The 9/11 Forum

T_Szamboti wrote:Oh, I looked carefully enough.

Doubt it.

... I am not sure if you fully appreciate what you are claiming though.

Possibly not.

The connections could not be stronger than the beams themselves.

I agree that would be most unlikely.

NCSTAR1--6D, Tables 4--27 and 4--29. Then find the wall load data for floor 98.

O.K., here's the WTC 7 velocity vs. time plot I mentioned last night. It's based on OWE's data (posted a while ago) for two roofline locations - one near the center and the other near the NW corner. The two sets of measurements were quite similar so I averaged them to get a bit of data smoothing. The three (or possibly four?) strange jolts are very reproducible so, while I believe they are real, I have no idea what's causing them. Interestingly they involve short sharp periods of acceleration, followed by more or less equal and opposite deceleration.

Curious indeed .....

Dr. G wrote:The three (or possibly four?) strange jolts are very reproducible so, while I believe they are real, I have no idea what's causing them. Interestingly they involve short sharp periods of acceleration, followed by more or less equal and opposite deceleration.

Curious indeed .....

Perhaps these compression block 'waves' have something to do with it:



I'd think so. Caution is advised on taking differentials on measurements from a very localized region. I'm working out the per-frame equation of the fitted line for that part of the roofline and the adjacent vertical corner so as to find the intersection, plot the NW corner motion and, as much as possible, and mitigate error from this source.

OneWhiteEye wrote:Caution is advised on taking differentials on measurements from a very localized region.

Very good advice.

Dr. G wrote:O.K., here's the WTC 7 velocity vs. time plot I mentioned last night. It's based on OWE's data (posted a while ago) for two roofline locations - one near the center and the other near the NW corner. The two sets of measurements were quite similar so I averaged them to get a bit of data smoothing. The three (or possibly four?) strange jolts are very reproducible so, while I believe they are real, I have no idea what's causing them. Interestingly they involve short sharp periods of acceleration, followed by more or less equal and opposite deceleration.

Curious indeed .....

It is interesting that the first abrupt negative slope change is at approximately 2.25 seconds, which is the same as the proven freefall timeframe. I have to wonder if this isn't an artifact of some sort, since others (Chandler, NIST, etc.) have shown the upper block was in freefall at that time and it is the second abrupt negative slope change, near 3.2 seconds into the fall, which corresponds with the end of the 2.25 second freefall.

Perhaps some of the WTC 7 jolts were induced in a similar way to box cars getting jostled back and forth on a speeding freight train, ...... they show short bursts of acceleration and deceleration but overall the "group" velocity remains fairly constant!

If this is the case the period of the jostles would be about 1 second, ...... hence a jolt at approximately 1.3, 2.3, 3.3 and 4.3 seconds!

A velocity burst before impact makes no sense.

Any ideas?

Perhaps some of the WTC 7 jolts were induced in a similar way to box cars getting jostled back and forth on a speeding freight train, ...... they show short bursts of acceleration and deceleration but overall the "group" velocity remains fairly constant!

If this is the case the period of the jostles would be about 1 second, ...... hence a jolt at approximately 1.3, 2.3, 3.3 and 4.3 seconds!

In that case I would look for periodic motion. There is a smaller sinusoidal wave pattern in the plot.
I still couldn't see how this can explain the abrupt velocity bursts seen just before a possible impact.

Dr. G, frame 334 and 335 are duplicates, and there's an unnatural jump between frames 380 and 381. This is essentially the interval between your biggest jolts, and they're located in about the right place, so I'm pretty sure that's the cause of those. Sorry I didn't tell you before now, I just discovered it. I'd been suspicious of this when the other curves showed a flat spot and I meant to check it out, but I didn't get around to it until now.

The other little bumps are more than accounted for by the waves I show above. There's simply no way to reliably see any jolts in this video.

I can't say this video copy is worthless because it was good enough to replicate the horizontal wobble NIST plotted, but it is of limited value for further drop analysis. I got to this point with WTC1 - tired of crappy internet copies - and bought the Sauret DVD, problem solved. It may not be so easy with CBS footage, although I assume Naudet's video can be had for a price.

Hmm, Sauret, Naudet... XauXet

Wait wait wait...

Sorry in case that a lot of time has been spent on that smearogram, but that was a quick and dirty dump.

No worries, einsteen. It's the WTC7 CBS video from NIST that has the duplications, not the CD smear.

But that CBS copy came from you, so be aware! I don't know if it's the source or my dump at this point. The Naudet 50 fps does not have the problem. Here is a measure of image difference from one frame to the next on the CBS video:



The downward spikes are duplicates, the upward some sort of adjustment tweak. By comparison, the 'clean' Naudet video:



Running this test is now the first thing I'll do on any video. I don't know the conditions of the Balzac CD videos. This seems to be a pretty common thing (resampling or whatever) and it vexes me. I want some good videos!!

Ok, that's great. I quickly scanned the thread. Before commenting on jolts I'll re-read it in-depth...

OWE:

Thanks for clearing that up! It was quite troubling to me that the spikes appeared to suggest a short burst of acceleration preceeded each jolt.

I was so puzzled by this I even lost some sleep over it .....

I hope you sleep well now... The first thing I was thinking about was incredible high peak forces. kind of delta functions...

Einsteen:

Rust Never Sleeps ..........