WTC7 and the girder walk-off between column 79 and 44

[Christopher7]

Deleted.

 

[pgimeno]

The stiffner plates on each side of the girder web, and the fact that even if the web had moved the maximum amount (which it couldnt) it would still be supported by the seat 'pf' . You wont find the stiffner plates in NISTs report though, you have to go to the drawings to see them.
My original problem was with you saying that 'What's wrong with the concept of a displaced girder with just 0.5" support rocking off the seat?' , which i presume you now retract.

I don't see a reason to retract it.

The removal of the girder after 5.5" displacement is done only in the ANSYS simulation (chapter 11), while the rocking is mentioned in section 8.8, which starts with this paragraph:

A finite element analysis of the northeast corner floor system was conducted to evaluate its response to elevated temperatures and to confirm which failure modes needed to be accounted for in the 16-story ANSYS model. A finite element model of the northeast corner was developed using the LS-DYNA software that included the design details described in the previous section such as shear studs on the beams and seat connections at the girder ends and exterior ends of the beams.
​

I'd say that the removal of the girder after 5.5" displacement in the ANSYS model can be seen as an approximation, as the rocking was not simulated in ANSYS.

It seems to me that NIST contemplated many possible failure scenarios and picked one for ANSYS.

But, since you're contending this, do you have a better theory? Or are you just pointing your finger and saying "Newton's physics is wrong and we can't use it" because it allows speeds greater than light's?

 

[Christopher7]

You are able to show that the stiffeners would be sufficient to prevent this at the temperatures expected?

To a support plate at the same temperatures that we expect the girder to be right? So you can show that the dynamic load of the girder dropping while it and the 2 inch wide support plate are at several hundred deg. would be insufficient to prevent girder failure?

Interesting point.

Here are the temperatures after 4 hrs of heating case B
The girder is in the 300^oC to 350^oC range.

These temperatures supposedly did this damage:

So to answer your question, The stiffeners would have all their strength at 350^oC and would do the job they were designed to do.

 

[Christopher7]

I don't see a reason to retract it.

The removal of the girder after 5.5" displacement is done only in the ANSYS simulation (chapter 11), while the rocking is mentioned in section 8.8, which starts with this paragraph:

A finite element analysis of the northeast corner floor system was conducted to evaluate its response to elevated temperatures and to confirm which failure modes needed to be accounted for in the 16-story ANSYS model. A finite element model of the northeast corner was developed using the LS-DYNA software that included the design details described in the previous section such as shear studs on the beams and seat connections at the girder ends and exterior ends of the beams.
​

I'd say that the removal of the girder after 5.5" displacement in the ANSYS model can be seen as an approximation, as the rocking was not simulated in ANSYS.

It seems to me that NIST contemplated many possible failure scenarios and picked one for ANSYS.

But, since you're contending this, do you have a better theory? Or are you just pointing your finger and saying "Newton's physics is wrong and we can't use it" because it allows speeds greater than light's?

This was the LS-DYNA test model, not the final ANSYS model and "The girder and beam temperatures were assumed to be 500 °C and 600 °C, respectively"

 

[NoahFence]

Question....

All this talk about rocking, and temperatures, and failures of girders on seats, etc....


How does explosive demolition fit into this? Are we forgetting there was none, and therefore the NIST explanation, while potentially flawed, is still the most accurate one?

I'm under the assumption that a term called "walk-off" is a bit too subtle a phenomenon to be attached to something "explosive".

 

[Christopher7]

Question....

All this talk about rocking, and temperatures, and failures of girders on seats, etc....


How does explosive demolition fit into this? Are we forgetting there was none, and therefore the NIST explanation, while potentially flawed, is still the most accurate one?

I'm under the assumption that a term called "walk-off" is a bit too subtle a phenomenon to be attached to something "explosive".

The NIST report is fatally flawed, it does NOT explain the collapse so other possibilities must be considered. Denial based on personal incredulity and lack of knowledge is no substitute for the laws of physics.

 

[NoahFence]

The NIST report is fatally flawed, it does NOT explain the collapse so other possibilities must be considered. Denial based on personal incredulity and lack of knowledge is no substitute for the laws of physics.

It does explain the collapse.

It doesn't explain it to your exacting specifications.

Neither do any of your "theories" by the way.

Care to actually answer the question, kiddo?

How does "walk-off" happen in conjunction with "explosive" demolition?

 

[tfk]

Not at all, the reason that we never took into account the expansion of the girder is that if it was left hard up to the column face (which it would have been) this goes against NISTs walk off theory. These videos have deliberatley moved the variables in favour of nist. However, having expanded and went into compression between 79 and 44, there is a need for a different explanation, and so now we have the rock off theory. Which one do you think happened, walk of rock?

What do you think happens if it expands up against col 79 & 44, then buckles since the ends are constrained, and then the fire moves on, and the girder cools down?

 

[tfk]

So to answer your question, The stiffeners would have all their strength at 350^oC and would do the job they were designed to do.

Doesn't matter one iota how much the stiffener plates do their job.

The seat plate is the weak link.

If the end of the girder moves about 1" north or about 6" west, the seat plate is going to shear, and the girder is going to drop.

 

[Jackanory]

Awww, it's nice that you missed me. Been busy with other things. The spalling thing is interesting, and I have a few papers on that here, but generally, i feel that the misrepresentation of the drawings on the part of NIST is just being ignored by most of you here. I do welcome an open and frank discussion of that, and will keep an eye on the thread to see if that happens.

I mentioned it because it could possibly be in the mix as a contributing factor.

I don't think that the drawings are being ignored at all. In fact, I think they are being given far too much attention in this context, certainly by you and Chris7.

It is interesting to me that you and Chris7 rely heavily on what NIST 'say'.

'Walk off' vs 'Rock off'. Perhaps it would have pleased you more if they had said 'fell off', 'slid off', 'knocked off', 'snapped off', 'pushed off'. Ultimately they 'came off' or where 'dislodged' or where simply 'snapped'. That is for shure.

What is NISTS reasoning behind their study? What was there objective? Was it to understand the failures to enable the structural industry to make corrections and improve the safety of future builds...or was it to hide the dastardly deeds of a corrupt gubmint?

What is striking to me is that the build era's of WTC7, 1 & 2 was sound for that time and they could have remained standing for many, many years to come. Yes, building regs and changes to safety requirements needed addressing and updating when possible but that isn't just specific to buildings affected by 911.

Does 'walk off' or 'rock off' change the way that WTC7 should have been built? I doubt it. It has little bearing on the soundness of the construction for the purpose it was intended. Perhaps if it was built with hindsight and with better fire protection, better sprinkler system, secondary fire measures, 'bunker busting' thicker walls etc, but it wasn't.

You and Chris 7 seem to be working from the same sheet and both refer to sagging and expansion yet then determine a resulting action that can't possibly be the case.

Seems to me that in some instances you both are looking at pieces of steel that are in a 'test environment' and projecting those 'test' results as if they are what we see happening. Steel pieces that are sagging, expanding, retracting, buckling in a test environment without any load or other factors involved? Chris7 thinks that fancy graphs showing the spread of heat in the chain of steel is somehow conclusive. What does it matter that a piece of steel was at 300 degrees or 500 degrees when a shearing bolt 100 feckin metres away will go first?

I referred to the ACME principal of levitaion in an earlier post. From what I have read so far from both you and Chris7 is that this would seem to be your conclusion. The steel pieces left their respective connections and levitated?

 

[Christopher7]

Doesn't matter one iota how much the stiffener plates do their job.

The seat plate is the weak link.

No, the plate is 2" thick and 14" high. It is welded to the column side plates. That is what has been holding up the girder. You have not a clue what you are talking about.

If the end of the girder moves about 1" north or about 6" west, the seat plate is going to shear, and the girder is going to drop.

At 350^oC the girder would expanded about 2" and be pushing against the column.
http://img853.imageshack.us/img853/9758/col79stiffenerse.jpg

Walk off would have been prevented by:
1) The girder flanges would be stopped by the column side plate after about 4" of travel.
2) The stiffeners would prevent the bottom flange from folding for a little further but even when they failed the girder would land on the support plate.
3) Steel looses about half its strength at 600^oC and the beam would have to be heated to 738^oC to expand 6". By then it would have lost ~70-80% of its strength and sag, reducing the length.
http://img823.imageshack.us/img823/4001/beamexpansionspreadshee.jpg

 

[GlennB]

No, the plate is 2" thick and 14" high. It is welded to the column side plates. That is what has been holding up the girder. You have not a clue what you are talking about.

The seat plate (part PF) to which tfk refers was 1" thick. PG was 2" thick and therefore projected only 2" out from the column, PF is only able to bear the weight of the girder while the girder is directly over PG. Beyond 2" PF is on its own and dealing with increasing leverage with every millimetre. And it's extremely unlikely the girder would be square on the plate at this point, so 2" is generous.

3) Steel looses about half its strength at 600^oC and the beam would have to be heated to 738^oC to expand 6". By then it would have lost ~70-80% of its strength and sag, reducing the length.
http://img823.imageshack.us/img823/4001/beamexpansionspreadshee.jpg

Precisely what tfk and others are saying when they talk about the cooling phase pulling the girder off to the N when it had already sagged and lost relative length previously, even at 600°C

eta: I previously asked whether you had the sag calculations your "engineer friend" provided, whether this was theoretical "stand alone" sag or real-world loaded sag. Any luck with that?

 

[ozeco41]

...Precisely what tfk and others are saying when they talk about the cooling phase pulling the girder off to the N when it had already sagged and lost relative length previously, even at 600°C...

They are still trying to treat is as "one factor in isolation"- the mistake I pointed out some time back.

...eta: I previously asked whether you had the sag calculations your "engineer friend" provided, whether this was theoretical "stand alone" sag or real-world loaded sag. Any luck with that?

The "engineer friend" must not be aware of the full issue - or he is not very competent.

 

[GlennB]

The "engineer friend" must not be aware of the full issue - or he is not very competent.

I suspect he was primed to provide an answer to a one-off purely theoretical question - "How much will girder X sag when heated to 600C ?"
That is was being pushed in a complex manner by expanding beams while heavily and eccentrically loaded makes it a tougher question. Hey ... maybe a FEA is called for ?

 

[Christopher7]

The seat plate (part PF) to which tfk refers was 1" thick. PG was 2" thick and therefore projected only 2" out from the column, PF is only able to bear the weight of the girder while the girder is directly over PG. Beyond 2" PF is on its own and dealing with increasing leverage with every millimetre. And it's extremely unlikely the girder would be square on the plate at this point, so 2" is generous.

You are ignoring the expansion of the girder which would put more of the girder over the support plate.

Precisely what tfk and others are saying when they talk about the cooling phase pulling the girder off to the N when it had already sagged and lost relative length previously, even at 600°C

We are discussing the NIST hypotheses, the fact that it is impossible and they falsified the data trying to get it to work.

eta: I previously asked whether you had the sag calculations your "engineer friend" provided, whether this was theoretical "stand alone" sag or real-world loaded sag. Any luck with that?

Yes, as it turns out my instincts were correct.
Above 600^oC the sagging would shorten the beam more than it was expanding.
http://img707.imageshack.us/img707/6996/shortvtempscreenshot.jpg

At 600^oC the beam would expand 4.68"
http://img823.imageshack.us/img823/4001/beamexpansionspreadshee.jpg

but sagging would shorten it 0.416" for a net length increase of 4.264".
At ~640^oC the beam would expand 5" but loose 2" to sagging.

In other words, Thermal expansion of the beam could not push the girder 5".

 

[Jackanory]

Yes, as it turns out my instincts were correct.
Above 600^oC the sagging would shorten the beam more than it was expanding.
http://img707.imageshack.us/img707/6996/shortvtempscreenshot.jpg

At 600^oC the beam would expand 4.68"
http://img823.imageshack.us/img823/4001/beamexpansionspreadshee.jpg

but sagging would shorten it 0.416" for a net length increase of 4.264".
At ~640^oC the beam would expand 5" but loose 2" to sagging.

In other words, Thermal expansion of the beam could not push the girder 5".

Is your 'sagging' calc refective of load bearing onto it or is it an unladen 'test piece' scenario? You seem to be discussing a uniformal effect. Is that really so?

 

[GlennB]

The seat plate is the weak link.

No, the plate is 2" thick and 14" high. It is welded to the column side plates. That is what has been holding up the girder. You have not a clue what you are talking about.

The seat plate (part PF) to which tfk refers was 1" thick. PG was 2" thick ....

So tfk did have a clue what he was talking about and you didn't. Glad we cleared that up.

 

[GlennB]

Is your 'sagging' calc refective of load bearing onto it or is it an unladen 'test piece' scenario? You seem to be discussing a uniformal effect. Is that really so?

That's at least 3 times he's been asked that. Don't expect an answer.

 

[Jackanory]

That's at least 3 times he's been asked that. Don't expect an answer.

The link he posted refers to a load of 59625lb. Not sure where this comes from. Is it a pressure test load, the load of one floor or the load of numerous floors?

 

[GlennB]

The link he posted refers to a load of 59625lb. Not sure where this comes from. Is it a pressure test load, the load of one floor or the load of numerous floors?

It appears to refer to an individual beam and its load. The load on the girder would be multiples of that, but absent C7's friend's methodology and working we have no idea.