AIRPOWER Magazine
January, 1975  Vol. 5 No. 1

Reprinted with permission

The Falcons Of Freeman Field
Testing Japan's Eleventh Hour Entrants
by Jack Dean


Photo at Freeman Field

Editor's Note:
Located in the fertile farmland of Indiana, Freeman Field became the Air Force's special testing ground for captured enemy aircraft. To be sure, other facilities, both at home and overseas, also evaluated captured enemy equipment. In the case of Japanese planes, the first was at San Diego. A short time later, testing was removed to Anacostia Naval Air Station, near Washington, D.C. But as more and more Japanese aircraft were brought back from preliminary tests, usually performed in Australia, Freeman Field was expanded to handle the rapidly increasing influx, many of which were new and exotic to technical intelligence officers and mechanics who had been led to believe that every other Japanese aircraft was either a Zero fighter or a clumsy, heavily spatted bomber. By the end of WW II, every Japanese and German type had passed through Freeman Field, and for as long as two years after the close of WW II, these aircraft were still being evaluated, tabulated and chronicled. In fact, so many enemy types were in the air, that had they not known better, former service pilots would have sworn that the Japanese were again attacking the U.S. Unfortunately, far too many of them were also destroyed, although a few were stored for later museum and display purposes. Here then is a survey of what Allied Technical Intelligence discovered about Japan's late war aerial threat. A threat that might have very well materialized into a grim reality, were it not for the development of America's heavy, long range bomber, the B-29.

While the Japanese Navy was having Mitsubishi design a naval interceptor, the Army, while giving the go-ahead for lull protection of Nakajima's Ki 43 Oscar, a classic dogfighting aircraft, also had that same company come up with their concept of an interceptor for defense against bombers. The result was the Ki 44 Shoki (Demon) codenamed Tojo. A typical Nakajima product, but with maneuverability sacrificed for climb and speed, its fuselage was altered to accommodate a brutish engine which, in the last Ki 44-III model, had an output of 2000 hp. for take off. Bomber busting was the task of this pugnacious-looking bird and its 40mm cannon, one of which is shown projecting from the right wind leading edge, was tailormade for the job.

When Japan entered WW II against the United States, her Air Forces possessed the best naval fighter in being, the Mitsubishi A6M Zero, the best torpedo bomber, the Nakajima B5N2 Kate, a solid, long ranging twin engined naval bomber, the Mitsubishi G4M2 Betty, and a host of satisfactory auxiliary naval types, but Japan's Imperial Army Air Force, was not so advanced. Its most numerous fighter, the Ki-43 Oscar, while one of the most maneuverable aircraft ever conceived, suffered from lack of speed, lack of armament and a near total lack of pilot protection. Despite considerable revamping throughout the war, these problems were never fully solved and Japan was forced to rely on newer replacement types then on the drawing boards or under construction.

Unfortunately for the Japanese war planners, while they had shown the foresight to undertake the design and construction of improved replacements, they had overlooked the possibility that these replacement aircraft would have to be built under other than ideal conditions. Raw materials, gasoline and other vital ingredients so crucial to running a successful aviation industry, quickly became scarce or disappeared altogether. Industries were short of skilled labor, with many factory assembly lines being manned by women with no previous training. Then, too, a multiplicity of designs, many of them deadend and wasteful, added a further strain to Japan's already overburdened technological resources, and the coup de grace came in 1944, when her production facilities were burned up from the air in a series of heavy bombing raids which caused more destruction than the nuclear bombings of Hiroshima and Nagasaki.
With stockpiles of raw materials practically denuded; with her work force decimated and her major assembly plants a shambles, it was a wonder that Japan produced any new designs of consequence during the period 1943-45, but she did, and many of her eleventh hour engines of war were not only remarkable in their being produced at all, but surpassed, in several respects, the best the United States its arsenals safe from attack could achieve.
When the U.S. entered the war with Japan late in 1941, our production facilities were just gearing up for mass delivery. As such, we were forced to fight the first six months to one year with designs which had been originated in the mid and late 30s. At the same time, the Japanese were already operating equipment which had been tested a year or two later. Although their planes were not as sophisticated and did not provide for much pilot or crew safety, they performed much better than our own, and because they outnumbered us in the Pacific, the Japanese Naval and Army Air Forces ran wild for the first six months. After a period of indecisiveness, when the balance seesawed first one way, then another, it became apparent by the beginning of 1943 that Japan's arsenals would be hard put to match those of the U.S., to say nothing of outstripping her more powerful enemy. By the end of 1943, it was a foregone conclusion that Japan would lose the war. The only question was . . . could she conclude a peace treaty that would leave her home islands intact.

Having lost control of the seas, the Japanese, by 1943, were already concentrating on production for home defense against aerial attack. To give their planners credit, the Army had foreseen this possibility and, in 1939, when the Nakajima Ki 43 was accepted for production, Nakajima was given a contract for an interceptor, the Ki 44, which would trade maneuverability, the heretofore primary Japanese fighter prerequisite, for speed and climb. By the end of 1942 the Ki 44 (Tojo) began to reach Japanese Army Air Force units. Despite the fact that it was restricted in its aerobatics, particularly at high speeds, service pilots liked its heavy armament, its rate of climb and diving speed, and if one had to get very close to insure hitting the target, the Ki 44's 40mm cannon proved highly successful against heavy American bombers such as the B-29.
In building the Ki 44, Nakajima had, in fact, again altered their basic fighter design of 1936, the Ki 27 Nate, which, together with the Mitsubishi A5M Claude, were Japan's first low-wing monoplane fighters. While the Claude was to be eventually eclipsed by the A6M Zero, a totally new design, Nakajima was to refine their Nate for nearly a decade, gradually improving it over the years. If one takes Severky's P-35 fighter as the first of the line and ends with the Republic P-47N Thunderbolt, by filling in the intervening changes with the P-43 Lancer, and the P-47B Razorback, a similar parallel development can be traced in this country. Within two years of the rather pugnacious-looking Ki 27 Nate's debut, a slimmer successor, its fuselage silhouette generally drawn out, appeared. This was Nakajima's Ki 43, Oscar. More powerful and faster, with a cleaner canopy design, the longer fuselage of the Ki43, still retained the wings of the Ki 27. With the introduction of the Ki44, in 1942, a still more powerful engine dictated a thicker fuselage, while speed considerations changed the shape of the wings, but the Ki 44 retained a distinct resemblance to the Ki 43. By 1944, yet-another Nakajima fighter arrived on the scene to once again alter, but not appreciably change, the basic concept. Resembling a beefed up Ki 43, but without the Ki 44's rotund fuselage, and featuring wings and tail group very similar to those of the Ki 43 Oscar, the Ki 84 Gale was, perhaps, the finest Japanese fighter of WW II. Certainly, it saw the most combat of those types introduced late in the war. Designed in 1942, debuting in 1944, the Ki 84 was fast enough at 392 mph, possessed of exceptional range, 1,053 miles, heavy armament, two 12.7mm machine guns and a pair of 20mm cannons, afforded protection for the pilot, was very maneuverable, and for emergencies, was capable of 427 mph for short periods.

Well designed and brilliantly conceived, the career of the Gale notonly exhibited the competence of native Japanese technology and creative genius, but also demonstrates how resourceful, even if handicapped, Japan's wartime industries were by 1944. With aluminum in short supply, plans to manufacture the Ki 84 from wood were made, the plywood skin to be heavily lacquered to insure a smooth finish. Other variants were to be built partly of steel (cockpit sections, ribs, bulk-heads) and partly from wood, with sheet steel skinning. Although production Gales were plagued by inferior workmanship, due to the fact that unskilled workers, many of them teenagers, were putting them together, Nakajima still managed to deliver 3,382 production Ki 84s in only 17 months, a rather amazing accomplishment when one considers the heavy B-29 raids and the need for the dispersal of factories through-out the home islands.
Meanwhile, as the Japanese Army was revamping its fighter strength, switching to interceptors, the Japanese Navy, after being forced to continue with improved models of the basic Zero, was also looking for a new fighter. Designed by Jiro Horikoshi's Zero team, the Mitsubishi J2M Jack, languished on the drawing boards for nearly a year, while its creators refined the Zero. When introduced in 1942, the Jack, with its laminar-flow wing, the first on a Japanese aircraft, proved stable and steep climbing, but engine trouble hampered full scale production and although it was possibly the best bomber destroyer the Japanese possessed, it lost nearly two years due to design lag. (See Oct Wings, 1973).

As a result, the hardpressed Japanese Navy turned to Kawanishi to provide it with an alternative to the Zero. Kawanishi had developed the N1K1 floatplane fighter, codenamed Rex. Extremely advanced, it had been built in very limited numbers, but Kawanishi had other ideas for their design. If the A6M Zero could be turned into the A6M-N Rufe float plane fighter by Mitsubishi, why couldn't the Rex be converted into an orthodox fighter like the Zero? In December, 1941, Kawanishi presented this proposal and in only seven months, the new landplane fighter was flown.

Although only 1,400 N1K1-Js were built, codenamed George, due mainly to landing gear and engine problems, the George was probably the fastest, and most maneuverable Japanese naval fighter of the war. In the hands of a veteran pilot it was a formidable adversary, particularly the later models with four 20mm wing cannon. It possessed good range (1,375 miles) and although maximum speed was under 370 mph. its handling was sterling. Because the power of its Homare 21 engine fell off rapidly at high altitudes, it was of little use against high flying B-29s, but against F6Fs and F4Us in middle and low altitude dogfights, it was spectacular. In February 1945, a lone Japanese ace, Warrant Officer Kinsuke Muto, engaged a dozen F6F Hellcats, single-handed, while flying a George. He destroyed four and forced the others to break off.
These last minute heroics, however, were the beginning of the end of Japan's aviation industry. While her newer fighters proved too little and too late, Japan persisted in building twin-engined fighters such as the Ki 44 Nick, which, despite improvements, still was basically an obsolete heavy fighter on the order of the German Me 110, one that could not maneuver with its single-seat counterparts. The naval air program also continued down the same engineering cul de sac. As the war news became grimmer, Japan introduced yet another three man torpedo and/or dive bomber. Also doubling as reconnaissance machines, these newcomers, such as the Yokosuka D4Y Judy, were generally slower than enemy fighters and, due to a lack of escorts, suffered heavily, even though they had a top speed of 350 mph.

They were also assigned to carriers which, by the fall of 1944, had ceased to exist. Their bomb loads were comparatively light and, when compared to American successors to the veteran Douglas SBD, such as the BTD-1 and XSB2D-1, which were abandoned in favor of the single-seat AD Skyraider, were inferior in every respect. Yet they continued to be produced. Had they been in service in 1942, the tide of battle in the Pacific might have been reversed. However, the same could be said of American types like the P-38, P-51 and Skyraider.

In the final reckoning, Japan's war planners had been correct in their belief that Japan's technology would be able to keep pace with that of the leading aircraft manufacturing nations of the world. What they did not allow for, was the consequence of their combat forces being unable to maintain control of those raw material producing areas seized during the first months of the war. Add to this the unrelenting strategic bombing raids conducted by the U.S. during 1944-1945, which all but totally destroyed their war-making potential, and you have the real reason why the best known Japanese combat types were those dating from the time of Pearl Harbor. Thus, many of the Falcons of Freeman Field were forced to show off their skills, over the peaceful Indiana countryside, instead of the Pacific. And judging by their performance, it was probably a good thing that they did.


In flight over Southern Indiana

Kawasaki's Ki 45 Toryu (Dragon Slayer) codenamed Nick, represented an old design that survived on the Japanese Army Air Force rolls, mainly because its robust airframe could be put to a number of tasks, which had been unforeseen when Nick was designed in 1937. The U.S. and Britain learned early in WW II that twin-engined, multi-seat fighters and attack planes could not compete with single-engined fighters and, moreover, were not worth producing, unless they could carry substantial bomb loads. The British Mosquito was the only exception, but even that exception could carry prodigious weights at speed. The Nick's bomb carrying capability was not much better than a Zero fighter's. It was not particularly fast, ranging between 300 and 330 mph. in its various models. However, it did have room in its fuselage for heavy armament and when the B-29 bombers began coming over Japan at night, it was the only night interceptor available with a service ceiling over 30,000 ft. and a range of over 1,000 miles. It could also accommodate radar and, as a result, the night fighting role was thrust upon it. As an attack bomber it was a dud, but as a nightfighter, it proved a welcome surprise, much like the German Me 110. It is shown here being evaluated after the war and, below, during its combat life, with twin 20mm cannon installed to fire upward into the bellies of marauding B-29s. The Necks type starter arrangement in the last photo was typical of all Japanese Army types which were conceived during the period of 1937-1941. The last meaningful fighter interceptions of the war were conducted in a Nick by Goldin Isamu Kashide, who watched the bombings of Hiroshima and Nagasaki from one.


At Freeman Field


Indiana's White River below


Probably not at Freeman Field


Probably not at Freeman Field


NAKAJIMA KI 44II SHOKI (TOJO)
China, 1943
Speed: 376 mph at 17,000 ft.
Service Ceiling: 36,000 ft.
Range, clean, at fast cruise: 565 miles
Armament: Two 12.7mm machine guns, two 40mm wing cannon


Ki 43 Oscar

Development of Ki 44 Shoki (Tojo) interceptor (top) from basic lines of Ki 43 Oscar (middle photos) bears striking parallel to lineage of our own P-35/P-47 Thunderbolt, and took place roughly during the same time frame. Larger Nakajima 14 cylinder radial, provided 50 percent more power to what was basically the same airframe, thus the enlarged and more rotund forward cowling, which in profile reminds one of the early P-43 Lancer. Although Tojo was not particularly maneuverable, when compared to orthodox Japanese fighters, it was the best climbing plane in the Army Air Force and its heavy armament made it that arm's most efficient bomber attacker, provided it could get close enough to fire the 20 rounds from its two 40mm, low velocity cannon. Two aircraft at bottom of page are training Ki 44s with old style tubular gunsight. In combat, a reflector sight was employed.


At Freeman Field


At Freeman Field


At Freeman Field

May be at Freeman Field.  The B-32 was known to have been at Freeman

If the Ki-44 was an interim Nakajima design to supply an interceptor for the Japanese Air Force, the Ki-85 Hayate or Gale, was the best Japanese fighter plane of the war and a true all around performer, ranking with the P-51 Mustang. Debuting in the spring of 1944, first in China and then in the Philippines, the Frank, as it was codenamed, fought a brilliant delaying action. Although vastly outnumbered in the Philippine campaign, more Ki 84s were lost to faulty hydraulic systems than to enemy air action. With a top emergency speed exceeding 400 mph., good armor plate, and up to four 20mm cannon, the Hayate also was possessed of exceptional range (over 1,000 miles) and could climb to 16,500 ft. in just under six minutes. Photo at top of page shows a captured Frank being tested by the Technical Air Intelligence Unit, Southwest Pacific Area. At middle left is one of the first Service Test batch at the factory. Next to it is a newly captured Frank, with a B-32 Convair Dominator behind it, while at the bottom is a lovingly restored Ki 84-la, which once belonged to the Ontario, California, Air Museum.


1. Nakajima B6N2 Jill. Although a satisfactory performer, Jill's size and high landing speed limited it to use on large carriers only. Although very similar to its predecessor, Kate, two years of teething trouble delayed Jill's combat debut until mid 1943, by which time attack aviation had gone far beyond its limited state of the art.

2 & 3. Two views of Jill, both showing its most vulnerable aspects. Four-bladed prop chewing into opponents, was its only means of frontal defense, since Jill had no forward firing armament. Although a ventral gun was installed on some models to fire to the rear, direct attacks from underneath could also be made with impunity, and with disastrous results, when wingroot tanks were exposed, as in this photo.

Aichi-Yokosuka D4Y3 Judy, also known as Judy 33, in flight over Freeman Field. Although fast, with a top speed of 360 mph., the Judy was badly handled during her first important combat debut at the Battle of the Philippine Sea in June, 1944. Caught far from the target carriers by energetic swarms of patrolling Hellcats, which swept aside the Judy's escorts, she was shot down in droves.
5. Big and commodious, yet weighing just over 5,500 lbs. empty, Judy had phenomenal range, nearly 2,400 miles in the liquid-cooled engined D4Y1, 1,800 miles in the -3 radial engined version, with tanks. Nevertheless, bomber was obsolete. Her payload was light: 700 lbs. She carried three crew members; she could not protect herself from attack, and represented the type of plane U.S. Navy planners had abandoned by 1942.

 


Over Freeman Field


At Freeman Field


At Freeman Field

6. Fast, maneuverable, fully protected and heavily armed, the NIKI Kawanishi Shiden (Violet Lightning) codenamed George, was the best Japanese naval fighter of the war. Almost as maneuverable as the Zero, it was much faster, carried four 20mm cannon, and did not fall apart or explode after one or two strikes.

7. In one of WW II's most unlikely non-sequitors, George was developed from a fine, but superfluous floatplane fighter, Kawanishi's N1K Kyofu (Mighty Wind) codenamed Rex. It had been designed to provide air cover for Japanese landing forces, but although fast (300 mph. at 18,000 ft.) a floatplane fighter was no match for single engined enemy fighters operating without the drag from one large main, and two outboard floats.

8. Thus, when the A6M Zero began to approach the end of its stretch, it was decided, in an alternate type of solution, to reconfigure the Rex into a landplane fighter. Within seven months, the X-1 prototype (shown here) rolled off assembly lines, its distinguishing feature the absence of an airscoop below the cowling.

9. The prototype appeared in December, 1942. By the time protection aircraft had their bugs worked out, it was late 1943. Never intended to fly from a carrier deck, the commitment of the Japanese Navy to the building of 1,435 Georges, admitted that it would now go over to the defensive. Troubles with the Nakajima Homare engine delayed really full scale production, and when these were finally ironed out, B-29 raids crippled the two major plants turning out the Shiden. For the U.S. it proved a blessing. With a speed of 370 mph. capable of hauling 2,000 lb. of bombs, or rockets, a battery of wing cannon, and brilliant handling qualities, the N1K1 shown here at the Air Force Museum, was a deadly competitor.

 

 

Successor to the B5N Kate, was Nakajima's B6N Jill. Basically the same aircraft, but with a much more powerful engine, Jill may have been 50 mph faster, but with puny defensive armament, she was a set up for Hellcat and Corsair fighters. Designed in 1939, the aircraft did not go aboard carriers until 1943 due to a rash of landing accidents involving weak arresting hooks.

Greatly influenced by the U.S. Army's A-18 twin engined attack plane,
Kawasaki Ki 45 Nick, a two-seat, heavy attack fighter was in production for more than five years, after a long and frustrating testing phase. Designed in 1937 as an answer to France's Potez 63 series and Germany's 110, Nick ended the war as a heavily armed nightfighter against B-29s.


At Freeman Field

One of Japan's fastest attack aircraft, the carrier based D4Y Judy, designed by Yokosuka Naval Air Factory, but built by Aichi, was the successor to Aichi's own Val dive bomber. With a top speed of 350 mph. in the D4Y3 model shown here, Judy was a swift striking machine, but a low normal bombload of only 685 Ibs, light armament, and lack of fighter protection hampered its career, which began inauspiciously when two pre-production inline engined D4Y1-Cs went down with the carrier Soryu at Midway.

Mitsubishi's J2M Jack, was Imperial Navy's first try at a modern, land based interceptor that sacrificed some maneuverability for climb and speed. An adjunct, rather than a replacement for the Zero, it lost a great deal of time between design and production, and illustrated quite unequivocally, that Japan realized, as early as 1940, that it might be forced to fight a defensive war.

Page last revised 01/10/2007