On a winter morning threatening snow, Matthew Davis stands in the atrium of the nearly complete Simons Center for Geometry and Physics on the Stony Brook University campus. The 39,000-square-foot research center on New York’s Long Island, designed by Perkins Eastman, is on track to achieve LEED Gold certification. Built as a showcase for not just sustainable practices but, more conceptually, for the celebration of mathematics and physical science, it’s the perfect spot to see examples of a newly developed kinetic shading system. The system was developed by the Adaptive Building Initiative (ABI)—a joint venture of Hoberman Associates, a multidisciplinary design practice, and engineering consultancy Buro Happold—in partnership with metals fabricator A. Zahner Co.
Davis, Hoberman’s vice president of engineering, stares intently at the four panels installed just inside the Simons Center’s glazed façade. Muted sun shines through a latticework that recalls ornament from almost every culture and era: interlocking Arabic motifs, Arts and Crafts textiles, and Op Art illusions. Simple geometries make up the diffusing mesh; for the Stony Brook installation, ABI designed four different panels. Each one offers rows and columns of a single geometric shape: hexagons, circles, squares, or triangles, both small and large. (Other shapes, such as the rectangles, are also possible.) But what starts out as straightforward quickly turns complex.
Each panel is made up of thin layers of perforated stainless steel; these are supported by a steel frame and encased in glass. The front and back layers of steel are fixed, and a linear actuator motor at the panel’s base powers a small piston that drives the “tumblers” that mechanically move the independent internal layers. As the tumblers rotate, the pattern varies. Thus each panel can display a multitude of kaleidoscopic patterns. According to Davis, each tumbler has a central axle that structurally ties the unit together from front to back and that also functions as a common pivot point, helping define the path of movement for the different layers. (The profile of each tumbler is designed to blend in with the retracted, simplest pattern of the unit.) Each petal of the flower-shaped tumbler connects to an individual layer by means of a pin-and-bearing connection, so the pattern changes as the pin moves.
Because the pattern’s opacity continually evolves and works to block daylight and heat gain at a variety of levels, the system is suitable as part of an environmental-control system. According to ABI, its opacity can range from 10 percent to 85 percent.
At roughly 52 inches wide and 11 feet tall each, the panels fill four bays of the façade, reaching from the floor (where they sit on a stainless steel riser) to the first structural member, where they bolt into the beam. By the time the Simons Center opens, four more, slightly shorter panels will span from the beam to the ceiling, for a total of about 380 square feet of shading. As Davis puts the system panels through their paces via a laptop computer, he programs them to run at a speed several times faster than a typical building installation, looking for glitches, catches, or bending as the layers shift and combine. When tied into a building-management system (BMS), the shading system’s motor will run slowly; its speed can be set to track the sun or respond to environmental factors such as energy use, solar gain, and daylight levels. Patterns can take from several minutes to several hours to switch position. If the panels are used sculpturally, however, the motors can cycle through the various geometries in under a minute.
The Stony Brook installation is a kind of demonstration of sorts for the Hoberman team. It’s a chance to test and fine-tune the ABI kinetic shading system, but it’s also a way to galvanize the company’s entry into the architectural market. Chuck Hoberman, founder and president of Hoberman Associates, is best known as an inventor and artist. His kinetic artworks are internationally shown, but his team also collaborates with marquee architecture firms such as Foster + Partners, Kohn Pedersen Fox Associates, and Shop Architects. The system, a self-contained, framed screen, isn’t an off-the-shelf component with a single-use profile. It’s designed to be totally customizable. Architects interested in it would collaborate with the team on the design, the manufacturing, and the installation. ABI touts the system as both decorative and performative, suitable for façade installations, glazing systems, and room partitions.
Like the obvious precedent, Jean Nouvel’s 1987 Arab World Institute, in Paris, the panels mechanically shift to open and close. But unlike Nouvel’s motor-controlled brise soleil, which relied on photosensitive controls to mediate the apertures, these dynamic panels can be directly integrated with a BMS to track changing light and weather conditions for energy savings. Because the panels are custom designed for each project, different panel configurations and patterns lend themselves to applications for controlling airflow, solar gain, and privacy. While shading is critical in almost any building looking to reach LEED certification or meet ASHRAE Standard 90.1 requirements, conventional shading systems based on blinds or louvers offer limited creative solutions. By contrast, this system’s patterns can be designed as a collaboration between the architect and ABI. Architects “are looking for something that is not just about the performance aspects, but they want something that gives them flexibility and allows them to put their own stamp on the shading,” notes Craig Holland, director of operations at Hoberman.
By teaming with Kansas City, Mo.based Zahner, ABI steps immediately into an arena where top designers can apply their imaginations to the system, pushing it to its conceptual limits. The manufacturer routinely works with some of the best-known international firms to create signature skins and wall surfaces. The dynamic shades are fabricated as thin-profile framed panels—at Stony Brook, they’re 4 inches deep—that can stand on their own or, in larger applications, be paired with a curtainwall assembly. At Stony Brook, ABI used five layers of metal that are a combined 1-1/8 inches in thickness. Adding and subtracting layers changes the range of pattern configurations as well as potential screen density. (Although current iterations of the system’s panels are not yet fully functioning exterior façade systems, the ABI team hopes to develop weathertight prototypes in the future.)
According to Zahner’s president and CEO, Bill Zahner, Hon. AIA, the system is straightforward and poses few technical or operational obstacles. “It is amazing how simple and easy they [the panels] are to fabricate,” he explains. Hoberman’s two decades’ worth of experience engineering moving parts is security against the units breaking down. If on the odd chance they do pose a mechanical problem, each panel has an individual access panel to reach the motor and driver—nonproprietary parts that can be easily replaced. Additionally, the glass panels can be easily removed to fix or swap out malfunctioning pieces, which makes the system’s maintenance no more complicated than more-standard shading.
For Zahner, material choices pose the biggest challenges because they directly affect cost. Making the layers thin and lightweight keeps material expenses down. The Stony Brook installation uses 14-gauge steel for the outer, static layers and 16-gauge steel for the inner, moving layers. (Gauge thickness is determined on a case-by-case basis, Holland says, and is dependent on unit size, perforation pattern, number of moving layers, and other factors.) A water jet cut the pattern, giving a less precise edge than a laser but leaving a more textured surface to reflect the light better. (A laser-cut edge is often mirrored, while a water-jet-cut surface is more pearly and thus less prone to creating glare spots.) Because each installation would be a collaboration between the commissioning architecture firm, ABI, and Zahner, not an off-the-shelf specification, it’s tricky to estimate costs. “Presently, I would put it in the $150-per-square-foot range, which is on the pricey side,” Zahner admits. “But it is very difficult to compare to something else, because there is nothing like it. If we are able to incorporate it into a façade or HVAC system, the additional cost of putting in a decorative screen is balanced by its energy-savings performance.”
Matthew Herman, head of Buro Happold’s Chicago office and an ABI board member, led the system’s performance testing. His goal was to determine the energy benefit and to compare the impact of an adaptive surface to a standard fixed-shading system. Using southeast-facing windows on the 16th floor of a New York City building as the test location, team members created two identical test rooms and placed a mock-up panel (featuring a square design) in the window of one; the window of the other room remained uncovered. No lights or equipment in either room were turned on, and the HVAC system was shut off as well. On two different days—one sunny, one cloudy—daylight readings were taken for 13 configurations of the screen’s layers at 11 a.m. Measurements were made 3 feet off the floor (an approximate desk-level height) at zero feet from the panel, 3 feet from the panel, and 6 feet from the panel (just past the midpoint of the 10-foot-deep room). Each pattern configuration represented a different level of opacity, from 15 percent to 85 percent, and the base acceptable light level was 30 to 50 foot-candles, considered optimal for office task lighting.
For example: A 15 percent opaque configuration reading showed 42 percent daylight reduction at zero feet, 58 percent reduction at 3 feet, and 85 percent at 6 feet, for an average daylight reduction of 62 percent. Foot-candle readings for the same configuration were 136.7, 100, and 34.8, respectively, compared with 236, 160, and 65 for the room with no shading. At 40 percent opacity, 93.9, 78, and 27.1 foot-candles translated to 60 percent, 67 percent, and 89 percent daylight reduction, for an overall average of 72 percent. (The diffuse nature of a cloudy day, Herman says, did not ultimately make for much of a comparison, as light levels in the room suggested that the system would be left opened as widely as possible.)
Temperature data were taken at 10-minute intervals throughout the test day (July 21), with the mock-up panel set at 40 percent opacity; these data points then served as the baseline for a dynamic thermal model, which measures energy reduction. As opposed to taking continual readings from a mock-up, a thermal model can simulate a full year in the life of a typical office building and apply to multiple geographic locations. Herman used the energy-modeling software IES Virtual Environment 6.1, which is compatible with a wide range of CAD programs, for the digital model. The adaptive-system model was set to minimize energy consumption and allowed more solar heat gain, measured in British thermal units, in the winter and blocked out eastern and afternoon sunlight in the summer.
Running the model through an 8,760-hour year, Herman tested the product’s range, comparing the system’s adaptive shading at each of the 13 fixed configurations over the course of a year against fixed shading—defined as external horizontal louvers with a 2:1 glass-to-projection ration—as well as no shading, the lowest condition allowed by ASHRAE 90.1 and represented by a glass coating with a solar-heat-gain coefficient of 0.39. All three models were identical: 10-story buildings with a window-to-wall ration of 60 percent; weather data came from John F. Kennedy International Airport (Typical Meteorological Year 3 data format).
The results were surprising. According to Herman, typical fixed-shading and high-performance façades reduce building energy consumption by 3 percent. The ABI’s kinetic system showed a 3 percent to 5 percent reduction at any one moment, but when it was added up over the course of a year for the New York City climate test building, the model produced a 6 percent energy reduction. Herman attributes the higher percentage to the mechanism’s ability to quickly adapt for better environmental performance. The benefits were most prevalent in the spring, summer, and fall, when less solar gain reduces the cooling load. The adaptive-façade model also showed an average internal temperature drop of 1 degree Fahrenheit—which may not seem very large, but the average internal thermal comfort range of a building can vary by 10 degrees (from 68 F to 78 F, per ASHRAE Standard 55).
In an actual building, a BMS would monitor temperature, energy usage, and daylight and, based on the readings, adjust the system’s panels so that they could immediately reduce solar gain or balance light levels. The kinetic shading’s control system is open source and can accept inputs from just about any digital source, and each panel has its own motor and controller. Open-source software means easy compatibility with BMS software, and nonproprietary mechanical parts keep construction and repair costs down. The panels link into a single CPU that can run off of a standard PC or Mac running Linux or Windows. The BMS tells the central computer how to array the shades in a language of digital signals or voltages from meter readings. Individual panels, or several in tandem, can also be set to create a specific, localized effect, making the sum of them integral to potentially all scales of a building’s performance.
Not ready to rest on the laurels of ABI’s findings, Herman sees the kinetic shading system’s ability to reduce the amount of energy expended on heating, cooling, and lighting as part of a full-building sustainability strategy. “For us, the starting point is solar shading, not the end. It is just one of the energy flows and mass flows through a building that impacts its carbon footprint,” he notes. “We have the technology and are comfortable with making these screens move, so what else … would benefit from a moment-by-moment dynamic adaptation? Other systems, such as water and acoustics, could be improved with changes in geometry, material, and mechanisms.”