As unlikely as it may seem, there are situations in which manufacturing top quality parts can cause just as much downtime, lost revenue and frustration as producing defective ones. Shop personnel at Accuma Corporation's Beatrice plant in Nebraska were all too familiar with such a recurring problem.

One of the premiere manufacturers of plastic injection molded components for batteries, Accuma, is headquartered in Italy and has sales and manufacturing operations in the United States (North Carolina and Nebraska), Germany, England and China. Accuma's Beatrice facility manufactures more than 300 types of battery containers, lids and accessories such as handles, vents, and caps for the automotive industry. Engineers at the Beatrice plant had relied on laser micrometers to determine whether a lead battery post was positive or negative, and to ensure that it was positioned correctly based on the diameter of the post. "The micrometers could detect battery lids with two positive or two negative posts or incorrect orientation when these conditions existed," says Scott Wagers, Maintenance Supervisor at Accuma's Beatrice plant. "The problem is that they were falsely rejecting an excessive quantity of parts - 50% - because any part that didn't align pefectly in front of the sensor was auomatically rejected."
Performing secondary inspections of rejected parts by gauging whether they were a go or no go was costly and time consuming. To find a solution to this ongoing problem, Wagers contacted Hartfiel Automation in Des Moines, IA.

Getting a Better Look

Hartfiel engineers provided Accuma with a Teledyne DALSA BOA BVS-0640M-INS vision sensor demo unit for a one-week trial that turned into a permanent solution. The BOA is a VGA (640x480) monochrome vision sensor that contains all of the elements of an industrial machine vision system within a tiny (44 x 44 x 39 mm), rugged IP-67 rated enclosure. Most importantly for Accuma's needs, the BOA vision system tracks each part throughout a view window so that the part does not have to be perfectly positioned during the inspection process.
Accuma's BOA, which is fitted with an 8 mm lens and uses a 5" x 5" red backlight, interfaces with a Windows-based, 6-axis robotic battery lid pick-and-place application. Each minute, the robot presents two battery lids to the BOA in a constantly rotated, 180-degree sequence (Lid #1's positive battery post, then Lid #2's negative battery post, and so on). Using its embedded inspection software, the BOA determines whether the battery post in question is correct. An operator packages all of the good parts and places any defective parts in a locked reject bin.
"Since installing the BOA, we have drastically reduced the number of false rejects and eliminated the need for a secondary inspection of rejected parts," Wagers says. "If the camera rejects a part, we are confident that it is a non-conforming part."
Installing the vision sensor was straightforward. "We learned how to program the vision system using the manual that came with the product and we haven't needed any training," Wagers comments. He and others in the maintenance department are the only ones who make adjustments to the vision system.

Gaining More Vision

"The biggest benefits to us as a result of using this vision system have been the overall repeatability of the BOA, the ease of set up to various part configurations and the confidence we have in the system to provide reliable results," Wagers says. Since the BOA has been up and running, Accuma hasn't shipped a single non-conforming part.
Accuma installed a second BOA to duplicate this successful solution on another battery lid line, and has plans to install additional vision systems throughout the facility to solve a variety of quality applications for battery box covers and containers.
"Hartfiel's recommendation to use the Teledyne DALSA system in this application was spot on," Wagers concludes. "The BOA has reliably and cost-effectively solved our false reject problems and we look forward to achieving more cost savings and productivity gains using BOA vision solutions in future applications."