REVISITING WIND UPLIFT TESTING OF WOOD ROOF SHEATHING - INTERPRETATION OF STATIC AND DYNAMIC TEST RESULTS

Following Hurricane Hugo (in 1989) and Hurricane Andrew (in 1992), several organizations initiated wind uplift testing on wood roof structural components. The results from those tests have informed many of the wind load design requirements for residential roof construction. Still, during recent (2004 and 2005) hurricanes, existing roof structures built in accordance with those “wind-resistant” provisions were damaged. This research is motivated by the need to understand the causes of such damage and to improve the prediction of structural failure of wood roof sheathing panels in extreme wind events. Current wind uplift tests typically apply spatially uniform, static pressures to roof specimens to predict their performance under actual wind pressures that are temporally varying (dynamic), and spatially nonuniform. A review of prior wind uplift tests and comparative data from the studies is presented. At the same time, the authors question the validity of an exercise that compares failure capacities from studies that used different test protocols. As a result, a proposed standardized test protocol (covering both static and dynamic test methods) was developed as a model for future wind uplift testing of wood roof sheathing panels. The protocol was developed in two phases; Phase 1 includes static uplift pressure tests conducted to compare the ultimate wind uplift capacities of lab-built wood roof sheathing panels versus wood roof panels harvested from existing residential structures. In Phase 2, a Pressure Loading Actuator was used to conduct static and dynamic testing and compare the wind uplift capacities of comparable sets of lab-built and existing wood roof panels between two protocols. Results from the uniform static pressure tests confirmed that that the failure capacity of roof panels is linearly dependent on interior fastener spacing. The withdrawal resistance of deformed (ring shank) nails was also validated as in previous test results. The authors recommend that future test reports should include documentation of the material properties of wood, (i.e. specific gravity and moisture content), and also the nail geometry, as these do affect failure capacity although no trends could be determined. The dynamic tests produced ultimate failures that were on average 20 to 30% lower than the static failure capacity for similarly constructed panels. This finding is significant because it suggests that current structural test protocols may overestimate the wind uplift capacity of wood roof systems, (and perhaps other roof and cladding systems) subjected to dynamic wind pressures. The premature failure of roof systems that are frequently observed during hurricanes may be related to this discrepancy, compounding issues, such as improper spacing or missing framing and overestimating the capacity of properly made connections. A more extensive comparison of the static versus dynamic performance is underway, which will include tests on additional panels harvested from existing residential roof structures.