• Quick note - the problem with Youtube videos not embedding on the forum appears to have been fixed, thanks to ZiprHead. If you do still see problems let me know.

Hardfire: Szamboti / Chandler / Mackey

R.Mackey said:
The reason we're not going to cover it here is because it's off-topic, and because Tony has already had it explained to him repeatedly.

You can start a new thread if you want, but the most productive use of your time would be to search for posts by Newtons Bit, where this has been hashed out in detail.

The core error, in brief, is Tony's utterly random and completely indefensible claim that buckling would occur at 95% of linear yield. When buckling occurs is a function of geometry, and there is no geometry in his first principles analysis. NIST's analysis is a full-on finite element simulation of the as-built structure, and is therefore credible. No argument Tony's proposed against it even makes it past the giggle test.

ETA: When two engineers give you the same answer independently, it's usually right. ;)
Click to expand...
Correlary to Asimov's law, do you think:
"If an elderly, but distinguished scientist says that a thing is impossible, is is almost certainly wrong"
CLarke's Corrilary:
"If a distinguished, but elderly scientist states that a thing may be possible, he is likely correct"
Mackey's postulate:
"If two engineers independently give the same answer to a question, it is most probably the correct one"
 
Last edited:
What you've just witnessed is why debating the Truth Movement is a waste of time.

The producer informed me that he should be able to salvage the second show, despite my equipment failures. Probably another week or so until they're done. I'll keep you posted.

I'll also open up a new thread, when the time comes, for future ideas and desires. Personally, I am rather disgusted with how direct debate turns out and would rather do more educational shows, but I will be interested to see what you all think. But that will be another thread.
 
For further reading on the significance of the DCR, I also recommend posts by Architect, most recently this one. The NIST calculations affirm that the safety factor in the core is more like 40-80%, not a factor of 3. The safety factor of 3 is a Heiwa delusion, and he was quite rightly raked over the coals for it.

This officially closes this question. We've learned something. As scientists, we now move on.
 
R.Mackey said:
The reason we're not going to cover it here is because it's off-topic, and because Tony has already had it explained to him repeatedly.

You can start a new thread if you want, but the most productive use of your time would be to search for posts by Newtons Bit, where this has been hashed out in detail.

The core error, in brief, is Tony's utterly random and completely indefensible claim that buckling would occur at 95% of linear yield. When buckling occurs is a function of geometry, and there is no geometry in his first principles analysis. NIST's analysis is a full-on finite element simulation of the as-built structure, and is therefore credible. No argument Tony's proposed against it even makes it past the giggle test.

ETA: When two engineers give you the same answer independently, it's usually right. ;)
Click to expand...

Well a quick look at the text in front of me isn't very telling. End fixity factor? Johnson formula? Column analysis spreadsheets? It's like you need some sorta training to figure this stuff out...

It's hard to learn from the back of the book when you've just read the first few pages. ;)
 
R.Mackey said:
For further reading on the significance of the DCR, I also recommend posts by Architect, most recently this one. The NIST calculations affirm that the safety factor in the core is more like 40-80%, not a factor of 3. The safety factor of 3 is a Heiwa delusion, and he was quite rightly raked over the coals for it.

This officially closes this question. We've learned something. As scientists, we now move on.
Click to expand...

lol
 
R.Mackey said:
What you've just witnessed is why debating the Truth Movement is a waste of time.

The producer informed me that he should be able to salvage the second show, despite my equipment failures. Probably another week or so until they're done. I'll keep you posted.

I'll also open up a new thread, when the time comes, for future ideas and desires. Personally, I am rather disgusted with how direct debate turns out and would rather do more educational shows, but I will be interested to see what you all think. But that will be another thread.
Click to expand...

It's impossible to judge how effective the debate was without seeing it ourselves. But I do think if you're able to get your message out there it's always a good thing.

If you're interested in collaborating on a comprehensive video debunking the Jones/Harrit paper, I'm up for it. Let me know.
 
R.Mackey said:
For further reading on the significance of the DCR, I also recommend posts by Architect, most recently this one. The NIST calculations affirm that the safety factor in the core is more like 40-80%, not a factor of 3. The safety factor of 3 is a Heiwa delusion, and he was quite rightly raked over the coals for it.

This officially closes this question. We've learned something. As scientists, we now move on.
Click to expand...

Just to show Architect does not understand what the NIST did take a look at this comment from the post you mentioned

Core columns in WTC typically had a Demand to Capacity Ratio (DCR) of 0.83, ie a safety factor of 1/0.83=1.20.

Architect is writing it like DCR is a straight reciprocal of factor of safety and in general it is but that isn't how the NIST did it. In their case they incorporated the mimimum factor of safety in the calculation and although the NIST does say that the factor of safety is inherent in the calculation, it might not be clear to those who aren't familiar.

Factor of safety is a straight forward relation of FoS = yield stress/actual stress or FoS = yield load/actual load. The AISC manual has always limited compressive loads to a maximum of 0.6Fy which is a minimum factor of safety of 1.67. No building like the WTC would have a factor of safety against gravity of just 1.20. That is ludicrous.

The reality is that the NIST used an equation like this: DCR = (actual stress x minimum factor of safety)/yield stress with the minimum factor of safety being part of the demand. Now if the mimimum factor of safety was 1.92 with a yield stress of 36,000 psi and an actual stress of 15,562 psi then the NIST DCR would have been shown as 0.83.

In the NIST calculations of DCR how much the DCR is less than 1.00 is telling you how much margin you have even after the minimum factor of safety is satisfied. If the value is 1.00 then it means you have just enough strength to meet the minimum factor of safety.

If you want to find the mimimum factor of safety from the NIST DCRs multiply it by the yield stress and divide by the actual stress. If you want the actual factor of safety then multiply the minimum factor of safety by the reciprocal of the NIST DCR. In this case multiply 1.92 by 1/0.83 which is 1.92 x 1.20 = 2.30

What the NIST did with the DCRs was similar to what we do in aerospace where we use the Margin of Safety which is MoS = [yield stress/(actual stress x factor of safety)] -1. In this case with a minimum factor of safety of 1.92 a yield stress of 36,000 psi and an actual stress of 15,562 psi the Margin of Safety would be MoS = [36,000 psi/(15,562 psi x 1.92)] - 1 = .20, with anything greater than zero meaning there is additional margin above the minimum factor of safety. If it was zero then it would mean there was enough strength to just meet the minimum factor of safety.

Seeing that you bought into what Architect was mistakenly saying Ryan, I am starting to think you don't understand this either. You also treated the NIST DCRs as a direct reciprocal of factor of safety in our debate.

I showed above that with a minimum factor of safety of 1.92 the actual factor of safety in the NIST DCR of 0.83 was 2.30, which would be 130% greater than minimum not 40 to 80%. It is also not hard to see that the NIST could have easily had a difference from the 3.00 I calculated if they had a little greater mass in their model than what Gregory Urich calculates in his analysis. I used Gregory Urich's analysis to do my calculations. I don't think we know what the mass in the NIST model was exactly.
 
Last edited:
A number of polls split to AAH for bickering. Please keep posts on topic and do not personalize the arguments.
Replying to this modbox in thread will be off topic  Posted By: Tricky
 
The actual DCR

If the NIST had not increased the demand by multiplying it by the factor of safety their DCRs for the core columns would have been in the 0.40 to .045 range.

I am not saying what they did was wrong but that one needs to understand just what they did by multiplying the actual demand by the factor of safety in their DCR calculation. Their DCR is thus a measure of how much margin one has beyond the minimum factor of safety. It is not a direct comparison of actual demand to actual capacity. This can be misleading if one doesn't understand their methodology.

A 0.43 DCR is the direct reciprocal of the factor of safety of 2.30 and is much closer to what I calculated using Gregory Urich's mass analysis. It would be interesting to see what mass the NIST actually used, as opposed to what Gregory Urich calculated in his rigorous mass analysis, which corresponds to the mass per unit area seen in other large buildings built in the same era, such as the Sears tower and the John Hancock building. The NIST factor of safety would have the upper block of WTC 1 weighing about 99 million pounds vs. Gregory's 68 million pounds and their mass per unit area would be 907 kg/sq. meter vs. Gregory's 626 kg/sq. meter. The Sears tower has an advertised mass per unit area of 476 kg/sq. meter and the Hancock building a 670 kg/sq. meter. My instinct is that Gregory's analysis is more accurate and the actual factor of safety of the core columns was approximately 3.00.
 
Last edited:
Tony Szamboti
If you were to calculate the buckling stress for each of the core columns about their least radius of gyration using the appropriate KL/r you would find that they all have buckling stresses near the yield stress.

What I was saying was that the buckling stresses of the core columns were in the 95% of the yield stress range because in the case of these relatively stocky columns the actual calculated buckling stress is close to the yield stress.
Click to expand...


Some people do not understand or believe the short version of the following, so here's the long version.

Heiwa, Gage, Chandler et al believe that if the smaller upper block fell onto the larger lower block, the lower block would have arrested the collapse. Since the collapse was not arrested, CD must have been used.

Tony Szamboti believes that if the smaller upper block fell axially onto the larger lower block columns, the lower block columns would have arrested the collapse. Since the collapse was not arrested, CD must have been used​

But This Is Not What Happened. This Is What Happened:​

1) WTC2. (and mutatis mutandis WTC1) It is not possible for all exterior column walls to have failed at the same time, as the combination of the columns and floors damage, redistributed weight and heat damage varied and failures begin at the weakest point first. In this case it was the east side that was weakest and failed first. We know this because the building began the tilt that direction. The airplane hit WTC2 south wall off center and at an angle, pushing flammable debris up against the east wall. The fires cooked this wall and the perimeter columns failure began here.

2) Once the east wall began to fall 12 feet, the upper block pivoted (tilted) down . The tilting upper mass applied a horizontal torque force to the remaining heat and impact damaged vertical columns that bent them, and the columns fractured at the point where the columns changed from vertical to where they were bent. When all the columns fractured, the upper mass stopped tilting and fell vertically down, and the load path changed (because the columns above were not located directly above the columns below, up to 3 feet off center per NIST) from column-above-to-column-below to column-above-to-floor slab-below and the entire upper mass fell, unattached to any of the columns below. Once the tilt occurred and the remainder of the columns fractured, the lower columns below the failure did not carry the weight of the upper block any longer. The top upper block columns punched through the lower floor slabs.

Before and during WTC2 initial collapse showing 3 foot columns displacement.
[qimg]http://911stories.googlepages.com/ST1.jpg/ST1-full.jpg[/qimg]

3) (3.3 feet column spacing x 17.5 feet floor area x 110 lb /sf LL+DL x 29 stories) / (1.2 sf x 1.2 sf of column area) = 128,000 lbs per columns square foot (static load only) landed onto the 2.5” to 4” thick floor slab x 14" x 14" area. The floor slab was designed to support (110 lb DL+LL /SF x assumed 3.0 static load safety factor = 330 lb / SF) and failed (128,000 lb/sf column load > 330 lb/sf slab design load ). When all the columns above fractured , the block stopped tilting, and they punched through the slab below. The displaced columns above did not hit the off center columns below.

4) We know the floor slabs failed first before the columns failed because of the visual evidence . Had the columns below failed before the floor slabs, the columns would have failed in buckling (or decapitated by CD) with portions of the floor structures still attached to the columns and the perimeter wall assemblies would have fallen vertically nearly “onto its own footprint”. The floor slabs pancaked, yes pancaked, before the unbraced columns failed in buckling at the bolted perimeter connections (36 feet oc) and welded core column locations (36 feet oc). There is no evidence of universal buckling of columns at the column body that would indicate column crushing forces or columns decapitation that would show CD. After the floors pancaked, the unsupported perimeter and core columns failed at the connections and toppled onto the ground.

Picture of WTC1 uncrushed columns, floors failed first then columns toppled outward, not onto “its own footprint” had the load path been axially through the perimeter columns or CD'd.
[qimg]http://www.911myths.com/assets/images/columns.jpg[/qimg]

5) Therefore it doesn’t matter if “ you would find that they all have buckling stresses near the yield stress.” because the columns below did not fail by crushing or CD. They failed by pancaked floors and toppling.

6) Therefore it doesn’t matter if “the 3.00 to 1 factor of safety for the core columns was calculated using Gregory Urich's mass analysis.” because the columns below did not fail by crushing or CD. They failed by pancaked floors and toppling.

7) Therefore it doesn’t matter if ” the columns on each story were designed to have the same unit stress to preclude differential deflections and floor warpage between the core and perimeter.” because the columns below did not fail by crushing or CD. They failed by pancaked floors and toppling.

8). Therefore it doesn’t matter if “ the NIST used a factor of safety of 1.67 to 1.92 against yield and buckling for the core columns.” because the columns below did not fail by crushing or CD. They failed by pancaked floors and toppling.

9) Therefore it doesn’t matter if “you were to calculate the buckling stress for each of the core columns about their least radius of gyration using the appropriate KL/r” because the columns below did not fail by crushing or CD. They failed by pancaked floors and toppling.

...
----------------------------------------------------------------------------------------------------------

There is great confusion about Bazant’s collapse theory.
​

“For our purpose, we may assume that all the impact forces go into the columns and are distributed among them equally. Unlikely though such a distribution may be, it is nevertheless the most optimistic hypothesis to make because the resistance of the building to the impact is, for such a distribution, the highest. If the building is found to fail under a uniform distribution of the impact forces, it would fail under any other distribution.”​

[qimg]http://911review.com/coverup/fantasy/imgs/figure4.gif[/qimg]
From Bazant - Fig. 4. Scenario of tilting of upper part of building ~South Tower - showing horizontal bending forces and displacement of upper and lower columns.

From Bazant’s first paper.
http://www.civil.northwestern.edu/people/bazant/PDFs/Papers/405.pdf



Summary:
WTC1,2 failed by gravity collapse only.
On the weakest wall the perimeter columns failed, the upper stories tilted 3 feet (WTC2), the remainder of the columns fractured, the tilt stopped , the upper stories fell straight down, the displaced columns above punched through the floor slab below - not hit the columns below, the floor slabs pancaked, and the unsupported perimeter and core columns then toppled to the ground.​

-----------------------------------------------------------------------------------------------------------------------------
You can’t reason someone out of something they were not reasoned into. -Swift
 
3bodyproblem said:
I'm sure you've gone over this before, but a little more explanation of l/d and slenderness ratio would be cool for the rest of us willing to learn a thing or two.
Click to expand...

R.Mackey said:
The reason we're not going to cover it here is because it's off-topic, and because Tony has already had it explained to him repeatedly.
Click to expand...

R.Mackey said:
And, again, you're assuming compressive fracture instead of buckling failure, which is wrong, or you're assuming pinned connections, which is also wrong. Been over this with you repeatedly and told you so again during the debate. NIST doesn't make these mistakes and is therefore entirely credible.
Click to expand...

I'm not sure what's off topic? The OP suggests the thread is about the debate and questions brought up therein.
 
Sigh. DCR, one more time.

The DCR that NIST computes is the Demand to Capacity Ratio, where the Demand is the actual load (the "in service load"), and the Capacity is the design load, which incorporates a factor of safety mandated by the ASCE7 AISC standard.

NIST's calculation is not a first-principles analysis, like Tony carried out. It is based on the actual SAP2000 model, and that means modeling the materials, the geometry, and the weight. NIST's estimate of the service load was based on material dimensions for the dead load, original WTC construction documents for the applied dead load, and the NYCBC 2001 estimate of average office contents for the live load. This weight, incidentally, is consistent with Gregory Urich's computations, so Tony claiming a difference here is another mistake on his part. NIST also included the wind loads as "in service," and it is important to note this is somewhat different from the wind loads estimated at the time of design.

There were two estimates of loading provided, one corresponding to the original design case, and the other, more relevant "state of the practice" case which primarily considers an ordinary wind load. The distinction is unimportant since the core loads are practically the same between them.

NIST found, as described in NCSTAR1-2A Table 5-3, that the DCR of the core columns averaged to 0.86, and that 9.9% of core columns exceeded a DCR of 1.0.

Now, to translate this to Factor of Safety, we need to understand what factor was built into these design equations. As AISC describes and was confirmed by NIST in its 2004 conference, the relevant FoS for the core situation is 1.67. So, to compute the effective, in-service FoS, we simply compute 1.67 / 0.86 = 1.94.

Not 3.0.

The Factor of Safety = 3 is a lie, and I called Tony on it during the debate.

This Factor of Safety was further reduced by impact, as you can read in NCSTAR1-6D. This is the parameter I referred to in our debate, as you will see when it's complete.

It's also totally irrelevant to the collapse, once again, because the mechanism of collapse removes supporting elements from the columns, and this reduces or eliminates their strength. The bulk of debris does not land squarely on the columns in the first place. The more relevant resistance is that of the floor system, which can support -- according to Tony -- a maximum 29 million pounds, or less than 45% of the mass of the upper block at the start of the collapse, and getting steadily worse as the descending material snowballs.

This is all information that has been presented here in the JREF Forums before, and available for years. I took advantage of it. I've learned. Tony has not. This is why debating him was rather disappointing.
 
Last edited:
Edited by LibraryLady: 
Edited for civility. Attack the argument, not the arguer.


R.Mackey said:
Sigh. DCR, one more time.

The DCR that NIST computes is the Demand to Capacity Ratio, where the Demand is the actual load (the "in service load"), and the Capacity is the design load, which incorporates a factor of safety mandated by the ASCE7 standard.

NIST's calculation is not a first-principles analysis, like Tony carried out. It is based on the actual SAP2000 model, and that means modeling the materials, the geometry, and the weight. NIST's estimate of the service load was based on material dimensions for the dead load, original WTC construction documents for the applied dead load, and the NYCBC 2001 estimate of average office contents for the live load. This weight, incidentally, is consistent with Gregory Urich's computations, so Tony claiming a difference here is another mistake on his part. NIST also included the wind loads as "in service," and it is important to note this is somewhat different from the wind loads estimated at the time of design.

There were two estimates of loading provided, one corresponding to the original design case, and the other, more relevant "state of the practice" case which primarily considers an ordinary wind load. The distinction is unimportant since the core loads are practically the same between them.

NIST found, as described in NCSTAR1-2A Table 5-3, that the DCR of the core columns averaged to 0.86, and that 9.9% of core columns exceeded a DCR of 1.0.

Now, to translate this to Factor of Safety, we need to understand what factor was built into these design equations. As ASCE7 describes and was confirmed by NIST in its 2004 conference, the relevant FoS for the core situation is 1.67. So, to compute the effective, in-service FoS, we simply compute 1.67 / 0.86 = 1.94.

Not 3.0.

The Factor of Safety = 3 is a lie, and I called Tony on it during the debate.

This Factor of Safety was further reduced by impact, as you can read in NCSTAR1-6D. This is the parameter I referred to in our debate, as you will see when it's complete.

It's also totally irrelevant to the collapse, once again, because the mechanism of collapse removes supporting elements from the columns, and this reduces or eliminates their strength. The bulk of debris does not land squarely on the columns in the first place. The more relevant resistance is that of the floor system, which can support -- according to Tony -- a maximum 29 million pounds, or less than 45% of the mass of the upper block at the start of the collapse, and getting steadily worse as the descending material snowballs.

This is all information that has been presented here in the JREF Forums before, and available for years. I took advantage of it. I've learned. Tony has not. This is why debating him was rather disappointing.
Click to expand...
 
Last edited by a moderator:
R.Mackey said:
Sigh. DCR, one more time.

The DCR that NIST computes is the Demand to Capacity Ratio, where the Demand is the actual load (the "in service load"), and the Capacity is the design load, which incorporates a factor of safety mandated by the ASCE7 AISC standard.
Click to expand...


I admit, most of this discussion is way outside of my area of expertise, but I do have a question. Are all of these calculations of DCR and safety factors meant to apply to the columns immediately below the area of impact? In other words, is Mr. Szamboti claiming that without additional assistance, the collapse would have been arrested immediately at the point of failure?

If so, regardless of the design DCR and safety factor, does it really make sense to assume those specific column segments in that area are actually performing to their design specifications?
 
This is as-built. That's what Tony's referring to, and it is useless on top of useless even if he'd get it right instead of inflating the numbers.

NIST in NCSTAR1-6D looks at the DCR of columns in the damaged section, particularly Figures 4-68 through 4-71 of NCSTAR1-6D, demonstrating that the impact and the fire afterward has a profound effect on the core. And these figures don't take into account additional weakening from creep and vertical displacement.

To treat both, it is more useful to look at strain than load and capacity, such as is done in 4-72 through 4-81. The strain conditions at the end of the run are of a structure at imminent failure no matter how you slice it. Tony's calculations, of course, don't address this at all.
 
R.Mackey said:
I have a rather shocking fluorescent green shirt and jittering paisley necktie combination. Wear it to weddings, mostly. It's quite nice, in its own way, though it tends to damage recording equipment.
Click to expand...

Please tell me you didn't wear this?
 
R.Mackey said:
This is as-built. That's what Tony's referring to, and it is useless on top of useless even if he'd get it right instead of inflating the numbers.

NIST in NCSTAR1-6D looks at the DCR of columns in the damaged section, particularly Figures 4-68 through 4-71 of NCSTAR1-6D, demonstrating that the impact and the fire afterward has a profound effect on the core. And these figures don't take into account additional weakening from creep and vertical displacement.

To treat both, it is more useful to look at strain than load and capacity, such as is done in 4-72 through 4-81. The strain conditions at the end of the run are of a structure at imminent failure no matter how you slice it. Tony's calculations, of course, don't address this at all.
Click to expand...


That's what I thought, thanks.
 
You must log in or register to reply here.

Back
Top Bottom