Need a book? Engineering books recommendations...

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

Re: asd vs lrfd

[Subject Prev][Subject Next][Thread Prev][Thread Next]
At 02:31 PM 10/15/97 -0400, A. Roger Turk, P.E.(Structural) wrote:
>Load factors and their accompanying resistance factors are used to
>provide a *factor of safety*, the same reason that allowable stresses
>are somewhat lower than yield or ultimate stress.

I guess we had better get very specific with our vocabulary lest we
lose our real-world points in academic specificity.  You are correct,
strictly speaking, that load factors don't predict *ultimate* loads.
A better definition (which is just a gnat hair away from saying the "u" 
word) might be as defined in my old concrete book:

   "In the strength design method (formerly called ultimate strength method),
   the service loads are increased sufficiently by factors to obtain the load
   at which failure is considered 'imminent'"
                  - "Reinforce Concrete Design" Salmon & Wang, Fifth Ed.

The load factors for concrete were not developed in exactly the same manner as
those in LRFD, but some of the intentions were obviously similar as PCI Notes on
ACI 318-95 states:

        "...the probability of overload on an understrength structure is
        about 1 in 100,000.  Load factors were derived to achieve this
        probability of overload"

But, to equate WSD factors of safety with load factors oversimplifies
things. Safety factors
used in WSD lack the thoughtfulness of load factors in some important ways: 
                1)      Differing degrees of uncertainty in loads can't be
handled
                        by a single "factor of safety".  As such, WSD results in
                        a a lot of variability in the ratio of strength to
service load.
                        This is obviously not a good thing with regard to
designing for
                        life-safety in strucutes subject to unpredictable
use and loading.
                2)      Creep and shrinkage effects can't be handled easily
by elastic
                        stress calculations.
                3)      Stress isn't proportional to strain at concrete
crushing strength,
                        so the SAFETY is really UNKNOWN when one simply uses
a percentage
                        of f'c as an "allowable stress".

The strength design method is a ***better model of reality*** than working
stress
design, hands down.  Concrete doesn't know or even care when its reached
it's allowable stress,
and the true "safety factor" in WSD is unknown.  WSD has a place in design
without a doubt,
but it is obvious that strength design should be the standard to better
protect life-safety.

Regards,
Joe McCormick