As I mentioned earlier, our roof was constructed using trusses rather as opposed to being stick-built in the field:
I don’t know if using roof trusses is less expensive than a roof built in the field. But I do know that the trusses allowed us to construct the house without any load bearing walls on the second floor (the second floor floor was constructed using floor trusses, and did require a single load bearing wall on the first floor). This made it much easier to seal the second floor ceiling (with Zip sheathing) because it gave us the ability to do it before any interior walls were constructed. But it also raised (at least with me) two concerns.
The first concern was something that I happened upon by accident, when I encountered an article that mentioned “truss uplift.” Evidently, truss uplift is a condition that can (but does not always) occur when the bottom chord of a roof truss (which is buried beneath the insulation) absorbs moisture at a rate that is different than the top chords (that run along the roof line). If my understanding is correct, when this happens (usually in winter), the upper chords bow causing the bottom chords to lift. If the interior walls and the drywall ceiling are attached to the bottom chord, this “uplift” can cause damage to the walls (which are presumably also attached to the floor. It can also cause the drywall ceiling to separate from the drywall on the walls.
Simpson makes a bracket that is used to attach the interior walls to the trusses in a way that prevents this problem (if the drywall ceiling is installed without using screws near the seam).
The problem is that these brackets can’t be used in our house because the Zip sheathing is attached directly to the trusses. So that left us with two options; either hope that it doesn’t become an issue, or come up with a way to allow the ceiling to move without tearing up the interior walls.
The solution that I came up with (which I hope will work should the problem occur) is as follows:
First, I screwed the top plates for the interior walls to the Zip sheathing and trusses (or blocking between the trusses where the walls ran parallel to the trusses):
Then I constructed the interior walls, screwed them to the floor, and attached them to the top plate with four-inch Timberlok screws, which I did not screw in completely:
If you look closely, you can see a couple of the Timberloks in the above photo. They have two inches of thread. The other two inches is a smooth shank. Therefore, if they are left slightly less than a half-inch “out,” all of the threads are embedded in the top plate and Zip sheathing, and there is room for the top plate to rise 1/2″ independently of the wall that is attached to it.
So that’s part of the solution. The other part was to ensure that the drywallers did not screw the drywall to the top plate and that they kept the ceiling (drywall) screws back, away from the interior walls. As with the Simpson brackets, the idea is to allow the trusses to move up, without forcing the drywall to separate at the seams.
Hopefully all of this was unnecessary. That may be the case because a) the interior of our house was unfinished for a year, (giving the trusses to dry completleyl) and b) unlike a typical house, our house isn’t going to allow any moisture to move into the attic during the winter months (or any other time during the year, for that matter). But then again, maybe the upper chords dry out more than the bottom chords in the winter.