Future Advanced Windows for Zero-Energy Homes

Reprinting of this proof for distribution or for posting on Web sites is not
permitted. Authors may request permission to reprint or post on their
personal or company Web site once the final version of the paper has been
published. A reprint permission form may be found at www.ashrae.org.
KC-03-12-3
Future Advanced Windows for
Zero-Energy Homes
Joshua Apte
Dariush Arasteh, P.E.
Yu Joe Huang
Member ASHRAE
Member ASHRAE
ABSTRACT
dential buildings. Approximately 2.7 quads (2.8 EJ; Appendix
A) of the total 9.6 quads (10.1 EJ) of source energy used for
Over the past 15 years, low-emissivity and other techno-
residential heating and cooling is attributable to today’s
logical improvements have significantly improved the energy
window stock. This amounts to nearly three percent of total
efficiency of windows sold in the United States. However, as
interest increases in the concept of zero-energy homes—build-
U.S. energy consumption, or a cost of more than $25 billion
ings that do not consume any nonrenewable or net energy from
(DOE 2002). If all windows in today’s residential stock were
the utility grid—even today’s highest-performance window
low-e, the energy use attributable to windows would drop to an
products will not be sufficient. This simulation study compares
estimated 1.6 quads (1.7 EJ; Appendix B).
today’s typical residential windows, today’s most efficient resi-
However, as interest increases in the concept of zero-
dential windows, and several options for advanced window
energy homes—buildings that do not consume any non-
technologies, including products with improved fixed or static
renewable or net energy from the utility grid—even today’s
properties and products with dynamic solar heat gain proper-
highest-performance window products will not be able to meet
ties. Nine representative window products are examined in
the requirements of a zero-energy home. A new generation of
eight representative U.S. climates. Annual energy and peak
highly efficient windows will require new technologies. As
demand impacts are investigated. We conclude that a new
today’s highly efficient (“super”) windows tend to be climate-
generation of window products is necessary for zero-energy
specific, one way to improve window energy efficiency would
homes if windows are not to be an energy drain on these homes.
be to develop dynamic fenestration systems that can alter their
Windows with dynamic solar heat gain properties are found to
solar heat gain properties according to seasonal/temperature
offer significant potential in reducing energy use and peak
variations. This paper describes a simulation study that
demands in northern and central climates, while windows with
very low (static) solar heat gain properties offer the most
compares the performance of currently available windows,
potential in southern climates.
future windows with dynamic solar heat gain properties, and
future windows that represent only improvements in the static
INTRODUCTION
properties of today’s highly efficient super windows. Perfor-
mance was studied for different U.S. climates.
During the past 15 years, low-emissivity (low-e) glazings
and other improvements in window technology have signifi-
In order to understand the advantages of a dynamic
cantly reduced window-related energy use and peak demand
window system, it is important to understand the limitations
in residential buildings. Estimates indicate that more than 40%
on the performance of current high-performance, low-e
of windows sold today have low-e coatings (Ducker 2000),
windows. Today’s low-e windows are designed to address the
and low-e products are expected to dominate the market in the
parameters that are typically used to quantify energy perfor-
near future. Despite the energy advantages of low-e coatings,
mance: U-factor (a measure of the heat lost because of indoor-
windows still represent a significant energy liability in resi-
outdoor temperature differences) and solar heat gain coeffi-
Josh Apte is a student research assistant and Dariush Arasteh and Joe Huang are staff scientists at Lawrence Berkeley National Laboratory,
Berkeley, Calif.
2003 ASHRAE. THIS PREPRINT MAY NOT BE DISTRIBUTED IN PAPER OR DIGITAL FORM IN WHOLE OR IN PART. IT IS FOR DISCUSSION PURPOSES ONLY
AT THE 2003 ASHRAE ANNUAL MEETING. The archival version of this paper along with comments and author responses will be published in ASHRAE Transactions,
Volume 109, Part 2. ASHRAE must receive written questions or comments regarding this paper by July 11, 2003, if they are to be included in Transactions.

---

cient (SHGC, which quantifies the fraction of heat from inci-
lating, and air-conditioning (HVAC) systems are employed to
dent solar radiation entering a space).
lower total building energy use. Impacts on peak demand are
By reflecting long-wave radiant energy, low-e windows
also examined.
reduce window U-factors and therefore reduce heat loss
through windows. Although valuable in all climates, reduced
EMERGING TECHNOLOGIES
U-factors are most useful in colder climates where heating
Technologies currently in research and development are
energy requirements, driven by large indoor-outdoor temper-
expected to lay the groundwork for the next generation of resi-
ature differences, are significant.
dential window products. These technologies are described
Many low-e coatings are tuned to reflect the solar infrared
briefly in the following paragraphs (Carmody et al. 2000;
(or invisible) portion of the sun’s energy in order to reduce a
Arasteh 1995).
window’s SHGC. Such products, known as “low-gain” or
Several technologies are currently being researched to
“spectrally selective” glazings, are effective in climates where
reduce the heat loss (U-factor) of windows. These include
cooling dominates energy bills, but these products also reduce
vacuum windows, aerogel windows, and improved multi-
solar gain through windows during the heating season, thus
layer low-e/gas-filled windows. Vacuum windows utilize low-
reducing the windows’ ability to provide free solar heat. Low
e coatings and an evacuated air space (which virtually elimi-
U-factors combined with solar gains are important in turning
nates conduction/convection), much like a thermos bottle.
windows from liabilities to assets during the heating season.
Aerogel is a silica-based, open-cell, foam-like material
Some low-e coatings (high-solar-gain, low-e) have been
composed of about 4% silica and 96% air; the microscopic
developed to maximize the SHGC; these products are optimal
cells trap air, maximizing the insulating value, but still allow-
in climates where heating dominates the energy used for space
ing light to pass. Multi-layer (two or more low-e coatings and
conditioning because high solar gains can offset heating loads.
gas-filled insulating gaps) highly insulating windows are
In short, currently available products represent a compro-
currently available as specialty products; new manufacturing
mise: they perform best in climates where either heating or
approaches (rigid polymeric “nonsealing” inserts, convection
cooling is the dominant space-conditioning use, but not both.
baffles, and thin glass) offer the potentials for more cost-effec-
Unfortunately, most U.S. climates require both heating and
tive products.
cooling during some periods of the year. Although it is rela-
Technologies to reduce solar heat gain include improve-
tively straightforward to say that low-solar-gain windows are
ments to existing low-e coatings, light redirecting layers, and
appropriate in the southern U.S. where air conditioning domi-
self-shading windows. Today’s current spectrally selective
nates space-conditioning needs and heating is rarely required,
low-e coatings minimize unwanted solar heat gain by trans-
the choice is less clear-cut in the rest of the country. Even in
mitting only the visible light and minimal near-infrared;
climates where heating is the dominant space-conditioning
sharper cutoffs will lead to small decreases in solar heat gain.
need, air conditioning is typically needed during some part of
Lower transmittances across the visible spectrum will reduce
the summer. A window that reduces solar gain will lower
solar heat gain further but may also make the glass appear
summer cooling energy consumption but also reduce the solar
darker. Consideration of frame/glazing shading patterns in
gains that can significantly offset wintertime heating costs.
window design may also help reduce solar heat gains for high
Thus, all low-e windows will provide less-than-optimal solar
sun angles.
gain performance during some portion of the year.
Technologies that enable windows to have dynamic prop-
A dynamic window system could optimize a window’s
erties include electrochromic glazings, operable shading
solar-gain characteristics according to weather conditions,
systems, and light-redirecting devices. Electrochromics are
taking advantage of passive solar effects in winter and reject-
typically multilayer coatings that change transparency over a
ing unwanted solar heat gain in summer. The study described
broad range (from as high as 70% down to a few percent);
in this paper used the energy simulation program DOE2.1 to
while researched in the past for commercial building applica-
evaluate the potential benefits of dynamic fenestration
tions, modified electrochromics have significant potentials for
systems and compared these hypothetical future systems to
use in high-performance residential buildings where seasonal
what might be the next generation of energy-efficient
solar switching is needed. Operable and controlled shading
windows based only on improvements in current technolo-
systems can be significant energy savers and can be built using
gies—that is, more highly efficient (“ultra”) windows with
currently available technologies. Light-redirecting glazings
“static” (fixed) rather than dynamic properties. In addition, we
can be utilized to transmit winter sun but reflect summer sun;
considered the impact of combining the properties of the
such products can take the form of angle selective films, modi-
dynamic and ultra windows. Although neither the dynamic nor
fied coatings, or refractive/reflective glazing geometries.
the ultra windows simulated in this study are currently avail-
SIMULATION DESIGN
able, they represent products that could, realistically, result
from research during the next decade.
For this study, the energy performance of dynamic fenes-
This study also quantifies the impacts of dynamic and
tration systems was simulated with RESFEN 5, an interface to
ultra static windows in typical residential applications where
the DOE-2.1E energy model (Mitchell et al. 2002). The simu-
shading strategies and improved insulation and heating, venti-
lated windows were placed in a single-story, frame-construc-
2
KC-03-12-3

---

tion residence with 2,000 ft2 (186 m2) of floor area and 75 ft2
SHGC of 0.1. Finally, a window was simulated that combines
(7 m2) of window area on each orientation (north, east, south,
the properties of the dynamic and ultra windows; it has the
west). The homes were simulated in eight U.S. cities that
heating season properties of the high-gain ultra window and
represent a range of climates: Boston, MA; Seattle, WA;
the cooling season performance of the low-gain ultra window.
Denver, CO; Washington, D.C.; Kansas City, MO; Sacra-
Products with properties similar to windows #6-#9 are not
mento, CA; Jacksonville, FL; and Phoenix, AZ. Specific
currently commercially available. The development of
levels of insulation were determined for each location based
windows with electrochromic coatings and automated shad-
on Model Energy Code standards (CABO 1993), and base-
ing systems is expected to lead to the development of the
ment or slab-on-grade construction was chosen in accordance
dynamic properties that windows #6 and #9 have. Windows
with local practice. Homes were heated with a gas furnace
with U-factors in the range of windows #4 to #6 can currently
(AFUE1 = 0.78) and cooled with a 10.0-SEER2 air-condition-
be achieved with multiple layers, low-e coatings and gas fills,
ing unit. This set of simulations is referred to as “typical” and
insulating spacers, and frames. Windows with U-factors in the
is based on past efforts by the National Fenestration Rating
range of windows #7-#9 will require the development of more
Council (Arasteh et. al. 1999). To compare the effects of future
insulating components and products, as discussed in the
high-performance fenestration and the effects of shading strat-
“Emerging Technologies” section.
egies for reducing energy, homes with large roof overhangs
RESFEN was used to calculate several values for the
and deciduous trees were also simulated. Finally, to compare
application of each window: whole-house heating energy
the effects of high-performance windows to the effects of
(Mbtu/GJ), whole-house cooling energy (kWh), primary
improved insulation and HVAC systems, homes were
home HVAC energy consumption3 (Mbtu/GJ), window heat-
modeled with approximately doubled insulation levels and
ing energy consumption (“energies”) (Mbtu/GJ), window
improved HVAC system efficiencies, as well as higher-
cooling energies (kWh), and peak cooling demand (kW).
performing windows. Table 1 lists the details of these simula-
Baseline energies were calculated—HVAC energy consump-
tions.
tion not attributable to windows—by subtracting the total
Nine windows were simulated to represent a range of
annual window HVAC energies from total annual whole-
current and potential future window types. The whole-window
house HVAC energies. Because this value varied slightly from
SHGCs and U-factors for the nine simulated products are
window to window within a city, the value presented here is a
listed in Table 2. The first five windows represent a range of
numerical average.
currently available fenestration systems: double-glazed
windows with clear glass and a wood/vinyl frame (#1) are
SIMULATION RESULTS AND DISCUSSION
midrange products, argon-filled low-e windows in a wood/
vinyl frame typify higher-performance products on the market
Although simulations for all nine windows were
today (#2 and #3), and triple-glazed super windows (#4 and
performed in all eight cities, all nine windows are not appro-
#5) represent the most efficient one to two percent of today’s
priate for each climate. For example, it would not be sensible
market.
to install a high-solar-gain window in a Phoenix home. For this
reason, only one low-e window (#2 or #3), one super window
The final four windows presented in Table 2 represent a
(#4 or #5), and one ultra window (#7 or #8) is shown for each
range of next generation products. The dynamic window is
climate. In heating-dominated climates,4 such as Washington,
assumed to have the heating-season performance of the high-
D.C., and Kansas City, these windows have high solar gain. In
gain super window (#4) and the cooling season performance
cooling climates and Sacramento, these windows have low
of the low-gain super window (#5). In other words, it takes the
solar gain.
characteristics of today’s most efficient windows and adds
dynamic properties. We also defined two future high-perfor-
3.
mance static “ultra windows” that represent further improve-
We calculated total annual HVAC energy consumption in MBtu
by adding heating energy consumption to cooling energy
ments in the energy-efficient characteristics of today’s most
consumption multiplied by the conversion factor for kWh to
efficient windows; that is, these future ultra windows have
MBtu (0.003412) and a site-to-source conversion efficiency
very low U-factors (0.10 Btu/ft2⋅h⋅°F / 0.57 W/m2⋅°C), and one
factor of 3.22. We calculated total annual HVAC energy consump-
has a relatively “high” SHGC of 0.35, while the other a low
tion in GJ by adding heating energy consumption to cooling
energy consumption multiplied by the conversion factor for kWh
to GJ (0.0036) and a site-to-source conversion efficiency factor of
1. AFUE (annualized fuel utilization efficiency) is the amount of
3.22.
heat delivered by a furnace, divided by the latent heat of the fuel
4. For the purposes of this paper, heating-dominated climates are
the furnace consumes. A furnace with AFUE = 1 is perfectly effi-
climates in which high-solar-gain super windows use less energy
cient.
on an annual basis than low-gain super windows. In cooling
2. SEER (seasonal energy efficiency ratio, expressed in Btu/W) is
climates, low-solar-gain super windows use less energy than high-
the cooling output of an air conditioner, divided by the energy
gain super windows. In mixed climates, there is little difference in
input to the air conditioner.
energy performance between the two types of windows.
KC-03-12-3
3

---

TABLE 1

Construction Schemes (Adapted from Mitchell et al. 2002)
Scheme
Characteristics
Typical
Insulation and building systems:
• 1993 Model Energy Code levels of insulation (described in text)
• Gas Furnace AFUE = 0.78; AC SEER = 10.0
Shading
• Interior shades (seasonal SHGC multiplier, summer value = 0.80, winter value = 0.90)
• 1 ft (0.3 m) overhang
• a 67% transmitting same-height obstruction 20 ft (6 m) away, intended to represent adjacent buildings
• To account for other sources of solar heat gain reduction (insect screens, trees, dirt, building and window self-
shading), SHGC multiplier further reduced by 0.1, resulting in a final winter SHGC multiplier of 0.8 and a final
summer SHGC multiplier of 0.7
Typical + Overhangs
Same as above, but with 2-ft (0.6 m) overhangs instead of 1-ft (0.3 m) overhangs
Typical + Overhangs
Insulation and building systems
+ Trees
• Same as “Typical”
Shading
• Interior shades (seasonal SHGC multiplier, summer value = 0.80, winter value = 0.90)
• 2 ft (0.6 m) overhang
• A 10 ft (3 m) diameter obstruction 4 ft (1.2 m) above ground level, located 8 ft (2.4 m) away from the house;
zero-percent solar transmittance, March 15-Oct. 15; 60% solar transmittance, Oct. 15-March 15
• To account for other sources of solar heat gain reduction (insect screens, trees, dirt, building and window self-
shading), SHGC multiplier further reduced by 0.1, resulting in a final winter SHGC multiplier of 0.8 and a final
summer SHGC multiplier of 0.7
Double Insulation
Insulation and Building Systems
• Insulation levels approximately double those of 1993 Model Energy Code standards; locally specific
Shading
• Same as “Typical”
Double Insulation with
Insulation and Shading
Efficient HVAC
• Same as “Double Insulation”
Building Systems
• Simulated ultra efficient systems furnace AFUE = 0.95*; AC SEER = 16.0†
*
The annual energy consumption for AFUE = 0.95 for a furnace was calculated from the energy consumption for the simulated AFUE = 0.78 by multiplying this value by
the ratio of the two furnace efficiencies (0.821).
†
Estimates of energy savings from higher-rated air-conditioning systems were conservative. SEER measurements are misleading in that air conditioners with a higher rated
SEER do not necessarily increase efficiency proportional to the increase in SEER; thus, efficiency improvements for a 16 SEER unit over a 10 SEER unit would be less than
60% (Kavanaugh 2002). The annual energy consumption of a 16 SEER AC unit was calculated by reducing the annual energy consumption of the simulated 10 SEER AC unit
by 20%. Peak energy demand for the 16 SEER unit was calculated by multiplying the demand of the simulated 10 SEER unit by 0.943 based on a 6% increase in peak EER
found by Kavanaugh (2002).
TABLE 2

Window Types
U-Factor
(Btu/ft2⋅h⋅F)/
Window
(W/m2 ⋅°C)
SHGC
1
Double Clear (static)
0.49 / (2.78)
0.56
2
Low-e, high solar (static)
0.36 / (2.05)
0.53
3
Low-e, low solar (static)
0.34 / (1.93)
0.30
4
Super, high solar (static)
0.18 / (1.02)
0.40
5
Super, low solar (static)
0.18 / (1.02)
0.26
6
Dynamic
0.18 / (1.02)
0.26 or 0.40
7
Ultra, high solar (static)
0.10 / (0.57)
0.35
8
Ultra, low solar (static)
0.10 / (0.57)
0.10
9
Dynamic + Ultra
0.10 / (0.57)
0.10 or 0.35
Glazing systems 2-5 have 90 percent argon gas fill. For all windows, U-factor and SHGC are whole-window values for a 60 x 150-cm generic wood-vinyl frame.
4
KC-03-12-3

---

Figure 1 Total annual HVAC energy use.
Figure 2 Percent of savings in annual whole house energy use over low-e windows.
Annual Energy Performance Comparisons
Thus, low-e windows typically saved about 40% of the energy
use attributable to windows. Simulated homes incorporating
As the simulation results in Figure 1 show, homes with
the dynamic and the ultra-window technologies studied here
low-e windows in most climates used 5 to 10 Mbtu (5-10 GJ)
can use as little as or less energy than a home with no windows
less energy (8% to 15% less total house energy use) for heating
whatsoever; in other words, these future technology advances
and cooling than homes equipped with standard double-
can convert windows from energy liabilities to energy assets,
glazed windows. As a comparison, baseline energy consump-
which will be key in zero-energy construction.
tion—the energy consumption that would exist even if the
In heating-dominated and mixed climates, homes with
home had no windows—was around 12 to 25 Mbtu (13-26 GJ;
super windows (#4 or #5) used 10% to 11% less total house
20% to 40% of total house HVAC energy use) lower than
energy than homes equipped with low-e (#2 or #3) windows
energy consumption for homes with double-glazed windows.
[see Figure 2]. In cooling-dominated climates, the energy use
KC-03-12-3
5

---

of today’s super windows was 6% to 7% (of total house
Peak Demand Performance Comparisons
energy) lower than low-e windows. In all climates, energy
Figures 3 and 4 show peak cooling energy consumption
savings from super windows result primarily from the
and demonstrate that windows with low annual energy
windows’ very low U-factor; because cooling climates have a
consumption do not necessarily draw less power during
smaller indoor-outdoor temperature differential, the savings
important peak cooling periods. As discussed previously, in
from super windows are smaller in these climates.
heating-dominated and mixed climates, high-solar super
As Figure 1 shows, energy savings from dynamic
windows (#4 in Table 2) performed nearly as well as dynamic
windows (#6) were greater than from superwindows, but not
windows on an annual energy basis. However, in these
always significantly. In climates like those of Phoenix, Seat-
climates, with the exception of Sacramento, peak demand
tle, and Boston, which are all heavily dominated by one
savings for the dynamic window (#6) were almost twice as
season, dynamic windows used only 1% to 2% less total house
large as savings for a super window (#4). This corresponded
energy on an annual basis than super windows. However, in
to peak power demand in homes with dynamic windows (#6)
mixed climates like that in Sacramento, dynamic windows
that was 0.25-0.4 kW lower than the peak power demand for
increase whole house energy savings over low-e windows by
homes with super windows (#4).
another 9%.
In cooling-dominated climates and Sacramento, the trend
Dynamic and static ultra windows were compared in
was different. Since the dynamic window (#6) had the same
order to assess which technology might be most promising to
SHGC as the climate-appropriate super window (#5), the
pursue as the basis for the next generation of highly efficient
dynamic window offered no additional savings in peak cool-
windows. It was found that in climates with one dominant
ing power demand in these climates.
season, the static ultra windows outperformed dynamic
On an annual basis, the high-solar-gain ultra window (#7
windows. As Figure 2 shows, in heating-dominated climates,
in Table 2) consumed less energy than the dynamic window
homes with ultra windows consumed 14% to 17% less total
(#6) in heating-dominated and mixed climates. However, the
house energy than homes with today’s low-e windows. By
home with the dynamic window had lower peak cooling power
contrast, dynamic windows saved 12% to 16%. In cooling-
demand because of the low summer SHGC for this window.
dominated climates, energy savings from ultra windows were
Homes with the ultra window (#7) had cooling power
significantly greater than those from dynamic windows.
consumption 0.15 to 0.25 kW higher than similar homes with
However, in Washington, D.C., which is typical of mixed
the dynamic window (#6).
climates, homes with the high-solar-gain ultra window (#4)
In cooling climates, the trend was again different. The
and the dynamic window consumed roughly similar amounts
low-solar-gain ultra window (#8) used in these climates has a
of energy. In another mixed climate studied—Sacramento—
lower SHGC than the dynamic window (#6). Homes in cool-
the low-solar-gain ultra window (#6) saved significantly less
ing climates with the low-solar-gain ultra window (#8) had
energy than the dynamic window (12% vs. 14%, respectively).
peak cooling power demand about 0.3 kW lower than in
When the sole concern is to minimize annual energy consump-
comparable homes with the dynamic window (#6).
tion, static ultra windows deliver performance roughly on par
The dynamic + ultra window (#9 in Table 2) offered the
with the dynamic window (#6) in mixed climates, slightly
deepest reduction in peak cooling power demand in heating-
superior performance in heating climates, and more signifi-
dominated and mixed climates (except Sacramento). Savings
cantly superior performance in cooling climates.
were greatest in the most extreme climates and ranged from a
Window 9, the “dynamic + ultra” window, represents an
35% to 65% reduction in peak HVAC power demand from
outer bound in product design and performance, combining
homes equipped with the high-gain low-e window (#2). In
the low U-factors of ultra static windows with the solar-heat-
cooling climates, peak power consumption for window 9 was
gain properties of the dynamic window. Energy savings from
no lower than that for the low-solar-gain ultra window appro-
this combined window were 18% to 30% of total house energy
priate for these cities (#8). In all cases, the dynamic + ultra
use, greater (depending on climate) than the savings from
window had very low peak cooling power demand. This
today’s low-e windows. As with all the dynamic windows we
demand was only slightly greater (11% to 25%) than that of a
simulated, the dynamic + ultra window’s energy savings were
home with no windows whatsoever.
greatest in mixed climates (see Figures 1 and 2). Notably, in
Effects of Shading Strategies
heating and mixed climates, homes with dynamic + ultra
windows consumed less energy than the baseline “no
Increased shading of windows through the use of large
window” case, turning windows into a net energy benefit for
overhangs and trees was found to decrease summer cooling
the home.
energy consumption and increase winter heating energy
6
KC-03-12-3

---

Figure 3 Peak cooling energy use.
Figure 4 Percent of savings in peak cooling load over low-e.
KC-03-12-3
7

---

Figure 5 Annual energy: shading schemes compared.
Figure 6 Peak cooling loads: shading schemes compared.
consumption. Figure 5 shows total annual energy consump-
Shading strategies significantly improved the perfor-
tion for windows in a heating climate (Boston), a mixed
mance of all windows in cooling-dominated climates. As
climate (Washington, D.C.), and a cooling climate (Phoenix).
Figure 5 shows, intelligent use of deciduous trees coupled with
a conventional low-solar-gain low-e window (#3) in a climate
Shading strategies increased annual energy consumption
such as Phoenix can yield savings comparable to those from
in all heating-dominated cities (not shown) because the result-
future window technologies when not shaded. A combination
ing increased winter energy use outweighed the summer cool-
of advanced window technologies and shading led to the great-
ing energy savings. In mixed climates, shading strategies did
est possible energy savings.
not significantly affect the performance of static windows,
As Figure 6 shows, deciduous trees were effective in
although shading slightly increased the annual energy
reducing peak cooling demand in all climates. In several heat-
consumption of dynamic windows.
ing and mixed climates, tree-surrounded homes with high-
8
KC-03-12-3

---

Figure 7 Total annual HVAC energy use: insulation schemes compared.
Figure 8 Peak cooling load: insulation schemes compared.
solar-gain low-e windows had lower peak cooling than homes
when combined with more efficient building systems (Figure
with “typical” shading and dynamic windows. In cooling
7). Doubling insulation and improving mechanical equipment
climates, savings were more modest.
in a typical home can reduce annual energy consumption up to
In general, in cooling-dominated climates, where shad-
37%, depending on climate. In these cases, the impacts of
ing/overhangs are most effective, advanced window technol-
advanced windows are slightly lower in absolute terms but are
ogies are still required to bring window energy consumption
approximately equal in relative terms. Advanced windows are
to “zero” levels, although the absolute savings are less. In all
also a necessity if the energy impacts of windows are to be
climates, shading/overhangs contribute more toward peak
brought to “zero” levels.
reductions than energy reductions.
Peak demand savings for the double insulation and double
Effects of Improved Insulation
insulation with efficient HVAC cases were not as great as the
and Building Systems
annual energy savings, as Figure 8 shows. In all climates,
Doubled home insulation levels were found to dramati-
doubled insulation and doubled insulation with efficient
cally reduce annual home energy consumption, especially
HVAC helped reduce peak loads, but advanced windows are
KC-03-12-3
9

---

the main driver in reducing peak loads. Absolute kW reduc-
Cooling Climates
tions from window strategies are roughly the same in all three
1.
Dynamic window capabilities offer few savings in climates
cases (base case, doubled insulation, double insulation with
where cooling loads dominate and low SHGCs are the
efficient HVAC).
primary drivers of energy efficiency. Compared to high-
performance windows with similar U-factors, dynamic
SUMMARY AND CONCLUSIONS
windows produce relatively small annual energy savings
and no peak cooling demand savings.
We considered three different technology trajectories for
2.
Future window developments for cooling climates should
future fenestration products:
focus on achieving very low SHGCs (i.e. on static ultra
1.
Dynamic Windows—Windows with seasonally variable
windows) without excessively compromising visible trans-
solar heat gain properties to minimize peak summer cooling
mittance.
demand and maximize winter passive solar gain to lower
3.
Where heating loads are appreciable, dynamic windows
heating costs. These windows would have heat transfer
can have much lower winter energy costs than static
properties similar to those of today’s most highly efficient
windows with the same U-factors, with no change in
“super” windows (U
summertime peak cooling demand. This decreases results
IP = 0.18/ USI = 1.02) and have solar
heat gain coefficients that varied between the high and low
because the difference between summer and winter SHGCs
values offered by today’s super windows, 0.4 and 0.26,
is significant with dynamic windows.
respectively.
4.
Substantial peak and annual energy savings can be realized
with large roof overhangs and deciduous trees. Simulations
2.
Static Ultra Windows—Windows with significantly lower
showed that future window technologies with typical shad-
heat-loss rates (UIP = 0.10/ USI = 0.57) and fixed solar-heat-
ing save more energy than do shading strategies coupled
gain properties. Two different products were modeled: a
with today’s best windows.
high-gain product (SHGC = 0.35) for heating climates, and
a low-gain product (SHGC = 0.10) for cooling climates.
Mixed Climates
3.
Dynamic + Ultra Windows—Windows combining the
1.
Comparing today’s super windows with similar U-factors,
properties of dynamic and ultra windows (1 and 2 above);
dynamic windows achieve significant annual energy
savings. Dynamic windows do not require the trade-offs
these windows would have dynamic solar-gain control and
inherent between high-solar windows (heating energy
ultra-low heat-transfer rates (UIP = 0.10 / USI = 0.57) and
savings) and low-solar windows (cooling and peak demand
SHGC varying from 0.35 to 0.10).
savings).
Although these products were compared based strictly on
2.
Static ultra windows use a roughly equal amount of annual
performance, other factors will be important in determining
energy and significantly more peak energy than dynamic
which future window technologies are appropriate. These
windows. As would be expected, dynamic window technol-
factors include development and production costs, aesthetic
ogy is a great advantage in mixed climates.
appeal, durability, and installation and maintenance. These
3.
In mixed climates, dynamic + ultra windows can save
factors are outside the scope of our simulations, so we did not
significantly more annual energy than is saved by static
consider them explicitly. It is important, however, to note that
ultra windows and can dramatically lower peak energy
significantly reducing U-factors below current levels and
consumption.
developing dynamic windows will both present technical
4.
Shading strategies have little effect on annual energy
challenges.
consumption for static windows and a slight negative effect
It should be noted that the first dynamic windows on the
on annual energy consumption for dynamic windows.
Cooling savings are generally offset by heating energy
market will most likely not have the U-factors as low as those
increases. Use of overhangs or trees lowers peak demand.
noted in this study. Their dynamic ranges will also be different.
Such products may be extremely effective in reducing cooling
Heating Climates
loads and peak demands but may not be as effective in reduc-
1.
Dynamic windows moderately reduce energy consumption
ing heating loads as well.
on an annual basis and significantly reduce peak demand
The primary conclusion is that the future advanced fenes-
relative to static windows with the same U-factors.
tration products that we studied—dynamic, ultra, and
2.
Static ultra windows achieve greater annual energy savings
dynamic + ultra windows—offer the potential for significantly
than dynamic windows; however, dynamic windows have
greater HVAC energy savings than can be achieved with
significantly lower peak cooling demands. Either technol-
currently available high-performance windows. Specific
ogy trajectory—dynamic window or ultra window—could
conclusions are presented below according to climate type.
deliver significant improvements in performance.
10
KC-03-12-3

---