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

[GlennB]

An engineer I know calculated that the beam would sag about 7 1/2" at
1300^oF and that would shorten the beam a little less than 1/2".

I assume you mean girder here.

Was that free-standing or loaded? I'm sure your engineer friend's figures would be of interest here.

 

[ozeco41]

...Anyone think that it's impossible for the combination of the expansion of the W24 beams AND the motion of their opposite ends AND any lateral movement of columns 79 and/or 44 could not possibly move the southern end of the W33 column 6" west and/or 1" north??....

Which is much of the reason I have been telling C7 and gerry they are wrong to ignore all the factors they have been ignoring....

...does anyone care to guarantee that Col79 or Col44 will not move out of strict vertical alignment given the heat stresses which were occurring?

etc

etc though all the other variables...

 

[Sunstealer]

No i was thinking more of 0.27 - 0.3
Maybe you should just look it up in the ANSYS manual. Or any other FEA manual, or steel study site or.... you get the idea.

I am well aware of what Poisson's Ratio is and the values for common engineeering alloys that is why I posted a link to the totally non-intuitive negative Poisson Ratio (and made a poor quip)

In mechanics, Poisson's ratio is the ratio of strain in the x & y direction to the strain in the z direction as a result of the load in the z direction.

If I put a tensile load on something in the z direction, then I will get a strain in the z direction that is proportion to the tensile (or Young's) modulus.

z direction strain = (Z direction Load) / (area * tensile modulus)

But this z direction load will also produce strains along the x & y axes. (Pulling the material inward for tensile z-axis strains or pushing it outward for compressive z-axis strains.) Think of how a piece of rubber responds to being compressed (expanding laterally) or pulled (thinning down laterally).

X direction strain = Nu * (z direction strain)
Y direction strain = Nu * (z direction strain)

Nu = Poisson's ratio

The ratio of strain in the x & y axis to the strain in the z axis is Poisson's ratio.

It's typically about 0.3 for most metals, and about 0.5 for rubbers & elastomers.

tom

Cork has a Nu of near enough 0. That's why it's used in wine bottles. (it's a good exercise for people to work out why the properties of cork are good for that application.)

This subject always reminds me of this:

About 20 years ago, as a metallurgy undergraduate, I sat in a lecture with 10 colleagues who became completely bamboozled. Poisson's Ratio was being taught and all was going well when the lecturer started on about a negative Poisson Ratio which is totally non-intuitive and goes against everything you and I learnt in the real world about materials.

For those who don't have the background I'll really simplify - if you pull a material, say a rubber band, it gets longer and thinner. Yes? Everyone done that at some point? Right, well Possion's Ratio is very simply the amount of stretching divided by the amount of thinning. This ratio is positive for just about every material you and I have ever known.

Now imagine the scenario of a material whereby if you stretch it, it gets fatter! How the **** does that work? Think about your elastic band, imagine pulling it and the band getting fatter. In this scenario Poisson's Ratio becomes negative.

Well 20 years ago us undergrads just refused point blank to believe this. We spent the next 15 mins arguing with the professor and doing calculations on the board to show why a negative Poisson's Ratio was impossible. Exasperated he just said, "right I'll photocopy some papers" and left. It took him 15 mins and we were still arguing when he got back. The papers were handed out and silence descended. The prof then just gathered his gear and left some 20 minutes before the end of the lecture.

We had a free period after this lecture. 30 minutes after the lecture time had ended we were still sat in this room (rather than in the pub) going through the maths together and understanding the internal structures that can cause a material to get fatter when you stretch it, completely amazed that it was possible. Weird but it's true, there are materials that do just that.

http://en.wikipedia.org/wiki/Auxetic

 

[LSSBB]

Glenn,

I saw that you wrote this earlier. Yup, that's all you need to do.

Just as soon as the bolts are sheared, the column is unsupported. Now, in spite of what someone might think from looking at drawings, column 79 & column 44 have about 3 side loads, 3 vertical loads & 3 moments at every story. The chance of the resultant forces & moments being directly downwards and zero, respectively, are zero. There will be a resultant side load & moment on the now unsupported columns 79 & southward unsupported column 44.

If column 79 should buckle 6" eastward, then the girder will slide off. But the big danger is if it buckles about 1" southward OR if column 44 buckles about 1" northward. If either of these happens, then the massive girder will simply shear off the 1" thick, flimsy seat, and fall.

Similarly, if the girder (W33 x 130) thermally expands until it contacts the body of either or both of col 79 or 44, and then buckles under the axial load because it is constrained, it will fall off the support plate, shear the seat & fall.


Similarly, if the girder merely sags due to heat after shearing the bolts, it'll fall off the support plate, shear the seat & fall.

And if it buckles from thermal expansion, somehow does not fall off, but then cools off, then it will absolutely fall off the support plate, shear the seat & fall.

The 1" seat (without a gusset plate) is the incredibly weak link in this design. It ONLY manages to hold up that girder as long as the girder is over the support plate. If the central rib of the girder ever manages to come to the edge of the support plate, it's all over. The 1" seat will fold up like a cheap beach chair.

Anyone think that the support structure elements are moving around sizable distances in that building?

Anyone think that it's impossible for the combination of the expansion of the W24 beams AND the motion of their opposite ends AND any lateral movement of columns 79 and/or 44 could not possibly move the southern end of the W33 column 6" west and/or 1" north??

I fully expect that C7 or gerry will guarantee that this combined, relative motion is "impossible".

That's because they have no experience with large structures. Especially ones on fire.

Good work, Glenn.

tom

Good analysis Tom, that put the situation into perspective for me. Like all things Truther, they try to isolate everything to just one "anamoly" to build a case on it, leaving out the bigger picture and all the interconnections, whether mechanical in this situation, or operational for the whole day. Holocaust deniers do the same thing.

 

[OCaptain]

Absolutley none of your business.

You will not be able to publish your paper and get the right kind of attention for it - especially if you think you have grounds to challenge NIST's results - if you plan on remaining anonymous, so who and your other researchers are and what your training is IS everyone's business.

 

[Gamolon]

tfk (or anyone else),

When you have a moment, can you explain to me this calculation provided by gerrycan? He used this to show that the shear studs on the floor beams should have restrained the beam from thermally expanding.

CSA of W24 x 55 beam=15.986 sq. inches, the modulus of elasticity of steel is 29 million lbs./sq. inch, length is 52 feet x 12 inches/foot = 624 inches
force generated by beam for a 5 inch expansion is force = [5 inches x 15.986 sq. inches x 29,000,000 lbs./sq. inch] / 624 inches = 3,714,753 lbs, and for five beams would be 18,573,768 lbs buckling force = [Pi^2 x modulus of elasticity of steel x area moment of inertia] / [(effective length factor x unsupported length of beam)^2] = [9.8696 x 29,000,000 lbs/sq. inch x 29.1 in^4]/[(2 x 624 inches)^2] = 5,347 lbs, so the max force of the 5 beams is 26,738 lbs.30 x 3/4" diameter shear studs on the girder, so their total cross sectional area was Pi x R^2 x 30 = 13.25 sq. inches. The shear stress is just force/unit area and is thus 26,738 lbs. / 13.25 sq. inches = 2,018 psi
The shear studs would have had a tensile yield of 36,000 psi and a shear yield of 57.7% of that at 20,772 psi. The shear stress would only be 10% of what the shear studs could take.

Is this correct? I see NIST did calculations in their report that show that the shear studs would have failed. I asked gerrycan to address this as he said that NIST OMITTED the shear studs in their calculations, but he has not yet addressed this. What is the difference between the two calculations and where did gerrycan's engineer friend go wrong (if he/she did get it wrong).

Thanks in advance.

 

[tfk]

tfk (or anyone else),

When you have a moment, can you explain to me this calculation provided by gerrycan? He used this to show that the shear studs on the floor beams should have restrained the beam from thermally expanding.



Is this correct? I see NIST did calculations in their report that show that the shear studs would have failed. I asked gerrycan to address this as he said that NIST OMITTED the shear studs in their calculations, but he has not yet addressed this. What is the difference between the two calculations and where did gerrycan's engineer friend go wrong (if he/she did get it wrong).

Thanks in advance.

I'll take a look at it later today, but my first reaction is that I believe that the failure mode is fracturing of the concrete around the studs.

"Weakest link" & all that...


tom

 

[Jackanory]

DOUBLE BINGO!!!

That makes three of us who have identified several times that it is not a single factor situation involving a single member. Correct me if I've missed anyone.

Me....me......me.. in post* 300. I don't just take the piss ya know.

 

[ozeco41]

Me....me......me.. in post* 300. I don't just take the piss ya know.

That's four - any more?

 

[LSSBB]

That's four - any more?

Me too, if I wasn't clear on it before (probably not).

 

[000063]

Yeah, and you just earned the right to be wrong. lol @ 3/4"
Oh, and is being a snob considered a good thing where you are from? Over here it isn't something that people would readily admit to. Please elaborate on the 3/4" inch thing for me. Interesting.

So you are willing to scoff at a claim you don't actually understand?

 

[gerrycan]

So you are willing to scoff at a claim you don't actually understand?

I suppose you would have to define failure. NIST for example would appear to disagree with you, maybe their definition of failure is different to yours. Mine is too.

 

[Christopher7]

Just as soon as the bolts are sheared, the column is unsupported. Now, in spite of what someone might think from looking at drawings, column 79 & column 44 have about 3 side loads, 3 vertical loads & 3 moments at every story. The chance of the resultant forces & moments being directly downwards and zero, respectively, are zero. There will be a resultant side load & moment on the now unsupported columns 79 & southward unsupported column 44.

If column 79 should buckle 6" eastward, then the girder will slide off. But the big danger is if it buckles about 1" southward OR if column 44 buckles about 1" northward. If either of these happens, then the massive girder will simply shear off the 1" thick, flimsy seat, and fall.

Good analysis Tom, that put the situation into perspective for me. Like all things Truther, they try to isolate everything to just one "anamoly" to build a case on it, leaving out the bigger picture and all the interconnections, whether mechanical in this situation, or operational for the whole day. Holocaust deniers do the same thing.

That makes sense to you?

Column 79 was surrounded by concrete slabs. It could not move laterally.

The fire was heating the slab on the east along with the girders so if column 79 could move it would move to the west - but there was this darn slab preventing that.

 

[jaydeehess]

prove
that including the steel pan would SIGNIFICANTLY
change the heat getting to the beams!!!

It would change the heat getting to the CONCRETE. It has just less than half of its surface area in contact with it. BIG heatsink.

Has anyone claimed that expanding concrete would affect how the girder came off its seat? In fact it seems to me that if your contention were correct and that the concrete and the steel pan expanded significantly that would CONTRIBUTE to the movement of the beams and the pushing of the girder.

Seems to me that your point is that the floor pan/concrete would have conducted heat such that the beams would not heat up and expand.
I fail to see how any heat sinking ability (sink it to where is another matter) of the steel pan or concrete would prevent heating of the beams.

Thus I ask you to prove that the conductivity of the steel pan/concrete would affect the expansion of the beams.

Please explain that.
(If you already attempted that please tell me what post number as I may not be back on the forum until tomorrow)

 

[jaydeehess]

That makes sense to you?

Column 79 was surrounded by concrete slabs. It could not move laterally.

The fire was heating the slab on the east along with the girders so if column 79 could move it would move to the west - but there was this darn slab preventing that.

Did you just produce a reasoning to do away with beams and girders Chris?

Does that maker sense to you?

 

[jaydeehess]

DOUBLE BINGO!!!

That makes three of us who have identified several times that it is not a single factor situation involving a single member. Correct me if I've missed anyone.

So far C7 has denied and gerrycan has ignored the advice.

Are we referring to the fact that WTC 7 succumbed due to heat damage that allowed the assymmetrically placed beams to push a girder off its seat and fail a large area, long span floor which in turn removed most lateral support on one side of col 79 at that level, which itself was heated at that level and thus weakened, which caused col 79 to fail drawing down all floor pans along its length collapsing them all the way to the roof, which caused thousands of tons of heavy rubble to impact adjacent floor areas and fail girders which also caused other columns to fail in the core area which removed the northern main support for the cantilever trusses over the Con-ed structure which in turn saw most of the entire north half of the building to begin falling as the Con-ed structure was crushed by the mass of the 40 storeys above it, while the southern half lost its connection to the core and also failed, doing so quicker than it would have had not the southern face already suffered major loss of structural integrity.

The northern half of WTC 7 falling to the NE while the southern portion falling generraly to the south.

Is that what we are referring to?

Or are we referring to the fact that the heat expanded the beams causing both an expansion and a twisting of those beams?
Is that what we are referring to?

 

[Christopher7]

Did you just produce a reasoning to do away with beams and girders Chris?

Does that maker sense to you?

According to NIST the beams and girders were breaking shear studs bolts when they expanded but they said nothing about the columns moving laterally. They knew that the columns could not move laterally because they were surrounded by slabs. If you can't comprehend that then I don't know what else to say.

 

[LSSBB]

According to NIST the beams and girders were breaking shear studs bolts when they expanded but they said nothing about the columns moving laterally. They knew that the columns could not move laterally because they were surrounded by slabs. If you can't comprehend that then I don't know what else to say.

Can't move laterally?

Say, what happens if one floor has slabs move due to expansion, and the one above it doesn't?

 

[Christopher7]

Are we referring to the fact that WTC 7 succumbed due to heat damage that allowed the assymmetrically placed beams to push a girder off its seat and fail a large area, long span floor which in turn removed most lateral support on one side of col 79 at that level, which itself was heated at that level and thus weakened, which caused col 79 to fail

No, we are talking about the fact that the girder was NOT pushed off its seat because:
1) the seat was 12" wide not 11" wide and
2) the girder had stiffeners that would have prevented the bottom flange from folding and
3) even if the beams could push the girder completely off the seat it would only drop 1" to the support plate.

NIST said the girder expanded until it pressed against column 79 so the top and bottom flanges would hit the side plate after about 4" of travel.

Or are we referring to the fact that the heat expanded the beams causing both an expansion and a twisting of those beams?
Is that what we are referring to?

No, that was found to be a preliminary test and not the final cause. You got to the party late and you missed a lot.

 

[DGM]

No, we are talking about the fact that the girder was NOT pushed off its seat because:

But, once the bolts sheared. Isn't the dimension of the plate a moot point?

If not, explain.