Category Archives: Through World War I

Paradox side

Deissner ‘Paradox’ Rotary Aircraft Engine

By William Pearce

Deissner Paradox running

Charles (Carl) Deissner stands next to the 30 hp (22 kW) Paradox engine during a test run. The frame around the engine enabled it to be run in such demonstrations and was not needed when the engine was installed in an aircraft (which probably never happened). Note the carburetor at the front of the engine.

As early aviators began to take flight, it quickly became apparent that most engines were not suitable for use in aircraft. A number of engineers and designers worked to create light and powerful engines that were ideal for aircraft use. Some of these designs, such as the Antoinette, lay the foundation for many engines to follow, but other engine designs were quickly abandoned. Like many others, Charles (Carl) Deissner of London, England designed one of the engines destined to go nowhere. Deissner called his engine the Paradox.

The Paradox was an air-cooled, four-cylinder, four-stroke, rotary engine. In this context, “rotary” refers to a radial engine in which the crankcase and cylinders rotate around the crankshaft. This is not to be confused with a Wankel engine, which uses a rotor spinning in a fixed crankcase to produce power. Most rotary aircraft engines had the crankshaft fixed to the airframe, while the propeller was attached to and spun with the crankcase. However, the Paradox was not like other rotary engines.

While the crankcase of the Paradox rotated just like any other rotary engine, inside the Paradox, the crankshaft turned at twice the speed of the crankcase and in the same direction. The engine’s internals were kept in order by eccentric gears on the crankshaft engaging ring gears inside the crankcase. The stroke of the crankshaft represented one quarter of the piston’s stroke. The path of the eccentric gear also represented one quarter of the piston’s stroke. The relative motions of the crankcase and crankshaft enabled the full stroke to be utilized and allowed the unique Paradox engine to function.

Paradox engine sectional

Sectional view of the 30 hp (22 kW) Paradox engine. The valves can be seen on the front of the L-head cylinders. Below the valves are the pushrods actuated by counter-weighted rockers. The rockers are driven by a short camshaft extending on each side of a pinion. The pinion rotates as its teeth mesh with a gear fixed to a stationary shaft at the front of the engine.

The easiest way to visualize the Paradox engine’s operation is to consider the piston at the top of the cylinder when the cylinder is at the 12 o’clock position. Here, the crankshaft and its throw are at top dead center. When the engine has rotated 180 degrees, putting the cylinder at the six o’clock position, the crankshaft has rotated 360 degrees. The crankshaft is again at top dead center, but since the cylinder is now at the bottom of the engine, the piston is now at the bottom of the cylinder. When the engine has rotated another 180 degrees, the cylinder is back at the 12 o’clock position, and the crankshaft, having rotated 360 degrees, is again at top dead center, returning the piston to the top of the cylinder.

The crankshaft had two throws 180 degrees apart, and each throw served a pair of cylinders. The cylinders of each pair were 180 degrees apart, and the two cylinder pairs were 90 degrees apart. A non-articulating, solid connecting rod served each cylinder pair so that when one piston was at the top of the cylinder, the piston in the opposite cylinder was at the bottom of the cylinder. Deissner stated that since the connecting rod did not articulate, Paradox engines could be made with relatively long strokes and achieve high compression ratios.

Paradox side

The 30 hp (22 kW) Paradox engine complete with propeller. Note the skew gear-driven magneto, counter-weighted rocker arms, and the cylinders’ L-head design.

Three versions of the Paradox engine were planned for construction with different outputs: 30 hp (22 kW), 70 hp (52 kW), and 100 hp (75 kW). However, it appears only the 30 hp (22 kW) and 70 hp (52 kW) engines were actually built. While both engines had four cylinders and shared the same rotary and eccentric crankshaft arrangement, each engine also had a number of unique features.

The 30 hp (22 kW) Paradox was a demonstration engine mounted in a metal frame. The engine utilized an L-head cylinder with side valves. The single intake and exhaust valves were positioned on the front side of the cylinder. Each valve was actuated by a pushrod driven by a large, counter-weighted rocker arm. Part of the rocker rode on a camshaft that extended through the axis of a pinion. The cam on one side of the pinion controlled the intake while the cam on the other side controlled the exhaust. The pinion was driven by a skew gear mounted on a stationary shaft that did not rotate with the engine.

Induction air was brought in through a carburetor at the front of the engine. The air/fuel mixture then passed through the crankcase, where it was warmed, and into separate manifolds for each cylinder. Exhaust was taken through a manifold from each cylinder, piped through the crankcase, and vented out the front of the engine’s propeller shaft, which was fixed to the crankcase.

Paradox Induction Exhaust

Schematic view of the induction and exhaust system in the 70 hp (52 kW) Paradox engine. For clarity, the valves are illustrated on the front and rear of the T-head cylinder, rather than its sides. Note the offset crankshaft.

A magneto was driven by a skew gear at the rear of the engine. The magneto fired the one spark plug installed in each cylinder. However, it appears the engine could accommodate two spark plugs per cylinder. Ball bearings were used throughout the engine. The 30 hp (22 kW) Paradox engine had a 2.76 in (70 mm) bore and a 7.17 in (182 mm) stroke. The engine displaced 171 cu in (2.8 L). Its 30 hp (22 kW) output was obtained at 1,200 rpm. Increasing the engine’s rpm to 1,400 resulted in an output of 40 hp (30 kW).

The 70 hp (52 kW) Paradox engine also used side valves but in a T-head arrangement, with the valves on opposite sides of the cylinder. The valves were actuated by the same method used on the 30 hp (22 kW) engine, but a pushrod and rocker was now positioned on each side of the cylinder. One schematic shows the valves on the front and back sides of the cylinder, rather than the left and right sides. This was most likely done for illustrative purposes, to show how the engine’s induction and exhaust systems worked. Induction air was brought in the front of the engine, passed through the crankcase (where it was warmed), and flowed through a fixed shaft at the rear of the engine. Here, it passed through a carburetor, and the air/fuel mixture flowed back through the shaft to manifolds at the rear of the engine; these manifolds led to each cylinder. The 70 hp (52 kW) Paradox engine had a 4.0 in (102 mm) bore and a 7.0 in (178 mm) stroke. The engine displaced 352 cu in (5.8 L). Its 70 hp (52 kW) output was obtained at 1,400 rpm, and it produced 60 hp (45 kW) at 1,200 rpm. The engine weighed 220 lb (100 kg).

Deissner Paradox Ad

1910 advertisement for the Paradox engine expressing its many virtues over other rotary engines. The pricing for the 70 hp (52 kW) engine is given, although the 30 hp (22 kW) engine is illustrated in the photograph. The price of the 70 hp (52 kW) engine was increased to £460 in March 1911. (via

The 30 hp (22 kW) Paradox engine was running by late 1910. It was run both with and without a 7 ft 6 in (2.3 m) Eta propeller. Some of the engine’s noted advantages were that standard lubricating oil could be used—other rotaries typically needed castor oil. The Paradox engine was also said to have good fuel economy, but no specifics were given. In early 1911, the 30 hp (22 kW) engine broke free during a test run, resulting in a destroyed propeller and a damaged engine. The engine was repaired in February, and the 70 hp (52 kW) Paradox engine was to be finished by March 1911. However, no further information has been found regarding any Paradox engine.

The Paradox engines may have offered some improvements in oil consumption, which was always quite high with standard rotaries, but its other unique features did not offer any tangible advantage over more popular engines. Rotary engines would continue to be widely used until after World War I. At that time, conventional engines had out-powered the rotary, and the inherent limitations of its spinning crankcase design could not be overcome.

– “The Paradox Rotary Engine” Flight (19 November 1910)
The Art of Aviation by Robert W. A. Brewer (2nd edition, 1911)

Antoinette VII with 100hp V-16 engine paris 1909

Antoinette (Levavasseur) Aircraft Engines

By William Pearce

Léon Levavasseur was one of those rare geniuses of early aviation who designed and built engines as well as aircraft. In the early 1900s, Levavasseur gave up his career as an electrical engineer to focus on aviation. He needed funding to pursue his interests, so in August (some say July) 1902, Levavasseur approached industrialist and fellow Frenchman Jules Gastambide. Levavasseur had impressed and then befriended Gastambide when he repaired generators at one of Gastambide’s power plants. At the July meeting, Levavasseur outlined his plans for an aircraft of his design to be powered by a new engine that he was also designing. Levavasseur suggested the new enterprise should be named Antoinette, after Gastambide’s young daughter. Gastambide was interested but thought the engine should be built first, as no aircraft had yet flown. Levavasseur was agreeable, and with financial backing secured, he set to work on the new engine.

Antoinette IV 50hp V-8 Latham Levavasseur Camp Chalons 5 June 1909

An Antoinette mechanic (left), Hubert Latham (middle), and Léon Levavasseur (right) with an Antoinette IV aircraft powered by a 50 hp (37 kW) Antoinette V-8 at Camp de Châlons in early June 1909. At the event, Latham flew continuously for 1:07:37, setting a French endurance record. Note the condenser under the aircraft’s wing.

Levavasseur quickly returned to Puteaux (near Paris), France, and set up a shop to work on the new engine. On 28 August 1902, Levavasseur applied for a secret patent on his engine, which consisted of eight cylinders laid out in a Vee pattern, forming two banks of four cylinders. This patent application became public on 28 August 1903 and was granted French patent no 399,068 on 30 September 1904.

From 1902 through 1910, Levavasseur produced several different Antoinette engines, but they all used the same basic layout. Technically, the first time the “Antoinette” name was applied during the Levavasseur-Gastambide partnership was in 1905 when a series of boats were so named. It would not be until 1906 that the Société Anonyme Antoinette was officially incorporated.

All of Levavasseur’s engines consisted of individual, water-cooled cylinders arranged in a 90 degree Vee on an aluminum crankcase. The cylinders were staggered on the crankcase to facilitate the use of side-by-side connecting rods. With this arrangement, the connecting rods for each left and right cylinder pair attached to the crankshaft on the same crankpin. This allowed the engine to be much shorter than if each connecting rod had its own crankpin. The connecting rods were of a tubular design, and the pistons were made from cast iron.

Cody Antoinette 50hp V-8 Nulli Secundus 1907

A 1907 image of Samuel Cody with a 50 hp (37 kW) Antoinette V-8 in the framework of the “Nulli Secundus,” Britain’s first airship.

Each cylinder had one intake and one exhaust valve positioned on the Vee side of the cylinder. The intake valve was situated above the exhaust valve, creating an “F-head” or “Intake Over Exhaust” (IOE) cylinder head. The intake valve was atmospheric (automatic)—drawn open by the vacuum created in the cylinder as the piston moved down. The exhaust valve was actuated by a pushrod driven by the camshaft located in the Vee of the engine. The top of the cylinder’s combustion chamber was hemispherical, with a single spark plug positioned at its center.

The Antoinette engines used a primitive type of direct fuel injection. A belt-driven fuel pump at the rear of the engine fed fuel into a small reservoir (injector) located above each intake valve. When the intake valve opened, the suction that drew in air also pulled in fuel from the reservoir via a narrow, capillary passageway .008 in (.2 mm) in diameter. By manipulating the fuel pump, the pilot could exert a certain degree of fuel flow regulation. However, the system had some issues, and Antoinette engines had difficulty running at low rpm. In addition, the small passageways in the “injectors” easily became clogged by impurities in the fuel. These complications led some Antoinette operators to convert the engine to a carbureted induction system.

To cool a normal Antoinette engine, a belt-driven water pump at the rear of the engine provided water to an inlet at the base of each cylinder on the outer side of the engine. The water then passed up through the water jacket and along the cylinder barrel. The water exited the top of the cylinder on the Vee side of the engine, where it flowed into a common manifold for each cylinder bank. The water was then taken to a radiator or to a reservoir tank, as engine cooling was not critical on very short flights.

Antoinette 50hp V-8 London Science Museum

This 50 hp (37 kW) Antoinette V-8 engine is on display at the Science Museum in London and may have been used by Cody in 1908 for the first flight in Britain. Note the aluminum cylinder heads, brass water jackets, and copper water manifolds. The fuel distributor can be seen on each vertical intake pipe. (Warbird Tails image)

For lubrication, oil was taken from the crankcase and pumped through a pipe inside the crankcase, just above the camshaft. The pipe ran the length of the engine and was pierced with numerous small holes. Oil sprayed from the pipe, lubricating all of the engine’s internal components. Any components not in direct contact with an oil spray were lubricated by the oil mist created inside the crankcase.

Some (but not all) of the Antoinette engines had the ability to reverse their running. This was particularly helpful for braking and maneuvering in dirigibles. The camshaft was normally locked into its driving gear at the rear of the engine. With the engine stopped, the camshaft could be unlocked, rotated 90 degrees, and locked into a second position in its driving gear. One position was for normal (counter-clockwise) rotation, and the other was for reverse rotation (clockwise).

Antoinette engine ad V-16

Antoinette engine ad circa 1907 illustrating the lightness of the 220 lb (100 kg), 100 hp (75 kw) V-16 engine.

Each Antoinette engine was made to the highest standards in a shop that boasted of tolerances down to .0004 in (.01 mm). Another feature of all Antoinette engines was that their components were engineered to be just strong enough for their individual tasks. By designing parts with operating stresses in mind, all extra material could be eliminated, resulting in complete engines that were much lighter than their contemporaries. This design philosophy also had a drawback: engine reliability could suffer because parts were more easily overstressed, resulting in a failure. Antoinette engines were relatively specialized, and when one was in need of repair, it had to be shipped back to the factory.

The early Antoinette engines used open cylinder barrels made from cast iron. An aluminum cylinder head was bolted to the barrel. Each cylinder had a spun brass water jacket shrink-fitted to the cylinder head. The engine’s spark plugs were fired by a battery-powered ignition coil distributor.

Levavasseur’s first engine was running by the end of 1902, but efforts to improve the engine were undertaken throughout 1903. The engine was a V-8 with a 5.12 in (130 mm) bore and stroke. The engine’s total displacement was 842 cu in (13.8 L), and it produced 80 hp (60 kW). Initially, the engine weighed 346 lb (157 kg), but refinements brought the weight down to 320 lb (145 kg).

Antoinette V-16 engine display

An Antoinette V-16 engine possibly in storage at the Musée de l’Air et de l’Espace in Le Bourget, France. Note how the V-16 utilized all the same components as the V-8, with the exception of the crankcase, crankshaft, camshaft, and water manifolds. This engine has rather unusual exhaust stacks. (image source)

The 80 hp (60 kW) Levavasseur engine underwent a military test in March 1903 in which it produced 63 hp (50 kW). This output was derived by using a alternative power calculation method, and Levavasseur objected to the results. A little later, the engine was tested again, using another power calculation method. During this test, the engine registered an output of 82 hp (61 kW). The engine’s performance and Levavasseur’s ideas sufficiently impressed General Louis André, France’s Minister of War, who then provided 20,000 Francs of secret funds to Levavasseur for the construction of his airplane.

From July to September 1903, Levavasseur built his airplane in Villotran, France, which is why the aircraft became known as the Aéroplane de Villotran. The aircraft was powered by the 80 hp (60 kW) engine. Unfortunately, the aircraft proved incapable of flight. By 15 September 1903, Levavasseur had decided the aircraft was a failure; the engine was removed, and the aircraft was burned.

Antoinette V-24 marine engine

The 360 hp (268 kW) Antoinette V-24 marine engine of 1906. Unlike the V-16 engines, the V-24 appears to be comprised of V-8 engine sections. The engine is labeled as follows: A) air intake pipe; B) exhaust; C) cooling water outlet; D) aluminum cylinder head; E) steel cylinder covered with a brass water jacket sleeve; F and G) ignition distributors; H) spark plug; h) cylinder head bolts; I) fuel distributor to the intake valve; J) cooling water inlet; K) cylinder mounting bolt; P) engine mounting flange; S and T) gears for the ignition distributors; U) crankshaft gear; V) camshaft gear.

The Aéroplane de Villotran’s failure to fly, and the fact that no other aircraft had yet flown, made the concept of an aircraft engine seem futile. However, Levavasseur and Gastambide knew the engine held great potential and turned to motorboat racing as a way for the engine to prove its worth. From 1904 through 1906, Levavasseur’s engines powered a number of motorboats that achieved various distance speed records. It was to some of these motorboats that the “Antoinette” name was first given. In some cases, V-8 engines were coupled in tandem to create a more substantial power unit. By 1908, the Antoinette company had ended its support for motorboat racing and focused on using its engines only for aviation.

In 1904, while his engine was beginning to gain fame, Levavasseur designed other engines. Using the 80 hp (60 kW) V-8 as a foundation and maintaining the 5.12 in (130 mm) bore and stroke, Levavasseur designed V-16, V-24, and V-32 engines. Levavasseur believed that engines with many cylinders created frequent power pulses that smoothed out the engine’s operation and caused less stress on its internal components. The V-16 displaced 1,685 cu in (27.6 L) and produced 155 hp (116 kW). The V-24 displaced 2,527 cu in (41.4 L) and produced 225 hp (168 kW). The V-32 was most likely never built, but it would have displaced 3,370 cu in (55.2 L) and produced around 300 hp (224 kW).

Bleroit IX 50hp V-16 1908

A Bléroit IX aircraft under construction in 1908 with a 50 hp (37 kW) Antoinette V-16 installed in its frame. From left to right, the Bléroit mechanics are Louis Peyret, Louis Paragot, M. Pelletier, Alfred Bertrand, and Julien Mamet.

Levavasseur also varied the bore and stroke to make engines of different sizes and power. In 1905, Levavasseur built a V-8 Antoinette engine with a 3.15 in (80 mm) bore and stroke for aviation pioneers Ferdinand Ferber and Alberto Santos-Dumont. The engine displaced 196 cu in (3.2 L) and produced 24 hp (18 kW). The engine went through some refining and eventually weighed only 79 lb (36 kg). Another V-8 used a 4.13 in (105 mm) bore and stroke to displace 444 cu in (7.3 L). This engine produced 50 hp (37 kW) and weighed 176 lb (80 kg). The engine was approximately 32 in (.81 m) long, 24 in (.62 m) wide, and 22 in (.55 m) tall. A large V-8 was also built with a 7.87 in (200 mm) bore and stroke. Undoubtedly for marine use, this engine displaced 3,067 cu in (50.3 L), produced 200 hp (149 kW), and weighed 838 lb (380 kg).

In 1906, Levavasseur built a large V-24 Antoinette engine for marine use. The V-24 had a 5.91 in (150 mm) bore and stroke and displaced 3,882 cu in (63.6 L). The engine produced 360 hp (268 kW) and weighed some 1,322 lb (600 kg). Some sources say this engine was too heavy for the intended boat, which ended up sinking. The specifics of this incident have not been found.

Antoinette 50hp V-8 engine Krakow

A later model 50 hp (37 kW) Antoinette V-8 with an extended propeller shaft on display at the Muzeum Lotnictwa Polskiego in Krakow, Poland. Note the different lengths of the air intake pipes and that the water jackets are made from copper. (Alan Wilson image)

An Antoinette automobile made its debut in the 1906 Salon de l’Automobile in Paris, and some Antoinette engines were built for automotive use, these being a slightly different design and heavier than the aviation engines. Adams Manufacturing Company in London built a small number of the automotive engines under license for cars they were manufacturing.

By 1907, V-16 versions of the 3.15 in (80 mm) bore/stroke and 4.13 in (105 mm) bore/stroke engines were being built. The 3.15 in (80 mm) bore and stroke V-16 displaced 393 cu in (6.4 L), produced 50 hp (37 kW), and weighed around 143 lb (65 kg). The 4.13 in (105 mm) bore and stroke V-16 displaced 888 cu in (14.5 L), produced 100 hp (75 kW), and weighed 220 lb (100 kg). The 100 hp V-16 engine’s approximate dimensions were 55 in (1.40 m) long, 24 in (.62 m) wide, and 22 in (.55 m) tall.

Antoinette engines circa1907

The basic specifications of various Antoinette engines available circa 1907.

In addition to the V-16 engines listed above, Levavasseur offered Antoinette V-8 engines with five bore/stroke combinations. The V-8 engines had bores/strokes of 3.15, 4.13, 5.12, 5.91, or 7.87 in (80, 105, 130, 150, or 200 mm); the smallest two were mainly used for aviation.

It seems that around 1908 Levavasseur ceased experimenting with different engines and began working to refine the popular types. Levavasseur incorporated many changes to the Antoinette engines, but not all of the changes were applied at the same time. An elongated propeller shaft housing of approximately 12 in (.30 m) was cast integral with the crankcase. This feature made for a very aesthetically pleasing installation when the engine was used in an Antoinette aircraft. The battery-powered coil ignition was replaced by an accumulator and high-frequency distributor. A new one-piece, steel cylinder was used in which the cylinder head was integral with the cylinder barrel. This construction allowed for a higher compression ratio. New one-piece water jackets made of either brass or copper surrounded the new cylinders. The water jacket was unusual in that it was made by electrolytically depositing metal onto a wax mold that had been coated with graphite as a conductive material. The wax was then melted out, leaving a formed water jacket as thin as .04 in (1 mm). While this method of construction could yield perfect parts, it was expensive, and there was a high rejection rate because of irregularities in the water jacket wall.

Antoinette VII with 100hp V-16 engine paris 1909

Antoinette VII aircraft with a 100 hp (75 kW) V-16 engine displayed at the Salon de l’Aeronautique in 1909. Most sources indicate the condenser used aluminum tubes and copper side manifolds, details that this image seems to support.

By 1908, Levavasseur had perfected a zero-loss steam-cooling method for the Antoinette engines that were installed in Antoinette aircraft. The system used the same basic routing as the normal cooling system, but the water was allowed to boil in the water jackets. The steam was then collected and sent to a water separating tank. From the tank, the steam was sent though large condensers made of aluminum tubing with copper side manifolds. The condensers were positioned horizontally on each side of the aircraft. The steam condensed back to water and was routed back to the separating tank by a second pump. From the separating tank, the water was pumped back to the engine. The system condensed .26 gallons (1 L) of water per minute, and its capacity was 3.17 gallons (12 L). By reducing the amount of water needed, the steam-cooling method weighed less than conventional water-cooling.

Four Antoinette engines were displayed at the first Paris Salon de l’Aeronautique, starting in December 1908. A V-8 and V-16 were still offered with bores and strokes of 3.15 in (80 mm). However, the engines’ weights had increased to 93 lb (42 kg) for the V-8 and 165 lb (75 kg) for the V-16. The other two engines were a V-8 and V-16 with a bore of 4.33 in (110 mm) and a stroke of 4.13 in (105 mm). The V-8 engine displaced 487 cu in (8.0 L) and weighed 209 lb (95 kg). The V-16 engine displaced 974 cu in (16.0 L) and weighed 264 lb (120 kg). Some sources list the larger bore engines as producing 50 hp (37 kW) and 100 hp (75 kW) respectively. However, other sources give the outputs as 67 hp (50 kW) and 134 hp (100 kW). Note that the kW values (50 and 100) of the second figures match the hp values of the first figures. Possibly a printing error, the higher power figures have been found in fairly early publications and have been repeated a number of times over the years. The weight increases for all four engines were a result of strengthening components to increase Antoinette engine reliability.

Antoinette 50hp V-8 close up

A late model 50 hp (37 kW) Antoinette V-8 on display in the Musée de l’Air et de l’Espace. Note that this engine used brass water jackets; the piping and water manifolds were made from copper. (Aerofossile2012 image)

Alberto Santos-Dumont, Paul Cornu, Louis Blériot, Gabriel and Charles Voisin, Henri Farman, Léon Delagrange, Samuel Cody, and Hubert Latham are just some of the pioneers who used Antoinette engines to power their flying machines and themselves into the record books. Outside of the Wright brothers, almost all early aviation “firsts” were achieved in machines powered by an Antoinette engine. It was a 100 hp (75 kW) V-16-powered Antoinette aircraft that Latham flew during the aviation meet at Reims, France in August 1909 and for the Gordon Bennett Cup at Belmont Park, New York in October 1910.

With the success of the engine, Levavasseur began to focus entirely on building aircraft. From 1908 on, little engine development was undertaken to keep the Antoinette engines at the forefront of aviation. The Antoinette aircraft were built with the same spare-no-expense mentality as the engines, which resulted in them being priced far above the competition. To make matters worse, the aircraft proved to be a challenge to fly. At the same time, other engine manufactures closed the developmental gap, and the expensive Antoinette engines were no longer the coveted power plant they once were. These factors conspired to put the Antoinette company out of business in 1912.

Latham Antoinette VII 100hp V-16 NY 1910

Latham’s 100 hp (75 kW) V-16-powered Antoinette VII aircraft at Belmont Park, New York in October 1910 for the Gordon Bennett Cup. Note the length of the condenser, which extends some 13 ft (4 m) along the side of the aircraft.

While Levavasseur was most likely the first to build a V-8 engine, it is a near certainty that he was the first to create V-16 and V-24 engines. Very little information can be found regarding the V-24 marine engine. Some sources state that Levavasseur also built a V-32 intended for marine use, while other sources claim the engine did not proceed past the design phase.

Some sources state that Levavasseur also built V-12 and V-20 engines. These engines, especially the V-20, depart from Levavasseur’s known engine layout with a V-8 at its core. While no photographs of Levavasseur’s V-12 (or V-32) engine have been found, the V-20 engine does still exist (albeit as a table). The V-20’s cylinders, valves, and valve train do not match any other engine built by Levavasseur. There was a commonality of components and configuration from Levavasseur’s earliest engine of 1903 to his last of circa 1910. The components that make up the V-20, reportedly built in 1905, are unique to that engine and are not common with any other Levavasseur engine. This component incompatibility would lead some to conclude that the V-20 engine was not built by Levavasseur.

Antoinette VII with 50hp V-8 Musee du Bourge

An Antoinette VII aircraft with a 50 hp (37 kW) V-8 engine on display in the Musée de l’Air et de l’Espace. Note how well the engine fits into the aircraft and that the condensers appear to be made entirely of copper. For cooling, steam flowed from the top of the water jackets into a separation tank. The steam then flowed from the tank into the condenser, where it returned to water. The water was then pumped from the condenser into the bottom of the separation tank and back to the engine. (Pline image)

Levavasseur’s Antoinette engines were essentially the first commercially available aircraft engine and represented the pinnacle of performance in the early days of aviation. Just about every early European aviation pioneer’s first flight was powered by an Antoinette, but the engines’ reliability left much to be desired. While some changes were incorporated over the years, and their reliability did improve, the basic engine did not change much. Just before 1910, other engines (the Gnome rotary in particular) offered similar power for a similar weight but were often more reliable than the Antoinettes.

Note: Many sources give similar but completely different values for Antoinette engine dimensions, bores, strokes, and outputs. Some discrepancies can be attributed to numerous unit conversions being applied, and some of the power discrepancies can be attributed to the engines having different outputs at different rpms. Sadly, it seems that detailed specifics were not recorded in those early days of aviation; therefore, there can be no absolute certainty about the various Antoinette engine models or their histories.

V-20 engine

The V-20 engine as displayed at Le Manoir de l’Automobile et des Vieux Métiers in Lohéac, France. Regardless of the V-20’s origins, it seems rather inglorious for such machinery to be turned into furniture. However, the engine would have probably been scrapped long ago had it not found favor as a conversation piece. Apparently, its conversion to a table consisted of nothing more than bolting on legs to preexisting mounts, something that could easily be reversed for a more befitting display. (ZANTAFIO56 image)

Bléroit: Herald of an Age by Brain A. Elliot (2000)
Les moteurs et aéroplanes Antoinette by Gérard Hartmann (13 August 2007) 7.4 MB pdf in French
The Art of Aviation by Robert W. A. Brewer (1910)
The Passion That Left the Ground by Stephen H. King (2007)
Vingt Cinq Ans d’Aéronautique française: 1907-1932 Tome 1 (1934)
– “The First Paris Aeronautical Salon: Engines for Aeroplanes” Flight (16 January 1909)
– “How Levavasseur Built his Light Motor” by Ferdinand Ferber The Automobile (28 March 1907)
– “Historie du moteur Antoinette” by Ferdinand Ferber l’Aérophile (15 February 1908)
– “Moteur à huit cylindres” French patent 339,068 by Léon Levavasseur (applied 28 August 1903)
– “Carbureter” US patent 878,297 by Léon Levavasseur (applied 16 may 1907)
– “Umsteuerungsvorrichtung für Mehrzylinder-Explosionskraftmaschinen” Austrian patent 25,610 by Léon Levavasseur (applied 14 September 1904)
Airplane Engine Encyclopedia by Glenn Angle (1921)

Roberts 6-X engine side

Roberts Motor Company Aircraft Engines

By William Pearce

In the early 1900s, the Roberts Motor Company of Sandusky, Ohio made a series of two-stroke engines for boats. With aviation gaining popularity, it was only natural for the company to adapt its engines for aircraft use. Roberts aircraft engines first appeared in 1911 and were designed by the company’s founder and president Edmund W. Roberts.

Roberts 4-X engine Smithsonian

A four-cylinder Roberts 4-X engine on display at the Smithsonian National Air and Space Museum in Washington, DC. Note the tubular housing to which the carburetor is attached. Inside the housing is the tubular distributor sleeve for delivering the air/fuel mixture to the cylinders. The water pump is mounted on the upper rear of the housing. (National Air and Space Museum image)

Roberts Motor Company’s aircraft engines differed from their marine counterparts in that they were engineered to be as light as possible. To keep parts (and associated points of failure) to a minimum, the Roberts two-stroke engines did not use poppet valves. In addition, Roberts’ engines incorporated unique designs to overcome drawbacks generally found with two-stroke engines, namely the air/fuel mixture pre-igniting as it entered the cylinder, causing a backfire.

The engines’ cast cylinder liners were constructed of a proprietary alloy called “Aerolite,” which Roberts said was as light as aluminum but twice as strong and had the wear properties of cast iron. The individual cylinder liners were covered by an aluminum water jacket. A ring around the base of the cylinder liner would fit into a recess around the bottom of the water jacket, securing the two together. The cylinder liner’s spark plug boss (and that of the decompressor if present) passed through the water jacket casting. The outer diameter of the boss was threaded, and a nut secured the top of the water jacket and cylinder liner. This nut would also draw up the base of the liner into the water jacket, securing the whole cylinder assembly.

Roberts 6-X engine side

The six-cylinder Roberts 6-X engine. The crankcase casting provided a space under each cylinder for the air/fuel mixture. Each space was sealed by crankshaft main bearings. Note the two exhaust ports for each cylinder.

The pistons were made of cast iron and were attached to drop forged I-beam connecting rods made of vanadium steel. The connecting rods were attached to the crankshaft by a bronze strap about a third the size of the crankpin. (Firing every revolution, the pistons of a two-stroke engine are not pulled down by the crankshaft and therefore do not need a full-size connecting rod bearing cap.) The crankshaft was hollow and made of drop forged steel. The cylinders were secured to the crankcase by four bolts, with adjacent cylinders sharing their bolts. The crankcase was made from Magnalium, an aluminum and magnesium alloy that made it lighter and stronger than an aluminum crankcase of the same thickness.

The carburetors were mounted to a tubular housing that ran along the right side of the engine. Inside of this housing was a tubular distributor sleeve (also called a rotary induction valve) driven by an intermediate gear that engaged the accessory drive gear mounted on the end of the crankshaft. The distributor sleeve rotated at crankshaft speed and had ports to control the air/fuel mixture flow from the carburetor into the crankcase. The crankcase was constructed so that a small space existed under each cylinder for the incoming charge. The carburetor aligned with multiple ports in the distributor sleeve to allow a constant flow into the distributor, but each cylinder matched up with a single port to control air/fuel delivery. For each cylinder, the incoming charge passed from the distributor though a port in the side of the crankcase. The distributor helped eliminate the risk of backfires and distributed the air/fuel mixture equally to all the cylinders, allowing the engine to run smoothly.

Roberts 4-X Rotary Distributor

The tubular distributor sleeve of a Roberts 4-X engine. The ports at the center of the tube aligned with the carburetor.

The distributor port opened as the piston moved up and drew in the air/fuel mixture. The port then closed as the piston moved down on its power stroke, compressing the incoming charge underneath. Two ports in the piston aligned with ports in the cylinder wall when the piston was near bottom dead center. This alignment allowed the pressurized, incoming air/fuel mixture to flow from the crankcase into a space on the outer side of the cylinder. In this space, Roberts installed what they called a “cellular by-pass” to prevent backfires. The incoming charge flowed through the cellular by-pass and then through another set of ports positioned above the piston. The piston’s top had a large deflector to send the incoming air charge toward the top of the combustion chamber to improve exhaust gas scavenging. This deflector was positioned on the intake port side of the piston. Two exhaust ports were located on the opposite side of the cylinder from the intake ports and were also controlled by the piston so that all ports were uncovered when the piston was near bottom center.

The cellular by-pass was a series of flat and corrugated plates creating a honeycomb mesh. The large surface area of the cellular by-pass extinguished any flame should a backfire occur, but it did not decrease the engine’s efficiency in normal operation. The Roberts engine’s resistance to backfiring allowed a leaner mixture to be used, thus increasing the engine’s fuel economy. The cellular by-pass also helped mix and vaporize the fuel in the incoming charge.

Roberts 6-X parts Weeks

Various parts for one of the two Roberts 6-X replica engines built for Kermit Weeks. Note the deflector on the top of the pistons and the two ports in the side of the pistons. A decompressor valve can be seen on the top of the cylinder in the lower left corner. (Fantasy of Flight image)

The water pump was mounted at the right rear of the engine and was driven from the gear driving the tubular distributor sleeve. The pump drew water from the radiator and then pushed it through a passageway on the right side of the crankcase. Each cylinder had an open port that aligned with a coolant passageway in the crankcase. The water flowed into a small channel in the cylinder and then around the exhaust ports and up into the cylinder’s water jacket. It then flowed out the top of the cylinder and into a manifold that led back to the radiator. The engine was lubricated by the oil and fuel mixing and via splash lubrication. Grease cups were used to lubricate the main bearings.

A single spark plug was mounted in the center of each cylinder’s semi-hemispherical combustion chamber. The spark plug was fired by a Bosch magneto mounted at the rear of the engine. The magneto was driven by a helical gear via the intermediate gear that meshed with the accessory drive gear on the end of the crankshaft. An advance fork mounted above the magneto shaft moved the helical gear of the magneto along its shaft to either advance or delay ignition—an adjustment that could be made by the pilot while in flight.

Roberts 6-X gears Weeks

The exposed accessory gears of the Robert 6-X replica. The crankshaft drove an intermediate gear, the backside of which engaged the magneto drive shaft. To advance or delay the timing, the helical gear for the magneto drive shaft could be adjusted by the brass advance fork above it. The intermediate gear drove the gear for the tubular distribution sleeve, which in turn drove the gear for the water pump. (Fantasy of Flight image)

The Roberts four-cylinder engine was known as the 4-X. It had a 4.5 in (114 mm) bore, a 5 in (127 mm) stroke, and a total displacement of 318 cu in (5.2 L). The engine produced 50 hp (37 kW) at 1,200 rpm. It used one carburetor, and its magneto turned at twice crankshaft speed. Its crankshaft was 2.5 in (64 mm) in diameter, with crankpins 1.75 in (44 mm) in diameter and 2.5 in (64 mm) long. The crankshaft was supported by five main bearings; the one toward the propeller was 6.375 in (162 mm) long. The crankshaft was 40 in (1,016 mm) long and weighed 17.5 lb (7.9 kg). The 4-X was 40.5 in (1.03 m) long, 25 in (.64 m) tall, 24 in (.61 m) wide, and weighed 170 lb (77 kg).

The six-cylinder engine was known as the 6-X. Like the 4-X, it had a 4.5 in (114 mm) bore and a 5 in (127 mm) stroke. The engine’s total displacement was 477 cu in (7.8 L), and it produced 75 hp (56 kW) at 1,200 rpm. The 6-X used two carburetors, and its magneto turned at three times crankshaft speed. Its crankshaft and crankpins were the same size as the 4-X’s. The crankshaft was supported by seven main bearings, was 52 in (1,321 mm) long, and weighed 27.5 lb (12.5 kg). The 6-X was 52.5 in (1.33 m) long, 25 in (.64 m) tall, 24 in (.60 m) wide, and weighed 240 lb (109 kg).

Roberts 6-XX engine

The Roberts 6-XX engine with exhaust manifolds and enclosed accessory gears.

A further development of the six-cylinder engine was the 6-XX. This engine had its accessory gears covered and bathed in oil. Its bore and stroke were enlarged to 5.5 in (140 mm) and 6 in (152 mm) respectively. The 6-XX’s total displacement was 588 cu in (14.0 L), and it produced 125 hp (93 kW) at 1,100 rpm. The engine used two carburetors. Its Bosch HL magneto turned at 1.5 times crankshaft speed to fire one of the two spark plugs in each cylinder. The other spark plug was fired by a Delco distributor. The 6-XX’s crankshaft was 3 in (76 mm) in diameter, with crankpins 2.5 in (64 mm) in diameter and 3.5 in (99 mm) long. The crankshaft was supported by seven main bearings; the one toward the propeller was 12 in (305 mm) long. The 6-XX was approximately 60.5 in (1.54 m) long, 27.5 in (.70 m) tall, 24 in (.60 m) wide, and weighed 390 lb (177 kg).

The Roberts engines were refined over time and used by a good number of early aviation pioneers. By 1913, all Roberts engines had the exposed accessory gears enclosed like those on the 6-XX engine. This change necessitated the magneto be repositioned. The cylinder liners were now made of cast iron, and the water jackets were made of Aerolite. The pistons were also made of Aerolite, which reduced their weight by over 2 lb (.9 kg) each. The tubular distributor sleeve was mounted on four sets of ball bearings to reduce friction. Dual ignition, like that used on the 6-XX, was available as an option on the 6-X engine. A starting crank attached to the end of the crankshaft was also an option.

Roberts 6-X 1913 rear

The rear of the 1913 Roberts 6-X engine showing the enclosed accessory gears, repositioned magneto, and optional hand starting crank on the end of the engine. The decompressors can be seen on the top of the cylinders. These were used to make starting the engine easier.

As two-stroke Roberts engines were surpassed by new four-stroke engines, like the Curtiss OX-5 and Hispano-Suiza 8, the company struggled to keep up. By 1918, the 6-X engine had its bore and stroke increased to 5 in (127 mm) and 5.5 in (140 mm) respectively, giving a total displacement of 648 cu in (10.6 L). The tubular distributor was replaced by more conventional intake manifolds, and the engine produced 100 hp (75 kW) at 1,200 rpm. The 6-X now weighed 368 lb (167 kg).

There is also some indication that a V-12 was planned and possibly constructed by Roberts in the late 1910s. This engine was known as the E-12. It had a 6 in (152 mm) bore, a 6.5 in (165 mm) stroke, and a total displacement of 2,205 cu in (36 L). The engine produced 350 hp (261 kW) at 1,200 rpm. Each cylinder had its own crankpin, and 13 main bearings supported the crankshaft. The E-12 weighed 990 lb (490 kg).

Roberts 6-X Weeks

One of Weeks’ Roberts 6-X replica engines. The engines were built to power a replica Benoist XIV flying boat. The Benoist is a pusher, which is why the outlet for the coolant pipe is toward the propeller shaft. Note the brass grease cups for lubricating the crankshaft main bearings. The cover on each cylinder is for the cellular by-pass. (Fantasy of Flight image)

The Roberts Motor Company left the aviation field by 1919 to focus on marine engines. Around this time, the company changed its name to Roberts Motors. A few years later, Roberts Motors went out of business. A number of early Roberts four- and six-cylinder aircraft engines still exist in museums.

Recently, Kermit Weeks of Fantasy of Flight in Polk City, Florida commissioned the creation of two replica Roberts 6-X engines. One of the engines was a test engine, and the other would be installed in a Benosit XIV flying boat replica. An original engine was reverse engineered, allowing these engines to be built. Below is a video from 2013 of Mr. Weeks checking on the progress of the Roberts 6-X engine construction at Vintage & Auto Rebuilds in Chardon, Ohio. Both Roberts 6-X replica engines have since been completed and test run.


Roberts Aviation Motors by The Roberts Motor Co. (1912)
– “Construction of Cylinder of Internal Combustion Engines” US patent 1,210,537 by Edmund W. Roberts (granted 2 January 1917)
The 1913 Model 6-X by The Roberts Motor Co. (1913)
Fantasy of Flight Blog (various entries relating to the Roberts engine and Benoist flying boat replica)
Aerosphere 1939 by Glenn Angle (1940)
(Jane’s) All the World’s Aircraft 1918 by C. G. Grey (1918)

REP 7-cylinder

R.E.P. Fan (Semi-Radial) Aircraft Engines

By William Pearce

Robert Esnault-Pelterie was born in France on 8 November 1881. In the early 1900s, he began experimenting with a glider modeled after the Wright Brothers’ glider of 1902. Esnault-Pelterie had experienced trouble with the Wrights’ wing warping technique and switched to ailerons in 1903. In 1906, Esnault-Pelterie began constructing a powered aircraft of his own design as well as an engine to power it. Esnault-Pelterie named the aircraft the R.E.P. 1 and it first flew in 1907. A new feature of the aircraft was a control stick, which Esnault-Pelterie patented. That patent made Esnault-Pelterie a rich man once royalties were paid after World War I. While the unique engine that he designed was the first in a family of engines known as R.E.P., their success would not endure like that of the control stick.

REP 7-cylinder

Front view of the R.E.P. seven-cylinder fan engine. Exhaust gases flowed out the holes around the top of the cylinders. The intake manifold can barely be seen behind the cylinders. In this view, the four cylinders on the right shared an intake manifold as did the three cylinder on the left.

Esnault-Pelterie’s first engine was an air-cooled, seven-cylinder fan engine. Sometimes referred to as a semi-radial, this engine had cylinders that were fanned-out on the top of the crankcase and not positioned around its entirety like a true radial. The cylinders were arranged in two rows—the front had four cylinders, and the rear had three cylinders. This configuration solved lower cylinder lubrication issues of radial engines, and air-cooling issues of inline engines.

The cast iron cylinders were attached to an aluminum crankcase. The upper part of the cylinder had cooling fins to dissipate heat. At the very top of the cylinder was a large, single valve. The valve was shaped like a piston and inverted, with the valve stem attached to underside of the head. When the valve was completely closed, a flange on its head would seat against the cylinder head and seal the cylinder. When the valve was partially open, exhaust gases flowed around the flange and escaped through ports in the cylinder head. When the valve was completely open, ports in its sides aligned with ports in the cylinder head to allow the intake mixture to flow into the cylinder.

REP 7-cylinder section

In this section view of the R.E.P. seven-cylinder engine, the cylinders are numbered by firing order. Cylinder 5 has the intake/exhaust valve completely closed. Cylinder 1 shows the valve partially open to allow exhaust gases to exit the cylinder. Cylinder 2 shows the valve completely open to allow the air/fuel mixture into the cylinder. Note the master/articulated connecting rod arrangement.

The valve was actuated by a rocker arm attached to the cylinder. The rocker arm was moved via a pushrod that was operated by a cam ring inside the engine. Each row of cylinders had its own cam ring positioned at the rear of the engine, and the cam rings had stepped lifts. The first step opened the valve part way to allow the exhaust gases to vacate the cylinder. The second, higher lift completely opened the valve to allow the fresh air/fuel mixture into the cylinder. For the intake, the cylinders were separated into left and right groups, with the left group (when viewing the engine from the rear) having an additional cylinder. Each group shared a common intake manifold with a carburetor attached to its end. The intake manifold was attached to the upper rear of the cylinder. Exhaust gases flowed out though ten holes around the cylinder’s top; there were no exhaust stacks.

A single spark plug was installed in the side of the cylinder and fired by an ignition coil powered by a battery. The pistons were made of steel and had two oil rings. They were attached to the connecting rods by trunnions bolted to the underside of the piston. The connecting rod for each row of cylinders had one master rod, and the rest were articulating rods. The crankshaft had two throws offset 180 degrees and was supported by two main bearings. To balance the crankshaft, Esnault-Pelterie left the crankpin solid for the row with an additional cylinder, and the crankpin for the row with one fewer cylinder was drilled hollow.

REP 10-cylinder side

This side view of the R.E.P. 10-cylinder engine illustrates how it was comprised of two five-cylinder engines bolted front-to-front. Note the rocker arm arrangement and the single valve. The pictured engine was equipped with magnetos.

The engine had a 3.35 in (85 mm) bore and 3.74 in (95 mm) stroke. The seven cylinders displaced a total of 230 cu in (3.8 L). The engine produced 30 hp (22 kW) at 1,500 rpm and weighed 150 lb (68 kg). A five-cylinder version was also built with three cylinders in the first row and two in the second. It produced 20 hp (15 kW) at 1,500 rpm from its 164 cu in (2.7 L) and weighed 118 lb (54 kg). Another version consisted of two five-cylinder engines joined at their front to create a 10-cylinder engine. Each engine group had its own intake manifold feeding five cylinders. The 10-cylinder engine produced 50 hp (37 kW) at 1,500 rpm from its 329 cu in (5.4 L) and weighed 214 lb (97 kg). Some sources indicate the same coupling treatment was applied to the seven-cylinder engine to create a 14-cylinder engine, but this cannot be confirmed. A 14-cylinder engine would have displaced 461 cu in (7.5 L) and produced around 70 hp (52 kW).

The five- and seven-cylinder engines powered a number of early aircraft (R.E.P.s, Bléroits, Kapferer-Paulhans, and Breguets), but it is unlikely the 10-cylinder engine ever flew. Esnault-Pelterie received an award in 1908 from the Société des ingénieurs civils de France (Society of Civil Engineers of France) for his seven-cylinder R.E.P. engine. However, the cylinder’s single valve proved unsatisfactory, and the engines were redesigned in 1909.

REP 10-cylinder back

Rear view of the R.E.P. 10-cylinder engine equipped with a coil ignition. Note that each engine section has its own intake manifold and carburetor.

The updated engines had two valves per cylinder, but they were still operated by a single rocker arm. The intake valve was in the front of the cylinder, and the exhaust valve was in the rear. The rocker arm pivoted between the valves so that it pushed the intake valve open and then rocked back to pull down on the exhaust valve to open it. This was achieved by a grooved cam-disc that could “pull” and “push” the pushrod.

The engine’s bore and stroke were increased to 3.94 in (100 mm) and 5.51 in (140 mm). The five-cylinder engine displaced 335 cu in (5.5 L) and produced 60 hp (45 kW) at 1,400 rpm. The seven-cylinder engine displaced 470 cu in (7.7 L) and produced 90 hp (67 kW) at 1,400 rpm. There is no indication that any attempt to couple the engines was made. The cylinders had revised cooling fins, and the spark plug was repositioned to the cylinder head. Magnetos were used in place of the coil ignition.

REP 5-cylinder Type D

An updated R.E.P. five-cylinder engine preserved in a R.E.P. Type D monoplane at the Musée de l’Air et de l’Espace. Note the two valves per cylinder and rocker arm arrangement. The unique induction system can be seen in which it drew air from the crankcase and delivered it to the cylinders via the copper pipes. The individual exhaust pipes can be seen at the rear of the engine. (John Martin image via the Aircraft Engine Historical Society)

Induction and exhaust were also updated. Intake air was fed from the crankcase (where it was warmed), through a distributor, and then to the front of each cylinder. Exhaust gases were collected in pipes at the rear of each cylinder and directed away from the cockpit. The many changes increased the weight of the engines to 243 lb (110 kg) for the five-cylinder and 287 lb (130 kg) for the seven-cylinder. The updated R.E.P. fan engine had no trouble running for 10 hours non-stop during various bench tests.

The five-cylinder engine seemed the more successful of the two and was installed in a number of aircraft (R.E.P.s and Farman-Neubauers). It was used in the Vickers R.E.P., which was the first aircraft made by Vickers and would have been the first aircraft to fly in Antarctica had its wings not been damaged during a demonstration flight in Australia. Even so, the wingless Vickers was taken to Antarctica and used as a powered sled, but with not much success. The remains of this aircraft were rediscovered there in January 2010.

Vickers REP in Antarctica

The Vickers R.E.P. in Antarctica in 1911. The engine clearly has two valves per cylinder and the unique induction system of the updated fan engine.

In 1911, Esnault-Pelterie refocused his design efforts on true radial engines, constructing five- and seven-cylinder engines. The fan/semi-radial engines were phased out in 1912. Over the next few years, Esnault-Pelterie stopped designing piston engines as he became more interested in rocketry. A few R.E.P. fan engines still exist in museums, including a seven-cylinder engine in Esnault-Pelterie’s original R.E.P. 1 aircraft from 1907 displayed at the Musée national des Arts et Métiers (National Museum of Arts and Crafts) in Paris, France. This museum may also hold another original seven-cylinder engine cutaway. An updated five-cylinder engine exists installed in an uncovered R.E.P. Type D monoplane from 1910 at the Musée de l’Air et de l’Espace (Air and Space Museum) in Le Bourget, France.

Note: Many sources list a variety of different bore and stroke combinations for the R.E.P. fan engines. Some sources list some of the early engines as having a 3.54 in (90 mm) stroke, while others list the updated engines as having a 4.33 in (110 mm) bore or 6.30 in (160 mm) stroke. While it is possible that such bore and stroke combinations were built, little supporting information has been found.

REP 1 with 7-cylinder engine

Esnault-Pelterie’s original R.E.P. 1 aircraft and its engine preserved in the Musée national des Arts et Métiers. (PHGCOM image via Wikimedia Commons)

– “Moteur Extra-Léger a Explosion” by Robert Esnault-Pelterie, Mémoires et Compte Rendu des Travaux de la Société des ingénieurs civils de France Bulletin (December 1907)
Les aéroplanes et moteurs R.E.P. by Gérard Hartmann (4 MB pdf)
Les Moteurs a Pistons Aeronautiques Francais Tome I by Alfred Bodemer and Robert Laugier (1987)
– “The First Paris Aeronautical Salon: Engines for Aeroplanes” Flight (16 and 23 January 1909)
Aero Engines by G. A. Burls (1916)

Lancia V-12 aircraft engine side

Lancia Tipo 4 and Tipo 5 V-12 Aircraft Engines

By William Pearce

Vincenzo Lancia was born near Turin, Italy in 1881. From an early age, he demonstrated an aptitude in mathematics, and his father encouraged him to become an accountant. However, Lancia was mainly interested in machinery and engineering. By the age of 17, he worked as a bookkeeper for a small bicycle and auto repair shop, operated by the Ceirano brothers. There, he became more of a mechanic’s assistant than a bookkeeper. When the shop built an automobile, FIAT bought the business and made Lancia, who was only 19, chief inspector of their new factory and also a test driver. His driving skills impressed FIAT, and they later made him a race car driver.

Lancia V-12 aircraft engine

The Lancia Tipo 4 V-12 aircraft engine displaying its individual cylinders and distinct valve train. The engine was configured for pusher installations, which is why exhaust was expelled toward the propeller.

But Lancia wanted to design. In 1906, at the age of 25, he and another FIAT worker founded their own car company: Lancia & Company. Lancia and his company produced a number of vehicles and engines and became known throughout Europe. Always experimenting and innovating, Lancia took out patents for a narrow Vee engine configuration and an offset crankshaft in 1915. World War I interrupted plans to use the design for an automotive engine but gave Lancia incentive to build an aircraft engine.

Known as the Tipo 4 (Type 4), the Lancia V-12 aircraft engine was water cooled and had a 50-degree angle between the cylinder banks. The engine’s individual steel cylinders were mounted to its aluminum crankcase, with a deep oil pan. Each cylinder had one intake and one exhaust valve perpendicular to the cylinder axis. These horizontal valves opened into a small, rectangular clearance space above the cylinders that extended the combustion chamber above the piston. The valves were actuated by long rocker arms positioned in the Vee of the engine. A single hollow camshaft positioned in the middle of the Vee acted upon the rocker arms. The camshaft was driven from the crankshaft at the rear of the engine. The valve train was very similar to that later used on the Duesenberg Model H aircraft engine.

Lancia V-12 aircraft engine stand

Two technicians stand next to the Tipo 4 engine. Note the straight exhaust stacks. This is thought to be the prototype Tipo 4, and it closely resembles the engine that is preserved in the FIAT/Lancia archives.

Each cylinder had two spark plugs that were positioned on the opposite side from the valves. Two magnetos were located at the rear of the engine, each firing one spark plug per cylinder. One of the magnetos could be replaced by a distributor. Two Claudel-Lancia carburetors were mounted on each side of the engine. Each carburetor supplied air to three cylinders via a manifold that looped above the cylinders. A section of the intake manifold was jacketed to use engine cooling water to heat the air/fuel mixture as it traveled to the cylinders. Exhaust was expelled via a short manifold extending above each cylinder.

The hollow crankshaft had six throws and used a side-by-side connecting rod arrangement. However, to compensate for the odd Vee angle, each cylinder had its own crankpin that was slightly offset (displaced) from the crankshaft’s center. Cast aluminum pistons and pressure lubrication were used. Cooling water was pumped into the jacket around each cylinder’s barrel via manifolds on each side of the engine. The water then flowed up into the cylinder head and finally into a manifold that took it back to the radiator.

Lancia V-12 aircraft engine side

Side view of the Lancia Tipo 4 illustrating the two carburetors and intake manifolds on the side of the engine. Note the two spark plugs for each cylinder.

The Tipo 4 had a 4.75 in (120.7 mm) bore and a 7.09 in (180 mm) stroke. Many sources state the bore was 4.72 in (120 mm); however, all primary source material from Lancia indicates the bore was 120.7 mm (4.75 in). The engine’s total displacement was 1,508 cu in (24.7 L). It produced 320 hp (237 kW) at 1,380 rpm and 380 hp (283 kW) at 1,420 rpm. The Tipo 4 engine was direct drive and weighed 740 lb (335 kg).

Lancia V-12 aircraft engine top

A good view of the Lancia Tipo 4 showing the two magnetos, open gear train, coolant manifolds, and hand crank at the rear of the engine used for starting.

The Tipo 4 aircraft engine was built in 1916. It was installed in the Caproni Ca 37 and Ca 38 aircraft. These relatively fast aircraft were light-bomber / ground attack prototypes. The Ca 37 flew in 1916, and the Ca 38 was a more refined version of the aircraft that flew in 1917. Neither aircraft entered production, and it is not clear if the Tipo 4 engine was installed in any other types.

At least one Tipo 4 engine was shipped to the United States in late 1917. Thomas Evarts Adams, Inc represented Lancia & Company in New York and initiated the process of producing the engine in the United States. The engine was on display until early 1918 when it was sent to McCook Field, Ohio for testing by the US government. The Tipo 4 was tested in May and July 1918 and did not develop the anticipated power. On test, the Tipo 4 produced 279 hp (208 kW) at 1,250 rpm and 305 hp (227 kW) at 1,400 rpm. Plans for producing the Lancia Tipo 4 V-12 never moved forward. The end of World War I caused a large influx of surplus aircraft engines that left aircraft engine manufacturers with a very small market. In addition, the US government was interested in the more powerful Tipo 5 (Type 5) engine that Lancia was designing. A Tipo 4 engine is preserved in the FIAT/Lancia Archives in Turin, Italy.

Caproni Ca37 Lancia Tipo 4

The Caproni Ca 37 was the first aircraft powered by the Lancia Tipo 4 V-12 engine. The Ca 37 first flew in the summer of 1916. Note the engine’s exhaust tips. The Ca 37 had a top speed of 103 mph (165 km/h.)

The design for the Lancia Tipo 5 V-12 engine was well underway by the end of 1918. The Tipo 5 was very similar to the Tipo 4; however, there were a number of differences between the two engines. The Tipo 5 had a larger bore and a 53-degree angle between its cylinder banks. The Tipo 5 did not have offset crankpins; the engine used an early style of a fork-and-blade connecting rod design. The straight fork rod was relatively thick, and this size allowed the blade rod to connect to the same crankpin via a cut-out section of the fork rod. The Tipo 5 did not have the deep oil pan like the Tipo 4. The Tipo 5 had a 5.91 in (150 mm) bore and a 7.09 in (180 mm stroke). The engine’s total displacement was 2,329 cu in (38.2 L), and it was forecasted to produce approximately 600 hp (447 kW) at 1,700 rpm. The Tipo 5 weighed 992 lb (450 kg). No verifiable proof has been found that a Tipo 5 engine was ever built.

Caproni Ca 38 front

The Caproni Ca 38’s fuselage and tail booms were more rounded and streamlined compared to those of the Ca 37, but the aircraft were otherwise very similar. The Ca 38 flew in 1917 and had a top speed of 106 mph (170 km/h). The Ca 37 and Ca 38 never entered production. They are the only known applications of the Lancia Tipo 4 engine.

Through the early 1920s, Lancia designed at least two additional V-12 engines for automotive use, one of which had a cylinder bank angle of 14 degrees (connecting rod angle was 22 degrees). None of the V-12 engines entered production. However, these engines led to a range of narrow V-8s and V-4s that Lancia produced starting in the 1920s. Narrow V-4 types were in production until the 1960s.

Lancia V-12 aircraft engine Section

Sectional view of the Lancia Tipo 5 V-12 aircraft engine showing a 53-degree angle between the cylinder banks. Note the long rocker arms, horizontal valves, and small space above the combustion chambers. This configuration was very similar to the Tipo 4 and early Duesenberg engines.

– Correspondence with Geoff Goldberg, Lancia Historian
Textbook of Aero Engines by E. H. Sherbondy and G. Douglas Wardrop (1920)
Aerosphere 1939 by Glenn Angle (1940)
– “To Build Lancia Airplane Engine,” Automobile Topics (17 November 1917)
Air Service Handbook by Iskander Hourwich (1925)
Los Motores Aeroespaciales: A-Z by Ricardo Miguel Vidal (2012)
The V-12 Engine by Karl Ludvigsen (2005)
Aeroplani Caproni by Rosario Abate, Gregory Alegi, and Giorgio Apostolo (1992)
Gli Aeroplani Caproni by Gianni Caproni (1937)