The 9/11 Forum

Original post

femr2 wrote:

Trippy wrote:Consider the point that an over-pressure isn't actually required to generate these dust jets

My point is to highlight to OWE and DBB that if the linear ejecta detailed was simply air being expelled, then having a focussed jet and the south side ejecta at the same time would be reasonably described as mutually exclusive. Data and opinion on the dust jets themselves is probably best held within a specific thread.

I'm not so sure of this, but you'll have to elaborate on your reasoning here.

femr2 wrote:As you've gone to some effort, I'll respond likewise, but again suggest as much specicivity in threads. (Especially ones I start )

Thanks, I appreciate it.

femr2 wrote:

WTC 1&2 were 63.4m on each side

63.0936m external. Less internally (using window insets).

I'm generally quite time poor, so sometimes I have to make do, and this was one of those occasions, thankyou for your correction, and the additional information.
Just FYI:
208'*12"*2.54/100 = 63.3984m = 63.4m (3sf)

femr2 wrote:

, and the floors are 3.66m apart (both figures rounded to 3sf)

3.6576m for standard floor, though of course 0.1016m of concrete slab must be subtracted outside the core (for standard floors), and 0.1270m within the core, also taking into account that the external floor region and core region did not descend together as a single piston.

As I stated, both figures were rounded to 3sf 3.6576=3.66 when rounded to 3 significant figures.
As far as the piston thing goes, I know, hence the effort I went to to try and emphasize this as a source of error.

femr2 wrote:(Thinking of piston... http://www.youtube.com/watch?v=acTcwA_YHuA ...motion detection by subtraction.)

Dam, I was working on something very similar to this, I was also using XOR as well as subtraction.

femr2 wrote:

this gives us a total volume of air between the floors of 63.4*63.4*3.66 m^3 or 14,700 m^3 (again, rounded to 3sf).

Rather less, but...

Not as much as you might think - technically I should be using a weighted average based on the percentage of floor occupied, but I lack the time to calculate that just now, however, using a simple average for floor thickness (I considered subtracting the floor thickness, but lacked the time to try and track down the information, never came back to it), and your corrected figures still gives us a shade of 14,000 m^3

femr2 wrote:

The first thing to realize is that this volume includes the space occupied by the core columns, which introduces an error in the form of a volume of air that will be included in the calculations, but can not be expelled by the collapse.

Second thing to realise is that the volume taken up by ALL of the other SDL, CDL and LL masses must also be accounted for, as do all air flow restrictions imposed by structure, partitioning, ...

I considered that as well, which would be one o fthe reasons that I went to the pains that I did to try and emphasize that this is only an approximation, although arguable my initial rounding implicitly allocates nearly 12m^3 to this (which is probably inadequite, however) and the figure of 14,000 m^3 (also rounded to 3sf) allocates about 20m^3 to this.

femr2 wrote:

Now, if we assume we are dealing with the first collapsing floor, and, for the sake of simplicity, assume that it falls as one intact piece onto the floor below it

An over simplification in terms of the subsequent justification IMO.

, then we can calulate (using s=(gt^2)/2 that it takes 864ms (3sf) for an object to free fall 3.66m, this tells us that we have 864ms for all 14,700 m^3 of air to be expelled from between the floors.

That would assume zero air-resistance, and also other resisting factors, further distorting the calculation, and assuming free-fall...hm.

If you have a better figure to use, and a reason why it should be used...

And no, IMO using 0.7g is not neccessarily a more accurate value (I considered that as well).

femr2 wrote:

Calculating a flow rate is simple, it's simple the volume moved/time taken, so in this case 14,700/0.864 which is 17,000 m^3/s (3sf).

Now, when we're dealing with a pipe, the flow velocity of the material in the pipe is equal to the flow rate/cross sectional area of the pipe. In the case of WTC 1&2 the pipe isn't a pipe per se, however there are a number of apetures that the air can move through, represented by the voids in the core

There are some core corridors which may have played a part, but that is why I suggest a separate thread, as detailing each jet individually is important to understanding their nature.

Agreed, corridors might have served to focus the flor through a few of the windows, which would increase the flow velocity, by reducing the cross sectional area under consideration - thus this simplifying assumption serves to reduce the flow velocity, and provides balance to the previous assumptions that i've made which might serve to increase the flow velocity.

femr2 wrote:

, and the windows.

Ah. For your purposes, the windows cannot be treated as an aperture really, as the focus must be produced elsewhere IMO (core corridors for example).

Focus is a seperate issue, and not something that I was generally considering (as I thought was clear when I stated the initial considerations - see above for further information).

{snip}

femr2 wrote:

But yes, what i'm suggesting is that dust jet ejections like that don't neccessarily require an over pressure to occur.

Being more specific, the dust jet being highlighted (for the purpose highlighted) occurs at the same level as the linear ejecta. Other dust jets occur in front of the linear ejecta. That makes no sense at all. All behaviours must be explainable. The linear ejecta represents the very lowest *crush front* and shows firstly that internal destruction was asymmetrical and did not result from complete storey collapses, but instead the south side descent was in advance of the north. Such behaviours mess with the simplifications significantly.

To me it fails to make sense only if you're assuming that the external collapse is happening in the same place as the internal collapse (i've never been a huge fan of the solid block model) or, to put it another way, it potentially makes sense if you stop assuming that the internal collapse and the visible collapse initiated at the same time.

(if that makes sense).

Dust jets are the result of an overpressurization of an enclosed space, no?

Therefore jets which favor specific locations (windows) to pop out of must be explained via a mechanism showing why the ejecta "chose" only particular windows from which to emerge.

Explaining those ejecta via pressurization of a floor space only makes no sense.

For example, why would dust jets as the result of floor overpressurization from the OOS w, sw destruction choose to come out only from the sw portion of the west face?

This must be the front of physical destruction and not just trapped air.


I think any thread dealing with dust jets should first look at the mother of all dust jets, shown in the linked video.

http://www.youtube.com/watch?v=acTcwA_YHuA

This extreme case should be remembered when studying isolated dust jets in general.

femr2 wrote:

Trippy wrote:I'm not so sure of this, but you'll have to elaborate on your reasoning here.

Well, the suggestion (from DBB) was that the linear/terminal velocity ejecta was not a *crush front* but instead a dust ejecta, similar to the more focussed *dust jets*. My viewpoint is that for both phenomena to be occuring around the same floor would seem to be mutually exclusive actions. Dust jets also occur ahead of the linear ejecta, so they are not directly correlated. If both were pressure based, then one would affect the other, would you not agree ?

Potentially, however, does that assumption not require treating the air between the floors as a single contiguous air space, rather than discretely packaged cells?

I imagine, for example, that some windows had passages connecting them directly to the core space, where others may have been effectively isolated from the floor space (consider the scenario posited below).

femr2 wrote:

Dam, I was working on something very similar to this, I was also using XOR as well as subtraction.

It's a useful method. Any view upon the implications and conclusions of the video ?

You mean aside from rash and hasty? (I was taking a more conservative approach, because i'm all to aware of the potential for errors due to changes in lighting).
Consider what no_body has had to say - that there is a rumble coinciding with the first movement of the antenna.
This video seems to imply that the Dust Jet began at the same time as the antenna moved, but was initially weak, and later strengthened.
Consider now my explanation for both the rumble and the ejecta - some degree of floor A falling onto floor B.

What I would like to know is if anybody jumped from that level - consider what we've been talking about with corridors focussing flows and 'artificially' increasing flow velocities, combined with the fact that right from the get go, we've agreed that air is going to be forced into the core of the building, as well as out through the windows.

And then consider what would happen if at some point in the previous 80 minutes (or however long the building burned for) someone who, for what ever reason, couldn't escape, and walked to the end of a corridor that effectively connected the core to the perimeter, opened a window, and threw themselves out (rather than face the alternative).

Once movement started above that point, and air started being displaced, would that corridor and open window not now form a path of least resistance for the rapidly building over pressure in the core?

Yes, it's speculation, but it has the advantage of being verifiable speculation (although I lack the time to undertake that endeavour) - it's verifiable (or potetially so) in that the person leaping from that window may have been caught on camera, and floor plans, including partioning may be available.

femr2 wrote:

And no, IMO using 0.7g is not neccessarily a more accurate value (I considered that as well).

Hmm. Perhaps the linear/terminal ejecta velocity would be a good value...25m/s to 29m/s.

The thing with that is it takes 136ms for an object to fall 3.66m at 27m/s, which means that you're now dealing with a flow rate almost 8 times bigger (14,000 m^s in 136ms = 103,000 m^3/s = flow velocity of 96 m/s, which is just shy of three times stronger than what's required for a hurricane force wind.

femr2 wrote:

To me it fails to make sense only if you're assuming that the external collapse is happening in the same place as the internal collapse (i've never been a huge fan of the solid block model) or, to put it another way, it potentially makes sense if you stop assuming that the internal collapse and the visible collapse initiated at the same time.

I'm pretty sure that internal descent initiated early, particularly for core members (I'm on the other side of the fence as I'm sure you are aware ). I'm fully comfortable with all sorts of internal processes occurring which are not directly visible from the exterior. The linear/terminal velocity ejecta certainly appears to behave as a *crush front* rather than anything describable as simply dust ejecta. I uploaded a better quality version of the BBC footage which lets you take a closer look:
http://www.megaupload.com/?d=CVDEAAJL

Awesome sauce.
I'll have a look at the video when I get home from work.

femr2 wrote:

Trippy wrote:Potentially, however, does that assumption not require treating the air between the floors as a single contiguous air space, rather than discretely packaged cells?

Not specifically, as the behaviour remains consistent over many floors. There is no known OOS partitioning which remains consistent over such a number of floors. Worth considering in more detail though...

I imagine, for example, that some windows had passages connecting them directly to the core space, where others may have been effectively isolated from the floor space (consider the scenario posited below).

Quite possibly, and I suggest that the location of each major dust jet is looked at in specific detail.

This is part of what I was alluding to/suggesting. Yes.

femr2 wrote:

You mean aside from rash and hasty?

No, I mean any opinion

In that case, for the time being at least, 'Rash and hasty' shall suffice.

femr2 wrote:

(I was taking a more conservative approach, because i'm all to aware of the potential for errors due to changes in lighting).

Hmm. The effect of lighting change should not affect the results of the subtraction test, which showed very early lateral ejecta at the dust jet location. Such pixel behaviour could not IMO be caused by change in lighting.

Quite. However I'm not attributing the behaviour of the dust jet to changes in lighting - merely pointing out that motion isn't the only thing that can give a result (in other words, pointing out a 'false positive' source), although, the subtraction test is simply based off the RGB values of the pixel, it's a measure of the difference in the colour of the pixel between frames (I have two sets of subtraction data for this reason - difference between subsequent frames, and differences between each frame and the first frame in the set i'm using).

femr2 wrote:

This video seems to imply that the Dust Jet began at the same time as the antenna moved, but was initially weak, and later strengthened.

That is what the subtraction test would appear to show, so it sounds fairly reasonable to me.
Replicating the process would allow clarification.

I'll have another look at what I've got when I get home. In all honesty, I was more interested in the initial movement in the impact zone that I noticed, than I was the dust jets.

femr2 wrote:

What I would like to know is if anybody jumped from that level...

Whoa. Would have to stop you there. I'm certainly resistant to thinking that the dust jets appeared when and where they did as a result of *jumpers* leaving the window open after they left. Geez

Again, quite, but I'm not positing this as a source for every jet (I don't posit that every jet neccessarily has the same source), i'm simply positing it as a source for this jet based on it's longevity.

femr2 wrote:I'm sure, absolutely, that a more technical explanation is more relevant, and am happy to begin documenting the specific jets in detail. (There are other threads here on the same subject. Will dig out conclusions)

Works for me.

femr2 wrote:

The thing with that is it takes 136ms for an object to fall 3.66m at 27m/s, which means that you're now dealing with a flow rate almost 8 times bigger (14,000 m^s in 136ms = 103,000 m^3/s = flow velocity of 96 m/s, which is just shy of three times stronger than what's required for a hurricane force wind.

Nevertheless, that is the *probable* rate at which (one of the) the *crush front(s)* traversed the height of the tower after the first 3 or 4 seconds.

Dude.
Seriously?
I had never contemplated that the winds could reach that sort of level.

femr2 wrote:Also think it prudent to drop in that airflow through the core itself is very restricted, limited to pretty well sealed lift shafts and stairways...additional aperture narrowing.

Quite. I was simply using DBB's 40% as an average figure/starting point.

femr2 wrote:If I was *air*, I wouldn't bother going through the core. I'd get out of the nearest window, whether I had to break it or not. If I had to travel upwards, fine. Easiest route for me

If you were air, you wouldn't neccessarily know the difference.
There has to come a point, as you move from the perimeter to the core where moving through the core is easier than trying to push the air between 'you' and the perimeter out of the way to get to it in time (think of it as the the air passing through the windows generating a back pressure which forces some air trhough the core. Also note that I have made no comments about what percentage goes in both ways, merely assumed that some will go in both directions.

The wall borad layers partioning (some of) the voids in the core certainly would not survive much overpressure. Niether would interior walls on the office floors. For that matter neither would stairwell fire doors and elevator doors.
One of the suvivors in the so-called lunar module reported being blown down several flights of stairs; see the writeup of his interview on firehouse.com.
Also consider the nature of the HVAC ducting; it provides open routes between the vertical HVAC ducts in the core and out to all parts of the ceilings.

femr2 wrote:

Dude. Seriously?

Well, yes. Can't avoid the rate at which each successive floor was destroyed I'm afraid, and the *crush front* would seem to traverse at that kind of rate (25->30m/s).

Yeah, I understand trhat :3 Although bear in mind that the 15 m/s was for near the beginning (initially in response to comments made about the expulsion of fire from below the impact zone when the antenna first moves),

femr2 wrote:

I had never contemplated that the winds could reach that sort of level.

Floors coming down at that rate...air has to give way at the same rate...
Is one reason why I am not keen on assumptions about vacuum based free-fall, as at that rate I would liken the behaviour of air towards a plastic solid than a gas, and treat it as an appropriately scaled resisting force. (Over a suitably short time period)

That's the advantage of the method that I'm using, is that it doesn't matter whether we're talking about air, water, or treacle. Moving 14,000 m^3 of anytthing through a pipe with a cross sectional area of 1,067 m^2 in 136 ms still implies a flow rate of 130,000 m^3/s and a flow velocity of 96 m/s. In this instance, the difference that viscosity and plasticity make is in how much work is required to do so (moving 14,000 m^3 of treacle that quickly would require a signifanctly larger pump that moving 14,000 m^3 of air).

Beaufort Force 12: ≥ 32.7 m/s Very widespread damage to vegetation. Some windows may break; mobile homes and poorly constructed sheds and barns are damaged. Debris may be hurled about. from
http://en.wikipedia.org/wiki/Beaufort_s ... dern_scale

Doesn't go anywhere near high enough.

Category 5 huricane is at least 69 m/s, from
http://en.wikipedia.org/wiki/Saffir-Sim ... Category_5
which reads Category 5 is the highest category a tropical cyclone can obtain in the Saffir-Simpson scale. These storms cause complete roof failure on many residences and industrial buildings, and some complete building failures with small utility buildings blown over or away. Collapse of many wide-span roofs and walls, especially those with no interior supports, is common. Very heavy and irreparable damage to many wood frame structures and total destruction to mobile/manufactured homes is prevalent. Only a few types of structures are capable of surviving intact, and only if located at least three to five miles (four to eight km) inland. They include office, condominium and apartment buildings and hotels that are of solid concrete or steel frame construction, public multi-story concrete parking garages, and residences that are made of either reinforced brick or concrete/cement block and have hipped roofs with slopes of no less than 35 degrees from horizontal and no overhangs of any kind, and if the windows are either made of hurricane resistant safety glass or covered with shutters.

Sounds as if office furnishings, partitions, interior walls and ceiling systems go out the windows or down the voids in the core.

David B. Benson wrote:Category 5 huricane is at least 69 m/s, from
http://en.wikipedia.org/wiki/Saffir-Sim ... Category_5
which reads Category 5 is the highest category a tropical cyclone can obtain in the Saffir-Simpson scale. These storms cause complete roof failure on many residences and industrial buildings, and some complete building failures with small utility buildings blown over or away. Collapse of many wide-span roofs and walls, especially those with no interior supports, is common. Very heavy and irreparable damage to many wood frame structures and total destruction to mobile/manufactured homes is prevalent. Only a few types of structures are capable of surviving intact, and only if located at least three to five miles (four to eight km) inland. They include office, condominium and apartment buildings and hotels that are of solid concrete or steel frame construction, public multi-story concrete parking garages, and residences that are made of either reinforced brick or concrete/cement block and have hipped roofs with slopes of no less than 35 degrees from horizontal and no overhangs of any kind, and if the windows are either made of hurricane resistant safety glass or covered with shutters.

Sounds as if office furnishings, partitions, interior walls and ceiling systems go out the windows or down the voids in the core.

I meant to look that up and post it to, because suddenly you have an explanation for people being knocked off their feet during the collapse (or rather, I meant to include that information in a reply to your previous post, but I was interrupted by work :3).

Trippy --- Yes, work does interfere...

BLGB points out that the overpressure could well be so extreme as to bend the office floor trusses, cracking the concrete and so providing another escape route, provided the floor pan also cracks. If so, I'm certainly willing to consider this effect as some sort of pre-crushing front, just in advance of the impacts of masses above.

NIST has written in a FAQ that one office floor truss could bear the weight of five (or six?) more. From that one might try to find the overpressure at which the lightweight concrete begins to break up due to the bending of the truss supporting such.

floor plan for floor 77:
http://911research.wtc7.net/wtc/evidenc ... -128_0.png
It turns out that some aspects of the floor plan were asymetrical, and so should we not expect the building to respond in an asymetric fashion?

Trippy --- On that floor plan, notice the AC ? ? Exhaust ? towards the lower left outside the core. It is a void of some sort throught the beamed floor (type 12).