Nozed, I think I agree except that I was basing some of my understanding of Bazant from his crush-up crush-down paper. Its my understanding that he said that an intact block was required the whole way down, -
I don't think there is any "Require[ment]" the block stay together the entire way down. Again, he created a series of simple assumptions to determine if - under circumstances he considered the best possible chance to arrest a collapse - the building still collapsed.
For the building to have the best chance of survival, the upper floors would contact the floors underneath them squarely. Truss hits truss evenly. All of that is simply an assumption of the model, not a requirement in order to ensure collapse. In fact, it is largely the opposite.
- then once that block completed crush down...crush up would occur at the end destroying the block. My problem is that when I look at the video of the tower I dont see a block crushing all the way down.
Second thing, am I understanding Bazant correctly that he thinks having only part of the upper block crushing down would be worse than having the whole thing crushing...is that what he was getting at with equal distribution?
There are conflated issues here. The three things to keep in mind are:
(1) Mass
(2) Time
(3) Pieces vs. "intact"
There is also velocity/acceleration, but lets stick with the simple stuff. The first two are, from my point of view, the important issues. The load striking the lower floors is a dynamic load. As people have pointed out, getting hit with 1,000 pounds of anything all at once is different than being hit with 1000, 1 pound packets if the packets are dropped one at a time and the impact is therefore spread out over a longer period of time. Just as a car's roof will stop rain, but -- as the embedded video shows -- will be absolutely crushed if a large amount of water is dumped all at once.
In this case, almost all of the material in the upper floors dropped within the space of the building floor. Almost all of the entire mass therefore hit the floor beneath it. And it is that one floor that matters, because when the weight hits, that one floor has to take all of the strain of the mass striking it as a dynamic load. So the Mass (No. 1 on the list) is not materially different than in the Bazant model.
With regard to time, I do not believe there is a material difference there, as well. The entire mass came down. Whether you consider it largely intact, or not, this is not a case where the material was dropped, small bit by small bit, over the course of minutes -- it all hit within the space of a second. If anyone feels this is a material difference, then they will have to use math to show it -- any difference is not something that could be arrived at by mere looking and guessing. Moreover, the amount of mass of the upper floors provided more than enough force to initiate the collapse (by a factor of ten or more), so any math will have to show the minimal delay reduced the force (I may be misusing the technical term, but I hope the meaning is clear) being applied to a mere fraction -- even cutting it in half or to a mere quarter of the model's assumptions still has the buildings collapse.
And number 3, the factor you are looking at, will not matter if the first two do not vary significantly. The "not intact" factor only matters if the amount of time can be shifted significantly enough to allow the lower floors to absorb the initial impact and stand long enough to then take the next one as a more-or-less separate event. Otherwise, being crushed under a ton of rocks is no better than being buried under a one-ton slab.
The final question, relating to equal distribution – is not about having “part” of the upper floors hitting, but is about the mass striking the floors below unevenly – not the clean, truss-onto-truss, beam-onto-beam even distribution he modeled.
In other words, instead of spreading out the impact evenly, more weight/mass/force would be applied in certain spots – making the collapse more likely because that one spot would be strained even more than in the Bazant model, where it is all spread out evenly.
(Whew. Had just finished typing that when the thread closed. Hated to lose it, not for the quality, but because of the size.)