How Buckling Led to "Free Fall" acceleration for part of WTC7's Collapse.

[Henkka]

It doesn't seem obvious to me that buckling columns would result in an acceleration 99% of g. This could be checked by measuring the acceleration of verinage demolitions when the columns are buckled. David Chandler has one video where he does such a measurement, where got 8.5 m/s^2. Probably not that many David Chandler fans here, but I'm not aware if anyone else has done this.

It would have also been nice if NIST rendered their simulation further, and showed that it also reached 9.8 m/s^2 at times, or very close to that.

 

[Oystein]

It doesn't seem obvious to me that buckling columns would result in an acceleration 99% of g. This could be checked by measuring the acceleration of verinage demolitions when the columns are buckled. David Chandler has one video where he does such a measurement, where got 8.5 m/s^2. Probably not that many David Chandler fans here, but I'm not aware if anyone else has done this.

It would have also been nice if NIST rendered their simulation further, and showed that it also reached 9.8 m/s^2 at times, or very close to that.

Can you find that Chandler video? I think 8.5 is pretty damned close to 9.8 (within 15%). Of course any verinaged building that Chandler could possibly have measured would be VERY different from the WTC steel-frame skyscrapers, and in particular would almost certainly not sport the 8th-floor transfer girders that so likely have something to do with the North wall's coming down around 9.8 m/s^2 over an elevation differential of (surprise!) 8 stories.

Also, Verinage demolition explicitly tries avoid having significant portions of the mass fall freely: What you want to do is control the collapse such that as much as possible of the falling top part actually hits structural members of the bottom part, so that the most amount of structural damage is incurred during collapse.

Also, measuring random details of what happens in an extremely complex simulation late into the collapse sequence is sure to be a fool's errand, as any slight imprecision in starting conditions is likely to result in ever larger deviations from reality later on.
The idea that such a simulation would recreate every feature that us pattern-loving, heuristics-minded little humans can think we observe, is just wrong.

You accept, uncritically, that free fall acceleration is a significant feature - but it is not, it is, like so many others, an emergent and collateral feature. It's a mantra.

 

[Henkka]

Can you find that Chandler video? I think 8.5 is pretty damned close to 9.8 (within 15%). Of course any verinaged building that Chandler could possibly have measured would be VERY different from the WTC steel-frame skyscrapers, and in particular would almost certainly not sport the 8th-floor transfer girders that so likely have something to do with the North wall's coming down around 9.8 m/s^2 over an elevation differential of (surprise!) 8 stories.

Also, Verinage demolition explicitly tries avoid having significant portions of the mass fall freely: What you want to do is control the collapse such that as much as possible of the falling top part actually hits structural members of the bottom part, so that the most amount of structural damage is incurred during collapse.

Also, measuring random details of what happens in an extremely complex simulation late into the collapse sequence is sure to be a fool's errand, as any slight imprecision in starting conditions is likely to result in ever larger deviations from reality later on.
The idea that such a simulation would recreate every feature that us pattern-loving, heuristics-minded little humans can think we observe, is just wrong.

You accept, uncritically, that free fall acceleration is a significant feature - but it is not, it is, like so many others, an emergent and collateral feature. It's a mantra.

Here's the video. Relevant part starts at around 1:50, acceleration of 8.5 m/s^2 is shown at 2:50.



Trying to adhere to the no-click policy, it's a video by prominent truther David Chandler where he compares the fall of the top of the North Tower to a verinage demolition. He argues that since the verinage demolition decelerates and the top of the North Tower didn't, that implies CD.

There's quite a bit of uncertainty here, though. To get an accurate measurement, he would have had to know how tall the French building was, or otherwise know the scale of the image. I'm guessing he just approximated it, since the point of the video wasn't to show the 8.5 m/s^2 acceleration exactly, but to show the deceleration. The deceleration would show up in the graph even if his scale was a bit off.

And yeah, the building is very different from WTC 7, of course. One could say the weight bearing down on the lower columns of WTC 7 was greater. But on the other hand, in the verinage demolition the columns are buckled simultaneously and symmetrically by machines. In WTC 7, I suppose the process would be a lot more chaotic and random, with some columns buckling before others and so on.

 

[econ41]

@Henkka Recall my repeated advice to understand the actual Twin Towers mechanisms as a basis for testing the validity of truther claims. In this case D Chandlers claims in the video. Then go back to the basis premises which underpin Chandler's claims. And his support of T Szamboti's "Missing Jolt" because both make the same fundamental errors.

Two fatal errors are that they both assume a mechanism of collapse which started by the "Top Block" (1) dropping to impact the columns of the lowers tower and (2) crushing the lower tower columns all the way to the ground. Both of those assumptions are wrong. The second one is most relevant to this video claim. Watch and listen for the number of times Chandler refers to "crushing" - he means the falling material crushing all below it including the columns. That did not happen. And the reality that it did not happen explains the rapidity of the progression of the collapse. Szamboti was looking for a "missing jolt". There was no "Missing Jolt" because the scenario Szamboti assumed never existed. The Top Block did not "drop" to impact the Lower Tower columns. The columns were never "crushed" And Chandler uses different terminology but he is also looking for a resistance that never existed. Neither the deceleration Chandler was looking for nor the "Jolt" that Szamboti looked for ever existed because the scenario they assume never existed.

I can explain that in fuller and comprehensive detail if you need. But lets now look at a couple of issues you raise remembering that Chandler is using a false setting for his claims.

..., it's a video by prominent truther David Chandler where he compares the fall of the top of the North Tower to a verinage demolition. He argues that since the verinage demolition decelerates and the top of the North Tower didn't, that implies CD.

Three issues in that argument:
(a) Like most truthers Chandler holds the mantra that "Free Fall Acceleration ["FFA"] proves CD". It doesn't as per @Oystein's previous comment. FFA, if it occurs, is a feature of the collapse mechanism not what started the collapse.
(b) The Verinage demolition was actually crushing the columns of the lower parts of the building. The North Tower (both towers actually) did not crush columns. Chandler's analogy is false.
(c) The delicious irony of a person who claims that lack of deceleration proves CD quotes a known CD example which decelerates... play with that one.

There's quite a bit of uncertainty here, though. To get an accurate measurement, he would have had to know how tall the French building was, or otherwise know the scale of the image. I'm guessing he just approximated it, since the point of the video wasn't to show the 8.5 m/s^2 acceleration exactly, but to show the deceleration. The deceleration would show up in the graph even if his scale was a bit off.

You are chasing down the derail or "rabbit burrow" of Chandler's false argument. The measurements prove nothing about the actual collapses. They are only in support of Chandler's false argument.

And yeah, the building is very different from WTC 7, of course. One could say the weight bearing down on the lower columns of WTC 7 was greater. But on the other hand, in the verinage demolition the columns are buckled simultaneously and symmetrically by machines. In WTC 7, I suppose the process would be a lot more chaotic and random, with some columns buckling before others and so on.

The topic is WTC7 - the explanatory recent comments have been about Twin Towers - specifically WTC1 - which is what your example video referenced.

 

[Oystein]

Here's the video. Relevant part ...

Relevant for what?
Comparing this specific Verinage demolition, of a not-so-tall concrete building, with the non-demolition of very tall steel-frame buildings is irrelevant from the get go.
They are different.
So what?
Did anyone expect them to be same?

You repeat, as allegedly relevant, the quirk that during the Verinage collapse, there is a brief moment of net deceleration - that is exactly what I told you: That's on purpose - direct the kinetic energy of the falling mass as squarely onto the columns below to get as much destruction as quickly as possible.

Since nothing was directed in the WTC collapses, since nothing impacted any columns to any significant degree, this did not happen at the WTC (at least not before sub-assemblies started hitting the ground).
Which neither means this was demolition, nor that it wasn't demolition.
It's just different accelerations, because the collapses were different and the structures were different.

So discussing the detailed motion of some concrete building Verinage demolition is and remains to be IRRELEVANT to what happened at the WTC.

-----------
[EDITED TO ADD:]

An apology to you - I asked you present the Chandler video with the Verinage measurments, you provided it, and I should thank you for it - thank you!

Looking at it, and the "-8.5 m/s/s" value, look what I found at 2 minutes 52 seconds:

This value is the AVERAGE over an arbitrarily chosen interval of 0.8 s.
It seems to vary within that interval: Between 1.8 s and 2.0 s, velocity changes from -3.3 m/s to -5.6 m/s, that's an acceleration of -11.5 m/s/s - higher than free fall acceleration!

The average acceleration is dependent on the interval you pick. For example:

• From 1.6 s to 2.2 s (0.6 s interval), v changes from -1.65 m/s to -6.9 m/s for an average acceleration of -8.75 m/s/s
• From 1.4 s to 2.0 s (0.6 s interval), v changes from -0.34 m/s to -5.6 m/s for an average acceleration of -8.77 m/s/s
• From 1.6 s to 2.0 s (0.4 s interval), v changes from -1.65 m/s to -5.6 m/s for an average acceleration of -9.875 m/s/s - whoa - that's freefall acceleration right there!!

The truth is of course that at no point is any part of the structure actually in free fall. Column resistance at one level went from 100% to 0% in a short but non-zero interval (in this case: initiated by hydraulic jacks), but whatever bit of the structure Chandler measured remained attached directly or indirectly to other columns, floor sections, etc.

(In the case of WTC, when the West core collapse, it no doubt progressed horizontally to the perimeter columns horizontally, by floor girders pulling or pushing, and then vertical resistance of the North wall columns in some portion of the wall dropped from 100% to 0% in a short but non-zero time interval. As most of the North wall columns did not go all the way to the ground but instead were cantilevered on transfer girders on the 8th floor, and that rapid column failure most likely occurred below the 8th floor, once the wall dropped, it dropped 8 floors essentially with no column resistance, which is why we saw that period of around free fall acceleration occur over 8 floors. And a brief bit within that interval even HIGHER acceleration than 9.8 m/s/s - because that is what happens when a horizontal girder is resisted on one end - the core end - only: The other end experiences >g acceleration. What I picture there is that the north wall was still connected via floor girders to core columns; the core columns went first and made the girders tug the wall below 8th floor inward, causing it to fail rapidly and send the wall down; then the falling core columns ran into the ground, an action transferred to the wall above the 8th floor via still connected girders, causing a brief >g episode. As a result, the entire 2.25 s / 8 floor interval AVERAGED about g.
Does the NIST full collapse model bear this sequence out? -> I don't know and I don't care. It's not going to be accurate this late into the collapse anyway. Chaos and all that.)

 

[Henkka]

This has strayed pretty far from what I was saying. I'm not making the argument Chandler was making in the video. I was basically just wondering about this statement in Mick's OP:

And that's what happened with WTC7 during those 2.25 seconds of "free fall". The columns below had buckled, and offered negligible resistance. Still tons of resistance, but compared to the mass they previously supported, it was practically nothing. Hence 99% freefall.

I was wondering if that was universally true in all cases of buckling, or was it specific to WTC 7, could it be verified somehow etc. That's when I remembered seeing that Chandler video where buckling columns resulted in 8.5 m/s^2 acceleration. So that would be about 87% of g, not 99%.

 

[Oystein]

This has strayed pretty far from what I was saying. I'm not making the argument Chandler was making in the video. I was basically just wondering about this statement in Mick's OP:

Fair enough. Apologies if I was out of line.

I was wondering if that was universally true in all cases of buckling, or was it specific to WTC 7, could it be verified somehow etc. That's when I remembered seeing that Chandler video where buckling columns resulted in 8.5 m/s^2 acceleration. So that would be about 87% of g, not 99%.

But again: the "87% of g" (8.5 m/s^2) is really just an artefact of the 0.8 s time interval that Chandler picked arbitrarily. His data points, rough as they may be, have a higher resolution, consisting of several 0.2 s time intervals - and one short 0.2 s interval has even an acceleration considerably (+17%) larger than g at 11.5 m/s^2, and depending on which intermediate interval you pick, you could get 11% below g, or essentially (+0.7%) at g.
Had Chandler measured in increments of 0.1 s, or 0.167 s, or whatever, or started his analysis a few frames sooner or later, and dabbled with various interval lengths, then you could data-mine any average acceleration between 50% and 120% of free-fall out of that video - and none of those values would be very meaningful or of much relevance.

---

(One egregious example of data-mining a video-analysis to its certain death is hidden in the "Missing Jolt" paper by A. Szamboti that Chandler mentions in the beginning of his video: A jolt is a "sudden" change in acceleration, such as acceleration going from negative to positive within a short time frame. This is, mathematically, the 3rd derivative of distance or position, the 1st being velocity and the 2nd derivative acceleration. To observe a change of acceleration in a data set with discrete data points for position, you need at least 4 neighboring data points. What Szamboti did in his paper was to first compute running average values for (I think) velocity - that is he assumed that there'd be measurement errors for position in his frame-by-frame data that get bigger with every step of derivation: The raw velocity plot would be jagged, the acceleration plot wild, and the "jolt" plot catastrophic. And so - he smoothed the data, by doing running averages. But the very smoothing algorithm over several data points guarantees with mathematical certainty that there would not appear a "jolt" in the 3rd derivation, even if the data were perfect and there really WAS a jolt in it! And sure enough: The raw data contains jolts, and a smarter smoothing algorithm than "running averages", that limits the loss of information, still sees a jolt in Szamboti's data.
Szamboti is missing a jolt because he worked with averages over arbitrarily chosen time intervals that guaranteed he would not find a jolt. Not finding a jolt was not a function of the data, much less reality; it was a function of the inept algorithm the author chose)

 

[Henkka]

Fair enough. Apologies if I was out of line.


But again: the "87% of g" (8.5 m/s^2) is really just an artefact of the 0.8 s time interval that Chandler picked arbitrarily. His data points, rough as they may be, have a higher resolution, consisting of several 0.2 s time intervals - and one short 0.2 s interval has even an acceleration considerably (+17%) larger than g at 11.5 m/s^2, and depending on which intermediate interval you pick, you could get 11% below g, or essentially (+0.7%) at g.
Had Chandler measured in increments of 0.1 s, or 0.167 s, or whatever, or started his analysis a few frames sooner or later, and dabbled with various interval lengths, then you could data-mine any average acceleration between 50% and 120% of free-fall out of that video - and none of those values would be very meaningful or of much relevance.

I'm not sure I would say it's arbitrary... He's drawing a best fit curve through a part of the data that is reasonably straight. Isn't it exactly the same as what NIST did here:

Now you can quibble over the acceleration over very brief periods of time like 0.2s or 0.4s, but I'm not sure what that accomplishes. Like if you look at the NIST graph and just draw a straight line between the data points at 2.25s and 2.5s, it looks like the building is going much faster than free fall. Should we conclude then there was some force (besides gravity) pushing or pulling WTC 7 down at that moment? It makes more sense to me to just consider it to be fuzziness in the data, measurement error, etc. It seems unlikely that the verinage demolition or WTC 7 were ever accelerating significantly faster than g.

 

[econ41]

I'm not sure I would say it's arbitrary... He's drawing a best fit curve through a part of the data that is reasonably straight. Isn't it exactly the same as what NIST did here:


Now you can quibble over the acceleration over very brief periods of time like 0.2s or 0.4s, but I'm not sure what that accomplishes. Like if you look at the NIST graph and just draw a straight line between the data points at 2.25s and 2.5s, it looks like the building is going much faster than free fall.

If you need to pursue this side track I recommend being clear as to whether you refer to the motion of Free Fall Acceleration OR to the state of Free Fall i.e. no forces other than gravity actring on the body. Your next comment touches on a situation where the distinction is important.

Should we conclude then there was some force (besides gravity) pushing or pulling WTC 7 down at that moment?

That may have been the situation for one specific measurement of the WTC7 collapse. But it is a detail not relevant to current discussion.

It makes more sense to me to just consider it to be fuzziness in the data, measurement error, etc. It seems unlikely that the verinage demolition or WTC 7 were ever accelerating significantly faster than g.

A reasonable comment if referring to the whole or large parts of either building. There is some evidence that one "spot" or "zone" (depending on whose measurmwents you refer to) may have exceeded "G" briefly. It is an intersting detail but probaly not significant in the context of the "bigger picture" issues you are discussung with @Oystein.

 

[econ41]

Szamboti is missing a jolt because he worked with averages over arbitrarily chosen time intervals that guaranteed he would not find a jolt. Not finding a jolt was not a function of the data, much less reality; it was a function of the inept algorithm the author chose)

I agree the criticism of the maths Szamboti adopted. However, as you know, I assert that there is a more fundamental problem. The "jolt" that Szamboti looked for never could have been measired even if he did use valid maths. He assumed a wrong mechanism of collapse. (D Chandler made essentially the same error.)

 

[Oystein]

I'm not sure I would say it's arbitrary... He's drawing a best fit curve through a part of the data that is reasonably straight. Isn't it exactly the same as what NIST did here:
[Image removed in Quote /Oy]

Yes - NIST did this not because they thought it was relevant or enlightning, but because David Chandler had submitted his analysis as part of the Public Comments phase after the Draft report was issued by NIST, in Summer or so of 2008.
It was, in my opinion, an unwise decision by NIST to pursue that line of analysis. You will notice that NIST draws no conclusions from the acceleration profile, so they could as well just not have done it: It's irrelevant.

Now you can quibble over the acceleration over very brief periods of time like 0.2s or 0.4s, but I'm not sure what that accomplishes. Like if you look at the NIST graph and just draw a straight line between the data points at 2.25s and 2.5s, it looks like the building is going much faster than free fall. Should we conclude then there was some force (besides gravity) pushing or pulling WTC 7 down at that moment?

Yes!

It makes more sense to me to just consider it to be fuzziness in the data, measurement error, etc. It seems unlikely that the verinage demolition or WTC 7 were ever accelerating significantly faster than g.

There is certainly "fuzziness" in the data - in all the data sets, NIST's, Chandler's various analyses... You know what? That "fuzziness" (error margins) itself can (in principle at least) be analysed - that's something Chandler, nor anyone in 9/11 "Truth" has never done. So if you compute raw velocities and accelerations directly from raw data, you are going to get results that are outside of reality.

BUT that doesn't mean that >g acceleration is not plausible or not in the data - in fact it is very likely that there were brief episodes of true >g acceleration both at WTC7 North wall roofline and the Verinaged building.

The quick explanation for why this is plausible is:

• The data shows acceleration wasn't constant - it increased and decreased
• Acceleration averaged g over some suitably chosen time interval, spanning enough data points to make this conclusion valid
• It seems implausible that acceleration was constant during this interval, as the whole data set shows acceleration varied all the time
• Therefore, it is plausible that at some moment(s) during the time interval, acceleration was <g
• To make up for the deficit and result in average = g, it follows that at some moment(s) during the time interval, acceleration was >g

And the argument for why it is possible that there were brief episodes of >g:
Consider a simple horizontal beam that is supported on and free to rotate around one end, and unsupported otherwise: the free end will descend, with the whole beam rotating around the pivot (the supported end)
The free end accelerates at >g!

Another way to get the same result is: let a horizontal beam fall freely, without rotation: All the beam will accelerate down at acceleration = g. Now aim the beam such that one end hits an obstacle: that end will experience a large upward acceleration (its velocity decreases sharply), but as the beam goes into rotation, the free end is now accelerating at considerably >g.
If this seems counterintuitive, consider a beam floating in space, at rest relative to you, no gravitation (or gravitation offset by centrifugal force in orbit): tap it on one end orthogonally to its long axis: this will make the beam spin, and also make its center of gravity move in the direction of the tapping - but the addition of rotation and translation results in the free end moving in the direction opposite to the tapping. So a force on a beam going this way --> can result in motion (i.e. acceleration) of the far end of the beam in the opposite direction <--.

The collapsing walls of the collapsing buildings that we are looking at may be / probably are attached to horizontal beams which in turn may have some support on their other ends
Thus the condition for >g acceleration of the descending end of those beams is met: As the core of the building runs into (increased) resistance, an upward force on the core-side of floor beams results in a DOWNWARD FORCE on the perimeter-side of the floor beams IN ADDITION to the force of GRAVITY.

 

[Oystein]

I agree the criticism of the maths Szamboti adopted. However, as you know, I assert that there is a more fundamental problem. The "jolt" that Szamboti looked for never could have been measired even if he did use valid maths. He assumed a wrong mechanism of collapse. (D Chandler made essentially the same error.)

My critique was meant to illustrate the perils of averaging over time intervals, not to discuss the Szamboti paper as such. That would be for another thread. I agree of course that Szamboti employed a useless model assumption. However, the pure measurement and analysis of observed motion is independent of what underlying model the author may have in mind for the cause of the motion. The point is: There may actually have been a jolt in reality and captured by the data if a competent analysis had been done, but Szamboti's approach with running linear averages made sure such a jolt, if it existed, would be hidden by the very method of analysis.

The wider point is that the numerical results of such an analysis of acceleration profils, especially when given as a single value, is meaningless when that value depends so much on the details of how you measure and over which data points you arbitrarily choose to average.

----

Summary conclusion for this and the last few threads:
- Average acceleration values are a function of what the author does with the data as much as of the data itself and thus not objective "fact".
- Acceleration of subassemblies observed during collapses, whether instantaneous, maximum or average accelerations, are generally irrelevant, except maybe in extreme cases. Having a brief episode of about g or slightly >g is not an extreme case at all.

 

[econ41]

My critique was meant to illustrate the perils of averaging over time intervals, not to discuss the Szamboti paper as such. That would be for another thread. I agree of course that Szamboti employed a useless model assumption.

Understood and agreed.

However, the pure measurement and analysis of observed motion is independent of what underlying model the author may have in mind for the cause of the motion.

Unless the author implies more than the validity of the maths or the reader is led astray with the implications about the underlying false mechanism representing reality.

The point is: There may actually have been a jolt in reality and captured by the data if a competent analysis had been done, but Szamboti's approach with running linear averages made sure such a jolt, if it existed, would be hidden by the very method of analysis.

Yes. But, though it could be "A jolt" (singular) and far more likely "a lot of jolts" it definitely could not be "the jolt" that Tony Sz. was looking for. It wasn't "missing" because it could never have been. And ignoring or denying that distinction, though irrelevant to the current discussion, has been the subject of extended controversy.

The wider point is that the numerical results of such an analysis of acceleration profils, especially when given as a single value, is meaningless when that value depends so much on the details of how you measure and over which data points you arbitrarily choose to average.

Yes.

Summary conclusion for this and the last few threads:
- Average acceleration values are a function of what the author does with the data as much as of the data itself and thus not objective "fact".
- Acceleration of subassemblies observed during collapses, whether instantaneous, maximum or average accelerations, are generally irrelevant, except maybe in extreme cases. Having a brief episode of about g or slightly >g is not an extreme case at all.

True except when "about g" or "slightly >g" or even the possibility of ">g" becomes the issue in contention.

 

[Mendel]

it looks like the building is going much faster than free fall. Should we conclude then there was some force (besides gravity) pushing or pulling WTC 7 down at that moment?

No, what you should conclude is that what happened wasn't free fall, but rather a mechanically complex sequence unfolding, which is obviously true. (See Oystein in post #171 for a simple mechanical system that exhibits >g acceleration without additional forces).

And you need to realize that it's a detraction, since demolitions don't explain excessive downward acceleration of whole building parts, either.

 

[Oystein]

Understood and agreed.

Unless the author implies more than the validity of the maths or the reader is led astray with the implications about the underlying false mechanism representing reality.

Yes. But, though it could be "A jolt" (singular) and far more likely "a lot of jolts" it definitely could not be "the jolt" that Tony Sz. was looking for. It wasn't "missing" because it could never have been. And ignoring or denying that distinction, though irrelevant to the current discussion, has been the subject of extended controversy.

Agreed, of course.

True except when "about g" or "slightly >g" or even the possibility of ">g" becomes the issue in cocontention.

But the correct way to address such contentions is not to measure the motion of arbitrarily chosen points at arbitrarily chosen times of arbitrarily chosen structures undergoing collapse.

The way to go is to point out that gravity pulls down and makes things fall, and that fall strives towards free fall once the structure is broken. There is no mystery, no contention to be had. There exists no law derivable from first principles that episodes of about freefall acceleration necessarily imply "everything has been blown up with explosives". That's simply a just-so story. A collapse of a large structure is going to be complex, and all sorts of episodes of something may occur that someone might find surprising, but that alone does not warrant a long-winded explantion.

 

[Henkka]

Yes - NIST did this not because they thought it was relevant or enlightning, but because David Chandler had submitted his analysis as part of the Public Comments phase after the Draft report was issued by NIST, in Summer or so of 2008.
It was, in my opinion, an unwise decision by NIST to pursue that line of analysis. You will notice that NIST draws no conclusions from the acceleration profile, so they could as well just not have done it: It's irrelevant.

Eh, it's not like Chandler had the power to force them to do anything. If they thought it was entirely irrelevant and uninteresting, they could have left it entirely out of the report. They could have also not drawn a big, red line through the Phase 2 datapoints and calculated its slope (32.196 ft = 9.81m). Your explanation of possible >g acceleration was interesting, though. But the fundamental thing I was wondering about is the idea that:

Buckling columns = 99% of g acceleration

How do we know that to be true? The only "experiment" like this on buildings I could think of was verinage demolitions. But there's 2 big problems...

1) Obviously, there were no hydraulic machines buckling the columns in WTC 7
2) All buildings that have been demolished through verinage are very, very different to WTC 7

So it's hard to say how comparable verinage demolitions are to the collapse of WTC 7. That also goes for soda cans. And we can't build an exact replica of WTC 7 to do an experiment, either. So it seems like the best and only thing you can do is create computer simulations of WTC 7 that could demonstrate the 99% of g acceleration. NIST created just such a simulation, but unfortunately we don't get to see much of the fall:

 

[econ41]

Eh, it's not like Chandler had the power to force them to do anything. If they thought it was entirely irrelevant and uninteresting, they could have left it entirely out of the report. They could have also not drawn a big, red line through the Phase 2 datapoints and calculated its slope (32.196 ft = 9.81m). Your explanation of possible >g acceleration was interesting, though.

Let me repeat my advice - Step #1 - "Understand the Collapse Mechanism". It is easiest to start with Twin Towers because more than sufficient evidence is on view in the graphic and video record. WTC 7 is less "out in full view" and needs some ability to comprehend the internals. But there is sufficient visible data to explain all three collapses without getting trapped down the three "dead ends" you are focusing on. Viz (1) experiments, (2) comparative other collapses and (3) maths that is both wrong for what it attempts and based on a false model of collapses.

Your current (implied ?) goal is to understand how the multiple storeys of the collapse of WTC7 at near "G" was caused. Trap #1 - it was not all of WTC7 - just the perimeter shell after the core has already collapsed internally.

So understand the mechanism, In gross overview, the core - internal structures - of WTC7 fell first leaving the perimeter shell mostly intact at the levels we can see.

We saw the perimeter shell fall rapidly resulting in the 7 or 8 storeys of near FFA. So what could cause that collapse?

Obviously cutting, buckling out of line or otherwise removing what supported most columns below that level. So recall one feature of the WTC7 design - it was bridged over the Con Ed substation using an arrangement of transfer trusses. And the failure of a transfer truss would immediately create near-zero support over a 7-8 storey gap that the structure above could fall through. But one column unsupported would not result in near FFA of the whole perimeter shell. OK so what are the plausible options?

1) Failure of all columns by buckling. Tho it looked "instantaneous" a plausible process would almost certainly have been a sequenced cascading failure led by the failure of less than the full number of columns. The remainder failing under load redistribution. (BTW this is the exact key process of the mechanism that initiated the Twin Tower's collapses. And part of the reason why I recommend starting with "Twin Towers" when gaining an understanding of all three WTC collapses. IF you understand the sub-mechanisms it is easier to put the full picture together.)

2) Failure of transfer trusses in combination with buckling failure of other columns.

AND
3) Either of those two could be precipitated by explosive CD cutting.

But the fundamental thing I was wondering about is the idea that:

Buckling columns = 99% of g acceleration

Buckling OR removal of Transfer Truss support OR cutting by explosive CD. We are trying to understand the mechanism.

How do we know that to be true?

Because we understand basic physics and can formulate explanatory hypotheses. "Experiments" may help but come well down the path - especially if they are redundant to explaining the collapse

The only "experiment" like this on buildings I could think of was verinage demolitions. But there's 2 big problems...

1) Obviously, there were no hydraulic machines buckling the columns in WTC 7
2) All buildings that have been demolished through verinage are very, very different to WTC 7

Those are only problems if you are locked down the path of force-fitting the inappropriate comparison into your hypothesis. You have actually shown why the comparison is not a viable path to understanding the near "free fall" episode in the collapse of the WTC7 perimeter shell.

So it's hard to say how comparable verinage demolitions are to the collapse of WTC 7. That also goes for soda cans. And we can't build an exact replica of WTC 7 to do an experiment, either. So it seems like the best and only thing you can do is create computer simulations of WTC 7 that could demonstrate the 99% of g acceleration.

If you don't take the first step of understanding the gross mechanism you will not be able to attempt the computer simulation.

So whether or not Transfer Truss failure was a contributor to the near "G" portion of collapse column buckling was also a part. And there are two aspects that need to be understood:
(a) How a single column buckles and how much strength it loses as it buckles; then
(b) How multiple columns fail in buckling - almost certainly involving a process of load redistribution,

We can progress the explanation of those two if you are interested. Otherwise, I can leave you to continue your discussions of Verinage, comparison of other building collapses, and criticism of maths which is irrelevant because, whether correct maths or not, it is applied to a wrong model of collapse and WTC1 not WTC7.

 

[Henkka]

Your current (implied ?) goal is to understand how the multiple storeys of the collapse of WTC7 at near "G" was caused. Trap #1 - it was not all of WTC7 - just the perimeter shell after the core has already collapsed internally.

So understand the mechanism, In gross overview, the core - internal structures - of WTC7 fell first leaving the perimeter shell mostly intact at the levels we can see.

Yeah I understand that. Here's the thing about it, though... I could understand that if the core was cut somehow, and suddenly the perimeter columns had to hold up the entire weight of the building, that they would just buckle instantly and the building would come down at 99% of g. This would be a lot like a person standing on top of a soda can, the weight on top overwhelms the resistance of the structure to such a degree that it comes down basically in free fall.

But if the core had already entirely collapsed, wouldn't the weight be much less? The "perimeter shell" of the building, as you put it, was only a fraction of the weight of the entire building. And if the core collapsed, the perimeter columns would only have to hold themselves up. Would that be maybe 5% of the weight of the whole building?

In such circumstance, it doesn't feel very intuitive that they would come smoothly down at 99% of g. It feels more intuitive that the walls would become sort of floppy and then fall one way or another. And just so I'm not going entirely off of intuition, this sort of thing could be seen in the Plasco building collapse:

You see that a portion of the perimeter shell remained standing for a while. The walls became floppy, and then it folded over like a tree. This can also be seen in another simulation NIST produced, without the debris damage:

You can see that as the core collapses, the walls don't stay rigid and then come smoothly down. Instead, they again become floppy and fold in as the core collapses. Now you can say I'm going down "dead ends", but I do think it's entirely valid to ask why the observed collapse of WTC 7 is quite different to other real collapses or attempted simulations of its collapse.

 

[Jeffrey Orling]

@Jeffrey Orling check out this snapshot from the NIST simulation and see if it changes your mindView attachment 51558

I see a collapse of the entire East side. For destruction to move laterally failures have to PROGRESS laterally. No?

 

[Mendel]

The walls became floppy, and then it folded over like a tree.

Are you claiming that trees do not fall at 9.81m/s²?

 

[Jeffrey Orling]

Let me repeat my advice - Step #1 - "Understand the Collapse Mechanism". It is easiest to start with Twin Towers because more than sufficient evidence is on view in the graphic and video record. WTC 7 is less "out in full view" and needs some ability to comprehend the internals. But there is sufficient visible data to explain all three collapses without getting trapped down the three "dead ends" you are focusing on. Viz (1) experiments, (2) comparative other collapses and (3) maths that is both wrong for what it attempts and based on a false model of collapses.

Your current (implied ?) goal is to understand how the multiple storeys of the collapse of WTC7 at near "G" was caused. Trap #1 - it was not all of WTC7 - just the perimeter shell after the core has already collapsed internally.

So understand the mechanism, In gross overview, the core - internal structures - of WTC7 fell first leaving the perimeter shell mostly intact at the levels we can see.

We saw the perimeter shell fall rapidly resulting in the 7 or 8 storeys of near FFA. So what could cause that collapse?

My hunch is that the interior collapsed slightly in advance of the perimeter... whatever the mechanism of the gutting of the interior. However 47 stories of material collapsed down inside the perimeter shell and largely bypassed the shell.
The shell connected to the foundations at 26 column locations. There were 57 columns above floor 7 in the moment frame. Axial loads were "transferred" to 26 locations using a 2 story belt truss.
The 47 stories of material landing on the ground contained by the facade pushed laterally and likely causes enough of the 26 columns to fail for the moment frame and attached curtain wall to collapse as a shell.
|
The NIST sim seems to show to failures INSIDE the perimeter not the perimeter, I could be wrong.

 

[Abdullah]

@Jeffrey Orling the caption says "all interior columns buckled"

 

[Jeffrey Orling]

@Jeffrey Orling the caption says "all interior columns buckled"

All failed but likely not all buckled... Some lost support.... some likely displaced laterally.

 

[Henkka]

Are you claiming that trees do not fall at 9.81m/s²?

Uh, I guess they do, but I was merely using it to describe the motion of that perimeter wall of the Plasco building as it fell. Maybe I shouldn't have, since trees being chopped down is not really analogous to what happened to the Plasco or WTC 7. I found a better angle of it btw:

 

[econ41]

My hunch is that the interior collapsed slightly in advance of the perimeter... whatever the mechanism of the gutting of the interior. However 47 stories of material collapsed down inside the perimeter shell and largely bypassed the shell.

Why is the need for a "hunch"? Surely it should be agreed that the core collapsed first, followed by the perimeter shell which remained mostly intact due to the strong moment frame?

The shell connected to the foundations at 26 column locations. There were 57 columns above floor 7 in the moment frame. Axial loads were "transferred" to 26 locations using a 2 story belt truss.
The 47 stories of material landing on the ground contained by the facade pushed laterally and likely causes enough of the 26 columns to fail for the moment frame and attached curtain wall to collapse as a shell.

Details aside there was a major discontinuity at those lower levels due to bridging the Con Ed sub-station. Such a discontinuity is a prima-facie argument that failure at that level resulted from or was very strongly influenced by the discontinuous structural arrangements.
|

The NIST sim seems to show to failures INSIDE the perimeter not the perimeter, I could be wrong.

We are not discussing whether NIST was right or wrong OR whether NIST's simulation helps. The topic is 'How Buckling Led to "Free Fall" acceleration for part of WTC7's Collapse.' The "part of WTC7" referred to is the perimeter shell - not the internal or "core" structure.

 

[econ41]

Your current (implied ?) goal is to understand how the multiple storeys of the collapse of WTC7 at near "G" was caused. Trap #1 - it was not all of WTC7 - just the perimeter shell after the core has already collapsed internally.

So understand the mechanism, In gross overview, the core - internal structures - of WTC7 fell first leaving the perimeter shell mostly intact at the levels we can see.

Yeah I understand that. Here's the thing about it, though... I could understand that if the core was cut somehow,

Do you accept as fact that the video record shows the core collapsing first? << If you don't we need to go back and establish that fact.

and suddenly the perimeter columns had to hold up the entire weight of the building, that they would just buckle instantly and the building would come down at 99% of g.

Not so. The perimeter only had itself to support - plus some remnant second-order issues as internal floor beams were still dragging on the perimeter.

This would be a lot like a person standing on top of a soda can, the weight on top overwhelms the resistance of the structure to such a degree that it comes down basically in free fall.

We know that the perimeter shell mostly exhibited near "free fall acceleration" ["FFA"]. Again I'm not sure that the example helps understand the specific question we are discussing viz why the perimeter of WTC7 exhibited FFA.

But if the core had already entirely collapsed, wouldn't the weight be much less?

It had already collapsed. Why the doubt? Yes the weight of the perimeter is much less.

The "perimeter shell" of the building, as you put it, was only a fraction of the weight of the entire building. And if the core collapsed, the perimeter columns would only have to hold themselves up. Would that be maybe 5% of the weight of the whole building?

Yes. Again why the "if" we know that the core collapsed first. Yes, the perimeter columns mostly only had themselves top hold up. More like 25% at an engineer's reasoned "guess". Certainly a lot more than 5%.

In such circumstance, it doesn't feel very intuitive that they would come smoothly down at 99% of g.

Accept that they did come down at near "G". You are actually questioning the very basis of the topic. Did "come down" at near FFA is what we are explaining.

It feels more intuitive that the walls would become sort of floppy and then fall one way or another.

But that intuition is wrong. See the many times posted video clips. The perimeter did not become excessively floppy not fall one way or another. It went straight down. That fact is implicitly accepted as the starting point for this thread's OP.

And just so I'm not going entirely off of intuition, this sort of thing could be seen in the Plasco building collapse:

We are not discussing Plasco. Why are you looking for third party examples which contradict the visual evidence of what did happen at WTC7?

You see that a portion of the perimeter shell remained standing for a while. The walls became floppy, and then it folded over like a tree. This can also be seen in another simulation NIST produced, without the debris damage:

There must be thousands of other collapses or simulations which do not resemble WTC7 collapse. So what? We are trying to explain what happened at WTC7 based on known facts.

You can see that as the core collapses, the walls don't stay rigid and then come smoothly down. Instead, they again become floppy and fold in as the core collapses. Now you can say I'm going down "dead ends",

Go down any trail that helps understand WTC76. Don't go down irrelevant dead end trails that do not help with WTC7

but I do think it's entirely valid to ask why the observed collapse of WTC 7 is quite different to other real collapses or attempted simulations of its collapse.

but that doesn't need answering. We already know that wTC7 looks different to others because it was different. The challenge is to understand why WTC7 el the way it actually did fall. NOT to review dozens of different and irrelevant other collapses.

 

[Edward Current]

Please, no more about Plasco. It was a very different structure in many ways, and the collapse sequences were very different in many ways.

 

[Mendel]

But if the core had already entirely collapsed, wouldn't the weight be much less? The "perimeter shell" of the building, as you put it, was only a fraction of the weight of the entire building. And if the core collapsed, the perimeter columns would only have to hold themselves up. Would that be maybe 5% of the weight of the whole building?

In such circumstance, it doesn't feel very intuitive that they would come smoothly down at 99% of g.

Galileo established in the 16th century that freefall acceleration does not depend on weight.
Experiment trumps intuition.

 

[Henkka]

Galileo established in the 16th century that freefall acceleration does not depend on weight.
Experiment trumps intuition.

We're talking about crushing here, like a man crushing a soda can.

The speed of the crushing depends on the weight of the object doing the crushing compared to how much resistance the object being crushed can put up, no?

Replying to econ41 tomorrow maybe, it's very late here

 

[Mendel]

We're talking about crushing here, like a man crushing a soda can.

"We" are not talking about crushing; in fact, I've pointed out to you that steel-frame buildings do not fail like that (mostly), but rather by the steel joints shearing apart. Half the time, you seem to acknowledge this yourself:

The walls became floppy, and then it folded over like a tree.

You can see that as the core collapses, the walls don't stay rigid and then come smoothly down. Instead, they again become floppy and fold in as the core collapses.

Once the structure is broken, there's next to no resistance.

 

[econ41]

We're talking about crushing here, like a man crushing a soda can.

No, we are not! The debate is about 'How Buckling Led to "Free Fall" acceleration for part of WTC7's Collapse.' And the recent discussions focused on what could have caused the period of near "g" acceleration that did in accepted fact occur with the perimeter shell of WTC7.

"We" are not discussing "soda can" crushing which is a side track you introduced and I and some other members have cautioned you against taking it. We need to comprehend why the WTC shell fell at that "speed".

The speed of the crushing depends on the weight of the object doing the crushing compared to how much resistance the object being crushed can put up, no?

No - for several reasons. The topic is "How buckling led...." Buckling not crushing. I have already identified two possible sub-mechanisms and we have no justification for the presumptive limitation to "buckling" as the only plausible trigger to FFA.

Replying to econ41 tomorrow maybe, it's very late here

No rush either way. I offered advice as to an alternate path to construct a reasoned hypothesis more efficiently. If you prefer to stay with the approach by discussing details - I'll leave you to that process.

 

[Abdullah]

Is speculating on our own really appropriate for Metabunk? Shouldn't we be looking for authoritative explanations? Like the NIST report.

The NIST simulation ended with all of the columns buckled, and the building essentially in freefall. What we then must do is probe their report to find out how this came to be.

 

[econ41]

Why accept NIST as authoritative? Why accept Bazant as authoritative? Or any other academic? What position do you take when they are wrong?

I made a decision in Nov 2007 to debate the true explanations of why the Twin Towers collapsed on 9/11. Those proposing CT claims would keep switching the discussion objective from "What Happened" to "Did NIST (or Bazant) explain it correctly. So I simply resolved NOT to assume NIST, Bazant or other academics were "right". Later I extended it to WTC7 BUT the criteria remain the same. What is the true explanation? THEN, if it is relevant, "Did NIST (or BAZANT or XYZ) explain it correctly"?

 

[Abdullah]

@econ41 the question of whether NIST is wrong is important to the sites audience. If it is wrong, unless the correct answer is fairly obvious to a layman (as is the case with "15 floors can't smash 95 floors" claim, or "jet fuel can't melt steel beams" or "aluminium can't cut steel beams") we are basically screwed.

But we haven't even reached that stage yet. We are speculating without even making good use of the information already available.

We must return to the NIST report, and read the relevant parts, and probe the figures, as closely as possible to figure out why, according to NIST, WTC7 entered freefall

 

[Mendel]

according to NIST, WTC7 entered freefall

I believe that's a bad paraphrase

NIST may have measured some parts at ~9.8m/s², but do they write anywhere that these parts are in free fall?

 

[Thomas B]

do they write anywhere that these parts are in free fall?

I don't have a dog in this race, but does this count?

The analyses of the video (both the estimation of the instant the roofline began to descend and the calculated velocity and acceleration of a point on the roofline) revealed three distinct stages characterizing the 5.4 seconds of collapse:

• Stage 1 (0 to 1.75 seconds): acceleration less than that of gravity (i.e., slower than free fall).
• Stage 2 (1.75 to 4.0 seconds): gravitational acceleration (free fall)
• Stage 3 (4.0 to 5.4 seconds): decreased acceleration, again less than that of gravity

https://www.nist.gov/pao/questions-and-answers-about-nist-wtc-7-investigation

 

[Jeffrey Orling]

I don't have a dog in this race, but does this count?

The analyses of the video (both the estimation of the instant the roofline began to descend and the calculated velocity and acceleration of a point on the roofline) revealed three distinct stages characterizing the 5.4 seconds of collapse:

• Stage 1 (0 to 1.75 seconds): acceleration less than that of gravity (i.e., slower than free fall).
• Stage 2 (1.75 to 4.0 seconds): gravitational acceleration (free fall)
• Stage 3 (4.0 to 5.4 seconds): decreased acceleration, again less than that of gravity

https://www.nist.gov/pao/questions-and-answers-about-nist-wtc-7-investigation

You can only measure what you can see/observe? What percentage of the building was "observed"?

 

[Thomas B]

You can only measure what you can see/observe? What percentage of the building was "observed"?

I was only answering Mendel's question about NIST:

do they write anywhere that these parts are in free fall?

 

[Abdullah]

Let's move forward. Here is the core at the moment of global collapse

You can see the buckling very clearly. Notice how the columns buckle in the same direction across multiple storeys.

Why did this happen?

 

[Jeffrey Orling]

I don't see it. It's a static image...