In this project, we focus on improving the quality control process for copper plates, formed through anode depletion in acid. During the depletion process, impurities can lead to the formation of nodules on the copper plates. The presence, number, and size of these nodules are key indicators of the plate's quality. Plates with larger or more numerous nodules are deemed defective and are rejected.
To automate this detection, we have implemented a vision system using three camerasmounted at a 90-degree turn on the conveyor. The cameras monitor the plates as they move along the production line, detecting and counting the nodules and measuring their sizes. Based on the nodule count and size, plates are categorised as either acceptable or defective.
Key Features
01
System Configuration
Cameras Setup: Three high-resolution cameras are strategically placed near the conveyor belt:
One camera positioned at the front of the plate.
Two cameras positioned at the back, each covering half of the plate’s surface.
This setup ensures full coverage of each plate, enabling thorough inspection.
Images are processed on a Local server attached in the network
02
Defect Detection and User Control
The system identifies nodules on the copper plates and counts both the number and size of the nodules. A user interface has been provided for operators, allowing them to:
Monitor a live view of the inspection process.
Review plates flagged as defective in real-time.
Set threshold limits for the size and number of nodules.
If the detected nodules exceed the set limits, the plate is flagged as defective. Each defective plate is automatically recorded in a MySQL database, where operators can review the defect details (such as nodule size, count, and time of detection) for future reference.
03
Integration with Transposing System
To seamlessly integrate with the existing production line, the system is connected to the machine's start/stop signals. This integration ensures that the nodule detection process begins when the machine is active and pauses when the machine is idle. Additionally, if a plate is detected as defective, the system sends a rejection signal, diverting the plate to a separate conveyor belt for rejected items.
Conclusion
This cathode nodule detection system greatly enhances the quality assurance process by automating the detection of impurities on copper plates. It provides a flexible, efficient, and reliable solution to ensure only high-quality plates proceed through the production line, while defective plates are rejected automatically, improving overall productivity and reducing manual inspection efforts.