From the Rolls-Royce experimental archive: a quarter of a million communications from Rolls-Royce, 1906 to 1960's. Documents from the Sir Henry Royce Memorial Foundation (SHRMF).
Journal page discussing engine component failure modes such as ring-sticking and bearing corrosion, with supporting test data graphs.
| Identifier | ExFiles\Box 154a\2\ scan0010 | |
| Date | 1st January 1939 | |
| 10 S. A.{Mr Adams} E.{Mr Elliott - Chief Engineer} JOURNAL (Transactions) Vol. 44, No. 1 Fig. 1 - Relation between initial side clearance of top piston ring and cylinder temperature required to produce ring-sticking in 5 hr. 250-cc. air-cooled J.A.P. engine No. 3, 3000 r.p.m., non-detonating conditions. Same lubricating oil used for all tests. [Column 1] rings on running time before sticking occurs in the single-cylinder Norton engine. Variations in piston skirt clearance also have a marked effect on ring-sticking temperature. Since an experimental error greater than ±2 deg. cent. could not be permitted on check tests, it became necessary to exercise the most careful control of mechanical conditions, and to use standardization or check tests at frequent intervals. These observations do not apply only to the small engines; they are equally applicable to the aircraft-engine tests. How little regard is paid to the importance of this aspect of the problem is evidenced by the occasional submission of results based on flight tests made in good faith on engines frequently of different age, probably on different aircraft operating on different routes. If flight tests of this nature are to be of any value, much closer control of the tests must be exercised, or mass data must be obtained over thousands of hours of engine operation. Bearing Corrosion Attention first was drawn to the question of bearing corrosion with the use of lead-bronze as a bearing material in automobile engines. The introduction of wholly solvent-extracted oils at about the same time appeared to aggravate the trouble. Laboratory tests have been developed for the determination of the corrosive action of oils on bearings, but there appears to be a lack of data to show how far these tests correlate with practice. It was clearly necessary, therefore, first to develop an engine test from which existing laboratory tests could be checked, or alternatively, to make an attempt to modify them so that reliable correlation could be obtained. In order to get bearing corrosion in a reasonable period of time, engine conditions had to be made unusually severe. Since the temperature of the bearing appeared to be the controlling factor, an engine was modified to operate with extremely high oil temperatures, ranging up to 170 deg. cent., as shown in Fig. 3. The tests consist of runs of 20 hr. preceded by 4 hr. of running in of the big end at normal oil temperature (50-60 deg. cent.), the oil temperature being found at which bearing corrosion just develops. With lead-bronze bearings an oil was considered satisfactory for most [Column 2] purposes, providing the temperature at which corrosion appeared on the bearing was not below about 145 deg. cent. The results on lead-bronze vary considerably with the brand of bearing metal, which appears to be very liable to “erosion” effects in the case of unsatisfactory metals. With cadmium-base bearings there appears to be an increased danger of corrosion from oil and, in view of the mechanical advantages of these bearings, tests in the special engines referred to have been conducted with them, and these tests have been followed by full-scale engine tests. There has been some difficulty in obtaining satisfactory correlation between small-engine and full-scale engine tests. Effect of Fuel on Carbon Formation Since the element carbon is the principal constituent by weight of gasolines, the possibility that the fuel may contribute to the formation of carbon deposits cannot be ignored. Some experimental results have been published² to show that the use of rich fuel-air mixtures gives somewhat greater total carbon deposits than those obtained with weaker mixtures. More recently Bouman, at the Delft Laboratory, has run a single-cylinder engine on hydrogen and on a non-leaded aviation gasoline without finding any appreciable difference in the amount of carbon obtained on the piston crown under comparable conditions of load and engine temperature. The carbon in the ring-grooves was actually greater with hydrogen than with gasoline. The general evidence of the experiments was to the effect that the fuel plays a relatively small part in the formation of carbon deposits, except in the case of leaded fuels when lead may contribute materially to the weight of deposit and also make it very adherent. Proportion of Oil Burnt to Carbon An attempt was made some years ago to estimate the proportion of the total oil consumed which burns to form carbon in the combustion space.³ The method of dealing with this problem consisted in adding a soluble metallic compound, Fig. 2 - Effect of piston-ring side clearance on duration of test at constant cylinder temperature - 490 cc. Norton engine [Footnotes] ² See S.A.E. TRANSACTIONS, Vol. 21, 1926, Part II, pp. 150-181: “Influence of Temperature, Fuel, and Oil on Carbon Deposition,” by S. P. Marley, C. J.{Mr Johnson W.M.} Livingstone, and W. A.{Mr Adams} Gruse; see also Aircraft Engineer, Vol. 2, May, 1930, pp. 109-113: “The Lubrication of Engines,” by O.{Mr Oldham} Thornycroft and C. H.{Arthur M. Hanbury - Head Complaints} Barton. ³ See Aircraft Engineering, Vol. 2, May, 1930, pp. 109-113: “The Lubrication of Engines,” by O.{Mr Oldham} Thornycroft and C. H.{Arthur M. Hanbury - Head Complaints} Barton. | ||