An Electrifying Century - Electrical Engineering: An Early History of the Caltech EE Department

AN ELECTRIFYING CENTURY The First Half-Century An Early History of the Caltech EE Department by Donna Fox, Robert J. McEliece, and Babak Hassibi Caltech's Electrical Engineering (EE) Department had its beginnings in the summer of 1910, when Throop Polytechnic Institute (TPI) moved from what is now known as Old Town Pasadena to a 22-acre quiet campus bordered by California Boulevard, Wilson Avenue, San Pasqual Street and Hill Avenue. In 1910, Throop Polytechnic, which became the California Institute of Technology a decade later, offered undergraduate degrees in mechanical, electrical, and civil engineering, but there was no administrative structure, so President James A. B. Scherer, at the behest of Caltech's founding father George Ellery Hale, invited Royal Sorensen to start up an Electrical Engineering Department. It is fair to say that this was one of the best appointments in Caltech's history: Sorensen remained the head of EE for more than 40 years and saw his fledgling Department become one of the powerhouses of electrical engineering in the United States. Over the years since its inception, the EE Department has shifted its main research focus several times in line with the demands for new technology in a rapidly changing world-from early work on power engineering, to the war effort, to computing and communications. Pioneering research Throop became Caltech. The so-called "Big Three" (Hale, Noyes, and Millikan) drafted a new educational philosophy that emphasized pure science, and Sorensen conceived a plan for graduate study in EE, foreseeing a need for education beyond the traditional baccalaureate degree. has led to major technological innovations, spawned other disciplines, and developed vitally important collaborations. Power Engineering and the High-Voltage Lab EE's cutting-edge science has endured for a hundred years. Engineering Based on the Fundamental Sciences By the early 1920s, master's and doctoral programs in EE had begun, and the achievements of the large group of In 1910, when Sorensen, a 28-year-old graduate of the University of Colorado, left General Electric in Pittsfield, MA, to start the TPI EE Department, he worked with Scherer, Hale, and chemistry professor Arthur A. Noyes to produce a unique engineering department based on the fundamentals of physics and mathematics. At first, there were only three engineering students— two EE graduate students, and one mechanical engineering student. In 1913, TPI became Throop College of Technology, and Robert Millikan started a few years later (1917) as Director of Physical Research. In 1920, Millikan succeeded Scherer as Chief Executive, and Royal Sorensen 14 EN Genious NO.7 2010 AN ELECTRIFYING CENTURY 2,000,000 volt surge generator of the High-Voltage Laboratory alumni these produced proved the merits of his concept of cation that is now Sloan Mathematics) to solve some of emphasizing physics and mathematics. Sorenson was one the emergent problems in power transmission. It was the of the first to realize the importance of a national grid sys- first laboratory in the country to have a million-volt power tem to distribute electrical power in large-scale and long- source, provided by a cascade system of transformers de- distance transmissions and participated in California's pi- signed by Sorensen himself. The High-Voltage Lab was oneering work on this. In a cooperative venture with the available for both research and industrial testing and was Southern California Edison Company (SCE) in 1923, a used to aid SCE in the development of high-voltage trans- high-voltage laboratory was erected on campus (in the lo- mission lines, to furnish lightning protection of oil storage 15 AN ELECTRIFYING CENTURY EE undergraduate student in the late 1920s, and stayed on to complete a PhD in Physics in 1936. At the time, radio communications and broadcasting were beginning to develop, but the use of electronics in control systems was not yet well known and, as the Department was still heavily oriented toward power engineering, radio communications and vacuum tubes were not considered very important. Mackeown, how-ever, could see the growing importance of this field and urged Millikan to appoint a professor whose research specialty was electronics. Mackeown even offered to sacrifice his own salary for such an appointment. Pickering was the obvious choice and joined the EE faculty shortly after graduating. During this period (1936), the Jet Propulsion Laboratory (JPL) began in Pasadena as the Guggenheim Aeronautical Laboratory, California Institute of Technology (GALCIT) rocket project, led by Professor Theodore von Kármán and financed by Harry Guggenheim. In 1940, the William Pickering, as grad student, with the experimental cosmic ray telescope he built, used to investigate the source and intensity of cosmic rays. GALCIT group became a rocket research, development, and testing facility. A few years later, a substantial contract from the U.S. Army Air Corps transformed it into the large, tanks for the oil industry, and for other research efforts to permanent laboratory it is now, and its name was changed industrialize Southern California. Shortly after Caltech's to JPL. Its purpose was to research, develop, and test mis- lab was completed, a group of Stanford alumni built the sile technology. There was a need for electronics to support Ryan high-voltage lab at Stanford in 1926, and for several missile guidance and tracking systems, and Pickering's spe- years, Caltech and Stanford became the two outstanding cialization in electronics drew him there. In 1954, he be- schools in the country in the area of high-voltage work. For came the director of JP L and remained in that position until a time, Caltech was giving more PhD degrees in Electri- 1976. Under his leadership, electronics grew in prominence cal Engineering than MIT. Professor Charles Lauritsen at the lab, which became a NASA facility in 1958, and work built high-voltage vacuum tubes and EE Professor Francis began on a worldwide, civilian satellite communications Maxstadt was involved in the testing of insulators and tow- network, known today as the Deep Space Network (DSN). ers, more so than even Sorensen himself. Professor Stuart Mackeown also spent a lot of time in the High-Voltage Lab. The War Years Sorensen's experiments with high-power current interruption in vacuum, designed to address the serious problems During the World War Il years, Caltech's EE program was of control and protection of high-voltage power systems, led put on hold. Several faculty members left to take various to his invention of the vacuum switch in 1923, although it was war jobs, and from 1943 until 1946, Pickering organized not until 1960 that new technologywould make this invention a three-year, year-round Navy "V-12" curriculum to meet commercially feasible. His original model is on permanent the military's need for radar training. Upon completion of display in the Smithsonian Institute in Washington, D.C. the program, students went to MIT's Radiation Laboratory In 1926, EE became part of the Physics and Math- (Rad Lab) to study the more specific techniques. Picker- ematics Division, as the modern structure of Caltech's Divi- ing visited the Rad Lab periodically to learn more about sions came into place. the electrical characteristics of waveguides. He also organized and managed the Engineering and Science Manage- Electronics ment War Training (ESMWT) program, a series of evening courses that were given all around Los Angeles, and person- William Pickering, a New Zealander, came to Caltech as an ally taught some of the approximately 20 different courses. 16 ENGenious NO.7 2010 AN ELECTRIFYING CENTURY It was mostly done at the junior college level and was presented at several high schools in the area. The war effort also caused other Caltech faculty to redirect their focus. Gilbert McCann (PHD EE '39) joined Westinghouse in Pittsburgh to study natural lightning phenomena. But when his research was diverted to support the military, he was set to work devising a way of doing complex engineering calculations using electrical circuits to simulate mechanical forces. This was to become the basis of the analog computer that he invented to do calculations that would previously have taken years. McCann used this computer to design a system for improving the tracking accuracy of anti-aircraft guns, which enabled the British to shoot down almost all the Left to Right: Frederick Lindvall, Lee DuBridge, Gilbert McCann, and Kendrick Hebert in the Willis H. Booth German V-1 bombers cross- Computing Center ing England's eastern coastline. After the war, Frederick Lindvall (PhD EE '28), who ested in control systems) and Bart Locanthi. Their computer was the first to receive a PhD from the Caltech EE Depart- became known as Caltech's "Direct Analogy Electrical Ana- ment, became Professor of Electrical and Mechanical En- log Computer," and it provided service for JPL, the military, gineering and Chairman of the Engineering and Applied and the entire Southern California aerospace industry. The Science Division, a position he held for 24 years. Lindvall computer was mainly used to solve design analysis problems pioneered and taught advanced engineering courses, and in the fields of solid mechanics, fluid mechanics, and heat encouraged students to take as many courses as possible in transfer. In particular, methods of structural stress analy- other departments and divisions at Caltech. He introduced sis, vibration analysis, and aero-elastic analysis of airframes the concepts of "interdisciplinary research," "systems engi- were developed, the latter carried out primarily by McCann, neering," and "management of technology." Wilts, and Richard MacNeal (MS EE '47, PhD EE '49). By 1950, the McCann's lab had become too busy for The Computing Era Caltech. The director of GALCIT, Clark Millikan, suggested spinning off a commercial company, and Com- In 1946, McCann left Westinghouse to join Caltech's EE puter Engineering Associates (CEA) was formed on faculty and began setting up an analysis lab to build an im- Halstead Street in Pasadena. McCann, the largest share-proved version of his analog computer. He negotiated a deal holder, could not run the company since he would have with Westinghouse to make two of everything, so that the to resign his faculty position to do so, but MacNeal and duplicate parts were sent to his Caltech lab at a reduced cost. William Dixon (BS EE '48, MS EE '49, PhD EE '52) The 33,000-pound computer was assembled with the help of became senior managers, and Locanthi was in charge EE professors Charles Wilts (who eventually became inter-of constructing the computers. They were built pri- 17 AN ELECTRIFYING CENTURY marily for the burgeoning American aircraft industry. into a group studying plasma physics and microwave in- Shortly before the spin-off, McCann recruited Stan- teraction with matter. After the microwave-tube era came ley Frankel to head a small, innovative digital comput- to an end in the late fifties, his research focused mostly on ing unit. A computer scientist with a PhD in Physics plasma physics. Charles Papas, who joined the EE faculty from Berkeley, Frankel had become interested in digital a year after Fields (1954), focused his work on electricity computing during his time at the Manhattan Project in and magnetism. In 1965, he wrote a book, Theory of Electro- Los Alamos. At Caltech, he worked with graduate stu- magnetic Wave Propagation, based on his Caltech lectures, in dent Bernard Alder (PhD EE '52) on his thesis proj- which he presented a number of newly important topics in ect-the simulation of the interaction of atoms using the theory of electromagnetic wave propagation and anten- statistical mechanics and liquid theory-for which they nas in a coherent and simple way, providing evidence that developed what is now called the Monte Carlo method these ideas developed from the Maxwell field equations. of statistical analysis. This pioneering work has become David Middlebrook, who joined the EE faculty im- essential to a wide range of scientific and technical work, mediately after gaining a Stanford PhD in 1955, became from nuclear weapons development to VLSI chip design. known as a power electronics icon. In 1970, he founded the In the 1950s, Frankel designed the logic for a gener- Power Electronics Group at Caltech, which developed into al-purpose "poor man's computer," called MINAC. Libra- an academic discipline. His teaching was widely lauded by scope, a Southern California company located in Glendale, both students and colleagues, and in 1997 he was awarded licensed the design from Caltech and hired Frankel and Caltech's highest teaching award, the Richard P. Feynman physics graduate student James Cass (MS EE '53) to turn Prize for Excellence in Teaching. He had a deep under- the design into a production-ready product. Marketed as standing of both power electronics and electronic circuit the LGP-30, it may well be considered to be the first per- design and taught analog and power electronics courses sonal computer and is now on display at the Smithsonian both publicly and at Caltech. His seminal book, An In- Institute and the Computer History Museum in San Jose, troduction to Junction Transistor Theory, helped engineers CA. understand how to apply transistors in circuit design. His extra-element theorem, which yields simple formulas for Growth in the 50s the effect of adding a single element to an electronic cir-cuit, is still widely used in circuit design and measurement. Robert Langmuir (PhD Physics '43) left the General This glimpse into the early evolution of Caltech's Electric Research Laboratories in New York in 1948 to EE Department highlights its heritage. Space limita- join the EE faculty and remained there until his retire- tions prevent us from casting a wider net and crediting all ment in 1980. He had invented and patented a mass spec- those whose work has been a part of this impressive his- trometer in 1940 and discovered the radioactive isotope tory. Since its beginnings in 1910, the EE Department has K37. While at General Electric, Langmuir had worked laid a strong foundation to support the scientific research on the development of CW (continuous wave) magne- and discovery that continues today. Since the 1960s, it has trons, which were later used by the Navy to jam enemy expanded into new areas, such as solid-state circuits and radar. In 1960, Langmuir participated in the building of devices, electromagnetics and optics, control, communica- the 1.5 BeV synchrotron at Caltech, the first to operate tions and information systems, and networks. It also played in the United States. He also taught courses in electric- a critical role in spawning the Applied Physics, Computer ity and magnetism, and electronics, while conducting re- Science, Control and Dynamical Systems, and Computa- search in various topics in applied physics and engineering. tional and Neural Systems Departments at Caltech and Five years later, in 1953, Lester Field arrived from retains close ties to these through several joint faculty ap- Stanford, where he had been one of the originators of their pointments. Our complementary article highlighting cur- electron-tube laboratory. He continued traveling-wave rent faculty research interests provides further details. EN a tube work at Caltech, setting up his lab in the basement of Spalding. Field developed several forms of beam plas- Visit: http://ee2.caltech.edu/centennial/memories ma and established a tube-research activity that expanded 18 ENGenious NO.7 2010 AN ELECTRIFYING CENTURY A research fellow in Electrical Engineering tests a microwave antenna. CC Electrical Engineering ALT FORNIA INSTITUTE OF TECHNOLOGY An Electrifying Century 19