Merged Applicability of Bazant's model to the real world

[leftysergeant]

Again, we see gaseous ejecta with little solid matter falling outside the tube. Here, it is mostly aluminum cladding, possibly shaved off as the steel slides down the inside of the tube. Notice the slight inward buckling at the left hand corner while the right still looks fairly straight. This has to pull the left margin of the failure toward the right, which is also reflected in the movement of the antenna.

 

[Newtons Bit]

The truss connections at the perimeter were indeed strong enough to pull a perimeter column inward before breaking. You are correct there. In fact, there was more strength than just the bolts when the gusset plates and strap anchors are considered along with the bolts.

However, the issue isn't about one set of bolts or one truss connection. NIST needs to disconnect nine out of ten trusses at the perimeter to get the overload needed for the sag they predict. The problem is then that they only have one truss connection to pull about twenty perimeter columns inward, and that can't happen before the connection breaks. That is the real dichotomy in their theory.

No they don't. Bazant has shown that the elements with the absolute greatest vertical load bearing capacity in the structure, the columns, could not resist a drop of the upper block. The floors are weaker than the column in carrying vertical forces, ergo, the floors fail.

This is simple logic, Tony Szamboti.

 

[leftysergeant]

The floors are weaker than the column in carrying vertical forces, ergo, the floors fail.

This is simple logic, Tony Szamboti.

In both of the YT clips I have posted, we also see hoizontal movement in the perimeter columns, most pronounced in WTC 2 but also measureable with the human eye on WTC 1. That means we have blades sliding down inside the tube chopping floors loose ahead of the visible line of failure. I don't thnk that we give that enough importance in any of the discussions so far.

There is no way that the floors arrest the collapse under any circumstances, once the process starts.

 

[leftysergeant]

delete double post.

 

[femr2]

That means we have blades sliding down inside the tube chopping floors loose ahead of the visible line of failure. I don't thnk that we give that enough importance in any of the discussions so far.

As I indicated previously, perimeter *kick-out* has been looked at in considerable detail at the911forum.

Of course the upper East face of WTC2 descended inside the lower, but it is clear that at least the Western half of the upper North face descended outside the lower (which would tend to strip floors from the upper block there, and help destroy the upper block rapidly).

I'll dig out some links for you.

There is no way that the floors arrest the collapse under any circumstances, once the process starts.

I don't think the upper perimeter being inside/outside lower will affect this, but it will affect the regional *crush front* behaviour. Split North-South (WTC1) and West-East (WTC2) would seem to confirm this.

 

[ozeco41]

In both of the YT clips I have posted, we also see hoizontal movement in the perimeter columns, most pronounced in WTC 2 but also measureable with the human eye on WTC 1. That means we have blades sliding down inside the tube chopping floors loose ahead of the visible line of failure. I don't thnk that we give that enough importance in any of the discussions so far....

Spot On!!

There are two complementary factors - first the fact of the "blade effect" and second being that the top block fell inside the outer tube. The latter is true whether or not the top block fell with its outer columns attached and inside the lower tube OR with those columns peeled off because those columns fell outside the lower tube.

I regard those two factors as the most significant issues about the transition from the initial collapse to the global collapse. However I know from personal frustrating experience that you can make those statements/claims in explicit detail and the significance is missed/overlooked/disregarded without comment.

...There is no way that the floors arrest the collapse under any circumstances, once the process starts.

Yes. That is one of the "undeniables" about Twin Towers collapses. The other one for my money is the fact, which should be self evident, that once the initial collapse starts with the top block moving downwards, there simply cannot be any significant number of columns, if any, that are still in axial contact.

Those two set clear definable stages in the collapse sequence which have undeniable consequences for discussion of collapse mechanisms. The second one by itself near fatal to Tony Sz's "missing jolt" which is one reason why he cannot acknowledge the simple truth of it as demonstrated multiple times in this thread.

 

[leftysergeant]

Of course the upper East face of WTC2 descended inside the lower, but it is clear that at least the Western half of the upper North face descended outside the lower (which would tend to strip floors from the upper block there, and help destroy the upper block rapidly).

The lower perimeter columns would probably also strip off upper perimeter columns so that you soon have all of it again falling inside the tube. This is why the collapse becomes more symetrical as it progresses.

 

[ozeco41]

...Of course the upper East face of WTC2 descended inside the lower, but it is clear that at least the Western half of the upper North face descended outside the lower (which would tend to strip floors from the upper block there, and help destroy the upper block rapidly)...

I have found it extremely difficult to get that concept across both here and on my longer term posting forums - the now defunct Richard Dawkins net and rationalskepticism. I personally regard it as the key to understanding the transition from "initial collapse" to "global collapse". One barrier could be the unwarranted in my opinion authority afforded to Bazant et al's "crush down crush up" - this concept cuts right across the validity of "crush down crush up" in the specifics of WTC Tower collapse. So possible we identify another "limit of applicability" for the OP of this thread.

.....I'll dig out some links for you...

...I would appreciate them also.

...I don't think the upper perimeter being inside/outside lower will affect this, but it will affect the regional *crush front* behaviour. Split North-South (WTC1) and West-East (WTC2) would seem to confirm this.

...agreed it matters not which in the broader picture. But it readily explains the availability of enough mass to start the ROOSD/"global collapse" if I use both terminologies - and recognise that core strip down appears to have been a separate process which trailed behind.

 

[femr2]

The lower perimeter columns would probably also strip off upper perimeter columns so that you soon have all of it again falling inside the tube. This is why the collapse becomes more symetrical as it progresses.

Again, not accurate I'm afraid, even though I've just stated issues with your interpretation. (Which of course can be confirmed)

WTC1 - North (for a while) and West (and probably East) upper block perimeter descended outside lower perimeter.

WTC2 - At least half North (and likely half South) upper block perimeter descended outside lower perimeter.

These details have been determined by close inspection of video, and include identifying and keeping track of specific large upper block perimeter sections from initiation to ground.

The split crush fronts do not re-synchronise later in descent, so there is not the symmetry you seem to be describing.

The apparent symmetry comes from the trailing elements of destruction, not the primary ROOSD.

 

[femr2]

One barrier could be the unwarranted in my opinion authority afforded to Bazant et al's "crush down crush up" - this concept cuts right across the validity of "crush down crush up" in the specifics of WTC Tower collapse.

Yes, another reason for attempting to put ROOSD out there.

I view the kick-out process as a rapid mechanism to strip upper block perimeter and *dump* multiple OOS floor masses onto the lower *block*.

I would appreciate them also.

No problem. Need to check with MT whether he has summarised anywhere, or it'll mean picking out relevant posts from lengthy discussions.

...agreed it matters not which in the broader picture. But it readily explains the availability of enough mass to start the ROOSD/"global collapse" if I use both terminologies

Yes, which is why I'm emphasising being accurate about which specific perimeter sections pass inside/outside.

and recognise that core strip down appears to have been a separate process which trailed behind.

Agreed.

 

[ozeco41]

Yes, another reason for attempting to put ROOSD out there...

Having the label of "ROOSD" is effective marketing. I have posted the concept untold times without any specific label and, unlabelled, it is easy to overlook and disregard. Putting the label out front makes it attract attention. Including the attention of those who choose to see it as a "truther" concept. Also MT takes the concept into zoning details which is at least one layer deeper into detail that I ever went.

...I view the kick-out process as a rapid mechanism to strip upper block perimeter and *dump* multiple OOS floor masses onto the lower *block*....

Similarly I viewed the "knife blade" explanations as doing the same thing. Whether upper block stripping OR lower tower stripping was not a significant concern to me - either would put sufficient falling mass into the mechanism. I comprehend why you would prefer greater accuracy. Also we have spoken of "inside" or "outside" the lower tube. There is the possibility of half and half though I tend to see it as unlikely. No strong reason other than the "in" or "out" choice results from gross positioning/alignment of top block to lower tower. So that would tend to leave the side walls of the blocks straight and preclude the "half and half' outcomes.

No problem. Need to check with MT whether he has summarised anywhere, or it'll mean picking out relevant posts from lengthy discussions.

Yes, which is why I'm emphasising being accurate about which specific perimeter sections pass inside/outside.

Agreed.

...all noted.

 

[ergo]

However, the issue isn't about one set of bolts or one truss connection. NIST needs to disconnect nine out of ten trusses at the perimeter to get the overload needed for the sag they predict. The problem is then that they only have one truss connection to pull about twenty perimeter columns inward, and that can't happen before the connection breaks. That is the real dichotomy in their theory.

Tony, I'm not clear on what you're saying here, as my understanding of the NIST theory is that the floors sagged due to "heat", not due to any truss disconnections, then that sagging supposedly caused the pull inward on the perimeter columns.

As described in this video:

http://www.youtube.com/watch?v=-RzwmD8uB8g&feature=player_embedded

esp. 0:59 - 1:20

 

[ergo]

Still a pretty implausible scenario, but I merely wanted to point out the contradiction in the arguments of some here suggesting that the floor to column connections were so weak as to precipitate a global pancaking collapse but were at the same time strong enough to hold while the trusses pull in and snap an entire row of steel box columns, even having been allegedly subjected to heat.

Pancakers here may also want to note the comments at 0:36 - 0:48 in the above linked video.

 

[ergo]

Most of it these days is particle board. It would be pretty finely ground up in that swirling mess of stuff.

What would be "swirling" in a pancaking collapse? I would suggest it's you folks who have difficulties properly visualizing what it is you're trying to describe.

Put a large bug of some sort in a gallon jar half-full of marbles and aquarium gravel and shake it like mad for about twenty seconds. You'll get the idea.

Again, if you are positing a pancaking mode of collapse, we would see a crushing and flattening between the floor pans and assemblies. Not things that have been "shaken" or "swirled".

I suggest you decide what mode of collapse you're arguing, and then stay true to that in your descriptions.

 

[ozeco41]

Still a pretty implausible scenario, but I merely wanted to point out the contradiction in the arguments of some here suggesting that the floor to column connections were so weak as to precipitate a global pancaking collapse but were at the same time strong enough to hold while the trusses pull in and snap an entire row of steel box columns, even having been allegedly subjected to heat.....

It's simple basic engineering. Probably we should have two threads - one for the engineers who know what they are talking about and one for those who question simple basic physics.

Then we could run a "Basic Structural Engineering 001" course on one thread and the real WTC 9/11 discussions on the other.

Beggar it - I have a few minutes - here goes with the basics.

• The trusses did not "...pull in and snap an entire row of steel box columns.." Some trusses pulled some columns out of line. Then those columns were far too weak to hold their applied loads and buckled to failure. The basic engineering concept needed for that being that an axial loaded column can carry a substantial load BUT the load carrying is very sensitive to "eccentric loading" and/or bowing out of line. The two can result from common causes. The load carrying capacity of a column falls off rapidly once either occurs. The visual evidence shows how far out of line some columns were pulled.
  those columns would carry less than 10% of their designed load. Someone may care to do the maths.
• The two effects:
  1)"...were so weak as to precipitate a global pancaking collapse but..";
  2)"...were at the same time strong enough to hold while the trusses pull in..."
  firstly were not "...at the same time..." in the chronological sense - they were different stages of the collapse. However the force needed to pull in a column and applied by tension through a connection is orders of magnitude less than the force applied to the same connector in shear when it is hit by a multi floor overwhelming overload with dynamic impact. Apart from the order of magnitude differences between the forces involved the basic engineering goes to understanding of dynamic impact loads. Impact loading is actually not an Engineering 101 topic - more like Engineering 301. Bottom line is that dropping a ruddy great load suddenly onto a rigid bit of welded on angle will easily shear of that bit of connector.

Those two should do for now.

 

[ergo]

[*]The trusses did not "...pull in and snap an entire row of steel box columns.." Some trusses pulled some columns out of line. Then those columns were far too weak to hold their applied loads and buckled to failure.

This is not what the video I linked to above states. That video clearly states and depicts that the columns snapped due to the strain.

The two effects:
1)"...were so weak as to precipitate a global pancaking collapse but..";
2)"...were at the same time strong enough to hold while the trusses pull in..."
firstly were not "...at the same time..." in the chronological sense -

And I did not say they were.

However the tensile strength needed to pull in a column is orders of magnitude less than the shear strength of the same connector when hit by a multi floor overwhelming overload with dynamic impact. Apart from the order of magnitude differences between the forces involved the basic engineering goes to understanding of dynamic impact loads.

You guys do like your "orders of magnitude", don't you? Do you have some evidence regarding this claim about tensile versus shear strengths of structural steel frame connections? I don't doubt that you do; I just haven't seen it.

 

[Dave Rogers]

One barrier could be the unwarranted in my opinion authority afforded to Bazant et al's "crush down crush up" - this concept cuts right across the validity of "crush down crush up" in the specifics of WTC Tower collapse. So possible we identify another "limit of applicability" for the OP of this thread.

There is a rather broader validity to Bazant's model, though. I agree that the collapse was not sufficiently homogeneous and well-behaved to follow Bazant's differential equations accurately. However, when faced by the wilful misapplication of Newton's Third Law by the "Net Force=0" idiots, it's worth pointing out that there were three components to the collapse, not two. The initial impact was of the upper block on the lower - necessarily, because there were no other components initially - but this impact, damaging both, created debris, and from then on the lower block suffered impacts from both the upper block and the debris. Since the upper block and the debris were both falling under the same gravity and were, to first order, both experiencing the resistive force of the lower block, collisions between the upper block and the debris were of much smaller effect. Therefore, there was an asymmetry between the forces experienced by the upper and by the lower block, leading to a predominance of crush-down over crush-up, from simple reasoning. The ROOSD model is simply enunciating the detail of how this occurred; the basic finding, which is the broader conclusion of Bazant's model, is still valid.

Dave

 

[Tony Szamboti]

Tony, I'm not clear on what you're saying here, as my understanding of the NIST theory is that the floors sagged due to "heat", not due to any truss disconnections, then that sagging supposedly caused the pull inward on the perimeter columns.

As described in this video:

http://www.youtube.com/watch?v=-RzwmD8uB8g&feature=player_embedded

esp. 0:59 - 1:20

I am glad you asked.

Another engineer and I did the calculations for a main double truss at high temperatures with a standard floor load and found that even at 700 degrees C the deflection is only about 2 inches.

700 degrees C is as high a steel truss temperature as NIST could credibly claim.

So how does NIST get the trusses to sag to 40+ inches in their model?

The only way a larger deflection could occur is with more weight. The only way to do that is to have disconnected trusses applying their weight to one that is still connected. Even with the additional load from nine disconnected trusses you only get 25 inches of sag. To get the sag to go to 40+ inches NIST invokes the nebulous notion of creep.

The NIST report claims a large number of the trusses disconnected, but not all. This was obviously done to gain the load needed to show a main truss could sag more than just 2 inches. As I said, there is a dichotomy here that shows the fallacy in their explanation, since if most of the trusses are disconnected to get a few to sag to the 40+ inch point with creep added in, then it begs the question of what would be available to pull in the perimeter columns. The truss sagging is a plausible scenario for NIST and people will buy it if they don't understand that they needed to have nine out of ten trusses disconnected in order to get one to sag that much. I guess we aren't supposed to ask the question I am asking, about what is then left to pull in the columns.

Interestingly, the actual NIST model did not cause perimeter column failure due to sagging trusses so they then applied artificial lateral loads of 5 kips to each perimeter column to get them to bow inwardly and buckle. They say they needed to do this as the model would not have captured all of the forces of the sagging trusses and they used the circular argument of saying they added the 5 kip lateral loads to the model to match the alleged photographic evidence of inward bowing.

I think what Dan Rather said about the collapse of WTC 7 applies to the NIST report's manipulations "Amazing, incredible, pick your word". Finally, it is nothing but unadulterated nonsense.

 

[ozeco41]

This is not what the video I linked to above states. That video clearly states and depicts that the columns snapped due to the strain....

I couldn't care less what some video states. My answer was given in good faith to clarify some basic engineering.

..And I did not say they were...

I was well aware of the ambiguity in the use of "same time" and merely disposed of the chronological option first before answering the one that I thought you meant.

...You guys do like your "orders of magnitude", don't you? Do you have some evidence regarding this claim about tensile versus shear strengths of structural steel frame connections? I don't doubt that you do; I just haven't seen it.

The critical aspect is not the difference between tensile and shear - maybe I should have left those comments out - it is the fact that the 'pull in' force to bow a column is at least two decimal orders of magnitude less than the available impact load of the falling mass. And it is not "you guys liking" rather a fact that for most engineering order of magnitude is a key issue. Most of the analysis of WTC collapse needed to answer the 'demolition or not' questions can be answered by order of magnitude assessments and comparisons.

 

[ozeco41]

There is a rather broader validity to Bazant's model, though. I agree that the collapse was not sufficiently homogeneous and well-behaved to follow Bazant's differential equations accurately. However, when faced by the wilful misapplication of Newton's Third Law by the "Net Force=0" idiots, it's worth pointing out that there were three components to the collapse, not two. The initial impact was of the upper block on the lower - necessarily, because there were no other components initially - but this impact, damaging both, created debris, and from then on the lower block suffered impacts from both the upper block and the debris. Since the upper block and the debris were both falling under the same gravity and were, to first order, both experiencing the resistive force of the lower block, collisions between the upper block and the debris were of much smaller effect. Therefore, there was an asymmetry between the forces experienced by the upper and by the lower block, leading to a predominance of crush-down over crush-up, from simple reasoning. The ROOSD model is simply enunciating the detail of how this occurred; the basic finding, which is the broader conclusion of Bazant's model, is still valid.

Dave

Understood - mostly agreed.

It seems a pity if we need to resort to false or even oversimplified engineering merely to satisfy less competent persons who have mendacious objectives.

I comprehend the three forces not two point that you make, also I am aware of all the misguided or untruthful misapplications of Newton 1, 2 and 3 and several other basic laws which are scattered throughout these threads. I try not to get confused by the confused thinking. I think I succeed most of the time. It gets a bit hard when the analogies suggested are actually impossible to implement.

I am comfortable leaving Bazant et al issues floating rather than distort my analysis into a defence of Bazant. He et al may be taken to task in the academic publishing arena at some stage but I don't hold my breath. I recall a small scale personal episode where, working in a design office, I had the temerity to question one of the "authorities" on which a lot of work had been built. That authority was wrong but I got no thanks for identifying the error. The experience never changed me It was a big organisation. I got transferred to another section shortly after. Never looked back from there....

Meanwhile, back on the topic, I agree with your comments about the inherent asymmetry upper to lower. The art of "simple reasoning" is something which often seems to be derided as "speculation" or some such.

I do get sensitive to comments such as "the basic finding, which is the broader conclusion of Bazant's model, is still valid." which may indicate that defence of Bazant is given higher priority than determining what happened with the WTC collapse. That may not be your intention but I tend to read carefully when similar statements are made.

Cheers.

Eric C