Need a book? Engineering books recommendations...

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

Section 1633.2.6 & 1612.4

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Dear Mark,

I'm glad you are raising interesting and important questions for discussion 
on this server.  Concerning your recent post (you knew this was coming), in 
which you wrote:

"----ITEM 3)----
If rho is greater than 1, don't forget that you can "take it back
out" before checking collectors, if your collector force is based
on E, which already includes the rho factor."

I would Like to make an observation.  If one is designing a structure with 
the idea that it will have a non-brittle yield mode, this approach can lead 
to trouble.  Say that two EBFs are designed, one using a  rho of 1.0 (one), 
and the other using a rho of 1.5 (one point five).  The first EBF is 
assured of having a link-yield mechanism; the second (given typical link 
over-strengths and the difficulty of finding sections that match demands 
perfectly) may experience collector failure prior to link yielding.

Of course, the omega factor for an EBF is large, and the calculated 
collector force is still substantially above the EBF demand.  My point, 
however, is that the reliability for the two cases is different.  If rho 
were used for the design of collectors, the  omega factor, and the 
calculated demands, for collectors in redundant structures could be 

I'm not saying that you are misreading the code.  However, I think any 
discussion of how to use the code can be informed by some attention to the 
meaning of the terms.  In this case, it is complicated by the fact that 
omega-sub-naught is used for two purposes.  The first is as it is used in 
this example: a factor to ensure that force-controlled elements remain 
elastic while a ductile mechanism is allowed to form.  In certain other 
contexts, it is used as a penalty factor for designs that have performed 
poorly in the past (slender braces in OCBFs, or brace lines with poor 
lateral-force distribution); I wonder if collectors in those cases should 
be factored by rho*[omega-squared].  (I'm only kidding, of course; doing a 
non-linear dynamic analysis using a suite of time-histories should 
demonstrate collector adequacy).

I welcome other opinions on this subject.


Rafael Sabelli

DASSE Design Inc.
33 New Montgomery, Suite 850
San Francisco, California 94105
Phone:	415-243-8400
Fax:	415-243-9165