Hydropower: 6-1 Introduction

CHAPTER 6
Hydropower
6-1 INTRODUCTION
Turbines have been used for centuries to convert freely available mechanical energy from
rivers and water bodies into useful mechanical work, usually through a rotating shaft.
The rotating part of a hydro turbine is called the runner. When the working fluid is water,
the turbomachines are called hydraulic turbines or hydro turbines. Large dams are built
in the flow path of rivers to collect water. The water having potential energy is run through
turbines to produce electricity. Such an installation is called a hydroelectric power plant.
Some dams are also used for irrigation of farms and flood control. The large dam takes a
long time and a large amount of investment to build but the cost of producing electricity by
hydropower is much lower than the cost of electricity production by fossil fuels.
The first hydraulic turbine-generator set was developed in 1882. It used a 10-m-high
water reservoir and provided a power of 6 kW to light bulbs. This is followed by the first
known hydroelectric power plant in 1883 in Northern Ireland. This plant had two turbines
with a power rating of 39 kW. An electric train was powered by this plant. In the same year,
two generators powered by waterwheels were installed in Diamantina, Brazil. The water
source was 5-m-high and 12 kW power were produced (Farret and Simoes, 2006). A good
review of the history of hydropower can be found in Lewis et al. (2014).
Most of the world’s renewable electricity production is due to hydropower with a global
installed capacity of over 1250 GW as of 2017. Hydropower accounts for 71 percent of
renewable power and it generated 17 percent of world’s electricity considering all energy
sources including fossil fuels. The installed capacity of hydropower is 341 GW in China,
103 GW in the United States, 100 GW in Brazil, and 81 GW in Canada. The United States
was generating 40 percent of its electricity from hydropower in 1930. This is dropped to
6.5 percent in 2017. Canada, Norway, Turkey, Nepal, and New Zealand produce 50 percent
or more of their electricity by hydropower (IEA, 2018).
Most large hydroelectric power plants have several turbines arranged in parallel. This
offers the power company the opportunity to turn off some of the turbines during times
of low power demand and for maintenance. Hoover Dam in Boulder City, Nevada, for
example, has 17 parallel turbines, 15 of which are identical large Francis turbines that can
produce approximately 130 MW of electricity each (Fig. 6-1). The maximum gross head is
590 ft (180 m). The total peak power production of the power plant exceeds 2000 MW while
about 4 billion kWh electricity is produced every year.
Building a hydroelectric power plant with an accumulation reservoir is expensive and
takes a long time to build. It also involves complex hydrological and topographical studies for
163
06_Kanoglu_Ch06_p163-194.indd 163 04/06/19 3:20 PM

164 Chapter 6. Hydropower
(a)
Figure 6-1 (a) An aerial view of Hoover

Dam. ( Corbis RF.) (b) The top (visible)
portion of several of the parallel electric
generators driven by hydraulic turbines at
(b) Hoover Dam. ( Brand X Pictures RF.)

6-2 Analysis of a Hydroelectric Power Plant 165
a careful assessment of site and plant characteristics. The size of the plant including the dam is
large, and environmental impacts of the plant need serious considerations and studies. Unlike
run-of-river plants, a plant with a dam can use most of the collected water for power produc-
tion. It should be noted that the water reservoir of a hydroelectric power plant is also used for
other purposes such as irrigation, fish farming, and recreation. For some hydroelectric instal-
lations, irrigation is more important than or as important as power production.
The power house of a hydropower plant has hydraulic turbines and generators. A regu-
lator, a water admission valve, an electrical command board, and an inertial flywheel are
among the supplemental equipment. A hydroelectric power plant project involves consid-
erable amount of civil works. This may include the construction of a dam, water intake
system for diversion of river from its normal course, a spillway to return water to its normal
course, various piping, a balance chimney, and a machine house. There is less amount of
construction involved in river plants. As a result, river plants can be built in a relatively
short time; they occupy less space; they have less environmental impacts; and they involve
lower initial cost compared to large hydropower plants.
In hydroelectric power plants, large dynamic turbines are used to produce electricity.
There are two basic types of dynamic turbine—impulse and reaction. Comparing the two
power-producing dynamic turbines, impulse turbines require a higher head, but can oper-
ate with a smaller volume flow rate. Reaction turbines can operate with much less head, but
require a higher volume flow rate. Before discussing characteristics of impulse and reac-
tions turbines, we provide general analysis of a hydroelectric power plant.
Much of the material in this chapter is taken from Çengel and Cimbala (2018), and the
readers are referred to this title for additional information and analysis.
6-2 ANALYSIS OF A HYDROELECTRIC POWER PLANT

Mechanical energy can be defined as the form of energy that can be converted to mechani-
cal work completely and directly by an ideal mechanical device such as an ideal turbine. The
mechanical energy of a flowing fluid can be expressed on a unit mass basis as
P V2
emech = + + gz (6-1)
ρ 2
where P/r is the flow energy, V2/2 is the kinetic energy, and gz is the potential energy of the
fluid, all per unit mass. Then the mechanical energy change of a fluid during incompressible
flow becomes
P − P V 2 − V12
∆emech = 2 1 + 2 + g (z 2 − z1 ) (6-2)
ρ 2
Therefore, the mechanical energy of a fluid does not change during flow if its pressure,
density, velocity, and elevation remain constant. In the absence of any irreversible losses,
the mechanical energy change represents the mechanical work supplied to the fluid (if
Demech > 0) or extracted from the fluid (if Demech < 0). The maximum (ideal) power generated
by a turbine, for example, is
W max = m ∆emech
where m is the mass flow rate of the fluid, as shown in Fig. 6-2. If we take point 1 as the
surface of the water reservoir and point 4 as the surface of discharge water from the turbine
(Fig. 6-2a), the maximum power that can be generated by the turbine is
W max = m ∆emech = mg
 (z1 − z 4 ) = mgh
 (6-3)

since P1 ≈ P4 = Patm and V1 = V4 ≈ 0
If we take point 2 as the turbine inlet and point 3 as the turbine outlet (Fig. 6-2a), the maxi-
mum power that can be generated by the turbine is
P − P3 ∆P
W max = m ∆emech = m 2 = m (6-4)
ρ ρ
since V2 ≈ V3 and z 2 ≈ z 3
In fluid systems, we are usually interested in increasing the pressure, velocity, and/or
elevation of a fluid. This is done by supplying mechanical energy to the fluid by a pump, a fan,
or a compressor. Or we are interested in the reverse process of extracting mechanical energy
from a fluid by a turbine and producing mechanical power in the form of a rotating shaft
that can drive a generator or any other rotary device. The degree of perfection of the conver-
sion process between the mechanical work extracted and the mechanical energy change of
the fluid is expressed by the turbine efficiency. In rate form, it is defined as
W shaft W shaft W W
ηturbine =  = = shaft =  shaft (6-5)
∆Emech,fluid m ∆emech mgh
 Wmax
where W shaft is the shaft power output from the turbine and m ∆emech is rate of decrease in the
mechanical energy of the fluid, which is equal to maximum power W max = mgh  , based on the
notation in Fig. 6-2a. We use the positive value for the mechanical energy change to avoid
negative values for efficiencies. A turbine efficiency of 100 percent indicates perfect conver-
sion between the shaft work and the mechanical energy of the fluid, and this value can be
approached (but never attained) as the frictional effects are minimized.
The mechanical efficiency of a turbine should not be confused with the generator
efficiency, which is defined as
W electric
ηgenerator = (6-6)
W shaft
W Turbine
h
2
W Turbine 4
3
Generator
Generator
(a) (b)
Figure 6-2 Mechanical energy is illustrated by an ideal hydraulic turbine coupled with an ideal generator.
In the absence of irreversible losses, the maximum produced power is proportional to (a) the change in
water surface elevation from the upstream to the downstream reservoir or (b) (close-up view) the drop
in water pressure from just upstream to just downstream of the turbine.

hturbine = 0.75 hgenerator = 0.97
Welect. out
Turbine
Generator
Figure 6-3 The overall efficiency of a turbine-generator is the

hturbine–gen = hturbine hgenerator product of the efficiency of the turbine and the efficiency of
= 0.75 × 0.97 the generator, and represents the fraction of the mechanical
= 0.73 power of the fluid converted to electrical power.
where W electric is the electrical power output from the generator. A turbine is usually
packaged together with its generator. Therefore, we are usually interested in the combined
or overall efficiency of turbine-generator combination (Fig. 6-3), which is defined as
 W   W  W
ηturbine-generator = ηturbineηgenerator =   shaft   electric  = electric (6-7)
 Wmax   Wshaft  Wmax
Most turbines have efficiencies approaching 90 percent. Large hydro turbines achieve over-
all efficiencies above 95 percent.
The analysis of a hydroelectric power plant involves that of the turbine and the penstock,
which is the piping system between the upper and lower water levels, as shown in Fig. 6-4.
The steady-flow energy equation on a unit-mass basis can be written for this penstock/
turbine combination as
emech, in = emech, out + emech, loss (6-8)
P1 V12 P V2
or + + gz1 = 2 + 2 + gz 2 + w turbine + emech, loss (6-9)
ρ1 2 ρ2 2
Penstock
h
.
W Turbine 2
Generator
Figure 6-4 The analysis of a hydroelectric power plant
involves that of the turbine and the penstock.

where wturbine is the mechanical work output (due to a turbine). When the flow is incom-
pressible, either absolute or gage pressure can be used for P since Patm/r would appear on
both sides and would cancel out. Multiplying Eq. (6-9) by the mass flow rate m gives
 P V2   P V2 
m  1 + 1 + gz1 = m  2 + 2 + gz 2  + W turbine + E mech, loss, total (6-10)
ρ
 1 2  ρ
 2 2 
where Wturbine is the shaft power output through the turbine’s shaft, and E mech, loss, total is the

total mechanical power loss, which consists of turbine losses as well as the frictional losses
in the piping network. That is,
E mech, loss, total = E mech, loss, turbine + E mech, loss, piping (6-11)
By convention, irreversible turbine losses are treated separately from irreversible losses due
to other components of the piping system. Thus, the energy equation is expressed in its most
common form in terms of heads by dividing each term in Eq. (6-10) by mg  . The result is
P1 V2 P V2
+ 1 + z1 = 2 + 2 + z 2 + hturbine,e + hL (6-12)
ρ1 g 2 g ρ2 g 2 g
w turbine,e W turbine,e W turbine

where hturbine,e = = = is the extracted head removed from the
g mg ηturbinemg
fluid by the turbine. Because of irreversible losses in the turbine, hturbine,e is greater than
W turbine /mg
 by the factor hturbine.
emech, loss, piping E mech, loss, piping

hL = = is the irreversible head loss between 1 and 2 due to
g mg
all components of the piping system other than the turbine.
Note that the head loss hL represents the frictional losses associated with fluid flow in
penstock, and it does not include the losses that occur within the turbine due to the inef-
ficiencies of this device—these losses are taken into account by hturbine. The total head loss in
the penstock is determined from
hL , total = hL , major + hL , minor = f

L V2
D 2g
+ ∑ KL
V2
2g (
L
= f + ∑ KL
D
V2
2g )
(6-13)
where f =Darcy friction factor. It can be determined from the Moody chart (or Colebrook
equation) for turbulent flow. For laminar flow, f = 64/Re, where Re is the
Reynolds number.
L = length of the penstock
D = diameter of the penstock
V = velocity of water in the penstock
KL = loss coefficient for minor losses in the piping system
The minor loss coefficients for various pipe components are available in Çengel and
Cimbala (2018).
We sketch in Fig. 6-5 a typical hydroelectric dam that utilizes Francis reaction tur-
bines to generate electricity. The overall or gross head Hgross is defined as the elevation
difference between the reservoir surface upstream of the dam and the surface of the water
exiting the dam,
Hgross = zA − zE(6-14)

Dam Gross
head
Head gate EGL in Hgross
(open)
Power station Net

Generator head zA
B H
•
Shaft
Turbine E
Penstock C
Draft tube D EGLout zE
Tailrace
Arbitrary datum plane (z = 0)
Figure 6-5 Typical setup and terminology for a hydroelectric plant that utilizes a Francis turbine to
generate electricity; drawing not to scale. The pitot probes are shown for illustrative purposes only.
If there were no irreversible losses anywhere in the system, the maximum amount of power
that could be generated per turbine would be
W max = ρ gVH gross (6-15)
where r is density, g is gravitational acceleration, and V is volume flow rate. Of course, there
are irreversible losses throughout the system, so the power actually produced is lower than
the maximum power given by Eq. (6-15).
We follow the flow of water through the whole system of Fig. 6-5, defining terms and
discussing losses along the way. We start at point A upstream of the dam where the water is
still, at atmospheric pressure, and at its highest elevation, zA. Water flows at volume flow rate V
through a large tube through the dam called the penstock. Flow to the penstock can be cut-off
by closing a large gate valve called a head gate at the penstock inlet. If we were to insert a pitot
probe at point B at the end of the penstock just before the turbine, as illustrated in Fig. 6-5,
the water in the tube would rise to a column height equal to the energy grade line EGLin at the
inlet of the turbine. Note that the energy grade line (EGL) represents the total head of the fluid
EGL = Pressure head + Velocity head + Elevation head = P/rg + V2/2g + z (6-16)
This column height is lower than the water level at point A, due to irreversible losses in the
penstock and its inlet. The flow then passes through the turbine, which is connected by a
shaft to the electric generator. Note that the electric generator itself has irreversible losses.
From a fluid mechanics perspective, however, we are interested only in the losses through
the turbine and downstream of the turbine.
After passing through the turbine runner, the exiting fluid (point C) still has appre-
ciable kinetic energy (velocity head), and perhaps swirl. To recover some of this kinetic
energy (which would otherwise be wasted), the flow enters an expanding area diffuser
called a draft tube, which turns the flow horizontally and slows down the flow speed,
while increasing the pressure prior to discharge into the downstream water, called the
tailrace. If we were to imagine another pitot probe at point D (the exit of the draft tube),

the water in the tube would rise to a column height equal to the energy grade line
labeled EGLout in Fig. 6-5. Since the draft tube is considered to be an integral part of the
turbine assembly, the net head across the turbine is specified as the difference between
EGLin and EGLout,
Hnet = EGLin− EGLout(6-17)
That is, the net head of a turbine is defined as the difference between the energy grade line
just upstream of the turbine and the energy grade line at the exit of the draft tube.
At the draft tube exit (point D) the flow speed is significantly slower than that at point
C upstream of the draft tube; however, it is finite. All the kinetic energy leaving the draft
tube is dissipated in the tailrace. This represents an irreversible head loss and is the reason
why EGLout is higher than the elevation of the tailrace surface, zE. Nevertheless, significant
pressure recovery occurs in a well-designed draft tube. The draft tube causes the pressure
at the outlet of the runner (point C) to decrease below atmospheric pressure, thereby ena-
bling the turbine to utilize the available head most efficiently. In other words, the draft tube
causes the pressure at the runner outlet to be lower than it would have been without the
draft tube—increasing the change in pressure from the inlet to the outlet of the turbine.
Designers must be careful, however, because sub atmospheric pressures may lead to cavita-
tion, which is undesirable for many reasons, as discussed previously.
If we were interested in the overall efficiency of the entire hydroelectric plant including
penstock flow, we would define this efficiency as the ratio of actual electric power produced
to maximum power, based on gross head. That is,
W electric W electric
ηplant =  = (6-18)
Wmax ρ gVH gross
This is equivalent to turbine-generator efficiency [Eq. (6-7)], defined earlier. Usually, we

are more interested in the efficiency of the turbine itself. The turbine efficiency given in
Eq. (6-5) is based on the gross head. However, by convention, turbine efficiency hturbine
should be based on net head Hnet rather than gross head Hgross. Specifically, hturbine is defined
as the ratio of actual turbine output shaft power to energy power extracted from the water
flowing through the turbine (mechanical power difference between the end of the penstock
and the exit of the draft tube),
W shaft
ηturbine = (6-19)
ρ gVH net
This is a more realistic form of turbine efficiency as the turbine is not held responsible for
irreversible losses in the penstock and its inlet and those after the exit of the draft tube.
When the total irreversible head loss hL in the piping is known, the corresponding power
loss is determined from
E mech,loss,piping = ρ gVhL (6-20)
The turbine efficiency is then expressed as
W shaft W shaft W shaft

ηturbine =  = = (6-21)
   
Wmax − Emech,loss,piping ρ gVH gross − ρ gVhL ρ gV ( H gross − hL )

The effect of irreversible head losses in the piping system can be accounted for using an
efficiency term hpiping as
E mech, loss, piping

ηpiping = 1 − (6-22)
W max
Remembering that the generator efficiency is ηgenerator = W electric /W shaft, the overall efficiency
of a hydroelectric power plant can be expressed as
ηplant = ηgeneratorηturbineηpiping
 W  W shaft   E mech,loss,piping 
=  electric     1 − 
 Wshaft   Wmax − E mech,loss,piping   W max 
 W  W shaft   E mech,loss,piping  (6-23)

=  electric     1 − 
 Wshaft  Wmax (1 − E mech,loss,piping / W max )  W max 
W electric
=
W max
Therefore, the overall efficiency of a hydroelectric power plant is defined as the electrical
power output divided by the maximum power potential, and it can be expressed as the
product of generator, turbine, and piping efficiencies.
EXAMPLE 6-1 River water is collected into a large dam whose water height is 240 ft. How much power can be
Power Potential produced by a set of ideal hydraulic turbines if water is run through the turbine at a total rate of
of a Dam 9000 gpm?
SOLUTION We take the density of water to be r = 62.4 lbm/ft3. The total mechanical energy the
water in a dam possesses is equivalent to the potential energy of water at the free surface of the dam
(relative to free surface of discharge water), and it can be converted to work entirely for an ideal opera-
tion. Therefore, the maximum power that can be generated is equal to potential energy of the water.
First, the total mass flow rate of water is determined as
 1 ft 3  1 min
m = ρV = (62.4 lbm/ft 3 )(9000 gal/min)
 7.4804 gal  60 s( )
= 1251.3 lbm/s
Then the maximum power is determined from
1 kJ   1 kW 
 = (1251.3 lbm/s)(32.2 ft/s 2 )(240 ft)  1 Btu/lbm
W max = mgh 
 25,037 ft 2 /s 2   0.94782 Btu   1 kJ/s 

= 408 kW
This is the maximum power that can be generated. Actual turbines generate less power. ▲
EXAMPLE 6-2 The pressures just upstream and downstream of a hydraulic turbine are measured to be 1400 and
Analysis of a 100 kPa, respectively. The turbine accepts water at the bottom level of a large water reservoir. (a) What
Hydraulic Turbine is the shaft work, in kJ/kg, that can be produced by this turbine? (b) If the efficiency of the turbine is
estimated to be 88 percent, what is the height of this reservoir?

SOLUTION (a) The density of water is taken to be r = 1000 kg/m3. The maximum work per unit
mass of water flow is determined from
P1 − P2 (1400 − 100) kPa  1 kJ 

w max = = = 1.30 kJ/kg
ρ 1000 kg/m 3  1 kPa ⋅ m3 
The actual shaft work per unit mass of water flow is determined using turbine efficiency to be
w shaft = ηturbinew max = (0.88)(1.30 kJ/kg) = 1.144 kJ /kg
(b) The height of the reservoir is determined from
w shaft 1.144 kJ/kg  1000 m 2 /s 2 

w shaft = gh →
 h= = = 117 m ▲
g 9.81 m/s 2  1 kJ/kg 
EXAMPLE 6-3 The water in a large water dam is to be used to generate electricity by the installation of a hydraulic
Analysis of a turbine. The elevation difference between the free surfaces upstream and downstream of the dam is 320 m
Hydroelectric (Fig. 6-6). Water is to be supplied to the turbine at a rate of 8000 L/s. The turbine efficiency is 93 percent
Power Plant based on the net head, and the generator efficiency is 96 percent. The total irreversible head loss (major
losses + minor losses) in the piping system including the penstock is estimated to be 7.5 m. Determine the
overall efficiency of this hydroelectric plant, the electric power produced, and the turbine shaft power.
SOLUTION The density of water is taken to be r = 1000 kg/m3 = 1 kg/L. The total mechanical energy
the water in a dam possesses is equivalent to the potential energy of water at the free surface of the
dam (relative to free surface of discharge water), and it can be converted to work entirely for an ideal
operation. Therefore, the maximum power that can be generated is equal to potential energy of the
water. Noting that the mass flow rate is m = ρV , the maximum power is determined from
 gross = ρVgH gross

W max = mgH
 1 kJ/kg 
= (1 kg/L)(8000 L/s)(9.81 m/s 2 )(320 m) 
 1000 m 2 /s 2 
= 25,115 kW
Penstock
H = 320 m
Turbine 2
.
V = 8000 L/s
Generator
hgenerator = 96%
Figure 6-6 Schematic for Example 6-3.

6-3 Impulse Turbines 173
The mechanical energy loss due to frictional and minor losses in the piping is
 L = ρVghL
E mech,loss,piping = mgh
 1 kJ/kg 
= (1 kg/L)(8000 L/s)(9.81 m/s 2 )(7.5 m) 
 1000 m 2 /s 2 
= 589 kW
Therefore, 589/25,115 = 0.0234 or 2.3 percent of power potential is lost due to piping losses. This is
expressed by the efficiency term hpiping as
 E mech, loss, piping   589 kW 

ηpiping = 1 −
W max  = 1 − 25,115 kW  = 0.977
 
The overall plant efficiency is the product of turbine, generator, and piping efficiencies:
ηplant = ηgeneratorηturbineηpiping = (0.96)(0.93)(0.977) = 0.872 or 87.2%
The electric power produced is determined from the definition of plant efficiency to be
W electric
ηplant =  W electric = ηplantW max = (0.872)(25,115 kW) = 21, 900 kW
→
W max
The definition of turbine efficiency based on the net head is
W shaft
ηturbine = 
Wmax − E mech, loss, piping
Solving for the shaft power output, we obtain
W shaft = ηturbine (W max − E mech, loss, piping )
= (0.93)(25,115 kW − 589 kW)

= 22, 800 kW
This plant has an overall efficiency of 87.2 percent, which is considerably higher than the efficiency of
steam and gas-turbine power plants (typically less than 50 percent). ▲
6-3 IMPULSE TURBINES

In an impulse turbine, the fluid is sent through a nozzle so that most of its available mechan-
ical energy is converted into kinetic energy. The high-speed jet then impinges on bucket-
shaped vanes that transfer energy to the turbine shaft, as sketched in Fig. 6-7. The modern
and most efficient type of impulse turbine was invented by Lester A. Pelton (1829–1908)
in 1878, and the rotating wheel is now called a Pelton wheel in his honor. The buckets of
a Pelton wheel are designed so as to split the flow in half, and turn the flow nearly 180°
around (with respect to a frame of reference moving with the bucket), as illustrated in
Fig. 6-7b. According to legend, Pelton modeled the splitter ridge shape after the nostrils of a
cow’s nose. A portion of the outermost part of each bucket is cut-out so that the majority of
the jet can pass through the bucket that is not aligned with the jet (bucket n + 1 in Fig. 6-7a)

r
w
Shaft
Bucket n + 1
Nozzle Bucket n
Vj rw
(a)
Splitter ridge
Vj – rw Figure 6-7 Schematic diagram of a Pelton-type impulse
turbine; the turbine shaft is turned when high-speed
fluid from one or more jets impinges on buckets
b mounted to the turbine shaft. (a) Side view, absolute
Vj – rw reference frame, and (b) bottom view of a cross section
(b) of bucket n, rotating reference frame.
to reach the most aligned bucket (bucket n in Fig. 6-7a). In this way, the maximum amount
of momentum from the jet is utilized. These details are seen in a photograph of a Pelton
wheel (Fig. 6-8). Figure 6-9 shows a Pelton wheel in operation; the splitting and turning of
the water jet is clearly seen.
Figure 6-8 A close-up view of a Pelton wheel showing the detailed design of the buckets; the electrical
generator is on the right. This Pelton wheel is on display at the Waddamana Power Station Museum
near Bothwell, Tasmania. (Courtesy of Hydro Tasmania, www.hydro.com.au. Used with permission.)

6-3 Impulse Turbines 175
Figure 6-9 A view from the bottom of

an operating Pelton wheel illustrating
the splitting and turning of the water jet
in the bucket. The water jet enters from
the top, and the Pelton wheel is turning
to the right. (Courtesy of Andritz Hydro.
Used with permission.)
We analyze the power output of a Pelton wheel turbine by using the Euler turbomachine
equation. The power output of the shaft is equal to
W shaft = ω Tshaft = ρω V (r2V2,t − r1V1,t ) (6-24)
We must be careful of negative signs since this is an energy-producing rather than an

energy-absorbing device. For turbines, it is conventional to define point 2 as the inlet and
point 1 as the outlet. The center of the bucket moves at tangential velocity rw, as illus-
trated in Fig. 6-7. We simplify the analysis by assuming that since there is an opening in
the outermost part of each bucket, the entire jet strikes the bucket that happens to be at
the direct bottom of the wheel at the instant of time under consideration (bucket n in Fig.
6-7a). Furthermore, since both the size of the bucket and the diameter of the water jet are
small compared to the wheel radius, we approximate r1 and r2 as equal to r. Finally, we
make the approximation that the water is turned through angle b without losing any speed;
in the relative frame of reference moving with the bucket, the relative exit speed is thus
Vj−rw (the same as the relative inlet speed) as sketched in Fig. 6-7b. Returning to the abso-
lute reference frame, which is necessary for the application of Eq. (6-24), the tangential
component of velocity at the inlet, V2,t, is simply the jet speed itself, Vj. We construct a
velocity diagram in Fig. 6-10 as an aid in calculating the tangential component of absolute
velocity at the outlet, V1,t. After some trigonometry, which you can verify after noting that
sin (b −90°) = −cosb,
V1,t = rω + (V j − rω )cos β

Splitter
ridge
Vj – rw
rw
Figure 6-10 Velocity diagram of flow into and out of a
Vj –rw b
Pelton wheel bucket. We translate outflow velocity from
V1, t the moving reference frame to the absolute reference
V1 frame by adding the speed of the bucket (rw) to the right.
Upon substitution of this equation, Eq. (6-24) yields
W shaft = ρ rω V {V j − [rω + (V j − rω )cos β ]}
which simplifies to
Output shaft power: W shaft = ρrω V(V j − rω )(1 − cos β ) (6-25)
Obviously, the maximum power is achieved theoretically if b = 180°. However, if that were
the case, the water exiting one bucket would strike the back side of its neighbor coming
along behind it, reducing the generated torque and power. It turns out that in practice, the
maximum power is achieved by reducing b to around 160° to 165°. The efficiency factor
due to b being less than 180° is
W 1 − cos β
Efficiency factor due to b : ηβ = shaft, actual = (6-26)
Wshaft, ideal 1 − cos(180°)
When b = 160°, for example, hb = 0.97—a loss of only about 3 percent. Finally, we see from
Eq. (6-25) that the shaft power output W shaft is zero if rw = 0 (wheel not turning at all). W shaft
is also zero if rw = Vj (bucket moving at the jet speed). Somewhere in between these two
extremes lies the optimum wheel speed. By setting the derivative of Eq. (6-25) with respect
to rw to zero, we find that this occurs when rw = Vj /2 (bucket moving at half the jet speed,
as shown in Fig. 6-11).
For an actual Pelton wheel turbine, there are other losses besides that reflected in
Eq. (6-26): mechanical friction, aerodynamic drag on the buckets, friction along the inside
walls of the buckets, nonalignment of the jet and bucket as the bucket turns, back splashing,
Vj
w = —–
2r r
Shaft
Nozzle
Figure 6-11 The theoretical maximum power achievable
by a Pelton turbine occurs when the wheel rotates at w =
Vj Vj /(2r), that is, when the bucket moves at half the speed
Vj rw = —–
2 of the water jet.

6-4 Reaction Turbines 177
and nozzle losses. Even so, the efficiency of a well-designed Pelton wheel turbine can
approach 90 percent. In other words, up to 90 percent of the available mechanical energy of
the water is converted to rotating shaft energy.
EXAMPLE 6-4 A Pelton wheel turbine has an efficiency of 84 percent. The average radius of the wheel is 1.22 m,
Flow Rate and and the jet velocity is 87 m/s from a nozzle of exit diameter equal to 6.7 cm. The turning angle of the
Shaft Power Output buckets is b = 160°. Determine (a) the volume flow rate through the turbine and (b) the output shaft
of a Pelton Turbine power.
SOLUTION (a) We assume that frictional losses are negligible, so that the Euler turbomachine
equation applies, and the relative exit speed of the jet is the same as its relative inlet speed. We
take the density of water as 1000 kg/m3. The volume flow rate of the jet is equal to jet area times
jet velocity:
π D 2j π (0.067 m)2
V = V j = (87 m/s) = 0.3067 m 3 /s
4 4
(b) The maximum output shaft power occurs when the bucket moves at half the jet speed (w r = Vj /2).
Thus, the rotational speed of the wheel for maximum power is
Vj 87 m/s
ω= = = 35.66 rad/s
2r 2(1.22 m)
The maximum shaft power is determined from
W shaft,max = ρ rω V (V j − rω )(1 − cos β )
= (1000 kg/m 3 )(1.22 m)(35.66 rad/s)(0.3067 m3 /s)

× (87 m/s − (1.22 m)(35.66 rad/s))(1 − cos160°)

 1 N   1 J/s   1 kW 
×  
 1 kg ⋅ m/s 2   1 N ⋅ m/s   1000 J/s 
= 1126 kW
The actual shaft power is determined using the turbine efficiency as
W shaft,actual = ηturbine W shaft,max = (0.84)(1126 kW) = 946 kW ▲
6-4 REACTION TURBINES

The other main type of energy-producing hydro turbine is the reaction turbine, which
consists of fixed guide vanes called stay vanes, adjustable guide vanes called wicket gates,
and rotating blades called runner blades (Fig. 6-12). Flow enters tangentially at high pres-
sure, is turned toward the runner by the stay vanes as it moves along the spiral casing or
volute, and then passes through the wicket gates with a large tangential velocity component.
Momentum is exchanged between the fluid and the runner as the runner rotates, and there
is a large pressure drop. Unlike the impulse turbine, the water completely fills the casing
of a reaction turbine. For this reason, a reaction turbine generally produces more power
than an impulse turbine of the same diameter, net head, and volume flow rate. The angle
of the wicket gates is adjustable so as to control the volume flow rate through the runner.

Vin, Pin Top view

Stay vanes
Wicket
In
gates
Out
w
Runner
Volute blades
w
r2
r1 Shaft
b2
Figure 6-12 A reaction turbine differs significantly
b1 Band from an impulse turbine; instead of using water
jets, a volute is filled with swirling water that drives
Draft tube
Out the runner. For hydro turbine applications, the axis
•
is typically vertical. Top and side views are shown,
Vout, Pout including the fixed stay vanes and adjustable wicket
Side view gates.
(In most designs the wicket gates can close on each other, cutting off the flow of water into
the runner.) At design conditions the flow leaving the wicket gates impinges parallel to the
runner blade leading edge (from a rotating frame of reference) to avoid shock losses. Note
that in a good design, the number of wicket gates does not share a common denominator
with the number of runner blades. Otherwise there would be severe vibration caused by
simultaneous impingement of two or more wicket gate wakes onto the leading edges of the
runner blades. For example, in Fig. 6-12 there are 17 runner blades and 20 wicket gates.
These are typical numbers for many large reaction hydro turbines, as shown in the photo-
graphs in Figs. 6-14 and 6-15. The number of stay vanes and wicket gates is usually the same
(there are 20 stay vanes in Fig. 6-12). This is not a problem since none of them rotate, and
unsteady wake interaction is not an issue.
There are two main types of reaction turbine—Francis and Kaplan. The Francis turbine
is somewhat similar in geometry to a centrifugal or mixed-flow pump, but with the flow
in the opposite direction. Note, however, that a typical pump running backward would
not be a very efficient turbine. The Francis turbine is named in honor of James B. Francis
(1815–1892), who developed the design in the 1840s. In contrast, the Kaplan turbine is
somewhat like an axial-flow fan running backward. If you have ever seen a window fan start
spinning in the wrong direction when a gust of wind blows through the window, you can
visualize the basic operating principle of a Kaplan turbine. The Kaplan turbine is named in
honor of its inventor, Viktor Kaplan (1876–1934). There are actually several subcategories

of both Francis and Kaplan turbines, and the terminology used in the hydro turbine field is
not always standard.
We classify reaction turbines according to the angle that the flow enters the runner
(Fig. 6-13). If the flow enters the runner radially as in Fig. 6-13a, the turbine is called a Francis
radial-flow turbine (see also Fig. 6-12). If the flow enters the runner at some angle between
radial and axial (Fig. 6-13b), the turbine is called a Francis mixed-flow turbine. The latter
design is more common. Some hydro turbine engineers use the term “Francis turbine” only
when there is a band on the runner as in Fig. 6-13b. Francis turbines are most suited for heads
that lie between the high heads of Pelton wheel turbines and the low heads of Kaplan turbines.
A typical large Francis turbine may have 16 or more runner blades and can achieve a turbine
efficiency of 90 to 95 percent. If the runner has no band, and flow enters the runner partially
turned, it is called a propeller mixed-flow turbine or simply a mixed-flow turbine (Fig. 6-13c).
Finally, if the flow is turned completely axially before entering the runner (Fig. 6-13d), the
turbine is called an axial-flow turbine. The runners of an axial-flow turbine typically have only
three to eight blades, a lot fewer than Francis turbines. Of these there are two types: Kaplan
turbines and propeller turbines. Kaplan turbines are called double regulated because the flow
rate is controlled in two ways—by turning the wicket gates and by adjusting the pitch on the
runner blades. Propeller turbines are nearly identical to Kaplan turbines except that the blades
are fixed (pitch is not adjustable), and the flow rate is regulated only by the wicket gates (single
regulated). Compared to the Pelton and Francis turbines, Kaplan turbines and propeller tur-
bines are most suited for low head, high volume flow rate conditions. Their efficiencies rival
those of Francis turbines and may be as high as 94 percent.
Crown Crown
w w
Wicket Stay vane

gate Band Band
(a) (b)
Hub
w
Hub
w
(c) (d )
Figure 6-13 The distinguishing characteristics of the four subcategories of reaction turbines: (a) Francis
radial flow, (b) Francis mixed flow, (c) propeller mixed flow, and (d) propeller axial flow. The main
difference between (b) and (c) is that Francis mixed-flow runners have a band that rotates with the runner,
while propeller mixed-flow runners do not. There are two types of propeller mixed-flow turbines: Kaplan
turbines have adjustable pitch blades, while propeller turbines do not. Note that the terminology used here
is neither universal among turbomachinery textbooks nor among hydro turbine manufacturers.

Figure 6-14 The runner of a Francis radial-flow turbine used at the Boundary hydroelectric power
station on the Pend Oreille River north of Spokane, WA. There are 17 runner blades of outer
diameter 18.5 ft (5.6 m). The turbine rotates at 128.57 rpm and produces 230 MW of power at a
volume flow rate of 335 m3/s from a net head of 78 m. (© American Hydro Corporation. Used with
permission.)
Figure 6-14 is a photograph of the radial-flow runner of a Francis radial-flow turbine.

The workers are shown to give you an idea of how large the runners are in a hydroelectric
power plant. Figure 6-15 is a photograph of the mixed-flow runner of a Francis turbine,
and Fig. 6-16 is a photograph of an axial-flow propeller turbine. The view is from the
inlet (top).
We perform preliminary design and analysis of turbines using the Euler turbomachine
equation and velocity diagrams. For a turbine, the flow direction is opposite to that of a
pump, so the inlet is at radius r2 and the outlet is at radius r1. For a first-order analysis we
approximate the blades as being infinitesimally thin. We also assume that the blades are
aligned such that the flow is always tangent to the blade surface, and we ignore viscous
effects (boundary layers) at the surfaces.
Consider for example the top view of the Francis turbine of Fig. 6-12. Velocity vec-
tors are drawn in Fig. 6-17 for both the absolute reference frame and the relative reference
frame rotating with the runner. Beginning with the stationary guide vane (thick black line
in Fig. 6-17), the flow is turned so that it strikes the runner blade (thick gray line) at abso-
lute velocity V2 . But the runner blade is rotating counterclockwise, and at radius r2 it moves
tangentially to the lower left at speed w r2. To translate into the rotating reference frame, we
form the vector sum of V2 and the negative of w r2, as shown in the sketch. The resultant is
vector V2, relative , which is parallel to the runner blade leading edge (angle β2 from the tangent

Figure 6-15 The runner of a Francis mixed-flow turbine used at the Smith Mountain hydroelectric
power station in Roanoke, VA. There are 17 runner blades of outer diameter 20.3 ft (6.19 m). The
turbine rotates at100 rpm and produces 194 MW of power at a volume flow rate of 375 m3/s from a
net head of 54.9 m. (Photo courtesy of American Hydro Corporation, York, PA. Used with permission.)
Figure 6-16 The five-bladed propeller turbine used at the Warwick hydroelectric power station in
Cordele, GA. There are five runner blades of outer diameter 12.7 ft (3.87 m). The turbine rotates
at 100 rpm and produces 5.37 MW of power at a volume flow rate of 63.7 m3/s from a net head of
9.75 m. (Photo courtesy of American Hydro Corporation, York, PA. Used with permission.)

b2
V2, t
w r2
V2, relative
V2, n
r1
V2 Figure 6-17 Relative and absolute velocity vectors

w
r2 and geometry for the outer radius of the runner of a
Francis turbine. Absolute velocity vectors are bold.

line of circle r2). The tangential component V2,t, of the absolute velocity vector V2 is required
for the Euler turbomachine equation [Eq. (6-24)]. After some trigonometry,
V2,n
Runner leading edge: V2,t = ω r2 − (6-27)
tan β 2
Following the flow along the runner blade in the relative (rotating) reference frame,
we see that the flow is turned such that it exits parallel to the trailing edge of the runner
blade (angle b1 from the tangent line of circle r1). Finally, to translate back to the absolute
reference frame we vectorially add V1, relative and blade speed w r1, which acts to the left as
sketched in Fig. 6-18. The resultant is absolute vector V1. Since mass must be conserved, the
normal components of the absolute velocity vectors V1,n and V2,n are related through
V = 2π r1b1V1,n = 2π r2b2V2,n
where axial blade widths b1 and b2 are defined in Fig. 6-12. After some trigonometry (which
turns out to be identical to that at the leading edge), we generate an expression for the
tangential component V1,t of absolute velocity vector V1 for use in the Euler turbomachine
equation,
V1,n
Runner trailing edge: V1,t = ω r1 − (6-28)
tan β1
b1 V2, relative
w r1
V1 V1, relative
r1
V2 Figure 6-18 Relative and absolute velocity vectors
w and geometry for the inner radius of the runner of a
r2
Francis turbine. Absolute velocity vectors are bold.

Figure 6-19 In some Francis mixed-flow turbines, high-

power, high-volume flow rate conditions sometimes lead to
reverse swirl, in which the flow exiting the runner swirls in
the direction opposite to that of the runner itself, as sketched
Reverse swirl here.
For some hydro turbine runner applications, high power/high flow operation can
result in V1,t < 0. Here the runner blade turns the flow so much that the flow at the runner
outlet rotates in the direction opposite to runner rotation, a situation called reverse swirl
(Fig. 6-19). The Euler turbomachine equation predicts that maximum power is obtained
when V1,t < 0, so we suspect that reverse swirl should be part of a good turbine design.
In practice, however, it has been found that the best efficiency operation of most hydro
turbines occurs when the runner imparts a small amount of with-rotations wirl to the
flow exiting the runner (swirl in the same direction as runner rotation). This improves
draft tube performance. A large amount of swirl (either reverse or with-rotation) is not
desirable, because it leads to much higher losses in the draft tube. (High swirl velocities
result in “wasted” kinetic energy.) Obviously, much fine tuning needs to be done in order
to design the most efficient hydro turbine system (including the draft tube as an integral
component) within imposed design constraints. Also keep in mind that the flow is three-
dimensional; there is an axial component of the velocity as the flow is turned into the
draft tube, and there are differences in velocity in the circumferential direction as well. It
doesn’t take long before you realize that computer simulation tools are enormously useful
to turbine designers. In fact, with the help of modern CFD codes, the efficiency of hydro
turbines has increased to the point where retrofits of old turbines in hydroelectric plants
are economically wise and common.
EXAMPLE 6-5 A Francis radial-flow hydro turbine has the following dimensions, where location 2 is the inlet and
Analysis of a location 1 is the outlet: r2 = 1.90 m, r1 = 1.35 m, b2 = 0.78 m, and b1 = 2.15 m. The runner rotates at
Francis Radial-Flow n = 165 rpm. The wicket gates turn the flow by angle a2 = 55° from radial at the runner inlet, and
Hydro Turbine the flow at the runner outlet is at angle a1 = 25° from radial (Fig. 6-20). The volume flow rate at
design conditions is 200 m3/s. Neglecting irreversible losses, calculate the inlet and outlet runner
blade angles b2 and b1, respectively, and predict the power output (MW) and required net head (m).
SOLUTION We assume that the flow is everywhere tangent to the runner blades and we neglect
irreversible losses through the turbine. We solve for the normal component of velocity at the inlet,
V 200 m3 /s
V2,n = = = 21.48 m/s
2π r2b2 2π (1.90 m)(0.78 m)
Using the figure provided with this problem as a guide, the tangential velocity component at the
inlet is
V2,t = V2,n tan α 2 = (21.48 m/s)tan55° = 30.67 m/s

V2, t
a2
V2, n
V2
V1, t
r2
V1 a1
V1, n
r1
Control volume
Figure 6-20 Schematic for Example 6-5.
The angular velocity is
60

60 s( )
ω = 2π n = 2π (165 rad/min) 1 min = 17.28 rad/s
The tangential velocity component of the absolute velocity at the inlet is obtained from trigonometry
to be
V2,n
V2,t = ω r2 −
tan β 2
From the above relationship, we solve for the runner leading edge angle b2,
 V2,n  21.48 m/s

β 2 = arctan  = arctan   = 84.3°
 ω r2 − V2,t  (17.28 rad/s)(1.90 m) − 30.67 m/s) 
We repeat the calculations for the runner outlet:
V 200 m3 /s
V1,n = = = 10.97 m/s
2π r1b1 2π (1.35 m)(2.15 m)
V1,t = V1,n tan α 1 = (10.97 m/s)tan 25° = 5.114 m/s
 V1,n  10.97 m/s

β1 = arctan  = arctan   = 31.1°
 ω r1 − V1,t  (17.28 rad/s)(1.35 m) − 5.114 m/s) 
The shaft output power is estimated from the Euler turbomachine equation as
W shaft = ρω V (r2V2,t − r1V1,t )
= (1000 kg/m 3 )(17.28 rad/s)(200 m3 /s)[(1.90 m)(30.67 m/s) − (1.35 m)(5.114 m/s)]

1 N ⋅ s 2  1 MW ⋅ s3 
= 1.775 × 108
1 kg ⋅ m  106 kg ⋅ m 2 
= 177.5 MW

6-5 Turbine Specific Speed 185
Finally, we calculate the required net head assuming that hturbine = 100 percent since we are ignoring
irreversibilities:
W shaft 177,500 kJ/s
H= = = 90.5 m
ρVg  1 kJ/kg 
(1000 kg/m )(200 m3 /s)(9.81 m/s 2 ) 
3
 1000 m 2 /s 2 
This is a preliminary design in which we are neglecting irreversibilities. Actual output power will be
lower, and actual required net head will be higher than the values predicted here. ▲
6-5 TURBINE SPECIFIC SPEED

A useful parameter in the preliminary selection of hydraulic turbines is the turbine specific
speed NSt. It is defined in terms of two independent dimensionless parameters for turbines,
namely power coefficient CP and head coefficient CH, as
1/2
 bhp 
C1/2  ρω 3 D 5 
ω (bhp)1/2
P
N St =
5/4 =
= 1/2 (6-29)
CH  gH 
5/2
ρ ( gH )5/4
 ω 2 D 2 
where w is the rotational speed, bhp is brake horsepower (shaft power W shaft), r is the fluid
density, and H is the net head.
Although NSt is by definition a dimensionless parameter, practicing engineers have grown
accustomed to using inconsistent units that transform NSt into a cumbersome dimensional
quantity. In the United States, most turbine engineers write the rotational speed in units of
rotations per minute (rpm), bhp in units of horsepower, and H in units of feet. Furthermore,
they ignore gravitational constant g and density r in the definition of NSt. (The turbine is
assumed to operate on earth and the working fluid is assumed to be water.) We define
(n , rpm)(bhp, hp)1/2
N St, US = (6-30)
( H , ft)5/4
There is some discrepancy in the turbomachinery literature over the conversions
between the two forms of turbine specific speed. To convert NSt,US to NSt, we divide by g5/4
and r1/2, and then use conversion ratios to cancel all units. We set g = 32.174 ft/s2 and
assume water at density r = 62.40 lbm/ft3. When done properly by converting w to rad/s,
the conversion is NSt,US = 0.02301 NSt or NSt = 43.46 NSt,US.
There is also a metric or SI version of turbine specific speed that is becoming more
popular these days and is preferred by many hydro turbine designers. It is defined in the
same way as the customary U.S. pump specific speed, except that SI units are used (m3/s
instead of gpm and m instead of ft),
(n , rpm)(V , m3 /s)1/2
N St, SI = (6-31)
( H , m)3/4
We may call this capacity specific speed to distinguish it from power specific speed [Eq. (6-29)].
Technically, turbine specific speed could be applied at any operating condi-
tion and would just be another function of CP. That is not how it is typically
Turbine specific speed is used to
used, however. Instead, it is common to define turbine specific speed only at the
characterize the operation of a
turbine at its optimum conditions
best efficiency point (BEP) of the turbine. The result is a single number that
(best efficiency point) and is useful characterizes the turbine.
for preliminary turbine selection. As plotted in Fig. 6-21, impulse turbines perform optimally for NSt near
0.15, while Francis turbines and Kaplan or propeller turbines perform best at

NSt, US
1 2 5 10 20 50 100 200
1
Francis Kaplan/propeller
Impulse
0.9
0.8
hmax
0.7
0.6
0.5
0.01 0.02 0.05 0.1 0.2 0.5 1 2 5 10
NSt
Figure 6-21 Maximum efficiency as a function of turbine specific speed for the three main types of
dynamic turbine. Horizontal scales show nondimensional turbine specific speed (NSt) and turbine
specific speed in customary U.S. units (NSt,US). Sketches of the blade types are also provided on the
plot for reference.
NSt near 1 and 2.5, respectively. It turns out that if NSt is less than about 0.3, an impulse
turbine is the best choice. If NSt is between about 0.3 and 2, a Francis turbine is a better
choice. When NSt is greater than about 2, a Kaplan or propeller turbine should be used.
These ranges are indicated in Fig. 6-21 in terms of NSt and NSt,US.
EXAMPLE 6-6 Calculate the turbine specific speed of the Francis turbine in Example 6-5. Provide answers in both
Turbine Specific dimensionless form and in customary U.S. units. The values for the turbine of Example 6-5 are
Speed of a w = 17.28 rad/s, bhp = 1.775 × 10 W, H = 90.5 m, n = 165 rpm.
8
Francis Turbine
SOLUTION We first calculate the nondimensional form of NSt,
1/2
ω (bhp)1/2 (17.28 rad/s)(1.775 × 108 W)1/2  1 kg ⋅ m 2 
N St = 5/4 =  3  = 1.50
ρ ( gH )
1/2
(1000 kg/m 3 )1/2 (9.81 m/s 2 )(90.5 m)  1 W ⋅ s 
5/4
From Fig. 6-21, this is in the range for a Francis turbine. From the conversion given in the text,
NSt,US = 1.50 × 43.46 = 65.2
Alternatively, we can use the original dimensional data to calculate NSt,US,

1/2
  1 hp  
(165 rpm) (1.775 × 105 kW) 
(n , rpm)(bhp, hp)1/2   0.7457 kW  
N St,US = = = 65.3 ▲
( H , ft)5/4
( )
5/4
 3.2808 ft 
(90.5 m) 1 m 
6-6 RUN-OF-RIVER PLANTS AND WATERWHEELS

The majority of the electricity from hydroelectric power plants is produced using accu-
mulation reservoirs, also called dams. Run-of-river plants (also called river plants or small
hydroelectric power plants) do not use water reservoirs. Instead, they are built along a water
stream such as a river. In this system, a proper portion of the water stream is diverted to a
turbine-generator unit, and the used water is returned to the river, as shown in Fig. 6-22
(Farret and Simoes, 2006).

6-6 Run-of-River Plants and Waterwheels 187
Canal Forebay
Penstock
Turbine and
generator
Weir
River
Figure 6-22 Working principle of a run-of-river plant. Weir is the barrier placed in river to control
water discharge. Forebay is part of canal from which water is taken to run the turbine.
The flow rate of a river is not constant as it varies with season, month, and day. As a con-
sequence, this scheme of hydropower generation does not make the full use of water stream
most of the time. This is because the capacity of system is normally selected for an average
water flow rate (i.e., rated flow rate). When the flow rate is more than the rated flow, excess
water is diverted from the turbines and not used for power generation. Installing a large
turbine-generator system to accommodate the highest possible water flow rate increases
the amount of power generation but this also increases the initial cost and the system oper-
ates at part load (and possibly lower efficiency) most of the time. Therefore, the selection of
system capacity is a tradeoff among the initial cost, power output, and efficiency.
A run-of-river plant is installed when there is sufficient velocity and flow rate for a
river. In order to estimate the power potential from a river, we consider a river water flow at
a mass flow rate of m, a velocity of V, and negligible elevation change. A hydraulic turbine
converts the kinetic energy of this water into power. Then the power potential of this water
stream can be determined from
2
V V 2
V 3
W available = m = ρ AV = ρA (6-32)
2 2 2
For a water density of r = 1000 kg/m3, a water velocity of V = 1 m/s, and per unit cross-
section area A, the rate of kinetic energy or power potential is calculated to be 500 W/m2
of flow cross section. Unfortunately, only about 60 percent of this potential is theoretically
available. In addition, the efficiency of turbine-generator units in river plants is lower than
that of large hydropower plants. Equation (6-32) shows that the power potential is propor-
tional to the cube of velocity. For a water velocity of 2 m/s, the power potential would be
4000 W/m2 of flow section. Unfortunately, water speeds in rivers are not high in most cases,
and the power potential is limited.
A waterwheel is different from a run-of-river plant in that a waterwheel is installed
directly into falling or free-flowing water to convert the energy of the water into power.
This is particularly illustrated for a lower bucket type of a waterwheel in which water flows
through the lower blades (called buckets) causing the wheel to rotate. A more common type
of a waterwheel involves upper buckets for which water flows through the upper buckets,
fills them, and forces the wheel to rotate due to the weight acting on the buckets (Fig. 6-23).

Upper penstock
Buckets
Wheel
rotation
Head
Water flow
Figure 6-23 Operation of an upper
bucket waterwheel.
The power potential from a waterwheel can be determined from
W available = mgH
 = ρVgH (6-33)
where m is the mass flow rate, V is the volume flow rate, r is the water density, and H is the
head of the water that is the height difference between water streams at the input and output
of the water channel. The actual power output can be estimated by multiplying Eq. (6-33) by
the efficiency of the waterwheel. The efficiencies of waterwheels can be taken to be about
60 percent. For a water density of r = 1000 kg/m3, a water flow rate of V = 1 m3/s, and a
head of H = 10 m, the power potential is calculated to be 98 kW. For an efficiency of
60 percent, the actual power output is 59 kW.
Waterwheels were used since ancient times for milling flour in gristmills, and they are
still used in some parts of the world for this application. Previously, they were also used
for water-lifting for irrigation, grinding wood into pulp for papermaking, machining,
hammering wrought iron, and powering mine hoists. They are usually made from wood
or steel, with shovels of blades fixed regularly around their circumference. Water pushes
the shovels (or buckets) tangentially around the wheel, causing a torque to develop on
the shaft. The rotational speeds are typically low and these machines have low efficiencies
due to losses such as friction, turbulence, and incomplete filling of the buckets (Farret
and Simoes, 2006).
Waterwheels are simple devices and operation is not affected by dirty water. Since
waterwheels have very low rotational speed, the generator requires high rates of speed mul-
tipliers for the production of electricity. This presents additional losses in the system. The
efficiency of a waterwheel for the production of electricity is very low.
EXAMPLE 6-7 A run-of-river plant operates on a water stream whose cross-sectional area is 18 m2. Water is flowing
Efficiency of a at a speed of 2.5 m/s. If the electrical power output from the plant is 61 kW, what is the overall effi-
Run-of-River ciency of this plant?
Plant
SOLUTION The power potential of this water stream is
3
V (2.5 m/s)
3
W available = ρ A = (1000 kg/m 3 )(18 m 2 ) = 140,625 kg ⋅ m 2 /s3
2 2

Hydropower: 6-1 Introduction

Uploaded by

Copyright:

Available Formats

Hydropower: 6-1 Introduction

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Hydropower: 6-1 Introduction

Uploaded by

Copyright:

Available Formats

CHAPTER 6

06_Kanoglu_Ch06_p163-194.indd 163 04/06/19 3:20 PM

Figure 6-1 (a) An aerial view of Hoover

06_Kanoglu_Ch06_p163-194.indd 164 04/06/19 3:20 PM

6-2 ANALYSIS OF A HYDROELECTRIC POWER PLANT

06_Kanoglu_Ch06_p163-194.indd 165 04/06/19 3:20 PM

since P1 ≈ P4 = Patm and V1 = V4 ≈ 0 

06_Kanoglu_Ch06_p163-194.indd 166 04/06/19 3:20 PM

hturbine = 0.75 hgenerator = 0.97

Figure 6-3 The overall efficiency of a turbine-generator is the

emech, in = emech, out + emech, loss (6-8)

06_Kanoglu_Ch06_p163-194.indd 167 04/06/19 3:20 PM

w turbine,e W turbine,e W turbine

emech, loss, piping E mech, loss, piping

hL , total = hL , major + hL , minor = f

06_Kanoglu_Ch06_p163-194.indd 168 04/06/19 3:20 PM

Power station Net

Draft tube D EGLout zE

06_Kanoglu_Ch06_p163-194.indd 169 04/06/19 3:20 PM

Hnet = EGLin− EGLout(6-17)

This is equivalent to turbine-generator efficiency [Eq. (6-7)], defined earlier. Usually, we

E mech,loss,piping = ρ gVhL (6-20)

The turbine efficiency is then expressed as

W shaft W shaft W shaft

06_Kanoglu_Ch06_p163-194.indd 170 04/06/19 3:20 PM

E mech, loss, piping

 W  W shaft   E mech,loss,piping  (6-23)

Then the maximum power is determined from

06_Kanoglu_Ch06_p163-194.indd 171 04/06/19 3:20 PM

P1 − P2 (1400 − 100) kPa  1 kJ 

w shaft = ηturbinew max = (0.88)(1.30 kJ/kg) = 1.144 kJ /kg 

(b) The height of the reservoir is determined from

w shaft 1.144 kJ/kg  1000 m 2 /s 2 

 gross = ρVgH gross

06_Kanoglu_Ch06_p163-194.indd 172 04/06/19 3:20 PM

 E mech, loss, piping   589 kW 

ηplant = ηgeneratorηturbineηpiping = (0.96)(0.93)(0.977) = 0.872 or 87.2% 

The definition of turbine efficiency based on the net head is

Solving for the shaft power output, we obtain

W shaft = ηturbine (W max − E mech, loss, piping )

= (0.93)(25,115 kW − 589 kW) 

6-3 IMPULSE TURBINES

06_Kanoglu_Ch06_p163-194.indd 173 04/06/19 3:20 PM

06_Kanoglu_Ch06_p163-194.indd 174 04/06/19 3:20 PM

Figure 6-9 A view from the bottom of

W shaft = ω Tshaft = ρω V (r2V2,t − r1V1,t ) (6-24)

We must be careful of negative signs since this is an energy-producing rather than an

06_Kanoglu_Ch06_p163-194.indd 175 04/06/19 3:20 PM

Upon substitution of this equation, Eq. (6-24) yields

W shaft = ρ rω V {V j − [rω + (V j − rω )cos β ]} 

Output shaft power: W shaft = ρrω V(V j − rω )(1 − cos β ) (6-25)

06_Kanoglu_Ch06_p163-194.indd 176 04/06/19 3:20 PM

W shaft,max = ρ rω V (V j − rω )(1 − cos β )

= (1000 kg/m 3 )(1.22 m)(35.66 rad/s)(0.3067 m3 /s)

The actual shaft power is determined using the turbine efficiency as

W shaft,actual = ηturbine W shaft,max = (0.84)(1126 kW) = 946 kW ▲

6-4 REACTION TURBINES

06_Kanoglu_Ch06_p163-194.indd 177 04/06/19 3:20 PM

Vin, Pin Top view

since P1 ≈ P4 = Patm and V1 = V4 ≈ 0

emech, in = emech, out + emech, loss (6-8)

Hnet = EGLin− EGLout(6-17)

E mech,loss,piping = ρ gVhL (6-20)

 W  W shaft   E mech,loss,piping  (6-23)

w shaft = ηturbinew max = (0.88)(1.30 kJ/kg) = 1.144 kJ /kg

ηplant = ηgeneratorηturbineηpiping = (0.96)(0.93)(0.977) = 0.872 or 87.2%

= (0.93)(25,115 kW − 589 kW)

W shaft = ω Tshaft = ρω V (r2V2,t − r1V1,t ) (6-24)

W shaft = ρ rω V {V j − [rω + (V j − rω )cos β ]}

Output shaft power: W shaft = ρrω V(V j − rω )(1 − cos β ) (6-25)

W shaft,actual = ηturbine W shaft,max = (0.84)(1126 kW) = 946 kW ▲

V2,t = V2,n tan α 2 = (21.48 m/s)tan55° = 30.67 m/s

NSt,US = 1.50 × 43.46 = 65.2