Draft 2 April 4, 2007. Corrects and updates previous draft.How long did it take the World Trade Center's two tallest towers to fall? The answer is crucial to the government's "modified pancake theory" of collapse. [See footnote.]
But, government information on fall times is difficult to locate. The National Institute of Standards and Technology web search engine revealed nothing for "NCSTAR, seconds." NCSTAR is the codeword for the main (2005) and backup reports on the collapses of the twin towers. Unfortunately, I don't have access to my hard copy of those reports to see whether fall times are given (which might be dicey, given that there is no real index). I do know that the NIST's computer models halt at the point "global collapse" begins.
The 9/11 commission report, its appendixes and transcripts of public hearings have been taken offline by the Bush administration. Various attempts at searching Google Scholar for "World trade center, collapses, seconds" and similar combinations brought up nothing usable from government reports.
I searched the Federal Emergency Management Agency in this manner and had a hard time finding its 2003 World Trade Center report, though the item eventually turned up when I used a phrase from the report's title.
That report gives some seismic data for the collapses, which it apparently got from the Lamont-Doherty Near Earth Observatory, an adjunct of Columbia University.
Lamont-Doherty's web site publishes the seismic data, which were picked up from several seismic stations and processed at a seismological data center, before being released.
LaMont-Doherty's data show that the two waveforms for the collapses were similar but of different duration. The duration of the seismic signal for the collapse of World Trade Center 2, which fell first, was 8 seconds. The duration for WTC1 was 10 seconds.
I reviewed the ABC videotape of WTC2's collapse on Google Video, and, carefully watching the timer on the screen, came up with a collapse time of 14 to 15 seconds. However, the quality of the video is somewhat open to question as the images have been degraded because of internet data compression requirements. Also, one must trust that there has been no tampering with the timer.
One source says that the 9/11 panel used 10 seconds for WTC2 but that NIST later revised the estimate to 12 seconds. The FBI's estimates of the fall times are not generally available.
Some thoughts concerning fall times:
The seismic signals may have somehow been truncated either because parts of the fall weren't recorded or because of problems at the data processing center.
The WTC earthquakes were each a little over 0.2 on the Richter scale and produced mainly surface waves (as opposed to waves that travel at angles down through the earth). That is, the quakes weren't terribly powerful and so perhaps several seconds of collapse didn't shake the ground enough to be recorded. This idea however should be weighed against the fact that steel conducts shock waves very efficiently and rapidly, so that the concussion of the initial blocks of floors crashing down onto the lower parts of the buildings would have been telegraphed to the ground almost immediately.
Now if we suppose that the quake begins with that initial event and ends when the roof hits a few meters above ground zero, then the seismic reading should bracket the entire duration of the collapse. But, this idea does not square with the 8-second time for WTC2, as the roof can't possibly hit the ground in less than 9.22 seconds, the free-fall time for 417 meters.
So for WTC1 we have a reported estimate of 12 seconds as opposed to 10 on the seismic reading.
The duration given for WTC1 seems anomalous. However, there seems to have been no published government effort to issue a better estimate.
If we suppose that WTC2 fell in 10 seconds, we then have a collapse close to the free-fall rate which -- because of inertial resistance -- is actually faster than possible. So then, is it reasonable to suppose that the tower could fall that fast without the aid of explosives at key points? Perhaps such an outcome is plausible, but the NIST avoided any technical analysis of the rate-of-fall issue.
Internet searches disclosed no videos of WTC1's collapse, a somewhat disconcerting situation in that probably hundreds of video cameras were focused on that tower after the collapse of WTC2.
Now as for the 47-floor World Trade Center 7, which collapsed at 5:20 p.m. Sept. 11, the seismic signal is recorded at 18 seconds. The waveform also looks different than the waveforms for the twin towers.
Some who have studied internet videos conclude that WTC7 fell in almost free-fall time. This is odd, considering the seismic results. Such a disparity raises the possibility of tampering with either the seismic or the video evidence.
I have checked two videos on Google Video for WTC7 fall times and come up with about seven seconds. The free-fall time for the 174-meter building is 6 seconds.
There seems to be no reason not to come up with a computer model that uses the NIST's conditions at the time of "global collapse" as initial conditions for a means of assessing probable fall rates. If any such models have been carried out, the results are closely held.
One idea is to simply check for inertial resistance and ignore the resistance of the floor-girder connectors and air resistance. My own very simple physics calculations show that when the upper block -- assuming a symmetrical "flat" fall -- strikes the next floor, that block is slowed noticeably by the inertia of the struck floor (Newton's third law). The struck floor, assuming no other resistance, jolts down to the next floor and bounces back up, thence meeting the downard moving block.