Anti-Aircraft Fire Control andthe Development of IntegratedSvstems at Sperry, 19251940dDavid A. MindellThe dawn of the electrical age broughtnew types of control systems. Able totransmit data between distributed compo-nents and effect action at a distance, thesesystems employed feedback devices aswell as human beings to close controlloops at every level. By the time theoriesof feedback and stability began to becomepractical for engineers in the 1930s a tra-dition of remote and automatic controlengineering had developed that built dis-tributed control systems with centralizedinformation processors [I]. These twostrands of technology, control theory andcontrol systems, came together to producethe large-scale integrated systems typicalof World War II and after.Elmer Ambrose Sperry (I 860-1930)and the company he founded, the SperryGyroscope Company, led the engineeringof control systems between 1910 and1940. Sperry and his engineers built dis-tributed data transmission systems thatlaid the foundations of today’s commandand control systems. Sperry’s fire controlsystems included more than governors orstabilizers; they consisted of distributedsensors, data transmitters, central proces-sors, and outputs that drove machinery.This article tells the story of Sperry’sinvolvement in anti-aircraft fire controlbetween the world wars and shows howan industrial firm conceived of controlsystems before the common use of controltheory. In the 1930s the task of fire controlbecame progressively more automated, asSperry engineers gradually replaced hu-man operators with automatic devices.Feedback, human interface, and systemintegration posed challenging problemsfor fire control engineers during this pe-108riod. By the end of the decade these prob-lems would become critical as the countrystruggled to build up its technology tomeet the demands of an impending war.Anti-Aircraft Artillery FireControlBefore World War I, developments inship design, guns, and armor drove theneed for improved fire control on Navyships [2]. By 1920, similar forces were atwork in the air: wartime experiences andpostwar developments in aerial bombingcreated the need for sophisticated fire con-trol for anti-aircraft artillery. Shooting anairplane out of the sky is essentially aproblem of “leading” the target. As air-craft developed rapidly in the twenties,their increased speed and altitude rapidlypushed the task of computing the lead outof the range of human reaction and calcu-lation. Fire control equipment for anti-air-craft guns was a means of technologicallyaiding human operators to accomplish atask beyond their natural capabilitiesDuring the first world war, anti-aircraftfire control had undergone some prelimi-nary development. Elmer Sperry, as chair-man of the Aviation Committee of theNaval Consulting Board, developed twoinstruments for this problem: a goniome-ter, a range-finder, and a pretelemeter, afire director or calculator. Neither, how-ever, was widely used in the field [3].When the war ended in I 9 18 the Armyundertook virtually no new developmentin anti-aircraft fire control for five to sevenyears. In the mid-1920s however, theArmy began to develop individual compo-nents for anti-aircraft equipment includ-ing stereoscopic height-finders,searchlights, and sound location equip-ment. The Sperry Company was involvedin the latter two efforts. About this timeMaj. Thomas Wilson, at the FrankfordArsenal in Philadelphia, began develop-ing a central computer for fire control data,loosely based on the system of “directorfiring” that had developed in naval gun-nery. Wilson’s device resembled earlierfire control calculators, accepting data asinput from sensing components, perform-ing calculations to predict the future loca-tion of the target, and producing directioninformation to the guns.Integration and DataTransmissionStill, the components of an anti-aircraftbattery remained independent, tied to-gether only by telephone. As Preston R.Bassett, chief engineer and later presidentof the Sperry Company, recalled, “nosooner, however, did the components getto the point of functioning satisfactorilywithin themselves, than the problem ofproperly transmitting the informationfrom one to the other came to be of primeimportance.” [4] Tactical and terrain con-siderations often required that differentfire control elements be separated by up toseveral hundred feet. Observers tele-phoned their data to an officer, who manu-ally entered it into the central computer,read off the results, and telephoned themto the gun installations. This communica-tion system introduced both a time delayand the opportunity for error. The compo-nents needed tighter integration, and sucha system required automatic data commu-nications.IEEE Control SystemsIn the 1920s the Sperry GyroscopeCompany led the field in data communi-cations. Its experience came from ElmerSperry’s most successful invention, a true-north-seeking gyro for ships. A significantfeature of the Sperry Gyrocompass was itsability to transmit heading data from asingle central gyro to repeaters located ata number of locations around the ship. Therepeaters, essentially follow-up servos,connected to another follow-up, whichtracked the motion of the gyro withoutinterference. These data transmitters hadattracted the interest of the Navy, whichneeded a stable heading reference and asystem of data communication for its ownfire control problems. In 1916, Sperrybuilt a fire control system for the Navywhich, although it placed minimal empha-sis on automatic computing, was a sophis-ticated distributed data system. By 1920Sperry had installed these systems on anumber of US. battleships [5].Because of the Sperry Company’s ex-perience with fire control in the Navy, aswell as Elmer Sperry’s earlier work withthe goniometer and the pretelemeter, theArmy approached the company for helpwith data transmission for anti-aircraft firecontrol. To Elmer Sperry, it looked like aneasy problem: the calculations resembledthose in a naval application, but the physi-cal platform, unlike a ship at sea, anchoredto the ground. Sperry engineers visitedWilson at the Frankford Arsenal in 1925,and Elmer Sperry followed up with a letterexpressing his interest in working on theproblem. He stressed his company’s expe-rience with naval problems, as well as itsrecent developments in bombsights,“work from the other end of the proposi-tion.” Bombsights had to incorporate nu-merous parameters of wind,