I believe residential roof trusses represent a false economy for a roof with a ridge height over 8 ft because the potential value of that space for future use is many times greater than the initial savings. In renovations, existing trusses have prevented me from adding elevators, skylights, and cathedral ceilings so they have not only cost me design fees but made me acutely aware of how shortsighted it is to use them. So, I would say roof trusses make sense for developers or for a custom designed house with a roof slope lower than 6 in 12 or if the owner doesn't plan to own the house for very long. But it is hard to put a price on drywall cracks from truss uplift. I haven't found a reason to specify a roof truss in 40 years nor has anyone ever suggested that I do so.

brickeyee Your point about RH of winter air seems well taken. What was missing from the Carson Dunlop explanation, I think, is that attic air in a home is often more humid than air outside the attic due to leakage from the conditioned to unconditioned space. Here's how another P.E. describes truss uplift: "A problem that can occur with this system causes cracks in wallboard over time. In recent years, attic insulation usage has increased, which ends up covering the bottom chord of the truss. This tends to keep the bottom chords warm and dry while the top chords absorb moisture and expand. The top cord expansion causes the top chords to lift, pulling up the bottom chord as shown in Figure 1, hence the term truss uplift. " Charles C. Roberts, Jr., Ph. D., P.E. My untrained guess would be that the lengthwise shrinkage of grain does not relate to the length of the wood but to which portion of the tree the wood was cut from. Here's a longer explanation: "Truss uplift is caused by wood's natural response to moisture exposure. This response occurs when humidity changes differentially in two areas (an attic floor and attic ceiling, for example). A truss's top chords expand as their moisture content increases. Problems resulting from truss uplift are most common with flat-bottomed trusses in homes during winter when a truss's bottom chords, often covered by insulation, are kept warm and dry while its top chords are exposed to moisture from condensation. During winter months, warm, humid air emanates from buildings' interior use and occupants. The uplift movement occurs because, as the truss's upper components resize themselves under the differential moisture conditions, internal tensile stress occurs in the truss's web members, which then pulls up the bottom chords (and the attached ceiling drywall), resulting in a separation along the wall-to-ceiling juncture of the drywall. To put this into perspective, consider a truss with top chords made of two members, each measuring 30 feet in length. The total movement anticipated from winter swelling caused by the truss's moisture content increasing is a 0.72-inch expansion per side, or a 1-inch total rise of ridge above the bottom chords. And once a crack forms inside a home, additional moisture generated by occupants and their activities can escape into the attic, travel to the cold attic ceiling surface, condense on the truss's top chords and cause further truss uplift. " Lucas J. Hamilton manager of building science applications, CertainTeed Corp.'s Insulation Group For interest, I will forward this thread with a request for comments to Carson Dunlop. lzerarc As I said, different places, different practices. I've changed only reluctantly. Seeing those spindly stuck-together 2x2s gives me a visceral reaction and I have to rationalize using roof trusses. However, my engineer says they're much better done now than ever and with wall lengths less than 50 feet uplift is rare. None of this applies to warm climate homes.

Take a closer look at the photos. The existing structure is, in fact, a very simple truss but only because the rafter chords sit on the ties just beyond the wall plates instead of directly on the plates. If the rafters did rest on the top plates, the ties could simply be raised until the rafters reached their maximum bending capacity (about 6" by the IRC tables with assumptions listed earlier). Figuring out how to efficiently raise the ties to the mid point of the barely adequate rafters and repair the rafter seat detail has been discussed at length earlier. I have attempted to give the OP sound structural advice within the parameters I listed because that is what I have been trained to do. If there is no building department oversight and he is willing to tolerate sag in the rafters and movement in the wall top plates during extended snow storms, all he needs to do is block or strap the rafter bottoms to the plates and then raise the ties. However, all it would take under full loading conditions is for enough tie connections to fail and the shed could collapse. In my opinion, the tie connection overlap is too small for adequate nailing or bolting and should have a metal or wood plate. Potential structural failure is not an option I would recommend even for a shed.

Usually, the dips(if the dips are between the trusses) are caused by roof sheathing that is too thin. That allows the weight of the shingles to bow the sheathing. Example: Roof trusses are set on 24" centers(24" from the center of a truss to the center of the next truss). That is allowable by building codes. The sheathing is 1/2" (actual 15/32") plywood. That is also within most building codes. However, the 1/2" sheathing is not strong enough to hold the shingle weight without bowing between the trusses. It is not going to fail, but looks terrible. Solution is to remove all the shingles, tear off the thin sheathing, and replace with 5/8" thick sheathing.