Automated inline inspection of 100% of production is the only way to be certain that nonconforming components are not passed on to downstream operations or to the end user. Implementing automated inline inspection comes with a number of challenges, including the often-uncontrolled factory environment, part fixturing requirements, high production rates and different types of part transfer. 3D sensors provide easy to implement solutions to meet these challenges.
Factory Floor Environment
Most manufacturing lines are subject to ambient temperature changes, dirt, shock and vibration, and changing ambient light conditions. All of these conditions can significantly affect accuracy and reliability of many types of sensors. To survive and operate accurately in the factory environment, look for 3D sensors that are:
Designed and tested by the manufacturer to maintain accuracy over their specified temperature range.
Packaged in sealed enclosures with no cooling vents or fans to ensure that dirt, moisture and other contaminants cannot enter the sensor housing.
Designed and tested by the manufacturer for resistance to vibration
Insensitive to changes in ambient light
The best of today’s 3D smart sensors are built to thrive in the demanding industrial conditions of the factory floor. They have all internal components mounted on a single rigid “spine” for rigidity and thermal stability, are provided in IP67 enclosures, and are shock and vibration tested by the manufacturer.
In high ambient light situations, consider laser based profile sensors. Some sensor manufacturers offer a selection of laser powers to suit specific needs.
Part Fixturing
Traditional metrology requires parts being precisely located in the gaging station. For inline gaging, precise fixturing is often expensive and frequently impossible for high speed continuously moving lines. The solution is found with the smartest 3D sensors, which have built in firmware providing a selection of part tracking options.
The first option is measurement anchoring, used to track the movement of parts within the field of view of the sensor, compensating for variations in the height and position of parts. The movement is calculated as an offset from the position of a measured feature, where the offset is then used to correct the positions of measurement regions of other measurement tools. This ensures that the regions used to measure features are correctly positioned for every part. Several anchors can be created to run in parallel. For example, you could anchor some measurements relative to the left edge of a target at the same time as some other measurements are anchored relative to the right edge of a target.
The second option is part matching, which accommodates measurement of parts in random orientation. At setup, a model or "golden" part is scanned and measurement tools applied. All further parts with similar geometries will automatically have their image rotated to the model part orientation, and measurements and decision-making tools applied. This feature also is useful in robot guidance, picking unoriented parts from a conveyor.
Part Manufacturing Speed
High volume manufacturing lines can often produce thousands of parts per hour, requiring very high sensor speeds to insure 100% measurement of every part produced. Today’s fastest scanning 3D smart sensors provide data at high speed, and buffer, align and deterministically deliver pass/fail results without dropping data during processing and communication.
Part Transfer Mechanisms
In automated manufacturing, part transfer is either incremental (such as a lift and carry) or continuous motion on a conveyor. For incremental transfer, snapshot 3D sensors provide gaging and inspection capability at the rest positions. For continuous motion, 3D laser line profiling sensors generate 3D point clouds of part geometry by taking sequential profiles as the parts move past the sensor. With continuous motion, use of an encoder on the conveyor to trigger the sensor at specific motion intervals is highly recommended, since conveyor speed frequently varies over time. In this case, select a 3D smart sensor that has a direct encoder input to simplify integration.
Inline inspection presents many challenges, which are solved by implementing smart 3D all-in-one sensors.
