A virtual model illustrating some aspects of the collapse of the WTC Towers

Cube Radio

Member
View attachment 26280

There's very little actual discussion going on over there. But I'm curious as to why they feature my simplistic illustrative models, and yet ignore Kai Kostack's much more impressive model from six years ago.
Source: https://www.youtube.com/watch?v=4_J7ak_IZXk
As I've mentioned before I've seen this same model used by the likes of AE911 to support their arguments. In my own opinion it's too "low" a tower in terms of stories to offer any real insight into the question of how the collapses could progress so rapidly and steadily through so many floors, despite very different initiation points.
I've no interest in splitting the discussion over two places. @Cube Radio, you are, I think, a reasonable proxy for the "collapses like that are physically impossible" crowd.
Thank you. But @aka is infinitely more suitable as should be wholly obvious -- I don't have the kind of interest that would support a whole subreddit dedicated entirely to just this question.

However I don't expect @aka to conduct discussion here because of the way his posts were moderated last time he was around, so I think that keeping discussion in the sub is much more sensible.

As you will have seen if you've visited the sub, though, he's very grateful for the modelling effort you've put in (as am I).

You are also, I think, not the type to fall silent when you finally realize that the towers might actually have collapsed by the floors being stripped from the columns which then failed in part due to slender column buckling at the joints.
I'm waiting for the global collapse of a taller tower model before I cough up the $100, at your suggestion.

In general though, I have a young family and a demanding job among other things, and these questions don't obsess me -- much as I have a real interest in them. So there will always be times when I don't check in here for extended periods. I get a bit tired of smug/ad hominem remarks typified by NoParty's and Jeffrey's comments upthread, and then when I'm banned for the nth time I tend to lose interest in Metabunk for long after my ban expires.

But don't worry -- I'm not going anywhere. This whole discussion may proceed more slowly than others for whatever reason, but I'm not leaving it!
 

Mick West

Administrator
Staff member
and these questions don't obsess me
Me neither. I find them interesting largely because I think they are essentially resolved and I'm interested in the challenge of explaining them. They certainly don't obsess the vast majority of structural engineers for the same reasons that the Flat Earth theory does not obsess the vast majority of geodetic surveyors. Hence the effort put into investigating and explaining is left more to people like myself - people interested in scientific skepticism and debunking false claims.

And I don't really spend that much time on 9/11. I have more fun things to do, like debunking the Chilean Navy. Which raises another topic we've discussed. Why have I come up with functional models (both physical and virtual), but AE911 still relies on cardboard boxes?
 

deirdre

Moderator
Staff member
I'm interested in the challenge of explaining them
I think you've explained everything quite well. And then Plasco shows even more nicely how collapse occurs.

If people simply don't want to believe the science and math, (and want to believe Iran staged Plasco to help out the US Gov. ) there isn't really anything you can do.
 

Mick West

Administrator
Staff member
I think you've explained everything quite well. And then Plasco shows even more nicely how collapse occurs.

If people simply don't want to believe the science and math, (and want to believe Iran staged Plasco to help out the US Gov. ) there isn't really anything you can do.
Well it's not like there's just one type of Truther. There's a strong small core that will likely be unconvinced. But in really talking to the more reasonable members

If the core loses support will it collapse?
 

deirdre

Moderator
Staff member
Well it's not like there's just one type of Truther.
True. I guess I'm mostly confused about why anyone would believe a virtual demonstration over a physical demonstration? Computer graphics can be programmed to do anything.. unlike in real life.
 

Mick West

Administrator
Staff member
True. I guess I'm mostly confused about why anyone would believe a virtual demonstration over a physical demonstration? Computer graphics can be programmed to do anything.. unlike in real life.
It's an open-source model they can experiment with themselves rather than just a video. I'd hope some Truthers do experiment with it.
 

Jeffrey Orling

Senior Member
The runaway collapse of the twin tower design only required a few floor masses free and falling to kick off the complete destruction. However consider that if say the 12-15 floors collapsed the towers might not arrest but the colunmns down there were so stiff that their buckling was less likely and the tower may have stood on "stilts".

The total destruction from a ROOSD had to lead to facade column peel and buckling... and that happened where the columns were not as stiff as down below.

This is a guess but an interesting thing to study.
 

Cube Radio

Member
Computer graphics can be programmed to do anything.. unlike in real life.
That's exactly why nobody should accept the (poor, highly approximate and incredibly brief) computer graphics offered by NIST as the sole and complete evidence for its "explanation" of the collapse of WTC7 unless they're given access to all its data -- data NIST refuses to supply for genuine, independent peer review for no good reason.

No genuine sceptic should accept that ridiculous state of affairs: Mick understands that his modelling effort must be kept open source for exactly this reason.

Another important way in which an open source computer model can be said to be superior to a physical model is that the computer model can be made to account for scale and the square cube law, which is quite impossible using small scale physical experiments.
 

deirdre

Moderator
Staff member
That's exactly why nobody should accept the (poor, highly approximate and incredibly brief) computer graphics offered by NIST as the sole and complete evidence for its "explanation" of the collapse of WTC7
I may be misremembering but the wtc graphics weren't proof, they were just visual demonstrations of what occurred. no?

the computer model can be made to account for scale and the square cube law, which is quite impossible using small scale physical experiments
well that's true of course. Although, no offense, I am assuming if any of those architect and engineer Truthers understood the formulas needed in a computer simulation then they would have built one by now. So if they don't even know what the data is supposed to be, then how are they going to evaluate any data?

ps. I also don't like the deflection to wtc7. I've stated before it's silly to assume CD in that building for a multitude of reasons IF you cant first prove CD in the Twins. Which you cant.
 

Jeffrey Orling

Senior Member
The data about the COLLAPSES is all known... that is to say... the actual structural design and loads (weights) WTC7 collapse was a failure... cause to be determined.... low down in the structure... and that failure set off a chain reaction and this undermined all support for the structure above and it simply collapsed.

However we saw that the EPH collapsed ahead of most of the tower... and this tells us that something supporting the EPH (columns) failed first. This also may have been a transfer truss failure (1 or 2)... both of which were what probably led to the chain reaction down between flrs 5&7 causing the whole tower to collapse.

What remains UNKNOWN is what cause tthose structures to fail. NIST claims heat on flr 13 near col 79. It could have been heat down in the transfer truss region failing one of those trusses... We don't have accurate fire data so it's been a sort of educated guess. What is known is that fire and steel don't get along and there WAS fire and it began just after the first plane hit 1WTC.

NIST doesn't know and likely are not hiding anything. What seems to be relevant is the design and how well it was protected from or vulnerable to fire.
 

Cube Radio

Member
I may be misremembering but the wtc graphics weren't proof, they were just visual demonstrations of what occurred. no?
No. The computer animation of the collapse of WTC7 is the sole and complete evidence to support NIST's WTC7 collapse theory. In other words, absolutely no physical evidence from the building was located or examined at any stage of that multi-million dollar investigation.

Apart from the fact that the input data for this model has never been supplied to independent experts for verification, "truthers" take issue with it because it fails entirely to represent or reproduce the period of freefall acceleration that occurred during the destruction of the building, among other things.
I am assuming if any of those architect and engineer Truthers understood the formulas needed in a computer simulation then they would have built one by now.
That's exactly what's happened in the case of WTC7.
I also don't like the deflection to wtc7. I've stated before it's silly to assume CD in that building for a multitude of reasons IF you cant first prove CD in the Twins.
Three buildings, three separate cases. There are no assumptions here: but it's ridiculous to assert that you know exactly why one building collapsed simply because you think you know why a totally different building did or didn't collapse.

But let's not get off topic. Mick has made all his models open source and that is the only legitimate and credible (not to say scientific) way to proceed. I am hoping to pay him $100 for a taller model than he has produced so far, which will also I hope be more representative of the actual towers than the models he has produced so far.

This model is certain to be analysed by truthers, as Mick desires, in the tower challenge subreddit; once this is done, and assuming it collapses straight down at speed (and is reasonably strong before it does so) I will genuinely be delighted to make such a humble contribution in respect of his time and intellectual curiosity.
 

Mick West

Administrator
Staff member
But let's not get off topic.
Agreed.

This model is certain to be analysed by truthers, as Mick desires, in the tower challenge subreddit; once this is done, and assuming it collapses straight down at speed (and is reasonably strong before it does so)
I think an interesting question here is what is "reasonably strong"? How might we verify this in a simplified Blender model?

You might recall I had a few tests of my physical model. The basic criteria for a floor slab was that it could support six times its own weight (sudden application, 12x if gradually applied). So I performed some simple tests along those lines.

In the final large scale test I did I used a wrench, banging it against the floors, and throwing it agains the side of the building with enough force to make the building sway an inch from side to side (a few feet, to scale).

So what would be the required tests for a virtual building?
 

Jeffrey Orling

Senior Member
The NIST animation is not evidence... it is simply a digital simulation of how a collapse might occur given whatever inputs NIST used wherever they were and how long and hot (fires) burned etc. Their sim was only ONE of many possibilities as there was not actual fire data available.

Again the issue is not if the building could totally collapse but where the structural failures were and what caused them. Heat was the only known input, plus some mech damage from falling debris.
 

Cube Radio

Member
what do you mean by this? are you saying that is why they haven't built one by now?
No, I'm saying a new WTC7 model has been built -- by the University of Alaska -- and the research is due to be published soon. I'm sure we'll return to the thread about it here when that happens.

I think an interesting question here is what is "reasonably strong"? How might we verify this in a simplified Blender model?

You might recall I had a few tests of my physical model. The basic criteria for a floor slab was that it could support six times its own weight (sudden application, 12x if gradually applied). So I performed some simple tests along those lines.

In the final large scale test I did I used a wrench, banging it against the floors, and throwing it agains the side of the building with enough force to make the building sway an inch from side to side (a few feet, to scale).

So what would be the required tests for a virtual building?
This is an interesting question, yes, and the quality of the tests are as important as the experiment. Clearly with lateral loads, wind applies pressure simultaneously across and along the entire face of a building -- this is unlike throwing a wrench and a localised impact.

Big Brother will get back to you on this shortly, if that's OK. In the meantime, do you feel the most recent tower fulfils the conditions of @aka's tower challenge, as posted upthread?
 

Mick West

Administrator
Staff member
This is an interesting question, yes, and the quality of the tests are as important as the experiment. Clearly with lateral loads, wind applies pressure simultaneously across and along the entire face of a building -- this is unlike throwing a wrench and a localised impact.
In my physical model it was essentially the same, as the building had very strong sides - much stronger than they would be at that scale - the result was a significant sway in the building.

Blender has wind generators, but my building has no sides. Perhaps just anything that results in a few feet sway int he top of the building.
 

deirdre

Moderator
Staff member
@aka has been following the discussion on reddit's tower challenge sub. You can visit here https://www.reddit.com/r/towerchall...pringtime_metabunkorgs_mick_west_opensources/ but his principal comment on the question of testing the model at this point is this:
His models stand up. Perhaps before quoting your special friend in the future you can get more details. How is your @aka quote at all constructive?

Unless you can answer "HOW are they 'seriously lacking'?" please don't post such things again.
 

Mick West

Administrator
Staff member
@aka has been following the discussion on reddit's tower challenge sub. You can visit here https://www.reddit.com/r/towerchall...pringtime_metabunkorgs_mick_west_opensources/ but his principal comment on the question of testing the model at this point is this:
I start the models with one floor section with offset columns - roughly simulating collapse initiation conditions.

There's a general problem in simulating buildings in going from all the parts in the correct position to all the parts fully loaded. It's like going from zero gravity to full gravity. The solution to this is to ramp up gravity over a few seconds from 0 to g, and let the building settle before applying damage.

Unfortunately with basic Blender I've not seen a way of doing this, and so I had to make the connections perhaps stronger than they would be in order to overcome this initial settling problem.

Another issue is that with very tall buildings and the current simple physics, then impulses tend to get magnified in the lower floors, leading to inaccurate failure.

Using the Bullet Constrains plugin should get around these issues, but I've not really looked into it yet.
 

Cube Radio

Member
His models stand up. Perhaps before quoting your special friend in the future you can get more details. How is your @aka quote at all constructive?

Unless you can answer "HOW are they 'seriously lacking'?" please don't post such things again.
[chit chat removed]

But please answer this question. How do you KNOW Mick's three models stand up under their own weight? Have you tested them? I don't think you have.
 
Last edited by a moderator:

deirdre

Moderator
Staff member
How do you KNOW Mick's three models stand up under their own weight?
that's not what your post says. your post says
one should try and start with making the model stand up; all three virtual models presented by Mick are seriously lacking in that department so far.
I can SEE them standing up.

This is a public forum Cube. If you wish to have a private chat with Mick then do so in Private Messages, otherwise post comments that outside readers can fully understand. cryptic messages are against posting guidelines.

if you are going to quote outside sources, then quote them in full
 
Last edited:

Cube Radio

Member
@aka notes:

 

Mick West

Administrator
Staff member
Added a bit of code to ramp up gravity for initial settling
Code:
#############################################################################
# gravity keyframing as per https://blender.stackexchange.com/questions/7123/add-keyframe-to-a-scene-property

scene = bpy.context.scene
if not scene.animation_data:
	scene.animation_data_create()
if not scene.animation_data.action:
	scene.animation_data.action = bpy.data.actions.new("GravityAction")

fcurve = None
for fcurve in scene.animation_data.action.fcurves:
	if fcurve.data_path == "gravity":
		break
if not fcurve or fcurve.data_path != "gravity":
	fcurve = scene.animation_data.action.fcurves.new("gravity")
# if you dont set the array index to 2 (Z) it defaults to 0 (X) and 
# see: https://docs.blender.org/api/blender_python_api_2_59_0/bpy.types.FCurve.html?highlight=fcurve#bpy.types.FCurve
fcurve.array_index=2;  

keyframe = fcurve.keyframe_points.insert(frame=0.0, value=0)
keyframe = fcurve.keyframe_points.insert(frame=50.0, value=-9.81)
keyframe.interpolation = "LINEAR"

##################################################################################
This seems to work very well in my limited tests. It's ramping gravity up from 0 to 9.81 over about two seconds (50 frames)

Remember the buildings themselves had the benefit of "settling in" over the entire course of construction.

File attached with the code in it, using 1x1 instead of 3x3 for speed, can do 36 assemblies. Sorry for size, but don't have time to figure out why it's so big.
 

Attachments

Cube Radio

Member
Do you happen to remember where and how?
I've been searching in vain -- I apologise. I should have taken note of it because I was struck by the fact that truthers (I'm afraid I can't even say for certain it was actually AE911 now, having scoured their stuff) were attempting to use Kostack's model for their own arguments -- it was in the context of an argument from Newton's Third Law -- it was claimed that you could see in Kostack's model that the upper, falling section was being destroyed at the same rate as it was destroying the structure below. I have to say I think Kostack's model is hardly conclusive of this, but it was definitely being offered as evidence to support the claim.

I'll keep looking and if/when I come across it again I'll be sure to post the link.
 

Mick West

Administrator
Staff member
I've been searching in vain -- I apologise. I should have taken note of it because I was struck by the fact that truthers (I'm afraid I can't even say for certain it was actually AE911 now, having scoured their stuff) were attempting to use Kostack's model for their own arguments -- it was in the context of an argument from Newton's Third Law -- it was claimed that you could see in Kostack's model that the upper, falling section was being destroyed at the same rate as it was destroying the structure below. I have to say I think Kostack's model is hardly conclusive of this, but it was definitely being offered as evidence to support the claim.

I'll keep looking and if/when I come across it again I'll be sure to post the link.
That seems a little odd, because his WTC1 collapse simulation seems pretty similar to the actual collapse.

20170415-071926-vv082.jpg

20170415-071844-g604l.jpg

How do they say it differs? And would that difference be anything that I should look into in my own (much simpler) model?
 

Cube Radio

Member
That seems a little odd, because his WTC1 collapse simulation seems pretty similar to the actual collapse.

View attachment 26403

View attachment 26402

How do they say it differs? And would that difference be anything that I should look into in my own (much simpler) model?
I agree they look pretty similar at initiation but there's one obvious difference: no clouds of dust/debris are obscuring the falling upper section in Kostack's model.

I don't think the argument is that they're significantly different at the point of initiation; the suggestion is that as Kostack's collapse progresses the top section is apparently or allegedly starting to lose structural integrity (in accordance with Newton's Third Law, it would be argued) and therefore would not be able to crush the healthy structure below it all the way to the ground.

I don't see any reason for us to go over this again here, though. As I said I think Kostack's model is far, far too brief in terms of its representation of the collapse progression to be conclusive of anything, and as such I don't see it as being hugely significant in terms of what you're investigating with your modelling effort.
 

Keith Beachy

Senior Member
...
I don't think the argument is that they're significantly different at the point of initiation; the suggestion is that as Kostack's collapse progresses the top section is apparently or allegedly starting to lose structural integrity (in accordance with Newton's Third Law, it would be argued) and therefore would not be able to crush the healthy structure below it all the way to the ground.

I don't see any reason for us to go over this again here, though. As I said I think Kostack's model is far, far too brief in terms of its representation of the collapse progression to be conclusive of anything, and as such I don't see it as being hugely significant in terms of what you're investigating with your modelling effort.
Starting to lose structural integrity does not offer an argument for Newton's Third Law to say it can't crush the healthy structure below. Mass is mass, and a floor in the WTC will fail when loaded past 29,000,000 pounds, even if it is pieces of WTC. 9/11 truth arguments which include some reference to Newton's third law don't make sense, seem to be a trick used to fool people.

The collapse of the towers continues partly due to overwhelming mass overwhelming the connections of the floors to the core and shell. The uniques design which made the WTC strong, was also a weakness for the terrorist event on 9/11. The design was strong, it would stop planes with 10 times the kinetic energy which broke into the ESB.

Good luck with the models
 

qed

Senior Member
Mick has produced an open-source 3D Blender model that stands indefinitely under gravity until an initial collapse event is triggered. After the collapse event is triggered, the building progressively collapses.

Note, that like his physical models, Mick simply implemented an abstraction of the key features of the building, and they all progressively collapse after the initial collapse event is triggered. Mick did not have to do anything "special". When modelled naturally, progressive collapse occurs.

  • The challenge surely now falls upon opponents (@Cube Radio, etc) to suggest/implement changes to the model that can prevent collapse. Tell us how to stop the progressive collapse.
 

Cube Radio

Member
Mick has produced an open-source 3D Blender model that stands indefinitely under gravity until an initial collapse event is triggered. After the collapse event is triggered, the building progressively collapses.

Note, that like his physical models, Mick simply implemented an abstraction of the key features of the building, and they all progressively collapse after the initial collapse event is triggered. Mick did not have to do anything "special". When modelled naturally, progressive collapse occurs.

  • The challenge surely now falls upon opponents (@Cube Radio, etc) to suggest/implement changes to the model that can prevent collapse. Tell us how to stop the progressive collapse.
Source: https://www.reddit.com/r/towerchallenge/comments/63jafo/its_springtime_metabunkorgs_mick_west_opensources/dgc8h77/
 

Ray Von Geezer

Senior Member
Maybe I've got this wrong, but from Mick's comments regarding the need to ramp up gravity it looks like the tower is effectively "built" in zero G, and consequently turning on full 'G' acts like hitting the tower with a huge force?

Can gravity be applied through the "build" process so it's always present? If not, is that due to limitations of the application or the impact on generation time?

Even if not, then (so long as the towers stand when G is reached), isn't this statement - "Don't "ramp up gravity", you don't "ramp up gravity" in the real world either. g is 9.81 m/s², deal with it." (from AKA?) just goalpost moving? The real tower wasn't built in zero G either, effectively he/she is asking for the model to simulate a situation that didn't exist.

Ray Von
 

Cube Radio

Member
Maybe I've got this wrong, but from Mick's comments regarding the need to ramp up gravity it looks like the tower is effectively "built" in zero G, and consequently turning on full 'G' acts like hitting the tower with a huge force?

Can gravity be applied through the "build" process so it's always present? If not, is that due to limitations of the application or the impact on generation time?

Even if not, then (so long as the towers stand when G is reached), isn't this statement - "Don't "ramp up gravity", you don't "ramp up gravity" in the real world either. g is 9.81 m/s², deal with it." (from AKA?) just goalpost moving? The real tower wasn't built in zero G either, effectively he/she is asking for the model to simulate a situation that didn't exist.

Ray Von
@aka explained to Mick how to overcome this issue and "ramp-up" gravity slowly upthread -- which Mick seemed grateful for and implemented.

Also, Mick's towers don't stand up when G is reached, as @aka also explains.
 

Mick West

Administrator
Staff member
Ramping up gravity is a pretty common technique for settling a building.
aka explained to Mick how to overcome this issue and "ramp-up" gravity slowly upthread -- which Mick seemed grateful for and implemented.
I'd used a different method found on Stackexchange. I prefer to do things in script.

I'll be able to spend a bit more time on this later. But I'm sure other people could make the towers stand. The last one did with 102 floor, I think (I'm on my phone right now)
 

deirdre

Moderator
Staff member
Your connections are not "stronger than they would be" - they are way too weak!
I'm confused by this statement. Even before Mick fixed his gravity, when the demos started the lower floors held up. If the connections were too weak we wouldn't see (in the virtual model) a progressive collapse, we would see all the floors collapse at once. But we don't.

The only thing the connections are 'holding up' is the weight of each floor. Is aka saying the connections wont hold up the weight of one floor?
 

benthamitemetric

Senior Member
Is there a way in blender to script in damage over time? It seems to my like Mick's model would stand indefinitely except for the fact that he designed it, from the outset, to have a catastrophic weakness that would induce collapse. It sounds to me like others are taking Mick's model and adding strength such that it stands for a longer period of time before it fails from the designed weakness. If instead of having the catastrophic weakness from the outset there was a way to script it such that the weakness was only implemented later (demonstrating that, absent the weakness, the tower stands), then that should resolve the issue.

As a side note, it is pretty jarring to read that @aka, a person who claims to have devoted years of his life to understanding the modeling of the collapse of the towers, was apparently completely ignorant of how gravity is applied in such models. NIST, WAI, Arup, Bailey all stated in their various reports the parameters they used to ramp up the gravity loading. Mick did not invent the problem--he was just looking for a way to solve it on this particular modeling platform.

EDIT: To be clear, I'm referring to @aka's rambling post here re his ignorance of how gravity is applied to complex finite element models. The portion that belies his ignorance of how this is done in every single complex finite element model of a building is as follows:

 
Last edited:

Mick West

Administrator
Staff member
The spurious floor failures are not from things dropping on them, they are from "impulses" at the connections which are not realistic.

Floor connection need to support 6x the weight of a slab suddenly applied.
 

Mick West

Administrator
Staff member
Is there a way in blender to script in damage over time? It seems to my like Mick's model would stand indefinitely except for the fact that he designed it, from the outset, to have a catastrophic weakness that would induce collapse. It sounds to me like others are taking Mick's model and adding strength such that it stands for a longer period of time before it fails from the designed weakness. If instead of having the catastrophic weakness from the outset there was a way to script it such that the weakness was only implemented later (demonstrating that, absent the weakness, the tower stands), then that should resolve the issue.
It really just needs to be run a few time. Once without damage, once paerhaps with some lateral force or impact to make it sway, and once with damage leading to a collapse.
 

benthamitemetric

Senior Member
It really just needs to be run a few time. Once without damage, once paerhaps with some lateral force or impact to make it sway, and once with damage leading to a collapse.
I agree--having an undamaged version of the building that stands indefinitely should satisfy the current criticism re how the version that is designed to fail from the outset does, in fact, fail.
 
Top