Although it was to make a name for itself as a dive-bomber, most notably the sinking of the German cruiser Konigsberg in Bergen Fjord on 10 April 1940, the Skua was also tasked with fleet air defence and the type was responsible for shooting down a Dornier Do 18 flying-boat off the Norwegian coast on 25 September 1939, the first enemy aircraft to be claimed by the Fleet Air Arm (FAA). The Skua was designed by G.E. Petty to Specification O.27/34. K5178, the first of two prototypes, was flown for the first time on 9 February 1937 by Captain A.M. ‘Dasher’ Blake. Both aircraft were powered by an 840 hp Bristol Mercury IX engine, but production machines, designated Skua II, featured a Bristol Perseus XII sleeve-valve radial of 890 hp, driving a de Havilland three-blade, two-pitch propeller. The top speed of the Skua II was 225 mph at sea level with an initial rate of climb of 1580 ft/min and a service ceiling of 20,200 ft.
Despite its seemingly modest performance, the Skua represented a radical departure from the fabric-covered biplanes that it replaced. It introduced the all-metal cantilever monoplane to Fleet Air Arm service and also pioneered the use of landing flaps, a retractable undercarriage and a variable-pitch propeller. The two crew members were accommodated under a long canopy, which was protected by two reinforced fuselage frames in case the aircraft came to rest inverted. The fuselage was of flush-riveted Alclad and incorporated two watertight compartments, one under the front cockpit and the other to the rear of the gunner’s station. These provided buoyancy in case of ditching. The wings comprised a centre section bolted to the fuselage, with tapered outer panels, detachable tips and fabric-covered ailerons. Wing folding was incorporated, whereby the wings moved back about an inclined hinge and also twisted so that the leading edge was pointing upwards. The tail surfaces were metal cantilever structures with fabric-covered elevators and rudder, and the aircraft was fully stressed for catapult take-offs and arrested landings.
The armament comprised four 0.303 in Browning machine-guns mounted in the wings and a single Mk. IIIE Lewis machine-gun mounted on a Fairey Battle-type pillar for the gunner/observer in the rear cockpit. In the dive-bombing role, a 500-lb semi-armour piercing (SAP) bomb could be carried under the fuselage on a retractable ejector arm or, alternatively, eight 30-lb practice bombs on racks under the wings.
The Skua was first seen in the New Types Park at the RAF Display at Hendon on 26 June 1937. K5178 went to A&AEE at Martlesham Heath for handling trials, which were carried out between 20 October and 8 November 1937. Entry to the cockpit was relatively easy via the wing root and the forward view was adequate, assuming that the seat was raised to its fullest extent. The control column was conveniently positioned, all controls could be moved without any friction or play (although the rudder adjustment was difficult to use), and trimmers were provided for the elevator and rudder. The engine and propeller controls were easy to use, but the controls for the combined flap and dive brake was located too close to the fuselage side, which tended to trap the pilot’s hand.
Ground handling at normal and aft loadings was straightforward, but with CG at the forward limit the aircraft became noticeably nose heavy when taxying over rough ground. As a result the brakes had to be handled with care to avoid nosing over. Take-offs were normally made with 30 degrees of flap and the rudder bias was easily able to cope with the swing to the right that was experienced. The aircraft became airborne at around 70 mph IAS, and once the undercarriage was retracted, a rather ponderous acceleration could be made to the best climbing speed of 120 mph IAS.
Flight trials did not involve assessing the Skua throughout its full performance envelope, but at the speeds flown aileron control was found to be generally effective, except near the stall. Aileron control tended to become quite heavy as speed was increased, but the forces involved were not excessive. The rudder and elevator both gave adequate control, but a major problem was discovered during stability tests. Although the aircraft was just stable at speeds higher than 140 mph IAS, below this mark a marked longitudinal instability was experienced, with increased tail heaviness, which had to be counteracted by moving the control column forward to prevent the nose rising excessively. This tendency was very marked when close to the stall at extended aft CG. The work of the test pilots was made more difficult by the fact that the aircraft was also unstable laterally. This meant that normal stick-free stability tests could not be carried out, as the control column had to be held to keep the wings level by use of aileron.
The Skua stalled at about 75 mph IAS with the flaps and undercarriage up and at 69 mph IAS with the flaps and undercarriage down. If the speed was reduced in a glide in the clean configuration, the aircraft tended to self-stall if the pilot did not push the stick forward. There was very little stall warning, with no airframe buffet, but when the stall did occur, a wing would drop, followed by the nose. This tended to occur even if the control column was eased forward, but if the aircraft was mishandled by pulling the stick back, it became quite violent and a falling leaf developed. With the flaps and undercarriage down the aircraft’s characteristics were very similar, but it was slightly more mild-mannered. Once the nose had dropped and speed had increased, control could quickly be regained.
When approaching to land, the undercarriage could be lowered in about forty-five seconds without any obvious change in trim. Should the pilot forget to lower the wheels, a warning bell sounded, but unfortunately this proved to be inaudible and was of no use whatsoever. Owing to the instability at low speeds, nose-down trim had to be applied, which was the opposite of what was normally required. After landing on anything other than a smooth surface, a bucketing motion was liable to set in and at times the aircraft also developed a rolling gait from bumps causing excessive compression of an oleo leg.
After completing its handling assessment, K5178 remained at A&AEE for gunnery trials, before being delivered to Gosport for buoyancy tests. It was replaced at Martlesham Heath by the second prototype K5179, which had been flown for the first time on 4 May 1938 at Brough. It featured a nose lengthened by 2 ft 4¾ in which brought CG forward and, together with a tensioning device on the elevator control circuit, was an attempt to overcome the aircraft’s habit of self-stalling. K5179 was also fitted with leading edge slats, which were tested locked and unlocked, but the benefits at the stall were only marginal and these were not fitted to production aircraft. In an attempt to improve lateral handling, the wing tips were cranked upwards.
In view of the Skua’s longitudinal stability problems, much was expected of the spring tensioning fitted to the elevator control, as this was designed to apply a force to lower the elevators to resist the tendency for the nose to rise of its own accord. To test its qualities K5179 was flown at a take-off weight of 7827 lb at extended aft CG. The effectiveness of the spring depended very much on whether the engine was on or off. With the flaps and undercarriage up, the Skua was still unstable when climbing at low speed, and if the control column was pulled back and released, the nose came up rapidly followed by a sudden stall. A wing tended to drop which, if not corrected immediately, could lead to loss of control. In the glide with the engine off, the spring was more effective as it improved the feel of the elevator control and just about overcame the aircraft’s tendency to self-stall, although a little forward stick was still needed. As the engine was opened up, however, the spring gradually became less effective, until it made no difference at all. The aircraft behaved in a similar fashion with the flaps and undercarriage down.
The first production Skua II (L2867) took to the air on 28 August 1938 and was quickly followed by L2868. Both machines were delivered to A&AEE the following month, taking part in performance and armament trials respectively. Subsequent testing showed that the Skua was steady in the dive, without any vibration or instability. It tended to become tail heavy, which required a considerable forward push on the control column to maintain the correct angle of dive, but at reduced throttle settings this force was much reduced. Owing to its steadiness in the dive, pilots had no difficulty holding the aircraft on to a target. Lack of gun heating was commented on during the armament trials, but it was thought that this was would not be too much of a problem, as the Skua was unlikely to have to operate as a fighter at anything other than low to medium levels.
The last Skua to be tested at Martlesham Heath was L2888. Pilots were pleased to discover that the bell warning that the undercarriage had not been lowered had been replaced by a klaxon that could actually be heard. There were still some minor niggles though: the cockpit heating was unsatisfactory; the pilot’s seat adjustment was inadequate; and the interconnected throttle and mixture controls were badly positioned. On take-off difficulty was experienced in raising the tail and the aircraft had to be deliberately flown off, otherwise it was likely to remain stubbornly attached to the ground. The rate of climb was poor and, with the power on, the same longitudinal instability as before was experienced, together with some fore-and-aft pitching. Control response and effectiveness were good, except at slow speeds, when there was a marked deterioration, but the rudder and ailerons were very heavy above 230 mph IAS. Despite the spring fitted to the elevator control, the aircraft was still rated as being unstable longitudinally when gliding with the flaps and undercarriage up. At low speeds it was also unstable laterally. Stability tended to become neutral in the same condition with the flaps and undercarriage down. The aircraft was directionally stable at all times.
Such was the desperate need to get the Skua into service, it was accepted with the spring tensioning device to the elevator control and the cranked wing tips as the only major modifications. The first deliveries were to No. 800 Squadron aboard HMS Ark Royal in late 1938. By the time that war was declared two more squadrons (801 and 803) were also operational. The Skua’s most successful action against the Konigsberg was carried out by eleven aircraft from No. 803 Squadron and five from No. 800 Squadron. The attack was launched from Hatston in the Orkneys and achieved complete surprise, with only one aircraft failing to return. Although the Skua’s endurance was officially quoted as 4 hours 20 minutes, several aircraft managed to exceed this figure by up to ten minutes. Further dive-bombing attacks against the Scharnhorst and the French battleship Richelieu at Dakar were failures, as the 500-lb SAP bombs carried were ineffectual against more heavily armoured capital ships. Skuas from Ark Royal also provided fighter cover for some of the first Malta convoys and acted as guides for Hurricanes on their way to the island. However, it had never been intended that the Skua should have to fight land-based fighters and its poor performance by comparison led to it being replaced by Fulmars and Sea Hurricanes in early 1941. The remaining aircraft were taken on by No. 806 Squadron at Eastleigh and were used mainly for training and target towing.
The Blackburn Roc was an adaptation of the Skua to meet Specification O.30/35. The most obvious difference was the inclusion of a Boulton Paul power-driven gun turret behind the pilot’s cockpit, mounting four 0.303 in Browning machine-guns. To accommodate the turret, the fuselage had to be widened slightly and the wings featured 2 degrees dihedral outboard of the centre section in place of the Skua’s upturned wing tips. Provision was also made for a streamlined 70-gallon fuel tank to be carried under the forward fuselage and there were attachment points for a float undercarriage. With Blackburn fully occupied with the Skua and the forthcoming Botha twin-engined reconnaissance bomber, production of the Roc was transferred to Boulton Paul, with the tail units being supplied by General Aircraft at Hanworth. The first Roc (L3057) was flown on 23 December 1938 by Blackburn test pilot Flight Lieutenant H.J. Wilson. This aircraft, together with L3059, was delivered to Martlesham Heath in March 1939 for handling trials, with L3058 being used for testing the turret.
The aircraft was flown at varying weights depending on CG position: 6930 lb (forward CG), 7350 lb (aft CG) and 7815 lb (normal CG). On take-off, very little swing was reported, although the run appeared to be rather long. In the air, the controls were all relatively heavy at high speeds, but at least harmonisation was good. The ailerons were light at low speeds and although they were effective, they were rather slow in response. The rudder was also light at the low end of the speed range, but it was not particularly effective. The elevator was light at low speeds, but was slow in response and ineffective.
Following the tribulations with the Skua, stability was closely monitored on the Roc. It was found to be directionally stable at all speeds, but after this things became a little more complicated. Once again, stability varied greatly throughout the speed range and at different CG positions, varying from neutral at a normal service load in level flight to slightly unstable at aft CG. The behaviour was similar with the flaps and undercarriage up and down. Lateral stability was difficult to assess as pilots were pre-occupied with sorting the aircraft out longitudinally, but it appeared to be just stable with the engine on and neutrally stable in the glide. With the flaps and undercarriage up, the stall speed was 82 mph IAS and with the flaps and undercarriage down, the stall occurred at 76 mph IAS. There was very little warning except for a gradual worsening in elevator effectiveness. When trimmed at 1.2 × stall speed the tendency to self-stall was apparent once again and forward stick was needed to prevent the nose rising too steeply.
All aerobatic manoeuvres could be flown, except aileron turns when diving, as this was prevented by control heaviness with increase in speed. However, when manoeuvring in the horizontal plane care had to be taken in steep turns as high accelerations could lead to tightening. Dives were carried out up to 290 mph IAS, with the flaps up and down, without difficulty and recovery was straightforward. In the approach there was insufficient trim available at speeds lower than 115 mph IAS with forward CG and since the glide was steep and the elevator not particularly effective, a wheel landing was recommended. After touchdown, it was advisable to delay using the brakes for as long as possible to reduce the risk of nosing over. The overall assessment of the Roc was slightly more complimentary than the Skua, as its handling qualities were somewhat better and its stability improved because of its lower all-up weight.
Testing of the Boulton Paul turret showed that movement of the turret in the air produced a slight yaw to the left, which was not dependent on the direction the guns were pointing at the time, and with rudder control heavy at high speed, this could be difficult to overcome. The turret performed reasonably well up to the aircraft’s service ceiling, although there was a noticeable drop in the speed of rotation in the colder temperatures at altitude. Problems were also experienced when used in the ‘high speed’ mode, as electrical fuses tended to blow with monotonous regularity. Continued use of the turret tended to drain the air pressure that was required to operate the fairings, so for a time movement had to cease while pressure built up again. The Roc was also stressed to carry light series bomb carriers under the wings.
In an attempt to improve performance, L3058 was flown with an experimental propeller, which consisted of a normal Skua hub with blades of the type fitted to the Tiger engines of an Armstrong Siddeley Whitley. Although the rate of climb was slightly better, there were no improvements in top speed and the proposal did not go any further. The performance figures for the Roc were very similar to the Skua. The maximum speed was 223 mph at 10,000 ft and the initial climb rate was 1500 ft/min. The service ceiling was 18,000 ft.
It had always been the intention to use the Roc as a seaplane fighter and L3057 and L3059 were sent to the Blackburn factory at Dumbarton in October 1939 to be converted. The wheel wells were covered over and Blackburn Shark floats were mounted on N struts under the centre section with separate front struts. Water rudders at the rear of the floats were connected to the aircraft’s braking system and operated pneumatically. By the time the conversion had been completed, the Marine Aircraft Experimental Establishment (MAEE) had conveniently moved from Felixstowe to Helensburgh on the Clyde and tests were commenced in November 1939. The trials were marred by L3059’s crash on 3 December as a result of marked directional instability. L3057 was modified to include an enlarged ventral fin in an attempt to improve the aircraft’s controllability in this respect. Without the fin, low-level turns were particularly dangerous, but with it in place there was a marked improvement, although the turn and slip indicator still had to be monitored closely as any sideslip could lead to disaster. L3060 was also fitted with floats, but with the withdrawal of British forces from Norway in 1940, the requirement for a floatplane fighter, for the time being at least, came to an end.
The Roc never did operate from aircraft carriers, as had been the original intention, and was only ever used from bases on the mainland. It entered service with No. 806 Squadron at Eastleigh in February 1940 and began flying with No. 801 Squadron at Hatston four months later. Experiences with the Defiant were soon to prove that the concept of a turret-armed fighter was not viable and the Roc was quickly downgraded to second-line duties. One of the largest users of the Roc was No. 2 Anti-Aircraft Co-operation Unit at Gosport, which received sixteen in June 1940, four of which had to suffer the indignity of being used as ground-based machine-gun posts as defence against air attack. The Roc also served with numerous FAA training squadrons and a number were converted as target towing aircraft, the last being withdrawn from use in mid 1943.