This is a great article on the history of machine tools and machine work. But it is also a history of the growth of early america and the United States. A Thank You needs to go, first to the author-unknown, to American Machinist for originally publishing this article, also to The Wayback Machine for archiving it, and to Pedro for finding it at The Wayback Machine and emailing me. My position here is that this article needs to be available for interested machinists and others to read, I don't think either of the other sites will be doing that. I gladly make it available while making no benefit from it -Note: The ads at the top of my pages are Angelfire's not mine. Enjoy, Pat McGuirk
Former link to the article at the American Machinist site. http://www.americanmachinist.com/library/features/aug96/lathes.html
Article was originally printed: American Machinist, August 1996
Microprocessor magic
Numerical control is a means of controlling machine tools that has received an inordinate amount of attention during the past 30 years. But there are ramifications of numerical control that are of critical significance. These involve disciplines and areas well beyond the realm of straight machine-tool operation.
Numerical control is a means of communication between people and machines. NC techniques form the basis for total computer-integrated manufacturing.
It was shortly after World War II, when computers were still in their infancy, that John T. Parsons envisioned the use of mathematical data to actuate a machine tool.
In June 1949, the U.S. Air Force funded a program to develop a mathematic or numerical control system for machine tools. This led to an electronic control system.
The first commercial production-based NC unit was built by Bendix Corp. It was produced in 1954 for machine tools introduced in 1955. In 1960, the first controller with transistor technology was introduced. These systems were able to control machines with three, four, and five axes and had new features such as circular and parabolic interpolation, cutter compensation and dial input.
Integrated circuits (ICs) came in 1967. These permitted a 90% reduction in the number of components, as well as an 80% reduction in writing. These systems were much more reliable.
The combination of lower control cost, simpler operation, and reduced programming complexity really opened up the market. Controls builders began to provide only the hardware, while software-oriented firms started to provide new programming languages and post-processors.
With more data processing being done within the control, it was inevitable that NC systems were heading for more real-time computing power. The NC unit became more than mere a digitally controlled, automatic servosystem. It resembled a rather complex hybrid (digital/analog) computer. Fourth generation controls were identified by either the use of an integral computer, for computer NC (CNC), or extensive use of memory technology.
The NC unit was rapidly becoming a machine-control unit. It was assuming a greater responsibility in the man-machine and control interface.
The ascension of the computer in controlling manufacturing gradually eroded the role of the individual machine tool as a major controlling factor in manufacturing economics. The computer forced on-going reassessments of machine design and application.
Computerizing the systemWith mass production came the development of automation. In turn, with increased automation, came increasingly specialized machinery for manufacturing processes that depend on the availability of unlimited markets and long model runs. But the day of black automobiles and white refrigerators was long over. The name of the game became product diversification and fast response to the changing needs of the marketplace. Mass production, as it was known, was not compatible with these demands. The limits of expansion of mass-production-type manufacturing essentially were reached and, under the influence of socioeconomic pressures, batch-type manufacturing was on the increase. But, while the most significant tool of the future (and the present) is the computer, the most significant task ahead is to understand and adequately define the manufacturing process.
Manufacturing has often been characterized as organized chaos. Manufacturing control has been defined as "simply a matter of solving one crisis after another with the most expedient solution available at the time."
The computer in manufacturing will be the one ubiquitous tool that will help bring order out of this chaos.
The organization - the infrastructure, if you will - and the modes of communication among people, processes, and individual machines took on an entirely new dimensionÑreal time. Making current information available, whenever and wherever it is needed, whether for control or for decision-making, is the most significant underlying principle of manufacturing management.
With the advent of the low-cost micro- and minicomputers and the development of standard communication software, such as Digital Equipment Corp.'s DECNET, distributed computer networks formed the basis for computer control in manufacturing.
In the network approach, minicomputers or stand-alone processors are programmed for controlling process units on the factory floor. As they control the process, they also transmit information on its status to higher levels of computers. The higher-level computers reduce the data to management-information format, enabling management to determine the actual operating status of the plant, what process units are down and why, and how long it took the maintenance personnel to bring the failed unit back on line.
Another important element in the distributed network is the growing number of applications for programmable logic controllers (PLCs). Again, the magic of large-scale integrated circuitry in the form of microprocessors has changed the PLC from a mere relay replacement to an intelligent industrial controller. Smart PLCs are tied to computers, and the PLCs themselves talk to other processors. One attraction of programmable control lies in the fact that it brings computer technology to the people controlling the factory in a language with which they are familiar.
Computers have become an integral part of manufacturing systems. As the cost of integrated circuitry declines, information, computation, and control becomes a cheap commodity.
General-purpose machine tools possess microprocessor-based diagnostic systems that monitor the machine tool's 'health.' Wear in the ways, backlash in the leadscrews, and other types of degradation of the machineÕs performance are continuously and automatically monitored, reported, and, where possible, compensated for.
Because there is really only one way to assure quality - by doing it right in the first place - the trend has been to eliminate inspection by controlling quality during the manufacturing process. Inspection is like a watchdog. It keeps out undesirable elements, but it does not, in itself, prevent the fault from occurring. The more you inspect, the more faults you may find, but thatÕs really closing the barn door after the horses have left. Quality control, in its ultimate conception, assures that every component and end product meets clearly established and well-defined criteria, no more and no less. Hence, the move from inspection to quality control and now toward quality assurance.
Quality assurance is, in a way, a predictive process. To be effective, it requires assessment, monitoring, and control. Every process in the manufacturing cycle of a product must first be assessed to determine its potential for meeting previously established quality requirements. Once the process is selected, it must be monitored to make sure that the assessment of its potential is still correct. Finally, the process must be controlled on the basis of such monitoring to keep it under control. Keeping the process under control will increasingly become the watchword as flexible automation, computer-aided manufacturing, and integrated manufacturing become more prevalent.
No doubt about it, just as the computer represents the manufacturing "tool" with the greatest potential for improving productivity, the resulting integration of the manufacturing process will require increasing automation of the quality-assurance function.
