IV curve tracing measures a photovoltaic string across its entire operating range, from short circuit to open circuit, producing a curve whose shape is a fingerprint of the string’s condition. A multimeter gives you two isolated points on that curve and tells you almost nothing about what lies between them. Many faults that cost real energy are invisible at those two points and obvious in the curve.
What the curve is and how it is taken
A healthy string’s current stays nearly flat as voltage rises from zero, then falls sharply near the open-circuit point, producing a characteristic knee. A curve tracer sweeps an electronic load across the string rapidly, sampling current and voltage at many points, and simultaneously records irradiance and module temperature so the measured curve can be corrected to a reference condition and compared with the expected curve from the module datasheet. The comparison is the diagnostic. A curve that matches the expected shape at slightly lower magnitude means uniform mild loss. A curve that is the wrong shape means something structural is wrong with the string.
Reading the deviations
Different faults distort the curve in different ways, and an experienced technician reads them like a chart. A reduced current across the whole curve points toward uniform soiling or general irradiance loss. A stepped curve with a notch suggests part of the string is bypassed, which points at a shaded or damaged module or a conducting bypass diode. A sloped section approaching short circuit suggests a shunt path within a cell or module. A rounded knee rather than a sharp one, or a slope near open circuit, suggests increased series resistance from degraded connectors, poor crimps, corroded terminations or damaged conductors. A voltage that is short by roughly one module’s worth points at a module bypassed or missing from the string.
Why the two-point measurement misses so much
Open-circuit voltage and short-circuit current are the endpoints. A string with significant added series resistance can show a perfectly normal open-circuit voltage, because no current is flowing and therefore no voltage drops across the resistance. It can show a nearly normal short-circuit current too. Yet at the operating point, where both voltage and current are substantial, that resistance is dissipating energy as heat and the maximum power the string can deliver has fallen. The lost power lives in the curvature between the two endpoints, which a multimeter never visits. This is why a string can test fine with handheld instruments and still be quietly costing energy and heating a termination.
When tracing is worth commissioning
Curve traces belong at commissioning, as a baseline for every string, and whenever a string underperforms its siblings, a monitoring platform shows an unexplained shortfall, a warranty claim is being prepared, or a system changes hands and the buyer wants to know what they are getting. The baseline matters enormously: comparing today’s curve against the same string’s curve from commissioning is far more sensitive than comparing it against a datasheet, because it eliminates the uncertainty in module tolerance and in the correction calculation. Any array large enough to justify per-string monitoring is large enough to justify baseline traces. Conditions matter as much as the measurement. A trace taken under broken cloud, at low sun angle or on a partially soiled array will not correct cleanly to reference conditions, so the technician needs stable irradiance above a reasonable threshold and an accurate cell temperature. A curve taken in poor conditions is not a weak result; it is an unusable one.
Why this is not a homeowner activity
Taking a curve requires disconnecting a string from the inverter and connecting a test instrument to conductors that are energized whenever the sun is up, at voltages high enough to sustain an arc that will not self-extinguish. It requires a calibrated tracer, an irradiance sensor, a temperature measurement, the ability to interpret corrected curves, and usually roof access. Every element of that is qualified-technician work with appropriate protective equipment. The owner’s role is to ask for the traces, keep the reports, and use them as the comparison point later. It is also worth asking whether the technician will trace every string or only the suspect one. Comparing a questionable string against a healthy neighbor measured minutes apart under the same sky removes most of the uncertainty in the correction, and it costs very little extra once the instrument is already on site.
What to check before you act
- Request baseline curve traces for every string at commissioning.
- Keep the reports with the irradiance and temperature conditions recorded.
- Ask for a trace whenever one string underperforms its identical sibling.
- Compare against the baseline curve, not only against the datasheet.
Common questions
Can a clamp meter find the same problems?
No. A clamp meter reads current at one operating point. Series resistance, shunt paths and partial bypassing all live in the shape of the curve between the endpoints.
How often should strings be traced?
At commissioning, then when monitoring flags a problem or a warranty claim is being built. Routine periodic tracing is more common on commercial arrays than on residential ones.
The practical takeaway
Ask for baseline curve traces at commissioning and keep them, because they turn every later investigation into a direct comparison. When a string underperforms, request a trace rather than accepting a voltage reading, and leave the measurement to a qualified technician.
References and further reading
- NREL: photovoltaics research including module characterization and field diagnostics
- NREL: a technical report from the laboratory’s photovoltaics publication library
- OSHA: Subpart S electrical standards covering work on energized equipment
Numerical scenarios are illustrative unless identified otherwise. Follow the exact product instructions; component ratings and local installation requirements can differ.
