Lightning Protection for a Solar Array

Lightning protection for a solar array is a design done before the modules are up, not a rod added after the first storm. IEC TR 63227, in the preview text that sets its scope, covers protection of photovoltaic power systems against lightning and against surge voltages of atmospheric origin. It says that putting a conventional PV system on or at a building does not, by itself, change the lightning-strike risk. It recommends designing the PV system and the lightning protection system together before they are built. A surge protective device, in that document’s definitions, contains at least one non-linear component and is intended to limit surge voltages and divert surge currents. That is a different object from a lightning rod, and it is a different object from the equipment grounding conductor that clears a fault.

What lightning protection has to include

Three ideas get collapsed in sales language. A lightning protection system is the air terminals, down conductors, and earth electrodes that give a strike a path, when a risk assessment says that system is warranted. Bonding and grounding are the fault-current and touch-safety paths the electrical code already requires for the array. A surge protective device limits what a transient does to the inverter, the combiner, and the communications lines. You can have grounding and still destroy an inverter with a surge. You can have a surge device and still have a bad equipment ground. IEC TR 63227 points at the lightning-protection series and at the surge-device series, including the part written for photovoltaic installations, as the documents that sit behind those choices. This article names them. It does not extract a shopping list of kiloamperes.

The array is a large conductor spread over a roof or a field, and it is energized by daylight even when the sky is dark enough that you would rather be indoors. The technical report states the operational consequence: the PV generator delivers current and voltage even at low irradiance. That is why a storm is the wrong time to “check the connections.” Modules are not off because the inverter has faulted. Pacific Northwest National Laboratory’s shutdown order, AC disconnect then DC disconnect, is for trained people, in conditions where they can do it safely. It is not a task to start when lightning is in the area.

NREL’s field guide on ground-fault testing adds a practical warning that belongs here. Insulation tests and other high-voltage checks can damage surge devices or be confused by them if the devices are left in circuit. Testing is planned with the devices in mind. It is not a reason for an owner to pull cartridges. The guide is about fire-hazard investigation by people equipped for it. A visual check that stops at a closed cover is the owner’s limit, and it is described in what to check on a solar system.

Where surge protection sits, without a recipe

Surges show up on the DC conductors from the array, on the AC conductors to the service, and on the thin cables used for monitoring and weather sensors. Protecting only the AC side leaves the DC side and the data side as the path into the inverter. Protecting a data line with a power device, or the reverse, is a mismatch of category. The technical report’s definitions separate power-system surge devices from devices for telecommunications and signalling. The person who specifies them has to match the voltage of that circuit, including the maximum PV voltage on the DC side, which is higher on a cold morning than the nameplate at standard test conditions. Voltage windows are the same discipline as for charge controllers, applied here to a surge device that must survive the normal operating voltage and still clamp the transient. Get that match from the listing, not from a universal ampere rating.

Distance matters in the sense that a long DC run is a different exposure from a microinverter under the module, but the spacing rule belongs in the standard and the manufacturer’s coordination study, not in a paragraph that someone could mistake for an installation drawing. If the inverter already contains a surge device, confirm what it protects and what it does not. An internal device is not automatically a device at the array, and a device at the array is not automatically a device on the monitoring cable. Severe weather is when people discover the gap. The storm behavior of the plant, including why the inverter may go offline correctly, is in severe weather and solar panels.

A grid-tied inverter that stops during a strike nearby may be obeying IEEE 1547-2018, which requires it to cease energizing an unintentional island within 2 seconds. That stop is not proof the surge devices worked, and it is not proof they failed. Look at the indicators and the fault log after the storm, with the array left alone until then.

What lightning protection does not do

It does not make the roof a safe place in a thunderstorm. It does not raise energy yield. It does not replace a performance warranty, and a warranty may exclude lightning or limit the remedy to hardware while you pay for the crane. Read the exclusion. The contract habits are in what a solar panel warranty covers. It does not change the fact that soiling, shade, and snow are the ordinary losses. Those are operations. Lightning is a rare, high-consequence event that is handled by design and by staying away, not by a maintenance calendar.

If a vendor offers a kit with a single kiloampere number for every roof, ask which standard’s risk assessment was done for this building, which circuits are protected, and what the maximum continuous voltage of each device is relative to this array. If those answers are missing, the kit is a box, not a design. Adding modules did not create a new strike risk by itself, in the words of the technical report. Leaving the surge path undesigned is still how inverters fail.

Frequently asked questions

Does a solar array attract lightning?

IEC TR 63227 says erecting a conventional PV system on or at a building does not change the lightning-strike risk. The array is still a set of conductors that can carry a surge if a strike or a nearby discharge couples into it. Design the protection with the array. Do not assume the building’s old rod, if it has one, covers the new DC wiring.

Is a surge protective device the same as a ground rod?

No. A ground electrode is part of the earthing and bonding system. A surge protective device limits transient voltage and diverts surge current, and it has to be listed for the circuit it is on. One does not substitute for the other.

Can the owner add lightning protection after a close call?

The right time was before construction, which is the report’s recommendation. An after-the-fact design is still a design, done by someone qualified, with the array isolated by people who can isolate it. It is not an afternoon of extra clamps on a live string.

Sources and references

Scope, the unchanged strike risk, joint design before erection, the definition of a surge protective device, and production at low irradiance: IEC TR 63227:2020, preview. Why surge devices complicate unplanned electrical tests: NREL field guide on photovoltaic ground faults. Live modules and shutdown order: Pacific Northwest National Laboratory. Islanding limit: NREL and Sandia IEEE 1547-2018 primer.