Need a book? Engineering books recommendations...

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

Re: Effects of the New Code on Wood structures - good or bad?????

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Dennis Wish wrote:

		Mark, the answer is very simple - none of these provision
were either 
		applicable to wood framed construction with wood diaphragms
until 1988 and 
		none of them were either enforced or consider the "Standard
of Practice" for 
		wood framing until this code cycle.
		The actual reason is not known just why the provisions were
not enforced 
		prior to 1997 UBC but it is suspected that it was because
there was no 
		rational means to calculate the deflection of a flexible
diaphragms (and 
		there is only a multiplier suggested by APA at this time)
and the provisions 
		in the code could not be satisfied.

Mark Swingle responds:
There is STILL no rational means to calculate the deflection of WOOD
diaphragms.  In addition, very few people calculate the stiffness or
deflection of ANY continuous diaphragms, because it is very difficult to
analyze even for relatively isotropic materials such as reinforced concrete
(at least it is isotropic at elastic levels).

The provisions in the code still cannot be satisfied absolutely, but a close
approximation of such an analysis could and can be satisfied in a
conservative way if we know the stiffness of the SHEAR WALLS.  The stiffness
and deflection of the diaphragm is immaterial if we bracket the solution by
totally flexible and infinitely rigid analyses.  See my post from earlier
today regarding this.

I still claim that there was no significant change in the 1988 UBC other
than to clarify that the 5% eccentricity was to be IN ADDITION to any actual
eccentricity. That was not true in the earlier codes, all that was required
was a MINIMUM eccentricity, if the actual eccentricity was small.  Before
1988, if the actual eccentricity was greater than 5%, then no additional
eccentricity was required.  All the other changes in the 1988 UBC were
merely attempts to QUANTIFY what was already in the code since the early
1960s. The reason that a "flexible" diaphragm needed to be quantified was
that there is no mathematical significance to "accidental eccentricity" when
the diaphragms are so flexible that redistribution of shears from the
tributary load model cannot occur.

Since the December 1959 SEAOC "Recommended Lateral Force Requirements" was
published, torsion needs to be considered ONLY if the diaphragm is
sufficiently stiff to transmit the torsion.  I think the original wording
helps to clarify my point that little has changed.  Here are excerpts from
the December 1959 RLFR:

		2312 (e)  Distribution of Horizontal Shear.  Total shear in
any horizontal plane shall be distributed to the various resisting elements
in proportion to their rigidities considering the rigidity of the horizontal
bracing system or diaphragm as well as the rigidities of the vertical
resisting elements.

		2312 (g)  Horizontal Torsional Moments.  Provisions shall be
made for the increased shear resulting from the horizontal torsion due to an
eccentricity between the center of mass and center of rigidity.  Negative
torsional shears shall be neglected.  In addition, where the vertical
resisting elements depend upon diaphragm action for shear distribution at
any level, the shear resisting elements shall be capable of resisting a
torsional moment assumed to be equivalent to the story shear acting with an
eccentricity of not less than five percent of the maximum building dimension
at that level.

Mark Swingle continues:
So, I will make my point yet another way.  If there is actual eccentricity
and/or if the deflection of the walls is not identical to the tributary load
model, the diaphragm will be unable to REdistribute shears FROM THE
TRIBUTARY CASE if the diaphragm is sufficiently flexible with respect to the
adjacent shear walls.  In other words, torsion can be ignored for
sufficiently flexible diaphragms because it is true mathematically, not
because the code says so.  The term "diaphragm action" refers to
redistribution that occurs in rigid diaphragms, as opposed to "beam action"
that occurs in flexible diaphragms.

So, starting in the 1960s (when the RLFR was adopted into the UBC), we could
design buildings with truly flexible diaphragms by tributary loads and still
meet every code through the present 1997 code.

The problem is that most wood diaphragms are no longer "flexible" because
wood buildings have changed in two fundamental ways: the diaphragms are
stiffer and the shear walls are more flexible than 30 years ago, so our
assumptions are no longer valid.

Older wood buildings had straight or diagonal board sheathing, but now we
have 3/4" or 1-1/8" plywood floors glued to the joists.  So the diaphragms
are much more stiff than 30 years ago.

Current wood frame buildings (especially residences) have more openings,
fewer walls, and the walls are narrower, than 30 years ago.  So the shear
walls are more flexible than 30 years ago.  In addition to this, the shear
wall deflection equation forces us to underestimate the actual stiffness of
the shear walls for two reasons, 1) the piers and spandrels are ignored, and
2) the relative fixity at the top of lower story walls is ignored.

That's all I have for now (back to work!)

Mark Swingle, SE
Oakland, CA

These views are my own and not necessarily those of my employer