Washington, DC, is known for its row houses—two- or three-story structures that are narrow and deep. Roof space is sparse, and fitting enough PV modules on a rooftop to zero out a household’s annual electricity usage can be challenging.
If modules are mounted at a tilt optimized for annual production—which is usually somewhat close to the degree of latitude—the rows of modules on flat roofs have to be widely spaced to avoid shading each other.
Reducing the tilt somewhat decreases output, but allows more modules in a given space. In this case, deviating from what is generally considered the optimal tilt can provide more net production since more modules can be placed on the roof—as long as interrow shading is avoided (see “Methods: Interrow Shading” in HP151).
Vendor D offered three options (see illustrations at left): tilting the PV modules on I-beams that would lay across the “party” walls to avoid any weight on the roof; a tilted, ballasted array; and an array that was mounted flat on the roof, resulting in an estimated 4,477, 5,086, and 6,983 kWh per year production, respectively. Each option was progressively more expensive initially, but progressively more profitable in terms of net present value and internal rate of return. If you can afford to make the larger investment, the payoff will also be larger.