J911 Studies Peer Reviewers

[Tony Szamboti]

Great post, NB. Even I, a non-engineer, could understand exactly why Tony the Twoofer is wrong.

I've also got one beverage of your choice that says he doesn't answer this criticism.

Your use of childish ad hominem along with unwarranted confidence did not add anything to this discussion. It turns out you owe someone a beverage.

 

[Newtons Bit]

You can argue with the AISC manual, which says floor slab connections constitute a braced frame. The reason is obvious and it is the stiffness and great inertia which the floor slabs impart to the connection. Newtons Bit's use of a K factor of 1.0 is conservative for design, as I pointed out earlier, and does not belong in determining what the real K factor was in the failure analysis.

No it doesn't. Floor slabs brace beams against buckling (both lateral torsional and compressive), a braced frame a slab does not make.

Stop making things up.

 

[Tony Szamboti]

First, download and read the commentary for AISC-360-05. At least this way we can be referencing the same document. Then, admit you've made a mistake because you've yet again failed completely to understand what you're talking about. It's a simple thing to do.

Let's look at the example you've provided. The author provides a diagram representing a building section:

[qimg]http://www.internationalskeptics.com/forums/imagehosting/thum_163294781014125ed2.jpg[/qimg]

The important thing to note is that point J is a pinned connection. The author is correct in saying that sidesway is inhibited because point J, and thus the beams connected to it, cannot translate. Point J represents a stiff lateral element, a braced frame or a shear wall perhaps. It does not represent a moment frame.

The author then shows how to calculate k for the top portion of the building, columns DE and GH. These he says are, "there is no sideways bracing for the top portion of the frame". That's because they are not connected to a braced frame or a shear wall but rather a moment frame (or a cantilevered column, it depends on what the connections look like).

You can try to twist it all you want. The reality is that the columns in the Twin Towers would not have behaved as pinned connections, as you try to claim. They would have acted as being fixed at both ends, and sidesway inhibited, due to the stiffness and inertia provided by the composite action of the floor slabs fastened to beams, in both the core, and the perimeter. I intend to show this and that you are wrong. Your paper will get a real peer review and it will be shown that you are using conservative design values for the slenderness ratios, instead of actuals, to support your contention that the columns could have easily buckled.

This post will be my last here on the subject.

 

[CHF]

This post will be my last here on the subject.

I can see why, Tony.

How about doing your cause a favour by submitting your "research" for peer-review in a REAL scientific or engineering publication.

 

[twinstead]

Your paper will get a real peer review and it will be shown that you are using conservative design values for the slenderness ratios, instead of actuals, to support your contention that the columns could have easily buckled.

I shudder to think what would happen to your work if it ever gets a 'real' peer review.

 

[Newtons Bit]

You can try to twist it all you want. The reality is that the columns in the Twin Towers would not have behaved as pinned connections, as you try to claim. They would have acted as being fixed at both ends, and sidesway inhibited, due to the stiffness and inertia provided by the composite action of the floor slabs fastened to beams, in both the core, and the perimeter. I intend to show this and that you are wrong. Your paper will get a real peer review and it will be shown that you are using conservative design values for the slenderness ratios, instead of actuals, to support your contention that the columns could have easily buckled.

This post will be my last here on the subject.

I never said they were pinned, I said they were MOMENT frames. That's sort of the opposite of pinned in case you didn't know. Don't put your failure to understand simple engineering on me. Even the example you linked to showed you're wrong and you twist and twist and twist to try to get it to say you're right. And when confronted about it, you run away.

And how will my "paper" get a real peer-review? Are you going to submit it under false pretenses? That's not very nice, Mr. Szamboti.

 

[Myriad]

I have the Eighth edition of the AISC manual and the Chapter 5 Commentary was effective 11/1/1978. Maybe they changed the paragraph number in later editions. Chapter 5 Section 1.8 of the Eighth edition is titled "Stability and Slenderness Ratios". You should look for a paragraph with that title.

Thank you for the additional information. It's clear that there is no single corresponding paragraph in the 2005 edition. The relevant Commentary chapter is now chapter C (chapters are lettered and appendices are numbered), "Stability Analysis and Design," and section C2 within that chapter discusses methods of caclulating required strengths, including "traditional" methods that use the Effective Length Factor K.

I see nothing in that chapter or elsewhere in the current manual that suggests that the WTC towers are any exception of the general rule that K values in braced frames as well as moment frames are normally greater than or equal to 1.0. (see paragraphs C1.3b, C1.3c). There is mention (in the text accompanying the C-C2.3 nomograph) that column ends "rigidly attached to a properly designed footing" (emphasis added) can be taken to have relatively low values of G, which would lead to relatively lower values of K. However, the above-ground floors in the towers, despite being partially constructed of concrete, were in no way "footings" and in any case the column ends were not (very) rigidly attached to them.

More generally, there is nothing that suggests that the general nature of the horizontal connections between column ends, such as their total mass, their length, the materials they're made of, or whether or not they happen to be used as floors, has any bearing (no pun intended) on the effective value of K for the columns. (Naturally, the structural properties of the horizontal members would affect other aspects of any stability analysis, but they don't figure into the K of the columns.)

I think you need to do an analysis to show that any of the columns in the towers would have performed as pinned connections rather than fixed on both ends. Diagonal bracing is not the only thing that causes a frame to act as braced.

"Fixed at both ends" doesn't imply K < 1.0. Actually, according to AISC, braced-frame systems are analyzed and designed as pin-connected systems, with a K of 1.0. Moment-frame systems, which better describes the WTC towers, generally use K values greater than 1.0:

Moment-frame systems rely primarily upon the flexural stiffness of the connected beams and columns, although the reduction in the stiffness due to shear deformations can be important and should be considered where column bays are short and/or members are deep. Except as noted in Section C2.2a(4), Section C2.2b and Appendix 7, the design of all columns and beam-columns must be based on an effective length, KL, greater than the actual length determined as specified in Section C2.

(emphasis added)

Lest you get the impression that Sections C2.2a(4), C2.2b, and Appendix 7 contain a statement that these rules don't apply if the columns in question happen to be connected to floors (aren't most columns in most buildings connected to floors?), the quote continues:

The Direct Analysis Method in Appendix 7. as well as the provision of Sections C2.2a(4) and C2.2b, provide the means for proportioning columns with K = 1.0.

In other words, those sections cover alternate "direct" analysis methods in which K is not used as a variable; instead the various influences of the physical parameters that figure into K are accounted for separately (generally, in a computer model).

So far, all the information from the AISC manual appears to support Newton's assessment.

Respectfully,
Myriad

ETA: Cross-posted with the last several posts by Newton, Tony, and others. It seems that learning the principles of structural engineering from reference sources can take some time. Who'd'a thought?

 

[Jonnyclueless]

Your paper will get a real peer review and it will be shown that you are using conservative design values for the slenderness ratios, instead of actuals, to support your contention that the columns could have easily buckled.

This post will be my last here on the subject.

Uh oh, does this mean Charlie Sheen will be reviewing it? Maybe Webster?

 

[jhunter1163]

Well, it seems I was wrong about Tony's response (or lack thereof). Since I'm nothing if not a gentleman of my word, I'll buy Tony a beverage of his choice. And NB too, for that matter.

However, Tony is still wrong.

 

[Newtons Bit]

ETA: Cross-posted with the last several posts by Newton, Tony, and others. It seems that learning the principles of structural engineering from reference sources can take some time. Who'd'a thought?

You seemd to have it figured out with no problems. It's pretty straight forward.

 

[Gravy]

Aww, I was looking forward to toying with the twoofer some more, but I had to go to bed. It's bad when a tour guide can immediately see why Szamboti, who claims to be a mechanical engineer, doesn't know what he's talking about.

Suppose I just made everything up on my tours. Wouldn't that be wrong of me?

 

[R.Mackey]

And how will my "paper" get a real peer-review? Are you going to submit it under false pretenses? That's not very nice, Mr. Szamboti.

We should be on the lookout for this... it wouldn't be the first time, as I recall. Truly those at the "Journal" for 9/11 Studies have a demented concept of peer-review!

 

[Newtons Bit]

I think we can officially call this closed. I'm thinking it might be useful to compile a list of errors that the peer-reviewers of the "journal" make so as to challenge their credibility.

I'm tired of truthers calling them experts on buildings. We need a big long list of links that show they're not. And while just linking to the papers on the "journal" would suffice for most rational folks to see they're B.S. I'd like to see something more substantial where it's easy to focus on just one point and show how they're completely wrong. Especially points where the truther himself won't admit he's wrong when even a tour guide can understand the math involved to see he is.

 

[JamesB]

We should be on the lookout for this... it wouldn't be the first time, as I recall. Truly those at the "Journal" for 9/11 Studies have a demented concept of peer-review!

No kidding, I write an e-mail to a psychology PhD candidate, and the next thing I know it was a "peer reviewed" paper.

 

[AZCat]

I think we can officially call this closed. I'm thinking it might be useful to compile a list of errors that the peer-reviewers of the "journal" make so as to challenge their credibility.

I'm tired of truthers calling them experts on buildings. We need a big long list of links that show they're not. And while just linking to the papers on the "journal" would suffice for most rational folks to see they're B.S. I'd like to see something more substantial where it's easy to focus on just one point and show how they're completely wrong. Especially points where the truther himself won't admit he's wrong when even a tour guide can understand the math involved to see he is.

Thanks to you (and the others) for taking the time to explain in detail the calculations for determining slenderness ratios for columns. I had this back in school but not as in-depth (went for the thermo track).

 

[Newtons Bit]

When inventing things on the internet....

Word of advice to Mr. Szamboti: when inventing things on the internet, make sure that the people reading it don't have access to what you are referencing.

Mr. Szamboti stated that:

Paragraph 1.8.2 of the Commentary of the AISC manual says connections to floor slabs constitute a braced frame for horizontal stability.

That's pretty specific. Unfortunately for him, I have access to an AISC 8th edition that has the spec he's quoting. And I've attached the page from the spec below. It doesn't say what he says it does. Specifically it states:

In trusses and in those frames where lateral stability is provide by adequate attachment to diagonal bracing, to shear walls, to an adjacent structure having adequate lateral stability, or to floor slabs or roof decks secured horizontally by walls or bracing systems parallel to the plane of the frame, the effective length factor, K, for the compression members shall be taken as unity, unless analysis shows that a smaller value may be used.

Bolding mine. What that is saying is that floor slabs or roof decks that connect to a gravity column AND a brace constitute a braced frame. This is pretty much common sense. The braced frame provides the lateral system and the gravity column just goes along for the ride.

The very next section in the book that Mr. Szamboti was referencing defines unbraced frames. It is very clear that the WTC exterior columns are in this category AND that K shall not be less than 1.

In frames where lateral stability is dependent upon the bending stiffness of rigidly connect beams and columns, the effective length KL of compression members shall be determined by a rational method and shall not be less than the actual unbraced length.

I think we have ourselves a charlatan.

 

[ref]

Thanks NB good work.

 

[beachnut]

very nice work -

 

[NicoS]

What about K in NIST computed simulations ?

Hello everyone.
(my first post in this forum...)

Didn't NIST compute a model of the towers and run a simulation to describe the very first instants of the collapse initiation? (I think it was WTC2, as far as I remember) They shall have choosen a K factor, or they shall have choosen a model to represent the structure (including the columns to beams connections) in whitch K can be calculated. What about this K factor in those simulations?
And more generally, isn't it obvious in those simulations that buckling occured?(it is an opened question, I'm not familiar enough with these NIST simulation to make it afirmative).

Thanks for your attention.
And sorry for my english speaking, I'm french (nobody's perfect)