Introduction to the Apollo Guidance Computer

Between 1963 and 1969, Raytheon Corporation churned out approximately 75 rectangular chunks of metal, plastic, and silicon. Bigger than a breadbox but significantly smaller than a UNIVAC, the Apollo Guidance Computer had been designed by the MIT Instrumentation Laboratory to safely guide the Apollo astronauts and spacecraft to the Moon and back.

Instrumentation Laboratory engineers first began designing the Apollo Guidance Computer (AGC) in August 1961. Digital computer design was an exciting and relatively young discipline, while integrated circuits using silicon transistors had only just arrived on the marketplace. The most significant buyer in that marketplace was the federal government, which, besides procuring integrated circuits for explicitly military digital computers, spent millions of dollars to help MIT and Raytheon explore these radical new technologies for peaceful space exploration.

In the early days of digital computing, power, ruggedness, and reliability required groundbreaking (in other words, expensive) technological breakthroughs in both design and manufacture. The designers of the AGC were presented with a number of challenging goals and constraints. Besides the overarching goals of "beating" the Soviets and successfully landing men on the moon, the engineers were given the task of providing a "reasonable" allocation of responsibilities to both man and machine in spacecraft guidance. Furthermore, the computing machine expected to take on these responsibilities had to have sufficient power and interface capability to not only calculate guidance equations but to provide for accurate navigation and control of the spacecraft independently of the Earth-based guidance network.

The engineers designing the AGC worked to achieve these goals constrained by a number of requirements. Since the fuel budget for a trip to the Moon was extremely tight, all spacecraft components had to remain as light as possible and consume as little power as possible. The computer had to be able to withstand the rigors of space flight, including extreme temperature shifts and extraordinary vibration. Above all else, the computer couldn't fail. Astronaut's lives could not be lost due to hardware failures or software bugs. Hundreds of MIT Instrumentation Laboratory engineers worked long hours under extraordinary pressure to design the hardware and software necessary to meet these goals and requirements, all fueled by the enthusiasm generated by the challenging endeavor.

Designing the computer proved to be only half the challenge. Raytheon Corporation, the industrial contractor responsible for production of the working AGC models, relied on a variety of innovations as well as traditional practices to ensure the ruggedness and reliability of the machines. The technology drew on the long-standing industrial infrastructure of New England.  Raytheon, located in a region of eastern Massachusetts famous since the 19th century for its textile industries, took advantage of the skills of laborers from the old technological regime and incorporated them into the new. Former textile workers, who were primarily women, proved invaluable to the Apollo project as they used skills--gained in lived experience--to weave wires into the miniscule washer-like magnetic cores that served as the AGC's erasable memory banks.

But even if their skills were invaluable, employees remained under strict discipline; Raytheon management and MIT engineers strove to keep quality of work as high as possible. Astronauts would occasionally visit production lines to remind the workers that real human lives depended on the quality and consistency of their work. 

Clean room technology, which reached impressive standards during the Apollo project, served both to limit undesirable contamination in the computer manufacturing environment and to impress personnel with a sense of discipline. Machines were brought in to replace humans in situations where the simplest human error could cause a complete computer failure, such as in the wiring of unerasable core rope memory modules. Nothing could be left to chance; neither dirt particles nor undisciplined employees nor human error could be allowed to jeopardize the mission.

By the mid-1970s, when the Apollo project wound down, digital computing and integrated circuitry had become established features of the American technological landscape. Digital computer design and manufacturing was quickly becoming one of the most dynamic industries in the United States; by the time the first AGC successfully guided astronauts to the lunar surface and back in Apollo 11 (1969), private industries had designed and built computers far more powerful and flexible than the AGC (albeit more bulky and less reliable). Clean rooms, workers with esoteric chip manufacturing and design skills, and ever-faster and smaller digital computers quickly became a standard feature of American high technology, forcing analog computers increasingly into the background.

Other significant changes had taken place in American society, politics, and technologies, as well. Most importantly, the Cold War justification for one of the most expensive government-sponsored projects in American history had lost the urgency so central to the national politics of the 1960s. MIT, like other universities and institutions across the nation, came under heavy criticism from students and other anti-war protestors in the late 1960s and early 1970s for its support of defense research. Even the ostensibly peaceful Apollo missions were attacked as covert military operations. Consequently, MIT administrators chose to sever ties with the Instrumentation Laboratory, which became a private institution and acquired its current name, The Charles Stark Draper Laboratory.

Meanwhile, the AGC itself began a new life as a control computer for digital fly-by-wire systems in the Air Force's F-8. The fly-by-wire features of the Apollo spacecraft proved so similar to the needs of high-speed jet aircraft that engineers simply commandeered an unused AGC, rewrote its software, and stuck it into an experimental jet. The success of the F-8 program led to a digital revolution of sorts in fly-by-wire systems for military and commercial aircraft, allowing for greater air speeds and radical redesigns of aircraft bodies (such as "flying wings").

The rest of the production models of the AGC, both the Block I models (for early Apollo test missions) and the Block II models (flown on Apollo missions 5 and 7 through 16), have found their way either to the lunar surface or to Earth-based museums and archives. A number are on display at the Smithsonian's National Air and Space Museum in Washington, DC, where they help to tell a cheering story of American technological achievement and imply that digital control computers with integrated circuitry provided the only logical path to success. As the primary sources listed on this site indicate, however, the design and manufacture of the Apollo Guidance Computer took place in a tangled web of political, social, and technological interactions; only the alignment of particular people with certain ideas and resources in a particular place at a particular historical moment made the AGC take the form it did. Exploring how those particular people, places, and things came together, to open a window on the world of high-technology design and manufacture in Cold War America, is the purpose of this website.

-- S.H./D.M.

site last updated 12-08-2002 by Alexander Brown