Because connectors make reliable and efficient electrical connections between parts of an electronic system, they’re critical to the quality, functionality, and reliability of the final electronic device in which they’re used. This application focuses on connector pin inspection after they’re inserted into the connector housing during the assembly stage.
Inserting pins during assembly, like any other precision manufacturing process, can encounter various defects that affect the quality, functionality, and reliability of the final electronic device in which it’s used. These defects include:
Pins are critical to electronic circuits– any defects can disrupt electrical reliability, cause failures of the electronic devices in which they’re used, shorten the lifespan of those devices, and lead to safety risks like short circuits, overheating, or fires. Manufacturers must prevent these defects to avoid costly recalls and comply with industry standards.
But pin defects can be difficult to detect– they are variable in type and location. Traditional machine vision requires programming hundreds of hand-crafted rules, causing them to fail to detect new or variable defects that don’t match their programmed parameters
Pins are made of metal with low contrast and reflective surfaces, making defects hard to see. Traditional machine vision systems struggle to capture clear images and distinguish between actual defects, reflective surfaces, and the background, ultimately missing defects or causing false rejections..
And in high-volume manufacturing environments, headers must be inspected quickly to keep up with production rates. Traditional machine vision products may fail to keep up with required cycle times.
UnitX’s AI-powered inspection effectively detects inserted pin defects where other solutions fail.
First, the OptiX imaging system illuminates and images the inserted pins. Then, the CorteX Central AI platform is trained on inserted pin defects. Lastly, those AI models are deployed to the CorteX Edge inference system to detect and classify defects in-line.
OptiX provides superior images that minimize reflectivity while maximizing defect visibility. It has 32 independently controllable lighting sources that can be optimized for metal pin surfaces and various defects via software. Its computational imaging capability can be used to take multiple shots and eliminate hotspots caused by highly reflective metal pin surfaces. And its lighting dome design supports a very acute incidence angle of projected light, causing even very tiny defects to cast shadows which increase their visibility.
CorteX accurately detects random, complex defects. It automatically normalizes for variability in positions and orientations and recognizes defects down to the pixel-level. It reduces false positives that lead to scrap and wasted product.
CorteX supports fast AI model development, deployment, and iteration. CorteX AI models are sample efficient– they only require a few images to train on new defect types.
UnitX optimizes yield. In CorteX, can tune quality criteria and visualize the impact on yield before rolling those changes to production. All inspection data is referenceable in one central platform for manufacturers to analyze and identify areas for process improvements.
UnitX provides rapid, 100% inline inspection. OptiX has bright LEDs and fast fly capture speeds of 1m/s for high speed imaging. And CorteX Edge supports high inference speeds (up to 100 MP) to quickly output an OK/NG decision, seamlessly communicating that decision via integration to all major PLC, MES, and FTP systems.
Manufacturers who use UnitX to automate inserted pin inspection are able to:
In this example, we inspected pins inserted into housing for missing and damaged pins.
Imaging
First, we used OptiX to capture images of the inserted pins. We used the software interface to test out different lighting patterns and ultimately settled with the following lighting pattern (interface intentionally blurred):
The software-defined lighting made it easy to experiment with various lighting patterns and identify the best lighting. OptiX allows for multiple lighting configured to be used in-line as well in case different defects are best illuminated by different configurations.
We selected the above lighting pattern and used it to capture images of inserted pins, including both defective and OK parts.
Training
Next, we used CorteX Central to train our models. We created labels for two defects: missing pins and bent pins.
We then labeled those defects in the images we captured from OptiX, using only 9 images of NG parts and 4 images of OK parts.
Because of CorteX’s user-friendly interface and the low number of images it requires to train its AI models, it only took us 24 minutes to complete the labeling for the three defects.
Detection
We then deployed those AI models to CorteX Edge to detect defects on new inserted pins, resulting in the accurate detection and classification of our two defects.
More inspection solution details
