1. Mick West

    Mick West Administrator Staff Member

    Not really, it's just a more powerful server computer. The headlines and descriptions are rather misleading as to its actually capabilities. It can't actually "manage" 160 petabytes of data in 250 ns, it can just randomly access any part of it in 250ns. And it's not likely to be ready for a few years.

    More powerful computers help, but only to a degree. If you make the model from 2x the number of segments, then you get 4x the number of possible interactions.

    However you could certainly throw far more computing power at it than was used by NIST (I think they just used a cluster of 10 fairly regular PCs). It's just rather expensive - and again, there's no real need, as there are too many unknowns, so all you can do anyway is verify the general principles.
     
    • Informative Informative x 3
  2. Mick West

    Mick West Administrator Staff Member

    And a practical illustration of this.
    [​IMG]

    The can is holding 30 lb, it can easily hold 100lb, or 150lb if very carefully added. I take a 2.5 lb weight and drop it from about three feet up. The can fails.

    [​IMG]

    So, we have just 2% of the maximum static load causing failure, because it's a dynamic load. And this is just a three foot drop.
     
    • Like Like x 3
  3. Cube Radio

    Cube Radio Banned Banned

    It's just a three foot drop but it's onto just one level, right? Aren't you saying that if you built a tower like this to just five levels the lowest can would fail under static load?
     
  4. Mick West

    Mick West Administrator Staff Member

    Yes, but we were talking about the strength of the floor connections here, not the supporting columns.

    And by "floor" here, I mean the open span area between the inner and outer columns. The "Angle Clip" in this simplified diagram:
    [​IMG]
    [​IMG]

    Take five (numbers are rough guesstimates) floors, and load them onto one floor, see how it does. Then take one floor, drop it from 12 feet onto another. How do those clips hold up?
     
    • Agree Agree x 1
  5. Mick West

    Mick West Administrator Staff Member

    The columns mostly failed by coming apart at the seams, not by being crushed. So in my simplified Jenga model the individual sections of column are essentially infinitely strong.

    [​IMG]

    The floor sections fail, the complex wave of rubble pushes sideways and presses down on the columns, which have no lateral support, so they come apart at the seams.
     
    • Useful Useful x 2
  6. Mick West

    Mick West Administrator Staff Member

    Would you agree that my Jenga model essentially replicates this? It's only a small section, yet you can see how the collapse would not be arrested.
     
    • Agree Agree x 1
  7. Jason

    Jason Senior Member

    I'm sure there are computers out there that could simulate the events that unfolded on 9/11 without question, but putting in the time and money it would cost programmers to write the software would probably be very expensive. Just curious, but wouldn't it make sense for the Sky Scraper community to invest money to come up with programs that could simulate airplane strikes and bomb threats at any floor for future builders. I'm sure insurance companies who insure these buildings are probably wanting their hands on that technology since it could outline weaknesses in bldgs they insure. If we have computers that can go back in time to the moment of singularity in the big bang, and work forward through simulations to see all the stars and galaxies we have today, we should be able to come up with programs to run a simulation of plane crashing into a bldg, NO?
     
  8. Mick West

    Mick West Administrator Staff Member

    I'm sure they do, and I'm sure the new Freedom Tower had all kind of simulations run on it. However the safety of building under such events really more of a function of meeting the building codes and recommendations that arose after 9/11. Simulations of disasters are probably not things they want to talk about.

    http://enr.construction.com/news/buildings/archives/050629.asp


    Wind tunnel:
    [​IMG]
     
    • Informative Informative x 1
  9. Mick West

    Mick West Administrator Staff Member

    I tried to make a taller model, and used some scrap strips of wood which I cut for length.

    [​IMG]

    However the tape "seats" did not work very well. It was very fiddly, so I ended up using too much tape, making the connections too strong, and then the floors would rotate instead of falling - shedding the load forward.

    [​IMG]

    Still, interesting experiment. I learned just how crucial the connection failures are to the whole thing - both their strength, and the elasticity. I feel very confident that I could make a model demonstrating collapse progression - however it would be very fiddly to set up.

    I think perhaps magnets might work in terms of strength and re-usability. But fail in other ways.
     
    • Agree Agree x 1
  10. Jason

    Jason Senior Member

    Nice work though. Way to put in the effort..
     
  11. deirdre

    deirdre Moderator Staff Member

    so those are giant jenga blocks? cant you put the support blocks 'horizontal' for more stability? the floors will collapse without the supports collapsing too (for more weight) I'm sure, if that's what youre trying to show.

    and you can move it away from the wall too, which has to be messing you up
     
  12. Mick West

    Mick West Administrator Staff Member

    They are normal Jenga blocks.

    I can't use them horizontal, as I want to demonstrate hinge buckling.

    I can't move it away from the wall as it needs to be somewhat constrained to a plane.
     
    • Like Like x 1
  13. Cube Radio

    Cube Radio Banned Banned

    [​IMG]

    Not to distract the thread from Mick's physical modelling efforts, but here is the collapse simulator created by femr2 that I mentioned yesterday.

    http://femr2.ucoz.com
     
    • Funny Funny x 1
  14. Keith Beachy

    Keith Beachy Active Member

    Why would you model the collapse, when once started it goes to the floor, as I know, and as the chief structual engineer knows.

    After doing the calculations for each floor, we find that 12 floors it the breaking point. You sum up the connections of the floor to the core and shell, it is math, a math model of why the collapse continues. If you wish to do a model fine, but the model is what a floor in the WTC can hold. It is that simple, and what we saw on 911.

    Where do you download the model? Who is the structural engineer who created the model, what journal is it in?

    It does not look like the collapse, but then models never are like the real world. The real model comes down to what a floor can hold. How many connections are there? What can they hold?

    This is math that can be checked, things you can estimate to see if NIST got close. This is why the WTC collapsed as seen, two full up models of the WTC collapsed meeting your goal of science, full up models cost way too much to use, but on 911, because terrorists did it, we have real models collapsing, no estimated models of math, real steel. Reality beats models everyday. That is an engineering fact.
     
    • Like Like x 1
    • Dislike Dislike x 1
  15. Mick West

    Mick West Administrator Staff Member

    Did femr2 reach any conclusions from his simulation? It seems like it has not been updated for five years.

    Here's his spreadsheet attached, for posterity. It seems like it's using some simple assumptions about conservation of momentum and/or energy. Probably a topic for another thread if anyone is interested.
     

    Attached Files:

  16. Jason

    Jason Senior Member

    I think he did it that way to try an correctly model the weight differential between the columns and floors.
     
  17. Cube Radio

    Cube Radio Banned Banned

    Keith, the collapses were not modelled events. They're the events that need to be modelled. I don't think that can be too hard to grasp. You're begging the question and appealing to authority. Those are two fallacies and they do not address the fact that no successful model of the event has been made.
     
  18. Jason

    Jason Senior Member

    OT; but pertinent. Why when researching material for the core columns and lay out, many CT sites or "research" for CT's constantly note that they are using a "best estimate" for the layout of the columns, and mass of the columns in relation to their decreases in mass and as you make your way up the core. They always state that there is little information regarding this area of the towers and the architectural designs were lost in the fire at the Port Authority Bldg when tower 7 came down... Is there any truth to this?
     
  19. Mick West

    Mick West Administrator Staff Member

    It was more to do with what materials I had to hand :)
     
    • Funny Funny x 1
  20. Cube Radio

    Cube Radio Banned Banned

    Jason, I have also seen this kind of complaint a lot. Psikeyhackr, who created the paper loop model posted upthread, has often claimed on the Center for Inquiry forums that information regarding the precise distribution of steel and concrete through the towers is not available to researchers and is not in the public domain, a fact he seems particularly disgusted by as he observes that this information would be basic to any modelling effort.
     
  21. Keith Beachy

    Keith Beachy Active Member

    No, you don't have to model what really happened, it is a waste of time. After the collapse begins, it does not stop, and this is based on the way the WTC is built. I understand that, NIST does, and it is understood by the chief structural engineer who built the WTC. Thus, making a model of what really happened is a waste of time.

    The collapses were real gravity collapses. No, I am not appealing to authority, I am exposing the fact, a floor can hold 29,000,000 pounds and the part above the impact zone exceeds what a floor can hold. This is not authority, it is called math.

    The chief structural engineer is not an appeal to authority, he building the WTC towers, and knows off the top of his head so many floors falling on one floor starts a collapse. Thus, modeling the collapse is a waste of time. NIST can't trust the chief structural engineer, so they calculated what a floor can hold, and found the chief structural engineer said is true, collapse continues with 12 floors of debris, even if it is small particles, the mass is too much for the floor, even if it was sand. I checked the numbers, you can too. Bingo, the floor can only hold 29,000,000 pounds - on 911 this happened, and the collapse then continues, because we have n+1 floors each impact.

    What I have done in this post is model the WTC collapse as a floor failure. Why this is important is, the WTC is a system, the floor fails and the system is broke. You call math an appeal to authority, and that is a mistake.

    You have this backwards. The gravity collapse of the WTC towers, are full up models, you can't do better. Thus a study of the collapse was seen on 911, twice. Sorry, but the real thing saves having to do models.

    You see a 767 fly, and you want to make a paper airplane to verify lift works. What is the purpose to model something we know is verified by two full up models, and is understood by the chief structural engineer for the WTC. He agrees with me, no model is needed to understand the collapse, if you know the WTC as a system of floors, shell, and core. The core can't stand without the floors and shell, and the same for each part. It is cool when the number one expert on the WTC in the world can guess the numbers of floors which would make the collapse continue to the ground, he said 14, NIST calculated 12. What did the real collapse have. 12 extra floors of mass, carefully placed on a WTC floor causes the collapse, On 911 the floors were moving, my simple model just got more likely to collapse due to movement, not an appeal to authority, it is math and physics I am using to make a point which is missed. Are you claiming the floor will not fail? Are you saying you did not see the WTC collapse due to gravity? Why do you need to go backwards and prove something with a lesser model than the real thing - it is not logical. Why do you make up an appeal to authority? Can you prove a floor will not fail with over 29,000,000 pounds added, and can you explain why it would stop?

    You can't give one reason for a model that makes engineering sense. I am not appealing to authority, it is math, one floor, finished. But go ahead, call understanding, math, physics and engineering an appeal to authority, it beats having to supply a goal, or purpose for needing model for something we saw happen, for real, the gravity collapse of the WTC. I think AE911T has all those experts, they can do the model and declare thermite did it, or some other failed CT. What is the purpose, and goals, based on engineering needs? Using some engineering stink, state the purpose please.

    Do you understand what a floor can hold? This is not an appeal to authority, yet engineers did not need a model of the collapse to know it was collapsing, and I agree as an engineer. This is not authority, is math, we understand why. I don't think being an engineer matters, it is math, 29,000,000 pounds. That is the model, prove my model wrong. Which weighs more a pound of WTC, or a pound of WTC dust.
     
    • Like Like x 2
    • Dislike Dislike x 1
  22. Keith Beachy

    Keith Beachy Active Member

    Anyone can estimate the information Psikeyhackr wants; guess he missed the thousands of pages NIST has. His model failed to model the WTC collapse, the full up models prove that.
     
    • Agree Agree x 1
  23. jaydeehess

    jaydeehess Senior Member

    @Jason,
    I believe you were asking for load numbers.
    From NIST FAQ. http://www.nist.gov/el/disasterstudies/wtc/faqs_wtctowers.cfm
    So, as far as modeling goes, as Keith mentions above, why bother? You could model for floor collapses but all you can do is prove an obvious point, to whit; a structure that depends on lateral bracing between columns will completely fail in any scenario in which those lateral braces are caused to progressively fail.

    What is supposedly wanted here though, is a model that shows this is possible for these specific structures.
    OK design a 3d (as opposed to Mick's 2d model) for which the floor/column connections can support up to 12 times the load of a typical floor. But wait, there's more: you need some way to model the actual floors themselves. So first you make a model with floor/column connections that are essentially infinitely stiff(will not fail). Now determine the load at which the floor itself will fail, divide by 10 for safety factor.
    Now design your floor to column connections to support 12 of your previously determined floor loads.

    As for columns: they should be designed, for a first model, as essentially unavailable, able to withstand say 150 times the load of the model in the vertical, and very stiff in the lateral. Now the only thing the floors are doing is supporting the vertical orientation of the columns, there will be no column buckling. If lateral support is lost they will fall over, but they won't buckle.

    Now , from a miniscule height above the top floor, drop 12 times the determined individual floor load. In this way, dynamic loading on that first floor is kept extremely low.
    Our design expects that first floor to fail, its static loading has exceeded its design max.
    Does the floor collapse arrest at any point?

    If not then there is no reason to model anything further.( personally this is as far as you'll need to go IMO). Note that subsequent floors below the top one will see dynamic loading. It will be more spread out over time as the distance between the components of your drop mass will be affected by the transfer of momentum into that top floor, the differential in time when floor supports fail, floor tilt, etc.

    If it does arrest, try again with the extra loading coming from 50% or 75% of the floor separation above the top floor. Does the extra, initial , dynamic loading disallow arrest?

    Still arrests? Continue with new models , increasing the height of of drop with each subsequent test up to twice the floor separation distance.

    Of course the sticky part is that gravity doesn't scale. Dynamic load is directly proportional to velocity but velocity is proportional to the square of the drop height.

    However I am pretty sure that you will see no collapse arrest with essentially zero dynamic load, certainly not if you have any appreciable dynamic load on initiation.

    You may wish to further 'play' with the model.
     
    Last edited: Jun 19, 2014
    • Like Like x 1
  24. deirdre

    deirdre Moderator Staff Member

    heres my little model:
    playing cards. scotch tape hinges. 1 piece per corner about 1 centimeter 'hinge'.
    1 tablespoon "chalk line" chalk per floor. loose on open floors
    for 10 floor 'stack' I drop, i sprayed cards with water, added tbsp chalk and squished together.
    two dowels on each side simple to hold cards upright (or it folded like an accordion)
    left side dowel taped to table leg to prevent falling at freestand.

    *top dowel section, on right collapsed without the floor supports.
    *bulk of powder/dust remained in the building (not to mention I didn't have a front of back wall)
    *floors collapse even without adding weight of side or center columns.

     

    Attached Files:

    • Like Like x 4
    • Winner Winner x 1
  25. jaydeehess

    jaydeehess Senior Member

    His paper loop model most definitely does not model the scenario described in the quote from NIST I posted above.
     
  26. Jason

    Jason Senior Member

    Knowing how precise you always try to be gave me the impression that you were trying to correctly scale it down. I guess I was wrong
     
  27. Mick West

    Mick West Administrator Staff Member

    It was more trying to illustrate the way in which the towers collapsed - i.e. not from columns being crushed, but from the floors failing.

    I could build a more accurate model, but I think my time might be better spent doing a re-usable computer version in JavaScript.
     
    • Agree Agree x 3
  28. jaydeehess

    jaydeehess Senior Member

    First off, as I said above, I have seen no one, who questions that the collapses were brought about by impact and fire damage, bother to model the scenario actually described by NIST and others. Instead I have seen models such as Psikeyhackers which models column buckling due to vertical over loading.
    The information on the strength of columns is irrelevant in a progressive floor collapse scenario. As I said in describing a possible model, you could use columns in your model that are 10,000 times strong get in their axial load capacity than is the mass of your entire model. If the floors are holding them vertically, ie.providing lateral support, they will fall over if the floors fail.

    Secondly, why does a model more faithful to what-actually-happened, progressive floor destruction, need to be modeled? The only reason would nbe to investigate possible mechanisms of collapse arrest. Given that falling mass will increase, and more importantly velocity of that mass will increase, its very unlikely that mass ejection will ever be significant enough to reduce floor loading to less than it was in n the first failing floor.
     
    • Like Like x 1
  29. jaydeehess

    jaydeehess Senior Member

    YES! I'm not sure cube radio has acknowledged that this is the scenario of the so called "official story".

    Quite definitely, if that's within your wheelhouse.
    Of course the reason why fea is used is precisely due to its reusability. That is to say that one can run various slightly different models without having to invest the time, effort and materials cost, to build physical models. In addition, computer modeling allows full scale rather than attempting a scale down.
    Because of limitations in computing power and uncertainties of variables, one needs to limit some things, and approximate others.

    For instance, if progressive floor failure is being investigated you might as well model full tower height columns rather than bother with column sections. You could even give the columns infinite axial and lateral strength
     
    • Like Like x 1
  30. Jason

    Jason Senior Member

  31. Mick West

    Mick West Administrator Staff Member

    I think a good reason for this would be as an education tool. I think a lot of the "truthers" don't really understand the actual mechanism of collapse - hence their focus on column crushing. If you could show what actually happened in a nice visual way, then it might help some of them move on.

    The purpose of my Jenga model was mostly illustrative, rather than investigative. However it did help investigate a number of issues with model building - particularly the difficulty of getting the connections right.
     
    • Agree Agree x 2
  32. Cube Radio

    Cube Radio Banned Banned

    I can't speak for Psikey, but it is the approximation of information rather than the actual information that he observes he is obliged to use. He also notes that he has burned the entire of the NIST report to DVD and searched it for the information he wants with respect to the distribution of steel and concrete in the towers; it is not there.

    This is the last time I'm going to point out that the collapse of the towers were not models of the collapse of the towers: they are the events that need to be modelled. Nobody else on this thread appears to support the absurd notion that they were what you bizarrely call "full up models"; if they do, I will address myself to them.
     
  33. jaydeehess

    jaydeehess Senior Member

    Probably not there because its not relevant , especially below the impact/fire floor levels, to the cause or mechanism of collapse.
     
  34. deirdre

    deirdre Moderator Staff Member

    I think they are actual physical models of how tube in tube towers collapse if a plane flies into them. Because they are models of how those types of buildings collapse under those conditions.
     
  35. Cube Radio

    Cube Radio Banned Banned

    No, they're the events that need to be modelled.
     
  36. deirdre

    deirdre Moderator Staff Member

    to me they have been modeled by people I consider experts. But I understand what you are saying and you don't agree; you need a scaled down physical model.
     
  37. Cube Radio

    Cube Radio Banned Banned

    Or a credible computer model, yes. Again, you can't take events as models of themselves.
     
  38. Cube Radio

    Cube Radio Banned Banned

    Or a credible computer model, yes. Again, you can't take events as models of themselves.
     
  39. Pete Tar

    Pete Tar Moderator Staff Member

    Is the definition of model being that you can re-use or rebuild it, and have focused observation on specific parts that weren't available to close observation in the real event?
    If model simply means 'replication' then obviously the real things aren't replicas of themselves, but they are 'models' of how the forces involved will act in a particular structure.
     
    • Like Like x 1
    • Agree Agree x 1
  40. jaydeehess

    jaydeehess Senior Member

    Given that there are two of them, they are examples of the susceptibility of open office space, tube-in-tube design to lateral support loss.