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 the design principles and fluid dynamics of a hydro-kinetic power transmitter, including diagrams of fluid flow.
| Identifier | ExFiles\Box 156\4\ scan0090 | |
| Date | 2nd September 1938 guessed | |
| 4 a vaned driving element to a vaned driven element rotating relatively to one another at greatly differing and constantly varying speed, it is evident that, if losses of prohibitive magnitude are to be avoided, the rotational velocity with which the liquid emerges from the driving element must be so controlled that the direction in which the liquid impinges upon the receiving ends of the vanes of the driven element is maintained, under all conditions, at an angle at which the liquid may be received without appreciable shock. Hitherto such control has not been achieved. Control of velocity variations implies, as a pre-requisite, steadiness of motion of the liquid, and it was held that in a centrifugal pump—and therefore in an hydro-kinetic power transmitter—the motion of the liquid could not be other than unsteady. The reason for this is given in a later section headed “Vane and Duct Characteristics.” Hence the accepted theory that under variable conditions an hydro-kinetic power transmitter can function with acceptable efficiency only within a comparatively narrow range of variation, and that efficient performance by a single transmitter of the widely differing functions of a clutch and of a variable ratio gear, is an impossibility. The fundamental characteristic of the transmitter which is dealt with herein is that its driving element is a centrifugal pump of such a construction that the motion of the liquid within its ducts remains steady for the velocities that obtain in the ducts, with the result that a steady stream can be maintained within the hydraulic circuit of the transmitter under all conditions, and velocity variations can be, and are, adequately controlled. By virtue of this characteristic both the fundamental and the consequential causes of loss of energy are virtually eliminated. The other two salient features of the transmitter in question relate, respectively, to its driven element and to its reaction element. To enable an hydro-kinetic transmitter to perform effectively the functions of a clutch, the driven element needs vanes having characteristics differing widely from those requisite for performance of the functions of a variable gear, and it has therefore been held that the combination of the two functions in a single transmitter must perforce involve resort to mechanically variable or deformable vanes. In the transmitter about to be described any such complication is avoided. The profile of the vanes is such that their hydraulic action, in co-operation with that of the reaction element, varies with varying conditions, although neither their disposition nor their conformation is variable. A similar difficulty existed in connection with the reaction element because the angle which the vanes of that element must occupy to enable them to perform their purpose when the transmitter functions as a variable gear is detrimental to efficiency when the transmitter functions as a clutch. This difficulty has been overcome by the evolution of reaction vanes of such a conformation that, when their action is not needed, the effect of their angular disposition becomes virtually neutralised, although they are in no way movable or deformable. Essential Points of the Design.—The vanes of the centrifugal pump or driving element are disposed in (6474-B) ENGINEERING Fig. 2. FACE VIEW OF ASSEMBLY OF DRIVEN ELEMENT AND REACTION ELEMENT, SHOWING VANES AND PAWLS OF REACTION ELEMENT. 9 ducts having the conformation of the ducts shown in Figs. 21, 22 and 23, respectively. The dimensions of the three ducts used for the tests were proportioned to fit a pump wheel of a diameter at outlet of 7 in. The values obtained in the tests, together with the corresponding values for the oil medium on the basis of kinematic viscosity above stated, are set out in the accompanying Table. From the values given, the corresponding values for ducts made to any larger scale can be calculated according to the formula of Reynolds. The photographic views, showing the nature of the motion of the liquid in the three ducts at different velocities of flow are indicated in the Table. The first view given of each of the three ducts was taken under conditions of perfect streamline motion, the band of coloured liquid appearing undisturbed throughout the length of the ducts; the second view was taken at the respective velocities of flow at which the streamlines begin to break down in each duct at the outlet. The third view was taken at higher velocities than the second, and illustrates the magnitude of the eddies forming at the outlets of each duct while streamline motion is still prevalent near its inlet. The fourth view of the ducts, shown in Figs. 21 and 22, illustrates the high degree of turbulence which is produced in these ducts, even at velocities of flow considerably lower than those obtaining in a transmitter under normal working conditions; and it should be borne in mind that the turbulence shown represents only that which results from the conformation of the ducts, quite apart from the further and much graver disturbance which is set up in the ducts illustrated in Figs. 21 and 22 by the angular motion of the wheel, and which does not arise in the duct shown in Fig. 23. The fourth view of this duct was taken at the maximum velocities at which the liquid could be drawn, having regard to the pressures which the test apparatus would withstand. Even at such inordinately high velocities the stream preserves a remarkable degree of steadiness, only minute eddies being present. The values given in the Table, in respect of the oil medium, considered in conjunction with the diagrams in Figs. 17 to 20, make clear that a condition of angular divergence in the construction of its ducts ensures in the transmitter steadiness of the stream under all normal working conditions, even for high velocities. The principle of convergence being adhered to in the construction of the ducts of all the elements of the transmitter, it is suggested that the liquid emerges from and is received by each element in a steady annular jet, the pre-requisite of control and of co-ordination of velocity variations of the stream under varying conditions being thus established. The effect of the disposition of the outlet ends of the vanes of the driving element of any hydro-kinetic transmitter at an angle backwards relatively to the direction of rotation is to impart to the liquid emerging from the driving element a component of velocity backwards relatively to the speed of the ends of a vane of a magnitude varying substantially as the speed of relative rotation of the driving and driven elements; but if inordinate eddy motion exists within the hydraulic circuit, this relationship will not hold good. In the transmitter, since the stream remains steady under all normal working conditions, the backwards component varies substantially as the speed at which the driving element rotates relatively to the driven, the rotational velocity of the liquid emerging from the driven element being thereby reduced, under all conditions, to the extent requisite to enable the bulbous ends of the vanes of the driven element to receive it virtually without shock. Any theory such as here suggested must, however, be fully justified by experiment and road tests and brief reference is made to these later. Mechanical Construction.—The overall dimensions of the transmitter, whether it comprises one stage only, or whether it be of the two-stage type, are identical. In both cases the construction is throughout sturdy and does not comprise parts liable to get out of order. The matter of wear and tear being virtually absent, the Fig. 17. Fig. 18. Fig. 19. (6474-F.{Mr Friese}) Fig. 20. "ENGINEERING" | ||