For a long time, the thermal performance of façades in research buildings has been undervalued because of the large volumes of air being moved through the space. Using the arsenal of strategies currently available to lab designers—such as chilled beams and low-flow fume hoods—air volumes in many contemporary research labs have been reduced to the minimums needed to maintain health and safety, and the performance of the building envelope is having an increasing influence on the building’s energy usage and thermal comfort.
Thermal bridging in building construction is a well-understood phenomenon frequently resulting from structural elements that penetrate through insulation layers and create a path for unimpeded heat transfer. While the construction industry has begun to develop materials and assemblies intended to mitigate this effect, there is little available research documenting the extent of the problem or the performance benefit that will result from the use of these new products. Anecdotal reports suggest that thermal bridges in conventional construction may reduce insulation effectiveness by as much as 40%.1
Payette received funding from the American Institute of Architect’s Upjohn Research Grant to better understand and quantify the impact of thermal bridging on the overall performance of the building envelope. The intent of this research is to bring rigor to the investigation of thermal bridges in commercial construction by using thermal imaging equipment to quantify actual performance of built installations; and use these results along with heat-transfer modeling software to suggest and then analyze performance improvements. Preliminary results suggest that through the use of readily available construction materials and careful detailing, it’s possible to effect a 50% or greater reduction in the impact of common thermal bridges (Figure 1).
In order to understand how façades are performing in the field, we used a thermal imaging camera to determine the R-value and identify sources of thermal bridges in recently completed projects designed by the firm. Teams were deployed to locate and document a range of façades and conditions. Using the methodology tested by Madding,2 we were able to measure the exterior air temperature, interior air temperature and radiant temperature in order to calculate the R-value of the assembly.
We collected thousands of images from visits to 15 buildings. These images were then organized by assembly type, and we noted conditions that were likely to affect performance, such as the transition to a foundation wall or adjacency of a window. Having established a library of data that was primarily focused on thermal bridging, the research team was able to identify typical problem areas thematically. We noted that they fell generally into two categories: one that is related to structure that supports façade and roof systems, and one that is more about material transitions.
- Existing building façade renovations.
- Masonry wall systems.
- Metal panel wall systems.
- Curtainwall systems.
- Rainscreen wall systems.
Transitions and penetrations:
- Transitions between new and existing façades.
- Transitions between different wall systems.
- Transitions between windows and walls.
- Foundation-to-wall transitions.
- Roof-to-wall transitions.
- Roof parapets.
- Roof penetrations.
- Seismic and movement joints.
- Louver openings.
Using heat flow simulation software, such as Lawrence Berkley National Laboratory’s THERM, it’s possible to study alternative designs. The research team prepared THERM models of the areas being studied which were calibrated to the performance measured in the field with the thermal imaging camera. With validated THERM models in place, the research team is currently testing the quantitative impacts of potential design improvements.
The THERM modeling platform enables us to probe a detail and effectively measure the changes in heat transfer associated with different detailing or materials selections. Using this approach, we’re able to make quantifiable recommendations for improvements that can then be rigorously evaluated on a lifecycle basis for a specific project.
Case study: Interior spray foam in existing buildings
Preliminary findings in our research show the actual R-value of many façades is approximately 40 to 70% less than the design-intended R-value, so our findings suggest far greater significance than was originally anticipated. As the amount of insulation we specify continues to increase, the conductive losses resulting from thermal bridging will continue to grow as a percentage of the building’s total energy load. Adding more insulation will have a diminishing return as the heat flow through the envelope is increasingly determined by the thermal bridges. Our research suggests that we are near a point now where the key to improving thermal performance lies in better detailing rather than increasing insulation thickness. A simple illustration of this is featured in the renovation case studies of three existing masonry façades.
Spray-applied insulation is once again gaining popularity particularly because of its ability to fill unseen voids and often provide an integral vapor barrier. In the northeast, it’s a particularly popular technology for renovating existing un-insulated masonry façades. Conventional details often call for metal studs to support interior gypsum board, and spray foam is installed between the studs following manufacturers’ recommendations. Unfortunately, this typical installation creates discontinuities at 16-in or 24-in center spacing. While the web of the steel stud is quite slender, they are highly effective heat-transfer devices because of the conductivity of the material and the flanges which provide significant contact area to collect and disperse heat.
Thermal images of three existing building’s renovations revealed dramatically different results. The first case, Building 1, had applied 3-in of insulation, Building 2 employed 2-in of insulation and Building 3 had used 4-in (Figure 2). While hand calculations of the thermal resistance would show the façade with the least insulation to be the poorest performer and the one with the most insulation to be the best, the thermal images revealed a different story. Building 1 had placed the steel studs flush against the exterior construction, resulting in an R-value that was 54% less than the calculated R-value. Building 2 pulled back the studs 1-in, allowing for half of the applied insulation to be continuous and decreasing the R-value by only 16%. As a result, Building 2 was observed to have a higher R-value than Building 1, despite having less insulation. Building 3 took the studs even further back, completely separating them from the insulation and resulting in a simulated R-value that was essentially identical to the design intention (Figure 3).
Our study showed that the continuity of the first inch is critical for the efficiency of the spray foam insulation performance. By simply pulling the studs in-board, even by a small amount, to allow a percentage of the insulation to be uninterrupted, the assembly R-value can be increased by about 70%. In the event that the studs are required to support exterior sheathing, it should be possible to fasten the sheathing using discontinuous shims or spacers so that once again, the majority of the insulation in that outer 1-in layer remains continuous. It’s important to remember that other factors—such as the continuity of the slab through the insulation, or window openings—will often decrease the thermal performance from our ideal conditions. But these too can be improved through careful detailing. The ultimate lesson is that small changes in the design that permit as little as an inch of continuous insulation can lead to dramatic improvement in overall performance.
Case study: Rainscreen supports
Rainscreens are a common exterior cladding used in the design of contemporary research labs. Because the insulation typically lies behind the cladding and ventilated cavity, but before the supporting stud or structural wall, the rails, hat channels, Z-girts and clips required to support that exterior rainscreen all become thermal bridges. Typically made of highly conductive aluminum, continuous supports such as Z-girts were observed to decrease the R-value of the assembly by 45 to 60% from the design-intended R-value. Rainscreens that employed discontinuous supports, like clips, showed a significant improvement in thermal performance, though the R-value was still determined to fall short of the theoretical design value by 15 to 25%.
Because supports like Z-girts are so widespread in commercial construction today, and have been noted in the industry as common thermal bridges, a number of manufactures have developed alternative solutions. These thermally broken, off-the-shelf products have been developed to meet the structural requirements to transfer the load of the exterior cladding, while not allowing continuous aluminum to pass through the insulation. By limiting the penetrations to only screws and fasteners, and providing insulating pads between metal attachments, we can now achieve the elusive goal of providing continuous insulation. Based on our heat flow simulations, these thermally broken support systems allow a wall to achieve an R-value that is only about 5% less than the clear wall R-value. While each manufacturer has their own strategy deterring the flow of heat, the options that exist give designers the opportunity to find the rainscreen attachment system that works best for their project conditions.
The significant impact observed from thermal bridging suggests a more substantial problem than simply one of condensation risk in a humidified lab environment. The data collected thus far shows that thermal bridging can easily double the conductive heat transfer over what was intended for the design. As we strive to reduce air change rates within the lab environment, we are seeing that envelope heating and cooling loads are an increasingly significant portion of the building energy demand. This research shows that thermal bridging already has a significant impact on the performance of our buildings, and that controlling this path for heat transfer is the key to achieving to truly high-performance façades.
Not surprisingly, the majority of the typical problems identified in this research have centered on transitions between systems and the structural assemblies necessary to support exterior cladding. While many building products exist on the market and it’s easy to implement alternative designs that improve thermal performance, it’s clear that not all manufacturers are equally sensitive to the magnitude of the problem and marketing material must be scrutinized carefully. As we seek to achieve higher-performing research buildings, careful attention and analysis is needed during design to minimize thermal bridges and deliver buildings that perform as anticipated.
1. Morrision Hershfield. (2011). “Thermal Performance of Building Envelope Details for Mid- and High-Rise Buildings” (ASHRAE 1365-RP). Atlanta: ASHRAE Technical Committee 4.4.
2. Madding, R. (2008). “Finding R-Values of Stud Frame Constructed Houses with IR Thermography.” Inframation 2008 Proceedings vol. 9. Reno.
Charles Klee drives Payette’s Research and Innovation Initiative, bringing a detailed understanding of emerging building science and sustainable technologies. Andrea Love leads Payette’s sustainability initiatives. Combining her background in architectural practice and building science, she provides sustainable design knowledge and energy expertise.