The 9/11 Forum

Major_Tom wrote:The only valid question is: Is there any physical reason why you would expect the fall of a 100 story building to be one smooth process? There is none.

Nonsense! It looks like a vertical avalanche; using vaf gives the best fit for the measurements and also runs out in the correct time (+-) to the lowest level. So to a good first approximation (but only that) it is one smooth process; same applies to flowing snow avalanches.

Considering the physics, your circular logic just inserted a constraint condition into your solution with no physical justification

I just justified it (since you were unable to do so yourself, despite my provding many, many hints). In fact, it is not a formal constraint, just a useful observation to note that the best fitting force function also does run out at least as well as other alternatives tried.

(and I doubt you were even aware of it).

You, sir, have no idea what I am aware of. Mind your language in this forum.

... if there is no physical reason to expect a single stage collapse scenario, ...

There certainly is. To a good approximation the tower is uniform from top to bottom. So parismony suggests starting with such a simple equation. Guess what? It works.

There is no evidence of a single smooth process in the WTC7 collapse.

Irrelevant. DIfferent design and a crush-up to boot.

We suspect quite a different crushing process in the first 3 seconds than than later into the collapse.

Suspect away. There is no difference in what is going on at 3 seconds with 29.2 m of drop than at 3.76 seconds with 44.5 m of drop.

OOS flooring runaway distruction (if any) is a totally different physical process than early crushing.

No evidence for it.

So what is really responsible for resisting force F during the earliest moments of collapse?

We wisely rule out floor by floor column buckling. Floor slab collisions are a very important factor.

For WTC1 we note one other important factor:

One perimeter sheet slips inside the other during the initial failure. The outer truss seats are especially vulnerable on the sheet that slips to the outside. Why?
Because the inside perimeter sheet can act as a hammer which strips the outer truss connections from the sheet which fell outwards.

In the case of WTC1 (and 2 n, w and s walls) the upper perimeter walls are seen to fall outside the lower wall on all observable faces.

This gives a clear advantage to the lower block in the initial collision. Inside positioning of the lower perimeter walls allows a hammering mechanism by which the upper perimeter walls are stripped of it's connected flooring, severely damaging the upper block as witnessed.


OWE, why do I post this info here? Because understanding what is colliding with what and how gives us an advantage in guessing resistance force F.

If we really understand the first few collisions, we can guess an appropriate form for force F much better than Dr Bazant could. In his dreams Dr Bazant didn't have the info that we possess. So why would we take this new info and stuff it inside that old musty shoebox of BV, BL?

The fact that the upper perimeter slipped outside the lower perimeter during initial failure may help us explain many observables, like the little jolts being examined on the other thread.

Perhaps these collisions causing the jolts (if any) are upper floor vs 97th floor spandrel plate impacts along the N face? As the lower wall slid inside the upper, 98th 99th, 100th.... floor outer truss connections became very vulnerable to lowwer perimeter (97th floor spandrel plate) "whacks".

Really helps explain upper perimeter kickout and early destruction of the upper block.

Just think about it, guys.

Major_Tom wrote:So what is really responsible for resisting force F during the earliest moments of collapse?

We wisely rule out floor by floor column buckling. Floor slab collisions are a very important factor.

Probably not. That gives (via homogenization) a constant force as does truss seat removal. That fits the data the least well when used in isolation. When used with other, more improtant resistive forces, such as comminution, the constant force is the least important, being almost negligible.

If we really understand the first few collisions, we can guess an appropriate form for force F much better than Dr Bazant could.

Actually, the first few collisions show no different behavior than any of the others.

In summary, in the last few posts I have introduced a probable description of the first few collisions consistent with all photo evidence. This allows us to visualize the collisions and possible forces involved in the earliest movement detected from the N and NW.

As usual I had David following close behind.

I cannot respond to many of his comments because I feel they act as an energy drain. This fellow who can't watch videos has quite a lot to say about what the visual evidence shows.


Whatever I type David will quickly follow with a diet of noise based on no visual evidence.


I introduce ideas and he craps all over them.


Reader, please try to understand carefully the dynamics I am trying to describe because I cannot keep retyping it through David's noise.

He says things like: "Actually, the first few collisions show no different behavior than any of the others." Based on nothing but hot air.

I welcome a conversation based on physics and careful examination of the visual and audio record. But I won't be wasting my time responding to such a windbag anymore.

David B. Benson wrote: To a good approximation the tower is uniform from top to bottom. So parismony suggests starting with such a simple equation. Guess what? It works.

Well, the tower is not uniform from top to bottom! It consists of abt. 95% very light air, 4% horizontal elements of mostly concrete (secondary structure of SG 2-3) and 1% vertical heavy steel elements (SG 7.85 ), the latter providing the primary support of the tower with a FoS of at least 3. Of course there are also furniture, fittings and humans in the towers.

If a top part C then is dropped on the bottom part A of the tower, you can be sure that strong, primary structure (columns) will locally damage weak, secondary structure (the horizontal elements), energy is absorbed causing a jolt, and soon locally damaged elements will rub against each other in both C and A, more energy is absorbed causing more deceleration of part C. Few elements of part A will become completely detached and start dropping on their own and if they do, they might drop outside.

So just do a proper energy balance of failures and friction and you will see that the destruction is arrested within a second. That's why dropping top part C (10 units tall) on bottom part A (100 units tall) of any structure (110 units tall) from a distance of one unit (drop height thus one unit) will always be arrested at once. Too little energy input available is easily absorbed as local failures and friction. That's why no vertical avalanches (sic) ever take place.

But you are always welcome to show me a homogeneous structure, light or heavy, weak or strong, that destroys itself in a vertical avalanche as per your 1-D theory/equations.

Naturally, no sooner than I'm charged with keeping the peace, tensions rise and people get jiggy.

Major_Tom, I'd say 'windbag' is over the top.

David B. Benson: as I see it, Major_Tom is trying to delve deeper than the myriad simplifications and formulate first a narrative of confirmed and likely events, then perhaps a corresponding model. It may seem folly to you, unnecessary, or may even contradict your understanding of events on several points. However, some weight has to be given to observables - a lot, in fact. There will always be some disagreement on particulars but certain things, like a 10+ story differential in crush front location north to south, plainly visible, are not addressed by the current state of the art.

Now, if you look around, you'll see plenty of people crowing about long-explained, phantom anomalies. Nobody really gives a rat's ass about that... However, having a crush front ahead of where a momentum transfer model with no energy sinks would predict is something to look at, attempt to understand, and explain. So is an upper block that got shredded; nothing intact passed through that spire. Now, I know you say you only need the mass present for VAF, but that's not the case for any of the Bazant series of articles. These are formulated based on a rigid top. It was not. Therefore, the mechanics, however good the results may be, is missing something very fundamental and it is legitimate to probe further.

Things like this will keep coming up! It may never be explained to the satisfaction of everyone, so be it, but if there's no attempt to explain it then it's moot.

Sure, one can say it's outside the scope of the model, or it's within the error band, or the model represents aggregate as opposed to specific behavior. Fine. Then it's unexplained. And the questions will keep coming forever.

Major_Tom wants to explore and sees you as raining on that parade by insisting that everything's fine, none of this matters. Against the backdrop of countless hours poring over images, seeing things which do not correspond to any model, that's bound to be frustrating. Likewise, I'm sure it's frustrating to do a great deal of textbook style analysis and parameter estimation, getting good fits to actuals, only to be told you're crapping on ideas by pointing out those things you believe to be true or already settled.

-------------

I still see both sides of this. This is like the two raging hemispheres of my brain. This is what synthesizes new ideas, pushes the envelope until it bends, and forges new models. But, at some point, it's going to devolve into ugliness, if it hasn't already.

Could we all please try to have more respect for the ideas of others? Somehow, I manage to juggle these seemingly contradictory ideas, maybe it's the choice of brandy in my snifter, who knows...

Truce? Maybe?

benson, i think i was referring to gp cherepanov when i said the avalanche guy. he is also the fracture wave guy, and i would say his fracture wave theory would have been a WAY better tack for the NISTian/JREFers to use as an explanation for the rapid collapse.

newton wrote:... i would say his fracture wave theory would have been a WAY better tack for the NISTian/JREFers to use as an explanation for the rapid collapse.

Already thoroughly refuted.

well, jrefers and nistians keep repeating things that are thoroughly refuted, too. i'm just saying, if i was a jrefer hired to lie about it, i'd pick the fracture wave lie. it at least explains the speed of collapse, unlike imaginary number "creative" math and simulations.

Might I suggest that everyone take a look at the thesis by A.G. Vlassis entitled:

"Progressive Collapse Assessment of Tall Buildings"

Available on-line at:

http://eprints.imperial.ac.uk/handle/10044/1/1342

See especially Chapter 6.

Most of this material has also been published as a paper entitled "Progressive Collapse of Multi-Storey Buildings due to Failed Floor Impact" in Engineering Structures 31, 1522, (2009)

A quote from this paper is very pertinent to the present discussion:

"It can be concluded that in the event of a failure and subsequent impact of a single floor plate onto the floor plate below , the lower impacted system is highly unlikely to possess sufficient dynamic load carrying capacity to resist the imposed dynamic loads and prevent progressive collapse. .... This is particularly true when the falling floor completely disintegrates and falls as debris without retaining any residual strength or spanning capability

Wow, thanks for that link.

This is particularly true when the falling floor completely disintegrates and falls as debris without retaining any residual strength or spanning capability.

This should spark some interesting discussion. Looks my intuition about a rubble driven collapse may be good (that's the step after demonstrating bi-directional crush, of course). Must read.

I do get the sense, with these new papers/studies and the existence of software like Extreme Loading, that we're behind the curve over here... in some ways. Still ahead of NIST, though.

Dr. G wrote:Might I suggest that everyone take a look at the thesis by A.G. Vlassis entitled:

"Progressive Collapse Assessment of Tall Buildings"

Available on-line at:

http://eprints.imperial.ac.uk/handle/10044/1/1342

I certainly will. Thank you for finding and sharing this!

You are most welcome, ....... but be warned, the thesis is 320 pages long and heavy going, .... (but aren't they all!)

I recommend you get the Engineering Structures paper if you can - it's a little more digestible at 12 pages.

""It can be concluded that in the event of a failure and subsequent impact of a single floor plate onto the floor plate below , the lower impacted system is highly unlikely to possess sufficient dynamic load carrying capacity to resist the imposed dynamic loads and prevent progressive collapse. .... This is particularly true when the falling floor completely disintegrates and falls as debris without retaining any residual strength or spanning capability"

OMG, It looks like that most of the demo companies are out of job now. What a wonderful paragraph suported by so much of experimental evidence!

Quote from the paper:

Progressive collapse may be initiated by a range of extreme loading hazards that can be either man-made or accidental. As already noted, most of these hazards have a low probability of occurrence and thus either they are not considered in structural design for economic reasons or they are indirectly accounted for through passive protection measures rather than explicit structural calculations.

Earlier, I said:

Architects and engineers spend a great deal of time and effort to calculate loads and moments in an attempt to create a structure that will remain intact in the normal expected environment. If they spend any time at all in calculating the performance under suboptimal conditions, it's an attempt to box in the boundaries of a good design, not to see what the mechanics of failure are.

And, by mechanics of failure, I meant continuation specifically, as opposed to initiation and should have specified. The statement referenced above seems to agree.