During World War I and thereafter, several air forces developed armoured “ground attack fighters” with the layout of a conventional fighter, but much heavier armour. However, these aircraft were not expected to be fully effective fighters.
One of the first dedicated single-seat fighters with armour installed was the Polikarpov I-16 Typ 4, which flew in 1934, with full-scale production starting in 1935. In the final batch of the Typ 4 and on later models, a small plate of 8 mm thick headrest armour was installed. The windscreen remained a simple sheet of curved plexiglass. Nevertheless the little Soviet fighter, the most advanced single-seat fighter of its day, was once more far ahead of its time. Aviation armour had been under consideration in the USSR since about 1930, with the development of suitable nickel-molybdenum steel alloys. Most nations did not install armour in their fighters before 1940, and some waited much longer.
For example, the Hawker Hurricane and Supermarine Spitfire both entered production without any armour plate. The necessity was quickly understood after the outbreak of the WWII, and modifications had a high priority. Most of the RAF fighters to participate in the Battle of France, and that was most of the strength, did not yet have armour installed, but all fighters were modified before the Battle of Britain began. For the Spitfire this included 33 kg of armour plate, and an externally bolted-on armourglass windscreen, which cost nearly 10 km/h in speed. Later the armoured windscreen was internalized, and the armour increased.
At about the same time the Germans installed armour in their fighters. The Messerschmitt Bf 109E-4 version introduced a more angular cockpit with an armoured windscreen and an angled armour plate behind the pilot’s head. The 8 mm armour plate was also retrofitted to older models. The later G-model introduced a cockpit canopy with even more armour and a 90 mm thick windscreen. The heavily framed and armoured Bf 109 canopies were criticised for restricting the view of the pilot, but they offered good protection. Much later, at the end of the war, the Erla Haube was fitted. This new canopy, also rather inaccurately called the “Galland Hood”, offered a considerably improved field of view.
Combat experience from Europe soon reached the USA. British representatives ordered aircraft from American manufacturers, but they demanded modifications to make them combat-capable, including armour. In addition, they supplied examples of captured German equipment for evaluation; this e.g. offered the US Navy the opportunity to perform firing tests on a Bf 110. Hasty modifications of US fighters followed. For example, the Bell P-39 Airacobra was first designed and flown without any armour, but in late 1939 not less than 120 kg of armour was added. Installing self-sealing fuel tanks added another 109 kg to the empty weight. The USAAF demanded these modifications for the P-39D model, but at first it insisted that they would be made at no extra cost and with no reduction in performance! It was soon forced to adopt a more realistic attitude. Perhaps Bell was being overly generous with armour plate. In the successor to the P-39, the P-63 Kingcobra, the weight of the armour was reduced to 55 kg.
Maybe a bit slower to react, the US Navy installed 68 kg of armour plate in the Grumman F4F Wildcat from the summer of 1941 onwards. But its main opponent, the Japanese Navy, neglected to armour its fighters. The first version of the Mitsubishi A6M Reisen (Zeke) “Zero” to carry armour behind the pilot’s seat was the A6M5c, which entered service in the autumn of 1944! By then even the Japanese Army had had 13 mm armour plate in its fighters for two entire years. The F4F was in many ways inferior to the A6M, but it could survive the fire of the Japanese fighter, while the A6M was incredibly vulnerable. Later US Navy fighters outperformed the A6M and were well protected against .50 and even 20 mm hits. This helped the USN pilots to survive, even if their aircraft were quite often impossible to repair on board of the carriers and had to be dumped. The failure of the IJN to protect the lives of the pilots contributed to the rapid and fatal depletion of its trained cadres. When the A6M5 finally entered service, there were few experienced pilots left and the training of the new pilots was very poor.
Of course fighters that entered service during the war had the benefit of experience, which allowed a more efficient distribution of armour. The Focke-Wulf Fw 190 had a 13 mm plate to protect head and shoulders of the pilot, 8 mm seat armour, some 5 mm and 6 mm plate to fill in the gaps around the seat, and an armoured windscreen 50 mm thick. Armoured rings of 5.5 mm and 6.5 mm were installed around the lip of the engine cowling. An unique modification was the Fw 190A-8/R-8, modified to attack US heavy bombers from a close distance. Most fighters were protected only against from the rear and front. But the /R8 modification provided protection against fire from the sides as well, because this could be expected when the fighters got close in the bomber formations. The nose and headrest armour were made heavier, 30 mm armourglass was fitted to the side of the canopy, and 5 mm plate was installed at the sides of the cockpit and behind the instrument panel. The wing ammunition boxes for the 30 mm cannon were also protected, for any explosion of this ammunition would be fatal.
Self-sealing fuel tanks were as important as armour. Early attempts involved covering the inside or outside of a metal tank with some soft material, which expanded in contact with fuel, to seal any bullet holes. But this was not very effective, and it was soon discovered that the bullet entry holes were a comparatively minor problem. The exit holes made by the tumbling bullets were considerably larger. Worse, the shock of impact and the pressure wave inside the tank caused it to rupture. In the first American tests, the entry holes were small, but the entire rear of the tank was knocked out. [The Story of the Self-Sealing Tank, in US Naval Institute Proceedings, February 1946. Page 205.] The answer was a flexible fuel cell of self-sealing material, with as few seams as possible, and suspended in straps so that it could absorb shocks without rupture. Such a tank should not be in direct contact with the fuselage skin, because the moving tank could cause the skin to buckle, the torn metal skin could cut into the tank, sparks were often generated when the projectiles passed through the metal skin, and the skin might trigger explosive rounds.
Evidently, self-sealing fuel tank installations were costly both in weight and in volume compared with conventional fuel tanks. And of course there was also a limit to their usefulness. The US Navy designed its self-sealing tanks to resist .50 hits and found that they also offered some protection against 20 mm hits. But if an explosive round blasted a large hole in the wall of the tank there was no hope to seal it. For high-altitude aircraft the fuel tanks had to be pressurised, but that made sealing far more difficult. Hence self-sealing tanks were increasingly replaced by integral fuel tankage after the war, despite the higher vulnerability.
The risk of explosion did only exist if there was a suitable fuel/air mixture. A leak would of course provide such a mixture, but there was also a risk if an incendiary or explosive projectile entered the tank. Soviet designers found a solution: The fuel tanks were pressurised with cooled and filtered exhaust gases. The Lavochkin LaGG-1 of 1940 had 10 mm seat armour and self-sealing fuel tanks with such a fire surpressing system. It was also installed in other Soviet fighters. A disadvantage was that the exhaust gases tended to react with the self-sealing material, and it was preferable to use the system only in combat zones. Another feature of the Soviet fighters was that instead of a headrest with armour plate they had a slab of armoured glass installed behind the pilot’s head, to improve the view towards the rear. Similar installations were made in the Bell P-39 and P-63, of which large numbers were delivered to the USSR.
How effective was the armour? It’s thickness varied from 8 mm to about 13 mm. The armour was certainly effective against rifle-calibre machineguns, but these weapons were increasingly replaced by far more powerful medium-calibre machineguns or by cannon. The American .50 AP M2 round, a projectile with a high muzzle velocity, was expected to penetrate 1 inch (24.5 mm) at 100 yards (91 mm) and the AP-I M8 round still 7/8 inch. However, such armour penetration figures are traditionally measured against a homogeneous “standard” plate, while the armour plate fitted to aircraft would be face-hardened plate of good quality, to achieve maximal protection for minimal weight. Also important was that before it could hit the armour, the projectile had to pass through the aircraft skin and maybe structural members, which would deflect it or slow it down and was likely to cause tumbling, which would considerable reduce armour penetration. In this way relatively thin plates could greatly increase the protection. Equipment in the aft fuselage could be carefully arranged so that the bullet would have to pass it first, before it could hit the pilot. Finally, typical firing distances were of the order of 300 yards. Most airforces seem to have felt that the armour of their fighters offered substantial protection against .50 and even 20 mm rounds.
The Spitfire F Mk.21, a late war model, was considered protected against German 20 mm AP rounds in a 20 degrees cone from the rear, and against 13 mm rounds from the front. The US Navy expected fighters to carry armour able to stop a .50 rounds at 200 yards. Early in the war the relatively slow projectiles of the Type 99-1 cannon were often stopped by the armour of the F4F. Protection against US .50 rounds was the required standard for German fighters. Indeed it would not have made much sense for most German aircraft to carry armour that would not stop the .50 at combat distances, for this was the standard weapon of the USAAF, the enemy that was most often met in daylight combat.
Armoured glass windscreens were more difficult to make in sufficient strength while maintaining good transparancy, and armoured glass is also very heavy. The laminated glass panels developed for the B-17 were about 40 mm thick, and they would stop a rifle-calibre bullet at 100 yards. But these large panels weighed 88 kg per square meter (18 lb per sq. ft.). Fighter windscreens were smaller, and could be thicker and better supported; armourglass of up to 90 mm was used. Even so the front remained less well protected than the rear. In single-engined fighters the pilot was protected against fire from the front by the engine. Protection of the engine itself and the vulnerable cooling systems of liquid-cooled engines was almost impossible.
There were a few exceptions. The most heavily armoured aircraft of the war were close-support types such as the Ilyushin Il-2 and the Henschel Hs 129. Their role brought these aircraft within reach of small-arms fire from the ground, which was highly dangerous, especially after specialized anti-aircraft vehicles appeared. The armour of the Il-2 was part of the fuselage structure itself, in an attempt the save weight. Pilot and engine were enclosed in in welded shell, with a thickness varying between 4 mm and 12 mm. The windscreen was 65mm thick. The vulnerable coolant radiator was protected by installing it inside the fuselage, behind the engine; air was ducted to it from an intake on top of the cowling. The weight of all this armour was no less than 990 kg, and accounted for the Il-2’s rather sluggish performance. The rear gunner’s cockpit was not armoured, and it is claimed that the casualties of the rear gunners were seven times higher than those of the pilots. Il-2 losses were rather high, because the aircraft did very dangerous work and could not hope to evade enemy fighters.
The German Hs 129 represented a slightly different philosophy. Again, the pilot sat in an welded armoured box, 6 mm to 12 mm thick and with a 75 mm windscreen. Although the box was as small as possible — so cramped indeed that the gunsight had to be installed outside the cockpit — the armour weighed 1075 kg. Two air-cooled engines were used instead of a single liquid-cooled engine, thus removing the problem of protecting the cooling system. In the Hs 129B production versions the engines were captured French Gnome-Rhône 14M radials, which turned out to be rather unreliable. Performance was poor, but like the Il-2 the Hs 129 was an effective anti-tank aircraft, although it too suffered high losses.