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).
Detailed specification for the design and operation of axial flow turbine and compressor blades based on fluid dynamics.
| Identifier | ExFiles\Box 147\2\ scan0204 | |
| Date | 21th August 1939 | |
| 4 511,278 dynamic design of rotor blades. Yet the principle condition, namely that the discharge from the rotor should be as nearly as possible a free circular vortex of uniform axial velocity, seems to have been generally disregarded. Some workers have recognised that the ideal condition is one in which the rotor impresses uniform energy upon all the fluid passing therethrough, which leads to the condition that the rotor discharge should be a vortex, but, for reasons best known to themselves, they appear to have utilised blade formations which cannot satisfy this requirement. Experiment has shown that in fact the rotational type of flow is extremely stable and tends to occur in the discharge from a stator nozzle ring or from the rotor of a compressor even if the formation of the blades is not one specially adapted to produce this kind of flow, but if the maximum of mechanical energy is to be obtained from the expansion of gas through a turbine, then it is theoretically necessary to arrange that the nozzle ring should conform to a flow of the fluid therefrom in a free vortex of uniform axial velocity, and that the rotor should produce a purely axial exhaust flow. In the case of a compressor, the corresponding condition to be achieved is that the discharge from the rotor should be a vortex of uniform axial velocity and the discharge from the stator should be purely axial. It will be appreciated that the condition specified, namely that the flow between a rotor and a stator should be either a vortex of uniform axial velocity or a purely axial flow, means that the rotational component of velocity, if any, varies inversely as the radius. The present invention is based fundamentally upon a realisation of the fact that in order to increase the efficiency obtainable in machines of the kind referred to by avoiding losses due to undesired changes in the flow of the operative fluid through such machines, it is necessary to adopt formations of the rotor and/or stator blades which correspond and conduce to the type of flow above mentioned. The invention therefore comprises an axial flow turbine, compressor, pump or like rotary power conversion machine, operating with compressible viscous fluids, wherein are employed blades which are so shaped that their effective entry and/or leaving angle or angles and pitch vary progressively at increasing radial distances along an individual blade in such a manner that the said effective angle or angles and pitch at each said radial distance correspond substantially to the motion of the operative fluid relative to said blade at each said radial distance upon the basis that the velocity of said fluid varies inversely with radial distance. The invention also comprises an axial flow turbine, compressor, pump or like rotary power conversion machine wherein are employed blades which are so shaped and positioned as to conform to the condition that, when the machine is operating under its normal design conditions, in flowing through the said blades, the operative fluid, being compressible and viscous, has its angular momentum about the axis of the machine changed from one value which is substantially uniform before engaging the blades, to another value which is also substantially uniform after leaving the said blades, whilst the axial velocity of said fluid is substantially uniform both before entering and after leaving the said blades, though the magnitude of the axial velocity is not necessarily the same before entering and after leaving the said blades. Other features will appear from this specification. Hitherto the accepted distinction between "impulse" turbines and "reaction" turbines has been that in the former for the highest diagram efficiency the ratio of the blade speed u to the rotational component of the fluid speed vw, which has been commonly designated by the expression u/vw (in the following description) is of the order of 0.5 whilst in the latter it is of the order of 1.0. In the light of my researches however this distinction is less marked, since it may be shown that a so-called "impulse" turbine which upon hitherto accepted assumptions as to fluid flow has a value for u/vw at mid-blade of 0.5 may have in fact a certain degree of reaction at the blade tips, whilst it is theoretically possible to provide a single-row low turbine, or one with a single row per stage, in which the value of u/vw at mid-blade is much less than 0.5 and still have axial discharge. It might be thought that it would only be important to allow for the rotational character of the flow in turbines where the blade length is very long radially in proportion to the mean diameter. It has, however, been shown by experiment with gaseous fluid that a substantial gain in efficiency can be obtained by providing in the design for rotational flow in a case where the blade length is approximately 511,278 5 only 8% of the mean diameter. From the foregoing it follows that, for impulse blading, to yield substantial efficiency there has to be recompression at the root. This is difficult to achieve without loss. Impulse blading possesses the big advantage over reaction blading that for a given wheel speed, a much greater energy conversion per stage can be obtained. It is a corollary of the foregoing that the wheel which would obtain the highest energy conversion per stage whilst avoiding the losses associated with recompression, is one in which u/vw at the root is 0.5 or slightly greater. This represents a well defined design which lies between the accepted form of impulse and reaction types, i.e. u/vw at mid-blade is higher than for an impulse blade and lower than for a reaction wheel. Moreover, such design is of a wheel on which the pressure difference across the disc is nearly zero. In such a blade the degree of reaction increases towards the tip, from a zero value at the root. Now, as a practical matter of axial-flow rotor blade design, such blades, whilst having a certain pitch angle, have also been previously given "twist", i.e. a progressive change of effective pitch angle from root to tip, in accordance with the different linear speeds of points on a blade at different radii; as hitherto it was assumed that the fluid velocity was constant, such twist was assumed to result in the blade meeting the gas at the same angle at all radii. According to this invention, such blades are formed with a much greater twist than that which is determined by the different linear velocities of different radial stations of the blade, such greater twist being that which takes account of the radial gradient of fluid velocity (which corresponds to, but is in inverse sense to, the radial pressure gradient). Further, the last mentioned feature being adopted, a clearance is preferably left between the leaving edges of turbine stator nozzle blades and the leading edges of the rotor blades, of at least a quarter of the chord of the rotor blades, and a clear space so formed is walled-in peripherally, preferably by cylindrical extension of the rim of the nozzle diaphragm. Other novel features are also preferably provided, briefly indicated below. A preferred construction of an axial flow turbine with blades formed as above, is such that u/vw at the blade roots is .5 or slightly greater. A preferred form of velocity-compounded turbine having the blades of successive rows so arranged, that only the last row creates axial discharge, whilst the angular momentum of the fluid in the spaces between rotor and stator blades is uniform in each said space. In fully applying the invention both rotor and stator blades are formed in accordance with the basic principles hereinbefore enunciated. It is to be understood that there is a possibility that the pitch as between two sections at different radii (i.e., radial stations) of a blade may in effect be varied either by shaping the blade with a constant axial dimension, or by tapering the blade, e.g. by curtailing its leading edge. In the drawings herewith:— Figure 1 is a formal diagram of turbine nozzle and rotor blades in conventional showing; Figure 2 is a vector diagram related to Figure 1; Figure 3 shows a turbine blade in end view from which can be seen a root section (thicker section) and a tip section (thinner section); Figure 4 is an elevation of the blade of Figure 3, viewed axially; Figure 5 is a diagrammatic and partial view of a blade arrangement showing stator and rotor blade relationships; Figures 6 and 7 are diagrams corresponding to Figures 1 and 2, in the case of a compressor. The invention is explained graphically with the aid of the accompanying diagrams, in which the method of arriving at the new blade formation is as follows, in the case of a single-row turbine to yield purely axial exhaust flow. Figure 1 diagrammatically illustrates turbine nozzle blades 1 with leading edges 2 and leaving edges 3, and rotor blades 4 with leading edges 5 and leaving edges 6. Certain angles show the fluid-flow direction, referred to the plane of rotation; A is the angle of the flow from the nozzle blades at 3, B is the angle at which this flow meets the rotor blades at 5, and C is the angle at which the fluid leaves the edges 6. The rotor blades move according to the arrow in this Figure. Figure 2 is a vector diagram enabling | ||