Return to index: [Subject] [Thread] [Date] [Author]


[Subject Prev][Subject Next][Thread Prev][Thread Next]
-----Original Message-----
From: Caldwell, Stan [mailto:scaldwell(--nospam--at)]
Sent: Thursday, June 14, 2001 10:38 AM
To: 'seaint(--nospam--at)'

There could also be a great design supported by wonderfully detailed and organized calculations, but the structure still failed...

Case in point:

I once saw a set of structural engineering calculations that could have been framed and put in a museum. They were gorgeous: Detailed sketches, every variable defined, every reference cited. Table of contents, index, the works.
It was when I was working at Brown & Root. The calculations were for a refinery pipe rack. And the design was fatally flawed.
I discovered to my dismay, that at B&R in the 1980s at least, no one knew or understood the principles behind the phenomenon of "lateral torsional buckling." Bear in mind that the types of structures they designed were essentially "bare steel frames." No concrete floors or steel decking to provide any sort of lateral resistance to the compression flanges of beams and girders.
All they thought you had to do was run a cross-beam between "stringers" and *voila*! You were now entitled to use L/2 for the "effective length" of the member for flexural design.
I actually had to "force" the civil/structural manager to allow me to explain my objections to this standard design practice. I had to get out my copy of Salmon and Johnson and get them to read the very cogent explanation contained therein. Even then, they were reluctant to admit that their designs were incorrect, because "we've been doing 'em this way for years, and there's never been a single problem."
But man, they sure made some pretty design notes.

William L. Polhemus, Jr., P.E.
Polhemus Engineering Company
Katy, Texas
Phone 281-492-2251
Fax 281-492-8203