The pre-collapse inward bowing of WTC2

We both must have stared at what I wrote simultaneously. :)


(Legging it here…)
I know the feeling. My response to Oystein at post #194 has me out at my fringe.....and I'm (supposed to be) the engineering physics person. 25 years back I could have answered it in my sleep. I wasn't kidding about the "rusty 73yo brain". I'm holding my breath in case I got some detail wrong such as the little bits of trigonometry. ;)
 


I know the feeling. My response to Oystein at post #194 has me out at my fringe.....and I'm (supposed to be) the engineering physics person. 25 years back I could have answered it in my sleep. I wasn't kidding about the "rusty 73yo brain". I'm holding my breath in case I got some detail wrong such as the little bits of trigonometry. ;)

So the IB occurred because fires under some areas caused sufficient sagging? With respect to 2wtc do we know where the extensive fires were? They appear to me to be along the east side.... but wasn't the observed bowing on the east side of the north facade?... west of the leaking liquid? Why wouldn't the IB be manifest on the east side first?

When the top really got moving it dropped to the S and E... No? Wasn't the strong... virtual hinge in the NW? Why would the bowing be seen first on the N side east end?
 
It would if they were at their point of buckling.
What you refer to is a concept I intuitively grasp but don't know what that is actually like... Maybe. So you take column A and restrain the ends and then load it up bit by bit... let's say in some container sitting on top which you pour sand into. At some point the column's axial compression capacity is reached. It has not buckled.. it's straight a and true. As add one more grain of sand and the capacity is exceeded. It would buckle and bow out in one direction the weakest one even in the tiniest way.

Now if we did the same experiment with a load applied to the side... which is not retrained... like a chunk of steel welded to one side. I think the bowing would be AWAY from the weight.

If the load was a beam spanning to another column which as the far end free to move or rotate.. un pinned.. it would still bow out opposite to the attached beam.

If there was a cable attached to the side of the column (under some tension) the column would fail and bow toward the cable (tension forces).

The issue with the WTC floors has to deal with the integrity and the nature of the end connections of the trusses. If the connections performed and the core side of the floor dropped or was pulled into the core with the belt girder... the trusses would pull the facade toward the core. If the connections failed the floors wold drop and break free This would leave facade with unbraced columns.

The IB looks like it was (a guess) an entire section of floor collapsing and pulling at the facade...It may likely be because the belt girder on the core side dropped or the beam stub outlookers supported the belt girder failed... the section of floor fell first at the core side and pulled at the facade maybe breaking the connections. There is no way the facade could bow 55" with the floors there unless they were crushed by the IB of facade (first tension then the columns compress the trusses...Maybe)

The fires seem to be well hidden... likely closer to the core... where the IB is seen and this suggests that the failure as the core side... and in 2wtc this is likely where the fires were raging with air from the HVAC and elevator risers.

Maybe
 
...a few mm of pull in would not have turned those columns to non performing.
It would if they were at their point of buckling.
Correct. The basic issues are well understood aspects of engineering.
1) Columns buckle if subjected to too much axial load; AND
2) Impose a little bit of sideways movement or eccentricity and the amount of axial load that can be carried is drastically reduced.

So a column loaded to far less than its design load will fail if pushed a little bit out of line.

In the technical language of engineering the "little bit out of line" is called p-delta.

p-delta depends on how much axial load the column is carrying. So a column already near to full axial load requires very little sideways movement to induce failing. A less fully loaded column will tolerate more sideways movement.

BUT once the sideways movement - or load eccentricity - exceeds p-delta - failure results.

I agree. No-one with any sense tries to put a finger on chaos. That's all it is possible to say here.
The WTC collapses were a long way short of true chaos Jazzy. We can be assured of all the main aspects of engineering physics relevant to understanding why the buildings collapsed, why they collapsed the way they did, why they collapsed as fast as they did and why CD (explosive or thermXte) was not required.

There is a lot of detail we can never know but none of it changes the fact that the main aspects are understandable.

Engineering physics can and IMO does show CD was not required but cannot prove there was no CD. That is not a failure of engineering. It is the logical problem of cannot prove a negative.
 
But my contention is that the 'bowing' is not before the collapse but during the collapse. Collapse was already in course once the core was compromised, and one observed effect of that was inward bowing.

Seems like a semantic argument. You could say the collapse of WTC2 started when the plane hit.

The inward bowing of the exterior was observed minutes (IIRC) before WTC2 fell. Are you suggesting the core was being slowly lowered for several minutes before the high speed portion of the "collapse", and this was slowly pulling in the exterior in the floors that were on fire?



The above image was six minutes prior to collapse.
 
Seems like a semantic argument. You could say the collapse of WTC2 started when the plane hit.

The inward bowing of the exterior was observed minutes (IIRC) before WTC2 fell. Are you suggesting the core was being slowly lowered for several minutes before the high speed portion of the "collapse", and this was slowly pulling in the exterior in the floors that were on fire?

The above image was six minutes prior to collapse.
I would say so! I would and I have suggested that the period from plane strike to actually collapse was one where the core ( and parts of the OOS flooring) were coming "apart"... load paths were destroyed and redistributed... this included local dropping of floors (and with support beams)
 
Last edited:
Correct. The basic issues are well understood aspects of engineering.
1) Columns buckle if subjected to too much axial load; AND
2) Impose a little bit of sideways movement or eccentricity and the amount of axial load that can be carried is drastically reduced.

So a column loaded to far less than its design load will fail if pushed a little bit out of line.

In the technical language of engineering the "little bit out of line" is called p-delta.

p-delta depends on how much axial load the column is carrying. So a column already near to full axial load requires very little sideways movement to induce failing. A less fully loaded column will tolerate more sideways movement.

BUT once the sideways movement - or load eccentricity - exceeds p-delta - failure results.
That is what I have been (poorly) trying to say. It is illustrated by Mick's pop can videos. Fully loaded, then small deformation initiates buckling.
In this case it may have been floor sag or, if I read Jeffrey correctly he postulates the floors were dragged down at the core side either by shortening core columns (various mechanisms by which that could occur) , or by the floors decoupling from the core and falling. Same effect, increased side load on perimeter columns towards the core. Add in increasing vertical loading and the columns begin buckling,,,, inwards. While we seem to focus on increased axial load THEN side load, there's no reason why side loading could not develop first or along with increased axial load.
Yes it could be partially driven by core shortening due to heat induced creep. Yes the core beams may have pushed core columns slightly out of vertical allowing them, already under greater than normal loading beginning with impact damage, to suffer creep deformation due to heat. That would, in turn, increase perimeter column axial loading as well as inward pull due to floor angle.
 
Last edited:
That is what I have been (poorly) trying to say. It is illustrated by Mick's pop can videos. Fully loaded, then small deformation initiates buckling.
In this case it may have been floor sag or, if I read Jeffrey correctly he postulates the floors were dragged down at the core side either by shortening core columns (various mechanisms by which that could occur) , or by the floors decoupling from the core and falling. Same effect, increased side load on perimeter columns towards the core. Add in increasing vertical loading and the columns begin buckling,,,, inwards. While we seem to focus on increased axial load THEN side load, there's no reason why side loading could not develop first or along with increased axial load.
Yes it could be partially driven by core shortening due to heat induced creep. Yes the core beams may have pushed core columns slightly out of vertical allowing them, already under greater than normal loading beginning with impact damage, to suffer creep deformation due to heat. That would, in turn, increase perimeter column axial loading as well as inward pull due to floor angle.

NB that the core columns were not comments to the trusses!... The trusses were carried on the core side by a belt girder which was cantilevered from the 24 perimeter core columns and beams stub out lookers... and the belt girder was as much as 36" or so from the face of the core perimeter columns up at the level of the IB. There were minimum 24 lookers but probably 32 as the corner columns had one in each direction.

This is not a lot of connection points to support 240 trusses.. so if a few of those outlookers went south they would take out the core side support for 10 trusses or over 60 linear feet of the floors and correspondingly at the facade.
 

Attachments

  • FRAMING.pdf
    69.5 KB · Views: 675
  • core cross bracing.PNG
    core cross bracing.PNG
    48.9 KB · Views: 522
I would say so!

Beachnut is fond of posting a pic showing column in WTC5 or 6 that apparently slowly deformed plastically. No real reason why tower core columns could not do the same.
Since inward movement of the perimeter seems to have continued occurring right up to 'release' , sure, the semantic of saying it is part of collapse. If the collapse clock is set to time zero at release then semantically the inward bowing might be regarded as a transition from static ( in the vertical movement vector) to global collapse, a time period characterised by some parts of the upper structure exhibiting downward movement. Global collapse beginning when all parts of the upper section of the structure are moving downwards.
 
NB that the core columns were not comments to the trusses!... The trusses were carried on the core side by a belt girder which was cantilevered from the 24 perimeter core columns and beams stub out lookers... and the belt girder was as much as 36" or so from the face of the core perimeter columns up at the level of the IB. There were minimum 24 lookers but probably 32 as the corner columns had one in each direction.

This is not a lot of connection points to support 240 trusses.. so if a few of those outlookers went south they would take out the core side support for 10 trusses or over 60 linear feet of the floors and correspondingly at the facade.
Don't you hate autocorrect sometimes?

Connection to core with outrigger supported steel belt, in between. That arrangement, in my strictly intuitive sense, looks more prone to heat failure than direct coupling to columns. Columns hot enough to deform , under side load from expanding beams plus increased axial load, would certainly imply the floor to column structures were hot enough to fail.
 
Don't you hate autocorrect sometimes?

Connection to core with outrigger supported steel belt, in between. That arrangement, in my strictly intuitive sense, looks more prone to heat failure than direct coupling to columns. Columns hot enough to deform , under side load from expanding beams plus increased axial load, would certainly imply the floor to column structures were hot enough to fail.

Don't have auto correct and don't proof. Sorry...
 
Interesting find. Does anybody have visuals for what happened at the top stories at this moment? Whas the top coming down when this happened?

It's the start of the collapse, so you can see it in the other videos, this is just a close-up
 
Just watching that vid sort of shows the whole 'dropped into it's own footprint' to be not true. As the top floors fall they cant over about 25-30 degrees and seam to be falling at least partly outside the foot print of the building as they are enveloped in the dust cloud. Compare that to any other vids of controlled demolitions and you'll see that the upper stories tend to fall straight, with little or no canting over. To me that says its a natural collapse, (or as natural as you can say after two of the worlds biggest buildings had been rammed by fully loaded jet-liners.)

(funny after all these years I still find that footage, along with the Hindenburg crash, some of the most disturbing images I've ever seen)
 
Last edited:
Just watching that vid sort of shows the whole 'dropped into it's own footprint' to be not true. As the top floors fall they cant over about 25-30 degrees and seam to be falling at least partly outside the foot print of the building as they are enveloped in the dust cloud. Compare that to any other vids of controlled demolitions and you'll see that the upper stories tend to fall straight, with little or no canting over. To me that says its a natural collapse, (or as natural as you can say after two of the worlds biggest buildings had been rammed by fully loaded jet-liners.)

(funny after all these years I still find that footage, along with the Hindenburg crash, some of the most disturbing images I've ever seen)

it's as clear as day. Bin Laden's initial plan was to topple the top portions. As an engineer, he knew by crashing planes into the upper floors this would happen. He got more than he bargained for. Thank you Mick West for posting this little gem!
 
The plan with the first bombing in the underground garage, was to destroy part of the foundation of one tower and topple it onto the other tower.
 
That is correct! Not to mention it was intended to kill 250,000 people. Crazy stuff.
The wrench in the works then was their inability to find a parking space at the correct location, otherwise it did have a chance of working. Instead it just blew a big hole through several levels of the garage and caused a difficult to fight fire.
 
The wrench in the works then was their inability to find a parking space at the correct location, otherwise it did have a chance of working. Instead it just blew a big hole through several levels of the garage and caused a difficult to fight fire.
Even with a thousand pounds of explosives?
 
Even with a thousand pounds of explosives?
My military engineer's "gut feeling" says "No".

The problem is that you need high explosives to cut steel which demands close contact because the cutting power falls of rapidly with distance away from contact. And the cutting level blast of HE ditto falls off rapidly - same thing stated in reverse.

So at even closest truck bomb distance you are relying on general overpressure blast which is the range where lower explosives - quarrying stuff like ANFO - come into their own. BUT they need surface area to act on and large truck load bombs of ANFO would certainly remove the "wall area" by blowing out panelling - even concrete partitions, floors. ceilings and the like. BUT it would not DIRECTLY cut columns. Maybe tear out a column or two due to the "sail area" of partition wall dragging a column out - probably failing in shear at connections. BUT the structural redundancy of the "Nest" of columns would load redistribute without overall failure.

Then - and especially clear with WTC Twins - it would probably take simultaneous failure of both core and perimeter. And WTC design says that EITHER by itself would not be enough. Remember we are only dealing with a few columns in a concentrated area either core or perimeter IF we have even a well placed single truck bomb...and BOTH core and perimeter if we had TWO well placed truck bombs. AND that is so whether using HE or LE.

My money would be on the Towers surviving.

Again reversing the scenario - If I was the Military Engineer tasked by the General to demolish either or both "Twins" no way would I consider even two well place truck bombs as my plan. Remember most Generals want guaranteed results. ***

....and as a Military Engineer Captain I don't want to spend what may be left of my Military career counting blankets in a stores battalion as the FORMER engineer officer who failed to demolish the WTC Towers. :rolleyes:



(*** An anecdote from RL. About 30 years back the AU Army was rewriting its "Pamphlets" - the "Manuals of Warfare" - and me - Army Reserve officer (Engineers) - was in my Civvy street job in charge of the largest water supply dam in AU. So I was consulted on "how would you blow up this dam?"

I said "I wouldn’t try because it cannot be done by either conventional or non-conventional (including terrorist) war-like activity. Anything short of Nuclear". They had in mind a "Dam Busters" style scenario . Wrong sort of Dam. Details if anyone interested.

BUT the relevant issue here in this thread is that I suggested to back off and re-examine what the objective is...
...and they didn't need to blow up the dam to achieve it.

So ask what was the Terrorist objective in such an attack...and can it be achieved by other means. 'Coz two truck bombs would not IMNSHO have collapsed one of the WTC Twins)
 
Last edited:
Just watching that vid sort of shows the whole 'dropped into it's own footprint' to be not true. As the top floors fall they cant over about 25-30 degrees and seam to be falling at least partly outside the foot print of the building as they are enveloped in the dust cloud. Compare that to any other vids of controlled demolitions and you'll see that the upper stories tend to fall straight, with little or no canting over. To me that says its a natural collapse, (or as natural as you can say after two of the worlds biggest buildings had been rammed by fully loaded jet-liners.)

Indeed a good view of the top floors tilting
I agree the footprint idea is not true. Large parts of the towers are ejected. Pictures of the debris are also not supportive.

Still, strange to see how the initially tilting top of the building is coming straight down through the rest of the building.

Offtopic: is there a topic examining the energy needed to pulverize the building and eject large parts of the debris in mid-air? Thanx.
 
Still, strange to see how the initially tilting top of the building is coming straight down through the rest of the building.
It doesn't seam strange at all, look at the vids closely. if you can slo-mo even better. You see the outer columns fracture in one corner, the failure runs quickly along one side or the tower, and that is the side that starts to fall first canting the top section over towards that side. This due to the fact that for a second or so at least the opposite side from the initial failure is still supporting the upper floors one one side as the other starts to fall . Over the next second or so the structural failure spreads to the other parts of the building, and when these go there is nothing to support the upper floors at all, and continue the canting over, so it then goes straight down.
 
It doesn't seam strange at all, look at the vids closely. if you can slo-mo even better. You see the outer columns fracture in one corner, the failure runs quickly along one side or the tower, and that is the side that starts to fall first canting the top section over towards that side. This due to the fact that for a second or so at least the opposite side from the initial failure is still supporting the upper floors one one side as the other starts to fall . Over the next second or so the structural failure spreads to the other parts of the building, and when these go there is nothing to support the upper floors at all, and continue the canting over, so it then goes straight down.

Wouldn't you expect the top to tilt further and fall outside the blueprint, leaving the rest of the building intact?
 
Not really given the speed the structure failed at. Had the far side structure held on a few seconds more then the upper floors may have tilted further out and fallen in more of an arc, especially if the centre of gravity had tilted outside the outer walls of the structure as a whole. But it didn't get that far, before it had chance to cant right over the remaining supporting structure gave way and it then fell straight smashing the floors beneath it.
 
Goes to my post in another thread.
Imagine a large box hanging in mid-air. Inside the box is a rod on a pivot that is attached to the floor of the box.
The rod begins the experiment almost vertical, say one degree from vertical. Now allow the rod to begin rotating about its fixed pivot.

When the rod has moved 7 degrees from vertical allow the box to beginnfree fall.

Questions: Does the rod continue to pivot?
If it continues to pivot, then; Does it pivot faster, slower or at the same rate as it was when it reached 7 degrees from vertical?

Slightly different set up.
Same as above but now the pivot point is only fastened to the box floor until the box is allowed to drop at which point it is free to move. Assume the pivot is massless.
Same questions as above plus one other. If rotation continues, is rotation still centered on the original pivot end of the rod?
 
My military engineer's "gut feeling" says "No".

The problem is that you need high explosives to cut steel which demands close contact because the cutting power falls of rapidly with distance away from contact. And the cutting level blast of HE ditto falls off rapidly - same thing stated in reverse.

So at even closest truck bomb distance you are relying on general overpressure blast which is the range where lower explosives - quarrying stuff like ANFO - come into their own. BUT they need surface area to act on and large truck load bombs of ANFO would certainly remove the "wall area" by blowing out panelling - even concrete partitions, floors. ceilings and the like.
The truck bomb was underground, no perimeter columns. Concrete foundation. Core columns however did run right down to lowest basement floor.
Idea was to take out enough of the perimeter foundation to cause a 110 storey tilt and overwhelm the ability of the core to withstand the stresses. I would assume wind direction and speed would play a part here. If the upper part of the tower was already swaying with the wind in the same direction you were trying to tilt the entire structure then maybe you have a chance. Wind opposite that might be enough to counter your efforts, at least until the wind dies off.
(*** An anecdote from RL. About 30 years back the AU Army was rewriting its "Pamphlets" - the "Manuals of Warfare" - and me - Army Reserve officer (Engineers) - was in my Civvy street job in charge of the largest water supply dam in AU. So I was consulted on "how would you blow up this dam?"

I said "I wouldn’t try because it cannot be done by either conventional or non-conventional (including terrorist) war-like activity. Anything short of Nuclear". They had in mind a "Dam Busters" style scenario . Wrong sort of Dam. Details if anyone interested.

BUT the relevant issue here in this thread is that I suggested to back off and re-examine what the objective is...
...and they didn't need to blow up the dam to achieve it.

So ask what was the Terrorist objective in such an attack...and can it be achieved by other means. 'Coz two truck bombs would not IMNSHO have collapsed one of the WTC Twins)
I'm going to search for a graphic of where they wanted the truck bomb and where they managed to park it.
,,,and I love anecdotes. I'm going to guess its a concrete encased, thick, earth and rock embankment dam. Probably around 100 feet thick at the top.
To cause havoc you attack the spillway. Crack its bottom surface close to the upper section of the spillway. allowing high flow rate water to erode the earth and rock embankment. Oh, and the control mechanism at the top of the spillway, you want flow to be wide open.

Pic. http://upload.wikimedia.org/wikiped..._spillway.jpg/420px-Llyn_Brianne_spillway.jpg
 
Last edited:
http://www.history.com/news/remembering-the-1993-world-trade-center-bombing

The plotters rented a storage locker in New Jersey, where they stockpiled urea, nitric acid, sulfuric acid and other ingredients for making bombs. They simultaneously concocted a nitroglycerin trigger at a nearby apartment and scouted out the World Trade Center’s underground floors.

Probably getting OT here. This article explains the truck was illegally parked on a ramp.

Still looking for description of their original plan.
 
,,,and I love anecdotes. I'm going to guess its a concrete encased, thick, earth and rock embankment dam. Probably around 100 feet thick at the top.
Try this http://www.sca.nsw.gov.au/water/visit/warragamba-dam - the large side spillway to the left is post my time - later upgrading work. One of my political moments of fame was standing up TWO "Ministers of the Crown" and telling them that the original dam wasn't strong enough. Politicians don't like bad news. My boss lacked the guts to tell them. That "bad news" was what resulted in that big auxiliary spillway.
To cause havoc you attack the spillway. Crack its bottom surface close to the upper section of the spillway. allowing high flow rate water to erode the earth and rock embankment. Oh, and the control mechanism at the top of the spillway, you want flow to be wide open.
Fully concrete gravity dam and 400ft through (horizontally) thick at base. So not vulnerable to erosion processes which could fail an earth and rock-fill structure.

The military objective was "Deny water to Sydney" - attack the pipelines was a lot easier. The method I proposed ended up in the Army Manual. :rolleyes:
 
Ahh, dam busters was to destroy hydroelectric production...permanently(wrt to the war)
Sure --- and that is an example of the point I was making.

Make sure you properly define your objectives......and that is a thread or three on its own

HOWEVER the topic here is "Inwards Bowing" ;)
 
Back
Top