1. Jedo

    Jedo Member

    Isn't it also a serious question how free fall acceleration could occur in a controlled demolition collapse? I am just asking this since most videos of controlled demolitions I have seen are slower than a free fall. This is quite plausible as in the controlled demolition you want to use the potential energy not be 100% converted into kinetic energy (100% conversion is equal to free fall) but use a considerable ratio of the kinetic energy to smash part of the structural elements that support the building.

    First I would like to emphasize that I understood that not the whole building descended at free fall and also not all of the descend was at free fall acceleration. The measurement of free fall is just verified for some distinct points of the facade, and only for some intermediate time after the start of descend. I hope we can agree on that.

    I can understand your doubts. But the question is how to verify if they are justified or not. This is not easy.

    How much time do you think is reasonable for the interior buckling or collapse to progress westward? What would you set as a maximal speed of progression?

    Maybe you want to have a look at these rack collapses, which IMO collapse quite rapidly:




    You see in the videos a collapse initiating event, followed by some intermediate transient phase, followed by rapid progression to global collapse. Once the support is lost at a local level, the structure shifts loads to the remaining columns, which can't support these loads and collapse very quickly.
     
    Last edited: Feb 14, 2018
    • Informative Informative x 1
    • Useful Useful x 1
  2. deirdre

    deirdre Moderator Staff Member

    let's not encourage John to go off topic. or make statements without providing backup evidence.
     
    • Agree Agree x 1
  3. John85

    John85 Member

    Then all I'll say @deirdre is that I agree. I'm glad you're making such comparisons, @Jedo, and asking that question.
     
    • Like Like x 1
  4. Jedo

    Jedo Member

    And what do you think about the horizontal progression of the collapse after seeing the examples of the rack collapses?
     
  5. John85

    John85 Member

    They fold like a house of cards. It's a good thing high rises aren't designed like that, or the falling debris would have taken out WTC 7 when there could have been people in the building, or the airplane impacts in 1 & 2 would have done the same.
     
  6. Jedo

    Jedo Member

    I also think it is a good thing that high-rise buildings are not designed like a house of cards, but neither so are industrial storage racks. There are many casualties each year from rack collapses, so design for safety needs a proper analysis of stability limits and possible failure modes. However, when you look closely at research papers studying these kind of failures, you may find many similarities (I guess) to the studies of building safety.

    I did just a quick web search and came up in five minutes with these examples. For one, lets start with A Case Study on the Collapse of Industrial Storage Racks, James P. Plantes ; Deepak Ahuja, P.E., M.ASCE ; and Ryan T. Chancey, Ph.D., P.E., M.ASCE Forensic Engineering 2012 : Gateway to a Safer Tomorrow . 2012 (you can google for the full pdf)
    [my emphasis]
    Or, let's go to the next paper: Collapse of a high storage rack,Ch. Affolter, G. Piskoty, L. Wullschleger, B. Weisse, Engineering Failure Analysis,Volume 16, Issue 6, 2009,
    and
    Or this one:
    Methods of restraining progressive collapse in rack structures, A.L.Y. Ng, R.G. Beale, M.H.R. Godley, Engineering Structures, Volume 31, Issue 7, 2009,
    That sounds all similar to me to the things we are discussing here for the collapse of WTC7, don't you agree?
     
    • Informative Informative x 2
    • Useful Useful x 1
  7. John85

    John85 Member

    I would say that there are several relevant differences: WTC 7's columns were braced and therefore not designed to resist only compression. The building was not likely overloaded, and having been evacuated, would probably have been 'underloaded'. Poor design I doubt, given the age of the building. Removing one column would not cause progressive collapse, and indeed when several core columns were supposedly cut by impacting debris, the building remained standing.
     
    • Disagree Disagree x 1
  8. Mick West

    Mick West Administrator Staff Member

    The reason columns are braced is largely so they can resist more compression. C79 failed because it lost that bracing, and the compressive load led to buckling. Other columns buckled from a combination of factors - increased load transferred to them, some floor collapses leading to loss of bracing, and the dynamic effects of nearby floors and columns collapsing which pushed and pulled on them.

    I think the live load of people (maybe 500 tons spread through 100,000 ton building) is pretty much irrelevant here. Lateral progressive collapse happened in part because individual columns became overloaded.
     
    Last edited: Feb 15, 2018
    • Agree Agree x 1
  9. John85

    John85 Member

    I see what you're saying, but I don't think that analysis will withstand scrutiny in Hulsey's study.
     
    • Disagree Disagree x 1
  10. Jedo

    Jedo Member

    First sentence of the third paper I cited:
    The are going on in describing standard racking systems:
    Here some picture of an industrial rack containing braces (15 seconds internet search): [EDIT: image link was broken after inserting; added other images from Wikipedia]
    [​IMG]
    [​IMG]


    Well, that depends on the column. In several of NIST's simulations they could achieve a global collapse of WTC7 by removing just column 79 only on two stories. (Btw, they used this as their minimal controlled demolition scenario for checking window breakage and sound levels of the charges would make for just removing the column 79 on two stories, all incompatible with the observed facts.)

    Column 79 was apparently the Achilles heel of the building. Probably columns 80 and 81 were also, but they were not tested afaik.
     
    Last edited: Feb 15, 2018
  11. Oystein

    Oystein Active Member

    Today, I installed two "heavy load racks" in my basement: 160 cm wide, 180 cm high, 60 cm deep, each of four levels good for 200 kg of (ideally well distributed) loads. The long 160 cm beams were really flimsy and easily bent, and the uprights wobbly, all gaining some precarious stability with 2 lateral bridges on each level. The whole stability came together only with the addition of the plywood floors (I haven't measured - 3 to 5 mm I'd guess), which themselves bend easily under gravity if unsupported.

    Only the integrity of all components and connections makes for a strong and stable rack.

    Remove corner connections, buckle an upright, or burn a floor, and all bets are off.
     
  12. Jedo

    Jedo Member

    In case you really do the experiment, please post the video
     
    • Funny Funny x 1
  13. Mick West

    Mick West Administrator Staff Member

    This type of thing?
    Metabunk 2018-02-15 13-22-50.

    It's easy to visualize how the remove of a couple of the shelves would lead to loss of stability. Might make an entertaining (but perhaps dangerous) demonstration to build it without those shelves and then load it to the design specification.
     
  14. Jedo

    Jedo Member

    No, make a fire on the lowest shelve! :D:D [where is the arsonist smiley?? How is that not part of the standard set???]

    It would be best if the shelf is metal itself, and you get the thing collapse just by the expansion of the heated shelf. Perhaps the second shelf would be even better, as when the third shelf fails before it can press the columns outward to buckle, it might at least by falling down cause a collapse.
     
  15. Jedo

    Jedo Member

    Ahh, someone already did it:


    Edit: Notice the onset of buckling around 2:50…
     
    Last edited: Feb 15, 2018
    • Like Like x 1
  16. John85

    John85 Member

    We look at the same video and see different things. I notice the supposed 900 degrees C temperature, small diameter metal supports and partial, gradual collapse. WTC 7 members were larger diameter and subjected to lower temperatures.
     
    • Disagree Disagree x 1
  17. Jedo

    Jedo Member

    I have noticed these things. But take care: Do not confuse the gas temperatures in the WTC7 fires, that probably had over 1000'°C, with the temperatures the structural elements got, which only reached 600°

    Also, shouldn't the comparison of larger diameters be done in comparison to the total lengths? 2cm (or whatever the rack has) might seems small compared to the 14in ~ 36cm (or how thick were the columns? I am now too lazy to look it up, but here we talk about orders of magnitude only, anyway), but so are the 2m of the shelves (or maybe 3m?) compared to the 15 foot beams etc.

    Moreover, we have fire for 10 minutes in the video, not for hours as for the WTC7.
     
  18. Mick West

    Mick West Administrator Staff Member

    Let's not drift too far off topic please. Feel free to chat on PM.
     
  19. Oystein

    Oystein Active Member

    Exactly.
     
  20. Jedo

    Jedo Member

    Maybe one last remark, partly to correct myself, partly because it can be relevant also for the discussion here and elsewhere.

    I have already thought about the idea of building a miniature WTC7, like 1:100, for experiments. Would be a big effort, of course, but surely fun. However, the problem of scaling is not easily addressed, and maybe unsolvable. You have to account for all kinds of physical laws, and not all go the same way when you miniaturize.

    For instance, assume we make all length dimensions a factor µ smaller.,say µ = 0.01, to have a model of WTC7 which is about 2m high. Critical shear forces then are a factor µ² smaller (as the shear has units N/m²). Makes sense, as you need about some ten kN to bend a real sized I-beam, say 40cm thick, while you need only a few N to bend a 4mm steal beam. This is all in accordance with Mick's experiments.

    However, the problem comes in with gravity, which you can't rescale. Of course, you could increase the loads by some proportionality factor, but that doesn't account for the gravitational forces that act on the building elements because of their self-weight. The mass of a beam scales with the cube µ³, and the total force due to self-weight is mg, So by miniaturizing, the gravitational self-weight forces are a factor µ = 0.01 too small! To check if this makes sense, we can figure it the other way round, using some common sense: A steal beam of 4mm diameter and 1 or 2m length is held on one end horizontally; it will bend a little, but sure still in the elastic regime. A steam beam of 40cm diameter and 100m length is unable to support its own weight horizontally without additional support columns, and it will surely yield inelastically. So to accurately model a horizontal beam you would have to make it 100 times heavier.

    [On the other hand, there is the interesting possibility that structural collapse is a universal phenomenon (in the sense of statistical mechanics phase transitions), leading to scale invariance close to criticality, meaning the behaviour of the system is independent from the details of it. However, I will not go into that as probably only theoretical physicists could follow ;) ]

    The take away is that when scaling things down, the self-gravitational effects do not scale correctly.
     
  21. cloudspotter

    cloudspotter Senior Member

    don't know if you've come across this thread but it was something that Mick was tackling https://www.metabunk.org/towards-a-...the-collapse-of-the-wtc-towers-on-9-11.t7396/
     
    • Like Like x 1
  22. Jedo

    Jedo Member

    • Like Like x 1
  23. Joe Hill

    Joe Hill Member

    Whether you think it possible or not is irrelevant. It is fact that the core/floors system and perimeter frame system collapsed as two separate entities. A simple study of structural remains/debris patterns proves this is what occurred.
    First, note the entire pile is draped with perimeter walls:
    wtc7-debrisoverheadfromnorth.
    B7rubbleaerial1.
    Even lower floors of the west wall have no floors attached:
    SWcornerbulge.
    B7rubblewestwall2.
    Remove the perimeter wall panels, and you can observe the rest of the structure, the interior structure, piled up in the foot print. Look closely, and you can make out the layering of floors:
    B7rubblelayering.
    The only way for perimeter walls to end up draped on top of the pile is for the interior to hit the deck first, and the walls fall on top.

    The real question, and a good one it is, should be, "How/why did floor to perimeter connections fail so easily as to not noticeably disturb the perimeter walls?"
     
  24. Jeffrey Orling

    Jeffrey Orling Active Member

    If the portions of the facade ended up on the top of the "debris pile" ... the pile has likely material from inside the facade as well as some of the facade... bits on top was the last to collapse.