Document BRvr5y87XV0NRRJ4Y38YpNKpX

Confidential i Donald J. Smith 3^91 Connaught Avenue Montreal 262, Quebec} Canada Tel.: C51*0 ^86-8524 PLAINTIFF'S EXHIBIT BIR-325 Progress Report June 16, '1972 INVESTIGATION OP CRACKING OF ASPHALT SHINGLES IN THE WEATHER To: iir. Carroll a . Homan oc: Messrs. W. K. Price A. K. Powers L. V/. Weaver D. E. Frazier R. E. Heim E. N. Zinkowski ) BIRD 011188 f-t e j Introduction Mr. V/. K. Price accompanied the writer on examina tion of roofs in Florida, South Carolina, and North Carolina. In addition, the writer examined roofs in iitianta, Ga. The cooperation of area salesmen has been wholehearted and essential. Mr. Price has collaborated in this progress report. .The detailed eports appended show the salesmen involved and the, reas covered. BIRD 011189 bu: L.iar;- of Observations From Ta:.na, Florida, north to Elizabeth City, N. C., all seal down shingles, both of Bird and of competitive manufac ture., are cracking horizontally, the start occurins two and one half to five years after exposure. The cracking pax-allels the butts of the shingles at a distance of one half to two and one half inches below the overlying shingles. This is about one ha?f wa; between where the exposed butt is held by adhesive and by the first rov; of nails. On one job, one course of shin ies was exposed for five inches at one end of the roof, taper ing to three inches at the other; the cracking gradually approached the butts of the overlying shingles as the exposure decreased and finally disappeared under the overlying shingles-, holy scaled, and hand sealed shingles behaved similarly. Ho horizontal ex*a eking of this type was observed on unsealed shin los. The severity of cracking appears to be affected by many fac tors. Heavier weight shingles developed wider cracks, and lighter weights have finer cracks. Darker colors have severer era chin than lighter1 ones of similar exposure. Lower slopes developed more severe cracking. At each particular location the most advanced crac!:in^ developed on low slopes on the southern to south-western exposures and was worst in Florida, least in Atlanta. Cracking right through the felt has occurr ed. in as little as four years, where all factors were unfavor able. Failure, i.e., splitting through the full thickness of the shin les, has occurred in as little as five years. It was noted, that Bird FK coatings are very friable. One Johns-hanville exposure was similarly friable.' A large area of black Firescreens, exposed in Florida on an apartment complex at about three and one half inches per foot slope for three vears has failed completely and will have to be replaced. Vertical cracking to the point of failure, particularly severe in embossed lines, was associated with the horizontal crackingon sealed shin.les. Craze type cracking developed with age in unsealed thick-butt shingles, but was not critical. Bwellin of felts was fairly general, more noticeable on ha ore 2g's and Jets. Tentative Conclusions 1. Life expectancy for Mark 25 shingles is much below the A 2y-year guarantee period., throughout the area surveyed. ' 2. Life expectancy for Windseal and Windseal Jets is likely to be less than the 15-year guarantee period, particul ar?.: in the southeast.- 1-. Competitive shingles are behaving similarly. 4. Shingles sealed by factory-applied adhesive and by roofappliec'. strips of adhesive are failing similarly. 5. Won-sealed shingles are not subject to horizontal type cracking. 6. Brittle coating may be contributing to the high rate of deterioration of sealed-down shingles. 7- Low proportions of saturant are contributing to swelling and shrinkage due to moisture absorption. This may be associated with the use of higher weight and variable kerosene felts. Tentative Sli jested Failure Mechanism The diurnal and seasonal increases and decreases in tempera ture and moisture cause shingles to expand and contract. When the exposed portions of these shingles are held by two hori zontal" lines of restraint, the expansion results in a bowing upv;ard between the lines. This results in sufficient strain to cause minute cracks. The cracks loosen the granules immediately adjacent to them, increasing the rate of attack by the sun. The cracks widen. Water reaches the felt. The extent of the expansion-contraction cycle increases. Perman ent contraction occurring with each drying'period finally results in sufficient strain to pull the felt apart. The thicker the coating, the greater is the strain at the surface for any degree of flexing, and thus the more rapid the crackin.., and subsequent failure. hecO'.imended Action Further investigation appears to be imperative.