Although some screw manufacturers have received ES reports including diaphragm tables there is some concern within the engineering community regarding screw performance in this application.
APA does not recommend screws for diaphragms (or shear walls) in high-seismic areas. Cyclic testing has shown that nails tend to bend and withdraw slightly rather than breaking and withdrawing completely. Even when partially withdrawn, nails retain some reserve withdrawal resistance. On the other hand, once a screw begins to withdraw it loses all withdrawal resistance due to the shearing of the threads.
Another issue is the potential of brittle failures from screws. Compared to nails screws are generally quite brittle. Limited cyclic testing has shown that screws can exhibit a non-ductile failure. This issue, however, may be a bit academic. For typical residential structures with light loads, it is questionable as to how much displacement the fasteners actually see. A cyclic event may not bend the fasteners enough to cause a brittle failure.
Some engineers suggest that the shank of screw should extend some depth into framing. In addition to pulling the panel tight against framing this is based on the assumption that screws tend to break at the end of the threaded portion. Another interesting factor is that fasteners with cut threads tend to be more brittle than those with rolled threads.
A few ICC ES reports on wood screws for wood structural panel diaphragms we know of (which may not include all applicable ES reports, last verified 02/2009) are:
ER 5053 - Quick Drive Screws
ER 5280 - Grabber Screws
ER-5634 - USP
The three evaluation reports have diaphragm tables directly in the text or reference building code tables. It also may be possible to derive diaphragm design values following principles of mechanics for some of the other reports covering screws. ICC reports are available from the ICC web site:
Section 2305.1.1 of the IBC permits calculation of resistance without limit based on the principles of mechanics. As long as a material (screws, staples or glue or anything else) has recognized lateral design values it may be used. This code provision, however, does not address the potentially important issues of fastener ductility or toughness.
I brought up a similar question a couple years ago. My question was related more to floor diaphragms, but the idea is still the same. I was asking because nailed floors tend to start squeaking, but I realized there could be ductility concerns with screws. I never got a clear answer, so this thread will be interesting.
A lot of floors get glued and screwed.... May not be the "best" way - but done a lot.... Especially with the availability of the extended screw guns allowing carpenters to install from a standing postion.
