The superstructure was even more clearly a departure. The redesign meant that the sweep back of the mainplane outer sections, in response to the changing service loads of later Southamptons, was no longer necessary and the earlier two-bay structure and warren girders were now reduced to a much simpler arrangement, whereby two of the four struts also supported the engines. These engines were now positioned directly under the top wing and the previous triple fins were now succeeded by two units with extended rudders.
Beverley Shenstone, Mitchell’s aerodynamicist, considered the resulting aircraft ‘perhaps the cleanest biplane flying boat ever built, with minimum struttage and clean nacelles faired into the wing’. He did not mention the very ‘boxy’ radiators which projected on either side to the rear of these nacelles, although this positioning least compromised the overall lines of the new design (see photograph above).
The Scapa prototype. (Courtesy of E.B. Morgan)
The new prototype made its first flight on 8 July 1932, and it was delivered at the end of October to the Marine Aircraft Experimental Establishment, Felixstowe, for service testing. In the following May, the prototype was flown to Malta, for overseas acceptance trials with No. 202 Squadron and these included a long-distance flight to Gibraltar and a cruise to Port Sudan via Sollum, Aboukir, and Lake Timsah. On its return, the Scapa took part in the 1934 fly-past of ‘the competition’ at the Hendon RAF display with, as Penrose reported, ‘the clean Supermarine twin-engined Scapa leading, followed by the four-engined Short Singapore, triple engined Blackburn Perth, the distinctive gull-winged Short Knuckleduster, Saro R24/31 London and the three Saro Cloud trainers’.
The Hendon event was clearly designed to impress foreign governments with Britain’s military capabilities and, during the 1936–1939 period, the Scapa fulfilled its required purpose with anti-submarine patrols to protect neutral shipping during the Spanish Civil War. Some of the aircraft of No. 202 squadron were later transferred to No. 204 Squadron and were sent to Egypt during the Italian-Abyssinian confrontation. There was also a single Scapa, attached to No. 228 Squadron, whose contribution to developing war preparations was its involvement in early radar trials.
The Stranraer
Soon after Supermarine had received orders for the Scapa, the Air Ministry issued another specification which was unlikely to be matched by a simple development of this last type. This latest specification was for another general purpose coastal patrol flying boat, of robust and simple construction with low maintenance costs, but capable of carrying a 1,000lb greater load and of maintaining height on one engine with 60 per cent of fuel on board.
An enlarged and substantially altered version of the Scapa had to be projected and a specification was submitted, alongside one from Shorts and one from Saunders-Roe. Only the last, the A27 London, was accepted and it was later ordered to replace the Southamptons and Scapas of Nos 201 and 202 Squadron respectively. The Short Singapore III had also been ordered as a replacement for other Scapas with Nos 204 and 240 Squadrons.
This Short machine had about the same speed as the Scapa and was powered by twice as many engines, and another Short machine, the Saraband, was only a few mph faster with six engines. Thus, given an economic situation in which orders for these larger flying boats were likely to be kept to a minimum, it seemed a distinct possibility that a performance from Mitchell’s smaller, twin-engined proposal, if significantly better than that of the Saunders-Roe London, might still stand a chance of winning some contracts, given the growing calls for British rearmament.
Another reason for anticipating orders for the proposed new design was not simply based on the good performance figures that the Scapa had returned but on Mitchell’s having come to believe in the virtue of employing a thin wing – for other than Schneider Trophy racers – contrary to the generally perceived wisdom of the day. The eventual outcome was an Air Ministry contract for a flying boat that outclassed all of its contemporaries of similar configuration.
The engines chosen initially to pull the new machine’s thinner aerofoil through the air, and to give it the one-engine performance required by the Air Ministry specification, were 820hp Bristol Pegasus IIIMs, providing an additional 590hp more than the Scapa’s Kestrels. These two engines were to be mounted with the same thrust line and in streamlined fairings but, being air-cooled radials, they did not incur the weight and drag penalties of the Scapa’s radiators. Long-chord Townend drag-reducing rings now surrounded the cylinder heads and their oil coolers formed part of the top centre-section leading edge. Against these improvements, there was the additional 12 per cent increase in wing area of the new machine. This extra drag and weight was added to by the two-bay strut arrangement required to support the extra 10ft of wingspan that was needed to meet the new specification’s load carrying requirements.
The larger size, nevertheless, did have some aerodynamic advantages. The extra depth of the hull allowed the top of the enclosed cockpit to form a continuous line with the midships gunner’s cockpit, which was now placed in the centre of the hull top (in the Southampton and Scapa there had been two mid-ship gunner’s cockpits, offset from the centre line). Now, for the first time, Supermarine had built a larger service aircraft which made it possible to install the second rear gunner more sensibly, in a faired-in cockpit in the tail. This had been proposed for the unsuccessful Vickers/Southampton X prototype, with its wingspan of 79ft, and so presented little difficulty for the new 85ft craft. Its ‘general purpose’ character was evidenced by the fitting of carriers below the inner sections of the lower wings for up to four 250lb bombs or extra fuel tanks. The flatter fuselage section between the lower wings was even more convenient than that of the Scapa for transporting supplies, such as a spare engine.
The Stranraer prototype. (Courtesy of RAF Museum)
By the time that the designing and the constructing of the new prototype was well under way, the Australian Seagull V order had not yet been completed and the Scapa flying boat contract was also being fulfilled. Nevertheless, the new prototype, K3973, was test flown by Summers on 27 July 1934 and delivered in very short time to the Marine Aircraft Experimental Establishment at Felixstowe, for service assessment.
The performance of the aircraft was such that an order for seventeen aircraft, K7287–K7303, was placed with Supermarine by the following year. The standard service machine was fitted with the more powerful 920hp Pegasus X engines, giving it a maximum speed of 165mph and making it the fastest biplane flying boat to enter RAF service – yet with a stalling speed of only 51mph.
Its maximum ceiling was 20,000ft and it could climb to the first 10,000ft in ten minutes – the Scapa had taken twice as long, two years earlier. As it had been necessary to withhold these performance details because of the developing international situation, company publicity had to be content with the by-no-means despairing comment that the aircraft ‘passed all its tests brilliantly’ and went on to claim that:
The outstanding feature of this flying boat is that the performance obtained during a series of extended service trials, whether in respect of speed, climb, ceiling or take-off, is unequalled by any other British flying boat. All the specification requirements were exceeded by large margins.
As the new aircraft had become quite distinct from the earlier Scapa from which it was developed, a new name was chosen and the machine entered service as the ‘Stranraer’. It must have been gratifying for Supermarine to see the Saunders-Roe London flying boat then replaced by their new aircraft with Nos 201 and 240 Squadrons, and to see another rival company’s aircraft, the Singapore III, superseded by the Stranraer with No. 209 Squadron. Other machines replaced the company’s own Scapas with No. 228 Squadron and so the total of Stranraers ordered from Supermarine, including the prototype, came to eighteen.
It thus came about that the Stranraer was actually operated in the Second World War, serving with No. 228 Squadron, when it was needed to patrol the North Sea. Additionally, some of the Stranraers of this unit were transferred
to No. 209 Squadron and, fitted with extra fuel tankage under one wing and bombs under the other, they conducted patrols against enemy shipping between Scotland and Norway, until replaced in April 1941 by the ubiquitous Short Sunderland. No. 240 Squadron was also equipped with the Stranraer and made the last operational patrol of the type on 17 March 1941, after which it was replaced by the American-built Catalina.
In addition to the British Stranraers, the Royal Canadian Air Force also adopted them; forty were built by Canadian Vickers and these saw a great deal more service than their British counterparts – in the battle against the U-boats in the Atlantic. The last RCAF Stranraer was retired as late as 20 January 1946, and fourteen of the aircraft were sold to the civil sector, particularly to private airline companies in Canada where the lakes of the Northern Territory provided ready-made runways. The last of the Stranraers served in these regions until 1958.
However, the longevity and performance of the Stranraer does not conceal the fact that, in this larger flying boat category, the influence of Mitchell was relatively short-lived – in many ways, a result of the lack of official encouragement for his Air Yacht (monoplane) approach to the reconnaissance flying boat type and of the cancellation of the Type 179 Giant cantilever monoplane.
Meanwhile, the early and single-minded approach of Shorts to all-metal aircraft had paid dividends and finally led to the military development of another cantilever monoplane type, the Sunderland, which dominated the wartime long-range sea patrol provision, with a total of 741 being built. In addition, Short’s clean, streamlined civilian version had monopolised the flying boat provision on the Imperial Airways routes just prior to the outbreak of the war and represented a major step forward in flying boat design, without a rival aircraft from Supermarine.
‘HE LOOPED THE BLOODY THING!’
After the Spitfire, the Walrus is probably the best-known RAF aircraft to come from Mitchell’s drawing board. Yet it owed its origin to an Australian order for a replacement for the Supermarine Seagull IIIs, which the Royal Australian Air Force had been operating since 1926 (including an annual race between these staid machines at a regatta in Hervey Bay, north of Brisbane).
The Seagull V/Walrus
In 1930, an Australian specification had been sent around British aircraft firms for a replacement reconnaissance amphibian which could be catapulted with full military load, and which was also capable of shipboard stowage. In view of the deck-landing limitations and porpoising characteristics of their aging Seagull III, the Seagull V, as the Australian aircraft was to be named, had to be a complete redesign for reasons other than obsolescence (the earliest version of the type, the Seal, first flew in 1921).
In fact, it might be fairly accurate to say that the only influence of the older type on the Seagull V was the basic layout of the last experimental Seagull II, N9644, a biplane amphibian which had reverted to the use of a pusher air-cooled power unit. Otherwise, the move to metal structures, slab-sided fuselages, and the experience of the intervening years, produced a quite distinct type within the older formula.
One significant aspect of the redesign was the employment of a fully retracting undercarriage, which had featured little in Air Ministry requirements – although, as we have seen, Mitchell had had to invent mechanisms since 1920 to lift his amphibians’ wheels out of the water. Now, Alan Clifton persuaded Mitchell to retract the wheels into the wing, saying ‘We shall have to do it eventually, why not now?’
Another aspect of the redesign was the hull. It now displayed the slab-sided features of the Scapa and the later Stranraer, but without their upward sweep to the tail unit and with a one-step planning surface. Various reasons contributed to the especially utilitarian appearance to the hull, but the foremost need inherent in Mitchell’s redesign was to improve on the ‘sea state’ performance of the Seagull III.
Supermarine’s move to metal structures would contribute to the sturdiness of the proposed Seagull V, signified by the direct attachment of the lower wings to the hull, as the new material would be less prone to failure at the join. Again, this feature was similar to that in the recent Scapa, and the Stranraer that followed, although in this case there was no elegant lower wing centre-section and the upper centre-section was functionally cut back for clearance of the propeller, and cut back again to allow for wing folding. But then, speed was not a requisite in a ship’s aircraft where spotting would be one of its main functions. On the other hand, the lower wing did not have the large cut-outs of its predecessors, as the inner portion of the wing behind the rear spar could be folded away in order to clear the hull sides for stowage; and the wings had single bay struts similar to the recent Seamew and Scapa designs.
Compared with these other two seaplanes, the new design overall had a much more ‘minimalist’ appearance which was not decreased by the engine having the traditional Supermarine position between the wings, supported by eight struts, and by its nacelle being noticeably offset to counteract the corkscrew pressure of the propeller slipstream on one side of the fin.
The new machine was clearly no beauty and its functional appearance, allied with its very traditional configuration, appeared to have won it no friends when, in June 1933, the prototype was completed and seen by the Air Ministry Director of Technical Development. Those at Supermarine with long memories of the inability of the company to win orders for their naval amphibian fighter type could hardly have been encouraged by the director’s comment to Clifton: ‘Very interesting; but of course we have no requirement for anything like this.’
The Seagull V prototype on Southampton Water. (Courtesy of Solent Sky Museum)
Perhaps this reaction had some bearing on the test pilot’s performance at the 1933 SBAC Show at Hendon. The Aeroplane nicely described the event:
This boat made its maiden flight on 21 June, five days before its first public appearance, but Mr Summers proved its qualities by throwing it about in a most carefree manner. Of its performance little is known but there can be little doubt about its amiability and general handiness in the air and on the ground. One must be prepared to see all sorts of aeroplanes looping and rolling with abandon nowadays, but somehow one has, up to now, looked to the flying boat to preserve that Victorian dignity which one associates with crinolines, side whiskers, bell-bottom trousers and metal hulls. The Seagull V destroyed all one’s illusions.
Henry Knowler, chief designer at Saunders-Roe, who watched the display in the company of Mitchell, reported the designer’s understandable surprise and anxiety at the low-level antics of the five-day old prototype. ‘He looped the bloody thing!’ Mitchell kept repeating to everyone he met.
He had obviously not heard that a disgruntled American pilot had once done the same over Killingholme in a 95ft Felixstowe flying boat. Nor did he live to see the production of the Walrus (its later British name) as flown by Supermarine test pilot George Pickering:
At the finish of the test he would fly very low over the river, by low I mean about 300ft or so, in front of the flight shed and loop the Walrus to signal that it had passed the test; if not, he would just land and return up the slip and it would be worked upon. If you haven’t seen a Walrus looped, you haven’t lived. [As recounted by Peter Weston, who began his varied flying career as a Supermarine apprentice in 1938.]
Thereafter, the Seagull V underwent modifications and trials. Test pilot ‘Mutt’ Summers criticised the rigidity of the undercarriage and the lack of steering capacity on the ground, and these deficiencies had had to be put right. By this time it had been discovered that another make of aircraft had been allocated its N1 number, so the one-and-only Supermarine prototype now became N2 and went to the MAEE on 29 July, just over six weeks after its maiden flight. Evaluation tests then lasted until the end of October, after which the Seagull went to the Royal Aircraft Establishment at Farnborough for the catapulting trials required by the Australian Government.
Now the empty weight of the aircraft had gone up to 5,016lb and Mitchell, having returned to work
at the end of 1933 after his operation for cancer, ordered an aerodynamic clean-up as parasitic drag had appreciably reduced the performance of the heavier machine.
Despite the lengthy testing, including successful catapult trials, no British order was expected. Webb, now in the business manager’s department, quoted an Air Ministry letter, saying that ‘we do not envisage any role for an aircraft of this type with HM Forces’. He also relates how serving officers at the nearby flying boat base at Calshot could also see no use for it. One of them asked, ‘What are you people doing wasting our time on a machine like that – it will be shot out of the skies by the fighters?’ So Webb pointed out that there would be no fighters with enough range to shoot anything down in mid-ocean and that, catapulted from a cruiser or battleship, it would be the eyes of the fleet.
The aircraft was then taken back to home waters for the continuation of trials at Sheerness and in the Solent until May, when it was returned to Supermarine for the fitting of redesigned wing-tip floats for improved buoyancy, the removal of the wheel brakes for lightness and for an improved layout of the observer’s compartment.
Further fleet operation trials continued, including ‘sea state’ landings in 30 knot winds and 6ft waves off the Kyles of Bute, and underway recovery onto a warship making up to 13 knots through rough water. In these trials the Royal Navy had been acting as ‘programme manager’ for the Australians and, as a result, their government ordered twenty-four production Seagull Vs, A2-1 through to A2-24.
Thus it would appear that the future of Mitchell’s design would rest solely with the Australian Government’s requirement, but movements were afoot nearer to home, as reported in Caspar John’s foreword to The Supermarine Walrus:
From Nighthawk to Spitfire Page 15
