Background and Buyer Wants
In electronics manufacturing, guaranteeing exact glue dishing out earlier than the PCB enters the soldering machine is essential for product high quality. A buyer required an automatic inspection system to confirm the accuracy of glue software, as defects in dishing out might result in product failures.
After glue is disbursed entrance dishing out machine, points resembling incomplete software, leakage, or unintended contact with video reminiscence elements might happen, requiring inspection. Beforehand, the inspection relied on guide monitoring, which was susceptible to human fatigue, inconsistencies, and a scarcity of traceable inspection knowledge.
Challenges
- Guide Inspection Limitations: Operators wanted to always monitor the method, growing the danger of fatigue and misjudgment.
- Excessive Employees Turnover: Totally different inspectors utilized various standards, making standardization troublesome.
- Lack of Traceability: Guide inspection didn’t permit for picture storage or knowledge monitoring, complicating manufacturing administration.
Resolution & Key Applied sciences
Techman Robotic’s cobot arm offered an automatic inspection resolution, integrating superior imaginative and prescient and AI classification know-how to make sure exact glue software detection. Successfully fixing the above drawbacks and obtain automated workflow.
- Imaging & Detection
- TM AI Cobot had been deployed to examine glue software on PCBs earlier than the soldering machine course of
- The cobot arm detected right and incorrect dishing out, distinguishing between “OK” (correctly utilized glue) and “NG” (glue lacking or misplaced)
- All photos and inspection outcomes had been recorded in a database for traceability and course of optimization
- AI Mannequin Coaching
- Coaching Time: Roughly 10 minutes
- AI Perform Used: AI Classification
- Dataset Measurement: 130 OK photos, 80 NG photos
- OK Merchandise: Handed inspection and continued to the following manufacturing stage
- NG Merchandise: Flagged for rework, stopping faulty items from advancing within the course of
- Efficiency
- Inspection Pace: 108 visible duties accomplished in simply 30 seconds, considerably bettering effectivity.
- Achieved 99.9% inspection accuracy
- False alarm price of lower than 0.1%
- Overkill price of lower than 0.1%
Utility Eventualities
- Automated verification of glue dishing out earlier than the PCB enters the Soldering machine
- Identification of lacking or misplaced glue to stop product defects
- Actual-time inspection with automated knowledge assortment for high quality management
Advantages
- Enhanced Accuracy & Consistency
- Eradicated human error by automating the inspection course of.
- Achieved 99.9% accuracy with AI classification, guaranteeing defect-free merchandise.
- Elevated Effectivity
- Totally automated inspection decreased reliance on guide operators.
- Processing time was considerably shortened, enhancing manufacturing pace.
- Improved Traceability
- AI-powered inspection saved all take a look at outcomes and pictures for high quality administration.
- Barcode studying and serial quantity monitoring enabled environment friendly manufacturing monitoring.
Conclusion
The cobot efficiently reworked PCB glue dishing out inspection by integrating good automation, AI-driven classification, and superior imaginative and prescient know-how. This resolution not solely enhances accuracy and effectivity but additionally offers worthwhile traceability knowledge, guaranteeing increased manufacturing requirements and decrease operational prices. The case highlights how producers can leverage AI to optimize manufacturing traces and preserve superior high quality management.
