Design of Nonballasted Low-Slope Roof Assemblies for Wind Resistance – The Current Situation and Recommendations for the Future

Design of Nonballasted Low-Slope Roof
Assemblies for Wind Resistance—the Current
Situation and Recommendations for the Future
Stephen Patterson, RRC, PE
Roof Technical Services, Inc.
1944 Handley Dr., Fort Worth, TX 76112
Phone: 817-496-4631, Ext. 102 • Fax: 817-496-0892 • E-mail: spatterson@rooftechusa.com
Madan Mehta, PE, PhD
University of Texas at Arlington
601 W. Nedderman Dr., Arlington, TX 76019
Phone: 713-842-0393 • E-mail: mmehta@uta.edu
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 P a t t e rs o n a n d Me h t a • 1 6 9
Abstract
This paper provides an historical perspective of wind design of roofs and an analytical
basis for the development of a design standard that meets the requirements of the
International Building Code and ASCE 7. It traces the development of wind design standards
from the early days of Factory Mutual, the building codes, and ASCE 7, as well as providing
an analysis of the current wind design standards and compliance with code. Finally,
it reviews the history of the safety factors involved in wind design for roofs and provides
recommended design standards based upon the code-required loads with safety provisions
consistent with historical design standards.
Speakers
Stephen Patterson, RRC, PE — Roof Technical Services, Inc., Fort Worth, TX
Mr. Pattterson is a registered professional engineer, a registered roof consultant,
and the president of Roof Technical Services, Inc., Fort Worth, Texas, an engineering firm
that specializes in roofing and waterproofing. Mr. Patterson has been heavily involved in
the design, testing, and inspection of roofs, as well as forensic engineering related to roofs
since 1973. He has also been organizing and presenting seminars and courses on roofing
at the University of Texas at Arlington. Patterson is the coauthor of the following books and
monographs: Roofing Design and Practice, published by Prentice Hall, 2001; Roof Drainage,
published by the RCI Foundation; and three editions of the monograph, Wind Pressures on
Low-Slope Roofs, published by the Foundation.
Dr. Madan Mehta, PE — University of Texas at Arlington – Arlington, TX
Dr. Madan Mehta is a professor in the School of Architecture at the University of
Texas at Arlington and teaches courses in structures and building construction. He is a
licensed professional engineer and author of several full-length books on architectural engineering.
He is also a coauthor (along with Stephen Patterson) of the following books and
monographs: Roofing Design and Practice, published by Prentice Hall, 2001; Roof Drainage,
published by the RCI Foundation; and three editions of the monograph, Wind Pressures on
Low-Slope Roofs, published by the RCI Foundation.
1 7 0 • P a t t e rs o n a n d Me h t a 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
1. INTRODUCTION
This paper follows the authors’ work on
the third edition of the monograph titled
Wind Pressures on Low-Slope Roofs,1 published
by the Roof Consultants Institute
Foundation (RCIF) in March 2013. The
monograph is keyed to the 2010 edition
of the American Society of Civil Engineers
(ASCE) standard titled Minimum Design
Loads for Buildings and Other Structures,2
referred to as “ASCE 7 standard.”
A U.S. standard that has an international
recognition, ASCE 7 provides guidance
for the determination of various types of
loads on buildings and nonbuilding structures.
It is referenced by the International
Building Code (IBC), the model code on
which the building codes of almost all local
jurisdictions in the U.S. are based. For the
specific case of designing roof assemblies to
resist wind pressures, the IBC requires that
the wind pressure be determined in accordance
with ASCE 7 standard, as mentioned
in the following excerpts:
2006/2009/2012 IBC, Section
1504.3 (Wind Resistance of Nonballasted
Roofs) states that the “roof
coverings installed on roofs…shall
be designed to resist the design
wind load pressures for components
and cladding in accordance with
Section 1609.” Section 1609.5.1
(“Roof Deck”) states, “The roof deck
shall be designed to withstand the
wind pressures determined in accordance
with ASCE 7. Section 16.9.5.2
(“Roof Coverings”) states that “roof
coverings shall comply with Section
16.9.5.1.”3,4,5
Because building codes are legal documents,
the reference in them to ASCE 7
standard gives this standard a legal status.
Therefore, the roof assemblies are required
to be designed for wind pressures obtained
from the use of the current edition of ASCE
7 standard.
In the 2010 edition of the ASCE 7 standard
(called “ASCE 7-10 standard”), substantial
changes were made in wind load
provisions. One of these changes relates to
the specification of the basic (design) wind
speed—from service-level (also called “nominal”)
wind speed to the strength-level (also
called “ultimate”) wind speed. This paper
examines the impact of this change on
the procedure(s) described in wind design
standards for nonballasted (i.e., adhered
or anchored) low-slope roofs. (The terms
“nominal,” “service-level,” “strength-level,”
and “ultimate” are elaborated in Section 6
of this paper.)
2. P ROCEDURE F OR T HE D ESIGN
OF WIND-RESISTANT LOW -SLOPE
ROO FS
Like all structural design, the design of
a low-slope roof for wind resistance is a twopart
process. The first part of the design
process consists of determining the wind
pressures on the roof. This determination
is accomplished through calculations based
on the building’s location, its dimensions,
envelope properties, and the roof geometry.
The second part of the design process
involves the selection of an appropriate roof
assembly whose uplift resistance equals
or exceeds the calculated wind pressures
on the roof. Safety margin is included in
the design to account for the uncertainties
inherent in the assessment of wind pressures
and in determining the resistance
of roof assemblies. The term “selection” is
used (not “calculation”) because under the
current industry practice, wind resistance
of a low-slope roof assembly cannot be calculated
using standard structural analysis
and design procedure similar to that used
for the design of the structural elements of
a building.
Instead, the roof assembly with the
required wind resistance is “selected” from
among several manufacturers’ assemblies
that have been tested for wind resistance
by an independent agency. The tests are
typically conducted on full-scale specimens
of roof assemblies as per standard test procedures.
3. STANDARDS FOR THE DESIGN
OF WIND-RESISTANT LOW -SLOPE
ROO FS
The two-part design process (wind uplift
calculations and the selection of roof assembly)
is theoretically simple but confusing in
practice. The most important factor contributing
to the confusion is the absence of
a design standard that has legal status by
virtue of its being adopted or referenced by
the building codes.
Currently, two standards for wind
design of low-slope roofs exist in the United
States. These are:
1. FM Global document, Property Loss
Prevention Data Sheets 1-286
2. ANSI/SPRI document, ANSI/SPRI
Wind Design Standard Practice for
Roofing Assemblies7
The FM Global document has a long
history and has been updated several times.
Its latest revision was published in 2012.
The current ANSI/SPRI standard was published
in 2012, preceded by its 2007/2008
version.
Note that ASCE 7 standard is not a
comprehensive design standard. It is a load
standard and addresses the first part of
the design process by providing minimum
values of loads on a building and its components
and hence does not provide any guidance
for the second part of the process—
specification of safety factor, specification
of tests used for determining the strength
of the assembly, guidance for anchorage
or adhesion of assembly components to
the roof deck, and so on. This part is the
responsibility of the material industry—in
our case, the roofing industry.
By comparison, both ANSI/SPRI and
FM Global documents are comprehensive
design standards because they address
both parts of the design process. Although
Design of Nonballasted Low-Slope Roof
Assemblies for Wind Resistance—the Current
Situation and Recommendations for the Future
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 P a t t e rs o n a n d Me h t a • 1 7 1
the two documents share several similarities,
they also contain several differences,
so the design obtained from the two documents
is generally quite different.
While this is one cause of confusion, a
greater sense of confusion results from the
fact that neither standard is referenced in
the building codes. Consequently, a roof
designer may be at a loss to determine
whether the design based on any one of
these documents is below-code, per-code,
or above-code.
Wind Pressures Obtained From FM
Global and ANSI/SPRI Standards
Part 1 of the design process in both FM
Global and ANSI/SPRI documents is loosely
based on the ASCE 7 standard, so that
the roof pressures obtained from the two
documents are different from each other
and also different from those obtained from
ASCE 7. Consequently, both documents
lack full compliance with ASCE 7 standard.
As an illustration of the differences,
the field-of-roof wind pressures for a hypothetical
building, as obtained from ASCE
7, FM Global, and ANSI/SPRI documents,
are shown in Table 1. The eave height of
this hypothetical building, with a low-slope
roof (slope ≤ 7°) is 100 ft. in Exposure C.
Roof wind pressures in the field of roof
have been shown for Dallas, Texas (nonhurricane-
prone region), and New Haven,
Connecticut (hurricane-prone region), for
Risk Categories, I, II and III.
Factor of Safety
A factor of safety (FOS) is applied to the
calculated design pressure to obtain the
required minimum wind pressure resistance,
i.e., the required test strength of roof
assembly (STR), from the following expression.
Thus:
STR ≥ (FOS)(DWP)
Where DWP is the calculated design
wind pressure on the roof. Both design
standards (ANSI/SPRI and FM Global) use
FOS = 2.0. Hence, for an adequate design:
STR ≥ (2.0)(DWP)
Thus, if the calculated DWP on a roof
zone is 40 psf, the required minimum
strength of the tested assembly used for
that zone must equal 2.0(40) = 80 psf.
It may be instructive to note that
FOS = 2.0 is endorsed by ASTM D6630,
Standard Guide for Low-Slope Insulated
Roof Membrane Assembly Performance,
which recommends a “minimum” factor of
safety = 2.0. However, as explained in the
following section, the effective values of
FOS used in both roof design standards is
greater than 2.0.
4. E FFECTIVE FACTORS OF S AFETY
USED IN DESIGN STANDARDS
When roof design is accomplished per
FM Global or ANSI/SPRI standards, the
effective FOS is higher than the 2.0 implied
in these standards, because the wind pressures
used in them are higher than the
corresponding pressures given by ASCE
1 7 2 • P a t t e rs o n a n d Me h t a 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
Location ASCE 7-10 ANSI/SPRI (2012) FM Global (2012)
Dallas, TX Risk Cat. I (-28.8 psf) Risk Cat. I (-33.6 psf*)
Risk Cat. II (-34.5 psf) Risk Cat. II (-40.4 psf) Min. acceptable design (-41.0 psf)
Risk Cat. III & IV (-37.6 psf) Risk Cat. III & IV (-44.0 psf*) Enhanced design (-59.0 psf)
N ew Haven, CT Risk Cat. I (-34.5 psf) Risk Cat. I (-40.4 psf)
Risk Cat. II (-40.8 psf) Risk Cat. II (-47.9 psf*) Min. acceptable design (-61.0 psf)
Risk Cat. III & IV (-47.6 psf) Risk Cat. III & IV (-55.8 psf*) Enhanced design (-72.0 psf)
Roof height = 100 ft.; Exposure C; Enclosed building; Effective wind area = 10 sq. ft.; Roof slope ≤ 7°;
No topographical feature (Kzt = 1.0); Parapet height ≤ 3 ft.
Table 1 – Comparison of field-of-roof wind pressures on a hypothetical low-slope roof.
Notes:
1. Negative sign implies uplift pressure.
2. ASCE 7 values have been obtained by multiplying ultimate wind pressures (using Part 3, Chapter 30 of ASCE 7-10) by 0.6 to
convert them to nominal pressures in order to provide a fair comparison with ANSI/SPRI and FM Global values, which are
nominal pressures.
3. Wind speeds for ASCE 7-10 and ANSI/SPRI are 105 mph, 115 mph, and 120 mph for Risk Category I, Risk Category II, and Risk
Category III-IV buildings respectively for Dallas, TX. For New Haven, CT, the corresponding wind speeds are 115 mph, 125 mph,
and 135 mph.
4. Wind pressures, given in this table under the column for ASCE 7-10, can be obtained from calculations or from the tables
provided in Reference 1.
5. F M Global values are based on basic wind speed of 90 mph for Dallas, TX, and 110 mph for New Haven, CT, for Minimum
Acceptable Design category. The corresponding values for Enhanced Design category are 108 mph and 120 mph (Ref. 6, p. 5).
6. Values marked with an asterisk have been obtained by interpolation or extrapolation of the values given in ANSI/SPRI standard.
7. L arge wind pressure values for New Haven, CT, as compared with the corresponding values in ANSI/SPRI are due to the
significant reduction in basic wind speeds in hurricane-prone regions of the U.S. in ASCE 7-10, as compared with those of ASCE
7-05.8 ANSI/SPRI (2012) is based on ASCE 7-10 basic wind speeds, while FM Global (2012) values are based on ASCE 7-05 basic
wind speeds. Note, however, that the lower design wind pressures based on ASCE 7-10 in hurricane-prone regions may not be
obtained if the building is exposed to coastline, because ASCE 7-10 has reintroduced Exposure D for such locations in place of
Exposure C in ASCE 7-05.
Equation 1
Equation 2
7, as shown in Table 1. In the ANSI/SPRI
standard, the effective FOS is approximately
2.35 for all risk categories of buildings.
This is because the roof pressures used
are approximately 1.175 times the ASCE
7 roof pressures, giving an effective FOS of
(2.0)1.175 = 2.35.
In other words, an additional safety
margin is embedded in roof pressures given
in the ANSI/SPRI standard. This additional
safety margin is due to ANSI/SPRI disregarding
the use of the wind directionality
factor, Kd = 0.85. The wind directionality
factor accounts for the extremely low probability
that the peak (i.e., the design) wind
speed will come from the least favorable orientation
of the building or the building component.
Disregarding Kd inflates the roof
pressures by a factor of (1/0.85) = 1.175.
While the effective FOS in the ANSI/
SPRI standards is 2.35 in all situations, in
FM Global standard, it follows a complex
pattern because:
1. The FM Global standard is based on
the earlier (2005) version of ASCE 7
standard (ASCE 7-05).
2. FM Global disregards risk category
classification of buildings and treats
all buildings as belonging to the
riskiest category (Risk Category III
and IV of ASCE 7-05). This disregard
increases the design wind pressure
by 1.15 for a Risk Category II building
and by 1.32 for a Risk Category
I building of ASCE 7-05.
3. FM Global uses Kd = 0.85, as given
by ASCE 7.
4. FM Global has established two
design categories: (a) Minimum
Acceptable Design and (b) Enhanced
Design. These categories are unique
to FM Global. The design wind
speeds for the Minimum Acceptable
Design category are generally the
same as the basic wind speeds in
ASCE 7-05, but are higher for the
Enhanced Design category.
The approximate values of effective FOS
used in both design standards are given in
Table 2.
5. COMPLIANCE OF ROO F DESIGN
STANDARDS WITH BUILDING
CODES—RECOMMENDATION S
UN DER THE PRESENT FORMAT
Wind-resistant roofing design should
use the same design procedures that are
employed in the structural design of buildings
in concrete, steel, wood, and masonry.
The design standards developed by the
respective associations of structural material
industries (American Concrete Institute,
American Institute of Steel Construction,
American Wood Council, and Masonry
Standards Joint Committee)9 use loads
obtained from ASCE 7 standard with no
modifications.
The roofing industry should follow the
practice of the structural material industries,
i.e., establish an industry-recognized
minimum value of FOS and use the roof
wind pressures as given by ASCE 7 without
modification. (If required, different minimum
values of FOS for different situations
may be recommended.)
Using a recognized FOS and unmodified
ASCE 7 pressures should automatically
satisfy building code requirement for compliance
with ASCE 7 and greatly simplify
the design standards. Additionally, revisions
in design standards can be made
independent of those made in ASCE 7 and
can be modified as ASCE 7 evolves.
Another benefit of the use of unmodified
ASCE 7 pressures in design standards
is that it avoids the confusion caused from
reading two different values of roof pressures
for the same situation—one value
given in ASCE 7-10 tables (e.g., Ref. 2,
p. 328) and a different value given in the
design standard (e.g., Ref. 7, p. 13).
Thus, if no change is sought in roof
design obtained from the two design standards
in their current versions, all that is
needed for the design standards to comply
with the building code is to revise them to
use the current ASCE 7-10 pressures and
alter the minimum FOS from 2.0 to 2.35 (for
ANSI/SPRI) and revise FM Global FOS from
2.0 to the values given in Table 2 (which
vary from 2.0 to 3.5).
6. RECOMMENDATION S
FOR THE FUTURE
The above discussion leads us to the
recommendations for the future. The wind
provisions of the ASCE 7-10 standard have
been revised significantly from its previous
edition (ASCE 7-05). As stated in the introduction,
the most significant revision pertains
to the definition of the basic (design)
wind speed for a location.
In ASCE 7-05, the basic wind speed
for a location has a 50-year return period
for a Risk Category II building. For a Risk
Category III or IV building in the same location,
we use the same basic wind speed but
multiply the wind pressure by an importance
factor of 1.15. The 15% increase
in design wind pressures is statistically
equivalent to increasing the return period of
design wind speed to 100 years.
For a Risk Category I Building, an
importance factor of 0.87 is applied to the
pressures obtained from the use of the basic
wind speed. This is equivalent to assuming
that the design wind speed for a Category
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 P a t t e rs o n a n d Me h t a • 1 7 3
Table 2 – Values of Effective Factors of Safety in ANSI/SPRI and FM Global
Standards
ANSI/SPRI (2012) FM Global (2012)
Minimum Design (Dallas, TX)—nonhurricane-prone region
Risk Category I: Effective FOS = 2.8
Risk Category II: Effective FOS = 2.4
Risk Category III & IV: Effective FOS = 2.2
E nhanced Design (Dallas, TX)—nonhurricane-prone region
Effective FOS = 2.35 Risk Category II: Effective FOS = 3.5
in all situations Risk Category III & IV: Effective FOS = 3.1
Minimum Design (New Haven, CT)—hurricane-prone region
Risk Category I: Effective FOS = 3.5
Risk Category II: Effective FOS = 3.0
Risk Category III & IV: Effective FOS = 2.6
E nhanced Design (New Haven, CT)—hurricane-prone region
Risk Category II: Effective FOS = 3.5
Risk Category III & IV: Effective FOS = 3.0
Note: The values of effective FOS for FM Global have been obtained by dividing FM
Global pressures by the corresponding ASCE 7 pressures given in Table 1 and multiplying
the ratio by 2.0.
I building has a return period of 25 years.
In other words, there is only one basic
wind speed for a location in ASCE 7-05,
whose return period is 50 years. Three different
importance factors (1.15, 1.0, and
0.87) have been used to obtain wind pressures
for three different risk categories.
Wind pressures obtained using ASCE 7-05
basic wind speeds are called “nominal pressures.”
(Incidentally, ASCE 7-05 uses the
term “occupancy categories” in place of the
more appropriate term “risk categories” of
ASCE 7-10.)
By contrast with ASCE 7-05, ASCE 7-10
uses three different basic wind speeds for
three different risk categories (as described
later). The three different wind speeds
reflect the three different return periods for
three different risk categories.
One Factor of Safety—The Allowable
Strength Design (ASD) Approach
Wind-resistant design of roof assemblies
requires that the calculated design wind
pressures (DWP) on a roof equal or exceed
the strength of the roof by an FOS. The FOS
accounts for the fact that our calculations
of DWP and our assessment of the strength
of the assembly entail a great deal of uncertainty.
Because DWP in ASCE 7-05 represents
nominal wind pressure, it does not contain
any safety margin, so that the entire
safety margin is included on the resistance
(strength) side of the assembly, as shown in
Equation 1. This design approach (in which
one FOS, is used on the strength side of the
equation) is known as the “allowable stress
design” or “allowable strength design” (ASD)
approach.
By including the entire safety on the
strength side, the ASD approach ignores
the reality of the situation. The reality is
that uncertainty exists on both sides: 1)
in determining the DWP (uncertainties in
design wind speeds and various coefficients
used to convert the design wind speed to
wind pressures on the building), and 2) in
determining the strength of the assembly
(uncertainties in workmanship, quality of
materials, and the differences between field
and test conditions). Because the quantum
of uncertainty on which the FOS value
depends is probabilistic, the statistically
correct approach is to include a part of the
factor of safety in DWP and a part in wind
resistance (strength) of the assembly.
Two Partial Safety Factors—
Load-Resistance Factor Design (LRFD)
Approach
The design approach in which two (partial)
safety factors are used—one on the
load (pressure) side and the other on the
resistance (strength) side—is called a “loadresistance
factor design” (LRFD) approach.
The structural engineering profession is
transitioning to the LRFD approach. Design
in concrete abandoned the ASD approach
several decades ago, and the other material
industries are headed in the same direction.
Several university programs teach only
LRFD. It is, therefore, recommended that
the roofing industry also transition to LRFD
as soon as practical.
In LRFD approach, the nominal loads
are increased by multiplying them with the
load safety factors, more commonly referred
to as the “load factors.” A load factor (LF)
is always greater than 1.0. The load (wind
pressure) so obtained is called the “ultimate
load” or “ultimate wind pressure.” The
ultimate wind pressure is the design wind
pressure in ASCE 7-10. Thus:
ASCE 7-10 (ultimate) DWP =
(LF)[ASCE 7-05 (nominal) DWP] The value of LF is a function of the
uncertainty in determining the design load.
Its value for wind loads (pressures) in ASCE
7-05 standard = 1.6.
Load Factor and Importance Factor
Absorbed in Basic Wind Speeds in ASCE
7-10
In ASCE 7-10 standard, LF has been
absorbed into basic wind speed. Because
the wind pressure is directly proportional
to the square of wind speed, the basic wind
speed for a location in ASCE 7-10 is larger
than the corresponding ASCE 7-05 basic
wind speed by a factor = √(1.6) = 1.2649.
In other words, the design wind pressure
in ASCE 7-10 has the load factor of 1.6
included in it. Therefore:
(ASCE 7-10 DWP) = (1.6)(ASCE 7-05 DWP)
The importance factor (IF) of ASCE 7-05
that distinguishes between the three risk
categories has also been absorbed in the
basic wind speed. Thus:
(ASCE 7-10 basic wind speed) =
√(1.6) √(IF) [ASCE 7-05 basic wind speed] Therefore, ASCE 7-10 has three basic
wind speeds for a location and one basic
wind speed for each risk category. The basic
wind speeds of ASCE 7-10 are called “ultimate
wind speeds.”
Partial Safety Factor—The Strength
Reduction Factor
The second partial safety factor in
LRFD approach is placed on the resistance
(strength) side. This factor accounts
for the uncertainty in determining the
strength of the assembly to resist wind
pressures. Thus, the strength of the assembly
(obtained from calculations, specimen
testing, or both) is reduced by multiplying
it by a factor called the “strength reduction
factor,” referred to as the “f-factor.” “f” is
always ≤ 1.0. Therefore:
Practical strength of roof assembly =
f (strength of roof assembly from test)
The value of “f” is obtained from a
detailed statistical (structural reliability)
analysis of structural failure and, as previously
stated, depends on workmanship,
quality of materials, type(s) of stress present
in the member, the consequences of
failure caused by the stress, and so on. For
example, the American Concrete Institute
recommends f = 0.9 for bending failure, f =
0.75 for shear failure, and f = 0.65 for compressive
failure. In the absence of a similar
analysis available for roof design, the values
of “f” can only be inferred from the currently
used values of overall FOS from the following
equation:
FOS = LF( )
Substituting LF = 1.6 in Eq. (3):
f =
Thus, if the overall FOS = 2.0, f = 0.8.
For overall FOS = 2.35, f = 0.68, and so
on. Values of f, corresponding to the values
of overall FOS in roof design, are given in
Table 3, which also gives the corresponding
values of .
1 7 4 • P a t t e rs o n a n d Me h t a 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
Equation 3
Equation 4
1
1
FOS
1.6
Suggested Roof Design Procedure as
Per ASCE 7-10 Standard Using the New
Safety Multiplier (NSM)
Because ASCE 7-10 basic wind speeds
are ultimate wind speeds, a partial safety
factor already exists in DWP. Therefore, the
only safety margin needed for roof design
should come from the value of “f,” so that:
f (roof STR) ≥ (ASCE 7-10 DWP)
Or,
Roof STR ≥ (ASCE 7-10 DWP)
The roof design procedure recommended
in this section is based on Equation 6. To
obtain the required minimum strength of
roof assembly using this equation, we first
obtain ASCE 7-10 DWP and then multiply it
with the required value of .
Therefore, should be considered as
the NSM.
Hence, Equation 6 may be written as:
Roof STR ≥ (NSM)(ASCE 7-10 DWP)
As shown in Table 3, NSM = 1.25 for an
overall FOS of 2.0. If the overall FOS is 2.35
(ANSI/SPRI value), NSM = 1.47. A value of
1.5 may be used as an approximation.
Example 1: Determine the minimum
required strength of roof assembly for
Dallas, Texas, given in Table 1. Risk
category is Category II. Overall, FOS = 2.0.
Solution: ASCE 7-10 pressure = 57.5
psf (Ref. 1, p. 96). NSM = 1.25 (Table
3). Therefore from Eq. (7), the minimum
required strength of assembly = 1.25(57.5)
= 71.9 psf ≈ 72 psf.
Example 2: Determine the minimum
required strength of roof assembly of
Example 1. Overall FOS = 2.35.
Solution: ASCE 7-10 pressure = 57.5
psf. NSM = 1.5 (Table 3). Therefore, from
Equation 7, the minimum required strength
of assembly = 1.5(57.5) = 86.3 psf ≈ 86 psf.
This procedure is similar to the one used
in the ANSI/SPRI standard but does not
require converting ASCE
7-10 pressure to nominal
pressure by multiplying it
with 0.6. It obviates the
need for the standard to
generate new, modified
tables for design wind
pressures. Note that 0.6 is
an approximation for 1/1.6,
whose exact value is 0.625.
7. CON CLUDING
REMARKS
The highlights of this paper are as
follows:
Wind Pressures
The structural design profession has
embraced the use of design based on the
ultimate load (LRFD) approach, gradually
discarding the nominal load (ASD)
approach. ASCE 7 and IBC have adopted
ultimate wind speeds for direct use with
strength (LRFD) design. Therefore, lowslope
roof wind resistance design standards
should fully embrace the use of ultimate
wind pressures. The wind pressure tables
provided in roof design standards should
give ultimate wind pressures in place of the
nominal wind pressures. This has the following
advantages:
1. The nominal wind pressure tables
provided in roof design standards
have been obtained using the multiplication
factor of 0.6 instead of
the more accurate factor of 0.625,
incurring an error of 4 percent. This
error can be avoided through the
use of ultimate pressures.
2. If ultimate wind pressure tables
are provided in roof design standards
instead of nominal pressures,
it will allow a curious and informed
designer to easily verify the design
standard tables with the simplified
tables in the current ASCE 7
standard, which give ultimate wind
pressures.
3. More importantly, because the roof
design standards provide wind pressure
tables for a limited set of conditions,
reference to ASCE 7 standard
becomes necessary when the building
does not meet those conditions.
Therefore, the more the design standards
are in consonance with ASCE
7, the easier it is for a designer
to switch from, and make crossreferences
between, the design standard
and ASCE 7.
4. The use of ultimate pressures in roof
design makes it consistent with the
current structural design philosophy
for buildings.
Separate Strength From Design
Pressures
As stated in Section 2, the design of
roofs for wind uplift resistance is a two-part
process: 1) determining the wind pressures
on the roof assembly, and 2) determining
the strength of roof assembly. By and large,
the research and development of standards
related to loads and pressures on buildings
are beyond the interest and expertise of the
roof design profession.
On the other hand, the expertise related
to the strength of roof assemblies to
resist wind pressures lies entirely with roofing
design and construction professionals.
Therefore, the roof design standards should
accept wind pressures as obtained from
ASCE 7 standard without any modifications
and deal only with the strength side of
the equation by specifying the appropriate
value(s) of partial safety factor to be applied
to the tested strength of an assembly, in
addition to dealing with several other related
factors included in the current design
standards, such as fastener arrangement,
adhesive ribbon layout, size and layout of
insulation boards, and so on.
Factor of Safety
The wide variation in the effective FOS
values currently used in wind design of
low-slope roofs (2.0 to 3.5 given in Table
2) points to the lack of consensus in the
roofing industry on the minimum value(s)
of FOS (despite ASTM D6630). As such, the
current situation is confusing, particularly
to those who are familiar with structural
design procedures, where a single minimum
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 P a t t e rs o n a n d Me h t a • 1 7 5
O verall FOS f 1
N ew Safety Multiplier (NSM)
2.0 0.8 1.25
2.4 0.67 1.5
2.8 0.57 1.75
3.2 0.50 2.0
3.6 0.44 2.25
Equation 5
Equation 6
Equation 7
1
1
1
Table 3 – Relationship Between FOS, f, and 1 for Wind
Design of Roofs
value of FOS is prescribed for a given mode
of failure.
There is a need, therefore, for the roofing
community to arrive at a consensus
value of FOS, which will make design for
wind uplift resistance of roofs consistent
with design standards formulated by other
material industries (concrete, steel, wood,
and masonry). Because codes and standards
must specify minimum requirements,
the consensus value must represent the
minimum value of FOS.
From the consensus value (or values) of
FOS, the partial safety factor (the NSM) to
be used with the ultimate wind pressures
of ASCE 7-10 can be determined as shown
in Table 3.
As a start (until a statistical reliability
analysis of wind-resistant roof design is
undertaken), the consensus value of FOS
may be arrived at through an informed
judgment of a panel of experts intimately
related to the field. To minimize personal
bias, conflict of ideas, interests, and
undue influence of articulate personalities,
a structured set of questions can be formulated
to extract anonymous responses from
panel members.
Multiple rounds of questioning to finetune
the questionnaire based on previous
responses through an unbiased facilitator
may be needed. Statistical analysis of
results so obtained should lead to the final
convergence of expert opinion on the value
of FOS. This commonly used procedure
generally requires the help of a knowledgeable
but unbiased facilitator.
REFERENCES
1. S. Patterson and M. Mehta, Wind
Pressures on Low-Slope Roofs, RCIF
Publication No. 01.01, 2013.
2. Structural Engineering Institute
of the American Society of Civil
Engineers, Minimum Design Loads
for Buildings and Other Structures,
2010.
3. International Code Council,
International Building Code, 2006.
4. International Code Council,
International Building Code, 2009.
5. International Code Council,
International Building Code, 2012.
6. FM Global, “Property Loss Prevention
Data Sheets,” January 2012.
7. American National Standard
Institute (ANSI) and Single-Ply
Roofing Institute (SPRI), “ANSI/SPRI
Wind Design Standard Practice for
Roofing Assemblies,” July 2012.
8. T. Smith, “Mapping the 2010 Wind
Changes,” Professional Roofing,
August 2010.
9. American Concrete Institute (ACI),
Building Code Requirements for
Structural Concrete; American
Institute of Steel Construction (AISC),
Specifications for Structural Steel
Buildings; American Wood Council
(AWC), National Design Specification
for Wood Construction; and Masonry
Standards Joint Committee (MSJC),
“Building Code Requirements for
Masonry Structures.”
1 7 6 • P a t t e rs o n a n d Me h t a 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