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).
Study on engine cranking resistance, 'break-away torque', and the behavior of different oils at various temperatures.
| Identifier | ExFiles\Box 145\3\ scan0257 | |
| Date | 18th February 1937 guessed | |
| - 4 - The results have been re-grouped to show that the "break-away torque", which has always been so clearly felt when hand-cranking an engine, is indistinguishable from the "just moving" torque on the new oils, with the exception of new Aero. The same effect, applying to all the new oils, is noticeable when the engine is hot. This is a pleasing feature of these new oils, and it may possibly mean less cylinder wear. Presumably a thicker oil film remains in situ between rings and walls than with the old range. (5) Engine Cranking Resistance. It will be remembered that it has been previously shown that under normal circumstances, the torque T lbs.ft. required to crank an engine at N r.p.m. with an oil of viscosity V centipoises is given by T = A + B √NV. As each test was completed, a sample of sump oil was extracted and tested both for dilution and for viscosity at the temperature of the test. The torque, derived from the calibrated starter, was plotted against √NV. The "just moving" torque is taken as the zero speed reading. It will be seen that with the exception of one oil, Old Shell Triple, a reasonably good straight line can be drawn through the points, corresponding to the equation. T = 5 + .062 √NV An attempt to find out why this oil behaves differently from the other oils was made by carrying out cranking tests at rather higher temperatures. It was found that it behaves quite normally at + 1.5ºC but deviates from the line, i.e. does not behave according to its viscosity, at temperatures below this. It is thought to be due to partial separation of wax from the oil which would account for a difference in performance between a thin film of the oil and the comparatively large jet of oil in the viscometer. It was hoped that tests with the Mitchell ball and cup viscometer would show up the difference, but this method gave results similar to those of the viscometer (see E.4270). Using the car resistance formula above and substituting viscosity values of the various used oils at exactly -1ºC, a family of curves has been drawn showing the relation between torque and engine speed under these conditions. Across them has been drawn the starter torque-speed calibration curve to engine speed and torque ordinates. The intersections will give the cranking speeds under the particular conditions concerned, i.e. with these various oils and at -1ºC. Similar curves can be constructed for any other temperature or for any other oil. It must be remembered that the curves do not hold good for Old Shell Triple below a temperature of + 1.5ºC. (6) General Conclusions: This particular engine as we received it is most obviously "over-startered". The engine is particularly free, but apparently is only capable of an indicated torque of about 25 lbs. ft., so that it will not start immediately unless the engine resistance is somewhere near this figure. Our starter turns the engine at as much as 97 r.p.m. with the heaviest of all the oils - Old Shell Triple - which is plenty fast enough for carburation. In spite of this, the engine took 25 secs., before it ran. The mechanism of such a start is undoubtedly that starter and engine run in parallel until firing has warmed up the oil film to such an extent that the engine indicated torque is able to overcome the resistance torque, which is steadily falling as firing goes on. The ideal start occurs when the engine is able to produce enough torque to maintain motion right from the start. | ||