Hyatt Regency disaster - 1981, Kansas City

[AELLC]

Example of PAY ATTENTION to basic statics - just wanted to bring this to attention of younger engineers.

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

I believe the engineer who first specified the original detail (which did seem to be safe) neglected to realize that it was impossible to fabricate an all-thread rod of that length, and expect someone to spend who knows how much time threading the nuts up to where they are shown to clamp the boxed C-sections.

When the detail was revised to the more constructible version version shown to the right, the force on the boxed C-sections was multiplied, and the failure mode was by bending the C-flanges and punching thru.

Basic statics. I mean really basic.

The definition of a structural engineer: overdesign by a factor of 1.999, instead of the usual 2.

 

[paddingtongreen]

Your sig line reminds me, we used to define a structural engineer as someone who multiplies 2 by 2, on a slide rule, gets 3.99 and rounds it off to 4 for convenience.



Michael.
"Science adjusts its views based on what's observed. Faith is the denial of observation so that belief can be preserved." ~ Tim Minchin

 

[ajh1]

In my opinion the primary fallacy with the connection was that the two C-sections should have been placed back to back with just enough room between them for the hanging rod. This is classic pipe support design and puts the load where the maximum load carrying capability would exist in the double channel. Alteratively, there should have been a stiffener plate installed between the channels next to the rod location to provide stiffness to the box section. Either approach would have been helpful regardless of whether the original design or the revised fabrication method was used.

 

[DamsInc]

I'm currently taking a construction law/documents class. The prof uses this failure as an example of confusion with "who does what" with regards to steel detailing and shop drawings.

The way it was explained to us was that the steel fabricator had a junior guy do a crappy design without it being properly checked, assuming that the EOR would check it thoroughly. The EOR assumed the fabricator was the expert in connection design and didn't run any calcs on it. Then when it failed the legal question was who was responsible for the design of the connections?

The legal principle: Risk should be assigned to whomever is best able to manage it.

The Canadian Construction Documents Committee has clarified this in their stipulated price contract (CCDC 2). The steel erector is the person with the expertise in connection design, they are the person responsible for the structural design of the connections, even though the consultant reviews the drawings.

"3.10.6 The Consultant's review is for conformity to the design concept and for general arrangement only."

"3.10.10: "The Consultant's review shall not relieve the Contractor of responsibility for errors or omissions in the Shop Drawings or for meeting all requirements of the Contract Documents"

 

[dik]

Make for some interesting case law... I think the EOR is the person best able to manage it.

Dik

 

[AELLC]

The way I hears it, a junior engineer did the original detail, contractor complained that it was impossible to build, and showed the revised detail on shop drawing, and jr engineer blindly approved it w/o thinking twice.

The definition of a structural engineer: overdesign by a factor of 1.999, instead of the usual 2.

 

[Bagman2524]

I've seen ajh1 design above used often. I don't get the use of flange to flange C sections as a support beam. Running hangers two stories doesn't seem like a good idea either.

“Faith is taking the first step even when you don’t see the whole staircase.” MLK

 

[AELLC]

the rods 2 stories - $100 says the Archi insisted upon it for good looks. Also the stupid C-box section.

The definition of a structural engineer: overdesign by a factor of 1.999, instead of the usual 2.

 

[paddingtongreen]

As a result of this tragedy, there was a move by a number of states to write into law a requirement that the EOR be involved, authoritatively, all the way through construction. This made me very happy, I was fed up with seeing my designs butchered in the field and having to buy off on them because "It will kill the schedule to re-do it".

The states backed down in the face of some fierce lobbying.

Michael.
"Science adjusts its views based on what's observed. Faith is the denial of observation so that belief can be preserved." ~ Tim Minchin

 

[CELinOttawa]

paddingtongreen: I have no idea when it happened here, being well before my time as an Engineer, but in Ontario the law states that all work conducted by an engineer must be reviewed by a P.Eng.

It is not EOR will full authority on the project all the way through, but it is close. Most authorities (cities, counties, etc) will require a final letter from the engineer. Makes a very big difference and helps both get the job done competently, and the engineer paid.