A
manufacturer of seatbeat retractor gears wanted to improve the quality
and throughput of their products by incorporating 100% vision inspection
onto their production line. Each tooth of each gear had to be inspected
to ensure minimum and maximum gear diameter, and to look for possible
burrs that might be present on the ends of any gear teeth.
Solution
The vision component of the solution used Sherlock integrated with a DALSA frame grabber inside a standard PC, interfacing to a standard analog camera. When the part arrives in the inspection bed, its presence is detected by a sensor, which triggers a command for the camera to take a picture. Custom ring lighting illuminates only the top surface of the gear on the part being inspected, without showing the background.
With the part in place, Sherlock analyzes the image and determines if it is "Good" or "Bad." It measures the diameter of the part, counts to make sure all the teeth are there, and makes sure that the teeth are the right size and shape. The software then shuttles a gate into the proper position to allow "Good" parts to pass into a pre-positioned shipping container or "Bad" parts to be automatically dropped into a locked reject bin. The entire process takes about 50 ms per part.
Aside from performing the actual inspection, Sherlock's built-in I/O interface eliminated the need for programmable logic controllers (PLCs). Most vision solutions require one or more PLCs to control I/O functions, such as sensing that a part is in the right place, and actuating the automation equipment to move it to other stations. In this system, these functions include the on/off feeder bowl operation, reject and pass automation, air cylinder actuators, and data collection. Sherlock was able to handle all of these activities without the cost of additional PLCs. Courtesy of Inosys, Inc.
A manufacturer of catalytic converters is required by their clients to ensure defect-free products. As part of this process the manufacturer needs to implement 100% inspection of their catalytic substrates for plugged channels. Original and processed views of the images are as follows:


Solution
The vision component of the solution, designed by engineers at Image Labs International, uses Sherlock integrated with a DALSA PC-DIG frame grabber inside a standard PC, interfacing to a standard digital line scan camera. As the part passes into the inspection area on a conveyor, a beam sensor triggers the beginning of the image.
When the image is complete, the image data is transferred to Sherlock for analysis and the acquisition hardware is prepared to acquire the next part on the conveyor. Sherlock analyzes the image to locate plugs in the part and counts the number of plugged cells in the part.
Crucial
to the success of this project was the ability to create customized
preprocessor and analyzer functions for Sherlock. Extensive processing
was required to differentiate between plugs and the substrate structure.
Sherlock allowed the engineers at Image Labs International to efficiently
build a fully customized inspection application.
A
manufacturer of camshafts needed to verify that all the parts being
packaged for shipment are the correct parts. Parts are difficult
to differentiate by human packagers. Incorrect parts in a shipment
can have disastrous results when installed into an engine.
Solution
The vision component of the solution, designed by engineers at Image Labs International, uses Sherlock integrated with a DALSA frame grabber inside a standard PC interfacing to a standard analog camera. Each part is placed by hand into an inspection fixture as it is packaged for shipment. A sensor in the fixture triggers an inspection.
The
SMART Search tool was used to locate and read the identification
lobes on each part. Parts with incorrect identification lobes were
rejected. The use of Sherlock in this application allowed a robust
application to be built and implemented rapidly and at low cost.
DALSA's Inspect Vision System Drives Scrap Reductions, Greater Efficiency At Polaris