The 9/11 Forum

[Enik]

I was following the topic WTC7 and the girder walk-off between column 79 and 44 on the jref forum. Even though I haven't been that interested in WTC 7, since I recently reacquired ABAQUS, I thought I would go ahead and model the area in question.

The first two images are full scale models of the columns and girders between column 79 and 44. One shows the floor removed.




Larger Image 1 Larger Image 2

I rested the girder between columns 79 and 44 on the support plates, free to move. I placed a 50 psf load on the floor at room temperature, heated to 500F (260 C), and 1000F (538 C).
The results are shown in the next three images which show the support plate on the built up Column 79:







Larger Image 3 Larger Image 4 Larger Image 5

Very little movement occurred at room temperature. At 500 F, the girder moved about 3.4 inches and at 1000 F, the girder moved 6.7 inches. The last two images show the overall model with and without the floor.




Larger Image 6 Larger Image 7

[SanderO]

Brilliant work! Thanks!

[kawika]

I was part of the team that put the following videos together to refute NIST's walk off theory.

Part 1--
http://www.youtube.com/watch?v=qGe0E9cjUbI&feature=plcp

Part 2--
http://www.youtube.com/watch?v=AvRKZO5o_dA&feature=plcp

Part 3--
http://www.youtube.com/watch?v=-Zsp0UcgMzs&feature=plcp

Thank you for these models. I do have several areas of concern with them.

1. The seat looks to be smaller than the flange. NIST claimed it was only 11" wide. The flange on a W33 x 130 is 11.5" wide. The seat was actually 12" wide. This is shown on Drawing 1091, Materials List.

2. The column 79 model does not show the 2" x 26" side plates that extend past the column face. There is 1.8" of overlap. These are shown on Drawing 9114, Section AA.

3. The girder does not appear to expand towards the face of col 79 as NIST said in Vol 1. NIST says the girder began to bear upon the face of col 79 and this expansion was why the bolts sheared.

4. The seat plate is not 90 degrees to the column face, but offset 4+ degrees, following the girder.

5. Most importantly, the girder had 3/4" x 5.5" x 18" stiffener plates welded into the end. NIST failed to see these on Drawing 9114 or failed to consider them in their model. These stiffeners would prevent the flange of the girder from deflecting once the girder moved upon the seat past the web.

6. There is no point using a 1000C temperature as the beams pushing on the girder would begin to sag at 600C. I think I understand why you used it as a hypothetical-- to show that even at that extreme the girder would not have been pushed far enough to drop off the seat. With the stiffeners, you need to go the full width of the flange plus 1/4", or 11.75".

NIST said the beams only reached 600C, the girder 500C. NIST said the girder pushed 5.5" (exactly half of the 11" wide seat they claimed) because they used the wrong formula. What is the true expansion at 600C? 4.67", not enough to push even past the web.

But see item #2 above. The girder can only move about 4" West before contacting the 2" side plate.

7. There is a 2" thick x 14" tall x 18.9" wide under-seat support not shown on the column. This would come into play if the girder could drop off the 1" seat.

Thank you for showing these. Please contact me if you would like to contribute models to Part 4, where we talk about the fires (or lack of) that acted upon these beams and girder.

[OneWhiteEye]

Tony, you got your ears on? Care to weigh in?

[T_Szamboti]

Tony, you got your ears on? Care to weigh in?

Yes.

I think Enik did a nice job on the model and it is certainly a big leg up on anything we have had up until now, but as Kawika pointed out it isn't fully reflective of the actual structural situation regarding

- the seat width, length, and orienation. The seat was 12" wide x 1" thick and was 17.25" from the center of the column to the bolts on the girder and extended at least another 1" beyond the bolts. The seat was also oriented parallel to the girder.

- the 2" x 26" build-up plates on column 79.

- the 2" thick x 18.875" wide x 14" tall plate welded to the build-up plates which supported the seat.

- the 3/4" thick x 5.465" wide x 18" tall stiffeners on each side of the girder 3/4" from its column
79 end.

The drawings which would give the details needed have been released and they are Frankel Steel drawings 9114 and 1091. They can be downloaded here http://ae911truth.org/en/news-section/4 ... quest.html

Enik, if you could add these features and rerun the analysis your model would be very valuable. If you don't want to download the full file of drawings and just want the two pertinent ones, I can e-mail them to you if you give me your e-mail address. Please PM me if you are interested.

[Enik]

To kawika and Tony
If either one of you can sketch up something with dimensions and upload it, I can update the model and rerun sometime next week.

Kawika
I used 1000 F not 1000 C.

[T_Szamboti]

To kawika and Tony
If either one of you can sketch up something with dimensions and upload it, I can update the model and rerun sometime next week.

Kawika
I used 1000 F not 1000 C.

Enik, I am not very practiced at uploading things but I can send them via e-mail.

My e-mail address is tonyszamboti@comcast.net. If you want send me a message, so I can send you the drawings which give the dimensions.

I also wanted to mention that the dead loads are 75 psf and live loads 50 psf giving a total floor load of 125 psf per a table in the NIST Report. Usually the live load is reduced by 50% so you should apply a 100 psf floor load.

[Enik]

I think I’ve included all of the missing information this time. The first image shows the dimensions of the seat that I used in the analysis (Click on any picture for a larger version).



The second image shows the setup including the 2” x 26” side plates, offset by about 2” from Column 79, along with the 2” thick plate that sits under the seat.



Note, I did this analysis using shell planar models to speed up the calculations and it appears that is what the NIST did as well. The analysis was performed on the mid-surface so understand this when you look at the 3D rendering of the model since it appears that parts are cutting into each other.



In the next two images, I show the dimensions of the W33X130 flange along with the ¾” thick stiffeners.





Here is an overview of the setup without the floor.





I did not take into account the temperature rise of the W33X130 girder in my original analysis. I figured all of the pushing was done by the W24X55 beams. I have since included the 500 C (932 F) temperature in the W33x130 girder and 600C (1,112 F) temperature in the W24X55 beams. I had to take the model up in steps, as shown below (all results are displacements in the X direction):

























I also ran the analysis without the stiffeners; it moved the girder out slightly more.



I did not include yielding in this analysis so I guess it is possible the lower flange on the W30x133 girder could give way and drop.

[Enik]

I have started a complete WTC 7 model (full scale of course) similar to the WTC 1 models I made earlier. Below are a couple of screen shots showing a typical floor.





However, at this time, I do not know how long it will take me to complete.

[SnowCrash]

Great work. But.. the girder isn't just resting unfixed is it? I don't get what makes it 'walk'. (If it's fixed)

And what about the (ostensible?) temperature discrepancy between the girder and the seat? Is that realistic?

[SanderO]

This is superb! I am waiting to see the structure depicted from street level to floor 8.

Side note... I was talking to an electrical contractor this afternoon who was called in after '93 bombing and had worked frequently in the towers. He commented that the floors were very bouncy on B7 and thought it was constructed very cheaply.

[T_Szamboti]

Enik,

I noticed you used 100 psf for the floor load and added in the temperature for the girder. That's good.

However, the seat needs to be parallel to the girder with the girder centered on it at room temperature, as that is the way it was oriented in reality. The seat was 12 inches wide and without a parallel orientation it only acts like about a 10 inch wide seat.

You don't say if you adjusted the modulus of elasticity for elevated temperature with the AISC retention data or if you used the average CTE between room temperature and 1,112 F of 7.01 x 10e-06 in/in/deg. F for the beams.

The seat orientation, modulus at elevated temperature, and average CTE will make a significant difference and it would be worth rerunning the analysis with them in there.

The flange without the stiffeners would have yielded shortly after the web was off the seat, as it could not take the loads on the girder and would have failed in bending. With the stiffeners the flange would not have yielded. Ron Brookman and I both did the calculations for the welds on the stiffeners and they carry the load and have more than a 2 to 1 safety factor with the girder having been pushed 9.5 inches west.

[T_Szamboti]

Great work. But.. the girder isn't just resting unfixed is it? I don't get what makes it 'walk'. (If it's fixed)

And what about the (ostensible?) temperature discrepancy between the girder and the seat? Is that realistic?

The bolted connections would have been broken by the expansion and the girder was alleged to not have shear studs, so it was just resting unfixed per NIST. Enik is trying to see if there would have been enough expansion to push it off under those circumstances.

The girder was at 500 degrees C, the beams at 600 degrees C, and the columns at 300 degrees C per the NIST fire simulation. The seat would have initially had the temperature of the column and would have been draining some heat from the girder and getting a little hotter itself, but the dimensional and strength changes would have been relatively insignificant to the issue here.

[SanderO]

The 100psf is the NY Code.. the PANY got a reduction to 58 PSF for the twin towers... it's stated in the correspondence. Perhaps there was load reduction at B7.

Note... a friend of mine from sailing.... I discovered is an electrical contractor who worked in B7 and the twins over the years. He remarked yesterday that he thought the floors at B7 were extremely *bouncy*... likely means an 1/360 or even 1/180 deflection. He also said the construction at B7 were very shoddy and cheap... whatever that means. He's not an engineer so his comments don't carry much weight about structure... but he was in and worked on various aspects of the electrical system including some of the main feeds to the tower.. which he said were 4 13.8KV and floors 4 and 6 contained the switch closets, relays and step down transformers (not on topic).

[SnowCrash]

From AE911Truth's two hour slide show: (click to enlarge)



Seems like the girder is resting inside a socket. Doesn't seem to me like it would just move sideways, should it expand, one would think it would bend and deform like railway tracks without expansion slots.

he was in and worked on various aspects of the electrical system including some of the main feeds to the tower.. which he said were 4 13.8KV and floors 4 and 6 contained the switch closets, relays and step down transformers (not on topic).

Interesting though. May deserve its own thread. Did one of those explode when that scene near the payphone was filmed?