Claim: Jim Hoffman's "9/11 progressive collapse challenge" can't be met

You might want to read up on scaling and the non-dimensional numbers. Understanding the Froude number would show you how to scale a 100 mph wind for a test on a model one-hundredth the size of the original structure, for example.
Help a guy out! What's the scaled-down wind speed if you understand the Froude number?
 
"Most of the gravity and lateral loads are normally taken by the outer tube because of its greater strength." (Wikipedia)

I'm not arguing one way or another, I'm just trying to make clear why some people think the outer shell was doing a lot of structural work without need of the floors for lateral bracing.

At the risk of being a bore, this is why I really would like a good, detailed book on the subject.
I actually agree that a laymen targeted book explaining the collapse of the three buildings is not a bad idea.
For sure there is much consensus about a lot of the processes, and so forth. But a fair amount of disagreement remains to this day.

There is consensus that the heat from unfought fires played a major role in all three collapses. There is not consensus about what that role was and how the heat effected the structures and even where it was doing its damage. And this is very important.

For example NIST used heat to explain sagging open floor trusses which pulled the facade causing it to buckle and the collapse ensued.

But others believe heat destroyed the core, leading to the collapse of the twin towers. In 1wtc the head hollowed out the core, pushed columns out of axial alignment which lead to buckling and the antenna coming down along with tens of thousands of tons of material from above the plane strike zone initiating the "ROOSD"... a version of runaway pancaking. In wtc the core was fatally weakened in the SE side and raging fires tipped the stability and the top 32 stories tipped and dropped (driven by gravity, like the antenna collapse and the floors above the plane strike zone in 1wtc. The loss of strength in the core led models... posits that columns buckled from heat or from the structure warping and columns displaced enough to lose sufficient bearing to cause core column buckling. This might have moved the entire upper blocks laterally enough for the perimeter columns to assist the falling mass in severing the slabs from the perimeter.

Most non NIST researchers agree on the ROOSD process to explain the runaway floor collapse process and demise of the core columns.

Essentially it is almost impossible to know how the heat was destroying the core's structural integrity.

NIST also places heating of the girder and beams framing into column 79 on floor 12 as the location and cause of 7wtc's collapse.

So a book might offer the various theories about what was the heat doing and where it was doing it. This is all about the initiation of the runaway collapse phase. No consensus and no proofs. It possible that any one of the several theories could be correct.
 
... Without bracing they would collapse.

Yup. Unstable equilibrium. As soon as the first small perturbation starts, the way it reacts to the continuing force from above is to perturb even more in the same direction, as there's no significant restorative force at all. Most obvious in vertical flat surfaces.
 
"Most of the gravity and lateral loads are normally taken by the outer tube because of its greater strength." (Wikipedia)
That comment from Wikipedia is misleading at the level of interest you are showing. And as far as I can tell the advice you have been given by other members misses the key point. Here are the separate points explained:
1) "Lateral Loads" i.e. wind loads are taken by the outer tube because of the location of the outer tube. See the picture. WIND from the LEFT will try to tip over the tower and will cause uplift forces on columns nearest the applied wind and downwards forces on the opposite sides. Almost entirely driven by the location of the columns. A common explanation of wind load effects asserts that the towers designers deliberately intended that the perimeter columns took the wind loads. True enough BUT the wind loads would go there anyhow because of location.
3ColsWINDLOADred.jpg
So take care when any lay person explanation says putting the wind loads onto the perimeter was intentional... it was a natural consequence of structural physics. And:

2) So the wind loads were NOT "taken by the outer tube because of its greater strength." That description is loose lay person language.
THEN
3) IF the first point "Most of the gravity.." is in fact true it would be for other somewhat more complicated reasons of the column layout. e.g. much of the gravity loading would end up on the perimeter because of the open office space "Tube" layout. Half the floor loads would go to perimeter and half to core. I'll set aside any more complicated explantions for now
 
Last edited:
I asked structural engineer Donald Friedman (who worked on the WTC cleanup) about this issue.
In the clip (at 15:10, in the same video I've embedded below), Friedman says that the core would be "self-supporting for gravity" but that it would be a "house of cards" laterally, so a moderate wind ("wind load much less than the code required") would knock it over. To me this makes it unlikely that the floor trusses transferred significant lateral loads (wind shear) to the core. The core wasn't designed to support them.

It's still possible that the floors provided lateral connections between the exterior columns and helped transfer loads from one face to the other. But everything I'm reading about Khan's tube structure suggests otherwise. The lateral connections between the individual columns came from the spandrel beams and the faces braced each other at the corners.
[Kahn] defined the framed tube structure as "a three dimensional space structure composed of three, four, or possibly more frames, braced frames, or shear walls, joined at or near their edges to form a vertical tube-like structural system capable of resisting lateral forces in any direction by cantilevering from the foundation." (Wikipedia)

And, indeed, at 13:02, Friedman says:
All of the lateral rigidity of the buildings was in the exterior.


So that's why truthers find models that depend on the floors for rigidity unconvincing. They imagine an exterior shell that you can pick up and move, push on, etc. A cage or box. A tube.

Mick says, "The outer skin was absolutely not stable by itself and would have folded in a stiff breeze - or probably just from natural variations." This suggests that the faces were something like sheets of paper.

But they were of course also joined at the corners. So the model I imagine an engineering (or physics) teacher teaching Euler columns would use is a sheet of ordinary writing paper, folded to make four faces and glued to a base. Much stronger than the very same sheet of paper to the wind trying to stand upright on its edge. Indeed, folding it turns it into a self-standing "tube".

Again, I don't have the qualifications to adjudicate this. It's just increasingly clear to me why Hoffman's challenge doesn't/can't settle anything between truthers and debunkers. Already the mental models of the buildings are too different.

Maybe ask Friedman back on TFTRH and talk it through with him?
 
In the clip (at 15:10, in the same video I've embedded below), Friedman says that the core would be "self-supporting for gravity" but that it would be a "house of cards" laterally, so a moderate wind ("wind load much less than the code required") would knock it over.
Remember that both Mick and Friedman are having a general discussion, using loose definitions and not a lot of detailed rigour. Then there is a massive quantum leap of misunderstanding between what you here them say and THIS next step of your interpretation>>>

To me this makes it unlikely that the floor trusses transferred significant lateral loads (wind shear) to the core. The core wasn't designed to support them.

It's still possible that the floors provided lateral connections between the exterior columns and helped transfer loads from one face to the other. But everything I'm reading about Khan's tube structure suggests otherwise. The lateral connections between the individual columns came from the spandrel beams and the faces braced each other at the corners.
I'll take it step by step BUT there is a central and fundamental misunderstanding involved - You are absolutely correct. The floor joists do NOT transfer horizontal load. I will outline what happens then explain step by step reference to the points you make.

How wind (lateral) loads are resisted. The resistance to lateral wind loads comes from the horizontal wind loading attempting to overturn the tower which causes an overturning moment. That overturning moment is resisted by forces in the columns. And in a conventional columns grid arrangement how the laod is distributed is complex. HOWEVER for a tube in tube design like the twin towers that moment is mostly the result of forces in the upwind and downwind perimeter column sheets. It can be legitmately represented by two forces - see the red arrows in my previous post. And the effect is a resisting MOMENT which results from the two forces AND the distance by which they are separated. The role of the floor joists plus the core is to maintain separation distance. The "moment arm". The ONLY force applied horizontally by the joists is the minimal force needed to keep the perimeter columns straight and which maintains the moment arm - the distance between the forces. And that force is an order of magnitude less than the axial forces in the columns which forming the moment resisting overturning. (I'll save the explanation for later if needed.) Please read that a couple of times and ask if you need more explantion. I'm aware that it is the key issue and did not come up in the interview in the video.

So lets address each of your points step by step:
1) "To me this makes it unlikely that the floor trusses transferred significant lateral loads (wind shear) to the core." << Correct - the floor trusses do NOT transfer horizontal loads that way.
2) "The core wasn't designed to support them." << Correct. They would primarily involve the perimeter. There would be some second order loading transferred to the core .- explanation if needed.
3) "It's still possible that the floors provided lateral connections between the exterior columns and helped transfer loads from one face to the other." << Not as a primary function. Sorry for the pedantry but any such transfer would be secondary and for a range of second or lower order issues we dont need to consider at this stage.
4) "But everything I'm reading about Khan's tube structure suggests otherwise." << You are possibly thinking about what actually happened and it conflicts with a simplified general explanation. Can you ignore Khan for now? Get the main principles clear. Lateral loads are resisted by moment from vertical forces in perimeter columns. And NOT by transfer via floor joists.
5) "The lateral connections between the individual columns came from the spandrel beams and the faces braced each other at the corners." << Yes and their primary effect was parallel with the outer face - NOT "radial" towards the core in the direction taken by floor joists. The effects are secondary. (They involve the much more subtle effects that Guy Nordenson was discussing in the video you liked a page or two back. Set those matters aside for now - they are also second order and not directly relevant to our current main theme.)

I'll reserve comments on the remainder of your post.
 
Last edited:
The role of the floor joists plus the core is to maintain separation distance. The "moment arm". The ONLY force applied horizontally by the joists is the minimal force needed to keep the perimeter columns straight.

But aren't the sides (the remaining two faces, neither upwind nor downwind) what are keeping the columns straight? At the corners, for example, it's not the floor joists that provide this minimal horizontal force but the perpendicular face, braced with spandrel beams.

The trick, as I understand Nordenson, to designing the upwind and downwind faces (which of course can be any face) is to make it transfer the shear forces (via the spandrel beams) into the interior of the face so that all the columns have to share the axial load, not just he corners. He says:
In a tube frame the idea is that the entire perimeter of the structure is mobilized so that the front and back faces of the building become like flanges of a box beam.

In any case, as I understand you, you're saying that if we asked Friedman the opposite question that Mick asked, i.e., "If we carefully remove the floors and the core would the exterior be self-supporting?" he would say either

(a) the hollow tube would immediately buckle (like a too-tall column under its own weight)
or
(b) just a little bit of wind would buckle the downwind face/columns,
but not
(c) it would be just as capable of resisting all the same wind loads (and earthquakes for the matter) because it's just a simple cantilever stuck into the ground (now with no gravity load but its own weight to carry) and it was designed precisely to resist all those lateral loads by itself.

Is that correct?
 
Let me respond separately to the two parts. I"ll take the easiest second part first:
In any case, as I understand you, you're saying that if we asked Friedman the opposite question that Mick asked, i.e., "If we carefully remove the floors and the core would the exterior be self-supporting?" he would say either

(a) the hollow tube would immediately buckle (like a too-tall column under its own weight)
or
(b) just a little bit of wind would buckle the downwind face/columns,
but not
(c) it would be just as capable of resisting all the same wind loads (and earthquakes for the matter) because it's just a simple cantilever stuck into the ground (now with no gravity load but its own weight to carry) and it was designed precisely to resist all those lateral loads by itself.

Is that correct?
The question could be just remove the floors but leave the core - whichever it would be hard to achieved HOWEVER

CORRECT it is (a) or (b). It is definitely NOT (c)

Don't ask me to pick between (a) and (b) I suspect that - IF you started removing floors from the Top down the perimeter might self buckle before you had removed all the floors. I could be wrong. I have no "engineers gut feeling" for the issue and it doesn't lend to quantified prediction. The perimeter shell would become increasingly critical as floor removal progressed. And in a hyper-critical situation the slightest bump or static lack of symmetry can trigger collapse. Are you familiar with the column "spires" that remained standing for a few seconds after the global collapse? They were way beyond critical length for Euler buckling.
 
But aren't the sides (the remaining two faces, neither upwind nor downwind) what are keeping the columns straight? At the corners, for example, it's not the floor joists that provide this minimal horizontal force but the perpendicular face, braced with spandrel beams.

The trick, as I understand Nordenson, to designing the upwind and downwind faces (which of course can be any face) is to make it transfer the shear forces (via the spandrel beams) into the interior of the face so that all the columns have to share the axial load, not just he corners. He says:
"In a tube frame the idea is that the entire perimeter of the structure is mobilized so that the front and back faces of the building become like flanges of a box beam."
I will need to look again at the video to see exactly what Nordenson is claiming BUT let me first analyse the issue qualitatively myself:
Recall also I suggested set aside Nordenson because he is describing a different aspect. Nordenson is describing how the spandrel plates force load distribution in the plane of the perimeter wall of columns. The spandrels created a very rigid sheet of columns in the plane of the perimeter. That same rigidity passed around the corners. Stiffer at the corner because the adjacent perimeter sheet forms a 90degree brace. But near the middle of the face the perimeter would still be relatively flexible "radially" or "inwards/outwards" EXCEPT if restrained by the floor joists.

Now Nordenson is describing forcing load distribution onto all the columns. And the load he is forcing is the axial load - the primary purpose. Recall I simplified the two overtrning moment forces to be a single force and taken by the upind and downwind faces. What Nordenson is describing is how some of that total overturning moment force will be carried round the corners into the "side" perimeter sheets. I hope that is clear - the explantion gets more difficult if I need to translate that I can visualise in 3D and put it into words or 2D graphics. So as yousay the corners are sartre only braced by the and

So let's retrace our steps. Your question arose because I identified that the floor joists do not transfer significant lateral force BUT they do have a minimal lateral force needed to keep the perimeter columns straight.

You correctly challenged that by noting that some "keep the columns straight" was achieved by the adjacent side wall sheet "around the corner" forming a brace. True BUT the effect of that bracing will reduce as you get some distance from the corner. Leaving the floor joists to do the bracing. Nordenson's comments refer primarily to how spandrel plates create stiffness in the plane of each face of perimeter. But as we move further from the corner that effect will diminish and "keep the columns straight" will depend on floor joists.
 
IMG_0198.JPG
Can you guess what the challenge is?

The paper is 30 cm x 41 cm and 90 g/m2.
The cardboard is basically what you get in a pizza box, cut into 7.5cm squares.

Those are AA Duracell batteries. You must put an equal amount of batteries on each floor and the floors must be equally spaced throughout the tower.

My (not rhetorical) question is: in order to use these sorts of materials to meet (something like) Hoffman's challenge, how big would you need the piece of paper & cardboard squares to be? (You can also specify the grade of paper if you like.)

Ideally, you will keep the floor heights and bases [spans] equal, so that the structure is a series of cubes, but I'm willing to hear why that's unreasonable.

If you want more or bigger (or smaller) batteries, just ask!
You can also cut holes in the paper as you choose to weaken the walls.
You can (and probably should) connect the tower to the base (the big piece of cardboard) using as much tape as you like. If you need a bigger or heavier base, that can also be arranged.

The structure must be strong enough in its initial state to let me poke a hole through the paper with the scissors at any point.

The structure must also be strong enough to let me shift the base about 10% of the width of the tower back and forth instantaneously (simulating an earthquake). I think that makes any "wind testing" moot.

The initiating failure must be brought about using the scissors in the ordinary way, cutting the paper, cardboard, or tape as much as you like somewhere in the top 20% of the structure.

This event must cause the whole structure to be destroyed. (It's not good enough that all the floors and batteries end up at the bottom of the tube, which remains stranding.)papertower.jpg

The solution using the smallest piece of paper wins.

(Note: you are of course free to ignore this challenge, ridicule it, or meet it as you choose. Once it occurred to me I thought it was too good to keep to myself. I don't think I'm going to be able to meet it with the 30x40cm piece of paper I have. But I'm going to try.)

[EDIT: The thing I like about this challenge is that it lets us scale up instead of down. That is, if we can't get the 41 cm tower to progressively collapse because gravity doesn't scale well, we can imagine it -- or even build it -- bigger until we reach floor heights where a gravity-driven collapse becomes possible.]
 
Last edited:
...the effect of that bracing will reduce as you get some distance from the corner. Leaving the floor joists to do the bracing.
The image I'm getting here is that without the floor joists the face would billow inwards like a sail. Is that right?
 
Let me take a stab

Image a flat plane the size of one face of the tower... It rests on its edge. effectively it is like standing a sheet of paper or piece of cardboard on its edge.... the slenderness ratio of the paper thickness to its height is about 1/1200. (twin tower perimeter columns were 14", (height of tower was 110 stories x 12'-9" x 12 or 16,830" tall) The facade steel was 14" thick.
The slenderness ratio is facade thickness / facade ht which would be
14"/ 16,830" or .000832 - that's quite thin and would exceed the threshold of 1/182 or .0055 for Euler buckling. Facades could not stand without bracing!

The issue is raised if the facade acted as a solid single plate... ie all the connections - column to column, spandrel to spandrel and column to spandrel were rigid enough to make the components act as single plate

So the forces on the face would be the wind pressure/ft x 208'x1400'
a wind speed of 40 mph (gale not hurricane) would exert 4 pounds per sq ft.
70 mph hurricane winds are over 3x the force of gale winds.

So the force of a wind normal to a facade for a 40 mph wind would be

208 x 1400 x 4# = 1,165,000 pounds.... applied as a uniform load over the entire face of the building. (winds aloft are stronger) Hurricane force wonds would be well over 3,000,000 pounds

The wind pressure on the facade is resisted by:
a) the two perpendicular walls connected at the diagonal corners​
b) the 20 connections of the tree columns to the foundations​
c) the floor plates/slabs connected via the bolts on the 20 web truss seats​

So this force was distributed to the multiple elements of the structure. The 110 floor plates in turn would "transfer" these wind forces to the connections to the perpendicular facade wall... and the core via the belt girder which surrounded it.

It's rather difficult (above my pay grade) to compute what the forces a, b & c were separately.... and individually at the connection level.

But if Euler's calculations control.... a side of facade itself would buckle from its own weight without the floors bracing it..
 
a side of facade itself would buckle from its own weight without the floors bracing it.
I'm not sure I follow all your calculations, but are you doing this for a facade on its own, or are you taking into account that they brace each other along their edges where they meet (to form a square tube)? As Nordenson puts it, each facade is basically a flange for the two facades it meets at the corners.

effectively it is like standing a sheet of paper ... on its edge
I did this experiment today. A single sheet of paper of course collapses under its own weight. But four pieces of paper, set at right angles to each other (to form a square tube), and taped together along their edges, stand up just fine.
 
I'm not sure I follow all your calculations, but are you doing this for a facade on its own, or are you taking into account that they brace each other along their edges where they meet (to form a square tube)? As Nordenson puts it, each facade is basically a flange for the two facades it meets at the corners.
Yes... this is the explanation of how a tube works as a beam. The issue as I mentioned is whether the entire side which was composed of panels bolted together would behave as a single plate. Suppose all the steel of the facade was turned into a thin plate? Not that it would be tapered... thicker at the bottom and thinner at the top where the winds were strongest! So I can't tell you how this would perform as web... or a flange.

The diagonal cut corners are interesting in this conception of web and flange and a completely rigid connection. This may or may not play a role.


I did this experiment today. A single sheet of paper of course collapses under its own weight. But four pieces of paper, set at right angles to each other (to form a square tube), and taped together along their edges, stand up just fine.
matter of scale of the connections... of the sheets at the corners.

I suspect that the tube by itself would buckle from Euler force and also the wind would bow the facade inward stressing connections. Those forces were resisted by the floor plates.
 
I did this experiment today. A single sheet of paper of course collapses under its own weight. But four pieces of paper, set at right angles to each other (to form a square tube), and taped together along their edges, stand up just
Try attaching 4 more of those paper tubes on top of the first one and see what happens.
 
Try attaching 4 more of those paper tubes on top of the first one and see what happens.

papertower.jpg
This one stands up also when I take out the floors (every 7.5 cm). But I could probably stack four of them on top of each other without causing them to collapse.

[EDIT: It's actually 41 cm tall. It's a page torn from a croquis pad with a coil binding. It turns out the pad is 42 cm, the page you tear off is 41 cm. Just in case you're doing the math at home.] [EDIT: I updated the picture.]

[EDIT: This one has about the right the aspect ratio for our purposes, almost 6:1. But all this is of course quite arbitrary since the strength of the materials haven't been chosen with any particular care. It's just the grade of paper that happened to be in the pad, and I used the dimensions I could get just by folding. I'm hoping I can push the materials to their breaking point by scaling up, or maybe finding some cheaper paper. The challenge will be getting it to be "strong" in its normal configuration and yet vulnerable to top-down collapse when properly loaded and wounded.]
 
Last edited:
now cut holes with a razor blade in that paper to mimic the mesh style frame of the twin perimeters. You'd have to ask the guys how many evenly spaced slits you'd need to match the twins though. looks liek a lot.
1616867300139.png 1616867436774.png
 
now cut holes with a razor blade in that paper to mimic the mesh style frame of the twin perimeters
Yes, I think something like that will be necessary to get it to work. But I don't think it's to "mimic" the windows as such. After all, that would require me to cut out the holes and then fold the paper into little box columns to recover the strength that is lost in order to have windows.

(I looked into this a while back, and I'm pretty sure that if you "unfolded" the perimeter box columns and welded them together into a solid sheet of steel, so the whole outer shell is just one big box column, kind of like my sheet of paper, the resulting structure would be exactly as strong as the WTC's outer shell actually was with windows.)

It's just because the paper is too damn strong at these scales that it will have to be artificially weakened. A few slits (not even holes) would probably do the trick.
 
As I've said before, I'm not an engineer, and consistently defer to those who know the field much better than I. Earlier in this thread, the ROOSD term came up again.
With due respect to econ41, Major_Tom and or anyone else who
may've had any hand in coining the term, it still kind of puzzles me.

R = runaway (a term I rarely associate with buildings...will do last)
O = open (next 3 seem only to describe where we're talking about)
O = office
S = space
D = destruction (well, obviously virtually everything was destroyed)

So, back to "runaway." I'm guessing that this means that once
floors began to fall, they did so kind of unstoppably (?)

The first time I ran across "ROOSD," years ago, without much context, I took it to be a (possibly disputed) theory re. how the
WTC towers came down. But once I looked up what the acronym
means...it has seemed to me to be saying little more than that
the building collapsed...which we all already know.
I'm sure not qualified to advocate one engineer's theories
over another's...but am I correct in assuming that ROOSD
is just a shorthand for the collapse we saw, rather than any controversial theory or explanation for the collapse?
 
it still kind of puzzles me.
Warning to outside readers: I am guessing here, so don't put too much faith in my interpretation.

i'm going to comment so that the engineers can correct me at the same time. MY take on what Roosd signifies is a distinction from "pancaking", which i think pancaking means the floors collapse on top of each other but without breaking up.. like Thomas' cardboard floors will fall because cardboard is incapable of breaking up into pieces.

My understanding is the 'destruction' bit in the Roosd is the chaotic nature of the cement etc breaking and smashing and pulling on supports etc.

so... not this:

Source: https://youtu.be/0aj7k1tv4zQ?t=17
 
but am I correct in assuming that ROOSD is just a shorthand for the collapse we saw, rather than any controversial theory or explanation for the collapse?
Roosd signifies is a distinction from "pancaking", which i think pancaking means the floors collapse on top of each other but without breaking up.. like Thomas' cardboard floors will fall because cardboard is incapable of breaking up into pieces.
In the OP I say:
At the time, [Hoffman's] target was "pancake collapse", not the ROOSD model that I know some people here hold to, but the challenge itself still seems to make intuitive sense whatever explanation you prefer.
My suggestion is that the challenge of designing a model that "reproduces" "whatever explanation you prefer" is an interesting and serious one. That is, if you think you understand how the WTC collapsed and that claims that the "official explanation defies the laws of physics" are nonsense, then just describe a simpler structure (so that we can keep track of its finite elements) that undergoes the process of destruction you imagine.

And describe one that we have a reasonable chance of building out of affordable materials. I continue to believe this is possible.

I personally don't need the cardboard to break because in my latest challenge I'm representing a concrete slab with some number of batteries taped to it. But maybe ROOSD supporters can think of a better way representing the "chaotic" action of the collapses.
 
ROOSD is an acronym for a more accurate description of a "pancake collapse".

The floor collapse sequentially and unstopped from a high floor which has been over loaded and fails. It's mass, and the contents on it become a dynamic load on the floor below which it lands on... and that floor does the same thing to the floor below it. The ROOSD can include parts of floors which drop onto the same area on the floor below.

It is hard to know if the overload to that first floor to collapse was distributed over the entire foot print at the same time. It seems possible that the some areas led and others was lagging... but they all began unstoppable collapse of every floor.

The dynamic load was so massive that the floor it acted on barely resisted of slowed the descent of the collapse debris. The mass seems to have reached a maximum velocity of about 65 MPH... or about 100' per second. And this aligns with the estimates for the collapse times.

Severing the slabs by column impacts when / if the top block translates is possible... but all the column connections were staggered and it seems that this may not be the created of unattached floors.
 
The image I'm getting here is that without the floor joists the face would billow inwards like a sail. Is that right?
Yes.....you are correct in the concept but let me identify a couple of limitations....
..It could be inwards or outwards; AND

..."billows" suggests large flapping movements of something like canvas or a sail cloth BOTH of which have very little rigidity.

The actual movements will still be small and restrained by the elasticity of the steel.

IF the perimeter was not held in place by the floor joists - if it could move a few inches that "freedom" would change the effective length of the column. I think you are familiar with the concept of "Euler Buckling". THe critical isaue is the effective length of a column. The perimeter columns retrained by the floor joists in the inwards outward direction had an effective length of one storey. (In the "sideways" direction they were restrained by the spandrel connectors.) Remove the floor joist or loosen it to allow a few inches movement and the net result is that the effective length is doubled. And the effect on column strength is worsened by four times. It is a "square" function.

No need to go into more detail. You have the concept - the perimeter columns away from the corners of the buildings would be relatively free to "flap in the breeze". Somewhat similar to a billowing sail. BUT constrained by the elastic properties of the steel - the "flapping' or "billowing" is still bending the steel tho it is only for a few inches. That few inches is all that is needed.
 
Let's try to lay this one to rest.
With due respect to econ41, Major_Tom and or anyone else who
may've had any hand in coining the term, it still kind of puzzles me. << econ - I wasn't involved

R = runaway (a term I rarely associate with buildings...will do last)
O = open (next 3 seem only to describe where we're talking about)
O = office
S = space
D = destruction (well, obviously virtually everything was destroyed)

So, back to "runaway." I'm guessing that this means that once
floors began to fall, they did so kind of unstoppably (?) <<econ CORRECT - both aspects "yes that is what it meant" AND "yes it is true"

.............but am I correct in assuming that ROOSD is just a shorthand for the collapse we saw, rather than any controversial theory or explanation for the collapse? <<econ both actually - just shorthand that happens to be correct AND controversial - see explantion below.
Yes there was a long history of controversy. Put simply the key points are:
1) Yes it is just a way of describing what actually happened.
2) Yes it was disputed for two main reasons BUT first put it in context;
3) Prior to about 2009 ALL explanation of the rapid progression collapse of the Twin Towers were dominated by gross approximations - one dimensional simplifications which paid no regard to the actual mechanism. Both the emerging "sides" of "truthers" v "debunkers" made the same errors. The "debunkers" divided into two sub camps of research. Both wrong.
4) In 2007 - on a side line forum and unaware of both the mainstream errors and the coming controversies - I posted an explanation of the real mechanism. Without the ROOSD acronym and ignoring the mainstream debate which I was naively unaware of. Near enough concurrently Major-Tom independently made the same observations and put it into mainstream debate on JREF in 2009 and complete with the ROOSD acronym. And walked head first into the anti-truther claque of JREF die-hard engineers.

So the two reasons it was controversial:
(a) Major Tom was agnostic but perceived as a "truther" and of course all JREF debunkers knew that no truther could ever be right. (I kid you not - it is still readable - the whole farce - on what is now ISF); AND
(b) a range of utterly childish objections to the acronym which were both linguistically stupid AND supported by false engineering claims.

So that is the back ground.
(i) ROOSD is a valid acronym to label a correct explanation of the key feature of the actual collapse mechanism;
(ii) Like all good acronyms it is both descriptive and saves a lot of typing. IF you can ignore the fact that a couple of truthers got something right when the mainstream debunkers were wrong. I have no such hangups.

AND the silliest part of the "controversy" was reached when a debunker claimed "It is right when we the debunkers say it BUT it is wrong when a truther says it". Again - I kid you not.
 
Last edited:
@deirdre - just a bit of clarification. You have two different issues sort of conflated in this:
i'm going to comment so that the engineers can correct me at the same time. MY take on what Roosd signifies is a distinction from "pancaking", which i think pancaking means the floors collapse on top of each other but without breaking up.
The two distinct topics that have caused two different areas of confusion are:
1) "pancaking" was controversial in the early days because a preliminary report by FEMA suggested that "pancaking" was what initiated the Twin Towers collapse. That was incorrect and NIST dismissed it in their report. But the "disagreement" between the authorities gave a point of attack for truthers..

Note that "pancaking" was and is wrong as an explanation of initiation of collapse.

BUT OUR current discussion is about "progression". And "pancaking" is a legitimate description for the "progression stage". Personally I prefer to not use the term because the historic controversy can be manipulated by truthers in discussion.

BOTTOM LINE: "pancaking" is wrong for "initiation stage". It is valid to describe "progression stage" but I recommend not using it OR, if you want to use it, making your meaning explicitly clear. THEN

2) "ROOSD" is a valid and technically correct descriptive acronym for the key feature of the Twin Towers "progresion stage". BUT it is NOT a distinction from "pancaking". Both are descriptors of the same process both or either can be used when appropriate to the context.
My understanding is the 'destruction' bit in the Roosd is the chaotic nature of the cement etc breaking and smashing and pulling on supports etc.
It was originally more general - the process which caused the overall "destruction of the towers".
 
Last edited:
2) "ROOSD" is a valid and technically correct descriptive acronym for the key feature of the Twin Towers "progresion stage". BUT it is NOT a distinction from "pancaking". Both are descriptors of the same process both or either can be used when appropriate to the context.
huh. I admit i never did grasp (after 5 years of reading MB 911 threads) how roosd is really different from the term pancaking. So now that i know they aren't different i'm going to have to agree with NoParty on this one. probably easier for laymen if you just call it Consecutive Floor Collapse or something.
 
huh. I admit i never did grasp (after 5 years of reading MB 911 threads) how roosd is really different from the term pancaking. So now that i know they aren't different i'm going to have to agree with NoParty on this one. probably easier for laymen if you just call it Consecutive Floor Collapse or something.
Your call - it could work on MB. But remember that some of us visit other places of outer darkness where yet another label only adds to the noise. Plus - in those other truther infested fora - using "pancaking" is just asking for trouble. ;)

Still - I have on occasions been accused of pedantry - I have this funny idea that we should be clear what we mean. :rolleyes:
 
your paper tower -even with out the appropriate steel holes- can hold up a cardboard floor with batteries taped to it? have you tried that yet?
IMG_0202.JPG
Yes, easily. With the floors in place, the tower can support at least ten batteries on the top. (And, as I discovered when taking the pictures, it can even sustain a mild earthquake while I rotated the base.) I'm pretty confident that the paper could bear 4 or 5 batteries on each floor (separated by about 7 cm).

Note: this is still a very early design, with some pretty ad hoc construction methods, especially at the corner! The strength of this seam (both vertically and laterally is of course something that needs to be thought about.
IMG_0216.JPG
 
the perimeter columns away from the corners of the buildings would be relatively free to "flap in the breeze".
The corner columns would be completely rigid, right? And as we move from center to perimeter each column would be successively less "free" to bend. Now, each column is laterally connected by the spandrel plates (about 100 of them along its length) to a stronger (less free) one beside it (closer to the corner), until, like I say, we get to the corner itself which, for all intents and purposes, can't bend and therefore won't Euler buckle.

The image I'm getting is that of a (very stiff) sail unfurled between two (even stiffer) "masts" with a "yard" at the top (hat truss) and bottom (base structure). Is that right?
 
The corner columns would be completely rigid, right? And as we move from center to perimeter each column would be successively less "free" to bend. Now, each column is laterally connected by the spandrel plates (about 100 of them along its length) to a stronger (less free) one beside it (closer to the corner), until, like I say, we get to the corner itself which, for all intents and purposes, can't bend and therefore won't Euler buckle.

The image I'm getting is that of a (very stiff) sail unfurled between two (even stiffer) "masts" with a "yard" at the top (hat truss) and bottom (base structure). Is that right?
Yes - correct on both the concept and the detail. The corner is near enough to "completely rigid" for our current discussion so I won't complicate the discussion. It is actually still "slightly elastic" and the difference between "completely rigid" and "still very slightly elastic" can be critical in some situations but not in your scenario. If you (or any other members) are interested I can post an example. But probably best not to derail the ongoing discussion.
 
[The corner] is actually still "slightly elastic" and the difference between "completely rigid" and "still very slightly elastic" can be critical in some situations but not in your scenario.
But now I'm puzzled. If the corners, base, and top are like "masts" and "yards" to the "sail" of the face, how does the bending, i.e., "billowing", of the columns near the center of the face cause the whole facade to collapse?
billow.jpgbillowlines.jpg IMG_0229.JPG
Edit: I thought maybe an illustration would be useful. In this model, I've removed all the floors, so it's just a hollow tube. Obviously, wind doesn't "poke" at buildings like this, so this, I imagine, is much worse than what the towers would have experienced. (Correct me if I'm wrong.) My hand is resting lightly on the top as I poke at the face. It is nowhere near to collapsing under these conditions.

NOTE: It would be interesting to work out how big a piece of a 90 g/m2 paper would need to be to get a tower made by folding it like this to collapse under its own weight. That would give us a great sense of the scaling problem.
 
Last edited:
But now I'm puzzled. If the corners, base, and top are like "masts" and "yards" to the "sail" of the face, how does the bending, i.e., "billowing", of the columns near the center of the face cause the whole facade to collapse?
Take care - we are confusing two vastly different situations and two different aspects of explanation. Specifically we (you and I in our discussion) are not at this stage discussing what caused the "whole facade to collapse". That is a distinct and separate topic. Yes I can explain it if you want. And I can also also discuss how it relates to our original discussion about where the wind load lateral forces are resisted. I would prefer to leave discussion of your paper model of perimeter tube stability between you and other members unless you specifically want my comments.

So let's clear up the confusion which is arising.

The "billowing sails" explanation arose out of my description of how the lateral forces of wind loads are handled by the structure. Put simply the lateral force applied to the tower by wind causes an overturning moment which is resisted by forces in the perimeter tube columns. Simply approximated by two single forces - tension - downwards force - in the upwind perimeter and increased compression in the down wind perimeter.

We then explored the Nordensen "complication" that recognises forces also arise in the "side" perimeters. And we discussed how the forces propagate around the corner. All that has nothing to do with causing the "facade to collapse". It is a separate process.

But the bottom line is that we showed that the floor joists had to remain connected so the real event mechanism DID NOT have "sails billowing effects". So "sails billowing" did NOT cause facade collapse. The situation for sails billowing did NOT exist when the facade collapsed.

Two other issues are relevant:
(a) The "sails billowing" topic arose when considering wind loads. Wind loads were not a significant factor of the real event on 9/11; AND
(b) Your paper model is not a model of the real event. IT is a research model to explore the structural stability of the perimeter "tube" of columns with NO floor joists and the side panels free to "flap in the breeze". THEREFORE a very different scenario from the one we were discussing.

As for what did "...cause the whole facade to collapse?" - that is another separate issue. I can explain but it is off-topic in this present thread. It MAY need to be undertood at some stage depending on the direction the debate takes.
 
Last edited:
That is a distinct and separate topic.
Just so we're on the same page: I don't think we're discussing how the facades actually collapsed. We're discussing whether the exterior shells of the WTC towers were a self-supporting structure without the floor joists in place.* As I understand our conversation at this point, you are explaining to me,

(a) why, if we (magically) removed the core and the floors, the exterior shell would collapse to the ground, either immediately, or when disturbed by a light wind, and/or

(b) why it would be impossible to construct, from the ground up, only the the exterior shell of the WTC towers (because at a certain height it would reach its Euler buckling limit).

If that's not what you're trying to explain then, yes, we're talking past each other.

*[EDIT: for those coming to the discussion late, the issue came to a head with @Mick West here, and @econ41 joined in here.]
 
Last edited:
The way to go is this:

  1. Construct the tube model with intact floors for lateral bracing
  2. Load it to the point where it starts buckling
  3. Measure the mass that caused the buckling
  4. Now, construct the same model (reuse the same, if it can be repaired without changing anything)
  5. Remove some floors
  6. Again, load it to the point where it starts buckling
  7. Measure the mass that caused the buckling this time
You will definitely find that the second time around, the mass is only a fraction of what it was the first time around.

It doesn't matter very much what the factor in your model will turn out to be, as it is not designed to have any aspect in scale with the original twins.

This will teach that removing lateral bracing in a tube design reduces vertical capacity by a significant factor. That's all you need to understand from this model.
 
...depending on the direction the debate takes.
Just a quick point of personal privilege: I'm not debating. I'm trying to understand your understanding of the exterior shells of the Twin Towers. I had always thought they were entirely self-supporting. And I didn't realize how committed @Mick West was to the idea that they weren't "shells" so much as "skins", unable to maintain rigidity without a "skeleton" inside the building. I thought the design concept was that the facade was an "exoskeleton".

If I understand this correctly, then there is indeed a debate or dispute between truthers and debunkers and I have (unknowingly) been on the "truther side" of it. Until this discussion, I thought everyone believed that the facade carried "all the lateral loads and most of the gravity loads" (quoting something from memory, probably in the Wikipedia article or its sources). This has been very illuminating.

But, to repeat, I don't have a position on this (because I didn't know I needed one). At this point, I'm still trying to make up my mind.
 
Last edited:
The way to go is this:

  1. Construct the tube model with intact floors for lateral bracing
  2. Load it to the point where it starts buckling
  3. Measure the mass that caused the buckling
  4. Now, construct the same model (reuse the same, if it can be repaired without changing anything)
  5. Remove some floors
  6. Again, load it to the point where it starts buckling
  7. Measure the mass that caused the buckling this time
You will definitely find that the second time around, the mass is only a fraction of what it was the first time around.

It doesn't matter very much what the factor in your model will turn out to be, as it is not designed to have any aspect in scale with the original twins.

This will teach that removing lateral bracing in a tube design reduces vertical capacity by a significant factor. That's all you need to understand from this model.
IMG_0202.JPG
Thanks. I've already done this, albeit in a slightly simpler form. I've tested the tube with floors and without, in both cases putting the entire load on the top. The point you're making is indeed confirmed. But that's not something I needed to learn.

What I'm looking for is a stable configuration of floors, equally loaded with batteries, such that it's possible to destroy the whole tower solely by weakening either the paper, the cardboard floors, or the connections between them, in the top 20% of the structure. [Edit: a crucial point here is that this also means that at least 80% of the mass will be bellow the level of the weakening.]

I'm expecting that this will require a significantly bigger structure (i.e., that it has to be "scaled up" to leverage the necessary gravitational forces) but, since I will be using the same materials, I hope I will not need to build a 1400' x 200' paper tower with tons and tons of batteries on each floor!
 
Just so we're on the same page: I don't think we're discussing how the facades actually collapsed. We're discussing whether the exterior shells of the WTC towers were a self-supporting structure without the floor joists in place.* As I understand our conversation at this point, you are explaining to me,

(a) why, if we (magically) removed the core and the floors, the exterior shell would collapse to the ground, either immediately, or when disturbed by a light wind, and/or

(b) why it would be impossible to construct, from the ground up, only the the exterior shell of the WTC towers (because at a certain height it would reach its Euler buckling limit).

If that's not what you're trying to explain then, yes, we're talking past each other.

*[EDIT: for those coming to the discussion late, the issue came to a head with @Mick West here, and @econ41 joined in here.]
Given the several overlapping discussions...

I agree your context of: "discussing whether the exterior shells of the WTC towers were a self-supporting structure without the floor joists in place."

And - in that context I agree both points (a) and (b). Recall in a previous post I was unsure of (a) - my "engineer's gut feeling" wasn't working. But Jeffrey Orling contributed some numbers and I think he is right. I have ZERO doubt that (b) is impossible.
 
Back
Top