Industrial robots in the plastics and electronics industries

Industrial Robot

ISSN: 0143-991x

Article publication date: 1 August 2004




Soska, G.V. (2004), "Industrial robots in the plastics and electronics industries", Industrial Robot, Vol. 31 No. 4.



Emerald Group Publishing Limited

Copyright © 2004, Emerald Group Publishing Limited

Industrial robots in the plastics and electronics industries

Industrial robots in the plastics and electronics industries

Keywords: Robots, Plastics, Electronics

Geary V. Soskais based at the Robotics Applications Consult Inc., North Canton, Ohio, USA.

The industrial robot is no stranger to the plastics and electronics industries. Industrial robots have been employed in a wide range of applications in these industries for many years. From unloading plastic injection molding machines to perform surface mounting of components on printed circuit boards, industrial robots have proven their ability to increase productivity and improve overall product quality, benefiting both the manufacturer and the end user.

What is really exciting is that industrial robots have only begun to scratch the surface of potential applications in the plastics and electronics industries. It is not that industrial robots have by themselves become more capable, but rather it is enabling technologies that have opened doors in these industries that were previously closed.

Enabling technologies are those technologies that allow industrial robots to do things that we previously thought either impractical or sometimes impossible. For example, over the past decade vision system technologies have made tremendous gains in both reliability and capability. Not that camera technology has made quantum leaps, but high-speed microprocessor technology has enabled signal-processing time to approach the real-time signal processing time of the human brain. In the not so distant past, a printed circuit had to be precisely fixtured in surface mounting applications to ensure the robot could place components into sockets without damaging the leads. That meant a significant investment in tooling which equated to added production system costs. With today's vision system technology, all that is required is that a printed circuit board be in relative close proximity of the robot. With its interface to the vision system, the robot can locate the socket and properly mount the component without damaging the leads. The obvious benefit is an instant reduction in tooling costs, as vision systems for 2D locating are relatively inexpensive. The same can be said for a robot system being used to pick electronic components from a conveyor and transfer them either to another location or a packaging station. Rather than invest in tooling that ensures each component is always in the exact physical location on the conveyor for the robot to pick up, a vision system will permit the robot to locate a component wherever it is on the conveyor. Couple that with line tracking and the robot can locate and grasp the component on the fly.

In the plastics industry the same enabling technologies come into play. For example, suppose a company is injection molding plastic boxes that will later be used to package tools that will be sold in a hardware retail store. Typically, when something is injection molded there are either sprues or flash that needs to be removed before the box is suitable for use. Years back it would take a lot of programming time to enter point after point of locations on the box for a robot to trim off the sprues or the flash. Today, and again due to ultra-high speed microprocessor technology, robots can use vision technology to locate the sprues or flash and guide the robot through the task of removing them. Once again, in almost human real-time and with a lot less programming.

An important concept to understand here is “almost human real-time”. While today's microprocessor are extremely fast, they still lag far behind in the signal processing time of the human brain. Granted you can capture data in real-time, but you need to process that data, calculate dimensional or position errors, compute robot trajectories and send commands to the robot to react. Unfortunately, the microprocessors needed to perform those tasks as fast as the human brain have yet to be invented. However, the microprocessor industry is making giant leaps almost on a daily basis. If you doubt that, consider that the personal computer you purchase today is almost obsolete by the time you get it home. Even if it is state-of-the-art, it will be obsolete within 18 months.

One other application to consider in the plastics and electronics industries is inspection. Inspection is still pretty much a human application as only humans can perform judgement by weighing certain criteria in real time. Robotic systems, even those equipped with vision technology, will only perform within specified parameters. If any variations outside of those parameters occur, the robot has no idea what to do. In most applications, when something unexpected happens, the robot simply stops and sends out a signal that something is wrong. It is now up to human being to intervene and determine what the problem is and correct it so that the robot can resume its task.

Industrial robots have come a long way over the past 10 years and from a mechanical standpoint not much is going to change in the future. What is going to change is robot control systems in terms of microprocessor technology. Computer chips with processing time in the picosecond range are not far off, and when they do become commercially available we may just reach a point where the only limitations to applying robots in the plastics and electronics industry will be those of our own imaginations. Because by then, we will have reached the real-time signal processing capabilities of the human brain.

Geary Soska is President of Robotic Applications Consulting in the US and the 1991 recipient of the Golden Robot Award. For further details contact: E-mail:

Related articles