Birth of Radio Location in the United States Navy - Chapter 4 of Radar and the Fighter Directors

By David L. Boslaugh, Capt USN, Retired

A Mysterious Interference

It was 24 June 1930 at the Naval Research Laboratory (NRL) on the Potomac River near the southernmost point of the District of Columbia. Laboratory engineers Leo Young and Lawrence Hyland were outside testing a series of focused radio beams they hoped to use to guide airplanes to landings in darkness or fog. Hyland had set up a field strength meter to check the intensity of one of the beams, and as a test airplane following the beam came nearer he noticed the field strength meter dial fluctuating widely. They both realized that their radio waves were being reflected from the plane and, what was surprising, the reflections were strong enough to cause very noticeable constructive and destructive interference with the strength of their beam. The two engineers reported the strange phenomena to their boss, Dr. A. Hoyt Taylor head of the NRL Radio Division. He encouraged the two to keep studying the reflected wave phenomena on a not-to-interfere basis with their other work. [1, p.62]

In the Fall of 1930 Dr. Taylor wrote a report to the Navy’s Bureau of Engineering saying he believed that a device using radio waves reflected from objects could be used to not only determine their locations at long distances, but also by measuring the Doppler shift in frequency of waves reflected from moving objects, the objects speed could be determined. He indicated that the Radio Division planned to continue research in the area using a small amount of their directly appropriated research funds and he requested additional funding support from the Bureau. The Bureau’s response was less than enthusiastic; no funding was offered. Taylor prevailed on NRL’s Acting Director, E. D. Almy, to send a second letter on 16 January 1931 describing his impressions of the value such investigations might have for the Navy. This time the Bureau responded in three days authorizing a laboratory “problem” to study the use of radio waves to detect the presence of enemy ships and airplanes. The reply also emphasized that the research was to be kept secret. [1, p.63]

The project had official sanction but no additional staff was authorized. It would have to compete with other Radio Division work, and for years would progress at a low level. The Navy airship Akron appeared over the laboratory in December 1931 to provide a target for calibrating high-frequency direction finders, and it was also an opportunity to verify Taylor’s theory that the wavelength of a signal used for radio detection should be about the same as the size of the object being detected. They found they could detect signals bounced off the dirigible at frequencies down to 1.4 megahertz, but a smaller Curtiss Condor passenger plane showed no echoes at that frequency; confirming the theory. Other experiments, using continuous wave transmitters and receivers, showed that the transmitter and receiver had to be separated from each other at distances exceeding even the length of large ships to keep the transmitted signal from saturating the receiver. This did not bode well for shipboard radio detection equipment. [1, p.64]

In mid February 1934, members of the Congressional Naval Appropriations Subcommittee visited NRL to review its efforts, and one of the projects they were shown was the radio detection device. Up to this time only three men had worked on the project, and only part time at that. A fourth person, newly hired junior engineer, Robert M. Page, was assigned to help the group get the equipment ready for demonstration. The Subcommittee members were apparently impressed with the demonstrations because they generously increased laboratory funding in their next appropriation. [1, p.79]

A Shift to Pulsed Radio Waves

Until this time, the radio detection equipment had used continuous waves of radio energy, and the results had not been spectacular. Engineer Leo Young had experimented for a while in 1933 with pulsed waves, and he convinced Dr. Taylor that radio detection should be tried with short pulses of radio energy. Taylor directed the new engineer, Robert Page, to experiment with pulsed waves to detect the presence of ships and aircraft. Working by himself, he first devised a means of showing range of a detected target. He borrowed an idea from the laboratory’s underwater sound ranging equipment, a circular trace around the periphery of a cathode ray tube. He would start the trace moving when the pulse was transmitted and an echo received back would show as an outward radial spike on the circular trace. The position of the blip from the starting point could be calibrated to indicate target range. [1, p.82]

Next, Page devised circuits to modulate a transmitter to send out pulses of high frequency energy lasting about ten microseconds, and then rest for ninety microseconds before sending out the next pulse. He found an existing transmitter at the lab that he could modify to send out unusually strong pulses. The head of the Lab’s receiver section said he would loan Page a receiver if he did not “modify it too badly.” His modifications were mainly to keep it from breaking into oscillations when it sensed the strong transmitter pulse. It had to recover quickly so it could sense an echo from a target at close range. He made his first test in December 1934, and the results were not encouraging. The transmitting side worked pretty well, but the receiver did not recover fast enough from the outgoing pulse. Page was going to have to design a receiver specifically to support radio detection, and he worked on the new receiver until November 1935. All this time, Page was still the only one working on the project, and even then he was sometimes detailed to other work. Furthermore, there was little priority in the electronics fabrication shop to build his receiver. [1, pp.82-83, pp.88-90]

An Increase in Priority

In early 1935, Hoyt Taylor, Head of the Radio Division, and Harvey Hayes, head of the Sound Division, got permission from the NRL Director and from the chief of the Bureau of Engineering to brief an influential member of the Congressional Naval Appropriations Subcommittee. They emphasized the need to increase the small annual Congressional direct appropriation to the laboratory because there was potentially much high payoff scientific work they should be doing in the fields of radio waves and underwater sound, but they did not have the resources. The subcommittee member listened to them but made no commitment, however, in a few days he phoned saying the Committee was increasing their annual appropriation by $100,000. This was a lot of money in 1935. This meant their 1936 appropriation would increase by fifty one percent compared to 1935. The laboratory proposed a priority order in fifteen fields of study, of which the “study of the detection of ships and aircraft by radio” was listed as ninth. Surprisingly, when the Bureau of Engineering reviewed the list, they directed that radio detection be moved up to third priority. [1, pp.90-91]

Thanks to the new priority and funding, a second engineer, Robert C. Guthrie, was added to radio detection research. He started working with Page on 22 November 1935, at which time the electronics shop finally finished Page’s receiver. In December, Page decided to replace his circular display sweep with a linear sweep similar to the Royal Navy’s solution, where range was easier to read. By December the two investigators had their electronics ready for a trial. The Naval Research Laboratory had an existing large directional curtain array antenna stretched between two 200-foot towers, and they decided it would be much easier to retune their electronics to match the array than it would be to build a new antenna. They dropped their operating frequency from sixty to twenty-eight MHz, yielding a wavelength of about ten meters. The transmitter was connected to the curtain array, and the receiver was put in a shack on top of a nearby building. The receiving antenna was suspended between two posts on the roof of the same building. [1, pp.92-94]

On 28 April 1936 they fired up the equipment, and immediately saw echoes from airplanes flying nearby. They could detect the planes out to more than two miles, and the returned echo pulses were sharp and distinct. The following day they drove their transmitter harder, by increasing plate voltage to 5,000 volts. Now an airplane could be tracked out to eight miles. It was time to show their bosses. On 6 May, Hoyt Taylor, Laboratory CO CAPT H. M. Cooley, and other senior lab personnel watched as Page and Guthrie tracked a plane out to seventeen miles. On 10 June the Bureau of Engineering sent CDR Wilbur J. Ruble and LT J. B. Dow of the Bureau’s Radio and Sound Division to watch a demonstration. Two days later a letter arrived from Bureau Chief RADM Harold G. Bowen. It stated that the radio detection program should be given highest priority and directed toward developing shipboard equipment “based on the use of a manual and motor driven beam” that could both detect objects and show their range. (Author’s note: since the Admiral’s letter mentioned a moveable beam it seems reasonable the Bureau also wanted target bearing measured.) The letter also directed that the project be classified secret and knowledge thereof be held to an absolute minimum number of people. [1, pp.94-96]


The Antenna Duplexer

Next, CAPT Cooley arranged a radio detection demonstrations for the Chief of Naval Operations (CNO), an Assistant Secretary of the Navy, the Commander in Chief of the U.S. Fleet and other senior naval officials. By June 1936 three more engineers had been added to the project to begin turning a fragile laboratory device into a robust piece of shipboard equipment. Page realized he would have to select a higher operating frequency to match the smaller rotating antennas that could be fitted aboard a ship. The higher frequency would also make the radio beam more focused and achieve improved target resolution. The highest workable frequency was dictated by the abilities of vacuum tubes on the market, and the best they could do with high power output was about 200 megahertz. This would result in a wavelength of about 1.5 meters. They also started investigating a number of antenna configurations that might be suitable for shipboard use. Radio Division Director Hoyt Taylor did not like the idea of having to burden a ship with two antennas; one for transmitting and one for receiving. He directed that Page must find a way to use only one antenna. Page’s first reaction was it was impossible; a receiver could not be designed to absorb that much power from the transmitter. In the end, he created an electronic switching circuit that isolated the receiver during the time the outgoing pulse was being transmitted. He called it the antenna duplexer. [23, p.175] [1, pp.98-100]

The radio detection team completed the new 200 MHz set and the duplexer switch in ten weeks and were ready to test both on 22 July 1936. The duplexer was a great success, and reflections from nearby buildings and towers were easily detected, but echoes from aircraft were weak. Page realized he needed more transmitter power, and he devised circuits using four transmitting tubes in place of one. By early 1937 the configuration was working in the transmitter, but Page was given little time to continue improving his equipment. It seems that senior naval officers were usually very conservative when it comes to accepting new equipment and devices; they had a reputation for resisting innovation. But in the case of radio location they seemed to have sensed its potential value early on. So much so that in 1936, the CNO, Vice Admiral A. J. Hepburn, directed the Chief of the Bureau of Engineering to arrange for at-sea testing of the new technology as soon as possible. [1, pp100-101]

First Test at Sea

Early in 1937 Page was told to start preparing his system for sea trials. He felt the order was premature, and he was concerned a poor showing could doom the project, but he had no choice. He had just a little time to “ruggedize” his equipment and build some weather-tight enclosures so the equipment could be placed on the deck near the antenna. The equipment was to be installed in the destroyer Leary tied up at the nearby Washington Navy Yard. A month of the ship’s schedule was set aside for the trials. The lab would have a week to install and prepare; a week at sea; a week in port to evaluate results and fine tune; and then a final week at sea. A five-inch gun mount made a convenient antenna base that could be trained and elevated. Various planar and Yagi fishbone antennas were clamped to the gun barrel, and airplane detections were made at about fifteen miles with the planar arrays, and shorter distances with the Yagi antennas. The results were disappointing and the reason was obvious; the transmitter needed more power. [1, pp.101-102] [23, p.180]

Testing of the Naval Research Laboratory’s prototype 200 mHz radio location set aboard the destroyer Leary in April 1937. Various types of antennas were tried; clamped to the handy five-inch gun barrel in the background. It offered convenient antenna training and elevation. Naval History and Heritage Command Photo # NH 51280

Page and his team made a thorough search of all available transmitting tubes and found a potential solution. The Eitel-McCullough Company made a tube, called the Eimac 150TH, for radio amateurs who were known for pushing their equipment to extremes. Page found that the tube could withstand short radio frequency pulses at plate voltages up to 15,000V. He also found that his four-tube ring configuration could be extended to any even number of tubes and he tried a six-tube configuration using the Eimac tube. They now had enough transmitter power for a 200 MHz set, and they set about fabricating a rotating directional antenna. Key components of the prototype antenna were a truck rear axle from a junk yard and a length of sewer pipe. The rear axle parts were to allow mounting the antenna so it could be tilted in elevation to measure target altitude. By February 1938 the developers were quite satisfied with the new system’s performance, but felt there were a number of possible improvements before committing to fleet production. But the Bureau of Engineering felt otherwise; they wanted work started immediately on a shipboard prototype. [1, pp.102-103]

On 24 February 1938, the Bureau ordered NRL to build a shipboard version of the 200 MHz set as soon as possible. The lab agreed on a 1 September completion date and a total cost, including salaries, of $25,000. The laboratory was accustomed to turning experimental equipment into shipboard equipment through a process of making the gear more rugged and reducing its size. One item however, the antenna, resisted “miniaturization” because it needed to be about seventeen-feet square to match the system’s wavelength, and it was going to weigh about 850 pounds. This meant it would be too large for anything but aircraft carriers, battleships, and cruisers. They did not have difficulty getting a test ship because there was already a volunteer. The commander of the Atlantic Squadron, Admiral A. W. Johnson, had been at previous radio location demonstrations, and volunteered his flagship, the Battleship New York, as the test platform. [1, pp.104-105]

Battleship New York Tests

The Bureau of Engineering designated the prototype set “XAF”, and NRL was ready to ship the equipment on 8 December. (Note: “X” and “CX” were prefixes assigned to electronic equipment developed by the Naval Research Lab, “A” probably stands for first in series and the meaning of “F” seems to be lost in the mists of time.) The NRL shops, guided by seven engineers and three draftsmen built the set, except for the antenna, in eight months. The Brewster Aeronautical Corporation built the antenna in accordance with the lab’s drawings, and total cost was almost exactly the originally estimated $25,000. Although the purpose of the large antenna mounted atop New York’s pilot house was secret, the sailors could not help noticing it, and to them it became the “flying bedspring.” A team of NRL engineers rode New York to the Caribbean in January 1939, and first determined that the XAF did not interfere with other on-board electronics. The XAF survived the shock of gunfire, high wind, and rain storms, and kept on running up to 24 hours a day. Ships could be tracked out to ten miles and airplanes out to forty-eight miles. Amazingly, it could follow fourteen-inch gun projectiles in flight. In simulated destroyer torpedo attacks, the set detected the attackers before they got within torpedo range. New York’s skipper wrote the following recommendation in his cruise report:

That [radio location] be installed at once on all [aircraft carriers] and as soon as possible on other vessels. I would make no reduction in size at the expense of range for the present, particularly for [carriers]. The device looks big, but really caused very little inconvenience. After all we can’t expect to get something for nothing. It is well worth the space it occupies.

[1, p.105, p.108, p.110]

The XAF radio location antenna installed aboard the battleship New York, late 1938. Naval History and Heritage Command photo # NH 77350

During the New York tests, Page noted that the need to measure target range off the A-scope and then to look away to get antenna bearing off another indicator slowed operators down and led to errors. He also realized that to get a complete picture of all ships and airplanes present within the system’s range called for manually stopping the antenna at each target, plotting range and bearing, and then moving on to the next target; a tedious and time consuming process. He felt there should be a better way of getting a map type presentation of the total target picture around the ship. When back at the laboratory he began research on a better type of radio location indicator. It would use a cathode ray tube in the same manner as the British plan position indicator (PPI), which has already been described but still unknown to Page. It would lead to a series of PPI indicators for the U.S. Navy. [23, p.193]

The XAF transmitter and receiver installed in New York. The A-scope is at top right. Naval History and Heritage Command photo # NH 105852

The CXAM Goes into Production

As a result of the New York tests, the Office of the Chief of Naval Operations (OPNAV) designated radio location as Special Project 1, and on 8 May 1939 the CNO directed the Bureau of Engineering to acquire from 10 to 20 XAF copies as soon as possible to install in fleet vessels for service test. The Bureau knew of only two contractors qualified to build the sets, RCA and Western Electric, and on 17 October 1939 RCA was awarded a contract for the first six sets, to be designated CXAM. (Note: the meaning of “CX” and “A” have previously been explained, and “M” most likely meant “radio transmitting and receiving equipment” from an early navy equipment designation system.) The first “engineering development” set was delivered in November 1939 and installed in the battleship California in the summer of 1940. The following five sets were for the carrier Yorktown, and heavy cruisers Chicago, Chester, Pensacola, and Northampton. Testing in these ships was encouraging and 14 more sets with minor improvements, and designated CXAM-1, were acquired for the carriers Lexington, Saratoga, Ranger, Enterprise, and Wasp, the battleships Texas, Pennsylvania, West Virginia, North Carolina, and Washington, the heavy cruiser Augusta, the light cruisers Albermarle and Cincinnati, and the seaplane tender Curtiss. Most would be installed before 7 December 1941. [1, pp.110-111] [50, p.40]

Click here to go to “The Beginnings of Naval Fighter Direction - Chapter 5 of Radar and the Fighter Directors.”