NASA Scientist Ryan Mackey Answers ATS Questions

VALHALL Questions to Mr. Mackey regarding the NIST report and his responses.

Mr. Mackey's responses are bolded.

Dear Sir,

My contentions with the NIST report due not lie within the framework of LIHOP or MIHOP. In fact, my contentions don't attach to any conspiracy theory at all. Where I take great umbrage with the NIST final report is in the fact that the conclusions drawn within the report do not align with, and at times soundly contradict, the data presented in the report. Since this investigation, analysis and resultant report were paid for by the American people, which include me, I take issue with being handed a report that reflects insufficient data collection (insufficent collection of failed members and testing on same), scientifically unsound statements (creep played a significant role in the failure of the supporting elements), and a failure model that contradicts the data they present themselves on what little testing they did perform.

This is awfully long-winded and short on actual questions. There are no "sound contradictions" between the conclusions and the data, and since this individual didn't cite any examples of this, I have no idea what he or she is talking about. There is nothing unscientific about creep playing a role in failure, and this is an expected result given the anticipated temperatures and loading of the structural elements. There is also nothing contradictory in the failure model.

This person, while complaining that NIST hasn't provided enough detail about its answers, has failed to provide any detail about his or her *questions.* I'm trying to avoid an "ad hominem tu quoque" response, but without more to go on, there's no way to respond to such a vague complaint.

What makes the incongruities of the NIST report even more exasperating is the fact NIST refuses to share their model and database with the U.S. citizens who wish to vet the process used. I noticed you state in your paper that reviews Dr. Griffin's work (a work I must admit I know little to nothing about), that while Dr. Griffin does not provide calculations to support certain accusations (calculations I fully agree are needed), you state "the NIST Report [7] is readily available, and while the report may be criticized, errors that Dr. Griffin makes regarding its contents may be factually verified with no uncertainty." While this statement is true as refers to the REPORT, it is absolutely untrue as regards the actual important information (i.e. the model, the empirical data, the assumptions, etc.) Yes, we can verify the CONTENT of the NIST report, but that content is made up of only sparse references to actual data, and by and large is INTERPRETATION of that undisclosed data.

This is patently false. NIST's models are in fact available to ordinary people. I've corresponded with someone who received the SAP2000 models simply by requesting them, and paying what he termed a "reasonable fee." They are available. Try it yourself.
The models are only useful if we suspect they are somehow constructed incorrectly. If we get NIST's models, and run them with NIST's inputs, we will get NIST's answers -- there is little additional insight that isn't already provided in the reports.
Besides, it is much more useful to either contest NIST's interpretations of those results, or to contest the assumptions with which they were derived. This is precisely what folks such as Dr. Usmani and Arup have done, as I describe in my whitepaper. These folks didn't have NIST's models either; they came up with their own. Not having access to NIST's models does not impede research, or the ability to challenge NIST's findings.

Stepping off my soap box I get to the matter hand. One particularly ornerous incongruity is the fact that the final failure mode on the macro level presented by NIST is in unacceptable contradiction (as far as this learned reader can tell) to the failure mode of certain components analyzed. In particular, the floor truss connectors at the perimeter walls and at the central core columns of WTC 1/2. The NIST report states that the floor truss connectors at the perimeter walls were sheared in a downward motion while none of the floor truss connectors on the central columns exhibited this failure mode. This would require that the floors fall in a downward motion at the perimeter wall connection while staying attached to the core columns. Since the perimeter walls couldn't possibly have "jumped up" to cause a downward shearing motion at the connectors, and since the connectors on the core columns did not shear in a downward motion, this would require that the floor trusses and core columns move in concert - which would be in a downward motion relative to the perimeter walls. The failure mode that has been presented by NIST does not account for this situation.

NIST does not claim this. NIST suggests that the floor connections were, in most locations (all of those indicated as "pull-in" locations) remained attached to both core and perimeter until collapse was already underway. After that, the precise timing of floor connections is academic.
If the complaint is about the failure modes observed during collapse, it is quite possible -- the perimeter columns essentially buckled inwards, whereas the core, being the heaviest part of the structure and the most compromised in the moments leading up to failure, would have been leading the collapse. The perimeter is expected to pull inwards before either the floors pulled free on their own, or were sheared off by the descending upper mass. This is consistent with NIST, but entirely peripheral to their investigation, and surely an irrelevant point of minutiae. Once the collapse began, there was no stopping it. The work of Bazant and Zhou proved this long before the NIST investigation began.

WOW! Too bad he didn't read the NIST report.

I think I have my answers...more than you'll ever know.

I've decided to let his comments stand alone.

After all, it could be unhealthy for my family's safety to confront an individual so desparate for funds.

[edit on 10-16-2007 by Valhall]

reply to post by Valhall



Valhall ~

It seems you are far from happy with his response. Please let me know where he is in error, honestly he's very fair and open to any mistakes he has made. (as you will see in a response he gave Griff that I will be posting)

C.O.

I have nothing further to say to this individual. He excused himself from making any real response with the excuse I did not do his research for him. I believe he has placed himself in a position to do his own research. As an amateur, I've already done mine, and it is presented in numerous threads on this board. His responses do not even show evidence of him actually reading the report. And him claiming to have friends with the models is nice anecdotal non-evidence, but it also tells me he doesn't. And I have lots I could say from an engineering standpoint concerning some of his technical statements...but let's just leave it at - I hope it all spends well for him.

I have nothing more I want to hear from this individual. Good luck, CO.

reply to post by Valhall



So, you are worried if you ask for the information, your family will be in danger? Hmmm.. not the response I would expect from you. But..it's your perogative.

No, I'm not worried. I just don't care to hear what this guy has to say. He's not even trying.

The quip above about my family was tongue-in-cheek.

reply to post by Valhall


Valhall, seems like a cop out to me. But who am I?

... I was trying to offer posters in this forum a way to communicate with a REAL scientist with REAL answers. I never said everyone would like them. You had the guts to state some issues you had with the NIST report. He answered your questions within 5 hours of me e-mailing him. You apparently do not like his answers.

Can I ask again....what was in error? What makes you think he didn't read the report?

You're not listening to me...I'm done. In my first post, which you've already shared with him, I was clear what bothers me about the NIST report. The conclusions do not match the data they present. I gave an example of the data not matching the conclusions, he responded with the NIST conclusions.

???

As I said, I've done my research. I'm not doing his.

And I question the quality of the "scientist" you came up with for us to converse with.

I'll wait for the scientific publication that reviews data on thick-walled steel structures failing from creep when the steel temperature does not exceed 650 F and the time duration of elevated temperature under loading is 1 hour. I'm sure he's probably going to whip that right out for us....after all, he said it happened and that it happens and so he's obviously got that on the shelf to his immediate right.

Originally posted by CaptainObvious
If the complaint is about the failure modes observed during collapse, it is quite possible -- the perimeter columns essentially buckled inwards, whereas the core, being the heaviest part of the structure and the most compromised in the moments leading up to failure, would have been leading the collapse.

A few points here. I was under the impression that the perimeter columns were pulled inward and that was what caused collapse? He is saying that the core was what caused collapse? Which goes along with my theory but is a contradiction with NIST I believe.

BTW, how was the core structure "the most comprimised"? Especially in WTC 2 where the plane practically missed the core?

The work of Bazant and Zhou proved this long before the NIST investigation began.

But, the work of Bazant and Zhou included many assumptions that have not been proven or disproven.

Can someone help me out here? I'm looking for the NIST loads. I had them at one time but can't find it. And I really don't have the time to sift through a thousand some page report. Also, the section you got it from would be helpful as I need more information but I don't want you to do all the homework.

I'm trying to put together calculations of what the factor of safety of the exterior columns of the towers would be in all seriousness. No more guessing around etc. (well, I'll probably have to assume some things, which I'll run by you guys).

Hint: When designing for beam-columns, you convert (equate) the moment caused by the horizontal forces (the wind loads etc.) into a P (force) effective. This Peff. is a concentric load down the column...i.e. like dead and live loads.

Edit: To take this a step further. The columns were the same the entire length of the tower on the exterior except for the tree at the bottom. So, from the 10th story up, they were the same dimensions, strength etc. So, the (say) 85th floor columns would have the same strength (yes, vertical and some horizontal) as the columns from the 10th floor. Imagine all that load on the 10th floor columns that the 85th floor columns were designed with the same strength as. It is a known fact that at least 40% of the buildings weight was supported on the exterior. This leaves a lot of factor of safety up in the higher floors IMO.

So, I'm still a little confused as to why Mr. Mackey would state the wind loads control (and he's probably right). But, the columns would be designed for the effective gravity loads imposed on by the wind forces. Thus making his claim a little erroneous IMO. But, that still doesn't mean Dr. Griffin is correct with his 2000% either.



[edit on 10/16/2007 by Griff]

[edit on 10/16/2007 by Griff]

I think I may have found where the 2,000% comes from.

The "Engineering News Record" is quoted as saying:

"A design procedure that will be used for structural framing of the
1,350-ft high twin towers of the World Trade Center in New York City gives the
exterior columns tremendous reserve strength. Live loads on these columns can
be increased more than 2,000% before failure occurs."

I'm trying to confirm this at the moment.

Here ya go Griff ~

Here is the E-mail I sent Mr. Mackey

And this is from "Griff" who claims you are wrong for stating that "Wind load is not considered as a live load."

Source: "Structural Analysis", Alexander Chajes, 1990, Prentice Hall

My structural analysis book from college.


"It is similar but not identical. In particular, the design rules allow "live load reduction" with height, which was a shorthand approach to avoid double-counting safety margin in tall structures. Wind load is not eligible for live load reduction, and therefore must be bookkept separately. Wind load is not a dead load or a superimposed load, so it is perhaps most similar to a live load, but for tall structures it is simply false to state that it is live load. "

Then why does my structures book say that wind load is a live load?
Because, for whatever reason, your textbook saw fit to group all loads into either "live" or "dead." I certainly won't argue that, if given only these two choices, a wind load is more like a "live" load than a "dead" one, but this grouping is insufficiently precise.

The actual definitions -- the requirements for the actual WTC design and construction -- are not responsive to this textbook. Instead, they follow building codes, which clearly provide different definitions regarding wind load.

For example, ASCE 7-02 states the following: "Live loads are those loads produced by the use and occupancy of the building or other structure and do not include construction or enviromental loads such as wind load, snow load, rain load, earthquake load, flood load, or dead load. Live loads on a roof are those produced (1) during maintenance by workers, equipment, and materials, and (2) during the life of the structure by movable objects such as planters and by people."

So, as you can see, wind load is NOT generally considered part of the live load. Perhaps there is some building code somewhere that uses such a definition, but it would be exceptional. There is no reason to assume this is true for the WTC construction, and no reason to assume it applies to Skilling's offhand comments from 1964.

Ryan's quote:"In modern design practice this is irrelevant as we no longer use load factors and ultimate strength design."

Also wrong. In modern times we use LRFD (Load Resistance Factor Design). Notice the load and factor in there? Also, modern times don't govern what was designed for back in the 60's and 70's. We used ASD (Allowable Strength Design) back then. Don't know where he gets Ultimate Strength Design from.

I'll concede this point, since I was imprecise. The ASD practice from the 60's and 70's (and still in use in some places) is concerned with the maximum elastic capacity of a structure. LRFD (also known as "Ultimate Stress Design," which is how my professors referred to it) can plausibly go beyond this limit, but uses multiple formulas to calculate likely worst cases rather than a single monolithic worst case, as in ASD.
The point I failed to clearly articulate is that, under ASD, since there is only one formula in use -- that of the maximum elastic stress -- there is a well defined maximum load as well, and thus we can easily express the total margin as a function of the total service load, the dead load only, or (as was done in Skilling's quote) a function of service live load only. In LRFD, the limit is not as well defined due to the competing limit cases, and it would be harder to make a "2000% margin" quote that had any meaning. We would also have to specify a scenario, i.e. minimal wind load, maximum earthquake load, etc. Under ASD there are no different scenarios.

In any case, I see no error with the respondent's corrections in this regard, so I'll leave it at that. I should have been more careful in my first reply.

What is relevant is that the "2000%" claim cannot refer to both service live loads and wind loads. If it did, that means the perimeter columns were overdesigned by more than a factor of 10. This is simply not true, as found by NIST in its baseline models, and can be verified by hand from the column cross-sections and composition provided in NCSTAR1-1. We therefore understand that the "2000%" could only have referred to the service live loads, in which case it isn't at all inconsistent with our expectations.
On the relatively windless day of September 11th, the perimeter columns were about six times stronger than the total load at the time, and that is consistent with our interpretation that the "2000% of live load" remark only refers to service live loads, not wind loads. This margin is why the Towers stood as long as they did, but it was not enough margin to be invincible, contrary to what Dr. Griffin claims.

Thanks,
Ryan Mackey

reply to post by CaptainObvious


Thanks. Also, if you correspond with him again, thank him for me would you?

He basically said what I was saying. Depending on the source wind load is considered live load or not. Could you ask him about the conversion of the moment caused by the wind load into a concentric load for me? I'd like to hear his explaination. Thanks.

The equation is:

P(eff) = P + M(ux)m + M(uy)mu

P(eff) is the equivalent axial load
P is the total factored concentric load (dead and live loads)
M(ux) is the moment due to the total factored bending load (wind load in this case) in the x direction
M(uy) is the moment due to the total factored bending load in the y direction
m and u are factors coinciding with the column properties.

That is how a column that is subjected to bending and axial compression is designed.

Thanks.

Originally posted by CaptainObvious
This is awfully long-winded and short on actual questions. There are no "sound contradictions" between the conclusions and the data

What physical tests did NIST do that corroborated their critical hypothesis (specifically that heated trusses could cause the required deflections in outer columns to initiate a failure), and how specifically was their hypothesis corroborated? Basically, I want to know where the meat is. I want to see a truss heated and subsequently pull a perimeter column out of place.

There is nothing unscientific about creep playing a role in failure, and this is an expected result given the anticipated temperatures and loading of the structural elements.

Emphasis mine. What temperatures are you anticipating in the steel in these fires, and where is the experimental data to support your figure?

If you look up the tests done on steel frame structures in Cardington in the UK, by the University of Edinburgh and a few other institutions, you'll see high-powered fires (intentionally created to generate high amounts of power directly upon the steel itself for extended periods of time -- whereas NIST states that the fires tended to flare up and die down in different parts of the buildings, among other differences) uniformly heat steel only to a max of between 600 and 650 C or so, hot enough to glow in broad daylight. Again, NIST found no evidence of this kind of heating in the WTC Towers, but U of E et. al. did this intentionally to study the effects on the structure.

It was found in the final study (here's a link: www.studyof911.com... ) that failures in steel frame structures occur primarily from stresses between the members caused by thermal expansion. This is the primary cause of warping, sagging, buckling, etc. in fires heating steel to ~600 C and below.

The study opens with this:

Steel beams in standard fire tests reach a state of deflections and runaway well below temperatures achieved in real fires. In a composite steel frame structure these beams are designed to support the composite deck slab. It is therefore quite understandable that they are fire protected to avoid runaway failures. The fire at Broadgate showed that this didn't actually happen in a real structure. Subsequently, six full-scale fire tests on a real composite frame structure at Cardington showed that despite large deflections of structural members affected by fire, runaway type failures did not occur in real frame structures when subjected to realistic fires in a variety of compartments.

www.studyof911.com...

They talk about local failures (ie warping, etc.) inducing local stresses that are "seen" less and less by members the further out you go in the structure, which makes sense. None of the stresses are enough to cause anything more than localized warping and buckling, of course, and this is only where high-powered fire is hitting right on the column for an extended period of time. They also put some numbers on these internal stresses if I'm not mistaken, and they show diagrams illustrating relations of stresses with temperature and things like that.

[edit on 19-10-2007 by bsbray11]

For reference, here are perimeter/truss connections:





Imagine the "floor" pulling the outer columns in, simply by heating the "floor" up. According to NIST, the truss was heated, sagged, and then the change in the angle at which the truss meets the column supposedly deflected the column. At least this is my understanding. I have never seen an FBD of this theory and I'm not exactly clear on where what forces are supposed to work.


NIST's engineers also helped put together this cute illustration, from a PBS NOVA presentation they did:




I would love to see a free body diagram on one of those, too. Apparently you even get a dynamic load from heating a truss, motion being almost completely horizontal and enough to yank out a column at both ends at once.

[edit on 19-10-2007 by bsbray11]

Originally posted by bsbray11

I would love to see a free body diagram on one of those, too. Apparently you even get a dynamic load from heating a truss, motion being almost completely horizontal and enough to yank out a column at both ends at once.

What's more likely? The above scenario that NIST is trying to sell. Or the angle clips shearing before the column shears? How is a 1/4 inch angle clip stronger than a 2 inch thick column? Don't even get me started about having to shear through the "sides" of the 2 inch thick column either. It's obsurd to say the least.

Edit to add: To take it one step further. What about the trusses that are suppossedly being weakened and doing the pulling? How are they so strong to pull in those columns?

Try this. Construct a tube column with cardboard. Put a piece of paper (imitating an angle clip) on the column and try to pull. What rips first?

NIST really is reaching IMO for that scenario to work.

[edit on 10/19/2007 by Griff]

Thanks guys, I will send this off to him soon. However, I will not be attaching any photo's in his e-mails. I do link the web pages if his is interested in looking there.

I have a couple of clarifications for Ryan Mackey.

He cites finding 13 of NISTAR1-2.

Finding 13: Impact of an empty wing segment from approximately mid-span of the wing normal to the exterior wall produced significant damage to the exterior columns but not complete failure. Impact of the same wing section, but filled with fuel, resulted in extensive damage to the external panels of the tower, including complete failure of the exterior columns.

• What was the fuel load of Flight 175 at the time of impact as a percentage of total capacity?
  
  
• If a wing 'filled with fuel, [would result] in extensive damage to the external panels of the tower, including complete failure of the exterior columns', can we assume that anything less than 100% of capacity at the time of impact would have the effect of significantly reducing the level of damage to the exterior columns upon impact?
  
  
• Given the

I want to know where the meat is.

apparently OVer's are vegans also.

/sarcasm

I've been looking for it high and low and can't find any on either side. Its pretty amazing to me since 'truthers' seem to have alterior motives for the most part, however OV'ers don't seem to think the burden of truth is on 'us'.

well, 'we' didnt go to war over it.