POPP Clamp Inspection
Ensure all spring band clamps are properly placed and attached
Related Products
Powered by In-Sight ViDi Deep Learning-Based Vision Software
A breakthrough in complex inspection, part location, classification, and OCR
POPP (Pre-Opened Pre-Positioned) clamps, or spring band clamps, are the most common hose clamp type in automobiles. They need no adjustment after installation but must be positioned precisely and then sprung completely for optimal function.
POPP clamps hold automotive hoses onto their fittings. These hoses carry fluids or gas to and from the radiator, heating core, brakes, and other systems. The fluids can be hot, corrosive, or toxic. Poorly clamped hoses that come loose from their fittings are common causes of vehicle malfunction and warranty claims.
POPP clamps can be in a variety of locations and orientations on hoses of various colors and be viewed from various angles. Both clamps and hoses come from a range of suppliers. Clamps that are positioned too high or low on the fitting, are incorrectly sprung, or have broken tabs, must be detected before the next manufacturing step. Conventional machine vision defect detection has difficulty with the wide range of possible clamp positions, orientations, and colors.
Cognex Deep Learning learns from image sets a range of both good and bad spring band clamp installations. The classification tool quickly determines if the clamp is functional, or flags potential problems that need to be corrected before the vehicle can proceed to further assembly.
If desired for process improvement, a more detailed application of the classification tool can distinguish between installation errors. For example, it can identify that a POPP clamp is gradually being placed farther down its hose, indicating a potential problem with the installation device that can be addressed before any actual defect occurs.
Cognex Deep Learning is quick to set up and can easily be retrained if clamp designs are altered, proper placement changes, or fittings move position.
