Read an Excerpt
Into the BlackJPL and the American Space Program, 1976-2004
By Peter J. Westwick
Yale University PressCopyright © 2007 Yale University
All right reserved.
Chapter OneThe Inheritance
THE JET PROPULSION LABORATORY (JPL) STARTED AS A GRADUATE-STUDENT rocket project at Caltech in the 1930s. At the time Caltech was already a center for science and engineering in the United States, a position it would occupy for the rest of the century. In 1930 Caltech lured Theodore von Kármán, a leading authority on aerodynamics, to become director of its Guggenheim Aeronautical Laboratory, or GALCIT. Von Kármán's lab at first studied airplane flight until a graduate student, Frank Malina, proposed thesis work on rockets. Malina banded together with two other rocket enthusiasts, John Parsons and Ed Forman; they fired their first rocket motor in fall 1936 at an isolated spot in the Arroyo Seco, a dry wash three miles above the Rose Bowl in Pasadena. The wisdom of using a remote site was confirmed when subsequent tests on campus misfired, one explosively. By spring of 1938 Malina and his group had a rocket that ran on the test stand for over a minute. The rocket work piqued military interest, and in January 1939 the National Academy of Sciences began funding GALCIT for work on rocket-assistedtakeoff for airplanes. The following year the Army Air Corps took over, and the expanding program soon shifted for good to the Arroyo Seco site.
Arsenal for the Army
The onset of World War II brought big budgets and secrecy to the rocketeers. It also led to the formal establishment of JPL. Following intelligence reports in 1943 of German rocket development, von Kármán, Malina, and Hsue-shen Tsien, a Chinese mathematician performing theoretical analyses for GALCIT, proposed a long-range program. The army responded with enthusiasm, although Caltech's trustees approved the contract only for the duration of the war. The arrangement nevertheless illustrated a basic watershed in the history of American science and technology: the wartime use of research contracts to enlist academic science in service for the federal government, particularly the military. In this case the army paid for new facilities and operating expenses, while Caltech contributed its administration, faculty, and graduates, as well as its name, to the enterprise. The army thus obtained access to expertise outside its ranks, lab staff got a technical challenge and the resources to pursue it, and both scientists and university administrators earned a patriotic sense of contributing to the war effort. The campus also received a fixed fee on top of the operating budget, a more concrete and compelling inducement that aided recovery from the Depression. The Jet Propulsion Laboratory officially opened on 1 July 1944, the new name shedding the speculative stigma of rockets. When von Kármán's increasing work for the air force took him to Washington later that year, Malina stepped in as director. JPL did not demobilize at the end of the war, although that required accommodations with campus. Caltech had agreed to only a wartime project, and many campus faculty wanted to return to peacetime research. Caltech barred classified research on campus and required military contracts to involve fundamental research instead of strictly applications. Malina and von Kármán suggested that Caltech set up its own rocket laboratory for unclassified, basic research on rockets for scientific use. Caltech's trustees instead decided to continue the existing arrangement, after the army assured them that JPL could focus on basic, unclassified research. Thus, like other large wartime labs, JPL survived through the postwar flux and was ready for service at the onset of the cold war. The decision to continue JPL as a cold war military lab had its costs. Domestic anticommunism perhaps encouraged Malina to resign as director in 1946, as he had moved in left-leaning circles in the 1930s and come under the suspicion of the FBI. It certainly cost the lab the later services of Tsien; when he sought to return to Maoist China, the federal government detained him in the United States and barred him from classified material and thus from JPL. Military work had programmatic consequences as well. The army had pushed for a broad program on guided missiles, and the lab began to acquire expertise in electronics as well as in aerodynamics and propulsion. To highlight the shift, in 1954 William Pickering, an electrical engineer from Caltech, assumed the directorship of JPL, replacing Louis Dunn, Malina's successor.
Pickering's low-key geniality and informality belied the increasing organization of the lab. From the war through the late 1950s JPL designed and built a series of larger and longer-range missiles, designated by military rank: from the Private to Corporal and finally Sergeant. The work initially entailed much basic research in chemistry, physics, aerodynamics, and electronics; but the army wanted an operational weapon in the end, and the national emergency of the early 1950s-the Soviet atomic bomb, war in Korea, and development of tactical nuclear weapons-increased the urgency for a tactical missile. In 1950 the army asked JPL to weaponize the Corporal, primarily to carry nuclear warheads, and over the next several years JPL moved from research into development and then production functions, and even into training troops in use of the weapon. Corporal started JPL's transition from a small, unclassified, academic research outfit to a large, secret, development organization. The transition intensifed in 1954 when JPL undertook Sergeant, which would use solid instead of liquid propellants. An ad hoc, academic design process and loose organization had proved insufficient to handle the many problems on Corporal, including component failure, integration of components into subsystems and systems, oversight of contractors, and operation and training. The solution was managerial, not technological. The Sergeant managers-Robert Parks and his deputy, Jack James-included reliability, testing, and maintenance factors in the component design process, standardized the test and safety procedures, and, a crucial step, insisted on a progressive design freeze, with documented control of all changes. These procedures enabled JPL to develop Sergeant largely on schedule. From a longer view, they represent the initial steps toward the techniques of systems engineering. By 1953 JPL had more than 1,000 staff and a budget of $11 million. Despite the army's assurances, most of the work was secret: by 1958 almost two-thirds of lab publications were classified. The increasing secrecy, formality, and production nature of the work weakened links with campus, but Caltech administrators and trustees rebuffed suggestions to transfer the lab to another contractor or to the army itself.
Onward and Upward with NASA
Von Kármán had initially intended to extend the missile series up through colonel, "the highest rank that works." But Pickering chafed under the pressure of developing weapons systems, and he and Caltech president Lee DuBridge determined not to go beyond Sergeant. The lab instead looked outward, and upward, for new opportunity. JPL rocketeers had always kept an eye on space as a destination for their hardware. In 1949 they reached it, with a version of the Corporal launched on top of a V-2 rocket to a height of 250 miles. In 1955 JPL renewed this collaboration with expropriated German rocket scientists at the Army Ballistic Missile Agency (ABMA), for whom JPL developed a radio-guidance system and reentry vehicle for an intermediate-range ballistic missile. This work led to a tracking system that could detect very faint radio signals thousands of kilometers away and to a proto-satellite vehicle.
The federal government was meanwhile prosecuting the crash program for an intercontinental ballistic missile and beginning to appreciate the appeal of space for international prestige as well as military uses. The United States declared its intent to launch a satellite as part of the International Geophysical Year in 1957-58, but President Eisenhower insisted on a civilian, science-oriented precedent and thus sank a collaborative proposal from the army's labs at JPL and ABMA. Then, on 4 October 1957, the Soviets launched Sputnik. When the hurried American response failed dismally on the launch pad, JPL and the army got the green light to enter the space race. JPL's tracking system and reentry vehicle earned it the right to build the satellite, known as Explorer 1. The triumphant launch of Explorer on 31 January 1958 propelled JPL into the public eye and also into a leading role in the nation's space program. JPL followed with more Explorers and two Pioneers, the last of which aimed for the moon and signaled JPL's intent to push beyond earth orbit. Meanwhile, after much debate, Eisenhower and Congress in mid-1958 created the National Aeronautics and Space Administration (NASA). NASA coveted JPL's space expertise, and on 1 January 1959 the lab transferred to the new agency. JPL would thence have to negotiate its role amidst the often overlapping missions of other NASA centers. NASA assigned JPL responsibility for automated spacecraft for lunar and planetary exploration, which solidified the shift away from its titular interest in propulsion. In the heady days of the early space race, JPL planners laid out a series of flights to the moon, Venus, and Mars, culminating in a manned flight around Mars and back in 1965. They worried that this program lacked ambition. NASA instead accepted a more measured plan for three main flight series: first, reconnaissance flights to the moon known as Ranger; then Surveyor, to soft-land a spacecraft on the moon; and, concurrently, Mariner probes to Venus and Mars. But even this scaled-back program would push JPL to the breaking point and beyond, and force the lab to forge a new regime.
The shift from rockets to spacecraft was not just a matter of mastering new technical fields. JPL engineers went from developing production-line weapon systems, with dozens or hundreds of test flights, to designing custom-built spacecraft that were too elaborate and expensive to test in flight. The Corporal and Sergeant programs in the 1950s had impelled the first steps toward systems engineering, but in the rush of the early space race JPL managers had dispensed with formal methods in favor of quick results, firing off spacecraft until project engineers learned how to make one fly.
JPL learned the hard way that space missions might not resemble production-line missiles. From 1961 through 1962 the Ranger program suffered a series of mishaps. The launch vehicles failed for Rangers 1 and 2. Ranger 3 survived launch vehicle problems only to be bitten by a bug in JPL's flight software: a single reversed sign altered the trajectory, sending the craft in a direction opposite to that intended. Ranger 4 flew the correct path, but its communications failed and the spacecraft sailed silently, dumbly, into its perfect impact on the moon. After Ranger 5 missed the moon altogether NASA called a halt. It was not just the cost the nation could not tolerate, but also the embarrassment. The space race put a premium on quick results, but above all on results alone, and each failure undermined international perceptions of American prestige.
NASA's failure review board traced the problem to JPL's "shoot and hope" approach. JPL needed to test components and verify systems on the ground beforehand to ensure that spacecraft would work right the first time. To correct what NASA called "a loose anarchistic approach to project management," Pickering assigned new managers to Ranger, bringing in Bob Parks, the former Sergeant manager, as head of the lunar program and Harris "Bud" Schurmeier as Ranger project manager. Parks and Schurmeier applied the rigorous methods of systems engineering, including a formal design review and failure reporting system. The most important management technique involved design freezes and change control: at particular stages the project manager froze the design of a component, allowing modifications only with his written approval. The project manager thus kept individual engineers and scientists from pursuing indefinite improvements at the expense of the overall schedule, budget, and reliability. All of this relied on formal documentation to record, report, and enforce management decisions. JPL thus helped originate the discipline, in both senses of the word, of systems engineering.
The crucible of Ranger also forged a new organizational structure. Through the 1950s Pickering had favored a functional organization, with staff divided among several technical divisions akin to disciplinary academic departments. Such a structure served well while JPL worked mainly on one large project at a time, but as the lab entered the space program and undertook a number of concurrent projects, it adopted a matrix organization. The matrix overlaid the technical divisions with a number of small, temporary project offices. Almost all the permanent staff resided in the technical divisions, which each project would draw on as necessary. The matrix thus allowed technical people to float from project to project while providing them a permanent home in the organization. In effect, the project offices controlled money and the technical divisions controlled people. The original matrix, however, was weak; the technical divisions kept most of their authority and left little to the project offices, which hired technical staff subject to the whim of the technical managers. After the failures Pickering gave project managers authority over all staff assigned to their project. Although Ranger may have been on the brink of success in its original mode, its subsequent results cemented the foundation of project management. The 1960s may have been the heyday of the technological fix-the tendency of American society to seek solutions to almost any problem through wonderful new technologies-but Ranger, though itself a technological tour de force, represented instead a managerial fix. Jack James was meanwhile already applying the formal approach to Mariner in parallel to Ranger, including failure reporting, design freezes, and change control. Although the launch vehicle failed on the first Mariner, doomed by the omission of a single hyphen in the guidance equations, Mariner 2 flew flawlessly to Venus in late 1962. After Mariner 3 failed, Mariner 4 in 1964 returned the first close-up pictures of Mars and showed up the Soviets after their five failures to reach the red planet. Subsequent Mariner flights-to Venus in 1967, Mars in 1969, to orbit Mars in 1971, and to Venus and Mercury in 1973-demonstrated JPL's mastery of high-reliability spacecraft and built up a cadre of experienced project managers.
There was another, less publicized product of the early space missions, albeit a physically large one. JPL had won the Explorer mission in part thanks to its radio tracking work, and it began setting up a worldwide network of radio antennas, capable of communicating at any time with satellites in any orbital position around the earth. Such a network required three sites about 120 degrees apart in longitude. For the first and main site communications engineers, led by Eberhardt Rechtin, a Caltech PhD in electrical engineering, chose the Goldstone dry lake bed in the Mojave desert, about a hundred miles east of JPL. For the other nodes of what was called the Deep Space Network, JPL eventually settled on stations at Tidbinbilla in Australia and near Madrid in Spain, each of which by the early 1970s had antennas 26 meters and 64 meters in diameter. The massive antennas of the Deep Space Network enabled reception of signals transmitted by very low power spacecraft transmitters across hundreds of millions of kilometers. But large aperture alone was not enough. Hydrogen masers provided a precise frequency standard to ensure phase coherence of uplinked and downlinked signals, and cryogenic cooling of the ground receivers helped reduce signal noise. JPL engineers also developed complex codes to apply to the signals to screen out noise and transmission errors, as well as ones for data compression and pseudo-noise codes to prevent anyone-say, the Soviets-from eavesdropping or hijacking the spacecraft. Such techniques made JPL an important early center for telecommunications coding, and the techniques and the people themselves would help drive the emergence of the telecommunications industry, especially cellular phones, decades later.
Excerpted from Into the Black by Peter J. Westwick Copyright © 2007 by Yale University. Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.