Visit the Adept Scientific web site
Click on the advert above to visit the company web site

Product category: Vision and scanning systems
News Release from: Cognex UK | Subject: In-Sight 2000 machine vision sensor
Edited by the Manufacturingtalk Editorial Team on 13 May 2002

Machine vision technology increases
quality

If you wince at the thought of a catheter kinking during insertion, or pulling apart upon removal, then you understand the necessity of test and inspection for catheter manufacturers.

If you wince at the thought of a catheter kinking during insertion, or pulling apart upon removal, then you understand the necessity of test and inspection for catheter manufacturers For example, Arrow International a manufacturer of disposable catheters for critical and cardiac care, says it recently increased productivity, quality, and reduced labor costs using machine vision technology

A short time ago, the company automated assembly of its catheter sheaths and began using an In-Sight 2000 machine vision sensor from Cognex to ensure proper assembly.

The sensor consists of a compact industrial hardened processing unit, a 640 x 480 x 8-bit progressive scan CCD camera, and a Nintendo-like handheld controller, as well as a library of vision software tools.

In-Sight automatically measures the distance between the end of a sheath body and a flex-straw-like piece of corrugated plastic extrusion that is insert-molded to the body to ensure the gap is within tolerance.

"The inspection was necessary," explains Kevin Grover, a control systems project engineer for Arrow, "because if the gap were too big, the corrugated extrusion may kink, instead of bending at the corrugation as it is being inserted into the patient." Grover integrated the vision sensor onto a custom-built machine that is designed for the ultrasonic welding of caps onto the bodies.

To prevent a second mode of failure, the automated assembly machine has a pull-test station.

"The pull test checks the strength of the insert-molded joint between the extruded catheter material and the molded sheath body," Grover explains.

"If the joint is too weak, the assembly might pull apart when withdrawn from the patient." During production, an operator puts a valve into a sheath body, caps it, then loads one sheath, valve and cap-assembly at a time into six separate transports on a rotary dial.

The rotary dial indexes the assembly to the ultrasonic welding station, which welds the cap to the body.

Next, the transport indexes to the pull-test station that checks the insert-molded joint.

Once the caps have been welded to the devices and the assemblies pass the pull test, the parts are indexed on their respective transports in an upright position, one at a time, in front of the side-mounted In-Sight vision camera.

Red backlighting behind the device helps optimise contrast between the device and background to give the vision sensor the clearest possible view of the measurement area, according to Grover.

As a device enters the camera's field of view, an Allen-Bradley MicroLogix 1000 PLC that is linked to the sensor through I/0 triggers the camera to capture an image of the device.

"The image capture trigger is based on input from the indexer," says Grover.

Then the image is instantly transferred to the vision-processing unit for analysis with a variety of vision software tools.

After inspection, the transport drops the catheter device into a tube that shifts between good and bad bins based on pass or fail outputs from the vision-processing unit to the PLC.

"If we get three failures in a row," explains Grover, "a signal is sent from In-Sight to the PLC to stop the machine.

This way, an operator can go in and investigate the problem on the spot." To set up the inspection, Grover used the In-Sight sensor's vision spreadsheet interface.

The process involved selecting vision tools, such as In-Sight's PatFind, and parameters from drop-down menus using the handheld control pad.

The vision spreadsheet then automatically generates tool results into worksheet cells, which are linked together to set up the measurement operation.

"To first find the pattern in the field of we used the PatFind tool, which determines the overall position of the de vice," says Grover.

"It remembers what that part looks like in any orientation." To measure the gap, Grover used one of the In-Sight Edge Detection tools, which finds two points and calculates the distance between them.

During operation, the edge-detection tool finds the two edges, measures the distance between them, and outputs a value.

"We had to set the high and low reject limits," explains Grover.

" Anything between those two values passes." A Cognex-supplied flat-panel LCD monitor that is mounted above the machine displays live images of each device being inspected and corresponding pass/fail data for the operator.

According to Grover, since the sensor has been installed, it has been inspecting approximately 6,000 devices per eight-hour shift with a high degree of reliability.

"By adding inspection capability to our automated assembly l process, we are able to ensure that all products leaving our facility are defect free.

This helps eliminate the possibility of failure in the field." . Request a free brochure from Cognex UK ...

(This was Manufacturingtalk's Top Story on 10 May 2002).

Cognex UK: contact details and other news
Email this article to a colleague
Register for the free Manufacturingtalk email newsletter
Manufacturingtalk Home Page

Search the Pro-Talk network of sites

Visit the Adept Scientific web site