Transistors to Chips

As often happens with a large project lasting several years, especially an electronic one, the basic technology for the Apollo computer was hardly static. Transistors, invented in 1948, had become commercially available and accepted in the 1950s, especially for military equipment. Integrated circuits (ICs) could put several transistors on a single semiconductor chip, doubling or tripling the package density. IL electronics guru Eldon Hall had incorporated an early IC into the Polaris Mark II computer and thought integrated circuits might have some application to the new project—especially since reduced size and weight would be such high priorities. The Minuteman ICBM was using integrated circuits with strict requirements for reliability, giving Fairchild credibility with military and aerospace systems. Fairchild Semiconductor put its ''MicroLogic'' family of ICs on the market in 1961, followed soon thereafter by Texas Instruments and Signetics.2

The Mod 3C prototype used a circuit design inherited from the Mars computer called ''core-transistor logic.'' Familiar, proven, conservative, and made out of discrete transistors, it seemed just the sort of circuitry on which to bet a multibillion-dollar spaceflight. Yet in 1962 Hall asked David Hanley, a young engineer recently come to the IL from the air force, to purchase a large supply of Fairchild's MicroLogic ICs and to look into designing a computer with them. Hall bet him he couldn't build a version of the machine with ICs that would be faster than the discrete version. Hanley began to build an exact copy of the Mod 3C with the MicroLogic devices, but found that a basic redesign would be easier and more efficient. So he set about designing and building a MicroLogic-based computer that would run the same instructions as the Mod 3C, and named it AGC for ''Apollo guidance computer.'' The name would stick. Logic designers Ray Alonso, Hugh Blair-Smith, and Albert Hopkins saw the redesign as ''an unusual second chance'' to tweak their work.3 Hanley won his bet with Hall. The AGC model began working in early 1963, occupying less space and running faster than the transistor version.

Late one day in 1963, engineer Herb Thaler was working on the new computer with the integrated circuits in the IL's large computer lab. ''I was debugging it when the word came that Kennedy was shot,'' he said. Thaler remembered the shock, and how afterward people on the project became rededicated to going to the moon as Kennedy's legacy.

Still, for all the impressive engineering, a model AGC working in a laboratory would not be sufficient to convince NASA. Hall recognized that his idea for using MicroLogic would require advocacy and persuasion. Hall based his arguments on three factors: (1) the speed of the MicroLogic computer would be more than twice as fast, (2) size and weight would be cut in half, and (3) costs would be comparable to the traditional approach. The integrated-circuit version of the computer, however, would consume nearly double the electrical power of the discrete version (about fifty watts, or the equivalent of a small light bulb).4 Still, the integrated circuits allowed a simpler, more elegant, and more modular design. One type of circuit, a two-input NOR gate, would repeat throughout the entire computer, rather than thirty-four different modules, as would be the case with the core-transistor logic.

While the benefits to integrated circuits seemed clear, the risks were uncertain. It was new technology. Only one company, Fairchild, made the required chips. Would they remain available for the entire length of the project, or were they just another passing fad in electronics? (As it happened, Fairchild did stop making the chips, and the IL ended up buying them from Philco.) Would Apollo be a sufficiently large customer to encourage the suppliers to stay in the business? To increase the demand, Hall proposed that all the ground test equipment for the computer be built with the MicroLogic circuits as well (sometimes chips that had failed the rigorous tests for flight hardware).

NASA was convinced, and in November 1962 allowed the integrated-circuit design to go forward.5 Integrated circuits were not limited to the computer itself, but also covered the analog circuits that brought data into the digital world. Hall soon added to the computer not only the digital NOR-gate IC, but also a new analog IC for a sense amplifier—to condition the analog signals from the spacecraft's numerous sensors.

Hall's relative ease at convincing NASA to accept integrated circuits did not mean that the decision was straightforward, or that the use of integrated circuits was ''obvious'' in any way. For no other major Apollo subsystem used the new technology to the same extent—not the computer that IBM built to stabilize the Saturn, not the radio or radar electronics, and not the LM control electronics outside of the computer. NASA would not allow integrated circuits in these systems, and at least one contractor envied the IL's ability to use the MicroLogic designs, having been turned down by NASA.6 The agency's approval to use them in the AGC, then, was a testament to Eldon Hall's competence in electronics, his confidence in his own analysis, and his skill at presenting the case and persuading his sponsor.

The AGC gave a boost to the fledgling IC business. The Minuteman missile program also used integrated circuits, but over twenty types of custom designed chips. Apollo, by contrast, used a single, standard type of chip. By 1963 Apollo was consuming 60 percent of the integrated circuit production in the United States.7 In 1964 Robert Noyce of Fairchild stated that the company had had shipped 110,000 ICs for Apollo.8 Noyce cited the Apollo computer as evidence of the reliability of the ICs he manufactured, and of their suitability for military electronics.9

The Block I AGC took shape during 1963. The IL created a formal organization for the project, under David Hoag as project director. Reporting to him were the Mission Development group under Battin, Digital Computation group under Alex Kosmala, and Digital Development group under Eldon Hall. There was still no group devoted to software.10 For the following three years, the bulk of the IL's manpower was devoted to hardware development.

Apollo's guidance and navigation team brought their system to the public in September 1963. A press conference at MIT—led by David Gilbert of NASA and including Trageser and Hoag of the IL, Ralph Ragan of Raytheon, and project directors from Kollsman, Sperry, and AC SparkPlug—reported the design as ''virtually complete'' and ''within spitting distance of meeting our schedule objectives.'' Hoag presented the system itself, with the computer at its core, and described the navigation and sighting techniques. Ragan described the manufacturing operations in Bedford, Sudbury, and Waltham, all in Massachusetts. When asked about what an astronaut would do during the mission, Hoag replied: ''He has to align the inertial system, set in the right sub-routine to the computer, and then let it go.''11

Telescopes Mastery

Telescopes Mastery

Through this ebook, you are going to learn what you will need to know all about the telescopes that can provide a fun and rewarding hobby for you and your family!

Get My Free Ebook


Post a comment