1. Mick West

    Mick West Administrator Staff Member

    Last edited: Sep 9, 2019
    • Like Like x 1
  2. Miss VocalCord

    Miss VocalCord Active Member

    If I understand it correct he explains what they did at this point in the presentation.

    Source: https://www.youtube.com/watch?v=TAEHhDCTaBw&t=3682


    It indeed is a static analysis where they one by one started removing the full columns and calculated/created the image and then made those images into this animation. It doesn't make any sense to me.
     
    Last edited by a moderator: Sep 9, 2019
    • Agree Agree x 1
  3. Crusher

    Crusher New Member

    I beg to differ, the penthouse is a non-issue in this video, especially when the more prominent and glaring issue is that the NIST model deforms far more than reality, and Hulsey's model hardly deforms at all because that's what the observation in reality is.

    Just read the comments of the video, and it looks they're roasting you on this issue too. Care to clarify such focus on the penthouse, while not paying equal heed to the NIST model deforming far more than reality?
     
    Last edited: Sep 7, 2019
    • Agree Agree x 1
  4. Mick West

    Mick West Administrator Staff Member

    Hulsey's model does not deform because it's obviously not an actual dynamic analysis. Again, look at Figure 4.16

    via GIPHY


    Source: https://giphy.com/gifs/VCn5AOzBAr2kgkkTcY/html5


    Do you think this is somehow more accurate than the NIST model? This is very obviously not an analysis.
     
    • Agree Agree x 1
  5. JohnP

    JohnP New Member

    In the audio (1:01:30 to 1:02:20) he maintains (as I understand it) that this is what he gets when he tries to model the NIST interpretation, and he uses this video as an argument *against* the NIST model of progressive beam collapse and in favour of his model of simultaneous collapse.
     
  6. Agonist

    Agonist New Member

    This sort of bizarre hyperbole is obstructive to real discussion Mick. Not an analysis? Come on. As someone dispassionately observing this thread it seems alarming that you felt the need to state this so strongly, especially when the fact remains that the observed collapse is much more visually similar to the Hulsey than the NIST. Any criticism of Hulsey must incorporate this or you run the risk of not looking impartial on the subject.
     
    • Agree Agree x 1
  7. Mick West

    Mick West Administrator Staff Member

    Okay, explain to me how that represents an analysis? It's a rotating block. There's no deformation. There are no collisions. No relevant physics is being simulated.
     
  8. Agonist

    Agonist New Member

    Mick, the proof of the analysis is that the simulation produced in the Hulsey model is a much closer match to the collpase than the NIST. There is no getting around this. You have defended a model for years which produced an inferior simulation. Hulsey even lasts longer: significant because the NIST model collapse could not possibly continue as it started without looking absurdly different to the actual spectacle. The building is warped so drastically by the end of the clip that it wouldnt have resembled what actually happened. You can't lament the lack of analysis in Hulsey if it observably outperforms the NIST!!
     
    • Agree Agree x 1
    • Disagree Disagree x 1
  9. Crusher

    Crusher New Member


    Can you point out where the real footage deforms? (aside from the penthouse which you've covered in depth already).
     
    Last edited by a moderator: Sep 7, 2019
  10. Keith Beachy

    Keith Beachy Active Member

    With that much tilt, the building would fail and fall down, not over. WTC 7 was not a short concrete reinforced building, it will not topple like a tree. The top would be failing.
    The purpose of computer models for WTC 7 are not to make a FX Hollywood visual version, it is to do the engineering. The math/physics in the model does not have to look like the actual collapse. NIST was not trying to make a visualization look like the collapse, they were doing an engineering analysis, not special effects. When I worked at AFWAL (old name for the USAF lab I worked at), I would be leary of a visual model of a complex event looking like the real world, as if they did the FX version, and the math/physics/engineering might be lacking. NIST was not making a cartoon to look like the WTC 7 collapse.

    The same goes for Hulsey's report and presentation. The visual effects should not be to replicate the WTC 7 video. It is the report, the engineering is the key, not the visual effects. The building tipping over like a tree is ridiculous, and does not happen in a WTC 7 type building. If visuals are the key, then Hulsey's tipping over like a tree would make the study invalid nonsense.

    If you witness a nuke going off, see the mushroom cloud, and I say E=mc2 explains it. I have modeled the blast in a simple equation (not the derivation). I never expect computer models to accurately represent what we might see in an actual physical model when it comes to dynamics and a chaotic event like a building collapsing. Except when the physics/math/engineering are behind the visuals and correct.

    [​IMG]
    Computer modeling has limitation. Steel fails in fire.

    Hulsey's results.
    With fires not fought the only culprit present, fire can't be ruled out. You can't have the penthouse fall through WTC 7 for more than 6 seconds and claim near simultaneous failure of all column. 6 seconds is not simultaneous when parts of WTC 7 had to fail before the penthouse collapses into WTC 7. The results make no sense based on the real WTC 7 collapse and fires. The simulation showing WTC 7 falling over is not reality, the top section and steel columns would fail and fall down, not over. The building was not made to stand laterally. Hulsey's visual make NIST visual look like better engineering was behind the visuals.

    Fire can cause buildings to collapse. To rule out fire as a the cause, or as a possible cause is not engineering.

    I have many question, and more if I was to make a thesis effort event of this, but I don't have 6 months.
    Did they model the damage to WTC 7 due to the debris which damaged WTC 7 and started the fires? Yes they added some debris damage. Are fire models adjusted for fires not fought? Why show a visual where the building acts like a tree, when it would fail and fall down, not tip over like a tree?

    What is troubling, reference to NIST 202 times in the report. Less of a study of WTC 7, more of a petty what NIST got wrong paper.

    I don't see find NIST errors. Why mention NIST at all based on the objective to "examine the structural response of WTC 7". I don't know what to make of 202 reference to NIST in a paper intended to be independent research.

    What is the definition of "near-simultaneous" - a question for the report authors. Is it over 6 seconds.

    What is troubling, they tried to match their models to the visual collapse video. This is not how to do a study unless you are Hollywood and you want your visuals to match the real world. Infact Hollywood FX departments have to not use the refined computer models for engineering to get real world effects. You don't try to match the visuals, you study the structure and fire science.

    Here in the report.
    "observed in videos" - To me this is bad engineering. The video is not what needs to be modeled to study how the collapse began. You don't make your model look like the video, that is not engineering, that is Hollywood FX or worse.

    Why can't fire be the cause? The report does not prove fire did not cause WTC 7 collapse. Back in 2015 Hulsey compares Windsor fire to WTC 7 fire. He says the fires in WTC 7 were smaller. Windsor had a concrete core which saved it from collapse, and steel sections outside the core quickly failed in fires. Windsor building fires were fought. Windsor building was totaled by fire. WTC 7 was totaled by fire. He can't compare them, they are different. Why compare things when they are not the same. The fire I saw pouring out of a floor of WTC 7 looked substantial, and has what looked like the flame color as if copper was being vaporized. For me his initial entry into this study looks more like he had made his conclusion before starting the study. If find it troubling Hulsey makes this building fire comparison in the favor of his preconceived conclusion - by not mentioning why the other building did not collapse, ignoring the steel sections that collapsed quickly, and failure to mention fire fighting efforts.
     
    • Agree Agree x 3
    • Winner Winner x 3
  11. benthamitemetric

    benthamitemetric Active Member

    No one is disputing Prof. Hulsey's intent. The question is whether the actual techniques he employed ultimately supported his stated conclusion. Mick and others in this thread have convincingly demonstrated that they could not. Hulsey did not create a fulsome physics-based model of any of the collapse scenarios he looked at (including the NIST scenarios). If he didn't even model the basic physics correctly, as is evident from the videos he released where, for example, 30+ floors of the building slowly rotate through another 5-10 floors of the building without any damage or deformation of any kind to any element of the building, then how can he possibly use what happens in his model to say anything about what should have happened in the NIST model, which was a fulsome physics-based model, or in reality? He can't. His nonsense models say nothing about the NIST model or about what should or could have happened in reality. His modeling techniques make no sense and the conclusions he draws from his nonsensical models are thus unsupported.
     
    Last edited: Sep 7, 2019
    • Agree Agree x 2
  12. Mick West

    Mick West Administrator Staff Member

    That's only true if the analysis is actually correct. Hulsey does not actually show any dynamic analysis, so how can you say it's better?
     
  13. Mick West

    Mick West Administrator Staff Member

    The point here is that in Hulsey's "dynamic" model, figure 4.16, the building rotates 45 degrees through itself and the floor.

    But with notably with the real footage, there's the kink, the inwards sagging of the north face that forms quite and angle. Hulsey says he struggled to get this in his model results, but could not really do it.

    WTC-7-sagging-loop. WTC7BBC. WTC7TiltBack.
     
  14. Mick West

    Mick West Administrator Staff Member

    • Useful Useful x 3
    • Informative Informative x 2
  15. Agonist

    Agonist New Member

    This is really a very alarming statement to make. I could understand if you had said 'the model never quite looks like the event, even when the engineering and data involved are sound'. But when you add that you would be particularly suspicious of a model BECAUSE it looked so much like the event, it sounds very much like deliberate misleading.

    What then is the worth of producing a simulation based on a model if you are assured of the accuracy of the engineering regardless of what it looks like?...unless it looks like the actual event in question in which case you disbelieve it? This is not a believable position at all.
     
    Last edited: Sep 7, 2019
  16. Agonist

    Agonist New Member

    Again how can a simulation look LESS like the event than another simulation but be considered superior? That invalidates the entire premise that simulations are in any way useful. And given that NIST refuse to be as open as Hulsey with their data you end up in the Bizarro world position of the less visually representative model with the less open investigation being considered the more accurate and superior position. This is an untenable position Mick. A model that incorporates the relevant physics and dynamic analysis is almost certainly not going to look LESS like the event than one which does not.
     
  17. Mick West

    Mick West Administrator Staff Member

    The superior model is the one that is most accurate, not the one that gives you the result you want in one situation. Hulsey has a model where if you remove six core columns then this happens.

    via GIPHY


    Source: https://giphy.com/gifs/VCn5AOzBAr2kgkkTcY/html5


    Since that looks entirely unlike anything possible in reality, then it's a pretty good bet that the model is wrong. Or that's not actually a proper model.
     
  18. deirdre

    deirdre Moderator Staff Member

    has Hulsey released his data yet?
     
  19. benthamitemetric

    benthamitemetric Active Member

    You are being presented with two very different ways to make a model of the collapse of WTC7: (1) build as accurate a model of the building as reasonably possible (including all major structural elements and connections) and then observe via physics-based dynamic analysis what happens to those elements when a reasonably accurate damage state is applied thereto, or (2) make a simplistic wire frame model of the steel frame of the building and manually animate it to superficially appear to match some (but not all!) aspects of the actual collapse event (or to match other pre-determined motions).

    The first approach is NIST's approach. The second approach is Hulsey's approach. Only the first approach can give you any insight into what may have plausibly happened to the building in reality, and I would argue that the second approach is of no scientific merit or interest whatsoever. @Keith Beachy is entirely correct that the second approach--Hulsey's approach--is akin to making CGI for a movie. Here the movie was decidedly low budget and the FX team decidedly incapable, however.
     
    • Agree Agree x 2
  20. econ41

    econ41 Active Member

    Keith Beachy has given an extensive explanation and supported by benthamitemetric and other comments.

    Put very simply there are two reasons for making models or doing simulation graphics. They are:
    1) To present something that LOOKS like - visually resembles the actual event. Usually for a lay person audience; OR
    2) To demonstrate some aspects of engineering. For professional research. These models will almost always diverge from visual representation to deal with aspects such as scaling and the need to allow of legitimate quantified measurements.

    Two very different needs and they often - in this case do - conflict. The distinction has been explored in much detail in previous threads on this forum. ( discussion with member Cube Radio - I'll see if I can locate relevant threads - there were three or four as the topic "evolved".)

    NIST's model is a rigorous engineering model. It grossly exaggerates the distortions - way beyond the ductile limits of the materials. For the simple reason it is showing directions of motion for the components and is NOT intended to be a visual replica.

    Hulsey's models appear to be more targeted towards visual resemblance. With many indications that he is targeting the AE911 "members" audience of lay persons - I will leave that issue for possible future discussion.

    So let's take your other points and test them against those two distinct and conflicting goals for modelling and simulations:
    You are conflating the two reasons for modelling. Your definition of "useful" implies - subsumes - "visually similar" and you clearly are of the opinion that visually similar simulations are superior. You are conflating - not distinguishing - two very different purposes - and the two purposes - "visually similar" and "engineering accurate" are in conflict. Hulsey's versions "look visually better" but they are not engineering rigorous.

    You comments about openness with data are not relevant here but they are a common issue of contention. The "more accurate and superior position" is inherent in the method of simulation. Which aspect is not affected by availability of data. The point is currently moot anyway - we don't have Hulsey's data at this stage.

    Sadly your conclusion is 180 degrees off course for the reasons I have given. A model that incorporates the relevant physics and dynamic analysis WILL almost certainly NOT LOOK like one intended to be a visual replication.
     
    • Agree Agree x 4
  21. econ41

    econ41 Active Member

    Not at this stage possibly. I note and agree your courteous and objective approach.

    However I foreshadow the need to debate Hulsey's intent. It is becoming increasingly clear that T Szamboti has had significant input. The incorporation of Tony's EPH and CD claims the most blatant example. The extent to which Hulsey may have compromised scientific rigour in order to satisfy AE911 expectations is IMO certain to become a sub-topic of increasing interest. One example being the "adjustment" of the modelling simulations to more closely resemble the real event.
     
    • Like Like x 1
  22. Mick West

    Mick West Administrator Staff Member

  23. Agonist

    Agonist New Member

    He says he will later this month
     
  24. Oystein

    Oystein Active Member

    I do not say he won't - he likely will - but I'd like to remind you that they promised a completely open and transparent study that would publish everything so that the public could follow the progress, and they created an expectation of what that would entail, by publishing the work diaries of the assistants as well as discussions of their daily progress - all in 2015. But then they promptly stopped doing that entirely - presumable after the money solicited with the help of such promises had come in. For very long stretches of time - 2 years, really, there was NO new information, the publich was kept in the dark about the project's progress, and never had a chance to provide feedback, as promised.

    So I am not totally confident the promise will be kept.
     
    • Agree Agree x 1
  25. Agonist

    Agonist New Member

    Which surely begs the question: why, if this is your stated position, would you create a comparison video of Hulsey NIST and the actual collapse side by side?? Please explain
     
  26. Kidda

    Kidda Member

    Thx Mick, Thx Metabunk.
    One second was enough for me. No building can ever take the Husley position without blowing to smithereens, unless Rem Coolhaas wants it. One second was enough.
    Mick’s analyses of the penthouse animation makes it even worse: it moves it’s feet outward, meaning that it pushes the floor aside, straight impossible.
    When this is signed by Architects and Engineers, they have not the faintest idea of structural behavior.
    Hulsey’s work is bad, very bad.
     
    Last edited by a moderator: Sep 8, 2019
    • Agree Agree x 2
  27. Oystein

    Oystein Active Member

    Agonist insinuates that Hulsey's simulation (presumably the one where he removes ALL columns at once) actually visually appeared similar to the event. But is that true?

    I think this is as good a time as any to list the visual characteristics of the real event, as are known from the video record, and check which of those are or aren't replicated by Hulsey's model (and would not be visible in the sim videos released so far). Perhaps let's also do an appraisal of which visual clues are significant, which aren't, and why. Because, you see, Hulsey argues this(page 2, Executive Summary, my bolding):
    Here are the details of the "behavior observed" that Hulsey mentions:
    1. "collapse of the east penthouse approximately 7 seconds prior to the collapse of the rest of the structure" (p. 5)
    2. "straight-down collapse" (p. 5)
    3. "2.25 and 2.5 seconds of free fall (i.e., gravitational acceleration) during its collapse" (p. 11)
    4. "The debris pile of WTC 7 was contained mostly inside the building’s footprint. ... (see Figure 1.7)" (p. 11)
    5. "it did not have large pieces of concrete flooring or intact structural framing that would be expected in a building collapse (see Figure 1.7)." (p. 11)
    6. "[Key Feature 1.]The collapse of the east penthouse, which begins approximately 6.9 seconds prior to the descent of the north face roofline" (p. 91)
    7. "[Key Feature 2.]The collapse of the screenwall and west penthouse, which begins approximately 0.5 to 1 second prior to the descent of the north face roofline" (p. 91)
    8. "[Key Feature 3.] The descent of the north face roofline, which progresses at a rate of free fall for approximately 2.25 to 2.5 seconds over a distance of approximately 105 feet or 8 stories," (p. 91)
    9. "[Key Feature 3., continued] during which the building’s sheathing remains attached to the exterior steel framing and does not experience visible differential movements" (p. 91)
    10. "window breakage, cracking of the facade, and exterior deformation, none of which were observed" (p. 91)
    11. "the minimal differential movement of the exterior seen in videos of the collapse" (p. 91)
    1. and 6. are the same feature
    2. and 4. overlap greatly
    3. and 8. are the same feature
    9. and 11. are the same feature, and 10. is closely related

    So this boils down to 6 distinct features, which Hulsey thought worthy of a mention:

    i) EPH descends several seconds before the rest - no details given, such as that it kinks
    ii) Screenwall and WPH start descent 0.5 to 1 s prior to north wall - no further details given
    iii) Collapse staight down, largely into the footprint
    iv) 2.25 to 2.5 s of FFA of the north face roofline
    v) minimal differential movement of the exterior, resulting in no window breakage, no cracking of the façade, and no exterior deformation
    vi) no large pieces of concrete flooring or intact structural framing in the debris pile post-collapse

    Hulsey notes (page 91):
    So...
    • Feature i) is covered by NIST. What's more, in the NIST analysis, this is a result (output) of the simulation of the damage accumulation, while in the Hulsey model, it is a premise (input) of the simulation: The EPH drops a few seconds before the rest because Hulsey artifically forces it to. NIST's model explains, Hulsey's doesn't
    • Feature ii) is covered by NIST. What's more, in the NIST analysis, this is a result (output) of the simulation of the damage accumulation, while in the Hulsey model, it is a premise (input) of the simulation: The west core drops a moment before the perimeter because Hulsey artifically forces it to. NIST's model explains, Hulsey's doesn't
    • Feature iii) is FALSE - as I showed in post #29, the building did NOT collapse into its own footprint, and NOT (entirely) straight down, instead it had major parts hitting buildings across two different streets. In addition, Hulsey FAILS to show that this feature is replicated by his preferred simulation (all columns removed).
    • Feature iv) is covered by NIST (according to Hulsey - I am not aware that NIST, or Hulsey, or anyone, ever actually analysed the acceleration of the north wall roofline in NIST's global collapse simulation). What Hulsey misses is the few tens of seconds before FFA is reached. He again forces the FFA, by artificially, and without explaining (stating a cause), removing all perimeter columns at once, thus making free fall a premise (input) of his model.
    • Feature v) is FALSE - there was deformation of the facade observed, there was differential movement, there was window breakage. See below.
    • Feature vi) is a bare assertion, not supported by a proper citation nor by own study. The only reference, Figure 1.7, actually DOES show very large pieces of still-connected structural framing. In addition, Hulsey FAILS to show that this feature is replicated by his preferred simulation (all columns removed).
    In effect, we see that Hulsey replicates NO actual, real feature that NIST doesn't - none anyway that he cared to mention himself.
    Worse yet: The few true features that he replicates arise not as a result (output) of the model from actual physical causes, but as a direct result of assuming (inputting) these features: Making the EPH fall seconds before the WPH, making the WPH descend moments before the perimeter, making the perimeter fall freely by removing all columns at once. This has zero explanatory value.

    I note with interest that nowhere in the report, Hulsey claims the collapse was "symmetric". He says at one point (page 11): "WTC 7 would be expected to experience a combination of axial rotation and bending of members, resulting in a disjointed, asymmetrical collapse. Asymmetrical, tipping behavior is especially likely because WTC 7 did not have planar symmetry", but falls short of making "symmetry" a feature to look for. He writes these two sentences in an explanation of why he thinks FFA is relevant.

    Now of course there are more major visual features of the collapse that Hulsey does NOT mention:
    1. The kink in the north face that forms as the EPH disappears
    2. The counter-clockwise motion of the entire structure (north-east corner falls to the north, south-west corner to the south)
    3. The east part of the north wall lags behind the center and west parts after a few seconds
    4. Numerous windows do break after the EPH, later the WPH, start to descend
    5. The eastern facade bulges dynamically as the EPH falls
    For visual references, see Mick's post #54.

    Also, prior to the rapid collapse of the EPH
    • The building was observed to bulge etc, show signs of distress
    • The roofline was already observed to move (up and down) for as much as 30 seconds.
    Hulsey of course replicates none of these, nor are any of these plausible as results of his "all core columns at once, then all perimeter columns at once" scheme.

    Edited to add:
    I asked, at the beginning of this post:
    "Agonist insinuates that Hulsey's simulation (presumably the one where he removes ALL columns at once) actually visually appeared similar to the event. But is that true?"

    The answer now is:
    Hulsey's simulation "visually appeared similar to the event" only with respect to those features that he explicitly forced on the model by making columns disappear at opportune time, without explanation why they would, how they could disappear like that.
    Hulsey fails to replicate any other features.
     
    Last edited: Sep 8, 2019
    • Like Like x 2
    • Informative Informative x 2
    • Useful Useful x 2
    • Winner Winner x 1
  28. Kidda

    Kidda Member

    Question:
    What does Peer Review mean in USA? Which parties Peer and who funds it?
     
  29. Oystein

    Oystein Active Member

    The sponsor of the study, AE911Truth, has selected a panel of reviewers. It is not clear to me who they are, whether this same panel does a peer review at this time, and whether they receive any funding.

    It is safe to assume that this panel is selected from people AE911Truth already reliably knows to believe in explosive demolition of WTC7.
     
  30. deirdre

    deirdre Moderator Staff Member

    Moderator deirdre

    All Members:

    Please follow posting guidelines.
    Do not paraphrase other members. If you question what their meaning is, then ask them. Do not state what their meaning is as if it is a fact.
    Provide backup source links for any claims you make.
    And please adhere to the Link Policy.


     
    Last edited: Sep 8, 2019
  31. deirdre

    deirdre Moderator Staff Member

  32. econ41

    econ41 Active Member

    Peer review is a standard step in Academic Publishing and it is practised in many countries and for similar reasons and processes.

    Simply stated the purpose is to assess whether or not a proposed paper is "good enough" to be put into formal academic debate. It is read and assessed by "peers" of the author(s) - AKA those who should be familiar with the professional context for the topic. After "peer review" is passed the paper is "published" - usually by printing in the professional journal of the relational body of expertise or some associated journal. "Debate" takes the form of responding papers - critiques which may be of or against - published in later editions of the journal. The normal ritual then allows the original author to make "closing comments".

    The key point often lost sight of in these on line discussions is that "peer review" is ONLY a preliminary step. It does little more than ensure no gross errors or embarrassing oversights are in the paper. It specifically DOES NOT guarantee that the claims asserted in the paper are correct. It allows the paper to be "published" formally after which counter claims or support may be presented. So it doesn't finish debate - merely allows debate to start.

    The main problem in the setting of these online 9/11 discussions is that both "sides" of debate have afforded "peer review" a status which it does not warrant or deserve. "Peer review" is far too often assumed to be a guarantee of accuracy. It is not.

    One example of the type of issue which arises from having "peer review" on too high a pedestal is the many times repeated claims that the NIST Reports were not "peer reviewed". That is true. And the reason should be obvious. The NIST Reports were "put out for public comment" at the draft stage. And that is a far more demanding test - anyone with relevant interest and expertise is free to comment. Peer review is in secret with a limited audience who are "in club". "Out for public comment" has no such limitations.

    So those are the broad explanations. There are differences of standard both between and within the various professions. Some "peers reviews" much more rigorous than others. I am not aware of any significant differences with US practice - but I'm AU based - so a US member may wish to give some US specific comments.
     
  33. Miss VocalCord

    Miss VocalCord Active Member

    If I understand correct Hulsey first modeled certain types of connections in ABAQUS. (from page 55 and on)
    He then applied a number of analysis to these connections (nonlinear, moment-rotation response and shear-displacement response).
    Based on these analysis the connections in the global model were replaced with rotational nonlinear springs (page 53):
    And schematic (from page 52):
    However in none of the analysis shown (page 59 to 62) I see any of the connections being disconnected at any time.
    Is this maybe what is being seen in some of the global models where materials are simply stretched to (in my opinion) unrealistic lengths, simply because none of the connections will fail?
    [​IMG]

    Edit:
    It appears the idea of the rotational nonlinear springs came from this paper:
    https://www.sciencedirect.com/science/article/pii/S0143974X02000214
     
    Last edited: Sep 9, 2019
  34. Mick West

    Mick West Administrator Staff Member

    So they say. However the word "nonlinear" is not used in reference to anything from chapter 4 onwards. The static analysis diagrams say "linear". The dynamic analysis diagrams don't specify, but since they don't actually look anything like a dynamic analysis using the full connection model, it's not the important point.
     
  35. Miss VocalCord

    Miss VocalCord Active Member

    A&E created this video where they compare the UAF, NIST and real life video of the building collapsing:

    Source: https://www.youtube.com/watch?v=Xd7tqpwdlpQ&feature=youtu.be&t=44


    In your video the UAF building seems to start slightly early in respect to the real video; however it also seems to fall much faster as it does in the A&E video. Is this just video frame rates not matching or something else going on?
     
  36. Mick West

    Mick West Administrator Staff Member

    I don't see the difference you are referring to. Can you maybe give timestamps of start and end of collapse, and see how much they differ?
     
  37. Miss VocalCord

    Miss VocalCord Active Member

    This is your video; the left shows the start. Both the NIST and UAF simulation have already started falling down. (red line)
    At the last point in the comparison the UAF model is far further down as the real video. (red lines)

    When looking at the A&E comparison between the UAF and real video they seem to have fallen down a more equal amount of height.
     
  38. Mick West

    Mick West Administrator Staff Member

    I think that's simply because I synced the videos on the start of the collapse of the penthouse (the first visible deformation of the top of the penthouse).

    The NIST model starts to descend at around the same time as reality. The Hulsey model is about one second ahead at this point.

    Significantly, the timing of the NIST model just arises from the simulation, whereas the timing in Hulsey's model was done manually, as they claim there's no connection between the penthouse collapse and the global collapse. So they would have to pick three points in time to inflict the mystery damage. 1) Columns under the East Penthouse, but high up. 2) interior columns (maybe just under the mid and west roof area). 3) all remaining columns.

    i.e. their timing is fudged to match observations. NIST's timing matches observations based on a single initial state.
     
    • Agree Agree x 1
  39. Miss VocalCord

    Miss VocalCord Active Member

    That makes sense.

    Too bad we have to deal with too little details from Hulsey right now. In the paper they claim their simulation follows Chandler's speed curve very closely:
    I'm wondering if they matched up the full simulation (including the east penthouse) or they only did it for the following collapse. Too bad they haven't released their own full side by side comparisons yet.
     
  40. benthamitemetric

    benthamitemetric Active Member

    I agree with your general sentiment about the significance of peer review; however, note that, for what it's worth, the NIST WTC7 report was peer reviewed by, and republished in, the Journal of Structural Engineering, which is the flagship structural engineering journal of ASCE. The published paper is of course highly truncated compared to the full public NIST report on WTC7, but it incorporates that whole report by reference while focusing on summarizing its key methodologies.
     
    • Agree Agree x 1