When you buy a new car, you can choose from a wide range of exterior and interior features, transmission and dynamics, and safety bundles, to configure your individual vehicle. From the point of view of engine manufacturing, this entails a wide range of designs and sizes. To meet demands for high quality, the manufacturer must test their engines according to industry-specific guidelines and quality standards. Industrial vision systems make an important contribution to quality assurance and increased efficiency in engine manufacturing. There used to only be a small range of products, whereas nowadays fully automated evaluation of a wide variety of large components and complex geometries is required. A single component may call for as many as 100 features to be inspected in different positions.

Figure 1. Flexible inspection system.

During final assembly, fully automated inspection is performed on engines for the presence of all components, their correct positions, and part types. Various inspection positions at different distances from the engines must be taken into account. The traditional approach would use an individual camera, optics, and illumination for each position so a new feature would require adding a camera and an illumination unit. However, a more flexible approach would be to use NeuroCheck's integrated inspection solution, RoboInspector. The camera to be used for visual inspection, is installed using the mounting flange of a Universal Robots series (UR5/UR10) robot. The control technology of the inspection line consists of RoboControl process control and NeuroCheck image processing. For safety, the operation of the UR robot is force-controlled and intrinsically safe. The camera module is designed with a touch-sensitive extraneous light screen that eliminates edges that could pose a risk of injury.

Simple Programming for Different Variants

Figure 2. RoboInspector with integrated liquid lens and light shield.

RoboInspector provides menu-guided access to “Teach” mode to enable easy programming of new inspection positions or entire inspection variants. Each variant can consist of a different robot position and imaging-evaluation routine. In teach mode, an interactive program guides the user through a custom setup. The various teach positions are adopted into an image processing table for that particular variant. An individual trajectory with any number of inspection and focus positions can then be defined. Defined variants and trajectories are finally merged in a database. Thus the application can be extended at will. In inspection mode, the NeuroCheck software accesses this database and creates the workflow that will be processed step by step for the component to be inspected.

Programming Variable Focus

Figure 3. By controlling the liquid lens, it is possible to focus on the barcode (left) or the PCB (right) without having to move the robot.

With standard fixed focus optics, it is difficult to achieve the proper depth of field for different working distances from a component's surface. Working distances had to be varied mechanically or by changing the focus of a lens using a motor-driven control. However, the required speed, robustness, and mechanical effort presented obstacles in a production setting. To overcome these limitations, NeuroCheck has integrated a variable-focus liquid lens manufactured by Optotune (Switzerland). The lens is controlled by a plug-in from the NeuroCheck visual inspection software.

The liquid lens works in a fashion similar to the human eye — without any moving parts. The lens consists of an optical fluid enclosed by an elastic polymer membrane. The shape of the lens is changed electrically by applying a low voltage electrical current to a voice-coil actuator that pumps liquid from the outside to the central portion of the lens. Within milliseconds, the lens can focus over large working distance ranges, providing an increase in depth of field on the order of 100x for a typical system. Several images can be captured sequentially and very rapidly at various focus levels. If, for example, you change the working distance from 500 to 100 mm, this results in a five times optical zoom.

Controlling Illumination

Figure 4. Working principle of Optotune's focus-variable liquid lens. A voice coil actuator applies a force to a membrane, forcing fluid into the clear aperture of the lens, changing its curvature and increasing optical power.

The RoboInspector's integrated extraneous light shield enables independence from sunlight or surrounding light. The camera module itself consists of a digital camera with an LED ring light featuring individually switchable elements. Electronic control of the optics and illumination can create the appropriate illumination for any inspection situation.

Conclusion

The integrated robot inspection station offers a high degree of flexibility regarding illumination, working distances, and inspection positions. By using a variable-focus liquid lens, complex inspection tasks can be created using dynamic processes with a broad range of variables. Unlike traditional multi-camera inspection systems, the robot inspection station can be extended in a cost-effective manner. When adding a new inspection parameter, all that needs to be done is to teach the new position, rather than to add a new camera. This is considerably less costly. The initial deployment is cost-effective as well and can be economically expanded to adapt to new variants.

This article was written by Doreen Kusche, Marketing Assistant, Neurocheck GmbH (Remsech, Germany). For more information, contact Ms Kusche at This email address is being protected from spambots. You need JavaScript enabled to view it. or visit here .


Photonics & Imaging Technology Magazine

This article first appeared in the September, 2017 issue of Photonics & Imaging Technology Magazine.

Read more articles from this issue here.

Read more articles from the archives here.