Laser versus vision centring
07 January 2006
Martin Kunz discusses the advantages and disadvantages of component centring using laser or vision systems
The continual miniaturisation of components is placing ever more stringent demands for accuracy on component centring. In addition to the precision of centring, decisive criteria are component recognition time, flexibility and susceptibility to environmental influences.
In modern, automated component assembly systems there are two established basic systems for component centring: laser-based and camera-based. Each system has advantages and disadvantages that have a considerable bearing on the quality, speed and reliability of the assembly process. For both laser and camera systems, the accuracy of placement is a direct function of the resolution and quality of the optics. However, particularly with the camera-based systems, there are numerous application variants, of which the properties differ significantly.
Camera mounted on the machine
With respect to camera mounting, there are three main categories: fixed camera, moving camera, and travelling camera fixed to the placement head. A fixed-camera system is mounted in a fixed position on the machine chassis and the placement head travels over the camera after a component is picked. A moving camera moves along the x-axis in such a way that the placement head can travel to the placement position from the pick position without diversion. If the camera is travelling on the placement head, the components may be measured from the x- and y-axes during the procedure.
Image processing sequence
A second subdivision into systems that measure only one component per image and those that measure multiple components per image allows for comprehensive monitoring. The two graphs show a rough evaluation of the various camera systems with respect to their flexibility and speed. As with most technical systems, the generalisation that a system can be either fast or flexible also applies here. Accuracy of recognition is the direct quotient of the resolution (number of pixels) and the size of the image. From this, it may be deduced that systems that measure several components simultaneously, and so measure all axes with a single large image, are less accurate than systems that measure every component individually. In practice, the various camera systems and configurations are used differently and often combined, depending on the machine manufacturer. For example, a single image is used for all axes for chip placement in order to achieve high speed and individual images are used for the placement of active components. On close examination, however, this precise arrangement can also conceal major disadvantages: e.g. for mounting 0201 components and/or achieving a high packing density the recognition accuracy is no longer sufficient
Laser centering
The images above contrast the images of a 0201 chip obtained using a typical camera with 50-micron resolution and a laser system with 16.7-micron resolution. The higher resolution of the laser allows significantly clearer component recognition and thus also more accurate positioning. This also describes the main advantage of modern laser systems - the resolution being comparatively higher than that of camera systems. A further important advantage is that they are able to measure in all three dimensions (X;Y;Z). This also makes it possible to compensate for variations in the thickness of components. Since the lasers are very compact, the systems are normally mounted on the placement head, so measuring is done during the movement of the x- and y-axes, i.e. in neutral time. Speed is thus a further advantage of laser systems. What can hardly be put into figures is the simpler programming of laser systems and their freedom from susceptibility to environmental interference. In a production environment, however, these characteristics are of great value, and ultimately affect the productivity of a system. The obvious disadvantage of laser systems in relation to camera systems is that only rectangular forms but no flat surfaces can be measured.
In order to exploit the advantages of both systems and to optimise them both for one task, Juki equips their assembly machines with both a special multi-nozzle laser system and a camera system. The patented multi-nozzle laser system measures all four z-axes at the same time, using one laser unit. When placing large components only every second axis will pick up a component which can then be measured using the double width. Since the laser measures the chip components precisely and in neutral time, the focus of the camera system is on flexibility. For camera systems installed permanently on the machine, the FlexLighting system - unique in the industry, with three illumination colours and angles - is particularly notable. This allows for components of any shape and also components with high glaze surfaces to be easily centred and checked. The use of multiple images allows the system to measure components of sizes up to 75x75mm or 50x150mm.
The multi-nozzle laser system is used for chip mounting on all Juki systems. The camera system is used on the new KE-2055R and the KE-2060R assembly machines. The KE-2055R allows components of up to 20x20mm or 26.5x11mm to be measured with the camera and the laser. The KE-2060R is also equipped with a high-precision head, which can be used for placing components of up to 75x75mm or 50x150mm.
The author worked for Juki Automation Systems until last December, when he moved on to working for a machine tool company. Martin has been replaced at Juki by Heinz Schlup.
Contact Details and Archive...
Related Articles...
Most Viewed Articles...