Need a book? Engineering books recommendations...

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

Re: Story Drift: 1994 UBC vs. 1997 UBC

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Bill:

The short answer:
Your "flagpole" designed using the 1994 UBC would really deflect 1.5 
to 2 inches in the design event; the UBC was lying about the expected 
deflection.

The long answer:
The 3Rw/8 multiplier significantly underestimates the expected drift. 
This has long (probably for more than 30 years) been recognized by 
researchers, code writers, and many designers.  The UBC 
continued to indicate smaller displacements because a realistic 
estimation of the expected drift concerned code writers.  ATC 3, 
which is the source document for modern seismic design in the U.S. 
was published in 1978 and in it the expected drift is calculated 
as Cd times the drift from code-level forces.  The NEHRP Recommended 
Provisions always carried this same calculation method (based on Cd). 

UBC code writers have continued to drag their feet for two reasons: 
1) the big numbers are "scary" and 2) they generally don't believe 
that practicing engineers are capable of anything that isn't dirt 
simple.  Given increasing concerns about the actual performance of 
buildings in earthquakes (which can be quite sensitive to the imposed 
displacements), it was decided that the first reason was no longer 
compelling.  That's why the 1997 UBC finally reflects knowledge 
concerning displacements that was state of the art in the late 
60s and early 70s (not bad; only 30 years too late).  On the basis 
of the second reason, it was decided to "simplify" by using 0.7 R 
instead of Cd.  I suppose this "simplification" is justified for two 
reasons: 1) we would all somehow be baffled if we had to look up Cd 
in Table 16-N, and 2) this way the deflection calculation procedure 
(in the western U.S.) can change in the 1997 UBC and change again in 
the 2000 IBC where Cd values are used.

-Mike

*******************
> From:          "Bill Allen" <Bill(--nospam--at)AllenDesigns.com>
> Subject:       Story Drift: 1994 UBC vs. 1997 UBC
> Date:          Wed, 11 Aug 1999 06:08:27 -0700

> Can this be true?
> 
> "Flagpole" columns (i.e., inverted pendulums)
> h=8 ft.
> D=Delta
> Ds=Delta sub s (service)
> Dm=Delta sub m (maximum inelastic)
> 
> 1994 UBC:
> Ds < 0.005*h=0.48 in
> Dm=Ds*3*Rw/8=0.48 * 3 * 3 / 8 = 0.54 in.
> 
> 1997 UBC (1630.9.2)
> Dm=0.7 * R * Ds < 0.025 * h
> Ds < 0.025 * h / 0.7 * R = 0.025 * 8 * 12 / (0.7 * 2.2) = 1.56 in
> 
> Is it true that we can now design flagpoles (and steel frames for that
> matter) for an allowable drift of three times than before or have I missed
> something???
> 
> Regards,
> 
> Bill Allen, S.E.
> ALLEN DESIGNS
> Laguna Niguel, CA

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Michael Valley                                   E-mail: mtv(--nospam--at)skilling.com
Skilling Ward Magnusson Barkshire Inc.                  Tel:(206)292-1200
1301 Fifth Ave, #3200,  Seattle  WA 98101-2699          Fax:        -1201