The Effects of Hail on Metal Roofing Systems

The Effects of Hail
on Metal Roofing Systems
Jim D. Koontz, RRC, PE, and Troy L. White, PE
Jim D. Koontz & Associates, Inc.
3120 North Grimes, Hobbs, NM 88240
Phone: 575-392-7676 • Fax: 575-392-7602 • E-mail: jim@jdkoontz.com and troy@jdkoontz.com
2 9 t h RC I I n t e r n a t i o n a l C o n v e n t i o n a n d T r a d e S h ow • Ma rc h 2 0 – 2 5 , 2 0 1 4 K o o n t z a n d Wh i t e • 1 1 3
Abstract
The purpose of this study is to evaluate the effects of hail size on metal roofing and its
ability to shed water over an extended period of time. Substantial indentation at side laps,
end laps, or at the juncture with concealed clips can result in conditions that may affect
water-shedding capabilities. Commonly used metal roof panels will be impacted by variably
sized ice spheres at multiple locations across the panel per National Bureau of Standards
23. Panels will then be visually inspected and subjected to laboratory testing to determine
effects on water-shedding capability and coating resilience and performance.
Speakers
Jim D. Koontz, RRC, PE – Jim D. Koontz & Associates, Inc.
Jim Koontz, president of Jim D. Koontz & Associates, Inc., is a graduate of Tulane
University with a bachelor of science degree in engineering and a master’s in business
administration. Koontz has been involved in the roofing industry since 1960 and began testing
roofing material in 1976. He has experience as a roofer, estimator, consultant, lecturer,
researcher, and expert witness. Koontz was first published in 1984 and has numerous
articles relating to roofing material/product research to his credit.
Troy L. White, PE – Jim D. Koontz & Associates, Inc.
Troy White received a master’s degree in civil engineering from Texas Tech University in
Lubbock in 2002 and became a registered professional engineer in Texas and New Mexico
in 2005 and 2008, respectively. He has worked as a construction observer, designer, and
project manager in civil and architectural firms since March 2000. White recently became
an associate with Jim D. Koontz & Associates, Inc. and currently serves as a construction
observer, field investigator, and research associate within the firm. His experience ranges
from industrial and public works infrastructure to roofing systems design, construction
observation, and damage and failure analysis.
1 1 4 • K o o n t z a n d Wh i t e 2 9 t h RC I I n t e r n a t i o n a l C o n v e n t i o n a n d T r a d e S h ow • Ma rc h 2 0 – 2 5 , 2 0 1 4
Abstract
The purpose of this study is to evaluate
the effect of hail on metal roofing. Metal
roofing systems utilized on both steep- and
low-sloped roofs constitute a major market
share of the roofing industry. The function
of metal roofs obviously can vary from a
simple water-shedding capacity that protects
the building, to wind resistance, fire
resistance, aesthetics, and potentially some
hail resistance.
Within a substantial area of the United
States, hail occurrence is common (refer to
the hail map1 in Figure 1). Dependent upon
geographical location, metal roofs are often
impacted by hail, resulting in permanent
indentations. Significant insurance claims
are filed as a result of these indentations,
with disputes arising as to whether the
indentations constitute actual physical loss
or damage. Substantial time and expense
are expended to resolve these claims. The
extent to which these indentations affect
the functional attributes of the metal roof
system—including water-shedding capacity,
wind resistance, aesthetic value, material
longevity, and corrosion resistance—is
a common issue in these claims debates.
In exceptionally rare cases, metal roofs
impacted with very large hail may split
at the site of impact.
Substantial indentations
at side laps, end laps, or
at the juncture with concealed
clips can result
in conditions that may
affect water-shedding
capabilities of a metal
roof. Minor indentations
may have little to
no effect on roof performance
over an extended
period of time. Aesthetics
of hail indentations to
metals roofs also come
into consideration when
assessing reported damage.
To evaluate the effect of hail impact on
various metal roofing systems, two methods
of impacting metal roofs were utilized. Both
ice and steel spheres were used to impact
selected metal roofing systems. A pneumatic
launcher was used to fire ice spheres at
metal roofs. The ice spheres were propelled
at velocities listed by the National Bureau
of Standards (NBS) Building Science Series
No. 23.2 The metal roofing systems were
also impacted with steel spheres by methods
listed by Underwriters Laboratories
(UL)3 and by procedures listed in the marketing
literature published by the metal
roofing manufacturers.
Some metal samples indented during
naturally occurring hail events were also
examined and impacted. Comparisons were
made of naturally occurring hail indentations
and indentations that were a result of
laboratory testing. Observations were made
of differences between impacting a metal
roof with steel spheres or with ice.
Intr oducti on
Mankind has used various types of metal
roofing for thousands of years. Historically,
the first metals used included both lead
and copper. In some cases, the roofs had
a service life of over a century. The temple
in Jerusalem was reportedly covered with
a copper roof in 970 BC.4 In more recent
years, corrugated metal and various types
of configurations of metal roofing have
been utilized. Copper, although still widely
used in some applications, has been more
commonly replaced with coated-steel metals.
The coating over the steel is to provide
enhanced corrosion resistance. Today,
metal roofs are commonly coated with galvanized
zinc, aluminized steel, aluminumzinc
coatings, and a variety of polymer
coatings and films.
Two basic configurations of metal roofing
are widely used today: metal roofing
mechanically attached with exposed fasteners,
and variations of standing-seam metal
roofing secured with concealed clips. Metal
roofing can be installed on slopes as low as
¼ in. per ft.
Manufact urers ’ Mar keti ng
Literat ure
In 1998, the Texas Department of
Insurance (TDI)5 instituted a program in
which homeowners would receive discount
insurance rates if their residences were
covered with hail-resistant roofing. The
reduction in insurance rates does not apply
to commercial structures.
The methodology used by
TDI to determine hail resistance
consists of impacting
roofing products according
to UL test procedure 2218.
The UL 2218, Class 4 test
requires dropping a 2-in.-
diameter steel sphere from a
height of 20 ft. The resultant
impact energy is equivalent
to the impact from a 2-in.
hailstone falling at terminal
velocity. The UL test requires
two impacts at the same
location. After impact, visual
damage observation is to
be made with 5x magnification.
The acceptance criteria,
The Effects of Hail
on Metal Roofing Systems
2 9 t h RC I I n t e r n a t i o n a l C o n v e n t i o n a n d T r a d e S h ow • Ma rc h 2 0 – 2 5 , 2 0 1 4 K o o n t z a n d Wh i t e • 1 1 5
Figure 1 – Hail climatology map.
according to UL 2218, are as follows:
The report requires measurements
of the depth of depression and a
determination of any tearing, fracturing,
cracking, splitting, rupture,
crazing, or other evidence of damage
to the roofing system. A pass/fail
determination is then made.
Numerous metal roofing manufacturers
have submitted their products for testing
and are marketing their products for compliance
with UL 2218. The TDI6 website lists
28 roofing manufacturers that represent
their metal roofing products as passing UL
2218, Class 4.
Metal roofing manufacturers may also
market their roofing systems as being
in compliance with Factory Mutual (FM)
4471.7 Under FM 4471, two steel-sphere
impact tests are utilized (see Table 1). The
impact energy of the resultant FM Severe
Hail rating is less than the UL 2218, Class
4 impact energy (4 ft-lbf. [foot-pound force] versus 23.71 ft-lbf). The FM Moderate Hail
rating is slightly higher than the UL Class
2: 8 ft-lbf. versus 7.35 ft-lbf.
Roof systems that pass the tests are
rated either Moderate Hail- or Severe Hailresistant.
FM also allows the use of ice
sphere testing according to FM 4473.8
Class 1 panel roofs shall be able to
withstand the effects of hail. Panels
shall show no evidence of puncture
or chipping, peeling, blistering,
cracking, or crazing of the coating
when examined under 10x magnification.
Some metal manufacturers warrant that
their products will be hail-resistant. These
manufacturers define hail damage as penetration
completely through the panel or
cracks/splits of the panel’s steel substrate
around the point of impact. Other metal
manufacturers may list a resistance to
hail up to hailstones less than 2.5 inches
in diameter. The claimed hail resistance of
some of these roofing systems is obviously
a part of the manufacturer’s marketing
features.
Buildi ng Codes
Per the International Building Code
(IBC)9 Section 1502, “roof covering” is
defined as “the covering applied to the roof
deck for weather resistance, fire classification,
or appearance.”
Under Section 1504, paragraph 1504.7,
roof coverings installed on low-slope roofs
(roof slope <2:12) shall resist impact damage
based on the results of test procedures
developed by The American Society for
Testing and Materials10 (ASTM) D3746 and
the Canadian General Standards Board11
(CGSB) 37-GP-52M.
ASTM D3746 and CGSB 37-GP-52M
involve impacting a roof system with a steel
dart dropped from a prescribed height. The
ASTM laboratory procedures were originally
developed for testing built-up roofing.
The Canadian test was developed for both
bituminous roofing and single-ply roofing.
The testing has been utilized for all types
of roofing. It is not entirely clear that this
part of the IBC would be applicable to metal
roofing.
The impact energy described in ASTM
D3746 and UL 2218, Class 4, are very
similar. ASTM D3746 uses a 2-in.-diameter,
5-lb. dart dropped from 4 ft., 5 in. that
generates impact energy of 22 ft-lbf. The
UL 2218 Class 4 uses a 2-in. steel sphere
dropped from 20 and develops impact energy
of 23.71 ft-lbf.
Field Studies of Hail Events
Data derived from real-world hail events
are somewhat limited. The Roofing Industry
Committee on Weather Issues (RICOWI)12
has conducted two Hail Investigation
Programs (HIP) following hail events. The
first HIP investigation was in April 2004
in Oklahoma City, Oklahoma (OKC). Jim
D. Koontz & Associates, Inc. (JKA) participated
in the OKC investigation conducted
by RICOWI.
Six metal roofs were examined. The
metal roofs were impacted by hail reported
to be 1 to 2.5 inch in diameter. Dents were
also observed on roofs impacted with hail
1.5 inches or larger.
On these roofs, the hail-caused
dents were found to be a cosmetic
issue, with no functional damage
to the paint or the metal plating.
On exposed fastener systems, there
were no instances found of fasteners
loosened by hailstone impacts.
Panel joints had not been distorted
sufficiently to affect the watershedding
ability of the panels.
A second RICOWI HIP program was
conducted in May 2011 in Dallas, Texas.
Sixteen metal roofs were examined during
this investigation. Hail 1 to 4 inches
in diameter was reported during this hail
event. Nine of the 16 roofs did not exhibit
dents in the metal systems. No fractures,
spalling, or punctures occurred on the
metal panels. There was also no evidence of
leakage within any of metal roofs.
Lab orat ory Testi ng
Ten metal roofing targets made of corrugated,
standing-seam, and R-panel roofing
of different gauges were constructed for test
purposes. Each of the metal roofing systems
was assembled over a 5-ft.-wide purlin
system similar to structural supports commonly
used in the construction industry.
1 1 6 • K o o n t z a n d Wh i t e 2 9 t h RC I I n t e r n a t i o n a l C o n v e n t i o n a n d T r a d e S h ow • Ma rc h 2 0 – 2 5 , 2 0 1 4
FM 4471
Rating Height of Drop Diameter of Impact
Steel Sphere Energy
Class 1 – Severe hail 17 feet 9.5 inches 1.75 inches 14 ft-lbf.
Class 1- Moderate hail 5 feet 2.0 inches 8 ft-lbf.
Table 1 – FM 4471.
Table 2 – Steel and ice sphere testing results.
Steel Sphere Testing Ice Sphere Testing
1.5-in.-diameter 2-in.-diameter 1.5-in.-diameter 2-in.-diameter
UL Class 2 UL Class 4 NBS 23 NBS 23
KE KE KE KE
7.35 ft-lbf 23.71 ft-lbf 7.35 ft-lbf* 23.71 ft-lbf*
*Impact energy +10%
The ten targets of metal roofing were
impacted using both 1.5- to 2-in.-diameter
steel spheres and 1.5- to 2-in.-diameter ice
spheres (refer to Table 2). The ice spheres
had a density of .91 g/cm3. The steel sphere
testing was per UL 2218 Class 2 and Class
4 test standards. The steel sphere impact
testing involves dropping a steel sphere
from a prescribed height to generate a
given kinetic energy (Figure 2). Since a steel
sphere of an exact weight is dropped from
the same height each time, the specific
kinetic energy produced is very reproducible.
The UL procedure calls for impacting
the same location twice. For the purposes of
this study, only one impact was performed
in order to have
a comparison
to a single ice
sphere impact.
The ice
sphere testing
was per NBS
No. 23 standards.
The ice
sphere impact
method propels
an ice sphere
with the use
of a pneumatic
launcher, as
seen in Figure 3. The weight of each ice
sphere is initially recorded. As the ice
spheres are propelled at the target, their
speed is recorded with a ballistic timing
device. The kinetic-impact energy of each
ice sphere is then calculated. The kinetic
energy of the ice spheres may vary by a
+10% kinetic energy value. The metal roof
targets are listed in Table 3.
Depending on the type of
metal roof target, impact locations
included flat locations, tall rib, and
short rib. If a metal roofing manufacturer
represents that its roofing
passes the UL and FM tests, then
an impact at any location within
the metal roofing is acceptable.
Following impacts, the indentation
diameters and depth of indentations
were measured for each type
of impacting steel or ice or sphere
(refer to Figures 4 and 5). The
surface of each sample target was
examined for cracking or damage
to the coating.
2 9 t h RC I I n t e r n a t i o n a l C o n v e n t i o n a n d T r a d e S h ow • Ma rc h 2 0 – 2 5 , 2 0 1 4 K o o n t z a n d Wh i t e • 1 1 7
Panel ID Gauge Configuration Panel Type
A 26 R-panel 36 in. Steel prefinished
B 24 Corrugated 36 in. Galvanized
C 24 Standing seam 16 in. Steel prefinished
D 24 Standing seam 18 in. Steel prefinished
E 26 R-panel 36 in. Galvalume
F 26 R-panel 36 in. Steel prefinished
G 25 Standing seam 16 in. Steel prefinished
H 32 Corrugated 22 in. Galvanized
I 24 R-panel 36 in. Galvalume
J 24 Standing seam 24 in. Galvanized
Figure 2 – Steel sphere. Table 3 – Metal roof targets.
Figure 3 – Pneumatic launcher and target.
Figure 4 – Cross section metal
indentation.
Figure 5 –
Indentation
measurement with
micrometer.
A typical ice sphere impact is depicted
in Figure 6. The resulting data from the
steel and ice sphere impacts is recorded in
Attachment 1.
There is a correlation between indentation
depth and dent diameter. The indentation
obviously varies depending upon gauge,
type of metal roof, and diameter of impacting
sphere. Refer to Figures 7 through 10.
The higher the impact energy (i.e., the
greater the diameter of hail or steel sphere),
the wider and deeper the indentation.
Lab orat ory Obser vati ons
• With the exception of a 32-gauge
corrugated metal penetration, splitting
of the metal did not occur with
any of the metal roofing tested per
the UL and NBS impact procedures.
• Indentations resulted in all metal
roofing tested when impacted by
either ice or steel spheres.
• The depth and width of dents varied,
depending upon impact location,
gauge, and type of metal. For the
most part, the characteristics of the
dents, diameter,
and depth are
generally random.
There are
some differences
between steel
and ice sphere
impacts; however,
it does appear
that with highdensity
ice
spheres, the degree
of indentation
is slightly
higher than that
observed with
steel spheres.
• The random
results of indentation, diameter,
and depth in two metal roofs with
the same gauge is mostly like due
to variation in yield strength in
kilopound force per square inch
(ksi = 1,000 psi). Most roofing metals
range from 33 ksi to 80 ksi.
Prior research13 has documented
that higher-yield strengths result in
metal roofs less vulnerable to indentation
from hail impacts.
• The steel sphere impact testing did
damage the coatings of some metal
roofing panels. Ice sphere impact
testing did not damage the coatings
of the metal tested.
1 1 8 • K o o n t z a n d Wh i t e 2 9 t h RC I I n t e r n a t i o n a l C o n v e n t i o n a n d T r a d e S h ow • Ma rc h 2 0 – 2 5 , 2 0 1 4
Figure 6 – Typical ice sphere impact.
Figure 7 – 1½-in. ice sphere.
Figure 9 – 1½-in. steel sphere.
Figure 8 – 2-in. ice sphere.
Figure 10 – 2-in steel sphere.
Long-Term Experie nce
Over the last 50 years, the staff of Jim
D. Koontz & Associates, Inc. has examined
thousands of metal roofs impacted by hail.
This has included all types of metal roofs
commonly used in the roofing industry.
For purposes of this study, five typical
metal roofs impacted by numerous hail
events of varying size of hail are reviewed:
Roof 1: Hobbs, New Mexico
The 24-gauge metal roof of standing
seam, two foot on center, was installed in
1980 (see Figures 11 and 12). Multiple hail
events have occurred at this location. Minor
dents observed have not resulted in any
apparent deterioration or loss of functional
or aesthetic value.
Roof 2: Hobbs, New Mexico
The 24-gauge, R-panel metal roof was
installed on an office area and warehouse in
1980. These metal roofs have been exposed
to multiple hail events over 30 years. Minor
dents in the metal roofs have not resulted in
any premature failure, corrosion, or loss of
functional or aesthetic value.
Roof 3: Dora, New Mexico
The corrugated metal Quonset hut
building has been in place for approximately
40 years. Although subject to numerous
hail events over the years resulting in
numerous indentations, premature failure
from corrosion did not occur. The roof was
replaced following a 2.5-in. hail that distorted
side and end laps, resulting in leakage
during wind-driven rain events.
Roof 4: Plainview, Texas
Multiple R-panel and corrugated metal
roofs installed in the 1950s and 1960s at a
warehouse location were examined. The various
buildings have been subjected to numerous
hail events of various magnitudes over
the last 60 years. Some surface corrosion
was observed at various locations not related
to hail impacts. Corrosion or deterioration of
the metal roofs at hail impact points was not
observed at any location.
Roof 5: Lovington, New Mexico
Metal roofs consisting of both R-panels
and corrugated metal have been subjected
to numerous hail events over the last 40
years. Minor dents in the metal roofs have
not resulted in corrosion, deterioration, or
premature failure.
During this 50-year time period, several
observations on metal roofs have been made
by Jim D. Koontz & Associates:
2 9 t h RC I I n t e r n a t i o n a l C o n v e n t i o n a n d T r a d e S h ow • Ma rc h 2 0 – 2 5 , 2 0 1 4 K o o n t z a n d Wh i t e • 1 1 9
Figure 11 – 24-gauge metal roof
installed in 1980.
Figure 12 – Minor dents
have not resulted in any
apparent deterioration.
• Fewer than ten metal roofs have
actually been punctured or split by
hail impact. Obviously, very large
hail is required to actually puncture
or split a metal roof.
• Long-term corrosion or deterioration
of metal roofs at the points of impact
has not occurred. Minor impacts or
dents have no effect on the longterm
performance of metal roofing
systems.
• H ail impact dents have to be of
significant size for side laps or end
laps to be distorted to the point
that leakage will occur during winddriven
rain events. Standing seams
have to be impacted to the point that
some distortion occurs from thermal
expansion and contraction at concealed
clips.
Opinions
• The impact test procedures for both
FM and UL should be better defined
for metal roofing. Criteria for precise
points of impacts, measurements
of dents, and visual examination
should be clarified.
• Pass/fail criteria should be better
defined by FM, UL, and by metal
roofing manufacturers.
• Metal roofing manufacturers should
provide consumers with test data
documenting FM and UL compliance.
• The use of steel spheres for impacting
metal roofing may result in surface
coating damage not consistent
with impacts from ice spheres with
the same kinetic energy.
• Consumers, contractors, and
designers rely upon information
that roofing products comply with
building code requirements for performance
criteria such as wind and
fire. Building code officials should
consider additional requirements to
ensure all roofing products—including
metal panel roofing systems—
will perform in reasonably expected
hail events.
• Naturally occurring hail dents in
metal roofing that are smaller than
dents created by test standards
listed in manufacturer marketing
literature should not be considered
damage, as the dents have no longterm
effect on the performance of
the metal roofing.
• Aesthetic concerns may be a consideration
for steep-sloped metal roofing
when hail dents are visible from
the ground. Aesthetics should not
be a consideration for low-sloped
commercial roofs.
• Additional research in this area
should be performed.
REFERENCES
1. Weather Forensics, Weather Decision
Technologies, 2003-10 Hail Climatology.
2. Sidney H. Greenfield, “Hail Resistance
of Roofing Products,” U.S.
Department of Commerce, National
Bureau of Standards, Build Science
Series 23, Washington, DC, August
1969.
3. Underwriters Laboratory, 2218,
“Impact Resistance of Prepared Roof
Covering Materials,” May 31, 1996.
4. C.W. Griffin and R.L. Fricklas, Manual
of Low-Slope Roof Systems, 2006.
5. Texas Department of Insurance,
Austin, TX, 1998.
6. Texas Department of Insurance website,
www.tdi.texas.gov, “Qualified
Metal Products.”
7. Factory Mutual Approvals, Class
Number 4471, Approval Standard
for Class 1 Panel Roofs, March 2010.
8. Factory Mutual Approvals, Class
Number 4473, Specification Test
Standard for Impact Resistance
Testing of Rigid Roofing Materials
by Impacting With Freezer Ice Balls,
July 2005.
9. International Building Code, 2009.
10. American Society for Testing and
Materials, D3746, Standard Test
Method for Impact Resistance of
Bituminous Roofing Systems, 1985.
11. Canadian General Standards Board,
37-GP-52M, August 1984.
12. Roofing Industry Committee on
Weather Issues, April 21, 2004,
2011.
13. United States Steel Corporation,
Technical Bulletin 2012.17.
1 2 0 • K o o n t z a n d Wh i t e 2 9 t h RC I I n t e r n a t i o n a l C o n v e n t i o n a n d T r a d e S h ow • Ma rc h 2 0 – 2 5 , 2 0 1 4
2 9 t h RC I I n t e r n a t i o n a l C o n v e n t i o n a n d T r a d e S h ow • Ma rc h 2 0 – 2 5 , 2 0 1 4 K o o n t z a n d Wh i t e • 1 2 1