PV I-V curve tracer — string diagnostics per IEC 62446-1

The [guide on insulation resistance testing on the DC side of PV strings](/guides/nen-3140/isolatieweerstandsmeting-pv-strings-dc-zijde) covers one of the mandatory measurements during the initial verification and periodic inspection of a PV installation. This article covers an additional diagnostic test from the same standard, IEC 62446-1, which is not mandatory by default but provides the most powerful tool for understanding an underperforming string: the I-V curve tracer.

Category 1 versus Category 2: mandatory versus additional

IEC 62446-1 distinguishes two test categories:

Category 1 (Cat 1) covers the tests that are mandatory for virtually every PV installation: continuity of protective conductors, insulation resistance measurement, polarity check, and a measurement of open-circuit voltage (Voc) and short-circuit current (Isc) per string.
Category 2 (Cat 2) extends this with more in-depth diagnostics, including I-V curve tracing and performance benchmarking — intended for situations where the Cat 1 measurements have already been carried out, but a string still shows lower-than-expected output and the cause is not immediately clear.

What an I-V curve tracer measures

An I-V curve tracer sweeps across the full operating range of a PV string — from short-circuit current (Isc, at zero voltage) to open-circuit voltage (Voc, at zero current) — recording hundreds of current-voltage pairs along the way. From this complete curve, the tracer derives Isc, Voc, the current and voltage at the maximum power point (Impp, Vmpp), the delivered power Pmax, and the fill factor (FF): the ratio between Pmax and the product of Isc and Voc, which indicates how "square" the curve is relative to the theoretical ideal.

Why the shape of the curve reveals the cause

The diagnostic power of an I-V curve tracer lies not only in the final values (Isc, Voc, Pmax), but especially in the shape of the deviation from a healthy reference curve:

Uniform shading across the entire string reduces Isc proportionally but largely leaves the shape of the curve otherwise intact.
Cell mismatch or partial shading (for example one shaded module in an otherwise unshaded string) produces a characteristic step or kink in the curve, as one or more bypass diodes of the shaded module start conducting.
Increased series resistance (for example from a corroding or loose MC4 connector, or a deteriorated solder joint) shows up as a reduced slope of the curve near Voc — the point where the curve normally bends nearly vertical becomes "softer".
A failed bypass diode produces a specific, recognizable step or dip at a fixed location in the curve, independent of the actual shading situation.
PID (potential-induced degradation) typically reduces both the fill factor and the power in a way that differs from ordinary cell ageing, making it recognizable as a distinct cause from an I-V curve.

In this way, the shape of the curve distinguishes causes that a single Isc/Voc measurement (Cat 1) cannot tell apart.

Correction to standard test conditions

A measured I-V curve is directly dependent on the actual irradiance and module temperature at the time of measurement, which are almost never equal to the standard test conditions (STC: 1000 W/m², 25 °C cell temperature) on which the panels' nameplate specifications are based. To meaningfully compare a measured curve with the expected, manufacturer-specified curve, the measurement must therefore be carried out simultaneously with an irradiance measurement and a module temperature measurement, and subsequently corrected to STC (see IEC 60891 for the correction procedure). An I-V curve tracer that does not apply this correction produces a curve that is only valid for the actual measurement conditions and cannot be directly compared with the manufacturer's datasheet.

Note: correctly interpreting an abnormal curve shape requires experience and, when in doubt, a comparison with the curve of a healthy reference string under the same conditions; this article covers the principle, not an exhaustive catalogue of every possible curve shape.

Practical relevance

For a string showing lower energy output than the other, similarly-oriented strings of the same installation, an I-V curve measurement is generally more efficient than visually inspecting every connector and every panel one by one: the shape of the deviation directly indicates the most likely cause (shading/mismatch, series resistance, or a failed bypass diode), allowing further investigation to be carried out in a more targeted way.

Common mistakes

Measuring an I-V curve without a simultaneous irradiance and temperature measurement — without this data the curve cannot be corrected to STC, making a meaningful comparison with the datasheet impossible.
Measuring an I-V curve during rapidly changing cloud cover — a quickly varying irradiance during the measurement (which takes a few seconds) can produce a distorted, non-representative curve.
Attributing a lower Pmax solely to "normal ageing" without assessing the shape of the curve — a recognizable step or slope deviation often points to a specific, fixable cause rather than general degradation.
Not consulting a reference curve from a healthy, comparable string when in doubt about the interpretation — a direct comparison under the same irradiance and temperature often makes a deviation clearer than the absolute curve alone.