Metatheory and the NIST report

[TellyKNeasuss]

I will preface my comments by acknowledging that I only skimmed parts of the NIST report. I am just throwing out some ideas to get the dialogue moving again, and my impressions may be way off. I don't even understand the context in which the simulations were used to describe the collapse mechanism. I also am not an engineer of any sort, nor am I a numerical modeler.

Can you answer the question of why NIST appealed to the higher fuel load of 5 lb/sq ft in WTC 1 after pointing out repeatedly that 4 lb/sq ft was the best estimate of the fuel load in the towers?

My interpretation was that the 4 lbs/sq ft was a guestimate based on a simple model of the layout of furniture, etc. The 4 lbs./sq ft most likely did not generate enough heat to match the observed temperatures (I could be wrong, but I probably won't have time to thoroughly read the section in the near future). If you make a rough guess at the input to any kind of model and your results don't match the observations, do you: 1) assume that the observations are wrong; 2) assume that your guess as to what the input should be is wrong. The inputs to the model are not that precise; for instance they were unable to include heat from the burning aluminum in the airplane. I doubt that this sort of adjustment is unheard of in situations where the baseline data is not known.

They also clearly state that increasing the fuel load would not produce the observed effects, nor would it allow for the fire to travel to the south side in time to initiate global collapse.

Do I misunderstand this statement, or are you acknowledging that increasing the fuel load from 4 lb/sq ft to 5 lb/sq ft wasn't important anyway. Because it was just an adjustment needed to generate the heat that was indicated by the observations?

No one here can answer this question. There is no good answer in any of the NIST literature. Whether you or anyone else here likes it or not, there is no good reason to increase the fuel load in the towers by 25%, yet the NIST authors do so, and claim they've answered the question of why the buildings collapsed.

Yet they published this in a publicly available document. If they were fudging numbers, why wouldn't have they just published the 5 lb/sq ft data? And just because no one here can answer a question, does that prove that the question cannot be answered by anyone?

The real question in all this seems to be, Why do you still believe the NIST report answered the questions posed from the collapse? It's not because of the solid case they make in the report. Is it the fear that you may have been fooled this whole time? The fear that people in official positions are not looking out for your best interests? The peace of mind that only official reports can produce? What exactly keeps you clinging to the results of a report that no one can properly make sense of?

No one can properly make sense of? You know for sure that here isn't a single structural engineer or fire protection engineer in the whole world that can make sense of the report? And I suppose that the reason why nobody has mentioned it is because every single one is too embaressed to admit it.

 

[twinstead]

It appears that Jay claims that NO ONE can make sense of the NIST report.

 

[cmcaulif]

For WTC 1, the changes were designed to create a "more severe" fire.

<snip>

The intent of these changes was to span the probable range of gas temperatures in the core area by considering a more severe case. The Case A simulation showed that the model was replicating the basic fire behavior at the building exterior, but there was no way to confirm the behavior deep within the building.

The rest of the page provides further details. This should be noted as well:

The assumptions made for all Cases fell within the range of uncertainty resulting from the impact analyses; the objective of the parameter changes was to generate results that would span as much as possible the set of probable outcomes given the range of plausible initial and boundary conditions

 

[GlennB]

The rest of the page provides further details. This should be noted as well:

Yep, and jay howard knows all this perfectly well.

He seems to be obsessively clinging to the 4lb figure, apparently believing that any variation from it somehow consigns NIST to the dustbin. Of course it doesn't.

 

[cmcaulif]

Yep, and jay howard knows all this perfectly well.

He seems to be obsessively clinging to the 4lb figure, apparently believing that any variation from it somehow consigns NIST to the dustbin. Of course it doesn't.

Especially given the nature of the FDS software, which can be downloaded for free here, just as an aside.

FDS is sensitive to a number of inputs, including fuel load, and after the first case was done for WTC 1, the NIST wanted to examine a case with higher temperatures at the core. Increasing the fuel load was one of the ways they achieved this through the FDS software.

 

[jay howard]

And yet, no one here can squarely reconcile the fuel load increases.

cm,

you quote the NIST report as if their "answers" justified the increase:

The intent of these changes was to span the probable range of gas temperatures in the core area by considering a more severe case.

And how would they know what the "probable range of gas temperatures in the core" are? From estimating the a reasonable fuel load--which they do on several occasions in the report as 4 lb/sq ft.

Now, if 4 lb/sq ft is a reasonable estimate given all the info known about the towers, what possible reason do they suppose that they underestimated this by 25%? That's a significant increase. That's 60 tons per floor versus 75 tons per floor. Where do you suppose an extra 15 tons of fuel per floor came from? Wouldn't you think they'd give some kind of justification for an extra 15 tons per floor?

The Case A simulation showed that the model was replicating the basic fire behavior at the building exterior, but there was no way to confirm the behavior deep within the building.

So they clearly admit that Case A fits the available evidence. However, they must contradict the visual evidence in favor of an estimate that "there was no way to confirm."

Not only is this an argument from ignorance--in that they appeal to an unknown factor to increase the fuel load--they must actually contradict the available visual evidence to make this appeal.

If a "twoofer" was making this kind of a case for ...whatever, you'd jump his **** so fast, he'd get a headache. And rightfully so. This is a weak argument for not only appealing to an unknown in order to justify what appears to be a predetermined conclusion, but contradicting the evidence at hand. The real question is "why doesn't this argument bother you?"

The assumptions made for all Cases fell within the range of uncertainty resulting from the impact analyses; the objective of the parameter changes was to generate results that would span as much as possible the set of probable outcomes given the range of plausible initial and boundary conditions

So what? Maybe you'd like to try to interpret this sentence in such a way that would justify the fuel load increase? I do not see any real answer here. Please help. The “set of probable outcomes” should be bound by reasonable estimates of fuel loads. This doesn’t explain why they decided to increase these estimates willy-nilly. Please help.

And just because no one here can answer a question, does that prove that the question cannot be answered by anyone?

Not necessarily, but it does make clear that you are willing to believe something that you cannot logically justify, i.e. you are treating the NIST theory on the same grounds that creationist treat the age of the universe: on Faith.

You admit that you can't find a reason for the increase, yet you still believe. The fuel load issue is not tangential: the entire argument hinges on it. Yet you cannot come to terms with the possibility that you've been sold a pile of *****.

Twinstead,

Yes, I am indeed saying that no one here can make sense of the fuel load increase or the fuel load redistribution in Cases C and D. If they could, I would like them to make their case, please.

Because no one here seems to be able to make sense of these pivotal problems, it is strikingly clear that you and others are clinging to the NIST report like a safety blanket. The fact is that despite the evidence that the NIST report could only get the steel to fail in exaggerated simulations, you will not even accept the possibility that the NIST report might be wrong.

You are not operating on logic and reason, you are making your judgments in this particular case on faith.

You cannot find fault with my reasoning--in which I show that the lynch pin of the NIST explanation is based on an unjustified assumption--yet you cling to it for dear life.

Again, this is exactly the kind crap thinking that you are all too happy to point out when it applies to a different conclusion. When it applies to your own conclusion, you tell me to take it up with the experts--so long as I leave you alone.

Your (and others') faith in the NIST report is a psychological issue: not a logical one. No one here can justify the fuel load increase, (because there is no justification for it) yet you refuse to come to terms with even the possibility that a government-funded agency would try to sell you something. Wake up.

 

[TellyKNeasuss]

And yet, no one here can squarely reconcile the fuel load increases.
.
.
.
Now, if 4 lb/sq ft is a reasonable estimate given all the info known about the towers, what possible reason do they suppose that they underestimated this by 25%? That's a significant increase. That's 60 tons per floor versus 75 tons per floor. Where do you suppose an extra 15 tons of fuel per floor came from? Wouldn't you think they'd give some kind of justification for an extra 15 tons per floor?

So they clearly admit that Case A fits the available evidence. However, they must contradict the visual evidence in favor of an estimate that "there was no way to confirm."

Not only is this an argument from ignorance--in that they appeal to an unknown factor to increase the fuel load--they must actually contradict the available visual evidence to make this appeal.

You're assuming that because they didn't specifically state the reason for re-running the simulation with a higher fuel load that there must be something sinister about it. Did you consider any other possibilities? Such as that people knowledgeable about fire simulation modeling would not need the reasoning spelled out? That it might be common in situations in which the true input data is not known to essentially work both forwards and backwards from the middle to try to find realistic estimates of the proper input data that enable the model to produce results that match the observations? That it might be acceptable to increase the fuel load to compensate for other heat sources (e.g., Dr. Greening's PVC reactions, the heat from burning aircraft parts) that could not be input into the model? Do you have any experience in numerical simulations that would enable you to answer these questions? Have you researched fire simulations or contacted any fire engineering experts? Have any experts complained about the way NIST handled the simulations?

It might be just my strange way of thinking, but if I were fudging input data into to produce a desired result, when I wrote the paper I would portray my fudged data as the best values for the input.

I will consider posing this topic to my niece, who is working on a Ph.D. in structural engineering and has taught steel design classes. However, when I've mentioned conspiracy theories in the past she has said that she considers them so ludicrous that she has no interest in discussing them.

 

[cmcaulif]

you missed the point entirely jay. You are clinging to the fuel load increase, but you forget this, along with the other adjusted factors for the more severe case, were done to produce a scenario with higher temps at the core, within the limits of the FDS software. They specifically wanted to see what would happen if heat was trapped in the core(the added fuel load actually slowed the spread of fire as well), which is why they made these decisions.The fuel load increase is merely a way to produce a desired environment with the available resources.

 

[jay howard]

you missed the point entirely jay. You are clinging to the fuel load increase, but you forget this, along with the other adjusted factors for the more severe case, were done to produce a scenario with higher temps at the core, within the limits of the FDS software. They specifically wanted to see what would happen if heat was trapped in the core(the added fuel load actually slowed the spread of fire as well), which is why they made these decisions.The fuel load increase is merely a way to produce a desired environment with the available resources.

So effing what? What is your point? That they wanted to play a $20M game of "Rampage"?

The salient point in all this is that the "most influential factor" in the simulations had to be arbitrarily and significantly increased in order to produce the white whale of their explanation: collapse initiation.

Without increasing these values 25%, there is no simulated collapse. It's possible their entire assessment is fundamentally wrong. However, if their fundamental assessment is acceptable, then why do they go against it in order to simulate the collapses?

You are on the horns of an uncomfortable dilemma: either

1. NIST's assessment of the conditions in the towers is flawed--in which case, we can't say whether 4 lb/sq ft is a reasonable estimate of the fuel load or not,

or

2. NIST's assessment of the conditions in the towers is reasonable, however, they must ignore and contradict data regarding the "most influential factor" in the simulations in order to produce a collapse.

So, either you have no simulated collapse with their reasonable data, or you have incompatible data with collapse initiation. You seem to want to say that they have good data AND they get to ignore their good data without any justification to produce a simulated collapse initiation. If this is what you mean to say, then you have removed their approach entirely from the scientific realm by making their conclusion irrefutable.

Despite the alluring sound of "irrefutable," it is not a desirable quality of theories. Astrology is irrefutable. Marxism is irrefutable. T.v. psychics are irrefutable. You do not want your cherished theory to be grouped in with these by making it irrefutable.

 

[cmcaulif]

Once again, you have ignored what I said, kept clinging to the fuel load increase despite this, and have thrown in more snide and pointless comments. This of course, is why most posters don't bother with you anymore.

The temperatures produced by in the core were within the ranges that they found in the any experiments they did within the subproject. To recreate a more severe case, using the FDS software, which has limitations, they decided to increase the fuel load. They are not claiming that an increased fuel load is needed for collapse, the load is only increased because that produced the sort of outputs the NIST wanted to examine, outputs which still remain within experimental uncertainty. So they have a simulation with reasonable data, albeit with limitations and uncertainties, which they are quite clear about.

If you are going to simply ignore what I'm saying and continue to repeat yourself, don't bother to respond, like many other posters here, I have no interest in talking to a brick wall. You can drop your snide remarks as well(my cherished theory is the one proposed by Arup/Edinburgh, not the NIST by the way)

 

[jay howard]

You're assuming that because they didn't specifically state the reason for re-running the simulation with a higher fuel load that there must be something sinister about it.

It's not a matter of "something sinister". That is a moral assessment. I'm making a determination about the structure of the NIST theory for the collapses. If I've been unclear about this, let me disabuse you of any misunderstandings on this: at some point, someone involved with the NIST Report on the WTC collapses probably knew that their theory hinged on unsubstantiated speculation. But that's speculation on my part. I'm speculating that the people who compiled the data and created the conclusions were smart enough to know that their data did not support their conclusions.

Is that an "unsubstantiated speculation"? Not really, and it's not pivotal to my criticisms anyway. Why they went ahead with the report without any clarification about the nature of these assumptions is unclear. I cannot make a good determination about their motivations. Are they morally flexible? It's possible. It's also possible that they were manipulated or coerced by more powerful forces. However, it's not very likely that they just didn't understand that the data did not support the conclusions. People are complicated. But it doesn't change the fact that their data does not support their conclusions.

Did you consider any other possibilities? Such as that people knowledgeable about fire simulation modeling would not need the reasoning spelled out? That it might be common in situations in which the true input data is not known to essentially work both forwards and backwards from the middle to try to find realistic estimates of the proper input data that enable the model to produce results that match the observations?

Sure, I considered that. All I needed to do was to find anywhere in any of the reports where it said 5 lb/sq ft fuel load was substantiated by any evidence or reasoning whatsoever. What I found were accounts to the effect that increasing the fuel load above 4 lb/sq ft would contradict multiple data sets: photo/video observations, fire-spread rates, and timing in the simulations--not to mention the caveat that fuel load was "the most influential factor" in the simulations.

In other words, the experts at NIST provided various substantiations for the fuel load of 4 lb/sq ft as well as reasons NOT to increase it. Why they went ahead and increased it anyway is never made clear--other than to "create a more severe fire".

On the other hand, you have no reason to doubt their fuel load estimate AND no reason to accept a higher estimate. Further, you should WANT to know why they increased this estimate, but instead you tell me I should look harder for a reason to justify it. I've looked for a reason to justify it. If it's there, please point it out. You don't have one, yet you accuse me of not looking hard enough to find one. The fact is, neither one of us can find a good reason why they went against their own estimates in order to produce the collapse initiation, except that they needed to produce a collapse initiation.

Are you still convinced that the fires burned hot enough for long enough to produce the collapses? Of course you are.

Why? Do you have a reason that doesn't hinge on your faith in other people and their titles?

That it might be acceptable to increase the fuel load to compensate for other heat sources (e.g., Dr. Greening's PVC reactions, the heat from burning aircraft parts) that could not be input into the model?

Your people at NIST already thought about that:

“… it would have been possible to further refine or ‘tune’ the properties of the simulated workstations to better match the experiments, but this would have been a waste of time and resources since the workstations were merely convenient surrogates for a variety of other combustible furnishings found throughout the buildings. Indeed, sensitivity studies revealed that the total mass loading of combustibles was more important than its composition.”(NCSTAR 1-5F, 36)

Do you have any experience in numerical simulations that would enable you to answer these questions? Have you researched fire simulations or contacted any fire engineering experts? Have any experts complained about the way NIST handled the simulations?

It might be just my strange way of thinking, but if I were fudging input data into to produce a desired result, when I wrote the paper I would portray my fudged data as the best values for the input.

It looks like it was fudged well enough to convince the "skeptics" here at jref. Maybe instead of asking me to look harder for a reason why my criticisms aren't valid, you should try a little harder to look for a reason why the NIST report's conclusions make sense.

This is the whole point of falsifiability: if, after your theory fails to predict expected outcomes, and you still think it's a good explanation, you can still cling to your theory, but you do so by making it irrefutable.

What, in other words, would convince you that the NIST explanation is deficient? If you answer "nothing", then you are not participating in a scientific endeavor--you are acting on faith. Faith is for you and your family--not for data analysis and theory construction.

 

[jay howard]

Once again, you have ignored what I said, kept clinging to the fuel load increase despite this, and have thrown in more snide and pointless comments. This of course, is why most posters don't bother with you anymore.

You're breaking my heart. I couldn't care less if you responded, yet you continue to carry on this conversation as if it had an effect on you. It's clear that despite a lack of directly corroborative forensic evidence and manipulated simulation data, you will not even consider the possibility that the NIST conclusions maybe just might be overstated.

I'm open to change my mind about the NIST conclusions if IF any substantive reason could be provided for the increased estimates used in the simulations. Yet, none have been pointed out. The piss poor attempts presented here don't come close to overcoming this fundamental problem. However, after being given reason after reason for why they SHOULD NOT HAVE changed their fuel load estimates, you still won't even consider the possibility that NIST's conclusions are flawed.

The temperatures produced by in the core were within the ranges that they found in the any experiments they did within the subproject.

What? "Temperatures produced by in the core" where? In the simulations? Where was the majority of the fuel located? In the cores? No. The majority of the fuel was located between the cores and the exteriors. They want to explore different damage estimates such that the soffits remained? Fine. That doesn't provide any grounds to increase the fuel load estimates by 25%.

To recreate a more severe case, using the FDS software, which has limitations, they decided to increase the fuel load. They are not claiming that an increased fuel load is needed for collapse,

In fact they are:

“The Investigation Team then defined three cases for each building by combining the middle, less severe, and more severe values of the influential variables. Upon a preliminary examination of the middle cases, it became clear that the towers would likely remain standing. The more severe case (which became Case B for WTC 1 and Case D for WTC 2) was used for the global analysis of each tower.”( NCSTAR 1, 144)

Just not explicitly. The fact remains that fuel load was determined to be the "most influential factor" in the simulations, and the middle cases "would likely remain standing". The fact is, the "more severe case" used for global analysis increased this fuel load without any apparent justification other than to "create more severe fires".

the load is only increased because that produced the sort of outputs the NIST wanted to examine, outputs which still remain within experimental uncertainty. So they have a simulation with reasonable data, albeit with limitations and uncertainties, which they are quite clear about.

How are you making the determination that these increased data parameters are "reasonable"? This is the part you are ignoring. You clearly have NO REASON to believe this, yet you believe it anyway for no other apparent reason than you want to believe it. That's called faith. You cannot argue from faith. And you cannot be convinced from faith either. If we were having a rational scientific discussion based on evidence and reason, you would have to admit that there is indeed no reason to assume higher fuel loads in the towers, yet the simulations use higher estimates to produce "collapse initiation".

 

[cloudshipsrule]

Jay,

You are wasting your time typing your crap. It does NOT matter what you type. If you dislike NIST's report then please point us to a more thorough investigation with better simulations that you actually like.

Waiting..............

What? There isn't one?

Well then, we'll just have to look to the NIST simulations until the TWOOF movement does something better, huh?

I imagine we'll be waiting a looooooong time.


Here's the thing, Jay.

Planes hit the building. Damage occurred. Fires started. Fires weakened steel. The building was NOT designed to withstand the damage it sustained, and it fell just as Mother Nature intended. (You know, gravity and inertia and all.) It's really not rocket science. It's just basic science with complicated variables.

Give it up. You're wasting your time.

What are you trying to say anyaway?

Explosives were planted? ZERO EVIDENCE.
Accelerants were used? ZERO EVIDENCE.
What else could you possibly be hinting at?

All of the evidence points to exactly what the rational individuals on this forum believe.

Gravity is the root of collapse initiation. Oh yeah, fire and structural damage helped too.

 

[cmcaulif]

Once again, drop the snide remarks and pointless comments on 'faith'. What is the point of that? I have stated what I think is potentially the best theory, so OBVIOUSLY I have considered that the NIST's conclusions are flawed.

What? "Temperatures produced by in the core" where? In the simulations? Where was the majority of the fuel located? In the cores? No. The majority of the fuel was located between the cores and the exteriors. They want to explore different damage estimates such that the soffits remained? Fine. That doesn't provide any grounds to increase the fuel load estimates by 25%.

And the disturbed combustibles were weighted towards the core, not evenly around the floor space, which is significant, AND the fuel load increase slowed the spread of the fire to concentrate it near the core to some degree. This was the case they wished to examine.

In fact they are:

...

Just not explicitly.

...

I realize the point you are making, but you are ignoring the my previous posts to make it. The simulation was very sensitive to the input of the fuel load. The temperatures produced by the simulation were within the limits of the laboratory experiments they performed. They increased the fuel load to produce the scenario they wanted to examine, but this is because a simulation can never perfectly capture real life, not because they thought an increased load was necessary.

How are you making the determination that these increased data parameters are "reasonable"?

1. Become familiar with FDS, its limitations, and what inputs the simulation is sensitive to.

2. Become familiar with the experiments performed in the subproject, the ones most relevant to this discussion are 1-5C and 1-5E.

3. Realize that there is a difference between 1, 2, and an uncontrolled environment such as the WTC.

4. Attempt to understand the uncertainties involved in all of the above, and make a judgment from there.

This is the part you are ignoring. You clearly have NO REASON to believe this, yet you believe it anyway for no other apparent reason than you want to believe it. That's called faith. You cannot argue from faith. And you cannot be convinced from faith either. If we were having a rational scientific discussion based on evidence and reason, you would have to admit that there is indeed no reason to assume higher fuel loads in the towers, yet the simulations use higher estimates to produce "collapse initiation".

If we were having a rational discussion you would drop these retarded accusations, and you might actually address my overall point, other that to say 'so effing what?'. I have said that the justification for the fuel load increase is based on the sensitivity of the software, and does not translate to an explicit or implicit claim that there was a necessary increase in fuel load in reality. I don't know why you refuse to acknowledge that a software package may not be able to capture a given event perfectly. It responds to different inputs, and the outputs it produced were still within the experimental ranges. I expect your answer to be more of the same, so this will probably be the last post from me on the topic.

 

[DGM]

Jay:
I'd sure like to see your controlled demolition evidence that can pass the extreme scrutiny your putting this through. Can you post this for us? After all if it can pass your review it must be good. Thanks in advance.

 

[jay howard]

I have stated what I think is potentially the best theory, so OBVIOUSLY I have considered that the NIST's conclusions are flawed.

Really? In what ways specifically, do you consider NIST's conclusions to be flawed?

I have said that the justification for the fuel load increase is based on the sensitivity of the software, and does not translate to an explicit or implicit claim that there was a necessary increase in fuel load in reality. I don't know why you refuse to acknowledge that a software package may not be able to capture a given event perfectly. It responds to different inputs, and the outputs it produced were still within the experimental ranges. I expect your answer to be more of the same, so this will probably be the last post from me on the topic.

If only this was your last response.

But again, in your words above, there is no clear justification for the fuel load increase.

I know perfectly well that the FDS software "may not be able to capture a given event perfectly." How does that answer the question of why NIST increased the fuel loads? It DOESN'T. period.

 

[jay howard]

Jay:
I'd sure like to see your controlled demolition evidence that can pass the extreme scrutiny your putting this through. Can you post this for us? After all if it can pass your review it must be good. Thanks in advance.

I'd sure like to see where you gathered that I'm making such a claim.

Can you post that? No. Because it's not the point I'm making.

This is the typical strategy here: reduce all alternatives to some kind of absurdity, and HURRAY! You win!

That's why this thread is about the theoretical fortitude of the NIST Report. I've shown time and again here that NIST's conclusion do not follow from their own data. Yet no one here is willing to say "hey, jay, you've got a point. Maybe we should look for a theory that better fits the data."

That's how scientific deliberation happens.

If you have a reason to suspect the criticisms of the NIST report I've presented are flawed, I'll take them seriously and respond as rationally as possible. If you want to dismiss what I say because you don't think alternative theories hold water, you have not successfully overcome my criticisms.

If you still believe in the conclusions of the NIST report but you cannot overcome these criticisms, you are participating in an article of faith. There is no way around this.

 

[DGM]

I'd sure like to see where you gathered that I'm making such a claim.

Can you post that? No. Because it's not the point I'm making.

This is the typical strategy here: reduce all alternatives to some kind of absurdity, and HURRAY! You win!

That's why this thread is about the theoretical fortitude of the NIST Report. I've shown time and again here that NIST's conclusion do not follow from their own data. Yet no one here is willing to say "hey, jay, you've got a point. Maybe we should look for a theory that better fits the data."

That's how scientific deliberation happens.

If you have a reason to suspect the criticisms of the NIST report I've presented are flawed, I'll take them seriously and respond as rationally as possible. If you want to dismiss what I say because you don't think alternative theories hold water, you have not successfully overcome my criticisms.

If you still believe in the conclusions of the NIST report but you cannot overcome these criticisms, you are participating in an article of faith. There is no way around this.

I'm not trying to win anything. It sounded to me that you had a better theory, that's all. I'll take your answer as a no.

 

[cmcaulif]

Really? In what ways specifically, do you consider NIST's conclusions to be flawed?

I meant to say may be flawed, I would need to see more work on the Arup/Edinburgh hypothesis before I accept it fully, though I have corresponded with some people from Arup on this. To answer your question, I believe that it is highly likely that the effects of thermal expansion are under weighted in the NIST global models. The reason for this is because these effects are included for support columns, but not for floor trusses. Arup/Edinburgh show that these effects would be quite significant, and would result in collapse without structural damage and with lower fire temps, though, of course, there are limits to there models as well. In general, I would like to see more research into this and I am happy that Dr. Quintiere of Univ. Maryland has proposed a review of the NIST, in which some of these things migt be explored.

If only this was your last response.

this is what I have been saying from the start, the actual load increase some sinister deed as you seem to think.

But again, in your words above, there is no clear justification for the fuel load increase.

I know perfectly well that the FDS software "may not be able to capture a given event perfectly." How does that answer the question of why NIST increased the fuel loads? It DOESN'T. period.

I gave you a four step mini journey which you can take that could perhaps satisfy you. Hopefully at the end, you will, at the very least have calmed down, or maybe even figured out that they just produced a scenario within the boundaries of their experiments, using the inputs sensitive to the software.

 

[chipmunk stew]

I'm speculating that the people who compiled the data and created the conclusions were smart enough to know that their data did not support their conclusions.

The 4 lb/sq ft figure is not a data point. It's a "reasonable estimate" based on data (which, given the size and complexity of the problem, is bound to be incomplete.) It gave them a starting point for building the model. The increase in fuel load was found to fit the observed data better than the 4 lb/sq ft estimate. This type of iterative model-building is typical in engineering analyses. R. Mackey illustrated this beautifully a number of pages back.