Determining wind loads on roof-mounted solar arrays for calculating ballast weight required to secure the system in place.
Since 2009 the interest in renewable energy in the form of arrays of photovoltaic (PV) solar collectors has grown rapidly. This was bolstered by government incentive programs in various provinces and states and the retirement of coal fired generation in the province of Ontario. Technology originally imported from Germany provided the basis for many players in the early days. The North American industry grew rapidly and it was recognized early on that the approval process required home-grown science behind the designs. hb Solar was one of the early players in the industry, and continues to provide well-designed systems for installations across North America.
Unused rooftops of industrial and commercial buildings represented a major source of real estate for the application of PV solar. However, wind climates of different regions, aerodynamics of the building and the specifics of the aerodynamics of the racking system all play a major role in the weight of ballast required to hold the solar array in place in a 50-year design wind storm. Since market economics did not justify site-specific wind tunnel testing for most installations, wind tunnel tests of generic configurations were needed to provide the design loading information that building code provisions were unable to suffice.
Using our well proven wind tunnel technique of measuring surface pressures on high-rise buildings, scale models of both a generic flat roof building, typical of big-box store commercial buildings and industrial buildings, and the array itself were installed in the boundary layer wind tunnel. Pressure taps on both the top and bottom surfaces of many rows of scale model PV modules were measured simultaneously at hundreds of locations. This was made possible through our familiar methods of routing internal pressure passages through solid models produced by stereolithography. Integrating rapidly fluctuating pressures real-time in the wind tunnel tests for each of 36 effective wind directions provided detailed information on the peak wind loads acting to lift and drag the solar array from the roof.
Compiling vast amounts of data from the wind tunnel into coefficients formulated to replace those provided for common building forms in building codes, tables representing the envelope of worst case values for various portions of the array were produced. This form of the data allowed design engineers to work within a familiar wind load calculation framework but with input specific to the racking geometry produced by the manufacturer and position of the module within the array.
The testing underlined the fact that for flat roof mounted solar arrays, the turbulent wind flow structures created by the building itself were the most important to the peak loading on the solar array. Most importantly vortices, or spinning flow, generated by the separation of the wind as it flows over a leading building corner for wind directions at an angle to the walls created the largest peak loads in general. Additionally, the position within the array of a specific solar module was critical to the interaction with these vortices.
The end result was a table of values to be used by designers to apply building code-like calculation procedures to determine wind loads. This enabled incorporating common site specific parameters such as local 50-year design wind speed, exposure of the site created by surrounding terrain and structures, and importance of the building. Through varying other features of the specific racking system and layout in the wind tunnel, additional tables providing guidance for variations such as parapet height, tilt angle, row spacing or array skew to the building edges were made available.