2

Hydraulic Pumps

2.1 INTRODUCTION
Hydraulics is defined as the science of the conveyance of liquids through pipes.
The pump is often used to raise water from a low level to a high level where it can
be stored in a tank. Most of the theory applicable to hydraulic pumps has been
derived using water as the working fluid, but other liquids can also be used. In this
chapter, we will assume that liquids are totally incompressible unless otherwise
specified. This means that the density of liquids will be considered constant no
matter how much pressure is applied. Unless the change in pressure in a particular
situation is very great, this assumption will not cause a significant error in
calculations. Centrifugal and axial flow pumps are very common hydraulic
pumps. Both work on the principle that the energy of the liquid is increased by
imparting kinetic energy to it as it flows through the pump. This energy is
supplied by the impeller, which is driven by an electric motor or some other drive.
The centrifugal and axial flow pumps will be discussed separately in the
following sections.

2.2 CENTRIFUGAL PUMPS

The three important parts of centrifugal pumps are (1) the impeller, (2) the volute
casing, and (3) the diffuser.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 48 Chapter 2

2.2.1 Impeller
The centrifugal pump is used to raise liquids from a lower to a higher level by
creating the required pressure with the help of centrifugal action. Whirling
motion is imparted to the liquid by means of backward curved blades mounted on
a wheel known as the impeller. As the impeller rotates, the fluid that is drawn into
the blade passages at the impeller inlet or eye is accelerated as it is forced radially
outwards. In this way, the static pressure at the outer radius is much higher than at
the eye inlet radius. The water coming out of the impeller is then lead through the
pump casing under high pressure. The fluid has a very high velocity at the outer
radius of the impeller, and, to recover this kinetic energy by changing it into
pressure energy, diffuser blades mounted on a diffuser ring may be used. The
stationary blade passages have an increasing cross-sectional area. As the fluid
moves through them, diffusion action takes place and hence the kinetic energy is
converted into pressure energy. Vaneless diffuser passages may also be used. The
fluid moves from the diffuser blades into the volute casing. The functions of a
volute casing can be summarized as follows: It collects water and conveys it to
the pump outlet. The shape of the casing is such that its area of cross-section
gradually increases towards the outlet of the pump. As the flowing water
progresses towards the delivery pipe, more and more water is added from the
outlet periphery of the impeller. Figure 2.1 shows a centrifugal pump impeller
with the velocity triangles at inlet and outlet.
For the best efficiency of the pump, it is assumed that water enters the
impeller radially, i.e., a1 ¼ 908 and Cw1 ¼ 0. Using Euler’s pump equation, the
work done per second on the water per unit mass of fluid flowing
W
E¼ ¼ ðU 2 Cw2 2 U 1 Cw1 Þ ð2:1Þ
m
Where Cw is the component of absolute velocity in the tangential direction. E is
referred to as the Euler head and represents the ideal or theoretical head
developed by the impeller only. The flow rate is
Q ¼ 2pr 1 Cr1 b1 ¼ 2pr 2 Cr2 b2 ð2:2Þ
Where Cr is the radial component of absolute velocity and is perpendicular to the
tangent at the inlet and outlet and b is the width of the blade. For shockless entry
and exit to the vanes, water enters and leaves the vane tips in a direction parallel
to their relative velocities at the two tips.
As discussed in Chapter 1, the work done on the water by the pump consists
of the following three parts:
1. The part (C22 – C21)/2 represents the change in kinetic energy of the
liquid.
2. The part (U22 – U21)/2 represents the effect of the centrifugal head or
energy produced by the impeller.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Pumps 49

Figure 2.1 Velocity triangles for centrifugal pump impeller.

3. The part (V22 2 V21)/2 represents the change in static pressure of the
liquid, if the losses in the impeller are neglected.

2.3 SLIP FACTOR

From the preceding section, it may be seen that there is no assurance that the
actual fluid will follow the blade shape and leave the impeller in a radial
direction. There is usually a slight slippage of the fluid with respect to the blade
rotation. Figure 2.2 shows the velocity triangles at impeller tip.
In Fig. 2.2, b2 0 is the angle at which the fluid leaves the impeller, and b2 is
the actual blade angle, and Cw2 and C w2 0 are the tangential components of
absolute velocity corresponding to the angles b2 and b2 0 , respectively. Thus, Cw2
is reduced to C w2 0 and the difference DCw is defined as the slip. The slip factor
is defined as
C w2 0
Slip factor; s ¼
Cw2
According to Stodola’s theory, slip in centrifugal pumps and compressors is due
to relative rotation of fluid in a direction opposite to that of impeller with the same

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 50 Chapter 2

Figure 2.2 Velocity triangle at impeller outlet with slip.

angular velocity as that of an impeller. Figure 2.3 shows the leading side of a
blade, where there is a high-pressure region while on the trailing side of the blade
there is a low-pressure region.
Due to the lower pressure on the trailing face, there will be a higher velocity
and a velocity gradient across the passage. This pressure distribution is associated
with the existence of circulation around the blade, so that low velocity on the high-
pressure side and high velocity on the low-pressure side and velocity distribution
is not uniform at any radius. Due to this fact, the flow may separate from the
suction surface of the blade. Thus, Cw2 is less than Cw20 and the difference is
defined as the slip. Another way of looking at this effect, as given by Stodola, is
shown in Fig. 2.4, the impeller itself has an angular velocity v so that, relative to
the impeller, the fluid must have an angular velocity of 2 v; the result of this being
a circulatory motion relative to the channel or relative eddy. The net result of the
previous discussion is that the fluid is discharged from the impeller at an angle
relative to the impeller, which is less than the vane angle as mentioned earlier.

Figure 2.3 Pressure distribution on impeller vane. LP ¼ low pressure, HP ¼ high

pressure.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Pumps 51

Figure 2.4 Relative eddy in impeller channel.

Hence, the slip factor s is defined as

0
C w2
s¼ ð2:3Þ
C w2
For purely radial blades, which are often used in centrifugal compressors, b2 will
be 908 and the Stodola slip factor becomes
p
s¼12 ð2:4Þ
n
where n is the number of vanes. The Stanitz slip factor is given by
0:63p
s¼12 ð2:5Þ
n
When applying a slip factor, the Euler pump equation becomes
W
¼ sU 2 Cw2 2 U 1 C w1 ð2:6Þ
m
Typically, the slip factor lies in the region of 0.9, while the slip occurs even if the
fluid is ideal.

2.4 PUMP LOSSES

The following are the various losses occurring during the operation of a
centrifugal pump.
1. Eddy losses at entrance and exit of impeller, friction losses in the
impeller, frictional and eddy losses in the diffuser, if provided.
2. Losses in the suction and delivery pipe. The above losses are known as
hydraulic losses.
3. Mechanical losses are losses due to friction of the main bearings, and
stuffing boxes. Thus, the energy supplied by the prime mover to

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 52 Chapter 2

impeller is equal to the energy produced by impeller plus mechanical

losses. A number of efficiencies are associated with these losses.
Let r be the density of liquid; Q, flow rate; H, total head developed by the
pump; Ps, shaft power input; Hi, total head across the impeller; and hi, head loss
in the impeller. Then, the overall efficiency ho is given by:

Fluid power developed by pump rgQH

ho ¼ ¼ ð2:7Þ
Shaft power input Ps
Casing efficiency hc is given by:

hc ¼ Fluid power at casing outlet/fluid power at casing inlet

¼ Fluid power at casing outlet/ðfluid power developed by
impeller 2 leakage lossÞ
¼ rgQH/rgQH i ¼ H/H i ð2:8Þ

Impeller efficiency hi is given by:

hi ¼ Fluid power at impeller exit/fluid

power supplied to impeller
¼ Fluid power at impeller exit/ðfluid power
developed by impeller
þ impeller lossÞ
¼ rgQi H i / rgQi ðH i þ hi Þ ¼ H i /ðH i þ hi Þ ð2:9Þ

Volumetric efficiency hv is given by:

hv ¼ Flow rate through pump/flow rate through impeller

ð2:10Þ
¼ Q/ðQ þ qÞ

Mechanical efficiency hm is given by:

hm ¼ Fluid power supplied to the impeller/power

input to the shaft
¼ rgQi ðhi þ H i Þ/Ps ð2:11Þ
Therefore,

ho ¼ hc hi hv h m ð2:12Þ

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Pumps 53

A hydraulic efficiency may be defined as

Actual head developed by pump
hh ¼
Theoretical head developed by impeller
H
¼ ð2:13Þ
ðH i þ hi Þ

The head H is also known as manometric head.

2.5 THE EFFECT OF IMPELLER BLADE SHAPE

ON PERFORMANCE
The various blade shapes utilized in impellers of centrifugal pumps/compressors
are shown in Fig. 2.5. The blade shapes can be classified as:
1. Backward-curved blades (b2 , 908)
2. Radial blades (b2 ¼ 908)
3. Forward-curved blades (b2 . 908)
As shown in Fig. 2.5, for backward-curved vanes, the value of Cw2 (whirl
component at outlet) is much reduced, and thus, such rotors have a low energy
transfer for a given impeller tip speed, while forward-curved vanes have a high
value of energy transfer. Therefore, it is desirable to design for high values of b2
(over 908), but the velocity diagrams show that this also leads to a very high value
of C2. High kinetic energy is seldom required, and its reduction to static pressure
by diffusion in a fixed casing is difficult to perform in a reasonable sized casing.
However, radial vanes (b2 ¼ 908) have some particular advantages for very high-
speed compressors where the highest possible pressure is required. Radial vanes
are relatively easy to manufacture and introduce no complex bending stresses
(Fig. 2.6).

Figure 2.5 Centrifugal pump outlet velocity triangles for varying blade outlet angle.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 54 Chapter 2

Figure 2.6 Characteristics for varying outlet blade angle.

2.6 VOLUTE OR SCROLL COLLECTOR

A volute or scroll collector consists of a circular passage of increasing cross-
sectional area (Fig. 2.7). The advantage of volute is its simplicity and low cost.
The cross-sectional area increases as the increment of discharge increases
around the periphery of the impeller, and, if the velocity is constant in the volute,

Figure 2.7 Volute or scroll collector.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Pumps 55

Figure 2.8 Cross-section of volute casing.

then the static pressure is likewise constant and the radial thrust will be zero. Any
deviation in capacity (i.e., flow rate) from the design condition will result in a
radial thrust which if allowed to persist could result in shaft bending.
The cross-sectional shape of the volute is generally similar to that shown in
Fig. 2.8, with the sidewalls diverging from the impeller tip and joined by a
semicircular outer wall. The circular section is used to reduce the losses due to
friction and impact when the fluid hits the casing walls on exiting from the
impeller.

2.7 VANELESS DIFFUSER

For the diffusion process, the vaneless diffuser is reasonably efficient and is best
suited for a wide range of operations. It consists simply of an annular passage
without vanes surrounding the impeller. A vaneless diffuser passage is shown in
Fig. 2.9. The size of the diffuser can be determined by using the continuity
equation. The mass flow rate in any radius is given by

m ¼ rAC r ¼ 2prbrCr ð2:14Þ

where b is the width of the diffuser passage,

r 2 b2 r2 Cr2
Cr ¼ ð2:15Þ
rbr

where subscripted variables represent conditions at the impeller outlet and the
unsubscripted variables represent conditions at any radius r in the vaneless
diffuser. Assuming the flow is frictionless in the diffuser, angular momentum

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 56 Chapter 2

Figure 2.9 Vaneless diffuser.

Figure 2.10 Logarithmic spiral flow in vaneless space.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Pumps 57

is constant and
Cw ¼ ðC w2 r 2 Þ/r ð2:16Þ
But the tangential velocity component (Cw) is usually very much larger than the
radial velocity component Cr, and, therefore, the ratio of the inlet to outlet
diffuser velocities CC23 ¼ rr32 .
It means that for a large reduction in the outlet kinetic energy, a diffuser
with a large radius is required. For an incompressible flow, rCr is constant, and,
therefore, tan a ¼ Cw/Cr ¼ constant. Thus, the flow maintains a constant
inclination to radial lines, the flow path traces a logarithmic spiral.
As shown in Fig. 2.10, for an incremental radius dr, the fluid moves through
angle du, then rdu ¼ dr tan a.
Integrating we have
u 2 u2 ¼ tan a logðr/r2 Þ ð2:17Þ
Substituting a ¼ 788 and (r/r2) ¼ 2, the change in angle of the diffuser is equal
to 1808. Because of the long flow path with this type of diffuser, friction effects
are high and the efficiency is low.

2.8 VANED DIFFUSER

The vaned diffuser is advantageous where small size is important. In this type of
diffuser, vanes are used to diffuse the outlet kinetic energy of the fluid at a much
higher rate than is possible by a simple increase in radius, and hence it is possible
to reduce the length of flow path and diameter. The vane number, the angle of
divergence is smaller, and the diffuser becomes more efficient, but greater is the
friction. The cross section of the diffuser passage should be square to give a
maximum hydraulic radius. However, the number of diffuser vanes should have
no common factor with the number of impeller vanes. The collector and diffuser
operate at their maximum efficiency at the design point only. Any deviation from
the design discharge will change the outlet velocity triangle and the subsequent
flow in the casing.

2.9 CAVITATION IN PUMPS

Cavitation is caused by local vaporization of the fluid, when the local static
pressure of a liquid falls below the vapor pressure of the liquid. Small bubbles or
cavities filled with vapor are formed, which suddenly collapse on moving
forward with the flow into regions of high pressure. These bubbles collapse with
tremendous force, giving rise to pressure as high as 3500 atm. In a centrifugal
pump, these low-pressure zones are generally at the impeller inlet, where the fluid
is locally accelerated over the vane surfaces. In turbines, cavitation is most likely

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 58 Chapter 2

to occur at the downstream outlet end of a blade on the low-pressure leading face.
When cavitation occurs, it causes the following undesirable effects:
1. Local pitting of the impeller and erosion of the metal surface.
2. Serious damage can occur from this prolonged cavitation erosion.
3. Vibration of machine and noise is also generated in the form of sharp
cracking sounds when cavitation takes place.
4. A drop in efficiency due to vapor formation, which reduces the
effective flow areas.
The avoidance of cavitation in conventionally designed machines can be
regarded as one of the essential tasks of both pump and turbine designers. This
cavitation imposes limitations on the rate of discharge and speed of rotation of the
pump.
A cavitation parameter is defined as sc ¼ pump total inlet head above
vapor pressure/head developed by the pump or at the inlet flange
 
sc ¼
p1 V 21 pv
þ
rg 2g rg
2 =H ð2:18Þ

The numerator of Eq. (2.18) is a suction head and is called the net positive suction

Figure 2.11 Cavitation limits for radial flow pumps.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Pumps 59

head (NPSH) of the pump. It is a measure of the energy available on the suction
side of the pump, and H is the manometric head. The cavitation parameter is a
function of specific speed, efficiency of the pump, and number of vanes.
Figure 2.11 shows the relationship between sc and Ns. It may be necessary in the
selection of pumps that the value of sc does not fall below the given value by the
plots in Fig. 2.11 for any condition of operation.

2.10 SUCTION SPECIFIC SPEED

The efficiency of the pump is a function of flow coefficient and suction specific
speed, which is defined as
23/4
N suc ¼ NQ 1/2 gðNPSHÞ
Thus,


h ¼ f Q; N suc
The cavitation parameter may also be determined by the following equation
N s /N suc ¼ ðNPSHÞ3/4 /H 3/4
¼ sc3/4 ð2:19Þ

2.11 AXIAL FLOW PUMP

In an axial flow pump, pressure is developed by flow of liquid over blades of
airfoil section. It consists of a propeller-type of impeller running in a casing.
The advantage of an axial flow pump is its compact construction as well as its
ability to run at extremely high speeds. The flow area is the same at inlet and
outlet and the minimum head for this type of pump is the order of 20 m.

2.12 PUMPING SYSTEM DESIGN

Proper pumping system design is the most important single element in
minimizing the life cycle cost. All pumping systems are comprised of a pump, a
driver, pipe installation, and operating controls. Proper design considers
the interaction between the pump and the rest of the system and the calculation of
the operating duty point(s) (Fig. 2.12). The characteristics of the piping system
must be calculated in order to determine required pump performance. This
applies to both simple systems as well as to more complex (branched) systems.
Both procurement costs and the operational costs make up the total cost of an
installation during its lifetime. A number of installation and operational costs are
directly dependent on the piping diameter and the components in the piping
system.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 60 Chapter 2

Figure 2.12 The duty point of the pump is determined by the intersection of the system
curve and the pump curve as shown above.

A considerable amount of the pressure losses in the system are caused by

valves, in particular, control valves in throttle-regulated installations. In systems
with several pumps, the pump workload is divided between the pumps, which
together, and in conjunction with the piping system, deliver the required flow.
The piping diameter is selected based on the following factors:
. Economy of the whole installation (pumps and system)
. Required lowest flow velocity for the application (e.g., avoid
sedimentation)
. Required minimum internal diameter for the application (e.g., solid
handling)
. Maximum flow velocity to minimize erosion in piping and fittings
. Plant standard pipe diameters.
Decreasing the pipeline diameter has the following effects:
. Piping and component procurement and installation costs will decrease.
. Pump installation procurement costs will increase as a result of
increased flow losses with consequent requirements for higher head
pumps and larger motors. Costs for electrical supply systems will
therefore increase.
. Operating costs will increase as a result of higher energy usage due to
increased friction losses.
Some costs increase with increasing pipeline size and some decrease.
Because of this, an optimum pipeline size may be found, based on minimizing
costs over the life of the system. A pump application might need to cover several

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Pumps 61

duty points, of which the largest flow and/or head will determine the rated duty
for the pump. The pump user must carefully consider the duration of operation at
the individual duty points to properly select the number of pumps in the
installation and to select output control.

2.12.1 Methods for Analyzing Existing

Pumping Systems
The following steps provide an overall guideline to improve an existing pumping
system.
. Assemble a complete document inventory of the items in the pumping
system.
. Determine the flow rates required for each load in the system.
. Balance the system to meet the required flow rates of each load.
. Minimize system losses needed to balance the flow rates.
. Affect changes to the pump to minimize excessive pump head in the
balanced system.
. Identify pumps with high maintenance cost.
One of two methods can be used to analyze existing pumping systems. One
consists of observing the operation of the actual piping system, and the second
consists of performing detailed calculations using fluid analysis techniques. The
first method relies on observation of the operating piping system (pressures,
differential pressures, and flow rates), the second deals with creating an accurate
mathematical model of the piping system and then calculating the pressure and
flow rates with the model.
The following is a checklist of some useful means to reduce the life cycle
cost of a pumping system.
. Consider all relevant costs to determine the life cycle cost.
. Procure pumps and systems using life cycle cost considerations.
. Optimize total cost by considering operational costs and procurement
costs.
. Consider the duration of the individual pump duty points.
. Match the equipment to the system needs for maximum benefit.
. Match the pump type to the intended duty.
. Do not oversize the pump.
. Match the driver type to the intended duty.
. Specify motors to have high efficiency.
. Monitor and sustain the pump and the system to maximize benefit.
. Consider the energy wasted using control valves.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 62 Chapter 2

Figure 2.13 Typical pump characteristics.

2.12.2 Pump System Interaction

The actual operating point on the pump system characteristic curve is defined by its
interaction with the operating characteristics of the installed system (Fig. 2.13).
The system characteristics will consist of:
. The total static head, being the difference in elevation between the
upstream and downstream controls (generally represented by reservoir
levels),
. The energy losses in the system (generally pipe friction), which are
normally expressed as a function of velocity head.
. The interaction point of these curves represents the actual operating
point (as shown later), defining the Head supplied by the pump and the
Discharge of the system. The efficiency of the pump under these
conditions will also be defined.
Note that the efficiency of the pump at this operating point is the critical
parameter in pump selection and suitability for a particular system (Figs. 2.14
and 2.15).

2.13 LIFE CYCLE ANALYSIS

Over the economic life of a pumped supply system, a number of design parameter
will change. System behavior for all possible operating environments is needed
(Fig. 2.16). Parameters that will change include:

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Pumps 63

Figure 2.14 Pump– system interaction point and pump efficiency.

Figure 2.15 Selection of pump type.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 64 Chapter 2

Figure 2.16 Variations in demand and operating characteristics.

. Seasonal variations in demand.

. Water demand increases as the system population expands.
. Increasing pipe friction as the system ages.
For all operating conditions, it is necessary to maintain pump operation
close to peak efficiency. This can be achieved using multiple pumps and timed
pumping periods.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Pumps 65

2.14 CHANGING PUMP SPEED

The most common pump – motor combination employed in water supply
operations is a close coupled system powered by an electric motor. These units
can only operate at a speed related to the frequency of the A.C. supply
(50 cycles/s or 3000 cycles/min), with the number of pairs of poles in the motor
(N) reducing the pump speed to 3000/N revolutions per minute.
Pumps driven through belt drives or powered by petrol or diesel motors are
more flexible allowing the pump speed to be adjusted to suit the operational
requirements. Analysis of system operation will require the head –discharge-
efficiency characteristic for the particular operating speed.
Given the head – discharge-efficiency characteristics for speed N (in tabular
form), the corresponding characteristics for any speed N0 can be established as
follows:
 0
N
Q0 ¼ Q flow points ð2:20Þ
N
 0 2
N
H0 ¼ H head points ð2:21Þ
N

h0 ¼ h efficiency points ð2:22Þ

The data set for the new pump speed can then be matched to the system
characteristics.

2.15 MULTIPLE PUMP OPERATION

The most common type of pumping station has two or more pumps operating in
parallel. This provides flexibility of operation in coping with a range of flow
conditions and allows maintenance of individual units while other units are in
operation.
Occasionally situations will be encountered where pumps are operated in
series to boost outputs. This is generally a temporary measure as any failure of
one unit will severely affect system operation.
Composite characteristics (head – discharge curves) are obtained by
combining the individual curves. The composite curve is employed in the same
manner (i.e., intersection with system curve) s an individual curve (Fig. 2.17).
Where pumps operate in parallel, the composite curve is obtained by
adding the flow rates for a given head.
Where pumps operate in series, the composite is obtained by adding heads
for a given flow rate.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 66 Chapter 2

Figure 2.17 Composite pump characteristics.

2.15.1 Net Positive Suction Head

Simply, NPSH is the minimum suction condition (pressure) required to prevent
pump cavitation. Conceptually, NPSH can be imagined as the pressure drop
between the pump inlet flange and the point inside the pump where the fluid
dynamic action, as it leaves the impeller, causes a pressure rise. Sufficient NPSH
allows for pumping without liquid vaporizing in the pump first-stage impeller eye
as the fluid pressure drops due to pump suction losses (Fig. 2.18).
The NPSH required is reported in head of fluid (being pumped) required at
the pump inlet. As such, NPSH required has units of length (height). Usually, the
datum line for pump NPSH is the centerline of the inlet. This is satisfactory for
small pumps. For larger pumps, the NPSH requirements reported by the
manufacturer should be verified for the datum line being discussed. The NPSH

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Pumps 67

Figure 2.18 The elements of Eq. (2.24) are illustrated with a pump taking suction from a
tower.

available differs from NPSH required. The NPSH required determined during the
manufacturers test and shown on the vendor’s pump curve is based upon a 3% head
pump differential loss. The NPSH available must be large enough to eliminate
head loss. The NPSH available is the excess of pressure over the liquid’s
vapor pressure at the pump suction flange. Except in rare circumstances,
centrifugal pumps require the NPSH available to be greater than NPSH required
to avoid cavitation during operation. Determining the NPSH available is the
process engineer’s job. Determining the NPSH required is the job of the pump
vendor.
Our concern here is the process system side of determining what NPSH is
available. Pressure balance and NPSH available derive from Bernoulli’s equation
for a stationary conduit where the total energy (head) in a system is the same for
any point along a streamline (assuming no friction losses). Incorporating friction
losses and restating the formula in a form familiar to process engineers, the NPSH
available in a system can be expressed as:
 
2:31ðP þ Pa 2 Pv Þ V2
NPSHa ¼ þ S2B2Lþ ð2:23Þ
g 2g
where NPSHa is the net positive suction head available (ft); P, pressure above
liquid (psi gage); Pa, atmospheric pressure (psi); Pv, vapor pressure of liquid at
pumping conditions (psia); g, specific gravity of liquid at pumping conditions;

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 68 Chapter 2

S, static height of liquid from grade (ft); B, distance of pump centerline (suction
nozzle centerline for vertical pumps); L, suction system friction losses (ft of
pumping liquid); and V is the average liquid velocity at pump suction nozzle (ft/s).
Converting to absolute pressures, fluid density and resetting the datum line
to the pump centerline results in:
144ðPabs 2 Pv Þ V2
NPSHa ¼ þH2Lþ ð2:24Þ
r 2g
where Pabs is the pressure above liquids (psia); r, fluid density (lb/ft3); and H is
the static height of liquid between liquid level and pump suction centerline
(datum line), ft.

Illustrative Example 2.1: A centrifugal pump runs at a tip speed of 12 m/s

and a flow velocity of 1.5 m/s. The impeller diameter is 1.2 m and delivers
3.8 m3/min of water. The outlet blade angle is 288 to the tangent at the impeller
periphery. Assuming that the fluid enters in the axial direction and zero slip,
calculate the torque delivered by the impeller.
Solution:
From Fig. 2.2, for zero slip b2 ¼ b12. Using Eq. (2.1), the Euler head
H ¼ E ¼ (U2Cw2 2 U1Cw1)/g. Since Cw1 ¼ 0, as there is no inlet whirl
component, head H is given by
   
U 2 C w2 U 2 1:5 12 1:5
H¼ ¼ U2 2 ¼ 12 2
g g tan 288 9:81 tan 288

¼ 11:23 m

J 1000ð9:81Þð3:8Þð11:23Þ
Power delivered ¼ rgQH ¼
s 60ð1000Þ

¼ 6:98 kW
Torque delivered ¼ Power/angular velocity ¼ 6980(0.6)/12 ¼ 349 Nm.

Illustrative Example 2.2: A fluid passes through an impeller of 0.22 m

outlet diameter and 0.1 m inlet diameter. The impeller is rotating at 1250 rpm, and
the outlet vane angle is set back at an angle of 228 to the tangent. Assuming that
the fluid enters radially with velocity of flow as 3.5 m/s, calculate the head
imparted to a fluid.
Solution:
Since fluid enters in the radial direction, Cw1 ¼ 0, a1 ¼ 908, b2 ¼ 228,
Ca1 ¼ 3.5 m/s ¼ Ca2

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Pumps 69

Head developed H ¼ Cw2U2/g

Impeller tip speed, U 2 ¼ pDN
60 ¼
pð0:22Þð1250Þ
60 ¼ 14:40 m/s
Whirl velocity at impeller outlet, from velocity diagram,

C w2 ¼ U 2 2 ðC a2 /tan b2 Þ
¼ 14:40 2 ð3:5/tan 228 Þ ¼ 5:74 m/s

Therefore, the head imparted is given by

H ¼ 5:74ð14:40Þ/9:81 ¼ 8:43 m

Design Example 2.3: A centrifugal pump impeller runs at 1400 rpm, and
vanes angle at exit is 258. The impeller has an external diameter of 0.4 m and an
internal diameter of 0.2 m. Assuming a constant radial flow through the impeller
at 2.6 m/s, calculate (1) the angle made by the absolute velocity of water at exit
with the tangent, (2) the inlet vane angle, and (3) the work done per kg of water
(Fig. 2.19).
Solution:
1. Impeller tip speed is given by

pD2 N pð0:4Þð1400Þ
U2 ¼ ¼ ¼ 29:33 m/s
60 60

Whirl velocity at impeller tip

C r2 2:6
Cw2 ¼ U 2 2 ¼ 29:33 2 ¼ 23:75 m/s
tan b2 tan 258

Figure 2.19 Velocity triangle at outlet.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 70 Chapter 2

Now, from the velocity triangle at impeller tip,

C r2 2:6
tan a2 ¼ ¼ ¼ 0:1095
Cw2 23:75
Therefore, a2 ¼ 6.258.
2. Impeller velocity at inlet
pD1 N pð0:2Þ1400
U1 ¼ ¼ ¼ 14:67 m/s
60 60
Cr1 2:6
tan b1 ¼ ¼ ¼ 0:177
U 1 14:67
Therefore, b1 ¼ 10.058.
3. Work done per kg of water is given by
C w2 U 2 ¼ 23:75ð29:33Þ ¼ 696:59 Nm ¼ 696:59 J:

Design Example 2.4: A centrifugal pump impeller has a diameter of 1.2 m;

rpm 210; area at the outer periphery 0.65 m2; angle of vane at outlet 258, and ratio
of external to internal diameter 2:1. Calculate (1) the hydraulic efficiency, (2)
power, and (3) minimum speed to lift water against a head of 6.2 m. Assume that
the pump discharges 1550 l/s (Fig. 2.20).
Solution:
1. Here, Q ¼ 1550 l/s, b2 ¼ 258, H ¼ 6:2 m; D2 /D1 ¼ 2; D2 ¼ 1:2 m,
N ¼ 210 rpm; A ¼ 0:65 m2 .
Velocity of flow at impeller tip
Q 1550
C r2 ¼ ¼ ¼ 2:385 m/s
A 1000ð0:65Þ

Figure 2.20 Velocity triangle at impeller outlet.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Pumps 71

Impeller tip speed,

pð1:2Þð210Þ
U2 ¼ ¼ 13:2 m/s
60
C r2 2:385
C w2 ¼ U 2 2 ¼ 13:2 2 ¼ 8:09 m/s
tan b2 tan 258
Assuming radial entry, theoretical head is given by
Cw2 U 2 8:09ð13:2Þ
H¼ ¼ ¼ 10:89 m
9:81 9:81
Assuming slip factor, s ¼ 1, hydraulic efficiency is given by
6:2ð100Þ
hh ¼ ¼ 56:9%:
10:89
2. Power P ¼ (1550)(10.89)(9.81)/1000 ¼ 165.59 kW.
3. Centrifugal head is the minimum head. Therefore,
U 22 2 U 21
¼ 6:2
2g
It is given that U1 ¼ U2/2. Therefore,
U 22 2 0:25U 22 ¼ 2ð9:81Þð6:2Þ
i.e., U2 ¼ 12.74 m/s
Hence, minimum speed is ¼ 12.74(60)/p(1.2) ¼ 203 rpm.

Illustrative Example 2.5: A centrifugal pump is required to pump water

against a total head of 35 m at the rate of 45 l/s. Find the horsepower of the pump,
if the overall efficiency is 60%.
Solution:
Total head; H ¼ 35m
Discharge; Q ¼ 45l/s ¼ 0:045 m3 /s
Overall efficiency;ho ¼ 60% ¼ 0:60
rgQH J rgQH
Power; P ¼ ¼ kW
ho s 1000ho
9:81ð0:045Þð35Þ
¼
0:746ð0:60Þ
¼ 34:5 hp

Illustrative Example 2.6: A centrifugal pump impeller has 0.3 m inlet

diameter and 0.6 m external diameters. The pump runs at 950 rpm, and the entry

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 72 Chapter 2

Figure 2.21 Velocity triangle at impeller outlet and inlet.

of the pump is radial. The velocity of flow through the impeller is constant at
3.5 m/s. The impeller vanes are set back at angle of 468 to the outer rim. Calculate
(1) the vane angle at inlet of a pump, (2) the velocity direction of water at outlet,
and (3) the work done by the water per kg of water (Fig. 2.21).
Solution:
1. Velocity of flow, Cr1 ¼ Cr2 ¼ 3.5 m/s. Let a1 be the vane angle at inlet.
Tangential velocity of impeller at inlet
pD1 N pð0:3Þð950Þ
U1 ¼ ¼ ¼ 14:93 m/s
60 60
From inlet velocity triangle
Cr1 3:5
tan a1 ¼ ¼ ¼ 0:234
U 1 14:93
Therefore, a1 ¼ 13.198.
2. Tangential velocity of impeller at outlet
pD2 N pð0:6Þð950Þ
U2 ¼ ¼ ¼ 29:86 m/s
60 60
For velocity of whirl at impeller outlet, using velocity triangle at outlet
C r2 3:5
C w2 ¼ U 2 2 ¼ 29:86 2 ¼ 26:48 m/s
tan 468 tan 468

and C 22 ¼ C 2r2 þ C 2w2 ¼ 3:52 þ 26:482 ; C 2 ¼ 26:71 m/s

where C2 is the velocity of water at outlet. Let a2 be the direction of

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Pumps 73

water at outlet, and, therefore, a2 is given by

Cr2 3:5
tan a2 ¼ ¼ ¼ 0:132
C w2 26:48
i.e., a2 ¼ 7.538.
3. Work done by the wheel per kg of water
W ¼ Cw2 U 2 ¼ 26:48ð29:86Þ ¼ 790:69 Nm

Design Example 2.7: A centrifugal pump delivers water at the rate of

8.5 m3/min against a head of 10 m. It has an impeller of 50 cm outer diameter and
25 cm inner diameter. Vanes are set back at outlet at an angle of 458, and impeller
is running at 500 rpm. The constant velocity of flow is 2 m/s. Determine (1) the
manometric efficiency, (2) vane angle at inlet, and (3) minimum starting speed of
the pump (Fig. 2.22).
Solution:
1. The manometric efficiency is given by
H
hman ¼
ðCw2 U 2 /gÞ

From outlet velocity triangle

pD2 N pð0:5Þð500Þ
U2 ¼ ¼ ¼ 13:0 m/s
60 60

Figure 2.22 Velocity triangle (a) outlet, (b) inlet.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 74 Chapter 2

Now, tan b2 ¼ Cr2/(U2 2 Cw2), or tan 458 ¼ 2/(U2 2 Cw2), or

1 ¼ 2/(13 2 Cw2), Cw2 ¼ 11 m/s.
H 10ð9:81Þ
Hence, hman ¼ ¼ ¼ 68:6%
ðC w2 U 2 /gÞ 11ð13Þ
2. Vane angle at inlet b1
Cr1
tan b1 ¼ and U 1 ¼ 0:5 £ U 2 ¼ 6:5 m/s
U1
2
[ tan b1 ¼ ¼ 0:308
6:5
i.e., b1 ¼ 178.
3. The minimum starting speed is

pD N  2
pD N  2
2
2 601
ðU 22 2 U 21 Þ/2g ¼ H or 60
¼ 10
2ð9:81Þ

Therefore, N ¼ 618 rpm.

Illustrative Example 2.8: A centrifugal pump impeller has 0.6 m outside

diameter and rotated at 550 rpm. Vanes are radial at exit and 8.2 cm wide.
Velocity of flow through the impeller is 3.5 m/s, and velocity in the delivery pipe
is 2.5 m/s. Neglecting other losses, calculate head and power of the pump
(Fig. 2.23).
Solution:
1. D2 ¼ 0.6 m, N ¼ 550 rpm, Cr2 ¼ 3.5 m/s, Cw2 ¼ U2.
Impeller speed at outlet
pD2 N pð0:6Þð550Þ
U2 ¼ ¼ ¼ 17:29 m/s:
60 60

Figure 2.23 Velocity triangle for Example 2.8.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Pumps 75

Figure 2.24 Velocity triangles for Example 2.9.

Head, through which the water can be lifted,

V2
H ¼ C w2g U 2 2 ðneglecting all lossesÞ
2g
ð17:29Þð17:29Þ 2:52
¼ 2 ¼ 30:47 2 0:319
9:81 2ð9:81Þ
¼ 30:2 m of water:
rgQH
2. Power ¼ kW
1000
where Q ¼ pD2 b2 Cr2 ðwhere b2 is widthÞ
¼ pð0:6Þð0:082Þð3:5Þ ¼ 0:54 m3 /s

Therefore, power is given by

rgQH 1000ð9:81Þð0:54Þð30:2Þ
P¼ kW ¼ ¼ 160 kW:
1000 1000
Illustrative Example 2.9: A centrifugal pump impeller has a diameter of
1 m and speed of 11 m/s. Water enters radially and discharges with a velocity
whose radial component is 2.5 m/s. Backward vanes make an angle of 328 at exit.
If the discharge through the pump is 5.5 m3/min, determine (1) h.p. of the pump
and (2) turning moment of the shaft (Fig. 2.24).
Solution:
1. Data
D2 ¼ 1 m,
U2 ¼ 11 m/s,

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 76 Chapter 2

a1 ¼ 908,
Cr2 ¼ 2.5 m/s,
b2 ¼ 328,
Q ¼ 5.5 m3/min.
First, consider outlet velocity triangle
C r2
Cw2 ¼ U 2 2
tan b2
2:5
¼ 11 2 ¼ 7 m/s
tan 328

Power of the pump is given by

rQC w2 U 2
P¼ kW
1000
ð1000Þð5:5Þð7Þð11Þ
P¼ ¼ 7 kW:
ð60Þð1000Þ
2pNT
2. Now, h:p: ¼
60
where T is the torque of the shaft. Therefore,
h:p: £ 60
T¼
2pN
But U 2 ¼ pD602 N or N ¼ 60£U
pD2 ¼ pð1Þ ¼ 210 rpm
2 60ð11Þ

Hence,
ð7Þð1000Þð60Þ
T¼ ¼ 318 Nm/s:
2pð210Þ

Illustrative Example 2.10: A centrifugal pump running at 590 rpm and

discharges 110 l/min against a head of 16 m. What head will be developed and
quantity of water delivered when the pump runs at 390 rpm?
Solution:
N1 ¼ 590,
Q1 ¼ 110 l/min ¼ 1.83 l/s,
H1 ¼ 16 m,
N2 ¼ 390 rpm,
H2 ¼ ?

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Pumps 77

pffiffiffiffiffiffi pffiffiffiffiffiffi
H1 H2
As N1 ¼ N2

pffiffiffiffiffi pffiffiffiffiffiffi
Then, 16 ¼ H 2
590 390

Therefore, H2 ¼ 6.98 m
Therefore, head developed by the pump at 390 rpm ¼ 6.98 m. In order to
find discharge through the pump at 390 rpm, using the Eq. (1.11)
pffiffiffiffiffiffi pffiffiffiffiffiffi
N 1 Q1 N 2 Q2
¼
H 3/4
1 H 23/4
pffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffi pffiffiffiffiffiffi
590 1:83 390 Q2 798:14 390 Q2
¼ or ¼
ð16Þ3/4 ð6:98Þ3/4 8 4:29
pffiffiffiffiffiffi
Q2 ¼ 1:097
i.e., Q ¼ 1.203 l/s

Illustrative Example 2.11: The impeller of a centrifugal pump has outlet

diameter of 0.370 m, runs at 800 rpm, and delivers 30 l/s of water. The radial
velocity at the impeller exit is 2.5 m/s. The difference between the water levels at
the overhead tank and the pump is 14 m. The power required to drive the pump
is 8 hp, its mechanical and volumetric effectiveness being 0.96 and 0.97,
respectively. The impeller vanes are backward curved with an exit angle of
458. Calculate (1) ideal head developed with no slip and no hydraulic losses and
(2) the hydraulic efficiency.
Solution:
1. Impeller tip speed
pD2 N
U2 ¼
60

or
pð0:37Þð800Þ
U2 ¼ ¼ 15:5 m/s:
60

As the radial velocity at the impeller exit ¼ 2.5 m/s.

Therefore, C w2 ¼ U 2 2 tanCr2b2 ¼ 15:5 2 tan2:5458 ¼ 13 m/s.
When there is no slip, the head developed will be
Cw2 U 2 ð13Þð15:5Þ
H¼ ¼ ¼ 20:54 m
g 9:81

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 78 Chapter 2

If there are no hydraulic internal losses, the power utilized by the pump
will be:
P ¼ ð0:96Þð8Þ ¼ 7:68 hp
Q 0:03
Theoretical flow rate ¼ ¼ ¼ 0:031 m3 /s
hv 0:97
Ideal head, Hi, is given by
ð7:68Þð0:746Þ
Hi ¼ ¼ 18:84 m:
ð9:81Þð0:031Þ

2. The hydraulic efficiency is

H 14
hh ¼ ¼ ¼ 0:746 or 74:3%:
H i 18:84

Illustrative Example 2.12: The impeller of a centrifugal pump has outer

diameter of 1.06 m and speed is 56 m/s. The blades are backward curved and
they make an angle of 208 with the wheel tangent at the blade tip. If the radial
velocity of the flow at the tip is 7.5 m/s and the slip factor is 0.88. Determine
(1) the actual work input per kg of water flow and (2) the absolute velocity of
fluid at the impeller.

Solution:
1. Exit blade angle, b2 ¼ 208
Cr2 7:5
[ C w2 ¼ U 2 2 ¼ 56 2 ¼ 35:4 m/s
tan b2 tan 208

Using slip factor, s ¼ 0.88, the velocity whirl at exit is, Cw2 ¼
s £ 35.4 ¼ 0.88 £ 35.4 ¼ 31.2 m/s.
Work input per kg of water flow
Cw2 U 2 ð56Þð31:2Þ
W¼ ¼ ¼ 1:75 kJ/kg:
1000 1000
2. Absolute velocity at impeller tip
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

2 ffi q
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C2 ¼ C r2 þ C 2W2 ¼ 7:52 þ 31:22
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C 2 ¼ 56:25 þ 973:44 ¼ 32:09 m/s

Design Example 2.13: A centrifugal pump impeller of 260 mm

diameter runs at 1400 rpm, and delivers 0.03 m3/s of water. The impeller has

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Pumps 79

a backward curved facing blades inclined at 308 to the tangent at outlet. The
blades are 20 mm in depth at the outlet, and a slip factor of 0.78 may be
assumed. Calculate the theoretical head developed by the impeller, and the
number of impeller blades.

Solution:
Assuming the blades are of infinitesimal thickness, the flow area is
given by
A ¼ impeller periphery £ blade depth
¼ p £ 0:26 £ 0:02 ¼ 0:0163 m2

Flow velocity is given by

Q 0:03
C r2 ¼ ¼ ¼ 1:84 m/s
A 0:0163
Impeller tip speed, U2, is
pD2 N pð0:26Þð1400Þ
U2 ¼ ¼ ¼ 19:07 m/s
60 60
Absolute whirl component, Cw2 is given by
C r2 1:84
C w2 ¼ U 2 2 ¼ 19:07 2 ¼ 15:88 m/s
tan 308 tan 308
Using Euler’s equation, and assuming Cw1 ¼ 0 (i.e., no whirl at inlet)
U 2 Cw2 ð19:07Þð15:88Þ
H¼ ¼ ¼ 30:87 m
g 9:81
Theoretical head with slip is H ¼ 0.78 £ 30.87 ¼ 24.08 m.
To find numbers of impeller blades, using Stanitz formula
0:63p 0:63p 0:63p
Slip factor, s ¼ 1 2 or 0:78 ¼ 1 2 or ¼12
n n n
0:78 ¼ 0:22
0:63p
[ n¼ ¼9
0:22
Number of blades required ¼ 9
Design Example 2.14: A centrifugal pump impeller runs at 1500 rpm, and
has internal and external diameter of 0.20 m and 0.4 m, respectively, assuming
constant radial velocity at 2.8 m/s and the vanes at the exit are set back at an angle
of 308. Calculate (1) the angle absolute velocity of water at exit makes with the
tangent, (2) inlet vane angle, and (3) the work done per kg of water.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 80 Chapter 2

Solution:
1. D1 ¼ 0.2 m, D2 ¼ 0.4 m, N ¼ 1500 rpm, Cr2 ¼ 2.8 m/s, b2 ¼ 308.
Impeller tip speed, U2, is
pD2 N pð0:4Þð1500Þ
U2 ¼ ¼ ¼ 31:43 m/s
60 60
Whirl component of absolute velocity at impeller exit is
C r2 2:8
C w2 ¼ U 2 2 ¼ 31:43 2 ¼ 26:58 m/s
tan 308 tan 308
2:8
tan a2 ¼ ¼ 0:1053
26:58
i.e., a2 ¼ 68.
2. Impeller speed at inlet
pD1 N pð0:2Þð1500Þ
U1 ¼ ¼ ¼ 15:7 m/s
60 60
2:8
tan b1 ¼ ¼ 0:178
15:7
i.e., b1 ¼ 10.18.
3. Work done per kg of water
C w2 U 2 ¼ 26:58 £ 31:43 ¼ 835:4 Nm:

Design Example 2.15: An axial flow pump discharges water at the rate
of 1.30 m3/s and runs at 550 rpm. The total head is 10 m. Assume blade
velocity ¼ 22 m/s, the flow velocity ¼ 4.5 m/s, hydraulic efficiency ¼ 0.87, and
the overall pump efficiency ¼ 0.83, find (1) the power delivered to the water,
and power input, (2) the impeller hub diameter and tip diameter, and (3) the inlet
and outlet blade angles for the rotor.

Solution:
1. Power delivered to the water
P ¼ rgHQ/1000 kW
¼ ð9:81Þð1:30Þð10Þ ¼ 127:53 kW
Power input to the pump
127:53
P¼ ¼ 153:65 kW:
0:83

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Pumps 81

2. Rotor tip diameter is given by

60U 2 ð60Þð22Þ
D2 ¼ ¼ ¼ 0:764 m
pN pð550Þ
Rotor hub diameter
Q 1:3
D21 ¼ D22 2 ¼ 0:7642 2 ¼ 0:216 m
ðp/4Þ £ C a ðp/4Þð4:5Þ
i.e., D1 ¼ 0.465 m.
3. Rotor velocity at hub is given by
D1 ð0:465Þð22Þ
U1 ¼ U2 ¼ ¼ 13:39 m/s
D2 0:764
Since, the axial velocity is constant, we have: rotor inlet angle at tip
a1t ¼ tan 21 ðCa /U 1 Þ ¼ tan 21 ð4:5/13:39Þ ¼ 18:588
Rotor outlet angle
a2t ¼ tan 21 ðCa /U 2 Þ ¼ tan 21 ð4:5/22Þ ¼ 11:568:

Design Example 2.16: A single stage, radial flow, and double suction
centrifugal pump having the following data:

Discharge 72 l/s
Inner diameter 90 mm
Outer diameter 280 mm
Revolution/minute 1650
Head 25 m
Width at inlet 20 mm/side
Width at outlet 18 mm/side
Absolute velocity angle at inlet 908
Leakage losses 2 l/s
Mechanical losses 1:41 kW
Contraction factor due to vane thickness 0:85
Relative velocity angle measured
from tangential direction 358
Overall efficiency of the pump 0:56

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 82 Chapter 2

Determine (1) inlet vane angle, (2) the angle at which the water leaves the wheel,
(3) the absolute velocity of water leaving impeller, (4) manometric efficiency,
and (5) the volumetric and mechanical efficiencies.
Solution:
Total quantity of water to be handled by the pump
Qt ¼ Qdel þ Qleak
¼ 72 þ 2 ¼ 74

Total quantity of water per side ¼ 74/2 ¼ 37 l/s

1. Impeller speed at inlet
pD1 N pð0:09Þð1650Þ
U1 ¼ ¼ ¼ 7:78 m/s
60 60

Flow area at inlet ¼ PD1 b1 £ contraction factor

¼ ðPÞð0:09Þð0:02Þð0:85Þ ¼ 0:0048 m2

Therefore, the velocity of flow at inlet

Q 37 £ 1023
C r1 ¼ ¼ ¼ 7:708 m/s
Area of flow 0:0048

From inlet velocity triangle

Cr1 7:708
tan b1 ¼ ¼ ¼ 0:9907
U1 7:78
b1 ¼ 44:738
2. Area of flow at outlet
A2 ¼ P £ D2 £ b2 £ contraction factor
Where b2 ¼ 18/2 ¼ 9 mm for one side.
So, A2 ¼ (P)(0.28)(0.009) (0.85) ¼ 0.0067 m2.
Therefore, the velocity of flow at outlet
Q 37 £ 1023
C r2 ¼ ¼ ¼ 5:522 m/s
Area of flow 0:0067

The impeller speed at outlet

pD2 N pð0:28Þð1650Þ
U2 ¼ ¼ ¼ 24:2 m/s
60 60

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Pumps 83

Now using velocity triangle at outlet

C r2 5:522
tan b2 ¼ ¼
U 2 2 Cw2 24:2 2 C w2
Cw2 ¼ 16:99 m/s
Further,
Cr2 5:522
tan a2 ¼ ¼ ¼ 0:325
Cw2 16:99
a2 ¼ 188:

3. The absolute velocity of water leaving the impeller

Cw2 16:99
C2 ¼ ¼ ¼ 17:8 m/s:
Cos a2 Cos 188

4. The manometric efficiency

ðgÞðH mano Þ 9:81 £ 25
hmano ¼ ¼ ¼ 0:596:
ðU 2 ÞðC W2 Þ 24:2 £ 16:99

5. The volumetric efficiency

Q2 72
hv ¼ ¼ ¼ 0:973
QTotal 74
Water power ¼ rgQH
¼ 1000 £ 9:81 £ 72 £ 25/1000
¼ 17:66 kW
Water power 17:66
Shaft power ¼ ¼ ¼ 31:54 kW
ho 0:56

Mechanical efficiency is
Ps 2 Ploss 31:54 2 1:41
hm ¼ ¼ ¼ 0:955 or 95:5%:
Ps 31:54

Illustrative Example 2.17: A single stage centrifugal pump is designed to

give a discharge of Q when working against a manometric head of 20 m. On test,
it was found that head actually generated was 21.5 m for the designed discharge,
Q. If it is required to reduce the original diameter 32 cm without reducing the
speed of the impeller, compute the required diameter to be reduced.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 84 Chapter 2

Solution:
Head generated by the pump
U 2 ðpDN/60Þ2
H¼ ¼
2g 2g
or
H / D2
 2
H D
¼
H0 D0

H ¼ 21:5 m; H 0 ¼ 20 m; D ¼ 32 cm

So,
 0 1/2  
0 H 20 2
D ¼D ¼ 32 ¼ 30:86 cm
H 21:5

Design Example 2.18: A two stage centrifugal pump is designed to

discharge 55 l/s at a head of 70 m. If the overall efficiency is 76% and specific
speed per stage about 38, find (1) the running speed in rpm and (2) the power
required to run pump.
If the actual manometric head developed is 65% of the theoretical head,
assuming no slip, the outlet angle of the blades 288, and radial velocity at exit
0.14 times the impeller tip speed at exit, find the diameter of impeller.

Solution:
1. The specific speed is
pffiffiffiffi
N Q
N s ¼ 3/4
H
N s H 3/4 38ð70/2Þ3/4 546:81
N¼ pffiffiffiffi ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ ¼ 2327 rpm:
Q 55 £ 10 23 0:235

2. Q ¼ 55 £ 1023 m3/s
Power required to drive pump

rgQH 1000 £ 9:81 £ 55 £ 1023 £ 70

¼ ¼
0:76 0:76 £ 1000
¼ 49:7 kW

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Pumps 85

Hmano ¼ 0.65 H.
Here b2 ¼ 288 and C r2 ¼ 0:14U 2 .
From velocity triangle at outlet
C r2
tan b2 ¼
U 2 2 CW2
0:14U 2
or tan 288 ¼
U 2 2 C W2
U2 0:5317
¼ ¼ 3:798 ðAÞ
U 2 2 C W2 0:14
As the flow at entrance is radial and a1 ¼ 908, the fundamental
equation of pump would be
H mano U 2 C W2
¼
hmano g
Where hmano manometric efficiency of pump which is 65%.
35
Therefore, 0:65 ¼ U 2 Cg W2
35 £ 9:81
U 2 CW2 ¼
0:65
528:23
CW2 ¼ ðBÞ
U2
Substituting for CW2 in Eq. (A) and solving
U2
¼ 3:798
U 2 2 528:23
U2

U 2 ¼ 26:78 m/s:
Also,
pD2 N
U2 ¼
60
or 26:78 ¼ p£D260£2327

D2 ¼ 0:2197 m or 21:97 cm:

Design Example 2.19: Two multistage centrifugal pumps are used in series
to handle water flow rate of 0.0352 m3/s, and total head required is 845 m. Each
pump is required to produce a head of half the total and run at 1445 rpm. If the
impeller in all the stages is identical and specific speed is 14, determine (1) head
developed per stage and the required number of stages in each pump,
(2) The required impeller diameters assuming the speed ratio based on the outer

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 86 Chapter 2

tip diameter to be 0.96 and the shaft power input, if the overall efficiency of each
pump is 0.75.
Solution:
Head developed in each stage is
pffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffi
N Q 1445 0:0352
H ¼
3/4
¼
Ns 14
H ¼ 51:93 m
Total head required ¼ 845 m (of water)
Number of stages needed ¼ 51:93
845
¼ 16
Number of stages in each pump ¼ 8
Impeller speed at tip is
U 2 ¼ 0:96ð2gHÞ0:5
¼ 0:96½2 £ 9:81 £ 51:930:5
¼ 30:6 m/s
Impeller diameter at tip, D2 ¼ p £ 60 £ 30:6 £ 1445. But
pD2 N
U2 ¼
60
or
U 2 £ 60 30:6 £ 60
D2 ¼ ¼ ¼ 0:4043 m or 40:43 cm:
p £ 1445 p £ 1445

Design Example 2.20: A centrifugal pump is required to be made to lift

water through 105 m heights from a well. Number of identical pumps having their
designed speed 900 rpm and specific speed 700 rpm with a rated discharge of
5500 l/min are available. Determine the number of pumps required and how they
should be connected?
Solution:
Specific speed for a single impeller is given by
pffiffiffiffi
N Q
N s ¼ 3/4
H
5500
Given; N s ¼ 700; H ¼ 105 N ¼ 900; and Q ¼ ¼ 91:67 l/s
60
Substituting,
pffiffiffiffiffiffiffiffiffiffiffi
900 91:67
700 ¼ ; H ¼ 28 m
H 3/4

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Pumps 87

Hence number of stages required

Total head to be developed
¼
Head per stage
105
¼ ¼ 4 stages in series
28

Design Example 2.21: The specific speed of an axial flow pump

impeller is 1150 and velocity of flow is 2.5 m/s. The outer and inner diameters
of the impeller are 0.90 m and 0.45 m, respectively. Calculate the suitable
speed of the pump to give a head of 5.5 m. Also, calculate vane angle at the
entry of the pump.
Solution:
Given,
D2 ¼ 0.9 m, D1 ¼ 0.45 m, Ns ¼ 1150, Cr ¼ 2.5 m/s, H ¼ 5.5 m.

As discharge; Q ¼ area of flow £ velocity of flow

¼ p4 ð0:92 2 0:452 Þ £ 2:5 ¼ 1:193 m3 /s
¼ 1193 l/s
Also,
pffiffiffiffi
N Q
N s ¼ 3/4
H

or
pffiffiffiffiffiffiffiffiffiffi
N 1193
1150 ¼
ð5:5Þ3/4

ð5:5Þ3/4 £ 1150
N¼ pffiffiffiffiffiffiffiffiffiffi ¼ 120 rpm
1193
In order to find vane angle at entry, using velocity triangle at inlet,
pD1 N p £ 0:45 £ 120
U1 ¼ ¼ ¼ 2:82 m/s
60 60
Cr1 2:5
tan a1 ¼ ¼ 0:8865
U 1 2:82
i.e.,
a ¼ 41:568:

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 88 Chapter 2

PROBLEMS
2.1 A centrifugal pump of 25 cm impeller diameter running at 1450 rpm,
develops a head of 15 m. If the outlet flow area is 480 cm2, and discharging
water 0.12 m3/s, and loss of head in the pump can be taken as 0.003C21 , find
the outlet blade angle.
(148)
2.2 A centrifugal pump having vane angles at inlet and outlet are 258 and 308,
respectively. If internal and external diameters of impeller are 0.15 and
0.30 m, respectively, calculate the work done per kg of water. Assume
velocity of flow constant.
(197.18 Nm)
2.3 A centrifugal pump discharges 50 liters/second of water against a total head
of 40 m. Find the horsepower of the pump, if the overall efficiency is 62%.
(42 hp)
2.4 A centrifugal pump delivers 26 l/s against a total head of 16 m at 1450 rpm.
The impeller diameter is 0.5 m. A geometrically similar pump of 30 cm
diameter is running at 2900 rpm. Calculate head and discharge required
assuming equal efficiencies between the two pumps.
(11.52 m, 11.23 l/s)
2.5 A centrifugal pump is built to work against a head of 20 m. A model of this
pump built to one-fourth its size is found to generate a head of 7 m when
running at its best speed of 450 rpm and requires 13.5 hp to run it. Find the
speed of the prototype.
(190 rpm)
2.6 Derive the expression for power required for a pump when it discharges a
liquid of specific weight w at the rate of Q against a head of H.
2.7 Show that the pressure rise in an impeller of a centrifugal pump is given by
C2r1 þU 22 2C 2r2 cosec2 b2
2g (where Cr1 ¼ velocity of flow at inlet, U2 ¼ blade velocity
at outlet, Cr2 ¼ velocity of flow at outlet, and b2 ¼ blade angle at outlet).
Assuming that friction and other losses are neglected.
2.8 Derive an expression for static head developed by a centrifugal pump
having radial flow at inlet.
2.9 A centrifugal pump discharges 0.15 m3/s of water against a head of 15 m.
The impeller has outer and inner diameter of 35 and 15 cm, respectively.
The outlet vanes are set back at an angle 408. The area of flow is constant
from inlet to outlet and is 0.06 m2. Calculate the manometric efficiency

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Pumps 89

and vane angle at inlet if the speed of the pump is 960 rpm. Take slip
factor ¼ 1.
(57.3%, 188)
2.10 A centrifugal pump of 35 cm diameter running at 1000 rpm develops a
head of 18 m. The vanes are curved back at an angle of 308 to the tangent at
outlet. If velocity flow is constant at 2.4 m/s, find the manometric
efficiency of the pump.
(76.4%)
2.11 An axial flow pump is required to deliver 1 m3/s at 7 m head while
running at 960 rpm. Its outer diameter is 50 and hub diameter is
25 cm. Find (1) flow velocity, which is assumed to be constant from
hub to tip and (2) power required to drive the pump if overall
efficiency is 84%.
(6.791 m/s, 81.75 kW)
2.12 An axial flow pump has the following data:

Rotational speed 750 rpm

Discharge of water 1:75 m3 /s
Head 7:5 m
Hub to runner diameter ratio 0:45

Through flow velocity is 0.35 times the peripheral velocity. Find the
diameter and minimum speed ratio.
(0.59 m, 0.83)
2.13 In an axial flow pump, the rotor has an outer diameter of 75 cm and an
inner diameter of 40 cm; it rotates at 500 rpm. At the mean blade radius,
the inlet blade angle is 128 and the outlet blade angle is 158. Sketch the
corresponding velocity diagrams at inlet and outlet, and estimate from
them (1) the head the pump will generate, (2) the discharge or rate of flow
in l/s, (3) the shaft h.p. input required to drive the pump, and (4) the
specific speed of the pump. Assume a manometric or hydraulic efficiency
of 88% and a gross or overall efficiency of 81%.
(19.8 m; 705 l/s; 230 hp; 45)
2.14 If an axial flow pump delivers a discharge Q against a head H when
running at a speed N, deduce an expression for the speed of a
geometrically similar pump of such a size that when working against unit
head, it will transmit unit power to the water flowing through it. Show that
this value is proportional to the specific speed of the pump.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 90 Chapter 2

NOTATION
b width of the diffuser passage
Cw2 tangential components of absolute velocity corresponding to
the angle b2
E Euler head
H total head developed by the pump
Hi total head across the impeller
Nsuc Suction specific speed
m mass flow rate
n number of vanes
Ps shaft power input
Q flow rate
r radius
U impeller speed
V relative velocity
a absolute velocity angle
b relative velocity angle
hc casing efficiency
hR hydraulic efficiency
hi impeller efficiency
hm mechanical efficiency
ho overall efficiency
hv volumetric efficiency
r density of liquid
s slip factor
v angular velocity

SUFFIXES
1 inlet to impeller
2 outlet from the impeller
3 outlet from the diffuser
a axial
r radial
w whirl

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 3
Hydraulic Turbines

3.1 INTRODUCTION
In a hydraulic turbine, water is used as the source of energy. Water or hydraulic
turbines convert kinetic and potential energies of the water into mechanical
power. The main types of turbines are (1) impulse and (2) reaction turbines. The
predominant type of impulse machine is the Pelton wheel, which is suitable for a
range of heads of about 150 –2,000 m. The reaction turbine is further subdivided
into the Francis type, which is characterized by a radial flow impeller, and
the Kaplan or propeller type, which is an axial-flow machine. In the sections that
follow, each type of hydraulic turbine will be studied separately in terms of the
velocity triangles, efficiencies, reaction, and method of operation.

3.2 PELTON WHEEL

An American Engineer Lester A. Pelton discovered this (Fig. 3.1) turbine in
1880. It operates under very high heads (up to 1800 m.) and requires
comparatively less quantity of water. It is a pure impulse turbine in which a jet of
fluid delivered is by the nozzle at a high velocity on the buckets. These buckets
are fixed on the periphery of a circular wheel (also known as runner), which is
generally mounted on a horizontal shaft. The primary feature of the impulse

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 92 Chapter 3

Figure 3.1 Single-jet, horizontal shaft Pelton turbine.

turbine with respect to fluid mechanics is the power production as the jet is
deflected by the moving vane(s).
The impact of water on the buckets causes the runner to rotate and thus
develops mechanical energy. The buckets deflect the jet through an angle of
about 160 and 1658 in the same plane as the jet. After doing work on the buckets
water is discharged in the tailrace, and the whole energy transfer from nozzle
outlet to tailrace takes place at constant pressure.
The buckets are so shaped that water enters tangentially in the middle and
discharges backward and flows again tangentially in both the directions to avoid
thrust on the wheel. The casing of a Pelton wheel does not perform any hydraulic
function. But it is necessary to safeguard the runner against accident and also to
prevent the splashing water and lead the water to the tailrace.

3.3 VELOCITY TRIANGLES

The velocity diagrams for the Pelton wheel are shown in Fig. 3.2.
Since the angle of entry of the jet is nearly zero, the inlet velocity triangle is
a straight line, as shown in Fig. 3.2. If the bucket is brought to rest, then the
relative fluid velocity, V1, is given by
V 1 ¼ jet velocity 2 bucket speed
¼ C1 2 U 1
The angle turned through by the jet in the horizontal plane during its passage over
the bucket surface is a and the relative velocity at exit is V2. The absolute

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 93

Figure 3.2 Velocity triangles for a Pelton wheel.

velocity, C2, at exit can be obtained by adding bucket speed vector U2 and
relative velocity, V2, at exit.
Now using Euler’s turbine Eq. (1.78)
W ¼ U 1 CW1 2 U 2 C W2
Since in this case CW2 is in the negative x direction,
W ¼ U fðU þ V 1 Þ þ ½V 1 cosð180 2 aÞ 2 U g
Neglecting loss due to friction across the bucket surface, that is, V1 ¼ V2,
then
W ¼ UðV 1 2 V 1 cos aÞ
Therefore
E ¼ UðC 1 2 UÞð1 2 cos aÞ/g ð3:1Þ
the units of E being Watts per Newton per second weight of flow.
Eq. (3.1) can be optimized by differentiating with respect to U, and
equating it to zero.
Therefore
dE
¼ ð1 2 cos aÞðC 1 2 2UÞ/g ¼ 0
dU
Then
C1 ¼ 2U or U ¼ C1 /2 ð3:2Þ

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 94 Chapter 3

Substituting Eq. (3.2) into Eq. (3.1) we get

Emax ¼ C21 ð1 2 cos aÞ/4g

In practice, surface friction is always present and V1 – V2, then Eq. (3.1)
becomes
E ¼ UðC 1 2 UÞð1 2 k cos aÞ/g ð3:3Þ
where k ¼ VV 21
Introducing hydraulic efficiency as
Energy Transferred
hh ¼
Energy Available in jet

E
i:e: hh ¼ ð3:4Þ
ðC 21 /2gÞ
if a ¼ 1808, the maximum hydraulic efficiency is 100%. In practice, deflection
angle is in the order of 160 – 1658.

3.4 PELTON WHEEL (LOSSES AND EFFICIENCIES)

Head losses occur in the pipelines conveying the water to the nozzle due to
friction and bend. Losses also occur in the nozzle and are expressed by the
velocity coefficient, Cv.
The jet efficiency (hj) takes care of losses in the nozzle and the mechanical
efficiency (hm) is meant for the bearing friction and windage losses. The overall
efficiency (ho) for large Pelton turbine is about 85– 90%. Following efficiency is
usually used for Pelton wheel.
Energy at end of the pipe
Pipeline transmission efficiency ¼
Energy available at reservoir
Figure 3.3 shows the total headline, where the water supply is from a
reservoir at a head H1 above the nozzle. The frictional head loss, hf, is the loss as
the water flows through the pressure tunnel and penstock up to entry to the nozzle.
Then the transmission efficiency is
htrans ¼ ðH 1 2 hf Þ/H 1 ¼ H/H 1 ð3:5Þ
The nozzle efficiency or jet efficiency is
Energy at nozzle outlet
hj ¼ ¼ C21 /2gH ð3:6Þ
Energy at nozzle inlet

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 95

Figure 3.3 Schematic layout of hydro plant.

Nozzle velocity coefficient

Actual jet velocity pffiffiffiffiffiffiffiffiffi
Cv ¼ ¼ C 1 = 2gH
Theoretical jet velocity
Therefore the nozzle efficiency becomes
hj ¼ C 21 =2gH ¼ C2v ð3:7Þ
The characteristics of an impulse turbine are shown in Fig. 3.4.
Figure 3.4 shows the curves for constant head and indicates that the peak
efficiency occurs at about the same speed ratio for any gate opening and that

Figure 3.4 Efficiency vs. speed at various nozzle settings.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 96 Chapter 3

Figure 3.5 Power vs. speed of various nozzle setting.

the peak values of efficiency do not vary much. This happens as the nozzle
velocity remaining constant in magnitude and direction as the flow rate changes,
gives an optimum value of U/C1 at a fixed speed. Due to losses, such as windage,
mechanical, and friction cause the small variation. Fig. 3.5 shows the curves for
power vs. speed. Fixed speed condition is important because generators are
usually run at constant speed.

Illustrative Example 3.1: A generator is to be driven by a Pelton wheel

with a head of 220 m and discharge rate of 145 L/s. The mean peripheral velocity
of wheel is 14 m/s. If the outlet tip angle of the bucket is 1608, find out the power
developed.
Solution:
Dischargerate; Q ¼ 145 L/s
Head; H ¼ 220 m
U1 ¼ U2 ¼ 14 m/s
b2 ¼ 180 2 1608 ¼ 208

Refer to Fig. 3.6

Using Euler’s equation, work done per weight mass of water per sec.
¼ ðC w1 U 1 2 C w2 U 2 Þ
But for Pelton wheel Cw2 is negative

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 97

Figure 3.6 Inlet and outlet velocity triangles.

Therefore
Work done / s ¼ ðC w1 U 1 þ C w2 U 2 Þ Nm / s
From inlet velocity triangle
C2
C w1 ¼ C1 and 1 ¼ H
2g
pffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Hence, C1 ¼ 2gH ¼ 2 £ 9:81 £ 220 ¼ 65:7 m/s
Relative velocity at inlet is
V 1 ¼ C 1 2 U 1 ¼ 65:7 2 14 ¼ 51:7 m/s
From outlet velocity triangle
V 1 ¼ V 2 ¼ 51:7 m/s(neglecting friction)
and cos b2 ¼ U 2 þC
V2
w2
or
14 þ C w2
cosð20Þ ¼
51:7
Therefore
C w2 ¼ 34:58 m/s
Hence, work done per unit mass of water per sec.

¼ ð65:7Þð14Þ þ ð34:58Þð14Þ ¼ 1403:92 Nm

ð1403:92Þð145Þ
Power developed ¼ ¼ 203:57 kW
1000

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 98 Chapter 3

Design Example 3.2: A Pelton wheel is supplied with 0.035 m3/s of water
under a head of 92 m. The wheel rotates at 725 rpm and the velocity coefficient of
the nozzle is 0.95. The efficiency of the wheel is 82% and the ratio of bucket
speed to jet speed is 0.45. Determine the following:
1. Speed of the wheel
2. Wheel to jet diameter ratio
3. Dimensionless power specific speed of the wheel
Solution:
Power developed
Overall efficiency ho ¼ Power available

[ P ¼ rgQH ho J/s ¼ rgQH ho

1000 kW
¼ 9:81ð0:035Þð92Þð0:82Þ ¼ 25:9 kW

Velocity coefficient

C1
Cv ¼ pffiffiffiffiffiffiffiffiffi
2gH
pffiffiffiffiffiffiffiffiffi
or C1 ¼ Cv 2gH ¼ 0:95½ð2Þð9:81Þð92Þ1/2 ¼ 40:36 m/s
1. Speed of the wheel is given by

U ¼ 0:45ð40:36Þ ¼ 18:16 m/s

2. If D is the wheel diameter, then

vD 2U ð2Þð18:16Þð60Þ
U¼ or D¼ ¼ ¼ 0:478 m
2 v 725ð2pÞ

Q 0:035
Jet area A¼ ¼ ¼ 0:867 £ 1023 m2
C 1 40:36

and Jet diameter, d, is given by

 1/2  1/2
4A ð4Þð0:867 £ 1023 Þ
d¼ ¼ ¼ 0:033 m
p p

D 0:478
Diameter ratio ¼ ¼ 14:48
d 0:033

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 99

3. Dimensionless specific speed is given by Eq. (1.10)

NP 1/2
N sp ¼
r 1/2 ðgHÞ5/4
     5/4
¼ 725 £ ð25:9Þð1000Þ 1/2 £ 1
60 103 ð9:81Þ £ ð92Þ
¼ ð12:08Þð5:09Þð0:0002Þ
¼ 0:0123 rev
¼ ð0:0123Þð2pÞ rad
¼ 0:077 rad

Illustrative Example 3.3: The speed of Pelton turbine is 14 m/s. The water
is supplied at the rate of 820 L/s against a head of 45 m. If the jet is deflected by
the buckets at an angle of 1608, find the hP and the efficiency of the turbine.
Solution:
Refer to Fig. 3.7
U1 ¼ U2 ¼ 14 m/s
Q ¼ 820 L/s ¼ 0.82 m3/s
H ¼ 45 m
b2 ¼ 180 2 1608 ¼ 208

Velocity of jet
pffiffiffiffiffiffiffiffiffi
C 1 ¼ Cv 2gH , assuming C v ¼ 0:98
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
¼ 0:98 ð2Þð9:81Þð45Þ ¼ 29:12 m/s

Figure 3.7 Velocity triangle for Example 3.3.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 100 Chapter 3

Assuming
b1 ¼ 1808
b2 ¼ 180 2 1608 ¼ 208
Cw1 ¼ C 1 ¼ 29:12 m/s
V 1 ¼ C1 2 U 1 ¼ 29:12 2 14 ¼ 15:12 m/s
From outlet velocity triangle,
U 1 ¼ U 2 (neglecting losses on buckets)

V 2 ¼ 15:12 m / s and U 2 ¼ 14 m/s

Cw2 ¼ V 2 cosa2 2 U 2 ¼ 15:12 cos 208 2 14

¼ 0:208 m/s
Work done per weight mass of water per sec
¼ ðCw1 þ C w2 ÞU
¼ ð29:12 þ 0:208Þ £ ð14Þ ¼ 410:6 Nm / s

[ Power developed ¼ ð410:6Þð0:82

1000
£ 103 Þ ¼ 336:7 kW

¼ 451 hP
Power developed
Efficiencyh1 ¼ Available Power

¼ ð1000Þð336:7Þ ¼ 0:930 or 93:0%

ð1000Þð9:81Þð0:82Þð45Þ

Illustrative Example 3.4: A Pelton wheel develops 12,900 kW at 425 rpm

under a head of 505 m. The efficiency of the machine is 84%. Find (1) discharge
of the turbine, (2) diameter of the wheel, and (3) diameter of the nozzle. Assume
Cv ¼ 0.98, and ratio of bucket speed to jet speed ¼ 0.46.
Solution:
Head, H ¼ 505 m.
Power, P ¼ 12,900 kW
Speed, N ¼ 425 rpm
Efficiency, ho ¼ 84%
1. Let Q be the discharge of the turbine
P
Using the relation ho ¼
9:81QH

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 101

or
12; 900 2:60
0:84 ¼ ¼
ð9:81Þð505ÞQ Q
or
Q ¼ 3:1 m3 /s
2. Velocity of jet
pffiffiffiffiffiffiffiffiffi
C ¼ Cv 2gH ðassume Cv ¼ 0:98Þ
or
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C ¼ 0:98 ð2Þð9:81Þð505Þ ¼ 97:55 m/s

Tangential velocity of the wheel is given by

U ¼ 0:46C ¼ ð0:46Þð97:55Þ ¼ 44:87 m/s
and
pDN
U¼ ; hence wheel diameter is
60
60U ð60Þð44:87Þ
D¼ ¼ ¼ 2:016 m
pN ðpÞð425Þ
3. Let d be the diameter of the nozzle
The discharge through the nozzle must be equal to the discharge of the
turbine. Therefore
Q ¼ p4 £ d 2 £ C
3:1 ¼ ðp4 Þðd 2 Þð97:55Þ ¼ 76:65 d 2
qffiffiffiffiffiffiffiffiffiffiffi
[ d ¼ 76:65 3:1 ¼ 0:20 m

Illustrative Example 3.5: A double Overhung Pelton wheel unit is to operate

at 12,000 kW generator. Find the power developed by each runner if the generator
is 95%.

Solution:
Output power ¼ 12,000 kW
Efficiency, h ¼ 95%
Therefore, power generated by the runner
12; 000
¼ ¼ 12; 632 kW
0:95

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 102 Chapter 3

Since there are two runners, power developed by each runner

12; 632
¼ ¼ 6316 kW
2
Design Example 3.6: At the power station, a Pelton wheel produces
1260 kW under a head of 610 m. The loss of head due to pipe friction between the
reservoir and nozzle is 46 m. The buckets of the Pelton wheel deflect the jet
through an angle of 1658, while relative velocity of the water is reduced by 10%
due to bucket friction. The bucket/jet speed ratio is 0.46. The bucket circle
diameter of the wheel is 890 mm and there are two jets. Find the theoretical
hydraulic efficiency, speed of rotation of the wheel, and diameter of the nozzle if
the actual hydraulic efficiency is 0.9 times that calculated above. Assume nozzle
velocity coefficient, Cv ¼ 0.98.
Solution:
Refer to Fig. 3.8.
Power output P
Hydraulic efficiency hh ¼ ¼
Energy available in the jet 0:5mC 21

At entry to nozzle

H ¼ 610 2 46 ¼ 564 m

Using nozzle velocity coefficient

pffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C1 ¼ Cv 2gH ¼ 0:98 ð2Þð9:81Þð564Þ ¼ 103:1 m/s
Now
W ¼U C 2U C
m 1 w1 2 w2

¼ U fðU þ V 1 Þ 2 ½U 2 V 2 cosð1808 2 aÞg

¼ U ½ðC 1 2 U Þð1 2 k cos aÞ where V 2 ¼ kV 1

Therefore, W/m ¼ 0.46C1(C1 2 0.46C1)(1 2 0.9 cos 1658)

Substitute the value of C1

W/m ¼ 5180:95
Power output
Theoretical hydraulic efficiency ¼
Energy available in the jet
5180:95
¼ ¼ 98%
0:5 £ 1032

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 103

Figure 3.8 Velocity triangle for Example 3.6.

Actual hydraulic efficiency ¼ ð0:9Þð0:98Þ ¼ 0:882

Wheel bucket speed ¼ ð0:46Þð103Þ ¼ 47:38 m/s

ð47:38Þð60Þ
Wheel rotational speed ¼ N ¼ ¼ 1016 rpm
ð0:445Þð2pÞ
Actual power ð1260 £ 103 Þ
Actual hydraulic efficiency ¼ ¼
energy in the jet 0:5 mC 21

Therefore, m ¼ ð1260 £ 103 Þ ¼ 269 kg/s

ð0:882Þð0:5Þð1032 Þ
For one nozzle, m ¼ 134.5 kg/s
rC1 pd 2
For nozzle diameter, using continuity equation, m ¼ rC 1 A ¼ 4
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Hence, d ¼ ð134:5Þð4Þ ¼ 0:041 m ¼ 41 mm
ðpÞð103 £ 103 Þ

Illustrative Example 3.7: A Pelton wheel has a head of 90 m and head lost
due to friction in the penstock is 30 m. The main bucket speed is 12 m/s and
the nozzle discharge is 1.0 m3/s. If the bucket has an angle of 158 at the outlet
and Cv ¼ 0.98, find the power of Pelton wheel and hydraulic efficiency.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 104 Chapter 3

Figure 3.9 Velocity triangle for Example 3.7.

Solution: (Fig. 3.9)

Head ¼ 90 m
Head lost due to friction ¼ 30 m
Head available at the nozzle ¼ 90 2 30 ¼ 60 m
Q ¼ 1 m3/s
From inlet diagram
pffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C1 ¼ Cv 2gH ¼ 0:98 £ ð2Þð9:81Þð60Þ ¼ 33:62 m/s

Therefore, V1 ¼ C1 2 U1 ¼ 33.62 2 12 ¼ 21.62 m/s

From outlet velocity triangle

V 2 ¼ V 1 ¼ 21:16 m/s (neglecting losses)

U 2 ¼ U 1 ¼ 12 m/s

Cw2 ¼ V 2 cos a 2 U 2 ¼ 21:62 cos 158 2 12 ¼ 8:88 m/s

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 105

and
Cr 2 ¼ V 2 sin a ¼ 21:62 sin 158 ¼ 5:6 m/s

Therefore,
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C2 ¼ C 2w2 þ Cr 22 ¼ ð8:88Þ2 þ ð5:6Þ2 ¼ 10:5 m/s

C21 2 C22 ð33:62Þ2 2 ð10:5Þ2

[ Work done ¼ ¼ ¼ 510 kJ/kg
2 2
Note Work done can also be found by using Euler’s equation (Cw1U1 þ
Cw2U2)
Power ¼ 510 kW
Hydraulic efficiency
work done ð510Þð2Þ
hh ¼ ¼ ¼ 90:24%
kinetic energy ð33:62Þ2

Design Example 3.8: A single jet Pelton wheel turbine runs at 305 rpm
against a head of 515 m. The jet diameter is 200 mm, its deflection inside the
bucket is 1658 and its relative velocity is reduced by 12% due to friction. Find
(1) the waterpower, (2) resultant force on the bucket, (3) shaft power if the
mechanical losses are 4% of power supplied, and (4) overall efficiency. Assume
necessary data.
Solution: (Fig. 3.10) pffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Velocity of jet, C 1 ¼ Cv 2gH ¼ 0:98 ð2Þð9:81Þð515Þ ¼ 98:5 m/s
Discharge, Q is given by
p
Q ¼ Area of jet £ Velocity ¼ £ ð0:2Þ2 ð98:5Þ ¼ 3:096 m3 /s
4
1. Water power is given by
P ¼ rgQH ¼ ð9:81Þð3:096Þð515Þ ¼ 15641:5 kW
2. Bucket velocity, U1, is given by
pffiffiffiffiffiffiffiffiffi
U 1 ¼ C v 2gH
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
¼ 0:46 ð2Þð9:81Þð515Þ ¼ 46 m/s ðassuming C v ¼ 0:46Þ

Relative velocity, V1, at inlet is given by

V 1 ¼ C1 2 U 1 ¼ 98:5 2 46 ¼ 52:5 m/s

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 106 Chapter 3

Figure 3.10 Velocity triangles for Example 3.8.

and
V 2 ¼ 0:88 £ 52:5 ¼ 46:2 m/s

From the velocity diagram

Cw2 ¼ U 2 2 V 2 cos 15 ¼ 46 2 46:2 £ 0:966 ¼ 1:37 m/s

Therefore force on the bucket

¼ rQðCw1 2 C w2 Þ ¼ 1000 £ 3:096ð98:5 2 1:37Þ
¼ 300714 N

3. Power produced by the Pelton wheel

ð300714Þð46Þ
¼ ¼ 13832:8 kW
1000

Taking mechanical loss ¼ 4%

Therefore, shaft power produced ¼ 0.96 £ 13832.8 ¼ 13279.5 kW
4. Overall efficiency
13279:5
ho ¼ ¼ 0:849 or 84:9%
15641:5

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 107

Figure 3.11 Outlines of a Francis turbine.

3.5 REACTION TURBINE

The radial flow or Francis turbine is a reaction machine. In a reaction turbine, the
runner is enclosed in a casing and therefore, the water is always at a pressure
other than atmosphere. As the water flows over the curved blades, the pressure
head is transformed into velocity head. Thus, water leaving the blade has a large
relative velocity but small absolute velocity. Therefore, most of the initial energy
of water is given to the runner. In reaction turbines, water leaves the runner at
atmospheric pressure. The pressure difference between entrance and exit points
of the runner is known as reaction pressure.
The essential difference between the reaction rotor and impulse rotor is
that in the former, the water, under a high station head, has its pressure
energy converted into kinetic energy in a nozzle. Therefore, part of the work
done by the fluid on the rotor is due to reaction from the pressure drop, and
part is due to a change in kinetic energy, which represents an impulse
function. Fig. 3.11 shows a cross-section through a Francis turbine and Fig.
3.12 shows an energy distribution through a hydraulic reaction turbine. In
reaction turbine, water from the reservoir enters the turbine casing through
penstocks.
Hence, the total head is equal to pressure head plus velocity head. Thus,
the water enters the runner or passes through the stationary vanes, which are
fixed around the periphery of runners. The water then passes immediately into
the rotor where it moves radially through the rotor vanes and exits from the
rotor blades at a smaller diameter, after which it turns through 908 into the draft
tube. The draft tube is a gradually increasing cross-sectional area passage. It
helps in increasing the work done by the turbine by reducing pressure at the
exit. The penstock is a waterway, which carries water from the reservoir to
the turbine casing. The inlet and outlet velocity triangles for the runner are
shown in Fig. 3.13.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 108 Chapter 3

Figure 3.12 Reaction turbine installation.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 109

Figure 3.13 (a) Francis turbine runner and (b) velocity triangles for inward flow reaction
turbine.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 110 Chapter 3

Let
C 1 ¼ Absolute velocity of water at inlet
D1 ¼ Outer diameter of the runner
N ¼ Revolution of the wheel per minute
U 1 ¼ Tangential velocity of wheel at inlet
V 1 ¼ Relative velocity at inlet
C r1 ¼ radial velocity at inlet
a1 ¼ Angle with absolute velocity to the direction of motion
b1 ¼ Angle with relative velocity to the direction of motion
H ¼ Total head of water under which turbine is working
C 2 ; D2 ; U 2 ; V 2 ; C r2 ¼ Corresponding values at outlet

Euler’s turbine equation Eq. (1.78) and E is maximum when Cw2 (whirl
velocity at outlet) is zero that is when the absolute and flow velocities are equal at
the outlet.

3.6 TURBINE LOSSES

Let
Ps ¼ Shaft power output
Pm ¼ Mechanical power loss
Pr ¼ Runner power loss
Pc ¼ Casing and draft tube loss
Pl ¼ Leakage loss
P ¼ Water power available
Ph ¼ Pr þ Pc þ Pl ¼ Hydraulic power loss

Runner power loss is due to friction, shock at impeller entry, and flow
separation. If hf is the head loss associated with a flow rate through the runner of
Qr, then
Ps ¼ rgQr hf ðNm/sÞ ð3:8Þ
Leakage power loss is due to leakage in flow rate, q, past the runner and therefore
not being handled by the runner. Thus
Q ¼ Qr þ q ð3:9Þ

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 111

If Hr is the head across the runner, the leakage power loss becomes
Pl ¼ rgH r q ðNm / sÞ ð3:10Þ
Casing power loss, Pc, is due to friction, eddy, and flow separation losses in the
casing and draft tube. If hc is the head loss in casing then
Pc ¼ rgQhc ðNm / sÞ ð3:11Þ
From total energy balance we have
rgQH ¼ Pm þ rg ðhf Qr þ hc Q þ H r q þ Ps Þ
Then overall efficiency, ho, is given by
Shaft power output
ho ¼
Fluid power available at inlet
or
Ps
ho ¼ ð3:12Þ
rgQH
Hydraulic efficiency, hh, is given by
Power available at runner
hh ¼
Fluid power available at inlet
or
ðPs þ Pm Þ
hh ¼ ð3:13Þ
rgQH
Eq. (3.13) is the theoretical energy transfer per unit weight of fluid.
Therefore the maximum efficiency is
hh ¼ U 1 Cw1 /gH ð3:14Þ

3.7 TURBINE CHARACTERISTICS

Part and overload characteristics of Francis turbines for specific speeds of 225
and 360 rpm are shown in Fig. 3.14
Figure 3.14 shows that machines of low specific speeds have a slightly
higher efficiency. It has been experienced that the Francis turbine has unstable
characteristics for gate openings between 30 to 60%, causing pulsations in output
and pressure surge in penstocks. Both these problems were solved by Paul Deriaz
by designing a runner similar to Francis runner but with adjustable blades.
The part-load performance of the various types are compared in Fig. 3.15
showing that the Kaplan and Pelton types are best adopted for a wide range of
load but are followed fairly closely by Francis turbines of low specific speed.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 112 Chapter 3

Figure 3.14 Variation of efficiency with load for Francis turbines.

Figure 3.15 Comparison of part-load efficiencies of various types of hydraulic turbine.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 113

3.8 AXIAL FLOW TURBINE

In an axial flow reaction turbine, also known as Kaplan turbine, the flow of water
is parallel to the shaft.
A Kaplan turbine is used where a large quantity of water is available at low
heads and hence the blades must be long and have large chords so that they are
strong enough to transmit the very high torque that arises. Fig. 3.16 and 3.17 shows
the outlines of the Kaplan turbine. The water from the scroll flows over the guide
blades and then over the vanes. The inlet guide vanes are fixed and are situated at a
plane higher than the runner blades such that fluid must turn through 908 to enter the
runner in the axial direction. The function of the guide vanes is to impart whirl to
the fluid so that the radial distribution of velocity is the same as in a free vortex.
Fig. 3.18 shows the velocity triangles and are usually drawn at the mean
radius, since conditions change from hub to tip. The flow velocity is axial at inlet
and outlet, hence Cr1 ¼ Cr2 ¼ Ca
C1 is the absolute velocity vector at anglea1 toU1, andV1 is the relative
velocity at an angle b1. For maximum efficiency, the whirl component Cw2 ¼ 0,
in which case the absolute velocity at exit is axial and then C2 ¼ Cr2
Using Euler’s equation

E ¼ UðC w1 2 C w2 Þ/g

and for zero whirl (Cw2 ¼ 0) at exit

E ¼ UC w1 /g

3.9 CAVITATION
In the design of hydraulic turbine, cavitation is an important factor. As the outlet
velocity V2 increases, then p2 decreases and has its lowest value when the vapor
pressure is reached.
At this pressure, cavitation begins. The Thoma parameter s ¼ NPSH H and
Fig. 3.19 give the permissible value of sc in terms of specific speed.
The turbines of high specific speed have a high critical value of s, and must
therefore be set lower than those of smaller specific speed (Ns).

Illustrative Example 3.9: Consider an inward flow reaction turbine in

which velocity of flow at inlet is 3.8 m/s. The 1 m diameter wheel rotates at
240 rpm and absolute velocity makes an angle of 168 with wheel tangent.
Determine (1) velocity of whirl at inlet, (2) absolute velocity of water at inlet, (3)
vane angle at inlet, and (4) relative velocity of water at entrance.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 114 Chapter 3

Figure 3.16 Kaplan turbine of water is available at low heads.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 115

Figure 3.17 Kaplan turbine runner.

Solution: From Fig. 3.13b

1. From inlet velocity triangle (subscript 1)
C r1 C r1 3:8
tana1 ¼ or C w1 ¼ ¼ ¼ 13:3 m/s
C w1 tana1 tan168
2. Absolute velocity of water at inlet, C1, is
C r1 C r1 3:8
sina1 ¼ or C 1 ¼ ¼ ¼ 13:79 m/s
C1 sina1 sin168
3.
ðpD1 ÞðNÞ ðpÞð1Þð240Þ
U1 ¼ ¼ ¼ 12:57 m/s
60 60

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 116 Chapter 3

Figure 3.18 Velocity triangles for an axial flow hydraulic turbine.

and
C r1 3:8 3:8
tan b1 ¼ ¼ ¼ ¼ 5:21
ðC w1 2 U 1 Þ ð13:3 2 12:57Þ 0:73

[ b1 ¼ 798 nearby
4. Relative velocity of water at entrance
C r1 C r1 3:8
sin b1 ¼ or V 1 ¼ ¼ ¼ 3:87m/s
V1 sin b1 sin 798

Illustrative Example 3.10: The runner of an axial flow turbine has mean
diameter of 1.5 m, and works under the head of 35 m. The guide blades make an
angle of 308 with direction of motion and outlet blade angle is 228. Assuming
axial discharge, calculate the speed and hydraulic efficiency of the turbine.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 117

Figure 3.19 Cavitation limits for reaction turbines.

Figure 3.20 Velocity triangles (a) inlet and (b) outlet.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 118 Chapter 3

Solution:
Since this is an impulse turbine, assume coefficient of velocity ¼ 0.98
Therefore the absolute velocity at inlet is
pffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C1 ¼ 0:98 2gH ¼ 0:98 ð2Þð9:81Þð35Þ ¼ 25:68 m/s
The velocity of whirl at inlet
Cw1 ¼ C1 cos a1 ¼ 25:68 cos 308 ¼ 22:24 m/s
Since U1 ¼ U2 ¼ U
Using outlet velocity triangle
C2 ¼ U 2 tan b2 ¼ U tan b2 ¼ U tan 228
Hydraulic efficiency of turbine (neglecting losses)
C w1 U 1 H 2 C 22 /2g
hh ¼ ¼
gH H

22:24U ðU tan 228Þ2

¼H2
g 2g
or
22:24U ðU tan 22Þ2
þ ¼H
g 2g
or
22:24U þ 0:082U 2 2 9:81H ¼ 0
or
0:082U 2 þ 22:24U 2 9:81H ¼ 0
or
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
222:24 ^ ð22:24Þ2 þ ð4Þð0:082Þð9:81Þð35Þ
U¼
ð2Þð0:082Þ
As U is positive,
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
U ¼ 222:24 þ 0:164
494:62 þ 112:62

¼ 222:24 þ 24:64 ¼ 14:63 m/s

0:164

Now using relation

pDN
U¼
60

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 119

or
60U ð60Þð14:63Þ
N¼ ¼ ¼ 186 rpm
pD ðpÞð1:5Þ
Hydraulic efficiency
C w1 U ð22:24Þð14:63Þ
hh ¼ ¼ ¼ 0:948 or 94:8%
gH ð9:81Þð35Þ

Illustrative Example 3.11: A Kaplan runner develops 9000 kW under a

head of 5.5 m. Assume a speed ratio of 2.08, flow ratio 0.68, and mechanical
efficiency 85%. The hub diameter is 1/3 the diameter of runner. Find the diameter
of the runner, and its speed and specific speed.
Solution:
pffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
U 1 ¼ 2:08 2gH ¼ 2:08 ð2Þð9:81Þð5:5Þ ¼ 21:61 m/s
pffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Cr1 ¼ 0:68 2gH ¼ 0:68 ð2Þð9:81Þð5:5Þ ¼ 7:06 m/s
Now power is given by
9000 ¼ ð9:81Þð5:5Þð0:85ÞQ
Therefore,
Q ¼ 196:24 m3 /s
If D is the runner diameter and, d, the hub diameter
p
Q ¼ ðD 2 2 d 2 ÞC r1
4
or
 
p 1 2
D 2 D 7:06 ¼ 196:24
2
4 9
Solving
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
D¼ ð196:24Þð4Þð9Þ ¼ 6:31 m
ðpÞð7:06Þð8Þ
pffiffiffi pffiffiffiffiffiffiffiffiffiffi
N s ¼ N 5/4P ¼ 65 9000 ¼ 732 rpm
H 5:55/4

Design Example 3.12: A propeller turbine develops 12,000 hp, and rotates
at 145 rpm under a head of 20 m. The outer and hub diameters are 4 m and 1.75 m,

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 120 Chapter 3

respectively. Calculate the inlet and outlet blade angles measured at mean radius
if overall and hydraulic efficiencies are 85% and 93%, respectively.
Solution:
4 þ 1:75
Mean diameter ¼ ¼ 2:875 m
2
pDN ðpÞð2:875Þð145Þ
U1 ¼ ¼ ¼ 21:84 m/s
60 60
Using hydraulic efficiency
C w1 U 1 ðC w1 Þð21:84Þ
hh ¼ ¼ ¼ 0:93Cw1
gH ð9:81Þð20Þ
or
Cw1 ¼ 8:35 m/s

Power ¼ ð12; 000Þð0:746Þ ¼ 8952 kW

Power ¼ rgQH ho
or
8952 ¼ 9:81 £ Q £ 20 £ 0:85
8952
Therefore, Q ¼ ð9:81Þð20Þð0:85Þ ¼ 53:68 m3 /s
Discharge, Q ¼ 53:68 ¼ p4 ð4 2 1:75 ÞC r1
2 2

[ Cr1 ¼ 5:28 m/s

C r1 5:28 5:28
tan b1 ¼ ¼ ¼ ¼ 0:3914
U 1 2 Cw1 21:84 2 8:35 13:49

b1 ¼ 21:388
and
C r2 5:28
tan b2 ¼ ¼ ¼ 0:2418
U 2 21:84

b2 ¼ 13:598

Illustrative Example 3.13: An inward flow reaction turbine wheel has

outer and inner diameter are 1.4 m and 0.7 m respectively. The wheel has radial
vanes and discharge is radial at outlet and the water enters the vanes at an angle of
128. Assuming velocity of flow to be constant, and equal to 2.8 m/s, find

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 121

Figure 3.21 Velocity triangles at inlet and outlet for Example 3.13.

1. The speed of the wheel, and

2. The vane angle at outlet.
Solution:
Outer diameter, D2 ¼ 1.4 m
Inner diameter, D1 ¼ 0.7 m
Angle at which the water enters the vanes, a1 ¼ 128
Velocity of flow at inlet,
C r1 ¼ C r2 ¼ 2:8 m/s

As the vanes are radial at inlet and outlet end, the velocity of whirl at inlet
and outlet will be zero, as shown in Fig. 3.21.
Tangential velocity of wheel at inlet,
C r1 2:8
U1 ¼ ¼ ¼ 13:15 m/s
tan 128 0:213
Also, U 1 ¼ pD602 N or
60U 1 ð60Þð13:15Þ
N¼ ¼ ¼ 179 rpm
pD2 ðpÞð1:4Þ

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 122 Chapter 3

Let b2 is the vane angle at outlet

pD1 N ðpÞð0:7Þð179Þ
U2 ¼ ¼ ¼ 6:56 m/s
60 60
From Outlet triangle,
C r2 2:8
tan b2 ¼ ¼ ¼ 0:4268 i:e: b2 ¼ 23:118
U2 6:56

Illustrative Example 3.14: Consider an inward flow reaction turbine in

which water is supplied at the rate of 500 L/s with a velocity of flow of 1.5 m/s.
The velocity periphery at inlet is 20 m/s and velocity of whirl at inlet is 15 m/s.
Assuming radial discharge, and velocity of flow to be constant, find
1. Vane angle at inlet, and
2. Head of water on the wheel.

Solution:
Discharge, Q ¼ 500 L/s ¼ 0.5 m3/s
Velocity of flow at inlet, Cr1 ¼ 1.5 m/s
Velocity of periphery at inlet, U1 ¼ 20 m/s
Velocity of whirl at inlet, Cw1 ¼ 15 m/s
As the velocity of flow is constant, Cr1 ¼ Cr2 ¼ 1.5 m/s
Let b1 ¼ vane angle at inlet
From inlet velocity triangle
C r1 1:5
tan ð180 2 b1 Þ ¼ ¼ ¼ 0:3
U 1 2 C w1 20 2 15

[ ð180 2 b1 Þ ¼ 168410
or
b1 ¼ 1808 2 168410 ¼ 1638190
Since the discharge is radial at outlet, ad so the velocity of whirl at outlet is
zero
Therefore,
Cw1 U 1 C2 C2
¼ H 2 1 ¼ H 2 r1
g 2g 2g

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 123

or
ð15Þð20Þ 1:52
¼H2
9:81 ð2Þð9:81Þ
[ H ¼ 30:58 2 0:1147 ¼ 30:47 m

Design Example 3.15: Inner and outer diameters of an outward flow reaction
turbine wheel are 1 m and 2 m respectively. The water enters the vane at angle of
208 and leaves the vane radially. Assuming the velocity of flow remains constant at
12 m/s and wheel rotates at 290 rpm, find the vane angles at inlet and outlet.
Solution:
Inner diameter of wheel, D1 ¼ 1 m
Outer diameter of wheel, D2 ¼ 2 m
a1 ¼ 208
Velocity of flow is constant
That is, Cr1 ¼ Cr2 ¼ 12 m/s
Speed of wheel, N ¼ 290 rpm
Vane angle at inlet ¼ b1
U1 is the velocity of periphery at inlet.

Therefore, U 1 ¼ pD1N ðpÞð1Þð290Þ ¼ 15:19 m/s

60 ¼ 60
From inlet triangle, velocity of whirl is given by
12 12
C w1 ¼ ¼ ¼ 32:97 m/s
tan 20 0:364

Hence, tan b1 ¼ C C2r1 12 12

U 1 ¼ 32:97 2 15:19 ¼ 17:78 ¼ 0:675
w1
i.e. b1 ¼ 348
Let b2 ¼ vane angle at outlet
U2 ¼ velocity of periphery at outlet
pD2 N ðpÞð2Þð290Þ
Therefore U 2 ¼ ¼ ¼ 30:38 m/s
60 60
From the outlet triangle
C r2 12
tan b2 ¼ ¼ ¼ 0:395
U2 30:38

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 124 Chapter 3

i.e.,
b2 ¼ 218330

Illustrative Example 3.16: An inward flow turbine is supplied with 245 L

of water per second and works under a total head of 30 m. The velocity of wheel
periphery at inlet is 16 m/s. The outlet pipe of the turbine is 28 cm in diameter.
The radial velocity is constant. Neglecting friction, calculate
1. The vane angle at inlet
2. The guide blade angle
3. Power.
Solution:
If D1 is the diameter of pipe, then discharge is
p
Q ¼ D21 C 2
4
or
ð4Þð0:245Þ
C2 ¼ ¼ 3:98 m/s
ðpÞð0:28Þ2
But C2 ¼ Cr1 ¼ Cr2
Neglecting losses, we have
Cw1 U 1 H 2 C22 /2g
¼
gH H
or
Cw1 U 1 ¼ gH 2 C22 /2
ð3:98Þ2
¼ ½ð9:81Þð30Þ 2 ¼ 294:3 2 7:92 ¼ 286:38
2

Power developed
P ¼ ð286:38Þð0:245Þ kW ¼ 70:16 kW

286:38
and Cw1 ¼ ¼ 17:9 m/s
16
3:98
tan a1 ¼ ¼ 0:222
17:9
i.e. a1 ¼ 128310
C r1 3:98 3:98
tan b1 ¼ ¼ ¼ ¼ 2:095
C w1 2 U 1 17:9 2 16 1:9
i.e. b1 ¼ 64.43 or b1 ¼ 648250

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 125

Design Example 3.17: A reaction turbine is to be selected from the

following data:
Discharge ¼ 7:8 m3 /s
Shaft power ¼ 12; 400 kW
Pressure head in scroll casing
at the entrance to turbine ¼ 164 m of water
Elevation of turbine casing above tail water level ¼ 5:4 m
Diameter of turbine casing ¼ 1 m
Velocity in tail race ¼ 1:6 m/s

Calculate the effective head on the turbine and the overall efficiency of the
unit.
Solution:
Velocity in casing at inlet to turbine
Discharge
Cc ¼ Cross 2 sectional area of casing

¼ 7:8 2 ¼ 9:93 m/s

ðp/4Þð1Þ

The net head on turbine

C2 2 C2
¼ Pressure head þ Head due to turbine position þ c 2g 1

¼ 164 þ 5:4 þ ð9:93Þ 2g

2 ð1:6Þ2
2

2 2:56 ¼ 174:3 m of water

¼ 164 þ 5:4 þ 98:619:62

Waterpower supplied to turbine ¼ QgH kW

¼ ð7:8Þð9:81Þð174:3Þ ¼ 13; 337 kW
Hence overall efficiency,
Shaft Power 12; 400
ho ¼ ¼ ¼ 0:93 or 93%
Water Power 13; 337
Design Example 3.18: A Francis turbine wheel rotates at 1250 rpm and net
head across the turbine is 125 m. The volume flow rate is 0.45 m3/s, radius of the
runner is 0.5 m. The height of the runner vanes at inlet is 0.035 m. and the angle of

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 126 Chapter 3

the inlet guide vanes is set at 708 from the radial direction. Assume that the
absolute flow velocity is radial at exit, find the torque and power exerted by the
water. Also calculate the hydraulic efficiency.
Solution:
For torque, using angular momentum equation

T ¼ mðCw2 r 2 2 Cw1 r 1 Þ

As the flow is radial at outlet, Cw2 ¼ 0 and therefore

T ¼ 2mC w1 r 1
¼ 2rQC w1 r 1
¼ 2ð103 Þð0:45Þð0:5Cw1 Þ
¼ 2225C w1 Nm

If h1 is the inlet runner height, then inlet area, A, is

A ¼ 2pr 1 h1
¼ ð2ÞðpÞð0:5Þð0:035Þ ¼ 0:11m2

0:45
Cr1 ¼ Q/A ¼ ¼ 4:1 m/s
0:11

From velocity triangle, velocity of whirl

Cw1 ¼ Cr1 tan708 ¼ ð4:1Þð2:75Þ ¼ 11:26m/s

Substituting Cw1, torque is given by

T ¼ 2ð225Þð11:26Þ ¼ 22534 Nm

Negative sign indicates that torque is exerted on the fluid. The torque
exerted by the fluid is þ 2534 Nm
Power exerted

P ¼ Tv
¼ ð2534Þð2ÞðpÞð1250Þ
ð60Þð1000Þ
¼ 331:83 kW

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 127

Hydraulic efficiency is given by

Power exerted
hh ¼ Power available
¼ ð331:83Þð10 Þ
3
rgQH
¼ 331:83 £ 103
ð103 Þð9:81Þð0:45Þð125Þ
¼ 0:6013 ¼ 60:13%

Design Example 3.19: An inward radial flow turbine develops 130 kW

under a head of 5 m. The flow velocity is 4 m/s and the runner tangential velocity at
inlet is 9.6 m/s. The runner rotates at 230 rpm while hydraulic losses accounting for
20% of the energy available. Calculate the inlet guide vane exit angle, the inlet
angle to the runner vane, the runner diameter at the inlet, and the height of the
runner at inlet. Assume radial discharge, and overall efficiency equal to 72%.
Solution:
Hydraulic efficiency is

Power deleloped
hh ¼ Power available

¼ mðC w1 U 1 2 C w2 UÞ
rgQH

Since flow is radial at outlet, then Cw2 ¼ 0 and m ¼ rQ, therefore

C w1 U 1
hh ¼
gH

ðCw1 Þð9:6Þ
0:80 ¼
ð9:81Þð5Þ

ð0:80Þð9:81Þð5Þ
Cw1 ¼ ¼ 4:09 m/s
9:6

Radial velocity Cr1 ¼ 4 m/s

tan a1 ¼ C r1 /C w1 ðfrom velocity triangleÞ

4 ¼ 0:978
¼ 4:09

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 128 Chapter 3

i.e., inlet guide vane angle a1 ¼ 448210

C r1
tan b1 ¼
ðC w1 2 U 1 Þ

4 4
¼ ¼ ¼ 20:726
ð4:09 2 9:6Þ 25:51

i.e., b1 ¼ 2 35.988 or 1808 2 35.98 ¼ 144.028

Runner speed is
pD1 N
U1 ¼
60
or
60U 1 ð60Þð9:6Þ
D1 ¼ ¼
pN ðpÞð230Þ
D1 ¼ 0:797 m
Overall efficiency
Power output
ho ¼
Power available
or
ð130Þð103 Þ
rgQH ¼
0:72
or
ð130Þð103 Þ
Q¼ ¼ 3:68 m3 /s
ð0:72Þð103 Þð9:81Þð5Þ
But
Q ¼ pD1 h1 C r1 ðwhere h1 is the height of runnerÞ
Therefore,
3:68
h1 ¼ ¼ 0:367 m
ðpÞð0:797Þð4Þ

Illustrative Example 3.20: The blade tip and hub diameters of an axial
hydraulic turbine are 4.50 m and 2 m respectively. The turbine has a net head of
22 m across it and develops 22 MW at a speed of 150 rpm. If the hydraulic
efficiency is 92% and the overall efficiency 84%, calculate the inlet and outlet
blade angles at the mean radius assuming axial flow at outlet.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 129

Solution:
Mean diameter, Dm, is given by
Dh þ Dt 2 þ 4:50
Dm ¼ ¼ ¼ 3:25 m
2 2
Overall efficiency, ho, is given by
Power develpoed
ho ¼
Power available
22
[ Power available ¼ ¼ 26:2 MW
0:84
Also, available power ¼ rgQH
ð26:2Þð106 Þ ¼ ð103 Þð9:81Þð22ÞQ
Hence flow rate, Q, is given by
ð26:2Þð106 Þ
Q¼ ¼ 121:4 m3 /s
ð103 Þð9:81Þð22Þ
Now rotor speed at mean diameter
pDm N ðpÞð3:25Þð150Þ
Um ¼ ¼ ¼ 25:54 m/s
60 60
Power given to runner ¼ Power available £ hh
¼ 26:2 £ 106 £ 0:92
¼ 24:104 MW

Theoretical power given to runner can be found by using

P ¼ rQU m Cw1 ðC w2 ¼ 0Þ
ð24:104Þð106 Þ ¼ ð103 Þð121:4Þð25:54ÞðC w1 Þ
ð24:104Þð106 Þ
[ C w1 ¼ ¼ 7:77 m/s
ð103 Þð121:4Þð25:54Þ
Axial velocity is given by
Q£4 ð121:4Þð4Þ
Cr ¼ ¼ ¼ 9:51 m/s
pðD2t 2 D2h Þ pð4:502 2 22 Þ

Using velocity triangle

Cr 9:51
tan ð180 2 b1 Þ ¼ ¼
U m 2 C w1 25:54 2 7:77

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 130 Chapter 3

Inlet angle,
b1 ¼ 151:858
At outlet
Cr
tan b2 ¼
V cw2
But Vcw2 equals to Um since Cw2 is zero. Hence
9:51
tan b2 ¼ ¼ 0:3724
25:54
that is,
b2 ¼ 20:438

Design Example 3.21: The following design data apply to an inward flow
radial turbine:
Overall efficiency 75%
Net head across the turbine 6m
Power output 128 kW
The runner tangential velocity 10:6 m/s
Flow velocity 4 m/s
Runner rotational speed 235 rpm
Hydraulic losses 18% of energy available
Calculate the inlet guide vane angle, the inlet angle of the runner vane, the
runner diameter at inlet, and height of the runner at inlet. Assume that the
discharge is radial.
Solution:
Hydraulic efficiency, hh, is given by
Power given to runner
hh ¼ Water Power available

¼ mðU 1 Crw1gQH
2 U 2 C w2 Þ

Since flow is radial at exit, Cw2 ¼ 0 and m ¼ rQ. Therefore

U 1 Cw1
hh ¼
gH
ð10:6ÞðCw1 Þ
0:82 ¼ or Cw1 ¼ 4:6 m/s
ð9:81Þð6Þ

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 131

Now
4
tan a1 ¼ Cr1 /C w1 ¼ ¼ 0:8695
4:6

that is; a1 ¼ 418

From Figs. 3.22 and 3.23

C r1 4
tan ð180 2 b1 Þ ¼ ¼ ¼ 0:667
U 1 2 Cw1 10:6 2 4:6

that is; b1 ¼ 33:698

Hence blade angle, b1, is given by

1808 2 33:698 ¼ 146:318

Runner speed at inlet

pD1 N
U1 ¼
60

Figure 3.22 Velocity triangles for Example 3.14.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 132 Chapter 3

Figure 3.23 Velocity triangles at inlet and outlet for Example 3.15.

Figure 3.24 Inlet velocity triangle.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 133

or
U 1 ð60Þ ð10:6Þð60Þ
D1 ¼ ¼ ¼ 0:86 m
pN ðpÞð235Þ
Overall efficiency
Power output
ho ¼
Power available

ð128Þð103 Þ
rgQH ¼
0:75
From which flow rate
ð128Þð103 Þ
Q¼ ¼ 2:9 m3 /s
ð0:75Þð103 Þð9:81Þð6Þ
Also,
Q ¼ pD1 hC r1
where h1 is the height of runner
Therefore,
2:9
h1 ¼ ¼ 0:268 m
ðpÞð0:86Þð4Þ

Design Example 3.22: A Kaplan turbine develops 10,000 kW under an

effective head 8 m. The overall efficiency is 0.86, the speed ratio 2.0, and flow
ratio 0.60. The hub diameter of the wheel is 0.35 times the outside diameter of
the wheel. Find the diameter and speed of the turbine.
Solution:
Head, H ¼ 8 m, Power, P ¼ 10,000 kW
Overall efficiency, ho ¼ 0.86
Speed ratio
U1 pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2¼ ; or U l ¼ 2 £ 9:81 £ 8 ¼ 25:06 m/s
ð2gHÞ 1=2

Flow ratio
C r1 pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
¼ 0:60 or C r1 ¼ 0:60 2 £ 9:81 £ 8 ¼ 7:52 m/s
ð2gHÞ 1=2

Hub diameter, D1 ¼ 0.35 D2

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 134 Chapter 3

Overall efficiency,
P
ho ¼
rgQH

Or
10000
0:86 ¼
1000 £ 9:81 £ Q £ 8

[ Q ¼ 148:16 m3 /s

Now using the relation

p 2
Q ¼ Cr1 £ D 2 D 22
4 1

Or
p 2

148:16 ¼ 7:52 £ D 1 2 0:35D 21
4
[ D1 ¼ 5:35 m

The peripheral velocity of the turbine at inlet

pD1 N p £ 5:35 £ N
25:06 ¼ ¼
60 60
60 £ 25:06
[ N¼ ¼ 89 rpm
p £ 5:35

Design Example 3.23: An inward flow reaction turbine, having an inner

and outer diameter of 0.45 m and 0.90 m, respectively. The vanes are radial at
inlet and the discharge is radial at outlet and the water enters the vanes at an angle
of 128. Assuming the velocity of flow as constant and equal to 2.8 m/s, find the
speed of the wheel and the vane angle at outlet.
Solution:
Inner Diameter, D2 ¼ 0.45 m
Outer Diameter, D1 ¼ 0.9 m

a2 ¼ 908ðradial dischargeÞ

a1 ¼ 128; Cr1 ¼ Cr2 ¼ 2:8 m/s

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 135

From velocity triangle at inlet (see Fig. 3.11), The peripheral velocity of the
wheel at inlet
C r1 2:8
U1 ¼ ¼ ¼ 13:173 m/s
tan a1 tan 128
Now,
pD1 N
U1 ¼
60
or
60U 1 60 £ 13:173
N¼ ¼ ¼ 279 rpm
pD1 p £ 0:9
Considering velocity triangle at outlet peripheral velocity at outlet
pD2 N p £ 0:45 £ 279
U2 ¼ ¼ ¼ 6:58 m/s
60 60
Cr2 2:8
tan b2 ¼ ¼ ¼ 0:426
U 2 6:58

[ b2 ¼ 23:058

Design Example 3.24: An inward flow reaction turbine develops 70 kW at

370 rpm. The inner and outer diameters of the wheel are 40 and 80 cm,
respectively. The velocity of the water at exit is 2.8 m/s. Assuming that the
discharge is radial and that the width of the wheel is constant, find the actual and
theoretical hydraulic efficiencies of the turbine and the inlet angles of the guide
and wheel vanes. Turbine discharges 545 L/s under a head of 14 m.
Solution:
Q ¼ 545 L/s ¼ 0.545 m3/s
D1 ¼ 80 cm, D2 ¼ 40 cm
H ¼ 14 m, a2 ¼ 908 (radial discharge)
b1 ¼ b2
Peripheral velocity of the wheel at inlet
pD1 N p £ 0:80 £ 370
U1 ¼ ¼ ¼ 15:5 m/s
60 60
Velocity of flow at the exit, Cr2 ¼ 2.8 m/s
As a2 ¼ 908, Cr2 ¼ C2
Work done/s by the turbine per kg of water ¼ Cwg£U1

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 136 Chapter 3

But this is equal to the head utilized by the turbine, i.e.

Cw1 U 1 C2
¼H2
g 2g
(Assuming there is no loss of pressure at outlet) or
Cw1 £ 15:5 ð2:8Þ2
¼ 14 2 ¼ 13:6 m
9:81 2 £ 9:81
or
13:6 £ 9:81
Cw1 ¼ ¼ 8:6 m/s
15:5
Work done per second by turbine
¼ rgQ Cw1 U 1

¼ 1000 £ 0:545 £ 8:6 £ 15:5

1000
¼ 72:65kW

rgQH
Available power or water power ¼ 1000
¼ 74:85

Actual available power ¼ 70 kW

70
Overall turbine efficiency isht ¼ £ 100
74:85
ht ¼ 93:52%

This is the actual hydraulic efficiency as required in the problem.

Hydraulic Efficiency is
72:65
hh ¼ £ 100 ¼ 97:06%
75:85
This is the theoretical efficiency
Q ¼ pD1 b1 Cr1 ¼ pD2 b2 Cr2
(Neglecting blade thickness)
D2 40
Cr1 ¼ Cr2 ¼ 2:8 £ ¼ 1:4 m/s
D1 20
Drawing inlet velocity triangle
C r1 1:4 1:4
tan b1 ¼ ¼ ¼ ¼ 0:203
U 1 2 C w1 15:5 2 8:6 6:9

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 137

i.e., b1 ¼ 11.478
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
0:5
C 1 ¼ Cw1 þ Cr1 ¼ 8:62 þ 1:42 ¼ 8:64 m/s
and
C w1 8:6
cosa1 ¼ ¼ ¼ 0:995
C1 8:64
i.e., a1 ¼ 5.58

Design Example 3.25: An inward flow Francis turbine, having an overall

efficiency
pffiffiffiffiffiffiffiffiffiof 86%, hydraulic efficiency of 90%, and radial velocity of flow at inlet
0.28 2gH . The turbine is required to develop 5000 kW when operating under a
net head of 30 m, specific speed is 270, assume guide vane angle 308, find
1. rpm of the wheel,
2. the diameter and the width of the runner at inlet, and
3. the theoretical inlet angle of the runner vanes.
Solution: pffiffiffiffiffiffiffiffiffi
Power, P ¼ 5000 kW; a1 ¼ 308; H ¼ 30 m; Cr1 ¼ 0:28 2gH , Ns ¼ 270,
hh ¼ 0.90, ho ¼ 0.86
1. Specific speed of the turbine is
pffiffiffi
N P
N s ¼ 5=4
H
or
Ns H5=4 270 £ ð30Þ1:25 18957
N¼ pffiffiffi ¼ pffiffiffiffiffiffiffiffiffiffi ¼ ¼ 267 rpm
P 5000 71
2. Velocity of Flow:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Cr1 ¼ 0:28 2 £ 9:81 £ 30 ¼ 6:79m/s

From inlet velocity triangle

C r1 ¼ C 1 sin a1
or
6:79 ¼ C 1 sin 308
or
6:79
C1 ¼ ¼ 13:58 m/s
0:5

Cw1 ¼ C1 cos308 ¼ 13.58 £ 0.866 ¼ 11.76 m/s

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 138 Chapter 3

Work done per (sec) (kg) of water

C w1 £ U 1
¼ ¼ hh £ H
g
¼ 0:9 £ 30
¼ 27 mkg/s

Peripheral Velocity,
27 £ 9:81
U1 ¼ ¼ 22:5 m/s
11:76

But U 1 ¼ pD1N
60
or
60U 1 60 £ 22:5
D1 ¼ ¼ ¼ 1:61 m
pN p £ 267

Power, P ¼ rgQHho
or
5000 ¼ 1000 £ 9.81 £ Q £ 30 £ 0.86
or
Q ¼ 19.8 m3/s
Also Q ¼ kpD1b1Cr1 (where k is the blade thickness coefficient and b1
is the breath of the wheel at inlet) or
Q 19:8
b1 ¼ ¼ ¼ 0:61 m
kpD1 C r1 0:95 £ p £ 1:61 £ 6:79

3. From inlet velocity triangle

C r1 6:79 6:79
tan b1 ¼ ¼ ¼ ¼ 0:632
U 1 2 C w1 22:5 2 11:76 10:74
i.e. b1 ¼ 32.308

Design Example 3.26: A 35 MW generator is to operate by a double

overhung Pelton wheel. The effective head is 350 m at the base of the nozzle.
Find the size of jet, mean diameter of the runner and specific speed of wheel.
Assume Pelton wheel efficiency 84%, velocity coefficient of nozzle 0.96, jet ratio
12, and speed ratio 0.45.
Solution:
In this case, the generator is fed by two Pelton turbines.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 139

Power developed by each turbine,

35; 000
PT ¼ ¼ 17; 500 kW
2
Using Pelton wheel efficiency in order to find available power of each
turbine
17; 500
P¼ ¼ 20; 833 kW
0:84
But, P ¼ rgQH
P 20833
Q¼ ¼ ¼ 6:07 m3 /s
rgH 1000 £ 9:81 £ 350
pffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Velocity of jet,C j ¼ C v 2gH ¼ 0:96 2 £ 9:81 £ 350
Cj ¼ 79:6 m/s

Area of jet, A ¼ CQj ¼ 6:07

79:6 ¼ 0:0763 m
2

 0:5  0:5
[ Diameter of jet,d ¼ 4A ¼ 4 £ 0:0763 ¼ 0:312 m
p p
d ¼ 31:2 cm

Diameter of wheel D ¼ d £ jet ratio ¼ 0.312 £ 12 ¼ 3.744 m

Peripheral velocity of the wheel
pffiffiffiffiffiffiffiffiffi
U ¼ speed ratio 2gH
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
¼ 0:45 £ 2 £ 9:81 £ 350 ¼ 37:29 m/s

But U ¼ p60
DN or
60U 60 £ 37:29
N¼ ¼ ¼ 190 rpm
pD p £ 3:744
Specific speed,
pffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
N PT 190 17; 500
N s ¼ 5=4 ¼ ¼ 16:6
H ð350Þ1:25

PROBLEMS
3.1 A Pelton wheel produces 4600 hP under a head of 95 m, and with an overall
efficiency of 84%. Find the diameter of the nozzle if the coefficient of
velocity for the nozzle is 0.98.
(0.36 m)

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 140 Chapter 3

3.2 Pelton wheel develops 13,500 kW under a head of 500 m. The wheel rotates
at 430 rpm. Find the size of the jet and the specific speed. Assume 85%
efficiency.
(0.21 m, 21)
3.3 A Pelton wheel develops 2800 bhP under a head of 300 m at 84% efficiency.
The ratio of peripheral velocity of wheel to jet velocity is 0.45 and specific
speed is 17. Assume any necessary data and find the jet diameter.
(140 mm)
3.4 A Pelton wheel of power station develops 30,500 hP under a head of 1750 m
while running at 760 rpm. Calculate (1) the mean diameter of the runner,
(2) the jet diameter, and (3) the diameter ratio.
(2.14 m, 0.104 m, 20.6)
3.5 Show that in an inward flow reaction turbine, when the velocity of flow is
constant and wheel vane angle at entrance is 908, the best peripheral
velocity is
pffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2gH / 2 þ tan2 a

where H is the head and a the angle of guide vane.

3.6 A Pelton wheel develops 740 kW under a head of 310 m. Find the jet
diameter if its efficiency is 86% and
C v ¼ 0:98:
(0.069 m)
3.7 A reaction turbine runner diameter is 3.5 m at inlet and 2.5 m at outlet.
The turbine discharge 102 m3 per second of water under a head of
145 m. Its inlet vane angle is 1208. Assume radial discharge at 14 m/s,
breadth of wheel constant and hydraulic efficiency of 88%, calculate
the power developed and speed of machine.
(128 MW, 356 rpm)
3.8 Show that in a Pelton wheel, where the buckets deflect the water through an
angle of (1808 2 a) degrees, the hydraulic efficiency of the wheel is given by
2UðC 2 UÞð1 þ cos aÞ
hh ¼
C2
where C is the velocity of jet and U is mean blade velocity.
3.9 A Kaplan turbine produces 16000 kW under a head of 20 m, while running
at 166 rpm. The diameter of the runner is 4.2 m while the hub diameter is
2 m, the discharge being 120 m3/s. Calculate (1) the turbine efficiency,

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Hydraulic Turbines 141

(2) specific speed, (3) the speed ratio based on the tip diameter of the blade,
and (4) the flow ratio.
(78%, 497, 1.84, 0.48)
3.10 Evolve a formula for the specific speed of a Pelton wheel in the following
form
pffiffiffiffiffiffiffi d
N s ¼ k h
D
where Ns ¼ specific speed, h ¼ overall efficiency, d ¼ diameter of jet,
D ¼ diameter of bucket circle, and k ¼ a constant.

NOTATION
C jet velocity, absolute
Cv nozzle velocity coefficient
Cw velocity of whirl
D wheel diameter
d diameter of nozzle
E energy transfer by bucket
Hr head across the runner
hf frictional head loss
Ns specific speed
P water power available
Pc casing and draft tube losses
Ph hydraulic power loss
Pl leakage loss
Pm mechanical power loss
Pr runner power loss
Ps shaft power output
U bucket speed
W work done
a angle of the blade tip at outlet
b angle with relative velocity
hi nozzle efficiency
htrans transmission efficiency
k relative velocity ratio

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 4
Centrifugal Compressors
and Fans

4.1 INTRODUCTION
This chapter will be concerned with power absorbing turbomachines, used to
handle compressible fluids. There are three types of turbomachines: fans, blowers,
and compressors. A fan causes only a small rise in stagnation pressure of the
flowing fluid. A fan consists of a rotating wheel (called the impeller), which is
surrounded by a stationary member known as the housing. Energy is transmitted to
the air by the power-driven wheel and a pressure difference is created, providing
airflow. The air feed into a fan is called induced draft, while the air exhausted from
a fan is called forced draft. In blowers, air is compressed in a series of successive
stages and is often led through a diffuser located near the exit. The overall pressure
rise may range from 1.5 to 2.5 atm with shaft speeds up to 30,000 rpm or more.

4.2 CENTRIFUGAL COMPRESSOR

The compressor, which can be axial flow, centrifugal flow, or a combination of
the two, produces the highly compressed air needed for efficient combustion.
In turbocompressors or dynamic compressors, high pressure is achieved by
imparting kinetic energy to the air in the impeller, and then this kinetic energy

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 144 Chapter 4

converts into pressure in the diffuser. Velocities of airflow are quite high and
the Mach number of the flow may approach unity at many points in the air stream.
Compressibility effects may have to be taken into account at every stage of the
compressor. Pressure ratios of 4:1 are typical in a single stage, and ratios of 6:1
are possible if materials such as titanium are used. There is renewed interest in the
centrifugal stage, used in conjunction with one or more axial stages, for small
turbofan and turboprop aircraft engines. The centrifugal compressor is not
suitable when the pressure ratio requires the use of more than one stage in series
because of aerodynamic problems. Nevertheless, two-stage centrifugal
compressors have been used successfully in turbofan engines.
Figure 4.1 shows part of a centrifugal compressor. It consists of a stationary
casing containing an impeller, which rotates and imparts kinetic energy to the air
and a number of diverging passages in which the air decelerates. The deceleration
converts kinetic energy into static pressure. This process is known as diffusion,
and the part of the centrifugal compressor containing the diverging passages is
known as the diffuser. Centrifugal compressors can be built with a double entry
or a single entry impeller. Figure 4.2 shows a double entry centrifugal
compressor.
Air enters the impeller eye and is whirled around at high speed by the vanes
on the impeller disc. After leaving the impeller, the air passes through a diffuser
in which kinetic energy is exchanged with pressure. Energy is imparted to the air
by the rotating blades, thereby increasing the static pressure as it moves from eye
radius r1 to tip radius r2. The remainder of the static pressure rise is achieved in
the diffuser. The normal practice is to design the compressor so that about half the
pressure rise occurs in the impeller and half in the diffuser. The air leaving the
diffuser is collected and delivered to the outlet.

Figure 4.1 Typical centrifugal compressor.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Centrifugal Compressors and Fans 145

Figure 4.2 Double-entry main stage compressor with side-entry compressor for cooling
air. (Courtesy of Rolls-Royce, Ltd.)

4.3 THE EFFECT OF BLADE SHAPE

ON PERFORMANCE
As discussed in Chapter 2, there are three types of vanes used in impellers. They
are: forward-curved, backward-curved, and radial vanes, as shown in Fig. 4.3.
The impellers tend to undergo high stress forces. Curved blades, such as
those used in some fans and hydraulic pumps, tend to straighten out due to
centrifugal force and bending stresses are set up in the vanes. The straight radial

Figure 4.3 Shapes of centrifugal impellar blades: (a) backward-curved blades, (b) radial
blades, and (c) forward-curved blades.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 146 Chapter 4

Figure 4.4 Pressure ratio or head versus mass flow or volume flow, for the three blade
shapes.

blades are not only free from bending stresses, they may also be somewhat easier
to manufacture than curved blades.
Figure 4.3 shows the three types of impeller vanes schematically, along
with the velocity triangles in the radial plane for the outlet of each type of vane.
Figure 4.4 represents the relative performance of these types of blades. It is clear
that increased mass flow decreases the pressure on the backward blade, exerts the
same pressure on the radial blade, and increases the pressure on the forward
blade. For a given tip speed, the forward-curved blade impeller transfers
maximum energy, the radial blade less, and the least energy is transferred by the
backward-curved blades. Hence with forward-blade impellers, a given pressure
ratio can be achieved from a smaller-sized machine than those with radial or
backward-curved blades.

4.4 VELOCITY DIAGRAMS

Figure 4.5 shows the impeller and velocity diagrams at the inlet and outlet.
Figure 4.5a represents the velocity triangle when the air enters the impeller in the
axial direction. In this case, absolute velocity at the inlet, C1 ¼ Ca1. Figure 4.5b
represents the velocity triangle at the inlet to the impeller eye and air enters
through the inlet guide vanes. Angle u is made by C1 and Ca1 and this angle is
known as the angle of prewhirl. The absolute velocity C1 has a whirl component
Cw1. In the ideal case, air comes out from the impeller tip after making an angle of
908 (i.e., in the radial direction), so Cw2 ¼ U2. That is, the whirl component is
exactly equal to the impeller tip velocity. Figure 4.5c shows the ideal velocity

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Centrifugal Compressors and Fans 147

Figure 4.5 Centrifugal impellar and velocity diagrams.

triangle. But there is some slip between the impeller and the fluid, and actual
values of Cw1 are somewhat less than U2. As we have already noted in the
centrifugal pump, this results in a higher static pressure on the leading face
of a vane than on the trailing face. Hence, the air is prevented from acquiring

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 148 Chapter 4

a whirl velocity equal to the impeller tip speed. Figure 4.5d represents the actual
velocity triangle.

4.5 SLIP FACTOR

From the above discussion, it may be seen that there is no assurance that the
actual fluid will follow the blade shape and leave the compressor in a radial
direction. Thus, it is convenient to define a slip factor s as:
C w2
s¼ ð4:1Þ
U2

Figure 4.6 shows the phenomenon of fluid slip with respect to a radial
blade. In this case, Cw2 is not equal to U2; consequently, by the above
definition, the slip factor is less than unity. If radial exit velocities are to be
achieved by the actual fluid, the exit blade angle must be curved forward
about 10– 14 degrees. The slip factor is nearly constant for any machine and
is related to the number of vanes on the impeller. Various theoretical and
empirical studies of the flow in an impeller channel have led to formulas for

Figure 4.6 Centrifugal compressor impeller with radial vanes.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Centrifugal Compressors and Fans 149

slip factors: For radial vaned impellers, the formula for s is given by Stanitz
as follows:
0:63p
s¼12 ð4:2Þ
n
where n is the number of vanes. The velocity diagram indicates that Cw2
approaches U2 as the slip factor is increased. Increasing the number of vanes
may increase the slip factor but this will decrease the flow area at the inlet.
A slip factor of about 0.9 is typical for a compressor with 19 – 21 vanes.

4.6 WORK DONE

The theoretical torque will be equal to the rate of change of angular momentum
experienced by the air. Considering a unit mass of air, this torque is given by
theoretical torque,
t ¼ Cw2 r2 ð4:3Þ
where, Cw2 is whirl component of C2 and r2 is impeller tip radius.
Let v ¼ angular velocity. Then the theoretical work done on the air may be
written as:
Theoretical work done Wc ¼ Cw2r2v ¼ Cw2U2.
2
Using the slip factor, we have theoretical Wc ¼ sU2 (treating work done
on the air as positive)
In a real fluid, some of the power supplied by the impeller is used in
overcoming losses that have a braking effect on the air carried round by the
vanes. These include windage, disk friction, and casing friction. To take account
of these losses, a power input factor can be introduced. This factor typically takes
values between 1.035 and 1.04. Thus the actual work done on the air becomes:
2
Wc ¼ csU 2 ð4:4Þ

(assuming Cw1 ¼ 0, although this is not always the case.)

Temperature equivalent of work done on the air is given by:
csU 22
T 02 2 T 01 ¼
Cp
where T01 is stagnation temperature at the impeller entrance; T02 is stagnation
temperature at the impeller exit; and Cp is mean specific heat over this
temperature range. As no work is done on the air in the diffuser, T03 ¼ T02, where
T03 is the stagnation temperature at the diffuser outlet.
The compressor isentropic efficiency (hc) may be defined as:
T 030 2 T 01
hc ¼
T 03 2 T 01

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 150 Chapter 4

(where T03 ¼ isentropic stagnation temperature at the diffuser outlet) or

0


T 01 T 030 =T 01 2 1
hc ¼
T 03 2 T 01
Let P01 be stagnation pressure at the compressor inlet and; P03 is stagnation
pressure at the diffuser exit. Then, using the isentropic P –T relationship, we get:
 
P03 T 030 g /ðg21Þ hc ðT 03 2 T 01 Þ g /ðg21Þ
¼ ¼ 1þ
P01 T 01 T 01
g /ðg21Þ
hc csU 2 2
¼ 1þ ð4:5Þ
C p T 01
Equation (4.5) indicates that the pressure ratio also depends on the inlet
temperature T01 and impeller tip speed U2. Any lowering of the inlet temperature
T01 will clearly increase the pressure ratio of the compressor for a given work
input, but it is not under the control of the designer. The centrifugal stresses in a
rotating disc are proportional to the square of the rim. For single sided impellers
of light alloy, U2 is limited to about 460 m/s by the maximum allowable
centrifugal stresses in the impeller. Such speeds produce pressure ratios of about
4:1. To avoid disc loading, lower speeds must be used for double-sided impellers.

4.7 DIFFUSER
The designing of an efficient combustion system is easier if the velocity of the air
entering the combustion chamber is as low as possible. Typical diffuser outlet
velocities are in the region of 90 m/s. The natural tendency of the air in a diffusion
process is to break away from the walls of the diverging passage, reverse its direction
and flow back in the direction of the pressure gradient, as shown in Fig. 4.7. Eddy
formation during air deceleration causes loss by reducing the maximum pressure
rise. Therefore, the maximum permissible included angle of the vane diffuser
passage is about 118. Any increase in this angle leads to a loss of efficiency due to

Figure 4.7 Diffusing flow.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Centrifugal Compressors and Fans 151

boundary layer separation on the passage walls. It should also be noted that any
change from the design mass flow and pressure ratio would also result in a loss of
efficiency. The use of variable-angle diffuser vanes can control the efficiency loss.
The flow theory of diffusion, covered in Chapter 2, is applicable here.

4.8 COMPRESSIBILITY EFFECTS

If the relative velocity of a compressible fluid reaches the speed of sound in the fluid,
separation of flow causes excessive pressure losses. As mentioned earlier, diffusion
is a very difficult process and there is always a tendency for the flow to break away
from the surface, leading to eddy formation and reduced pressure rise. It is necessary
to control the Mach number at certain points in the flow to mitigate this problem.
The value of the Mach number cannot exceed the value at which shock waves
occur. The relative Mach number at the impeller inlet must be less than unity.
As shown in Fig. 4.8a, the air breakaway from the convex face of the
curved part of the impeller, and hence the Mach number at this point, will be very
important and a shock wave might occur. Now, consider the inlet velocity
triangle again (Fig. 4.5b). The relative Mach number at the inlet will be given by:
V1
M1 ¼ pffiffiffiffiffiffiffiffiffiffiffi ð4:6Þ
gRT 1
where T1 is the static temperature at the inlet.
It is possible to reduce the Mach number by introducing the prewhirl. The
prewhirl is given by a set of fixed intake guide vanes preceding the impeller.
As shown in Fig. 4.8b, relative velocity is reduced as indicated by the
dotted triangle. One obvious disadvantage of prewhirl is that the work capacity of

Figure 4.8 a) Breakaway commencing at the aft edge of the shock wave, and
b) Compressibility effects.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 152 Chapter 4

the compressor is reduced by an amount U1Cw1. It is not necessary to introduce

prewhirl down to the hub because the fluid velocity is low in this region due to
lower blade speed. The prewhirl is therefore gradually reduced to zero by
twisting the inlet guide vanes.

4.9 MACH NUMBER IN THE DIFFUSER

The absolute velocity of the fluid becomes a maximum at the tip of the impeller
and so the Mach number may well be in excess of unity. Assuming a perfect gas,
the Mach number at the impeller exit M2 can be written as:
C2
M2 ¼ pffiffiffiffiffiffiffiffiffiffiffi ð4:7Þ
gRT 2
However, it has been found that as long as the radial velocity component (Cr2) is
subsonic, Mach number greater than unity can be used at the impeller tip without
loss of efficiency. In addition, supersonic diffusion can occur without the
formation of shock waves provided constant angular momentum is maintained
with vortex motion in the vaneless space. High Mach numbers at the inlet to the
diffuser vanes will also cause high pressure at the stagnation points on the diffuser
vane tips, which leads to a variation of static pressure around the circumference
of the diffuser. This pressure variation is transmitted upstream in a radial
direction through the vaneless space and causes cyclic loading of the impeller.
This may lead to early fatigue failure when the exciting frequency is of the same
order as one of the natural frequencies of the impeller vanes. To overcome this
concern, it is a common a practice to use prime numbers for the impeller vanes
and an even number for the diffuser vanes.

Figure 4.9 The theoretical centrifugal compressor characteristic.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Centrifugal Compressors and Fans 153

4.10 CENTRIFUGAL COMPRESSOR

CHARACTERISTICS
The performance of compressible flow machines is usually described in terms
of the groups of variables derived in dimensional analysis (Chapter 1). These
characteristics are dependent on other variables such as the conditions of
pressure and temperature at the compressor inlet and physical properties of
pffiffiffiffiffi
the working fluid. To study the performance of a compressor completely, it is
necessary to plot P03/P01 against the mass flow parameter m PT0101 for fixed
speed intervals of pNffiffiffiffiffi
T 01
. Figure 4.9 shows an idealized fixed speed
characteristic. Consider a valve placed in the delivery line of a compressor
running at constant speed. First, suppose that the valve is fully closed. Then
the pressure ratio will have some value as indicated by Point A. This
pressure ratio is available from vanes moving the air about in the impeller.
Now, suppose that the valve is opened and airflow begins. The diffuser
contributes to the pressure rise, the pressure ratio increases, and at Point B,
the maximum pressure occurs. But the compressor efficiency at this pressure
ratio will be below the maximum efficiency. Point C indicates the further
increase in mass flow, but the pressure has dropped slightly from the
maximum possible value. This is the design mass flow rate pressure ratio.
Further increases in mass flow will increase the slope of the curve until point
D. Point D indicates that the pressure rise is zero. However, the above-
described curve is not possible to obtain.

4.11 STALL
Stalling of a stage will be defined as the aerodynamic stall, or the breakaway of
the flow from the suction side of the blade airfoil. A multistage compressor may
operate stably in the unsurged region with one or more of the stages stalled, and
the rest of the stages unstalled. Stall, in general, is characterized by reverse flow
near the blade tip, which disrupts the velocity distribution and hence adversely
affects the performance of the succeeding stages.
Referring to the cascade of Fig. 4.10, it is supposed that some
nonuniformity in the approaching flow or in a blade profile causes blade B to
stall. The air now flows onto blade A at an increased angle of incidence due
to blockage of channel AB. The blade A then stalls, but the flow on blade C
is now at a lower incidence, and blade C may unstall. Therefore the stall
may pass along the cascade in the direction of lift on the blades. Rotating
stall may lead to vibrations resulting in fatigue failure in other parts of the
gas turbine.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 154 Chapter 4

Figure 4.10 Mechanism of stall propagation.

4.12 SURGING
Surging is marked by a complete breakdown of the continuous steady flow
throughout the whole compressor, resulting in large fluctuations of flow with time
and also in subsequent mechanical damage to the compressor. The phenomenon
of surging should not be confused with the stalling of a compressor stage.
Figure 4.11 shows typical overall pressure ratios and efficiencies hc of a
centrifugal compressor stage. The pressure ratio for a given speed, unlike the
temperature ratio, is strongly dependent on mass flow rate, since the machine is
usually at its peak value for a narrow range of mass flows. When the compressor
is running at a particular speed and the discharge is gradually reduced, the
pressure ratio will first increase, peaks at a maximum value, and then decreased.
The pressure ratio is maximized when the isentropic efficiency has the
maximum value. When the discharge is further reduced, the pressure ratio drops
due to fall in the isentropic efficiency. If the downstream pressure does not drop
quickly there will be backflow accompanied by further decrease in mass flow. In
the mean time, if the downstream pressure drops below the compressor outlet
pressure, there will be increase in mass flow. This phenomenon of sudden drop
in delivery pressure accompanied by pulsating flow is called surging. The point
on the curve where surging starts is called the surge point. When the discharge
pipe of the compressor is completely choked (mass flow is zero) the pressure
ratio will have some value due to the centrifugal head produced by the impeller.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Centrifugal Compressors and Fans 155

Figure 4.11 Centrifugal compressor characteristics.

Between the zero mass flow and the surge point mass flow, the operation of the
compressor will be unstable. The line joining the surge points at different speeds
gives the surge line.

4.13 CHOKING
When the velocity of fluid in a passage reaches the speed of sound at any cross-
section, the flow becomes choked (air ceases to flow). In the case of inlet flow
passages, mass flow is constant. The choking behavior of rotating passages

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 156 Chapter 4

differs from that of the stationary passages, and therefore it is necessary to make
separate analysis for impeller and diffuser, assuming one dimensional, adiabatic
flow, and that the fluid is a perfect gas.

4.13.1 Inlet
When the flow is choked, C 2 ¼ a 2 ¼ gRT. Since h0 ¼ h þ 12 C 2 , then
C p T 0 ¼ Cp T þ 12 g RT, and
 
T gR 21 2
¼ 1þ ¼ ð4:8Þ
T0 2C p gþ1
Assuming isentropic flow, we have:
     ð12gÞ=ðg21Þ
r P T0 1

¼ ¼ 1 þ g 2 1 M2 ð4:9Þ
r0 P0 T 2
and when C ¼ a, M ¼ 1, so that:
  " #1=ðg21Þ
r 2
¼
 ð4:10Þ
r0 gþ1

_ 1=2
A ¼ rC ¼ r gRT
Using the continuity equation, m , we have

 _
m 2 ðgþ1Þ=2ðg21Þ
¼ r0 a0 ð4:11Þ
A gþ1

where (r0 and a0 refer to inlet stagnation conditions, which remain unchanged.
The mass flow rate at choking is constant.

4.13.2 Impeller
When choking occurs in the impeller passages, the relative velocity equals the
speed of sound at any section. The relative velocity is given by:
 
gRT U2
V 2 ¼ a 2 ¼ gRT and T 01 ¼ T þ 2
2C p 2C p
Therefore,
    
T 2 U2
¼ 1þ ð4:12Þ
T 01 gþ1 2C p T 01

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Centrifugal Compressors and Fans 157

Using isentropic conditions,

 
r T 1=ðg21Þ
¼ and; from the continuity equation:
r01 T 01
 _
m T ðgþ1Þ=2ðg21Þ
¼ r0 a01
A T 01
   ðgþ1Þ=2ðg21Þ
2 U2
¼ r01 a01 1þ
gþ1 2C p T 01

 ðgþ1Þ=2ðg21Þ ð4:13Þ
2 þ g 2 1 U 2 =a201
¼ r01 a01
gþ1
Equation (4.13) indicates that for rotating passages, mass flow is dependent on
the blade speed.

4.13.3 Diffuser
For choking in the diffuser, we use the stagnation conditions for the diffuser and
not the inlet. Thus:
 _  ðgþ1Þ=2ðg21Þ
m 2
¼ r02 a02 ð4:14Þ
A gþ1

It is clear that stagnation conditions at the diffuser inlet are dependent on the
impeller process.
Illustrative Example 4.1: Air leaving the impeller with radial velocity
110 m/s makes an angle of 258300 with the axial direction. The impeller tip speed
is 475 m/s. The compressor efficiency is 0.80 and the mechanical efficiency
is 0.96. Find the slip factor, overall pressure ratio, and power required to drive the
compressor. Neglect power input factor and assume g ¼ 1.4, T01 ¼ 298 K, and
the mass flow rate is 3 kg/s.
Solution:
From the velocity triangle (Fig. 4.12),

 U 2 2 C w2
tan b2 ¼
Cr2
475 2 C w2
tanð25:58Þ ¼
110
Therefore, C w2 ¼ 422:54 m/s.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 158 Chapter 4

Figure 4.12 Velocity triangle at the impeller tip.

Now, s ¼ CUw2 ¼ 422:54 ¼ 0:89

2 475
The overall pressure ratio of the compressor:

g=ðg21Þ 3:5
P03 hc scU 22 ð0:80Þð0:89Þð4752 Þ
¼ 1þ ¼ 1þ ¼ 4:5
P01 C p T 01 ð1005Þð298Þ

The theoretical power required to drive the compressor:

mscU 22 ð3Þð0:89Þð4752 Þ
P¼ kW ¼ ¼ 602:42kW
1000 1000

Using mechanical efficiency, the actual power required to drive the

compressor is: P ¼ 602.42/0.96 ¼ 627.52 kW.

Illustrative Example 4.2: The impeller tip speed of a centrifugal

compressor is 370 m/s, slip factor is 0.90, and the radial velocity component at the
exit is 35 m/s. If the flow area at the exit is 0.18 m2 and compressor efficiency is
0.88, determine the mass flow rate of air and the absolute Mach number at the
impeller tip. Assume air density ¼ 1.57 kg/m3 and inlet stagnation temperature
is 290 K. Neglect the work input factor. Also, find the overall pressure ratio of the
compressor.
Solution:
C w2
Slip factor: s¼
U2
Therefore: Cw2 ¼ U2s ¼ (0.90)(370) ¼ 333 m/s
The absolute velocity at the impeller exit:
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C 2 ¼ C2r2 þ C 2v2 ¼ 3332 þ 352 ¼ 334:8 m/s
The mass flow rate of air: m_ ¼ r2 A2 Cr2 ¼ 1:57* 0:18* 35 ¼ 9:89 kg/s

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Centrifugal Compressors and Fans 159

The temperature equivalent of work done (neglecting c):

sU 22
T 02 2 T 01 ¼
Cp
sU 2
Therefore, T 02 ¼ T 01 þ C 2 ¼ 290 þ ð0:90Þð370 Þ ¼ 412:6 K
2

p 1005
The static temperature at the impeller exit,

C22 334:82
T 2 ¼ T 02 2 ¼ 412:6 2 ¼ 356:83 K
2Cp ð2Þð1005Þ

The Mach number at the impeller tip:

C2 334:8
M2 ¼ pffiffiffiffiffiffiffiffiffiffiffi ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ 0:884
gRT 2 ð1:4Þð287Þð356:83Þ
The overall pressure ratio of the compressor (neglecting c):
3:5 3:5
P03 hc scU 22 ð0:88Þð0:9Þð3702 Þ
¼ 1þ ¼ 1þ ¼ 3:0
P01 C p T 01 ð1005Þð290Þ

Illustrative Example 4.3: A centrifugal compressor is running at

16,000 rpm. The stagnation pressure ratio between the impeller inlet and outlet
is 4.2. Air enters the compressor at stagnation temperature of 208C and 1 bar.
If the impeller has radial blades at the exit such that the radial velocity at the exit
is 136 m/s and the isentropic efficiency of the compressor is 0.82. Draw the
velocity triangle at the exit (Fig. 4.13) of the impeller and calculate slip. Assume
axial entrance and rotor diameter at the outlet is 58 cm.

Figure 4.13 Velocity triangle at exit.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 160 Chapter 4

Solution:
Impeller tip speed is given by:
pDN ðpÞð0:58Þð16000Þ
U2 ¼ ¼ ¼ 486 m/s
60 60
Assuming isentropic flow between impeller inlet and outlet, then
T 020 ¼ T 01 ð4:2Þ0:286 ¼ 441:69 K
Using compressor efficiency, the actual temperature rise


T 020 2 T 01 ð441:69 2 293Þ
T 02 2 T 01 ¼ ¼ ¼ 181:33K
hc 0:82
Since the flow at the inlet is axial, Cw1 ¼ 0

W ¼ U 2 C w2 ¼ C p ðT 02 2 T 01 Þ ¼ 1005ð181:33Þ
ð181:33Þ ¼ 375 m/s
Therefore: C w2 ¼ 1005486
Slip ¼ 486 – 375 ¼ 111 m/s
C w2 375
Slip factor: s ¼ ¼ ¼ 0:772
U2 486
Illustrative Example 4.4: Determine the adiabatic efficiency, temperature
of the air at the exit, and the power input of a centrifugal compressor from the
following given data:

Impeller tip diameter ¼ 1 m

Speed ¼ 5945 rpm
Mass flow rate of air ¼ 28 kg/s
Static pressure ratio p3 /p1 ¼ 2:2
Atmospheric pressure ¼ 1 bar
Atmospheric temperature ¼ 258C
Slip factor ¼ 0:90
Neglect the power input factor.
Solution:
The impeller tip speed is given by:
DN ¼ ðpÞð1Þð5945Þ ¼ 311 m/s
U 2 ¼ p60 60
The work input: W ¼ sU 2 ¼ ð0:9Þð311 Þ ¼ 87 kJ/kg
2
2
1000

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Centrifugal Compressors and Fans 161

Using the isentropic P – T relation and denoting isentropic temperature by

T3 , we get:
0

 0:286
P3
T 30 ¼ T 1 ¼ ð298Þð2:2Þ0:286 ¼ 373:38 K
P1
Hence the isentropic temperature rise:

T30 2 T 1 ¼ 373:38 2 298 ¼ 75:38 K

The temperature equivalent of work done:

 
W
T3 2 T1 ¼ ¼ 87/1:005 ¼ 86:57 K
Cp

The compressor adiabatic efficiency is given by:


T 30 2 T 1 75:38
hc ¼ ¼ ¼ 0:871 or 87:1%
ðT 3 2 T 1 Þ 86:57
The air temperature at the impeller exit is:
T 3 ¼ T 1 þ 86:57 ¼ 384:57 K

Power input:
P ¼ mW
_ ¼ ð28Þð87Þ ¼ 2436 kW

Illustrative Example 4.5: A centrifugal compressor impeller rotates at

9000 rpm. If the impeller tip diameter is 0.914 m and a2 ¼ 208, calculate
the following for operation in standard sea level atmospheric conditions: (1) U2,
(2) Cw2, (3) Cr2, (4) b2, and (5) C2.
1. Impeller tip speed is given by U 2 ¼ p60 DN
¼ ðpÞð0:914Þð9000Þ
60 ¼ 431 m/s
2. Since the exit is radial and no slip, Cw2 ¼ U2 ¼ 431 m/s
3. From the velocity triangle,
Cr2 ¼ U2 tan(a2) ¼ (431) (0.364) ¼ 156.87 m/s
4. For radial exit, relative velocity is exactly perpendicular to rotational
velocity U2. Thus the angle b2 is 908 for radial exit.
5. Using the velocity triangle (Fig. 4.14),
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C2 ¼ U 22 þ C 2r2 ¼ 4312 þ 156:872 ¼ 458:67 m/s

Illustrative Example 4.6: A centrifugal compressor operates with no

prewhirl is run with a rotor tip speed of 457 m/s. If Cw2 is 95% of U2 and hc ¼ 0.88,

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 162 Chapter 4

Figure 4.14 Velocity triangle at impeller exit.

calculate the following for operation in standard sea level air: (1) pressure ratio, (2)
work input per kg of air, and (3) the power required for a flow of 29 k/s.

Solution:
1. The pressure ratio is given by (assuming s ¼ c ¼ 1):
g=ðg21Þ
P03 hc scU 22
¼ 1þ
P01 C p T 01
3:5
ð0:88Þð0:95Þð4572 Þ
¼ 1þ ¼ 5:22
ð1005Þð288Þ
2. The work per kg of air

W ¼ U 2 C w2 ¼ ð457Þð0:95Þð457Þ ¼ 198:4 kJ/kg

3. The power for 29 kg/s of air

P ¼ mW
_ ¼ ð29Þð198:4Þ ¼ 5753:6 kW

Illustrative Example 4.7: A centrifugal compressor is running at

10,000 rpm and air enters in the axial direction. The inlet stagnation temperature
of air is 290 K and at the exit from the impeller tip the stagnation temperature is
440 K. The isentropic efficiency of the compressor is 0.85, work input factor
c ¼ 1.04, and the slip factor s ¼ 0.88. Calculate the impeller tip diameter,
overall pressure ratio, and power required to drive the compressor per unit mass
flow rate of air.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Centrifugal Compressors and Fans 163

Solution:
Temperature equivalent of work done:
scU 22 ð0:88Þð1:04ÞðU 22 Þ
T 02 2 T 01 ¼ or :
Cp 1005
Therefore; U 2 ¼ 405:85 m/s
60U 2 ð60Þð405:85Þ
and D¼ ¼ ¼ 0:775 m
pN ðpÞð10; 000Þ
The overall pressure ratio is given by:
hc scU 22 ð Þ
g= g21
P03
¼ 1þ
P01 C p T 01
3:5
ð0:85Þð0:88Þð1:04Þð405:852 Þ
¼ 1þ ¼ 3:58
ð1005Þð290Þ
Power required to drive the compressor per unit mass flow:

ð1Þð0:88Þð1:04Þð405:852 Þ
P ¼ mcsU 22 ¼ ¼ 150:75 kW
1000

Design Example 4.8: Air enters axially in a centrifugal compressor at a

stagnation temperature of 208C and is compressed from 1 to 4.5 bars. The
impeller has 19 radial vanes and rotates at 17,000 rpm. Isentropic efficiency of the
compressor is 0.84 and the work input factor is 1.04. Determine the overall
diameter of the impeller and the power required to drive the compressor when the
mass flow is 2.5 kg/s.
Solution:
Since the vanes are radial, using the Stanitz formula to find the slip factor:
0:63p 0:63p
s¼12 ¼12 ¼ 0:8958
n 19
The overall pressure ratio
g=ðg21Þ
P03 hc scU 22
¼ 1þ ; or 4:5
P01 C p T 01
3:5
ð0:84Þð0:8958Þð1:04ÞðU 22 Þ
¼ 1þ ; so U 2 ¼ 449:9 m/s
ð1005Þð293Þ

60U 2 ð60Þð449:9Þ
The impeller diameter, D ¼ ¼ ¼ 0:5053 m ¼ 50:53 cm.
pN pð17; 000Þ

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 164 Chapter 4
csU 22 ð0:8958Þð1:04Þð449:92 Þ
The work done on the air W ¼ ¼ ¼
1000 1000
188:57 kJ/kg
Power required to drive the compressor: P ¼ mW
_ ¼ ð2:5Þð188:57Þ ¼
471:43 kW
Design Example 4.9: Repeat problem 4.8, assuming the air density at the
impeller tip is 1.8 kg/m3 and the axial width at the entrance to the diffuser is
12 mm. Determine the radial velocity at the impeller exit and the absolute Mach
number at the impeller tip.
Solution:
C w2
Slip factor: s ¼ ; or Cw2 ¼ ð0:8958Þð449:9Þ ¼ 403 m/s
U2
Using the continuity equation,
m_ ¼ r2 A2 C r2 ¼ r2 2pr 2 b2 C r2
where:
b2 ¼ axial width
r2 ¼ radius
Therefore:
2:5
C r2 ¼ ¼ 73:65 m/s
ð1:8Þð2pÞð0:25Þð0:012Þ
Absolute velocity at the impeller exit
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C 2 ¼ C2r2 þ C 2w2 ¼ 73:652 þ 4032 ¼ 409:67 m/s

The temperature equivalent of work done:

T 02 2 T 01 ¼ 188:57/C p ¼ 188:57/1:005 ¼ 187:63 K
Therefore, T02 ¼ 293 þ 187.63 ¼ 480.63 K
Hence the static temperature at the impeller exit is:
C 22 409:672
T 2 ¼ T 02 2 ¼ 480:63 2 ¼ 397 K
2C p ð2Þð1005Þ
Now, the Mach number at the impeller exit is:
C2 409:67
M2 ¼ pffiffiffiffiffiffiffiffiffiffiffi ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ 1:03
gRT 2 ð1:4Þð287Þð397Þ
Design Example 4.10: A centrifugal compressor is required to deliver
8 kg/s of air with a stagnation pressure ratio of 4 rotating at 15,000 rpm. The air
enters the compressor at 258C and 1 bar. Assume that the air enters axially with
velocity of 145 m/s and the slip factor is 0.89. If the compressor isentropic

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Centrifugal Compressors and Fans 165

efficiency is 0.89, find the rise in stagnation temperature, impeller tip speed,
diameter, work input, and area at the impeller eye.
Solution:
Inlet stagnation temperature:
C 21 1452
T 01 ¼ T a þ ¼ 298 þ ¼ 308:46 K
2C p ð2Þð1005Þ
Using the isentropic P – T relation for the compression process,
 ðg21Þ=g
P03
T 03 ¼ T 01
0 ¼ ð308:46Þð4Þ0:286 ¼ 458:55K
P01
Using the compressor efficiency,


T020 2 T 01 ð458:55 2 308:46Þ
T 02 2 T 01 ¼ ¼ ¼ 168:64 K
hc 0:89

Hence, work done on the air is given by:

W ¼ C p ðT 02 2 T 01 Þ ¼ ð1:005Þð168:64Þ ¼ 169:48 kJ/kg
But,
ð0:89ÞðU 2 Þ
W ¼ sU 22 ¼ ; or :169:48 ¼ 0:89U 22 /1000
1000
or:
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ð1000Þð169:48Þ
U2 ¼ ¼ 436:38 m/s
0:89
Hence, the impeller tip diameter
60U 2 ð60Þð436:38Þ
D¼ ¼ ¼ 0:555 m
pN p ð15; 000Þ
The air density at the impeller eye is given by:
P1 ð1Þð100Þ
r1 ¼ ¼ ¼ 1:17 kg/m3
RT 1 ð0:287Þð298Þ
Using the continuity equation in order to find the area at the impeller eye,
m_ 8
A1 ¼ ¼ ¼ 0:047 m2
r1 C 1 ð1:17Þð145Þ
The power input is:
P ¼ m_ W ¼ ð8Þð169:48Þ ¼ 1355:24 kW

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 166 Chapter 4

Figure 4.15 The velocity triangle at the impeller eye.

Design Example 4.11: The following data apply to a double-sided

centrifugal compressor (Fig. 4.15):

Impeller eye tip diameter: 0:28 m

Impeller eye root diameter: 0:14 m
Impeller tip diameter: 0:48 m
Mass flow of air: 10 kg/s
Inlet stagnation temperature: 290 K
Inlet stagnation pressure: 1 bar
Air enters axially with velocity: 145 m/s
Slip factor: 0:89
Power input factor: 1:03
Rotational speed: 15; 000 rpm

Calculate (1) the impeller vane angles at the eye tip and eye root, (2) power
input, and (3) the maximum Mach number at the eye.
Solution:
(1) Let Uer be the impeller speed at the eye root. Then the vane angle at
the eye root is:
 
Ca
aer ¼ tan21
U er
and
pDer N pð0:14Þð15; 000Þ
U er ¼ ¼ ¼ 110 m/s
60 60
Hence, the vane angle at the impeller eye root:
   
21 C a 21 145 0
aer ¼ tan ¼ tan ¼ 528 48
U er 110

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Centrifugal Compressors and Fans 167

Impeller velocity at the eye tip:

pDet N pð0:28Þð15; 000Þ

U et ¼ ¼ ¼ 220 m/s
60 60

Therefore vane angle at the eye tip:

   
21 Ca 21145 0
aet ¼ tan ¼ tan ¼ 338 23
U et 220

(2) Work input:

W ¼ m_ csU 22 ¼ ð10Þð0:819Þð1:03U 22 Þ

but:
pD2 N pð0:48Þð15; 000Þ
U2 ¼ ¼ ¼ 377:14 m/s
60 60
Hence,
ð10Þð0:89Þð1:03Þð377:142 Þ
W¼ ¼ 1303:86 kW
1000
(3) The relative velocity at the eye tip:
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
V1 ¼ U 2et þ C2a ¼ 2202 þ 1452 ¼ 263:5 m/s

Hence, the maximum relative Mach number at the eye tip:

V1
M1 ¼ pffiffiffiffiffiffiffiffiffiffiffi ;
gRT 1
where T1 is the static temperature at the inlet
C21 1452
T 1 ¼ T 01 2 ¼ 290 2 ¼ 279:54 K
2Cp ð2Þð1005Þ
The Mach number at the inlet then is:

V1 263/5
M1 ¼ pffiffiffiffiffiffiffiffiffiffiffi ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ 0:786
gRT 1 ð1:4Þð287Þð279:54Þ

Design Example 4.12: Recalculate the maximum Mach number at the

impeller eye for the same data as in the previous question, assuming prewhirl
angle of 208.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 168 Chapter 4

Figure 4.16 The velocity triangle at the impeller eye.

Solution:
Figure 4.16 shows the velocity triangle with the prewhirl angle.
From the velocity triangle:

145
C1 ¼ ¼ 154:305 m/s
cosð208 Þ

Equivalent dynamic temperature:

C 21 154:3052
¼ ¼ 11:846 K
2C p ð2Þð1005Þ

C w1 ¼ tanð208Þ Ca1 ¼ ð0:36Þð145Þ ¼ 52:78 m/s

Relative velocity at the inlet:

V 21 ¼ C 2a þ ðU e 2 C w1 Þ2 ¼ 1452 þ ð220 2 52:78Þ2 ; or V 1

¼ 221:3 m/s

Therefore the static temperature at the inlet:

C 21
T 1 ¼ T 01 2 ¼ 290 2 11:846 ¼ 278:2 K
2C p
Hence,
V1 221:3
M1 ¼ pffiffiffiffiffiffiffiffiffiffiffi ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ 0:662
gRT 1 ð1:4Þð287Þð278:2Þ

Note the reduction in Mach number due to prewhirl.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Centrifugal Compressors and Fans 169

Design Example 4.13: The following data refers to a single-sided

centrifugal compressor:
Ambient Temperature: 288 K
Ambient Pressure: 1 bar
Hub diameter: 0:125 m
Eye tip diameter: 0:25 m
Mass flow: 5:5 kg/s
Speed: 16; 500 rpm

Assume zero whirl at the inlet and no losses in the intake duct. Calculate the
blade inlet angle at the root and tip and the Mach number at the eye tip.
Solution:
Let: rh ¼ hub radius
rt ¼ tip radius
The flow area of the impeller inlet annulus is:



A1 ¼ p r 2t 2 r 2h ¼ p 0:1252 2 0:06252 ¼ 0:038 m2

Axial velocity can be determined from the continuity equation but since the
inlet density (r1) is unknown a trial and error method must be followed.
Assuming a density based on the inlet stagnation condition,

P01 ð1Þð105 Þ
r1 ¼ ¼ ¼ 1:21 kg/m3
RT 01 ð287Þð288Þ
Using the continuity equation,
m_ 5:5
Ca ¼ ¼ ¼ 119:6 m/s
r1 A1 ð1:21Þð0:038Þ
Since the whirl component at the inlet is zero, the absolute velocity at the
inlet is C1 ¼ Ca.
The temperature equivalent of the velocity is:

C21 119:62
¼ ¼ 7:12 K
2Cp ð2Þð1005Þ
Therefore:

C21
T 1 ¼ T 01 2 ¼ 288 2 7:12 ¼ 280:9 K
2Cp

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 170 Chapter 4

Using isentropic P– T relationship,

 g=ðg21Þ  3:5
P1 T1 5 280:9
¼ ; or P1 ¼ 10 ¼ 92 kPa
P01 T 01 288
and:
P1 ð92Þð103 Þ
r1 ¼ ¼ ¼ 1:14 kg/m3 ; and
RT 1 ð287Þð280:9Þ

5:5
Ca ¼ ¼ 126:96 m/s
ð1:14Þð0:038Þ
Therefore:
C 21 ð126:96Þ2
¼ ¼ 8:02 K
2C p 2ð1005Þ

T 1 ¼ 288 2 8:02 ¼ 279:988 K

 
279:98 3:5
P1 ¼ 10 5
¼ 90:58 kPa
288

ð90:58Þð103 Þ
r1 ¼ ¼ 1:13 kg/m3
ð287Þð279:98Þ
Further iterations are not required and the value of r1 ¼ 1.13 kg/m3 may be
taken as the inlet density and Ca ¼ C1 as the inlet velocity. At the eye tip:
2pr et N 2p ð0:125Þð16; 500Þ
U et ¼ ¼ ¼ 216 m/s
60 60
The blade angle at the eye tip:
   
U et 216
bet ¼ tan21 ¼ tan21 ¼ 59:568
Ca 126:96
At the hub,
2pð0:0625Þð16; 500Þ
U eh ¼ ¼ 108 m/s
60
The blade angle at the hub:
 
108
beh ¼ tan21 ¼ 40:398
126:96

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Centrifugal Compressors and Fans 171

The Mach number based on the relative velocity at the eye tip using the
inlet velocity triangle is:
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
V 1 ¼ C2a þ U 21 ¼ 126:962 þ 2162 ; or V 1 ¼ 250:6 m/s
The relative Mach number

V1 250:6
M ¼ pffiffiffiffiffiffiffiffiffiffiffi ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ 0:747
gRT 1 ð1:4Þð287Þð279:98Þ

Design Example 4.14: A centrifugal compressor compresses air at ambient

temperature and pressure of 288 K and 1 bar respectively. The impeller tip speed
is 364 m/s, the radial velocity at the exit from the impeller is 28 m/s, and the slip
factor is 0.89. Calculate the Mach number of the flow at the impeller tip. If the
impeller total-to-total efficiency is 0.88 and the flow area from the impeller is
0.085 m2, calculate the mass flow rate of air. Assume an axial entrance at the
impeller eye and radial blades.
Solution:
The absolute Mach number of the air at the impeller tip is:
C2
M2 ¼ pffiffiffiffiffiffiffiffiffiffiffi
gRT 2
where T2 is the static temperature at the impeller tip. Let us first calculate
C2 and T2.
Slip factor:
C w2
s¼
U2
Or:
Cw2 ¼ sU 2 ¼ ð0:89Þð364Þ ¼ 323:96 m/s
From the velocity triangle,
C22 ¼ C 2r2 þ C2w2 ¼ 282 þ 323:962 ¼ ð1:06Þð105 Þ m2 /s2
With zero whirl at the inlet
W
¼ sU 22 ¼ C p ðT 02 2 T 01 Þ
m
Hence,
sU 22 ð0:89Þð3642 Þ
T 02 ¼ T 01 þ ¼ 288 þ ¼ 405:33 K
Cp 1005

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 172 Chapter 4

Static Temperature
C 22 106000
T 2 ¼ T 02 2 ¼ 405:33 2 ¼ 352:6 K
2C p ð2Þð1005Þ
Therefore,
 12
ð1:06Þð105 Þ
M2 ¼ ¼ 0:865
ð1:4Þð287Þð352:6Þ
Using the isentropic P– T relation:
    g=ðg21Þ
P02 T 02
¼ 1 þ hc 21
P01 T 01
  3:5
405:33
¼ 1 þ 0:88 21 ¼ 2:922
288
   3:5  3:5
P2 T2 352:6
¼ ¼ ¼ 0:614
P02 T 02 405:33
Therefore,
  
P2 P02
P2 ¼ P01
P02 P01
¼ ð0:614Þð2:922Þð1Þð100Þ
¼ 179:4 kPa
179:4ð1000Þ
r2 ¼ ¼ 1:773 kg/m3
287ð352:6Þ

Mass flow:
m_ ¼ ð1:773Þð0:085Þð28Þ ¼ 4:22 kg/s
Design Example 4.15: The impeller of a centrifugal compressor rotates at
15,500 rpm, inlet stagnation temperature of air is 290 K, and stagnation pressure
at inlet is 101 kPa. The isentropic efficiency of impeller is 0.88, diameter of the
impellar is 0.56 m, axial depth of the vaneless space is 38 mm, and width of the
vaneless space is 43 mm. Assume ship factor as 0.9, power input factor 1.04, mass
flow rate as 16 kg/s. Calculate
1. Stagnation conditions at the impeller outlet, assume no fore whirl at the
inlet,
2. Assume axial velocity approximately equal to 105 m/s at the impeller
outlet, calculate the Mach number and air angle at the impeller outlet,

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Centrifugal Compressors and Fans 173

3. The angle of the diffuser vane leading edges and the Mach number at
this radius if the diffusion in the vaneless space is isentropic.
Solution:
1. Impeller tip speed
pD2 N p £ 0:56 £ 15500
U2 ¼ ¼
60 60
U 2 ¼ 454:67 m/s
Overall stagnation temperature rise
csU 22 1:04 £ 0:9 £ 454:672
T 03 2 T 01 ¼ ¼
1005 1005
¼ 192:53K

Since T03 ¼ T02

Therefore, T02 2 T01 ¼ 192.53K and T02 ¼ 192.53 þ 290 ¼ 482.53K
Now pressure ratio for impeller
 3:5  
p02 T 02 482:53 3:5
¼ ¼ ¼ 5:94
p01 T 01 290
then, p02 ¼ 5.94 £ 101 ¼ 600 KPa
2.
C w2
s¼
U2
Cw2 ¼ sU 2
or
Cw2 ¼ 0:9 £ 454:67 ¼ 409 m/s
Let Cr2 ¼ 105 m/s
Outlet area normal to periphery

A2 ¼ pD2 £ impeller depth

¼ p £ 0:56 £ 0:038
A2 ¼ 0:0669 m2

From outlet velocity triangle

C 2 2 ¼ C r22 þ Cw22
¼ 1052 þ 4092
C 2 2 ¼ 178306

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 174 Chapter 4

i:e: C 2 ¼ 422:26 m/s

C 22 422:262
T 2 ¼ T 02 2 ¼ 482:53 2
2C p 2 £ 1005

T 2 ¼ 393:82 K
Using isentropic P –T relations
 g21 g  
T2 393:82 3:5
P2 ¼ P02 ¼ 600 ¼ 294:69 kPa
T 02 482:53

From equation of state

P2 293:69 £ 103
r2 ¼ ¼ ¼ 2:61 kg/m3
RT 2 287 £ 393:82
The equation of continuity gives
m_ 16
C r2 ¼ ¼ ¼ 91:63 m/s
A2 P2 0:0669 £ 2:61
Thus, impeller outlet radial velocity ¼ 91.63 m/s
Impeller outlet Mach number
C2 422:26
M2 ¼ pffiffiffiffiffiffiffiffiffiffiffi ¼
gRT 2 ð1:4 £ 287 £ 393:82Þ0:5

M2 ¼ 1:06

From outlet velocity triangle

C r2 91:63
Cosa2 ¼ ¼ ¼ 0:217
C2 422:26
i.e., a2 ¼ 77.478
3. Assuming free vortex flow in the vaneless space and for
convenience denoting conditions at the diffuser vane without a
subscript (r ¼ 0.28 þ 0.043 ¼ 0.323)

Cw2 r 2 409 £ 0:28

Cw ¼ ¼
r 0:323

C w ¼ 354:55 m/s

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Centrifugal Compressors and Fans 175

The radial component of velocity can be found by trial and error. Choose as a
first try, Cr ¼ 105 m/s

C2 1052 þ 354:552
¼ ¼ 68 K
2C p 2 £ 1005

T ¼ 482.53 2 68 (since T ¼ T02 in vaneless space)

T ¼ 414.53K
 3:5  
T2 419:53 3:5
p ¼ p02 ¼ 600 ¼ 352:58 kPa
T 02 482:53

p2 294:69
r¼ ¼
RT 2 287 £ 393:82

r ¼ 2:61 kg/m3

The equation of continuity gives

A ¼ 2pr £ depth of vanes
¼ 2p £ 0:323 £ 0:038
¼ 0:0772 m2

16
Cr ¼ ¼ 79:41 m/s
2:61 £ 0:0772
Next try Cr ¼ 79.41 m/s

C2 79:412 þ 354:552
¼ ¼ 65:68
2Cp 2 £ 1005

T ¼ 482:53 2 65:68 ¼ 416:85K

 3:5  
T 416:85 3:5
p ¼ p02 ¼ 600
T 02 482:53

p ¼ 359:54 Pa

359:54
r¼ ¼ 3 kg/m3
416:85 £ 287

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 176 Chapter 4

16
Cr ¼ ¼ 69:08 m/s
3:0 £ 0:772
Try Cr ¼ 69:08 m/s
C2 69:082 þ 354:552
¼ ¼ 64:9
2C p 2 £ 1005

T ¼ 482:53 2 64:9 ¼ 417:63 K

 3:5  
T 417:63 3:5
p ¼ p02 ¼ 600
T 02 482:53
p ¼ 361:9 Pa
361:9
r¼ ¼ 3:02 kg/m3
417:63 £ 287
16
Cr ¼ ¼ 68:63 m/s
3:02 £ 0:772
Taking Cr as 62.63 m/s, the vane angle
Cw
tan a ¼
Cr
354:5
¼ ¼ 5:17
68:63
i.e. a ¼ 798
Mach number at vane
 
65:68 £ 2 £ 1005 1=2
M¼ ¼ 0:787
1:4 £ 287 £ 417:63

Design Example 4.16: The following design data apply to a double-sided

centrifugal compressor:
Impeller eye root diameter: 18 cm
Impeller eye tip diameter: 31:75 cm
Mass flow: 18:5 kg/s
Impeller speed: 15500 rpm
Inlet stagnation pressure: 1:0 bar
Inlet stagnation temperature: 288 K
Axial velocity at inlet ðconstantÞ: 150 m/s

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Centrifugal Compressors and Fans 177

Find suitable values for the impeller vane angles at root and tip of eye if the
air is given 208 of prewhirl at all radii, and also find the maximum Mach number
at the eye.
Solution:
At eye root, Ca ¼ 150 m/s
Ca 150
[ C1 ¼ ¼ ¼ 159:63 m/s
cos208 cos208
and Cw1 ¼ 150 tan 208 ¼ 54.6 m/s
Impeller speed at eye root
pDer N p £ 0:18 £ 15500
U er ¼ ¼
60 60
U er ¼ 146 m/s
From velocity triangle
Ca 150 150
tan ber ¼ ¼ ¼ ¼ 1:641
U er 2 C w1 146 2 54:6 91:4
i:e:; ber ¼ 58:648
At eye tip from Fig. 4.17(b)
pDet N p £ 0:3175 £ 15500
U et ¼
60 60
U et ¼ 258 m/s

Figure 4.17 Velocity triangles at (a) eye root and (b) eye tip.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 178 Chapter 4

150 150
tan aet ¼ ¼ ¼ 0:7375
258 2 54:6 203:4
i:e: aet ¼ 36:418

Mach number will be maximum at the point where relative velocity is

maximum.
Relative velocity at eye root is:
Ca 150 150
V er ¼ ¼ ¼
sin ber sin 58:648 0:8539
V er ¼ 175:66 m/s
Relative velocity at eye tip is:
Ca 150 150
V et ¼ ¼ ¼
sin aet sin 36:418 0:5936
V et ¼ 252:7 m/s
Relative velocity at the tip is maximum.
Static temperature at inlet:
V 2et 252:72
T 1 ¼ T 01 ¼ ¼ 288 2 ¼ 288 2 31:77
2C p 2 £ 1005
T 1 ¼ 256:23 K
V et 252:7 252:7
Mmax ¼
1=2 ¼ ¼
gRT 1 ð1:4 £ 287 £ 256:23Þ 1=2 320:86

Mmax ¼ 0:788

Design Example 4.17: In a centrifugal compressor air enters at a

stagnation temperature of 288 K and stagnation pressure of 1.01 bar. The impeller
has 17 radial vanes and no inlet guide vanes. The following data apply:
Mass flow rate: 2:5 kg/s
Impeller tip speed: 475 m/s
Mechanical efficiency: 96%
Absolute air velocity at diffuser exit: 90 m/s
Compressor isentropic efficiency: 84%
Absolute velocity at impeller inlet: 150 m/s

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Centrifugal Compressors and Fans 179

Diffuser efficiency: 82%

Axial depth of impeller: 6:5 mm
Power input factor: 1:04
g for air: 1:4

Determine:
1. shaft power
2. stagnation and static pressure at diffuser outlet
3. radial velocity, absolute Mach number and stagnation and static
pressures at the impeller exit, assume reaction ratio as 0.5, and
4. impeller efficiency and rotational speed
Solution:
1. Mechanical efficiency is
Work transferred to air
hm ¼
Work supplied to shaft

W
or shaft power ¼
hm
for vaned impeller, slip factor, by Stanitz formula is
0:63p 0:63 £ p
s¼12 ¼12
n 17
s ¼ 0:884
Work input per unit mass flow

W ¼ csU 2 C w2

Since C w1 ¼ 0

¼ csU 22

¼ 1:04 £ 0:884 £ 4752

Work input for 2.5 kg/s

W ¼ 1:04 £ 0:884 £ 2:5 £ 4752

W ¼ 518:58K

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 180 Chapter 4

518:58
Hence; Shaft Power ¼ ¼ 540:19kW
0:96
2. The overall pressure ratio is
" #g=ðg21Þ
p03 hc csU 22
¼ 1þ
p01 C p T 01

3:5
0:84 £ 1:04 £ 0:884 £ 4752
¼ 1þ ¼ 5:2
1005 £ 288
Stagnation pressure at diffuser exit
P03 ¼ p01 £ 5:20 ¼ 1:01 £ 5:20

P03 ¼ 5:25 bar

 g=g21
p3 T3
¼
p03 T 03

W ¼ m £ C p ðT 03 2 T 01 Þ

W 518:58 £ 103
[ T 03 ¼ þ T 01 ¼ þ 288 ¼ 494:4 K
mC p 2:5 £ 1005
Static temperature at diffuser exit

C 32 902
T 3 ¼ T 03 2 ¼ 494:4 2
2C p 2 £ 1005

T 3 ¼ 490:37 K

Static pressure at diffuser exit

 g=g21  
T3 490:37 3:5
p3 ¼ p03 ¼ 5:25
T 03 494:4

p3 ¼ 5:10 bar

3. The reaction is
T2 2 T1
0:5 ¼
T3 2 T1

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Centrifugal Compressors and Fans 181

and  2 
C1 2 C3 2 W 1502 2 902
T 3 2 T 1 ¼ ðT 03 2 T 01 Þ þ ¼ þ
2C p mC p 2 £ 1005

518:58 £ 103
¼ þ 7:164 ¼ 213:56 K
2:5 £ 1005
Substituting
T 2 2 T 1 ¼ 0:5 £ 213:56
¼ 106:78 K
Now
C 12
T 2 ¼ T 01 2 þ ðT 2 2 T 1 Þ
2Cp
¼ 288 2 11:19 þ 106:78
T 2 ¼ 383:59 K
At the impeller exit
C2 2
T 02 ¼ T 2 þ
2C p
or
C2 2
T 03 ¼ T 2 þ ðSince T 02 ¼ T 03 Þ
2C p
Therefore,

C 2 2 ¼ 2C p ½ðT 03 2 T 01 Þ þ ðT 01 2 T 2 Þ
¼ 2 £ 1005ð206:4 þ 288 þ 383:59Þ
C 2 ¼ 471:94 m/s

Mach number at impeller outlet

C2
M2 ¼
ð1:4 £ 287 £ 383:59Þ1=2

M2 ¼ 1:20
Radial velocity at impeller outlet

C r22 ¼ C 2 2 2 Cw22
¼ ð471:94Þ2 2 ð0:884 £ 475Þ2

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 182 Chapter 4

C r22 ¼ 215:43 m/s

Diffuser efficiency is given by

h30 2 h2 isentropic enthalpy increase T 30 2 T 2
hD ¼ ¼ ¼
h3 2 h2 actual enthalpy increase T3 2 T2
T   g21=g
T2 3
0
21 T2 p3
p2 21
T2
¼ ¼
T3 2 T2 ðT 3 2 T 2 Þ

Therefore
  3:5
T3 2 T2
p3
p2 ¼ 1 þ hD
T2
 
0:821 £ 106:72 3:5
¼ 1þ
383:59
¼ 2:05
5:10
or p2 ¼ ¼ 2:49 bar
2:05
From isentropic P –T relations
 3:5  
T 02 494:4 3:5
p02 ¼ p2 ¼ 2:49
T2 383:59
p02 ¼ 6:05 bar
4. Impeller efficiency is
 g21 g
T 01 pp0201 21
hi ¼
T 03 2 T 01
"  #
6:05 0:286
288 21
1:01
¼
494:4 2 288
¼ 0:938
p2 2:49 £ 105
r2 ¼ ¼
RT 2 287 £ 383:59

r2 ¼ 2:27 kg/m3

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Centrifugal Compressors and Fans 183

m_ ¼ r2 A2 Cr2
¼ 2p r 2 r2 b2

But
pND2 pN m_
U2 ¼ ¼
60 r2 pCr2 b2 £ 60
475 £ 2:27 £ 246:58 £ 0:0065 £ 60
N¼
2:5
N ¼ 41476 rpm

PROBLEMS
4.1 The impeller tip speed of a centrifugal compressor is 450 m/s with no
prewhirl. If the slip factor is 0.90 and the isentropic efficiency of the
compressor is 0.86, calculate the pressure ratio, the work input per kg of
air, and the power required for 25 kg/s of airflow. Assume that
the compressor is operating at standard sea level and a power input
factor of 1.
(4.5, 182.25 kJ/kg, 4556.3 kW)
4.2 Air with negligible velocity enters the impeller eye of a centrifugal
compressor at 158C and 1 bar. The impeller tip diameter is 0.45 m and
rotates at 18,000 rpm. Find the pressure and temperature of the air at the
compressor outlet. Neglect losses and assume g ¼ 1.4.
(5.434 bar, 467 K)
4.3 A centrifugal compressor running at 15,000 rpm, overall diameter of the
impeller is 60 cm, isentropic efficiency is 0.84 and the inlet stagnation
temperature at the impeller eye is 158C. Calculate the overall pressure ratio,
and neglect losses.
(6)
4.4 A centrifugal compressor that runs at 20,000 rpm has 20 radial vanes, power
input factor of 1.04, and inlet temperature of air is 108C. If the pressure ratio
is 2 and the impeller tip diameter is 28 cm, calculate the isentropic efficiency
of the compressor. Take g ¼ 1.4 (77.4%)
4.5 Derive the expression for the pressure ratio of a centrifugal compressor:
g=ðg21Þ
P03 hc scU 22
¼ 1þ
P01 C p T 01

4.6 Explain the terms “slip factor” and “power input factor.”

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 184 Chapter 4

4.7 What are the three main types of centrifugal compressor impellers? Draw
the exit velocity diagrams for these three types.
4.8 Explain the phenomenon of stalling, surging and choking in centrifugal
compressors.
4.9 A centrifugal compressor operates with no prewhirl and is run with a tip
speed of 475 the slip factor is 0.89, the work input factor is 1.03,
compressor efficiency is 0.88, calculate the pressure ratio, work input per
kg of air and power for 29 airflow. Assume T01 ¼ 290 K and Cp ¼ 1.005
kJ/kg K.
(5.5, 232.4 kJ/kg, 6739 kW)
4.10 A centrifugal compressor impeller rotates at 17,000 rpm and compresses
32 kg of air per second. Assume an axial entrance, impeller trip radius is
0.3 m, relative velocity of air at the impeller tip is 105 m/s at an exit angle
of 808. Find the torque and power required to drive this machine.
(4954 Nm, 8821 kW)
4.11 A single-sided centrifugal compressor designed with no prewhirl has the
following dimensions and data:
Total head / pressure ratio: 3:8:1
Speed: 12; 000 rpm
Inlet stagnation temperature: 293 K
Inlet stagnation pressure: 1:03 bar
Slip factor: 0:9
Power input factor: 1:03
Isentropic efficiency: 0:76
Mass flow rate: 20 kg/s

Assume an axial entrance. Calculate the overall diameter of the impeller

and the power required to drive the compressor.
(0.693 m, 3610 kW)
4.12 A double-entry centrifugal compressor designed with no prewhirl has the
following dimensions and data:
Impeller root diameter: 0:15 m
Impeller tip diameter: 0:30 m
Rotational speed: 15; 000 rpm
Mass flow rate: 18 kg/s

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Centrifugal Compressors and Fans 185

Ambient temperature: 258C

Ambient pressure: 1:03 bar
Density of air
at eye inlet: 1:19 kg/m3

Assume the axial entrance and unit is stationary. Find the inlet angles of
the vane at the root and tip radii of the impeller eye and the maximum
Mach number at the eye.
(a1 at root ¼ 50.78, a1 ¼ 31.48 at tip, 0.79)

4.13 In Example 4.12, air does not enter the impeller eye in an axial direction
but it is given a prewhirl of 208 (from the axial direction). The remaining
values are the same. Calculate the inlet angles of the impeller vane at the
root and tip of the eye.
(a1 at root ¼ 65.58, a1 at tip ¼ 38.18, 0.697)

NOTATION
C absolute velocity
r radius
U impeller speed
V relative velocity
a vane angle
s slip factor
v angular velocity
c power input factor

SUFFIXES
1 inlet to rotor
2 outlet from the rotor
3 outlet from the diffuser
a axial, ambient
r radial
w whirl

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 5
Axial Flow Compressors and Fans

5.1 INTRODUCTION
As mentioned in Chapter 4, the maximum pressure ratio achieved in centrifugal
compressors is about 4:1 for simple machines (unless multi-staging is used) at an
efficiency of about 70– 80%. The axial flow compressor, however, can achieve
higher pressures at a higher level of efficiency. There are two important
characteristics of the axial flow compressor—high-pressure ratios at good
efficiency and thrust per unit frontal area. Although in overall appearance, axial
turbines are very similar, examination of the blade cross-section will indicate a
big difference. In the turbine, inlet passage area is greater than the outlet. The
opposite occurs in the compressor, as shown in Fig. 5.1.
Thus the process in turbine blades can be described as an accelerating flow,
the increase in velocity being achieved by the nozzle. However, in the axial flow
compressor, the flow is decelerating or diffusing and the pressure rise occurs
when the fluid passes through the blades. As mentioned in the chapter on diffuser
design (Chapter 4, Sec. 4.7), it is much more difficult to carry out efficient
diffusion due to the breakaway of air molecules from the walls of the diverging
passage. The air molecules that break away tend to reverse direction and flow
back in the direction of the pressure gradient. If the divergence is too rapid, this
may result in the formation of eddies and reduction in useful pressure rise. During
acceleration in a nozzle, there is a natural tendency for the air to fill the passage

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 188 Chapter 5

Figure 5.1 Cutaway sketch of a typical axial compressor assembly: the General
Electric J85 compressor. (Courtesy of General Electric Co.)

walls closely (only the normal friction loss will be considered in this case).
Typical blade sections are shown in Fig. 5.2. Modern axial flow compressors may
give efficiencies of 86– 90%—compressor design technology is a well-developed
field. Axial flow compressors consist of a number of stages, each stage being
formed by a stationary row and a rotating row of blades.
Figure 5.3 shows how a few compressor stages are built into the axial
compressor. The rotating blades impart kinetic energy to the air while increasing
air pressure and the stationary row of blades redirect the air in the proper direction
and convert a part of the kinetic energy into pressure. The flow of air through the
compressor is in the direction of the axis of the compressor and, therefore, it is
called an axial flow compressor. The height of the blades is seen to decrease as
the fluid moves through the compressor. As the pressure increases in the direction
of flow, the volume of air decreases. To keep the air velocity the same for each
stage, the blade height is decreased along the axis of the compressor. An extra
row of fixed blades, called the inlet guide vanes, is fitted to the compressor inlet.
These are provided to guide the air at the correct angle onto the first row of
moving blades. In the analysis of the highly efficient axial flow compressor,
the 2-D flow through the stage is very important due to cylindrical symmetry.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow Compressors and Fans 189

Figure 5.2 Compressor and turbine blade passages: turbine and compressor housing.

Figure 5.3 Schematic of an axial compressor section.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 190 Chapter 5

The flow is assumed to take place at a mean blade height, where the blade
peripheral velocities at the inlet and outlet are the same. No flow is assumed in the
radial direction.

5.2 VELOCITY DIAGRAM

The basic principle of axial compressor operation is that kinetic energy is
imparted to the air in the rotating blade row, and then diffused through passages
of both rotating and stationary blades. The process is carried out over multiple
numbers of stages. As mentioned earlier, diffusion is a deceleration process. It is
efficient only when the pressure rise per stage is very small. The blading diagram
and the velocity triangle for an axial flow compressor stage are shown in Fig. 5.4.
Air enters the rotor blade with absolute velocity C1 at an angle a1 measured
from the axial direction. Air leaves the rotor blade with absolute velocity C2 at an
angle a2. Air passes through the diverging passages formed between the rotor
blades. As work is done on the air in the rotor blades, C2 is larger than C1. The
rotor row has tangential velocity U. Combining the two velocity vectors gives the
relative velocity at inlet V1 at an angle b1. V2 is the relative velocity at the rotor
outlet. It is less than V1, showing diffusion of the relative velocity has taken place
with some static pressure rise across the rotor blades. Turning of the air towards
the axial direction is brought about by the camber of the blades. Euler’s equation

Figure 5.4 Velocity diagrams for a compressor stage.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow Compressors and Fans 191

provides the work done on the air:

W c ¼ UðCw2 2 C w1 Þ ð5:1Þ

Using the velocity triangles, the following basic equations can be written:
U
¼ tan a1 þ tan b1 ð5:2Þ
Ca

U
¼ tan a2 þ tan b2 ð5:3Þ
Ca
in which Ca ¼ Ca1 ¼ C2 is the axial velocity, assumed constant through the stage.
The work done equation [Eq. (5.1)] may be written in terms of air angles:

W c ¼ UC a ðtan a2 2 tan a1 Þ ð5:4Þ

also,

W c ¼ UC a ðtan b1 2 tan b2 Þ ð5:5Þ

The whole of this input energy will be absorbed usefully in raising the pressure and
velocity of the air and for overcoming various frictional losses. Regardless of the
losses, all the energy is used to increase the stagnation temperature of the air, KT0s.
If the velocity of air leaving the first stage C3 is made equal to C1, then the
stagnation temperature rise will be equal to the static temperature rise, KTs. Hence:

UC a
T 0s ¼ DT s ¼ ðtan b1 2 tan b2 Þ ð5:6Þ
Cp

Equation (5.6) is the theoretical temperature rise of the air in one stage. In reality,
the stage temperature rise will be less than this value due to 3-D effects in the
compressor annulus. To find the actual temperature rise of the air, a factor l, which
is between 0 and 100%, will be used. Thus the actual temperature rise of the air is
given by:

lUC a
T 0s ¼ ðtan b1 2 tan b2 Þ ð5:7Þ
Cp

If Rs is the stage pressure ratio and hs is the stage isentropic efficiency, then:
g=ðg21Þ
hs DT 0s
Rs ¼ 1 þ ð5:8Þ
T 01

where T01 is the inlet stagnation temperature.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 192 Chapter 5

5.3 DEGREE OF REACTION

The degree of reaction, L, is defined as:
Static enthalpy rise in the rotor
L¼ ð5:9Þ
Static enthalpy rise in the whole stage
The degree of reaction indicates the distribution of the total pressure rise into the
two types of blades. The choice of a particular degree of reaction is important in
that it affects the velocity triangles, the fluid friction and other losses.
Let:
DT A ¼ the static temperature rise in the rotor

DT B ¼ the static temperature rise in the stator

Using the work input equation [Eq. (5.4)], we get:
W c ¼ Cp ðDT A þ DT B Þ ¼ DT S
)
¼ UC a ðtan b1 2 tan b2 Þ
ð5:10Þ
¼ UC a ðtan a2 2 tan a1 Þ
But since all the energy is transferred to the air in the rotor, using the steady flow
energy equation, we have:
1
W c ¼ C p DT A þ ðC 22 2 C 21 Þ ð5:11Þ
2
Combining Eqs. (5.10) and (5.11), we get:
1
C p DT A ¼ UC a ðtan a2 2 tan a1 Þ 2 ðC 22 2 C 21 Þ
2
from the velocity triangles,
C 2 ¼ C a cos a2 and C 1 ¼ C a cos a1
Therefore,
C p DT A ¼ UC a ðtan a2 2 tan a1 Þ 2 12 C2a ðsec2 a2 2 sec2 a1 Þ
¼ UC a ðtan a2 2 tan a1 Þ 2 12 C2a ðtan 2 a2 2 tan 2 a1 Þ

Using the definition of degree of reaction,

DT A
L¼
DT A þ DT B
UC a ðtan a2 2 tan a1 Þ 2 12 C 2a ðtan 2 a2 2 tan 2 a1 Þ
¼
UC a ðtan a2 2 tan a1 Þ
¼ 1 2 CUa ðtan a2 þ tan a1 Þ

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow Compressors and Fans 193

But from the velocity triangles, adding Eqs. (5.2) and (5.3),
2U
¼ ðtan a1 þ tan b1 þ tan a2 þ tan b2 Þ
Ca
Therefore,
 
C a 2U 2U
L¼ 2 þ tan b1 þ tan b2
2U C a Ca
Ca ð5:12Þ
¼ ðtan b1 þ tan b2 Þ
2U
Usually the degree of reaction is set equal to 50%, which leads to this interesting
result:
U
ðtan b1 þ tan b2 Þ ¼ :
Ca
Again using Eqs. (5.1) and (5.2),
tan a1 ¼ tan b2 ; i:e:; a1 ¼ b 2
tan b1 ¼ tan a2 ; i:e:; a2 ¼ b 1

As we have assumed that Ca is constant through the stage,

Ca ¼ C 1 cos a1 ¼ C 3 cos a3 :
Since we know C1 ¼ C3, it follows that a1 ¼ a3. Because the angles are equal,
a1 ¼ b2 ¼ a3, and b1 ¼ a2. Under these conditions, the velocity triangles become
symmetric. In Eq. (5.12), the ratio of axial velocity to blade velocity is called the
flow coefficient and denoted by F. For a reaction ratio of 50%,
(h2 2 h1) ¼ (h3 2 h1), which implies the static enthalpy and the temperature
increase in the rotor and stator are equal. If for a given value of C a =U, b2 is chosen
to be greater than a2 (Fig. 5.5), then the static pressure rise in the rotor is greater
than the static pressure rise in the stator and the reaction is greater than 50%.

Figure 5.5 Stage reaction.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 194 Chapter 5

Conversely, if the designer chooses b2 less than b1, the stator pressure rise will be
greater and the reaction is less than 50%.

5.4 STAGE LOADING

The stage-loading factor C is defined as:
Wc h03 2 h01
C¼ ¼
mU 2 U2
lðC w2 2 C w1 Þ
¼
U
lC a
¼ ðtan a2 2 tan a1 Þ
U ð5:13Þ
C ¼ lF ðtan a2 2 tan a1 Þ

5.5 LIFT-AND-DRAG COEFFICIENTS

The stage-loading factor C may be expressed in terms of the lift-and-drag
coefficients. Consider a rotor blade as shown in Fig. 5.6, with relative velocity
vectors V1 and V2 at angles b1 and b2. Let tan ðbm Þ ¼ ðtan ðb1 Þ þ tan ðb2 ÞÞ/2. The
flow on the rotor blade is similar to flow over an airfoil, so lift-and-drag forces will
be set up on the blade while the forces on the air will act on the opposite direction.
The tangential force on each moving blade is:
F x ¼ L cos bm þ D sin bm
 
CD
F x ¼ L cos bm 1 þ tan bm ð5:14Þ
CL
where: L ¼ lift and D ¼ drag.

Figure 5.6 Lift-and-drag forces on a compressor rotor blade.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow Compressors and Fans 195

The lift coefficient is defined as:

L
CL ¼ ð5:15Þ
0:5rV 2m A
where the blade area is the product of the chord c and the span l.
Substituting V m ¼ cosCba m into the above equation,
 
rC 2 clC L CD
Fx ¼ a sec bm 1 þ tan bm ð5:16Þ
2 CL
The power delivered to the air is given by:
UF x ¼ mðh03 2 h01 Þ
¼ rCa lsðh03 2 h01 Þ ð5:17Þ
considering the flow through one blade passage of width s.
Therefore,
h03 2 h01
¼
U2
Fx
¼
rCa lsU
 
1 C a  c
¼ sec bm ðC L þ C D tan bm Þ
2 U s
ð5:18Þ
1 c
¼ sec bm ðCL þ CD tan bm Þ
2 s
For a stage in which bm ¼ 458, efficiency will be maximum. Substituting this
back into Eq. (5.18), the optimal blade-loading factor is given by:
w  c
Copt ¼ pffiffiffi ðC L þ C D Þ ð5:19Þ
2 s
For a well-designed blade, CD is much smaller than CL, and therefore the optimal
blade-loading factor is approximated by:
w  c
Copt ¼ pffiffiffi CL ð5:20Þ
2 s

5.6 CASCADE NOMENCLATURE

AND TERMINOLOGY
Studying the 2-D flow through cascades of airfoils facilitates designing highly
efficient axial flow compressors. A cascade is a row of geometrically similar blades
arranged at equal distance from each other and aligned to the flow direction.
Figure 5.7, which is reproduced from Howell’s early paper on cascade theory and
performance, shows the standard nomenclature relating to airfoils in cascade.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 196 Chapter 5

Figure 5.7 Cascade nomenclature.

a1 0 and a2 0 are the camber angles of the entry and exit tangents the camber
0 0
line makes with the axial direction. The blade camber angle u ¼ a1 2 a2 . The
chord c is the length of the perpendicular of the blade profile onto the chord line.
It is approximately equal to the linear distance between the leading edge and the
trailing edge. The stagger angle j is the angle between the chord line and the axial
direction and represents the angle at which the blade is set in the cascade. The
pitch s is the distance in the direction of rotation between corresponding points on
adjacent blades. The incidence
angle i is the difference between the air inlet angle
0  0
(a1) and the blade inlet angle a1 . That is, i ¼ a1 2 a1 . The deviation angle
(d)
0 
is the difference between the air outlet angle (a2) and the blade outlet angle a2 .
The air deflection angle, 1 ¼ a1 2 a2, is the difference between the entry and
exit air angles.
A cross-section of three blades forming part of a typical cascade is shown in
Fig. 5.7. For any particular test, the blade camber angle u, its chord c, and the pitch
0 0
(or space) s will be fixed and the blade inlet and outlet angles a1 and a2 are
determined by the chosen setting or stagger angle j. The angle of incidence, i, is
0
then fixed by the choice of a suitable air inlet angle a1, since i ¼ a1 2 a1 .
An appropriate setting of the turntable on which the cascade is mounted can
accomplish this. With the cascade in this position the pressure and direction
measuring instruments are then traversed along the blade row in the upstream and
downstream position. The results of the traverses are usually presented as shown

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow Compressors and Fans 197

Figure 5.8 Variation of stagnation pressure loss and deflection for cascade at fixed
incidence.

in Fig. 5.8. The stagnation pressure loss is plotted as a dimensionless number

given by:
P01 2 P02
Stagnation pressure loss coefficient ¼ ð5:21Þ
0:5rC21
This shows the variation of loss of stagnation pressure and the air deflection,
1 ¼ a1 2 a2, covering two blades at the center of the cascade. The curves of
Fig. 5.8 can now be repeated for different values of incidence angle, and the whole
set of results condensed to the form shown in Fig. 5.9, in which the mean loss and
mean deflection are plotted against incidence for a cascade of fixed geometrical
form.
The total pressure loss owing to the increase in deflection angle of air is
marked when i is increased beyond a particular value. The stalling incidence of
the cascade is the angle at which the total pressure loss is twice the minimum
cascade pressure loss. Reducing the incidence i generates a negative angle of
incidence at which stalling will occur.
Knowing the limits for air deflection without very high (more than twice
the minimum) total pressure loss is very useful for designers in the design of
efficient compressors. Howell has defined nominal conditions of deflection for

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 198 Chapter 5

Figure 5.9 Cascade mean deflection and pressure loss curves.

a cascade as 80% of its stalling deflection, that is:

1* ¼ 0:81s ð5:22Þ
where 1s is the stalling deflection and 1* is the nominal deflection for the cascade.
Howell and Constant also introduced a relation correlating nominal
deviation d* with pitch chord ratio and the camber of the blade. The relation is
given by:
sn
d* ¼ mu ð5:23Þ
l
For compressor cascade, n ¼ 12, and for the inlet guide vane in front of the
compressor, n ¼ 1. Hence, for a compressor cascade, nominal deviation is
given by: s12
d* ¼ mu ð5:24Þ
l
The approximate value suggested by Constant is 0.26, and Howell suggested a
modified value for m:
 2  *
2a a
m ¼ 0:23 þ0:1 2 ð5:25Þ
l 50
where the maximum camber of the cascade airfoil is at a distance a from the
leading edge and a*
2 is the nominal air outlet angle.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow Compressors and Fans 199

Then,
a*2 ¼ b2 þ d*
s1
¼ b2 þ m u 2
and, l

a*1 2 a*2 ¼ 1*
or:
a*1 ¼ a*2 þ 1*
Also,
i* ¼ a* *
1 2 b1 ¼ a2 þ 1* 2 b1

5.7 3-D CONSIDERATION

So far, all the above discussions were based on the velocity triangle at one
particular radius of the blading. Actually, there is a considerable difference in
the velocity diagram between the blade hub and tip sections, as shown in
Fig. 5.10.
The shape of the velocity triangle will influence the blade geometry, and,
therefore, it is important to consider this in the design. In the case of a compressor
with high hub/tip ratio, there is little variation in blade speed from root to tip. The
shape of the velocity diagram does not change much and, therefore, little
variation in pressure occurs along the length of the blade. The blading is of the
same section at all radii and the performance of the compressor stage is calculated
from the performance of the blading at the mean radial section. The flow along
the compressor is considered to be 2-D. That is, in 2-D flow only whirl and axial
flow velocities exist with no radial velocity component. In an axial flow
compressor in which high hub/tip radius ratio exists on the order of 0.8, 2-D flow
in the compressor annulus is a fairly reasonable assumption. For hub/tip ratios
lower than 0.8, the assumption of two-dimensional flow is no longer valid. Such
compressors, having long blades relative to the mean diameter, have been used in
aircraft applications in which a high mass flow requires a large annulus area but a
small blade tip must be used to keep down the frontal area. Whenever the fluid
has an angular velocity as well as velocity in the direction parallel to the axis of
rotation, it is said to have “vorticity.” The flow through an axial compressor is
vortex flow in nature. The rotating fluid is subjected to a centrifugal force and to
balance this force, a radial pressure gradient is necessary. Let us consider
the pressure forces on a fluid element as shown in Fig. 5.10. Now, resolve

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 200 Chapter 5

Figure 5.10 Variation of velocity diagram along blade.

the forces in the radial direction Fig. 5.11:

 
dP du
du ðP þ dPÞðr þ drÞ 2 Pr du 2 2 P þ dr
2 2
C 2w ð5:26Þ
¼ r dr r du
r
or
 
dP
ðP þ dPÞðr þ drÞ 2 Pr 2 P þ dr ¼ r dr C 2w
2
where: P is the pressure, r, the density, Cw, the whirl velocity, r, the radius.
After simplification, we get the following expression:
1
Pr þ P dr þ r dP þ dP dr 2 Pr þ r dr 2 dP dr ¼ r dr C 2w
2
or:
r dP ¼ r dr C2w

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow Compressors and Fans 201

Figure 5.11 Pressure forces on a fluid element.

That is,
1 dP C 2w
¼ ð5:27Þ
r dr r

The approximation represented by Eq. (5.27) has become known as radial

equilibrium.
The stagnation enthalpy h0 at any radius r where the absolute velocity is C
may be rewritten as:

1 1

h0 ¼ h þ C 2a þ C 2w ; h ¼ cp T; and C 2 ¼ C 2a þ C2w
2 2

Differentiating the above equation w.r.t. r and equating it to zero yields:

 
dh0 g 1 dP 1 dC w
¼ £ þ 0 þ 2C w
dr g 2 1 r dr 2 dr

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 202 Chapter 5

or:
g 1 dP dCw
£ þ Cw ¼0
g 2 1 r dr dr
Combining this with Eq. (5.27):
g C 2w dC w
þ Cw ¼0
g21 r dr
or:
dC w g Cw
¼2
dr g21 r
Separating the variables,
dC w g dr
¼2
Cw g21 r
Integrating the above equation
Z
R dCw g dr
¼2
Cw g21 r
g
2 ln C w r ¼ c where c is a constant:
g21

Taking antilog on both sides,

g
£ Cw £ r ¼ e c
g21
Therefore, we have
C w r ¼ constant ð5:28Þ
Equation (5.28) indicates that the whirl velocity component of the flow varies
inversely with the radius. This is commonly known as free vortex. The outlet
blade angles would therefore be calculated using the free vortex distribution.

5.8 MULTI-STAGE PERFORMANCE

An axial flow compressor consists of a number of stages. If R is the overall
pressure ratio, Rs is the stage pressure ratio, and N is the number of stages, then
the total pressure ratio is given by:
R ¼ ðRs ÞN ð5:29Þ
Equation (5.29) gives only a rough value of R because as the air passes
through the compressor the temperature rises continuously. The equation used to

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow Compressors and Fans 203

find stage pressure is given by:

g
g21
hs DT 0s
Rs ¼ 1 þ ð5:30Þ
T 01
The above equation indicates that the stage pressure ratio depends only on inlet
stagnation temperature T01, which goes on increasing in the successive stages. To
find the value of R, the concept of polytropic or small stage efficiency is very
useful. The polytropic or small stage efficiency of a compressor is given by:
 
g21  n 
h1;c ¼
g n21
or:
 n   
g
¼ hs
n21 g21
where hs ¼ h1,c ¼ small stage efficiency.
The overall pressure ratio is given by:
n
NDT 0s n21
R¼ 1þ ð5:31Þ
T 01
Although Eq. (5.31) is used to find the overall pressure ratio of a
compressor, in actual practice the step-by-step method is used.

5.9 AXIAL FLOW COMPRESSOR

CHARACTERISTICS
The forms of characteristic curves of axial flow compressors are shown in
Fig. 5.12. These curves are quite similar to the centrifugal compressor.
However, axial flow compressors cover a narrower range of mass flow than the
centrifugal compressors, and the surge line is also steeper than that of a
centrifugal compressor. Surging and choking limit the curves at the two ends.
However, the surge points in the axial flow compressors are reached before the
curves reach a maximum value. In practice, the design points is very close to the
surge line. Therefore, the operating range of axial flow compressors is quite
narrow.

Illustrative Example 5.1: In an axial flow compressor air enters the

compressor at stagnation pressure and temperature of 1 bar and 292K,
respectively. The pressure ratio of the compressor is 9.5. If isentropic efficiency
of the compressor is 0.85, find the work of compression and the final temperature
at the outlet. Assume g ¼ 1.4, and Cp ¼ 1.005 kJ/kg K.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 204 Chapter 5

Figure 5.12 Axial flow compressor characteristics.

Solution:
T 01 ¼ 292K; P01 ¼ 1 bar; hc ¼ 0:85:
Using the isentropic P – T relation for compression processes,
g
0 g21
P02 T 02
¼
P01 T 01
where T 020 is the isentropic temperature at the outlet.
Therefore,
g 21
0 P02 g
T 02 ¼ T 01 ¼ 292ð9:5Þ0:286 ¼ 555:92 K
P01
Now, using isentropic efficiency of the compressor in order to find the
actual temperature at the outlet,

0 
T 02 2 T 01 ð555:92 2 292Þ
T 02 ¼ T 01 þ ¼ 292 þ ¼ 602:49 K
hc 0:85

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow Compressors and Fans 205

Work of compression:
W c ¼ Cp ðT 02 2 T 01 Þ ¼ 1:005ð602:49 2 292Þ ¼ 312 kJ/kg

Illustrative Example 5.2: In one stage of an axial flow compressor, the

pressure ratio is to be 1.22 and the air inlet stagnation temperature is 288K. If the
stagnation temperature rise of the stages is 21K, the rotor tip speed is 200 m/s, and
the rotor rotates at 4500 rpm, calculate the stage efficiency and diameter of the
rotor.

Solution:
The stage pressure ratio is given by:
g
g21
hs DT 0s
Rs ¼ 1 þ
T 01
or
3:5
hs ð21Þ
1:22 ¼ 1 þ
288
that is,
hs ¼ 0:8026 or 80:26%
The rotor speed is given by:
pDN ð60Þð200Þ
U¼ ; or D ¼ ¼ 0:85 m
60 pð4500Þ

Illustrative Example 5.3: An axial flow compressor has a tip diameter of

0.95 m and a hub diameter of 0.85 m. The absolute velocity of air makes an angle
of 288 measured from the axial direction and relative velocity angle is 568. The
absolute velocity outlet angle is 568 and the relative velocity outlet angle is 288.
The rotor rotates at 5000 rpm and the density of air is 1.2 kg/m3. Determine:
1. The axial velocity.
2. The mass flow rate.
3. The power required.
4. The flow angles at the hub.
5. The degree of reaction at the hub.
Solution:
1. Rotor speed is given by:
pDN pð0:95Þð5000Þ
U¼ ¼ ¼ 249 m/s
60 60

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 206 Chapter 5

Figure 5.13 Inlet velocity triangle.

Blade speed at the hub:

pDh N pð0:85Þð5000Þ
Uh ¼ ¼ ¼ 223 m/s
60 60
From the inlet velocity triangle (Fig. 5.13),
C w1 ðU 2 C w1 Þ
tan a1 ¼ and tan b1 ¼
Ca Ca
Adding the above two equations:
U
¼ tan a1 þ tan b1
Ca
or:
U ¼ C a ðtan 288 þ tan 568Þ ¼ C a ð2:0146Þ
Therefore, Ca ¼ 123.6 m/s (constant at all radii)
2. The mass flow rate:
m_ ¼ pðr 2t 2 r 2h Þr C a
¼ pð0:4752 2 0:4252 Þð1:2Þð123:6Þ ¼ 20:98 kg/s
3. The power required per unit kg for compression is:
W c ¼ lUC a ðtan b1 2 tan b2 Þ
¼ ð1Þð249Þð123:6Þðtan 568 2 tan 288 Þ1023
¼ ð249Þð123:6Þð1:483 2 0:53Þ
¼ 29:268 kJ/kg

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow Compressors and Fans 207

The total power required to drive the compressor is:

W c ¼ mð29:268Þ
_ ¼ ð20:98Þð29:268Þ ¼ 614 kW

4. At the inlet to the rotor tip:

C w1 ¼ C a tan a1 ¼ 123:6 tan 288 ¼ 65:72 m/s

Using free vortex condition, i.e., Cwr ¼ constant, and using h as the
subscript for the hub,
rt 0:475
Cw1h ¼ C w1t ¼ ð65:72Þ ¼ 73:452 m/s
rh 0:425
At the outlet to the rotor tip,

Cw2t ¼ C a tan a2 ¼ 123:6 tan 568 ¼ 183:24 m/s

Therefore,
rt 0:475
Cw2h ¼ C w2t ¼ ð183:24Þ ¼ 204:8 m/s
rh 0:425
Hence the flow angles at the hub:
C w1h 73:452
tan a1 ¼ ¼ ¼ 0:594 or; a1 ¼ 30:728
Ca 123:6
ðU h Þ 223
tan b1 ¼ 2 tan a1 ¼ 2 0:5942 ¼ 1:21
Ca 123:6
i.e., b1 ¼ 50.438
C w2h 204:8
tan a2 ¼ ¼ ¼ 1:657
Ca 123:6
i.e., a2 ¼ 58.898
ðU h Þ 223
tan b2 ¼ 2 tan a2 ¼ 2 tan 58:598 ¼ 0:1472
Ca 123:6
i.e., b2 ¼ 8.378
5. The degree of reaction at the hub is given by:
Ca 123:6
Lh ¼ ðtan b1 þ tan b2 Þ ¼ ðtan 50:438 þ tan 8:378Þ
2U h ð2Þð223Þ
123:6
¼ ð1:21 þ 0:147Þ ¼ 37:61%
ð2Þð223Þ

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 208 Chapter 5

Illustrative Example 5.4: An axial flow compressor has the following

data:
Blade velocity at root: 140 m/s
Blade velocity at mean radius: 185 m/s
Blade velocity at tip: 240 m/s
Stagnation temperature rise in this stage: 15K
Axial velocity ðconstant from root to tipÞ: 140 m/s
Work done factor: 0:85
Degree of reaction at mean radius: 50%

Calculate the stage air angles at the root, mean, and tip for a free vortex
design.

Solution:
Calculation at mean radius:
From Eq. (5.1), Wc ¼ U(Cw2 2 Cw1) ¼ UKCw
or:
Cp ðT 02 2 T 01 Þ ¼ C p DT 0s ¼ lUDCw
So:
C p DT 0s ð1005Þð15Þ
DC w ¼ ¼ ¼ 95:87 m/s
lU ð0:85Þð185Þ
Since the degree of reaction (Fig. 5.14) at the mean radius is 50%, a1 ¼ b2
and a2 ¼ b1.
From the velocity triangle at the mean,
U ¼ DCw þ 2C w1
or:
U 2 DC w 185 2 95:87
Cw1 ¼ ¼ ¼ 44:57 m/s
2 2
Hence,
C w1 44:57
tan a1 ¼ ¼ ¼ 0:3184
Ca 140
that is,
a1 ¼ 17:668 ¼ b2

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow Compressors and Fans 209

Figure 5.14 Velocity triangle at the mean radius.

and
ðDCw þ Cw1 Þ ð95:87 þ 44:57Þ
tan b1 ¼ ¼ ¼ 1:003
Ca 140
i.e., b1 ¼ 45:098 ¼ a2
Calculation at the blade tip:
Using the free vortex diagram (Fig. 5.15),
ðDC w £ UÞt ¼ ðDCw £ UÞm
Therefore,
ð95:87Þð185Þ
DCw ¼ ¼ 73:9 m/s
240
Whirl velocity component at the tip:
C w1 £ 240 ¼ ð44:57Þð185Þ
Therefore:
ð44:57Þð185Þ
C w1 ¼ ¼ 34:36 m/s
240
Cw1 34:36
tan a1 ¼ ¼ ¼ 0:245
Ca 140

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 210 Chapter 5

Figure 5.15 Velocity triangles at tip.

Therefore,
a1 ¼ 13:798
From the velocity triangle at the tip,
x2 þ DC w þ Cw1 ¼ U
or:
x2 ¼ U 2 DC w 2 Cw1 ¼ 240 2 73:9 2 34:36 ¼ 131:74

DCw þ x2 73:9 þ 131:74

tan b1 ¼ ¼ ¼ 1:469
Ca 140
i.e., b1 ¼ 55.758
ðC w1 þ DCw Þ ð34:36 þ 73:9Þ
tan a2 ¼ ¼ ¼ 0:7733
Ca 140
i.e., a2 ¼ 37.718
x2 131:74
tan b2 ¼ ¼ ¼ 0:941
Ca 140
i.e., b2 ¼ 43.268
Calculation at the blade root:
ðDCw £ UÞr ¼ ðDC w £ UÞm

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow Compressors and Fans 211

or:
DCw £ 140 ¼ ð95:87Þð185Þ and DCw ¼ 126:69 m/s
Also:
ðC w1 £ UÞr ¼ ðC w1 £ UÞm
or:
C w1 £ 140 ¼ ð44:57Þð185Þ and C w1 ¼ 58:9 m/s
and
ðC w2 £ UÞt ¼ ðC w2 £ UÞr
so:

C w2;tip ¼ C a tana2 ¼ 140 tan 37:718 ¼ 108:24 m/s

Therefore:
ð108:24Þð240Þ
C w2;root ¼ ¼ 185:55 m/s
140

58:9
tan a1 ¼ ¼ 0:421
140
i.e., a1 ¼ 22.828
From the velocity triangle at the blade root, (Fig. 5.16)
or:
x2 ¼ Cw2 2 U ¼ 185:55 2 140 ¼ 45:55

Figure 5.16 Velocity triangles at root.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 212 Chapter 5

Therefore:
U 2 Cw1 140 2 58:9
tan b1 ¼ ¼ ¼ 0:579
Ca 140
i.e., b1 ¼ 30.088
C w2 185:55
tan a2 ¼ ¼ ¼ 1:325
Ca 140
i.e., a2 ¼ 52.968
x2 45:55
tan b2 ¼ 2 ¼2 ¼ 20:325
Ca 140
i.e., b2 ¼ 2 188
Design Example 5.5: From the data given in the previous problem,
calculate the degree of reaction at the blade root and tip.

Solution:
Reaction at the blade root:
Ca 140
Lroot ¼ ðtan b1r þ tan b2r Þ ¼ ðtan 30:088 þ tan ð2188 ÞÞ
2U r ð2Þð140Þ
140
¼ ð0:579 2 0:325Þ ¼ 0:127; or 12:7%
ð2Þð140Þ

Reaction at the blade tip:

Ca 140
Ltip ¼ ðtan b1t þ tan b2t Þ ¼ ðtan 55:758 þ tan 43:268 Þ
2U t ð2Þð240Þ
140
¼ ð1:469 þ 0:941Þ ¼ 0:7029; or 70:29%
ð2Þð240Þ

Illustrative Example 5.6: An axial flow compressor stage has the

following data:
Air inlet stagnation temperature: 295K
Blade angle at outlet measured from the axial direction: 328
Flow coefficient: 0:56
Relative inlet Mach number: 0:78
Degree of reaction: 0:5

Find the stagnation temperature rise in the first stage of the compressor.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow Compressors and Fans 213

Solution:
Since the degree of reaction is 50%, the velocity triangle is symmetric as
shown in Fig. 5.17. Using the degree of reaction equation [Eq. (5.12)]:
Ca Ca
L¼ ðtan b1 þ tan b2 Þ and w¼ ¼ 0:56
2U U
Therefore:
2L
tanb1 ¼ 2 tan 328 ¼ 1:16
0:56
i.e., b1 ¼ 49.248
Now, for the relative Mach number at the inlet:
V1
M r1 ¼
12
gRT 1
or:
 
C2
V 21 ¼ gRM 2r1 T 01 2 1
2Cp
From the velocity triangle,
Ca Ca
V1 ¼ ; and C1 ¼
cosb1 cosa1
and:
a1 ¼ b2 ðsince L ¼ 0:5Þ
Therefore:
Ca Ca
C1 ¼ ¼
cos328 0:848

Figure 5.17 Combined velocity triangles for Example 5.6.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 214 Chapter 5

and:

Ca Ca
V1 ¼ ¼
cos 49:248 0:653

Hence:

C 2a C2a
C21 ¼ ; and V 21 ¼
0:719 0:426

Substituting for V1 and C1,

 
C 2a
C2a ¼ 104:41 295 2 ; so : C a ¼ 169:51 m/s
1445

The stagnation temperature rise may be calculated as:

C 2a
T 02 2 T 01 ¼ ðtan b1 2 tan b2 Þ
Cp w

169:512
¼ ðtan 49:248 2 tan 328Þ ¼ 27:31K
ð1005Þð0:56Þ

Design Example 5.7: An axial flow compressor has the following

design data:

Inlet stagnation temperature: 290K

Inlet stagnation pressure: 1 bar
Stage stagnation temperature rise: 24K
Mass flow of air: 22kg/s
Axialvelocity through the stage: 155:5m/s
Rotational speed: 152rev/s
Work done factor: 0:93
Mean blade speed: 205m/s
Reaction at the mean radius: 50%

Determine: (1) the blade and air angles at the mean radius, (2) the mean
radius, and (3) the blade height.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow Compressors and Fans 215

Solution:
(1) The following equation provides the relationship between the
temperature rise and the desired angles:
lUC a
T 02 2 T 01 ¼ ðtan b1 2 tan b2 Þ
Cp
or:
ð0:93Þð205Þð155:5Þ
24 ¼ ðtan b1 2 tan b2 Þ
1005
so:
tan b1 2 tan b2 ¼ 0:814
Using the degree of reaction equation:
Ca
L¼ ðtan b1 þ tan b2 Þ
2U
Hence:
ð0:5Þð2Þð205Þ
tan b1 þ tan b2 ¼ ¼ 1:318
155:5
Solving the above two equations simultaneously for b1 and b2,
2 tan b1 ¼ 2:132;

so : b1 ¼ 46:838 ¼ a2 ðsince the degree of reaction is 50%Þ

and:
tan b2 ¼ 1:318 2 tan 46:838 ¼ 1:318 2 1:066;

so : b2 ¼ 14:148 ¼ a1

(2) The mean radius, rm, is given by:

U 205
rm ¼ ¼ ¼ 0:215m
2pN ð2pÞð152Þ

(3) The blade height, h, is given by:

m ¼ rACa, where A is the annular area of the flow.
Ca 155:5
C1 ¼ ¼ ¼ 160:31 m/s
cosa1 cos14:148

C 21 160:312
T 1 ¼ T 01 2 ¼ 290 2 ¼ 277:21 K
2C p ð2Þð1005Þ

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 216 Chapter 5

Using the isentropic P– T relation:

 g21
g
P1 T1
¼
P01 T 01
Static pressure:
 
277:21 3:5
P1 ¼ ð1Þ ¼ 0:854 bar
290
Then:
P1 ð0:854Þð100Þ
r1 ¼ ¼ ¼ 1:073 kg/m3
RT 1 ð0:287Þð277:21Þ
From the continuity equation:
22
A¼ ¼ 0:132m2
ð1:073Þð155:5Þ
and the blade height:
A 0:132
h¼ ¼ ¼ 0:098m
2pr m ð2pÞð0:215Þ

Illustrative Example 5.8: An axial flow compressor has an overall

pressure ratio of 4.5:1, and a mean blade speed of 245 m/s. Each stage is of 50%
reaction and the relative air angles are the same (308) for each stage. The axial
velocity is 158 m/s and is constant through the stage. If the polytropic efficiency
is 87%, calculate the number of stages required. Assume T01 ¼ 290K.
Solution:
Since the degree of reaction at the mean radius is 50%, a1 ¼ b2 and
a2 ¼ b1. From the velocity triangles, the relative outlet velocity component
in the x-direction is given by:

V x2 ¼ C a tan b2 ¼ 158tan 308 ¼ 91:22 m/s

1
V 1 ¼ C2 ¼ ðU 2 V x2 Þ2 þ C2a 2

1
¼ ð245 2 91:22Þ2 þ 1582 2 ¼ 220:48 m/s
Ca 158
cos b1 ¼ ¼ ¼ 0:7166
V 1 220:48
so: b1 ¼ 44.238

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow Compressors and Fans 217

Stagnation temperature rise in the stage,

UC a
DT 0s ¼ ðtan b1 2 tan b2 Þ
Cp
ð245Þð158Þ
¼ ðtan 44:238 2 tan 308Þ ¼ 15:21K
1005
Number of stages
n
NDT 0s n21
R¼ 1þ
T 01
n g 1:4
¼ h1 ¼ 0:87 ¼ 3:05
n21 g21 0:4
Substituting:
3:05
N15:21
4:5 ¼ 1 þ
290
Therefore,
N ¼ 12 stages:

Design Example 5.9: In an axial flow compressor, air enters at a stagnation

temperature of 290K and 1 bar. The axial velocity of air is 180 m/s (constant
throughout the stage), the absolute velocity at the inlet is 185 m/s, the work done
factor is 0.86, and the degree of reaction is 50%. If the stage efficiency is 0.86,
calculate the air angles at the rotor inlet and outlet and the static temperature at
the inlet of the first stage and stage pressure ratio. Assume a rotor speed of
200 m/s.
Solution:
For 50% degree of reaction at the mean radius (Fig. 5.18), a1 ¼ b2 and
a2 ¼ b1.
From the inlet velocity triangle,

C a 180
cos a1 ¼ ¼ ¼ 0:973
C 1 185

i.e., a1 ¼ 13.358 ¼ b2
From the same velocity triangle,

1
1
C w1 ¼ C21 2 C2a 2 ¼ 1852 2 1802 2 ¼ 42:72 m/s

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 218 Chapter 5

Figure 5.18 Velocity triangles (a) inlet, (b) outlet.

Therefore,
ðU 2 C w1 Þ ð200 2 42:72Þ
tan b1 ¼ ¼ ¼ 0:874
Ca 180
i.e., b1 ¼ 41.158 ¼ a2
Static temperature at stage inlet may be determined by using
stagnation and static temperature relationship as given below:
C1 1852
T 1 ¼ T 01 2 ¼ 290 2 ¼ 273 K
2Cp 2ð1005Þ
Stagnation temperature rise of the stage is given by
lUC a

DT 0s ¼ tanb1 2 tanb2
Cp
0:86ð200Þð180Þ
¼ ð0:874 2 0:237Þ ¼ 19:62 K
1005

Stage pressure ratio is given by

g=g21 3:5
hs DT 0s 0:86 £ 19:62
Rs ¼ 1 þ ¼ 1þ ¼ 1:22
T 01 290

Illustrative Example 5.10: Find the isentropic efficiency of an axial flow

compressor from the following data:

Pressure ratio: 6
Polytropic efficiency: 0:85
Inlet temperature: 285 K

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow Compressors and Fans 219

Solution:
Using the isentropic P – T relation for the compression process,
 g 21
P02 g
T 020 ¼ T 01 ¼ 285ð6Þ0:286 ¼ 475:77 K
P01
Using the polytropic P– T relation for the compression process:
n21 g21 0:4
¼ ¼ ¼ 0:336
n gh1;c 1:4ð0:85Þ
Actual temperature rise:
 ðn21Þ=n
p02
T 02 ¼ T 01 ¼ 285ð6Þ0:336 ¼ 520:36 K
p01
The compressor isentropic efficiency is given by:
T 020 2 T 01 475:77 2 285
hc ¼ ¼ ¼ 0:8105; or 81:05%
T 02 2 T 01 520 2 285

Design Example 5.11: In an axial flow compressor air enters the

compressor at 1 bar and 290K. The first stage of the compressor is designed on
free vortex principles, with no inlet guide vanes. The rotational speed is 5500 rpm
and stagnation temperature rise is 22K. The hub tip ratio is 0.5, the work done
factor is 0.92, and the isentropic efficiency of the stage is 0.90. Assuming an inlet
velocity of 145 m/s, calculate
1. The tip radius and corresponding rotor air angles, if the Mach number
relative to the tip is limited to 0.96.
2. The mass flow at compressor inlet.
3. The stagnation pressure ratio and power required to drive the
compressor.
4. The rotor air angles at the root section.
Solution:
(1) As no inlet guide vanes
a1 ¼ 0; C w1 ¼ 0
Stagnation temperature, T01, is given by
C1
T 01 ¼ T 1 þ
2C p2
or
C1 1452
T 1 ¼ T 01 2 ¼ 290 2 ¼ 288:9K
2Cp 2 £ 1005

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 220 Chapter 5

The Mach number relative to tip is

V1
M ¼ pffiffiffiffiffiffiffiffiffiffiffi
gRT 1
or
V 1 ¼ 0:96ð1:4 £ 287 £ 288:9Þ0:5 ¼ 340:7 m/s
i.e., relative velocity at tip ¼ 340.7 m/s
From velocity triangle at inlet (Fig. 5.3)

0:5
V 21 ¼ U 2t þ C 21 or U t ¼ 340:72 2 1452 ¼ 308:3 m/s
or tip speed,
2pr t N
Ut ¼
60
or
308:3 £ 60
rt ¼ ¼ 0:535m:
2p £ 5500
U t 308:3
tanb1 ¼ ¼ ¼ 2:126
Ca 145

i:e:; b1 ¼ 64:818
Stagnation temperature rise

tUC a

DT 0s ¼ tan b1 2 tan b2
Cp
Substituting the values, we get
0:92 £ 308:3 £ 145

22 ¼ tan b1 2 tan b2
1005
or


tanb1 2 tanb2 ¼ 0:538
(2) Therefore, tan b2 ¼ 1.588 and b2 ¼ 57.88
r m 2 h/2
root radius/tip radius ¼ ¼ 0:5
r m þ h/2
(where subscript m for mean and h for height)
or rm 2 h/2 ¼ 0.5 rm þ 0.25 h
[ rm ¼ 1.5 h
but rt ¼ rm þ h/2 ¼ 1.5 h þ h/2

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow Compressors and Fans 221

or 0.535 ¼ 2 h or h ¼ 0.268 m
and rm ¼ 1.5 h ¼ 0.402 m
Area, A ¼ 2prmh ¼ 2p £ 0.402 £ 0.268 ¼ 0.677 m2
Now, using isentropic relationship for p – T
 g/ðg21Þ  
p1 T1 288:9 3:5
¼ or p1 ¼ 1 £ ¼ 0:987 bar
p01 T 01 290
and

p1 0:987 £ 105
r1 ¼ ¼ ¼ 1:19 kg/m3
RT 1 287 £ 288:9
Therefore, the mass flow entering the stage
m_ ¼ rAC a ¼ 1:19 £ 0:677 £ 145 ¼ 116:8 kg/s

(3) Stage pressure ratio is

g/ðg21Þ
hs DT 0s
Rs ¼ 1þ
T 01
3:5
0:90 £ 22
¼ 1þ ¼ 1:26
290

Now,
W ¼ C p DT 0s ¼ 1005 £ 22 ¼ 22110J/kg
Power required by the compressor
P ¼ mW
_ ¼ 116:8 £ 22110 ¼ 2582:4 kW

(4) In order to find out rotor air angles at the root section, radius at the
root can be found as given below.
rr ¼ r m 2 h/2
¼ 0:402 2 0:268/2 ¼ 0:267m:

Impeller speed at root is

2pr r N
Ur ¼
60
2 £ p £ 0:267 £ 5500
¼ ¼ 153:843 m/s
60

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 222 Chapter 5

Therefore, from velocity triangle at root section

U r 153:843
tanb1 ¼ ¼ ¼ 1:061
Ca 145

i:e:; b1 ¼ 46:98
For b2 at the root section
tU r C a

DT 0s ¼ tanb1 2 tanb2
Cp
or
0:92 £ 153:843 £ 145

22 ¼ tanb1 2 tanb2
1005
or


tanb1 2 tanb2 ¼ 1:078

[ b2 ¼ 20:9748

Design Example 5.12: The following design data apply to an axial flow
compressor:
Overall pressure ratio: 4:5
Mass flow: 3:5kg/s
Polytropic efficiency: 0:87
Stagnation temperature rise per stage: 22k
Absolute velocity approaching the last rotor: 160m/s
Absolute velocity angle; measured from the axial direction: 208
Work done factor: 0:85
Mean diameter of the last stage rotor is: 18:5cm
Ambient pressure: 1:0bar
Ambient temperature: 290K

Calculate the number of stages required, pressure ratio of the first and last
stages, rotational speed, and the length of the last stage rotor blade at inlet to the
stage. Assume equal temperature rise in all stages, and symmetrical velocity
diagram.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow Compressors and Fans 223

Solution:
If N is the number of stages, then overall pressure rise is:
n21
NDT 0s n
R¼ 1þ
T 01
where
n21 g
¼ ha c
n g21
(where hac is the polytropic efficiency)
substituting values
n21 1:4
¼ 0:87 £ ¼ 3:05
n 0:4
overall pressure ratio, R is
3:05
N £ 22
R¼ 1þ
290
or
1 N £ 22
ð4:5Þ3:05 ¼ 1 þ
290

[ N ¼ 8:4
Hence number of stages ¼ 8
Stagnation temperature rise, DT0s, per stage ¼ 22K, as we took 8
stages, therefore
22 £ 8:4
DT 0s ¼ ¼ 23:1
8
From velocity triangle
Ca8
cos a8 ¼
C8
or
C a8 ¼ 160 £ cos20 ¼ 150:35 m/s
Using degree of reaction, L ¼ 0.5
Then,
C a8

0:5 ¼ tanb8 þ tanb9
2U

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 224 Chapter 5

or
150:35

0:5 ¼ tanb8 þ tanb9 ðAÞ
2U
Also,
tUC a8

DT 0s ¼ tanb8 2 tanb9
Cp
Now, DT0s ¼ 22K for one stage.
As we took 8 stages, therefore;
22 £ 8:4
DT 0s ¼ ¼ 23:1 K
8
0:85 £ U £ 150:35

[ 23:1 ¼ tanb8 2 tan20 ðBÞ
1005
Because of symmetry, a8 ¼ b9 ¼ 208
From Eq. (A)


U ¼ 150:35 tanb8 þ 0:364 ðCÞ
From Eq. (B)
181:66
U¼ ðDÞ
tanb8 2 0:364
Comparing Eqs. (C) and (D), we have

 181:66
150:35 tanb8 þ 0:364 ¼

tanb8 2 0:364
or

 181:66
tan 2 b8 2 0:3642 ¼ ¼ 1:21
150:35
or
tan2 b8 ¼ 1:21 þ 0:1325 ¼ 1:342
pffiffiffiffiffiffiffiffiffiffiffi
[ tanb8 ¼ 1:342 ¼ 1:159

i:e:; b8 ¼ 49:208
Substituting in Eq. (C)
U ¼ 150:35ðtan 49:208 þ 0:364Þ
¼ 228:9 m/s

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow Compressors and Fans 225

The rotational speed is given by

228:9
N¼ ¼ 393:69rps
2p £ 0:0925
In order to find the length of the last stage rotor blade at inlet to the stage, it
is necessary to calculate stagnation temperature and pressure ratio of the
last stage.
Stagnation temperature of last stage: Fig. 5.19
T o8 ¼ T 01 þ 7 £ T 0s
¼ 290 þ 7 £ 23:1 ¼ 451:7 K
Pressure ratio of the first stage is:
3:05
1 þ 1 £ 23:1
R¼
451:7
Now,
p08 /p09 ¼ 1:1643
p09
¼ 4; and p09 ¼ 4bar
p01
4
p08 ¼ ¼ 3:44bar
1:1643
and
C28
T 08 ¼ T 8 þ
2Cp
or
C 28
T 8 ¼ T 08 2
2C p
1602
¼ 451:7 2
2 £ 1005
¼ 438:96 K

Figure 5.19 Velocity diagram of last stage.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 226 Chapter 5

Using stagnation and static isentropic temperature relationship for the last
stage, we have
 1:4/0:4
p8 T8
¼
p08 T 08
Therefore,  
438:96 3:5
p8 ¼ 3:44 ¼ 3:112bar
451:7
and p8
r8 ¼
RT 8
3:112 £ 105
¼ ¼ 2:471 kg/m3
287 £ 438:9
Using mass flow rate
m_ ¼ r8 A8 C a8
or
3:5 ¼ 2:471 £ A8 £ 150:35
[ A8 ¼ 0:0094m 2
¼ 2prh
or
0:0094
h¼ ¼ 0:0162m
2p £ 0:0925
i.e., length of the last stage rotor blade at inlet to the stage,
h ¼ 16.17 mm.

Design Example 5.13: A 10-stage axial flow compressor is designed for

stagnation pressure ratio of 4.5:1. The overall isentropic efficiency of the
compressor is 88% and stagnation temperature at inlet is 290K. Assume equal
temperature rise in all stages, and work done factor is 0.87. Determine the air
angles of a stage at the design radius where the blade speed is 218 m/s. Assume a
constant axial velocity of 165 m/s, and the degree of reaction is 76%.
Solution:
No. of stages ¼ 10
The overall stagnation temperature rise is:
 g21 


T 01 R g 2 1 290 4:50:286 2 1
T0 ¼ ¼
hc 0:88
¼ 155:879 K

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow Compressors and Fans 227

The stagnation temperature rise of a stage

155:879
T 0s ¼ ¼ 15:588 K
10
The stagnation temperature rise in terms of air angles is:
tUC a
T 0s ¼ ðtan a2 2 tan a1 Þ
Cp
or
T 0s £ C p 15:588 £ 1005
ðtan a2 2 tan a1 Þ ¼ ¼
tUC a 0:87 £ 218 £ 165
¼ 0:501 ðAÞ

From degree of reaction

Ca
^¼ 12 ðtan a2 þ tan a1 Þ
2U
or
165
0:76 ¼ 1 2 ðtan a2 þ tan a1 Þ
2 £ 218
0:24 £ 2 £ 218
[ ðtan a2 þ tan a1 Þ ¼ ¼ 0:634 ðBÞ
165

Adding (A) and (B), we get

2 tan a2 ¼ 1.135
or tan a2 ¼ 0.5675
i.e., a2 ¼ 29.578
and tan a1 ¼ 0.634 2 0.5675 ¼ 0.0665
i.e., a1 ¼ 3.808
Similarly, for b1 and b2, degree of reaction
tan b1 þ tan b2 ¼ 2.01
and tan b1 2 tan b2 ¼ 0.501
[ 2 tan b1 ¼ 2.511
i.e., b1 ¼ 51.468
and tan b2 ¼ 1.1256 2 0.501 ¼ 0.755
i.e., b2 ¼ 37.038

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 228 Chapter 5

Design Example 5.14: An axial flow compressor has a tip diameter of

0.9 m, hub diameter of 0.42 m, work done factor is 0.93, and runs at 5400 rpm.
Angles of absolute velocities at inlet and exit are 28 and 588, respectively and
velocity diagram is symmetrical. Assume air density of 1.5 kg/m3, calculate mass
flow rate, work absorbed by the compressor, flow angles and degree of reaction at
the hub for a free vortex design.
Solution:
Impeller speed is
2prN 2p £ 0:45 £ 5400
U¼ ¼ ¼ 254:57 m/s
60 60
From velocity triangle


U ¼ C a tan a1 þ tan b1
U 254:57
Ca ¼ ¼ ¼ 119:47 m/s
tan a1 þ tan b1 ðtan 288 þ tan 588Þ
Flow area is
A ¼ p r tip 2 r root
¼ p 0:452 2 0:422 ¼ 0:0833 m2

Mass flow rate is

m_ ¼ rACa ¼ 1:5 £ 0:0833 £ 119:47 ¼ 14:928 kg/s

Power absorbed by the compressor

¼ tU ðC w2 2 Cw1 Þ
¼ tUC a ðtan a2 2 tan a1 Þ
¼ 0:93 £ 254:57 £ 119:47ðtan 588 2 tan 288Þ
¼ 30213:7 Nm

m_ £ 30213:7
Total Power; P ¼ kW
1000
¼ 451 kW
and whirl velocity at impeller tip Cwt ¼ C a tan a 1 ¼ 119.47 £
tan 288 ¼ 63.52 m/s

Now using free vortex condition

r Cw ¼ constant
[ rhCw1h ¼ rtCw1t (where subscripts h for hub and t for tip)

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow Compressors and Fans 229

or
r t C w1t 0:45 £ 63:52
C w1h ¼ ¼ ¼ 71:46 m/s
rh 0:4

Similarly
C w2t ¼ Ca tana2 ¼ 119:47 tan588 ¼ 191:2 m/s
and
r h C w2h ¼ r t C w2t
or
r t C w2t 0:45 £ 191:2
C w2h ¼ ¼ ¼ 215:09 m/s
rh 0:4
Therefore, the flow angles at the hub are
C w1h
tan a1 ¼ ðwhere C a is constantÞ
Ca
71:46
¼ ¼ 0:598
119:47

i.e., a1 ¼ 30.888
U h 2 C a tana1
tanb1 ¼
Ca
where Uh at the hub is given by
2 £ p £ 0:4 £ 5400
U h ¼ 2pr h N ¼ ¼ 226:29 m/s
60

226:29 2 119:47 tan30:888

[ tanb1 ¼
119:47
i.e., b1 ¼ 52.348
Cw2h 215:09
tana2 ¼ ¼ ¼ 1:80
Ca 119:47
i.e., a2 ¼ 60.958
Similarly,
U h 2 C a tana2 226:29 2 119:47 tan 60:958
tanb2 ¼ ¼
Ca 119:47
i.e., b2 ¼ 5.368

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 230 Chapter 5

Finally, the degree of reaction at the hub is

Ca
 119:47
^¼ tanb1 þ tanb2 ¼ ðtan52:348 þ tan5:368Þ
2U h 2 £ 226:29
¼ 0:367 or 36:7%:

Design Example 5.15: An axial flow compressor is to deliver 22 kg of air per

second at a speed of 8000 rpm. The stagnation temperature rise of the first stage is
20 K. The axial velocity is constant at 155 m/s, and work done factor is 0.94. The
mean blade speed is 200 m/s, and reaction at the mean radius is 50%. The rotor
blade aspect ratio is 3, inlet stagnation temperature and pressure are 290 K and
1.0 bar, respectively. Assume Cp for air as 1005 J/kg K and g ¼ 1.4. Determine:
1. The blade and air angles at the mean radius.
2. The mean radius.
3. The blade height.
4. The pitch and chord.
Solution:
1. Using Eq. (5.10) at the mean radius
tUC a

T 02 2 T 01 ¼ tan b1 2 tanb2
Cp

0:94 £ 200 £ 155


20 ¼ tanb1 2 tanb2
1005


tanb1 2 tanb2 ¼ 0:6898
Using Eq. (5.12), the degree of reaction is
Ca

^¼ tanb1 þ tanb2
2U
or

 0:5 £ 2 £ 200
tanb1 þ tanb2 ¼ ¼ 1:29
155
Solving above two equations simultaneously
2 tanb1 ¼ 1:98

[ b1 ¼ 44:718 ¼ a2 ðas the diagram is symmetricalÞ

tanb2 ¼ 1:29 2 tan44:718

i.e.,
b2 ¼ 16:708 ¼ a1

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow Compressors and Fans 231

2. Let rm be the mean radius

U 200 £ 60
rm ¼ ¼ ¼ 0:239m
2pN 2p £ 8000
3. Using continuity equation in order to find the annulus area of flow
Ca 155
C1 ¼ ¼ ¼ 162 m/s
cosa1 cos16:708
C21 1622
T 1 ¼ T 01 2 ¼ 290 2 ¼ 276:94 K
2Cp 2 £ 1005
Using isentropic relationship at inlet
 g21 g
p1 T1
¼
p01 T 01
Static pressure is
 3:5
276:94
p1 ¼ 1:0 ¼ 0:851bars
290
Density is
p1 0:851 £ 100
r1 ¼ ¼ ¼ 1:07 kg/m3
RT 1 0:287 £ 276:94
From the continuity equation,
22
A¼ ¼ 0:133m2
1:07 £ 155
Blade height is
A 0:133
h¼ ¼ ¼ 0:089m:
2pr m 2 £ p £ 0:239
4. At mean radius, and noting that blades b, an equivalent to cascade, a,
nominal air deflection is
1 ¼ b1 2 b2
¼ 44:718 2 16:708 ¼ 28:018
Using Fig. 5.20 for cascade nominal deflection vs. air outlet angle, the
solidity,
s
¼ 0:5
c
span
Blade aspect ratio ¼
chord

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 232 Chapter 5

Figure 5.20 Cascade nominal deflection versus air outlet angle.

Blade chord,
0:089
C¼ ¼ 0:03m
3
Blade pitch,
s ¼ 0:5 £ 0:03 ¼ 0:015 m:

PROBLEMS
5.1 An axial flow compressor has constant axial velocity throughout the
compressor of 152 m/s, a mean blade speed of 162 m/s, and delivers 10.5 kg
of air per second at a speed of 10,500 rpm. Each stage is of 50% reaction
and the work done factor is 0.92. If the static temperature and pressure at the
inlet to the first stage are 288K and 1 bar, respectively, and the stagnation
stage temperature rise is 15K, calculate: 1 the mean diameter of the blade
row, (2) the blade height, (3) the air exit angle from the rotating blades, and
(4) the stagnation pressure ratio of the stage with stage efficiency 0.84.
(0.295 m, 0.062 m, 11.378, 1.15)

5.2 The following design data apply to an axial flow compressor:

Stagnation temperature rise of the stage: 20 K
Work done factor: 0:90
Blade velocity at root: 155 m/s

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow Compressors and Fans 233

Blade velocity at mean radius: 208 m/s

Blade velocity at tip: 255 m/s
Axial velocity ðconstant through the stageÞ: 155 m/s
Degree of reaction at mean radius: 0:5
Calculate the inlet and outlet air and blade angles at the root, mean radius
and tip for a free vortex design.
(188, 45.58, 14.848, 54.078, 39.718, 39.188, 23.568, 29.428, 53.758, 2 208)
5.3 Calculate the degree ofreaction at the tip and root for the same data as Prob. 5.2.
(66.7%, 10%)

5.4 Calculate the air and blade angles at the root, mean and tip for 50% degree
of reaction at all radii for the same data as in Prob. [5.2].
(47.868, 28.378, 43.988, 1.728)

5.5 Show that for vortex flow,

C w £ r ¼ constant
that is, the whirl velocity component of the flow varies inversely with the
radius.
5.6 The inlet and outlet angles of an axial flow compressor rotor are 50 and 158,
respectively. The blades are symmetrical; mean blade speed and axial
velocity remain constant throughout the compressor. If the mean blade
speed is 200 m/s, work done factor is 0.86, pressure ratio is 4, inlet
stagnation temperature is equal to 290 K, and polytropic efficiency of the
compressor is 0.85, find the number of stages required.
(8 stages)

5.7 In an axial flow compressor air enters at 1 bar and 158C. It is compressed
through a pressure ratio of four. Find the actual work of compression and
temperature at the outlet from the compressor. Take the isentropic
efficiency of the compressor to be equal to 0.84
. (167.66 kJ/kg, 454.83 K)

5.8 Determine the number of stages required to drive the compressor for an
axial flow compressor having the following data: difference between the
tangents of the angles at outlet and inlet, i.e., tan b1 - tan b2 ¼ 0.55. The
isentropic efficiency of the stage is 0.84, the stagnation temperature at
the compressor inlet is 288K, stagnation pressure at compressor inlet is
1 bar, the overall stagnation pressure rise is 3.5 bar, and the mass flow rate is
15 kg/s. Assume Cp ¼ 1.005 kJ/kg K, g ¼ 1.4, l ¼ 0.86, hm ¼ 0.99
(10 stages, 287.5 kW)

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 234 Chapter 5

5.9 From the data given below, calculate the power required to drive the
compressor and stage air angles for an axial flow compressor.
Stagnation temperature at the inlet: 288 K
Overall pressure ratio: 4
Isentropic efficiency of the compressor: 0:88
Mean blade speed: 170 m/s
Axial velocity: 120 m/s
Degree of reaction: 0:5

(639.4 kW, b1 ¼ 77.88, b2 ¼ 2 72.698)

5.10 Calculate the number of stages from the data given below for an axial flow
compressor:
Air stagnation temperature at the inlet: 288 K
Stage isentropic efficiency: 0:85
Degree of reaction: 0:5
Air angles at rotor inlet: 408
Air angle at the rotor outlet: 108
Meanblade speed: 180 m/s
Work done factor: 0:85
Overall pressure ratio: 6
(14 stages)
5.11 Derive the expression for polytropic efficiency of an axial flow
compressor in terms of:
(a) n and g
(b) inlet and exit stagnation temperatures and pressures.
5.12 Sketch the velocity diagrams for an axial flow compressor and derive the
expression:
g
g21
P02 hs DT 0s
¼ 1þ
P01 T 01
5.13 Explain the term “degree of reaction”. Why is the degree of reaction
generally kept at 50%?
5.14 Derive an expression for the degree of reaction and show that for 50%
reaction, the blades are symmetrical; i.e., a1 ¼ b2 and a2 ¼ b1.
5.16 What is vortex theory? Derive an expression for vortex flow.
5.17 What is an airfoil? Define, with a sketch, the various terms used in airfoil
geometry.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow Compressors and Fans 235

NOTATION
C absolute velocity
CL lift coefficient
Cp specific heat at constant pressure
D drag
Fx tangential force on moving blade
h blade height, specific enthalpy
L lift
N number of stage, rpm
n number of blades
R overall pressure ratio, gas constant
Rs stage pressure ratio
U tangential velocity
V relative velocity
a angle with absolute velocity, measured from the axial direction
a*2 nominal air outlet angle
b angle with relative velocity, measure from the axial direction
DTA static temperature rise in the rotor
DTB static temperature rise in the stator
DT0s stagnation temperature rise
DTs static temperature rise
D* nominal deviation
e* nominal deflection
es stalling deflection
w flow coefficient
L degree of reaction
l work done factor
c stage loading factor

SUFFIXES
1 inlet to rotor
2 outlet from the rotor
3 inlet to second stage
a axial, ambient
m mean
r radial, root
t tip
w whirl

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 6
Steam Turbines

6.1 INTRODUCTION
In a steam turbine, high-pressure steam from the boiler expands in a set of
stationary blades or vanes (or nozzles). The high-velocity steam from the nozzles
strikes the set of moving blades (or buckets). Here the kinetic energy of the steam
is utilized to produce work on the turbine rotor. Low-pressure steam then
exhausts to the condenser. There are two classical types of turbine stage designs:
the impulse stage and the reaction stage.
Steam turbines can be noncondensing or condensing. In noncondensing
turbines (or backpressure turbines), steam exhausts at a pressure greater than
atmospheric. Steam then leaves the turbine and is utilized in other parts of the
plant that use the heat of the steam for other processes. The backpressure turbines
have very high efficiencies (range from 67% to 75%). A multi-stage condensing
turbine is a turbine in which steam exhausts to a condenser and is condensed by
air-cooled condensers. The exhaust pressure from the turbine is less than the
atmospheric. In this turbine, cycle efficiency is low because a large part of the
steam energy is lost in the condenser.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 238 Chapter 6

6.2 STEAM NOZZLES

The pressure and volume are related by the simple expression, PVg ¼ constant,
for a perfect gas. Steam deviates from the laws of perfect gases. The P-V
relationship is given by:
PV n ¼ constant
where:
n ¼ 1:135 for saturated steam
n ¼ 1:3 for superheated steam

For wet steam, the Zeuner relation,

 x
n ¼ 1:035 þ
10
(where x is the initial dryness fraction of the steam) may be used.
All nozzles consist of an inlet section, a throat, and an exit. The velocity
through a nozzle is a function of the pressure-differential across the nozzle.
Consider a nozzle as shown in Fig. 6.1.
Assume that the flow occurs adiabatically under steady conditions. Since
no work is transferred, the velocity of the fluid at the nozzle entry is usually very
small and its kinetic energy is negligible compared with that at the outlet. Hence,
the equation reduces to:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C 2 ¼ f 2ð h1 2 h2 Þ g ð6:1Þ
where h1 and h2 are the enthalpies at the inlet and outlet of the nozzle,
respectively. As the outlet pressure decreases, the velocity increases.
Eventually, a point is reached called the critical pressure ratio, where the
velocity is equal to the velocity of sound in steam. Any further reduction in
pressure will not produce any further increases in the velocity. The temperature,
pressure, and density are called critical temperature, critical pressure, and critical

Figure 6.1 Nozzle.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Steam Turbines 239

density, respectively. The ratio between nozzle inlet temperature and critical
temperature is given by:
T1 2
¼ ð6:2Þ
Tc n þ 1
where Tc is the critical temperature at which section M ¼ 1. Assuming
isentropic flow in the nozzle, the critical pressure ratio is:
 n21
n
P1 T1
¼ 0 ð6:3Þ
Pc Tc
where Tc0 is the temperature, which would have been reached after an isentropic
expansion in the nozzle. The critical pressure ratio is approximately 0.55 for
superheated steam. When the outlet pressure is designed to be higher than the
critical pressure, a simple convergent nozzle may be used. In a convergent nozzle,
shown in Fig. 6.2, the outlet cross-sectional area and the throat cross-sectional
areas are equal. The operation of a convergent nozzle is not practical in high-
pressure applications. In this case, steam tends to expand in all directions and is
very turbulent. This will cause increased friction losses as the steam flows through
the moving blades. To allow the steam to expand without turbulence, the
convergent– divergent nozzle is used. In this type of nozzle, the area of the section
from the throat to the exit gradually increases, as shown in Fig. 6.1.
The increase in area causes the steam to emerge in a uniform steady flow.
The size of the throat and the length of the divergent section of every nozzle must
be specifically designed for the pressure ratio for which the nozzle will be used.
If a nozzle is designed to operate so that it is just choked, any other operating
condition is an off-design condition. In this respect, the behavior of convergent
and convergent –divergent nozzles is different. The temperature at the throat,
i.e., the critical temperature, can be found from steam tables at the value of Pc and
sc ¼ s1. The critical velocity is given by the equation:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C c ¼ f 2ð h1 2 hc Þ g ð6:4Þ
where hc is read from tables or the h – s chart at Pc and sc.

Figure 6.2 Convergent nozzle.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 240 Chapter 6

6.3 NOZZLE EFFICIENCY

The expansion process is irreversible due to friction between the fluid and walls
of the nozzle, and friction within the fluid itself. However, it is still approximately
adiabatic as shown in Fig. 6.3.
1 –20 is the isentropic enthalpy drop and 1 – 2 is the actual enthalpy drop in
the nozzle. Then the nozzle efficiency is defined as
h1 2 h2
hn ¼
h1 2 h2 0

6.4 THE REHEAT FACTOR

Consider a multi-stage turbine as shown by the Mollier diagram, Fig. 6.4.
The reheat factor is defined by:
Cumulative stage isentropic enthalpy drop
R:F: ¼
Turbine isentropic enthalpy drop
P
Dh0 stage
¼
½Dh0 turbine
 0
  0
  0

h1 2h2 þ h2 2h3 þ h3 2h4
¼
 ð6:5Þ
h1 2h"4

Figure 6.3 Nozzle expansion process for a vapor.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Steam Turbines 241

Figure 6.4 Mollier chart for a multi-stage turbine.

Since the isobars diverge, R.F. . 1.

The reheat factor may be used to relate the stage efficiency and the turbine
efficiency.
Turbine isentropic efficiency is given by:
Dh
ht ¼ ð6:6Þ
Dh0
where Dh is the actual enthalpy drop and Dh0 is the isentropic enthalpy drop.
From diagram 6.4 it is clear that:
P
Dh ¼ ½Dhstage
Dh1 – 4 ¼ ðh1 2 h2 Þ þ ðh2 2 h3 Þ þ ðh3 2 h4 Þ
if hs (stage efficiency) is constant, then:
P P
hs Dh0 stage hs Dh0 stage
ht ¼ ¼
½Dh0 turbine ½Dh0 turbine
or ht ¼ hs £ ðR:FÞ: ð6:7Þ

Equation 6.7 indicates that the turbine efficiency is greater than the stage
efficiency. The reheat factor is usually of the order of 1.03 –1.04.

6.5 METASTABLE EQUILIBRIUM

As shown in Fig. 6.5, slightly superheated steam at point 1 is expanded in a
convergent– divergent nozzle. Assume reversible and adiabatic processes. 1 – 2 is

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 242 Chapter 6

Figure 6.5 Phenomenon of supersaturation on T –S diagram.

the path followed by the steam and point 2 is on the saturated vapor line. Here, we
can expect condensation to occur. But, if point 2 is reached in the divergent
section of the nozzle, then condensation could not occur until point 3 is reached.
At this point, condensation occurs very rapidly. Although the steam between
points 2 –3 is in the vapor state, the temperature is below the saturation
temperature for the given pressure. This is known as the metastable state. In fact,
the change of temperature and pressure in the nozzle is faster than the
condensation process under such conditions. The condensation lags behind the
expansion process. Steam does not condense at the saturation temperature
corresponding to the pressure. Degree of undercooling is the difference between
the saturation temperature corresponding to pressure at point 3 and the actual
temperature of the superheated vapor at point 3. Degree of supersaturation is the
actual pressure at point 3 divided by the saturation pressure corresponding to the
actual temperature of the superheated vapor at point 3.
Illustrative Example 6.1: Dry saturated steam at 2 MPa enters a steam
nozzle and leaves at 0.2 MPa. Find the exit velocity of the steam and dryness
fraction. Assume isentropic expansion and neglect inlet velocity.
Solution:
From saturated steam tables, enthalpy of saturated vapor at 2 MPa:
h1 ¼ hg ¼ 2799:5 kJ/kg and entropy s1 ¼ sg ¼ 6:3409 kJ/kg K

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Steam Turbines 243

Since the expansion is isentropic, s1 ¼ s2: i.e., s1 ¼ s2 ¼ 6.3409 ¼ sf2 þ

x2sfg2, where x2 is the dryness fraction after isentropic expansion, sf2 is the
entropy of saturated liquid at 0.2 MPa, sfg2 is the entropy of vaporization at
0.2 MPa. Using tables:

6:3409 2 1:5301
x2 ¼ ¼ 0:8595
5:5970
Therefore,
h2 ¼ hf2 þ x2hfg2 ¼ 504.7 þ 0.8595 £ 2201.9 ¼ 2397.233 kJ/kg
Using the energy equation:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C 2 ¼ f2ðh1 2 h2 Þg
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
¼ fð2Þ £ ð1000Þ £ ð2799:5 2 2397:233Þg
or: C 2 ¼ 897 m/s

Illustrative Example 6.2: Dry saturated steam is expanded in a nozzle

from 1.3 MPa to 0.1 MPa. Assume friction loss in the nozzle is equal to 10% of
the total enthalpy drop; calculate the mass of steam discharged when the nozzle
exit diameter is 10 mm.
Solution:
Enthalpy of dry saturated steam at 1.3 MPa, using steam tables,
h1 ¼ 2787:6 kJ/kg; and entropy s1 ¼ 6:4953 kJ/kg K:
Since the expansion process is isentropic, s1 ¼ s2 ¼ sf2 þ x2 sfg2 , hence
dryness fraction after expansion:
6:4953 2 1:3026
x2 ¼ ¼ 0:857
6:0568
Now, the enthalpy at the exit:
h2 ¼ hf2 þ x2 hfg2 ¼ 417:46 þ ð0:857Þ £ ð2258Þ

¼ 2352:566 kJ/kg
Therefore enthalpy drop from 1.3 MPa to 0.1 MPa
¼ h1 – h2 ¼ 2787:6 – 2352:566 ¼ 435:034 kJ/kg
Actual enthalpy drop due to friction loss in the nozzle
¼ 0:90 £ 435:034 ¼ 391:531 kJ/kg
Hence, the velocity of steam at the nozzle exit:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C 2 ¼ fð2Þ £ ð1000Þ £ ð391:531Þg ¼ 884:908 m/s

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 244 Chapter 6

Specific volume of steam at 0.1 MPa:

¼ x2 vg2 ¼ ð0:857Þ £ ð1:694Þ ¼ 1:4517 m3 =kg
(since the volume of the liquid is usually negligible compared to the
volume of dry saturated vapor, hence for most practical problems, v ¼ xvg)
Mass flow rate of steam at the nozzle exit:
AC 2 ðpÞ £ ð0:01Þ2 £ ð884:908Þ £ ð3600Þ
¼ ¼ ¼ 172:42 kg=h:
x2 vg2 ð4Þ £ ð1:4517Þ

Illustrative Example 6.3: Steam at 7.5 MPa and 5008C expands through
an ideal nozzle to a pressure of 5 MPa. What exit area is required to accommodate
a flow of 2.8 kg/s? Neglect initial velocity of steam and assume isentropic
expansion in the nozzle.
Solution:
Initial conditions:
P1 ¼ 7.5 MPa, 5008C
h1 ¼ 3404.3 kJ/kg
s1 ¼ 6.7598 kJ/kg K
(h1 and s1 from superheated steam tables)
At the exit state, P2 . Pc ¼ ð0:545Þ £ ð7:5Þ ¼ 4:0875 MPa; and
therefore the nozzle is convergent. State 2 is fixed by P2 ¼ 5 MPa, s1 ¼
s2 ¼ 6.7598 kJ/kg K
T2 ¼ 4358K, v2 ¼ 0.06152 m3/kg, h2 ¼ 3277.9 kJ/kg (from the super-
heated steam tables or the Mollier Chart).
The exit velocity:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C 2 ¼ fð2Þ £ ð1000Þ £ ðh1 2 h2 Þg
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
¼ fð2Þ £ ð1000Þ £ ð3404:3 2 3277:9Þg ¼ 502:8 m/s

Using the continuity equation, the exit area is

mv2 ð2:8Þ £ ð0:06152Þ
A2 ¼ ¼ ¼ ð3:42Þ £ ð1024 Þ m2
C2 502:8

Illustrative Example 6.4: Consider a convergent –divergent nozzle in

which steam enters at 0.8 MPa and leaves the nozzle at 0.15 MPa. Assuming
isentropic expansion and index n ¼ 1.135, find the ratio of cross-sectional area,
the area at the exit, and the area at the throat for choked conditions (i. e. , for
maximum mass flow).

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Steam Turbines 245

Solution:
Critical pressure for maximum mass flow is given by Fig. 6.6:
 n21
n  8:41
2 2
P c ¼ P2 ¼ P 1 ¼ 0:8 ¼ 0:462 MPa
nþ1 2:135

From the Mollier chart:

h1 ¼ 2769 kJ/kg
h2 ¼ 2659 kJ/kg
h3 ¼ 2452 kJ/kg
Enthalpy drop from 0.8 MPa to 0.15 MPa:

Dh123 ¼ h1 2 h3 ¼ 2769 2 2452 ¼ 317 kJ/kg

Enthalpy drop from 0.8 MPa to 0.462 MPa:

Dh1 – 2 ¼ h1 2 h2 ¼ 2769 2 2659 ¼ 110 kJ/kg

Dryness fraction: x2 ¼ 0.954

Dryness fraction: x3 ¼ 0.902
The velocity at the exit,
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C3 ¼ fð2Þ £ ð1000Þ £ ðDh1 – 3 Þg
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
¼ fð2Þ £ ð1000Þ £ ð317Þg ¼ 796m/s

The velocity at the throat

pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C2 ¼ fð2Þ £ ð1000Þ £ ðDh1 – 2 Þg ¼ fð2Þ £ ð1000Þ £ ð110Þg

¼ 469m/s

Figure 6.6 Convergent – divergent nozzle.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 246 Chapter 6

Mass discharged at the throat:

A2 C 2
m_ 2 ¼
x2 vg2

Mass discharged at the exit

A3 C 3
_3 ¼
m
x3 vg3
Therefore
A3 C 3 A 2 C 2
¼
x3 vg3 x2 vg2
Hence,
A3 C2 x3 vg3 469 ð0:902Þð1:1593Þ
¼ ¼ ¼ 1:599
A2 C3 x2 vg2 796 ð0:954Þð0:4038Þ

Illustrative Example 6.5: Dry saturated steam enters the convergent –

divergent nozzle and leaves the nozzle at 0.1 MPa; the dryness fraction at the exit
is 0.85. Find the supply pressure of steam. Assume isentropic expansion
(see Fig. 6.7).

Figure 6.7 h – s diagram for Example 6.5.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Steam Turbines 247

Solution:
At the state point 2, the dryness fraction is 0.85 and the pressure is 0.1 MPa.
This problem can be solved easily by the Mollier chart or by calculations.
Enthalpy and entropy may be determined using the following equations:
h2 ¼ hf2 þ x2 hfg2 and s2 ¼ sf2 þ x2 sfg2 ;
i.e.: h2 ¼ 417.46 þ (0.85) £ (2258) ¼ 2336.76 kJ/kg
and s2 ¼ 1.3026 þ (0.85) £ (6.0568) ¼ 6.451 kJ/kg K
Since s1 ¼ s2, the state 1 is fixed by s1 ¼ 6.451 kJ/kg K, and point 1 is at the
dry saturated line. Therefore pressure P1 may be determined by the Mollier
chart or by calculations: i.e.: P1 ¼ 1.474 MPa.

6.6 STAGE DESIGN

A turbine stage is defined as a set of stationary blades (or nozzles) followed by a
set of moving blades (or buckets or rotor). Together, the two sets of blades allow
the steam to perform work on the turbine rotor. This work is then transmitted to
the driven load by the shaft on which the rotor assembly is carried. Two turbine
stage designs in use are: the impulse stage and reaction stage. The first turbine,
designated by DeLaval in 1889, was a single-stage impulse turbine, which ran at
30,000 rpm. Because of its high speed, this type of turbine has very limited
applications in practice. High speeds are extremely undesirable due to high blade
tip stresses and large losses due to disc friction, which cannot be avoided. In large
power plants, the single-stage impulse turbine is ruled out, since alternators
usually run speeds around 3000 rpm. Photographs of actual steam turbines are
reproduced in Figs. 6.8 – 6.10.

Figure 6.8 Steam turbine.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 248 Chapter 6

Figure 6.9 Pressure velocity-compounded impulse turbine.

6.7 IMPULSE STAGE

In the impulse stage, the total pressure drop occurs across the stationary blades
(or nozzles). This pressure drop increases the velocity of the steam. However, in
the reaction stage, the total pressure drop is divided equally across the stationary
blades and the moving blades. The pressure drop again results in a corresponding
increase in the velocity of the steam flow.
As shown in Figs. 6.10 and 6.11, the shape of the stationary blades or
nozzles in both stage designs is very similar. However, a big difference exists in
the shapes of the moving blades. In an impulse stage, the shape of the moving
blades or buckets is like a cup. The shape of the moving blades in a reaction
stage is more like that of an airfoil. These blades look very similar to the
stationary blades or nozzles.

6.8 THE IMPULSE STEAM TURBINE

Most of the steam turbine plants use impulse steam turbines, whereas gas turbine
plants seldom do. The general principles are the same whether steam or gas is the
working substance.
As shown in Fig. 6.12, the steam supplied to a single-wheel impulse turbine
expands completely in the nozzles and leaves with absolute velocity C1 at an
angle a1, and by subtracting the blade velocity vector U, the relative velocity
vector at entry to the rotor V1 can be determined. The relative velocity V1 makes
an angle of b1 with respect to U. The increase in value of a1 decreases the value
of the useful component, C1 cos a1 and increases the value of the axial or flow
component Ca sin a1. The two points of particular interest are the inlet and exit of
the blades. As shown in Fig. 6.12, these velocities are V1 and V2, respectively.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Steam Turbines
Figure 6.10 Steam turbine cross-sectional view.

249
Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved
 250 Chapter 6

Figure 6.11 Impulse and reaction stage design.

Vectorially subtracting the blade speed results in absolute velocity C2. The steam
leaves tangentially at an angle b2 with relative velocity V2. Since the two velocity
triangles have the same common side U, these triangles can be combined to give a
single diagram as shown in Fig. 6.13.

Figure 6.12 Velocity triangles for turbine stage.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Steam Turbines 251

Figure 6.13 Combined velocity diagram.

If the blade is symmetrical then b1 ¼ b2 and neglecting the friction effects

of blades on the steam, V1 ¼ V2. In the actual case, the relative velocity is
reduced by friction and expressed by a blade velocity coefficient k. That is:
V2
k¼
V1
From Euler’s equation, work done by the steam is given by:
W t ¼ UðC w1 þ Cw2 Þ
Since Cw2 is in the negative r direction, the work done per unit mass flow is
given by:
W t ¼ UðC w1 þ Cw2 Þ ð6:9Þ
If Ca1 – Ca2, there will be an axial thrust in the flow direction. Assume that Ca is
constant. Then:
W t ¼ UC a ðtan a1 þ tan a2 Þ ð6:10Þ

W t ¼ UC a ðtan b1 þ tan b2 Þ ð6:11Þ

Equation (6.11) is often referred to as the diagram work per unit mass flow and
hence the diagram efficiency is defined as:
Diagram work done per unit mass flow
hd ¼ ð6:12Þ
Work available per unit mass flow
Referring to the combined diagram of Fig. 6.13: DCw is the change in the velocity
of whirl. Therefore:
The driving force on the wheel ¼ mC
_
w ð6:13Þ

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 252 Chapter 6

The product of the driving force and the blade velocity gives the rate at which
work is done on the wheel. From Eq. (6.13):
Power output ¼ mUDC
_
w ð6:14Þ
If Ca1 2 Ca2 ¼ DCa, the axial thrust is given by:
Axial thrust : F a ¼ mDC
_
a ð6:15Þ
The maximum velocity of the steam striking the blades
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C 1 ¼ f2ðh0 2 h1 Þg ð6:16Þ
where h0 is the enthalpy at the entry to the nozzle and h1 is the enthalpy at the
nozzle exit, neglecting the velocity at the inlet to the nozzle. The energy supplied
to the blades is the kinetic energy of the jet, C21 =2 and the blading efficiency or
diagram efficiency:
Rate of work performed per unit mass flow
hd ¼
Energy supplied per unit mass of steam
2 2UDC w
hd ¼ ðUDC w Þ £ ¼ ð6:17Þ
C 21 C 21
 
Using the blade velocity coefficient k ¼ VV 21 and symmetrical blades
(i.e., b1 ¼ b2), then:
DC w ¼ 2V 1 cos a1 2 U
Hence
DC w ¼ 2ðC 1 cos a1 2 U Þ ð6:18Þ
And the rate of work performed per unit mass ¼ 2(C1 cos a1 2 U )U
Therefore:
2
hd ¼ 2ðC1 cos a1 2 U ÞU £ 2
C1
4ðC1 cos a1 2 U ÞU
hd ¼
C21
 
4U U
hd ¼ cos a1 2 ð6:19Þ
C1 C1
U
where is called the blade speed ratio.
C1
Differentiating Eq. (6.19) and equating it to zero provides the maximum
diagram efficiency:


d hd 8U
  ¼ 4 cos a1 2 ¼0
d CU1 C1

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Steam Turbines 253

or
U cos a1
¼ ð6:20Þ
C1 2
i.e., maximum diagram efficiency
4 cos a1  cos a1 
¼ cos a1 2
2 2
or:
hd ¼ cos2 a1 ð6:21Þ
Substituting this value into Eq. (6.14), the power output per unit mass flow rate at
the maximum diagram efficiency:
P ¼ 2U 2 ð6:22Þ

6.9 PRESSURE COMPOUNDING (THE RATEAU

TURBINE)
A Rateau-stage impulse turbine uses one row of nozzles and one row of moving
blades mounted on a wheel or rotor, as shown in Fig. 6.14. The total pressure drop
is divided in a series of small increments over the stages. In each stage, which
consists of a nozzle and a moving blade, the steam is expanded and the kinetic
energy is used in moving the rotor and useful work is obtained.
The separating walls, which carry the nozzles, are known as diaphragms.
Each diaphragm and the disc onto which the diaphragm discharges its steam is
known as a stage of the turbine, and the combination of stages forms a pressure
compounded turbine. Rateau-stage turbines are unable to extract a large
amount of energy from the steam and, therefore, have a low efficiency. Although
the Rateau turbine is inefficient, its simplicity of design and construction makes it
well suited for small auxiliary turbines.

6.10 VELOCITY COMPOUNDING (THE CURTIS

TURBINE)
In this type of turbine, the whole of the pressure drop occurs in a single nozzle,
and the steam passes through a series of blades attached to a single wheel or rotor.
The Curtis stage impulse turbine is shown in Fig. 6.15.
Fixed blades between the rows of moving blades redirect the steam flow
into the next row of moving blades. Because the reduction of velocity occurs over
two stages for the same pressure decreases, a Curtis-stage turbine can extract

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 254 Chapter 6

Figure 6.14 Rateau-stage impulse turbine.

more energy from the steam than a Rateau-stage turbine. As a result, a Curtis-
stage turbine has a higher efficiency than a Rateau-stage turbine.

6.11 AXIAL FLOW STEAM TURBINES

Sir Charles Parsons invented the reaction steam turbine. The reaction turbine
stage consists of a fixed row of blades and an equal number of moving blades
fixed on a wheel. In this turbine pressure drop or expansion takes place both in the
fixed blades (or nozzles) as well as in the moving blades. Because the pressure
drop from inlet to exhaust is divided into many steps through use of alternate
rows of fixed and moving blades, reaction turbines that have more than one stage
are classified as pressure-compounded turbines. In a reaction turbine, a reactive
force is produced on the moving blades when the steam increases in velocity and
when the steam changes direction. Reaction turbines are normally used as

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Steam Turbines 255

Figure 6.15 The Curtis-stage impulse turbine.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 256 Chapter 6

Figure 6.16 Velocity triangles for 50% reaction design.

low-pressure turbines. High-pressure reaction turbines are very costly because

they must be constructed from heavy and expensive materials. For a 50%
reaction, the fixed and moving blades have the same shape and, therefore, the
velocity diagram is symmetrical as shown in Fig. 6.16.

6.12 DEGREE OF REACTION

The degree of reaction or reaction ratio (L) is a parameter that describes the
relation between the energy transfer due to static pressure change and the energy
transfer due to dynamic pressure change. The degree of reaction is defined as the
ratio of the static pressure drop in the rotor to the static pressure drop in the stage.
It is also defined as the ratio of the static enthalpy drop in the rotor to the static
enthalpy drop in the stage. If h0, h1, and h2 are the enthalpies at the inlet due to the
fixed blades, at the entry to the moving blades and at the exit from the moving
blades, respectively, then:
h1 2 h2
L¼ ð6:23Þ
h0 2 h2
The static enthalpy at the inlet to the fixed blades in terms of stagnation enthalpy
and velocity at the inlet to the fixed blades is given by
C2
h0 ¼ h00 2 0
2Cp
Similarly,
C2
h2 ¼ h02 2 2
2Cp

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Steam Turbines 257

Substituting,
ð h1 2 h2 Þ
L¼  
C20

C2
h00 2 2C p2 h02 2 2C2p
But for a normal stage, C0 ¼ C2 and since h00 ¼ h01 in the nozzle, then:
h1 2 h2
L¼ ð6:24Þ
h01 2 h02
We know that h01Re1 ¼ h02Re2. Then:

2 
V 2 V 22
h01Re1 2 h02Re2 ¼ ðh1 2 h2 Þ þ 1 ¼0
2
Substituting for (h1 2 h2) in Eq. (6.24):

2 
V 2 2 V 21
L¼
½2ðh01 2 h02 Þ

2 
V 2 2 V 21
L¼ ð6:25Þ
½2U ðC w1 þ Cw2 Þ
Assuming the axial velocity is constant through the stage, then:

2 
V w2 2 V 2w1
L¼
½2U ðU þ V w1 þ V w2 2 U Þ
ðV w2 2 V w1 ÞðV w2 þ V w1 Þ
L¼
½2U ðV w1 þ V w2 Þ


Ca tanb2 2tanb1
L¼ ð6:26Þ
2U
From the velocity triangles, it is seen that
Cw1 ¼ U þ V w1 ; and C w2 ¼ V w2 2 U
Therefore, Eq. (6.26) can be arranged into a second form:
1 Ca

L¼ þ tan b2 2 tan a2 ð6:27Þ
2 2U
Putting L ¼ 0 in Eq. (6.26), we get
b2 ¼ b1 and V 1 ¼ V 2 ; and for L ¼ 0:5; b2 ¼ a1 :
Zero Reaction Stage:
Let us first discuss the special case of zero reaction. According to the
definition of reaction, when L ¼ 0, Eq. (6.23) reveals that h1 ¼ h2 and Eq. (6.26)
that b1 ¼ b2. The Mollier diagram and velocity triangles for L ¼ 0 are shown in
Figs. 6.17 and 6.18:

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 258 Chapter 6

Figure 6.17 Zero reaction (a) Mollier diagram and (b) velocity diagram.

Now, h01r01 ¼ h02r01 and h1 ¼ h2 for L ¼ 0. Then, V1 ¼ V2. In the ideal

case, there is no pressure drop in the rotor, and points 1, 2 and 2s on the Mollier
chart should coincide. But due to irreversibility, there is a pressure drop through the
rotor. The zero reaction in the impulse stage, by definition, means there is no
pressure drop through the rotor. The Mollier diagram for an impulse stage is shown
in Fig. 6.18, where it can be observed that the enthalpy increases through the rotor.
From Eq. (6.23), it is clear that the reaction is negative for the impulse
turbine stage when irreversibility is taken into account.
Fifty-Percent Reaction Stage
From Eq. (6.23), Fig. (6.19) for L ¼ 0.5, a1 ¼ b2, and the velocity diagram
is symmetrical. Because of symmetry, it is also clear that a2 ¼ b1. For L ¼ 1/2,
the enthalpy drop in the nozzle row equals the enthalpy drop in the rotor. That is:
h0 2 h1 ¼ h1 2 h2

Figure 6.18 Mollier diagram for an impulse stage.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Steam Turbines 259

Figure 6.19 A 50% reaction stage (a) Mollier diagram and (b) velocity diagram.

U
Substituting b2 ¼ tan a2 þ into Eq. (6.27) gives
Ca
Ca
L¼1þ ðtan a2 2tan a1 Þ ð6:28Þ
2U
Thus, when a2 ¼ a1, the reaction is unity (also C1 ¼ C2). The velocity diagram
for L ¼ 1 is shown in Fig. 6.20 with the same value of Ca, U, and W used for
L ¼ 0 and L ¼ 12. It is obvious that if L exceeds unity, then C1 , C0 (i.e., nozzle
flow diffusion).
Choice of Reaction and Effect on Efficiency:
Eq. (6.24) can be rewritten as:
C w2 2 C w1
L¼1þ :
2U
Cw2 can be eliminated by using this equation:
W
Cw2 ¼ 2 C w1 ;
U

Figure 6.20 Velocity diagram for 100% reaction turbine stage.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 260 Chapter 6

Figure 6.21 Influence of reaction on total-to-static efficiency with fixed values of

stage-loading factor.

Figure 6.22 Blade loading coefficient vs. flow coefficient.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Steam Turbines 261

yielding:

W C w1
L¼1þ 2
2 ð6:29Þ
2U U

In Fig. 6.21 the total-to-static efficiencies are shown plotted against the degree of
reaction.
W
When 2 ¼ 2, hts is maximum at L ¼ 0. With higher loading, the
U
optimum hts is obtained with higher reaction ratios. As shown in Fig. 6.22 for a
high total-to-total efficiency, the blade-loading factor should be as small as
possible, which implies the highest possible value of blade speed is consistent
with blade stress limitations. It means that the total-to-static efficiency is heavily
dependent upon the reaction ratio and hts can be optimized by choosing a suitable
value of reaction.

6.13 BLADE HEIGHT IN AXIAL FLOW MACHINES

The continuity equation, m_ ¼ rAC, may be used to find the blade height h. The
annular area of flow ¼ pDh. Thus, the mass flow rate through an axial flow
compressor or turbine is:

m_ ¼ rpDhC a ð6:30Þ

Blade height will increase in the direction of flow in a turbine and decrease in the
direction of flow in a compressor.

Illustrative Example 6.6: The velocity of steam leaving a nozzle is

925 m/s and the nozzle angle is 208. The blade speed is 250 m/s. The mass flow
through the turbine nozzles and blading is 0.182 kg/s and the blade velocity
coefficient is 0.7. Calculate the following:
1. Velocity of whirl.
2. Tangential force on blades.
3. Axial force on blades.
4. Work done on blades.
5. Efficiency of blading.
6. Inlet angle of blades for shockless inflow of steam.
Assume that the inlet and outlet blade angles are equal.
Solution:
From the data given, the velocity diagram can be constructed as shown in
Fig. 6.23. The problem can be solved either graphically or by calculation.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 262 Chapter 6

Figure 6.23 Velocity triangles for Example 6.6.

Applying the cosine rule to the KABC,

V 21 ¼ U 2 þ C 21 2 2UC 1 cosa1
¼ 2502 þ 9252 2 ð2Þ £ ð250Þ £ ð925Þ £ cos208
so: V 1 ¼ 695:35 m/s
But,
V2
k¼ ; orV 2 ¼ ð0:70Þ £ ð695:35Þ ¼ 487m/s:
V1
Velocity of whirl at inlet:
Cw1 ¼ C1 cosa1 ¼ 925cos208 ¼ 869:22m/s
Axial component at inlet:
Ca1 ¼ BD ¼ C 1 sina1 ¼ 925sin208 ¼ 316:37m/s
Blade angle at inlet:
C a1 316:37
tanb1 ¼ ¼ ¼ 0:511
Cw1 2 U 619:22
Therefore, b1 ¼ 27.068 ¼ b2 ¼ outlet blade angle.
Cw2 þ U
cos b2 ¼ ;
V2
or: C w2 ¼ V 2 cos b2 2U ¼ 487 £ cos 27:068 2 250
¼ 433:69 2 250 ¼ 183:69 m/s
and: Ca2 ¼ FE ¼ (U þ Cw2) tan b2 ¼ 433.69 tan 27.068 ¼ 221.548 m/s

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Steam Turbines 263

1. Velocity of whirl at inlet, Cw1 ¼ 869.22 m/s;

Velocity of whirl at outlet, Cw2 ¼ 183.69 m/s
2. Tangential force on blades
¼ ṁ (Cw1 þ Cw2) ¼ (0.182) (1052.9) ¼ 191.63 N.
3. Axial force on blades
¼m _ ðC a1 2 C a2 Þ ¼ ð0:182Þ ð316:37 2 221:548Þ ¼ 17:26N
4. Work done on blades
¼ tangential force on blades £ blade velocity
¼ (191.63) £ (250)/1000 ¼ 47.91 kW.
Work done on blades
5. Efficiency of blading ¼
Kinetic energy supplied

47:91 ð47:91Þð2Þð103 Þ
¼1 ¼
2
2 mC 1
ð0:182Þð9252 Þ

¼ 0:6153 or 61:53%

6. Inlet angle of blades b1 ¼ 27.068 ¼ b2.

Design Example 6.7: The steam velocity leaving the nozzle is 590 m/s
and the nozzle angle is 208. The blade is running at 2800 rpm and blade diameter
is 1050 mm. The axial velocity at rotor outlet ¼ 155 m/s, and the blades are
symmetrical. Calculate the work done, the diagram efficiency and the blade
velocity coefficient.
Solution:
Blade speed U is given by:

pDN ðp £ 1050Þ £ ð2800Þ

U¼ ¼ ¼ 154 m/s
60 ð1000Þ £ ð60Þ

The velocity diagram is shown in Fig. 6.24.

Applying the cosine rule to the triangle ABC,

V 21 ¼ U 2 þ C21 2 2UC 1 cos a1

¼ 1542 þ 5902 2 ð2Þ £ ð154Þ £ ð590Þ cos 208

i.e. V 1 ¼ 448:4m/s:

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 264 Chapter 6

Figure 6.24 Velocity diagram for Example 6.7.

Applying the sine rule to the triangle ABC,

C1 V1
¼
sin ðACBÞ sin ða1 Þ
But sin (ACB) ¼ sin (1808 2 b1) ¼ sin (b1)
Therefore,
C1 sin ða1 Þ 590 sin ð208Þ
sin ðb1 Þ ¼ ¼ ¼ 0:450
V1 448:4
and: b1 ¼ 26.758
From triangle ABD,
C w1 ¼ C 1 cos ða1 Þ ¼ 590 cos ð208Þ ¼ 554:42 m/s
From triangle CEF,
Ca2
¼ tan ðb2 Þ ¼ tan ðb1 Þ ¼ tan ð26:758Þ ¼ 0:504
U þ C w2
Ca2 ¼ 155 ¼ 307:54
or: U þ C w2 ¼ 0:504 0:504

so : C w2 ¼ 307:54 2 154 ¼ 153:54 m/s

Therefore,
DCw ¼ Cw1 þ C w2 ¼ 554:42 þ 153:54 ¼ 707:96 m/s
Relative velocity at the rotor outlet is:
C a2 155
V2 ¼ ¼ ¼ 344:4 m/s
sin ðb2 Þ sin ð26:758Þ

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Steam Turbines 265

Blade velocity coefficient is:

V 2 344:4
k¼ ¼ ¼ 0:768
V 1 448:4
Work done on the blades per kg/s:
DC w2 U ¼ ð707:96Þ £ ð154Þ £ ð1023 Þ ¼ 109 kW
The diagram efficiency is:
2UDC w ð2Þ £ ð707:96Þ £ ð154Þ
hd ¼ ¼ ¼ 0:6264
C 21 5902

or, hd ¼ 62:64%

Illustrative Example 6.8: In one stage of an impulse turbine the velocity

of steam at the exit from the nozzle is 460 m/s, the nozzle angle is 228 and the
blade angle is 338. Find the blade speed so that the steam shall pass on without
shock. Also find the stage efficiency and end thrust on the shaft, assuming
velocity coefficient ¼ 0.75, and blades are symmetrical.
Solution:
From triangle ABC (Fig. 6.25):
C w1 ¼ C 1 cos 228 ¼ 460 cos 228 ¼ 426:5 m/s
and:
C a1 ¼ C1 sin 228 ¼ 460 sin 228 ¼ 172:32 m/s

Figure 6.25 Velocity triangles for Example 6.8.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 266 Chapter 6

Now, from triangle BCD:

Ca1 172:32
BD ¼ ¼ ¼ 265:5
tan ð338Þ 0:649
Hence, blade speed is given by:
U ¼ C w1 2 BD ¼ 426:5 2 265:5 ¼ 161 m/s
From Triangle BCD, relative velocity at blade inlet is given by:
C a1 172:32
V1 ¼ ¼ ¼ 316:2 m/s
sin ð338Þ 0:545

Velocity coefficient:
V2
k¼ ; or V 2 ¼ kV 1 ¼ ð0:75Þ £ ð316:2Þ ¼ 237:2 m/s
V1
From Triangle BEF,
BF ¼ V 2 cos ð338Þ ¼ 237:2 £ cos ð338Þ ¼ 198:9
and
Cw2 ¼ AF ¼ BF 2 U ¼ 198:9 2 161 ¼ 37:9 m/s

Ca2 ¼ V 2 sin ð338Þ ¼ 237:2 sin ð338Þ ¼ 129:2 m/s

The change in velocity of whirl:
DC w ¼ C w1 þ Cw2 ¼ 426:5 þ 37:9 ¼ 464:4 m/s
Diagram efficiency:
2UDC w ð2Þ £ ð464:4Þ £ ð161Þ
hd ¼ ¼ ¼ 0:7067; or 70:67%:
C 21 4602
End thrust on the shaft per unit mass flow:
Ca1 2 Ca2 ¼ 172:32 2 129:2 ¼ 43:12 N

Design Example 6.9: In a Parson’s turbine, the axial velocity of flow of

steam is 0.5 times the mean blade speed. The outlet angle of the blade is 208,
diameter of the ring is 1.30 m and the rotational speed is 3000 rpm. Determine the
inlet angles of the blades and power developed if dry saturated steam at 0.5 MPa
passes through the blades where blade height is 6 cm. Neglect the effect of the
blade thickness.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Steam Turbines 267

Figure 6.26 Velocity triangles for Example 6.9.

Solution:
pDN p £ ð1:30Þ £ ð3000Þ
The blade speed, U ¼ ¼ ¼ 204 m/s
60 60
Velocity of flow, Ca ¼ (0.5) £ (204) ¼ 102 m/s

Draw lines AB and CD parallel to each other Fig. 6.26 at the distance of
102 m/s, i.e., velocity of flow, Ca1 ¼ 102 m/s.
At any point B, construct an angle a2 ¼ 208 to intersect line CD at point
C. Thus, the velocity triangle at the outlet is completed. For Parson’s turbine,

a 1 ¼ b2 ; b 1 ¼ a2 ; C1 ¼ V 2 ; and V 1 ¼ C2 :

By measurement,
DCw ¼ Cw1 þ C w2 ¼ 280:26 þ 76:23 ¼ 356:5 m/s
The inlet angles are 53.228.Specific volume of vapor at 0.5 MPa, from the
steam tables, is
vg ¼ 0:3749 m3 /kg
Therefore the mass flow is given by:
AC 2 p £ ð1:30Þ £ ð6Þ £ ð102Þ
m_ ¼ ¼ ¼ 66:7 kg/s
x2 vg2 ð100Þ £ ð0:3749Þ
Power developed:
_
mUDC w ð66:7Þ £ ð356:5Þ £ ð102Þ
P¼ ¼ ¼ 2425:4 kW
1000 1000

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 268 Chapter 6

Figure 6.27 Velocity triangles for Example 6.10.

Design Example 6.10: In an impulse turbine, steam is leaving the nozzle with
velocity of 950 m/s and the nozzle angle is 208. The nozzle delivers steam at
the rate of 12 kg/min. The mean blade speed is 380 m/s and the blades are
symmetrical. Neglect friction losses. Calculate (1) the blade angle, (2)
the tangential force on the blades, and (3) the horsepower developed.
Solution:
With the help of a1, U and C1, the velocity triangle at the blade inlet can be
constructed easily as shown in Fig. 6.27.
Applying the cosine rule to the triangle ABC,
V 21 ¼ U 2 þ C 21 2 2UC 1 cosa1
¼ 9502 þ 3802 2 ð2Þ £ ð950Þ £ ð380Þ £ cos208 ¼ 607m/s
Now, applying the sine rule to the triangle ABC,
V1 C1 C1
¼ ¼
sinða1 Þ sinð1808 2 b1 Þ sinðb1 Þ
or:
C 1 sinða1 Þ ð950Þ £ ð0:342Þ
sinðb1 Þ ¼ ¼ ¼ 0:535
V1 607
so:
b1 ¼ 32:368
From Triangle ACD,
Cw1 ¼ C1 cos ða1 Þ ¼ 950 £ cos ð208Þ ¼ ð950Þ £ ð0:9397Þ

¼ 892:71m/s

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Steam Turbines 269

As b1 ¼ b2, using triangle BEF and neglecting friction loss, i.e,: V1 ¼ V2

BF ¼ V 2 cos b2 ¼ 607 £ cos 32:368 ¼ 512:73
Therefore,
C w2 ¼ BF 2 U ¼ 512:73 2 380 ¼ 132:73 m/s
Change in velocity of whirl:
DCw ¼ Cw1 þ C w2 ¼ 892:71 þ 132:73 ¼ 1025:44 m/s
Tangential force on blades:
ð12Þ £ ð1025:44Þ
F ¼ mDC
_
w ¼ ¼ 205N
60
ð12Þ £ ð1025:44Þ £ ð380Þ
Horsepower, P ¼ mUDC
_
w ¼ ¼ 104:47 hp
ð60Þ £ ð1000Þ £ ð0:746Þ

Design Example 6.11: In an impulse turbine, the velocity of steam at the

exit from the nozzle is 700 m/s and the nozzles are inclined at 228 to the blades,
whose tips are both 348. If the relative velocity of steam to the blade is reduced by
10% while passing through the blade ring, calculate the blade speed, end thrust on
the shaft, and efficiency when the turbine develops 1600 kW.
Solution:
Velocity triangles for this problem are shown in Fig. 6.28.
From the triangle ACD,
C a1 ¼ C1 sin a1 ¼ 700 £ sin 228 ¼ 262:224 m/s

Figure 6.28 Velocity triangles for Example 6.11.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 270 Chapter 6

and
C a1 262:224
V1 ¼ ¼ ¼ 469:32 m/s
sin ðb1 Þ sin 348
Whirl component of C1 is given by
Cw1 ¼ C1 cos ða1 Þ ¼ 700 cos ð228Þ ¼ 700 £ 0:927 ¼ 649 m/s
Now, BD ¼ Cw1 2 U ¼ V1 cos b1 ¼ (469.32) £ (0.829) ¼ 389
Hence, blade speed
U ¼ 649 2 389 ¼ 260 m/s
Using the velocity coefficient to find V2:
i:e:; V 2 ¼ ð0:90Þ £ ð469:32Þ ¼ 422:39 m/s
From velocity triangle BEF,
Ca2 ¼ V 2 sin ðb2 Þ ¼ 422:39 sin 348 ¼ 236:2 m/s
And
U þ Cw2 ¼ V 2 cos 348 ¼ ð422:39Þ £ ð0:829Þ ¼ 350:2 m/s
Therefore,
Cw2 ¼ 350:2 2 260 ¼ 90:2 m/s
Then,
DC w ¼ C w1 þ Cw2 ¼ 649 þ 90:2 ¼ 739:2 m/s
Mass flow rate is given by:
P ¼ mUDC
_
w

or

_ ¼
ð1600Þ £ ð1000Þ
m ¼ 8:325 kg/s
ð739:2Þ £ ð260Þ
Thrust on the shaft,
F ¼ m_ ðC a1 2 Ca2 Þ ¼ 8:325ð262:224 2 236:2Þ ¼ 216:65 N
Diagram efficiency:
2UDC w ð2Þ £ ð739:2Þ £ ð260Þ
hd ¼ ¼ ¼ 0:7844; or 78:44%:
C 21 7002

Illustrative Example 6.12: The moving and fixed blades are identical in
shape in a reaction turbine. The absolute velocity of steam leaving the fixed blade
is 105 m/s, and the blade velocity is 40 m/s. The nozzle angle is 208. Assume axial

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Steam Turbines 271

Figure 6.29 Velocity triangles for Example 6.12.

velocity is constant through the stage. Determine the horsepower developed if the
steam flow rate is 2 kg/s.

Solution:
For 50% reaction turbine Fig. 6.29, a1 ¼ b2, and a2 ¼ b1.
From the velocity triangle ACD,
C w1 ¼ C 1 cos a1 ¼ 105 cos 208 ¼ 98:67 m/s
Applying cosine rule to the Triangle ABC:

V 21 ¼ C21 þ U 2 2 2C 1 U cos a1
so:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
V1 ¼ 1052 þ 402 2 ð2Þ £ ð105Þ £ ð40Þ £ cos208 ¼ 68:79 m/s

Now,
BD ¼ C w1 2 U ¼ V 1 cos b1 ¼ 98:67 2 40 ¼ 58:67
Hence,

58:67
cosb1 ¼ ¼ 0:853; and b1 ¼ 31:478
68:79

Change in the velocity of whirl is:

DCw ¼ Cw1 þ C w2 ¼ 98:67 þ 58:67 ¼ 157:34 m/s

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 272 Chapter 6

Figure 6.30 Velocity triangles for Example 6.13.

Horsepower developed is:

ð2Þ £ ð157:34Þ £ ð40Þ
P ¼ mUDC
_
w ¼ ¼ 16:87 hp
ð0:746Þ £ ð1000Þ

Illustrative Example 6.13: The inlet and outlet angles of blades of

a reaction turbine are 25 and 188, respectively. The pressure and temperature
of steam at the inlet to the turbine are 5 bar and 2508C. If the steam flow rate
is 10 kg/s and the rotor diameter is 0.72 m, find the blade height and
power developed. The velocity of steam at the exit from the fixed blades is
90 m/s.
Solution:
Figure 6.30 shows the velocity triangles.
a1 ¼ b2 ¼ 188; and a2 ¼ b1 ¼ 258
C1 ¼ 90 m/s
From the velocity triangle,
Cw1 ¼ C1 cos ða1 Þ ¼ 90 cos 188 ¼ 85:6 m/s
Ca1 ¼ CD ¼ C 1 sin a1 ¼ 90 sin 188 ¼ 27:8 m/s
From triangle BDC
C a1 27:8 27:8
BD ¼ ¼ ¼ ¼ 65:72 m/s
sin ðb1 Þ sin ð258Þ 0:423
Hence blade velocity is given by:
U ¼ C w1 2 BD ¼ 85:6 2 65:62 ¼ 19:98 m/s:
Applying the cosine rule,
V 21 ¼ C 21 þ U 2 2 2C 1 U cos a1
¼ 902 þ 19:982 2 ð2Þ £ ð90Þ £ ð19:98Þ cos 188
V 1 ¼ 71:27m/s

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Steam Turbines 273

From triangle AEF,

C w2 ¼ C 2 cosða2 Þ ¼ 71:27 cos 258 ¼ 64:59 m/s

Change in the velocity of whirl:

DCw ¼ Cw1 þ C w2 ¼ 85:6 þ 64:59 ¼ 150:19 m/s

Power developed by the rotor:

ð10Þ £ ð19:98Þ £ ð150:19Þ
P ¼ mUDC
_
w ¼ ¼ 30 kW
1000
From superheated steam tables at 5 bar, 2508C, the specific volume of
steam is:

v ¼ 0:4744 m3 /kg

Blade height is given by the volume of flow equation:

v ¼ pDhC a

where Ca is the velocity of flow and h is the blade height.Therefore,

0:4744 ¼ p £ ð0:72Þ £ ðhÞ £ ð27:8Þ; and

h ¼ 0:0075 m or 0:75 cm

Design Example 6.14: From the following data, for 50% reaction steam
turbine, determine the blade height:

RPM: 440
Power developed: 5:5 MW
Steam mass flow rate: 6:8 kg/kW
h
Stage absolute pressure: 0:90 bar
Steam dryness fraction: 0:95
Exit angles of the blades: 708
(angle measured from the axial flow direction).
The outlet relative velocity of steam is 1.2 times the mean blade speed. The
ratio of the rotor hub diameter to blade height is 14.5.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 274 Chapter 6

Solution:
Figure 6.31 shows the velocity triangles.
From the velocity diagram,
V 2 ¼ 1:2U
C a2 ¼ V 2 cos ðb2 Þ
¼ 1:2U cos 708
¼ 0:41U m/s
At mean diameter,
pDN 2pN ðDh þ hÞ
U¼ ¼
60 ð60Þ £ ð2Þ
where Dh is the rotor diameter at the hub and h is the blade height.
Substituting the value of U in the above equation,
ð0:41Þ £ ð2pÞ £ ð440Þð14:5h þ hÞ
Ca2 ¼ ¼ 146:45 h m/s
ð2Þ £ ð60Þ
Annular area of flow is given by:
A ¼ phðDh þ hÞ ¼ phð14:5h þ hÞ

or

A ¼ 15:5ph 2

Specific volume of saturated steam at 0.90 bar, vg ¼ 1.869 m3/kg.

Figure 6.31 Velocity triangles for Example 6.14.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Steam Turbines 275

Then the specific volume of steam ¼ (1.869) £ (0.95) ¼ 1.776 m3/kg.

The mass flow rate is given by:

ð5:5Þ £ ð103 Þ £ ð6:8Þ

m_ ¼ ¼ 10:39 kg/s
3600
But,
C a2 A Ca2 15:5ph 2
m_ ¼ ¼
v v
Therefore:
ð146:45Þ £ ðhÞ £ ð15:5Þ £ ðph 2 Þ
10:39 ¼
1:776
or:
h 3 ¼ 0:00259; and h ¼ 0:137 m

Design Example 6.15: From the following data for a two-row velocity
compounded impulse turbine, determine the power developed and the diagram
efficiency:
Blade speed: 115 m/s
Velocity of steam exiting the nozzle: 590 m/s
Nozzle efflux angle: 188
Outlet angle from first moving blades: 378
Blade velocity coefficient ðall bladesÞ: 0:9

Solution:
Figure 6.32 shows the velocity triangles.
Graphical solution:
U ¼ 115 m/s
C1 ¼ 590 m/s
a1 ¼ 188
b2 ¼ 208
The velocity diagrams are drawn to scale, as shown in Fig. 6.33, and the
relative velocity:
V1 ¼ 482 m/s using the velocity coefficient
V2 ¼ (0.9) £ (482) ¼ 434 m/s

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 276 Chapter 6

Figure 6.32 Velocity triangle for Example 6.15.

The absolute velocity at the inlet to the second row of moving blades, C3, is
equal to the velocity of steam leaving the fixed row of blades.

i:e:; : C3 ¼ kC 2 ¼ ð0:9Þ £ ð316:4Þ ¼ 284:8

Driving force ¼ ṁ DCw

For the first row of moving blades, ṁDCw1 ¼ (1) £ (854) ¼ 854 N.
For the second row of moving blades, ṁDCw2 ¼ (1) £ (281.46)
N ¼ 281.46 N
where DCw1 and DCw2 are scaled from the velocity diagram.

Figure 6.33 Velocity diagram for Example 6.16.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Steam Turbines 277

Total driving force ¼ 854 þ 281.46 ¼ 1135.46 N per kg/s

Power ¼ driving force £ blade velocity
ð1135:46Þ £ ð115Þ
¼ ¼ 130:58 kW per kg/s
1000

Energy supplied to the wheel

mC 21 ð1Þ £ ð5902 Þ
¼ ¼ ¼ 174:05kW per kg/s
2 ð2Þ £ ð103 Þ
Therefore, the diagram efficiency is:
ð130:58Þ £ ð103 Þ £ ð2Þ
hd ¼ ¼ 0:7502; or 75:02%
5902
Maximum diagram efficiency:
¼ cos2 a1 ¼ cos2 88 ¼ 0:9045; or 90:45%
Axial thrust on the first row of moving blades (per kg/s):
_ a1 2 C a2 Þ ¼ ð1Þ £ ð182:32 2 148:4Þ ¼ 33:9 N
¼ mðC
Axial thrust on the second row of moving blades (per kg/s):
_ a3 2 C a4 Þ ¼ ð1Þ £ ð111:3 2 97:57Þ ¼ 13:73 N
¼ mðC
Total axial thrust:
¼ 33:9 þ 13:73 ¼ 47:63 N per kg/s

Design Example 6.16: In a reaction stage of a steam turbine, the blade

angles for the stators and rotors of each stage are: a1 ¼ 258, b1 ¼ 608,
a2 ¼ 71.18, b2 ¼ 328. If the blade velocity is 300 m/s, and the steam flow rate is
5 kg/s, find the power developed, degree of reaction, and the axial thrust.
Solution:
Figure 6.34 shows the velocity triangles.
The velocity triangles can easily be constructed as the blade velocity and
blade angles are given.From velocity triangles, work output per kg is given
by:
W t ¼ UðC w1 þ Cw2 Þ
¼ ð300Þ £ ð450 cos 258 þ 247 cos 71:18Þ
¼ 14; 6; 354 J

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 278 Chapter 6

Figure 6.34 Velocity diagram for Example 6.17.

Power output:

ð5Þ £ ð1; 46; 354Þ

t ¼ ¼ 732 kW
_
mW
1000
Degree of reaction is given by:

V 22 2 V 21 4432 2 2202
L¼ ¼ ¼ 0:5051; or 50:51%
2 £ Wt ð2Þ £ ð14; 6; 354Þ

Axial thrust:

F ¼ mðC
_ a1 2 Ca2 Þ ¼ ð5Þ £ ð190:5 2 234Þ ¼ 2217:5 N

The thrust is negative because its direction is the opposite to the fluid flow.

Design Example 6.17: Steam enters the first row of a series of stages at a
static pressure of 10 bars and a static temperature of 3008C. The blade angles for
the rotor and stator of each stage are: a1 ¼ 258, b1 ¼ 608, a2 ¼ 70.28, b2 ¼ 328.
If the blade speed is 250 m/s, and the rotor efficiency is 0.94, find the degree of
reaction and power developed for a 5.2 kg/s of steam flow. Also find the static
pressures at the rotor inlet and exit if the stator efficiency is 0.93 and the carry-
over efficiency is 0.89.
Solution:
Using the given data, the velocity triangles for the inlet and outlet are
shown in Fig. 6.34. By measurement, C2 ¼ 225 m/s, V2 ¼ 375 m/s,
C1 ¼ 400 m/s, V1 ¼ 200 m/s.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Steam Turbines 279

Work done per unit mass flow:

W t ¼ ð250Þ £ ð400 cos 258 þ 225 cos 70:28Þ ¼ 1; 09; 685 J/kg
Degree of reaction [Eq. (6.25)]
V 22 2 V 21 3752 2 2002
L¼ ¼ ¼ 0:4587; or 45:87%
2 £ Wt ð2Þ £ ð1; 09; 685Þ
Power output:
ð5:2Þ £ ð1; 09; 685Þ
P ¼ mW
_ ¼ ¼ 570:37 kW
1000
Isentropic static enthalpy drop in the stator:

2 

0 C 1 2 C 22 4002 2 ð0:89Þ £ ð2252 Þ
Dhs ¼ ¼
hs 0:93

¼ 1; 23; 595 J/kg; or 123:6 kJ/kg

Isentropic static enthalpy drops in the rotor:

W 1; 09; 685
Dhr 0 ¼ ¼
hr hs ð0:94Þ £ ð0:93Þ

¼ 1; 25; 469 J/kg; or 125:47 kJ/kg

Since the state of the steam at the stage entry is given as 10 bar, 3008C,
Enthalpy at nozzle exit:
h1 2 Dh0 stator
¼ 3051:05 2 123:6 ¼ 2927:5kJ/kg
Enthalpy at rotor exit:
h1 2 Dh0 rotor
¼ 3051:05 2 125:47 ¼ 2925:58kJ/kg
The rotor inlet and outlet conditions can be found by using the Mollier
Chart.
Rotor inlet conditions: P1 ¼ 7 bar, T1 ¼ 2358C
Rotor outlet conditions: P2 ¼ 5 bar, T2 ¼ 2208C

PROBLEMS
6.1 Dry saturated steam is expanded in a steam nozzle from 1 MPa to 0.01 MPa.
Calculate dryness fraction of steam at the exit and the heat drop.
(0.79, 686 kJ/kg)

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 280 Chapter 6

6.2 Steam initially dry and at 1.5 MPa is expanded adiabatically in a nozzle to
7.5 KPa. Find the dryness fraction and velocity of steam at the exit. If the exit
diameter of the nozzles is 12.5 mm, find the mass of steam discharged per
hour.
(0.756, 1251.26 m/s, 0.376 kg/h)
6.3 Dry saturated steam expands isentropically in a nozzle from 2.5 MPa to
0.30 MPa. Find the dryness fraction and velocity of steam at the exit from
the nozzle. Neglect the initial velocity of the steam.
(0.862, 867.68 m/s)
6.4 The nozzles receive steam at 1.75 MPa, 3008C, and exit pressure of steam is
1.05 MPa. If there are 16 nozzles, find the cross-sectional area of the exit of
each nozzle for a total discharge to be 280 kg/min. Assume nozzle
efficiency of 90%. If the steam has velocity of 120 m/s at the entry to the
nozzles, by how much would the discharge be increased?
(1.36 cm2, 33.42%)
6.5 The steam jet velocity of a turbine is 615 m/s and nozzle angle is 228, The
blade velocity coefficient ¼ 0.70 and the blade is rotating at 3000 rpm.
Assume mean blade radius ¼ 600 mm and the axial velocity at the
outlet ¼ 160 m/s. Determine the work output per unit mass flow of steam and
diagram efficiency.
(93.43 kW, 49.4%)
6.6 Steam is supplied from the nozzle with velocity 400 m/s at an angle of 208
with the direction of motion of moving blades. If the speed of the blade is
200 m/s and there is no thrust on the blades, determine the inlet and outlet
blade angles, and the power developed by the turbine. Assume velocity
coefficient ¼ 0.86, and mass flow rate of steam is 14 kg/s.
(378 500 , 458, 310 , 1234.8 kW)
6.7 Steam expands isentropically in the reaction turbine from 4 MPa, 4008C to
0.225 MPa. The turbine efficiency is 0.84 and the nozzle angles and blade
angles are 20 and 368 respectively. Assume constant axial velocity
throughout the stage and the blade speed is 160 m/s. How many stages are
there in the turbine?
(8 stages)
6.8 Consider one stage of an impulse turbine consisting of a converging nozzle
and one ring of moving blades. The nozzles are inclined at 208 to the blades,
whose tip angles are both 338. If the velocity of the steam at the exit from
the nozzle is 650 m/s, find the blade speed so that steam passes through
without shock and find the diagram efficiency, neglecting losses.
(273 m/s, 88.2%)

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Steam Turbines 281

6.9 One stage of an impulse turbine consists of a converging nozzle and one
ring of moving blades. The nozzle angles are 228 and the blade angles are
358. The velocity of steam at the exit from the nozzle is 650 m/s. If the
relative velocity of steam to the blades is reduced by 14% in passing
through the blade ring, find the diagram efficiency and the end thrust on the
shaft when the blade ring develops 1650 kW.
(79.2%, 449 N)
6.10 The following refer to a stage of a Parson’s reaction turbine:
Mean diameter of the blade ring: 92 cm
Blade speed: 3000 rpm
Inlet absolute velocity of steam: 310 m/s
Blade outlet angle: 208
Steam flow rate: 6:9 kg/s

Determine the following: (1) blade inlet angle, (2) tangential force, and
(3) power developed.
(388, 2.66 kW, 384.7 kW)

NOTATION
C absolute velocity, velocity of steam at nozzle exit
D diameter
h enthalpy, blade height
h0 stagnation enthalpy, static enthalpy at the inlet to the fixed
blades
h1 enthalpy at the entry to the moving blades
h2 enthalpy at the exit from the moving blades
h00 stagnation enthalpy at the entry to the fixed blades
h0l stagnation enthalpy at the entry to the fixed blades
h02 stagnation enthalpy at the exit from the moving blade
k blade velocity coefficient
N rotational speed
R. F. reheat factor
U blade speed
V relative velocity
a angle with absolute velocity
b angle with relative velocity
DCw change in the velocity of whirl
Dh actual enthalpy drop
Dh0 isentropic enthalpy drop

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 282 Chapter 6

hd diffuser efficiency
hn nozzle efficiency
hs stage efficiency
ht turbine efficiency
hts total - to - static efficiency
htt total - to - total efficiency
L degree of reaction

SUFFIXES
0 inlet to fixed blades
1 inlet to moving blades
2 outlet from the moving blades
a axial, ambient
r radial
w whirl

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 7
Axial Flow and Radial Flow
Gas Turbines

7.1 INTRODUCTION TO AXIAL FLOW TURBINES

The axial flow gas turbine is used in almost all applications of gas turbine power
plant. Development of the axial flow gas turbine was hindered by the need to
obtain both a high-enough flow rate and compression ratio from a compressor to
maintain the air requirement for the combustion process and subsequent
expansion of the exhaust gases. There are two basic types of turbines: the axial
flow type and the radial or centrifugal flow type. The axial flow type has been
used exclusively in aircraft gas turbine engines to date and will be discussed in
detail in this chapter. Axial flow turbines are also normally employed in industrial
and shipboard applications. Figure 7.1 shows a rotating assembly of the Rolls-
Royce Nene engine, showing a typical single-stage turbine installation. On this
particular engine, the single-stage turbine is directly connected to the main and
cooling compressors. The axial flow turbine consists of one or more stages
located immediately to the rear of the engine combustion chamber. The turbine
extracts kinetic energy from the expanding gases as the gases come from the
burner, converting this kinetic energy into shaft power to drive the compressor
and the engine accessories. The turbines can be classified as (1) impulse and
(2) reaction. In the impulse turbine, the gases will be expanded in the nozzle and
passed over to the moving blades. The moving blades convert this kinetic
energy into mechanical energy and also direct the gas flow to the next stage

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 284 Chapter 7

Figure 7.1 Axial flow turbine rotors. (Courtesy Rolls-Royce.)

(multi-stage turbine) or to exit (single-stage turbine). Fig. 7.1 shows the axial
flow turbine rotors.
In the case of reaction turbine, pressure drop of expansion takes place in the
stator as well as in the rotor-blades. The blade passage area varies continuously to
allow for the continued expansion of the gas stream over the rotor-blades. The
efficiency of a well-designed turbine is higher than the efficiency of a
compressor, and the design process is often much simpler. The main reason for
this fact, as discussed in compressor design, is that the fluid undergoes a pressure
rise in the compressor. It is much more difficult to arrange for an efficient
deceleration of flow than it is to obtain an efficient acceleration. The pressure
drop in the turbine is sufficient to keep the boundary layer fluid well behaved, and
separation problems, or breakaway of the molecules from the surface, which
often can be serious in compressors, can be easily avoided. However, the turbine
designer will face much more critical stress problem because the turbine rotors
must operate in very high-temperature gases. Since the design principle and
concepts of gas turbines are essentially the same as steam turbines, additional

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow and Radial Flow Gas Turbines 285

information on turbines in general already discussed in Chapter 6 on steam

turbines.

7.2 VELOCITY TRIANGLES AND WORK OUTPUT

The velocity diagram at inlet and outlet from the rotor is shown in Fig. 7.2. Gas
with an absolute velocity C1 making an angle a1, (angle measured from the axial
direction) enters the nozzle (in impulse turbine) or stator blades (in reaction
turbine). Gas leaves the nozzles or stator blades with an absolute velocity C2,
which makes and an a2 with axial direction. The rotor-blade inlet angle will be
chosen to suit the direction b2 of the gas velocity V2 relative to the blade at inlet.
b2 and V2 are found by subtracting the blade velocity vector U from the absolute
velocity C2.
It is seen that the nozzles accelerate the flow, imparting an increased
tangential velocity component. After expansion in the rotor-blade passages, the
gas leaves with relative velocity V3 at angle b3. The magnitude and direction of
the absolute velocity at exit from the rotor C3 at an angle a3 are found by
vectorial addition of U to the relative velocity V3. a3 is known as the swirl angle.

Figure 7.2 Velocity triangles for an axial flow gas turbine.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 286 Chapter 7

The gas enters the nozzle with a static pressure p1 and temperature T1. After
expansion, the gas pressure is p2 and temperature T2. The gas leaves the rotor-
blade passages at pressure p3 and temperature T3. Note that the velocity diagram
of the turbine differs from that of the compressor, in that the change in tangential
velocity in the rotor, DCw, is in the direction opposite to the blade speed U. The
reaction to this change in the tangential momentum of the fluid is a torque on the
rotor in the direction of motion. V3 is either slightly less than V2 (due to friction)
or equal to V2. But in reaction stage, V3 will always be greater than V2 because
part of pressure drop will be converted into kinetic energy in the moving blade.
The blade speed U increases from root to tip and hence velocity diagrams will be
different for root, tip, and other radii points. For short blades, 2-D approach in
design is valid but for long blades, 3-D approach in the designing must be
considered. We shall assume in this section that we are talking about conditions at
the mean diameter of the annulus. Just as with the compressor blading diagram, it
is more convenient to construct the velocity diagrams in combined form, as
shown in Fig. 7.3. Assuming unit mass flow, work done by the gas is given by

W ¼ U ðC w2 þ Cw3 Þ ð7:1Þ

From velocity triangle

U
¼ tan a2 2 tan b2 ¼ tan b3 2 tan a3 ð7:2Þ
Ca

In single-stage turbine, a1 ¼ 0 and C1 ¼ Ca1. In multi-stage turbine, a1 ¼ a3 and

C1 ¼ C3 so that the same blade shape can be used. In terms of air angles, the stage

Figure 7.3 Combined velocity diagram.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow and Radial Flow Gas Turbines 287

work output per unit mass flow is given by

W ¼ U ðC w2 þ C w3 Þ ¼ UCað tan a2 þ tan a3 Þ ð7:3Þ


or W ¼ UCa tan b2 þ tan b3 ð7:4Þ
Work done factor used in the designing of axial flow compressor is not required
because in the turbine, flow is accelerating and molecules will not break away
from the surface and growth of the boundary layer along the annulus walls is
negligible. The stagnation pressure ratio of the stage p01 /p03 can be found from
"  ðg21Þ/g #
1
DT 0s ¼ hs T 01 1 2 ð7:5Þ
p01 /p03
where hs is the isentropic efficiency given by
T 01 2 T 03
hs ¼ ð7:6Þ
T 01 2 T 003
The efficiency given by Eq. (7.6) is based on stagnation (or total)
temperature, and it is known as total-to-total stage efficiency. Total-to-total stage
efficiency term is used when the leaving kinetics energy is utilized either in the
next stage of the turbine or in propelling nozzle. If the leaving kinetic energy
from the exhaust is wasted, then total-to-static efficiency term is used. Thus total-
to-static efficiency,
T 01 2 T 03
hts ¼ 0 ð7:7Þ
T 01 2 T 3
where T 30 in Eq. (7.7) is the static temperature after an isentropic expansion from
p01 to p3.

7.3 DEGREE OF REACTION (L)

Degree of reaction is defined as

Enthalpy drop in the moving blades

L¼
Enthalpy drop in the stage
h2 2 h3 Ca

¼ ¼ tan b1 2 tan b2 ð7:8Þ
h1 2 h3 2U

This shows the fraction of the stage expansion, which occurs in the rotor, and it is
usual to define in terms of the static temperature drops, namely
T2 2 T3
L¼ ð7:9Þ
T1 2 T3

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 288 Chapter 7

Assuming that the axial velocity is constant throughout the stage, then

Ca2 ¼ Ca3 ¼ Ca1 ; and C3 ¼ C 1

From Eq. (7.4)


C p ðT 1 2 T 3 Þ ¼ Cp ðT 01 2 T 03 Þ ¼ UCa tan b2 þ tan b3 ð7:10Þ

Temperature drop across the rotor-blades is equal to the change in relative

velocity, that is
1
2 
C p ðT 2 2 T 3 Þ ¼V 2 V 22
2 3
1

¼ Ca 2 sec2 b3 2 sec2 b2
2
1

¼ Ca 2 tan2 b3 2 tan2 b2
2
Thus
Ca

L¼ tan b3 2 tan b2 ð7:11Þ
2U

7.4 BLADE-LOADING COEFFICIENT

The blade-loading coefficient is used to express work capacity of the stage. It is
defined as the ratio of the specific work of the stage to the square of the blade
velocity—that is, the blade-loading coefficient or temperature-drop coefficient c
is given by

W 2C p DT os 2Ca

c¼1 2
¼ 2
¼ tan b2 þ tan b3 ð7:12Þ
2U
U U

Flow Coefficient (f)

The flow coefficient, f, is defined as the ratio of the inlet velocity Ca to the
blade velocity U, i.e.,
Ca
f¼ ð7:13Þ
U
This parameter plays the same part as the blade-speed ratio U/C1 used in the
design of steam turbine. The two parameters, c and f, are dimensionless and

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow and Radial Flow Gas Turbines 289

useful to plot the design charts. The gas angles in terms of c, L, and f can be
obtained easily as given below:
Eqs. (7.11) and (7.12) can be written as


c ¼ 2f tan b2 þ tan b3 ð7:14Þ
f

L¼ tan b3 2 tan b2 ð7:15Þ
2
Now, we may express gas angles b2 and b3 in terms of c, L, and f as follows:
Adding and subtracting Eqs. (7.14) and (7.15), we get
 
1 1
tan b3 ¼ c þ 2L ð7:16Þ
2f 2
 
1 1
tan b2 ¼ c 2 2L ð7:17Þ
2f 2
Using Eq. (7.2)
1
tan a3 ¼ tan b3 2 ð7:18Þ
f
1
tan a2 ¼ tan b2 þ ð7:19Þ
f
It has been discussed in Chapter 6 that steam turbines are usually impulse or a
mixture of impulse and reaction stages but the turbine for a gas-turbine power
plant is a reaction type. In the case of steam turbine, pressure ratio can be of the
order of 1000:1 but for a gas turbine it is in the region of 10:1. Now it is clear that
a very long steam turbine with many reaction stages would be required to reduce
the pressure by a ratio of 1000:1. Therefore the reaction stages are used where
pressure drop per stage is low and also where the overall pressure ratio of the
turbine is low, especially in the case of aircraft engine, which may have only
three or four reaction stages.
Let us consider 50% reaction at mean radius. Substituting L ¼ 0.5 in
Eq. (7.11), we have
1
¼ tan b3 2 tan b2 ð7:20Þ
f
Comparing this with Eq. (7.2), b3 ¼ a2 and b2 ¼ a3, and hence the velocity
diagram becomes symmetrical. Now considering C1 ¼ C3, we have a1 ¼ a3 ¼ b2,
and the stator and rotor-blades then have the same inlet and outlet angles. Finally,
for L ¼ 0.5, we can prove that
c ¼ 4f tan b3 2 2 ¼ 4f tan a2 2 2 ð7:21Þ
and c ¼ 4f tan b2 þ 2 ¼ 4f tan a3 þ 2 ð7:22Þ
and hence all the gas angles can be obtained in terms of c and f.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 290 Chapter 7

Figure 7.4 Total-to-static efficiency of a 50% reaction axial flow turbine stage.

The low values of f and c imply low gas velocities and hence reduced
friction losses. But a low value of c means more stages for a given overall turbine
output, and low f means larger turbine annulus area for a given mass flow. In
industrial gas turbine plants, where low sfc is required, a large diameter, relatively
long turbine, of low flow coefficient and low blade loading, would be accepted.
However, for the gas turbine used in an aircraft engine, the primary consideration is
to have minimum weight, and a small frontal area. Therefore it is necessary to use
higher values of c and f but at the expense of efficiency (see Fig. 7.4).

7.5 STATOR (NOZZLE) AND ROTOR LOSSES

A T –s diagram showing the change of state through a complete turbine stage,
including the effects of irreversibility, is given in Fig. 7.5.
In Fig. 7.5, T02 ¼ T01 because no work is done in the nozzle,
ðp01 2 p02 Þ represents the pressure drop due to friction in the nozzle. ðT 01 2
0
T 2 Þ represents the ideal expansion in the nozzle, T2 is the temperature at the
0
nozzle exit due to friction. Temperature, T2 at the nozzle exit is higher than T 2 .
The nozzle loss coefficient, lN, in terms of temperature may be defined as
0
T2 2 T2
lN ¼ 2 ð7:23Þ
C 2 /2C p

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow and Radial Flow Gas Turbines 291

Figure 7.5 T– s diagram for a reaction stage.

Nozzle loss coefficient in term of pressure

p01 2 p02
yN ¼ ð7:24Þ
p01 2 p2
lN and yN are not very different numerically. From Fig. 7.5, further expansion in
0
the rotor-blade passages reduces the pressure to p3. T 3 is the final temperature
00
after isentropic expansion in the whole stage, and T 3 is the temperature after
expansion in the rotor-blade passages alone. Temperature T3 represents the
temperature due to friction in the rotor-blade passages. The rotor-blade loss can
be expressed by
00
T3 2 T3
lR ¼ ð7:25Þ
V 23 /2C p

As we know that no work is done by the gas relative to the blades, that is,
T03rel ¼ T02rel. The loss coefficient in terms of pressure drop for the rotor-blades
is defined by
p02 rel 2 p03 rel
lR ¼ ð7:26Þ
p03 rel 2 p3

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 292 Chapter 7

The loss coefficient in the stator and rotor represents the percentage drop of
energy due to friction in the blades, which results in a total pressure and static
enthalpy drop across the blades. These losses are of the order of 10 – 15% but can
be lower for very low values of flow coefficient.
Nozzle loss coefficients obtained from a large number of turbine tests are
typically 0.09 and 0.05 for the rotor and stator rows, respectively. Figure 7.4 shows
the effect of blade losses, determined with Soderberg’s correlation, on the total-to-
total efficiency of turbine stage for the constant reaction of 50%. It is the evident
that exit losses become increasingly dominant as the flow coefficient is increased.

7.6 FREE VORTEX DESIGN

As pointed out earlier, velocity triangles vary from root to tip of the blade
because the blade speed U is not constant and varies from root to tip. Twisted
blading designed to take account of the changing gas angles is called vortex
blading. As discussed in axial flow compressor (Chapter 5) the momentum
equation is
1 dP C 2w
¼ ð7:27Þ
r dr r
For constant enthalpy and entropy, the equation takes the form
dh0 dCa dCw C 2w
¼ Ca þ Cw þ ð7:28Þ
dr dr dr r
For constant stagnation enthalpy across the annulus (dh0/dr ¼ 0) and constant
axial velocity (dCa/dr ¼ 0) then the whirl component of velocity Cw is
inversely proportional to the radius and radial equilibrium is satisfied. That is,
C w £ r ¼ constant ð7:29Þ
The flow, which follows Eq. (7.29), is called a “free vortex.”
Now using subscript m to denote condition at mean diameter, the free
vortex variation of nozzle angle a2 may be found as given below:
C w2 r ¼ rCa2 tan a2 ¼ constant

Ca2 ¼ constant
Therefore a2 at any radius r is related to a2m at the mean radius rm by
r 
m
tan a2 ¼ tan a2m ð7:30Þ
r 2
Similarly, a3 at outlet is given by
r 
m
tan a3 ¼ tan a3m ð7:31Þ
r 3

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow and Radial Flow Gas Turbines 293

The gas angles at inlet to the rotor-blade, from velocity triangle,

U
tan b3 ¼ tan a2 2
Ca
r   
r Um
¼
m
tan a2m 2 ð7:32Þ
r 2 r m 2 Ca2
and b3 is given by
r   
m r Um
tan b2 ¼ tan a3m þ ð7:33Þ
r 3 r m 3 Ca3

7.7 CONSTANT NOZZLE ANGLE DESIGN

As before, we assume that the stagnation enthalpy at outlet is constant, that is,
dh0/dr ¼ 0. If a2 is constant, this leads to the axial velocity distribution given by
2
Cw2 r sin a2
¼ constant ð7:34Þ
and since a2 is constant, then Ca2 is proportional to Cw1. Therefore
2
Ca2 r sin a2
¼ constant ð7:35Þ
Normally the change in vortex design has only a small effect on the performance
of the blade while secondary losses may actually increase.

Illustrative Example 7.1 Consider an impulse gas turbine in which gas enters at
pressure ¼ 5.2 bar and leaves at 1.03 bar. The turbine inlet temperature is 1000 K
and isentropic efficiency of the turbine is 0.88. If mass flow rate of air is 28 kg/s,
nozzle angle at outlet is 578, and absolute velocity of gas at inlet is 140 m/s,
determine the gas velocity at nozzle outlet, whirl component at rotor inlet and
turbine work output. Take, g ¼ 1.33, and Cpg ¼ 1.147 kJ/kgK (see Fig. 7.6).

Figure 7.6 T-s diagram for Example 7.1.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 294 Chapter 7

Solution
From isentropic p –T relation for expansion process
0  ðg21Þ/g
T 02 p02
¼
T 01 p01
 ðg21Þ/g  
0 p02 1:03 ð0:248Þ
or T 02 ¼ T 01 ¼ 1000 ¼ 669 K
p01 5:2
Using isentropic efficiency of turbine
 0

T 02 ¼ T 01 2 ht T 01 2 T 02 ¼ 1000 2 0:88ð1000 2 669Þ

¼ 708:72 K
Using steady-flow energy equation
1
2 
C 2 2 C 21 ¼ C p ðT 01 2 T 02 Þ
2
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Therefore, C2 ¼ ½ð2Þð1147Þð1000 2 708:72Þ þ 19600 ¼ 829:33 m/s
From velocity triangle, velocity of whirl at rotor inlet
C w2 ¼ 829:33 sin 578 ¼ 695:5 m/s
Turbine work output is given by
W t ¼ mC pg ðT 01 2 T 02 Þ ¼ ð28Þð1:147Þð1000 2 708:72Þ

¼ 9354:8 kW

Design Example 7.2 In a single-stage gas turbine, gas enters and leaves in axial
direction. The nozzle efflux angle is 688, the stagnation temperature and
stagnation pressure at stage inlet are 8008C and 4 bar, respectively. The exhaust
static pressure is 1 bar, total-to-static efficiency is 0.85, and mean blade speed is
480 m/s, determine (1) the work done, (2) the axial velocity which is constant
through the stage, (3) the total-to-total efficiency, and (4) the degree of reaction.
Assume g ¼ 1.33, and Cpg ¼ 1.147 kJ/kgK.
Solution
(1) The specific work output
W ¼ C pg ðT 01 2 T 03 Þ
¼ hts C pg T 01 1 2 ð1/4Þ0:33/1:33
¼ ð0:85Þð1:147Þð1073Þ 1 2 ð0:25Þ0:248 ¼ 304:42 kJ/kg

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow and Radial Flow Gas Turbines 295

(2) Since a1 ¼ 0, a3 ¼ 0, Cw1 ¼ 0 and specific work output is given by

W 304:42 £ 1000
W ¼ UC w2 or C w2 ¼ ¼ ¼ 634:21 m/s
U 480
From velocity triangle
Cw2
sin a2 ¼
C2
or
C w2 634:21
C2 ¼ ¼ ¼ 684 m/s
sin a2 sin 688
Axial velocity is given by
Ca2 ¼ 684 cos 688 ¼ 256:23 m/s
(3) Total-to-total efficiency, htt, is
T 01 2 T 03
htt ¼
T 01 2 T 003
w ws
¼  s ¼
C 23
T 01 2 T 3 þ 2Cpg w s C2
2 3
hts 2C pg
304:42
¼ ¼ 92:4%
304:42 ð256:23Þ2
2
0:85 2 £ 1147

(4) The degree of reaction

Ca

L¼ tan b3 2 tan b2
2U
     
Ca U Ca U Ca
¼ £ 2 tan a2 þ £
2U Ca 2U Ca 2U
(from velocity triangle)
Ca 256:23
L¼12 tan a2 ¼ 1 2 tan 688 ¼ 33:94%
2U ð2Þð480Þ
Design Example 7.3 In a single-stage axial flow gas turbine gas enters at
stagnation temperature of 1100 K and stagnation pressure of 5 bar. Axial velocity
is constant through the stage and equal to 250 m/s. Mean blade speed is 350 m/s.
Mass flow rate of gas is 15 kg/s and assume equal inlet and outlet velocities.
Nozzle efflux angle is 638, stage exit swirl angle equal to 98. Determine the rotor-
blade gas angles, degree of reaction, and power output.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 296 Chapter 7

Figure 7.7 Velocity triangles for Example 7.3.

Solution
Refer to Fig. 7.7.
Ca1 ¼ Ca2 ¼ Ca3 ¼ Ca ¼ 250 m/s
From velocity triangle (b)
Ca2 250
C2 ¼ ¼ ¼ 550:67 m/s
cos a2 cos 638
From figure (c)
Ca3 250
C3 ¼ ¼ ¼ 253 m/s
cos a3 cos 98
C w3 ¼ Ca3 tan a3 ¼ 250 tan 98 ¼ 39:596 m/s
U þ C w3 350 þ 39:596
tan b3 ¼ ¼ ¼ 1:5584
Ca3 250
i:e:; b3 ¼ 57:318
From figure (b)
C w2 ¼ Ca2 tan a2 ¼ 250 tan 638 ¼ 490:65 m/s

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow and Radial Flow Gas Turbines 297

and
Cw2 2 U 490:65 2 350
tan b2 ¼ ¼ ¼ 0:5626
Ca2 250
0
[ b2 ¼ 29821
Power output
W ¼ mUCað tan b2 þ tan b3 Þ

¼ ð15Þð350Þð250Þð0:5626 þ 1:5584Þ/1000

¼ 2784 kW
The degree of reaction is given by
Ca

L¼ tan b3 2 tan b2
2U
250
¼ ð1:5584 2 0:5626Þ
2 £ 350

¼ 35:56%

Design Example 7.4 Calculate the nozzle throat area for the same data as in the
precious question, assuming nozzle loss coefficient, TN ¼ 0.05. Take g ¼ 1.333,
and Cpg ¼ 1.147 kJ/kgK.
Solution
Nozzle throat area, A ¼ m/r2Ca2
p2
and r2 ¼
RT 2
C22 ð550:67Þ2
T 2 ¼ T 02 2 ¼ 1100 2 ðT 01 ¼ T 02 Þ
2Cp ð2Þð1:147Þð1000Þ
i.e., T 2 ¼ 967:81 K
From nozzle loss coefficient
0 C22 0:05 £ ð550:67Þ2
T 2 ¼ T 2 2 lN ¼ 967:81 2 ¼ 961:2 K
2Cp ð2Þð1:147Þð1000Þ
Using isentropic p –T relation for nozzle expansion
 0
g/ðg21Þ
p2 ¼ p01 = T 01 /T 2 ¼ 5/ð1100/961:2Þ4 ¼ 2:915 bar

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 298 Chapter 7

Critical pressure ratio

   
g þ 1 g/ðg21Þ 2:333 4
p01 /pc ¼ ¼ ¼ 1:852
2 2

or p01 /p2 ¼ 5/2:915 ¼ 1:715

Since pp012 , pp01c , and therefore nozzle is unchoked.

Hence nozzle gas velocity at nozzle exit
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C2 ¼ 2C pg ðT 01 2 T 2 Þ
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
¼ ½ð2Þð1:147Þð1000Þð1100 2 967:81Þ ¼ 550:68 m/s

Therefore, nozzle throat area

m p2 ð2:915Þð102 Þ
A¼ ; and r2 ¼ ¼ ¼ 1:05 kg/m 3
r2 C 2 RT 2 ð0:287Þð967:81Þ

Thus

15
A¼ ¼ 0:026 m 2
ð1:05Þð550:68Þ

Design Example 7.5 In a single-stage turbine, gas enters and leaves the turbine
axially. Inlet stagnation temperature is 1000 K, and pressure ratio is 1.8 bar. Gas
leaving the stage with velocity 270 m/s and blade speed at root is 290 m/s. Stage
isentropic efficiency is 0.85 and degree of reaction is zero. Find the nozzle efflux
angle and blade inlet angle at the root radius.
Solution
Since L ¼ 0, therefore
T2 2 T3
L¼
T 1 2 T 3;
hence
T2 ¼ T3
From isentropic p –T relation for expansion
T 01 1000
T 003 ¼
ðg21Þ/g ¼ ¼ 863:558 K
p01 /p03 ð1:8Þ0:249

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow and Radial Flow Gas Turbines 299

Using turbine efficiency


T 03 ¼ T 01 2 ht T 01 2 T 003

¼ 1000 2 0:85ð1000 2 863:558Þ ¼ 884 K

In order to find static temperature at turbine outlet, using static and
stagnation temperature relation

C 23 2702
T 3 ¼ T 03 2 ¼ 884 2 ¼ 852 K ¼ T 2
2C pg ð2Þð1:147Þð1000Þ

Dynamic temperature

C 22
¼ 1000 2 T 2 ¼ 1000 2 852 ¼ 148 K
2Cpg
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C2 ¼ ½ð2Þð1:147Þð148Þð1000Þ ¼ 582:677 m/s

Since, C pg DT os ¼ U ðC w3 þ C w2 Þ ¼ UC w2 ðC w3 ¼ 0Þ
ð1:147Þð1000Þð1000 2 884Þ
Therefore, C w2 ¼ ¼ 458:8 m/s
290
From velocity triangle

C w2 458:8
sin a2 ¼ ¼ ¼ 0:787
C2 582:677

That is, a2 ¼ 518540

Cw2 2 U 458:8 2 290

tan b2 ¼ ¼
Ca2 C2 cos a2
458:8 2 290
¼ ¼ 0:47
582:677 cos 51:908
i.e., b2 ¼ 25890

Design Example 7.6 In a single-stage axial flow gas turbine, gas enters the turbine
at a stagnation temperature and pressure of 1150 K and 8 bar, respectively.
Isentropic efficiency of stage is equal to 0.88, mean blade speed is 300 m/s, and
rotational speed is 240 rps. The gas leaves the stage with velocity 390 m/s.
Assuming inlet and outlet velocities are same and axial, find the blade height at the
outlet conditions when the mass flow of gas is 34 kg/s, and temperature drop in the
stage is 145 K.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 300 Chapter 7

Solution
Annulus area A is given by

A ¼ 2 pr m h

where h ¼ blade height

r m ¼ mean radius
As we have to find the blade height from the outlet conditions, in this case
annulus area is A3.
A3
[ h¼
2 pr m
U m ¼ pDm N

ðU m Þ 300
or Dm ¼ ¼ ¼ 0:398
pN ð pÞð240Þ
i.e., r m ¼ 0:199 m
Temperature drop in the stage is given by
T 01 2 T 03 ¼ 145 K
Hence T 03 ¼ 1150 2 145 ¼ 1005 K

C23 3902
T 3 ¼ T 03 2 ¼ 1005 2 ¼ 938:697 K
2C pg ð2Þð1:147Þð1000Þ
Using turbine efficiency to find isentropic temperature drop
145
T 003 ¼ 1150 2 ¼ 985:23 K
0:88
Using isentropic p– T relation for expansion process
p01 8 8
p03 ¼
g/ðg21Þ ¼ ¼
0
T 01 /T 03 ð1150/985:23 Þ 4 1:856

i.e., p03 ¼ 4:31 bar

Also from isentropic relation
p03 4:31 4:31
p3 ¼
 ¼ ¼ ¼ 3:55 bar
T 0 /T g/ðg21Þ ð985:23/938:697Þ4 1:214
03 3
p3 ð3:55Þð100Þ
r3 ¼ ¼ ¼ 1:32 kg/m 3
RT 3 ð0:287Þð938:697Þ
m 34
A3 ¼ ¼ ¼ 0:066 m 2
r3 Ca3 ð1:32Þð390Þ

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow and Radial Flow Gas Turbines 301

Finally,
A3 0:066
h¼ ¼ ¼ 0:053 m
2 pr m ð2 pÞð0:199Þ
Design Example 7.7 The following data refer to a single-stage axial flow gas
turbine with convergent nozzle:
Inlet stagnation temperature; T 01 1100 K
Inlet stagnation pressure; p01 4 bar
Pressure ratio; p01 /p03 1:9
Stagnation temperature drop 145 K
Mean blade speed 345 m/s
Mass flow; m 24 kg/s
Rotational speed 14; 500 rpm
Flow coefficient; F 0:75
Angle of gas leaving the stage 128
Cpg ¼ 1147 J/kg K; g ¼ 1:333; lN ¼ 0:05
Assuming the axial velocity remains constant and the gas velocity at inlet and
outlet are the same, determine the following quantities at the mean radius:
(1) The blade loading coefficient and degree of reaction
(2) The gas angles
(3) The nozzle throat area

Solution
C pg ðT 01 2 T 03 Þ ð1147Þð145Þ
ð1Þ C¼ ¼ ¼ 1:4
U2 3452
Using velocity diagram
U/Ca ¼ tan b3 2 tan a3
1
or tan b3 ¼ þ tan a3
F
1
¼ þ tan 128
0:75
b3 ¼ 57:18
From Equations (7.14) and (7.15), we have


C ¼ F tan b2 þ tan b3

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 302 Chapter 7

and
F

L¼ tan b3 2 tan b2
2
From which
1
tan b3 ¼ ðC þ 2LÞ
2F
Therefore
1
tan 57:18 ¼ ð1:4 þ 2LÞ
2 £ 0:75
Hence
L ¼ 0:4595

1
ð2Þ tan b2 ¼ ðC 2 2LÞ
2F
1
¼ ð1:4 2 ½2½0:459Þ
2 £ 0:75
b2 ¼ 17:88
1
tan a2 ¼ tan b2 þ
F
1
¼ tan 17:88 þ ¼ 0:321 þ 1:33 ¼ 1:654
0:75
a2 ¼ 58:88

ð3Þ Ca1 ¼ U F
¼ ð345Þð0:75Þ ¼ 258:75 m/s
Ca1 258:75
C2 ¼ ¼ ¼ 499:49 m/s
cos a2 cos 58:88
C22 499:492
T 02 2 T 2 ¼ ¼ ¼ 108:76 K
2Cp ð2Þð1147Þ
ðT N Þð499:492 Þ ð0:05Þð499:492 Þ
T 2 2 T 2s ¼ ¼ ¼ 5:438 K
ð2Þð1147Þ ð2Þð1147Þ
T 2s ¼ T 2 2 5:438
T 2 ¼ 1100 2 108:76 ¼ 991:24 K
T 2s ¼ 991:24 2 5:438 ¼ 985:8 K
 g=ðg21Þ
p01 T 01
¼
p2 T 2s

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow and Radial Flow Gas Turbines 303

 4
985:8
p2 ¼ 4 £ ¼ 2:58
1100
p2 ð2:58Þð100Þ
r2 ¼ ¼ ¼ 0:911 kg /m 3
RT 2 ð0:287Þð991:24Þ

m 24
ð4Þ Nozzle throat area ¼ ¼ ¼ 0:053 m 2
r1 C 1 ð0:907Þð499:49Þ
m 24
A1 ¼ ¼ ¼ 0:102 m 2
r1 Ca1 ð0:907Þð258:75Þ
Design Example 7.8 A single-stage axial flow gas turbine with equal stage inlet
and outlet velocities has the following design data based on the mean diameter:
Mass flow 20 kg/s
Inlet temperature; T 01 1150K
Inlet pressure 4 bar
Axial flow velocity constant through the stage 255 m/s
Blade speed; U 345 m/s
Nozzle efflux angle; a2 608
Gas-stage exit angle 128

Calculate (1) the rotor-blade gas angles, (2) the degree of reaction, blade-
loading coefficient, and power output and (3) the total nozzle throat area if the
throat is situated at the nozzle outlet and the nozzle loss coefficient is 0.05.
Solution
(1) From the velocity triangles
C w2 ¼ Ca tan a2
¼ 255 tan 608 ¼ 441:67 m/s
C w3 ¼ Ca tan a3 ¼ 255 tan 128 ¼ 55:2 m/s
V w2 ¼ C w2 2 U ¼ 441:67 2 345 ¼ 96:67 m/s
V w2 96:67
b2 ¼ tan 21 ¼ tan 21 ¼ 20:88
Ca 255
Also V w3 ¼ C w3 þ U ¼ 345 þ 55:2 ¼ 400:2 m/s
V w3 400:2
[ b3 ¼ tan 21 ¼ tan 21 ¼ 57:58
Ca 255

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 304 Chapter 7

F

ð2Þ L¼ tan b3 2 tan b2
2
255
¼ ð tan 57:58 2 tan 20:88Þ ¼ 0:44
2 £ 345
Ca

C¼ tan b2 þ tan b3
U
255
¼ ð tan 20:88 þ tan 57:58Þ ¼ 1:44
345

Power W ¼ mU ðCw2 þ Cw3 Þ

¼ ð20Þð345Þð441:67 þ 54:2Þ ¼ 3421:5 kW

0
Cp T 2 2 T 2
ð3Þ lN ¼ 1 2
; C 2 ¼ Ca seca2 ¼ 255sec608 ¼ 510 m/s
2 C2

0 ð0:05Þð0:5Þð5102 Þ
or T2 2 T2 ¼ ¼ 5:67
1147

C22 5102
T 2 ¼ T 02 2 ¼ 1150 2 ¼ 1036:6 K
2Cp ð2Þð1147Þ

0
T 2 ¼ 1036:6 2 5:67 ¼ 1030:93 K

 g=ðg21Þ  
p01 T 01 1150 4
¼ ¼ ¼ 1:548
p2 T2 1030:93

4
p2 ¼ ¼ 2:584 bar
1:548

p2 2:584 £ 100
r2 ¼ ¼ ¼ 0:869 kg/m3
RT 2 0:287 £ 1036:6

m ¼ r2 A 2 C 2

20
A2 ¼ ¼ 0:045 m 2
0:869 £ 510

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow and Radial Flow Gas Turbines 305

Illustrative Example 7.9 A single-stage axial flow gas turbine has the
following data
Mean blade speed 340 m/s
Nozzle exit angle 158
Axial velocity ðconstantÞ 105 m/s
Turbine inlet temperature 9008C
Turbine outlet temperature 6708C
Degree of reaction 50%
Calculate the enthalpy drop per stage and number of stages required.
Solution
At 50%,
a2 ¼ b3
a3 ¼ b2
U 340
C2 ¼ ¼ ¼ 351:99 m/s
cos 158 cos 158
C22 2 C23 ð351:99Þ2 2 ð105Þ2
Heat drop in blade moving row ¼ ¼
2Cp ð2Þð1147Þ
123896:96 2 11025
¼
ð2Þð1147Þ
¼ 49:2 K
Therefore heat drop in a stage ¼ ð2Þð49:2Þ ¼ 98:41 K
1173 2 943 230
Number of stages ¼ ¼ ¼2
98:41 98:4
Design Example 7.10 The following particulars relate to a single-stage turbine of
free vortex design:
Inlet temperature; T 01 1100K
Inlet pressure; p01 4 bar
Mass flow 20 kg/s
Axial velocity at nozzle exit 250 m/s
Blade speed at mean diameter 300 m/s
Nozzle angle at mean diameter 258
Ratio of tip to root radius 1:4

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 306 Chapter 7

The gas leaves the stage in an axial direction, find:

(1) The total throat area of the nozzle.
(2) The nozzle efflux angle at root and tip.
(3) The work done on the turbine blades.
Take
C pg ¼ 1:147 kJ/kg K; g ¼ 1:33

Solution
For no loss up to throat
 g=ðg21Þ  
p* 2 2 4
¼ ¼ ¼ 0:543
p01 gþ1 2:33

p * ¼ 4 £ 0:543 ¼ 2:172 bar

 4
2
Also T * ¼ 1100 ¼ 944 K
2:33
C2
T 01 ¼ T * þ
2C pg

rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 ffi
C* ¼ 2Cpg T 01 2 T *
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
¼ ð2Þð1147Þð1100 2 944Þ ¼ 598 m/s

p* ð2:172Þð100Þ
r* ¼ ¼ ¼ 0:802 kg/m3
RT * ð0:287Þð944Þ

(1) Throat area

m 20
A¼ ¼ ¼ 0:042 m 2
rC * ð0:802Þð598Þ

(2) Angle a1, at any radius r and a1m at the design radius rm are related by
the equation

rm
tan a1 ¼ tan a1m
r1

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow and Radial Flow Gas Turbines 307

Given
Tip radius rt
¼ ¼ 1:4
Root radius r r
Mean radius
[ ¼ 1:2
Root radius
a1m ¼ 258
r mean
tan a1r ¼ £ tan a1m
r root
¼ 1:2 £ tan 258 ¼ 0:5596

[ a1r ¼ 29:238
 
rr 1
tan a1t ¼ £ tan a1r ¼ ð0:5596Þ ¼ 0:3997
rt 1:4

[ a1t ¼ 21:798

rm r m Ca2 250
ð3Þ C w2 ¼ xC w2m ¼ ¼ 1:2x ¼ 643 m/s
rr r r tan a2m tan 258
ð20Þð300Þð643Þ
W ¼ mUC w2 ¼ ¼ 3858 kW
1000

7.8 RADIAL FLOW TURBINE

In Sec. 7.1 “Introduction to Axial Flow Turbines”, it was pointed out that in axial
flow turbines the fluid moves essentially in the axial direction through the rotor.
In the radial type the fluid motion is mostly radial. The mixed flow machine is
characterized by a combination of axial and radial motion of the fluid relative to
the rotor. The choice of turbine depends on the application, though it is not
always clear that any one type is superior. For small mass flows, the radial
machine can be made more efficient than the axial one. The radial turbine is
capable of a high-pressure ratio per stage than the axial one. However, multi-
staging is very much easier to arrange with the axial turbine, so that large overall
pressure ratios are not difficult to obtain with axial turbines. The radial flow
turbines are used in turbochargers for commercial (diesel) engines and fire
pumps. They are very compact, the maximum diameter being about 0.2 m,
and run at very high speeds. In inward flow radial turbine, gas enters in the radial
direction and leaves axially at outlet. The rotor, which is usually manufactured of

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 308 Chapter 7

Figure 7.8 Radial turbine photograph of the rotor on the right.

Figure 7.9 Elements of a 908 inward flow radial gas turbine with inlet nozzle ring.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow and Radial Flow Gas Turbines 309

Figure 7.10 A 908 inward flow radial gas turbine without nozzle ring.

cast nickel alloy, has blades that are curved to change the flow from the radial to
the axial direction. Note that this turbine is like a single-faced centrifugal
compressor with reverse flow. Figures 7.8 –7.10 show photographs of the radial
turbine and its essential parts.

7.9 VELOCITY DIAGRAMS AND

THERMODYNAMIC ANALYSIS
Figure 7.11 shows the velocity triangles for this turbine. The same nomenclature
that we used for axial flow turbines, will be used here. Figure 7.12 shows the
Mollier diagram for a 908 flow radial turbine and diffuser.
As no work is done in the nozzle, we have h01 ¼ h02. The stagnation
pressure drops from p01 to p1 due to irreversibilities. The work done per unit mass
flow is given by Euler’s turbine equation
W t ¼ ðU 2 Cw2 2 U 3 C w3 Þ ð7:36Þ
If the whirl velocity is zero at exit then
W t ¼ U 2 C w2 ð7:37Þ

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 310 Chapter 7

Figure 7.11 Velocity triangles for the 908 inward flow radial gas turbine.

Figure 7.12 Mollier chart for expansion in a 908 inward flow radial gas turbine.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow and Radial Flow Gas Turbines 311

For radial relative velocity at inlet

W t ¼ U 22 ð7:38Þ
In terms of enthalpy drop
h02 2 h03 ¼ U 2 Cw2 2 U 3 Cw3
Using total-to-total efficiency
T 01 2 T 03
htt ¼ ;
T 01 2 T 03 ss
efficiency being in the region of 80– 90%

7.10 SPOUTING VELOCITY

It is that velocity, which has an associated kinetic energy equal to the isentropic
enthalpy drop from turbine inlet stagnation pressure p01 to the final exhaust
pressure. Spouting velocities may be defined depending upon whether total or
static conditions are used in the related efficiency definition and upon whether or
not a diffuser is included with the turbine. Thus, when no diffuser is used, using
subscript 0 for spouting velocity.
1 2
C ¼ h01 2 h03 ss ð7:39Þ
2 0

1 2
or C ¼ h01 2 h3 ss ð7:40Þ
2 0
for the total and static cases, respectively.
Now for isentropic flow throughout work done per unit mass flow
W ¼ U 22 ¼ C20 /2 ð7:41Þ

or U 2 /C 0 ¼ 0:707 ð7:42Þ
U2
In practice, U2/C0 lies in the range 0:68 , , 0:71.
C0

7.11 TURBINE EFFICIENCY

Referring to Fig. 7.12, the total-to-static efficiency, without diffuser, is defined as
h01 2 h03
hts ¼
h01 2 h3 ss
W
¼
W þ 12 C23 þ ðh3 2 h3ss Þ þ ðh3s 2 h3 ss Þ ð7:43Þ

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 312 Chapter 7

Nozzle loss coefficient, jn, is defined as

Enthalpy loss in nozzle
jn ¼
Kinetic energy at nozzle exit
h3s 2 h3 ss
¼ ð7:44Þ
0:5C22 ðT 3 /T 2 Þ
Rotor loss coefficient, jr, is defined as
h3 2 h3s
jr ¼ ð7:45Þ
0:5V 23
But for constant pressure process,
T ds ¼ dh;
and, therefore
h3s 2 h3ss ¼ ðh 2 h2s ÞðT 3 /T 2 Þ
Substituting in Eq. (7.43)
21
1
2 
hts ¼ 1 þ C þ V 23 jr þ C 2 jn T 3 /T 2 W ð7:46Þ
2 3
Using velocity triangles
C 2 ¼ U 2 cosec a2 ; V 3 ¼ U 3 cosec b3 ; C 3 ¼ U 3 cot b3 ; W ¼ U 22

Substituting all those values in Eq. (7.44) and noting that U3 ¼ U2 r3/r2, then
" (  2 )#21
1 T3 r3

hts ¼ 1 þ jn cosec a2 þ
2
jr cosec b3 þ cot b3
2 2
2 T2 r2
ð7:47Þ
Taking mean radius, that is,
1
r 3 ¼ ðr 3t þ r 3h Þ
2
Using thermodynamic relation for T3/T2, we get
  "  2 #
T3 1
 U2 2 r3
¼12 g21 1 2 cot a2 þ
2 2
cot b3
T2 2 a2 r2
But the above value of T3/T2 is very small, and therefore usually neglected. Thus
" (   )#21
1 r 3 av 2

hts ¼ 1 þ jn cosec a2 þ
2
jr cosec b3 av þ cot b3 av
2 2
2 r2
ð7:48Þ

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow and Radial Flow Gas Turbines 313

Equation (7.46) is normally used to determine total-to-static efficiency. The hts

can also be found by rewriting Eq. (7.43) as
h01 2 h03 ðh01 2 h3 ss Þ 2 ðh03 2 h3 Þ 2 ðh3 2 h3s Þ 2 ðh3s 2 h3 ss Þ
hts ¼ ¼
h01 2 h3 ss ðh01 2 h3ss Þ

2  2
¼ 1 2 C3 þ jn C 2 þ jr V 3 /C 0
2 2 ð7:49Þ
where spouting velocity C0 is given by
1 h
g21=g i
h01 2 h3 ss ¼ C20 ¼ C p T 01 1 2 p3 /p01 ð7:50Þ
2
The relationship between hts and htt can be obtained as follows:
W ¼ U 22 ¼ hts W ts ¼ hts ðh01 2 h3ss Þ; then
W 1
htt ¼ ¼
W ts 2 12 C23 1 C2
2 2W3
hts
 2
1 1 C2 1 1 r 3 av
[ ¼ 2 3 ¼ 2 2 cot b3 av ð7:51Þ
htt hts 2W hts 2 r2
Loss coefficients usually lie in the following range for 908 inward flow turbines
jn ¼ 0:063–0:235
and
jr ¼ 0:384–0:777

7.12 APPLICATION OF SPECIFIC SPEED

We have already discussed the concept of specific speed Ns in Chapter 1 and
some applications of it have been made already. The concept of specific speed
was applied almost exclusively to incompressible flow machines as an important
parameter in the selection of the optimum type and size of unit. The volume flow
rate through hydraulic machines remains constant. But in radial flow gas turbine,
volume flow rate changes significantly, and this change must be taken into
account. According to Balje, one suggested value of volume flow rate is that at
the outlet Q3.
Using nondimensional form of specific speed
NQ1/2
Ns ¼ 0
3
ð7:52Þ
ðDh0 Þ3/4
where N is in rev/s, Q3 is in m3/s and isentropic total-to-total enthalpy drop
(from turbine inlet to outlet) is in J/kg. For the 908 inward flow radial turbine,

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 314 Chapter 7

U 2 ¼ pND2 and Dh0s ¼ 12 C 20 ; factorizing the Eq. (7.52)

  1/2
Q1/2 U2 U2
N s ¼
3 3/4
1 2
C pD2 pND2
2 0
pffiffiffi3/2  3/2  
2 U2 Q3 1/2
¼ ð7:53Þ
p C0 ND32

For 908 inward flow radial turbine, U 2 /C0 ¼ p1ffiffi2 ¼ 0:707; substituting this value
in Eq. (7.53),
 
Q3 1/2
N s ¼ 0:18 ; rev ð7:54Þ
ND32
Equation (7.54) shows that specific speed is directly proportional to the square
root of the volumetric flow coefficient. Assuming a uniform axial velocity at rotor
exit C3, so that Q3 ¼ A3 C 3 , rotor disc area Ad ¼ pD22 /4, then
pffiffiffi
C0 2
N ¼ U 2 /ð pD2 Þ ¼
2 pD2
Q3 A3 C3 2 pD2 A3 C3 p2
¼ pffiffiffi ¼ pffiffiffi
ND32 2C 0 D22 Ad C 0 2 2
Therefore,

1  1
C 3 2 A3 2
N s ¼ 0:336 ; rev ð7:55Þ
C0 Ad
 12  12
C3 A3
¼ 2:11 ; rad ð7:56Þ
C0 Ad

Suggested values for C3/Co and A3/Ad are as follows:

0:04 , C 3 /C0 , 0:3

0:1 , A3 /Ad , 0:5

Then 0:3 , N s , 1:1; rad
Thus the Ns range is very small and Fig. 7.13 shows the variation of efficiency with
Ns, where it is seen to match the axial flow gas turbine over the limited range of Ns.

Design Example 7.11 A small inward radial flow gas turbine operates at its
design point with a total-to-total efficiency of 0.90. The stagnation pressure and
temperature of the gas at nozzle inlet are 310 kPa and 1145K respectively. The
flow leaving the turbine is diffused to a pressure of 100 kPa and the velocity of

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow and Radial Flow Gas Turbines 315

Figure 7.13 Variation of efficiency with dimensionless specific speed.

flow is negligible at that point. Given that the Mach number at exit from the
nozzles is 0.9, find the impeller tip speed and the flow angle at the nozzle exit.
Assume that the gas enters the impeller radially and there is no whirl at the
impeller exit. Take
Cp ¼ 1:147 kJ/kg K; g ¼ 1:333:
Solution
The overall efficiency of turbine from nozzle inlet to diffuser outlet is
given by
T 01 2 T 03
htt ¼
T 01 2 T 03 ss
Turbine work per unit mass flow
W ¼ U 22 ¼ Cp ðT 01 2 T 03 Þ; ðCw3 ¼ 0Þ
Now using isentropic p –T relation
  "  g21=g #
T 03ss p03
T 01 1 2 ¼ T 01 1 2
T 01 p01

Therefore
"   #
p03 g21=g
U 22 ¼ htt C p T 01 1 2
p01
"   #
100 0:2498
¼ 0:9 £ 1147 £ 1145 1 2
310

[ Impeller tip speed, U2 ¼ 539.45 m/s

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 316 Chapter 7

The Mach number of the absolute flow velocity at nozzle exit is given by
C1 U1
M¼ ¼
a1 a1 sin a1

Since the flow is adiabatic across the nozzle, we have

C22 U 22
T 01 ¼ T 02 ¼ T 2 þ ¼ T2 þ
2Cp 2Cp sin 2 a2
T2 U 22 gR
or ¼12 ; but Cp ¼
T 01 2C p T 01 sin 2 a2 g21



T2 U 22 g 2 1 U 22 g 2 1
[ ¼12 ¼12 2
T 2gRT 01 sin 2 a2 2a01 sin 2 a2
 01 2
But T2 a2 a2
¼ ¼ since T 01 ¼ T 02
T 01 a01 a02
and a2 U2
¼
a02 M 2 a02 sin a2
 2

U2 U 22 g 2 1
[ ¼12 2
M 2 a02 sin a2 2a02 sin 2 a2
 2 
 
U2 g21 1
and 1¼ þ 2
a02 sin a2 2 M2
 2 
 
U2 g 2 1 1
or sin 2 a2 ¼ þ 2
a02 2 M2
But a202 ¼ gRT 02 ¼ ð1:333Þð287Þð1145Þ ¼ 438043 m 2 / s 2
 
539:452 0:333 1
[ sin 2 a2 ¼ þ 2 ¼ 0:9311
438043 2 0:9
Therefore nozzle angle a2 ¼ 758

Illustrative Example 7.12 The following particulars relate to a small inward

flow radial gas turbine.
Rotor inlet tip diameter 92 mm
Rotor outlet tip diameter 64 mm
Rotor outlet hub diameter 26 mm
Ratio C 3 /C0 0:447

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow and Radial Flow Gas Turbines 317

Ratio U 2 /C 0 ðidealÞ 0:707

Blade rotational speed 30; 500 rpm
Density at impeller exit 1:75 kg/m3
Determine
(1) The dimensionless specific speed of the turbine.
(2) The volume flow rate at impeller outlet.
(3) The power developed by the turbine.

Solution
(1) Dimensionless specific speed is
 12  12
C3 A3
N s ¼ 0:336 ; rev
C0 Ad
Now


p D23t 2 D23h
A3 ¼
4


p 0:0642 2 0:0262
¼ ¼ ð2:73Þð1023 Þ m 2
4
pD22  p
Ad ¼ ¼ ð0:0922 Þ ¼ ð6:65Þð1023 Þ m 2
4 4
Dimensionless specific speed
 1
½0:447½2:73 2
N s ¼ 0:336
6:65
¼ 0:144 rev
¼ 0:904 rad
(2) The flow rate at outlet for the ideal turbine is given by Eq. (7.54).
 
Q3 1=2
N s ¼ 0:18
ND32
 1=2
½Q3 ½60
0:144 ¼ 0:18
½30; 500½0:0923 

Hence

Q3 ¼ 0:253 m 3 /s

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 318 Chapter 7

(3) The power developed by the turbine is given by

W t ¼ mU
_ 2
3

¼ r3 Q3 U 23
 2
pND2
¼ 1:75 £ 0:253 £
60
 
½ p½30; 500½0:092 2
¼ 1:75 £ 0:253 £
60
¼ 9:565 kW

PROBLEMS
7.1 A single-stage axial flow gas turbine has the following data:
Inlet stagnation temperature 1100K
The ratio of static pressure at the
nozzle exit to the stagnation
pressure at the nozzle inlet 0:53
Nozzle efficiency 0:93
Nozzle angle 208
Mean blade velocity 454 m/s
Rotor efficiency 0:90
Degree of reaction 50%

C pg ¼ 1:147 kJ/kgK; g ¼ 1:33

Find (1) the work output per kg/s of air flow, (2) the ratio of the static
pressure at the rotor exit to the stagnation pressure at the nozzle inlet,
and (3) the total-to-total stage efficiency.
(282 kW, 0.214, 83.78%)
7.2 Derive an equation for the degree of reaction for a single-stage axial flow
turbine and show that for 50% reaction blading a2 ¼ b3 and a3 ¼ b2.
7.3 For a free-vortex turbine blade with an impulse hub show that degree of
reaction r 2
h
L¼12
r
where rh is the hub radius and r is any radius.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 Axial Flow and Radial Flow Gas Turbines 319

7.4 A 50% reaction axial flow gas turbine has a total enthalpy drop of 288 kJ/kg.
The nozzle exit angle is 708. The inlet angle to the rotating blade row is
inclined at 208 with the axial direction. The axial velocity is constant through
the stage. Calculate the enthalpy drop per row of moving blades and the
number of stages required when mean blade speed is 310 m/s. Take Cpg ¼
1:147 kJ/kgK; g ¼ 1:33:
(5 stages)
7.5 Show that for zero degree of reaction, blade-loading coefficient, C ¼ 2.
7.6 The inlet stagnation temperature and pressure for an axial flow gas turbine
are 1000K and 8 bar, respectively. The exhaust gas pressure is 1.2 bar and
isentropic efficiency of turbine is 85%. Assume gas is air, find the exhaust
stagnation temperature and entropy change of the gas.
(644K, 2 0.044 kJ/kgK)
7.7 The performance date from inward radial flow exhaust gas turbine are as
follows:

Stagnation pressure at inlet to nozzles; p01 705 kPa

Stagnation temperature at inlet to nozzles; T 01 1080K
Static pressure at exit from nozzles; p2 515 kPa
Static temperature at exit from nozzles; T 2 1000K
Static pressure at exit from rotor; p3 360 kPa
Static temperature at exit from rotor; T 3 923K
Stagnation temperature at exit from rotor; T 03 925K
r2 av
Ratio r2 0:5
Rotational speed; N 25; 500 rpm

The flow into the rotor is radial and at exit the flow is axial at all radii.
Calculate (1) the total-to-static efficiency of the turbine, (2) the impeller tip
diameter, (3) the enthalpy loss coefficient for the nozzle and rotor rows, (4)
the blade outlet angle at the mean diameter, and (5) the total-to-total
efficiency of the turbine.
[(1) 93%, (2) 0.32 m, (3) 0.019, 0.399, (4) 72.28, (5) 94%]

NOTATION
A area
C absolute velocity

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved

 320 Chapter 7

C0 spouting velocity
h enthalpy, blade height
N rotation speed
Ns specific speed
P pressure
rm mean radius
T temperature
U rotor speed
V relative velocity
YN nozzle loss coefficient in terms of pressure
a angle with absolute velocity
b angle with relative velocity
DT0s stagnation temperature drop in the stage
DTs static temperature drop in the stage
1n nozzle loss coefficient in radial flow turbine
1r rotor loss coefficient in radial flow turbine
f flow coefficient
hs isentropic efficiency of stage
L degree of reaction

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved