AE911 Truth's WTC7 Evaluation Computer Modelling Project

Status
Not open for further replies.
Please don't delete this again Mick as it is very on topic.
These break elements are crucial and I think it would inform us to know what type of element NIST used (COMBIN37) and how that may have changed the analysis. I would also then like to put Dr Hulsey's model under the same scrutiny to compare them. I am going through the notes at the moment for any of Dr Hulsey's model details that might be there.
But I think it would do no harm to establish just how many directions a unidirectional element can move in.
As I said earlier, I think it's one, but I am prepared to be wrong.
But I don't think you are ever going to know exactly how that connection was set up. Unless you've got some hard evidence then it's pointless speculation.
I've gone through the notes, they had build some simple models and bent them and heated them a bit. They were trying to figure out how to model the connections. The last connection related entry is:

Date: 08/21/2015 Working Hours: 8
Submitted by Zhili Quan

Today, “1 joint link” elements were studied. However, despite tons of effort being put into modeling the elements to simulate the stiffness between multiple elements that have one of their joints that share the same location, it cannot be done. Later, after a phone call consultation session with the company that provides SAP2000, it was found out that “1 joint link” elements can only be used to connect elements to the ground. Therefore, it cannot be used to simulate the stiffness between different elements that are connected.

“2 joint link” elements were also studied. After numerous test-modeling, it was found that a “2 joint link” element can be put into one shared location of two different joints of two different elements therefore giving the stiffness between these two elements.

Figure 1 shows a triangular “shell” element with one of its joints that shares the same location as one of the joints of a “frame” element. A “2 joint link” element has been inserted between these two elements.
However, when there are multiple elements that each has one of their joints sharing the same location as what is shown in Figure 2. The highlighted location has 6 different joints all clustered together. It takes very complicated connecting condition to model the partially composite elements. More study has to be conducted to resolve this issue.
Content from External Source
Then two weeks later they had another hard drive crash, and nothing was heard from them since.
 
I just watched the last "Lab Video" from August 5, 2015. People probably don't watch them because they are boring and hard to understand with the accents and technical details. But since this was the last one I thought I'd give it a go.


Source: https://www.youtube.com/watch?v=nsLCaFMvOU0&t=85s


It did not inspire confidence. There's a whole almost comedic bit where Hulsey seems to misunderstand Kelvin vs. Celsius.
Video at 08:27

Hulsey: How many degrees C is there for one Kelvin?
Zhili: one?
Hulsey: what?
Zhili: one. I mean the relationship isn't, I mean the slope
Hulsey: Not one! It is 273 degrees C per Kelvin I believe
Zhili: I mean the increments the same
Hulsey: I know i understand but take a look online, let's take a look at kelvin which is absolute and what we want is what's the relationship between degrees kelvin and degrees C. I think it's 273
Zhili: yeah plus something but
Hulsey: well I'm not sure
Zhili: it's plus something well it depends, I mean, Celsius Equals Kelvin minus 1, 27 .. degrees
Feng: So, one C is [Feng has looked up 1°C in K, getting 274.15°K]
Hulsey: So it's one, see, there's no add on! It's 274.15 not 273, I thought it was 273. So anyway that's a direct relationship. You've got C, you add 274.5 and you've got degrees kelvin.
So let's assume you have 500 degrees Kelvin, how many degrees C is that?
[Feng looks it up, it's 226 degrees C]
Hulsey: See, that's just above boiling water, right? 100 degrees C is boiling.
Content from External Source
In fact to convert C to K you add 273.15, a number that was never mentioned, and one Kelvin IS one degree C.

Clearly there was a bit of a communication problem there. Hulsey was confused (he's asking the wrong question, he wants to know what the temperature in Kelvin is at 0°C, but instead asks "How many degrees C is there for one Kelvin", which would make sense if he were asking about Fahrenheit)). His students were trying to respectfully explain things to him as he smoothly transitions for "Not one, it's 273", to "So, it's one, see!" as if he was right all along. Then boldly notes that 226°C is just above boiling (again confusing C and F) then instantly notes 100°C is boiling, without acknowledging any mistake.

The problem here is that people have touted his qualifications as a part of the evidence. If Hulsey says it, then since he's so well qualified, then it must be true, right? However this example does not give much confidence that his qualifications continue to be relevant at this date. He seems incapable of recognizing his own mistakes.
 
Last edited:
But I don't think you are ever going to know exactly how that connection was set up. Unless you've got some hard evidence then it's pointless speculation.
I've gone through the notes, they had build some simple models and bent them and heated them a bit. They were trying to figure out how to model the connections. The last connection related entry is:

Date: 08/21/2015 Working Hours: 8
Submitted by Zhili Quan

Today, “1 joint link” elements were studied. However, despite tons of effort being put into modeling the elements to simulate the stiffness between multiple elements that have one of their joints that share the same location, it cannot be done. Later, after a phone call consultation session with the company that provides SAP2000, it was found out that “1 joint link” elements can only be used to connect elements to the ground. Therefore, it cannot be used to simulate the stiffness between different elements that are connected.

“2 joint link” elements were also studied. After numerous test-modeling, it was found that a “2 joint link” element can be put into one shared location of two different joints of two different elements therefore giving the stiffness between these two elements.

Figure 1 shows a triangular “shell” element with one of its joints that shares the same location as one of the joints of a “frame” element. A “2 joint link” element has been inserted between these two elements.
However, when there are multiple elements that each has one of their joints sharing the same location as what is shown in Figure 2. The highlighted location has 6 different joints all clustered together. It takes very complicated connecting condition to model the partially composite elements. More study has to be conducted to resolve this issue.
Content from External Source
Then two weeks later they had another hard drive crash, and nothing was heard from them since.

I just skimmed it. The above is about how composite the elements are with each other using a friction value.
I am not seeing anything at all to do with the actual connection elements in the notes. Pity there's not more ABAQUS stuff in there.
Oh, and never lend these guys a hard drive.
 
...

Then two weeks later they had another hard drive crash, and nothing was heard from them since.

Here's a question Feng asked re ABAQUS in January 2016:


ABAQUS Plastic - Elastic Question?
Hello,

I have built a steel connection, and it works well in linear condition by using Static, General Step, however, it cause problems when I added material behavior with plastic.

Could you tell me what can cause those problems (have refined mesh, and reduced the initial step, but it doesn't help):

1.The strain increment has exceeded fifty times the strain to cause first yield at 32 points.

2.The strain increment is so large that the program will not attempt the plasticity calculation at 91 points.

3. Too many attempts made for this increment.

Thanks,

Feng

Content from External Source
It appears one response has been deleted and next Feng replies:


Thank you for your answer. I have just figure out it because of stable problem.

The model is stable in linear condition, but not for nonlinear.

And I added some boundary condition then it works.

Best,
Content from External Source
Two more responses then question this resolution:

Angus Ramsay · Ramsay Maunder Associates Limited
Sounds like Finite Element Malpractice to me! The boundary conditions for the plastic case should be identical to the elastic case...
Content from External Source
and


Raafat E S Ismail · Beirut Arab University
i think in your model there is a severe local yeilding due to stress concentration where applied load and /or support are prescribed. first you check your result to find its location then if so try to use regid plate to avoid it.

Content from External Source
It looks like there was still a learning curve to climb several months after the last journal entry...
 
It looks like there was still a learning curve to climb several months after the last journal entry...
Your point ?
Everyone has to learn, and if you stop doing it you start forgetting.
Students climbing learning curves is great. It's the ones that don't that aren't.
 
Your point ?
Everyone has to learn, and if you stop doing it you start forgetting.
Students climbing learning curves is great. It's the ones that don't that aren't.

I think it's great that they were learning. But the learning curve they faced begs the question, which Mick has raised before, as to why they, under the direction of a bridge expert, were the best people to carry out this particular project when there are many people in the world who have already climbed that learning curve.
 
Last edited:
I think it's great that they were learning. But the learning curve they faced begs the question, which Mick has raised before, as to why they, under the direction of a bridge expert, were the best people to this particular project when there are many people in the world who have already climbed that learning curve.
You should show Dr Hulsey more respect. you are aware of his qualifications so I won't repost them.
 
I just skimmed it. The above is about how composite the elements are with each other using a friction value.
I am not seeing anything at all to do with the actual connection elements in the notes. Pity there's not more ABAQUS stuff in there.
Oh, and never lend these guys a hard drive.
If only they had kept the public informed even after the fall of 2015, so that we, the public, could now monitor the progress!
 
I agree, esp if private donors are paying for it.

And where would you suggest as an alternative? I think the project represents good value.

I think, for me, the point is understanding the software and software limitations is important in studies. I would feel more confident in their results if they had an experienced mentor helping them.
Which they do. Dr Hulsey.
What is important to me are things like....
Including partial height stiffener plates.
Knowing the difference between 11" and 1ft when it's written on a bill of materials on paper in front of you.
Including lateral beam supports.... the list goes on.

[response to deirdres off topic comment removed]
 
You should show Dr Hulsey more respect. you are aware of his qualifications so I won't repost them.
I'm aware that he has made sweeping conclusions (a year ago) that he currently is not sure about 100%. he said so himself.
So I would think he was even more unsure a year ago when he first made the conclusive statement. You gotta admit it doesn't bode well.
 
I'm aware that he has made sweeping conclusions (a year ago) that he currently is not sure about 100%. he said so himself.
So I would think he was even more unsure a year ago when he first made the conclusive statement. You gotta admit it doesn't bode well.
What specific statement?
And I believe in going where the evidence takes you. If the model shows that fire can do it, for example, then the project will have answered the question and potentially enhanced public safety. So who would object to that?
It's sure not going to conclude that NIST were right about the girder initiation, whatever happens. The girder's stuck there.
 
That [his study proves] fires could not have caused the collapse of wtc7
He knows that a falling girder won't progress the collapse by failing the connection directly below it. Not unless you mess with the stiffness constant. Given how much further down this road he is now, he clearly remains confident in his statement.
 
He knows that a falling girder won't progress the collapse by failing the connection directly below it. Not unless you mess with the stiffness constant. Given how much further down this road he is now, he clearly remains confident in his statement.

Depends on the condition of the girder below it.
 
Depends on the condition of the girder below it.

This 215,211 lb. impact force is only 34% of the 632,000 lb. force required and thus insufficient to shear the girder bearing seat support welds. The northeast corner of floor 12 would not have collapsed if a girder at floor 13 came off its seat at column 79 and fell onto it. Thus the ARUP analysis does not show a basis for propagation, even if the girder were to fall off its seat at column 79. These findings illustrate that ARUP’s explanation is invalid.



The NIST WTC7 report also states that the falling girder would break through the next floor down. Therefore the NIST explanation is also invalid on this ground.
Content from External Source
[link add: ine.uaf.edu/media/92216/wtc7-structural-reevaluation_progress-report_2017-9-7.pdf]

So you agree with the above? It's just the building condition directly below that concerns you ?
 
Last edited by a moderator:
You should show Dr Hulsey more respect. you are aware of his qualifications so I won't repost them. (gerrycan)



Actually, as a non-scientist myself, I am especially deferential and open-minded towards those who are.
I made the following 100% sincere inquiry (Post #80 12/9/15) almost 2 years ago in this very thread:

**************************************************************
Is anyone aware of any rationale for the choice of Dr. Hulsey for this project?
I see a lot of work with bridges on his resume, but not much that would explain why he would
be among the better candidates for an inquiry into WTC 7.

I'm not disparaging him, just wondering if any explanation has been given... (?)
**************************************************************


I'm still not clear as to which of his specific qualifications should give us solid confidence
in his assertions. I don't think you were here in Dec. '15 (?) Gerry, so I would welcome your
views on this now, since you've made respect an issue.
 
You should show Dr Hulsey more respect. you are aware of his qualifications so I won't repost them. (gerrycan)



Actually, as a non-scientist myself, I am especially deferential and open-minded towards those who are.
I made the following 100% sincere inquiry (Post #80 12/9/15) almost 2 years ago in this very thread:

**************************************************************
Is anyone aware of any rationale for the choice of Dr. Hulsey for this project?
I see a lot of work with bridges on his resume, but not much that would explain why he would
be among the better candidates for an inquiry into WTC 7.

I'm not disparaging him, just wondering if any explanation has been given... (?)
**************************************************************


I'm still not clear as to which of his specific qualifications should give us solid confidence
in his assertions. I don't think you were here in Dec. '15 (?) Gerry, so I would welcome your
views on this now, since you've made respect an issue.

Dr. Hulsey focuses on working with students to achieve a high quality education. His research is in the fields of bridge engineering and effects of temperature extremes on structural systems like composite wall panels for buildings. His work blends a strong experimental component with the fundamentals of theory. Dr. Hulsey has expertise in mathematical modeling using state-of-art methods in finite element, finite difference and theoretical solid mechanics. Funding isavailable for graduate students to work on the analysis or testing of bridges.
Content from External Source

  • 1976 Ph.D. Structural Engineering, University of Missouri-Rolla
  • 1968-1971 Post Graduate, University of Illinois
  • 1966 M. S. Civil Engineering, University of Missouri at Rolla
  • 1965 B. S. Civil Engineering, Missouri School of Mines and Metallurgy
Content from External Source
The rest of his CV is here http://cem.uaf.edu/cee/people/leroy-hulsey.aspx
It's extensive, as you are no doubt aware.
 

This 215,211 lb. impact force is only 34% of the 632,000 lb. force required and thus insufficient to shear the girder bearing seat support welds. The northeast corner of floor 12 would not have collapsed if a girder at floor 13 came off its seat at column 79 and fell onto it. Thus the ARUP analysis does not show a basis for propagation, even if the girder were to fall off its seat at column 79. These findings illustrate that ARUP’s explanation is invalid.



The NIST WTC7 report also states that the falling girder would break through the next floor down. Therefore the NIST explanation is also invalid on this ground.
Content from External Source
[link add: ine.uaf.edu/media/92216/wtc7-structural-reevaluation_progress-report_2017-9-7.pdf]

So you agree with the above? It's just the building condition directly below that concerns you ?

How much force would it take to unseat the lower girder if it were damaged to the extent it actually damaged was on 9-11? Please explain exactly how you determined how damaged it was. If you can accept one girder can become damaged enough from fires to fail and fall, pretty silly to deny another girder cannot become damaged enough to fail given similar fires + the first falling girder.
 
...the stiffness constant...
Stiffness is not constant in a building that experiences devastating fire and accumulating damage, nor is the strenth of connections.

Since Hulsey does not run fire simulations, his study cannot possibly come to the global negative conclusion.
 
Dr. Hulsey focuses on working with students to achieve a high quality education. His research is in the fields of bridge engineering and effects of temperature extremes on structural systems like composite wall panels for buildings. His work blends a strong experimental component with the fundamentals of theory. Dr. Hulsey has expertise in mathematical modeling using state-of-art methods in finite element, finite difference and theoretical solid mechanics. Funding isavailable for graduate students to work on the analysis or testing of bridges.
Content from External Source

  • 1976 Ph.D. Structural Engineering, University of Missouri-Rolla
  • 1968-1971 Post Graduate, University of Illinois
  • 1966 M. S. Civil Engineering, University of Missouri at Rolla
  • 1965 B. S. Civil Engineering, Missouri School of Mines and Metallurgy
Content from External Source
The rest of his CV is here http://cem.uaf.edu/cee/people/leroy-hulsey.aspx
It's extensive, as you are no doubt aware.
So he has been paid for forensic investigations re. fire and large building failures before...yes?
 
Last edited:
How much force would it take to unseat the lower girder if it were damaged to the extent it actually damaged was on 9-11?
What the damage was in the model that NIST used to sell us this tale is what matters. Personally, I am all for ramping conditions up until collapse (theoretically) progresses, then seeing just how far past 10 I had to turn it up by.
Please explain exactly how you determined how damaged it was.
Theoretically, I would have looked at NIST's damage pattern for their model, prior to their alleged initiation.
If you can accept one girder can become damaged enough from fires to fail and fall,
I don't. nobody has evidenced that happening yet in a model, or in reality.
pretty silly to deny another girder cannot become damaged enough to fail given similar fires + the first falling girder.
Yeah that would be silly. Although I do accept that if the girder above is most damaged, and cannot be proven to fall due to fire, then the lower girder should be fine.
 
Dr. Hulsey focuses on working with students to achieve a high quality education. His research is in the fields of bridge engineering and effects of temperature extremes on structural systems like composite wall panels for buildings. ...
Content from External Source
...
This most likely refers to weather extremes - conditions typical for arctic projects.

His complete lack of experience with fire conditions is probably best illuminated by his choice to not even do any fire simulations.
 
...in the model that NIST used to sell us ...
You are mixing up things, it seems.
NIST did not sell us anything, or did they ever ask you for MONEY?
Hulsey sold a study to AE911truth - the price tag was 316,000 dollars (and probably counting), and they are trying to further sell it to us by calling for more MONEY.
 
What the damage was in the model that NIST used to sell us this tale is what matters. Personally, I am all for ramping conditions up until collapse (theoretically) progresses, then seeing just how far past 10 I had to turn it up by.
Theoretically, I would have looked at NIST's damage pattern for their model, prior to their alleged initiation.

I don't. nobody has evidenced that happening yet in a model, or in reality.

Yeah that would be silly. Although I do accept that if the girder above is most damaged, and cannot be proven to fall due to fire, then the lower girder should be fine.

Great--a whole bunch of "I don't knows". Remarkable that just a few lines ago you knew enough to declare the NIST model invalid on this point.
 
Great--a whole bunch of "I don't knows". Remarkable that just a few lines ago you knew enough to declare the NIST model invalid on this point.
= You don't like the answers you got.
There's no "I don't knows" in there. Address my last point please.
"I do accept that if the girder above is most damaged, and cannot be proven to fall due to fire, then the lower girder should be fine."
 
= You don't like the answers you got.
There's no "I don't knows" in there. Address my last point please.
"I do accept that if the girder above is most damaged, and cannot be proven to fall due to fire, then the lower girder should be fine."

I don't care what you accept. I care what you can prove.
 
I don't care what you accept. I care what you can prove.
Well, so far I have proven that a small research team can take an issue like this and find some good patriotic Americans who are open minded enough to question their government's story.

Do you think that UAF should release their data inputs and model details ?
 
if the girder above is most damaged, and cannot be proven to fall due to fire,
This "if" has not actually been proven by Hulsey. His model is woefully incomplete and thus incapable of approximating the real-world geometry changes of the assembly due to the fire histories on 6 nearby floors.

then the lower girder should be fine."
Even if the condition above held true, then this does not follow, as the lower girder, too, was subjected to a long time of extreme fire exposure and quite surely not "fine".

Hulsey and you are too focused on a single detail, you miss the bigger picture.
 
Well, so far I have proven that a small research team can take an issue like this and find some good patriotic Americans who are open minded enough to question their government's story.

Do you think that UAF should release their data inputs and model details ?

I think they should release a higher level of detail than they have. I don't care about having their actual models if they describe them sufficiently. Even based on the level of detail we have, it is easy enough to spot glaring errors.

Do you think Hulsey's team should have actually modeled NIST's heating scenario if trying to test NIST's theory?
Do you think Hulsey's team should have modeled more than two corners of two floors?
Do you think Hulsey's team should have treated everything outside of those two corners of two floors as fixed?
For the temperatures Hulsey's team did apply to the building, even accepting that they failed to model NIST's heating scenario, do you think they should have chosen those temperatures from a time after the key elements they were modeling had begun to cool according to NIST's model?
 
Last edited:
I think they should release a higher level of detail than they have.
Great. Just checking that you don't think it could jeopardise public safety.
I don't care about having their actual models if they describe it sufficiently. Even based on the level of detail we have, it is easy enough to spot glaring errors.
So you don't have enough detail, but you dismiss it. At least you have the details from the drawings and you know the elements are correct and present.
Do you think Hulsey's team should have actually modeled NIST's heating scenario if trying to test NIST's theory?
Yes. They should replicate NIST's stated conditions in order to make a comparison. Whilst making sure that they get the structure correct as per the drawings, unlike NIST.
Do you think Hulsey's team should have modeled more than two corners of two floors?
Yes. If they were analysing more than 2 corners on 2 floors they should have.
Do you think Hulsey's team should have treated everything outside of those two corners of two floors as fixed?
No, and I think that the equivalency method will serve them well in that regard. Interesting that they did small scale simulation runs on those 2 particular areas, isn't it.
For the temperatures Hulsey's team did apply to the building, do you think they should have chosen those from a time after the key elements they were modeling had begun to cool?
No, they should have chosen whatever particular temperature they wanted to replicate in the structure and used that.
 
wonder if the UAF study showed that the K3004 girder also expanded to the East slightly, as can be seen would happen below, when you examine the (K3004 - column 38) connection.C38.jpg (NIST didn't)
 
Last edited by a moderator:
How much force would it take to unseat the lower girder if it were damaged to the extent it actually damaged was on 9-11? Please explain exactly how you determined how damaged it was. If you can accept one girder can become damaged enough from fires to fail and fall, pretty silly to deny another girder cannot become damaged enough to fail given similar fires + the first falling girder.
The 2" thick x 14" deep x 18.875" wide support plate below the bearing seat is the item which would need to fail for the 12th floor girder to be collapsed by an impact from the girder above.

That plate was welded to column 79 with 3/8" fillet welds on both sides (28 inches of weld). I have attached the drawing which shows the plate and its welds on column 79. The plate has the material code of "PG" in the drawing and the weld type, size, and location is called out. These welds would need to be sheared and that shear force is 632,000 lbs..

The support plate was welded to column 79 which never got hotter than 300 C in the fire simulation. The support plate would have been at approximately the same temperature as column 79. At 300 C steel has not yet lost any yield strength, so the force required would still have been about 632,000 lbs.

The impact could not have provided more than about a 1/3rd of that force.
 

Attachments

  • 1091.pdf
    2.1 MB · Views: 657
Last edited:
For the temperatures Hulsey's team did apply to the building, even accepting that they failed to model NIST's heating scenario, do you think they should have chosen those temperatures from a time after the key elements they were modeling had begun to cool according to NIST's model?
You keep trying to say Hulsey did not use all of NIST's fire simulations as though that presents a problem. Your logic is flawed here, as Hulsey used NIST's worst case. By definition that envelopes all other situations.

An analogy for others to understand is that if one were to do a test on what the forces were during the torquing of a lug nut to failure all one ultimately needs to use in the analysis is the worst case force, which would be that which occurred at failure. It envelopes all of the previously measured forces.

[Off topic material removed]
 
Last edited by a moderator:
You keep trying to say Hulsey did not use all of NIST's fire simulations as though that presents a problem. Your logic is flawed here, as Hulsey used NIST's worst case. By definition that envelopes all other situations.

An analogy for others to understand is that if one were to do a test on what the forces were during the torquing of a lug nut to failure all one ultimately needs to use in the analysis is the worst case force, which would be that which occurred at failure. It envelopes all of the previously measured forces.

Fire is not a simple thing. It changes over time, so it heats different things at different times. Differential heating leads to larger differences in expansion between adjacent members. So a more accurate simulation would account for that.

Fire induced thermal expansion is not a simple force either. To extend your analogy, suppose you were removing a screw, you stick in your Hole Hawg with a screwdriver bit and you strip the head smooth. Or you take a handheld screwdriver and apply pressure firmly and remove the screw. The "worst case force" did not remove the screw.

The variations over time need to be considered. Hulsey does not even seem to have considered differential heading rates for steel and concrete in one spot, let alone the spread of the the fire.
 
I keep wondering about the heat conductivity in a steel construct.
Does it act like a big heatsink where fires above and below transfer their heat via this big heatsinky steel construct, and if so, to which degree, and how much gets through through the connections?
 
The 2" thick x 14" deep x 18.875" wide support plate below the bearing seat is the item which would need to fail for the 12th floor girder to be collapsed by an impact from the girder above.

That plate was welded to column 79 with 3/8" fillet welds on both sides (28 inches of weld). I have attached the drawing which shows the plate and its welds on column 79. The plate has the material code of "PG" in the drawing and the weld type, size, and location is called out. These welds would need to be sheared and that shear force is 632,000 lbs..

The support plate was welded to column 79 which never got hotter than 300 C in the fire simulation. The support plate would have been at approximately the same temperature as column 79. At 300 C steel has not yet lost any yield strength, so the force required would still have been about 632,000 lbs.

The impact could not have provided more than about a 1/3rd of that force.

Assuming it fails for the seat shearing as opposed to a girder that is already walking being knocked off in some direction. Of course, there were multiple ways it could have failed given that it, like the girder above it, was being subjected to fires that could have been intense enough to fail it by themselves (and, even if you continue to pretend that you have established the NIST walk-off scenario was impossible, Arup has demonstrated multiple other scenarios for that same girder to fail). Plus, as you know, NIST and Arup also found that multiple beams on floor 13 around girder 79-44 would fail as well under expected fire conditions (either in the heating or cooling phases) and so we can expect these floor systems were heavily damaged. Insisting that the only mode of failure in such a system is to shear the girder seat is just a transparent attempt to get around the fact that neither you nor Hulsey (at least as far as we know) made a serious attempt to understand how damaged the lower floor would be.

You keep trying to say Hulsey did not use all of NIST's fire simulations as though that presents a problem. Your logic is flawed here, as Hulsey used NIST's worst case. By definition that envelopes all other situations.

An analogy for others to understand is that if one were to do a test on what the forces were during the torquing of a lug nut to failure all one ultimately needs to use in the analysis is the worst case force, which would be that which occurred at failure. It envelopes all of the previously measured forces.

Nope. Refer to my post here on the last page. Not only is Mick right that of course the sequence of heating, and thus the sequence of expansion, buckling, cooling, contraction, failure, etc. across this complex, interconnected system matters, but Hulsey didn't even properly model NIST's worst case scenario on floor thirteen even if all he was trying to do was heat each element to its hottest point in a simplistic way without regard for sequence. The temperatures he indicated that he used from the 6:00 time frame were an hour after the area had started to cool in NIST's model.
 
Last edited:
Assuming it fails for the seat shearing as opposed to a girder that is already walking being knocked off in some direction. Of course, there were multiple ways it could have failed given that it, like the girder above it, was being subjected to fires that could have been intense enough to fail it by themselves (and, even if you continue to pretend that you have established the NIST walk-off scenario was impossible, Arup has demonstrated multiple other scenarios for that same girder to fail). Plus, as you know, NIST and Arup also found that multiple beams on floor 13 around girder 79-44 would fail as well under expected fire conditions (either in the heating or cooling phases) and so we can expect these floor systems were heavily damaged. Insisting that the only mode of failure in such a system is to shear the girder seat is just a transparent attempt to get around the fact that neither you nor Hulsey (at least as far as we know) made a serious attempt to understand how damaged the lower floor would be.



Nope. Refer to my post here on the last page. Not only is Mick right that of course the sequence of heating, and thus the sequence of expansion, buckling, cooling, contraction, failure, etc. across this complex, interconnected system matters, but Hulsey didn't even properly model NIST's worst case scenario on floor thirteen even if all he was trying to do was heat each element to its hottest point in a simplistic way without regard for sequence. The temperatures he indicated that he used from the 6:00 time frame were an hour after the area had started to cool in NIST's model.
I am going to presume that you are referring to NIST's ANSYS model here.
Here is how the parameters were applied.
I look forward to your comments on it re the above.ANSYS loading.jpg
 
wonder if the UAF study showed that the K3004 girder also expanded to the East slightly, as can be seen would happen below, when you examine the (K3004 - column 38) connection. (NIST didn't)
Haven't we been through this already?

I showed you that your claim is FALSE and the opposite is true: NIST DID model the gap and allowed for the beam to expand east!

You then moved the goal post and SPECULATED that perhaps the control elements didn't work properly - a claim for which you have zero evidence.

So stop making this claim until you have actual evidence that these beams did not expand east when they should have!
 
Status
Not open for further replies.
Back
Top