ATLANTA partial parking deck collapse

[msucog]

i haven't seen this posted yet. rather unfortunate for the contractor (same one as botanical garden walkway collapse)...they're actually "better" than most contractors i run across.

http://www.wsbtv.com/news/19892813/detail.html

 

[Ron]

Yeah, Hardin's a pretty good contractor...certainly above average. Noticed that a precast soffit beam popped out. Could be connection deterioration, design, construction...who knows at this point.

 

[hokie66]

I know it's a speculative generalisation, but I have never felt comfortable in these precast parking structures. There is not enough robustness in the connections or redundancy in the system for my liking. Give me a steel framed or cast in place structure anytime.

 

[Ron]

hokie66...I agree. I've wondered if the traffic vibration would push a beam off its corbel/haunch. Not much holding these things together.

 

[dik]

I've done many Precast, Steel and Reinforced Concrete parkades and nary a problem, yet! The first one I did using 12" Hollowcore about 30 years back just had an couple of floors added.

Dik

 

[GalileoG]

I'm working on one right now.....

I'll be tossing and turning tonight.

Clansman

If a builder has built a house for a man and has not made his work sound, and the house which he has built has fallen down and so caused the death of the householder, that builder shall be put to death." Code of Hammurabi, c.2040 B.C.

 

[hokie66]

I'm not saying they can't be built so they behave satisfactorily, and I am confident that you fellows will do your best, tossing and turning included, to make sure they are competent structures. But the first thing I think of when I hear of a parking structure collapse is "not another precast house of cards".

 

[ishvaaag]

I am even less confident than hokie66 in this kind of structure. Here it has been used not only in parking but multilevel supermarkets of the "we are the cheapest" type full of people and, locally, racks to the ceiling. Yet till now, and even in spite of too small seats by any *** subjective*** standards, I am not aware of any notorious failure.

The same can't be said of bridges with standard to small seats, some have collapsed upon rotation of the foundation and column, or upon impact of a truck.

 

[Teguci]

The lack of redundancy in a precast parking structure won't come in to play until a major seismic event or explosion.

The seating of the haunches are more than adequate for normal loading and the welded connections with the chord/diaphragm steel will hold the pieces together under loading.

When I think of a precast double-tee I would consider it better than an open web steel joist floor system which are used all the time for supermarkets.

Problems with precasst DT systems - Welded steel connections need to be maintained. Flange reinforcing is WWR which may degrade without any visual sign. Diaphragms may span great distances without lateral support (250 ft between shearwalls).

 

[msucog]

here's my suggestion to help everyone sleep a little better at night..."completely/strictly" follow the special inspections program (with a good/qualified testing firm on board essentially full time if possible) to help document the construction and so that you/designer are able to say it was built like it was designed. i have not seen the same urgency to follow SI on precast as for steel (which is still lacking by the way). from my independent third party eyes, some of the precast field work is "scary" to say the least(basically iron worker mentality except with precast--you know what i mean if you've been around construction long enough)...all i'm going to say at this point.

been to that parking deck before...nothing out of the ordinary that i noticed (but wasn't looking for anything). and apparently pretty vacant of people at the time of collapse so maybe not traffic vibration. the parking deck only few years old.

and i really do hate to see the hardin folks wrapped up in this one...but it should be an eye opener to everyone as to what the rest of the construction really looks like.....(don't fool yourself--again from these independent eyes)

 

[ishvaaag]

Just an anecdote,

once I was asked to start design for some maintenance buildings holding some trucks and in whatever the way, I came with some solution that had precast L and inverted T beams; I did preliminary checks and determined the sections had to be truly massive and of 60 MPa characteristic strength or near 9 ksi. I scrapped the idea.

10 years passed and my job then (last year's) went through a design of akin loads and structural system, then I found there was at town a glistening (wholly new facilities) firm making those gross beams at such 60 MPa strength and seemed to have enough success to stay working.

Simply, that one doesn't want to do something is not warranty of that this won't be done by another. Nor a warrant of that one is right on using it or the contrary.

 

[Mike1144]

Did it seem odd to anyone else where that beam landed on the grass? It looked like it was yanked straight out. It didn't look like it just fell after the floor gave way.

 

[hokie66]

Mike,
Not knowing anything about the structure inside the spandrel beam, I would speculate that the spandrel "ejection" was initiated by either a failure of the spandrel to column connection in torsion, or by a bearing/shear failure of beams bearing on the spandrel, which would have resulted in a big lateral force as the beams fell.

 

[GalileoG]

Spandrel beam takes a lot of torsion if it has a ledge as opposed to being pocketed. As someone mentioned, the spandrel to column connection can take significant tension and can overturn if that connection fails.

Clansman

If a builder has built a house for a man and has not made his work sound, and the house which he has built has fallen down and so caused the death of the householder, that builder shall be put to death." Code of Hammurabi, c.2040 B.C.

 

[jorton]

But the beam appears to have overturned in a direction opposite of the torsion. Even then, it looks like the beam moved 30' to 40' laterally. How? It's not a small piece of concrete.

 

[hokie66]

jorton,
Why do you say that? It ended up on the outside face, but could have come to rest either way.

Looks like from the attached video that there wasn't much tying the floors to the spandrel, and the connection at the spandrel to column connection wasn't enough to prevent the beam from twisting out. As the floor slipped past the spandrel, boom, the spandrel was ejected.

http://www.wsbtv.com/video/19919886/index.html

 

[jorton]

hokie66,
You're right. It could have come to rest either way. I just have trouble envisioning the beam traveling with enough backspin to land as it did. Perhaps it only made a 1/2 turn in the air and then rolled the other 1/2. OR, maybe it took a bounce off the building. I understand that there is some surveillance video from within the deck. Hopefully it will be released soon.

 

[Teguci]

There are two things I'd be looking for -
1. Any rusting at the double-tee to L-beam connections. If the welded connection (pretopped) or coil rod connection (field topped) between the elements failed, then the torsion of the ledge connection would be transferred to the column coil bolts and consequent failure.
2. Failure of the coil bolt connections at the columns. If the column pockets for the coil bolts do not allow sufficient lateral deflections of the coil bolts then every time the spandrel is loaded a shearing action will act on the coil bolts due to beam end rotation.

 

[vandede427]

CIP makes me sleep easier too over Precast. But, well said by MSUCOG about sleeping better: special inspections.

 

[NomLaser]

Update to this story if you are still following...

http://www.wsbtv.com/news/19915945/detail.html