INTRODUCTION

A fluid machine is a device which converts the energy stored by a fluid into mechanical energy or vice versa . The energy stored by a fluid mass appears in the form of potential, kinetic and intermolecular energy. The mechanical energy, on the other hand, is usually transmitted by a rotating shaft. Machines using liquid (mainly water, for almost all practical purposes) are termed as hydraulic machines. n this chapter we shall discuss, in general, the basic fluid mechanical principle governing the energy transfer in a fluid machine and also a brief description of different kinds of hydraulic machines along with their performances. !iscussion on machines using air or other gases is beyond the scope of the chapter.

CLASSIFICAITONS OF FLUID MACHINES

The fluid machines may be classified under different categories as follows"

Classification Based on Direction of Energy Con ersion!

The device in which the kinetic, potential or intermolecular energy held by the fluid is converted in the form of mechanical energy of a rotating member is known as a turbine. The machines, on the other hand, where the mechanical energy from moving parts is transferred to a fluid to increase its stored energy by increasing either its pressure or velocity are known as pumps, compressors, fans or blowers.

Classification Based on "rinci#le of O#eration

The machines whose functioning depend essentially on the change of volume of a certain amount of fluid within the machine are known as positive displacement machines. The word positive displacement comes from the fact that there is a physical displacement of the boundary of a certain fluid mass as a closed system. This principle is utili#ed in practice by the reciprocating motion of a piston within a cylinder while entrapping a certain amount of fluid in it. Therefore, the word reciprocating is commonly used with the name of the machines of this kind. The machine producing mechanical energy is known as reciprocating engine while the machine developing energy of the fluid from the mechanical energy is known as reciprocating pump or reciprocating compressor. The machines, functioning of which depend basically on the principle of fluid dynamics, are known as rotodynamic machines. They are distinguished from positive displacement machines in requiring relative motion between the fluid and the moving part of the machine. The rotating element of the machine usually consisting of a number of vanes or blades is known as rotor or impeller while the fi$ed part is known as stator. mpeller is the heart of rotodynamic machines, within which a change of angular momentum of fluid occurs imparting torque to the rotating member. %or turbines, the work is done by the fluid on the rotor, while, in case of pump, compressor, fan or blower, the work is done by the rotor on the fluid element. !epending upon the main direction of fluid path in the rotor, the machine is termed as radial flow or axial flow machine . n radial flow machine, the main direction of flow in the rotor is radial while in a$ial flow machine, it is a$ial. %or radial flow turbines, the flow is towards the centre of the rotor, while, for pumps and compressors, the flow is away from the centre. Therefore, radial flow turbines are sometimes referred to as radially inward flow machines and radial flow pumps as radially outward flow machines. &$amples of such machines are the %rancis turbines and the centrifugal pumps or compressors. The e$amples of a$ial flow machines are 'aplan turbines and a$ial flow compressors. f the flow is party radial and partly a$ial, the term mixed-flow machine is used. %igure (.( (a) (b) and (c) are the schematic diagrams of various types of impellers based on the flow direction.

Fig! $!$ Sc%e&atic of different ty#es of i&#ellers

Classification Based on Fl'id Used

The fluid machines use either liquid or gas as the working fluid depending upon the purpose. The machine transferring mechanical energy of rotor to the energy of fluid is termed as a pump when it uses liquid, and is termed as a compressor or a fan or a blower, when it uses gas. The compressor is a machine where the main ob)ective is to increase the static pressure of a gas. Therefore, the mechanical energy held by the fluid is mainly in the form of pressure energy. %ans or blowers, on the other hand, mainly cause a high flow of gas, and hence utili#e the mechanical energy of the rotor to increase mostly the kinetic energy of the fluid. n these machines, the change in static pressure is quite small. %or all practical purposes, liquid used by the turbines producing power is water, and therefore, they are termed as water turbines or hydraulic turbines . Turbines handling gases in practical fields are usually referred to as steam turbine, gas turbine, and air turbine depending upon whether they use steam, gas (the mi$ture of air and products of burnt fuel in air) or air.

ROTOD(NAMIC MACHINES
n this section, we shall discuss the basic principle of rotodynamic machines and the performance of different kinds of those machines. The important element of a rotodynamic machine, in general, is a rotor consisting of a number of vanes or blades. There always e$ists a relative motion between the rotor vanes and the fluid. The fluid has a component of velocity and hence of momentum in a direction tangential to the rotor. *hile flowing through the rotor, tangential velocity and hence the momentum changes. The rate at which this tangential momentum changes corresponds to a tangential force on the rotor. n a turbine, the tangential momentum of the fluid is reduced and therefore work is done by the fluid to the moving rotor. +ut in case of pumps and compressors there is an increase in the tangential momentum of the fluid and therefore work is absorbed by the fluid from the moving rotor.

Basic E)'ation of Energy Transfer in Rotodyna&ic Mac%ines

The basic equation of fluid dynamics relating to energy transfer is same for all rotodynamic machines and is a simple form of , -ewton .s /aws of Motion, applied to a fluid element traversing a rotor. 0ere we shall make use of the momentum theorem as applicable to a fluid element while flowing through fi$ed and moving vanes. %igure (.1 represents diagrammatically a rotor of a generalised fluid machine, with 232 the a$is of rotation and the angular velocity. %luid enters the rotor at (, passes through the rotor by any path and is discharged at 1. The points ( and 1 are at radii and from the centre of the rotor, and the directions of fluid velocities at ( and 1 may be at any arbitrary angles. %or the analysis of energy transfer due to fluid flow in this situation, we assume the following" (a) The flow is steady, that is, the mass flow rate is constant across any section (no storage or depletion of fluid mass in the rotor). (b) The heat and work interactions between the rotor and its surroundings take place at a constant rate. (c) 4elocity is uniform over any area normal to the flow. This means that the velocity vector at any point is representative of the total flow over a finite area. This condition also implies that there is no leakage loss and the entire fluid is undergoing the same process. The velocity at any point may be resolved into three mutually perpendicular components as shown in %ig (.1. The a$ial component of velocity is directed parallel to the a$is of rotation , the radial is

component is directed radially through the a$is to rotation, while the tangential component directed at right angles to the radial direction and along the tangent to the rotor at that part.

The change in magnitude of the a$ial velocity components through the rotor causes a change in the a$ial momentum. This change gives rise to an a$ial force, which must be taken by a thrust bearing to the stationary rotor casing. The change in magnitude of radial velocity causes a change in momentum in radial direction.

Fig $!* Co&#onents of flo+ elocity in a generalised fl'id &ac%ine

0owever, for an a$isymmetric flow, this does not result in any net radial force on the rotor. n case of a non uniform flow distribution over the periphery of the rotor in practice, a change in momentum in radial direction may result in a net radial force which is carried as a )ournal load. The tangential component only has an effect on the angular motion of the rotor. n consideration of the entire fluid body within the rotor as a control volume, we can write from the moment of momentum theorem ((.() where T is the torque e$erted by the rotor on the moving fluid, m is the mass flow rate of fluid through the rotor. The subscripts ( and 1 denote values at inlet and outlet of the rotor respectively. The rate of energy transfer to the fluid is then given by ((.1) where is the angular velocity of the rotor and which represents the linear velocity of the rotor.

Therefore and are the linear velocities of the rotor at points 1 (outlet ) and ( (inlet) respectively (%ig. (.1). The &q, ((.1) is known as &uler.s equation in relation to fluid machines. The &q. ((.1) can be written in terms of head gained .H. by the fluid as

((.5)

n usual convention relating to fluid machines, the head delivered by the fluid to the rotor is considered to be positive and vice3versa. Therefore, &q. ((.5) written with a change in the sign of the right hand side in accordance with the sign convention as

((.6)

Components of Energy Transfer t is worth mentioning in this conte$t that either of the &qs. ((.1) and ((.6) is applicable regardless of changes in density or components of velocity in other directions. Moreover, the shape of the path taken by the fluid in moving from inlet to outlet is of no consequence. The e$pression involves only the inlet and outlet conditions. A rotor, the moving part of a fluid machine, usually consists of a number of vanes or blades mounted on a circular disc. %igure (.5a shows the velocity triangles at the inlet and outlet of a rotor. The inlet and outlet portions of a rotor vane are only shown as a representative of the whole rotor.

Fig $!/ ,aFig $!/ ,.-

,a,.0elocity triangles for a generalised rotor ane Centrif'gal effect in a flo+ of fl'id +it% rotation is the

4ector diagrams of velocities at inlet and outlet correspond to two velocity triangles, where velocity of fluid relative to the rotor and

are the angles made by the directions of the absolute and are the angles

velocities at the inlet and outlet respectively with the tangential direction, while

made by the relative velocities with the tangential direction. The angles and should match with vane or blade angles at inlet and outlet respectively for a smooth, shockless entry and e$it of the fluid to avoid undersirable losses. -ow we shall apply a simple geometrical relation as follows" %rom the inlet velocity triangle,

or, 8imilarly from the outlet velocity triangle.

((.7)

or,

((.9)

nvoking the e$pressions of and in &q. ((.6), we get H (*ork head, i.e. energy per unit weight of fluid, transferred between the fluid and the rotor as) as

((.:)

The &q ((.:) is an important form of the &uler.s equation relating to fluid machines since it gives the three distinct components of energy transfer as shown by the pair of terms in the round brackets. These components throw light on the nature of the energy transfer. The first term of &q. ((.:) is readily seen to be the change in absolute kinetic energy or dynamic head of the fluid while flowing through the rotor. The second term of &q. ((.:) represents a change in fluid energy due to the movement of the rotating fluid from one radius of rotation to another.

More A.o't Energy Transfer in T'r.o&ac%ines

&quation ((.:) can be better e$plained by demonstrating a steady flow through a container having uniform angular velocity as shown in %ig.(.5b. The centrifugal force on an infinitesimal body of a fluid of mass dm at radius r gives rise to a pressure differential dp across the thickness dr of the body in a manner that a differential force of dpdA acts on the body radially inward. This force, in fact, is the centripetal force responsible for the rotation of the fluid element and thus becomes equal to the centrifugal force under equilibrium conditions in the radial direction. Therefore, we can write

with dm ; dA dr < where < is the density of the fluid, it becomes

%or a reversible flow (flow without friction) between two points, say, ( and 1, the work done per unit mass of the fluid (i.e., the flow work) can be written as

The work is, therefore, done on or by the fluid element due to its displacement from radius to radius and hence becomes equal to the energy held or lost by it. 8ince the centrifugal force field is responsible for this energy transfer, the corresponding head (energy per unit weight) head. The transfer of energy due to a change in centrifugal head the static head of the fluid. is termed as centrifugal causes a change in

The third term represents a change in the static head due to a change in fluid velocity relative to the rotor. This is similar to what happens in case of a flow through a fi$ed duct of variable cross3sectional area. =egarding the effect of flow area on fluid velocity relative to the rotor, a converging passage in the

direction of flow through the rotor increases the relative velocity and hence decreases the static pressure. This usually happens in case of turbines. 8imilarly, a diverging passage in the direction of flow through the rotor decreases the relative velocity occurs in case of pumps and compressors. and increases the static pressure as

The fact that the second and third terms of &q. ((.:) correspond to a change in static head can be demonstrated analytically by deriving +ernoulli.s equation in the frame of the rotor. n a rotating frame, the momentum equation for the flow of a fluid, assumed ,inviscid, can be written as

where

is the fluid velocity relative to the coordinate frame rotating with an angular velocity

*e assume that the flow is steady in the rotating frame so that coordinate system to

. *e choose a cylindrical

with #3a$is along the a$is of rotation. Then the momentum equation reduces

where

and

are the unit vectors along z and r direction respectively. /et and s be a coordinate along the stream line. Then we can write

be a unit vector in the

direction of

More A.o't Energy Transfer in T'r.o&ac%ines

Taking scalar product with it becomes

*e have used

. *ith a little rearrangement, we have

8ince v is the velocity relative to the rotating frame we can replace it by . %urther is the linear velocity of the rotor. ntegrating the momentum equation from inlet to outlet along a streamline we have

(1.() or, Therefore, we can say, with the help of &q. (1.(), that last two terms of &q. ((.:) represent a change in the static head of fluid.

Energy Transfer in A1ial Flo+ Mac%ines %or an a$ial flow machine, the main direction of flow is parallel to the a$is of the rotor, and hence the inlet and outlet points of the flow do not vary in their radial locations from the a$is of rotation. Therefore,
and the equation of energy transfer &q. ((.:) can be written, under this situation, as

(1.1)

0ence, change in the static head in the rotor of an a$ial flow machine is only due to the flow of fluid through the variable area passage in the rotor.

Radially O't+ard and In+ard Flo+ Mac%ines

%or radially outward flow machines, , and hence the fluid gains in static head, while, for a

radially inward flow machine, and the fluid losses its static head. Therefore, in radial flow pumps or compressors the flow is always directed radially outward, and in a radial flow turbine it is directed radially inward.

I&#'lse and Reaction Mac%ines The relative proportion of energy transfer obtained by the
change in static head and by the change in dynamic head is one of the important factors for classifying fluid machines. The machine for which the change in static head in the rotor is #ero is known as impulse machine . n these machines, the energy transfer in the rotor takes place only by the change in dynamic head of the fluid. The parameter characteri#ing the proportions of changes in the dynamic and static head in the rotor of a fluid machine is known as degree of reaction and is defined as the ratio of energy transfer by the change in static head to the total energy transfer in the rotor. Therefore, the degree of reaction,

(1.5)

I&#'lse and Reaction Mac%ines

%or an impulse machine R ! , because there is no change in static pressure in the rotor. t is difficult to obtain a radial flow impulse machine, since the change in centrifugal head is obvious there. -evertheless, an impulse machine of radial flow type can be conceived by having a change in static head in one direction contributed by the centrifugal effect and an equal change in the other direction contributed by the change in relative velocity. 0owever, this has not been established in practice. Thus for an a$ial flow impulse machine . %or an impulse machine, the rotor can be made open, that is, the velocity "# can represent an open )et of fluid flowing through the rotor, which needs no casing. A very simple e$ample of an impulse machine is a paddle wheel rotated by the impingement of water from a stationary no##le as shown in %ig.1.(a.

Fig *!$

,a- "addle +%eel as an e1a&#le of i&#'lse t'r.ine ,.- La+n s#rin2ler as an e1a&#le of reaction t'r.ine

A machine with any degree of reaction must have an enclosed rotor so that the fluid cannot e$pand freely in all direction. A simple e$ample of a reaction machine can be shown by the familiar lawn sprinkler, in which water comes out (%ig. 1.(b) at a high velocity from the rotor in a tangential direction. The essential feature of the rotor is that water enters at high pressure and this pressure energy is transformed into kinetic energy by a no##le which is a part of the rotor itself. n the earlier e$ample of impulse machine (%ig. 1.(a), the no##le is stationary and its function is only to transform pressure energy to kinetic energy and finally this kinetic energy is transferred to the rotor by pure impulse action. The change in momentum of the fluid in the no##le gives rise to a reaction force but as the no##le is held stationary, no energy is transferred by it. n the case of lawn sprinkler (%ig. 1.(b), the no##le, being

a part of the rotor, is free to move and, in fact, rotates due to the reaction force caused by the change in momentum of the fluid and hence the word reaction machine follows.

Efficiencies
The concept of efficiency of any machine comes from the consideration of energy transfer and is defined, in general, as the ratio of useful energy delivered to the energy supplied. Two efficiencies are usually considered for fluid machines33 the hydraulic efficiency concerning the energy transfer between the fluid and the rotor, and the overall efficiency concerning the energy transfer between the fluid and the shaft. The difference between the two represents the energy absorbed by bearings, glands, couplings, etc. or, in general, by pure mechanical effects which occur between the rotor itself and the point of actual power input or output. Therefore, for a pump or compressor,

(1.6a)

(1.6b)

%or a turbine,

(1.7a)

(1.7b)

The ratio of rotor and shaft energy is represented by mechanical efficiency Therefore

(1.9)

"rinci#le of Si&ilarity and Di&ensional Analysis

The principle of similarity is a consequence of nature for any physical phenomenon. +y making use of this principle, it becomes possible to predict the performance of one machine from the results of tests on a geometrically similar machine, and also to predict the performance of the same machine under conditions different from the test conditions. %or fluid machine, geometrical similarity must apply to all significant parts of the system vi#., the rotor, the entrance and discharge passages and so on. Machines which are geometrically similar form a homologous series. Therefore, the member of such a series, having a common shape are simply enlargements or reductions of each other. f two machines are kinematically similar, the velocity vector diagrams at inlet and outlet of the rotor of one machine must be similar to those of the other. >eometrical similarity of the inlet and outlet velocity diagrams is, therefore, a necessary condition for dynamic similarity. /et us now apply dimensional analysis to determine the dimensionless parameters, i.e., the ? terms as the criteria of similarity for flows through fluid machines. %or a machine of a given shape, and handling compressible fluid, the relevant variables are given in Table 5.( Table 5.( 4ariable @hysical @arameters of %luid Machine 0aria.le #%ysical #ara&eters Di&ensional for&'la

$ ; any physical dimension of the machine as a measure of the / machine.s si#e, usually the rotor diameter % ; volume flow rate through the machine & ; rotational speed (revAmin.) /5 T 3( T 3(

H ; difference in head (energy per unit weight) across the machine. This may be either gained or given by the fluid / depending upon whether the machine is a pump or a turbine respectively

=density of fluid
; viscosity of fluid ' ; coefficient of elasticity of fluid g ; acceleration due to gravity

M/35 M/3( T 3( M/3( T31 /T 31

( ; power transferred between fluid and rotor (the difference between ( and H is taken care of by the hydraulic efficiency M/1 T35

n almost all fluid machines flow with a free surface does not occur, and the effect of gravitational force is negligible. Therefore, it is more logical to consider the energy per unit mass gH as the variable rather than H alone so that acceleration due to gravity does not appear as a separate variable. Therefore, the number of separate variables becomes eight" $, %, &, gH, ), *, ' and ( . 8ince the number of fundamental dimensions required to e$press these variable are three, the number of independent ? terms (dimensionless terms), becomes five. Bsing +uckingham.s ? theorem with $, & and < as the repeating variables, the e$pression for the terms are obtained as,

*e shall now discuss the physical significance and usual terminologies of the different ? terms. All lengths of the machine are proportional to $ , and all areas to !1. Therefore, the average flow velocity at any section in the machine is proportional to . Again, the peripheral velocity of the rotor is proportional to the product &$ . The first ? term can be e$pressed as

Si&ilarity and Di&ensional Analysis

Thus, represents the condition for kinematic similarity, and is known as capacity coefficient or term is known as the head coefficient since it e$presses the rotor velocity, the term becomes by

discharge coefficient The second

head H in dimensionless form. Considering the fact that &$

, and can be interpreted as the ratio of fluid head to kinetic energy of the rotor, !ividing the square of we get

The term

can be e$pressed as

and thus represents the =eynolds number with rotor and , it becomes is

velocity as the characteristic velocity. Again, if we make the product of

which represents the =eynolds.s number based on fluid velocity. Therefore, if kept same to obtain kinematic similarity, fluid velocity. The term

becomes proportional to the =eynolds number based on

e$presses the power ( in dimensionless form and is therefore known as power coefficient

. Combination of and in the form of gives . The term .@Dg0. represents the rate of total energy given up by the fluid, in case of turbine, and gained by the fluid in case of pump or compressor. 8ince ( is the power transferred to or from the rotor. Therefore hydraulic efficiency get for a turbine and becomes the term, we

for a pump or a compressor. %rom the fifth

Multiplying

, on both sides, we get

Therefore, we find that

represents the well known +ach number , Ma.

%or a fluid machine, handling incompressible fluid, the term can be dropped. The effect of liquid viscosity on the performance of fluid machines is neglected or regarded as secondary, (which is often sufficiently true for certain cases or over a limited range).Therefore the term dropped.The general relationship between the different dimensionless variables ( e$pressed as can also be terms) can be

(5.()

Therefore one set of relationship or curves of the performance of all the members of one series.

terms would be sufficient to describe the

Si&ilarity and Di&ensional Analysis

or, with another arrangement of the ? terms,

(5.1)

f data obtained from tests on model machine, are plotted so as to show the variation of dimensionless

parameters with one another, then the graphs are applicable to any machine in the same homologous series. The curves for other homologous series would naturally be different.

S#ecific S#eed The performance or operating conditions for a turbine handling a particular fluid are usually e$pressed by the values of & , ( and H , and for a pump by & , % and H . t is important to know the range of these operating parameters covered by a machine of a particular shape (homologous series) at high efficiency. 8uch information enables us to select the type of machine best suited to a particular application, and thus serves as a starting point in its design. Therefore a parameter independent of the si#e of the machine $ is required which will be the characteristic of all the machines of a homologous
series. A parameter involving & , ( and H but not $ is obtained by dividing this parameter be designated by as by . /et

(5.5)

8imilarly, a parameter involving & , % and H but not $ is obtained by divining represented by as

by

and is

(5.6)

8ince the dimensionless parameters and are found as a combination of basic ? terms, they must remain same for complete similarity of flow in machines of a homologous series. Therefore, a particular value of or relates all the combinations of & , ( and H or & , % and H for which the flow conditions are similar in the machines of that homologous series. nterest naturally centers on the conditions for which the efficiency is a ma$imum. %or turbines, the values of & , ( and H , and for pumps and compressors, the values of & , % and H are usually quoted for which the machines run at ma$imum efficiency.

The machines of particular homologous series, that is, of a particular shape, correspond to a particular value of for their ma$imum efficient operation. Machines of different shapes have, in general,

different values of . Thus the parameter is referred to as the shape factor of the machines. Considering the fluids used by the machines to be incompressible, (for hydraulic turbines and pumps), and since the acceleration due to gravity dose not vary under this situation, the terms g and are taken out from the e$pressions of and . The portions left as and

are termed, for the practical purposes, as the specific speed Therefore, we can write,

for turbines or pumps.

(specific speed for turbines) ; (specific speed for turbines) ;

(5.7) (5.9)

The name specific speed for these e$pressions has a little )ustification. 0owever a meaning can be attributed from the concept of a hypothetical machine. %or a turbine, is the speed of a member of the same homologous series as the actual turbine, so reduced in si#e as to generate unit power under a unit head of the fluid. 8imilarly, for a pump, is speed of a hypothetical pump with reduced si#e but representing a homologous series so that it delivers unit flow rate at a unit head. The specific speed is, therefore, not a dimensionless quantity. The dimension of can be found from their e$pressions given by &qs. (5.7) and (5.9). The dimensional formula and the unit of specific speed are given as follows" 8pecific speed (turbine) (pump) !imensional formula M (A1 T 37A1 /3(A6 / 5A6 T35A1 Bnit (8 ) kg (A1A s7A1 m(A6 m 5A6 A s5A1

The dimensionless parameter from .

is often known as the dimensionless specific speed to distinguish it

3as T'r.ine Syste& 4 Centrif'gal and A1ial Flo+ Co&#ressors

Introd'ction
A turbofan engine that gives propulsive power to an aircraft is shown in %igure 6.( and the schematic of the engine is illustrated in %igure 6.1. The main components of the engine are intake, fan, compressor, combustion chamber or burnner, turbine and e$haust no##le. The intake is a critical part of the aircraft engine that ensures an uniform pressure and velocity at the entry to the compressor. At normal forward speed of the aircraft, the intake performs as a diffusor with rise of static pressure at the cost of kinetic energy of fluid, referred as the .ram pressure rise.. Then the air is passed through the compressor and the high pressure air is fed to the combustion chamber, where the combustion occurs at more or less constant pressure that increases its temperature. After that the high pressure and high temperature gas is e$panded through the turbine. n case of aircraft engine, the e$pansion in the turbine is not complete. 0ere the turbine work is sufficient to drive the compressor. The rest of the pressure is then e$panded through the no##le that produce the require thrust. 0owever, in case of stationery gas turbine unit, the gas is completely e$panded in the turbine. n turbofan engine the air is bypassed that has a great effect on the engine performance, which will be discussed later. Although each component have its own performance characteristics, the overall engine operates on a thermodynamic cycle.

Fig're 5!$ 3as T'r.ine ,Co'rtesy 6 ae!gatec%!ed'-

Fig're 5!* ,Co'rtesy 6 NASA 3lenn Researc% Centren this chapter, we will describe the ideal gas turbine or aircraft propulsion cycles that are useful to review the performance of ideal machines in which perfection of the individual component is assumed. The specific work output and the cycle efficiency then depend only on the pressure ratio and ma$imum cycle temperature. Thus, this cycle analysis are very useful to find the upper limit of performance of individual components. %ollowing assumptions are made to analysis an ideal gas turbine cycle.

(a) The working fluid is a perfect gas with constant specific heat. (b) Compression and e$pansion process are reversible and adiabatic, i.e isentropic. (c) There are no pressure losses in the inlet duct, combustion chamber, heat e$changer, intercooler,

e$haust duct and the ducts connecting the components. (d) The mass flow is constant throughout the cycle. (e) The change of kinetic energy of the working fluid between the inlet and outlet of each component is negligible. (f) The heat3e$changer, if such a component is used, is perfect.

7o'le or Brayton Cycle

The ideal cycle for the simple gas turbine is the Eoule or +rayton cycle which is represented by the cycle (156 in the p3v and T38 diagram (%igure 6.5). The cycle comprises of the following process. (31 is the isentropic compression occuring in the compressor, 135 is the constant pressure heat addition in the combustion chamber, 536 is the isentropic e$pansion in the turbine releasing power output, 63( is the re)ection of heat at constant pressure 3 which closes the cycle. 8trictly speaking, the process 63( does not occur within the plant. The gases at the e$it of the turbine are lost into the atmosphereF therefore it is an open cycle.

C3 Compressor +3 +urner or Combustion Chamber T3 Turbine /3 /oad

Fig're 5!/ Si&#le gas t'r.ine cycle! n a steady flow isentropic process,

Thus, the Compressor work per kg of air Turbine work per kg of air 0eat supplied per kg of air

The cycle efficiency is,

or, Making use of the isentropic relation , we have,

*here, r is pressure ratio. The cycle efficiency is then given by,

Thus, the efficiency of a simple gas turbine depends only on the pressure ratio and the nature of the gas. %igure 6.6 shows the relation between , and r when the working fluid is air (- ;(.6), or a monoatomic gas such as argon( - ;(.99).

Fig're 5!5 Efficiency of a si&#le gast'r.ine cycle The specific work output w, upon which the si#e of plant for a given power depends, is found to be a function not only of pressure ratio but also of ma$imum cycle temperature T5. Thus, the specific work output is,

/et

and

Then

at

means

i.e., no heat addition

Fig're 5!8 S#ecific +or2 o't#'t of a si&#le gas t'r.ine To get the ma$imum work output for a fi$ed temperature ratio t and inlet temperature T(,

or, or, or, Thus, the work output will be ma$imum when the compressor outlet temperature is equal to that of turbine. %igure 6.7 illustrates the variation of specific work output with pressure ratio for different values of temperature ratio. The work output increases with increase of T5 for a constant value of inlet temperature T(. 0owever for a given temperature ratio i.e constant values of T ( and T5, the output becomes ma$imum for a particular pressure ratio. Si&#le Cycle +it% E1%a'st Heat E1c%ange CBT9 Cycle ,Regenerati e cyclen most cases the turbine e$haust temperature is higher than the outlet temperature from the compressor. Thus the e$haust heat can be utilised by providing a heat e$changer that reduces heat input in the combustion chamber. This saving of energy increases the efficiency of the regeneration cycle keeping the specific output unchanged. A regenerative cycle is illustrated in %igure 1.9

for heat e$change to take place *e assume ideal e$change and

Fig're 5!:

Si&#le gas t'r.ine cycle +it% %eat e1c%ange

*ith ideal heat e$change, the cycle efficiency can be e$pressed as,

or,

or,

we can write or,

&fficiency is more than that of simple cycle *ith heat e$change (ideal) the specific output does not change but the efficiency is increased

3as T'r.ine Cycle +it% Re%eat

A common method of increasing the mean temperature of heat reception is to reheat the gas after it has e$panded in a part of the gas turbine. +y doing so the mean temperature of heat re)ection is also increased, resulting in a decrease in the thermal efficiency of the plant. 0owever , the specific output of the plant increases due to reheat. A reheat cycle gas turbine plant is shown in %igure 7.(

Fig're 8!$ Re%eat cycle gas t'r.ine #lant The specific work output is given by,

The heat supplied to the cycle is

Thus, the cycle efficiency,

Therefore, a reheat cycle is used to increase the work output while a regenerative cycle is used to enhance the efficiency.

3as T'r.ine Cycle +it% Inter;cooling

The cooling of air between two stages of compression is known as intercooling. This reduces the work of compression and increases the specific output of the plant with a decrease in the thermal efficiency. The loss in efficiency due to intercooling can be remedied by employing e$haust heat e$change as in the reheat cycle.

8pecific work output

Fig're 8!* Cycle +it% intercooling

0eat supplied ;

is constant and not dependent on temperature, we can write"

-ote 0ere heat supply and output both increases as compared to simple cycle. +ecause the increase in heat supply is proportionally more, decreases. *ith multiple inter3cooling and multiple reheat, the compression and e$pansion processes tend to be isothermal as shown in %igure 7.5

Fig're 8!/ M'lti#le re%eat and intercool cycle The cycle tends towards the &ricsson cycle, the efficiency is same as that of the Carnot cycle

The use of intercoolers is seldom contemplated in practice because they are bulky and need large quantities of cooling water. The main advantage of the gas turbine, that it is compact and self3contained, is then lost.

Act'al 3as T'r.ine Cycle

G &fficiency of the compression and e$pansion processes will come into consideration. G @ressure losses in the ducting, combustion and heat e$changer. G Complete heat e$change in the regenerator is not possible. G Mechanical losses due to bearings au$iliary etc are present. G 8pecific heat of the working fluid varies with temperature. G Mass flow throughout the cycle is not constant.

3as T'r.ine Cycles for "ro#'lsion

n aircraft gas turbine cycles, the useful power output is in the form of thrust. n case of turbo)et and turbofan (%igure 6.(), the whole thrust is generated in the propelling no##el, whereas with the turboprop most is produced by a propeller with only a small contribution from the e$haust no##le. A turboprop and a turbo)et engine is illustrated in %igure 7.6and 7.7 respectively. ,i- T'r.o#ro#

Fig're 8!5 T'r.o#ro# Engine @ower must eventually be delivered to the aircraft in the form of thrust power, )ust as it is with a piston engine driving a propeller. The thrust power ( T( ) can be e$pressed in terms of shaft power ( .( ), propeller efficiency ( ) and )et thrust / by

where

is the forward speed of the aircraft .

,ii- T'r.<et Engine

Fig're 8!8 T'r.o<et Engine

Fig're 8!: "ro#elling No==le

a3( 3H ram effect in intake. (31 3H isentropic compressor 135 3H constant pressure heat addition in the combustion chamber 536 3H sentropic e$pansion in the turbine 63 3HConstant pressure heat addition in the after burner 37 3H isentropic e$pansion in the no##le

Fig're 8!> T'r.o<et Cycle @ropelling no##le refers to the component in which the working fluid is e$panded to give a high velocity )et. +etween the turbine e$it and propelling no##le, there may be a )et pipe. *hen thrust boosting is required, an afterburner may be introduced in the )et pipe as shown in figure 7.:. %igure 7.: indicates the ideal turbo)et cycle on the T38 diagram, which is often used to evaluate the design performance of a turbo)et engine. After reviewing the thermodynamic cycle for a gas turbine or aircraft engine, characteristic features and performance of individual components such as the compressor, turbine, combustion chamber and no##le (in case of aircraft engine) will be discussed in the following section.

Co&#ressors
n Module (, we discussed the basic fluid mechanical principles governing the energy transfer in a fluid machine. A brief description of different types of fluid machines using water as the working fluid was also given in Module (.0owever, there e$ist a large number of fluid machines in practice, that use air, steam and gas (the mi$ture of air and products of burnt fuel) as the working fluids. The density of the fluids change with a change in pressure as well as in temperature as they pass through the machines. These machines are called .compressible flow machines. and more popularly .turbomachines.. Apart from the change in density with pressure, other features of compressible flow, depending upon the regimes, are also observed in course of flow of fluids through turbomachines. Therefore, the basic equation of energy transfer (&uler.s equation, as discussed before) along with the equation of state relating the pressure, density and temperature of the working fluid and other necessary equations of compressible flow, are needed to describe the performance of a turbomachine. 0owever, a detailed discussion on all types of turbomachines is beyond the scope of this book. *e shall present a very brief description of a few compressible flow machines, namely, compressors, fans and blowers in this module. n practice two kinds of compressors"centrifugal and a$ial are generally in use.

Centrif'gal Co&#ressors

A centrifugal compressor is a radial flow rotodynamic fluid machine that uses mostly air as the working fluid and utili#es the mechanical energy imparted to the machine from outside to increase the total internal energy of the fluid mainly in the form of increased static pressure head. !uring the second world war most of the gas turbine units used centrifugal compressors. Attention was focused on the simple turbo)et units where low power3plant weight was of great importance. 8ince the war, however, the a$ial compressors have been developed to the point where it has an appreciably higher isentropic efficiency. Though centrifugal compressors are not that popular today, there is renewed interest in the centrifugal stage, used in con)unction with one or more a$ial stages, for small turbofan and turboprop aircraft engines. A centrifugal compressor essentially consists of three components.

1. A stationary casing 2. A rotating i&#eller as shown in %ig. 9.( (a) which imparts a high velocity to the air. The impeller
may be single or double sided as show in %ig. 9.( (b) and (c), but the fundamental theory is same for both. 3. A diff'ser consisting of a number of fi$ed diverging passages in which the air is decelerated with a consequent rise in static pressure.

Fig're :!$,a-

,.,c,dFig're :!$ Sc%e&atic ie+s of a centrif'gal co&#ressor

Fig're :!* Single entry and single o'tlet centrif'gal co&#resssor %igure 9.1 is the schematic of a centrifugal compressor, where a single entry radial impeller is housed inside a volute casing.

Co&#ressors
"rinci#le of o#eration6 Air is sucked into the impeller eye and whirled outwards at high speed by the impeller disk. At any point in the flow of air through the impeller the centripetal acceleration is obtained by a pressure head so that the static pressure of the air increases from the eye to the tip of the impeller. The remainder of the static pressure rise is obtained in the diffuser, where the very high velocity of air leaving the impeller tip is reduced to almost the velocity with which the air enters the impeller eye.
Bsually, about half of the total pressure rise occurs in the impeller and the other half in the diffuser. Iwing to the action of the vanes in carrying the air around with the impeller, there is a slightly higher static pressure on the forward side of the vane than on the trailing face. The air will thus tend to flow around the edge of the vanes in the clearing space between the impeller and the casing. This results in a loss of efficiency and the clearance must be kept as small as possible. 8ometimes, a shroud attached to the blades as shown in %igure.9.((d) may eliminate such a loss, but it is avoided because of increased disc friction loss and of manufacturing difficulties. The straight and radial blades are usually employed to avoid any undesirable bending stress to be set up in the blades. The choice of radial blades also determines that the total pressure rise is divided equally between impeller and diffuser. +efore further discussions following points are worth mentioning for a centrifugal compresssor. (i) The pressure rise per stage is high and the volume flow rate tends to be low. The pressure rise per stage is generally limited to 6"( for smooth operations. (ii) +lade geometry is relatively simple and small foreign material does not affect much on operational characteristics. (iii) Centrifugal impellers have lower efficiency compared to a$ial impellers and when used in aircraft engine it increases frontal area and thus drag. Multistaging is also difficult to achieve in case of centrifugal machines.

?or2 done and #ress're rise

8ince no work is done on the air in the diffuser, the energy absorbed by the compressor will be determined by the conditions of the air at the inlet and outlet of the impeller. At the first instance, it is assumed that the air enters the impeller eye in the a$ial direction, so that the initial angular momentum of the air is #ero. The a$ial portion of the vanes must be curved so that the air can pass smoothly into the eye. The angle which the leading edge of a vane makes with the tangential direction, , will be given by the direction of the relative velocity of the air at inlet, , as

shown in %ig. 9.5. The air leaves the impeller tip with an absolute velocity of component . Bnder ideal conditions,

that will have a tangential or whirl

, would be such that the whirl component is equal to the impeller speed .

at the tip. 8ince air enters the impeller in a$ial direction,

Fig're :!/ 0elocity triangles at inlet and o'tlet of i&#eller .lades Bnder the situation of as and , we can derive from &q. ((.1), the energy transfer per unit mass of air

(9.() !ue to its inertia, the air trapped between the impeller vanes is reluctant to move round with the impeller and we have already noted that this results in a higher static pressure on the leading face of a vane than on the trailing face. t also prevents the air acquiring a whirl velocity equal to impeller speed. This effect is known as slip. +ecause of slip, we obtain J . The slip factor 0 is defined in the similar way as done in the case of a centrifugal pump as

The value of 0 lies between 2.K to 2.K1. The energy transfer per unit mass in case of slip becomes (9.1) Ine of the widely used e$pressions for 0 was suggested by 8tanit# from the solution of potential flow through impeller passages. t is given by

, where n is the number of vanes.

"o+er In#'t Factor

The power input factor takes into account of the effect of disk friction, windage, etc. for which a little more power has to be supplied than required by the theoretical e$pression. Considering all these losses, the actual work done (or energy input) on the air per unit mass becomes (:.() where is the power input factor. %rom steady flow energy equation and in consideration of air as an ideal gas, one can write for adiabatic work w per unit mass of air flow as (:.1)

where and are the stagnation temperatures at inlet and outlet of the impeller, and specific heat over the entire temperature range. *ith the help of &q. (9.5), we can write (:.5)

is the mean

The stagnation temperature represents the total energy held by a fluid. 8ince no energy is added in the diffuser, the stagnation temperature rise across the impeller must be equal to that across the whole compressor. f the stagnation temperature at the outlet of the diffuser is designated by . Ine can write from &qn. (:.5) , then

(:.6)

The overall stagnation pressure ratio can be written as

(:.7)

where,

and

are the stagnation temperatures at the end of an ideal (isentropic) and actual is the isentropic efficiency defined as

process of compression respectively (%igure :.(), and

(:.9)

Fig're >!$ Ideal and act'al #rocesses of co&#ression on T;s #lane 8ince the stagnation temperature at the outlet of impeller is same as that at the outlet of the diffuser, one can also write in place of in &q. (:.9). Typical values of the power input factor lie in the region of

(.257 to (.26. f we know we will be able to calculate the stagnation pressure rise for a given impeller speed. The variation in stagnation pressure ratio across the impeller with the impeller speed is shown in %igure :.1. %or common materials, is limited to 672 mAs. at the eye tip has to be

%igure :.5 shows the inducing section of a compressor. The relative velocity

held low otherwise the Mach umber (based on

) given by

will be too high causing

shock losses. Mach number should be in the range of 2.:32.K. The typical inlet velocity triangles for large and medium or small eye tip diameter are shown in %igure :.6(a) and (b) respectively.

Fig're >!* 0ariation in stagnation #ress're ratio +it% i&#eller ti# s#eed

Fig're >!/ Ind'cing section of a centrif'gal co&#ressor

Fig're >!5 ,a-

Fig're >!5,.Fig're >!5 0elocity triangles at t%e ti# of eye

Diff'ser

The basic purpose of a compressor is to deliver air at high pressure required for burning fuel in a combustion chamber so that the burnt products of combustion at high pressure and temperature are used in turbines or propelling no##les (in case of an aircraft engine) to develop mechanical power. The problem of designing an efficient combustion chamber is eased if velocity of the air entering the combustion chamber is as low as possible. t is necessary, therefore to design the diffuser so that only a small part of the stagnation temperature at the compressor outlet corresponds to kinetic energy. t is much more difficult to arrange for an efficient deceleration of flow than it is to obtain efficient acceleration. There is a natural tendency in a diffusing process for the air to break away from the walls of the diverging passage and reverse its direction. This is typically due to the phenomenon of boundary layer separation and is shown in %igure. :.7. &$periments have shown that the ma$imum permissible included angle of divergence is ((L to avoid considerable losses due to flow separation.

Fig're >!8 Accelerating and decelerating flo+s n order to control the flow of air effectively and carry3out the diffusion process in a length as short as possible, the air leaving the impeller is divided into a number of separate streams by fi$ed diffuser vanes. Bsually the passages formed by the vanes are of constant depth, the width diverging in accordance with the shape of the vanes. The angle of the diffuser vanes at the leading edge must be designed to suit the direction of the absolute velocity of the air at the radius of the leading edges, so that the air will flow smoothly over vanes. As there is a radial gap between the impeller tip and the leading edge of the vanes , this direction will not be that with which the air leaves the impeller tip. To find the correct angle for diffuser vanes, the flow in the vaneless space should be considered. -o further energy is supplied to the air after it leaves the impeller. f we neglect the frictional losses, the angular momentum remains constant. 0ence decreases from impeller tip to diffuser vane, in inverse proportion to the radius. %or a channel of constant depth, the area of flow in the radial direction is directly proportional to the radius. The radial velocity will therefore also decrease from impeller tip to

diffuser vane, in accordance with the equation of continuity. f both and decrease from the impeller tip then the resultant velocity " decreases from the impeller tip and some diffusion takes place in the vaneless space. The consequent increase in density means that proportion to the radius as done by continuity. , and the way will not decrease in inverse

varies must be found from the equation of

Losses in a Centrif'gal Co&#ressor

The losses in a centrifugal compressor are almost of the same types as those in a centrifugal pump.

0owever, the following features are to be noted. Frictional losses6 A ma)or portion of the losses is due to fluid friction in stationary and rotating blade passages. The flow in impeller and diffuser is decelerating in nature. Therefore the frictional losses are due to both skin friction and boundary layer separation. The losses depend on the friction factor, length of the flow passage and square of the fluid velocity. The variation of frictional losses with mass flow is shown in %igure. M.(. Incidence losses6 !uring the off3design conditions, the direction of relative velocity of fluid at inlet does not match with the inlet blade angle and therefore fluid cannot enter the blade passage smoothly by gliding along the blade surface. The loss in energy that takes place because of this is known as incidence loss. This is sometimes referred to as shock losses. 0owever, the word shock in this conte$t should not be confused with the aerodynamic sense of shock which is a sudden discontinuity in fluid properties and flow parameters that arises when a supersonic flow decelerates to a subsonic one. Clearance and lea2age losses6 Certain minimum clearances are necessary between the impeller shaft and the casing and between the outlet periphery of the impeller eye and the casing. The leakage of gas through the shaft clearance is minimi#ed by employing glands. The clearance losses depend upon the impeller diameter and the static pressure at the impeller tip. A larger diameter of impeller is necessary for a higher peripheral speed between the casing and the impeller eye tip. and it is very difficult in the situation to provide sealing

The variations of frictional losses, incidence losses and the total losses with mass flow rate are shown in %igure.M.(

Fig're @!$ De#endence of ario's losses +it% &ass flo+ in a centrif'gal co&#ressor The leakage losses comprise a small fraction of the total loss. The incidence losses attain the minimum value at the designed mass flow rate. The shock losses are, in fact #ero at the designed flow rate. 0owever, the incidence losses, as shown in %ig. M.(, comprises both shock losses and impeller entry loss due to a change in the direction of fluid flow from a$ial to radial direction in the vaneless space before entering the impeller blades. The impeller entry loss is similar to that in a pipe bend and is very small compared to other losses. This is why the incidence losses show a non #ero minimum value (%igure. M.() at the designed flow rate.

Co&#ressor c%aracteristics
The theoretical and actual head3discharge relationships of a centrifugal compressor are same as those of a centrifugal pump as described in Module (. 0owever, the performance of a compressor is usually specified by curves of delivery pressure and temperature against mass flow rate for various fi$ed values of rotational speed at given values of inlet pressure and temperature. t is always advisable to plot such performance characteristic curves with dimensionless variables. To find these dimensionless variables, we start with a implicit functional relationship of all the variables as (M.()

where $ ; characteristic linear dimension of the machine, & ; rotational, m ; mass flow rate, stagnation pressure at compressor inlet, stagnation temperature at compressor inlet, characteristics gas constant. ; stagnation pressure at compressor outlet,

; ;

; stagnation temperature at compressor outlet, and R ;

+y making use of +uckingham.s ? theorem, we obtain the non3dimensional groups (? terms) as

The third and fourth non3dimensional groups are defined as .non3dimensional mass flow. and .non3 dimensional rotational speed. respectively. The physical interpretation of these two non3dimensional groups can be ascertained as follows.

Therefore, the .non3dimensional mass flow. and .non3dimensional rotational speed. can be regarded as flow Mach number, and rotational speed Mach number, .

*hen we are concerned with the performance of a machine of fi$ed si#e compressing a specified gas, and $ may be omitted from the groups and we can write

(M.1) %unction

Though the terms and are truly not dimensionless, they are referred as .non3 dimensional mass flow. and .non3dimensional rotational speed. for practical purpose. The stagnation pressure and temperature ratios and are plotted against in the form of

two families of curves, each curve of a family being drawn for fi$ed values of . The two families of curves represent the compressor characteristics. %rom these curves, it is possible to draw the curves of isentropic efficiency definition of the isentropic efficiency as for fi$ed values of . *e can recall, in this conte$t, the

(M.5)

+efore describing a typical set of characteristics, it is desirable to consider what might be e$pected to occur when a valve placed in the delivery line of the compressor running at a constant speed, is slowly opened. *hen the valve is shut and the mass flow rate is #ero, the pressure ratio will have some value. %igure M.1 indicates a theoretical characteristics curve A+C for a constant speed. The centrifugal pressure head produced by the action of the impeller on the air trapped between the vanes is represented by the point .A. in %igure M.1. As the valve is opened, flow commences and diffuser begins to influence the pressure rise, for which the pressure ratio increases. At some point 121, efficiency approaches its ma$imum and the pressure ratio also reaches its ma$imum. %urther increase of mass flow will result in a fall of pressure ratio. %or mass flows greatly in e$cess of that corresponding to the design mass flow, the air angles will be widely different from the vane angles and breakaway of the air will occur. n this hypothetical case, the pressure ratio drops to unity at 131 , when the valve is fully open and all the power is absorbed in overcoming internal frictional resistances. n practice, the operating point .A. could be obtained if desired but a part of the curve between 1A1 and .+. could not be obtained due to surging. t may be e$plained in the following way. f we suppose that the compressor is operating at a point .!. on the part of characteristics curve (%igure M.1) having a positive slope, then a decrease in mass flow will be accompanied by a fall in delivery pressure. f the pressure of the air downstream of the compressor does not fall quickly enough, the air will tend to reverse its direction and will flow back in the direction of the resulting pressure gradient. *hen this occurs, the pressure ratio drops rapidly causing a further drop in mass flow until the point .A. is reached, where the mass flow is #ero. *hen the pressure downstream of the compressor has reduced sufficiently due to reduced mass flow rate, the positive flow becomes established again and the compressor picks up to repeat the cycle of events which occurs at high frequency. This surging of air may not happen immediately when the operating point moves to the left of .+. because the pressure downstream of the compressor may at first fall at a greater rate than the delivery pressure. As the mass flow is reduced further, the flow reversal may occur and the conditions are unstable between .A. and .+.. As long as the operating point is on the part of the characteristics having a negative slope, however, decrease in mass flow is accompanied by a rise in delivery pressure and the operation is stable.

Fig're @!* T%e t%eoretical c%aracteristic c'r e There is an additional limitation to the operating range, between .+. and .C.. As the mass flow increases and the pressure decreases, the density is reduced and the radial component of velocity must increase. At constant rotational speed this means an increase in resultant velocity and hence an angle of incidence at the diffuser vane leading edge. At some point say .&., the position is reached where no further increase in mass flow can be obtained no matter how wide open the control valve is. This point represents the ma$imum delivery obtainable at the particular rotational speed for which the curve is drawn. This indicates that at some point within the compressor sonic conditions have been reached, causing the limiting ma$imum mass flow rate to be set as in the case of compressible flow through a converging diverging no##le. Choking is said to have taken place. Ither curves may be obtained for different speeds, so that the actual variation of pressure ratio over the complete range of mass flow and rotational speed will be shown by curves such as those in %igure. M.5. The left hand e$tremities of the constant speed curves may be )oined up to form surge line, the right hand e$tremities indicate choking (%igure M.5).

Fig're @!/ 0ariations of #ress're ratio o er t%e co&#lete range of &ass flo+ for different rotational s#eeds

A1ial Flo+ Co&#ressors

The basic components of an a$ial flow compressor are a rotor and stator, the former carrying the moving blades and the latter the stationary rows of blades. The stationary blades convert the kinetic energy of the fluid into pressure energy, and also redirect the flow into an angle suitable for entry to the ne$t row of moving blades. &ach stage will consist of one rotor row followed by a stator row, but it is usual to provide a row of so called inlet guide vanes. This is an additional stator row upstream of the first stage in the compressor and serves to direct the a$ially approaching flow correctly into the first row of rotating blades. %or a compressor, a row of rotor blades followed by a row of stator blades is called a stage. Two forms of rotor have been taken up, namely drum type and disk type. A disk type rotor illustrated in %igure K.( The disk type is used where consideration of low weight is most important. There is a contraction of the flow annulus from the low to the high pressure end of the compressor. This is necessary to maintain the a$ial velocity at a reasonably constant level throughout the length of the compressor despite the increase in density of air. %igure K.1 illustrate flow through compressor stages. n an a$ial compressor, the flow rate tends to be high and pressure rise per stage is low. t also maintains fairly high efficiency.

Fig're A!$ Dis2 ty#e a1ial flo+ co&#ressor The basic principle of acceleration of the working fluid, followed by diffusion to convert acquired kinetic energy into a pressure rise, is applied in the a$ial compressor. The flow is considered as occurring in a tangential plane at the mean blade height where the blade peripheral velocity is 4 . This two dimensional approach means that in general the flow velocity will have two components, one a$ial and one peripheral denoted by subscript w , implying a whirl velocity. t is first assumed that the air approaches the rotor blades with an absolute velocity, , at and angle to the a$ial direction. n

combination with the peripheral velocity 4 of the blades, its relative velocity will be at and angle as shown in the upper velocity triangle (%igure K.5). After passing through the diverging passages formed between the rotor blades which do work on the air and increase its absolute velocity, the air will emerge with the relative velocity of at angle which is less than . This turning of air towards the a$ial direction is, as previously mentioned, necessary to provide an increase in the effective flow area and is brought about by the camber of the blades. 8ince is less than due to diffusion,

some pressure rise has been accomplished in the rotor. The velocity the absolute velocity at the e$it from the rotor at an angle

in combination with 4 gives

to the a$ial direction. The air then at an

passes through the passages formed by the stator blades where it is further diffused to velocity

angle which in most designs equals to so that it is prepared for entry to ne$t stage. 0ere again, the turning of the air towards the a$ial direction is brought about by the camber of the blades.

Fig're A!* Flo+ t%ro'g% stages

Fig're A!/ 0elocity triangles Two basic equations follow immediately from the geometry of the velocity triangles. These are"

(K.()

(K.1)

n which is the a$ial velocity, assumed constant through the stage. The work done per unit mass or specific work input, w being given by (K.5) This e$pression can be put in terms of the a$ial velocity and air angles to give (K.6) or by using &qs. (K.() and (K.1) (K.7)

This input energy will be absorbed usefully in raising the pressure and velocity of the air. A part of it will be spent in overcoming various frictional losses. =egardless of the losses, the input will reveal itself as a rise in the stagnation temperature of the air is made equal to that at the entry. temperature rise of the stage, . f the absolute velocity of the air leaving the stage will also be the static

, the stagnation temperature rise

, so that

(K.9)

n fact, the stage temperature rise will be less than that given in &q. (K.9) owing to three dimensional effects in the compressor annulus. &$periments show that it is necessary to multiply the right hand side of &q. (K.9) by a work3done factor N which is a number less than unity. This is a measure of the ratio of actual work3absorbing capacity of the stage to its ideal value. The radial distribution of a$ial velocity is not constant across the annulus but becomes increasingly peaky (%igure. K.6) as the flow proceeds, settling down to a fi$ed profile at about the fourth stage. &quation (K.7) can be written with the help of &q. (K.() as (K.:)

Figure 9.4 Axial velocity distibutions

8ince the outlet angles of the stator and the rotor blades fi$ the value of of . Any increase in will result in a decrease in and and hence the value

and vice3versa. f the

compressor is designed for constant radial distribution of

as shown by the dotted line in %igure

(K.6), the effect of an increase in in the central region of the annulus will be to reduce the work capacity of blading in that area. 0owever this reduction is somewhat compensated by an increase in in the regions of the root and tip of the blading because of the reduction of at these parts of the annulus. The net result is a loss in total work capacity because of the adverse effects of blade tip clearance and boundary layers on the annulus walls. This effect becomes more pronounced as the number of stages is increased and the way in which the mean value varies with the number of stages. The variation of with the number of stages is shown in %igure. K.7. Care should be taken to avoid confusion of the work done factor with the idea of an efficiency. f is the e$pression for the specific work input (&quation. K.5), then is the actual amount of work which can be supplied to the stage. The application of an isentropic efficiency to the resulting temperature rise will yield the equivalent isentropic temperature rise from which the stage pressure ratio may be calculated. Thus, the actual stage temperature rise is given by

(K.M)

and the pressure ratio

by

(K.K)

where,

is the inlet stagnation temperature and

is the stage isentropic efficiency.

Fig're A!8 0ariation of +or2;done factor +it% n'&.er of stages E1a&#le6 At the mean diameter, 4 ; 12 mAs, and ; (5.7 L. The factor ; 2.M9 and , ; 65.K L

; 2.M7 and inlet temperature

is 1MM '. Calculate the pressure ratio.

; 11.16 '

and

of air has been taken as (227 EAkg 'P

Degree of Reaction
A certain amount of distributionof pressure (a rise in static pressure) takes place as the air passes through the rotor as well as the statorF the rise in pressure through the stage is in general, attributed to both the blade rows. The term degree of reaction is a measure of the e$tent to which the rotor itself contributes to the increase in the static head of fluid. t is defined as the ratio of the static enthalpy rise in the rotor to that in the whole stage. 4ariation of over the relevant temperature range will be negligibly small and hence this ratio of enthalpy rise will be equal to the corresponding temperature rise. t is useful to obtain a formula for the degree of reaction in terms of the various velocities and air angles associated with the stage. This will be done for the most common case in which it is assumed that the air leaves the stage with the same velocity (absolute) with which it enters ( This leads to respectively, . f and ).

are the static temperature rises in the rotor and the stator

then from &qs (K.6),(K.7),(K.9),

((2.()

8ince all the work input to the stage is transferred to air by means of the rotor, the steady flow energy equation yields,

*ith the help of &q. ((2.(), it becomes

+ut

and

, and hence

((2.1)

The degree of reaction

((2.5)

*ith the help of &q. ((2.1), it becomes

and

+y adding up &q. (K.() and &q. (K.1) we get

=eplacing

and

in the e$pression for

with

and

((2.6)

As the case of 72Q reaction blading is important in design, it is of interest to see the result for R ;2.7 ,

and it follows from &qs. (K.() and (K.1) that i.e. i.e. ((2.7a) ((2.7b)

%urthermore since

is constant through the stage.

And since we have initially assumed that

, it follows that

. +ecause of this equality of

angles, namely, and , blading designed on this basis is sometimes referred to as symmetrical blading .The 72Q reaction stage is called a repeating stage. t is to be remembered that in deriving &q. ((2.6) for , we have implicitly assumed a work done factor of unity in making use of &q. ((2.1). A stage designed with symmetrical blading is referred to as 72Q reaction stage, although will differ slightly for N.

Ele&entary Cascade T%eory

The previous module dealt with the a$ial flow compressors, where all the analyses were based on the flow conditions at inlet to and e$it from the impeller following kinematics of flow e$pressed in terms of velocity triangles. 0owever, nothing has been mentioned about layout and design of blades, which are aerofoil sections. n the development of the highly efficient modern a$ial flow compressor or turbine, the study of the two3 dimensional flow through a cascade of aerofoils has played an important part. An array of blades representing the blade ring of an actual turbo machinery is called the cascade. %igure ((.(a shows a compressor blade cascade tunnel. As the air stream is passed through the cascade, the direction of air is turned. @ressure and velocity measurements are made at up and downstream of cascade as shown. The cascade is mounted on a turn3table so that its angular direction relative to the direction of inflow can be changed, which enables tests to be made for a range of incidence angle. As the flow passes through the cascade, it is deflected and there will be a circulation ((.(b S ((.(c). and thus the lift generated will be with the a$ial direction. (%ig

is the mean velocity that makes an angle

Fig're $$!$a A Cascade T'nnel

Co&#ressor cascade 6
%or a compressor cascade, the static pressure will rise across the cascade, i.e.

Fig're $$!$. Co&#ressor Cascade

3 ;chord of the blade . ; pitch

Fig're $$!$c 0elocity triangle Circ'lation6

/ift ;

/ift coefficient,

from velocity triangles,

where .,3 3depend on the design of the cascade 3 flow angles at the inlet and outlet /ift is perpendicular to line

T'r.ine Cascade6 8tatic pressure will drop across the turbine cascade, i.e

/ift coefficient, Above discussion is based on 'utta3Eoukowski theorem Assumption 3 nviscid flow n reality, we face viscous flow together with formation of wakes. Thus, the viscous flow is the cause of drag which in turn affect the lift .

Effect of isco's flo+

As the fluid passes through the cascade, there will be a decrease in total pressure between the inlet to the cascade and at a section well downstream of the cascade due to frictional effect on aerofoils and also losses due to mi$ing of blade wades (i.e. the effect of viscous flow). 8ince at up and downstream, the flow is uniform and kinematics of flow remains unchanged, the dynamic pressure at up and downstream remains the same. Thus loss in total pressure is the same as that of static pressure.

The loss in total pressure

consists of two components"

(() %rictional loss due to the formation of boundary layer on blades. (1) Mi$ing of the blade wades. .

Co&#ressor Cascade ,0isco's Casen compressor cascade, due to losses in total pressure , there will be an a$ial force figure below. Thus the drag, which is perpendicular to the lift, is defined as as shown in

The lift will be reduced due to the effect of drag which can be e$pressed as" &ffective lift ; The lift has decreased due to viscosity,

Actual lift coefficient

where,

;drag coefficient,

n the case of turbine, drag will contribute to work (and is considered as useful).

T'r.ine Cascade ,0isco's case-

!rag ; &ffective lift ;

Actual lift coefficient, The drag increases the lift. Thus, thedrag is an useful component for work.

Blade efficiency ,or diff'sion efficiency%or a compressor cascade, the blade efficiency is defined as"

!ue to viscous effect, static pressure rise is reduced

from velocity triangle"

Also we get

OAppro$imation" i.e.in the e$pression for lift, the effect of drag is ignoredP

3ma$imum, if

The value of for which efficiency is ma$imum, The blade efficiency for a turbine cascade is defined as"

%or very small ratio of

which is same as the compressor cascade

-ature of variation of -ote"

wrt mean flow angle provides fle$ibility in

does not vary much in the range design.

n the above derivation for blade efficiency of both the compressor and turbine cascade, the lift is assumed as , neglecting the effect of drag. *ith the corrected e$pression of lift, actual blade efficiencies are as follows"

Cascade No&enclat're
An aerofoil is build up around a basic camber line, which is usually a curricular or a parabolic arc (figure below). An camber line is thus the skeleton of the aerofoil. A thickness t is distributed over the camber line with the leading and trailing edge circles that finally form an aerofoil.

n the above figure, the dotted line indicates the camber line and .a. is the distance from the leading edge for ma$imum camber and .b. is the ma$imum displacement from the chord line. A cascade geometry is defined completely by the aerofoil specification, pitch3chord ratio (pitch is the spacing between two consecutive blade) and the chosen setting i.e. stagger angle is called the aerofoil camber angle i.e. %or a parabolic arc aAc J 2.7. (shown below). and aAc ; 2.7.

. %or a circular arc,

Co&#ressor Cascade
The different geometric angles, blade setting and their relationship with the flow angles for a compressor cascade are defined below.

N ; stagger angle ( positive for a compressor cascade) ; blade inlet angle ;

; blade outlet angle ; The angle of incidence .i. is the angle made by the inlet flow with the camber line. Bnder a perfect situation, the flow will leave along the camber line at the trailing edge of the blade. +ut it does not really happen so and there is a deviation which is denoted by and air outlet angle, 0ence, T ; deflection of flow ; . Thus, the air inlet angle,

T'r.ine Cascade
The different geometric angles and the blade setting of a turbine cascade are shown in the figure below.

N stagger is the stagger which is negative for a turbine cascade. To .e noted6 The difference between the orientation of Compressor cascade and that of the Turbine cascade.

3AS TURBINE
A1ial Flo+ T'r.ine
A gas turbine unit for power generation or a turbo)et engine for production of thrust primarily consists of a compressor, combustion chamber and a turbine. The air as it passes through the compressor, e$periences an increase in pressure. There after the air is fed to the combustion chamber leading to an increase in temperature. This high pressure and temperature gas is then passed through the turbine, where it is e$panded and the required power is obtained. Turbines, like compressors, can be classified into radial, a$ial and mi$ed flow machines. n the a$ial machine the fluid moves essentially in the a$ial direction through the rotor. n the radial type, the fluid motion is mostly radial. The mi$ed3flow machine is characteri#ed by a combination of a$ial and radial motion of the fluid relative to the rotor. The choice of turbine type depends on the application, though it is not always clear that any one type is superior. Comparing a$ial and radial turbines of the same overall diameter, we may say that the a$ial machine, )ust as in the case of compressors, is capable of handling considerably greater mass flow. In the other hand, for small mass flows the radial machine can be made more efficient than the a$ial one. The radial turbine

is capable of a higher pressure ratio per stage than the a$ial one. 0owever, multistaging is very much easier to arrange with the a$ial turbine, so that large overall pressure ratios are not difficult to obtain with a$ial turbines. n this chapter, we will focus on the a$ial flow turbine. >enerally the efficiency of a well3designed turbine is higher than the efficiency of a compressor. Moreover, the design process is somewhat simpler. The principal reason for this fact is that the fluid undergoes a pressure drop in the turbine and a pressure rise in the compressor.The pressure drop in the turbine is sufficient to keep the boundary layer generally well behaved, and the boundary layer separation which often occurs in compressors because of an adverse pressure gradient, can be avoided in turbines. Iffsetting this advantage is the much more critical stress problem, since turbine rotors must operate in very high temperature gas. Actual blade shape is often more dependent on stress and cooling considerations than on aerodynamic considerations, beyond the satisfaction of the velocity3triangle requirements. +ecause of the generally falling pressure in turbine flow passages, much more turning in a giving blade row is possible without danger of flow separation than in an a$ial compressor blade row. This means much more work, and considerably higher pressure ratio, per stage. n recent years advances have been made in turbine blade cooling and in the metallurgy of turbine blade materials. This means that turbines are able to operate successfully at increasingly high inlet gas temperatures and that substantial improvements are being made in turbine engine thrust, weight, and fuel consumption.

3AS TURBINE
T+o;di&ensional t%eory of a1ial flo+ t'r.ine!
An a$ial turbine stage consists of a row of stationary blades, called no##les or stators, followed by the rotor, as %igure (5.( illustrates. +ecause of the large pressure drop per stage, the no##le and rotor blades may be of increasing length, as shown, to accommodate the rapidly e$panding gases, while holding the a$ial velocity to something like a uniform value through the stage. t should be noted that the hub3tip ratio for a high pressure gas turbine in quite high, that is, it is having blades of short lengths. Thus, the radial variation in velocity and pressure may be neglected and the performance of a turbine stage is calculated from the performance of the blading at the mean radial section, which is a two3dimensional ,pitch3line design analysis ,. A low3pressure turbine will typically have a much lower hub3tip ratio and a larger blade twist. A two dimensional design is not valid in this case. n two dimensional approach the flow velocity will have two components, one a$ial and the other peripheral, denoted by subscripts .f. and respectively. The absolute velocity is denoted by 4 and the

relative velocity with respect to the impeller by . The flow conditions .(. indicates inlet to the no##le or stator vane, .1. e$it from the no##le or inlet to the rotor and .5. e$it form the rotor. Absolute angle is represented by and relative angle by as before.

Fig're $/!$ A1ial T'r.ine Stage A section through the mean radius would appear as in %igure.(5.(. Ine can see that the no##les accelerate the flow imparting an increased tangential velocity component. The velocity diagram of the turbine differs from that of the compressor in that the change in tangential velocity in the rotor, , is in the direction opposite to the blade speed B. The reaction to this change in the tangential momentum of the fluid is a torque on the rotor in the direction of motion. 0ence the fluid does work on the rotor.

3AS TURBINE
Again applying the angular momentum relation3ship, we may show that the power output as, ((5.()

n an a$ial turbine,

The work output per unit mass flow rate is

Again, !efining *e find that the stage work ratio is

((5.1)

%igure (5.1 illustrates a combined (inlet to and e$it from the rotor ) velocity diagram of a turbine stage.

Fig're $/!* Co&.ined elocity diagra& The velocity diagram gives the following relation"

Thus,

i.e,

((5.5)

The &q ((5.5) gives the e$pression for

in terms of gas angles associated with the rotor blade.

-ote that the ,work3done factor, required in the case of the a$ial compressor is unnecessary here. This is because in an accelerating flow the effect of the growth of boundary layer along the annulus passage in much less than when there is a decelerating flow with an adverge pressure gradient. nstead of temperature drop ratio Odefined in &q ((5.1)P, turbine designers generally refer to the work capacity of a turbine stage as,

((5.6)

is a dimensionless parameter, which is called the ,blade loading capacity, or ,temperature drop coefficient,. n gas turbine design, called ,the flow coefficient, . is kept generally constant across a stage and the ratio is

Thus, &q ((5.6) can be written as, ((5.7) As the boundary layer over the blade surface is not very sensitive in the case of a turbine, the turbine designer has considerably more freedom to distribute the total stage pressure drop between the rotor and the stator. 0owever, locally on the suction surface of the blade there could be a #one of an adverse pressure gradient depending on the turning and on the pitch of the blades. Thus, the boundary layer could grow rapidly or even separate in such a region affecting adversely the turbine efficiency. %igure (5.5 illustrates the schematic of flow within the blade passage and the pressure distribution over the section surface depicting a #one of diffusion. !ifferent design groups have their own rules, learned from e$perience of blade testing, for the amount of diffusion which is permissible particularly for highly loaded blades.

Fig're $/!/a Sc%e&atic diagra& of flo+ t%ro'g% a t'r.ine .lade #assage

Fig're$/!/. "ress're distri.'tion aro'nd a t'r.ine .lade

Degree of reaction
%rom the velocity diagram,

we can see that

i.e.

Then, ((6.1) The &q ((6.1) illustrates the effect of the no##le outlet angle on the impulse turbine work output. t is evident, then, that for large power output the no##le angle should be as large as possible. Two difficulties are associated with very large unit frontal area), it is evident that large . %or reasonable a$ial velocities (i.e., reasonable flow per creates very large absolute and relative velocities throughout is

the stage. 0igh losses are associated with such velocities, especially if the relative velocity supersonic. n practice, losses seem to be minimi#ed for values of around

. n addition, one can

see that for large the absolute e$haust velocity will have a swirl in the direction opposite to B. *hile we have not introduced the definition of turbine efficiency as yet, it is clear that, in a turbo)et engine where large a$ial e$haust velocity is desired, the kinetic energy associated with the tangential motion of the e$haust gases is essentially a loss. %urthermore, application of the angular momentum equation over the entire engine indicates that e$haust swirl is associate with an (undesirable) net torque acting on the aircraft. Thus the desire is for a$ial or near3a$ial absolute e$haust velocity (at least for the last stage if a multistage turbine is used). %or the special case of constant e$haust velocity and , the &q.(6.1 becomes, and a$ial

%or a given power and rotor speed, and for a given peak temperature, &q. ((6.1) is sufficient to determine appro$imately the mean blade speed (and hence radius) of a single3stage impulse turbine having a$ial outlet velocity. f , as is usually the case, the blade speed is too high (for stress limitations), or if the mean diameter is too large relative to the other engine components, it is necessary to employ a multistage turbine in which each stage does part of the work.

Degree of reaction
t has been shown that the 72Q reaction compressor stage (with constant ) has symmetrical blading. The same is true for the 72Q reaction turbine stage. As the change in static enthalpy is same in both

stator and rotor blades, the change in kinetic energy relative to each blade row must be the same. The velocity diagram for a 72Q reaction stage with a constant a$ial velocity is shown in %ig (6.1.

Fig're $5!* Fifty ;#ercent reaction t'r.ine +it% constant a1ial elocity 8ince the velocity diagram is symmetrical,

i.e or, or,

or,

((6.5)

Degree of reaction
Again the desirability of large is indicated and the same limitations are encountered, so that typical and are

values of are near . %or the special case of a$ial outlet velocity and constant #ero and the velocity diagram becomes a rectangle. The stage work output is then

Thus, for the same blade speed and for a$ial outlet velocities, the impulse stage work is twice that of the 72Q reaction stage. 0owever, we can e$pect the impulse stage to have somewhat greater loss, since the average fluid velocity in the stage is higher and the boundary layer on the suction side of the rotor blades may be significantly thicker and closer to separation, depending on the turning angle and blade spacing. The 72Q reaction stage is not uniquely desirable, of course. Ine can use any degree of reaction (greater then #ero) to design a turbine of acceptable performance. The gas flow angles at inlet and e$it of blades can be e$pressed in terms of %or the rotor blade, the relative total enthalpy remains constant and we have, and

or,

f the a$ial velocity is the same upstream and downstream of the rotor, then

The &q.((6.() becomes,

Again from the velocity triangle (%ig (5.1),

Thus,

((6.6)

((6.7) 8olving &q.(5.7 and &q.(6.7, we have ((6.9) ((6.:) and from geometric relation

((6.M) ((6.K)

0ence, from given values of

and we can estimate gas flow angles and the blade layout.

Again, combining &q.(6.7 and &q.(6.M, we have ((6.(2) *hich is the e$pression for = in terms of the e$it air angles. %or the special case of symmetrical blading , and we have case of #ero e$it swirl, (6.(2 because . %or the case of and it follows that we have , i.e. . -ow for the special and &q.

((6.(() Again for #ero e$it swirl, the blade loading capacity, &q.(5.7 reduces to

((6.(1) since &quations ((6.(() and ((6.(1) have been used in plotting %ig ((6.5), which pertains to design conditions only. 0ere we see that for a given stator outlet angle, the impulse stage requires a much higher a$ial velocity ratio than does the 72Q reaction stage. n the impulse stage all flow velocities are higher, and that is one reason why its efficiency is lower than that of the 72Q reaction stage.

Fig're $5!/ ?or2 ca#acity B and degree of reaction R of a1ial t'r.ine stages design for =ero e1it s+irl!

STA3E EFFICENC(
The aerodynamic losses in the turbine differ with the stage configuration, that is, the degree of reaction. mproved efficiency is associated with higher reaction, which tends to mean less work per stage and thus a large number of stages for a given overall pressure ratio. The understanding of aerodynamic losses is important to design, not only in the choice of blading type (impulse or reaction) but also in devising ways to control these losses, for e$ample, methods to control the clearance between the tip of the turbine blade and the outer casing wall. The choices of blade shape, aspect ratio, spacing, =eynolds number, Mach number and flow incidence angle can all affect the losses and hence the efficiency of turbine stages.

Two definitions of efficiency are in common usage" the choice between them depends on the application for which the turbine is used. %or many conventional applications, useful turbine output is in the form of shift power and the kinetic energy of the e$haust, would be conditions, is used. , is considered as a loss. n this case, ideal work , based on the inlet and e$it static

and a total to static turbine efficiency,

Thus,

((7.()

The ideal (isentropic) to actual e$pansion process in turbines is illustrated in %ig (7.(.

Fig're $8!$ T;S diagra&6 e1#ansion in a t'r.ine %urther,

((7.1)

n some applications, particularly in turbo)ets, the e$haust kinetic energy is not considered a loss since the e$haust gases are intended to emerge at high velocity. The ideal work in this case is then rather than total turbine efficiency , defined by . This requires a different definition of efficiency, the total3to3

((7.5)

Ine can compare

by making the appro$imation,

and using &qs. ((7.1) and ((7.5) it can be shown that

Thus The actual turbine work can be e$pressed as,

((7.6)

or,

STA3E EFFICENC(
!espite the development in computational methods to predict the flow field in turbine blade passages, the estimation of stage losses and thus efficiency is still a matter of considerable difficulty. n addition to the primary flow through the blade passage, there are secondary flows which move fluid across the blade passages under the action of centrifugal and coriolis forcesF blade loading effects causing incidence and deviationF leakage between the moving blade tip and the stationary shroudF the boundary layers and wakes shed by bladesF and for transonic blades, shock waves in the blade passage and at the trailing edges. Another class of effects is the unsteady generated mainly by the interaction of ad)acent blade rows. All these contribute to losses. n general, cascade tests of different blade geometries are performed.The results from these cascade tests can be correlated to define the loss coefficients for the stator (no##le) and rotor blade of turbines. *ith reference to the %ig(7.(, the effects of loss and thus irreversibility through the stator and rotor are e$pressed by differences in static enthalpies, and ,enthalpy loss coefficient, for the no##le can be defined as, respectively. -on3dimensional

8imilarly, the ,enthalpy loss coefficient, for the rotor,

Thus, the e$pressions for the efficiency can be appro$imated as"

((7.7)

((7.9)

*hile designing a turbine stage for a particular application, the restriction arises from the view point of blade stress rather than from the aerodynamics to achieve the ma$imum possible efficiency. n short, the blade speed is limited by the blade stress, particularly in high temperature applications. The turbine designers will often work to a ma$imum value of blade speed defined by temperature and material properties. Thus in modern times, the turbine blade cooling in very vital, which determines the life of an engine ( particularly for the turbo)et engine). n many applications, the characteristics of the compressor which the turbine drives also impose limits on the turbine speed.

T'r.ine "erfor&ance
%or a given design of turbine operating with a given fluid at sufficiently high =eynolds member, it can be shown from the dimensional analysis as,

where, stagnation states 21 and 25 are at the turbine inlet and outlet, respectively. %igure ((7.1) shows the overall performance of a particular single3stage turbine.

Fig're$8!* Ty#ical c%aracteristics of a t'r.ine stage Ine can see that pressure ratios greater than those for compressor stages can be obtained with satisfactory efficiency. The performance of turbines is limited principally by two factors" compressibility and stress. Compressibility limits the mass flow that can pass through a given turbine and, as we will see, stress limits the wheel speed B. The work per stage, for e$ample, depends on the square of the wheel speed.

0owever, the performance of the engine depends very strongly on the ma$imum temperature. If course, as the ma$imum temperature increases, the allowable stress level diminishesF hence in the design of the engine there must be a compromise between ma$imum temperature and ma$imum rotor tip speed B. %or given pressure ratio and adiabatic efficiency, the turbine work per unit mass is proportional to the inlet stagnation temperature. 8ince, in addition, the turbine work in a )et or turboshaft engine is commonly two or three times the useful energy output of the engine, a (Q increase in turbine inlet temperature can produce a 1Q or 5Q increase in engine output. This considerable advantage has supplied the incentive for the adoption of fairly elaborate methods for cooling the turbine no##le and rotor blades.

A brief note on >as Turbine Combustors

Iver a period of five decades, the basic factors influencing the design of combustion systems for gas turbines have not changed, although recently some new requirements have evolved. The key issues may be summari#ed as follows.

The temperature of the gases after combustion must be comparatively controlled to suit the highly
stressed turbine materials. !evelopment of improved materials and methods of blade cooling, however, has enabled permissible combustor outlet temperatures to rise from about ((22' to as much as (M72 ' for aircraft applications. G At the end of the combustion space the temperature distribution must be of known form if the turbine blades are not to suffer from local overheating. n practice, the temperature can increase with radius over the turbine annulus, because of the strong influence of temperature on allowable stress and the decrease of blade centrifugal stress from root to tip. G Combustion must be maintained in a stream of air moving with a high velocity in the region of 52392 mAs, and stable operation is required over a wide range of airAfuel ratio from full load to idling conditions. The airAfuel ratio might vary from about 92"( to (12"( for simple cycle gas turbines and from (22"( to 122"( if a heat3e$changer is used. Considering that the stoichiometric ratio is appro$imately (7"(, it is clear that a high dilution is required to maintain the temperature level dictated by turbine stresses G The formation of carbon deposits (.coking.) must be avoided, particularly the hard brittle variety. 8mall particles carried into the turbine in the high3velocity gas stream can erode the blades and block cooling air passagesF furthermore, aerodynamically e$cited vibration in the combustion chamber might cause si#eable pieces of carbon to break free resulting in even worse damage to the turbine. G n aircraft gas turbines, combustion must be stable over a wide range of chamber pressure because of the substantial change in this parameter with a altitude and forward speed. Another important requirement is the capability of relighting at high altitude in the event of an engine flame3out. G Avoidance of smoke in the e$haust is of ma)or importance for all types of gas turbineF early )et engines had very smoky e$hausts, and this became a serious problem around airports when )et transport aircraft started to operate in large numbers. 8moke trails in flight were a problem for military aircraft, permitting them to be seen from a great distance. 8tationary gas turbines are now found in urban locations, sometimes close to residential areas. G Although gas turbine combustion systems operate at e$tremely high efficiencies, they produce pollutants such as o$ides of nitrogen , carbon mono$ide (CI) and unburned hydrocarbons (B0C) and these must be controlled to very low levels. Iver the years, the performance of the gas turbine has

been improved mainly by increasing the compressor pressure ratio and turbine inlet temperature (T T). Bnfortunately this results in increased production of . &ver more stringent emissions legislation has led to significant changes in combustor design to cope with the problem. @robably the only feature of the gas turbine that eases the combustion designer.s problem is the peculiar interdependence of compressor delivery air density and mass flow which leads to the velocity of the air at entry to the combustion system being reasonably constant over the operating range. %or aircraft applications there are the additional limitations of small space and low weight, which are, however, slightly offset by somewhat shorter endurance requirements. Aircraft engine combustion chambers are normally constructed of light3gauge, heat3resisting alloy sheet (appro$. 2.M mm thick), but are only e$pected to have a life of some (2222 hours. Combustion chambers for industrial gas turbine plant may be constructed on much sturdier lines but, on the other hand, a life of about (22222 hours is required. =efractory linings are sometimes used in heavy chambers, although the remarks made above regarding the effects of hard carbon deposits breaking free apply with even greater force to refractory material.

Fig're $:!$ Co&.'stion c%a&.er +it% s+irl anes %igure (9.( indicates the schematic of a combustion chamber. The primary air is introduced through twisted radial vanes known as .swirl vanes., that results in a vorte$ motion with a low3pressure region along the a$is of the chamber. The fuel is in)ected in the same direction of air. The vorte$ motion is some time enhanced by in)ecting the secondary air through short tangential chutes in the flame tube. The burning gases tends to flow towards the region of low pressure and some portion of them swept round towards the )et of fuel as indicated by the arrow. The ob)ective is to obtain a stable flame.

Aircraft Engines and "ro#'lsion Syste&

The modern aircraft engine have the ability to actuate massive airstream and thus to produce high thrust. The engine airflow rate is perhaps 72 times the fuel flow rate, and the term air breathing engine is quite appropriate. Thus, a continuous stream of air flows through the air3breathing engine. The air is compressed, mi$ed with fuel, ignited, e$panded through a turbine and then e$pelled as the e$haust gas. The following four types of aircraft engines are generally used

G Turbo)et &ngine G Turboprop &ngine G Turbofan &ngine G =am)et &ngine At low speeds, propeller propulsion is more efficient than )et propulsion. Conventional propellers, however, inefficient and noisy at flight speeds higher than 2.7 or 2.9 times the speed of sound. n contrast, turbo)et and engines can function efficiently and quietly at flight speeds as high as 2.M7 times the speed of sound. Turbo)ets operate at supersonic flight speeds. =am)et, which is the simplest of all air3breathing engines can operate at speed than turbo)et engines and is mostly suitable for supersonic flight. become turbofan can also a higher

$! T'r.o<et Engine

Fig're $:!* T'r.o<et The turbo)et engine consists of a gas turbine, the output of which is used solely to provide power to the compressor. The compressor and the turbine are normally mounted on common shaft. Air is taken into the engine through an appro$imate diffuser duct, passes through the compressor and enters the combustions chamber, where it is mi$ed and burned with fuel. Most common fuels are hydrocarbons (Aviation kerosene). The ratio of fuel to air is determined by the ma$imum allowable gas temperature permitted by the turbine. -ormally, a considerable e$cess air is used. The hot high pressure gases are then e$panded through the turbine to a pressure which is higher than the ambient atmosphere, and yet sufficiently lower than the combustion chamber pressure, to produce )ust enough power in the turbine to run the compressor. After leaving the turbine, the gas is e$panded to the ambient pressure through an appropriate no##le. As this occurs, the gas is accelerated to a velocity, which is greater than the incoming velocity of the ingested air, and therefore produces a propulsive thrust.

*! T'r.o#ro# Engine

Fig're $>!$ T'r.o#ro# n this engine, a higher proportion of the total available pressure drop from the combustion chamber to the atmosphere is taken through the turbine and a smaller proportion through the propulsive no##le. This strategy produces e$cess power in the turbine. The e$cess power is used not only to drive the compressor, but also to drive a propeller, in the same way as in the conventional reciprocating engines. 0ere the ma)or portion of the thurst is generated by the propeller.

/! T'r.ofan Engine

Fig're $>!* T'r.ofan This is another variety of gas3turbine aircraft engine. This is very similar to the turboprop in principle, e$cept that a fan is used instead of a propeller, and this fan is contained within a duct as shown in the above figure. Most airliners use modern turbofan engines because of their high thrust and good fuel efficiency. %igure below shows the picture of +oeing :6: aircraft that uses a turbofan engine. n a turbofan engine , the air is sucked by the engine inlet. 8ome of the incoming air passes through the fan and continues on into the core compressor and then the burner, where it is mi$ed with fuel and combustion occurs. The hot e$haust passes through the core and fan turbines and then out the no##le, as in a basic turbo)et. The rest of the incoming air passes through the fan and bypasses, or goes around the engine, )ust like the air through a propeller. The air that goes through the fan has a velocity that is slightly increased from free stream. 8o a turbofan gets some of its thrust from the core and some of its thrust from the fan. The ratio of the air that goes around the engine to the air that goes through the core is called the .y#ass ratio. +ecause the fuel flow rate for the core is changed only a small amount by the addition of the fan, a turbofan generates more thrust for nearly the same amount of fuel used by the core. This means that a turbofan is very fuel efficient. n fact, high bypass ratio turbofans are nearly as fuel efficient as turboprops.

Boeing >5>6Ty#ical e1a&#le of a T'r.ofan Engine

5! Ra&<et Engine

Fig're $>!/ Ra&<et At higher forward speeds, the ram pressure of the air is already very large, and the necessity for a compressor tends to disappear. A turbo)et engine minus the compressor and turbine, but with a combustion chamber, is known as a ram)et engine. 8uch engines simply consists of (. A duct designed to diffuse the incoming air, slowing its velocity and raising its pressure 1. A combustor, designed to heat the air, normally by combustion with a liquid fuel 5. A no##le, designed to e$pand and accelerate the heated gases rearwards The ram)et engine does not accelerate itself from a standing start but requires some other form of propulsion, usually a rocket, to accelerate it to near its operating speed.

Stea& T'r.ine
Introd'ction A steam turbine converts the energy of high3pressure, high temperature steam produced by a steam generator into shaft work. The energy conversion is brought about in the following ways"
(. The high3pressure, high3temperature steam first e$pands in the no##les emanates as a high velocity fluid stream. 1. The high velocity steam coming out of the no##les impinges on the blades mounted on a wheel. The fluid stream suffers a loss of momentum while flowing past the blades that is absorbed by the rotating wheel entailing production of torque. 5. The moving blades move as a result of the impulse of steam (caused by the change of momentum) and also as a result of e$pansion and acceleration of the steam relative to them. n other words they also act as the no##les.

A steam turbine is basically an assembly of no##les fi$ed to a stationary casing and rotating blades mounted on the wheels attached on a shaft in a row3wise manner. n (M:M, a 8wedish engineer, Carl >. @. de /aval developed a simple impulse turbine, using a convergent3divergent (supersonic) no##le which ran the turbine to a ma$imum speed of (22,222 rpm. n (MK: he constructed a velocity3compounded impulse turbine (a two3row a$ial turbine with a row of guide vane stators between them. Auguste =ateau in %rance started e$periments with a de /aval turbine in (MK6, and developed the pressure compounded impulse turbine in the year (K22. n the B8A , Charles >. Curtis patented the velocity compounded de /avel turbine in (MK9 and transferred his rights to >eneral &lectric in (K2(. n &ngland , Charles A. @arsons developed a multi3stage a$ial flow reaction turbine in (MM6. 8team turbines are employed as the prime movers together with the electric generators in thermal and nuclear power plants to produce electricity. They are also used to propel large ships, ocean liners, submarines and to drive power absorbing machines like large compressors, blowers, fans and pumps. Turbines can be condensing or non3condensing types depending on whether the back pressure is below or equal to the atmosphere pressure.

Flo+ T%ro'g% No==les

A nozzle is a duct that increases the velocity of the flowing fluid at the e$pense of pressure drop. A duct which decreases the velocity of a fluid and causes a corresponding increase in pressure is a diffuser . The same duct may be either a no##le or a diffuser depending upon the end conditions across it. f the cross3section of a duct decreases gradually from inlet to e$it, the duct is said to be convergent. Conversely if the cross section increases gradually from the inlet to e$it, the duct is said to be divergent. f the cross3section initially decreases and then increases, the duct is called a convergent3divergent no##le. The minimum cross3section of such ducts is known as throat. A fluid is said to be compressible if its density changes with the change in pressure brought about by the flow. f the density does not changes or changes very little, the fluid is said to be incompressible. Bsually the gases and vapors are compressible, whereas liquids are incompressible .

No==le4 Stea& No==le and Stea& T'r.ine

STA3NATION4 SONIC "RO"ERTIES AND ISENTRO"IC E9"ANSION IN NOCCLE
The stagnation values are useful reference conditions in a compressible flow. 8uppose the properties of a flow (such as T, p, < etc.) are known at a point. The stagnation properties at a point are defined as those which are to be obtained if the local flow were imagined to cease to #ero velocity isentropically. The stagnation values are denoted by a subscript #ero. Thus, the stagnation enthalpy is defined as

%or a calorically perfect gas, this yields, ((M.()

which defines the stagnation temperature. t is meaningful to e$press the ratio of

in the form

((M.1) or, f we know the local temperature (T) and Mach number (Ma), we can fine out the stagnation temperature . Consequently, isentropic relations can be used to obtain stagnation pressure and stagnation density as.

((M.5)

((M.6)

n general, the stagnation properties can vary throughout the flow field.

0owever, if the flow is adiabatic, then

is constant throughout the flow. t follows that the

and are constant throughout an adiabatic flow, even in the presence of friction. 0ere a is the speed of sound and the suffi$ signifies the stagnation condition. t is understood that all stagnation properties are constant along an isentropic flow. f such a flow starts from a large reservoir where the fluid is practically at rest, then the properties in the reservoir are equal to the stagnation properties everywhere in the flow (%ig. (M.().

Fig $@!$

An isentro#ic #rocess starting fro& a reser oir

There is another set of conditions of comparable usefulness where the flow is sonic, Ma;(.2. These sonic, or critical properties are denoted by asterisks" and. . These properties are attained if the local fluid is imagined to e$pand or compress isentropically until it reachers Ma;(. *e have already discussed that the total enthalpy, hence , is conserved so long the process is and density

adiabatic, irrespective of frictional effects. n contrast, the stagnation pressure decrease if there is friction. %rom &q.((M.(), we note that

((M.7a) or, is the relationship between the fluid velocity and local temperature (T), in an adiabatic flow. The flow can attain a ma$imum velocity of

((M.7b)

As it has already been stated, the unity Mach number, Ma;(, condition is of special significance in compressible flow, and we can now write from &q.((M.1), ((M.5) and ((M.6).

((M.9a)

((M.9b)

((M.9c)

%or diatomic gases, like air

, the numerical values are

The fluid velocity and acoustic speed are equal at sonic condition and is ((M.:a)

( (M.:b ) or, *e shall employ both stagnation conditions and critical conditions as reference conditions in a variety of one dimensional compressible flows.

Effect of Area 0ariation on Flo+ "ro#erties in Isentro#ic Flo+

n considering the effect of area variation on flow properties in isentropic flow, we shall concern ourselves primarily with the velocity and pressure. *e shall determine the effect of change in area, A, on the velocity ", and the pressure p. %rom +ernoulli.s equation, we can write

or,

!ividing by

, we obtain

((K.()

A convenient differential form of the continuity equation can be obtained from &q. ((6.72) as

8ubstituting from &q. ((K.(),

or,

((K.1)

nvoking the relation (

) for isentropic process in &q. ((K.1), we get

((K.5)

%rom &q. ((K.5), we see that for MaJ( an area change causes a pressure change of the same sign, i.e. positive dA means positive dp for MaJ(. %or MaH(, an area change causes a pressure change of opposite sign. Again, substituting from &q.((K.() into &q. ((K.5), we obtain ((K.6) %rom &q. ((K.6), we see that MaJ( an area change causes a velocity change of opposite sign, i.e. positive dA means negative d" for MaJ(. %or MaH(, an area change causes a velocity change of same sign. These results are summari#ed in %ig.(K.(, and the relations ((K.5) and ((K.6) lead to the following important conclusions about compressible flows"

1.

At subsonic speeds (MaJ() a decrease in area increases the speed of flow. A subsonic no##le should have a convergent profile and a subsonic diffuser should possess a divergent profile. The flow behaviour in the regime of MaJ( is therefore qualitatively the same as in incompressible flows. 1. n supersonic flows (MaH(), the effect of area changes are different. According to &q. ((K.6), a supersonic no##le must be built with an increasing area in the flow direction. A supersonic diffuser must be a converging channel. !ivergent no##les are used to produce supersonic flow in missiles and launch vehicles.

Fig $A!$

S%a#es of no==les and diff'sersin s'.sonic and s'#ersonic regi&es

8uppose a no##le is used to obtain a supersonic stream staring from low speeds at the inlet (%ig.(K.1). Then the Mach number should increase from Ma;2 near the inlet to MaH( at the e$it. t is clear that the no##le must converge in the subsonic portion and diverge in the supersonic portion. 8uch a no##le is called a convergent-divergent nozzle. A convergent3divergent no##le is also called a de 5aval nozzle,

after Carl >.@. de /aval who first used such a configuration in his steam turbines in late nineteenth century (this has already been mentioned in the introductory note). %rom %ig.$A!* it is clear that the Mach number must be unity at the throat, where the area is neither increasing nor decreasing. This is consistent with &q. ((K.6) which shows that d" can be non3#ero at the throat only if Ma;(. t also follows that the sonic velocity can be achieved only at the throat of a no##le or a diffuser.

Fig $A!* A con ergent;di ergent no==le

The condition, however, does not restrict that +a must necessarily be unity at the throat, According to &q. ((K.6), a situation is possible where at the throat if d";2 there. %or an e$ample, the flow in a convergent3divergent duct may be subsonic everywhere with +a increasing in the convergent portion and decreasing in the divergent portion with at the throat (see %ig.(K.5). The first part of the duct is acting as a no##le, whereas the second part is acting as a diffuser. Alternatively, we may have a convergent3divergent duct in which the flow is supersonic everywhere with +a decreasing in the convergent part and increasing in the divergent part and again at the throat (see %ig. (K.6).

Fig $A!/ Con ergent;di ergent d'ct +it%

at t%roat

Fig $A!5

Con ergent;di ergent d'ct +it%

at t%roat

Iscentro#ic Flo+ of a a#or or gas t%ro'g% a no==le

%irst law of thermodynamics"

(if

where

is enthalpy drop across the no##le Again we know, Tds ; dh 3 dp %or the isentropic flow, dh ; dp

or, (12.() or, Assuming that the pressure and volume of steam during e$pansion obey the law p n ; constant, where n is the isentropic inde$

(12.1)

-ow, mass flow rate

Therefore, the mass flow rate at the e$it of the no##le

(12.5)

The e$it pressure, p1 determines the

for a given inlet condition. The mass flow rate is ma$imum when,

%or ma$imum

(12.6)

n;

(.6,

for diatomic gases for super saturated steam for dry saturated steam for wet steam with dryness fraction $

%or ,

(72Qdrop in inlet pressure)

f we compare this with the results of sonic properties, as described in the earlier section, we shall observe that the critical pressure occurs at the throat for Ma ; (. The critical pressure ratio is defined as the ratio of pressure at the throat to the inlet pressure, for checked flow when Ma ; (

Stea& No==les

Fig're *$!$ S'#er Sat'rated E1#ansion of Stea& in a No==le

The process (31 is the isentropic e$pansion. The change of phase will begin to occur at point 1 vapour continues to e$pand in a dry state 8team remains in this unnatural superheated state untit its density is about eight times that of the saturated vapour density at the same pressure *hen this limit is reached, the steam will suddenly condense @oint 5 is achieved by e$tension of the curvature of constant pressure line from the superheated region which strikes the vertical e$pansion line at 5 and through which *ilson line also passes. The point 5 corresponds to a metastable equilibrium state of the vapour. The process 135 shows e$pansion under super3saturation condition which is not in thermal equilibrium t is also called under cooling At any pressure between and i.e., within the superheated #one, the temperature of the vapous is lower than the saturation temperature corresponding to that pressure 8ince at 5, the limit of supersaturation is reached, the steam will now condense instantaneously to its normal state at the constant pressure, and constant enthalpy which is shown by the hori#ontal line where is on normal wet area pressure line of the same pressure .

is again isentropic, e$pansion in thermal equilibrium. To be noted that 6 and are on the same pressure line. Thus the effect of supersaturation is to reduce the enthalpy drop slightly during the e$pansion and consequently a corresponding reduction in final velocity. The final dryness fraction and entropy are also increased and the measured discharge is greater than that theoretically calculated.

!egree of super heat ; ; limiting saturation pressure ; saturation pressure at temperature degree of undercooling 3 3 shown on T3s diagram

is the saturation temperature at ; 8upersaturated steam temperature at point 5 which is the limit of supersaturation.

(1(.()

(1(.1)

8upersaturated vapour behaves like supersaturated steam and the inde$ to e$pansion,

STEAM TURBINES
T'r.ines
*e shall consider steam as the working fluid 8ingle stage or Multistage A$ial or =adial turbines Atmospheric discharge or discharge below atmosphere in condenser mpulseAand =eaction turbine

I&#'lse T'r.ines
mpulse turbines (single3rotor or multirotor) are simple stages of the turbines. 0ere the impulse blades are attached to the shaft. mpulse blades can be recogni#ed by their shape. They are usually symmetrical and have entrance and e$it angles respectively, around 12 L . +ecause they are usually used in the entrance high3pressure stages of a steam turbine, when the specific volume of steam is low and requires much smaller flow than at lower pressures, the impulse blades are short and have constant cross sections.

T%e Single;Stage I&#'lse T'r.ine

The single-stage impulse turbine is also called the de 5aval turbine after its inventor. The turbine consists of a single rotor to which impulse blades are attached. The steam is fed through one or several convergent3divergent no##les which do not e$tend completely around the circumference of the rotor, so that only part of the blades is impinged upon by the steam at any one time. The no##les also allow governing of the turbine by shutting off one or more them. The velocity diagram for a single3stage impulse has been shown in %ig. 11.(. %igure 11.1 shows the velocity diagram indicating the flow through the turbine blades.

Fig're **!$ Sc%e&atic diagra& of an I&#'lse Tr'.ine and ; nlet and outlet absolute velocity

and

; nlet and outlet relative velocity (4elocity relative to the rotor blades.)

B ; mean blade speed ; no##le angle, ; absolute fluid angle at outlet

t is to be mentioned that all angles are with respect to the tangential velocity ( in the direction of B )

Fig're **!* 0elocity diagra& of an I&#'lse T'r.ine

and and

; nlet and outlet .lade angles ; Tangential or whirl component of absolute velocity at inlet and outlet

and

; A$ial component of velocity at inlet and outlet

Tangential force on a blade, (11.() (mass flow rate U change in velocity in tangential direction) or, (11.1)

@ower developed ; +lade efficiency or !iagram efficiency or Btili#ation factor is given by

(11.5)

or,

(11.6)

(15.() stage efficiency

(15.1) or, or, Iptimum blade speed of a single stage turbine (15.5)

(15.6)

where,

; friction coefficient

; +lade speed ratio

(15.7)

is ma$imum when

also

or, (15.9) or, is of the order of (M2 to 112

-ow,

(%or single stage impulse turbine)

The ma$imum value of blade efficiency

(15.:)

%or equiangular blades,

(15.M)

f the friction over blade surface is neglected (15.K)

Co&#o'nding in I&#'lse T'r.ine

f high velocity of steam is allowed to flow through one row of moving blades, it produces a rotor speed of about 52222 rpm which is too high for practical use. t is therefore essential to incorporate some improvements for practical use and also to achieve high performance. This is possible by making use of more than one set of no##les, and rotors, in a series, keyed to the shaft so that either the steam pressure or the )et velocity is absorbed by the turbine in stages. This is called compounding. Two types of compounding can be accomplished" (a) velocity compounding and (b) pressure compounding &ither of the above methods or both in combination are used to reduce the high rotational speed of the single stage turbine.

T%e 0elocity ; Co&#o'nding of t%e I&#'lse T'r.ine

The velocity3compounded impulse turbine was first proposed by C.>. Curtis to solve the problems of a single3stage impulse turbine for use with high pressure and temperature steam. The 3urtis stage turbine, as it came to be called, is composed of one stage of no##les as the single3stage turbine, followed by two rows of moving blades instead of one. These two rows are separated by one row of fi$ed blades attached to the turbine stator, which has the function of redirecting the steam leaving the first row of moving blades to the second row of moving blades. A Curtis stage impulse turbine is shown in %ig. 15.( with schematic pressure and absolute steam3velocity changes through the stage. n the Curtis stage, the total enthalpy drop and hence pressure drop occur in the no##les so that the pressure remains constant in all three rows of blades.

Fig're */!$ 0elocity Co&#o'nding arrange&ent 4elocity is absorbed in two stages. n fi$ed (static) blade passage both pressure and velocity remain constant. %i$ed blades are also called guide vanes. 4elocity compounded stage is also called C'rtis stage! The velocity diagram of the velocity3compound mpulse turbine is shown in %igure 15.1.

Fig're */!* 0elocity diagra&s for t%e 0elocity;Co&#o'nded I&#'lse t'r.ine The fi$ed blades are used to guide the outlet steamAgas from the previous stage in such a manner so as to smooth entry at the ne$t stage is ensured. ', the blade velocity coefficient may be different in each row of blades

*ork done ;

(15.(2)

&nd thrust ;

(15.(()

The optimum velocity ratio will depend on number of stages and is given by

 *ork is not uniformly distributed ((st H1nd ) The fist stage in a large (power plant) turbine is velocity or pressure compounded impulse stage. "ress're Co&#o'nding or Ratea' Staging
T%e "ress're ; Co&#o'nded I&#'lse T'r.ine
To alleviate the problem of high blade velocity in the single3stage impulse turbine, the total enthalpy drop through the no##les of that turbine are simply divided up, essentially in an equal manner, among many single3stage impulse turbines in series (%igure 16.(). 8uch a turbine is called a Rateau turbine , after its inventor. Thus the inlet steam velocities to each stage are essentially equal and due to a reduced Vh.

Fig're *5!$ "ress're;Co&#o'nded I&#'lse T'r.ine

"ress're dro# 3 takes place in more than one row of no##les and the increase in kinetic energy after each no##le is held within limits. Bsually convergent no##les are used *e can write

(16.()

(16.1)

where

is carry over coefficient

Reaction T'r.ine
A reaction turbine, therefore, is one that is constructed of rows of fi$ed and rows of moving blades. The fi$ed blades act as no##les. The moving blades move as a result of the impulse of steam received (caused by change in momentum) and also as a result of e$pansion and acceleration of the steam relative to them. n other words, they also act as no##les. The enthalpy drop per stage of one row fi$ed and one row moving blades is divided among them, often equally. Thus a blade with a 72 percent degree of reaction, or a 72 percent reaction stage, is one in which half the enthalpy drop of the stage occurs in the fi$ed blades and half in the moving blades. The pressure drops will not be equal, however. They are greater for the fi$ed blades and greater for the high3pressure than the low3pressure stages. The moving blades of a reaction turbine are easily distinguishable from those of an impulse turbine in that they are not symmetrical and, because they act partly as no##les, have a shape similar to that of the fi$ed blades, although curved in the opposite direction. The schematic pressure line (%ig. 16.1) shows that pressure continuously drops through all rows of blades, fi$ed and moving. The absolute steam velocity changes within each stage as shown and repeats from stage to stage. %igure 16.5 shows a typical velocity diagram for the reaction stage.

Fig're *5!* T%ree stages of reaction t'r.ine indicating #ress're and elocity distri.'tion @ressure and enthalpy drop both in the fi$ed blade or stator and in the moving blade or Rotor

!egree of =eaction ; (16.5) or, A very widely used design has half degree of reaction or 72Q reaction and this is known as @arson.s Turbine. This consists of symmetrical stator and rotor blades.

Fig're *5!/ T%e elocity diagra& of reaction .lading

The velocity triangles are symmetrical and we have

&nergy input per stage (unit mass flow per second)

(16.6)

(16.7) %rom the inlet velocity triangle we have,

*ork done (for unit mass flow per second) (16.9) Therefore, the +lade efficiency

(16.:)

Reaction T'r.ine4 Contin'ed

@ut

then

(17.()

%or the ma$imum efficiency

and we get (17.1)

from which finally it yields

(17.5)

Fig're *8!$ 0elocity diagra& for &a1i&'& efficiency Absolute velocity of the outlet at this stage is a$ial (see figure 17.(). n this case, the energy transfer (17.6)

can be found out by putting the value of

in the e$pression for blade efficiency

(17.7) (17.9) is greater in reaction turbine. &nergy input per stage is less, so there are more number of stages.

Stage Efficiency and Re%eat factor

The Thermodynamic effect on the turbine efficiency can be best understood by considering a number of stages between two stages ( and 1 as shown in %igure 17.1

Fig're *8!* Different stage of a stea& t'r.ine The total e$pansion is divided into four stages of the same efficiency and pressure ratio.

(17.:)

The overall efficiency of e$pansion is

. The actual work during the e$pansion from ( to 1 is

or,

(17.M)

=eheat factor (=.%.);

or,

(17.K)

=.% is (.25 to (.26 f remains same for all the stages or is the mean stage efficiency.

(17.(2)

(17.(() or,

*e can see" (17.(1) This makes the overall efficiency of the turbine greater than the individual stage efficiency. The effect depicted by &qn (17.(1) is due to the thermodynamic effect called ,reheat,. This does not imply any heat transfer to the stages from outside. t is merely the reappearance of stage losses an increased enthalpy during the constant pressure heating (or reheating) processes AU, +W, CX and !1.

IM"ULSE TURBINE

Fig're *:!$ Ty#ical "ELTON ?HEEL +it% *$ B'c2ets 0ydropower is the longest established source for the generation of electric power. n this module we shall discuss the governing principles of various types of hydraulic turbines used in hydro3electric power stations. Impulse Hydraulic Turbine 6 T%e "elton ?%eel The only hydraulic turbine of the impulse type in common use, is named after an American engineer /aster A @elton, who contributed much to its development around the year (MM2. Therefore this machine is known as @elton turbine or @elton wheel. t is an efficient machine particularly suited to high heads. The rotor consists of a large circular disc or wheel on which a number (seldom less than (7) of spoon shaped buckets are spaced uniformly round is periphery as shown in %igure 19.(. The wheel is driven by )ets of water being discharged at atmospheric pressure from pressure no##les. The no##les are mounted so that each directs a )et along a tangent to the circle through the centres of the buckets (%igure 19.1). !own the centre of each bucket, there is a splitter ridge which divides the )et into two equal streams which flow round the smooth inner surface of the bucket and leaves the bucket with a relative velocity almost opposite in direction to the original )et.

Fig're *:!* A "elton +%eel %or ma$imum change in momentum of the fluid and hence for the ma$imum driving force on the wheel, the deflection of the water )et should be . n practice, however, the deflection is limited to about

so that the water leaving a bucket may not hit the back of the following bucket. Therefore, the camber angle of the buckets is made as . %igure(19.5a)

The number of )ets is not more than two for hori#ontal shaft turbines and is limited to si$ for vertical shaft turbines. The flow partly fills the buckets and the fluid remains in contact with the atmosphere. Therefore, once the )et is produced by the no##le, the static pressure of the fluid remains atmospheric throughout the machine. +ecause of the symmetry of the buckets, the side thrusts produced by the fluid in each half should balance each other. Analysis of force on the bucket and power generation %igure 19.5a shows a section through a bucket which is being acted on by a )et. The plane of section is parallel to the a$is of the wheel and contains the a$is of the )et. The absolute velocity of the )et with which it strikes the bucket is given by

Fig're *:!/

,a-Flo+ along t%e .'c2et of a #elton +%eel ,.- Inlet elocity triangle ,c-O'tlet elocity triangle

where, is the coefficient of velocity which takes care of the friction in the no##le. H is the head at the entrance to the no##le which is equal to the total or gross head of water stored at high altitudes minus the head lost due to friction in the long pipeline leading to the no##le. /et the velocity of the bucket (due to the rotation of the wheel) at its centre where the )et strikes be 4 . 8ince the )et velocity is tangential, i.e. and 4 are collinear, the diagram of velocity vector at inlet (%ig 19.5.b) becomes simply a straight line and the relative velocity is given by

t is assumed that the flow of fluid is uniform and it glides the blade all along including the entrance and e$it sections to avoid the unnecessary losses due to shock. Therefore the direction of relative velocity at entrance and e$it should match the inlet and outlet angles of the buckets respectively. The velocity triangle at the outlet is shown in %igure 19.5c. The bucket velocity 4 remains the same both at the inlet and outlet. *ith the direction of 4 being taken as positive, we can write. The tangential component of inlet velocity (%igure 19.5b)

and the tangential component of outlet velocity (%igure 19.5c)

where and are the velocities of the )et relative to the bucket at its inlet and outlet and outlet angle of the bucket.

is the

%rom the &q. ((.1) (the &uler.s equation for hydraulic machines), the energy delivered by the fluid per unit mass to the rotor can be written as

(19.()

(since, in the present situation,

The relative velocity becomes slightly less than mainly because of the friction in the bucket. 8ome additional loss is also inevitable as the fluid strikes the splitter ridge, because the ridge cannot have #ero thickness. These losses are however kept to a minimum by making the inner surface of the bucket polished and reducing the thickness of the splitter ridge. The relative velocity at outlet is usually

e$pressed as where, ' is a factor with a value less than (. 0owever in an ideal case ( in absence of friction between the fluid and blade surface) ';(. Therefore, we can write &q.(19.() (19.1) f D is the volume flow rate of the )et, then the power transmitted by the fluid to the wheel can be written as

(19.5) The power input to the wheel is found from the kinetic energy of the )et arriving at the wheel and is given by . Therefore the wheel efficiency of a pelton turbine can be written as

(19.6)

t is found that the efficiency constant values of the condition given by

depends on

and

%or a given design of the bucket, i.e. for only, and we can determine

and ', the efficiency at which

becomes a function of

becomes ma$imum.

%or

to be ma$imum,

or,

(19.7)

is always negative. Therefore, the ma$imum wheel efficiency can be written after substituting the relation given by eqn.(19.7) in eqn.(19.6) as (19.9)

The condition given by &q. (19.7) states that the efficiency of the wheel in converting the kinetic energy of the )et into mechanical energy of rotation becomes ma$imum when the wheel speed at the centre of the bucket becomes one half of the incoming velocity of the )et. The overall efficiency will be less

than because of friction in bearing and windage, i.e. friction between the wheel and the atmosphere in which it rotates. Moreover, as the losses due to bearing friction and windage increase rapidly with speed, the overall efficiency reaches it peak when the ratio is slightly less than the theoretical value of 2.7. The value usually obtained in practice is about 2.69. The %igure 1:.( shows the variation of wheel efficiency with blade to )et speed ratio for assumed values at k;( and 2.M, and

. An overall efficiency of M73K2 percent may usually be obtained in large machines. To obtain high values of wheel efficiency, the buckets should have smooth surface and be properly designed. The length, width, and depth of the buckets are chosen about 1.7.6 and 2.M times the )et diameter. The buckets are notched for smooth entry of the )et.

Fig're *>!$ T%eoretical ariation of +%eel efficiency for a "elton t'r.ine +it% .lade s#eed to <et s#eed ratio for different al'es of 2

Specific speed and wheel geometry . The specific speed of a pelton wheel depends on the ratio of )et diameter d and the wheel pitch diameter. $ (the diameter at the centre of the bucket). f the hydraulic efficiency of a pelton wheel is defined as the ratio of the power delivered ( to the wheel to the head available H at the no##le entrance, then we can write.

(1:.()

8ince O

and

The specific speed

The optimum value of the overall efficiency of a @elton turbine depends both on the values of the specific speed and the speed ratio. The @elton wheels with a single )et operate in the specific speed range of 63 (9, and therefore the ratio !Ad lies between 9 to 19 as given by the &q. ((7.17b). A large value of !Ad reduces the rpm as well as the mechanical efficiency of the wheel. t is possible to increase the specific speed by choosing a lower value of !Ad, but the efficiency will decrease because of the close spacing of buckets. The value of !Ad is normally kept between (6 and (9 to maintain high efficiency. The number of buckets required to maintain optimum efficiency is usually fi$ed by the empirical relation.

(1:.1) n(number of buckets) = Govering of elton Turbine 6 %irst let us discuss what is meant by governing of turbines in general. *hen a turbine drives an electrical generator or alternator, the primary requirement is that the rotational speed of the shaft and hence that of the turbine rotor has to be kept fi$ed. Itherwise the frequency of the electrical output will be altered. +ut when the electrical load changes depending upon the demand, the speed of the turbine changes automatically. This is because the e$ternal resisting torque on the shaft is altered while the driving torque due to change of momentum in the flow of fluid through the turbine remains the same. %or e$ample, when the load is increased, the speed of the turbine decreases and vice versa . A constancy in speed is therefore maintained by ad)usting the rate of energy input to the turbine accordingly. This is usually accomplished by changing the rate of fluid flow through the turbine3 the flow in increased when the load is increased and the flow is decreased when the load is decreased. This ad)ustment of flow with the load is known as the governing of turbines. n case of a @elton turbine, an additional requirement for its operation at the condition of ma$imum efficiency is that the ration of bucket to initial )et velocity has to be kept at its optimum value of

about 2.69. 0ence, when B is fi$ed. has to be fi$ed. Therefore the control must be made by a variation of the cross3sectional area, A, of the )et so that the flow rate changes in proportion to the change in the

flow area keeping the )et velocity same. This is usually achieved by a spear valve in the no##le (%igure 1:.1a). Movement of the spear and the a$is of the no##le changes the annular area between the spear and the housing. The shape of the spear is such, that the fluid coalesces into a circular )et and then the effect of the spear movement is to vary the diameter of the )et. !eflectors are often used (%igure 1:.1b) along with the spear valve to prevent the serious water hammer problem due to a sudden reduction in the rate of flow. These plates temporarily defect the )et so that the entire flow does not reach the bucketF the spear valve may then be moved slowly to its new position to reduce the rate of flow in the pipe3line gradually. f the bucket width is too small in relation to the )et diameter, the fluid is not smoothly deflected by the buckets and, in consequence, much energy is dissipated in turbulence and the efficiency drops considerably. In the other hand, if the buckets are unduly large, the effect of friction on the surfaces is unnecessarily high. The optimum value of the ratio of bucket width to )et diameter has been found to vary between 6 and 7.

Fig're *>!*

,a- S#ear al e to alter <et area in a "elton +%eel ,.- 7et deflected fro& .'c2et

Li&itation of a "elton T'r.ine6 The @elton wheel is efficient and reliable when operating under large heads. To generate a given output power under a smaller head, the rate of flow through the turbine has to be higher which requires an increase in the )et diameter. The number of )ets are usually limited to 6 or 9 per wheel. The increases in )et diameter in turn increases the wheel diameter. Therefore the machine becomes unduly large, bulky and slow3running. n practice, turbines of the reaction type are more suitable for lower heads.

Francis T'r.ine
Reaction T'r.ine6 The principal feature of a reaction turbine that distinguishes it from an impulse turbine is that only a part of the total head available at the inlet to the turbine is converted to velocity head, before the runner is reached. Also in the reaction turbines the working fluid, instead of engaging only one or two blades, completely fills the passages in the runner. The pressure or static head of the fluid changes gradually as it passes through the runner along with the change in its kinetic energy based on absolute velocity due to the impulse action between the fluid and the runner. Therefore the cross3sectional area of flow through the passages of the fluid. A reaction turbine is usually well suited for low heads. A radial flow

hydraulic turbine of reaction type was first developed by an American &ngineer, Eames +. %rancis ((M(73 K1) and is named after him as the %rancis turbine. The schematic diagram of a %rancis turbine is shown in %ig. 1M.(

Fig're *@!$ A Francis t'r.ine A %rancis turbine comprises mainly the four components" (i) sprical casing, (ii) guide on stay vanes, (iii) runner blades, (iv) draft3tube as shown in %igure 1M.( . S#iral Casing 6 Most of these machines have vertical shafts although some smaller machines of this type have hori#ontal shaft. The fluid enters from the penstock (pipeline leading to the turbine from the reservoir at high altitude) to a spiral casing which completely surrounds the runner. This casing is known as scroll casing or volute. The cross3sectional area of this casing decreases uniformly along the circumference to keep the fluid velocity constant in magnitude along its path towards the guide vane.

Fig're *@!* S#iral Casing This is so because the rate of flow along the fluid path in the volute decreases due to continuous entry of the fluid to the runner through the openings of the guide vanes or stay vanes. 3'ide or Stay ane6 The basic purpose of the guide vanes or stay vanes is to convert a part of pressure energy of the fluid at its entrance to the kinetic energy and then to direct the fluid on to the runner blades at the angle appropriate to the design. Moreover, the guide vanes are pivoted and can be turned by a suitable governing mechanism to regulate the flow while the load changes. The guide vanes are also known as wicket gates. The guide vanes impart a tangential velocity and hence an angular momentum to the water before its entry to the runner. The flow in the runner of a %rancis turbine is not purely radial but a combination of radial and tangential. The flow is inward, i.e. from the periphery towards the centre. The height of the runner depends upon the specific speed. The height increases with the increase in the specific speed. The main direction of flow change as water passes through the runner and is finally turned into the a$ial direction while entering the draft tube. Draft t'.e6 The draft tube is a conduit which connects the runner e$it to the tail race where the water is being finally discharged from the turbine. The primary function of the draft tube is to reduce the velocity of the discharged water to minimi#e the loss of kinetic energy at the outlet. This permits the turbine to be set above the tail water without any appreciable drop of available head. A clear understanding of the function of the draft tube in any reaction turbine, in fact, is very important for the purpose of its design. The purpose of providing a draft tube will be better understood if we carefully study the net available head across a reaction turbine. Net %ead across a reaction t'r.ine and t%e #'r#ose to #ro iding a draft t'.e . The effective head across any turbine is the difference between the head at inlet to the machine and the head at outlet from it. A reaction turbine always runs completely filled with the working fluid. The tube that connects the end of the runner to the tail race is known as a draft tube and should completely to filled with the working fluid flowing through it. The kinetic energy of the fluid finally discharged into the tail race is wasted. A draft tube is made divergent so as to reduce the velocity at outlet to a minimum. Therefore a draft tube is basically a diffuser and should be designed properly with the angle between the walls of the tube to be limited to

about M degree so as to prevent the flow separation from the wall and to reduce accordingly the loss of energy in the tube. %igure 1M.5 shows a flow diagram from the reservoir via a reaction turbine to the tail race. The total head at the entrance to the turbine can be found out by applying the +ernoulli.s equation between the free surface of the reservoir and the inlet to the turbine as

(1M.()

or,

(1M.1)

where is the head lost due to friction in the pipeline connecting the reservoir and the turbine. 8ince the draft tube is a part of the turbine, the net head across the turbine, for the conversion of mechanical work, is the difference of total head at inlet to the machine and the total head at discharge from the draft tube at tail race and is shown as H in %igure 1M.5

Fig're *@!/ Head across a reaction t'r.ine Therefore, H ; total head at inlet to machine (() 3 total head at discharge (5)

(1M.5)

(1M.6)

The pressures are defined in terms of their values above the atmospheric pressure. 8ection 1 and 5 in %igure 1M.5 represent the e$its from the runner and the draft tube respectively. f the losses in the draft tube are neglected, then the total head at 1 becomes equal to that at 5. Therefore, the net head across

the machine is either or . Applying the +ernoull.s equation between 1 and 5 in consideration of flow, without losses, through the draft tube, we can write.

(1M.7)

(1M.9)

8ince , both the terms in the bracket are positive and hence is always negative, which implies that the static pressure at the outlet of the runner is always below the atmospheric pressure. &quation (1M.() also shows that the value of the suction pressure at runner outlet depends on #, the height of the runner above the tail race and , the decrease in kinetic energy of the fluid in

the draft tube. The value of this minimum pressure should never fall below the vapour pressure of the liquid at its operating temperature to avoid the problem of cavitation. Therefore, we fine that the incorporation of a draft tube allows the turbine runner to be set above the tail race without any drop of available head by maintaining a vacuum pressure at the outlet of the runner.

R'nner of t%e Francis T'r.ine

The shape of the blades of a %rancis runner is comple$. The e$act shape depends on its specific speed. t is obvious from the equation of specific speed that higher specific speed means lower head. This requires that the runner should admit a comparatively large quantity of water for a given power output and at the same time the velocity of discharge at runner outlet should be small to avoid cavitation. n a purely radial flow runner, as developed by Eames +. %rancis, the bulk flow is in the radial direction. To be more clear, the flow is tangential and radial at the inlet but is entirely radial with a negligible tangential component at the outlet. The flow, under the situation, has to make a K2o turn after passing through the rotor for its inlet to the draft tube. 8ince the flow area (area perpendicular to the radial direction) is small, there is a limit to the capacity of this type of runner in keeping a low e$it velocity. This leads to the design of a mi$ed flow runner where water is turned from a radial to an a$ial direction in the rotor itself. At the outlet of this type of runner, the flow is mostly a$ial with negligible radial and tangential components. +ecause of a large discharge area (area perpendicular to the a$ial direction), this type of runner can pass a large amount of water with a low e$it velocity from the runner. The blades for a reaction turbine are always so shaped that the tangential or whirling component of velocity at the outlet becomes #ero . This is made to keep the kinetic energy at outlet a minimum. %igure 1K.( shows the velocity triangles at inlet and outlet of a typical blade of a %rancis turbine. Bsually the flow velocity (velocity perpendicular to the tangential direction) remains constant throughout, i.e. and is equal to that at the inlet to the draft tube. The &uler.s equation for turbine O&q.((.1)P in this case reduces to (1K.()

where, e is the energy transfer to the rotor per unit mass of the fluid. %rom the inlet velocity triangle shown in %ig. 1K.( (1K.1a) and (1K.1b)

8ubstituting the values of

and

from &qs. (1K.1a) and (1K.1b) respectively into &q. (1K.(), we have (1K.5)

Fig're *A!$ 0elocity triangle for a Francis r'nner

The loss of kinetic energy per unit mass becomes equal to blade efficiency becomes

. Therefore neglecting friction, the

since

can be written as

The change in pressure energy of the fluid in the rotor can be found out by subtracting the change in its kinetic energy from the total energy released. Therefore, we can write for the degree of reaction.

Osince

Bsing the e$pression of e from &q. (1K.5), we have

(1K.6)

The inlet blade angle

of a %rancis runner varies

and the guide vane angle angle

from

. The ratio of blade width to the diameter of runner +A!, at blade inlet, depends upon the required specific speed and varies from (A12 to 1A5. &$pression for specific speed. The dimensional specific speed of a turbine, can be written as

@ower generated ( for a turbine can be e$pressed in terms of available head H and hydraulic efficiency as

0ence, it becomes (1K.7)

Again, 8ubstituting

, from &q. (1K.1b)

(1K.9)

Available head H equals the head delivered by the turbine plus the head lost at the e$it. Thus,

since

with the help of &q. (1K.5), it becomes

or,

(1K.:)

8ubstituting the values of 0 and - from &qs (1K.:) and (1K.9) respectively into the e$pression by &q. (1K.7), we get,

given

%low velocity at inlet

can be substituted from the equation of continuity as

where + is the width of the runner at its inlet

%inally, the e$pression for

becomes,

(1K.M)

%or a %rancis turbine, the variations of geometrical parameters like have been described earlier. These variations cover a range of specific speed between 72 and 622. %igure 1K.1 shows an overview of a %rancis Turbine. The figure is specifically shown in order to convey the si#e and relative dimensions of a typical %rancis Turbine to the readers.

Fig're *A!* Installation of a Francis T'r.ine

DA"LAN TURBINE Introd'ction

0igher specific speed corresponds to a lower head. This requires that the runner should admit a comparatively large quantity of water. %or a runner of given diameter, the ma$imum flow rate is achieved when the flow is parallel to the a$is. 8uch a machine is known as a$ial flow reaction turbine. An Australian engineer, 4ikton 'aplan first designed such a machine. The machines in this family are called 'aplan Turbines.(%igure 52.()

Fig're /E!$ A ty#ical Da#lan T'r.ine

De elo#&ent of Da#lan R'nner fro& t%e C%ange in t%e S%a#e of Francis R'nner +it% S#ecific S#eed
%igure 52.1 shows in stages the change in the shape of a %rancis runner with the variation of specific speed. The first three types O%ig. 52.1 (a), (b) and (c)P have, in order. The %rancis runner (radial flow runner) at low, normal and high specific speeds. As the specific speed increases, discharge becomes more and more a$ial. The fourth type, as shown in %ig.52.1 (d), is a mi$ed flow runner (radial flow at inlet a$ial flow at outlet) and is known as !ubs runner which is mainly suited for high specific speeds. %igure 52.1(e) shows a propeller type runner with a less number of blades where the flow is entirely a$ial (both at inlet and outlet). This type of runner is the most suitable one for very high specific speeds and is known as 'aplan runner or a$ial flow runner. %rom the inlet velocity triangle for each of the five runners, as shown in %igs (52.1a to 52.1e), it is found that an increase in specific speed (or a decreased in head) is accompanied by a reduction in inlet velocity . +ut the flow velocity at inlet increases allowing a large amount of fluid to enter the turbine. The most important point to be noted in this conte$t is that the flow at inlet to all the runners, e$cept the 'aplan one, is in radial and tangential directions. Therefore, the inlet velocity triangles of those turbines (%igure 52.1a to 52.1d) are shown in a plane containing the radial ant tangential directions, and hence the flow velocity represents the radial component of velocity.

n case of a 'aplan runner, the flow at inlet is in a$ial and tangential directions. Therefore, the inlet velocity triangle in this case (%igure 52.1e) is shown in a place containing the a$ial and tangential directions, and hence the flow velocity represents the a$ial component of velocity .The tangential component of velocity is almost nil at outlet of all runners. Therefore, the outlet velocity triangle (%igure

52.1f) is identical in shape of all runners. 0owever, the e$it velocity while it is the radial one in all other runners.

is a$ial in 'aplan and !ubs runner,

,a- Francis r'nner for lo+ s#ecific s#eeds

,.- Francis r'nner for nor&al s#ecific s#eeds

,c- Francis r'nner for %ig% s#ecific s#eeds ,d- D'.s r'nner

,e- Dal#an r'nner

,f- For allreaction ,Francis as +ell as Da#lan- r'nners

O'tlet elocity triangle Fig! /E!* E ol'tion of Da#lan r'nner for& Francis one %igure 52.5 shows a schematic diagram of propeller or 'aplan turbine. The function of the guide vane is same as in case of %rancis turbine. +etween the guide vanes and the runner, the fluid in a propeller turbine turns through a right3angle into the a$ial direction and then passes through the runner. The runner usually has four or si$ blades and closely resembles a ship.s propeller. -eglecting the frictional effects, the flow approaching the runner blades can be considered to be a free vorte$ with whirl velocity being inversely proportional to radius, while on the other hand, the blade velocity is directly proportional to the radius. To take care of this different relationship of the fluid velocity and the blade velocity with the changes in radius, the blades are twisted. The angle with a$is is greater at the tip that at the root.

Fig! /E!/ A #ro#eller of Da#lan t'r.ine Different ty#es of draft t'.es incor#orated in reaction t'r.ines The draft tube is an integral part of a reaction turbine. ts principle has been e$plained earlier. The shape of draft tube plays an important role especially for high specific speed turbines, since the efficient recovery of kinetic energy at runner outlet depends mainly on it. Typical draft tubes, employed in practice, are discussed as follows. Straig%t di ergent t'.e FFig! /E!5,a-G The shape of this tube is that of frustum of a cone. t is usually employed for low specific speed, vertical shaft %rancis turbine. The cone angle is restricted to M 2 to avoid the losses due to separation. The tube must discharge sufficiently low under tail water level. The ma$imum efficiency of this type of draft tube is K2Q. This type of draft tube improves speed regulation of falling load. Si&#le el.o+ ty#e ,Fig! /E!5.- The vertical length of the draft tube should be made small in order to keep down the cost of e$cavation, particularly in rock. The e$it diameter of draft tube should be as large as possible to recover kinetic energy at runner.s outlet. The cone angle of the tube is again fi$ed from the consideration of losses due to flow separation. Therefore, the draft tube must be bent to keep its definite length. 8imple elbow type draft tube will serve such a purpose. ts efficiency is, however, low(about 92Q). This type of draft tube turns the water from the vertical to the hori#ontal direction with a minimum depth of e$cavation. 8ometimes, the transition from a circular section in the vertical portion to a rectangular section in the hori#ontal part (%ig. 52.6c) is incorporated in the design to have a higher efficiency of the draft tube. The hori#ontal portion of the draft tube is generally inclined upwards to lead the water gradually to the level of the tail race and to prevent entry of air from the e$it end.

Fig're /E!5 Different ty#es of draft t'.es

Ca itation in reaction t'r.ines

f the pressure of a liquid in course of its flow becomes equal to its vapour pressure at the e$isting temperature, then the liquid starts boiling and the pockets of vapour are formed which create vapour locks to the flow and the flow is stopped. The phenomenon is known as cavitation. To avoid cavitation, the minimum pressure in the passage of a liquid flow, should always be more than the vapour pressure of the liquid at the working temperature. n a reaction turbine, the point of minimum pressure is usually at the outlet end of the runner blades, i.e at the inlet to the draft tube. %or the flow between such a point and the final discharge into the trail race (where the pressure is atmospheric), the +ernoulli.s equation can be written, in consideration of the velocity at the discharge from draft tube to be negligibly small, as

(5(.()

where,

and

represent the static pressure and velocity of the liquid at the outlet of the runner (or at , the smaller is the value of and the cavitation is

the inlet to the draft tube). The larger the value of more likely to occur. The term

in &q. (5(.() represents the loss of head due to friction in the draft tube

and # is the height of the turbine runner above the tail water surface. %or cavitation not to occur where is the vapour pressure of the liquid at the working temperature.

An important parameter in the conte$t of cavitation is the available suction head (inclusive of both static and dynamic heads) at e$it from the turbine and is usually referred to as the net positive suction head .-@80. which is defined as

(5(.1)

with the help of &q. (5(.() and in consideration of negligible frictional losses in the draft tube &q. (5(.1) can be written as

(5(.5)

A useful design parameter known as Thoma.s Cavitation @arameter (after the >erman &ngineer !ietrich Thoma, who first introduced the concept) is defined as

(5(.6)

%or a given machine, operating at its design condition, another useful parameter cavitaion parameter is define as

known as critical

(5(.7)

Therefore, for cavitaion not to occur f either z or H is increased,

(since,

is reduced. To determine whether cavitation is likely to occur in a particular is greater than the value of for a

installation, the value of may be calculated. *hen the value of particular design of turbine cavitation is not e$pected to occur.

n practice, the value of is used to determine the ma$imum elevation of the turbine above tail water surface for cavitation to be avoided. The parameter of increases with an increase in the specific speed of the turbine. 0ence, turbines having higher specific speed must be installed closer to the tail water level.

"erfor&ance C%aracteristics of Reaction T'r.ine

t is not always possible in practice, although desirable, to run a machine at its ma$imum efficiency due to changes in operating parameters. Therefore, it becomes important to know the performance of the machine under conditions for which the efficiency is less than the ma$imum. t is more useful to plot the basic dimensionless performance parameters (%ig. 5(.() as derived earlier from the similarity principles of fluid machines. Thus one set of curves, as shown in %ig. 5(.(, is applicable not )ust to the conditions of the test, but to any machine in the same homologous series under any altered conditions.

Fig're /$!$ #erfor&ance c%aracteristics of a reaction

t'r.ine ,in di&ensionless #ara&eters%igure 5(.1 is one of the typical plots where variation in efficiency of different reaction turbines with the rated power is shown.

Fig're /$!* 0ariation of efficiency +it% load

Co&#arison of S#ecific S#eeds of Hydra'lic T'r.ines

8pecific speeds and their ranges of variation for different types of hydraulic turbines have already been discussed earlier. %igure 51.( shows the variation of efficiencies with the dimensionless specific speed of different hydraulic turbines. The choice of a hydraulic turbine for a given purpose depends upon the matching of its specific speed corresponding to ma$imum efficiency with the required specific speed determined from the operating parameters, namely, & (rotational speed), p (power) and H (available head).

Fig're /*!$ 0ariation of efficiency +it% s#ecific s#eed for %ydra'lic t'r.ines

3o erning of Reaction T'r.ines >overning of reaction turbines is usually done by altering the position of the guide vanes and thus controlling the flow rate by changing the gate openings to the runner. The guide blades of a reaction turbine (%igure 51.1) are pivoted and connected by levers and links to the regulating ring. Two long regulating rods, being attached to the regulating ring at their one ends, are connected to a regulating lever at their other ends. The regulating lever is keyed to a regulating shaft which is turned by a servomotor piston of the oil

Fig're /*!* 3o erning of reaction t'r.ine

B'l. T'r.ine

The bulb turbine is a reaction turbine of 'aplan type which is used for e$tremely low heads. The characteristic feature of this turbine is that the turbine components as well as the generator are housed inside a bulb, from which the name is developed. The main difference from the 'aplan turbine is that the water flows in a mi$ed a$ial3radial direction into the guide vane cascade and not through a scroll casing. The giude vane spindles are normally inclined to 922 in relation to the turbine shaft and thus results in a conical guide vane cascade contrary to other types of turbines. The runner of a bulb turbine may have different numbers of blades depending on the head and water flow. The bulb turbines have higher full3 load efficiency and higher flow capacity as compared to 'aplan turbine. t has a relatively lower construction cost. The bulb turbines can be utili#ed to tap electrical power from the fast flowing rivers on the hills. %igure 51.5 shows the schematic of a +ulb Turbine @ower @lant.

Fig're /*!/ Sc%e&atic of B'l. T'r.ine "o+er 3enerating Station

"'&#s
Rotodyna&ic "'&#s
A rotodynamic pump is a device where mechanical energy is transferred from the rotor to the fluid by the principle of fluid motion through it. The energy of the fluid can be sensed from the pressur and velocity of the fluid at the delivery end of the pump. Therefore, it is essentially a turbine in reverse. /ike turbines,

pumps are classified according to the main direction of fluid path through them like (i) radial flow or centrifugal, (ii) a$ial flow and (iii) mi$ed flow types.

Centrif'gal "'&#s
The pumps employing centrifugal effects for increasing fluid pressure have been in use for more than a century.The centrifugal pump, by its principle, is converse of the %rancis turbine. The flow is radially outward, and the hence the fluid gains in centrifugal head while flowing through it. +ecause of certain inherent advantages,such as compactness, smooth and uniform flow, low initial cost and high efficiency even at low heads, centrifugal pumps are used in almost all pumping systems. 0owever, before considering the operation of a pump in detail, a general pumping system is discussed as follows.

3eneral "'&#ing Syste& and t%e Net Head De elo#ed .y a "'&#

The word pumping, referred to a hydraulic system commonly implies to convey liquid from a low to a high reservoir. 8uch a pumping system, in general, is shown in %ig. 55.(. At any point in the system, the elevation or potential head is measured from a fi$ed reference datum line. The total head at any point comprises pressure head, velocity head and elevation head. %or the lower reservoir, the total head at the free surface is and is equal to the elevation of the free surface above the datum line since the velocity and static pressure at A are #ero. 8imilarly the total head at the free surface in the higher reservoir is ( reference datum. ) and is equal to the elevation of the free surface of the reservoir above the

The variation of total head as the liquid flows through the system is shown in %ig. 55.1. The liquid enters the intake pipe causing a head loss for which the total energy line drops to point 2 corresponding to a location )ust after the entrance to intake pipe. The total head at 2 can be written as

As the fluid flows from the intake to the inlet flange of the pump at elevation

the total head drops further

to the point C (%igure 55.1) due to pipe friction and other losses equivalent to . The fluid then enters the pump and gains energy imparted by the moving rotor of the pump. This raises the total head of the fluid to a point ! (%igure 55.1) at the pump outlet (%igure 55.(). n course of flow from the pump outlet to the upper reservoir, friction and other losses account for a total head loss or down to a point ' . At ' an e$it loss occurs when the liquid enters the upper reservoir, bringing the total heat at point / (%igure 55.1) to that at the free surface of the upper reservoir. f the total heads are measured at the inlet and outlet flanges respectively, as done in a standard pump test, then

Fig're //!$ A general #'&#ing syste&

Fig're //!* C%ange of %ead in a #'&#ing syste&

Total inlet head to the pump ;

Total outlet head of the pump ; where and are the velocities in suction and delivery pipes respectively.

Therefore, the total head developed by the pump, (55.()

The head developed H is termed as manometric head . f the pipes connected to inlet and outlet of the pump are of same diameter, and therefore the head developed or manometric head H is simply the gain in pie#ometric pressure head across the pump which could have been recorded by a manometer connected between the inlet and outlet flanges of the pump. n practice, ( ) is so small in

comparison to that it is ignored. t is therefore not surprising o find that the static pressure head across the pump is often used to describe the total head developed by the pump. The vertical distance between the two levels in the reservoirs is known as static head or static lift. =elationship

between , the static head and H , the head developed can be found out by applying +ernoulli.s equation between A and 3 and between $ and / (%igure 55.() as follows"

(55.1)

+etween $ and / ,

(55.5)

substituting we can write

from &q. (55.1) into &q. (55.5), and then with the help of &q. (55.(),

(55.6) Therefore, we have, the total head developed by the pump ; static head Y sum of all the losses.

The simplest from of a centrifugal pump is shown in %igure 55.5. t consists of three important parts" (i) the rotor, usually called as impeller, (ii) the volute casing and (iii) the diffuser ring. The impeller is a rotating solid disc with curved blades standing out vertically from the face of the disc. The impeller may be single sided (%igure 55.6a) or doublesided (%igure 55.6b). A double sided impeller has a relatively small flow capacity.

Fig're //!/ A centrif'gal #'&# The tips of the blades are sometimes covered by another flat disc to give shrouded blades (%igure 55.6c), otherwise the blade tips are left open and the casing of the pump itself forms the solid outer wall of the blade passages. The advantage of the shrouded blade is that flow is prevented from leaking across the blade tips from one passage to another.

,a- Single sided i&#eller ,.- Do'.le sided i&#eller ,c- S%ro'ded i&#eller Fig're //!5 Ty#es of i&#ellers in a centrif'gal #'&# As the impeller rotates, the fluid is drawn into the blade passage at the impeller eye, the centre of the impeller. The inlet pipe is a$ial and therefore fluid enters the impeller with very little whirl or tangential component of velocity and flows outwards in the direction of the blades. The fluid receives energy from the impeller while flowing through it and is discharged with increased pressure and velocity into the casing. To convert the kinetic energy or fluid at the impeller outlet gradually into pressure energy, diffuser blades mounted on a diffuser ring are used. The stationary blade passages so formed have an increasing cross3sectional area which reduces the flow velocity and hence increases the static pressure of the fluid. %inally, the fluid moves from the diffuser blades into the volute casing which is a passage of gradually increasing cross3section and also serves to reduce the velocity of fluid and to convert some of the velocity head into static head. 8ometimes pumps have only volute casing without any diffuser. %igure 56.( shows an impeller of a centrifugal pump with the velocity triangles drawn at inlet and outlet. The blades are curved between the inlet and outlet radius. A particle of fluid moves along the broken curve shown in %igure 56.(.

Fig're /5!$ 0elocity triangles for centrif'gal #'&# I&#eller /et be the angle made by the blade at inlet, with the tangent to the inlet radius, while and is the blade

angle with the tangent at outlet. respectively, while and respectively. Therefore,

are the absolute velocities of fluid at inlet an outlet

are the relative velocities (with respect to blade velocity) at inlet and outlet

*ork done on the fluid per unit weight

(56.()

A centrifugal pump rarely has any sort of guide vanes at inlet. The fluid therefore approaches the impeller without appreciable whirl and so the inlet angle of the blades is designed to produce a right3angled velocity triangle at inlet (as shown in %ig. 56.(). At conditions other than those for which the impeller was designed, the direction of relative velocity does not coincide with that of a blade. Consequently, the fluid changes direction abruptly on entering the impeller. n addition, the eddies give rise to some back flow into the inlet pipe, thus causing fluid to have some whirl before entering the impeller. 0owever, considering the operation under design conditions, the inlet whirl velocity and accordingly the inlet angular momentum of the fluid entering the impeller is set to #ero. Therefore, &q. (56.() can be written as *ork done on the fluid per unit weight = (56.1)

*e see from this equation that the work done is independent of the inlet radius. The difference in total head across the pump known as manometric head, is always less than the quantity of the energy dissipated in eddies due to friction. The ratio of manometric head H and the work head imparted by the rotor on the fluid because

(usually

known as &uler head) is termed as manometric efficiency . t represents the effectiveness of the pump in increasing the total energy of the fluid from the energy given to it by the impeller. Therefore, we can write

(56.5)

The overall efficiency

of a pump is defined as (56.6)

where, % is the volume flow rate of the fluid through the pump, and ( is the shaft power, i.e. the input power to the shaft. The energy required at the shaft e$ceeds because of friction in the bearings and other mechanical parts. Thus a mechanical efficiency is defined as

(56.7)

so that (56.9)

Sli# Factor
Bnder certain circumstances, the angle at which the fluid leaves the impeller may not be the same as the actual blade angle. This is due to a phenomenon known as fluid slip, which finally results in a reduction in the tangential component of fluid velocity at impeller outlet. Ine possible e$planation for slip is given as follows. n course of flow through the impeller passage, there occurs a difference in pressure and velocity between the leading and trailing faces of the impeller blades. In the leading face of a blade there is relatively a high pressure and low velocity, while on the trailing face, the pressure is lower and hence the velocity is higher. This results in a circulation around the blade and a non3uniform velocity distribution at any radius. The mean direction of flow at outlet, under this situation, changes from the blade angle at outlet outlet to a different angle is reduced to as shown in %igure 56.1 Therefore the tangential velocity component at , as shown by the velocity triangles in %igure 56.1, and the difference is defined as

is defined as the slip. The slip factor

Fig're /5!* Sli# and elocity in t%e i&#eller .lade #assage of a centrif'gal #'&# *ith the application of slip factor , the work head imparted to the fluid (&uler head) becomes

. The typical values of slip factor lie in the region of 2.K.

Losses in a Centrif'gal "'&#

Mechanical friction power loss due to friction between the fi$ed and rotating parts in the bearing and
stuffing bo$es. G !isc friction power loss due to friction between the rotating faces of the impeller (or disc) and the liquid. G /eakage and recirculation power loss. This is due to loss of liquid from the pump and recirculation of the liquid in the impeller. The pressure difference between impeller tip and eye can cause a recirculation of a small volume of liquid, thus reducing the flow rate at outlet of the impeller as shown in %ig. (56.5).

Fig're /5!/ Lea2age and recirc'lation in a centrif'gal #'&#

C%aracteristics of a Centrif'gal "'&#

*ith the assumption of no whirl component of velocity at entry to the impeller of a pump, the work done on the fluid per unit weight by the impeller is given by &quation( 56.1). Considering the fluid to be frictionless, the head developed by the pump will be the same san can be considered as the theoretical head developed. Therefore we can write for theoretical head developed as

(57.()

%rom the outlet velocity triangle figure( 56.() (57.1) where % is rate of flow at impeller outlet and A is the flow area at the periphery of the impeller. The blade speed at outlet can be e$pressed in terms of rotational speed of the impeller & as

Bsing this relation and the relation given by &q. (57.1), the e$pression of theoretical head developed can be written from &q. (57.() as

(57.5)

where,

and

%or a given impeller running at a constant rotational speed.

and

are constants, and therefore head with % is

and discharge bears a linear relationship as shown by &q. (57.5). This linear variation of plotted as curve in %ig. 57.(.

f slip is taken into account, the theoretical head will be reduced to . Moreover the slip will increase with the increase in flow rate % . The effect of slip in head3discharge relationship is shown by the curve in %ig. 57.(. The loss due to slip can occur in both a real and an ideal fluid, but in a real fluid the shock losses at entry to the blades, and the friction losses in the flow passages have to be considered. At the design point the shock losses are #ero since the fluid moves tangentially onto the blade, but on either side of the design point the head loss due to shock increases according to the relation (57.6)

Fig're /8!$ Head;disc%arge c%aracteristics of a centrif'gal #'&#

where is the off design flow rate and e$pressed as

is a constant. The losses due to friction can usually be

(57.7)

where,

is a constant.

&quation (57.7) and (57.6) are also shown in %ig. 57.( (curves and 4) as the characteristics of losses in a centrifugal pump. +y subtracting the sum of the losses from the head in consideration of the slip, at any flow rate (by subtracting the sum of ordinates of the curves and 4 from the ordinate of the curve at all values of the abscissa), we get the curve 4 which represents the relationship of the actual head with the flow rate, and is known as head3discharge characteristic curve of the pump.

Effect of .lade o'tlet angle

The head3discharge characteristic of a centrifugal pump depends (among other things) on the outlet angle of the impeller blades which in turn depends on blade settings. Three types of blade settings are possible (i) the forward facing for which the blade curvature is in the direction of rotation and, therefore, (%ig. 57.1a), (ii) radial, when (%ig. 57.1b), and (iii) backward facing for which the blade

curvature is in a direction opposite to that of the impeller rotation and therefore, (%ig. 57.1c). The outlet velocity triangles for all the cases are also shown in %igs. 57.1a, 57.1b, 57.1c. %rom the geometry of any triangle, the relationship between and can be written as.

which was e$pressed earlier by &q. (57.1).

Fig're /8!* O'tlet elocity triangles for different .lade settings in a centrif'gal #'&# n case of forward facing blade, . n case of radial blade, and hence cot and is negative and therefore is more than and

n case of backward facing blade,

Therefore the sign of , the constant in the theoretical head3discharge relationship given by the &q. (57.5), depends accordingly on the type of blade setting as follows"

For forward curved blades

%or radial blades %or backward curved blades *ith the incorporation of above conditions, the relationship of head and discharge for three cases are shown in %igure 57.5. These curves ultimately revert to their more recogni#ed shapes as the actual head3 discharge characteristics respectively after consideration of all the losses as e$plained earlier %igure 57.6.

%or both radial and forward facing blades, the power is rising monotonically as the flow rate is increased. n the case of backward facing blades, the ma$imum efficiency occurs in the region of ma$imum power. f, for some reasons, % increases beyond there occurs a decrease in power. Therefore the motor used to drive the pump at part load, but rated at the design point, may be safely used at the ma$imum power. This is known as self3limiting characteristic. n case of radial and forward3facing blades, if the pump motor is rated for ma$imum power, then it will be under utili#ed most of the time, resulting in an increased cost for the e$tra rating. *hereas, if a smaller motor is employed, rated at the design point, then if % increases above the motor will be overloaded and may fail. t, therefore, becomes more difficult to decide on a choice of motor in these later cases (radial and forward3facing blades).

Fig're /8!/ T%eoretical %ead;disc%arge c%aracteristic c'r es of a centrif'gal #'&# for different .lade settings

Fig're /8!5 Act'al %ead;disc%arge and #o+er;disc%arge c%aracteristic c'r es of a centrif'gal #'&#

Flo+ t%ro'g% 0ol'te C%a&.ers

Apart from frictional effects, no torque is applied to a fluid particle once it has left the impeller. The angular momentum of fluid is therefore constant if friction is neglected. Thus the fluid particles follow the path of a free vorte$. n an ideal case, the radial velocity at the impeller outlet remains constant round the circumference. The combination of uniform radial velocity with the free vorte$ ( ;constant) gives a pattern of spiral streamlines which should be matched by the shape of the volute. This is the most important feature of the design of a pump. At ma$imum efficiency, about (2 percent of the head generated by the impeller is usually lost in the volute.

0anned Diff'ser
A vanned diffuser, as shown in %ig. 59.(, converts the outlet kinetic energy from impeller to pressure energy of the fluid in a shorter length and with a higher efficiency. This is very advantageous where the si#e of the pump is important. A ring of diffuser vanes surrounds the impeller at the outlet. The fluid leaving the impeller first flows through a vaneless space before entering the diffuser vanes. The divergence angle of the diffuser passage is of the order of M3(2 L which ensures no boundary layer separation. The optimum number of vanes are fi$ed by a compromise between the diffusion and the frictional loss. The greater the number of vanes, the better is the diffusion (rise in static pressure by the reduction in flow velocity) but greater is the frictional loss. The number of diffuser vanes should have no common factor with the number of impeller vanes to prevent resonant vibration.

Fig're /:!$ A anned diff'ser of a centrif'gal #'&#

Ca itation in centrif'gal #'&#s

Cavitation is likely to occur at the inlet to the pump, since the pressure there is the minimum and is lower than the atmospheric pressure by an amount that equals the vertical height above which the pump is situated from the supply reservoir (known as sump) plus the velocity head and frictional losses in the suction pipe. Applying the +ernoulli.s equation between the surface of the liquid in the sump and the entry to the impeller, we have

(59.()

where, is the pressure at the impeller inlet and is the pressure at the liquid surface in the sump which is usually the atmospheric pressure, 6# is the vertical height of the impeller inlet from the liquid surface in the sump, is the loss of head in the suction pipe. 8trainers and non3return valves are

commonly fitted to intake pipes. The term must therefore include the losses occurring past these devices, in addition to losses caused by pipe friction and by bends in the pipe. n the similar way as described in case of a reaction turbine, the net positive suction head .-@80. in case of a pump is defined as the available suction head (inclusive of both static and dynamic heads) at pump inlet above the head corresponding to vapor pressure. Therefore,

(59.1)

Again, with help of &q. (59.(), we can write

The Thomas cavitation parameter s and critical cavitation parameter in case of reaction turbine) as

are defined accordingly (as done

(59.5) and (59.6)

*e can say that for cavitation not to occur,

n order that s should be as large as possible, z must be as small as possible. n some installations, it may even be necessary to set the pump below the liquid level at the sump (i.e. with a negative vale of z ) to avoid cavitation.

A1ial Flo+ or "ro#eller "'&#

The a$ial flow or propeller pump is the converse of a$ial flow turbine and is very similar to it an appearance. The impeller consists of a central boss with a number of blades mounted on it. The impeller rotates within a cylindrical casing with fine clearance between the blade tips and the casing walls. %luid particles, in course of their flow through the pump, do not change their radial locations. The inlet guide vanes are provided to properly direct the fluid to the rotor. The outlet guide vanes are provided to eliminate the whirling component of velocity at discharge. The usual number of impeller blades lies between 1 and M, with a hub diameter to impeller diameter ratio of 2.5 to 2.9. The %igure 5:.( shows an a$ial flow pump. The flow is the same at inlet and outlet. an a$ial flow pumps develops low head but have high capacity. the ma$imum head for such pump is of the order of 12m.The section through the blade at U3U (%igure 5:.() is shown with inlet and outlet velocity triangles in %igure 5:.1.

Fig're />!$ A #ro#eller of an a1ial flo+ #'&#

Fig're />!* 0elocity triangles of an a1ial flo+ #'&# Analysis The blade has an aerofoil section. The fluid does not impinge tangentially to the blade at inlet, rather the blade is inclined at an angle of incidence (i) to the relative velocity at the inlet . f we consider the conditions at a mean radius then

where

is the angular velocity of the impeller.

*ork done on the fluid per unit weight ;

%or ma$imum energy transfer ,

, i.e

.Again , from the outlet velocity triangle,

Assuming a constant flow from inlet to outlet

Then, we can write Ma$imum energy transfer to the fluid per unit weight (5:.() %or constant energy transfer over the entire span of the blade from hub to tip , the right hand side of &quation (5:.() has to be same for all values of . t is obvious that increases with radius , therefore

an equal increase in must take place, and since Therefore , the blade must be twisted as the radius changes. Matc%ing of "'&# and Syste& C%aracteristics

is constant then

must increase.

The design point of a hydraulic pump corresponds to a situation where the overall efficiency of operation is ma$imum. 0owever the e$act operating point of a pump, in practice, is determined from the matching of pump characteristic with the headloss3flow, characteristic of the e$ternal system (i.e. pipe network, valve and so on) to which the pump is connected. /et us consider the pump and the piping system as shown in %ig. (7.(M. 8ince the flow is highly turbulent, the losses in pipe system are proportional to the square of flow velocities and can, therefore, be e$pressed in terms of constant loss coefficients. Therefore, the losses in both the suction and delivery sides can be written as (5:.1a) (5:.1b)

where,

is the loss of head in suction side and and

is the loss of head in delivery side and f is the !arcy.s

friction factor,

are the lengths and diameters of the suction and delivery pipes

respectively, while and are accordingly the average flow velocities. The first terms in &qs. (5:.(a) and (5:.(b) represent the ordinary friction loss (loss due to friction between fluid ad the pipe wall), while the second terms represent the sum of all the minor losses through the loss coefficients and which include losses due to valves and pipe bends, entry and e$it losses, etc. Therefore the total head the pump has to develop in order to supply the fluid from the lower to upper reservoir is

(5:.5) -ow flow rate through the system is proportional to flow velocity. Therefore resistance to flow in the form of losses is proportional to the square of the flow rate and is usually written as (5:.6)

; system resistance ;

where 7 is a constant which includes, the lengths and diameters of the pipes and the various loss coefficients. 8ystem resistance as e$pressed by &q. (5:.6), is a measure of the loss of head at any particular flow rate through the system. f any parameter in the system is changed, such as ad)usting a valve opening, or inserting a new bend, etc., then 7 will change. Therefore, total head of &q. (5:.1) becomes, (5:.7) The head H can be considered as the total opposing head of the pumping system that must be overcome for the fluid to be pumped from the lower to the upper reservoir. The &q. (5:.6) is the equation for system characteristic, and while plotted on H-% plane (%igure 5:.5), represents the system characteristic curve. The point of intersection between the system characteristic and the pump characteristic on H-% plane is the operating point which may or may not lie at the design point that corresponds to ma$imum efficiency of the pump. The closeness of the operating and design points depends on how good an estimate of the e$pected system losses has been made. t should be noted that if there is no rise in static head of the liquid (for e$ample pumping in a hori#ontal pipeline between two reservoirs at the same elevation), origin. is #ero and the system curve passes through the

Fig're />!/ H;H C%aracteristics of #'&# and syste&

Effect of S#eed 0ariation

0ead3!ischarge characteristic of a given pump is always referred to a constant speed. f such characteristic at one speed is know, it is possible to predict the characteristic at other speeds by using the principle of similarity. /et A, 2, 3 are three points on the characteristic curve (%ig. 5:.6) at speed %or points A, 2 and 3 , the corresponding heads and flows at a new speed are found as follows" .

Fig're />!5 Effect of s#eed ariation on o#erating #oint of a centrif'gal #'&# %rom the equality of term O&q. (5.()P gives (since for a given pump $ is constant) (5:.9)

and similarly, equality of

term O&q. (5.()P gives (5:.:)

Applying &qs. (5:.9) and (5:.:) to points A, 2 and 3 the corresponding points and then the characteristic curve can be drawn at the new speed Thus,

and

are found

and which gives

or

(5:.M)

&quation (5:.M) implies that all corresponding or similar points on 0ead3!ischarge characteristic curves at different speeds lie on a parabola passing through the origin. f the static lift becomes #ero, then the curve for system characteristic and the locus of similar operating points will be the same parabola passing through the origin. This means that, in case of #ero static life, for an operating point at speed , it is only necessary to apply the similarity laws directly to find the corresponding operating point at the new speed since it will lie on the system curve itself (%igure 5:.6).

0ariation of "'&# Dia&eter

A variation in pump diameter may also be e$amined through the similarly laws. %or a constant speed,

and or, (5M.()

"'&#s in Series and "arallel

*hen the head or flow rate of a single pump is not sufficient for a application, pumps are combined in series or in parallel to meet the desired requirements. @umps are combined in series to obtain an increase in head or in parallel for an increase in flow rate. The combined pumps need not be of the same design. %igures 5M.( and 5M.1 show the combined H-% characteristic for the cases of identical pumps connected in series and parallel respectively. t is found that the operating point changes in both cases. %ig. 5M.5 shows the combined characteristic of two different pumps connected in series and parallel.

Fig're /@!$ T+o si&ilar #'&#s connected in series

Fig're /@!* T+o si&ilar #'&#s connected in #arallel

S#ecific S#eed of Centrif'gal "'&#s

The concept of specific speed for a pump is same as that for a turbine. 0owever, the quantities of interest are &, H and % rather than &, H and ( like in case of a turbine. %or pump (5M.1)

Fig're /@!/ T+o different #'&#s connected in series and #arallel The effect of the shape of rotor on specific speed is also similar to that for turbines. That is, radial flow (centrifugal) impellers have the lower values of compared to those of a$ial3flow designs. The impeller, however, is not the entire pump and, in particular, the shape of volute may appreciably affect the specific speed. -evertheless, in general, centrifugal pumps are best suited for providing high heads at moderate rates of flow as compared to a$ial flow pumps which are suitable for large rates of flow at low heads. 8imilar to turbines, the higher is the specific speed, the more compact is the machine for given requirements. %or multistage pumps, the specific speed refers to a single stage.

%ans and blowers (%ig. 5K.() are turbomachines which deliver air at a desired high velocity (and accordingly at a high mass flow rate) but at a relatively low static pressure. The pressure rise across a fan is e$tremely low and is of the order of a few millimeters of water gauge.The upper limit of pressure rise is of the order of 172mm of water gauge.The rise in static pressure across a blower is relatively higher and is more than (222 mm of water gauge that is required to overcome the pressure losses of the gas during its flow through various passages. A blower may be constructed in multistages for still higher discharge pressure.

Fig're /A!$ A centrif'gal fan or .lo+er A large number of fans and blowers for relatively high pressure applications are of centrifugal type. The main components of a centrifugal blower are shown in %ig. 5K.1. A blower consists of an impeller which has blades fi$ed between the inner and outer diameters. The impeller can be mounted either directly on the shaft e$tension of the prime mover or separately on a shaft supported between two additional bearings. Air or gas enters the impeller a$ially through the inlet no##le which provides slight acceleration to the air before its entry to the impeller. The action of the impeller swings the gas from a smaller to a larger radius and delivers the gas at a high pressure and velocity to the casing. The flow from the impeller blades is collected by a spiral3shaped casing known as volute casing or spiral casing . The casing can further increase the static pressure of the air and it finally delivers the air to the e$it of the blower.

Fig're /A!* Main co&#onents of a centrif'gal .lo+er The centrifugal fan impeller can be fabricated by welding curved or almost straight metal blades to the two side walls (shrouds) of the rotor. The casings are made of sheet metal of different thickness and steel reinforcing ribs on the outside. 8uitable sealing devices are used between the shaft and the casing.

A centrifugal fan impeller may have backward swept blades, radial tipped blades or forward swept blades as shown in %ig. 5K.5. The inlet and outlet velocity triangles are also shown accordingly in the figure. Bnder ideal conditions, the directions of the relative velocity vectors and are same as the blade angles at the entry and the e$it. A #ero whirl at the inlet is assumed which results in a #ero angular momentum at the inlet. The backward swept blades are employed for lower pressure and lower flow rates. The radial tipped blades are employed for handling dust3laden air or gas because they are less prone to blockage, dust erosion and failure. The radial3tipped blades in practice are of forward swept type at the inlet as shown in %ig. 5K.5. The forward3swept blades are widely used in practice. In account of the forward3swept blade tips at the e$it, the whirl component of e$it velocity in a higher stage pressure rise. The following observations may be noted from figure 5K.5. is large which results

, if

, backward swept blades

, radial blades

, if

, if

, forward swept blades

Fig're /A!/ 0elocity triangles at inlet and o'tlet of different ty#es of .lades of an i&#eller of a centrif'gal .lo+er

"ara&etric Calc'lations
The mass flow rate through the impeller is given by (5K.()

The areas of cross sections normal to the radial velocity components

and

are

and

(5K.1) The radial component of velocities at the impeller entry and e$it depend on its width at these sections. %or small pressure rise through the impeller stage, the density change in the flow is negligible and the flow can be assumed to be almost incompressible. %or constant radial velocity (5K.5) &qs. (5K.1) and (5K.5) give (5K.6)

?or2
The work done is given by &uler.s &quation (refer to Module3() as (5K.7) t is reasonable to assume #ero whirl at the entry. This condition gives and hence, Therefore we can write, (5K.9) &quation (5K.7) gives

(5K.:)

%or any of the e$it velocity triangles (%igure 5K.5)

(5K.M)

&q. (5K.:) and (5K.M)

(5K.K)

where

is known as flow coefficient

0ead developed in meters of air ; &quivalent head in meters of water ;

(5K.(2) (5K.(()

where

and

are the densities of air and water respectively.

Assuming that the flow fully obeys the geometry of the impeller blades, the specific work done in an isentropic process is given by (5K.(1) The power required to drive the fan is

(5K.(5) The static pressure rise through the impeller is due to the change in centrifugal energy and the diffusion of relative velocity component. Therefore, it can be written as (5K.(5) The stagnation pressure rise through the stage can also be obtained as" (5K.(6) %rom (5K.(5) and (5K.(6) we get

(5K.(7) %rom any of the outlet velocity triangles (%ig. 5K.5),

or, or,

(5K.(9)

or,

*ork done per unit mass is also given by (from (5K.:) and (62.6))"

(5K.(:)

Efficiency

In account of losses, the isentropic work Therefore the stage efficiency is defined by

is less than the actual work.

(5K.(M)

N'&.er of Blades Too few blades are unable to fully impose their geometry on the flow, whereas too many of them restrict the flow passage and lead to higher losses. Most of the efforts to determine the optimum number of blades have resulted in only empirical relations given below

(i)

(5K.(K)

(ii)

(5K.12)

(iii)

(5K.1()

I&#eller Si=e
The diameter ratio of the impeller determines the length of the blade passages. The smaller the ratio the longer is the blade passage. The following value for the diameter ratio is often used by the designers

(5K.11)

where

The following relation for the blade width to diameter ratio is recommended" (5K.15) f the rate of diffusion in a parallel wall impeller is too high, the tapered shape towards the outer periphery, is preferable.. The typical performance curves describing the variation of head, power and efficiency with discharge of a centrifugal blower or fan are shown in %igure 5K.6

Fig're/A!5 "erfor&ance c%aracteristic c'r es of a centrif'gal .lo+er or fan

Fan La+s
The relationships of discharge % , head H and @ower ( with the diameter $ and rotational speed & of a centrifugal fan can easily be e$pressed from the dimensionless performance parameters determined from the principle of similarity of rotodynamic machines as described before . These relationships are known as %an /aws described as follows (62.() (62.1)

(62.5)

where

and

are constants.

%or the same fan, the dimensions get fi$ed and the laws are

and %or the different si#e and other conditions remaining same, the laws are

(62.6) and These relationships are known as the %an3laws. The %an3lows can be summari#ed as %or the same fan" !ischarge 8peed 0ead developed (8peed)1 @ower (8peed) 5 %or the fans of different si#es" !ischarge (!iameter)5 0ead developed (!iameter)1 @ower (!iameter)7

Performance of Fans
%or all three cases (backward, radial and forward swept blades) in %igure 5K.5, we can write (62.7) The work done is given by &uler.s equation (refer to Modue3() as (62.9)

-oting that

(#ero whirl at the entry) we can write

or,

(62.:)

The volume flow rate (assuming no density change between the inlet and outlet)

Thus

+y substitution in (62.:)

(62.M)

/et us define, @ressure Coefficient,

(62.K)

4olume Coefficient, (62.(2) and @ower Coefficient, (62.(() 8ubstitution of (62.M) in the above yields

(62.(1)

or,

(62.(5)

(62.(6)

&quations (62.(5) and (62.(6) are plotted in %ig 62.( for different values of volume coefficient as a parameter.

with

Fig're 5E!$ "erfor&ance c'r es of a Fan

The characteristics in %igure 62.( depict the following

(i) %orward curved fans speed.

develop the highest pressure for a given impeller diameter and

(ii) @ower requirement of a forward curved fan increases steeply for a small change in flow rate. (iii) @ressure developed decreases fast with increasing flow rate in a backward curved fan n conclusion, the forward curved fans have large volume discharge and pressure rise but they demand higher power. 0owever, forward curved fans are unstable for off3design operating conditions. +ackward curved fans are very efficient and the drooping power characteristic makes them suitable for a better off3design performance =adial curved fans are preferred for dust3laden fluids. !ue to their shape, the solid particles are not stuck and deposited on the blade surface.