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).
Patent specification by Frank Whittle for improvements in the design and efficiency of axial flow turbines and compressors.
| Identifier | ExFiles\Box 147\2\ scan0203 | |
| Date | 15th December 1937 | |
| 2 511,278 momentum should be approximately equal in each row of moving blades. The construction of rotor blades is preferably such that centroids of progressive radial sec- 5 tions remain on a common radius to avoid bending loads. The invention includes the application of a machine characterised as above, in reaction propulsion systems. Dated the 15th day of December, 1937. For the Applicant, F.{Mr Friese} J.{Mr Johnson W.M.} CLEVELAND & COMPANY, Chartered Patent Agents, 29, Southampton Buildings, Chancery Lane, London, W.C.2. PROVISIONAL SPECIFICATION No. 1063 A.D. 1938. Improvements relating to Axial Flow Turbines and like Machines 10 I, FRANK WHITTLE, a British Subject, of Broomfield, Bilton Road, Rugby, Warwickshire, do hereby declare the nature of this invention to be as follows:— 15 This invention relates to axial flow turbines and like machines. It seeks to improve the efficiency more particularly of axial flow turbines, and possibly to render manufacture easier. It may also 20 tend to prevent flow leakage around the periphery of the rotor or rotors. The invention may be found to be cognate with that of Application No. 34687/37 (Serial No. 511,278), and in any case is preferably 25 employed together with the subject-matter thereof. According to this invention broadly stated, the nozzle ring or the nozzles of an axial flow turbine or like machine are 30 spaced from the plane of the adjacent edges of the blades of the rotor by a substantial uninterrupted gap, preferably of the order of half or more of the chord of the turbine vanes and preferably also 35 increasing in axial dimension or clearance with the radius. Thus a space is provided between trailing edges of nozzle vanes and leading edges of rotor blades, and/or vice 40 versa. Further according to the invention, vanes in a machine of the type stated forming a nozzle ring and/or rotor blades, are shortened chordwise as their radius increases, in such a manner that 45 for blades of a given formation, progressive curtailment at the trailing or down-stream extremities radially of nozzle vanes, or leading extremities of rotor blades, results in effect in a change of 50 pitch angle so as to produce fluid flow of natural vortex nature, i.e. with uniform axial components of velocity, and a rotational component which varies inversely with the radius. The blade ring of a tur- 55 bine characterised as above, is preferably inset in the casing in an internal circumferential groove. The formation and relation to the rotor of the nozzle ring characterised as above, 60 is intended to be such that pressure differences due to wake effect behind each guide vane, subside to almost or completely negligible value before the fluid impinges on the rotor blades, so that the 65 gap or space between nozzle and blades allows or causes stable flow conditions to be reached, with the intention that the blades are subjected only to a natural vortex flow from which maximum advan- 70 tage can be attained. Although nozzles or nozzle vanes may be set away from the rotor by a gap, the inner and outer walls defining the nozzle ring may (and preferably are) continued axially quite 75 close to the wheel or rotor blades, in the form of an annular space. Dated the 12th day of January, 1937. For the Applicant, F.{Mr Friese} J.{Mr Johnson W.M.} CLEVELAND & COMPANY, Chartered Patent Agents, 29, Southampton Buildings, Chancery Lane, London, W.C.2. COMPLETE SPECIFICATION Improvements relating to Turbines and Compressors I, FRANK WHITTLE, a British Subject, of Broomfield, Bilton Road, Rugby, 80 Warwickshire, do hereby declare the nature of this invention and in what manner the same is to be performed, to be particularly described and ascertained in and by the following statement:— This invention relates to axial flow tur- 85 bines, compressors, pumps and like rotary power conversion machines, operating with compressible viscous fluids. It is concerned to increase the efficiency obtainable in such machines by the employment therein of novel forms and arrangements of blades. 511,278 3 In determining the form of the blades 5 employed in turbines of the kind above referred to, it has hitherto been assumed that the operative fluid issues from the blades forming the ring of nozzle jets, in a series of discrete and separate jets, and 10 that these jets tend to continue to travel in straight lines. The usual text books accept this assumption. As the result of the general acceptance of this assump- tion blade formations have been adapted 15 to engage or discharge the operative fluid at constant fluid velocity and pressure over all radii. Further, it has been the practice in such turbines to keep the clearance between stator blades and rotor blades 20 very small, with a view to inducing the fluid to pass from one to the other without escaping outside the periphery of the rotor, and for the same reason to make the rotor blades radially longer than their co- 25 operating nozzle blades on the stator. In the case of impulse turbines the acceptance of the assumption already mentioned has led to the further assumption that the fluid pressure is very little 30 different on the two sides of the rotor wheel and no special effort has been made to guard against tip leakage. Also in impulse turbines the assumption of sub- stantially equal pressure on the two sides 35 of the rotor wheel has materially affected the design of thrust bearings, and in many cases has been responsible for the provision of pressure equalising holes through the wheel. From theoretical con- 40 siderations I have found that this assumption of straight line flow from the nozzle ring is unjustified, and this has been proved to be the case in fact by experiment. It has been found that even from 45 a nozzle ring with which there is not associated any bounding cylinder to constrain the radial flow of the fluid emerging from the nozzle ring, the flow is in fact substantially rotational, and does not 50 show the hitherto assumed tendency to flow tangentially, with the result that the velocity and the pressure of the fluid flowing from the nozzle ring vary in manner complementary each to the other as the 55 radial distance from the axis varies. Thus, even when no special provision in the formation of the blades is made to produce this result, the characteristics of the flow from a stator nozzle ring of a 60 turbine, or the discharge from the rotor of an axial flow compressor, tend to approximate to the characteristics of a vortex. That is to say, even though the blading is unsuited to such an effect, a 65 tendency exists for the product of the linear rotational velocity of the fluid and the radius to be uniform. Wherever in this specification a quantity is said to be uniform, it is to be 70 understood thereby that the magnitude of the said quantity is the same at all points along a radius within the path of flow of the operative fluid. Some of the consequences of a failure 75 to appreciate the true nature of the flow of a compressible and viscous fluid from the stator nozzle ring, in, for example, an impulse turbine may be summarised as follows:— 80 (a) It has not been appreciated that on entering the rotor blades the pressure of the fluid is in fact substantially higher near the tips than it is near the roots of the blades, and, as the result of this, the 85 assumed rotational velocity of the fluid at the tips has been too high, and, at the roots, has been too low. The result of this has been that a degree of reaction existed at the tips of the blades for which no 90 provision was made in the blade formation hitherto utilised. Thus, with the blade formations hitherto employed, the fluid has struck the back of the blades near the tips thereof, and, near the roots, has 95 struck the blades with a substantially larger angle of attack than that for which the said blades were designed. At the tips, therefore, of impulse turbine blades there has been a substantial leakage of 100 fluid unless other measures were taken to prevent it. (b) The assumed pressure between the rotor wheel disc and the stator nozzle diaphragm has been higher than was the 105 pressure which in fact existed, and this, in the case of an imperforate wheel disc, has produced a thrust towards the diaphragm, to meet which no provision was made, or, in the case of a perforated 110 disc, has produced a wasteful flow of fluid from the exhaust side to the nozzle side of the wheel. (c) Either the pressure at the blade roots on the nozzle side has been lower than that 115 of the exhaust, in which case there has been a root recompression, for which no provision was made, or, what is more probable, the exhaust pressure has been substantially lower than it might have 120 been for a given output owing to a failure to provide for the necessary recompression. This is probably the most serious loss, hitherto unrealised, which occurs in impulse turbines. 125 Similar consequences in the case of other types of turbines of the kind specified will be readily appreciated. In the case of axial flow compressors considerable importance has in some cases 130 of late been attached to the correct aero- | ||