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).
Piston failures, including scuffing and seizure, across various engine models.
| Identifier | ExFiles\Box 107\3\ scan0132 | |
| Date | 27th June 1938 | |
| - 2 - BY.10/G.27.6.38. judged by the above facts to our experiences when the aluminium piston was introduced on the 'Silver Ghost' when R.{Sir Henry Royce} provided a deep top land somewhere round about 5/8" in axial depth, with the object of preventing the top ring from gumming, the result was the top land picked up in a precisely similar manner to the pistons under discussion, and large portions were torn away, but to a greater extent owing to the increased depth of land. We have had a somewhat similar complaint in regard to the Bentley car B.3-EJ. Col. Hancock, total mileage 40,000, decarboned at 26,000, seized up at a further 14000 miles. In this case, however, there was the added feature that five pistons showed serious 'scuffing' but it was evident that the 'scuffing' was not responsible for the final failure, the condition of the cylinder making it clear that picking up had started on the top land of No.6. We have had no failures of this type on the 20 HP. or 20/25, the fault being peculiar in Rolls-Royce cars to the 25/30. The differences between the pistons of the different types are that on the 25/30 an attempt was made to copy the general proportions of the Aerolite piston in regard to rings and depth of top land, but the top land was made .002 smaller than the Aerolite piston for the Bentley, apparently following our usual Rolls-Royce clearance dimension. With the 25/30 was introduced the Stromberg carburetter with its petrol pump coupled with the accelerator. The oil hole in the rod was deleted at the commencement of the series to keep down oil consumption, but since 'scuffing' has not taken place on any of the three RR. cars affected the failure could not be ascribed to the deletion of this feature. We measured the piston over the carbon taken from one of the engines which had seized, not including the seized piston itself, the increase in dimension of the piston on the top land due to carbon, which had apparently worked satisfactor-ily was .008 to .009, and one would suggest an increase of .004 or .005 to the upper land as being a reasonable addition to prevent the deposition of a thick carbon band as the thickness of the carbon seems to be a factor in the failure. This point is more or less proved by the fact that the first two chassis GXM.54 and GXM.18 were decarboned twice before failure took place, seizure taking place at a total mileage of 21,500 in the case of GXM-54, 27,000 - GXM-18, and 19,400 - GWN-56. This represents a record of the facts, and some of the suggestions put forward to account for the failures. | ||