Machine vision applications typically enhance existing quality improvement techniques by making inspections that fall within three categories:
- Defect Detection
- Tolerance Monitoring
- Component Measurement
When a large manufacturer of fluorescent lighting products needed a new automated inspection system for identifying defective glass tubes, they called upon Automation Software & Engineering (AS&E) to develop a solution that would both improve the inspection process, and be easier to maintain than the 15-year-old technology in their current system.
Drawing upon our extensive experience in implementing process monitoring, control, and machine vision systems, AS&E responded with a new, cutting-edge quality improvement process that leveraged advanced Gigabit Ethernet camera technology to deliver high-speed, high resolution images and ensure that the highest standards of quality are met.
The manufacture of fluorescent lighting is a multi-step process that starts with the creation of a precisely-formed glass tube. When caught early, a less-than-perfect tube can simply be melted down and re-used in the manufacturing process. But if a defective tube progresses too far in the process, it becomes extremely expensive to salvage and handle the electrodes, phosphorus, mercury, and other hazardous substances.
This AS&E quality improvement solution uses six networked area scan cameras working with AS&E’s VisTrax™ visual recognition system to detect tube defects early in the process, while keeping the production line moving at the rate of 70 tubes per minute. The glass tubes, which have been cut to length, heated, and squeezed at each end to form a slightly smaller concentric opening, are sent past the inspection system where a walking beam positions each backlit tube under the six-camera apparatus.
Two cameras, placed perpendicular to one another over the middle of the tube, send images to the host PC to detect any bowing. The remaining four cameras – two at each end – provide images of the formed openings, which are analyzed for size, shape and concentricity. All six cameras are networked to an Ethernet hub which is connected via an Intel NIC to a PC running the VisTrax system. The entire inspection process, including capturing images from all six cameras and analyzing the images with VisTrax, takes place in 820 milliseconds.
Meanwhile, the custom-designed quality improvement plan had other benefits, as well. The ability to use affordable Cat5E cabling instead of 6-pair, twisted-shielded cables, reduced cabling costs by nearly 75 percent for the 50-foot cabling run between the cameras and the PC. Further cost reductions resulted from the use of the hub/NIC combination, which has eliminated the need for more expensive frame grabber technology.
The result was a powerful, yet cost-effective quality improvement solution with numerous reusable components capable of being adapted to other factory inspection applications. AS&E has developed similar customized part inspection systems for the steel mill, high performance metals, quartz tubing, medical, food processing, power generation, solar energy, electronics and industrial coatings industries.