Designing PV Systems for Environmental Extremes

Have you ever wondered how integrators and engineers adapt system designs to withstand high snow loads, extreme wind speeds or corrosive environments?
After hearing from readers who asked just that, we reached out to industry experts with real-world experience.

While every PV system must withstand tough environmental conditions—from high humidity to elevated rooftop temperatures—some systems are tested to the extreme. For instance, design wind speeds in Florida generally range between 110 and 185 miles per hour (even higher in some cases), depending on the location and building risk category. The design wind speeds in much of the rest of the country are tame by comparison.

Except when they are not: Even if the majority of your projects have a nominal design wind speed of 90 mph, which is typical for much of the continental United States, a number of scenarios might require you to design to higher wind speeds. These include buildings in a high-risk category, such as schools, hospitals, fire stations and government facilities. Design wind speeds also increase based on building height or roof zone. You might even win a once-in-a-blue-moon contract to install PV in the Bahamas, in which case it would help to know how veteran integrators design for the dual threat of hurricane wind speeds and coastal corrosion.

For this article, I reached out to quality-minded solar professionals who routinely design and deploy systems subject to extreme environments, as well as independent engineers who have studied the effects of extreme environmental exposure on system components. I was specifically interested in those environmental conditions that exert structural or mechanical stresses on PV systems, such as snow, wind and corrosion. Further, I was less interested in the engineering analysis required to account for these forces and more interested in their empirical effects. 

There is no better test of our products and installation practices than the real world, which imposes itself in ways both expected and unexpected. Though there are engineering calculations for describing snow loads on a PV system or structure, they will not tell you what might happen when that snow slides off PV modules—or when well-intentioned owners use the wrong tool to clean snow off their array. In effect, this article picks up where engineering calculations leave off, and its lessons are all the more valuable because they point out potential mistakes as well as unexpected consequences. Best of all, our contributors share their proven recipes for success when deploying PV systems in extreme environments.

Snow Loads

What type of damage due to excessive snow loads have you experienced with your fielded PV systems? How do you think you could have avoided these problems?

We have not had any PV system damages or failures in western Canada as a result of snow loading. However, we have seen damage caused by customers trying to brush and knock snow off an array. One customer actually managed to put the corner of a push broom through a module while trying to knock off the ice.

—David Kelly, chief executive officer, SkyFire Energy

From my experience, bent frames due to snow and ice are the most common cause of PV module failures. This is especially true when there is thawing and refreezing. However, snow alone can cause module breakage. The probability of this type of damage is highly dependent on local conditions.

—Sarah Kurtz, PV module reliability test and evaluation group manager, National Renewable Energy Laboratory (NREL)

The mechanical failure of racking hardware and fasteners is the most common issue I have encountered associated with heavy snow loading. The weight of the snow and the perpetual winter freeze-thaw cycle puts tremendous downward and lateral forces on flush-mounted rooftop arrays. In Colorado, for example, I came upon an array with so much snow backed up behind it that a number of mid-clamp T-bolts had pulled right out of the top channel of the rail, which left modules dislodged.

In Vermont, I’ve seen L-foot fasteners pried 2 inches out of their associated rafters or trusses and bent down at 90°; this caused noticeable drooping on a rail that was unintentionally cantilevered about 6 feet beyond the last fully secured L-foot. I have also seen ice buildup that was almost 3 inches thick under the top portions of arrays; where this ice fully encases module-to-module interconnections, it can lead to electrical shorts.


Article Discussion