Cobb Napier-Railton Endurance Racer

By William Pearce

After John Rhodes Cobb made a small fortune as a fur broker, he started auto racing. Early in Cobb’s racing career, he served as a riding mechanic for Ernest Eldridge and his Mephistopheles racer, and he occasionally drove John Godfrey Parry-Thomas’ Babs racer on the Brooklands raceway in Surrey, England. In the late 1920s, Cobb had established himself as a capable, gentleman racer at Brooklands. His cars were often serviced by Thomas at his shop, located at the Brooklands raceway. The company was formed by Thomas and Ken Thomson, and renamed Thomson & Taylor in 1927, with Ken Taylor joining the firm after the death of Thomas during a Land Speed Record attempt.


John Cobb sits behind the wheel of the Napier-Railton at the Brooklands track. The exhaust system with mufflers was a requirement for Brooklands and did a good job of muting the engine. Note the vertical bars covering the radiator.

In late 1932, Cobb ordered a special car from Thomson & Taylor that would be able to set lap records at Brooklands as well as establish endurance records up to 24 hours, with sustained speeds in excess of 150 mph (241 km/h). Cobb had previously set the Outer Circuit lap record at Brooklands three times, and it was a record that was special to Cobb. Cobb and Thomson & Taylor gave the task of designing the car to Reid Antony Railton, head engineer. Railton knew he would need to come up with a design that was strong, durable, and reliable to stand up to the rough Brooklands track and the prolonged endurance runs. The car Railton designed would be known as the Napier-Railton.

In selecting an engine for the new racer, Railton wanted something that was powerful and reliable—a high-performance engine capable of running at high-power for 24 hours. Railton selected the normally aspirated Napier Lion XIA. The Lion was a 12-cylinder aircraft engine with three banks of four cylinders. The center bank extended vertically from the crankcase, with the left and right banks angled at 60 degrees from the center bank. Normally fitted with a propeller gear reduction, the Lion XIA for the Napier-Railton was modified by Napier with a special, elongated crankshaft and the removal of the gear reduction. As tested by Napier, the special Lion XIA produced 502 hp (374 kW) at 2,200 rpm, 564 hp (421 kW) at 2,350 rpm, and 590 hp (440 kW) at 2,700 rpm. The engine was fitted at the front of the car and mounted between the chassis’ two large frame rails, which were 10 in (254 mm) tall. Five cross members secured the car’s frame.


The chassis of the Napier-Railton with its Napier Lion engine and three-speed transmission. The two levers by the transmission were for the gear shift and driveshaft brake. The oil tank can be seen extending below the driveshaft and under what would become the cockpit.

Behind the engine was a single-plate clutch and the three-speed transmission. Since the car was to operate almost entirely at high speed, the first and second gears were much weaker than the robust third gear. This enabled the transmission to be smaller and lighter. The transmission drove the rear axle’s very strong differential, which had a 1.66 drive ratio. The forged rear axle housing was made of three sections: a center section that carried the differential, and left and right sections that carried the full-floating axle shafts. An oil sump, finned for cooling, was attached to the bottom of the axle’s center section. The car’s front and rear axles were positioned above the underslung frame rails, which enabled the car to have a low center of gravity. The suspension for the front axle used half-elliptical leaf springs, and the suspension for the rear axle used two sets of cantilever leaf springs on both sides of the car. The Napier-Railton was fitted with drum brakes on the rear axle and no brakes on the front axle. A driveshaft brake was operated by a hand lever and acted as a parking brake.

The chassis was covered by an aluminum body made by Gurney Nutting Ltd. The radiator at the front of the car was encased by the body, with a large opening for cooling air. At various times, the radiator opening was covered with vertical bars, a single bar, or no bars at all. The engine cowling had large humps for the left and right cylinder banks, a louvered top, and was secured by leather straps. Exhaust gases from each of the three cylinder banks were collected into separate manifolds, with the manifold for the center bank located on the left side of the car. An exhaust system consisting of a muffler and tailpipe extending to the rear of the car could be attached to each manifold. This system was used when the car competed at the Brooklands track. An undershield covered the bottom of the chassis.


At 6 ft 3 in and around 240 lb, Cobb was one of the few that could make the large Napier-Railton look almost normal-size by comparison. The leather straps that secured the engine cowling passed through the humps covering the left and right cylinder banks.

The cockpit was behind the engine and offset to the right, with the driver’s feet to the right of the transmission. The throttle pedal was in the center, with the brake pedal on the right and the clutch pedal on the left. A raised scuttle panel and windscreen protected the driver. At times, an enlarged scuttle and a shield to the cockpit’s right rear were added to protect the driver from a burst tire. In addition, a covered mirror was occasionally fitted to the scuttle left of the cockpit. An 18 US gallon (15 Imp gal / 68 L) oil tank was positioned to the left of the cockpit. The tank extended under the driveshaft and below the driver’s seat, and its underside was finned for cooling. Behind the cockpit, the body of the car tapered to a short wedge. Housed behind the driver was a 78 US gallon (65 Imp gal / 295 L) fuel tank. The Napier-Railton had a 10 ft 10 in (3.30 m) wheelbase, a track of 5 ft (1.52 m), and was 15 ft 6 in (4.72 m) long. The car weighed approximately 5,000 lb (2,268 kg). Various tire sizes ranging from 19 x 7 in (483 x 178 mm) to 35 x 6 in (889 x 152 mm) were used throughout the car’s career, with smaller tires used for acceleration and larger tires fitted for top speed. The wheels were mounted to the car with knock-off hubs. For long record runs, the throttle could be held open via a cable, and lights could be added to the car. Push starting was employed to bring the Napier-Railton’s Lion engine to life.

The newly completed Napier-Railton made its debut for the press on 6 June 1933. Minor testing by Cobb and Railton occurred before the debut, and serious testing was carried out in July. The car’s public debut was at the Brooklands track on 7 August 1933. Cobb set a Brooklands standing start lap record on the first lap of the Napier-Railton’s first race, covering the 2.75-mile (4.43-km) course at an average of 120.59 mph (194.07 km/h). The Napier-Railton went on to win the short race.


A builder and team photo of the Napier-Railton at Brooklands. Cobb is in the driver’s seat; Ken Taylor is on the far left; Ken Thomson is third from left; Reid Railton is fourth from left. Note the single vertical bar on the radiator housing.

Cobb then took the Napier-Railton to the 1.58-mile (2.55-km) speed ring at the Autodrome de Linas-Montlhéry track south of Paris, France for an attempt on the 24-hour record. Over 6 and 7 August 1933, American Ab Jenkins had established a new 24-hour record of 117.821 mph (189.615 km/h) driving a Pierce-Arrow V-12 at the Bonneville Salt Flats in Utah. This was the speed to beat. Cobb had previously arranged to use some equipment provided by George Eyston, a friend and fellow racer who was familiar with endurance runs at Montlhéry. The Napier-Railton’s exhaust mufflers were removed, and individual stacks were used. An angled shield was added to the scuttle left of the cockpit to block the exhaust flame glare from the center bank during night running. For the 24-hour attempt, Cobb’s co-drivers were Brian Lewis, Cyril Paul, and Tim Rose-Richards. Starting the record run on 2 October 1933, the car tore through its tires, and some difficulty was experienced with changing them. Push-starting the car after pit stops was also problematic. Rules stipulated that the car needed to travel forward under its own power. After shutting the car off during a pit stop, the crew needed to push it back some distance so that it could be pushed forward and started before it reached its original stopping point. Although several records were set with the Napier-Railton, including 200 miles (322 km) at 126.84 mph (204.13 km/h), 500 miles (805 km) at 123.27 mph (198.38 km/h), and six hours at 122.62 mph (197.34 km/h), the 24-hour attempt was abandoned after the radiator developed a leak and parts of the Montlhéry circuit began to break up under the car’s relentless pounding.

Back at Brooklands, Cobb and the repaired Napier-Railton set a new standing-start mile (1.6 km) record at an average of 102.52 mph (164.99 km/h) on 31 October 1933. On 4 November, the standing-start kilometer (.6 mi) record fell at 88.521 mph (142.461 km/h). Cobb was also timed covering 1 km (.6 mi) at 143.67 mph (231.21 km/h), the fastest speed recorded at Brooklands up to that point. On 2 April 1934, the Napier-Railton established a new Brooklands Outer Circuit lap record of 139.71 mph (224.84 km/h). Later that month, the Napier-Railton was back at Montlhéry for another 24-hour attempt. Cobb was supported by co-drivers Charles Brackenbury, Freddie Dixon, and Cyril Paul. Starting on 16 April, six hours passed at an average of 123.01 mph (197.97 km/h), 12 hours at 121.19 mph (195.04 km/h), and 2,000 miles (3,219 km) at 120.71 mph (194.26 km/h). On 17 April, after 19.5 hours had elapsed, Dixon lost control of the car, hit a guardrail and wound up in an infield ditch. Dixon was unharmed, but the Napier-Railton was damaged, and the record run was over. An AMR 33 light Army tank was required to pull the heavy car from the ditch. The Napier-Railton racer returned to the Thomson & Taylor works where it was repaired. At Brooklands on 6 August 1934, Cobb won the Championship Race and set a new Outer Circuit lap record at 140.93 mph (226.77 km/h).


Cobb takes flight as the Napier-Railton transitions over the River Wey to the Railway Straight and Brooklands. The bridge over the river created a bump that caused faster cars to become airborne, an indication of how Brooklands was a rough track. The image illustrates both the enlarged scuttle and the rear shield added to protect the driver. Note the bar-less radiator housing.

In mid-August 1934, Jenkins increased the 24-hour record to 127.229 mph (204.756 km/h). Cobb still wanted to set his own 24-hour record, and Jenkins’ success on the 10-mile (16-km) circular track in the wide expanses of the Salt Flats convinced Cobb to make the trip to Bonneville in mid-1935. For the Bonneville record attempt, a 120 US gallon (100 Imp gal / 455 L) fuel tank with two filler necks replaced the 78 US gallon (65 Imp gal / 295 L) tank, and the side panels covering the engine were removed for additional cooling. Like at Montlhéry, individual exhaust stacks were used.

Cobb, his team, and the Napier-Railton arrived at Bonneville in early July 1935. Ever the sportsman, Jenkins had a lot of equipment already setup on the Salt Flats and left it there for Cobb to use. On 12 July 1935, Cobb established a new 1-hour record at 152.70 mph (245.75 km/h) and a 100-mile (161 km) record at 152.95 mph (246.15 km) while testing the car on the salt. Backed by co-drivers Charlie Dodson and Rose-Richards, Cobb and the Napier-Railton set 16 records over 15 and 16 July 1935. The average speed for 500 miles (805 km) was 147.66 mph (237.64 km/h); 1,000 miles (1,609 km) was 144.93 mph (233.24 km/h); 12 hours was 139.84 mph (255.05 km/h); 2,000 miles (3,219 km) was 137.86 mph (221.86 km/h); 3,000 miles (4,828 km) was 134.56 mph (216.55 km/h); and 24 hours was 134.85 mph (217.02 km/h). In that 24-hour period, the Napier-Railton covered 3,236 miles (5,208 km).


The Napier-Railton in front of Gus F. Koehler’s Hudson dealership in Salt Lake City in 1935. The Hudson Motor Car Company provided courtesy vehicles to Cobb and his team. In its Bonneville configuration the Napier-Railton had a larger fuel tank, individual exhaust stacks, and its engine side covers removed. American and British flags were painted atop the radiator housing. The anti-glare shield appears in place on the left side of the car, but the windscreen is missing.

While Cobb achieved his goal, the record did not stand for long. At the end of August 1935, Jenkins increased the 24-hour record to 135.580 mph (218.195 km/h), covering 3,354 miles (5,398 km) in his new Duesenberg Special. In mid-September, the record was broken again at Bonneville, this time by George Eyston in Speed of the Wind, averaging 140.52 mph (226.15 km/h) and covering 3,372 miles (5,427 km). In three months, three groups of racers in three separate cars established three new 24-hour records, which varied by less than six mph.

After returning to England, Cobb and Rose-Richards won a 500-mile (805-km) race at Brooklands on 22 September 1935. The Napier-Railton averaged 121.28 mph (195.18 km/h), a speed that would not be bettered in a 500-mile (805-km) race until the 1949 running of the Indianapolis 500. On 7 October 1935, Cobb and the Napier-Railton set a final lap record at Brooklands of 143.44 mph (216.36 km/h). This speed was not exceeded before the track was partially torn up during World War II. During the attempt, Cobb covered 1 km (.6 mi) at 151.97 mph (244.57 km/h), the fastest speed recorded at Brooklands.


Cobb starting an attempt for the 1-hour record in 1936. The electric starting motor can be seen just before the rear tire. The driver would pull the lever that pressed the roller against the tire. The electric motor would then be turned on, driving the entire car forward. With a little bit of speed, the clutch could be let out, forcing the ever-reliable Lion engine to turn over and fire.

In mid-July 1936, Eyston increased his 24-hour record with an average speed of 149.096 mph (239.947 km/h), covering 3,578 miles (5,759 km). Cobb had already planned to make another attempt on the 24-hour record. By early September 1936, Cobb was back in Bonneville with Brackenbury, Johnny Hindmarsh, and Rose-Richards as his co-drivers. The Napier-Railton had a new external electric starting motor that, when engaged, drove the right rear tire to effectively push-start the car. Also, an exhaust manifold (without mufflers) was fitted to the center bank to reduce the glare from the flames at night. On 10 September, using the 12-mile (19-km) course, Cobb set a new 1-hour record at 167.69 mph (269.87 km/h) and covered 100 miles (161 km) at 168.59 mph (271.32 km/h). On 12 and 13 September, the Napier-Railton established four new records, including averaging 156.85 mph (252.43 km/h) over 1,000 miles (1,609 km), 149.27 mph (240.23 km/h) over 2,000 miles (3,219 km), and 150.16 mph (241.66 km/h) over 24 hours, covering 3,604 miles (5,800 km). Cobb’s new 24-hour record was less than one mph faster than the previous record set by Eyston; once again, the record did not stand for long. In late September 1936, Jenkins took back many of the records and averaged 153.823 mph (247.554 km/h) for 24 hours, covering 3,692 miles (5,942 km).

The Napier-Railton raced only at Brooklands in 1937. On 29 March, it won a race averaging 136.03 mph (218.92 km/h), the fastest race ever run at Brooklands. On 18 September 1937, Cobb, co-driver Oliver Bertram, and the Napier-Railton won a 500 km (311 mi) race averaging 127.05 mph (204.47 km/h). This was the last time the car was run on the track. Cobb retired from circle-track racing to focus attention on his Land Speed Record (LSR) car, the twin-Lion powered Railton. Eyston and Jenkins continued their duel for endurance records, and Eyston tried for absolute LSR records with his Thunderbolt car. The Napier-Railton was stored through World War II and acted as an LSR car for the 1951 film Pandora and the Flying Dutchman. Installed for the film were a streamlined radiator housing, a headrest behind the cockpit, and an elongated tail.


The Napier-Railton being utilized by the GQ Parachute Company to test aircraft braking parachutes. The pylon atop the rear of the car could automatically retract the parachute and store it for reuse. The streamlined nose was made for the 1951 film Pandora and the Flying Dutchman and was removed in the mid-1950s.

After Cobb’s death while attempting a water speed record in September 1952, the Napier-Railton was used by Geoffrey Quilter of the GQ Parachute Company. The car remained mostly as it had appeared in the movie, but a smaller fuel tank was fitted, and a parachute testing structure was mounted above the rear axle. To improve stopping, discs replaced the drum brakes on the car’s rear axle. Quilter used the car for a number of years to test aircraft braking parachutes. Eventually, the original radiator housing replaced the movie nose.

The Napier-Railton was purchased by Patrick Lindsay in 1961. Lindsay competed in various Vintage Sports Car Club meets and other events, and was clocked at 165 mph (266 km/h) in the Napier-Railton. After Lindsay passed, the Napier-Railton was acquired by Bob Roberts in 1971. The car was restored to a configuration similar to how it appeared while being raced at Brooklands by Cobb. After Robert’s death, the car was purchased by Victor Gauntlett in 1987 and was subsequently acquired at auction by a German collector in July 1991. Following a protracted three-year negotiation, the Napier-Railton returned to England under the ownership of Lukas Hüni in early 1997. Under an agreement with the Brookland Society, Hüni held the car until funds could be raised to purchase the Napier-Railton for the Brooklands Museum. The car’s purchase was finalized in December 1997, and the Napier-Railton was officially handed over to the Brooklands Museum on 6 May 1998. The Napier-Railton, still equipped with its original engine, is on display at the Brookland Museum and is occasionally run for special events. Over its career, the Napier-Railton set seven records at Brooklands, 11 records at Montlhéry, and 29 records at Bonneville.


The Napier-Railton in its current form enjoying some sun. The car has been mostly returned to how it appeared for its various runs at Brooklands and is occasionally run at special events. (Dave Rogers image via Wikimedia Commons)

Reid Railton: Man of Speed by Karl Ludvigsen (2018)
Brooklands Giants by Bill Boddy (2006)
The 1933 24-litre Napier-Railton, Profile Publications Number 28 by William Boddy (1966)
Napier: The First to Wear the Green by David Venables (1998)
The Fast Set by Charles Jennings (2004)
The John Cobb Story by S. C. H. Davis (1953)
Napier: Lions at Large 1916 – 2016 by Alan F. Vessey (2016)
– “King of Brooklands: The ex-John Cobb Napier-Railton Impressions” by Don Vorderman, Automobile Quarterly Volume IX, Number 1 (Fall 1976)


Curtiss XF14C Carrier-Based Fighter

By William Pearce

On 30 June 1941, the United States Navy, in preparation for the future of aerial combat, ordered prototypes of the Grumman F6F Hellcat carrier fighter and the F7F Tigercat heavy fighter. The Hellcat was intended to replace the F4F Wildcat and counter the Japanese Mitsubishi A6M Zero. The Tigercat was intended to out-perform and out-gun all other fighters. The Hellcat and Tigercat went on to serve with distinction for many years. Also on 30 June 1941, the Navy ordered two prototypes of the Curtiss XF14C.


The Curtiss XF14C-2 with its contra-rotating propellers and four 20 mm cannons appears as an imposing aircraft. However, its performance did not meet expectations. Note the stagger of the cannons and the glazed, rearward-sliding canopy.

Since 1939, the Navy had been supporting the development of the 2,300 hp (1,715 kW) Lycoming XH-2470 engine. The XH-2470 was a liquid-cooled, 24-cylinder engine in a vertical H configuration. The Navy’s support for the XH-2470 was unusual, as it had a long history of exclusively using air-cooled radial engines. In addition, the Navy had no applications for the engine until the XF14C was proposed as a high-performance fighter.

The Curtiss-Wright XF14C was designed at the company’s main facility in Buffalo, New York. The two XF14C-1 prototypes ordered were assigned Navy Bureau of Aeronautics numbers (BuNo) 03183 and 03184. Most sources state that the XF14C-1 was to be powered by the XH-2470-4 engine. Lycoming documents indicate that the -4 featured contra-rotating propellers. However, some sources state the XF14C-1 had a single rotation propeller that was 14 ft 2 in (4.32 m) in diameter. The XH-2470-2 used a single rotation propeller, but no sources have been found specifically stating that this was the engine for XF14C-1.

Regardless of the exact engine model and propellers, the XF14C-1 was an all-metal, low-wing aircraft with standard landing gear and a conventional layout. The gear was fully retractable, including the tail-wheel, and the main legs had a wide track. The arrestor tail hook extended from the extreme rear of the fuselage. The outer panels of the wings had around 7.5 degrees of dihedral and folded up for aircraft storage on an aircraft carrier. The fixed wing section had a flap along its trailing edge, and the folding section had a small flap on its inner trailing edge. The rest of the folding section had an aileron along its trailing edge. Just inboard of the wing-fold was the aircraft’s armament. Initially, each wing would house three .50-cal machine guns, but this was revised to two 20 mm cannons with 166 rounds per gun.


Side profile of the XF14C-2 illustrates the large exhaust pipe from the turbosupercharger under the aircraft. The inscription under the diving figure on the cowling reads “Coral Princess.” Note the large wheel covers and the retracted tail hook.

The XF14C-1 had a 46 ft (14.02 m) wingspan, was 38 ft 4 in (11.68 m) long, and was 14 ft 6 in (4.42 m) tall. With the wings folded, the aircraft’s span was 22 ft 6 in (6.89 m). The XF14C-1 had an estimated speed of 344 mph (554 km/h) at 3,500 ft (1,067 m) and 374 mph (602 km/h) at 17,000 ft (5,182 m). Its initial rate of climb was 2,810 fpm (14.3 m/s), and it had a service ceiling of 30,500 ft (9,296 m). The aircraft had an empty weight of 9,868 lb (4,476 kg), a gross weight of 12,691 lb (5,757 kg), and a maximum weight of 13,868 lb (6,290 kg). The XF14C-1 had a range of 1,080 miles (1,738 km) at 176 mph (283 km/h) on 230 US gallons (192 Imp gal / 871 L) of internal fuel. With two 75-US gallon (62 Imp gal / 284 L) drop tanks, range increased to 1,520 miles (2,446 km) at 164 mph (264 km).

Wind tunnel tests conducted by the Navy in October 1942 indicated that the Curtiss-provided performance specifications for the XF14C-1 were optimistic, but the program moved forward. The first airframe (BuNo 03183) was mostly complete by September 1943. However, delays with the XH-2470 left the XF14C-1 without an engine. The engine delay gives some credence to a contra-rotating version of the XH-2470 being used in the XF14C-1. A single rotation XH-2470 had passed a Navy acceptance test in April 1941, and a single rotation XH-2470 that was delivered to the Army Air Force had made its first flight in the Vultee XP-54 on 15 January 1943. With the availability of the single-rotation XH-2470 for the Army Air Force, it seems that such an engine could have been supplied to Curtiss for the XF14C-1 if that is what the aircraft needed. The Navy subsequently dropped its participation in the XH-2470 engine program, and the XF14C-1 was cancelled in December 1943.

Curtiss and the Navy negotiated to proceed with the XF14C program by changing the engine to the experimental Wright XR-3350-16. The -16 was turbosupercharged and used contra-rotating propellers. Rated at 2,250 hp (1,678 kW) at 32,000 ft (9,754 m), the 18-cylinder, air-cooled, radial engine offered a higher service ceiling than the XH-2470. This interested the Navy, as they were looking toward developing a high-altitude interceptor. With the new engine, the Curtiss aircraft became the XF14C-2 and was pushed into a high-altitude fighter role. The cancellation of the XF14C-1 terminated all work on the second prototype, BuNo 03184, which was never built.


The XF14C-2’s outer wing section folded up just outside of the cannons. Note the gap around the spinner for cooling the two-row, 18-cylinder R-3350 engine and that the second set of propeller blades have cuffs to aid cooling.

BuNo 03183 became the XF14C-2 and was modified to accept the new engine. A six-blade, contra-rotating Curtiss Electric propeller with a diameter of approximately 12 ft 10 in (3.91 m) was installed on the XR-3350-16 engine. The cowling incorporated an intake scoop under the engine. Oil coolers were placed in extensions of the XF14C-2 wing roots. The turbosupercharger was installed directly behind the engine in a housing that extended back from the lower cowling. A large exhaust pipe from the turbosupercharger extended below the aircraft behind the main wheels.

The Curtiss XF14C-2 had the same 46 ft (14.02 m) wingspan as the XF14C-1 but was shorter at 37 ft 9 in (37.75 m) long and 12 ft 4 in (3.76 m) tall. The aircraft had an estimated speed of 317 mph (510 km/h) at sea level and 424 mph (682 km/h) at 32,000 ft (9,754 m). The XF14C-2’s initial rate of climb was 2,700 fpm (13.7 m/s), and it had a service ceiling of 39,500 ft (12,040 m). The aircraft had an empty weight of 10,582 lb (4,800 kg), a gross weight of 13,405 lb (6,080 kg), and a maximum weight of 14,950 lb (6,781 kg). At a cruising speed of 172 mph (277 km/h), the XF14C-2 had a range of 950 miles (1,529 km) on 230 US gallons (192 Imp gal / 871 L) of internal fuel and 1,355 miles (2,181 km) with two 75-US gallon (62 Imp gal / 284 L) drop tanks.

The XF14C-2 was first flown in July 1944 and delivered to the Navy on 2 September 1944. Testing quickly revealed that the aircraft did not meet the expected performance and offered no advantage over fighters already in service. Top speeds of only 300 mph (483 km/h) at sea level and 398 mph (641 km/h) at 32,000 ft (9,754 m) were achieved. The aircraft’s engine and propeller combination also caused a bad vibration throughout the airframe. With the XF14C-2 underperforming, no urgent need for a high-altitude fighter, and all the R-3350 production dedicated for the Boeing B-29 Superfortress and Convair B-32 Dominator bombers, the Navy cancelled the XF14C-2. The airframe was eventually scrapped. The XF14C-2 was the last piston-engine fighter built by Curtiss.

Curtiss proposed the XF14C-3 to truly fulfill the role of a high-altitude fighter. It had a pressurized cockpit and could operate at 40,000 ft. Studies of the XF14C-3 were conducted at Navy expense until early 1945, but no aircraft was built.


The XF14C-2 had oil-coolers in the wing roots. Note the dihedral angle of the outer wing sections. The engine and propeller combination caused an unacceptable level of vibration.

Curtiss Fighter Aircraft by Francis H. Dean and Dan Hegedorn (2007)
US Experimental & Prototype Aircraft Projects: Fighters 1939-1945 by Bill Norton (2008)
American Secret Projects 1 by Tony Buttler and Alan Griffith (2015)
To Join with the Eagles by Murry Rubenstein and Richard M. Goldman (1974)
The American Fighter by Enzo Angelucci and Peter Bowers (1987)

Latecoere 631-03

Latécoère 631 Flying Boat Airliner

By William Pearce

On 12 March 1936, the civil aeronautics department of the French Air Ministry requested proposals for a commercial seaplane with a maximum weight of 88,185 lb (40,000 kg) and capable of carrying at least 20 passengers (with sleeping berths) and 1,100 lb (500 kg) of cargo 3,730 miles (6,000 km) against a 37 mph (60 km/h) headwind. In addition, the aircraft needed a normal cruising speed of 155 mph (250 km/h). This large passenger aircraft was to be used on transatlantic service to both North and South America. Marcel Moine, head engineer at Latécoère (Société Industrielle Latécoère, SILAT) had already been working on an aircraft to meet similar goals. In late 1935, Moine had designed an aircraft for service across the North Atlantic with a maximum weight of 142,200 lb (64,500 kg). However, the design was seen as too ambitious. Moine modified the design to meet the request issued in 1936, and the aircraft was proposed to the Air Ministry as the Latécoère 630.

Latecoere 631-04

The Latécoère 631 was one of the most impressive flying boats ever built. Unfortunately, its time had already passed before the aircraft could enter service. Laté 631-04 (fourth aircraft) F-BDRA is seen here, and it was the second of the type in service for Air France. Note the configuration of the flaps and ailerons.

The Laté 630 was an all-metal, six-engine flying boat with retractable floats. The 930 hp (694 kW), liquid-cooled Hispano Suiza 12 Ydrs was selected to power the 98,860 lb (44,842 kg) aircraft, which had a 187 ft (57.0 m) wingspan, was 117 ft 9 in (35.9 m) long, and had a range of 4,909 miles (7,900 km). On 15 November 1936, order 575/6 was issued for detailed design work of the Laté 630 and a model for wind tunnel tests. This was followed by order number 637/7 for a single Laté 630 prototype on 15 April 1937. However, the Air Ministry cancelled the Laté 630 on 22 July 1937, stating that advancements in aeronautics enabled the design and construction of a larger and more capable aircraft. Construction of the Potez-CAMS 161, which was designed under the same specifications as the Laté 630, was allowed to continue.

Taking aeronautical advancements into consideration, the Air Ministry issued an updated request for an aircraft with a maximum weight of 154,323 lb (70,000 kg) and capable of transporting 40 passengers and 11,000 lb (5,000 kg) of cargo with a normal cruising speed of over 186 mph (300 km/h). To meet the new requirements, Moine and Latécoère enlarged and repowered the Laté 630 design, creating the Laté 631. In October 1937, detailed design work and a wind tunnel model of the Laté 631 were ordered. Order number 597/8 for a single prototype was issued on 1 July 1938. A Lioré et Olivier H-49 (which became the SNCASE SE.200) prototype was also ordered under the same specification as the Laté 631.

The Latécoère 631 was an all-metal flying boat with a two-step hull. The monocoque fuselage consisted of an aluminum frame covered with aluminum sheeting. The interior of the hull was divided into numerous passenger compartments and included a lounge/bar under the radio/navigation room (may have been in the nose in some configurations) and a kitchen at the rear. The cockpit and radio/navigation room were located above the main passenger compartment and just ahead of the wings. The cockpit was positioned rather far back from the nose of the aircraft. Numerous access doors were provided, including in the nose, side of the cockpit, and in the sides of the fuselage.

Latecoere 631 cockpit

The cockpit of the Laté 631 was rather spacious. Note the six throttle levers suspended above the pilot’s seat. The copilot could not reach the levers, but the flight engineer had another set of throttles. The central pylon contained the trim wheels and controls for the floats and flaps. At left in the foreground is the navigation station, and the radio station is at right.

The high-mounted wing was blended to the top of the fuselage and carried the aircraft’s six engines in separate nacelles. The wing had two main spars and a false spar. Each wing consisted of an inner section with the engine nacelles and an outer section beyond the nacelles. The outer engine nacelle on each wing incorporated a retractable float that extended behind the wing’s trailing edge. Due to interference, the float needed to be at least partially deployed before the flaps could be lowered. A passageway in the wing’s leading edge was accessible from the radio/navigation room and allowed access to the engine nacelles. Each nacelle had two downward-opening doors just behind the engine that served as maintenance platforms. A section of the firewall was removable, allowing access to the back of the engine from within the nacelle. Between the inboard engine and the fuselage was a compartment in the wing’s leading edge designed to hold mail cargo.

Originally, 1,500 hp (1,119 kW) Gnôme Rhône 18P radial engines were selected to power the Laté 631. However, the availability of these engines was in question, and a switch to 1,600 hp (1,193 kW) Wright R-2600 radial engines was made. The Gnôme Rhône 14R and the Pratt & Whitney R-2800 were also considered, but the 14R was also unavailable, and the export of R-2800 engines was restricted. Each engine turned a three-blade, variable-pitch propeller that was 14 ft 1 in (4.3 m) in diameter and built by Ratier. Later, larger propellers were used, but sources disagree on their diameter—either 14 ft 5 in or 15 ft 1 in (4.4 m or 4.6 m). It is possible that both larger diameters were tried at various times.

At the rear of the aircraft were twin tails mounted to a horizontal stabilizer that had 16.7 degrees of dihedral. All control surfaces had an aluminum frame with a leading edge covered by aluminum. The rest of the control surface was fabric covered. Movement of the control surfaces was boosted by a servo-controlled electrohydraulic system, which could be disengaged by the pilot. The slotted aileron on each wing was split in the middle and consisted of an outer and an inner section. The ailerons also had Flettner servo tabs that were used to trim the aircraft and could be engaged to boost roll control.

Latecoere 631-01 German 63-11

Laté 631-01 (F-BAHG) in German markings as 63+11. The openings for the large passenger windows existed in the airframe but were covered on Laté 631-01. The prototype aircraft was destroyed during an allied attack while in German hands on Lake Constance in April 1944.

Six wing tanks carried 7,582 gallons (28,700 L) of fuel, and each tank fed one engine. During flight, these tanks were replenished by pumping fuel from six tanks in the hull that carried 5,785 gallons (21,900 L) of fuel. The Laté 631’s total fuel capacity was 13,367 gallons (50,600 L). Each engine had its own 111-gallon (422-L) oil tank.

The Latécoère 631 had a 188 ft 5 in (57.43 m) wingspan, was 142 ft 7 in (43.46 m) long, and was 33 ft 11 in (10.35 m) tall. The aircraft had a maximum speed of 245 mph (395 km/h) at 5,906 ft (1,800 m) and 224 mph (360 km/h) at sea level. Its cruising speed was 183 mph (295 km/h) at 1,640 ft (500 m). The Laté 631 had an empty weight of 89,265 lb (40,490 kg) and a maximum weight of 163,347 lb (75,000 kg). The aircraft had a 3,766-mile (6,060-km) range with an airspeed of 180 mph (290 km/h) against a 37 mph (60 km) headwind.

Construction of the Laté 631 was started soon after the contract was issued. However, work was halted on 12 September 1939 so that Latécoère could focus on production of desperately needed military aircraft after war was declared on Germany. After the French surrender, work on the Laté 631 resumed in July 1940 but was halted again on 10 November by German order. The French and Germans negotiated over continuing work on the aircraft, which was purely for civil transportation. The Germans allowed construction to continue, and a second prototype was ordered under the same contract as the first (597/8) on 19 March 1941. The 35 Wright R-2600 engines that had been ordered were stranded in Casablanca, Morocco by the outbreak of the war in 1939. Amazingly, the hold on these engines was released, and they were delivered at the end of 1941.

Latecoere 631-02 stripes

Laté 631-02 (F-BANT) was finished at the end of the war and painted with invasion stripes for (hopefully) easy identification. The aircraft is at Biscarrosse undergoing tests, probably around the time of its first flight on 6 March 1945. Like on the prototype, the passenger windows are covered, but the windows were later added. Note the retractable float and that engine No. 5 is running.

The Laté 631-01, the first prototype, was registered as F-BAHG and completed at Toulouse, France in the summer of 1942. The aircraft was then disassembled and transported, with some difficulty, 310 miles (500 km) to Marignane in southern France. The aircraft was then reassembled for subsequent tests on Étang de Berre. The SNCASE SE.200, the Laté 631’s competitor, was built at Marignane and was nearing completion at the same time. The reassembly of Laté 631-01 was completed in October 1942, and the aircraft made its first flight on 4 November with Pierre Crespy as the pilot. Seven others, including Moine, were onboard as crew and observers. A second flight was made on 5 November, and flutter of the aileron and wing was encountered at 143 mph (230 km/h). The issues were traced to an improperly made part in the aileron control circuit that had subsequently failed.

Laté 631-01 was repaired, but German occupation of the French free zone on November 1942 brought a halt to further flight tests. On 23 November, order 280/42 was issued for two additional Laté 631s, bringing the total to four aircraft. The Germans lifted flight restrictions, and Laté 631-01 was flown again in December 1942. Test flights continued but were halted on several occasions by German orders. In April 1943, the tests were allowed to continue provided the aircraft was painted in German colors with German markings and a Lufthansa pilot was on board during the flights. Germany had essentially seized Laté 631-01 (and the SE.200) at this point and believed the aircraft could be used as a commercial transport once the “quick” war was concluded. The Germans were also interested in ways to add armament to the flying boat and make it a maritime patrol aircraft. Laté 631-01 was repainted and carried the German code 63+11 (for 631-01).

Laté 631-01 flight testing resumed in June 1943. On 20 January 1944, the aircraft took off on its 46th flight, and it was the first flight in which its gross weight exceeded 154,323 lb (70,000 kg). A second flight was made at 157,630 lb (71,500 kg). The tests had demonstrated that at 88,185 lb (40,000 kg), the Laté 631 could hold its course with three engines on the same side shut down. At 154,323 lb (70,000 kg), the course could be held with the outer two engines shut down on the same side. Some additional indications of flutter had been encountered but not understood.

Latecoere 631-02 Brazil

Laté 631-02 at Rio de Janeiro, Brazil in late October 1945. Note the open nacelle platforms, which were accessible through a wing passageway. A Brazilian flag is attached to the forward antenna mast.

Around 22 January 1944, Laté 631-01 was taken over by German forces and flown to Lake Constance (Bodensee) and moored offshore from Friedrichshafen, Germany. The SE.200 had already suffered the same fate on 17 January. On the night of 6 April 1944, Laté 631-01 and the SE.200 were destroyed at their moorings on Lake Constance by an Allied de Haviland Mosquito. The Laté 631 prototype had accumulated approximately 48 hours of flight time.

Construction of other Laté 631 aircraft had continued until early 1944, when German forces wanted Latécoère to focus on building the Junkers 488 bomber (which was never completed and was destroyed by the French Resistance). The disassembled second Laté 631 (631-02) was hidden in the French countryside until the end of the war. On 11 September 1944, order 51/44 was issued for five additional Laté 631 aircraft, which brought the total to nine. In December 1944, the components of Laté 631-02 were transported to Biscarrosse, where the aircraft was completed and assembled for testing on Lac de Biscarrosse et de Parentis. On 6 March 1945, Crespy took Laté 631-02 aloft for its first flight. While testing continued, the aircraft was christened Lionel de Marmier and was registered as F-BANT in April 1945. On 31 July, Laté 631-02 started a round trip of over 3,730 miles (6,000 km) to Dakar, Senegal, returning to Biscarrosse on 4 August. On 24 August, material for two additional Laté 631s was added to order 51/44, enabling the production of up to 11 aircraft.

On 28 September 1945, an issue with the autopilot in Laté 631-02 caused a violent roll to the right that damaged the wing, requiring the replacement of over 8,000 rivets to affect repairs. The aircraft was quickly fixed so that a scheduled propaganda flight to Rio de Janeiro, Brazil could be made on 10 October 1945. On that day, Laté 631-02 collided with a submerged concreate mooring block while taxiing and tore a 6 ft 7 in (2 m) gash in the hull. Upset over this incident, French authorities took the opportunity to nationalize the Latécoère factories. Production of the last six Laté 631 aircraft was spread between AECAT (which was formed from Latécoère), Breguet, SNCASO, and SNCAN. SNCASO at Saint-Nazaire would be primarily responsible for the production of aircraft No. 6, 8, and 10, and SNCAN at Le Havre would be primarily responsible for aircraft No. 7, 9, and 11. Laté 631-02 eventually made the flight to Rio de Janeiro, with 45 people on board, arriving on 25 October 1945.

Latecoere 631-03

Laté 631-03 (F-BANU) was the third aircraft completed. Its first flight was on 15 June 1946, and it crashed during a test flight on 28 March 1950 while investigating the loss (in-flight break up) of Laté 631-06 on 1 August 1948. Investigation of Laté 631-03’s crash revealed vibration issues with the engines and wings, and led to a solution to prevent further accidents.

On 31 October 1945, the first tragedy struck the Laté 631 program. While on a flight between Rio de Janeiro and Montevideo, Uruguay with 64 people on board, Laté 631-02 suffered a propeller failure on the No. 3 (left inboard) engine. The imbalance caused the No. 3 engine to rip completely away from the aircraft. A separated blade damaged the propeller on the No. 2 engine (left middle), which resulted in that engine almost being ripped from its mounts. Another separated blade flew through the fuselage, killed one passenger, and mortally wounded another (who later died in a hospital). An emergency landing was performed on Laguna de Rocha in Uruguay. The failure of the Ratier propeller was traced to its aluminum hub, which was subsequently replaced with a steel unit. The recovery of the aircraft was performed by replacing the missing engine with one from the right wing. The four-engine aircraft, with a minimal crew, was flown to Montevideo on 13 November for complete repairs, which took three months.

In February 1946, three Laté 631 aircraft were purchased by Argentina, but this deal ultimately fell through, with Argentina never paying for the aircraft. In May 1946, an agreement was reached in which Air France would take possession of three Laté 631 aircraft. On 15 June 1946, Jean Prévost made the first flight of Laté 631-03 at Biscarrosse. The aircraft was registered as F-BANU, christened as Henri Guillaumet, and soon transferred to Air France.

Laté 631-04 was registered as F-BDRA, and its first flight occurred on 22 May 1947 at Biscarrosse. The aircraft was the second Laté 631 to go to Air France. Laté 631-05 was registered as F-BDRB, and its first flight occurred on 22 May 1947. Laté 631-06, registered as F-BDRC, made its first flight on 9 November 1947, taking off from the Loire estuary near Saint-Nazaire, France. Laté 631-06 F-BDRC was the third aircraft for Air France.

Latecoere 631-05

Laté 631-05 (F-BDRB) first flew on 22 May 1947. The aircraft was slated to be converted into a cargo transport, but that never occurred. The aircraft was damaged beyond economical repair during a hanger collapse in February 1956.

Laté 631-07, registered as F-BDRD, made its first flight on 27 January 1948. The aircraft was lost on 21 February during a test flight from Le Havre to Biscarrosse. Laté 631-07 had taken off in poor weather and was not equipped for flying on instruments alone. It crashed into the English Channel (Bay of Seine) off Les-Dunes-de-Varreville (Utah Beach). A definitive cause was never found, but it was speculated that either the pilot lost spatial orientation and crashed into the sea, or that the pilot was flying very low or trying to land after the weather closed in and struck wreckage left behind from the D-Day landings at Utah Beach. Regardless, all 19 on board, which were the crew and Latécoère engineers, were killed.

On 1 August 1948, Air France Laté 631-06 F-BDRC was lost over the Atlantic flying between Fort-de-France, Martinique and Port-Etienne (now Nouadhibou), Mauritania. Wreckage was recovered that indicated an in-flight breakup that possibly involved a fire or explosion, but a definitive cause was never determined. None of the 52 people on board survived. F-BDRC had accumulated 185 flight hours at the time of the accident, and Air France subsequently withdrew its two other Laté 631s from service. Laté 631-04 F-BDRA participated in the search for survivors, flying a total of 75 hours, including a single 26-hour flight.

The flying boat era had ended during the 10 years between when the Latécoère 631 was ordered in 1938, and when the aircraft went into service with Air France in 1947. The advances in aviation during World War II had shown that landplanes were the future of commercial aviation. Following the accidents, there was no hope for the Laté 631 to be used as a commercial airliner. With four completed aircraft and another four under construction, the decision was made to convert the Laté 631 into a cargo aircraft.

Latecoere 631-06 Air France

Laté 631-06 (F-BDRC) made its first flight on 9 November 1947. It was the third (and final) aircraft to be received by Air France. On 1 August 1948, Laté 631-06 disappeared over the Atlantic with the loss of all 52 on board. Air France withdrew its remaining Laté 631 aircraft as a result. Note the access hatch atop the fuselage. Another hatch existed behind the wings.

On 28 November 1948, Laté 631-08 F-BDRE was flown for the first time, taking off from Saint-Nazaire. Laté 631-08 was originally intended as an additional aircraft for Air France but was orphaned after the crash of Laté 631-06. Laté 631-08, along with Laté 631-03, were eventually given to a new company, SEMAF (Société d’Exploitation du Matériel Aéronautique Français / French Aircraft Equipment Exploitation Company). SEMAF was founded in March 1949 and worked to develop the Laté 631 as an air freighter. Laté 631-08 F-BDRE was converted to a cargo aircraft by strengthening its airframe and installing a 9 ft 2 in x 5 ft 3 in (2.80 x 1.60 m) cargo door on the left side of the rear fuselage. The aircraft was first flown with the modifications on 8 June 1949. Laté 631-08 soon began hauling fabric and manufactured products between France and various places in Africa. The aircraft had completed 12 trips by March 1950.

Laté 631-09 F-BDRF preceded Laté 631-08 into the air. Laté 631-09’s first flight occurred on 20 November 1948 at Le Harve. Laté 631-10 F-BDRG made its first flight on 7 October 1949 from Saint-Nazaire. Both of these aircraft were flown to Biscarrosse and stored with the never completed Laté 631-11 F-BDRH. Laté 631-09 and -10 were later reregistered as F-WDRF and F-WDRG.

Laté 631-03 F-BANU was reregistered as F-WANU when it underwent tests to measure vibrations of the airframe and engines. This was done in part to discover what led to the loss of Laté 631-06 F-BDRC. On 28 March 1950, Laté 631-03 made its second flight of the day, taking off from Biscarrosse. With engine power pushed up, the left wing began to flutter, and the outer section of the left aileron broke away. Laté 631-03 began to spin, turned on its back, and continued to spin until it impacted the water inverted. The 12 people on board, which included the crew and engineers from Latécoère and Rotol, were killed instantly. Many witnessed the crash, and the wreckage of Laté 631-03 was recovered. Examination revealed that the engines with a .4375 gear reduction and operating at 1,925 rpm during cruise flight turned the propeller at 840 rpm. This resonated with a critical frequency of the wings, ailerons and Flettner tabs, which was 840 cycles per minute. The interaction rapidly fatigued parts in the outer aileron control system and caused them to fail. The damaged aileron system allowed the aileron to flutter, breaking the control system completely and leading to a complete loss of aircraft control.

Latecoere 631-08

Laté 631-08 (F-BDRE) is seen here with its updated registration of F-WDRE. Laté 631-08 was the only aircraft that operated as an air freighter.

At the time if the accident, Laté 631-03 had been reengined with R-2600 engines incorporating a .5625 gear reduction. These engines were installed on later Laté 631 aircraft and retrofitted on the earlier aircraft. However, nearly all of the Laté 631-03’s 1,001 hours were with the other engines, which was enough to have fatigued the aileron control to its breaking point. The loss of Laté 631-03 led to the collapse of SEMAF.

With the cause of the crash known, a new company was formed to upgrade the Laté 631 fleet and modify them for cargo service. La Société France Hydro (France Hydro Company) was given charge of Laté 631-02 and Laté 631-08, which was reregistered as F-WDRE. Modifications to prevent a reoccurrence of Laté 631-03’s crash were incorporated into the aircraft, and Laté 631-08 returned to cargo service in late 1951. Laté 631-08 flew a Biscarrosse-Bizerte-Bahrain-Trincomalee-Saigon route of some 7,460 miles (12,000 km) starting in March 1952. The aircraft departed Bizerte, Tunisia with a takeoff weight of 167,000 lb (75,750 kg), the highest recorded for a Laté 631. By 1953, Laté 631-08 was hauling cotton from Douala, Cameroon to Biscarrosse. This had proven somewhat lucrative, and a cargo-conversion of Laté 631-02 was started. Laté 631-05 was also transferred to France Hydro, but little was done with the aircraft. On 10 September 1955, Laté 631-08 broke apart during a violent thunderstorm while over Sambolabo, Cameroon. All 16 people on board were killed. The Latécoère 631 was withdrawn from service after this accident, and no further attempts were made to use the aircraft.

In February 1956, Laté 631-05, -10, and -11 were damaged beyond economical repair when the roof of the Biscarrosse hangar collapsed after heavy snowfall. All of the remaining Latécoère 631s were subsequently scrapped, most in late 1956. In 1961, the remains of Laté 631-01 and the SE.200 prototype were raised from Lake Constance by a Swiss recovery team and subsequently scrapped.

Latecoere 631-08 France-Hydro

Laté 631-08 while in service with France Hydro. The aircraft crashed in a storm on 10 September 1955; this was the last flight of any Laté 631. The remaining aircraft were later scrapped. Note the open door on the bow and the open hatch forward of the cockpit that led to a cargo hold.

Les Paquebots Volants by Gérard Bousquet (2006)
Latécoère: Les avions et hydravios by Jean Cuny (1992)

Yokosuka YE2H front

Yokosuka YE2H (W-18) and YE3B (X-24) Aircraft Engines

By William Pearce

After World War I, the Japanese Navy established the Aircraft Department of the Hiro Branch Arsenal, which was part of the Kure Naval Arsenal. These arsenals were located near Hiroshima, in the southern part of Japan. The Aircraft Department was the Japanese Navy’s first aircraft maintenance and construction facility. In April 1923, the Hiro Branch Arsenal became independent from the Kure Naval Arsenal and was renamed the Hiro Naval Arsenal (Hiro).

Kawanishi E7K1 floatplane

The Kawanishi E7K1 floatplane served into the 1940s and was powered by the Hiro Type 91 W-12 engine. The Type 91 was based on the Lorraine 12Fa Courlis.

In 1924, the Japanese Navy purchased licenses from Lorraine-Dietrich in France to manufacture the company’s 450 hp (336 kW) 12E aircraft engine. The Lorraine 12E was a liquid-cooled, W-12 aircraft engine, and Hiro was one of the factories chosen to produce the engine. Hiro manufactured three different versions of the Lorraine engine, appropriately called the Hiro-Lorraine 1, 2, and 3. In the late 1920s, Hiro started designing its own engines derived from the Lorraine architecture. Hiro also produced engines based on the updated Lorraine 12Fa Courlis W-12. It is not clear if Hiro obtained a license to produce the 12Fa or if the production was unlicensed. The most successful of the Hiro W-12 engines was the 500–600 hp (373–447 kW) Type 91, which was in service until the early 1940s. Modeled after the 12Fa Courlis, the Type 91 had a bank angle of 60-degrees and four valves per cylinder. The engine had a 5.71 in (145 mm) bore, a 6.30 in (160 mm) stroke, and displaced 1,935 cu in (31.7 L).

Like Lorraine, Hiro also produced W-18 engines. Hiro’s first W-18 engine was built in the early 1930s and used individual cylinders derived from the type used on the 12Fa Courlis / Type 91. While Hiro’s W-18 engine may have been inspired by the Lorraine 18K, the engine was not a copy of any Lorraine engine. Reportedly, Hiro’s first W-18 had a 60-degree bank angle between its cylinders. The engine did not enter production and was superseded in 1934 by the Type 94. The Type 94 replaced the earlier engine’s individual cylinders with monobloc cylinder banks and used a 40-degree angle between the banks. The W-18 engine had a 5.71 in (145 mm) bore and a 6.30 in (160 mm) stroke. The Type 94 displaced 2,902 cu in (47.6 L) and produced 900 hp (671 kW) at 2,000 rpm. The engine was 86 in (2.18 m) long, 44 in (1.11 m) wide, 43 in (1.10 m) tall, and weighed 1,631 lb (740 kg). Only a small number of Type 94 engines were produced, and its main application was the Hiro G2H long-range bomber, of which eight were built. The engine was found to be temperamental and unreliable in service.

Hiro G2H1 bomber

The Hiro G2H1 bomber was the only application for the company’s Type 94 W-18 engine. The engine was problematic, and only eight G2H1s were built. Note the exhaust manifold for the center cylinder bank.

By the mid-1930s, the Navy’s aircraft engine development had been transferred from Hiro to the Yokosuka Naval Air Arsenal (Yokosuka). For a few years, the Navy and Yokosuka let aircraft engine manufacturers develop and produce engines rather than undertaking development on its own. However, around 1940, Yokosuka began development of a new W-18 aircraft engine, the YE2.

The Yokosuka YE2 was based on the Hiro Type 94 but incorporated many changes. The liquid-cooled YE2 had an aluminum, barrel-type crankcase, and its three aluminum, monobloc cylinder banks were attached by studs. The cylinder banks had an included angle of 40 degrees and used crossflow cylinder heads with the intake and exhaust ports on opposite sides of the head. All of the cylinder banks had the intake and exhaust ports on common sides and were interchangeable.

Each cylinder had two intake and two exhaust valves, all actuated by a single overhead camshaft. The camshaft for each cylinder bank was driven via a vertical shaft from an accessory section attached to the drive-end of the engine. The YE2 had a 5.71 in (145 mm) bore, 6.30 in (160 mm) stroke, and displaced 2,902 cu in (47.6 L). The YE2A, B, and C variants had a rated output of 1,600 hp. However, very little is known about these engines, and it is not clear if they were all built.

Yokosuka YE2H front

The Yokosuka YE2-series was developed from the Hiro Type 94. The YE2H was built in the early 1940s, but no applications for the engine have been found. Note the output shaft on the front of the engine that is bare of its extension shaft. The vertical fuel injection pump is just above the horizontally-mounted magnetos. (Smithsonian Air and Space Museum image)

The Yokosuka YE2H variant was developed around 1942 and given the Army-Navy designation [Ha-73]01. It is not clear how the YE2H differed from the earlier YE2 engine. The YE2H was intended for installation in an aircraft’s fuselage (or wing) in a pusher configuration. The rear-facing intake brought in air to the engine’s supercharger. Air from the supercharger was supplied to the cylinders at 12.6 psi (.87 bar) via three intake manifolds—one for each cylinder bank. A common pipe at the drive-end of the engine connected the three intake manifolds to equalize pressure. Fuel was then injected into the cylinders via the fuel injection pump driven at the drive-end of the engine. The two spark plugs per cylinder were fired by magnetos, located under the fuel injection pump. An extension shaft linked the engine to a remote gear reduction unit that turned the propeller at .60 times crankshaft speed.

The YE2H had a maximum output of 2,500 hp (1,864 kW) at 3,000 rpm. The engine had power ratings of 2,000 hp (1,491 kW) at 2,800 rpm at 4,921 ft (1,500 m) and 1,650 hp (1,230 kW) at 2,800 rpm at 26,247 ft (8,000 m). The YE2H was approximately 83 in (2.10 m) long, 37 in (.95 m) wide, and 39 in (1.00 m) tall. The engine weighed around 2,634 lb (1,195 kg). The YE2H was completed and run around March 1944, but development of the engine had tapered off in mid-1943. At that time, Yokosuka refocused on the YE3 engine, which was derived from the YE2H.

Yokosuka YE2H side

The YE2H’s rear-facing intake scoop (far left) indicates the engine was to be installed in a pusher configuration. Note the intake manifolds extending from the supercharger housing. (Smithsonian Air and Space Museum image)

Development of the Yokosuka YE3 started in the early 1940s. The engine possessed the same bore and stroke as the YE2, but the rest of the engine was redesigned. The YE3 was an X-24 engine with four banks of six cylinders. The left and right engine Vees had a 60-degree included angle between the cylinder banks, which gave the upper and lower Vees a 120-degree angle. The YE3’s single crankshaft was at the center of its large aluminum crankcase.

Each cylinder bank had dual overhead camshafts actuating the four valves in each cylinder. The camshafts were driven off the supercharger drive at the non-drive end of the engine. The supercharger delivered air to the cylinders via two loop manifolds—one located in each of the left and right engine Vees. Two fuel injection pumps provided fuel to the cylinders where it was fired by two spark plugs in each cylinder. The fuel injection pumps and magnetos were driven from the drive end of the engine. Exhaust was expelled from the upper and lower engine Vees. Like the YE2, the YE3 was designed for installation in an aircraft’s fuselage or wing, with an extension shaft connecting the engine to the remote propeller gear reduction.

Yokosuka YE3B front

The drive end of the Yoskosuka YE3B gives a good view of the engine’s X configuration. The fuel injection pumps are below the output shaft. (Larry Rinek image via the Aircraft Engine Historical Society)

The YE3A preceded the YE3B, but it is not clear if the YE3A was actually built. The Yokosuka YE3B was given the joint Army-Navy designation [Ha-74]01. The YE3B had a 5.71 in (145 mm) bore and a 6.30 in (160 mm) stroke. The engine displaced 3,870 cu in (63.4 L) and produced 2,500 hp (1,864 kW). The YE3B was rated at 2,150 hp (1,603 kW) at 6,562 ft (2,000 m) and 1,950 hp (1,454 kW) at 16,404 ft (5,000 m). The engine was approximately 79 in (2.00 m) long, 43 in (1.10 m) wide, and 28 in (.70 m) tall.

The YE3B was run by October 1943. The engine used a two-speed remote gear reduction that drove contra-rotating propellers. No real applications for the YE3B are known. However, the engine is often listed as the powerplant for the S-31 Kurowashi (Black Eagle), which was a purely speculative propaganda aircraft. The S-31 was designed as a heavy bomber, and its four YE3B engines were buried in its fuselage.


Side view of the YE3B illustrates the engine’s loop intake manifold. Spark plug leads and fuel injector lines can be seen in the Vee between the cylinder banks. Note the camshaft-driven water pump mounted on the end of the lower cylinder bank. (Tom Fey image)

A further development of the YE3-series was the YE3E. The YE3E was given the joint Army-Navy designation [Ha-74]11. The engine was similar to the earlier YE3-series except that it had two crankshafts. Some sources indicate the engine essentially consisted of two V-12s laid on their sides in a common crankcase with their crankshafts coupled to a common output shaft. The YE3E produced 3,200 hp (2,386 kW) and had power ratings of 2,650 hp (1,976 kW) at 4,921 ft (1,500 m) and 2,200 hp (1,641 kW) at 26,247 ft (8,000 m). The YE3E was approximately 79 in (2.00 m) long, 51 in (1.30 m) wide, and 39 in (1.00 m) tall. The engine was scheduled for completion in spring 1944, but no records have been found indicating it was finished.

A YE2H [Ha-73]01 W-18 engine and a YE3B [Ha-74]01 X-24 engine were captured by US forces after World War II. The engines were sent to Wright Field in Dayton Ohio for further examination. The United States Air Force eventually gave the YE2H and YE3B engines to the Smithsonian National Air and Space Museum, where they are currently in storage.


Detail view of the supercharger mounted to the end of the YE3B. Note the updraft inlet for the supercharger. Camshaft drives can be seen extending from the supercharger housing to the cylinder banks. (Tom Fey image)

Japanese Aero-Engines 1910–1945 by Mike Goodwin and Peter Starkings (2017)
Japanese Secret Projects 1 by Edwin M. Dyer III (2009)

Piaggio P119 engine

Piaggio P.119 Experimental Fighter

By William Pearce

Founded in 1884, Piaggio was an Italian industrial firm that began making aircraft under license in 1917. In 1923, Piaggio began building aircraft of its own design, led by Giovanni Pena. In the early 1930s, Piaggio began to manufacture aircraft engines under license. In 1936, Pena left the company and was replaced by Giovanni Casiraghi. Casiraghi had previously worked for the Waco Aircraft Company in the United States for several years.

Piaggio P119 mockup

Mockup of the Piaggio P.119 in the Finale Ligure plant. Note the guns in the wing. They appear to be 7.7 mm (.303-cal), but it is not clear. Only two machine guns are in the nose.

In 1938, Casiraghi began to design a new single-seat fighter of a rather unconventional configuration. He aspired to create a fast and maneuverable fighter that utilized as many Piaggio-sourced components as possible—the aircraft, engine, and propeller were all manufactured by Piaggio. Designated as the Piaggio P.119, the fighter design was submitted to the Regia Aeronautica (Italian Royal Air Force) on 18 March 1939. While the Regia Aeronautica was busy with other projects, Casiraghi continued to refine the fighter. The experimental P.119 was not ordered until 2 June 1941.

The P.119 had a conventional layout with the exception of the engine installation. The air-cooled, radial engine was located in the fuselage, behind the pilot. An extension shaft extended from the engine, under the cockpit, and to the propeller gear reduction at the front of the aircraft. This configuration provided good pilot visibility and enabled the armament to be centrally located in the aircraft’s nose and the engine to be located at the aircraft’s center of gravity, which enhanced maneuverability.

Piaggio P119 construction

The P.119 under construction at Finale Ligure. Note the tubular-steel center section of the engine mount and the frame of the aileron awaiting its fabric covering.

The P.119 had an all-metal airframe made up of three sections. The front and rear fuselage sections had an aluminum frame covered with aluminum panels, creating a monocoque structure. The center section, which supported the engine and wings, consisted of a tubular steel frame covered with aluminum panels. The entire fuselage possessed a circular cross section. Under the conventional tail was a non-retractable tailwheel. The all-metal wings had two spars and housed the fully retractable main wheels. Large ailerons occupied the outer half of the wings’ trailing edge, with split-flaps running along the remaining trailing edge of the wing. All control surfaces had an aluminum frame and were covered with fabric. Each wing contained an 87-gallon (330 L) fuel tank, and a 90-gallon (340 L) fuel tank was located in the fuselage behind the engine.

The cockpit was placed above the wings’ leading edge and covered with a canopy that hinged to the side (some sources state the canopy slid back). However, it does not appear that the hinged canopy covering was installed. Behind the cockpit was a tubular-steel frame that supported the air-cooled radial engine and connected the aircraft’s nose section, wings, and tail section. Originally, a 1,700 hp (1,268 kW) Piaggio P.XXII engine was to be used, but delays with that engine resulted in the substitution of a 1,500 hp (1,119 kW) Piaggio P.XV. Both engines had 18 cylinders and displaced 3,237 cu in (53.0 L). A scoop located under the aircraft’s nose brought in cooling air that was distributed annularly into the cooling fins of the engine’s cylinders with baffles helping to direct the airflow. The cooling-air exited via a semi-annular line of cowl flaps set atop the fuselage. Just behind the cockpit was the engine’s intake, and the exhaust was expelled from four stacks forward of the cowl flaps. The P.119’s variable-pitch, three-blade propeller was made by Piaggio and was 10 ft 10 in (3.3 m) in diameter.

Piaggio P119 engine

Nicolò Lana in the cockpit of the P.119 preparing for an engine run. The canopy has been removed, and only two machine guns are installed in the nose. The two left-side exhaust stack openings are visible in front of the open cowl flaps.

The aircraft’s armament consisted of four 12.7 mm (.50-cal) machine guns positioned in the nose above the propeller gear reduction and a 20 mm cannon that fired through the propeller hub. The machine guns had 500–550 rpg (the number varies by source), and the 20 mm cannon had 110 rounds. Some sources state that provisions existed to install two additional machine guns in each wing with 400 rpg. However, those sources disagree on whether the guns were 7.7 mm (.303-cal) or 12.7 mm (.50-cal). A mockup of the P.119 included the wing guns, which appear to be 7.7 mm (.303-cal), but the mockup also appears to have only two nose machine guns. Images of the P.119 prototype do not indicate any provisions for wing guns. Reportedly, the prototype did not have the cannon or two of the four nose machine guns installed. Consideration was given to a ground attack version with a 37 mm cannon firing through the propeller hub, and a bomb rack under each wing and under the aircraft’s centerline.

Piaggio P119 rear

Rear view of the P.119 illustrates the aircraft’s relatively clean exterior. The aircraft is at Villanova d’Albenga, presumably before its first flight.

The Piaggio P.119 had a wingspan of 42 ft 8 in (13.0 m), a length of 31 ft 10 in (9.7 m), and a height of 9 ft 10 in (3.0 m). The aircraft had a top speed of 398 mph (640 km/h) at 22,310 ft (6,800 m) and a stalling speed of 81 mph (130 km/h). The P.119 had an empty weight of 5,886 lb (2,670 kg) and a maximum weight of 9,039 lb (4,100 kg). The aircraft had an initial rate of climb of approximately 3,077 fpm (15.6 m/s), and a climb to 19,685 ft (6,000 m) took 7 minutes and 15 seconds. The P.119’s ceiling was 41,011 ft (12,500 m), and it had a maximum range of 932 miles (1,500 km).

Some sources indicate that two P.119 prototypes were ordered and given the Matricola Militare (military registration number) of MM 496 and MM 497, with MM 496 used on the mockup and MM 497 applied to the actual prototype. It is not clear why a mockup would need a serial number, and other sources contend that MM 496 was assigned to the prototype. However, MM 496 appears to have been assigned to the Piaggio P.108C prototype four-engine transport, and the majority of sources state that MM 497 was the P.119 prototype.

Piaggio P119 painted

The P.119 undergoing an engine run. Note the scoop that brought in cooling air for the engine. The aircraft had a fairly wide-track landing gear.

The P.119 was built at Piaggio’s Finale Ligure plant in western Italy. The aircraft was completed in late 1942 and underwent ground tests in mid-November. The P.119’s first flight occurred on 19 December 1942. The aircraft was flown at Villanova d’Albenga by Nicolò Lana. The initial flight testing revealed that the P.119 suffered from engine cooling issues, requiring the cowl flaps to stay open. The open flaps slowed the aircraft and caused its nose to pitch up. Other issues included vibrations from the engine and extension shaft installation and general instability of the P.119. These issues resulted in complete flight trails not being conducted, and aerobatic maneuvers were not attempted. On 2 August 1943, the P.119 was damaged when the brakes locked up on landing, causing the aircraft to nose over. The damage was minor and mostly limited to the propeller and a wing, but the aircraft was not repaired before the Italian surrender on 8 September 1943. Problematic and difficult to fly, the P.119 subsequently disappeared and was presumably scrapped.

Piaggio P119 noseover

The P.119 after it nosed over during landing on 2 August 1943. While the aircraft has been painted, it does not appear that the canopy cover has been installed. Note the deployed split flaps, and the intake scoop behind the cockpit.

Dimensione Cielo 3: Caccia Assalto by Emilio Brotzu, Michele Caso, Gherardo Cosolo (1972)
Volare Avanti: The History of Piaggio Aircraft by Paolo Gavazzi (2000)
War Planes of the Second World War: Fighters, Volume Two by William Green (1961)
Italian Civil and Military Aircraft 1930-1945 by Jonathan Thompson (1963)