Build better machines with machine vision

Cost-effective, high-performance machine vision tools help machine builders increase accuracy and throughput on a new machine design using four cameras, one controller, and high-speed communications.

04/02/2012


If you haven’t integrated machine vision on an automation line or machines recently, you could be missing some powerful and cost-effective solutions. Andy Wright, manager of automation engineering at Istech (Dover, Penn.), helps the company provide turnkey custom automation solutions for a variety of industries using advanced computer and automation technologies.

Istech provides turnkey custom automation solutions for a variety of industries using advanced computer and automation technologies. Courtesy: Teledyne Dalsa

“We’ve been integrating vision for more than 15 years at Istech,” Wright said, and he’s been open about considering lower cost, higher performance systems recommended by distributor Faber Industrial Technologies.

Aligning features

Istech’s first project using a new vision platform vision consisted of designing and building a custom assembly machine for Conductive Technologies, a long-time customer in the medical device industry. The machine applies lids—clear plastic strips that come on a roll and peel off like labels—to sheets of plastic substrates. Each sheet has six silk-screened substrates, or coupons, on it. To apply the lids reliably and accurately to meet production goals, a series of features on each lid must be aligned with each corresponding coupon.

Plant floor personnel had been performing this process by hand; however, this method lacked consistency, was time consuming, and did not provide the required throughput and accuracy. As a result, Istech engineers designed a machine vision system into the assembly machine.

Four high-resolution, GigE cameras are connected to an industrial controller, which provides expandable Gigabit Ethernet camera interfaces that significantly decrease system costs for multicamera applications like this one. Two of the cameras are positioned on opposite sides of the sheet, while the other two are located at each end of the lid.

After each plastic substrate sheet is loaded onto a moving vacuum table, two GigE cameras identify and locate the six coupons. Courtesy: Teledyne Dalsa

After a sheet is loaded onto a moving vacuum table, the sheet cameras identify and locate the six coupons. The lid cameras identify and locate the corresponding features on each lid. Vision system software performs calculations and provides X, Y, and rotational correction values. These values are transferred to an Epson robot, which positions the lid on each coupon accordingly. Heat sealers then attach each lid. Once all six coupons are complete, the sheet is offloaded to a stack for further processing.

High-speed synchronization

The machine vision system’s ability to process information asynchronously was essential to the success of this application. To achieve the required cycle rates, Istech programmed the robot to work ahead, so that while one sheet is being taken off the vacuum table, a new lid is ready to go as the next sheet is loaded.

Normally, the vision values are completed first for a given coupon; however, during the sheet exchange sequence, the lid data is calculated first. As a result, information from the cameras must be transmitted asynchronously.

Two additional GigE cameras identify and locate the corresponding features on each lid. Sherlock vision software performs calculations and provides X, Y, and rotational correction values, which an Epson robot uses to position the lid on each coupon accord

Using the machine vision software, the system handles asynchronous communications quite well, Wright said. “The algorithms in this application were straightforward,” and data handling is what made the machine vision system so successful in this case.  

Faster than 30 sec

When the machine was first installed, Istech’s customer, Conductive Technologies, stringently evaluated the product that came off the acceptance run. Both accuracy and throughput were deemed excellent. Before the system was installed, human assemblers at Conductive Technologies required several minutes to align and apply lids to each sheet.

“Our spec was to run one sheet in 30 seconds, and we were able to beat that rate,” Wright said. And, since the machine has been installed, Conductive Technologies reports that lid placement accuracy far exceeds the project’s initial goals.

“I am particularly impressed” with the machine vision software since it provides an “extremely powerful set of programming tools in a cost-effective product,” Wright said, adding that he’s looking forward to integrating it in future projects.

- Maureen Clancy is marketing project manager, Teledyne Dalsa. Edited by Mark T. Hoske, content manager, CFE Media, Control Engineering

www.conductivetech.com

www.istech.us.com 

www.robots.epson.com 

www.teledynedalsa.com 

www.youtube.com/user/ISTECHInc 

http://controleng.com/machinevision

ONLINE extra - Additional Istech machine vision project notes

- Four high-resolution, Gig E cameras are connected to a Teledyne Dalsa Geva industrial controller.

- Teledyne Dalsa’s Sherlock vision system software performs calculations and provides X, Y, and rotational correction values.

- “Using the Sherlock software, the Teledyne Dalsa system handles asynchronous communications quite well,” Wright said.

- “Based on how happy we were with the cost and performance of our first project, we’ll definitely use Teledyne Dalsa products in the future,” Wright said.