Solar module degradation and warranties, decoded

Every module datasheet carries two sets of promises: a product warranty covering material and workmanship defects, typically for 10 to 25 years, and a performance (or "power output") warranty guaranteeing a minimum percentage of rated power at specific points across 25 to 30 years. Underneath both promises sits a set of physical degradation mechanisms that reduce a module's output over time regardless of manufacturing quality, plus a set of IEC test sequences designed to catch modules that would degrade faster than expected. Understanding the mechanisms makes the warranty documents legible, and understanding what the qualification tests do and do not prove is essential to judging whether a "25-year warranty" is a meaningful guarantee or a marketing document with little practical recourse.

This article works through the three degradation mechanisms most relevant to crystalline silicon modules — light-induced degradation (LID), light- and elevated-temperature-induced degradation (LeTID), and potential-induced degradation (PID) — then explains how warranty structures are typically written and what the underlying IEC 61215/61730 qualification sequence actually demonstrates.

Light-induced degradation (LID)

LID is a rapid, largely one-time degradation that occurs in the first hours to weeks of a module's exposure to light, caused primarily by boron-oxygen defect complexes in boron-doped p-type silicon. When illuminated, these complexes shift into a configuration that increases recombination of charge carriers, reducing cell efficiency. Typical PERC modules lose roughly 1.5–3% of their rated power to LID within the first weeks of operation before stabilising; this is why many manufacturers state a first-year degradation allowance in their performance warranty noticeably larger than subsequent annual steps — commonly around 2% in year one against roughly 0.4–0.6% per year afterward.

Gallium-doped p-type wafers and n-type wafers (used in TOPCon and HJT, see TOPCon vs HJT vs PERC) are substantially more resistant to boron-oxygen LID because they either lack the boron-oxygen complex mechanism or use a different majority dopant, which is one of the reasons n-type architectures publish flatter, more favourable degradation curves.

Light- and elevated-temperature-induced degradation (LeTID)

LeTID is a slower-developing, and more concerning, degradation mode discovered more recently than LID. It is driven by hydrogen-related defects that migrate and activate under combined light exposure and elevated temperature (typically 50–75°C) over weeks to months, rather than the hours-to-days timescale of LID. Unlike LID, LeTID can, in susceptible cell designs, continue degrading output by several percent over the first one to two years before partially or fully recovering as the defect population re-anneals, though the degree of recovery is manufacturing-process dependent and not guaranteed.

LeTID susceptibility is influenced by the hydrogenation processes used during cell passivation (particularly in PERC manufacturing) and has prompted manufacturers to adjust firing and passivation recipes specifically to suppress it. Because LeTID manifests over a timescale beyond typical factory qualification testing, it has historically been under-represented in standard IEC 61215 sequences, and industry test houses have since developed supplementary extended LeTID test protocols to close that gap.

Potential-induced degradation (PID)

PID is caused by leakage currents driven by the voltage difference between cells and the grounded module frame, particularly in systems using negatively grounded or high-voltage ungrounded inverter topologies where a module can sit at several hundred volts relative to the frame. Under humid conditions, this voltage stress drives sodium ion migration from the glass through the encapsulant to the cell surface, increasing shunting and reducing shunt resistance, which can cause severe and sometimes irreversible power loss — in extreme, poorly mitigated cases, 20% or more within a single season.

PID risk is mitigated at three levels: cell and encapsulant design (anti-PID cell coatings, PID-resistant EVA/POE encapsulants), system design (inverters with PID-resistant grounding schemes, or a PID recovery box that reverses the bias overnight), and installation practice (ensuring the array's electrical midpoint voltage relative to ground stays within the levels the module was qualified for). PID is tested via IEC TS 62804, described below, and a module's PID test result should be checked against the specific system voltage and grounding scheme planned for a project rather than assumed safe by default.

Comparison of degradation mechanisms

Reading a warranty document properly

Product (materials and workmanship) warranty

Covers manufacturing defects such as delamination, junction box failure, frame corrosion or glass breakage not caused by external events, for a stated period, typically 10–15 years for older PERC lines and increasingly 25 years for premium TOPCon and HJT products as manufacturers align product and performance warranty lengths. This warranty is only as good as the manufacturer's continued existence and the terms of any local distributor or insurance-backed warranty scheme, which matters more than the number on the page for a homeowner assessing real recourse risk 15 years into ownership.

Performance (power output) warranty

Guarantees a minimum percentage of rated power at specific milestones, typically expressed as a linear or two-stage degradation curve: an initial step (covering LID) followed by an annual linear decline. A common current-generation TOPCon warranty might read: 98% of rated power guaranteed at year 1, declining linearly at 0.4% per year, guaranteeing 87.4% of rated power at year 30.

Worked example

Take a module rated at 440 Wp with a warranty specifying 98% of rated power at year 1 and linear decline of 0.45% per year thereafter, and ask what power the warranty guarantees at year 20.

Year 1 guaranteed power: 440 × 0.98 = 431.2 W. Years 2 through 20 is 19 further years of decline at 0.45% of the original rated power per year: 440 × 0.0045 × 19 = 37.62 W. Guaranteed power at year 20: 431.2 − 37.62 = 393.58 W.

As a percentage of original rated power: 393.58 ÷ 440 = 89.4%. If a monitored system's measured output at year 20, after correcting for irradiance and temperature, falls meaningfully below 89.4% of rated power, the module would be within its rights to a warranty claim, subject to the manufacturer's measurement and claims procedure — a process that is often more administratively demanding than the number itself suggests, and worth understanding before a claim is needed.

What IEC qualification testing actually proves

It is worth being precise about what passing IEC 61215 and IEC 61730 demonstrates, because it is commonly overstated. These are design qualification and safety tests performed on a small sample of modules from a specific design, not a guarantee about every unit shipped, and not a direct simulation of 25 years of field exposure.

Passing these tests is a necessary baseline for any credible module, but the performance warranty is a separate commercial commitment from the manufacturer, not something IEC certification alone guarantees. Buyers comparing modules for a project should check both the qualification test reports and the specific warranty document, and should factor realistic degradation assumptions, rather than headline warranty numbers, into any yield or payback calculation, particularly for older PERC stock where LID and LeTID risk is higher than for current n-type products.

Practical implications for system design

Degradation and warranty terms interact with system design choices covered elsewhere on this site. Grounding scheme and inverter topology affect PID risk and should be checked against the module's PID test rating during inverter sizing. Battery-coupled systems, discussed in sizing a home battery without guessing, should use degraded, not rated, module output when modelling year-15 or year-20 self-consumption, since a 20-year-old array producing 90% of its original output changes the sizing arithmetic for a battery meant to serve the household for a similar horizon.

Standards and references

• IEC 61215 — Design qualification and type approval for terrestrial crystalline silicon and thin-film PV modules, covering thermal cycling, damp heat, mechanical load and hot-spot tests.
• IEC 61730 — PV module safety qualification, covering electrical shock, fire and mechanical hazard protection.
• IEC TS 62804 — Test methods for the detection of potential-induced degradation in crystalline silicon photovoltaic modules.
• IEC 63209 series — Extended stress testing methodology for comparative long-term reliability assessment of PV modules.
• IEC 62548 — PV array design requirements, relevant to system-level PID mitigation through grounding and voltage design.

Frequently asked questions

Is LID something to worry about when buying a module today?

For current n-type modules (TOPCon, HJT) LID is a minor consideration, since n-type wafers largely avoid the boron-oxygen mechanism that causes it. For p-type PERC modules, still widely available, LID remains a real first-year effect but a well-characterised and stabilising one, already reflected in the manufacturer's stated first-year degradation step. It is not a defect to dispute, simply a known physical effect factored into the rated numbers.

Can PID be fixed once it has occurred?

In many, though not all, cases PID is at least partially reversible if caught early: reversing the electrical bias on the affected modules, sometimes with a dedicated PID recovery device connected overnight, can recover a meaningful fraction of lost performance as the migrated ions redistribute. Severe or prolonged PID exposure can cause permanent shunting damage that does not fully recover, which is why system-level PID prevention through correct grounding and inverter selection is preferable to relying on recovery.

Does a longer warranty period always mean a better module?

A longer warranty period is a reasonable signal of manufacturer confidence, but it is only as valuable as the manufacturer's financial stability and the practicality of the claims process over that period. A 25-year performance warranty from a well-established manufacturer with a track record and clear claims documentation is generally more valuable than a 30-year warranty from a newer entrant with no comparable history, regardless of the number printed on the datasheet.

What is the practical difference between LID and LeTID for a homeowner?

LID happens quickly, in the first days to weeks, and then stabilises, so it shows up as a one-time step down in output that is already accounted for in the module's rated and warranted figures. LeTID develops more slowly over months to roughly two years and can continue reducing output during that window, sometimes followed by partial recovery, making it harder to distinguish from normal early-life variation without careful monitoring, which is why choosing a module with a documented low-LeTID manufacturing process matters more than it might first appear.

How much annual degradation is normal for a modern module?

Most current-generation modules, whether PERC, TOPCon or HJT, specify a first-year degradation step of roughly 1–2% followed by linear annual degradation of about 0.35–0.6% per year, with n-type products generally clustering toward the lower end of that range. A system consistently degrading faster than its warranted curve, after ruling out soiling, shading and measurement error, is a legitimate basis for a warranty claim.

Sources

• NREL — PV Reliability and Degradation Rates Research
• Fraunhofer ISE — Photovoltaics Report
• PVEducation — Degradation of Solar Cells
• IEC — TC 82 Solar photovoltaic energy systems