There is no universal interval
Owners often ask for the correct number of months between thermal inspections. The honest answer is that no single interval is right for every array. Guidance for field infrared inspection exists, for example in IEC TS 62446-3, but it concerns how to carry out an inspection, not a mandatory frequency for every site. The schedule is a judgment about what a missed fault costs and how quickly faults tend to appear.
Start with a baseline
The most valuable scan is the first one. A baseline thermal inspection after construction or at handover shows what the array looked like when it was new, and it catches installation problems while the installer is still accountable. Later scans are compared against it, which makes change visible. Without a baseline, every later finding has to be judged without knowing whether it is new.
Factors that push the interval shorter
- Age. Equipment ages, connections loosen and faults become more likely over time, so older arrays are reasonable candidates for more frequent checks.
- Harsh environment. Heat, dust, hail exposure, salt air or heavy thermal cycling stress modules and connections.
- Past findings. A site that has already shown hot cells or diode faults deserves a closer follow-up on those areas.
- Value of output. Where lost production is costly or contractually sensitive, faults should be found sooner.
- Warranty and performance terms. Claims often depend on documented condition within a stated period.
- Low monitoring detail. If monitoring only reports at a coarse level, faults can hide inside the averages.
Factors that allow a longer interval
- A young array with a clean baseline.
- Detailed monitoring that flags string or module-level changes.
- A mild environment and few past issues.
Even then, a periodic check is useful, because monitoring shows a loss of output but not always where the loss sits.
Event-driven scans
Some triggers should prompt an inspection regardless of the schedule: a storm or hail event, a fire or electrical event nearby, a noticeable drop in production, a change to the array such as repowering or expansion, and a pending sale or financing. These scans are tied to a question and produce a more focused result.
What each scan should deliver
For a scan to be worth repeating, it must be comparable. That means the same method each time, a record of the conditions at the time of the flight, findings tied to module locations on the as-built layout, and a thermal image paired with a visible image for each finding. A short summary should rank findings by their likely effect so that the owner knows what to repair first.
Thermal is one tool of several
Infrared finds heat-producing faults. It does not find everything. Cell cracks that do not heat up, surface soiling and physical damage need other methods. The note on thermal scan versus EL test compares what each catches, and the visual inspection checklist covers what imagery of the module surface should record. A sensible program combines methods at different intervals: infrared on the regular schedule, and the others when a question calls for them.
For help choosing the flight timing and conditions, see when to fly a drone thermal scan.
Setting your schedule
A practical way to decide is to list the array's age, environment, monitoring detail and the cost of lost output, then pick an interval you can defend, write it down, and revisit it after each scan. If the scans keep finding little, lengthen the interval. If they keep finding new issues, shorten it.
Keeping the program simple
A thermal program does not need to be elaborate. Write down the baseline date, the interval you have chosen and the reasons for it, the events that trigger an extra scan, and who receives the report. Keep every report in one place so the history is available when a warranty question, a sale or a lender review arrives. A simple program that runs on schedule is worth more than an ambitious one that lapses after the first year.
Where Solar Survey AI fits
Our thermal inspection service uses infrared thermography, including drone-based aerial capture across every module, with findings located against the array so each scan can be compared with the last. Tell us about the array and what you need protected, and we will propose a schedule that suits the site.