Debunked: UAF Study Shows WTC7 Could Not Have Collapsed from Fire

[Mick West]

Mod Note

This thread was to discuss the 2017 presentation by Hulsey, and has been superseded by the publication of the draft report, discussed here: https://www.metabunk.org/sept-3-2019-release-of-hulseys-wtc7-draft-report-analysis.t10890/

The Claim Being Debunked Here

In a Sept 8 2017 article shared by Architects and Engineers for 9/11 Truth, the Daily Mail says:

new study shows it was impossible that the third tower collapsed from fire
...
This week a team of experts said that fire could not have caused the collapse of WTC7
...
This week, eminent Alaska University engineers dismissed [the NIST] explanation. Dr J. Leroy Hulsey, Chair of the university's Civil and Environmental Engineering Department, said: 'Fire did not and could not have caused the failure of this building.'

There are numerous problems with the claim that the study shows that WTC7 could not have collapsed from fire:

• The study is unfinished. Nothing has been published other than Dr. Hulsey giving a presentation on YouTube, and a pdf file of the slides for that presentation.
• The study is largely not new. While there is some new material, the bulk of the slides were used by Dr. Hulsey nearly a year ago, in October 2016. Most importantly the "UAF conclusions" slide is totally unchanged.
• The study only focuses on one connection. Dr. Hulsey focuses on the connection that NIST identified as a "probable initiation event" in some of its reports, but in fact NIST identified several potential connection failures. This particular connection was not the initiating one in NIST's global collapse models.
• The study makes incorrect displacement comparisons. In both 2016 and 2017 Dr. Hulsey made much of a difference in the displacement at column 79 (5.5" west vs. 2" east). But he appears to be comparing the wrong values — global instead of local displacements. https://www.metabunk.org/posts/210992/
• The study makes incorrect temperature related buckling comparisons. Dr. Hulsey claims (slide 82) his study shows col 79 did not buckle due to temperature. He lists this as a point of comparison with NIST. However NIST explicitly makes the exact same observation. https://www.metabunk.org/posts/211186/
• The study does not model fire progression. Dr. Hulsey only used one static temperature distribution, where the actual fires moved around heating unevenly.
• The study mischaracterizes NIST's modelling of the exterior. Dr. Hulsey claims the exterior columns were fixed when they were not. https://www.metabunk.org/posts/210990/
• The study mischaracterizes NIST connection modeling in the LS-DYNA model. Dr. Hulsey claims that volumes of the full-building LS-DYNA model did not have connections modeled, but his evidence for this is a misrepresentation of a different model, the ANSYS model. https://www.metabunk.org/posts/210990/
• The study was not open. At the start of the study we were told "WTC 7 Evaluation is a completely open and transparent investigation into the cause of World Trade Center Building 7's collapse. Every aspect of the scientific process will be posted here and on the university's website so that the public can follow its progress." The last such release was in 2015. Nothing has been released since then except videos of Dr. Hulsey giving versions of this slideshow.
• The study neglects unknowns. Impact damage from falling WTC1 debris, the actual fire spread and temperatures, the state of the insulation at every spot, and differences between drawings and constructions are all factors that are unknown, and make it impossible make a determination of the exact cause of the collapse.

While it is possible that Dr. Hulsey's study will eventually yield some interesting results, it is factually incorrect to say that it proves that fire could not have caused the collapse.



Background

On 9/11/2001 the two World Trade Center towers (WTC1 and WTC2) fell within two hours of being hit. When WTC1 (the North Tower) fell at 10:28AM, it damaged a third building: World Trade Center Building 7 (WTC7). Multiple fires were ignited in WTC7, firefighters were hampered by a lack of water. After observing deformation of the building as early as 2:00PM, firefighters withdrew from the area around the building at 3:30PM, fearing collapse. WTC7 collapsed at 5:20PM, after varied burning for nearly seven hours, seemingly from an interior collapse, followed by exterior. The visual resemblance to some controlled demolitions led people to think it might have been brought down deliberately.

In November 2008, NIST released their final reports on the collapse of WTC7, totaling over a thousand pages.

• NCSTAR 1-9 Structural Fire Response and Probable Collapse Sequence of World Trade Center Building 7, Volume 1 and 2
• NCSTAR 1-A Final Report on the Collapse of World Trade Center Building 7
• NCSTAR 1-9A Global Structural Analysis of the Response of World Trade Center Building 7 to Fires and Debris Impact Damage

The conclusion of NIST was that multiple damaged connections led to floors collapsing, leading to lack of support and subsequent buckling failure of a single column (column 79) which spread to adjacent columns, leading to the interior collapse, quickly followed by the exterior collapse.

Architects and Engineers for 911 Truth (a group that thinks that the World Trade Center was destroyed by explosives) approached Dr. Leroy Hulsey (a professor at University of Alaska, Fairbanks) in 2013 to:

Conduct sophisticated computer modeling of World Trade Center Building 7 to demonstrate, first, the impossibility of the collapse initiation mechanism put forth by the National Institute of Standards and Technology (NIST) and, second, that a controlled demolition controlled more readily replicates the observed destruction.

Source: http://archive.is/ZR9S5 and https://web.archive.org/web/20150330080428/www.ae911truth.org/membership-2015

The study took on more official status in 2015. AE911 was asking for over $200,000 to finance the project, and UAF put the project budget at $316,152. A web site was set up in October 2015, with the less loaded description:

WTC 7 Evaluation is a study at the University of Alaska Fairbanks using finite element modeling to evaluate the possible causes of World Trade Center Building 7's collapse.

Source: https://web.archive.org/web/20151126053055/http://www.wtc7evaluation.org/


Note: This post is largely derived from this larger thread:
https://www.metabunk.org/ae911-truths-wtc7-evaluation-computer-modelling-project.t5627/
Where relevant I've added links to that thread to some point above. Please check the discussion there before posting here.

 

[FFTR]

I agree with your problem list. There are many unknowns when dealing with WTC7. Such as fire behavior on the floor, fuel loading for the fire, damage done by the fallen debris from WTC2, only modeling fire on two floors when we know fire existed on several floors, exact temperatures, etc.

Modeling is a useful tool when one remembers the models limits and assumptions.

 

[Christopher 7]

The study only focuses on one connection. Dr. Hulsey focuses on the connection that NIST identified as a "probable initiation event" in some of its reports, but in fact NIST identified several potential connection failures. This particular connection was not the initiating one in NIST's global collapse models.

That is incorrect.

NIST said that the girder between columns 79 and 44 was pushed of its seat and that set off a chain of cascading floor failures that left column 79 unbraced on 3 sides over 9 floors which led to its buckling and that led to the total collapse of the building. NIST offers no other explanation for what started the collapse. Without that initial failure of the girder at column 79 there would be no collapse. Therefore, the NIST report does not establish that the probable cause of the collapse was fire as they posit.

"Further thermal expansion of the floor beams pushed the girder off its seat, which led to the failure of the floor system surrounding Column 79 on Floor 13. The collapse of Floor 13 onto the floors below—some of which were already weakened by fires—triggered a cascade of floor failures in the northeast region. This, in turn, led to loss of lateral support to Column 79 in the east-west direction over nine stories (between Floors 5 and 14). The increase in unsupported length led to the buckling failure of Column 79, which was the collapse initiation event."

NISTSTAR 1-9 Vol. 2 page 611 [PDF page 677]

 

[Mick West]

What you quote is just NIST's description of the "probable initiation event". NIST identifies several failed connections from the ANSYS simulation, and these were the ones used for the global collapse LS_DYNA simulation. It's the black connections here:

Note, that's just one floor, failed connections were also found on other floors.

 

[Christopher 7]

What you quote is just NIST's description of the "probable initiation event". NIST identifies several failed connections from the ANSYS simulation, and these were the ones used for the global collapse LS_DYNA simulation. It's the black connections here:

View attachment 28937

Note, that's just one floor, failed connections were also found on other floors.

Yes, there were many failed connections but none of them triggered the collapse around column 79 which caused a cascade of floor failures and led to its buckling. It was the walk-off of the girder on the north side of column 79 that triggered the collapse and without it column 79 would not have lost lateral support over 9 floors and buckled.

 

[Mick West]

Yes, there were many failed connections but none of them triggered the collapse around column 79 which caused a cascade of floor failures and led to its buckling. It was the walk-off of the girder on the north side of column 79 that triggered the collapse and without it column 79 would not have lost lateral support over 9 floors and buckled.

This was the damage case used by NIST to create their LS-DYNA simulations that showed the building collapse.

Yes, they discussed the C79/A2001 seat push-off as a possible initiating event, but they did not use it in their global analysis, and they certainly did not say it was the only possible trigger for collapse.

 

[deirdre]

It was the walk-off of the girder on the north side of column 79 that triggered the collapse and without it column 79 would not have lost lateral support over 9 floors and buckled

if that were definite, they wouldn't use the word 'probable' in their paper title. That's the point; if Husley proves eventually NIST was wrong about that, then all Husley proved was that NIST was wrong about that.

Proving a mistake in the NIST report, does not equal 'proving fires did not cause the collapse'.

 

[Christopher 7]

Yes, they discussed the C79/A2001 seat push-off as a possible initiating event

NIST said that it was the (not the possible) initial failure that led to the collapse. Here is the whole paragraph.

Initial Local Failure for Collapse Initiation. The simple shear connection between Column 79 and the girder that spanned the distance to the north face (to Column 44) failed on Floor 13. The connection failed due to shearing of erection bolts, caused by lateral thermal expansion of floor beams supporting the northeast floor system and, to a lesser extent, by the thermal expansion of the girder connecting Columns 79 and 44. Further thermal expansion of the floor beams pushed the girder off its seat, which led to the failure of the floor system surrounding Column 79 on Floor 13. The collapse of Floor 13 onto the floors below—some of which were already weakened by fires—triggered a cascade of floor failures in the northeast region. This, in turn, led to loss of lateral support to Column 79 in the east-west direction over nine stories (between Floors 5 and 14). The increase in unsupported length led to the buckling failure of Column 79, which was the collapse initiation event.

NCSTAR 1-9 Vol.2 page 611 [PDF p. 677]

On page NCSTAR 1-9 Vol. 2 page 572 [PDF p. 638] they refer to this

Initial Local Failure for Collapse Initiation

in their global analysis. Just because they didn't spell it out again doesn't mean that they didn't use it in their global analysis.

and they certainly did not say it was the only possible trigger for collapse.

It's the only trigger NIST mentioned or used. There was no need to have more than one trigger so that argument is moot.

 

[Christopher 7]

if that were definite, they wouldn't use the word 'probable' in their paper title. That's the point; if Husley proves eventually NIST was wrong about that, then all Husley proved was that NIST was wrong about that.

Proving a mistake in the NIST report, does not equal 'proving fires did not cause the collapse'.

NIST used the term "Probable collapse sequence" because they know that they did not prove that fire caused the collapse.

Although proving that NIST did not provide a valid "Probable Collapse" scenario does not prove that fires did not cause the collapse, Hulsey's analysis may well do so.

 

[deirdre]

Hulsey's analysis may well do so.

maybe, but he has a LONG way to go based on what he seems to have accomplished (and not accomplished) so far. Not sure what proving terrorists planted bombs in wtc7 will help anything, but it certainly can't hurt to check out all possibilities.

 

[Oystein]

That is incorrect.

You say that Mick's claim "The study only focuses on one connection" is incorrect, but then, to explain this, ...

NIST said that the girder between columns 79 and 44 was pushed of its seat and that set off a chain of cascading floor failures that left column 79 unbraced on 3 sides over 9 floors which led to its buckling and that led to the total collapse of the building. NIST offers no other explanation for what started the collapse. Without that initial failure of the girder at column 79 there would be no collapse. Therefore, the NIST report does not establish that the probable cause of the collapse was fire as they posit.

"Further thermal expansion of the floor beams pushed the girder off its seat, which led to the failure of the floor system surrounding Column 79 on Floor 13. The collapse of Floor 13 onto the floors below—some of which were already weakened by fires—triggered a cascade of floor failures in the northeast region. This, in turn, led to loss of lateral support to Column 79 in the east-west direction over nine stories (between Floors 5 and 14). The increase in unsupported length led to the buckling failure of Column 79, which was the collapse initiation event."

NISTSTAR 1-9 Vol. 2 page 611 [PDF page 677]

...you focus on that one connection!

Creating memes is not my thing, but you know the guy I have in mind with this:

I am not saying the study focuses on one connection

but

it focuses on one connection!​

 

[Mick West]

Here's a simplification of the differences in the studies, based on Hulsey saying everything moves together 2"

 

[Christopher 7]

Here's a simplification of the differences in the studies, based on Hulsey saying everything moves together 2"

Hulsey was incorrect when he compared the 2" of eastward movement of column 79 with the westward movement of the girder. They are two separate issues.

As for your animated graphic, [Discussion here: https://www.metabunk.org/nist-sideplates-feasibility.t9069/#post-211936]

 

[Nada Truther]

Here's a simplification of the differences in the studies, based on Hulsey saying everything moves together 2"

Doesn't Husley's example of everything moving (assuming that Mick's interpretation is correct) indicate that there are OTHER connections that might be stressed, instead of column 79? If this is the case, couldn't there be failures elsewhere that led to the collapse? I have been trying to keep up on the other thread, but I don't know how people are able to keep that going so intently. I know what he is trying to say is that NIST is wrong, but how much movement could the other connections handle? Maybe they were wrong and it wasn't 79 due to heat, but it might have been a different column that failed due to movement at 79... and that was due to heat!!

Sorry if I am chiming in way late with info that has already been discussed. like I said, I am trying to keep up, but most of the discussions are way above my pay grade. Or just back and forth argument about semantics.

 

[Mick West]

Doesn't Husley's example of everything moving (assuming that Mick's interpretation is correct) indicate that there are OTHER connections that might be stressed, instead of column 79? If this is the case, couldn't there be failures elsewhere that led to the collapse? I have been trying to keep up on the other thread, but I don't know how people are able to keep that going so intently. I know what he is trying to say is that NIST is wrong, but how much movement could the other connections handle? Maybe they were wrong and it wasn't 79 due to heat, but it might have been a different column that failed due to movement at 79... and that was due to heat!!

Sorry if I am chiming in way late with info that has already been discussed. like I said, I am trying to keep up, but most of the discussions are way above my pay grade. Or just back and forth argument about semantics.

Hulsey shows a chart of displacements that is smooth.

There's no indication in that chart of discontinuities. He explicitly states that the area around C79 "all moves together". He does not discuss any connection failures. It's not clear at all if he even modeled failure criteria.

 

[Me2]

How is the connection held together? What has to fail for the left animation to be possible? Bolts, welds, both?

 

[Oystein]

How is the connection held together? What has to fail for the left animation to be possible? Bolts, welds, both?

Four bolts. Two on the bottom, two on the top flange of the girder.

I am currently not sure if NIST gave details about how the connection failed in the Chapter 11 ANSYS model that matter. But a Chapter 8 preliminary FEA showed that said bolts would shear.

These were construction bolts, meant to keep the girder from falling off its seat during construction, before lateral beams were attached. I suppose that means they were not desjgned and dimensioned to resist the kind of lateral loads we are talking about here - such were not design considerations.

 

[Me2]

This is an open web joist girder or solid I shape?

Clearly I need to read the respective reports.

 

[Oystein]

This is an open web joist girder or solid I shape?

Clearly I need to read the respective reports.

Solid I-shape.

Reports are linked to in the OP. Check out NCSTAR 1-9!

 

[Thomas Green]

Has this criticism been sent to Dr. Hulsey?

 

[Mick West]

Has this criticism been sent to Dr. Hulsey?

Yes, I've emailed him directly. I've also been in contact with people who had contacts with him via AE911, and they have forwarded it to him. Tony Szamboti is aware of the various criticisms, and partially acknowledged the issue of local vs. global displacement in a recent podcast discussion with me.

 

[Thomas Green]

Yes, I've emailed him directly. I've also been in contact with people who had contacts with him via AE911, and they have forwarded it to him. Tony Szamboti is aware of the various criticisms, and partially acknowledged the issue of local vs. global displacement in a recent podcast discussion with me.

Do you have a link to the podcast?

No respons from dr. Hulsey?

 

[Mick West]

Do you have a link to the podcast?

Source: https://www.youtube.com/watch?v=u1CZmtR8gno&feature=youtu.be&t=36m47s

The local vs. global discussion starts at 38:22

No respons from dr. Hulsey?

No

 

[Thomas Green]

"Tony Szamboti ... partially acknowledged the issue"

I don't hear any acknowledgment. It's rather schooling of an amature by a professional.

 

[Mick West]

I don't hear any acknowledgment.

Tony says at 41:18 "he didn't give you the relative movement. He will give you that, and I know what it is"

The point the is about the UAF vs. NIST comparison slide where Hulsey compares numbers that cannot be compared.

The study makes incorrect displacement comparisons. In both 2016 and 2017 Dr. Hulsey made much of a difference in the displacement at column 79 (5.5" west vs. 2" east). But he appears to be comparing the wrong values — global instead of local displacements. https://www.metabunk.org/posts/210992/

@Tony Szamboti knows this comparison is invalid, but says it's unimportant, a "typo" that's irrelevant to the overall argument.

 

[benthamitemetric]

"Tony Szamboti ... partially acknowledged the issue"

I don't hear any acknowledgment. It's rather schooling of an amature by a professional.

Can you present an argument for the assertion that this error could have simply been a mere typo and not a fundamental misunderstanding on Hulsey's part? Not only does the typo defense not square with how the PPT presents the figure in context, but Hulsey also harped on this specific comparison in his extemporaneous remarks.

 

[Ray Von Geezer]

"Tony Szamboti ... partially acknowledged the issue"

I don't hear any acknowledgment. It's rather schooling of an amature by a professional.

What did you hear? I heard Tony cycle through typo, unimportant, out of context then when Mick points out that Dr. Hulsey has used the typo/unimportant/out of context comparison multiple times as a key point in presentations, Tony gets angry.

Tony said he knows what the relative movement is but doesn't give it during the discussion. I wonder if he's told Dr. Hulsey what it is yet?

Ray Von

 

[Thomas Green]

They can't be compared since NIST made the eastwall fixed. It has nothing to do with reality and therefore has nothing to do with UAFs model.

 

[Oystein]

They can't be compared since NIST made the eastwall fixed. It has nothing to do with reality and therefore has nothing to do with UAFs model.

No, this is FALSE. NIST did not model the east wall as fixed (i.e. infnitely rigid). Hulsey does know how to read their model and thus made up that false claim.

 

[benthamitemetric]

For reference, here is the post where where Oystein, actually citing and quoting the NIST report at length, demonstrates that NIST did not "fix" the east wall in its model of WTC7. It's not clear why Hulsey thought they did, but he's wrong. Mick even links to this exact post in the original post in this thread.

Thomas, there is a ton of material in that thread, but, as the original post in this thread makes clear, you should read (or at least skim) it before posting in this thread. You should at least read the material Mick linked to above before you repeat Hulsey's errors.

And, by the way, even if NIST had "fixed" the east wall in its model of WTC7 (which it didn't) or even if that's just a mistaken assumption upon which Hulsey erroneously relied, Hulsey's use of the comparison in question in his presentation still makes no sense and the evolving list of excuses Tony offered for such nonsensical comparison doesn't stand up to scrutiny. I think you can prove this to yourself by simply trying to cogently articulate how any of Tony's proffered excuses would actually give rise to Hulsey's nonsensical uses of the comparison in both his presentation and extemporaneous remarks. The excuses simply do not square with the reality of what has been written and said.

 

[Contrairement]

What you quote is just NIST's description of the "probable initiation event". NIST identifies several failed connections from the ANSYS simulation, and these were the ones used for the global collapse LS_DYNA simulation. It's the black connections here:

View attachment 28937

Note, that's just one floor, failed connections were also found on other floors.

In Juanary 2018, you found that the ANSYS simulation results for the same floor (13) and for the same case temperatures (but at 4.0h) were photoshoped (to be consistent with the "probable collapse event" scenario).
https://www.metabunk.org/wtc7-is-ae...f-it-was-not-key-in-nists-global-model.t9427/

Have you checked that the figure 11-27 (for 3.5h) is not also manually edited ?

I think it's best to quote the diagrams from the draw report rather than the diagrams from the final report's version, because this last one seems to twist the results of the ANSYS simulation to make them stick with the hypothesis of thermal expansion.

 

[Mick West]

were photoshoped (to be consistent with the "probable collapse event" scenario).

You don't know WHY they were manually edited, it might have been to add data from later runs that did not make it into the first draft.

I think it's best to quote the diagrams from the draw report rather than the diagrams from the final report's version, because this last one seems to twist the results of the ANSYS simulation to make them stick with the hypothesis of thermal expansion.

Or perhaps have the most up-to-date data.

 

[Svartbjørn]

You don't know WHY they were manually edited, it might have been to add data from later runs that did not make it into the first draft.

I know we use change logs the IT sector for software patches, but I also use them for process documentation updates showing what was changed and why. Do documents like this usually come with a change log of some kind and an explanation or would the sheer number of potential changes be too much as new information became available etc?

 

[Christopher 7]

There are numerous problems with the claim that the study shows that WTC7 could not have collapsed from fire:

• The study is unfinished. Nothing has been published other than Dr. Hulsey giving a presentation on YouTube, and a pdf file of the slides for that presentation.

Sarns: That is not a problem. The study was not finished in 2017, so this was simply a progress report.


mod edit: paraphrasing and unsourced assertions removed.

 

[benthamitemetric]

Sarns: That is not a problem. The study was not finished in 2017, so this was simply a progress report.


mod edit: paraphrasing and unsourced assertions removed.

It was a problem because, even though it was unfinished, AE911Truth and Hulsey were touting its purported conclusions as definitive findings.

 

[Christopher 7]

It was a problem because, even though it was unfinished, AE911Truth and Hulsey were touting its purported conclusions as definitive findings.

They had already done enough to determine that fire could not have been the cause. However, I understand your having a problem with the data to support that claim not being released after it was promised that it would be.

 

[deirdre]

They had already done enough to determine that fire could not have been the cause.

in 2017 (above) you said differently. Did I misread your comment back then? when you said "may well do so" I thought you meant "in the future with more work", but did you mean Hulsey's 2017 study already-in 2017- does so and we just didn't know it yet. ?

Although proving that NIST did not provide a valid "Probable Collapse" scenario does not prove that fires did not cause the collapse, Hulsey's analysis may well do so.

I only ask because i agreed with you back then, but if i misunderstood then i don't agree anymore.

 

[Christopher 7]

The study is largely not new. While there is some new material, the bulk of the slides were used by Dr. Hulsey nearly a year ago, in October 2016. Most importantly the "UAF conclusions" slide is totally unchanged.

Sarns: The reason the conclusions were still "unchanged" is that throughout the study Hulsey's research team kept coming up with a consistent set of findings no matter how many different aspects of WTC 7's failure they explored in depth.

 

[Christopher 7]

in 2017 (above) you said differently. Did I misread your comment back then? when you said "may well do so" I thought you meant "in the future with more work", but did you mean Hulsey's 2017 study already-in 2017- does so and we just didn't know it yet. ?

You misread what I said. I was responding to:

... if Husley proves eventually NIST was wrong about that, then all Husley proved was that NIST was wrong about that.

This is a prime example of nit-picking for the sake of argument.

 

[benthamitemetric]

They had already done enough to determine that fire could not have been the cause. However, I understand your having a problem with the data to support that claim not being released after it was promised that it would be.

So you think that, in order to determine whether fire could in any scenario cause a progressive collapse of the building, it was sufficient to test a single, unrealistic fire scenario (using a simplified heating curve that uniformly, and without any progression simulated, ramps area temperatures to certain temperatures pulled from a point in NIST's simulation when the area in question was cooling) in a limited model of part of less than half the floor space on two floors of the building? Well, good news--there may be an opening for you on the Titanic's PR team!

And, by the way, just a reminder that Hulsey announced his conclusion before he even finished the limited model described above.