From: Roger Turk <73527.1356(--nospam--at)COMPUSERVE.COM>
Date: Thu, 20 Jan 2000 11:18:12 -0500
Yes, the commentary to the SEAOC Blue Book and ATC-19, "Structural Response
Modification Factors," both describe the method in which the response factor,
R, was developed, and both indicate that all it did was take the K
coefficients between the extremes, 1.33 buildings and 0.67 buildings, and
manipulate the formula, modifying some other factors, to come up with
approximately the same answers for 1.33 buildings, 1.0 buildings, 0.80
buildings, and 0.67 buildings.
Rationality? Experience showed that using (ZIKCS)W, 0.67 buildings (ductile
steel and concrete) buildings performed well, 0.80 buildings (dual steel and
concrete) buildings had performed well, 1.33 buildings (shear wall
buildings, aka "box") had performed well, so the coefficients were about
right, so they shouldn't be changed. And they weren't changed, just put in
another form, consistent with the form in ATC-03, to give the same answer.
A. Roger Turk, P.E.(Structural)
Bill Sherman wrote:
Actually, Rw is 8/K. But the simple act of moving the coefficient to the
denominator improves the formula by identifying the coefficient more
directly with a structural property.
Following is a conversion from the "old UBC" formula to the "new" formula,
that I developed a number of years ago; the "new" formula has changed some
V=ZIKCSW can be converted to the new formula by replacing the coefficients
with the following:
Z=2.5*Z (e.g., 2.5*0.40=1.0 zone factor)
C=C/20 (e.g., CSmax=2.75/20=0.14)
K=8/R (e.g., 8/6=1.33 for shear walls)
(Based on the SEAOC Commentary.)
> -----Original Message-----
> From: Roger Turk [mailto:73527.1356(--nospam--at)compuserve.com]
> Sent: Wednesday, January 19, 2000 6:36 PM
> To: seaint(--nospam--at)seaint.org
> Subject: RE: Philosophy: Seismic Design Standards
> Bill Sherman wrote:
> >>Code formulas and coefficients should correspond to
> identifiable quantities.
> This is one area that I think improvements have been made in
> the seismic
> codes over the years, e.g. I couldn't relate to "K" in the
> formula ZIKSW,
> but I can relate "R" to ductility in the formula ZICW/R; and
> I can relate to
> Aa=0.40 better than "Zone 4". Also, commentaries and design
> examples help
> greatly when trying to interpret the real purpose of code language.<<
> Most AMAZING!
> R is simply 1/K! And was developed that way so that the
> results of the
> formula wouldn't change.
> "A rose is a rose ..."
> A. Roger Turk, P.E.(Structural)
> Tucson, Arizona