Photovoltaic panels are typically attached to the roof of a building with a fixed mount. The panels are optimized to face south (in the northern hemisphere) and tilted at an angle that optimizes production. Unfortunately, a fixed mount only optimizes the array for one time of the day one day of the year. PV panels work best when the sunlight hits the panel exactly perpendicular to the panel itself. As you can see in this picture, this panel is optimized to deliver maximum power on the spring and fall equinoxes, while performing sub-optimally during the summer and winter months.
Solar farms use tracking motors or a robotic method of adjusting the panels so they always face the sun directly. Dual-axis tracking can improve solar production by 30% to 80%, depending on local conditions. This is fine for utility-scale solar farms, but for individual buildings and small scale production the motors add a lot to the cost and complexity of the system, while not increasing the production enough to justify those costs. In those cases it’s often more cost-effective to simply purchase more PV panels, especially since their price is decreasing. Of course, that requires a larger footprint, and land isn’t cheap.
Researchers at MIT are wondering why PV designers insist on thinking in only two dimensions. They’ve come up with a novel approach: 3D structures.
Image: Allegra Boverman
Sunlight can reach panels directly, as shown in the top photo, but light also reflects off of various surfaces on the ground. With panels facing the sun, the reflected light reaching the PV panels is negligible. Using the 3D arrangement, the reflected light reaches the panels that are angled downwards. The downward-facing panels also grab sunlight when the sun is lower in the sky, such as early morning, late afternoon, and in the winter.
MIT researchers first developed a computer model and ran simulations to determine the optimum configuration of panels for consistent energy production throughout the day and the year. Then they built several models and attached them to the roof of an MIT lab building. Field testing showed that the 3D array generated 200% to 2000% more than a fixed array with the same footprint. The 3D arrays require about twice as many panels (maybe 4 times, depending on the configuration), but they get much more bang for the buck in terms of production.