The manufacturing process faces ongoing challenges in detecting thread defects in cylinder head holes. The most common defects include missing threads and holes drilled too shallow, both of which cause assembly failures and reduce overall product reliability. Although excessively deep holes occur less frequently, they remain a secondary concern.
The current inspection process relies on manual Go/No-Go gauges and requires between 15 and 45 minutes per part. This approach slows production, increases labor costs, and introduces the potential for human error. Additional factors, such as metal shavings, residual coolant, and inconsistent lighting, further complicate defect detection and have contributed to the failure of previous automation attempts.
Team ThreadHeadz is developing an automated proof-of-concept system to detect shallow or missing threads while maintaining false positive and false negative rates below 5%. A successful system will reduce reliance on manual inspection, improve measurement consistency, and enhance overall manufacturing efficiency for both producers and end users.
The manufacturing process produces cylinder heads with approximately 327 threaded holes, each required to meet strict depth and threading specifications. The current inspection process relies on manual Go/No-Go gauges to detect defects such as shallow or missing threads. This method operates slowly, requires significant labor, and remains vulnerable to human error caused by environmental conditions, including coolant residue, metal shavings, and inconsistent lighting.
The engineering objective focuses on designing a reliable and efficient system capable of detecting shallow or missing threads in blind holes during production. The system aims to reduce the average inspection time from 30–45 minutes per part while maintaining or improving detection accuracy.