Need a book? Engineering books recommendations...

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

Re: Why LRFD vs ASD (was: We're Not Getting Older, We're Getting DUMBER)

[Subject Prev][Subject Next][Thread Prev][Thread Next]
> From: Christopher Wright <chrisw(--nospam--at)>

> On Jan 26, 2006, at 11:02 AM, Paul Ransom wrote:
> > more consistent "factor of safety" in ALL design checks (e.g. ASD is
> > not consistent for beam vs. connection
> That's only half an answer. What have been the effects? Fewer bodies?
> Lower costs? Safer construction.

I think that if you are looking for benefits, you have to ensure that
you are comparing apples to apples. Comparing designs for ASD 9th and
LRFD 12th is not realistic. 

> And I guess my experience with connections has been that connection 
> analysis is far less accurate than even analysis of plastic hinge 
> formation or instability. Arguably the consequences of connection 
> failure seem to be more of a safety threat.

Connections are a special issue. We reduce the connection capacity by
1/3 in either case, so we consider the accuracy and consequences,
already. In either approach, bolts/welds are based on ultimate, so
there's nothing there.   

> > However, the "law" is interpreted to permit a designer to meet ASD 9th
> > minimum, in some cases, to the detriment of public safety.
> Now we're down to it. Does the 9th edition truly represent a safety 
> issue? If so are all buildings designed to the 9th edition present a 
> significant hazard?

Significant ...? Strong word. However, that's my take on the issue.

Consider if the ASME standards had been split into two different
approaches when the major changes were made some years back. One was
left to rot (but remained popular due to commercial/regulatory inertia -
it costs money to change and it is supported by legal acceptance so
there is no incentive) but the other was upgraded and supported by
developing research and failure response, such as the current ASME.
That's essentially what has been happening between ASD 9th and ASD/LRFD
13th. If the ASD 9th had been maintained in parallel, we would not be
having this discussion.

Now, to go a step farther, let's say that one tried to apply the "old"
ASME to new materials and processes. Where the "new" ASME considers
these explicitly, the practitioner may be extrapolating the clauses of
the "old" ASME and missing known failure modes. On a projected life
cycle basis, which design would be consdered to be less expensive?

The issues may not be "significant" today, but what about after the next
standard cycle?

> I'm with you on the development issue: Codes should be improved as 
> significant issues come up. But I daresay that ASD Code development 
> only slowed because LRFD was going to be the Next Big Thing. It isn't
> of course. Limit analysis was written into all the old steel codes and
> into the nuclear codes back in the 70's. it's been around for dogs 
> years. And damn useful for special cases like impact. Makes me wonder
> why the change came all of a sudden.

Why ASD or LRFD ... I'm going to cop out on that one. I have detailed
what I know. A lot of people have been convinced in a lot of countries.

Having a level of consistent statistical risk across all analysis
points, in all materials and construction forms, MUST lead to economic
efficiencies. In the aggregate, those efficiencies can show up in
regulatory cost, constructed cost, insurance premiums, litigation
awards, etc. Not much of which can be directly correlated to the cost (#
of bolts, size of welds, thickness of plate) of the connection that I
just designed. 

There's a cost to maintaining a dual track. That, more than anything
else, is why Canada orphaned the steel WSD (the plan was to maintain
parallel for some period of time before dropping it). Our building codes
still permit the use of WSD design load cases (mostly for foundations,
now, but that may disappear) but the steel standard has explicitly
removed the equivalence exemption because it was being abused.

The next time that you use software that has dual standards or code
checks for the same regulatory region, think about the chain of people
that have to support the software, write dual standards, review dual
standards for code acceptance, issue permits with knowledge of dual
standards, etc.

I have seen the transition to limit states analysis confused with:
1) switching to metric
2) switching to "higher efficiency" materials (e.g. A36 to A992)
3) less expensive
4) more expensive
5) academic hijacking
6) more work
7) seismic driven (e.g. does not really apply in low seismic regions)
8) plastic analysis
9) not relevent to certain applications
There are probably more that I don't remember.

Disclosure: I work with ASD, LRFD, LSD.

R. Paul Ransom, P. Eng.
Burlington, Ontario, Canada
<mailto:ado26(--nospam--at)> <>

******* ****** ******* ******** ******* ******* ******* ***
*   Read list FAQ at:
*   This email was sent to you via Structural Engineers 
*   Association of Southern California (SEAOSC) server. To 
*   subscribe (no fee) or UnSubscribe, please go to:
*   Questions to seaint-ad(--nospam--at) Remember, any email you 
*   send to the list is public domain and may be re-posted 
*   without your permission. Make sure you visit our web 
*   site at: 
******* ****** ****** ****** ******* ****** ****** ********