
Edited on Thu Nov1209 04:40 AM by eomer
You say that the Part A columns have to decelerate the Part C mass to zero before buckling. But, again, decelerate it to zero over what time interval? If Part C is a rigid block as per Bazant's assumption then that deceleration has to be accomplished over a very short (virtually instantaneous) time interval. This is the calculation that Bazant makes. With no give (assumed) in Part C, the time it would take (theoretically) before the Part C columns fail or accomplish deceleration has got to be a small fraction of a second.
If, on the other hand, Part C is not a rigid block and the Part C structure fails before the Part A structure does then the time interval over which the deceleration must occur is drastically different. If, hypothetically, the leading onestory structure of Part C fails before the leading onestory structure of Part A does then the result of the first such contest is that Part A fends off collapse for a time interval and a partial deceleration has been accomplished during that interval. Now there is a Part B in existence and a new leading onestory structure of Part C. If Part C wins again in this second collision then another interval has occurred during which Part A fends off collapse, partial deceleration has occurred and we have to keep going to see what happens. An iterative process ensues of successive Part C onestory structures failing before the leading onestory structure of Part A does. During all these iterations the deceleration of Part C is occurring gradually. The time for this process to occur may be at least several seconds based on the visual evidence.
Between Bazant's assumption and my hypothetical, there is a drastic difference in the time that the deceleration of Part C must occur. Bazantts interval would be a small fraction of one second; mine would be at least several seconds. It seems there could easily be an order of magnitude of difference, perhaps more.
I'm not claiming that collapse will not occur under my alternative explanation. I'm just claiming that we do not know whether collapse will occur until we do the calculations correctly. Bazant's calculation clearly requires the deceleration to occur within an unrealistically (erroneously, in other words) short period of time and therefore does not demonstrate what it claims to demonstrate.
