Boise Airport Hangar 7

[TigerGuy]

https://www.cityofboise.org/news/fi...veral-injuries-following-a-building-collapse/

Hangar under construction collapsed killing 3 and injuring more.

 

[bones206]

You can see the heads of the bolts that popped off laying nearby in the photo. They probably failed in tension, then the base plate sheared along the flange and zippered down the web. Agree with JAE that it could easily be a result of the collapse, but I wouldn't rule it out as an initiating failure either.

 

[driftLimiter]

The crane looks rather small to cause so much damage. The open frame of the building would catch a lot of wind.

The unbraced frame is substantially weaker than the complete and properly braced flanges and columns.

The top flange of the beams needs to be braced by purlins, metal roof deck, and horizontal bracing elements. Without those elements in place the frames are subject to instability failure.

A relatively small point load in the center of a PEMB frame that is not braced could cause enough instability to initiate a collapse.

If this is in fact the initiating failure of the collapse, it should be a wake up call for the PEMB industry.

I've seen many columns with the same baseplate details. 1000s of them are installed per year without failing under frame self weight. There simply has to be something more going on than just baseplate failure. These frames have some serious joint rigidity at the eave, under self weight there is no way that the joint wants to rotate by 30 degrees.

 

[dik]

In particular with PEMBs... it's not called a Systems Building for nothing.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[SwinnyGG]

Judging what failed first here is difficult with the limited information we have, but high wind and multiple cranes is not a good recipe. especially depending on how the cranes are sized. Based on the typical extreme level of stiffness at the joint between PEMP columns and girders, I would be looking at a hoisting mistake, high wind, or lack of temp bracing as likely contributors. Failure of the steel or foundations is unlikely if everything is being hoisted correctly and the temp bracing is being installed correctly.

Not a rigging guy but using multiple smaller cranes to lift a large load (if that's really the case?) sounds like a terrible idea.

For very long steel it's the better way to do it... You can't control the rotation of a 100' long, 8' deep steel girder with a tag line, and if it gets out of control it can easily rotate into the boom and take the crane out.

I will say, however, that adding a third crane is generally not great planning; that's a last resort move. The third crane makes it very difficult to control what percentage of the load is being taken up by each cable.

 

[hokie66]

Like JAE implied, the failure was due to instability of the structure. The base plate failure is a red herring.

 

[bones206]

I've seen many columns with the same baseplate details. 1000s of them are installed per year without failing under frame self weight. There simply has to be something more going on than just baseplate failure. These frames have some serious joint rigidity at the eave, under self weight there is no way that the joint wants to rotate by 30 degrees.

I guess my point is that if it was base plate initiated failure, that should set off some alarm bells. We all know PEMB base plates are skimpy compared to conventional steel, and I've seen a few base plate failures in PEMB's.

 

[JStructsteel]

It seems the foundation engineer is off the hook.....but i am sure will be sued anyway.

 

[Sym P. le]

The possibility of base plate failure is very much in play even if only from a last line of defense perspective. Note that the purlin work was incomplete at the time of failure.

Consider the diagonal wind load experienced that day. For at least five hours prior to the incident, the winds blew diagonally across the main span at speeds up to 26 or 32 mph.

What failure mechanism resulted in the wave pattern buckling of the main span? Did the wind force cause bowing in the main span, pulling in at the perimeter supports causing uplift and thus bolt failure from the outside in?

Just asking for a friend. [URL unfurl="true"]https://youtu.be/c3cnJXRLsBI?t=1370[/url]

 

[SWComposites]

Them purlins look very wimpy compared to the deep web beams.

Lateral deflection induced a buckling mode perhaps.

 

[hokie66]

They took a big risk and lost. At least one bay should have been completed, including roof, side wall, purlins, and bottom flange bracing. Then the balance. Those deep and thin frame sections are subject to all sorts of buckling.

 

[enginesrus]

So is this on the construction team or the designers? My initial thought was gee they can't even design these things correctly any more. What would have happened with the weight of a roof and a snow load on it?
Is there supposed to be a brace that ties the members that meet at the center or peak, or just flanges? Pretty scary deal.

 

[Sym P. le]

This incident would fall under partially completed construction. I imagine it's upon the builder to be knowledgeable about appropriate precautions in the event of wind loading. In tilt-up construction, until the roof deck was fully installed, our eyes were always on the weather to ensure there were no loose ends if the wind came up.

As far as the erection process, it would seem to be standard procedure to erect pairs of main elements with full cross bracing before falling back and filling in the balance of the structure. A single one of these main beams would have no lateral stiffness whatsoever. The completed structure would act as a diaphragm with minimal moments transferred to the base.

The OSHA study would likely consider what if any indications precluded the collapse and go from there as to whether the construction company left their employees in harms way.

The only anomaly I see in the collapse pattern is in the north west pair where there is some northward tipping and rotation about the corner base. The news team dutifully applied chirons to help us in our understanding of their news.

 

[Aesur]

It seems the foundation engineer is off the hook.....but i am sure will be sued anyway.

For every PEMB I have seen, the foundation engineer is the project engineer of record and from my understanding technically is somewhat responsible for the building engineering as well, even though they are only reviewing the deferred submittal. As the EOR you are responsible for all structural aspects of the projects. In fact some jurisdictions are starting to make it clearer that the foundation engineer is the EOR and responsible for the building aspects as well. That is why PEMB foundation designers are idiots for charging so little to design foundations only in my opinion.

 

[Brian Malone]

Big D Builders of Meridian has erected a similar structure. The Skywest hangar completed in 2015:

https://www.bigdbuilders.com/single-post/2015/02/18/skywest-hanger-boise-id

 

[human909]

Those beams are extraordinarily slender beams. What twist restraint if any was in place for them. Without suitable twist restrain they strike me as incredibly unstable... Sure there was purlins on the top flange, but that isn't enough to prevent twist of a member that slender.

The outcome I believe proves the point.

 

[Brian Malone]

https://www.idahostatesman.com/news/local/community/boise/article284954822.html

Access to this article is limited by number of visits and may be thru a paywall.

 

[JedClampett]

This kind of shakes my confidence. I've always said that if you assume it pinned, it will act pinned (due to the rigidity provided elsewhere in the frame).
Another thing: If you hit the PEMB manufacturer with a hammer, asking for more bolts, they'd still only provide 4 bolts with minimum spacing, or less, at the base connection. Why did they provide eight bolts here? Maybe this has something to do with the problem.
If there's any good to come out of this, due to the loss of life, OSHA will perform a analysis, available to the public. And you can bet someone will change their procedures due to it.

 

[Sym P. le]

What twist restraint if any was in place for them.

There was plenty of cross bracing employed during this operation, both across the main span, and on the verticals (heavier at the corners).

This photo from the Big D site shows how the main span is assembled on the ground as two halves (in pairs) and then lifted into place where the final connections are made. Although four cranes are utilized, only two are used for each half. Oddly, with the open web trusses, I don't see any cross bracing but typically, I would expect to see it. The purlins are easier to install on the ground but don't provide the same construction assist. The verticals are also assembled in pairs with a top and bottom half.

Of particular note is the heavy duty anchor system used in the example. This may point to a design liability though the moment design principles of the two main span systems differ.

 

[human909]

There was plenty of cross bracing employed during this operation, both across the main span, and on the verticals (heavier at the corners).

True. And I did miss that and I do agree there is "plenty" and it seems significant. However in its present state to provide twist restraint it is reliant on the axial compression strength/stiffness of the top purlins. Which is SFA given they have no rod bracing/bridging or a roof over it. Which might be getting closer to the root cause.

From my perspective those deep girders are still have pretty significant in strength in pure bending, and for them to fail (wind and/or crane impact) they had severe LTB issues.

When you are relying on tiny purlins for LTB stability on a span of that size you are asking for trouble. They seem so slender that LTB failure under dead load seems plausible. All they needed was a little trigger to produce the cascading LTB failure.

 

[3DDave]

Even with that bracing in place it just ties the two beams together to double the resistance to side movement; double the beams is double the wind load. The bracing prevents twist, but this doesn't look like it started with twist.

What it required was horizontal x-bracing to take what would be a shear load with the roofing installed. Without it they left one mode of buckling unconstrained.