Oral-History:Edwin Lee White
About Edwin Lee White
Edwin White, a radio pioneer, received his AB and MS degrees from George Washington University. He subsequently joined the Naval Aircraft Radio Laboratory. Leaving NRL, White spent three years as an administrative radio engineer with the Signal Corps. He then joined the Federal Radio Commission as a senior engineer, remaining with the FRC, and its successor, the FCC, until his retirement as chief of the Safety and Special Radio Services Bureau in 1955.
The interview covers White's early acquaintance with Dr. A. Hoyt Taylor at the University of North Dakota and his subsequent work with Taylor at the Naval Aircraft Radio Laboratory (NRL). White discusses his work on shortwave communications for naval fleets, where his primary focus was on circuitry analysis. He goes on to detail developments relating to frequency accuracy and high-frequency standards for naval communications. White discusses his work with L.A. Gebart in the design of crystal-controlled transmitters and his successful efforts in crystal temperature control. The interview continues with a discussion of White's work with both the FRC and the FCC, focusing primarily on frequency regulation. The interview concludes with a discussion of international aviation and the establishment in 1949 of an international agreement concerning the regulation of international aviation frequency assignment.
About the Interview
EDWIN LEE WHITE: An Interview Conducted by Kenneth Van Tassel, IEEE History Center, July 1974
Interview # 019 for the IEEE History Center, The Institute of Electrical and Electronics Engineers, Inc.
This manuscript is being made available for research purposes only. All literary rights in the manuscript, including the right to publish, are reserved to the IEEE History Center. No part of the manuscript may be quoted for publication without the written permission of the Director of IEEE History Center.
Request for permission to quote for publication should be addressed to the IEEE History Center Oral History Program, IEEE History Center at Stevens Institute of Technology, Castle Point on Hudson, Hoboken, NJ 07030 USA. It should include identification of the specific passages to be quoted, anticipated use of the passages, and identification of the user.
It is recommended that this oral history be cited as follows:
Edwin Lee White, an oral history conducted in 1974 by Kenneth Van Tassel, IEEE History Center, New Brunswick, NJ, USA.
Interview: Edwin Lee White Interviewer: Kenneth Van Tassel Date: July 1974
Dr. A. Hoyt Taylor
This afternoon we are interviewing Mr. Edwin Lee White. He was born in Valley City, North Dakota in 1896. He attended various colleges and received his AB and MS degrees from George Washington University. After receiving his bachelor's degree from George Washington University in 1922, he joined the staff of the Naval Aircraft Radio Laboratory, which later became a part of the Naval Research Laboratory at Anacostia. From there he went to Hawaii for a three-year stint as an administrative radio engineer with the Signal Corps. He returned to join the Federal Radio Commission's staff as a senior engineer in their newly created engineering department. He remained with that organization and its successor, the Federal Communications Commission, until his retirement in 1955 as chief of the Safety and Special Radio Services Bureau, having taken time out to serve in the Air Force during World War II as a colonel. Since his retirement from the FCC and the Air Force he has done consulting engineering and teaching.
In our informal discussions you told me of your early interest in radio. You were a student of Dr. A. Hoyt Taylor at the University of North Dakota. During the school year of 1916-17 he broadcast regular programs of music from the university's radio station. Will you tell me about what he did, and some of his problems?
Although I was only a junior student at the university, I became quite well acquainted with Dr. Taylor. The year before I had taught school at Ariska, North Dakota, and among other things had arranged for Dr. Taylor to lecture to the townspeople on radio. Of course, as the one who had arranged for his appearance, I was his host, and when I became his student in engineering physics, the friendship ripened. In 1914 to 1915 Dr. Taylor had built a Poulsen arc transmitter at the physics laboratory of the university. For a power supply he arranged connection to the overhead wire of the city electric railway system, which was 550 volts DC with a negative ground. This posed problems because normally in an arc transmitter the positive was grounded to simplify cooling problems. As I remember, he water-cooled the positive pole of the arc, using a column of pure water as the insulator. Another problem was the frequency he was required to use. His license from the Department of Commerce required him to use frequencies higher than he would normally use in the transmitters, which made an unstable arc. How that problem was solved I do not know, but it was. I believe he used frequency shift keying rather than shifting power to a dummy antenna. In the winter of 1916-17 he did start a regular program of broadcasting music from that station. He built a carbon microphone into the tone arm of a phonograph and connected that microphone as well as another one for announcing purposes into the antenna circuit of his arc transmitter. A switch shorted out the microphones not in use. I think he had several antenna microphones ready for service because the antenna current burned out one in a very short time. All had to be rebuilt every day for the next night's program. These programs were received over a radius of about 200 miles from Grand Forks, North Dakota. All this radio work was closed down with the entry into World War I. The summer of 1917, Dr. Taylor was commissioned as Commander of the US Naval Reserves and was put in command of the Naval Radio Training Center at the Great Lakes Radio Station just north of Chicago, Illinois.
After the end of World War I, Dr. Taylor, instead of returning to teaching, being released from his naval commission, was made head of the Naval Aircraft Radio Laboratory at Anacostia. I understand you joined that staff there after receiving your bachelor's degree from George Washington University. What work was being carried on at this early date?
Broadcasting. The radio station there was started on a scheduled basis from 1921, under the call letters NSF, and continued through the winner of 1922 and 1923. I remember that they had a great deal of trouble. They used local people — some people wanted to come back too often. They weren't selected because they didn't have enough time, and they went to their congressmen because they weren't selected. So they ended up by using a naval band only. But they did have the programs. The main thing aside from this broadcasting was the conversion of the fleet from spark to vacuum tubes, initially to five-hundred-cycle, self-rectified tube sets. Dr. Taylor was a firm believer in continuous waves, but the top naval communications staff was not easy to convince. I believe it was Admiral S. E. Hooper who stated that CW could not be practical for the Navy until both transmitter and receiver could be set on a designated frequency within five hundred cycles. This figure was arrived at by assuming that one set might be set five hundred cycles low, and the other might be five hundred cycles high, and that they were supposed to be set with a fifteen-hundred cycles offset, with a nominal beat tone of fifteen hundred cycles. Then the total they would get would be somewhere between five hundred and two thousand cycles, both well within the audible response range of the personnel and the equipment, the receivers. With the tube circuitry then in use, this kind of stability and accuracy of equipment setting was unheard of. However, with the development of radio direction finding, the sending of strings of signals, of calls, the tuning in could no longer be tolerated.
While this was going on, the Naval Research Laboratory at Anacostia was established, with the Naval Radio Lab at the Bureau of Standards, another lab at Dahlgren, one at New London, and perhaps others whose identity have slipped my mind. In any case, the lab at the Air Station was closed, and I, with Warren Burgess and a few others, were the first load of men and equipment which landed at the dock at the new laboratory and took possession of the new buildings. Al Crossly, Dr. E. B. Stevenson, and I were with Dr. Taylor at the University of North Dakota, 1916-1917, and were now again together at Annapolis. Dr. Stevenson was in the underwater sound division. Al was working in radio, as was I. Of that first group, I think L.A. Gebart, who was still active at the lab, Leo Young, and I are the only ones still alive. The problems at that time, in addition to those growing out of the requirements of stability and reset ability, were those generated by new equipment and by the discovery that the short waves were good for something other than the amateurs. I had my fingers in all these problems.
The problems of circuit stability, particularly oscillator stability, were first attacked. I had done some circuit analysis in my graduate work at George Washington University, which I was continuing at night, which indicated that the worst offender in oscillator instability was the grid-filament impedance. So, at home, I designed an oscillator completely shielded, except for a single turn at the ground level of the oscillator coil, using a modified Colepits circuit with the highest possible ratio of the grid filament, grid plate capacitance. I believe it was ten-to-one.
To maintain stable oscillation over the tuning range, which was from just under a thousand kilohertz to just over two thousand kilohertz — one octave — I used one of the best variable capacitors available to the amateur — Nashville's was the brand name — which had a friction territory ratio system without backlash between the knob and the dial. I cut a vernier to divide the dial range into tenths. I also put a voltmeter on the filament tube and used batteries as a plate voltage source to ensure stable voltage there. When this was completed, I could set it at a standard signal, turn it off, reset it by the numbers within some two hundred cycles, leave it on, and it would stay put. I took it into the lab and turned it over to Dr. Taylor to be tested. My circuitry was adopted, and was the basis for future stable tube oscillators. The first modification was the addition of the crystal calibrator, and the use of a screen grid as the oscillator plate, with a tube plate as the output pickup. Both these were NRL developments, but not at my time.
Transmitter with Reset Ability
As for the reset ability of the transmitter, a tube transmitter was designed, originally being a self-rectifying job to replace the spark transmitter used in emergency sets in the fleet. It ended up as a CW transmitter, as I remember it, in the two- to four-megacycle band. The oscillator had a tapped coil. One end was signal-turned taps, the other was ten turned taps. In the center was a silver disk, mounted so that it could be turned either parallel to the winding or vertically to the winding. The coil was mounted at a forty-five degree angle, and the rod perpendicular to the panel, with the disk-mount also at a forty-five degree angle. In this combination, the disk provided for the spaces between the settings on the turn taps, and a smooth inductance setting through the whole range was possible. This transmitter required that two myths be dispelled. First, as the superstructure of a battleship cast so many radio shadows, that short-waves were impractical aboard ships of the line. The second was that short-wave transmitters could not be located below the water line because of the long antenna leads required. This transmitter fed into what we would now call a transmission line, to a tuned circuit at the base of the antenna top-side. There was a remote antenna tuning motor control and a remote reading antenna current-reading meter, both of which worked.
Field Work with Shortwave and Continuous Wave
<flashmp3>019 - white - clip 1.mp3</flashmp3>
Robert Myers, another of Dr. Taylor's staff, and I were sent to the fleet onto the West Coast with a small shortwave transmitter and a portable shortwave receiver to match. The receiver was fitted with a constant impedance signal measuring bridge. I've not seen one outside of a museum for many years. It had some hundred pairs of contacts and an arm to reach them. The receiver arm was connected to one end of the box and the headset to the other. In use you adjusted the arm and you could just recognize the signal. To broadcast, the transmitter was set up between the bridge and the forward mast, just aft of the two main forward turrets, which was deemed by the pessimist to be the worst possible place. Bobby manned the transmitter. I went out in a boat about five miles with a crew of sailors and started circles. This was off Long Beach, California. On the U.S. California, where the transmitter was located, a man with a Polaris kept watch on our position. Bobby sent Vs, nonsense, and stuff, and every ten degrees, on signal from the watchman, sent a dash, at which time I would record the signal strength. We circled that ship at least half a day, at various distances, and needless to say, we found no shadows. The evidence was now only in on CW, and basically on short wave, but still the top naval staff was not sold.
The fleet was scheduled to make a trip to Australia. Dr. Taylor arranged for the fleet command ship to have a high frequency transmitter and receiver aboard. Fred Snell, then with the ARRL, at Harvard, who was holding a commission in the Naval Reserves, was called to duty and sent with the fleet. Transmitters and receivers matching that with the fleet were sent to Pearl Harbor, Leo Navy Yards, Naval Headquarters in Washington, and of course the laboratory. Fred maintained schedules with all of them. The arc transmitters faded out about half-way to Australia from Hawaii. The high frequency stayed on and was able to handle all of the fleet traffic the whole trip. Short waves had arrived, and the advice of NRL was never afterwards questioned. So much for stability.
The other urgent question was the frequency accuracy. The first detectors I remember were based on some of Dr. Cady's work in 1922, when he discovered quartz resonators. Dr. Taylor and his receiver staff, probably Leo Young, and perhaps Thomas McKell-Davis, who died just the other day at the age of 83, were involved. They designed a receiver with a circular box built in with spaces in the rim for a number of quartz crystal rods at specified resonant frequencies. These rods could be connected one at a time to the oscillating circuit of the CW receiver. As the alternator flipped through the frequency of the crystal, of the crystal rod, the rod was excited, and a chirp was heard in the receiver headset, which gave a spot frequency check of the receiver calibration.
As I understand, Mr. White, this was the first use of crystals to calibrate receivers at that time, to use these for purely calibrating receivers.
As far as I know, yes it is.
What range of frequencies, again, did you say that these crystals were adjusted for?
If I remember the crystals they were using, they probably ran from two hundred kilocycles to maybe fifteen hundred kilocycles, hertz. And of course they only gave approximate spot checks because the chirp only lasted as you passed through. You had to visually make a spot as to where the place was as you went by in either direction, and then mark that with your eyes and use that as a calibration.
Well, this project showed promise, but was abandoned with the announcement of Dr. G. W. Peirce's work using quartz crystals to control the frequency of oscillations. A whole new field then opened, and they dropped this other business of using the quartz crystals. Interesting to note, the first crystals that were used were discarded quartz eyeglasses.
Dr. J. M. Miller — I think his initials were J. M. M. — had been assigned to the development of a satisfactory high-frequency standard for permanent installation for the Navy, for a portable high-frequency standard for use in the fleet, and for other naval installations. At this time, the best frequency standard was the general radio wave meter, which was an inductive capacitor, and a beautiful capacitor. One thing they learned was that the capacitor itself had a temperature coefficient, and it varied between capacitors. Some of them had a positive capacity coefficient and some of them had negative, so under Dr. Miller's direction they started finding why. They found that it was minor variations in the tolerances of making the thing up, in the spaces between the leaves.
As I understand, we are talking about air condensers?
They could make those condensers with a zero coefficient. That helped a lot of things. There was a chap by the name of Eisenhower. I don't remember his first name, but everybody called him Eisenhower. He was no relation to General Eisenhower, who became president. He was charged at the Tate laboratory, and he cut our crystals for us. I think that it was under his direction that they finally got the AT cut crystals. I don't know whether his laboratory did it or whether the Bell Laboratory did the final work on the AT crystals. But at any rate, they were both working on it at the same time, and who finally got it I don't know.
I do know, as you say, that Bell Labs and the Navy were working on AT cuts.
At the same time. But at this time the crystals were still temperature conscious.
But before that, Dr. Miller was working on a temperature-controlled frequency standard. It was closely temperature controlled. It was a storage battery, the power supply, to ensure voltage stability. He had a chronograph which he used to compare the signals from his standard with the signals from the Bureau of Standards, and shortly his time agreed better with itself than the Bureau of Standards to itself, which must have been an embarrassment. This set the Bureau's research program, which resulted in their finest time standards in the world. The stable oscillator then used as a frequency meter would supply the crystal calibrator, and it became the standard frequency meter for all purposes during World War II.
L. A. Gebart was assigned the task of designing a crystal-controlled transmitter with a twenty-kilowatt output to operate from a four thousand kilocycle — hertz. I do not remember the exact frequency, perhaps four thousand fifteen. This was a success. The crystal controlled a fifty-watt tube, which drove two two-hundred-and-fifty watt tubes in parallel, which drove in turn a twenty-kilowatt tube. I was assigned to assist Gebart in this project. We were told that when the laboratory industry was asked to bid on this project, GE refused, stating that their staff had investigated and believed that there was no future in crystals for transmitter control. This may be a myth, but it was believed at the time. I was assigned to build the twenty-kilowatt transmitter to complement the one just finished, to operate on eight, twelve, and sixteen megacycle — hertz.
I came up with one, with much the same tube line-up, but by driving the first tube harder I doubled or tripled frequency at the two-hundred-and-fifty watt level. By driving both first and second stages hard I doubled at both stages, producing the desired final frequency. To produce sufficient doubling or tripling the grid wires for this must be raised much above cut-off, and the grid voltage raised as well. This works the crystal much harder, as might be expected. When the circuit was first tested, everything was fine until we went to full power, when there was a split frequency produced, traced to the crystal, not a parasitic. I sent the crystal back to Eisenhower and it checked out okay. We had a consultation. I found a brass plate, screwed a coiled brass tubing to it, lapped the other side smooth, mounted it on the insulator, put it in the set, set the control crystal holder on it, ran cool water through the tube, and it went full power, with no more trouble.
This, I would take it, is a good way of stabilizing the crystal parameters and taking the heat out of the crystal.
I believe that was the first case of crystal temperature control. To keep the temperature cool.
Some time before that, I had, under Dr. Taylor's direction, designed and built an aircraft radio receiver with a built-in crystal set for the frequency of the aircraft's control station on the ground. It worked beautifully, but when the pilot took altitude and the crystal drifted off frequency, communication was lost if things warmed up. About that time I left NRL to go to Hawaii to work for the Signal Corps. I applied for a patent on the crystal control and receiver circuitry, and ran into interference. My patent had been assigned to Wire and Radio, that is, a commercial interest. The interference was settled by agreement, I found out later, and my application was withdrawn, although I had my receiver in full flight before the successful applicant even claimed he had conceived the idea.
Federal Radio Commission
After three years in Hawaii I joined the staff of the engineering division of the FRC. That was an interesting time. There were several landmark decisions made in the early years of the FRC, which may not be appreciated. On the first day of the wireless field, in 1913, I didn't need a license under the then-current concept of law and radio usage. But during the early years of the FRC it was finally determined by the courts that radio by its very nature was interstate in character and required national standards and regulating. A decision based on sound engineering. The airlines had a major hassle with the FRC, which stood adamant that no airline could have its own frequencies. All airlines using the same route had to use the same frequencies to hear each other, and that the frequency was for safety only, not for chit-chat by the passengers. That's the same position that was assigned to the FRC in the police service. Metropolitan areas got the same frequencies, and the local police departments had to co-operate. It was hard to convince senators, congressmen, and local interests that there were not just enough frequencies.
Minneapolis-St. Paul Case
In connection with the Federal Radio Commission, or FCC work, there are some cases which are non-technical. You told us a little while ago about an interesting story in discussing with a senator who was objecting to frequencies used in Minneapolis and St. Paul. How did this turn out?
Minneapolis and St. Paul were set on the same frequency. The FCC said that as far as the criminals were concerned they were one area, and one radio system should handle it because all the police should know what was going on in the criminal field. The police seemed to think so too, but the city councils thought that they ought to be different, so they approached the senators and all that. They came down to see us, with blood in their eyes, saying that the city of St. Paul was very important. So was Minneapolis, and they should have their own radio systems and frequencies. So we pointed out the problems about the police needing to co-operate, and how they were co-operating, and how one of the things that had helped them to co-operate was their single radio system. After they heard the story, that suited them. There were other occasions where the frequency was a problem. Michigan was very much worried. At one time Michigan was going to turn out the National Guard because the commission restricted the power that they could use in the state police radio system, but they didn't get away with it. Later, when they began to get interference from somebody elsewhere, they wanted the FCC to step in and help them cut the interference down. But as they began to get communications people on their staff rather than politicians, they began to see the light.
Inter-City Radio Case
<flashmp3>019 - white - clip 2.mp3</flashmp3>
The Inter-City Radio case was another one of the early cases. It's not too much engineering, but a very important case. Mostly legal. In the Inter-City Radio case a corporation called Inter-City Radio was formed, and the FRC in the early days authorized it over opposition by Western Union, the Post and Telegraph, RCA, Mackey Radio, AT&T, and some others. It was appealed. This company was going to provide domestic radio service between selected points in the United States. Not full coverage like Western Union, but selected coverage. The radio companies opposed it on the grounds that the frequencies should be reserved for jobs which could not be done by wire. The others objected on competition grounds because here was a company competing for the cream of the crop and not serving all the little places like they had to. At any rate, the important thing was that it went to court, and the court started to hear the case, what they called de novo I believe, as lawyers call it. In other words, the court started right from the ground up, just like the FRC did, called in engineering witnesses, accountants and everything else, hearing the whole story. They had a whole room full of records, and they took so long that the Inter-City Company went bankrupt and the case folded because there was nobody else involved. As a result, the law was amended, which gave the FRC power, and enabled the FCC to have final judgement and the weighing of the facts. The courts can only overrule the FCC on errors of procedure and failure to consider the facts.
This gave them a great deal more power than they had before, and as you say, it was final.
Actually, the way it was before was impossible because on occasions we were silly. The litigation completely killed a company before it had a chance to do anything about it.
International Agreements: World War II and 1948
Now, as far as the other subject, about using the radio before wires could be used, later the international treaties came along and put in international law that, as a general principle, radio should not be used where wire could serve. Of course, nowadays with the use of satellites and microwaves and the resulting larger availability of radio frequencies domestically, wire can be used but is not entirely suitable.
The decisions about aviation, that the airlines could not have separate frequencies, had to be for safety. Before World War II, western hemisphere aviation and European aviation developed in different ways. Following the U.S. lead established in 1930, communication with aircraft followed the route's flow in the United States and Canada. In 1936 there was an inter-American aviation conference in Peru. I was with the U.S. delegation. Lufthansa was there, that's a German outfit, and urged the adoption of the European concept, which was basically point-to-point flying, but the U.S. concept prevailed. To expand on the European concept, with the large number of countries there were no long flights, such as are common with us. As a result, the custom was for aircraft to fly from airport to airport requiring navigation aids only at the two airports, and communications only there. The frequency requirements for short range could be repeated frequently and were therefore small. World War II changed this picture. The long-range bombers and military transports came into being. Air transport routes were established circling the globe. Long-range navigation systems came into being and were installed. In the war situation the frequencies needed were occupied. In 1945 the radio frequency world was in chaos. There were new nations; other nations had disappeared. The U.S. suggested a meeting of the ITU at Atlantic City, which was held.
I was on the delegation.
At that meeting a place had been made for aviation, both military and civil. No one wanted to release frequencies. Generally, the maritime nations were not interested in aviation and wanted to keep the marine frequencies. Many nations wanted to hold on to the higher frequencies used for wartime propaganda broadcasting, for continued propaganda. The U.S.A. was among those, the Voice of America. The United States was really the only world power with a going aviation industry. The compromise was reached. Aviation got frequencies, not enough in some bands, more than enough in others. Starting in the summer of 1948, a world international administrative radio conference convened under the auspices of the ITU in Geneva. I was vice chairman of the U.S. delegation. Art Labelle was chairman, and was elected chairman of the conference, so the task of U.S. spokesman fell to me. At that time I was chief of the aviation division, the engineering department of the FCC, having served during the war as a colonel of the USAF. The skeleton which was the U.S. position prior to this conference was developed in that division.
Basically the principle was that all aircraft on a route, regardless of nationality or ownership, would use the same frequencies. Frequencies were furnished a route at such bands as might be necessary to ensure communication at any time of the day, year, or sun-spot cycle, to one end or other of the route. The number of frequencies to be assigned would be in proportion to the number of aircraft flown over the certain route. Further, frequencies would be assigned in two classes. The major world areas, or route areas, were such as in Europe, North America, and regions such as Scandinavia, Benelux, and so forth. The United States’ position was finely polished, and I presented it to the conference. I said we realized that under the formula proposed, and considering the number of flights actually being flown in the United States at the time, the U.S. would require all of the available frequencies. However, we would agree in advance that under whatever formula the conference would agree to, the U.S. would accept only half the frequencies the formula showed it was entitled to, and would make do by using VHF to supplement its communications system. The Iron Curtain bloc refused to give any information about their current flight patterns. All the other countries co-operated and worked like beavers. We suggested that all we could do was assume that the Soviets had a flight density similar to the rest of the world and assign them frequencies accordingly. We pointed out to those who said that if they didn't want to co-operate we should leave them out, that any plan without them would never last as they would go as they pleased, cause interference, and blame the rest of the world. After three months of hard work a draft plan was worked out, and the assembly adjourned in the summer of 1949. In the interim, countries were to try out the plan for size, be prepared to modify or improve it in 1949. In 1949 the plan was approved with the ten countries of the bloc failing to sign. They were, however, some of the first to conform, and within five years formally joined. In 1964 the plan was renewed by another conference, and although the frequency assignments were changed as developments of world aviation indicated, the plan itself was reaffirmed.
- 1 About Edwin Lee White
- 2 About the Interview
- 3 Copyright Statement
- 4 Interview
- 4.1 Dr. A. Hoyt Taylor
- 4.2 Naval Aircraft Radio Laboratory
- 4.3 Signal Corps
- 4.4 Federal Radio Commission
- 4.5 International Agreements: World War II and 1948