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

Re: More "furrin code" stuff

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Do US jurisdictions accept projects using Canada's LS design method?

Thor A Tandy P.Eng, MCSCE
Victoria BC
e-mail: vicpeng(--nospam--at)
----- Original Message -----
Sent: Friday, May 05, 2000 12:24 PM
Subject: More "furrin code" stuff

Again, I don't know if this ever went out or got lost with the list server
problems. Here it is, perhaps for the second time, just in case. And I
promise not to speak on the topic again.

Peter Higgins, SE

If you're interested, the best work I have ever seen on the subject of
comparative steel codes is "Effective Length and Notional Load Approaches
for Assessing Frame Stability: Implications for American Steel Design"
written by the Task Committee on Effective Length of the ASCE Structural
Institute. There are also several papers in the December, 1997 edition of
the ASCE Structural Journal. Just remember, this was written by American
engineers who had no interest in performing a hatchet job on LRFD. You can
draw your own conclusions about the viability of LRFD after reading them.

Canadian limit states design (CSA S16.1) is one of many codes which use the
"equivalent imperfection" approach. It shares a lot with the Eurocodes and
UK/Australian/NZ standards.  Only Japan and the USA continue to use the
"effective length" approach for steel design. Effective lengths are old
technology. They have their uses, but the rest of the world moved on to
better things a long time ago. I suspect Japan will leave the fold soon,
leaving us as the only dinosaur.

If you're willing to make the transition, CSA S16.1 is probably the best
around. It is everything you asked for: Less work, faster, more accurate,
saves steel, and easier to apply to unusual structures. I doubt if anyone
who has used it would ever revert to LRFD. That being said, there is an ASD
version of CSA S16 which is still widely used by Canadian engineers (myself
among them). Both coexist with little debate on the topic. Indeed, it isn't
all that unusual to switch back and forth. ASD for mostly determinate
structures, and LS for the highly indeterminate ones.

I'm also tired of debating ASD vs. LRFD. It is a fools errand. However,
what troubles me is that the debate assumes LRFD is a viable replacement
for ASD. It isn't. LRFD is in bad need of a successor. The real problem is
that AISC isn't working on one.

They both should be replaced. And the sooner the better. Then I'll switch.

Peter Higgins, SE