Category Archives: Aircraft

VEF I-16 Front right

VEF I-16 Light Fighter Aircraft

By William Pearce

VEF (Valsts Elektrotehniskā Fabrika or State Electro-Technical Factory, often spelled Valsts Elektrotechniskā Fabrika) was a large industrial manufacturer founded in Latvia in 1919. Kārlis Irbītis was an engineer at VEF during the 1930s and 1940s. Irbītis had designed aircraft since the 1920s and continued the practice in his spare time while employed at VEF. In 1935, Irbītis suggested that VEF should enter the aircraft manufacturing business, and VEF management was responsive.

VEF I-12 International Expo 1938

The VEF I-12 was a light sport plane and a follow-on to the I-11 aircraft. Powered by a 90 hp (67 kW), air-cooled, four-cylinder Cirrus Minor engine, the I-12 had a top speed of 143 mph (230 km/h).

VEF solicited a manufacturing contract from the Latvian government, but the request was denied. Frustrated, VEF decided to build Irbītis’ latest aircraft design, the I-11, as a private venture. The I-11 was the first in a series of fixed-gear, low-wing aircraft designed by Irbītis and built by VEF. The I-11 was a two-seat sport plane that first flew in late June 1936. Some aspects of the I-11 design were less than ideal, as aircraft components had to be made to fit out the workshop’s small door. After the I-11’s successful flight tests, an improved model was designed and designated I-12.

The I-12 was seen as a stepping stone to the design and construction of future military aircraft. VEF approved of Irbītis’ plan to build a military trainer after the I-12, and a light fighter would follow after construction of the trainer. The I-12 was first flown on 26 June 1937 and demonstrated good performance and handling. Ultimately, around 12 I-12 aircraft were built. Construction of the I-14 military trainer started in April 1937, and the aircraft made its first flight on 19 November 1937. The I-14 prototype was damaged beyond repair during an emergency landing on 23 April 1938. Undeterred, VEF authorized the design and construction of two new aircraft types: the I-15 trainer and the I-16 fighter.

VEF I-14

The VEF I-14 was developed as a military trainer and was powered by a 200 hp (149 kW), air-cooled, six-cylinder Menasco B6S Buccaneer engine. The I-14 had an estimated top speed of 186 mph (300 km/h), but a crash prevented the completion of flight tests.

Irbītis continued to improve his aircraft designs. The I-15 and I-16 shared a very similar layout and employed the same construction techniques. Manufacture of the I-15 started in the summer of 1938, and two aircraft were built. The I-15a had a wooden, fixed-pitch propeller and gun camera, but it had no armament. The I-15a was powered by an air-cooled, 200 hp (149 kW) de Havilland Gipsy Six series I engine. The I-15b was powered by a 210 hp (157 kW) Gipsy Six series II engine and used a metal, constant-speed propeller. The I-15b could accommodate a single 7.7 mm machine gun, and its cockpit was moved forward slightly to improve pilot visibility.

The I-15a first flew in April 1939. The first flight of the I-15b was delayed because of the late delivery of its constant-speed propeller but finally occurred around November 1939. The Latvian Aviation Regiment decided to purchase the I-15a and I-15b aircraft based on the favorable experience with four I-12 aircraft given to the Aizsargu Aviācija (Aviation Guard) by VEF in late 1938. The I-15a had a top speed of 186 mph (300 km/h), and the I-15b had a top speed of 205 mph (330 km/h). Both I-15 aircraft had successfully completed flight testing by the time Latvia was invaded by Soviet forces in June 1940; the invasion stopped further development.

VEF I-15a

The VEF I-15a on skis to enable flight testing during the Latvian winter. The I-15A carried the Latvian military serial number 190, while the I-15b carried 191.

Design work on the VEF I-16 light fighter began in late 1938. The aircraft was a single-seat, low-wing monoplane with fixed undercarriage. The landing gear was covered in streamlined fairings, and a retractable gear design was to be incorporated on a later model. The I-16 was comprised of a wooden structure covered in plywood and had fabric-covered control surfaces. Each wing accommodated a 10.6 gallon (40 L) fuel tank, and a single fuel tank in the fuselage held 58.1 gallons (220 L). The I-16 had two 7.7 mm machine guns mounted in the upper cowling in front of the cockpit. Provisions were made for an additional 7.7 mm machine gun to be mounted under each wing.

The I-16 was powered by a Sagitta I-SR engine built by Walter in Czechoslovakia. The Sagitta was an air-cooled, inverted V-12 engine that had a 4.65 in (118 mm) bore and a 5.51 in (140 mm) stroke. The supercharged engine displaced 1,121 cu in (18.4 L) and produced 520 hp (388 kW) at 12,467 ft (3,800 m). Air was fed into the supercharger via two scoops on the upper cowling. Engine exhaust was discharged through ejector stacks positioned at the bottom of the cowling. The shroud around the exhaust stacks also allowed cooling air to exit the cowling after passing through the cylinders’ fins. Additional cooling air exited via a vertical slit at the rear of the cowling. The engine turned a two-blade, wooden, fixed-pitch propeller, but a three-blade, metal, constant-speed propeller was planned for future use.

VEF I-16 construction

The VEF I-16 was a continuation of Kārlis Irbītis’ light, sleek monoplane design. The aircraft was the only monoplane fighter designed and built in Latvia. The wooden, fixed-pitch propeller was considered temporary. The cockpit canopy hinged open toward the right.

The I-16’s wingspan was 26 ft 11 in (8.2 m); its length was 23 ft 11 in (7.3 m); and its height was 8 ft 10 in (2.7 m). With the three-blade propeller, the aircraft had an estimated maximum speed of 286 mph (460 km/h) at 13,123 ft (4,000 m) and 249 mph (400 km/h) at sea level. The I-16 had an initial rate of climb of 2,187 fpm (11.1 m/s), and its ceiling was 26,247 ft (8,000 m). The aircraft had an empty weight of 2,425 lb (1,100 kg) and a loaded weight of 3,417 lb (1,550 kg). The I-16’s range was around 497 miles (800 km).

Construction on the I-16 continued through 1939, and the aircraft made its first flight in the spring of 1940 at Riga, Latvia. The pilot for the first flight was Konstantīns Reichmanis, and the I-16’s engine quit after about 20 minutes of flight time. Reichmanis managed to get the aircraft back on the ground without any damage. Poor fuel distribution was thought to have caused the engine trouble, as similar issues had been encountered during ground runs. Reichmanis praised the aircraft’s handling during his short flight. Changes were made to the I-16’s fuel system, and a few more flights were accomplished before testing was halted by the Soviet invasion.

VEF I-16 rear left

The completed I-16 with German markings during an engine runup in 1941. The two intake scoops for the engine are visible on the top of the cowling, with the left gun port immediately below. Armament was never installed in the aircraft, but VEF did have possession of the Browning machine guns until they were removed by Soviet occupying forces.

The Soviets expressed some interest in the VEF aircraft, and the I-15a, I-15b, and other aircraft were shipped to the Soviet Union in March 1941. The I-16 remained in Latvia to resolve the fuel distribution issues. Before the I-16 could be sent to Russia, the Germans attacked the Soviets and took over Latvia in June 1941. Irbītis escaped deportation, but many of his colleagues, including Reichmanis, were not so fortunate and disappeared into the work camps in Siberia. Under German occupation, the I-16 was returned to an airworthy status and carried the identification code AW+10. The aircraft made two test flights before it was taken over by the Luftwaffe. Some accounts list the aircraft as being part of a training school in Toruń (in occupied Poland) until 1942, but its final disposition is not known.

In late 1939, Irbītis began designing a new fighter, the I-19. With a wingspan of 36 ft 1 in (11 m) and a length of 29 ft 6 in (9 m), the I-19 was larger than the I-16. The I-19 also featured retractable gear, two machine guns in each wing, and a top speed of 404 mph (650 km/h). Irbītis persisted with the wooden construction, but the lack of a suitable power plant led him to consider building his own engine comprised of three V-12 engine sections, with each engine section based on a Ranger V-770. The cylinder banks of each engine section were spaced at 120 degrees. The resulting 36-cylinder engine employed three crankshafts and three superchargers. Irbītis estimated that the engine would displace around 2,200 cu in (36 L) and produce 1,450 hp (1,081 kW) at 3,250 rpm. However, Irbītis switched to an Allison V-1710 engine when more serious design work was undertaken. The I-19 never proceeded beyond the preliminary design phase. After World War II, Irbītis immigrated to Canada and helped develop the tiltwing Canadair CL-84 Dynavert.

VEF I-16 Front right

The sleek I-16 aircraft resembled similar light fighters developed in France (Caudron) and Italy (Ambrosini) during the same period. Note the streamlined fairings covering the gear.

Of Struggle and Flight: The History of Latvian Aviation by Kārlis Irbītis (1986)
Latvian Air Force 1918–1940 by Richard Humberstone (2000)

Republic XP-72 No 2 front

Republic XP-72 Super Thunderbolt / Ultrabolt Fighter

By William Pearce

In 1941, the Republic P-47 Thunderbolt had just entered production, and hundreds of the aircraft had been ordered. However, led by Alexander Kartveli, the design team at Republic stayed at the forefront of fighter development by incorporating new engines into new airframe designs. In July 1941, Republic submitted two new designs to the United States Army Air Force (AAF), the AP-18 and the AP-19. The AP-18 was a unique interceptor fighter powered by the Wright R-2160 Tornado engine. The AP-19 design was more conventional and was powered by the Pratt & Whitney (P&W) Wasp X (R-4360). Both engines were under development, but the R-2160 was anticipated first and received much interest from the AAF. As a result, the AP-18 design was ordered on December 1941 as the Republic XP-69.

Republic XP-72 No 1 roll out

The first Republic XP-72 prototype soon after being completed. The 14 ft 2 in (4.23 m) diameter Curtiss propeller was one of the largest used during World War II.

By 1943, the R-2160 engine had encountered major issues, but development of the R-4360 engine was steadily progressing. Republic felt the AP-19 design held more potential and wanted to end work on the XP-69. The AAF agreed, and the XP-69 project was cancelled on 11 May 1943. Two prototypes of the AP-19 design were ordered on 18 June 1943, and the aircraft was designated XP-72 (it also carried the “Materiel, Experimental” project designation MX-189). In addition, Republic felt the XP-72 was superior to the XP-47J, an interceptor derivative of the P-47, and asked that the second XP-47J prototype be cancelled. The AAF approved this request, and Republic focused on the XP-72.

The XP-72 was often called the Super Thunderbolt, or Superbolt, or Ultrabolt, and it benefitted from everything Republic had learned with the P-47 series, including the XP-47J. The XP-72 was essentially the wings, fuselage, and tail of a bubble-canopied P-47D combined with the close-fitting cowling used on the XP-47J. Of course, numerous internal changes made the XP-72’s resemblance to the P-47 a superficial oversimplification of the new aircraft’s design. Under the close-fitting cowling was a 28-cylinder P&W R-4360 engine. The engine drove a fan at the front of the cowling to assist cooling. A small cowl flap was positioned on each side of the cowling. The cowl flaps were automatic but could be manually controlled. At the cowl flaps, air exiting the cowling was combined with exhaust gases being expelled through ejector stacks and provided a small amount of thrust.

PW R-4360 remote supercharger

The Pratt & Whitney R-4360-13 and -19 engines had a remote, variable-speed, first-stage supercharger. This large supercharger was installed behind the XP-72’s cockpit and was connected to the engine via a fluid coupling.

The XP-72’s R-4360 engine used two-stage supercharging. The first stage was a mechanically-driven, variable-speed, remote supercharger positioned behind the cockpit, where the turbosupercharger was located on the P-47. To power the remote supercharger, a covered shaft extended from the unit, through the lower cockpit, and connected to the engine via a fluid coupling. The remote supercharger’s impeller was around 3 ft (.9 m) in diameter. The second stage was the standard supercharger that was integral with the engine.

A scoop positioned under the fuselage and in line with the wings leading edge split air into three ducts. The left and right ducts delivered air to oil coolers positioned on the bottom sides of the scoop. The outlet for each oil cooler was on the lower side of the scoop and about at the midpoint of its length. The larger, center duct fed air to the intake on the back of the remote supercharger and to the intercooler. The intercooler was positioned behind the remote supercharger. After being compressed in the supercharger, the air exited via two outlets and passed through the intercooler. After leaving the intercooler, the cooled induction air was split into two ducts and delivered to the R-4360’s downdraft intake, which is where the two ducts merged. The air then passed through the engine’s integral supercharger and into the engine’s cylinders. Cooling air that passed through the intercooler was discharged via an outlet in front of the tailwheel. No exhaust-driven turbosupercharger was installed on or planned for the XP-72 prototypes or the P-72 production aircraft.

Republic XP-72 No 1 left side

This side view of the first XP-72 illustrates the aircraft’s resemblance to the P-47 Thunderbolt. The notch just before the tailwheel is the air outlet from the intercooler. The serial number painted on the tail should actually be “336598” to conform to AAF guidelines. Neither XP-72 had the “correct” serial number painted on their tails.

Some sources state the XP-72 had strengthened landing gear compared to the P-47, while other sources say it was the same landing gear used on the P-47. The wings incorporated six .50-cal machine guns (three in each wing) with 267 rpg. However, the gun package could be changed to four 37-mm cannons (two in each wing). A hardpoint under each wing could carry a 150-gallon (568 L) drop tank or up to a 1,000 lb (454 kg) bomb. Just like on the P-47, an inlet for cabin air was located on the leading edge of the right wing. Dive recovery flaps were fitted under the wings, just behind the main gear wells.

The XP-72 was roughly the same size and weight as the P-47D but was more aerodynamic and possessed about 50% more power. The XP-72 aircraft had a 40 ft 11 in (12.47 m) wingspan, was 36 ft 8 in (11.18 m) long, and was 16 ft (4.88 m) tall. The aircraft had an empty weight of 11,375 lb (5,160 kg), a normal weight of 14,760 lb (6,695 kg), and a maximum takeoff weight of 17,492 lb (7,934 kg). The XP-72 had a top speed of 490 mph (789 km/h) at 25,000 ft (7,620 m) and an initial rate of climb of 5,280 fpm (26.8 m/s), decreasing to 3,550 fpm (18.0 m/s) at 25,000 ft (7,620 m). The aircraft could reach 20,000 ft (6,096 m) in under five minutes. The XP-72’s service ceiling was 42,000 ft (12,802 m). With 370 gallons (1,401 L) of internal fuel and two 150-gallon (568 L) drop tanks, the aircraft had a range of 1,200 miles (1,931 km) at a 300 mph (483 km/h) cruise speed.

Republic XP-72 No 1 right front

The XP-72 was a formidable aircraft with amazing performance. The scoop under the fuselage brought air to the oil coolers, intercooler, and supercharger. The duct in the wing was for cabin air. The close-fitting engine cowling was one of the best installations of an R-4360 and used an engine-driven fan to assist cooling.

The first XP-72 prototype (serial number 43-36598) was completed with a single-rotation propeller and a P&W R-4360-13 engine. The Curtiss Electric four-blade propeller was 14 ft 2 in (4.23 m) in diameter, which was one of the largest propellers used during World War II and was probably the largest propeller fitted to a fighter. The propeller left only 5 in (127 mm) of ground clearance, and the pilots employed three-point takeoffs and landings to make sure there were no propeller ground strikes. The R-4360-13 engine could accommodate the remote, variable-speed supercharger, but sources disagree regarding whether or not the remote supercharger was installed in the XP-72. The -13 engine produced 3,450 hp (2,573 kW) with the remote supercharger and 3,000 hp (2,237 kW) without it. The first XP-72 was finished on 29 January 1944. The aircraft’s first flight was made from Republic Field in Farmingdale, New York on 2 February 1944.

The second XP-72 prototype (serial number 43-36599) used a 13 ft 6 in (4.11 m) diameter, six-blade, contra-rotating Aeroproducts propeller. This propeller gave 9 in (229 mm) of ground clearance, and three-point takeoffs and landings were still the standard practice. Sources disagree on which dash number engine was used in the second prototype. Some sources claim a 3,000 hp (2,237 kW) R-4360-3 engine was used on the second XP-72. The -3 had a single-speed, single-stage (non-remote) supercharger and accommodated contra-rotating propellers, but the -3 engine used SAE #60-80 spline shafts. The Aeroproducts propeller used SAE #50-70 spline shafts, so it seems unlikely that the -3 engine was used. Many sources state the second XP-72 used a R-4360-13 engine, the same type fitted to the first prototype. The -13 engine was single-rotation with a SAE #60 spline shaft and could not accommodate contra-rotating propellers. However, it is possible that a contra-rotating gear case from another engine could have been fitted to the -13. The -8 (Douglas XTB2D) and -10 (Boeing XF8B) engines built for the Navy both used SAE #50-70 spline shafts. It is odd that another dash number was not assigned for such a change, but the -13 engine seems like the most likely candidate to have powered the second XP-72. Other sources propose that the engine was a R-4360-19 (see below), but there is no indication that any -19 engines were built.

Republic XP-72 No 2 front

With its six-blade, contra-rotating propellers, the second XP-72 is an impressive sight. Even with its hollow blades, the propeller still weighed around 765 lb (347 kg). Note the installed underwing pylons.

Regardless of the exact dash number, the second XP-72 was first flown on 26 June 1944. The contra-rotating propellers had a slight destabilizing effect on the aircraft, but the effect was manageable, and the aircraft still exhibited excellent flight characteristics. It is often reported that the second XP-72 was damaged beyond repair because of an emergency landing following an inflight fire. However, Ken Jernstedt, the pilot on that flight, has stated the incident never happened. In Jernstedt’s account, he was making a high-performance takeoff from Caldwell, New Jersey when an oil seal on the supercharger failed and caused a massive oil leak. Hot oil sprayed into the cockpit, on Jernstedt’s legs, and on the outside of the canopy. When Jernstedt brought the aircraft around for a quick landing, a Vought F4U Corsair suffering from an inflight fire crossed his path. Jernstedt had to veer around the Corsair to land the XP-72, which was damaged in the incident. While the aircraft could have been repaired, the XP-72 program ended soon after.

The Republic XP-72 was noted as exceptionally fast with amazing performance and for being a beautifully flying airplane. It is often reported that test pilot Carl Bellinger attained a speed of 480 mph (772 km/h) at sea level, but this speed was most likely recorded at altitude. Almost all sources indicate that both XP-72 prototypes achieved 490 mph (789 km/h) at 25,000 ft (7,620 m).

Republic XP-72 no 2 right side

The second XP-72 shortly after an engine run. Note that the tail of the aircraft is tied down. The air outlet from the oil cooler is visible on the lower fuselage, just under the wing’s trailing edge. Another outlet was positioned in the same spot on the opposite side of the aircraft.

An order for 100 P-72 aircraft was placed in late 1944. Production P-72s were to be powered by the P&W R-4360-19 engine, which used the Aeroproducts contra-rotating propeller and had an engine-driven, variable-speed, remote supercharger similar to the one used on the -13. The -19 engine was planned to provide 3,650 hp (2,722 kW) at sea level and 3,000 hp (2,237 kW) at 25,000 ft (7,620 m), allowing the P-72 to attain an estimated speed of 504 mph (811 km/h) at 25,000 ft (7,620 m). Further engine development resulting in 4,000 hp (2,983 kW) would reportedly enable the P-72 to reach a speed of 540 mph (869 km/h) at 25,000 ft (7,620 m). The 540 mph (869 km/h) speed seems a little optimistic. An upgraded wing, similar to that used on the P-47N, was to be applied to production P-72s. The P-47N wing held more fuel and increased the aircraft’s span by about 2 ft (.61 m). The speeds mentioned above were most likely estimated with the original, smaller wing.

As excellent as the P-72 may have been, the war situation indicated the aircraft was not needed, and the emergence of jet aircraft indicated that the P-72’s speed would soon be outclassed. The order for 100 P-72 aircraft was cancelled on 4 January 1945 so that Republic could focus on the P-84 Thunderjet fighter. On the day the P-72 was cancelled, the AAF ordered 100 P-84 jet aircraft (25 pre-production YP-84As and 75 production P-84Bs). The two XP-72 aircraft survived the war but did not last much longer. One airframe, without its engine, was given to a local (on Long Island, NY) chapter of the Air Scouts in August 1946. The other airframe was eventually scrapped.

Republic XP-72 No 2 right rear

With the war winding down and jet aircraft on the horizon, the XP-72 never entered production, despite the aircraft’s impressive performance. Production P-72 aircraft could have been the ultimate piston-engine fighter.

Correspondence with Tom Fey
U.S. Experimental and Prototype Aircraft Projects: Fighters 1939–1945 by Bill Norton (2008)
R-4360: Pratt & Whitney’s Major Miracle by Graham White (2006)
Republic’s P-47 Thunderbolt by Warren M. Bodie (1994)
American Secret Projects: Fighters, Bombers, and Attack Aircraft, 1937-1945 by Tony Buttler and Alan Griffith (2015)
US Army Air Force Fighters Part 2 by William Green and Gordon Swanborough (1978)

NAA XA2J Super Savage top

North American XA2J Super Savage Medium Bomber

By William Pearce

At the close of World War II, the United States Navy lacked the ability to carry out a nuclear strike. The nuclear bombs of the time were large and heavy, and no aircraft operating from an aircraft carrier could accommodate the bomb’s size and weight. The Navy did not want nuclear strikes to be the sole responsibility of the Army Air Force (AAF). In addition, the Navy felt that launching an attack with a medium-sized aircraft from a carrier that was hundreds of miles from the target offered advantages compared to large AAF bombers traveling thousands of miles to the target. On 13 August 1945, the Navy sponsored a design competition for a carrier-based, nuclear-strike aircraft. The competition was won by the North American AJ Savage.

NAA AJ Savage

Typical example of a production North American AJ-1 Savage, with its R-2800 engines on the wings and J33 jet in the rear fuselage. The intake for the jet was just before the vertical stabilizer and was closed when the jet was not in use.

First flown on 3 July 1948, the AJ Savage was a unique aircraft that spanned the gap between the piston-engine and jet-engine eras. The Savage was powered by two Pratt & Whitney R-2800 engines and a single Allison J33 turbojet that was mounted in the rear fuselage. The jet engine was used for takeoff and to make a final, high-speed dash to the target. In December 1947, before the AJ prototype had even flown, North American Aviation (NAA) proposed an improved version of the Savage that benefited from the continued advancement of turboprop engines. Designated NA-158 by the manufacturer, a mockup was inspected in September 1948, and the Navy ordered two examples and a static test airframe in October 1948—only three months after the AJ Savage’s first flight. The new aircraft was designated XA2J Super Savage, and the two prototypes ordered were given Navy Bureau of Aeronautics (BuAer) serial numbers 124439 and 124440.

Originally, the North American XA2J Super Savage was to be very different from the AJ Savage, but the jet engine in the rear fuselage would be retained. As the project moved through 1949, emphasis was placed on improving the XA2J’s deck performance over that of its predecessor. As a result, the XA2J became an entirely new aircraft but still resembled the AJ Savage. A mockup of the updated XA2J design, the NA-163, was inspected by the Navy in September 1949, and approval was given for NAA to begin construction.

NAA XA2J Super Savage Apr 1949

Concept drawing of the XA2J Super Savage from April 1949. Note how the aircraft bears little resemblance to the AJ Savage. The intake for the jet engine can be seen just before the vertical stabilizer. The pilot sat alone under the canopy, and the co-pilot/bombardier and gunner sat in the fuselage, behind and below the pilot.

The XA2J had the same basic configuration as its predecessor but was a larger aircraft overall. The Super Savage was of all metal construction and utilized tricycle landing gear. The high-mounted, straight wing was equipped with a drooping leading edge and large trailing edge flaps. To be brought below deck on a carrier, the aircraft’s wings and tail folded hydraulically. The pressurized cockpit accommodated the three-man crew, which consisted of a pilot, a co-pilot/bombardier, and a gunner. The pilot and co-pilot/bombardier sat side-by-side, and the rear-facing gunner sat behind them. Cockpit entry was via a side door, and an escape chute provided emergency egress out of the bottom of the aircraft. The co-pilot/bombardier was responsible for the up to 10,500 lb (4,763 kg) of bombs stored in a large, internal bomb bay. The gunner managed the radar-equipped tail turret with its two 20 mm cannons and 1,000 rpg. The defensive armament was never fitted to the prototype.

The XA2J did away with the mixed propeller and jet propulsion of the earlier AJ Savage; instead, it relied on two wing-mounted Allison T40 turboprop engines. The T40 engine was made up of two Allison T38 engines positioned side-by-side and coupled to a common gear reduction for contra-rotating propellers. Either T38 power section could be decoupled from the gear reduction, and the remaining engine could drive the complete contra-rotating propeller unit. The engine produced 5,332 hp (3,976 kW) and 1,225 lbf (4.7 kN) of thrust, for a combined output equivalent to 5,850 hp (4,362 kW). The Aeroproducts propellers used on the XA2J had six-blades and were 15 ft (4.57 m) in diameter.

NAA XA2J Super Savage ground

The XA2J Super Savage as built only had turboprop engines. In this image, the wide propellers installed on the aircraft have different cuff styles. Markings on the propeller installed on the right engine would seem to indicate that the propeller (rounded cuff) is being tested. Note the cockpit entry side door and open bomb bay doors.

The Super Savage had a 71 ft 6 in (21.8 m) wingspan and was 70 ft 3 in (21.4 m) long and 24 ft 2 in (7.4 m) tall. Folded, the wingspan dropped to 46 ft (14 m), and height decreased to 16 ft (4.9 m). The aircraft had an empty weight of 35,354 lb (16,036 kg) and a maximum takeoff weight of 61,170 lb (27,746 kg). Two fuel tanks at each wing root and two fuselage fuel tanks gave the aircraft a total fuel capacity of 2,620 gallons (9,918 L). The XA2J’s estimated top speed was 451 mph (726 km/h) at 24,000 ft (7,315 m), and its cruise speed was 400 mph (644 km/h). The aircraft had a ceiling of 37,500 ft (11,430 m) and a combat range of 2,180 miles (3,508 km) with an 8,000 lb (3,629 kg) bomb load.

NAA believed that the Super Savage airframe could be more than just a carrier-based medium bomber. The company developed designs in which various equipment packages could be installed in the aircraft’s bomb bay. The XA2J could be changed into a photo-recon platform with the installation of a camera package. Or the aircraft could become a tanker once it was outfitted with a 1,400 gallon (5,300 L) fuel tank in the bomb bay and a probe-and-drogue refueling system. A target tug system was also designed.

NAA XA2J Super Savage top

The Super Savage over the desert of California. The Allison T40 engine created trouble for every aircraft in which it was installed. The jet exhaust divider between the T38 engine sections can just be seen at the rear of the engine nacelle. Both propellers installed on the aircraft have square cuffs.

Construction of the first XA2J Super Savage prototype (BuAer 124439) began in late 1949 and progressed rapidly. However, Allison experienced massive technological problems developing the T40 engines, and they were not delivered until late 1951. The XA2J finally made its first flight on 4 January 1952 and was piloted by Robert Baker. The aircraft took off from Los Angeles International Airport and was ferried to Edwards Air Force Base (Edwards) for testing. By the time of the XA2J’s first flight, superior aircraft designs, namely the Douglas A3D (A-3) Skywarrior, were nearing completion. In addition, Allison never solved all of the T40’s issues, and the engines were limited to 5,035 hp (3,755 kW).

Testing at Edwards revealed some difficulties with the Super Savage. All aircraft powered by the complex T40 experienced numerous power plant failures, and the XA2J was no exception. The Super Savage was around 4,000 lb (1,814 kg) overweight and was never tested to its full potential. The highest speed obtained during testing was just over 400 mph (644 km/h). Even the aircraft’s estimated performance did not offer a significant advantage over that of the AJ Savage already in service. The XA2J project was cancelled in mid-1953, and the second prototype (BuAer 124440) was never completed.

NAA XA2J Super Savage in flight

The Super Savage had an aggressive appearance that gave the impression that the aircraft could live up to its name. However, it was outclassed by the Douglas A3D (A-3) Skywarrior and had performance on par with the AJ Savage it was intended to replace.

North American Aircraft 1934-1999 Volume 2 by Kevin Thompson (1999)
Aircraft Descriptive Data for North American XA2J-1 (June 1953)
American Attack Aircraft Since 1926 by E.R. Johnson (2008)
The Allison Engine Catalog 1915–2007 by John M. Leonard (2008)
“They didn’t quite… 5: Turbine-Driven Savage,” Air Pictorial Vol. 21 No. 12. (December 1959);all

Riout 102T wings up

Riout 102T Alérion Ornithopter

By William Pearce

French engineer René Louis Riout was interested in ornithopters—aircraft that used flapping wings to achieve flight. His first ornithopter, the DuBois-Riout, was originally built in 1913, but testing was delayed because of World War I. The aircraft never achieved sustained flight and was destroyed in an accident in 1916.

Riout 102T wing frame

The nearly-finished Riout 102T Alérion is just missing the fabric covering for its wings and tail. Note the wing structure and how the spars are mounted to the fuselage.

After the war, Riout designed a new ornithopter that had two sets of flapping wings. He continued to refine his ornithopter design, but no one was interested in producing such a machine. Riout worked for a few other companies, including a time with Société des Avions Bernard (Bernard Aircraft Company) from 1927 to 1933. While at Bernard, Riout was involved with their Schneider Trophy racer projects.

In 1933, Riout presented his ornithopter designs and research to the Service Technique de l’Aéronautique (STAé or Technical Service of Aeronautics). Riout’s presentation included designs and models of two- and four-wing ornithopters. The models weighed 3.5 and 17.6 oz (100 and 500 g) and performed flights up to 328 ft (100 m). As a result of these tests, STAé ordered a 1/5-scale model with wings powered by an electric motor.

Riout 102T wings up

Completed, the Riout 102T ornithopter resembled a dragonfly. An engine cylinder and its exhaust stack can be seen behind the rear wing. Note the enclosed cockpit; the rear section slides forward for entry.

The 1/5-scale model was built in 1934. From 11 November 1934 to 1 February 1935, the model underwent 200 hours of testing in the wind tunnel at Issy-les-Moulineaux, near Paris, France. The successful tests established the feasibility of Riout’s design and indicated the ornithopter would be capable of 124 mph (200 km/h) if it were powered by a 90 hp (67 kW) engine. Based on the test results, STAé ordered a full-scale ornithopter to be built and tested in the wind tunnel for research purposes. On 23 April 1937, Riout was awarded a contract for the construction of an ornithopter prototype.

The ornithopter was designated Riout 102T Alérion. The word alérion, or avalerion, is the name of a mythical bird about the size of an eagle. The single-place ornithopter had a cigar-shaped fuselage. Its frame was made of tubular-steel and skinned with aluminum. The enclosed cockpit occupied the nose of the aircraft. Two wheels on each side of the aircraft retracted into the fuselage sides. The landing gear had a 4 ft 3 in (1.3 m) track.

Behind the cockpit were two pairs of flapping wings. The two-spar wings had metal frames and were fabric-covered. A hinge at each spar mounted the wing to a large structure in the center of the fuselage. Immediately behind the wings, a 75 hp (56 kW) JAP (John Alfred Prestwich) overhead valve V-twin engine was installed with its cylinders exposed to the slipstream for air-cooling. The exact engine model has not been found, but the 61 cu in (996 cc) JAP 8/75 is a good fit. The 102T ornithopter had conventional vertical and horizontal stabilizers that were made of tubular steel frames and covered with fabric.

Riout 102T wind tunnel

On 12 April 1938, the wings of the 102T failed during a wind tunnel test. Stronger wings could have been designed and fitted, but the impractically of the ornithopter left little incentive to do so. The landing gear was removed for the tests. Note the engine cylinder behind the rear wing.

A drawing indicated the wings had 50 degrees of travel—40 degrees above horizontal and 10 degrees below. However, a detailed description of how the wings were flapped has not been found. The method appears to be somewhat similar to the system used on the DuBois-Riout ornithopter of 1913, in which the engine was geared to a crankshaft that ran between the wings. A connecting rod joined each wing to the crankshaft, but each wing was on a separate crankpin that was 180 degrees from the opposite wing. However, images of the 102T show both sets of wings in the up position, as well as one set of wings up and the other down. If a crankshaft was used for the wings, it must have employed clutches and separate sections for each pair of wings. It appears the standard operating configuration was for the wings to be on different strokes: one pair up and one pair down. Wing warping was used to achieve forward thrust, with the portion of the wing behind the rear spar moving.

The Riout 102T had a 26 ft 3 in (8.0 m) wingspan and was 21 ft (6.4 m) long. At its lowest position, the wing had 2 ft 2 in (.67 m) of ground clearance. At its highest point, the wingtip was 13 ft 5 in (4.1 m) above the ground. The aircraft’s tail was 8 ft 2 in (2.5 m) tall. The ornithopter weighed 1,102 lb (500 kg) empty and 1,389 lb (630 kg) fully loaded.

The aircraft was built in Courbevoie, at the company of coachbuilder Émile Tonnelline (often spelled Tonneline). Final assembly was completed in late 1937 by Bréguet (Société des Ateliers d’Aviation Louis Bréguet or Luis Bréguet Aviation Workshop) in Villacoublay. With its four wings and side-mounted landing gear, the completed ornithopter resembled a dragonfly.

Riout 102T frame

Restoration efforts provide a good view of the Riout 102T’s frame. Note how neatly the landing gear folded into the fuselage. The ornithopter’s aluminum body was saved, but the original wings were lost. (Shunn311 image via

After some preliminary testing, the 102T was moved to the wind tunnel at Chalais-Meudon in early 1938. First, tests lasting two minutes with the wings stationary were conducted. These tests were followed by wing flapping tests. Eventually, the ornithopter test sessions lasted a continuous 20 minutes, but all tests were conducted without the wings warping (providing thrust). It was noted that the engine was only producing around 60 hp (45 kW), but the tests were continued. On 12 April 1938, the 102T was in the wind tunnel undergoing a flapping test with a wind velocity of 81 mph (130 km/h). When the engine speed was increased to 4,500 rpm, one wing folded, quickly followed by the other three. The outer third of all the wings bent, with the right wings folding up and the left wings folding down. At the time of the mishap, the ornithopter had operated in the wind tunnel for around three hours and had satisfied initial stability tests.

Before the wings failed, Riout had notified the STAé of some modification he would like to make to the ornithopter. However, there was no interest to fund repairs or continue the project after the aircraft was damaged. The damaged wings were discarded, but the fuselage of the 102T was somehow preserved. Today, the Riout 102T Alérion is undergoing restoration and is on display at the Espace Air Passion Musée Régional de l’Air in Angers, France. While a few manned ornithopters flights have been made, the aircraft type has been generally unsuccessful.

Riout 102T frame restoration

The frame of the ornithopter consisted of small diameter steel tubes that were welded together. The aluminum wing supports may not be original. The Riout 102T is currently on display in the Espace Air Passion Musée Régional de l’Air. (Jean-Marie Rochat image via

“Avion à ailes battantes Riout 102T” by Christian Ravel Le Trait D’Union No 225 (January-February 2006)
Les Avions Breguet Vol. 2 by Henri Lacaze (2016)
“Flying Machine with Flapping Wings” US patent 1,009,692 by René Louis Riout (granted 21 November 1911)

Dubois Riout front wings down

DuBois-Riout Ornithopter

By William Pearce

When humans began to contemplate heavier-than-air flight, it was only natural to emulate birds. However, the complications of an ornithopter—using flapping wings to achieve flight—proved to be insurmountable. By 1900, most aviation pioneers focused their efforts on propellers and fixed wings; however, some persisted with the ornithopter.

Riout 1911 Patent

Drawings from René Riout’s US patent of 1911. Fig 1 shows the ornithopter design, which had a passing similarity to the aircraft built in 1913. Fig 2 and Fig 3 show the wing flapping mechanism. Fig 4 and Fig 5 show the wing in a gliding position. Fig 6 and Fig 7 show the wing warped for thrust.

In the early 1900s, French engineer René Louis Riout shifted his focus from automobiles to aviation. Initially, Riout designed models of gliders and propeller-driven aircraft, but his attention soon turned to ornithopters. By 1907, Riout was successfully flying his model ornithopter designs. In 1909, one of Riout’s models flew 164 ft (50 m) at an altitude of 10 ft (3 m). In 1910, his ornithopter model was flying 558 ft (170 m), and the distance expanded to 722 ft (220 m) in 1911.

In late 1910, Riout was granted French patent 419,140 for his flapping wing mechanism and ornithopter design. The same invention was patented in Great Britain (191117951) and the United States (1,009,692) in 1911. Riout’s patent described how power from an engine was geared at a reduced speed to a crankshaft. The crankshaft had two crankpins that were positioned 180 degrees apart. A connecting rod linked each crankpin to the pivoting mechanism of one wing. As the crankshaft turned and the crankpin moved to the horizontal position nearest the wing, the wing was moved to its highest position. As the crankpin moved to the horizontal position farthest from the wing, the wing moved to its lowest position. Thus, the up and down movement of the wing was controlled by the speed of the engine. The drive system incorporated a heavy flywheel to smooth out power pulses from the engine. For small aircraft, a heavy spring could be substituted for the flywheel.

Dubois Riout front

Front view of the DuBois-Riout ornithopter with the three-cylinder Viale engine. The engine cylinders can be seen protruding above the cowling. The wings are positioned around 20 degrees above horizontal. Note the quarter-turn belt drive for the wheel axle.

The patent details how the wings would warp as they moved. The upstroke was made in a neutral, gliding position. On the downstroke, the wing’s trailing edge would deflect up to provide thrust. Springs in the wing regulated the warp to match the power of the downstroke. A slow downstroke would result in the wing maintaining its glide form. The warp of the wing was greatest at the tip, tapering to very little warp at the root.

By 1913, Riout had partnered with Jean Marie DuBois, and a full-scale ornithopter was built. Exactly what role DuBois played in the creation of the ornithopter has not been found, but the resulting machine was known as the DuBois-Riout monoplane. The DuBois-Riout ornithopter had a slender, streamlined airframe that was made from tubular-steel and covered in fabric. A vertical stabilizer with a rudder protruded from below the fuselage. A horizontal stabilizer extended to the sides from the top of the fuselage and incorporated an elevator. The single-place cockpit was positioned between the ornithopter’s wings. The wings had a tubular-steel frame and were fabric-covered. The aircraft was supported by taildragger landing gear.

Dubois Riout front wings down

The ornithopter’s wings in the down position were about 20 degrees below horizontal, which was enough to make them contact the ground. This is why wing flapping would only be initiated after the aircraft was airborne, having been propelled to takeoff speed by the wheels. A shroud can be seen covering the top part of the drive belt.

The ornithopter was powered by a three-cylinder Viale Type A engine. The three cylinders were spaced 65 degrees apart in a fan configuration. The air-cooled engine had a 4.13 in (105 mm) bore and a 5.12 in (130 mm) stroke. Its total displacement was 206 cu in (3.4 L), and it produced 35 hp (26 kW) at 1,500 rpm. The engine was positioned in the nose of the ornithopter and encased in a cowling, but its cylinders protruded into the air stream for cooling. The engine drove a crankshaft to flap the wings, just like the patent described.

A major problem facing ornithopter designs was how to start the takeoff roll and gain enough forward speed to achieve flight. Via a belt, the DuBois-Riout used engine power to drive the main wheels during the takeoff run. The drive pulley was positioned behind the engine, and the follower pulley was positioned on the main wheels’ axle and perpendicular to the drive pulley. The follower pulley was offset to the left so that its front edge was directly below the left side of the drive pulley. As the belt came off the rear of the follower, it traveled to the right to reconnect with the right side of the drive pulley. The belt twisting 90 degrees enabled the longitudinal rotation of the engine’s crankshaft to be converted to transverse rotation for the aircraft’s wheels. Once the ornithopter was up to speed, the machine was glided off the ground. Via clutches, engine power was transferred from the pulley to the flapping wings for sustained flight. The DuBois-Riout ornithopter had a 34 ft 5 in (10.5 m) wingspan and a predicted max speed of 84 mph (135 km/h). The machine weighed 794 lb (360 kg).

Dubois Riout side

Side view of the DuBois-Riout ornithopter illustrates the vertical stabilizer under the fuselage and the elongated horizontal stabilizer. Note the large pulley on the wheel’s axle.

In late 1913 or early 1914, Riout initiated tests of the ornithopter but encountered issues with the engine. It is not clear if the engine was not running correctly or if more power was needed. Before the issues were resolved, Riout left to serve in World War I. In 1916, Riout was granted permission to restart tests on the ornithopter. A 50 hp (37 kW) Gnome-Rhône engine was acquired and installed in the aircraft. No information has been found as to what modifications were made to the ornithopter to handle the rotary engine or its gyroscopic torque. Reportedly, the ornithopter made it into the air but quickly came down hard and was wrecked. No one was injured in the mishap, but Riout needed to return to the war, and no further work was done on the ornithopter.

One might think that with the destruction of the DuBois-Riout machine and conventional aircraft proving their worth throughout World War I, Riout would move away from the ornithopter design. However, he persisted, but 20 years passed before his next ornithopter, the Riout 102T Alérion, was built.

Dubois Riout rear

The ornithopter’s rudder can be seen in this rear view. Note the large control wheel in the cockpit and the fabric gap between the wings and fuselage.

“Flying Machine with Flapping Wings” US patent 1,009,692 by René Louis Riout (granted 21 November 1911)
“French Monoplane with Flapping Wings” Popular Mechanics (February 1913)
French Aeroplanes Before the Great War by Leonard E. Opdycke (2004)
Rotary Wing Aircraft Handbooks and History Volume 11: Special Types of Rotary Wing Aircraft by Eugene K. Liberatore (1954)
“Avion à ailes battantes Riout 102T” by Christian Ravel Le Trait D’Union No 225 (January-February 2006)