Whether you’re machining, molding, stamping or bending, quality control is a vital part of the production cycle. In most facilities, operators send a first-off, last-off or other selected parts to the quiet, climate-controlled lab at the back of the shop for testing. This could mean a trip on the CMM, 3D measuring arm or other process.

This conventional system costs time and is limited in the data it can capture. For example, if your first-off was a go, but later parts returned no-go, it isn’t clear when the problem occurred. With in-line testing, deviation from specifications can be detected before it exceeds tolerances, potentially empowering the production line to avoid making bad parts altogether.

How In-Line Quality Control Inspection Works

There’s a well-known adage in the realm of quality management called the 1:10:100 rule. This rule states that the cost of prevention will be ten times less than the cost of correction and 100 times less than failure. That’s why in-line inspection is effective: even in a worst-case scenario, it can prevent waste by stopping production of bad parts.

The Bluewrist system featured in the above video is called the Preciscan3D Inspection Station. Following what seems to be a recent trend in robotics deployments, the system is available as a plug-and play integrated system, including collaborative robot, laser scanner mounted on a cart, with scanning and analysis software also included.

In the past, quality testing delegated to operators, such as with verniers, height gauges or micrometers was not high-accuracy. Now, portable measuring systems such as the Hexagon Romer arm do not require strictly controlled conditions. According to Bluewrist, the Preciscan3D system is accurate to 0.1mm, and only requires calibration to the cart-mounted monument sphere.

While it’s sold as a package, the software itself is hardware agnostic. In the video, it’s shown with a Universal Robots UR-10 cobot and LMI Gocator 2340 laser scanner. The system also could be configured with a non-collaborative, traditional industrial robot, but the company does not recommend this option. A non-collaborative robot would increase deployment time because of the required fencing or other safety hardware, and the testing would have to take place in a dedicated cell. 

The software used in this system recently passed certification by the National Institute of Standards and Technology (NIST). The test examined the algorithms used to generate the 3D point cloud of the part geometry.

Although the system is intended to reduce downstream defects, it’s easy to see another application: fast inspection of incoming parts from a supplier. For example, if you machine parts before sending them out for an operation such as heat treating or powder coating, this solution could quickly verify the quality of the parts returning to your plant.

If you want to read more about in-line QC, check out Why Choose 3D Laser Scanning over Touch Probes in Manufacturing Quality Control.