Smartphone Imaging for Solar Cell Microcracks

ISEF Category: Energy: Sustainable Materials and Design

Subcategory: Solar Process, Materials, and Design  ·  Difficulty: Intermediate  ·  Setup: School Lab  ·  Time: 1 to 2 Months

The Hook

A tiny crack can cut a solar cell’s output long before you can see it by eye. That makes used PV cells a great target for image-based testing. You can turn a phone camera into a simple inspection tool and ask which defects it catches best. This kind of project connects optics, materials, and clean energy.

What Is It?

Photoluminescence imaging means you make a material glow, then measure that glow with a camera. In solar cells, damaged spots often glow differently from healthy spots because cracks change how charge carriers move. Think of it like shining a flashlight through a cracked phone screen, except you are reading the pattern of light the material gives back.

Your goal is not to build a commercial inspection system. Your goal is to test whether a smartphone camera can separate cracked regions from intact regions well enough to map damage. You can compare images from different phone settings, different filters, or different lighting setups, then see which setup gives the clearest defect contrast. That gives you a real measurement problem, not just a picture-taking task.

Why This Is a Good Topic

This is a strong science fair topic because you can test image quality, defect detection, and analysis method without needing a full solar factory. The real-world link is clear, since cracked PV cells waste material and lower solar output. You can learn optics, image analysis, and experimental design while working with a question that matters in renewable energy.

Research Questions

• How does smartphone camera exposure affect the contrast between cracked and intact regions in reused PV cells?
• How does the choice of optical filter affect the visibility of microcracks in photoluminescence images?
• Does image averaging improve crack detection compared with a single photo?
• To what extent do different phone camera models change defect contrast in the same solar cell?
• Which image thresholding method separates cracked and uncracked regions most accurately?
• How does crack density relate to the fraction of the cell area that shows low photoluminescence signal?

Basic Materials

• Used PV cells salvaged from broken modules.
• Smartphone with manual camera controls.
• Tripod or phone stand.
• Dark box or darkened room setup.
• Uniform excitation light source matched to the sample setup.
• Simple optical filter or infrared-pass filter if your setup uses one.
• Ruler or printed scale for image calibration.
• White card or matte background for alignment.
• Computer for image analysis.
• Spreadsheet software for data logging.

Advanced Materials

• Used PV cells with known damage levels.
• Smartphone with RAW capture support.
• Fixed mount with repeatable geometry.
• Calibrated excitation source.
• Optical filters for emission selection.
• Reference photoluminescence standard.
• Higher-resolution camera or scientific imaging sensor.
• Image analysis software for segmentation and contrast metrics.
• Microscopy or magnified inspection tool for ground-truth comparison.
• Electrical testing setup for pairing image data with performance loss.

Software & Tools

• ImageJ: Measures contrast, segments crack regions, and compares image features across samples.
• Python: Automates image cleanup, threshold testing, and batch analysis.
• Google Sheets: Organizes image scores, sample labels, and comparison data.
• GeoGebra: Helps plot calibration curves and compare trend lines.
• NIH ImageJ plugins: Adds extra tools for filtering, measurement, and image math.

Experiment Steps

1. Define the defect signal you will measure, such as crack contrast, crack area, or edge sharpness.
2. Choose one imaging setup first, then lock the phone position, sample position, and lighting geometry.
3. Plan a set of reference samples that span low to high damage so you can compare signal changes across a range.
4. Build a calibration approach that turns pixel values into a repeatable measurement.
5. Design controls that separate true crack signal from shadows, glare, and background texture.
6. Choose one analysis method and one backup method so you can check whether both give the same ranking of samples.

Common Pitfalls

• Using auto exposure or auto focus, which changes image brightness from sample to sample.
• Letting room light leak into the setup, which can hide weak photoluminescence contrast.
• Comparing cells with different surface dirt or module residue, which can look like crack signal.
• Skipping a ground-truth check, which makes it hard to tell whether bright or dark regions really match damage.
• Treating every dark line as a crack, which inflates false positives from solder lines, shading, or texture.

What Makes This Competitive

A stronger project goes beyond simple before-and-after photos. You can compare multiple phones, filters, or analysis methods and report which one gives the best defect detection. You can also test how image metrics track real damage labels, not just visible cracks. That turns your project into a measurement study with clear engineering value.

Project Variations

• Test whether infrared-pass filters improve crack contrast more than visible-light imaging.
• Compare cracked cells from different module ages or failure types to see whether the imaging method works across sample groups.
• Pair image-based crack maps with power output data to test how well visual damage predicts electrical loss.

Learn More

• NASA NTRS: Search for reports on solar cell defect imaging and photoluminescence methods.
• PubMed: Search review articles on photoluminescence imaging and semiconductor defect detection.
• NOAA Solar Resource tools: Use background material on solar energy systems and performance factors.
• MIT OpenCourseWare: Search materials science and imaging courses for background on semiconductors and image analysis.
• Solar Energy Materials and Solar Cells: Read peer-reviewed papers on defect imaging and PV materials through your library or journal abstracts.