I've been rather disengaged from this thread due to illness (nothing major), so I've got a bit of catching up to do. Sorry for the compound post but there are several comments I wanted to reply to.
I for one have not pushed explosive or thermite demolition, but don't be silly. Explosives or thermite can fail any structural component just as well as a "NIST fire".
Explosives can cause point failures, but they can't cause softening and sagging. Arguably, neither can thermite, because it burns too hot; it's either melt the steel or nothing. I have a hard time envisaging any explosive or thermite related mechanism that could cause the perimeter column bowings, and I've yet to see anyone propose a viable one.
First, you are assuming the collapse floor only fell 12 feet in a 2 second time frame during the inward pull of the perimeter columns. That assumption isn't supported anywhere. The 12 foot drop of the antenna by that amount when it was 140 feet further up is not indicitive of the exact fall of the collapse floor and the core columns they were attached while the perimeter columns were being pulled inward.
By the way, the core was 137 x 87 feet and with the 209 foot square building with 14 inch square perimeter columns at each side the floor trusses were 35 and 60 feet long.
Putting aside the truly inconsequential nitpicking, your reference for the prior fall of the antenna is:
http://911research.wtc7.net/wtc/evidence/videos/wtc1_close_frames.html
The text on that page reads: "The radio tower descends about 10 feet relative to the facade." The time given is two frames in the sequence rather than two seconds; the actual time spacing is one second, so that was the only error I made. Since your own source gives a 10 foot drop, I could perhaps be forgiven for assuming a greater drop - which favours your hypothesis - in evaluating the conclusions you drew from it.
As for the fall of the antenna not being indicative of the fall of the floors where bowing took place, if that's the case then you require that the core must have been severed at two places at least - one below and one above the region where the perimeter columns bowed inwards - and that for some reason the upper and heavier part of the core fell more slowly than the lower and lighter part. Any ideas why that should have happened?
For your paper on the antenna drop, try not to come up with too high a value, because that'll leave you with a tough choice: do you try and explain away the fact that the perimeter columns didn't bow inwards above the initiation zone, or do you hypothesise that all the floor trusses from the initiation zone to the roof were severed at the beginning of the drop? I'll await your conclusions with mild interest.
I still stand by my rejection of your conclusion that the collapses required assistance. You have made two implicit assumptions in reaching it: firstly, that collapse cannot initiate unless the global safety factor falls below unity, and secondly, that the only relevant property of the columns is the resistance to purely compressive stress. The former amounts to an assumption that (a) the structure redistributes loads perfectly whatever the damage profile, and that (b) progressive failure cannot occur. The latter amounts to an assumption that there are no lateral forces on any columns, and is incidentally incompatible with the former. Since the observed collapse was initiated by a progressive failure begun by lateral forces on the perimeter columns, your line of reasoning therefore reduces to: If we eliminate the entire class of phenomena that includes the actual cause of collapse, then we cannot hypothesise a viable alternative cause of collapse. This is logically defensible but utterly irrelevant.
Thanks Ryan!
Here's the new, improved summary:
- Jets destroyed perimeter and core columns;
- Jets dislodged SFRM;
- Fires heated poorly-protected floor trusses;
- Heated floor trusses expanded and sagged;
- Heated core columns experience creep and thermal shortening;
- Action of hat trusses redistribute load away from the cores to the perimeter columns
- Sagging floor trusses exerted catenary forces on perimeter columns;
- Contracting floors after local burnout accentuate pull-in forces, particularly along long-span trusses;
- Catenary forces caused perimeter columns to bow inward;
- The eccentricities of inward bowing perimeter columns exceeded the moment capacity of the column splice bolts, causing the bolts to fail;
- The dominant failure mode for inward bowing columns was to separate and pivoted at the failed splices;
- However, there were cases of bowing and buckling in less than three-story lengths. (Bowing at initiation, for instance, was spread through many "waffle sections," and since these were staggered, the bowing was sometimes greatest near connections, and sometimes in the middle.)
- Once the upper block gets moving, the dynamic load exceed the carrying capacity of the structure, and total collapse ensues.
Sounds reasonable enough to me, but I'm not a structural engineer so there may be details I've missed.
Dave