# Cam Design and Pin Defect Detection of Cam Pin Insertion Machine in IGBT Packaging

## TL;DR

This paper presents an automated pin insertion machine for IGBT modules, improving productivity and quality in semiconductor packaging.

## Contribution

A novel pin insertion machine with cam design and image-based defect detection for IGBT packaging is developed and tested.

## Key findings

- The pin insertion machine achieved a 99.75% pass rate for pin insertion.
- The average insertion time per pin was 2.84 seconds with 0.02 mm guidance accuracy.
- The pin cutting pass rate reached 97% with defect detection using image algorithms.

## Abstract

Packaging equipment plays a crucial role in the semiconductor industry by enhancing product quality and reducing labor costs through automation. Research was conducted on IGBT module packaging equipment (an automatic pin insertion machine) during the pin assembly process of insulated gate bipolar transistor (IGBT) modules to improve productivity and product quality. First, the manual pin assembly process was divided into four stages: feeding, stabilizing, clamping, and inserting. Each stage was completed by separate cams, and corresponding step timing diagrams are drawn. The profiles of the four cams were designed and verified through theoretical calculations and kinematic simulations using a seventh-degree polynomial curve fitting method. Then, image algorithms were developed to detect pin tilt defects, pin tip defects, and to provide visual guidance for pin insertion. Finally, a pin insertion machine and its human–machine interaction interface were constructed. On-machine results show that the pin cutting pass rate reached 97%, the average insertion time for one pin was 2.84 s, the pass rate for pin insertion reached 99.75%, and the pin image guidance accuracy was 0.02 mm. Therefore, the designed pin assembly machine can reliably and consistently perform the pin insertion task, providing theoretical and experimental insights for the automated production of IGBT modules.