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).
The problems and considerations of high-speed impeller design for superchargers.
| Identifier | WestWitteringFiles\V\December1930-February1931\ Scan069 | |
| Date | 23th December 1930 guessed | |
| - 3 - on the problem of high speed impeller design has proven to me that this distortion in scale is not possible without a serious sacrifice of efficiency. In other words, it is not possible successfully to adapt the well known low speed impeller design; which is intended for 2500 to 3500 R.P.M. to supercharger service in which close ten times these speeds are necessary. A consideration of the following facts will make this clear: The peripheral velocity, particularly near the O.D. of a supercharger impeller eye, is necessarily very large. This causes the velocity through the impeller, particularly in the vicinity of the inlet, to be very much larger than in the case of the commercial low speed impeller. Since the energy represented by this high velocity flow is quite considerable, being several times the corresponding energy found in low speed practice, it is highly important that the flow through the impeller be treated kindly if serious losses are to be averted. To obtain the all important nicety of flow through an impeller of the type hitherto used is inherently impossible. This will be clear from the following:- (a) The inlet angle cannot be made accurately correct throughout the radial extent of the vane inlet edge. The primitive "hooks" of the vanes at the inlet give very poor passages on account of their abrupt flare and the absence of any parallel portions or correct throat". (b) The flow through the impeller is forced to undergo two very sharp eddy-producing deflections, one at the entrance, in the "hook", and the second one at the point where the flow changes from an axial direction into a radial direction. Due to the aforementioned high peripheral speed near the O.D. of the impeller eye: u 2 - u 1 ( u 1 is the peripheral velocity of the impeller inlet & u 2 the peripheral velocity at the outlet ) which represents the "centrifugal effect" the impeller has on the air passing through it is very much different for a particle travelling near the O.D. compared with a particle entering near the centre of the impeller eye. If each of the two could be made to travel in a separate pressure-tight conduit, it would be found that on account of the difference in (u 2^2 - u 1^2), the inner one would be compressed to a higher pressure than the outer one. contd. | ||