I wanted to talk about structural collapse for a moment, and this seemed a good place to do it.
Several years ago, one of the barracks on Fort Bragg, NC, near where I lived, was declared unsafe, and evacuated. Safety inspectors came because of a complaint of cracks in a wall, and a mirror shattering for no known reason, and concrete debris in the basement.
As it turns out, one of the concrete support beams in the basement had developed fractures. The inspectors said it can happen with concrete if, for some reason, the concrete sets improperly. If there had been any artillery firing near the main base area on one of the days during construction, vibrations could have caused minute cracks to begin forming, and load stress would have increased those cracks.
Under normal circumstances, a single cracking support column wouldn't have been a problem. But the barracks in question had been build a long time ago, when it was one soldier to a room, with a bed and wall locker, no air conditioning, etc. Since then, they estimate that the live load had more than quadrupled - four man rooms, much furniture, tons of personal belongings, and air conditioning units.
The building stood almost a week after evacuation, as the engineer slated to tear it down were out of the state. Then, one evening, it just fell down. Straight down, in an even collapse.
A lot of the guys asked the same exact questions the CTists do here. The answer was simply that, once a column fails to be able to support its load, it increases stress on surrounding columns. In the case of a building where loads were already exceeded - like this one - many of those columns were damaged as well, though not visibly. They, too, were unable to handle the sudden stress, and failed, transferring even more load to neighboring columns.
This transferral takes seconds.
The result is total structural failure.
The same thing happened in a poorly designed bank years ago - I saw it on a documentary on Discover Channel. The building hadn't properly accounted for dead load, and eventually underwent spontaneous structural failure due to a single column cracking.
In this regard, steel is no different. When steel is weakened by fire to half its strength - a temperature of merely 600 degrees - it loses its ability to bear its assigned load, and that load transfers to other supports. If enough supports fail, all of them will fail; meaning that the building will go from standing to not standing, just like that. No slow sag, no gentle lean, no domino-effect collapses.
Just wanted to talk... don't mind me.