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

Re: ASD vs. LRFD

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Bill Polhemus wrote:
> > -----Original Message-----
> > From: Christopher Wright [mailto:chrisw(--nospam--at)]
> > Sent: Wednesday, June 16, 1999 10:20 AM
> > To: SEAOC Newsletter
> > Subject: Re: ASD vs. LRFD
> >
> >
> > The reason they're ASD now and likely to remain so is because service
> > loads govern.
> This continues to be an amusing facet of the argument.
> I don't remember EVER being taught (and I was taught in ASD) to pay careful
> attention to service loads (meaning, in this case, deflection criteria) in
> my instruction with design codes. We were too busy figuring out all the
> myriad ways stresses could be computed.
> This newfound preoccupation with service loads as a governing criteria
> stems, I suppose, from the assumption that LRFD methods yield lighter
> sections that must be checked for serviceability.
> But the fact is that serviceability is one of the "limit state" criteria
> that the LRFD-based design codes I'm familiar with require to be checked as
> a matter of course (as opposed to viz. the AISC code which throws it in
> there for secondary consideration).
> If service loads (meaning "deflection") govern, then so be it, and you will
> know it if you follow LRFD-based design code criteria. In fact, you will
> know pretty much everything you need to know about the behavior of the
> structure.
> Which is more than can be said for the ASD design codes.
Should I say, Hurrah for diversity in engineering education?  When I
went to school (early-mid 70's) we were taught to understand the
deflected shape 1st, as a reality check.  This also continued in Grad
School at Berkeley.  By starting w/ deflections we would both cover
servicability requirements AND check the analysis for major errors.  The
value of this approach was reinforced by the Hancock Tower (Boston)
curtain wall problems which occurred concurrently.

Whether or not LRFD or ASD, I recommend always beginning with
deflections and satisfying those requirements, then checking stresses
(or strengths).  

One More Thought: In seismic analysis/design, the Code pseudo-force
method obscures the fact that we are really designing for
deflection/rotation limitation and maintenance of continuity under
maximum deflections, not forces/stresses per-se.

Perhaps the discussion should not be ASD vs. LRFD, but displacement vs.
force based analysis and design.  Any thoughts out there?

Laurence Oeth, P.E.