P.H.Smith and the Smith Chart
Phillip Hagar Smith was born in Lexington, Massachusetts on April 29, 1905, to George and Rose Whitney Smith of Scotch and English ancestry. Rose Whitney was a descendant of Eli Whitney, the inventor of the cotton gin. While attending Tufts College, Phil was an active amateur radio operator with the call sign 1ANB. He also played the cornet in the Tufts College band. To commute between Lexington and Tufts, he drove a reconstructed model T Ford and later a 4 cylinder Harley Davidson motorcycle. He received the BSEE degree from Tufts College (now Tufts University) in 1928, majoring in electrical communications.
In 1928 he joined the technical staff of Bell Telephone Laboratories with the Radio Research Department, Deal, NJ where he worked under J.C. Schelleng and E.J. Sterba. In these early days, Phil became involved in the design and installation of directional antenna equipment for commercial AM radio broadcasting. In 1929 he was working in Lawrenceville, NJ, on an antenna system which was designed to communicate by short wave with Europe and South America. The antenna was connected to the transmitter by a two-wire transmission line. Perhaps the major reference at the time was J.A. Fleming's 1911 telephone equation, which expressed the impedance characteristics of high frequency transmission lines in terms of measurable effects of electro-magnetic waves propagating thereon, i.e., the standing wave amplitude and the wave position.
In the reprint of an article entitled "Transmission Lines for Short-wave Radio Systems," presented at the IRE 20th anniversary convention in April 1932, there was a footnote, which read "Disclosed to the writers by P. H. Smith, Bell Telephone Laboratories". The footnote referred to a paragraph in the article, which began; "There is another effective way for transforming line impedance by means of short line devices...." It was the first published report of work done by Phil, work that ultimately led to the creation of the Smith Chart.
In spite of his long identification and association with antenna activities, Phil was basically a transmission-line type. He relished the problem of matching the transmission line to the antenna; a component, which he considered, matched the line to space. Considering the frequency and the consequent large size and resultant cumbersomeness of the antenna, the measurements were not simple. In those early days, the sensing element was a thermocouple bridge with about 6 or 8 thermocouples coupled to two coils, whose dimensions were determined by the frequency of transmission. The indicator was a microvoltmeter, which measured the magnitude of the signal.
The entire assembly was then moved along the transmission line to determine the relative magnitude and location of the maximum and minimum signals. For transmission lines high in the air, this required one individual to move the sensing device along at the end of a long pole, while a second individual would read the signal through a telescope. It was primitive, but it worked. This was the early environment that Phil faced as an electrical engineer with the Bell Telephone Laboratories. For those who knew him best, it was no surprise that he would doggedly pursue his goal of creating a chart to simplify the work. From Fleming's equation, and in an effort to simplify the solution of the transmission line problem, he developed his first graphical solution in the form of a rectangular plot.
Phil persisted in his work, the diagram gradually evolved through a series of steps. The first rectangular chart was limited by the range of data it could accommodate. He was aware of the limitations and kept working on the problem until some time in 1936, when he developed a new diagram that eliminated most of the difficulties. The new chart was a special polar coordinate form in which all values of impedance components could be accommodated. The data for this diagram was scaled from the earlier rectangular diagram. The impedance coordinates in this case were not orthogonal and were not true circles, but, in the form chosen, the standing wave ratio was linear. The chart closely resembled what ultimately became the final result.
Phil, however, suspected that a grid made up of a system of orthogonal circles might be more practical. He felt it would have distinct advantages, particularly as regards reproducibility. With this in mind, he spoke to two of his co-workers, E.B. Ferrell and J.W. McRae. Because they were familiar with the principles of conformal mapping, they were able to develop the transformation whereby all data from zero to infinity could be accommodated.
Fortunately, curves of constant standing wave ratio, constant attenuation and constant reflection coefficient were all circles coaxial with the center of the diagram. The scales for these values, while not linear, were entirely satisfactory. A diagram designed along these lines was constructed in early 1937. It was essentially the form still being used today.
Smith approached a number of technical magazines with regard to publication of the Chart, but acceptance was slow. There were not many technical magazines at the time, and none in the microwave area. However, in January of 1939, after a delay of two years, the article was printed in Electronics magazine.
A fact one cannot ignore is that many highly competent people proposed charts for use in solving transmission line problems. Some of their charts had brief periods of popularity, but it is a comment on Phil's persistence in searching out the ultimate solution, that his Chart stands out above all others in its use and usefulness.
It took a while for Phil to convince other people of the utility of his chart. One of the first individuals to see its value was A.G. Fox at Bell Labs, who in 1939 found it useful in some early work he was doing on the new subject of waveguides. When the M.I.T. Radiation Laboratory was formed in 1940, the value of the Smith Chart was recognized immediately and it was put into general use. According to Phil, the M.I.T. workers were his first customers. It would be hard to visualize many of the achievements of the M.I.T. Rad Lab without some help from the Smith Chart. For microwave people at that period, the Smith Chart had the equivalent impact of turning on a bright light in a previously dark room.
Phil published a second article in 1944 which incorporated further improvements including the use of the chart with either impedance or admittance coordinates. In 1958, in the first issue of the Microwave Journal, a biography of Phil was published to acknowledge the importance of his contribution. In a series of 6 subsequent issues of the magazine, Dr. George Southworth described the importance and some of the applications of the Smith Chart.
According to Dr. Southworth, the Smith Chart, even in its earliest form, was no sudden flash of genius. Phil's first ideas were imperfect and they required time for full maturity. However, as Dr. Southworth wrote, "it was to his everlasting credit that he did not allow his idea to die on the vine, but nourished it until he had brought it to a high degree of perfection".
Today's emergence of the digital computer as a dominant design tool has in no way diminished the importance of the Smith Chart. The Smith Chart has become the ultimate background for both computer and measurement instrument displays.
Phillip Smith Beyond the Smith Chart
Had he not invented the Smith Chart, Phil would still deserve to be honored for his many contributions to technology. Just before America's entry in World War II, he was sent with a small group of engineers to Fort Hancock to work with the Signal Corps Laboratories on a most important secret weapon - radar. He spent a year on Sandy Hook designing antennas and related components for production of the SCR-268 radar. Later, he worked on early microwave radar antenna developments for submarine use under W.H. Doherty at Whippany, NJ. In his early professional career, while developing 500 kW coaxial line components for radio station WHAS in Louisville, Kentucky, he obtained a basic patent on the optimum conductor diameter radio for a coaxial transmission line. This is the outer to inner diameter ratio of a coaxial line, which results in maximum power handling capability for a given outer conductor diameter. Smith said this was one of the simplest patents ever granted - the only claim was the single number 1.65. Another basic patent he obtained was for the adjustable matching stub tuner.
After World War II he worked on the design of FM broadcasting antennas for Western Electric broadcasting equipment. During that period he invented the famous "Cloverleaf" antenna. Later he became involved in military weapon radar systems studies and designed and supervised groups responsible for the electrical design of the DEW LINE, NIKE ZEUS and the ABM System, which became SAFEGUARD.
One of the programs he worked on that can help to illustrate his creativity in microwave technology was an acquisition radar system on the Island of Kwajalein, in the South Pacific. This was an experimental system in the early days of the SAFEGUARD program. The design of the antenna involved using a Luneburg lens technique. The classical Luneburg lens is a spherical lens that has the property that when the lens intercepts a plane wave, the focal point of the wave will always appear at a point perpendicular to the wave itself on a line through the center of the sphere at a point on the opposite surface of the sphere, regardless of the direction from which the plane wave approaches the lens.
This made it possible that when a signal was received, by virtue of the location of the receivers and the action of the Luneburg lens, one could determine the azimuth and elevation of the target.
The technique that was used at Kwajalein was to build one half of the sphere - that is a hemispherical Luneburg lens - with a ground plane significantly larger than the diameter of the sphere itself. The lens was made up of a series of polyfoam cubes about 2' x 2' x 2' loaded with aluminum slivers, so that the polyfoam block had a uniform dielectric constant throughout. By varying the amount of aluminum slivers, one could vary the dielectric constant of the block. The required values of dielectric constant were then determined to achieve the Luneburg lens performance. It turned out for their system they needed about 10 to 12 different values of dielectric constant and perhaps dozens of each value. The system worked as predicted by theory.
The operation of the antenna relied on the ability to build the homogeneous aluminum-loaded polyfoam blocks of different, but precise dielectric properties. The idea for the blocks came from Phil. This episode helps to highlight one of Phil personality traits. As a friend of his commented, "he could be oh so stubborn." And "on occasion that stubbornness had a profound effect." Against the wisdom of some of the most distinguished consultants at Bell Labs, Phil maintained that by the random distribution of the aluminum slivers the dielectric constant could be controlled both as to homogeneity and value so as to serve the needs of the project. The test proved he was right.
Phil married Rosine Rittenhouse around 1930. They had three children. Donald was born in 1932.. Stephen, born in 1936, is an engineer. in 1993. A daughter, Sharon was born just after the war.
Stephen recalls that around 1945 Phil used surplus components to assemble a television on top of a card table in their home. NBC was broadcasting 1 to 2 hours a night on channel 4 out of New York. As town folk regularly gathered to view the new marvel, Phil enclosed the entire tabletop to prevent contact with the HIV circuitry. Phil also had an interest in building boats, usually small boats with outboard motors.
In 1950 Phil took up private flying, eventually purchasing his own plane. He loved to fly and accumulated over 1500 flight hours in the U.S., Bahamas, Cuba, Mexico and Canada. On one eventful trip to Lexington, Phil, with son Stephen and Wally Smith (an unrelated coworker), ran afoul of a weather front and was forced to make an ungraceful landing in New London, Connecticut. Wally apparently refused to fly with Phil again.
Phil was continuously active in the IRE and later the IEEE from 1947 on. Phil served on and chaired numerous IEEE committees, including technical standards Committee 2 on Antennas and Waveguides. In 1952 he was elected IEEE Fellow "for his contributions to the development of antennas and graphical analysis of transmission line characteristics". He was secretary-treasurer of the Antennas and Propagation Society in 1954. He is a past member of Commission 6 of URSI, and a member of the Delta Chapter of Tau Beta Pi.
In March of 1958, he and his bride, Anita Macpherson from Maplewood, N.J.,flew in their private plane to Cuba for a honeymoon. In 1964, their daughter Penny was born. Penny is also an electrical engineer.
Toward the end of his career he continued to work as an individual contributor. Although he had the perks of a supervisor, he chose not to be a manager. His function was to look at anything and everything and contribute. And he did, in the very best sense of the word. He was completely happy in his environment.
He was a hands on engineer and was not particularly mathematical. When he had a problem to solve and recognized that he needed some special help, he would not hesitate to seek it out. He was highly organized and super meticulous. When he was sure he was right, there was no way to make him back down.
The first edition of this book Electronic Applications of the Smith Chart in Waveguide, Circuit, and Component Analysis, was published by McGraw-Hill in1969. He also authored an article on the Smith Chart for The Encyclopedia of Electronics published by Reinhold Publishing Company in 1962 and 35 papers on antennas and transmission lines. Phil has 20 U.S. patents in the microwave field including the basic patent on the transmission line matching stub, the Cloverleaf antenna, and the optimum power ratio coaxial transmission line.
Phil retired from Bell Labs in 1970.
At its annual symposium in 1975 the MTT presented him with a Special Recognition Microwave Application Award for his invention and application of the Smith Chart.
The Smith Chart was eventually manufactured and sold by at least two companies. When Phil retired from Bell Labs he organized Analog Instruments Company of New Providence, NJ - which initially merchandised simple navigational instruments for light aircraft, but later began supplying his charts and a dozen or more chart-related items. Through 1975 Analog Instruments has sold about 9 million copies to engineers and educators all over the world. The Smith Chart is currently selling at the rate of about a ton per year. The company is still operated by his wife Anita.
Phil Smith passed away on August 29, 1987.
In 1989, the 50th anniversary of the Smith Chart was celebrated at the MTT International Microwave Symposium in Dallas, Texas. Much of the material in this biography was taken directly from material prepared for that celebration.
In 1994, Phil was elected to the New Jersey Inventors Hall of Fame.
Much of this material was collected by Randy Rhea, N4HI. Randy is the founder of Eagleware, a company specializing in RF simulation software and the author of "HF Filter Design and Computer Simulation".