Mind the Gap: Summary of Window Residential Retrofit Solutions
Improving the insulation, solar heat gain, and infiltration characteristics of windows can significantly improve overall thermal performance by reducing heat transfer through the window and by decreasing air leakage into and out of a home. As approximately 43% of existing homes (~50 million houses) still have low-performing, single-pane clear-glass windows, and millions of other homes have only double-pane clear-glass windows, improving window performance can also save a lot of energy.
Today, various energy-saving window retrofit solutions are available to homeowners, ranging from window shades and storm windows to highly insulating triple-pane R-5 window replacements. Many of these technologies have been evaluated in the field, in the Lab Homes at Pacific Northwest National Laboratory (PNNL), and through modeling to prove their cost-effectiveness and performance in different climate regions. The resulting data are used to increase market penetration of these efficient technologies.
Recently, for example, PNNL’s Regional Technical Forum approved a utility measure for low-e storm windows based on such data. This was a watershed moment for increasing fenestration options in utility programs—especially low-income programs. This article will review the most recent field test and discuss future rating efforts. For the complete study, including a review of past research and full references to previous work, as well as detailed descriptions of window technologies, window testing, and efforts to support window retrofits and show their cost-effectiveness, see “learn more” at the end of the article.
PNNL Lab Homes Experiments
The experiments described below were conducted in the PNNL side-by-side Lab Homes, which form a platform for precisely evaluating energy-saving and grid-responsive technologies in a controlled environment. The PNNL Lab Homes are two factory-built homes installed on the PNNL campus in Richland, Washington. Each Lab Home has seven windows and two sliding glass doors, for a window-to-floor area of 13.7%. To be representative of the Pacific Northwest climate, clear double-pane windows were installed as the baseline technology for both PNNL Lab Homes. For the primary experiments examined in this study, the “experimental home” was retrofitted with the fenestration technology under evaluation, while a matching “baseline home” remained unaltered. The floor plan of the Lab Homes as constructed is shown in Figure 1. The Lab Homes are meant to represent typical Pacific Northwest housing stock and include R-11 cavity insulation and R-22 floor and ceiling insulation. However, the building shell is relatively tight (~3.2 ACH50) due to the capabilities of the manufactured housing fabrication process.
The metering approach includes metering and HVAC system control activities taking place at the electrical panel. Heat transfer through the primary glass and window fenestration will be aggregated by differing temperature sensors. For all experiments, metering was completed using Campbell Scientific data loggers and matching sensors. Two Campbell data loggers were installed in each home, one allocated to electrical measurements and one to temperature and other data collection. Data from all sensors were collected via cellular modems that were individually connected to each of the loggers. All data were captured at one-minute intervals by the data loggers. The one-minute data were averaged over hourly and daily time intervals to afford different analyses. Occupancy in the homes was simulated via a programmable commercial lighting breaker panel (one panel per home), using motorized breakers. These breakers were programmed to activate connected loads on schedules to simulate human occupancy by introducing heat to the space.
R-5 and Low-E Storm Windows
From March 2013 through December 2014, PNNL evaluated commercially available exterior and interior low-e storm windows in the PNNL Lab Homes for their thermal performance, including energy savings and impact on interior temperature distributions. The interior and exterior windows were tested separately over the experimental period. Each product was installed to the manufacturer’s specifications. To demonstrate and visualize the thermal impact of the low-e storm windows, infrared images were taken of both the baseline windows (in the baseline home) and baseline windows equipped with interior or exterior low-e storm windows (in the experimental home). These images, presented in Figures 2 and 3 for interior low-e storm windows, help to visualize the temperature differential between the conditioned space and outdoor air, and the interior images (Figure 3) help to visualize the impact of the storm windows on thermal comfort.
The images were taken on February 2, 2015. During this day, the average outside temperature was 34ºF with a low of 17ºF and a high of 40ºF. Reviewing the exterior images (Figure 2) shows the effect of the interior storm windows on the envelope. The external surface temperature of the baseline window in the baseline home was measured to be 43ºF as compared to 39.4ºF for the baseline window with interior storm windows in the experimental home. The change in temperature between the two surfaces is 3.6ºF, demonstrating the increased insulating quality of the baseline window with the interior storm window. This interior storm window reduces the amount of heat that is transferred though the baseline window and thus keeps the temperature of the baseline window closer to that of the outdoors.
On the same day, measurements were taken of the internal temperature of the master bedroom glazing (Figure 3). In the baseline home, the internal temperature of the baseline window was 59.9ºF, while the internal temperature of the baseline window in the experimental home was 66.6ºF, a differential of 6.7ºF. This differential can also be attributed to the insulating properties of the interstitial space and the low-e coating on the interior storm windows. The low-e surface was more effective at trapping and reflecting the internal heat back into the space, resulting in higher interior surface glass temperatures.
Both interior and exterior storm windows were evaluated in the PNNL Lab Homes throughout one winter heating and one summer cooling season. During the heating season, the heat pump was disabled, and a forced-air electric-resistance furnace supplied the required heating to the Lab Homes. In general, measured HVAC savings due to the exterior storm windows averaged 10.5% for the heating season and 8.0% for the cooling season for identical occupancy conditions. Yearly savings for this technology is estimated to be 10.1 ±1.4% of the HVAC Load. Because of limitations on the manufactured size of interior storm windows, only an estimated 74% of the window area could be covered during the test. The collected data showed that interior low-e storm windows installed over 74% of the window area in the experimental home resulted in an 8.1 ±1.9% and a 4.2 ± 0.7% reduction in HVAC energy use during the heating and cooling seasons, respectively. Yearly savings for this technology is estimated to be 7.8 ±1.5% of the HVAC Load.
The average annual energy savings calculated from the measured PNNL Lab Homes data for the interior and exterior low-e storm window evaluations are summarized in Table 1.These savings are compared to the average annual HVAC energy savings, 12.2 ±1.3%, estimated from primary window replacement with highly insulating R-5 windows. Generating annual savings estimates for the individual technologies is done through simple heating-and-cooling degree- day calculations coupled with the experimental savings shown for each season. Details of this process can be found in the research papers highlighted in the “learn more” section.
High-Efficiency Cellular Shades
PNNL evaluated cellular shades in the PNNL Lab Homes during the 201516 heating and cooling seasons. The technology examined as part of this study was the Hunter Douglas Duette Architella Trielle honeycomb fabric shade, which is made with six layers of fabric, including two opaque layers and five insulating air pockets. The inclusion of insulating air pockets as well as the layer of metallized Mylar that lines the air pockets, minimizes conductive and radiant heat transfer and effectively increases the R-value of the fabric.
Table 2. Average HVAC Savings Produced by Cellular Shades During the 2016 Heating and Cooling Seasons
These cellular shades were also equipped with the Hunter Douglas Green (HD Green) automated scheduling technology, which allowed researchers to evaluate both the thermal improvement attributable to the shades and the impact of automated shading devices on the optimal management of solar gains. To independently evaluate the functionality of the shades and the associated automation schedules, as well as their collective performance, PNNL tested the cellular shades in comparison with three other baseline technologies—no window attachments, vinyl blinds operated per the HD Green schedule, and vinyl blinds that remained closed. The results of the tests are shown in Table 2.
JM Petersen, GP Sullivan, KA Cort, MB Merzouk, and JM Weber. October 2015. Evaluation of Interior Low-E Storm Windows in the PNNL Lab Homes. PNNL - 24857, Pacific Northwest National Laboratory, Richland, WA.
JM Petersen, GP Sullivan, KA Cort, and CE Metzger. December 2016. Evaluation of Cellular Shades in the PNNL Lab Homes. PNNL – 24857 Rev 2, Pacific Northwest National Laboratory, Richland, WA.
Knox, JR, and SH Widder. 2014. Evaluation of Low-E Storm Windows in the PNNL Lab Homes. PNNL-23355, Pacific Northwest National Laboratory, Richland, WA.
Widder, SH, and GB Parker. 2012. Side-by-Side Field Evaulation of Highly Insulating Windows in the PNNL Lab Homes. PNNL-21678, Pacific Northwest National Laboratory, Richland, WA.
Because the data presented in Table 2 are preliminary results, the savings have not been annualized over the heating and cooling seasons. Insulating values associated with cellular shades alone resulted in a reduction in total HVAC system load of 10.5–16.6% when compared to standard vinyl blinds, as shown in the static-operation experiment. The implementation of the HD Green schedule as shown in the optimum-operation comparison experiment, resulted in savings of 15.3 – 16.6%. The addition of automated insulating shades to a home with no window attachments at all resulted in achievable savings—14.4 – 14.8%.
Studies conducted at the PNNL Lab Homes have demonstrated the energy savings potential of low-e storm windows and cellular shades with automated scheduling. Additional studies have modeled the cost-effectiveness of these technologies, making them a compelling retrofit option for homes, including multifamily buildings. The addition of energy ratings and labels for window attachment programs should help consumers, designers, and home performance contractors make informed decisions about window retrofit options. Current data prove the energy-saving potential and cost-effectiveness of most window attachment technologies. Further utility incentives and market transformation efforts will encourage builders to adopt these technologies—which is the next step to saving some of vast amount of energy we lose through our windows each year.
Enter your comments in the box below:
(Please note that all comments are subject to review prior to posting.)
While we will do our best to monitor all comments and blog posts for accuracy and relevancy, Home Energy is not responsible for content posted by our readers or third parties. Home Energy reserves the right to edit or remove comments or blog posts that do not meet our community guidelines.