Heat Shrinkable Tube Production: Automated Quality Control

Abstract
This project is an excellent example of tech transfer: Heat shrinkable tubing is used in a variety of applications in the electronics, automotive, military and aerospace sectors. One typical application is the electrical and mechanical insulation of cable splices. We developed a camera-based measuring system for an automated control of the cut length in production.

Introduction
In production, the heat shrink tubes are cut to specific lengths from a continuous tube. Especially in the automotive sector, the accuracy demands are very high and the tubes have to meet tolerances not larger than +/- 0.3 mm in length. A camera-based measuring system has been developed for an automated control of the cut length in production.

Implementation
First, a tube is located in an image applying a fast 1D profile analysis. Then, the left and right tube edges are detected simultaneously with subpixel accuracy. Special dynamic tube edge templates guarantee a reliable detection even under the presence of background clutter.

Depending on the production velocity, 2-10 individual measurements are performed on each tube. The total length is computed by averaging, thus reducing the influence of image noise to a minimum. The measuring system consists of an industrial camera, LED light line, light barrier and a powerful standard PC. Requirements are as follows:

Tube lengths 20 – 100 mm
Diameter 4 – 25 mm
Color black, transparent, colored
Up to 2.000 parts/minute Length tolerances up to +/- 0.3 mm

Results
The system allows for a contact-free and fully automatic control of 100% of produced heat shrink tubes according to the given requirements with a measuring precision of +/- 0.03 mm. After the measurement process, defective parts are sorted out using air pressure to blow them aside. Dependent on the color, length and diameter of the shrink tubes, a specificity (true negative rate) of 99,8% to 100% is reached.

Beyond even sorting, the measurements can be used as feedback to control the cutting process, should any deviations occur, thereby eliminating the production of defective parts. Thus, one does not even produce defectives.

The measuring results are displayed on a monitor in real-time in terms of the camera image, diagrams, and statistical data. The continuously adapted values include means value, standard deviation, min./max. length or the number of good pieces. In addition, these values can be stored and documented in a production order file.