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).
Analysis of stress concentration and uneven force distribution on a roller component.
| Identifier | WestWitteringFiles\R\2October1927-November-1927\ 95 | |
| Date | 1st May 1927 guessed | |
| -5- The circles of radii r1, r2 touch the ellipse at the points P and Q which are the feet of the two real normals which can be drawn from O¹ to the ellipse. If, however, we wish to construct the surfaces traced out by the generators of a roller which lie nearest to and furthest from the axis of revolution respectively, we find that these generators pass through the points H, K of fig. 2 which are the extremities of the minor axis of the ellipse, and the corresponding radii of the sections are O¹H and O¹K.{Mr Kilner} The diagram, however, shows clearly that O¹P < O¹H and O¹Q > O¹K, so that, if the bearing surfaces were constructed in this manner, e.g., by revolving a tool set at the given angle α to the axis, there would be no room for the roller at a distance from the plane of symmetry, if it fitted tightly in that plane, or, in other words, the roller would be gripped at the ends and loose in the middle. This is a point of some importance, as it would involve the concentration of stress on the ends of the rollers. 5. UNEVEN DISTRIBUTION OF FORCE ON THE ROLLER. Even if all surfaces have been accurately shaped in the manner indicated above, there will still be, when jamming occurs, a concentration of force on the end of the roller. It has already been shown that, when the clutch acts, the collar tends to roll up (or down as the case may be) the shaft. In this process the surface AC of the collar is displaced by translation to A¹C¹ and fig. 3 shews that in consequence the interval between collar and neck is narrowed far more in the neighbourhood of, say, C than in that of A.{Mr Adams} FIG. 3. | ||