30 Proceedings of the North American Conference …

1 30 Proceedings of the North American Conference on Roofing Technology long -TERM EFFECTS OF hail impact ON. ASPHALT SHINGLES AN INTERIM REPORT. SCOTT J. MORRISON. Haag Engineering Co. Carrollton, Texas, on new and weathered shingles. Shingle types tested A laboratory study was begun in 1983 to determine the long -term effects of hail on asphalt shingles. Simulated included 20-year-warranty, three-tab shingles with organic reinforcements; 25-year-warranty, three-tab shingles with hailstones were launched at speeds of same-sized natural fiber glass reinforcements and 30-year-warranty, laminated hailstones perpendicular to roofing panels constructed to shingles with fiber glass reinforcements.

2 Findings and replicate standard applications. Impacts were made on new methods derived from observations throughout this study fiber glass-reinforced three-tab, organic-reinforced three- include the following: tab and fiber glass-reinforced laminated shingles at the Identification of the characteristics of impact -caused beginning of the study and after approximately 11 years of damage to shingles. natural weathering exposure. The study confirmed that Documentation of the attributes of impact -caused frac- dents in asphalt shingles caused by impacts that did not ini- tures initially and after prolonged exposure to natural tiate fractures did not change measurably over time and, weathering.

3 Hence, were not functional damage. Additionally, it was demonstrated that hail -damaged areas in shingles do not Evaluation of granule loss at impact locations to deter- increase measurably in size with increased time of expo- mine the effect of granule loss on expected shingle life. sure. This has been found to be true whether the impact Originally, the study was anticipated to have been com- damage occurred to new shingles or to weathered shingles. pleted after 20 years, with impacts to be made on roofing It is important that functional damage to shingles caused at years 0, 7 and 14.

4 Initial impacts were made on new roof- by impacts was immediate and identifiable. This interim ing shingles in Year 0; all shingles then were allowed to report is issued after 15 years of testing, which is a typical weather naturally. At 7 years, the impacting schedule was service life for many shingles in the southern United States. re-evaluated and modified based on observations of the The study is continuing, and further reports may be issued extent of weathering of the shingles and re-assessment of after additional years of exposure.

5 The study goals. It was decided that more useful data would be generated with impacts made on roofing shingles near- KEY WORDS ing the end of their expected service lives; hence, impact - ing on shingles was postponed from Year 7 until Year 11. Asphalt Shingles, Granule Loss, hail , hail -caused Damage, The impacting regime originally slated for Year 14 similarly hail Damage, Hailstones, Ice Balls, Weathering. has been postponed based on the extent of weathering of the shingles and re-evaluation of study goals. PURPOSE.

6 Functional damage to any roof covering is defined by The purpose of this study is to assess the effects of weather- Haag Engineering Co. as a diminution of water-shedding ing on asphalt shingles impacted with simulated hailstones capability or reduction in the expected long -term service Photo 1. Naturally occurring hail . Photo 2. Shingle impacted by a simulated hailstone. Proceedings of the North American Conference on Roofing Technology 31. life of the material. More specifically, applied to asphalt dimensions, and the shingles were outlined on the acetate shingles, impact -caused damage is rupture of the reinforce- sheets.

7 Individual shingle tabs or segments of upper or ment or displacement of granules sufficient to expose lower laminates of laminated shingles were designated underlying bitumen. The former is a penetration of the beginning at the upper left corner of the test panel (refer- roofing shingles that, in effect , removes a layer of the shin- ences reflect a viewer looking upslope at installed roofing gles; the latter reduces the expected service life of the shingles) with alphabetic rows and numeric columns. material. (Haag has further refined the definition of func- Conditions of shingles were recorded on the acetate over- tional damage for laboratory analyses of asphalt shingles lays to form a permanent record.)

8 Additionally, conditions in which reinforcements are solvent-extracted for examina- were notated and photographed. The various conditions tion to include strain [permanent deformation] in the included effects of weathering and/or manufacturing and reinforcement over an area greater than 1 4 inch [6 mm] in damage caused inadvertently in handling or during storms. diameter. This is a rare circumstance detectable only in See Photo 4. extracted reinforcements.) Year 0 impacts on the shingles were made with freezer ASPHALT SHINGLES USED FOR TESTING.

9 Three types of asphalt shingles were tested: both fiber glass three-tab shingles, organic-reinforced three-tab shingles and fiber glass-reinforced, laminated shingles. Specifics about the shingles are listed in Table 1. The three selected shingle types were produced by major manufacturers and purchased through local material sup- pliers. Selection of the three shingle types was intended to represent prevalent types of products in the roofing mar- ket. The three shingle products are without known durabil- ity problems.

10 TEST PROCEDURES. In 1983, six panel frames were constructed of steel angles and clad on top sides with 3 4-inch- (19-mm-) thick plywood decking treated with a sealer. The 4-foot by 6-foot (1,219. mm by 1,829 mm) decking was covered with No. 15 Photo 3. View of test panel CO-FS-D7/14. asphalt-saturated felt. New shingles were applied over underlayment in strict accordance with manufacturers'. requirements. Finally, guards for natural hail (steel frames covered with hardware cloth) were secured approximately 5-1 2 inches (140 mm) above the test assembly.