The majority of low-rise residential structures in the U.S. are constructed with wood. Wood-based composites are primary building materials in these structures, used as structural sheathing, joists, and beam components. Wood composites are susceptible to degradation upon exposure to high levels of moisture. Moisture durability is routinely assessed with accelerated weathering (AW) procedures. The goal of AW is to expose wood composites to the degrading influences of moisture in a suitable assessment time-frame, which can require severe conditions. Small specimens are evaluated to reduce the time necessary to observe degradation and increase replications within constraints of quality control and research laboratory equipment. Applying AW results to wood composites exposed to adverse conditions in service is hindered by severity of AW conditions and small specimen sizes evaluated. Moisture uptake can differ between small and large specimens. Edge effects, or the propensity for lower resistance to moisture uptake at the edges, are assumed to be less detrimental as specimen size increases. The influence of edge effects, and in particular, the ability to apply AW results of small specimens to sizes more representative of those in service are largely unknown. Moisture durability of larger wood composites and assemblies produced with wood composites have been investigated, but indicate that a knowledge gap remains.
The influence of edge effects on moisture durability along with performance of structural-size specimens and assemblies were studied in this work with a multi-scale approach. At the small scale, the influence of specimen size on AW results was evaluated in three wood composite products commonly used in residential construction: laminated veneer lumber (LVL), oriented strand board (OSB), and plywood. Moisture durability of wood composite I-joists was evaluated using laboratory AW and outdoor weathering. The impact of moisture degradation in engineered wood structural panels on full-size shear walls was investigated by subjecting OSB and plywood to AW prior to shear wall assembly. Finally, a numerical model predicting moisture transport in OSB and plywood was developed to better understand moisture transport in these materials during cyclic changes in relative humidity.
Specimen size was most influential on water absorption, where AW method-dependent relationships between specimen size and water absorption were observed. However, the edge effects observed for water absorption did not directly translate into mechanical property loss. Lack of distinct trends regarding edge effects indicates that a wider range of specimen sizes may be necessary to distinguish these effects. Short-span bending strength of I-joists was reduced by both forms of weathering. One month of outdoor exposure in a rainy, cool climate reduced bending strength twice as much as the full laboratory AW procedure. In addition, moisture exposure resulted in a shift in failure mode, indicating degradation of the OSB web and web-flange joint. Weathering OSB prior to shear wall construction resulted in statistically significant reductions in yield load, maximum load, and energy dissipation. Weathering plywood sheathing prior to shear wall construction had no influence on wall properties. The numerical model developed for moisture transport in OSB and plywood was validated with experimental measurements, where the average deviation between measured and predicted values was 9.0% and 10.2% for OSB and plywood, respectively.
Information gained on the influence of specimen size will help guide future experiments to better understand these effects. Results from I-joist and shear wall tests could provide engineers with residual properties after exposure to adverse conditions to determine remedial actions. The moisture transport model can aide in developing and refining AW procedures and provide moisture field input into damage evolution models.