The U.S. Navy Radar Training Facility at Jamestown, Rhode Island, made important contributions to winning the war in the Pacific and to making today’s air travel safer. The very existence of the station (known to those who worked there or with it as “Mickey”) was so secret that few Rhode Islanders knew of it. The habit of secrecy is so strong that today some documents in the National Archives dealing with it still are stamped with security restrictions, although not enforced. Ground for the facility was broken in 1942; it was razed in the 1970s. Only someone familiar with local Navy history could find it today.
After Pearl Harbor, Japanese air fleets could attack U.S. ships at night with impunity. U.S. Navy radar was inadequate for creating fighter interceptions, and no night fighters were capable of defending the fleet. The British had anticipated this problem in defending their island from the Germans, and had created a radar defense screen. The “Chain Home” system (a series of towers extending more than 100 miles along the English Channel coast) was sufficient to win the Battle of Britain. Radar operators on the ground coached night fighters by radio to within visual contact with enemy raiders. This system was inappropriate for defense of ships at sea.
Commander William E. G. Taylor was an experienced U.S. Navy aviator who volunteered for service in the Royal Air Force (RAF), and soon commanded the U.S.-volunteer Eagle Squadron in the Battle of Britain. During this tour of duty, Taylor flew numerous night interception missions and became thoroughly familiar with British techniques. He became convinced that airborne intercepts with radar were vital, and approached Navy brass with his ideas about radar on his return to the United States in 1941.
Meanwhile, British Prime Minister Winston Churchill was desperate for U.S. help. He sent the Tizard Mission in September 1940, which brought with it a new device that would make radar more powerful and compact: the magnetron. U.S. experts were amazed when this device was demonstrated. With President Franklin Roosevelt’s blessing, a committee, chaired by Dr. Vannevar Bush, established a national research laboratory at MIT—the Radiation Laboratory (RadLab). The RadLab was staffed by some of the brightest physicists from universities throughout the United States. This lab was the parent of Mickey, and it engaged in many kinds of radar research. Its initial product, in December 1940, was the first effective U.S. radar receiver.
In January 1941, the lab turned its attention to airborne- intercept radar. In May it demonstrated its first airborne radar set, mounted in an obsolete B-18 bomber; the set presented respectable images of Boston Harbor. Mickey was started by the RadLab as a civilian research facility and initially was dubbed the Spraycliff Laboratory. Management was in the hands of E. C. Pollard, a former Yale professor, and later Professor E. C. Byers (both were RadLab scientists). The property, located on 14 acres of Beavertail Point about a half-mile north of the lighthouse, was leased “for the duration” from the estate of Joseph Wharton.
All the while, the RAF was using airborne radar in twin- engine Beaufighters, manned by a pilot and a radar operator. The British ground equipment used by fighter directors determined the range and bearing of an enemy, but it was very inadequate for determining the enemy’s altitude. This was not a problem for the British, as the Germans habitually flew at certain altitudes. This scheme, for a number of reasons, was inappropriate for fleet aircraft carriers. In May 1941, the RadLab proposed to the Navy to develop a radar set, called Al, for single-seat fighters.
Taylor had persuaded the Navy to create a night air- defense system for the Pacific Fleet. In January 1942, Taylor reported for duty to Rear Admiral Gerald Bogan, Commander, Fleet Air Quonset. Taylor was assigned to create a night-fighter-training unit, dubbed Project Afirm. Mickey was merged as a supporting unit. Ownership of Mickey was transferred on an unknown date from the RadLab to the Navy as of 1 January 1942, but management remained with RadLab scientists. Mickey was to have an air-search radar manned by fighter directors. As the first fighter-director trainees, Lieutenants (junior grade) Hugh Jones and C. M. Martenson, did not arrive until March 1943, interceptions initially were ground-controlled by RadLab scientists. The facility had a dual role: to improve fleet night-fighter intercepts by training both fighter directors and pilots, and to research fleet and airborne night air defense. It also defended the East Coast from air attack, which was a real concern at the time. On 24 March 1942, construction of Mickey began in earnest. To maintain secrecy, the building was a facsimile of a New England farmhouse with a “water tank,” which concealed the radar antennas.
At the same time, the RadLab was experimenting with night interception with radar systems for landing aircraft at night. Professor Luis F. Alvarez tried a primitive system, called PGP, at a National Guard field in East Boston (now Logan Airport) by employing a pilot flying an amphibious J2F-3 scout plane. This experiment was discredited because it did not use equipment suitable for fighter aircraft. In March 1942, Alvarez used newly developed equipment to conduct an experimental ground-controlled landing (by radar) at Quonset Point. It was deemed a partial success, because the work was done with experimental radar managed by a RadLab scientist.
By this time Project Afirm, housed in a shed abutting a hanger, needed more room and Quonset Point was running out of space. In May 1942, an auxiliary airfield at Charlestown, about 20 miles south of Quonset, was started that would be devoted entirely to night fighters. By July, Mickey was reliably plotting aircraft 52 miles away (about 20 minutes flying time) using a radar developed for the Army. This was progress, but not good enough for effective air defense. In September the first AI radar sets for installation in aircraft were delivered to Quonset; they went in F4U aircraft. At about the same time, Mickey received the first S-series search radar, the SM.
With techniques and equipment at Mickey and Quonset assured, Project Afirm was ready for staff to train pilots and fighter directors. In November, the first six pilots reported for duty. Staff was recruited as well. Lieutenant William Slater Allen (my father), a long-time reserve officer and fighter pilot, was brought in from New York to be administrative officer. The Navy had no provision for such a position in a small command, but Taylor wanted no shadow cast on his authority by an executive officer. The Navy acquiesced. Scott Truesdale, another reserve officer, was installed as operations officer, but Mickey remained largely in the hands of RadLab personnel. Project Afirm was given a new name: Night Fighter Training Unit (NFTU).
In December, Alvarez returned to Quonset with new equipment (called the “two-truck unit”) and conducted a ground-controlled landing by Ensign Bruce Griffin. In January 1943, Byers of the RadLab reported to Cambridge that Mickey, using a CXBL search radar, was operating successfully with the cadre at Quonset. He noted in his report that a second storey had been added to the farmhouse and part of the interior had been remodeled to mimic the air combat center on new carriers, the space in which fighter directors would work. Byers also reported that Mickey was training Royal Navy personnel. The student had begun teaching the teacher.
In March 1943, the first Navy officers, Jones and Marten- son, reported to Mickey for training as fighter directors. On 21 April, Pollard, the RadLab supervisor of Mickey, visited the facility and reported back that it was running night intercepts successfully, with “enemy” aircraft often seen at 60-75 miles. This was not enough time to launch defensive fighters, but it was enough to achieve an intercept with an airborne combat air patrol. The altitude problem remained unresolved.
On 27 June, the law of unintended consequences led to another “first” for Mickey. Lieutenant Fred Dungan, flying an SNB aircraft, ran into trouble landing at Quonset at night and in dense fog. Going too low, his left wing hit the tops of some trees somewhere near Saunderstown (about four miles south of Quonset). Righting the plane, Dungan knew Quonset was out of the question and called Mickey for help. A fighter director coached Dungan and the damaged plane to Hartford, Connecticut, where Dungan landed successfully. This was the first U.S. cross-country flight managed by ground-control radar.
The training regimen for fighter directors and night- fighter pilots was set. Lieutenants Spero T. Constantine or Harold S. Durfee would take off in an SNB shortly after sundown and fly south or southeast about 200 miles past Block Island. On reaching this point, the SNB pilot would radio Mickey he was turning and coming back. Mickey would telephone Quonset (later Charlestown) that a “bogie” was approaching and fighters would be sent up. As soon as Mickey spotted the SNB with search radar it would start the intercept process. Mickey could distinguish the SNB from other aircraft that might chance to be in the area by IFF (Identification, Friend or Foe)—a device invented by the British, revealed by the Tizard Mission, and greatly improved by RadLab research.
The airborne fighters were directed to fly in an arc to approach the SNB from its rear. When within three or four miles, Mickey gave the signal “punch,” meaning the fighters were to use their own AI radar sets to close on the target. When the fighters sighted the target on their AI sets, they switched the radar to “gun aim” mode. When the AI indicated they were in position and within firing range, the fighters pulled away and radioed Mickey with “splash” (meaning the target had been shot down), and then returned to base. This was nearly a four-hour exercise, conducted every night of the week.
This was not as routine as it appeared. Constantine, Durfee, and the fighters were instructed to fly with no lights whatsoever, so the interceptors could not cheat. There was a real possibility that planes could have collided in midair. Everyone was in real danger; happily, that catastrophe never happened. On the other hand, the risks of night flying cost 48 pilots at Charlestown during the war years. In September, the RadLab and manufacturers delivered to Quonset an improved radar for fighters—the AIA—which remained basically unchanged as the AN/APS-6 I used during the Korean War. The AIA was installed on F6F fighter aircraft.
NFTU was ready to train fleet-defense pilots. Lieutenant Commanders W. J. Widhelm, a veteran of air combat at the Battle of Midway, and R. E. Harmer formed the core of Night Fighter Squadron (VF[N])-75. Twelve other pilots joined them. On 2 October 1943, six pilots of the squadron under Widhelm were deployed in F4U aircraft to a base at Munda, New Georgia, for night-fighter duty. They proved less effective than hoped, because the original plan—to deploy with a fighter director from Mickey with whom they had trained—was not followed. They had to make do with an Army officer who was unfamiliar with their capabilities and techniques.
In the summer of 1943, the first group of pilots specifically intended for fleet night-fighter squadron duty at sea was brought in. Lieutenant Commander Evan “Pete” Aurand was designated commander of VF(N)-76. Lieutenant Commander Robert Freeman was assigned to command VF(N)-77. Lieutenant Commander James Gray, another veteran of Midway, was assigned to form VF(N)-78. In November, Lieutenant Commander Turner F. Caldwell, of Guadalcanal fame, reported to NFTU to form VF(N)-79 (later designated VF[N]-41).
December witnessed the first ground-controlled approach landing entirely by Navy personnel at Quonset. This exercise also was unique in that the pilot was completely blind all the way to a full stop. The plane, an SNJ trainer, was flown by Harmer with all the glass masked by paper, with Dungan in the back seat as safety pilot. After landing, they seemed to be headed for a collision with a snowplow clearing the runway, but they stopped just short. The frightened plow operator dove head first into a snow bank.
In 1944, the rush was in full swing. On 16 January, Harmer took the rest of VF(N)-75 on board the Enterprise (CV-6), activating the first Navy night fighters in the fleet. This time they took a fighter director from Mickey with them. Later in January, Aurand took part of VF(N)- 76 on board the Bunker Hill (CV-17), while another detachment went on board the Yorktown (CV-10). Fighter director Martenson from Mickey went to sea with Aurand; Jones went with the group on the Yorktown. The rest of VF(N)-76 had to wait until March, when they went on board the Lexington (CV-16). With no one to spare, they were unable to get a fighter director from Mickey.
Mickey soon had a little brother: “Cousin.” A ground- controlled approach radar station was set up at Charlestown, located just west of the tennis courts there now. This was the result of the efforts of Alvarez, Aurand, and Harmer to create a blind landing system. Lieutenant Dick Thompson was brought from Mickey to be Cousin’s controller. In poor weather, pilots called Cousin by radio for landing help while over Block Island. Thompson or one of his colleagues would reply with directions as seen on search radar to bring the pilot down to 100 feet at the end of the runway.
Mickey was becoming less dependent on the RadLab, which was attenuating its management functions but still providing radar maintenance. Pollard, its chief Rad- Lab adviser, left in May to help prepare for D-Day. VF(N)-41 was deployed, with six Mickey fighter directors, to the Independence (CVL-22) in August 1944. This was the first time a night-fighter squadron was deployed as an intact unit.
Squadrons and fighter directors came, trained, and went to sea; 26 squadrons in all. In Hawaii, VF(N)-106 was integrated into Carrier Air Group (Night)-90 on board the Enterprise, the first carrier devoted entirely to night action. This was followed shortly by CAG(N)-52, with Mickey fighter directors under my father on board the Bon Homme Richard (CV-31), a second night carrier. Mickey’s initial purpose finally had reached fruition.
The excellence of Mickey’s training was demonstrated by an extraordinary event. In early fall 1944, Dungan and Lieutenant Johnny Dear were assigned a mission of harassing Chichi Jima. A Japanese fighter surprised them and Dungan and his plane were hit. He could not see because oil smeared his windshield, and then felt faint from loss of blood. In desperation, he called his ship. The fighter director, Hugh Jones, coached Dungan back to the ship. When deck crews and corpsmen went out to Dungan’s F6F, they found him unconscious from his wounds. Dungan had learned to obey the fighter director’s directions reflectively; indeed, they may well have penetrated his subconscious and brought him home safely. CBS radio dramatized this event in July 1945 on a national program.
Mickey continued training fighter directors for some months, then was converted to an entirely different training purpose. In 1949, the command of Mickey was transferred to Commander, Naval Training Station Newport, which signaled yet another purpose. It was used as a communications station for U.S. submarines worldwide. This technology became obsolete and Mickey sat idle until it was torn down. Only shards of typical Navy wall board tell its history. The buildings are gone and many of the men have cashed their last check, but the lessons learned keep military and civilian air travelers alike safer today.
Sources for this article include the authors interviews with many of those associated with Mickey, including William E. G. Taylor, C. M. Martenson, Fred Dungan, E. J. Hrube, R. E. Harmer, Harrison Wright, Spero T. Constantine, Harold Durfee, John Gilman, Russell Reiserer, Richard Thompson, Arthur M. Hayes, and L. G. Belisle. The surviving members of squadron VF(N)-76 will hold their reunion for five days starting on 8 October 2003 at Quonset Point, Rhode Island.