Case Sharing: FA FVI Broken Half Detection for PCBs

Background and Buyer Wants

Within the fast-paced world of electronics manufacturing, making certain the standard of merchandise earlier than packaging is vital. A serious buyer required an answer to detect broken or lacking elements on printed circuit boards (PCBs) with excessive accuracy and effectivity. Guide inspection strategies struggled to maintain up with manufacturing calls for and infrequently ignored small defects.

Challenges

1. Detection of Small Anomalies: Human inspectors discovered it difficult to constantly determine tiny defects on PCBs.
2. Excessive Quantity Manufacturing: The necessity for fast, scalable inspections to match manufacturing cycle occasions.
3. Labor-Intensive Course of: Dependence on handbook inspections elevated prices and launched inconsistencies.

Answer & Key Applied sciences

Techman Robotic’s TM AI Cobot supplied a complete automated answer, integrating superior imaging and AI classification expertise.

1. Imaging & Detection

   • The Eye-in-Hand (EIH) digicam enabled exact positioning and the exterior digicam enabled multi-point visible inspection, performing picture seize from a number of angles to make sure each element was inspected precisely.
   • Pictures had been analyzed utilizing the AI mannequin to categorise elements as Cross (OK) or Fail (NG).

2. AI Mannequin Coaching

   • Leveraged classification AI to coach the system with a dataset of 70 pictures (40 OK, 30 NG).
   • Coaching time was minimized to only quarter-hour, enabling fast adaptation to adjustments in manufacturing necessities.

3. Automated Workflow

   • OK Merchandise: Robotically directed to the subsequent station.
   • NG Merchandise: Recognized and the cobot arm will select the faulty half to a devoted cycle space for additional course of.
   • Outcomes had been computed on the AOI Edge after which transmitted to the robotic, which executes the decision-making to make sure a seamless manufacturing movement.

Utility Eventualities

• Detecting lacking or broken elements earlier than packaging
• Making certain product high quality by figuring out small anomalies early within the course of

Advantages

1. Enhanced Accuracy

   • Achieved an inspection accuracy fee of 99.99%.
   • False alarm and overkill charges had been lowered to lower than 1%, making certain reliability.

2. Elevated Effectivity

   • Automation improved inspection pace and lowered manpower necessities by 50%.
   • Excessive-speed inspection aligned seamlessly with manufacturing cycle occasions.

3. Price Discount

   • Decrease reliance on handbook labor minimized operational prices whereas enhancing consistency.

Conclusion

The half detection answer utilizing TM AI Cobot demonstrates how sensible automation transforms inspection processes. By combining AI-powered classification with exact imaginative and prescient expertise, this case exemplifies how producers can obtain unparalleled effectivity, accuracy, and price financial savings in trendy manufacturing strains.