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

Diaphragms and rho Factors

[Subject Prev][Subject Next][Thread Prev][Thread Next]

Continuing the discussion with Charles and Ben:

I can not comment on why the Blue Book says what it says, except that it was
written by a somewhat different group of people than those who actually put the
1997 UBC together, and some years after the basic effort.

With regard to diaphragm strength vs. vertical element strength - I disagree
with Ben Youssefi.  BY seems to believe that if you make the diaphragms
stronger, this will deliver more load to the vertical frames (walls) and cause
failure of vertical elements.  He likens this to the strong column weak beam
requirements, that encourage yielding of beams rather than columns to protect
the columns.

I agree with BY that the structure has to yield somewhere and will yield in the
weakest element.  I do not agree that the diaphragm is the best place for this
to happen.  There are several reasons for this -

1) The R factors we use to proportion our systems are based on the deatiling and
qulaity of the vertical elements of the seismic system. The fact is diaphrgams
are often not detailed in a very ductile manner. As Charles notes, part of the
reasons for failure of parking garages in Northridge (not just the Cal State
garage, by the way) was that the diaphragms were poorly detailed (thin slabs,
big bar collectors, etc.)    We tend to use the same detailing of diaphragms
regardless of the vertical system selected.  Now if we pump up the strength of
the vertical elements, we will deliver more demand to the diaphragms and these
often nonductile elements will be asked to do more yielding, not a good

2) The code presumes that yielding will occur in the vertical elements, not the
diaphragms.  This is why we have all of the ductile detailing requirements for
vertical elements.  If you allow the diaphragms to yield prematurely (by making
the vertical elements stronger) than you are violating the basic principles of
"capacity design" as promulgated by the New Zealanders, in which you decide
where you want yielding to occur, then detail to accomodate this.

Finally, the strong column weak beam requirement is not just to protect columns
from yielding, though this is an important part.  It is also to prevent the
formation of story mechanisms.  If you have weak columns, then you will form
hinging in the columns in a single story and all of the inelastic deformation
will occur in that story, concentrating demands and forcing very large ductiltiy
demands in that story. The best designs distribute plasticity throughout the
structure, so that each element only has to accomodate modest inelastic

Anyway- there is certainly room for disagreement.

Where are your advocates?  You are your advocates.  The listserver is widely
read, though many prefer not to respond as it draws them into long discussions
such as this one.  The arguments put forward are listened to, and weighed with
other evidence in the process.