Keeping them in

THOSE who have experienced difficulty in keeping bearings in high-speed engines should find much to interest them in the report of the I.A.E. Research Committee on “Oil Flow Through Engine Bearings,” prepared by J. Spiers, M.Sc., and published in the January issue of the I.A.E. Journal. The tests were made on a 4-litre 85 x120 mm. six-cylinder engine having thin-walled, steel-backed bearings, with the exception of one main bearing lined with copper-lead alloy, and a four-bearing counter-balanced crankshaft. Bearing clearances varied from approximately 0.0007″ to 0.0055″, engine speed from 1,000-3,500 r.p.m., lubrication pressure from 35 lb. to 100 lb. per square inch, and the oil inlet temperature from 40° to 100°C. Two grades of oil, S.A.E. 50 and S.A.E. 20 were used. An approximate linear relationship existed between pump pressure and speed over the temperature range studied. Big-end pressures differ from main-bearing pressures on account of the pressure built up by the oil column under centrifugal load on the side of the crankshaft centre line remote from the big-end, and if this pressure is not overcome by the supply pressure, feed to the big-end may fail. In addition, the presence of oil spray holes or gudgeon pin feed holes and relatively small end clearance; as against circumferential grooves and large end clearance in main bearings, also affect flow. Oil flow is proportional to pressure at any given speed, and for any given pressure increases with engine speed. Flow rapidly increases with temperature rise, rising from just over 0.1 lb./min. at an inlet temperature of 40°C., to 0.2 lb./min. at 70°C., on the test engine, at 3,000 r.p.m. Using a light oil greatly increases the flow, but this does not vary in proportion to the viscosity change at the inlet, because, for a given inlet temperature, the oil film within the bearing will be at a lower temperature in the case of a light oil, and consequently viscosity is not decreased in a proportional manner with oils of varying inlet viscosity. On the test engine, a flow rate of under 0.25 lb./min. at 3,500 r.p.m. and 40°C. approximately using S.A.E. 20 oil changed to about 0.36 lb./min. at over 70ºC., whereas, using S.A.E. 50 oil under identical conditions, the change was from about 0.18 lb./min. to almost the same figure of approximately 0.36 lb./ min. For these tests a bearing clearance of 0.0007″ was used and the engine run at a b.m.e.p. of 30 lb. per square inch, the effect of load having been found negligible. The flow in the big-end bearings was generally similar to that in the mains, but the use of heavy oil had less effect on the rate of flow expected as big-end bearings operate at a higher temperature than the mains and so over a flatter part of the viscosity curve. Bearing clearance also had less effect on flow speed, but the big-ends ran at greater clearances than at the minimum used for the mains. The flow increase in the test engine consequent upon increasing the main bearing clearance by .0038″, from 0.0007″ to .0045′, was approximately 2.45 lb./min., at 3,000 r.p.m. and a pressure of 60 lb. per square inch. The only instance in which flow in a big-end followed different characteristics from that of flow in a main was in the case of a connecting rod which had a considerable oil escape from the feed to the little-end bush, and in which this was fed by a column of oil 7.875″ long, the inertia of which would raise the pressure above that in the big-end itself, which column was in direct connection with the main oil supply for 30° of crankshaft rotation. In this case the flow rose to approximately 1.3 lb./min. at 2,000 r.p.m. and a pressure of 60 lb. per square inch against a flow rate of 0.5 lb./Min. for the other rods, and the flow increased in greater proportion as speed rose, speed predominating over all other factors. Big-end clearances were 0.0015″ to 0.0020′ and 0.0050″. General findings were that flow rapidly increases with speed, even when inlet temperature and, in the case of main bearings, the supply pressure remained constant. This is also true of big-end bearings, even after making allowance for the effect of centrifugal pressure, and the author strongly discredits the pumping theory. The temperatures in the bearings vary linearly with speed to a first approximation, under conditions of constant inlet temperature. Sump temperatures and bearing temperatures vary considerably, and viscosity of the bearing film is of importance in determining flow values; a rational explanation of flow increase with speed may lie in the reduced film viscosity due to increased operational temperature. At a surface speed of 2,000 ft./min. bearing temperature decreased from 130°C. to 80°C. with a diametral clearance increase of 0.004″ for a big-end and from 143°C. to 86°C. (approximate figures) for a main bearing, for an inlet temperature of 70’C., using S.A.E. 50 oil. Flow is influenced by supply pressure, film viscosity, and diametral clearance, while total flow may be influenced by design features, and when a certain value of “clearance area” is exceeded, there appears to be a disproportionate increase in flow. Operating temperature is independent of load and varies linearly with speed. Flow is directly proportional to supply pressure, inversely proportional to film viscosity and, as tested, directly proportional to the square of the diametral clearance, up to 0.005″. A report on the effect of clearance on temperature is in course of preparation which will doubtless prove of great interest.