QUOTE (metamars+Jan 11 2008, 01:57 PM)
Charles M. Beck is up to version 8 in his WTC avalanche paper.
Didn't he get laughed out of this forum? He came in talking about avalanches... I think I need to go back and read those posts again, it's been quite a while. Could be very interesting.
Didn't he get laughed out of this forum? He came in talking about avalanches... I think I need to go back and read those posts again, it's been quite a while. Could be very interesting.
QUOTE (OneWhiteEye+Jan 11 2008, 06:33 PM)
OK, thanks. It's interesting to see how the analysis system as a whole functions. No camera angle correction at all versus final perspective correction results in... no significant rank changes?
Hmmm. That is correct, noting the 'significant'. What changes is that good hypotheses fare better with the better corrections and bad hypotheses fare worse. So there is greater discriminatory power.
Hmmm. That is correct, noting the 'significant'. What changes is that good hypotheses fare better with the better corrections and bad hypotheses fare worse. So there is greater discriminatory power.
QUOTE (einsteen+Jan 11 2008, 12:53 PM)
... a thruther who probably thought I was one of the bad guys.
Haha. You should stick the knife in and twist it. Get the satisfaction of an OCTer, just once.
By the way, your recent post is not lost on me. I think it deserves a very close look but, for me, that could make it weeks off. So don't be surprised if I pull it back out of the closet at an unexpected time in the future. Just like the 'real' F4 data.
Haha. You should stick the knife in and twist it. Get the satisfaction of an OCTer, just once.
By the way, your recent post is not lost on me. I think it deserves a very close look but, for me, that could make it weeks off. So don't be surprised if I pull it back out of the closet at an unexpected time in the future. Just like the 'real' F4 data.
QUOTE (David B. Benson+Jan 12 2008, 06:00 PM)
What changes is that good hypotheses fare better with the better corrections and bad hypotheses fare worse.
This statement says a lot to me. It implies a meaningful convergence.
As I've mentioned, I'm at least as interested in the process as the results, maybe more so.
This statement says a lot to me. It implies a meaningful convergence.
As I've mentioned, I'm at least as interested in the process as the results, maybe more so.
Here is an important on-the-edge discrimination: Using C449 the poorer hypothesis is only 3.9 dB down, but using C447 we have
CODE
Sef-K+Z+ZSS-F-exp-pow-stretch dB= 0.0 sd= 0.101
Sef-K+Z+SS-F-exp-pow-stretch dB= 5.0 sd= 0.261
Sef-K+Z+SS-F-exp-pow-stretch dB= 5.0 sd= 0.261
which is 3:1 odds against, just enough to say that the poorer hypothesis is substantially disconfirmed.
This is important, because the poorer hypothesis is the one corresponding to crushing only occurring at the crushing front, versus the better, vertical avalanche, hypothesis.
By the way, Beck might have called his work 'avalanche', but he does not have the correct force function to make it so...
QUOTE (OneWhiteEye+Jan 12 2008, 11:10 AM)
It implies a meaningful convergence.
I over-simplified for the general trend. In the specific case of the SS hypothesis, before the optical axis correction it was significantly disconfirmed by both C447 and C449. Afterwards, it only failed just C447, as posted above. (I'm now running C448). However, the parameters were significantly different for the two data sets, which is viewed with deep suspicion.
I over-simplified for the general trend. In the specific case of the SS hypothesis, before the optical axis correction it was significantly disconfirmed by both C447 and C449. Afterwards, it only failed just C447, as posted above. (I'm now running C448). However, the parameters were significantly different for the two data sets, which is viewed with deep suspicion.
QUOTE (David B. Benson+Jan 12 2008, 06:11 PM)
Here is an important on-the-edge discrimination: Using C449 the poorer hypothesis is only 3.9 dB down, but using C447 we have
CODE
Sef-K+Z+ZSS-F-exp-pow-stretch dB= 0.0 sd= 0.101
Sef-K+Z+SS-F-exp-pow-stretch dB= 5.0 sd= 0.261
Sef-K+Z+SS-F-exp-pow-stretch dB= 5.0 sd= 0.261
which is 3:1 odds against, just enough to say that the poorer hypothesis is substantially disconfirmed.
This is important, because the poorer hypothesis is the one corresponding to crushing only occurring at the crushing front, versus the better, vertical avalanche, hypothesis.
OK, now, this is pure conjecture: what if an enormous body of data is eventually taken - for the dark band - and it all agrees very well with C449?
(I realize the superficial answer is to simply adjust your way of thinking but I'm considering the emergence of results as part of the results)
Another thing that occurs to me is that some functional forms are more capable generally of fitting classes of graphs than others. I offer that statement without explanation, for now.
QUOTE (David B. Benson+Jan 12 2008, 06:19 PM)
I over-simplified for the general trend. In the specific case of the SS hypothesis, before the optical axis correction it was significantly disconfirmed by both C447 and C449. Afterwards, it only failed just C447, as posted above. (I'm now running C448). However, the parameters were significantly different for the two data sets, which is viewed with deep suspicion.
The edge cases are fascinating to me. Again, my perspective is data-centric. The very minor final adjustment pushed a fence-sitter farther away. I can see how, on the whole, a few hypotheses on the edge are not exciting in the body of analysis but that's the interesting thing to someone who's trying to get a feel for the I/O.
I'll accept that. But he was the first I saw that had the idea, and the idea seemed quite novel and potentially useful at the time. Just from the conceptual level. My reaction was that dissipative forms in general make a lot of sense; from mechanics there are a lot of cases (n-body systems and non-conservative potentials? it's been a long time, my lingo is gone) which tend towards simplified solutions of a dissipative form. Behavior of ensembles of particles in the statistical limit.
The motion of a solid body through a fluid medium and an avalanche are quite different but both fall under this general study. What I propose is that the collapse of a large structure may be in this class of problems, but could be a sub-class in the sense of both of these above examples. A new functional form may be required, unless a closer analog has already been studied.
The edge cases are fascinating to me. Again, my perspective is data-centric. The very minor final adjustment pushed a fence-sitter farther away. I can see how, on the whole, a few hypotheses on the edge are not exciting in the body of analysis but that's the interesting thing to someone who's trying to get a feel for the I/O.
QUOTE
By the way, Beck might have called his work 'avalanche', but he does not have the correct force function to make it so...
I'll accept that. But he was the first I saw that had the idea, and the idea seemed quite novel and potentially useful at the time. Just from the conceptual level. My reaction was that dissipative forms in general make a lot of sense; from mechanics there are a lot of cases (n-body systems and non-conservative potentials? it's been a long time, my lingo is gone) which tend towards simplified solutions of a dissipative form. Behavior of ensembles of particles in the statistical limit.
The motion of a solid body through a fluid medium and an avalanche are quite different but both fall under this general study. What I propose is that the collapse of a large structure may be in this class of problems, but could be a sub-class in the sense of both of these above examples. A new functional form may be required, unless a closer analog has already been studied.
QUOTE (OneWhiteEye+Jan 12 2008, 11:25 AM)
OK, now, this is pure conjecture: what if an enormous body of data is eventually taken - for the dark band - and it all agrees very well with C449?
More data gives greater discriminatory power to the naive Bayes factor method. So instead of being just 3.9 dB down, the SS hypothesis would be further down. The dividing point is 5.0 dB down.
Edited to add: The specific functional form called K+Z+ZSS is novel and seems to give the best fits to the data.
Further edited to add: As is typical, C448 does not offer much discriminatory power between the two hypotheses in just prior posts. This might be because it contains fewer data points than the other two.
More data gives greater discriminatory power to the naive Bayes factor method. So instead of being just 3.9 dB down, the SS hypothesis would be further down. The dividing point is 5.0 dB down.
Edited to add: The specific functional form called K+Z+ZSS is novel and seems to give the best fits to the data.
Further edited to add: As is typical, C448 does not offer much discriminatory power between the two hypotheses in just prior posts. This might be because it contains fewer data points than the other two.
QUOTE (David B. Benson+Jan 12 2008, 06:47 PM)
More data gives greater discriminatory power to the naive Bayes factor method. So instead of being just 3.9 dB down, the SS hypothesis would be further down. The dividing point is 5.0 dB down.
OK, makes perfect sense. Agreement is a loose term. Here it means 'preserve ranking' so additional data which is in agreement will only sharpen the view.
I'll bet if I gave you highly accurate data multiplied by a square term (in t, x, or v) with a small coefficient, it would do well, too. Just my gut.
OK, makes perfect sense. Agreement is a loose term. Here it means 'preserve ranking' so additional data which is in agreement will only sharpen the view.
QUOTE
Edited to add: The specific functional form called K+Z+ZSS is novel and seems to give the best fits to the data.
I'll bet if I gave you highly accurate data multiplied by a square term (in t, x, or v) with a small coefficient, it would do well, too. Just my gut.
QUOTE (OneWhiteEye+Jan 12 2008, 12:01 PM)
I'll bet if I gave you highly accurate data multiplied by a square term (in t, x, or v) with a small coefficient, it would do well, too. Just my gut.
I'm not sure what you are suggesting, but please don't do that to me!
The form has three parameters to fit the data. If I had more data, I would like to add another, for an additional SS term to account for air resistance.
I'm not sure what you are suggesting, but please don't do that to me!The form has three parameters to fit the data. If I had more data, I would like to add another, for an additional SS term to account for air resistance.
QUOTE (David B. Benson+Jan 12 2008, 07:16 PM)
I'm not sure what you are suggesting, but please don't do that to me!No, I wouldn't. But I don't expect you to take my word for it; the process, I trust, will be brainless enough that anyone will be able to replicate it. I'll provide all details necessary to recreate the data from the source video.
It's sometimes easier to open a debate on the effect of error when mentioning a source that's automatically the kiss of death.
QUOTE (OneWhiteEye+Jan 12 2008, 12:31 PM)
It's sometimes easier to open a debate on the effect of error when mentioning a source that's automatically the kiss of death.
Systematic errors are difficult to deal with.
But probably the most troublesome aspect is the form of the stretch function. I have chosen an decaying exponential, just because it is an easy way to have a monotonically strictly decreasing positive function. The argument of the exponential is the momentum raised to a parametrized power. This makes it easy to take the needed derivative used to help compute the acceleration at each Runge-Kutta step.
Incidentally, the power, for the best hypothesis, is always close to one. I don't know what to make of this.
But the main point is that I, nor anyone else it seems, have any better idea of what to do. Given that this stretch function is purely empirical, it seems a poor idea to make overly much out of one resistive functional form being better than another as a matter of actual fact. Despite this caveat, the vertical avalanche does seem to do the best, and has some physical intuition behind it.
Systematic errors are difficult to deal with.
But probably the most troublesome aspect is the form of the stretch function. I have chosen an decaying exponential, just because it is an easy way to have a monotonically strictly decreasing positive function. The argument of the exponential is the momentum raised to a parametrized power. This makes it easy to take the needed derivative used to help compute the acceleration at each Runge-Kutta step.
Incidentally, the power, for the best hypothesis, is always close to one. I don't know what to make of this.
But the main point is that I, nor anyone else it seems, have any better idea of what to do. Given that this stretch function is purely empirical, it seems a poor idea to make overly much out of one resistive functional form being better than another as a matter of actual fact. Despite this caveat, the vertical avalanche does seem to do the best, and has some physical intuition behind it.
QUOTE (David B. Benson+Jan 12 2008, 07:16 PM)
I'm not sure what you are suggesting, but please don't do that to me!I'll give you a more thoughtful response.
You know the adage you can't believe everything you see on the Internet? It applies here, too. The data I provide is no better than a YouTube video, both literally and figuratively. After this post or the next, I'm going to order the DVD because it's time. I know the steps I'm taking to ensure the data is good.
I am just words on a screen. I may be nice, and I may be able to write more conversational chat-bots than some of the posters I've seen, but I may not have a clue what I'm doing or have some agenda - as far as you know. So I always advise caution.
That wasn't why I said that, I really wanted to discuss systematic error in greater detail and it was like dropping a cocktail glass when you want to demonstrate a floor cleaner. Still worth mentioning for the advisory aspect. I have a Google video right now where a few pixels of a few frames imply something extraordinary - if it's not fake. But I do have a copy of the same video but better pedigree, lying here amongst hundreds of CDs I can't be bothered looking through, and if it's in that, too, my need to find an explanation will increase dramatically. But even my better copy could be faked, you see?
QUOTE (David B. Benson+Jan 12 2008, 07:59 PM)
Systematic errors are difficult to deal with.
I believe I could quantify some systematic errors in C447-449. I don't know what the results would be until I did it, and it would take some effort which is why I haven't. I'm not sure it would make a difference in what you're doing. If I knew it would, I'd be more inclined.
A 'better' set of data is closer at hand than is the correction of the previous sets, which may not ever happen. I'm trying to judge the value of correction without doing it. Possibly you don't want any correction and I can understand a lot of valid reasons for that. Clearly, that would reduce my motivation.
The dark band is roughly a hook shape that slides across the three columns in the course of the measurement frames. If you were to jam the graph of a (i.e.) parabola into a few pixels, it might look like that section of the dark band. To me, that's like knowing the data is doped in the manner I suggested above, not by any nefarious agent but by circumstance - which can be quantified.
I believe I could quantify some systematic errors in C447-449. I don't know what the results would be until I did it, and it would take some effort which is why I haven't. I'm not sure it would make a difference in what you're doing. If I knew it would, I'd be more inclined.
A 'better' set of data is closer at hand than is the correction of the previous sets, which may not ever happen. I'm trying to judge the value of correction without doing it. Possibly you don't want any correction and I can understand a lot of valid reasons for that. Clearly, that would reduce my motivation.
The dark band is roughly a hook shape that slides across the three columns in the course of the measurement frames. If you were to jam the graph of a (i.e.) parabola into a few pixels, it might look like that section of the dark band. To me, that's like knowing the data is doped in the manner I suggested above, not by any nefarious agent but by circumstance - which can be quantified.
It may be worth emphasizing that the existing sets are manual, and therefore even I cannot replicate them! All I could do is add another trial on the same columns.
I'd have no qualms about directly merging multiple trials of individual columns, especially if many people were doing it, but I would disagree with the interpretation that it would necessarily converge on the correct solution. The discriminatory power would be focused on selection of a hypothesis most capable of fitting a few highly accurate transcriptions of a warped record.
Imagine these three column datasets as somewhat accurate pictures, from slightly different angles, of an image viewed through a pane of glass with a wrinkle in it. The image you want does not have the little distortion, which may be small in magnitude compared to the image as a whole, but not insignificant in the localized region it affects.
By contrast, forthcoming automated data not only won't have the wrinkle, it will be totally replicable, front to back. There is no hand-eye-person subjectivity, no associated systematic errors introduced as factors which then must be accounted for by more numbers in any stochastic method. There will be greater quantity of points, which will help with the introduction of video noise not present in the manual data.
There may be some subjectivity in terms of choice of fixing number of pixel clusters in a spatial region, use of manhattan or euclidean distance, or even using clustering at all in turning video data into numeric values representing feature motion in the image plane. Nevertheless, it's self documenting; change the parameters, change even the algorithms, run again. By producing graphic overlays of the calculated results onto the source video, the process ensures a result worse than manual placement is never accepted.
I'd have no qualms about directly merging multiple trials of individual columns, especially if many people were doing it, but I would disagree with the interpretation that it would necessarily converge on the correct solution. The discriminatory power would be focused on selection of a hypothesis most capable of fitting a few highly accurate transcriptions of a warped record.
Imagine these three column datasets as somewhat accurate pictures, from slightly different angles, of an image viewed through a pane of glass with a wrinkle in it. The image you want does not have the little distortion, which may be small in magnitude compared to the image as a whole, but not insignificant in the localized region it affects.
By contrast, forthcoming automated data not only won't have the wrinkle, it will be totally replicable, front to back. There is no hand-eye-person subjectivity, no associated systematic errors introduced as factors which then must be accounted for by more numbers in any stochastic method. There will be greater quantity of points, which will help with the introduction of video noise not present in the manual data.
There may be some subjectivity in terms of choice of fixing number of pixel clusters in a spatial region, use of manhattan or euclidean distance, or even using clustering at all in turning video data into numeric values representing feature motion in the image plane. Nevertheless, it's self documenting; change the parameters, change even the algorithms, run again. By producing graphic overlays of the calculated results onto the source video, the process ensures a result worse than manual placement is never accepted.
QUOTE (OneWhiteEye+Jan 12 2008, 02:28 PM)
There may be some subjectivity in terms of choice of fixing number of pixel clusters in a spatial region,
use of manhattan or euclidean distance, ...
By producing graphic overlays of the calculated results onto the source video, the process ensures a result worse than manual placement is never accepted.
The process is new. So long as it is documented.
Explain more, please.
You'll have to do the overlays. This looks to be a winner of an idea. Check your PM in about 5 minutes.
use of manhattan or euclidean distance, ...
By producing graphic overlays of the calculated results onto the source video, the process ensures a result worse than manual placement is never accepted.
The process is new. So long as it is documented.
Explain more, please.You'll have to do the overlays. This looks to be a winner of an idea. Check your PM in about 5 minutes.
QUOTE (David B. Benson+Jan 12 2008, 09:51 PM)
Explain more, please.The most sensible things to track are identifiable features, things on the antenna and such. But what are features to programs that look at:
FrameNumber, Column, Row, Red, Green, Blue
as input? Plenty of literature on the subject, too much. The essence of most of it is similarity across a number of dimensions, leading to a more rigorous definition of similarity. For spatial distance in a pixel plane, two natural choices are manhattan and euclidean metrics. I've used euclidean so far.
QUOTE
You'll have to do the overlays
That's the easy thing, really. It only takes a few lines of code to add the calculated results (translucently) back on to the current frame being analyzed. So you can look right at it, basically.
QUOTE (OneWhiteEye+Jan 12 2008, 03:23 PM)
The most sensible things to track are identifiable features, things on the antenna and such.
... So you can look right at it, basically.
Ah, yes. I encourage starting out by tracking a single dish antenna on the antenna tower, if that is possible.
Hmmm. The calculated results are in vertical meters of drop, so you'll need code to project that to the simulated camera. I guess that was obvious, now that I've written it...
... So you can look right at it, basically.
Ah, yes. I encourage starting out by tracking a single dish antenna on the antenna tower, if that is possible.
Hmmm. The calculated results are in vertical meters of drop, so you'll need code to project that to the simulated camera. I guess that was obvious, now that I've written it...
QUOTE (David B. Benson+Jan 12 2008, 10:46 PM)
Ah, yes. I encourage starting out by tracking a single dish antenna on the antenna tower, if that is possible.
My thoughts exactly. There are, even with such a simple feature, a variety of possible data products. I'll outline those in a bit.
Actually I was talking about the intermediate calculation of pixel locations in the source video image, but round-tripping through a 3D render with the data in meters is possible, too. A little farther away, I'm afraid, and more of visualization than a verification. The overlay I refer to ensures the automated process produces results good enough to satisfy the eye, at least.
One dish is a roughly elliptical blob a few pixels across. It suffices to place a reasonably sized dot on top of it to show where the routines have determined the location of the feature to be. There are many ways to arrive at that number (vector), and all will fall within the visually detectable boundary of the ellipse if functioning correctly. The overlay is foremost a means of ensuring proper functionality - at least staying within the bounding region. Secondarily, it could allow determination of whether jitter in the data corresonds to image jitter or artifact in the detection algorithms, something that might be eliminated by an adjustment of parameters.
If it can do better than the eye, it ought to look really good to the eye.
My thoughts exactly. There are, even with such a simple feature, a variety of possible data products. I'll outline those in a bit.
QUOTE
Hmmm. The calculated results are in vertical meters of drop, so you'll need code to project that to the simulated camera. I guess that was obvious, now that I've written it...
Actually I was talking about the intermediate calculation of pixel locations in the source video image, but round-tripping through a 3D render with the data in meters is possible, too. A little farther away, I'm afraid, and more of visualization than a verification. The overlay I refer to ensures the automated process produces results good enough to satisfy the eye, at least.
One dish is a roughly elliptical blob a few pixels across. It suffices to place a reasonably sized dot on top of it to show where the routines have determined the location of the feature to be. There are many ways to arrive at that number (vector), and all will fall within the visually detectable boundary of the ellipse if functioning correctly. The overlay is foremost a means of ensuring proper functionality - at least staying within the bounding region. Secondarily, it could allow determination of whether jitter in the data corresonds to image jitter or artifact in the detection algorithms, something that might be eliminated by an adjustment of parameters.
If it can do better than the eye, it ought to look really good to the eye.
QUOTE (David B. Benson+Jan 12 2008, 10:46 PM)
Hmmm. The calculated results are in vertical meters of drop, so you'll need code to project that to the simulated camera. I guess that was obvious, now that I've written it...
Yes, the values you calculate from a hypothesis could be round-tripped back onto the 2D-image, too. Just need the inverse of your perspective correction. I think it would be useful and cool.
Edit to add: it might be more than the inverse of correction. The crush front, if that's your spatial coordinate, could be mapped onto (any) video .
Yes, the values you calculate from a hypothesis could be round-tripped back onto the 2D-image, too. Just need the inverse of your perspective correction. I think it would be useful and cool.
Edit to add: it might be more than the inverse of correction. The crush front, if that's your spatial coordinate, could be mapped onto (any) video .
QUOTE (OneWhiteEye+Jan 12 2008, 04:22 PM)
Yes, the values you calculate from a hypothesis could be round-tripped back onto the 2D-image, too. Just need the inverse of your perspective correction. I think it would be useful and cool.
Edit to add: it might be more than the inverse of correction. The crush front, if that's your spatial coordinate, could be mapped onto (any) video .
Yes.
There are two distinct calculated values, the drop (of the top) and the location of the crushing front. These differ due to the stretch function. So being able to draw the calculated location of the crushing front on suitable videos ought to help to decide on a more realistic stretch function. Eventually.
Edit to add: it might be more than the inverse of correction. The crush front, if that's your spatial coordinate, could be mapped onto (any) video .
Yes.
There are two distinct calculated values, the drop (of the top) and the location of the crushing front. These differ due to the stretch function. So being able to draw the calculated location of the crushing front on suitable videos ought to help to decide on a more realistic stretch function. Eventually.
This is an extreme blowup of the top-center dish on the antenna:
http://i10.tinypic.com/71qfiis.png
The dish is approximately 4x7 pixels and ellipsoidal, at least it is in this particular frame. It's very easy to follow visually frame by frame until it goes out of sight. What is the x,y numeric position of this feature in each frame?
There are several approaches with automation, I'm hoping the pros and cons of each will be self-evident.
1) coordinate of the leftmost/topmost pixel of the ellipse
2) coordinate of the rightmost/bottommost pixel of the ellipse
3) other permutation of each 1 & 2 above
4) center of the bounding box defined by the combination of 1 & 2 above
5) unweighted geometric center of the outermost boundary of the ellipse
6) geometric center, weighted by luminous intensity, of the outermost boundary
7) unweighted geometric center of pixels interior to the ellipse which fall within a (variable) intensity range
8) as in 7, but weighted by intensity
9) unweighted geometric center of all pixels which are members of the ellipse, by variable inclusion criteria
10) weighted variant of 9
11) position of unique pixel within ellipse, say brightest
12) any weighted combination of the above
This skims over the difficulty in obtaining any of the coordinates described above from a little blob that changes shape (luminosity) quite a bit over the frames of motion. If you look at the dish in a 3D graph where intensity = 'elevation' you can see the question is much like the question "Where is the mountain, precisely?"
http://i5.tinypic.com/81sccb9.png
Except this is how it looks in just one frame. When you introduce time, the question becomes "Where is the wave?" and simply following the peak may not do as well as placing points by hand. This is what the dish looks like in the same format as it's about to disappear:
http://i18.tinypic.com/837c7ed.png
One does have to select a method for producing the desired scalar, they aren't all created equal. A few give only pixel resolution, some are jumpy. Like I said, all of them should place a point somewhere on or in that ellipse. Some will do a better job of staying fixed on the same relative point within the feature, subject to the limits of information contained in the video. I think the best results will come from a composite of cluster boundaries weighted by intensity. I'm guessing sub-pixel noise and at least 1/10th pixel average resolution.
http://i10.tinypic.com/71qfiis.png
The dish is approximately 4x7 pixels and ellipsoidal, at least it is in this particular frame. It's very easy to follow visually frame by frame until it goes out of sight. What is the x,y numeric position of this feature in each frame?
There are several approaches with automation, I'm hoping the pros and cons of each will be self-evident.
1) coordinate of the leftmost/topmost pixel of the ellipse
2) coordinate of the rightmost/bottommost pixel of the ellipse
3) other permutation of each 1 & 2 above
4) center of the bounding box defined by the combination of 1 & 2 above
5) unweighted geometric center of the outermost boundary of the ellipse
6) geometric center, weighted by luminous intensity, of the outermost boundary
7) unweighted geometric center of pixels interior to the ellipse which fall within a (variable) intensity range
8) as in 7, but weighted by intensity
9) unweighted geometric center of all pixels which are members of the ellipse, by variable inclusion criteria
10) weighted variant of 9
11) position of unique pixel within ellipse, say brightest
12) any weighted combination of the above
This skims over the difficulty in obtaining any of the coordinates described above from a little blob that changes shape (luminosity) quite a bit over the frames of motion. If you look at the dish in a 3D graph where intensity = 'elevation' you can see the question is much like the question "Where is the mountain, precisely?"
http://i5.tinypic.com/81sccb9.png
Except this is how it looks in just one frame. When you introduce time, the question becomes "Where is the wave?" and simply following the peak may not do as well as placing points by hand. This is what the dish looks like in the same format as it's about to disappear:
http://i18.tinypic.com/837c7ed.png
One does have to select a method for producing the desired scalar, they aren't all created equal. A few give only pixel resolution, some are jumpy. Like I said, all of them should place a point somewhere on or in that ellipse. Some will do a better job of staying fixed on the same relative point within the feature, subject to the limits of information contained in the video. I think the best results will come from a composite of cluster boundaries weighted by intensity. I'm guessing sub-pixel noise and at least 1/10th pixel average resolution.
The ultimate proof, that you can see, goes back to the orders for the military that were changed three months and ten days before 911.
http://www.dtic.mil/doctrine/jel/cjcsd/cjcsi/3610_01a.pdf
It has yet to be changed, or superceded, according to the Gov's own web site. It could all happen again today.
With Bush reading Goat Stories to children, Cheney running to hide in his bunker and Rumsfeld in a meeting with instructions to not be disturbed, America was defenseless.
They created the situation, and the rules, where it could not be stopped. The fighter jets sat on the runway waiting for permission that came too late.
But, you won't see that on TV....
http://www.dtic.mil/doctrine/jel/cjcsd/cjcsi/3610_01a.pdf
QUOTE
4. Policy.
a. Aircraft Piracy (Hijacking) of Civil and Military Aircraft. Pursuant
to references a and b, the Administrator, Federal Aviation Administration
(FAA), has exclusive responsibility to direct law enforcement activity
related to actual or attempted aircraft piracy (hijacking) in the “special
aircraft jurisdiction” of the United States. When requested by the
Administrator, Department of Defense will provide assistance to these
law enforcement efforts. Pursuant to reference c, the NMCC is the focal
point within Department of Defense for providing assistance. In the
event of a hijacking, the NMCC will be notified by the most expeditious
means by the FAA. The NMCC will, with the exception of immediate
responses as authorized by reference d, forward requests for DOD
assistance to the Secretary of Defense for approval. DOD assistance to
the FAA will be provided in accordance with reference d. Additional
guidance is provided in Enclosure A.
a. Aircraft Piracy (Hijacking) of Civil and Military Aircraft. Pursuant
to references a and b, the Administrator, Federal Aviation Administration
(FAA), has exclusive responsibility to direct law enforcement activity
related to actual or attempted aircraft piracy (hijacking) in the “special
aircraft jurisdiction” of the United States. When requested by the
Administrator, Department of Defense will provide assistance to these
law enforcement efforts. Pursuant to reference c, the NMCC is the focal
point within Department of Defense for providing assistance. In the
event of a hijacking, the NMCC will be notified by the most expeditious
means by the FAA. The NMCC will, with the exception of immediate
responses as authorized by reference d, forward requests for DOD
assistance to the Secretary of Defense for approval. DOD assistance to
the FAA will be provided in accordance with reference d. Additional
guidance is provided in Enclosure A.
It has yet to be changed, or superceded, according to the Gov's own web site. It could all happen again today.
With Bush reading Goat Stories to children, Cheney running to hide in his bunker and Rumsfeld in a meeting with instructions to not be disturbed, America was defenseless.
They created the situation, and the rules, where it could not be stopped. The fighter jets sat on the runway waiting for permission that came too late.
But, you won't see that on TV....
QUOTE (ArchAngel+Jan 12 2008, 10:07 PM)
They created the situation, and the rules, where it could not be stopped. The fighter jets sat on the runway waiting for permission that came too late.
Total Horse Puckey.
No one asked (or needed) Rumsfeld's permission to launch the Otis jets and they didn't wait to take off, they scrambled first and hot footed it to NY even though they had no idea which plane they were looking for and would have had no authority to shoot a US civilian airliner down even if they had got there in time, which of course they didn't.
So, nothing could have stopped the first two jets from hitting their targets.
Only luck or 20/20 hind-sight could have stopped the jet that hit the Pentagon.
Arthur
Total Horse Puckey.
No one asked (or needed) Rumsfeld's permission to launch the Otis jets and they didn't wait to take off, they scrambled first and hot footed it to NY even though they had no idea which plane they were looking for and would have had no authority to shoot a US civilian airliner down even if they had got there in time, which of course they didn't.
So, nothing could have stopped the first two jets from hitting their targets.
Only luck or 20/20 hind-sight could have stopped the jet that hit the Pentagon.
Arthur
QUOTE (adoucette+Jan 13 2008, 05:35 AM)
Total Horse Puckey.
No one asked (or needed) Rumsfeld's permission to launch the Otis jets and they didn't wait to take off, they scrambled first and hot footed it to NY even though they had no idea which plane they were looking for and would have had no authority to shoot a US civilian airliner down even if they had got there in time, which of course they didn't.
So, nothing could have stopped the first two jets from hitting their targets.
Only luck or 20/20 hind-sight could have stopped the jet that hit the Pentagon.
Arthur
really?
what about all those terrorist watch lists? they ought to be good for stopping terrorists.
or enabling them.
No one asked (or needed) Rumsfeld's permission to launch the Otis jets and they didn't wait to take off, they scrambled first and hot footed it to NY even though they had no idea which plane they were looking for and would have had no authority to shoot a US civilian airliner down even if they had got there in time, which of course they didn't.
So, nothing could have stopped the first two jets from hitting their targets.
Only luck or 20/20 hind-sight could have stopped the jet that hit the Pentagon.
Arthur
really?
what about all those terrorist watch lists? they ought to be good for stopping terrorists.
or enabling them.
Only luck or 20/20 hind-sight could have stopped the jet that hit the Pentagon.
There are always armed jets just outside of DC, waiting, just in case.
They had enough time to scamble a man on a bike with a shoulder launched missile.
Total Horse Puckey.
All of it is true. You are the one sitting in a pile of poo...
There are always armed jets just outside of DC, waiting, just in case.
They had enough time to scamble a man on a bike with a shoulder launched missile.
Total Horse Puckey.
All of it is true. You are the one sitting in a pile of poo...
QUOTE (ArchAngel+Jan 13 2008, 07:06 AM)
There are always armed jets just outside of DC, waiting, just in case.
Really?
Source from 2001 showing this assertion is TRUE please.
Arthur
Really?
Source from 2001 showing this assertion is TRUE please.
Arthur
Here are two hypotheses which cannot be distinguished by the naive Bayes factor method (C447 data):
CODE
Sef-K+Z+ZSS-F-exp-pow-stretch dB= 0.0 sd= 0.101
Sef-K+Z+avalanche-F-exp-pow-stretch dB= 1.1 sd= 0.144
Sef-K+Z+avalanche-F-exp-pow-stretch dB= 1.1 sd= 0.144
Both are vertical avalanche style, the difference being the mass used in the avalanche part of the resistive force function; the best used just M_B, the mass of the crushed materials; the other uses the total moving mass (M_A+M_B).
Here, by naive Bayses factor method (C447 data) the poorer hypothesis is right on the edge of being substantially disconfirmed. But other statistics (not shown) show that the curve shape is poor, which tends to help disconfirm it.
CODE
Sef-K+Z+ZSS-F-exp-pow-stretch dB= 0.0 sd= 0.101
Sef-K+Z+ZSS-F-const-stretch dB= 4.9 sd= 0.255
Sef-K+Z+ZSS-F-const-stretch dB= 4.9 sd= 0.255
Edited to add: WIth quite different parameters, the poorer hypothesis is 3.6 dB down on the C449 data. This parametric instability is certainly a mark against the hypothesis.
QUOTE (David B. Benson+Jan 13 2008, 05:58 PM)
Both are vertical avalanche style, the difference being the mass used in the avalanche part of the resistive force function; the best used just M_B, the mass of the crushed materials; the other uses the total moving mass (M_A+M_B).
Reducing the effective mass makes a better fit? Interesting. I'd be curious as to fitness as a function of mass for that hypothesis. Any other values of mass you've run on these?
Reducing the effective mass makes a better fit? Interesting. I'd be curious as to fitness as a function of mass for that hypothesis. Any other values of mass you've run on these?
QUOTE (David B. Benson+Jan 9 2008, 10:22 PM)
I have never seen this before. Thank you for your effort.
Yes, this is air being forced down the voids in the core (or around the ends of paritially detached floors). This causes overpressure sufficient to break the windows. (Do recall that not all floors were the same height. In particular floors 74, 75, 76 and 78 were 14 feet high.)
Can you tell how thick it is?
Yes. For WTC 2 about 2--3 floors all collapsed at about the same instant. Recall how massive the upper portion was.
DBB,
I never replied this one. This originates about 10 pixels below the impact zone, or about 4-6 meter. The object is about 3-7 meter and the initial horizontal speed is about 13.6 m/s, I didn't calculate the real error margin but I guess > 10 m/s
The tower wind cannot account for that. Released strain energy or a kind of scattering effect or wedge effect sounds more plausible. Tiddlywink effect....?
Yes, this is air being forced down the voids in the core (or around the ends of paritially detached floors). This causes overpressure sufficient to break the windows. (Do recall that not all floors were the same height. In particular floors 74, 75, 76 and 78 were 14 feet high.)
Can you tell how thick it is?
Yes. For WTC 2 about 2--3 floors all collapsed at about the same instant. Recall how massive the upper portion was.
DBB,
I never replied this one. This originates about 10 pixels below the impact zone, or about 4-6 meter. The object is about 3-7 meter and the initial horizontal speed is about 13.6 m/s, I didn't calculate the real error margin but I guess > 10 m/s
The tower wind cannot account for that. Released strain energy or a kind of scattering effect or wedge effect sounds more plausible. Tiddlywink effect....?
QUOTE (OneWhiteEye+Jan 12 2008, 06:04 PM)
Haha. You should stick the knife in and twist it. Get the satisfaction of an OCTer, just once.
By the way, your recent post is not lost on me. I think it deserves a very close look but, for me, that could make it weeks off. So don't be surprised if I pull it back out of the closet at an unexpected time in the future. Just like the 'real' F4 data.
There was someone who mention the triplet structure of the perimeter columns, that is indeed a possible debunk, but i'm not yet convinced, because that still doesn't explain why that set of dust ejections is also a row at the right hand side. If I can find the time and the videos I want to synchronize videos at all 4 sides. I also would like to see how fast that collapse really was. That's also very useful for David's fitting function I guess.
By the way, your recent post is not lost on me. I think it deserves a very close look but, for me, that could make it weeks off. So don't be surprised if I pull it back out of the closet at an unexpected time in the future. Just like the 'real' F4 data.
There was someone who mention the triplet structure of the perimeter columns, that is indeed a possible debunk, but i'm not yet convinced, because that still doesn't explain why that set of dust ejections is also a row at the right hand side. If I can find the time and the videos I want to synchronize videos at all 4 sides. I also would like to see how fast that collapse really was. That's also very useful for David's fitting function I guess.
QUOTE (OneWhiteEye+Jan 13 2008, 05:52 PM)
Reducing the effective mass makes a better fit? Interesting. I'd be curious as to fitness as a function of mass for that hypothesis. Any other values of mass you've run on these?
Only slightly better. The reasoning is that the crushed materials are continually re-crushed resulting in the term kM_Bv^2. I have also tried kv^2 (for crushing only at the crushing front) which is even worse.
Here is the vertical avalanche resisting force applied to C447 with a constant stretch of 0.18:
Substantially disconfirmed, but ok with a larger stretch, but not parametrically stable.
Only slightly better. The reasoning is that the crushed materials are continually re-crushed resulting in the term kM_Bv^2. I have also tried kv^2 (for crushing only at the crushing front) which is even worse.
Here is the vertical avalanche resisting force applied to C447 with a constant stretch of 0.18:
CODE
Sef-K+Z+ZSS-F-exp-pow-stretch dB= 0.0 sd= 0.101
Sef-K+Z+ZSS-F-const-stretch dB= 5.1 sd= 0.264
Sef-K+Z+ZSS-F-const-stretch dB= 5.1 sd= 0.264
Substantially disconfirmed, but ok with a larger stretch, but not parametrically stable.
QUOTE (einsteen+Jan 14 2008, 06:57 AM)
Tiddlywink effect....?
I opine it was a piece of aluminum cladding, somehow tiddlywinked out...
I opine it was a piece of aluminum cladding, somehow tiddlywinked out...
QUOTE (David B. Benson+Jan 14 2008, 07:20 PM)
Substantially disconfirmed, but ok with a larger stretch, but not parametrically stable.
Parametric stability, intuitively, seems a good indicator of fitness of model. Is there any formal argument in support of the notion?
Somewhat related:
I've recently played around with conventionally fitting a known generated dataset injected with known error (so far just time and value offsets). The sensitivity of fit is significant, particularly wrt time as there is an asymmetry about t0, where an analytic representation of a curve diverges wildy from the graph it attempts to fit.
Trimming the 'constant' baseline at the beginning of a dataset is obviously not the same as t0 determination, which is crucial, and I appreciate that your method attempts to minimize the variance in the high grade hypotheses wrt time offset.
I screwed around with real numbers and a curve fitting app rather than try to develop any sort of formal analysis to generalize the work. What I found in checking polynomial fits might be old news to some and not even surprising to me, but two things were most interesting:
1) the asymmetry of the graph against the time coordinate made the fit far less sensitive to time offset error in one direction versus the other
2) higher order polynomials often provided the best fit for even slight time errors, despite the fact that test data was generated with certain coefficients identically zero
Intuitively, for the limited case I examined, I'm inclined to cast similar suspicions on any terms with small coefficients. However, my broader common sense tells me that this is an invalid operating principle, as the reverse is probably also true: simpler forms can fit better than the true, underlying complex form, if the influences of the terms are but subtle.
Edit: I did not state #2 correctly. What I meant was non-zero coefficients provided a better fit than did the real form with no term at that order. I do want to check the fit of higher order unconstrained solutions versus lower order constrained.
Parametric stability, intuitively, seems a good indicator of fitness of model. Is there any formal argument in support of the notion?
Somewhat related:
I've recently played around with conventionally fitting a known generated dataset injected with known error (so far just time and value offsets). The sensitivity of fit is significant, particularly wrt time as there is an asymmetry about t0, where an analytic representation of a curve diverges wildy from the graph it attempts to fit.
Trimming the 'constant' baseline at the beginning of a dataset is obviously not the same as t0 determination, which is crucial, and I appreciate that your method attempts to minimize the variance in the high grade hypotheses wrt time offset.
I screwed around with real numbers and a curve fitting app rather than try to develop any sort of formal analysis to generalize the work. What I found in checking polynomial fits might be old news to some and not even surprising to me, but two things were most interesting:
1) the asymmetry of the graph against the time coordinate made the fit far less sensitive to time offset error in one direction versus the other
2) higher order polynomials often provided the best fit for even slight time errors, despite the fact that test data was generated with certain coefficients identically zero
Intuitively, for the limited case I examined, I'm inclined to cast similar suspicions on any terms with small coefficients. However, my broader common sense tells me that this is an invalid operating principle, as the reverse is probably also true: simpler forms can fit better than the true, underlying complex form, if the influences of the terms are but subtle.
Edit: I did not state #2 correctly. What I meant was non-zero coefficients provided a better fit than did the real form with no term at that order. I do want to check the fit of higher order unconstrained solutions versus lower order constrained.
QUOTE (einsteen+Jan 14 2008, 01:57 PM)
Tiddlywink effect....?
Could not have termed it better, good turn of phrase. It doesn't have to be a wedge effect as you graphically depicted, it can be pure strain energy.
In looking at the north tower video we've kicked around, I swear I see obvious deformation of the 'undamaged' lower portion in response to the increasing eccentricity of the load from above. Unable to quantify yet - I've tried, and to this point, I don't even have a way of making it visually obvious, though applying translucent regular patterns to produce artificial moire seems promising.
Even large components could be launched by the tiddlywink effect. I just don't think a significant amount of material can go that route. Tiddlywinks: small mass, large strain, externally imposed.
Could not have termed it better, good turn of phrase. It doesn't have to be a wedge effect as you graphically depicted, it can be pure strain energy.
In looking at the north tower video we've kicked around, I swear I see obvious deformation of the 'undamaged' lower portion in response to the increasing eccentricity of the load from above. Unable to quantify yet - I've tried, and to this point, I don't even have a way of making it visually obvious, though applying translucent regular patterns to produce artificial moire seems promising.
Even large components could be launched by the tiddlywink effect. I just don't think a significant amount of material can go that route. Tiddlywinks: small mass, large strain, externally imposed.
My grandma had the "vlooienspel" at her home, flea game if you translate it literally but glad I got a dictionary, I still need it every day!
In that controlled demolition video from national geographic where they placed charges on two levels also such a piece pops out, but then caused by explosives. During the rest of the drop the ejections don't look very violently. It mainly stays within the footprint.
In that controlled demolition video from national geographic where they placed charges on two levels also such a piece pops out, but then caused by explosives. During the rest of the drop the ejections don't look very violently. It mainly stays within the footprint.
QUOTE (OneWhiteEye+Jan 14 2008, 01:10 PM)
Parametric stability, intuitively, seems a good indicator of fitness of model. Is there any formal argument in support of the notion?
I've looked for 'parametric sensitivity'. Not the right notion.
What I call parameter estimation is called 'parameter calibration' by some. That didn't help.
I'll try some more and then likely have to devise a suitable argument.
However, the intuition is clear: adjacent pixel columns describe essentially the same phenomenon. If noticeably different parameter values are required for these by some model, then no matter how appealing the physics underlying the model, it is simply to sensitive to provide stable predictions.
I've looked for 'parametric sensitivity'. Not the right notion.
What I call parameter estimation is called 'parameter calibration' by some. That didn't help.
I'll try some more and then likely have to devise a suitable argument.
However, the intuition is clear: adjacent pixel columns describe essentially the same phenomenon. If noticeably different parameter values are required for these by some model, then no matter how appealing the physics underlying the model, it is simply to sensitive to provide stable predictions.
QUOTE (einsteen+Jan 14 2008, 09:32 PM)
My grandma had the "vlooienspel" at her home, flea game if you translate it literally but glad I got a dictionary, I still need it every day!.
Thank you for making the effort. Vlooienspel would not have made the same impression.
A certain JPL scientist we know (not Jack Parsons) has made some pronouncements on the abilities of explosives to propel structural elements. I haven't looked at it, but I wonder how well it jibes with a structure that is known to have members propelled by explosive.
Is the piece you refer to ejected too early to be due to anything other than the charge(s)?
Thank you for making the effort. Vlooienspel would not have made the same impression.
QUOTE
In that controlled demolition video from national geographic where they placed charges on two levels also such a piece pops out, but then caused by explosives. During the rest of the drop the ejections don't look very violently. It mainly stays within the footprint.
A certain JPL scientist we know (not Jack Parsons) has made some pronouncements on the abilities of explosives to propel structural elements. I haven't looked at it, but I wonder how well it jibes with a structure that is known to have members propelled by explosive.
Is the piece you refer to ejected too early to be due to anything other than the charge(s)?
QUOTE (David B. Benson+Jan 14 2008, 10:42 PM)
I'll try some more and then likely have to devise a suitable argument.
Excellent. And I'll try to follow it!
Agreed. Yet I cannot shake the idea that it could be sensitivity to the error in the data as opposed to sensitivity to the true signal. If it were true that the greatest share of variance between C447-449 sets were due to something other than artifact I've described then I've little doubt the principle you describe would apply.
Imagine three datasets (I do get tiresome with this, don't I?) that are independent renderings of the same underlying signal. Each are equally accurate and subject to small random variations we'll call noise in the normal sense (random magnitude, no particular bias). With sufficient points in each set to allow the method to properly extract signal from noise, performing your analysis, I'd expect, would show no significant rank changes from one set to the next. Combining the sets would also not change rank, but provide better discriminatory power.
Now, take one set and multiply the points by (1 + f(t)) and another set by (1 - f(t)) where f(t) = ct^2 and c is chosen such that f(t), over the interval of the sets, has a maximum magnitude roughly equal or slightly greater than the peak noise value in each set. What happens to the analysis results?
Continued....
Excellent. And I'll try to follow it!
QUOTE
However, the intuition is clear: adjacent pixel columns describe essentially the same phenomenon. If noticeably different parameter values are required for these by some model, then no matter how appealing the physics underlying the model, it is simply to sensitive to provide stable predictions.
Agreed. Yet I cannot shake the idea that it could be sensitivity to the error in the data as opposed to sensitivity to the true signal. If it were true that the greatest share of variance between C447-449 sets were due to something other than artifact I've described then I've little doubt the principle you describe would apply.
Imagine three datasets (I do get tiresome with this, don't I?) that are independent renderings of the same underlying signal. Each are equally accurate and subject to small random variations we'll call noise in the normal sense (random magnitude, no particular bias). With sufficient points in each set to allow the method to properly extract signal from noise, performing your analysis, I'd expect, would show no significant rank changes from one set to the next. Combining the sets would also not change rank, but provide better discriminatory power.
Now, take one set and multiply the points by (1 + f(t)) and another set by (1 - f(t)) where f(t) = ct^2 and c is chosen such that f(t), over the interval of the sets, has a maximum magnitude roughly equal or slightly greater than the peak noise value in each set. What happens to the analysis results?
Continued....
Perhaps a better example of distortion to the signal would be f(t) as above on one set and a g(t) which is of the form k(Tf - t)^2 where Tf is the final time in the dataset. This puts the largest magnitude of systematic distortion at the beginning of the affected set, where it is at the end of the other set.
Hypotheses which are best able to accommodate the small, but unrealistic, distortions at ends will perform better overall. Hypotheses which most closely model the real signal may not fare as well, because the test is actually unfair.
If a hypothesis which best fits the true signal can adapt to the distortions, it may have to do so by significant variation of parameters. The only real test is when one is comparing against a dataset which accurately portrays the signal, even if noisy.
If three thousand datasets such as above were prepared and intentionally distorted with functions f and g, but with random coefficients between 0-c and 0-k, respectively, I think the Bayesian method would have no problem distinguishing the signal and selecting the best hypothesis... but with only three?
Hypotheses which are best able to accommodate the small, but unrealistic, distortions at ends will perform better overall. Hypotheses which most closely model the real signal may not fare as well, because the test is actually unfair.
If a hypothesis which best fits the true signal can adapt to the distortions, it may have to do so by significant variation of parameters. The only real test is when one is comparing against a dataset which accurately portrays the signal, even if noisy.
If three thousand datasets such as above were prepared and intentionally distorted with functions f and g, but with random coefficients between 0-c and 0-k, respectively, I think the Bayesian method would have no problem distinguishing the signal and selecting the best hypothesis... but with only three?
QUOTE (OneWhiteEye+Jan 14 2008, 05:28 PM)
... but with only three?
There is a huge literature on attempting to understand various distortions in the data in attempts to find simple models which explain 'most of' the variance.
I find your example forms of systematic error to be rather contrived. While there may indeed be various errors, timing is not likely to be a problem. I previously suggested some form of mass-sprig system acting between the hat truss and the antenna tower. In effect, a good model is able to shrug off the effects of various high frequency vibrations. Poor ones are effected by such.
There is a huge literature on attempting to understand various distortions in the data in attempts to find simple models which explain 'most of' the variance.
I find your example forms of systematic error to be rather contrived. While there may indeed be various errors, timing is not likely to be a problem. I previously suggested some form of mass-sprig system acting between the hat truss and the antenna tower. In effect, a good model is able to shrug off the effects of various high frequency vibrations. Poor ones are effected by such.
QUOTE (David B. Benson+Jan 15 2008, 01:15 AM)
I find your example forms of systematic error to be rather contrived.
Very contrived, by design, but probably remarkably close to the real C447-449 situation, also by design. The purpose being to try to describe an effect when all factors are known a priori, so the effects can be appraised rather than assumed.
The statements of timing were more general in nature. It was a little exploratory venture that taught me something.
The statements of timing were more general in nature. It was a little exploratory venture that taught me something.
I previously suggested some form of mass-sprig system acting between the hat truss and the antenna tower. In effect, a good model is able to shrug off the effects of various high frequency vibrations. Poor ones are effected by such.
That, too, makes sense. How about low frequency 'vibrations' where 1/2 the period is the time interval of the whole data set?
http://i16.tinypic.com/8e6ecgn.jpg
I think the winter garden is about 64 meter * 122pix/55pix=142 meter from the perimeter of wtc1
wfc3 is sqrt(137^2 + 13^2)*64 meter/55= 160 meter from the perimeter of wtc1
and that was even not at ground zero but 1/3-1/2 way the building
160 > 142 >> 92. the wfc3 beam is even more than a factor 1.7 too far, Powermetal!
O.W.E., I'll have a look later
where not only is the poorer hypothesis substantively disconfirmed, but also the stretch parameter value is dramatically different than for the C447 data.
Very contrived, by design, but probably remarkably close to the real C447-449 situation, also by design. The purpose being to try to describe an effect when all factors are known a priori, so the effects can be appraised rather than assumed.
QUOTE
While there may indeed be various errors, timing is not likely to be a problem.
The statements of timing were more general in nature. It was a little exploratory venture that taught me something.
QUOTE (->
| QUOTE |
| While there may indeed be various errors, timing is not likely to be a problem. |
The statements of timing were more general in nature. It was a little exploratory venture that taught me something.
I previously suggested some form of mass-sprig system acting between the hat truss and the antenna tower. In effect, a good model is able to shrug off the effects of various high frequency vibrations. Poor ones are effected by such.
That, too, makes sense. How about low frequency 'vibrations' where 1/2 the period is the time interval of the whole data set?
QUOTE (David B. Benson+Jan 15 2008, 01:15 AM)
some form of mass-sprig[sic] system
[sic]
[sic]
QUOTE (David B. Benson+Jan 14 2008, 10:42 PM)
However, the intuition is clear: adjacent pixel columns describe essentially the same phenomenon. If noticeably different parameter values are required for these by some model, then no matter how appealing the physics underlying the model, it is simply to sensitive to provide stable predictions.
The important part for my argument is this:
"adjacent pixel columns describe essentially the same phenomenon"
A reasonable statement. The degree to which this assumption is true with these particular smear-o-grams is the question. I know you get that. When the effect arising from deviation from the assumption is equal to or greater than any other source of error, what is to be expected?
I think the curves were digitized with an accuracy of about +/- 1 pixel. BUT - the measurements are of the smeared curves, making them accurate but indirect measurement of motion. Measurements of smear curves are not automatically synonymous with measurements of feature motion. They can be, but to limited accuracy.
Smear-o-grams follow a feature only so well, and must be chosen as such. Here, the feature moves out of the column(s). I've quantified an approximate magnitude of 2 pixels error but no quantification of how this error is distributed to the data. Doesn't seem like much, but...
It is NOT a gaussian distribution. It is most likely a monotonic or simple convex function. (edit: sorry, HAS to be monotonic or have at most one change of sign of first derivative) It is continuous and slow-varying, but not necessarily evenly distributed. If it had to be thought of as having a frequency (not true), it would have to be described as having a period of twice the dataset time duration. The maximum magnitude could occur early, when true deflection is less even than the error!
It does not affect the columns in the same way at the same time. It is as systematic as my contrived example, and it is roughly the magnitude of the measurement error band.
(Major edits done)
The important part for my argument is this:
"adjacent pixel columns describe essentially the same phenomenon"
A reasonable statement. The degree to which this assumption is true with these particular smear-o-grams is the question. I know you get that. When the effect arising from deviation from the assumption is equal to or greater than any other source of error, what is to be expected?
I think the curves were digitized with an accuracy of about +/- 1 pixel. BUT - the measurements are of the smeared curves, making them accurate but indirect measurement of motion. Measurements of smear curves are not automatically synonymous with measurements of feature motion. They can be, but to limited accuracy.
Smear-o-grams follow a feature only so well, and must be chosen as such. Here, the feature moves out of the column(s). I've quantified an approximate magnitude of 2 pixels error but no quantification of how this error is distributed to the data. Doesn't seem like much, but...
It is NOT a gaussian distribution. It is most likely a monotonic or simple convex function. (edit: sorry, HAS to be monotonic or have at most one change of sign of first derivative) It is continuous and slow-varying, but not necessarily evenly distributed. If it had to be thought of as having a frequency (not true), it would have to be described as having a period of twice the dataset time duration. The maximum magnitude could occur early, when true deflection is less even than the error!
It does not affect the columns in the same way at the same time. It is as systematic as my contrived example, and it is roughly the magnitude of the measurement error band.
(Major edits done)
QUOTE (OneWhiteEye+Jan 15 2008, 02:17 AM)
the measurements are of the smeared curves, making them accurate but indirect measurement of motion. Measurements of smear curves are not automatically synonymous with measurements of feature motion. They can be, but to limited accuracy.
"They can be, but to limited accuracy."
No, ignore that. My point will never come across unless I express myself precisely.
Smears do not measure features, they are a picture composed of slices. The smears, in turn, offer features of their own for measurement. It's easier to follow an edge in a single smear than a 2D blob from frame to frame, but it's not the same thing. The degree to which it gives the same numeric results determines the degree of equivalency.
The feature motion has to be collinear with the column in order to preserve feature motion in the smear curve. It is not precisely so in C447-449 but may be good enough. The measured curve is in reality a wandering point on the dark band, which wanders further than my hand.
Think of smear-o-grams as being like the Alta Vista translator. The last sentence of the paragraph above, when translated to Portuguese and back gives:
The measured curve is in the reality a point wandering in the dark band, that wanders more more of the one than my hand.
If that's good enough to get sufficiently similar meaning as the original statement, then so be it. But no amount of accuracy in further (real) translations or transcriptions will remove the existing distortions.
"They can be, but to limited accuracy."
No, ignore that. My point will never come across unless I express myself precisely.
Smears do not measure features, they are a picture composed of slices. The smears, in turn, offer features of their own for measurement. It's easier to follow an edge in a single smear than a 2D blob from frame to frame, but it's not the same thing. The degree to which it gives the same numeric results determines the degree of equivalency.
The feature motion has to be collinear with the column in order to preserve feature motion in the smear curve. It is not precisely so in C447-449 but may be good enough. The measured curve is in reality a wandering point on the dark band, which wanders further than my hand.
Think of smear-o-grams as being like the Alta Vista translator. The last sentence of the paragraph above, when translated to Portuguese and back gives:
The measured curve is in the reality a point wandering in the dark band, that wanders more more of the one than my hand.
If that's good enough to get sufficiently similar meaning as the original statement, then so be it. But no amount of accuracy in further (real) translations or transcriptions will remove the existing distortions.
DBB -
I won't bug you about this anymore, I promise! It's not that big of a deal.
I won't bug you about this anymore, I promise! It's not that big of a deal.
QUOTE (OneWhiteEye+Jan 14 2008, 11:19 PM)
Thank you for making the effort. Vlooienspel would not have made the same impression.
A certain JPL scientist we know (not Jack Parsons) has made some pronouncements on the abilities of explosives to propel structural elements. I haven't looked at it, but I wonder how well it jibes with a structure that is known to have members propelled by explosive.
Is the piece you refer to ejected too early to be due to anything other than the charge(s)?
That piece is ejected from the zone where the building breaks, could be the same pop-out effect
A certain JPL scientist we know (not Jack Parsons) has made some pronouncements on the abilities of explosives to propel structural elements. I haven't looked at it, but I wonder how well it jibes with a structure that is known to have members propelled by explosive.
Is the piece you refer to ejected too early to be due to anything other than the charge(s)?
That piece is ejected from the zone where the building breaks, could be the same pop-out effect
ps. I found a very interesting video that shows the south-tower's top section falling exactly taken perpendicular, here it is
http://rapidshare.com/files/84004160/911.w...close.mpeg.html
this makes it possible to measure angle1(time). it is also from far away since they zoom out later. It looks like the building affects the top's toppling a little bit. Further there is a kink in the top section's corner. Get it!
http://rapidshare.com/files/84004160/911.w...close.mpeg.html
this makes it possible to measure angle1(time). it is also from far away since they zoom out later. It looks like the building affects the top's toppling a little bit. Further there is a kink in the top section's corner. Get it!
Einsteen:
Ejection of steel (or concrete) sections during the collapse of WTC 1 or 2 is caused by reflected elastic compression waves when they create tension above the yield stress.
Now the elastic strain energy capacity of steel is readily determined to be about 50 J/kg. If all the elastic strain energy is converted to kinetic energy we have:
K.E. = 1/2Mv^2, or mass specific kinetic energy = 1/2 v^2 = 50 J/kg
Therefore the maximum ejection velocity of a steel section is 10 m/s.
It's as simple as that!
Maybe......... perhaps..........
However, please note that this ignores strain rate effects!
Ejection of steel (or concrete) sections during the collapse of WTC 1 or 2 is caused by reflected elastic compression waves when they create tension above the yield stress.
Now the elastic strain energy capacity of steel is readily determined to be about 50 J/kg. If all the elastic strain energy is converted to kinetic energy we have:
K.E. = 1/2Mv^2, or mass specific kinetic energy = 1/2 v^2 = 50 J/kg
Therefore the maximum ejection velocity of a steel section is 10 m/s.
It's as simple as that!
Maybe......... perhaps..........
However, please note that this ignores strain rate effects!
Then the maximum distance it can reach is 92m
QUOTE (einsteen+Jan 15 2008, 02:47 PM)
ps. I found a very interesting video that shows the south-tower's top section falling exactly taken perpendicular, here it is
http://rapidshare.com/files/84004160/911.w...close.mpeg.html
this makes it possible to measure angle1(time). it is also from far away since they zoom out later. It looks like the building affects the top's toppling a little bit. Further there is a kink in the top section's corner. Get it!
Good find!
There's a good view of that exterior perimeter section that remains standing as the "avalanche" bypasses it. It's a bit hard to tell as the camera moves, but I was surprised at how close it seems to be to the first failing corner. In fact when it emerges from the smoke, it seems to be almost directly below where the corner was.
On a side note, any chance you can do a slow motion close up of the perimeter columns (to the right) like you did in this one?
http://rapidshare.com/files/84004160/911.w...close.mpeg.html
this makes it possible to measure angle1(time). it is also from far away since they zoom out later. It looks like the building affects the top's toppling a little bit. Further there is a kink in the top section's corner. Get it!
Good find!
There's a good view of that exterior perimeter section that remains standing as the "avalanche" bypasses it. It's a bit hard to tell as the camera moves, but I was surprised at how close it seems to be to the first failing corner. In fact when it emerges from the smoke, it seems to be almost directly below where the corner was.
On a side note, any chance you can do a slow motion close up of the perimeter columns (to the right) like you did in this one?
QUOTE (OneWhiteEye+Jan 14 2008, 11:44 PM)
It's not that big of a deal.
It's slowly sinking in.
Here is substantial disconfirmation of constant force with constant stretch via C447 data:
Moreover, the two parameters are highly unstable in comparison to C449 data.
It's slowly sinking in.
Here is substantial disconfirmation of constant force with constant stretch via C447 data:
CODE
Sef-K+Z+ZSS-F-exp-pow-stretch dB= 0.0 sd= 0.101
Sef-const-F-const-stretch dB= 7.8 sd= 0.308
Sef-const-F-const-stretch dB= 7.8 sd= 0.308
Moreover, the two parameters are highly unstable in comparison to C449 data.
Einsteen:
Well, if the projectile was a perimeter column, it would land (92 + 64) = 156 meters from the central vertical axis of a tower; this, I believe, is about the distance from the North Tower to the Winter Garden....
Well, if the projectile was a perimeter column, it would land (92 + 64) = 156 meters from the central vertical axis of a tower; this, I believe, is about the distance from the North Tower to the Winter Garden....
http://i16.tinypic.com/8e6ecgn.jpg
I think the winter garden is about 64 meter * 122pix/55pix=142 meter from the perimeter of wtc1
wfc3 is sqrt(137^2 + 13^2)*64 meter/55= 160 meter from the perimeter of wtc1
and that was even not at ground zero but 1/3-1/2 way the building
160 > 142 >> 92. the wfc3 beam is even more than a factor 1.7 too far, Powermetal!
O.W.E., I'll have a look later
QUOTE (NEU-FONZE+Jan 15 2008, 01:05 PM)
(92 + 64) = 156
92 + 32 = 122 meters from the center of the tower.
92 + 32 = 122 meters from the center of the tower.
If the constant resistive force is constrained to be 'reasonable', that is. almost as large as in Greening's paper, we obtain, using the C447 data,
CODE
Sef-K+Z+ZSS-F-exp-pow-stretch dB= 0.0 sd= 0.079
Sef-const-F-const-stretch dB= 8.9 sd= 0.324
Sef-const-F-const-stretch dB= 8.9 sd= 0.324
where not only is the poorer hypothesis substantively disconfirmed, but also the stretch parameter value is dramatically different than for the C447 data.
DBB:
Yes, sorry about that...
I'm having a "bad-math" day!
Yes, sorry about that...
I'm having a "bad-math" day!
QUOTE (einsteen+Jan 15 2008, 02:47 PM)
ps. I found a very interesting video that shows the south-tower's top section falling exactly taken perpendicular, here it is
http://rapidshare.com/files/84004160/911.w...close.mpeg.html
this makes it possible to measure angle1(time). it is also from far away since they zoom out later. It looks like the building affects the top's toppling a little bit. Further there is a kink in the top section's corner. Get it!
Got it! A great find, indeed. I watched again and again. Wow.
http://rapidshare.com/files/84004160/911.w...close.mpeg.html
this makes it possible to measure angle1(time). it is also from far away since they zoom out later. It looks like the building affects the top's toppling a little bit. Further there is a kink in the top section's corner. Get it!
Got it! A great find, indeed. I watched again and again. Wow.
QUOTE (NEU-FONZE+Jan 15 2008, 03:56 PM)
Einsteen:
Ejection of steel (or concrete) sections during the collapse of WTC 1 or 2 is caused by reflected elastic compression waves ...
I love that, on so many levels. Bad math day or not.
Ejection of steel (or concrete) sections during the collapse of WTC 1 or 2 is caused by reflected elastic compression waves ...
I love that, on so many levels. Bad math day or not.
QUOTE (lozenge124+Jan 15 2008, 06:12 PM)
...I was surprised at how close it seems to be to the first failing corner. In fact when it emerges from the smoke, it seems to be almost directly below where the corner was.
No doubt.
There appears to be a considerable volume of the glowing yellow liquid expelled from the area. Looks like a lot more than the total that had spilled out over the few minutes eariler. Given the apparent dimensions, a swimming pool of lava would not be a stretch for a rough description. Maybe a kiddie-pool.
No doubt.
There appears to be a considerable volume of the glowing yellow liquid expelled from the area. Looks like a lot more than the total that had spilled out over the few minutes eariler. Given the apparent dimensions, a swimming pool of lava would not be a stretch for a rough description. Maybe a kiddie-pool.
I never thought about the ejection due to the 50 J/kg but as a percentage of the front speed, but it is interesting, the ejection speed would be independent then. If a random cloud of mass hits an object then there can also be other effects, lemme put on the debunkers glasses,
- rotational energy of the debris cloud breaking perimeter columns
- fall over theory
- a real wedging, i.e. like a square block you drop on a triangle (with a 45 degree corner)
- rotational energy of the debris cloud breaking perimeter columns
- fall over theory
- a real wedging, i.e. like a square block you drop on a triangle (with a 45 degree corner)
QUOTE (David B. Benson+Jan 15 2008, 07:01 PM)
It's slowly sinking in.
Right on. It will all come out in the wash, eventually.
I expect the Etienne Sauret DVD to show up soon.
Right on. It will all come out in the wash, eventually.
I expect the Etienne Sauret DVD to show up soon.
QUOTE (einsteen+Jan 16 2008, 08:01 AM)
- rotational energy of the debris cloud breaking perimeter columns
- fall over theory
- a real wedging, i.e. like a square block you drop on a triangle (with a 45 degree corner)
Hey, einsteen. How's it going?
The equipartition theorem I guess is not supposed to apply. But I bet it does in some way.
Edit: some very twisted way, no longer recognizable as such...
- fall over theory
- a real wedging, i.e. like a square block you drop on a triangle (with a 45 degree corner)
Hey, einsteen. How's it going?
The equipartition theorem I guess is not supposed to apply. But I bet it does in some way.
Edit: some very twisted way, no longer recognizable as such...
OWE,
Going good here. With equipartition I guess you refer to those lines...
Lozenge,
I'll make a slow motion one (I didn't see it was you that you replied)
Going good here. With equipartition I guess you refer to those lines...
Lozenge,
I'll make a slow motion one (I didn't see it was you that you replied)
QUOTE (einsteen+Jan 16 2008, 08:36 AM)
With equipartition I guess you refer to those lines...
Equipartition of energy between all modes - translational, rotational, vibrational, etc. Far from equilibrium, yes, so maybe UNequipartition, but still with some sort of distribution for 'particle' masses and energies. The extremes of translational kinetic energy could be pretty high, at least once the collapse has been going a bit - avalanche and all that.
Equipartition of energy between all modes - translational, rotational, vibrational, etc. Far from equilibrium, yes, so maybe UNequipartition, but still with some sort of distribution for 'particle' masses and energies. The extremes of translational kinetic energy could be pretty high, at least once the collapse has been going a bit - avalanche and all that.
Looks like an almost undoable model, 1d is already difficult enough 
but if the ejection of mass requires energy maybe it needs to be added to the E1, for the total collapse it probably will not really differ because the collapse does not realy depend on it (unless the whole top section is ejected... or Blachard is right with his 95%....)
Lozenge,
Is that the kind of video you need ? http://rapidshare.com/files/84207748/lozenge.avi.html
but if the ejection of mass requires energy maybe it needs to be added to the E1, for the total collapse it probably will not really differ because the collapse does not realy depend on it (unless the whole top section is ejected... or Blachard is right with his 95%....)Lozenge,
Is that the kind of video you need ? http://rapidshare.com/files/84207748/lozenge.avi.html
QUOTE (OneWhiteEye+Jan 16 2008, 07:56 AM)
No doubt.
There appears to be a considerable volume of the glowing yellow liquid expelled from the area. Looks like a lot more than the total that had spilled out over the few minutes eariler. Given the apparent dimensions, a swimming pool of lava would not be a stretch for a rough description. Maybe a kiddie-pool.
IT is also interesting it is the right color for low melt ferrous chlorides.
There appears to be a considerable volume of the glowing yellow liquid expelled from the area. Looks like a lot more than the total that had spilled out over the few minutes eariler. Given the apparent dimensions, a swimming pool of lava would not be a stretch for a rough description. Maybe a kiddie-pool.
IT is also interesting it is the right color for low melt ferrous chlorides.
QUOTE (einsteen+Jan 16 2008, 08:01 AM)
I never thought about the ejection due to the 50 J/kg but as a percentage of the front speed, but it is interesting, the ejection speed would be independent then. If a random cloud of mass hits an object then there can also be other effects, lemme put on the debunkers glasses,
- rotational energy of the debris cloud breaking perimeter columns
- fall over theory
- a real wedging, i.e. like a square block you drop on a triangle (with a 45 degree corner)
How about naturally generated explosive gasses, they would be present in the collapse as well.
One of those is carbon monoxide.
- rotational energy of the debris cloud breaking perimeter columns
- fall over theory
- a real wedging, i.e. like a square block you drop on a triangle (with a 45 degree corner)
How about naturally generated explosive gasses, they would be present in the collapse as well.
One of those is carbon monoxide.
QUOTE (einsteen+Jan 16 2008, 09:40 AM)
Lozenge,
Is that the kind of video you need ? http://rapidshare.com/files/84207748/lozenge.avi.html
Thanks einsteen, that's the area I was looking for. I wouldn't mind zooming in a bit more though, but instead of taking up your time again, what software do you do this video editing in? (so I can play around myself a bit)
Is that the kind of video you need ? http://rapidshare.com/files/84207748/lozenge.avi.html
Thanks einsteen, that's the area I was looking for. I wouldn't mind zooming in a bit more though, but instead of taking up your time again, what software do you do this video editing in? (so I can play around myself a bit)
Recently I discovered the ffdshow codecs, that makes it possible to open
most avi files (the mpeg4/xvid/divx related ones), codecs are in general a nag. There is
a command line tool called ffmpeg that I use in cases that the files are .rm/.mov/.mp4/.asf/.wmv
because virtualdub only is able to process avi's. There is a workaround to open windows media (using
direct show) but that is a whole story.
virtualdub is the main program I use (a free program developed by a whizkid),
I have several versions and use them for a long time. The whole program
is in fact two screens, a left one for the input video and the other for output.
You can use filters to resize/crop/adjust saturation/rotate/remove compression blocks,
especially the MSU filters are very useful. But it takes some time to learn it, I only
know the things that I have to use and have no idea about its potential capacities.
The output file will be an avi (I see that the new version can even create a .gif directly) and there you have to choose your compression. Always work with raw avi files until you are finished.
What I did with the original mpeg was
1) opening with virtual dub
2) cropping the area
3) changing framerate
4) saving as raw avi
5) make a mirrored avi (reverse duplicate)
6) paste them to a new avi
7) use 6) and make 4 copies and paste them again to one big one
8) compress the thing
For 5) I had to use the recently discovered avisynth but this is enough for now.
I'm sure that premiere and all those fat packages can also do it, maybe even more
user friendly but they are huge. A normal out-of-the-box virtualdub is already
very useful if you want to see a movie frame by frame. If a codec is missing then
it will complain and you have to google for it.
most avi files (the mpeg4/xvid/divx related ones), codecs are in general a nag. There is
a command line tool called ffmpeg that I use in cases that the files are .rm/.mov/.mp4/.asf/.wmv
because virtualdub only is able to process avi's. There is a workaround to open windows media (using
direct show) but that is a whole story.
virtualdub is the main program I use (a free program developed by a whizkid),
I have several versions and use them for a long time. The whole program
is in fact two screens, a left one for the input video and the other for output.
You can use filters to resize/crop/adjust saturation/rotate/remove compression blocks,
especially the MSU filters are very useful. But it takes some time to learn it, I only
know the things that I have to use and have no idea about its potential capacities.
The output file will be an avi (I see that the new version can even create a .gif directly) and there you have to choose your compression. Always work with raw avi files until you are finished.
What I did with the original mpeg was
1) opening with virtual dub
2) cropping the area
3) changing framerate
4) saving as raw avi
5) make a mirrored avi (reverse duplicate)
6) paste them to a new avi
7) use 6) and make 4 copies and paste them again to one big one
8) compress the thing
For 5) I had to use the recently discovered avisynth but this is enough for now.
I'm sure that premiere and all those fat packages can also do it, maybe even more
user friendly but they are huge. A normal out-of-the-box virtualdub is already
very useful if you want to see a movie frame by frame. If a codec is missing then
it will complain and you have to google for it.
Thanks einsteen.
QUOTE (OneWhiteEye+Jan 16 2008, 01:01 AM)
Etienne Sauret
Is this the operator of the video we are extracting information from?
Is this the operator of the video we are extracting information from?
Just wanted to post this here for comment, you can call me insane, and after a couple of these experiments I would probably agree with you. 
http://forums.randi.org/showthread.php?p=3343056#post3343056
http://forums.randi.org/showthread.php?p=3343056#post3343056
QUOTE (David B. Benson+Jan 16 2008, 06:57 PM)
Is this the operator of the video we are extracting information from?
Yes. Turns out Etienne Sauret is a filmmaker by profession. I may contact Sauret about the exact location if it's not obvious from watching the DVD. The one frame shagster posted, I believe, has the right and bottom of the frame bordered by building(s), cropped out of my internet copy.
There are some consequences of going to the original source. New frame numbers are likely, which can be synced with the old set if desired. I'm going redo scaling but it probably won't change much. The scale needed a little attention, anyway. New data must come from the best copy possible.
Yes. Turns out Etienne Sauret is a filmmaker by profession. I may contact Sauret about the exact location if it's not obvious from watching the DVD. The one frame shagster posted, I believe, has the right and bottom of the frame bordered by building(s), cropped out of my internet copy.
There are some consequences of going to the original source. New frame numbers are likely, which can be synced with the old set if desired. I'm going redo scaling but it probably won't change much. The scale needed a little attention, anyway. New data must come from the best copy possible.
QUOTE (Chainsaw+,Jan 16 2008, 07:27 PM)
Just wanted to post this here for comment, you can call me insane, and after a couple of these experiments I would probably agree with you.
http://forums.randi.org/showthread.php?p=3343056#post3343056
OK, you're insane. The undisputed champion of back-40 chemistry.
http://forums.randi.org/showthread.php?p=3343056#post3343056
OK, you're insane. The undisputed champion of back-40 chemistry.
QUOTE (Chainsaw+,Jan 16 2008, 12:44 PM)
IT is also interesting it is the right color for low melt ferrous chlorides.
I'll defer to you and NEU-FONZE for that. I know more voodoo than chemistry. Let me see if I can estimate the possible bounds on the volume of material. It looked like an awful lot, but that might be an impression from a few runs at full speed. Seems worth a second look, frame by frame.
Certainly an interesting phenomena. You'll have to cut me some slack because I've had only a dim awareness of the discussions concerned with globs/thermite/UPS/lead etc. I know someone made a gross estimate of volume of the globs prior to collapse, and then there were arguments about whether or not temp can be reliably inferred from image color and what that temp might be, and some discussion about impurities and color. Here or elsewhere, don't recall.
What sort of temperature would the ferrous chlorides be at? Is this an ongoing combustion process, by-product of a combustion process, externally heated office soup, something else? Iron + PVC partial combustion products + high temp? Which direction is the heat flowing, in or out?
I'll defer to you and NEU-FONZE for that. I know more voodoo than chemistry. Let me see if I can estimate the possible bounds on the volume of material. It looked like an awful lot, but that might be an impression from a few runs at full speed. Seems worth a second look, frame by frame.
Certainly an interesting phenomena. You'll have to cut me some slack because I've had only a dim awareness of the discussions concerned with globs/thermite/UPS/lead etc. I know someone made a gross estimate of volume of the globs prior to collapse, and then there were arguments about whether or not temp can be reliably inferred from image color and what that temp might be, and some discussion about impurities and color. Here or elsewhere, don't recall.
What sort of temperature would the ferrous chlorides be at? Is this an ongoing combustion process, by-product of a combustion process, externally heated office soup, something else? Iron + PVC partial combustion products + high temp? Which direction is the heat flowing, in or out?
Chainsaw:
Looking at the liquid frame by frame, there's not as much as it seems when the video is rolling.
http://i7.tinypic.com/8br4yzb.png
Some of it at the top is flame (about 20%), which goes UP, no surprise. Of what goes down after this frame, I measure roughly a 20x70 pixel area in this frame. The width of five columns plus their windows is 37 pixels. If the width of a column plus window is 40 inches, then the liquid presents an apparent cross-sectional area of:
20 px x 70 px
=> 2.748 m x 9.618 m
~ 26 m^2
The thickness is unknown, but it's reasonable to think it's at least 5cm. It is gushing outward toward the camera but just call it a vertical sheet to stay to a minimum. Then, there's the question of how much of the volume is void. Say the material occupies only 25% of the splash volume (that could be high). Then,
(26 m^2) x (0.05m) x (0.25) = 0.325 m^3
or a cube about 7/10 meter on a side. Not even a kiddie pool, more like a barrel and a half. That is the absolute minimum, though, on what's visible in one particular frame. There's no reason to believe that anywhere near all of the material contained in the corner is visible at the surface of the debris cloud, or that all of it was ejected instead of remaining interior. Later frames look like greater volume continues to be disgorged, but it's more difficult to distinguish flame from the liquid with which it seems associated.
That which has already spilled out has to be added to the total, which then must be of the order of barrels, at a minimum.
Looking at the liquid frame by frame, there's not as much as it seems when the video is rolling.
http://i7.tinypic.com/8br4yzb.png
Some of it at the top is flame (about 20%), which goes UP, no surprise. Of what goes down after this frame, I measure roughly a 20x70 pixel area in this frame. The width of five columns plus their windows is 37 pixels. If the width of a column plus window is 40 inches, then the liquid presents an apparent cross-sectional area of:
20 px x 70 px
=> 2.748 m x 9.618 m
~ 26 m^2
The thickness is unknown, but it's reasonable to think it's at least 5cm. It is gushing outward toward the camera but just call it a vertical sheet to stay to a minimum. Then, there's the question of how much of the volume is void. Say the material occupies only 25% of the splash volume (that could be high). Then,
(26 m^2) x (0.05m) x (0.25) = 0.325 m^3
or a cube about 7/10 meter on a side. Not even a kiddie pool, more like a barrel and a half. That is the absolute minimum, though, on what's visible in one particular frame. There's no reason to believe that anywhere near all of the material contained in the corner is visible at the surface of the debris cloud, or that all of it was ejected instead of remaining interior. Later frames look like greater volume continues to be disgorged, but it's more difficult to distinguish flame from the liquid with which it seems associated.
That which has already spilled out has to be added to the total, which then must be of the order of barrels, at a minimum.
QUOTE (Chainsaw+,Jan 16 2008, 12:44 PM)
IT is also interesting it is the right color for low melt ferrous chlorides.
Which color, this one:
http://i16.tinypic.com/6jx15op.png
or this one less than a half second later:
http://i16.tinypic.com/72j4sa1.png
Or this one where I tinkered with the gamma and contrast:
http://i11.tinypic.com/7xj7wbs.png
With or without video compression? Or fries?
These are prior to the frame above.
Which color, this one:
http://i16.tinypic.com/6jx15op.png
or this one less than a half second later:
http://i16.tinypic.com/72j4sa1.png
Or this one where I tinkered with the gamma and contrast:
http://i11.tinypic.com/7xj7wbs.png
With or without video compression? Or fries?
These are prior to the frame above.
QUOTE (OneWhiteEye+Jan 17 2008, 04:33 AM)
I'll defer to you and NEU-FONZE for that. I know more voodoo than chemistry. Let me see if I can estimate the possible bounds on the volume of material. It looked like an awful lot, but that might be an impression from a few runs at full speed. Seems worth a second look, frame by frame.
Certainly an interesting phenomena. You'll have to cut me some slack because I've had only a dim awareness of the discussions concerned with globs/thermite/UPS/lead etc. I know someone made a gross estimate of volume of the globs prior to collapse, and then there were arguments about whether or not temp can be reliably inferred from image color and what that temp might be, and some discussion about impurities and color. Here or elsewhere, don't recall.
What sort of temperature would the ferrous chlorides be at? Is this an ongoing combustion process, by-product of a combustion process, externally heated office soup, something else? Iron + PVC partial combustion products + high temp? Which direction is the heat flowing, in or out?
OnewhiteEye,
Any time iron oxidizes your going to have a significant amount of heat produced by the oxidizing iron, the amount of heat is directly dependent on the amount of oxidation that takes place over a given area.
Slow oxidation and the Iron chloride would be orange to yellow, quick oxidation and it would appear white. If you have oxidizing Iron with aluminum or carbon definitely white.
So long as you have reactants the reactions can continue.
It is the explosive Carbon monoxide that starts the reaction though, it has almost the explosive range of hydrogen.
In fact it is the only explosive gas that would work besides hydrogen.
Certainly an interesting phenomena. You'll have to cut me some slack because I've had only a dim awareness of the discussions concerned with globs/thermite/UPS/lead etc. I know someone made a gross estimate of volume of the globs prior to collapse, and then there were arguments about whether or not temp can be reliably inferred from image color and what that temp might be, and some discussion about impurities and color. Here or elsewhere, don't recall.
What sort of temperature would the ferrous chlorides be at? Is this an ongoing combustion process, by-product of a combustion process, externally heated office soup, something else? Iron + PVC partial combustion products + high temp? Which direction is the heat flowing, in or out?
OnewhiteEye,
Any time iron oxidizes your going to have a significant amount of heat produced by the oxidizing iron, the amount of heat is directly dependent on the amount of oxidation that takes place over a given area.
Slow oxidation and the Iron chloride would be orange to yellow, quick oxidation and it would appear white. If you have oxidizing Iron with aluminum or carbon definitely white.
So long as you have reactants the reactions can continue.
It is the explosive Carbon monoxide that starts the reaction though, it has almost the explosive range of hydrogen.
In fact it is the only explosive gas that would work besides hydrogen.
QUOTE (OneWhiteEye+Jan 17 2008, 07:55 AM)
Which color, this one:
http://i16.tinypic.com/6jx15op.png
or this one less than a half second later:
http://i16.tinypic.com/72j4sa1.png
Or this one where I tinkered with the gamma and contrast:
http://i11.tinypic.com/7xj7wbs.png
With or without video compression? Or fries?
These are prior to the frame above.
Those pictures are similar to what I have observed, in the ferrous chlorides, a range of colors dependent on the reaction with different elements in the mix.
Dr. Joneses experiments could not have produced ferrous chlorides, they require entrapment of the HCl to react with the Fe, creating FeCl3 and carbon monoxide, if carbon smoke is present.
After the carbon monoxide gathers at the ceiling an air movement it required to oxidize it and set off the explosion, that could occur easily it is an unstable gas layer.
Basically just opening a door could do it, or a slight gust of wind, a helicopter down wash, anything that allowed oxygen to the carbon monoxide.
I just can not figure out why NIST did not model this in the fire tests that they performed.
http://i16.tinypic.com/6jx15op.png
or this one less than a half second later:
http://i16.tinypic.com/72j4sa1.png
Or this one where I tinkered with the gamma and contrast:
http://i11.tinypic.com/7xj7wbs.png
With or without video compression? Or fries?
These are prior to the frame above.
Those pictures are similar to what I have observed, in the ferrous chlorides, a range of colors dependent on the reaction with different elements in the mix.
Dr. Joneses experiments could not have produced ferrous chlorides, they require entrapment of the HCl to react with the Fe, creating FeCl3 and carbon monoxide, if carbon smoke is present.
After the carbon monoxide gathers at the ceiling an air movement it required to oxidize it and set off the explosion, that could occur easily it is an unstable gas layer.
Basically just opening a door could do it, or a slight gust of wind, a helicopter down wash, anything that allowed oxygen to the carbon monoxide.
I just can not figure out why NIST did not model this in the fire tests that they performed.
Chainsaw:
Interestingly NIST gives some carbon monoxide data for its multiple workstation burns in Chapter 7 of NCSTAR 1-5E. The concentrations are less than 5 vol %. I am sure the CO concs would be higher if PVC was burning - Masuda et al. show that in the combustion of PVC-Fe2O3 mixtures there is enough CO to reduce the iron oxide to metallic iron!
Interestingly NIST gives some carbon monoxide data for its multiple workstation burns in Chapter 7 of NCSTAR 1-5E. The concentrations are less than 5 vol %. I am sure the CO concs would be higher if PVC was burning - Masuda et al. show that in the combustion of PVC-Fe2O3 mixtures there is enough CO to reduce the iron oxide to metallic iron!
QUOTE (NEU-FONZE+Jan 18 2008, 11:23 AM)
Chainsaw:
Interestingly NIST gives some carbon monoxide data for its multiple workstation burns in Chapter 7 of NCSTAR 1-5E. The concentrations are less than 5 vol %. I am sure the CO concs would be higher if PVC was burning - Masuda et al. show that in the combustion of PVC-Fe2O3 mixtures there is enough CO to reduce the iron oxide to metallic iron!
I know that is what my fire tests were all about finding a mechanism that would produce a solid metallic iron micro sphere with a good deal of sulfur, from the reduction of iron oxide with carbon.
I was never able to do it because at 800c the sulfur is about all oxidized to SO2, although I did produce iron micro spheres and molten iron-iron oxide.
Steel when oxidized produces enough heat to melt steel, that is basically how an oxygen lance works.
Chlorides and Acids break down the Crystalline structures of steel and lower the initiation temperature of steel oxidation, but the heat produced at oxidation remains constant.
What do you think the Co would be if PVC, and sulfuric acid from a lead acid battery power source were involved, you will have a reaction similar to chlorosulfonic super acid.
Aluminum chloride and iron chloride easily mix, and oxidize to Iron and Aluminum oxide at a temperature that will even melt aluminum oxide.
Here are some good link I found on steel oxidation.
http://www.ingentaconnect.com/content/klu/...020005/00463539
http://en.wikipedia.org/wiki/Oxy-fuel_welding_and_cutting
In Oxy-fuel cutting, a cutting torch is used to heat up ferrous metal to kindling temperature (about 980°C). A stream of pure oxygen is trained on the hot metal which chemically combines with the iron which then flows out of the cut, or kerf, as an iron-oxide slag
http://en.wikipedia.org/wiki/Thermic_lance
Thermal lances burn at 7000 to 8000°F (3870 to 4420°C)[citation needed], melting even rocks.
Is it any wonder that when doing experiments that recreate the conditions in the buildings I see molten steel, and explosions?
Interestingly NIST gives some carbon monoxide data for its multiple workstation burns in Chapter 7 of NCSTAR 1-5E. The concentrations are less than 5 vol %. I am sure the CO concs would be higher if PVC was burning - Masuda et al. show that in the combustion of PVC-Fe2O3 mixtures there is enough CO to reduce the iron oxide to metallic iron!
I know that is what my fire tests were all about finding a mechanism that would produce a solid metallic iron micro sphere with a good deal of sulfur, from the reduction of iron oxide with carbon.
I was never able to do it because at 800c the sulfur is about all oxidized to SO2, although I did produce iron micro spheres and molten iron-iron oxide.
Steel when oxidized produces enough heat to melt steel, that is basically how an oxygen lance works.
Chlorides and Acids break down the Crystalline structures of steel and lower the initiation temperature of steel oxidation, but the heat produced at oxidation remains constant.
What do you think the Co would be if PVC, and sulfuric acid from a lead acid battery power source were involved, you will have a reaction similar to chlorosulfonic super acid.
Aluminum chloride and iron chloride easily mix, and oxidize to Iron and Aluminum oxide at a temperature that will even melt aluminum oxide.
Here are some good link I found on steel oxidation.
http://www.ingentaconnect.com/content/klu/...020005/00463539
http://en.wikipedia.org/wiki/Oxy-fuel_welding_and_cutting
In Oxy-fuel cutting, a cutting torch is used to heat up ferrous metal to kindling temperature (about 980°C). A stream of pure oxygen is trained on the hot metal which chemically combines with the iron which then flows out of the cut, or kerf, as an iron-oxide slag
http://en.wikipedia.org/wiki/Thermic_lance
Thermal lances burn at 7000 to 8000°F (3870 to 4420°C)[citation needed], melting even rocks.
Is it any wonder that when doing experiments that recreate the conditions in the buildings I see molten steel, and explosions?
QUOTE (Chainsaw+,Jan 17 2008, 11:10 AM)
OnewhiteEye,
Any time iron oxidizes your going to have a significant amount of heat produced by the oxidizing iron, the amount of heat is directly dependent on the amount of oxidation that takes place over a given area.
Slow oxidation and the Iron chloride would be orange to yellow, quick oxidation and it would appear white. If you have oxidizing Iron with aluminum or carbon definitely white.
So long as you have reactants the reactions can continue.
It is the explosive Carbon monoxide that starts the reaction though, it has almost the explosive range of hydrogen.
In fact it is the only explosive gas that would work besides hydrogen.
Thanks for your response. My curiousity is aroused but I'll have to back-burner this for awhile. Before I do, let me ask a couple more questions and say why I find this so odd; perhaps you or someone else can explain this or describe a scenario in which it makes sense.
The melting point of FeCl2 is 677C. Allow the melting point to be depressed by the introduction of impurities into the already melted material, so that the material does not become solid until below 600C. That means there is directly observable evidence that temperature, somewhere in the tower, reached above 600C.
If you are correct about the nature of substance to the extent that it produced some or all of the heat via its own formation, then it's a different case than coming to equilibrium with the environment. There's no need to postulate ambient temps that high. But I have to wonder about the environment in that case, as well. Would the reactants be mixed together already, or is constant mixing required to sustain the reaction? Are the blobs STILL reacting as we see them emerge, or have they concluded at a prior time and are in the process of cooling?
Some important situational distinctions:
- Produced some heat through reaction / formation vs heated exclusively by the environment
- Produced at the location observed vs migrated there from another location
Ponder that and I'll continue in a moment.
-
Any time iron oxidizes your going to have a significant amount of heat produced by the oxidizing iron, the amount of heat is directly dependent on the amount of oxidation that takes place over a given area.
Slow oxidation and the Iron chloride would be orange to yellow, quick oxidation and it would appear white. If you have oxidizing Iron with aluminum or carbon definitely white.
So long as you have reactants the reactions can continue.
It is the explosive Carbon monoxide that starts the reaction though, it has almost the explosive range of hydrogen.
In fact it is the only explosive gas that would work besides hydrogen.
Thanks for your response. My curiousity is aroused but I'll have to back-burner this for awhile. Before I do, let me ask a couple more questions and say why I find this so odd; perhaps you or someone else can explain this or describe a scenario in which it makes sense.
The melting point of FeCl2 is 677C. Allow the melting point to be depressed by the introduction of impurities into the already melted material, so that the material does not become solid until below 600C. That means there is directly observable evidence that temperature, somewhere in the tower, reached above 600C.
If you are correct about the nature of substance to the extent that it produced some or all of the heat via its own formation, then it's a different case than coming to equilibrium with the environment. There's no need to postulate ambient temps that high. But I have to wonder about the environment in that case, as well. Would the reactants be mixed together already, or is constant mixing required to sustain the reaction? Are the blobs STILL reacting as we see them emerge, or have they concluded at a prior time and are in the process of cooling?
Some important situational distinctions:
- Produced some heat through reaction / formation vs heated exclusively by the environment
- Produced at the location observed vs migrated there from another location
Ponder that and I'll continue in a moment.
-
I realize the spillage of this material was observed on video years ago and is the source of some controversy. What ramps up my curiousity from NFI to what the hell is the quantity of material located in that corner. Not the quantity of material, per se, but the quantity being located there.
Let me start with the second of the two considerations I listed above. Does the material form in place or flow to there from somewhere else? While it's not possible to see the conditions interior to the corner, comparatively speaking, it is not as hot (as observed from the outside) as many other portions of the tower at that time. In fact, I think it's adjacent to the noted cool spot. The window holes are dark. So, while it's technically impossible to rule out high enough ambient temperatures, it doesn't seem as likely as the formation of this substance in another location. This really includes the case of chemical reaction in place, as well, because any exothermic reaction capable of producing that quantity in situ should be observable from the outside (and I'd argue that point vigorously).
Without fully eliminating the possibility, let's examine the case of formation elsewhere with migration to observed location.
- how did it get there?
- what is containing it at the corner?
Ignore the fact that migration to a cooler area implies material loss due to boundary condensation and initial higher material temps than those directly observed. What is the structural channel to conduct the flow to this corner? Granted, the blueprints are of little use when debris has been plowed around, partial floor collapses may have occurred; so this question is less interesting than the second: what part of the structure in this corner is holding this substantial reservoir of extremely hot liquid?
Given the estimated minimum quantity of material, the minimum temp, the known period of time the material was present at the corner, and further surmising that the eariler spillage was overflow from a reservoir as opposed to simple leakage of flow to the outside, I think it's also fair to ask:
What is the impact on the structural integrity of even a fully intact tower corner?
Let me start with the second of the two considerations I listed above. Does the material form in place or flow to there from somewhere else? While it's not possible to see the conditions interior to the corner, comparatively speaking, it is not as hot (as observed from the outside) as many other portions of the tower at that time. In fact, I think it's adjacent to the noted cool spot. The window holes are dark. So, while it's technically impossible to rule out high enough ambient temperatures, it doesn't seem as likely as the formation of this substance in another location. This really includes the case of chemical reaction in place, as well, because any exothermic reaction capable of producing that quantity in situ should be observable from the outside (and I'd argue that point vigorously).
Without fully eliminating the possibility, let's examine the case of formation elsewhere with migration to observed location.
- how did it get there?
- what is containing it at the corner?
Ignore the fact that migration to a cooler area implies material loss due to boundary condensation and initial higher material temps than those directly observed. What is the structural channel to conduct the flow to this corner? Granted, the blueprints are of little use when debris has been plowed around, partial floor collapses may have occurred; so this question is less interesting than the second: what part of the structure in this corner is holding this substantial reservoir of extremely hot liquid?
Given the estimated minimum quantity of material, the minimum temp, the known period of time the material was present at the corner, and further surmising that the eariler spillage was overflow from a reservoir as opposed to simple leakage of flow to the outside, I think it's also fair to ask:
What is the impact on the structural integrity of even a fully intact tower corner?
QUOTE (OneWhiteEye+Jan 18 2008, 05:43 PM)
Thanks for your response. My curiousity is aroused but I'll have to back-burner this for awhile. Before I do, let me ask a couple more questions and say why I find this so odd; perhaps you or someone else can explain this or describe a scenario in which it makes sense.
The melting point of FeCl2 is 677C. Allow the melting point to be depressed by the introduction of impurities into the already melted material, so that the material does not become solid until below 600C. That means there is directly observable evidence that temperature, somewhere in the tower, reached above 600C.
If you are correct about the nature of substance to the extent that it produced some or all of the heat via its own formation, then it's a different case than coming to equilibrium with the environment. There's no need to postulate ambient temps that high. But I have to wonder about the environment in that case, as well. Would the reactants be mixed together already, or is constant mixing required to sustain the reaction? Are the blobs STILL reacting as we see them emerge, or have they concluded at a prior time and are in the process of cooling?
Some important situational distinctions:
- Produced some heat through reaction / formation vs heated exclusively by the environment
- Produced at the location observed vs migrated there from another location
Ponder that and I'll continue in a moment.
-
They are produced by the entrapment of iron with HCl gas given off from PVC.
a small amount mixed with say lead and aluminum chloride would be sufficient.
The oxidation of the steel and aluminum would produce enough heat for that and it would continue as long as the substance was in motion, exposing fresh hot chloride containing metals to oxidation.
The boiling HCI bubbles would form hollow micro spheres of Fe 3O4 I believe.
They are produced by the entrapment of iron with HCl gas given off from PVC.
a small amount mixed with say lead and aluminum chloride would be sufficient.
The oxidation of the steel and aluminum would produce enough heat for that and it would continue as long as the substance was in motion, exposing fresh hot chloride containing metals to oxidation.
The boiling HCI bubbles would form hollow micro spheres of Fe 3O4 I believe.
- Produced at the location observed vs migrated there from another location
What I have seen is FeCl3 formed at the ceiling floor pan level with Zinc chloride as well as Zinc oxide raining down onto anything below.
It would definitely be a very complicated, mix hard to duplicate without doing actual experiments of similar conditions, such as I have done.
Without entrapment of the HCl gas there is no formation of the compound, with it, it can not be prevented from forming.
Dr. Jones mixed plastic pipe PVC and dry wall into a pan and heated it, he did not try any gas entrapment experiments as far as I know.
If you had damaged hollow steel structures you could literally form tons of Fe Cl3 in fires with tons of PVC.
water tanks or metal filing cabinets would be especially susceptible, and thin steels.
I have seen the reactants that look exactly similar to the ones in the towers, Aluminum Chloride and Iron chloride will mix without oxidation-thermite reaction, but if iron chloride oxidizes in presence of aluminum it will cause an aluminum oxidation reaction as it heats from its own oxidation.
I am trying to figure it out OneWhiteEye however it is complicated and dangerous, not easy to do, I do not know if anyone though of complicated Chloride metal reactions before, it is not something one would normally think of, just stumbled across it be accident.
The melting point of FeCl2 is 677C. Allow the melting point to be depressed by the introduction of impurities into the already melted material, so that the material does not become solid until below 600C. That means there is directly observable evidence that temperature, somewhere in the tower, reached above 600C.
If you are correct about the nature of substance to the extent that it produced some or all of the heat via its own formation, then it's a different case than coming to equilibrium with the environment. There's no need to postulate ambient temps that high. But I have to wonder about the environment in that case, as well. Would the reactants be mixed together already, or is constant mixing required to sustain the reaction? Are the blobs STILL reacting as we see them emerge, or have they concluded at a prior time and are in the process of cooling?
Some important situational distinctions:
- Produced some heat through reaction / formation vs heated exclusively by the environment
- Produced at the location observed vs migrated there from another location
Ponder that and I'll continue in a moment.
-
QUOTE
- Produced some heat through reaction / formation vs heated exclusively by the environment.
They are produced by the entrapment of iron with HCl gas given off from PVC.
a small amount mixed with say lead and aluminum chloride would be sufficient.
The oxidation of the steel and aluminum would produce enough heat for that and it would continue as long as the substance was in motion, exposing fresh hot chloride containing metals to oxidation.
The boiling HCI bubbles would form hollow micro spheres of Fe 3O4 I believe.
QUOTE (->
| QUOTE |
| - Produced some heat through reaction / formation vs heated exclusively by the environment. |
They are produced by the entrapment of iron with HCl gas given off from PVC.
a small amount mixed with say lead and aluminum chloride would be sufficient.
The oxidation of the steel and aluminum would produce enough heat for that and it would continue as long as the substance was in motion, exposing fresh hot chloride containing metals to oxidation.
The boiling HCI bubbles would form hollow micro spheres of Fe 3O4 I believe.
- Produced at the location observed vs migrated there from another location
What I have seen is FeCl3 formed at the ceiling floor pan level with Zinc chloride as well as Zinc oxide raining down onto anything below.
It would definitely be a very complicated, mix hard to duplicate without doing actual experiments of similar conditions, such as I have done.
Without entrapment of the HCl gas there is no formation of the compound, with it, it can not be prevented from forming.
Dr. Jones mixed plastic pipe PVC and dry wall into a pan and heated it, he did not try any gas entrapment experiments as far as I know.
If you had damaged hollow steel structures you could literally form tons of Fe Cl3 in fires with tons of PVC.
water tanks or metal filing cabinets would be especially susceptible, and thin steels.
I have seen the reactants that look exactly similar to the ones in the towers, Aluminum Chloride and Iron chloride will mix without oxidation-thermite reaction, but if iron chloride oxidizes in presence of aluminum it will cause an aluminum oxidation reaction as it heats from its own oxidation.
I am trying to figure it out OneWhiteEye however it is complicated and dangerous, not easy to do, I do not know if anyone though of complicated Chloride metal reactions before, it is not something one would normally think of, just stumbled across it be accident.
QUOTE (OneWhiteEye+Jan 18 2008, 06:19 PM)
I realize the spillage of this material was observed on video years ago and is the source of some controversy. What ramps up my curiousity from NFI to what the hell is the quantity of material located in that corner. Not the quantity of material, per se, but the quantity being located there.
Let me start with the second of the two considerations I listed above. Does the material form in place or flow to there from somewhere else? While it's not possible to see the conditions interior to the corner, comparatively speaking, it is not as hot (as observed from the outside) as many other portions of the tower at that time. In fact, I think it's adjacent to the noted cool spot. The window holes are dark. So, while it's technically impossible to rule out high enough ambient temperatures, it doesn't seem as likely as the formation of this substance in another location. This really includes the case of chemical reaction in place, as well, because any exothermic reaction capable of producing that quantity in situ should be observable from the outside (and I'd argue that point vigorously).
Without fully eliminating the possibility, let's examine the case of formation elsewhere with migration to observed location.
- how did it get there?
- what is containing it at the corner?
Ignore the fact that migration to a cooler area implies material loss due to boundary condensation and initial higher material temps than those directly observed. What is the structural channel to conduct the flow to this corner? Granted, the blueprints are of little use when debris has been plowed around, partial floor collapses may have occurred; so this question is less interesting than the second: what part of the structure in this corner is holding this substantial reservoir of extremely hot liquid?
Given the estimated minimum quantity of material, the minimum temp, the known period of time the material was present at the corner, and further surmising that the eariler spillage was overflow from a reservoir as opposed to simple leakage of flow to the outside, I think it's also fair to ask:
What is the impact on the structural integrity of even a fully intact tower corner?
, because any exothermic reaction capable of producing that quantity in situ should be observable from the outside (and I'd argue that point vigorously).
No it would not be noticed the gases that formed the substances are invisible, only when the oxidation started would you see any temperature or light increases.
In a reducing atmosphere there would be no reaction at all, until oxygen was supplied.
The chlorides form over time from entrapped gasses rain down.
Iron III Chloride with not react with aluminum until oxygen is present, except in a replacement reaction. Ferrous Chloride, plus Aluminum = Iron plus aluminum chloride. Ferrous Chloride plus Oxygen, = Fe 3O4 +heat, plus Aluminum = heat and Al2O3.
I do not have all the answers the more experiment I do the more confused I get as to what is actually going on it is getting to technical for me at this point, however I am trying to describe accurately what I have observed.
Let me start with the second of the two considerations I listed above. Does the material form in place or flow to there from somewhere else? While it's not possible to see the conditions interior to the corner, comparatively speaking, it is not as hot (as observed from the outside) as many other portions of the tower at that time. In fact, I think it's adjacent to the noted cool spot. The window holes are dark. So, while it's technically impossible to rule out high enough ambient temperatures, it doesn't seem as likely as the formation of this substance in another location. This really includes the case of chemical reaction in place, as well, because any exothermic reaction capable of producing that quantity in situ should be observable from the outside (and I'd argue that point vigorously).
Without fully eliminating the possibility, let's examine the case of formation elsewhere with migration to observed location.
- how did it get there?
- what is containing it at the corner?
Ignore the fact that migration to a cooler area implies material loss due to boundary condensation and initial higher material temps than those directly observed. What is the structural channel to conduct the flow to this corner? Granted, the blueprints are of little use when debris has been plowed around, partial floor collapses may have occurred; so this question is less interesting than the second: what part of the structure in this corner is holding this substantial reservoir of extremely hot liquid?
Given the estimated minimum quantity of material, the minimum temp, the known period of time the material was present at the corner, and further surmising that the eariler spillage was overflow from a reservoir as opposed to simple leakage of flow to the outside, I think it's also fair to ask:
What is the impact on the structural integrity of even a fully intact tower corner?
, because any exothermic reaction capable of producing that quantity in situ should be observable from the outside (and I'd argue that point vigorously).
No it would not be noticed the gases that formed the substances are invisible, only when the oxidation started would you see any temperature or light increases.
In a reducing atmosphere there would be no reaction at all, until oxygen was supplied.
The chlorides form over time from entrapped gasses rain down.
Iron III Chloride with not react with aluminum until oxygen is present, except in a replacement reaction. Ferrous Chloride, plus Aluminum = Iron plus aluminum chloride. Ferrous Chloride plus Oxygen, = Fe 3O4 +heat, plus Aluminum = heat and Al2O3.
I do not have all the answers the more experiment I do the more confused I get as to what is actually going on it is getting to technical for me at this point, however I am trying to describe accurately what I have observed.
Again, thanks for the fascinating and informative replies, Chainsaw. It makes the work you're doing more relevant to me when I see such obvious outward manifestations.
See, I wasn't following your work closely enough to understand the potential quantites involved, from the standpoint of pooling. The other explanations I've heard are lead from batteries and aluminum skin from the fuselage, other aircraft components (magnesium, etc). These also seem capable of supplying the needed quantity.
Are you aware of any estimate of the minimum volume where the amount is similar to what I've calculated (for that specific area)?
See, I wasn't following your work closely enough to understand the potential quantites involved, from the standpoint of pooling. The other explanations I've heard are lead from batteries and aluminum skin from the fuselage, other aircraft components (magnesium, etc). These also seem capable of supplying the needed quantity.
Are you aware of any estimate of the minimum volume where the amount is similar to what I've calculated (for that specific area)?
No it would not be noticed the gases that formed the substances are invisible, ...
Please note that I mean literally formed at the observed containment region, as if in a big vat, as opposed to being materials distributed throught the environment and unconstrained in their ability to be exposed to any freely moving chemical species. If the region in which the reactions to form tons of product material are localized over time, the near environment must come close to thermal equilibrium, which in this case is visibly hot for non-gases (solid, liquid, and plasma). As such, the effects should be visible in close proximity, if only indirectly. But, continuing with your statement:
We are in agreement. essentially. If the oxidation ongoing in one small region were sufficient to produce tons of material, it should be visible according to what you're saying, right? The gases participating in the reaction would be invisible, but I'd expect the nearby environment to be in near thermal equilibrium, making the effects of heat of oxidation directly visible. Remember, we're talking about a reservoir right inside the perimeter which has been spilling out.
In contrast, there's the idea that it formed elsewhere and flowed to the location observed and pooled there. I think that is the case when you say:
We are in agreement. essentially. If the oxidation ongoing in one small region were sufficient to produce tons of material, it should be visible according to what you're saying, right? The gases participating in the reaction would be invisible, but I'd expect the nearby environment to be in near thermal equilibrium, making the effects of heat of oxidation directly visible. Remember, we're talking about a reservoir right inside the perimeter which has been spilling out.
In contrast, there's the idea that it formed elsewhere and flowed to the location observed and pooled there. I think that is the case when you say:
...water tanks or metal filing cabinets would be especially susceptible, and thin steels.
Not dissimilar from racks of UPS batteries - which I could see, after sufficient heating of containing plastics, might be like tons of lead sheets on a barbecue rack. Forming somewhere else and pooling at the observed location at least weakly implies somewhat higher temps and more material than what is observed. Note the rapidity with which the material cools upon emerging from the building.
QUOTE
If you had damaged hollow steel structures you could literally form tons of Fe Cl3 in fires with tons of PVC.
See, I wasn't following your work closely enough to understand the potential quantites involved, from the standpoint of pooling. The other explanations I've heard are lead from batteries and aluminum skin from the fuselage, other aircraft components (magnesium, etc). These also seem capable of supplying the needed quantity.
Are you aware of any estimate of the minimum volume where the amount is similar to what I've calculated (for that specific area)?
QUOTE (->
| QUOTE |
| If you had damaged hollow steel structures you could literally form tons of Fe Cl3 in fires with tons of PVC. |
See, I wasn't following your work closely enough to understand the potential quantites involved, from the standpoint of pooling. The other explanations I've heard are lead from batteries and aluminum skin from the fuselage, other aircraft components (magnesium, etc). These also seem capable of supplying the needed quantity.
Are you aware of any estimate of the minimum volume where the amount is similar to what I've calculated (for that specific area)?
No it would not be noticed the gases that formed the substances are invisible, ...
Please note that I mean literally formed at the observed containment region, as if in a big vat, as opposed to being materials distributed throught the environment and unconstrained in their ability to be exposed to any freely moving chemical species. If the region in which the reactions to form tons of product material are localized over time, the near environment must come close to thermal equilibrium, which in this case is visibly hot for non-gases (solid, liquid, and plasma). As such, the effects should be visible in close proximity, if only indirectly. But, continuing with your statement:
QUOTE
... only when the oxidation started would you see any temperature or light increases.
We are in agreement. essentially. If the oxidation ongoing in one small region were sufficient to produce tons of material, it should be visible according to what you're saying, right? The gases participating in the reaction would be invisible, but I'd expect the nearby environment to be in near thermal equilibrium, making the effects of heat of oxidation directly visible. Remember, we're talking about a reservoir right inside the perimeter which has been spilling out.
In contrast, there's the idea that it formed elsewhere and flowed to the location observed and pooled there. I think that is the case when you say:
QUOTE (->
| QUOTE |
| ... only when the oxidation started would you see any temperature or light increases. |
We are in agreement. essentially. If the oxidation ongoing in one small region were sufficient to produce tons of material, it should be visible according to what you're saying, right? The gases participating in the reaction would be invisible, but I'd expect the nearby environment to be in near thermal equilibrium, making the effects of heat of oxidation directly visible. Remember, we're talking about a reservoir right inside the perimeter which has been spilling out.
In contrast, there's the idea that it formed elsewhere and flowed to the location observed and pooled there. I think that is the case when you say:
...water tanks or metal filing cabinets would be especially susceptible, and thin steels.
Not dissimilar from racks of UPS batteries - which I could see, after sufficient heating of containing plastics, might be like tons of lead sheets on a barbecue rack. Forming somewhere else and pooling at the observed location at least weakly implies somewhat higher temps and more material than what is observed. Note the rapidity with which the material cools upon emerging from the building.
I know I've seen one IR image of that area before. Even if this is just two garbage cans of eggnog at 100C, and even if raging fires had just passed through the area, it should show up on a sufficently sensitive thermal image.
QUOTE (Chainsaw,+Jan 18 2008, 07:11 PM)
I am trying to figure it out OneWhiteEye however it is complicated and dangerous, not easy to do, I do not know if anyone though of complicated Chloride metal reactions before, it is not something one would normally think of, just stumbled across it be accident.
It's cool that someone is doing it. You might be charmed, but exercise caution anyway!
Very interesting theory, what little I understand.
It's cool that someone is doing it. You might be charmed, but exercise caution anyway!
Very interesting theory, what little I understand.
QUOTE (OneWhiteEye+Jan 18 2008, 08:39 PM)
Again, thanks for the fascinating and informative replies, Chainsaw. It makes the work you're doing more relevant to me when I see such obvious outward manifestations.
See, I wasn't following your work closely enough to understand the potential quantites involved, from the standpoint of pooling. The other explanations I've heard are lead from batteries and aluminum skin from the fuselage, other aircraft components (magnesium, etc). These also seem capable of supplying the needed quantity.
Are you aware of any estimate of the minimum volume where the amount is similar to what I've calculated (for that specific area)?
Please note that I mean literally formed at the observed containment region, as if in a big vat, as opposed to being materials distributed throught the environment and unconstrained in their ability to be exposed to any freely moving chemical species. If the region in which the reactions to form tons of product material are localized over time, the near environment must come close to thermal equilibrium, which in this case is visibly hot for non-gases (solid, liquid, and plasma). As such, the effects should be visible in close proximity, if only indirectly. But, continuing with your statement:
We are in agreement. essentially. If the oxidation ongoing in one small region were sufficient to produce tons of material, it should be visible according to what you're saying, right? The gases participating in the reaction would be invisible, but I'd expect the nearby environment to be in near thermal equilibrium, making the effects of heat of oxidation directly visible. Remember, we're talking about a reservoir right inside the perimeter which has been spilling out.
In contrast, there's the idea that it formed elsewhere and flowed to the location observed and pooled there. I think that is the case when you say:
Not dissimilar from racks of UPS batteries - which I could see, after sufficient heating of containing plastics, might be like tons of lead sheets on a barbecue rack. Forming somewhere else and pooling at the observed location at least weakly implies somewhat higher temps and more material than what is observed. Note the rapidity with which the material cools upon emerging from the building.
OneWhiteEye also do not forget about lead Chloride it is almost identical to Ferrous chloride only the oxidation temperature is a lot less.
http://en.wikipedia.org/wiki/Lead(II)_chloride
What I am looking at is a mixture of chlorides that can form and then go though complex chemical reactions while flowing.
I have seen the phenomena only once and then while dogging chunks of concrete flying though the air.
Actually I was cowering behind a large oak tree after the first blast, of carbon monoxide.
I know it involved Aluminum, lead, and other metals with Iron Chloride which is passive when mixed with aluminum but oxidizes into Black iron oxide, producing high temperature, and if it oxidizes in the presence of aluminum a Exothermic thermite like reaction occurs.
It is just some very complex and unexpected Chemistry that I am not sure that I have the abillity to fully understand it.
See, I wasn't following your work closely enough to understand the potential quantites involved, from the standpoint of pooling. The other explanations I've heard are lead from batteries and aluminum skin from the fuselage, other aircraft components (magnesium, etc). These also seem capable of supplying the needed quantity.
Are you aware of any estimate of the minimum volume where the amount is similar to what I've calculated (for that specific area)?
Please note that I mean literally formed at the observed containment region, as if in a big vat, as opposed to being materials distributed throught the environment and unconstrained in their ability to be exposed to any freely moving chemical species. If the region in which the reactions to form tons of product material are localized over time, the near environment must come close to thermal equilibrium, which in this case is visibly hot for non-gases (solid, liquid, and plasma). As such, the effects should be visible in close proximity, if only indirectly. But, continuing with your statement:
We are in agreement. essentially. If the oxidation ongoing in one small region were sufficient to produce tons of material, it should be visible according to what you're saying, right? The gases participating in the reaction would be invisible, but I'd expect the nearby environment to be in near thermal equilibrium, making the effects of heat of oxidation directly visible. Remember, we're talking about a reservoir right inside the perimeter which has been spilling out.
In contrast, there's the idea that it formed elsewhere and flowed to the location observed and pooled there. I think that is the case when you say:
Not dissimilar from racks of UPS batteries - which I could see, after sufficient heating of containing plastics, might be like tons of lead sheets on a barbecue rack. Forming somewhere else and pooling at the observed location at least weakly implies somewhat higher temps and more material than what is observed. Note the rapidity with which the material cools upon emerging from the building.
OneWhiteEye also do not forget about lead Chloride it is almost identical to Ferrous chloride only the oxidation temperature is a lot less.
http://en.wikipedia.org/wiki/Lead(II)_chloride
What I am looking at is a mixture of chlorides that can form and then go though complex chemical reactions while flowing.
I have seen the phenomena only once and then while dogging chunks of concrete flying though the air.
Actually I was cowering behind a large oak tree after the first blast, of carbon monoxide.
I know it involved Aluminum, lead, and other metals with Iron Chloride which is passive when mixed with aluminum but oxidizes into Black iron oxide, producing high temperature, and if it oxidizes in the presence of aluminum a Exothermic thermite like reaction occurs.
It is just some very complex and unexpected Chemistry that I am not sure that I have the abillity to fully understand it.
QUOTE (Chainsaw+,Jan 18 2008, 09:39 PM)
OneWhiteEye also do not forget about lead Chloride it is almost identical to Ferrous chloride only the oxidation temperature is a lot less.
http://en.wikipedia.org/wiki/Lead(II)_chloride
What I am looking at is a mixture of chlorides that can form and then go though complex chemical reactions while flowing.
I have seen the phenomena only once and then while dogging chunks of concrete flying though the air.
Actually I was cowering behind a large oak tree after the first blast, of carbon monoxide.
I know it involved Aluminum, lead, and other metals with Iron Chloride which is passive when mixed with aluminum but oxidizes into Black iron oxide, producing high temperature, and if it oxidizes in the presence of aluminum a Exothermic thermite like reaction occurs.
It is just some very complex and unexpected Chemistry that I am not sure that I have the abillity to fully understand it.
Wow.
Heh heh. There's part of me, a small part, that envies your activities. The larger part prefers the armchair. Good luck.
I'm beginning to get the chemistry, a little, thanks. I confess that the quantity of chem I've taken is small, and unfortunately coincided with early mornings and raging hangovers.
It's hard for me to wave away the implied consequences of what I see, regardless of the actual temperature, quantity, or chemical composition of the substance in question. There is direct visual evidence of a reservoir present to disgorge material which, under minimum estimates for all parameters, suggest:
- The presence of high temperatures is confirmed directly, rather than inferred, for cases involving more than isolated metal fires.
- I think the above elevates whatever probability your theory already had by some amount, however small. As well others like lead, etc.
- There is potential for influence on structural integrity; however localized, the significance may be exaggerated by virtue of the specific location
- There are rather pressing obligations to understand and quantify the hazards associated with the phenomena if it does exist, and involves anything other than the introduction of a commercial airliner.
For example, if the source is lead from banks of power supplies, should that be a consideration in the future placement of such installations? What are the consequences for a 'normal' office fire with the emerging (in two or three decades past) office component assemblage today? I see evidence of a new migration path for fire that could span many floors downward in short order.
http://en.wikipedia.org/wiki/Lead(II)_chloride
What I am looking at is a mixture of chlorides that can form and then go though complex chemical reactions while flowing.
I have seen the phenomena only once and then while dogging chunks of concrete flying though the air.
Actually I was cowering behind a large oak tree after the first blast, of carbon monoxide.
I know it involved Aluminum, lead, and other metals with Iron Chloride which is passive when mixed with aluminum but oxidizes into Black iron oxide, producing high temperature, and if it oxidizes in the presence of aluminum a Exothermic thermite like reaction occurs.
It is just some very complex and unexpected Chemistry that I am not sure that I have the abillity to fully understand it.
Wow.
Heh heh. There's part of me, a small part, that envies your activities. The larger part prefers the armchair. Good luck.
I'm beginning to get the chemistry, a little, thanks. I confess that the quantity of chem I've taken is small, and unfortunately coincided with early mornings and raging hangovers.
It's hard for me to wave away the implied consequences of what I see, regardless of the actual temperature, quantity, or chemical composition of the substance in question. There is direct visual evidence of a reservoir present to disgorge material which, under minimum estimates for all parameters, suggest:
- The presence of high temperatures is confirmed directly, rather than inferred, for cases involving more than isolated metal fires.
- I think the above elevates whatever probability your theory already had by some amount, however small. As well others like lead, etc.
- There is potential for influence on structural integrity; however localized, the significance may be exaggerated by virtue of the specific location
- There are rather pressing obligations to understand and quantify the hazards associated with the phenomena if it does exist, and involves anything other than the introduction of a commercial airliner.
For example, if the source is lead from banks of power supplies, should that be a consideration in the future placement of such installations? What are the consequences for a 'normal' office fire with the emerging (in two or three decades past) office component assemblage today? I see evidence of a new migration path for fire that could span many floors downward in short order.
QUOTE (OneWhiteEye+Jan 18 2008, 09:58 PM)
Wow.
Heh heh. There's part of me, a small part, that envies your activities. The larger part prefers the armchair. Good luck.
I'm beginning to get the chemistry, a little, thanks. I confess that the quantity of chem I've taken is small, and unfortunately coincided with early mornings and raging hangovers.
It's hard for me to wave away the implied consequences of what I see, regardless of the actual temperature, quantity, or chemical composition of the substance in question. There is direct visual evidence of a reservoir present to disgorge material which, under minimum estimates for all parameters, suggest:
- The presence of high temperatures is confirmed directly, rather than inferred, for cases involving more than isolated metal fires.
- I think the above elevates whatever probability your theory already had by some amount, however small. As well others like lead, etc.
- There is potential for influence on structural integrity; however localized, the significance may be exaggerated by virtue of the specific location
- There are rather pressing obligations to understand and quantify the hazards associated with the phenomena if it does exist, and involves anything other than the introduction of a commercial airliner.
For example, if the source is lead from banks of power supplies, should that be a consideration in the future placement of such installations? What are the consequences for a 'normal' office fire with the emerging (in two or three decades past) office component assemblage today? I see evidence of a new migration path for fire that could span many floors downward in short order.
This file http://www.firetactics.com/NRC%203D.pdf
OneWhiteEye,
Will help you understand what I am looking into, it is all about trapped gasses, and the reaction of said gases chemically with metals.
The World Trade Center used mechanical purging of gases in fires, however the system was not working on Sept 11 2001 because of the planes impacts.
The towers had electrical conduits running under the floor pans on top of the trusses, perfect for the entrapment of gasses in fires.
I discovered this just by accident and Called Dr. Jones about it years ago just did not understand it then, since then I have been looking into it as a source to created the micro spheres, every other discovery is just accidental.
I have made thousands of these.
Because basically burning steel melts steel.
Burning oxidizing steel next to aluminum produces a thermite like reaction in the aluminum.
I was going to give up on this before I blew myself up, but some one who shall not be named talked me out of it.
The towers had what was called live decking, electrical passages over the trusses built into the floor pans, this is a crude drawing.
From what I have seen there needs to be more testing of live decking in fires because it seems to be a design flaw in fires.
Nist did not include Live decking or the right amount of plastics in the fire models that changes the entire fire dynamic.
I wish other would actually do some testing into this because the one thing I do know is I am not qualified to really do the tests.
Heh heh. There's part of me, a small part, that envies your activities. The larger part prefers the armchair. Good luck.
I'm beginning to get the chemistry, a little, thanks. I confess that the quantity of chem I've taken is small, and unfortunately coincided with early mornings and raging hangovers.
It's hard for me to wave away the implied consequences of what I see, regardless of the actual temperature, quantity, or chemical composition of the substance in question. There is direct visual evidence of a reservoir present to disgorge material which, under minimum estimates for all parameters, suggest:
- The presence of high temperatures is confirmed directly, rather than inferred, for cases involving more than isolated metal fires.
- I think the above elevates whatever probability your theory already had by some amount, however small. As well others like lead, etc.
- There is potential for influence on structural integrity; however localized, the significance may be exaggerated by virtue of the specific location
- There are rather pressing obligations to understand and quantify the hazards associated with the phenomena if it does exist, and involves anything other than the introduction of a commercial airliner.
For example, if the source is lead from banks of power supplies, should that be a consideration in the future placement of such installations? What are the consequences for a 'normal' office fire with the emerging (in two or three decades past) office component assemblage today? I see evidence of a new migration path for fire that could span many floors downward in short order.
This file http://www.firetactics.com/NRC%203D.pdf
OneWhiteEye,
Will help you understand what I am looking into, it is all about trapped gasses, and the reaction of said gases chemically with metals.
The World Trade Center used mechanical purging of gases in fires, however the system was not working on Sept 11 2001 because of the planes impacts.
The towers had electrical conduits running under the floor pans on top of the trusses, perfect for the entrapment of gasses in fires.
I discovered this just by accident and Called Dr. Jones about it years ago just did not understand it then, since then I have been looking into it as a source to created the micro spheres, every other discovery is just accidental.
I have made thousands of these.
Because basically burning steel melts steel.
Burning oxidizing steel next to aluminum produces a thermite like reaction in the aluminum.
I was going to give up on this before I blew myself up, but some one who shall not be named talked me out of it.
The towers had what was called live decking, electrical passages over the trusses built into the floor pans, this is a crude drawing.
From what I have seen there needs to be more testing of live decking in fires because it seems to be a design flaw in fires.
Nist did not include Live decking or the right amount of plastics in the fire models that changes the entire fire dynamic.
I wish other would actually do some testing into this because the one thing I do know is I am not qualified to really do the tests.
Thank you. Now I'm beginning to get some of the process and implications.
http://www.irinfo.org/Articles/article_9_11_2001.html
Haven't read the article yet, but this is the thermal imaging I'd heard about.
Haven't read the article yet, but this is the thermal imaging I'd heard about.
QUOTE (OneWhiteEye+Jan 18 2008, 11:26 PM)
http://www.irinfo.org/Articles/article_9_11_2001.html
Haven't read the article yet, but this is the thermal imaging I'd heard about.
That is interesting, thanks for posting it will have to read it all a little later, right now working on taxes.
Haven't read the article yet, but this is the thermal imaging I'd heard about.
That is interesting, thanks for posting it will have to read it all a little later, right now working on taxes.
QUOTE (OneWhiteEye+Jan 18 2008, 11:26 PM)
http://www.irinfo.org/Articles/article_9_11_2001.html
Haven't read the article yet, but this is the thermal imaging I'd heard about.
Hadn't seen that article before, interesting.
This South tower thermal image appears to have been taken at 9:18am, or 15min after impact. Is there something to be made of a comparison with the NIST fire model predictions? I don't know enough about thermal imaging, for example are those temperatures the air and smoke just outside the WTC facade, or do they reflect interior temperatures somewhat?
The thermal scale indicates temperatures < ~100°C if it is accurate.
Haven't read the article yet, but this is the thermal imaging I'd heard about.
Hadn't seen that article before, interesting.
This South tower thermal image appears to have been taken at 9:18am, or 15min after impact. Is there something to be made of a comparison with the NIST fire model predictions? I don't know enough about thermal imaging, for example are those temperatures the air and smoke just outside the WTC facade, or do they reflect interior temperatures somewhat?
The thermal scale indicates temperatures < ~100°C if it is accurate.
QUOTE (lozenge124+Jan 19 2008, 03:24 AM)
This South tower thermal image appears to have been taken at 9:18am, or 15min after impact. Is there something to be made of a comparison with the NIST fire model predictions?
I would think. Perhaps they even used the info to assist in modeling, I don't know.
I've no idea, either. Speculation: Mostly the temperature of the facade and the gases escaping. To the extent the interior is not masked by smoke and gas, I'd think it would register as well. FLIR, anyone?
I've no idea, either. Speculation: Mostly the temperature of the facade and the gases escaping. To the extent the interior is not masked by smoke and gas, I'd think it would register as well. FLIR, anyone?
The thermal scale indicates temperatures < ~100°C if it is accurate.
It is early, and it is the exterior. I wish it were much later.
I would think. Perhaps they even used the info to assist in modeling, I don't know.
QUOTE
I don't know enough about thermal imaging, for example are those temperatures the air and smoke just outside the WTC facade, or do they reflect interior temperatures somewhat?
I've no idea, either. Speculation: Mostly the temperature of the facade and the gases escaping. To the extent the interior is not masked by smoke and gas, I'd think it would register as well. FLIR, anyone?
QUOTE (->
| QUOTE |
| I don't know enough about thermal imaging, for example are those temperatures the air and smoke just outside the WTC facade, or do they reflect interior temperatures somewhat? |
I've no idea, either. Speculation: Mostly the temperature of the facade and the gases escaping. To the extent the interior is not masked by smoke and gas, I'd think it would register as well. FLIR, anyone?
The thermal scale indicates temperatures < ~100°C if it is accurate.
It is early, and it is the exterior. I wish it were much later.
QUOTE (OneWhiteEye+Jan 18 2008, 11:00 PM)
I would think. Perhaps they even used the info to assist in modeling, I don't know.
I've no idea, either. Speculation: Mostly the temperature of the facade and the gases escaping. To the extent the interior is not masked by smoke and gas, I'd think it would register as well. FLIR, anyone?
It is early, and it is the exterior. I wish it were much later.
OWE,
Have you read NIST NCSTAR 1-5A????????
You seem to be all about the photography and that Sub-Report is ALL ABOUT the VISUAL evidence and TIMELINE.
It discusses these IR pictures and what can (and can't) be derived from them.
http://wtc.nist.gov/oct05NCSTAR1-5index.htm
FLIR is the company name.
The pictures are in general over-saturated in the IR region (as one would expect, these are far hotter objects than one would typically use an IR camera to film)
The data is thus not very quantitative, you can tell that these red areas (or white in the false positives) are hot, but not how hot they are relative to the other areas.
From their angle they give little info about the interior, but there is some that is interesting, particularly if you read NIST's evaluation of these pictures.
The only input from the visual clues to the FDS was the window openings. As the FDS progressed through its simulation they removed windows (increasing ventilation) that corresponded to the missing windows in the photos.
Arthur
I've no idea, either. Speculation: Mostly the temperature of the facade and the gases escaping. To the extent the interior is not masked by smoke and gas, I'd think it would register as well. FLIR, anyone?
It is early, and it is the exterior. I wish it were much later.
OWE,
Have you read NIST NCSTAR 1-5A????????
You seem to be all about the photography and that Sub-Report is ALL ABOUT the VISUAL evidence and TIMELINE.
It discusses these IR pictures and what can (and can't) be derived from them.
http://wtc.nist.gov/oct05NCSTAR1-5index.htm
FLIR is the company name.
The pictures are in general over-saturated in the IR region (as one would expect, these are far hotter objects than one would typically use an IR camera to film)
The data is thus not very quantitative, you can tell that these red areas (or white in the false positives) are hot, but not how hot they are relative to the other areas.
From their angle they give little info about the interior, but there is some that is interesting, particularly if you read NIST's evaluation of these pictures.
The only input from the visual clues to the FDS was the window openings. As the FDS progressed through its simulation they removed windows (increasing ventilation) that corresponded to the missing windows in the photos.
Arthur
PhysOrg scientific forums are totally dedicated to science, physics, and technology. Besides topical forums such as nanotechnology, quantum physics, silicon and III-V technology, applied physics, materials, space and others, you can also join our news and publications discussions. We also provide an off-topic forum category. If you need specific help on a scientific problem or have a question related to physics or technology, visit the PhysOrg Forums. Here you’ll find experts from various fields online every day.
To quit out of "lo-fi" mode and return to the regular forums, please click here.