Need a book? Engineering books recommendations...

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

Re: Questions about Residential design and 1997 UBC

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Dennis -

Sorry it has taken me so long to throw in my 2 cents on this one.  Thanks for
the seminar update.

So here goes:

In a message dated 98-10-01 19:07:10 EDT, you write:

<< 1. Rw and residential construction - I have always taken the safe road
 determining the Rw value I use. Regardless of whether or not I use gypsum or
 stucco (which I don't) in a custom home with plywood shearwalls I assume the
 Rw to be 6. The code allows an Rw of 8 for a box system with plywood shear
 Am I being overly conservative and how many would stick with code rated Rw
 of 8?>>

We always use Rw=6, because we usually have an ordinary steel frame in the
building which drives the Rw=6 into the equation.  And if we don't have steel,
I'd hate to design for Rw=8 and tell the buliding owner that the holddowns and
plywood on the far side of the building don't work because of the change in
the living room wall that caused us to add a frame!

<< 2. Rigid Diaphram and shearwall stiffness analysis in residential
 Construction - The 1997 UBC (and proposed Los Angeles Regional Codes)
 requires a diaphragm and shearwall deflection analysis.>>
THIS IS A VERY BAD IDEA in my opinion, of course.

A)  Typically, the weak link in a plywood building is the short wall.  Using a
rigid diaphragm analysis, this wall gets even less load and will now be even
weaker!  Our fancy analysis will help to create a soft story in the building!!

B)  The code prohibits (or at least frowns on) rotation in wood diaphragms.
This is justified because rotating wood diaphragms tend to perform poorly (see
Northridge Meadows).  The rigid analysis encourages rotation.

C)  Wood buildings are not like commercial steel and concrete buildings where
the architectural finishes have little effect on the structural system.  In
wood buildings, the drywall and stucco have an enormouse effect on the
stiffness of the walls, at least for the first few cycles of the earthquake.
Where does that go in the model?  The same is true for the ceilings and the
floors.  Is your floor diaphragm stiffer with hardwood or tile or carpet over
your plywood?  You have to think it matters.  Do we model it?  What if the
owner changes it?

D)  Wood buildings (residential) tend to get many design changes over their
life and even the original construction period.  Does the addition of a new
opening in one wall during construction, which weakens the wall and
redistributes forces to the other walls mean that I have to increase the
nailing and HD's for those walls?

E)  Wood buildings are not always built per plan.  Whereas if I specify a
W12x50 column, I will usually get it, the saem is not true in wood.  I may
specify nailing at 6" o.c, and get 4".  I may specify 1/2" roof plywood and
get 5/8".  This is impossible to know and to model.

F)  How do all the "nonshear" walls in the building effect the response?

G)  Conclusion:  We must differentiate between Analysis and Design!!  While
rigid diaphragm analysis may be closer to reality the limitations in the
modeling assumptions make it a fruitless task.  And why are we doing this
anyway?  Do well designed and detailed wood buildings under the current code
have severe structural problems in an earthquake?  Let's not be afraid to jump
off the analysis horse and focus on the design issues.

<< 4. Prescriptive Residential Construction Vs New Design Standards >>

These shoudl be limited to small, regular structures.
 <<5. Perforated Shearwall Design - APA #157:  American Forest and Paper
 Association (AF&PA) presented a methodology for the design of perforated

See my response to (2) above.  Again, the construction irregularities (tight
blocking?, plumbing lines?, straps installed properly?)  can cause problems.
We should be aiming to simplify the design procedure not make it into a
doctoral thesis!

That's it for me.  Now back to work.

Bruce Resnick, SE
Parker Resnick Str. Eng.