Sept 3, 2019 release of Hulsey's WTC7 draft report: Analysis

Status
Not open for further replies.
Did he contribute any insight?
Not really, he kept saying that the I was wrong to say the static analysis was inappropriate because A) static analysis is commonly used to simulate building response to single common removal, and B) figure 4.14 (the leaning rainbow building with fewer glitches) was "not the analysis, just a visualization of the analysis".
Fig 4.14 Visialization of linear static analysis.jpg
I tried to explain that linear static analysis works fine for limited column removal if the loads can be redistributed without much movement or deformation. But he kept saying I was a liar and that I didn't understand.

I also tried to explain that the visualization of a static analysis would show the actual positions and orientation of the building structural members. But he seemed to keep saying that it was "a visualization" like it was some kind of artistic interpretation and that I was an idiot/liar he was going to expose to the real engineers.


Source: https://twitter.com/gerrycan1/status/1176274424271708160


However, it was a very helpful discussion as my attempts to explain things to him let me to discover the information I discussed in the above few posts.

To be clear (since people might suspect I'm suppressing his Truth) Gerrycan was banned (several times) for being very rude (not just impolite, but hurling streams of colorful insults in PMs.) Perhaps he's just a passionate Scot, but it was highly inappropiate, so he had to go. But he does read Metabunk, to gather more evidence of how wrong I am.
 
There would be a lot more if that was the case.
Not sure; if you look at this image:
WTC7stretch.png
If you look inside the green circle either a connection is disconnected (as in 'removed' by Hulsey by hand), or the connection is stretched to whatever length the simulation calculated it to be. Of course those connections would have failed long time before.

The same seems to be with the green curved beams highlighted; the floor doesn't have support anymore, so the simulation calculated how for it would bend and stretch the beams/connections accordingly.

As for the glitches in the full purple view I'm not sure, but it still seems alike; just no failure is allowed other then ones made by hand, so everything else is stretched to whatever the calculations tell it to be.

So yes you could call it a bug; but to me it seems to be a data bug; one as a result of the decisions made by Hulsey by making all connections 'non-failable', and indeed they have very little to do with reality.

Edit:
I think what I'm trying to say is that it is not so much a software bug as it is much more a data input bug.
 
My recent re-examination also helped clarify what Hulsey is doing wrong. The biggest problem is that he's using static modeling without structural failures for a dynamic event with structural failures. He is ONLY distorting members as if they are linear springs. NIST, on the other hand, despite all their problems, simulated the collapse using dynamic analysis, where horizontal structural members would fall.
N1-9 Fig 12-43 background.jpg

The impacts and the time sequence are entirely absent in Hulsey's model. When replicating what he sees as the NIST scenario he just sets up a bunch of connected springs, removes a few of them, then does a static analysis to see where the load goes. In his world no floor fails, nothing snaps, nothing falls, nothing hits anything else, connections deform linearly as far as you like. That's not the NIST scenario at all, and nor is in any way reflective of reality.

Then when he's simulating his own scenario (instantaneous controlled demolition of columns under the Penthouse and then simultaneous removal of all the columns) he does not do any analysis of the collapse. He just does heavy-handed removal of as many columns are needed so that the unsupported sections will fall in a way that resembles the actual collapse.

It's like a forensic scientist trying to recreate a blood-spatter pattern by painting the blood on the wall with a red sharpie.
 
Frankly, I had all along understood the colors of the circles this way: Red means "fails structurally - a result of analysis", green means "did not fail, we select this for manual take-out in the next iteration"

I only just now noticed that, other than the first six columns, there are no green circles in the report.

In fact, it's totally different. in the Sept 3 presentation, the column failure sequence is different after the removal of columns 76 to 81, to the extent that they were forced to remove more columns in order to keep the collapse going. But in the report, there's no manual removal after the first six.
Hulsey presentation vs.jpg

So it looks like they had two different models. The later model (used in the report) was adjusted so it collapsed. The earlier one required significant (and, I think, quite unrealistic) help in order to collapse.

This raises the very real possibility that they tweaked the model until it did what they wanted.


 

Source: https://youtu.be/a-DadyW-LR4


This is a follow-up to "Some Problems with the UAF/Hulsey/AE911Truth WTC7 Draft Report" published on Sept 7th 2019. Source: https://www.youtube.com/watch?v=7OClixCTdDw

In that video, I noted that Hulsey was using static analysis where he should have been using dynamic analysis and that his dynamic analysis was suspiciously simple.
It has since emerged that he has not actually done ANY real global dynamic analysis, despite several figures in the report being labeled as dynamic analysis - they are actually just simply box physics, set up to look like a particular result based on the static analysis.
The static analysis is also problematic in that it is linear, and the large deformations seen are non-linear.
But perhaps most seriously, there are TWO VERSIONS of the static analysis being presented - one in the report and one in the presentation. These were presented on the same day. The one shown by Hulsey shows a model that will not collapse, and requires the manual removal of columns to force a result. The one in the report is different and has no manual removal. So if they are using two very different models, which one is correct? How do we know EITHER of them is correct? What criteria is used to validate the model? Why is Hulsey using the manually adjusted model in his presentation? Why are there horrific glitches in the geometry in his final result diagram?
Simulating a collapsing building requires a dynamic analysis. NIST did one. It's not perfect, but setting up a dynamic analysis is complex, and runs are computationally expensive. NIST spent months running dynamic simulations.
You can argue their results don't look perfect. But, based on the report and presentation, Hulsey DID NOT EVEN TRY to do a global dynamic analysis. Instead, he used a dubious linear static model, of which there are multiple versions, he (or his students) pushed it way beyond anything that is linear or static, and then they used the "results" of that analysis to manually create some rotating and falling boxes, extrapolating what they thought might happen, which he then labeled "dynamic analysis".
Content from External Source
 
Mick,
It seems as if Hulsey got in way over his head... [...] Thanks for the analysis. I am not the least bit surprised they are passing this junk off as engineering or science with rigor. Most truthers couldn't tell the difference so it may work for the choir.
 
Last edited by a moderator:
Hulsey claims that their static analysis shows this: When you remove columns 79-81, the remaining structure will settle into a new static equilibrium without failing any more column(s).

But maybe I missed, probably Hulsey failed to explain: Why? How?
He removes columns over 8 stories, so the other 39 stories worth of column (or at any rate the lengths of column above and below the removed part) are still there. Presumably still connected to girders.
What happens with the girders that used to tie into the removed lengths of column - do they hang in the air? Drop to the floor below (but how could they, in a static analysis)?
Wouldn't the entire area sag, and put all the floor assemblies around col79-81 into tension? Would that not pull in walls, or tear out some girder-column connections?

Hulsey is ominously silent about this.

NIST did a (dynamic - LS-DYNA!) simulation where they removed columns 79 only, over 2 floors, without fire damage, without debris impact damage from the North Tower, just to see what would happen - and the building still ended up collapsing completely. This is Section 12.4.7 "Classic Global Progressive Collapse Analysis Results Without Debris Impact and Fire-induced Damage"
of NCSTAR 1-9, pages 594 through 598:

...
The only imposed damage in this analysis was the removal of Column 79 between Floors 11 and 13 after gravity loads were imposed. After removing the two-story segment of Column 79 at -0.8 s (4.5 s), the floors around Column 79 collapsed all the way to the roof. A kink formed in the east penthouse roof framing at 0 s (5.3 s). Column 80 buckled between Floors 37 and 39 at 6.6 s, followed by sections of Column 81 failing at the roof level due to lateral loads. As Column 81 buckled at lower stories, there was a sharp reduction in its axial load (at Floor 8) between about 2.6 s (7.9 s) and 5.1 s (10.4 s), followed by a continued but more gradual reduction in load.

The Truss 2 diagonal partially buckled at 4.9 s (10.2 s) due to falling debris from the failure of Columns 79, 80, and 81. However, the Truss 2 damage did not lead to buckling of Columns 77 and 78.

Figure 12–71 shows the structural response of the interior columns between 0.7 s (6.0 s) and 12.7 s (18.0 s). The east penthouse fell into the building at 7.5 s (12.8 s). Most of the east penthouse above Columns 79 and 80 sagged into the building after Columns 79 and 80 buckled, but the east penthouse did not fall into the building until the upper sections of Column 81 failed. The east penthouse was not disconnected from the screening wall structure until 7.5 s (12.8 s). Column 76 was impacted by the east penthouse and other structural debris around Floor 35 at about 10.8 s (16.1 s), causing the column to deflect northward. Column 77 buckled at about 11.0 s (16.3 s) around Floor 36. Column 78 then buckled at about 9.5 s (19.3 s) between Floors 25 and 33.
NCSTAR 1-9 Fig 12-71 Collapse of the north interior columns from 0.7 s (6.0 s) to 12.7 s.jpg
Figure 12–72 shows the structural response of the interior columns between 15.7 s (21.0 s) and 21.2 s (26.5 s). Column damage continued to spread westward with Column 73 buckling at about 15.0 s (20.3 s) between Floors 35 and 37. Column 58 buckled at 14.7 s (20.0 s), following a column splice failure at Floor 31 that occurred 8 s earlier, at about 6.9 s (12.2 s), but did not cause column failure. The column splice initially failed due to relative twisting and shearing motions between floors that were most acute in the northwest corner of the building. The global building motions grew in response to the column and floor structure failures in the eastern interior. The relative motions at the Column 58 splice were magnified compared to other surrounding columns due to the relatively large area supported by Column 58 and the limited shear stud connections in the northwest corner of the building, which together provided relatively less twisting resistance from the composite floor slab. As a result, the Column 58 splice above Floor 31 and some surrounding framing connections sustained damage, which continued to accumulate with subsequent building motions.
NCSTAR 1-9 Fig 12-72 Collapse of the north interior columns from 15.7 s (21.0 s) to 21.2 s.jpg
After Column 58 buckled, floor and column collapse propagated horizontally throughout the interior. By 21.2 s (26.5 s), all the interior columns had buckled above Floor 23, and the west penthouse and screening wall had fallen into the building, as indicated in Figure 12–73. Vertical motion at the roofline indicated buckling of the exterior columns, and the start of global collapse.
NCSTAR 1-9 Fig 12-73 Exterior column buckling - without debris impact and fire damage.jpg
Figure 12–74 is a plot of the kinetic energy in the global model during the analysis, where the kinetic energy is due to falling debris and debris impact. An initial peak kinetic energy occurred around 3.7 s (9.0 s), when Columns 79, 80, and 81 had buckled between 1.3 s (6.6 s) to 5.1 s (10.4 s) and the surrounding floors and columns were falling. After the debris had fallen, the collapse progression slowed, as Columns 76, 77, and 78 did not buckle until 10.8 s (16.1 s) to 14.0 s (19.3 s), respectively. The kinetic energy levels decreased during this period until between 14.7 s (20.0 s) to 16.7 s (22.0 s), after which all the interior columns buckled and led to a sharp increase in kinetic energy.

The bi-modal kinetic energy plot illustrates the complex structural system interactions that affected global stability of the building. Had the analysis been stopped at some time between 6.7 s (12.0 s) and 14.7 s (20.0 s), an incorrect and non-conservative conclusion would have been drawn that the structural damage would have been limited to the east side of the building. This type of analysis should be continued until the structure reaches a point of equilibrium (i.e., global stability) and the kinetic energy drops to zero.

This global analysis showed both a vertical and horizontal progression of failure, following the removal of a two-floor section of Column 79, as well as subsequent downward movement at the roofline due to the buckling of exterior columns. The analysis results were similar to the sequence of events shown for the results of the global analyses with debris impact damage and/or fire-induced damage, where the failure of Column 79 was due to the fire-induced damage at 4.0 h. The results of this analysis more closely matched the results for the global analysis with fire-induced damage but without debris impact damage, as the buckling of columns took place near the center of the building height.

This analysis showed that WTC 7 was prone to classic progressive collapse upon the failure of Column 79. WTC 7 was not evaluated for progressive collapse, as it was not (and still is not) standard practice to evaluate commercial office buildings for such an event. Unless a potential hazard is known, a typical progressive collapse analysis includes a critical failure location and identifies areas where the structural design could be strengthened or made more redundant to mitigate against a global or disproportionately large collapse.
NCSTAR 1-9 Fig 12-74 Global kinetic energy - without fire or debris impact damage.jpg
Content from External Source
Now that's analysis - and we even get a sense (from the kinetic energy development) that total collapse wasn't inevitable, it almost arrested between about 10 and 17 seconds after column removal, before picking up more dynamics again.

Hulsey simply does not analyse what his models actually do.
 
Something new and potentially very interesting.

Hulsey replicated a NIST diagram and table [EDIT: Table 2.1 is actually by Hulsey] in his report, pages 42-43
Metabunk 2019-09-25 09-33-46.jpg
Metabunk 2019-09-25 09-34-09.jpg
(converted to spreadsheet, attached)

"Axial force" basically means how much weight the column is carrying. So these are all the columns on floor 13, and how much weight they had to support. C79 (the one that seemed to fail first) carries the largest load, mostly because it's furthest from other columns.

The units are "kips", a kip is "1000 pounds-force", which in the case of a static column is essentially 1,000 pounds of weight being supported by that column, at that floor. So C79 is supporting 5,142,000 pounds, according to NIST's table. Columns 3-14, on the West wall, are supporting less than 1,000,000 pounds (<1,000 kips).

Hulsey does a static analysis of the whole building, and displays the result with just impact damage, before removing any core columns. Figure 4.8

4.8 Linear static analysis of Columns 79 to 81 (circled in green) being removed.png

You can zoom in on this. It has four similar numbers, which I think are the axial loads for the column on four floor. The top one is the lowest value, so probably floor 13. (with the other numbers being floors 12,11 and 10, perhaps).

Look at C79
Metabunk 2019-09-25 09-46-36.jpg

and column #3 - Figure 4.5
Metabunk 2019-09-25 09-47-49.jpg

So, if this is showing axil load in kips, then Hulsey has
  • C79 at 3,898 while NIST [Hulsey Table 2.1] has it at 5,142
  • C3 at 2,979 while NIST [Hulsey Table 2.1] has it at 916
Looking at the broader distribution, Hulsey's values range from about 2,500 to about 4,000, while the NIST [Hulsey Table 2.1] values range from around 700 to over 5000.

If they are measuring the same thing then this is a HUGE difference, and it can't be accounted for by the impact damage. It would seem to indicate the Hulsey's [static] model is very different to NIST [Hulsey Table 2.1], and probably very different to reality. Offhand, I'd say the likely explanation is Hulsely's model is a lot stiffer, so more evenly distributes loads. BUt it might also be that the mass is distributed differently.

I don't think this is just Hulsey's model vs. NIST's model. It's more like Hulsey vs. reality. The initial loads on the columns would have been calculated when the building was designed, and I'm pretty sure they will match NISTs figures.
 

Attachments

  • Column loading.xlsx
    10.6 KB · Views: 484
Last edited:
If they are measuring the same thing then this is a HUGE difference, and it can't be accounted for by the impact damage.
Minor follow-up, I was just reading the Weidlinger report, and it gave a diagram showing the change in load at the BASE of each column, accounting for damage. The missing columns we obvious massively unloaded, and the nearby columns had a higher load, but most of the building did not seem to show a significant change.
http://s3.amazonaws.com/tt_assets/pdf/WTC_7_Collapse_Analysis_and_Assessment_Report.pdf
Metabunk 2019-09-25 10-28-35.jpg
 
Last edited:
Something new and potentially very interesting.

Hulsey replicated a NIST diagram and table in his report, pages 42-43
Metabunk 2019-09-25 09-33-46.jpg
Metabunk 2019-09-25 09-34-09.jpg
(converted to spreadsheet, attached)
Mick, I could not find Hulsey's Table 2.1 sourced in the NIST report, nor does the caption for Table 2.1 refer to NIST.
Contrast with Figure 2.23 on page 42, which is taken from NIST NCSTAR 1A, where it is Figure 1-5 on page 6: Hulsey makes the reference in the caption: "Figure 2.23 Column number layout of Floor 13 (NIST, 2008, NCSTAR 1A)."

Hulsey writes "The loading condition for the Floor 12 and 13 assembly was calculated by imposing axial forces acting on the top of Floor 13" (page 42) and then on the next page: "The boundary condition for the Floor 12 and 13 assembly was obtained by analyzing the Floor 3 to 12 combination. The computer model included spring stiffnesses to simulate each connection type".

This sounds to me as if Hulsey calculated the numbers in Table 2.1, not NIST!

Hulsey goes on:
"The spring stiffness at the four far corners of Floor 12 was then checked by imposing unit forces in both horizontal directions x and y at each corner, then calculating the responses of the model under unit forces, and then solving the matrices by assembling the responses of the model to obtain spring stiffnesses."
I am not exactly sure what this means. Sounds to me they fudged with values for spring stiffness until the load distribution was "ok", although I would have no idea how they would know what "ok" is - what to compare to, if this doesn't come from NIST.
But then:
"After modeling the boundary conditions for Floors 12 and 13, we modeled the concrete slab of Floors 12 and 13."
Adding the slabs after the stiffness has already been "fudged" to conform to some target makes the assemblies TOO stiff, I think. Hulsey does not revisit the issue of column loading before playing with removing columns in Chapter 4, and so misses that loads are more evenly distributed than they should be.

"Axial force" basically means how much weight the column is carrying. So these are all the columns on floor 13, and how much weight they had to support. C79 (the one that seemed to fail first) carries the largest load, mostly because it's furthest from other columns.

The units are "kips", a kip is "1000 pounds-force", which in the case of a static column is essentially 1,000 pounds of weight being supported by that column, at that floor. So C79 is supporting 5,142,000 pounds, according to NIST's table. Columns 3-14, on the West wall, are supporting less than 1,000,000 pounds (<1,000 kips).
The only case of column loading (before fire damage) that I can find in the NIST report is Figure 11-21 in NCSTAR 1-9, page 493, which has (higher) values at floor 8 instead of floor 13:

NCSTAR 1-9 Fig 11-21 Column forces (kip) above Floor 8 for gravity loads.jpg

A bit down, in Figures 11-29, 11-37 and 11-45, they show what has changed after fire damage in Cases B and C after 3.5 or 4 hours in the ANSYS model. For example Figure 11-37:

NCSTAR 1-9 Fig 11-37 Column forces (in kip) above Floor 8 at t eq 4.0 h for Case B temperatures.jpg

The differences are not huge. Most strikingly, the corners have unloaded. Load has shifted from core and north- and south-face to east- and west-face.

...Looking at the broader distribution, Hulsey's values range from about 2,500 to about 4,000, while the NIST values range from around 700 to over 5000.

If they are measuring the same thing then this is a HUGE difference, and it can't be accounted for by the impact damage. It would seem to indicate the Hulsey's model is very different to NIST's, and probably very different to reality. Offhand, I'd say the likely explanation is Hulsely's model is a lot stiffer, so more evenly distributes loads. BUt it might also be that the mass is distributed differently.

I don't think this is just Hulsey's model vs. NIST's model. It's more like Hulsey vs. reality. The initial loads on the columns would have been calculated when the building was designed, and I'm pretty sure they will match NISTs figures.
Again, my hunch is that the invalid step was to add slabs AFTER "modeling the boundary conditions for Floors 12 and 13" by "solving the matrices ... to obtain spring stiffnesses".
 
This sounds to me as if Hulsey calculated the numbers in Table 2.1, not NIST!
Hmm, I though it was NIST based on the way it was discussed. But it seems not. Is he then at odds with his own model?

The only case of column loading (before fire damage) that I can find in the NIST report is Figure 11-21 in NCSTAR 1-9, page 493, which has (higher) values at floor 8 instead of floor 13:


That distribution does seem consistent with the table.

I did a scatter plot of the 2.1 table values (orange) against the values in figure 4.8. (blue)
Metabunk 2019-09-25 17-50-02.jpg

The totals are interesting. Figure 4.8 (Hulsey's static linear analysis with impact damage) totals 239,613. Table 2.1 (Hulsey's calculation of column loads for the undamaged building) totals 162,356.
 
Last edited:
Hmm, I though it was NIST based on the way it was discussed. But it seems not. Is he then at odds with his own model?
That's how I see it, yes.

That distribution does seem consistent with the table.
Eye-balling, I agree.

The totals are interesting. Figure 4.8 totals 239,613. Table 2.1 totals 162,356.
NIST has a total of 176,550 at floor 8 (before any damage).
Floor 8 has 39 floors above it, that's 5 (14.7%) more than floor 13, but NIST has (only) 8.7% more total load than Table 2.1. The difference is hardly significant. The sum of 239,613 is 36% higher than NIST's, where it should be 13% less. That IS significant. Again, Hulsey's loads in Table 2.1 are before adding concrete slabs.
 
Here are Hulsey floor 13 figures, scaled so they sum to the same as the floor 8 figures from 11-12, so they should be the same.
Definite similarities, but some MAJOR differences. Like
Metabunk 2019-09-25 23-19-11.jpg
16, 27, 43, 57, and 79 are the biggest difference.

16, 27, 43, 57 are all one away from a corner on a long side.


Metabunk 2019-09-25 21-57-30.jpg

Data attached.
 

Attachments

  • Column loading.xlsx
    33.4 KB · Views: 474
Last edited:
But with notably with the real footage, there's the kink, the inwards sagging of the north face that forms quite and angle.
Mick, the "kink" is THE biggest clue in determining how Building 7 collapsed. It is vital to understand what we are seeing. That is not an "inwards sagging". The east part of the building is falling over to the north, and it is pivoting far below what we can see. I covered this topic here: https://www.metabunk.org/debunked-t...7-was-pulled-inward.t10796/page-2#post-232255
Hulsey says he struggled to get this in his model results, but could not really do it.
Because it's virtually impossible to achieve that motion with an intact trapezoid, or any 4 sided geometric solid. I notified Hulsey in Feb. 2017 of same and explained why. He ignored my communication. Interestingly, NIST couldn't replicate that motion either.
 
Quotes from "A Structural Reevaluation of the Collapse of World Trade Center 7"
4.7 Summary and Conclusion
In summary, several findings were made from the analyses above:
5. The simultaneous failure of all core columns over 8 stories followed 1.3 seconds later by the simultaneous failure of all exterior columns over 8 stories produces almost exactly the behavior observed in videos of the collapse.
Content from External Source
False statement. Hulsey threw in the term "almost exactly" (to give himself wiggle room?), but it isn't near "exact". He left out the initiating motion of the visible perimeter frame collapse:B7visualanalysis2Annotated.png
The east half of the visible perimeter frame is falling over to the north to begin it's collapse. There has been no downward motion of the structure. Approximately 3/4 seconds of collapse expired before descent. Columns are failing on the east part of the structure, but not on the west part; hardly "simultaneous".
That initiating motion is the east half of the structure falling over to the north, explained here:
https://www.metabunk.org/debunked-t...7-was-pulled-inward.t10796/page-2#post-232255
Hulsey claims the visual record of collapse is "straight down":
4.7 Summary and Conclusion
...let alone the observed straight-down collapse with approximately 2.5 seconds of free fall and minimal differential movement of the exterior.
Content from External Source
He then minimizes "differential movement of the exterior", painting a quite different picture than reality.
I advised Hulsey of these details in Feb. 2017. He ignored my communication, and ignores the most important part of any structural collapse; onset. I need dig no deeper to know this report is worthless relative to the collapse of Building 7.
 
B7visualanalysis2Annotated.png
The east half of the visible perimeter frame is falling over to the north to begin it's collapse. There has been no downward motion of the structure. Approximately 3/4 seconds of collapse expired before descent. Columns are failing on the east part of the structure, but not on the west part; hardly "simultaneous".
Here's a GIF animation I just made, taking a still of the split second before the exterior face starts to fall, placing that still over @ 45% opacity to act as a referent against which to see the changes. I've omitted the initial penthouse collapse because this is just to examine the exterior face, as Joe is pointing out.



What's most notable to me is how the interior collapse runs across the building, from left to right, marked first by the initial penthouse collapse (not included above) and windows breaking below it as the floors collapse; and then the other roof penthouse structure gives way running left to right a split second before the exterior face goes down; and then exactly as that happens windows break on the far-right side of the exterior face.

So there was obviously a sustained interior collapse running left to right from this view, coring out, or caving in, the interior, and during which no explosives are heard. So by all impressions we're seeing a spontaneous progressive interior collapse, not a controlled demolition, that eventually pulls the exterior face down with it.

And here's precisely the frame from the overlay-comparison above that makes Joe's point, that at the last moment when the right edge has not yet fallen, the left side had already been falling out. So there's not a simultaneous global exterior collapse. Indeed, the collapse, both onset interior and subsequent exterior, runs left to right from this view...



Seems to me the left side wants to fall outwards, but the right side is pulled inwards, falling in unison with the interior progressive collapse as it reached the right side, having progressed from left to right. In contrast, the left side did not fall in unison with the interior collapse there. The exterior face held tight horizontally until the progressive interior collapse reached the right, at which point there was no more interior support for the face.
 
Last edited:
Careful observation seems to support the notion that the interior DID collapse down INSIDE the facade and then the perimeter comes down. This seems to suggest that the perimeter was reasonably stiff on its own without the floor bracing support. It was a structural moment frame. It would most likely fail at the locations where spandrel beams attached to the columns. It would collapse if what supported it failed. The interior first collapse suggest that the perimeter's support at its base was undermined by the interior floor collapse. A hollowed out building leaving the perimeter frame and attached curtain wall would / could flex at the column joints not mid spandrel.

It does look like the NE interior collapsed just before the rest of the interior and the kink showing the NE rotating appears to be at the boundary between the perimeter opposite the core and the corner region or column 79. This may explain the kink or rotation of the perimeter moment frame. The design below floor 8 suggests that the transfer structures determined the form of the collapse.
 
Here's a GIF animation I just made, taking a still of the split second before the exterior face starts to fall, placing that still over @ 45% opacity to act as a referent against which to see the changes. I've omitted the initial penthouse collapse because this is just to examine the exterior face, as Joe is pointing out.

Excellent illustration!!! Thank you sir. I have downloaded your looped overlay, and with your permission, will use it elsewhere, and use images from it in the book on this topic I am writing. Message me with credit details if you wish to be credited.
What's most notable to me is how the interior collapse runs across the building, from left to right,...
Yes, that is an important observation in showing the collapse was indeed progressive from east to west. More important in my view though, is behavior of the perimeter frame.
First, your illustration clearly shows the NE vertical corner was pivoting far below what is visible, negating any notion this was a "localized distortion". The entire east half of the structure fell over to the north.
Second, it shows that something "snapped" the instant the west penthouse descended, and the east half of the structure immediately lurched over to the north, bowing, then folding the north face vertically. Absent a rocket motor on the south side pushing the east half north, it had to already be leaning north for this motion to occur.
Third, and most important, the west half did not react to the east half falling north, and the west half did not follow the east half over to the north; the west half actually fell to the south. This all means the perimeter frame was not a complete 4 sided geometric solid at initiation; either a vertical break occurred out of view at onset, or was already present.
 
Excellent illustration!!! Thank you sir. I have downloaded your looped overlay, and with your permission, will use it elsewhere, and use images from it in the book on this topic I am writing. Message me with credit details if you wish to be credited.
...
Seconded!!
 
Excellent illustration!!! Thank you sir. I have downloaded your looped overlay, and with your permission, will use it elsewhere, and use images from it in the book on this topic I am writing. Message me with credit details if you wish to be credited.
Thanks. Please use it as you wish and if you want to cite a source, cite that post. The reverse 'fall up' really adds to one's ability to see what's going on better than if is just passes by in one direction.

Yes, that is an important observation in showing the collapse was indeed progressive from east to west. More important in my view though, is behavior of the perimeter frame.
First, your illustration clearly shows the NE vertical corner was pivoting far below what is visible, negating any notion this was a "localized distortion". The entire east half of the structure fell over to the north.
Good point! And I think that shows the 'kink' is complex. It seem to be not just a kink along one horizontal axis, but two horizontal axes because the left side of the face seems to be rotating forward with a pivot at the kink. And yes, that would seem to be accounted for by significant interior collapsing much lower down.

Is Hulsey actually saying that the initial penthouse collapse is just some random unrelated event that just happened to occur right before his global-failure scenario? If so, that's amazingly incoherent.

Second, it shows that something "snapped" the instant the west penthouse descended, and the east half of the structure immediately lurched over to the north, bowing, then folding the north face vertically. Absent a rocket motor on the south side pushing the east half north, it had to already be leaning north for this motion to occur.
Third, and most important, the west half did not react to the east half falling north, and the west half did not follow the east half over to the north; the west half actually fell to the south. This all means the perimeter frame was not a complete 4 sided geometric solid at initiation; either a vertical break occurred out of view at onset, or was already present.
Yes, I agree with all your observations. There's a lot more twisting and rotation going on than a lazy examination might suggest, and the kink is more of a crack, like of you had a big piece of cardboard and folded it to have a crease along it. And it's a crack formed by differential falling on the left vs right (I've not spent enough time on this issue to confidently fire off the cardinal directions for these sides or corners, so I'm just saying left and right from this point of view).
 
Is Hulsey actually saying that the initial penthouse collapse is just some random unrelated event that just happened to occur right before his global-failure scenario? If so, that's amazingly incoherent.
He's not saying it's random. He's essentially saying it was part of a planned controlled demolition. It's coherent, but ridiculous.
 
He's not saying it's random. He's essentially saying it was part of a planned controlled demolition. It's coherent, but ridiculous.
He actually rather explicitly says that he is not saying that (nor anything else).
In his September 03 presentation, he is asked to comment on what may be the cause for columns "disappearing", so to say, and his answer is "I am not going there" - which earns him approving and knowing laughter.
 
He actually rather explicitly says that he is not saying that (nor anything else).
In his September 03 presentation, he is asked to comment on what may be the cause for columns "disappearing", so to say, and his answer is "I am not going there" - which earns him approving and knowing laughter.

Yeah, so he's essentially saying that's what it is. He's just saying that he's not actually going to physically say that. Wink wink.
 
Yeah, so he's essentially saying that's what it is. He's just saying that he's not actually going to physically say that. Wink wink.
He's been skirting around that issue from the start of the project.

It is - de-facto - a prudent move for AE911. Whether deliberate or not. As it stands he has given AE911 the semblance of a professional report which meets the AE911 goals. And so far - as far as I can see - no serious debunking comment from high grade professionals. If he did say "CD" there is little doubt that serious people would line up to counter him from a range of perspectives. Including technical and ethical grounds.
 
This seems to suggest that the perimeter was reasonably stiff on its own without the floor bracing support.
Yes. Stiffness of the perimeter frame can also be ascertained visually by observing the bend of the north face as the east half falls over. The north face was considerably longer than the south wall, so the south wall would be even more resistant to bending.
 
If he did say "CD" there is little doubt that serious people would line up to counter him from a range of perspectives. Including technical and ethical grounds.

if he proved fire didn't cause the collapse, and that all the columns across 8 floors simultaneously disappearing is the only way the building could have fallen that way, then what other alternatives are there? witchcraft? lasers from space? I don't think lasers from space could wipe out every column without causing exterior damage. I guess witchcraft is the best alternative. Sounds reasonable.
 
Thanks. Please use it as you wish and if you want to cite a source, cite that post. The reverse 'fall up' really adds to one's ability to see what's going on better than if is just passes by in one direction.
Thank you, and the fall up indeed helps. Loops are the best invention ever for studying motion.;)


I think that shows the 'kink' is complex. It seem to be not just a kink along one horizontal axis, but two horizontal axes because the left side of the face seems to be rotating forward with a pivot at the kink.
It's really not complex, and if you are referring to the apex of the bend as the "axis", it is single. Hold a stiff sheet of paper up in front of you, and pull the upper left corner toward you. You can alter the bend apex by moving the corner up or down as you pull it. Keep the corner up, and you'll mimic Building 7. The bend was diagonal from the point where it eventually folded at the roof line, down to the lower left (NE) corner.
What's complex is determining how the unseen parts of the structure behaved in order to accommodate the motion of what we can see, because there was no reaction in the SW corner or the west half to the east half falling over.
But I suppose that's getting off topic. For sure, collapse wasn't "global" or "near instantaneous" as Hulsey claims, but indeed progressive.
 
if he proved fire didn't cause the collapse,
He cannot prove that claim so the surmise is moot.

"Fire didn't" (or "fire couldn't") is a global negative claim. In effect saying "there is NO scenario in which fire could cause collapse."

It is generally not possible to prove a negative. And the only exception in the WTC 7 scenario is that he has to falsify EVERY alternative. Put simply if there are (say) 8 possible scenarios call them A, B, C, D, E, F and G and he falsifies A, C, E and F he has not proved it was not B, D or G. All he can legitimately claim is that he has NOT identified a scenario where fire could cause collapse. And even that requires that he gets the engineering argument right. As several members have shown his engineering analyses are dubious at best. And to add to the challenge he faces (a) There are many more that 8 possible scenarios; AND (b) it is not pragmatically possible to identify all of them.

...and [If he proved that] that all the columns across 8 floors simultaneously disappearing is the only way the building could have fallen that way,
He hasn't proved "all columns failed" and it is effectively a strawman. Recall the two relevant aspects of strawman: (a) The building didn't fall that way - the 8 floors of near free fall acceleration applied to the perimeter shell AND (b) Hulsey is glossing over/ignoring the extant hypotheses that the core collapsed first. The long standing hypothesis that Joe Hill and igoddard have been revisiting and expanding on.


.... then what other alternatives are there? witchcraft? lasers from space? I don't think lasers from space could wipe out every column without causing exterior damage. I guess witchcraft is the best alternative. Sounds reasonable.
Sure. There are reasoned explanations but this thread is about analysis of Hulsey's Report. My comment that you responded to was my agreement with Oystein and Mick West's reflections on why Hulsey has been coy and avoiding explicitly stating CD.
 
if a high rise collapses down... all you can say is that the axial system did not transfer the loads through it to the foundations. The collapse of the region below the plane strikes of the twin towers appears to be not related to the failure of columns but the inability of the floor system to support excessive loads presented when the top sections became disassociated from the axial system. The floors collapsed while the columns stood by and did nothing to prevent or inhibit the floor collapse. Then the columns remaining after the floor collapse came down because they no longer had the stability that the floor provided as lateral bracing.

7wtc appears to be a collapse initiated low down in the structure. It likely was a progressive structural failure of the axial system within the four walls of the moment frame. The moment frame was like the exteriors of the twin towers... essentially a "victim" of the interior's failure/collapse. And it may have lost axial support as a result of the interior's collapse.

If the failure of the interior columns/axial structures was progressive it may mean that one failure led to another and not all columns/axial structure failed simultaneously. That is there likely was be no need to "take out" every column/axial structure ... just the ones to get the runaway collapse going.

A runaway progressive failure of the columns/axial structures could conceivably initiate from mechanical a mechanical failure, shearing connections, lateral distortion and the like.
 
There are reasoned explanations but this thread is about analysis of Hulsey's Report. My comment that you responded to was my agreement with Oystein and Mick West's reflections on why Hulsey has been coy and avoiding explicitly stating CD.

I am talking about Hulsey's report. I'm not talking about "free fall acceleration" or anything else. In his report HE CLAIMS he proved fire did not cause the collapse and he "proved" this by making columns disappear. That was his methodology in the report, removing (disappearing) columns from the model.

My point is, that I disagree with you that his not-saying-"CD" makes his report seem more professional. Because there are no other reasonable explanations for a bunch of columns disappearing from a building at the same time than CD.

and I don't recall Mick and Oystein saying "why" Hulsey is being coy. Although I've only skimmed parts of this thread, perhaps I missed it.
 
September 30th has come and gone and, as far as I can tell, the modeling data that Hulsey promised to upload and make available to the public by that date has not yet been made available to the public. Hulsey has made a habit of missing the deadlines he publicly proclaims, but this one is perhaps a more glaring miss than the earlier delays in that this blown deadline makes it impossible for reviewers to submit comments on whatever is contained in those files during the public review period set by Hulsey. If the release of those files is to be touted as an important element of Hulsey's report, then the review of those files is an important element of the review of that report.

At the end of the day, given the obvious flaws in the report itself, I don't think the files really matter at all; I'm sure that more flaws would be uncovered if the files were released, but I can't imagine a scenario in which the files somehow buttress the flawed methodologies in the report to an extent that overcomes those flaws. But, that said, Hulsey and AE911Truth can't use the release of those files to claim transparency while not timely releasing them during the public review period.
 
Last edited:
Being a creator of 3D simulations for 30 years, including for the Department of Defense, I notice some simulation terminology confusion. There are major differences between static and dynamic models in a simulation and static state finite element analysis and stochastic dynamic simulations. From a theoretical point-of-view everything is static at some level of computation (and reality). The different is accuracy and the cost of computation. Stochastic dynamic simulation is mostly what you see, whether it is from academia or hollywood. It is much easier to "coax" in biased ways, and has the added benefit of not requiring ten thousand years to compute with a 256 Xeon blade network. Without the data on the random variables used, you can only guess on the accuracy. I would wait for the data.
 
Stochastic dynamic simulation is mostly what you see, whether it is from academia or hollywood.

Nobody is talking about stochastic dynamic simulations. The difference that's important is:

  • Hulsey simulated the collapse initiation with a static linear analysis, then used a simple dynamic box mode to create visualization of what he though would happen next.
  • NIST simulated the collapse initiation and progression with dynamic non-linear analysis.
The key point being that NIST used a far better suite analysis model, and that by inspecting the output from Hulsey's model, it's obviously wrong (stretchy columns, and impossibly strong pivots, for example)
 
The AE project page has added this update in-text:
(Update: Please be advised the data has not been posted yet due to technical issues encountered while uploading several hundred gigabytes of data. It will be available as soon as the technical issues are resolved.)
Content from External Source
 
The AE project page has added this update in-text:
(Update: Please be advised the data has not been posted yet due to technical issues encountered while uploading several hundred gigabytes of data. It will be available as soon as the technical issues are resolved.)
Content from External Source

I wonder if they are going to extend the deadline for comments? Originally they had:

Sept 15-30 - Deadline for release of data
Nov 1 - Deadline for public comments.

So every day late for the data dump, that's a day less for the public to look at the data and comment upon it. Only three days late so far, but you can do a lot of work in a day.
 
Status
Not open for further replies.
Back
Top