Need a book? Engineering books recommendations...

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

Re: Plywood rigid diaphragms

[Subject Prev][Subject Next][Thread Prev][Thread Next]
In a message dated 98-08-29 11:10:59 EDT, you write:

<< What about plywood shear walls that do not have any holdowns?  (there is no
 calculated uplift.)  And what if those wall are in line with a wall that does
have
 a holdown?  How about Simpson PHDA holdowns that supposedly limit holdown
 deflection?  How about strap type holdowns?  Will a strap with 500 lb. uplift
have
 the same deflection as an HD20A with 15,000 lb. uplift?  Do you need to
calculate
 the deflection of each holdown type based on your guess of the loads, run the
 analysis, and then re-run it?  And when the Architect moves a door or window
at
 the last minute do you tell him it will cost $2,000 to re-design the walls?
>>

The deflection of the holdown assembly includes the holdown body (example:
Simpson catalog provides deflections for their holdowns at maximum load listed
in the table) and the immediately attached wood members (post, sill plate,
support under wall.)  For sources of holdown slip you can look in the Simpson
catalog, which are more sources than I think anyone can reasonably include.
The primary ones I would think to consider are the 1/16" minimum oversized
hole in the post, crushing of the sill plate, holdown deflection, and any wood
shrinkage or sill plates or joist under the sill plate.  On long walls the
crushing will be less of a problem, but remember that most sill plates are
treated hem-fir, and therefore crush at a lower load than douglas fir.

The PDA will have less slip than a traditional holdown because you have
eliminated the 1/16" oversized hole in the post and the body of the holdown
has less deflection.  But you still have deflection from lumber shrinkage and
crushing.  Some current building jurisdictions require that you wait until
just before closing in the wall to tighten up the holdowns to account for
lumber shrinkage.  In evaluating older buildings with the traditional
holdowns, it is not uncommon to find them loose since the wood joist have
shrunk.  The resulting shear wall deflection is even greater because now the
wall has to uplift furthur before the holdown will engage. 

To analyize the shear walls,  as you mention below,  you will have to
calculate the deflection of the wall line, which may consist of one or more
shear walls in the same line.  The only thing that they must have in common is
that the top of wall deflection must be the same for all walls along the same
line.  The double top plate distributes the load based upon rigidity of  the
individual shear walls along that wall line.  The uplift of the individual
shear walls will depend upon the gravity loads and the length of the shear
wall, therefore it is quite possible that some walls may not have uplift (long
walls with large gravity loads) while other may require large holdowns (short
walls with insignificant gravity loads) along the same wall line.  So based
upon the load distribution along a wall line, a strap may have the same
movement as a HD20 because of the load distributed to the individual walls.
Note that the deflection between two parallel wall lines can and most likely
will be different.  

The only practical way to distribute the loads for each wall line is to write
yourself a spreadsheet program which can iterate the loads between the walls
along the same wall line (reducing loads to some walls, increasing loads to
other walls) until you reach a convergence of where the deflections are all
within an acceptable tolerance range (say 0.05 to 0.10 inches between the
maximum and minimum shear wall deflections)  You can do it by hand if there
are only two or three walls along the same wall line, but even this can be
time consuming, especially if you keep changing holdown sizes.

As you noted, when an architect suddenly changes a door location or adds a
window, etc. this causes quite a few problems.  I don't have a practical
answer other than revising the nailing and holdowns for the walls along this
particular wall line until you get about the same deflection for that wall
line.  I will have the same problem as you trying to collect addtional money. 

<<< You won't necessarily be calculating the deflection of a single wall, but
the
 deflection of a "wall line".  But another problem is that these plywood walls
do
 not distribute loads like concrete walls do.  Due to slippage of nails and
other
 factors, the forces tend to equalize themselves along the run of a wall.  The
 deflection at the top of the wall at one point may not be the same as
another.
 The horizontal diaphragm must also be "flexible" enough to accommodate this
 re-distribution of stresses or the whole theory goes out the window.>>>

I agree that the forces will equalize themselve along the run of the wall, if
you do not have holdown slip.  When you take the same wall line and continuely
increase the lateral loads while keeping the framing constant, eventually all
the walls take the same uniform load per foot.  But by this point you have
greatly exceed the allowable deflection of 0.005H.  Now once one wall line
deflects, the plywood diaphragm will drag load to the next adjacent stiffer
shear wall lines adding load to these wall lines.  

To me it is this redistribution which should be accounted for somehow if the
diaphragm is going to be considered somewhat rigid.  The added load to this
adjacent shear line can not be more than its tributary area plus the tributary
area of the wall line which deflected.  I would think we could redistribute
the load base upon the difference in deflection between wall lines by some
simple formula.  Ideally all wall lines in a given direction would have equal
translational deflection, which will not happen in the real world.  I might
propose, that when the deflection difference between parallel wall lines
exceeds 10%, then some sort of redistribution should be done.
 
 <<<<In my opinion, this falls in the category of trying to fix something that
is not
 broke.  From what I have seen and read about the performance of wood framed
 structures during recent earthquakes, wood framed horizontal and vertical
 diaphragms did very well with the exception of tall slender walls.  I tend to
lean
 toward the "where are the bodies" camp when making significant code changes.
 
 You are really going to open a pandora's box if we try and go down this road
too
 far.  Near as I can tell, most buildings perform fine when horizontal and
vertical
 diaphragms are considered flexible.
 
 Lynn
  >>

I believe that pandora's box has been opened and that there's probably no
going back.  I would agree with you on where are all the bodies, but I think
we have uncovered something about the holdowns which we must consider.
Hopefully this discussion will bring to light problems with holdowns and
reasonable solutions, not necessarily codifed solutions.  Generally, for a one
story building I would think the holdown issue is going to be less of a
concern then for multistory wood buildings.  Multistory buildings are going to
require longer shear walls with aspect ratios of 1:1 or 1:2 or more to
maintain deflections limitations which most architects are likely to find
unacceptable.   As a side point, I imagine the enforcement of the deflection
criteria will likely only be by the larger building departments since
habitability after an earthquake and not just life-safety is going to be a
growing concern.  Where do you put 100,000 people after an earthquake in a
large city and its suburbs if the wood frame housing is not considered
habitable, remember Kobe had over 300,000 homeless.

Hopefully we can devise something relatively simple to account for the holdown
assembly movement and shearwall deflection which will not impact our design
practice too much.

Any other thoughts?

Michael Cochran