Need a book? Engineering books recommendations...

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

[SEAOC] Re: [SEAOC] Re: [SEAOC] Re: [SEAOC] Re: [SEAOC] Re: [SEAOC] Re: Reroofing (Was: Basic Roof Rafter Ana...

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

Some possible considerations regarding whether wood structural panel sheathing will stiffen/strengthen the structure:

a.  Addition of panel sheathing should stiffen the roof structure to resist lateral loads from wind/seismic.  The question is, how much "diaphragm" action will be developed, since there are typically no continuous tension or compression chords, no perimeter nailing/stapling of the panel sheathing at the boundary of the "diaphragm" to transfer diaphragm shear forces to the framing (chords), no direct connection of the roof sheathing to the framing (and intermediate slip plane is created when the spaced sheathing is left in place), and no connections across the ridge of the roof (assuming a typical pitched roof), where not only the panel sheathing but also the framing is discontinuous.  Typically, the only way to get the diaphragm shear forces transferred to chords (doubled top plates of the wall) is if truss hold-down connectors are added, such as Simpson H10 which can not only reinforce against uplift, but also transfer some shear.
b.  Ideally, the plywood sheathing should be fastened to blocking between trusses/rafters, to create the diaphragm perimeter fastening to the chords.  At the ridge, a formed sheet metal strip can be stapled to the panel sheathing over the ridge, to transfer shear forces parallel to the ridge from one side to the other.  If a ridge vent strip is used, the gaps for ventilation can be spaced so that the intermediate sheathed portions can transfer the shear forces.
c.  At end walls, the blocking should be fastened to the wall top plates with a Simpson A35 or similar to transfer diaphragm shear to the end shear walls.  Is this typically done?  Probably not.  The load path needs to be followed from the diaphragm to the end shear walls regardless of whether the end roof structure is a gable end wall, french hip or hip roof.  If you really want to consider diaphragm action, a structural analysis would be needed (too expensive and complicated), but for "conventional construction," simplifying assumptions are made in UBC requirements (e.g., "deemed to comply").
d.  APA has a technical note available on the installation of wood structural panel sheathing over spaced boards.  An ICBO chapter in No. CA is developing this into a guideline for contractors.
e.  In my opinion, the best solution is to forget heavy tile roofing and instead install panel sheathing over the spaced boards as a substrate for laminated asphalt shingle roofing (Class A) which is lightweight (3psf), fire resistant, and looks like wood shake roofing.  As an alternate, there are several sources (I think) for metal tile roofing that looks like the real thing but is lightweight.

Comments anyone?

John Rose/APA, Tacoma, WA
_______


>In response to RLFOLEY(--nospam--at)aol.com:
>I believe the analogy is still the same. The shear transferred by 1x
>skipped sheathing is less than the same diaphragm with plywood sheathing.
>Furthermore, it's unreasonable to think that you can transfer more shear
>than the capacity of the diaphragm will allow. If the demand of the
>diaphragm exceeds the capacity of the diaphragm, the diaphragm will fail -
>i.e., panel buckling, nail popping etc. This is the concept between
>capacity and demand. 
>Therefore, if the capacity of the diaphragm were 100 pounds per foot, the
>resisting wall (whether it be stucco, plywood or masonry) must be able to
>resist the accumulation from the diaphragm at 100 plf. Any more than this,
>the diaphragm will fail. If the diaphragm were 20 feet long then it can
>only impose a maximum of 20*100 or 2000 pounds to all shearwalls in that
>line. This makes one assumption that I can be faulted on, and that is to
>assume that the resisting walls are capable of a capacity equal to the
>connected length of diaphragm times it's capacity (plf). Most of the time,
>it is much less.
>But, following one step further, if you indiscriminately sheath the roof
>and increase the demand by the diaphragm to 2.5 times it's original
>capacity, you must increase the shearwall capacity accordingly - which we
>don't normally do.
>So, without seeming too verbose, the increase in the demand of the
>diaphragm can be detrimental to the existing shearwalls and should be a
>consideration before recommending a new diaphragm - whether above the
>existing or in place of it.
>Hope this clarifies my point. There is little difference in the URM and
>wood frame when talking about flexible diaphragm action. The main
>difference lies in that we automatically design from tributary shear in
>wood framing and in a modified distribution that considers the diaphragms
>capacity in URM work.
>
>Thanks for the comments
>Dennis Wish PE
>wish(--nospam--at)cyberg8t.com
><html><head></head><BODY bgcolor="#FFFFFF"><p><font size=2 color="#000000" face="Arial">In response to RLFOLEY(--nospam--at)aol.com:<br>I believe the analogy is still the same. The shear transferred by 1x skipped sheathing is less than the same diaphragm with plywood sheathing. Furthermore, it's unreasonable to think that you can transfer more shear than the capacity of the diaphragm will allow. If the demand of the diaphragm exceeds the capacity of the diaphragm, the diaphragm will fail - i.e., panel buckling, nail popping etc. This is the concept between capacity and demand. <br>Therefore, if the capacity of the diaphragm were 100 pounds per foot, the resisting wall (whether it be stucco, plywood or masonry) must be able to resist the accumulation from the diaphragm at 100 plf. Any more than this, the diaphragm will fail. If the diaphragm were 20 feet long then it can only impose a maximum of 20*100 or 2000 pounds to all shearwalls in that line. This makes one assumption that I can be faulted on, and that is to assume that the resisting walls are capable of a capacity equal to the connected length of diaphragm times it's capacity (plf). Most of the time, it is much less.<br>But, following one step further, if you indiscriminately sheath the roof and increase the demand by the diaphragm to 2.5 times it's original capacity, you must increase the shearwall capacity accordingly - which we don't normally do.<br>So, without seeming too verbose, the increase in the demand of the diaphragm can be detrimental to the existing shearwalls and should be a consideration before recommending a new diaphragm - whether above the existing or in place of it.<br>Hope this clarifies my point. There is little difference in the URM and wood frame when talking about flexible diaphragm action. The main difference lies in that we automatically design from tributary shear in wood framing and in a modified distribution that considers the diaphragms capacity in URM work.<br><br>Thanks for the comments<br>Dennis Wish PE<br>wish(--nospam--at)cyberg8t.com<br><br></p>
></font></body></html>

...