The early development and use of aircraft cannon


© Anthony G Williams


This article is a summary of information in 'Flying Guns – World War 1: Development of Aircraft Guns, Ammunition and Installations 1914-32' by Anthony G Williams and Emmanuel Gustin.

Last amended: August 2014


The rifle-calibre machine gun was the supreme air-fighting weapon during the First World War and for the next two decades, but some nations put a great deal of effort into developing weapons of 20 mm and larger calibre, known today as 'cannon'. Aircraft equipped with cannon were never available in large numbers during the Great War and most weapons developed saw little more than experimental use. Between the Wars, use was even more desultory. However, these weapons represent a fascinating side-branch from the main trunk of armament development, which was only to come to fruition during the next great conflict a quarter of a century later.

Most of these large-calibre guns were manually-loaded; they only fired one shot at a time, and a gunner had to insert each cartridge into the breech by hand. A few of these were recoilless, firing a counterweight out of the back of the gun to balance the recoil of the shell leaving from the front. Some, but not all, featured automatic breech opening and ejection of the fired case immediately after firing, powered by the recoil, a feature known in artillery terms as "semi-automatic". Fully automatic aircraft cannon, which were in effect big machine guns, were only just achieving service status at the end of the war and their development saw only very gradual progress until the 1930s.

The smallest calibre of cannon available before the First World War was 37 mm because of the Declaration of St Petersburg in 1868, which renounced the use of any projectile weighing under 400 g which was either explosive or "charged with fulminatory inflammable matter" (observance of this agreement, at least as far as aircraft ammunition was concerned, effectively ended during World War 1). Most cannon shells were hollow in order to contain high explosive (HE) and were fitted with a fuze to detonate the contents. Other types of ammunition used were incendiary shells, some of which had large holes drilled near the nose of the shell from which the long-burning incendiary flamed out as the shell flew towards its target; and canister shells, which were filled with a quantity of small balls just like a giant shotgun. In British practice most cannon were referred to by the approximate weight of the shells rather than by calibre, so guns were for example known as 'One pounders' (1 Pr or 1 pdr).


French Pioneers

The earliest and the most persistent users of cannon-armed aircraft were the French.  Despite the fact that early aircraft had considerable difficulty in lifting themselves and their pilots into the air, experiments with fitting large-calibre cannon took place even before World War 1. In 1910 the Frenchman Gabriel Voisin installed a 37 mm cannon in one of his aeroplanes for publicity purposes, but it never flew with this armament. The next year, ground tests of a 37 mm Hotchkiss mounted in front of the propeller of a Blériot 11 wrecked the aircraft. In 1913, a 37 mm gun was finally fired from a "Voisin Canon," a modified Voisin model 1913 with a 37 mm Hotchkiss cannon and a 200 hp engine.

The first of the French 37 mm airborne cannon was the powerful M1902 "Tube Canon" which fired a 37x201R cartridge. The other common French aircraft weapon in this calibre was the short-barrelled naval Hotchkiss M1885, chambered for the low-velocity 37x94R cartridge. The M1885 was later modified with a smoothbore barrel to fire the "shotgun cartridge" shells in some fighters, as a rifled barrel disturbed the pattern of shot. 

Various sub-types of the Voisin aircraft were available with these cannon throughout the war, flexibly mounted in the front gun position of the pusher aircraft. The Breguet 5Ca2 was also originally fitted with a 37 mm cannon for bomber escort purposes, as was the Caudron R.14 which appeared too late to see action.

  37mm Hotchkiss M1885 in the nose of a Breguet Type 5 (courtesy Harry Woodman)

Even larger guns were tried, including 47 mm Hotchkiss guns which were fitted to some of the last Voisins (for ground attack) and a Tellier flying boat (for anti-submarine use; in British tests this gun managed to hole a submarine in Portsmouth harbour). Even a 75 mm gun, with reduced muzzle velocity compared with the artillery piece, was reportedly tested in a huge four-engined Voisin triplane in 1918. The ultimate cannon plane was still under construction at the end of the war, and never flew; the four-engined "Henri Paul" was intended to be fitted with a 75 mm gun  and several 37 mm for self-defence.

 47mm Hotchkiss in the nose of a Tellier flying boat, for anti-submarine use (courtesy Harry Woodman)

The original purpose of the French cannon was apparently to attack aircraft, and cannon-armed planes were used as bomber escorts. The first recorded engagement between a Voisin and a German plane was in May 1915, but without result. Some success was achieved in shooting down balloons during the Verdun battle, and in March 1916 a Breguet 5Ca2 shot down an LVG with its cannon. Despite this success, the slow-firing cannon stood very little chance of hitting another aeroplane in the air, but the French persevered, drawing up in the spring of 1916 detailed specifications for cannon-carrying aircraft, designated Class D. The D1 types used the low-velocity M1885 37 mm gun and were intended for air combat, while the D2 models had the high-velocity M1902 version and were used for anti-balloon work and ground attack. Cannon-armed aircraft were never very common; by 1 February 1916 there were just twenty-five at the front, and only about sixty by August 1917. One of the last of the pusher planes, the Breguet 12Ca2 of 1917, was equipped with a searchlight as well as a 37 mm gun and stationed for defending Paris against Zeppelins.

Despite the problems of air-to-air shooting, the French fighter ace Guynemer was interested in the possibility of installing an engine-mounted cannon between the cylinder banks of the geared Hispano V8 aero engine, firing through the hollow propeller hub, and he inspired such an installation in the SPAD 12Ca1. There were two different types of 37 mm cannon available; some confusion as to their origins exists but it appears that one was a conventional SAMC design with a rifled barrel, the other was a modified M1885 smoothbore firing canister shot. They are often referred to as "Puteaux" guns but this might just refer to the arsenal where they were made. A Vickers machine gun was also carried. The plane emerged in July 1917 and a number were built (although nothing like the 300 ordered), several pilots, including Guynemer, achieving some successes with it. These weapons were still manually loaded, however, and unpopular with most pilots because of their awkward loading and the propellant fumes which filled the cockpit on firing. Only eight were reported to be at the Front on 1 October 1918.

Forty SPAD 14 floatplanes, and some of the SPAD 24 landplane version, were also ordered with a 37 mm cannon, of which a few may have reached service. Attempts were made to develop automatic loading cannon, but these were too late for the war. Some French airships reportedly carried 47 mm and even 75 mm guns but it is doubtful if they were ever used in action.

The guns used by the French were all manually loaded. Despite being the major users of large-calibre aircraft cannon, the French failed to get an automatic version into service before the end of the War. In 1918, two long-recoil AMC models were being developed at the Puteaux Arsenal, a low-velocity one using the usual 37x94R cartridge and a high-velocity model designed around the 37x190 ammunition of the British C.O.W. gun. The end of the War reduced the impetus for these developments and it appears that neither saw service.


British Experiments

The British also put much effort into developing cannon-armed aircraft but with far fewer results, probably because they ambitiously aimed for fully-automatic cannon at a time when these were too primitive to be reliable. Despite this, there was some interest in manually-loaded guns and the first installations, specially designed pusher "Gun-carrier" seaplanes from Shorts (S.81) and Sopwith (No.127) equipped with a semi-automatic 1½ Pr (37 mm) Vickers Mark B gun (the Mark A being the fully-automatic naval version). These were tested by the Admiralty, with reasonable success, by firing them at a floating target in May 1914, although the effect of the recoil on the flying characteristics was reportedly noticeable. The British were also the main users of the Davis recoilless guns (RCLs).

The British interest in cannon was principally for use in ground and naval attack rather than air fighting. They were also keen on upward-firing weapons to attack airships, although this proved ineffective. At the outbreak of the war Vickers had started work on a heavy fighter designed around the 1 Pr Vickers (Maxim) gun; the automatic 'Pom pom'. The F.B.7 was a relatively conventional twin-engined biplane, with the gunner and gun in the nose. In August 1915 a dozen were ordered, but Vickers soon recognised that the type had no future and convinced the War Office to cancel the contract. The concept was carried over the in smaller and lighter F.B.8, armed only with a Lewis machine gun, which it was hoped would possess superior performance to single-engined aircraft, but it did not and was abandoned.

The British made hardly any use of manually-loaded cannon, except for the Davis guns mentioned below. An exception was the 1.59" Vickers-Crayford (originally known as the 1.45" Vickers Light Gun) which had the attraction of being very small and light for the calibre. This became popularly known as the Rocket Gun after an early misunderstanding as to the nature of the weapon; in fact it was a conventional gun firing a 40x79R cartridge (a shortened version of the common 40x158R naval AA round). The nickname was probably the result of firing incendiary shells, which were set alight on firing and burned throughout the flight, doubtless looking spectacular at night. It was fitted with an imposing muzzle brake to reduce the recoil as it was intended that the mounting would be interchangeable with the .45" Maxim which Vickers then had in hand. Although the Crayford was approved for air service in 1917, the simple, artillery-type manual breech mechanism was so slow to reload that the gun saw little use. They were fitted to some FE2bs of Nos. 100 and 102 squadrons in April 1917 and tested on night operations. Experience was mixed; some liked it and there is one report of stopping a train after firing 30 rounds at it, but other reports refer to unreliable ignition, leading to delayed 'hang-fires' due to weak firing springs and defective primers. The Crayford was subsequently fitted to (or planned for) three different aircraft; the Parnell Zeppelin Scout of 1916, R.A.F. N.E.1 of late 1917 and the Vickers F.B.25 of early 1917. In all of these cases, the gun was intended for anti-Zeppelin use and could be fired upwards, but none of the aircraft entered production.

The 1.59inch Crayford Gun (top) and 1½ Pr C.O.W. gun (Courtesy MoD Pattern Room

More interest was shown in automatic cannon and two different types saw service; the 1½ Pr C.O.W. gun (so-called because it was developed by the Coventry Ordnance Works) and the 1 Pr Vickers Mk III. The predecessor of the C.O.W. gun – the 1 Pr – was actually test-fired from an F.E.3 in 1913 (although the plane was suspended from a gantry rather than flying) and was also intended for the F.E.6 of 1914, but probably never was fired in the air, and attention soon switched to the 1½ Pr. The appearance was distinctive in that the recoil spring was wrapped around the barrel in the interest of compactness. Ammunition was fed to the gun via a five-round clip, meaning that only a short burst could be fired before reloading (the cyclic rate of fire was 100-120 rpm). Development work took a long time and it was not until August 1917, following successful trials of the Mk.III gun, that an order was placed for 90 of these weapons; the maximum which could be produced without affecting the output of other guns. By August 1918 the orders had increased to 450 and the urgency of the order was stressed but problems continued and only fourteen had been delivered. The end of the war saw the cancellation of the orders with only those under construction being completed, and in all about 76 guns were delivered.

Only in October 1918 was a British machine available with a cockpit designed for the gun; it was fitted in the rear cockpit of three D.H.4 light bombers, angled upwards at 45º for use against the giant R-plane bombers as well as Zeppelins. A hole was cut into the upper wing for the gun to fire through. Aiming was the job of the pilot, who instructed the gunner/loader when to fire. Various mounting, handling, sighting and buffer problems were resolved and over 1,000 rounds fired but the muzzle blast caused problems and the plane required light alloy plating to protect it, as well as some structural strengthening. In the event, just three D.H.4 aircraft fitted with this gun entered service in November, and only two of these reached the front, but they had no time to see action before the Armistice.

 1½ Pr COW gun mounted in a DH4, late 1918 (courtesy Harry Woodman)

The gun was also tested in the Tellier flying boat, a Voisin 8, two D.H.3A and a D.H.10 twin-engined bomber and was intended for the three-seat F.E.4 of 1916. Two C.O.W. guns were proposed in August 1918 as defensive armament for the big Handley-Page V/1500 bomber.

After the War, the C.O.W. gun was tried in various installations, especially in large flying boats in which the gun was intended primarily for anti-submarine work. The twin-engined Armstrong Whitworth Sinaia, the Short Cromarty and the experimental Vickers Valentia, all made in 1921, had provision for one in the nose. In the late 1920s the Blackburn Iris and Perth, similar three-engined biplanes built in small numbers, were also available with a C.O.W. gun in their bow position instead of the usual Lewis, and in 1932 the Short Sarafand had a similar installation.

1½ Pr COW gun mounted in a Blackburn Perth flying boat

Some landplanes were proposed as mounts for the big cannon. The Bristol Bagshot and Westland Westbury were twin-engined three-seat heavy fighters, designed to a 1924 specification for bomber-destroyers and based around the C.O.W. gun. The Bagshot was a high-wing monoplane, the Westbury a biplane, but both were armed with two flexibly-mounted cannon in nose and dorsal positions. The muzzle blast was recognised to be severe enough to damage any normal structure so strengthening had to be applied, and because of severe vibration on firing it was necessary to fit instruments and sensitive equipment on flexible mountings.

Perhaps the most remarkable were the two experimental C.O.W. gun fighters of 1931, designed to specification F.29/27 by Vickers and Westland. These single-seat, single-engined planes not only had a 37 mm C.O.W. gun as their sole armament, it was angled upwards at 45-55º for attacking bombers from below. The installation was based on the theory of the "no allowance angle," at which the body lift from the projectile would compensate for gravity drop over an important part of the trajectory.

Despite all of this work, the only significance the C.O.W. gun achieved was to be used as the starting point for the Vickers 40mm Class S design, after Vickers had taken over the Coventry Ordnance Works. The remaining C.O.W. guns saw out their days on AA mountings as airfield defence weapons (some mounted on lorries) in the Second World War. They were not finally declared obsolete until 1948.

Vickers 1 Pr Mk III with its 37x69R ammunition (Courtesy R.W.Clarke)

The Vickers 1 Pr Mk III was in many ways the most promising of the First World War aircraft cannon, as it was light and compact, belt fed (so no need to change or reload magazines) and fired a shorter and less powerful cartridge (37x69R instead of 37x94R) than the usual 1 Pr Maxim with which it is often confused. It was first installed in the Vickers F.B.7 of 1915, then in 1917 was fitted together with searchlights in five F.E.2b aircraft for trials in the home defence role. The gun was flexibly mounted in a widened front cockpit, with the gunner sitting next to the pilot. Two other F.E.2b with this gun were sent to France and carried out night ground attack missions during the summer of 1917 with some success, particularly against trains. It was also planned for the Curtis C and the Martinsyde F.1 and Vickers F.B.24 two-seat fighters of 1917.

The use of the gun was not without problems; the recoil caused damage to the structure and loose spent shell cases were sometimes blown back into the pusher propeller. The Mk III might have made a viable fighter weapon in a fixed installation, but the heavy mounting, plus the weight of the gunner, rather negated its compactness and light weight.


The Davis Recoilless Guns

Early aircraft were very frail and found it difficult to cope with the recoil as well as the weight of a powerful large-calibre cannon. There was therefore much interest even before the war in the recoilless gun (RCL) developed by the USN Commander Cleland Davis, which was available in three sizes; 2 Pr (40 mm), 6 Pr (57 mm) and 12 Pr (76 mm). The Davis guns had a breech which was open to the rear and used the counterweight principle, in which the cartridge fired a weight (normally lead shot) to the rear which was equal to the projectile weight, cancelling out the recoil.

The lack of recoil force meant that the Davis guns could be made very light (although long, as they effectively had two barrels, back to back) and also only needed a light and simple mounting. However, the cartridge cases were very big as RCLs required much more propellant, and reloading was cumbersome and very slow. It was achieved by unlocking the two halves of the gun, then pulling the rear half backwards before rotating it along an axis parallel to that of the barrel to clear the breech for extracting and replacing the cartridge.

A sectioned 6 pr Davis gun round

In the USA, a gun was mounted in the front of the cockpit of a Curtiss Twin JN aeroplane, aimed by means of a tracer-firing Lewis gun attached to the barrel.

The British carried out trials with the 2 Pr and 6 Pr Davis guns in late 1916, and tested incendiary shells in these and the 12 Pr calibre. They were envisaged as being useful against airships (an order was placed for 12 Pr incendiary shells to be tested against a "Zeppelinette"; an airship-type target) and also ground attack, with "buildings, stores, ammunition dumps and trains" being identified as possible targets. The 2 Pr (40 mm) version was successfully fired from a Short 184 in April 1916, hitting a seaplane target which was being towed by a motor boat, but this installation could not be reloaded in flight.

Various anti-Zeppelin night fighter aircraft were intended to use this weapon, probably the best known examples being the large quadruplanes built by Pemberton-Billing. The P.B.29E of late 1915 was a twin-engined aircraft in which the fuselage was attached to the second wing, counting from below, and a streamlined gunner's nacelle filled the gap between the third and fourth wings – a 'fighting top' in the contemporary parlance. It was succeeded by the P.B.31E, which flew in early 1917. Its cockpit was faired into a towering structure that connected the fuselage between the second and third wings to gun positions on the upper wing carrying a Davis gun forward and a .303" Lewis on a gun ring aft. The nose contained another gunner, with a Lewis. The purpose was to mount standing patrols against airships, and it was claimed that the P.B.31E could stay in the air for up to 18 hours. But the ponderous aircraft was too slow to catch an intruder and only the prototype was completed.

Also intended to be armed with the Davis gun was the Robey Peters R.R.F.25, designed to an Admiralty requirement. This was a large single-engined tractor biplane, with streamlined nacelles for two gunners attached to the upper wing, outboard of the propeller disk. The armament of the prototype consisted of a Lewis in the port nacelle and a Davis 2 Pr gun in the starboard nacelle, with ten rounds. Two prototypes were built in 1917, the second being intended to carry a Davis gun in each nacelle, but both aircraft crashed during testing. The Dutch designer Koolhoven created the Armstrong-Whitworth FK.5 and FK.6, sizeable single-engined triplanes, with two small nacelles attached to the central wing, outboard of the propeller, so that the gunners would sit in front of the wing. The FK.6 may have been intended for Davis recoilless cannon, for in the spring of 1916 Armstrong-Whitworth received wooden mock-ups of the 2 Pr and 6 Pr weapons, but it was only briefly tested.

The only service installations of the 2 Pr Davis gun appear to have been in a couple of D.H.4s at Dunkirk in 1917, and an R.E.8 in the Middle East in 1918 in which the gun was aimed forwards and downwards at a 45º angle for ground attack.

The 6 Pr (57 mm) Davis gun was tested in the Short S.81 Gun-carrier in 1915 and subsequently fitted to a number of other aircraft for trial purposes; a B.E.2c in 1916 (it could be angled to fire 45º up or down), a Short 310-B seaplane in April 1917 (aimed upwards for anti-Zeppelin use), a Felixstowe Porte Baby flying boat, and a B.E.12 (in which the gun was mounted at an upward angle of 45 degrees for firing but could be lowered to the horizontal for loading). Trial installations of both 2 Pr and 6 Pr weapons were made in Handley-Page O/100 and O/400 and several of these large bombers were fitted with the 6 Pr, for ground attack and anti-submarine duties. It appears they were of little use, however, and were withdrawn before the end of the war.

A Davis gun fitted to a Felixstowe fling boat for anti-submarine use. The Lewis gun was used for sighting.

There does not appear to be any record of the 12 Pr Davis gun being used in aircraft apart from initial tests. In 1915 the Admiralty issued a contract for a large seaplane, according to specifications drawn up by the Air Department for an anti-shipping aircraft. It was desired that the aircraft would be able to carry three alternative weapon loads: An 18-inch Whitehead Mk.VII torpedo and one Lewis; twenty-two 100 lb bombs and one Lewis; or a Davis 12 Pr recoilless cannon, firing 5.4 kg shells, and two Lewis machine guns. Extensive protection with armour plate was called for. To meet this specification Samuel Whites constructed the Wight A.D.1, a very large twin-boom seaplane. For some reason, perhaps because there were concerns about the blast and counterweight shot to the rear by the Davis gun, the design was then changed to accommodate a conventional 12 Pr Naval Landing Gun, with elevation ranging from ‑38 to +49 degrees. The first prototype turned out to be so overweight that it could barely fly; the second prototype was better streamlined and dispensed with some of the armour plate, but it never flew. The 12 Pr Davis gun never entered service although there was a British scheme in 1916 for a Dyott "Battleplane" to be fitted with one of these cannon, firing sideways while the aircraft circled the target. This was not built, but is remarkably reminiscent of the USAF's gunships of more than half a century later.

Work on the Davis recoilless guns continued for a while after the War. Following a series of experiments with all three calibres of weapon the British concluded in April 1919 that "no type of machine could be found which would enable a useful zone of fire to be obtained owing the restrictions imposed by the rear charge" and the RAF took the decision to abandon all Davis guns.  The USA also continued experiments with these weapons and were still testing them in 1921.


German Automatics

In complete contrast to the French, the Germans concentrated almost entirely on automatic weapons, only testing one large-calibre manually-loaded gun. A 13 cm Ausstossrohr ('launching tube') was mounted to fire downwards from the VGO.II R.9/15 bomber, and was tested on the ground and in the air during October 1916. This was apparently a recoilless weapon, but interest was then shown in a higher-velocity and almost certainly conventional 10.5 cm weapon, which was ordered in November 1917. Lt. Dr. Ernst Neuber designed an R-plane to carry this gun, firing downwards. It would have been a large but fairly conventional, eight-engined aircraft with a structure reinforced to absorb the 3000 kg recoil force. Besides the cannon and 110 to 120 rounds of ammunition, eight machine guns would have been carried for self defence. This aircraft was never built.

In June 1915 the German War Ministry issued a specification for an airborne machine cannon with a calibre not to exceed 3.7 cm, which should be able to fire ten rounds in quick succession and weigh not more than 70 kg.

Two automatic 20 mm cannon which met this specification were already under development, the Ehrhardt and the Becker. The 2 cm Ehrhardt FlzK (Flugzeugkanone, or aircraft cannon) was made by Rheinmetall. This fired a 20x70RB cartridge. It employed a scaled-up version of the Dreyse short-recoil design, introduced in 1912 in an infantry machine gun of that name, which used a pivoting block lock designed by Louis Schmeisser, a famous gun designer who ironically is best remembered in the popular name of a sub-machine gun with which he was not in fact involved. The Ehrhardt weighed 36 kg, was 150 cm long (with a 101 cm barrel) and fired at 250-300 rpm from a side-mounted box magazine. The locked-breech mechanism was considered superior to the Becker's by the APK (Artillerie Prüfungs Kommission = Artillery Test Commission) but it did produce more recoil and vibration, which led to many jams.

The Ehrhardt did make it into production and at least fifty-one were built (one surviving example is stamped '51'), but it was too late to see service. After successful attempts to conceal its existence from the Allied Control Commission, it was developed postwar by Solothurn, the Swiss subsidiary set up by Rheinmetall in 1929 in order to develop armaments away from the restrictions then in force in Germany. It formed the basis of several automatic canon used by Germany in World War 2, most notably the FlaK 30 AA gun.

2 cm Ehrhardt FlzK aircraft cannon (Courtesy MoD Pattern Room)

The most important German development specifically intended for aircraft use was undoubtedly the 2 cm Becker; it was the only German cannon to see air service in the First World War. The key design feature of the Becker was that the cartridge was fired just before it was fully chambered, while the bolt was still moving forwards at high speed. The initial rearwards impulse of the 20 x 70RB cartridge case (similar to the Ehrhardt's, but slightly slimmer) was therefore used up in stopping the forward movement of the bolt, before it could begin to drive the bolt backwards again. This delay gave sufficient time for the projectile to leave the barrel. Because the primer is fired before the cartridge is fully chambered, this operating principle is usually called advanced primer ignition, or API, blowback.

The API blowback is not only very simple, it has the added advantage of spreading the recoil energy over a long period of the firing cycle. Mountings for these weapons could therefore be simpler than with most other types of mechanism of equivalent power. The Becker was also light at 30 kg, plus 5 kg for a loaded 15-round box magazine; a 10-round magazine was also available. APIB has some disadvantages; it works best with straight cases with parallel sides, because the case needs to slide to and fro in the chamber as the cartridge is being fired. As with other blowbacks, cartridges usually need to be lubricated with grease or oil to prevent the chamber pressure from sticking the case to the walls. One inherent feature of the APIB design is that it has to fire from an open bolt, so these weapons could never be used in synchronised mountings since the lock time is far too long. The Becker was fitted with spade grips having two triggers; the right trigger for single shots, the left for fully automatic fire (the gun had a cyclic rate of around 300 rpm).

The concept for the Becker was reportedly developed by the Cönders brothers but was patented by Dr. Rheinhold Becker of the Stahlwerke Becker AG in the Rhineland in September 1913. At first, the new weapon faced some resistance from the German armament authorities because of the lack of relevant experience of the company, but the German Army Air Service showed interest and firing tests were ordered in mid-1915, in a Gotha-Ursinus G.I. These tests were unsatisfactory but the concept showed promise so more technical assistance to Becker was provided by the nearby Spandau Arsenal in order to develop a satisfactory weapon. In June 1916 Becker received a production contract for 120 cannon, while Spandau continued developing the gun. This was followed-up by field trials of five cannon in August of that year, in which explosive tracer ammunition showed good results (it originally fired solid projectiles to avoid the prohibition on small explosive ammunition, but HE shells were soon developed). There were still ammunition feed problems which were not resolved until the introduction of the Type 2 Becker in November 1916. In 1917 it was agreed that Becker would build guns for air service, while Spandau and MAN-Nürnberg would make them for the army, mainly for anti-tank use. The ammunition was modified, with a slightly fatter case being adopted for the final version; this bears a close resemblance to the Ehrhardt's cartridge. By May 1918 the development of the gun was considered essentially complete, although work continued on the feed arrangements with various drum and belt feeds being tested.

  2 cm Becker (Courtesy MoD Pattern Room)

There was a large number of projects and test installations of the Becker cannon, but very few of these saw any use. Although some applications were for air fighting purposes - it was fitted to some bombers and naval airships - it was primarily seen as a ground attack weapon, particularly for use against tanks. There were therefore two basic types of installation; in the usual ring-type flexible mountings, for use by a gunner, and those fitted to fire downwards for ground attack. In addition, they were a few projects for using the Becker as a fixed, forward-firing gun in fighters.

The main users were the Friedrichshaven G.IIIa night bomber and the armoured AEG G.IVk, also a twin-engined plane. The G.IIIa, several of which were delivered in September 1918, had a Becker in a nose mounting and used it to strafe ground targets when returning from bombing raids. A test was also made of a rear installation, and a ball turret for this aircraft was designed but not built. Five G.IVk were ordered in March 1918, following successful tests in a G.IV. These were fitted with one Becker in a dorsal ring mounting (able to fire downwards through a 'Gotha tunnel') and another in an armoured turret in the lower part of the nose. In addition, two MG 14 were used for self defence. The planes were successfully tested in April, and four survived to be handed over in January 1918, but their service use is not known.

  A 2 cm Becker (on the left) mounted along with an early model of the 7.92 mm MG 14 Parabellum (courtesy Philip Jarrett)

Unsuccessful or uncompleted projects using ring-mounted Beckers included the AEG J.I and C.IV, Dornier Gs.I flying boat, Gotha WD 7 floatplane (two completed and tested) and the Hansa-Brandenberg W 19 floatplane (twenty ordered in June 1918, but never used). Hansa-Brandenburg also planned to use the Becker cannon in its W.23, installed to starboard with a LMG 08/15 to port, but it never saw service as its flying characteristics were unacceptable. A Becker was also proposed for arming the W.35 twin-engined flying boat, but this was not built. The German naval air service was particularly interested in cannon in order to deal with the large and heavily armed Curtiss-type flying boats which were a proving a nuisance over the North Sea.

The first downward-firing installation was a rather makeshift effort in an Albatros J.I ground-attack plane, in which the Becker was fitted to the left-hand side of the aircraft, outside the observer's cockpit, allowing a limited range of movement. This was tested in the air in December 1917 and eight examples of this model were delivered to the Front for combat evaluation in the Spring of 1918. They were used successfully against rear-area transportation systems but it was noted that fighter escort was essential. The installation was criticised for its limited field of fire and the difficulty of changing the big magazines against the slipstream (a large handle was accordingly added to the magazines). One of these aircraft subsequently had the gun relocated to the floor of the observer's cockpit, which became the preferred location in single-engined ground-attack projects.

Other aircraft with downward-firing guns fitted in the observer's cockpit were the AEG J.II (twenty delivered in September 1918, use not known), Albatros J.II (not delivered), the LVG C.V (tested May-June 1918) and the AGO S.I Schlachtflugzeug (heavy ground attack - two completed October 1918, possibly never flown). The Albatros C.V was also tested with this installation, but with the curious twist that the pilot's controls were also located in the rear cockpit; it seems that the front observer was only expected to change the magazines. A (possibly fixed) downward-firing installation was also tested in the Gotha G.IV in mid-1918.

The Becker never saw service in fighters, although four projects involved it. The first was rather bizarre; it was fitted to an Albatros D.II fighter and tested in November and December 1916. The breech was in front of the pilot as usual, but because the gun could not be synchronised, it was angled up by thirty or forty degrees to miss the propeller. The tests reportedly included combat evaluation, but this was presumably unfavourable as nothing more was heard of the installation. The only satisfactory configuration for installing this weapon in a single-engined fighter was the pusher layout, and this was adopted by the AGO C.I (planned in October 1918) and the Albatros D.VI. The D.VI actually flew in February 1918 but was damaged on landing. It was intended to carry an LMG 08 in its nose as well as the cannon, but work on it was cancelled in June of that year. Finally, the Hansa D.I was a twin-engined fighter with a fixed Becker in the nose, It was still under construction at the end of the war.

It is unclear how many Beckers were made. Some 362 were discovered by the Allies after the war, of which approximately one-third were supplied to the Army for anti-aircraft and anti-tank use. However, 428 aircraft Beckers were reportedly made by MAN alone (despite the earlier agreement). Becker's own production was very small until late in 1918, 111  being delivered in September of that year.

In 1921 the patent for the Becker was acquired by the Swiss company SEMAG, who continued to market the weapon and developed their own, more powerful version, firing a 20x100RB cartridge instead of the original 20x70RB. In 1924 the rights to the guns were acquired by another Swiss company, Oerlikon, named after the suburb of Zurich in which the factory was based. They added a third weapon to the range, chambered for an even more powerful 20x110RB bottlenecked cartridge. By the late 1920s the three models were known as the Oerlikon F (20x70RB), L (20x100RB) and S (20x110RB) and were offered in various versions as aircraft, anti-aircraft and anti-tank guns. All three weapons were steadily improved through the 1930s, with increases in rates of fire and muzzle velocity and reductions in weight (during which their appearance was significantly altered), and they saw considerable use with several nations in the Second World War.

One rare interwar installation of the Oerlikon L was in the Savoia-Marchetti S.62bis flying boats in the early 1930s, which carried one in the bows. A manufacturer's catalogue promoting this installation adds some interesting detail. The ring-type mounting weighed 55 kg and could traverse through 360 degrees while  providing up to 90 degrees of elevation and depression (although depression was limited to 60 degrees in the SM.62bis installation). The ammunition feed was via a 15-round drum which weighed 2 kg empty (each round weighed 220 g). Sights weighed another 2 kg and a sack to hold 30 fired cases weighed 1.2 kg (necessary because the SM.62 was a 'pusher', so it was important to prevent fired cases from flying back and striking the propeller). Although the gun had a cyclic rate of 350 rpm, the practical rate, allowing for magazine changes, was 125 rpm. The complete installation weight, including six loaded magazines, was 135 kg.

The Oerlikon L in a flexible mounting (Courtesy Harry Woodman)

Some experimental 37 mm German automatic cannon were designed to the June 1915 specification.  It is known that prototype guns to this specification were made by Krupp, DWM (Deutsche Waffen und Munitionsfabrik) and  Rheinmetall (Rheinische Metallwaren und Maschinenfabrik), but no further details of the weapons appear to have survived. It would seem that they were considered too heavy for the aircraft in service at the time, although some were tested in aircraft. A Gotha-Ursinus G.I was tested with the 3.7 cm Rheinmetall cannon at the same time as the Becker trials in late 1915, and an AEG G.II carrying a Krupp or DWM cannon was shown to members of the Reichstag in April 1916. In 1917 a 3.7 cm DWM cannon was tested in two different Gotha WD 7 floatplanes, with air firing trials taking place between July and October, but the gun was rejected due to ammunition feed problems. No aerial tests of the 3.7 cm Krupp are reported, although four guns were made.

A lighter version of the Maxim pom-pom did see service; the Sockelflak. This fired its own ammunition, of 37x101SR calibre (450 g at 355 m/s) which was fed via ten-round clips inserted vertically over the breech. Although reportedly intended for aircraft (or at least airship) use, it was apparently only used in the AA role. It weighed 260 kg (with mounting), and fired at 120 rpm. The barrel was only 54 cm long.

3,7cm Sockelflak

In an advanced state of development at the end of the war was the Szakats, a 19 mm gun which was intended to be produced in various versions for ground and air use. These all fired 19x114R cartridges, and used a form of retarded blowback mechanism with an additional piston and cylinder (liquid-filled in the later models) to check the violent recoil of the bolt.


Other nations' efforts

The Austro-Hungarians were active in the development of large airborne cannon. They were responsible for fitting the 66 mm Skoda C95 L/18 Marine landing gun to the front cockpit of two large three-engined flying boats designated G3 and G6. It was used to fire at Italian naval craft (apparently without result) and thereby earned the distinction of being probably the largest-calibre weapon to be fired in anger from an aircraft in the First World War.  Experiments and proposals were also made concerning weapons in 37 mm, 47 mm, 50 mm and 70 mm calibre. The Hansa-Brandenburg G.I bomber was used for various experiments in 1916, including a 50 mm gun and (separately) a 70 mm Skoda cannon weighing 200 kg, both being fitted in the nose, and a 37 mm Skoda was test-fired from a dorsal mounting early in 1917. A 37 mm weapon was also installed in a small flying boat. By the end of the war they reportedly had one 20 mm gun (given as the Szanáts, but this might have been the Szakats) and three different 25 mm cannon (the Boykow-Czerney, the Szebeny and one designed by the Technischen Militärkomitee) being developed, but no details of these weapons or their ammunition have emerged.

The Italians used the French M1885 (as did various other nations) and also developed their own Vickers Terni aircraft gun around the more powerful 37x201R cartridge (Vickers and the Italian armaments works at Terni had set up a joint-venture factory). They referred to these guns as the 37/20.6 and 37/40 respectively, after the barrel lengths in calibres (i.e. the 37/40 barrel was 37mm x 40 = 1,480 mm long). The Vickers Terni plant also manufactured a Vickers 1 inch gun (not produced in the UK), a belt-fed long-recoil automatic which fired 25x87R ammunition at 150 rpm. This weighed 50 kg and was 137 cm long overall, with a barrel of 76 cm. The only recorded user of this gun was the Swedish Navy which adopted it postwar as the M/22 for arming submarines, but a photo does exist of what is clearly a Vickers 1 inch mounted on an Italian Caproni bomber, although the original caption identified this as a Revelli-FIAT (see below).

Vickers 1 inch gun in a aircraft mounting (Courtesy MoD Pattern Room)

Vickers 1 inch gun mounted on a Caproni Ca.3 bomber. Note the sturdiness of the mounting, the degree of protection for the gunner, and the attention given to shielding the pilot from the muzzle blast! (Courtesy Philip Jarrett)

Another automatic cannon made in Italy was the 25 mm Revelli-FIAT M1917. This also used a long-recoil mechanism, fired 25x87 ammunition very similar to the Vickers Terni, and was offered loaded with similar projectiles, but the gun was fed by a top-mounted 8-round magazine. It weighed 45 kg and was 133 cm long with a 67 cm barrel; the rate of fire was the same as the Vickers Terni. The appearance of the gun in different photos varied considerably. 

25 mm Revelli-FIAT M1917

The M1917 is stated to have been flexibly mounted in the nose of a Macchi L flying boat, a dozen Savoia Pomilio S.P.2, and a similar number of Caproni bombers, mainly Ca.3 but also Ca.4 and Ca.5, although given the error in identifying the Vickers 1 inch in the previous photo there must be some doubt about which guns were used. However, a sensitive nose fuze caused problems with premature detonations of the HE shells, and production had to be stopped when Vickers threatened to sue for breach of patent due to the similarities between this weapon and the Vickers 1 inch gun.

25 mm Revelli-FIAT cannon mounted in the nose of a Macchi L. Note the different appearance from the previous photo, and the twin-barrelled Villar Perosa (courtesy Harry Woodman)

The Russians did produce small numbers of one locally-designed gun, the manually-loaded 37 mm Obukhov, named after the steelworks near St Petersburg where the gun was built. It was produced just before or during the First World War, and reports indicate that ten were sent to the Baltic Sea Division in 1917 and probably a similar number to the Black Sea Division. The gun was fitted to, at least, one Grigorovich M.9 flying boat and presumably other aircraft as well. Ten were still on charge at the Krasnoyarsk hydro aviation base in 1919. No data have survived, except that it was available with case shot ammunition and in 1917 work was underway on a time-fuzed shell.  Apart from the Obukhov the only service use of cannon in Russia was of some Voisins armed with 37 mm Hotchkiss, probably the M1885. The Russians also fitted a 37 mm naval gun to an Ilya Muromets as early as 1914, mounted below the fuselage for ground attack, but tests showed that it was far too inaccurate so it was removed. Further tests in the following year of a 76 mm recoilless gun on the same type of aircraft were also fruitless, although that did not appear to discourage the development in 1916 of the unsuccessful Morskoi Kreiser (naval cruiser) floatplane, designed for a 75 mm gun of uncertain provenance.

37 mm Obukhov on a Grigorovich M.9 flying boat from the carrier Orlitsa, based in the Baltic (courtesy Harry Woodman/Y.J.Toivanen/A.Alexandrov)


            First World War cannon cartridges: 7.92x57 for scale, 2cm Becker (20x70RB), 25mm Revelli-Fiat (25x87), Vickers 1 Pr Mk 3 (37x69R; with sensitive No.131 fuze), Maxim 1 Pr pom-pom (37x94R), Vickers 1½ Pr Mark B (37x123R), French "Tube canon M1902" (37x201R), 1½ Pr C.O.W. gun (37x190), 1.59" Vickers-Crayford (40x79R), 47 mm Hotchkiss (47x131R)