First Attempts & Experimental Machines

Sep 14, 2015

Numerical control—the precursor to CNC—was developed by John Parsons as a method of producing integrally stiffened aircraft skins. Parsons, while working at his father’s Traverse City, Michigan-based Parsons Corp., had previously collaborated on the development of a system for producing helicopter rotor blade templates. Using an IBM 602A multiplier to calculate airfoil coordinates, and inputting this data to a Swiss jig borer, it was possible to produce templates from data on punched cards.

Parsons’ work lead to numerous Air Force research projects at the Massachusetts Institute of Technology (MIT) starting in 1949. Following extensive research and development, an experimental milling machine was constructed at MIT’s Servomechanisms Laboratory.

A 28-inch Cincinnati Hydro-Tel vertical-spindle contour milling machine formed the base of the experimental system. Extensive modifications were made, including the removal of all table, cross-slide, and head drives and controls, and the installation of three variable-speed hydraulic transmissions, which were connected to leadscrews. Through gearing and leadscrew, each transmission could produce a 0.0005” motion of the table, head, or cross-slide for each electrical pulse it received from the director.

To ensure that this experimental machine was functioning as directed, a feedback system was added. Synchronous motors geared to each motion generated voltage in response to movement. This voltage was sent back to the detector for comparison to the original command voltage.

By 1953, enough data had been collected through application studies to indicate the practical possibilities of the nascent technology. However, a different, less accurate, experimental NC machine also developed at MIT—one that utilized a Flexowriter, eight-column paper tape, a tape reader, and vacuum-tube electronic control systems—became the true prototype for the future of CNC.