US3010642A - Radial flow supersonic compressor - Google Patents

Description

Nov. 28, 1961 J. F. E. DICKMANN ETAL 3,010,642

RADIAL FLOW SUPERSONIC COMPRESSOR Filed Feb. 15, 1956 3 Sheets-Sheet 1 Nov. 28, 1961 J. F. E. DICKMANN EI'AL 3,010,642

RADIAL FLOW supsasouxc COMPRESSOR 3 Sheets-Sheet 2 Filed Feb. 15, 1956 1c MAC/l. A a

Nov. 28, 1961 J. F. E. DICKMANN ETAL 3,010,642

RADIAL 110w SUPERSONIC COMPRESSOR Filed Feb. 15, 1956 3 Sheets-Sheet 3 FIG. 8

F16. 1f F16. 12

IN vslvfozS United States Patent 3,010,642 RADIAL FLOW SUPERSONIC COMPRESSOR Johannes Franz Eduard Dickmann and Hans Joachim Klaue, Karlsruhe, and Theo Heim, Frankenthal, Germany, assignors to Rheinische Maschinen und Apparate G.m.b.H., Mannheim, Germany Filed Feb. 15, 1956, Ser. No. 565,742 Claims priority, application Germany Feb. 16, 1955 4 Claims. (Cl. 230-127) The present invention relates to a radial flow supersonic compressor, to a method of compressing a fluid, and more particularly to a compressor conveying a compressible fluid, such as a gaseous medium or steam, in a direction of flow which has a radially and outwardly directed component.

Compressors have the purpose to transport a medium through the machine and at the same time to increase the static pressure of the medium. Generally, the amount of energy required for increasing the pressure of the medium is considerably greater than the energy required for transporting the medium. The energy produced in the compressor can be transferred in two manners from the rotor to the fluid medium:

(1) The energy may be transferred as centrifugal energy which produces a direct increase of the static pressure and thereby of the enthalpy of the conveyed medium.

(2) The energy may be transferred as kinetic energy which produces at first only an acceleration of the conveyed fluid medium.

Compressor rotors in which the fluid moves in axial direction transfer the energy predominantly in the form of kinetic energy, whereas rotors conveying the fluid in a stream having a radial component, utilize both kinds of energy transfer.

-With the exception of the amount of energy required for transporting the fluid medium, the kinetic energy is also transformed into pressure. Consequently, a reducing of the kinetic energy results in an increase of the enthalpy. In accordance with the construction of the machine, this is done by delaying the stream between the rotor vanes and a downstream arranged stator. Frequently, the rotor and the stator are used for obtaining this result.

An increase of the efficiency of the compressors according to the known art is limited by the following facts which are characteristic for the compressors according to the known art.

(1) The speed of of the flowing fluid approaches the speed of sound, resulting in local compression shocks and other disturbances causing losses.

(2) Limits of channel cross sections, or as vane angles are reached at which the flowing stream is interrupted.

(3) The gap losses increase between rotor inlet and rotor outlet.

It has been proposed to exceed the above mentioned limits at least partly in such manner that the conveyed medium is accelerated to a speed above the speed of sound for producing a considerable part of the total pressure increase by compression shocks. Compression shocks result in losses of useful energy, because they are connected with a more or less pronounced increase of the entropy, but they permit a reducing of the length of the friction faces of the walls, and the friction losses are under certain circumstances reduced as compared with compressors operating at speeds below the speed of sound.

Supersonic compressors are mainly used in connection with airplane engines and consequently have been built as axial compressors. However, axial compressors are not able to produce a useful centrifugal energy in the conveyed medium.

It is one object of the present invention to overcome the disadvantages 'of the compressors according to the 3,010,642 Patented Nov. 28, 1961 known art, and to provide a compressor capable of transferring to a conveyed fluid much greater amounts of energy as has been possible with compressors according to the known art of the same size.

It is another object of the present invention to use energy produced in a fluid by centrifugal action for accelerating the fluid to supersonic speed.

' It is another object of the present invention to use energy produced in a fluid by centrifugal action for accelerating the fluid to a supersonic speed of such magnitude that the meridian component of the flowing fluid exceeds the speed of sound.

It is another object of the present invention to use static pressure produced in a fluid by centrifugal action for accelerating the fluid to supersonic speed.

It is another object of the present invention to use energy produced in a fluid by centrifugal action during passage through a rotor for accelerating the fluid within the rotor to supersonic speed.

It is also an object of the present invention to obtain the last mentioned objects without accelerating the fluid in the rotor by external means such as a vacuum pump arranged downstream of the rotor.

It is a further object of the present invention to produce by a meridian supersonic component of the flowing fluid a stable annular compression shock in an annular space surrounding the rotor.

It is a further object of the present invention to eliminate the necessity of providing downstream arranged stator vanes.

It is a further object of the present invention to produce a stable compression shock by continuously accelerating a fluid in a converging passage to supersonic speed. a

It is a further object of the present invention to continuously increase the supersonic speed of a fluid flowing from a rotor to a stator of a compressor without any sudden changes in pressure or Mach number.

With these lobjects in view, one aspect of the present invention consists in a rotor means for a rotary compressor for a compressible fluid, the rotor means being formed with passages extending from an inner portion of the rotor means to an outer portion of the rotor means, the passages having cross sections gradually decreasing from the inner portion to the outer portion of the rotor means for increasing in the passages the relative speed of the fluid with respect to the rotor means to supersonic speed during rotation of the rotor means.

The present invention also consists in a method of producing a stable compression shock in a compressible fluid, the method comprising the steps of guiding the fluid through passages of gradually decreasing cross section, and continuously increasing by centrifugal acceleration the speed of the fluid to supersonic speed.

According to another aspect of the present invention, a rotary compressor comprises stator means including a pair of annular walls having opposite faces defining an annular channel, and the fluid discharged from the rotor means passes into the annular channel of the stator means. The pressure of the fluid moving through the annular channel in the stator means is gradually decreased. According to the present invention, the pressure of the fluid is slightly decreased in the rotor means before entering the channel of the stator means so that the pressure gradually and continuously changes as the fluid moves out of the rotor and into the stator. By this arrangement, a smooth and continuous increase of the Mach number of the speed of the fluid is obtained.

In the arrangement of the present invention, it is possible Without any difliculty, to increase the speed of the fluid to such extent that even the meridian component of the flowing fluid reaches the speed of sound. Consequently, very simple compression shock arrangements can be obtained.

The fluid in the compressor according to the present lar channel in the stator, an annular compression shock" which is coaxial with the axis of the rotor. Due to this ariaagemestir is possible to eliminate shock producing stator "varies. A'change in the operating conditions prodices only a displacement of the annular shock downstreamer upstream. In the event that the annular compression shock, and the pressure'increase connected therewith, takes place in the stator, the gap losses are low, and the gap widtli between rotorand stator can be made somewhat larger. 7

' Injth rotor according to the present invention in which fluid flows in radial direction, or at least in partly radial direction, the absolutespeed as well as the relative speed of the conveyed fluid are increased within the rotor from speeds below the speed of sound to supersonic speeds. It is necessary to increase the speed of the flowing fluid to such extent that the meridian component of the fluid reaches supersonic speed if .a coaxial annular compression shock is to be produced.

In order to increase the relative speed of the flowing fluid with respect to the rotor passages, and also the meridian component speeds in the rotor from speeds below speeds of sound to supersonic speed, a particular shape of the rotor passages is necessary.

The passages of a rotor according to the, present invention may be compared with a Laval nozzle. However, it must be understood that in the rotor of the present invention the total energy of the conveyed fluid increases in the. relative flow, whereas in the Laval nozzle only a transformation of the existent potential, energy takes p e- The present invention will be better understood, when compared with a Laval nozzle. A Laval nozzle is arranged between two spaces between which a minimum critical pressure diflerential exists. This potential energy is used for accelerating the fluids passing through the Laval nozzle from one space to the other space. The variation of the cross sections of the Laval nozzle is es- (1 The local speed of flow. An increase of the speed of flow at constant density necessitates a reduced cross section ascompared with an inlet cross section.

(2) The volume increase of the flowing fluid which is caused by.a reduction of the density due to the acceloration and necessitates an increase of the cross section.

'As long as the speed of the expanding fluid medium is'below the speed of sound, the first factor prevails. It is then necessary to reduce the cross section of the Laval nozzle in order to produce an increase of speed. However, as the speed of sound is approached, the influence of the second factor increases. In the moment in which the speed of sound is reached, the two factors have the same influence, and consequently the cross section must be constant at this place. If it is desired to exceed the speed of sound, the second factor must be more considered than the first factor, in other words the cross section of the passage must widen. At the point at which speed of the flowing fluid exceeds the speed i of sound, the narrowest cross section must be located. The 'arrangement according to, the Laval nozzle is the only way bywhic'h a flowing mediumcan be accelerated to the speed of sound by transforming its own potential sn r y- The arrangem ent of the present invention according to which a rotor 'isprovided for accelerating the fluid to supersonic speed, distinguishes inasmuch over the Laval 4 nozzle as the fluid medium is provided with additional mechanic energy during its acceleration in the radial passages, or the radial passages of the rotor which are formed by the motor vanes. The energy thus added, is used for producing the supersonic speed so that the fluid medium need not have potential energy for reaching supersonic speed.

This difierence in the operation and function results in a completely different shape of the passages through which the fluid flows in accordance with the present invention as compared with the cross sections of the Laval sential, and the" dimensions of the cross sections are de- V termined by two opposing factors.

nozzle. 7

Particularly, the constricted narrowest cross section which is essential in the Laval nozzle, is rather unimportant in the arrangement of the present invention, and in many cases need not or must not be provided. In the event that sufficient mechanic energy is provided for the fluid medium, an expansion of the medium does not necessarily take place, on the contrary during the acceleration a compression may take place, and consequently the reason for a narrowest constricted cross section is obviated.

The novel features which are considered as characteristic for the invention are set forth in particular in the appended claims. The invention itself, however, both as to its construction and its method of operation, tw gether with additional objects and advantages thereof, will be best understood from the following description of specific embodiments when read in connection with the accompanying drawings, in which:

FIGS. 1-4 show schematically cross sectional views of passages for a flowing medium, a graph showing the increase of the Mach number along the length of the passage, and another graph showing the variation of the pressure of the flowing medium along the length of the passage;

FIG. 5 shows a passage for a flowing medium, and a graph illustrating the distribution of the pressure along the length of the passage under certain operational conditions; V

FIG. 6 is a fragmentary meridian sectional view illustrating the outermost portion of a rotor passage and the adjacent stator channel according to the present invention;

FIG. 6a is a diagram illustratingthe variation of the static pressure, the. variation of the Mach number of the speed of the flowing fluid, and the variation of the cross section of the passages in rotor and stator along the radial extension of such passages;

H68. 7 and 7a are views corresponding to FIGS. 6 and 6a respectively and illustrate Jan arrangement according to a preferred embodiment of the presentinvention;

FIG. 8 is an axial sectional View of an embodiment of the present invention;

FIG. 9 is an end view of the rotor and. the stator of the embodiment shown in FIG. 8;

FIGS. 10, 11 and 12 are vector diagrams illustrating the distribution of the velocities of the "fluid medium at various cross sections of the rotor passages;

FIG. 13 is a cross-sectional view taken on line 13-13 in FIG. 8; and

FIG. 14 is a cross-sectional view taken on line 14-14 in FIG. 8.

Referring now to the drawings, FIGS. 1-4 schematically illustrate passages N N- in'which a fluid is accelerated from a speed helow'the speed of sound to supersonic speed. The fluid is assumed to flow from the left to the right as indicated by the arrows, and the rate of acceleration is assumed to be the same in all four examples. The amount of outer mechanical energy which is introduced during the acceleration of the liquid is different in the four examples illustrated in FIGS; 1-4;v

In the arrangement of FIG; l, it is assumed that no outer energy is added to the flowing medium, and-that a pressure difierential produces the acceleration of the fluid. Consequently, the arrangement of FIG. 1 corresponds to a conventional Laval nozzle. The entire required kinetic energy must be provided by the pressure differential between nozzle inlet and nozzle outlet as illustrated by the graph P in FIG. 1. The cross section is gradually reduced to a narrowest constricted portion and is then again increased. As shown in the graph M, the Mach number is increased according to linear function during the passage of the fluid through the Laval nozzle N and exceeds in the constricted portion the value 1, in other words the speed of sound reached at the constricted portion indicated by a broken line.

In the arrangement of FIG. 2, it is assumed that in addition to the pressure differential, the flow of fluid is accelerated by a small amount of mechanical energy which is added during passage through the passage N The passage N may be provided in a rotor and arranged in radial or partly radial direction, and the rotor may be rotated at a moderate speed. Consequently centrifugal force is produced, and the energy of the centrifugal force added to the energy of the fluid. The required kinetic energy can be partly covered by the added mechanical energy produced by centrifugal force. Consequently the pressure differential between nozzle inlet and nozzle outlet can be smaller as compared with the arrangement of FIG. 1. This corresponds to a smaller enthalpy diflerence of the conveyed fluid. As shown in the graph P in FIG. 2, the pressure drops in the passage between the passage inlet and the passage outlet. However, the pressure drop is much much smaller than in the Laval nozzle according to FIG. 1. In the conventional classification of compressor rotors, the rotor can be classified as a rotor having a negative reaction degree assuming an adiabatic flow. The decrease of the cross section of the first part of the passage must be more rapid as is the case in the arrangement of FIG. 1. It will be noted that the narrowest constricted cross section is now located in the supersonic range. The Mach number exceeds the value 1 already before the fluid passes the narrowest cross section indicated by a broken line. In the arrangement of FIG. 2, the passage widens again downstream of the narrowest cross section while the Mach number further increases and the pressure drops.

In the arrangement illustrated in FIG. 3, the amount of added mechanical energy is further increased to such extent, that the required kinetic energy is provided by the added mechanical energy. Consequently, no pressure differential is required for accelerating the fluid in the passage N Such increased mechanical energy is produced by the rotor rotating at a higher speed as compared with the rotor speed used in the arrangement of FIG. 2. Since the fluid has no potential energy, the pressure in the passage does no drop as is the case in the arrangement of FIGS. 1 and 2, but is constant. The arrangement corresponds to a constant pressure rotor having the reaction degree 0. In this arrangement an even more pronounced cross section reduction is required as compared with the arrangement of FIG. 2. A narrowest constricted cross section is not provided in the passage, but is in theory reached at an infinite distance.

FIG. 4 illustrates an arrangement in which a great amount of mechanical energy is added to the fluid during passage through the passage N; which corresponds to a very high rotary speed of the rotor in which the passage N, is provided. The amount of added mechanical energy exceeds the amount required for accelerating the fluid passing through the passage. Consequently not only the Mach number is increased, but also the static pressure and the enthalpy are increased. The reaction degree is positive in the arrangement of FIG. 4 and corresponds to an over-pressure rotor. In this case, the passage cross section must be rapidly reduced and asymptotically ap proach value. A narrowest constricted portion does not exist. In accordance with the present invention, certain mathematical conditions are. necessary for obtaining in a rotor having radial or partly radial passages, the desired supersonic speeds of the fluids flowing through the passages. The cross sections of the passages must follow a certain law, and the equation governing such design of the cross section will now be developed. The following characters and symbols will be used in the equations:

FChannel cross section wRelative speed of the fluid that is the speed with respect to the rotor u-Tangential speed at the respective radius aLocal speed of sound MMach number of the relative speed according to the equation M =w/a v M Mach number of the tangential speed according to the equation M =u/ a TAbsolute temperature rRadius of the rotor at the respective cross section S Density of the conveyed fluid medium gAcceleration due to gravity kAd.iabatic exponent of the fluid medium, which is the ratio of the specific heat of the fluid at constant pressure with respect to the specific heat at constant volume according to the equation k c /c RGas constant As pointed out above F indicates a given cross section which may be assumed, and w. is the relative speed of the fluid at the cross section F The quotient of the relative speeds w and w'can be assumed, for instance by assuming a desired increase or decrease of the flow speed of the fluid in the passage. The expression however, depends on the required relative speed w and on the Mach number M respectively, as well as from the amount of added energy which corresponds to the tangential speed u at the respective cross section F:

It would be possible to assume also the density and the pressure, respectively, in which event the relative speed w and the Mach number M respectively, could be obtained from an equation similar to the Equation 2. The same is true if a cross section of the passage F is selected. However, for constructive reasons it is advantageous to work with the Equation 2 since in the event a rotor operating under stable conditions can be designed as will be explained hereinafter. The Equation 2 is obtained by equating the pressure members in Eulers equation with the thermo dynamic adiabatic equation as follows:

is derived an expression for the speed of sound a:

a sure increase within the rotor.

With the above equations the following equation is obtained:

The Equation 5 indicates for any place in the rotor the necessary cross section F depending on the Mach number M which is desired at the respective cross section and also depending on the energy available at this place, such energy resulting from the tangential speed Mach number M and irom the radius r at the respective cross section F. All other values are given limits, such as F M M and r The character M corresponds to the Mach number of the relative speed at the cross section F which is given, and character M corresponds to the Mach number of tangential speed at the cross section F The adiabatic exponent k is known and is assumed to be constant during the passage of the fluid through the rotor passages.

The Equation 5 has general validity. It comprises the speed range below supersonic speed, the passage through the speed of sound, and supersonic range. The Equation 5 makes possible the designing'of a rotor for a compressor in which the relative speed between the flowing fluid and the rotor vanes is increased from speeds below the speed of sound to supersonic speed and does not require the meridian component of the speed of the fluid to exceed the speed of sound. V

The Equation 5 is valid also for an axial flow compressor rotor. However, since the radius in the direction of flow is constant corresponding to r=r.,,. the member corresponding to the energy becomes 0, as follows:

ter- 1 The equation for designing a Laval nozzle remains. An axial flow rotor consequently corresponds to the arrangementshown in FIG. 1. In this arrangement an outer pressure differential is requiredbetween rotor inlet and outlet. A compressor rotor in which the fluid flows in va radial flow and which is designed according. to Equation 5 can be used for producing supersonic streams of fluid in different manners... Some constructions have been found particularly advantageous.

In designing the shape of; the rotor passages, constructions are preferred in which the bounding Walls of the passages have a distance from each other which decreases in outward direction as seen in a meridian section. In accordance with the present invention, a particularly high density of the energy in the machine is preferably obtained by decreasing in outward directions the cross sections of the passages in planes perpendicular to the flow of fluid.

Such construction has advantages as regards the stresses on the rotor and the corresponding elements produced by the high centrifugal force. The reducing of the mass of the varies in outward direction have a favorable influence on the produced stressing forces.

Constructions which also consider possible energy losses have particular advantages. The supersonic flows discussed above refer to isentropic flows, but it is possible that non-isentropic flows may occur in the motor. In this event a particular condition must be maintained if a stable operation of the rotor is to be assured. Nonisentropic flows occur particularly in the event that within the stream of the fluid compressiqn shocks occur. Such compression shocks. may be desired, and may be for instance a system of shock wavesrotating with the rotor and provided for the purpose of producing a sudden pres- On the other hand the compression shocks may be non-stationary streams which enter the rotor passage in upstream direction. Such operational conditions can be covered by the additional rewithin the rotor passages. The constructive fulfillment of this condition is an essential part of the present invention.

The arrangement will be best understood by explaining the known conditions in the Laval nozzle as shown in FIG. 5. FIG. 5 shows a Laval nozzle N It is assumed in FIG. 5 that in the divergent right portion of the Laval nozzle, a normal compression shock has'been produced by a corresponding counter-pressure at the outlet of the Laval nozzle. Such compression shock is indicated by a solid vertical line s. The'pressure variation is shown below the cross section of the Laval nozzle in FIG. 5 in the graph T, as is clearly shown in the graph, a sudden pressure increase takes place due to the compression shock s. The pressure increase results in a corresponding volume decrease. The volume decrease is compensated by a speed decrease which results in the reduction of the Supersonic speed before the compression 7 shock to a speed below the speed of sound downstream of the compression shock. The continuous flow of the fluid is maintained, since the cross sections are filled with a fluid due to the reduction of the speed. The com: pression shock is consequently a real shock. A dist urbance of the flow of fluid, which may be caused by a compression shock running from the right in FIG. 5 through the nozzle outlet against the stream of fluid, displaces the front of the normal compression shock in upstream direction for a short distance as indicated by the broken line s in FIG. 5. Thereby, the thermo dynamic condition of the fluid medium prevailing after the compression shock are changed in such manner that the condition of continuity of flow is again achieved. After the disturbance has ended, that is after the originally prevailing pressure at the nozzle outlet is. again established, the front of the compression shock is again displaced and assumes its original, position indicated by reference character s. A corresponding reaction is produced by a disturbance running in the opposite direction in the form of a underpressure wave as indicated by the dash and dot line s The type of compression shock shown in FIG. 5 is considered a stable shock. Such stable shocks are only possible in the divergent portion of a Laval nozzle unless energy is added to the fluid during passage to the nozzle.

By examining flow conditions Without the. addition of outer energy, it can be. shown that a normal compression shock is only stable in a divergent passage in which the pressure and the density decrease and the supersonic speed increases. A shock is always unstable in a convergent passage during pressure and density increase and during decrease of the supersonic speed. In the event that no energy is added, three interrelated conditions have to be considered for producing a stable perpendicular shock:

(I) A divergent passage.

(2) Decrease of pressure.

(3) Increase of the Mach number.

Extensive tests have proven that these interrelated conditions for a stable shock and thereby for the high resistance against disturbances are not true in the event that mechanical energy is added to the fluid flowing in the passage. Consequently, a normal: shock can be stable in a convergent channel in the event that mechanical energy is added to a fluid.

In contrast to the Laval nozzle it is immaterial during such conditions whether the static pressure is increased or decreased. The present invention is based on the fact that the only condition for the possibility of a stable shock is a continuous increase of the flow speed, and in other words of the Machnumber of the isentropic 1 supersonic flow. Only in this manner it is possible to quirement that the rotor passage should be constructed in such a manner that a stable compression shock is possible obtain a stable shock in the rotor which produces a meridian component at supersonic speed, or respectively produces a supersonic relative speed of the flowing fluid in the rotor passages defined by the rotor vanes. The teaching of the present invention to increase the Mach number and thereby the speed of the fluid within the rotor in a continuous and monotonous manner (the rate of change of the Mach number along the path of the flow always has to be positive) is contrary to the constructions of the know compressor rotors, since in such known compressor rotors the fluid medium is delayed or retarded within the rotor in order to increase the pressure.

A rotor in which the Mach numbers of the fluids in direction of flow are continuously increased, is capable of absorbing purposely produced or undesired disturbances or shocks without a collapse of the supersonic flow. For example, if in the rotor a system of normal compression shocks is maintained which rotates with the rotor, the shock fronts will contract or expand a little under the influence of disturbances from the outside relative to the axis of the rotor. The rotor works not only stable, but moreover may be regulated within a wide range as regards the pressure to be produced by the entire compressor.

If it is desired to produce the compression shocks not in the rotor itself, but in a stationary stator arranged downstream of the rotor, for instance in the form of a coaxial annular compression shock within an annular stator channel which is free of vanes, the consequences of the uneven flow of the supersonic stream out of the rotor can be substantially reduced in the rotor according to the present invention. Such irregular flow out of the rotor cannot be entirely avoided due to the fact that the finite number of vanes having a minimum thicknws must be provided.

The requirement of a continuous increase of the speed of the isentropic flow, which is caused by the necessary stability of the flow, still permits the choice of the way in which the Mach number is to increase within the rotor. For example, the Mach number may increase according to a linear function along the length of the passage, or according to a parabolic function.

Such liberty in choosing the constructive details in accordance with the desired variation of the Mach number in the rotor, may be used for fulfilling the requirement to diminish the unfavorable results of the uneven flowing out of the fluid from the rotor. A construction according to this requirement, is a further essential part of the present invention. The flow in the stator is determined by its construction, as well as by the condition and direction of flow of the fluid medium at the time the fluid medium is discharged from the rotor. Such a flow takes place without further addition of energy, for instance at a pressure decrease during further acceleration in the manner of a divergent part of a Laval nozzle. In order to reduce the disturbances at the points at which the fluid is discharged from the rotor, it is advantageous to provide for a flow within the rotor in the region of the fluid discharge which accomplishes a smooth transition of the absolute flow within the rotor into the absolute flow within the stator. The expression smooth is used in a certain sense to indicate that in the graphic illustration the variation of the absolute Mach number and the variation of the pressure at the point of discharge from the rotor is continuous or substantially continuous without any sudden changes, which would cause a sudden change of direction in the respective graph. For obtaining this result, it may be advantageous under certain circumstances to increase the Mach number shortly before the discharge of the fluid from the rotor to such extent, that potential energy is taken from the fluid medium. Under such conditions, a pressure drop takes place within the rotor towards the discharge point of the fluid, and if the rotor and stator are accordingly designed, the pressure drop at the discharge points of the rotor continuously emerges into the pressure drop within the stator. Consequently, there is no sudden change of pressure at the point at which the fluid passes from the rotor into the stator.

FIG. 6 illustrates in the meridian section, an outwardly located portion of the rotor 8 and an annular channel 10 of the stator. FIG. 6a shows in three graphs the static pressure, the cross sections and the Mach number of the absolute speed of the fluid along the radial extension of the elements shown in FIG. 6. In the construction according to the present invention shown in FIGS. 6 and 6a, a transition free of disturbances from rotor flow to stator flow is desired. Since the flow in the annular channel 10 of the stator takes place without any addition of energy, the flow conditions in channel 10 are determined by the condition of the fluid medium at the time of entering the channel 10, on the angle of the flow, on the Mach number at the time of entry, and on the shape of the walls bounding the annular channel 10. If the meridian component of the flowing fluid exceeds supersonic speed in accordance with the present invention, the fluid medium expands in the annular stator channel 10 and is simultaneously accelerated. In accordance with the present invention the region of the fluid discharge of the rotor is to be constructed for a desired pressure decrease.

The horizontal broken line in FIG. 6a indicates the point at which the rotor flow changes into the stator flow in the channel 10. In the region of the channel 10, the pressure decrease and the Mach number increase are clearly determined. However, it is possible to influence the stream of fluid within the rotor within the framework of the previously explained conditions, and such variation can be carried out by suitably choosing the variation of the change of the Mach number.

As shown in FIG. 6a, the variation of the Mach number in the rotor is so chosen that the static pressure is continuously and gradually changed from a pressure increase to a pressure drop. The thus accomplished pressure drop has an extent corresponding to the already determined pressure drop in the stator, so that the graph showing the pressure is continuous and does not have a step or sudden change in the region of the broken line which indicates the transition from the rotor to the stator. In order to obtain such decrease of the pressure within the rotor directly before the discharge of the fluid, it is necessary to reduce the cross section reduction in the region of the discharge from the rotor as is indicated by the curvature of the graph illustrating the cross sections in FIG. 6a. 7

Referring again to FIG. 2, it will be understood that a pressure reduction necessitates a narrowest cross section of the rotor passage, such narrowest cross section is located outside of the rotor in the arrangement shown in FIGS. 6 and 6a, as indicated by the broken line extension of the graph illustrating a variation of the cross section. In the arrangement shown in FIGS. 6 and 6a, the fluid is discharged from the rotor before reaching the narrowest cross section of the passage," and consequently the fluid does not reach such narrowest cross section but enters the channel 10 in which the cross section is increased in a linear function. The linear increase of the cross section of the stator channel 10 is caused by the fact that the stator faces are parallel and circular. The arrangement illustrated in FIG. 6 and explained with reference to FIG. 6a, obtains a considerable reduction of the unavoidable losses during the transition of the fluid from the rotor to stator, and is an important aspect of the present invention.

According to a further aspect of the present invention, the smooth transition from rotor to stator is not only obtained by the construction of the rotor passages as explained with reference to FIGS. 6 and 6a, but also by suitably constructing the channel of the stator. V

In accordance with a preferred embodiment of the present invention illustrated in FIGS. 7 and 7a, the opposite faces of the channel 10 are not parallel as in FIG; 6, but converge in outward direction. Consequently a "substantially continuous variation of the cross sections of the passage for the fluid is obtained. It will be noted that the graph in FIG. 6a illustrating the cross section variation has a sharp bend in theregion of the broken line where the transition from rotor to stator takes place. In contrast thereto the transition in the arrangement of FIG. 7a' is smooth. However, it should be noted that the cross section is gradually reduced within the rotor, whereas it is increased within the stator channel. Such increase is due to the fact that the channel expands in layers of the fluid. In practical constructions it is necessary to make the cross sections of the passages somewhat greater than in accordance with theoretical conditions derived from the adiabatic laws and the equations derived therefrom, in order to considerthe cross section reducing effect of the boundary layer of the fluid whose thickness increases in outward radial direction. Also, the minimum thickness of the vanes has to be considered. The correcting factors, and the thickness of the boundary layer of the fluid along the walls of the passage can be computed by known equations. In a similar manner, the grid effect can be compensated by introducing a corresponding factor into the equations.

In considering the effects of the boundary layers of the fluid, another phenomenon must be considered, namely, the separation of the boundary layer whose detrimental results are mitigated or completely overcome by the construction according to the present invention. A peeling-01f of the boundary layer caused by a compression shock, will not have the same detrimental eflect in a radial compressor in accordance with the present invention as in axial compressors. Due to the fact that the. width of the passages, considered in the meridian section, is not very great in the rotor and stator according to the present invention, the particles of the boundary layers which may peel ofl" will accumulate behind the compression-shock on the channel walls and produce a noticeable reduction of the free cross section available for the fluid flow. .Since the radial component of the flowing fluid has a speed smaller thanthe speed of sound after a shock, the narrowing of the cross section caused by displaced particles of the boundary layers will again result ina local-acceleration of the flowing fluid. Therebythe particles of the boundary layer are again pressed against the wall surfaces and rapidly driven'away in the direction of the flowing fluid so that they cannot have a detrimental influence on the flow of fluid. The displaced particles of the boundary layer cannot travel in a direction, transverse to the stream of the flowing fluid, as is the case in axial compressors, because in radial compressors there is no pressure gradient in the plane of the cross sections.

Referring; now to FIGS. 8 and 9 which illustrate an embodiment of the present invention, a stator portion 4 is provided for guiding the fluid in axial direction into the rotor passages 91. A set of stator vanes is provided in the stator portion. 4, which is. capable of reducing the Mach number of the fluid at the points at which. the fluid enters the rotor 8. Due to. the high rotary speed of the rotor, otherwise the danger is present that the speed of sound is exceeded at an undesired place.

The curved fluid guiding surface of the stator has a 'first surface portion 41, and a second surface portion 42.

The. fluid guiding curved rotor surface 71 is located opposite the stator surface portion 42. Rotor vanes 9 project from the rotor surface 71 toward the stator surface portion 42 so that the surfaces 42, 71 and the rotor vanes 9 define rotor passages. The stator 4 further includes a stator portion 43 which is located in the axis of the machine and has another stator surface 44 which is a continuation of the rotor surface 71. Consequently the surfaces 41 and 44 cooperate to gradually deflect the fluid from axial flow into a partly radial flow which is then further deflected into a radial flow in the rotor passages 91. Due to the fact that in accordance with the present invention, the fluid is already partly deflected in the stator portion 4, an extreme curvature of the rotor passages is avoided'in the region in which the fluid achieves the speed of sound whereby disturbances are prevented.

The shaft 6 carries the rotor and drives the same. The stator portion 7 houses the rotor, and it will be noted that no cover wall is provided on the rotor opposite the stator surface 42. The rotor vanes 9 extend in radial direction as bestseen in FIG. 9 and are slightly bent in the region of the rotor inlet. The cross sections of the rotor passages 91 which are bounded by the rotor surface 71, by the stator surface 42 and by the faces of the vanes 9 are designed according with the previously discussed equation. The requirements for obtaining a continuous'increase of the Mach number and the conditions explained with reference to FIGS. 7 and 7a are considered in the construction of the rotor passages 91.

FIGS. 10, l1 and 12 are diagrams illustrating the speed conditions in the cross sections 1, 2, and 3, respectively. The characters applied in FIGS. 10, 11 and 12 have the following significance:

uTangential speed,

wRelative speed,

w Meridian component of the speed, c-Absolute speed,

aLocal speed of sound.

The indices 1, 2 and 3 refer, respectively, to the cross sections 1, 2 and 3.

FIG. 10 shows that at the inlet cross sections 1 of the ,rotor passages 91, allqspeeds are belowthe speed of sound.

In the cross section 2, the relative speed is equal to the local speed of sound a. Since the stream of the flowing fluid has passed beyond the curved inlet portions 92 of the rotor vanes 9, and has entered the portions of the passages 91 which are separated from each other by wall portions located in planes passing through the rotor axis, the meridian speed component w' is equal to the relative speed w. The absolute speed c has exceeded the speed of sound before the fluid passes through the cross section 2. In the cross sections 3 of the rotor passages, all speeds and'speed components are supersonic. An annular stator channel 10 surrounds the rotor 8, and it will be noted that the stator surface 42 merges substantially into one face of the stator channel 10, whereas the rotor surface 71 merges substantially into the other face of the stator channel 10. The results of this arrangement' have been described with reference to FIGS. 6 and 7. The fluid discharged through cross sections 3 of the rotor passages 91 enters the stator channel 10 and expands in the same under further acceleration without any energy being added in the stator channel 10. The pressure of the fluid in the stator channel 10 is substantially increased due to a coaxial compression shock, and the pressure is further increased in the spiralshaped tubular guide means 11 of the stator.

It will be understood that the equations given, for the necessary cross sections of the rotor passages according to the present invention are valid for other rotors different from the rotor illustrated in FIG. 8. For instance the rotor vanes may be bentforwardly or rearwardly. It is also of no consequence for the validity of the equations for dimensioning the rotor passages, whether only an 13 isentropic flow passes through the entire compressor, or whether one or several compression shocks are provided in the stator or in the rotor. It is only necessary to suitably choose the limits designated by the index e in the preceding discussion of the theoretical conditions.

Under certain circumstances, it is advantageous, although not absolutely necessary, to arrange other compressors in series with the compressor according to the present invention. Such compressors may be constructed in accordance with the present invention, or may be different compressors, md arranged upstream and downstream of the compressor according to the present invention.

A compressor according to the present invention need not have a strictly radial flow of the fluid, but may be constructed in such a manner that the fluid flows only in partly radial directions. It is however necessary that the flowing medium has a component in radial direction within the rotor.

It will be understood that each of the elements described above, or two or more together, may also find a useful application in other types of compressors diifering from the types described above.

While the invention has been illustrated and described as embodied in a rotary compressor including rotor means formed with at least partly radial passages having cross sections gradually decreasing in outward directions for increasing the relative speed of the fluid within the passages to supersonic speeds, it is not intended to be limited to the details shown, since various modifications and structural changes may be made without departing in any way from the spirit of the present invention.

Without further analysis, the foregoing will so fully reveal the gist of the present invention that others can by applying current knowledge readily adapt it for various applications without omitting features that, from the standpoint of prior art, fairly constitute essential characteristics of the generic or specific aspects of this invention and, therefore, such adaptations should and are intended to be comprehended within the meaning and range of equivalence of the following claims.

What is claimed as new and desired to be secured by Letters Patent is:

1. In a rotary compressor for a gaseous medium, in combination, a stator formed with an annular stator space having an inner restricted annular shock-creating portion and an outer annular collector portion, said stator having an inwardly facing annular opening communicating with said inner restricted annular shockcreating portion and having also discharge means communicating with said outer annular collector portion of said annular space; a rotor arranged for rotation co-axially within said stator and having vanes terminating radially inwardly of said inwardly facing annular opening of said stator so as to communicate witth said inner restricted annular shockcreating portion of said annular stator space, said vanes defining substantially radial passages having inlets adjacent the axis and discharge openings at the periphery of the rotor, the cross-sectional areas of the passages decreasing in direction from said inlets to 'said discharge openings and said rotor having a predetermined speed of rotation at which said gaseous medium is discharged through said discharge openings at a velocity the radial component of which exceeds supersonic speed, said gaseous medium thus entering said inner restricted annular shock-creating portion of said annular stator space at a velocity having a radial supersonic speed component whereby an annular shock wave is created due to said radial supersonic speed component in said inner restricted annular shock-creating portion of said annular stator space, said annular shock wave resulting in high compression of said gaseous medium also in said outer annular collector portion of said annular stator space from where the thus highly compressed gaseous medium will be discharged through said discharge means; and a shaft mounted for rotation coaxially with said rotor and supporting the same, said shaft serving as means for rotating said rotor at said predetermined speed of rotation.

2. A rotary compressor as set forth in claim 1 wherein said cross sectional areas of said passages between said vanes decrease toward said discharge openings in such a manner that the Mach numbers of the meridian component of the velocity of the gaseous medium in said passages continuously and gradually increase toward said discharge openings.

3. A rotary compressor as set forth in claim 1 wherein said annular shock-creating portion of said stator space is bounded by annular surfaces approaching each other in outward direction in such a manner that sections of said rotor passages and of said stator opening taken in radial planes passing through the axis of said rotor have outlines merging into each other.

4. A rotary compressor as set forth in claim 1 wherein i i ldY l b wherein F is a cross sectional area to be determined, M is the Mach number of the meridian component of the velocity of the gaseous medium flowing in said passages, the Mach number M for each cross sectional area of each passage being selected so that the Mach numbers M in all cross sectional areas of the respective passage continuously and gradually increase toward said discharge openings of said passages, wherein M is the Mach number of the peripheral velocity at said dis charge openings, said Mach numbers being based on the local speed of sound, wherein r is the radial distance between the rotor axis and the respective cross sectional area F, and k is the adiabatic exponent of the gaseous medium, the index e indicating that the respective values refer to the cross sectional area F References Cited in the file of this patent UNITED STATES PATENTS Germany Aug. 29-, 1942

Images