interesting but the same ground is now covered, in greater and more recent detail, by the earthquake codes and by GSA 2003. Fire Fire adds a huge new dimension to the issues canvassed above and distinguishes the events at Ronan Point and Oklahoma City (no fi re) from that at the World Trade Center. Note, however, that fi re can follow explosions and fire frequently follows earthquakes; indeed the writer’s memory is that there were more fatalities at San Francisco in 1906 from the ensuing fi res than, directly from the earthquake. This raises issues of the robustness of lifeline utilities such as water reticulation systems for fire fi ghting after extreme events. These issues are too large to pursue here. The subject of fire and the prospective direction of structural engineering practice in the light of the NIST 2005 document is addressed in a companion paper (Gurley 2008). Conclusions and suggestions Lost column analysis (notional load-path analysis) will provide a useful tool towards ensuring that buildings are robust against disproportionate or progressive collapse. The structural engineering profession is already familiar with the collapse mechanisms relating to extreme wind and earthquake events and the analytical techniques required to ensure that, when damage is inevitable, it is proportionate damage. The Oklahoma City attack is, for code writers, a more reasonable precedent than the attack on the World Trade Center. The profession needs now to develop a comparable understanding of the lost support/double span mechanisms associated with bombs, blast and fi re and to develop analytical techniques to ensure robust buildings and damage that is proportional to the cause. The writer believes that simple hand calculations can be developed for fl exural “lost column” Concrete in Australia Vol 34 No 4 47