Electroluminescence (EL) imaging is the most powerful quality control tool available to solar procurement managers and EPC engineers. A 5-second EL exposure reveals micro-cracks, broken cell fingers, bypass diode failures, and inactive cell areas that are invisible to visual inspection, IV curve tracing, and thermal imaging alike. For large procurements — whether 5,000 modules for a commercial project or 500,000 modules for utility-scale — including EL testing in your quality control plan is one of the highest-return quality investments you can make. This guide explains the physics, the testing protocol, how to interpret results, and when to require EL testing in your procurement contracts.
Electroluminescence (EL) testing involves passing a direct current through a solar cell in the dark at a forward bias — typically at the module's short-circuit current (Isc) — causing the silicon cells to emit near-infrared (NIR) light at approximately 1,100 nm wavelength. This light is captured by a silicon-sensitive camera (usually a CCD or InGaAs detector) to produce a greyscale image of the entire module. Areas with good silicon crystal structure appear bright; defective areas appear dark because recombination centres in cracked, broken, or deactivated regions consume charge carriers without emitting photons.
EL testing is completely non-destructive and leaves no mark on the module. Each module takes approximately 10–60 seconds to test, including transport time, making factory-level 100% testing feasible at production volumes of 1,000+ modules per day.
Silicon solar cells are direct-band-gap semiconductors (with an indirect band gap of approximately 1.12 eV). When forward-biased, minority carriers injected into the junction recombine radiatively, emitting photons with energy corresponding to the band gap — hence the near-infrared emission.
The intensity of EL emission at any point in the cell is proportional to the minority carrier concentration, which is in turn determined by:
Modern n-type TOPCon and HJT cells have higher minority carrier lifetime than p-type PERC, resulting in higher EL intensity and better imaging contrast — making EL imaging even more powerful on premium module technologies like LONGi Hi-MO 7 and Jinko Tiger Neo.
EL imaging is sensitive to a wide range of defects, many of which would not manifest as power loss until years into the module's operating life:
EL testing is performed at two distinct stages in the project lifecycle, with different objectives and protocols:
Factory EL testing is conducted at the module manufacturer's facility before modules are packed for shipment. It is the primary quality gate for detecting manufacturing defects. Key aspects:
Field EL testing is performed after modules are installed on the mounting structure, using portable EL imaging systems. It serves two purposes:
Field EL testing requires the PV system to be isolated from the inverter (open-circuit condition) and a portable DC power source connected to each string. Darkness is required (night-time testing) unless a sun-blocking tent is used for small strings. Typical field EL imaging systems weigh 8–15 kg and test one module in 10–30 seconds.
Several international standards and industry guidelines address EL testing:
| Defect Type | EL Appearance | Severity | Expected Power Impact | Action |
|---|---|---|---|---|
| Micro-crack (non-crossing, type A) | Thin dark line not crossing any finger | Minor | < 1% initially; monitor | Accept; document baseline |
| Micro-crack (finger-crossing, type B) | Dark line crossing one or more fingers; partial cell darkening | Moderate | 2–8% per affected cell | Accept at < 5% cell area; reject if > 10% |
| Inactive cell area (type C) | Large dark patch; > 20% of cell area affected | Severe | 5–20% per cell; cascades via bypass diode | Reject module |
| Inactive cell row / bypass diode failure | Entire row of cells uniformly dark | Critical | Up to 33% module power loss | Reject; replace module |
| Broken finger / contact ribbon failure | Dark horizontal or vertical line along edge or busbar | Moderate | 1–5%; risk of hot spot under partial shade | Evaluate; usually reject |
| Shunt (resistive) | Very bright small spot | Low–Moderate | Usually < 0.5%; risk of hot spot | Accept if isolated; reject if multiple |
| Grid finger interruption | Short dark segments within a cell | Minor | < 1% per cell | Accept; common in PERC manufacturing |
EL testing should be contractually specified at the following project stages:
Econo Solar works exclusively with module manufacturers who operate automated EL testing on their production lines — including all major suppliers LONGi, Jinko, JA Solar, and others. Our factory audit process verifies that EL testing equipment is calibrated, that rejection criteria meet IEA PVPS Task 13 severity-C thresholds, and that EL images are traceable to module serial numbers in the manufacturer's QMS.
For projects requiring third-party pre-shipment inspection, we co-ordinate with your preferred TPIC agency and ensure factory access, EL imaging setup, and documentation flow — so your lender's independent engineer receives the complete quality evidence package without schedule delays.
To discuss EL testing requirements for your next module procurement, contact Econo Solar. Our technical team will review your quality plan and propose a procurement and inspection schedule aligned with your project timeline.
EL testing is excellent at detecting defects that affect minority carrier recombination — cracks, broken fingers, deactivated cell areas, and bypass diode failures. However, it does not detect all defects: surface contamination (soiling, anti-reflection coating failure), discoloured encapsulant (yellowing EVA), backsheet cracking, junction box defects, or frame delamination require visual inspection or IV curve analysis to detect. Similarly, PID (Potential-Induced Degradation) requires a dedicated IEC 62804 test rather than EL alone. A comprehensive quality control plan uses EL testing in combination with visual inspection, IV curve tracing, and IR thermography.
Acceptable EL defect rates depend on the defect severity classification. For a large project procurement, the following guidelines are widely used: Type A defects (non-crossing micro-cracks) — acceptable up to 10% module incidence, with documentation; Type B defects (finger-crossing cracks with <5% inactive cell area) — acceptable up to 3–5% module incidence per production lot; Type C or critical defects (large inactive areas, bypass diode failures, complete cell rows inactive) — zero tolerance, with 100% replacement of affected modules required. If TPIC sampling reveals critical defect rates above 0.25% of the sample, require 100% EL re-testing of the entire lot before shipment.
Field EL testing of entire utility-scale plants (50+ MW) using traditional ground-level equipment is time-consuming — at 30 seconds per module and 100,000 modules for a 50 MW plant, a 100% test would take over 2,500 crew-hours. Practical approaches include: (1) Risk-based testing — 100% EL testing on all strings that show anomalies in I-V curve scanning (a rapid preliminary screen), plus random 5% sampling on all strings; (2) Drone-mounted EL systems — commercial drone EL platforms (several now at commercial availability for open-field plants) can cover 1 MW per night, enabling full-plant EL testing of large projects within one to two weeks; (3) Annual risk-prioritised testing — focus EL testing annually on the lowest-performing 10% of strings as identified by SCADA performance monitoring.
Econo Solar sources modules with factory EL testing as standard, with third-party inspection coordination and full QC documentation for your lender package.
Get a Free Quote