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).
Aircraft engine lubrication experiments studying ring-sticking and piston cleanliness with various oils.
| Identifier | ExFiles\Box 154a\2\ scan0009 | |
| Date | 1st January 1939 | |
| January, 1939 AIRCRAFT-ENGINE LUBRICATION 9 most suitable test conditions to employ on these engines. It was found that, to obtain ring-gumming in any period less than 5 hr., involved engine conditions so severe that considerable difficulty would be encountered in maintaining accuracy of results on account of excessive wear and tear of the engine. A 5-hr. test period was therefore adopted. At the time the experiments were initiated little was known about the effects of weak-mixture running on lubrication—in fact aircraft engines were not then capable of running continuously at the weak mixtures commonly employed under cruising conditions today. It was known, however, that detonation had an appreciable influence on ring-gumming, and investigation showed that, at the same mixture strength, the occurrence of detonation required the lowering of cylinder temperatures by about 10 deg. cent. to obtain the same degree of ring-gumming as with no detonation. The first set of test conditions was therefore established, that is, using a fuel of an antiknock value just to give moderate detonation at the beginning of the test, the intensity of detonation being maintained constant throughout the test by adjustment of ignition timing. Subsequent experience on aircraft engines showed the mixture strength to have an appreciable influence on ring-gumming, and the subject was examined on the test engines. The experiments made showed that the relative rating of oils could be changed according to the mixture-strength conditions under which they were tested. As a result, it was decided that oils should be tested both under rich-mixture (full-power mixture strength) detonating conditions, and at weak mixture (cruising mixture strength) without detonation. Experience gained with these engines was valuable, and correlation with practice was obtained with certain exceptions. To extend the investigation, particularly in the direction of studying the behavior of oils in tests of longer duration, a Norton racing motorcycle engine has been installed (for details see Appendix 2). This engine, as well as the J.A.P. engines previously mentioned, are operated at the works of Messrs. Ricardo & Co., Shoreham-by-Sea, England. In addition to comparative tests on ring-sticking and piston cleanliness, this engine is being used to study the building up of carbon in the ring-grooves. The results so far indicate that there is little connection between ring-sticking and the amount of “ring-packing,” that is, the quantity of carbon formed behind the ring. Incidentally, the same conclusion has been reached from the J.A.P. engine work. It will be evident that results obtained on these small engines had to be correlated with full-scale engine results before they could be considered of any value. A certain amount of data were obtained from various tests and service results, but there were always too many gaps in these data to permit of reliable correlation being obtained. We were fortunate at this stage in having the cooperation of The Bristol Aeroplane Co., Ltd. A single-cylinder test unit was installed for oil tests and a test program devised. This program entailed not only the actual running conditions to be used but also the control of component parts of the engine used for each test. See Appendix 3. The table opposite shows some comparative results obtained with the small laboratory engines and the single-cylinder Bristol unit. The lubricating oils used for the foregoing experiments were chosen for their good performance with respect to ring-sticking at high temperatures. In the single-cylinder Bristol engine the tests on oils A to E were each 65 1/2 hr. For the test on F onwards, each oil was run for 75 hr. For the test on oils F to M the cylinder-head was modified and, for the check test on oil H and succeeding work, a new type of piston was used. From the point of view of ring-sticking the best oils in the Bristol engine are E, D, H, J, and F.{Mr Friese} The first four of these oils also show the highest resistance to sticking in the J.A.P. engines under condition (a), but oil F is under-rated. Oils A, B, and K come out slightly inferior to the first group both in the Bristol engine and in the J.A.P.’s under test condition (a). In the J.A.P. units, according to test condition (b)) D, H, and F come into the first class, but E gives less satisfactory results. As regards the cleanliness of the piston and, in the case of the Bristol engine, the engine as a whole, the oils A, B, K, L, and N are the best in the Bristol tests. Excepting oil L, which gave a dirty piston under condition (a), all these oils gave good results in the J.A.P.’s. The dirtiest oils in the Bristol unit were D, E, and M; these oils also gave similar results in the J.A.P.’s excepting oil M under condition (b). In general the appearance of the J.A.P. pistons is likely to vary considerably from test to test even when the running conditions are maintained as constant as possible. In this direction the longer tests in the Norton unit seem likely to give more definite results. Even at this stage, it was not certain that correlation would exist between results on the single-cylinder unit and in service. The next stage, therefore, was full-scale engine bench tests. These, consisting of 200-hr. runs simulating service conditions, have given results in good agreement with the single-cylinder unit. Further data obtained in airline service under comparable operating conditions are being collected. It will be appreciated that, in the development and pursuance of the tests enumerated, much incidental data have been accumulated. Effect of Mechanical Conditions The influence of piston-ring side clearance is shown in Fig. 1. In Fig. 2 is shown the effect of side clearance of the [TABLE] Oil | Single-Cylinder Bristol Engine | J.{Mr Johnson W.M.} A.{Mr Adams} P. Engines (5-Hr. Tests) Ring-Sticking Temperatures, deg. cent. --- | --- | --- | Rings (Top) | Rings (Second) | Cleanliness | (a) Maximum power with detonation | (b) 10 percent weak, no detonation Fuel-air mixture A | Free | 75 percent stuck | Very clean | 295 Clean | 275 Clean B | 30 percent stuck | 25 percent stuck | Very clean | 295 Clean | 275 Clean D | Free | 40 percent stuck | Dirty | 300 Dirty | 280 Dirty E | Free | Free | Very dirty | 300 Dirty | 275 Dirty F | Free | 40 percent stuck | Clean | 285 Clean | 280 Clean G | Free | 100 percent stuck | Clean | 285 Dirty | 280 Clean H | Free | 30 percent stuck | Fairly clean | 300 Dirty | 280 Clean J | Free | 30 percent stuck | Clean | 300 Dirty | 280 Dirty K | Free | 60 percent stuck | Very clean | 295 Clean | 270 Clean L | Free | Ring broke | Very clean (rather better than K) | 295 Dirty | 290 Clean M | Free | 90 percent stuck | Rather dirty | 295 Dirty | 280 Clean H (check) | Free | Free | Clean | 295 Clean | N | Free | 100 percent stuck | Very clean indeed | 295 Clean | 275 Clean | ||