Study Shows That Expensive Windows Yield Meager Energy Returns
An engineer investigating ways to optimize the design of net-zero-energy homes concludes that inexpensive triple-glazed windows are good enough
An architectural cliché from the 1970s — the passive solar home with large expanses of south-facing glass — is making a comeback. In recent years, we’ve seen North American designers of PassivhausA residential building construction standard requiring very low levels of air leakage, very high levels of insulation, and windows with a very low U-factor. Developed in the early 1990s by Bo Adamson and Wolfgang Feist, the standard is now promoted by the Passivhaus Institut in Darmstadt, Germany. To meet the standard, a home must have an infiltration rate no greater than 0.60 AC/H @ 50 pascals, a maximum annual heating energy use of 15 kWh per square meter (4,755 Btu per square foot), a maximum annual cooling energy use of 15 kWh per square meter (1.39 kWh per square foot), and maximum source energy use for all purposes of 120 kWh per square meter (11.1 kWh per square foot). The standard recommends, but does not require, a maximum design heating load of 10 W per square meter and windows with a maximum U-factor of 0.14. The Passivhaus standard was developed for buildings in central and northern Europe; efforts are underway to clarify the best techniques to achieve the standard for buildings in hot climates. buildings increase the area of south-facing glass to levels rarely seen since the Carter administration.
What’s the explanation for all this south-facing glass? We’re told that there’s no other way for designers to meet the energy limit for space heating required by the Passivhaus standard: namely, a maximum of 15 kWh per square meter per year.
Struggling to meet this goal, many Passivhaus designers have found that the typical triple-glazed windows sold in North America have U-factors that aren’t quite low enough (or SHGCs that aren't quite high enough) for their designs to meet the standard. Because of this, these designers often end up specifying very expensive triple-glazed windows from Germany or Austria.
What about cost-effectiveness?
As I have often noted, these Herculean efforts to meet the Passivhaus standard pay no attention to cost-effectiveness. Even when designers find it necessary to invest in measures that are much more expensive than a photovoltaic(PV) Generation of electricity directly from sunlight. A photovoltaic cell has no moving parts; electrons are energized by sunlight and result in current flow. array, they plow ahead because they have to meet the numbers dictated by the PHPP software.
These investments in very expensive building materials are probably a waste of money. An excellent paper by Gary Proskiw, “Identifying Affordable Net Zero Energy Housing Solutions,” looks into the cost-effectiveness of large expanses of south-facing glazing as well as the cost-effectiveness of low-U-factor windows. Proskiw, a mechanical engineer from Winnipeg, Manitoba, who specializes in residential energy issues, concludes that heroic window measures don’t pay worthwhile dividends.
Proskiw’s analysis and conclusions are fascinating and thought-provoking, and I believe that most designers of low-energy homes will want to read Proskiw's paper. (I'd like to credit GBAGreenBuildingAdvisor.com reader Sasha Harpe for alerting me to the existence of the report.)
Proskiw’s paper focuses on ways to optimize the design of a net-zero-energy (NZE) house. While the paper mostly focuses on Canadian climates, it considers one U.S. home (the net-zero-energy house built by Habitat for Humanity in Wheat Ridge, Colorado) in its analysis.
The value of extra south-facing glazing
Proskiw asked an interesting question: should the designer of a superinsulated home add extra south-facing windows “to increase solar gains and reduce the space heating load”? He tackled the question by analyzing the cost of this measure and then comparing the cost to the energy benefit.
For the purpose of his analysis, he considered an 1,800-square-foot net-zero-energy house located in Winnipeg, Manitoba. He assumed that the base-case house had R-44 exterior walls and south-facing glazing with an area equal to 6% of the floor area. According to his analysis, the cost to build an R-44 wall is $170 per square meter ($15.80 per square foot). What if the designer chose to add another south-facing window — one measuring 1 square meter (10.76 square feet)?
Proskiw estimated that a triple-glazed, low-e, argonInert (chemically stable) gas, which, because of its low thermal conductivity, is often used as gas fill between the panes of energy-efficient windows. -filled fixed window measuring one square meter costs $488. Proskiw estimated that the net cost of adding this window to the house would equal the cost of the window minus the cost of the displaced wall area; he calculated the incremental cost this way: $488 - $170 = $318. (Proskiw knows that this calculation method underestimates the cost of adding a window, since it ignores the costs associated with framing the rough opening, the cost of the header, and the cost of trimming the opening; however, if these costs are included, Proskiw's argument is only strengthened.)
The base-case house had a modeled energy consumption of 1,462 kWh per year. Adding the extra window resulted in a modeled energy consumption of 1,443 kWh per year. In other words, the extra window saved only 19 kWh per year, which Proskiw valued at $1.90. He calculated that the payback period for this measure is 167 years. “Given that the life expectancy of an insulated glazing unit (IGU) is about 25 years, it is clear that inclusion of the extra 1 square meter of south‐facing window area can never be economically justified,” Proskiw wrote. “From a design perspective, these results indicate that increasing the amount of window area in a NZE house, as an energy saving measure, has to be examined extremely carefully since it is unlikely to be economic relative to other options.”
According to energy expert Marc Rosenbaum, adding a south-facing window measuring 1 square meter (gross area) on one of his Massachusetts house designs (a house from the Eliakim's Way development on Martha's Vineyard) would save 120 kWh a year (worth about $12 a year, according to Proskiw's method). If the window could be installed for an incremental cost of $318 — (the actual incremental cost is likely to be higher) — the simple payback period would be 26 years.
What about investing in really good windows?
Proskiw also compared two different types of triple-glazed window:
- The less expensive option ($360 per square meter) was a “relatively conventional triple-glazed unit with an insulated spacer.” This window did not include argon gas or any low-e coatings.
- The more expensive option ($488 per square meter) was a “triple-glazed unit with one low-e coating, two argon fills, and an insulated spacer.” The incremental cost for this window was: $488 - $360 = $128.
Proskiw assumed that the window measured one square meter and faced south. The energy savings attributable to the glazing upgrade was 8 kWh per year, which Proskiw valued at $0.80. The upgraded glazing had a simple payback period of 160 years.
Expensive glazing doesn’t make economic sense
Proskiw wrote, “The reason the two window upgrades fared so poorly, from an economic perspective, is that the space heating load in a NZE house is very small compared to any other type of house. By adding window area or upgrading window performance, the space heating load is reduced but it is already so small that there is little opportunity for further savings.”
He went on to note, “The preceding discussion used the incremental analysis of costs and
benefits to illustrate the economics of adding glazed area and of upgrading windows in a Net Zero Energy House. Although it used single examples, the process could be easily used for other windows in other houses. A more rigorous analysis, using a wider range of windows, houses, locations, etc., would yield similar results in most cases.”
What’s the lesson for designers of superinsulated homes? “Since windows and their upgrade options are so expensive, the investment would often be better spent on improving the energy performance of some other conservation or renewable energy option,” Proskiw concludes.
Inexpensive triple-glazed windows are fine
One way to summarize Proskiw’s findings: builders of superinsulated homes in cold climates should choose affordable (usually triple-glazed) windows rather than exotic windows with extremely low U-factors.
“From an energy perspective and based on the incremental costs and energy savings, window selection should be based solely on the need to control condensation,” Proskiw wrote. In most cases, that means that you should choose glazing with a warm-edge spacer. “Further, the window area should be limited to that necessary to meet the functional and aesthetic needs of the building. As such, south‐facing glazing area should be restricted to 6% [of the conditioned floor area] (to control overheating) and total window area should also be limited to that required for functional and aesthetic considerations. On a broader level, these results indicate that our long‐held belief in the merits and value of passive solar energy as a key component of Net Zero Energy House design need to be carefully re‐examined and likely challenged.”
Simplicity is a virtue
Proskiw’s paper includes many other nuggets of wisdom that are worthy of attention from designers of superinsulated homes. For example, he noted that low levels of air leakage are easier to achieve if the building’s envelope has a simple shape. Proskiw advised, “Draw out complicated details. If you can’t draw it, you probably can’t build it.”
When it comes to space heating and water heating equipment, Proskiw advocates simplicity. “One of the most common problems (both observed and reported) with Net Zero Energy Housing has been the complexity of the mechanical systems (space heating, domestic hot water heating, ventilation and cooling). While there may be a temptation to use every thermodynamic opportunity to maximize performance, the reality is that complex mechanical systems almost always prove to be problematic, expensive and far too unreliable. Perhaps the most trouble-prone example has been seasonal heat storage systems which attempt to capture and store large amounts of energy between seasons. While technically feasible, such systems are usually extremely expensive, produce nominal savings and may require the homeowners to adopt a full‐time repairman as a live‐in family member.”
Solar thermal systems and buyers' remorse
Proskiw went on, “The need to simplify mechanical systems was arguably the most consistent comment offered by designers during the interview phase of this project. For example, one designer had used a solar thermal system in conjunction with a GWHR [gray water heat recovery] system and a desuperheater on a heat pumpHeating and cooling system in which specialized refrigerant fluid in a sealed system is alternately evaporated and condensed, changing its state from liquid to vapor by altering its pressure; this phase change allows heat to be transferred into or out of the house. See air-source heat pump and ground-source heat pump. — three separate technologies to heat water. Overall, he found the solar thermal system to be leak‐prone and not as effective as originally hoped. In retrospect, he felt that it would have been preferable to use additional photovoltaic capacity in place of the solar thermal system since most of the DHW [domestic hot water] heating was provided by the geothermal and GWHR systems. Perhaps the issue of mechanical system complexity was best captured up by one NZEH designer who summed up his approach: ‘Just say no!’”
Among the lessons learned: “Some of the surveyed designers who had used solar thermal systems for DHW heating stated that given the problems with reliability and performance, they would have replaced the system with a larger PVPhotovoltaics. Generation of electricity directly from sunlight. A photovoltaic (PV) cell has no moving parts; electrons are energized by sunlight and result in current flow. array and a (relatively) conventional electric DHW heater.”
An interesting postcript
In January 2013, I received copies of an e-mail exchange between Peter Amerongen, the developer of the Riverdale Net Zero project in Edmonton, and Gary Proskiw.
Amerongen wrote (in part), “I think Gary is underestimating the benefits of additional south glazing area. ... That is a small quibble. The bigger issue is that if you limit conservation to benefits that are less expensive than PV, you will never reach net zero in our part of the world.” (Peter Amerongen's full comments can be read below; they are posted as Comment #62.)
Gary Proskiw responded, “Let me try to explain the problem with windows. Imagine a somewhat conventional house with a gross monthly heating load as shown by the red line in the graph below:
“Also shown are the monthly base loads (averaging around 500 kWh/month). The difference between the gross space heating load and the base loads is the heating load which has to be supplied by the heating system. For the conventional house, this means that the heating season will extend for about nine months per year - as shown by the range ‘A.’
“For a net-zero-energy house, the gross monthly heating load is much smaller, as shown by the green line. This means that the heating season for the net-zero house will last about four months per year, as shown by the range ‘B.’
“Now, when additional window area is added to the house there is greater opportunity for passive gains. However, they are only of value if the house can actually utilize the gains. In the conventional house, that means the windows could theoretically reduce the heating load over a nine month period. But in the net-zero-energy, the ‘window’ during which these benefits can be realized is less than one-half that of the conventional house. Further, the months in which the net-zero house could theoretically use these additional solar gains occurs during the period of the year when solar gains are at a minimum.
“In other words, adding a square meter of window area to a net-zero house will produce much less benefit than adding that same square meter of window area to a conventional house. Of course, the same holds true for conservation (adding one more batt of insulation to a net-zero house will produce less benefit than if that batt were added to a conventional house). The difference is that one square meter of window area costs about ten times as much as (say) an equivalent amount of wall area, so the consequences of excess glazing area are economically more problematic than excess insulation.
“Basically, here's the problem. When one square meter of window area is added to a house, two things happen: the gross space heating load is increased (since the window has a lower R-valueMeasure of resistance to heat flow; the higher the R-value, the lower the heat loss. The inverse of U-factor. than the wall area which it replaced) and the passive solar gains are increased. Unfortunately, the increased space heating load is present 24/7 during the heating season. The extra passive gains are also available, but can only be utilized during a portion of the heating season (unless overheating is allowed to occur). As the house becomes more energy efficient, that ‘portion’ becomes increasingly smaller. To illustrate, the percentage of the solar gains which were actually usable by the house in the two cases below was 33% for the conventional house and 28% by the net-zero house, as calculated by HOT2000.
“One can then make the argument for thermal storage to facilitate greater utilization of passive gains; however that is usually a very expensive proposition which produces nominal benefits.”
[Editor's note: to read Peter Amerongen's response to these comments by Gary Proskiw, see Comment #93, posted below.]
Martin Holladay’s previous blog: “Who Deserves the Prize for the Greenest Home in the U.S.?”
- Image #1: GO Logic
- Image #2: Martin Holladay
Sep 14, 2012 6:34 AM ET
Sep 14, 2012 7:39 AM ET
Sep 14, 2012 10:03 AM ET
Sep 14, 2012 10:20 AM ET
Sep 14, 2012 10:44 AM ET
Sep 14, 2012 10:48 AM ET
Sep 14, 2012 2:24 PM ET
Sep 14, 2012 2:45 PM ET
Sep 15, 2012 1:29 AM ET
Sep 15, 2012 4:28 AM ET
Sep 15, 2012 6:03 AM ET
Sep 15, 2012 7:38 AM ET
Sep 15, 2012 7:42 AM ET
Sep 15, 2012 8:08 AM ET
Sep 15, 2012 9:00 AM ET
Sep 15, 2012 9:13 AM ET
Sep 15, 2012 9:23 AM ET
Sep 15, 2012 9:53 AM ET
Sep 15, 2012 2:05 PM ET
Sep 15, 2012 2:17 PM ET
Sep 15, 2012 3:37 PM ET
Sep 15, 2012 7:25 PM ET
Sep 16, 2012 12:19 PM ET
Sep 16, 2012 3:16 PM ET
Sep 16, 2012 3:32 PM ET
Sep 16, 2012 4:00 PM ET
Sep 16, 2012 5:56 PM ET
Sep 16, 2012 8:54 PM ET
Sep 16, 2012 9:25 PM ET
Sep 16, 2012 10:32 PM ET
Sep 17, 2012 4:28 AM ET
Sep 17, 2012 9:33 AM ET
Sep 17, 2012 9:55 AM ET
Sep 17, 2012 10:56 AM ET
Sep 17, 2012 12:16 PM ET
Sep 17, 2012 12:41 PM ET
Sep 18, 2012 12:17 AM ET
Sep 18, 2012 3:23 AM ET
Sep 18, 2012 8:57 AM ET
Sep 18, 2012 3:13 PM ET
Sep 19, 2012 10:52 AM ET
Sep 19, 2012 7:02 PM ET
Sep 19, 2012 11:39 PM ET
Sep 19, 2012 11:49 PM ET
Sep 20, 2012 7:37 AM ET
Sep 20, 2012 3:03 PM ET
Sep 20, 2012 3:18 PM ET
Sep 20, 2012 3:25 PM ET
Sep 20, 2012 5:11 PM ET
Sep 20, 2012 8:02 PM ET