Wind-Borne Debris: Impact Resistance of Residential Glazing

FULL TEXT: Adobe Acrobat (*.pdf, 648 KB)

NOTE: Adobe Reader is required to download, view, and/or print PDF files. If your computer does not have this software, you must first download Adobe Reader, and follow the installation instructions before accessing PDF files from PATH's Web site.

Since the disaster caused by Hurricane Andrew in 1992, protection of glazing (i.e., windows and doors on the exterior of homes and other buildings) from wind-borne debris damage has received much attention. The resulting impact on product standards and building regulations has affected the balance of affordability and performance of new homes in some hurricane-prone regions of the United States with uncertain costs and benefits. In part, this uncertainty is due to the lack of an objective calibration of performance criteria to historically accepted levels of risk. In general, historically accepted levels of risk are represented by the actual built environment and are not easily determined by supposition based on anecdotal observations and "expert" opinion. Data on the impact resistance of typical residential glazing materials (i.e., annealed glass) is needed to realistically analyze and benchmark an acceptable level of risk of glass breakage due to wind-borne debris. From such data, risk-consistent levels of performance (i.e., impact load criteria) can be objectively determined and applied to specify glass (or glass protection devices such as shutters) used in hazardous conditions, particularly when the benchmarked level of accepted risk is exceeded with typical glazing materials such as annealed glass.

The objective of this research was to provide needed data on the fragility (i.e., impact magnitude vs. glass breakage probability) of typical residential glass using field-observed and standardized missile types representing wind-borne debris. In this experimental study, representative sources of debris hazards, such as pieces of roof shingles and nominal 2 in. by 4 in. dimension lumber (hereafter referred to as a '2x4') were used to impact "standard" (i.e., non-impact resistant) glass, namely annealed glass. The test matrix included both 3/32-in. and 5/32-in. glass thickness tested at 1:1 and 2:1 aspect ratios (2 ft x 2 ft and 2 ft x 4 ft panels). Impact speed was varied as necessary to characterize fragility (i.e., glass breakage probability vs. impact magnitude). In addition, the response of annealed glass to multiple impacts was investigated. Representations of impact magnitude using kinetic energy and momentum are also compared with respect to the ability to predict glass behavior or fragility.

From this work, it is anticipated that wind-borne debris hazards and standardized performance criteria for wind-borne debris protection of glazing in residential buildings will be improved or at least better understood. In particular, this information is intended to improve hazard-modeling assumptions used to assess the risk of glass breakage in hurricane-prone environments of the United States. Ultimately, such research should lead to optimized solutions for wind-borne protection of window and door glazing in homes and other buildings.

Content updated on 3/24/2006