Rolls-Royce R Srs
“R” of 1931
This special racing engine, first developed for the Supermarine S.6s participating in the Schneider Trophy Contest of 1929, was a development of the Buzzard, and having most moving parts in common with that engine, it differed considerably in appearance. External modifications, made in order to allow better streamlining, affected the valve covers, cylinder blocks, bottom cover and bearer feet. Auxiliaries were redisposed to reduce the depth, and the length of the reduction-gear housing was increased in order to fine down the cowling lines. The gear itself was of 6 : 1 ratio. A major factor in achieving the output of 1,900 h.p. at 2,900 r.p.m. was an entirely new supercharger.
The 1931 model gave 2,300 h.p. at 3,200 r.p.m. for a weight of only 1,630 lb. This meant that the power had increased 21 per cent for a weight increase of only 6.5 per cent, an advance obtained largely by increasing the engine speed, the supercharger gear ratio and the size of the air intake.
Rolls-Royce decided that a one-hour run at full power should be the first mark to be passed before the 1931 "R" engines could be delivered as airworthy. By the end of April the early specimens were usually lasting about 20 min before some kind of failure. By the middle of July they would still only do about half an hour non-stop; but on August 3rd a run of 58 min was accomplished at 2,360 h.p., and on August 12th one hour non-stop was completed at 2,350 h.p.
As part of the re-design of the 1931 engine, an entirely new type of connecting rod was developed, and very considerable modifications were effected in the crankshaft and crankcase, to withstand the terrific loads. "Some idea of the magnitude of the crank-chamber stresses," Flight remarked, "may be gained from the fact that the load on the centre main bearing due to centrifugal and inertia forces was no less than nine tons. Alternative bearing metals were tried, but white metal was finally made to stand up. An obscure trouble experienced at one time was sidelashing of the big-end bearings against the webs, which caused failure of the white metal on either side of the big ends."
Considerable trouble occurred with valve springs, more than was expected, considering that the speed was raised by only 300 r.p.m.; in fact, at one time two or three would be found broken after a ten-minute run. Extensive tests were run on a valve-spring rig consisting of one cylinder block. and all aspects of design and material were thoroughly reviewed. The springs finally arrived at were somewhat reactionary, but no further failures at all were experienced on the engines after fitting them.
"The oil consumption," Flight reported, "rose to terrific figures at the 1931 speeds and powers, partly owing to great quantities lost through the breathers. On one 25-min run the oil consumption was at the rate of 112 gallons an hour. By weeks of work on combinations of scraper rings and crankcase breathers, by modification of the scavenging system, and by the final adoption of a deeper sump which filled all the available space in the machine, the consumption was reduced to about 14 gallons an hour for the final race engines. It also effected a considerable reduction in the oil temperature rise through the engine; oil entered the engine at about 80 deg C and came out at about 140 deg C. The oil used was pure castor."
The fuel to be used was the subject of many tests, for a compromise had to be effected between power and specific consumption. Invaluable assistance was rendered by the Ethyl Export Corporation. The plugs used in both 1929 and 1931 were Lodge Type X170, which gave very little trouble indeed. During the course of development it was found that an actual engine-run was a more severe test of insulation and general performance than anything the plug-makers could impose themselves.
It was decided to give all plugs an endurance test in an engine before sending them to Calshot. After this run they were returned to the Lodge firm, who examined the inners, repolished them and fitted new outer bodies, before despatching them to Calshot. A few doubtful cases, which might have caused trouble in flight, were found in this way. Actually, no plug troubles of any description were experienced in the air, and the same plugs could be used for warming-up and taxying as for full-throttle flying.
Both B.T.H. and Watford magnetos were used on the race engines. Some trouble was experienced with these on the test-bed, due to the vibration to which they were subjected; consequently every nut on them had to be split-pinned, and certain parts had to be stiffened up.
Before despatch the engines were tested in a hangar with an airscrew fitted. The tick-over speed was set at 475 r.p.m., and the engines were remarkable (as racing units) for their tick-over and flexibility. Once in the aircraft they were started by compressed air or with a Bristol gas starter.
Hiduminium alloys (R.R.50 series) were extensively used and, the life of the engines being short, aluminium-alloy forgings replaced bronze and steel in many parts. From the experience gained from many smashed-up engines, a "life of parts" list was drawn up, and pieces were scrapped after their allotted span, whether broken or not. This procedure was found to be cheaper and quicker than rebuilding smashed engines. There was hardly a single component part in the engine which did not receive design attention and which was not improved in someway-from large units like the crankcase and cylinders down to individual nuts and bolts all over the engine.
For short periods powers appreciably greater than 2,350 b.h.p. were obtained from the "R," and on September 9th, 1931, a power of 2,530 h.p. was taken from a "sprint" engine, enabling an S.6B to establish a world record of 407.5 m.p.h.
The total production of "R" engines was 20.
Engines of the "R" type were used not only in the S.6, S.6A and S.6B seaplanes, but for land-speed records in Bluebird, Speed of the Wind and Thunderbolt and for water-speed records in Miss England If and III and Bluebird II and III.