Shannon, Claude E. [Elwood]

A Study of the Deflection Mechanism and Some Results on Rate Finders [Typescript carbon]

[New York, NY]: [Bell Telephone Laboratories, Incorporated] circa Feb 10, 1941. [2], 2-37 leaves + 15 figures on 8 leaves of plates. 10 7/8 x 8 3/8 inches. Carbon typescripts on thin paper with four holes punched at the spine as issued. The plates are cyanotypes, and the sizes are somewhat larger. Each plate is stamped on the verso "Printed Feb 10, 1941, New York." Very Good. Wraps. [28663]


"Report to the National Defense Research Committee, Div., 7-311-M1" (circa April 1941) was the first publication of this paper. (a modern copy is provided for comparison) This item is the original typescript carbon for the report.

"The NDRC supported differential analyzers as computing facilities because their ability to produce firing tables, as well as data for ballistic cams, made them a legitimate part of fire control research. Firing tables, indeed, were numerical solutions to the machine-control problem of aiming a gun; producing and using them required a series of conversions between continuous curves, numerical outputs, and mechanical motions (although not in real time). Indeed, the differential analyzers had a technical similarity to mechanical fire control computers … Early in its program, D-2 sought to apply the academic knowledge about the analyzers, including theories of feedback, to the industrially produced military machines...On Thornton Fry's suggestion, D-2 let a contract, Project 7, to Shannon for "Mathematical Studies Relating to Fire Control." The NDRC asked him to look at Sperry's antiaircraft director and at another designed by the Frankford Arsenal and suggest how to improve the smoothness of tracking. Shannon analyzed the calculating mechanisms of the gun directors, especially their smoothing circuits, as feedback networks. Using stability and transient analysis, Shannon treated both the gun directors and the differential analyzers as dynamic systems. He wrote five papers on the topic. [ including the current paper ] Shannon compared the devices to electric circuits, found 'the use of electrical analogues very useful in understanding these devices,' and devised a circuit-like notation to describe them. He drew on network theory and his own relay algebra to define 'analysis and synthesis' for differential analyzer setups, as though they were electrical circuits. For the first time in a detailed technical study, Shannon made explicit the equivalence that Bush and Ford Instrument had intuited years before the mechanical calculators created at universities and the fire control computers built for the military were analytically similar machines … in 1941 Shannon joined Thornton Fry's mathematics department at Bell labs, thus making an institutional and intellectual link between MIT's differential analyzers and Bell Labs' new work in fire control" (Mindell, pp 289-290)

This paper, "A Study of the Deflection Mechanism and Some Results on Rate Finders," was one of the five papers Shannon wrote for this contract. The other four were: "Theory and Design of Linear Differential Equation Machines,” "A Height Data Smoothing Mechanism,” "Some Experimental Results on the Deflection Mechanism," and "Backlash in Overdamped Systems.”(Mindell, p 386)

PROVENANCE: The personal files of Claude E. Shannon (unmarked). The only example in Shannon's files.

REFERENCES:
Sloane and Wyner, "Claude Elwood Shannon Collected Papers," #7
Mindell, David A., "Feedback, Control, and Computing before Cybernetics : between human and machine," Johns Hopkins University Press: Baltimore and London, 2002

ITEM SOLD