Westland-F7-30-front-left

Westland F.7/30 (PV.4) Biplane Fighter

By William Pearce

On 1 October 1931, the British Air Ministry issued Specification F.7/30 for a single-seat day and night fighter. Further requirements of Specification F.7/30 were for the aircraft to be capable of at least 195 mph (314 km/h) at 15,000 ft (4,572 m), have a landing speed of 50 mph (80 km/h), carry an armament of four machine guns, and offer the pilot an unobstructed, all-around field of view. Aircraft developed from Specification F.7/30 were expected to outperform then-current contemporary fighters in respect to handling, maneuverability, range, rate of climb, and service ceiling.

Westland-F7-30-in-flight

The Westland F.7/30 (PV.4) on an early test flight. The aircraft is seen with its original open canopy and exhaust manifolds. Note the good visibility from the pilot’s position.

Numerous manufacturers responded to Specification F.7/30, including Blackburn with the F.3 biplane, Bristol with the Type 123 biplane and Type 133 monoplane, Gloster with the SS.37 biplane, Hawker with the P.V.3 biplane, Supermarine with the Type 224 monoplane, and Westland with the F.7/30 biplane. In addition to the F.7/30 biplane, Westland also submitted a high-wing monoplane design. However, the monoplane’s long wingspan and comparatively high landing speed resulted in the biplane being selected by the Air Ministry.

Although an engine type was not listed in Specification F.7/30, the Air Ministry informally expressed a strong preference for the Rolls-Royce Goshawk. The Goshawk was a development of the Rolls-Royce Kestrel IV V-12, which was the most advanced British liquid-cooled production engine at the time. While both the Kestrel and the Goshawk had a 5.0 in (127 mm) bore, a 5.5 in (140 mm) stroke, and displaced 1,296 cu in (21 L), the Goshawk employed evaporative (steam) cooling.

Westland-F7-30-front-left

The Westland F.7/30 with an enclosed canopy but still with the original exhaust manifolds. The leading-edge slats are visible on the upper wing.

The basic premise behind evaporative cooling for aircraft engines was that cooling water would enter the engine and be allowed to boil as it flowed through the hottest part of the engine, the cylinder heads. The conversion of water to steam allowed for more heat to be taken from the engine compared to traditional water cooling. The steam would then be fed though a condenser (heat exchanger), in which the steam would cool and convert back to water. The condenser could be incorporated into the aircraft’s structure, such as the wing’s leading edge and skin, and the condenser would be cooled by the slipstream of air flowing over its surface. In theory, the end result of evaporative cooling compared to conventional water cooling was a more efficient system that eliminated the drag of a conventional radiator. In practice (with various installations), evaporative cooling systems were complex, leaked steam, had inefficient or ineffective condensers, and could suffer from vapor lock.

The Westland F.7/30 originally carried the company designation PV.4, and it was designed by Arthur Davenport, Westland’s chief designer, with input from Harald Penrose, Westland’s chief test pilot. Most of the aircraft designed for Specification F.7/30 were fairly conventional. However, the Westland F.7/30 incorporated some unconventional features. A biplane arrangement was selected to satisfy the 50-mph (80-km/h) landing speed and maneuverability requirements of Specification F.7/30. The cockpit was placed in the extreme nose of the aircraft and forward of the upper wing to provide an unobstructed view for the pilot. This necessitated moving the engine from the nose of the aircraft to behind the cockpit, with an extension shaft driving the propeller.

The aircraft’s center and forward fuselage had a tubular aluminum frame and were covered by removable aluminum panels. The rear fuselage and tail had an aluminum frame and were cover by fabric. The upper and lower wings attached to the center fuselage, which was strengthened to support the engine. The upper wing was of the gull-type to improve the pilot visibility and was equipped with ailerons and automatic leading-edge slats. A 49 US gal (41 Imp gal / 186 L) fuel tank was located in the inboard section of each upper wing. The lower wing was staggered behind the upper wing, had a reduced span, and was straight with no control surfaces. Both upper and lower wings had aluminum frames with aluminum skin covering the leading edge and fabric covering the rest of the wing. All control surfaces had fabric-covered aluminum frames.

Westland-F7-30-side

Side view of the Westland F.7/30 with the updated exhaust manifolds. Note the barrel of the upper machine gun extending almost to the propeller, while the lower machine gun was almost completely recessed. The odd structure forward of the canopy is the gun sight.

The Goshawk II engine produced 600 hp (447 kW) at 2,600 rpm and was installed in the center of the aircraft. An extension shaft ran from the engine, under the cockpit, and to a propeller reduction gear in the nose of the aircraft where it turned a 10 ft 4 in (3.15 m) diameter, two-blade, wooden propeller. The condenser (radiator) was mounted under the lower wing’s center section. The engine’s exhaust manifolds protruded from each side of the fuselage between the upper and lower wings. Two machine guns were mounted on each side of the aircraft. The lower machine guns were mounted below the exhaust manifolds, and the upper machine guns were mounted by the cockpit in the forward fuselage. All four Vickers .303 machine guns, each with 140 rounds, were synchronized to fire through the propeller. The cockpit was originally open but was later fitted with an enclosed canopy. The main landing gear was mounted just forward of the lower wing. The main gear and tailwheel were covered with aerodynamic fairings.

The Westland F.7/30 had a wingspan of 38 ft 6 in (11.7 m), a length of 29 ft 6 in (9.0 m), and a height of 10 ft 9 in (3.3 m). The aircraft had an estimated top speed of 185 mph (298 km/h) at 15,000 ft (4,572 m) and a landing speed of 55 mph (89 km/h). Time to climb to 20,000 ft (6,096 m) was 18 minutes. The Westland F.7/30 had an empty weight of 3,624 lb (1,644 kg) and a loaded weight of 5,170 lb (2,345 kg).

In July 1932, Westland received an order from the Air Ministry to build a F.7/30 prototype, and the aircraft was given serial number K2891. Construction of the prototype was undertaken at the Westland factory in Yeovil in southern England. Behind schedule, the aircraft was completed in early March 1934. Ground handling tests were undertaken in mid-March. After some adjustments, the aircraft was transported by ground to Royal Air Force Station Andover for its initial flight. The first flight was made on 23 March 1934 with Harald Penrose at the controls. The aircraft handled well, and Penrose soon flew it back to Yeovil.

Westland-F7-30-front

Front view of the Westland F.7/30 illustrates the condenser (radiator) under the lower wing. The landing gear supports did nothing to improve airflow into the condenser, and the additional struts for the gull wing did nothing to improve the aircraft’s aerodynamics.

Continued flight testing revealed that the pilot experienced an uncomfortable amount of buffeting in the open cockpit at cruising speed or higher. Also, the cooling system was barely adequate, with engine temperature always running near its upper limit. Flight testing continued, and the Westland F.7/30 was found to have a top speed of only around 145 mph (233 km/h).

The cockpit was enclosed to improve pilot comfort and also in an effort to increase the aircraft’s speed. More flight testing occurred, and it was discovered that as the aircraft rolled to 90 degrees, the engine’s hot exhaust would burn the fabric covering the rear fuselage. The exhaust manifolds were subsequently modified to resolve this issue. The new manifolds had a single exit that was positioned toward the front of the engine, allowing the hot exhaust to flow back against the non-flammable metal manifold so that it was sufficiently cooled by the time it reached the rear fuselage’s fabric covering.

The Westland F.7/30 was evaluated by the Aeroplane and Armament Experimental Establishment at Martlesham Heath Airfield in mid-1934. Maximum speeds of only 146 mph (235 km/h) at 10,000 ft (3,048 m) and 122 mph (196 km/h) at 20,000 ft (6,096 m) were recorded. Development of the Westland F.7/30 was abandoned in 1935, as its performance was far below expectations and that of the other F.7/30 competitors. For all of the aircraft powered by Goshawk engines, the complex evaporative cooling system was prone to overheating and found to be unsuitable for service aircraft. The Westland F.7/30 was ultimately scrapped. The winner of Specification F.7/30 was the Gloster SS.37, which was powered by a conventional radial engine. Put into production as the Gloster Gladiator, it was Britain’s last biplane fighter and an aircraft that served nobly in World War II.

Westland-F7-30-rear

Although the Westland F.7/30 flew well, its performance was far below contemporary fighter aircraft. The winner of Specification F.7/30, the Gloster Gladiator, was 100 mph (161 km/h) faster.

Sources:
Westland Aircraft since 1915 by Derek N. James (1991)
Interceptor Fighters by Michael J. F. Bower (1984)
British Piston Aero-Engines and Their Aircraft by Alec Lumsden (2003)
British Flight Testing by Tim Mason (1993)
https://www.alternatehistory.com/forum/threads/specification-f-7-30.368722/
https://www.secretprojects.co.uk/threads/british-f-7-30-fighter-specification.29612/

2 thoughts on “Westland F.7/30 (PV.4) Biplane Fighter

  1. jiro y

    Wow! A major failure of performance expectations. Great write up of an interesting failure. Considering that the Westwind was abandoned in 1935 and the P-39 made it’s first flight in April 1938,I wonder if you found any references to contact between Larry Bell, Bell Aircraft or H.M. Poyer, designer for project leader Robert Woods of the P-39? I am guessing that the P-39/P63 were the most successful of the mid engined piston pursuit planes. Thank you for sharing your scholarship.

    Reply
    1. William Pearce Post author

      Hello Jiro – Glad you liked the article. I have not seen anything to indicate Bell’s P-39 was influenced by the F.7/30. The Bell team always seemed to think a little outside the box.

      Reply

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.