The Flying Radiators: Pt.1 – R. J. Mitchell's S.6

In the 1927 Schneider Trophy contest, Britain and Mitchell had been more fortunate than the Italians in that the winning Napier Lion engine had not been a new and unreliable design. It had been continuously developed since its use in the Supermarine 1922 Schneider Trophy winner and, although it had never failed in any of the racing aircraft it powered, it was nevertheless reasonable to ask whether this fine engine had now reached the end of its development potential, especially as a supercharger had not yet been produced.

     By now, Rolls-Royce had produced the successful 490 hp Kestrel, in response to the American Curtiss D -12 engine which powered the Schneider Trophy winning aircraft of 1923 and 1925 and were now offering supercharging and reduction gearing. Mitchell and Major G.P. Bulman, the Air Ministry official responsible for the development of aero engines, decided to take the gamble of ordering a racing engine from Rolls-Royce – identified as the 'R' engine.

 S.6.

Mitchell decided to rely on the basic suitability of his previous design (see my Blog: The High Speed Designer Confirmed – R.J. Mitchell’s S.5) for the new engine and so his main design effort was in respect of an all-metal version, but larger than the S.5, in order to accommodate the projected heavier engine: the 930 lbs. of the Napier Lion in 1927 was to be superceeded by an engine weighing 1530 lbs. 

     An immediately obvious alteration to be made was to the cowling required by the change from the ‘arrow’ shape of the Lion engine to the ‘V’ of the new Rolls-Royce unit and, as the empty weight of the S.6 was 1791 lb. heavier than that of the S.5, the wingspan was increased by more than three feet and the front float struts had to be moved further forward on the fuselage to support the heavier and longer engine. Also, the increase in fuel consumption from the Lion engine with a 22 litre capacity to the proposed Rolls-Royce R engine of 36.7 litres would require both floats to be used for the fuel tanks.  The extra weight of the new engine also required moving the pilot's position further back.

 The S.6 airframe receiving the new Rolls-Royce R engine.

     Solving the above mentioned constructional and loading problems in itself had justified the new design's ‘S.6’ designation, but these matters were relatively straightforward compared with contending with the heat produced by the new engine which represented a power increase of 211% over the previous Lion engine (for what turned out to be a loaded aircraft weight increase of only 78%). Mitchell was reported to have said: ‘Go steady with your horsepower’ – no doubt anticipating the cooling problems that would be encountered. 

    The areas of oil-cooling channels which had run along the sides of the S.5 were now increased and new ones were added to the underside of the fuselage. Particular attention was given to the efficiency of these radiators. Mitchell’s Chief Metallurgist, Arthur Black, came up with a method whereby the oil, which ran from and to the engine, passed around copper tongues soldered in the channels at right angles to the oil flow – in such a way that they did not impede this flow, whilst ensuring maximum contact of the oil with the surfaces being cooled by the slipstream of the aircraft. Additionally, the hot oil was sprayed from piping at the top of the fin to trickle down guttering to the return channels, thereby causing the fin to act as both oil tank and radiator. Nevertheless, the High Speed Flight pilot, Greig reported that the position of the oil pipes, attached to a very narrow fuselage, ‘turned the inside of the cockpit into something approaching an extremely hot Turkish bath’ with the oil temperature gauge reading ‘around 136 degrees centigrade.’]

    Meanwhile, Supermarine were anxious to point out that their move to metal construction was not just with respect to the framework of their machine but that it placed them in the forefront of the use of load-bearing external skinning – particularly in respect of the wings where, instead of being plywood-covered with the radiator panels externally attached, as with the S.5, they were now covered by the radiator panels alone. Additionally, made now of aluminium, they took torsional loads and, as Supermarine announced, ‘saved a considerable amount of weight over previous practice’.

Despite there now being a gap of two years between competitions, the scheduled start of the eleventh event was less than six weeks away before Mitchell’s new airframes could be tested in the air. In May, the new R engine had reached 1545 hp but failures began to occur and it was only at the end of July that the new engine passed the one hour mark at full throttle and supercharger boost. A few days later, with the blending of a special fuel, an engine run of 100 minutes and 1850 hp was achieved – much to the relief of the citizens of Derby, as the tests had also required the simultaneous running of three Kestrel engines. They drove fans to cool the new R engine, to dispel fumes from the test shed, and to enable the carburettors to be set up in simulated flight conditions. People living up to fifteen miles away reported still being able to hear the engine runs and and the ears of the Rolls-Royce workers were plugged with cotton wool. They were also well supplied with milk to counteract the laxative effect of breathing in the engine oil fumes, particularly in the early stages of testing: the engine would consume 60 gals of castor oil in 25 minutes, the majority of which was ejected out of the exhaust ports and deposited on the walls of the test cell.

     Meanwhile, the High Speed Flight now consisted of Greig's fellow members of the aerobatic team for the last RAF Display at Hendon : Fl. Lts. G. H. Stainforth and R. D. H. Waghorn and Fl. Off. R. L. R. Atcherley. The command was then handed over to Sq. Ldr A. H. Orlebar with the new title: ‘Officer Commanding the High Speed Flight’. Waghorn has described how they used  practice machines to devise the best method for cornering, with the help of scientists from the Royal Aircraft Establishment, who installed instruments to measure speed, acceleration, and climb. A compromise between high G very tight turns, with loss of speed, and wider arcs, which incurred less drag, was thus worked out. But, with the increased speed in the turns, pilots had now to get used to blacking out.

     These preparations received a considerable setback when the actual contest aircraft finally arrived and it was discovered that the possible effects of the much greater torque of the new engine had not been fully realized.  Against the turn of the fixed pitch propeller, the S.6 would dig in the left float, describing circles in the water which Orlebar reported ‘had rather shaken’ Mitchell. One can easily imagine the Chief Designer’s feelings, seeing his aircraft quite unwilling to fly and when Orlebar pointed out to him that 247, the number of the first of the new machines, added up to 13, ‘the poor chap replied with feeling that he had not designed that’.

     Before the invention of variable pitch propellers, it was discovered that a breeze, kept on the left quarter, allowed the aircraft to get onto the step and lift off. To assist the pilots, Mitchell lengthened the starboard float by a foot so that it could contain 90 gallons of fuel and allow the capacity of the submerging float to be reduced to 25 gallons. But having achieved take-off, engine over-heating became a problem and so radiator piping had to be fitted along the sides of the floats. Small wing-tip scoops were also fitted under the wing tips, facing forwards and with exhaust ports at the wing roots, thus creating a flow of air over the inner surfaces of the radiators – an unexpected bonus for using the radiators as load-bearing wing surfaces.

Waghorn's winning S.6

The Italians also received their new aircraft late and it was soon found that the hydrofoil equipped Piaggio Pc.7 was unable to take off at all. Also, one of the two Fiat C.29s caught fire and later stalled on take-off and sank. On the other hand, the first of the Macchi M.67s was looking much more promising, reaching a speed of over 360 mph, but then it too crashed, killing its pilot. Accordingly, Italy requested a one month postponement on August 22 but the Royal Aero Club stuck to the rules and refused the next day; the FAI concurred and the eleventh Schneider Trophy contest was still set to begin  at Calshot on 7 September.

   

To prevent Britain winning by a fly-over, Italy had decided not to withdraw despite their problems and sent over their remaining, as yet, untested aircraft – the second Fiat C29, a Savoia-Marchetti S.65, and the two Macchi M.67s – as well as two older Macchi M.52s. By this time, both the Supermarine aircraft were ready but, as the Glosters’ fuel problems could not be solved in time, it was decided to call up one of the 1927 S.5s to complete the country's permitted three contest aircraft.

    British tactics were to take account of a compromise that had had to be worked because cooling was so critical that a temperature of 95 degrees had not to be exceeded by the new Rolls-Royce engines It was decided that Waghorn would therefore fly as fast as possible, consistent with keeping to a safe engine temperature; Atcherley, in the second S.6 would risk a higher temperature if the performances of the two preceding Italians made it necessary to go faster than Waghorn; it was expected that Greig, in the slower S.5, would provide additional back-up, as would Molin in the M.52.

     In the event, these tactics were unnecessary: Cadringher retired on lap two as he had been nearly blinded and suffocated by fumes from his exhaust and Monti, in the second M.67, also had fume problems and was forced to made a hasty landing when steam and nearly boiling water blew back from the engine.  
    Atcherley also had a visibility problem – caused by spray from a long take-off run; he tried to replace his goggles with a substitute pair but these were swept away in his slipstream. Thus he saw the pylon at the first turn rather late and unfortunately turned inside it; he was therefore disqualified.   
     And so Waghorn, in the new S.6, was first with an average speed of 328.63 mph; Molin was second in the 1927 M.52, at an average speed of 284.2 mph; and Greig, also in an aircraft from the previous contest, averaged 282.11 mph. Atcherley, in the second new S.6  completed the course, although disqualified, and thus Mitchell was denied the satisfaction of having both of his new aircraft design coming first and second as well as being the only 1929 machines to complete the course. However, he and Atcherley were compensated for the disqualification of N248 by its setting up World Closed Circuit Speed Records for 50 km and 100 km at 332.49 and 331.75 mph respectively, on the sixth and seventh laps. 
    Three days later there followed a competition between the Gloster VI, whose fuel supply problems were now being overcome, and the S.6 to establish a new World Absolute Air Speed Record. The Gloster machine achieved 336.3 mph but Orlebar took the record with 355.8 mph. The previous Italian record of 318.62 mph was further improved on two days later, on September 12, when the S.6 achieved 357.7 mph. 

* * * * *

For reference sources, see my Blog: “Source Material and References." An extended bibliography is included in my R.J.Mitchell at Supermarine; Schneider Trophy to Spitfire  which also provides material for wider reading, grouped according to specific areas of interest.