top of page
Writer's pictureSascha Lummitsch

Steinbeis Qualitätssicherung und Bildverarbeitung GmbH has developed a new image processing system

Springs - They’re found in almost all areas of our daily lives and ensure mechanical parts move correctly. For example, you’ll find them in cabinets used as tension springs, or used as compression springs behind light switches next to staircases. The average car contains around 8,000 springs. This makes the quality and accuracy of springs particularly important when it comes to production. Our Clusterpartner and imaging specialist Steinbeis Qualitätssicherung und Bildverarbeitung GmbH has developed an image processing system that cuts manufacturing costs and safeguards product quality.


The goal in spring manufacturing is to produce springs in the required quality, to continuously monitor quality, and to improve product quality on an ongoing basis. To safeguard quality in spring production, quality control loops can be introduced. These involve the use of quality control lines, quality control devices, quality reference variables, quality control deviations, and quality control parameters. For example in spring production, quality control reference variables are based on target values and a range of tolerances for the factors that dictate quality. To introduce quality control loops in practical terms, control systems need to be automatic and adaptable. They also have to match different processes and work pieces. Performance adhering to quality requirements thus depends on a combination of product and process quality control loops. One way to control quality in spring production is to use quality control charts.


Quality factors are managed by setting intervention points (UIP, LIP), warning points (UWP, LWP), and a center line (C) on a quality control chart. If spring measurements remain within the intervention and warning points, processes can continue as before. If they fall beyond those limits, which are calculated using distribution methods, systems are automatically interrupted to regulate the process and a search begins for the cause of deviations. If a measurement lies beyond a warning point, it’s important to observe processes more carefully. The effectiveness of controls depends on the position of intervention points on the control chart.


Retrofitting image processing systems to detect spring types

Springs are subject to extremely high quality standards, including spring geometry. This is where the expertise offered by Steinbeis Qualitätssicherung und Bildverarbeitung comes into play. The Steinbeis experts from Ilmenau have developed an image processing system called SpringTest. Their system can be fitted on modern spring coiling machines, delivering 100% spring geometry inspection in production. An essential component of SpringTest is an automatic, in-line “camera-and-lens” measuring unit, which recognizes different types of springs (cylindrical, conical, double conical) in captured images and measures the geometry of springs, such as length and diameter. “The entire system comes with measurement software based on a special image processing system that compares target geometries with tolerances and feeds control deviations into the machine control unit,” explains Managing Director Steffen Lübbecke.


Steinbeis system checks quality attributes

Monitoring schedules should be kept straightforward and safe. At the same time, they should minimize downtime on spring coiling machines. Once machine operators have set up the machine to process a new spring, they just need to press a monitoring schedule button and the unit starts to detect the type of spring. This enables the system to set up a new monitoring schedule based on appropriate parameters – and the machine is ready for production. The software automatically detects the type of spring and its position, determining suitable measuring points for each spring under observation.


Another feature of the test software is that it can automatically track measuring points. Springs involve fast-moving manufacturing processes and as a result, the exact positioning of each spring tends to deviate from the previously produced spring due to vibration. The job of the software is to determine the new position of each spring as a result of that vibration and track measuring points accordingly. This allows up to 900 produced springs to be inspected per minute with regard to geometry such as length or diameter. To do this, measurements are compared with control and tolerance limits. For example, if a spring is too long the software in the image processing system interrupts the process and automatically adjusts the spring length to ensure the next spring is of the correct length again. Tolerance limits stored by the system make it possible to sort springs automatically.


The system includes a special image processing software module for analyzing and measuring different spring geometries. To do this, it compares values for the diameter and length of each spring (which are supplied by the measuring unit with the camera) looking at target and tolerance values. If a spring deviates from given quality control values, i.e. if a measurement lies outside tolerance limits, a command is transmitted to the machine control unit.


Flexibility to adapt to other monitoring requirements

The optical in-line inspection method described here for spring coiling machines has performed excellently in practical application, time after time, and it can also be used for other components. For example, it can also be used to conduct visual inspections on nails, screws, and other bent wire components,” says Managing Director Professor Dr.-Ing. Gerhard Linß. One of the biggest benefits offered by the special image processing system is that it makes it possible to cut the cost of scrapping and reworking parts. With a small number of adjustments, the system can also be transferred to a variety of intermittently and continuously manufactured components, thus offering the Steinbeis Enterprise from Ilmenau a whole host of opportunities to keep developing its solution.


For more information visit our Clusterpartner website.

11 views0 comments

Comments


bottom of page