Abstract
Daylight electroluminescence (dEL) inspection using InGaAs cameras has proven to be a powerful technique for assessing the condition of photovoltaic (PV) modules in the field. Recent advancements have shown it suitable for quality control and evaluation tasks in large-scale solar installations. The quality of dEL images is crucial for accurately identifying potential defects. Therefore, it is important to determine which camera optical stack yields stronger signals in dEL imaging. Camera optical stacks typically include specialized short-wave infrared (SWIR) lenses and bandpass filters to reduce background sunlight. To further limit the light intensity reaching the sensor and prevent saturation,options include adjusting the lens iris, using neutral density (ND) filters, or reducing exposure time. The choice among these depends on system constraints. Even though reducing exposure time is the easiest way to accomplish no saturation, high exposure time reduces noise, so ND filters and the iris are interesting options independent of the camera’s internal controller. This study compares the light intensity reduction methods between ND filters and iris providing higher EL and dEL image quality using signal-to-noise ratio (SNR) as metric. Two SNR metrics (SNRkari and SNR(25)) are used to evaluate the configurations. We compare two setups: one using a C-RED 3 InGaAs camera with a SWIR lens (F-stop range 1.4–16) and a bandpass filter, and another using the same camera and lens fixed at Fstop 1.4 (fully open) combined with ND filters of varying transmittance (0.73 to 0.02). Indoor EL data is used to characterize light attenuation for each configuration. Subsequently, dEL images are captured under 600-800 W/m²irradiance for both setups. The results show that using the iris to reduce light intensity yields higher image quality. This is attributed to the increased depth of field resulting from a smaller optical aperture, which enhances the focus range and sharpness of the captured images. In conclusion, the study demonstrates that using a lens with an adjustable iris is more effective for dEL imaging with InGaAs cameras. This finding is valuable for optimizing optical setups to achieve high-SNR images in PV module inspections.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of 42nd European Photovoltaic Solar Energy Conference and Exhibition |
| Number of pages | 6 |
| Publisher | EU PVSEC |
| Publication date | 2025 |
| Article number | 020191 |
| ISBN (Print) | 3-936338-93-0 |
| DOIs | |
| Publication status | Published - 2025 |
| Event | 42nd European Photovoltaic Solar Energy Conference and Exhibition - Bilbao Exhibition Centre, Bilbao, Spain Duration: 22 Sept 2025 → 26 Sept 2025 |
Conference
| Conference | 42nd European Photovoltaic Solar Energy Conference and Exhibition |
|---|---|
| Location | Bilbao Exhibition Centre |
| Country/Territory | Spain |
| City | Bilbao |
| Period | 22/09/2025 → 26/09/2025 |
UN SDGs
This output contributes to the following UN Sustainable Development Goals (SDGs)
-
SDG 7 Affordable and Clean Energy
Keywords
- Daylight Electroluminescence
- InGaAs Camera
- EL Signal-to-Noise Ratio
- Photovoltaic Module Inspection
Fingerprint
Dive into the research topics of 'Evaluating the Efficacy of Neutral Density Filters and Camera Lens Aperture for Reducing Sensor Saturation and Increase Image Quality in Daylight EL'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver