University of Florida researchers  tested three levels of spray-foam roof underside retrofit: a “fillet” application at the joint, a one-inch application across the whole  cavity (as might be used in a "flash and batt" application), and a three-inch application filling the bay. The thicker foam application provides the most uplift strength (far stronger than nails alone). But the thicker foam also tends to trap moisture in the sheathing in the case of a roof leak. However, applying self-adhering membrane roof underlayment to the top of the sheathing can sharply limit this moisture risk, the researchers report.
University of Florida researchers tested three levels of spray-foam roof underside retrofit: a “fillet” application at the joint, a one-inch application across the whole cavity (as might be used in a "flash and batt" application), and a three-inch application filling the bay. The thicker foam application provides the most uplift strength (far stronger than nails alone). But the thicker foam also tends to trap moisture in the sheathing in the case of a roof leak. However, applying self-adhering membrane roof underlayment to the top of the sheathing can sharply limit this moisture risk, the researchers report.

Applying closed-cell spray insulation to the underside of a roof deck at the sheathing-to-truss connection can double or triple the strength of the assembly, boosting the roof sheathing’s resistance to wind uplift in a hurricane. But foam also traps moisture in the roof in case of a leak: a sealed and insulated roof deck can’t dry inward to the attic if the roofing happens to let water through.

So there’s an opportunity, but also a risk — and that problem is getting a lot of scrutiny these days from researchers at the University of Florida, led by Professor David O. Prevatt, a P.E. and Ph.D. who is an expert in wind engineering.

In 2011, Prevatt and colleagues Peter Datin and Weichiang Pang studied the strengthening effect of high-density adhesive insulating foam on the sheathing-to-framing joint and published the results in a paper for the Journal of Architectural Engineering (“Wind-Uplift Capacity of Residential Wood Roof-Sheathing Panels Retrofitted With Insulating Foam Adhesive,” by Datin, Prevatt, and Pang). The team looked at a variety of attachment methods in combination, examining three different hail types, two nail spacings, and three ways of applying the closed-cell foam. They placed test assemblies in a steel pressure chamber to simulate the suction of high winds flowing over a roof. What they learned was that adding foam insulation to a nailed-down roof could increase the uplift resistance of the sheathing by 250% to 300%.

But what if the roof leaks? In a 2012 study, Prevatt and some University of Florida colleagues looked at that issue (“Wind Uplift Capacity of Foam-Retrofitted Roof Sheathing Subjected to Water Leaks,” by David O. Prevatt, Kenton McBride, David B. Roueche, and Forrest J. Masters).

The team built five full-scale attic structures at the University of Florida’s Hurricane Research Laboratory in Gainesville. The roofs got a spray-foam underside application in various thicknesses. Then the researchers drilled 100 half-inch leak holes into half of the roofs and left the roofs exposed to natural weather conditions — plus 150 days in a row of a 15-minute water spray rain simulation.

The results? A foam-adhesive roof joint holds its strength even when the roof leaks. Wrote the study authors: “Increased moisture contents in wood did not produce statistically significant changes in panel failure pressures over the 150-day weathering period. No strength reduction was seen between roofs exposed to leakage compared to non-leaking roofs containing the same retrofit. Furthermore, no correlation was found between moisture content and failure pressure of individual panels within leaking roofs under wide ranges of moisture contents during structural testing of 8% to over 60%, suggesting that the presence of water has no measurable effect on wind uplift capacity within a 150-day period.”

But wet wood in a building is never a good thing, of course — over time, it’s a recipe for rot and deterioration. So the researchers looked for ways to make the foam-retrofitted roofs more resilient in the face of water intrusion. They set up four more test roofs, with somewhat less extensive leak holes. One roof got a venting system to allow the underside of the roof to dry, while another got an adhesive waterproofing membrane. Complete results of that test aren’t available quite yet, but the authors say that both methods show promise: “In the roof incorporating vents, moisture content was reduced from 40% to below 20% in 3 days, as compared to 12 days for the roof that did not have vents in its closed-cell spray polyurethane foam retrofit. It was also observed that installing self-adhered waterproofing membrane as roofing underlayment substantially reduced the amount of moisture intrusion into the roof system. Even with comparable number of leak gaps cut into the membrane, the moisture content in Roof 2 remained below 20% for the time period.”

The University of Florida research is partly funded by the Florida Building Commission. And while it may take some time for public policy and building codes to reflect the results of the study, there is a lot to be gained from finding a way to make Florida homes more wind-resistant and moisture-tolerant at the same time. Upgrades in the Florida building code that were instituted after Hurricane Andrew have significantly reduced the damage caused by later hurricanes. But as Prevatt’s team points out, 80% of the homes in Florida were built before those new codes took effect. And for those older houses, spray foam — once all the bugs are worked out — offers a simple way to strengthen many of those existing roofs.

In addition, there are other benefits to insulating the underside of a roof deck. When HVAC ductwork runs through the attic, research shows a big energy savings from insulating the roof deck to keep the ductwork within a conditioned space. And wildfire researchers say that eliminating roof vents and building a “hot roof” also eliminates one of a house’s greatest vulnerabilities in wildfire: the risk of windblown embers entering the attic and igniting the house. But the risk of moisture damage from roof leaks stands on the other side of the equation — so the research into that problem by Prevatt’s University of Florida team could be significant in the years to come.