Hail Resistant Roof: Hyload Roofing Membrane Holds Up Against Texas Hail
In 1987 it was time to put a new hail resistant roof on the Texas Workforce Commission, TWC, (formerly the Texas Employment Commission) building at 301 West 13th Street in downtown Fort Worth. TWC consulted with Armko Industries, Inc. regarding their best roofing choice. The decision was made to go with a hot-applied Hyload Coal Tar Elastomeric Membrane (CTEM) system, made with DuPont™ Elvaloy® KEE roofing resin.
Sixteen years ago the existing roof was torn off down to the structural concrete deck and a new CTEM built-up roof system was laid up as follows:
- 2-in.-thick polyisocyanuate insulation ribbon-mopped with hot asphalt to the deck
- 1/2-in.-thick perlite mopped in place with hot asphalt
- 2 plies of Type IV fiberglass felt mopped in hot asphalt
- 1 ply of Hyload H150E cap sheet mopped in hot asphalt
- Flood coat of hot coal tar
- Aggregate surfacing of washed river rock spread into the flood coat
From October 1987 through October 2003, numerous hail storms and other severe weather events passed through downtown Fort Worth. There were 12 readings with hail stones of 2-in. or larger. Notably, in May, 1995 over the span of 35 minutes, the 8 year-old roof was pummeled with hail stones between 2 1/2-in. and 4 in.. (Hail-fall data compiled from Computerized Hail Area Searches, CHAS, ©1997 by Haag Engineering Co.)
Hyload and Armko had been monitoring this roof through the years. Of obvious interest were the roof properties that contribute to successful long-term performance. When most commercial roof membrane systems are new, one can reasonably expect to get good results if cores are taken and tested. However, this roof has been in place for 16 years.
The natural and expected results of aging for roof membrane systems include increasing embrittlement and deterioration. Further, the presence of the underlying insulation will accelerate this embrittlement by holding higher temperatures in the roof assembly. Testing and evaluation of this roof would include examination of any residual impact damage, subjecting a test core to new impacts, evaluating the interply mopping asphalt, and testing the residual strength of the bituminous built-up roof membrane system.
In July, 1995 after the severe May storm, the cutting of a random core sample from the roof was witnessed and evaluated by Maxim/Southwestern Laboratories. The M/SwL report made particular note of softball and baseball size impacts that were apparent over the entire roof. It also stated that no penetrations due to hail impacts were detected in the roof field. The bituminous membrane system was de-saturated and scrutinized. No punctures, tears, or any other identifiable damage to the Hyload membrane or the glass felts were found. Further, the report stated "the interply bitumen appeared to be newly applied."
In October 2003, it had been 16 years since installation, 8 years since an extreme hail event, plus numberous other stoms that had passed through the area. Questions to be answered were:
- Was there any residual damage from the hail exposure this roof had experienced?
- Could this aged roof withstand significant impacts today and remain effective?
- After 16 years of service, was the general condition of the roof acceptable?
To answer the first two questions, in October, 2003 new cores were taken from the roof. One of the cores was sent to Haag Engineering in Carrollton, TX. Haag has developed an ice ball impact test that can be run in the controlled laboratory environment. Artificial hail stones are made in the freezer and shot at samples at a speed that corresponds to the terminal free-fall velocities of similar-sized, naturally occurring hail stones. Although laboratory techniques limit the size of the artificial ice balls to 2 1/2-in., the laboratory testing is more demanding than natural hail stones of the same size for two reasons:
- The artificial stones are denser than natural stones due to the differences in how they are formed. This higher density results in a higher energy at impact with the artificial stone.
- The laboratory impacts are made perpendicular to the sample which assures the maximum energy transfer. In actual hail events, the vast majority of the stones falling are impacting roof surfaces at an angle less than perpendicular due to the high winds that almost invariably accompany hail storms. This type of glancing impact does not deliver as much energy to the roof as a perpendicular impact.
The laboratory impacts are made perpendicular to the sample which assures the maximum energy transfer. In actual hail events, the vast majority of the stones falling are impacting roof surfaces at an angle less than perpendicular due to the high winds that almost invariably accompany hail storms. This type of glancing impact does not deliver as much energy to the roof as a perpendicular impact.
The Haag Engineering report on this testing concluded that there was no latent damage of any kind after sixteen years of exposure. Haag also concluded that impacts by 2 1/2-in. simulated hailstones against the Hyload core where ballasted with gravel left the Hyload 150E membrane and underlying reinforcements and insulations intact.
Both the July, 1995 and the October, 2003 evaluations performed on cores from the roof provide strong evidence of the ability of an aggregate-surfaced Hyload built-up roof system to withstand hail impacts for many years without punctures or tears. A remaining question was whether the general condition of the roof was still acceptable after 16 years in service.
Although built-up roofing systems have been in use for decades, there is no published standard in existence that contains performance properties for this type of roof. However, in November, 1974 the National Bureau of Standards published a paper titled "Preliminary Performance Criteria for Bituminous Membrane Roofing". The authors tested and evaluated both good performing and poor performing 4-ply built-up roofs and analyzed the results. A recommendation that came out of their study was that the tensile strength of a built-up roof should not be less than 200 lb/in in the weakest direction when tested at 0°F.
Note: The manufacture of roofing materials often results in the "machine direction" of the material having a higher tensile strength than the "cross-machine direction. This is the reason for the stipulation of the "weakest direction" requirement in the proposed criteria. Another test core from the TWC roof was sent to PRI Asphalt Technology in Tampa, FL for evaluation of the tensile strength property of the membrane system. When the roofing membrane was tensile tested at 0°F the results were 402 lb/in. in the machine direction and 275 lb/in. in the cross-machine direction. These results, exceeding the proposed requirements by a minimum of 37%, provide evidence that this roof 16-year-old insulated roof assembly has maintained good strength properties.
To evaluate the condition of the interply asphalt, a section of a test core from the roof was sent to Trumbull Asphalt, the major supplier of mopping asphalt. Years of experience in roof asphalt evaluations reveal that interply asphalt will age (harden) when in place on the roof, but the aging should limit the rise in softening point to approximately 240°F and the drop in penetration to 6 dmm when tested at 77°F. When tested at the Trumbull laboratory, the interply asphalt from the test core results were 228°F softening point and 13 dmm penetration at 77°F, results that fall within normal expectations. The Hyload 150E cap sheet protected the interply asphalt and kept it "alive" even though the combination of insulation and solar-induced thermal loading provided significant aging influence for 16 years. (For those wanting more information on softening point and penetration testing, please refer to ASTM standards D36 and D5, respectively.)
A summary of the evaluation of the 16 year-old TWC roof is as follows:
- The roof is in place and currently performing well.
- Evaluation of a sample core shows there is no residual damage from hail exposure, including the 4-in. hail that occurred in 1995 when the roof was 8 years old.
- Controlled and measured impacts of 2 1/2-in. artificial hail stones on a 16-year-old core produced no damage to the membrane system.
- The strength of the membrane system exceeds recommended values by over 37%.
- The interply mopping asphalt has exhibited no more than normal aging in this insulated roof system.
Roofing Manufacturer Link
Hyload Roofing Systems » Visit HyLoad Roofing Systems Web site