VI.

DESIGN CONSIDERATIONS

GENERAL
The reference data presented here is of a general nature and is for use as a design check or guideline. It is not intended to replace the
careful studies, which are normally made for each project by the responsible design professionals.
Local building codes and federal and state regulatory agencies establish minimum design loads and other safety requirements for
most structures. The design professional should investigate the adequacy of these regulatory stipulations for a given project or portion
of a project. For tall buildings, buildings with an unusual shape or buildings exposed to unusual surroundings, wind tunnel tests are
recommended as a means of establishing the probable, and frequently unusual, wind load forces on the building.
The continued push for more energy conservation and the federally mandated state energy codes have brought heat transfer design
into the limelight. Testing and computer modeling of complete assemblies are becoming much more common. Various standards-
writing and regulatory agencies have developed criteria to provide guidance and mandatory regulations regarding heat transfer.
Intimate working knowledge of local and state energy codes is a vital requirement in building design. Glass products are available
which provide substantial heat transfer reduction without a corresponding reduction of glass area.

STRUCTURAL PERFORMANCE OF GLASS

For typical sizes and thicknesses used in windows, curtain walls and skylights, glass reacts to loads as a combination plate
and membrane. Conventional engineering procedures for thin plates may not be applicable when the maximum lateral deflection
exceeds half the glass thickness. Glass strength can only be determined by extensive testing or complex engineering analysis.
Glass is a brittle, elastic material up to the point of fracture. Its strength is controlled by the interaction of tensile stresses
with randomly occurring stress-raising discontinuities on the surface or in the body of the glass. If a large number of samples
of nominally identical lites are tested to failure, there will be a significant variation in the measured failure strength. This
means that glass strength must be addressed using statistical methods.
It is generally accepted for weathered annealed glass that about two-thirds of the lites exposed to a uniform pressure will fail
in the range of +/- 20 to 25 percent of the average breaking pressure for each size and thickness. For heat-strengthened glass
the range is +/-15 percent and for fully tempered glass it is +/-10 percent. These percentages are termed coefficients of variation
and are useful for the design of glass and prediction of the probability of breakage.
The guidelines for selecting glass have gone through an evolution since the early 1960s. In the early 1960’s, thousands of
lites of new annealed glass were tested to failure. The resulting data formed the basis for a set of empirical glass thickness
selection charts, which were widely referenced by manufacturers, building codes, and standards up to and including the
1990’s. Additional research and testing of weathered glass that was begun in the 1970’s eventually led to a new set of glass
thickness selection charts for annealed glass that were first presented in ASTM E 1300. The glass thickness selection charts
presented in ASTM E 1300-00 Standard Practice for Determining Load Resistance of Glass in Buildings have since become
the accepted basis for vertical and sloped glazing subjected to wind, snow, and self-weight loads. The ASTM glass thickness
selection charts are based on a theoretical glass failure prediction model and the results of weathered glass tests. The ASTM
charts incorporate detailed information relating to the effect of aspect ratio on the strength of glass. ASTM E 1300 continues
to evolve as more information becomes available.
The glass failure prediction model that serves as the basis for ASTM E 1300 assumes that the probability of glass breakage
is a function of the distribution and severity of stress-raising surface discontinuities and the distribution of surface tensile
stresses over the glass area. If the maximum stress levels in two lites with different dimensions and thicknesses are the same,
the lite with the maximum stress over the greater area is more likely to fail. It is not appropriate to base the structural adequacy
of glass used in buildings solely on its modulus of rupture as determined through the testing of small-scale laboratory specimens.
The ASTM E 1300 standard retained some concepts from earlier work, including the use of strength factors for heat-
strengthened, fully tempered, laminated, and insulating glass constructions. In addition, the ASTM E 1300 standard introduced
load share factors for insulating glass units that incorporate glass lites of the same and different thicknesses and types. ASTM
E 1300 also incorporates procedures to address the variation of glass strength with load duration using glass type factors for
long and short durations. The load resistance factor for laminated glass depends on the relationship of the glass dimensions
and thickness, the interlayer temperature, and the load duration.
Unlike most metal and other engineering materials, glass is usually designed on the basis of an acceptable probability of
breakage or on the basis of historical experience and engineering judgment. ASTM E 1300 presents general procedures to
determine the load resistance of glass for the most common design probability of breakage of 8 lites per 1000. In addition, an
optional procedure is presented that allows other design probabilities to be addressed.

GANA Glazing Manual - 35 2004

 Frequently, it is necessary to design glass of unusual shape or construction. In such cases, calculation techniques such as
finite element, finite difference, or standard engineering mechanics formulas can be used to determine the maximum principal
tensile stress on the surface of the glass as a result of a specific load. ASTM E 1300 presents conservative glass design stresses
corresponding to a probability of breakage of 8 lites per 1000 and a load duration of 60 seconds for the 2000 and previous versions.
Starting with the 2002 version of ASTM E 1300, the load duration was changed to 3 seconds. These values were combined with
statistical methods and the assumption that failure stresses are normally distributed to develop the glass design stresses presented
in Table 6 for different probabilities of breakage. A calculated maximum principal stress can then be compared to the values
presented in Table 6 to conservatively estimate the probability of breakage. Table 6 presents probabilities of breakage ranging
from 1 lite to 8 lites per 1000. While a probability of breakage of 8 lites per 1000 is common for vertical glazing, designers and
architects typically use a probability of breakage of 1 lite per 1000 for sloped glazing, skylights, and other critical applications.
Probabilities of breakage greater than 8 lites per 1000 are not generally recommended for the design of flat glass.

The design stresses presented in Table 6 can be converted to 3-second duration design stresses by multiplying them by a load
duration transformation factor of 1.21.
In addition to strength, the deflection characteristics of the glass should also be examined by the designer. Deflections are a
consideration in maintaining proper gasket engagements that are required to maintain continuous edge support of the glass. If
the glass is wet-glazed, excessive deflections can lead to improper sealant performance. Specific issues relating to overall glass
movement under design load conditions may affect placement of draperies, blinds, or other shading devices. In certain
circumstances, excessive glass deflections can become an aesthetic concern. In the evaluation of glass deflections, the designer
should be aware that annealed, heat-strengthened, and fully tempered glass share a common modulus of elasticity of 10.4 x 106
psi and a Poisson’s ratio of 0.22, and therefore exhibit the same deflection characteristics under the same load. For a given
glass size and load, a thicker glass is required to reduce deflection. ASTM E 1300 presents methods for calculating deflections
in glass with 1, 2, 3, and 4 sides of continuous support.
For selection of relatively thick glass as used in viewing windows for large aquariums, glass railings, glass mullions, and animal
enclosures, finite element analysis, finite difference analysis, or conventional engineering mechanics equations can be used to
calculate stresses. Experience has shown that allowable stress ranges for such applications are as follows:
Annealed Glass 600 - 1,200 psi
Heat-Strengthened Glass 1,200 - 3,000 psi
Fully Tempered Glass 2,400 - 6,000 psi
The suggested annealed glass design stresses are below the generally accepted static fatigue limit for long term loads, and the
suggested design stresses for the kinds of heat-treated glasses are below the minimum requirements of the residual surface
compressions for these kinds of glass. The probabilities of breakage associated with these design stresses will be much smaller
than 1 lite per 1000. These same allowable stress ranges can be used in the design of structural glass railings and glass mullions
supporting all-glass walls.

GANA Glazing Manual - 36 2004

 IN-SERVICE EXPOSURES OF GLASS
Various service conditions justify special considerations. These conditions may increase glass stresses and probability of
breakage. If they are not considered, glass may be selected which may not be adequate for the conditions. These conditions
include the following:
1. Screens, eyebrows, louvers, shutters, etc., may increase or decrease wind loads and thermal stresses.
2. Windborne roof gravel, hail and windborne debris may lead to surface damage, reduced strength and increased breakage
under subsequent impact, wind load or thermal load.
3. Severe temperature exposures, uneven temperature exposures, glazing stresses, sonic boom, seismic action, mechanical
stresses from door or window operation, pressure effects of air conditioning system operation, stack effects of ventilating
systems and impact load such as that caused by window washing ladders or equipment, hose streams, etc., may impose
significant stresses.
When the effect of service conditions cannot be accurately predicted for the life span of the building, it is generally prudent to
specify a lower probability of breakage, e.g., 4, 2, or 1 lite per 1000.

DESIGN LOAD
Various types of loads and combinations of loads must be considered for the design of glass. These types of loads include: 1)
wind loads, both positive (inward) and negative (outward); 2) dead loads; 3) thermal loads; 4) snow/ice loads; 5) impact loads;
6) seismic loads; 7) interior pressures from HVAC equipment; 8) interior pressures due to building stack effect (these tend to be
inward at the base of a building and outward near the top of the building); and 9) live loads (generally glass is not designed for
people to walk on it; however, occasionally design professionals specify a live load to account for a distributed load that may be
applied for maintenance). Load duration is also of great importance.
Although all of the above types of loads must be considered by the design professional, the discussion that follows is limited
to wind load, interior pressures, missile impact loads and loads associated with sloped glazing.

WIND LOAD
The principal load applied to glass in an exterior wall is the net pressure differential caused by local wind conditions.
Therefore, it is important to understand this type of loading to ensure proper design. The design wind load for a lite of glass on
the side of a building is dependent on the following: 1) wind speed, 2) importance of the structure, 3) type of exposure, 4)
building height, 5) building shape and orientation, 6) location on the building, and 7) size of the glass. Each of these is
discussed briefly below.

WIND SPEED
Wind pressure (loading) is proportional to the square of the wind speed. At standard atmospheric pressure [29.92 inches of
mercury (101.325 kPa) atmospheric pressure and 59 degrees F (15 degrees C) temperatures] the relationship between wind
speed and wind pressure is found in Ensewiler’s Formula: P = 0.00256 V2 where P is the pressure in pounds per square foot
exerted on a stagnation point on a flat surface oriented normal to the wind direction and V is the wind speed in miles per hour.
Wind speeds have been recorded by the National Weather Service and are summarized on a U.S. map in American Society of
Civil Engineers (ASCE) 7 Minimum Design Loads for Buildings and Other Structures. Formerly, wind speeds were measured in
a manner known as the “fastest mile of wind” in which the procedure was to measure the time it takes for a mile long sample of air
to pass a fixed measuring point. During the last couple of decades, wind speeds have been measured using anemometers that are
able to measure gust speeds in a relatively short 2 to 3 second duration. These anemometers are located at the standard height of
33 ft. (10 m) above ground level in open terrain with scattered obstructions having heights generally less than 30 feet (9.1 m).
Now wind speeds are summarized as their “gust speeds” on a 3-second duration time basis. In the United States, the 1995
edition of ASCE 7 first introduced the 3-second gust speed wind speed maps for use in wind engineering design. With this
approach, the gust factors used were significantly reduced when compared to gust factors used with fastest mile wind speed
data. In most cases gust factors equal to 0.85 are used. Based upon information presented in ASCE 7-02, use of a gust factor of
0.85 with a 3-second gust wind speed results in design pressures on cladding with durations somewhat greater than 3 seconds.
Thus, the gust duration associated with the 3-second versions of ASCE 7 are consistent with the gust durations associated with
the fastest mile wind speed versions of ASCE 7. In ASCE 7-88 (fastest mile version), it is stated:
”The analytical procedure provides pressures that are expected to act on components and cladding for durations
in the range from 1 to 10 seconds. Peak pressures acting for a shorter duration may be higher than those
obtained using the analytical procedure. The gust response factors, pressure coefficients, and force coefficients
of this standard are based on a mean wind speed corresponding to the fastest-mile wind speed.” —ASCE 7-88,
ASCE 7-93 (Last two fastest-mile versions of ASCE 7)
Cladding pressure durations of 1 to 10 seconds have been associated with wind load procedures since at least the early 1960’s.
Thus, it can be concluded that the durations associated with cladding pressures are not directly a function of the averaging period

GANA Glazing Manual - 37 2004

used to report wind speed data. Further, it is clear that introduction of the 3-second versions of ASCE 7 resulted in no significant
change in the durations associated with design pressures for cladding.
The 3-second wind speed map presented in Figure 8 presents the maximum 3-second gust wind speed that has a probability of
occurrence of 0.02 in a single year or a mean recurrence interval of 50 yrs. However, severe wind gusts are not isolated events.
Rather, they are a part of a windstorm system such as a thunderstorm, tornado, downburst, hurricane, etc. It is well understood
that multiple wind gusts are associated with severe windstorm events. Because of the static fatigue properties of glass, all of the
wind gusts that affect the glass have a cumulative effect at least over the duration of the windstorm event. So it is not enough to
simply know the magnitude and duration of a single gust event. The combined durations of all of the gusts embedded in the
windstorm event must be recognized in the design of glass.
Historically, glass thickness selection charts have been based on a 30-60 second duration glass strength with most code
bodies referencing a 60 second duration. The reason that this is the case is that a windstorm does not consist of a single
maximum gust. Rather, a windstorm consists of a series of gusts several of which are at or near the maximum value. Information
presented in ASTM E 1886-02 Standard Test Method for Performance of Exterior Windows, Curtain Walls, Doors and Storm
Shutters Impacted by Missile(s) and Exposed to Cyclic Pressure Differentials suggests that there can be as many as 100 short
gusts near the maximum wind speed in a single severe windstorm event. Thus, it is very reasonable to assume that combined
durations of the maximum wind speed gusts in a single windstorm event will sum to a value in the range of 60 seconds. This would
involve six 10-second gusts, or twelve 5-second gusts, twenty 3-second gusts, etc.
The latest version of ASTM E 1300 standard incorporates non-factored load charts corresponding to a 3-second duration.
When a wind event is of a sustained nature, the actual wind load will have more than one maximum 3-second gust. This
situation will present pressures that when accumulated will have an effect on the glass that will be greater than the 3 second
duration. The specifying authority must consider this when determining the load duration to be used for the glass load resistance
analysis. It may not be appropriate to use the ASTM E 1300 3-second duration load charts for all wind load conditions. The
standard allows for calculation of strength requirements at longer load durations.
Since the capacity of many materials to resist a load is often time dependent, it is important to ensure that wind loads
reference a duration. Glass, for example, can resist more uniform loading for a three second duration than it can for a one
minute duration, and it can resist more loading for a one minute duration than it can for a one month duration.
Figure 8, 8a, 8b, 8c on pages 40-43 are wind speed maps taken from ASCE 7-02.

They show basic wind speed as 3-second gust speeds that are associated with an annual probability of 0.02.

IMPORTANCE
An annual probability of 0.02 corresponds to a mean recurrence interval of 50 years, which is the most common design
interval selected for wind design on buildings in the U.S. Through the use of an Importance Factor multiplier, adjustments are
made to the annual probability of occurrence for buildings whose occupancies reflect different levels of hazard to human life.

TYPE OF EXPOSURE
The type of exposure is categorized to reflect the characteristics of ground surface irregularities for the building site. Variations
in ground surface roughness arise from natural topography and vegetation, as well as from constructed features. The categories
of exposure classified in ASCE 7 are:
1. Exposure A. This exposure condition, which was originally intended to represent highly built up down town areas, has
been removed from consideration.
2. Exposure B. Urban and suburban areas, wooded areas or other terrain with numerous closely spaced obstructions having
the size of single-family dwellings or larger. Use of this exposure category shall be limited to those areas for which terrain
representative of Exposure B prevails in the upwind direction for a distance of at least 1,500 feet or 10 times the height of the
building or structure, whichever is greater.
3. Exposure C. Open terrain with scattered obstructions having heights generally less than 30 feet. This category includes
flat, open country and grasslands.
4. Exposure D. Flat, unobstructed coastal areas directly exposed to wind flowing over large bodies of water. This exposure
shall be used for those areas representative of Exposure D extending inland from the shoreline a distance of 1,500 feet or 10
times the height of the building or structure, whichever is greater.
The ASCE standard also makes reference to increased pressures (or suctions) that can occur from channelization of wind by
nearby buildings and obstructions. Also see “Windborne Missiles.”
BUILDING HEIGHT
Wind speed, and therefore wind pressure, increases with height above ground level and the rate of increase is related to the
type of terrain over which the wind flows, i.e. the terrain applies a frictional drag to the air moving over it. The increase is not a

GANA Glazing Manual - 38 2004

straight-line relationship to the height nor is it a proportional relationship to the height. The rate of increase in speed decreases
with height until, at the gradient level, the wind speed is assumed to be constant. See ASCE 7 for more information.
BUILDING SHAPE & ORIENTATION
When the wind blows perpendicular to a building face, it is slowed down with a consequent build-up of pressure against that
face. At the same time, it is deflected and accelerated around the end walls and over the roof, creating a suction or negative
pressure on these areas. A large eddy is created behind the building, which exerts suction on the leeward face.

Figure 7
Negative Pressure Effects

The magnitude of the negative pressures created on square or rectangular buildings in Exposures B and C are relatively well
known and generally can be anticipated in design. Considerably more obscure are the effects on buildings with corners that are
not 90o, buildings with other than four sides, and buildings not oriented to face the direction of the prevailing high winds and
the channelization effects of adjacent buildings. Negative pressures are significantly greater than positive pressures at building
corner zones.

LOCATION ON THE BUILDING

Corner zones and other areas of discontinuity, such as roof ridges, have higher negative (outward) loads than flat walls away
from discontinuities. Therefore, the negative loads on glass at corners are generally higher than at intermediate areas.

SIZE OF THE GLASS

Glass is a component of the wall system. As such, it experiences only the load applied locally to it. Smaller areas experience
greater wind loads than those averaged over large areas. Therefore, the pressure per square foot on a lite of glass is higher than
the average pressure over a larger wall area. ASCE 7 has tables that make adjustments for tributary areas.

MINIMUM DESIGN WIND LOADING

Per ASCE 7, the design pressure for components and cladding shall be not less than 10 psf (0.48 kN/m2) acting in either
direction normal to the surface.
More complete discussions regarding wind load on buildings may be found in the American Society of Civil Engineer’s
publication ASCE 7 and the American Architectural Manufacturers Association (AAMA) publication, Design Wind Loads for
Buildings and Boundary Layer Wind Tunnel Testing.

GANA Glazing Manual - 39 2004

 Figure 8
Wind Speed Map*
Basic Wind Speed
miles per hour (meters/second)

NOTE: Consult design professional for determination of appropriate design wind loads.

*Basic Wind Speed map is taken from Figure 6-1, ASCE 7-02 Minimum Design Loads for Buildings and other Structures and is used
with permission from American Society of Civil Engineers.

GANA Glazing Manual - 40 2004

 Figure 8a
Wind Speed Map*
Basic Wind Speed – Western Gulf of Mexico Hurricane Coastline
miles per hour (meters/second)

NOTE: Consult design professional for determination of appropriate design wind loads.

*Basic Wind Speed – Western Gulf of Mexico Hurricane Coastline map is taken from Figure 6-1a, ASCE 7-02 Minimum Design
Loads for Buildings and other Structures and is used with permission from American Society of Civil Engineers.

GANA Glazing Manual - 41 2004

 Figure 8b
Wind Speed Map*
Basic Wind Speed – Eastern Gulf of Mexico and Southeastern US Hurricane Coastline
miles per hour (meters/second)

NOTE: Consult design professional for determination of appropriate design wind loads.

*Basic Wind Speed – Eastern Gulf of Mexico and Southeastern US Hurricane Coastline map is taken from Figure 6-1b, ASCE 7-02
Minimum Design Loads for Buildings and other Structures and is used with permission from American Society of Civil Engineers.

GANA Glazing Manual - 42 2004

 Figure 8c
Wind Speed Map*
Basic Wind Speed – Mid and Northern Atlantic Hurricane Coastline
miles per hour (meters/second)

NOTE: Consult design professional for determination of appropriate design wind loads.

*Basic Wind Speed – Mid and Northern Atlantic Hurricane Coastline map is taken from Figure 6-1c, ASCE 7-02 Minimum
Design Loads for Buildings and other Structures and is used with permission from American Society of Civil Engineers.

GANA Glazing Manual - 43 2004

INTERNAL BUILDING PRESSURES
When determining the thickness or type of glass needed to resist a specified load, internal building pressures must also be
considered. These internal pressures become especially significant in partially enclosed buildings and in buildings located in
hurricane prone regions, especially when the openings are not designed to resist the impact from windborne debris. For more
information reference ASCE 7.

WINDBORNE MISSILES
Current design standards and codes acknowledge the existence of small and large missile impacts upon building facades.
ASTM standards E 1886 Standard Test Method for Performance of Exterior Windows, Curtain Walls, Doors and Storm Shutters
Impacted by Missile(s) and Exposed to Cyclic Pressure Differentials and E 1996 Standard Specification for Performance of
Exterior Windows, Curtain Walls, Doors and Storm Shutters Impacted by Windborne Debris in Hurricanes provide guidance
and information relating to the testing of glass to demonstrate its resistance to the effects of both small and large missile impacts.
The small missiles consist primarily of roofing stones and other urban debris that can easily become windborne. The large missiles
are primarily intended to represent sections of loose timber members that can be propelled by the wind.
Investigations at the Institute for Disaster Research at Texas Tech University indicate that a 0.2 oz. (6.0 gm) roofing stone can,
within a distance of 70 feet, attain a speed of 35 mph in an 80 mph wind, a speed of 42 mph in a 100 mph wind, and a 51 mph in a
120 mph wind. Further, it was shown that all types, kinds, and thicknesses of glass are susceptible to breakage and damage when
impacted by roofing stones at relatively low speeds. For example, most common thicknesses of annealed monolithic glass will
break when impacted by a roof stone missile with a speed of about 26 mph and most common thicknesses of fully tempered glass
will break when impacted by a roof stone missile of 45 mph.
Negative wind pressures can readily lift stones and other debris from roofs and inject them into the wind stream. Gravity
will cause the stones to assume a curved trajectory, and they will be caught by winds channelized between buildings. The speed
of wind channelized between buildings can be substantially greater than the wind speed registered at a recording station for the
general area. Therefore, it can be conclude that all common types and kinds of window glass are susceptible to breakage as the
result of small missile impact in relatively mild windstorms.
National and local building codes have acknowledged the existence of large and small missile impact upon building facades
under high wind loads. Certain jurisdictions in the Atlantic and Gulf of Mexico coastal areas have developed building codes
that require stringent impact testing and certifying of glazing systems under certain circumstances dependent on building
location and wind speeds. Specialized glazing and framing products have been developed to meet these test requirements.

SLOPED GLAZING
Inward and outward sloping of partial or entire building facades evolved from skylights. “Sloped Glazing,” the modern-day
terminology for fenestration that slopes 15o or more from the vertical, has attained tremendous popularity in all types of
building construction. The sloped glazing system must entail a broader range of design considerations than the conventional
vertical window, store front or curtain wall system. The design and selection of glass for skylights and sloped glazing requires
special attention for a number of reasons:
1. Due to the great degree of solar energy exposure, both the glass and metal framing system must be carefully engineered
to accommodate differential thermal stresses and opposing building reactions. For most orientations, sloped glazing
may reach substantially higher temperatures than vertical glazing because the solar radiation is more nearly perpendicular
to the glass surface, and because of the stratification of warm air under the glass. Consequently, the thermal stresses
created usually require heat-treated glass.
2. The various geometric shapes and geographic locations require that dead loads, snow loads, seismic loads, live loads
and wind loads be analyzed in certain combinations in accordance with state and local building codes.
3. Good effective drainage of both condensation and water infiltration is essential. The glazing rabbet must be drained
into a gutter, and the gutter must freely allow water to exit to the exterior. Sealants alone should not be relied upon to
prevent water infiltration. Generally, conventional vertical window or curtain wall systems do not perform adequately
when installed in a sloped configuration.
4. The drainage systems of sloped glazing systems generally do not function well when installed in a vertical position.
5. Horizontal caps, if used, must be designed to allow proper and complete runoff of water. If not, glass staining, premature
failure of insulating glass and water infiltration may result.
6. Sloped glass is more susceptible to impact from falling objects, windborne debris and missiles than vertical glass.
7. Sloped glazing, in most cases, is more likely to fall from the opening when it breaks than is vertical glass. Good design
indicates that the choice of glass must be based on eliminating or minimizing, to the degree practical, any potential
hazards.
8. Snow loads, unlike wind and live loads, may be imposed on the glazing for extended periods; strength of glass and
contribution of the plastic interlayer in laminated glass are both time dependent, i.e., under long-term loading, the
strength of these materials is less than under short-term loading, such as wind.

GANA Glazing Manual - 44 2004

 9. If the sloped glazing is close to walkways, consideration must be given to snow and ice sliding off, or water cascading
off, striking pedestrians and potentially causing injury.
Proper glass selection is probably the single most important consideration when dealing with a sloped glazing system. Factors
such as life safety resulting from glass fallout after breakage and the potential liability for the owner, architect, glass and skylight
manufacturers, and glazing contractor must be carefully considered. Consult applicable building code for commercial and residential
construction guidelines.

GLASS THICKNESS AND SIZE SELECTION

In order to select the appropriate glass type and thickness for a specific job, the design professional(s) must determine the
design load on the glass. There are three criteria used to determine appropriate loads:
1. Model wind tunnel studies,
2. Requirements as defined in the current version of ASCE 7, Minimum Design Loads for Buildings and Other Structures,
3. Applicable local building code requirements
Model wind tunnel studies are job specific and are often done for large, complex building designs. Applicable local building
codes are minimum requirements that must be met. Determining the design load on a building in accordance with ASCE 7 is the
most often used method for selecting glass. Specifiers must be aware that a number of loads affect design including wind load,
snow load, dead load, seismic loads and live loads. ASCE 7 provides information in determining the appropriate loads and
combinations of loads to be applied.
Once the design load and duration have been determined and a suitable probability of breakage selected, the appropriate glass
thickness and glass type can be chosen. The industry standard to assist in the selection of glass thickness is ASTM E 1300
Standard Practice for Determining Load Resistance of Glass in Buildings. ASTM E 1300 provides glass thickness charts relating
length, width and thickness of glass to equivalent design loads of both short duration (3 seconds in versions 2002 or later and 60
seconds for versions 2000 and sooner) and long duration (up to one month). In addition, the standard provides information on:
1. Calculating maximum glass deflection,
2. Estimating the probability of breakage of rectangular glass (subjected to a design load),
3. Multipliers for heat-treated, laminated and insulating glass,
4. Effect of aspect ratio on glass strength.
Glass is a brittle material. It will act elastically until it ruptures at ultimate load. That ultimate load will vary, depending
upon the type and duration of the loads applied and the distribution, orientation and severity of the surface discontinuities.
Because the ultimate strength of glass will vary, its strength can best be described statistically for many applications. The
commonly referenced probability of breakage of 8 lites per 1000 (Pb = 0.008) describes the statistical probability of a fracture
in an annealed lite at design load or greater. It should not be confused with a statement describing the average number of lites
that will fail. ASTM E 1300 does provide for determining allowable load with alternative probabilities of breakage.
The specifier should be aware that the maximum center of glass lateral deflection of a lite is often a consideration in the
selection of glass and should be addressed. Excessive deflection can cause poor performance of glazing gaskets or tapes and
glass-to-metal contact, causing glass breakage. For the same thickness, heat-treating will not change the deflection characteristic
of glass; it only changes its breaking strength.
While the glass manufacturers have the appropriate data for determining the performance of their products, it remains the
responsibility of the design professional to review these performance criteria and determine if they are suitable for the intended
application.
All glass products have size limitations. Consult manufacturer/fabricator.

THERMAL PERFORMANCE
The high-performance capabilities of today’s architectural glass products require a strong working knowledge of optical and
thermal performance terminology. The following terms are commonly used to describe and analyze the performance of architectural
glass products.
Visible Light Transmittance: The percentage of visible light within the solar spectrum (390 to 770 nanometers) that is transmitted
through glass. A generic listing of visible light transmittance of varying glass products is noted in Table 9, page 48.

Visible Light Reflectance: The percentage of visible light within the solar spectrum that is reflected from the glass surface.

Solar Energy Transmittance: The percentage of ultraviolet, visible and near infrared energy within the solar spectrum (300 to
2100 nanometers) that is transmitted through the glass.

GANA Glazing Manual - 45 2004

 Solar Energy Reflectance: In the solar spectrum, the percentage of solar energy that is reflected from the glass surface(s).

Shading Coefficient: The ratio of the solar heat gain through a specific fenestration to the solar heat gain through a lite of 1/
8 (3mm) clear glass. Glass of 1/8 (3mm) thickness is given a value of 1.0; therefore, the shading coefficient of a glass product is
calculated as follows:
S.C. = Solar Heat Gain of the Glass in Question
Solar Heat Gain of 1/8” Clear Glass

U-Value: A measure of air-to-air heat transmission (loss or gain) due to thermal conductance and the difference in indoor and
outdoor temperatures. As the U-value decreases, so does the amount of heat that is transmitted through the glazing material. A
lower U-value reduces the amount of heat transferred through the fenestration product.
U-values are expressed in Btu/hr/ft2/oF (W/m2/oC). Glass manufacturers and fabricators publish center-of-glass U-values. Since
the area within 2.5 (64mm) of the glass edge may have a higher U-value due to the influence of an insulating unit spacer material
and framing material, window manufacturers publish total window U-values. The U-value is the reciprocal of the R-value. U-
values can be converted to R-values as follows: R-value = 1/U-value.
U-values are calculated on the basis of the standard American Society of Heating, Refrigerating and Air-Conditioning Engineers
(ASHRAE) conditions as shown in Table 7. A generic listing of center of glass U-values for various glass products is noted in
Table 8, page 47.

Relative Heat Gain (RHG): The amount of heat gain through a glazing material taking into consideration the effects of solar
heat gain (shading coefficient) and conductive heat gain (U-value). The value is expressed in Btu/hr/ft2 (W/m2).
The relative heat gain is calculated as RHG = (Summer U-value x 14oF) + (Shading Coefficient x 200). The lower the relative heat
gain, the more the glass product restricts heat gain.
R-Value: The thermal resistance of a glazing system is expressed in ft2/hr/oF/Btu (m2/W/oC). The R-value is the reciprocal of the
U-value. The higher the R-value, the less heat is transmitted throughout the glazing material.

Solar Heat Gain Coefficient (SHGC): The ratio of the solar heat gain entering the space area through the fenestration product
to the incident solar radiation. Solar heat gain includes directly transmitted solar heat and absorbed solar radiation, which is then
reradiated, conducted, or convected into the space.
SHGC = Solar Energy Gain Through the Glazing

Solar Energy Incident on the Glazing

Glass manufacturers and fabricators provide center-of-glass Solar Heat Gain Coefficients. Total fenestration product values
require consideration of all frame components, the edge of the glass construction and center of glass conditions. A generic listing
of Solar Heat Gain Coefficients for various glass products is noted in Table 9, page 48.

Emissivity (e): The measure of a surface’s ability to emit long-wave infrared radiation or room temperature radiant heat energy.

Emittance: The ratio of the rate of radiant emission of the body, as a consequence of temperature only, to the corresponding
emission of a black body at the same temperature.
Coolness Index (Luminous Efficacy or Light-to-Solar Gain Ratio): The visible transmittance of a glazing system divided by
the solar heat gain coefficient. This ratio is helpful in selecting glazing products for different climates in terms of those that
transmit more heat than light and those that transmit more light than heat.

GANA Glazing Manual - 46 2004

 Industry published performance values for shading coefficients, U-values, and SHGC are centerpane values based on laboratory
measurements and calculations from WINDOW 5.2 Program for Analyzing Window Thermal Performance, developed by the
Windows and Daylighting Group of Lawrence Berkeley National Laboratory at the University of California through contract
support from the United States Department of Energy.

GANA Glazing Manual - 47 2004

GANA Glazing Manual - 48 2004
 GLAZING CONSIDERATIONS FOR SYSTEMS IN SEISMIC REGIONS

The design professional must specify the various loads anticipated during a seismic event. The following should be considered.

GLASS CORNER AND/OR EDGE CUSHIONING

Padding consisting of 50-70 Shore A durometer hardness material should be placed in the glazing channel or on the glass
edges/corners to avoid any glass to frame contact due to the anticipated sway of the frame. This is important for both dry glazed
and wet sealed systems, since the building structure will experience movements and vibration, e.g., during an earthquake.

GASKET PERFORMANCE
For a dry glaze system consisting of a key-in gasket on one side and a roll-in (wedge) gasket on the other side of the glass,
the wedge gasket should have a positive lock-in method so that the gasket will not disengage from the metal framing system
during the up and down and side-to-side movement that occurs during a seismic event. The gasket must remain in place when
the glass moves due to the sway of the building.

SETTING BLOCKS AND SUPPORTS

The setting blocks should be positioned in a permanent manner by keying into the horizontal framing member, using a
compatible sealant for placement or other method that will not allow movement of the block. When a setting chair is used to
support the block, it should be permanently anchored or secured to the horizontal framing member and support the setting
block as referenced.

SNAP-ON OR APPLIED FINISH STRIPS

Members of the glazing system that act as finish strips and/or glass support members that rely on compression of gaskets or
metal fit to remain in place should be attached and designed so that the member will not become loose or cease to provide glass
support from vibration and sway due to seismic loads.

SEALANT DESIGN AND APPLICATION

Sealant joint and application practices should be used that incorporate the anticipated movement of the glazing system and
provide structural capacity (for structural glazing systems) along with weatherproofing requirements and glass retention ability.

WALL SYSTEM CONNECTION AND BUILDING ANCHOR SYSTEMS

Wall system connection and building anchor systems should be designed to consider the inertia loads of the cladding and
the distortion and vibration sources from in-plane and out-of-plane loads. Building code requirements must be reviewed and
understood for issues relating to cladding isolation and story drift limitations. Seismic movement considerations may require
special analysis for the connections and building attachment methods to insure ductile behavior of the system components.

GLASS HOLDING FRAMES

The glass holding frames should be designed to accommodate the inter-story drift and racking anticipated during the
seismic event and allow for proper edge clearance to avoid loads on the glass edges and surfaces.

GANA Glazing Manual - 49 2004