Capacitor Soldering Inspection
Detect soldering defects with deep learning
Graphical programming environment for deep learning-based industrial image analysis
Capacitors are crucial electronic parts that are soldered into inverters, chargers, and other circuits in electric vehicles (EVs). They may also be interconnected to supercapacitors or ultracapacitors. The low resistance and high current carrying capacity of these soldered electrical connections are essential to EV operation. If a connection to a capacitor is weak and conducts poorly, the vehicle’s efficiency will suffer. If a crucial connection breaks completely, it can lead to serious malfunction. If that broken connection is located somewhere like the auxiliary battery, the vehicle might be completely disabled, and require service.
Connections created by soldering can vary significantly in appearance without affecting function, while unacceptable connections can visually resemble functional ones. Given the consequences of connectivity problems, soldered parts with suspected defects must often be pulled and X-rayed to check their connections, with all the expense and delay that entails.
Cognex Deep Learning’s defect detection and classification tools are trained on a wide range of good and defective solder connection variations and learn to accurately classify and distinguish functional flaws from merely cosmetic ones. Using an example-based approach instead of traditional rule-based machine vision allows shortens application development time.