Industrial Engineering

Chain-Link Monitoring System Prevents Unplanned Downtime

Chain-Link Monitoring System Prevents Unplanned Downtime
Assembly Magazine May 2021
By Jim Camillo
https://www.assemblymag.com/articles/96383-chain-link-monitoring-system-prevents-unplanned-downtime

Chain drives are essential to automotive manufacturing. These drives are typically used to transport automotive frames down kilometer-long production lines. Unfortunately, constant vibration and pull on the chain links can cause them to stretch, be distorted and break, resulting in unplanned downtime, as well as lost productivity and profits.

The traditional approach to chain drive inspection is to have a technician hold up a metal rectangle of the same length as the link and visually compare the two. If a link looks problematic, he marks the chain and hopes it continues to function until the next scheduled maintenance. If the chain breaks in the meantime, the manufacturer incurs expensive downtime until the chain is repaired.

In 2018, engineers at the General Motors of Canada Co. (GM Canada) looked into alternate inspection methods for chain drives used on the company’s power transmission assembly line. Their first attempt involved using discrete photoelectric sensors with laser pointers to measure chain link lengths.

After that, the engineers worked closely with vision specialists at Cognex Corp. Together, they developed an In-Sight 2000-vision-sensor-based system that measures the stretching of every chain link in real time and automatically marks noncompliant links without slowing or stopping the line.

The system consists of vision and proximity sensors, a red LED illuminator, a red filter and image processing software. As the illuminator backlights the chain—to provide a crisp outline of the link under inspection—the filter helps eliminate the effects of ambient light and the proximity sensor triggers the camera to take several images of each link when it is within the field of view.

After this, Cognex In-Sight Explorer image processing software uses edge-detection algorithms to identify the center of each chain link. A caliper tool then measures the link’s internal length.

“Basic pass-fail inspections can be accomplished with intuitive, point-and-click setup tools,” says Christopher Eid, a system designer at GM Canada’s electrical engineering paint department. “In this case, when edge detection finds both link edges, the distance between the two connections is measured, and the result is placed under the respective chain-stretch limit category. The HMI displays the relevant information.”

Caliper measurement data helps GM engineers determine each link’s percentage of stretch variation. When the variation reaches 4 percent or higher, the vision system sends a signal to a nearby PLC through an optional Cognex I/O module.

The PLC, in turn, directs a downstream paint sprayer to mark these links for replacement during the next planned maintenance cycle. During inspection, the vision system automatically counts how many chain links need to be replaced, as well as the number of rotations for a 1,680-foot chain.

“We started on our fastest conveyor at 70 percent coverage because the camera runs faster than the PLC,” notes Eid. “Today, we’re doing most of the analysis on the camera, and we use a nine-frame buffer bit on the vision sensor. [This gives] us more coverage per scan. Based on the system’s success, we plan to use several more in the near future and eventually modify it [to measure] overhead chains for painting stations.”