View Single Post
Old 19th August 2007, 04:27 PM   #785
Join Date: Apr 2006
Posts: 7,854
Originally Posted by rwguinn View Post
Ever stop to think that an airplane full of JET-A might contribute a bit to the situation.
Note that they said ""Simulations performed with doubled fuel loads slowed the fire spreadwell below the observed rates. Combined with the above results, this suggested that the estimated overall combustible load of 4 lb/sq ft was reasonable."(NCSTAR 1, 129)"
Normal fire simulations use a spread from a source location (Someone with credentials correct me if I'm wrong). The JET-A would have spread the firea lot more quickly--requiring an increase in fuel load in the sim to match observations.

The most important thing is that the sim must match reality. The other way around never works.
So, with that demo of reading incomprehension on the part of a twooist, I am done here.
Actually, it's simpler than that.

The quote that "the Towers would likely remain standing" is based on the preliminary models -- and also doesn't mean "likely to remain standing forever." It actually means that, under those cases, the Towers would still have collapsed, but it would have taken longer than it did in reality.

As it turns out, the fuel loading is actually not that relevant. More relevant is the fuel placement, and how the fuel load determines the location -- not the intensity -- of the fires.

An increase in fuel load from 4 lb/ft2 to 5, i.e. 25%, is small compared to the increase that was tested during model sensitivity testing. That sensitivity testing, in which the fuel load was increased by 33%, found that the fuel load was relevant but only with respect to the duration of fires. See NCSTAR1-5F section 5.2 for the sensitivity analysis.

The overall effect of fuel load on the fires is described in Chapter 6 of NCSTAR1-5F:

Originally Posted by NCSTAR1-5F page 78
The results of the Case B simulation of WTC 1 are included on the following pages (Figs. 6-18 through 6-25). In general, the results were similar to Case A because the fires in WTC 1 were limited by the supply of air from the exterior windows. As the window breakage pattern was not changed in Case B, the extra combustibles within the building did not contribute to a larger fire, but they did delay the spread slightly because the fires were sustained longer in any given location due to the increase in combustible load.
In Case C and D, of course, there was no change in the combustible load at all:

Originally Posted by NCSTAR1-5F page 100
Unlike WTC 1, the designated combustible load in WTC 2 had a noticeable effect on the outcome of the simulation. Because most of the windows on the impact floors were broken out by the airplane debris and the ensuing fireball, there was an adequate suply of air for the fires. In the Case D simulation of WTC 2, the combustible load was kept at 20 kg/m2, the same as case C, but the aircraft debris and "rubble" were spread out over a wider area. In Case C, the debris pile was concentrated in the northeast corner of the 80th, 81st, and 82nd floors, whereas in Case D the pile was less concentrated. Also, in Case D, the furnishings away from the impact areas were assumed to be undamaged.
Therefore, the complaint about increasing the fuel load in WTC 1 is groundless, and stems from lack of comprehension about the report.

It is similarly incorrect to state that the fuel loading or placement defined Cases A and C versus B and D. More importantly, B and D involved more impact damage, especially fireproofing damage. That makes a far greater difference than any change in the fuel load.

The reasons to prefer the higher impact damage estimates are given in NCSTAR1-2B. We've discussed them to death in this thread (skipping most of the crap in that thread), and besides that cases B and D were seen to better simulate the leaning and floor behavior than cases A and C. This is why A and C were rejected:

Originally Posted by NCSTAR1-6, 8.3, pg. 235
WTC 1 and WTC 2 global models were subjected to Case B and Case D aircraft damage and fires, respectively. The results of the isolated wall, core, and full floor analyses indicated that structural responses to Case B and Case D more closely matched observed structural behavior in photographs and videos than did Case A or Case C respectively.
For more information about this, please see Chapter 7 of NCSTAR1-6.
R.Mackey is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top