0 Q

LIBRARY
Kashmiri Gate, Delhi*11000fi

Accpssion No._

Class No._

Book No.._

Borrower is roquisfed

to check the book and

get the signatores on the

fomed pages, if any.

FREDERICK T. MORSE, M.E.. E.E.
Professor of Mechanical Engineering
University of Virginia
 Copyright ® 1932, 1942, 1953, BY

LITTON EDUCATIONAL PUBLISHING, INC.

No reproduction in any form of this book, in whole or in

part {except for brief quotation in critical articles or reviews)^
may be made without written authorization from the publishers

Affiuatfd East-West Press Pvt. Ltd.

LaU'West Student Edition-1964
Second Cast-West Reprint-l96S
Third East-West Reprint-1974
(In MRS Units)

Price in India Rs 15 00

Sales Territory India, Bangla Desh, Burma,

Ceylon, Malaysia, Pakistan and Singapore

Reprinted in India with the special permission of the original

publishers, Litton Educational Publishing, Inc,, New York^ U.S A
and the copyright holders

This book has been published with the assistance of the Joint
Indian-American Textbook programme.

Published by K S Padmanabhan for AFFILIATED EAST-

WEST PRESS PVT LTD , 9 Nizamuddin East, New Delhi 13,
India, and printed by Mohan Makhijani at Rekha Printers,
New Delhi
 PREFACE

This book is the result of work that started as an intended revision of the
author's Power Plant Engineering and Design, Out of the experiences of
twenty years of teaching and consultive engineering which have followed pub¬
lication of the first edition have come many ideas for change—additions,
deletions, and modifications. These ideas have been incorporated into the
present book, and the result must be considered as more of a new book than
a revision.
The objective continues to be the presentation of a thorough study of
stationary power and heating plants, including public service, industrial, and
institutional varieties. The needs of the engineering student, the consulting
engineer, and the plant operator were equally considered in the preparation
of the present volume. Fundamentals are emphasized, leaving details and
dimensions of equipment for more specialized books and manufacturers' litera¬
ture. The power plant is emphasized as an integrated assembly. Economic
factors are pointed out, but specific costs and prices are avoided as being sub¬
ject to rapid change. The treatment will give due consideration to recent
trends and theories in the field of power generation. Much greater emphasis
is placed on the small plant than in the former editions.
Those who are familiar with Power Plant Engineering and Design, this
oook's predecessor, will note the absence of former sections dealing with hy-
<lraulic power and electrical equipment. On the other hand, steam plant equip-
nent is awarded expanded coverage. The inclusion of a short chapter on the
gas turbine is an indication of regard for its potentiality as a future factor
m the stationary power plant field. When used as a college'textbook, a full
two-semester course may be based on this book. A one-semester course on
steam power plants would employ Chapters 1, 6, 8, 9, 10, 11, 12, and 13 only.
Power plant engineering is a subject of great detail. Throughout this book
^.he author has endeavored to summarize wherever possible, then proceed to
the separate explanations, if any, in the belief that this method will be of
major benefit both to student and teacher. Sample examples explain most
of the principles of the equipment and its assembly. These are further ampli¬
fied by a large selection of chapter-end problems provided for the reader's
practice.
Although Power Plant Engineering, like its predecessor, presumes that the
reader has actively studied or otherwise become acquainted with the subject
of Elementary Heat Power or Applied Thermodynamics, it will be found that
more attention has been given to review and explanation of these funda¬
mentals. In this and other ways the author acknowledges valuable construc¬
tive criticism tendered by reisers. Thanks are due the several friends who
have contributed comments, suggestions, and reviews of parts of the new
manuscript.
V
VI PREFACE
Many industrial finns, associations, and publishers have been asked to
contribute data and illustrations. It is a pleasure again to be able to give
testament to the splendid cooperation the author has received from all. The
Diesel Engine Manufacturers’ Association, Heat Exchange Institute, Hy¬
draulic Institute, and other associations have freely donated material from
their Standards. An especial acknowledgment is due Power magazine, whose
editors have provided several illustrations of good educational value.
Although carefully prepared and reviewed, this work may contain errors
of typography, mathematics, or statement. Any report of these errors will be
welcomed by the author.
The assistance of the author’s wife, Genevieve F. Morse, in the preparation
of the manuscript and proofreading is most gratefully acknowledged.

F, T. M.
University of Virginia
Charlottesville, Va.
March 1, 1958
CHAPTER PAGE
1 Introduction 1
Energy and power Background of power study. Working proc¬
esses. The power age. Power jilant engineering and design. Draw¬
ings, specifications, and correspondence Electric power systems.
Problems.

2 The Variable Load Problem 36

Industrial production and power generation compared. Ideal and
realized load curves Terms and factors. Effect of variable load
on power plant design, f^ffect of variable load on power plant op¬
eration. Methods of meeting the load. Problems.

3 Power Plant Economics 63

Source of income. Effect of plant type on costs. Rates. Fixed ele¬
ment. Energy element. Customer element. The investor’s profit.
Depreciation and replacement. Theory of rates. Making up rate
structures. Problems.

4 The Power Plant Bxulding 86

General design. Materials. Reference lines Architectural fea¬
tures. Chimney Masonry chimney stability. Foundations. Mis¬
cellaneous equipment. Problems.

5 Fuels AND Combi stion 116

Fuels. Gas. Oil Internal combustion engine fuel Furnace fuel
oil Coal. Fuel tests for heating value Proximate analysis. Ash
and refuse analyses. Combustion. Illustrative examples in com¬
bustion. Approximations and short-cuts Smoke. Problems.

6 Internal Combustion Engine Power Plant 149

Internal combustion engines. Stationary Diesel engine power. Com¬
bustion. Engine performance. Plant design. Auxiliary systems.
Evaporative cooling. Testing plant performance. Problems.

7 Gas Turbine Power Plant 191

Stationary gas turbine. Constant pressure combustion cycle. The
simple gas turbine plant. Thermal refinement of the gas turbine
cycle. Field for gas turbine power. Problems,
viii CONTENTS

CHAPTER page
8 Vapor Cycles .... 208

Energy from external combustion. Vapor cycles. Rankine cycle.

Regenerative cycle. Design of regenerative cycles. Heat balance,
regenerative cycle. Reheating cycle. Superposed power units. In¬
dustrial heat-power balance. Combination cycles. Variable load
operation. Problems.

9 Energy Flow in the Steam Power Plant 266

The steam power plant. Functional relation of equipment. Pro¬

duction of heat energy. Thermal level. Heat flow. Conduction.
Surface convection. Mean temperature difference. Convective
heat transfer from products of combustion. Convective heat trans¬
fer between condensing steam and water. Examples of surface
convection calculations. Radiation. Apportionment of heating
surface. Problems.

10 Steam Generators 290

Components. Design criteria. Boiler types. Representative fire-

tube boilers. Representative water-tube boilers. Water walls.
Boiler installation. Furnace. Superheat. Boiler accessories and
trim. Combustion equipment. Spreader stoker. Conveyor stoker.
Underfeed stoker Coal burners. Gas and oil burners. Operation
and performance. Heat balance of a steam generator. Problems.

11 Steam Prime Movers 369

Prime mover types. Steam engine applications. Steam engine

capacity and performance, ynaflow engine. Steam turbines.
Thermal efficiency. Governing steam turbines. Part-load per¬
formance. Mechanical-drive turbines. Turbine installation and
operation. Turbine tests. Vacuum. Contact condensers. Surface
condensers. Air removal. Condenser auxiliaries. Condenser op¬
eration and testing. Problems.

12 The Gas Loop . 440

Function of the gas loop. Coal storage. Coal conveyors. Pulver¬

ized coal systems. Oil and gas supply systems. Flue gas cleaning.
Ash handling. Air preheater. Draft. Gas conduits. Draft by
fans. Draft by chimneys. Control of gas loop flows. Combustion
control equipment. Problems.

13 The Feedwater Loop 610

Function of the feedwater loop. Water contamination—its effect.
CONTENTS ix
CHAPm PAGE
Water tests and analyses. Deooncentration. Water treatment.
Feedwater heating. Contact heaters. Surface heaters. Evapora*
tots. Water pumps. Reciprocating pumps. Turbine pumps.
Centrifugal pumps. Feedwater regulation. Problems.
14 The Piping System 571
Pipe system classification. Commercial pipe. Pipe connections.
Pipe fittings. Common valves. Special valves. Boiler safety
valves. Thermal expansion. Heat insulation. Determining pipe
sise. Pipe support. Drips and drains. Designing the piping sys¬
tem. Pipe drawings. Problems.
15 Ikstbvmentation 624
Purpose of instrumenting the power plant. Classification of instru¬
ments. Temperature and pressure measurement. Measurement of
fluid flow. Commercial* fluid flow meters. Gas analysis. Combi¬
nation meters. Electrical measurements. Remote readings. Selec¬
tion of instruments. Records and their use.

Biblioobaphy 657
Appendix 660
Index 679
 A Area. Atomic mass number.
a Acc(‘leration. Unit area.
C A constant or coefficient. Mol specific heat.
c Unit sp(‘eifie heat. Velocity of light.
D Diameter. Draft.
(] Density. Thickness (of wall, tube, etc.).
E Voltage.
e Energy.
F Force.
/ Fretpiency. Power factor. Friction factor. Stress (structural).
G Mass flow.
g Acceh'iation of gravity.
H Height.
h Enthalpy. Surface coefficient of heat transfer.
I Amperes. Moment of inertia.
J Mechanical equivalent of heat.
K A constant or coefficient.
k Cyoefficient of conductivity.
L Distance, linear.
M ^Moment. Torque. Molecular weight.
m Mass.
N Rotative speed, revolutions per unit time.
n Poly tropic exponent. Number of units, i.e., cylinders, tubes, etc
P Power.
p Unit fluid pressure.
Q Transfc'rred heat energy. Volume rate of flow.
Qy, Higher heating value (fuel).
Ol Lower heating value (fuel).
q Heat flow rate.
R Gas constant. Cut-off ratio. Resistance.
Hydraulic radius.
r Radius. Compression, or expansion, ratio.
S Stress (pipe).
s Unit entropy.
T Absolute temperature.
t Temperature.
U Thermal conductance.
u Unit internal energy. Tangential velocity.
XI
xii SYMBOLS
V Velocity.
V Specific volume.
W Work. Total weight.
w Weight. Unit work.
X Drynei>s factor.
P (Beta) Diameter ratio,
y (Gamma) Adiabatic polytropic exponent.
A (Delta) An increment.
€ (Epsilon) Emissivity (Radiant Energy).
yj (Eta) Efficiency.
B (Theta) Temperature, special.
X (Lambda) Wave length.
/X (Mu) Absolute viscosity.
V (Nu) Kinematic viscosity.
p (Rho) Mass density.
S (Sigma) A summation operator.
or (Sigma) Stephan constant (Radiant Energy).
T (Tau) Time.
^ (Psi) No-load factor.
Q (Omega) Radiant heat fraction,
w (Omega) Rotative speed, radians per unit time.

ms/teo Cubic metres per seconcL

m/min Metres per minute.
aiD/«eo Centimetres per second*
m/see Metres per second**
kw hr Kilowatt hours.
In Natural logarithm,
km/h Kilometres per hour*
NPSH Net positive suction head,
kg/m* Kilograms per cubic metre,
kg/m* Kilograms per square metres
kg/om* Kilograms per-square centimetre.
TDH Total dynamic head.
TDS Total dissolved solids.
K Degrees Kelvin.
Mw Megawatts*
J Joule*
kgm Kilogram metre*
g Gram,
t Tonnes,
These in addition to obvious abbreviations, such as m for metre.
 CHAPTER 1

1-1 Energy and Power. Both the historical and the present-day civiliza¬
tion of mankind are closely interwoven with energy, and there is little reason
to doubt but that in the future our existence will be ever more dependent upon
this thing called energy. Mere existence requires that even an animal jiroduce
and expend it. Until man found ways to utilize energy from sources outside
and beyond his own physical efforts, his status on earth was quite animal-like.
What is this quantity that has been so essential to life and progress? Energy
probably was the original stuff of creation. As we encounter it about us, energy
appears in many forms, but has one thing in common—energy is possessed of
the ability to produce a dynamic, vital effect.
Energy is associated with physical substance, but is not a substance itself.
It shows itself by the excited, animated state assumefl by material which
receives energy. We see the manifestations everywhere around us, in bodily
physical activity produced by the consumption of energy-containing food, or
in masses in motion, wind and wave, projectiles and vehicles; we see it in the
form of heat received from the sun, also as heat generated from fuels and other
sources; we see it in the sudden galvanizing into action of an idle electric motor
which has been connected to a far distant source. The examples that can be
mentioned are almost endless in number. In this era energy is being produced in
enormous quantities. As more and more of it is produced per person, the com¬
forts, conveniences, and pleasures of life are ever enhanced.
All these manifestations are conveniently classified as: (1) mechanical
work; (2) heat; (3) electricity; and (4) radiation. Technical details of these
various forms of energy and their computations are included in this chapter.
The average person knows, or thinks he knows, the meaning of power, and
he uses the term frequently in everyday language. But what is the exact
technical meaning of this word, and how is it best defined? If a person has a
good and comprehensive idea of “energy,” it is quite simple for him to under¬
stand the technical meaning of power, for powder is the rate 'at which energy is
produced and consumed. Any physical unit of energy when divided by a unit of
time automatically becomes a unit of powder. However, it is in connection with
the mechanical and electrical forms of energy that the term “power” is generally
used. The rate of production or consumption of heat energy and, to a certain
extent, of radiation energy is not ordinarily thought of as power. Power is
1
2 INTRODUCTION
primarily associated with mechanical work and electrical energy. Therefore,
we will define power as the rate of flow of energy and state that a i)ovv(*r i)]ant
is a unit built for the production and delivery of a flow of nieclianieal and
electrical energy.
In common usage, a machine or assemblage of equij)ment tliat i)roduces and
delivers a flow of mechanical or electrical energy is a power plant. Hcaice, an
internal combustion engine is a power plant, a water wheel is a power plant,
etc. However, what wc generally mean by the term is that assemblage of
equipment, permanently located on some chosen site, which receives raw
energy in the form of a substance capable of being operated on in such a way
as to produce electrical energy for delivery from the power plant. It is this kind
of power plant to which this book is chiefly, but not (‘xclu^^ively, addressed. The
problems of the mobile power plant are similar in general features, but the
need for mobility usually imposes requirements of simjdicity at the sacrifice of
some efficiency. For these and other reasons, the mobile power plant has be¬
come rather specialized along paths too numerous to cover in a book of this
character.
We here propose to inquire into the nature of the stationary jiower plant,
its economics, design, and performance. Considerable emphasis will be placc'd
upon the public utility type of stationary power plant, hut we do not intend to
concentrate on that type to the point of lu'glecting the smaller industrial anil
institutional units. Although possibly not providing illustrations of tlie maxi¬
mum technical progress made in this field, nevertlieless the latter account for
a significant portion of the remunerative engineering efforts iu power ])lant
engineering and design.
Furthermore, the many plants in w’hich no mechanical or electrical ])ower
is produced, but from which energy is delivered in the form of the heat of
steam, are considered to be within the purview of thi^' book, for any power plant
using fuel as the raw material is, fir.st of all, a heat-producing plant. If the
heat is produced in some convenient carrying medium, such as steam, th<*n that
would be the end product of a plant which the owner had erected as a ^‘heating
plant.” Thus, considerations of production of heat in steam which will l>e sub¬
sequently developed in connection with the production of mechanical power
are, for the most part, equally valid for heating plants. Although ue set out to
encompass the engineering of power plants, incidentally we also shall have
covered the engineering of heating plants.
We shall use the term “efficiency” fairly frequently. Unless specifically de¬
fined otherwise, efficiency will be taken to mean that fraction of the energy
supplied to a unit or a system which is finally usefully eonverted to the purpose'
for which the project was conceived. That is, efficiency is energy efficiency. In
its simplest form it is output divided by input. As power is the rate of energy
flow, efficiency may be expressed equally well as output power divided by input
power.
Man looks to the following sources for energy to make power:*
1. Fuels.
2. Flairing streams of water.
* Not including food, which body metabolism converts to muscular power.
ENERGY AND POWER 3

3. Ocean tides and waves.

4. Winds.
o. Solar rays.
6. Terrestrial heat.
7. Atomic nuclei.

Fvels. Currontly, fuels provide* more energy than any of the other sources
list(‘(l. Fuels arc substances having sufficient carbon or hydrogen for the
chemical oxidation to i)roduce, exothermically, worth-while quantities of heat.
The princiind fuels are coal, natural gas, and petroleum, and their derivatives.

300

in
q:
zy
200 ?
H
K
<
o

z
o
too 3
m

1920 1930 1940 1950

Fkj. 1-1 Origin of electricity used in the U.S. 1920-1950.

It is not expedient to enter into a detailed discussion of fuels here, but at the
appiopriate point the nature of fuels will be given further attention. Bituminous
coal, the principal steaming coal, is found in twenty-eight states of the United
States, and anthracite and semi-bituminous in half a dozen others. Although it
is believed that the United States contains more than one-half of the known
coal resources of the world, it must be realized tliat a great deal of this is un¬
marketable under present conditions of production. The relative proportions of
coal, gas, and oil as fuels are shown in Fig. 1-1. It will be seen that coal is the
principal fuel, of the stationary power plant. However, there are many factors
which enter into the relative use of different fuels. Labor conditions in the in¬
dustry which produces the fuel may generate differences in the economics of that
4 INTRODUCTION
fueFs utilization from one decade to another. Productive capacity of the com¬
mercial fuel industry also has a bearing on prices, but although these may in¬
fluence the initial selection of, say, fuel oil in preference to coal, once the instal¬
lation is made as a one-fuel plant, it is diflScult to shift to cheaper fuel, even if
changing economic conditions produce one. Although there has been a rising
trend to the use of fuel oil, much of it has been the result of possible temporary
surpluses of petroleum products, the convenience of use of a liquid, ash-free
fuel, and the uncertain status of coal production.
In a typical year, 1947, the production (in tonnes) of bituminous coal and
lignite in the United States was consigned as iollows:
Railroads 99.151^
Coke making 94,949.800
Electric power plants 78,020,700
Steel mills 9,115.300
Industry (general) 112,907,400
Domestic users* 89,958,900
Cement mills 7,141,400
Colliery, bunker, and export 66,030.«00
Total 557,275^ tonnes

* In addition, domestic consumers use many tons of anthracite coal.

Flowing Streams of Water, The energy contained in flowing streams of water

is a form of mechanical energy. It may exist as the kinetic energy of a moving
stream or as the potential energy of water at some elevation with respect to a
lower datum level, an example of which would be the water held behind a dam.
Hydraulic plants are slowly increasing in number, although the number of
new plants of this type built annually is quite small compared with those which
exploit heat energy. As a usual thing, the most desirable hydroelectric sites are
the first to be utilized; consequently, as ipore hydroelectric plants are built, the
owners must pay increasingly higher development costs From the standpoint
of capitalistic economics, it is often hard to justify the development of hydro¬
electric power in comparison with steam power, but from the standpoint of the
conservation of a fixed natural resource, namely, its mifteral fuels, it is obvious
that every effort should be made to harness the water power of the country,
since if unharnessed it goes to waste, whereas fuel, if unmined, remains intact
and undiminished in value in the ground.
Ocean Tides and Waves, That there is an enormous amount of energy in
waves and in tides is apparent to anyone who witnesses these oceanic phenom¬
ena. But here is a case where it is diflScult to harness and control energy, even
though it is in a high-grade form and one which is readily visible, and whose
magnitude is easily appreciated. The use of tides for tKe generation of electrical
energy might be practical in a few favorably situated sites where the geography
of an inlet or bay favors the construction of a large-scale hydroelectric plant.
To harness the tides, a dam would be built across the mouth of ihe bay. It
would have large gates in it, and there would be installed, in connection with
it, low-head hydraulic turbines. Capture of tidal power then could be accom¬
plished by opening the gates when the tide comes in, then closing them and
letting the water so trapped in the bay out through the turbines. After the
ENERGY AND POWER 6
tide has receded there is a working hydraulic head between the bay water and
open ocean. Some plans have been devised with the idea of turning the turbines
on the inflow as well as on the outflow. On account of the relatively low heads
developed (although there is certainly no dearth of water quantity), the size
of the hydraulic and mechanical installations is so enormous that the sheer over¬
all expense of such projects has been a strong deterrent.
Wave power remains practically undeveloped, with the exception of a few
experimental plants which have done little more than to indicate that it is
possible to devise systems for converting the mechanical energy of water waves
into electrical energy. There is an enormous amouflt of energy present in ocean
waves, but it is so distributed and variable that the development of it on a
large industrial scale presents complexities that have not yet been solved.
Winds. Power from the winds has served man for many centuries, but the
total amount of energy generated in this manner is small. The expense of instal¬
lation and the variability of operation have tended to limit use of the windmill
to intermittent services where its variable output has no serious disadvantage.
The principal services of this nature are the pumping of water into storage
tanks and the charging of storage batteries. Windmill power equipment might
be classified under four heads, as follows:
1. The multi-bladed turbine wheel.
2i The high-speed propeller type.
3. The rotor.
4. The Dutch sail type.
The foremost type in use is the multi-bladed turbine wheel, called the American
type in order to distinguish it from the older Dutch type. The efficiency of the
multi-bladed turbine wheel is about 10% of the kinetic energy of the wind pass¬
ing through it. It is said to be about 20% for the Dutch type. The propeller and
rotor types are suitable for the generation of electrical energy, as both of them
possess the ability to start in very low winds. The propeller type is more likely
to be used in small units such as the driving of small battery-charging genera¬
tors, whereas the rotor, which is rarely seen, is more practical for large installa¬
tions, even of several hundred kilowatts’ capacity.
Solar Rays. In a few instances the direct rays of the sun have been used to
generate power by absorbing energy first as heat and using the heat in some
type of heat engine. A serious fault of this source of energy is, of course, that it
is effective only during the day, so that if a continuous output is needed, some
large reservoir of energy, such as a storage battery or a heat accumulator tank,
must be drawn upon at night. Also, the output is handicapped if there is cloudy
weather. Nevertheless, there are some locations in the world where strong solar
radiation is received very regularly, and where the sources of mineral fuel are
either scanty or entirely lacking. Such locations offer more interest to the solar
power plant builder than the more favored regions of earth.
Terrestrial Heat Natural steam escapes from surface vents in many places
on the earth. Such natural steam wells suggest the possibility of tapping ter¬
restrial heat in this form and using it for the development of power. Unfortu¬
nately, the locations where the steam-producing substrata seem to be fairly
close to the surface are far removed from centers of civilisation where the
power could be usefully employed.
6 INTRODUCTION
Nevertheless, there arc probably many places where, although no natural
steam vents or liot springs are showing, deep drillings might tap a source of
underground steam. The cost oi such explorations and the great likelihood of
an Unsuccessful conclusion arc not very conducive to exploitation of this source
of energy. There has, however, been some limited use of terrestrial heat in a
steam power plant.
Atomic Nucleus, It is commonly accepted that the physical universe is
composed of molecules which, in turn, are composed of some 92 basic types of
atoms, called the elements. Customarily these are arranged for display in the
well-known Atomic Table* where the sequential numbering is by the atomic
number Z, Each atom is thought to contain a nucleus and one or more electrons.
The nucleus is itself composed of sub-particles, but all these are compactly
assembled at the center of the atom. The atomic dimensions are provided by the
electrons. In an imaginary super atom built like a sphere 6.1 ID in diameter, the
nucleus would be a tiny sphere of pinhead dimensions located at the center.
The stable nuclear particles arc the protons and neutrons, but other fleeting
nuclear jiarticles, such as mesons, neutrinos, etc., are postulated by the nuclear
physicists and believed to have existence. The atomic number assigned to an
atom is the number of protons in its nucleus. Its atomic mass number is the
sum of the neutrons and protons. Many elements have atoms with variations in
the number of neutrons. These variants are termed isotopes. Although there are
only 92 natural elements, more than 250 isotopes have been found in nature.
The isotopes of an element arc cliemically identical, since the chemical charac¬
teristics of an atom arc derived from its electronic structure and this is
determined by the protons, not the neutrons.
Within the atom nucleus there are tremendously large attractive and re¬
pulsive forces. A detailed description of the importance of these, and their
place in atomic structure, is beyond the present scope, but it is common
knowledge now that the magnitude of this kind of energy compared to the
mass of substance involved is incredibly higher than any other source men¬
tioned.
Some of the properties of atomic particles are set forth in Table 1-1.
Table 1-1. PROPERTIES OF ATOMIC PARTICLES
Symbol Rest Mass^ Weight (g.) Charge"^
Electron e'^ 0.000548 9.10 X lO"® -1
Neutron on^ 1.00894 1.67 X lO"®^ 0
Proton 1.00758 1.67 X 10~" +1
' Based on the mass of the oxygen atom 0^ — 16.00000.
•One electronic charge = 4.8024 X 10"’® electrostatic units, or 1.60 X 10"’® coulombs.

The nature of a particle or nucleus is written with a subscript and a super¬

script, thus tN'® represents a nitrogen nucleus with a charge of 7 and a compo¬
sition of 7 protons and 6 neutrons. The proton is the nucleus of hydrogen,
charge 1, proton 1, neutron 0.

♦See Appendix. In addition to the natural elements, some artificial elements have been
created.
ENERGY AND POWER 7
The combination of particles into atoms, or of atoms and particles into other
atoms, can result in end products of slightly smaller mass than the components.
Likewise the jarring of complex nuclei into fragments can end in products of
slightly decreased total mass. These fusion and fission activities are, however,
capable of liberating an enormous quantity of energy for they involve that
astronomically large number, the energy equivalence of mass.*
Thus if it were possible to assemble a helium nucleus 2H^ from 2 hydrogen
nuclei and 2 neutrons, the component masses would be 2 X 1.00758 4* 2 X
1.00894 = 4.03304. However, the carefully measured mass of the helium nucleus
is only 4.00279, making a discrepancy of 0.03025 mass units, or 0.0501 X
10-24 g.
Each atom of helium thus formed would be accompanied by the liberation of
0.0501 X 10-24 X 2*497 x 10’ kw hr, which is 0.00125 x lO*'* kw hr per
atom. This is, of course, an insignificant energy quantity when viewed individu-

0 50 100 150 200 250

ATOMIC WEIGHT-A

Fig. 1-2 Binding energy of atomic nuclei.

ally. But when assembled as, say, from a leg of helipm formed in fusion^ the
energy is startlingly great. The result is 0.00125 x 10"^® X 6060 X lO^® f =
7.575 X 10® kw hr per kg mol, or 1.894 x 10® kw hr per kg. These energy
(liia^ntitics per unit of the raw ingredients are revolutionary compared to the
entirely different realm of combustion energy we now tap so freely and which
is obtained solely by electron rearrangements, leaving the nuclei unchanged.
That this energy potentially exists is no longer doubted; however, someone has
yet to devise a means for obtaining it in useful form. It is conceivable that this
problem can ultimately be solved.
The energy that has just been calculated is the “binding energy,” liberated
when helium nuclei are formed. The corresponding energies of other atoms are
shown in Fig. 1-2. Man has been successful in liberating energy from elements
♦See page 14.
t This is the Avogadro Number in the Metric System. A mass of experimental evidence
places the number of molecules in gas at 6.06 X 10^ per gram mol.
8 INTRODUCTION
at the other end of the periodic table—but by fission rather than fusion. The
energy liberated is vast (atomic bomb), but not as great as by fusion. In fact,
were it not for the peculiar shape of the curve, no surplus energy could come
from fission. But because it is convex the binding energy of the unstable high
mass elements is less than that of stable elements such as barium and krypton;
hence the fission into these fragments can liberate energy.
Of course, it should be pointed out that this use of nuclear energy which
may be impending is representative of something that has always existed
Nuclear energy contributing to the welfare of man is not a new thing. From
time immemorial, the earth has been receiving nuclear energy radiated to it
from the sun. The origin of this energy at the sun is from nuclear reactions
which are now beginning to be understood fairly well. As an energy machine,
the sun depends on the fact that deep within it hydrogen is being converted into
helium for the production of energy, which is immediately absorbed inside the
sun as heat generated at terrifyingly high temperatures. This energy, on its way
out from the body of the sun is, of course, degraded to the surface temperature,
which is about 5000® C. Scientists and engineers ha\e only recently been able
to produce energy in the same manner, that is, from within the atom, and then
only explosively as a weapon. However, research is going forward at a rapid
rate not only on problems of improving on this source of energy as a military
weapon, but also in developing ways to generate industrial energy.
Except for atomic energy, and possibly terrestrial heat, all the sources of
energy may be traced back to the sun. The sun^s heat causes evaporation of
surface water to form rain clouds which continually replenish the flow of water
in streams. Gravitational effects account for tides Warming and cooling of dif¬
ferent parts of the earth^s atmosphere cause winds, and thereby waves. Solar
rays nourished tropical vegetation through the prehistoric ages, and from this
vegetation we inherit the deposits of coal which loom so largely as a present
source of energy. Likewise, some theories of the origin of oil and gas, although
not so definitely formulated as for coal, do imply that, without incident energy
falling on the earth from the sun^s nuclear furnace, such fuels would not have
been formed.
1-2 Background of Power Study. This study of power plant engineering
is not intended to be ^^from-the-ground-up.” The author will presume that the
reader has acquired a working knowledge of energy and its transfers and trans¬
formations by means of an adequate study of some basic introductory work on
applied heat power. Titles of several books that have been published in the
introductory field are appended to this chapter. The underlying heat engineer¬
ing here assembled is designed for review or refresher use and carries no claim
of intelligibility for the -unprepared reader.
Properties are basic physical quantities describing the state of a fluid sub¬
stance. They are:
1. Pressure, p,
2. Volume, v,
3. Temperature, L
4. Enthalpy, h.
6. Entropy, s.
To these basic five quantities others may be added, such as:
BACKGROUND OF POWER STUDY 9
6. Specific heat, c,
7. Density, d,
8. Internal energy, u.
These properties pertain both to gaseous and vaporous media. In addition, the
physical state of a vapor may require, for its definition, the dryness factor, x,
to be given. This is the fractional part of a sample of a wet vapor that is true
vapor. The liquid portion is, of course, equal to 1 — x. This factor has generally
been called “quality.” The change to dryness factor, which is made in this
volume, is in recognition of its increasing adoption by engineers. Quality may
be either dryness factor or superheat.
In general, the properties of gases may be derived from equations, such as
the General Gas Law, with the use of relatively few constants. However, vapors
do not obey simple equations unless very highly superheated; hence the need
for tabulations or charts of their physical properties. The unabridged steam
tables of Keenan and others* are recommended as the reference for the proper¬
ties of vapor and have been used in effecting the numerical solution of prob¬
lems in this book.

Pressure vessel

Absolute Zero

Fig. 1-3 Review of pressure and temperature scales.

Fluid pressures are ordinarily given in kg per square cm, being abbreviated
kg/cm* ab if an absolute pressure and kg/cm* ga if a gauge pressure. However,
a conversion of this fluid pressure unit into kg per square metre (kg/m*) is fre*
quently in order for dimensional homogeneity. Manometric pressures are
commonly expressed in mm of mercury (mm Hg) for which the multiplier to
obtain kg per square cm is 1.36 x 10-*.
Density is ordinarily used in kg per cub^c metre, abbreviated kg/m», and
specific volume is its reciprocal. Temperature scales are denoted by letter
symbol, thus:
F—Fahrenheit.
R—Rankine (absolute Fahrenheit scale),
C—Centigrade.
K—Kelvin (absolute centigrade scale).
^Thermodynamic Properties of Water Including Vapour, Liquid and Solid Phans, Keenan,
Keyes, Hill and Moore, Jeto Wiley and Sons.
 10 INTRODUCTION
Enthalpy is a tenn expressing the combination of internal molecular energy,
expansion work, and flow work. By definition.

Enthalpy, h = u + ^Joules per gram (abbreviated J/g){l-l)

J

Entropy measures the fraction of heat energy that is unavailable for ideal
conversion to mechanical work. Entropy has no absolute scale of values; it is
always the change of entropy that is pertinent. Where it seems to have a zero
origin a study of the data will reveal that entropy entries are relative to some
“datum"’ that was arbitrarily assumed to be the zero value. Thus the steam
tables show entropy of liquid water at 38® C to be 0.S458. However, the same
tables give entropy a zero value atO.Ol® CThe interpretation of 0.5458 is that
A« from 0.0Pto38®Cis 0.5458. Again, a subtraction of liquid entropy at 16® C
from that at 38® C obviously cancels the assumed data and yields:

*= 0.5458 — 0.2390 « 0.3068 Joules per gram K.

In a similar fashion, enthalpy subtractions cancel the datum.
Mechanical Work. Three common forms of mechanical work are:
1. The force overcoming a resistance linearly, or the torque applied rota-
tively.
2. The kinetic mass, either in form of a fluid jet or a solid body.
3. The potential mechanical energy of mass poised above a datum elevation.
If the rate at which the work is produced or consumed per minute is known,
the power is readily obtained, for one metric horsepower is 735.5 watts (—735.5
joules/sec). Examples of these cases of mechanical work will now be given.
Example 1: A force of 45 kg pushing against a body moves the body at a uniform
speed horizontally a distance of 4 m. The force, however, is applied at an angle of 30®
to the horizontal.
Mechanical work is done bv the force component which acts in the direction of
motion and equals the product of this effective force and the distance moved. In this
case,

Work » 45 cos 3(r X 3.5 « 155.9 kg m

45 cos 30® X 2
Were the body to be moved at a speed of 2 m/mip the power would be-
or 0.0173 hp (metric). 60x75

Example 2: A shaft rotating at 1200 rpm receives an applied torque (via gears,
belt, etc.) of 80kg*in. What is the received horsepower?

Power = torque X rotative sj^ed in radian units

Power SO X 2it X* 1200/60 x 75 « 134 hp (metric)
Example 3: A water jet which is 5 cm in diameter emerges from a nozzle with a
velocity of 230 m/s. The kinetic energy represented by one pound of water in this let is
/2mV , where m — 1/9.807 hyls. Hence the energy is % (1/9.807) X230* kg m per kg
water. To find power, note that there are ii>kg per min issuing from the jet, where w =
velocity X area* X density.
^ CrossHsectional area of flow.
BACKGROUND OF POWER STUDY 11
0.0S««
w 230 X X 1000
4
230* X 0^ ^

Therefore, power =
wsa
75
» 16240 hp (metric) » 11.94 Mw
Example 4: A rotating mass, say a turbogenerator rotor, has a total weight of
.11,000 kg. and a composite radius of gyration of 50 cm. The rotative speed is 1800 rpm.
The kinetic energy is in which I is the mass moment of inertia of the rotating
system, and o> is the angular velocity in radian units.

I = mr*(r = rad. of gyration)

I = X 0.5* »• 280.5 kgm/sec*

1 /1800
Energy = -X 280.5 X ( — X 2w 4983300 kgm
)■-
Were this rotor to be slowed from 1800 to 1780 rpm in 10 sec, the average power
released would be — a>2i7»)/10 kgm per sec

i X 280.5 X (^YdSOO* - 1780>)/10

Power =-^“ 147 hp(mctric>=108kw
75
Example 5: A 1400 kg automobile having a speed of 72 kmph possesses a certain
store of kinetic energy due to this motion. To determine this energy, one uses the com¬
mon kinetic energy expression,
For the data given.

Energy = -j" ^ ( ^806 }* ” ***"

This is the energy, regardless of the direction of travel.

Example 6: The potential energy of a mass poised above a datum plane is illustrated
by the rise of an elevator weighing 700 kg from first floor at grade 0.00. m to fifth floor
at grade 14.8 m. The energy expended in making the lift is the simple product of
weight and height, or
Energy ~ 700 x 14.8 = 10,360 kgm
This energy which was expended as a force overcoming a resistance then resides
potentially in the mass of the elevator while it remains at grade 14.8 m^

Example 7: The power represented by a river flow of 5m*/iecat a site where a 40 m

head can be developed between reservoir and discharge water levels is another example
of potential work. Taking the discharge level as datum, each cubicmetrebf water in the
reservoir carries potential energy of 1000 x 40 — 40,000 kgm.
The potential power is 40,000 x 5/75 ** 2667 hp (metric) « 1961 kw.

Heat The common unit of this form of energy, as employed by most engi¬
neers, is the Joule, abbreviated J. This is the unit used in the steam tables,
also to express such quantities as calorific value of fuels^ heat transfer
rates, and the like. Heat can exist internally in a fluid as random mloec-
12 INTRODUCTION
ular motion, and externally as a volume under pressure. Also, quantities of
heat energy are involved in processes accompanying a change of physical state
—as from liquid to vapor, or solid to liquid. Transferred heat energy, desig¬
nated by Q, is definitely associated with temperature changes in gases, but may
occur at constant temperature in vapors. The transferred heat may be shown
to equal the change of enthalpy if the process occurs at constant pressure.
The Ah of a gas is therefore always CpAT, Constant pressure changes are a
common occurrence in steam cycles; for example, the processes in boilers and
condensers are of this nature. The heat transfers in these cases are therefore
readily evaluated as the differences in enthalpy read from tables.
It is the function of a heat power plant to convert as much of the heat
energy produced in it into high-grade mechanical and electrical energy as is
economically feasible. The supply of heat energy is commonly produced from
combustion, a chemical reaction of exothermic nature. Of all of the reactions
which are exothermic, those of carbon and hydrogen with oxygen have been
most in the service of man. In fact, except for some quite recent experiments
in exothermic reactions for rocket propulsion, the manufactured heat supply
has been developed with these simple oxidation reactions:

C + JO2 = CO + 10,304 J/g C (1-2)

C + O2 = CO2 + 33,820 J/g C (1-3)

CO + iOj = CO2 + 10,187 J/g CO (1-4)

H2 + i02 = H2O + 144,212 J/g H, (1-5)

Substances containing these elements in either pure or compound form in

sufiicient strength to result in a high temperature reaction are the fuels of
commerce.
The higher heating value, or calorific value Qu, of a fuel is the heat given
out during combustion, with equal initial and final temperatures, usually at
constant pressure. It includes the latent heat of any water vapor present when¬
ever the fuel contains H2; and which is necessarily condensed. If a deduction
is made for the aforesaid latent heat for the reason that it is not available in
an engineering process, then the net heat is termed the lower heating value,
Ol. For some time it has been the common practice of engineers to use Qh as
a standard in the external combustion field, and Qr, in the internal combustion
field.
Electrical Energy, The common electrical form of energy is a flow of
electrons forced against or urged along by an electrical potential (voltage).
The common unit of potential is the volt. A coulomb is a unit quantity of elec¬
trons. A flow of one coulomb per second is the ampere of electrical measurement.
The product of the flow and potential is the basic unit of energy, one volt-
coulomb being a joule. The corresponding unit of power is the joule per second,
called the watt. As this is a small unit in terms of modern electric power utiliza¬
tion, the standard electric power unit for the power plant field is the kilowatt,
i.e., 1000 watts. In a simple d-c electric circuit, the watts of power are the
product of volts and amperes. Where the current is alternating in character,
BACKGROUND OF POWER STUDY 13
and the circuit not necessarily single phase, power calculations include consid¬
erations of power factor and polyphase arrangement.
Radiant Energy. Radiant energy may consist of particles in motion through
space, or it may exist as a propagated wave. Radiant heat energy is wave-like
in clyiracter, being defined by its frequency, /, in cps, and its wave length, A,
in cm. The product A/ equals the speed of light. High numerical values of /
are characteristic of energy radiations; hence the frequent use of kilocycles and
megacycles to describe frequency.
Radio, radar, heat, light, and other radiations are all of the same family.
They have principal characteristics in common and can be arranged in a con¬
tinuous series according to wave length or to frequency. In brief explanation
of the electromagnetic wave, recourse is had to the concept of a “field.” A field
is defined as a region in which a particular kind of force is exerted. An acceler¬
ated electric charge produces a changing magnetic field, which, in turn, also
produces an electric field. An electromagnetic wave is an oscillating electric
field and a similar magnetic field, each depending on the other for existence.
A wave “source” in some way accelerates electric charges. This produces a
magnetic field in the vicinity, and an electromagnetic disturbance is propagated
outwards. The source which has been mentioned here may actually be:
1. Molecules undergoing combustion.
2. Electric discharges in gas.
3. Electromagnetic oscillators.
4. Molecules in an active kinetic state—^temperature radiation.
5. Nuclear disintegration—gtoma radiation.
A significant feature of all electromagnetic waves is that their velocity
of propagation is uniformly 3 X 10'® cm per sec. Their frequencies, /, and their
wave lengths. A, extend over an enormous numerical range, but for all,

A/ = 3 X 10^° cm per sec

The “radiant energy” of thermodynamics consists of electromagnetic waves
having frequencies between 1 X 10" and 7 5 x 10" cps. If these waves fall on
a body or a substance, they may be reflected, absorbed, or transmitted. A
“perfect blackbody” is one which would absorb all the incident energy, leaving
none of it reflected or transmitted, but perfect blackbodies are difficult to
achieve. A good “radiator” may be possibly 90% blackbody.
The interchange of energy between two parallel planes is given by the
equation
Q = 20408.4 X 10-» - T^) J per m« per hr (1-6)

in which temperatures are Kelvin, and F® is a factor connected with the black¬
body qualities of the surface. F« is designated the “emissivity factor.” For non-
planar and nonparallel bodies, a shape factor also multiplies this equation.
Actually, the production and absorption of electromagnetic radiation ac¬
count for heats of combustion, for the waves are produced by rearrangements
of molecular electrons occurring during the chemical reactions we call com¬
bustion.
Transformation of Energy, Since work, heat, electricity, all are forms of
14 INTRODUCTION
energy, they are mutually interdiangeable at fixed ratios. The kgm is 9.807
Joules, and one joule is 0.102 kgm.
Electrical energy can be converted to heat by passing the flow of electrons
through an electrical resistance, R (measured in ohms).

Power = PR watts ^ (1-7)

From this the following electricity-heat relation is developed:

q = PR/AMI kilocalories per sec (1-8)

A metric horsepower is 735.5 watts. A kilowatt hour is 860 k cal. When
energy is changed from one form to another, it is done at these ratios.
When, subatomically, mass goes into energy or vice versa in accordance wit!
the Einstein theorem, the equivalence is of the order of
91.7 X 10^^ kgm per gram mass

1-3 Working Processes. To effect the transformation of high-grade

forms of energy such as work or electricity into heat is quite easy (friction,
resistance heating), but to upgrade heat energy into the higher forms is difficult.
This fact is described in the familiar Second Law of Thermodynamics. En¬
gineers have, in general, used fluid processes to secure the action, although they
have learned to expect that less than one-half of the heat energy can be trans¬
formed. The upgrading actions use an expansible fluid which is caused to
undergo, if possible, a reversible adiabatic type of process, during which it
decreases in pressure, temperature, and enthalpy, is theoretically unaltered in
entropy, and increases in volume. The ideal adiabatic process is one which can
be carried out in thoroughly heat-insulated equipment. If an ideal, nonviscous
fluid is assumed, the process may be completely reversible, that is to say,
isentropic. On the other hand, if frictibn and turbulence are present to a large
degree, and no work is performed, the process is completely irreversible. Such
are known as throttling processes. Between these two extremes may be found
processes of various degrees of irreversibility.
Maximum transformation of the heat, initially
indicated by entropy as being potentially avail¬
able for work-making, is secured by the
completely reversible type of process, which,
therefore, becomes the goal of designers of
engines and turbines. The extent to which the
irreversibility of friction, turbulence, and heat
transfer due to cooling action are kept out of
the process is reflected in the thermal efficiency
Fig. 1-4 The polytropic gas secured and by other measures of performance.
expansion process. On any chart having entropy as one axis,
ideal work-producing processes will appear as
lines perpendicular to the entropy axis, desirable real processes will be those
of minimum entropy change, and inefficient processes will exhibit large in¬
creases of entropy.
WORKINO PROCESSES 15
Hie working nonflow processes for ideal gases are known as the polyUvpic
proceaaes. The family equation is:
pV = C (1-9)
The range of numerical values for n is infinite; however, n’s of 0,1, Cp/Cr,*
and infinity, are important, for they fix the processes asoither isobaric, isother¬
mal, isentropic, or isometric. A table of equations applicable to the polytropic
process for an ideal gas is given in the appendix. Calculations for real gases
may also be made with these equations, provided the temperatures are not
extreme, say in the range from 0®-260®C. Otherwise, special charts, tables,
coefficients, or empirical equations of specific heat must be employed.'
Example 1: Air at 82® C and 3.S kg/cnP ab pressure expands isentropically to
1 03kg/cm*ab With this information, the volume of the initial state, and the other
properties of the final state, may be computed, as well as the conges in other
properties created by the process. The equations of Fig. A-17 of the appendix will
be ctmsulted when necessary.
Here p, = 3,5 X Ifl* T. = 640® R, p, = 1.03x10* y = 1A,R = 29.3355It
»i = RTi/pi = ».3 X 335/3.5 X 10* - 0,297 m*perkg
(iiW0.2»7)>-« = 3,5/l.0^whence e, = 0.712 m* per kg
640/Tj = (3.3/1.03)“*^; Tj - 249 K (-2^0.
Since the process is adiabatic, Q = 0
A« = wCrAT = 0.71(335 - 249) - 75.26 J per g
M » taCpAT ° 1.00(355 - 249) - i06 Jper g
Work = (3.3x10* x 0.297 - 1.03 x 10* x .0.712)/(1.4 - 1) - 7653 kg/mp«kg
Investigations of properties and processes of a vapor require the use of a
table of tlie thermodynamic properties of the vapor. It is also convenient to
employ charts of tlie properties, which charts may be constructed from the
tabular data. The common properties of a vapor are pressure, volume, tempera¬
ture, enthalpy, and entropy. Temperature is dependent on pressure if the vapor
is saturated; hence it is not an independent variable property as it is for the
case of a superheated vapor. In its place, one uses the other quality, dryness
factor. The reader is presumed to be familiar with the arrangement of these
properties in vapor table^, at least for water vapor.
The enthalpy of liquid water is generally taken directly from the saturated
steam temperature tables, although the actual pressure may be above satura-'
tion. For example,at271C'the tables give hf-113.25 Jpefgat saturation pressure,
O.OS€34-kg/dii*ab. Nevertheless, one customarily assumea that the enthalpy of wa •
ter at27®Cat higher pressures, say atmospheric, is 11325 J perfl. Actuidly, the
correct enthalpy at 27®C and 1.09 kg/cm* ab is 113.25+ the flow work, vLp.
0.102
To illustrate this, Vt at 27° C is 0.0010035 in*/kg; hence an enthalpy incremoit of
O.OO10O35(l.O3xio*-363i4VO.lOZXJtP«« &09m Jahoutd be added to 113.2SJ
toobtainthe exact enthalpy of water existing above the saturation pressure

* Symbol y is assumed to this ratio.

16 INTRODUCTION
corresponding to its temperature.* Unless pressures are very high, the correc¬
tion is negligible in view of other tolerances in heat power work. However, the
trend of steam power has been to higher jircssures, and this trend should
always be kept in mind, to be taken into account whenever necessary.
The Mollier Diagram is a chart of all the common properties except volume.
Basically, it is an enthalpy-entropy graph, with contours of pressure and
quality. The arrangement is illustrated in Fig. 1-5. This chart is valuable in

mollier diagram temperature-entropy diagram

Useful for fracing Processes and Useful in studies of Cycles
determining States

2900

2800

2 2600

2500

ENTROPY PER g ENTROPY PER g

ABOVE ®C ABOVE

Fig. 1-5 Vapor charts: a. isothermal process; b. iscntropic process; c. isobaric process;
d. throttling process.

the solution of various vapor processes, and the reader should endeavor to
familiarize himself with the form of the processes as they appear on this chart.

Example 2: From steam tables, determine the following properties of steam:

(a) Saturatitm temperature when the pressure Consult the saturated
steam table, having pressure the independent variable. The temperature listed for this
pressure is 181.16® C.
(b) Total temperature cklOSkglivfab and IKT C superheat. Saturation temperature
for (a) was 181.16® C; then total temperature is 181.16 -f 110 = 291.16 "'C.
(c) The enthalpy and volume of wet steam having 10% moisture at 10,5 kglcm^ ab,

/i = /if -f x/ifg = 768.46 + 0.90 x 2010.7 = 2578.1 J per g

♦This correction assumes that water is incompressible within the pressure range em¬
ployed. Where this is not permissible or for very high pressures, Table 4 of the Steam
Tables may be used.
WORKING PROCESSES 17
Alternately, ^ ^ - (1 - x)hf^ « 2779.25 - 0.10 x 2010.7 - 2578.2 J per g
V ^ vt + xvtg « rwg + (1 — x)v{

Except at high pressure, Vf is quite small compared to Vtg and the second term can
be dropped, making v = xVg. Do not use this approximation for very high pressures or
moisture contents.
i; = 0.90 X 189.22 - 170.3 cm*/g
(d) The entropy of steamaJt los Aglciff-aSoind superheat. The volume, enthalpy,
and entropy of superheated steam will be found listed in superheated steam tables.
These will be found to be double-entry tables, since there are two independent variables,
viz., pressure and total temperature. In this case, the table must be entered AtlOL8kg/om8
aband291.l^C.Thistemperature cannot be expected to be a standard tabular entry, so
an interpolation is in order. The result is an entropy of 7.0758 J per g pcr®K.
In many cases, neither pressure nor temperature will coincide with tabulated entries;
then a double interpolation is required.

Example 3: By use of a published chart of enthalpy vs entropy for steam, trace

the following steam processes:
(a) An isobaric process from an initially dry and saturated state at 19C until
16T* C superheat is obtained. Then find t^h. Locate the initial state by the intersection
of the “saturation line” with the 193* C total temperature line. This lands on 2788 J/gand
18.8 kg/omS afarpressure,which, if followed along the pressure line to >67®C superheat, gives,
finally, 3170 J/g enthalpy. The change of enthalpy for this process is:
^h = 3170 - 2778 » 382 J/g
(b) An isentropic expansion of steam witiaUyat 31(^0
Thedrynessandenthalpyof the steam at the conclusion of the process may be found by
first locating the initial state with pressure and temperature lines, then tracing along
constant entropy to intersection with the final pressure, kg/cm* ab- The point
representing the final state will be found to occur at an enthalpy of 2148 J/g and a moi¬
sture content of 18.9%. The dr3riies8 factor is, therefore, 0J311.
(c) A constant enthalpy or **throttIina'^ process fronianinUialstateof7iji/ctJ‘CL^
dry and saturated to a final pressure,After tracing the process, deterwh
ine^As.The following procedure may be followed on the chart. Locate the initid state
by the intersection of 7 kg/om* pressure with the saturation line The corresponding
entropy is 8.71. Then trace along constant enthalpy to intersection with CAkgjcmK
Here read 8.81. Then Aa=: 0.1J/g/^k.
Example 4: Determine the magnitude of the error made in assuming the enthalpy
of subcooled liquid water to be that for saturated liquid at the temperature of ITT* C*
Take the water as being under a pressure of 34 kg/oms ab» (This state might Well be
that of the feed water for a high-pressure boiler.)
At ITT* C , ht from tables = 862.69 J/g* This heat is the exact enthalpy of
water at 177* C and 9jS3llSkglmfi ab^The pressure difference of 84 — 9.538iniplies a flow
work of t;Ap/0.102.
The correct enthalpy for the state mentioned is therefore:
h == 862.69 f 0.0011235(84 X 10* - 9S380)/0.102 X 10^ » 870.89 J/g

M The Power Age. The measuring stick of man’s progress in this

world has many facets. In the spiritual and artistic fields, man has progressed
as well, if not better, ia ages where the masses were living very rudely, but
where the classes had leisure and freedom from toil and the vexations and
18 INTRODUCTION
turmoil of life in this more hif^y civilized era. The production of artistic
masterpieces in all fields of art was fully as great in some of the past ages as
at present, and we here have no intention of extolling the power age as the
golden age of civilization. We do wish to enter the claim, however, that only
in modem times have the masses of people in any country been placed in
possession of leisure time, comfortable living, adequate food, and a share of
entertainment and educational opportunity. This advantage, we think, is one
not to be taken lightly, and reflects enormous credit upon the power age.
Historically, the progress of mankind in the production and application of
power might be divided into three phases. First came a long era of toilsome
labor, in which slowly man learned to build many and various tools—^in the
beginning, just those which came immediately to his direct uses, that is, the
hand tools with which the getting of food, clothing, and shelter were somewhat
simplified. Then, in the second phase, certain inventions such as the wheel
and the subjugation of the labor of animals to man’s use greatly extended his
bodily power. In the latter part of this era, man’s scientific talents began to

Fig. 1-6 Growth of population and power use—U.S.

discover and to interpret certain of the laws of nature upon which could be
built that era in man’s historical career that we call the power age, and in which
we are now living. This power age marks the success of man in acquiring a
mastery over the forces of nature and an ability to create mechanical slaves
in the form of electrical or mechanical power-producing units. The last half
century, although not the entire scope of the power age, does cover that part
of it. in which the greatest progress has been registered. The kilowaU of
mechanically produced power per worker has increased several fold in this time.
This increase, of course, usually produces an increase as well of the productifin
per worker and a greater store of consumer goods for the average indivdual.
In 1900, the kilowatt in basic industries per worker was about l.Skasrin 1950
it is up to aboutS.8kw;It is this increasing availability of the labor of nonhuman
“slaves,” coupled, of course, with improvements in production machinery, that
spells the ever-widening variety of consumer products available to the average
man.
The introduction of mechanical power soon produced a centralization of
industry near the source of its power. This was because mechanical power is
THE POWER AGE 19
not readily transiAitted efficiently over long distances. The industrial revolu¬
tion therefore led to poor living conditions due to overcrowding and, had it
continued, would have produced far greater social problems than we have in
this day and time. But then came electrical energy in the alternating-current
form. It offered flexible and economic transmission of energy from a central
generating station to distributed industries. Decentralization became possible
and has occurred very extensively, thus markedly bettering the working condi¬
tions of workers of the power age over those that were predominant before
the long-distance transmission of energy was economical.
This is one of the reasons why much of the energy used in industry is being
generated centrally by corporations whose business is the production and sale
of electrical energy. Industries find it possible to locate in small communities,
and even rurally, and enjoy the advantage of adequate power transmitted to
them electrically from gen^^rating stations which may be located many miles
away. Smaller factories and shops can produce competitively with larger firms.
Small-scale enterprises often create a greater sense of responsibility and a feel¬
ing of individuality on the part of the average workman. The heart of an
individual system such as this is the central electric station, with its radiating
network of transmission lines, instantly ready to deliver power at a low price
—so low, in fact, that oftentimes large industries find it unprofitable to gener¬
ate their own power.
The application of power to the service of man has been the means of
elevating his standard of living. Where power has been universally and exten¬
sively employed by the population, the standard of living is highest. Unfor¬
tunately, also, such societies exhibit the greatest complexity of the social
system. The influence of mechanical power is felt today mainly in transporta'-
tion. Most of the activation of the ^'wheels of industry'^ nowadays is with the
electric motor. In agriculture, powered farm machinery has made it possible
for a farmer to cultivate and harvest ten times or more the acreage that he
could have handled without tractors and mechanical planting, cultivating, and
harvesting equipment. The domesticated draft animal is by no means an obso¬
lete source of power even today, but most of the intensively cultivated areas
or the United States do exhibit a definite trend from the draft animal to the
mechanical power unit. Then, of course, all large-scale methods of rapid trans^
portation of passengers and freight, by land, sea, and air, are, for the most
part, motivated by mechanical power units, although a certain amount of
surface transportation is accomplished by electrical power.
In communication lines, the telephone, the telegraph, also radio, radar, and
television, are certain to be included in any list of developments which have
added to the safety, comfort, and enjoyment of living. Rapid and reliable
communication is essential to the conduct of modem business. These are all
provided with electrical power, but in relatively small amounts as compared
to its use for industrial power, lighting, and heating.
It was in the field of lighting that the central station production of power
made the first headway. The first commercial station began operation in New
York about 1882. This was steam driven and was soon followed by a hydro¬
electric station in Wisconsin. The incandescent lamp invented in 1879 gave the
20 INTRODUCTION
impetus necessary for the rapid expansion of this infant industry. Seventy
years later the number of customers exceeds thirty million,* and the industry
that provides electrical energy to the general public customer is one of the
biggest and the most essential in these United States. Although the uses of
electricity are manifold and are increasing daily, lighting in various forms has
continued to be the main source of income of the public utility company. Be¬
cause electric lights are so convenient and so superior to all other sources in
producing adequate illumination, and because that form of lighting offers a
minimum hazard of fire, it has practically displaced all other lighting systems;
old fashioned, indeed, is the home, institution, or industry which is not today
illuminated electrically.
No more striking instance of the part played by electric power in elevating
the standard of living can be found than in the convenience of electric house¬
hold appliances. Cooking utensils, fans, refrigerators, stoves, vacuum cleaners,
and various small motors all help to make modern life more comfortable and
free countless people from drudgery.
The knowledge of the production and utilization of power should hold, we
think, more than usual interest for the engineer and for the student of engineer¬
ing. It is hard to imagine any other single factor which has had more influence
in shaping the present social structure of the United States than has mechanical
and electrical power. The special interest of the engineer, in particular, is that
his is the responsibility for a state of society that is requiring the services of
trained technical men to an ever-increasing extent. While the generation and
marketing of large blocks of power itself furnishes direct employment to many
engineers, a comprehensive view of modem engineering activity reveals that
nearly all of it is dependent upon an abundant supply of power.
1-5 Power Plant Engineering and Design. It is already evident that
the type of power plant to be treated in this book is the stationary plant which
produces electrical energy. With minor exceptions, the term “Power Plant^^
may be taken to imply an Electrical Power Generating Station. As noted before,
this type has developed into the most important power-producing unit. The
power plant must function as a unit, not as a collection of individual pieces of
equipment. This statement is significant, for it governs the method of treatment
of the subject in the following pages. Concentration on the external operating
characteristics of each piece of apparatus and a study of its relation to the
rest of the plant will eliminate the apparently endless maze of details which
accompanies a study of power plant equipment from the individaul design
standpoint. Power plant engineering, as construed here, would not include the
design of a steam trap, or of a motor or crane, but rather the selection of a
trap, motor, or crane manufacturer’s product for the specific service in mind.
Briefly, it is the art of selecting and placing the necessary power-generating
equipment so that a maximum of return mil result from a minimum of ex¬
penditure over the working life of the plant; and the operation of the completed
plant in a manner to provide cheap, reliable, and continuous service.
The scope thus set forth, it will be noted, implies the placing of professional
en^neers in three categories, viz., (1) Designing Engineers; (2) Construction

*26 'million residential; 5 million commercial and industrial.

POWER PLANT ENGINEERING AND DESIGN 21
Engineers; and (3) Operating Engineers. In some instances the same person
has successively served in all three categories on the same project, although it
is more common to find a person acting in the capacities of only the last two
mentioned. Corporately, many examples can be cited where the three phases
of work mentioned are performed by separate firms, but also other cases can
be found showing partial or complete combination. For example, a certain
new power station might eventually be brought into production by being de¬
signed by a firm of consulting engineers, then erected by a construction con¬
tractor, and finally operated by a public utility company. Alternately, the
large utility company has been known to design in its own engineering offices,
supervise the construction and the work of various separate contractors, and,
of course, finally operate the station. Other possibilities are (1) the design and
construction by a firm of engineering contractors who finally turn over a work¬
ing plant to the Public Service Company and (2) design and operation by
company engineers with erection by a construction contractor.
Except for the smallest industrial or institutional units a power plant design
is an exceedingly complex activity. A visit to a public service power station and
a view of the maze of equipment there is sufficient to convince anyone of the
need for complete and comprehensive initial planning. An adequate portrayal,
by drawings, schedules, reports, or letters, of the designer's conceptions will
insure that the design is equally intelligible to all the others whose labors are
essential to the building and operation of the plant. Also, clarity of the written
record is highly desirable in consideration of the contractual agreement system
usually employed to get the work done. The design engineer's work will become
part and parcel of legal documents and he should keep this in mind constantly
while proceeding with the drawings and specifications.
The design drawings must always be preceded by a considerable amount
of engineering work. Preliminary estimates are made and bids prepared, for
most power plant construction is contracted for through the processes of
competitive bidding The leading dimensions of the plant are usually well
established at the time of preparing the preliminary estimates; however, they
are subject to change during the progress of the design.
Assume that a bid has been submitted on a certain project and the contract
is awarded. The method of handling the subsequent design rests almost entirely
on the size of the job. The small plant can be completely designed by a few men
working in the same office. Employer and employee often labor at adjacent draw¬
ing tables, and the fine distinctions of engineer, assistant engineer, draftsman,
checker, or designer which play an important part in the large organization do
not exist. Each man is in continual touch with the work of the others, and
cross-reference or cross-checking becomes a simple matter. There is no necessity
for an elaborate design office organization and routine because of the limited
number of men at work on one project and the comparative simplicity of the
small job.
On the other hand, when a large station design is undertaken, failure to
set up a workable procedure for handling the routine features of the design
job is courting confusion and much waste of effort in the drawing rooms. The
result is high drafting office overhead cost. Those who are unacquainted with
the magnitude of the design labor required for even a modest^sized central
22 INTRODUCTION
station (say 15,000 kw) cannot be expected readily to appreciate the diffi¬
culties that attend securing a thorough check on ail the hundred or more draw¬
ings so that the final blueprints, as sent to the field, will be a clear set of in¬
structions for the erection of the station. Nothing must be left to guesswork
on the part of the erectors; and no conflicts or interferences, such as conduits
crossing a space where another print shows a ventilating duct, should appear
during the construction period. Of course it is a rare occurrence when some
changes, additions, or corrections do not have to be made on a set of drawings
for a large station after they reach the field. Ordinarily, a small field drafting
office is maintained on the job location to take care of these field changes. The
personnel of the design office should render a quality of work such that the
field office is confined to field changes and does not have to supply omissions or
make corrections on the design proper.
As many as 50 or 75 men may be employed at one time on the design of a
large power station. They will take from six months to a year to design a sta¬
tion and complete the drawings. The attempt should always be made to
organize the office so that these men will be employed steadily and the design
will proceed logically from beginning to end. In addition to this, a procedure
must often be worked out to correlate the office with the field work, because
sometimes the construction work starts as soon as the main drawings (layout,
foundation, etc.) are completed and while detail drawings, wiring plans, etc.,
are still in preparation. The order in which the drawings are to be completed
must be established in advance so that they may pass from the office to the
field as they are needed during the course of construction. Control of this
feature can be secured by interlocked drawing and construction Progress
Schedules, backed up by insistence that those in immediate charge of work
progress meet the schedule '‘or else.^’ Obviously those entrusted with the draw¬
ing up of Progress Charts need to have had much actual experience in this line
of work.
The organization of a large firm engaged in engineering, construction, and
supervisory work is shown in Fig. 1-7. The organization of departments re¬
sponsible for engineering and construction are carried out in more detail than
others. This diagram pictures the entire organization and not that which may
be assembled for one particular job. It is about as extensive an organizational
plan as will be found and in practice one would find various modifications, all
tending to consolidate functions shown separately in Fig. 1-7 and thus simpli¬
fying the organizational plan. Most of the detail design work is done by design
and drafting groups consisting of designers, checkers, detailers, and tracers
working under a chief who is generally classified as an Assistant Engineer.
Let us assume that a firm of contracting engineers has been awarded a
contract for the construction of a large power station, to be ready for operation
by a specified date. The first step after the award of the contract is the assign¬
ment, by the Engineering Manager, of the engineers who arc to work on that
particular job. One of these engineers will be responsible for the entire job,
not only for its engineering design but for its cost and success. He is the Super¬
vising Engineer and is usually a mechanical engineer in the case of a steam
station design. The first step in the engineering department is the preparation
of estimates, which are essential in order to keep each phase of the plant within
 TREASUA£fi

Fig. 1-7 General organization of a large engineering-contracting corporation.

24 INTRODUCTION
a certain allotted amount, so that the entire cost will not exceed the anticipated
figures upon which the contract is based. Each engineer prepares an estimate
for his class of work and his estimate is reviewed by his chief engineer. The
separate estimates are then sent to the Chief Estimator who assembles them
and adds the proper amount for engineering and drafting. The total estimate
is then reviewed by the Engineering and Construction Managers.
The next step is the preparation of schedules for construction, drafting,
and purchasing. These are worked up backwards; that is, the date on which
the plant is to be finished is the starting point and everything is scheduled

Show in red on light line percentage of work completed.

Show date of report by heavy vertical red line.
Show estimated time schedule in black on upper light line.
Show in red on lower light line actual time spent and estimated time required to complete.

Fig. 1-8 Construction progress chart covering the main subdivisions for a 60,000-kw
power plant. The chart is revised periodically and the percentages of work completed
are indicated for each revision.

backwards from that point so as to know when construction must start to meet
this date, when the drawings must be started in order to be completed in time
for construction purposes, and when specifications must be started in order to
receive bids with the proper time allowance for the placing of the bid, drawing
up the contract, manufacturing the equipment, and delivering it to the plant
site so that it is on hand when the date of erection arrives.
The respective engineers are then put in possession of the main points of
design and the dates upon which certain information must be ready for the
field. They and their assistants elaborate on the main details, imtil they can
pass on to the chief draftsman a fairly complete outline of the chief features
of the station. The chief draftsman in conjunction with the various engineers
DRAWINGS, SPECIFICATIONS, AND CORRESPONDENCE 26
then makes up a list of drawings which, to the best of their knowledge, will be
required for the complete delineation of the station design. In making this list
they must draw heavily on experience and judgment at this early stage. A
Drawing Progress Chart, similar to the Construction Progress Chart shown
in Fig. 1-8, but, of course, with appropriate headings, is then prepared. Suf¬
ficient technical men are then assigned to the work to meet the progress
plan. The design groups may consist of designers only, to begin with, but as
the work gets going, checkers, detailers, and other draftsmen are added as
needed. Meanwhile the engineers have made out a purchasing schedule in order
to deal with manufacturing contractors or subcontractors on a chronological
basis.
Designers are first engaged in preliminary studies and sketches which de¬
termine the rough characteristics of the plant and, when approved by executive
engineers, serve as an outline to guide preparation of final drawings. In the
course of final design work there is necessarily a considerable amount of cross
checking to be done between different individuals or groups so that interfer¬
ences will be eliminated and information can be mutually exchanged. This is
a formal routine on a job of the size we are now contemplating, but could be
quite informal and spontaneous in a small design office. Needless to say, the
various engineers must keep closely in touch with the drafting room to watch
the progress of the drawings and to direct the work. This is necessary in order
that the plant have embodied in it the features they desire, and to insure not
only a plant of the caliber contemplated, but one which may be constructed
within the amount allotted in the estimate.
The efficiency with which drawings can be produced is greatly dependent
upon the close attention the engineer gives to the drafting. Otherwise it is
likely to mean that the engineer gets into the position of “pulling apart” drawings
instead of directing them. Also, the prompt attainment of manufacturers^ prints
by the engineer is highly essential to good drafting room efficiency.
1-6 Drawings, Specifications, and Correspondence. A design or con¬
sulting engineer's plan for the construction of a power plant is set forth in his
drawings and specifications. Specifications are a written set of rules, regula¬
tions, explanations, standards of quality, etc., which accompany a set of draw¬
ings. Many items of information the engineer wishes to convey are far more
cheaply and conveniently set forth in multigraphed typewritten booklets than
blueprinted from hand-lettered drawings. Conversely, there are other matters
that can be briefly and succinctly shown by drawing which would re¬
quire unreasonably extensive wordage to convey in writing. A combination of
drawings and written specifications is more compact and definite than either
could be if relied on entirely. The drawings and specifications should be ade¬
quately cross-referenced so that the reader will be able to integrate the
information they contain with minimum effort.
In general, specifizations carry the following type of information and direc¬
tions:
1. General infonnatiem on what the contractor can expect at the site and the condi¬
tions of his use of the premises.
2. Basis of legal relations between contractor and purchaser. Conditions of payment,
insurance, guarantees, arbitration, etc.
26 INTRODUCTION
3. General description of the planned project, expected mode of operation,
sequence of construction, etc.
4. Standard of workmanship expected and method of insuring that it is furnished.
6. Qualities of materials, performance of components, etc.
6. Schedules of equipment in multiple which would overload the drawings if at¬
tempted to be placed there.
7. Conditions of acceptance tests.
Insofar as possible, the engineer should restrict his instructions to a con¬
tractor to ^^what-to-build,” and not be drawn into specification of *^how-to-
build,” for he cannot enter into the latter activity and still expect the responsi¬
bility for the complete job to rest on the shoulders of the contractor.
Similarly the engineer in his dealings with equipment manufacturers should
be careful not to write specifications that in the end could be legally interpreted
as relieving the contractor of responsibility for performance because the en¬
gineer had not only specified results, but also the design itself.
The major items of equipment in a power station are generally purchased
through competitive bidding. Since the basic principles of this form of buying
are well known to most persons, they need not be dealt with here. Equipment
specifications should set forth, as briefly and concisely as the individual case
permits, the requirements of the purchaser and the information which he desires
the manufacturer to submit in his bid. The specification should be clear on
matters of capacity, performance, over-all dimensions, and quality, but should
not unduly restrict freedom of design on the part of the manufacturer unless
added cost resulting from special design can be accepted. Lack of space prevents
printing of sample specifications here; however, they are frequently available
in printed form.*
When completed, power plant design drawings are put to three distinct uses:
(1) Purchases of material are made on the basis of the items indicated on the
drawings; (2) drawings furnish the erector with the information he needs to
build the plant in accordance with the designer’s ideas; and (3) certain of the
drawings will be used as a source of information for the guidance of the
operating staff. The drawings should furnish all the data required for these
services, but little more. The methods of orthographic projection are not neces¬
sarily followed except for large-scale detail drawings.
Layout drawings generally consist of one or more plans and a number of
elevations. The sections represented on the elevations are indicated on the
plan by section lines. The sections do not have to be taken completely along one
plane, but may be offset if such serves the designer’s purpose better.
Supplementing the plans and elevations, details, drawn to a large scale, give
important construction information that cannot be shown on the general draw¬
ings. In addition, there will be wiring diagrams, flow diagrams, etc. When
equipment manufacturers have supplied prints or drawings of their equipment,
such may be incorporated into the design as details by a note of reference to
the manufacturer's drawing by title and number.
Jn the large station design several drawings arc often made covering the
same plairWt each bringing out some special phase of the equipment. For
instance, one may be made primarily to show machine location, another to
* For example, in publications of the old NELA and later the Edison Electric Institute.
DRAWINGS, SPECIFICATIONS, AND CORRESPONDENCE 27
show main steam piping, another conduits, another lighting, and so forth. It
is customary to bring out the primary information by drawing the lines showing
it much heavier than the remainder. Such information is then said to be in
function.
The character of drawings required for a steam power station design may be
classified as follows:
Mechanical Electrical Steel
Machine layout Conduit Floor plans
Steam piping Equipment Stairs
Water piping Wiring Trusses
Boiler settings Lighting Cranes
Coal and ash handling Substation
Draft Control boards Concrete
Turbogenerator Details Floor plans
Condenser and 4iuxiliaries Wall elevations
Water conduits Building Service Foimdations
Cranes and hoists Plumbing Details
Heating
Ventilating
There are, at present, no drafting standards that can be said to have been
widely adopted by firms doing power plant design work. Many of the individual^
firms maintaining a large drafting organization have their own drafting stand¬
ards which they follow on all jobs. Some standards should be adopted at tiie
beginning of any job if such are not already in existence. Most firms take an
interest in completing a set of drawings that is uniform in appearance and neat
in execution. Good lettering will contribute more towards the external ap-
oearance of a drawing than anything else. Without good lettering an excellent
lesign will, offhand, create an unfavorable reaction. To the nontechnical man
(city managers, institutional superintendents, etc.) the appearance of the draw-
ng sometimes means more than the design itself. Most executive engineers,
formerly capable of first-class work themselves, react unfavorably to a poorly
executed drawing even though the technical substance of it is sound. By giving
due regard to the importance of lettering, and by following a standard drafting
procedure throughout the design, the designer can produce a set of drawings
which does not belie the technical skill and care that may have been employed
in its preparation.
Another thing which may reflect credit or discredit on a power station design
is the correspondence. Correspondence plays an important part in the engineer¬
ing of any job. Countless letters must pass between the engineers and the field.'
in order to keep the construction of the character desired, to explain those parts
of the dengn that are not covered by drawings, and also to explmn the drawings.
Correspondence with the client deserves the engineer’s careful attention. Care
must be exercised to get the client’s approval on all major points of design
since the plant will be his to operate. Well-written letters go a long way in
holding the good will of the client. Purchasing letters must be clear and deci¬
sive, leaving no point unsettled. This is essential to avoid any dispute whea, cm
delivery, the equipment is not found to be as expected, and to avdd any mis-
understwdings of price.
28 INTRODUCTION
The engineer’s duty is not only to design and construct, but to leave after
him a complete record of his work, and this record is largely his drawings and
his letters. Hence the circumspect engineer will attend carefully to the form
as well as the engineering of his drafting and correspondence.
1-7 Electric Power Systems. If power could be generated for the same
cost at any point in the country there would be no difficulties arising from
power distribution. The condition that early prevailed would still exist—each
power user would operate his own plant. But, unless use may be made of the
exhaust steam, the small privately owned plant is hardly able to compete with
the central station on an economic basis because of the inherently higher ef¬
ficiency of large generating units and the lower overhead cost of quantity
production. Hence, although large numbers of small plants are in operation at
present, the major portion of installed power capacity is to be found in central
stations. Power generated by industries may or may not be converted into the
electrical form before use, but that generated by central stations is invariably
electrical to permit transmission to distant points.

As Figs. 1-9 and 1-11 show, the central station is but one link of a chain
joining the source of energy and its ultimate user. The system of distribution
extending from the j)ower plant to the consumer is of varying complexity,
depending on the number of consumers and their location relative to the plant.
From the standpoint of economy in power transmission it is desirable to locate
the plant near the geograj)hical center of the load, but there are other factors
to be taken into account Some plant sites are more suitable than others from
the srandj)oint of gen(‘ration, even though many miles may separate the plant
and its customers. Nearness to natural waterfalls and sources of fuel, land cost,
supply of condensing wi^er—these are prominent influences in the location
of a central power jdant.
The distribution system may be separated into two parts, the primary and
secondary systems. The primary distribution system generally consists of a
transmission line carrying the three-phase current from the switchyard of the
plant to a substation located near the load served. The purpose of the substation
is to transform the high voltage necessary for economical long distance trans¬
mission to voltages suitable for lines in residential districts and for the pri¬
maries of the light pole-top distribution transformers. The secondary distnbu-
 ELECTRIC POWER SYSTEMS 29
tion gyatem extends from the substation to the customer’s meter through the
transformers, strategically located with respect to a small group of customers
which each one supplies. Small companies will transmit at such voltages that
the substation may sometimes be omitted. Large systems may require two or
more step-down transformations of voltage.
Power supply systems are owned and controlled by municipalities, institu¬
tions, government agencies, and public utility companies. Public-service power

Fio. 1-10 Typical organization of the engineering-management corporation operating

several subsidiary companies.

and li^t plants, at one time mostly municipally owned, have come more and
more into the hands of public utility companies. A utility company may own
one or more generating stations and a network of transmission lines serving its
territory. The company should be so organised as to weld generating, distribut-
mg, and public relations departments into a smooth working unit. The small
company operating one station and servicing one community easily accom¬
plishes this, but the lar^ company requires a complex organization. One com¬
pany divides its operating responsibility into five departments as followa:
30 INTRODUCTION
1. The system operating department which directs the operation of the
electrical system and has the care and operation of the electrical system, the
load dispatching, the provision of safety about high tension equipment, and
the keeping of electrical records as its responsibility.
2. The steam department which has charge of mechanical equipment and is
expected to have mechanical power ready for the generators at all times.
3. The substation department which maintains and operates all substations.
4. The test departmenty which runs all tests, obtains and analyzes experi¬
mental data, checks station instruments, and is responsible for keeping the
operation of the station at the highest possible efficiency.
5. The office department which handles clerical details, records, and
statistics.
Sales and mergers of public utility companies have resulted in holding or
management companies, which may be state-wide or even national in their
scope. The holding company, as a device for controlling more than one corpora¬
tion, is of comparatively recent origin. The decade prior to the economic de¬
pression of the 1930^8 saw a widespread expansion of financial control of public
utilities occur through the use of the holding company. Organized as a corpora¬
tion, it secured control over its subsidiaries by purchase of a controlling block
of stock. Although, theoretically, 51 per cent of outstanding stock might be re¬
quired, usually far less than this is sufficient for effective control. In view of
the widespread distribution of public utility securities, it is probable that
ownership of 10 per cent of the voting securities is tantamount to control.
Furthermore, through exchange of stock certificates, the holding company
could be created and controlled by promoters having relatively small personal
investment in it.
The management company structure lent itself to promotional pyramiding,
and the public in general derived little or no benefit from that situation. Abuses
of holding companies in taking advantage of corporate structure for law evasion
and obstruction of regulation, as well as a needless pyramiding and centraliza¬
tion of control of a large and essential national service, finally brought about
a public reaction to this form of business. By the Public Utility Holding Com¬
pany Act of 1935 the United States Congress opposed the abuses of holding
companies without administering the ''death penalty.” The Securities Exchange
Commission by that act was assigned duties which included the elimination
of uneconomic company structures, and the supervision of security transactions
and acquisition of securities by holding companies. A uniform system of ac¬
counts was also prescribed. The operations of a holding company are now
limited to a single, integrated (geographically) public utility system. The
present philosophy of governmental regulation of this type of corporate struc¬
ture is that the function of the holding company is the economic and efficient
administration of an integrated system, and not that of an investment company
which minimizes the risks of its investments through diversification of its port¬
folio of securities.
The cost of a central power station is high. Only a large organization can
finance the new construction. Problems of design and construction which are
involved in power plant engineering require the overcoming of difficult tech¬
nical problems and the exercise of a high degree of intelligence. The engineer
 POWER STATION

Fm. 1-11 A public electric supply system. (Courtesy West Penn Power Co.)
32 INTRODUCTION
is liable to feel that these problems constitute his major burden. It happens,
however, that the financing of power plant construction, as in any other field
of building, is at first the paramount consideration. Utility stock, sold to the
public, and particularly to the customer public, is a common means of financ¬
ing new work or reconstruction. The investor requires of power plant engineer¬
ing that it show him how the new construction will give a reasonable rate of
return, as well as overcome successfully the purely technical difficulties. In
other words, the power plant engineer should have breadth of outlook and the
ability to present to the public facts showing that the proposed construction
will return profits upon the money invested as well as showing that it is tech¬
nically possible.
Interconnections and mergers are slowly uniting public electric supply sys¬
tems into a network of distribution. The justification for interconnection can
be presented on a financial basis, but the ability to render intersystem as¬
sistance during local trouble and in that way to prevent interruption of service
to the customer should be the basic reason for interconnection, even though
no definite financial expression may be attached to it. Of more tangible value
is the use of the more efficient plants as base load stations. Off-peak power
may be exchanged on some prearranged basis when the load peaks on one sys¬
tem do not occur simultaneously with those on another. This has the effect of
delaying, for a time, the purchase of new equipment to care for increasing
peaks. Also, it renders economical the installation of additional capacity be¬
yond that justified by the gain in the individual system; that is, a new station
or unit causes temporary surplus capacity and consequent annual charges for
idie equipment which may be reduced by adjacent systems installing their
new equipment alternately. A decrease in the total of emergency standby capac¬
ity is possible when two or more systems operate interconnected.
Interconnection provides some relief from the necessity of splitting up the
plant capacity into a number of small units for the sake of uninterrupted serv¬
ice. No very difficult operating problems have been encountered in intercon¬
nected systems, provided the length of the interconnecting lines was kept to a
reasonable figure. To compensate for line drop in voltage under the condition
of a possible power flow in either direction, tap changing transformers have
been installed. Complete metering stations at the points of interconnection
provide records upon which rates and charges for exchange energy are made.
A superpower system is a vast interconnected system, which has for its
basis the maximum exploitable water power of the country, relying oh excess
flows at one point to counteract low water at another, and havihg i^team plants
suitably located to care for deficiencies in water power.

PROBLEMS
1. A train weighing 14501 is pulled up a 2% grade by 4475 kw. Train resistance,
8710 kg» What is the speed,kmph?
2. A pump is lifting water through5t50i to fill a 570 m* tank. The over-all energy
efficiency is 80%. Calculate the length of time that 7.5 kw applied to the pump will re¬
quire to complete the job.
3. A body is being dragged uniformly along a horizontal surface by a force of 45
kg acting at an angle of 20® to the horizontal. Find the work done in moving the body a
km.
 PROBLEMS 33
4. The difference in tension between the sides of a belt running over a 750m diameter
pulley is 25 kg,Pulley speed 500 rpm. What power in kw is transmitted? Is the direction
of the belt leaving the pulley a factor?
5. A truck “free-wheeling'* at BOkmph is stopped in 90 m by the application of a
uniform retarding force. Weight, 1800 kg« What is the magnitude of the force?
6. What is the power of a steam jet 13mm in diameter moving at 750 m/sec? Steam
condition, 1.4 kg/om< ab dry and saturated.
7. Forty percent of the electrical input to a moto^-driven pump is converted into
a hydraulic jet, If mmindiamcter, for the purpose of washing down ashes. Find the jet
velocity in m/ieo. The motor has a 3-phase, 220-v, 7.5-amp rating. Power factor, 85%.
8. How much kinetic energy is released when a flywheel, as here described, is
slowed from 250 rpm to 235 rpm? Mass of flywheel is 60 hyls ; radius of gyration,53cm.
9. A chain weighing 15 kg/m supports a load of 1500 kg at its lower end. The
upper end of the chain passes around a powered drum60cm in diameter. The weight is
120 m below the drum. How much mechanical energy must be applied to the drum to
raise the weight 75 m?
10. A turbogenerator rotating mass has a moment of inertia of 555 hyls/m*. It is
delivering 2500 kw at 1800 rpm. The load then suddenly increases to 25^ kw, the
developed steam power remaining unchanged. What is the resulting speed in rpm after
10 sec ?
11. Prove that 1 kw » 14.33 kcal/mtn or 860 kcal/hr.
12. A hoist is to raise a 1135 kg mine cage at the rate of 4.6 m/sec Mechanical efficiency
of the hoist is 92%. What is the power in kw required to drive at this speed?
13. How much power is there in the kinetic energy of the atmosphere (density
1.217 kg/m*) at56kmph wind velocity? Consider that section that passes through a3m^
diameter circular area normal to the velocity.
14. The flow of a riveri84.26 m/seosta site where a 23 m hydrostatic head can be
created by the erection of a dam. What is the potential capacity of a hydroelectric
power plant if installed at this site? Consider that the hydraulic efficiency of energy
conversion can be 80%, and that the electric efficiency can be 90%.
15. A locomotive is tested with a dynamometer car coupled between the tender and
the train. Traveling over a level track at a speed of 130 kmpVhe dynamometer registers
10615 kg . What horsepower is the locomotive then delivering? On an extended run the
dynamometer integrator gives 2945 kw hr, and over the same stretch 6215 kg of 27920
J/g coal were consumed. Calculate the thermal efficiency of the locomotive.
16. An electric heater is to heat 11 kg of oil per min from 4.5® C to 65.5® C*Specific
heat of the oil ifi 2,1 J/g pcr®C. How many watts should this heater consume?
17. Mechanical work in the form of a torque of 346 kg m is applied to an electric
generator at a speed of ISOO rpm. The machine loses some of this energy in the form
of a heat loss of 2640 kJ* per min. The remainder of the energy is transformed to the
electrical form. Find the kilowatts of electrical power produced.
18. An electric motor converted' 700 w of electrical injmt into work at 58 kgm per
sec. The speed was 1750 rpm. Find the kgm of driven torque and the motor efficiency.
19. In a certain industrial process 85 kg of a liquid (spec, heat: 3.56 J/g per
) are required to be heated through 36® C in a 42-min period. If an immersed elec¬
tric coil is used to accomplish this, how many watts will it draw while in operation, and
what must be its working resistance in ohm.s? (110-v circuit).
20. In a change-over from steam to electric heating, a unit which had been con¬
densing 4.5 kg dry and saturated steam at 2.1'kg/cm* each 15 min is to be electrically
heated from a 220-v circuit. How many r2-ohm resistors in parallel would be needed to
supply the same heat?
21. What power mkw is represented by a flow of water from a Sem diameter nozzle
at a velocity of 135 m/iec?
*kJ«rkiio jotties*
34 INTRODUCTION
22. A certain electric motor receives 12 kw of electrical power and delivers mechani¬
cal energy at the rate of 1000kgm per sec. What is its efficiency?
23. Water drops perpendicularly 365 m in a certain waterfall. Assuming all available
energy absorbed as heat after striking the lower pool, find rise in temperature of the
water passing the falls.
24. How many kg-m of torque are given to a turbine shaft transmitting 30,000 kw at
1800 rpm? At 3600 rpm?
25. What water horsepower(jiietric) is required to>raisel900 l/minthroughaheightof38m?
26. A power plant serves a factory having two 22 kw motors and ten 3.7kw motors.
Assume the efficiency of motors 80%, of transmission line 95%, of generator 92%. (a)
What should the rated capacity of the generator be if it is assumed that all motors
might be delivering their rated power simultaneously? (b) What should the rated
capacity of the engine be?
27. If 2685200 kl* per hr are given to an engine which can convert 12% of the
heat into work, what is the power in kw developed by the engine?
28. The over-all efficiency of an electric system, coal pile to lights, is 11%. What
fraction of a pound of coal containing 302S0 J/g must be consumed to light a 100-w
lamp 45 min?, min?
29. A plant containing one 600-kw and one 200-kw generating unit is being operated
at rated capacity. The coal consumption at the same time is 860 kg per hr. Heating
value of the coal is 22106 J/g • What fraction of the heat in the coal is converted into
useful energy?
30. Compare the heat content of 270 m* natural gas at a temperature of 16® C and
a pressure of 340 g above atmosphere with 15111 of fuel oil of 446691/g Gas has
35.21 J/cm« at standard conditions of 16®C and 1.03 kg/cm* pressure. Oil weighs 880 kg/
m*.
31. A power dam creates a head of 25 m in a stream whose normal flow is 4200 cfs.
Hydraulic turbines that can convert 92% of the water power into shaft power are in¬
stalled. They are connected to generators whose efficiency is 97.5%. What is the kilowatt
capacity of the plant?
32. A farmer has a small stream on his property which he thinks might supply
(iiough power to light his buildings. The stream is gauged and found to be flowing 0.03
m3/seo. The available fall is 2.i m. A small hvdraulic turbine and generator will have
a combined efficiency of 70%. I’aking the transmission line efficiency as 80%, how many
50-w lamps can be connected to the generator simultaneously?
33. Find, by use of steam tables, the temperature of steam at the following states:
(a) 195®C superheat at 35 kg/cm* ab(b) 85% dry atgdA^i'^^S ab(c) Dry aodsaturated at
7 kg/cm* ga.
34. Find, by use of sceam tables, the temperature of steam at the following states:
(a)3.2l<g/cm2abhavingl40® C superheat, (b) 739 mm Hg vacuum, 95% dry. Barometer
751mmHg. (c) Pressure of 3.2kg/om* ab entropy of 6.7R38 J/g/k.
35. Calculate the enthalpy change when water at 16® C is converted to wet steam of
90% dryness factor at 18 kg/cm* ab.
36. Water at 88®C andi Olkg/om* ab is pressurized andheated to the state of 42rl^8/
cm* abl49®C«(a)Bymeans of the usual approximation, find the increase of enthalpy, (b)
Allow for subcooling of liquid, and find the exact increase of enthalpy.
37. Calculate, for the following cases, the ]ier cent error in assuming water to contain
the heat of saturated liquid, (a) 42 kg/om* ab ]28®C (b) 84kg/cm* abO® F. (c) 200 psia,
43®C-
38. Calculate the enthalpy and volumeof£teamat6.3kg/om*gband98%dryne8Bfactor.-
39. At what dryness factor does steam at.8.8 kg/cm* gahaveameiiitropy;ot4.4488 J/g/lfc
40. A pipe line carries a flow of 69 m*/iniii of steam at28kg/<an* gaiand:ill®C super-
t l/min litres per minute.
^kJ » kilo joulee » lOODJoulci.
PROBLEMS 35
heat. Find the flow in kg per min. Diagram tho interpolation method employed on the
steam tables.
41. A drum of 0.S5 m* volume containn steam U).5 kg/om* ga.95% dry. What is
the weight of wet steam in the drum?
42. Interpolate for the enthalpy of steam at31kg/om*aband4l2’ CDiagram the sys¬
tem of interpolation used.
43. Steam is expanded isentropically from 10.6 kg/cm*»b 96% dry, to l.03kg/cmSab.
Sketch' the process on plane. Label with properties found for the final state.
44. Repeat Prob. 43 except for isothermal process.
45. Repeat Prob. 43 except for completely throttling process.
46. Steam is cooled at constant pressure from 227®C, 2908 J/gi, until its moisture is
5.5%. What are the final enthalpy and entropy? Chart solution, illustrated by labeled
sketch.
47. Wet steam at 7 kg/cm* ah is irreversibly expahded, without work, to 1.03 kg/omt
ab, where its temperature is n6®C What is the initial quality? Chart solution, illust-^
rated by a labeled sketch.
48. What is the highest pressure steam can have at 482®C not to have a moisture
content in excess of 15% after a reversible adiabatic expansion to 38mm Hg. Chart solu¬
tion, illustrated by labeled sketch.
49. In a certain throttling type steam calorimeter, the calorimeter pressure is 1.03
kg/om* abj temperature, 107*C. The line steam being sampled is at 7 kg/cm* ga. What
is its dryness factor? Chart solution, illustrated by a labeled sketch.
50. Steam flowing ideally through a nozzle undergoes a constant entropy expansion.
Initial state: e.gkg/cm* ab, 193''C Final state, 2.8 kg/cm* ab. Determine by chart the
change ol enthalpy. Illustrate with sketch showing how the chart was employed.

REFERENCE TEXTS

Heat Engines^ Allen and Bursley, McGraw-Hill.

Steam and Gas Engineeringy Butterfield, Luce, and Jennings, D. Van Nostrand.
Steam Power and Internal Combustion Engines, Craig and Anderson, McGraw-HilL
Applied Thermodynamics, Faires, Macmillan.
Heat Power Fundamentals, Leonard and Maleev, Pitman.
Elements of Applied Energy, Morse, D. Van Nostrand.
Heat Power, Norris and Therkelson, McGraw-Hill.
Elements of Steam and Gas Engineering, J^otter and Calderwood, McGraw-Hill.
Elementary Heat Power, Solberg, Cromer, and Spalding, Wiley and Sons.
 CHAPTER 2

THE VARIABLE LOAD

2-1 Industrial Production and Power Generation Compared. Most

of the perplexities and complexities of modern power plant operation arise from
the inherent variability of the load demanded by the users as they each and
severally require their small or large blocks of energy in accordance with the
demands of their activities. It was pointed out in Chapter 1 that the power
[)hint should be thought of as a factory for the manufacture of useful energy
from fuels or hydraulic energy as the raw niaterial. Modern industrial execu-
tiv(*s have come to the realization that their power plant is an important branch
of their manufacturing process and arc thinking of their power plant, not as a
n(‘C(^s^'ary evil, but as another production department. But one essential and
im|)ortant difference exists between other production departments and the
power plant. The power plant product is not completed until the instant it is
nti (led, and then only in quantities exactly equal to the instantaneous demand,
"riien* is no simj)le way to warehouse an extra supply of kilowatt hours against
some future period of extraordinary demand, yet it is certain that such a period
will occur, for it is rare, indeed, that a power demand is uniform. While there are
some industries using electrochemical decomposition or electroplating, both
chmacterize<l by ratlu'r steady loads, such instances are exceptions to the
g('n(‘ral rule of variable demand. Consequently the accepted ideas of commodity
manufacture must be considerably modified in the case of kilowatt-hour pro¬
duction.
Many advantages are to be gained from the ability to warehouse products in
such a mannc'r as to promote a uniform rate of manufacture. Labor may be
most gainfully employed, efficient routines set up, and fair rate scales de-
fHrmiiHMi. Mle invt'siment capital is reduced to the minimum and plant and
<‘(|uipmcnt can lie us(‘d to the most profitable advantage. Tnterdepf^rtmental
( xchanges may be placed upon a rational basis. Of course, the warehoused
material r(‘preM‘nls capital, b'lnporarily frozen, but the interest costs on it are
slight compar(‘d t<» costs which variable rate of manufacture presents to the
powtT industry.
2-2 Ideal and Realized Load Curves. The ideal load, from the standpoint
of c<|nij)ment ne(Mlc<l and operating routine, would be one of constant magnitude
36
IDEAL AND REALIZED LOAD CURVES 37
and steady duration. Such an ideal load is shown in Fig. 2-la. The cost to pro¬
duce an elementary area of tliis load curve (i.e., one kilowatt hour) could be
from Yj to % of that to produce the same unit under the more frequently
realized condition illustrated in Fig. 2-lb. Hence the problem of variable load
is a vital one, for, from the industrial viewpoint, the cost of manufactured
articles includes an energy charge as an element of no inconsidei’able proportion,

8
6
■■■
■■■ :ss■HI
mn
■ ■P 5iR|i PklPP
■ra mmmmm
4

i&ii ■■■ ■■KS

2
6
12 2 4 6 8 10 12 2 4 6 8 10 12
Noon Noon
Co) Idooi Lood (b) Rtolfzod Lood
Fig. 2-1 Ideal and realized load curves compared. Note: Each represents the same
quantity of energy.

while from the utility viewpoint, the chief concern is to put each kilowatt hour
on the transmission line at as low a production cost as possible. The reason for
this difference in unit output cost is to be found not only in the improved oper¬
ating conditions, such as turbines and generators operating at their best
efficiency, or uniform rate of driving the boilers, but may be even more the
result of lower first cost of equipment, brought about by simplified control and
the elimination of various auxiliaries and
regulating devices. ■ ■ 1H ■ ■ ■
As any power plant operator is well
■ ■ ■H ■ ■ ■
aware, the shape of the actual load curve
■■1 ■■■ IS1
departs far from this ideal. He also knows
■ ■ pj ■ S
III
that its mountainous outline is of tre¬ o
■ 11 R/l n ■
ri II
mendous importance to him as a guide I
£ u ai i ■ H
II II
M n ■ iii i n ■ 1 U
and index to his various duties con¬

5 B 1 i ■ 1 !l ■ ■ K
nected wdth the generation and delivery
of electrical energy to the users. What, II■IB
■■
then, is the fundamental condition which
produces the variable, rather than the Tim#
steady, load? Fig. 2-2 Individual customers* load
The general co7ich(sion is that indus- curves.
tiial processes and domestic uses impose
highly variable demands upon the capacity of the plant. The exceptions do not
disprove this as a basic operating condition of most equipment. Even though
the characteristics of the demand made by any one user will hardly be under¬
stood until his conditions of use arc fully investigated, we might suppose, for
purposes of illustration, that this has liecn accomplished. Then Fig. 2-2 might
represent the domestic demands of two adjacent residences. There is, appar¬
ently, no great similarity between them, nor would one expect a similarity unless
he knew the family life of the two sets of occupants to be similar. Furthermore,
36 THE VARIABLE LOAD PROBLEM
the next reBidence might have a still different form of load curve, and so on for
still other residences. This is all the result of the natural fact of the individual
differences of persons. However, as the number of connected customers increases,
the effect of individual differences is submerged to the general use conditions of
the community, and the resulting load curve is not unlike that of the average
in any other community where similar living conditions obtain. This fact is
illustrated in Fig. 2-3.
This similarity will be apparent only where the number of connected users
is large enough to smooth out the individual factor, and it also requires that no
one individual load be large enough to be considerable in comparison with the
sum of the others. Consequently, generators serving one industry, or just a few
large industries, will meet a demand which can be predicted only after a careful
survey of the industrial processes, this survey including not only demand, but
also diversity factors. The demand upon generators serving a community of

12 2 4 6 8 to 12M 2 4 6 8 10 12 12 2 4 6 8 1012M 2 4 6 8 10 12
July DtCfmbtr

Fig. 2-3 A comparison of the actual average load curves of two towns A and B of
about 4000 population each, situated 45 miles apart, and having similar community
life. Both are served by municipally owned and operated Diesel plants,

residences, or a business district, etc., could be predicted with a fair degree of

approximation by an examination of load curves for similar service.
Unusual conditions of service are clearly reflected in the corresponding load
curves. Sudden thundershowers which darken the skies about the time of maxi¬
mum power demand superimpose a heavy lighting demand on it, producing, in
some metropolitan areas, an extraordinary peak. Holiday crowds may tax the
electrically motivated transportation equipment and reflect their presence in an
unusually large railway load. Daylight saving creates a queer offset of one hour
in the time at which the peak occurs. In many cases daylight saving diminished
the magnitude of the peak by creating a diversity between the times of maxi¬
mum power and maximum lighting demands. Some companies maintain what
is, in effect, a three-dimensional graph of kw vs hr per day vs days per year by
plotting the daily load curves upon cardboard, cutting away the cardboard so
as to leave the load outline, and stacking these in chronological order. The
result is a highly interesting topography of mountains and valleys representing
the load conditions for the year. Fig. 2-4 is a photograph of a public display
model constructed from the profile exhibited by the year’s accumulation of
terms and factors 89
cards, stacked in their filing case. It will be noted that, with the exception of a
few unusual days, probably caused by cloudy weather as just explained, there
are no abrupt changes; that is, the effect of individual differences and of varia¬
tion of daily activities has been completely submerged to the effect of mass
action.
2-3 Terms and Factors. The variable load problem has injected into the
language of power plant men a number of terms which need definition. The
basic information is, of course, the operating data of demanded load plotted
against time sequence. This is commonly referred to as a load curve and usually

Brooklyn Edison Co,

Fig. 2-4 A mountain range produced by variable load.

appears with kilowatts as the power unit and hours as the time unit, the se¬
quence being the 24 hr beginning with midnight. The most important variations
of this curve are the monthly and annual load curves, each of which is the
average of the daily load curves over the period named. The monthly load
curve is sometimes used in establishing rates; the annual load curve for
annual load factor. The load duration curve, useful in financial studies, is ob¬
tained from the same data as the daily load curves for one year (8760 hr).
The ordinates of this curve may extend from 0 to maximum demand in kw or
from 0 to 100% maximum demand. The abscissa reaches from 0 hr (the length
of time of the peak demand) to 8760 hr. The number of hours during which
1000 kwj 2000 kw, 3000 kw, etc., is demanded is recorded from the daily load
curves, then totaled for the year and plotted as the load duration curve. Hence,
the interpretation of point A (see Fig, 2-5), is that 12,000 kw were demanded
for 5266 hr of the year, or that, through 60% of the year, not more than 40%
 OJ 0<P<MCD
JP (MO <D(0
(Mn^lOfDFCKO

Hours of tho Yoor

Fig. 2-5 Load duration curve.

As was mentioned m connection with Fig. 2-1, two load curves may represent
the same kw hr production, yet the unit cost of production for one be more than
for the other. Some information other than the magnitude of kw hr energy pro¬
duced is needed to describe an operating condition. Evidently the relation of
the peak load to the average in some
measure satisfies this requirement.
This relationship is expressed in the
load factor. The daily, monthly, or
annual load factor is the average
load over the time specified divided
by the maximum peak. This peak is
seldom taken as the maximum in¬
stantaneous value but rather as
the maximum 15-min, half-hour, or
hour-long peak.
Ti mt Load factor must not be con¬
Fk;. 2-() T.ond factor fioui nua^mcnientj fused with ‘"power factor,” with
on the load curve. which it has little in common. Power
factor is (‘mj)loycd to describe cer-
tain t(‘chiii(al aspects of a-c circuits aiid has no meaning in variable load
nomenclature.
The effect of the submergence of individual difference and the effect of
grouj) or community difference are also cared for by variable load nomenclature.
terms and factors 41
The manner in which the energy output of a single power plant is subdivided
and sent to thousands of individual customers is diagrammed in Fig. 2-7. First,
the energy is sent to substations which are located at the ends of the primary
distribution system. Feeders, which lead out from the substations, may each
supply several distribution transformers, only one of which (per feeder) is
actually included in the diagram. Each transformer is connected by short low-
voltage lines to the systems of one or more customers.
Now each customer has a “connected load” that is the sum of the continuous
ratings of all the equipment and outlets on the customer's circuits. For example,
the connected load for the distribution transformer shown on Feeder No. 1 is
the sum of the connected loads of customers, a, b, and c. A customer's demand
factor is his actual maximum demand divided by his connected load. As in the

Customers- domestic, industrial, business, etc.

Fig. 2-7 Elements of an electric power system.

case of peak load, the time interval for maximum demand measurement can
be a quarter, half, or full hour. In consideration of the variable, even random
manner of use of electric energy by the hundreds of thousands of customers
located at the ends of the distribution system one would judge that normal
diversification of use would mitigate against simultaneous existence of peak
loads on the components of a system fed from a common source. Thus the peak
load on the distribution transformer would be expected to be less than the sum
of the peak loads of customers a, b, and c, for a^s peak might occur at 11:00 a.m.,
b^B at 11:60 a.m., and c’s at 1:30 p.m.
The diversity factor between customers gives numerical expression to the
ratio of the sum of the individual peaks to their joint peak load. Diversity
factor is the sum of the individual maximum demands of the subdivisions of a
system taken as they may occur during the daily cycle, divided by the maximum
simultaneous demand of the system. The “system” may be a group of customers
served by a certain transformer, a group of transformers served by a feeder,
42 THE VARIABLE LOAD PROBLEM
Table 2-1. TYPICAL DEMAND FACTORS
Type of CoMumer Fador
fikw. 1.00
Readence Lighting U kw. 0.60
l^Over 1 kw. 0.60
^ Restaurants .. .. 0.70
Stores and offices. 0.70
Theaters. 0.60
Commercial Lighting
Small industry... 0.60
Schools, churches 0.65
^ Hotels. 0.50
7.5 kw- 0.75
General Power Service
7.5-15 kw . . 0.65
15 kw-75 kw 0.55
.. Over 75 kw 0.50
Table 2-2. TYPICAL DIVERSITY FACTORS

Residence Commercial General

Lighting Lighting Power Service
Between consumers 3-4 1.5 1.5
Between transformers 1.3 1.3 1.3
Between feeders 1.2 •1.2 1.2
Between substations 1.1 • 1.1 1.1

etc. Since individual variations have diminishing' effect as one goes farther
from the ultimate consumer in making measurements, one should expect de¬
creasing numerical values of diversity factor as the power plant end of the
system is approaches. Typical diversity factors between the different elements
of the system shown in Fig. 2-7 are set forth in Table 2-2.
Load and demand factors are always less than unity; diversity factors,
more than unity. High load and diversity factors are desirable qualities.
By way of explanation of the use of these statistical factors in predicting
load, the maximum individual demands'of customers a, h, and c are the products
of the connected load of each and an appropriate demand factor. The maximum
demand on the transformer is the sum of a, b, and c’s peak loads divided by
the diversity factor between consumers. Similarly, the maximum demand on
Feeder No. 1 is the sum of the peak loads bn the transformers connected to it
divided by the diversity factor between transformers. Likewise diversification
between feeders is recognized when obtaining substation peaks; and substation
diversification when predicting maximum load on the power plant.
Another term which is useful in variable load studies is the '^plant capacity
factor.*^* Since the load and diversity factors are not involved with the reserve
capacity of the plant or system, a factor is needed which will measure the re¬
serve, likewise the degree of utilization of the installed equipment. This need is
met by the plant capacity factor, which is defined as the actual energy produc¬
tion divided by the maximum possible energy that might have been produced
during the same period. Thus the annual capacity factor would be the annual
kilowatt hours produced divided by the kilowatts of plant capacity times hours
TEJIMS and factors 43
of the year. The difference between load and capacity factors is an indication
of reserve capacity. A modification of capacity factor is obtained by using only
the actual number of hours the plant was in operation. This is the ^'plant use
factor,” defined as follows. Plant Use Factor is the annual kilowatt hours pro¬
duction divided by the kilowatts of capacity times number of hours plant was
in operation.
Example 1: A daily load curve which exhibited a 15-min peak of 6520 kw is
drawn to scales of 1 cm 2 hr and 1 cm 500 kwThe area under the curve is measured
by planimeter and found to be 48.72 cm*'. The load factor based on the 15>min peak
will be found.

Each square cm represents 500 X 2 — 1000 kw hr.

Average load « 48.72 x 1000/24 = 2030 kw.
Load factor - 2030/6520 = 0.312.
Example 2: At the end of a power distribution system a certain feeder supplies
four distribution transformers, each one supplying a group of customers whose con¬
nected loads are listed as follows:

Transformer 1 Transformer 2 Transformer 3 Transformer 4

General Power
Service and
Commercial Residence Store Lighting Residence
Lighting Lighting and Power Lighting
a: 7,5 kw, 5 kw e: 5 kw j: 10 kw, 3.7 kw m: 15 kw
h: 5.5 kw, 4 kw /; 4 kw k: 8 kw, 18.5 kw n: 5 kw
c: 11 kw g: 8 kw 1: 4 kw o: 2 kw
d.'3.7kw, 2 kw h: 15 kw p: 5 kw
i: 20 kw
Using data from Tables 2-1 and 2-2, the maximum demand on the feeder will be pre¬
dicted. The individual maximum demands of the group of customers connected to
transformer 1 are obtained with factors from Table 2-1, as follows:
a: (7 J /0.72*) X 0.65 + 5 X 0.60 .9.77 kw
b: ( 5L} /0.72) X 0.75 + 4 X 0.60 8.27 kw
c; (11/0.72) X 0.66 9.93 kw
* (3.7 /0.72) X 0.76 + 2 X 0.60 5.05 kw
Total 33.02 kw
The diversity factor between consumers for this type of service is (from Table 2-2)
1.5. Hence peak load for transformer 1 is 33.02/1.5 = 22.02 kw.
In a similar manner the other transformer loads are determined to be:
ToUd SimiUtaneoiM
Transformer 2.26 kw 7.43 kw
Transformer 3 . 33.38 kw 22.35 kw
Transformer 4.13.50 kw 3.86 kw
The diversity factor between transformers is 1.3.

Maximum feeder load - 22.02 + 7.4^+ 22.35 +

* Motor efficiency assumed uniformly 72% for simplification.

44 THE VARIABLE LOAD PROBLEM
Example 3; The load shown by the duration curve of Fig. 2-5 is to be carried by a
base-load station having a capacity of 18,000 kw, and a standby station having a
capacity of 20,000 kw. The annual load factors, use factors, and capacity factors of
these two power stations will be evaluated.
The shaded area of the figure represents the annual output of the standby station.
The remaining area under the duration curve is the base-load station output. These
areas have to be measured and interpreted in terms of kw hr. This being done, it is
found that
Annual standby station output = 7,350,000 kw hr
Annual base-load station output = 101,350,000 kw hi‘
Peak load on the standby station is shown to be 12,000 kw. Hours that the standby
station was in use during the year are measured from the figure and found to be 2190
hr.
Standby station factors:
7,350,000
Load factor = = 0.279
12,000 X 2190
7,350,000
Use factor = = 0.167
20,000 X 2190

Capacity factor =- = 0.043

20,000 X 8760
Base-load station factors:
^ , 101,350,000
Load factor = = 0.642

Since this base-load station has no reserves above peak load, and is in continuous use,
its Use and Capacity Factors are also 0.642.

2-4 Effect of Variable Load on Power Plant Design. The necessity of

having to cope with a variable load in large measure influences the character¬
istics and method of use of power plant equipment. By way of illustration,
consider a steam central station. The essential elements involved in variable
power adjustments arc diagrammed in Fig. 2-8.
Air, fuel, and water arc the raw materials for otepm power. To produce
power variably it follows that the supply of these materials will be required to
be correspondingly varied. The sequence of control may be followed by refer¬
ence to the diagram. Assume that the power required at the customer's motor
has just been increased. Obviously the increased torque at the motor pulley
must be soon followed by an increased flow of fuel, air, and water to the steam
generator. However, the response is not instantaneous; time lags arc introduced
by thermal storage in the steam and water contained in the boiler and in the
coal bed and hot walls of the furnace. Although the governing response is
prompt from load to turbine, steam generator responses are considerably
slower.
The first event which occurs after an increased driven torque is put on the
motor pulley is a slowing down of the rotor. This slowing down is the means
by which the common induction motor is loaded. The effect is to increase the
frequency of the rotor circuit and induce therein more current. More current is
effect of variable load on power plant design 45
then drawn from the supply line by the stator winding. Hence, the slight decel¬
eration of the motor rotor under increased load has been the means of trans¬
ferring that increased load to the supply mains, and tiience through transmission
line, substation, station buses, and generator leads to the generator itself.
This increased current flow in the generator stator winding is immediately
felt as a greater electromagnetic drag on the generator rotor. The added torque
or braking action on the revolving portion of the turbogenerator in turn reduces
its speed momentarily.
The lower rotational speed is insufficient to hold the weights of the spring-
loaded, centrifugally operated governor in the position they had before the load
increase, and so, under the influence of the spring, they move to a new position.
Their movement is mechanically or otherwise converted into a steam valve
movement, admitting more steam and increasing the turbine speed to its normal

Fig. 2-8 Diagram of element.s responsive to variable load.

value. This governing action may be direct, but is often accomplished through
the intermediary of a pilot valve and motor in the case of a unit where con¬
siderable force is needed to move the governor valve, but where a lightweight,
sensitive governor is desired.
The governing response through to this point has followed rapidly the
original change of load. Beyond this point changes are not so rapid, for the
steam generator must operate with unbalance between heat transfer and steam
demand long enough to suffer a slight but definite decrease of steam pressure.
This is because steam pressure variation is the primary signal used by most
automatic combustion control systems. The controller, be it automatic or
human, must then increase fuel, air, and water flow in the proper amount, thus
affecting the operation of practically every piece of auxiliary equipment in the
plant.
46 THE VARIABLE LOAD PROBLEM
In the modem power plant there is much equipment devoted entirely to
adjusting the rates of supply of air, fuel, and water to the boilers in accordance
with tlic demands made upon them for steam. Efficient combustion is readily
obtained under steady steaming conditions. With fluctuating steam demand, it
becomes very difficult to secure good combustion and steady steam pressure,
because efficient combustion requires the coordination of so many various
services. Although it is not expected that the steam pressure will always be kept
to within as close regulation limits as, for example, turbine speed, yet a severe
pressure fluctuation results in lowered efficiency of both boiler and turbine. The
coordination just mentioned is not as simple as the supplying of more air and
fuel when more steam is required, the reason being that there is a certain time
lag element present in combustion that is not present in electrical generators.
Although this picture of control for variable load is necessarily elementary,
enough has been stated to indicate the important effect of load variation on
plant design.

Lood
Fig, 2-9 Characteristic efficiency-load curves.

The operating characteristics built into equipment are modified by variable

load requirements. The turbogenerator might be cited as an example. Un¬
doubtedly, turbogenerators could be constructed to operate at greater efficiencies
than they do if it could be certain that they would actually operate at the
designed load points. But, as we know, the ordinary load is anything but
steady; hence a flat-topped efficiency curve is more desirable than a peaked one,
even though it be obtained at the sacriflee of some of the economy possible
under ideal conditions. Turbogenerator efficiency curves usually take the form
of a curve, convex upwards, rather flat from %to% rated load. Fig. 2-9 shows
typical thermal efficiency curves for different types of power-generating units.
The range of load through which the units operate will depend, in a large
measure, upon the number and the capacity of the units installed. If the num¬
ber and size of units have been selected to fit a known or correctly predicted
load curve, it may be possible to operate them at or near the point of maximum
efficiency. But a close fit of the operating curve can usually be secured only by
subdividing the total plant capacity into several power units of different sizes.
This is partially illustrated in Fig. 2-10. The block outline showing the capacity
being used would more nearly coincide with the load curve if more units of
EFFECT OF VARIABLE LOAD ON POWER PLANT DESIGN 47
smaller size had been selected and the smaller units could be loaded somewhere
near their most efficient operating point. However, it must be remembered that
the investment cost per kilowatt of capacity increases as the size of the unit
decreases. Probably duplicate units will not meet the load requirements as well
as units of dissimilar capacities, but there is to be considered the saving in first
cost brought about by duplication of sizes and dimensions of pipes, foundations,
wires, insulators, etc., when identical units are installed. These facts must neces¬
sarily modify the selection of number and size of units of turbogenerator or
boiler capacity. What has been said of the steam plant applies in like manner
to the Diesel and hydraulic plants, although the application of these principles
to the latter needs to be done in conjunction with a study of stream flow and
reservoir area.

12 2 4 6 e 10 12 2 4 e S 10 12 12 2 4 6 8 10 12 2 4 8 0 ID tt
Timt Timf

Fio. 2-10 Selection of number and size of units to lit the load curve.

Example: To illustrate some of the principles underlying selection of niunber and

size of units, assume that the load whose variation is shown in Fig. 2-10 is to be sup¬
plied from a Diesel power plant, the equipment of which is to be selected from a certain
line of engines which the manufacturer has available in 625-kw, 500-kw, 275-kw, and
150-kw sizes.
Of several possible selections which could carry the load those illustrated in Fig.
2-10 are typical. Call these alternates Plant A and Plant B. One 625-kw imit and two
275-kw units are selected to compose Plant A, whereas two 500-kw units and one 150-kw
unit are selected for Plant B. llie operating schedules assumed can be read from the
curves and are as follows:
Plant A Plant B
12K)0 MU>. 1230 MID.... — No. 2 is operating
7:15 A.M. 1:00 A.M. .. .Start No. 3; stop No. 2
12«) P.M. 430 AM. .. .Start No. 2; stop No. 3
1:15 P.M. 4.<X) P.M_ .. .Start No. 1
5.-00 POi. .Start No. 1 8.*00 P.M. .. .Start No. 3
1030 P.M. .Stop No. 1 9:15 P.M. .. .Stop No. 3
1130 P.M. .Stop. No. 2 10:40 P.M_ .. .Stop No. 1
Since capacity /actor measures the ^rc^tage of maximum plant output that the
load cun'o allows the units to produce, it will be calculated for comparative purposes.
Use factor will also be calculated, for it shows to what extent the units approach their
48 THE VARIABLE LOAD PROBLEM
full rapacity during the time they are in service. The area under the load curve repre¬
sents 10,620 kw hr. The average load is, therefore, 10,620/24 = 442 kw. Peak load of
15-min duration is 1050 kw.
The load factor = 442/1050 = 42.0%
On account of the use of several units, the use factor will be higher than the load
factor, but because the number is limited, the capacity factor will be lower.
The kw hr represented by the rectangular outlines are measured* and found to be
14,970 kw hr for Plant A; 14,400 kw hr for Plant B.

Plant A Plant B

Capacity factor 10,620 10,620

= 0.377 0.385
24(625 275 -f 275) 24(500 4* 500 + 150)

10,620 10,620
Use factor - 0.762 = 0.735
13,940 14,470

2-5 Effect of Variable Load on Power Plant Operation. As was pointed

out in the preceding section, the variable load condition imposes certain design
requirements. When the power plant has been built and connected to a distribu¬
tion syvstem, the variable load condition imposes operating problems. Beginning
with the assignment of generating units to service, all the operation activities
on down to the final financial contact with the customer, which is the “rate”
used to compute his bill, variability of the load is influential in shaping operat¬
ing procedure.
The reliability of modern prime movers and other power plant equipment is
incredibly good, “Availability for service^^ of central station units is usually
over 96%, meaning that they were unavailable for use less than 6% of the
time. This desirable quality is helped along by appropriate inspection and main¬
tenance at such times as the plant load allows the equipment to be idle.
Public utility systems commonly remain on the “readiness-to-serve” basis.
This plays a much more important part in the operation of an isolated plant
than in one which is part of a power system network. The industrial plant is
usually an isolated one, but absolute dependability of service is not always
regarded so highly. Furthermore, management can arrange for such plants to
receive notification from production departments before some large motor,
furnace, or other load is connected. Operators of public service plants cannot
expect this consideration: instead, they must keep certain of their reserve
capacity in readincss-to-serve. This is called the “spinning reserve^^ and
represents equipment standing by at normal operating cjonditions of speed,
pressure, etc. Normally the spinning reserve should be at least equal to the
largest unit actively canying load. While ^^floating” on the line this reserve
causes standby expense from steam, oil, and water consumption, wear, super¬
vision, etc. The burden of these expenses falls on the customer public, of course,*
♦ By planimeter or by the sum of individual rectangles.
effect of variable load on power plant operation 49

because that is the kind of service it wants—electrical response without prior

notification.
The variable load placed upon the electric generators of a power plant is
mirrored in the steam demand on the boilers and on various other components.
Because the operating characteristics of such equipment are not exactly linear
with load the reflection is somewhat distorted. The following example will point
out in a graphical way how the characteristic plant curve of steam required at
various loads can be used to produce a boiler load curve.

steam— kg. per hr. Time—hrs.

Fig. 2-11 Graphical solution of a derivative of the plant load curve. In this example
a boiler load curve is constructed by use of .the steam flow versus load characteristic of
the plant.

Example: A plant load curve is given, together with a plant characteristic curve
showing rate of steam consumption vs plant output. These are plotted in Fig. 2-11 side
by side with a common ordinate—^kilowatts. Points are selected on the plant load curve.
Then each is projected vertically and horizontally, the latter defining a steam consump¬
tion which is graphically brought up and erected over the former, so fixing a point on
the boiler load curve. The area under the plant load curve represents the daily produc¬
tion of electrical energy, whereas that under the boiler load curve is the daily steam
consumption. For annual studies a load duration curve replaces the daily load curve.
As electrical power supply systems grew and stations were interconnected,
the individual plant superintendents ceased to have sufficient cognizance of the
entire generation and load characteristics of the system to make the best use
of the equipment under their control. Both for safety and for achieving the
utmost economy in the whole system, a certain amount of operating super-
60 THE VARIABLE LOAD PROBLEM
vision was concentrated in a ‘‘load dispatcher.” The load dispatcher was at first
primarily a load forecaster who advised the superintendents of the power stations
of what to expect. Subsequently, as systems became more complex, the dis¬
patcher’s duties and importance increased. The load dispatcher knows inti¬
mately the thermal and electrical characteristics of the stations and intercon¬
necting transmission lines. He has also, before him, complete meter information
regarding the load on the system. When the system is comprised of one base
load station and one or more peak load stations, the load dispatching is chiefly
distributing the load in excess of base load plant capacity to the best peak load
plants; but when there are a number of plants in the system all of which are
nearly equally efficient, the best distribution of load requires much study and
full Imowledge of system geography and technology.

Virginia Electrtc & Power Co,

Fig. 2-12 A load dispatching center.

To assist the dispatcher in a quick visual examination of the condition of

the electric system, dispatch boards came into use. At first these were large-
scale, single-line drawings of the whole electrical system with colored pins or
plugs to indicate location and condition of switching points. From this begin¬
ning, more elaborate panels with inset strips representing lines, colored lamps
representing the switches, and the like were set up to assist the dispatcher. Such
boards were found to suffer from inflexibility in following system changes, and
recent trends to simpler more flexible arrangements are noted.
To keep the dispatcher fully informed as to load conditions on the system,
telemetering instruments which record loads on generating stations and tie
lines are mounted in plain view of the dispatcher, and be is furnished with a
private telephone system with which to communicate with operators in generat¬
ing plants and substations. This system, which makes use of the power trans¬
mission lines and a high-frequency carrier current, is sometimes called the
“high-line phone.”
methods of meeting the load 61
2-6 Methods of Meeting the Load. The s^pecific methods of meeting the
variable load problem are irrevocably involved with the financial or economic
aspect of energy production, for the peaks add but little to the gross income of
the plant and may actually detract from the net. The methods which have been
either successfully tried or proposed, or are under experimentation at present,
are listed here without any special significance being attached to their order.
Some of them apply to carrying peak loads only, and others tend to equalize
all variations as well as carry peaks. They are:
1. Complete hydro development of a stream.
2. Steam and hydro plants in parallel operation.
3. Operation of plants with thermal efficiencies inferior to the more modem
plants and the use of obsolete equipment.
4. Purchase of energy from neighboring systems.
5. Installation of peak load Diesel engines or gas turbines.
6. Use of storage batteries (electrical accumulators).
7. Use of steam accumulators.
8. Development of high-head hydraulic accumulator sites.
The ideas incorporated in each of these methods will be discussed, using the
same topic numbers as above.
1. The possible hydroelectric development sites along the flow of a stream
are of two types, namely, those suitable for run-of-the-river plants and those
offering natural impoundment basins for storage plants. In general, the run-of-
the-river plant is cheaper than the storage plant of equal capacity, but it
suffers seasonal variation of output more or less proportional to the variation
of stream flow.
If all the run-of-the-river plants were located upstream from the storage
plants, they would be operated continuously on a base load plan, because, were
they idle, their small reservoirs would quickly overflow and water would be
wasted over the crest gates. If, however, they are located between storage
plants, the run of the river, as far as they are concerned, is just what thp
storage plants are passing on to them. Hence, located downstream from a stor¬
age plant, a run-of-the-river plant will produce an increase in output when the
storage plant increases its output. With a development approximating full
stream control, variable demand can be met by holding back this or that
storage plant, letting others take more of the load in accordance with their
respective locations relative to the run-of-the-river plants, and also in accord¬
ance with the state of depiction of the storage reservoirs, the whole being so
coordinated by centralized data and control that a minimum of water is
wasted over spillways or crests.
2. This is only a brief introduction to the theory of parallel operation of
steam and hydro plants. Considerable published material is available on this
topic. The conditions under which the hydro plants are erected are highly indi¬
vidual and the manner in which they are operated in parallel with steam
plants to meet the daily and seasonal load variations will be, in each case, the
logical outgrowth of those system conditions which have shaped themselves into
being with the growth of load. These facts present a combination of circum¬
stances which have developed different principles of paralleling the two types
52 THE VARIABLE LOAD PROBLEM
of plants in almost every system. There are, however, a few jirinciplcs that may
be generally applied.
In many hydro sites, land damage and the cost of dam and hcadvvorks
constitute much of the cost of development so that a peak load plant, in which
the capacity of the plant is much greater than that api)arcntly justified by the
stream flow, presents the best economic utilization of the site. To illustrate this,
we shall suppose a system supplied by a 10(),000-kw station to be requiring
more installed capacity and the predicted i)eak load for the next season is 120,-
000 kw. Allowing 10,000 kw for reserve, a 30,000-kw addition is in order. The
chances are that if a hydro site is available, the cost of land, dam, and head-
works for a 30,000-kw development would render the hydro plant more expen¬
sive than a steam station. If, however, the expensive hydro features just
mentioned were constructed for a 10,000-kw ])lant, with tlie installed turbo¬
generator cai)acity remaining 30,000 kw, a hydro development, instead of steam,
might be financially justified. Yet the hydro plant, by ranidlv drawing down its

Time Time
Low Flow day High Flow doy

Ficj. 2~13 TypuMl diMMon of load on a h>dro-steam system.

relatively small reservoir, couhi produce the required peak capacity and then
refill its depicted reservoir during off-peak hours. There arc many hydro sites
which are not above the level of economic competition with steam plants when
designed for base load but which, when developed as peak load plants, would
serve to carry the system peaks with the minimum of expenditure.
If hydro and steam are on about the same economic level, wdicn water is
plentiful, as in spring and summer, the hydro part of the system is given the
base load to carry, with steam jilants taking the peaks. When w'atcr is low, as
in fall and winter, the steam plants arc pven the base load with the maximum
hydraulic capacity realized by impounding during all but peak hours. Hydro
units are more reliable than steam units and usually give some advance indica¬
tion of loss of pow’cr; hence there need be less spinning reserve. Large hydro¬
electric units arc not able to respond to sudden swings of load, and it is desira¬
ble to have some of the more sensitive steam turbine equipment operating in
jiarallel for the purpose of stabilizing electrical frequency.
3. The lower the capacity factor of a station, the less is the utilization of
the investment. This diminished utilization represents less loss when applied
to a plant that has been in service some years, and consequently is of less
capital value because of depreciation, than if it affected a modem plant. The
METHODS OF MEETING THE LOAD 53

older, less efficient plants, or the older, less efficient units in one plant, keep
investment loss due to low use factor at a minimum. This fact is well known by
power men, and their use of it is reflected in curves showing the life history of
the output of different plants or of different units in the same plant. Operation
under this condition implies that the older units are still considered as normal
operating units. The capacity factor of peak load units is very nearly zero, and
the economic aspect of the use of this equipment resides not in the fact that it
is of a certain efficiency, or that its capacity factor is so and so, but that it is
forestalling the purchase of new equipment upon which low capacity factor
would stand as a distinct financial loss. This gives a direct key to the use of
obsolete equipment. If there is a sharp peak of short duration, it matters little
what the efficiency of the equipment carrying the peak is, as long as it can
eliminate an expenditure of capital for new equipment on which the earnings
would be quite inadequate. In so far as obsolete equipment may have sufficient
capacity to carry short peaks, it is excellent since its first cost will already have
been largely written off in depreciation.
4. One of the advantages of electrical interconnection is the ability to
exchange off-peak power. Extending the idea to the interconnection of two
systems, intersystem exchange would be effective only if the peaks did not
occur simultaneously. As a matter of fact, they will occur simultaneously unless
the geographical extent of the systems is sufficient to cover two sections remote
enough to offer different types of loads. The author believes there is a tendency
to place heavy demand charges on intersystem exchange between independent
companies, but even if this is untrue, it seems best to keep purchased power as
a reserve in case of inadequacy of other methods of carrying peaks.
6. Internal Combustion Power Plants have been advocated as peak load
units on account of the rapid starts possible, i.e., no long warm-up, as with
steam turbines. However, the size and cost of the Diesel engine are strong ad¬
verse factors and, consequently, not many have been used for this purpose.
There are bright prospects, nevertheless, for the gas turbine plant in the peak
load field, for reasons as follows: (1) ability to take a quick start from cold,
thus eliminating long warm-up fuei cost; (2) moderate first cost coupled with
compactness and minimum foundation requirements; and (3) high rotational
speed, which results in low generator cost.
The simple, compact gas turbine unit has quite low thermal efficiency, but
this is not a serious objection for load carrying at extremely low capacity fac¬
tors, as in peak load service. Also, the large quantities of hot exhaust gas that
are a by-product of low efficiency can sometimes be put to use in raising the
capacity of an adjacent steam plant. A public utility which installed a 3500-kw
gas turbine was able to increase its load peak by about 7000 kw. Preheating the
steam plant feedwater with heat transferred from the exhaust gases accounted
for the other 3500 kw gained in capacity. In this unusual case the steam plant's
generators were not fully loaded when the steam boiler, unassisted by feed
preheating, was operating at maximum output. However, many and varied
conditions exist throughout the industry, and many cases probably invite
consideration of the gas turbine.
6. The principal use of storage batteries for peak load is in connection with
traction systems which, for speed control reasons, use direct current. These
54 THE VARIABLE LOAD PROBLEM
traction peaks can be extremely sharp, and there may be many of them during
the day, but the kw hr they represent are not many. Storage battery reserve is
not often practical in an a-c system, for rotary conversion would be required
both during charging and discharging periods. Battery reserve incurs heavy
first cost and maintenance per kw hr of storage. However, they are often used
in the d-c field, sometimes in large capacity.
7. Under certain favoraldc conditions a heat energy “flywheer^ may be
interposed between a steam generator and a variable use of steam. A severely
fluctuating steam demand renders it difficult to fire a steam boiler with uni¬
formly good efficiency. Heat accumulator systems have been devised to smooth
out the fluctuations, the most used being the Ruths system of live steam accu¬
mulation.

Energy storage in the Ruths accumulator is* actually in the form of very hot
water, but it is charged with live steam, and releases live steam during the
discharge period; hence it can be called a steam accumulator. One essential to
its operation is a pressure difference between boiler and steam mains. The
greater this difference, the larger the quantity of energy storage in a given size
of vessel. The most attractive prospects for accumulators are those with sharp
but short-lived steam peaks, i.e., a load curve similar to that shown in Fig. 2-14
There are numerous cases like this in the industrial field, but the accumulator
is not well adapted for service with a central station. However, in textiles, sugar
refining, dyeworks, and many other industries, steam accumulation can render
triple service, viz.: (1) reduce required boiler capacity; (2) permit steadier
more efficient combustion ; and (3) maintain more constant steam pressure in
METHODS OF MEETING THE LOAD 66
an industrial process. There is a small but steady sale of these units to industry.
Some detailed treatment of the accumulator is given here, as it will not be sub¬
sequently considered.
The principle of storage is illustrated in Fig. 2-14. But first, to explain why
the energy is transferred to hot water, consider the energy storage possibilities
in a volume of one cubic metre. If one fills it with dry saturated steam at 10.5 kg/cm* ab
pressure (0.189 m’/kg, 181.2 **0, it will contain an enthalpy of 14.7x 10* J above 0 ^C. How*
ever, a cubic metre of water at 181.2'’C and 10.5 kg/cm* ab (886kg/in*) holds 681x10* J
above 0®C.
Basically, a steam accumulator is an unfired pressure vessel containing
water to which excess steam from the boiler is sent during a charging cycle.
The entering steam condenses in the water, raising the temperature. When the
water being charged reaches saturation temperature at boiler steam pressure,
the accumulator is fully charged. Steam draw-off during a discharge period
lowers the tank pressure, causing a spontaneous evaporation of part of the hot
water content in the en(leavor to maintain thermal equilibrium in the face of
diminishing pressure.
The action of the accumulator system in smoothing out the boiler load is as
follows:
The water in the accumulator will absorb or give up steam, depending upon
the pressure in the steam line leading to it and its saturation temperature. If
the pressure in the steam line rises above that in the accumulator, the accumu¬
lator will absorb steam until the heat added raises the saturation temperature,
and therefore the pressure, to that of the steam line. Normally the water in the
accumulator is at the saturation temperature. If the pressure in the steam line
decreases below that in the accumulator, the accumulator pressure is also
lowered, its water is able to hold less heat of the liquid, and consequently heat
is liberated, forming the steam supplied by the accumulator. The water in the
accumulator has assumed a lower thermodynamic level.
The reducing valve B regulates the amount of steam passed into the low-
pressure main in accordance with the low-pressure steam demands by maintain¬
ing constant pressure on the low-pressure side. The regulating valve A will pass
steam into the accumulator if the boiler pressure tries to rise above standard
because of the closing of valve B. It will also act as a reducing valve, passing
low-pressure steam in the event that the accumulator is unable to supply all
the steam demanded during a peak period and the pressure between A and P
sinks to a predetermined level.
The action of the accumulator in smoothing out the peaks is illustrated at
6. The irregular steam demand is assumed to be that of an industrial process
using considerable quantities of low-pressure steam, while the desired boiler
output is shown as the straight line. If the irregular demand is to be supplied,
and yet have the boilers operating at uniform rate of combustion, the accumula¬
tor must absorb steam during the valleys which occur below the steady boiler
output line and supply steam during the peaks. In the diagram shown it is as¬
sumed that the accumulator is large enough to absorb completely the variations.
The direction of steam flow when charging the accumulator is shown at D, while
at E is the flow diagram for accuinulator discharging.
56 THE VARIABLE LOAD PROBLEM
Example: Assuming that the maximum loop area above or below the average line
of Fig. 2-14 (b) represents 9980 kg of steam, the required size of the accumulator tanks
will be predicted. Boiler pressure, kg/om*ab; process steam pressure, 3.2 kg/cm^ ab*
Let W = weight of water in the tanks when they are fully charged to 10,6 kg/cm*
tkh Then W= 9980 4- weight remaining when fully discharged at8.2kg/em*ab. Assume
dry and saturated steam.
At 10.42 kg/em*ab, h =2779.3 J/g; at 8.8kg/om*ab, h = 2727.4 J/g . The average ig
2753.4 J/g.
Enthalpy discharge per cycle = 2753.4 x 9980 x 10® = 27.48 x 10® J
Now this is provided by the flashing of saturated water. Initially the water had an
enthalpy of 768.51 J/g; finally 568.15 J/g. Hence,
768.51 X 10® X IF - 568.15 x 10®(>F- 9980) =» 27.48 x 10®
fV = 108850 kg
A battery of four tanks, each2.5 min diameter and 7.2 m long, would contain this
mass of water with sufficient space left above the water line to provide enough disen¬
gagement area for the generation of reasonably dry and saturated steam. Although
this may seem to be an excessively large storage volume to have to provide, 9980 kg
of steam might represent approximately 13600 kg per hr maximum rate of steam flow
for a peak demand over the uniform boiler output of about an hour^s duration, or be
even greater for peaks of shorter duration. In effect, the tanks would be a substitute
for 1000 or more boiler metric horsepower.
8. Another interesting proposal is one which, although not absolutely un¬
tried, is, as yet, seldom employed in the United States. The Hydraulic Accumu¬
lator System pumps water into a reservoir with off-peak power in order later
to produce on-peak power. Quoted costs from similar European installations
seem to show that in favorable locations the efficiency of conversion and
storage need not be greater than 50fo to justify the project. As shown in Fig.
2-15, the centrifugal pump, hydraulic turbine, and motor-generator are
mounted on the same shaft. Sometimes horizontal shaft units are connected by
clutches so that one or the other may be disconnected from the motor-generator
when not in use. Another variation of the idea is to have two separate units,
and this, of course, applies to the use of hydraulic accumulators to carry steam
station peaks. One unit is a steam-turbine-driven centrifugal pump, the other
an ordinary hydraulic turbogenerator. Certain topographical features are essen¬
tial to the accumulator site. The hydraulic storage of power is necessarily a
high-head development, low-head equipment and hydraulic losses being too
expensive. Hence a storage site should offer a head better than 23 m with suffi¬
cient reservoir storage capacity to carry the peaks. The nature of the storage
basin also is important. There should be no run-off through the soil. An exten¬
sive shallow basin will cause the over-all efficiency to be lower than for a smaller
deeper basin of equivalent capacity because of both percolation and surface
evaporation. These losses are the same as energy evaporated, for energy had
been expended in placing the water in storage against the hydraulic head. Then,
too, the reservoir must be near enough to the station to minimize line cost and
keep pipe friction losses low. Pumped storage of energy can bring to a steam
plant some of the advantages previously cited for the steam accumulator. It
will furnish a load for the boilers during periods of minimum demand and
promote a more uniform rate of combustion.
METHODS OF MEETING THE LOAD 67
Having examined the causes underlying variable demand^ we perceive them
so deeply rooted in our daily occupations that creating a imiform use of electri¬
cal power is a well-nigh impossible task. Then, turning from the consumer to
the producer, it is interesting to note the various means which power plant
men have taken to cope with and partially offset the evil effects of load varia-

EItv.of
Upper Pool

Generator*
Motor

Fig. 2-15 Hydraulic accumulator.

bility; methods which arc even more interesting when we remember that eveiy
one of them must stand the acid test of an economic study before it can be
adopted.

PROBLEMS
1. A central station is supplying energy to a community through two substations.
One substation feeds four distributing circuits; the other, six. The maximum daily
recorded demands are:
58 THE VARIABLE LOAD PROBLEM
Power Station... 12,000 kw
Substation A. . 6000 kw Substation B. . 9000
Feeder 1. . 1700 Feeder 1. . 620
2. . 1800 2. . 1500
3. . 2800 3. . 1000
4. . 600 4. . 2900
5 . . 2200
6. . 3000
Calculate the diversity factor hotwocn (a) substations, (b) feeders on Substation A, (c)
feeders on Substation B.
2. The daily load curve of a power sy stem is given by the following data.

Time. 12 3 6 8 10 11 12 1 3 4 5 6 7 9 12

Load, kw X lO-**. . . . Oo 00 (K) 150 175 180 175 150 200 250 310 170 140 100 105

Plot the load curve to scales of I cm » 2 hr, I cm = 50,000 kwDetermine the load factor.
What is the daily load factor of the standby equipment that takes all peak load over
200,000 kw?
3. A daily load curve is defined as follows:

Time. 12 2 4 6 8 10 12 2 4 6 8 10 12

Load, kw.140 150 160 180 170 170 170 160 240 365 320 220 160

Plot the load curve to scales of Icm «- 2 hr, 1cm — 50 kw.Determine the load factor and
energy produced.
4. All investigation of the way of transforming daily load curves into annual load
duration curves may be made if one of the daily load curves described in these problems
is considered to be duplicated every other day of the year. This untrue but simplifying
assumption will eliminate need for a mass of load data and the reader will learn the
principle.
5. The daily load curve of a power plant is given by these coordinates.

Time.12 2 4 6 8 10 12 2 4 6 8 10 12

Load, kw X lO-**. ...2 2.5 3 4 o' 6.5 6.5 5 6 8 9 5 3

(a) Find th(3 daily load factor, (b) All loads in excess of 4000 kw are carried by Unit
No. 2, rated at 6000 kw. Find its use factor.
6. The annual jieak load on a 15,000-kw power plant is 10,600 kw. Two substations
are supplied by this plant. Annual energy dispatched through Substation A is 27,500,-
000 kw hr with a peak at 8900 kw, while 16,500,000 are sent through B with a peak at
6650 kw. Neglect line losses. Find (a) diversity factor between substations and (b)
capacity factor of the power plant.
7. A power plant load is represented by an average daily load given by the fol¬
lowing coordinates
PROBLEMS 59
Hour Kw Hour Kw Hour Kw Hour Kw
1 AM 220 7 300 1*PM 500 7 900
2 200 8 410 2 620 8 870
3 190 9 560 3 670 9 850
4 180 10 590 4 760 10 720
5 180 11 610 5 1000 11 600
6 200 12 m 605 6 930 12 380

This is carried by one 1200-kw steam turbogenerating unit which has a steam demand
represented by the equation
kg steam per hr = 997.2 -I- 4.99 kw

Find (a) the load factor, (b) capacity factor, and (c) kg steam used per day.
8. A distributing transformer supphes a group of general power customers having
a connected load of 186 kw. Demand factor = 0.75. If the load factor for the group will
average 45% and energy sells at SVJj cents per kw hr, what will be the monthly (30-
day) income from energy delivered through this transformer? Assume average motor
efficiency 75%.
9. An electrical feeder line serves four distribution transformers which have the
following connected loads.

No, 1 No. 2 No. 3 No. 4

Residence 10 kw Residence 20 kw Commercial 15 kw Power 32 kw
Residence 20 kw Residence 25 kw Commercial 5 kw Residence 24 kw
Residence 25 kw Residence 28 kw Power 4 kw Residence 35 kw
Commercial 35 kw Commercial 15 kw Residence 18 kw
Commercial 25 kw

Using tabulated factors, find peak load on the feeder.

10. A certain distributing transformer has connected load as follows: Residence
No. 1, % kw; No. 2, 2^ kw; No. 3, 500 w; No. 4, 3 kw. Restaurant, 5 kw lighting
and n Jew motors. Store, 15 kw lighting and3.7 kw motors. Efficiency of motors, 70%.
Calculate the maximum expected demand on the transformer.
11. A power plant is said to have had a use factor of 48.5% and a capacity factor
of 42.4%. How many hr did it operate during the year?
12. A central station has annual factors as follows: Load 58.5%, capacity 40.9%,
use 45.2%. The rescrvfe carried over and above the peak load is 8900 kw. Find (a)
installed capacity, (b) annual energy production, (c) hr per year not in service.
13. The daily load curve of a certain region is given by the following data.

A.M... 1:00 2:00 3:00 4:00 5:00 6:00 7:00 8:00 9:00 10:00 11:00 12:00 12:30

kw . ..220 200 190 180 180 200 300 410 560 590 610 605 490

P.M.. .IKX) 1:30 2:00 sm 4:00 5:00 5:30 6:00 7.‘00 8:00 9:00 10:00 11:00 12:00
kw . ..500 590 620 670 760 1000 960 930 900 870 850 720 600 380

It is proposed to carry this load with a new Diesel engine power station. A certain
line of engines is offered in the following sizes, depending on the number of cylinders*.
60 THE VARIABLE LOAD PROBLEM
360 kw, 480 kw, 600 kw, 720 kw, 960 kw. Plot the load curve and fit it with selections
from the engine line mentioned so that there will be a good balance between capacity
factor and number of units installed, (a) Determine the capacity factor of the plant,
(b) Tabulate an ''operating schedule.” (c) Determine the use factor of each unit.
14. The load sho\ni in Fig. 2-11 goes to a peak of 24,000 kw. It is to be carried by
one 10,000-kw imit and two 7500-kw units. Work out a use scheme similar to Fig. 2-10
which will be best from the standpoint of number of paralleling operations and plant
use factor. What is the resulting daily load factor for each unit?
15* A 50,000-kw steam plant delivers an annual output of 238,000,000 ^w hr with a
peak load of 42,860 kw. (a) What is the annual load factor? (b) What is the capacity
factor?
16. The load shown by the data of Prob. 13 is carried by one steam turbogenerator
having the following steam characteristic: Total steam =907.2 + 4.99 kw kg per hr,
where P = kwi Generator efficiency = 96%. Steam costs 67j^ per 500 kg to produce.
What is the cost of the steam supplied to this unit per day of 24 hr? What is the cost
of steam used per kw hr?
17. Given load factor 0.48, installed capacity 35,000 kw, reserve over peak 3000
kw, hours out of service per year 410. Find the capacity and use factors.
18. If the load on a system such as that shown in Fig. 2-7 were composed entirely
of residence lighting, what quantity of customer’s connected load could be cared for
by each kw of generating capacity installed in the power station?
19. The load mentioned in Prob. 2-3 is carried by two 150-kw and one 100-kw
engine-generators. Show how these would be operated to carry the load and compute
the resulting plant use factor.
20. The load duration curve of a system is given by the following data:

Per cent of the year.... 0 10 20 30 40 50 60 70 80 90 100

Load, kw X IQ-*... .36.0 33.0 32.2 29.8 29.6 29.2 28.5 28.0 22.0 12.0 8.0

This load is supplied by a steam plant whose fuel consumption is tabulated as follows:

Load, kw X 10“®. 0 5 10 15 20 25 30 35 40

Coal, lb per hr X 10“® 2.0 10.0 17.5 23.5 27.5 32.0 35.0 39.0 42.0

Assume that you are the purchasing agent and you want to contract in advance for
the year’s supply of coal to be delivered in equal semimonthly deliveries. How many
tonnes would you specify per delivery?
21. The system shown in Fig. 2-7 consists in part of a transformer serving customers
c, /, and g. Estimate the peak load on the transformer.
6. Store building with 6 kw lighting, 25-kw small motor power.
/. Store building with 18 kw lighting, 35-kw small motor power.
p. Office building with 65 kw lighting, 80-kw large motor power.
22. Assume that the maximum feeder loads of the system shown in Fig. 2-7 are (1)
18 kw, (2) 255 kw, (3) 115 kw, (4) 95 kw, (5) 37 kw, (6) 45 kw. Predict the peak
load for the power plant.
PROBLEMS 61
23. A city is supplied by a 15,000-kw plant whose load duration curve is defined
by the following data:

Duration, hr. 0 1000 2000 3000 4000 5000. 6000 7000 8760

Load, kw X 10~^... 114.5 91.6 81.0 74.5 69.4 62.0 52.0 40.8 29.8

The plant has one 15,000-kw unit supplied with steam by one boiler. Auxiliaries are
electrically driven. Draw the coal consumption duration curve from the load duration
curve with intermediate graphical modification from three plant characteristics, viz.
1. Auxiliary power. No load, 250 kw; full load, 600 kw.
2. Steam to turbine. No load, 9070per hr; full load, 90700 kg per hr.
3. Coal to stoker. No load, 907 kg per hr; full load, 9070 kg per hr.'
Assume straight-line variation between no-load and full-load conditions. Determine
annual coal consumption, tons.
24. The load mentioned in Prob. 23 is distributed through two substations, A and
B, A supplies the business and industrial district. Its annual peak load is 3050 kw; its
annual load factor 40%. The diversity factor between A and B is 1.177. Neglect losses
in transmission lines. Find (a) peak load on B; (b) kw hr sent through B.
25. Steam flow meters on an industrial process line recorded maximum flow rates
of 26260 kg per hr for cycles of 15 sec duration. To keep this fluctuation off the boilers
and allow them to steam evenly at an average rate of 3400 kg per hr, a Ruths steam
accumulator was installed, working between 3.2 and 2.5 kg/cm* ga.A water surface area,
of sufficient size to limit rate of steam disengagement to 0.305 m* per see per m*, was
considered necessary to insure dry steam production. Find suitable dimensions for
the accumulator tank (i.e., dia. X length). Note: As actually installed the tank was
L 33 m X 7,32 m long.
26. A horizontal steam accumulator tank isl62oin diameter by 6.1mlong. When fully
charged the water levelis 114 cm above the bottom, and the pressure is 7kg/cm* gi^ During
a test for maximum discharge it was found that the pressure could be dropped to6.33
kg/cm* ga in 10 sec without throwing water into the discharging steam. Calculate the
average rate of steam discharge from the accumulator in kg per hr.
27. The maximum demand upon a steam accumulator (i.e., one of the sectional
areas in Fig. 2-14) is 29500 kg steam.Maximumchargingpressureisf 5kg/om^ab dry and
saturated steam. The delivery pressure, beyond valveB,isgkg/om*ab Assuming that the
accumulator is fully charged just previous to the period of maximum demand, calculate
the kg water the accumulator must contain if this maximum demand is to be only 75%
of the accumulator's steam capacity.
28. A hydraulic storage plant has a unit similar to that shown in Fig. 2-15. The
efficiency of the generator-motor is 96%, of the turbine 80%, of the pump 75%.
Average elevation between upper and lower pools =30 ra. Assume a 2% loss of head in
pipe friction. This imit was installed to carry a daily peak load of 1400 kw hr. There is
a daily evaporation loss of stored water amounting to 907 t. Calculate the over-all effi*
dency of conversion.
29. Find the hydraulic impoundment volume necessary to meet a peak of 15,000
kw hr with pumped storage. Turbogenerator efficiency = 82%, head = 70 m. Neglect
friction and evaporation losses.
30. The boiler load curve shown in Fig. 2-11 is to be smoothed out by operating
an hydraulic accumulator unit in parallel. The accumulator plant will have an average
head of 76 m. Ne^ect pipe friction and evaporation losses. Assume the steam turbine
62 THE VARIABLE LOAD PROBLEM

driving the accumulator pump operates as part of the plant characteristic shown
and that the over-all efficiency of hydraulic storage from steam turbine shaft back to
hydro turbogenerator output is 65%.
(a) What steam plant base load capacity is needed?
(b) What boiler capacity is needed?
(c) What is the capacity of the accumulator turbine?
(d) How much more steam is generated per 24 hr than would have been used with¬
out the accumulator?
 CHAPTER 3

3-1 The Source of Power Plant Income. The revenues which pay for
the cost of producing electrical energy are derived, in the case of the industrial
power plant, in a different manner than that existing in a i)ublic utility. The in¬
dustrial plant is producing hut one of the items which enter into the final cost
of the finished article, w’hcreas the public service plant produces the electrical
energy itself as the finished product, ready for sale. The revenue of the indus¬
trial plant is that portion of the manufacturing returns which cost accounting
indicates as the portion of production costs assignable to the power used. There
are many factories where there is no attempt made to separate the cost of power
from the cost of the other items entering into production. It is in these plants
that high energy costs arc often found to exist, not so much because the im¬
portance of keeping them low is neglected, but because the identity of their
magnitude is lost in the unsegregated general costs. Many industrial plants,
however, do have metering equipment and a cost-accounting routine which
have been planned and selected for the specific purpose of evaluating the pro¬
portion of manufacturing cost that is due to the energy used.
The revenues of the public service plant are derived directly from the cus¬
tomers it serves.* Unlike the industrial plant wdierein the revenues may be
shown on paper only, there is an actual transfer of money from the user to the
producer of the energy. The electrical energy is the article of manufacture, and
the determination of its selling price is made upon a more complicated basis
than in the majority of industrial plants. Income from public electric service is
derived chiefly from domestic and power customers. Because of the more
favorable load characteristics of the latter, the per kw hr costs of serving them
are lower than for domestic service. The common method of collecting the
revenues is by monthly installments, on the basis of bills which are rendered to
the customer, usually at the end of each month. A discount is allowed for
prompt payment and, as nearly all customers will take advantage of it, the
discount is added to the actual cost of supplying the customer and becomes, in
reality, a sort of penalty for delinquent payment.
3-2 Effect of Plant Type on Costs. Included in the customer’s bill must
be an element of charge which will pay for the plant and system investment,

^Sales of electricity to ultimate consumers are divided as follows: industrial, 48J2%;

residential, 262%; commercial, 17.6%; others, 8%. (1960 data.)
62
 64 POWER PLANT ECONOMICS
ODe which will pay for the labor and raw materials used, and one which will
reimburse the investors with the profit which applied capital is exj^cted to
produce. From a business viewpoint, the supply of electric energy is public
service, and the industry is possessed of those characteristics which mark the
so-called ^‘public utility.” These utilities in general must make their particular
product or service available at all times and in accordance with the needs of
the public served. They are characterized by elaborate and expensive distribu¬
tion systems, and by a comparatively large ratio of capital outlay to labor cost.
Readiness to serve at fair and equitable rates is demanded of these industries.
Many utilities are publicly owned and operated, but the majority are private
businesses. The specific characteristics mentioned would act to increase un¬
reasonably the cost of service to the customer were normal competition allowed

AVERAGE ANNUAL CONSUMPTION PER DOMESTIC

CUSTOMER
— AVERAGE RATE PER KILOWATT HOUR

Fig. 3-1 Cost and use of domestic electric service.

to exist. Parallel and competing distribution systems to the same group of cus¬
tomers would be the chief cause of this. Therefore, in these fields the
public, acting through its governing organization, has found it expedient to
create an artificial monopoly to bar needless competitive investment. This is
accomplished by the award of an exclusive right to provide the specific product
or service in a stated region. This is commonly called a franchise, and its privi¬
leges are enforceable at law. In the presence of such a monopoly the customer
must be protected by regulation from monopolistic abuse. The electric service
industry operates under monopoly conditions by reason of the long-term
franchises granted to the individual companies. In nearly all the states govern¬
ment authority, acting through public service (or corporation) commissions,
has assumed a degree of control over privately owned public service corpora¬
tions in the public interest, so that excessive profits will not be made by com¬
panies which are free from the restraint of normal competitive business condi-
RATES 66
tions.* These commissions do not have final authority and sometimes have to
argue their decisions before higher courts.
The municipally owned and operated plant comes under no such restrictions^
it being public property, and the rates of such will not necessarily include a
profit on the investment other than a nominal rate of interest. However, the
rates may, and often do, contain a large charge which goes to defraying the
operating costs of the municipal government in part or in whole. The bulk of the
electrical plant customers will also be taxpayers and since the costs of municipal
government would have to be borne by taxes if plant revenue were insufficient,
it is seen where the justification of tax collection in the form of electric service
bills has had its origin. The best that can be said of it is that it is the cheapest
and most efficient way of tax collection as the delinquent customer may be
threatened with severance of electric service. The argument used in connection
with the collection of highway construction and maintenance funds by gasoline
tax is equally applicable to the inclusion of a tax element in the charge made
for municipal plant electric service. It is that, of all methods of taxation, that
method which will receive least public opposition is the one whereby the pay-
ments are spread out in small installments and attached to the purchase price of
some inexpensive, widely used, commodity.
A large power user will be granted, under this plan, a lower rate, not only
because of his more favorable load conditions, but because the tax function of
the rate must be modified to meet the case of a single large user of electrical
service.
From the foregoing it is seen that an economic comparison of the costs of
electric service is unintelligent unless the taxational features are equivalent.
Equally unintelligent is the comparison of portions, only, of the over-all costs.
For example, one may see Diesel and steam power plants compared on the basis
of fuel consumption only; or the operating costs of hydro and steam power
systems compared without any reference to fixed costs. The publication of
partial comparative data on different types of plants is a device sometimes used
to produce an advantageous reaction towards the equipment or system w^hich
is favored by the data. An inexperienced reader may be placed in a dilemma by
contradictions of sets of cost data and by discrepancies between printed infor¬
mation and actual facts. Only by considering all the elements entering into the
cost of power will a fair, economic comparison be made. The nature of these
elements of cost and their incorporation into electric rates will now be explored.
3-3 Rates. That an industry will install its own power plant if the utility
rates are unfavorable is a competitive aspect always present in the supply of
power by a utility company to an industry. That ever-present threat coupled
with the guardianship of public interests exercised by the regulating commis¬
sions makes it highly important that the rates in effect be a fair apportionment
of costs to one and all. The variable load problems, presented in Chapter 2,
have served to show that the cost of producing a kw hr is not the same for all
users but increases with increasing departure of the customer’s load conditions
from the ideal. These facts make the establishment of suitable rates a task of
some magnitude.
estate Commkaioa Jurisdiction and Regulation of Electric and Gas Utilities. Federal
Power Ckimmission.
66 POWER PLANT ECONOMICS
In general, it may be stated that the rate of charge for electrical energy
should satisfy the following conditions:
Fairness. The rates should distribute the costs where they rightfully belong,
taking full account of the variable conditions of the customer's demand.
Simplicity. The rates should not be so complicated as to cause an opposition
or antagonism from a public which is frequently distrustful of ^‘new fangled”
ideas.
Cost The financial outlay in meters and cost-accounting systems by the
electric company should not be so heavy as to increase costs to more than can
be saved by their installation.
The following elements enter into the cost of electrical energy to the con¬
sumer:
1. Fixed element.
2. Energy ele nent.
3. Customer element.
4. Investors* profit.
The first of these is governed by the extent of plant investment and the
current financial rates. It remains a fixed sum regardless of the amount of
energy sold. The second is directly proportional to the plant output. The cus¬
tomer element will be proportional to the number of customers and nearly inde¬
pendent of both the plant investment and its kw hr production. The profit is
that which sound normal business is expected to make.
3-4 The Fixed Element. The factors which must be taken into account
in arriving at the value of the fixed element of cost are:
1. Capital cost of the power plant.
a. Real estate.
b. Building and equipment.
c. Cost of installation.
d. Engineering fees.
2. Capital cost of primary distribution system.
a. Cost of right of way.
b. Cost of line.
c. Cost of substations.
3. Interest, taxation, and insurance rates.
4. The rate at which capital cost is written off to depreciation and ob¬
solescence. Salvage values.
5. Management cost,
6. The amount of general maintenance which would be required whether or
not the plant were operated.
Principal items* comprising the capital cost of the power plant will include
preliminary costs, real estate, buildings and equipment, engineering fees, and
cost of installation which, in addition to labor cost, will involve transportation
(freight and express), stores and storekeeping, interest during construction, and
overhead.
Tlui investment cost in a power plant varies widely, even for similar types
*8oc; also the more cxiensivo listing of items to bo considered in fixed cost: ‘^Definition
of a Unit of Fixed Capital,” N.E.L.A. Pnhlicalion No. 18, January 1931, and Table XCI
Uydroelcctric Ilnndhook, Cn'.-tfor and Justin, John Wiley and Sons, Inc.
THE FIXED ELEMENT 67
of plants. A Diesel engine type may cost from $130 to $260 per kw of capacity.
The simple, noncondensing steam engine plant may cost as little as $70 per
kw, but can be much more expensive if the more efficient unaflow type engine
is used. A stcam-elcctric power station of modern design will not likely be
obtainable for less than $120 per kw and may be much higher, say up to $150.
Hydroelectric developments show still greater variation, for land costs vary
widely. The upper limit may be $300 and more per kw of capacity. The author
hesitates to quote costs, realizing that these can be altered rapidly when there
arc inflationary trends in the nation’s currency. Cost data tend to obsolesce
far more rapidly than technical information. In recognition of this, costs are
considered in only a comparative way in this book.
The use of published cost information is not safe practice unless full condi¬
tions appertaining to the project arc given. Usually some conditions are radi¬
cally different from those of the estimator’s own problem. Building space
provided for future units, foundation construction difficulties, and the extent
to which duplicate auxiliary equipment is provided are typical of the indi-

Fig. 3-2 Typical subdivisions of investment costs in various types of plants. Com¬
parative unit costs are represented by the areas.

vidual conditions which will vary costs and which might not be mentioned in
published figures. Estimates arc of little value unless based on the estimator’s
first-hand knowledge. When cost estimates arc to be prepared, unless the
project is of minor importance and slight cost, the competent engineer will not
estimate the cost of a detail on which he has had no cost experience without
first consulting one whose experience has included the proposed item. Even after
careful estimating an addition of from 10% to 20% of the total cost should be
made to care for contingencies which may arise from unforeseen complications,
from the sum of many small items, each of which may have been considered
negligible by itself, or from possible errors in estimating.
When the power plant is not situated in proximity to the load served, the
cost of a primary distribution system will be a part of the initial investment.
The primary distribution system should be laid out to meet the expected
capacity of the plant; therefore its first cost will be proportional to plant cost
and not to the number of connected customers.
During the period in which a sinking fund may be accumulated to retire
the investment in a plant, the investment is treated as a loan upon which in-
68 POWER PLANT ECONOMICS
terest will have to be paid at the current rates. The severity of this rate is of
importance to the fixed element. In the case of a plant where a long working
life is expected and where a high rate of interest is being paid, the interest
component may be more than a quarter of the total fixed charges. Taxes will
also have to be paid at the local rate but will not be a seriously large charge
except for stations located on land of high value, as in metropolitan districts.
Of about the same magnitude as taxes will be the insurance upon boilers,
buildings, turbines, which, like the taxes, is proportional to the capacity of
the plant. Taxes and insurance are frequently lumped together in cost estimates
and are put at from 2% to 5% of the cost of building and equipment per
annum.
A certain amount of power plant maintenance is required whether any
salable energy is produced or not. As such, it is proportional to the size of the
building and the amount of equipment in it.
3-5 The Energy Element. The energy element, sometimes called operat¬
ing cost to distinguish it from the fixed costs of the investment element, is based
on the energy output as measured in kw hr. The magnitude of energy cost will
be in direct proportion to the number of kw hr used by the customer; hence
the charge is made as a unit charge per kw hr upon the consumption as re¬
corded on the customer's watthour meter. When all of those expenses which are
incurred in the actual production of energy are summed and divided by the
kw hr used, the quotient is the energy element of unit cost. When using the
plant output in kw hr as the denominator, it should be multiplied by the efii-
ciency of the primary distribution system in order that each customer pay his
part of transmission line losses. The steam and Diesel plants have energy
costs as a large part of the total; however, radically different conditions prevail
in the hydraulic plant where investment cost overshadows the relatively small
energy cost.
The components of the energy cost are:
1. Cost of fuel.
2. Cost of labor.
3. Cost of water for
a. Boiler feed.
b. Condensers.
c. Cooling and house service.
4. Oil, waste, and supplies.
5. Maintenance.
Labor is a small part of the cost of a kw hr. This being so, the increased
labor charge per kw hr is not pronounced when well-paid, skilled labor is sub¬
stituted for poorly paid, unskilled labor. As a matter of fact, the skill in at¬
tendance, as for instance in the firing of boilers, may reduce fuel costs to a
point where the sum of fuel and labor costs is a minimum when the labor cost
is a maximum. For that reason it is urged, on an economic basis purely, that
a large proportion of skilled labor be used in power plant attendance, and that
it be paid well enough to put forth the best effort that its skill sponsors.
At present the labor requirement of the average steam central station larger
than 10,000 kw is about one employee for each 1000 kw installed capacity. A
THE CUSTOMER ELEMENT 69
larger ratio of employees to installed capacity is met in the smaller stations.
Labor costs are truly operating costs, that is, they would cease, in the major
part, were the plant shu^ down for more than just a temporary period; never¬
theless, with the plant in steady operation and enjoying an average use factor,
the labor cost is more proportional to kw of plant capacity than to kw hr of
energy output.
Separated from the energy element in the causes underlying its existence,
but still usually classed with it in rate making, is a variable load clement.
Factors which govern it are:
1. Extent of reserve capacity carried.
2. The operating status of the plant considered as a member of a power
system.
3. Being in '‘readiness to serve.”
4. Amount of starting, stopping, and banking of power units that is neces¬
sary.
The reserve capacity carried is largely dependent upon the interconnections
of the system and upon the policies of management of those responsible for the
system operation. To be "ready to serve” is a state required of the public serv¬
ice plant and of some industrial plants. To remain in this state means to create
certain losses which will add to the variable load element of charge. Lastly,
the cost of starting, stopping, and banking of power units, especially the large
ones, being a fuel, labor, and maintenance expense, is independent of the energy
output of the plant. These factors have been mentioned in detail to show that
they are not true energy costs. Their magnitude is usually small in comparison
with the energy element. They are difficult to separate from energy costs, and
since there is no suitable rate vehicle upon which to load them, as the energy
element is loaded on kw hr, they are included in the energy charge.
3-6 The Customer Element. The third general clement of cost is pro¬
portional to the number of customers. Its components are:
1. Cost of the secondary distribution system.
a. Depreciation, interest, taxes, and insurance, upon the capital cost of
the secondary distribution system.
b. Line and transformer maintenance and inspection.
2. Labor cost of collecting revenue.
a. Meter reading.
b. Office (clerical).
(1) Records.
(2) Billing, collecting, and accounting.
3. Cost of franchise (nominal value), amortized over its life.
4. Publicity.
a. Public relations.
b. Advertising, etc.
The extent and, therefore, the cost of the secondary distribution system are
proportional to the number of customers it serves. It has been found that the
best method of charging for the secondary distribution system is to divide its
cost equally between the different customers unless, of course, one customer
greatly dwarfs the others in the size of the secondary system necessary to serve
70 POWER PLANT ECONOMICS
him. The expense of distributing through the secondary system is the deprecia¬
tion, interest, taxes, and insurance upon the investment and operating costs
such as line maintenance, transformer maintenance, and inspection.
Franchise cost depends more on the time and circumstances surrounding
the granting of the franchise than on the actual value of the franchise itself.
Sometimes a company will have received a franchise for less than its actual
worth to them; occasionally they will have paid dearly for it. On account of
the close control exercised over the utility business by the regulating commis¬
sions, the franchise of today is stripped of the financial importance formerly
attached to it, and it is the practice to allow for it in rate schedules only to the
extent to which the utility has actually paid for franchise rights. Under modern
regulatory conditions, public utility franchises have no value since there is no
special incentive for capital to seek the public utility field because of excep¬
tional profits.
Under the head of cost of publicity could be included public educational
campaigns to bring the public to a greater use of electricity and paid advertis¬
ing designed to create and maintain public good will. As a public utility, it
behooves any company to look well to the nature and character of its advertis¬
ing, for public bodies, both protective and otherwise, have demonstrated their
interest in guarding the public against big business propaganda. One of the
most important executives of the public utility is that officer who has charge
of the advertising, publicity, and public relations activities.
The sum of these customer costs divided by the number of customers is the
customer element. It frequently appears as a fixed, monthly service charge.
3-7 The Investors* Profit. The last element whi^h was mentioned as
being a part of the cost of electrical energy to the consumer was the investors’
profit. As would be expected, this element is also variable owing to variable
business conditions prevailing in different localities at different times. After
interest has been paid and all contingencies, present or expected, cared for,
the surplus may be declared as a dividend. The industrial plant is rarely expected
to show a profit except as its profit on energy used in the manufacturing process
is included in the general manufacturing profit. But the public service plant is
expected by those who have invested funds in its development to produce a
profit. The private industrialist is entitled to all the profit he can make in the
face of business competition. This business competition has been nearly re¬
moved in the case of the public utility by the nature of the franchise it holds;
hence an artificial control, such as regulation by a specially created public
commission, is substituted for the competition of private industry. Forty-one
states have established public service commissions to regulate profits.
Whereas the profits of private competitive industry range upward to 25%,
depending on the financial hazard of the enterprise, the relatively safe invest¬
ments in public utility companies arc restricted to 8% or less by state regula¬
tion. An examination of state commission regulation practices discloses that
the approximate rate of return prescribed in most cases is between 5% and
/ %.
61 2
3-8 Depreciation and Replacement. The largest component of the fixed
charge is the amount to be collected and set aside in a fund which, through the
useful life of the plant, will accumulate a sum equivalent to the money borrowed
depreciation and replacement 71
for investment. This cost is brought about through the reduction in value of
the property, through depreciation, from its initial cost to its salvage value at
the end of its economic usefulness. This depreciation is caused by age of service,
wear and tear on the machinery, by corrosion; or it may be due to obsolescence
of equipment or inadequacy of the equipment, necessitating its replacement
with larger, better adapted units. The depreciation charge, that is, the amount
which must be set aside from the income each year and placed in the deprecia¬
tion reserve, may be based on either the straight line or sinking fund theory.
The former is the more widely used in power plant cost accounting and is
simpler, in application, than the latter. The straight line method assumes a
constant depreciation each year while the sinking fund is based on a variable
depreciation. The straight line method is based on the assumption that de¬
preciation occurs according to a straight line law such as is expressed at a in
Fig. 3-3. The amount of this charge depends only on the total depreciation and
the expected life and is independent of the rate of interest it may draw during

.
UJ
3

AGE AGE
(0l DEPRECIATION BY STRAIGHT (b) DEPRECIATION BY SINKING
LINE METHOD FUND METHOD

Fig. 3-3 Methods of computing depreciation.

accumulation. Such interest moneys, if earned, are to be treated as income. In

the sinking fund method a fixed, predetermined sum is set aside each year and
interest compounded on it periodically. The fixed sum must be such that the
total of annual installments plus the interest accumulations will equal the
total depreciation.
Straight line annual depreciation reserve = —^— (3-1)
where P = The principal sum.
S = The final salvage value.
n = The term in years.
Average rate of interest paid over a period of n years, when r rate of interest
is charged on the unpaid balance is tav.

(3-2)
§(^)
The amount which, when annually placed at r rate of compound interest,
will amount to (P — S) in n years, is
(3-3)
Sinking fund payment = (P — S) ^
<(l -I- r)- t)
72 POWER PLANT ECONOMICS
The parenthetical term in Eq 3-3 is frequently referred to as the ^^sinking
fund factor.” It and useful variations are often entered in Interest Tables
having n and r as independent variables. However, for the few examples in this
book one may prefer to use the formula, as is done in the following example:
Example 1: A 7500-kw steam plant is erected at a cost of $125 per kw. Assume
that bonds in the amount of the total cost were sold. They are to mature in 15 years,
which is also the estimated life of the plant. Salvage value is estimated at 5% of first
cost. Interest on bonds is 4%; on sinking fund deposit The amount of annual
payment on the investment, also the sinking fund accumulation after 5 years, will be
computed.

Sinking fund factor = —r = 0.052

(l.uoo;*® — 1
Sinking fund payment - 0.052(125 X 7500 X 95%) * $46,312.50
Annual interest payment = 0.04(125 X 7500) = $37,500.00
Total = $83,812.50
0.035
Five-year accumulation factor = 0.186
(1.035)‘ - 1
46 312
Accumulated amount = —^ ' = $248,991.93
0.186

Although the function of depreciation is apparently simple in theory, in

reality there are many troublesome angles to it, bearing on its meaning and on
its evaluation. The accumulation of a fund sufficient to retire a given investment
may be the purpose of depreciation reserve; however, the history of power
plants does not show that the investors were returned their money and that the
plant was scrapped at the end of its working life. What actually happens, in
the majority of cases, is the replacement of the depreciated article by a new
one. In that sense, then, depreciation becomes the sum set aside to replace the
worn article. This introduces into the depreciation calculations the changing
value of money, a factor which will become more pronounced, the longer the
useful life of the article. Also, the history of power plant economics has been
that the estimated working life was a possibility, but not a probability, because
the rapid development of more eflScient, more flexible, more reliable, or more
compact equipment by the manufacturers has accentuated the inadequacy of
existing equipment and caused its replacement long before the end of its work¬
ing life. Furthermore, this working life itself may be no more than a conjecture,
as is the case where the units in service have been developed recently enough so
that none of them has reached the point of replacement. It should be remem¬
bered that, in depreciation accounting, the accounting for depreciation is rarely
a representation of the actual depreciation itself.
Replacement. Questions concerning the advisability of replacing component
parts of a power plant may be expected to arise from time to time. As new and
improved forms of equipment become available, owners may question the econ¬
omy of retaining some item of existing equipment long before the conclusion
of its useful life. In such cases application of the principles of engineering
economy dictates that those costs which are unaffect^ by the decision to re¬
place (or not) should be discovered and eliminated from consideration. Thus,
DEPRECIATION AND REPLACEMENT 73
a depreciation allowance or a sinking fund payment on existing equipment
ought not to be considered, since it is an expense already incurred and not
cancellable by any decision to replace. Consequently, the “book value'' of
equipment, as established by a depreciation accounting system, is not con¬
sidered since its magnitude depends on assumed life span and salvage values,
and not at all on current replacement economy.

Table 3-1. LEGALLY APPROVED DEPRECIATION RATES AND THE

USEFUL LIFE THEY REPRESENT

(From Standard Handbook for Electrical Engineers^ McGraw-Hill Book Co., Publishers)

Depreciation
Property (Straight-line) Life, Years
Per Cent

Air compressors. 4 to 5 20 to 25
Boilers, fire tube.. 6.7 to 10 10 to 15
Boilers, water tube. 5 20
Breechings. . 3.5 to 10 10 to 28.5
Buildings. 2 50
Coal and ash machinery. 5 to 10 10 to 20
Condensers. 5 20
Conduits. 2 50
Steam engines. 3 to 7.5 13.3 to 33.3
Gas engines. 6.7 15
Foundations, machinery... same as its same as its
machine machine
Fuel oil handling machinery. 4 25
Generators. 3to5 20 to 33.3
Steam turbogenerators. 5 to 10 10 to 20
Feed water heaters, closed. 3.3 33
Feed water heaters, open. 3.5 28.5
Electric meters . 5 to 8 12.5 to 20
Piping and covering... 1 4 to 6.67 15 to 25
Pumps. 5 to 6.7 15 to 20
Stacks. 3 to 10 10 to 33.3
Stokers, fixed parts. 5 20
Stokers, moving parts. 20 5
Storage batteries. 5 20
Switchboard and wiring.... 3 to 8 12.5 to 33.3
Transformers, station service. 5 to 6.7 15 to 20
Turbines, steam. 5 to 6.7 15 to 20
Turbines, hydraulic. 3.33 30
Wire, weatherproof. 6.25 to 7.5 13.3 to 16
Motors. 5 20

A simple method of analysis of these problems is to compare costs which

are incurred by a decision to replace with those which are prevented by the
same decision. Grant* specifically warns against the following faulty practices
in replacement studies.
* £. L. Grant, Principles of Engineering Economy, Ronald Press,
74 POWER PLANT ECONOMICS
1. Considering the excess of present book value over the net realizable value
of the old asset as an addition to the investment in the new asset.
2. Calculating depreciation and interest on the old asset on original cost
rather than present net realizable value.
3. Assuming that a reduction of direct costs will effect a corresponding
saving in indirect costs
4. Comparing calculated unit costs realizable only with full capacity opera¬
tion, where actual operation fails to reach this capacity.
Example 2: An engine costing $7200 is 5 years old. Its working life and salvage
value have been assumed to be 15 years and $500, respectively. The average operating
cost per year thus far has been $4200. A new engine having a cost of $12,000, an es¬
timated life of 15 years, and a salvage value of $800 is estimated to have an annual
operating expense of $3600. The new engine is considered as a replacement of the
old. It will be determined whether the replacement is advisable if all that can be ob¬
tained by sale of the old engine is $4500. Use straight line depreciation and 5% interest
rate. In this case the depreciation on the $4500 sale value is prevented if the old
engine is sold. The (7200 — 500)/15 == $446.66 depreciation item for the old engine
continues whether or not a replacement is made. It is seen that the costs actually
prevented or incurred are:

Prevented by replacement Incurred by replacement

4500 - 500 ^ 12,000 - 800 ^
Dep.
15-5
0.05/
Av. mt. —(^^~^)x4000« 110 X 11,200 = 299
2 >V. 15 /
Int. on salvage .05 X 500 = 25 .05 X 800 = 40
Operation = 4200 = 3600
. Total $4736 $4686
The difference in favor of replacement is so slight that it is doubtful that a change
should be made.
3-9 Theory of Rates. It has been stated that the electric utility derives
its revenues directly from the customers it serves on the basis of monthly bill¬
ings. The customer's meter readings are put into the rate structure and the
amount due from that customer determined. The apparent simplicity of the
nrocess is misleading, for the establishment of the rate structure that will fulfill
Table 3-2. STATE COMMISSION REGULATION PRACTICES

Rate Base Determination

Depreciation
State Method Treatment of Method
Applied Depreciation

Alabama. Prudent inveslment Deducted St. line

Arizona. Prudent investment St. line generally
Arkansas. Prudent investment Various
California. Reasonable hist, cost Not deducted Sinking fund
Colorado. All elements considered Deducted St. line or ret.
Connecticut. None
, .
THEORY OF RATES 76
Table 3*2 (Continued)

Rate Base Determination

Method Treatment of Depreciation

State Applied Depreciation Method

r)Alaware.
District of Columbia.... Original cost Deducted St. line
Florida.
rronroria . All elements considered
Idaho. Original cost Deducted St. line
TllinoLs.
Indiana.. Deducted
Iowa.
ICanAnn. Fair value Deducted
Kentucky.
Louisiana. Prudent investment Not deducted Judg. and hist.
record
Maine. Retirement
Maryland. Fair value ^23 plus net Inc. over '23
add. ded.
Massachusetts. Prudent investment Deducted St. line generally
Michigan. Original cost Deducted St. line
Minnesota.
Mississippi.
Missouri. Original cost Not deducted St. line
Montana. Fair value Deducted St. line
Nebraska.
Nevada. Prudent investment Deducted St. line generally
New Hampshire. All elements considered Deducted St. line
New Jersey. All elements considered Deducted St. line generally
New Mexico. Original cost adjusted Deducted St. line generally
New York. Original cost Deducted St. line
North Carolina. All elements considered Deducted St. line
North Dakota. Prudent investment Deducted
Ohio. Reproduction cost Deducted Observed
Oklahoma. Original cost Deducted
Oregon. Original cost Deducted St. line generally
Pennsylvania. Fair value
Rhode Island.
South Carolina. Fair value
South Dakota.
Tennessee. Prudent investment Deducted St. line
Texas.
Utah. Prudent investment Deducted
Vermont.
Virginia. Fair value Deducted Comp, interest
Washington. Original cost Deducted St. line generally
West Virginia. All elements considered Deducted Inq>. & CO. exper.
Wisconsin. Fair value Deducted St. line generally
Wyoming.
76 POWER PLANT ECONOMICS
the requirements of a successful working rate is a matter of considerable diffi¬
culty. From the public^s standpoint the rates should meet the following condi¬
tions:
1. Rate schedules should be simple. The problem of setting up a schedule
that will fairly distribute the costs is aggravated by the necessity of its being
comprehensible to the public as well as to the rate expert.
2. Rate schedules should be uniform over large territorial areas. There is
much yet to be accomplished here. Persons in one community frequently are
paying on one basis, and those in the neighboring community on another which
is so different as to be unintelligible to the first.
3. Direct service from producer to consumer. This requires the elimination
of the energy jobber, subcontractor, or middleman.

Sangamo Electric Co.

Fig. 3-4 Domestic watthour meter.

4. Distribution of costs in such a way that persons creating a desirable and

relatively inexpensive type of load may enjoy the full use and benefit of electri¬
cal appliances.
Scientific electric rate-making might be said to have originated with Dr.
John Hopkinson, an Englishman who lived in the last half oi the nineteenth
century. His great contribution to rate-making was embodied in an address
delivered in 1892 before the Junior Engineering Society of London. This ad¬
dress contained a clear analysis of the various elements entering into a fair
electric rate. The Hopkinson rate theory was based on two charges, one a fixed
annual charge per kw of maximum demand, the other a small unit charge
against each kw hr of energy used. Other early leaders in rate making theory
were W. J. Green and Arthur Wright in 1896, and H. L. Doherty in 1900.
THEORY OF RATES 77
During the first half century of public electric service many rate forms
were tried, of which the following received some considerable application.
Straight line Meter Rate. The term ''straight line” indicates that the price charged
per unit is constant, that is, does not vary on account of an increase or decrease in
the number of units. This is the simplest of all rates.
This rate is based on a flat rate per kw hr, and under no circumstances could it be
considered as encouraging the use of electricity, unless the rate were so low that it is
out of line with the usual charges in such rates.
Block Meter Rate. The term "block” indicates that a certain specified price per
unit is charged for all or any part of a block of such units, and reduced prices per unit
are charged for all or any part of succeeding blocks of units, each such reduced price
per unit applying only to a particular block, or portion thereof. This is now the form
for the majority of residential and small commercial customers. Its principal defect is
that it lacks a measure of the customer’s demand.
Example: First 60 kw hr per month or less at 44 per kw hr; next 50 kw hr at
3^ per kw hr; and all remaining use at 1.5^ per kw hr.
Flat Demand Rate. The term "flat demand” applies to a charge for electric service
based upon the customer’s installation of energy consuming devices. This is usually so
much per watt or per kw per month or per year. Sometimes this type of rate is
nominally so much per customer per year or per month, for each of the various classes
of customers. It has had some adoption for large customers of hydroelectric plants,
for in such cases the fixed element overshadows all others.
Block Hopkimon Demand Rate. A demand charge based on a maximum kw per
month, plus a follow-on energy block rate.
Example: $2.40 per month per kw for the first 50 kw of the maximum demand in
the month, $2.00 per month per kw for the excess of the maximum demand over 50
kw, plus an energy charge of 54 per kw hr for the first 1000 kw hr used per month, and
34 for the next 4000 kw hr used per month and further block steps where desired.
With this r^te any size of customer can receive an equitable charge compared to other
similar customers. As a demand meter is required, it is not well suited to the small
residential customer.
Three Charge or **Doherty” Rate. Any of the foregoing types of rates may be
modified by the addition of a customer charge. When such a charge is introduced in
the Hopkinson Demand rate, it becomes a Three Charge rate or "Doherty” rate, which
consists of a customer, or meter, charge, plus a demand charge, plus an energy charge.
Many people consider that, theoretically, this is an ideal type of rate.
The principal objections to it are that the charges are in three classes—customer
charge, demand charge, and energy charge. As it requires two meters, it is better suited
for industrial than for residential customers.
Example: $1.50 per kw of maximum demand per month plus 754 per month cus¬
tomer service charge plus 1^ per kw hr energy charge.
Room Rate Charge. Blank k\v hr per active room for the top rate per kw hr.
Blank kw hr per room for the next block rate per kw hr and following blocks where
it seems desirable.
Example: 10^ per kw hr per month for the first 3 kw hr per counted room; plus
74 per kw hr per month for the next three kw hr per counted room; plus 44 per kw hr
per month for the excess over 6 kw hr per counted room.
This rate is an attempt to introduce a demand charge without the cost of an ex<-
pensive demand meter, llie assumption is that a customer’s maximum demand will be
proportional to the number of dwelling rooms he occupies. Although the individual
case may belie this assumption, it is doubtless approximately correct in the statistics
of an assemblage.
78 POWER PLANT ECONOMICS
Special Features, Since rate limitations are prescribed by commissions and
since investors in electric companies are to have a “fair” return, rates often
contain clauses automatically revising the rate in the event operating costs
vary. There are also special provisions for steady loads, seasonal effects, etc.
Some of these are listed.
1. Higher demand charges in winter.
2. Fuel price adjustment to provide a rate change when fuel prices deviate
from a standard.
3. Special energy rates for electric water heating for controlled off-peak
service.
4. Penalties or bonuses for power factor deviation from an 80% standard.
3-10 Making-up Rate Structures. A rate should be sufficient to obtain
from the customers sufficient income to meet the costs detailed in this chapter
and provide the allowable return on the capital basis, whatever that may be
(see Table 3-2). From the simple case of a straight meter rate where the cost

Sanuamn FJeclrtc Co

Fig. 3-5 Combination watthour and wattmeter with maximum demand feature.

per kw hr is simply the §ium of the four elements mentioned divided by the
kw hr estimated to be registered on customers’ meters the job of rate-making
advances in complexity as attempts are made to place the costs equitably on
the customers, that is to say, to charge in accordance with the real costs
imposed on the company by the specific nature of the customers’ load. Then
the correct rate may be psychologically bad for use on the average domestic
customer. For example, the customer element, represented in a “service charge,”
may come out to be large compared to the energy charge, and the customer
public may feel thereby that it is being gouged for overhead. No company feels
MAKING-UP RATE STRUCTURES 79
that it wants to make a separate charge for profit, though any sane-minded
customer must understand that it is always present. Profit could with reason be
assigned to the elements that invohe capital outlay, i.e., fixed and customer, in
proportion to the investment in each. Again, this is not always done for reasons
of ‘‘appearance’^ or “balance’^ of the form of rate. The following examples are
rate problems much skeletonized and simplified in order to show how the rate
is devised to return the necessary income. An actual problem of this type re¬
quires a mass of detailed information and data to be available.
Example 1: The costs of a certain electric system will be summarized by elements
and translated into a straight meter rate.
A city of 150,000 customers (commercial customers being reduced to equivalent
domestic customers on the basis of about 50 kw hr per month per customer) is served
by a 30,000-kw plant through the medium of a 25 jcm transmission line. Cost of the
plant is $145 per kw; salvage value at the end of an 18-year useful life, 10% of its
first cost. Cost of the primary distribution system is $ 1200per km; salvage value at
the end of a 25-year useful life, 30%. The secondary distribution system has a capital
cost of $3,250,000 and an estimated salvage value of 20% at the end of 15 years. In¬
terest rate, 6%%; taxes plus insurance, 5%. Labor costs for the power plant and
primary distribution system:
30 men at $150 month
16 men at $200 month
8 men at $350 month
Management cost is $68,000 annually; maintenance and repairs, $50,000 annually,
oil, waste, and supplies, $25,000 annually. Cost of franchise and publicity, estimated

: ’5
£
- M01
- iio i
< 1
<A
// / —■ g 1
X
// / // ..-r “ V 5 -1-
1
// / h
1/ //
/
/ //
<
. o _l_
1

1
Llj_u_'''
r
/ /"
_1_ u 1
o
0
1
1
1-
1
•7e
1
i! 1
f t
(A
1
1
1 1 1
(
rf m ■B ts
gH n
im
!n111
—- o
H 20
\.i
1
nBB B
nB B ‘ 1 ' ■ ■1111
_ji- ■—
-i - BassB B
15 10 5
□ 6
►-
0 n n 8700
COAL - THOUSAND kg/h HOURS OF THE YEAR

Fig. 3-6 A graphical solution for annual coal consumption from load duration and
plant characteristic curves.
80 POWER PLANT ECONOMICS
$1 per customer. Collecting revenue, $225,000 annually; operating secondary distribu¬
tion system, $110,500 annually, cost of coal, delivered to the plant, $486 per tonne.
The load duration curve and plant fuel charactenstic are as shown m Fig. 3-6. By
the graphical method, these two curves are combined to give the third, the coal con¬
sumption duration curve The area under this curve represents coal used per year

Fixed element:
Capital cost = cost of plant + cost of transmission line
Capital cost = 145 X 30,000 + 25 x 1200 = $4,380,000
Depreciation = capital cost — salvage value
Depreciation (plant) = (145 X 30,000) (100 — 0 10) = $3,910,000
Depreciation (line) = (25 X 1200) (100 — 0 30) = $21,000
Annual depreciation reserve will be calculated on the straight line basis
Plant depreciation reserve = 3,910,000/18 = $217,000
lone depreciation reserve = 21,000/25 = $840
Annual depreciation reserve $^17,000 + $840 $217,840
Interest, taxes, and insurance (0 065 H- 0 05) X 4,380,000 505,000
Maintenance (10% of 50,000) 5,000
Management . 68,000
Total annual cost for fixed element $795,840
Energy element:
The area under the coal consumption duration curve was found tobe72000tcmiiM
and that under the load duration curve, 95,000,000 kw hr Labor costs per annum
are estimated to be 30 X 150 + 16 X 200 + 8 X 350 = $126,000
Labor $126,000
Fuel cost 72/)D0 X $4 86 350,000
Oil, waste, and supplies 25,000
Maintenance (50,000 — 5000) 45,000
Total energy element charge $546,000

f^ustomer element
Dep. of the secondary distribution
system = (100 - 0 20) X 3,250,000 = $2,600,000
Annual depreciation reserve 2,600,000/15 $173,500
Interest, taxes and insurance (0 065 + 0 05) X 3,250,000 374,000
Operating costs 110,500
Franchise and publicity 150,000 X $1 150,000
Cost of collecting revenue 225,000
Total customer element charge $1,033,000

Investors^ profit.
Assume annual profit on capitalization, over and above interest, to be 8%
Cost of plant, primary, and secondary distnbution systems*
Capitalization = 4,380,000 + 3,250,000 = $7,630,000
Profit element * 0.08 X 7,630,000 = $610,400.
Straight line meter rate:
Summing the various elements of cost:
MAKING-UP RATE STRUCTURES 81
Fixed element. S 795,840
Energy. 546,000
Customer. 1,033,000
Profit. 610,400

Annual production cost. $2,985,240

Assuming 80% of the i)lant output to be registered on the customers’ meters (20%
energy losses in line, transformers, etc.) the 95,000,000 kw hr plant production becomes
95,000,000 X 0.80 = 76,000,000 kw hr delivered to customers.

2 Q8^ non
Rate = y0 Q^ QQQ = 0.0393. Say, a 4c per kw hr rate.

Example 2: The elements of cost, as computed in Example 1, will be set into

the structure of a three-charge rate. For this rate let the profit element be placed with
fixed and customer elements in proportion to the capital investment in each.
Demand charge ~ fixed element + a portion of profit
438
= 795,840 + X 610,400 = $1,146,840
7oo
Assume that the over-all diversity factor is 4.2. Then a peak station demand of
25,000 kw represents a total of 25,000 X 4.2 = 105,000 kw of customer peaks. The
unit demand charge is 1,146,840/105,000 = $10 per kw approximately. Actually the
demand charge must be more than this, for customers are billed monthly, not annually,
and the maximum annual peak is not registered every month. Statistical operating
factors containing this information would be applied to the problem at this point. As¬
sume that such would raise the demand charge to $12 per annum. Then this portion
of the rate would become: $1.00 per kw of maximum demand per month. The energy
element was $546,000 for 76,000,000 kw hr. Hence the unit energy charge is 546,000/76,-
000,000, or $0.0072 per kw hr, say 7Vi mills per kw hr.

Service charge = customer element -|- remainder of profit

= 1,033,000 + X 610,400^ = $1,293,000

Dividing this equally between 150,000 customers, the result is $8.62 per year, or $0,718
per month. Here again the theoretical needs to be increased, this time to allow for such
factors as vacancies, temporary absence of such increase due to customer being dis¬
connected while away on vacation, etc. Assuming that this charge needs to be $0.75
per month, we have the final rate as follows:
$1.00 per kw of maximum demand per month, plus 7% mills per kw hr used, plus a
service charge of 754 per month.
Example 3: The same basic data are here worked into a room rate form. Suppose
it is desired to obtain the fixed element and the customer element in the first three kw
hr per room. (It is assumed that this will be the minimum individual consumption.)
The annual sum to be collected this way is $795,840 -I- $1,033,000 = $1,828,840.
Now, assuming all customers to be domestic (this would not be true, but it is as¬
sumed here for purposes of simplification), and each having an average of five rooms
per home, the rate is found as follows:
Charge per kw hr per month for the first three kw hr per counted room =
1,828,840/(150,000 X 5 X 3 X 12) = $0.0676, say 74.
82 POWER PLANT ECONOMICS
Amount realized from thia would be 160,000 X 6 X 3 X 12 X 0.07 = $1,890,000.
Remaining as a strai^t energy charge $2,985,240 — $1,890,000 = $1,095,240.
Kilowatt-hours remaining after this preliminary consumption =
76,000,000 - 160,000 X 6 X 3 X 12 = 49,000,000.

Energy rate - -= 0.0224, say 2H.

The room rate, then, is 74 per kw hr per month for the first three kw hr per counted
room and 2%^ per kw hr for all energy used in excess of this.

PROBLEMS
1. A power plant cost $375,800 to build. Its life is assumed 20 years; salvage, 15%.
Find the sinking fund payment. Interest compounded annually at 5%.
2. Find the production cost per 1000 kglsteam in a steam plant when the evapora¬
tion rate is 7.2 kg steam per kg coal; initial cost of plant, $150,000; annual operation
cost exclusive of coal, $15,000. Assume life of 20 years; no final value; interest on bor¬
rowed capital, 4%; on sinking fund, 3%. Average steam production is 14 500 kg per hr;
cost of coal, ^.00 per toime,
3. Determine the annual cost of a feedwater softener from the following data: Cost,
$15,000; life, 10 years; salvage value, 0; annual repairs and maintenance, $500; annual
cost of chemicals, $825; labor, $20.00 month; sinking fund depreciation with r = 4%.
4. The bonds issued to build a certain power plant have face value of $2,500,000
and bear interest at 4Vi%. They are to be retired at end of 20 years by an accumulated
sinking fund which will yield 4% compounded annually. Find the annual payment to
the account of capital investment.
5. The fixed element invested in a 100,000-kw power plant is $15,000,000. Find its
average annual cost per kw, based on straight line deprecisftion. Estimated life, 15 yr;
salvage, 20%; r, 5%.
6. An ash disposal system of a steam plant cost $30,000 when new. It is now 4 years
old. The annual maintenance costs for the four years have been $2000, $2250, $2675,
$3000. Interest rate = 6%. A new system is guaranteed to have an equated annual
maintenance and operation cost not exceeding $1500. Its cost is $47,000 installed. Life
of each system, 7 years; salvage value, 5% of first cost. Present sale value of old system
is same as salvage value. Would it be profitable to install the new system?
7. Ten years ago a small steam plant of 2000-kw capacity costing $125 per kw was
erected. The life was estimated at 15 years and the salvage value at 5%. At present,
abandoning the old plant in favor of a new Diesel plant is being considered. A market
has been found for the old engine and boiler equipment at $15,000. The remainder of
the old plant can be utilized in the new and is given a valuation of $8500. Depreciation
has been figured on the straight line basis. What is the difference between the depreci¬
ated book value of the old plant and its sale value? How would the difference be taken
care of in the reconstruction?
.
8 The load duration curve and fuel characteristic for a 45,000 kw plant are given
in Prob. 20, Chapter 2. Other data for the same station are: Initial cost, $135 per kw;
estimated life, 20 years; salvage value, 8% of first cost; interest rate, 5%; taxes and
insurance, 3%%; labor, 45 men; average salary, $175 per month; maintenance and
repairs, $65,000 per annum; oil, waste and supplies, $20,000 per annum; coal, $8.25
per tonne Find production cost per kw hr.
9. The annual costs expected by a utility system in supplying a certain residential
suburb of 45,000 customers are: Fixed element, $345,000; energy element, $180,000;
customer element, $300,000; return on investment, $200,000.
PROBLEMS 83
17,050,000 kw hr will be registered on customers’ meters during the year and their
tpftTiinum demand on the power plant will be of the order of 5500 kw. Diversity factor
from Table 2-2.
(a) Form a straight line meter rate, (b) Form a three charge rate, putting % of
the profit in the energy element, in fixed element, (c) Form a room rate in which
the customer element is a fixed monthly service charge and the fixed element is obtained
in the first 4 kw hr per room. (Assume average home, 6 rooms.) Energy element is
uniformly distributed.
10. A town of 4500 population has 2000 connected customers. The maximum de¬
mand is 350 kw and the annual output 1,950,000 kw hr. The plant is operated by the
municipality and its rate includes an element which goes towards defraying the ex¬
penses of town administration. The rate is expected to produce: $12,500 per annum
for fixed element; $20,000 per annum for energy element; $30,000 per annum for cus¬
tomer element; $^,000 per annum for town expenses element.
Work out a rate of the following form:
(A) 4 per kw hr for the first 25 kw hr per month; (B) 4 per kw hr for all energy
used in excess of 25 kw hr.
1st block to return fixed element, customer element, and $20,000 of town expenses.
11. The annual costs of operating an electric system are:
Coal, $300,000
Franc^se and publicity, $70,000
Station depreciation, $190,000
Station supplies, $30,000
Primary line dep., $190,000
Interest, taxes and insurance: on plant and primary line, $50,000; on secon¬
dary lines, $85,000
Secondary lines dep., $120,000
Plant maintenance: Fixed $1000, variable $2000
Secondary lines maintenance, $50,000
Labor: Generation, $105,000; Distribution $80,000; Accounting $150,000
Superintendence and management, $50,000
Dividends to stockholders, $350,000
Annual station output 70 X 10® kw hr, 10% energy loss in transmission; peak
load, 20,000 kw; diversity, 3.33; 100,000 customers
Compute a three-charge rate in which one-half of the dividends are in the service
charge, the other half in the demand charge.
12. A customer owning an 8-room house has 365 kw hr recorded on his meter during
a certain month. The local electric rate is: 6^ per kw hr for the first 6 kw hr per
counted room, plus 44 per kw hr for the next 8 kw hr per counted room, and 24 per kw
hr for all remaining energy consumed during the month. Compute the electric bill for
the month.
13. Annual costs in a certain power system are:
For fixed costs: Plant, $1,750,000; primary lines, $600,000; secondary lines,
$1,250,000.
For operating cost: Plant, $75,000 indirect and $950,000 direct.
For distribution system, $500,000.
Direct customer expense, $400,000; profit, 8% of fixed cost.
Peak load on plant, 45,000 kw; diversity factor, 4; annual plant output, 1.2 X 10®
kw hr. Assume 50,000 customers and 20% transmission loss.
Find the straight line meter rate.
14. Using data of Prob. 13, construct a Doherty rate, putting profit into the proper
elements, in proportion to the investment.
15. An air preheater installation will cost $12,500. Its life is assumed to be 8 years.
84 POWER PLANT ECONOMICS
Salvage value is nothing. Annual maintenance and repair is estimated to average $150.
Use compound interest at 6% and find the annual cost of the preheater.
16. What initial cost will an annual saving of $675 per year for ten years amortize at
4% interest compounded annually?
17. A 30nihp* condensate pump motor has been burned beyond repair. The plant
superintendent has two replacement alternatives. Manufacturer ^*A” offers to replace
the original (which was an “A” motor) for $510. Manufacturer ‘‘B” offers a cheaper
motor at $400 but can only guarantee 87% efficiency whereas the ‘"A” motor is guaran>
teed for 89%. The installation operates 25% of the time at full load, and 75% of the
time at half load where the two efficiencies become 85% and 84% respectively. Assume
a motor comparison period of 5 years, interest rate 8%, equal maintenance costs.
Electric energy is charged for at the rate of 1%^ per kw hr.
(a) Which motor is the more economical buy? (b) At what energy cost do they be¬
come equal alternatives?
18. Make a comparative analysis of the production cost per kw hr of the two plants
for which data are given. Annual production = 1 X 10® kw hr.
Diesel Plant Steam Plant
Engine and generator .... $53,000 Turbogen. and condenser . $24,000
Swbd. and wiring. $5,600 Boiler and stoker. $20,000
Miscellaneous . $8,000 Swbd. and wiring . $5,600
Building . $10,500 Miscellaneous . $10,000
Labor, per month . $350 Building . $12,000
Fixed charges . Labor, per month. $450
Oil, per Ut6W0.62l^/ml). . 4.i Fixed charges . 12%
Fuel economy . .49 Ib/kw hr Coal, per ton. $3.50
Fuel economy . 1.72 Ib/kw hr
19. Using data of the example in Sec 3-10, form a block meter rate with the fixed
element and onerhalf the customer element collected in the first 12 kw hr per month
per customer. In the second block, reduce the rate to % of the first block for the
collection of the energy element and the remainder of the customer element. Assume
that % of the customers will average full consumption of the second block. Collect the
profit element in the third and final block.
20. A 500-kw electric lighting plant cost $95 per kw installed. Fixed charges, 14% ;
operating costs, 1.3^ per kw hr. The plant averages 150 kw for 5000 hr of the year,
420 kw for 1000 hr, and 20 kw for the remainder. What is the unit cost of production
of electric energy?
21. The load duration of a group of residential customers served by a substation
is given by the following data:

Percent of the year. 0 10 20 30 40 50 60 70 80 90 100

Kilowatts. 95 50 40 30 29 29 28 28 22 12 A

Average efficiency of distribution is 95%. Customer's rate is 8^ per kw hr for the first
25 kw-hr, 54 per kw hr for the next 30 kw hr, 34 per kw hr for the next 50 kw hr,
and 24 per kw hr for all remaining energy. What is the average monthly gross income
from this group of customers if 20 of them average taking 300 kw hr per month, 50
of them take 130 kw hr per month, while the remainder average 70 kw hr per montii
each? How many customers are there in the whole group?
*lfatrio horse power
PROBLEMS 85
22. A customer’s meter reads 29,543 kw hr ou May 1, and 29,598 kw hr on June 1.
Find the amount of his electric bill for May based on the following rates.
(a) 74 per kw hr.
(b) 104 per kw hr for the first 35 kw hr; 54 per kw hr for the next 25 kw hr; 34 per
kw hr for all in excess of 60 kw hr.
23. Assume that a customer’s maximum monthly demand was recorded as 120 kw.
His energy consumption for the same period was 40,500 kw hr. His rate is: $2.40 per
month per kw for the first 50 kw of maximum demand; $2.00 per month per kw for
the excess of maximum demand over 50 kw; plus 54 per kw hr for the first 1000 kw hr
per month; 34 per kw hr for the next 4000 kw hr per month; and 24 per kw hr for all
energy in excess of 5000 kw hr. What is his bill for the month considered?
24. A customer having a 7-room house used 55 kw hr during a certain month. What
is his electric bill for that month if his rate is that given as the example of “Room
rate charge,” Sec 3-9, and there is a 5% discount for payment in 15 days
25. A customer owning an eight-room house has electric service under the following
rate: 94 per kw hr for the first 3 kw hr per room; 54 per kw hr for the next 5 kw hr
per room; 34 per kw hr for all in excess of 8 kw hr per room. His meter readings for
three consecutive months were: May 5—2789 kw hr; June 5—2984 kw hr; July 5—
3154 kw hr. What is the amount of his bill for May 5-June 5? for June 5-July 5? What
is the average cost per kw hr for each of the two i)eriods?
26. The rate for a commercial customer is $6.00 per kw per month for the first
kw of maximum demand, plus $5.00 per kw per month for the next 6 kw of maximum
demand, plus $4.00 per kw per month for all of the maximum demand in excess of 7 kw,
l)lus energy charge as follows; First 100 kw hr at 44 per kw hr. All remaining energy
at 14 per kw hr. What type of rate is this? How much is the customer’s bill in a
month when he registers 15-kw maximum demand and consumes 1850 kw hr?
 CHAPTER 4

THE POWER PLANT BUILDING

4-1 General Design. There are different approaches to the architecture

and structural design of a power plant. The industrialist will usually specify
a purely functional building, one that is but a weatherguard for the equipment
it houses. Sheet metal construction is of this character. The equipment of such
a plant may be just as excellent, as intelligently arranged, and as efficient as in
another plant whose exterior is consciously made architecturally beautiful.
Institutions such as schools, asylunts, or sanatoria will frequently want
the heating or power plant building to be in harmony with the architecture of
the institution. Such buildings may have to cost nearly as much as the equip¬
ment they house.
The central electric station field exhibits a great variation of ideas from the
purely functional to the impressively refined architecture seen in so many of
the modem plants built by public utility companies. Pride of appearance of a
large and prominent project such as a central power station is only natural
on the part of company officials. In approving the expenditure of money for
good-looking buildings the cold facts of economics cannot always govern, for
who can say exactly how much a utilitarian project deserves beautification.
Many personal and psychological factors are incorporated in the returns
upon money spent for beautification. One may, however, state that the expense
of architectural beauty has usually been overestimated. As a matter of fact,
simplicity and ruggedness of structure may contribute more to pleasing ap¬
pearance than a wealth of costly materials and ornaments. The power plant
which generates electricity for public use is a public building and, as such, is
deserving of some architectural embellishment. When the power plant was a
dirty, unsightly stracture it was relegated to the outskirts of the community
and seldom visited by the general public. The majority of the customers hardly
knew where it was and certainly had no desire to invade the grimy locality.
That condition has largely disappeared; power plant interiors are clean, airy,
and attractive; exteriors are imposing and invite inspection.
That there is a return on the money expended in making the power plant
presentable is hardly to be doubted, even though the return is not tangible.
Public good will, which is a major asset of any public utility, can better be
secured and retained by an attractive structure than by all the refinements of
design of the machinery it houses. If it is admitted that dignified beauty has
86
GENERAL DESIGN 87
a place in power plant architecture, the next logical question might be, ''How
can beauty be obtained in a utilitarian building, handicapped by such features
as chimneys and coal piles?*^ Of chimneys we shall speak later. It is difficult to
efface the ugliness of a coal pile and, yet, most plants will require a coal storage
to equalize the variable rates of delivery and usage. With a fairly constant coal
supply, large storage bins or coal silos may eliminate the outdoor coal storage.
A plant located in a city could be served directly by railroad cars and still
maintain, in the outskirts of the city, a coal storage yard of sufficient capacity
to insure against shut-down from nondelivery of coal.

Fig. 4-1 Heavy massive lines suggest the architectural motif of power and permanence.

Another school of thought in the public electric service field maintains

that the governing criterion should be the achievement of minimum investment
consistent with reliability in order that the final cost of electric energy delivered
to the customer can be the least possible. It is interesting to note that when
working towards this objective engineers economize by omitting parts of the
conventional building stmeture but continue to give the remaining structure
an architectural treatment in excess of purely functional requirements. The
semi^outdoor plant in which some of the equipment is left outside the building
proper will be discussed later.
Overfall Size. The obvious method to use for securing a building of satis*
88 THE POWER PLANT BUILDING
factory dimensions is to lay out, on paper, the best position of the necessary
equipment and then design the covering structure. The pursuance of this plan
will be more successful if the building is to be single-storied than if there are a
number of floors and mezzanine floors. This method is logical for a new plant
which is to be built upon a lot of ample proportions, for it undoubtedly leads to
a neat and systematic arrangement of the machinery and other building equip¬
ment. On the other hand, the designer occasionally must fit the equipment to a
predetermined lot size by carrying the building vertically until sufficient volume
is provided.

West. Fenn. Fewer Co.

Fig. 4-2 Central power station of the steam turbine type.

The dimensions of the power plant building can be allotted after a plan of
the site is made and the dimensions of the principal pieces of equipment secured.
For this purpose, manufacturers issue catalogues and supply prints of their
respective products showing dimensions that may be used for preliminary
estimating. These preliminary prints are seldom drawn with sufficient accuracy
to be used for final construction drawings. The designers make several studies
of possible schemes of equipment location, since it is improbable that first ideas
would be the best. In all layouts, allowances must be made for sufficient clear¬
ances and for walkways. Generator and motor rotors, boiler, heater, and con¬
denser tubes are examples of parts for which clearance must be allowed in the
layout. Walkway clearances around hot objects and rapidly moving machinery
should be wider than those just necessary to allow passage. Similarly, galleries
in the neighborhood of high tension buses should be as ample as available
space permits. Throughout the design, probable future extensions need to be
MATERIALS 89
kept in mind, and the equipment should be placed so as to allow for the addi¬
tions with a minimum of change or relocation and with no interruption of
service. After the equipment has been arranged, the whole should be restudied
from the standpoint of operating routine. Convenience to operators, short hauls
of supplies, and safety provisions are typical items deserving attention.
Architecture. If the power plant building is not to be purely functional,
then its exterior should receive a treatment appropriate to the field. The archie
tectural motif of a power plant should suggest, in its heavy massive lines, power
and permanence. Each wall should receive a symmetrical treatment in window
opening and ornamentation as far as it is possible to do so. Symmetry and
proportion are far more effective than expensive marbles in lending the proper
dignity to the structure. Simple contrasting stone trimmings can be used very
effectively to relieve the monotony of brick walls. Even the chimney has come
in for architectural treatment, as is seen in Fig. 4-8. A chimney placed to one
side of the plant k less conspicuous than one mounted on the roof. However,
many plants have a steel stack over the boiler room. Mechanical draft mak^
it possible to employ very short stacks unless the products of combustion must
be delivered at an elevation to free the vicinity of soot and gases.
Grounds adjacent to the hydroelectric plant lend themselves nicely to land¬
scaping, with an expanse of water the main theme. Unfortunately, most of the
hydroelectric plants are located far from centers of population and their natural
beauty is seen by relatively few persons. The surroundings of the steam plant
may also be beautified with lawns and shrubs. Where this has been carried out,
the former ugliness associated with power plants is largely forgotten. However,
many plants are located in metropolitan districts with no possibility of land¬
scaping.
Within comparatively recent years considerable thought has been expended
in the beautification of substations. The substation is often located in a resi¬
dential district where an unsightly building would cause a decrease in real
estate values and arouse the opposition of property owners. As a result, the
tendency is towards pleasing substation design.
4-2 Materials. Brick, cinderblock, hollow tile, concrete, and steel are the
principal building materials. Wood is unsuitable for power plant construction,
although it is occasionally used in the smaller plants for roof trusses. Wooden
construction presents a fire hazard, it is relatively short-lived, and its strength
as a building material is not great. Sheet metal is incombustible, if not fire¬
proof, but it is unsightly in sidewall construction. As a roofing material, it is
too good a conductor of heat. However, it is occasionally used because of its
low cost. Although clay bricks may be had in various shapes and sizes, the
standard brick measures 210 X 100 X 55 mm. The better class of brick, called
facing brick, forms a wall that is quite impervious to all weather if well laid
in good mortar. Considerable attention has been given to the ] possibility of
achieving various architectural effects by the use of different colors of brick¬
work in wall and trim.
Brickwork laid to a structural steel frame is a modem type of building con¬
struction. The inclosing walls between adjacent horizontal girders are called
curtain and panel walls, and are much thinner than for the bearing wall type
of construction where the bottom walls support not only their own weight, but,
90 THE POWER PLANT BUILDING
in addition, the weight of the superimposed walls. Masonry blocks are used for
walls (interior;, floors, and ceilings. One cubicaMta* of brick masonry will re¬
quire approximately 650 standard brick and fromiso to ^ Ute« of mortar,
depending on the thickness and style of the joints. Proportions for the mortar
are from one to three parts of dry sand to one part of Portland cement, depen¬
ding on the strength needed. For preliminaiy estimating purpo8es,acubic metre of
mortar may be taken as made from acubicmetwofsandand 13 sacks of cement.
The addition of a small amount of lime to cement mortar will not materially
impair its strength and will render it more easily worked. In the absence of
union rates, a bricklayer with helper will lay from UXX) to 2000 brick daily, the

Table 4-1. DATA ON CONCRETE MIXES TO YIELD 1 CU M CONCRETE

1
Safe
Mixture Cement, Sand, Application Wei^t, Comp.
sacks t/at» StreBB,t/iikt

HB
1:2:3 9.2 Roofs, sills, tanks, 266 882.78
tunnels

1:2:4

laiA
7.8

7.3
B
0.52
0A8

0A3
R. C. floors, beams,
and columns

Building walls
2.66

2.66
32840

273.42

218.73
1:3:5 6.2 0.52
B Foundations and
footings
2.66

1:2:4 8.6 B Cinders,

CU yd
0.96 R. C. floors 1.99 15311

Slag,
ouyd
1:2:4 8.6 0.49 0.96 R. C. floors 2.13 15311
1

Table 4-2. APPROXIMATE WEIGHTS OF BUILDING MATERIALS

kg/m^
Plain concrete, stone. 2323-2403
Plain concrete, slag. 2082
Plain concrete, cinder. 1002
Reinforced concrete, stone. 2403
Brick masonry, jMened. 2243
Brick masonry, common. 1922
Adilar masonry, granite. 2643
Ashlsr masonry, limestone. 2563
Ashlar masonry, sandstone. 2243
&nd and gravel. 1002-1762
Cinder fill. 641-721
REFERENCE LINES 91
former figure being for facing work and the latter for maarive work such as
foimdations or buttresses. Unions tend to regulate to much lower rate of cmi-
struction. Brick masonry varies in weight from 1.9toiuiec pern^ to2.5toimea
per m*. Facing bricks, have the greatest density. The building code committee
of the Bureau of Standards reconunends a maximiun stress in brick masonry
ofl2bg/cm*for bricks set in Portland cement mortar, but this may be increased
to 175 kg/em*iftheeffectsof eceentricloadingandlateralforcesare fully analysed.
Standard wall thicknesses are 100iimit210inm,322nBm,432nun and 544 nim
Theminimum thickness of bearing walls is 322 mm
Con<uete is extennvely employed in power plant construction, perhaps not
so much for wails as for foundations and fioors. For single-story walls, it is
poured .as a monolith both with and without reinforcement. For a building of
more than one story, steel framing is better ctmstruction since the walls of a
power idant usually have to support more than building and dead weight.
Concrete may be pven a smooth clean finish which will harmonize with foun¬
dations and floors of the same noaterial. Supports for buses, pipes, etc., can be
screwed into inserts placed in the walls at the time of pouring. Hence, ctmerete
is especially suitable for such locations as bus compartmoits, circuit breaker
cells, and pipe galleries. For strcmg permanent concrete the aggregates should
be clean, coarse, mid well graded. Table 4-1 gives typical concrete mixes with
the characteristics of each. This table is suitable for preliminary estimates or
for small amounts of concrete, but it should be remembered that research and
development in the science of concrete proportitming have advanced to the
point where little short of a laboratory analyris is required to establish the best
and most economical mix for a given condition. Ttp^ gradation of ag¬
gregate, moisture c<mtent of the sand, and water-cemrat ratio are typical
factors taken into account in a complete analysis for the specification of large
amounts of concrete work.
4-3 Reference Lines. The orderly dimensioning of derign <drawing3 re¬
quires the choice of reference lines from which to originate dimensions to
important subreference axes such as the center-lines of machines, etc. Such
references are usually either the interior walls of rectangular buidings, or
imaginary lines connected with some important part of the structure such as
the steel wall columns. Unless the shape of the building lot prevents it, wall
colunms are set in rectangular arrangement as shown in Fig. 4-3. An ima^-
nary line drawn through the center-line df a row of columns is called a column
line, and it may be designated by a number or a letter. By assigning letters to
east and west column lines, and numbers to north and south coliunn lines, any
particular point may be designated by the combination of a letter and a number.
For instance, point M would be designated “E22,” and N “D5.” A point cm the
second floor, 9 m above M, would be **E3-Orade 9 mif referred to the first
floor as datum grade.
The column lines may be accurately located by transit, at the beginning of
construction and readily re-established at any time; hence thsy form a con¬
venient stable reference source for the whole detdgn. It is usual to locate the
<^ter-lines of the principal equipment by dimennoning them to the eolurrm
line system. The foundations may thus ^ poured as soon as the cohimn lines
92 THE POWER PLANT BUILDING

iAii dimensions in umbl exeept where otheradee mentioned)

Fio. 4-3 Plan of small plant, illustrating the center-line to column line method of
locating equipment.
are located, so that they will be well cured before the equipment is placed. For
details of steel design the reader is referred to any standard text on structural
design.
4-4 Architectural Features. The architectural motif and the materials
in common use have been mentioned. The size, arrangement, and shape of
power plant buildings are principally affected by the type of plant, i.e., internal
combustion engine, steam turbine, etc., but one may generalize on common
features such as floors, roofs, etc. Roofs are generally flat, with covering of
built-up tar and gravel composition. The roof deck can be carried on trusses
and purlins, on reinforced concrete slabs and beams, on bar joists, or other
standard construction methods. The structure should be fireproof, and it is
often involved with other parts of the plant such as ventilators, stacks and
breechings, safety valve vents, etc. Floors are generally concrete, or concrete or
tile. Concrete is especially suitable, as it is fire resistant, impervious to rot, and
may have electrical conduits embedded in it. In accordance with the policy of
omitting combustible material, the doors and window frames should preferably
be of steel construction. Much use is made of pivoted steel window sash.
The heating and ventilating of power plant buildings are usually given
careful thought in the design of central stations, but can be lost sight of in
small plants, especially heating plants consisting of one main room housing one
or more boilers. With the thought that such rooms tend to become too warm
anyway, the designer is tempted to dismiss the heating problem. However,
many of these plants have b^n too chilly for winter-time operator comfort.
Two factors contribute to this. One is that modem boilers are usually very ex-
ARCHITECTURAL FEATURES gS
cellently insulated against heat leakage and often have the furnace walls lined
internally with water tubes. The exterior is cased in sheet metal, often painted
aluminum for decreased heat radiation. The other reason is the practice, in
plants of this type, of drawing the combustion air from the room instead of
through ducts and tunnels from outside. This exhausts the heated air so rapidly
that the boiler room becomes cold and drafty. Proper air duct design can elimi¬
nate this condition. Elsewhere heating may be needed in offices, electrical
rooms, control rooms. Turbine rooms are readily heated by discharging part of
the generator cooling air into the room. The hydroelectric station may some-

Fig. 4-4 Typical architecture of the Diesel plant.

times be heated thus, but it has been found necessary in some northerly situ¬
ated stations to install a steam heating system. Diesel engine plants can be
heated by passing the exhaust gas through a waste heat boiler and heating by
steam. Alternately the cylinder cooling water may be circulated through
radiators, or air may be circulated over finned sections of the exhaust duct
The illumination problem in power stations is somewhat specialized and the
installations are different from industrial lighting systems for reasons as noted.
1. It is not necessary for the worker to examine closely the product being
manufactured.
2. Considerable intensive local illumination is required, as, for instance,
instrument panels, gauge glasses, and control boards.
3. Emergency lighting of a dependable nature is absolutely essential be¬
cause the power plant is the source of lighting service.
4. No open, or mill type, wiring is permissible.
5. High ceilings eliminate any consideration of indirect lighting.
6. Fumes and moisture may corrode lighting units.
7. Rapidly moving machine parts are enclosed and protected, but walkways
near highly charged parts such as buses and switches will be hazardous unless
well illuminated.
8. Many small lamps are used for safety and for signaling purposes.
Internal Combustion Engine Plants. These plants may receive very simple
treatment. The auxiliaries to the Diesel engine are few and small in size. No
94 THE POWER PLANT BUILDING
coal pile hinders the landscape treatment; it is possible to locate the oil tanks
underground, although the reason for doing so is more often for convenience in
unloading from tank car or protection against the fire hazard than for a desire
to hide the tank. If the water for the jackets must be recirculated, a cooling
tower or spray pond will, of necessity, be located near the plant. The capacity
of the Diesel engine plant rarely exceeds 7500 fcw, the design of the building
being more or less directly determined by the dimensions of the engine units it
houses. The rectangular block shape illustrated in Fig. 4-4 is the usual result.
Hydroelectric Plants. Building and foundation of the hydraulic plant are
most important. In hydraulic nomenclature it is customary to designate the
building and foundation as superstructure and substructure. The substructure
of the hydraulic plant is the most massive and expensive single item of the
plant, excluding the dam and certain very high-head projects. The volumes

Hydro Plant- Hydro Plant-

SuPERSTRUCTURC BeLOWHeADWAT£RLevEL SUPERSTRUCTURE ABOVE HEADWATER LevEL

Fig. 4-6 Hydroelectric plant structure (when plant is located at the dam).

of water handled are large; consequently the water passages in the substruc¬
ture are large. Since the bulk of concrete involved prevents rearrangement of
design after construction, extreme care is necessary in proportioning water
passages. Compared to the substructure, the superstructure assumes minor im¬
portance, as is evidenced in Fig. 4-5. In the outdoor type of plant the super¬
structure has shrunk to a sheet metal hood protecting the exposed parts of the
generators from the weather, with a small auxiliary building housing switch¬
board and bus structure.
Steam Plants. The typical steam plant embodies three classes of building
treatment: boiler room, turbine room, and electrical bays. Head room required
in the boiler room will be greater than in the others, and ventilation presents
greater difficulty because of the heat liberated from the boiler surfaces. Usual
drafty conditions and the presence of considerable moisture in the air further
complicate the problem. Coal handling and draft equipment make it difficult
to illuminate the boiler room from above. Conversely, the turbine room becomes
the show room of the plant. Overhead, the turbine room is kept in the clear
ARCHITECTURAL FEATURES 96
to allow operation of the traveling crane. It is characteristic of the most modem
plants that practically no piping shows in the turbine room.
In central stations the room will usually be spacious^ uncrowded, well
lighted, and clean. It is, however, somewhat noisy, since this is inherent in the
action and construction of the rapidly turning turbine and generator. A typical
arrangement of the central station building structure is shown in Fig. 4-6. Also
other illustrations in this book, notably Fig. 1-11, may be examined in this
light.
Electrical bays contain switchboards, control boards, relay panels, motor-
generator-battery sets, bus galleries, and circuit breaker cells. There is usually
a large number of circuit breaker cells arranged in rows, sometimes all on one
level, sometimes in tiers on successive floors. If built of concrete, the reinforc¬
ing steel placement should receive the study of an electrical engineer to elimi¬
nate the chance of induced currents.

Fig. 4-6 Typical building arrangement, central steam-electric station.

Mezzanine floors of open metal grating are often used in the boiler room to
carry draft equipment, economizers, air preheaters, and coal handling equip¬
ment. In the turbine room, switchboards and control boards may also be ele¬
vated above the turbine floor on a platform or mezzanine floor. Below the tur¬
bine room floor the condenser and auxiliaries are located on one or more4evels.
Access to these various levels calls for widely different stairway design. Space
is often restricted; hence intermediate landings are employed to change the
direction of stair slant. Stairways infrequently used are made single width to
economize space. Steep slants and higher risers than those usually encountered
are permissible in the power plant.
Outdoor-type Sieam Plants. It is possible to save a considerable part (up
te a third) of the cost of the power plant building by an outdoor type of plant.
Many such plants have now been built. Boilers, turbines, tanks, fans and other
bulky pieces of ^am plant equipment are normally enclosed by a fairly
weather-tight casing, even wTien located indoors. The modifications necessary
06 THE POWER PLANT BUILDING
to weather-proof such equipment are not expensive. Of course, some equipment
must be housed; also, a building space is needed for offices, control rooms, etc.
However, the physical size is greatly reduced under that required for fully
covering all equipment.
A certain amount of inspection and maintenance becomes outdoor work for
operators of this kind of plant. The prospect is more attractive in the southern
than in the northern part of the United States, for heavy rainfall is then the
only weather obstacle to maintenance. Nevertheless, this type will be found
in states having cold seasons, such as Utah and Kansas. The greatest saving
in building cost is obtained where the fuel is oil or gas, and it will be noted that
most outdoor-type plants use these fuels. The simplification of building can be
judged by examining the arrangement shown in Fig. 4-7. No basement is ex¬
cavated; all equipment is set on a concrete slab laid at ground level. The boiler
and most of its auxiliaries are outdoors, as is the turbogenerator which is set
on the roof of the comparatively small building. The absence of coal bunkers
and ash basement is a major aid in holding the building to such modest dimen¬
sions. The illustration shows the extreme extent to which equipment is exposed.

Woftr Tunntit

Fig. 4-7 Building arrangement for an outdoor-type steam-electric station.

In many cases the building is carried up over the turbine, but leaving the crane
above the roof. Then roof hatches are provided over the heavy turbine and
generator in order to handle them with the crane.
4-5 Chimney. The products of combustion from the furnace of a steam
generator must be discharged to the atmosphere in an appropriate manner. In
some cases this duty is performed by a short stack or exhaust pipe, this being
possible where the gaseous products are not dust or smoke laden, and where the
furnace draft is produced by fans. But in many plants the draft, or part of it, is
obtained by the confinement of a tall column of the hot gas in a chimney. Con¬
sequently, between draft producing and the need to carry the dusty products of
a coal fire to a required altitude before discharge into the atmosphere, the
chimney liecomes a usual feature, one might almost say a trade mark, of the
steam power station.
From the architectural standpoint, there is nothing unlovely in the cylin¬
drical rise of a chimney, towering 30 m to 60 m above the plant. Indeed, the
stately shaft of a concrete or masonry chimney visible from af^r is a monument
to po^er development. Yet there are some plants, chiefly those connected with
educational institutions, hospitals, and the like, where both the power plant and
CHIMNEY 97
stack should blend in with the general architectural treatment of the building
group. Fig. 4*8 shows what can be done with the chimney from an architectural
standpoint. The important thing is to have the combustion equipment engi¬
neered so that the chimney top is not continually surmounted by a plume of
smoke.
Where mechanical draft is used, the necessary stack height is just that
needed to discharge the gases above surrounding buildings Except in metropoli¬
tan districts, this permits the use of a stub stack of sheet iron construction
mounted directly over the boiler room When a masonry stack is located over
the boiler room, its weight must be earned down to suitable foundations by a
steel or masonry structure in the boiler room.
The necessary diameter and height of a chimney are determined by gas flow
and draft. These topics are given due consideration m Chapter 12. Where the
chimney tapers, for stability reasons, the '^thermal diameter” is the internal
diameter at the top of the chimney. Neglecting seismic disturbances, the struc-

Fio. 4-8 Architectural treatment of a chnnney St. Paul’s School, Concord, N. H.

98 THE POWER PLANT BUILDING
tural loads on a chimney are caused by its own weight and by wind pressure.
Dead weight is treated as a single vertical force acting through the mass center;
wind pressure, as a concentrated horizontal force acting through the centroid of
the vertical projected area. The loading conditions will next be discussed, with¬
out reference to the type of chimney, then the applications to steel, concrete,
and masonry forms will be shown.
The following symbols will be used in chimney analysis:
W => Weight of chimney,kg.
Pw Horizontal wind lo^,kg.
H B Hei^t chimney, n.
h B Point application of resultant of horizontal wind load above
base,m.
r B Ind^ radius of chimney, m.
R B Outside racUus of chimney, m.
z B Factor such that Wz b P,A, z may be considered as the virtual
diq>lacement of W due to wind action.
fa B Ckimpiessive stress due to W, kg/m‘
tat BencUng stress due to Pw, kg/m*
k B Kem radiu8,m.
Experimental data show that wind pressure on a cylinder is approximately
two-thirds of that on a plane surface of the same projected area and shape.
However, there is a lack of definite and correlated information to guide the de¬
signer. The unit pressures existing against a 90 m chimney are different from
those against a 15 m one, even in the same wind. Therefore, some building
codes specify design pressures variable with chimney height, rather than wind
velocity being the variable. Where a local building code does not furnish the
eng^eer with a specific design load the following may be logical. At 160km/h
wind speed (a common specification for chimneys) ,150kg/m*windpi'essureisset
up on a flat plate of a shape similar to the projected area of a common chimney.
Applying the two-thirds rule, one obtains therefrom 100 kg/m* design wind
pressure.lThe total wind load is, therefore,
Pw=100X2PXH (4-1)
As most chimneys taper from Pi at the base to R2 at the top, the following
equation is more useful than Eq 4-1.
Pw =100H(Pi-I-B*) (4-2)
The location of Pw at the centroid distance h above the base may be found by
graphical or anal}rtical methods. The product P^h, which is the moment of wind
load, is derived using h, the centroidal distance of a trapezoid, as a function of
Rt, R%t and H.
Pwk = ^ (2P, + Pi) (4-3)

The forces W and Pw have a resultant determined as shown in Fig. 4-9. This
resultant, projected, intersects the horizontal section (above which W and Pw
exist) at a distance z from the chimn^ center-line, or neutral axis. This may
be required to be less than P for stractural safety in some types of chimneys,
CHIMNEY 99
but may safely exceed R in others. The intersection may also be thought of as
locating that point through which IF would have to,act if Pw were to be left out
of structural analysis (except for the determination of z). This is because
PwA = Wz (44)
It will be noted from Fig. 4-9 that the effect of W, alone, is to create uniform
compression /,. over the horizontal sectional area, while ^W» alone, creates a
bending action that results in variable compression, -h/b, on the leeward side of
a neutral axis and variable tension, —/b, to wind¬
ward. Steel and reinforced concrete chimneys can
have /b > /c, i.c., a net tension, but masonry con¬
struction should not be considered to possess any
tensile strength.
Steel Chimneye. These, if short, are made self-
supporting and are called stacks. High steel chim¬
neys are usually built for snfall, natural draft
boilers (because of low initial expense to the
owner)* and are braced with guy wires. Their small
diameter would render their design as a self-sup¬
porting structure difficult. Large-diameter tall
chimneys should be lined to extend their life, but
seldom are. Consequently, they need frequent re¬
placement and, as a long-range project, actually
become more expensive than a durable masonry
chimney. The large-diameter, short exhaust stacks,
often used at central stations where draft is cre¬
ated altogether by fans, are lined with brick and
consequently have a longer life expectancy. The
guyed steel chimneys are built with wall thickness
determined by rules of sheet metal shop practice.
3,35,,.2^180 and 2.0miii sheets, arej adequate for. MiMJe
most cases; in - fact' thickne88^ofJ/25-m«^^^ Dtod weight is distributed os
0 uniform compression due to
p;5int{n;| are often used where long life is not of the symmetrical shape.
importance, but light weight is. Steel chimneys are
generally built in welded sections and erected in
the field by riveting or welding horizontal joints.
Reinforced Concrete Chimneys. This type, like wind lood creates o food shodmg
the steel chimney, can withstand tension in the ^rom windward tension to leeward
structural column; therefore it can be built with chTmney^are^a^
thinner walls and less taper than masonry types.
The principles of stress analysis are those of stand- Chimney loads,
ard reinforced concrete construction. However, the
equations covering reinforced chimneys have been used to construct design
charts of relative simplicity. The concrete chimney is more costly to construct
than masonry because of the need for forming the structure. Furthermore, con¬
crete is not as fully impervious to weather as are the hard burned bricks of
the chimney trade.
Masonry Chimneys, Formerly, many chimneys were constructed of eommon
100 THE POWER PLANT BUILDING
brick, but the excellence of construction which may be achieved with perforated
radial brick has caused its general adoption for masonry chimneys. The bricks
are molded to suit the diameter of the chimney. The perforations aid struc¬
turally because the mortar partially enters them, giving good vertical bond.
The heat insulating properties of the dead air spaces formed are of advantage
in securing maximum draft performance of the chimney.
The maximum compressive strength of masonry is greatly influenced by
the mortar, and the actual crushing strengUi of the radial brick is many times
the safe design value for the masonry. A conservative value is indicate for a
structure of chimney proportions. With good lime-cement mortar, a radial brick
chimney that will not be stressed abovel5*6kg/em*a8the result of dead and wind
load should be satisfactory in any case where thermal strains are not unusual.
Radial brick masonry weighs betweenig22and3i03kg/nri Other physicalproper-
ties of radial chimney brick are mentioned in Chapter 12.
The conditions of stability of this type of chimney will be considered in the
following sectidn, where the reasons for chimney taper will become clear. Design
of these chimneys is a trial process; first a chimney section is assumed, then it
is analyzed to determine its economy and stability. As a guide to estimation of
the over-all shape of a chimney for a given inside top diameter da and height H,
the data of Table 4-3 may be consulted.
Table 4-3. TYPICAL RADIAL BRICK CHIMNEY PROPORTIONS
(iMkoa/h wind)
Top wall thickness,mm.180 to 230, dfyndiin; on
the dia^ter
Minimum wall thickness
at base, mm. 0-2517

Batter, (Ri — Rt),mm per metre,

as a fimction of H/dt ratio
BW/dt(w) BaKer(mm)
15 12
20 »
30 23
The table implies that the taper is uniform, i.e., a vertical element in the
outer surface is a straight line. This is frequently true, but not necessarily so,
for designs may sometimes be made more economical if the structural column
is considered in sections with sligjit differences in outside batter.
Exaniple 1: An approximation of the shape and wo^t of a radial brick chimney
column will be made for a 36 X 1.5 m chinmey.
Taking the top Ihickness atl90mm> >'* ■■ 750inm>B*~040tiim»0.04 m.
H/dt =^30/1.5 24. Estimating from Table 4-^, Batter » 21i! mm per m or
763mm for the chimney.
IlienRi - 940 + 763 1703mm - 1.7 m.
Bottom thickness »36x925 ■> 333mm. Use 322 mm(li)1}ridc).

r, - 1.7-0322- l.SSm
CHIMNEY STABILITY 101
Although the chimney as actually built would vary the thickness by steps, it is
here assumed that the interior is smoothly tapered so that the volume may be com*
puted as the difference of the frusta of two cones.

Outside conical volume «■ (1.1^ + 0.g4* + 1.7 X 0-94) — gOS*tg to?

Inside conical volume * (1.19* + 0.75* + l-*9 X0.75)— MI16 m*

o

Volume of masonry » 70.47 m*

4t IWkgjw?, chimney weight « 136210 kg
This particular case will be explored for stability in the next section.

4-6 Chimney Stability. As steel and reintorced concrete chimneys are

stabilised in the presence of high wind by tension in the windward side of the
shell, they will be stable if structurally safe and set on an adequate foundation.
It is the special feature of masonry chimney stability which we now consider.
As a starting point let it be assumed that the mortar in a horizontal layer be¬
tween radial brick courses is incapable of transmitting vertical tension. This is
substantially true because of the hard dense surface of these brick and the
slight adhesion of mortar to such surfaces. Referring to Fig. 4-9, it is seen that
the dead weight W produces a uniform compression over the horizontal area
— r®). Its magnitude is given by the following equation, with -h indicating
compression.
/c = +W/[t(R^ - r^)] (4-5)
If the chimney is considered as a hollow cantilever beam, the wind load
produces a compressive stress +/b on the leeward side, and a tensile stress —/b
on the windward side of the chimney. The common flexure formula of structural
theory has /b given by multiplying the stressing moment by the distance from
the neutral axis (centroid of the stressed area) to the point of maximum stress,
and dividing by the moment of inertia I of the stressed area around this same
axis, so the bending stress becomes MR/I.
For an annulus area, such as the cross section of a chimney column.
/ = ~ (R* — r*). M is the stressing moment Fwfc which, as has been previously
mentioned, may be replaced by Wz. These relations may be combined into the
following:
mzR
/b (4-6)
t{R^ - r*)

Note that, when a wind blows, /« and U act simultaneously—subtractively up¬

wind and cumulatively downwind.
In general a masonry chinmey may fail by crushing of the masonry on an
overstressed leeward section, by tipping downwind or laterally due to gust
buffeting, or by horizontal sliding. SUding is practically unknown. Downwind
tipping is rare if the masonry is safe against crushing. Lateral stability will be
unimpaired as long as i^itive compreasion acts along the neutral axis. Hence
the stmetural detign criteria are chiefly these: Provide sufficiant sectional area
102 THE POWER PLANT BUILDING
to maintain total maximum leeward compression safely within the limits of
masonry strength and spread it far enough from the center of the chimney that
the horizontal section under load is not
decreased to the point of impairing
lateral stability.
Other causes of chimney failure are
uneven settlement of the foundation,
lack or failure of protection against
lightning strokes, and weathering of
joints due to age or poor mortar. As
these do not directly bear on shape
analysis, and as the remedies are ob¬
vious, this section will be devoted
primarily to the principal design cri¬
teria.
Upon consideration of the assump¬
tion of no possible tensile stress we see
that a portion of the stressed cross sec¬
tion will suffer a horizontal crack to
windward if the wind moment is suffi¬
ciently great to cause ft to exceed L in
magnitude. At one specific value of z,
—/b will exactly balance +/c, and the
windward edge will 'Afloat,” while 2/c
acts to leeward. This value of z is
known as the “kem” radius, k, and is
the limiting virtual displacement of W
if the masonry is to be everywhere in
compression.
To derive an equation for k, equate
the right-hand side terms of Eqs 4-6
and 4-6.
Fig. 4-10 Conditions for stability of the
economic masonry chimney. k = (4-7)
4R
In the general case, z will not equal fc, except by coincidence. The total
maximum stress will be
W 4WzR

This is reducible to

Upon consideration of the weakness of mortar in tension, one could arrive

at a rule of design that the maximum allowable value of z is fc. However,
chimneys designed in accordance with this will be found to be oversized com¬
pared to commercial chimneys erected by the several successful large chimney
contractors in this country. We shall now examine the basis of their economy.
CHIMNEY STABILITY 108
If the resultant of wind and dead weight forces lies safely within the base
ring of the chimney, and if the safe unit compreteive pressures are not ex¬
ceeded, then a partial crack to windward will not necessarily imply failure.
There is no reduction in stability transverse to the wind direction, provided the
crack does not open up over one-half of the area. This maximiun cracked condi¬
tion is shown in Fig. 4-10. Naturally, the neutral axis is shifted and the region of
zero stress shifts to NN. Let y be the limiting displacement of z to obtain this
condition. Lange,* whose equations are widely used by chimney designers, gives
this limiting value for 2,

If 2 y, the maximum compressive stress is

/.-y = 2/,y/A: (4-10)
Lange gives the following equation ior y > z > k:

(‘- (^ ’]
0 0
The factor of safety against tipping downwind is the ratio of the stabilizing
moment to the wind moment, both taken about the downwind edge.

WR
FSo
P^h
but since P^h = Wz,
FSor = R/z (4-12)
This factor of safety should not be less than 1.5.

Example 1: The chimney proportions roughed out in Ex 1, Sec 4-5, will be

analyzed for stress in alOOkm/h wind. From Eq 4-3, Py^h = ——(2 X O.es
o
4- ].7 ) = i6S5201[g tn . Add to the weight previously determined for the structural
column the weight of9 m of a J00«nabrick lining (12480 kg). Then gross W on base =
136210 + 12480 = 148600 kg.
Calculate z, k, and y from the respective equations.

165620
148690
1.06 m; k *4 y = ^(2 + -l.I98m

The unit dead load stress is

148600
47.810 kSlm
•" t(6U» - 1.3«*)
From Eq 4-0:

Mrraokg/ta^
From Eq 4-12:
FS, 1.7/UBI - l.(
. *Der Sekomsteinbau. Gustav Lange, He'-wing’sche Verlags-Buchandlung. 1896.
104 THE POWER PLANT BUILDING
These results indicate a fairly economic design without, however, pushing the theory
oi the ^ludf-crack to windward’* to the limit. Since z is nearer to y than to k, a con¬
siderable windward floating may be expected at lUOfcjBi/h, but
not to reach as far as the MM axis. Therefore full lateral
D*. i4om stability is preserved. The FSor being well in excess of 1.5,
—j- dj* 1,82 m there is no chance of tipping instability, while an A of
I 0 195780/195800, or 80 % of crushing strength, is certainly
I working the masonry conservatively in compression.

In the foregoing examples the chimney has been

!-j og-Mow considered to be conical on both outside and inside.
S'* I Actually, the inside taper would be obtained by step-
s -J—220mm pmg the courses. The way this chinmey was actually
•f f laid out by the contracting builder is shown in Fig. 4-11.
I i D4.2aim ^ complete chimney analysis will include an investi-
E I L * gation of each section where the thickness changes.
* E These must meet the same stability conditions as the
S I base itself. An example of the section by section method
1 K ,will
03s2«68m
now be given.
I -302 mm
Example 2: The upper two batter sections of the chim-
* L ^LiNiNo shown in Fig. 4-11 will be analyzed for compressive stress
r and lateral stability. Masonry volumes are found by the prod-
f L n uct of average mean annular area of a section and its height.
02«3*04m
I 02*3«04m intermediate bases will be designated by sub-
gl ^ —330mm scripts, such as 4 for the annulus at Z> = 9.91 m and 46 for
m IM the upper 15 m section.

" ^ AnoZyais of section at A, t.c., upper ISm:

»46 + »66 ="7.5 X 1.997rX<kll»+7.5 X l-SOir X0.1J9L. 90,95 m»

Fig. 4-11 Profile—
radial bhek chimney. TF46 =20.96x 1922 «■ 40965 kg
Employing same equations as in Ex 1

Pw« -lOOX = *15®'^

PJ^ « = 22950 kg m

22930 „„ , 1.16*+ 0.94* „ 1.16/„, 0.94 \ ...

=0.48m;i„= — (^2 + --j = 0.g m

= 402«6/ir(l.I3* - a94») - 29178 kg/m»

U +^) - (i - j:5)(|i)’] -
Since y^> Za> K lateral stability is unimpaired. As /m is only 63900 when 195800 is
allowable, the masonry is definitely safe against crushing.
Analysis of section at Dt, t.e., upper 30

«!n + Vkt -7.6 X2.S8T X 0.S +7.5 X 24 t Xa27 - 8246 m>

FOUNDATIONS 106
Wu ~ X246X 19S9>Tetl80kg
TTm- 62196 +4C26K a± 102461 kg
Note that
PwAm ** P'Jhl + P'mt»(.h4» + 1®) “• PJht + Puhn + 16P»46
100 X 16* /2 X I.3II + 3;04\ .
Pwhu --g-(-2- 1 = 28725 kgn.
P,Am « 28725 + 22950+ IS X3160« 08926 kg m
98826 1^ + t.22*
it x^0J168m;ifc, »0.62Sm; yj = 1.0ft m
102461 ' 4 X 1.62
/.I = 102461/ira.62» - 1J2*) •= 40984 kg/nt

The chimney is thus also determined to be safe and stable at section D2. In a similar
manner the analysis could be extended to section Di, with reisults but slightly different
from the approximations of Ex 1.
4-7 Foundations. The foundation is a most important part of the power
plant building. Determination of the bearing power of the subsoU, selection of a
working factor of safety, and proportioning the wall footings to economical
construction, all are matters requiring a refinement of engineering judgment.
The foundation is that part of a structure which transmits the loads to the
supporting material. In the design of a foundation it is essential that the settle¬
ment shall be reduced to a minimum and that this settlement shall be uniform at
all points. The first requirement may be fulfilled by providing a bearing area
which is large enough to reduce to a safe value the bearing pressure on the
underlying material. In the case x)f soils having low bearing values, pile founda¬
tions can be used to reduce settlement. The second requirement may be secured
by designing the foundations so that the resultant of the vertical loads passes
through the center of gravity of the foundation. If the material under the foun¬
dation is structurally sound rock, having a bearing value within safe limits,
there will be no appreciable settlement of the structures; but there is bound to
be settlement in structures whose supporting medium is earth since it is a com¬
pressible material. A bearing wall rests upon a continuous fouildation, whereas

Table 4-4. SAFE BEARINGTOA^ER OF SOILS ^

Native rock. 1960 up

Ashlar masonry. 245-290
Best brick masonry. 145*195
Common brick masonry. i50*98
piay, compact. 60-73
Clay, soft. K5-19.5
Gravel and sand. 78-98
Sand.
Quicksand, loam, etc. 5-9.6
^Ira O. Baker, Tredtiae on Masonry Construction, John Wiley and Sons, publishers.
 106 THE POWER PLANT BUILDING
skeleton steel construction has piers at the main column bases, with lighter
footings from pier to pier to carry the lowermost curtain walls. Test hole drill¬
ings will establish the bearing character of the underlying strata. Sometimes
the soil is tested by loading a definite area, laid open by excavation, until it
reaches the yield point, settling being determined by running weekly or monthly
levels from a bench mark to a marked station on the load. A widely quoted
reference on the safe bearing power of soils is given in Table 4-4.
When excavating to suitable bearing soil, spread footings are used to dis¬
tribute the load over sufficient area. It is assumed that any concentrated loads
on a bearing wall (roof trusses,
floor beams, etc.) are given uni¬

Unit compres-
1 formly to the foundation unless
the proximity of the i)oint of ap¬
isive stress p
LJ plication to the foundation line is
’.A ‘ * I
along this line
such that it is apparent that a
concentration must exist. Column
Fig. 4-12 Spread footing. piers must distribute the axial
load in the columns and be thick
enough to resist the punching shear of the column on the concrete. Whether
required by the character of the subsoil or not, the column footings should be
spread wider than the wall to give stability. Fig. 4-12 illustrates the spread
footing. In the following example, the necessary offsets are determined. The
spread can be divided into as many offsets as the designer thinks expedient upon
comparison of the extra form cost against the saving in concrete mass. After
dividing the total spread into equal a^s, per block, the depth b is found by the
use of the following equation
b = a/V48//p (4-13)
where a and b are dimensions as in Fig. 4-12.

p = Unit compressive stress on the base of the block to which a and b per¬
tain,
j = Allowable stress, taken as 0.146 t/m^for concrete.
Example 1: Assuming that the wall in Fig. 4-12 is i60mmwide and carries 17^858
kg/m to the foundation, the dimensions of a two-step footing are determined. Soil
bearing is taken at i«.6 t/m^. The wall reaction is 17.86 t/wi, which would require
17.88/14.6,011.22 m width of the lower step of the footing, neglecting the footing weight
itself. However, this gives some indication of the footing size and we here allow for
a section of 0.76 m* area. At 2402 kg/m* this furnishes an estimated L8 t/m addi-
10 668
tional soil load. Trial width =rrr— 1.35m^
14.0

Offsets, a „ Lgft M§ ^ 0323m - 223mm

4
Width of upper step *=0.460+ 2 X 0.223«* 0.906 m
18.456
Estimating its weight at 0.6 aomie, pi 10.37 t/ro*
0.906
FOUNDATIONS 107
From Eq 4-13, 6i « 0.t*S/V48 x 0.1i«/iD.f7 i. 0J8 m
pj ** 1.5 tsf (specified), bt * 0.1tS/V4$ x 0.146/14.<5 m 0.99 m
Next check the assumption of J .8t/m of foundation.

W »r0.006 X 0.88 4. 1.859 x0.it)9.40i « i.87/m

This is nearly the same as assumed, but a larger discrepancy would have indicated
the need for a recalculation.
In case the soil condition is such as to require an extremely wide footing,
timber or I beam grillage is placed under it to assist the bearing power of the
soil. In fluid soils, piling is used to give sufficient bearing power. Both wooden
and concrete piling are in use, the former being the cheaper but the les^ desirable
of the two. Some large power house foundations are constructed on a mat of
wooden piles driven op 900 mm or 1200 mm centers, sawn off just below the
ground water line, and surmounted by a monolithic concrete capping which ad¬
ds to the bearing of the piles by making available whatever bearing power the
soil between the piles may possess. A common formula, known as the Eng^
ineering News formula, gives

Safe load on a pile = (4-14)

In this forumla w = weight of pile driver hammer,kg; H = its fall in m; S =

penetration under last blow,cm*The author would emphasize that the selection
of length and spacing of piles used to
stabilize the foundation of heavy build¬
ings requires specialized, experience and
geological training.
Static foundations which may be
eccentrically loaded require adequate
analysis of the effects of external mo¬
ments or eccentric loadings. A chimney
Square Sloped Octagon Stepped Octagon
foundation will illustrate the point.
These are generally of monolithic con- Fig. 4-13 Representative chimney foun-
cretc and should spread the chimney dations.
load sufficiently to prevent overstressing
the soil or completely neutralizing the windward compression. Typical shapes
are shown in Fig. 4-13. Foundations for round chimneys are generally octag¬
onal, being aboutSOOmmlargei than the chimney at the top base. From this they
should increase to a lower base sufficiently large to hold the maximum soil
pressure to design value, but not to have a width less than one-tenth the
chimney height plus chimney diameter at its top. The depth of the foundation
may be made 4% of the chimney height, but should not be less than 1.2 m-
Example 2: Given a sloping octagonal foundation gm across flats on the base,
and 9.4 iDd66p,weighingi45toiin6s.Thi8 foundation will be tested for its ability to carry
safelyaiea iotuM chimney en which the calculated wind moment is 179795 kg m and
wind pressure is 8888 kg. Foundation rests on a subsoil of compact clay.
108 THE POWER PLANT BUILDING
The wind moment must be transferred to the foundation base which is 2.4m below
the chimney base.
Moment M on foundation » 1707SO + 9888 X 8.4- 208461 kgm.
Weight W on subsoil * (146 -f 168) x 1000 « - 800000 hg-
208461
0.66 m
* “ 808000 “
Since k * 0.133 X 0 ~ 0.8 m , A; > 2 and there is no windward floating. Using
Eq 4-8,

Area of the octagonal base = 0.828 X 0* = 80 m».

^ 808000 , __ , .
/max = —rr— X 1.826 18786 kg/m*
80
This is less than20t/m* which.is seen from Table 4-4 to be a conservative loading
for compact clay.
Machine Foundations. The machine foundation performs far more than
the simple bearing function of the wall footing. The machine foundation
must:
1. Distribute the weight of the machine, the machine bed plate, and its own
weight over a safe subsoil area. If heavy unbalanced vertical kinetic forces are
produced by the machine, they should be added to the dead weight to obtain
bearing area. The dead weight must be well in ex( css of these vertical forces.
2. Provide sufficient mass to absorb machine vibration. Satisfactory founda¬
tion weight for this factor is not readily calculable. Table 4-5 is given to pro¬
vide an indication of these weights.
Table 4-5. WEIGHT OF MACHINE FOUNDATIONS
PER BRAKE METRIC HORSEPOWER

Prime Mover Single Cylinder MultincyUnder

Gas engine. 1186 kg 780 kg
Diesel engine. 910 kg 670 kg
Steam engine. 820 kg 280 kg
Steam turbine. Not to exceed permissible
deflection as stated by
turbine manufacturer.

These are handbook values which were evidently chosen to provide suflBcient
mass to dampen vibration almost completely. The concrete foundations yielded
by the application of the above data are sometimes grotesquely huge, obviously
unnecessary for the average installation. The data should be regarded as the
extreme upper limit of necessary foundation weight and be decreased appro¬
priately when complete dampening is unnecessary. If the shaking forces can be
calculated, a mass of weight equal to 10 to 20 times the forces should be ade¬
quate to dampen vibration.
Safe bearing power of soils for machine foundations are from a quarter to a
half of those given in Table 4-4, depending on the machine. Unless a uniform
FOUNDATIONS IO9
foundation soil is available, the concrete in the bottom of the heavy foundation
should have reinforcement.
Example 3: To illustrate, in a simple manner, the principles involved in founda^
tion design, the foundation of the 250-kw, 4-cylinder, vertical, unaflow engine shown in
Fig. 4-14 will be designed. The weight of this engine, together with generator and
exciter, is 45800 kg. The bedsoil is found to be dry sand, not exceptionally firm.
It is noted in Table 4-4 that the safe bearing power of the soil could be taken as
10.5t/iiiP,but, as this is a machine foundation, that will be halved, giving an allowable unit
soil stress of 0750 kg/ms Table 4-5 gives foundation weight 080 kgparbmhp for this type
of engine. The bhp of a 250-kw unit will be about 875 m bp, making the
Total weight of unit and foundation 46850 -f 875 x 080 -• 181600 kg
Base area of the foundation = 181600/0750 = IMm*

Fig. 4-14 Unaflow engine foundation.

A base 8.05 m X 4.05 m gives 18.0 mS. The sides are tapered to a top width of
8.70 m, making top area 11.7 m*.
Considering the volume of the pit left under the generator to equalize the projection
of the bearing and exciter pedestal, and neglecting the 500 nun rise under the engine, we
may find, roughly, the depth d of a foundation which will give the required mass.
The foundation weight is to be 88915 kg ; hence concrete volume == 85 m* .
Average horizontal area = <11.7 + 18,8)/8 = 12.75 m*. Then 10.75 d = 86 , and
d = 0:75 m depth. Now as this is a rather deep foundation, it is *»vident that foundation
mass, and not bearing area, is the governing factor.
By increasing the top to 8.0 m X 4.5 m and the base to 3.5 m x 4i.5 m

Average area ~ (18.5 + 15.75)/2aB14.605 mZ

and
d » 86/14.685 0.4 m
The 46860 kg of weight is not uniformly distributed; probably two-thirds of it is in
the engine itself. The average soil stress »181609/15.75 — 8866 kg/ni8* The moment of
imbalanced weight distribution will increase this somewhat. From the previous state¬
ments one can take one-third of the machine weight as an eccentric load. From Fig. 4-14,
tio THE POWER PLANT BUILDING
eBtiinate that the eccentricity from midbase is 1 m . Then to use the common flexure
Mz
formula, /b * “T'* the following are computed.

M » ISlSOkgm

z 2.f5m
2
>>5 x 4.5»
I 26.55 ml
12
161S0 X 2.25
/b I260kg/iii>
26.55
Hence extreme soil stress = 8256 -f 1260 = 2616 kg/mt, which is less than the
2750 kg/w^ taken to be allowable.

Large turbine foundations are not required to contain foundation masses

comparable with similar reciprocating units. Still, the necessity of providing
space beneath the turbogenerator for condenser pumps, generator air cleaner
or cooler, and sometimes the throttle lead, materially complicates the design of

Fig. 4-15 Typical turbogenerator foundations.

the turbine foundation. The turbine foundation does not carry the turbine upon
a lieavy bedplate, and hence a study of the foundation deflections is all-im¬
portant. No two foundations are alike. Reinforced concrete and structural steel
foundations cacli have their advocates, but any installation should be figured
upon a basis of comparable costs of the two types, because each is suited to a
particular field of utilitj^ and eeonoiny. The concrete base gives more rigidity
to the turbine, but it is claimed for hteel that its flexibility is an advantage in
large units as preventing distortivc bowing of the shaft and attendant difficul¬
ties. The concrete foundation will require less maintenance; the steel type yields
more available space below the unit.
Although the vibration problem is not so serious in the turbine as in the
reciprocating engine, there is another factor of importance. Referring to Fig.
4-15, it is seen that the portion of the low-pressure shell above the condenser
inlet is subjected to a downward thrust, while the corresponding area across
FOUNDATIONS 111
the turbine outlet is enibjeoted to condenser pressure. The resultant downward
thrust is exactly as though atmospheric pressure acted on one side of a hori-
sontal plate whose area equaled that of the exhaust nozzle cross section, and
absolute condenser pressure on the other. If the absolute condenser pressure is
Pc kg/cm', and the cadiaust nozzle area is Acm*, then the vacuum pull is
F - A(1.03 - p.) kg (4-16)
If the condenser is bolted directly to the turbine, this pull is neutralized in
internal stresses in the shell. In such cases the turbine foundation must support
the weight of the condenser as well as that of the turbine. Sometimes the con¬
denser weight is supported partially by springs, but the thrust produced by
them must not be sufficient to take all the gravity load off the turbine flange
when the condenser is empty of water.
Example 4: Vacuum pull can be a rignificant factor in the deagn of turbine
foundations. Consider the ease of a 15/XX)-kw turbine with exhaust pressure ot^Akgimf
eb. A geiieiatinc unit of this oae would have an exhaust opening about M4 xSdSem.
Substituting in Eq 4-15,
F - 244 X 366 X (1.03-0.1) 1-83053 kg
Setting Machinery to Foundations. Customarily, machines are attached to
their foundations by anchor bolts which are east in the c<Hicrete and project
above the foundatitm surface far enough to pass through holes in machine bed
plates and be secured by nuts. Two forms of foundation bolts are illustrated in
Fig. 4-16. The simpler, cheaper form is permanently set in the concrete and the
machine must be Ufted over the bolts and carefully let down so that they pass

Fio. 4-16 Attachment of equipment anchor bolts to concrete foundations.

through the holes provided. The other type is more expensive to construct, but,
being removable, the machine can be skidded sideways into position, then the
anchor bolts inserted into the prepared threads. Small machines for which an
anchor bolt setting template might be prepared directly from the machine could
be successfully set on foundaticms, the anchors to which were rigidly cast in
concrete. However, the bolt settinp for larger machines must be prepared from
blueprint information and should always be set in pipe sleeves so that some
minor adjustment of position can be made if necessary when setting the
machine.
When bringing a machine into alignment on its foundation, shims or sole
112 THE POWER PLANT BUILDING
plates are placed beneath the frame or bedplate and adjusted until the elements
being aligned are correct. The elements might be the edges of the bedplate, the
center-line of the shaft, a casing joint, etc. After alignment is secured, a tempo¬
rary dam is built around the top edge of the foundation, and a grout flowed
beneath the bedplate.
Grouting is the process of filling a small clearance between machine and
foundation with a hardening mixture. Grout is usually made of a 2:1 mixture
of sand and cement, mixed wet enough to be fully fluid. The machine having
been shimmed up level about an inch above
its foundation, and a board or clay dam
having been run around the edge of the
foundation, the grout is poured on and
allowed to run into and fill the clearance,
and to rise slightly above the lower edge of
^ rILLER BLOCKi
(if required) the machine base. After the grout is hard,
the dam is removed (shims also, if pos¬
FOUNDATION FORMS sible), nuts are tightened on anchor bolts,
DIMENSIONS SHOWN ARE and the exposed edge of foundation smooth
LAID OFF CAREFULLY FROM
FOUNDATION DRAWINGS
finished with a neat cement plaster. It is
preferable to pour the grout in two steps:
to the edge of the bedplate, then to the
edge of the foundation after removing the
Fig. 4-17 Setting of anchor bolts. shims. However, mechanics are not always
willing to take this trouble.
4-8 Miscellaneous Equipment. It is customary to provide the turbine or
engine room with a traveling crane of sufficient capacity to handle the heaviest
single piece of equipment in the room. Since these rooms are usually clear of
overhead obstructions, the crane in greatest favor is the bridge type with rails
supported at opposite walls. If bearing walls are used, the crane load is carried
down to foundations on brick or concrete columns called pilasters. In steel
construction the crane rail supports are either brackets attached to the building
columns, or are short columns made integral with the building columns.
Any coal-fired power plant poses a major problem in cleaning. From the
time in which the chief tool used to keep the plant clean was the shovel, plant
owners have progressed to a stage which recognizes the economic and moral
losses occasioned by dust and dirt. A dark, oilproof, enameled surface is one
which is easily wiped clean and does not readily show fingerprints. Floor paints
give concrete floors a surface that can be mopped clean. Aluminum paint on all
boiler room steel and duct work greatly improves illumination. The boiler room
offers the major problem in cleaning. Because of its finely divided state, coal
dust may prove an explosion hazard; and of course it excels as a begriming
agent. It is difficult to remove except by water flushing or suction. Confinement
of coal dust by functionally separating fuel preparation and handling from the
remainder of the building, use of covered conveyors and hoppers, and similar
measures are helpful. Plant design should give careful attention to illumination
and to provisions for cleanliness, but these have all too often been neglected.
Wash rooms, shower baths, toilet rooms, and offices may or may not be
provided in the plant, depending upon its size and upon local conditions. The
MISCELLANEOUS EQUIPMENT 113
central station in the country must be very nearly a self-contained community
with its operators' village, its store rooms, and its repair shops. When a power
plant is privately owned by an industry, the same repair shop may serve for
both factory and power plant; otherwise the power plant must contain its own
workshop, small or large as the case may be. A public utility plant generally
requires enough maintenance to employ a repair crew regularly.
Power plant maintenance requires large stocks to be kept on hand, since the
plant is a specialized industry. How to account for and issue the many small
and large parts that must be carried in stock is a problem. Periodic inspection to
detect wear, thus enabling a part to be ordered in advance, will aid in reducing
the stock necessary to be carried. The storekeeping in a power plant should be
as simple as possible, and yet should maintain a stock that will be insurance
against prolonged breakdown.
Building arrangements of central power stations often must include execu¬
tive offices, a control laboratory, equipment control rooms, and load dispatching
centers.

PROBLEMS
(t/m^cz: tonnes per square metre ;kg/m> = kg per cubic metre)
1. How high can a solid, unloaded brick wall (S.08 tonnee per m? ) be carried with¬
out the compressive stress on the lower course exceeding (a) 12 kg/om>, (b)l7.5 kg/emt?
2. The brick walls of a small power plant are ofthebeanng wall type,482 mm thick.
They arelO.Sm high to the eaves and take a reaction of B400kg from roof trusses spaced
4.Smapart, (a) What is the maximum compression in the brick work? (b) The safe
bearing power of the subsoil is 9788 kg/m>»and the wall footings weigh approximately
495 kg/m. How wide should the footings be ?
3. A brick power plant building islBmxll m x7.5 mhigh, with flat roof. Walls are
644 m thick for lower 4.6 m,482 mm for upper 8m.Thereiarel0window openings 1.5mX2.0m,
and 2 door openings 2.0 m x 2.5m.Thfi walls are built with thin cement mortar joints. Calcu¬
late the cost of building materials for the walls, given the following unit costs: Brick,
$40.00 per 1000; cement, $4.00 per bbl* sand, $3.50 per oubio metre.
4. What is the cost, in place, of a brick walls mhigh,21 mlong,482 mm thick?Thin
cement mortar joints. Bricklayer, $15.00 per day; helper, $8.00 per day. Briclc,
$32.00 per 1000; cement, $4.00 per bbl* sand, $3.00 per tonne (1720 kg » 1 ).
5. An engine is to be set on a S3rmmetrical concrete foundation!.2mx8.5m top face,
2.1 mXSOlmbottom face,1.6m deep. Make a list of the materials to be used in the construc¬
tion of the foundation and calculate the foundation weight.
6. A 80.5 m circular water tunnel of l«8m inside diameter and 805 mm thick walls is to be
constructed of concrete. Allowing 4 bags cement to the bbl, find cost of materials based
on the following unit prices: Cement, $4.50 per bbl; sand, $3.50 per tonne; stone,
$4.00 tonne. Sand and stone weigh 1720 kg/m*.
7. Estimate the quantities of stone, sand, and cement required to build a circular
concrete water tank (wood cover) to hold 7200 litres water. Interior depth of water,
1.5 m; of tank, 1.8 m. Walls and floor, 800 mm thick. Neglect volume occupied by steel
reinforcement.
8. A89.5mhigh chimney of radial brick masonry is described by the following top
and bottom dimensions. Dz = 1.9 m, ^2 = 1.5 mi, = 8.2 m, =2.Sm. Uniform
batter, weight 180 tonnes«Find the maximum compressive stress under 180 km/h wind
load. Will a base crack be opened to windward?

^Allowing 4 bags cement to the bbl

114 THE POWER PLANT BUILDING
9. The upper 15.25 m of a radial brick chinmey has Dj =: 2.2 m,2 m,wi^ thicl^
ness 178 mmJs it fully stable laterally at thei5.25naBectionand3afeagam8tcrush)ngwbeD
subjected tol60 lun/h wind? Density,2408 kg/afl; allowable compressive stress,12.66 kg/om*.
10. Using Table 4-3, estimate the proportions of a radial brisk eUmney of 2.4 m
inside diameter X 45.5 m high. Masonry weight, 1922 kg/m^-Wind,!60 km/h«Test the base
and mid-height sections for maximum compressive stress. No lining.
11. Repeat Prob. 10, except for 2.75 m x 60 m chimney.
12. Analyze the chimney base section as mentioned at the end of Ex 2, Sec 4-6.
Lining to weigh 12.4 tonnes.
13. A 45 m X 2.4 m chimney has been laid out as follows.

Section 123456 789

Ht. above ground, m 0 1.5 4.5 10.5 16.5 22.5 80 87.5 45
Batter, mm pel m 80 80 80 27 27 18.5 18.5 18.5
Wall, mm 464 425 887 887 298 278 216 178

Lay out this chimney to scale 1 cm=1 m-Determine its weight, cumulatively section by
section, using 1922 kg/m* both for main column and lining. Lining extends from 18.5 m
to 29 mabove ground, is carried on corbells at |8.5 mand21jl m^ is 100 mmthick and built
to give minimum 50 mm air space.
14. Analyze the stability of the chimney of Prob. 13 at sections 1, 4, and 6 when
subjected to 160 km/h wind load.
15. Proportion, from Table 4-3, the c’himney whose dimensions are given economic
study in the example, Sec 12-12. Consider chinmey foundation to be 2m below furnace
grates. Will the chimney, as so laid out, be safe in 160 km/h wind, or will the economic
dimensions be jettisoned because of structural requirements ?1922kg/m4. Top thickness,
178 mm. Uniform inside and outside taper.
16. * A block-type concrete foundation must support a static load of 22.5 Unmes trans¬
mitted to it on a thick steel plate 610 mm x 916 mm-Design a foundation of minimum weight
that will be safe against settlement and cracking. Estimate the concrete’ materials
needed. Bearing power of soil, 124/m2.
17. A482mmbhck wall 18.25mhigh rests uponaconcrete footing The soil has a bearing
power of 12t/m6. Design a two-step offset footing of concrete and detail to a scale
of 10 cm 1m
18. A reinforced concrete wall carries a load to foundation of 80t/mrun. The wall
is 500 mm thick. A subsoil bearing power of 12.5t/ii6 is allowed. Design and draw a
sketch of the wall footing (two-step).
19. The following data refer to a 6-cyHnder, 450-kw,650bmhp Diesel engine alterna¬
tor unit having direct connected exciter: Weight of engine and flywheel, 88500 kg;
weight of alternator and exciter, 6800 kg; bedplate of engine, 8m x 7.2 m; length of
whole unit,10.8mwi<ith of generator bedplate, 8 m.The subsoil is average sand. Design
and detail a foundation for this engine. The foundation top is all in one horizontal
plane. Calculate the amount of sand, stone, cement, and form Imnber required for its
construction.
20. The condenser <rf a 10,000-kw turbine is rigidly carrio^i by its own foundation
and is connected to the turbine by a copper expansion joint whose dimensions are
2.1 m X 23 m, approximately rectangular. Tbe water and steam normaUy in the con¬
denser weigh9072 kg. Condenser pressure ^0.07 kg/txiii8 ah. What is the variation in
weight on the condenser foimdation between in-servioe and out-of-service conditions T
21. Design a sloping octagonal concrete foundation for the chimney of Ex 1, Sec
4-6. Bearing power of soil I93ft/iii6. Justify any necessary assumptions.
22. A46.75 m chimney with Di ,4.71 weighs 272 toiiiies.The chimney is set
PROBLEMS 115
on a two^p octagonal concrete foundation, laid on soil having safe bearing
power. Wind load, 160 km/b. Design and make a drawing of the foundation.
23. A 60-kw vertical, single^ylinder, steam engine-generator set has a bedplate
1.6 m X t.S6m, Its wright isiSMkg. Design a foundation to rest on average sandy
soil. Foundation mass in accordance with Table 4-5.
24. The foundation under the engine of Fig.4-14i6S00iBm higher than that under the
generator. Considering the generator pit to have the same volume as the exciter pedestal,
find the amounts of sand, crushed stone, and cement required for this foundation.
25. A Diesel engine sets on a plain block foundation a m x Am x 176 m deep. Find
cost of foundation and form lumber (forms of *6 mm lumber). Lumber, S36.00 per 100
board metre; cement, $1.50 per sack; stone, $3.00 pertonne; sand, $2.25 per tonne.
26. Estimate the weight and cost of materials for foundations for the following
300mbp prime movers, using 70% of foimdation masses from Table 4-5. Get local unit
costs.
(a) Vertical multi-cylinder unailow engine.
(b) Vertical multi-cylinder Diesel engine.
(c) Horizontal single-cylinder gas engine.
(d) Horizontal single-cylinder Corliss engine.
27. Piles are driven in a quicksand (bearing = 6t/m*) on 000 mm centers. They
are driven until the penetration under the last hammer blow is 06 mm. The driver's
1-tonne hammer dropi 0.6 m . I^les are sawn off and surmounted by a concrete dab
900 ram thick. What average bearing power, int/mi*, does this foundation provide?
 CHAPTER 5

FUELS AND COMBUSTION

5-1 Fuels. The raw materials of the heat power plant are the poK ntial
energy dormant in a fuel and the oxygen necessary to convert it to a dynamic
high-temperature condition. A fuel is composed of chemical elements whicli, in
rapid chemical union with oxygen, produce combustion. As this definition is
contingent upon the meaning of combustion, the latter term will now be defined.
Combustion is that rapid chemical union with oxygen of an element whose
exothermic heat of reaction is sufficiently great and whose rate of reaction is
sufficiently fast that useful quantities of heat arc liberated at elevated tem¬
peratures, In all the fuels that are in common use, the elements which are sought
because they do have the necessary heat of formation and rate of reaction are
carbon and hydrogen. These two chemical elements are combined with others in
various w^ays, producing the fuels of commerce. Solid, liquid, and gaseous fuels
are employed, usually separately. Coal is the most used solid fuel for generating
steam, but sometimes industrial wastes are used. Fuel oil is the most important
liquid type of fuel. Natural, by-product, and manufactured gases form the third
group, but manufactured gases are seldom employed for they are high in cost.
The use of by-product gases is occasionally possible. Generally they are em¬
ployed only by the producing industry and are not distributed.
5-2 Gas. Natural gas is widely distributed in the United States, and a
growing network of pipe lines is making it increasingly available to industry.
Its cost per unit heat content at the wells compares favorably with other fuels,
but the cost of transportation through expensive pipe lines, having compressor
stations at periodic intervals, adds a delivery charge which causes it to be an
expensive fuel, compared to oil or coal. Its advantage is its convenience in use,
stemming from ease of control, smokeless combustion, and absence of ash. The
cost is a deterrent to any considerable use of it for producing steam. Methane
(CH4) is the major constituent of natural gas. In addition, this gas sometimes
contains ethane, nitrogen, and carbon monoxide. Most samples of natural gas
average about 37260Joules/litres measured at standard conditions of76«min,Hg
andl5.6®C.It is generally sold by volumetric measurement with the actual gas
flow corrected to standard conditions by use of temperature and pressure
measurements.
5-3 Oil. The liquid fuels of power plant practice are nearly always by¬
products of petroleum. Crude petroleum oil contains mainly carbon and hydro-
116
INTERNAL COMBUSTION ENGINE FUEL 117
gen, along with some small amounts of oxygen, nitrogen, and sulfur. The
hydrogen is present always in the form of hydrocarbon mixtures. This crude
oil is seldom used in its original form. Its decomposition into numerous com¬
mercial products is a major industry in the United States. The approximate
chemical composition of petroleum and its derivatives is: carbon, 85%;
hydrogen, 15%. Among products and by-products of petroleum the chemical
properties vary but little, although the physical properties, such as viscosity,
density, color, may vary greatly. Many special fuels are produced from crude
oil by distillation and by chemical cracking, which is decomposition under
conditions of high temperature and pressure. Cracking of crude oil yields a
larger fraction of I.C. engine fuel than does distillation. The tremendous demand
for engine fuel is reflected in the processing of most crude oil by the cracking
process. Thus it will be found that most fuel oils are the products or by-products
of a cracking process. The products produced by cracking refineries must be
separated according to boiling range into light oils and residual or heavy oils.
As an end result there are certain heavy, tarry refinery sludges which are not
conveniently utilizable in anything but very special equipment, used either in
the refinery power plant or in a near-by station.
The total proved reserves of crude oil in the United States in 1947 were 24.7
billion barrels. Comparing that with the current withdrawals, it would appear
that the liquid fuel reserve will be able to last only until 1960. However, pessi¬
mistic predictions like this have been repeatedly delivered in the past, and all
have ultimately been wrong, mainly because of the discovery of new reserves.
However, it certainly seems to be a fact that, before this century closes, the
pumping of pool oil from the ground will gradually end. Major oil companies
are, of course, cognizant of this fact and have planned programs that seem to
indicate that the demands for liquid fuels will be met for a long time in the
future. Such programs include large-scale importations from abroad, thus en¬
abling the companies to reduce to temporary inactivity some of our domestic
fields, and the production of synthetic liquid fuels from coal and gas. Further¬
more, pilot plants are investigating the costs of extracting oil from the oil shales
which represent a large, but at present economically unusable, source of fuel
oil. Further, the possibility of ultimately being able to harness nuclear energy
and thus replace the chemical energy of oxidation is very definitely a live one
at the present time, and one which beclouds the analysis of fuel type selection
based on a long-range economic plan. Nevertheless, it is noted that the cost of
fuel oil is currently higher than coal on a straight Joule comparison. In most
cases this throws the decision to coal, but sometimes factors such as the
simplification of the steam generator and its auxiliaries, lower initial cost of
equipment, reduced space requirements, elimination of some labor, and more
nearly smokeless combustion have been influential in favor of oil. This has oc¬
curred principally in industrial, commercial, and institutional plants and not
in central power stations.
5-4 Internal Combustion Engine Fuel. The I.C. engine as used in the
United States is operated almost exclusively on fuels derived from petroleum.
Hie exception is fuel gas for an occasional gas or dual-fuel engine. Although
Diesel originally intended to operate hia new engine on coal dust, neither he nor
anyone since has solved the ash problem. Crude petroleum has been abundant
118 FUELS AND COMBUSTION
in this country, but the mounting use of the liquid-fuel-powered I.C. engine on
land, sea, and air is making grievous inroads into natural supplies of crude.
The largest demand has been for gasoline on account of the automobile, truck,
and airplane. Refineries process the crude for maximum gasoline production.
Up to the present time Diesel fuel has been chiefly distillate, but, as the demand
for it increases, doubtless more of it will be blended straight-run distillate and
catalytically cracked oil.
The typical petroleum substance is a complex of carbon and hydrogen, mixed
in various compounds each having its own boiling range. The compounds are
classified in four groups; paraffins, olefins, naphthalenes, and aromatics. Carbon
is about 85% by weight: hydrogen, 15%. Fig. 5-1 shows the range of volatility
of the commercial fuel products derived from petroleum. Liquid-fueled spark-

BOILING POINT - DEG. C

Fig. 5-1 Products of distillation of mule oil

ignition engines require a highly volatile fuel for carburetion and therefore
need gasoline. Diesel engines can use a less volatile, cheaper fuel. Normally
their fuel has been a distillate composed of blended kerosene and gas oil, but
increasingly this is being mixed with cracked distillates. The resulting fuel,
though not favored by operators used to straight-run distillate, appears to
jierform satisfactorily if properly selected for the engine. Since the Diesel is the
principal I.C. engine for stationary power plants, the important comparative
properties of fuels suitable for it are to be considered. They arc:
1. Density. 5. Calorific value.
2. Viscosity. 6. Pour, flash, and color.
3. Volatility and residue. 7. Purity.
4. Ignition quality.
Density, Mass per unit volume is the universal definition of density. Kg
per litre is in common use for liquid fuels, also specific gravity (S.G.).
Density at 15,6®C= (S.G. at 15.6®/15«6®) kg per litre, (5-1)
The hydrometer furnishes the .simplest method of determining gravity. The
gravity bottle, or picnometer, is far more accurate, as are special balances, but
where its accuracy is adequate the hydrometer is preferred on account of low
cost and the ease and rapidity of use. In reporting S.G.data the 15.6®/I5.6®
standard is common, that is, the oil is atl5.6®Cand is referred to the density of
water taken atl5.6®G If the measurement is made at some other temperature, a
correction is in order. The volumetric coefficient of expansion of oil is 0.0007 per
®G.
Hydrometer scales will be found to be marked in “degrees” as well as S.G.
The degrees may be expressed in Baume or API units.
INTERNAL COMBUSTION ENGINE FUEL 119
140
® Baum4 = - 130 (6-2)
S.G. at 15.6715.6'
141.5
“API = - 131.5 (5-3)
S.G. at 15.6/15.6'
The American Petroleum Institute scale of degrees is accepted as the standard
by the petroleum industry. There is little difference between results using °B4
or' API.
Ex2miple 1: A 28° API oil has a temperature of i.4°C. Its density in kg per litre will
be calculated.
141
S.G. at 0.887

Eq 5-1: Density at 15.6®C = 0.887 kg per litre

Contraction fromll.2®C cooling = 0.0007 x 11.2= 0.008
Density at 4.4®0 ~ 0-887/0.002«b 0.894 kg/1

Viscosity, This property measures a fluid^s resistance to flow. In the C.G.S.

system the absolute unit is the ^^poise,” which dimensionally is expressed in
; in the English system, its counterpart is defined in terms of
sq cm sq tt
or alternately in 7-—. Kinematic viscosity is absolute viscosity divided by den-
ft-sec
cm^
sity. In the C.G.S. system the unit is “stokes^^ and dimensionally is-. In the
sec
sq ft
English system, dimensions are . Centipoises and centistokes are more
sec
commonly used.
The time taken by a fluid flowing through a calibrated tube at specified
temperature and head is proportional to kinematic viscosity. Laboratory meas¬
urement of viscosity is made with the Saybolt viscosimeter, consisting of a
temperature bath surrounding a standard oil tube. The Saybolt rating is ob¬
tained by timing the interval required for 60 cc of oil to flow through the tube.
As the viscosity varies with the temperature it is necessary to add a tempera¬
ture to Saybolt seconds when reporting viscosity. Most Diesel fuels have vis¬
cosity between 30 and 45 Saybolt Seconds Universal (SSU) at 37.8®C. In this
range centistokes = 0.308 (SSU—26). Another viscosity index is Saybolt Seconds
Furol. 62 SSF = 600 SSU.
Viscosity is an important quality of fuel oil. It is a fair indication of how
readily the oil will atomize and how it will affect the injection pump. It affects
the fuel spray droplets, thus governing the atomization as well as the penetrat¬
ing qualities of the spray.
Volatility and Residue. Diesel fuel being a mixture of hydrocarbons, it is
important to know the distribution of volatility, especially the two extremes of
low- and high-volatile compounds. This can be tested by the ASTM Distillation
Test in which a measured sample of the oil is slowly heated. As the various
boiling points of its compounds are reached they distill off, are condensed, and
120 FUELS AND COMBUSTION
measured. After complete distillation the data are used to construct a distilla^
tion curve of temperature vs. percent evaporated. The average range between
initial boiling point and end point is 163*C to 371®C. The distillation curve
should be smooth for good balance of compounds. High-speed engines require
larger fractions of low-boiling-point compounds than do the low-speed, since
the latter allow considerably more time for combustion.
A test is sometimes made for residue by carbonizing the fuel in a covered
dish and weighing the dry remainder. The result is an indirect indication of
the quantity of heavy, low-volatile material in the fuel and is especially useful
where the high-temperature end of the distillation test is inconclusive, possibly
because of the presence of residual oils containing high-boiling-point fractions.
The principal value of the test is to distinguish between distillate and residual
fuels, for distillates have very little carbon residue.
Ignition Quality, Of many factors influencing fuel ignition, the chemical
nature of the fuel itself is one of the most important. Means of measuring and
controlling this quality in a fuel are not yet perfected, but Cetane Number and
Diesel index are extensively employed. Cetane number scale is derived from
the practice of testing a fuel in a single-cylinder variable-compression test
engine and comparing its ignition quality with that of a standard fuel composed
of cetane (good ignition quality) and alpha-methylnaphthalene (poor ignition
quality). When the proportions of the standard fuel have been adjusted so that
the ignition quality* matches that of the fuel being tested. Cetane number is
the percent of cetane in the standard fuel. Present-day fuels range from 30 to
60 on this scale. The average high-speed engine requires a fuel of better than
45 Cetane. Full-compression ignition engines will scarcely run on fuels rated
lower than 25. Cracked oils show lower Cetane number than straight-run dis¬
tillates.
It has been found possible to get a comparative indication of ignition quality
by a rapid simple test requiring but little equipment. The “aniline point” is
that temperature where equal parts of oil and aniline fa coal tar derivative)
will dissolve in each other.
Diesel index = 0.018 X APItap+0.32*API (5-4)
where tap = Aniline point in °C.
This index has a range of approximately 20 to 60. In the 50-60 range Diesel
index is 5-10 units higher than Cetane number of the same fuel. It correlates
closely between 35 and 45 and is a few points lower below 30.
Recently the practice of adding components to fuel oil to improve ignition
quality has been growing. When these additions have been made, Diesel index
or other formulae are without value and ignition quality determinations must
be made by the CFR Diesel method. The additive in principal use is Amyl
Nitrate. One percent of this in fuel oil will increase Cetane number approxi¬
mately 10 points.

Determined as the time lag between injection and ignition of the fuel imder running
conditions. This is the CFR Diesel method. (CFR is from the title of the SAE Cooperative
Fuel Research Committed.) Alternately the lowest or critical compression ratio at which
fuel will ignite is used.
INTERNAL COMBUSTION ENGINE FUEL 121
Calorijio Value, Determination of the heats of cortibustion of petroleum oils
may be made with the bomb calorimeter (see Sec 5-7), results so obtained being
the higher heating values Qh. The net or lower heating value is a better indica¬
tion of the heat energy available to the engine, since exhaust gases are hot
enough to contain the H2O in the form of vapor. 1
The relative carbon and hydrogen content of oil is closely allied to its
density, for the impurities in commercial engine oil are an insignificant fraction
of the total. It follows that the density is the main variable in heating value, and
empirical formulae expressing Qh as a function of density are successful. The
following equations give Qh to a probable accuracy within 1% of actual value.
(ASME Formula) Qh = 41,130 + 139.6 x'^API J/g (5-5)
(Bureau of Standards Formula) Qh = 51,716—8793.8 (S.G.)^ J/g (5-6)
Table 5-1. ASTM DIESEL FUEL-OIL CLASSIFICATION
Grade of Fuel

1-D 2-D 3-D 4-D

Flash point ° C, min 87.8 48.8 4S.8 eo.o

or legal or legal or legal
Water & sediment, % by voL, max 0.05 0.05 0.1 0.5
Viscosity at 87.8*C
Centistokes, min 20
Equivalent Saybolt 32.6
Centistokes, max 60 12
Equivalent Saybolt. 45.5 65 140
Final boiling point, ®o , 810
90% Point ®0, max 857
Carbon residue, % by wt, max 0.5 2.5
Carbon residue on 10% residue
% by wt, max 0.25
Ash, % by wt, max O.Ol 0.01 0.02 0.05
Pour point ®c, max 0 -6.7* 1.7^ 1.7
Sulfur, % by wt, max 0.5 1.0 1.5 2.0
Cu strip corrosion at 60®O Pass Pafe«?
Cetane number, min 45 45 35 30
Alkali and mineral acid Neutral Neutral Neutral Neutral

♦Lower Pour Points may be .specified whenever reciuiied by local tempeiature condi¬
tions to facilitate storage and use

ASTM Diesel Fuel Recommendations

Graae Type of Engine

1-D Solid-injection engines operating at more than 1000 rpm.
3- D Solid-injection engines operating from 360-1000 rpm.
4- D Solid-injection engines with cylinders more than406 mm in (liainotc-i, and oper¬
ating at less than 240 ipm.
Air-injection engines operating at less than 400 rpm.
5- D Oil heavier than 4-D may be specified by referring to No. 5 Fuel Oil with an
agreement between buyer and seller regarding Carbon Residue and Sulfur
limits.
122 FUELS AND COMBUSTION
Fuel oils vary from 40705 to 46520 J per g ; the higher the density, the
lower the heat content. High-gravity fuel will have more heat units per liter,
although fewer per gram than light fuel. Lower heating value is obtained from
Qh as explained in Chapter 1. Latent heat of water vapor is taken as 2442 J
per g, and hydrogen content from the following equation:

Percent H2 == 26 — 15 (S.G.) (5-7)

Pour, Flash, and Color. This is a group of qualities which bear little relation¬
ship to the performance of fuel in an engine, but may be specified for some
other reason, such as handling.
The flash point is the temperature at which ignition of the fuel vapors rising
above the heated oil will occur when exposed to an open flame. Safety in han¬
dling and storing will usually be adequate if this is 65.6*0 or more.
Pour point is the minimum temperature at which the fuel will no longer
pour freely. It is more important to the fuel for mobile engines whose fuel
tanks are generally exposed and unheated than for stationary units.
Color is sometimes specified on the basis that good color indicates clean
fresh stock of Satisfactory volatility.
Purity. Although oil is a relatively pure fuel it may contain some sediment
and water, ash, or sulfur, all of which are undesirable impurities and are
limited to low permissible amounts in specifications. Sediment and water are
determined by diluting the fuel with benzol and centrifuging. They are reported
together, sometimes by the abbreviation B.S. and W., meaning bottom sediment
and water. Fuels have no B.S. and W. when initially manufactured, so this is a
test for the possibility of subsequent contamination. Ash is the incombustible
content of fuel oil. As it is quite abrasive and detrimental to the finely finished
surfaces of injection equipment, it must be limited to minute amounts by oil
specifications.
Sulfur to an excessive amount may produce corrosion and deposits in the
engine. All oils have some sulfur, but the allowable content is small, less than
2% in any case. Tests are sometimes carried out for the sulfur content and for
the corrosive effect on polished copper strips.
Manufacturers establish minimum fuel specifications for their engines,
based on their individual experience. In addition there are some general
standards. (See Table 5-1.) Operatois endeavor to purchase fuel supplies which
satisfy the minimum requirements without greatly exceeding them, for the
superior oils cost the most but do not outperform the poorer, provided the
latter meet appropriate minimum standards.
5-5 Furnace Fuel Oil. The characteristic qualities employed to measure
the value of furnace oil are mainly the same as those mentioned in Sec 5-4.
They are: heating value, specific gravity, viscosity, flash and pour points,
carbon residue, and impurities. Commercially, fuel oils for furnace use are
marked by a number index, specifications for which are to be seen in Table 5-2.
There it will be noted that six grades of fuel oil are designated. In general, the
higher the grade number the lower the cost of the oil per Joule of heat content.
Also, the higher the index number, the more diflicult it is to get proper atomiza¬
tion and smokeless combustion of the oil, more maintenance is required on
Table 5-2. DETAILED REQUIREMENTS FOR FUEL OILS*. (A voluntary commercial standard.)
124 FUELS AND COMBUSTION
burners, and the burning equipment may be additionally complicated by the
need for heaters to reduce the viscosity of the oil before spraying.
The common unit of measure for oil sales is volumetric, either the litre or
the ^^barrel,^^ which is 159 litres (42 U.S gallons). The pricing on such a
volumetric basis is commonly made at a 15.6®G volume, and the volume therefore
sold in any case is the actual volume delivered corrected by a coefficient of ex¬
pansion. The coefficient for fuel oil is about 0.0007; hence an oil that was
delivered into a tank of known volume, and gauged at 19900 litres in the tank at a
temperature of 4.4®C, is actually a commercial volume of 19900 + 19900 X
0.0007(15.6--4.4), or 20055 Utxes
The relation of the API and Baume scales to specific gravity is stated in
Eqs 5-2 and 5-3. The Bureau of Standards formula for Qh of all petroleum
products is Eq 5-6. Sherman and Knopff suggest the following for fuel oil:
Qh = 42450-93(Bc'+10) J per g at 15 6®G (5-8)
A^’here the prospect of oil as a fuel is considered, management frequently
inquires whether the necessary supplies of oil will be forthcoming in the future,
for it has been generally understood that the reserves of petroleum in the
ground are far less than those of coal. This is a question which cannot be ade¬
quately answered by statements like ‘'there is plenty of oil in sight for the next
five or ten years^^ for two good reasons. The first is that the equipment which
will be installed to burn oil rather than coal as a fuel is of a fairly permanent
character and might conceivably be expected to last for fifteen to twenty years,
so that changes of a fuel from liquid to solid could not be made without sub¬
stantial major alterations to the steam generating unit. Second, there is a
question as to the effect of abnormal future demands. Such questions are stimu¬
lated by the knowledge that previous wars have made increasingly large inroads
into the reserves of liquid fuels of the country, and by the instability, from the
labor standpoint, of the coal industry. Such considerations are the concern of
owners considering the types of fuel to use, and continuity of supply may be
more important than relative cost. The use of oil fuel is attended by several
tangible advantages which will sometimes outweigh the uncertainties.
5-6 Coal. Coal is a heterogeneous compound, difficult to define. While its
constituents are always carbon, hydrogen, oxygen, sulfur, nitrogen, and certain
mineral noncombustiblcs, not ordinarily analyzed but lumped together as
“ash,” the many ways in which these elements are compounded in coal as
found at different sites precludes any simple definition. In fact, coal is quite
a variable substance, and the classification of its different forms has not been
('a.>y. There are several classification systems which attempt to find some com¬
mon principle of variation among different kinds of coal. The inventor of a
practical system will necessarily have had to eliminate moisture and ash, for
the.^^c arc strictly of random occurrence in coal. The presence in coal, as found
in storage i)ilcs, or ready-to-fire, or as-received at the laboratory, of both mois¬
ture and ash gives rise to certain variations of analysis of the constituents.
These variations are:
1, As-received. As sampled from car deliveries, storage piles, stoker hoppers,
then tightly scaled for delivery to the analyzing laboratory.
2. Air-dried. When the as-received sample, partly ground or crushed, is laid
COAL 125
out in the laboratory room until it no longer loses moisture to the surrounding
atmosphere, its condition is said to be air-dried. In this form it contains some
surface moisture, but not much. It can be satisfactorily ground to fine consist¬
ency in sample grinders (which wet coal cannot) and will remain stable while
being weighed on analytical balances.
3. Moisture-free. This coal could actually be produced by holding the sam¬
ple at slightly above 100®C for a sufficient length of time to drive off all the
surface moisture. The moisture content of air-dried coal varies from time to
time, but the moisture-free basis is a duplicatable condition.
4. Moisture- and ash-free. This is a hypothetical substance. Actually one
is not able to separate ash from coal except by the process of combustion.
Moisture- and ash-free coal is generally called combustible, since it represents
the fraction of the original sample that can be gasified during proper combus¬
tion.
5. A variation of Item 4, known as moisture- and mineral-free, is occasion¬
ally used. Certain of the mineral content of coal, such as pyrites, will be gasi¬
fied during combustion. To some extent this mineral content may be eliminated
without destroying the coal.
The sulfur content of coal varies from as low as 0.5% to as much as 5%.
It adds a little heating value, but furnishes some undesirable characteristics.
The purchaser may well view with suspicion a coal whose analysis shows high
sulfur. It occurs as pyrites, as sulfates, and as organic sulfur compounds. Fre¬
quently high sulfur coal is priced low, but the operating engineer should shun
large contracts for it until a thorough trial has demonstrated the effect of the
coal on operation and maintenance costs, service availability, and capacity.
This is because the sulfur may be responsible for clinkering and slagging, cor¬
rosion, spontaneous combustion, and air pollution. In spite of these disagreeable
features, it sometimes happens that the attractive price of high sulfur coal more
than outweighs the difficulties.
Based upon the composition of the hypothetical ^^moisture- and ash-free
coal,^^ there have been several systems devised for the classification of the dif¬
ferent grades or “ranks.” The ASTM Standard Specification is shown in Table
5-3.
Two forms of coal analysis are used by combustion engineers. The first, the
proximate analysis, is a separation of the coal into moisture, volatile material,
fixed carbon, and ash. This analysis is readily made by thermal means with a
moderate amount of equipment and is or should be within the capacity of any
mechanical engineer to perform. The addition of an analysis of sulfur content,
and grindability index, is desirable in evaluating firing characteristics of the
coal.
The grindability index is an important characteristic in pulverized coal
plants, since it is a measure of the .relative ease or difficulty of pulverizing dif¬
ferent kinds of coal. One of the tests, the Hardgrove, uses a standard miniature
pulverizer into which is sent a definite amount of grinding energy. The weight
of the ground product that will pass a 75 micron sieve is employed to calculate
an index number called the Hardgrove Grindability (HG). Typical values of
this index are 50 to 60 for good bituminous coal. Lower values mean more diffi¬
cult grinding. This index is principally comparative, and the power and energy
 •If II
‘I’S
3 05
•§.!
li) ^ 3

o -g 3

§ 's pL <
o o ^ ^ ^ lili« =
I- s.^5 I
I
<o g s? s £ ■
“I'a.S I E
CLASSIFICATION OF COALS BY RANK-

Jill I .1 il £« S'
s S I s> ®.
tllll •«S aS'?S “q2 ^^3 -3 -2^ 'o 43

sisX
» C-"* 18 a S§ =
^ 3 a
o o k< V sfelsss
a fi ® s. E E ..i Z
5 o O 43 ^1-5

343 303 3O) fes

C3 «
33
^ C) 3. i s.i a§-i'
|S5 |_w5 s, Xs X
P<
3< » CD TJ i
.K 1^ 00 |.S
s §■ - 2 00
^ *-k' ^I w^hTt-s ^ 3 S I
p !».p >.pc-
& Cs^. Q^ EZrf. Q^ ts^w g.
rN * n CO GO M o 3
bba^a >.ri r^'w r'* S *o *S *S *o "o 3 ^ u t*
a Si «.g
OQ O
5 jD w
3 - ® a h'
" J s ■ § ■S
5 "■«•!: ®
3 O.S
? §“1 Sg
Table 5-3.

2 |52 e
3 c4 O
3-2 « IS
j T3 orj 3 c8
3 «"
5 |.S.S J
! g.S
5-25 a-s
.J3 JS ^ *3 if
5-s § s £
xnmm h5 W 11
i-H eq CO rH c^
; ® 5 J fO
’•11^^
:p-ii

126
COAL 127
requirements of actual pulverizers are not readily ascertained from the HG
index
Another important physical characteristic is the ash fusion temperature.
Other things being equal, the coal with highest ash fusion temperature would
be the most desirable, for it would give a stable, solid ash particle in the pres¬
ence of high furnace temperature. In some equipment, the fusion temperature
produces a limitation upon the capacity at which the equipment can be oper¬
ated. If a low ash fusion coal is contemplated at the time of designing the
plant, the proper provisions can be made for successfully using it. This is often
done on account of the attractive price usually offered by the producers of low-
fusion coal. However, to switch to low-fusion coal on a unit that has been
operating satisfactorily with high-fusion coal is usually inviting trouble from
slagging and clogging of the gas passages.
The proximate analysis is of little use in combustion calculations, which

T.iblc 5-4 ANALYSIS OF TYl'IC’AL AMERICAN FUELS*

A Coal as Reedvert

Proximate Analysis Ultimate Analysia—Mouture Included

Fixed Hydro¬ Nitro¬ Joule

Cartxm
Ash
gen
Carbon Oxygen per
gen
gram
1 Penna Anthraate 2 19 5 $7 86 24 5 90 0 57 2 70 86 37 0 91 3 55 32265
2 Ark Scju Anthracite 1 ^ 12 84 73 69 12 21 2 01 3 74 77 29 1 39 3 36 wfmm
3 Penna Seim Bituminous, Gearfield 2 95 21 29 66 92 8 84 1 35 4 74 78 51 1 19 5 37 32435
4 W Va Semi Bitununous, New River 2 89 2S 61 69 18 2 32 0 55 4 99 84 11 1 64 6 39 34524
5 W Va. Seim Bituminous, Pocahontas 3 10 17 91 75 26 3 73 0 55 4 50 84 02 1 17 6 03 34356
6 W Va Semi Bitummous, Pocahontas 2 18 16 36 73 02 8 44 0 69 4 20 80 70 1 44 4 53 32680
7 Penna Bitummous, Appalachian 2 44 28 44 60 68 8 44 1 32 5 07 76 91 1 31 6 95
8 Ala Bitummous 2 53 26 94 59 48 11 05 0 79 4 80 74 44 1 59 7 33 31000
9 HI Bitummous 13 31 33 62 41 34 11 73 3 75 5 19 59 07 0 95 19 31 24612
10 Ind Bitummous 9 69 38 59 41 04 10 68 4 79 5 39 62 36 1 28 15 50 26628
11 N Dakota Ligmte 36 18 29 77 25 35 mMbm 0 68 6 76 39 45 0 59 43 82 15633
12 Texas Brown Lignite 13 04 42 75 29 00 14 85 1 04 5 51 52 06 0 95 25 33 21835

B UltunaU Analysis of Oils

Carbon Hydrogen
Oxygen and
Sulphur
Degrees Baume Joule per
Nitrogen 15 6®C gram
1 Tex Crude 84 60 10 90 2 87 1 63 21 56 44279
2 Tex Fuel 83 26 12 41 3 83 0 50 21 25 45859
3 Penna Crude 84 90 13 7b 1 40 0 0 28 01 44823
4 Cal Fuel 81 52 11 51 6 42 0 55 14 93 43556

C Volumetric Analysis of Natural Gas

Joule per
CHa CsH4 CO Ha Oa Na COa
litre

1 Manon, Ind 93 57 0 15 0 60 1 20 3 42 0 30 37263

2 Pittsburgh, Pa 72 18 3 00 1 00 20 00 0 80 34282
3 Findlay, 0 93 35 0 35 0 41 1 64 0 39 3 41 0 25 37636
4 Monroe, La 78 80 9 50t 0 0 0 0 0 34 11 30 0 06 35772

D Volumetric Analysis of Manufactured and By Product Gas

1
Jouipper
CH4 CaH4 CO Oa Na COa

1 Average blast furnace gas 1

mumii 27 2
___ 58 12
litm

3726
2 Carburetted water gas IS 34 35 2 1 20495

HBH■■
3 Coke oven gas . 32 6 48 8 2 204951
4 Producer gu 22 11
IHH 59 6 $590

* Data mostly that of U S. Bureau of Mines. See Professional Paper 48, 1906 and Bulletin 85, 1914
fCi Hi.
128 FUELS AND COMBUSTION
are essentially chemical equations. A chemical, or ultimate, analysis, is re¬
quired for that kind of work. The quantitative separation of such a compound
as coal into its chemical elements requires not only a well-equipped chemical
laboratory, but expert chemists and chemical technology, and thus is not a test
ordinarily in the repertoire of the engineer. However, it is not always possible or
convenient to get such an analysis or to have it prepared on short notice. Studies
have been made of the possibility of constructing a synthetic ultimate analysis
from the proximate analysis. If this could be done, it would eliminate much
inconvenience in combustion calculations, particularly field or preliminary cal¬
culations, for the proximate analysis can usually be quickly obtained. There
have been some solutions to this problem published in the form of empirical
conversion equations. Those of F. C. Evans, summarized in Table 5-5, have
been found by the author to give excellent results. Nitrogen is not included in
this table, but one of the following empirical equations may be used to calculate
the percent of N in the combustible using V, the percent of volatile matter in
the combustible.
N = 0.07 y for anthracite (5-9)
N = 0.0210 — 0.012y for bituminous and lignite (5-10)
In order to eliminate the random elements, the transfer is made from the proxi¬
mate to the ultimate analysis of combustible; hence the first step is a reduction
of the analysis of coal to one of combustible, implying the ash- and moisture-
free basis. After the application of the transfer equations, the resulting ultimate
analysis of the combustible is then reconverted to the desired basis by the inser¬
tion of ash or ash and moisture. The following problem will illustrate the
transfer of a proximate to an ultimate analysis.
Example 1: Check the ultimate analysis of the Illinois bituminous coal given in
Table 5-4 by applying the conversion formulae to the proximate analysis.
To reduce the proximate analysis to a combustible basis divide by

(1 — ash — moisture)

Then the proximate analysis of the combustible is

Volatile matter, (k448; fixed carbon, 0.552

From Table 5-5, we find that the equations to use.for this case are:

II = 0.0457 + 0.02067 C = 0.953 ~ 0.3627

H - 0.0457 + 0.0206 X 0.448 = 0.0549 C = 0.953 - 0.362 X 0.448 * 0.7905
From Eq 5-10
N * 0.0210 - 0.012 X 0.448 « 0.0156
The ultimate analysis of combustible has become
C, 0.7905; H, 0.0549; N, 0.0156

The ultimate analysis as fired is found by including ash and moisture. Multiply the
above by (1 — ash — moisture), then compare this with the analysis from Table 5-4,
with moisture separate.
 Table 5-5 PROPERTIES OF COAL BASED ON PROXIMATE ANALYSIS
(F C Evans)
Cornell University Bulletin 3 (1925)

129
130 FUELS AND COMBUSTION
Converted: C, 59.3; Hs, 4.1; N2, 1.2; H20, 13.3; ash, 11.7; other, 10.4.
Analyzed: C, 59.1; H2,3.7; N2, 1.0; H2O, 13.3; ash, 11.7; other, 11.2.
The firing qualities of a coal are highly important when selection of the coal
or combustion equipment is being considered, for if it were not for the effects
of these qualities, coal could be purchased on the basis of heating value alone.
Some of the factors that must be considered are:
1. Sizes and grades for economical and satisfactory performance. Sizing practice
in the coal industry is to state the screen limits through which that sample of coal
will pass, and on which it will he retained. For example, a 60 mm x 80 mm size is that
size of coal which will pass through a 60 mm round hole opening or its equivalent, but be
retained on a SO ram round hole screen. Typical standard sizing practice and the use
of those sizes are given in Table 5-6.
2. Constituents. This includes such information as proximate and ultimate analyses,
ash analysis, and uniformity over a range.
3. Physical characteristics such as (a) friability—resistance to degradation; (b)
grindability, as measured by Hardgrove Grindability or some other standard; (c) size
consistency.
4. Storage characteristics. Deterioration, spontaneous combustion.
5. Performance characteristics—coking and caking tendency—the retention of
original shape during combustion vs softening or crusting.
6. Slagging characteristics—dependent on ash fusion temperature and iron content
7. Corrosive characteristics—the sulfur content.
8. Ignition characteristics—the presence‘of certai© minerals of low-ignition tempera¬
ture in the volatile.
The purchaser of small quantities of coal rarely pays any attention to
other than its cost per tonne, and, occasionally, if firing characteristics have forced
him to it, the sulfur and moisture content. The large-scale purchaser needs a
definite contract basis, for the cost of coal is a large item in total operating
expense. The contract between the purchaser and seller should set forth clearly
the coal to be furnished by stating its trade name, region mined, and size limits,
as well as the quantity. Ranges of acceptable physical and chemical qualities
are fixed and, of course, contract prices are given.
Table 5-G. COAL SIZING PRACTICE
Anthracite Bituminous
llOmrax 80 ram—Broken Run of mine —Rarely sold because of de¬
80 mm X 60 mm —^Egg mand for graded sizes.
60 mm x 40 mm —Stove 125 mm Lump —Hand firing
40 mm X 20 mm —Chestnut 125 mm x 60 mm Egg —^Hand firing
‘io mm X 14 ram —Pea 50 mm X 30 mm Nut —Industrial stokers
14 mm X 6 —No.l Buckwheat 30 mm X 20 mm Stoker—Domestic and industrial
8 mm X 6 ram—No.2 Buckwheat stokers
6 mm X 2.5 mm—No.3 Buckwheat 20 mm X 0 mm Slack —Industrial stokers.
Screenings —Culm Pulverizers
50 mm Nut & Slack—Industrial stokers

Typical form of a coal purchasing specification:

1. Specific designation of coal field or seam.
2. Size and range, such as “50 mm, nut and alack.’*^ Smallest size that will perf-
orm properly in furnace is usually economic.
FUEL TESTS FOR HEATING VALUE 131
3. Quality—^acceptable ranges of analysis of (1) moisture as loaded, (2) volatile
matter, (3) fixed carbon, (4) ash, (6) sulfur.
4. Firing qualities.
Ash fusion temperature—^minimum acceptable
Higher heating value—^minimum acceptable.
6. Contract cost. Per delivered tonne. Per delivered million Joules of higher heating
value. Bonus and penalty clauses.
5-7 Fuel Tests for Heating Value. Most modern central power stations
have a laboratory in which routine tests are made, including tests of the fuel.
The same is occasionally found in industry, where some technical department
of the industry makes fuel tests. This is also sometimes done by central public
authority in the case of institutional plants, but by and large, outside of the
public utility field, the check-ups on deliveries under coal contracts are quite
irregular.

Fig. 5-2 Oxygen bomb calorimeter.

Heating value tests are performed by the use of one of the standard fuel
calorimeters. These are likely to go much better for solid and gaseous fuels than
for liquid fuels, ^^hlch art' inherently difficult to test accurately for heating
value. On the other hand, fuel oil is a much more homogeneous substance than
coal, and good estimates of heating value can be obtained from hydrometer
readings.
The apparatus for heating value testing falls into two general categories:
(1) continuous flow and (2) individual sample or “batch” apparatus.
Solid fuels are tested with individual sample apparatus, gaseous fuels by
continuous flow apparatus, and liquid fuels by either type, depending upon the
volatility.
Most types of calorimeters measure the heat released by absorbing it in
water. If the maximum temperature of the absorbing medium is low enough
to cool the products of combustion below lOO^C, the latent heat of vaporiza¬
tion in the products will be released. Since this is always the case, it is the
higher heating value, rather than the lower, that is obtained by calorimetry.
132 FUELS AND COMBUSTION
Coal Calorimeter, The heating value of coal is determined either by the
peroxide or oxygen bomb calorimeter. Use of the latter type is here described.
The apparatus, illustrated in Fig. 5-2, employs gaseous oxygen under pressure
as the oxidant, and electric ignition. The bomb is construct^ of two pieces held
together with a screwed collar. The bomb has a fuel tray, oxygen valve, two
external electrical leads, and internal poles to which fuse wire can be connected.
A polished container holds the bomb and about».26kgof water. This, in turn,
k ..urrounded by a heat insulating jacket. A stirring device is provided so that
the water bath will be homogeneous in temperature.
About Ig of finely ground coal is selected for testing. The sample must
be briquetted with a measured amount of fuse wire, put in the fuel tray, and
weighed before being assembled into the bomb. It has been found advisable to
briquet the sample and embed the fuse wire to (1) insure ignition and (2) pre¬
vent loss of fuel from the pan before ignition. It is customary to put a few drops
of water in the bottom of the bomb when charging it. This is done because

Finol obttrvofions
3to5 min.

PO ploctcd sothot
Oi «02 Thtn PQ»aT
TIME AXIS

Fig. 5-3 Interpretation of temperature rise in the bomb calorimeter.

otherwise part of the latent heat of vaporization of the products would be held
as a humidity in the atmosphere of the bomb. The bomb is charged with oxygen
to a pressure of about 14 kg/om*. The electrical connections are made, then the
bomb is put in its container along with a weighed quantity of water. The water
used for the bath should be about 3® C below room temperature, since it may
be expected to rise about 6® C when the bomb is fired. Adjusting the water
bath thus minimizes the effect of radiation. External electrical leads are at¬
tached, and the sample is ignited. Meanwhile, the- water bath is being electri¬
cally stirred. Combustion is presumed to be almost instantaneous; however, the
thermal capacity of the outfit prevents any rapid temperature rise upon ignition.
Fig. 5-3 shows about what rate of temperature change may be expected.
The heat released by the ensuing combustion is absorbed by the water, by
the bomb, and by the container via a small rise in temperature. It is convenient
to express the heat absorbing capacity of the bomb and container per degree
temperature rise as that of an equivalent weight of water which would absorb
the same heat with the sarne temperature rise. This value is called the “water
equivalent” of the bomb and is added to the water weight in the heating value
calculations. It is necessary that the heating value of the fuse wire be known,
for the heat so released must be deducted from the gross heat initially calcu¬
lated. The apparent heat release is:

Q — w'c6t, (5-11)
FUEL TESTS FOR HEATING VALUE 133
in which ly' = Weight of the water bath + water equivalent of the outfit.
c « Specific heat of water, which may be taken as 1.
At = Observed temperature rise of the bath.
After adding or deducting any necessary corrections, Q is divided by the
initial weight of the coal sample to get the higher heating value.
It is important that the temperature rise- be rationally determined. Before
the bomb is ignited, the temperature rise is gradual. The temperature is plotted
agaitist time, resulting in a curve similar to that shown in Fig. 5-3. The ob¬
served temperature approaching the firing point, and following the combustion,
will usually plot as a straight line against time, however, the temperature rise

Fig. 5-4 Gas calorimeter.

following ignition may be expected to bo nonlinear. After the straight parts of

the curve are extended, and a line PQ is drawn vertically so as to intersect
equal areas Ui and 02, a height measurement PQ is produced. It can be theorized
that this indicates the temperature rise At more accurately than can be done
by any disorganized observations of the temperature. The first part of the
curve slopes slightly upward due to energ>^ being imparted by the stirring de¬
vice. If £he last portion of the record slopes downward appreciably, it is an
indication that the calorimeter is poorly insulated.
Gas Calorimeter, The apparatus shown in Fig. 6-4 is a Junkers continuous
flow gas calorimeter. Fuel is burned in a steady flame, with air as the oxidant.
To test gas in the continuous flow calorimeter, a special gas burner was devised
for perfect combustion. In this calorimeter the heat released is absorbed in water
as the latter circulates in tubes surrounding the « umbustion chamber. A con-
134 FUELS AND COMBUSTION
stant rate of water flow through the calorimeter at any given valve setting is
obtained by supplying the water under constant head. This is obtained by ad¬
mitting an excess of water to a small supply chamber having an overflow.
The quantity of gas burned is indicated by a “wet type^^ gas meter, which
shows the pressure and temperature of the gas as well as the volume consumed.
The customary way of expressing the heating value of a gas is Joules per litre
under standard conditions of temperature and pressure. These conditions have
been designated by the ASTM asl5.6®Oand762nuDHg. The volume measured by
the gas meter can be reduced to standard conditions by the use of gas laws.
The calorimeter is equipped with thermometers in wells for the measure¬
ment of temperature rise of tlie water and temperature of the outgoing products
of combustion. A special valve allows the water leaving the calorimeter to be
discharged at will into a receptacle or wasted. The heat absorbed by the water
during test is:
Q = wcM (5-12)
in which w = Weight of water caught during the test.
c = Specific heat of water, which may be taken as 1.
At = Temperature rise of the water passing through the calorimeter.

If the air used for combustion is not saturated, a part of the heat released
will be carried off in the latent heat of the vapor required for saturation of
the products. The correction for this is small Although it is tlieoretically cal¬
culable, one usually employs prepared correction tables.
In this calorimeter the products are cooled nearly to room temperature.
Ordinarily, no correction for the residual heat in the dry or wet products of
combustion is required.
The heating value of the fuel is found as follows:

Qh Q/vk + corrections (if any) (5-13)

Where Vg is the volume of fuel flow which produced Q Joule®, corrected to stan¬
dard conditions.
The lower heating value is obtained by deducting the latent heat of the
moisture in the products For this jiurpose the water of condensation must be
caught and measured.
5-8 Proximate Analysis. This analysis is made by heating the coal until
it decomposes successively into three of the four complex items of the proximate
analysis. Since the fourth is found by difference, the only feasible check on test
results is to analyze duplicate samples and compare results. The laboratory
must be provided with a sample grinder, a good muffle furnace with temperature
control and indicating pyrometer, a thermostatically regulated drying oven,
crucibles, desiccator jar, and analytical balances. The following description is
founded on the standard ASTM procedure, modified for speed in routine check
work. Six clean sample crucibles are used, four of shallow and two of deep shape.
Covers are used in some cases. All of these items are to be empty-weighed to
at least the nearest milligram.
Moisture. Two samples are weighed and placed in the shallow crucibles in
COMBUSTION 135
a 104,4®C oven for one hour, then removed to a desiccator for cooling and re¬
weighing. The desiccator is a closed vessel containing a dehumidifying stib-
stance which will keep the test samples from absorbing moisture from the at¬
mosphere as they cool. Samples must be cool, otherwise rising thermal currents
would disturb the analytical balances.
To check effectiveness of the drying the samples are reheated for 10 min,
cooled, and reweighed. This is continued until constant weight is indicated for
two consecutive heatings. The percent of moisture contained is the loss of weight
of the sample divided by the net weight of the original sample in the crucible.
Volatile Matter. Two samples are placed in deep crucibles with lids and
weighed. These are inserted in a 955oven for exactly 7 min, then removed
and cooled. The loss of weight of the sample in this instance is due both to the
moisture and the volatile which were driven off. Moisture having been previ¬
ously determined, volatile is found by difference.
Ash. The remaining two samples are heated in a 760® C oven for 2 hr,
then removed, cooled in a desiccator, and reweighed. The process should be
repeated, as with the moisture samples, until constant weight is obtained. The
percent ash content is the weight of the sample remaining divided by the net
original weight.
Finally, all other items having been analytically determined, the fixed car¬
bon is considered to be the difference between 100% and the sum of all the
percentages evaluated for the other three constituents.
5-9 Ash and Refuse Analyses. The complete ultimate analysis of coal
ash is a major job in the chemistry of quantitative analysis, which would not
be expected of a power plant engineer. Indeed, it so seldom is needed that the
information is obtainable only on special order. Ash fusion temperature is,
however, often needed. This test is made by preparing small triangular pyra¬
mids of ground ash mixed with binder, then heating a group of them slowly in
a furnace until a central specimen of the group softens and deforms in a manner
prescribed in the standard ASTM test. A special furnace is necessary, since the
heating is to be carried out in a reducing atmosphere.
Analysis of the refuse of a coal fire is needed after a heat balance test in
order to determine the combustible content, if any. No more laboratory equip¬
ment is needed than for the proximate analysis. Samples of ground refuse are
given the same processing as described heretofore for moisture and ash in the
proximate analysis of coal. The combustible content of the refuse is generally
assumed to have been carbon, although this is probably not always exactly
true. The refuse analysis is used to determine the “dry refuse per kg coal
bumed.^^ In the event that the refuse of a test is also weighed, the two methods
serve as a check on one another. Sometimes the refuse is wet down or sprinkled
with water before removal from the ash pit. Since the ash sample is conse¬
quently damp, it should be kept tightly scaled until deliverd to the laboratory.
5-10 Combustion. As was recorded in the beginning of this book, the
reader is assumed to have done preliminary reading in the fundamentals of
heat power. Consequently, the chemistry of oxidation, which we call combus’-
tion, is not repeated here, although the basic equations of combustion will be
found summarized in Fig. A-4 of the Appendix.
Combustion, which was somewhat elaborately defined in a previous section,
136 FUELS AND COMBUSTION
occurs when an atom of fuel and one or more atoms of oxygen are joined through
the medium of their electrons. After combustion, these molecules possess the
same number of electrons as the combining elements had originally, but the
electronic configuration is different—^its energy level is lower. This energy
originally got away from the molecule as electromagnetic radiation, but quickly
changed into the high velocity activity of surrounding molecules on which it
impinges. It is then heat. Before the combination can take place, the atoms of
fuel and oxygen must be well mixed together, and be in a highly agitated state
which, externally, is the ignition temperature. This is a simplification of the
real nature of combustion. Further insight on this phenomenon is furnished by
Sec 9-3.
To obtain good, efiicient combustion of a fuel, the following are required:
1. Thorough mixing of fuel and air in proportions which will insure com¬
plete combustion.
2. Exposure of fuel particles to oxygen throughout a period of time suf¬
ficient for their combustion.
3. Combustion zone maintained above the ignition temperature of the fuel.
The atmospheric combustion of coal is of a more complex nature than for
fuel oil; therefore this section will relate mainly to the combustion of coal.
The combustion of liquid fuel in the pressurized cylinders of engines is de¬
scribed in Sec 6-3.
When a coal is subjected to combustion conditions, it first absorbs the heat
necessary to cause volatilization of the hydrocarbons. This includes both sen¬
sible heat and latent heat necessary to vaporize any moisture in the coal. After
the distillation of the volatile hydrocarbons is complete, the remainder is coke,
a mixture of fixed carbon and ash. It is the hydrocarbons which must be most
carefully handled to obtain freedom from smoke and incomplete combustion.
Hence a major function of any furnace and stoker, grate, or burner installation
is the proper mixing of air with this distilled volatile matter, and holding it at
the ignition temperature sufficiently long to obtain complete combustion.
The fixed carbon burns in a somewhat complicated way, the nature of
which may be expressed most simply by saying that the manner of carbon
combustion is an incomplete oxidation to carbon monoxide, which itself is a
fuel. The carbon monoxide will then be reduced to carbon dioxide, provided
enough oxygen is mixed with it, and it remains hot enough to continue the com¬
bustion. During combustion each small piece of incandescent carbon becomes
blanketed with either carbon dioxide or carbon monoxide, and unless this gas
blanket can be continuously scrubbed off so as to expose a fresh surface the
combustion will stop. Continuous combustion is accomplished in stokers and on
grates by moving the air past a stationary fuel bed at high velocity induced
by draft pressure. In pulverized coal burners, where the streams of fuel and air
are moving into the furnace at something like the same velocity, this particle
cleaning action must be done by gaseous turbulence created by design of the
burner, jet impingements, etc.
When combustion of coal is complete, the gaseous products aie CO2, N2, O2
from excess air, and SO2. Water vapor is also present, originating from three
sources: (1) moisture originally contained in the coal; (2) vapor produced by
combustion of hydrogen; (3) the water vapor of atmospheric humidity. Cases
COMBUSTION 137
of incomplete combustion will, in addition, exhibit some CO, possibly also
some hydrocarbon gases.
The quantity of vapor from items (1) and (2) is nine times the total hydro¬
gen content, by weight, of the fuel analysis. This moisture can absorb both
sensible and latent heat. The vapor from (3) carries sensible heat only, since it
is already vaporized in the atmosphere. To find the quantity of atmospheric
moisture in a 1kg. of combustion air, one uses psychrometric readings and a
chart, or the following equation:

kg vapor per kg dry air (5-14)

in which ht^ *= Latent heat of vapor at wet bulb temperature(J/g)

M = Depression of wet bulb below dry bulb temperature (*0
= 0.6?5p,ftt/(l.03 — p«at) vapor required to saturate 1 kg
of dry air at wet bulb temperature.
Psat = Saturation pressure at wet bulb temperature (kg/cm*)
The air required for combustion may be considered as being composed of two
parts: that which is used to supply oxygen suflScient to unite with all the fuel
atoms; and that which is supplied in excess, so that there will be a certainty
that an oxygen atom will be adjacent to a fuel atom when required. Since at¬
mospheric air is only 20.9% oxygen by volume, the remaining. 79.1%* is inert
and merely absorbs heat, causing a loss of thermal efficiency. A cubic metre of
oxygen burns to a cubic metre of carbondioxidc; hence the theoretical maximum
carbon dioxide in the products of combustion is 20.9, say 21%, by volume. This
is if the combustion is perfect, if there is no excess air, and if the fuel is pure
carbon. In actual combustion the percent of carbon dioxide is considerably
lower; 14% CO2 is considered to represent good practice.
Combustion calculations of the steam plant engineer are fairly elementary
since temperatures do not rise high enough to cause much chemical dissociation
to exist. Although it is possible, as many operators regretfully admit, to have
situations where incomplete combustion with smoke and carbon monoxide are
never completely eliminated, the power plant engineer's design computations
are generally predicated on the assumption that he will be able to achieve the
proper technological conditions and obtain a fuel burning installation that wilh
produce complete combustion. His study will include estimates of flue gas
quantities and volumes, temperatures, densities, heating values, heat transfers,
etc. In studies of gas temperatures, reliable specific heat data are, of course,
indispensable. Fig. 6-6 shows the mean specific heat at constant pressure for
the products commonly resulting from the combustion of commercial fuel. These
are mean specific heats over the range from 15.6*0 to the abscissa temperature.
As a subsequent example will show, this chart can be made to furnish the
proper data when the initial temperature is higher than the datum of the chart.
Combustion calculations involve weights and volumes. Coal analyses, of

* Chiefly nitrogen, but with small, generally negligible quantities of argon, carbon
dioxide, uid other gases.
138 FUELS AND COMBUSTION
course, are gravimetric. So, also, are those of liquid fuels, but gas analyses are
customarily volumetric. The gaseous products of combustion are analyzable
by test, for which the portable Orsat apparatus is commonly used. Where a
gaseous state is involved, either as a product of combustion, or in the fuel itself,
the calculations may be made either on volumetric or gravimetric bases, or in
combination. Volumetric calculations may be written in cubic cm units; also
in mols, Mols are especially useful in the case of gaseous fuels, for the equa¬
tions of chemical reaction are also mol equations. On the other hand, gravi¬
metric calculations may always be checked by the law of Conservation of
Mass, whereas there is no law of conservation of volume, and mol equations
may take this unarithmetical form: 1 mol + % mol = 1 mol. This would never
happen in gravimetric calculations. Either of these forms of gas calculations
will, if correctly handled, give the same result, and the individual can feel at

2400

2350

2300

[2250

2200*i
\
j
[2150 :

2100
’o
2050

2000

1950

Fig. 6-5 Mean specific heat of gases between ld.a*0 and final temperature,
keal/Kg- *0 (at constant pressure).
COMBUSTION 139
liberty to employ his personal preference. Actually there is not much difference
in the two methods. Reviewing the basic facts of mols and mol volumes, we
have:
1 A mol is- weight of the substance, gtn
* molecular weight of the substance
2. A mol volume is 23.693 Iitreatatl5.6^G and 1.03kg/cm* a6. This is
independent of the kind of gas.
0 0845 T
3. At T® R and p kg/cm* ab^ the mol volume is - litres
V
Several standard equations are put forth for use in combustion calculations,
and many others may be invented to suit the need by using two simple princi¬
ples, namely;
1. Analysis of equations for dimensional homogeneity.
2. Principle of continuity of mass in steady flow.
As an example of equation invention, suppose an equation were desired, in¬
terpreting the results of an Orsat analysis of the products of combustion of a
certain coal, in terms of the air-fuel ratio employed during combustion. First,
note that the Orsat is a volumetric analysis and, therefore, likewise a molal
analysis. Next, the idea of continuity of carbon mass is introduced by relying
on the equivalence of all the carbon in the flue gas to the carbon actually burned
in the furnace. Also, with another continuity idea, that of equating the nitrogen
in the products to that coming from the air* (knowing that the dimensional
unit of the A:F ratio is ^^air per kg coaPO we can set up an equation involving
carbon and nitrogen continuity.

mols N2 ^ kgN2 ^ kgC

kg air _ mol gas mol N2 kg coal
kg fuel mol C ^ kgC ^ kgN2
mol gas mol C kgair

This equation will be found to be dimensionally homogeneous. In it the

“gas” is the dry gaseous product of combustion and the kg C per kg coal is the
carbon actually burned, which excludes any combustible in refuse or dust.
Some of the terms being constants, the equation can be simplified for work¬
ing purposes;

A:F Na X 28 X C' ^ T on pT (5-15)

(CO2 + CO) X 12 X 0.768 Leo, -1- CO.
in which N,, CO,, CO = Orsat analysis.

C' = Carbon burned per kg fuel.

For an equation of weight of dry gas formed per kg coal burned, note that:
. 1000 UttM
* This is true enough in the case of coal, although not for a fuel like by-product or manu¬
factured gas, which itself contains considerable nitrogen.
140 FUELS AND COMBUSTION
itggas ^ tgC
itg gas _ mol gas kg coal
kg coal kgC ^ inol C
mol C mol gas

= 44 CO, + 28 CO + 32 0, + 28 N,
mol gas
kg gas ^ (44 CO2 + 28 CO + 32 O2 + 28 N2)C^
kg coal 12(C02 + CO)
This reduces to;
kg gas ^ rU CO2 + 8 O2 + 7(C0 + NslC^
(5-16)
kg coal 3(C02 + CO)

The symbols have the same meanings as for Eq 5-15.

The foregoing equation does not include SO2. When S in the coal is not
negligible, the SO2 formed should be considered. In the Orsat equipment SO2
is absorbed in the CO2 chamber; hence the Orsat CO2 is really CO2 + SO2.
When this is given the proper consideration,* Eq 5-16 becomes:

kgBM llCO, + 8(), + 7(N, + CO)/'_, , 3„\ , 5„

-3(Cft + CO) +
in which C' and S are decimal fractions.
5-11 Illustrative Examples in Combustion. The conditions existing in
a case of actual combustion, the air supplied, and other quantities are obtain¬
able from calculations starting with an ultimate analysis of the fuel and an
Orsat analysis of the products. One method is shown in the following illustra¬
tive example.

Example 1: The various products of combustion, the A:F ratio, and the excess
air will be established, mainly from flue gas and fuel analyses. All computations will
be maintained on a weight basis in the first solution.

Data: Coal, as-fired: C, 77; Hg, 5; O2, 6; S, 1; Ash, 11.

Refuse: Ash, 80; combustible, 20.
Orsat: COg, 11.90; CO, 0.36; Og, 7.13; N2, 80.61.

C' of Eq 5-15 is found by applying the principle of continuity to ash flow, assuming
here no fly-ash.
Ash * 0.11 of the coal « 0.80 of the refuse.

Refuse per kg coal = C in refuse *= — 0.11 « 0.0275 kg

C' * 0.77 - 0.0275 « 0.7425 kg

Next transfer the volumetric analysis of products to a gravimetric basis by Avogadro’s
Law.
♦P. M. Ross, “Effect of Sulphur on Flue Gas Analysis,” The Engineer, July 14, 1944.
ILLUSTRATIVE EXAMPLES IN COMBUSTION 141

Item CO, CO 0, N, Total

1. Orsat (volumetric) 11.90 0.36 7.13 80.61 100.00

2. Molecular weights 44 28 32 28
3. Product, (1) X (2) 523 10.1 228 2257 3018
(31
4. Gravimetric analysis, .1740 .0033 .0756 .7471 1.000
5. Divisor for C content 3.67* 2.33*
6. C in products 0.0474 0.0014 0.0488
7. Products of combustion
15.24t X (4) 2.65 0.05 1.15 11.39 15.24

♦See Fig. A-4, Appendix,

t See following calculation.

kgjM ^ kgc burned jES.ga^ « 07425 X ^ 15.24

kg coal kg coal kg C burned ‘ ^ 0.0488
Hence dry gas produced per kg coal burned = 15.24kg.*
Air supplied per lb coal burned == 11.39t/.768 = 14.86 kg.

Theoretical air needed perkgcoal = 11.5 X .77 + 34.5 ^0.05 — + 4.3 X 0.01
= 10.32 kg. (11.5, 34.5, and 4.3 are constants from Fig. A-4.)
Excess air = 14.85 — 10.32 = 4.53 kg(43.8% excess).
Alternate solution: Were the air supplied and dry gas produced items of the fore¬
going calculations our only objective, a shorter, quicker solution is possible.

From Eq 5-15, AV = 3.03 X 0.7425 ) = 14.85 kg.

\ 12.20 /
Dry gas perkgcoal = 14.85 + 1 ~ wet gas — SO2 — refuse
- 15.85 - 9 X 0.05 - 2 X 0.01 - (0.11 -f 0.0275)
= 15.24 kg

Example 2: Exact and approximate calculations of the heat absorbed and carried
off by the products of combustion found in Ex 1 will be made. Assume 14W*C preheated
air temperature and 899^ discharge temperature. Atmospheric humidity is defined
by psychrometer dry and wet bulb temperatures of 15.6 and l(W .
Approximate method. The water vapor content of the atmosphere is small enough
to be neglected in approximate calculations.
In Ex 1 it is seen that there are 15,24 kg dry gas per kg coal. The ASME Code
permits a composite Cp of 1005 J per kg .per deg C to be used.
The water vapor from combustion is 9 X 0.05 = .45 kg •
Average Cp for water vapor is taken as 8094 Jper kg per 0.
Sensible heat above liW in products:

(15.24 X1005+ 0.45X8094)(899 - 149) = 4065 k Jper kg coal

♦ Here SOi is neglected.
t If fuel contains appreciable Ns, subtract at this point.
142 FUELS AND COMBUSTION
Latent heat present in products:
Qfg = 0.45 X hfg J per kg coal

The specific value of hfg to employ is uncertain. If conditions warrant, a value can
be calculated by thermodynamic methods, employing Dalton^s Law. Usually some value
between hfg at 15,6*0 and at 100® O is assumed. The ASME boiler test code uses hfg at
d5.6^C .

Exact method. All sulfur is assumed to be in SO2, with a Cp of 0.2 (exact value not
important); atmospheric moisture being calculated from Eq 5-14 or picked from
psychrometric chart. Fig A-12, Appendix.

= 0.625 X 0.0V(1.08-0.01) « 0.00766 kg

(0,00766 X 1065,6 x" 2.326)^5,6
(1066,6-2.SS6)+(1.8 x5.6)
= 0.0054 kg

The manner of use of Fig. 5-5 for average Cp when initial temperature exceeds 15.6®C
is as follows. Subscripts a and h refer to initial and final states.
Mean Cp =
— Ca

For the case of nitrogen, with U = 148H1 4 = WW

1.072(899-15.6)^1.06(149-15.6)..
Mean Cp «=-899^149-^ ^ P®' ^
In a similar fashion the mean Cp^s for the other gases are determined and entered in
the following tabular calculations.

Item CO, CO 0, N, H2O SO, Total

1

Weight 2.65 1.15 11.39 0.02

Mean Cp 0.252 0.238 0.258 0.2
w X Cp 0.668 BQil 0.274 •2.940 0.215 0.004 4.114

Sensible heat above 140*C in products:

Qb = 17.21(899-149)«4303 kJ per kg coal
Latent heat same as in approximate solution.

Cases sometimes arise where the engineer needs an estimate of the products
of combustion and their composite physical characteristics at some definite
temperature. The next two illustrative examples will show how these character¬
istics may be computed for solid and for gaseous fuels.

Example 3: The as-fired ultimate analysis of a certain coal that will be burned
with an estimated 50% excess air is given as follows:

C, 0.8411; H2,0.0499; O2, 0.0639; Nj, 0.0164; S, 0.0055; Ash, 0.0232

The probable results of combustion will be calculated, assuming complete combus¬

tion and final temperature. Combustion constants are used from Appendix,
Fig. A-4.
ILLUSTRATIVE EXAMPLES IN COMBUSTION 143
(n
0.0499 — —

+ 2.0 X 0.0056 « 5.04 kg

Gaseous products of combustion are CO2, HiO, N2, O2, SO2.

CO2 « 1.7 X 0.8411 « 1.48 kg
K2O =4.1X 0.0499 = 0.20 kg
N2 = 6.0 X 150% X 0.768 + 0.016 = 5.84 kg
O2 * 5.0 X 50% X 0.232 = 0.69 kg
SO2 » 0.9 X 0.0055 = 0.01 kg

8.07 kg total

Mols of products * ^ ^ ^ ^ *0.272 xi0»molsperkgcoal

At 1.03 kg/om* and 335^0 the molal volume is 49.88 litm

Volvme and density. The products of one kg of the coal will have a volume of
0.272 X 10* X 49.88 = 13837 litres. The density will be 8.07/13.887 =0.683 kg/m»
Dry gas analysis. The dry gas will consist of 0.0703 mol CO2, 0.0403 mol O2, and
0.458 mol N2, a total of 0.5686 mol. By dividing individual mols by the total, a volu¬
metric analysis is produced. (Probable Orsat analysis.)

CO2, 12.4%; O2, 7.1%; N2, 80.5%

Example 4: The products of combustion of the Pittsburgh natural gas, Table

6-4, when it is completely burned with 50% excess air are next cakulated; also a
prediction of the flue gas analysis; and the flue gas density at 2944*C. Neglect moisture
in air.
The mol equations are:

0.7218 CH4 + 2 X 0.7218 O2 = 0.7218 CO2 + 2 X 0.7218 H2O

0.0300 C2H4 + 3 X 0.0300 O2 = 2 X 0.0300 CO2 + 2 X 0.0300 H2O
0.0100 CO -f i X 0.0100 O2 = 0.0100 CO2 -h 0 H2O
0.2000 H2 + i X 0.2000 02 = 0 CO2 + 2000 H2O

0.9618 mol gas -t- 1.639 mol O2 = 0.792 mol CO2 + 1.704 mol H2O
The O2 required from the air = 1.639 — 0.008 = 1.631 mol.
Since the composition of air is 0.791 mol N2 to 0.209 mol O2, each mol O2 supplied
791
brings with it — = 3.78 mols N2. We may now write the products of combustion of

one mol of the natural gas.

The wet gas is composed of 1.704 mols H20 plus the dry gas.
The dry gas is composed of:
CO2: 0.792 + 0.008 = 0.80 mol
N2: 3.78 X 1.631 * 6.16 mol from theoretical air
N2: J X 6.16 = 3.08 mol from excess air
O2: J X 1.631 * 0.82 mol from excess air

Total 10.86 mol dry gas

The dry flue gas analysis is obtained by dividing each of the above by 10.86. Then flue
gas analyBis is:
CO,, 7.4%; 0,,7.6%; N,, 85.1%
144 FUELS AND COMBUSTION
At S04«4^ the molal volume has become M.14 litres^ And total volume of dry flue gas per
mol gas burned » x — 426 litres.
The weight of wet flue gas per 10.86 mols of dry gas is:
CO2: 0.80 X 44 » 35.2 gm
O2: 0.82 X 32 = 26.2
N2: 9.24 X 28 = 258.7
H20:1.704 X 18 = 30.7
Total 350.8 gm
In this case, the volume of the 30.7 gmof steam is obtained by converting the weight
to mols and multiplying by the molal volume at 204.4^0, i.e., by 30.14 1.
30 7
Volume of the flue gas =“ 426 + X 39.14*to401
lo
350 8
Density of the flue gas at 204.4^0 and 1.03 kg/om* * » 0.712 gm/1

In both this and the previous illustrative examples, no attempt is made to

apply Dalton’s Law of Partial Pressures to the gas-water vapor mixture. In
most combustion problems the vapor in the products is so highly superheated
that it can be treated like a gas.
5-12 Approximations and Short-cuts. There will be times when the more
exact combustion calculations can be dispensed with in favor of less accurate
but more rapid methods of approximation. Sometimes this arises in connection
with quick check-ups at the plant, with roughing out preliminary work, or
simply because the conditions do not warrant more precise calculations. Some
of these short-cuts will now be mentioned.
It has been found that the air theoretically required for combustion of or¬
dinary coals is approximately a kg for each 746 kilooalorief of higher heating
value. The composite molecular weight of the products of combustion of coal
fires is always about 30. This is useful in quick estimates of volume of flue
gas, etc.
If combustion is complete, the percent of excess air can be determined from
the carbon dioxide content of the flue gas and the fuel ratio of the coal, C:H.
20.9 g R + 2.37
Excess Air = (5-18)
C02{R + 3) JK + 3
in which CO2 = percent by volume in the dry flue gas.

R = fuel ratio C'/H. Anthracite 50-25; Semi-anthracite 25-20;

Semi-bituminous 20-16; Bituminous 16-12; Lignite 12-9.
Example 1: By means of approximations, estimate percent of excess air indicated
when the Alabama bituminous coal given in Table 5-4 is burned to a 12% CO2 analysis
by volume. The refuse contains 15% combustible.
In Eq 5-18, R, the fuel ratio, is CVH, wherein C' is the weight of carbon actually
burned, and H is the weight of hydrogen that is burned by oxygen from the air. If
some of the oxygen in the fuel is needed to burn the sulfur, a correction is readily
applied as follows:

H = Total hydrogen -
8
SMOKE 145
Therefore,
C' 0.7444 -(|XO.U05)' 0.07249

H = 0.048 - O Q P97-? „ 0.0398; whence R = 18.2

O

20.9 X 18.2 18.2 + 2.37

Excess air = 52.9%
12(18.2 4- 3) 18.2 + 3

5-13 Smoke. The discharge of smoke to the atmosphere is an unnecessary

evil. The effects of smoke may be considered under three headings, viz.: (1)
effect on health; (2) financial loss due to incomplete combustion, deleterious
effect on plant growth, and begriming of buildings; and (3) effect on standard
of living.
The smoke nuisance is at its worst, of course, in metropolitan districts. The
smoke is produced from both industrial and domestic fires. Smoke abatement
workers have found it much easier to render the former class smokeless than
the latter. For instance, through building permit control, through smoke in¬
spectors, and through technical cooperation with industrial plant owners, it
has been found that industrial smoke can be controlled. Often this has meant
major changes in boiler settings or in coal-burning equipment. But it can usu¬
ally be shown that bad smokers are wasting enough monqy by their incomplete
combustion to pay for the smokeless equipment. Domestic smoke is much harder
to control, owing to the larger number of individual installations and to the
layman^s ignorance of combustion theory rudiments. Publicity campaigns and
individual cooperation and inspection, and the enlistment of large numbers of
citizens in smoke abatement leagues, have been used to solve this phase of the
problem.
There has been no definite coordination discovered between diseases of the
respiratory tract and smoke density; however, common sense would indicate
that a smoky atmosphere was bound to be less healthful than smoke-free air.
It is, however, a well-known fact that smoke has a deadly effect on vegetation,
principally because of the sulfur products it carries. Smoke corrodes metals,
darkens paints, and in many other ways creates a tremendous economic loss
besides that due to loss of heating value of fuel. Those who have lived in smoky
cities know that a lower standard of cleanliness is prevalent. Neither building
interiors nor exteriors, clothing, hangings, furniture, etc., can be kept clean.
One of the oldest methods of classifying the density of smoke emitted by a
chimney involves use of a color comparison chart.
The Ringelman Chart is composed of four sets of gratings, black on white.
When these are placed about 15*24iiifrom the observer the gratings merge to a
solid color ranging from light gray to a dense black. The chart enables a smoke
inspector to rate the smoke emitted from a chimney on an arbitrary numerical
scale. Smoke ordinances are often based on the Ringelman rating. A typical
smoke and dust ordinance will contain:

/ 0.1105
♦ Alternately C* = 0.7444 - 0.1105 0.7249 as in Ex 1> Sec 5-11.
\ 0.85
146 FUELS AND COMBUSTION
1. A prohibition against emitting a smoke the shade or appearance of which is equal
to or darker than No. 2 Ringelman except
a. Equal to No. 2 may be permitted for periods aggregating 4 min in any 30
min, and
b. Equal to No. 3 may be permitted for periods aggregating 3 min in any 15 min
when building new fire, having maintenance difficulties, etc., not reasonably
preventable.
2. A prohibition against emitting dust in gases to exceed 6.85 kg per 1000 kg gases,
adjusted to 50% excess air, except that dust collector efficiencies of greater than
85% will not be required.

Furnaces burning coal in suspension (pulverized coal burners and spreader

stokers) will always throw out dust in the form of fly-ash, and collectors will
have to be installed in the breeching, if there are any dust emission regulations
to meet. However, no reasonably priced equipment can take smoke out of the
products of combustion. The proper way is to burn the fuel in the furnace
completely.

PROBLEMS
1. (a) Six hundred twenty three cubic metres of a fuel gas are passed through
a meter at 0.35 kg/om> • Barometer, inmHg.Find the commercial sales volume
of this gas.(b) Tank contains 214m*of fuel oil atJL7®C.S.G. = 0.945. Find the volume
and weight of this quantity of oil measured at 15.6°/15.6**
2. (a) At a natural gas metering station the temperature is steady at2a9*C during
a period in which 1281 m3 are registered. Pressi're in meter averages 0.84 kg/cm^ ga.
Barometcr,775inin Hg. Find the commercial sales volume of this gas. (b) How mu(;h
volume ( m3) will 3785 liters fuel oil of 18° Be at 15.6*/15.6* occupy when heated to
37.8*C ?
3. During a gas calorimeter test there was a 7.62 ® o rise of water temperature.
While 56,63 litres gas registered on the meter, 6.78 kg water were collected. Meter
pressure,203 mmHg; temperature,27.8°C. Find Qi,, corrected to 762 mm and 15.6°C.
4. Given this proximate analysis data sheet, get the complete proximate analysis,
air-dried basis, to nearest 0,1%.

4m-DRiED Sample

Moisture and
Test Moisture Volatile Ash

Crucible weight, g 16.025 17.176 15.821

Cruciblf Initial 17.116 18.155 16.862

+
coal Final 17.094 17.842 15.936

5. The following are the results of a proximate analysis test run on air-dried coa^.
All weights in grams. Calculate the proximate analysis of dry coal and of as-fired coal,
PROBLEMS 147
Empty Crucible and coal
Item crucible Initial Final
Moisture test 13.001 14.686 14.610
Moisture and volatile test 30.275 33.603 32.522
Ash test 11.255 12.800 11.378
(as-fired moisture test) 14.420 15.928 15.906
6. Testing a certain West Virginia Pocahontas coal by bomb calorimeter yielded
these results. AT, 3.49® C. Fuel sample, 1.0535 g. Fuse wire consumed, 0.0073 g. Water
charged, 1855.68 g. Water equivalent of bomb, 470 g. Test coal had 0.96% moisture.
Calculate Qu for dry coal. {Q of fuse wire, 1600 Cal per g.)
7. A cylindrical oil tank2.4indiaineterX6 mlongis filled to the neck with fuel oil
which is checked at 21* Be at $1* C. Estimate the kilo JouIm heating value stored in
this tank.
8. Fuel oil of 16* Be is stored in a tank to supply a boiler whose rated max. steam¬
ing rate is 6604 kg per hr at 150 psig. Feedwater, O. Average thermal efficiency,
70%. Consider that plant’s capacity factor is 45% and that- 1% months’ supply is
carried. Make any necessary assumptions and calculate the number of 1.8 m diaX4.6 m
long cylindrical tanks needed to hold this quantity.
9. A horizontal cylindrical fuel oil storage tankM m dia. x a mlong is gauged at 825.5
mm oil depth. A delivery of oil is made which raises the level to 2165 mm, measured after the
temperature of the. contents had attained ground temperature, i.e., 7.2 ®C. This oil is
purchased for $3.31perbblatl5.6*/15,6*. The delivery is billed at $380.50. Is this correct
to the nearest dollar?
10. Reduce the analysis of a coal designated from Table 5-4 to (a) moisture-free
basis (b) moisture- and ash-free basis.
11. Check the ultimate analysis and heating value of a coal designated from Table
5-4 by applying conversion formulae from Table 5-5. Also estimate the air required
for combustion at 50% excess.
12. Given a coal with proximate analysis as follows, find heating value and kgaii
required per kg coal at 50% excess. Moisture, 4.47; volatile, 22.74; fixed carbon,
53.72; ash, 19.07.
13. Predict the flue gas analysis to be obtained when a coal of the following proxi¬
mate analysis is completely burned with 85% excess air.
Moisture, 1.75; volatile, 21.70; fixed carbon, 69.13; ash, 7.42.
14. The as-fired proximate analysis is taken of a coal of West Virginia origin.
Moisture, 1.75; volatile, 21.70; fixed carbon, 69.13; ash, 7.42. When burned with what
is presumed to be sufficient air for complete combustion, the CO2 tests 10% by volume.
What percent excess air is indicated?
15. Apply the data of Table 5-3 to a coal designated from Table 5-4 and fix its
rank thereby.
16. Predict the Orsat analysis resulting from the combustion of a coal designated
from Table 5-4, with 50% excess air. 10% combustible in the refuse.
17. A coal is analyzed in a plant laboratory and found to have Qi, = 14^866 kJ ;
C, 58.93; VM, 31.81; ash, 9.26. All on a dry basis. Find the rank of this coal. Estimate
the A-F ratio used when a CO2 meter shows 12%. Use approximate methods.
18. A Kansas coal, reported on the combustible basis, has 5.7% S, 79.6% C. As-
fired, the moisture is 11%, ash 14%. Orsat of combustion gases is: CO2, 12.1 ; O2, 7.1;
CO, 0.8; N2, 80; carbon in refuse, 10%. Find dry gas per kg coal burned (a) neglecting
SO2 and (b) including 8O2.
19. From the analysis of a coal sample taken during a test it is estimated that the
as-fired total carbon was 76.5% and free hydrogen was 4.7%. From other tests C =
0.7263, Orsat: CO2, 10.53; Og, 3.74; CO, 0.49; Ng, 85.24. Find A:F ratio.
148 FUELS AND COMBUSTION
20. Boiler plant test data, in part:
Proximate analysis as fired: FC, 60.65; VM, 33.45; Moist., 0.83; ash, 5.07. Combusti¬
ble in dry refuse, 29.8%. Orsat: CO2,11.3; O2, 6.75; CO, 0.2. Find the air-fuel ratio and
the percent excess air.
21. Further analysis of the coal mentioned in Prob. 20 reveals a sulfur content of
3.2%. Find the kg dry gas produced per kg coal during this test.
22. Using the test data mentioned in Prob. 20, change proximate to ultimate analysis
and calculate the weights of the individual products formed by the combustion of a
kg of coal.
23. A coal designated from Table 5-4 is burned with 60% excess air. Atm. humidity,
5 g per kg. No incomplete combustion. Combustion air at 176.7®C. Flue gas at
441 '"C . (a) Estimate sensible heat loss by approx, method; (b) calculate sensible heat
loss by exact method.
24. Humid air which enters a combustion chamber at 27.2®C is transformed into
products of combustion at Psychrometer readings are27.2®Candtefi.2f®C'-'How many
kJ are carried off per kg coal burned by the superheating of this moisture? A:F =
15.5.
25. A coal as designated from Table 5-4 is burned in a pulverized coal furnace
with *20% excess air. Preheated air temperature, ]48.0**0; furnace outlet temperature,
1010® C No incomplete combustion. Calculate the kJ absorbed by the gases leaving
the furnace, per kg coal.
26. Calculate the volume ( m* ) of the products of combustion of a coal designated
from Table 5-4 if at 260°C aher complete combustion with 100% excess air.
27. What is the volume ( m* ) at 176.7®C of the products of combustion of a cubic
metro of fuel gas at I.07 kg/cm* ab, 15.6®C, whose volumetric analysis is C2H(j, 14.5; CH4,
59.1 ;C02, 25.1; N2, 1.3. 60% excess air.
28. A coal of following composition is burned to an Orsat gas analysis of 0% CO2.
Refuse analysis, 15% combustible. Flue gas at282.2r®C. Find the density of this flue gas.
C, 76.82; H2, 4.96; O2, 6.38; N2, 1.46; S, 1.39; ash, 6.26; moisture, 2.73.
29. WHte mol equations of combustion of a gas designated from Table 5-4, when
burned with 25% excess air. Then calculate the specific volume of the products, m*-
perkg, at 204® 0.
30. To what temperature, above an initial 16.6®0, would the products of combustion
of the California fuel oil. Table 5-4, be raised in a furnace where 45% of the sensible
heat developed is immediately absorbed radiantly. 45% excess air, complete combus¬
tion. Specific heat from Fig. 5-5.
 CHAPTER 6

INTERNAL COMBUSTION

6-1 Internal Combustion Engines. Power can be generated by heat en¬

gines wherein the products of combustion form the working fluid of the thermal
cycle. The influence of these internal combustion engines on life in a technically
advanced country such as the United States is profound. They are largely re¬
sponsible for the success of the self-propelled vehicle and have their major ap¬
plication in that field. Small I.C. engines power mobile equipment and stationary
units where electrical service is not convenient. Although, increasingly, station¬
ary power needs are being met by electrical motors, an important exception is
where the mechanical power is employed to drive generators originating the
supply of electrical power. It is to this use of the I.C. engine that the present
chapter is addressed. This application has the following variations:
1. Portable generating units which may be moved from site to site where
electrical power is required temporarily.
2. Standby units, normally idle, which can be activated when there is a
failure of central station power where an interruption would mean financial loss
or danger. (Tunnel lighting, operating rooms, key industrial processes, etc.)
3. Engine-generator units installed in power plants where they are the
normal primary source of electrical power generated for public, industrial, or
institutional consumption.
Internal combustion engines are built around a variety of thermodynamic
cycles. One type may vary from another in characteristics such as flexibility,
fuel used, ease of starting, weight, cost, etc., but all have many features in com¬
mon such as:
1. Use of the piston and cylinder to create a variable volume chamber in
which to work out the cycle.
2. Gaseous working medium.
3. Open thermal cycles, meaning a flow of air as well as fuel into the engine
and a discharge of gaseous products of combustion from it.
4. Two- or four-stroke mechanical cycles.
6. Cyclic rather than uniform power output, necessitating multiple cylinders
and/or heavy flywheels for steady delivery.
The principles and comparative features of the nine commercial variations
149
 150 INTERNAL COMBUSTION ENGINE POWER PLANT
used nowadays are summarized in Fig. 6-1. Engines using gaseous fuels are less
frequently seen than the liquid fuel types, as liquids are more readily trans¬
ported. The two most-used liquid fuel engines are the gasoline engine and the
Diesel engine. Since the Diesel is the more important to the field of electric
generation, this chapter is mainly devoted to it. However, when local conditions
justify, engineers are not hesitant to employ the other types, for all are suc¬
cessful commercial developments. For example, an aluminum producer, to
whom electricity is a principal manufacturing expense, located a factory in a

FUEL \ HIGH - PRESSURE

INTRODUCED BY/ CARBURETOR MIXING VALVE L-P INJECTION INJECTION

Hiia OUTSIDE CYLINDER.BEFORE COMPRESSION INSIDE CYLINDER .EARLY IN COMPRESSION

1 COMPRESSION LOW LOW 1

liGNITION METHOD SPARK SPARK 1

OASOLINI ENGINC GAS ENGINE INHCTION OAS ENGINE INJECTION GASOLINE

1 Caribvrvtvr vapoclMi f«Ml t AUainp valva Mandt air I Air antarc cylindar thraagh 1 Svetlan ttraka. with inlat
mm6 iiiiact it wMi air la and gat faal bi prapar bilat parti In cylindar wall valva apan filli cylindar
^mpmr prarartlwia pr apart lant 3 aiitan clatat Inlat and an with air
7 ItKflM itrali*, wllli inl«t 7 Swcfian ttraka wlik intaka hawtt ttarft fa caaipratt 7 Inlat valva clatat cam
valva epan filU iyllarfar valva apaa flllt lyllndar 3 Gat fual it injactad at law prattian haglnt fual an*
witli ailatwra with ittialwra prattura aarly in catnprat tart vndar high prattwrc
OUTLINE OF 9 Caaiprattlan itroka raltat
miatara prattwra ta J-
3 Cantprattlaa ttraka roltat
miHtwra prattwra taV<9>
tian ttraka
4 latulting miitwra it cam
3 fual and air min at cam
prattian ttraka raliat prat
ENGINE’S •S t ta/****" prattad taf^l tura taT-fd
4 Spark ignitat miatura aaar 4 Spark ignitat mlatwra naar panding an faal 4 Spark ignitat miftvra naar
CYCLE OF aa4 at <ampra**ian ttroka and af catnprettian ttraka 5 Spark ignitat mintara naar and aT camprattian ttraka
EVENTS 9 Tha flra4 miafwri capotidt
patklap plitan dawn
5 Ttia firad mtatura aapandt and af camprattian firad 5 Tha fired miatura anpandt
pwthing pittan Pawn miatara aapandt puthing pittan dawn
4 lakaatf valva apant rU 4 lahawtt valva apant rit A Piftan mavat dawn an 4 lahautt valva apant rIt
lap piftan claar* cylindar tag plitan ilaart cylindar warking ttraka wncav ing pittan ciaort cylindar
af Iwmad pai af btiraad gat art aahautt partt naar and af burnad gat

Fki. (3-1 Principles of commercial I C. engines. (Courtesy Power.)

gas field territory and built a power supply plant in which were installed more
than a hundred 1150-kw engine-generator units driven by 2-cycle, outside-mix¬
ing gas engines.
As employed in stationary power plants, engines are direct-connected to
generators of relatively low speed. The speeds employed (2(K)-1000 rpm) neces¬
sitate a large-diameter, salient-pole generator of rather short axial length. The
heavy, rugged slow-speed construction is productive of high reliability and, al¬
though costly, can be justified in central station practice.
In a patent dated 1892, Dr. Rudolf Diesel, a German engineer, described an
engine to operate on the Carnot cycle. Coal dust was the fuel, and it was to be
fed slowly enough so that isothermal expansion would result. After fuel cutoff,
an adiabatic expansion would continue, followed by a compression made iso-
INTERNAL CX)MBUBTION ENGINES 161
thermal by the injection of water into the cylinder. An adiabatic compression
then brought the cycle back to its beginning. A further claim of the patent cov¬
ered the use of liquid fuels and the spray valve. Early attempts to build this
engine* resulted in the adoption of a modified cycle which, after much experi¬
mentation, was built into a successful working engine. Since then the Diesel has
slowly but surely established itself as a practical prime mover of high efficiency
which can use a comparatively cheap fuel.
Engine installations may be divided into mobile and stationary. The class

HIGH - PRESSURE INJECTION

MISSflMAN INOINI VAMRIZING Oil INCINC CAS-DIISIl INOINI DUAl.nia INOINI

t liKflon tiroli*. wirii Intof 1. lactlaa atraha, with bilat 1. iactlaa ftraha, arMi lalat I Air aatara cyllaAar t. lalat aalaa apaaay aaa*
valvt fill* cjrlinAar walaa mptm, fill* aylhMiar aalva aytaa, HNa tyllaAar thraafh iaiat yarta riaa atraha fWa cyNadar
with air wHh air with air 3. yiatati alaaaa mlai aaA with air and faa
3 Compraitian «tr«li« rwiMt 3 Ia|a<ta4 f«al aa^arlaa*, 3. CaaifrataiaaatraliaralBaa aahavat caaaaraaaaa air 3 CamyraaaianatrafcaraltaB
tof-4KNAl# ailaat with haraa4 vaa, fraaaara f ahaalJTI^J- «a ahaai praaaara af adalara fa
3 P«mI It (wSlpIct. 4aaaa'l bam 9. Vaal lafactlaa afarta at ar 9 Naar and vl aamyraaalaa. ahaat
9 Cam^raaaiaa atraha faal faa and yila* all ara 9. ttaar and af caaiyraaalaa.
itroha mnd ••49UUH faraat air Mta iniaatad yilat a4 la iniactad fa hd*
4. Saarhf Im*! hafara cam. aa^iariiar 4. Naat af cantyraaaian If. tiata aaiwhaatlaw
^aitiaa aa4>, lanitat 4 Maafa4 aarlaaa af va^« laaaati hy aaatyraaaiaa, nitaa faal; yllat oil ata* 4 Naat af caaapraaalaa If*
fwal «a^riia4 hy com. laar l9altaa miafara aaar Ifaitaa faal hiliaaa camhaatian nitaa pllat all, which
praiflaa mmd ttararf haat •amarataiaa aa4 •. laralaf artataraaaiMada, 9. lamlnf mlniaraaKyaada. caaaaa mlatara to haaa
I. Ca^aadan |Mthai ^itaa 5 Caaihaatiaa la aaplaaiaa, faahiaf yialaa 4awa aa yaiMnf ylatan dawn 9 Raaalthif aayoaalaapaHh*
4awa ^aahaa piataa 4awa “warhiaf” atraha 4. Naar and af atraha. pit* aa yitfaa dawn
4 Cahawtt «al«a a[»an», rl>> 4 lahaait valaa afaway ria* 4. lahaaat valva ayaaai ria> tan ancavara aahaaat 4. lahaaat valva apaaay rlt>
laf |rt**an «laar« cyllnAar af flataa claara ayWaAar hif fiataa alaara cylhiAar 1af plalaa alaara cyttadar

of stationary services with which we are here concerned used to be the principal
outlet of Diesel engine manufacturing, but acceptance of this prime mover in
the railway, truck, and motor bus fields now accounts for a large share of the
output of engine builders.
The Diesel is an excellent prime mover for electrical generation in capacities
of from iOl hp to 5070 hp. As such, it is widely used by private industry, hotels,
utility companies, and municipalities.
The advantages of the Diesel engine are:
1. Low fuel cost.
2. No long wanhing-up period.
3. No standby losses.
* Difficulties in isothermal control, in addition to bulkiness of a Carnot engine caused
by its small mean effective pressure (thin cycle).
162 INTERNAL COMBUSTION ENGINE POWER PLANT
4. Uniformly high efficiency of all sizes.
5 Simple plant layout
6 Needs no large water supply
The Diesel can extract more work out of each heat unit than other engines.
For that reason it becomes an attractive prime mover wherever first cost is
written off slowly enough so that operating costs are influential Central sta¬
tion service is of that type On the other hand, the Diesel is not the answer to
all power supply problems Unlike the steam turbine, an increase of capacity
increases the plant floor area and cost nearly in the same proportion This fixes

Powe^

Fig 6-2 Elements of the stationary Diesel engine It can be seen that engine parts
break down mto several groups (1) structural parts—^bedplate, frame, liners, heads,
(2) major moving parts—piston, connecting rods, crankshaft, and their respective
bearings, (3) arrangements for getting air m and exhaust out—valves, valve mechanism,
manifolds, scavenging and supercharging systems, and (4) fuel-mjection system—
pumps, nozzles, control devices
DIESEL ENGINE POWER 153
an economic limit to the size of Diesel central stations. Also, the steam plant is
not necessarily inefficient compared to the Diesel. Large steam central stations
have net thermal efficiencies nearly as high as that of the Diesel. Another objec¬
tion to the Diesel is the exhaust noise. Proper exhaust silencers can minimize
this fault.
Where fuel prices or reliability of supply favor oil over coal, where water
supply is limited, where loads are relatively small, and where electric line
service is unavailable, or is available at too high rates, there one will frequently
find informed engineers recommending Diesel engine power.
6-2 Diesel Engine Power. The ideal thermal cycle of the Diesel engine
is illustrated in Fig. 6-3. Beginning with the working medium at state 1, it is
first polytropically compressed to state 2, then heat is added during a limited
isobaric expansion, after which a polytropic expansion to the initial volume
reduces the pressure to state 4. The ideal work produced by the cycle is repre¬
sented by its area, and the mean effective pressure is its average height.
Polytropic processes 1-2 and 3-4 in the ideal cycle are isentropics with air
as the fluid. Thus, for the air standard performance, n = 7 = 1.4
The ideal air standard efficiency,

(6-1)

Ideal mean effective pressure,*

yry^HR - 1) - - D"
pmi = ViT (6-2)
(r - 1)(7 - 1)
These equations show that high efficiency is promoted by high r and low R,
but that engine size (which may be considered proportional to p^i) is increased
as R decreases. The requirement of adequate fuel combustion imposes a limi-

Fig. 6-3 Ideal Diesel cycle.

* 83nnbo]8 as described in Fig. 6-3.
154 INTERNAL COMBUSTION ENGINE POWER PLANT
tation on maximum R that can be used. Since Diesels are load-governed by
varying point of cutoff, the ideal efiSciency increases at part load. Tliis partially
offsets other tendencies, and a fairly flat eflSciency-load characteristic results.
Real engines have cylinder cooling for mechanical reasons; also they work
on an open cycle, meaning that the fluid at state 1 is not the same as that which
completed the previous cycle. The products of combustion of each working
cycle are discharged as exhaust gas, and fresh air is inducted for use in the fol¬
lowing cycle. Instead of heat being transferred between states 2 and 3, fuel is
injected into the air and its heat of combustion provides the energy input. The
real engine will have nonisentropic compression and expansion processes; n =
1.35 is a fair average in practice. Although Eq 6-1 has-its uses in Diesel engine
studies, the actual thermal efficiencies are considerably less than those of the
air standard.

Example 1: An air standard Diesel cycle will be analyzed for state of the working
fluid and performance.Using the nomenclature of Fig. 6-3, state 1 is at 0*9 kg/emSab and
27* C. The volume quantity is that of a single-cylinder engine with 25 cm bore and
38 cm stroke. It will be assumed that, after a compression sufficient to produce 638*C ,
heat is added during the first 10% of the working stroke. Equations of polytropic
processes in Appendix Fig. A-17 are used wheic necessary without further reference,

1
Ratio of compression, r = Vi/th = = (8n/300)^/.4 s= 12.0
pi/Pi “ = 12^*^ = 32.4; — 29.2 kg/cm* ab
The piston displacement {vx — 1^2) = 12 5*it x 38/10 s0.01865 m*
Clearance volume, V2 - \vx — V2)/{r — 1) = 0.01865/11-Ob. 0.00167 m*
Vi == 0.01865 + 0.00167=0.02032 m*

Vz - V2-\- 10%(f;i — V2) = 0.00167 f 0-00187=0.00354 m*

T3/T2 = Vz/V2 = 0.00354/0.00167*=2.12; I’a = 1719*K

Pa/Pz = (*^3/1^4)“’' =(0.00354/0.02032)1**=0.0869; Pa »* 2.54 kg/cm* ab
Ta/Tz = {vz/vaP’~'^ = (0.00354/U.02032)-4--0 498; Ta =

Principal physical states of the cycle are summarized;

State 1 2 s 4
Pressure, kg/cm* ab 0.9 29.2 29.2 2.54
Volume, m» 0.02032 0,00167 0.(K)3r>4 0.02032
Temperature, ®K 300 811 1719 856
Temperature, °C 27 5is 1446 583
From Eq 6-2,

Pmi ^ 0.9- X 12.0 X

r 1.4x 12.0*(2.12-U- (2.121**- 1) 1
••5.83 kg/om*
{12.0~1)(1.4--1)

Net work done per cycle = 5.83; X10* X 0.01865 = 1087 kg-m

1 r
Ideal thermal efflciency. i?i = 1
1.4x12-* L 2.12-11 J A K£t
0.56

* As a check, investigate the isometric process 4-1. pjpx should equal 7\/Tx\ 2.54/0.9
=* 856/300 *2
DIESEL ENGINE POWER 166
The outstanding characteristic of the Diesel engine is compression ignition.
It is possible to raise the air sufficiently in temperature, by polytropic compres¬
sion, to ignite hydrocarbon fuels. Control of ignition timing is obtained by in¬
jecting fuel into the air after the compression. Atomization rather than carbure-
tion is employed for dispersal, and inexpensive low-volatile liquid fuel, such as
petroleum distillate, can be used in place of the gasoline of spark ignition
engines. Timing of combustion is accomplished by timing of fuel injection. The
extent of compression required for ignition may be understood from Fig. 6-5.
Here the volumetric compression ratio is shown plotted against the pressure
and the temperature at the end of compression, assuming n = 1.35 rather than

Nordberg Mfg. Co.

Fig. 6-4 Municipal power plant two-cycle Diesel engine, 3600 hp

the ideal 1 4. The effect of initial p and t is considerable. For example, a super¬
charge of about 0.2 kg/cm* ab over the naturally aspirated suction pressure of
about 1 kg/cm*ab raises the pressiiie at state 2 by 8 kg/cm* when the compression
ratio IS 14. Even more significant it. the initial temperature. At r = 14, adecreaseof
suction temperature from 15° to 18"'Glowers final compression temperature from
452 to 369*^0 The graph a Iso will explain wny relatively high compression ratios
are inbuilt by Diesel manufacturers. Assume an initial air condition of 15® C.
Then, for the typical ignition data given, a compression ratio of 11 7 would be
needed (intersection of ignition and compression temperature curves). This
would allow no margin over variables such as lower inlet temperature or air
leakage. An engine with r = 11.7, using this fuel, would not have reliable igni¬
tion. Some margin of assurance of ignition is necessary. If this were set by the
designer as 115®, then a compression ratio of about 15 is indicated, for at this
 156 INTERNAL COMBUSTION ENGINE POWER PLANT
ratio the compression temperature is
115® above the fuel ignition tempera¬
ture. It is hoped that the foregoing
discussion will have pointed up the
problem of compression ignition and
disclosed the variables involved.
The mechanical heart of the Diesel
is the fuel injection system. The en¬
gine can perform no better than its
fuel injection system. A very small
TEMPERATURE'DEGREES

quantity of fuel must be measured

out, injected, atomized, and mixed
with combustion air. The mixing
problem becomes more difficult the
larger the cylinder and the faster the
rotative speed. Fortunately the high¬
speed engines are the small-bore
automotive types; however, special
combustion arrangements such as pre¬
combustion chambers, air cells, etc.,
arc necessary to secure good mixing.
Engines driving electrical generators
have lower speeds and simple com¬
bustion chambers.
Atomization of fuel oil has been

Fiu. 6-6 Two type's of common-rail injection. Left. A single pump supplies high-
pressure fuel to header; a relief valve holds pressure constant. The control wedge ad¬
justs lift of mechanically operated valve to set amount and time of injection. Right,
Controlled-pressure system has pump which maintains set header pressure. Pressure-
relief and timing valves regulate injection time and amount. Spring-loaded spray valve
acts merely as check. (Courtesy Power.)
DIESEL ENGINE POWER 167
secured by (1) air blast and (2) pressure spray. Early Roekf arm
Diesel engines used air for fuel injection at about
70 kg/cni*.This is suflBicient not only to inject the oil,
but also to atomize it for a rapid and thorough com¬
bustion. The expense of providing an air compressor
and tank led to the development of ‘‘solid” injection,
using a liquid pressure of between 105and 210kg/cm^,
which is suflSciently high to atomize the oil it force§
through spray nozzles.
Comshoft
Great advances have been made in the field of
solid injection of the fuel through research and prog¬
ress in fuel pump, spray nozzle, and combustion Fig. 6-7 Unit injector.
chamber design. This has stimulated a trend to solid Mechanically actuated
pump plunger raises fuel
injection, and comparatively few new installations
to a high pressure, me¬
are foimd with air injection. ters the quantity and
The fuel injection system must pressurize, meter, controls injection tim¬
and time the fuel. These fimctions are achieved in a ing. Nozzle contains a
variety of ways, but the common rail, the individual spring-loaded delivery
pump, or the unit injector systems are usual on power valve actuated by
station Diesels. Although these can properly be classi¬ change in the fuel-oil
fied as internal characteristics, their great importance pressure. (Courtesy
to successful operation justifies reviewing them as is Power,)
done in accompanying illustrations.
Engine types in commercial use may be classified as follows:
1. Cycle, It is possible to build the
2-cycle engine without valve gear. The
absence of this feature is indeed the vir¬
tue of the 2-cycle principle. In the 4-cycle
engine the exhaust and inlet valves are
mechanically operated from a camshaft.
Since the Diesel engine is commonly rather
large, the valves are correspondingly large
in diameter and are operated from a mas¬
sive cam-shaft. The first successful Diesels
operated on the 4-cycle principle; then
2-cycle engines made their appearance. For
variable speed operation the 2-cycle engine
lacks tht flexibility of 4rcycle, but, for
electrical generation which is a constant-
speed service, it is quite successful. Its
capacity is not twice that of the 4-cycle
Fig. 6-8 Pump-injection ' system.
engine of the same dimensions because of
An individual pump or pump cylin¬
der connects directly to each fuel its lower volumetric efiiciency, but there is
nozzle. Pump meters charge and not much difference in thermal efficiency.
control injection timing. Nozzles 2. Number and arrangement of cylin-
contain a delivery valve actuated ders. It is customary to design a line of
by fuel-oil pressure. (Courtesy engines on the basis oi a fixed bore and
Power.) stroke, and to vary capacity by adding
158 INTERNAI. COMBUSTION ENGINE POWER PLANT
cylinders. In-line arrangement is common, but increasingly the V-bank ar¬
rangement is seen.
3. Single^ or double-acting. The double-acting principle was adopted to ob¬
tain larger capacity than the single-acting without much increase in over-all
dimensions. Although chiefly used in marine service, there have been some
stationary applications. Most Diesels are single-acting.
4. Governing. Constant stroke with variable by-pass or suction, or variable
stroke. Governing is accomplished by control of the quantity of fuel oil spray.
The governor speed control is derived from centrifugal flyweights whose action
is to vary the metering function of the injection system. The load may be
divided at will between units. Governors are sensitive enough to permit exact
load adjustment. For electrical generation even the multicylinder engines
must be equipped with a heavy flywheel to prevent cyclic variation of speed.
5. Frames, (a) Horizontal or vertical; (b) A-frame or box-frame. A few
Diesels, chiefly single-cylinder engines, have horizontal cylinders, but general
practice prescribes a vertical or Vee engine. The box-type frame is standard.
Of all the questions which arise in purchasing an engine, the one least
amenable to precise analysis is the selection of the engine type. Some answers
are furnished by local requirements or prior usage, but any man with an under¬
standing of Diesel engines can And good and bad arguments for any type of
engine. The greatest assurance of success lies not in the type of an engine, but
in the details of design, materials, and workmanship, as proved by the engine’s
record, by the manufacturer’s reputation, or both. An engine should therefore
be as conservatively rated in speed and mean effective pressure as the purchaser
can afford to use. Crowding a high horsepower rating into a given engine lowers
the cost per horsepower and shortens the life of the engine. Every step in the
direction of simple design, if proved successful, is an advantage for the pur¬
chaser.
6-3 Combustion. In the previous chapter the nature of fluid fuels was
examined and the principles of combustion were presented. Here the special
features of combustion as carried out in the Diesel engine cylinder are to
receive attention. However, first it appears desirable to repeat and summarize
the equations pertaining to fuel oil.

Specific density (at 15.0'*C)8.33722 X (S.G. @ 15-6716-6®) kg/I (6-3)

140
Density scales: B4 » r- 130 (64)
S.G. @16.6yi5.0«
141.5
•API - 131.5 (M)
S.G. @15.«yi6.6®
Ignitbn quality: Diesel index — o.ois x ®API x «apt H- 0-32 x °AP1 (<WI)

Heating value: On - 4|1S0 + 139.6 x '‘API J/g (6-7)

also Qh - 51716 - 8793.8 (S.G.)« J/g

Hydrogen content: Ht - 26 - 15 (S.G.) percent by wt (6-0)

Combustion in the Diesel engine cylinder begins theoretically at the instant

injection starts and continues, at constant pressure, until injection ceases. The
COMBUSTION 169
distillate fuel used may be considered to have an average chemical formula of
CiaHsz* for which the ideal air quantity is found as follows:
ClJIm + 24 Oi = 16 CO* + 16 H*0
Considering the numerical prefixes to be mols, the equation of combining weights
is written as follows:
224 kg Ci*Hm + 24 X 32 kg O* = 16 X 44 kg CO* + 16 X 18 kg H*0
Since 1 kg air provides 0.232 kg 0*,

A‘rP»kf - o.S2x*L " ‘A®**'

As was explained in Sec 6-2, early fuel cutoff is necessary to good thermal
efficiency, but early cutoff is not possible with the ideal A:F ratio of 14.8. This
is due to the need for limiting maximum temperature of the cycle for mechanical
and thermal reasons, under circumstances as set forth in the following example.

j, / Dtpth of jtt

Fig. 6-9 Schematic analysis of the disintegration of a fuel jet. (Courtesy Trans, SAE,)

Example 1: The ideal maximum temperature of combustion of a fuel of 24* API

is calculated, on the assumption of 427°C compression temperature and 14.8 kg air per
kg fuel.
141 *5
Fvel calculations: S.G. = = 11«®1
131.5 + 24
Eq 6-8: Qh = 61716 - 8793.8 x 0.91* » 44434 J/g
Ql = 44484 - 2442.7 x 9 (.26 - 0.15 x 0.91) = 41719
During isobaric combustion, using Cp » i*oo, the sensible heat, Qht will raise the products
At degrees, according to the relation
Q *= taCpAf
41719
At * 264(re
(14-8 *4- 1) X I
Maximum temperature (t * (j + Ai = 427 -f 2640 « 3067*C. Since T,/T2 F,/Fjdur.
ing an isobaric process,
3067 4- 273
4.77
42T 4- 273

* Cetane is CiaHM, alpha-methylnaphthalene Ci&Hm. Either Ci«Hm or CioHu may be used

to represent average fuel oil.
160 INTERNAL COMBUSTION ENGINE POWER PLANT
Not only is the calculated higher than can be considered practical in the engine
cylinder, it is high enough so that thermal dissociation of the products would have
prevented its attainment. Furthermore, an R of 4.8 is large enough to impair ideal
efficiency seriously. The solution is to use A:F ratio, higher than the chemical ideal.
1650 4^ 273
Assume that 1660*C is the limit of h. Then R = 8.76.This is compat-
427 + 278
ible with good efficiency.
w = Q/CjAT = 41719/1(1050 - 427) -84ag
A:F ratio = 34.1 — 1 = 33.1
Although this example is based on an ideal engine, the actual engine is
similar. Diesels work on from 2 to 5 times the ideal air ratio at rated power,
more at part load. When the engine cannot aspirate charges of air of standard
density on account of altitude condi¬
O 100 tions, the effect is to curtail the
G quantity of fuel that can be injected

<
UJ
%f%
N >

11
««“ A per power stroke if the same maxi¬
mum temperature limit is to prevail.
The power output is therefore re¬
duced. Fig. 6-10 shows manufactur¬
er's practice in derating engines for
1 - A• altitude operation.
B- UNSUPERCHARGED __ Some of this loss of power can be
tt ) *5 VO 1*5 2-0 2<5 3^ 3« 4*0 4-5 avoided by using a- supercharger
Ul
Q. ALTITUDE ABOVE SEA LEVEL - THOUSAND nr with the engine, driven either me¬
Fig. 6-10 Decrease of power at altitudes. chanically from the crankshaft or by
(DEMA Data.) an exhaust gas turbine. Also super¬
charging is practiced on sea level
engines in order to raise their mean effective pressure. Substantial increases in
power per cylinder are possible.One manufacturerwho8e22.86ciDx30.46cmx450
rpm, 6-cylinder engine is rated at 197 hp is able to increase the sea level rating
to 313 hp with a turbo-supercharger and oil cooling for the pistons.
Combustion of Diesel fuel goes on in two stages, first an ignition delay stage,
then a stage of inflammation of the fuel-air mixture. For brevity, the actions
occurring in each stage are diagrammed in Fig. 6-11. The implication there is
that the process of combustion is complex. This is substantiated by research.
Although the whole delay period may be only microseconds long, its character¬
istics are quite influential to external operating characteristics of the engine,
such as detonation, starting, products of combustion, and smoothness of opera¬
tion. Injection must begin several degrees of crank angle before dead center, the
advance being governed by engine speed, compression temperature, spray
shape, delay period of the fuel, and other lesser factors. Following ignition
there is a rapid pressure rise during ^‘inflammation,” the extent of this being
inversely to the delay period, for the longer the delay the more fuel in the cylin¬
der to be suddenly inflamed. In an exaggerated case the pressure rise is sharp
enough to constitute an engine knock. Higher Cetane rating of the fuel is one
cure for this. After inflammation the incoming spray bums directly without lag.
Whether this final stage is at constant pressure or not depends on the rate of
nozzle flow.
ENGINE PERFORMANCE 161

Mlxtur*
-ignitiOR- -Rnol eombuftiofi •
•Otioy ptriod-
rnyticoi osioy *i*untniiccii OMay*i

•mini iTOtlon I
of tfroom of , Inflommoflon,
injoetod futi

Mlxin9 0^ I
liquid fuol
_ with oif

Mixturo of
products of
portiol oxidotionl
Mixmo iEFD or of thtrmol
docomposition
with oir

Tomporoturs Tomporoturs
ond oxygon ond oxygon
-I'liii iiiimni] I coneontrotion
fovoroMo unfovorohit
for for complots|
combustion combustion
(chilling,
ovorloon,
ond owor-rleh
mix turns)

Products
(ts of
oomplol
dots
combustion
stion

Fig. 6-11 Outline of combustion process in the Diesel engine. (Courtesy Trans. SAE.)
Products of combustion are theoretically CO2, H2O, and N2, but, as Fig. 6-13
shows, the products may contain carbon monoxide, unbumed fuel, and the
(odoriferous) aldehydes. However, con¬
centrations of CO are never very much, PRESSURE, kg/cm* ABSOLUTE
and there is an absence of the H2, CH4,
and CgHe found in spark ignition en¬ FUEL CUTOFF

gine exhaust. As the fuel-air ratio in¬ o-b INFLAMMATION

creases there is a tendency to production b~c CONTROLLED

COMBUSTION
of smoke from unburned carbon. This
and cooling requirements may be the
limiting factors on the maximum power
of an engine. -120 -90 -60 -30 TDC 30 60 90 120
6-4 Engine Performance. The rapid CRANK ANGLE FROM TDC

cyclic action in an actual engine modifies Fig. 6-12 Time record of the Diesel
the ideal cycle of Fig. 6-3. The diagram combustion process.
comers are no longer sharp, fuel must
be injected ahead of dead-center, etc. The forms of ^indicator cards,” which
are instrument-drawn p-t/’s from running engines, are shown in Fig. 6-14 for
both 2- and 4-cycle types.
A Diesel engine is used as a source of power which can utilize a relatively
162 INTERNAL COMBUSTION ENGINE POWER PLANT

Fig. 6-13 Products of combustion as affected by fuel-air ratio. CFR Diesel engine.
Note: IJI = 1 for clear exhaust; oo for opaque exhaust. (Courtesy Trans. SAE.)

cheap fuel. Therefore its power capacity and thermal efficiency are paramount
engineering considerations.
The developed cylinder power is the indicated horsepower, abbreviated, ihp.
After engine friction and accessory power needs have been satisfied, the net
power available at the engine shaft
is the brake horsepower, bhp, so
called because it can be (but nowa¬
days seldom is) measured by a brake
dynamometer. Indicated power is
difficult to measure directly on small
Diesels with great accuracy, because
Fig. 6-i4 Typical engine indicator dia¬ the volume of indicator lead through
grams. the cylinder head appreciably lowers
the normal compression ratio. Also,
the cylinder heads of small-bore high-speed engines have special chambers,
valves, and other equipment, leaving little or no space available for an indi¬
cator lead. Since none of these difficulties exists with the large stationary Diesel,
indicator openings are provided as standard equipment.
A hypothetical pressure, known as brake mean effective pressure, bmep, can
be employed to show the magnitude of mean effective pressure. The true pres¬
sure, pmep, is higher on account of engine friction losses.
engine performance 163
ibp - p„^ LANJ449 702hp (6-10)
bhp - 2wWrNI**9 702hp (6-11)
bmep = bhp Xi4S7(t2/LANpkg/cai^ (6-12)
in which pm.. — Indicated mep, kg/cai*.
L » Piston stroke,.un.
A = Piston face area, cm*.
Np = Number of power strokes per min.

N = Rotative speed, rpm.

W — Net dynamometer force, kg^

r = Dynamometer arm length, cm.

As is true of all prime movers, there are a number of efficiency expressions
applying to Diesels. Mechanical efficiency is the i atio, bhp/ihp.

2648
Indicated thermal efficiency, rja = —(6-13)
viiQ

Brake thermal efficiency, i7tb *= (6-14)

in which Wi, Wh = Fuel consumption, kg per hr per ihp or bhp.

Q = Fuel heating value, J/g*, either Qh or Qi., according to

policy.
The Diesel Engine Manufacturers’ Association (DEMA) is the trade group
in this field whose standards and policies usually prevail. DEMA power
guarantees are made on the presumption of altitudes less than 317.5 mabove
sea level, barometer 71 .6 mm mercuryormore, airmtake temperature 32.2®C
or less. Power available at higher altitudes is indicated by Fig. 6-10. Fuel oil con¬
sumption guarantees are made in kg per net bhp hr at %, and full-load rat¬
ing, assuming fuel with Qh = 45016 J/g
Example 1: A 6-cylinder Diesel engine on dynamometer test was found to use
3S kg of fuel, having = *6016 J/g, in a one-hour test at steady load. The
brake thermal efficiency and the brake mep will be determined from the following test
data and measurements. Cylinder is 21.6 om x 26.7 om, 4-cycle type. Speed, 600 rpm.
Dynamometer torque, 25000 kg-om.

bhp » 2ii X 26000 X 600/449702 = 209.6 hp

, // 21.6*n 600\
bmep « 209.O X 449702/(26.7 X —X 6 X 1 — 6.36 kg/om^

Wh « 38/209*6 . 0.181 kg per bhp hr

i|ib » 2648/(0.181 X 45016) - 32.4%

* Joules per gram.

164 INTERNAL COMBUSTION ENGINE POWER PLANT
The energy supplied to I.C, engines in the form of Qb of the fuel input is
generally broken down into the following items for heat-balance* purposes.
1. Useful work, the actual net shaft output.
2. Cooling, Heat absorbed by water jackets. Sometimes this is not separable
from some of the friction and exhaust loss due to merging streams of water that
cool cylinders, exhaust manifolds, and lube oil heat exchangers. A wcAt type of
loss.
3. Exhaust gas loss. Heat carried off as sensible and latent heat in the prod¬
ucts of combustion. Sensible heat is a wcAt loss; latent heat is of the form
2442.7 X 9Ht (in J/g units).
4. Mechanical friction. Crankshaft and connecting rod bearings, piston,
etc. Usually taken to include energy supplied to engine-mounted auxiliaries
such as pump, governor.
5. Radiation and unaccounted-for. Heat radiated from engine and incom¬
plete combustion loss due to presence of fuel in exhaust gas.

Table 6-1. TYPICAL FULL-LOAD HEAT BALANCES (%)

(Based on Qh)
OUo Cycle Diesel Cycle
Spark Compression
Ignition Ignition
1. Useful work 25 34
2. Cooling 30 30
3. Exhaust 37 26
4. Friction, radiation, and
unaccounted 8 10

Input; heating value of fuel 100 100

Items 4 and 5 are not often separable nor measurable. They arc usually
combined, then the balance is made by subtracting Items 1, 2, and 3 from the
heat of the fuel and considering the difference to be Items 4 and 5.

4- 44-' 4“
1 si L. _ — ___
——
5° — --—

— __— _ U Wd '#i fd
fi rm JT. WA fM w, WA fM WA fM
—
^33 -- - -- ~\'d
8* il-
ss m
rv/j
tti X *

cSSSSiMwywtfiMM
-OiMvlft ought not to bo
-oporotod continuously i
JjffLJL.U-LJJU_U_l 1
for long poriodo of \
.timo in this load rang#
w^Ammmm

0 10 20 30 40 50 60 70 60 90 IOC
H) 20 30 40 50 60 70 60 90 100
PERCENT PLANT CAPACITY FACTOR
PERCENT RATED UOAO

Fig. 6-15 Range of variable load performance of Diesel plants.

♦Sec Sec 8-2 for definition of Heat Balance
ENGINE PERFORMANCE 166
An advantage of the Diesel is its fairly constant efficiency over considerable
range of load; also the achieved efficiencies do not vary greatly with the size of
the engine <as they do in steam plants). In Fig. 6-15 are shown these variable
load characteristics for a large group of units. The areas represent the range
and include approximately 90% of all data, Fuel consumption data are for the
engine alone, under optimum or test condi¬
tions. The other graph represents plant
operating experience, taken as it was found,
with gross output per litre of fuel given
against plant capacity factor. Net plant
output per litre would be somewhat lower
due to energy consumption by plant auxil¬
iaries. Note that a certain plant capacity
factor, say 30%, does not imply units oper¬
ating at 30% of rated load. If there were
several units in the plant, it would be pos¬
sible to operate some of them at the point
of minimum specific fuel consumption.
The variation of engine performance of
a specific engine (a slow-speed, 2-cycle Fig. 6-16 Variable load perform¬
crankcase compression unit) is shown in ance characteristic of slow-speed sta¬
Fig. 6-16. Note the characteristic low ex¬ tionary Diesel engines.
haust temperature. Generally the higher
the speed, the higher this will be. The exhaust temperature is important, for
it indicates: (a) magnitude of exhaust loss, but not comparatively between
two engines unless they use the same A:F ratio; (b) extent of problem of
shielding or cooling the exhaust manifolds and ducts; (c) proper adjustment
of injection pumps.
In a carbureted engine all cylinders get approximately the same mixture,
whether it be correct or otherwise. Hence carburetor adjustment can bring

Fig. 6-17 Thermocouple measurement of individual cylinder exhaust temperatures.

(Courtesy Illinois Testing Laboratories,)
166 INTERNAL COMBUSTION ICNOINE POWER PLANT
equally good combustion to all cylinders. But the compression ignition engine
has individual cylinder injectors. Fuel-air mixture will not be the same for all
cylinders unless the injection mechanism is individually adjusted, using the
exhaust temperature as a guide. Other cylinder difficulties, such as clogged ports,
gummed piston rings, scaled water jackets, can also be detected and isolated
with multiple exhaust thermocouples.

Powet
Fig. 6-18 Auxiliary equipment of the Diesel engine power plant.

6-5 Plant Design. The I.C. engine is almost a self-contained prime mover
in that it requires very few auxiliaries beyond those built onto the engine by
the manufacturer. Although '-oine are built as complete power units on a single
base, this is not cu.^-tomary if the engine is a stationary type Diesel designed
for liigh-cafiaeity-factor service in a power station. Fig. 6-18 shows the usual
auxiliary eciuipmcnt that must be selected and worked into the complete plant
by the power plant engineer.* These auxiliaries may be grouped into the follow¬
ing systems: (1) fuel supply; (2) cooling; (3) intake and exhaust; (4) lubrica¬
tion; (5) starting.
It is common practice to set the engine generator units in a power plant on
♦ A few of the auxiliaries shown are standard equipment furnished by the engine manu-
faeturer. Standard Prartirvs, published by DEM A, conlains a complete list of the standard
cquipnu'nt which the engine irmnufacturere propose to furnisli. In genc’ral it does not in¬
clude any apparutas which might require different treatment for different installations.
PLANT DESIGN 167
parallel center-lines. As the average plant has two or more engines, the parallel
arrangement leads to a building of somewhat square shape. Ample clearance
should be allowed in the layout for dismantling of the engine, generator, and
exciter. Manufacturers* prints indicate extreme dimensions for removal of
rotors, pistons, etc.
Engine foundations were described in Chapter 4. The heavy reciprocating
parts of the Diesel will cause objectionable vibration unless sufficient foundation
mass is provided. It is better to err on the side of too large a foundation than to
have complaints of vibration, because nothing can be done about it after the
foundation is poured and the engine is in place. Foundations of engines in the
basements of hotels, and in similar locations wliere any vibration transmitted
from the foundation would be undesirable, must have the foundations insulated
from the rest of the building by corkboard or vibration dampeners. For the
same reason, piping connections to the engine should be through short flexible
sections.
Power station engines are always direct-connected to their generators, usu¬
ally close-couidcd. The standard engine-type generator has its own bearings
and is connected by flexible coupling to the engine shaft, outboard of the fly¬
wheel. Flywheels are furni>hed with jacks for rotating the engine to its starting
position and for use when repairing or adjusting. Engines large enough so that
one man could not operate the manual device have pneumatic turning appara¬
tus.
Voltages i)referred for a-c generators from about 100 to 600 kw are either
480, 600, or 2400 v. Larger units may go up to 4160 v. Excitation is at 125 v d-c,
with the exciters being belt-driven to increase speed, thereby reducing physical
size. The relation between number of electrical poles ??, the frequency /, and the
rotative speed N is:
N = 120///? rpm (6-15)

A typical layout of a multi-engine Diesel power plant is shown in Fig. 6-19.

This design has a pipe trench below the floor to accommodate auxiliary piping.
In other plants one may find a basement has been excavated which will provide
room for piping, conduits, storage, w’orkshops, etc. The concrete sides of the
engine foundations are visible in the basement.
The Diesel plant arrangement must include a number of instruments for the
guidance of operators. These are chiefly tem])erature, pressure, flow, and
electrical instruments. Discuvssion of the latter i^ deferred to Chapter 15. If
reliability, economy, and safety are to be obtained, a continual check must be
maintained on operation by means of instruments. Thermometers and pyrom¬
eters are highly useful as an index to operation. They may be used to indicate:
1. Individual exhaust temperatures, and so tell wdiether one cylinder,
through faulty pump setting, is performing less than its share of work.
2. Jacket water temperatures. The rate of flow of cooling water is adjusted
by valves for proper outlet and inlet temperatures.
3. Injection air compressor temj^eratures to check on w^orn rings, l)roken
valves, or carbon deposits in the compressors.
4. Lubricating oil temperatures to check on oil cooler operation.
5. Fuel oil temperatures.
168 INTERNAL COMBUSTION ENGINE POWER PLANT
Pressure gauges are used on the injection air system to gauge both final and
interstage pressures. The starting air system is also equipped with a pressure
gauge, as is the lubricating oil system if pressure oiling is used. Fuel oil day
tanks are equipped with glass liquid level gauges and the main storage tanks
have a sounding opening so that the level, and thereby the volume, of the oil in
storage may be determined by a staff gauge.
The engine brake horsepower itself is generally determined only indirectly,
that is, either from the generator output and its efficiency as stated by the manu¬
facturer, or from indicated horsepower, and an assumption of mechanical
AUXILIARY SYSTEMS 169
efficiency. For operating records fuel flow meters are used. Provision should be
made for primary weighing for calibration and performance tests.
6-6 Auxiliary Systems. The equipment here described is not ordinarily
supplied with the engine. It is the work of the designing engineer to select and
locate such equipment properly. The operating engineer must imderstand it,
know how to control it,, and detect any malfunctioning.
\P(lel Supply. The Diesel plant may involve considerable outlay for fuel oil
equipment. To select and arrange this equipment one needs: (1) to be able to
predict the maximum rate of fuel flow; (2) to know the engine characteristics
and accessory equipment; (3) to know details of piping system layouts in order
to arrange equipment connections for reliability and flexibility of fuel supply;

Fig. 6-20 Installation of fuel tank underground.

(4) to determine what fuel supply is available, viz., size of deliveries, as well as
their frequency and reliability. Then the design of the fuel supply system is an
arrangement of equipment and piping to accomplish a flow described briefly
thus: Delivery from tank car or truck through the unloading facility to main
storage tanks, thence by transfer pumps to smaller service storage tanks from
which the fuel passes through cleaning equipment to the engines which bum it.
This main flow, so described, is made workable and practical by arranging the
piping equipment with the necessary heaters, by-passes, shut-offs, drain lines,
relief valves, strainers and filters, flow meters, pressure and temperature indi¬
cators.
The actual flow plans depend on type of fuel, engine equipment, size of plant,
etc., and are quite varied as they are subject to engineers’ individual ideas.
Most arrangements, however, can be classified as:
1. Simple suction system by means of engine-driven pump from service
tank locat^ near engine and few feet below it. Such pumps are constant volume
delivery, and an overflow line is required back to the tank.
170 INTERNAL COMBUSTION ENGINE POWER PLANT
2. Transfer system by which a motor-driven transfer pump moves oil from
main storage to an elevated service tank, sometimes called a day tank, from
which it can flow by gravity to (a) engine injection pumps or (b) engine
auxiliary oil pump.
3. Like the preceding, a transfer to a
day tank with gravity flow to a small en¬
gine oil tank from which the engine pump¬
ing system takes what it needs.
The volume of oil storage should be
suflBcient so that the maximum rate of fuel
consumption in the plant will not empty
the storage during the expected period be¬
tween deliveries. Truck trailers can usually
drop 11300-19900 1 at a delivery. Railway
tank cars vary from 30300 to 45400 1 in
capacity. Whether the tank shall be located
above ground or below is a matter of local
conditions, including method of oil delivery,
local ordinances, building location, etc. The
National Board of Fire Underwriters has
regulations covering the construction and
installation of fuel oil storage tanks, as do
Fig. 6-21 Service tank for gravity many local authorities. Comparison of ex¬
fuel supply to engine. posed vs buried tanks reveals the following
advantages.
Exposed Tanks Buried Tanks
Cost of excavation is avoided. Oil can be delivered by gravity.
Leakage is readily detected. Fire hazard is reduced.
Maintenance is easier. Plant grounds more easily landscaped.
Water and sediment are easily drained.

Horizontal welded steel tanks are quite common, the large-diameter, exposed
vertical tanks being seen only with very large plants. Exposed tanks are set on
concrete saddles and can have openings from the bottom as well as the top.
The buried tank, as shown in Fig. 6-20, has all openings on the top. More de¬
tails on storage tanks are given in Sec 12-6.
The number and size of tanks are established from the plant size, capacity
factor, and delivery schedules. It would be well to have sufficient tank capacity
to carry the plant for twice the normal time between fuel deliveries. Also the

a. Gear type h. Lobe type c. Screw t3rpe

Fig. 6-22 Some types of fuel transfer pumps.
AUXILIARY SYSTEMS 171
tanks must be chosen so that there can be oil remaining for operating when
empty storage volume is sufficient to accommodate the delivery. Capacity of
the fuel oil unloading pumps and lines will be determined by the size of delivery
and the time wished to be spent on unloading. Transfer pumps and the oil sup¬
ply lines from storage to engines must have sufficient capacity to be well in
excess of the maximum rate of flow at full plant capacity.
Example 1: A Diesel plant will have one 776-kw and two 400-kw units. Delivery
of the fuel oil is to be arranged on a monthly basis, the vendor expecting to deliver by
tank car. Expected plant capacity factor, 35%. The tankage and transfer system will
be sized.
The capacity of this plant is 775 -f 400 4- 400, or 1675 kw. Average output at 36%
load factor is 550 kw. Examination of Fig. 6-15 shows that the greatest fuel consump¬
tion to expedt at 35% load factor is 1.72 kw hr per litp^ oil, the lowest ftr hr per I
550
Required storage =- X 24 X 30 « 1816601
^ ® (l.72 + 2.64)/2
Five tanks of approidmately 878601 each are indicated by these figures, presuming
that the tank car capacity is 878501*
At full-rated load the fuel consumption can be expected to be aboutO.18 kg per bhp
hr (Fig. 6-15). 1575 generated kw is 2880 engine horsepower if a generator efficiency
of 94% is assumed.
Maximum rate of fuel usage =8280 x 0.18 « 410.4 kgper hr. At 0.9 kg per I this
would be a flow of 7.S l/min. Transfer pumps and piping should have at least tUs capacity,
probably more in view of the tendency of electric service plants to grow in size with
the passing of time.

Fig. 6-23 Multiple-unit fuel handling system. Relief, vent, drip, and other secondary
lines not shown.

A study of Fig. 6-23 will show how equipment is interconnected through

pipe headers for maximum flexibility of use, how pumps and filters are provided
in duplicate^ where relief valves are placed so as to prevent excessive pump*
pressures, and the location of flow meters for checking deliveries, measuring oil
in transfer, and metering flow to individual engines.
172 INTERNAL COMBUSTION ENGINE POWER PLANT
Whether oil must be heated for flow either in tank car or storage tank, for
atomisation, depends on the climate and the viscosity. Refer to Fig. 12-21 for
need for heating. If heating is required, hot water coils can be connected into the
engine jacket water circulation.
Fuel oil transfer pumps are positive displacement, rotary types, direct-con¬
nected to motors. There are several other practical designs in addition to those
here illustrated. Being of the positive displacement type, all these pumps need
to have pressure relief valves on their outlets, draining to storage tanks, or to
their own suction lines.
Satisfactory operation of a fuel oil supply system is impossible unless the
designer and builder competently handle provisions (1) for cleanliness, (2) for
valving to alternate lines during emergencies, and (3) for insuring tight pipe
joints in all suction lines. Imperfect pipe installations that would give no
trouble with water will sometimes suck air at the joints when carrying oil at a
vacuum. Before being covered, all oil lines should he put under air pressure and
the joints tested with soap solution. Small air leaks into the air lines can be the
source of exasperating operating difficulties and are hard to remedy once the
plant is in operation.
Great emphasis is placed on the need for cleanliness in handling bulk fuel
oil. Distillate fuel is clean as it leaves the refinery, but the multiple handling
before the ultimate user gets it may introduce dirt. There is no comparison be¬
tween the purity required for oil engines and that for gasoline engines. The
latter have relatively large jet openings in their carburetors. The fuel is light
and gravity will remove almost all dirt. If there is dirt that passes the coarse
screens used, the most it can do is cause the engine to misfire or stop. But Diesel
fuels are more viscous and hold dirt in suspension. They are rammed at high
pressure through small, finely finished injection pumps and then sprayed
through nozzles with minute orifice openings. Dirt particles will ruin the fine
lap of injection pumps or plug the injection nozzle orifices. Hence, high-grade
filters are of great importance to the Diesel oil supply system. Engines are
equipped with one or more, and the plant designer should arrange for one or
more external filters in his piping system. These should not be by-passed. In¬
stead they should be in duplicate so that one may be in use while the filter ele¬
ment of the other is being renewed. The piping between filter and engine should
be thoroughly oil-flushed before being first placed in service. Waste- and cellu¬
lose fiber-packed filters with renewable elements, bag filters, and cleanable
metal filters (edge type) are able to clean fuel oil to the extent of removing all
particles over 5 microns in size.
Lubrication. This is important in any mechanism. It is especially vital to the
Diesel engine because of the high pressures and small clearances common in
these engines. The life of the engine, the efficiency, and the extent to which it
may be expected to provide continuous service are dependent on the effective¬
ness of the lubricating system inbuilt in the engine, the success with which ex¬
ternal features of the system are selected and installed, and the quality of the
lubricating oil.
In a Diesel engine the lubrication requirements naturally fall into the fol¬
lowing grouping.
1. Piston and cylinders.
AUXILIARY SYSTEMS 173
2. Crankshaft and connecting rod bearings.
3. Gears or other mechanism designed to transmit motion to auxiliaries.
4. Integral injection or scavenging air compressors.
Lubrication may be achieved in different ways, i.e., full pressure lubrication
similar to that employed in the automobile engine, mechanical force-feed lubri¬
cators, or gravity circulation from an overhead tank. The pressure circulation
system, in which an oil pump supplies the lubricant under pressure to many
parts of the engine through a duct system, and to the crankshaft and wrist pin
bearings by drilled passages in the shaft and rods, is used more than any other
type. The cylinder walls will be lubricated by oil mist slung outward from the
connecting rod bearings, or by splash of the rod ends into oil pools. Large, slow-
speed engines are not readily adaptable to cylinder lubrication by this method
and may have mechanical force-feed lubricators for the cylinders.
The complete lubrication system will usually include some of the following

auxiliaries external to the engine: pump, oil cleaners, oil coolers, storage and
sump tanks, gauges, and safety devices. It is quite common practice to have
enclosed crank cases which serve as oil sumps into which the oil supplied for
engine lubrication drains, and from which it may be withdrawn by a pump.*
As oil passes through the lubrication cycle it accumulates impurities in the
form of carbon particles, water, and metal scrap. Continuous reliable operation
may not be expected in a recirculating system unless definite attention is given
to oil cleaning. For this purpose filters, centrifuges, or chemical action have
been employt d. Typical of mechanical filters are cloth bags, wool felt pads,
paper discs, and cartridges of porous material. Many engineers prefer to rough-
clean the oil with simple screen type strainers, and then pass the oil through
high-speed centrifuges for ultimate cleaning. Centrifuging may be done by peri¬
odic cleaning of the entire charge of lubricating oil or by continuous cleaning of
a smaller fraction which is split off from the main flow and returned to the
crankcase after cleaning. In either case, the oil should be hot before being ad¬
mitted to the centrifuge.
The friction losses of an engine will appear in part as the heating of the
lubricating oil during its circulation throu^ the engine. This must be removed
by cooling the oil before recirculation. Ordinarily, not much over of the
^Deeoriptioiis of engme hibrioation systems are not posaible in this book. Many of the
oil refining companies publish excellently illustrated literature on this subject.
174 INTERNAL COMBUSTION ENGINE POWER PLANT
heat in the fuel will find its way into the lubricating oil. By assuming this and
taking the useful output as 30%, it would appear that the heat to be removed
from the oil would be (632.4 X0.025)/0.30 or 52.7 kcal per bhp per hr. Shell-and-
tube coolers are employed, generally being part of the engine equipment. The
water used to absorb the heat may be part of the engine cooling water.
Lubricating oil replacement is no small item of Diesel engine operating cost.
Data from several stationary power plants show an average consumption of a
gallon per 1600 kw hr generated at full-load rating. Thus lubricating oil con¬
sumption is about 1% of fuel consumption. The type and quality of the lubricat¬
ing oil are always specified by the engine manufacturer. In general, slow-speed
engines operate satisfactorily on a well-refined straight mineral oil, whereas
specially treated and fortified oil is required for best results in heavy-duty
medium-speed and in high-speed engines. An engine lubricating oil, in addition
to preventing metal-to-mctal contact, must dissipate heat from the upper
cylinder and bearings, wash deposits from the surface (detergency), and hold
them in suspension (dispersion). The qualities of “oiliness,’^ film strength, and
viscosity are produced by the selection and processing of the mineral oil base.
Resistance to oxidation and foaming, detergency, dispersion, and anticorrosion
qualities are provided by adding detergents and inhibitors to the straight
mineral oil.
Lube oil tests are (1) oxidation, (2) sedimentation, (3) acidity, and (4)
viscosity and viscosity index. The fourth item is a most important physical
property. It governs both manufacture and application of the oil. It is commonly
graded by viscosimeter tests, but reported as an SAE number. SSU viscosities
may be converted to SAE numbers by tabulations of which Table 6-2 is an
abstract. Most refining companies offer Diesel lubricants in SAE grades from
10 to 60 in ten-number increments. Viscosity index is an, empirical number
which rates the change in viscosity with oil temperature. This index was
originally established at 0 for asphalt base oils and 100 for paraffin base oils.
Toda^ome solvent refined oils have viscosity indices over 100.
yjntake and Exhaust Passages. A large Diesel engine requires no inconsider¬
able amount of air for combustion. An engine needsfromO 056 to 0.084 m® of air
per min per hp developed. In its natural atmospheric state the air will vary in
temperature and in dust content. An air intake system thus becomes a necessary
part of a Diesel plant installation, and this will include a suitable engine mani¬
fold in the case of multicylinder engines. The air system begins with an intake
Table 6-2. VISCOSITY CONVERSION
SAE Number SSU at .U.4^C
Min Max
10 90 120
20 120 185
30 185 255
40 255
SSU at 98^9^0
40 80
50 80 105
60 105 125
AUXILIARY SYSTEMS 176
located outside the building provided with a filter to catch dirt which would
otherwise cause excessive wear in the engine. In northern states, the outside air
may sometimes reach such low temperatures that it promotes misfiring at low
loads; consequently, the air intake system may need to have a heating element
using exhaust gas. The filters used may be conveniently classified by types as:
oil-impingement, oil-bath, and dry. The impingement type consists of a frame
filled with crimped wire or metal shavings. These are coated with a special oil
so that the air in passing through the frame and being broken up into numbers
of small filaments comes into intimate contact with the oil, whose property is
to seize and hold any dust particles being carried by the air. When in service,
the cleansing capacity of this type of filter becomes j^rogressively exhausted,
and it is necessary periodically to refresh it by removing, washing, and re-oiling.
In the oil-bath type of cleaner, the air is swept over or through a pool of oil so

that the particles of dust become coated. The air is then caused to flow through
a filter, which retains the oil-coated dust particles. The dry type of filter is made
of cloth, felt, glass wool, etc. The effect of such filters is to catch dirt by causing
it to cling to the surface of the filter material. This type also needs to be cleaned
at intervals, the period between cleanings being governed by the amount of air
which has been used by the engines and the dust concentrations in it. Occa¬
sionally engine noise may be transmitted back through the air intake system to
the outside air. In such cases, a silencer is required between the engine and the
intake. Lightweight steel pipe is the material for intake ducts.
The exhaust system also requires the attention of the designing engineer
because, although the engine manufacturer will provide an exhaust manifold
for the engine, the plant designer must arrange for connection of the manifold
to a suitable exhaust system which will convey the exhaust gases to the atmos¬
phere with proper provision for the following:
1. Silencing of the exhaust noise to the required degree.
176 INTERNAL COMBUSTION ENGINE POWER PLANT
2. Discharge of the exhaust sufficiently high above ground level.
3. High temperature of the exhaust gas, which may necessitate water-cooled
exhaust lines or special high-temperature materials.
4. Expansion and contraction due to changes in temperature between the
extremes of full load and cold.
6. Possible by-product heat utilization. Where the exhaust is to be employed
for building heating, or other sources of heat, such conditions necessarily modify
,the exhaust system.
6. Arrangement of the exhaust system to minimize the back pressure created
by the exhaust system itself. (Header pressures ought not to be over 0.143 to
0.21 kg/cm‘ gauge pressure.)
7. Isolation of engine vibration from building and muffler system by use of
a flexible section of exhaust pipe.
The exhaust system must carry approximately 0.168-0.224mVniin of gases
per hp developed, this volume being at the average exhaust temperature.
Where the exhaust from a single cylinder of a 2-cyole engine is carried a dis*
tance to the header, there is some possibility of resonance of pressure wave
creating some impairment of engine operation. This may need investigating in
some instances, but is of little importance in the case of 4-cycle engines or 2-cycie
multicylinder engines having an .engine manifold.
Muffling of the exhaust noise is the main problem met in the exhaust system.
This is done in a silencer usually located outside the building. It is customary to
provide individual silencers for each engine. They may be of cast iron, sheet
steel, or concrete, the latter being simply
pits provided with baffles. A pipe or stack
slightly smaller than the exhaust line should
extend vertically from the silencer and
carry the gases above the building roof.
Commercial silencers, built by firms spe¬
cializing in acoustical engineering, are
usually made of steel. Several principles are
to be found in the types now available on
the market. Some operate by absorption
of pressure waves by acoustical material;
some by expansion through carefully de¬
signed inner chambers in the silencer body;
and some by providing baffles, or in other
ways creating internal friction. All designs
of silencers must necessarily create some
back pressure.
One system of supercharging, the Buchi,
Fig. 6-26 Exhaust-driven super¬ becomes part of the exhaust and intake
charger. systems. In order to obtain increased power
from an engine the intake air is pressur¬
ized. This has been covered in Sec 6-3. It is practiced mainly with 4-cycle
engines and by an exhaust-gas-turbine-driven centrifugal compressor. The
construction of this unit is shown in an accompanying figure. The single-
stage turbine and comnressor wheels'are mounted on the same shaft, uncon-
AUXILIARY SYSTEMS 177
nected externally. Exhaust gas passing through the turbine develops the power
to give the combustion air an initial plenum of 381-508 mm. Hg. Typical location
of the supercharger is illustrated in Fig. 6-33. Sometimes mechanically driven
blowers are used for supercharging, but the presence of such does not neces¬
sarily mean a supercharged engine, for all but the slow-speed, crankcase-
compression, 2-cycle Diesels require blowers for scavenging air.
Starting LC. Engines, The cycles of these engines must be externally moti¬
vated until the essentials of fuel-air-ignition are correlated and power ensues.
Spark-ignition engines are employed mainly in the smaller sizes—on auto¬
motive and portable equipment where the compression ratio to be overcome in
cranking is only 5 to 7. Hand and electric motor (6-12 v d-c) cranking are
practical.
The high compression pressures of the full Diesel make it difficult to start
even the smallest Diesels by hand cranking and, in general, Diesel engines must
be started by a mechanical cranking system. Compressed air, electric cranking
motors, and auxiliary gasoline engines are used. Large stationary Diesels are
started with compressed air.
In addition to compressed air for starting, the air injection types require an
injection air system; also some solid injection types have a scavenging air sys¬
tem. A mechanical injection engine, having no air compressor cylinder as an
integral part, requires the separately driven compressor for starting air.
Two or more compressed air storage tanks, or ^‘bottles,are provided. A
small compressor is installed either as a normal or emergency supply of com¬
pressed air to recharge the storage tanks. An air injection engine has an integral
compressor, and air may be bled from the injection system for the purpose of
recharging the starting tanks. Starting air pressure needs to be between 14 and
42 kg/cm^.
During the starting period, compressed air is admitted to one or more of the
engine^s cylinders. This powers the engine much the same as steam works in a
steam engine. The injection pumps are inoperative while speed is being gained
under air power. Then air is turned off, and oil injection is started while the
engine '^drifts” on its momentum. An engine in proper condition will fire im¬
mediately if it is a high-compression type with considerable margin of assurance
of ignition. When cold, the oil engines with lesser compression ratios and smaller
cylinders require the use of expedients such as electric glow plugs in the cylin¬
ders, continued operation by external power from a gasoline engine until
warmed by the heat of compression, or auxiliary flame heaters for the intake air.
Cooling System. The temperatures existing inside engines would disintegrate
the film of lubricating oil on the cylinder liners and otherwise render the engine
unserviceable by warping of valves, pistons, etc., were they not cooled by circu¬
lating water through jackets surrounding the heated parts. The operation of
cooling and recirculating this water accounts for much of the plant auxiliary
equipment. Small engines may be served with a cellular heat exchanger,^
through which the air is drawn by means of a fan, the water being broken up
by the flow through the radiator cells into numerous small streams offering
large surface exposure and being rapidly cooled by the conduction of heat to
the air. Although this is a compact form of heat transfer equipment for small
^ Often erroneously called a "radiator.”
178 INTERNAL COMBUSTION ENGINE POWER PLANT
engines, it is inferior to evaporative cooling for power plants. Cost, bulk, an4
auxiliary power required all favor the latter.
The calculations necessary to determine the amount of cooling water re¬
quired are simple. The most difficult part is to fix upon the jacket inlet and
outlet temperatures. If the water is too cold, the efficiency suffers; and it
should not rise above82.2*Goiiherwise there may be danger of insufficient cool¬
ing with resulting failure of the lubricating film. The theoretical limits might
be put at 37.8® inlet to65*6® outlet, giving a27*8®rise. As a matter of fact, more
uniform thermal conditions are obtained by circulating the water more rapidly
than that would indicate. An examination of current practice discloses the fact
that temperature rises in actual use vary all the way from IT to 27.8®, with a
majority in the range from 19.4® to 25® rise. Inlet water temperatures are gen¬
erally from26.7®to32.2®and outlet temperatures from 43.3° to 54.4®*
If 30% is taken as the thermal efficiency and 32% as the loss to cooling
water, then the weight of water per hour per brake horsepower is

631.1 IX 0.32
kg per hr per hp
{ti - ii) X 0.30
From this comes the following equation relating cooling water to brake horse¬
power.

«> = 674.58 (6-16)

ti- h

in which w = cooling water, 1 per hr^

bhp = Rated brake horsepower.

tiy = Inlet, outlet, water temperatures ®C.

Sarie or Expantioii Tank E-~Raw Water Batin Lr—Raw Water Pomp

(Hm Well May Be Used F-~By<patt Vaivinf M—ThemwnMten
Intlead of Surge Tank) C--Tower Bypatt Valve N^Water Diacharge froni Engine
B—Open Vent H—Lnkrieating Oil Cooler 0---Coolinf Tower
C>~Anloniatie Air Vent I->Raw Water Make-np Connectioa P-~Drain Vahret
D^eac Eachanger K<-Jacket Water Pnmp

Fig. 6-27 Closed cooling system with tower. (Courtesy DEMA.)

EVAPORATIVE COOLING 179
The engine builder, of course, provides the jackets around the cylinders,
their interconnections, exhaust manifold cooling, lube oil coolers, and any
other water cooling passages which are considered standard engine equipment.
But the pumps which create the flow through these passages, the external
circuits for cooling water flow, and the water cooling equipment are the respon¬
sibility of the plant designer. Sometimes the same water that undergoes evapo¬
rative cooling is circulated through the engine (open system), but this is not
recommended on account of the variable impurity content of water so used.
The closed system wherein the engine cooling water circulates around a closed
pipe circuit is best. The water absorbs heat in the engine and releases it to raw
water flowing in another circuit, via a heat exchanger.
The water in the primary, or engine, circuit must be free from impurities
which would form a scale on the inside of the water jackets. The action of scale
on the Diesel engine jackets is similar in effect to that on a boiler tube. Re¬
duction of heat transmission causes the surfaces in contact with hot gases to
become overheated, even though ample water circulates through the water pas¬
sages. In the boiler, the result is a burned tube; in the Diesel, it is scored cylin¬
ders, twisted crankshafts, etc. An indirect or double-circuit system is shown in
Fig. 6-27. The water circulated through the engine jackets is not exposed to
evaporation loss, but is indirectly cooled by flowing through pipe coils over
which raw cooling water drips. This heat exchanger is made up of standard
SQmm. pipe allowing about 1.5 m of pipe per engine horsepower. Instead of a pipe
heat exchanger as shown, a commercial shell-and-tube cooler or a submerged
pipe coil in the cooling tower basin may be used. The equipment of such a
system consists of pipes and fittings of appropriate size, pumps (generally
motor-driven centrifugals), expansion tank, raw water makeup, emergency
water feed into the primary circuit, pressure gauges, thermometer, and a water
cooler. As the latter is the most expensive, bulky, and thermodynanfiically
technical portion of the cooling system, the process of evaporative cooling will
be considered in some detail in the next section.
6-7 Evaporative Cooling. The atmosphere is a mixture of air and water
vapor, in proportions described by the term humidity. It rarely contains all the
vapor it is capable of holding, and when it does its vapor-holding capacity can
be further increased by warming it. In any process wherein more water is
vaporized into the air the proper latent heat of evaporation must be supplied
from some source. Under certain conditions this source can be the internal
energy of the liquid water from which the vapor is being produced. What
liquid remains after the vapor has passed off is found to be considerably cooled.
The practical application of this to the Diesel plant is worked out in cooling
apparatus which is designed to expose a large surface of warm water to an air
flow, thereby humidifying the air and cooling the remaining water. The forms
of this apparatus* are: (1) atmospheric cooling towers, (2) mechanical draft
cooling towers, and (3) evaporative coolers.
The cooling action is the same in all of them. Air is humidified; some of the
warm water goes off into the atmosphere and fresh water must be added to the
system to take its place. The makeup is 2-5% of the water flow.
* Cooling ponds, once in considerable use, are not favored now because of the com¬
paratively large area they require. Spray ponds are much smaller for the same cooling job,
but they are infrequent compared to the apparatus described.
180 INTERNAL COMBUSTION ENGINE POWER PLANT
Atmospheric towers are long, narrow structures of considerable height
placed with the long axis normal to the prevailing wind. They are built to
utilise horizontal wind movements. Decks of wooden lattice inside the struc-

OvtrKtdd
TonK
LovvWaftr
Alortn.

wifiTwr

I Drain t Raw Water Supply

Emcr^-Row * '
Wa^er to Cmer^. Row
Sewer Water Supply Woter Softener

Fig. 6-28 Closed cooling water system using soft water coolers and spray pond.

ture serve to break up a stream of water released at the top of the tower so that
it rains down through the air currents and is evaporatively cooled. This type is
not often used with Diesel plants.
The cooling tower in Fig. 6-27 is on natural draft. Air circulation is obtained
through the levity of a confined column of warm air. The performance of these
is best with high water temperatures
WARM AIR OUT and if built as a high structure. They
SATURATED SPRAY
-ELIMINATORS are fairly common as adjuncts to small
plants.
WARM WATER
With a fan providing dependable air
SRAY NOZZLES
movement, a cooling tower does not
•TOWER RLLING
SLATS TO AID have to be as high as with natural draft.
WATER EXPOSURE According to the fan location, the
AIR IN AT tower is classified as forced or induced
COOLED I, J H , 1 NORMAL
WATER ‘ ‘ ATMOSPHERIC draft. The principle is explained and a
HUMIDITY typical cross-section shown in accom¬
a ** I panying illustrations. When this cooler
is built entirely of steel, including a
FAN.
CATCH BASIN MOTOR 6RIVEN heat exchanger for cooling the water of
Fia. 6-29 Principle of forced-draft cool¬ the engine circuit, it is called an “evap¬
ing tower. orative cooler.” With sheet metal ducts
arranged to lead air to and from it, this
cooler may be located indoors in engine rooms, utility rooms, etc. It has some
use in Diesel practice, but most coolers are either natural or forced draft out¬
door towers.
Some sensible heat is transferred from water to air in the tower, but the
latent heat of evaporation of the water vaporised is the main source of heat
EVAPORATIVE COOLING 181
transfer. The degree of cooling action is limited by the vapor that can be ab¬
sorbed before the air reaches saturation humidity at its leaving temperature.
Counterflow principle is employed because the natural direction of air being
heated is up-flow whereas gravity assists the down-flow of water. The outgoing
air comes in contact with the warmest water, thus giving it the maximum
capacity to absorb vapor.
The water vapor carried by air is determined by hygrometry, customarily
with wet-and-dry bulb thermometers such as the sling psychrometer. With
psychrometric charts (see appendix) these temperatures can be converted into
humidity data. The chart can be used to obtain relative humidity (a percent¬
age) and specific humidity, i.e., the water vapor in kg per kg dry aii The
following symbols will be employed in explaining the cooling action.

SHi, SH2 = Specific humidity of entering, leaving air.

lOw —'Water circulating per kg dry air, kg.
Aw = Makeup water per kg dry air, kg.
t' = Makeup water temperature, °C
<a, th = Temperature of water in, out, ®C.
hif hi = Enthalpy of moist air, in, out, J/kg diy air.
An equation of mass balance for the cooling tower is:
1 kg air -j- SHi -f- + Att; = 1 kg air -+• SH2 -t tiv (6-17)
The corresponding heat balance is:
hi 4“ Wwhfa "t" Awhv= hi -|- Wyrhfb (6-18)
With these equations and the chart, the flows needed at assumed psychro¬
metric data can be calculated for design purposes, and the working perform¬
ance of a cooling tower can be determined from test data.
Example 1: The flows through a cooling tower suitable for the cooling system of
a 1270-hp Diesel engine will be investigated. Consider the equipment to be represented
by Fig. 6-18. Data for the installation are: Engine water into heat exchanger 65*6®, out
37.8“; tower water into heat exchanger32.2* out48.9*, all Centigrade. Make-up water at
16.6*C. Atmospheric pressure, 1.03 kg/om* ab; dry bulb, 29.4*c; wet bulb, 21.1*C. From
Eq 6sl6,
874.58 X 1270
w ■* 30818 1 per hr «
65.6 - 37.8

Let tOw be the water flow in the cooling tower circuit.

(48.9 - 32.2)ti;^ . (65.6 - 37,8) 30818 ; = 01300 1/hr = 865 kg/min

This enters the tower at 48.9*C. The theoretical maximum humidified state of the air
leaving is 48.9°C at 100% humidity. This cannot be obtained as actual performance.
There will be a temperature difference of several degrees: the smaller and cheaper the
tower, the greater this temperature differential. Also it is not possible to achieve a fully
saturated state on the outgoing air. Assume here a 0.0* differential and 90% outgoing
humidity.
182 INTERNAL COMBUSTION ENGINE POWER PLANT
From a psychrometric chart (Appendix) at 29.4® dry bulb and 21.1® wet bulb tem¬
peratures SHi = 0.0123 kg and = 79098 j/kg. Since the final state falls out¬
side the limits of this chart, the following equations are provided.

SH = .00625 RH (—^—) kgper kgdry air (6-19)

\Pa p./
h = 1013.4 -(- 18016.7 4- SH x per kg dry air (6-20)

where fd ®= Dry bulb temperature, ®C.

RH = Percent relative humidity,
p, * Saturation pressure of water vapor at U.
Pa = Atmospheric pressure.
Ag = Enthalpy at fa, dry and saturated, J/kg
Substitute in Eq 6-19,
0 0895
SH2 =? 0.00626 X 90 X = 0.0535 kg.

ht = 1013.4 X 43.3 f- 18015.7 + 0.053.1 x 2580140 = 200933 J/kg.

Now Eqs 6-17 and 6-18 are set up.

Eq (6-17) Aia = 0.0535 — 0.0123 = 0.0412 kg make-up per kg air
Eq (6-18) 79098 + 204730 -f 65510 x 0.0412 = 200033 4- 134986 v>^

from which = 1,7 kg water per kg dry air flowing.

Since 866 kg water are needed per min, the air flow is 856 /1.7 ~ 503 kg per min.*
The specific volume of air at 29.4'^and tu, =21.r is 0 862 m’/kg hence the air flow
is 503 x 0.862 » 433.6 mS/min .
Makeup water is 503 x 0.0412 = 20.7 kgper min.
A performance coefficient for evaporative coolers, called “cooling efficiency,”
is given by the following equation.

actual cooling
Cooling efficiency =
theoretical cooling range

Cooling efficiency = (6-21)

where = Atmospheric wet bulb temperature; other symbols as previously used.

The average cooling efficiency of spray ponds is approximately 60%; of
natural draft cooling towers, 60%; and of forced draft cooling towers, 70%.
Manufacturers base their efficiency guarantees on winds of 3 to 5 mph normal
to the long axis.
The efficiency of the apparatus in the sample example is (48.9- 32.2)/(48.9—
21.Dor 60%.
The practical cooling range fa- 4is5.6-16.7®C.Fanpoweri8typically3.75 hp
per 1000 water cooled. The value of to be used in design needs to be a
compromise between the highest for the vicinity, and some lesser value that will
866(48.9 -32.2)4.19
'^Notethatina dry cooler theminimuni air requirements would have been
1 (66*6 - 20.4)
= 1663 kg per min.
TBSnNQ PLANT PERFORMANCE 188
^ve a less expensive cooling tower. One authority reconunends that tw be
selected as that recorded wet bulb temperature which will not be exceeded for
more than 200-300 hr per annum. The differential fb — fw is also important to
cooling tower size and cost. The smaller this difference, the larger the tower.
The economic range is 10®-20“ F.
Mon

Foster Wheeler Co,

Fig. 6-30 Low-head forced-draft cooling tower.

6-8 Testing Plant Performance. The operation of a Diesel plant is a

routine of supervision, inspections of the equipment, and readings of pressures,
temperatures, and electrical data. Verification of normal operating conditions is
achieved mostly by watching pressure and temperature instruments. Faults
and impending troubles usually show up by abnormal temperatures. If these
are detected soon enough, some may be corrected without interruption of
service; in other cases damage may be avoided by a preventive shut-down.
The Diesel is an all-metal machine enclosing fiery gas without benefit of
refractory protection; hence its temperatures are most important. It is critical
to the malfimctioning of any of its auxiliaries; hence indicating thermometers,
preraure gauges, and automatic warning signals for dangerous conditiims idiould
be liberally, albeit intelligently, applied by the plant designer.
From time to time the operating staff may wish to make tests of individual
elements of the plant for the purpose of determining their condition or checking
performance after some adjustments have been made. The acceptance of newly
purchased equipment is customarily based on tests which se^ to discover
whether equipment lives up to claims made for it, or substantiates contractual
guaimitees offered by seller to purchaser. The ASME Test CJode forms a con¬
venient, carefully worded set of directions for such tests. This code is too vo-
184 INTERNAL COMBUSTION ENGINE POWER PLANT
luminous to be exhibited here, but the following may be said of it. The Code
prescribes tests with appropriate preliminary calibrations and agreements on
tolerances, methods, etc. It covers, of course, the determination of specific
thermal performance, but also leads to secondary testing which may be of even
greater importance to operators, viz., adequacy of speed control, cyclic irregu¬
larity, and governor characteristics. The Diesel Engine Manufacturers’ As¬
sociation also has a suggested Field Test Code which is often used as the basis
of acceptance tests of stationary engines.
Diesel engines for power plants are usually equipped for indicator testing
of the cycle. Electronic indicators are required for high speed. They are also
good for low speed, but there the mechanical indicators are practical, as well
as much simpler and cheaper. For speeds up to 360 rpm the pressure-volume
type is satisfactory. For higher speeds, up to 1000 rpm, the mechanical pressure¬
time indicator is used. High-speed electronic indicators also give pressure-time
data. This is of limited value for cycle analysis; however, it can be converted
to the p~v type by graphical means, then used in the customary manner for
determining mean effective pressure, power, valve action, etc.
Example 1: Given the engine p-r diagram of Fig. 6-31 A, to construct a p-v
diagram. The Rod .-Crank ratio of this engine is 4.
1. Locate TDC (Top Dead Center) of the successive cycles and divide the time
axis between them into a suitable even number of parts.
2. Select a suitable length for the p-v diagram and, using it as a diameter, draw a
semicircle (Fig. B).
3. Divide the arc of this semicircle into half the number of parts used in sub¬
dividing the p-r diagram.
4. From these points of division swing arcs which will locate the horizontal position
of the ordinates upon which pressures are laid off. The radius R is the semicircle
radius multiplied by the Rod:Crank ratio. Finally transfer the ordinates from the p-r
diagram and draw in the p-v outline.

The Diesel indicator diagram, such as that shown in Fig. 6-32, has compres¬
sion and expansion lines roughly polytropic in nature. The average value of
the exponent n for the expansion or compression curve is sometimes wanted for
the purpose of analyzing part of the engine performance. Theoretically the
same n should exist all along each curve. Actually heat transfer conditions

■o

Fiq. 6-31 Transformation of indicator diagram from pressure-time to pressure-volume.

TESTING PLANT PERFORMANCE 185
between gas and cylinder walls vary during the stroke and the polytropic ex¬
ponent fails to be constant. The best way of determining the average value is
to plot the curve on logarithmic axes, as is done in the illustration. Then a
straight line is averaged through the plotted points. The tangent of this line is
n. In the example used, the tangent
was measured and found to be 1.32.
Hence the expansion line is: =
C.
Performance testing includes the
taking of many data, much of which
is '^just for the record” and is not
subsequently employed in calcula¬
tions of eflSciency and heat balance.
Data pertinent to determining en¬
ergy distribution are those which
implement calculations for heat bal¬
ance. These are power, flows of fuel
coolant, and exhaust gas, tempera¬
tures of these flows, and quality of
the fuel. Engine builders are able to
make shop tests for bhp with dy¬
namometers direct-coupled to the
Fig. 6-32 Indicator card of 10'' X 15" two-
engines, but when these are plant- cycle Diesel engine. Inset: Logarithmic
installed they are part of an engine- plot of the expansion curve.
generator unit and not ordinarily
open to direct measurement of shaft power. The generator output is measurable
by electrical instruments, leaving bhp to be determined (1) by use of a gen¬
erator efficiency or (2) by adding to generated output the individual generator
losses, e.g.:
1. PR losses of generator stator and rotor coils.
2. Core losses.
3. Stray load losses.
Table 6-3. STANDARD DEDUCTIONS *
(Engine Generator Efficiencies)
DEMA ^^Standard Practices”
Deductions from NEMA
Generator Efficiencies
FuU Load NEMA Full Three-Quarter Half
Generator Efficiency Load Load Load
88.1 to 89 4.0 5.1 8.0
89.1 to 90 3.5 4.6 7.2
90.1 to 91 3.1 4.0 6.3
91.1 to 92 2.6 3.4 5.4
92.1 to 93 2.1 2.8 4.6
93.1 to 94 1.7 2.2 3.7
94.1 to 95 1.2 1.8 2.8
95.1 to 96 0.8 1.1 2.0
♦See Appendix for NEMA Standard Efficiencies. Fig. A-16.
 186 INTERNAL COMBUSTION ENGINE POWER PLANT
4. Exciter losses (in^plies engine-driven exciter).
5. Field rheostat losses.
6. Bearing friction and windage losses.
7. Cable losses (leads from generator to switchboard).
These may be determined by electrical testing in advance of engine testing;
however, if the generator is built to trade standards, the NEMA standard
generator eflSciencies are used since they are correct within the limits of ac¬
curacy of field tests. However, these standard eflSciencies do not include items
4-7 of the foregoing list; consequently from them should be deducted the al¬
lowances shown in Table 6-3, which are an estimate of these items.
If a test is made at constant load for t min, the watt-hour meter readings
being Wi and W2 at the start and end of the test, then

Generated Power = {W2 — Wi) X -j

Calculations of power and eflSciency were treated in Sec 6-4.

Heat distribution is determined by calculating all items except the “friction.

■aciM
OoMrator
Hi■

n
Kilowatt
Type Ratiat at Net
4S0M.iC. W^t A B D K
(IjfJ m m mi
m m H >n
FB>13S 345
S.7IS soil 1 a-f1 tSffl en 0-61?
vaiae 430
BB5 i-W 44*0 03041la 4 1 l-BTZ Bt o«W
n-137 490 BUB! 6.359 m BSi 0-305- I^Q 1-372 EQ 0-6W
iMt &.o3a O'3o«'! 39 BS 1-37^ ra 0«<?
f8-13f 560
s
fgEi 0-9 O-W
. rt>1354C 525

Fl-136^ 1 63S
\ui7tt C-SOL b-oo ♦■«oj mtsss^ I17Z 1 104 oTg.^
m-137^
2e<isc •7.010 9^ tio* [o»3oy 1 [Em x m 1 mi 0794^
n-iMC 1 tso 30m
Fig. 6-33 Dimensions of Nordberg 4-cycle unsupercharged and supercharged engines.
5, 6, 7, and 8 cylinders,S>,02f»ibore X4ieistroika. ^ '
TESTING PLANT PERFORMANCE 187
radiation, and unaccounted” losses, then finding these by the difference between
heating value of the fuel and the sum of all other items.
Cooling loss is found by testing for quantity of cooling water flow and its
rise of temperature. Piston coolant, if employed, should be included, but not
lube oil cooling, as that is friction energy. Exhaust loss includes heat of the
weiit form, as well as the latent heat of water vapor in the products, which may
be taken as the difference between Qh and Ql* The quantity flowing may be
determined from an orifice box attached to the air inlet, or from exhaust gas
measurements which, coupled with fuel analysis, can give the air-fuel ratio.*
This ratio, used with the measured rate of fuel consumption, gives the mass flow
to the engine which necessarily is also that of the exhaust gas.
Fuel measurements for acceptance tests are required to be made by direct
weighing. Temperature and gravity of the fuel oil should be taken frequently
during a test.
Example 2: A Diesel-engine power plant was given a test of 45 min duration,
with results here given. The various performance quantities that these data fix will be
calculated. The engine is 8-cylinder, 2-cycIe, 37.47 om x 46.72 cm X 267 rpm, direct-
connected to a 760-kw generator. Electrical output was read by watthour meter, Wi
7562 kw hr, W2 8087 kw hr. Fuel quantity was measured by direct weighing,!S8.7 kg
for the run. Water flow by Venturi meter 1136 l/min in at 36®C , out at 46.1*1! . Average
fuel temperature, 23.0K!; S.G., 0.921. Atmospheric temperature, 1B.3*C ; exhaust gas,
398.9^0. Average mean effective pressure of indicator cards taken during t^, 6.73kg/om3
Power used for plant pumps, 4.0 kw. During the test exhaust gas analysis determined
that the Air-Fuel ratio was 26.62.
Power CcdculaHons:
Generator output == (8087 — 7562) X 60/45 « 700 kw
This is 93.4% of rated load. NEMA efficiency is 94.0%. Correction from Table 6-3 is
1.4%; net efficiency « 92.6%.
bhp = 700/(0.926 X 0.736) = 1026 hp

6.73 X 46.72 x X 267 X 8

ihp = = 1371 hp
449702
Fuel Oil Calculations:

Correction factor on S.G. of fuel oil = ■:—TTTilirrr::-1 = 1.006

1 —0.00072(23.9 >16.6)
S.G. atl6.6yi5.6®« 0.921 X 1.006 = 0.927
From Eq 6-8, Qh = ®1716 - 8793.8 x 0.927* 44169 J/g
From Eq 6-9, = 26 - 15 X 0.927 = 12.1%
Calculate Ql 44169 - 9 x 0.121 x2442.7 - 41499 J/g
Rate of fuel consumption = 168.7X =* 224.9 kgper hr
Wi - 224.9/1372 -* 0.164 kg per ihp hr
tOh - 224.9/1026 » 0.219 kg per bhp hr

These calculations, equivalent to those for furnace combustion as set forth in Chap¬
ter 6, are not duplicated here.
 188 INTERNAL COMBUSTION ENGINE POWER PLANT
Efficiency CaleidaHoM:
From Eq 6-13, i}ti 1648/(0.164 x 44100) » 86.6% based on Qh
From Eq 6-14, i;tb = 2648/(0.829 x 44150) « 27.5% based on Qt
Plant efficiency, ijp *= (700 — 4) x 8.6 x 104/(224.9 x 104 x 44150) ** 25Jt%
Engine mechanical efficiency, rim = 1026/1818 s 74.8%

ENGINE HEAT BALANCE—^DISTRIBUTION OP HEAT ENERQT ^

Item koal^ %
1. Ueelhl outpnt : 1626 x 682.4 648,840 27.6
2. Cooling low: 1136 x 60 x 104(46.1 ^ 86) 766,600 81.8
Exhaust loss : (Gas flow = 224.9 x 26.62 4- 284.9
« 6211.7 kg per hr)
wc A* heat, 6211.7 x 0.266(898.9-*18.8) = 002,710
Latent heat, 224.9(44150 - 41490) x O.OSOf- 143JMO

746,270 746,870
4. Friotion, radiation and unaooounted for (diff) 216,900 04

Total x0.Si9t il,SM.aoo lao

PROBLEMS
1. Plot the cycle analyzed in Example 1, Sec 6-2, to scales of Icm^Skg/cm^ifandlQm
« 0.001*iii4 Calculate a similar cycle having same initial state, same r, and same
temperature at point 3, but with n = 1.35. Then superimpose a plot of this cycle on the
original example for comparison of pressures and areas.
2. A hypothetical Diesel cycle is to be plotted as was done in Prob. 6-1 for data
as follows: Pi, 0.9 kg/cm* ab; compression pressure, 88.7 kg/om* ab ; tj, 15.6’*C; R, 2J6if
n, 135; 85 om x 38 cm cylinders.
3. Using Prob. 6-2 data, calculate (1) ideal thermal efficiency (with n = 1.35 in
place of y), (2) mean effective pressure, (3) power from a 6-cylinder, 4-cycle, 327-rpm
engine operating on this cycle.
4. An ideal air standard Diesel cycle (except n = 1.35) begins with air at 4<rC,
supercharged to 1.2 kg/om4 ah, and attainsanefficieney of 47%. With the help of Fig '
6-5. determine compression ratio, minimum permissible suction air temperature, and
maximum temperature of the cycle for operation with a “margin of assurance’’ of
ignition of 83.3°C
5. Plot the air standard Diesel cycle beginning with pj 1,02 kg/omt ab, 32.2*D, Vi
0.028 m4. Max. cycle temperature, 1927^0; r, 15.Scales: lom^ 3 kg/om4 icm == 0.002m4
^d work done per cycle, and i^.
6. The specific gravity of oil tested at 29.4^0 is 0.852. Find its API degrees.
7. How many heat units are there in a standard 16.6* litre of fuel oil? When this
was tested at 6.6®C ,a |5.6®hydrometer sank to the 0.896-specific gravity level.
8. Estimate the Ql of 18® API fuel oil.
9. Using median data from Fig. 6-15, find the oil storage volume, in» , needed for
a two-week supply of 25® API fuel oil to operate a 1014 hp engine 70% of the time at
full load, 10% at % load. It is idle 20% of the time.
10. The load curve given in Prob. 2-13 is to be- considered typical for a plant con-
* This is jCp for the 18.8-399**C range for typical exhaust gas. Method of estimating
average Cp of products of combustion is given in Chapter 5.
@ Heat balance worked out in koal/hr.
t Multiidier for conveiting J/g to kcal/kg.
PROBLEMS 189
taining two 350-kw and one 600-kw Diesel engine generators. With help from Fig. 6-15,
estimate the litceB minimum daily consumption of 20*" API oil. Compute the average
daily thermal efficiency.
11. Find the higher and lower heating value per litre of distillate fuel oil of some
assumed, or assigned, API degrees.
12. An engine on test developed bhp steadily for 30 min during which it con¬
sumed 15.84 kg fuel. This same fuel when tested at 28.9®Cwith.a 15.6* hydrometer, showed
S.G* of 0.905. Calculate thermal efficiency, based on bhp and Q^.
13. Write "the ideal combustion reaction of fuel oil considered as C^eHao; a^iso the
reaction with sufficient air to yield A:F ratio of 40:1.
14. Find the percent excess air represented by 32:1 A-F ratio with Cetane as the
fuel.
15. The air flow to aDiesel engine is measured as 87.66jn>/fiilii at is.S^C 1.02 kg/om* ab
at a time when the fuel flow is 1*02 kgper min. Assume that the fuel is CieHsoand deteiv
mine the percent excess air.
16. Assume that the cycle analyzed in Ex 1, Sec 6-2, is produced with a fuel of
Qj^ z= 41870 J/g per lb. Assuming Cp = 1.06 J/g®C, find A:F ratio.
17. What ideal thermal efficiency is possible from a Diesel having r = 15, A:F =
40:1, Ql = 42800 J/g? Take Cp = 1.06 J/^C, = 16.6®C.
18. A 89.87 om x 66.88 om X 327-rpm, 16-cylinder, 4-cycle stationary Diesel engine
is connected to a 3125-Kva (80% power factor) generator. It also drives a 30-kw
exciter. Assume a gefficiency of 92% and determine bmep at rated load.
19. Specifications of a 4-cylinder, 4-cycle gas engine are 10.16omxl6.24om 900 rpm.
Mechanical efficiency, 86%. Find the kw output of a direct-connected generator of 90%
efficiency, when the indicated mep is 7.78 kg/om*.
20. An engine-type generator with 30 poles generates 3-phase, 60-cycle current at
600 V. Its rated output is 4(X) kw. Engine specifications: 40.64oinx 60.8cm. Find the brake
mep existing when line current is 360 amp; power factor, 0.8.
21. Estimate the electrical power output ot a unit consisting of a 1270-bhp, 277-rpm
Diesel engine direct-connected to a 60-cycle, 2400-y, 0.8 power factor generator.
22. Make a scaled layout, plan and elevation, of a Diesel power plant as assigned
or, alternately, to carry the load described in Prob. 2-13. No auxiliaries need be shown.
Proportion engine foundations in accordance with principles of Chapter 4. Use informa¬
tion from Figs. 19, 25, and 33 of this chapter.
23; Diagram a fuel oil supply system suitable for the plant of Prob. 6-22, and
determine (1) storage tank requirements; (2) day tank sizes based on 6 hr running
capacity at 50% use factor. Monthly deliveries by trailer truck.
24. Find lytb and bmep of an 8-cyIinder, 4-cycle Diesel engine which is direct-con¬
nected to a 2300-v, 3-phaM generator rated at 1250 kw. Engine specifications are 89.87 om
x66.88om X 327 rpm. During a test this imit used 269.6 kg fuel oil of 44660 J/g heat¬
ing value. Wattmeter readings at the start and finish ot the one-hour test were 25,156
and 26,378 kw. For electrical efficiencies use NEMA data, corrected by Tabfe 6-3.
25. Diagram the three fuel supply systems described in Sec 6-6.
26. Estimate the fuel storage tank capacity for a Diesel plant having 5000 kw
installed capacity. Expected plant capacity factor, 55%. Fuel contract is to be made for
smi-monthly delivery, but allow 100% extra for contingency. Also determine the gpm
transfer pump capacity.
27. Find the necessary litres capacity of a day tank large enough for 4-hr full-load
operation of the engine described in Prob. 6-24. Assume an average fuel rate from Fig.
6-15.
28. Diagram a fuel supply system for a plant having one engine with supply from
day tank, two storage tanks, fuel delivery from tank truck. Include all necessary
meters, filters, pumps, etc.
 190 INTERNAL COMBUSTION ENGINE POWER PLANT
29. Cooling water for a 607 hp Diesel engine is pumped to a cooling tower at 00*
C. It is desired to cool the water to a maximum temperature of 37.8° under an atmos¬
phere condition *of 82.2° dry bulb temperature, 27.8° wet bulb temperature. Find re¬
quired capacity and efficiency of the cooling tower.
30. A Diesel plant has a cooling system employing a cooling tower that loses to the
atmosphere approximately 6% of the wafer circulating. When the two 507-hp engines
are operated at full load on an average day the tower cools the water from 54.4' to 35°.
What should the capacity of a water softening plant for the makeup be, in l/min -
31. Diagram the water cooling system of the plant of Prob. 6-22. Include forced
draft cooling tower, shell-and-tube heat exchanger, raw water softener. Determine I/min
flow in various parts of the system. Tower cooling efficiency, 70%. Design wet-bulb
temperature,22.2°0. Engine cooling water discharge, 65.6°C ; inlet, 43.3°C . Tower inlet
water 6(fC.
32. Repeat Prob. 31 for a plant containing one 600-hp engine.
33. A cooling tower, needed in conjunction with an I.C. engine plant, will be re¬
quired to cool 681 l/min water at 65.6°C entering temperature. Atmospheric pressure,
1.03kg/om»; dry-bulb, 27.8°C; wet-bulb, 22.2°C. Forced draft tower with 68% cooling
efficiency. Calculate make-up (15.6°C) flow, air flow, and estimate fan power. Assume
air out of tower at 60°C , 90% humidity.
34. Find l/min loss of water due to evaporatively cooling an inlet flow ofi704]/mir.
water in a cooling tower from 60°C to 48.1°C. Atmospheric dry-bulb, 26.7°C; wet-bulb,
21.rc. Air discharge at 67.2°C ; 95% humidity. No make-up.
35. The p-r diagram of a 2-cycle Diesel engine is given herewith by coordinates.
This engine has 4.5 R-C ratio. Construct its p-v cycle grajihically.

T-sec 0 0.2 0.5 0.8 1.0 1.2 1.4 1.6 1.64 1.68 1.7 1.8 1.9 1.96

p-kg/om< ah 1.09 1.23 1.79 2.95 3.59 6.68 13.15 25.31 29.88 33.4 33.75 33.04 31.71 30.94
T-sec 2.0 2.0 2.1 2.3 2.6 2.9 3.0 3.1 3.3 3.33

p kg/ems ab 28.1218.9811.25 7.03 4.92 4.08 3.16 1.41. 1.09

36. Solve Prob. 34 altered to include makeup of evai)oration losses with water at
2i.rc.
37. The following are coordinates taken from the expansion portion of an I.C engine
cycle. By means of logarithmic plotting find the average polytropic exponent r?.
p, kg/om<ab 35.15 28.12 18.98 7.03 3.52 1.76
V, m* 0.00255 0.00340 0.00425 0.00878 0.01501 0.02832

38. Analyze an assigned indicator diagram for the average polytropic n for both
expansion and compression curves.
39. The shop test of an engine whose record is here reported was made for heat
balance. Calculate a four-item balance. Dynamometer torque, 651 kg-m at 450 rpm.
A:F ratio, 27:1. Speed, 450 rpm. Temperatures: Atmosphere, 26.7°C exhaust gas,464.4°
C; coolant in, 65°C ; out, 66.7°C ; fuel,28.9°C Fuel flow, 87 kg per hr; coolant flow (c
= 1), 98 gpm. Hydrometer test of fuel oil, 0.945 S.G.
40. The following readings were taken during a 60-min test at steady load of 300-
rpm Diesel engine—generator unit rated at 650 kw. Initial fuel meter reading, 14.0154
m4 final. 14.1811 m* . Initial watthour meter, 86,547 fcw hr; final, 87,002 kw hr.
Average temperatures: atmosphere, 4.4°C ; fuel, 16.6°C; exhaust gas, 416.7®0; cooling
water in, 46.1°C; out, 6l.l°C . Water flow, 549 l/min S.G. of fuel by hydrometer, 0.905.
A:F ratio, 22:1. Calculate a five-element heat balance, including electrical losses as one
item. Use NEMA efficiency data and Table 6-3.
 CHAPTER 7

7-1 The Gas Turbine. This type of prime mover derives its energy from
heat, commonly supplied by combustion. The products of combustion form
the working medium,* but the combustion region is external to the prime mover.
Hence the gas turbine is removed from the internal combustion machines of
Chapter 6, as well as from the external combustion cycles of Chapter 8 whose
working media are not the products of combustion. •
The industrially successful gas turbine power plant is relatively a newcomer
to the power field. Although gas turbines have been the subject of experimental
development for decades, only in recent years have there been gas turbine plants
of performance and cost that are commercially acceptable. The obstacle to
success resided principally in the condition that the plant uses approximately
70% of its prime mover output internally and, if all processes are not highly
efficient, the losses easily consume the potential net output, leaving the machine
unattractively inefficient. Early experimental plants sometimes failed to pro¬
duce any net output. Furthermore, good efficiency of the cycle used demands
high initial temperatures of the working medium. The manufacturer is con¬
fronted with a series of high-temperature problems of difficult solution. Com¬
mercial development followed when science and engineering had progressed
to where*
1. Manufacturers would build high-speed turbines for use with gases hot
enough to be incandescent ( 538®C and up).
2. The theory of blade action in turbines had developed to the point where
as much as 80% of the theoretically available energy could be converted into
shaft work.
3. High-speed air compressors of about 80% compression efficiency or better
could be built.
The aeronautical jet engine is a specialized gas turbine plant whose net
output is a kinetic propulsion jet of the working medium rather than shaft
work. The high-performance aircraft which could be made possible by an ex¬
tremely compact prime mover for jet propulsion interested governments in
expending large sums for development, much of which was to perfect the solu¬
tion of the three problems mentioned above. Some of the resulting know-how
has been beneficially applied to the manufacture and use of stationary gas
* This is not an absolute requirement of gas turbine plants, as will be shown.
191
192 GAS TURBINE POWER PLANT
turbine power plants. Although currently few in number, the prospects of in¬
creasing use of gas turbines in this field are considered sufficiently bright to
devote a short chapter to this type of stationary power plant.
At this point it is well to review the meaning of working process by reference
to Sec 1-3, for the critical action in any gas turbine plant is this process per¬
formed in the turbine. A reversible adiabatic expansion of a gas is theoretically
possible if the gas is made to pass through a properly shaped nozzle from a
region of high to one of low pressure. The ratio of high to low pressure will be
called the pressure ratio, whether the process is an expansion or a compression.
Actual expansions are almost reversible, there being but little friction or turbu¬
lence generated by nozzle action. The result of this process is a jet of gases con¬
taining high-grade mechanical (kinetic) energy. Turbine wheels have blades
onto which the jet is directed in order that this kinetic energy may be trans¬
formed into a blade thrust and thence to a shaft torque.

NOZZLl
A. SHAFT LOOEtt
NO MADE
MOTION,
NO ASSORmOH
OF JETS
KINETIC ENOCY

BLADE

HIGH BLADE
SPEED,
low RESIDUAL
JET VELOCITY
AND ENttCY

Fig. 7-1 Principle of the gas turbine.

In the gas turbine a stationary nozzle discharges a jet of gas (usually prod¬
ucts of combustion) against the blades on the periphery of a turbine wheel,
as shown in Fig. 7-1. The jet is thereby deflected and slowed while the blades
receive an impulse force which is transmitted as a mechanical torque to the
shaft. The prospective jet speed is sometimes, sufficiently high to warrant divid¬
ing the expansion into a series of stages with a set of nozzles and a row of blades
in each stage, all blade wheels being mounted on the same shaft. By limiting
the gas expansion per stage, the blade speed and rpm of the shaft are suitably
decreased. Were the blades themselves so shaped as to be virtual nozzles, some
expansion would also take place in the gas as it went through the blading.
The latter would, in consequence, receive a ^'reaction thrust" distinct from im¬
pulse action. Many gas turbine designs have employed the reaction principle.
The heat made available for conversion to work during an isentropic work¬
ing expansion is
AA == CpAT (7-1)
the gas turbine 193
in which ■■ Mean specific heat of gases for the temperature range of the
expansi<m, J/g**C<
AT •= Temperature drop, *C.
Since gas turbines work with a high-temperature fluid medium, the best ex¬
perimental data should be used to establish the value of the specific heat.
Many special charts and tables are to be found in the literature of gas turbine
engineering. Here we will present only instantaneous specific heats derived from
good spectroeeopie data, these being for the products of combustion of an as-

l-Mi •S2
l>32
-4 *1
IIIIIIICirafQIEEESaSiSSi
vjuuiMiMMmiiiii^siKsq 4 *30

•2f

•26 *

UJJJJ o
UU"' H *27 «

UTdyA^jZL} JUL H *26

. irm
:;^if5eaiLa__ a ULUL • 25

[H?^«1LLLJLLLLUJ LIJ LLLl

^JJJ_LJLL1JLLI±U
I'OOi
u:_I LI L •24

,,mTrrrnnnnrnTnniTrrnrm
I 400 600 600 1000 1200 UOO 160
16001
•23

275 . 11665
temperature, K

Fig. 7-2 Constant pressure specific heat for air and combustion products.

sumed liquid hydrocarbon (C12H26) with various fuel-air ratios. A mean Cp

value is obtainable from Fig. 7-2 to any required degree of accuracy.

Bacample 1: The available energy of an isentropic expansion of products of com¬

bustion of kerosene, considered to be Ci2H2s, will he computed. Fuel-air ratio, 0.015;
initial temperature, lOCO^’K; pressure ratio, 6.
By a trial solution it is guessed that the final temperature will be about S11*K.
Fig. 7-2 gives Cp « 1.177 at 1000”; 1.080 at 611”. An arithmetical average* is ^ « 1.188.
The molecular weight of air is usually given as 28.97. Air-fuel ratios used with the gas
turbine are lean in order to hold the temperature to that allowed by the turbine de»gn;
hence, the product’s molecular weight is but little more than air’s. Taking 29.0 as the
molecular weight, mol specific heat Cp » 89.0 x 1.188 »» 88.718 J per mol-^. From
* More accurate values of Cf are obtained by giving due consideration to the curvature
of the graph through the estimated temperature range.
194 GAS TURBINE POWER PLANT
the kinetic theory of gases, Cp ^ Cr ^ 8.845 J per mol (independent of temperature).
Hence Cr « 84.867 per mol and y » 82.718/84.867«!.341. Now with y 1.341 and
Sp » 1.188 the following calculations lead to the desired available energy.
ami
Ti/Tt « (pi/pt) y = « 1.576
Since Ti *= lOOO^K, Tj «1000 /1.576 « 684.5®K.
Although the Cp value is based on an 611^ estimate of T2, a recalculation is not
considered necessary in this example.
AT « 1000 ^ 684.5 » 865.5
From Eq 7-1,
Ah « 1.128 X 865.5 » 418.8 J per g of gas available energy
The turbine efficiency* r/r is taken to be the ratio of actual work output of
the turbine to the work obtainable from a reversible adiabatic process such as
that investigated in Example 1. The inability of a gas turbine to convert all
the ideal energy into work is mainly attributable to the blading, with its leakage
through clearance spaces, friction, irreversible .turbulence, and residual gas
velocity. If a turbine were to operate under the conditions of Ex 1 at an effi¬
ciency of 80%, the gas flow required per hp hr would be 2.648/(412.3 x 0.80) or
8.03 kg.The compressed air required would be8.03/1.015 = 7.91 kg per hp hr.
It will be found later that the air rate of a, pas turbine plant is much higher
because it is the air per net hp hr, and the net power is approximately a third
of the turbine power. The remainder is used to compress the air for combustion.
Ideal reversible compressions and expansiong emphasize the convertibility
of velocity and thermal state. For example, the equation

V.. _ r,..
shows that ideal reversible nozzle flow from pi, Ti, to pa produces the velocity
increase Vi to F2. Similarly, a reversible velocity diffusion from V2 to Vi would
produce a pressure increment Ap = pi — p2 and a temperature rise AT (= Ti
— T2). Thus, we have “total” quantities, which are the sum of the static quanti¬
ties and the increments representing the velocity. Using prefix t to designate a
total quantity,

th = h+ F*/2gJ (7-3)

tP = P (1 + ^ (7-4)

(7-5)

The symbol M represents Mach Number, a velocity descriptive term defined

as follows:

*Noie that this is not a basic thermal efficiency, but is more like ^^engine efficiency”
(see p. 367).
CONSTANT PRESSURE COMBUSTION CYCLE 195
M = V/VygRT (7-6)

It relates the gas velocity to the velocity of sound, which is VygRT,

In the internal combustion engine and vapor cycles the fluid velocities are
usually low enough to render negligible any difference existing between total
and static quantities. However, this is not the case everywhere in the gas tur¬
bine plant, and exact analyses may need to be based on total quantities.
7-2 Constant Pressure Combustion Cycle. It has been seen that the gas
turbine must receive a continuous supply of gas under a pressure above that of
the exhaust region. A gas turbine plant consists of the turbine and a means of
supply of the working medium. Modem gas turbine plants are based on a cycle
of continuous flow of the medium. Air is compressed in a machine driven by the
turbine. The compressed air then goes to a combustion chamber where fuel is
burned in it, raising the temperature and increasing the volume at constant
pressure. The working medium is then ready for the turbine. After the working
expansion the gas is discharged to the
atmosphere. A closed cycle is not pos¬
sible if the fuel is burned directly in the
air, but is if surface heat exchangers arc
used. We shall consider first the direct
combustion or *^open” cycle.
Referring to Fig. 7-3, air existing at
state 1 in the atmosphere is taken into
the compressor and compressed to state
2, where it is ready for delivery from
the compressor to the combustion cham¬
ber. The compression work is repre¬
sented by the area al26. The fuel-air Fir,. 7-3 Open cycle of the simple gas
ratio used is such as to produce the tem¬ turbine plant.
perature Tsy which is the limiting high
temperature of the cycle. Currently the upper limits run between 649®C and
816®C in stationary plant practice. The products of combustion^ now at
Ps and Ts, are expanded in the turbine to state 4, which is the discharge state.
Ideally ps = P2, Pa = Pi, and the processes 1-2 and 3-4 are isentropic.
Different types of compressors have been used in gas turbine plants (i.e.,
centrifugal, positive displacement, rotary), but the axial flow type is favored
for stationary plants because of its efficiency and inherent high rotative speed.
This is a multi-stage, axial-flow compressor, resembling in action a reversed
turbine. Fig. 7-4 shows a plant with a turbo-compressor. Air flows over a set
of airfoils* (arranged circumferentially as blading). The airfoil blades turn the
air stream through an angle. A diffusion thus effected slows down the air ve¬
locity and increases pressure. If air speeds remain nearly constant, then blade
heights can diminish in succeeding states (as witnessed in Fig. 7-4) 'owing to
increasing density of the compressed air. Operating the blades at high angles
of attack on the air helps build up pressure rapidly, but operation near the
stalling angle would be undesirable since small variations might occur which
could burble the airfoils and cause an unstable, rough, or even hazardous condi¬
tion to exist. Turbocompressors may be employed under conditions where ut-
196 GAS TURBINE POWER PLANT
most efficiency is imperative, as in gas turbine power units, and should create
llio o])timum favorable balance between good downwash and minimum turbu¬
lent airfoil wake. They have been built with energy efficiencies as high as 85%.
They may be operated effectively at high speeds, i.e., 5000-10,000 rpm.
In order to produce compressed air the compressor executes a polytropic
compression of the air, followed by delivery at constant pressure. The perform¬
ance of actual compressors is described by compressor efficiency, 70.
_Ideal work of compression_
^ Work actually required for compression
It is also called “adiabatic efficiency” or “internal efficiency.”
It can be shown that the ideal thermal efficiency of the simple gas turbine
cycle is given by the equation:

1 - (7-7)
x=l
ipi/pi) y
From Eq 7-7 one might draAv the conclusion that, for maximum 71, make pi/pi
maximum. But there is a temperature ceiling Ti for the cycle, set by the struc¬
tural limitations of the turbine. This impoees a limit on the pressure ratio that
can be used. Theoretically a pressure ratio of pi/pi — (Ti/Ti)y-^ would produce
the maximum efficiency, but it could not be tolerated because the useful work
of the ideal cycle then vanishes. Furthermore, the optimum pressure ratio
becomes less than (Tz/Ti)y-i when the turbine and compressor are given
realistic efficiencies.
To analyze the simple combustion gas turbine cycle the following nomen¬
clature is employed:

7t, vc = Internal efficiencies of the turbine and compressor.

Qi = Heat added in the combustion chamber, J per g air.
Wtj Wc — Ideal reversible work of the turbine and compressor, kg-m per kg
air flow.
Wn = Net work of cycle,kg-ro per kg air flow.
/ = Fuel-air ratio.
The thermal efficiency of a gas turbine plant is net work done divided by the
heat input to the combustion chamber.

WtvtH +/) - Wc/vc - W,

(7-8^
102*/Q
where Waux = Work to auxiliaries,kg-mpei kg air flow.
Q = Heating value of the fuel, J per g*.
Fuel-air ratios employed are very small, as also is the auxiliary power sup¬
plied directly from the machine shaft. Combustion efficiency with liquid hydro-
* Conaiant will be 426.9 if Qf is in keal per kg.
THE SIMPLE GAS TURBINE PLANT 197
carbon fuel is quite high. This being true, the Eq 7*8 loses little of its meaning if
simplified to

Wr>mc — T^c
(7-9)
HKVQi
To find the ratio pt/pi for this efficioacy to reach its highest value, replace
TTt and Wc with'lOZ^lhr andlQil^Ahc, reiq>ectivdy, but using as an ex¬
pression fw AA. Then
M ~ Tj) — CiiTt — 7*i)
’* eoCB(r, -
Consider c^, Tm, and. Ti constant, and use T4 » Tt(Ti/Tt) to eliminate 7*4,
thereby obtaining
(yrecn - ro(T. - ro
ift
iKj(r.T, - T/)
By performing the operation diit/dTt «« 0, the value of 7s for maximum fix,
when 7*4 is fixed, is found to be contained in
(2i>Tifcri7*i)7*s — (7*1 — (1 — iKJ*rr)r4)7*s* *»* rrmcT^Ti (7-10)
Using 7*s obtained from this equation, the best pressure ratio is

ps/pi « (7*s/7’i)t-i
The closed cycle would be similar in appearance to the p-t; diagram of
Fig. 7-3. A high-temperature surface heat exchanger would take the place of
the combustion chamber. Combustion could be carried out in an external fur¬
nace, using coal if desired, for the products of combustion are kept separate
from the working medium. Likewise, a low-temperature heat exchanger would
accomplish the constant pressure cooling along process 4-1. Being in a com¬
pletely closed unit the gas could be employed at higher pressures, thus reducing
the size of the equipment needed. For example, a pressure ratio of 6 might be
incorporated into a cycle working between pressure limits ofl6«8and2.Bkg/cm*
The gas would not even need to be air, and it is likely that other media might
be found to be superior for the purpose.*
The closed cycle had not received commercial development in the United
States prior to 1950. Partially closed cycles have been built, but are involved
with compound units, a type discussed in Sec 7-4.
7-3 The Simple Gas Turbine Plant. Fig. 7-3 has indicated the princi¬
pal elements needed to produce power in such a plant. Fig. 7-4 shows how the
elements may be assembled compactly in a single machine. The illustration is
that of an open-cycle unit currently being offered commercially. It is rated at
3500 kw when connected through gearing which changes the 6700-rpm turbine
speed to 3600-rpm generator speed. At 704® C to 760^0 maximum gas
temperatures the thermal efficiency is approximately 15%. Atmospheric air is
compressed in the multi-stage turbocompressor at approximately 5.9 pressure
ratio. Air leaving the compressor flows into the multiple combustor (6 chambers)
’^Helium, for example.
* OoDsteot will bo 426,9 if mto in koal/kg.
Fio. 7-6 Rotors of the unit shown in Fig. 7-4.
THE SIMPLE GAS TURBINE PLANT 199
where it bums the fuel necessary to produce the design temperature. Bunker
C (black) oil, pressurized and preheated, is fed in through air atomizer spray
nozzles. Ignition is by spark plugs located in two of the chambers, the other
chambers being ignited through cross tubes connecting them. Combustion is
self-sustaining, and spark ignition is used only during starting. Also, Diesel oil
is the fuel during starting and stopping periods of operation. Primary governor
control is by varying the fuel oil pressure to the nozzles. Secondary controls arc
provided to limit overspeed and maximum temperature. Upon leaving the com¬
bustion chamber the products of combustion are expanded to atmospheric
pressure in a two-stage gas turbine. The turbine and compressor rotors are
shown in Fig. 7-5. An extension of the generator shaft is connected to a 187 kw
electric motor which is needed for starting. Until a critical pressure is developed
by the compressor the unit is not self-energizing. This occurs at about 3400 rpm
in this unit.
Example 1: Some of the operational factors for a simple gas turbine plant will be
analyzed for a case where maximum permissible temperature is .760*^'and a fuel of
43156 J/g. lower heating value will be used. Other data are: atmospheric temperature,
26.7®C riT, 030; np, 0.78; combustion efficiency, 95%. Subscripts employed will refer to
state points on Fig. 7-3.
On the assumption that it is desired to use a pressure ratio that will yield the maxi¬
mum thermal efficiency, Eq 7-10 is first employed to determine T2.

(2 X 0.80 X 0.78 X 1038 x SGOOTj - [ 300 - (1 - 0.80 X 0.78) 10331Ti* *

0.80 X 0.78 X 10389 x 300
Solution of this equation gives T2 being the ideal temperature of com-
1.4
pression of air, the pressure ratio ps/pi *= 466.8/300 TT^ « 4.7. Peak efficiency is not
critical to exact pressure ratio; furthermore, considering the assumptions of Cpc ~
made in deriving Eq 7-10, the computed optimum ratio should be rounded off. Thus, a
pressure ratio of 5 is taken. This gives an ideal T2 of 472.2* which will be used in further
calculations.
The actual Tt is higher on account of compressor losses which add to the enthalpy
of the compressed air.
Real Tt leaving compressor and entering combustion chamber «=
472.2 - 300
300 4 62PK.
0.78
Qt needed to raise products to lOSS^'K « (1033 521) 1.118 «i 572.36 J/g.
Fuel-air ratio - 583.07/(43156 x 0.05) » 0.014.
By means of a preliminary trial it is intimated that the ideal T4 will approximate
OOT’K. Now Fig. 7-2 is used to obtain the mean Cp for expansion. Average Cp * 1.139
for / « 0.014 and temperature range 1033 — 667® K. Then
Cp «= 29 X 1.139 - 33.03 J per mol .®0, and y « 7.88/(7.88 - 2) « 1.34.
(7 - l)/7 » 0.254.
Calculated ideal » 1033/(5)® *“ = 686.4®K.t

* 1.118 = mean Cp, 621® to 1033®C range, for estimated / = 0.015, Fig. 7-2.
t Use of pa/p4 =; Pa/Pi, of course, is a neglect of pressure loss in combustor and exhaust
 200 0A8 TURBINE POWER PLANT
Wt - 101(1 +/)c^r, — r«) - los X 1.014 x i.i» (loas -'6w.4> kg^psr l(g air.
- loa^ (Ta - Ti) -> lot X 1.01S(47t.t - tOO) kg^ p«r kg •&.
Substitute in Eq 7*9 for thermal efficiency:
1.014 X I.IM X t4«.« X 0.M X 0.7S - l.Olt x 17t.t
Vt ---- - 17.8%
0.78 X 672.80
Including combustion losses, plant efficiency . 17.8 x 0.08 — 18.8%
Net output per^air flour. l.014 xi.188 x t4B.6x 0.80xlUi - i.oitx 178.2 x 10*/0>7S
• <880.8 - 881.1)10* . 90.1 x 10* J/kg.
/tir rate, lUi - 8.8M xl0*/00.1 x 10* - 88.7 kg pnr kp hr.
Fraction of turbine power used internally in the plant -i 881.1/880.8 0.801.
The actual exhaust temperature exceeds T4 because turbine leases reheat the gas. Actual
Oxhaust temperature » 1088 — (1088 - 886.4)0.80 ,=1 788‘K.

7-4 Thermal Refinemmt of the Gas Turbine Cycle. Although the

performance of the gas turbine which has been described in the previous three
sections is not especially attractive compared with efficiencies possible in Diesel
and steam power plants, a simple gas turbine has advantages in weight, sise, and
\4bration compared to the engine, and in sise and cost compared to the Bmall

REGENERATOR
TURBINE

ft GAS COOLER
(INTERCOOLER)
COMPRESSOR

ft geared A.C.
SHAFT a. COUPLING ■ff€) GENERATOR

COMBUSTION —© STARTING MOTOR

CHAMBER

Fig. 7-6 Symbols for gas turbine plant flow diagrams.

steam plant. It is also superior to both in quantity of water used, for the simple
gas turbine plant uses almost no cooling water. However, large steam power
stations will be more compact and less expensive than had they been powered
with gas turbines.
The efficiency handicap is surmountable, at the expense of adding complexity
to the gas turbine plant. The principal refinements which accomplish this are
regeneration, intercooling, and reheat.
Regeneration is the transfer of heat energy from exhaust gases to compressed
air flowing between the compressor and combustion chamber. A surface heater
called the “regenerator” is required. Resulting cooler final exhaust gases repre¬
sent a reduction of waste heat.
Intercooling is the removal of heat from compressed air between stages of
compression. This reduces the internal consumption of turbine power by the
plant. Usually water-cooled surface coolers are employed. Ideally the objective
is isothermal compression, but actually plant designers must be with
 THERMAL REFINEMENT OP THE GAS TURBINE CYCLE 201
staging, consisting of adiabatic compressions with intervening cooling.* Only
a few of these are economically practicable. The compressor must necessarily
be compounded, i.e., low- and high-pressure sections separately encased.
Reheating is the increase of temperature of partially expanded gas by burn¬
ing more fuel in it. It necessitates a compounding of the turbine, and a partner¬
ship with regeneration, for it raises the temperature of the turbine exhaust.
It is especially good for improving the efficiency of plants having high pressure
ratio but moderate maximum cycle temperature.

Fig. 7-7 Flow diagram of a regenerative plant having intercooling and reheat. Open
cycle.

These improvements to the simple cycle may be applied separately or to¬

gether. They are capable of raising plant efficiency to over 30%, thereby erasing
any advantage of fuel efficiency possessed by Diesel or condensing steam plants.
The relative location of the equipment of a high-performance gas turbine
plant will be seen in the diagram of Fig. 7-7. The state of a unit quantity
' of the working medium is shown in the next illustration.
The basic simple cycle for pressure ratio P2/P1, and temperatures Ti, Ts,
would have been l-2'-3-4'. The enclosed area would have been net work output,
while heat input would be represented by the constant pressure combustion
process, 2'-3. But with the thermal refinements actually incorporated, the work
area 1-2-3-4 is larger, while the combustion effect, 2x-3 plus 3a-36, may be
less. The result is more power per unit of flow, at greater efficiency.
The actions of the components of this plant are explained with Fig. 7-8 as
follows:
Regeneration, Exhaust gas is cooled in the regenerator from 4 to 4x, deliver¬
ing heat to the compressed air and raising its temperature from 2 to 2x. The
temperature fall of the gas is approximately equal to the temperature rise of
the air. The pressure losses in the regenerator are small and are not shown on
the diagram.
Intercooler. The total pressure rise is divided into two increments (maybe
more in -some plants) by compounding the compressor into low- and high-
pressure sections. Between the two sections the air flows through an intercooler
and is reduced in volume at constant pressure. This action is represented by the
portion la-16 of the cycle. Intercooling is usually accomplished with water-
cooled surface heat exchangers, thus imposing a water requirement on the plant.
* Wet compression accomplished by water spray into the hot air has been suggested. It
would appear that condensate would be required in order to prevent scaling up of the
compressor blades.
202 GAS TURBINE POWER PLANT
Reheat, If the working expansion is carried out in two or more turbine
stages, the turbine may be divided into high-pressure and low-pressure sections
so that the gas may be intermediately reheated to the maximum limit in a
combustion chamber. Because of the low fuel-air ratios, there is adequate
oxygen in the products to support the reheater combustion. The portion of the
cycle representing this reheat action is 3a-3f).

Fig, 7-8 Cyct^ for gas turbine plant with regeneration, intercooling, and r(‘heat. See
Fig. 7-7.

The foregoing indicates that the efficient gas turbine plant must be provided
with heat exchanger equipment, regenerator, intercooler, or both. The principal
requirements may be summarized as follows:
Regenerators Intercoolers
Gas-to-gas heat transfer Gas-to-liquid heat transfer
High temperatures Moderate temperatures
Lower coefficients of heat transfer than Water circuit required
for intercoolers (approx. 44 koal as Less heat to be transferred than for
against 73 kcal per hr m> ) regenerator

The extent to which one fluid is cooled while the other is heated is described
by the effectiveness of a heat exchanger. This is defined as follows:

^_Temperature drop in heat releasing fluid_

“ Maximum temperature difference between the two fluids
where Eex — Heat exchanger effectiveness.
Rather high (say 0.80-0.90) values of Eex are used to design intercoolers,
whereas 0.50 to 0.75 is more common for regenerators. Because of the low
coefficient of transfer in gas-to-gas cases, the regenerator will tend to become
bulky on account of the large surface involved. Both types are of the shell-and-
tube construction. If the pressurized air is inside the tubes, the shell is not
stressed and may be made oval, rectangular, etc. However, often the gas is
sent through the tubes, and the air through the shell. In such cases the shell
THERMAL REFINEMENT OF THE GAS TURBINE CYCLE 203
must be cylindrical to withstand the air pressure. This arrangement makes the
gas side easier to clean, should it become soot-laden.
Example 1: Temperatures of turbine and compressor discharges are and
177^, respectively. Maximum cycle temperature, 816*’C; E^x ~ 0.75; combustion
efficiency 0.05; Ql of fuel, 43270 J/g • With these data the temperatures of the
regenerator and the air-fuel ratio are found in the following manner.
427 — tix
Eex = 0.75 = ^27^ri77* " 289.5®C
Assume/ will be about 0.015. Cp (Fig. 7-2) = 1.08 for exhaust gas.
Heat release perkg air f]oWai(l -f 0.016)(427 ^ 230.6) 1.08*x 10* 205.68 x 10* J«
Air will be heated approximately 427 — 280.6 or 187.5^ estimated f2jc**177-H87.5*'364.5=*
687® F. Cp on the air side of the regenerator may now be estimated.
Cp (Fig. 7-2) « 1.067. At (air) = 205.58x 10»/(l.067X10*) «107.2®C, 177107.2« 374.2®C
For range 374* *816^0, Cp - 1.16 for an estimated / of 0.015.
1.16(816 ^ 374.5)(1+0.015)
Fuel-air ratio, / = - — 0.0126 .
43270 X 0.05
Although Fig. 7-7 has shown all rotating elements of a compounded system
coupled together, the mechanical connection is not necessary; in fact, it is ad¬
vantageous to have separated sections that may be operated at different speeds,
as well as at variable speed for part-load control. When both compressor and
turbine are compounded, this can lead to a number of variations of plant ar¬
rangement. Several plans which have been used are set forth in Fig. 7-9. In an
electric power plant, operations are confronted with the usual condition of

A. CROSS COMPOUNDED PLANT B. EUROPEAN DESIGN OF 27,000 KW COMPOUNDED PLANT

C. IN-LINE DESIGN FOR 5000 KW COMPOUND 0. PLANT WITH DIVIDED GAS FLOW TO TURBINES.
REGENERATIVE PLANT WITH INTERCOOLING TWO SHAFTS GIVE FLEXIBILITY. UNCOMPOUNDEO.

Fig. 7-9 Flow diagrams illustrative of high-efficiency, open-cycle gas turbine plants
(See list of symbols in Fig. 7-6.)
204 GAS TURBINE POWER PLANT
variable load (Chapter 2). However, the a-c generator must be driven at con¬
stant speed. Part-load operations are more efficient if the compressor speed
can be varied; therefore, a two-shaft plant may be expected to be less adversely
affected by part-load operations, and the starting power required is considerably
less with divided shafts. For example, a commercial design of a 3500-kw,
single-shaft machine specifies 253.5 hp in the starting motor, whereas another
of SOOO kw, but a two-shaft design, has specification of 40.6 hpon the low-pressure
compressor shaft and 101.4 hp on the high. Divided shaft plants permit several
advantageous arrangements of the equipment for compound-closed cycle and
semi-closed cycle plants.
7-5 Field for Gas Turbine Power. The important application of gas
turbine power, so far as the average citizen is now concerned, is in the airplane
where the gas turbine plant is used either to supply gas for jet propulsion or to
drive an airscrew, as in the so-called turboprop engine. These are highly special¬
ized and very progressive applications due to their association with military
WASTE PRODUCTS
OF COMBUSTION

Fig. 7-10 Closed-cycle gas turbine plant. Possible thermal efficiency, 32%.
needs. Railway and marine propulsion are also fields of active experimentation
with gas turbines. Some operating experiences have been accumulated with the
gas turbine as a heavy-duty truck engine. But though less known to the public,
the application of the gas turbine to stationary power generation is the oldest
field of successful use and one in which steady progress is being made today.
This is especially interesting since these units must compete actively on a com¬
mercial basis with the highly perfected Diesel and condensing steam power
plants.
The likelihood of many variations from the basic simple cycle of Fig. 7-3
is implied by Sec 7-4. These variations, however, only begin to cover the pos¬
sibilities, for the gas turbine idea is amenable to advantageous combinations
with industrial heat-power needs, with steam plants, with pipe-line gas trans¬
mission, and many others. A few of the cases that could be mentioned are:
1. In place of the regenerator of Fig. 7-9 (C), a waste heat boiler could
generate industrial steam. The gas turbine could operate on an oi>en cycle, as
shown, or on a closed cycle, using as the gas precooler, ahead of the compressor,
a water heater or other low-temperature heat exchanger. Unbalance of heat
and power demand could more readily be met with the open cycle, for then, if
FIELD FOR GAS TURBINE POWER 206
steam demands exceed those which waste heat will supply, an auxiliary oil
burner in the boiler could be used to make up the deficiency.
2. As an auxiliary to a large steam plant, the gas turbine plant could be a
simple unit whose exhaust heat was largely absorbed by heating steam-plant
feedwater, super-heating steam, or supplying steam for low-pressure station
auxiliaries.
3. Several combination steam turbine-gas turbine cycles have been proposed,
all highly interesting from the thermodynamic standpoint.
4. The extensive pipe-line transmission of natural gas now carried on in
this country offers a fertile field for the gas turbine as a drive to high-speed gas
compressors in the intermediate compressor stations required along the pipe
line. The same can be said for oil pipe-line pumping stations. These have been
powered by large reciprocating oil or gas engines.
5. The closed gas turbine cycle may become a suitable vehicle for the
nuclear power plant. Unless a feasible method of a large-scale direct conversion
of nuclear radiation into electricity
is found, any nuclear power scheme >100|
SIMPLE I I *
for electric generation must include I CLOSED
a heat power cycle for the conversion
80
of heat energy into mechanical
work. It would appear that the closed < ^§2=55^1— COMPOUNDED J
s
DC
OPEN evae
gas turbine cycle operating on com¬ 60 I (TWOSHAFTI

pressed helium might fill the need as

r ,
.,
well as the steam plants presently
contemplated for this field. j 40|
3 LfiWi f:
A summary of the comparative REGENERATIVE
OPEN CYCLE
features of the gas turbine as a prime 5 0w
mover includes: S u
u
1. Mechanically it is simple
compared to steam and I.C. engine 20 40
n 60 80 100
plants, but in the endeavor to reach
competitive efficiencies some of this PERCENT RATED LOAD

advantage is lost. Fig. 7-11 Part load operating efficiencies

2. An electric motor or I.C. en¬ of gas turbine power plants.
gine is required to start the gas tur¬
bine plant. As the starter must bring the compressor well up toward operating
speed, starting is not as simple as for I.C. engines.
3. Like steam turbines, the gas turbine is not readily reversible. Steam en¬
gines and two-cycle, I.C. engines are best in this respect.
4. Turbine plants have less vibration than engine plants of similar size, but
high-frequency noise emanating from the compressor intake is likely to be
objectionable.
6. The gas turbine uses high temperatures. Even though the pressures are
moderate, service conditions are complicated by the existence of these high
temperatures.
6. With certain types of compressors, efiSciency of the gas turbine plant is
not as well maintained at part load as with steam or the I.C. engine. However,
m GAS TURBINE POWER PLANT
compounding, with variable speed control on one shaft, is a means of overcom¬
ing this disadvantage.
7. Water consumption favors the gas turbine.
PROBLEMS
(MW « molecular weight)
1. Calculate the work done perkgof gas expanding from 6.33kg/om> ab tol.05 kg/cm*
abin a gas turbine of 82%intemal efficiency.Initial temperature.750®C;7“1.34; MW -29.
2. Products of combustion with y of 1.35, 556“K, MW 29, are moving Win an
exhaust pipe at 174 m/sec 1.12 kg/cm> ab static pressure. Find the total pressure and
temperature.
3. The exhaust temperature of a gas turbine, taken with a good total temperature
probe, is recorded at 636"C. Air-fuel ratio, o0.015; ;fuel is oil.. The gas velocity having
been determined to be 650 fps, calculate the static temperature.
4. Kerosene is the fuel of a gas turbine plant. / »= 0.012 7| « 972°K pressure
ratio, 4.5, exhaust to atmosphere. Find the available energy, kg-m per kg air flow.
5. A gas with y = 1.35, temperature 649®C, is expanded in a gas turbine from
3.52 to 1,05 kg/om* ab. 0.80. What is the temperature of the exhaust gas? Is this
static or total temperature? Give the reason.
6. Sketch the p-v and T-s cycles of an ideal open gas turbine cycle having: pressure
ratio, 5; fi, 649°C. Atmospheric state is standard for sea level, y = 1.37. Calculate the
ideal thermal efficiency.
7. An open gas turbine cycle is to be operated with a maximum temperature of
750®C. Intake air, i kg/cm* ab, 29.4®0; fuel, ChHm; Qu 43155 J/g ijt, «= 0.84;
fic *= 0.80; combustion efficiency, 95%. Draw the full load rt% vs. pressure ratio character¬
istic through the pressure ratio range of 4 to 10.
8. Consider that Fig. 7-11 is characteristic of all gas turbine plants and determine
the efficiency of a plant at half load. Plant has a pressure ratio of 6; other data same as in
Prob. 7.
9. Find the full-load efficiency, air rate, and exhaust temperature of an open-cycle
gas turbine plant to be built for optimum pressure ratio with Ts = 1028“K Ti = 178®K ,
ffT = 0.82, Tie ~ 0.79. Oil fuel with Ql » 43166 J/g • Combustion efficiency, 94%.
Standard sea level atmosphere. Specific heat data as in Fig. 7-2.
10. An open-cycle regenerative gas turbine plant has: Pi, 1.02 kg/om* ab; fi,32.2®C; ps,
7.03kg/om*abl8,816®C ^t,0.85; rja 0.84; fuel, Bunker C oil; combustion efficiency, 90%;
Ent 0.60. Find i/t and Wt^,
11. To the plant described in Prob. 9 is added a regenerator of 50% effectiveness.
Calculate the plant thermal efficiency.
12. An intercooler of 90% effectiveness is added to the plant of Prob. 10 at p = Vplp^.
Cooling water available at Jrc . Find
13. Graph the jhv cycle of Prob. 10 with scales of 1 oin=i kg/om>andl ora«60 om®
Quantity = 1 kg air at intake. Consider that compressor and turbine losses appear as
constant pressure reheats following isentropic processes.
14. Draw to scale a p-v cycle similar to Fig. 7-8 for a plant of the following data:
Scales 1 om ■■ 0.5 kg/cm®,quantity, 1 kg air at intake I’l, 0.9Skg/om® ab ; 15.6°C ; p^, 5.91
kg/om® ab ;/8 760®C ; interoooling at 2.45 kg cm® abto37.8*’C.Reheatat2.11kg/cm®ab- 29.4®C
Regenerator Eex = 0.75; tfr, 0.85; t/c, 0.85. Consider that turbine and compressor losses
appear as constant pressure reheats following isentropic processes.
15. An open-cycle gas turbine plant will exhaust into waste heat boilers which receive
65.6®C feedwater. The plant is otherwise like that described in Ex 1, Sec 7-3. The gas
turbine plant produces 8550 bp. How much saturated steam at 3.52 kg/om* ga can be
PROBLEMS 207
prodaoed per hoar, eiwiining tbet the flnel gM temperature will be S7.8°C above the
•team temperatura?
16. Sketch the shape of the p-o and T-$ graphs of cycles for the plants of Fig. 7-9(A)
and 7-9(C).
17. Rqieat Prob. 7-16, except use the other two plants shown.
18. Repeat Prob. 7-16, except use the plant shown in Fig. 7-10.
 CHAPTER 8

VAPOR CYCLES

8-1 Energy from External Combustion. Heat energy, produced by com¬

bustion of a fuel, can be converted to work under circumstances where the prod¬
ucts of combustion do not form the working medium of the heat cycle. This
external combustion principle implies a working medium interposed so as to
transfer heat from the region of combustion to the heat engine. The principle
of external combustion can be represented thus:

Heat produced by combustion is transferred by the heater to the working

medium; by it conveyed to a prime mover, which is able, in some way, to trans¬
form part of the heat energy, carried by the medium, into mechanical work.
The working medium is a fluid substance capable of absorbing and dis¬
charging heat energy by physical or chemical processes such as:
1. A change in temperature, but not in physical phase or chemical composi¬
tion.
2. A change in chemical composition.
3. A change in physical phase.
The first of these could be accomplished by a pressurized permanent gas
being heated to a high temperature in the heater, then flowing to the prime
mover, where it would expand, do work, and be cooled. The apparatus would
be bulky and much of the prime mover output would be needed to compress
the cool gas before heating. This method is exemplified by the closed gas turbine
cycle. It has been proposed that working media be employed which would
absorb and release heat by chemical composition and decomposition. However,
these have not been utilized because the third alternate is so well suited for an
external combustion cycle.
The change of physical phase which has been so widely used is vaporization
and condensation. The fluid is taken into the heater as a liquid under pressure
and there vaporized* by the heat energy. It then flows as a high-pressure vapor
to the prime mover which can be either of the engine or turbine form. There
♦ Hence the heater is a ''boiler.”
208
VAPOR CYCLES 209
it expands in a working process whereby the available energy it carries is
transfQP0Mi to mechanical work.
important to understand why this system is superior to one using a
permanent gas. Only a small part of the prime mover output is needed to in¬
troduce the fluid into the heater. The work of pumping the comparatively small
volume of liquid is far more favorable to equipment size (not necessarily to
thermal efiSciency) than is that of compressing a ^as:
The low-pressure vapor discharged from the prime mover is not diflScult to
condense to a liquid; hence a closed cycle may be produced. The thermodynamic
state of the.wij^:king^jnaejdium as^it circ^ylates from heater to prime mover and
back again constitutes a vapor cycle, de^j|d as ^.succession of thermodynamic
processes^'occurririg in a vaporizable fluid by means of which mecjbaid&arwdik
is’^pfoduceJ froih^ energy and after which the fluid is returned to its initial
state ready for another cycle. An engine cycle considers only the changes oc¬
curring within the en^Ue, but a vapor cycle involves, in addition, all changes
in the vapor state from the point of leaving the engine until it is again ready
to enter it.
There are several vapors which have physical properties suitable for work¬
ing media. Those which have been either seriously proposed or actually used
are steam, mercury vapor, sulfur dioxide, diphenyl oxide, and certain of the
hydrocarbons. The requirements for an ideal medium are:
1. Ample amounts should be available at low cost.
2. It should be noncorrosive and nontoxic.
3. The pressures corresponding to the maximum temperature range possible,
that is, from safe maximum temperature as dictated by the properties of the
metals of boiler, pipes, etc., to the minimum temperature as dictated by the
temperature of condensing water, should not be so high as to be commercially
impracticable, nor so low as to render maintenance of vacuum di£Scult.
4. It should wet the boiler surfaces enveloping it and should be stable,
chemically, at the maximum temperatures of the boiler.
5. Its heat of the liquid should be a minor part of the total heat since the
available heat is contained in latent form.
6. There should be considerable decrease in volume upon condensation.
Although water does not satisfy the third requirement as well as other
available vapors, its low cost and general abundancy have precluded, until
recent years, the use of any of the others. It satisfies the other requirements
well. It is in widespread use and its properties and cycles are well known to the
rank and file of practicing engineers. Nevertheless one should bear in mind that
rising fuel costs and smaller margins of profit will emphasize the importance
of the third-mentioned requirement, for we recognize that any possible increase
in the temperature range of the working medium itself will bring, without
further perfection of equipment, an increase of thermal efficiency. For instance,
mercury, although expensive and of toxic nature, has attractive possibilities
because of its moderate vapor pressures near the present-day temperature limit.
8-2 Vapor Cycles. Several thermodynamic combinations are currently in
use for vapor cycles, but all of them have absorption of heat at constant pressure
in a boiler, and a working expansion in a prime mover. An expansion of the
reversible adiabatic form would be desirable since, of all possible processes, it
210 VAPOR CYCLES
alone achieves the conversion of all the “available” heat energy into mechanical
work. Since friction destroys perfect reversibility, actual expansions approach
the ideal only to the extent that fluid and mechanical friction and heat transfer
are minimized during the working expansion.
A reversible adiabatic process is isentropic in nature. On temperature-
entropy plots, therefore, an ideal working process for a vapor cycle is repre¬
sented by a line perpendicular to the entropy axis'. Actual expansions are
accompanied by some entropy increases, those most inclined to veer from the
isentropic being least efficient. It can be thermodynamically proven that the
work done by an expanding medium equals its decrease of enthalpy, whether
ideal or otherwise, providing it is a simple expansion concerned solely with
transforming heat into work. The area enclosed by a cycle on temperature-
entropy axes also is work performed (inJoule units), provided the expansion is
isentropic. However, if the working process is partly irreversible, area of the
cycle has no significance.
The eflBciency of any working cycle is its output energy divided by its input
energy.
^ W
(8-1)
’ 0.102*Q,
where W = Work performed, kg-m per cycle
Qi = Heat supplied, Joulespercycle.
In this chapter we will describe the various vapor cycles in commercial use,
showing how to compute their performance and break down the energy flow in
“heat balances.”
yWMIh rtprtMnft A heat balance is an accounting of all energy
^ Entrgy
units involved in a system, a cycle, or an indi¬
Arrow tliowt flow vidual piece of equipment. It is the bookkeeping
of energy credits and debits with Joules as the unit
of currency. The most-used principle in heat bal¬

HI
ancing is the First Law of Thermodynamics. This
law is so‘generally employed and is so simple and
obvious that engineers often employ it without
thinking each time of its significance. The value
of the first law to heat balance calculations is
Figurt obovt shows
correct construction that, in the accounting for energy, simple addi¬
of 0 bronching hoot
stroom tions or subtractions may be used to account for
Figure of right
eontoins ssysrol of the total of heat, mechanical, or electrical energy.
the mistokss thot One way of showing a heat balance is in a tabular
form; another shows the energy as a stream, prop¬
erly branched and subdivided to indicate the
Fig. 8-1 Principles of heat distribution.
stream construction. A heat stream is often the best way to present
a heat balance, since it enables the reader to grasp
at once the relative magnitude of energy involved in the different sections of
heat-power equipment. To construct the diagram, one must have at hand a
calculated breakdown of all energy quantities, being sure to test the calcula-
* OonstHit will be 42(1.9 if la In kenl.
VAPOR CYC3LES 211
tions at all points possible for obedience to the aforesaid law. In other words,
one must be careful to account for aU energy. The widths of the heat stream
represent energy graphically so the scale could be any unit of energy. A com¬
mon basis, where combustion supplies the heat, is to consider the heating
value of the fuel as 100% and have the widths of the energy streams as per¬
centages of that heat.
Since stream width is energy, any tapering of the stream is in violation of
the first law. Streams may be split off or added together, but they may not be
tapered. Hence, where streams are to be curved, it is important that the same
center be used for both inner and outer arcs. Never use square comers for heat
streams, for the width perpendicular to the center-line would not be constant.
The performance of vapor cycle plants may be expressed in several ways.
The over-all thermal efficiency will probably convey the most meaning to the
novice who has a basic knowledge of thermodynamics and heat engineering
without much practical plant experience. On the other hand, the plant engineer
usually thinks in terms of his rate in Joule (or kcal) per iw hr, meaning
by that the number of Joule (or kcal) used to (xroduce one kw hn The Joule
(or kcal) are those represented in the heating value of the coal used, or diose
tnmsferred to the vapor cycle in the steam generator, deoending on the
purpose for which heat rate is stated.
Neither of these expressions is entirely descriptive of a power plant’s worth
in comparison with others in that they do not consider the eff^ of partial
capacity operation.
Sadi Camot, a French engineer of the early nineteenth century, during the
course of studies in natural philosophy recorded in a small brochure a descrip¬
tion of an ideal cycle. The basic soundness of his conclusions and the worth
of his cycle were not appreciated until after his death. Today his name is
perpetuated in the Carnot cycle. Although it has not been possible to constract
a practical plant operating on this cycle, it is of great value to heat power
theory. Its value resides in the high thermodynamic efficiency it possesses, the
highest, in fact, possible for any cycle operating between a heat source at
Ti °K and a heat rejection at Ta ‘’K. It is a standard of comparison for all other
cycles, whose possibilities may then be gauged by their approach to the Carnot
cycle efficiency, expressed by

The low efficiency common to all heat power cycles, including even the Car¬
not, may be understood if thought be given to the conditions which would be
required in order to make the value of igj 100%. Either Ti would have to be
infinitely large or Ta zero. As we can scarcely expect to work a plant at these
temperattire extremes, there is but little prospect of working the Camot cycle
efficiency much higher than it is now without some radical change, an event not
to be expected in the light of present experience.
Along with Camot and Joule, W. J. M. Rankine* is one of the pioneers of
heat power science. Rankine’s modification of the Camot cycle is the basis of
the modem steam plant, even though the Rankine cycle itself has been modi-
* Professor at the University of Qlasgow; b. 1820, d. 1872.
212 VAPOR CYCLES
fied and changed with the passing of time. For instance, Cotterill perceived
that the extraction of some of the steam from an engine for the purpose of
heating the boiler feedwater nearer saturation temperature would result in
considerable gain over the simple Rankine cycle. This idea was first applied to
reciprocating steam engine plants, but it did not enjoy its present widespread
use until the advent of the high-capacity, steam turbine central station. This
cycle is called the regenerative vapor cycle and is now applied to nearly every
modem central station and to a great many industrial plants as well.
The reheating cycle is a logical outgrowth of the trend to higher pressures.
The reheat vapor cycle is designed to keep the steam dry throughout more of
the expansion. This is accomplished by removing the steam from the tur¬
bine or engine before it reaches the state of saturation, resuperheating it
and readmitting to the prime mover for further working expansion.
There is another class of cycles which might be termed multiple vapor cy¬
cles. In 1899 Josse used the steam-sulfur dioxide cycle; in 1913 Emmet began
experimentation with a mercury-steam cycle. These are binary vapor cycles.
In such cycles the heat rejected from the leading vapor becomes the heat ab¬
sorbed by the following vapor. Thus, there are two working media, one of
which absorbs heat from the fuel and rejects its unavailable heat to the other
which, in turn, rejects as unavailable heat at the lowest possible temperature a
certain proportion of that which it receives.
The reader’s attention is directed to a compilation of diagram symbols
carried in the appendix. Some of these will be frequently, but not exclusively,
employed in the flow diagrams by means of which the equipment connections
required for the various cycles will be here illustrated.
8-3 Rankine Cycle. The closed Rankine vapor cycle can be carried out
in four pieces of equipment which are appropriately joined with pipes for con¬
veying the working medium from one to another. The basic arrangement of

Rump

Fig. 8-2 Elements of the Rankine vapor cycle.

this, the simplest of all vapor cycles, is shown by Fig. 8-2. This ideal plant
consists of a steam generator which receives feedwater under pressure from a
pump, a prime mover in which to obtain the working expansion, and a con¬
denser to reduce the exhaust steam to liquid, ready for pumping.
A closed cycle plant wdinarily contains auxiliary equipment added for
economic reasons; hence wc turn to an open cycle for an example of the
simplest form of an actual steam power plant. Illustrated by Fig. 8-3, such a
plant has a boiler generating steam for an engine which then discharges the
waste steam to atmosphere, thus losing the fluid. Replacement water, however,
RANKINE CYCLE 213
is taken in by the pump and pressurised for boiler use. Thus the condenser is
eliminated, but the cycle is open. A small refinement is the inclusion of a water
heater in which a small portion of the engine exhaust heat is salvaged by the
feedwater.
The efiSciency of such a plant is necessarily poor; yet for small amounts of
power it represents a type that has minimum invested capital and can be
operated in a successful manner by nonprofessional attendants.
When water is pumped into the boiler, it has the heat of the liquid, Af, in it
corresponding to its temperature. This temperature is that of the condensate
out of the condenser in an arrangement such as is shown in Fig. 8-2, but would
be the heated water temperature in case there is a heater. When vaporized by
the steam generator, this water has an enthalpy determined by its pressure and
quality, i.e., wet, dry, or superheated. The enthalpy entering the prime mover
is nominally the same as that leaving the steam generator. Call this enthalpy hi.

Fig. 8-3 Plant equipment for producing the open Rankine cycle.
In an ideal Rankine cycle the next action is an isentropic expansion to exhaust
pressure. Final enthalpy ^2 may be calculated after determining quality from
the equation Si = or it may be traced out on the Mollier diagram. After ex¬
haust from the prime mover the steam is condensed to a liquid at constant
pressure and temperature. Then it has enthalpy hfs. This assumes that the
enthalpy at state 4 (entrance to steam generator) is the same as at state 3,
since the temperature is the same. Actually some pump flow work is added, a
factor which is not entirely negligible for a high fluid pressure as was demon¬
strated in Sec 1-3. The efficiency of this Rankine vapor cycle, neglecting pump
work, is expressed as follows:

VR (8-3)

in which Ai — /i2 = Ideal heat-work transformation of an isentropic expansion.

Af, — Heat of the liquid at exhaust pressure.
A slightly lower net ijr is the result if pump work, Wp, is included. Eq 8-3 then

WJJ* h,-ht,-Wp/0.102 .

in which Wp is the pump work per kg fluid per cycle, kg-m Eqs S-Z and 8-4 will
be in substantial agreement except for high-pressure cycles.

* 3 mm aMehaaioal egaivalemt of baa» and 4M.9 if heat nnlta are in WL

214 VAPOR CYCLES
Where the working expansion is not carried isentropically fully to the ex¬
haust pressure, must be determined by other means than entropy equality.
Incomplete expansion, the custom in steam engine practice, is illustrated by the
cycle l-a-6-3-4, Fig. ^2. In place of fci-Aa of Eq 8-3, one must use

Pb)
^ 7710.2*“
in which p«, pb » Pressure at states a, b in kg/cm* ab.
Vft » Specific volume of steam at state a, cm* per g«
= Enthalpy of steam after isentropic expansion to pa, J/g.
Actual thermal efficiency of a vapor cycle based on the Rankine is given by
the following equation:*
2648 ®
(8-5)
w(hi — ^f)
where w « Steam roXe^ i.e., kg steam consumed by the prime mover per hp hr
of output.
hi = Heat of the liquid of the fluid entering the steam generator, J/g.
The efficiency lyt is based on .either indicated or shaft power, depending on
whether w is steam rate in kg per hr per ihp or bhp.
The combined thermal and electrical efficiency of a prime mover-generator
unit is
S599t
tjo — <8-6)
Wk(^i ~ K)
where is steam rate in kg per kw hr. Steam rates need to be established by
timed tests during which output power and steam consumption are both
measured.
Fiwnimr*- 1: The ideal Raakine effioieaoy for a vapor oyole operating between 11.6
kg/om* ab dry and saturated steam state and atmospherio pressure will be determmed.
Neglect pump work.
At 11.6 kg/cm* ab, 2783 J/g. After an isentropic expansion .to 1.03 kg/omt ab. Ag ««
2373 J/g, Afs » 41S.7 J/g. Using Eq 8-3,

2783 ~ 2378
17.3%
2783 418.7

Next, the thermal efficiency of an open-cycle plant operating at the same terminal
conditions is calculated. Feedwater is heated to 93.8®C . The steam flow, when producing
76 bhp, was measured as 1261 kg per hr.

Wh ** 1261/76 « 16.6 kg per bhp hr

Then with Eq 8-5,
^ 2648 ^ (170/
“ 16.e(S7SS - 390.8)

* If Aft is used, the equation becomes one of actual thermal efficiency of the prime mover,
@ Ckmstant will be 682.4 if A| and hf are in kocd/kg.
f Constant will be 869.9 if hi and Aft are in koal/kg.
 rankine cycle 215
Sxami^e 2: The calculaticm g{ Rankine efficiency of an incomplete expansion
is now stiowB* Assmns 10.05 kg/em* sb dry and saturated initial state ; release
pressiure 2.11 kg/em* ab ; eKhanst pressure «• 0.21 kg/omt ab.
First determine enthalpy and volume at states 1 and a.
A| - 2770.4 Jig. *1^ - 2404 J/g. « 0.90S» »» •» O.OOS x 857.8 » 774.6 em* per g.
Than with Eq. 8*2. modifled fsr incomplete expansion.
- 0.21)
^ ..= 17.0%
2770.4 254.5
Example 3: Consider that the following data pertain to the open-cycle plant
illustrated by 8-3. Heating value of coal. 82407 J/g; boiler and furnace
efficiency, 72.9%; engine output, 81 bhp; steam rate, 17 kg per bhp hr; steam to pump,
48.1kgper hr; equivalent pumping head,02.85 m; steam pressure, 125 psig, 99% dry;
fe^ temperature, 08.8K); cold water supply, 15.6®C. These are used to construct a heat
balance of this plant.
Enthalpy of steam = 756.6 -h .99 x 2021 = 2756.4 J/g
Increase of enthalpy by steam generator = 2756.4 - 390.8 = 2365.6 J/g.
Steam used = 81 x 17 4* 43.1 » 1420 kg per hr.
HEAT BALANCE OF ENGINE AND PUMP
Thousand Thousand
kccdthr Percent kcallhr Percent
^Reoeived from boiler, 1420 x 2756.4 x 0.280* .. . 932.7 84.9
Input to engine. 1877 x 2756.4 x 0.289 « 904.5 x 10»
Engine output. 1877 x . ^-7
Friction Iocs assumed os 4% of input. 36.2 3.3
Exhaust loss (difference). 817.1 74.4
Inrat to pump, 4S.1 x 2756.4 x 0,239-*28.3 x 10»
Fump output. 1320 x 92.85/426.9 . 0.3 0.0
Exhaust loss (difference). 27.9 2.5
Balance. 932.7 932.7
Before computing' the heater balance, the portion of the engine exhaust used by the
heater must be found. The mixing and heating are at constant pressure. Enthalpy of
engine exhaust = (817100/1377) 4.19 2490.3 J/g.
Let Wg = Exhaust steam entering heater; 1420 — Wg = Cold water entering. Then
1420 xl0>x 390.8 » 2490.3 x 10^ ^^0+65.51 (1420.1 ~ Wg}10’
Wg = 190.6 steam condensed per hr
Cold water entering heater = 1420—190.5 = 1229.5 kg per hr
Atmospheric exhaust * 1877 - 190.5=1186.6 kg per hr
HEAT BALANCE OF FEEDWATER HEATER
Thousand Thoueand
kcai/hr Percent kcallhr Pereeni
Reoeived in engine exhaust 817.1 74.4
Heat entering in cold water,.
1229.5 X 65.51 x 0.239 19.2 1.7
Delivered to boiler, 1420 x 890.8 x 0.239_ 132.2 12.0
Exhaust loss, 1196.5 x 2490.8 x 0Ji89 .... 704.1 64.1
Balsuoe. . 886.8 886.8

* Multiplier to oonvert J/g to keal/kg.

216 VAPOR CYCLES
HEAT BALANCE OF BOILER

Thoummd Thatuaind
healfhr Percent heailhr Percent
Received from heater. . 132,2 12.0
In heating value of coal,
(982.7 - lS2.2)I0*/.729 = . . . . 1097.9 100.0
Output in generated steam. 982.7 84.0
Boile** losses, 1007.9 x 10* x .271 = • 297.4 27.1

Balance*. . 1280.1 1280.1

By dividing all items of these balances by 4371.1, they are reduced to a basis of per¬
cent heat in the fuel.

8-4 Regenerative Cycle. Study of the Rankine vapor cycle has shown
that improvement requires either an increase in the thermal state of the high
pressure steam, a decrease of that of the exhaust steam, or both. In other words,
increased efficiency requires increased boiler pressure, initial superheat, and
condenser vacuum. These improvements reached natural limits of boiler

NEAT UNITS
IN THE FUEL
BOILER

ill NET ELEC.

OUTPUT

HEAT REJECTED TO
CONOENSINO WATER

Fio. 8-4 Flow diagram and graphic heat balance of simple condensing steam power
plant.

strength, high-temperature metallurgy, and available condensing water tem¬

perature some time ago. While the upper temperature limit is slowly being
increased, the ideal Rankine cycle performance is relatively static. Progress
has come with other cycles, primarily the Regenerative. This cycle is definitely
more efficient than the Rankine, for the same terminal steam conditions. Its
principal feature is a thermal regeneration of condensate into high-ftemperature
feedwater by the use of steam bled from the prime mover at points intermediate
between throttle and exhaust.
The advantage of this cycle over the Rankine is that the heating steam, as
extracted from the prime mover, has released considerable of the mechanical
work represented in its available energy while retaining most of its feedwater
heating ability. By extracting a portion of the main flow before it reaches too
regenerative cycle 217
low a thermal state it still contains most of the initial enthalpy. When the
feedwater is progressively heated nearly to boiler saturation temperature, the
necessary heat input to the cycle is decreased. As this decrease is greater than
the mechanical work lost by prematurely extractmg a small quantity of steam
'from the turome, the cycle eiiciency is mgner than that of the nonextractin v
ilWr-t her the boiler pressure em-
ployej^jfor thra heat of Ihe licuiidia.a-lart^r part of total enthalpy^ The extrac-
tlra of steam at a number of points is readily accomplished with steam turbines,
although not with steam engines. This, plus the fact that the regenerative cycle
plant is costly to construct, has meant that its prime mover is a steam turbine.
The regenerative cycle shows improvement over the Rankine in the item

of heat lost in the exhaust steam. It does, however, require more auxiliary
equipment in the form of heaters, pumps, traps, and piping. The heaters are
generally of the closed, or surface, type, but open, or contact, type heaters,
have been used. In order that the steam bleeder lines may be short in length,
the heaters are set as close to the turbine as the plant layout allows.
The elements of the regenerative vapor cycle are shown in Fig. 8-5. As
pictured there, feedwater heating from condenser to boiler feed temperature is
accomplished in three stages, using closed heaters. Heaters 2 and 3 are high-
pressure heaters, that is, they are on the boiler side of the boiler feed pump;
heater 1 is a low-pressure heater. This designation is based on water, not steam,
pressure. The condensed heater steam is pmnped from each heater into the
main feedwater stream leaving the heater. Thus all of the heat from the ex¬
tracted steam is given to the feedwater before it enters the next higher heater.
218 VAPOR CYCLES
The surge tank absorbs the irregularities of cycle flow. Its water level does not
alter except during changes of load on the generating unit, at which time water
either accumulates in it or is drawn from it, due to the time lag in rate of change
of flow in the different elements of the vapor cycle. Alternate schemes for the
handling of heater condensate are shown in Figs. 8-5 at (a) and (b). Traps are
float- or bucket-operated chambers which can pass condensate and will hold
back steam, but there must be a pressure difference between the steam chamber
and the condensate discharge. These traps do not involve as much initial or oper¬
ating expense as pumps, but their use results in a vapor cycle of slightly lower
eflSciency on account of the flashing of the condensate to a lower pressure, with
attendant increase of entropy. In scheme (a) the heater condensate is relumed
to the condenser hotwell through a water leg seal. This is possible only if the

Fig. 8-6 Ideal regenerative cycle employing infinite number of heaters.

lowest extraction pressure is sufficiently near the condenser pressure so that the
length of the vertical leg of the water seal is not too great. Otherwise it could
be cared for as in scheme (b), in which the condensate from all the heaters is
returned to the main feed stream just beyond heater 1 by a pump.
The terms regenerative vapor cycle and extraction cycle are not synonymous.
Extraction cycle refers to any airangement whereby steam is bled from a
^rbine at one or more pressures for any purpose whatsoever, i.e., feedwat^
^seating, process steam, Keating steam, etc. Regenerative cyde implies that the
extracteg steam liTused for one turpose, the thermal regeneration of the con*
densate to a temperature level approaching that of the boilerjwatgr. The terms
*^led steam^^ and '^extracW»E?^?!THny§’;'i^ synonymously, as may also
‘‘bleeder point” and “extraction point.” The effect of bleeding steam to heat the
feedwater is shown in Fig. 8-6 (a) which is for a cycle having an infinite num¬
ber of heaters. Were there a finite number of heaters, the line FJ would become
topped instead of smooth. In so showing the regenerative cycle on the tempera-
regenerative cycle 219
ture-entropy plane we assume that, by ei^actiojuthe ^tropy of ««panHing
at^Awi is decreased by the extraction of heat at constant temperatw^,. TTiis re¬
quires arbitrarily assuming tttat ordy heat pastes through the extractionJtnes
to the heaters. ' ’he extracted heat is immediately given up to the fee3water:
hence the line fj parallels tne uquid line CB BinceWeTieat~^at will be bl^ is
equal to the he Vj^aded to the feedwater beiween condenser temperature )T^and
feedwater temperature [Tf. But note, meanwhile, that as a definite steam wei^ET
is actually extracted ihkead of the heat only, as assumed, the actual condition
of the remaining expanding steam follows the line EFO and not EFJ.
This theoretical regenerative cycle (a) assumes, in addition to an infinite
number of heaters, no pressure or heat loss in the extraction lines, no terminal
temperature difference between water and steam in the heaters, no imdercooling
of the condensate below condenser saturation temperature, and an isentropic
expansion. An equation for the efiBciency of the regenerative cycle can be de¬
rived cm the basis of these assumptions. Pump work will be neglected. Referring
again to Fig. 8-6 (a), let it first be noted that

Tx Condensate temperature, ‘*K.

Tt»« Feedwater temperature, ®K.
hx » Initial enthalpy, J/g. Ana OADEH.
Kx » Final ^thalpy, J/g. Ana OABOH.
kt, Enthalpy of condraisate, J/g. Area OABL.
htw ~ Enthalpy of the liquid at Tt, J/g. Ana OACK.
Enthalpy of extracted steam, J/g. throttle flow. Ana IJFH.
Feed heating also ish, = Area LBCK.

Heat equivalent of net work of the cycle. Area BDEFJ.

W BDEFJ
Efficiency of the cycle — ijnt
J(hx - A,w) hi hti,
The area BDEFJ will be interpreted as follows:
BDEFJ = OADEH - OABGH - JGF
Howevei, since FJ parallels CB, JGF = BCM = LBCK — LBMK. Hence
BDEFJ = OADEH - OABGH - LBCK + LBMK.
Substitute in the efficiency equation, with replacement of areas by equivalent
enthalpies, noting that LBCK = hfw — ha.

— hi — hi — hur + hfi + TxAs

■ hi- hu,
Make the following substitution: -»{A2 — AfJ Af^«»lTf~273)4.18
As »4.187hi-^*
*V aai meohani^ equivalent of heat.
* Aasuming constant pressure heating with > 1 oal per » 4.187 J per g*0.
220 VAPOR CYCLES

hi-xjkfgt - 4.\&m+ 1143.1 + 4.187!r,ln

Iht, II (8-7)
ft, - 4.1872’f 4- 1143.1

Seeking the feedwater temperature for maximum efficiency, make 0)

dTi
then it will be found that Ti = Ti, meaning that the feed should be regenerated
to boiler saturation temperature for maximum ideal cycle efficiency. Should
Tt * Tiy the area MCDEG would become a rectangle equivalent to the Carnot
cycle. The following example will show that under this condition jjreg = vc-
While the ideal regenerative cycle of maximum efficiency equals the Carnot,
provided the prime mover uses saturated steam as in Fig. 8-6, the two cycles are
not equivalent when the steam is initially superheated because all the energy is
received at the high temperature in a Carnot engine and this is not true of the
steam cycle.
Example 1: The maximum efficiency of an ideal regenerative cycle working on
Mturated steam between 14.1 kg/om^ ab and 25.4 mm Hg abs will be compared with the
Carnot effioiency.

X2 = 0.747 dryness factor after isentropic expansion (from Mollier Chart)

At 14.1 kg/cm* ab Tj = 194,5 + 273 == 467.5'K.
At 25.4 mmUg Tf ~ 26.19+273— 299.2®K, Substitute in Eq 8-7, using steam table data.
.1198.4 - 0.747 X 1049.2 - 842 -j- 492 + 539 In m _
2789.7 - 4.187 x 487.3 + U48.1
V . •W7..3 - 299.2 „„„
1' or comparilBon: rjc — —— = 36%.
467.5

In the modern power station, where initial superheat is the rule, extraction
to the heaters begins in the superheated steam region. This alters the conclu¬
sions made lor saturated extraction. Referring to Fig. 8-6 (b) in which the final
feedwater temperature is Tf, y represents the initial extraction point on the
turbine expansion line. That is, if all the steam were taken by the heater, the
cycle would follow the line yy', which represents a constant pressure line on
the T-8 plane. But since only a portion of the steam is extracted, and that in
a series of heaters operating at constantly decreasing temperatures, the cycle
lollows a line yY to the saturation line, after which the remainder is the same
as in (a). The location of YY' is so that the shaded areas will be equal. No
reasonably simple equation which will represent the efficiency of the cycle
BCDEyYJ may be formed; but it is found that a line FF\ drawn from the
expansion line at Tt parallel to the liquid line approximates the cycle and has
the advantage of offering the same thermodynamic conditions as pictured in
Fig. 8-6 (a). Applying the previously derived equation for i^reg to this case, it is
again found that the maximum theoretical efficiency occurs when Tt equals the
saturation temperature at boiler pressure. But now the Carnot cycle efficiency
is considerably larger than the regenerative, the truth of which is demonstrated
in the following example.
regenerative cycle 221
Example 2: The maximum efficiency of an ideal regenerative cycle working be-
tween29-«*fg/cm* ab,371''Cand49.5aQjnHgabswillbe compared with the Carnot efficiency.

X2 = 0.812 dryness factor after isentropic expansion (from Mollier Chart).

At29.6 kg/cni« ga, « ^1-89 + 279 = 604.89, say 605°K.

At 49.6 mm, Hg,r2 = 37.9 4- 273 « 310.9, say 3irK .
8166.3 -- .812x2412.3 — 4.187 x 606 + 1143.1 -f 4.187 x 311 In gJi
= 39.5%
3166.3 ~ 4.187 x 606 + 1143.1
371 - 38
But now Vc = = 51.7%.
371 + 273

Example 3: It is now shown that failure to regenerate the feedwater fully to Tj

is not a severe penalty on regenerative cycle efficiency. Assume feedwater at 182.2®C.
Tf = 455°K. Other conditions as in Ex 2.
456
3166.3 - .812 X 2412.3 - 4.187 x 465 + 1143.1 + 4.187 x ®11
39.2%
3166.3 - 4.187 x 466 + 1148.1

Turbine Extraction
Pressure Heat Flow
Stage J/g kg/hr.
kg/cm 2
5 20*5 3154-6 22135
B 9*9 2935-9 9888
11 3*68 2812*7 19595
14 0-72 2568-4 15286

Fig. 8-7 Flow rliagram—Watts Bar steam plant. Flows are hourly, pressures absolute.
Heat added by feed pump is included.
222 VAPOR CYCLES
A difference of 89® in feedwater temperature caused only 0.3% difference in
efficiency.

Because of the excessive cost of heating surface and piping connections, it

is not possible to use sufficient extraction points even to approach the condition
represented by the line FJ. The number of heaters used in the modem plant
operating on the regenerative vapor cycle ranges from one to five. As higher
pressures are used, more heaters may be justified, but not many more on account
of the ever smaller increments of efficiency to be gained by increasing the num¬
ber of heaters. The criterion governing the best number of heaters is one of
economic justification requiring studies to be made for cycles using different
numbers of heaters, such studies taking into consideration costs and complica¬
tions added with each heater as well as cycle efficiencies. The optimum feed-
water temperature which theoretically equals Ti is reduced by the use of a finite
number of heaters, nonisentropic expansion line, throttling losses, etc. Calcula¬
tions for the most economic feed temperature and the proper number of extrac¬
tion heaters are inextricably involved with cost data and details too extensive
to be given in this book. However, there have been several studies* of this
problem. The results of studies of R. L. Reynolds are set forth in Figs. 8-8 to
8-11.
If the actual turbine expansion line is to be used and heat and pressure losses
in the bleeder piping not disregarded, the construction of the vapor cycle dia¬
gram is thereby rendered sufficiently complex that a T-s diagram is of question¬
able value. The actual case will be handled by vapor cycle calculations based
on assumed flow diagrams. Problems of this type are included in this chapter.
Fig. 8-8 shows the improvement in heat consumption obtained with different
numbers of stages of extraction. These curves show that the improvement due
to stage extraction increases as the final feed temperature increases, up to a
certain point, after which it decreases. As the number of stages of extraction in¬
creases, the most efficient final temperature of the feedwater also increases.
Because of the fact that the pressures in the turbine increase or decrease with
the amounts of steam passing through the various blade groups, the saturation
temperature of the steam in the turbine, and consequently the final feed tem¬
perature, will increase or decrease with the load carried by the unit. For the
average central station turbine, the most efficient feed temperature should be
obtained at from 75% to 85% of the full-load rating. Fortunately, the curve is
fairly flat for a considerable range of temperature on either side of the ideal,
which means that the pressure in the turbine can be varied almost without
appreciable loss in efficiency. The curves in Fig. 8-8 have been based on the as¬
sumption that all the feedwater heating is done with steam extracted from the
turbine.
Fig. 8-10 shows the effect of throttle pressure on the gain which can be ob¬
tained by the extraction of steam from the turbine from various numbers of
stages. In each case the feedwater is assumed to be heated to its most eco¬
nomical temperature. It can readily be seen that the gain due to the addition of
heaters decreases as the number of heaters increases. Also, additional heaters
can be used more advantageously with the higher throttle pressures than with
’^‘Also J. K. Salisbury, "The Steam Turbine Regenerative Cycle,” Trans, ASMS,
224 VAPOR CYCLES
the lower. The data are derived from a thermodynamic standpoint and do not
take into account the increased cost of installation and maintenance, due to the
addition of extra heaters and their resultant piping, foundations, etc. Fig. 8-11
shows the effect of throttle pressure on the temperature to which the feedwater
should be heated to obtain the maximum gain with stage extraction. These

Muttipb>r lo b» uftpd on
•0 improvomont -Pbrcontogo to bo oddod to or
obtoinod ot 736*6 mm Hg Qr 4.0*4-oubtroctod from the percentage
O 4_improvement obtoined at 899*CI
on fig 8-8 y .0*2fibJ_throlH. t«np*ratur* as shown I
B»s£s,«issi!S!s;;
2 o!E!!!S^seHiniHiiiHi

V. Lood
■■■■.■in
, ■■■■■■■vaflajsv.Lood 8
S*0 690 720 750 -0-6 ■■■■■■■■■■■■■■■■■■■■■■■
375 435 495 545
VACUUM - mm Hg
THROTTLE STEAM TEMPERATURE-’c

Fig. 8-9 Correction factors for vacuum, superheat, and partial load.

o I Z 3 4 5 6 7 8 9 »0
NUMSCa OF STAOeS or extraction
Fig. 8-10 Reduction in heat consumption obtained with various numbers of stages
and various steam pressures. (R.L. Reynolds.) Steam conditions: Pressure at throttle,
14.06-84.37 kg/cm* ga ,* total steam temperature, 399°C/ vacuum, 736.6 mm . Hg.
regenerative cycle 225
curves show that the most efficient final feed temperature increases quite
rapidly as heaters are added for the first four heaters, while additional points of
extraction above four affect the efficient final feed temperature only slightly.
The location of the turbine zones to which the heater should be connected de¬
pends upon the steam pressure at the turbine throttle. From Figs. ^8 and 8-11
the final temperature of the feedwater may be estimated. To obtain the maxi¬
mum efficiency it is desirable to locate the other feedwater heaters at such points
that the amounts of steam bled from the turbine to each heater are as nearly
equal as possible. This means that the temperature rises over each heater
would be approximately the same.
The higher the boiler pressure or the lower the condenser pressure, the better
will be the gain effected by the regenerative cycle, and more stages of regenera¬
tion will be justified. Certain secondary effects of a desirable nature result from
bleeding opieration. A reduction in the volume of steam passing the final stages

number of stages of extraction'

Fig. 8-11 Final feed water temperatures for maximum gain with various number of
stages and steam pressures from 14.00-84.87 kg/om* g». (R.li. Reynolds.) Total steam
temperature, 65.0*C; vacuum, 786.6 mm. Hg.
226 VAPOR CYCLES
of the turbine blading allows the turbine casing at the exhaust end to be much
smaller, an important factor because size of the turbine casing at the low-
pressure end is frequently a limiting dimension of the unit. Another result of
less steam in the exhaust is the higher vacuum possible without increase of
condensing water to the condenser.
Regenerative Turbines. It is common knowledge with those who ai'e familiar
with turbines that the expansion of steam is not isentropic. Friction loss in each
stage causes loss of availability and increasing departure of the steam condi¬
tions from the isentropic in the direction of increased entropy.
The extent of this defect is governed by the moisture in the steam, velocity
of the steam, the number of stages, and the general excellence of design and
construction. Consequently it is to be expected that turbines of different makes
and types will have varied expansion lines upon the Mollier Diagram. Some
typical lines are shown in Fig. A-13. These lines form the basis of regenera¬
tive cycle calculations and should be thoroughly understood. The exact form
of the line is a curve, but considering it a broken line between the known points
is accurate enough for practical purposes. The use of nonextraction expansion
lines for extraction calculations is an assumption frequently made. Turbines
from which steam is to be extracted are provided with suitable outlets to which
the extraction lines may be connected.
There are two types of extraction, i.e., extraction at constant steam pressure
and extraction at whatever pressure exists in the turbine at the extraction point.
Extraction at constant pressure requires that an extraction valve gear be pro¬
vided to regulate the area through which steam flows into the lower pressure
stages. This is necessary because, not only would ^he extraction pressure vary
with different amounts of extracted steam demanded, but varying loads
on the turbine would cause the casing pressure at the extraction nozzle to vary.
The extraction valve gear is often complicated by the use of a control or pilot
vaive to operate the main extraction valve. Turbines equipped with extraction
valve gear are naturally more expensive than the simpler forms which have no
pressure governing on the extraction lines. Industrial use of extracted steam
often requires that the pressure of the bled steam be kept constant. Also, in¬
dustrial use of the extraction turbine differs from central station practice in
that frequently a large portion of the total flow is extracted, whereas in the
power plant only a small fraction of the total is used for feedwater heating.
Therefore, regenerative cycle turbines are of the simplest form, the pressure and
temperature of the steam in the extraction lines varying in accordance with the
load on the turbine and the quantity extracted. This does not measurably affect
the regenerative cycle as its only effect is to vary the final feedwater tempera¬
ture within moderate limits, and it has already been said that considerable
variation in the final feedwater temperature does not seriously affect the
eflSciency of the cycle.
The conditions of extraction of steam are schematically set forth in Fig.
8-12 for an assumed three-heater cycle. Thermal state of the steam flowing
through the turbine is depicted on a section of the Mollier diagram. Steam
enters the turbine at pressure pi, total temperature ^i, and begins its working
expansion towards the exhaust end. After passing a group of stages its pressure
is reduced to its enthalpy to Ac. At this point a casing opening is provided so
regenerative cycle 227
that a small portion of the steam flow can be extracted to Heater C. The re¬
maining flow continues past another group of stages to point b where more of it
is extracted for use in Heater B. Similarly, after an ejrtraction at point a, the
remaining steam continues to expand until it flows through the exhaust nozzle
with pressure p2, dryness factor X2f and enthalpy ^2- The enthalpies At, and
he may be determined from the condition line if the pressures Pa, Pb, and Pc are
known. We shall next explain what sets the magnitude of these pressures.
The heaters are usually shell-and-tube construction, with steam contained
within the shell and water flowing through the tubes. The steam is considered
to be at the saturation temperature corresponding to shell pressure and is
everywhere warmer than the water in the tubes. The temperature difference.

Heoter C
Fig. 8-12 Conditions of steam extraction.

steam to water, is least where the water leaves the heater. This least M is called
the ^'terminal difference.'’ To produce a given outgoing water temperature to,
the turbine steam pressure must be appropriate to allow for the terminal differ¬
ence, as well as for the fluid friction loss in the extraction line. (See Fig. 8-12.)
Extraction pressure = Saturation pressure at (to + At) + Ap.
The turbine condition line, together with calculated extraction weights
(method to be shown later) and pressures, furnishes the enthalpy information
by means of which the gross mechanical work delivered to the turbine shaft is
estimated. The steam turbine is so compact a heat engine that radiation and
convection heat loss from its surface is negligible. Therefore, relying on the law
of conservation of energy, we can write the following equation for mechanical
work done per g of steam flow to the throttle.
W ^ (hi — he) + (1 — Wc)(hc — hh) + (1 — ““ y>h)(hh — ht) -j-
(l — We. — Wh - Wa)(K h) J per g (8-8)
The net energy delivered to the generator = gross energy developed — mechani¬
cal friction losses. Brown and Drewry give the following empirical formula for
mechanical losses, including bearing friction, gland resistance, and oil pump
drive.
4 O’**
Friction % normal rating (8-9)
/Rated kw
y 1000
* For ratings of over 25,000 kw, increase the numerator to 6.0.
228 VAPOR CYCLES
The following formula for generator efficiency is taken from the same
source.

Generator efficiency = 0.98 — X (8-10)

3/Rated kw Load
V 1000
Generator efficiencies have been increased since Eq 8-10 was proposed,
principally through the use of hydrogen cooling to reduce windage losses. (See
Fig, A-16.)
8-5 Design of Regenerative Cycles. The preliminary layout of a pro¬
posed regenerative cycle plant includes the selection of principal units of
equipment and their incorporation into a flow diagram, then an estimate of the
magnitude of flows in the different lines. An actual turbine condition line is
needed for a realistic appraisal of such a cycle. Several such graphs on the h^s
plane are included in the Appendix. Furthermore, in Chapter 11, w^ shall find
that the engineer can predict the condition line from general specifications for
the type of turbine.
In actual practice, a working knowledge of this cycle requires familiarity
with many details. One of the first questions to be raised after terminal pres¬
sures and temperatures, feedwater temperature, and number of heaters have
been decided is ^^How much steam is to be bled to each heater, and what are the
heater temperatures to be?’^
A regenerative heater resembles a small surface condenser. The terminal
difference used depends on how much can be paid for the heater. The economical
range for large central vstation practice is aboutl,7'^to4.4°C,Surf ace heaters with
desuperheat zones are sometimes justified if the extraction steam contains a
high superheat. These heaters have zones baffled off in the shell to confine the
incoming superheated steam to contact wdth tubes carrying the warmest water.
By this means the outgoing water may be heated to higher than heater satura¬
tion temperature. Heaters so modified are more expensive per unit of heat trans¬
fer capacity but are frequently economic on account of the extra value of high
temperature heat they recover.
Heaters are seldom at any great distance from the turbine so that there is
but little pressure drop to the heater. In the case of long extraction lines, a
calculation should be made for pipe line pressure drop. Calculations of this type
are given in Chapter 14. The pressure loss due to pipe friction and throttling at
the extraction nozzle can be made the subject of an empirical allowance. The
assumed heater terminal temperature difference can be increased by 1.7®to2,8®or
from 5% to 7% pressure drop (Ap, Fig. 8-12) can be allowed.
Another point of considerable interest in cycle layout is the disposal of the
condensed extraction steam. The condensate can be handled by pumps as in
Fig. 8-5. This system, indeed, is used to some extent; it has the advantages of
logical cycle layout, elimination of available energy loss due to flashing of con¬
densate to a lower pressure, and simplification of piping. Its disadvantages are
the cost of numerous pumps and motors, their wiring and control, and their
electric energy consumption. But principally pumps are objected to because the
operation of the plant can be tied up by such a small auxiliary, and that the
DESIGN OF REGENERATIVE CYCLES 229
requirements of availability and reliability do not warrant taking such risks.
Simpler and more foolproof systems are used; for instance, those illustrated in
the alternate schemes (a) and (b), Fig. 8-5.
Modifications of the regenerative cycle are so numerous that a complete
resume of them is beyond the possibilities of this chapter. Two might be men¬
tioned. (1) Heat which would otherwise be wasted is often recovered by passing
the flow from the condenser hotwell through generator air coolers, oil coolers,
steam jet inter- and after-condensers, gland steam condensers, etc., before it
reaches the first regenerative heater. (2) When distilled water is used for feed-
water make-up, evaporators using extracted steam for vaporizing the raw water
are included in the cycle.

Example 1: The probable flows and thermal performance of a three-heater regen¬

erative cycle having a flow diagram such as Fig. 8-5 will be determined, with the aid
of turbine expansion line A shown on the Appendix chart. The condensate from each
heater will be assumed to be at the same temperature as the outgoing feedwater, and
to be handled by pumping. Also a 5.6°C temperature difference between outgoing water-
and saturation temperature at the extraction point in the turbine will be considered to
cover both the heater terminal difference and the effect of pipe friction in the extraction
lines. This turbine has steam condition of 24.61 kg/om^ ab360*'Cexhaustat25.4mEn HgThe
generator output at rated load (represented by the expansion lin^) is 20,(XX) kw.
Fig. 8-11 is consulted to determine the maximum economic feedwater temperature.
The throttle pressure is24.61 kg/cm* ab, or 23.68 kg/cm* ga, for which the temperature is
found to be 161.7®0. The saturation temperature at 25.4 mm. Hg is 26.l°a The feed-
water must be heated from 26.r to 151.7% a range of 125.6 . It follows that, for three
heaters, the desirable rise of temperature in each is approximately 125.6/3 = 41.9®C.
Then the intermediate temperatures are: Between heaters A and B,26.1 + 41.9 —68%
and between heaters B and C, 68 -f 41.9 = 109.9. Applying the 5.6® difference between
feedwater tepiperature and extraction point saturation temperature, one obtains satura¬
tion temperatures at the turbine extraction points and, from steam tables, the corre¬
sponding pressures. Then by the use of the condition line the total heat in the extracted
steam is found at the intersection of condition line and extraction pressure. These are
recorded along with the pressures, the heat of the liquid hf, and the difference h — hf.

Pressure
Saturation i h at hi at
<0 at h-h,
Temperature turbine to
(°C) Turbine, ( J/gi)
at Turbine Abs ( j/g) ( j/g )

Throttle 222.07 24.61 3149.9

Extraction C 151.7 151.7+ 5.6 = 157.3 5.88 2884.7 6.39.6 2245.1
Extraction B 109.9 109.9+ 6.6=11.5.5 1.75 2698,7 460.9 2237.8
Extraction A 68 68 + 6.6 = 73.b 0.371 2619.5 284.6 2234.9
Condenser 26.1 25.4 mm 2280 109.5
1

The extraction flows per kg throttle flow are designated ta», tab,
Heat balances for the three heaters are written thus:*

’'‘For simplification M x 4.1868 hae been assumed to equal Mt. This is not striotbr true.
230 VAPOR CYCLES
Heater C: (l - - 109.9 )4.1868 =:^ 2245.1

Heater B: (1 - - 68)4.1868 = 2237.8

Heater A:(i - - wb“ “ 26.1)4.1868 - 2234.9
These equations, solved progressively, give:
Wt, =0.0626 kg; Wh =0.0674 kgj =0.0723 kg
Next, a substitution in Eq 8-8 gives the energy converted into work.
-= 1(3149.9 - 2884.7) + (l - 0.0723)(2884 7 - 2698.7)
-H(l - 0.0723 - 0.0674)(2698.7 - 2519.6)
-f (1 - 0.0723 - 0.0674 — 0.0626)(2519.5 — 2280)
783.1 J per g throttle flow ^
Friction loss = —X 783.1 _ 7.0 ,1 per g
/20,000
V 1000
loput to generator per g steam flow at throttle - 783.1 — 7 — 776.1 J*
The expansion line is for an 1800-rpm unit.Trdm Fig. A-15 the generator efficiency
is found to be 97.1% for an air-cooled generator, which type will be assumed here.
Energy in electrical output per g flow =-776.1 x 0.971 ^ 745 .1 . Since 20,000 kw are
equivalent to 7.2 x lO^o J per hr the fhll throttle flow ia 7.2 x I010/745 x 10^ ~ 96643 kg per hr.
Full extraction flows are
Wa = 96643 y 0.0626 = 6050 kg per hr

l^b = 96643 X 0.0674 . 6514 kg per hr

Wo = 96643 X 0.0723 — 6987 kg pei hr
Regenerative cycle thermal efficiency = 776.1/(3149.9 — 639.6)= 30.9%.

This is the gross efficiency. Net efficiency would include consideration of energy to
pumps and other auxiliaries.

8-6 Regenerative Cycle Heat Balance. The simple symmetry of feed-

water regeneration outlined in the two preceding sections is not often observed
in an actual plant, for water loss make-up, feed treatment, and other features
not seen in Fig. 8-5 arc often introduced. The basic plan of regeneration re¬
mains, but the simple flow diagrams heretofore described are altered as needed
to accommodate the desired additions. An example of a typical plant flow dia¬
gram is shown in Fig. 8-13. Here the turbine is provided with four points of
extraction, although the cycle has but three strictly regenerative heaters.
Extraction from point 3 is split, part being used to evaporate raw water for
make-up purposes. The extraction from point 2 is used to heat feedwater for
thermal deaeration (removal of dissolved gases, especially oxygen). The
evaporator vapor also goes to the deaerator. This deaerator employs a contact,
or mixing, type of process so that no condensate is separately removed from it.
It not only degasifies the water but also serves as an evaporator condenser and
a regenerative heater.
The heat of the steam used for jets in condenser air ejectors is generally
salvaged in an ejector condenser, as is done in this case. Here the jet steam is
derived from an auxiliary steam supply outside the main flow plan, but usually
such steam is taken from the main steam supply via pressure-reducing valves.
The deaerating heater tank acts as a partial surge tank, it usually having
REGENERATIVE CYCLE HEAT BALANCE 231
considerable waiter storage volume. External surge capacity is here provided by
an elevated storage tank (not shown) with valve in the surge line operated by
float control from the deaerator water level. Feedback from the storage tank
enters the system at the condenser hotwell via a valve operated by hotwell
water level.
This is one plan of regenerative flow—a typical one, but not a prototype of
such plants in general, for the only generalization that can be made of real flow
diagrams is that no two will be alike. There are many local considerations and
numerous individual engineering ideas that create the variations from plant to
plant.

STEAM CONDITIONS
Station Pressure {kglem^ ab) Enthalpy (Jig) Plow (kgfhr)
Throttle 88.94 3380.3 320.240
Extr. point 4 23.20 3089.5 28.850
Extr. point 3 9.14 2896.4 27,900
Extr. point 2 3.41 2729 11,200
Gland seal 3105.8 820
Extr. point 1 0.759 2.524.2 24.500
Condenser 0.0517 2259 226,970
Fig. 8-13 Sample of regenerative cycle plant flow diagram.

Example 1: Assuming that the data entered on Fig. 8-13 represents the result of
a test of the vapor cycle, it will now be shown how such information can be worked
into a heat balance and expressed graphically. All the separate pieces of equipment in
which heat transfers or transformations occur must be surveyed by use of the law of
conservation of energy. Where the data are sufficient for the purpose, a balance can be
struck, thereby validating the accuracy of testing, for the “unaccoimted for^^ item
necessary to make an exact balance should be within the accuracy of instrumentation
of the test. In some instances the balance is found by assigning the necessary difference
to one of the items. In these cases the achievement of an exact balance is not, of course,
proof of accurate test procedure. The following tabulated calculations are recorded in
abbreviated form making it necessary for the reader to employ prior knowledge of steam
232 VAPOR CYCLES
properties in correlating data from the diagram with these computations. Note dividon
of the solution into six sections.
1. Enthalpy of steam and blowdown leaving the steam generator is first found;
then by subtracting feedwater enthalpy, the heat absorption is determined.
Heat absorbed in steam generator = 320,240 x 3380.3 x 0.239 -j- 3180 x 1389.3^
X 10® — 323,420 X 912,8 x 0.239 = 188,661,500 koal per hr. The enthalpy oorreotion
explained in £x 4, See 1-3, is not applied' here because the magnitude of its effeot
is of the same order as of 1% accuracy in steam flow measurement, whereas
1% is good for such test data.
Heat input in coal 188.661,500/0.825 = 228,^80,500 koal per hr.
Combustion losses in steam generator = 228,680.500-188,661,500=40,019,000koalperhr.
2. Disposition of the enthalpy delivered to the turbine is next analyzed.
Thousand local per hr
Enthalpy at turbine throttle = 320.240 x 3380.3 x 0.239 = 257,956
Extraction No. 4 : 28.850 x 3089.5 x 0.239 «. 21,240
Extraction No. 3 : 27,900 x 2896.4 x 0.239 19,256
Extraction No. 2:/ll.200 x 2729 x 0.239 = 7,283
Gland leakoff: 820 x 3105.8 x 0.239 = 607
Extraction No. 1: 24.500 x 2524.2 x 0.239 = 14,737
Exhaust: 226,970 X 2259 x 0.239 « 122,180
185,303

Difference is gross work delivered to shaft 72,655

Generator output = 81,250 x 869.9 — 69,863

Difference is combined mechanical-electrical loss 2,792

3. Analysis of the condenser and ejector condenser is next. In the ejector condenser
the steam (used to activate the jet compression of noncondensable gas out of the main
condenser) which is exhausted by the ejector jets is condensed by absorbing its latent
heat in condensate from the main condenser. Beginning at this point the main conden¬
sate flow will be referred to as Feedwater,

Condenser heat balance Thousand local per hr

Turbine exhaust. 122,180
Condensate from ejector cond. 270 x 390.8 x 0.289i. = 25
Condensate from A: (820 -f 24,500)376.9 x 0.239 _ = 2,274
Feedwater out: 252,660 x 139.68 x 0.239. = 8,401
Absorbed by condensing water (difference). 116,078

124,470 124>479

Ejector condenser heat balance Thousand heal per hr

Jet steam: 270 x 3237.8 x 0.239 .. = 20H
Feedwater in... „ . 8401
Condensate to main condenser. 25
Feedwater out (difference). 8584

6098. 8609

4. Heat balances of the three extraction heaters A, B, and C are now tabulated.
Data are sufficient for a calculated balance. An ^^unaccounted for” item is inserted where
necessary in order to strike an exact balance.
REGENERATIVE CYCLE HEAT BALANCE 233
Healer A balance Thousand heal per M
Extraction steam (see part 2) 14,787
Gland steam 607.
Feedwater in* 8,408
Condensate out 2,274
Feedwater out 247,390 x 305.14 x 0.239 B- 21,526
Unaccounted for 48

23,800 23,800
Healer B balance Thousand heal per hr
Extraction steam 19,040 x 2890.4 x 0 239 13,556
Feedwater in 323,420 x 570.3 x 0.239 43,958
Condensate from 28,850 x 979.2 x 0.239 6,782
Feedwater out 323,420 x 724 x 0.239 55,799
Condensate to deaerator 48,490 x 731.5 x 0.239 8,452
Unaccounted for 10

64,251 64,251
Healer C balance Thousand heal per hr
Extraction steam 21,240
Feedwater in 65,709
Feedwater out 323,420 x 911.8 x 0 230 70,272
Condensate to B 6,732
Unaccounted for 35

77.039 77,039
5. Finally, the distribution of heat quantities in the evaporator and deaeratory which
are placed m this cycle for feed purification is computed
Evaporator heat balance Thousand heal per hr
Extraction steam 8260 x 2896.4 x 0 239 = 5,701
In make-up 8890 x 507 x 0.239 = 1,074
Vapor out 8070 x 2726 1 x 0.239 = 5,242
Blowdown 820 X 507 X 0.239 = 66
Condensate to deaerator 8260 x 730.4 x 0,239 1,438
Unaccounted for 4

6,779 6,779
Deaerator heat balance Thousand herd per hr
Extraction steam (see part 2) 7,283
Evaporator vapor 5,242
Evaporator condensate 1,438
Condensate from B 8,452
Feedwater from A 21,526
Feedwater to B 4S.963
Unaccoimted for 12

48,953 43,953
*This 18 less than enthalpy of feedwater out of ejector condenser because of the flow
of 5,170 /hr to storage Enthalpy of this flow = 8,584,000/252,500 = ^34 koal per kg.
Enthalpy of feedwater* to A = 8,584,000 — 5170 x 34 = 8,408,220 koal per kg,
t State assumed to be dry and saturated at 3.25 kg/om* ab.
234 VAPOR CYCLES
6. The flow of heat energ>' in this plant is more readily grasped if the results are
displayed in a heat stream, Fig. 8-14. In the arrangement of such a figure the engineer
may employ much originality and ingenuity, but he must be careful to respect the
principle of conservation of energy since the scaled widths of the heat stream repre¬
sent heat quantities.

Fig. 8-14 Heat balance of typical regenerative cycle.

B-? The Reheating Cycle. An adiabatic working process, such as is

approximated in steam nozzles, engines, and turbines, implies that the mechani¬
cal energy which appears has been produced at the expense of the heat in the
steam. Superheated steam will lose its superheat, then become more wet as the
expansion proceeds to lower pressures. It is a matter of record that moisture
in expanding steam is undesirable.
Development of high temperature-high pressure equipment has proceeded
steadily on account of the greater work availability contained in high potential
steam. Since structural limitations imposed by the metals available for super¬
heaters, pipes, etc., limit upper temperatures, and since higher pressures mean
also higher saturation temperatures, it is evident that the maximum degree of
superheat possible must ultimately decrease as higher pressures are adopted.
The inevitable effect of higher pressures is that the saturation line is reached
more quickly in an adiabatic expansion and more of the turbine stages operate
in the relatively undesirable saturated steam region. This explains why reheat-
 THE REHEATING CYCLE 235
ing the steam is sometimes practiced in high-pressure plants. The reheating is
accomplished by constructing the turbine so that all of the steam may be
extracted at a suitable point, resuperheated, then readmitted to the remaining
stages for further expansion.
Resuperheating is a feature associated with high steam pressures where
there is insufficiently high initial steam temperature to yield an expansion that
will end at the desirable state B, Fig. 8-15. It will be observed that this diagram
is on the Mollier plane and that a number of turbine condition lines are shown.
There is a minimum practical limit of final dryness
factor, X2f beyond which the accumulating moisture
droplets in the steam cause mechanical difficulties,
and there is a minimum exhaust pressure set by the
temperature and quantity of the condensing water
available to the steam condenser. These limits are
roughly about 86%and38.1mmHg obs.They produce a
“floor” under the permissible expansion line, of which
point B is obviously the most desirable, as it has the
minimum practical exhaust enthalpy. A turbine re¬
ceiving steam of initial state A .would expand it to B
without any resuperheating. State A, ab 19.3 kg/cm*
and 296®C , is typical of nonreheating plants built in
the 1920^8. Seeking higher efficiencies, companies
naturally followed progress to higher pressures and
temperatures when manufacturers were ready to
supply the equipment. In the same decade, equip¬
ment capable of producing and using steam at state
C was made available. Used without reheat the tur¬
bines would have produced an expansion CD', result¬
ing in maintenance difficulties because of excessive
moisture near the exhaust state. The expansion would
have had to be terminated at D" by increasing the
exhaust pressure, resulting in an unfavorable energy
utilization. But by exhausting the steam at D and Fig. 8-15 Reheat vs.
resuperheating at constant pressure to D, the remain¬ nonreheat turbine condi¬
ing expansion to F allowed full utilization of the tion lines: (1) early non-
reheat; (2) early reheat
condenser vacuum. Thus it can be seen that the pur¬
(1920’s); (3) nonreheat
pose of reheat is to take full advantage of higher (1930’s^:. (4) recent re¬
jnitial thermal states as they become commercially heat practice.
bailable, in the 193Q^s metallurgical improvements
extended the possibilities of steam generationto63.3kg/cm*and482®G—typically
state G, This State again permitted expansion to an exhaust state approximat¬
ing point B without reheat, and for a time there was little interest in the
reheat cycle. Recently increases in available steam pressures have outstripped
the temperature increases, and reheat is again being adopted, leading to ex¬
pansion conditions such as line HJ,
The high-pressure, high-temperature steam generator, together with extra
cost of turbines, piping, and controls, makes plants built to this cycle more
expensive than for nonrebeat. The cycle does not become an economic one
236 VAPOR CYCLES
unless constructed as a boiler-turbine unit of approximately 50,000 kw or more
capacity intended to be used as a base load plant.
The principal advantage is a gain of 4 to 7% thermal eflSciency over an

Fig. 8-16 Flow diagram of the reheat cycle.

equivalent nonreheat cycle. This advantage tends to justify greater initial

expenditures for equipment as the price of fuel advances.
Doubtless it will already have occurred to the reader that any power plant
of a type that would be using reheating
would also employ feedwater heating to a
high-temperature level. This indeed is the
case and all reheating vapor cycles also em¬
ploy regenerative feedwater heating. A
combination of the two vapor cycles would
retain all the advantages of each and incur
no additional disadvantages beyond those
already mentioned for these cycles.
An example of the equipment arrange¬
ment of the reheat cycle is illustrated in
Fig. 8-16. The turbine stages must be
grouped for convenience in removing the
entire flow at an appropriate intermediate
point. Formerly this could be done between
the high- and low-pressure sections of a
Fig. 8-17 The reheating cycle. compound turbine, but today^s high initial
steam conditions require reheat before the
low-pressure section is reached. Hence a modern arrangement is to divide the
high-pressure casing by a diaphragm, so converting it to high- and intermedi¬
ate-pressure sections with reheat between these. The average higher pressure
of reheating has helped by reducing the required size of reheat piping.
SUPERPOSED POWER UNITS 237
Some earlier reheat cycles employed live steam reheaters at the turbine to
dispense with reheat at the boiler and with reheat pipe lines to and from the
turbine. As reheat temperature when using live steam cannot be much higher
than boiler saturation temperature, this method has been abandoned in favor of
gas reheating sections in the steam generator or separately fired reheaters. The
latter are seen in marine practice, but not often in stationary plants.
Steam storage in the turbine, reheater, and reheat piping presents a problem
in speed control, for surplus energy can continue to be generated in the IP and
LP sections for some time after the main governor has reacted to a load de¬
crease. This has possibilities of producing dangerous overspeeding, to correct
which governor-operated intercept valves are installed ahead of the point of
reintroduction of the reheated steam to the turbines. These are adjusted to come
into action after the normal speed regulation has been exceeded, but before
speed has increased to the point (approximately 110% normal rpm) where an
emergency overspeed trip would shut down the whole unit.
The theoretical performance of this cycle is established by study of the
ideal cycle. Steam enters the turbine at a condition corresponding to point D,
Fig. 8-17, and, after an isentropic expansion to is reheated at constant
pressure to Trj, after which the expansion is completed to the exhaust state G
for which the temperature is To. Increase of entropy by reheating is As, and
reheat action is the portion EF.
The heat converted into work is the area BCDEFG, while that supplied in
the steam generator is hi -b hr — ht^^ where hr is the reheat, i.e., area EFHL
The thermal efficiency of the reheat cycle is given by the equation

hi hr — hi
(8-11)
hi hr — hff
in which K = rp(TR, - Tr.) = At. “
h^ = .T2fefgj + /if,.

By combining Eqs 8-7 and 8-11, one obtains the efficiency of an ideal re¬
heating-regenerative cycle (infinite number of heaters) having no feed heating
extraction until after reheating. It is also assumed that reheating begins at the
saturation line.

h + K- - 4.i87rf + 1143.1 + 4.187r, In TtlT,

ft ar ^ (o—)
hi H- hr - 4.187Tf + 1143.1
8-8 Superposed Power Units. Power demands are seldom static. Every
power plant, whether of public service or industrial type, may not only expect
to serve a variable load, but one whose incidence is either in the direction of a
growing or a diminishitig load. Commonly the change is in the direction of a
growth of load, this being almost always true for the public service plant. The
capacity of an existing power plant may be expanded in a number of ways, viz.:
1. Replace existing plant with a new one having adequate capacity.
2. Extend the existing plant by purchasing additional equipment similar to
that already installed.
All h*H in J/g,
238 VAPOR CYCLES
3. Modernize the existing equipment, thereby increasing its output and
efficiency.
4. Purchase power from systems having excess capacity.
6. Superpose a high-pressure steam plant on the existing plant.
The superposition mentioned above refers to the thermodynamic position of
the new equipment, rather than its physical location. Where an existing plant is
becoming inadequate chiefly through growth of load rather than by natural
depreciation or obsolescence, superposed power may be the economic method
of increasing capacity.
Sometimes plants in need cTf increased capacity have moderate pressure
turbines in excellent condition operating at efficiencies not too inferior to new
equipment. Such plants may be enlarged by a high-pressure, noncondensing

Fio. 8-18 Flow diagram of a high-pressure unit superposed on existing plant.

addition. It is possible that the superposed unit can pass sufficient steam into
the original plant header for the old boilers to be discarded. In this way, not
only are the excellent qualities of existing turbines retained, but also it is possi¬
ble to obtain the benefits of the progress that has occurred in high-pressure
steam generation. Many plant extensions have been secured in this way and
have displayed remarkable economies of operation when compared to the
original units. Not the least of the factors contributing to the success of super¬
position has been the development and standardization of equipment for high
pressures and high temperatures. Pressures of 56 to 98 kg/cm* and temperatures
to 538'*C I are no longer considered to be unusual or hazardous. Manufacturers
are ready to supply commercially tried and tested equipment in this range.
The capacity of the superposed unit and the throttle pressures required are
arrived at by a coordinated study of load growths, the cycle of the bating
 plant, and thermodynamic studies of the probable steam expansion in the
superposed unit.
The steam conditions for superposition are shown plotted on the Mollier
diagram in Fig. 8-19. Here point A is the established throttle state of an existing
plant. An ideal superposed turbine with throttle state B would liberate BA Joules
of work per g flow. Since turbine expansions
produce some irreversibility, the throttle state
needed is point C, so chosen that line CA is rep¬
resentative of the possibilities of a real turbine
able to produce the same work, per g flow,
as the ideal. It is seen that the effect of internal
friction is to increase pressure and temperature,
but in such a way that the degree of superheat
at C is* less than at B.
The pressure Pmax and temperature fma, are
assumed to be the highest for which commercial
equipment is available. The maximum power
capacity would be had in the superposed tur¬
bine were the throttle state C extended upward
until it reached either the limiting pressure or
temperature as at C\ The state Cmax could be
used only if by coincidence state A were so lo¬
cated that a normal turbine expansion Cmnx A
were possible.
The expansion is wholly in the superheated
steam region, and line CA should be approxi¬
mately straight if the superposed turbine is of
the usual multi-staged form. Stage eflSciency
remains substantially constant in the super¬
heat region. The slope of line CA is therefore a
function of the average stage efficiency. The
selection of point C will be illustrated by
example.

Example 1: Steam conditions of a superposed Determmation of

plant will be estimated based on an assiuned 30,000 throttle steam state of a super-
kw existing plant having Po ^^*1 kg/om* ab, posed plant:
<0«. 254.4*’G. The steam rate of this plant is A, Throttle state of original
C.tt7kg per kw hr. An additicmal 12,500 kw is plant.
%fianted from the superposed unit. Turbines for C, Throttle state of super-
this duty may be assumed to have average stage ^ posed plant,
ellloiensy between 75 and 80%. 78% is used.
The existing plant fixes the exhaust condition of the new unit, i.e., 15.1
kg/om* ab, 254.4°C, 2934.4 J/g, 170100 kg per hr. To yield 12,500 kw on this fiow the
superpceed unit must convert „ 864.6 J per g into output. This
would be ^e length AB of Fig. 8-19 exeept for meehanioal-eleotrioal lossee, and ^e
effeot of extraction to high-pressure heaters. If *6% of the fiow were extracted,
tlm aywage fiow toough the snporpoeed turbine might be of the order of 97 to 98%
of ite tbrotHe flow. A factor of 0.975 is thus introduced into the pretlminaiy ostiniet<~
240 VAPOR CYCLES
of point B.* AMuming a meontinieal-eleotrieal effloioncy of 96%, tbo heat releaee
AB is found.
Aab = 264.6/(0.976 x 0.96) m 282.7 J/g

Ab » 2924.4 + 282.7 - 2217.1 J/g

Now point C is located by trial as that pressure at 8217.1 J from which an isentropic
282 7
expansion to the 15.1 kg/om* line will produce Ah * r-zr * 302.4 J

At h « 8217.1 - 362.4 • 2864.7 J. and p = 16.1 kg/om*; 8 = 6.580(= «c)* On a

MoUier chart it will be seen that at 3217.1 J and 6.6S0 entropy, pc = 61.2kg/om* ah and
to = 417’*0. These are the approximate steam conditions requir^atthe throttle of the
superposed turbine.

Example 2: An extension of Ex 1 will show the limit of superposed power realiza¬

ble at this plant. Assume the same average stage efficiency, and commercial limitation
of 08.4 kg/om* ab and 510°C. The pointa A and O from Ex 1 when plotted as in
Fig. 8*19 define a condition line that may then be extended until it reaohee the
thermodynamic «<oeiling*’ at In this case the pressure Pmax hi the limiting factor.
A chart solution fields
= 476®C 3316 J/g

Thenjoultsonverted into work perkg exhaust flow, employing same assumptions as

in Ex 1, are
h « 3815 > 2034.4
356.3 X 10» X 170100
aqd superposed power « 16830 kw
3.6 X 10«

It is possible to increase the capacity of an older plant by a third, at the

same time decreasing the over-all steam rate by a third. The new capacity is
obtained at a moderate cost because no additional condensing equipment is
required and the extra high-pressure equipment is rather compact because of
the dense steam in that region.
Control presents a problem which is solved in different ways, depending on
the flow plan. Where the superposed turbine exhaust is not the entire flow to
the low-pressure section, some of the original boilers will be delivering steam
into the original steam header jointly with the topping unit. Low-pressure
boiler production will be held to the minimum by loading the topping unit fully.
While it is doubtless true that the old power plants suitable for topping
with a superposed turbine are becoming fewer, this technique in economic powtx
development may continue to be applied from time to time in allied fields. A
utility primarily producing steam for district heating could generate cheap
kilowatt hours if topped. Institutions with a heating steam load can produce
part of their electrical power requirements by the installation of a steam tur¬
bine, although this, of course, would be topping a heating cycle instead of a
vapor power cycle. Industries often have similar situations, which brings us to
the topic of the next section.
8-9 Industrial Heat-Power Balance. It is in the industrial field that the
* Subsequently an exact calculation could be carried out, using the necessary extraction!
data.
INDUSTRIAL HEAT-POWER BALANCE 241
ingenious designer may find the greatest opportunity of interrelating the heat
and power needs, so that these services may be obtained with the minimum
expenditure for fuel or other purchases. An almost unlimited number of possi¬
bilities are open in any project calling for supply of these two forms of energy.
Only a certain fraction of the heat available at any temperature may be
converted into power by a working expansion. It is as follows:

Fraction available energy = (8-13)

hi

However, heat originates from the combustion of fuels at very high thermal
levels. Although the availability feature may limit the extent to which energy
can be converted into mechanical power, it is possible that the requisite power
may be developed during the process of providing the heat where a need exists
for both low-temperature heat and power.
In other words, the thermal level of the origin of the energy may be placed
high enough so that the available energy may be skimmed from the heat and
diverted to power service before the remainder, or less available portion, is
delivered for heat service.
The possibilities opened up by this fact are numerous and have led to many
ingenious plant designs. Naturally enough, some of the simpler, but less
efficient methods, such as a low-pressure boiler for the supply of heating steam,
and Diesels for power, have been abandoned in favor of a single, balanced heat
and power system. The simpler system might be especially good for meeting
variable load flexibly, but will usually show higher investment costs and lower
efficiencies, due to the necessity of wasting at least part of the heat from the
engine cooling and exhaust systems.
Where it can be satisfactorily worked out, a balanced heat and power supply
will generally be economical. In this respect, industry enjoys an advantage
over the public service utility, whose plants almost invariably are based upon
the idea of extracting the maximum available fraction of the heat energy by
use of high pressures, condensers, extraction cycles, etc., because there is no
opportunity to use the vnavailable fraction of heat, which is abandoned as
heated condensing water. With expensive installations and skilled operation, it
scarcely expects to skim off more than 30% of the heat value of the coal as
mechanical work.
The industrial operator is ordinarily not under the same necessity of obtain¬
ing the most efficient operation because his rejections of heat from the prime
mover can be used. Frequently, however, the heat and power needs do not
coincide in variability; that is, the maximum power demand may occur at an
entirely different hour of the day from that for heat, and it is not always possi¬
ble to secure a storage or an averaging effect which will allow heat and power to
be balanced. Furthermore, there sometimes exist seasonal variations with the
result that, although there may be a synchronization of heat and power demands
at one season of the year, at another season they are out of step. In such cases a
portion of the power may be generated by the industry in the course of produc¬
ing its beat needs, and the balance purchased from a utility.
In a simple case where the right proportion exists between steam and power
242 VAPOR CYCLES
needs, it is often possible to obtain a reasonably satisfactory balance between
steam and power generation by the simple expedient of selecting boiler pressure
and temperature so that the requisite power may be produced by operating the
prime mover something in the manner of a reducing valve, except that in the
process of pressure reduction the requisite mechanical energy is extracted from
the steam. In some cases it may be advisable to float a steam accumulator on
the process header in parallel with the steam turbine designed for variable back
pressure. This would serve to equalize short-time, out-of-step fluctuations of
heat and power demands. This same objective is usually attained by relief
valves, alternate power sources, and pressure reducing valves, but under some
circumstances the accumulator would be more economical.

Fig. 8-2Gt Flow diagrams for industrial plants, (a) Simple case of proportional heat
and power demand, (b) Example of system for industrial heat and power. 1. When
power requirements exceed process steam, load is partially shifted to condensing turbine.
2. When process steam is insufficient, load is shifted to noncondensing turbine and
steam deficiency, if any, supplied through pressure reducing valve.

The setting of boiler operating conditions is a procedure similar to that of

the superposed cycle. The exhaust state is prescribed by the industrial heat
requirements, and turbine throttle state is enough higher to yield the desired
power.

Example 1: An industrial plant having equipment connected as shown in Fig. 8-20

(a) will use 4536 ka per hr of dry saturated process iteam at 2.11 kg/om* ab pressure.
At this output 507 turbine hp wUl be needed. The entire flow from the turbine is to
be exhausted into the prooess mains. The steam conditions that should prevail at the
boiler outlet in order to secure a balance of heat and power wiU be invesUgated,
It is obvious that there must be a flow of 4530/507 8.95 kg steam though the
turbine for each hp hr developed.
Let Ah = heat to be skimmed from each pound of steam flow. Then Ah X rim of
turbine * output energy. Assume rim = 94%.

Ah = 2.65 X 10*/8.95 xl0» x 0.94 * 315.1 J/g

From tables, entropy of process steam = 7.1157, and*enthalpy = 2708.2 J/g

Enthalpy of steam at throttle = 2708.2 -f 315.1 — 3023.3 J/g.
The throttle state is now chosen so that at an enthalpy of 3023.3 J/g the en¬
suing turbine expansion will end with dry and saturated steam at 2.11 kg/om*. Taking
into oonsideration the prevailing values*of stage effioiency (70-80%), the expansion
COMBINATION CYCLES 242
ig thmi workttd out on tho MoQlor Ohort m osploinAcI in 8eo 8*8. An initial itato of 18.47
kg/omt gk and 88l%l ahonld bo about rtgh^

This elementary case has been presented for the purpose of showing how
such an item as boiler pressure may be selected to help obtain a balance between
power and heat. In practice, an industrial heat ix)wer balance is generally a
matter involving more complexity and some considerable amount of compro¬
mise. At higher process steam pressures or temperatures, and where power needs
are larger in relation to process steam, the boiler pressure and superheat will
necessarily be increased. Such increases may leave the terminal condition im¬
practicably high.
An industrial process will frequently be found to require steam at two or
more pressures. Extraction-type turbines have received considerable applica¬
tion in these cases.
Other possibilities for industrial steam cycles are provided in instances
where considerable quantities of distilled water are employed. The use of multi¬
ple-effect evaporators for this service provides the opportunity of utilising the
exhaust of a back-pressure turbine.
In general, where industry furnishes some process steam requirements, it
will be good practice to furnish generating equipment sufficient to pass the
process steam, but it may not necessarily be economical to install condensing
equipment beyond this point if purchased power is available at reasonable
rates.
8-10 Combination Cycles. Modifications of the basic vapor cycles have
occasionally been used in the interest of greater economy of operation. The
reheat and regenerative principles are readily combined, as has been mentioned.
Certain combinations of vapor cycles with the gas turbine are technically in¬
teresting and may possibly be developed to some extent. A series combination
of two vapor cycles has certain advantages and a limited commercial importance
at present.
Water vapor has physical properties not altogether desirable at either end
of the expansion range, but no other common substance is a better compromise
for both extremes than water vapor. Thus, as the development of vapor cycles
has proceeded, it has been with the steam boiler and prime mover. Nevertheless,
the fact has been constantly before engineers that there were vapors which
were thermodynamically better suited than steam to one or the other ends of
the expansion range. On the low-pressure side sulfur dioxide, hydrocarbons such
as butane, propane, petroleum ether, etc., have been mentioned, while mercury
and diphenyl oxide are attractive in the high-pressure region. In order to realize
better efficiencies in heat power plants the temperature range ri-r2 has to be
increased. The increase can be accomplished by the following means, and at
the expense of disadvantages as noted.
1. Increase of degree of superheat.
(a) More investment in superheaters.
(b) Only the heat of superheat is at the elevated thermal potential.
2. Increase of boiler pressure.
(a) Quickly arrived at point beyond which major pressure increase was
needed for minor temperature increase.
244 VAPOR CYCLES
(b) High pressures mean thicker, heavier, and more expensive construc¬
tion in boilers, piping, turbines.
(c) Higher temperatures at high pressures weaken the metal in tensile
strength, aggravating the previously named disadvantage.
3. Decrease of exhaust pressure.
(a) Enormous volumes of steam made design of low-pressure end of
turbine or engine a compromise between size and possible efiBciency.
Condenser large and costly.
(b) Maintenance of the high vacuum is troublesome due to air leakage
and noncondensable gases in feedwater.
The mercury-steam is the only binary vapor cycle operated on a commercial
scale at the present time. As excellent results have been obtained in actual
operation, it is judged advisable to describe it.
The advantages of mercury as a vapor cycle fluid are these;
1. It has moderate vapor pressure at higher fluid temperature. For example,
at 538^ C its saturation pressure is less than 14.1 kg/cm* ab.
2. The liquid has high density, a desirable property in securing separation
of vapor from liquid in the boiler, in feeding liquid back to the boiler under
hydrostatic rather than pumping head, and other advantages.
3. Low specific enthalpy results in moderate jet velocities in the turbine,
enabling simple turbines ot tew stages to be used.
4. It is an element, therefore stable. Also it has been found possible to give
mercury a chemical treatment that causes it to spread in a film over the walls
of the heating surface with attendant rapid heat transfer.
Against these advantages one must consider the high cost of mercury, the
apparent limitation of supply, its toxic qualities, and its pervasiveness.* Since
the whole boiler, not just the superheater, operates at the very high initial tem¬
perature, the support and expansion problems are more difficult of solution than
in steam boilers.
Mercury systems have been built as superposed units on existing steam
plants whose pressures were too high to warrant consideration of superposed
steam cycles. They have also been built as binary vapor cycle plants where
mercury and steam sections were designed and built together as a unit. There
are not many mercury plants in the United States at present, but enough to
have established their practicability and commercial success. Some recent
developments in mercury vaporization have greatly reduced the ratio of system
charge to system circulation. This is of benefit in reducing the cost of the
initial charge of mercury per megawatt* of capacity.
The flow diagram in Fig. 8-21 may be taken as representative of the
mercury-steam cycle in its present stage of development. Here the liquid mer¬
cury feed is under hydrostatic head, but mercury feed pumps have been suc¬
cessfully used where it is not convenient to arrange the necessary static head.
The boiler is mainly radiant heating surface. The high density of mercury
creates much more pressure differential between the bottom and top of the
boiler than in steam equipment. Liquid mercury expands far more than water
when heated and this, coupled with some flashing action as the mercury rising

Mercury can seep through joints or cracks that would be impervious to water or steam.
COMBINATION CYCLES 245
in the furnace tubes meets decreasing pressure, causes strong circulation and a
^^foggy” rather than a liquid mixture in the upper sections of the heating sur¬
face. Because of the resulting increase of volume, the heating surface is suc¬
cessfully cooled by a charge of mercury that fills only about 10% of the internal
volume of the boiler initially.
The saturated mercury vapor flows from the boiler directly to the turbine,
which is usually a few-stage, simple, impulse turbine overhung in its casing
from the low-pressure end so that there is no high-pressure seal required. Alter
expansion to a high vacuum the vapor is exhausted to a '^condenser-boiler.^'
This is a compact, high-duty heat exchanger which will boil water with heat
transferred from the still-hot condensing mercury vapor. That this is readily
possible is shown by the operating condition of a recently installed plant. The
mercury vapor is exhausted at 0.19 kg/cm* ab, while the steam boiler section

Fig. 8-21 Elements of a binary vapor power plant.

is under 47.8 kg/cm* ab pressure. However, at 0.19 kg/cm* ab the saturation

temperature of mercury is 275.6°C, while the saturation temperature of water
at 47.8 kg/cm* is 260°C.This leaves 15.6® temperature difference for driving
heat out of the mercury into the water.
The condensed mercury is returned by gravity to the boiler drum. To over¬
come a difference of 8.1 kg/cm* between the mercury boiler drum and the
condenser-boiler, the mercury liquid level in the latter would not need to be much
over 6.1 m above that in the drum.
The steam from the boiler is first sent through a superheater located in the
gas passage of the mercury boiler, then delivered to the plant steam header
from which, in an industry, process and power steam could be drawn.
A binary vapor cycle power plant requires more investment per kilowatt
capacity than a steam power plant and should be given a base loading in order
for its high thermal efficiency to justify the cost. The steam portion of the
plant should be of a high-performance design also. The regenerative cycle is
commonly employed.
To illustrate the thermodynamics of the binary vapor without involvement
 246 VAPOR CYCLES
in practical details a case will be analyzed, making the following assumptions:
1. The mercury vapor expands from a dry and saturated condition isen-
tropically to the saturation temperature of the steam boiler. In condensing it
gives up its latent heat only, then is returned to the boiler.
2. The boiling water is at the temperature of the condensing mercury. (No
temperature difference in the heat exchanger.)
3. Steam expands isentropically from a dry and saturated condition.
4. The regenerative cycle has an infinite number of heaters which heat the
feedwater to the saturation temperature of the steam boiler.
Example 1: The mercury boiler pressure will be taken at7.0S kg/era* ab, quality dry
and saturated, with ^25 4 mm Hg aba as the pressure in the steam condenser In order
to keep the steam pressure as low as possiMe tiie mercury vapor is expanded to a pressure

Fig. 8-22 Ideal mercury-steam vapor cycle working between pressures of

7.03kg/om2 ab and 25.4 mm Hg abs.

of 0.07 kg/om*ab oorreepODding to a temperature of 230.7^0/and a saturated siSeam

pressure of 32.13 kg/om* ab. Kg for kg, the heat content of mercury vapor is about a
tenth that of steam. If the T-s diagram is to be drawn for 1 g mercury, only a fraction
of a g of steam is involved. The fractional part can be determined as follows using Figs.
A.2, A.3.
Tracing the isentropic expansion line on the Mollier diagram for mercury vapor, we
find the enthalpy at 0.07 kg/cm* ab » 253.5 J/g. The heat of the liquid being 32.6 J, 220.9 J
are availaUe for vaporizing the water per g mercuiy flouring. The latent heat of water at
22.18 kg/em*ab is 1781.1J ; therefore the steam flowing pergHg 229,9/178l.l =
.0.124 g. In laying out the T-a diagram, this fractional part of the entropies must be
used. The zero entropy axis for mercury will not coincide with that for steam. However,
by virtue of the calculations just made, the entropy of vaporization CF is the same;
heiice the T-s cycle for mercury is s’uperimposed on that for steam.
VARUBLE LOAD OPERATION 247
Heat input » JCDEFI ■■ Ai — ^ (for mercury).
Heat input - S66.9 - 3S.6 » 1SI.9 J/g.
Work output ■> Area BCDEFO, but since by the assumptions made for the regenera-
tive cycle, BCK * OFH.
Work output « KCDEFH - JCDEFI - JK X KH, where KH - 0.124 X s#g.

Work output « 82t.a - mi x 0.4SM « 193.9 J/g.

Efficiency - 193.6/393J -60.0%.

Efficiency of Carnot cycle KD'EH, re - 60.6%.

The difference between Carnot and binary vapor cycle efficiencies is to be found in
the area CUD, This fact illustrates the advantage of the steep slope of the mercury
liquid line CD. The efficiency of the steam cycle alone would have been equal to
between 830.7^and 26.1^.
936.7 - 96.1
17r«g = 41.3%
‘ 986.7 + 278

By adding a meroiury vapor eyole working aS 7.08 kg/om* tbe efficieney of tlie38.13
kg/cm> plant has bean inoreaasd (60.0 — 41.3)/4U ■■ 45.4%^

In practice the performance would fall short of these very favorable results.
The reasons for the discrepancy may well be understood by again examining
the assumptions made for the cycle. It is advisable to superheat the steam.
Between 11® and 17® temperature difference is used to transfer the heat from
condensing mercury to boiling water. The simplicity of the T-s diagram is
further destroyed if some feedwater heating is performed in the boiler because
of nonregeneration to the saturation temperature: Yet, withal, this cycle is of
interest to heat power engineers because of the high thermal efficiency of which
it is capable.
Published data on a 40,000*kw plant placed in service in 1950, which was
designed as a binary cycle plant, not a topping of an existing unit, give it a
9707x 10* Joules per kw hr over-all heat rate on base load,^ with oil fuel.This
corresponds to an over-all efficiency fi^re of 3.6 x 10*/970’7 x 10*, or 37.1%.
The vapor cycle efficiency of this plant is doubtless in excess of 40%.
8-11 Variable Load Operation. Turbines are most efficient in their use
of steam when the normal inlet valves are fully open, i.e., when theje is no
throttling of the steam flow entering. The previous discussions of vapor cycles
have assumed this condition, as have the turbine condition lines in the Appen¬
dix. In use, these machines are subjected to variable power output demands by
their generators and must be governed to continue at constant speed. The
governing action* is usually a ‘^throttling” process, a constant enthalpy pres¬
sure drop produced by a partially closed valve. Turbines can also carry some
overload if they are arranged for admitting some throttle steam at intermediate
stages. This is not the place to describe turbine details; howeveri some general

*Tbe govemiiig action of huge turbines is sometimes a combination of throttlmg and

nossle cutout features, for which no simple method can predict the part4oad oonditimi
line.
 248 VAPOR CYCLES
information is necessary background to an understanding of part load operation
of vapor cycles. When operating at part load the plant conditions are these:
1. Throttle steam pressure, condenser vac¬
uum, turbine speed, generator voltage and
frequency all remain constant.
2. Generated superheated steam tempera¬
ture will vary, but variation may be neutralized
with superheat control.
3. Rate of flows throughout cycle will de¬
crease.
4. Steam pressure on downstream side of
governor valve (i.e., the throttled pressure) will
vary approximately in a straight line with
intercept relation to load. Pressures at extrac¬
ENTHALPY J/g

tion openings similarly vary.

5. Total steam passed is directly propor¬
tional to inlet pressure.
At part load in a regenerative cycle there
is a general decrease of pressures and feedwater
temperatures. The condition line of the turbine
will be displaced on the Mollier diagram sub¬
stantially along constant enthalpy lines to
pressures corresponding to the degree of throt¬
tling at part load. To estimate the feedwater
temperature and flows in a regenerative cycle
at part load, the displaced condition line is
approximated by estimating a throttled inlet
pressure, then paralleling* the full-load line
from that point. The extraction pressures are
Fig. 8-23 Part load steam tur¬
then estimated and enthalpies located on the
bine condition line.
condition line. Then heater conditions are com¬
puted, using appropriate terminal differences and extraction line losses.
Example 1: The working conditions of Heater C of the regenerative cycle con-,
sidered in the example, Sec 8-5, will be estimated for operation at 14,500-kw output.
The intercepts mentioned in the foregoing item 4 will be assumed to be 6% of the full¬
load pressures. The full-load throttle pressure was 24.61 kg/om* ab; extraction
pressure, 5.88 kg/om* ab; hence the following equations are formed for part-load pressures.

Vi ^ 0.06 X 24.61 + —^ (0.95 x 24.61i 1.281 f 0.0011692L kg/om* ab

p.' » 0.06 X 5.88 + ^ (0.96 x 6.88)= 0.294 + 0.0002791£, kg/cm* ab

tfUyUUU

At load L ~ 14,500 kw, pi wl8.21kg/omS aband po » 4.86 kg/om* ab

Fig. 8-23 shows how the part-load expansion line is estimated and the new state of
the extraction to Heater C determined.
Again assuming 6.6* difference between saturation temperature at casing pressure
♦This assumes that stage efficiencies are not adversely affected by part load flows, a
jreasonable assumption for part loads in excess of 75% of rating, but one of increasing un¬
certainty below 50% load.
 VARIABLE LOAD OPERATION 249
and feedwater temperature out of C, the feedwater temperature is 294^ — 10, or
146 — 6.6 or 140.4®C. is determined similarly fo and is found to be 1.30 kg/cni* ab,
for which saturation temperature is 106.6®. Hence feedwater temperature entering Heater
G is 106.6 6.6 == 101®C.
Extraction flow to C at part load = at enthalpy he' taken from the condition
line at Pc»
Heater C balance: (1 — (690.86 — 423.26) == w'^ (2823.7 - 600.86)«
Extraction flow to C =0.0696 kg per kg throttle flow.

Heat Rates, For variable load analysis the Heat Rate, HR, is more useful
than thermal efficiency, even though they are related in a simple manner.

3.6x1(PVHR (8-14)
The unit of HR is Joules per kw hr. The heat input to a vapor cycle, or a plant,
typically takes the form shown in Fig. 8-24. A no-load or idling input is

Fig. 8-24 Typical heat plant input- Fig. 8-25 Heat- and increment-rate curves
output relation. derived from Fig. 8-24.

required to satisfy friction and radiation losses. As output load L is produced,

the injiut increases somewhat proportionately over a range of load increase,
then tends to more than proportional increase as the output approaches the
limit of which the system is capable. The heat rate curve is derived from these
data by dividing the rate of heat input at any point by the load. The heat rate
curve will show a minimum value at the most efficient load.
A derivation of the heat rate is used to fix the best division of total load
between two or more producing units. The increment heat rate, R, is the addi¬
tional heat used to produce the next kw hr of energy. Mathematically it is the
slope of the input-output curve. The R vs Jj curve, therefore, can be drawn by
mathematical or graphical analysis of the input-output characteristic, or by a
step-by-step procedure whereby-^—is plotted against the mid-
increment load
* 1] « 869.9/HR. if HR is in koal per kw hr.
260 VAPOR CYCLES
point of the merement load chosen. The increment rate curve of necessity 'will
pass through the minimum heat rate. It can be shown* that the most economical
division of load between two producing units occurs when their incremental heat
rates are equal. For this reason incremental heat rate data are frequently cal¬
culated as an aid to system load dispatching.

_m

0 too 200 300 400 800 600 700

788
Docroasw in plont not hoot rato, ke<kl/k« hr
.3168
J_I_L J_1_L-.-I_L
0 800 1000 1800 2000 2800 3000
Oocroaot in plant n«t haat rotO^ k Joulot/kw hr

Fig. 8>26 Chart for estimating justifiable added investment in equipment to effect a
reduction in heat rate.

Station incremental rate curves may be constructed for different combina¬

tions of operating boilers and turbines if the equipment increment rates are
available.

fisT = (8-15)
1 — /v*

where Rt and Rb *= turbine and boiler increment rates.

Rm = Ratio of increment Of auxiliary power required to incre¬
ment station load.

PROBLEMS
1. Plot an ideal Rankine vapor cycle, as specified, to scales of 1 cm a(r&aiidieiii=
0.6 J/g/^K entropy. Initial state at 14.1 kg/om> Bh, 66.6* superheat. Quantity 1 kg. Atmos-
pherto exhaust. Find :Ya) i}|^; (b) work done per cycle.
2. A simple steam engine plant has boiler feedwater at 05.6^.. Steam is supplied
to the engine at 7.0S kg/caiS‘ab, 97%diy.Atm^herio exhaust. Steam rate» 9.3 kg per ihp hr.
^, 87%. Generator of 91.5% efficiency is direct connected to the engine. Find: (a) lya,*
1
(b) lyt baaed on ihp and bhp; (c)
* Steinberg and Smith, Economy LdodUng of Power Plants and Electric Syeteme, ‘John
Wilay and Sons, Publishers.
PROBLEMS 251
3. Plot a Rankine vapor cycle, as specified, to scales of 1 om » WK and 1 em »
0.5 J/gr^ entropy. Also to eoale of 1 om 0.5 kg/cm* and I om or 0.35 m*. Initial etate^
10.6 kg/om* ab, 27.8* euperheat ; exliaiiet,0,,a kg/em* ab »2%d?y, <bedta.S*C quantity 1 kg.
Find ; (a) ; (b) work per cycle ; (e) ijt.
4. Plot the incomplete expansion Rankine vapor cycle to scales of 1 cnn«0.5kg/em>
and lem»0.05m< for a quantity of 1 kg.Calculate of this cycle. Initial state, 5.27
kg/cm< ga, dry and saturated ; exhauec preemre, 1.05 kg/omS ab ; release at 1.05 kg/omt ga.
5. Repeat Prob. 4. but lor initial state 9.14 kg/om^ ab, 98% dry,exhaust at1.05 kg/cm>
ab, release at 1.41 kg/om* ab.
6. A 75-kw turbine-generator has steam rate of 12.7 kg per kw hr. Steam at 12.8
kg/cm* ga, 66.6® superheat; exhaust, 50.8 mm Hg abs. Find : of a simple vapor cycle
incorporating this unit.
7. Construct a scaled heat stream setting forth the results of Ex 3, Sec 8-3. Scale
6 cm = heat input in coal.
kgfatFai
8. A steam power plant has operating 7m-0.90
conditions as shown in Fig. 8-8P. Determine £
all flows for a boiler output of 1 kg per
17-0.75
min and record same on a copy of the flow
diagram. Calculate the gross vapor cycle
efficiency and the plant efficiency. Main 0

exhaust at 0.21 kg/cm* ab. 90% dry. Nag.

Fto. 8-8P
lect pii>e line friction and static heads.
9. Construct a scaled heat stream of the energy flows found in Prob. 8. Scale:
6 cm = heat in coal.
10, Draw a flow diagram of a Rankine vapor cycle steam power plant. Steam engine
drives 150-kw generator of 90% electrical efficiency. Steam rate,6.7 kgper bhp hr. Steam
^pressure, 10.55 kg/om> ga, 65.6* superheat; exhaust to condenser at 15.2 cm Hg abs.
feedwater heating. Motor-driven boiler fbed pump. Find : (a) ; (b) iQtb * (®) ^<c.
11. Find the over-all heat rate and thermal efficiency of the plant shown in Fig.
8-1 IP. Boiler efficiency is 75%. Steam flow to pump, when generating 350 kw,is 181.4
kgper hr. Working head on pump =112.8 m. Neglect mechanical losses of pump.

12. Using Eqs 8-9 and 8-10, determine the steam rate of a turbine whose condition
hne is assigned from the group given in the Appendix.
13. A Rankine vapor cycle type power plant has a turbine which operates on the
condition line B, Appendix, when generator output is 9450 kw. Other plant test data
are: Steam generator efficiency, 0.875; steam flow, 719 kg per min; auxiliary power
used, 225 kw. Calculate the heat balance of tlus plant and express same as a heat stream
using scale of5 eiii*«Heat input in fuel.
262 VAPOR CYCLES
14. Plot the ideal regenerative cycle of maximum efficiency to scales of 1 cm «.
4(PK and 1 cm ss 0.75 J/g/°K entropy. Initial state, 14.06 kg/om^ ab dry and saturated;
exhaust 0.11 kg/cm* ab. Caloulate ^reg.
15. Compare the Rankine vapor cycle with the regenerative cycle of maximum
efficiency on the following bases. For each, initial condition, 28.12 kg/cm* ab, 816.0®C,
0.14 kg/cm^ ab exhaust. Find ; (a) vapor cycle efficioncy ; (b) relative volumes of exhaust
steam ; (c) kg-m work obtained per kg generated steam.
16. Calculate the efficiency of an ideal regenerative cycle operating with steam at
31.64 kg/om^ ab, 398.9®C ; exhaust pressure 25.4 mm Hg abs. 187.8®C. Sketch this
cycle on T-s plane.
17. Plot the ideal regenerative cycle on T-$ axes to scales of l cm 40®Kand 1 cm
ss 0.5 J/g/*’K entropy Tor a quantity of 1 kg.Initial pressure, 42.18 kg/cm* ab ; temperature,
537.8®C ; condenser pressure, 0.07 hg/cm* ab ; 2C4.4®C. Find lOreg.
18. The flow diagram of a one-heater regenerative cycle is shown in Fig. 8-18P.
Calculate the necessary quantities and draw a scaled heat stream of the plant, based
on 1-kg coal input. Scale 1 cm** 2500 x lO^ J. All pumps are motor-driven and,combined,
take 1% of the generator outjuit.

pJL-, 28*12 kg/cm^ab 337'B’c

27912J/oJ^°°‘^j 9*62 k ./enfob

E«h.<M)7li9/em»ob
0*696
U8<9C
Fig. 8-1 8P

19. A power plant installation reports turbine operating conditions as §hown by

Fig. 8-19P. Lay out a flow diagram of a plant that would use this unit. Condensate
plan as in Fig. 8-6(b). a and b to be low-pressure heaters. Assume 7% pressure drop
in extraction lines, 2.8® heater terminal differences, and label diagram with resulting
temperatures.
Extr T/g
Slo"c • 86
d
c 7.03 7
b 2-07 2698 7
4'9-9t»twi Hg 0 0'63

Fic. 8-19P
20. The turbine of Prob. 8-19 has the following no-load pressures. Inlet
6.62 kg/om* ab; e, 1*76 kg/cm^ ab; d, 0.84 kg/om^ ab ; o, 0,11 kg/om2 ab ; b,
71.1 mm Hg ; a, 69,8 min Hg, Plot the condition line to scale of 1 cm = 50 J/gand
1 om ■■ 0‘25 J/g/®K entropy. Then add an estimated condition line for 70% of
the oHginal load and locate thereon the extraction states e\ d\ etc.
21. Lay out the flow diagram for a four-heater regenerative cycle. TurbiiK^ to be
employed has condition line E,
Appendix. Condensate as in Fig. 8-5(b). Assume 5.6®
difference between feedwater out of heater and saturation •temperature at the ex¬
traction pressure. Also take condensate out of heaters at same temperature as feed-
water out. Record temperatures on all flow lines of the diagram, also pressures at bleed
points.
22. Employing data and results of Prob. 21, estimate the throttle and extraction
flows, kg per hr at 94,000-kw load, using Eqs 8-9 and 8-10. Also calculate the gross vapor
cycle efficiency. Report results in a manner similar to Example 1, Sec 8-5.
23. Design a two-heater regenerative cycle for a power plant whose turbine condi¬
tion line is assigned from the Appendix chart. Heater condensate is handled by traps.
Allow 2.8^heater terminal difference and 7% pressure loss in extraction lines. Assume
heater condensate cooled to outgoing feedwater temperature. Report the* design in
PROBLEMS 263
approximately the same fashion as used for the sample example, Sec 8-5. Draw the
flow diagram and label it with hourly flows at rated load. Neglect any make-up water
supply features when drawing the flow diagram.
24. Find the following heat rates for the data of Fig. 8-7.
(a) Station heat rate, Joules per kw hr.
(b) Vapor cycle heat rate. Joules per brake hp hr.
25. Calculate heat balances for the following elements of Fig. 8-7: (a) turbine-
generator; (b) deaerating heater.
26. Each of the following initial states is a possibility in a reheating cycle plant.
Consider that the final quality should not be less than 86% dry. Exhaust pressure-
0.07 kg/cm* ab. Determine the reheat conditions for maximum utilization of the avail¬
able energy if (1) reheat begins at the saturation point; (2) reheat is carried to the
original temperature.
(a) 112.6 kg/cm* ab, 663.6*0. (b) 66.2 kg/om* ab» 687.8®C. Work out each case on the
Mollier Chart, and dhow rei^ts by (no scale) sketch of same.
27. An ideal reheating cycle, based on the Rankine cycle, has initial state 87.9
kg/cm* ab, 398.9^C ; reheat at 28.9 kg/cmt ab to S98.9°C ; exhaust at 25.4 mm Hg abs.
Find the thermal efficiency of this cycle and compare with that of a Rankine <^e
operating between the same terminal conditions.
28. Plot an ideal reheating Rankine oyole to scales of 1 om == 50*C and 1 cm
« 0.6 J/g/^K entropy. Initial state, 112.6 kg/cm^ ab, 565.6^C. Quantity, 1 kg. Reheat
from saturation line, sufficient to produce a final exhaust condition of 0.07 kg/cm*
ab, 14% moisture. Superimpose a Rankine cycle having same initial state and same
final quality. Calculate thermal efficiencies of these cycles ; also that if the final state
of the Rankine cycle had been 0.07 kg/cm^ ab.
29. In a central power station having reheat and regeneration the steam generator
delivers 195,045 kg steam per hr at 103 kg/cm^ ab and 537.8X* to the turbine. After partial
expansion a flow of 162.552 kg per hr is returned fbr reheating from 27.63 kg/cm> ab,363.9*’C
to 25.62 kg/om^ ab. 637.8°0,after which it is readmitted to the turbine for complete expansion.
Feedwater is regeneratively heated to 230.6®C. Ix>»d,64,511 kw. Boiler blowdown, 1960 kg
per hr. Generator efficiency, 96%, Find the heat rate of the vapor cycle.
30. Assume that the plant shown in Prob. 8-18 is to have a superposed unit of maxi¬
mum capacity added to it. Draw the flow diagram of the complete plant. Superposed
unit stage efficiency » 72% ; thermal ceilings are 87.9 kg/cm^ ab and dlO'^C. Feed to high-
pressure boiler to remain 148,9''C. New boiler efficiency, 80%. By what percent are the
plant capacity and thermal efficiency increased by the superposition? Assume mech-elec
efficiency of superposed turbine, 90%. 112-5 kg/cr»f ga 5io*c
31. A recent installation of a superposed tur¬
bine operates with conditions as shown in Fig.
8-31P. Estimate the power that this unit ought
to be able to furnish. Mech-elec efficiency = 0.925.
32. The capacity of a power plant is to be
increased by superposing a high-pressure addi¬
tion on the existing 17.58 kg/cm3 ga plant. Existing
turbines are rated to take 369,678 kg steam per hr
at 287,8®C. Assume average stage efficiency of superposed turbine will be 76%; mech-
elec efiiciency, 0.93. What initial steam conditions would suffice to add 20,000 kw to
the plant capacity?
33. The plant of Ex 1, Sec 8-9, uses 283 hp at a time when process needs are
3674 kg steam per hr. Draw flow diagram, showing how a pressure reducing valve
would be installed to meet process steam requirements. Find: (a) weight of live steam
through reducing valve per hr; (b) quality of the 2,11 kg steam in the process mains.
34. The nonextracting condition line of an industrial plant turbine may be con-
254 VAPOR CYCLES
•id«r(Kl M ft line joining the following points on theXollier Chftrt.(l) 14.06 kg/omi nb,d and
8;(2) 5.t6kg/omSftb,0.M7 dry;(3)1.0$kg/oin*ftb,0.9a8diy Induetrial stenm is needed as fol¬
lows :544«3 kg per hr at 4«92 kg/em> ga, 2813.3 kgperhr at atmoepherio preasure. Extraction
pressureof 4.92 kg/om8 gait maintained by an Inbuilt valve gear. Assume that the action
of this gear is to throttle the steam to 4.57 kg/om* ga before passing it into the low-pres¬
sure section of the turbine. Find the available shaft horsepower. i)m 0.965.
35. The flow plan of an industrial plant is like Fig. 8-20 (b). Boiler pressure|17.6
kg/om* ab, 260^. Botbunits 100 kw rated oapaoity.Condenser vacuum. 63.5 mm Hgabs.Process
main pressure, 0*35 kg/om* ga. Full-load condition lines to be based on average stage efficiency
of 0.76 in superheated region, 0.70 in saturated region. Find division of load between
units when 1588 kg per hr process steam is required and 165 kw power is needed. Allow
0.90 for mech-elec efficiency.
36. Solve Prob. 35 if the steam demand had been 907 kg pei hr and the power
demand 115 kw.
37. Assuming that the answers to Prob. 36 are 71 kw, noncondensing, and 44 kw
condensing. Copy the flow diagram and record thereon all flows, in kg per hr. Assume
3J31 kg/cm* ab nozzle inlet pressure for no-load condition on condensing turbine.
38. Using the data of Fig. 2-14, draw an industrial flow diagram to meet variable
load operation by use of an accumulator. Process steam^ at 2.11 kg/cm^ab ; boiler j)ressure,
21.1 kg/cm* ab. At one manufacturing rate, power and steam are in balance, but at
other rates unbalance may occur in either direction.
39. Find the boiler steam conditions that would allow an industry, using 7484
kg per hr of steam at 7.03 kg/cm* ga, dry and saturated, to take 350 kw from this steam
before using it for heating. Make and record any necessary assumptions.
40. Mercury vapor, 98% dry, is expanded from 10.55 to 0.08 kg/cm* ub.It is then con¬
densed, giving up its latent heat to produce dry and saturated steam which is allowed
to expand to 25.4 mm Hg abs in an ideal regenerative cycle of maximum efficiency. Draw
the T-s diagram of the mercury-steam cycle and compute its efficiency. Make the
same assumptions as were used in the construction of Fig. 8^22.1 cm 40°K; 1 cm »
0.075 J/g/®K entropy.
41. Dry and saturated mercury vapor at 8.16 kg/om> ab is expanded in a turbine
to 0.19 kg/cm* ab. 'the following steam cycle finally rejects beat at 33.2'’C. Making the
same assumptions fuswere used in the construction of Fig. 8-22, except for 16.7°l9mpef»ture
difference in condenser-boiler, compute the binary vapor cycle efficiency, sketch the cycle
on T-s axes.
42. Find the working conditions of Heater B in the example, Sec 8-5, when the
generator output is 14,500 kw.
43. Find the working conditions of Heater C of the plant, diagrammed in Fig. 8-13,
at 6Q,000-kw load. Make any necessary assumptions and record'justification of same.
44. Diagram the probable condition line (h-s plane) of the turbine. Fig. 8-13, for
60,000-kw load. Assume simple throttling governing.
45. Work out a heat balance and draw a heat stream for the vapor cycle shown
in Fig. 8-7.
46. Find the heat rates and the station heat rates of the vapor cycles shown by
Figs. 8-7 and 8-13.
47. Two vapor cycles are to produce jointly, from the generators driven by their
prime movers, an output of 8000 kw. Their variable load characteristics are described
as follows:
(htlput, kw 0 mo JflOO 6000 8000
i/k*/No. 1 21.1 33.8 53.8 81.2 116.1
Input, m milhon boal/hr | ^ ^^ ^ ^ ^ j
Determine most economical division of load by the increment rate method*
PROBLEMS 266
48. Construct curves of station heat rate and increment rate vs. output between
0 and 40,000-kw load. Steam generator characteristic is represented by the equation:
Input Joules » 1.06X + 4.74 where X is the Joulc«*output. Obtain the in¬
crement rate of the turbine from its characteristic of steam'consumption, i.e., W =
15875 + 3.72Ir kg per hr« where L is kw load. Throttle steam has 3170 J/g.*
The increment rate of station auxiliaries is considered constant at 0.030 <^ouJe per Joule,
all auxiliaries being electrically driven.

Bqio will be Input keal i» l.OOX + > X 10-«X>, where X is koal output.
3170 J/g 757.8 koal/kg.
 CHAPTER 9

ENERGY FLOW IN THE STEAM

9-1 The Steam Power Plant. To obtain high-grade energy from fuel
via the external combustion, or vapor cycle, process takes a considerable ag¬
gregation of equipment if the transformation of energy is to be accomplished
as efficiently as possible. Something of this was implied by the vapor cycle
studies of the preceding chapter. A modern power station represents a large in¬
vestment in, literally, thousands of separate items, yet nearly all function to
transform or transfer energy. Transformations occur when one form is
changed to another. They are: (1) latent chemical energy into heat energy by
oxidation; (2) heat energy into mechanical work by expansions of a fluid
medium; (3) mechanical vork into electrical energy through the electromag¬
netic action in a generator; (4) electrical energy back to mechanical work in
electric motors; and, occasionally, (5) electricity to heat for convenience or
accurate spot heating. Transfers of energy are likewise numerous—^heat energy
transferred from fluid to fluid in steam generators, in condensers, coolers, and
heaters; and carried by these fluids from place to place in the plant’s pipe and
duct systems; electrical energy transferred by conductors comprising the many
circuits of the electrical system; etc.
When one has studied the steam power station via visualization of the fluid
flows and their changes of state, followed by an appreciation of the energy
flows also involved, he begins to comprehend the plant. The whole installation
can take on meaning and rationality when it is viewed as a plant for the purpose
of achieving certain energy flows that will ultimately produce the upgrading of
a considerable portion of low-grade raw material into the desirable electrical
output of the public utility system.
Some idea of this can be gained by inspection of Fig. 1-11. Coal which con¬
tains elements that oxidize with high heat release during the reaction is fed
into the plant. After preliminary preparation it undergoes this reaction with
oxygen, supplied by air, in the furnace of the steam generator. Here high-
temperature energy is produced, but almost immediately transferred to the
water which, on account of its confinement, becomes high-pressure steam. The
energy flow now rides the steam through pipes to the prime mover, a turbine in
this case. In this machine a working expansion to low pressure effects a trans-
256
functional relations 267
formation of some of the heat into highgrade mechanical work available as a
torque at the turbine shaft. Most of the remaining heat is transferred by the
condenser to masses of river water which it heats a few degrees.
The turbine came from the hands of its inventors already a highly developed
machine, and subsequent improvements have largely been in the perfection of
details and the building of large units at low cost. In the turbine we have the
heart of the steam electric station. The boiler and condenser are heat-transfer¬
ring devices, one operating at the high-temperature end of the cycle, the other
at the low-temperature end; the turbine, however, is the heat-xdilizing device.
Boiler and condenser transfer heat energy from one medium to another, but
the turbine transfers energy from one form to another—from heat to mechanical
work. The turbine’s job is, theoretically, vastly more difficult and the results,
consequently, much less impressive than those of its partners, the boiler and the
condenser. However, as has been shown by the contents of the previous chapter
on the vapor cycle, the conditions under which energy transfer is accomplished
in the vapor cycle show that the turbine is actually a highly perfected machine.
Next the mechanical torque is transferred directly to the rotor of the
electrical generator and used to overcome the drag of electromagnetic attrac¬
tion involved in the generator action whereby torque energy is consumed and a
current of electrons appears at an electric potential, this representing electrical
energy. This form is then readily transferred long distances, especially if raised
in potential (voltage) by transformers. This is the main energy stream, but
there are many secondary energy flows connected with increasing the percentage
yield of high-grade energy, motivating essential auxiliaries, etc. In this chapter
we shall examine some of the technical features of this energy flow, with special
emphasis on the transfer of heat, since that form predominates in steam power
plant equipment.
9-2 Functional Relations. We shall endeavor here to outline the method
to be used to cover the subject of steam power plant equipment while keeping
the functional relation to the main energy flow ever in mind.
As the more efficient power cycles came into use, the auxiliary equipment
required became more numerous and expensive, and its disposition in the plant
more of a problem. Equipment has been developed to the point where much of
it is justified on economic grounds. The complication attending the large amount
of auxiliary equipment tends to obscure the relationship and relative importance
of the component parts of steam cycle equipment.
Let it be remembered, however, that the whole action of modem electric
power production centers around three pieces of equipment: the boiler, the
turbine, and the condenser. No matter how extensive the remainder of equipment,
these three constitute the main power-producing group, and all the remaining
mechanical equipment services them. The modem trend is towards a boiler of
sufficient capacity to supply the entire steam requirements of one turbine; hence
the boiler-turbine-condenser can be thought of as the central unit of the power
plant. The design of any one of these involves some consideration of at least one
of the others. They are treated in Chapters 10 and 11. The principal factor
tying these three together is the working medium, which is delivered in suc¬
cession from boiler to turbine to condenser.
In order that the heat-absorbing and heat-utilising fimctions of the boiler-
258 ENERGY FLOW IN THE STEAM POWER PLANT
turbine-condenser group may be efficiently, safely, and economically performed
under the generally prevailing condition of variable load, it is supplied with an
eittensive number of auxiliaries; i.e., feed heaters, pumps, traps, fans, stokers,
etc. When the purpose and action of all these auxiliaries are understood, it will
be found that they naturally classify themselves into two groups: first, those
associated with the flow of the working medium of the cycle (i.e., water and
steam); second, those associated with the combustion of the fuel and the flow
of the resulting gases. Proceeding on this basis, the auxiliary equipment will
assume a position in one or the other of the two loops shown in Fig. 9-1. We
may call these the water and gae loops, and, by keeping them in mind, find that
the station layout and action, as far as auxiliary equipment is concerned, are
considerably clarified.

Fio. 9-1 Diagram of the steam power plant, illustrating how the equipment is viewed
as a boiler, prime mover, condenser group, with gas and v>ater service loops.

For the purpose of explaining the relation of Chapters 12 and 13 to this one,
let each be thought of as dealing^ with one of the loops diagrammed in this il¬
lustration.
9-3 Production of Heat Energy. The kinetic molecular nature of matter
and the character of heat energy have been alluded to in Chapter 1. While not
essential to the practice of power plant engineering, a picture of the processes
by which the heat energy is created should be of interest. Fuel molecules, say
H2, can exist without combustion in an atmosphere of oxygen at temperatures
below the ignition temperature. What, then, is the circumstance that produces
their union accompanied by the liberation of free energy? Also, how does this
differ from the origin of the nuclear energy which science and engineering are
so actively studying?
Combustion, Between the two nuclei of hydrogen gas there are both attrac¬
tion and repulsion forces. These vary differently with the distance between
nuclei and in such manner that there is one spacing of minimum potential
energy. This is the normal, unexcited, atomic configuration for the molecule.
The molecular energy then is mainly electronic. If temperature is increased by
gas compression or heat transfer,from an outside source, molecular vibrational
energy begins to appear. However, the molecule remains stable, although ex¬
cited, until the vibrational energy produces atomic motion violent enough to
overcome the force of attraction. There the molecule flies apart in an action
PRODUCTION OF HEAT ENERGY 260
called dmodatum. In the ease of hydrogen, free, excited atoma moving at hi|^
speeds are a product of this dissociation. These will 'collide with other hydtagen
or oxygen molecules, producing a variety of short*lived products, all of which
by further similar action go to a final stable end product, H2O.
As molecules of various velocities exist in a sample of gas exhibiting a eertiun
temperature, a few molecules may dissociate while the remainder remain stable.
The enei^ of the few dissociation products is absorbed without producing an
equal amount of active products, and there is no chain reaction. But at some
higher temperature the dissociation products are sufficiently numerous to pro¬
duce by impact more than are consumed, and a complex chain reaction follows.
The point at which this becomes possible is known as the ignition temperature.
It is affected by the type of molecules, the mean free path between ttem, and
their motion. Externally it is the point where the process of inflammation or
explosion begins.
Electromagnetic radiatitm is emitted both fimn the dissociated excited atoms
and the various intermediate reaction products. The region is tmporarily
saturated with high-speed collisions of particles. Electronic energy levels are
altered. The electromagnetic waves can absorbed by the more complex mole¬
cules in the vicinity, although diatomic gases seem to be transparent to it.
Finally, the end pr^uct of water vapor is reached with liberation of a finite
quantity of energy which is partly present as high moleculmr speed (tempera¬
ture) , the remainder having been discharged as infra red or visible li^t radia¬
tion. The combustion of carbon and of hydrocarbons is an even more complex
pattern of intermediate reactions thui for the single, diatomic gas molecule, Hg.
Nuclear Energy. Another source of heat energy is available in the nucleus
of the atom. Unsuspected in 1930, the subject of laboratory exploration in 1940,
H is, a decade later, undergmng large-scale study which may lead to the nuclear-
energy power plant. The possibility of realizing large quantities of energy from
a small mass of atomic “fuel” was introduced at the beginning of this book. The
most promising field lies in the controlled fission of the heavier, unstable atoms
such as Th**® and U®*®, or the artificial element plutonium, Pu®**. Because of
the terrific explosive power of a fast chain reaction in nuclear fissionable
material, the field of atomic energy in this country was, in 1946, brought under
strict governmental control. The Atomic Energy Commission was created to
direct and control work in research, production, engineering, and military ap¬
plications of nuclear energy. The Commission has power over the dissemination
of information that should be restricted in order to preserve national security.
This Commission succeeded the wartime Manhattan Engineer District. It has
continued the physical development of production of nuclear materials, licens¬
ing of manufacturing firms to operate government facilities, and other matters
associated with this, the latest and most important new field of physical science
and armament. It is clear that private industry cannot enter at will into wide¬
spread development of nuclear power plants. Only such development as is
federally authorized and licensed by the Commission can be undertaken. In
1962 there were several such official projects under way, the outcome of which
may be the production of kilowatt hours from nuclear energy.
In this conversion of energy it is thought that the nuclear reactor will re¬
place the combustion equipment and furnace of the conventional power plant.
260 ENERGY PLOW IN THE STEAM POWER PLANT
but the nuclear energy will appear as heat to be absorbed by some working
fluid and used in conventional prime movers. Thus the first nuclear power plants
might follow a plan somewhat like that implied in Fig. 9-2. The natural source
of fissionable material is uranium, a rather common element in that it is present
in the earth’s crust in about the same proportions as copper and lead. The
natural element is a mixture of and U“®. Energy is liberated when a free
neutron strikes the uranium nucleus; however, it is only the U*®® that is fis¬
sionable and natural uranium contains only one part in 140 of this isotope, the
remainder being nonfissionable IP®®. As each fissioned nucleus yields more

exchongsr

Fig. 9-2 Assumed heat-producmg end of an atomic-nred power plant. (From Good¬
man, Science and Engineering of Nuclear Power, Vol 1, Addison-Wesley Press, Inc.
Cambridge, Mass.)

than one new free neutron, there is a possibility of a chain reaction resulting in
continuous fission, just as when combustion is ignited it is possible to sustain
continuous combustion as long as any fuel remains. The question of whether
or not the U®®® and neutrons continue to ‘‘bum” or go out depends on the results
of a competition for free neutrons by: (1) Escape from the reactor; (2) non¬
fission capture by U®®®; (3) nonfission capture by impurities in the reactor;
(4) fission capture by U®®®. If item 4 produces more neutrons than are lost by
the others, then the chain reaction continues, otherwise it does not.
Fission is accomplished better by neutrons moving more slowly than those
which the fission itself produces. The fast neutrons are slowed suflSciently by
being passed through a moderator material (carbon, heavy water) and colliding
with molecules. Hence the uranium needs to be mixed with moderator or inter-
PRODUCTION OF HEAT ENERGY 261
spersed in a latticed moderator. Reactors are therefore homogeneom or hetero¬
geneous types. A chain reaction will not occur unless the uranium is present in
excess of a critical mass. The homogeneous reactor may be a sphere filled with
a solution of uranium salt in water, or with a slurry of moderator and uranium.
A heterogeneous reactor might be a large cube of pure graphite (a 'pile) with
numerous openings into which containers with purified uranium metal could be
inserted in a quantity to exceed the critical mass. Through other openings would
be inserted adjustable rods of material having the property of high neutron
absorptivity. Cadmium is such a material. Control of the rate of energy produc¬
tion is vested in movements of these control rods in or out of the pile.

Fig. 9-3 Fission by chain reaction.

Fig. 9-3 shows the chain reaction, as well as the production of a new arti¬
ficial element plutonium, Pu^®®, which, like is fissionable by slow neutrons,
but which may be chemically separated from U®®®, whereas U®®® can be isolated
only by expensive physical processes. In the pile large numbers of the slow
neutrons are absorbed by the U^®®, creating a short-lived new element, nep¬
tunium, which decays to plutonium.
The reactions are:*

92U®®® + on^ 92U®*® wNp»» +

9,Np2» 94Pu«® +

It is this ‘‘breeding” of new fissionable material during a fission process that

is largely responsible for the great interest in atomic power plants. Present sup-
* •n' represents a neutron, .le** an electron.
262 ENERGY FLOW IN THE STEAM POWER PLANT
plies and prices of fuels would make it unlikely that nuclear processes would
be developed to perform the same fimctions for many decades, except for the
likelihood of net gain breeding wherein some fissionable material is consumed,
heat energy is produced, but more new fissionable material is created than was
consumed. This aspect, obviously, has enormous implications for the future of
civilisation.
This whole subject is so complex, so changing, and so restricted on account
of the associated military potentialities of fissionable material that it is deemed
expedient to hold the present treatment to the few generalizations just covered,
plus the following picture of the source and nature of the energy.
The fission of a nucleus has been found to liberate about 200 Mev of
energy. The appearance of energy is necessary, for mass disappears in the
process. The fission products which are in the 35-57 atomic number range have
less combined mass than the original atom, and since energy + rest mass must
be the same before and after any change, the difference appears as energy.
Fission is preponderantly into two fragments of unequal mass. These both bear
strong electrc^atic charges which, being mutually repellant, propel the frag¬
ments in opposite directions at high velocity. Collisions of fragment with neigh¬
boring molecules then raise the thermal level of the whole region. This is said
to account for about 170 Mev of fission energy, the remainder being the energy
of fast neutrons evaporated from the fragments, gamma radiation, and other
minor results of the violence of fission.
This energy, which raises the thermal level of the region of fission, was re¬
moved and wasted by cooling water in the first atomic plants where the
emphasis was on bomb material. Now, with more knowledge of nuclear proc¬
esses, and with the accumulation of experience in operating nuclear reactions,
this cooling energy is being looked at with a view to its becoming the input of a
heat power plant. The reactor temperature for such service must be comparable
with that of the high-temperature fluids of modem power plants, i.e., 425- 656*
C, and this will introduce hundreds of difficult problems that scientists and
engineers must solve before'nuclew energy can be fed into a steam or gas turbine
power plant.
9-4 Thermal Level. The working process in a heat plant operatcjs be¬
tween a high-temperature source and a low-temperature sink of heat. At the
low-temperature end the level is set either by (1) the required exhaust
pressure or (2) by the temperature and quantity of natural water available for
condensing. Operating conditions of pressure and temperature at the upper
thermal level are more amenable to selection by designers having certain ends in
view. The value to heat cycles of high initial pressure and temperature has al¬
ready been assayed.
Selection of economic operating conditions for the boiler-turbine-condenser
group is the first and most important step in station design. The working
medium of the vapor cycle will cover a portion of the temperature drop avail¬
able between the temperature of the combining molecules in the furnace and
that of the circulating water in the condenser. The more of this range the de¬
signer includes in the vapor cycle, the more efficient the plant will be—also the
more expensive in first cost.
This temperature range is of the order of 1425®C down to 21®C , and it is
THERMAL LEVEL 263
the designer’s problem to decide what portion of this range will give him the
most economical installation. Limitations of available metals, necessary tem¬
perature differences, and cost of equipment cause the designer to use a small
fraction of the available range for the working cycle. The turbine exhaust
pressure is carried as low as the condensing water permits, on the basis of 2.8''
to 5.6" temperature difference on the tearm end of the condenser tubes. Boiler
operating conditions are much more debatable. The trend of steam power plant
practice is towards higher pressures and temperatures because of the advantages
of heat efficiency of the vapor cycle and of the decreased size of some of the
equipment. Standardization of operating conditions at the high-temperature
end of the cycle, although restricting the possible operating conditions to rela¬
tively few combinations, is nevertheless economical since industrial standardi¬
zation results in decreased manufacturing costs.
Generally the steam pressure at the turbine throttle valve is taken to be
the official plant pressure. Boiler drum steam pressure is, of course, higher be¬
cause of friction losses of pipe lines, valves, superheater, connecting tubes, and
steam purifiers, if any. For example, a need for steam at 26.4 kg/cm* ga at the
turbine would require a boiler pretture high enough to overcome losses which
might be .
Loss through piping and valves i.55 kg/cm*
Loss in superheater and connections 0.98 kg/cm«
Allowance for dry pipe in boiler drum 0.14 kg/oms
Total 2.67 kg/cm*

As the safety valve will not open until pressure has risen some 4% above
operating pressure, the maximum drum pressure is (26.4+2.67)1.04 ==30.2kg/cm*
g«. The boiler construction industry has a recommended pressure standard for
the construction of power boilers. These standardized pressures are 11.25,12.66,
14.06, andin increments of 1.76 kg above this. Hence the boiler design pressure
for the foregoing case would be 31.64 kg/cm* ga.
The pressure-temperature ranges used for steam power can be grouped as
follows:
8.8- 17.6 kg/om* : saturated steam to 83.S'’C superheat,
17.6-28.1 kg/cm^ ; saturated steam to 400°C total.
28.1- 66*2 kg/om> ; 400*^0 to 440®C.
66.2- 70.3 kg/cm* ; A2TC to 482‘»C.
70.8- 140.6 kg/cm* ; 449®C to 506*C.

Conditions recommended for large 3600-rpm turbogenerators in the AIEE-

ASME ^‘Preferred Standards” are:

Turbogenerator Throttle Pressure Throttle Temperc

Rating {kw) (kglcm^ ga) CC)
11,600 42.2 440
15,000 59.8 482
20,000 59.8 482
30,000 59.8 482
40,000 69.8 or 87.9 9 482 or 510
60,000 .69.8 or 87.9 9 482 or 610
90,000 J01.9 .538
264 ENERGY FLOW IN THE STEAM POWER PLANT
The use of high temperatures, while desirable from the point of energy utili¬
sation, poses many problems of mechanical design arising out of thermal
expansion, change of structural properties of metals, and thermal growth.
The maximum steam temperature has been increased through the years
until it is now at 538 to 566^0 in some of the latest central stations. But
designers are finding that increasing difficulties and costs bar the way to much
higher temperatures where the steam pressure is also high. The chief limitation
is the strength of metals at these temperatures. Expensive alloys are introduced
and these often have secondary undesirable characteristics. It should be re¬
membered that a 538°C temperature corresponds to the beginning of incan¬
descence. Materials at 566°G have a dull red glow.
Creep. The characteristics of metals at ordinary temperatures are well un¬
derstood: proportional limits, yield points, ultimate strength, all are familiar
terms. At higher temperatures, before the yield point is reached, there comes a
creep point above which additional stress causes a slow flow, or ^'creep,** of the
metal. Some alloy steels have less creep than plain carbon steel but are more
expensive. The tensile strength of steels drops rapidly for temperatures above
400°, and most metals at 538° are subject to creep. The creep limit has ar¬
bitrarily been set as that at which the stresses (with factor of safety of at least
three on the yield point) produce a creep per hr of times the actual length.
This limit is too high for some portions of the plant, such as bolts for flanges.
Baumann gives the following recommended creep limits. Flange bolts, lO””®;
turbine cylinder, 10”®; parts depending on shrink fits, 10”®; steam piping,
lO”*^; superheater tubes, 10“® to 10“®. The theory of high-temperature design
seems to be shaping itself into an acceptance of creep but with a proportioning
of parts to limit the creep to certain predetermined rates.
Expansion, Problems introduced by high temperatures are primarily those
of strength of materials but there are secondary problems of only slightly less
importance. The expansion and contraction problem is emphasized by the
higher temperatures. Expansion develops problems in the turbine, steam piping,
heaters, and boiler parts. All high-temperature parts must freely expand and
contract. The supports of high-temperature steam turbines must be specially
designed to permit longitudinal expansion of the cylinder and still maintain
the bore concentric with the spindle which is supported independently of the
cylinder. The provision for suitable means to care for pipe-line expansion be¬
comes very important. For instance, at 538° the expansion of chrome-nickel
steel becomes about 1% of the length. Other problems arise through the tend¬
ency of metals to stick together at high temperatures. These problems must
be met through the use of special alloys for the contacting parts. Chemical
stability is another problem sii\ce it is known that chemical affinity is more
pronounced at higher temperatures.
9-5 Heat Flow. A power plant, being in effect a factory for the processing
of low energy into high grade, consists mainly of equipment for energy flow and
transformation. A large part of this equipment is associated with the flow of
heat energy in the vapor cycle and in the transfer of heat between different
fluids. Hence the technical aspects of heat transfer are of importance to the
power plant engineer, especially the designer. The basic modes of heat transfer
are conduction, convection, and radiation.
HEAT FLOW 266
In conduction the heat must diffuse through solid materials or through
stagnant fluids, whereas in convection the heat is carried from one point to
another by actual movement of the convecting substance. This is commonly
carried out by flow of fluids, for the heat transfer may then be a continuous
process. The nature of radiant heat energy was noted in Chapter 1.
Heat transfer is a matter of great importance to industry, as well as power
generation. Its physical constants, governing laws, and use in various and sun¬
dry processes have received thorough study from scientists and engineers. The
following notes on heat transfer for the power plant engineer are but a small
facet of heat transfer theory, so completely has this subject been analyzed and
reported.
All three forms of transfer are found in power plant engineering, usually
in combination with each other. The principal heat-transfer calculations are
associated with;
1. Radiation from fuel bed and luminous gases to boiler tubes and water
walls.
2. Convection of heat from the combustion region to the more remote boiler
tubes, economizer tubes, and air-preheater surfaces.
3. Conduction of heat through boiler, economizer, and preheater surfaces.
4. Conduction of heat through condenser or heater tubes.
5. Conduction and radiation in so-called ^^heat insulators” such as refrac¬
tories and pipe coverings.
The first three of these are met in boiler design. Energy quanta are emitted at
a high thermal potential by the combining molecules of fuel and oxygen, then
partially absorbed as direct radiation by various water-cooled surfaces bound¬
ing the furnace. The portion unabsorbed by the envelope is conveyed to the
gas, via absorption by CO2, H2O, and thermal diffusion, raising its temperature
to that with which it enters the convection zone. During its passage through
the boiler and its auxiliaries, the gas gradually loses its thermal potential until
it reaches the stack potential. A characteristic of modern boilers is that a large
portion of this heat is given to surfaces which “see” the incandescent region.
The thermal potentials of the absorbing surfaces vary; for instance, that of the
boiler is all at the saturation temperature, that of the superheater is variable
from saturation to superheat temperature, while those of the economizer and
air preheater offer further instance of variable temperatures on both sides of
the conducting surfaces.
Simple Conduction. The transfer of heat by conduction through a flat homo¬
genous wall can be determined from Fourier’s Law;

q = kAS/dkcal per hr (9-1)

where A « Wall area normal to heat flow, m* •
k » Coefficient of conductivity of the wall material, kcal per hr-m-
deg C.
$ «> Temperature difference between the two outside faces of the wall,
degC. ^
d Wall thickness, m.
266 ENERGY FLOW IN THE STEAM POWER PLANT
This equation may be modified to cover eases of multiple layer flat walls and
curved walls such as are presented by pipe insulation in both single and multi¬
ple layer. The coefficient of conductivity is not a true constant, being affected
by the temperature of the material. This effect is not a pronounced one and may
often be neglected if the range of temper¬
atures is not large. A table (9-1) imply¬
ing that the conductivity of the materials
tt. listed is a constant, is given. But where
fr
Heat the temperature effect should be consid¬
Fluid
neoeiver ered, the author furnishes charts of the
medium
variation of conductivity with mean tem¬
perature. The material of furnace walls is
Fig. 9-4 Simple conduction and con¬ to be treated in this way. If the dimensions
vection. of length were taken as themetreor both A
and dy then conductivity would have the
dimension kcal per hr-m-deg C.

Simple Convection, This implies a carrying medium of heat capacity wc,

where w is the rate of flow and c the specific heat. Then if the flow is continuous,
through a heat receiver, with entering temperature ti, discharge temperature t2y
the heat delivered is:
q = wc{ti ~ fe) kcal per hr. (9-2)
where w = kg per hr.
c = kcal per kg-deg C.
tiy U = Fluid temperatures, degC.
The convection just described is an over-all conception of heat delivery.
Nothing is implied as to the mechanism of loading or unloading the heat. One
of the most frequently used modes of transfer in or out of the fluid is surface
convection, whereby heat is transferred between moving fluid and a stationary
surface. One finds this in various heat exchangers where heat energy is passed
from one fluid to another through a separating partition. Usually the surface
is tubular, one fluid flowing inside, the other outside. Such a case involves sur¬
face convection on both sides of the partition, and conduction through it; there¬
fore, surface convection between two fluids is actually conduction-convection.
This’important case of heat transfer is explained in Sec 9-7,
9-6 Conduction. Fourier’s Law given in the last section may also be
written
q « UA$ kcal per hr (9-3)

in which 17 is a coefficient of conductance, having dimensions of kcal per hr-m*-

deg C. A comparison of Eqs 1 and 3 shows that U is simply Jc divided by d.
The reciprocal of conductance is resistance. It is the resistance that is addi¬
tive when the wall through which heat is conducted consists of several layers
of differing materials. In such a wall, assuming no contact resistance between
faces of adjacent layers, the over-all conductance is related to the conductances
of individual layers as follows:
CONDUCTION 267
1
±+± + ± + ...1. (9A)
V Uy^ Ut^ Ut Un

Also
^ ^ ^ , ,,, (9-6)
U ki fe k* kn
For flat walls the area Ai * Aj = • • • An", hence
_Ae
3 = kcal per hr (9-6)

ky^ h k» kn

In Eq 9-6, 0 is the difference between the average temperatures on the two

outside faces of the wall.
Now if the wall is curved, as in a pipe or tube. Ay yi A2 ^ An. Using
Fourier's Law in differential form, the following equations, of conduction
through cylindrical walls are secured.

q = kcal per hr (9-7)

In Ti/ry

where ri, r* = Inner and outer radii of cylindrical surface.

L = Axial length of cylinder, m.

The case of double-layer cylindrical walls having radii of ri, r2, r#, is covered
by the equation:
2ir$L
3 kcal per hr (9-6)
In rt/ry In nM
ky k»

Table 9-1. CXDNDUCTIVITIES OF SOME COMMON MATERIALS

kcal per hr-xn-deg C
(at moderate temperatures)
Air (atmospheric). . .. 0*022
Aluminum. 182.3
Concrete. 0.89
Copper (pure).. !322.4
Copper alloys (condenser tubes). 03.7
Cupro-nickel. 29.8
Glass (sheet). 0.82
Glass wool. U.037
Hydrogen. o.i74
Iron (cast). 40.9
Iron (wrought). rt2,l
Magnesia (powdered). . u.Oo
Nitrogen and Oxygen. o.o22
Steel. 39.7
Water-;. 0.53
268 ENERGY FLOW IN THE STEAM POWER PLANT
9-7 Surface Convection. When a solid partition is conveying heat from
one fluid to another, as in Fig. 9-5, a thin layer of fluid remains stagnant against
each face and becomes, in effect, another conducting layer outside ‘of which
the free stream fluid temperature exists. Although
this film is extremely thin, its resistance to heat
flow may be high relative to that of the solid
partition. This is especially true of heat ex¬
changers where the wall is metal of good con¬
ductance. Although film conductance is difficult
to isolate or to predict (being under the control
of several variables such as fluid velocity, vis¬
cosity, density), its importance to heat transfer
is considerable. A basic method of calculating
this type of heat transfer is to treat the films
as conducting layers and determine an over-all
coefficient of conductance, C/.
1
U (9-9)
Fig. 9-5 Section of conduct¬ L+^+l
ing partition bathed by warm hi k. h.>
and cool fluids. hi and are film coefficients, d/k is the resist¬
ance of the solid wall.* This is then used with
Eq 9-3 to calculate heat transfer. There may be an uncertainty whether to
use the outside or inside of cylindrical surfaces for the transfer area A. This
may be resolved by adhering to the usual convention that the transfer surface
is taken to be that which is in contact with the warmer fluid.
This approach to surface convection appears to be simpler than it actually
is, for the estimation of film conductance, h, is difficult and results are not likely
to be closely confirmed by actual operation.
The thickness of the fluid film and, consequently, the rate at which it trans¬
fers heat are affected by the velocity of the fluid stream across the surface. If
this velocity is created solely by density differentials in the fluid due to heat
transfer, its magnitude is usually small and somewhat indeterminate. An ex¬
ample is the rise of heated air around a room-heating radiator. This is called
free convection, in distinction to forced convection, where the fluid velocity is
created by pumps, fans, or other positive means, and whose magnitude is con¬
trollable. Heat transfer calculations are simpler and more definite for forced
convection. It also happens to be the type in principal use where heat is trans¬
ferred to or from fluids in steam power plant equipment. Furthermore, as be¬
tween convection velocities that would represent turbulent or viscous stream
flow, power plant equipment almost always functions in the turbulent flow
range.
A rational equation, derivable by dimensional analysis, which often appears
in the literature of forced convection heat transfer is
Nu = CRe<^ Pr^ (9-10)
♦ Flat walls implied, but also satisfactory for thin-walled tubing of relatively high con¬
ductivity. Otherwise use rlnrs/nnift for d, where r is the radius of the chosen transfer
surface, and correct the h’s to the r surface.
SURFACE CONVECTION 269
Nu, Re, and Pr are dimensionless parameters representing certain natural
groupings of physical factors having an effect on film conductance.
Nu is the Nueaelt number, which contains the film coefficient of heat trans¬
fer, h.
Re is the Reynolds number, a measure of the ratio of inertia to viscous
forces in the fluid.
Pr is the Prandtl number. It is a property of the fluid and not a function of
the physical conditions of flow.*
hDf, „ DeV p
Nu —, Re =-; Pr
/C u A;/3600
where C^x^y = Constants which depend on the arrangement of the heating
surface and the direction of fluid flow relative to it.
c = Fluid specific heat, kcal per hr-m‘^-dcg C x 9.81.
De = A characteristic dimension of the fluid conduit, m,
h = Film conductance, kcal per hr-m-deg C.
k = Fluid conductivity, kcal per hr-m^-deg C*
V = Fluid stream velocity, m/sec.
p = Fluid density, mass density, kg-sec*/ni**
p = Fluid viscosity,! kg-sec/m*.
While Eq 9-10 is rationally derived, it must be implemented by experi¬
mentally determined constants before being usable. Constants for some ar¬
rangements of surface are given in Table 9-2.J What the characteristic dimen¬
sion De is and at what temperature the fluid physical properties are to be
determined depends on the policy of the author of the C, x, y, constants as he
interpreted his test data. These policies have not been uniform in the different
studies. De may be an internal or external tube diameter, a surface dimension,
or an equivalent diameter found as follows:

2) ^ /free cross-sectional area of flow

\ perimeter of this area ) (9-11)

The temperature for p, fc, c, and p may be the average .mixed temperature of
the fluid streams or the average fluid film temperature. It is necessary to
attach these conditions to tabulation of the constants.
The physical properties p, k, c, and p are dependent on fluid temperature,
posing the question of what temperature is to be used for this evaluation. Since
the transfer process occurs through the fluid film, the mean temperature of this
film should be used. However, this is somewhat uncertain and difficult to
measure. In turbulent flow the film is very thin, despite its high resistance.

* Tables of Pr vs pressure-temperature for a number of fluids have been published.

tThis is a form of d3mamic viscosity. To convert oent^ises into kg-sec/
multiply by 10.2 x
t Also in Trans, A8ME, Vol. 59 (October).
 270 ENERGY FLOW IN THE STEAM POWER PLANT
Table 0-2. CONSTANTS FOR USE IN EQ 0-10
(For Turbulent Flow and Clean Surfaces)

Arrangement of Temperature for

heating surface X y D, fluid properties

Longitudinal flow in tubes

and ducts 0.023 0.8 0.4 Eq 9-11 (tt -j- do)/2

Cross flow over tube banks 0.33 0.6 0.33 D* Average film
temperature

('ross flow over single tube 0.3 0.57 0 Do* Average film
1
temi)erature

*/>o = Outside diameter of tube, m*

Thi'rofore it is usual to consider the controlling temperature as that of the main

fluid stream, and if this is a variable the mean temperature is employed.
iig. 9-6 gives physical data required to calculate the heat transfer param¬
eters for air and steam. Calculations where products of combustion arc the
fluids may be made with the properties of air substituted for flue gas. Data for
O2, CO2, N2 might have been placed on Fig. 9-6, but were not because they

Fig. 9-6 Thermal properties of air, water, and steam.

would be relatively valueless in any calculation of /a and Id of a composite gas.

These properties of a mixture may not be calculated by the simple arithmetic
of mixing (as suffices for t, c, R, etc.) and actual tests on a specific mixture
would be required in order to determine its fi, and k. An example of calculations
for surface convection heat transfer will be given later, but at this point a dis¬
cussion of temperatures in heat transfer is needed. We have used the symbol B
for temperature difference effective in heat transfer. What determines $ when
the fluid temperatures are variable?
MEAN TEMPERATURE DIFFERENCE 271
9-8 Mean Temperature Difference. There are some few cases where 6 is
simply the difference between two uniform temperatures. Examples are a section
of a building wall or a heat exchanger with condensing vapor on one side and
boiling liquid on the other side of the heating surface. Far more numerous are
the examples of heat transfer where one or both fluids undergo a change of
temperature while receiving or discharging heat. A few examples:
Boiler. Gas temperature decreases while water temperature remains approxi¬
mately constant.*
Steam Condenser. Condensing steam temperature is constant everywhere at
the condenser pressure saturation temperature, while condensing water tempera¬
ture rises.
Steam Superheater. Steam temperature is increased and gas temperature
decreased.
Oil Cooler. Oil temperature decreased; water temperature increased.
Referring to Fig. 9-7 for nomenclature and physical arrangement of a heat¬
ing surface (total area A), the area is-seen surrounded by a constant tempera¬
ture medium at t\ The flow through the tube has a temperature designated t,
and it is assumed that ^ so that heat is ab¬
sorbed, raising t from ti to to. The mean value of
^ (= — 0 will be investigated for this case.
One rather obvious procedure is to subtract the
average t temperature from t\ This arithmetical
0 is in some error, for it assumes a uniform True t riee
^Assumption if
temperature rise inside the tube, whereas the onthmeficqllv overooed
rate of temperature rise is higher at the inlet Surface troversed

end and becomes smaller as 6^ decreases. For Fig. 9-7 Study of mean tem¬
very small temperature rises, to — ti, the error perature difference.
is not large, but becomes increasingly serious
with increasing At. The shape of the temperature rise curve is readily deter¬
mined. The arithmetical mean temperature difference is the average ordinate 6.
Through elementary area dA there is a heat flow dq that is also the heat
absorbed by the flow w, experiencing a temperature rise dt.
Now dq = UOxdA, where U is an over-all coefficient of conductance, con¬
sidered constant over the whole heating surface. Also, dq = wcdt, = wcdOx, for
dt = d$x if f is constant. As a result,

UOxdA = wcd$x (I)

As has been stated, the true mean temperature difference over the entire
area A is the average 6x- Therefore

9 -

might seem that water temperature would rise from feed temperature to saturation
temperature, and, of course, this is t ue of any specific mass of water. But on account of
recirculation (10-15 times around the circuit for complete evaporation) the incoming feed
is mixed with so much saturated water that the mixture temperature at the tube surface is
nearly that of saturation.
272 ENERGY FLOW IN THE STEAM POWER PLANT
If expression (I) is used to replace both A and dA in (II), we get the follow¬
ing result.
^max *tnin
d = (9-12)
In ^max/^iT

This is usually denoted the ^'logarithmic** mean temperature difference on ac¬

count of the natural logarithm it contains. It can also be written
U- ti
(9-13)
t' - tj
In
t* - to

If the t* fluid were inside and the t fluid were flowing over the outside of the
tube, the same equations would have resulted.
Both t* and t variable. Although a derivation for 0 may follow the same
principle as above, it becomes con
ti tn "to_ sidcrably more complicated, for in
this case one is not entitled to use
the dOx = dt relation. The mean tem¬
perature difference now involves four
temperatures. There are two possible
Parallel flow Counterflow types of flow as illustrated in Fig.
FiCJ. 9-8 Both fluids of varying tempera¬ 9-8, the parallel and the counter-
ture. flow. Counterflow is used where pos¬
sible because it places the highest t
and V together, permitting a larger range of temperature change. In either
case, however, the average temperature is given by a solution of ^BxdA
which, as it turns out, is identical with Eq 9-12, but not with Eq 9-13.
Example 1: The mean temperature difference existing for the convective heat trans¬
fer in a boiler tube bank will be calculated assuming that the case is approximated by
the situation shown in Fig. 9-7 inverted, that is, with t' inside the tubes and t outside.
Boiler pressure, 10.55 kg/om* ga; gas temperature, 982,2*C initially, final 287.8*’C.
At 10.55 4-1.03 kg/om2 ab- the saturation temperature is 185.6*0. This is
The gas temperature t varies from 982.2*0 to 287.8*0.

On 982.2 - 185.6 = 796.6 ; = 287.8 - 185.6 « 102.2*0

796.6 - 102.2
e = = 339*0.
In (796'6/102.2)

Example 2: Steam flowing through superheater tubes and being superheated with
heat transferred from gases flowing over the superheater tube bank is an approximation
of the counterflow case of Fig. 9-8. The mean temperature difference will be calculated
for the following data : Boiler steam pressure, 17.58 kg/om2 ga. Final steam temperature,
315.6*0. Gas temperature before superheater, 81.5.6*0 ; after superheater, 657.2*0.
At 17.58 kg/cm3 ga, saturated steam temperature is 207.8*0. The temperatures are
identified thus : = 815.6*0. « ^ = 657.2*0, « 207.8*0, «. 315.6*0.

BmMx « 815.6 -315.6 = 500*0 : 657.2 —207.8 » 449.4

^ 500 - 449.4
, / 500 \
» 476*0
CONVECTION HEAT TRANSFER, PRODUCTS OF COMBUSTION 273
It will be noted that in the case of counterflow it is possible to have a situa¬
tion where it is difficult to evaluate 0 because is nearly equal to Should
the two be equal, B would be indeterminate. However, for such cases the arith¬
metical mean is practically the same as the true mean and can be used as a
substitute. In Ex 2 the arithmetical temperature difference is 475®.*
There are cases which resemble, from an over-all viewpoint, the flows illus¬
trated in Fig. 9-8, but which are quite different for a specific increment of the
heating surface. For example, the economizer shown in Fig. 10-26, in general, is
counterflow, that is, the water progresses oppositely to the gas, but actually is
flowing transverse to file gases considered tube by tube. The operational factors
(soot, temperature, etc.) may be sufficiently variable not to warrant finely
drawn calculations of mean temperature difference. This is usually true of cases
of heat transfer involving products of combustion, but correction factors for the
logarithmic 6 are in order for shell-and-tube heat exchangers where heat trans¬
fer conditions are fairly uniform and controllable. Charts of these factors are
published by the Tubular Exchanger Manufacturers^ Association.
9-9 Convection Heat Transfer from Products of Combustion. It has
been found that the coefficient of heat transfer in boilers, economizers, and air
preheaters follows approximately a straight line with intercept relation to load.
This is a pure empiricism, based on plots of test results. These empirical rela¬
tions are of the general form:
U = A -h Bf? (9-14)
where A and B are experimentally determined constants and G is mass flow.
The mass flow is usually given as kg of fluid per hr per m* free area
normal to the flow. It brings in the velocity factor, but is more readily de¬
termined than velocity; also for gas flow through a tube bank the mass flow
may be constant throughout (if tube spacing is), whereas velocity changes from
point to point due to shrinkage cooling.
The constants A and B differ for the various classes of heat transfer. The
Babcock formula for the U of boiler heating surface has A*»9.76^B = 0.(X)14.t
These were determined for boilers having mainly convection heat transfer (no
water walls). Although not applicable to modem steam generators as a whole,
this coefficient may be employed to analyze heat transfer in the purely convec¬
tion zone.
Consider a case of convection between fluids (a) and (b) for the purpose of
relating required heating surface to the desired heating job. It is a counterflow
arrangement, having coefficient of U, Let the temperature of the heat releasing
fluid (a) be i', of heat absorbing fluid (b) be t. We will introduce a term z, de¬
fined as
zt - (9-16)

CbtOb
or Z “ ■ (9-16)

in which w is rate of flow and c the specific heat.

♦But note that in Ex. 1 it would have been H <962.2+287*8)-185.6 or 4 4•«4^ an error of
some ll(ra.
t Valid only for tranafer of sensible heat,
@ in koal per m>-hr-deg C
t in koal/kg«deg O
274 ENERGY FLOW IN THE STEAM POWER PLANT
Since q = VAB = one will find that, upon substitution of
Eqs 9-12, 15,16 into this expression, the heating surface is found to be

CtflVu Bnmx
A* (9^17)
U{Z ^ 1) ^„ln

Heat transfer from products of combustion of coal fires to various heat

transfer surfaces such as boiler, superheater, economizer, and air preheater may
not justify exact rational calculations because of the continually varying oper¬
ating conditions and their influence on heat transfer. These are mainly:
1. Varying accumulation of scale on the water side.
2. Varying accumulation of soot on the gas side.
3. Varying mass flows, G, at different boiler loads and at different percent¬
ages of excess air supplied.

However, unless the empirical data available are based on cases very similar
to the study being undertaken, the designer must revert to the rational methods
obtained from Thermodynamics.
9-10 Convective Heat Transfer Between Condensing Steam and
Water. This type of heat transfer has received full study in theory and experi¬
ment. The factors that affect the transfer have been individually studied and
evaluated; consequently predictions of heat transfer are on a fairly definite
basis in the steam condenser and water heater field. These factors have been
examined in investigations of vapor film resistance, effect of water velocity in
tubes, etc. We shall now devise an equation to assemble the results of such
studies in a relation between heating surface and the various influences on heat
transfer.
The customary arrangement of equipment for transferring heat from a
condensing vapor to water through a dividing surface is to arrange that surface
in tubular form with the water inside the tubes and the vapor outside. This
gives controlled water flow and is much more effec¬
tive in promoting high rates of transfer than is the
submerged steam coil arrangement. It furthermore
facilitates prompt and direct delivery of condensate
to the hotwell. This yields the characteristic ar¬
rangement of a bank of tubes expanded or packed
into tube sheets, supplied with water from a water
box which is essentially a water supply header for
Fig. 9-9 Cross section of the tubes. Encompassing the whole is the shell
condenser tube. which confines the vapor to the tubes. When prop¬
erly supplied with flanges for ingress of steam and
water and egress of condensate and water, we have the basis of both the
condenser and the heater.
Basically we have the case shown in Pig. 9-7. The tube is carrying water in
turbulent flow and is surrounded by condensing steam. The conditions attend¬
ing heat transfer where impure water has to be used are depicted on the tube
cross section. Fig. 9-9.

♦ Valid only for transfer of %entihle heat.

TRANSFER BETWEEN CONDENSING STEAM AND WATER 276
First the heat must be delivered from the vapor to the outside tube surface.
This is affected by the film of condensate flowing down the surface, the quantity
of noncimdensable gas that might be in the film, and sludge or scale on the tube
surface. If the tube is one of a large number contained in a nest or bank, the
effect of these variables is different for different tube locations and an average
for the entire bank of tubes is wanted more than expetimental results reported
for an isolated tube.
Next, the heat is transferred across the metal of the tube wall by conduc¬
tion. This is by far the simplest and most exact to compute of all the component
temperature drops. Also, it consumes but little of the total temperature drop,
even when steel tubes are used.
Althou^ a water heater will have little or no deposits on the tube interiors,
condenser tubes are subject to fouling from a number of sources such as silt
or sand in the water, organic waste, algae, slime, and scale. This interior ac¬
cumulation forms a coating whose heat conductivity is relatively poor. In some
extreme instances, the principal part of the temperature difference may be
consumed in driving heat through this layer. The magnitude may vary from
very small, e.g., mill scale on an otherwise clean tube, to great, e.g., a condenser
tube badly in need of cleaning. As in the tube itself, the action is one of con¬
duction, with the cmductivity of the scale being considered instead of that of
the metal.
Some further temperature difference is required to convey the heat from
the inside of the layer of scale or slime to the water. This (mnponent has been
investigated fully in the laboratory and workable formulae have been evolved.
It is by no means a minor part of the resistance to heat transfer and is affected
by velocity of the water, viscosity (or temperature) of the water, and diameter
of the tube. The tube diameter enters by virtue of its being a measure of the
thickness of the film of water next to the tube, relative to the dimensions of the
main body of flow.
Water enters at an inlet temperature fi, and leaves at an outlet temperature
of which, of course, is higher than U. lie condensing vapor surrounding the
tubm is at a saturation temperature of t'.
The over-all mean temperature difference $ is the sum of several resistances,
each being manifested as an incremental temperature drop. These are:
the drop from vapor to tube surface.
fit, the drop across the tube wall.
fi„ the drop across the layer of scale or sediment.
fiw, the drop from scale to water film.
These have the relationship fi = fir + fit + fit + fiw But, since q — UAfi, or fi =
q/UA
fi - (9-18)

As Fig. 9-9 shows, q is (xnnmon to all these temperature drops, since they
are in aeries; that is, the same amount of heat that enters a givoi segmoit of
wall must emerge from the same segment and heat the water. q/A would be
common except that the area of the izmer surface is less than the outer; there¬
fore the intenrily of heat flow is greatw.
276 ENERGY FLOW IN THE STEAM POWER PLANT
By taking Ay as on the outside of the tube, A. and A« at the inside, and At
as the average, all surfaces may be expressed in terms of the outer surface At.
Calling Z)w the inside diameter of the tube and Dy the outside diameter. A*
and At may both be expressed as a function of At and the diameters. Then
Eq 9>18 is reducible to

$ = H-^-h (9-19)
At LC/v ^ UtiDyy + Dy) ^ UJ)y, ^ Uy,D„j
This apparently unwieldy formula is, nevertheless, in useful form, and mere
length is not necessarily a handicap. Some of our shortest formulae are most
difficult to employ because of their inflexibility. The form of Eq 9>19 is useful
to the degree that research and tests supply the values of the separate It
is in a form such that, as new experimental data become available, they may
be used to supersede old data. Also, experimental data may be applied directly
instead of being converted into abstract coefficients. For instance, Ub would be
the actual conductance of a layer of scale in kcal per m* per hr per deg C tem¬
perature difference. It would be difficult to convert such information, were it
available, into an abstract multiplier such as a “cleanliness coefficient.”
A. T. Brown, Sherwood, McAdams, Turner, Partridge, White, and others
have contributed information on the various conductances. Their work is not
necessarily final and probably will be repeated and improved on by others as
time goes on.
Uti Conductance of the tube is fct/dt where dt is wall thickness. Consult
Table 9-1 for fct.
C/g! The amount of scale on the water side of condenser tubes varies with the
source of the water, season of the year, and frequency of cleaning. The coeffi¬
cient is therefore widely variable with a possible range of 4880-19530 kcal per
m*-hr-dcg C. An average value to use lacking definite local information would
bel4650.The mill scale on new tubes having no fouling has U of about 29250.
Uw* It has been established that this is some function of water velocity.
Sherwood and Turner give:
f/w = (9-20)

in which b = 1379.7 (9-21)

V = Water velocity through tube,m/scc.

J Water fluidity in reciprocal centipoises taken at average water
temperature.
Dw = Inside tube diameter, m.
Uri A study of empirical data* showed that the vapor film resistance vs.
heat loading characteristic is well matched by the equation
Uy =111910/(g/Av)®'^®® (9-22)
In most design problems the heat to be transferred and the mean tempera-
♦ A. T. Brown, Combustion, October 1030.
TRANSFER BETWEEN CONDENSING STEAM AND WATER 277
ture difference available will be known in advance. By substituting q and $ in
Eq 9-19, we find the required surface Ay. Another application of this equation
would be the determination of the B required to produce a heat fiow of g/A. A
further usefulness of it is the solution for water velocity necessary to produce
some desired q/A with a given 6.
The logarithmic mean temperature difference is to be employed for B, This,
of course, is somewhat of an approximation, since the temperature of the con¬
densing steam throughout the condenser is not constant on account of pressure
drops caused by tube resistance.
Eq 9-22 is for relatively gas-free condensate film. Orrok’s emi)irical equation
for over-all coefficient of conductance has a term p, designed to allow for con¬
siderable air entrainment.
U = 3024.8 (9-23)
where C = Cleanliness coefficient, range 0.5 to 1.0.
K = Material coefficient, about 0.97 for copper alloy.
V = Water velocity, fps.
p = Ratio of vapor pressure to the total pressure of the air-vapor
mixture.
Realizing the necessity of establishing heat transfer rates which are possible
of attainment in average commercial operation with efficient condensers, manu-

T*mp*rature of inlot water-*C

Fro. 9-10 Steam tuibine coadenser heat transfer rates. (For various tubes and'water
velocities.) (Courteqr Stdt. Heat Exchange Institute—mt.)
278 ENERGY FLOW IN THE STEAM POWER PLANT
facturer members of the Heat Exchange Institute have conducted tests and
arrived at values of over-all conductance which they consider to be maximum
safe design limits. The coefficients used by this association are shown by Fig.
9-10. The coefficient U is in kcal per hr-m*-dcg C, and is such that
Ue = q/Ay (9-24)
The data are depicted for21.rinlet water temperature, and are based on
clean, new 1.24 mm(18 BWG) Admiralty tubing. Other operating conditions
may be corrected for with the use of temperature and tubing multipliers. One
could judge from the purpose for which these coefficients are promulgated that
condensers in service could produce higher values of the coiAcient.
Very similar to the surface condenser is the steam water heater, referred to
in Chapter 8 as a regenerative or extraction heater. The heat transfer analysis
is practically the same. Whereas basically both accomplish the delivery of heat
to a fluid of rising temperature by the action of a condensing vapor delivering
its latent heat of evaporation while being reduced to the state of condensate,
actually they are distinct for the following reasons:
1. The condenser must accommodate practically the entire flow of the
vapor cycle as a low-pressure vapor, whereas the heaters are required to hold
only a small fraction of the flow as a vapor. Furthermore, the vapor density in
the heaters is higher than in the condenser, and thus there is a large difference
in mere physical size of equipment.
2. Heat transfer in a condenser is seriously complicated by formation of
scale and sludge on the transfer surfaces, whereas there is very little of that in
the heaters.
3. Condensers are generally one- or two-pass, while heaters are never less
than two-pass, and may be four- or six-pass.
4. Condenser tubes are copper alloy of high heat conductivity, but high-
pressure heater tubes may need to be of the stronger (but less conducting) steel.
9-11 Examples of Surface Convection Calculations.
Example 1: The film coefficient of water flowing in a 19.1 mm QD^ 1,66 mm tube at
1.67 m/seo will be calculated from Eq 9-10. Water enters this tube at 16.6^5 and leaves at
29.4‘»0.
The fluid properties will be evaluated for a temperature of (15.6 ^ 994^ ^
Conductivity from Table 9-1; viscosity from Table 14-8 or Fig. 9-6. Specific heat,
1.0koal par kg-deg C density, 697.7 kglmK
k »0.533 kcal per ha>m*iieg C,
H s 0.95 centipoises, or 0^95 x 10.1 x lO»-»kg
c » 1.0 X 9.81 * 9.81,
p « 997.7/9.81 » 101.7 kgWl/n^.
D, » (19.1 - 2 X 1.6»>/iOOOOJOMm.
These properties were computed in order to find the parameters Nu, Re, Pr, and thus
to use Eq 9-10.
Nu » 0.0158^/0.533 » O.OMk.
Re « 0.0158 X L67 x tOlJKkM X iOJ x 10^ - tWO.
Pr ^ 9.81 X 3600 X 0.95 X 10.2 x lO-s/0.583 « 6.47.
EXAMPLES OP SURFACE CONVECTION CALCULATIONB m
Substitute in £q 9>-10 with eonstanti Mieeted from Ti|4ble S-2.
O.OtMh - Mn X fftOOM X 6;47«4
h ^ 0906 Inel per hi^mWeg C

In £q 6*20, is also the film eoefficient. It will be computed and compared with the
foregoing result.
(l/D.60)e*M
From Eq 6-21, h * 1279.7 i’a5"‘'CT *■
1009
4290.4 X 1.67 01* » 6200 keel per iMbiaMegO
Example 2: A convection steam superheater is an example of e^ple heat transfer
between a gas and a vapor. Here a problem is devised which will involve an estimate
of the over-all coefficient. The surface arrangement is diagrammed in Fig. 6-11, and
some of the data required for a solution are to be found on the figure. The surbee
ares needed and tube bank arrangement will be determined.

STEAM
HEADERS

As the tube is thin walled and of very high conductivity compared to the steam
and gas films, the curvature will be neglected in the equation for over-all conductance.

i- = — -f- -i- 4- -
U hx Ut hz
where is steam film conductance; h2 the gas film conductance.
Steam Film Coefficient. Eq 9-10: Nu = O.O23/?e0-
This is capable of simplification, for Pr i^ approximately unity for superheated
steam. Also the pV of the Reynolds number can be exchang^ for G,/9^S1, where G,
is the steam mass flow, kg per ml-mo. With these modifications, £q 9-10 becomes

hi « 8.7 X 10-»A-^ »ib,(G./Mi)®'*

A mass flow must be assumed. High values promote heat transfer, reduce the number
df superheater elements in parallel, but also increase fluid friction winch may lead
to undesirably high steam pressure drop through the superheater. Not wisfa^g to
introduce friction loss calculations at this point, G» is assumed without further ado
at m.9* kg per mMeo, and the steam properties are assumed to be tbose at 17.6

0Tliis oorresponds to a velodty of about W9 mfmin.

280 ENERGY FLOW IN THE STEAM POWER PLANT
The average steam temperature is 260*0. From Fig. 9-6, i?! = 4 x 10-*, 22 x
10-7, Dy = 20.07/1000 n= 0.02007 m.
*1 == 3.7 X 10-3 X 0.02007-0.*x4xl0-*x[390.6/22X 10'7]0 * == 1283 koal per hr-
m*-deg C. Transfer this to the 25.4 mm diameter surface with the multiplier 20.07/ 26.4,
makmg hy ^ 1014.
Gas Film Coefficient. Eq 9-10: Nu = 0.33Ee^
The gas properties arc to bo those of the average film temperature, and the
characteristic length the outside tii))e diameter. The gas temperature is found with
use of a 0.25kcalperkg-degCspecific heat. (enthalpy gained by steam).

Since = 2wn;2 x 0.25 (815.6-t'o ) == (3064.4t - 2797.5t)0.239 = 63.8.

t'o = 688*0 . Average t' == 751.8°C.
44 6
The free area for gas flow, per element =2.44 x = 0.1088 m* .

Steam flow per element 390.6 x 0.000316 ^ 0.01234 kg per seo.

Gas flow per element == 2 (steam flow) := 0.02468 kg per sec.
Mass flow of gas .= 0.2468/0.1088 - 2.268 kg per m^-sec.

Gas film temperature is the average of tube tem])eraturc and gas stream tempera¬
ture. Preliminary trials having shown that /i, is much larger than hj,
the 751.8° to 260*
drop between fluids is assumed to take place mainly m th(' gas film Tube tem])erature
IS estimated to be 316*C.
Film temperature = (751.8-f 316)/2 ^ 534° C. From Fig. 9-6 at this temperature,
A’2 = 4.6 X = 38.33 X 10“7. Using air properties ns a substitute for those of
combustion products, = 0 884 *
Eq 9-10 is now modified to
h. = 0.33/)r’A-2[7)2Gg/(9.8lM2)?'’Pr2«''’^ =0.0841 W ^3
Substituting the values already determined, with D2 =26.4/1000
Ih = 0.0841 X 4.6 xlO-2 X (0.0254)-0-4 [2.268/38.33 xl0-7p «x0.884
whence A2 =43.19 kcal per hr-ni*-deg C.
Overall Coefficient. The tube coefiScient is approximately k/dt. From Table 9-1 for
steel, k = 39.7 kcal per hr-m-deg 0. Ut = kcal per hr-m*-deg C.
Now 1/U = 1/1014 + 1/14897 + 1/43.19 whence, U = 41.31 kcal per hr-m*-d6g C.
Number and Length of Elements. Assume that this superheater is to be in a 13610 kg
per hr steam generator. Superheater transfers 13610 [(3064.4t—2797.6)0.239] « 0.868 x 10*
koal per hr*
^max = 815.6 ~ 316.6 = 600*0, = 688 ~ 208 = 480*0,
500 - 480
488*0.
^ In (600/480)

A 0.868 X 10*/(41-31x488) = 48.6 m* *

Number of elements = 13610/3600/890.6 x0.000316 « 81.

Area per element = 43.5/81 » 1.4 m* «

Element length «1-4/0-07086 « 17*6 m •

'^Author used Nat. Bureau of Stds. Table 2.44, but Pr might also be calculated from
alt, ka, and ca.
tValues in Joules/gram. Multiply 0.238 to convert to koal/kg.
RADIATION 281
To Booure this in an %M m high paoBage will reqaire 8 paBBOB or 2.2 m per paBB. This
arrangement would probably have a drop of 1.4-2.1 kg/om< steam preBsure between super¬
heater headers, and a draft loss of 7.6 mm water, or less. The major heat flow reostance
is on the gas side on account of the high mass flow of steam and the low mass flow
of gas. Nevertheless, steam pressure is simpler and easier to create than draft.
Example 3: Illustrating the use of Eq 9-19, the rate of' heat transfer through
lO.lmmX 1.24mm. copper alloy clean condenser tubes will be determined.
Given 6 ~ B.1**C , »= 6885 koal per hr-m*- deg C •
Assume the Dy/(UJ)^) term negligible for clean tubes.
<y r 1 / q , 2x19.1 191 "1
Eq 9-19:
Av 1111910 VAv/ ^ 98.7 ^1^.62+19.1) ^ 98^ x 16.62 J
1.24xlO-»
Arrange as follows for solution:

fjL\ 906500 - 20*4

Uv/
This equation may be solved by trial, but less tedious is the graphical solution
given in Fig. 9-12.
Rale of heat transfer, g/A^ =■ 33000 kcai per hr-m* .

9-12 Radiation. The outstanding example of this method of transferring

heat is to be found in the furnace of the steam generator of a power plant. There
the elements combining in combustion liberate energy in the form of intensive
radiations in the infra red and visible light spectra, accompanied by the pro-

Fig. 9-12 Solution of [|-] ^

duction of combustion at high temperatures. Although this offers the most

complex case of radiation in the power plant, there are many other situations
where the discharge of radiant energy, usually in the infra red range, takes
place. Sometimes this is aided and abetted by design (coolers), whereas else¬
where it is undesirable and is opposed by whatever means seem to be practical,
such as covering the hot surface of a boiler casing with aluminum paint of low
emissivity.
282 ENERGY FLOW IN THE STEAM POWER PLANT
Let us consider the high-level radiations in a furnace. The trend in steam
generator design has been^ for several years, to transfer moie of the heat of
combustion by radiation, less by convection. Many years ago the heat was re¬
ceived by the water mainly via convection from heated gases produced in the
furnace. Today large steam generators are often built with convection surface
a minor part of the boiler heating surface. Radiant heating surface accounts for
nearly all the evaporative heat transfer in such units. Even small- and medium¬
sized units are customarily designed with a certain amount of water wall
ranged around the furnace envelope and radiant transfer may account for a
third to a half of the total.
The law of radiation expressed by Eq 1-6 is more useful in determining
the heat radiated from hot surfaces to the surrounding air than it is for radia¬
tions from luminous particles through a cloud of furnace gases. The application
of the Stefan-Boltzmann relation becomes exceedingly complex and difficult to
justify in furnace combustion. However, radiation from pipes, hot walls, etc., is
calculable on the assumption of radiator and receiver of infinite expanse.*
Factors of emissivity are then ascribed to the radiating surface, these really
being for combined wall and ambient atmospheric emissivity. The re-radiating
temperature of the atmosphere should be considered as that outside the layer
next to the wall wherein heat is carried away by convective currents.
Example 1: The emissivity ot bare brickwork is sometimes quoted as 0.93. The
heat radiated from a brick wall at 98‘3"C surface temperature to surrounding air at
a9*4®C will be computed. Eq 1-6 includes 04*874 x x These are converted to
a single coefficient. 4.874x0,93xlO-* =458.8xl0~w.
q = 458.3 X 10'la[(93.3 -f 273)4 _ (29.4 + 273)4] ^ 4Z7,i) kcal per br-m*

We will now consider the radiant heat transmission in boiler furnaces in

some detail. The possible rate of absorption of heat by a square foot of water-
cooled tube surface is so much more if the heat is implanted radiantly than by
convection that boiler designers have undertaken to utilize this mode of transfer
more and more. The heat quantities involved are diagrammed in Fig. 9-13. Fuel

Fio. 9-13 Heat flow in the boiler furnace.

* Not true, of course, but a practical method for many cases.
RADIATION 283
is combined with oxygen in combustion, with a possible liberation of Qi», the
higher heating value. Some of this, however, is unavailable for raising tenipera-
ture, and since radiation depends on temperature, we elect to work with the net
heating value, Ql. If any part of this is unrealized as active heat inside the
furnace, it should be deducted from the net heating value in order to determine
Qi, the sensible heat actually available. This sensible heat can leave the
furnace in one of these ways: (1) Direct net radiation to the boiler's effective
radiant heating surface. (2) Absorption by molecules of furnace gas which are
thereby heated to the jurnace aperture temperature^ i.e., that average tempera¬
ture possessedJ[)y the products of combustion leaving the furnace region.
We will designate the fraction of that is radiantly absorbed as 12. Then,
of course, the sensible heat remaining in the products is Q,(l — Q), based on
the datum of air temperature entering the furnace.
The above is summarized in the equation:

*= Ql — Q' kcal per kg fuel or J per g fuel (9-25)

in which Q, = Available sensible heat kcal per kg fuel or J per g fuel.

Ql = I^ower or net heating value, kcal per kg or J per g.

Q' = Heat losses, principally radiation and convection from setting,

incomplete combustion, and combustible in refuse.

Radiant energy travels in a straight line. Therefore to transmit it to an

object out of sight of the radiator requires a reflector, such as a furnace wall, to
deflect the rays to their objective.
It is possible to set up controlled laboratory radiation between simple plane
surfaces and determine therefrom accurate coefficients to incorporate into radia¬
tion equations. However, the radiation gf heat from furnace gases, consisting
of nonluminous gases, luminous carbon particles in flame, ash globules, etc.,
to the walls and tubes of a steam generator in commercial operation at variable
load, is another matter. Here, empirical data gathered and interpreted from
field tests on similar equipment must still be resorted to however great the de¬
signer's urge to return to basic laws of heat transfer.
Radiant hfeat transfer in furnaces is roughly proportional to the difference
in the fourth power of the absolute temperatures of the radiating and receiving
surfaces. The water wall surface is approximately at boiler saturation tempera¬
ture, whereas the superheater surface varies from this to somewhat above the
temperature of the steam at the superheate** outlet. However, the mean radiating
temperature of the furnace gases is usually over 1200®C . The fourth power of
the reoeiving surface temperature is*thus seen to be small compared to the
fourth power of the transmitting surface temperature; consequently the latter
controls the transmittance, and boiler tube temperature does not need to be con-
indered a variable to be accounted for.
Manufacturers must still rely on methods which are empirical, although
founded on the known laws of radiant heat transmission. There have been some
empirical formulae covering the case of fumaee radiation, of which the most
widely quoted is the Orrok-Hudson formula:
284 ENERGY FLOW IN THE STEAM POWER PLANT
1
O (9-26)
m .VCr
1 + 60
where O = Fraction of the available heat in the furnace which is transmitted
by radiation. The available heat is to be taken as the sensible
heat actually liberated.
Wt, = kg of air used per kg coal.
Ct == kg of coal per m* of equivalent radiant surface per hr. This con¬
templates the use ot t^ usual bituminous steaming coals. For
other fuels, replace VCr/60 with \/^/55 where C/ is the
combustible per m* per hr.
The chief objection to formulae of this nature is that, although they may fit
many cases of heat transfer well, there are other cases where the conditions of
heat transfer are different, and the results computed by formulae become er¬
roneous. For this reason manufacturers of equipment in which radiation is im¬
portant tend to rely mainly on their background of prior knowledge obtained
through field tests and manufacturing experience. These data may be worked
into curves expressing the effect of the variables involved, and the curves inter¬
polated or extrapolated to fit the individual case of each new design. Naturally
such data are mainly of a confidential nature, as they become the individual
firm^s “stock in trade.” Fig. 10-13 is exemplary of this type of information.

ooc#
a he
6^ d
(n rows deep)

Fig. 9-14 Arrangement of radiant heat absorbing surfaces.

The design of pulverized coal furnaces endeavors to provide sufficient receiv¬

ing surface to absorb enough of the sensible heat liberated from the coal to cool
the products of combustion to a temperature at which the ash will not cause
slagging difficulties. Radiation from small globules of ash may causo them to
have a frozen crust wliile preserving a molten center. Thus it is possible for ash
to behave like dry ash even though the exit furnace gas temperature exceeds the
a^li fusion temperature. Ordinarily, conservative design would not rely on
i , but would provide sufficient surface to reduce the gas temperature safely
0^ fusion temperature; however, in large units it is difficult to provide suf¬
ficient water wall surface to freeze the ash completely and the velocity of
furnace gases at the boiler aperture must be kept sufficiently low to minimize
“squashing” of ash globules as they collide with boiler tubes.
^'Ejfective' Surface. Figs. 9-14 and 10-12 show some of the arrangements in
which radiant heat absorbing surface is disposed. These illustrate another of
the difficulties which beset the designer following a rational or semirational form
RADIATION 285
of radiation analysis. Projected radiant surface is one thing; actual radiant
energy receiving surface may be quite a different area. For example, suppose
the tubes of case (a) Figure 9-14 to be separated and spaced Immon centers.
The projected areas would then be the same, but it is obvious that the exposure
of some of the rear circumference of the tube to reradiation from the wall will
increase the effective radiant surface. To quantize this action, let 8 be a factor
correcting projected tube area to equivalent absorbing surface. Then
Ar = (9-27)
where Ar = Equivalent radiant heating surface.
Ap = Heating surface projected on the plane of the furnace envelope.
8 = Slagging factor. 1 for clean tubes, 0.9 for tubes in normally satis¬
factory operating condition.
S = Area factor previously defined. For arrangements in Fig. 9-14:
(a) ir/2
(a) ' with tul)es spaced /j, where /, > d; ir/2, increased up fo .*>()';.
(b) ir/2 “h
(o) between 0.5ir and O.Tw.
(d) 2 up (no general agreement on extent of radianl ])en(‘Trati(>n). This
case is of relatively small concern to large steam generator <iesign
because of the use of thinner tube banks and th(‘ r(‘lativel\ >mall
aperture which minimizes the tube bank area ])enmtt(Ml lo “see”
the furnace.

The nomenclature of furnace radiation just given cannot be said to be in

common use. There has not evolved as yet a widely accepted singU* line of a])-
proach to this problem. For example, the literature contains the following terms
associated with radiant heating surface.
Ap = Projected area in the plane of the water wall tubes.
BS = Boundary surface of the furnace interior in the plane of the water
wall tubes, or roughly the refractory wall area.
ERS = Equivalent radiant surface (Ar).
EPR8 = Effective projected radiant surface. This is the BS nuiltiplie<l by
an ‘‘effectiveness^^ dependent on tube arrangenu'nt.
See Fig. 9-15 for the effectiveness of three tube arrangements.
FC = “Fraction cold.” This is A^/BS,
The object of these various approaches to furnace radiation is the same—
to relate furnace aperture temperature to water wall surface and conditions of
combustion. The empirical Eq 9-26 docs this directly, but could not be con¬
sidered as valid as other more modem approaches that bring c\irrent operating
data to bear on the relation between furnace temperature and heating surface.
Some of these data will be presented in the next chapter in an analysis of fur¬
nace water walls; however, it will be found that modified ladiant heating
surfaces, available heat quantities, etc., as here presented, will also be required.
 ENERGY FLOW IN THE STEAM POWER PLANT

Fig. 9-15 ^'Effectiveness” factors of bare-tiibe water walls. (Courtesy Combustion

Engineering-Superheater, Inc.)

9-13 Apportionment of Heating Surface. The heat absorption in a

steam-generating unit takes place not only in the boiler, including water walls
and superheater, but also in the economizer and air preheater if they are pro¬
vided. A combination of boiler and economizer, or of boiler, economizer, and air
preheater may, under certain conditions, be more economical for the same heat
absorption than a boiler alone. The fraction of the total heating surface to be
assigned to boiler, economizer, or air preheater should be such as to result in
lowest first cost per unit output. An economic study would naturally include
such factors as.coelficients of heat transfer and annual costs of the sq metre incre¬
ments of boiler, economizer, or air heater surface. The temperatures of the flue
gas are indicative of the heat-absorbing capacity. The following equations have
been developed * for finding the gas temperatures leading to the most econom¬
ical apportionment of heating surface.
Let subscripts b, e, and a refer to boiler, economizer, or air heater.

C as Annual charges per sq metre, increment of surface.

U « Heat transfer coefficient for incremental surface kcal per hr-m*-deg C.

C{ « Annual money saving per kcai of heat recovered per hr.

* Linn Helander, “Dividing Heating Surface Among Boiler, Economizer, and Air
Heater,” Power, March 4, 1930.
PROBLEMS 287

(7^ * of ^ l^cal X hr of ^rvice per yr (9-28)

Efficiency of steam-generating unit
f Gas temperatures.
t = Water or air temperatures.
tu » Saturation temperature of the steam.
Other symbols as used previously.
To determine the economical temperature of the outgoing gas from the
economizer or air heater when the inlet gas temperature and the inlet water or
air temperature are fixed; that is: Given i'l, to find t'©-

Ce(a)/(C/e(a)Cr) = (^o ~ UWi ^ O/^'i ^ U) (9-29)

To determine the economical boiler-outlet gas temperature for a boiler-
economizer, or boiler-air heater combination when the temperature of gases out
of the air heater or economizer and the entering air or water temperature' are
fixed; that is: Given i'©, ii, to find t'l.

= (^'i ~ to)/(t'i - Q (9-30)

To determine the economical temperatures out of boiler and economizer for

a boiler-economizer-air heater combination when the temperature of gases out
of the air heater and the inlet air and water temperature are fixed; that is:
Given f'oa, to find ^'le and ^'ia*

CJK ^ ie_^ i ia ^oi

(9-31)
CMn /'•
* i»* — /
to<* /'•
* la — /•
‘n

The foregoing equations, together with 9-15 and 9-16, can be used to set up
a series of simultaneous equations which can then be solved for the values of the
most economical temperatures. The solution of the simultaneous equations will
be simple for the first two cases, but will require more extensive labor for the
last.

PROBLEMS
1. Devise a diagram to show roughly, but quantitatively, the rate of flow of
energy in a power system (Fig. 1-11) per kw being used by a customer. At appropriate
points on the diagram indicate the magnitude of each of the following: (a) generator
output; (b) torque at turbine coupling at 18(X) rpm; (c) waste heat discharged
through condenser; (d) weight of coal (bituminous) used per min. Data: steam
generator thermal efficiency, 75%; auxiliary steam^ 5% of tW generated; thermal
efficiency of vapor cycle, 35%; generator efficiency, 95%; electrical auxiliaries use 2%;
power line efficiency, 93%.
2. A deuteron is an atomic particle composed of a neutron and a proton held
together by nuclear binding energy. On the atomic mass scale the deuteron mass
is 2.0146. Find the binding energy in Mev per deuterwi.
3. Assume that all the energy released in the fission of U*®® in 25 g ot normal
uranium is captured for use. How many kw hr should be obtain^ 25 g of normal
uranium?
288 ENERGY FLOW IN THE STEAM POWER PLANT
4, A new steam power plant of 30,000 kw capacity in one unit is being consideted.
Assuming that the pipe friction loss between boiler drum and turbine would not exceed
1% of the throttle gauge pressure, predict the probable boiler design pressure and
the degree of superheat to be used.
5. Bolts in a high-temperature pipe flange which had begun to leak after five
years* continuous service were removed and the length compared with installation
records. Stressed lengths that were installedat 130*83 mm were found to be 140*46 mm long.
Determine whether this is within the creep limit mentioned in text.
6. A flat partition is made in two layers, X and Y in close
contact (Fig. 9.6P). Thermocouple measurements are taken
at stations as shown. Material Y is known to have k =7.2
4.,5* koal per hi*—^m—des; C Find: (a) (b) contact re¬
sistance in hr-m^-deg C per kcal.
7. A steam pipe 219.08 mm OD is covered with 50.8 mm of
material for which k 3.47, over which is 63 5 mm of
Fio. 9-eP
material with k 2.54, both k'a in kcol per hr- m^-deg C.
Outside surface of pipe at 457“C ; of covering, 79*C. Length-of pipe, 4.72 m. How much
heat is lost through it per hr ?
8. Given Fourier’s Law as qdr = -—kAdO; derive Eq 9-7.
9. A thin, flat metal wall is bathed with water on one side, air on the other. It is
found that 3300 koal are transferred to the water in forced convection per in’* per
hr. Mean temperature difference, 27.8®C. The water film conductance is estimated at
4638 kcal per hr-m^-dog C Make an estimate of the air film conductance.
10. Write an equation for q transmitted by forced convection under
a mtd of $ for the case shown in Fig. 9-lOP.
11. Air at an average stream temperature of 82®C is flowing at
762 m/min in a rectangular duct 38cm x 76 cm. Estimate the conductance
of the air film.
12. Diagram a <*ase of heat transfer from condensing steam to
water flowing in a tube. Derive Eq 9-12 for this case, showing all Fio. 9-iOP
necessary steps.
13. Find the mean temperature difference in a steam condenser where the absolute
pressure is 0.12 kg/cm^and the condensing water terminal temperatures are 18.3*’Cand
3l.l“C.
14. In a two-fluid heat exchanger, operating counterflow, one fluid rises in tempera¬
ture from 82°c to I2rc j w^hile the other changes from 32rc to 218®C Diagram
the flows. Calculate the true and arithmetical mean temperature differences.
15. Repeat Prob. 14, except make the case one of parallel flow.
16. A lubricating oil cooler of the shell-and-tube type has the following flow
temperatures: Oil in, 66°C; out, 33.6®0. Water in, 15.8®C; out, 47.8®C. Arrange¬
ment, counterflow. Find the mean temperature difference.
17. The rate of water discharge from a 5.4 mm copper alloy tube 3.05m long surround¬
ed by steam was 97 kg per min. Water terminal temperatures, 18.3®C and 28.9®C. Steam
temperature, 104.4*0. What coefficient of conductance was developed?
18. With the use of Fig. 9-10, determine the required surface for a condenser if
43,090 kg steam per hr are to be condensed at 50.8 mm Hg aba, uaing water at 21.1*0
which will be heated to 4.4* less than steam temperature. Steam quality, 0.875 ; water
velocity, 2.44 m/sec ; 19.1 mm 1.24 mm Muntz tubes.
19. Determine the over-all 17 of a condenser from individual predictions of Uy, U^,
Ut, and U^, Data: 22.2 mm x 1.24 mm tubes; 2.l3m/seo water velocity; 7*G mtd;
23.9*0 average water temperature.
PROBLEMS 289

20. Calculate the film coefficient of steam flowing through S8.lmm x 2.67 mm tubes
at 10.64 kg/om* ga 22YX)t at the rate of 1220 m/min.
21. Determine, with Eq 9<*10, the film coefficient of conductance of water in a
60.6 mm x 3.18 mm Steel tube. Average water temperature, 110*C; velocity, 1.6S m/seo.
Compare result with a coefficient calculated from Eq 9-20.
lU. An assumption of 816”C was made in Ex 2, Sec 9-11. Using the results
computed in that example, what would the tube temperature actually have been? What
percentage correction does this lead to in M2 and k2^
23. Make an estimate of the over-all conductance, gas to water, of surface con¬
vection through a 60*8 mm x 3.05 mm steel tube having an Interior water flow at 0.61
m/sec, with water entering at 99^C, leaving at 16rC. Gas flow is transverse at G » 24,410
kg{Mrhr-mr and has temperature change 308*^0 to 274*C. Gas is to be considered as
having same thermal properties as air.
24. Employ Eq 9-19 to predict q/A^ of a condenser heat transfer surface where the
tubes are 22.2 mmx 1.24 mm copper alloy. 32.2^0; t|, IC.T'C 27.8T. Water velocity^
1.83 m/seo. Ug» 9766 koal per m^^hr-deg C.
25. Find the ERS and EPRS of radiant surfaces for tubes on a furnace wall in
arrangement (a), Fig. 9-14. There are 20 tubes ss.Ommdia x 3.05 m long. Assume a
slag factor of 0.92.
26. Find the ERS of a furnace envelope consisting of 3 walls arranged as in Fig.
9-15 (3), one wall refractory, floor refractory, and roof a tube bank. BS area of walls,
10.7 m* ^ach; of floor and roof, 8.4 m* each. 6 == 0.55v for walls; 1.5v for tube bank.
8 = 0.90; D = 76.2 mm; L = 178 mm.
27. Solve Ex 9-26, finding EPRS instead of ERS.
28. Find the “fraction cold'^ of the furnace of Prob. 9-26.
29. With the Orrok equation, find temperature of gases at furnace exit for a case
of combustion where bituminous coal of Qj^ = 6255kcal/kg is burned on 15:1
A:F ratio. Air temperature, 26.7®C; 0^9.8 kglnflERS. Allow for combustionlheat losses
of 280 kcal per kg coal. Cp of gases = 026 kcal per kg-deg C.
30. A steam generator is equipped with air preheating auxiliary heating surface.
Find the most economical temperature to which to cool the gas leaving preheater,
using the following data: fia, 371*C; t,, 21.rc Coal, $4.92 pertonne ; Q,, 7225 kcal;
Cg, $.3.23 I7„, 8.8 Operation 6000 hr per yr. z (Eq 9-16) = 0.70; over-all thermal
efficiency, approximately 80%.
 CHAPTER 10

10-1 Components. The modern steam generator is an integrated assembly

of several essential components. Its function is to convert water into steam at a
predetermined pressure and temperature. Tliis is a physical change of state,
accomplished by transferring heat, produced by combustion of a fuel, into the
water. Commonly it is a constant pressure i)rocess. The steam generator is a
pressure vessel into which liquid water is pumped at tlie operating pressure.*
After the heat has vaporized the liquid, the resulting steam is then ready either
for delivery to the user or for further heating in a superheater. Heat transfer
and the containment of a fluid pressure are the chief functions of a steam gen¬
erator.
A complete unit consists of (1) pressure parts; (2) enclosure, or setting; (3)
combustion equipment; (4) auxiliaries. It is an unusual purchaser who can, by
examination, determine whether a builder's design is adequate. Since “fired
pressure vessels^^ are an explosive hazard if insufficiently strong, reliance is
universally placed on requiring the manufacturer to conform with safe construc¬
tion practices as set forth in some authoritative set of rules. The ASME Con¬
struction Code for Power Boilers is the accepted standard in the United States.
Buyers of boilers automatically obtain the benefit of years of specialized experi¬
ence when they insist that their new bcnler bear the ASME stamp of conformity.
The pressure parts are (1) the boiler heating surface with attached drums or
shells for storage of water and steam and (2) the superheater surface, which is
simply more heating surface through which the steam must pass after leaving
the boiler if a final superheated state is desireil. In reheating cycle plants an
additional superheating surface is provided, but with this difference—the re¬
heater carries steam at much lower pressure than the superheater.
The heating surface itself is mainly in the form of seamless steel tubes for
which standard sizes most frequently seen are between 50.8 mm X 101.6 mm,
outside diameter, varying by 12.7 mm increments
These tubes are manutactured in standard wall thicknesses, varying by 0.26
in. increments to2.67 mm, then byO.38 mm. increments to4.57 mm., and by
larger increments above that. The maximum allowable working pressure is

* Subseqvently the reader will see hpw this pressure is maintained by combugtion con^
troL
COMPONENTS 291
governed by a combination of diameter and thickness, expressed in formulae
of the boiler construction code.
Drums are cylinders with spherical ends, of welded steel plate construction,
except in designs for very high pressures where required wall thickness exceeds
the practical limit of welded construction and forgings are used.
The rate at which heat can be transferred through heating surfaces varies
greatly in practice, for it is affected by the many factors mentioned in Chapter
9. The comparative rates at which surfaces located in different parts of a steam
generator can be instrumental in transferring heat energy from gas to water are
illustrated in Fig 10-1 This shows a pulverized coal-fired unit with water-
cooled furnace, superheater, and air preheater.

Fig 10-1 Typical conditions of temperature and heat transfer, koal per ^qm-hr.

The pressure jiaxts-Aca properly connected to produce some desired pattern

of flow of the water, then provided with suitable inlets, outlets, and “trimi! The
weight must be supported by an adequate steel structure. A furnace region for
the combustion of fuel and burners or stokers to produce that combustion are
essential components. The whole unit is surrounded and enclosed by a setting.
This is a wall, or casing, suitably shaped, to enclose and direct the flow of com¬
bustion gases, to support other components (but usually not the pressure parts),
and to insulate against flow of heat from the interior to the boiler room atmos¬
phere. The whole assembly is a steam generator^^Although it is oi^ referred
to as a ^'boiler,” actually only the pressure parts wherein vaporisation occurs
fliould be so designated.
292 STEAM OENERATORS
10-2 Design Criteria. A survey of steam generators may be quite confus¬
ing to the uninitiated because of the considerable' variations in shape, size, and
arrangement of the essential components i)re.sent in this very competitive field.
A purchaser may select his unit from many different designs. Except for
space requirements, nmst of them could fit a new installation equally well. The
purchaser of a water-tube boiler is faced with the necessity of choosing between
straight tube or bent tul)e, single or multii)le drum, header or headerless types.
Favorable and unfavorable points may be arlvanced for each of these and in
the final analysis the type is adopted which seems to promise the most advan¬
tageous features for the individual installation and, at the same time, provide
for meeting the unfavorable characteristics which will also be present.
The principal objectives for the design of central station steam generators
are set forth as the following desirable characteristics:

1. High efficiency rcprosf’ntcd l)y transfer of the required heat with minimum
losses. This involves the use of auxiliaiy'^ heat transfer surfaces, well-insulated casings,
effective baffling, etc.
2. High availability for service. The tendency to build steam generators of sufficient
capacity per unit to serve a turbine generator, unaided by other boilers, emphasizes
the need for inbuilt ability to remain ^'on the line^^ continuously. Ability of the
materials, and especially the joints, to stand the high temperature service reflects on
this. Also, the degree to which the design was successful in preventing slagging of
tubes by molten ash and the performance of the ash removal system are critical to
availability. Furthermore, the availability of the boiler auxiliaries must be equal to
that of the boiler itself,
3. Ability to deliver clean steam. See steam purifiers, Sec 10-10.
4. Uniformity of superheat, indicated by ability to deliver superheated steam at
the specified temperat\ire at full load, with superheat controlled within narrow limits
over a siiecified load ramre. Superheat is treated in Sec 10-0.
5. Ability to accommodate variations in rate of steaming without unsteadiness in
steam pressure, surging of water levels, development of localized overheating and other
transient phenomena.

In boiler selection, thermal, hydraulic, and structural factors are to be con¬

sidered and duly weighed. Heat transfer is the primary purpose of the boiler.
The theory of heat transfer in boilers has been considered in the preceding chap¬
ter. Fig. 10-1 has shown typical disposition of the heating surface of a large
steam generating unit, including data on the temperatures to be found in such
an installation. Heat transfer for simple boilers cannot be effected at such high
rat(*s as foi* boil(‘rs with auxiliary surface since the average temperature dif-
f(‘rence, gas to water, may be left higher when there is auxiliary surface to
finxnve the heat in the gas leaving the boiler. The higher temperature differ¬
ence pennits the use of a smaller boiler to accomplish the required steaming.
Another way to increase the rate of steaming per square metre is to increase
the speed of gas flow past the tubes at the cost of increased draft loss.
Physical factors also have some bearing on the selection of boiler type and
details. Head room and floor space requirements must be fulfilled, especially
where the equipment selected is a replacement of obsolete, inadequate, or worn-
out bmlers. Conversely, when floor space is at a premium, the b<^er designer is
DESIGN CRITERIA 203
able to supply any required capacity by extending the boiler and its auxiliary
surfaces vertically.
Where neither head room nor floor space is limited, draft may be an im¬
portant factor in selection of boiler type. In cases where a boiler serviced by
natural draft is to be replaced by a new one, the draft resistance of the new
steam generating unit must be kept carefully within limits if it is intended that
the unit shall continue to operate on natural draft.
The tendency of large boiler design toward use of higher pressures, water
walls, ash screens, and radiant heat transfer accentuates the need for a good
hydraulic design. The hydraulic circuit of a boiler consists of the path of water
flow induced by the difference between heads of water and water-steam mixture.
Steaming tubes and their risers are filled with a mixture of water and steam
which, having a lower density than water alone, is displaced in the direction of
flow by an equal head of water. The higher the steam pressure, the denser the
^steam becomes, and the less the ^fference between the density of the steam-
TmYf.iit!Q.AnrfTliM ivmr aion^ijonsequeiiwjT-Is 11101 e difficult to
achieve in high-jiressure than in low-pressure boilers, and it is in
I^rgssure boilers th&jU:)Ump ciJCg.ulation has been oiiCasiiuiatiy-^ployfd,,
factor created by high steam pressure is tbe^ncrea8edTate of steaming causeetby
decrease of the latent heat. Water must, therefore, be supplied more rapidly to
the tubes, and steam must have easy clearance from the tubes to the steam
drum.
High rates of heat transfer require carefully designed hydraulic circuits in
order that the flow will be ample to prevent premature conversion of all water
in the tube into steam with likelihood of a burnout due to over heating. Also,
the use of water walls and complicated tube arrangements, sometimes seen in
multiple-drum, bent-tube boilers, makes the hydraulic circuits less obvious and
more difficult to predict.
Considerable interest has been displayed in controlled circulation boilers in
the field of the high-capacity high-pressure boiler. Several such boilers are in
operation. This is not the once-through, forced circulation principle in which
4I0W must be produced by the feed pump forcing water through the heating
ciiTuits at a velocity i)roportional to load. Rather circulation is produced by a
separate pump running at constant speed. 6nly part of the water pumpeff’ is
evaporated, the remainder is recirculated. The circulation ratio is designed to be
l^slhanTor liaturaT circidation boilers. Orifices are employed to regulate' rate
of water flow m SiffcTent s^^ Of'the lieating surface. Small-diameter tubes
(38*lmmdi)2ire practical to use.Circulation is produced by one or more pumps in
. parallel, with a spaic in reserve. TTich^uxS in hi(^-premtrc boilers, where
steam density approaches water density is made smoother and more reliable by
the substitution pf controlled for natural circulation.
In spite of the wide variations in designs Ihere are certain requirements
fundamental to all boilers. The plant designer may turn to these for guidance in
investigating the boilers offered for his particular installation, for such require¬
ments should be met by any design deserving consideration.
First, there are the conditions governing behavior of the water within the
boiler. Most important of these is good water drculatum. The process of evolu-*
294 STEAM GENERATORS
tion in boiler development has eliminated types with faulty water circulation;
yet ‘‘tube starvation^' is a condition not unknown in the upper tubes of some of
the most modem, high-capacity boilers. The disengagement surface, where the
steam breaks through the surface of water in the drum, should be unrestricted.
Priming may be due to faulty design in this respect.
Provision of suitable storage space for steam within the boiler is a require¬
ment indirectly connected with the water conditions. The volume of steam
storage should be equal to the demands of the load served. Engine loads with
their pulsating cutoff will require more steam space than turbines to which the
same amount of steam is supplied. Insufficient storage space has an adverse
effect on the steadiness of steam pressure under variable load.
Another point to which attention should be given is the baffling. The path
of the gases through the boiler should be so baffled that they pass the tubes a
sufficient number of times to give up their heat to the required degree. This
degree is less when auxiliary heat transfer surfaces are provided. Practice has
determined the best baffle arrangement for most standard boilers under ordinary
firing conditions.
Certain features of a boiler should be investigated with respect to the possi¬
bility of undetermined thermal stresses being set up. Feedwater should dis¬
charge into the boiler at as near the saturation temperatufe as ^ssible. Cold
wat^ discharged against the boiler shell sets up contraction stresses. Joints and.
seams should be well protected from the direct action of flames or hot gas. The
setting of burners should never be made in such a way that the flames may play
directly upon tube surfaces.
To provide for intelligent and safe operation of the boiler, the engineer
should see to it that a full complement of leads, gauges, and safety devices is
provided. These include blow-off, steam lead, feedwater lead, water gauge
pressure gauge, superheated steam thermometer, safety valves and fusible plugs.
The setting of a boiler should provide for the introduction of hand or mechanical
soot-blowing devices.
Last, but not least, is the necessity of having an accessible boiler. This pro¬
vision is a requirement of all boilers which are expected to be insured so that the
insuring company's inspectors may determine from time to time the state of the
risk. Also accessibility should be provided for maintenance, inspection, and
repair by the regular operating force of the plant. The design should conform to
the ASME Boiler Construction Code, but within the limits of the Code there is
much latitude allowed the designer.
The purchaser of a steam generator naturally wants his new equipment to
be able to deliver the necessary quantity of steam, but will not want to invest in
unneeessary surplus capacity; hence, means for describing the production
capacity are needed. Steam generators are now being rated by the kg per hr of
actual steam generating capacity and by ASME units* of evaporation, at the
same time specifying operating pressures and temperatures.
Anothef capacity unit of long standing but dwindling usage is the boiler
horsepower. This was defined by the technical committee at the Philadelphia
Centennial Exposition in 1876. At that time the average engine would operate

* The unit is kB, a thousand Btu heat transfer (oonverted in ko»l or Jonlae).
DESIGN CRITERIA 2B5
onlS.4kg6team per hp per hr. Because it was thought desirable to have a
'^boiler horsepower” supply the steam for an engine horsepower, about iM kg
per hr of steam-raising capacity was selected as the unit of boiler horsepower.
On account of the variety of steam and feedwater conditions met in practice, a
hypothetical standard of 15.44kg steam per hr generated dry and saturated at
lOOX! from feedwater at the same temperature was taken for the purpose of
defining a developed boiler horsepower. This operating condition is now famil¬
iarly referred to as “from and atlOO®C” The heat transfer represented byl5.44
kg team generated from and atlOOX! is 15.44x1000x2257.0x2.3885 per hr.
X 10* kcal/hr.^ When the actual heat transfer per hr in a boiler is divided by
2257.0 the quotient is said to be the ‘‘Equivalent Evaporation.” The term
“Factor of Evaporatbn” is sometimes employed. It is the actual heat absorption,
per kg steam generated, divided by 2257.0.
At the time of defining the standard developed boiler horsepower, the
average rate of evaporation in boilers was 14*65 kg steam per hr per sq metre.
Comparing this with the average steam rate of engines at the time, it was
judged that 0.91 m* of heating surface would be needed to generate steam for
an engine horsepower.Consequently this surface (1.1 m* for fire tube boilers) was
adopted as a rating for “boiler horsepower.” The 10:30 ratio of heating surface
to steam rate has long been obsolete because steam rates have been reduced to
less than 4.5 kg per hp hr, while evaporation rates per sq metre have doubled and
tripled. Because the normal developed horsepower began frequently to exceed
the rated horsepower, the term “per cent of rating” came into use. The modem
high-capacity boilers may operate at as much as 400% rating, and 200% is
common in numerous installations. Unlike ratings in other fields, thisiMs not
construed as an overload on the boiler, but rather is the normal continuous load¬
carrying capacity.
Boiler horsepower and per cent rating are terms w’^hich have been abandoned
in large steam generator nomenclature. As water wall, economizer, and super¬
heater surfaces increasingly account for total heating surface these terms be¬
come meaningless. However, in the field where the unit is a steam boiler of
standardized design this nomenclature is still used. In some small boiler fields
maximum developable boiler horsepower is a standard method of rating.
Low-pressure boilers intended for heating use are frequently rated in terms
of the number of square metre of equivalent direct radiation (EDR) they will
supply.

Example 1: A boiler with 140 m* heating surface was found to evaporate S720 kg
oftteaaoLperhrat 18*(lkg/em>aband 0.9S diy ness factor. The feedwater was at 66.0*C.
The vanous capacity expressions previously mentioned will be determined. The steam
atote ooriesponds to an enthalpy of 2757x10* J/kg; hence increase of enthalpy produced
by the boiler was (2767—274.24) 1000 » 2483x104 J per kg evaporated
Rated boiler hp = 140/0«91 — 153-9 bo bp.
Developed boiler hp =3720x2483x104/36322x104 « 281.6 bo bp.
Percent rating = 281.6/163.8 » 170%.
ASME evaporation units = 3720x2483x104 «. 8237x104 J/hr,.

4 Actually. 8323x 10* koal per bt bp..

296 STEAM GENERATORS
Factor of evaporation 248Sxl0*/2257xl0* 1.1.
Equivalent evaporation = 8720x1.1 » 4002 kg per hr, from and at 100*C.

The plant operator should be interested in the efficiency with which steam
generating equipment performs its duty, for fuel costs are minimum when the
eflSciency of this equipment is maximum. There can be a great range of
efficiency. Many plants can be found which day after day get no more than
60fo of the potential heat of the fuel into steam whereas others operate con¬
sistently at better than 75%. Steam generator thermal efficiency is usually,
though incorrectly, called “boiler efficiency.”
By ^^boiler efficiency** is meant the measure of ability of a boiler or steam
generator to transfer the heat given it by the furnace to the water and steam.
But boiler and combustion equipment are so much a unit that ^^boiler and
furnace efficiency** is more important than boiler efficiency. To most persons in
this field the term “boiler efficiency” connotes over-all thermal efficiency, that
is, the percentage of the higher heating value of the coal which will be in the
steam. Efficiency can also be designated by kg evaporation per kg coal fired.
The evaporation may be based either on actual evaporation or on equivalent
evaporation “from and at 100®C”
Example 2: Three methods of expressing steam generator efficiency will be illus¬
trated using, in addition to the data of Ex 1, the hourly coal consumption of that
boiler. This is taken to have been 481 kg of coal of 28498.5 k J per kg higher heating
value.

Over-all thermal efficiency = 75-1%

Actual specific evaporation = 3720/431 » 8.63 kg per kg coal.

Equivalent specific evaporation = 8-63 xM *=» 9-49 kg pt*r kg coal.
An efficiency such as the 75.1% computed above implies a 24.9% waste of
potential heat energy. liater in this chapter we shall show how plant testing
permits the engineer to break down the total loss into components.
Grate efficiency is a term sometimes employed to describe the effectiveness
with which the combustion system prevents unburned carbon sifting from the
combustion region into the refuse. It may be calculated as follows:

Grate I'ificioncy = 1 —CrefQc/Qh (10-1)

in which Tref = Weight of combustible in refuse, kg per kg coal burned.
Qe = Heating value of the combustible. It is usually assumed
that the combustible loft in ashes is fixed carbon, for* which
3382xl0M/kg.
10-3 Boiler Types. Mention has already been made of different ways in
which tubular heating surface may be assembled in boiler units. In order to
cover the main variations as concisely as possildc, a classification system is
needed. To begin with, we shall eliminate from our description that large seg¬
ment of boiler practice which is connected with the production of low-pressure
heating steam. Cast-iron l>oilers and other fonns eminently suitable, durable,
and economicahfor generating steam at pressures up tol.05kg/cm* ga constitute a
BOILER TYPES 297
class of boilers which may well be omitted from our consideration, for the nature
of this book implies “power” boilers, i.e., those types which can safely stand
high pressures—7 kg/cm* ga upward.
In the pressure boiler fi^d A.pjdlXlftry classification would be according to
contehtTbt the"fiibuTarTieating surface—^water or gas. The result is a grouping
into fire-tube and water-tube boilers.
Fire-tube boilers arc those in which the products of combustion pass through
tlig tubes and the water Tics around the outside of tjy;m. This rcquircs~that the
tubes be surrounded by a shell so as to confine the water and contain the fme-
sHre.^Thc shell thus becomes a support for the heating surface and sometimes for
the combustion equipment. If the reqiiijcd capacity is not above that for which
this prmHple'ls“practicairthfi fire-tube boiler has advantages of compactness,^
unit construction, portability, and inexpensiveness that are continuing it in use
in spite of certain disadvantages. In most fire-tube boiler construction a nest of
“luTics is built into a shell. The tubes are straight and parallel to each other, and
to the axis of the shell. Variations are: (1) horizontal or vertical axes; (2) ex¬
ternal or internal furnaces; (3) fully cylindrical or partially cylindrical shells.
For low design pressure, some shells are built as a combination of cylin¬
drical and oval sections, the latter having to be internally stayed to hold their
shape. The building in of a furnace section may also require a variation from
the cylindrical, and again internal staybolts are always much in evidence.
File-tube boilers have a relatively large ratio of water content to steaming
capacity; hence fluctuations of steam demand cause only little unsteadiness of
steam pressure or water level. Simple automatic combustion control systems
may be employed, or if the control is manual the supervision does not have to
be so close and continuous as with water-tube boilers, which can boil dry during
a few minutes of inattention if under manual operation.
Water-tube boilers, which consist of tubes and drums, may be classified as
straight or bent tube.
“Straight tube boilers have a parallel group of straight equal-length tubes,
arranged in a uniform pattern and joined at either end to headers. These
headers in turn are joined to one or more horizontal drums. According to their
construction, headers may be classified as box or sectional types. Their chief
characteristics are:
Box /leader.Least expensive; must be internally stayed against the fluid
pressure; header surface must be perpendicular to tube axis, hence cannot be
vertical, since'liabes must be inclined to the horizontal in order to control circu¬
lation.
ak
Sectional header. Suitable for highest pressures;, since header surface is not
a flat sheet, the section can be so cast or forged that although the header is
Wtical it has a surface at the tube hole that is normal to the tube axis. Sections
are made sinuous in order to stagger tubes vertically.
Typical arrangement of the pressure parts of straight tube boilers are shown
in Fig. 10-2. The hoY header is seen to resemble a large, shallow, struc^ral steel
Imx, whereas the seotienafheader is a vertieal casting Gt forging df 'small trans¬
verse dimensions, each section accommodating a group of tubes in a vertical
r^, the width ol the tube bank being determined’By the number of sections
stacked side by tide.
208 STEAM GENERATORS
During assembly the tubes are inserted in holes bored to receive them, then
rolled * tightly into the headers. Holes in the headers opposite the tubes are
closed by tapered caps driven from the inside, but some rows must have oval
hand holes so that caps may be passed through them for use in the circular
holes; then the oval ones are finally closed with oval capsf having threaded
stems so that they may be secured from the outside. When caps are applied to
the inside of handholes, the steam pressure acts to tighten them. Small initial
leaks at tube joints are generally quickly ended by the formation of a rust
coating.

Fig. 10-2 Heating surface of straight tube boilers.

Top, Box header, longitudinal drum.
Bottom, Sectional header, cross drum. (Note diagrams of header construction.)
process wherein a multiple-roller tool is placed inside the end of the tube, after
which, by simultaneous revolving ^d expansion of the rollers against the tube wall, the
wall is slightly plastically deformed so that it is pressed tightly against the header wall and
locked in place. This is superior to welding in that subsequently a tube can be cut and
removed for replacement, leaving the header undamaged. However, it does require that
the plane of the tube hole be normal to that of the tube axis.
t The reader will see that an oval cap of sufficient sise to cover an oval hole may be
passed through that hole.
BOILER TYPES 299
Bent^tube boilers are headerless. The drums serve the same function as the
headers. A few, bent-tube boilers are built with one drum and two oi^ more
cylindrical headers, but the latter could be said to be undersized drums, so that
as a general principle the bent-tube boiler is multiple-drum. The tube holes are
bored directly in the drum walls. The drums must be large enough (l.rm dia.
and up) to have a manhole in the ends, for the mechanic must get into the drum
to perform the tube rolling procedure during assembly. Drilling of the holes
and “rolling in” of the tubes are most practical where the axis of the tube inter¬
sects the drum surface radially. Hence all rows of tubes except one must be
bent. The nature of the bends is best understood by examining drawings of this
type of boiler. The common heating surface arrangements are:
Drum-and-a-half. A long upper drum is paralleled by a shorter lower drum
(Fig. 10-8). The space below the upper and ahead of the lower drum is utilized
for a furnace. Water wall * tubes cover the furnace side walls and are connected
to the top drum and to lower headers. A large bank of bent tubes is connected
between the two drums. This is a compact inexpensive design for small boilers
where a water-cooled furnace is permissible.
Two-drum. Two parallel horizontal drums of equal length but not neces¬
sarily equal diameter are set one above the other and joined by multiple rows
of bent tubes (Fig. 10-7). If the plane of the tube bank is parallel to that of the
drum center-lines, the tubes will have two bends, .but many two-drum designs
require three bends per tube.
Three-drum. Two upper drums and one lower are arranged so that one upper
drum carries the water level and the other, being lower, really acts as a header.
Circulation tubes connect the upper two drums. A wide latitude of arrangement
is permitted by the three-drum arrangement. Formerly boilers were built with
four or more drums, but recent trends are toward one-, two-, or three-drum
designs, even in the largest units.
Purchasing Standards. Although the number of thermally successful boiler
types is almost endless, differences in construction cost, bulk, repairability, and
the like, which are determining factors in competitive selection, have greatly
narrowed the field of choice. Except where unusual conditions prevail, the
number of boiler types that can be offered in economic competition, based on
broad specifications set forth by the prospective purchaser, is not large. To be
able to write specifications broad enough to invite the competition of several
different boiler types, and at the same time adequately protect the dwner^s
interest, not only on first cost, but in the “long pull,” is a talent not possessed by
the rank and file of engineering superintendents or consultants. This is particu¬
larly true of the small boiler field where the purchasers are industries, institu¬
tions, mercantile establishments, and government agencies. Here purchase cost
is often the main factor influencing the purchaser. Those unseen inbuilt charac¬
teristics which lead to a long, trouble-free life are often difficult to incorporate
into competitive bidding without restricting the number of competitors. This
poses a particularly difficult problem for public agencies which by law are
required to place their capital outlays on the broadest possible, nonrestrictive
competitive bidding.

♦ See Sec 10-6.

300 STEAM GENERATORS
In the boiler manufacturing business there is some standardization on small-
and medium>capacity units. Fire-tube designs lend themselves well to this,
especially the multiple-pass, intemal-fumace horizontal type. This type has
recently received great impetus through its wide acceptance as a packaged
steam generator for pressures up to 17,6 kg/cm‘in an oil burning unit where the
‘^package^’ includes everything but fuel oil tanks, making the necessary design
and installation work at the customer's premises practically nil. A degree re¬
moved from this unit are the small water-tube designs, usually drum-and-a-
half, or two-drum, which are also sometimes called packaged units, but with
less cause. These are also of standardized designs, and the manufacturer may
keep a stock of the components on hand, but modifications are possible in each
sale so that different combustion equipment and casing designs may be offered
to suit the owner's desire.
Roughly the limits of the package fire-tube steam generator might be said
lobe 17.6 kg/cm* and 9072 kg steam per hr, while that of the standardized
water-tube boilers could be placed at 17.6 kg/cm* and 22680 kg per hr.
A boiler for a capacity of 45360 kg per hr would be classified by manufac¬
turers as a large boiler, and they would call the pressure ''low" if it were less
than 28.1kg/cm*.Mo8t large boiler8\wouldbebuiltfi>rinorethan28.1 kg/cm* oper-

Combustum Engineering-Superheater, Inc

Powm.
Fio. 10-3 Typical setting of an HRT boiler; and above, the boiler itedf.
REPRESENTATIVE FIRE-TUBE BOILERS 301
ating pressure, and they would be tailor-made tofitthe customer’s needs. It canbe
assumed that the customer lor such a unit would have retained adequate
technical counsel to guide him in specifying his needs and evaluating the offer¬
ings. This counsel is equally important for the purchaser of a small boiler, but
unfortunately many purchasers of small units prefer to substitute the uncertain
economy of purchasing a standardized unit, without any in-place testing to
determine whether it fulfills claims made for it, for the cost of professional ad¬
vice.
10-4 Representative Fire-tube Boilers. The more important representa¬
tives of this field are the horizontal return tubular (HRT) boiler, the horizontal
two-pass or economic boiler, the locomotive type, the round upright type, and
the horizontal multi-pass or modified Scotch marine type The first and last of
these are chosen for further description

CombusUon Engineertng-iiupcrhcaUrf Inc

Fig 10-4 “Economic*' type boiler

The HRT boiler is characterized by simplicity and cheapness As the

furnace is external to the shell, almost any kind of combustion equipment can
be accommodated. The pressure parts consist of a long cylindrical shell with
flat end sheets which are bored to receive longitudinal tubes. The boiler itself is
usually suspended from overhead girders and a brick setting built around it as
shown in Fig, 10-3. The tubes themselves act to stay the end sheets, but where
in the upper part there are no tubes, the manufacturer must add stay braces to
resist deformation of the ends by steam pressure. The water level is carried high
in the shell, well above the highest tube. The furnace gases flow horizontally in
contact with the lower half of the shell, then reverse direction and pass back
througl the fire tubes, finally leaving at the front or firing end, where a cylin¬
drical metal extension of the shell serves to guide them into the smokepipe or
302 STEAM GENERATORS
breeching. Heating surface is partly the shell and partly the tubes. A longer
journey through tubes is provided in a modification illustrated in Fig. 10-4. This
two-pass arrangement is often called an “economic” boiler, probably because
lower final gas temperatures are achieved in two-pass boilers compared to
HRT’s of the same over-all length.
The package oil-burner, fire-tube boiler is shown in Fig. 10-5. The shell is
horizontal and supported by brackets from a structural steel base. All other
equipment is mounted on the same base or on the shell itself, so that the entire
unit may be built in the manufacturer’s shops. It is also tested and adjusted
there, including capacity and efficiency runs, and is shipped as a single unit to

Prefentd VHUHm Co

Fig. 10-5 Unit steam generator. This fire-tube, intemal-fumace type is frequently
called a “Package Jailer.”

the customer, who, after preparing a simple foundation, is ready to receive it

and has a complete working ^am plant shortly after the unit is deliva«d. This
is in agreeable contrast to the cu^mary water-tube construction, where the
boiler is, so to speak, built on the customer’s foundation, then auxiliaries and
accessories added roundabout on previously prepared bases or mountings. Were
it not that the fire-tube design has limitations of pressure and capacity, it
would loom up more importuitly in the medium and large boiler field.
Inside the shell the water level is carried fairly high. Tubes extoid from
front to back and a large flue, at or near the geometric center, has the oil burner
attached to it. By partitioning built into the reversing gas passages, front and
REPRESENTATIVE WATER-TUBE BOILERS 303
rear^ the tubes are separated into groups which in succession carry the gas from
back to front and vice versa. Considering the large central flue as one pass,
these boilers are built as two-, three-, and four-pass units. A four-pass unit is
illustrated.
Gas velocities are kept high, thus promoting rapid heat transfer. Multi¬
passing keeps gas in the tubes long enough to transfer most of its heat. Eighty
percent over-all efiiciency is a common manufacturer's guarantee. The high
velocity and numerous turns require more draft than in almost any other boiler,
so that fan power is disproportionately high. Draft losses of 51tol02 nun water
are incurred in these units.
The combustion equipment consists of a high-quality oil burner set in a
small refractory chamber attached to the boiler-front at the main flue. The small
combustion space completely gasifles the oil, but combustion continues through
the large flue. Absence of ash simplifies the installation. Complete combustion
is achieved if the equipment is in proper adjustment. All necessary draft is
created by the unit’s fan and, as there need be no smoke, only a simple vent
pipe to atmosphere is needed to dispose of the products of combustion. Oil
pumps, combustion control, and operational safeguards are part of the standard
un^, but feedwater pump must be installed separately by the purchaser.
nO-5 Representative Water-tube Boilers. Straight-tube Boiler, Straight-
tube boilers have advantages in ease of access to tubes for cleaning or removal.
Since all tubes are the same length, the stocking of spare tubes is simplified.*
Access to tube ends is more readily obtained through headers than drums.
However, straight-tube boilers in small- and medium-sized units, where the
boiler tubes are the principal heating surface, tend to be more expensive per sq
metre of heating sur&ce b^use of the cost of the headers themsdves.
The heating surface of a straight-tube boiler was shown in Fig. 10-2, in
both long-drum and cross-drum arrangements. A cross-drum boiler is also shown
in Fig. 10-6,t with setting arranged for firing by coal stoker. This is a sectional
header boiler, with an added feature of a water-cooled rear furnace wall. The
boiler tubes are inclined about 15® to secure the necessary hydraulic circulation.
Tube flows are derived from, and delivered to, the two headers. The rear or
supply header is connected to the bottom of the drum; the front header is con¬
nected slightly above the normal water level by means of horizontal circulation
tubes extending from the top of the header to the drum. This setting is ‘‘brick-
set,’* with sheet steel casing and transverse baffling. Here the baffles are a
combination of refractory barriers and alloy steel sheet. These direct the gas
flow crosswise of the tubes, with three passes over the tube surface before the
outlet is reached. Refractory tiles are laid on the rear wall tubes and on the
circulation tubes. The former are part of the baffle system and also create a
cinder pocket in the rear. The upper tiles, with insulation covering, form the

* Exo^t for plants remotely located, this advantage is more theoretical than real. Usu¬
ally a replacement tid>e can be quickly obtained from the manufacturer. Also, tubes may
be removed and tube holes temporarily plugged. Welding repairs are now possible.
t If the reader will carefully compare the two pa^ of this illustration (also.Fig, 10-7),
he can learn how to visualise the internal construction of a boiler in three hy
t3rpical fiat sectional drawings, which are the usual form in which tl^ information is
available.
304 STEAM GENERATORS
roof of the boiler. A drain and blow-off connection is required from each of
the lower rear header and lower wall header.
Bent-tube Boiler. There are more bent-tube than straight-tube boilers sold.
This is probably because purchase costs tend to be a little lower and many con¬
tracts are let on a strictly competitive basis. There are more variations of
heating surface arrangement available to the designer. This is of assistance in
arranging the elements of a steam generator which may have superheaters,
water walls, and economizers in addition to evaporative surface. Also there is
more latitude in baffle arrangement than in straight-tube designs. Tubes are
more difficult of access for cleaning and removal, especially if the steam drum
has internal fittings for steam purification, for these must be removed before
access is had to the tube ends.

A boiler typical of small- and medium-sized bent-tube steam generators is

shown in Fig. 10-7. This consists of a two-drum boiler, water walls, and super¬
heater. The setting is arranged for pulverized coal firing.
The evaporating surface of this unit is composed of water walls on four sides
and the roof * of the furnace, and two banks of bent tubes extending between the
two drums. Refractory baffles direct the products of combustion upward through
the tube bank facing the furnace, then down over the superheater tubes, and
finally in a three-pass transverse flow over the large rear bank of tubes. The
water wall and front tube bank receive much heat by radiation, and some by
convection. Superheater and rear bank tubes are mainly dependent on convec¬
tion heat transfer. The major portion of steam produced is generated in the wall,

Floor is the ash pit with screen of ash-cooling tubes.

306 STEAM GENERATOHS
roof, and front bank tubes. Upper water wall headers, and superheater headers,
are suspended from the structural steel frame of the setting. Wall areas which
remain comparatively cool because of the overlay of water tubes are built of
large, flat, refractory tiles backed by heat insulation, such as mineral wool.
Other wall areas in tiie setting are firebrick with insulation. The outside surface
is then paneled with thin sheet-metal casing which makes it comparatively air¬
tight and which may receive finish painting.
Two bent-tube designs that have very successfully competed in the small
boiler field are those shown in Fig. 10-8. The Riley Type RO boiler is a two-
drum design with four water wail headers, all headers and drums being parallel
to the axis of the boiler. Furnace wall tubes cross over to the opposite top
neader, thus forming roof cooling. Circulation tubes connect the headers with
the drums. The Keeler Type CP has one long upper drum and a half lower

Riley Stoker Carp E. Keeler Co

Equal-drum type Dnim-and-a-half type

Fig 10-8 Longitudinal two-drum water tube boilers

drum There are two lower water wall headers. Drums and headers are parallel
to the boiler axis. Part of the steam drum is exposed to furnace temperature,
but there are few circulating tubes required.
Large steam generators are built to the purchaser's specification and are so
involved with other heat transfer surface that boiler tube banks are often a
minor portion of the whole unit. Examples of such units are given, but they
dannot be said to be ^‘typical,” because of the vast opportunity for individual
differences in this field.
10-6 Water Walls. The evaporative capacity of a water-tube boiler
be greatly extended, and at the same time a protection against high temperature v
can be afforded the furnace walls, by partially shielding the walls with vertical
tubes connected into the general circulation of the boiler.
This system of furnace cooling tubes is a water wall. Typical of the circula
tory connections to it are those shown in Fig. 10-11, in which the nonheating,
circulating tubes are emphasized As these are outside the furnace, they expand
WATER WALLS 311
Assumptions are; Air-fuel ratio, 11.4; pulverized coal firing with coal of Qh ■■ T$U
koal/kiyQL=7M7kcal/kg; slagging factor, 0.95; radiation plus incomplete combustion
losses, 1.5%. A rectangular furnace S.7 m X 4.1 m X 6.2 m high is taken, and for
simplicity is assumed to be covered on all six sides with water wall of the type shown
in ng. 19»12b»with 88.9 mm tubes spaced to 152.4 mm on oeiiters.This speoing permits
22 tubes on each of two side wells end 26 on each of the other side we11s,the roof; end
the floor. In this ecKemple, the date of Fig. 10-lSA ere employed.
The projected radiant heating surface, Ap, is calculated.
Proj. area (side wall tubes) = (26 + 22) x 2 x 6.2 x 0.0889 » 48.6 mt«
Proj. area (top and bottom) = 8.7 x 26 x 2 x 0.0889 ^ 16.6 mt.
Total Ap •m 60 m<.
To use Fig. 10-13A it will be necessary to calculate available heat, gas flow, and
effective radiant surface. A,.* The notation of Chapter 9 will be employed without
jr
redefinition. It would appear from comments in Chapter 9 that « = 1.2 ^ is a reasonable
assumption for this type of wall.

From Eq 9-27, At ^ flO X 1.2 ^ X 0.95 = 107.4 mi.

From Eq 9-25, Q» = 7867 (1 — 0.015) = 7256 kcal.

Gas per kg coal = 11.4 + 1 — ash = 12.3 kg if ash is 10%.
Available heat pei kggas = 7266/12.8 » 590 koal.
Coal consumed = 16.76 x 10®/(0.86 x 7634) • 2581 kg per hr.
Gas flow ~ 2581 X 12.3 » 31746 kg per hr.

ERS per 1000kg gas = Atj fiM = _ 3.64 m».

Entering Fig. 10-13A with ERS = 3.64 and avsilsble he»t of 590 ko»l, the furnace
exit temperature is found to be 1132^ This is based on 15.6*C datum of combustion
air. Were combustion air preheated, the enthalpy of preheat would have been added
to Qg.
From the furnace exit temperature the heat transfer to water walls could be
calculated, as could’the necessary transfer duty of the convection zone.

Example 2: The same case will be analyzed successively by the data of Figs.
B and C. Previous calculated results are employed where necessary. Area of the
furnace envelope, BS » 5.2 <3.7 + 4.1) x 2 -h 2 x 8.7x4.1-«112 m*. Reduce by 0.96
factor for slag effect. Available heat per sq metre BS » 2581 x 7256/(112 x 0.96)
175968 koal per hr.Entering Fig. B with 175.96 x 10* kcal, and xpulverized coal practice,'*
the f|imaoe exit temperature is determined to be 1099*0. Since the datum of enthalpy for
F**- is 26.7*0, a emperature of 1088*0 should be used to comparethis result with meth*-
V sed on 15.6*0 datUm.
Po use Fig. C it is necessary to reduce BS surfaces to EPRS, From Fig. 9-16,
i^th tube spacing ratio of 88.9/152.4, the ««peroent effectiveness" is found to be 0.986.
EPRS « 112 X 0.935 X 0.95 = 99.6 m«.
Available heat2681 x 7266/99.6 ■> 188656 koal per hr per m* EPRS. Entering
ig. C with 188.7,the limits of experience are l^alradiant heat abaoiption per

♦If different tube arrangements occupied different walls, then Ar = XsdAp.

312 STEAM GENERATORS
l^er m* Sensible heat in the gas at exit from the furnace =
X 99.5/31746 J} heal per kg. Aesuming Cp ■■ 0.20,
= {’j7*}/0.26 =
For a datum of 15.6*C, this would place the furnace exit temperature in the range
916 — 1170*C. The higher value would probably go with long flame burners or
stokers; the lower, with short flame burner.
Example 3: A solution of the same case by Eq 9-26 is presented for comparative
purposes.
Using data as previously determined,

0 ==-- ^-- = 0.515, fraction of Qa absorbed radiant!)’.

11.4\/258I/I07T
60

Then sensible heat left in gas at furnace exit = (1 — 0.515) X 7256 /12.3—284.5
koal per kg.
Exit temperature = 15.6 4 284.5/.26=s 11J0®C.

These examples imply that, although the physics of heat transfer from a
luminous cloud to water-cooled tubes is quite involved, practical solutions
have been evolved that are fairly reliable.
10-7 Installation. The installation of a steam generator in a boiler room
may range from the exceedingly simple to the extremely complex. The extremes
are:
Simple, A truly packaged unit such as a 150-hp oil burning fire-tube boiler
is received via railroad car fully assembled and pretested at factory. It is
handled in one piece and slid on its own base into place on a flat foundation.
When covered with heat insulation, joined to water, steam, blowdown, flue gas
vent, and oil supply lines, and supplied with electrical service and ignition gas,
it is ready to use.
Complex. A 45360 kg per hr power plant steam generator consisting of two-
drum boiler, superheater, water walls, setting, trim, combustion equipment, and
control equipment is received completely disassembled. One drum of this unit
may weigh as much as the entire boiler of the first example. There can be over
a thousand separate pieces to be received, inventoried, aifd finally used during
the erection of such a boiler which is, in a sense, built in place on the owner’s
foundations. The foundations themselves are complicated by special provisions
for ash disposal, air passages, and footings for carrying heavy concentrated
loads. In some cases the manufacturer’s contract includes erection; in others
an independent contractor erects, or the purchaser’s forces do so. The job may
consume upwards to a year’s time, and the unit may finally be tested for com¬
pliance with guarantees as much as a year and a half after the manufacturer
shipped the principal items (drums, tubes, and headers) from his factory.
The setting of a boiler may be thought of as all that material contributing
to the encasement of the heating surface and furnace region. Sometimes it is
built as a solid brick wall with inner courses of refractory material, but more
often nowadays it is a combination of refractory tiles on the inner surface
backed up by layers of heat insulation with a sheet-metal exterior casing. The
combustion equipment, reflecting arches, and other parts of the steam generator
FURNACE 313
may be built into, or be supported by, a brick setting, but the thinner, insulated
setting is not used for the mounting of any but light equipment. Combustion
equipment is separately supported from the foundation. The baffles which are
built against or across the tube banks to guide the gas flow are also part of
the setting, as are ash pits, soot hoppers, etc.
A major problem in the design of setting and support of pressure parts of
large boilers is thermal expansion. Tubes, baffles, tile walls and other parts at¬
tain different operating temperatures and proceed to different linear expansions,
all of which requires special attention in both design and installation in order
to preserve tight joints, avoid cracked walls, overstrained tubes, etc.
A fully installed steam generator will have its weight adequately supported
with allowance for thermal expansion. It will be completely encased with a
gas-tight wall capable of retarding the flow of heat from the hot interior to
a suflicient degree. This wall will, however, be pierced with numerous openings
for observation ports, lancing ports, soot blowers, draft gauge leads, etc. The
installation will have provided a furnace region and burners or stokers, and
a way of controlling gas flow through to a discharge conduit, called the uptake.
The complete installation includes disposal of waste products, viz., ash, cinder,
soot, and the proper introduction of combustion air. Numerous accessories,
either required or desirable, complete the installation. These are either for
instrumentation or are accessories collectively known as boiler trim*
10-8 Furnace. A furnace is a chamber for combustion. In addition it
provides support and enclosure for the combustion equipment—burners or
stokers. It surrounds the region where the combustion reaction takes place,
confining and isolating it so that it remains a controlled, albeit highly dynamic,
activity.
There are many interesting, puzzling, and difficult problems in the field of
furnace design. Many of the questions that arise have already been answered
by scientific testing and by experience, but some points are still met by the
liberal allowance method.
The design of a furnace cannot be carried out independently of other equip¬
ment, for its success will require coordination of several important factors,
among which may be mentioned:
1. Type of combustion equipment.
2. Character of the fuel used, especially its ash content.
3. Draft equipment employed.
4. Air supply and degree of preheating.
6. Boiler and its baffling arrangement.
Presuming that suflicient data on these factors are at hand, the furnace de¬
signer has to solve next a number of major problems before arriving at a
satisfactory result. What physical size is required for a particular installation
in order to provide sufficient furnace volume? What is the best possible shape
that will still satisfactorily accommodate boilers, stokers, burners, ash equip¬
ment, etc.? What wall type do the local conditions indicate? Will this type
give satisfactory combustion over the full range of variable boiler load ex¬
pected? Will the wall selected be satisfactory from the maintenance stand-

<•860 See 10-10.

314 STEAM GENERATORS
point, and will it be sufficiently cool on the atmospheric side? The correct
answer to these and other questions that will arise during the course of furnace
design are important, lor the first cost is no small item and alterations are
expensive in themselves, as well as removing from active service the whole
steam generating un|t.
In former years, especially when steaming equipment had been standardised
for a time, and l>efore pulverized coal firing had inspired the remarkable
progress in combustion, furnaces were customarily designed on the basis of
certain volumetric requirements per boiler horsepower or certain rates of
combiistion per m*of stoker surface coupled with experience-approved furnace
height to be allowed above the grates. Now furnace volumes have come to be
based on a unit of heat liberation called the heat release rate. This is taken
to be the higher heating value of the fuel, expected to be burned per hour
in the furnace, divided by the active furnace volume. To obtain a furnace
volume by this factor, the potential hourly heat liberation is divided by
some acceptable heat release rate. This release rate is not deducible from
theory, but fortunately there are available data on rates existing in successful
designs, i.e., those which have not experienced operating difficulties attributed
to undersized furnaces. Every time a ne^ steam generator is placed in operation,
another item is added to the accumulated heat release data.
It has been found possible to group these data under the influential factors
of (1) type of combustion equipment and (2) fusion temperature of the ash.
One such compilation is furnished in Table 10-1.

Table 10-1. FURNACE HEAT RELEASE RATES

OEBION VALUES—KCAL FBB U* FXB HB

Pulveriaed coal Below lUll^C Above

Ash fusion temp. . Hire 13J6 isie°c
Solid refractory. . Never 1US80 188500
Air-cooled refractory. . lOWOO 188500 155750
Water-cooled bottom. . 142400 155750 178000
Water walls. . 1M750 105800 887000
Lump coal*
Stoker firing Very large plant. 287000-8115001
Stoker firing Large plant. 178000-267000
Stoker firing Small plant. 188500-222500
Oil or gas. 222500445000
* Release par metre grate enrfiaee varies from M8S00-1S50000 keel per hr.
t Range for lump coal givot to cover various ranks of coal and wall types. Low rank
coals take the lower release value. Water-cooled walls permit the higher release values.

The applicaticm of heat release data will be shown for a case where it is
used to determine furnace hei^t.
Example 1: The dimensioDs of a furnace of a steam gennator, rated at 187S x 10*
kesl of imt transfer, are assumed to be determined horizontally by the shape of the
pressure parts and to be *Jim x The required hn^t inade this furnace will
be estimated, assinning water-coded walls. The coal has ash fusion temperature of
FURNACE 816
ISMI!. The thennal effioeney u usumed to be 80^. Pulverised coal firing. Hie fuel
must be potentially capaUe of liberating 1875 x 10*/0.8 = SMSTSOO keal per tor.
From Table 10-1 a heat rdease rate of 186800 koal per m* per hr is sdeeted as
appropriate to this case. Hmh fuinaoe volume = 8SU7000/105800 — no.? ni>.
I^uired furnace height = ii8>7/4^ x 44 -■ 6.8 m.
The type of boiler greatly influences the furnace shape. The firing equipment
itself may partially dictate the furnace shape. A stoker frequently fixes the
horisontal furnace dimensions. Pulverised coal burners, on the other hand,
allow the designer more latitude in the placement and number of combustion
units; hence the furnace shap^ is more ^Justable to other influences.
The shape should promote turbulence as much as possible. Needed turbu¬
lence may result from proper combination of furnace shape, burner placement,
and burner design. Furnace shapes are still further affected by the necessity
of ash disposal and by arches n^ed to promote ignition and to help bulm the
volatiles.
There are divergent ideas of how fuel burners should be located in the
furnace, but in general they should be aimed away from the opening leading
into the boiler tube banks (called the furnace aperture) and away from the
possibility of a direct impingement on the heating surface. The highly luminous
phase of the flame should not occupy a large part of the furnace volume. Gas
turbulence resulting from design or placement of burners is mueh.sou^t for
as an.aid to complete combustion.
The suspended ash particles (fly-ash) of pulverised coal and spreader stoker
furnaces may become critical to operation. The gas velocity in the furnace
should be made low enough to permit coarser particles of fuel to "hover,” thus
promoting completeness of combustion and effectiveness of radiant heat trans¬
fer. A low gas velocity through the boiler aperture will minimize the driving
of partly solidified ash particles against boiler tubes. Also, the tubes at the
boiler entrance should be spaced on sufficiently wide centers to delay brid^g
of such ash or slag as may stick.
The character of the furnace walls is an important factor in maintaining
continuous ignition and complete combustion, especially in furnaces unequipped
with arches. Two opposed considerations intrude themselves at this point. Tlie
walls should be as hot as possible, so that combustion will be complete and
heat transfer to water rapid. The walls should be as cool as possible for struc¬
tural strength and minimum midntenance. A judicious use of solid refractory
sections in water-cooled furnaces or of special refractory-covered water-walls
has helped, but the high-capacity, water-jacketed furnace is not extronely
flexible in carrying variable lo8d.
Many types of furnace walls are in use today. A primary classificati<»i of
them would be as follows:
1. iSoiuf JkfoMmrv.—This type langw from a sin^e homofeneous refraetoiy section
to one containing special insulation sandwiched between refraetoiy and earing. This
used to be a ecmmon type for small steam generating unite. As the furnace ride of
the refractory will ord^rily be ineandeeeent, it is important to prbvide a section
wlueh will be aatiafaetorily tuflit against beat leaks. Moat designers will either qieeify
the wall eeetion or cheek the manufacturer's figures en temperature gradient. Prririems
of this nature afford a dwnee for cfireet aam^ by theory and eemputaticn.
 316 STEAM GENERATORS
2. Aif'-cooled Masonry Walls.—^An air-cooled wall consists of a thin refractoty
section backed by an air space through which circulate cooling air currents. The air
is positively circulated bj’^ the furnace draft since it is desirable to use the heated air
for combustion. Obviously, an excellent feature of this construction is that the heat
flow which cools the refractories is again returned to the furnace. The cool wall front¬
ing the air sjiace is the furnace casing. The refractory section must be independently
supported by a steel or cast-iron skeleton from which the bricks (which are of special
shapes) are hung. Many ingenious methods of doing this have appeared, and a number
of successful typos are on the market. They are patented, and in the main this is the
field of the proprietary wall.
3. Partially Water-cooled Walls.—This type of wall is adaptable to a wide variety
of firing conditions. It is usually a solid type with a portion of the surface covered by
water tubes. Proper balance between the water-cooled and refractory sections will
allow high-capacity firing of fuel with low excess air by keeping the furnace temperature
sufficiently under the ash melting temperature. The refractory sections act as a
stabilizer to temperature, aiding in ignition and hampering overcooling. Partially
water-cooled walls are in common use for both stoker and burner-fired boilers. The
inner Burfaoe of refractory tile may be 50.8 to 101.6 mm thick, with 101.6 mm to 854 mm
heat insulation in block form backing it up. Such a wall could be used back of the water
tubes. Bare walls would require a thicker brick section.
4. Water-jacketed Furnace.—The combustion region is completely surrounded by
heat absorbing surface. As a type, the water-jacketed furnace is suitable for pulverized
coal firing, employing highly preheated air, and serving a fairly steady load.

Materials which resist change of shape, weight, or physical properties at

high temperatures arc known as refractories. Materials chiefly used are fire¬
clay, silica, kaolin, diaspore, alumina, and certain products of the electric
furnace. These are mostly used in the form of bricks. A fire brick is built pri¬
marily to withstand high temperatures and secondarily to have heat insulating
value. Uhfortunately, since high refractory quality is incompatible with low
conductivity, a w^all to withstand extremely high furnace-side temperatures is
built of more than one material. High-tcraperature refractories are built into
the furnace side to sufficient thickness that their cool side temperature is within
the maximum temperature limit of less refractory but more highly insulating
wall materials.
Fire-clay bricks are preferred because of their low cost whenever they will
give satisfactory service. Fire-clay brick are classified for temperature duty on
the basis of the pyrometric cone scale, high-temperature brick being those
possessing a pyrometric cone rating equivalent to about 1677®C, and low-duty
must show refractoriness at not less than 1580®C. Extremely high-temperature
service sometimes requires a material more refractory than fire-clay. Some
of these ^^super-refractories” are suitable for temperatures 166®C higher than
for the best fire-clay.
In addition to resisting fusion, furnace refractories should not spall under
rapid fluctuations of temperature, should resist erosion from ash-laden gases
and the fluxing action of slag, and hold up well in structural strength as its
temperature approaches the fusion temperature. The standard size of fire and
insulating brick is 228.6 mm x 114.3 mm x 63.5 mm.
Because of variations of temperature and the material comprising the
furnace walls, the expansion of materials in the walls continually tends to open
FURNACE 817
up cracks in the brickwork. Ezperioieed bricklayers have various methods of
bonding thor brickwork together so that a mmimum of air infiltration will
seep thi^gh the furnace walls.
Flow of heat through a wall of specified composilion may be calculated
frmn the principles of heat transfer. Tl^ is done for the purpose of determining
temperatures at critical points in tiie wall (as where the material changes in
a multi-layer wall), for an estimate of the outside surface temperature and
for calculation of heat loss through specific sections of a steam generator setting.
However, the frequently irregular shape of the whole setting exterior is such
as to discourage heat loss computations on a wall-by-wall basis. Instead the
over-all radiation loss established by numerous heat balance tests can be
studied and generalised. The radiation chart of Fig. 10-41 is often used. There
one notes that radiation loss, as a percentage of heat input, is expressed as a
function of load and wall cooling.
Heat flows through a wall by conduction and is discharged from the outer
surface by radiation and convection, mainly radiation. While the radiation
effect* is readily evaluated by the Stefan-Boltsmann law, direct analysis of
convection is elusive because it is affected by ambient air density, condition
of the wall, surface position (whether horisontal or vertical), and proximity of
obstructions to naturally rising, heated air currents. The effect of convection
can be included in a rou^ way by applying a convection factor as a multiplier
of the radiation. Also there have b^ published charts for detem^ing the
convection losses.
Temperatures occurring in this instance of beat transfer are shown in Fig.
10-14. The temperature of the furnace atmosphere is designated tf The hot
face of the wall is h; the cold face, and the ambi¬
ent ur atmosphere is f.. The intermediate face con¬
tacts of multiple layers are considered to be f', f',
etc.
The term “furnace temperature" has been used
loosely in power plant nomenclature to mean several
different temperatures. Probably it cannot be exactly
defined because of the dynamic energy conditions to
be found in the ordinaiy furnace. The temperature of
the flame will be one thing, the temperature of the
refractories another, and the temperature of the
products of combustion another. The only tempera¬ Fra. 10-14 Heat trans¬
ture which one may predict with some degree of fer through multiide layer
certainty is that of the gases entering the convection flat wall.
Bone. We know the flame temperature to be much
higher, and we suspect that there may be a considerable difference be¬
tween refractory temperatiu% and gas temperature. Ordinarily, optical ther¬
mometers used with furnaces give readings of wall temperatures, since the
flames—^der good combustion conditions—are fairly transparent to the radia¬
tion from the walls.
The temperature ti is often assumed equal to tt for bare refractory walls,
but is much, lower when walls are shielded with water wall tubes. The ambient
air temperature is not taken close to the walls where there are rising air cur-
 S28 STEAM GENERATORS
rents, but outside this disturbed region. The cold face temperature to is naturally
adjusted by heat transfer so that the rate of heat discharge from the outside wall
equals the rate of conduction through the wall. That is, Qr = qo where:

Qr = CFMTo^ - Ta") kcal per hr (10-2)

A (ti - U
qo = kcal per hr (10-3)
I ^ I

kl h kn

in which the Fs are as described in Fig. 10-14, and

A = Wall area normal to direction of heat flow, m*.

C = Coefficient of radiation; dependent on outside surface.

Plain brick wall, 464.1 X 10*"^®
Asbestos mill board, 455.7 X 10“^®
Trowel-finished insulating cement, 427.2 X 10“^®
Aluminum paint,313.8 X 10~^®
Black paint, glossy, 388.7X 10“^®
Black paint, rough, 467.1 X 10~i®

Fe == Convection factor. Average value about 1.5.

d *= Thickness of layers of homogeneous composition (m).

k = Heat conductivity of materials in homogeneous layer.*

The heat transferred through a wall can only be as much as that which is
discharged from the cool surface. Since the thermal conductivities of wall
materials vary somewhat with their mean temperature, the analysis of thermal
conditions in a furnace wall generally becomes a trial-and-error problem.
Often the temperature to must be repeatedly assumed imtil the calculated
Qr equals the calculated Qc,.

Example 2: As an illustration of the computation of heat transfer through a

specified furnace wall section, it will be assumed that of Fig. 10-14 is 4 in. of
refractory tile with thermal properties of average fire-clay, that do is 5 in. of block
insulation composed of mineral fiber. The outside surface is sheet-metal cased and
painted aluminum. Because of water wall shielding, hot face temperature is only
1500® F. Ambient air is 80® F. With these data the outside surface temperature t^, will
be determined; also the face temperature t\ An area A of 1 sq ft is assumed.
The sheet-metal casing is thin, with a k value so large that its d/k is negligible.
As a preliminary estimate, ki is assumed 8.5; kz, .55. Then equating = Qq

1500 - to
q = 11.0 X 10“>« X 1.5(To^ - 54(P)
(4/8.5) 4- (5/0.55)
When is assumed = 200®, Qr = 175, Qc = 136 Btu per hr.
This having revealed qR > Qq, the assumption of outside wall temperature needs
decreasing.
Upon assuming = 175, Qr = 127, qc = 139.
As the two q's are now in approximate balance, it will be well to re-estimate the
*See Figs. 10-15 and 10-16 for typical materials.
FURNACE 319
conductivity values before another trial. For this purpose assume g = 138 Btu per hr.
J5QQ _
Considering Eq 10-2 applied to the tile layer alone, 138 = —whence t' = 1435°.

The mean temperature of the tile layer is (1500 -f 1435)/2 = 1467° F. Neglecting
contact resistance, the mean temperature of the block insulation is (1435 + 175)/2 =
805° F. For these temperatures, ki = 9.0 and k2 = 0.64. As these new k's will raise
the rate of heat flow, an assumption of = 190 is made, in the view of the previous

Mean temperature of inftulation>Deg C

Service limit Muterinl

Hot face,
1371 Molded calcined lirick of diatomaceous sil¬
ica.
Molded calcined brick of diafomaceou.s sil¬
ica.
Natural diatomaceou.s silica brick (Cellite).
Plastic adhesive insulation of mineral wool
base.
Molded block of diatomaceous silica bonded
with asbestos fiber.
Molded block of magnesium carbonate and
binder. (859^ magnesia.)
Molded block of mineral fiber.
Mineral wool sectional blanket insulation.

Fig. 10-15 Conductivity of heat insulation often used in boiler settings.

1. Silicon carbide. Upper limit,

lUVQ.
2. Kaolin (abOsSiOs). Upper limit,
-^1538^1.
3. Fused silica brick. Upper limit,
1371*^.
4. Red brick, hard burned. Upper
limit, 871*^,
6. Average fireclay. Upper limit,
1427®C-
6. Red brick, soft. Upper limit,
7«0“C.
7. Castable refractory concrete (cal¬
cined refractory material.) Upper
300 500 700 900 1100 1300 1500 i
1650 limit, 1427^ k
Meon l•mparatur» of rofroctory-Oeg C 8. insulating firebrick (porous fireclay
refractory). Upper limit, 1427H?
Fig. 10-16 Conductivity of refractory materials
320 STEAM GENERATORS
calculations. Now = 153.5, Qq = 158. Although these are not exactly balanced,
further arithmetical adjustment is not justified by the nature of the underlying data.
A heat flow of 150 Btu per hr and an outside face temperature of 190 are probably
within 10% of the real quantities, if the wall were actually built and its hot face
temperature were 1500® F.
A recheck of t' shows that it remains about the same as previously calculated. Snce
the block insulation specified is capable of standing 1800® F, its hot face temperature is
sufliciently low. The casing temperature of 190® F is rather high for operator comfort.
It could be decreased by increasing the thickness d2, or by adding a layer of blanket
insulation, i.e., dg at A^g.
10-9 Superheat. Superheated ^am is vapor whose temperature has been
increased above that of its boiling point. To accomplish this the steam produced
in a boiler is led through a superheater, consisting of tubular heating surface
exposed to the gases pf combustion.
Superheating is A means for raising the thermal level of water higher than
is possible by the use of steam pressure alone. In view of the limitation of
saturation temperature of water<i.e.300.2*'C*^^ kg/oni> ab),it is obvious that the
currently used temperatures of 427-538®C cannot be achieved without
superheating.
•"The thermodynamic advantages of high temperature were implied by the
text of Chapter 8, especially where studies of energy, made available by work-

Wld« vorloty of standcn^ shcq^if

•ofos furnoce bvIMina problomt

B
Spaciol shapas ora doslpnod ond
ordorod of noodod from bvildor
Into forme tho somo oe concroto

notfic rofroctory moy bo poond- aofroctorjr mortor bonde Mclit#

•d Into flhopo In final location «nay bo otod for coating corfaco

Fio. 10-17 Refractories (Courtesy Power.)

SUPERHEAT 321
ing expansions, were considered. The improved performance of all steam prime
movers obtained by using initially superheated steam is almost axiomatic, so
frequently has it tn^n cited. For example, the steam rates of engines and tur¬
bines are improved by superheating as indicated in Table 10-2.
The rate of heat transfer through superheating surface has already been
shown to be dependent on steam and gas flow rates. The steam velocity is
dependent on what pressure decrease can be allowed between the inlet and
outlet headers. More pressure difference is acceptable on high-pressure than
low-pressure boilers. Although the high steam velocity is beneficial to heat

Table 10-2. RANGE OF STEAM RATES--KO PER HP-HR

Saturated 55 6® iir
Type of Prime Mover Steam Superheat Superheat

Noncondensing engine, throttle governed. 19.5-20.2 9.0-17.U 8.1.15.7

Noncondensing engine, automatic. 12.1-18.1 8.1-15.7 7.2-13.6
Simple Corliss engine, noncondensing. 11.2-15.7 8.1-13-5 8.7-11.2
Compound Corliss engine, noncondensing. 9.0-12.8 8.7-11.2 5.8-9.9
Compound engine, condensing. 5.4-9.0 4.5-9.0 4.0-7.8
Noncondensing turbine. llJl-22.5 9.0-20.2 8.1-18.0
Condensing turbine, small. 2.0-14.8 8.7-13.5 5.8-11.2
Condensing turbine, large. 4.5-8.1 4.0.8.7 3.14-5.4

transfer, the gas film has more effect on the over-all coefficient, as a previous
example has demonstrated. Also high mass flow, Gg, tends to reduce number of
superheater elements in parallel and to require greater length per element. The
effect of this is more folds per element between the two headers and greater
draft loss. For example,where 7 kg/cm*per 30.5m of element might be allowable
in a high-pressure unit, a low-pressure design might not be desirable with over
0.7 kg/cm* per 30.5 m. Flow velocities are of the magnitude of 1219-3048
m/min, with heat transfer coefficients of 24.4-73.2 kcal/m* hr deg C.
Tubes of 25.4 to 50.8 mm diameter arc generally used, although formerly
many superheaters were built of larger diameter tubing with external fins to in¬
crease heating surface.
Although separately fired superheaters have made an appearance in marine
units, the superheater is enclosed in the boiler setting in stationary practice.
This creates the need to design the boiler itself so that its setting will provide
sufficient room for the superheater and a means of supporting the elements.
There is little difficulty in arranging standard boilers to receive a superheater
for 37.8®to 93.3®C superheat, butthe superheater spaceneedsfor high temperature
units have considerable effect on boiler and setting arrangement.
Example 1: A convection superheater is composed of 12 elements in parallel, each
consisting of 1% in. X 0.105 in. tubing, 64 ft long. Gas temperature = 1850* F. Enter¬
ing steam has 0.985 quality at 250 psig. Superheated steam has 600® F at 235 psig.
The superheater transfer coefficient U will be estimated, assuming evaporation of 9.55*
lb per lb fuel and 18 lb gas per lb fuel. Steam flow 15,000 lb per hr.
This will be considered as a case of counterflow heat transfer, explained in Chapter 9.
The equation U =; q/AB will be used. To find tf, the gas temperature drop over the
 322 STEAM GENERATORS
superheater must be found. Assume Cp of gases = .25 Btu per Ib-deg. Equating heat
release of gases to heat absorption of ste.tm;

18 X 0.25(1850 - <'„) = 9.55(1318.5 - 1189.4)

t'o = 1576" F; = 1576 - 406 = 1170“ F; = 1850 - 600 = 1250” F

e= - «™.)/ln^; e= = 1220“ F

q = 15,000(1318.5 - 1189.4) = 1,938,000 Btu pei hr

Area A = 12 X 64 X (area per lineal ft at 1J in. diameter)
A = 12 X 64 X 0.338 = 260 sq ft
V = 1,938.000/(260 X 1220) = 6.1,Tltu per hr-ftMeg F

Also: Steam mass flow G, = Vo = 38 lb per sec-ft*

X u.uuyi
Steam velocity == On X = 78.6 fps for the average specific volume, which is
2.07 cu ft per lb.
The location of superheating surface creates the classification of convection,
radiant, and interdeck types. Convection superheaters are those which receive
heat transfer entirely by convection repre¬
sented by the relation q = UA6, Radiant
superheaters are located on furnace walls
where heat is implanted on them by direct
radiation from luminous products of combus¬
tion. Interdcck superheaters are primarily
convective but also receive some radiant heat
as they are screened from the furnace by only
a few rows of tubes. Headers are located so
that header-tube joints are removed from di¬
rect heating action, usually by placing them
outside the setting. As convenient header loca¬
tions are found near the top of the setting,
Fig. 10-18 Interdeck location of superheater tubes are not ordinarily drainable,
superheater in a sectional-header However, the drainable feature is not impor-
tant for superheaters expected to be in con¬
tinuous operation. Radiant superheaters are
not employed so frequently as the convection type and not oTlen outside the
Antral station field. When installed they resemble water walls in appearance
aiiSTare connected between upper and lower headers, without looping. CqW-
■ '!^gfilion and interdeck superheaters are multiple loop to gain length sufficient
tt>*iransfer the requisite heat.
Th£4KX-JiqjaiCtrc £Q$tpfsuperh^t surface is higher than boiler surface; also,
the cost per kcal transferred. Superheated vapor does not wet the tube surface,
so that the interior film convection coefficient is relatively small. Tube walls
therefore operate at higher average tejnperatures than boiler tubes and are
required to be of alloy steel --
The heat transfer conditions vary with load. Assuming that air-fuel ratios
and thermal efficiency remain relatively constant, then, as load is decreased,
SUPERHEAT 323
fras mass flow decreases proportionately. But filrp coefficients vary as some
power of G less than 1; hence rate of heat transfer decreases more rapidly than
steam flow and fewer degrees superheat are
produced by the convection type. On the other _
hand, air-fuel ratios remaining constant, the
furnace temperature does not vary propor- fs f\(
tionately wdth load—it falls more slowly. Ra-
diant heat transfer follows roughly the fourth I
power law and therefore decreases more *
slowly than the steam flow. As a result, the I *
steam receives more superheat than at higher | \
loads. Typically, then, the all-radiant super- 1 i
heater has a falling-\Mth-load temjierature 9|\||yy
characteristic; the convection, a rising-with- ji V
load; and the interdeck is likely to show both || ^ \^J
trends at different load ranges. W
The typical rising temperature character- 9^ J | \
istic of the convection superheater is accepted _
without correction unless the temperatures are
quite high and close control is required for Schematic diagram
the safety and efficiency of the steam using of Iwat'on of both convection
^uipn^t supp^. i. de,ig„^ to " *
take steam at 510®C. This turbine should
not be supplied with steam hotter than510®C for such designs are critical to
small excess of temperature. Hence, a choice is faced of either uncontrolled
324 STEAM GENERATORS
superheat with maximum temperature at 510*C and temperature deficiency at
part loads, or acceptance of a surplus superheating capacity cancelled by de¬
superheat or other form of control. This is illustrated in Fig. 10-22.

Fio. 10-21 Performance of combination Fig. 10-22 Superheat characteristics of

superheater. (Courtesy Foster Wheeler conventional convection superheater.
Corp.)

The conventional convection superheater with no control can produce rated

superheat only at full load. At other loads the steam temperature is less and
the prime mover efficiency suffers correspondingly. With temperature control
the superheater is made sufficiently large to produce the 510®C at the lower end
of the load range over which control is to be exercised. This is assumed to be

GAS BY-PASS twin FURNACE

Fig. 10-23 Superheat control without desuperheat.

SUPERHEAT 325
60% load in Fig. 10-22. At higher loads the surplus temperature is generated,
then cancelled by controlled desuperheat, or the rate of heat transfer through
the superheater is controlled by methods which will receive brief mention later.
Radiant and convection superheaters in series connection have opposing tem¬
perature characteristics that tend to cancel the variation of final superheat.
When used in series the final superheat is produced in the convection zone.
There are many methods in use for controlling final steam temperature. As
these are frequently rather complicated, they will merely be summarized here
and only a few illustrated.
1. Follow the superheater with a water spray desuperheater operated by a tem¬
perature regulator. This method, once widely used, is not greatly favored at present on
large units.
2. Gas by-passing. This method is employed for regulating convection superheaters.
It is necessary to employ an oversized superheater, one which will
give the required superheat at the lowest specified load. At higher
loads the by-pass is opened, allowing some of the gas to flow
around rather than over the superheater. Although the gas passing
the superheater is hotter, its volume is less, hence the compensa¬
tion. ^metimes used with spray desuperheater for fine control.
3. Combine convection and radiant types in series as has
already been mentioned, or place an interdeck type so as to
receive the proper proportions of radiant and convection energy.
Neither of these methods is adequate to maintain uniform tem¬
perature over a load range, but the variation will be substantially
less than for either type alone. May be combined with gas by¬
passing.
4. Other methods emplo3dng a combination of radiant and
convection superheaters with twin furnaces which may be dif¬
ferentially fired are said to have given excellent results over a
wide range of superheat control.
5. By-pass a varying portion of the steam, at an intermediate
point in the superheater, through a tubular desuperheater cooled
by boiler water or steam.
6. Control moisture quality of steam entering superheater by
condenser control using boiler feedwater as the heat absorber.
The superheater inlet header contains circulating tubes for the
feedwater. Fig. 10-24
7. Furnace exit temperature adjustment. The decreasing con¬ Spray type de¬
vection coefficient could be offset by higher mean temperature superheater with
difference, thereby holding degree of superheat more nearly con¬ temiioriiture ele¬
stant. In large, water-walled pulverized coal furnaces having ment and spray
tangential burners, the burners are sometimes designed to be regulator. (Cour¬
adjustable so as to tilt the flame in the vertical plane. A downward tesy Northern
flame tilt increases the heat absorption by water walls and de¬ Equipment Divi^
creases furnace exit temperature. Conversely, at lower loads an sion Continental
up-tilt of the flame reduces furnace heat absorption, increases Foundry & Mach,
temperature of gases leaving furnace, and thereby furnishes the Co.
needed extra temperature difference at the superheater.
Example 2: The superheat characteristic of a steam generator is represented by
t =:Z18J) -f- L/Sl^where t is superheat temperature, * 0, and L is the steam flow, kg
per hr.SteampressuresstSM kg/em^ Spray desuperheating can be used to maintain
326 STEAM GENERATORS
constant tcmi)tTaturc over a range of flow. The desuperheat action required to main¬
tain constant temi)erature between 6804 and 11840 kg, per hr flow will now be inves¬
tigated. At 6804 kg flow, t = 218.3 4* 6804/81.66 = 801.7^. This is the temperature at
the lower end of the proposed flow range; hence it is the temperature that can be
maintained by desuperheating.
Assuming spray water available at 83.3* C, the quantity required for a desuper¬
heated flow of 11340 kg per hr is determined by successive approximations. Let 11840
—■ L be the water sprayed in the desuperheater. The temperature of the L kg leaving
the superheater is somewhat less than given by218.8+ 11840/8L65, say 348® C. At this
temperature /i = 3117 J parg . The mixture leaving the desuperheater at801.7“C has
h = 30i21 J per g The desuperheater has a heat balance as follows:
3117xIO*L+ (11340 — L)391X10»=3021x10«x 11340
L ~ 10648 kg per hr
At this flow, t =218.3+ 10943/8166 =362.82*0. A recalculation using this tempera¬
ture gives L = 10886 ; hence spray water = 464 kg per hr

10-10 Boiler Accessories and Trim. The steam boiler consists basically
of heating surface and shell or drums. These are enclosed in a setting. Then
the ^^trim’’ is added, this consisting of valves and piping essential to the safe
operation of the boiler. These are supplied with the boiler when a comiilete or
*^tum key’^ installation is purchased. On the other hand, the purchaser fre¬
quently buys a bare boiler, then assembles the setting and trim himself. The
minimum trim is usually the following:
1. Feedwater entrance. A stop valve and a ch^ck vnlvo, with the former installed
nearest the boiler.
2. Blow-off (water), including special valves not vulne’’ahle to dirt in the water.
3. Safety valves (steam) to prevent rise of steam prehsurc above a predetermined
limit.
4. Water column, with gauge gla.ss, high- and low-water alarms, etc. The primary
purpose of a water column is to inform the boiler operator where the water level is in
his boiler. As this level should be maintained within a few mm , or les^’, of a normal
water line, the display of water level need only cover a heightot 3ooto 460mm., centered
about this normal line.
5. Stop-and-clieck valve. This is mounted on the main steam outlet and to it is
connected the steam pipe line.
6. Steam vent. A small outlet on the top drum normally closed by a valve, which
may be opened when it is desired to hold atmospheric pressure in the boiler, as during
initial boiling out.
7. Steam pressure gauge with siphon, cock, and inspector’s test cock.
8. Various special openings, in accordance with purchaser’s requirements, for the
accommodation of feedwater control elements, reagent feed, water sampling, etc.

Boiler accessories are low-water safeguards, soot blower, steam purifier,

economizer. Low water level is, of course, an invitation to disaster. It is not
as hazardous as overpressure in that, although low water can be detrimental to
the boiler, it is not so liable to cause an explosion that will wreck the premises,
at least for water tube types. But it can cause partial or complete ruination of
the boiler by overheating; hence alarms and automatic safeguards are fre¬
quently provided by cautious owners. Some of these arc: (1) the low-high water
alarm whistles usually placed in water columns; (21 float-operated, low-water
BOILER ACCESSORIES AND TRIM 327

Vtiieea
Rotating and retracting dement eoat blower

Fig. 10-25 Some boiler auxiliaries.

328 STEAM OENERATOBS
cutoffs for stopping electric motors driving fuel feeders; (3) fusible plugs of
low melting temperature alloy placed in the boiler shell l^low the minimum
safe water line. In spite of well-engineered safeguards, one reads occasional
reports of boilers being ruinously overheated because of failure of preventative
maintenance of automatic safety equipment or through operator indifference
and carelessness.
Soot Blowers. The heating surfaces of boilers, especially coal-fired, water-
tube boilers, have a tendency to become coated with debris of combustion, i.e.,
soot, cinder, fly ash. This loosely adhering layer is a definite impediment to the
transfer of heat and should be removed after it has built up sufiiciently to result
in a significant increase in flue gas temperature. The conventional method is
to blast the deposits from the heating surface with jets of steam from a hand
lance inserted through setting openings or by one or more permanently mounted
‘‘soot blowers.” The usual form for the smaller boilers is a dead-end tube pro¬
jecting through the setting and extending into the tube bank. A number of
small lateral nozzles are provided and the external head is arranged so that
high-pressure steam can be admitted to the tube and at the same time the tube
rotated around its axis. Externally then the device incorporates a turning
mechanism (manual or motorized), a valve, and the packing gland necessary
between fixed steam supply and rotating tube. Steam jets clean soot from heat¬
ing surface in their vicinity, but the action is limited in extent in a close array
such as a tube bank; hence several elements may be needed to clean a boiler
adequately. It is not uncommon to see 8 to 16 units in a large boiler. These
would be tedious to operate manually, and, since they must not all be operated
simultaneously,* the operating heads are electric-motor-powered and operated
by remote control on an automatic sequential basis. On some boilers the
automatic draft system controls have to be positioned for abnormally high
draft before beginning the soot blowing operation; otherwise the release of
steam into the setting may blow soot and smoke into the boiler room.
When not in use, the soot blower tube of the system described would be
dry, uncooled, and subject to gas temperatures. Even with alloy tubes the
deterioration is relatively rapid on units normally experiencing high gas tem¬
peratures. Retractable elements, though more expensive initially, have the
advantage of protection from high temperature when not in use. Also, using
only two traveling nozzles, the jet size can be larger without excessive steam
drain, and the projected blast distance greater. The eroding action of blower
jets on boiler tubes, sometimes seen with the fixed rotary type, is said to be
much less with the retractable type. Principal disadvantages are (1) clearances
required around outside of setting and (2) more difficult steam packing job
(travel as well as rotation). In spite of the cost of compressed air, some re-
tractables have been operated on air to reduce boiler water make-up, and
blower packing gland maintenance.
Soot blowers should be operated frequently enough to keep the tubes clean.
Instead of a once-a-shift or twice-a-shift rule-of-thumb routine, the operating
engineer should observe the flue gas temperature drop, estimate the quantity of
steam per blow, and calculate the frequency on an economic basis. Obviously if
’’‘This would place too heavy a steam drain on boiler; also, draft ss^ems are unable
to cope with extra volume of wet flue gas.
BOILER ACCE880R1EB AMD TRIM 829
the temperature drops 24*C after blowing, the interval might well be decreased,
whereas if only—12X1 were observed, it could represent a net loss becauite the
value of the steam used was more than that of the increment of flue gas heat.
Steam Purifiera. Generated steam might be unsuitable for some tues by
virtue of its quality or its lack of purity. Quality is the measure of amount ot
entrained moisture carried by the steam, whereas purity is the absence of solid
or dissolved matter in the wet steam. Impurity is measured as parts per'million
(ppm) of foreign matter in steam, by weight. It is present in steam mainly
because the generated steam entrained droplets of boiler water containing solid
or dissolved matter. After passing throu^ a superheater the impurity is al¬
together solid dust particles. If the steam is to be used in saturated state for
heating or process, its quality is not critical. Also, in plants of this type the rate
of boiler blowdown to steam generation can be, and usually is, maintained
sufiSciently high that the concentrations of dissolved matter in boiler water
remain fairly low and wet steam is not therefore highly charged with ppm of
solids. However, the use of high-pressure, highly superheated steam in modem
power plants does require very pure steam. The carryover of solid matter
becomes a source of operating di£Sculties in superheater and tmbine.
Typically in boilers without special purifying apparatus, 98.5% quality is
guaranteed, whereas for power service 99.5% has b^n considered necessary.
But in the latest power stations operating on boiler water concentrations like
2000 ppm, the steam ought not to contain over Buch water if the
impurity content of the steam flow is to be limit^ to 1 ppm. This is practically
impossible to secure unless steam purifiers are inserted in boiler drums. The
obtaining of clean steam from boiling water can be considered in three steps.
1. Primary separation of the steam from the circulating water—at the dis¬
engagement surface.
2. lYashing of the steam with fresh and relatively clean feedwater. After
washing, the quality may be approximately the same, but the ppm are low
because feedwater ppm are but a small fraction of boiler water concentrations.
3. Removal of excess moisture in a steam “drier” by multiple bafiBes, screens,
etc.
All boilers must have primary separation. As has been mentioned, this is a
function of the drum size and tube arrangement and is sufiicient for most genera¬
tion where saturated steam is the final product. Power boilers have driers, even
if it is no more than a simple dry pipe. The steam washer is increasingly re¬
quired on today’s large power boilers. 'iVashers and driers are required to be
built into the steam drum above the water level.
Economizers. The boiler economizer is a feedwater heater deriving heat
from gaseous products of combustion discharged from the boiler. It always
consists of tubular heating surface, frequently arranged as shown in Fig. 10-26,
but sometimes being inside a bent tube boiler casing and bring similar in ap¬
pearance to one of ^e rear banks of tubes. The significant difference between
economizer and evaporation surface is that there is no steam generation in the
economizer, only water heating; also, there is no recirculation path.
Economizer tubes are steel—smooth tubes in some designs, fin-covered in
others. Location is dependent on gas flow design of the boiler, external auxil¬
iaries, and available boiler room space. The degign of the economiser is o(m-
330 STEAM GENERATORS
cemed with providing sufficient heating surface to reduce the gas temperature
to the required degree. The problem is primarily convective heat transfer with
secondary problems of gas-side cleaning (soot blowers), accessibility, interior
corrosion,* exterior corrosion due to vapor condensation out of the low tem¬
perature gas at light loads, and draft loss from friction between gas and tubes.
The equations of heat transfer by convection between the fluids, both of
which vary in temperature, have been developed in Chapter 9. The typical
empirical equation for over-all conductance (9-14) has often been implemented
with constants derived from plots of test data. For example, Maynz gave, for

• I—.. '
-II "
-X

K-VWW
- c
1*1 I.'

1^'iG. 10-26 Sch ,ic economizer

A = sq m water heating surface
Aq =: sq m free area for gas flow at any horizontal section such as XX.
Tube bank x tubes wide by y tubes deep.

smooth tube economizers, A =1.22’^nd B = 0.00075. Kreisinger reported A =

10.74+ B = 0.(XX)95 for longitudinally finned tubes having two diametrically
opposite fins. The extra heating effect of circumferentially finned tubes is
so dependent on fin spacing and depth that no general values of A and B
are quoted.
The relationship between heating surface and temperatures shown by Eq
9-17 is capable of resolving the conditions of economizer performance into re¬
quired heating surface, provided that z were determinable either from estimates
of the temperatures involved or from empirical data, such as the following
(for plain tubes):
W
2 = 4.45-^ - 0.10 (10-4)
IF K
TFw
where = Ratio of water to gas flow, by weight.
Wg

z = Temperature ratio (t'l — t'o)/(to — U).

It has been seen that high gas velocity, represented by large mass flow,
O, is an aid to heat transfer since the principal resistance is in the gas film.

♦Usually solved by adequate feedwater deaeration (Chapter 13).

♦# ^
koal perm* hrdeg G
1 koal per kg». deg q
BOILER ACCESSORIES AND TRIM 331
However, gas velocity produces fluid friction which is manifested externally as
a draft loss* in the economizer, ^he ordinary measurement of draft loss is by
“ mm of water,referring, of course, to the displacement in a U-tube manom¬
eter. Draft losses in economizers rise sharply with gas velocity, approximately
as the square of G; therefore there is a practical limit to increase of the con¬
ductance by employment of high mass flow. If it were attempted to reduce
economizer size and cost by designing for high G, the draft loss through it,
as well as the cost of providing the draft, might become uneconomically large.
It appears that the heat transfer conditions selected for an economizer should
involve some attempt at economic balance among the several factors involved.
Although the introduction of these economical factors is not practicable here,
the economizer problem may be illustrated if certain assumptions replace the
involved economic analysis. Thus we find that an assumption of mass flow
simplifies the situation so that an illustrative example can be shown.
Example 1: Assuming that it is desired to operate a boiler with high gas exit tem¬
perature, say 899° C, the operating conditions of an economizer designed to reduce the
final temperature to 274° c will be investigated. The following data are to be used (see
Fig. 10-26 for nomenclature).
e'l = 399^0; 274° C; it = 99° C; Cg = 0.25, (V = l.Okoal per kg-degc;
G * 24267 kg per hr- m*
PTg/Tl^. = 2.0.
By heat balance: 0.25>Fg(899 — 274) = l.OTTwCfo — 99).
Since TTg/lFw is given as 2.0; fo = 161® C.
^max = 399 - 161 = 238*^ C ; = 274 - 99 = 175® C; ^nmx/^min == 1.36.
z = At'/At = (899 •“ 274)/(i6l - 99 ) = 2.0.
U = 1.22-1- 0.00075 X 24 267= 4.00 if plain tubes.
For an assumed 18144 kg steam per hr boiler the required economizer heating surface
by Ec] i;-17 :
18144 In 1.36
= 276 m*
(2.0 - 1)(4.00)
(With finned tubes having 1/ = 6.95; A = 164 m*)
Example 2: The arrangement of the heating surface for the economizer of Ex 1 is
]>rincipally the determination of a-, y, C, and D of Fig. 10-‘26. Assume60.8 mm X 3 mm
tubes on lOO mm centers horizontally.
Water velocity through the tubes is relatively unimportant to heat transfer for the
principal resistance is in the gas film. Low water velocities make for many more tubes
in parallel carrying the flow and promote a more practicable economizer shape to which
to attach the gas conduits. C and D must be chosen with regard to the assumed mass
flow G, while y involves additionally the heating surface, A.

Free area for gas flow: CD — x ^06 = W^/G *= X

Approximately, C = 0.1 (a: -f 1) or x =10(7 — 1.

♦ Draft is fully treated in Chapter 12.

 332 STEAM GENERATORS
Substituting WJW^ ^2,0 and assuming a shape of D •• ZC, x is then
calculated to 8 tubes. As the water velocity required to carry 18144 kg per hr
through eight 50«8 mmx 3mm tubes in parallel iso.acm/8eo^x«>i8{f apractic«al value CXD
is found to be 0.91 m X9.7in. To allow for the fact that tiie tube anangement is stag-
geredy C X D will be taken aai m X 3 m.

Heating surface per row « 8 X0.(MHr X 3 * 3*77 m*.

y - A/3.77 « 276/3.77 « 73 rows.

It is obvious, from the foregoing, why manufacturers frequently use finned econo¬
mizer tubes. The nominal heating surface required is decreased; consequently, the
number of rows of tubes, the size, and the cost of the economizer are far more favor¬
able. As was determined in Ex 1, A = 164 m> lor longitudinally finned tubes. B, D,
and X are unaffected, but y = 164 /3.77 = 42 rows.
Finally, it might be desirable to test for gas friction loss. An empirical equation for
draft loss on plain tubes (approximately correct for longitudinal fins, but not for
circumferential) is:
Ap » 5.65 X in. water (10-5)
For 0 «24267, Ap » 1.4 in HjO, a reasonable figure if forced draft is employed, not
80 for natural draft.

10-11 Combustion Equipment. At the outset of this chapter we men¬

tioned combustion equipment as a component of the steam generator. Since the
source of heat is the combustion of a fuel, a working unit must have whatever
equipment is necessary to receive the fuel and air, proportioned to each other
and to the boiler steam demand, mix, ignite, and perform any other special
combustion duties, such as distillation of volatile from coal prior to ignition.
Fhiid fuels are handled by burners; solid lump fuels, by stokers. In boiler plants
hand firing on grates is practically unheard of nowadays in new plants, al¬
though there are many small industrial and institutional plants still in service
with hand firing. The numbers are fewer each year, and hand firing will eventu¬
ally disappear.
Since so many different principles are used in combustion equipment, Table
10-2 has been prepared to summarize the more important currently manufac¬
tured tyi>es of stokers and burners. Several of these will be described and
illustrated.
Although the nature of fuels and the theory of combustion are treated ex¬
tensively in Chapter 5, a brief summary of the requirements of combustion
is pertinent to an understanding of the combustion equipment of a steam boiler.
The fuels are mainly bituminous coal, fuel oil, and natural gas, mentioned in
order of importance. All are composed of hydrocarbons, and coal has, as well,
much fixed carbon and little sulfur. To bum these fuels to the desired end
products, CO2 and H2O, requires air in sufficient proportions, a good mixing of
fuel and air, and a turbulence or relative motion between fuel and xiir. The
combustion equipment must fulfill these requirements and, in addition, be
capable of close regulation of rate^ firing^ the _iueL_fdl_^(aler^^
qpgi^_gi[yar^ load. Coal-firing equipment must also have a means for
holding and discharging the ash residue.
Burners and stokers are mechanisms of great ingenuity and mechanical de^
tCOMBUSTION EQUIPMENT :m
Tahlo COMMON COMBUSTION EQUIPMENT FOB STKAM |{«>I1.KKS

Muhiplo jet*
( f.TH Fan mix
Pru-mix rMultiple* burner
PreflHur(‘ utomiziuK* 'i Heplaeeable tip

Fluid I
1 Wide nuiKe lip

luc'Is , Oil -< Steam atomizing* j OutsMe mix

(JiuriHTs) I Clip* 1 Inside ini\
; W iclv
\'*iponyer
I Hot ptole
( j Short flame
j l*ulv(‘rize<I coal’*'
{ Lon^ flame
, ( Tangential
' (Vushed and
fineb sized I IIi|i:h turbuleiKM* turnaee (evc-loiii )*
<‘0{d I )uinp nral<‘
f M(M*hanieal throw* !
Spread(‘r* I (Indertlunw 'rr.ivelmjj jjrale
stoker
Lump I Jet throw J Air jet
eoal JOv(‘rfeed •; [Steam jet
(Stokers) 1 Convex or* , Traveling grate J Natural draft
stoker j Chain grate Forced draft
.. . , , , 1 Single j Ram f(HHi I Orates stalionar\
Hon.<.Mtul •
I'nderfeed-^
Sloping r<‘torl* -Multiple retort, large* eapaeitv.

* OesrrilM'd in text.

tail. They arc reduced to successful forms by their maiiufacUirer> chit^fly

through experimental development of prototypes followed by minor job-by-tob
improvements or refinements. In surveying the ecpiipnK'nt currently availnlilc in
this field, the reader should not become engrossed in meebanictil detail^ to tin*
extent of overlooking the manner in which each fulfills the basic reciufrcnientN
necessary to any sueecssful installation, which arc:
1. Thorough mixing of fuel and air.
2. Optiiniiin fuel-air ratios leading to most complete combust ion pos>iblc
maintaiiu'd over full load range.
3. Ready and accurate response ot rate of fuel feed to Ioa<l di'inand < usually
as reflecteil in boiler steam pressure).*
4. Continuous and r<*liabh’ ignition of fuel,
Ti. Practical distillation «d’ volatile components of coal followed bx adi<tuate
action coxered in Items 1 and 4 above.
t). Ad(‘quale control over point of formation and aeeiimulaticni ol a-h, xvhen
coal is the fu(‘l.
Ntitural gas is used as a boiler fuel in gas well regions where the fuel is
relatively clienp eoinimratively distant. 'Flu' transport at ion
of natural iia*' overland to supjdy cities with domestic and industrial heat lias
made the* gas in tla* wa ll nmre valuable and the gas-fired stt am generator luore
difficult to justify in compaiison xvilh ctial, on fuel cost alone, (dt'anlincs.s and

Here tiu' jinth<tr implies eombiistinii (muiIuiI sx.slems riesigii<*il i<» ni,iint:iiti iie.nrly
eonst:int steam pressuri* iu the boiler, a enrnplex siib|(*<*l rel(*gati*ii In S(s* 12-16.
 334 STEAM GENERATORS
convenience in use are other criteria of selection, but more decisive in small
plants than in central power stations.
Transportation costs add less to the delivery price of oil than gas; also fuel
oil may be stored in tanks at a reasonable cost, whereas gas cannot. Hence al¬
though fuel oil is usually more costly than coal per kg of steam generated,
many operators select fuel oil burners rather than stokers because of the
simplicity and cleanliness of storing and transporting the fuel from borage tp
burner. Also, many have decided on installations that can be readily switched
from coal to oil and vice versa as a safeguard against interruptions arising from
shortages created by strikes, bad weather, etc.
Most of the steam generated in the United States is produced from coal fuel.
In small plants the horizontal retort underfeed stoker has been an old standby
but is a poor competitor, at present, to the spreader stoker. Pulverized coal is
successful in small plants, but costs favor stokers up to approximately 22680 kg
per hr rate of steam generation per boiler. Above 45360 kg per hr the majority
of new units are pulverized coal-fired. In the past the multiple-retort underfeed
stoker was widely employed f6r large boilers, but this author has not seen any

QOlDittillatioii EESiCefn butt Ion £SM3Ath

I
t Fig. 10-27 Comparison of combustion principles.

recent evidence that they are of importance in the new plant field. Spreaders
and pulverized coal seem to have taken over the large boiler field. Spreader
stokers are now receiving the greatest interest and sales efforts of any stoker
type. The spreader is also very popular with small boiler operators; hence it
may be said currently to be the leading type.
10-12 Spreader Stoker. The principle of spreader stoking is the sprin¬
kling of coal, evenly and thinly, over a grate which forms the floor of the furnace.
'Desirable spreader coal consists of a mixture of fine particles and small lumps
up to31.8 nun screen sue. Air is fed up through the grates frenn wind boxes
located below. The thin fuel bed offers but little resistanoe and the coal is
quickly cauumed. Only the lumps reach the grates, for the fine paitides bum
In suspension ri^ng on the column of ps and fiame riidng from the grates.
TViibnlenoe, frhieh is quite InIjiMHtant in this stoker, is seeured liy ow-flre
w or steam jets.
A^natural resulfbf suspenmon burning of small fuel particles is the entnun-
ment of ash in the products of cmnbustion. There is enougd^ of this from spreader
stdiers to create a nuisance if discharged from chimnejrs. Also many of tiie
gas-bmne particles contain carbon which titould be 'tnqiped and retained to
 SPREADER STOKER 835
the furnace for bura^. Jhus one sees a disadvantage of these stdkers—a dust
ocdQMor 18 almost a nSbessary auxiliary.
A spreader stoker consists of (1) Infers and feeders, (2) distributors, and
(3) grates. The hopper receives the supply of coal. Underneath the hopper is a
feeder to measure out the coal in accordance with need and drop it onto the
rotating distributor.* The distributor then impels the coal into the furnace.
Because the coal is a mixture of fine and coarse pieces it can be launched into
the furnace so as to descend fairly uniformly over the grates, with the heavier
pieces, of course, farthest from the disti^utdr. This (ould not be achieved if
the coal were all one size. Although these stokers are not critical to coal com¬
position (as are many), the author has found them to be fairly sensitive to coal

Detroit Stoker Co,

Fio. 10-28 Spreader stoker setting, with detail of feeder.

sizing and percentage size distribution. The range of transverse distribution

limits the width of furnace that can be served by a single distributor; hence
this stoker usually has two or more feeder-hopper-distributor units. It is well
to employ two units, even in a small boiler, on account of the method of re¬
moving ash. When, upon a few hours of operation, an accumulation of ash. is
Pllgsent on tKe~grates, (me feeder is shut down. After the section of grates in
front of it is buiped. clean of coal, that section^ ash is dumped. 'Dien fire is
raked onto the clean grates, from the adjacent working SMtiBO, ^dja new fire
quickly established. After that, the other grate sections can be cleaned one by
<me. The wind box under the grates is, of course, zoned so that draft can be
Removed from the section to be cleaned. As proporlioning of aur to fud is ae-
c(»nplished by vaiying the air plenum in uiis wind box, ach must not be allowed
to tUddy and inteispee a .temporary hig^ draft resistanoe. Large
i|!i«ader dnkeis nave continuous ash deaningl^ the use ondl(Zg • fadtiravi^
*]a the path of a jet for noameehaaiod dinrftuhifs.
336 STEAM GENERATORS
jng grates. The feeders are always set in the front wall, and the grates move to¬
ward the front waif for ash discharge. Such spreader stokers are less likely to
have uneven coal distribution on the grates with defective coal sizing than are
the dump grate stokers.
"" The coal feeder is the mechanical heart of the stoker. It is here that the
automatic combustion control is applied; also here the manufacturers differ
considerably in equipment detail. Feeders may be reciprocating rams, endless
belts, spiral worms, etc. They are customarily operated from the same line-
shaft as the distributors with the variable speed device controlling combustion
rate located between line shaft and feeder.
The mechanical type distributor is a rotor with vanes or paddles, angled so
as to throw some coal diagonally as well as straight ahead. The rotor is
lineshaft-driven, while the lineshaft itself is usually motor-driven. Motor-to-
lineshaft connection has a speed adjustment whereby the rotor speed may be
varied to obtain proper coal throw.
Spreader stokers have quick response to load change, being more like
burners in this respect than other stoker types because there is only a small
araoimt of fuel on the grates at any time. The power required is also compara¬
tively small, not over J hp per tonne per hr fired. Typically 5000 to 8750 kcal
furnace heat release and 675000 to 1350000* kcal/m* h grate surface is possible.
Disadvantages include need for distributed coal sizing, dust collection
previously noted, and low-load limitations.f
This is a stoker that can, in emergency, be hand-fired to at least 50% of its
rated capacity in small- and medium-sized units having dump grates. (Not
true of other stokers.)
10-13 Conveyor Stoker. This stoker consists of an endless grate which
continuously receives coal at its ^Tront” end and carries it into the furnace sup¬
ported on the links, or bars, composing the grate surface. On most types | the
fuel bed remains undisturbed on the grates, the ash being discharged over the
rear end.
Chain grate stokers have longitudinal links or bars so connected by trans¬
verse pins as to become a continuous flat chain of the width of the furnace. The
chain is supported at the front and rear by multiple sprockets and intermedi¬
ately by tracks or skid plates. The front sprocket drives, and the grate is in
effect shoved into the furnace.
Thegratestravel8lowiy,i.e.,iay300inm/iiiin,withspeed varied for combustion
control. For example, in a drive where a steadily reciprocating pawl engaged
a ratchet wheel on the sprocket shaft, combustion control could be introduced
by regulating whether the pawl picked up one, two, three, etc., ratchet teeth
per throw.
Traveling grate stokers are similar in action to the chain grate type, differ-
* Higher for continuous ash discharge..
t Reference is to combustion difficulties, say below 25% rating. There is a tendency to
smoke excessively; also, firemen are likely to lose the fire due to extreme thinness of fuel
bed and uneven fuel distribution.
I In the "compensating feed” t3rpe8 there are alternate traveling and stationary sections,
the effect of which is to disturb the fuel bed and to compact it as combustion proceeds.
The drag of the stationary sections causes the whole fuel bed to move more slowly toward
the rear of thestoker.
CONVEYOR STOKEH 337
ing only in method of supporting the grate. Strong transverse bars are carried
at their ends by continuous chains which travel around the supporting frame.
These support bars carry transverse cast-iron grates, but there is no connection
between adjacent grates save through the bars and chains which carry them.
Traveling stokers are set in the furnace with the working surface of the
grates horizontal or nearly so. At the feed end the stoker projects from the
furnace sufficiently to accommodate the loading hopper and drive mechanism.
Air is supplied to the region between the w’orking and returning grates.
(^ombustion on anv stoker ,jnay be^ con^dered to be composed of three
s^q.ges; vjz., distillation (coking)^ combusts
Ignition cannot be continued satisfactorily on the conveyor stoker solely by
heat conduction from lump to lump because the area for ignition is relatively
small due to the tl)in fuel bed. Rate of propagation of the ignition line is liable

to be slower than the rate of travel of the stoker, thereby tending to shunt the
imrjiing section of the fuel bed rearward on the stoker. Consequently, furnaces
for these stokers arc built with ignition arches so arranged as to reflect heat
from.the incandescent region to the surface of the incoming coal.
Good distillation of the volatile matter in the coal must be followed by
equall^Tgood combustion of that volatile. The ignition arch constrains the ^s
t5“^e hottest portion of the furnace w’here it will be burned if over-fire air is
adnntted.to this regign. This is generally done by letting the over-fire air in
liErougii openings in the arch, so disposed as to promote turbulence in the
burning volatile. The main part of the combustion air is supplied from below
thevrorkitig gratc.'To control this air supply effectively (i.e., vary the plenum
along the length of the stoker in accordance with the combustion needs) the
plenum chamber should be zoned with cross partitions so that air may be fur¬
nished as and where needed.
The fuel bed is thin, though not so thin as in spreader stokers. Forced draft
coityeyOT stokers have approximately 150 mm thick green fuel ,bed ; those few
 338 STEAM GENERATORS
built ibr natural draft must operate with thinner beds and lower rates of com-
bustion per square metre of grate sur&ce. forctd draft stokers can develop
810000-1350(]^ kcal per hr perm* grate surface, with drive power of 3-5 hp per
tonne per hr. They will bum fine screenings, anthracite, lignite, and special
coals or sizings not readily handled by other stokers. Bituminous coal can be
used if it is free burning, i.e., noncaking, but as caking coals are the main
steaming coals of this country this is a rather serious limitation. Because of the
required ignition arches, the furnace will be more costly than for other stokers.
10-14 Underfeed Stoker. This type ot stoker is primarily a volatilization
retort to which may be attached overfeed grate sections where much of the

Riley Stoker Corp,

Fig 10-30 Singlo-retort imderfocd stoker.

jBokcd product is burned Fresh coal is supplied from below under pressure,
causing the coal already in the retort to rise, passing in succession through
'20008 of ^tillation, ignition, and combustion. Air is supplied along the sides
of the retorts and below the overfeed sections.
" Underfeed stokers fall into two main groups, the single retort and the
multiple retort stokers. As is seen from Fig. 10-30, the single retort stoker con¬
sists of a trough-shaped retort to which coal is delivered by a reciprocating
ram. A systepa of auxiliary distributing pushers operating simultaneously with
the main ram push the coal longitudinally into the retort and prevent all of it
from rising near the front end. This stoker is not confined to the use of free-
burning coals since the cver-upward motion of the fuel bed supplies a heaving
action which will break up the crusts which form during the volatilization of
coking coals. Air from the tuyeres penetrates the fuel bed, mixing thoroughly
UNDERFEED STOKER 339
with the evolved gaees after which the combustible mixture passes throu^ tfie
bed of ineandescent carbw in order to reach the tumace is^on. This aeeoimts
' for the ability of this stoker to bum a wide variety of free-bumii^ and coking
coals smokelessly. Fine coals which might pack too tightly under the ram action,
anthracite, and coals with relatively low ash fusion temperatures are not suited
to the underfeed stoker.
The fuel bed surface builds up to a natural angle of repose of the burning
carbon. Ash appears at the surface from which it moves to the ash plates. Since
the main purpose of the retort is volatilization, the coal-burning Cgpacity may

Fio. 10-31 Multiple-retort underfeed stoker. (Courtesy Power.)

be increased by providing side overfeed grates to support the coke during

combustion. The thick fuel beds over the tuyeres require that the underfeed
stoker be supplied with forced draft.
Op account of the thick fuel bed, windbox air pressures arejiecessarily high:
,§1 to 127 mm water. This is reflected in considerable fiweed draft fan power;
also it' bars any natural draft underfeed stoker design. Sudter drive takes
about 5 hp per tonne of coal per hr if the ram is motor-driven through gearing
and a cranlahaft.* However, some units are direct-driven by a steam piston.
Combustion control is introduced into the stoker drive either by varying the ram
stroke or by chan^ng the rate of reciprocation, usually the latter. On motor-
driven units this means an on-off control; on steam cylinders, a throttling of
the steam supply. Air-fuel ratio is adjusted by variation of air plenum at tiie
* Ezchisive of any fan power.
340 STEAM OENERATORS
tuyeres. Smokeless operation is possible at very light load. Furnace shape can
be simpT^Because of nonexposure of stoker meclianisin to the furnace, it can
be used with all-refractory furnaces under conditions tliat would endanger
other types.
Some horizontal-retort stokers have stationary side grates; however, only
in small capacities, for the agitated side overfeed grates arc capable of burning
much more coked coal, thus realizing the maximum distillation action in the
retort. The larger stokers sometimes have twin I'ctorls side by side.
Until the advent of spreader stokers the horizontal retort underfeed was the
mainstay in industrial and institutional i>lants burning bituminous coal. It has
more heat lag and, therefore, less sensitive control response; however, it is
less exi)ensive and performs better over a wide load l ange.
This principle of coal firing is applicable to very small units, even dwelling
heating units where ratings as low as 22.7kgcoal per hr are seen.The smallunits
have a pot-like retort which is screw-fed with coal. The retort is surrounded
by dead plates on which ash accumulates. Air-supplying tuyeres are set in the
rim and side of the retort. Control is on-off and drive is electric. In such stokers
the retort, feed, drum, control, and air fan are designed and sold as a single
unit.
Increase of capacity in an underfeed cannot be obtained simply by building
larger, single-retort stokers. A limitation in retort size exists by virtue of ina¬
bility of obtaining even air distribution from the sides of wide retorts. In the
past, sloping multiple-retort stokers, such as the one pictiirM in Fig. 10-31,
were installed under large boilers. Overfeed sections are extensions of the retort,
and in general the coal flow is from front to back of the furnace.
10-1S Coal Burners. Finely divided coal can be burned with certain ad¬
vantages that result in economic, flexible operation of large steam generators.
The method of prei)aration of pulverized coal and systems of transportation are
given in Chapter 12. Here we are concerned with the final piece of apparatus in
a pulverized coal system—the burner—b^Cjaugeit^is built,into and becomes part
of the,boiler furnace. It may be assumed that methods and equipment capable
COAL BURNERS 341
of preparing a stream of pulverized coal and air, regulated in quantity to the
boiler demand, are currently well developed and commercially available.
One manufacturer lists^the following functions of a successful burner. The
coal and air need to be handled so that.(l) there is stability of ignition ; (2)
with effective adjustment for control of point of ignition and resulting flame
shape,; (3) complete combustion; (4) heat uniformly developed in the flame
(no superhot spots); and (5) adequate protection against overheating, internal
fires, and excessive abrasive wear.
Coal floated on a portion of the air supply (primary air) comes to the burner
in one stream, while the remaining, or secondary, air is another stream. Both
ducts attach to the burner, from which the operator expects to get a uniform,
stable flame ’’dth practically no incomplete combustion products. Many designs
of burners arc available which fulfill the requirements rather perfectly on as
little as 10^ excess air—a rather remarkable accomplishment which represents
the i^csult of three decades of intensive dcVcI()pment of this form of combustion
in central power stations.
Ignition is product by radiation and flame propagation from fuel already
burning in tH^umace. Distillation quickly follows because jif^the high teca-
perature and smidLsize of coal particles. (60-70% through a 75 mfcron screen),
lotion point is a function of the velocity of the primary stream and the rate

Pat which it fans out and slows down after leaving the burner tip. Flame shape is
'^'^ntrollable by the adjustments of secondary air swirl vanes and other control
justments incorporated into the burners.
Pulverized coal burners are frequently classifed as long-flame, short-flame
(or turbulent), and tangential burners. The principal differences are in the
methods used for, -and rapidity of, air-fuel mixing, i.e., in turbulence. In the
long-flame system, the firing is downward in the furnace and, since there is,
therefore, a considerable length of flame travel permissible before combustion
is complete,, the burner itself ia ftot required to make an intensive mixture.
Tv^ulent humers are usually set into furnace walls and launch the flame
hdn^htalTy into the furnace. Full and complete mixing is required to be com¬
pleted At the burner. In flame is short and intensely hot. Tangential burners^
are set in the furnacrcomers and directed tangentially to an imaginary circle
in the center of the furnace. The swirling action produces such a turbulent
furnace atmosphere that high-turtojdeQQa^.the bumex itself is unno^ssary.
This type of* burner is sometimes constructed with tips that can be angled
through a small vertical arc so as to xaise or lower the position of the turbulent
combustion region in the furnace. This, in connection with a water-cooled
furnace, gives some control over temperature of gases at the furnace aperture
and is an assistance in maintaining constant superheated steam temperature
as the load varies.
Control of combustion in pulverized coal furnaces is secured by placing coal
feeders under the combustion control system. Air-fuel ratio' is controlled by
secondary air plenum, through dampers or fan control.
Cyclone Burner. Two disadvantages of pulverized coal are: (1) cost of
pulverization and (2) entrainment of 60-70% of the ash as “fly ash,” requiring
ej^nsive dust collectors in the jw^circuiti avoid these C(gm-
plications there has recently been developed a system of burning crushed, not
342 STEAM GENERATORS
pulverized, coal in vortex suqpeniion. Ck>al is crushed to 6.4 mm maximum size
and blown into a cylindrical ^'cyclone furnace.” Air admitted tangentially at
some 762 mm water plenum creates a strong vortex, highly turbulent. Extremely
high heat liberations and the use of preheated air cause high temperatures in
the cyclone. The fuel is quickly consumed, and liberated ash forms a molten
film flowing over the inner wall of the cylinder. Owing to the inclination of the
furnace, the molten ash flows tp an appropriate disposal system. Scrubbing of
ash particles over the molten ash film results in capture by adhesion. Reduc¬
tion of fly ash to 10-20% of the total ash is vouchsafed by tests on the flue gas.
The use of a pulverized coal burner involves first establishing the flame.
This is done by an auxiliary gas or oil burner. These should be engineered with
the same care and safeguards as any other burner. When the pilot flame is well
established, the coal pulverizer or feeder is started, and ignition of the coal
stream is prompt. Although not a difficult operation, the lighting-off of coal
QAS AND OIL BUENERS 84S
burners needs to be competently and carefully handled by experienced person*
nel, for should the setting take in a considerable quantity of unbumed coal
dust, and ignition occur before the setting had been thorougly purged, a damag¬
ing explosion might result. After a burner is in operation, it is under automatic
combustion control and operators need only
attend to occasional secondary adjustment
to prevent pulsations, unnecessary air, etc.
10-16 Gas and Oil Burners. Oil
Burner, The function of a fuel oil burner
is to receive a supply of fuel, meter it in
accordance with the boiler load, and condi¬
tion it for thorough mixing with air. Oil
humors are sometimes used with large sta¬
tion boilers, but more frequently with the
smaller boilers of industrial, commercial,
and institutional heating plants. In the
smaller capacities they and their air supply,
ignition, and control are incorporated into
an integrated oil burner unit, ^veral man¬
ufacturers have developed highly reliable
automatic units suitable for power boilers
6f 50-500 hp. A large number of light oil Fig. 10-34 Turbulent pulverized
units suitable for dwelling heating plants coal burner camng.
have been marketed.
The main difference in oil burners is the method of preparing the fuel for
close mixing with air. This can be done by vaporizing or by atomizing. Vapor¬
izing docs not appear in the power field. Atomizing can be accomplished (1) by
mixing the oil with a steam jet; (2) by forcing it through a small orifice imder
high pressure; and (3) by centrifuging it. The equipment is known as steam
atomizer, pressure atomizer, and rgtaiy cup atomizer, and examples are given
in Figs. 10-35. That the atomizing action is highly successful is demonstrated
by the willingness of most manufacturers to guarantee 13% CO2 minimum in
the products of combustion of their units. This represents low excess air, say

A. Steam pressure B. Oil pressure C. Rotary cup

Fiq. 10-35 Oil burner atomizers.
344 STEAM GENERATORS
10*25%. The following factors are instrumental in determining which of the
many designs of oil burners are suited to some contemplated installation.
1. Will the fuel be the fluid, easily atomized, refined “light'^ oil, or is a
cheaper viscous oil to be used? Most medium and large plants bum the heavy
oil (“black,” Bunker C, No. 6) for economy's sake, for it is as good a fuel in
heat imits and much cheaper in price.
2. Assuming a widely variable load, is the steam generator to be under
continuous supervision of operators who can start fires, change tips, etc., or is
it to be fully automatic?
In smaller plants light furnace oil is often burned for convenience, ease of
procurement, etc. The equipment is simpler than for heavy oil, which has to be
heated to 65.6-82.2®C in order to make it suflSciently fluid to atomize perfectly.
The pressure atomizing type is better adapted to light oils than to the heavy,
residue-laden fuels, especially in the smaller capacity units. The principle of
atomization consists of forcing the oil through a small orifice, or orifices, at
from 10.5*24.6 kg/cm* pressure.Oils laden with carbon-forming residue and tars
often carbonize at the hot tip, thus plugging the small orifice. Steam jet and
rotary cup atomizers give less operating trouble where heavy oil is used.
Mechanical atomizing generally seeks to produce a conical mist of oil
thorough which the air for combustion can be whipped. For this reason the tips
are designed to whirl the oil before releasing it through the orifice. Load control
may, of course, be secured by varying the pressure of the oil back of the tip.
However, in simple tips the range of oil delivery over which good atomization
is possible is limited to about 1.5:1, that is, maximum delivery is 1.5 times
minimum rate of combustion where oil is still satisfactorily atomized. In such
installations wide-range firing is met by having total load carried by several
burners set in the same furnace, some of which may be turned off at part load,
or by changing burner tips from time to time. Manufacturers have striven for
wide-range pressure atomizing designs by various ingenious methods, including
variable orifice, constant flow, etc. An example of the latter is shown in Fig.
10-35B. Oil is supplied with high vorticity just inside the orifice. The rate of
inflow to the tip is constant and greatly in excess of the maximum rate of com¬
bustion. Quantity sprayed is governed by* adjusting control valve in return
line, thus varying pressure at the tip. However, the vorticity is practically the
same over a wide range of flow through the tip and good atomization is there¬
fore secured over a much greater load range, say 4:1.
Steam atomizers are relatively insensitive to oil viscosity and have wide
range characteristics, i.e., 6:1. They are simple in form and readily controlled,
but there are two distinct disadvantages. One is the cost of the steam used
(0.1-0.8 kg steam per kg oil); the other the impossibility of making a cold start
in a dead boiler plant. Neither of these is important in central station oil burn¬
ing practice, for on large units the steam consumption is near the low value
mentioned. As more and more residue oils from catalytic cracking are placed
in the heavy oil market, the large oil-fired boiler operation tends toward steam
jet atomization. The maximum capacities per tip arc also favorable.
Finally there are the rotary cup burners which have been developed for
small- and medium-sized boilers. These are more complicated mechanically
than pressure atomizers. However, they (1) use low oil pressure; (2) have no
GAS AND OIL BURNERS 345
bmall orifices to plug up; (3) arc more economical to operate than steam
atomizers; and (4) are basically wide range (10:1) and readily controlled for
variable load. As will be seen in Fig. 10-35C, the principal element is a hori¬
zontal, slightly conical, rotating cup running at 3000-4000 rpm. Oil delivered
through the hollow shaft is picked up on the inside surface and moved to the
rim of the cup, where the centrifugal force that has been imparted to it flings
it into the air stream in a finely atomized mist. A primary air fan supplies some
air immediately at the rim in order to drive the oil into the furnace and start
the combustion. Secondary air is admitted to complete the combustion. This
type of oil burner has become relatively important recently due to the wide¬
spread use of Scotch type package steam generators, and No. 6 fuel oil, a
combination which is quite effectively fired by rotary cup burners.

Fig. 10-36 Furnace wall mounting of pressure atomizing burner.

As the purchasers of these steam generators usually want a completely auto¬

matic unit, even to the point of intermittent supervision for labor-saving
reasons, a complete unit will be described in some detail. Generally no small oil
burner is expected to go much below 20-30% full delivery and maintain efficient,
smokeless combustion. Thus there have arisen the following systems of opnjra-
tion.
1. On-off, When on the fuel is steadily delivered at the maximum rate of combus¬
tion until steam pressure or water temperature limits are reached; then the unit is
completely stopp^ until a new cycle of operation is called for by diminishing tem¬
perature or pressure.
2. High-low. Two rates of combustion are provided: one about 30% capacity; one
rated capacity. As long as load fluctuates above low-fire point, the pressure wnll be
maintained by a series of high-low-fire operations of variable duration. Below the low-
fire point the unit goes into on-off operation.
3. Modulating. This system has a low-fire point as in high-low, but above that the
fuel is varied in quantity delivered to suit the steam demand. This is similar to the
contrpl of coal furnaces and gives smoother operation with less thermal strain from
alternately high- and low-temperature conditions.
It tvill be seen that in all three of the foregoing systems it is possible to
operate in such a manner as to require repeated shutdowns and starts (herein-
346 STEAM GENERATORS
after called cycles) of the flame. If the reignition is not properly done, a serious
furnace explosion can be produced; hence automatic units are necessarily en¬
cumbered by much safety and precautionary equipment.
The burner unit should determine that the boiler is not low in water, that
the oil is sufficiently warm for spraying, and that the setting is purged of any
inflammable gas. Then it must light off a pilot gas flame and test the operation
to be certain that the ignition spark was followed by a pilot flame. When this
has been verified, the oil spray is started by opening the solenoid valve and a
photoelectric cell flame scanner put into operation. If the cell sees no oil flame
in a certain number of seconds, it operates to shut down the unit, which usually
has to be hand-checked before it will recycle. After a short time on low fire
(determined by the low-fire relay holding the modulating valve at minimum
opening) the low-fire relay permits the modulating valve to open, and combus¬
tion rate is taken over by a “pressure-troF^ element operating through a valve

As long as lood is obovt low

- fire point, flomo modulofts - When lood drops below low
to moot vorioblt lotfd ‘fire point, steom pressure
limit switch ehuts unit down
Moduloting oil volve
SolojH>id oil volv
Flomo scanner When steom pressure
has dropped below
Burner motor
■uEcnoi control limit I
cycle con be initioted

mErmsEsmmmm
-•rzmuL.
Next cycle
50 too 150
Time-seconds

Fk;. 10-37 Example of program control, automatic oil burner with modulating control
and safety ignition.

modulating motor. Air flow proportioning is taken from the same linkage. As
long as the steam demand remains above that corresponding to the low-fire
point, the burner will ^‘float^’ under the combustion control described. Should
load then dwindle below that point a short period of operation on minimum
modulating valve position will raise steam pressure, causing a pressure limit
control to shut dowm the unit. Later, when pressure has dropped below the preset
differential, a new cycle of operations will be initiated by closure of contacts in
the limit control. This program of operations is set forth in chronological se¬
quence in Fig. 10-37. The timing sequence during ignition can be obtained from
a cam driven by a synchronous motor. A schematic diagram of a typical ar¬
rangement of equipment and controls for a modulating type rotary cup burner
is given by Fig. 10-38.
^ Gas Burners. Gas is the only fuel which is originally in suitable physical
condition for use in a burner. However, as with other burner fuels, the gas sup¬
ply must be regulated to the load, air in proper proportion admitted, and the
two thoroughly mixed. The intermingling of a fuel gas and air is comparatively
easy since both are in the gaseous state. A conunon method for boiler practice
OPERATION AND PERFORMANCE 347
is to divide the gas stream into many small filaments by a system of manifolds
and nozzles located in the burner. This system produces gas streams of small
enough dimension that surrounding air readily mixes with them. The mixture is
launched into the furnace through a nest of short mixing tubes of refractory
material and burning is in short-flame form. There are also other systems of
gas combustion, but natural gas is not often used for steam generation because
of cost Its convenience is unexcelled, and the boiler plant is considerably
simplified compared to oil or coal In certain businesses and industry these
considerations may outweigh fuel cost, and gas, if available, may be used. In

Fig 10-38 Equipment connections on fully-automatic oil burner having modulating

control above the low-fire point and on-off below Note Each circmt represents a cable
consisting of the required number of conductors

such cases the puchaser is usually looking for a fully automatic plant, and
equipment similar to that of Fig 10-38 is readily modified to control and safe¬
guard gas burners
10-17 Operation and Performance. The operation of a modem steam
generator is a job for trained, intelligent personnel Gone are the black gangs
of former years, so called from their general appearance after spending a shift
passing and firing coal by hand. With stokers, burners, and automatic control,
operations have become mainly supervisory in nature, although hand loading of
a coal-handling system, and hand removal of ash are frequently to be found in
smaller plants, including those of recent design and modem equipment. Fully
mechanized plants are not always financially justifiable. In the larger and in
many of the smaller plants, operations are divided into shifts, for continuous
348 STEAM GENERATORS
production is the rule. In some of the smaller plants it is the practice to bank
fires overnight.
As was said, operation is mainly supervisory. Since loads are usually varia¬
ble, operation consists not only in ascertaining that the equipment is following
the load and functioning normally, but also in making secondary adjustments
which refine the thermal efficiency beyond the normal abilities of the usual com¬
plement of automatic equipment. Automatic equipment must be watched
through instruments or periodic inspections, load shifts between multiple units
made, critical points for pressure, temperature, etc., inspected from time to time
and other precautionary actions taken. In most plants, especially central
stations, part of supervision is “taking the log,” i.e., the record of flows, pres¬
sures, material quantities, and other physical data. These are supplemented by
samples of coal, water, etc., for testing in the plant laboratory. Performance of
the plant as a whole, or parts of it, are computed at intervals from such data.
Then there is always a certain amount of maintenance and repair, this being as
true of a boiler room as any other part of a power station. The larger the instal¬
lation, the more the need for division of operational duties between specialized
groups such as operations, tests, plant improvement, maintenance, repair, etc.
Professional engineers are needed for many of these duties; others can be
performed by nonprofessional technicians with special training, but there are not
many jobs for common labor in the modem boiler room.
Among the duties of professional nature are the testing of steam generators
and related equipment for performance. Although this often means “thermal”
performance, such is not always the case. For example, the performance of a
dust collector is not a matter of thermal eflSciency. However, the major operat¬
ing expense is fuel, so that the eflSciency with which it is converted into steam
or kw hr is of prime importance to operations. Performance tests are conducted
initially to determine whether sales guarantees are met, subsequently to test
whether a plant is performing as well as can be expected, to isolate faulty per¬
formance, and to check on changes designed to improve performance. Tests of
steam generators and firing equipment are among the most necessary and
complex of all such tests.
In a previous section of this chapter the meaning of “eflBciency” as applied
to steam generators was defined. Although it is of interest and importance to
know, for example, that the efficiency was 76.5%, what is often wanted is a
breakdown of the 23.5% losses so that each may be individually considered for
normal range, possibility of decrease by making operating changes, better main¬
tenance, etc.
The most widely accepted subdivision of losses is that of the ASME Test
Code in which the losses are segregated seven ways as follows: (a) Raising the
dry chimney gas from atmospheric temperature to flue gas temperature; (b)
boiling off the moisture in the fuel, and superheating to flue gas temperature;
(c) forming water vapor from free H2 in the fuel and superheating it to flue gas
temperature; (d) failure to realize the full heat of formation of carbon and
oxygen, as indicated by the presence of some CO in the products; (e) loss of
potential heating value represented by the presence of unconsumed combustible
in the ash refuse; (£) heat transfer from the setting to the ambient atmosphere
by the process primarily of radiation, with some secondary induced convection;
 OPERATION AND PERFORMANCE 349
(g) a miscellaneous group of relatively small magnitude, frequently not directly
measurable. It is designated ^^unaccounted-for-loss” and consists of the,super¬
heat taken on by the humidity in the air used for combustion, sensible heat in
ashes, free carbon floating in the gaseous products of combustion, and some
other small items.
When tests show that one of these losses has become excessive, then a
knowledge of the sources of the loss enables the operator to look intelligently
for his difficulty among the many items which constitute the total boiler loss.
Table 10-4 is a resume of the more common causes of thermal loss associated
with steam generators.
Large power and heating plants are invariably designed and constructed
with allowances for testing, consisting of instrumentation, with provision for
Table 10-4. CAUSES OF HEAT LOSS IN STEAM GENERATORS

^A. Loss due to moisture in coal.

a. Excessive wetting down of coal before firing.
b. High moisture absorption by coal in yard storage.
B. Loss due_to moisture formed by combustion of hydrogen. Irreducible for any spe¬
cific fuel. This loss is larger for oil and gas fuels than for coal.
C. Loss due to heat carried away in dry chimney gas.
^ a. High excess air as revealed in low CO2 content of flue ga‘<.
b. High flue gas temperature.
1. Dirty heating surfaces.
2. Poor water circulation. Scale on water side.
3. Dead gas pockets. Leaky or ineffective baffles.
4. ^Gas velocity too high.
D. Loss due to incomplete combustion.
a. Insufficientjdr.^uppLy. , /
b. Fuel bed in poor condition. aeo-rvLa,
c. Undercooling of.furnace at low ratings^ ^
d. Improper setting of boilers.^
JB. Loss due to combustible in ashpit.
a. Grate or stoker not proportioned toicind ofjfuel used.
b. Too higlij:ate of combustion attempted.
c. Grates dumped or fuel bed sliced too frequently.
d. Furnace temperature is aboveJusionJ^perature of ash^
F. Loss from radiation and convecdonrirom boiler and setting.
a. toiler drums uninsulated.
b. Wall u£ setting too thin or of poor quality.
c. Furnace refractories in need of repair or renewal.
G. Loss duerto moisture in the air.
a. Moisture-laden air as from steam isL
b. High eycess air nn days nf high humidity. This loss is small and frequ^itly in¬
cluded with several other small losses, usually unaccounted for, such as soot
or cinder in the chinmey gas, heat in ashes, etc.

checking instruments, openings for taking samples, etc. The author, however,
has visited several small plants which were assembled with apparently no
thought of any future tests, for when such tests were wanted piping alterations,
breeching openings, etc., were required, and the cost was far more than had they
3B0 , STEAM GENERATORS
been included in the original design. Although not describing in detail the neces*
sary provisions for testing, Fig. 10-39 is intended to summarize the minimum
requirements. Prior to an important test for which the engineer professionally
guarantees the results, instruments should receive calibration, limits of accuracy
be investigated, and general procedure methods be agreed on by parties in¬
volved. This subject is very adequately covered in the ASME Test Code for
Stationary Steam Generating Units and will not be repeated here.
A boiler test for efficiency is of questionable accuracy unless conducted for
several hours at the rate of steaming for which the test is being made. This is
not only to insure that thermal equilibrium is reached, but also to reduce the
effect of tolerances in measurement of fuel consumed. For example, with under¬
feed stokers there is always a large mass of coal in the stoker, with no way to

Orific stotion

Fig. 10-39 Location of boiler test points.

be certain that it contains the same quantity at the end of the test as at the
beginning. But if the range of variation were, say, 454 kg, then the possible
average error in measurement of coal burned would be 45.4 kg per hr on a 10-hr
test, but 454 kg per hr for a 1-hr test.
Considering that each load point takes several hours to test, and that numer¬
ous readings and samples are taken, the reader may see that a full-dress steam
generator test is an undertaking of large magnitude. After the test the data are
rendered by calculation into a heat balance, load curves, or any other form of
report desired. An example of graphical display of performance data is given
by Fig. 10-40, and a typical rendition of a load test into a heat balance is
given in the next section.
10-18 Heat Balance of a Steam Generator. It is quite important to con¬
duct a test at a steady rate of steam generation and fuel feed. To achieve the
highest possible efficiency, small manual adjustments of the fuel-air ratio set by
HEAT BALANCE OP A STEAM GENERATOR 361
automatic control may be necessary. Repeated checks of flue gas composition
test the value of these adjustments. Steady output can be maintained if the
unit is part of a system containing other units which may be caused to carry
all the variable part of system load. An isolated unit should be provided with
an atmospheric discharge line having a conveniently located hand valve. Then
steam may be wasted at variable rate during the test in order to compensate
for variable demand on the plant and thereby hold boiler steaming rate con¬
stant.
The ASME Code provides a “long” and “short” form of report. The long
form lists 154 items and leads to a complete breakdown of the heat losses, appli¬
cable not only to boilers, but to complete generators having superheaters, re¬
heaters, economizer, and air preheater. Except for acceptance tests on central
station units, the short form is usually adequate for all purposes. It is the test
we propose to illustrate here.

Pro. 10-40 isample ])erfonnance of large steam generator

The short-form heat balance consists of eight items, the sum of which must
be the higher heating value of the fuel. These items are, first, the heat usefully
absorbed by the unit, followed by the seven losses described in the previous
section. We shall list these in the ASME order:
1. Heat absorbed by unit.
2. Heat loss due to dry gas.
3. Heat loss due to moisture in fuel.
4. Heat loss due to HjO from combustion of H2.
5. Heat loss due to carbon monoxide.
6. Heat loss due to combustible in refuse.
7. Heat loss due to radiation (including convection) from setting.
8. Unaccoimted for.
Total = Qh, the hiidier heating value of fuel as fired.
362 STEAM GENERATORS
The first five of the losses are calculable from fiue gas analysis, temperatures,
refuse analysis, and fuel analysis. It is usually not troublesome to find or make
stations for getting these data. There are two methods of procedure for jointly
determining the other items, depending on which one is found by the difference
between Qi, and the sum of the other seven items.
Input-output Method. Rate of steam generation is determined by steam
fiow meter or feedwater measurement. Time of test and weight of fuel used
during that time are kept. Then the heat absorbed, per kg fuel, is calculated and
items 7 and 8 found by difference.* This method requires a test of several hours
in order to produce reliable fuel data. Results are subject to the tolerance of
the steam flow meter, the best of which are not guaranteed to closer than ±-\%.
Feedwater measurement by weight or volumetric means is quite accurate, but
plant arrangements are frequently such that feedwater weighing is impossible.
Nevertheless, this method is to be recommended, for it is the more direct route
to efficiency, considered as an output-input relationship.
Heat Balance Method The steam generator is brought up to a stable uni¬
form operating condition by several hours^ preliminary operation, and the
same data taken as for the input-output method, except that rate of steam
generation and fuel quantity are not measured.! This, of course, eliminates
timing as an element of the test. Extra care is taken to accumulate any data
fchat might have a bearing on accurate estimation of items 7 and 8. The average
of readings for one or more hours* operation is then used to calculate all the
losses, and item 1 is found by difference.
It is always preferable to calculate the results by both methods wherever the
data suffice. Of course they should agree within the limits of flow meter ac¬
curacy and setting heat loss calculation, but it is a proficient engineer indeed
who can repeatedly achieve such results.
An ‘^unaccounted for** item is frequently inserted in manufacturers* proposals
as a basis for arriving at efficiency when employing the heat balance method.
However, most of the “unaccounted for** can be closely estimated, and often
totals less than the 1% to 2% allowance seen in formal proposals. Radiation
and convection losses are commonly estimated from the ABAI chart. Fig. 10-
41. Another method is to measure the surface temperature and area of the
outside of the setting, in sections; also to measure the ambient atmospheric
temperature, then calculate radiation and convection from some reliable heat
transfer data.|
Example 1: This is to exemplify the test data and calculations! for heat balance
of a water-tube boiler fired by a spreader stoker employing bituminous coal. The heat
balance basis will be used. Item numbers are those of the short-form code, with any
non-pertinent items omitted.

* Item 7 can be separated from item 8 by estimating ''setting loss.*' See Fig. 10-41.
t Where fly-ash is appreciable, measurements of coal fired and refuse produced per
hour, as well as a sampling of the fly-ash, may be required. Plant coal scales are suffi¬
ciently accurate for this coal measurement since the weight is used to determine carbon
loss, not total heat input.
t Heilman, "Surface Heat Transmission,” Trans. ASMS, Vol. 61.
f The reasons for several of the forms of calculation have appeared in Chapter 5.
 _s

immimwmtmr/',A O
immmtmamyy,-.-
nmmim9vyy'..
nmmtw/yjum
LV r/yyy.'i'J.
mm_
Mil
l_E±J
/iwaaB^wv
■/ys
k9 tm' a m''jm
mikm'm
aMtw.tmrAwm
bm_
■ r. IV # M ■ ■§ iJWw;
5 if; 1S” ”u!zq:

j._
■rr
^VATa
aOD.
-\~AMWAKr ^■IKlVJU' '
_ .U-U_L-U_U
ZF^FdrAM'Am^ ^ W«PMr»li^V*V-U-P-LJZJ

mp^wmimuM

^iSKi!i!!ll<
PSaS/iSSSSiB ■■■■■■
I—1. ^ y.<r4T^r--_ ^-
^MrArAW.m:^Mr.m'sr^m%_

- -
niiS^ISeiillllllW
i ■ I. r—w:jir^■r.T^
IHItJJi
i—i— ■■ ■■ ^_i „ ^ «
L|_u.4a!:Bi^v»aaBBBaiHaaiiHArlxo P-b,
Lp^vBKiKMHBaBBaBiaaaaiiiiiHj ' ^

tT'iiv«!»^l5Hiiia»aBiHiiiii;j rs
ij.i^MiiiimMHiiii^JH n»<
0.illHBHilllllBHHl ^||||||j:S£J1=.
kLUUHHlIlllllIBHI |jjjl||ll»ll
^iJLIlilKHIk^lMIlHmKIIICIIMilKlHka^
„U.IKIk1Mllimilk1BA?Billk^lli:ili:i»r~~
jiiii^i»i9iii»ranniniiii!iii;iiifMi^
« ri •laai^ipiA’B e
inUNI 1.V3H t90HO JO iJNSMOA MOl NOiUrlOVM

S58
 354 STEAM GENERATORS
TEST DATA

1. Steam pressure in boiler drum, kg/em* ab S.1

4. Water temperature entermg boiler, ** 0 108.9
6 Steam quality, percent dry 98.5
6 Air temperature aroimd boiler, ambient, ® 0 86.7
7. Temperature at combustion air intake, ^ C IBA
8. Temperature of fuel, ® C 18.8
9 Gas temperature leaving boiler, ® O 268.9
14 Enthalpy of steam, kcal per kg 647.4
16 Enthalpy of feedwater, kcal per kg 1019
16 Heat absorbed per kg (koa))steam 660.4
17. Combustible content of refuse sample 20 2% by analysis Dry refuse
per kg as-fired fuel, kg (Eq 10-6) .. Oil
18. Heating value of refuse, kcal/kg, gll7.6 X 0 202 1640
19. Actual water evaporated, kg per hr 4819

Flue gas analysiSy percent by volume

83-86 CO2, 12 7, O2, 7 0, CO, 0, Nj, 80 3

FUEL DATA

38. Heating value as hred, koal/kg 7266

Uliimaie analysis as firedf percent by weight

88. H20,2 5, 31 Ash, 9 0, 34 C, 73 6, 36 H2,4 7, 36 02,8 6,
37. N2, 1 0, 38. S, 0 6.

HEAT BALANCE

kcal per
. kgfuel Percent
66. Heat absorbed by unit (by difference) 81 1
67. Heat loss due to dry gas 806 111
68 Heat loss due to moisture m fuel 17 02
69. Heat loss due to H2O from free H2 294 40
60 Heat loss due to carbon monoxide 0 0
61 Heat loss due to combustible in refuse ISO 25
68. Heat loss due to radiation and convection 68 08
63 Unaccounted for (from manufacture! s’ allowance) 03

7256 100 0

Formulae and Calculations Used for Heat Balance

Dry refuse per k, as-fired fuel, R = Irlefuse (10-6)

B = 9.0/(100 - 20.2) = 0.11 kg

Carbon burned per kg as-firedI fuel, C

C" - - ^7^-' (10-7)
’ 100 14,600
73.6 11 X 1640
Whence C' O.MSkg
100 SU7.«
ASME CODE 355
Diy gas per kg as-flred fuel,
_ rilC0, + 80, + 7(N3 + (’0)
L 3 (CO, + CO)
%r.ls] - ^ 1 (i(W)
|~11 X 12.7 + 8 X 7.0 + 7(80.3 + 0)
3(12.7 + 0) ][ ojita + g X 0.006;] +1X 0.006

s e.46 kg

Item 67 = TTdgCpAT (10-9)

Cp given as 0.24 in the Code; AT = item 9 — item 7.
Hence item 57 = 14.23 X .24(263.9-18*3) » 805.73 koal.
Heat to vaporize and superheat moisture in coal —
(506.6— item 8 + 0.46 X item 9)k0al per kg HjO (10-10)
Since there are 0.025 kg moisture as hred,
Item 58 « 0.025(505.6-18.8-f 0.46 X263.9) «17.5 koal
Item 59 = 9 (item 35) X (Eq 10-10) (10-11)
Item 60 = 6649 (CO X C')/(C02 + CO) koal (10-12)
Item 61 = Item 17 X item 18 (10-13)
Item 61 = 0.11 X 1640 = 180 kcal.
Item 62. Surface area of setting is approximately 69.€6id*
Estimated average temperature, 68.3K1.
Since Eq 10-2 is available, it is employed to find setting heat loss, using C « 46|5.T
Fe = 1.5.
gu » 465.7 X 1.5 X 69.6(348^ - 300O 26600kcal per hr
Estimate fuel per hr, assuming efficiency = 80%.

Fuel = Steam generated X ^ = 4819 X =460.4 kg per hr

kg steam 80% X 7266
Then radiation loss = 26600 / 460.4 — 58 koal per kg

PROBLEMS

(w.t. = water tube)

1. Make a logical classification, in outline form, of some assigned steam generator.
2. The operator of a 17.58 kg/cm» ga watwr tube lx)iler needs i replncemont tube,
76.2 mm dia., of seamless low carbon 8te<4. What wall thickness slu/ . he got ? Oonsult
ASME Power Boiler Code for thickness formulae.
3. The HRT boiler shell,Fig. 10-3, is 6.79 m long, 1.98m dia. The 101.6 mm X 2.67 nun
tubes are on. 127 mm center-lines vertically, 133 mrn horizontally, in staggorod arrangement.
Setting walls touch the shelf at the horizontal diameter. Find sq metro heating surface
and rated bo hp.
4. A boiler similar to that of Fig. 10-6 has a bank of 132 tubes, each 76.2 mm x‘3
m and 12 water wall tubes, each 76.2 mm X 4.6 m The heat transfer capacity is said to be
3856 kg per hr *Trom and atl00®0/’ What i)erccnt rating is implied?
5. A fuel of Qi, = 6888 kcal/kgwas burned in the furnace of a 223 m* nvater tube
boiler which was generating544S kg^steam per hr atl4di6kg/cmSab,d and a, fromlOO^
feed, at a time when fuel consumption was 6i2kg per hr. (a) Find koal rating and
percent rating, (b) Calculate over-all thermal efficiency.
6. Measurements of a cross-drum, box header boiler are:4.6m between headers;
366 STEAM GENERATORS
101.0mm tubes, bsnk 20 tubes wide, 90 tubes deep ; pressure. 15.1 kg/om* ab ; O.OSdry.
ness factor ; feed. 82.2**C ; furnace and chimney temperatures. 954**C and 296*^. During
a 3-hr test on this boiler 5715 kg of 2375 koal coal were burned, producing steam at an
average rate of 14969 kg per hr. Find (a) the thermal efficiency ; (b) the over-all coefficient
of heat transfer.
7. A 2323 w.t. boiler produces 14.5150 kg per hr steam at 24.6 kg/ems ga. 399®C
from feed at 177®C. What is its kcalrating?
8. A simple w.t. boiler of 1858 m* heating surface is offered by a salesman for
installation in a steam plant where 36287 kg of d and s steam are to be produced at
13.86 kg/omSgafrom99^:; feedwater. Using the Babcock formula verify whether this area
is adequate by assuming mass flow 24412 kg per h per m* per hr per sq ft, furnace and
chimney temperatures of 982^ and 232*^.
9. A boiler like Fig. 10-6(except no wall cooling) has 101.6 mmx6m long tubes in
a bank 22 wide x 16deep,with 162.4 mm horizontal spacing. Transverse baffles divide the
tube surface into three equal gas passes. Steam,12.3 kg/om* ga,d and«;93.3*0 feedwater.
Fuel: coal at Qs = 2500 kcal using 19 kg air per kg coal, (a) Use the Babcock formula
to find the over-all, heat transfer coefficient, (b) Assume that when burning 3402 kg
coal per hr the furnace temperature will be 1010*C. Then find probable temperature
of gases leaving the setting, also the rate of steam generation, kg per hr.
10. An oil-fired steam boiler generates 5897 kg; steam per hr at 21.lkg/om9ab, 343°C
when the fuel is burned at 464 l/h. Feedwater, 149®C. Fuel oil, 20® API at 15.6/15.6. Cal¬
culate thermal efficiency and equivalent evaporation per kg oil. Consult index for fuel
oil references.
11. A 260 mS w.t. boiler is rated at 8165 kg steam per hr at 7kg/omSgai 0.98
dryness factor. Feedwater, 107®C . Express the capacity as (a) koal,(b) bo hp, (c)
percent rating, (d) equivalent evaporation.
12. Repeat Prob. 10-11 except that the steam generated is 6804 kg per hr at 10*5
kg/ems ga.
13. Testing of a stoker-fired boiler revealed that, during a period when 2064 kg of
3.100 kcal coal were burned, refuse accumulated in the ashpit to the extent of 225 kg,
although the coal as fired contained only 10.0% ash by analysis. Determine the grate
efficiency of this stoker. (No fly-ash.)
14. A large steam generator equipped with superheater and economizer received
feedwarter at 14l®c . After a temperature rise to 163®C in the economizer the water
entered the boiler and was evaporated at 45.7kg/cmS mb» 97%dry. It then passes through
the superheater, emerging at482^ 447 kg/cm* ab. What percent of total heat transfer
did each of the surfaces mentioned contnbute?
15. The boiler outlet steam of a large unit is 99% dry at 103.5 kg/cm* ab. Feed-
water, 216®C. Superheater outlet state is 98.4kg/cinS ab» 538®c . The unit also includes a
steam reheater, the inlet and outlet states of which are 9.1 kg/cmt ab. 249®C and 7.0 ks/
oaskab«371®C respectively. Compute the percentage of total heat transfer contributed
by each element of the heating surface.
16. Find the heat release rate in the stoker-fired furnace shown diagrammatically
in vertical section in Fig. 10-16?. The energy release is sufficient to generate 6304 kg
per hr d and 8 steam at 7.0 kg/emSga from l00®Cfeed at an over-all thermal efficiency
of 0.789.
17. Given that the furnace in Fig. 10-16P has 40% “fraction cold” in water walls
of a type for which S = 0.6x. Assume clean walls and no incomplete combt^on. If,
at the rate of firing mentioned in Prob. 10-16, the A:F ratio were 13, what would the
furnace outlet temperature be, as predicted by Eq 9-26. Coal, as fired, has Qa = 7360
koal/kg,Q,^ = 7280 koal/kg,10% ash. Cp = 0*26 koal par kg.deg C; atmoephere,S17^.
18. Solve Prob. 10-17 by use of Fig. 10-13B, assuming that the ccmibustion equip¬
ment is a spreader stoker.
PROBLEMS 367

Fig. 10-16P

19. The furnace of a certain large boiler is diagrammed in Fig. 10-19P. Ap of tube
banks*7.enix jm. Side walls each shielded by76.2mmx3m tubejon 162mincenteis.Front wall
refractory, Bridge wall, refractory covered tubes, having A, = 8.7 m*. Fuel consump¬
tion, 2668 kg per hr good bituminous coal; A:F = 12. Preheated air at 08.8^. Qj, =
2626 koal. Assume Cp = 0.24 koal per kg-deg C mean between 15.6^C and 98.3^ ; 0.26
between98«8*C and Find the temperature of the gases at the furnace aperture, C,
using Eq 9-26.
20. Afumace4.6mx4.5inx6mhighhas all vertical walls shielded by tubes arranged
as in Fig.l0-12A. Tubes are 76.2 mm dia.onivrSmmcenters. The top of the furnace opens
intoa tubebank for which 9 can betaken as s •• ir Furnace bottom is an ash screen layer of
76.2innitubes on 177.8 mm centers, 9 The available heat released =192465 kcal per hr
perm* furnace volume. Combustion air flow = 72614^ per min at 15.6*0, preheat^ to
148* 0. Slag factor = 0.90. Fuel consumption, 4082 kg coal per hr, pulverized. With the
aid of Fig. 10-13A, determine the temperature of gases leaving the furnace and the
kcal beat transfer by radiation. Record any necessary assumption of data.
21. Solve Prob. 10-20 by the use of Fig. 10-13C.
22. Taking Qb of the coal burned in Prob. 10-21 as 2852 koal, determine the heat
release rate in the furnace mentioned. What does this suggest as a minimum ash fusion
temperature for coal purchasing specifications?
23. Estimate the required furnace volume for a pulverized-coal-fired steam gener¬
ator where 1275x10* koal per hr beat transfer at 80% thermal efliciency is the expected
thermal performance. Boiler has water walls. Ash fusion temperature, 1204* C.
24. Specify the insulation thickness for a furnace wall composed of 228*6 mm fire
clay brick, ini^tion #5, Fig. 10-15,and sheet-metal casing with glossy black paint. Atm¬
ospheric temperature, 29.4*c; furnace wall face,954.4*0.The insulation is to be suflicient
to hold CMing temperature to 66,6*0.
25. Find ^e kcal per hr-m* heat loss through a furnace wall from 922.2* C hot
face temperature to 16.2*C ambient temperature. The wall is composed of 222.6 mm
fire-elay brick, 114.2 mm diatomaceous silica brick, andf02Ji mm red brick,
26. A water-cooled furnace wall such as Fig. 10-12A has 76J mm costaUe lefra*
etory concrete, 50.2 moi mineral fiber block insulaticm, and 50.2 mm mineral wool blanket
insulation; aD covered by a steel casing painted aluminum. What is the earing tempe¬
rature when the temperature of the inner surface is 760*0? N6ri6et irregularities'due to
tube sockets and assume 22.ira, ambient temperature.
27. One wall of a pulveris^ coal furnace is bare refractory having 1204*C face
temperature. The 228.6 mm of fused silica refractory is bsdmd up by ttS.Omin of
358 STEAM GENERATORS
insulating fire brick, then by 88.9 mm of mineral wool blanket over which is a metal
casing painted dull black. TTie casing temperature is60®C. Find the heat leakage thro¬
ugh this wall, kcal per hr*
28. In Fig. 10-14, assume t, = 648,9*0, =32.2°C, =50.8 mm refractory fire clay
tile, ^2 = 101.6 mm of block insulation (#7, Fig. 10-15). The casing is metal, painted
aluminum.Find Q, to, t^.
29. A convection superheater with tube elements 38.1 mm x2.7 mm x 6 mlong is to
be placed in a boiler setting where it will be able to superheat steam of 24 6 kg/cm*
ga^ 98% dry, to 260®c when G, = 196 kg steam per sec-m® . Coefficient of conductance
expected to be realized is 73 kcal perm* hr deg C. = 1.8; Cg = 0.25. Find requ¬
ired t'l. Given 22680 kg per hr rate of steam flow, how many elements should the super¬
heater have?
30. Predict an over-all coefficient of convection for heat transfer in a superheater
employing 88.1 mm x 2.7 mrntubes. Pressure,!7.6 kg/om* ga; average steam temperature,
260®C,velocity,2l34 mpm. Average gas temperature, 482°C. Mass flow of gas, 43922 kg
per hr-m*-
31. The temperature characteristic of a certain convection superheater rises from
37l*C at 27216 to 454®C at 54432 kg per hr steam flow through it. Pressure, 42 kg/om*
ab. Spray water, 93.3°C. What constant superheat can be maintained by <k‘Miperheat
control for a delivery range of 27216 to 54432 kg per hr? What part oi the 54432 kg
flow will be evaporated spray water? By what percent does the regulalod steam volume
at 54432 kg per hr flow differ from the superheater discharge volume ?
32. Diagram the following systems of superheat regulation, including method of
control: (a) contact water spray; (b) initial condensation.
33. In a proposed countorfiow steel tube economizer the W^/W^ ratio is to be
1.5. = 316®C ; fj = 82.2°C. It is desired to proportion the economizer so that At is
54.4" c. Find the necessary sq m heating surface per 454 kg per hr feedwater. Gas
mass flow, 58563 kg per hr m*.
34. Design and sketch a counterflow economizer composed of'88.9 mm dia. longitudi¬
nally finned tubes, tj =26.7*C; = 371.rc; t^ =176.7°C; W^/W^ = 1.72. Maximum
allowable draft loss 88.9mm. Feedwater flow, 4536 kg per hr. Assume rectangular duct
D = 4C. Tubes on 114.8 mm centers.
35. Determine z, y, (\ D(Fig. 10-27) for an economizer with 38.1 mm dia. tubes, longi¬
tudinally finned. t\ = 371.1 C; tj = 2l.l®0. It is desired to reduce the gas temperature
to 232.2°C. A:F ratio 16:1; evaporation, 9.5 kg per kg coal. Allowable draft loss,50.8
mm. Evaporation of boiler, 6804 kg per hr. Tube spacing, 63.5 mm; D = 3C.
36. Assume that the test covered in Ex 1, Sec 10-18, was to provide data for the
input-output method. In addition to the data of the example the test produced infor¬
mation to show that the coal con.«umption was 442 kg per hr. Formulate the new heat
balance, omitting any calculations that would duplicate those in the example.
37. Calculate a boiler heat balance from these data:
Coal analysis: Coal No. 11, Table 5-4.
Refuse analysis: 20% combustible (no fly-ash).
Flue gas analysis: CO2,10.48; CO, 0.28; Oo, 9.(X); N2, 80.24.
Atmospheric temperature,21.1*C; flue gas temperature, 232.2*C.
Assume that items 7 and 8 total 0.05Qh.
38. Calculate a boiler heat balance from these data:
Coal analysis: C, 60.74; H2, 4.00; S, 1.32; O2, 8.24; Nj, 1.15; HgO, 12.85;
ash, 11.70. Qy^ 6293 kcal i^r kg.
Flue gas analysis: COo, 12; CO, 1; O2, 7; N2, 80.
Atmospheric temperature, 11.1*0; flue gas temperature, 260*c.
Heat transferred to steam per kg coal, 4620*
Refuse in pit per tonno coal bumad 42.2 kg, free of combustible* Flyrash sampling
showed 25% combustible in it. The generating unit has capacity of 29484
kg per hr, was tested at a load of 22880 kg’per hr. Two of the furnace side
walls are watereooled.
 CHAPTER 11

11-1 Prime Mover Types. A machine which originates mechanical mo¬

tion using some natural force is a prime mover. All heat engines are prime
movers, as are hydraulic turbines. An electric motor would not meet the speci¬
fication. Steam-using prime movers are the engine and the turbine. The engine’s
piston furnishes a wall against which the bombardment of the closely packed,
highly kinetic steam molecules build up a force—a working force because,
unlike the cylinder which receives force of the same magnitude, the piston
moves, so allowing the steam force to accomplish mechanical work. The effect
on the steam is to cool it (reduced molecular velocity) and expand it to greater
volumes and lower pressures. In the turbine the random kinetic energy we call
heat is first partially organized in the form of a steam jet, after which the con¬
version to torque energy is a problem of mechanics, met by the use of bladed
wheels which deflect and slow the jet, thereby receiving a torque force. Nozzles
are the elements of a turbine in which the jets are formed. These are quite
simple; almost any orifice could serve as a nozzle of sorts, but for realization
of maximum kinetic jet energy the nozzle shape must be that dictated by
thermodynamic principles of isentropic flow\ To organize the heat energy and
so transform it io unidirectional energy of a jet the pressure at the nozzle exit
must be less than at the inlet. A molecule moving through the nozzle towards
the lower pressure will have fewer unfavorable collisions with its neighbors
than one trying to travel oppositely. In this manner the molecules of the steam
arc set into motion downstream. Under a few kg per square cm pressure
difference the steam discharges from the nozzle as a high-velocity jet.
Although the molecular action whereby these two prime movers create mo¬
tion is quite different, the statistics of the molecular assemblage are the same.
Steam is required to be available at a pressure in excess of the final or exhaust
pressure. The working process is adiabatic and ideally isentropic. For similar
terminal pressure states the percentage organization of pressure energy into
mechanical work is approximately the same. This may surprise some, for so
much emphasis has b^n placed on the high efficiency of the large condensing
turbines that it may have been supposed that turbine action is therraod3mam-
ically superior to engine action. Such is not the case; rather, the compactness
of the turbine in large sizes, especially when in condensing operation at high
vacuum, is the reason for its preempting certain fields.
3S0
360 STEAM PRIME MOVERS
The engine is essentially a pressure machine; the turbine, a flow machine.
The reciprocating parts of engines limit their speed to a comparatively low
value, but turbine energy is obtained from a number of small forces working at
high velocity. Smaller dimensions and freedom from vibration give the turbine
an advantage in first cost, space, and foundation requirements. The turbine’s
superiority over the engine in compactness of large-sized units, and the Diesel
engine’s superiority in fuel consumption and quick-starting ability, are factors
which would have eliminated the steam engine had it not been for certain ad¬
vantages possessed by the latter which are not matters of thermal efficiency or
first cost.
When exhaust steam at or above atmospheric pressure has commercial value
the steam engine comes into a range where it performs to advantage compared
to the turbine. Lowering of the thermodynamic efficiency due to steam leakage
past blade clearances is the cause of poor turbine performance in the* high-
pressure range. On the other hand that range is well suited to the engine in
which clearance can be sealed with packing and rings, and for the further reason
that the reciprocating engine has no trouble in handling the volumes of high-
pressure steam. But in the low-pressure range the engine’s inability to handle
low-density steam through the ports rapidly enough to be efficient without
being unduly large is a disadvantage. The efficiency of the steam turbine,
however, is excellent. The long blades have comparatively low clearance losses
and can handle large volumes of steam because of the high steam velocity and
the proportionately large area that can be devoted to exhaust steam flow.
The less it has of this uneconomical low-pressure range, the more able the steam
engine is to compete with the turbine. In sizes larger than 1000 hp the inher¬
ently low speed of the engine begins to affect unfavorably the cost of the
electrical part of the installation. Serving highly variable loads, as industrial
loads are likely to be, the engine shows good economy over a wide range of
load. In summing up this comparison of the two types of prime movers it is
evident that the advantages cited for the engine will not he of moment in
central-station practice hut should he carefully weighed for each industrial
project.
11-2 Steam Engine Applications. The operating engineer’s field is stud¬
ded with small power plants in which the motive power is the steam engine.
Many of these are old installations, but still quite serviceable to their owners.
Some mechanically drive the machinery of a factory, while others are direct-
connected to generators. The steam engine is also found as an auxiliary source
of power, either because it can be direct-connected to equipment requiring
relatively slow speed, such as 300 rpm or less, or because it can be furnished as
a reversible drive, or to render the driven equipment less vulnerable to power
interruptions from outside sources.
The two principal uses for new steam engines in stationary service seem
to be:
1. Auxiliary power service. Small, upright, single-cylinder engines, with
simple valve. Thermal efficiency rather poor, but frequently unimportant be¬
cause exhaust heat is salvaged.
2. The larger engine embodsring modem refinements of mechanical and
thermodynamic nature, usually direct-connected to an electric generator and
STEAM ENGINE APPLICATIONB 861
serving as the indqiendent power source of many an industry, likewise many
an institution or booness.
It could be noted in passmg that the steam engine is also an important
modem prime mover in marine service, and many new installations are mnda
in ore-carriers, tugs, ferries, etc. Such units are available in sizes up to 5100 hp.
Because it is more efficient in its use of steam than the older dual-flow en¬
gines, the unafiow is the type on which attention is focused when the reciprocat¬
ing engine is considered as a principal power unit. Engines of this category are
built as single-cylinder horizontals and as multi-cylinder verticals. As the
unaflow is inherently long, the horizontal form must be chosen unless an un-

SkHmer Engine Co.

Fro. 11-1 Three-cylinder, vertical, 760-kw imaflow eng^ine-generator power unit.

usual amount of headroom is available in the engine room. The horizontal

single-cylinder engine is likely to be the choice up to 455 hp; above that the
vertical multi-cylinder units prevail.
Most engine power plant applications fall in the steam pressure range 8.8-
21.11^/cm*gawithupto315°C steam temperature. Sales of new equipment are
principally to those industries and institutions which need low-pressure steam
not only for cold weather space heating, but also for round-the-year processing
of various sorts. The types of industries which frequently fall into this category
are textile manufacturing, woodworking mills, paper mills, and chemical in¬
dustries. Hospitals, schools, prisons, hotels exemplify the institutions.
Not all power plant work involves new prime movers. Steam engines possess
362 STEAM PRIME MOVERS
a demonstrated ability to withstand* the years if properly maintained and
lubricated.
The power plant consultant may be retained by an industry which has out¬
grown its belt-and-lineshaft system of power transmission, but whose engine is
mechanically good, fairly efiBcient, and apparently serviceable for many more
years. These engines are usually found to be large, slow-speed horizontal en¬
gines, such as the Corliss, with belt drive taken from the rim of a large flywheel.
In some of these cases the owner^s interests are best served by a new plant;
however, factory electrification can often
be accomplished and the expense of a
new engine avoided by the installation
of a generator and control panel. As
slow-speed generators are unduly ex¬
pensive, a V-belt drive from the flywheel
rim to a small-diameter generator sheave,
as in Fig. 11-2, may be the economical,
practical way to electrify. The small
crown placed on the flywheel rim to
Fig. 11-2 Generator drive from slow-
make the flat belt run true needs to be
speed engine flywheel. Distance X cut off, in order that the multiple strands
chosen to give equal grip of belts on of a short-center, V-belt drive can run on
flywheel and sheave. the flywheel without crowding.
Industrial use of the steam engine is
associated with noncondensing operation, often at exhaust pressures of several
kg/cm* gauge in order to supply steam to some manufacturing process. Provided
that the engine incorporates a means for adjustment of the degree of compres¬
sion, the same unit can be operated alternately condensing and noncondensing.
This is desirable when the demand for process or heating steam is irregular
and high-efficiency condensing operation is employed when exhaust steam is
not in demand. Under these variable conditions the engine can usually show
better steam economy than the turbine.
Installation, It will be as'^umed that the common mechanical features of
the steam engine, i.e., its kinematics, valve types, and customary configuration,
are already known to the reader. A bare engine, onc(^ selected and purchased,
needs a well-engineered installation, including provisions for lubrication, drips
and drains, moisture separation, and testing. The foundation massf needs to
be rather heavy for single-cylinder, horizontal engines.
The steam pipe lead from boiler or steam header should be sized for lower
average steam speeds than are used with turbines on account of the pulsating
steam flow to an engine. A well-anchored steam lead, with velocity not over
610m/minjonthe average, should not vibrate, but might for certain configura¬
tions, especially where the lead is long. Receiver tanks located near the engine
throttle are sometimes necessary to eliminate line vibration. Moisture in the
♦The result of absence of high temperatures and sharp pressure waves. Temperatures
are more uniform than in I.C. engines. Speeds are slow. Erosion and unbalance are un¬
likely to occur.
t See Sec 4-7.
t Somewhat higher speeds are practical with multi-cylinder engines.
STEAM ENGINE APPLICATIONS 363
steam line should be prevented, if possible, by ^ood heat insulation and by
taking the lead from the top of the main steam header. However, a sepkrator
ought to be installed near the engine to catch water that might accidentally
get into the line while operating, and to remove condensation when warming
up the line. A gate valve should 1^ installed, in addition to the throttle; a drip
line is needed to drain the condensation from the pressure side of the throttle
valve before a start.
Much water of condensation is formed when first admitting steam to a cold
engine. Cylinder drains are provided for releasing this water before the throttle
is sufiSciently opened to revolve the engine. The valves in these drain lines are
of the “snifter” type which are spring loaded for closure and can be opened
either manually or by abnormally high cylinder pressure caused by accumulat¬
ing water when in operation. Snifter valves will not protect a piston or cylinder

Fig. 11-3 Equipment and piping for steam engine installation.

head from damage when a large slug of water appears and the engine is up to
speed, but are safeguards for lesser cases of water knock. The engineer should
not overlook the need to provide drains from low spots in the exhaust line when
laying out an engine installation.
Lubrication. The rubbing and sliding surfaces in a steam engine are numer¬
ous. Lubrication is therefore highly important—and somewhat complex. Lubri¬
cation of a steam engine is divided into (1) cylinder and valve lubrication and
(2) bearing lubrication. The smaller engines can receive lubrication of all parts
in contact with steam by oiling the steam entering the engine with the common
hydrostatic lubricator attachment. This is a simple system but subject to
operational fluctuations and hence is not used on modem main power units.
Instead a force-feed oiler with mechanical drive from the engine crosshead is
used to feed oil to the valve, cylinder walls, and piston rods. As it comes into
contact with hot steam this oil is heated and thinned; therefore, it is a heavy¬
bodied oil with high viscosity at atmospheric temperature. Best lubrication is
achieved by a com^imded oil with good emulsifying properties. On the other
hand, bearing lubrication is by straight mineral oil which will not readily
864 STEAM PRIME MOVERS
emulsify, for, whether splash system or force feed is used, the bearing lubricant
can be salvaged and reused if it separates readily from water. On some of the
older engines the bearing lubrication will be found to consist principally of
gravity-fed oil. These “all-loss’’ systems are mainly sight-feed oil cups, one
for each point of lubrication. Minimum oil consumption requires careful ad¬
justment of feed to least need plus use of a medium heavy-body mineral oil
having exceptionally good persistence-of-film property.
Practically all the oil injected into the steam appears in the exhaust. There
are relatively few uses of exhaust steam that can accept this contamination.
An oil separator to catch and divert the emulsified oil particles is needed if the
exhaust steam has further usage. By sparing use of cylinder lubrication and a
good separator on the exhaust line, the steam is fit for most purposes, although
its condensate may need further oil cleansing before becoming a safe boiler
feedwater. Oil separators may work on the principle of gravity separation, if
space for large reservoirs is available, or by centrifugal action provided by
cyclones or baffles. The condensate can further be filtered through sand or
charcoal, diatomite, or other materials which absorb oil. It is possible to re¬
duce oil content of filtered water to the limit of error of usual analytical
methods, a contamination of less than 0.1 ppm.
Governing. There have been two systems of steam engine governing for
constant speed: the constant cutoff, variable inlet pressure called throttling
governing; and the constant inlet pressure-variable cutoff method called cutoff
governing. Nowadays new engines, large and small, are equipped with cutoff
governing since this system is more efficient. Governors are of the centrifugal
type. They are attached to the flywheel and revolve at flywheel speed. Valve
drives are taken from shaft-mounted eccentrics (or their kinematic equivalents).
Mechanical Details. These, being problems of the engine manufacturer, are
mentioned here only in summary. All steam engines are double-acting and
mostly single-cylinder if of the horizontal type. Principal differences lie in
the design of valves for admitting and releasing steam and the mechanical
valve gear which produces the valve motion. Mechanical variations are great
in number, but we shall consider these to be internal features of the engine
and avoid descriptions. Valves are usually sliding piston type on the smaller
engines used for auxiliary drive, and either, drop-piston or double-beat poppet
types on the larger engines used for generator drive. Modem valves are non-
releasing, and the cylinder flow may be classified as counterflow or unaflow.
Releasing-type valve gear such as the Corliss, although very efficient in eliminat¬
ing wire drawing, are confined to use on relatively slow-speed engines, say
up to 125 rpm, whereas the trend has been to higher speeds, i.e., up to 300 rpm.
Although it is a more costly type to build, the unaflow engine is the most
efficient. Since thermal efficiency affects annual fuel costs, it is often the de¬
ciding factor in selection. For this reason the unaflow has come to the fore as
a main power unit where engines are warranted.
11-3 Steam Engine Capacity and Performance. Steam is admitted to
the engine cylinder at throttle pressure during the first part of the working
stroke, then cut off by closure of the steam valve. The steam so trapped in the
cylinder expands adiabatically to the release pressure, then is exhausted from
the cylinder during part of the return stroke. Compression is practiced in steam
STEAM ENGINE CAPACITY AND PERFORMANCE 366
engines to cushion the shock of incoming high-pressure steam. This cycle, de¬
picted in Fig. 11-4, is the Rankine engine cycle. The Rankine vapor cycle of
Chapter 8 was the thermodynamic cycle of a unit quantity of the working
fluid, but this engine Rankine cycle is
the cylinder steam p-v relation during
one revolution of the crankshaft. The
expansion is polytropic. It has been
found that the form of this expansion is
approximately hyperbolic, i.e., pv = C,
if the steam is initially saturated or only
slightly superheated. If originally highly
superheated, then n 1 and values murt
be determined from the isentropic lines
of steam expansions. Thermodynamic Fig. 11-4 Rankine eng^e cycle.
Properties of Steam (Keenan and Keyes)
charts isentropic exponents for steam expansions.
Steam engines always have small pipe taps into the clearance space so that
an engine indicator may be attached for the purpose of testing. The indicator
would alternately be applied to both head and crank ends with resulting dia¬
grams as pidtured in Fig. 11-5. Reference should be made to this figure for
the following nomenclature and relations.

e =• Fractional clearance, »»/(»« — r»).

D B Cylinder bore (also taken to be piston diameter in capacity calcular
tions), cm.
ly = Piston rod diameter, cm.
L » Piston stroke (»> 2 cranks), cm.
N <= Revolutions per minute of double-acting engine.
n = Number of cylinders.
PD = Piston displacement = 0.7854 (D* — D'*)L, cm*.
Pi, Pi ^ Throttle and exhaust pressures, kg/cm* ab.

Pm I ■= Ideal mean effective pressure.

Pmop Indicated mean effective pressure.
Jf2 «» Apparent ratio of expansion, (v, — r,)/(vb — «»»)• (R = Reciprocal
fractional stroke at cutdf.)
R = r/(l -I- c - re) (11-1)
r n Ratio of expansitm, v./vb.
Effective pressure is the difference between the fluid pressure acting against
the piston on the working stroke and the pressure at the same point on the.
return stroke. Effective pressure varies. Its average is the mean effective
pressure.
Fig. 11-6 shows head end and crank end indicator diagrams. The indicator
366 STEAM PRIME MOVERS
will also place an atmosphere line on the card, but the zero pressure and volume
axes must be drawn in by hand, using barometric pressure and clearance data.
The ideal conventional cycle, assuming no clearance, wire-drawing, or fluid
friction, has also been added to this figure. This cycle is useful as a standard
of comparison and is necessary in order to define diagram factor which will be
introduced subsequently. The mean effective pressure of the conventional cycle
is given by the following equation, assuming hyperbolic expansion.

Pm t Pi - P* (11-2)

Mean effective pressure of the indicated cycle is calculated by obtaining the

area graphically, then dividing by
its length to get the average height.
This height, interpreted as kg/cm*
with the use of the indicator spring
constant, is j^mep.
Effidendea, When in operation
the engine draws heat energy from
the boiler in the form of high-pres¬
sure steam, expands it to the exhaust
pressure, then dumps the heat of the
Fig. 11-6 Indicator diagram with super¬ exhaust. This rejected heat is the
imposed conventional cycle. main loss, but there are others which,
although of considerable lesser mag-
nitude, are more interesting to engineers because -improved designs can reduce
them. The energy losses of a steam engine are:
1. Exhaust.
2. Initial condensation.
3. Incomplete expansion.
4. Wiredrawing.
5. Friction and windage.
6. Radiation and convection.
The unaflow engine was developed to reduce initial condensation loss—one
of the largest—and all of its advantages result from its one-way flow of steam
as opposed to the counterflow of other engines. The gain shown by it over
counterflow engines is better the more the expansion, i.e., the more the difference
in temperature of inlet and exhaust steam.
Incomplete expansion will be minimized if the engine is properly chosen
for the work it is to perform. Engines may carry large overloads continuously,
but should not unless fuel cost is of no importance because large incomplete-
expan .sion losses are brought about by moving cytoff up toward full stroke.
Normal engine load for moderate speed noncondensing engines with cutoff
governing might be based on an apparent expansion ratio of 4. In general, the
greater pi — pa, the larger this should be.
Wiredrawing is avoided by the use of properly proportioned steam passages
and by using quick-acting valves. Leakage past valves is a serious loss, es¬
pecially after the engine has been in service several years and the valves have
become worn. High steam pressures accentuate this loss.
STEAM ENGINE CAPACITY AND PERFORMANCE 367
Friction and windage can account for between 5% and 15% of t^e in¬
dicated horsepower. Radiation and convection from the cylinder are minor
items if the original insulation is maintained in good condition. The efficiency
of an ideal steam engine is the same as for the simple vapor cycle, Eqs 8-3 and
8-4. The actual thermal efficiency equations are the same as Eqs 8-5 and 8-6,
but with heat of the liquid at exhaust pressure used for ftf.
The term “engine efficiency” is frequently applied both to engine and turbine
prime movers to denote perfection of thermod5mamic and mechanical design.
If Ci is the ideal quantity of energy made available by the possibility of ex¬
panding a fluid between specified initial and final states, whereas the actual
energy produced by a prime mover operating between these terminal conditions
is e, then we define engine efficiency as
Tie = c/ci (11-3)
The engine efficiency of steam engines or turbines operating on the Rankine
cycle is the ratio of i;t to i/r; hence
ij. = 2648/(w(fci -ht))* (11-4)
i/e will have an “indicated” or “brake” basis, depending on the basis of w, the
steam rate. Where the unit is direct-connected to a generator and tested steam
rate has had to be left as kg per kw hr, the over-all or combined engine efficiency
is
= 3599 /(v\(hi - hi)) * (11-5)
In both Eqs 11-4 and 11-5, ^2 is the result of an isentropic expansion
to exhaust pressure.
The steam rate of an engine decreases with increasing load to the point
of maximum efficiency, after which it increases because of excessive incomplete
expansion losses. The steam consumption rises with increase of load. This
characteristic generally plots against load as a straight line with i/-intercept,
known as the Willans line. The thermal efficiency curve is an inverted refiection
of the steam rate curve.
Power Capacity, A steam engine expected to have a certain mean effective
pressure will develop internal power which we will designate ihp since engine
indicators provide the only means to measure it. After the deduction of energy
for mechanical friction, flywheel windage, and auxiliaries, the power develops
a working rotating torque available at the shaft. This is sometimes measurable
by dynamometers, especially the Prony brake; therefore shaft power is com¬
monly designated bhp, Eq 6-11 is equally valid for steam as for internal com¬
bustion engines. The equation for indicated horsepower is
ihp = PmepLNnSA/449702 hp (11-6)
or, if pxnep is different on two ends,
ihp » 2(Ap«ap)LJNrn/449702 hp (11-6A)

where ZA = sum of 7 (D* — D^) for head and crank ends, sqom. Other ^m-
4
bols as previously defined.
* hg and hg ara in joides per gram. >648 and 3599 ohange to 982.4 and 859.9 respectively
if hg and hg are in koal/kg.
368 STEAM PRIME MOVERS
Indicators and dynamometers are associated with power capacity of in¬
stalled engines. Occasionally it becomes desirable to estimate the power that
some available engine would develop were it installed in a prospective pur¬
chaser’s plant. There is much traffic in used steam engines. As long as the manu¬
facturer’s design pressure and speed are not exceeded, engines can be employed
on a variety of steam terminal conditions.* Unlike turbines, they are not
adversely affected by use on steam at a pressure different from that for which
they were designed. To decide upon the power capacity of an engine of known
bore and stroke, an estimate of mean effective pressure is required. The value
of Pmep is affected by the shape of the cycle. This shape is primarily the result
of the valve gear employed and the engine speed. However, the effective
pressure of the craventional standard of comparison is independent of mechan¬
ical detail and Pmi is calculable from pi, p2 and r. The ratio of Pmep to Pmi,
called diagram factor, is therefore characteristic for an engine type.
Table 11-1. ENGINE FACTORS
PISTON SPEEDS OF ENGINES

200-350 rpm
High-speed mmple automatic
183-137 m per min
80-125 rpm
Low-speed releasing gear_
229-152 m per min
100-300 rpm
Unaflows.
^ 274-183 m per min

DIAGRAM FACTORS
High-speed, simple automatic. 0.70-0.85
Low-speed, releasmg gear. 0.80-0.90
Unaflows
Full compression, condensing. 0.75-0.85
Full compression, noncondensing. 0.70-0.80
Control!^ compression, condensing. 0.85-0.90
Controlled compression, noncondensing. 0.80-0.85
Example 1: The power capacity of a 80.5 cm. x eicro. single-cylinder Corliss engine,
when employed on 8.8 kg/cmt ga saturated steam is to be estimated. Atmospheric
exhaust.Release pressure, about 1.4 kg/om* ga.,Speed, 100 rpm. Piston rod, 3.8 omno
tail rod. Clearance, 4.0%.
Table 11-1 furnishes mep data. Diagram factor will be assumed at 0.85. With the
usual assumption for saturated steam expansion,

PhVh = Pcf'o; r = Vc/ch = Pb/Po * 9.83/2.48 = 4,02

Pmep
.r _
0;85pnri * 0.85 I 9,83 ( ^
/I + ln4.02\
^ I
.
4.09 kg/om*

2A « * 130.6* + (30.5* - 3.8*)] » 1450 om*

4
4.09 X 61 X 1450 X 100 X 1
ihp 80.4 hp
449702
* Engines designed for. condensing service cannot be shifted to noncondensing unless
there are special provisions for adjustment of clearance space so that overcompression will
not occur.
STEAM ENGINE CAPACITY AND PERFORMANCE 309
Hub engine would be capable of about 71 shaft hp. The apparent ratio of expansion is:
R « 4.02/(l + 0.04 - 4.02 X 0.04) - 4.66
This corresponds to a cutoff of 1/4.56, or 22% of stroke.
The shaft speed may be limited by range of governor adjustment, as well
as by maximum safe rotative speeds of the flywheel. Conventional practice in
steam engine speed is shown in Table 11-1. Direct-connected a-c generators
impose a speed requirement based on frequency of the alternating current. The
formula for speed in terms of the frequency / in dps and the number of electrical
poles n is
N = 120//nrpm (11-7)
The frequency is ordinarily 60 cps; the number of poles, usually one of the
following: 24, 26, 28, 30, 36, 48, 60.
Engine Tests. Field testing of .steam engines may have several objectives,
but primarily testing is to (1) determine the effect of valve action or (2) investi¬
gate the magnitude of output and efficiency.
Throttl*
fcg /em^ ga

Fig. 11-6 Typical steam rates for simple, noncondensing, high-sp^d engine-alter¬
nator sets. Based on dry saturated steam at 8.Skg/eiBi^ ga, aunospheric exhaust. Steam
rate corrections for other conditions: 1.8% per5®C superheat; l%por O.l kg/om» back
pressure; see inset for throttle pressure.

Valve action affects the shape of the indicator card. Correct mechanical
setting of the valves is often possible through a trial-and-error process, taking
indicator cards between each adjustment so as to note the effect. An engine
expert* is able to so interpret the cards he takes from an operating engine
that he can effect the settings of the valve gear linkage sufficient to get the
cycle events of the admission, cutoff, release, and compression correctly timed,
and the power production properly divided between head and crank end of the
cylinder. Also, leakage and lubrication defects in the valve gear are detectable
Years ago expert steam engine mechanics were fairly numerous, due to the large num¬
bers of engines in use. Increasing use of I.C. engmes, turbines, and electric motors has
sharply restricted the field. Outside of the service departments of the few active engine
builders, there are relatively few practicing master mechanics, most of these being ^*oId
timers.” In consequence, the professionally trained mechanical engineer, having mastered
the underlying principles of kinematics and steam power as part of his training, is more
frequently called in on plant engine problems.
370 STEAM PRIME MOVERS
throu^ indicating. Steam engine speeds are low enough that the ordinary
cyclic indicator which draws a p>v diagram is adequate. The p-t indicators so
necessary for the higher speed I.C. engines might have to be used if a takeoff
for stroke motion can not be obtained. Many engines have exposed crossheads;
manufacturers of fully enclosed engines usually provide a built-in indicator
drive.
Output and efficiency testing involves dynamometers or generators to
absorb power, as well as measurement of steam conditions and flow. Flywheel
Prony brakes can be built on the spot, using wooden brake blocks, but can
absorb power for brief intervals only, unless the flywheel rim is constructed
with internal flanges so as to retain cooling water. Frequently the engine will
be direct-connected to an electric generator; then amperes, volts, and power
factor give the generator output. Electrical losses in such cases are not always
separable from the mechanical.

nSOO

-l«00 5

H300 I

-fioo 8
E
o*
HlOO (a

lo

Fio. 11-7 Performance of simple, counter-flow 22 hp steam engine, 7 kg/om» ab'satu-

rated steam; 508 mm Hg condenser vacuum.

The variable load performance of an engine is readily displayed as a graph

of all other quantities plotted against output (bhp, kw, dr percent rated load).
The rated power of an engine is not always obvious. It has not been customary
to placard an engine with a horsepower rating. One arbitrary rating is the
power at 26% cutoff. A better method for engines with cutoff governing is to
take a series of net brake load and speed readings, as dynamometer torque is
increased. Eq 6-11 shows that the power of an engine is proportional to WN]
therefore that dynamometer load at which this product is greatest is maximum
power. A dynamometer load of % of this maximum is a reasonable rated load
point since it allows 25% overload capacity and occurs before a serious re¬
duction of rpm due to the govemer characteristic. Fig. 11-7 represents the
results of a test on a dynamometer-loaded 17.8 cm x 25.4 cm slide valve
engine taking saturated steam at approximately 7 kg/cm* ab and exhausting to a
condenser at 508 mm Hg vacuum. Steam consumption was obtained by
XJNAFLOW ENGINE 371
weighing condensate. Calculation of points for the three-quarters rated load
are the subject of the next example.
Example 2: A 17.8 om x 25.4 om, sini^e-cylinder, simple steam engine, rated at
22.2 bhp and 280 rpm, gave the following test data. Dynamometer arm, 01.4 om; tare,
18.7 kg; 81.8 mm piston rod.
Duration of test 15 min; average throttle pressure,6.38 kg/cmSga average vacuum,
620.7 mm Hg. Harometer, 752.0 mm. Steam quality by open throttling calorimeter with
average temperature reading, iod.7*C. Speed, 280.5 rpm; gross dynamometer load,
60kg. Average HE cards, 3.120kg/en^mop;CE cards, 2.422kg/cm* mep. Condensate
collected, 83.2 kg. Principal items of performance will be computed.
Pressures: Throttle 6.33+1.03 ->7.36kg/om4 ab; condenser,752.0 - 520.7 = 232.2mm. Hg.
Enthalpies: Throttle steam, hi * 630.5 + 0.47 (106.7 — 00.6) i- 642.8 kcal per kg
(nnee hi equiUs enthalpy in calorimeter).
Ideal hi after isentropic expansion to 232.2 mm c 526.5 koal per kg
ht at 232.2 mm s 70.4 koal/kg.
i>\4x280
Power: (Eq ll-OA) ihp - [3.l20xl7.8»+ 2.422 (i7.8* - 3.18*)] ic/4X

» 21.5 hp.

(Eq 6-11) bhp - 2* (BO -13.7) X 91.4 x~|- 16.5 hp.

Percent rating « (16.5/22.2) X 100 = 74.2%.

Steam Rate: Hourly consumption 83.2 x 60/16 = 332.8 kg.
tab » 332.8/16.5 » 20.2 kg.
Efficiencies: ffm = bhp/ihp « 76.8%.
i/t = 632.4/[20.2(642.8 - 70.4fl “ 6.5% based on bhp.
i/e = 632.4/[20.2{642.8 — 526.5)] “ 27.1% based on bhp.
Speed regulation = iN75% — Nioo%)/Niqo% = (280.5 — 280)/280 = 0.2%.
11-4 Unaflow Engine. The steam rate of engines for main power service
is an important factor in purchasing deci¬
sions. Although the slow-speed, multiple-
expansion engines of an earlier day had the
necessary good economy, they were too
bulky, heavy, and costly to survive the
competition of other prime movers. Then
the unaflow engine with steam-jacketed
cylinder appeared, with a steam rate as
good or better than that of a triple expan¬
sion engine, and in addition was reasonably
simple mechanically—and not cumber-
8ome. As this engine type has survived ^ Cylinder section, four-
competition, in fields favorable to it, some valve (Corliss) engine,
details of its construction will be presented.
The object of the unaflow principle is the elimination of initial condensation
loss. In the counterflow steam engine the relatively cold expanded steam washes
372 STEAM PRIME MOVERS
the cylinder walls during the return stroke, cooling them to such an extent that
a portion of the cycle’s admission steam is condensed in heating up cold ports
and walls. The higher the ratio of expansion used, the more there is of this
condensation. It is also aggravated by condensing operation. A counterflow
engine such as the Corliss, Fig. 11-8, with its four valves, eliminates valve-and-

Fig. 11-9 Performance of 500-kw vertical imaflow engine-generator. Based on dry

saturated steam at u.lkg/oms ga. Steam rate corrections for superheat: 0.72% per 6®C.

port initial condensation of the simple slide-valve engine, but suffers it on the
walls and cylinder heads
In the unaflow engine the steam enters the cylinder through admission ports
at the ends. After expansion it is exhausted through center ports as they are
uncovered by the piston High expansion ratios can be used without much
initial condensation, and favorable steam rates can be obtained. The cylinder

Skbtntr Enfln* Co.

Fiq. 11-10 C!ontrolled compression unaflow engme.

of an engine having all exhaust valving done by the piston is shown in Fig.
11-11. Compression begins as soon as the piston has covered the central ports
and a long compression stroke ensues. This is designated fuU comprestion and
tends to produce overly high compression pressures at admisnon unless (1)
the engine is operated condensing, (2) high throttle steam pressure is empl<^ed,
or (3) very large clearance space is built into the engine. Fig. 11-12 will explain
this.llirottle pressure isassumedto beM kg/om* gs.Small clearances are desirable,
UNAFLOW ENGINE 373
but, if as little as 5% is employed, compression is satisfactory only if operating
condensing. The other exhaust cases in a result in overcompression. It is noted
in b that an increase of clearance to 15% cures this difficulty, but leaves com¬
pression inadequate in condensing operation. This explains the reason for
auxiliary clearance pockets that can be opened and closed manually in order
to accommodate the engine to different exhaust pressures. The valve should be
spring-loaded so as to yield and safeguard the engine against excessive com¬
pression should vacuum accidentally be lost when the engine is arranged for
low-clearance operation.
Another disadvantage of full compression is the smaller diagram factor
created by the absorption of energy for the compression.* An engine manufac¬
turer has developed a controlled com¬
pression unaflow which is adaptable
to either of the exhaust conditions
without alteration of clearance. As
shown in c, back pressure and non¬
condensing operations do not result in
overcompression if the point of com¬
pression is delayed for about % of the
return stroke. This action is accom¬ Fig. 11-11 Full-compression unaflow
plished at the expense of some me¬ engine cylinder.
chanical complication, for auxiliary
exhaust valves must be added to create the delay. These must be linkage-
driven since they have to be closed on the working stroke, and open only
on the return. A small increase of initial condensation is to be expected over
the full compression type; however, diagram factor, adaptability to vary¬
ing exhaust pressure, and cost are all favorable. The compression delay is not
needed for condensing operation, and the auxiliary valves automatically remain

Fig. 11-12 Study of unaflow engine compression. Assumed ports covered at 5% of

return stroke. Back-pressure, 0.35 kg/om* ga condensing, 0.35 kg/om* ga, (Hyperbolic
oompresfflon assumed.)
♦ Not significant for condensing operation on account of the slow rise of pressure duiine
the first 2/3 of the stroke.
374 STEAM PRIME MOVERS
closed as long as the vacuum is adequate to prevent overcompression. It will
be noted that the engine shown in Fig. 11-10 is a controlled compression una-
flow.
11-5 Steam Turbine Applications. This most versatile prime mover is
capable of an almost endless variety of applications. It is a practical power
source when built in as small a size as 5 hp or as large as 100,000. It is relatively
quiet and smooth in operation. Its compactness is unexcelled in the high-
capacity region, and it can be operated at rotative speeds which permit direct
connection of generators of economic design. Yet the turbine is not without
disadvantages. It is nonroversible. Steam rates are high unless in condensing
operation. It suffers in efficiency if used on any terminal steam conditions other
than those for which it was designed. Unlike the I.C. engine it is not a complete
power plant, but must be associated with a steam generator.
These disadvantages, notwithstanding, the turbine can be ranked as the
principal prime mover on the basis of currently installed capacity in stationary
power jdants.
Here, as elsewliere in this book, there is an assumption of reader knowledge
of fundamentals of the turbine, principally the character of steam flow whereby
heat energy is transformed into kinetic energy in the nozzles, with subsequent
transfer to shaft torque by means of
the turbine blades. The elements of a
turbine will be recalled by Fig. 11-13.
While avoiding involvement in the
internal mechanical construction of
turbines, we shall investigate meth¬
ods of evaluating the physical state
of the working medium during the
course of its travels through the tur¬
bine.
The principal parts of steam tur¬
bines are these:
Fig. 11-13 Elements of the steam turbine. Rotor. This is the main moving
clement of a turbine. In impulse tur¬
bines it is a shaft on which are mounted wheels carrying the blades. The rotor
of a reaction turbine is a drum. It will be stepped or tapered so as to increase
in diameter toward the low-pressure end.
2, Casing. This is the principal stationary element, often called the cylinder.
It surrounds the rotor and holds, internally, any nozzles, blades, and dia¬
phragms that may be necessary to control the path and physical state of the
expanding steam. The bearings, auxiliaries, and steam lines are attached to
the casing or are an integral part of it. It also is shaped to become the main
frame and support of the assembled turbine. Routing the steam flow through
two or more cylinders is compounding.
3. Bearings. The main bearings of a single-cylinder turbine are two in
number, placed outboard of the shaft seal. Most journals run in plain babbitted
bearings. Some small turbines are ring-oiled from reservoirs, others follow large
turbine practice with pressure oiling systems. Thrust is carried by separate
thrust bearings (plain or ball). Where large end thrusts are produced, as in
STEAM TURBINE APPLICATIONS 376
the case of reaction turbines, they are mainly neutralized by steam-loaded
balance plates on the rotor.
4. Shaft Seals, Where the shaft emerges from the casing it needs sealing—
to prevent steam outflow at the high-pressure end and air inflow at the vacuum
end. On small noncondensing turbines this is accomplished by mechanical
sealing rings; however, these are not too practical if the shaft diameter is large.
Labyrinth glands with steam leak-off at the high-pressure end and steam or
water sealing at the condenser end are employed on all large turbines. Multi¬
stage impulse turbines must also be internally sealed between the shaft and
diaphragms.
5. Steam Control. Flow of the steam through a stationary turbine is usually
regulated so as to produce constant rotative speed in the presence of variable
power demand. This is always the case where the power is used for electric
generation. Control is exercised by varying the quantity and pressure of the
steam flowing through the turbine. If quantity control could be had alone, it
would be employed, but the turbine has fixed-size nozzles and pressure control
is the most practical method of varying quantity. In large turbines power is
varied with minimum throttling by subdividing the first-stage nozzles into
groups which come into action in sequence as load is increased. However, be¬
yond the first stage the entire nozzle group is always in action, and pressure as
well as quantity is variable when power is changed.
In a typical installation the steam line leading to the turbine will contain
one or more gate valves for the purpose of isolating the turbine from the steam
header. These are fully opened when preparing to start the turbine. Next in

Fiq. 11-14 Some types of turbines. Among types not illustrated are: cross-compound,
steeple-compound, mixed-pressure, and double automatic extraction.
376 STEAM PRIME MOVERS
aequence is the turbine throttle valve. This is manually operated and used to
regulate the rate of starting the turbine. When the turbine is warmed and up
to speed, this valve* is left fully open.
An emergency trip valve may either be built into the throttle valve or
become a separate unit following it. This is tripped shut if the turbine over¬
speeds because of governor defect. It can also be connected to act on other
emergencies such as water in extraction lines, generator undervoltage, loss of
oil pressure, etc. Following the emergency valve the steam flows through the
governor valve and into the first-stage nozzles, after which it expansively flows
through the turbine without further control except for two cases: (1) constant-
pressure extraction turbines which have automatically regulated internal valves
in the path of the steam flow at the extraction point; (2) the dilution of
partially expanded steam with high-pressure steam by-passed from the throttle
to intermediate stage points. This by-pass is under governor control and is a
method of carrying load over and above that power produced by full un¬
throttled expansion.
6. Oil System, Oil is required for lubricating the bearings. Most turbines use
the same oil pressure system for both bearing lubrication and governor servo¬
mechanism operation. An integral oil pump, driven from the main shaft,
provides the pressure for oil relays and governor valve-operating cylinders. The
Table 11-2. CLASSIFICATION OF STEAM TURBINES
A. By Size and Application,
1. Auxiliary service, direct-connected, small capacity.
2. Mechanical drive, geared and direct-connect^. Standard units available in range
10-1515 hp.
3. Industrial power. Usually impulse turbines with gear drive to generator in the
smaller sizes. Usual range 100^^000 kw.
4. Central station power; 2000-200,000 kw. Impulse and reaction; direct drive to
generator.
a. Smaller capacities, single cylinder, 1800 rpm.
b. Larger capacities, 10,000-60,000 kw, usually single cylinder, 3600 rpm.
c. Extreme sizes 100,000-200,000 kw, multi-cylinder, 3600 rpm.
B. By Steam Flow.
1. Impulse, impulse-reaction, and reaction-blading.
2. Single stage (smaller capacities) and multi-stage.
3. Re-entry. Partial multi-stage advantages without added mechanical complica¬
tion.
4. Subdivided steam flow in multiple cylinders in parallel or in tandem.
C. By Degree of Expansion,
1. Straight expansion, condensing and non-condensing.
2. Back pressure. Exhaust to heating or process steam mains at gauge pressure.
3. Bleeder. Steam extracted for feedwater heating.
4. Automatic extraction. Steam extracted at controlled pressure for process use.
6. Low pressure. Condensing turbine receiving steam from prior prime mover, say
steam engine.
6. Superposed. High-pressure turbine exhausting at throttle pressure of original
plant (Sec 8-8).
* Large high-pressure units often do not have such a valve, but have a manual over¬
ride on the governor system for use in starting.
STEAM TURBINE APPLICATIONS 377
same oil, when reduced somewhat in pressure, serves for circulating to the
bearings. An oil reservoir, oil filter, and oil cooler are included in this system.
Sometimes a separately driven emergency oil pump is provided, for should the
main oil supply fail on a large turbine, the bearings would be ruined before
the rotor came to rest, even though the fault were immediately detected and
the emergency valve tripped. This is because of the enormous store of energy in
the massive rotor turning at 1800 or 3600 rpm.
Turbines may all have the aforementioned components, yet vary in several
ways other than size. A classification of steam turbines is therefore de.sirable.
This is introduced by Table 11-2 and implemented in part by Fig. 11-14.
Power Rating, Mechanical drive turbines are rated in horsepower; turbine-
generator units, in kilowatts. There is no internal power comparable to the
indicated horsepower of engines, although the product of torque and rotor speed
is internal power. The nominal rating is a declared power capacity expected to
be the maximum load. The capability of a turbine is the manufacturer's
guaranteed maximum continuous output for a clean turbine, operating under
specified throttle and exhaust conditions, with full extraction at any openings,
if provided. The difference between capability and rating is considered to be
overload capacity. Maximum efficiency usually occurs near the rating point
since methods used to secure maximum capability generally interfere with
smooth streamline flow of steam in the turbine. In many cases the rating point
occurs with full first-stage admission, overload being carried by by-passing
throttle steam to downstream stages. This, however, is not always the case and
some turbines have full first-stage admissions at 80-909f of rating.
A common practice has been to design for a turbine capability of 125%
nominal rating and to provide a generator that will absorb rated power at 0.80
power factor. By raising power factor to unity, the generator wilt absorb the
full turbine caiiability. Recent experiences in the central station field, where
new installations are usually large, hydrogen-cooled generators, indicate that
average operating power factors are about 0.85 and that generator capacities
can be raised in emergencies by increasing the cooling gas pressure. It is seen
from the preferred standards for large, 3600-rpm turbogenerators (Table 11-3)
tiiat the overload allowance is 10% and the normal power factor 0.85.*
Standardized Units. Where a line of equipment lia^^ iirospective sah ^ '
different sizes and with variations of operating condition*^, an endless succe-^uPi
of new ^^tailor-made” designs arc posrible, each differing in detail ivom u
othci’S and incurring much individual design, molding patterns, i)eif(':mMue
estimates, etc. After years of manufacturing exiierience, interested parti(‘^ often
formulate standards to (1) reduce first cost to the purchasers, (2) reduce
number of parts sizes to be stocked, and (3) reduce time interval between
contract and delivery.
In the turbogenerator field the smaller sizes (up to 75(X) kw) have b(*en
defined for some time in the standards of the NE^IA. More recently a joint
AIEE-ASME committee has promulgated preferred standards for large 3600-
rpm, 60-cycle condensing steam turbine generators, specifying not only a series
’‘‘As the International Electro-technical Commission employs turbine capability as the
Rating, this difference should be remembered when comparing U. 8. data with international
standards.
378
GOVERNING STEAM TURBINES 391
11-7 Governing Steam Turbines. The electric power system is devised
of elements automatically responsive to variable load. (Sec Fig. 2-8.) One
of these is the prime mover governor. The steam turbine output is controlled
by its governor, which, for stationary power plants, functions to control speed
to nearly constant value. The speed responsive element used is either a centrifu¬
gal weight or hydraulic impeller. These must necessarily produce some speed
regulation. However, modifications may be added which permit manual or
automatic resetting to any required constant speed.
The governing system acts to translate slight speed changes, resulting from
load variation, into steam control. The nature of such control is fundamentally
a throttling process. Its effect on the condition line was described in Sec 8-11.

JTrip- throttlu
volva

oil pump*

btorinpt

lltliff volvt Mf
•t so IP

General Electric Co.

Fig. 11-24 Simple hydraulic relay speed governor.

392 STEAM PRIME MOVERS
The constant enthalpy throttling there mentioned is productive of losses of
available energy, which losses may be partially decreased by substituting for
the single governor valves shown in Figs. 11-13 and 11-23 a more complex
arrangement that reduces some of the throttling in the first turbine stage.
Steam control can be accomplished by one of the following:
1. Throttling by a single inlet valve. Constant enthalpy action. Suitable
for small turbines, even up to several hundred horsepower.
2. Varying the number of first-stage inlet nozzles in action, of which only
the last one opened is throttled at any time. Of course, there are irreversible
actions between the first and the second stages (which take full entry) that
place the condition line somewhere near where it would have been for simple
inlet valve control. More efficient than (1) and frequently used for medium
and large turbines having velocity-compounded first stage.

3. Varying duration of full-pressure puffs {blasts), of which there ar(

several per second. This is practiced on turbines having full peripheral admis*
sion to the first stage, mainly straight reaction types.
4. By-passing. Additional blade torque may be developed beyond thaf
existing at full, unthrottled flow by admitting some throttle steam downstream,
i.e., by secondary and tertiary valves by-passing some high energy steam
around the first stage and following high-pressure stages. This live stean^
'^peps up” the partially spent steam admitted through the primary valve. Bla^fe
thrust in the downstream stages is increased but with greater internal turbulence
and some loss of efficiency.
5. Manual control. The nozzle chest of a turbine controlled as in item 1
may be divided into sections, with entrance to nozzle groups from each section
controlled by a hand valve. At light loads the operator closes one or more of
these hand valves, thereby reducing the total amount of throttling.
Hydraulic relay action predominates in governor systems. The fluid is
lubricating oil, with both governor and lubrication systems supplied from the
same oil pump. Note that the hydraulic governor impeller is separate from this
(Fig. 11-38). The supply to lubricate the turbine and generator is reduced to
OOVERNINQ STEAM TUBBINES 383
ft much lower pressure then is used by the governor servomeehftnism system.
The need for the speed changer has been mentioned. As is seen in Fig. 11-23,
a speed changer is a bias of some sort applied to the governor system. It
usually takes the form of (1) an adjustment of a spring tension or (2) a
modification of the mechanical linkage connecting governor element with the
pilot valve. If this adjustment is made by a reversible motor, then it could be
accomplished while the turbine was running, and from a remote control point.
Thus by controlling the motor to run briefly in one direction, the governor
speed could be raised, while opposite rotation of the speed changer motor
would lower turbine speed. This is needed, not only for paralleling units, hut
also for controlling the division of load between a-c generators operating in
parallel.
The typical regulation of a centrifugal-govemor-controlled steam turbine is
shown by the drooping characteristic, Fig. 11-26. A turbogenerator may have
this characteristic rather flat, but limited by governor hunting, the fault in-

3635 1 5 y
_*o
3630 ^
3625h^
100%
3620n‘
z
a, 36!5 -
ec
3610
36051
3600
5000 10,000 15,000
Kilowotfs

Fio. 11-26 Governor characteristics.

duced by oversensitivity. The speed regulation would be (Si — S2)/Sj. Any

desired speed could be produced by operating the speed changer, the effect of
which is to shift the regulation characteristic nearly parallel to itself up or
down.
ParaUel Operation of Turbo-alternators. This, of course, implies two or
more units which can, by electrical switches, be connected to a common bus.
Now d-c generators can be paralleled on a bus after matching polarity and
voltage, and load can be divided between them at will through adjustment of
the shunt fields. A-c generators must also have a match of frequency and phase
rotation' before it is safe to connect the incoming machine. Frequency being
proportional to shaft rpm, control resides in the governors and speed changer.
Furthermore, when paralleled, the division of load between such machines can¬
not be accomplished by variation of excitation, which only causes more reactive
current to circulate between machines. Load division is accomplished by speed
changer adjustment. The machines must continue to operate in s3rnchronism,
but the one receiving an “increasing speed” signal from the speed changer will
pick up more load from the others. If we have a unit A operating in a system
at 3600 rpm and canying 10,000 kw (see figure), then a sUft of the regulation
394 STEAM PRIME MOVERS
characteristic to A' would necessitate that unit producing 11,400 kw if still
operating at 3600 rpin. The division of load between turbo-alternators is the
subject of the next example.

Example 1: Coiifeidcr two cquipolar alternators. A, 10,000 kw, 0.8% regulation;

B, 15,000 kw, 0.6% regulation. These operate m parallel to carry a system load of
20,000 kw Initially they are dividing the load at 3605 rpm as follows: A, 8400 kw;
Bf 11,600 kw. Let it be required to determine the adjustment needed so that B will
cairy 12,000 kw at 3610 rpm, and A the balance of the 20,000-kw system load. The
adjustments will be expressed m terms of the new no-load speeds.
The characteristic curves A and B are drawn to the prescribed regulation, as
shown m Fig 11-26, it being their slope that is desired, not any particular position
on the spevd-Ioad plane New characteristic B' is now passed through coordinates
12,000 k\\, 3610 rpm, parallel to B Since A must produce 8000 kw, also at 3610 rpm,
line A' IS passed through that point, parallel to A By extending A', B' to the no-load
axis, the new no-load speeds are determined to be* for unit A, 3633 rpm; B, 3627.5 rpm.
A speed indicator index plate could be marked with a scale of no-load speeds produced
at each setting

An emergency governor which is, in fact, an '^overspeed trip valve^^ is an

essential accessory for all turbines, large and small. The trip device is usually
a rotating eccentric weight on the governor shaft, held out of contact with a
stationary trigger by a carefully calibrated spring. When for some reason there
is a failure of the normal speed responsive governor, the speed may suddenly
begin to increase, but then the higher centrifugal force will cause this trip
weight to overcome its spring tension and move into contact with the trigger.
The trigger action then releases a stop valve which is normally cocked open

Euum Co.

Fig. 11-27 Industrial turbines in a paper mill. 2000-kw back pressure turbines with
controlled extraction.
396 STEAM PRIME MOVERS
against a spring force. This action may be inbuilt in a special throttle valve, or
it may be a separate valve in series with the throttle and governor.
To permit speed adjustment, the actual no-load speed of a turbine should
be adjustable between 95% and 105% of rated no-load speed. Within this range
the overspeed trip should be inoperative. The governing of the turbine should
be made sufficiently quick-acting to hold speed rise below the trip point on any
sudden loss of load.
Pressure responsive governors are occasionally used, some instances being:
1. Throttle governors on turbines driving pumps—for speed variation of
imit to produce constant pump discharge pressure or excess pressure, as across
an independently regulated valve in the pump line.
2. Throttle governors on turbines required to exhaust at constant back
pressure. For speed control, unit must be operated in parallel with other syn¬
chronous generators. Load control not possible independent of exhaust line
flow.
3. Extraction governors on Hli^eder turbines for automatically regulated
pressure extraction.
11-8 Part-load Performance. Throttling of incoming steam affects the
work delivered to the shaft in two ways—a reduction of rate of flow and a
decrease of energy conversion per kg steam. Hence a small amount of throttling
can effect a considerable change in power output. When the pressure at the
outlet of the first stage nozzles is less than the critical nozzle pressure, flow is
determined by the fixed nozzle throat area and by the heat release in expansion
to the critical pressure. With straight throttling control.

(11-15)
In this relation (which is founded on constant nozzle throat area) Wi is
the throttled flow ihto the nozzle at part load, Vt is specific volume at the

Fig. 11-29 Longitudinal section of 60,000-kw reaction steam turbine.

part-load performance 397
nozzle throat, and AA the isentropic heat release from nozzle inlet to throat.
Throat pressure is readily determined if first-stage nozzles are divergent,
for pt = 0.55pi, or 0.58pi, depending on whether steam is superheated or
saturated. When the first stage is composed of reaction blading, which is al¬
ways convergent nozzling, the calculation of reduced flow might be based on
the '‘law of constant volumes,” which is: The volume of steam flow into the first
row of blading remains constant over moderate variations of initial steam
state.*
Example 1: The decrease of output caused by initially throttling steam to the
turbine of Ex 1, Sec 8-11, to 18.Slkg/omZab^ will be investigated.
By paralleling the full-load line (as was done in Fig.S-^) t beginning at 18^
ab,81 A0.4 Joules^ the end point atpart load is determined to be Z8IZ JoiJep. Since,the full-load
end point is 8878.5 joules, tHrottling has caused a decrease of energy conversion to
(3149.4—2318)/(814914-2879.5) = 0.963 of the full-load amount.
To find the percentage flow, solve Eq 11-15 for Using MoUier Chart, and
noting that throat pressures are pt = ^^-55 X84-91 =18jl4kg/©in* abjpt'=^0416X18,81 =
10.01 kg/omS ab A/l = 3149,4- 2995.0 = 163.5joiiles; = 3149A-* 2098^5 = 166.8joules Ot at
134Mkg/om> ah,879.4®C|= 0.1844m*; atl)0.0lkg/om* ab,886,7‘*C =0.84816 m*.

2£l _ £1. „
Wi ~Vt'\Ah ~ O-SiBlsV 168.6 ”
Net work delivered to shaft is reduced to 0.963 X 0.754 or 0.727 of full load.f
Generator output would be similarly curtailed
In multiple-valve governing, the flow will be reduced in direct proportion to
the number of valves remaining open, plus the throttled flow through the one
valve to be partially open.
The Willans line, described for steam engines, is a characteristic also of
steam turbines. Test data yield straight lines on the Steam consumption vs
Load plane for full expansion turbines, and nearly straight lines for ordinary
regenerative cycle turbines. For obvious reasons no correlation is to be expected
of data taken from industrial extraction turbines. Any straight line is defined if
two points on it are known. We will now show how to predict a Willans line
by calculating two special points on it.
Engine efficiency, rated load, and steam conditions can be employed to de¬
termine steam consumption at rated load. The consumption at no-load is
estimated from a no-load factor if/j whose unit is kcal per hr. Actually ij/ is no-
load steam flow multiidicd by AK - c bct\veen throttle and exhaust steam condi¬
tions. This means it represents available energy consumed to idle the turbo¬
alternator. When xp calculated from tests is plotted against rated capacity,
which is assumed to be the principal factor affecting it, the data are not too
well correlated, indicating that other factors than turl>ine size affect it. Averag-
*The alteration of steam flow caused by variation of initial steam slate, for both
divergent and convergent first-stage nozzling, as here described for governing action, is
also valid tor purposes of estimating change of turbogenerator rating occasioned by changes
of plant steam pressure or superheat.
t Header will note that the part load line of Fig. 8-23 was constructed by means of
the empirical turbine rules of that section for an assumed 14,500-kw load, and that, in
good agreement, this is 0.725 of full load (20,000 kw).
398 STEAM PRIME MOVERS
ing of test data and consultations with manufacturers led to Fig. 11-30. The
following example predicts a Willans line from preliminary turbine information.
Example 2: The Willans line and its equation will be determined for n 2000-kw
straight condensing, 3600-rpm geared, multi-stage turbine. pi tlM kg/cni*g^, lll^O
super- heat, ps 68.i vam Ug.

Rixted load point. From Eq 8-10, electrical efficiency = 0.936.

Turbine power = 2000/(.936 X 0.7865*) = 2905bhp.
Fig. 11-16 shows an of 0 75 at 2905 hp.
Over-all A/i* . o =(l0e4Ll~ 2145) 0.239 = 224.8 koal/kg.
Work available at shaft =:224.8X 0.75 =168.2 kcal/kg steam.
Rated load steam flow, BD = 2905 X 682.4/166.2 = 10,920 kg per hr.
No'doad point. No-load factor for 2000-kw unit, ij/ = 204,100 koal per hr.
No-load steam flow, OA, = 204,100/224.3 =910 kg per hr.

These two points are plotted as shown in Fig. 11-31, and connected by a straight
line AB. Slope of this line is (BD — OA)/OD = 6 kg per kw hr. It is therefore repre¬
sented by the equation: W = 910 + 5 L kg per hr; where L = kw of generator
load.
1 [ _[_LJ_JL-ULU_i_ 1 1 1 1 111
20
1 M ■n11n
-
■ ■■ ■ 1 1 1 1 ■ 1 ■1 1 1
- ■ ■ >■■■ VI ■ ■ ■
_ 10
— _ ■ i IBS S5 SQBSis
—
- ——

“1 ■ ■ IM s S ■ ■1 8 '
- __ _ ■ ■1 ft £
■ ■ 1 11 ■■ ■ II
■ ■ 1 II B ■i 1 II 4
z ■■ ■ 1 1 II ■ ■1 11 -*
I ■ ■■1 ■ 11 i1 2 7
I
lZ 11 ■■■ 11 r
- 11 1 j ■11 11 a1
■ >■■■1 ■•1
7' ■ ilHBI ■1
•o
1 ■ IHHH ■ ■1
■ ■ ■ II •o
tij ■■ ■ 1 II
cj ■ 1 1 ilihh 1 1 II .4
, 3 2 3 4 ! >84 3 4 j e
lA

Rated Capoerty- kw

Fig. 11-30 Steam turbine no-load factor.

This line is probably a fair representation of actual performance of such a

turbine above 25% load, but the no-load point is an imaginary one obtained
by projecting the flnite load plot to the axis. There have been very few no-load
steam consumption tests of turbines (probably because they rarely operate
below 25% rating), but those that are made on turbines with near-optimum
velocity ratio R indicate a higher no-load steam flow (by 50%) than is implied
by the Willans line.
Characteristic performance curves which may be derived from either the
predicted or measured Willans line are those of steam rate and thermal
efficiency,
* 1 mfitric knrflft paver = 785 6 watte
PART-LOAD PERFORMANCE m
Example 3: The Willans line of Ex 2 furnishes information sufficient to predict
steam rate and over-all efficiency. For instance, at 1500 kw, steam consumption = 910
-I- 5 X 1500 = 3410 kg per hr. Steam rate, Wj^ = 8410/1500 = 5.6 kg per kw hr.
The combined efficiency may be calculated also.
^_859.9 _
^ 5^1(3084.1-178.4)0339*]
Other loads being similarly chosen for calculations, the steam rate and efficiency curves
can be plotted.
JZ

Load -kw
Fig. 11-31 Willans line and derived performance.

The small turbogenerator unit of this example probably would not have any
overload valves, but to show the effect of by-passing some steam around the
first stages to secure turbine capability beyond rating, the dotted overload
extensions are drawn on Fig. 11-31. For a 25% overload the steam rate will be
approximately 5% higher than for straight unthrottled flow without by-passing.
The point C then would be at 1.25 X 2000, or 2500 kw. Its ordinate would be

Lood Load

Singit volvt govtrning with Multipit valvt govorning Voivtt

monuol shut-off on holf of opon in soquonco of 2.3,ond4.
first stogs nozzlos with tomt ovtrlop on ono onothtr

Fig. 11-32 Variable load performance with grouped nozzles.

* factor for conyexting to koal/ka.

400 STEAM PRIME MOVERS
found as follows: at 2000 kw = 5.46 kg per kw hr. Steam consumption at
2600 kw = 5M X 1.06 X 2600 = 14,330 kg per hr. The dotted line BC is an
approximation only.
The analysis of Ex 2 and 3 implies a single governor valve system. If part
of the first-stage nozzles could be closed off by a hand valve for low loads, the
Willans line would be affected as shown in Fig. 11-32. A large turbine governed
by multiple inlet valves has these arranged with overlap, that is, the next one
starts to lift before the last one is completely open. This produces a scalloped
steam rate curve and a Willans line that is really a series of lines with breaks
at the points of successive valve openings. However, unless the line is plotted
to very large scale, the breaks are not distinguishable, and the line appears
straight throughout the normal load range.

Fk;. 11-83 Condensing turbine design s\iitahle for 500-5000 kw.

The Willans line has lost inucli of its foiiner significance under the modem
practice of considering heat rates in regenerative operation rather than straight
condensing steam rates. It remains, liowever, a facile tool for rajiid estimating
purposes.
As load on a turbine decreases, the exhaust pressure extends through the
casing toward the inlet end, until, at no-load, nearly the entire expansion occurs
in the first-stage nozzles and the remainder of the stages are idling in steam
nearly of exhaust or condenser pressure.
11-9 Mechanical-drive Turbines. Several manufacturers build standard
lines of small turbines designed for direct connection to auxiliaries such as
draft fans, pumps, pulverizers, and designed to be used to power miscellaneous
industrial equipment. These are always impulse turbines, mainly single-wheel,
MECHANICAL-DRIVE TURBINES 401
although the blading may be velocity compounded in Curtis or re-entry ar¬
rangement. The power range is about 5-1520 bhp in the single-wheel units, but
ratings of several thousand horsepower are built into multi-stage units. Turbines
are sometimes selected to drive auxiliaries in lieu of motors because of the
benefit to station heat balance of the exhaust steam. The turbine drive is more
readily adapted to variable speed control than is the motor; also, it is safer
in locations where there is some expectation of explosive gases in the atmos¬
phere.
The single-stage units are generally operated at atmospheric or back pressure,
with the exhaust steam used to heat feedwater or in an industrial process. The

Power

Fig. 11-34 Single-wheel re-entry turbines with details of blading. Frequently seen
for the smaller mechanical drives.

steam rates therefore are quite high and such units would be expensive to operate
if their exhaust heat were thrown away. These turbines compare with engines
for auxiliary drive as follows:
Mechanical Drive Turbines Steam Engines
Higher rotative speed is useful for direct Low rotative speed may be an advantage
connection to centrifugal pumps, fans, etc. in some applications. Lower steam rate.
Oil-free character of exhaust steam. Less Valve gear can be adapted to reversible
lubrication and maintenance. More com¬ operation. Usually quieter in operation.
pact.
Mechanical drive turbines are characterised by short rigid shafts and large
blade clearances which enable them to be started quickly. They have spe^
governors and overspeed trips as standard equipment. The speed governor may
be topped by a pressure-responsive element. When this is done, the unit operates
402 STEAM PRIME MOVERS
at variable speed up to the limit set by the speed governor. Variable speed
governing may also be introduced in connection with combustion control.
The lubrication systems are simple. Reservoir and ring oiling is common,
although some have circulating oil systems. Mechanical steam seals are stand¬
ard. Governors can be supplied for almost any desired speed, but 1800 and
3600 rpm are quite common. Where driven equipment is of low speed, the
geared turbine will provide more economical power than will a direct-connected,
low-speed machine. Double helical gears are quiet and have balanced end
thrust. Reduction ratios up to 10:1 are possible, so if the turbine were to operate
at the efficient speeds of 6000-8000 rpm, the drive could be as low as 600-
800 rpm.

Fk;. 11-35 Single-automatic-extraction condensing turbine and section at extraction

gear.

11-10 Turbine Installation and Operation. Basically the layout and

installation of a steam turbine unit consist of the work connected with (1)
foundation; (2) steam connection from high-pressure header; (3) exhaust con¬
nection; (4) extraction steam piping; (5) auxiliary piping (drip and drain,
gland, oil); (6) leads to remote instruments and controls; and (7) electrical
circuits for the generator and exciter.
Small turbines and their driven equipment are customarily mounted on a
rigid cast or wolrU’d bedplate which itself must be properly supported by a
foundation* of the power plant engineer’s design. This can be a solid concrete
block type for noncondensing units, as well as for small condensing turbines
with side exhaust. Large turbines have fore and aft supports at the bearings with
the main casing bridging between as a cylindrical beam. These supports are in
the form of brackets or pedestals and are carried on prepared foundations de¬
signed against specified allowable deflection under load. Condensing turbines
usually exhaust downward and therefore require an open form of foundation,
for space must be available immediately below the turbo-alternator for such
items as condenser, cxiraction steam piping, generator cooler, conduits for
generator electrical leads, etc
See Soc 4-7.
404 STEAM PRIME MOVERS
between turbine and heater to absorb expansion. There may be no exhaust
piping as such, for the condenser inlet flange is frequently bolted directly to the
turbine exhaust flange. Sometimes a flexible expansion joint is inserted at that
point—all depending on the method used to support the condenser weight. Small
piping, some of which may be supplied by the turbine manufacturer, will con¬
sist of:
1. Condensate drains from casing, steam chest, etc.
2. Oil piping for lubrication and governing system. There is a certain amount
of high-frequency vibration in a large steam turbine; hence, care is needed in
the layout of the oil piping. A broken high-pressure oil line could start an oil
fire. Welding is frequently used for joints; also a guard pipe is used over the oil
line.
3. Gland piping for leak-off steam of high-pressure gland and for supplying
water or low-pressure steam to exhaust-end gland.

AUU Chalmen Ufgr. Co,

Fiq. 11-37 Turbine gauges and controls.

TURBINE INSTALLATION AND OPERATION 405
4. Pressure lines to instruments for steam and oil pressures. Much of this
may be supplied by the manufacturer as an integral part of the turbine. If the
plant design requires some of these pressure readings on a separate turbine con¬
trol panel, then extension lines are needed.
Operating Supervision, Fig, 11-37 illustrates the complement of gauges and
controls used with a large compound turbine. Smaller turbines will have fewer
of these, the ultimate in simplicity being the small, mechanical drive turbine
with throttle valve, casing and throttle drain, and steam supply pressure gauge.
The starting of a noncondensing turbine is done by slowly opening the
throttle valve until sufficient steam is striking the blades to overcome static
friction. Then the shaft starts to rotate and the throttle should be adjusted so
that a slow speed (200-500 rpm) will be maintained until pipe lines, casing, and
rotor are thoroughly warmed up. But even before this is done the steam line
should be thoroughly drained of condensate, the oil supply checked in the reser¬
voir, and the driven machine inspected for “ready” condition. The driven
machine should be started in unloaded condition if possible. On small machines
with an oil relay in the governing system, and no auxiliary oil pump that can
be previously started, the operator will need to hold the governor valve open
against its spring force until the shaft-driven oil pump can produce sufficient oil
pressure on the operating piston to take over. When the machine is warm and
drains have been closed off, if oil pressure gauges show normal indication the
throttle may be slowly opened until the speed governor takes over, then com¬
pletely opened. At this point it is desirable to test the overspeed trip, even
though it may mean repeating the starting cycle. Two forms of trip test are*
(1) operate the trip trigger manually, which verifies the valve mechanism but
not the centrifugal actuator; and (2) overpower the governor, or with'speed
changer raise the shaft speed until the overspeed element operates the trip re¬
lease. This tests the entire trip system. To stop a noncondensing turbine one
merely needs to close the throttle valve.
Back pressure turbines are started with the atmosphere relief line fully open.
After the turbine is started and the generator paralleled and connected to the
plant electrical bus, the unit is said to be “on-the-line.” The atmospheric relief
valve is then adjusted to raise the exhaust pressure to its operating value. Auto¬
matic extraction turbines are put on-the-line on a straight condensing or non¬
condensing basis, after which the extraction gear is placed in service.
All the precautions and actions already mentioned are used in starting, con¬
densing turbines, plus several others arising from the need for simultaneous
activation of condenser auxiliaries, or from large size, from high temperature,
or other extremes of practice so often met in central station units. Manufactur¬
ers furnish the turbine purchaser with specific instructions for operation and
supervision. These are covered in a general way by the following:*
Turbine Starting Procedure
1. Start auxiliary oil pump and check oil pressure
2. Check level in oil reservoir
3. Open all cylinder drain valves
4. Open gland leakoff valves
* Power, December, 1945. McGraw-Hill.
406 STEAM PRIME MOVERS
5. Drain condensate from the main steam header and the steam leads
6. Establish circulating-water flow through condenser
7. Start condensate pump
8. Establish seal on h-p gland for starting condition
9. Establish seal on 1-p gland for starting condition
10. Start condenser air ejector and close vacuum breaker
11. Close cylinder drains to stages under vacuum
12. With partial vacuum established, quickly admit enough steam to start rotor
and then shut off
13. Listen for rubs on casing and at seal locations
14. If no rubs are evident, admit enough steam to establish rotor speed of about
200 rpni. Maintain about one-half hour to warm up rotor and casing evenly
15. Trip emergency hand control to check operation
16. Reestablish steam flow and slowly increase speed towards rated rpm during
next 15 minutes. If rotor vibrates severely decrease speed and continue warming
up until no objectionable vibration appears on speed increase
17. Adjust h-p and 1-p seals for operating condition
18. When cylinder condensation ceases close drain valves
19. Turn on cooling water to oil cooler to maintain about 43P c outlet oil tempera¬
ture
20. As turbine reaches rated speed make sure that the governor takes control
21. Place unit on line quickly and apply about 20% load
22. Open bleed-line valves and place heaters in operation

Turbine Shutdown Procedure

1. Reduce turbine load gradually to zero and quickly take the unit off the line
2. Close bleed-line valves and take heaters out of service
3. Shut off steam by manual tripping of overspeed trip
4. Open vacuum breaker
5. Shut off air ejector
6. Check that auxiliary oil pump starts at proper speed
7. Shut off gland seal water
8. Shut down condensate pump
9. Shut off gland sealing steam
10. Open all atmospheric drains
11. Shut off water to oil coolers
12. Shut down condenser circulating-water pumps
13. Keep auxiliary oil pump in operation until unit is cool

Because of close clearances between rotor and casing blades in large turbines,
it is necessary to avoid acts that would cause unequal heating and expansion,
such as passing steam through a turbine or drawing air through the glands with
the rotor at rest. Most of the large turbines have spindle-turning motors that
assist in maintaining even expansion or contraction by turning the rotor slowly
during preliminary warm-up, cooling off, or shut-down periods. The lube oil flow
can be greatly reduced during such periods of slow motion.
The maintenance supervision required during the lon^ periods of time that
central station turbo-alternators are on-the-line consists of a verification of
bearing oil temperature, pressure, and purity; a logging of steam conditions as
indicated by gauges and instruments; plus a corresponding set of generator and
exciter checks. Special instruments such as shaft eccentricity, bearing vibration,
TURBINE INSTALLATION AND OPERATION 407
and rubbing (noise) detectors are frequently installed on large turbines and
interpreting their readings is part of supervision. Careful watching of these items
can usually detect incipient troubles before real difficulties develop. Loading
supervision is generally carried out by remote control from the plant control
room or, in some cases, from the office of the load dispatcher which may be miles
away. Periodic checks of thermal efficiency and load-carrying ability are made,
generally by the engineering supervisor or an efficiency-improvement group
working for him. As long as steam conditions are maintained at standard levels,
vacuum is normal, and steam is clean, there is not much to turbine operation
except watchful waiting. Unclean steam represented by dust carryover from

To gowmof Speed-
control
machamsm
Oilcooltr

Oil strainer^ O'! itromer • Rphef volve

Check valve Check valve
Power

Fkj. 11-38 Oil system of large steam turbine having hydraulic pressure-type
governor.

the superheater may cause a slow loss of load-carrying ability, even vibration.
Eventually deposits may have to be removed by internal washing with water or
wet steam, or by opening and cleaning.
iMbricating Oil. Checking the lube oil has been mentioned as necessary
supervision. This is because the oil may deposit sludge, become acidified, foam,
or form an oil-water emulsion. Although the lubrication of a turbine can be
said to be relatively simple (compared to other prime movers), it is never¬
theless highly important and vital to operation. The turbine is a high-speed
machine. Large units have tremendous rotative momentum, and bearings would
be ruined before the machine could come to rest, even if the throttle were
tripped simultaneously with failure of the lubricating system. Brakes are not
built into steam turbines.
Steam turbine oil is highly refined mineral oil with good antioxidation and
408 STEAM PRIME MOVERS
demulsibility properties. It can be obtained in conventional or fortified forms,
the latter having certain additives for restraining or preventing rust and
oxidation. Less, frequent oil cleaning (4-6 months) is required with the in¬
hibited oils than with conventional (1 month). However, the best of oils
cannot be used continuously without purification. Either the entire charge
can be drained at some time when the unit is shut down or a small percentage
of the oil flow may be continuously by-passed to a purifier. The purification
can be accomplished by filtering, centrifuging, or settlement. When a new tur¬
bine is ready for service, and after a turbine overhaul, the lubricating system
should be cleaned, the oil cooler tube bundle inspected, orifices and strainers
checked to see that no sludge or scale is encrusted. Special flushing fluids are
sometimes run through the system before filling with new or renovated lubri¬
cating oil.
Table 11-5. STEAM TURBINE LUBRICATING OIL
IHrectHMnnected Ring-oiled and
Units Geared Units
Viscosit3^ SSU at 37.8” C 140-170 270-325
Neutralization No.* 0.05 0.05
ASTM Steam Emulsion No.f 120 max 150 max
API degrees 30 27
Cooling range approx. 83°-66“C 83®-66®C
♦ Acid number of ASTM-D663. The mg of KOH used to neutralize 1 g oil sample,
t ASTM.D157.
11-11 Turbine Tests. Turbine tests are commercial, educational, or ex¬
perimental. In the manufacturers^ shops turbine tests are made for power,
valve setting, speed regulation, overspeed trip setting, and running balance.
From time to time commercial tests may be made by the operating company
to determine what increase of steam rate or heat consumption has followed
operation of the unit over a certain length of time. Experimental testing is
carried on in manufacturers’ shops and in research laboratories for the purpose
of checking new designs or of improving the old. During the. initial period of
operation, the purchaser of a turbine may conduct an acceptance test to verify
any or all of the guarantees made for the unit by its manufacturer. The ac¬
ceptance test, as ordinarily performed, is a check on output, steam rate (or
heat consumption), speed regulation, and overspeed trip setting. The ASME
Test Code jor Steam Turbines has been a standard for acceptance tests. This
code covers the testing of turbines for (1) thermal economy characteristics
and (2) capacity; and provides specific instructions for the testing of three
classes of turbines, i.e. (1) complete expansion turbines, (2) reheating turbines,
and (3) regenerative turbines. The load on a turbine generator is defined in
the code as the net output of its generator. Net output = Electrical output
of generator (kw) minus that portion of the excitation power that is separately
supplied (kw) minus power for ventilation separately supplied (kw). The
total steam is determined in the case of a turbine exhausting to a surface
condenser by weighing or measuring the condensate.
The readings of commercial steam flow meters connected to orifices in the
steam line leading to the turbine may be used if the meter is checked and veri-
TURBINE TESTS 409
fied before the test and if the usual tolerance of these instruments, 1.6%
of full-scale reading, is permissible for the test. For noncondensing turbines
the steam flow meter or weighing of boiler feedwater are ways to get steam
consumption. Since steam turbo-alternators are close-coupled, permanently
installed units, it is not practicable to employ any sort of dynamometer load
as a means of measuring output. The generator is a dynamometer of sorts.
However, the instruments customarily installed give generator output; hence,
unless data on electrical efficiency are available, turbine performance cannot
be separated from that of the whole unit.
A test wherein the variations of steam consumption, steam rate, combined
efficiency, and engine efficiency with load are determined for a straight con-

Fig. 11-39 (connections for small condensing turbogenerator and auxiliaries. Special
instructional installation.

densing unit would present no unusual problems. If the unit were an extract-
ing or regenerative turbine, and if the testing were to include determining
the state of steam at extraction points, as well as extraction flows, the field
work would be much more of a job. Steam condition is readily and easily
determined by pressure gauge and thermometer where it is superheated. In
the saturated stages, it is pressure and quality that must be measured. Test¬
ing goes very well with throttling or separating calorimeters until the point is
reached where casing pressure is too low to operate the separating calorim¬
eter. The reader will recall that the principle of this calorimeter includes flow
measurement through an orifice as a function of the high-side pressure. This
requires the low side to be less than 58% of the high side, and for reliability
it would be well to limit use to pressures greater than twice condenser pressure.
Of course, the calorimeter discharge must be piped tightly to the condenser for
410 STEAM PRIME MOVERS
all casing pressures less than about twice atmospheric. The presence of steam
jet air ejectors, discharging some of the condensed jet steam back to the
condenser, and the possibility of ejector condenser as well as main condenser
tube leakage are further complications if condensate weighing is to establish
steam flow.
Example 1: Fig. 11-39 represents a 75-kw, condensing, geared, steam turbo-
alternator employed educationally in the author’s institution. During a certain* scries
of tests data were taken which led to the calculation of the following performance
characteristics at approximately %, and full-load. The average data for each
of the test runs and the derived calculations are shown in Table 11-6. The tests were
made to obtain variable load,data for (1) Willans line, (2) steam rate, (3) combined
efficiency, and (4) engine efficiency. Performance curves showing these items against
percent load can be plotted from the tabular calculations.
Preliminary static test with turbine exhaust valve closed, auxiliaries in operation,
and full condenser vacuum established was undertaken in order to find the ^Hare”
condensate flow. This was later subtracted from condensate weighings made during
load tests.

Manufacturer's guarantees. The purchaser invests large amounts of money

in a turbogenerator. He naturally wishes to hold the manufacturer to fulfillment
of claims for performance, capacity, etc. Therefore a guarantee is required and

Table 11-6. TESTS OF Tfi-kw, CONDENSING, STEAM TURBO-ALTERNATOR

TEST RUN NUMBER

TESTS
1 2 3 4

Elapsed time, min 30 30 30 30

Load fraction, approx X 1
4 i t
Initial steam pressure, kg/om* ab 18.87 13.01 13.01 13.57
Initial steam temperature, ° C 222;2 208.3 220 220*6
hif koal per kg 684.4 676.1 683.9 683.3
Observed vacuum, mm Hg 693.9 682.8 690.9 692.2
Vacuum, corrected to 15.6*C 691*7 679.7 688,8 691.4
Barometer, corrected to 16,6®0 747,0 748.3 750.6 748.3
Exhaust pressure,mm Hg abs 64.4 68.6 6i.V 66.9
Exhaust sat. tempera tuie, *0 39.65 44.05 41.95 40.55
Weighed condensate, kg 217.7 291.7 362.9 426.8
Condensate flow, kg i)er hr 435.4 583.4 726.8 863.6
Tare flow, kg per hr 54.4 54.4 54.4 54.4
Net flow (= throttle flow) kg per hr 381 529 671.4 799.2
Load, kw 17.6 34.1 55.0 70.4
Load fraction (of 75 kw) ! .235 .45:) .734 .938
Power factor 1.0 1.0 1.0 1.0
kg per kw hr 21.65 15.52 12.21 11.35
hfi at exhaust, koal per kg 39.67 44*06 41.94 40.56
Entropy of throttle steam 1.59 1.57 1.59 1.58
h2 (at p2 ^nd S\) kcal per kg 496.6 495.6 498.3 493.9
Vc = S6M/(w)c{hi - /?f;)) 6.2 8;8 10.9 11.8
T?ec = 860*9/(w^k(^l — /i2^) 21.1 30.7 38.2 40.0
.1
TURBINE TESTS 411
furnished which may cover many items,* but economy in the use of steam is
usually the purchaser’s main interest. This may be a guaranteed steam rate,
over the load range, or a guaranteed heat rate with specified extraction.
These guarantees are made for specified design values of throttle pressure,
throttle temperature, and exhaust pressure. However, these operating condi¬
tions may not be exactly achieved by the steam generator and the condenser at
the time of testing the new installation. Conditions may be either more or less
favorable and result in a test steam rate that w'ill need correction before it can
be legitimately compared with the guarantee. Hence the contract for a turbine
should include the basis to be used in establishing the facts of compliance.
Fig. 11-40 illustrates typical correction factoj^ which the manufacturer will

Fig. 11-40 Typical correction factors for steam turbine having performance guaran¬
teed at 42*2 kg/cin*ga,426*7*C throttle steam,38.1 min.Hg exhaust pressure, at rated
steam flow.
include in a turbine contract. Better steam or heat rates are provided by pres¬
sure, temperature, and vacuum higher than design specifications, and vice
versa. A test steam rate lower than guarantee might actually represent a
failure to meet the guarantee, if the test conditions were more favorable to
efficiency than those specified in the contract. How the test results will be
corrected for comparison with the guaranteed rate will be illustrated in the
following example.
Example 2: A straight condensing steam turbine was tested for steam rate at a
loading producing a rate of flow for which the corrections in Fig. 11-40 apply. Over
a 2-hr test the average load was 7515 kw, and the steam flow to the throttle was
64,180 kg. Steam at throttle, 42.9 kg/tmi> gfi^415.6‘’C; average exhaust pressure,44.5 thiii
Hg; operating conditions, steady. The guarantee at this load was 4.24 kg per kw hr.
Was it met?
From the correction curves one finds that the actual test conditions should produce
a steam rate that is 0.5% better because of overpressure, 1 3%; poorer because of
undertemperaturc, and 0.8% poorer because of higher back-pressure. These net
1.6% poorer steam rate than had the design conditions prevailed. A “poorer’' steam
rate is a higher steam rate; consequently the tested steam rate should be decreased
by 1.6% for comparison with the guarantee.
Test steiim rate = 64,180/(2 X 7515) = 4.28 kg per kw hr
Corrected steam rate = 4.28 — 0.016 X 4.28 = 4.21 kg per kw hr

This meets and slightly exceeds the guaranteed performance.

A better performance than contracted for means savings to the operator,

which is usually rewarded by a bonus to the manufacturer. Conversely a
♦ Maximum safe pressure and temperature, maximum capacity, extraction, etc.
412 STEAM PRIME MOVERS
poorer steam rate entails a penalty charge. A full set of correction curves
covering the load range would exhibit multiple curves, like those dashed in the
figure, for the purpose of verifying part-load guarantees.
In the absence of detailed correction factors such as these, engineers must
use arbitrary corrections like 1.8% per io®C difference in superheat, 1.5% per
10% variation in initial pressure, and0.l6%per 1 mm vacuum difference.
11-12 Vacuum. Vacuum is a subatmosphcric pressure, measured as the
pressure depression below atmospheric. (See Fig. 1-3.) Vacuum is created in a
closed chamber if the gas or vapor whose molecules are responsible for the
internal fluid pressure is partially withdrawn. To withdraw a permanent gas
much work of compression must be expended on the gas, but water vapor may
be readily removed with low energy expenditure by first condensing it. One
kg of water vapor at 0.07 kg/cm* aboccupies 20.92 m? whereas after condensat¬
ion it is only0.001 m*of liquid. This stupendous shrinkage accomplishes two
practical results. First the liquidating of most of the vapor molecules creates
a region of emptiness, thus a vacuum; second, but little pump work is needed
to elevate the condensate from vacuum to plenum.
A steam condenser is designed to accomplish this action. It is able con¬
tinuously to swallow large volumes of low-pressure vapor and, with the aid
of a flow of cool water, “blot up” the heat, effect condensation, and maintain
a high vacuum at the steam inlet. Now all this has a beneficial effect on the
conversion of heat into work in the prime mover. If the working expansion can
be continued below atmospheric pressures, extra work energy will be wrested
from each kg of steam. Notice how much of the total energy converted
into work per kg of steam, shown by the typical condensing turbine lines
(Fig. A-13), is released in the vacuum region. For example, line A represents
a 20,000-kw turbine with nearly 9000 kw^ of this d(‘veloped in the vacuum region,
thereby being made possible by condensing operation.
Condensers arc used with both steam turbines and engines, although es¬
pecially with turbines because their thermodynamic advantages occur chiefly
in low-pressure expansions. Engines are not ordinarily oi^erated at exhaust
pressures lower than a660 inm*Hg vacuum, but turbines employ condensers de¬
signed to produce up to as much as 749 mm-Cylinder dimensions are the limiting
factors for engines because, being of positive displacement, they must provide
piston displacement equal to exhaust steam volume.
Condenser vacuums are measured with mercury manometers or with vac¬
uum gauges having dials calibrated in mm -of-mercury units. In the measure¬
ment of condenser vacuums great precision is required, both with manometer
and barometer, for the absolute pressure, on which the steam proi)erties depend,
is extremely small—the result of subtracting two nearly equal numbers, such
as 761.5--740.2. At high vacuum, 2.5 mm. Hg pressure difference accounts for
about 0.5 *0 change in saturation temperature, anfU.75kcaJ available heat. As
16.4®Ctemperature expansion alone will vaiy a mercury columnby 2.5 mm, it
is seen that barometer and manometer columns must be corrected to a common
temperature before they can be subtracted to give absolute pressure. The
common temperature to use is the one for which the linear scale on the ba¬
rometer is correct. It will be noted that the mercury pressure entries in the
published steam tables are based on 0°C mercury.
CONTACT CONDENSERS 413
Linear correction to mercury column=s(d:40ff XlO'*^X1.8mm(ll-16)
where At is temperature difference of the correction, phis if observed tempera¬
ture is lower than standard, mirms if vice versa; and H = nmiHg observed
height of mercury column.
Example 1: The absolute pressure of a condenser will be determined from the
following data. Barometer, T49.8 mm at ; manometer, 707,4 mm Hg at 82.8M
Both will be corrected to 0*C .
True barometric pressure = 749,8 4. (-.80) x 749.8 x 1.8 x lO-* ■■ 746.6 mm
True condenser vacuum = 707,8-1- (—82J2) x 707.4 x 1,8x10“^ • 708.8 mm
Absolute condenser pressure = ^ 43,8 mm
Absolute condenser pressure = 48‘3X" ■■ 0,0689 kg/om* ab
Example 2: Suppom a vaomim is reported to a 16.6*C (762 mm) barometer eolomn
as 698,6 mm. We will now see what this would be if referr^ to a 760 mm column.

Absolute pressure in 15.6^ mercury units = 63.60 mm

This same pressure would be inchoated by a 0®C column of height: 63*6 +
(-15.6)x 63.6 X 1.8 X 10'«-i 63-40mm.
Therefore the vacuum referred to 0®O barometer = 760->68.40 696,60 mm
Air entrainment in the condenser has its effect upon the vacuum. The ad¬
dition of air lowers the vacuum possible with a fixed mixture temperature.
By Dalton^s Law, in a mixture of gas and vapor at a given temperature each
constituent contributes to the observed pressure the same amount that it would
have exerted by itself at the same temperature had the other not been present.
That is, the condenser pressure
Pc = Pv 4- Pa (11-17)
The application of this law will now be illustrated by example.
Example 3: Suppose the measured temperature of a steam-air mixture in a
condenser were 86.7®C , whereas the vacuum, referred t() a 768.0-inm barometer was
709.9 mm Hg. The weight of entrained air can be found as follows:
Mixture pressure, — (762—709.9)(1.033/762)—0.0706 kg/om* ab.
The vapor pressure at 86.7®C (tables), = 0.0629 kg/om* ab
Partial pressure of air, = 0.0077 kg cm* ab
Vfr at 36.7*C (tables) = 23.098 m* This is also v,^.
From w = pv/RT\ w, = ^ kgair per kg steam.
Note: Had the mixture pressure been erroneously used as saturation pressure, it
would have indicated a temperature of 88,8*0, or 2,1® too high. However, a 0.196:1
air-vapor ratio is far richer than would be encountered in condenser practice, for
which a ratio 0.0004:1 would be more normal. Hence in normal operations, condenser
pressure can be considered to determine mixture temperature.
11-13 Contact Condensers. Condensers can be divided into contact and
surface types. In the contact type, exhaust steam and condensing water are
actually mixed together. Heat transfer is by direct conduction. The steam
quickly condenses in the water, and the mixture is continuously withdrawn
414 STEAM PRIME MOVERS
from the condenser as a stream of warm water. Surface condensers interpose
a wall between steam and water, through which heat must be convectively
transferred. The condensate and condensing water are separately withdrawn.
The contact condensers (jet condensers) may be further subdivided into
those from which the water is extracted directly by pumps, those from which
it is forced by the ejector action of a high velocity jet, and those from which
it flows by virtue of a head of water maintained in a “tail” pipe. The first is
exemplified by the ordinary low level condenser, the second by the ejector
condenser, and the third by the barometric condenser. Elements of the jet
condenser are: (1) nozzles or distributors for the condensing water; (2) steam
inlet; (3) mixing chamber; (4) hotwell; and (5) in some cases, a diffusing
chamber or a tailpipe.
Mixing chambers are designated as parallel flow or counterflow, depend¬
ing on whether the steam and water move in the same direction before con¬
densation or whether the flows are opposite. Both chamber types are built in

Mox Wottr Lfvel 1 Condenser Shell 10 Removal Pump Submerged

Above Weir 2 Injection Water Inlet (Meosured Vertically)
3 Air and Vopor Outlet 11 Min. Eff Height of Toil Pipe
4 Steom Inlet Required for Grovity Droinogi
5 Water Outlet 12 Water Distributing Meons
6 Toil Pipe 13 Removal Pump Suction Pipe
7 Seol Tonk 14 Toil Pipe Submergence
8 Seal Tont( Drain 15 Vocuum Breoker
9 Removol Pump Air’Cooler

Borometric Type Low Level Jet Type

Fig. 11-41 Examples of contact condensers. (Courtesy Side, Heat Exchange

Institute,)
CONTACT CONDENSERS 416
both the barometric and low-level condensers. We here show a low-level con¬
denser with a parallel flow chamber and a barometric condenser with counter¬
flow.
In contact condenser terminology, the condensing water is called injection
water. In the low-level type having a removal pump, the injection water may
be induced into the mixing chamber by vacuum head. Part of the water lift
to the high-level (barometric) condenser is similarly generated, but a booster
pump will usually be needed on the injection water. The vacuum will provide
a theoretical suction lift,hi metres of water,of 13.596x10"^ the condenser vacuum
in in. Hg. Only 86% of this can be reliably counted on for lift, and from this
must be deducted the static heads shown in the figure, plus all friction and
water velocity head in the injection water line. Any resulting negative net
head must be made up by a booster pump. The height of setting (dimension 11,
Fig. 11-41) ought to be at least equal to (0.012 times mm maximum expected
barometric pmssure + 1.5241 n*
The head on the low-level condenser removal pump consists of the external
static head, the friction and velocity head in the pipe line, and the vacuum
head. From this should be subtracted the submergence head* to obtain net
pump head.
The water used for injection may be water previously removed from the
condenser if it has been cooled in the meantime. Usually injec^^ion water is
taken from some available natural source such as a stream or river, is used
only once, and is unfit for use as boiler feedwater. The submergence volume
in the tail pipe seal tank (dimension 14) should be at least 60% greater than
the water volume contained in the entire tail pipe.
Compressors for the removal of noncondensable gas are attached to open¬
ings marked 3, for unless this gas is continuously expelled its accumulation
will result in rich air-vapoi mixtures, thereby reducing the possible condenser
vacuum. To the normal sources of air inflow to the low-pressure steam (to be
mentioned later) there must be added the air released from injection water.
These combined air quantities form the load on the air removal apparatus.
The vacuum obtainable will be governed by outlet water temperature
which, in turn, varies with the amount of condensing water used per kg steam
and its initial temperature. Although it is theoretically possible to heat the
water to the saturation temperature of the steam, this is not practicable and
terminal differences exist. Terminal difference is saturation temperature corre¬
sponding to the pressure at the steam inlet minus outlet water temperature. This
is governed by condenser design and operating conditions. The Heat Exchange
Institute recommendations for condensers with vacuum pumps are shown in
Table 11-7. Ejector type condensers operate on g.T®—8.3^ terminal difference.
Table 11-7. MINIMUM TERMINAL DIFFERENCE
FOR JET CONDENSERS, ® F
Condenser pressure mm Hgabs. 25«4 38.1 50 8 76*2 101
Counterflow mixing 2.8 2.8 2.8 2.8 2.2
Parallel flow mixing g'l 3^9 2.8 2.8 2*8
Recommendation of the Heat Exchange Institute is t6 deduct0.9lDt l^than tte
actual submergence head.
416 STEAM PRIME MOVERS
The ejector type condenser is a low-level type. The water pump operates
with water under pressure instead of vacuum. This type is shown in Fig.
11-42. Discharge of water from this condenser is created by the kinetic energy
of the water jets it employs. However, this also means that condensing water
requirements are steady at maximum flow rate, regardless of quantity of
steam being condensed. Condensed steam and air both are entrained by the
multiple water jets and removed without further use of air pumps or com¬
pressors. This figure also shows methods of safeguarding the prime mover
should water pumps fail. Uj^on any water accumulating in the mixing chamber
a vacuum breaker float will ride up on the water, opening the chamber to the
atmosphere. This would be followed automatically by opening of the atmos¬
pheric relief valve, permitting the prime mover to exhaust noncondensing.
Furthermore, the path of water backing up toward the engine or turbine is
barred by a nonreturn or check valve.

Fig. 11-42 Reciprocating engine with low-level ejector condenser.

If exhaust steam enters a condenser at the rate of kg per hr at a condition

represented by fractional latent heat X2htg^ then the heat to be absorbed by
the injection water is (X2hte^ + W) Kcal per hr, where is the terminal
difference. Calling injection water per hr, w^y and its rise in temperature to — ^i,
the heat it absorbs is w^{to — U), The equation of heat released to heat ab¬
sorbed gives:
y^w(to ti) = WB{^x2hfg^ td )

Let be the imit water flow, in kg per kg steam, i.e., then

Wyr' = (X2hfg, + id)/{to — fi) (11-18)
The quantity X2htg^ is the exhaust enthalpy of the prime mover, less the
heat of the liquid at saturation pressure. This may be determined by the prime
mover efficiency studies which have been previously explained, although the
manufacturers of condensers have concluded to use the constant 628* for X2htg^
* Keal/kg.
SURFACE CONDENSERS 417
for turbine condenser design, and 556^ for engine condensers. This trade prac¬
tice applies likewise to surface condensers, although Eq 11-18 does not.
Example 1: A barometric jet condenser is being planned for a condensing opera¬
tion involving iggo kg per hr of engine exhaust steam. Desired vacuum, 67S lam. Hg
referred toW mm .Water to be cooled to HHJ and recirculated. Maximum recorded
barometric pressure, 750.6 muuSome leading specifications for this equipment may be
formulated from the data of this section.
Assuming counterflow chamber, = 2.8®0. Saturation temperature at 708—673
mm Hg = 40.8*0. Discharge temperature, = 49.2 — 2.8 ^ 46.4*C. Make the
assumption of 568 for ; then
= (556+2.8)/(46.4-82) «>88j9 leg water per kg steam.

Wy, = 38.9 X 1590-181.850 kg per hr.

Effective height of tail pipe = 0.012x750.8 + 1.524 m 16.82 m.
Assume /12 + -f (Fig. 11-41) =1.07 m velocity head and pipe friction 0.48 m
and Ai = io.52 + 1.83* m
Net pumping head = 0.86(18.598xl0*‘>x873)-(1.07+0.48+12.86)——1^1 m.
This implies that a booster pump will be needed for 8.1 m of the lift.
11-14 Surface Condensers. The surface steam condenser is invariably an
essential unit in the steam-turbine-powered central power plant. It is also
frequently applied to turbines elsewhere. Together with the steam generator
and turbo-alternator, it forms the central triumvirate of any vapor cycle
system. These three units are the
largest, and most of the remaining
mechanical equipment serves as aux¬
iliary to one or the other. The sur¬
face condenser is an airtight shell
enclosing a highly concentrated heat
transfer surface in the form of a com¬
pact bundle of small-diameter copper
alloy tubes. A condenser for a large
central station turbine will have thou¬
sands of these tubes from 3 to 7.6
m long. The other components of a
Fig. 11-43 Components of a surface con¬
condenser are those elements which
denser. (1) Steam inlet. (2) Outlet to air
enclose and support the heating sur¬ ejector. (3) Condensate outlet. (4) Con¬
face, direct the condensing water flow, densing water inlet. (5) Condensing water
admit the steam, and collect the con¬ outlet. (6) Water box. (7) Water box
densate. These components are iden¬ cover. (8) Shell. (9) HotwelL (10) Tube
tified and schematically illustrated in sheet. (11) Tub^. (12) Tube support
Fig. 11-43. The condenser is installed plate. (13) Inspection doors.
close up under the turbine, frequently
with its long axis transverse to that of the turbine. The shell is welded up in
plate steel construction, although formerly cast iron was the rule.
Attached to the bottom of the shell is a collection chamber for condensate,
called a hotwelL In central station practice this will usually be arranged for
* An tisiiumed static head, verifiable after initial layout is completed.
t Iml/kg.
418 STEAM PRIME MOVERS
de-aerating and reheating the condensate. Steam lanes are left through the
tube bank to provide direct steam flow to the reheating hotwell.
Whenever the condensate temperature is less than the saturation tempera¬
ture corresponding to steam inlet pressure, the difference is termed condensate
depression.
Surface condensers are classified as horizontal or vertical by the position
of their tubes. Horizontal are the more common. They are single-pass or two-
pass, according to the number of times the condensing water passes the length
of the condenser. The diagram illustrates single-pass. In the two-pass, water
enters and leaves the same water box, which, of course, must be divided by
a partition. Condensers are also classified by the shape of the shell, for example,
cylindrical, oval, U-shaped, rectangular, etc.
Condensing Surface. The tubular heat transfer surface is composed of copper
alloy tubes of small diameter. The sizes in ordinary use are included in Table
11-8. The tube size is its external diameter. Practically all condenser tubing
in use is of either 1.65 or 1.24 mmwall thickness. A great deal of research has
gone into the metallurgy of these tubes. The mixture developed by the British
Admiralty office, copper 70, zinc 29, tin 1, had come into general use in sta¬
tionary plants by 1930, but stepping up of some of the operating limits re¬
vealed weaknesses which have since been met with newly developed alloys.
Corrosion, de-zincification, oxygen pitting, and grooving were encountered—
serious because of the high cost of tube replacements* and loss of use of plant
capacity during replacement. Among the newer alloys, some of which are
patented, are arsenical Admiralty, antimonial Admiralty, aluminum brass,
and cupro-nickel mixtures.
Table 11-8. DATA FOR COMMON SIZES AND GAUGES
OF CONDENSER TUBES

Tube Size Wall Thickness Cross Section Ext. Surface Flow at

Ext. Dia. 1 Int. Dia. Area of Bore 0.6m|seo Hues/
(nun)
(mm ) BWG wntn (emS) (iii8) ((min )

I6.S 16 1.65 19.S7 1.941 0^X6657 9.799

18 1.94 19.99 1.408 0.0I96S7 6JU
19.1 16 1-66 16w7S ISSB 0.060867 6.844
18 1.24 ie.S7 9.156 0.0BIIM7 6469
ns 16 1.S5 19.99 9.611 9.498
18 1.94 19J4 9.060 0.069806 9.160
UA 16 1^ 92.19 8.886 ojnoToo 11J06
18 1.94 22.99 4.196 0.079796 19699

The tubes in a condenser are all the same length and are arranged in a
bundle having steam entry lanes created by the omission of certain groups of
tubes. The difference between performance of two condensers of identical
surface would usually be traceable to the effectiveness with which these lanes
’^The individual tube is not too costly, but there are so many of them in a large
condenser. A 50,000-kw unit’s condenser may containSl Ummm of tubing.
SURFACE CONDENSERS 419
let the steam flow easily to all tubes. With adequate steam lanes each tube
performs its share of the condensing job. Fig. 11-44 indicates some of the various
plans that manufacturers have evolved to accomplish this.
Tubes are fastened into tube sheets at their ends and loosely supported by
one or more bitermediate plates. The shell being steel, and the tub^ copper
alloy, differential thermal expansion is encountered. This is accommodated by:
1. Rolling tubes tightly into one tube sheet, packing in the other.
2. Rolling tubes into both sheets, with flexible joint between one sheet and
the supporting shell.

Power Plant Engineering

Fig. 11-44 Various arrangements of condenser surface are indicated by these condenser
tube sheet layouts.

Inlet-rolled and flared Outlet-packed In1et-i>acked

Fig. 11-45 Examples of tube fastening.
420 STEAM PRIME MOVERS
3. Fixed tube ends but tubes installed bowed.
4. Both ends packed.
The official surface area is the external tube area. Let be this surface
per linear metre of tube. Then for n tubes, each L m long,
Av == OvtiL sq m (11-19)
Each pass* of the condenser must present sufficient cross-sectional area
to carry the condensing water flow at the assigned velocity.

Wyr = 3600 X lO^nOwVw kgper hr (11-20)

where = Flow of condensing water, kg per hr.
n = Number of tubes per pass.
Ow = Tube cross-sectional flow area, sq metre.
Vw = Water velocity, m/scc.
To determine the number of passes assume that the condenser is one pass.
Then get the tube length L from Eq 11-19, having first found the necessary
transfer surface. If L is larger than would be convenient or possible, due to
construction or available space, a two-pass arrangement is employed. Tube
length is half, and the number of tubes double, that of an equivalent single¬
pass condenser.
Because of the practice of segregating some of the tubes surrounding the
air off-take, for the purpose of temperature depression of the air-vapor mix¬
ture taken by the ejector system, the total surface provided may be 5% to 20%
more than that calculated for condensation purposes. This segregation is ac¬
complished by internally baffling the tube bundle to provide a rather long
contact flow with the air-vapor mixture about to be ejected. The absolute
pressure under this baffle is the lowest anywhere in the condenser; hence the
noncondensable gas naturally flows in that direction.
Heat Transfer, It will be noted that the subject of condenser heat transfer
has been extensively covered in Sec 9-10. The reader is expected to employ
that background for the present treatment, as well as Sec 9-8 for mean tem¬
perature difference.
q = Heat transfer, kcal per hr.
U = Over-all coefficient of conductivity, based on Ay.
ho = Enthalpy of exhaust steam at condenser inlet.
Af«. = Enthalpy of liquid water at hotwcll temperature.
= Saturation temperature at condenser inlet pressure, ® C.
tif to = Condensing water temjieratures, in and out, ®C.
tVnyWw = Steam and condensing water flows, kg per hr.

♦Portion of the tube bundle through whieh all the condrn-ing waior flows.*
SURFACE CONDENSERS 421

r* Foster Wheeler Corp.

Fig. 11-46 Section through reheating hotwell and tube hank of large inward-flow
fcjurface condenser.

The equations to be employed, in addition to those previously presented,

when determining water flow and surface area are:

q = Wt{ht — hte) = w^(to — <i) kcal per hr

Since c « 1 for water at ordinary temperatures,
Ww = - hu)/{Us — U) kg per hr (11-21)
Ar * q/UB sq metre (11-22)
* q/{q/Ay) sq metre (11-23)
Eq 11-22 would be employed when using statistical methods (Fig. 9-10), or
a theoretical calculation based on Eq 9-9.
422 STEAM PRIME MOVERS
Eq 11-23 represents the method of converting a solution of Eq 9-19 into
surface area. In any event the mean temperature difference 0 v*ill involve use of
tif and ^2 may follow calculations on end conditions of the turbine steam
expansion, but manufacturers are said to use the a8sumed528kcaA perkg^ pre¬
viously introduced, for (^2 — Afc).
If sea water is used for condensing c 1, and density ^ 1000 kg/m* Inste^*
ad,0.94 and 1024*
The terminal difference ta** is the difference between steam saturation tem¬
perature at condenser inlet pressure and condensing water outlet temperature.
id to ai-24)
Terminal difference employed is an economic matter. Higher vacuums can
be carried with smaller but larger condensers. In practice, appears to
range from 2.8 to 5.6®C.The Heat Exchange Institute indicates that perform¬
ance cannot be exactly predicted if ta is less thang.Sn islessthan 18 mm.
Hg, or if is less than 0.9 m/*ec.
Example 1: The principal specification of a two-pass condenser (n, L, At, Ww) will
be determined, using the meth^ of Eq 9-19. Data given are: Wn = 18^140 kg per hr;
hi « 517 koal; pc «=86.4 mm Hg; h > IS.e^’C ; <d » 4.4^0; « 1.13 m/mo; no couden-
sate depression, U, Tubes mmm x
t, = a6.1°C i. = (. -4.4 — M 1-4.4 « tt.rv = M keal
The following calculations are made to implement Eq 9-19.

Ur =lll,910/(g//lv)»“
Ut - 9S.7/l.f4 X 10-» - 76.666

C/w = ^1879.7 X X 10“ — 4914

Then, with Eq 9-19:

<1 r(<7/Av)°->” , 2x88,8 22.8 . 28.2 _1
Ay L 75,546 <82*8+14.74) + 18.140 x 19.74 J
X 19.74

This reduces to (q/AyY = 794,540—88.7 iq/Ay),

For which a solution is (,q/Ay) » 33,330 keal per m2—5r
q » 1%140(517-26)«68.907x 10* koal per hr
Ay ^ 8.907 x10*/88,890 -878-9 eqm
Wy, = 8.907xl02/(21.7-15.4)->1.456xl02kgper hr
This represents a condensing water need of 6430 gpm.
Eq 11-20: n — 1.454x102/(8400x102 x 8.06x10-4x2.18) » 620 tubes per pass
The two-pass condenser will therefore have 1240 tubes.
From Eq 11-19, L » 87841/(0.049806 xl240)«4J8 m^
Since some length is required for end connections, and varying tube lengths by half
motra, the specific tube length will be 4.5 m .
Example 2: We shall now repeat the calculation for required surface, using a
coefficient taken from Fig. 9-10. At 2.18 m/seo, U for 22.8 mmtulm = 8898 koal per h^—
*960 B(u par lb.
tkoal per m8-.hr—deg C.
Am REMOVAL 423
mt-dagO.Inlet water is only 15.^, for which a correction factor of is found.
Hence corrected V — 3S93 x 0.93 ■■ 3154>
&907 X10^
If the tubes were clean, Ar - jH54~y 7. | "^57.8 sq m

Now if a eWnlinwiB factor of 0A5 is included (this is manufacturer’s jmctice), and

if {is taken at 18,140 x 528, then

Ar
0.85xil54x7.l
By comparing Ex 1 and 2, it can be seen that' standard derign practice in¬
cludes a liberal allowance for the effects of fouling, uneven steam loading, and
the uncertainty of exact exhaust enthalpy.
11-15 Air Removal. Heat transfer action in a surface condenser is hindered
by the presence of noncondensable gases which mix with the film of condensate
on the tube surface. The sources of air and other noncondensable gas leakage
are numerous. Some may come over with the boiler steam, or leak in through
turbine packing gland or exhaust nozzle connection. Ck>ndensing water leakage
past tute packings is another source, for raw water contains dissolved gases
which are released under a vacuum. Vents back from low-pressure heaters and
leaks in the condenser shell itself may provide still more of the undesirable gas.
Unless atmospheric relief valves are water-sealed, air will back-flow through
them.
Air infiltration is a serious factor and should be kept down as much as possi¬
ble. Since it is not practicable to eliminate it entirely, arrangements are made
for continuously drawing it out of the condenser and compressing it up to
atmospheric pressure where it can be released.
Recommendations of the Heat Exchange Institute are given in Table 11-9.
These are advanced to serve as a guide in specifying the capacity of the air re¬
moval equipment and are based on operating experience.
An additional source of noncondensable gas is to be foimd in the release from
injection water in contact condensers. For design purposes one canuseO.01864
free air from a ir* of 25*C water, increasing this linearly to 0.02689 m> for
water injected at 5*C •
The work required to compress a gas, a vapor, or a mixture of these is pro¬
portional to the volume to be compressed. Gas withdrawn from a condenser is
fully mixed with water vapor which could have been more cheaply removed by
TaWe 11-9. DESIGN ALLOWANCE FOR AIR INFILTRATION TO
CONDENSERS SERVING STEAM'ENGINES AND TURBINES—
M»/BaiaATBl.l*C(TO*F)

Maximum kg Free Dry Air Maximum kg Free Dry Air

of Steam Leakage of Steam Leaka^
Ckindenaed Condensed
per Hr Turbinea En^nes per Hr Turbines Engines

Uploll,M0 O.0W U10 119.401-490400 OJIIS

0.119 0497 49401-1111400 OM ....
sa,«8i>4Meo 0.149 0J8I 990401-Up. OJBH ....
424 STEAM PRIME MOVERS
condensation and pmnping were it possible to separate vapor from gas. While
not separable, the vapor content of the jnixture can be 'leaned” by cooling the
mixture as much as possible. This is the function of the air-cooler sections of
surface condensers. Given the mixture pressure and temperature, the gas rich¬
ness of the mixture can be calculated with Dalton’s principle of partial pressures.
Example 1: The pressure under the air baffle of a surface condenser is 61 mm Hg.
Temperature of the mixture leaving the cooler section is 28.9^. Assuming available
cooling water at 16.6*C, an external cooler might lower the temperature further, say
to The effect of this on the quantity of vapor accompanying the air to the
compressor will be studied.
The vapor being saturated, its partial pressure is given by saturated steam tables.
At 88.9^, Pv = 0,0902 kg/om* ab, = 46.152 m^/kg.
Pmix = Pa + PvJ Pa == 51 X 0.00186-0.0802-.0.0892 kg/om>ab.
Vn = tfg for the quantity of air associated with 1 kg vapor.
ti’air = 0.0892xl0«x46.168/(29.8x296.9)—8.08 kg per kg. vapor. Now if an external
cooler with negligible friction loss cooled the mixture from28*9*to 18.8“0, then
=(K0214;t;g:= 68.916 mS/kg;p^r=: 51 x0.00186~0.0214«0.0480 kg/omtab
Wgir = 0.0480x14^x09.0151(99.3x191.9)^,9.69 kg per kg. vapor
These calculations show that, by cooling the mixture from 28.9* to 18.8*0, the
vapor accompanying each lb of air withdrawn from the condenser is reduced
from 0.479 kg to 0.279 kg. This is bound to have a significant effect both on
ejector capacity required for the job and the consumption of energy to motivate
it.

1. lat stage body. 11. 2Dd stage nozzle. 20. Inter cooler hot well.
2. 1st stage nozzle extension. 12. 2nd stage diffuser. 21. After cooler hot well.
3. 1st stage nozzle head. 13. Elbow. 22. Water gage.
4. Pipe plugs. 14. Tube plate. 23. Cover nuts.
5. 1st stage nozzle. 15. Crown head. 24. Pipe plug.
6. 1st stage diffuser. 16. Water box. 25. Safety valve.
7. 2nd stage body. 17. Water box cover. 26. 101.6 mm evacoalor.
8. 2nd stage nozzle extension. 18. Tubes. 27. 101.6 mm vaivo.
9. 2nd stage nozzle head. 19. Oioler shell. 28. 2nd stage gate valve.
10. Pipe plug.

Fio. 11-47 Worthington two-stage air ejector with inter- and after-cooler.
AIR REMOVAL 426
Air compressors may be classified as reciprocating, rotary, or ejector types.
The reciprocating type, similar in principle to the air compressor, becomes in¬
conveniently large for other than small or low vacuum condensers.
The steam jet ejector is a widely used air pump, especially on large con¬
densers. When the heat liberated by condensing the ejector exhaust is recovered
in the feedwater, the ejector becomes a compact, efficient, and simple machine
for exhausting the noncondcnsable gases. Its first cost compares favorably with
other types of exhausters. Ejectors are usually operated on steam pressures of
from 100 to 250 psi, taking steam from reducing valve or bleeder nozzle. Up to
26-in. vacuum a single jet is satisfactory; above that two-stage ejectors are re¬
quired. The first-stage exhaust is condensed and cooled in an intercondenser;
the second-stage condenser is called the aftercondenser. The condensing water
should preferably be condensate from the hotwell of the main condenser. Jets
are arranged in parallel when capacities beyond that of a single jet are required.
Partial capacity operation is better in a multijet ejector than in a large single-
jet ejector because the extra jets can be turned off and each of the remaining
can work at its point of best efficiency. Ejectors are rated on the basis of their
free air handling capacity. Actually they handle an air-vapor mixture. The
calculations relating the actual operating conditions to the free air rating are
based on the ejector’s being equally able to handle free air or air-vapor on a
weight basis.
Example 2: A condenser has 51 mm abs pressure. The air-vapor mixture has been
cooled to 28*»* in an air cooler section. The ejector capacity required to handle0.086 m»
free air per min leakage will be established. From steam tables, the specific volume of
vapor at 28.0^ 46.162 m^/kg and py = O.OS02 kg/om>. Then
Pn = 61 x0.00136-0.0202-0.0892 kg/om*
then from pv = wRT
= 1.08xl0*x0.086/(29.8x288.6)—0.1086 kg per min
In the condenser, this air has volume = 0.1086 X 29.8x896.0/(0.0892 X 10*) =
2 358mS .

Mixed with it will be Wy lb of water vapor having specific volume of 46.162 m*

/kg at 0.0802 kg/om> ab • In 2.863 there would, therefore, be 2.863/46.162. or 0.0610 kg
vapor. Total weight of mixture carrying 0.086 m>free air is 0.1086+0.0610 —0.1546 kg.
Air-vapor capacity:air capacity = 0.1646 i 0.1086—1.492
Then free air rated capacity of ejector must be 0.086x 1.492» 0.1268 m^/min.

In recent years power plant designers have renewed an interest in mechanical

vacuum pumps. Several types had been formerly employed, and then steam jets
preempted the field. There have been developments in central station design that
have again put the mechanical compressor forward as a condenser auxiliary.
This has chiefly been the trend to high generated steam pressure and motor-
driven auxiliaries. Steam jet ejectors require steam of moderate pressure, which
must be taken from a reducing valve if the plant steam pressure is higher.
Turbine extraction steam would serve, once it was in operation, but would not
be available for starting vacuum. As higher steam pressures and temperatures
come into use there seems to be a desire to eliminate steam lines to accessory
426 STEAM PRIME MOVERS
equipment and have a single lead from boiler to turbine. Above 1000 psi operat¬
ing pressure the motor-driven, oil-sealed rotary vacuum pump appears to many
plant operators more desirable than a pressure-reducing, desuperheating station
to supply jet ejector steam. Sometimes the operator wishes to have much of the
plant equipment controlled from a central panel. Motor-driven vacuum pumps
can be set in operation by pressing a button on the control panel, a feature not
easily adapted to jet ejectors.
One make of mechanical vacuum pump is an oil-sealed rotary compressor,
motor-driven, with a circulating sealing oil system. A separator is provided to
recover the oil discharged with the compressed air-vapor. Compression is car¬
ried out in a hot machine so that vapor does not condense in the casing. These
units are more expensive than jet ejectors and probably require more inspection
and maintenance. However, they are receiving increasing acceptance in the
high-pressure central station.
11-16 Condenser Auxiliaries. Both contact and surface condensers need
pumps for the movement or pressurizing of water. The surface condenser will
always require a condensate pump and nearly always a condensing water
pump.* The condensate is withdrawn from the condenser and delivered to stor¬
age or to boiler feed pump suction. The -condensate pump serving a large con¬
denser is always of the centrifugal type, but reciprocating pumps may be used
for small condensers. The working head is the vacuum plus friction of the piping
to the surge tank, ])lus the velocity head, plus the difference in elevation between
the discharge to the surge tank and the condenser hotwell.
Example 1: The power to drive a condensate pump which is to serve the con¬
denser for a rated 60,000-kw Preferred Standard turbine is to be estimated; also the
pump capacity. Maximum exhaust pressure 63,6, minimum 88,1 mmHg abs, for which
condenser steam rates at full capability are given as 3.36 and 8.28 kg per kw hr for
straight condensing operation; 2.84 ancT 2.74 kg per kw hr in regenerative operation.
Pipe friction, velocity head, and static head are given as 18.3 mof water. Pump efficiency,
55%. S.O. of Ilg is 13.0,

Vacuum head = {723 9} ^ 13.6/W- {J.S} H2O

Total head = 18.3+ {1%} = m H,0
The condensate pump ought to be capable of carrying the unit at maximum rating
when for some reason the regenerative heaters are partially or wholly inoperative; hence
we use the flow of {3:28} X 66,000 = {m:iloo}>^g«per hr.
Head X Flow __
Power input to pump {i\i} hp.
Efficiency 75 X 60X 60 X 0.65

Two pumjjing conditions have been calculated, but it turns out that the pump
power is practically the same whether operating at 88.1 mm or 63.6 mm.Hg condenser
pressure. Maximum continuous flow of. condensate, barring condenser leakage, is
66,000 X 8.36/60«»8685 litres/min. A 3786 litToo/min pump good for 30 m dynamic
head at some specified minimum inlet submergence would give adequate capacity with
ample pressure over nomial rating to meet any contingency, yet would not be over¬
rated as the above calculations have shown. When operating at 60,000 kw regenerative.

’^Only exception is where water flow can be taken from 6—15 m higher elevation than
the discharge pool level. A dam or falls in a river would produce this condition.
CONDENSER AUXILIARIES 427
the condenser steam rate is 2.86 kg per kw hr. Condensate pump load is then only
2860lUvM per min.
The condensate to be removed varies with load; yet customarily condensate
pumps are constant-speed centrifugals. At part load, then, control must be
exercised on quantity of flow in order to maintain some uniformity of hotwell
liquid level. Three methods are practiced:
1. Throttling of flow in pump discharge line, deriving control from hotweU
float
2. Recirculation, This is automatic by-passing of part of the pump discharge
back to the hotwell. It is governed by a hotwell float-controlled valve in the
by-pass line.
3. Submergence control. The constant speed centrifugal vacuum pump has a
rapidly diminishing delivery as inlet submergence is lowered. For example, one

Motor

Traveling
| ;| Wofer-Intake Screen

Water Intake ^|p Woter Way to

to Troveling Condensers
Water Scre*en.^a

Link-Belt Co.

Fig. 11-48 Section through screen house.

of these pumps able to deliver 1B14 l/mio ati^Smsubrnergencewouldbe down to

'76i7 i/nain if the submergence should sink to 0.70 h>. This is a form of self-regula¬
tion, but one which has disadvantages, to wit, at light to medium loads the pump
operates imder cavitation conditions, and at very light loads a manual by-pass
or throttling of the discharge must be practiced.
A high-vacuum surface condenser requires 60-100 kg of water perkgof steam
condensed. Supply of circulating water is often a deciding fflictor in station loca¬
tion and a limiting factor m extension of existing plants. These large amounts of
circulating water make the condensing water circulating system of considerable
importance. The pumps, motors, and pipes serving the circulating water system
are among the largest in the plant. This flow passes through (1) intake works,
(2) screens, (3) intake conduit, (4) circulating water pump, (5) condenser-tubes
and water boxes, and (6) discharge conduit. The sources of circulating water
are rivers, lakes, and oceans. Sea water is suitable for condensers. It is of fairly
 428 STEAM PRIME MOVERS
even temperature the year round if the intake is located far enough offshore.
Tidewater location is not always the most convenient from other viewpoints
than that of water supply. Lakes as a source of condensing water are liable to
more temperature variation and to ice trouble in the winter. Rivers offer the
further problem of sand in suspension, floating debris, shifting currents, and
extremely variable water levels. By-passing some of the warm discharge water
to suitable points near the intake works takes care of ice formations. An intake
location should be made where silt is least likely to settle. The intakes should
be located in the deepest water possible since there the inlet circulating water
temperature will be lowest. Fish and debris are removed by trash racks and
screens. At some seasons of the year the screens need continuous cleaning, for

0 1-0 2-0 3*0 i-O 0 1*0 2-0 3-0 3-75

Velocity Through Tubes> m/s Velocity Through Tubes m/s
Ingersoll-Rana Cameron Hydraulic Data Book

Fio. 11-49 Friction loss through condenser tubes and water boxes.

which traveling screens with automatic cleaning features are employed. Fig.
11-48 illustrates a screen house located at the circulating water intake. The
traveling screen is composed of a series of small rectangular screens mounted
on two endless chains. When the trash-laden screens reach the cleaning position,
they are flushed clear by a jet of water directed through them from the inside.
Trash and water are caught in a trough and carried away. Unless thoroughly
flushed the debris will be carried over the top of the screen and into the con¬
densing water.
When natural supplies of condensing water are insufiicient, cooling towers
may be employed to cool the water for recirculation through the condenser, thus
making the atmosphere the ultimate medium to which the unavailable heat is
rejected. Cooling towers may be classified as atmospheric, natural draft, and
forced draft. The construction of cooling towers and their thermodynamics were
presented in Chapter 6 and vrill not be repeated here. They serve well for small
plants, which do not have adequate condensing water supply. Central stations
CONDENSER AUXILIARIES 429
are better located at the water, relying on electric transmission to reach the load
center.
Concrete tunnels and steel or iron pipe are used for the condensing water
conduits. A common method is to lead the water to and from the plant in con¬
crete tunnels into which the suction and discharge pipes dip. The head on the
circulating pump is the total of the hydraulic resistances of the trash racks and
screens, the velocity head, friction in conduits, water boxes, and condenser tubes
less the external head due to the difference in elevation of the intake and dis¬
charge. This external head can usually be neglected, and when it is appreciable,
it aids the pump. The friction heads in water conduits can be estimated from
Fig. A-14. Losses in the condenser itself can be taken from Fig. 11-49.

Example 2: To illustrate working conditions of the circulating pump, we shall

calculate the 'Vater horsepower’* required of a pump to circulate 288.800 litTO/mm.Con-
denser is of two-pass design with 28.4 mm 1.24 mm tuhos, 5.5 m long. Water velocity
throughtubes, 1.52 m/seo; water velocity in conduits,3.05 in/««c The water conduits total
88.5 m of steel pipe. Assume intake and discharge *at the same elevation and take
friction loss through screens, valves, elbows = 0.16 m

Volume flow «i288.800/10Sx60»4.818 mS/aeo .

/4.8I8 4
Diameter of water conduits 1.417 m
V 3.05 TT

With the use of Fig. A-14 in Appendix, the friction loss in 83.5 m of 1.417 mdia. steel
pipe can be found to be0.262raH2O. The velocity head= 3.05*/2(7 = 0.472 m. From Fig.
11-49, loss in tubes isO.46 mperZM m of tubes and 0.52 min the water boxes. Then friction
loss in condenser = 0.46/3.05 x 2x6.5 -f 0.62^ m = 2.17 m
Gross pump head — 0.15 -f- 0.262 + 0.472 + 2.17 on 3.054 m
288,800 x 3,054
Water power = -196 hp.
76x60

The circulating pumps are usually motor-driven, with direct drive. It is well
to be able to reduce ]>ump speed at partial loads as an economy measure through
motor control or fluid couplings. Condensers with divided water boxes (i.e.,
double water circuits) may carry light loads on one circuit.
Atmospheric relief valves are provided to permit partial-load noncondensing
operation and to prevent high back pressure on the weak cast-iron sections at
the low-pressure end of the turbine. As low-pressure steam is carried, the valve
and its piping are large and cumbersome, but nevertheless desirable wherever
they can be accommodated. The size required can be estimated with an assumed
2440—3050 m/min exhaust steam velocity, although probably half this relieving
area would suffice for protection only.
Vacuum breakers are provided to admit air to the condenser shell to break
the vacuum and allow the atmospheric relief valve to open. The purpose of
breaking the vacuum is to jirevent condensate backing up into the turbine should
the condensate pump break down. The vacuum breaker is usually a float-
operated device which remains out of service until condensate rising in the steam
space reaches a predetermined level. In order to have the same device protect the
turbine against damage from excessive circulating water leakage resulting from
430 STEAM PRIME MOVERS

1. Atmospheric relief valve 15. Air removal line from main condenser to
2. Expansion joint first stage ejectors
3. Gate valve 16. Air ejectors (first stage)
4. Circulating water discharge from main 17. Air ejectors (second stage)
condenser 18. Stop valve (steam)
5. Condensate from condenser to conden¬ 19. Throttle valve (steam)
sate pump 20. Steam strainer
6. Check valve 21. Steam pressure gauge
7. Condensate from condensate pump to 22. Intermediate condenser condensate drain
intermediate and after condenser loop
8. Condensate from intermediate and after 23. After condenser condensate drain drip¬
condensers ping
9. Condensate line to heaters 24. After condenser drainer
10. Condensate recirculating line to main 25. After condenser drainer float chamber
condenser piping
11. Condensate control for recirculating line 26. Aft(’r condenser drainer float chamber
12. Vent from control valve float chainber vent
13. Condensate line to conlrol valve float 27. Intermediate and after condensate return
chamber to main condenser
14. Vent from condensate pump to main 28. Main condenser support springs
condenser 29. Main condenser spring support
Fig. 11-50 Typical connection of auxiliaries for a large condenser.

a major internal failure, the vacuum breaker may be electrically connected to

the turbine throttle trip and to an alarm system.
The relation of the auxiliaries to the condenser itself, and the necessary con¬
nections can be studied in Figs. 11-39 and 11-50.
11-17 Operation and Testing of Condensers. For obvious reasons, the
operation of a condenser and its auxiliaries is tied to turbine operation and vice
OPERATION AND TESTING OF CONDENSERS 431
versa. The starting procedure for turbines, set forth earlier in this chapter, was
likewise that of the condenser, and was given for the case of a large unit whbse
turbine was to be started under vacuum. Operators of smaller turbines may
elect to start them noncondensing, not drawing the vacuum until the turbine
is warmed up and idling at several hundred rpm.
When the turbine-condenser is in normal operation, supervision consists of
inspecting and logging the readings of the several instruments and gauges with
which the operation of the condenser and its auxiliaries is verified. Many of
these may be delivered to a central instrument panel by remote indicating in¬
struments ; others will be locally installed. Thermometers should take the tem¬
perature of incoming and outgoing condensing water, hotwell temperature, and
air offtake. The condensing water thermometers should give the water tem¬
perature to the nearest 1/20 degree if these temperatures are to be used
quantitatively, for the heat carried off by this water is the product of a large flow
and a small temperature difference. Pressures to be watched are condenser
vacuum, ejector motive steam, condensate discharge, condensing water, and the
differential pressure between condenser inlet and outlet. Flows that are usually
metered are condensate, condensing water, and air ejector discharge. Water
sealing of condensate pump glands, the atmospheric relief valve, and the ejector
condenser drains require occasional inspection. Purity of the condensate may
be checked from time to time, or continuously recorded by a conductivity meter.
This gives an indication if sudden raw water leaks develop in the condenser.
Maintenance supervision endeavors to keep the vacuum system air tight by
detecting and scaling all possible air leaks. But mainly this service is concerned
with keeping the heat transfer surface clean and intact.
Fouled tubes, air leakage, or insufficient circulating water are all causes of
low vacuum. The fouling of tubes may arise from numerous sources, such as
silt or sand in the water, organic and inorganic slimes, or scale. When fouled, the
tubes are cleaned according to the characteristics of the deposit or scale. The
principal methods arc: (1) bake by draining the water side (certain scales will
crack loose and can be washed out after baking); (2) wash out slimes by in¬
creasing the water velocity through the tubes to4.6m/scc or more; (3) force rubber
or metal plugs through the tubes under water or air pressure, thus scraping out
the scale; and (4) wire brushes.
Surface waters contain vegetable matter and micro-organisms, principally
algae. The conditions within a condenser tube are favorable for the rapid growth
of these organisms, resulting in the formation of a gelatinous deposit on the
inside surfaces of the tubes. Chlorine has a germicidal action on these organisms
and, if fed daily in amounts sufficient to produce some residual chlorine content
in the water leaving condenser, will destroy slime. Chlorine treatment will not
eliminate the necessity of cleaning tubes of deposits other than from micro¬
organisms.
Because of the corrosive action of chlorine solution on metals, excessive con¬
centrations which may seriously damage condenser tubes must be avoided.
Special automatic equipment is available for controlling the addition of chlorine
to the circulating water.
Chlorination is most economically carried out with the unit operating at re¬
duced load. Then the circulating water pump can be set at low speed and its dis-
432 STEAM PRIME MOVERS
charge valve only partially opened. Liquid chlorine is then automatically fed
into the stream of circulating water at a rate sufficient to maintain a residual
chlorine content of from 1 to 15 ppm at the outlet. Some operators feed the
chlorine into the circulating water on the inlet side of the circulating water pump
for a period of about 30 min every 24 hr.
After chlorination the discharge valve is usually opened wide and the circu¬
lating pump increased to high speed in order to rinse out the chlorine and dead
slime.
Operators must also be able to detect tube leaks and cope with them with
little or no attendant outage to the turbogenerator. Air bubble impingement,
corrosion, dezincification, pitting, while not common, may suddenly appear in a
condenser. Troubles near the inlet end of the tube are often overcome by the
installation of thin, smooth, flared entrance sleeves which cover pitting that may
not have progressed too far. Small leaks, if detected by condensate analysis,
may temporarily be remedied by injecting special stop-leak compounds into the
condensing water. In other cases it may be necessary to stop the unit, unwater
the boxes, and plug the ends of the leaking tubes until such time as scheduling
of the unit allows enough time out for retubing
Tests. Operating tests for tube leakage are carried out in several ways, all
requiring the turbine to be at rest. One method is to unwater the condenser,
establish a steam side vacuum, plug one end of a tube, and insert a “tell-tale^^ in
the other. The vacuum works in through the leaks and is reported by the device.
Another method is j^ossible when the water box covers are fitted with diverts
lights and port glasses. The steam side is put under air plenum, the water boxes
being full. Leaks show as bubble streams leaving tubes which can therefore be
identified visually.
Engineering tests for the developed coefficient of heat transfer require the
establishment of heat transfer, usually from water side measurements, con¬
denser steam i)ressure, and knowledge of the condensing surface installed. Water
inlet and outlet temperatures must be accurately obtained and flows measured
correctly. These tests, which are fairly simple of execution with small con¬
densers, become on large units a real challenge to experimental engineering.
When steam consumption of air ejectors is wanted, the same may be readily
found from measurements of the nozzle throat diameter and the principles of
isentropic nozzle flow. The jet flow is set by throat conditions.
Example 1: Supi)ose it ha'i been a.-cert a mod that one of the nozzles for motive
steam m an ejector has a throat of 3.18 mm diameter. The normal steam pressure is
7.08kg/omSga. dry and saturated. The steam consumption of this nozzle can be deter*
mined with fair accuracy by solving the equation for flow at the throat.
ics == AtVtM (11-25)
Throat pressure, pt = 0.58 x 8.06»4.67 kg/om^ ab
The following steam properties are found from the Mollier Chart by locating the
condition of isentropic expansion from inlet to throat.
hi =661 a* Si = 1.590, ht =636.1, Xt = 0.962.
Then Vt =8i.61\/661.1-686 l -^457.56 0^4071 x0.068»0.8916 mt per kg

^ -0.00928 kg pw MO (83.4 kg par hr)

*Tn keal/kg
PROBLEMS 483
Were the steam initially superheated, the throat pressure would be 0.55pi, unless
higher accuracy is required for the critical pressure ratio. If so, pt/pi =* [2/(7 —
Thermodynamic ProperticB of Steam (Keenan and Keyes) charts the steam exponent 7
as a function of p and t.

Some published test results on a large condenser will be given in the next
example to illustrate the testing of these units for heat transfer.
Example 2: On a certain 5110 sqm condenser, the averages of {periodic readings
taken during a 2-hr test at stead}^ load are as here reported. The over-all coefficient
of heat transfer is to be computed.
Barometric pressure (corrected tol5.6°0), 770.9 mm
Metered steam flow to throttle, i2,i26 l^per hr.
Vacuum at neck of condenser (corrected to 16.6®C),761.8 mm.
Condensing water in at 8.94®C, out at t6.83*0*
Condensing water quantity 835»400 litrea/min.The heat transferred =885,400 x0,99918
X00(15.38-a94)«1284.8xl0i koal per hr.
Steam pressure = (770.9— 751,8) • 19.6 mm Hg abs atl5.6**; also 19.6 mm at 0^.

U « 1284,8 X 104/(9.88 X 5110) 2709 koal per hr - mMeg C-

It is possible to make a breakdown of the components of over-all coefficient
by a test method suggested by Eqs 9-4 and 9-20. If on separate test runs con¬
ducted with steam loads that would
make q/A^ approximately constant, the .004|
water velocity is varied from run to run,
91
then the principal variable is which
varies as a power of water velocity.
While this exponent is 0.73 in Eq 9-20,
good test results are had by assuming
Vw®*®. is « and 1/C7w = 0 l/V
when = 00 . This test point cannot Fig. 11-51 Reciprocal plotting of
be approached, for 3—4.6 m/sec will be overall convection coefficient.
found to be the limit in most installa¬
tions. However, by reciprocal plotting as illustrated in Fig. 11-51, the probable
values at Fw = ? can be determined since the test points usually fall along
a straight line which can be projected to the axis where l/F^®*® = 0. The
intercept is 1/Uy -h 1/C/t + 1/J7t. After cleaning the condenser tubes l/I/s =
0 (practically), and another series of runs so plotted would provide an inter¬
cept equal to 1/Uy -f l/f/t* The intercept increment on the two plots would
be 1/Un. Ut is readily calculated; hence a subtraction from 1/Uy + 1/Ut
leaves 1/Uy determined.

PROBLEMS
1. Find the required bore x stroke of a high-speed, simple steam engine for
direct connection to a 60-cycle, 25-kw generator. Governor can be adjusted to
operate in 300-350 rpia range. pi 7.03k^cm*ga, saturated steam; atmospheric
exhaust, deuanoe, 5%; cutoff, 25%; i| ^7^77.
434 STEAM PRIME MOVERS
2. A singIc>«ylinder50.8ciB>^101.6ciiiX240»rpm controlled compression unaflow
engine being available for a certain pump drive, it is desiredi to determine the gpm of
water that can be delivered against 11.6 m head. Clearance, 4%; ifm 88%; hydraulic
efficiency, 76%; Pi, 7.03 kg/cm* ab saturated steam; exhaust, jt52 mm Hg abs.
Release at 0.35 kg/cm* ga and 96% stroke.
3. A dealer otters an industrialist a single-c3'linder, full-compression, noncon¬
densing used unaflow engine of30.5cmX‘45.7 cmX 200-rpm dimensions. Engine would be
usedto belt-drive a piunp requiring 127kp6t1800 rpm.Normal cutoff, 25%; clearance.

-r—iXJ...'t... ;j> -J: ^

WeaUnghouMC Elec, ix Mfgr. Co.

Fig. 11-52 A small two-pass surface condenser.

Allit Chalmeri Mfgr, Co.

JP'iQ, 11-63 Water-box and water connections to large two-pass, divided flow surface
condenser. Note vertical motor-driven condensing water pumps.
PROBLEMS 435
10%. Advise the prospective purchaser whether the engitie would be suitable for his
job, knowing that the factory plant has 1S.S kK/em* g» boilers.
4. Using Fig. 11-6, derive Willans lines (TF = a -f bP) of a 10-kw noncondensing
unit for steam conditions as given:
(a) Superheated steam at M kg/em* ga» 218.9^ ; atmospheric exhaust.
(b) Superheated steam at 10.8lGg/omS ga, 218.8^; atmospheric exhaust
(c) Saturated steam at 7.08 kg]eiiit£a.0,85 kg/cmiXga back pressure.
(d) Saturated steam at7.08ka/oiBPga,0.7 kg/om^ga back pressure.
5. Determine, with use of Fig. 11-9, the based on hp ana of this unaflow
when operated (a) at 507 bhp, pi 14J kg/omtga and saturated steam; atmospheric
exhaust; or (b) 600-kw output, Pil4.1 kg/om*ga^82.2*C;0*56 kg/om*gt^chaust pressure.
6. The piston of a full-compression unatlow engine will cover the exhaust port
at 8% return stroke. Admission, 3% before DC; clearance, 7%. What minimum
throttle gauge pressure should be used if (a) exhaust is atmospheric; (b) exhaust is
at 508 mm Hg vacuum?
7.0iven a 80.5 omx4847om controlled compression horizontal unaflow having i;*based
oh ihp = 0.70wben operated between 8,8 kg;/omagaand atmosphere;steam initially dry
saturated. Piston rod,44.5 mm; tail rod,81.8 0.90. When this is connected to a
standard 240-rpm NEMA a-c generator, what is the proper kw rating? Find
Consider rated load to be such that r = 4.0.
8. Construct to scales i em—10% FD,1 om**5 kg/om^, superimposed synthetic HE
and CE indicator diagrams to the following specifications!

Clearance Admission Cutoff Release Compression

HE 4.8 2.0 39 90 35
CE 5.2 1.5 32 88 38

HE events specified from IlEDC; CE events from CEDC. Throttle steam state, 56J
kg/oms •bdl5.6*C:exhaust, 8.8 kg/cm^ ah (hence the assumption of hyperbolic expansion
IS not warranted). Find the average mep and diagram factor.-
9. Determine, from this single-cylinder engine test data, the same items of
performance as in Ex 2, Sec 11-3. Time, 45 min; readings averaged for test period.
Load is a belted electric dynamometer with76 cm brake arm. Allow 1% belt loss. Scale
force, 102 kg net. p, 10.65 kg/om?; quality by throttle calorimeter which shows 115.6^3.
Barometer748mmHg. Speeds: engine,!^ rpm;dynamometer,945 rpm.Pmep*«:j?.S,8i,66
kg/om*;C£,5.81 kg/oin*. Increase shown by steam flow meter integrator during test, 1048
kg. Ex^ust,0.88 kg/om4g«. Engine specitlcations :30.50m X45.7 omx 150rpm. rated at 127
bhp.Piston rod|51 mm no tail rod.
10. An engineer was asked to measure the performance of a Coriiss engine
driving a 125 kva a-c generator in a lumber mill, belted as in Fig. 11-2. There was no
arrangement by which steam or water flow
could be measured, but the engine could be
indicated,and the panel instruments were reliab'
le. Boileri.8 kg/eo^gi^probably about 97% dry,
long exhaust line.Engine was found to be 86.6 om x
9L4omwith51 mmpistonrod.FiQ.ll-lOPisan ac¬
curate reproduction oi the cards taken at what
Atm. lint i02mmlon9
was stated to be ^‘normal load.” Manufacturer Fio. 11-lOP
gives max cutoff of this engine, 25%. By assum¬
ing 0.80 power factor, 25% initial condensation loss, 5% clearance and 0.90 generator
efficiency, the engineer was able to (1) report a quantitative estimate of all performance
items mentioned in Sec 11-3 and (2) render a considered opinion whether the capacities
of engine and generator were well matched.
436 ^ STEAM PRIME MOVERS
11. As a familiarization exercise, any one of the turbine diagrams in Fig. 11-14 may
be selected for redrawing on the more detailed basis suggested by Fig. 11-13.
12. Calculate -qt of the turbine, Fig. 8-13.
13. Calculate the heat rate of the turbine in Fig. 8-7.
14. Review the results of Ex 1, Sec 8-5, then lay out the flow diagram of that
plant, properly labeled with the data pertinent to turbine heat rate. Calculate said
rate.
15. Obtain the following equation for the ratio of extracting to nonextracting '^in¬
ternal” steam rate. Assume condition line same in either case.

= (Ai - A2)/[^1 - ^2 - ^r(1 - (V^av))]

^1, ^2 are enthalpies at beginning and end of condition line, = (hi -I- h2)/2] =
keal extraction per kg throttle flow,
16. Note the equation in Prob. 15. Consider that is the difference between feed-
water and condensate enthalpies. Implement this equation with data from Ex 1, Sec
8-5 and find WexJw. Test the equation by calculation of w and TText separately; -it;
from the A turbine line, straight condensing; from data and results of the afore¬
said Ex 1.
17. Find the nonextracting steam rate, la^, for a 3600-rpm turbine of 30,000-kw
rating. Pi 45.7 kg/oiii<gA,464.4*X);P261 mm. Employ experience curves. Fig. 11-15. Assuming
that this turbine was used regeneratively, what steam rate is to be expected? (See
Prob. 15.) Feedwater at 171.I^«
18. How much steam will a 75-hp mechanical drive turbine need per hour? Steam
supplied at 114) kg/om* ge saturated. Exhaust, 0.56 kg/om* ga. Single-stage,1800-rpm type.
19. A large, 1800-rpm, noncondensing turbine 158»750 kg steam per hr at
50,58 shaft hp. P]58.1 kgfaufi ab,815 6^; 5.1 kg/om* gft exhaust. Is the developed engine
efficiency in agreement with Fig. 11-17? Assume = 0.97.
20. Estimate the stage efficiency in the dry steam stages of a pressure-compounded
impulse turbine of average economy. Do the same for a reaction turbine of high com¬
mercial efficiency.
21. A pressure-compounded impulse turbine for relatively low speed is designed
with blade velocity ratio of 4/10. Estimate the efficiency of a group of stages (15 keal
A5) in the wet region, beginning at 0.965 quality.
22. Predict the condition line of a straight reaction turbine operating between 50.8
kg/om* aby426.7^nd51 mmHg abs.Include numerically labeled sketch. Use an
average commercial velocity ratio.
23. Predict the condition line of a two-row velocity compounded turbine of three
pressure6tages.P]15.8 kg om* ga,37.8^superheat,exhaustmgat51iiimHgabs. Small unit,
warranting use of lowest commercial blade velocity ratio.
24. Assume that the steam end point in the turbine. Fig. 8-13, was not known and
attempt to determine it by heat balance.
25. By means of a heat balance determine the end point of the condition line of a
1014-bhp, geared, noncondenstng turbine. Mechanical friction, 5% of internal work.

Pi 17.6 kg/om* gi^ dry and saturated; P2,14)5 kg/om* ga.

26. Considering each isentropic beat decrement of Example 5, Sec 11-6 as a stage,
compute (1) reheat factor; (2) internal engbe efficiency; (3) over-all stage efficiency.
Find the blade velocity ratio represented by the top of line B, Fig. A-13.
Nozzle efficiency, 0.94; leakage and rotation loss, 0.05: Two-row Curtis stage.
28. Determine the blade velocity ratio represented by linq F, Fig. A-13. This
turbine is pressure-compounded. Nozzle efficiency, 0.94; leakage and rotation loss, 0.04.
29. Find the speed changer settings, in terms of no-load rpm, that will enable
PROBLEMS 437
machines A and B to divide a 10,000-kw load at 60 cps as follows: A of 5000-kw ca¬
pacity, 15% regulation, t^kes 3000 kw; B of 8000-kw capacity, 10% regulation,
takes 7000 W Both have 4-polc alternators
30. Three turbo-alternators are «haring a 33,000-kw, 60-cycle station load All are
normal 1800-rpm units Were the load then to decrease to 22,000 kw with no change
in governor setting, ho\^ would the load be divided between the three machines and
what would the frequency be^

Unit Speed legulaUon Capacity Load at 60 cps

A 09% 15,000 kw 12,000 kw
B 12% 15000 kw 14,000 kw
C 19% 10,000 kw 7,000 kw

31. Operating coii(iition& for i 2(),l)00-k\\ turbini with dnergent firist-stage nozzles
areJ4.1/om*»h,5fi.6®6uperheat,exhuist,88.1 mmHg^ i)WithfuHunthrottled flow, 16,500
kw are generated on 81,200kg ‘-team per hr Find the capacity and steam rate of this
unit were the steam tempenturt riiscd to 815.0®C
32. Repeat Prob 31 except for a rc iction turbiiit with full peripheral admission to
the first stage
33. A single-stage, 1800-rpm mechapicil drive turbine is rated at 81 bhp with
steam conditions of jo 6 kg/om* ga, saturated steam atmospheric exhaust What w^ould it
be rated at 01) 8.8 kg/cm* ga saturated steam and same exhiu^t*!^ Assume rfe constant
(Actually the chinge might improve it Whv
34. The turbine whose rated load condition line i*- GG, given by Fig A-13, has
divergent first-stage nozzles and single governor v live Whit prreent of its rated load
can It develop when the governor is throttling the fir^t-stigeadmi^HiontolO.Skg/om^v
35 Derive an equation fo'r the Willans line of a 50,00()-kw, 3600-rpm conoensing
turbine Steam pres&urey58.2 kg/om*gajtemperature^510°C Exhaust, 61 mrollgabs. CaP
dilate Vc at 40,000-kw load
36. Plot a predicted Willans line ind the derived curves of steam rate and thermal
efficiency, for a 3600-rpm, rxted 20,000 kw suiierposed turbine As'sume mech-elec effi¬
ciency of 0 92, Pi,87.9 kg/om> ab510^ , P2> 17.6 kg/cm» ab Overlo'»d condition capacity
24,000kw
37. Plot a Willans line from the data of Table 11-6, then derive values of
and ^
38. The results of i 00-min test run it stead} load on a ToOO-kvv condensing steam
turbine are hen Mimmarized Calculate the following (1) Exhiu>t saturation tempera¬
ture to 1/20 of a degree, (2) ideal and actual exhaust enthdpy, (3) (4) rjc,
(5) (6) HR
D^ta Pii28.1kg/cm8ga,3l6,6®C.Condenser manometer,715,2 mmHgat26*7®C baromet¬
er, 761.2 mmHg atlo®0 barometer scale correct ato°C Steam flow, 82,520 kg per hr
Watt hour meter readings initial, 265,217 kw hr, finil, 275,942 kw hi. Condensing
water flow, 24,510 litreg/mm; m at 20.1*C,out at80.8*X3
39. Assume that Fig 11-40 applies to a 12,^-kw turbine guaranteed for 4.42 kg
per kw hr at 42.2 kg/em> ga,426,7*0 with38,immHg,txh lu-tpn-sun Di'iing a performa¬
nce test at this load the steam rate was found to be 4.40 kg per kw hr Did this performance
meet the guarantee if the operating conditions vvcre41.5 kg/om8ga,446*Oi44 mmHg ^
40. A condenser manometer reads720.8ininHgat the same time that a mercunal
barometer shows762.5 mm Temperatures are manometer,29.4*C; barometer, 18.8*0. Find
correct vacuum, referred to 15.6X5 standard
41. Find the exact saturation temperature corresponding to the following: Con¬
denser manometer,674.6 mm Eg at 87.8*0; barometer, 758.7 mm at 27.2*0.
438 STEAM PRIME MOVERS
42. The enthalpy of steam entering a condenser is 2228Jfg. Condeoser has . a
vacuum of 724.2 mm Hg on a25.6®C column. Atmosphere, 1.010 Icg/omt »b- Determine
steam quality fo nearest 1/10%.
43. At what height would a 35"C mercury manometer column be sustained by a
vacuum in a region of 0.058 kg/om*/ab pressure? Barometer, 756.7 mm at 15.6®C.
44. The condenser pressure in a system with abnormal air leakage is accurately
determined to be 0.088 kg/om> ab. The contents areat 4l#7*C. Net interior volume, 84 m».
Find the air-vai)or mixture and the weight of air in the condenser.
45. What is the air richness of an air-vapor mixture entering an air ejector atl8.1*
Cand702.8 mm Hg vacuum. Manometer temperature,88.0*C. Barometer,712 mm at 16*0®C.
46. Body height of a parallel flow barometric condenser is approximately 1*22m-
Steam pressure, 0.3 kg/om* ab.Working barometric pressure787 mm.Injection water at40.<r
0 available at0.36kg/cm*ga relative to hotwell level. Load isll34 kg of engine exhaust steam
per hr.Pipe friction,o.01miwatcr,Estimate the following: Effective tail pipe height; kg
injection water per hr; noncondensable gas to be ejected; head required of a booster
pump, if any.
47. A certain low-level, counterflow mixing, jet condenser is to serve an engine of
characteristics given bj Fig. 11-7. Injection water available at 29.4*C. Pump suction
suhmergcnce4.22m,^?tatic discharge head,0.91m. Neglect pipe friction. Find required water
horsejiower of the removal pump.
48. Repeat Prob, 47, assuming an ejector condenser requiring 0.7 kg/om*gapressure
at the spray nozzles. Pump suction,o.l4 kgfom* ga.
49. Determine n, L, Ay, wv for a two-pass surface condenser, using Eq 9-19 for heat
transfer. Specifications: Condense 68i,040 kg steam per hr at 61 mm Hg abs 'vith 4.4*0
terminal difTerence. CondenMiig water available at 20*C* F^ = 1.83 m/eao; f/B=18.8l0 *
Tubes 19-1 mmX 1.24mm* Steam enthalpy,7277 J/g,l.l*C condensate depression. Neglect
air cooling requirements.
50. Determine n, L, for a surface condenser by employing data of Fig. 9-10.
90J20 kg. steam are to be condensed jier hr at 61 mmHgabs. 26.4miilx 1 *951.52m/
15'9®CAvailable space limits tubes to not over 7.62 mlong Neglect air cooling.
51. During a test a two-pass condenser with 20006.1 m X 22.2 mm xl.24imm tubes per
pass received a flow of 81,890 litrea/miniofl8-8°C condensing water which wasdischarged at
28.7®C • Condenser manometer, 727.2 mm llg at 82.2H3- Barometer,768.8 mm corrected to
16.6°C.How does the developed C compare with a prediction from Fig. 9-10?
52. Find the capacity rating, m* free air, of a steam jet ejector for a 30,000-kw
turbine's condenser. It is expected that the air-vapor mixture will be cooled to 26.7*C
at 88.1 mm Hg abs.
53*. Assume that the free air leakage including injection water release, for the con¬
denser mentioned in Prob. 47 will be0.0779m*/mmat 81,1 What rated capacity .of steam
jet ejector is needed if (1) there is no external mixture cooler (2) there is an external
cooler, as in Fig. 11-41, able to reduce mixture tempeiature to 61.7*0 ?
54. Find the water horsepower for the circulating pump of a two-pass condenser
with flowof24,830 litres/minicondensing water.There are 1200 22.2 mmxi.24mmx 4.6m tubes
in the condenser. Water conduits are steel pipes 610 mm with 68.6 m total length. Consider
final discharge at same elevijtion as screen house water level.
55. The following are test data on a 3066 m* , two-pass condenser having 6720
26.4 mmx 1.24mnitubes. Condensate fiow was 118,340 kg per hr at 87.8*0 ; condensing
water, 118,660 litres/^tn Temperatures: t, 88,3; ti 2^8 *, to 32.06 (all *0). Vapor pressure,
61 mm Hgabs.(l) Calculate (/; (2) assuming tnat Uy was 17,680<'*», find C/,.
56. A certain tube in a condenser, having been selected for testing, was isolated from
the main flow and these water quantities were measured:

*koal per mi—hr—deg 0

PROBLEMS 439
Test No. Water flow, hf/min Inlet *0 Outiet '0
1 66.7 m 18.76
2 86.1 uja 80.60
3 86.7 16.6 21.86
4 61.6 16.6 28.80
5 80.8 16.6 88.86
6 80.6 16.6 84.80

Between test No. 3 and 4 the inside of the tube was thoroughly cleaned. Tube is SU
mm X i,(M nun thiokneM x 4.7 m Condenser vapor pressure, tfi.4 mn Hg abs. From
these data, find U, and Vy,
 CHAPTER 12

THE GAS LOOP

12-1 Function of the Gas Loop. The equipment considered as **Gas

Loop” has the function of creating the pro^r conditions for combustion, of sup¬
plying the ingredients for combustion in the proper proportions, and of moving
tlie products of combustion to the atmosphere. I'hese actions necessarily involVe
the combustion equipment itselt, i.e., the iurnace, anTlSurner, or sthket; how¬
ever, this portion of the gas low Islio^mGmatery"‘aV^^ steam
generating unit that it was considered expedient to place the study of it in
COAL STORAGE 441
arrangements for supplying fuel and air to the combustion equipment and re¬
moving ash from it, together with control and regulatory apparatus, are what
constitute the Gas Loop.
The elements of the gas loop equipment shown in the customary relative
positions are schematically diagrammed in Fig. 12-1, where it will be observed
that the loop is imaginary. Actually, the flow is open to the atmosphere, and
once having been discharged from the*chimney, the pr?rd\icts of combustion are
not wanted again at the inlet of the combustion air system. However, the earth’s
atmosphere, blanketing the vegetation which covers much of the earth’s surface,
undergoes a sort of rejuvenation which undoes, to a certain extent, the action
of combustion. This is the continual synthesis, by green vegetation, of oxygen
through botanical action, whereby sunlight, carbon dioxide, and water produce
oxygen. Hence, outside the power plant, nature is at work changing carbon
dioxide, the proper product of the combustion of carbon, back to oxygen. Thus,
by the exercise of some imagination, and the taking of some liberty with the
meaning of “loop,” we have contrived to refer to the gaseous flow connected
with a steam power plant as a loop.
12-2 Coal Storage. To place coal in storage can be as simple as dumping
loads of it from motor trucks onto the ground in an area assigned to a “coal
pile.” However, at many industrial and all utility plants a mechanical system
for receiving coal from the delivery system, conveying it to storage, holding it
safely, then reclaiming it for use will have been provided.
The purpose of fuel storage is twofold. First, and most important, fuel
storage is insurance against complete shut-down of a plant occurring from
failure of normal operating supplies to arrive. Second, storage permits some
choice of the date of purchase, allowing the purchaser to take advantage of
seasonal market conditions. Storage of coal is primarily a matter of protection
against coal strikes, failure of the transportation system, and general coal
shortages.
The cost of fuel storage is inexpensive insurance against the loss occasioned
by complete cessation of a manufacturing process, even for a short time, caused
by lack of fuel. But coal storage is even more important to the public service
station. Electric light and power have become so vital in everyday domestic and
industrial life that cost of creating coal storage is submerged in the general effort
to secure reliable and continuous service. A storage of 10% of the annual con¬
sumption might suffice for most cases, although, of course, factors such as
nearness to coal fields, transportation facilities, amount coal will weather, and
the availability of substitute fuels must be considered. The frequency of in¬
terruption of coal mining by labor action in the past has tended to induce
management to store more coal than would otherwise be considered economical.
Where the area is available, it is not uncommon to find a full twelve-month
coal supply in storage.
Example 1: Taking the yearly load duration curve and related data as in the
example, Sec 3-10, the volume of coal storage necessary for an emergency reserve of
7% of the annual consumption and the required acreage of storage will be determined.
Storage depth,1.44m. From the problem referred to we find the annual consumption to
be71000 tonnes. Bituminous coal weighs approximately sOOkg/xnt,
442 THE GAS LOOP
72000 X 0.07 X 1000
Volume of coal storage 6300 mS
800

Storage area = 24^ ""

Actual area would be slightly larger because of sloping sides of the pile.
Storage from which coal may be withdrawn to supply combustion equip¬
ment with little or no rehandling is live storage. Dead storage is that holding'
area from which coal can only be taken by the exercise of a definite reclaiming
and conveying activity. Dead storage is ordinarily an exposed pile, laid di¬
rectly on the ground.
When coal is piled in storage it ^Veathers.^^ It has a tendency to become
slack; to oxidize with accompanying liberation of heat. When loosely piled
in shallow piles with large lumps, natural circulation of air currents may carry
the heat off rapidly enough so that the temperature does not rise dangerously.

Cor Oumpor
Shevolt

Fig. 12-2 Outline of coal-handling equipment.

Coals having high sulfur content are liable to prove troublesome to store in
volume because local heating apparently originates in a reaction between
sulfur, air, and water. Deep piling with exposure to winds aggravates spon¬
taneous heating. The best remedy known at present is to exclude air from the
pile by (a) underwater storage, (b) bituminous surface coatings, (c) rolling
and packing an air-tight layer of fine coal on the surface. At the Port Washing¬
ton plant 70307tonnci were satisfactorily protected by covering the pile with
30«5cm of fine coal dust overlaid by a thin layer of lump coal to prevent wind
and rain erosion.
The cost of maintaining coal in storage is the sum of interest and taxes on
capital invested in land, interest and depreciation on handling equipment, in¬
terest and insurance on coal stored, the energy costs of moving the coal, and
deterioration in value due to weathering.
The bulk of coal burned is received either by boat or railroad, supple¬
mented, to a limited extent, by trucks, mainly to supply institutional plants
that are accessible neither to rail nor water. Methods of handling coal con-
COAL STORAGE 443
stitute an important part of power plant equipment, and the coal handling
equipment is generally one of the major components of plant cost.
Coal for either live or dead storage is received intermittently at the plant;
consequently, the coal_ stocking and tedaimipg. equipment must have cjqjacity
to unload the expected shipments within a length of time wUch will not bear
any definite relation to maximum rate of usage in the furnaces. The unload¬
ing plant should be arranged so that the coal as received can be stocked to
dead storage or fed directly on to the conveying system leading to live storage.
In almost every case of coal handling, equipment may be selected from
among a large number of devices radically different from one another so that
no standard systems may be described. Each installation is an individual
problem based on the immediate surroundings.

nCCCAiMINC)
TVpicau SCCTiom.
Sauerman Broi,

Fig. 12-3 Storage and reclamation by drag scraper.

When coal is delivered to the plant by barge, unloading is readily ac¬

complished by grab bucket operated from a high coal tower adjacent to the
wharf. The coal is hoisted high enough before being dumped into a receiving
hopper that it can then flow by gravity to either live or dead storage.
Delivery by railroad may be (1) to track hoppers at ground level, (2) to
unloading spur where locomotive crane unloads it with grab bucket, (3) to
storage pile beneath elevated trestle, or (4) to open dump at ground level.
The last method is often used in small plants, where the railroad spur can be
carried to an overhead trestle. Coal is shoveled by hand from the cars or
dumped to the ground through hopper bottoms, then elevated to coal buggies
or wagons by portable loaders. Handling to storage ig^enerally by one of the
following methods;
1. Drag line scraper. Has the advantage of low first cost and maintenance,
but large energy consumption per ton is a disadvantage. It is adaptable to
irregular storage areas and can store deeply or thinly as required. There will
be a front post with swiveling sheaves located beside the reclaiming hopper
and several tail posts or, as shown in Fig. 12-3, a continuously movable tail
post. The two-drum winding machine is usually driven through reduction gears
by a reversible motor. The haulage cable is wound on one drum; the return,
on the other. By friction clutches the idling drum is always loosed from the
drive. The scraper itself is designed so that when hauled it dips and fills itself,
then floats on the material for conveying. When returned it quickly empties
itself. The change from stocking out to reclaiming is made simply by reversing
the cable hitches on the scraper.
2. Bridge crane. Flexible and of high capacity but requiring a heavy in¬
vestment cost.
3. Locomotive cranes. These both on rails and crawler treads. Flexible but
usually of higher cost than drag scrapers. Do not require formation of a
stocking-out pile as do drag scrapers.
4. Bulldozers and industrial trucks.

Link-BOt Co.

Fig. 12-4 Apron feeder shown removing coal from a track hopper and discharging
it to a crusher.

The stocking-out equipment is invariably used for reclaiming. Where a

track hopper is used; the reclaiming system ordinarily leads to it because
coal crushers and apron feeders at the beginning of the conveying system are
conveniently installed beneath the track hopper.
Live storage can be provided with bunkers, coal silos, and coal bins. The
word ^^bunkcr’’ is used here to describe coal storage that is not a part of a
building structure, but rather which is separately constructed and attached to
the building structure
The most common form of bunker is an overhead suspension type, usually
madeof 6.3mm plate steel,although reinforced concrete has also been employed.
Bunkers should receive a protective coating inside to guard their structure
against the corrosive effects of sulfur in the coal. Fig. 12-6 shows some forms
of steel bunkers. The vertical cylindrical type is usually mounted outside the
plant building, on the roof, or in its own separate structure. Troughed sus¬
pension bm kers arc inside mounted. The Berquist form approximates the ideal
 446 THE GAS LOOP
izontal transportation and dumpiiif? into live storage; and (3) gravity flow to
the combustion equipment. Besides transportation, this system may be in¬
volved with lump crushing, removal of tramp iron, weighing, and prevention
of segregation of sizes.
Common examples of hoisting equipment are:
1. Bucket elevator. Moderate lifts—continuous operation.
2. Grab bucket with crane or tower.

Fig. 12-7 Typical sj^oed reduction to con\c\or (lri\(‘ '-half (Shown with belt
con\ cyor.)

3. Skip hoifit. App]ical)le \\hen lifts are too high for bucket elevators.
Simple and compact.
4. Belt conveyor. Limited to less than 20° to the horizontal; therefore it
is necessary to start at considerable distance from the plant.
5. Electric hotfits to elevate larries or coal buggies which have been filled
at the storage pile and wheeled in to the plant by hand. They will be hoisted
until their contents may be discharged into the stoker hoppers.
Horizontal movement of coal to bunkers is by conveyors, of which the
following arc typical:
1. Belt convenors. Continuous system; belt usually troughed; high capac¬
ities possible.
_ 2. Spiral conveyors. Endless helicoid
screw in a trough. Can easily be made
dust-tight. Not having a return strand, it
requires a minimum of space. Limited in
length. Considerable wear.
3. Flight or scraper conveyors. Low in
first cost but having large energy consump¬
tion. There is considerable wear, caused by
friction and abrasion.
4. Pivoted bucket carriers. Probably
ideal from a good many standpoints, al¬
Stephena^Adamson Mfgr. Co.
though expensive in first cost. Material is
Fig. 12-8 Section through a carried and buckets are supported on rol¬
troughed belt conveyor. lers which reduce friction to a minimum.
The pivoted bucket carrier can both ele¬
vate and convey. Since it is run at low speed, the operation is both silent and
free from vibration.
COAL CONVEYORS 447
5. Larries. Suspended rail types are generally used because even ,though
they require more headroom than the floor type, they leave the boiler aisle free
from obstructions.
Design of equipment for coal handling involves capacity calculations as
well as selection of equipment type. What tonnage per hour should the coal
conveying machinery be abl(‘ to handle? The minimum requirements (not
allowing for the effect of load factor) would be the maximum rate at which
coal can be burned under the boilers. Expecterl plant extension should be

Cycle of oporotion for skip hoist

Sff phrriH-Adamton Mf^r. Co.

Fi(. J2-!) Skij) hoi^t.

anticipated by an initial installation of such dimensions that by increasing

oi)erating sjK‘e(l, or some other Minple modification, the conveying system can
continue to meet the needs ()therwi>e, enlarg(‘ments of coal conveying equip¬
ment may well prove an expensive proce.ss. Tlu* ultimate capacity of a power
plant is often initially uncertain; conseciuently conveyor systems usually have
eapaciti(*s of several times the maximum continuous demand of the combustion
equipment. Personnel duties planned for the station can effect selection of
conveyor capacity, for a daily stint at the coal conveying equipment may be
only one of the duties to be jierformed by a certain operator.
Coal conveying (‘quipment is not duplicated. Enough liv(‘ storage should

Fig. 12-10 Pivoted bucket conveyor. Combined coal and ash handling in a large
plant.
448 THE GAS LOOP
be interposed between it and the furnaces to allow for several hours' repair on
conveyors without running out of coal.
Full mechanical coal handling will not be justified in all boiler plants. But
even in the smaller plants, where the complete systems cannot always be af¬
forded, portable loaders, larries, etc., representing partial conveying, will be
found to be good investments.
Tramp iron is removed magnetically before coal is passed into pulverizers
or crushers
Means for weighing coal can be incorporated into the conveying system
(1) at track scales; (2) by weigh larries; (3) by coal spout volumetric meters;

and (4) by scale stations between bunker and dowmspouts. The weigh larry
has a small coal hopper suspended on knife edges in a scale beam weighing
system. Hopper, beam, and all are a portable structure on rails, whose align¬
ment suffices to carry the hopper under bunker discharge gates and then over
stoker or pulverizer hoppers. Some larries are motor propelled; small ones are
hand pushed. The hopper is filled from a bunker gate, the beam is put into
balance and read, and the hopper is discharged by gravity to the combustion
equipment. Larry scales are gener¬
ally built so that a prepared record
card when inserted will be marked
with the weight at which the beam
is balancing.
The distribution from bunker to
several hoppers is not as simple with
the automatic coal scales as with the
weigh larry. The automatic scales
Fig. 12-12 Flight conveyor. are best where one weigh station is
continuously and exclusively em¬
ployed with one combustion unit.
Hence, automatic coal scales are more likely to be employed in large plants,
and weigh larries in small. As Fig. 12-14 shows, the automatic scale feeds coal
into a weigh hopper which is suspended from a weigh beam. When a preset
load is registered, the beam comes to balance, stops the feeder automatically,
and the load is dumped into the lower hopper. The dump action causes a
count to be registered on the recording apparatus.
Coal Oates and Spouts, Coal will readily flow by gravity through square
COAL CONVEYORS 449
or round spouts of 30.5 cm internal dimension, provided the inclination is not
too flat. If the spout is more than 85® to the vertical, it should be arranged so
that operators may rap it if the coal should lodge. Gates should open easily
and close tightly, and be designed so that the gate and the cut-off plate can
be separately renewed. Coal discharge spouts are sometimes
fanned out so that the width of a stoker hopper is supplied by a
single dust-tight downspout. Special designs are required on these
distributing spouts in order to keep coal sizes from segregating
in them.
Conveyor Layout, The design and detailing of mechanical
conveyors are sufficiently specialized to lie outside the ordinary
scope of power i>lant engineering, and the plant designer cus¬
tomarily relies on conveyor manufacturers for advice on layouts.
However, space allowance, capacities, and the leading data must
come from the plant designer. Where the equipment itself is not
dust-tight, its vicinity should be isolated with dust-tight walls
and partitions so that the fine coal dust, which invariably flies up
wherever coal is dumped off a conveyor, will not settle over the
whole plant and discourage operators from maintaining a high
standard of cleanliness. Walkways need to be provided along¬
side conveyors for maintenance and inspection. After layout of S-A Co.

the system has been made, it should be studied with reference to Yig, 12-13
the possibility of coal piling up at conveyor discharges in the Bucket ele-
event of failure of any portion of the conveying system. Inter- vator.
locking relays are often applied to the conveyor motors to make
certain that conveyors and machines in scries with them are started and
stopped according to a predetermined order.
Belt Conveyors. This type is probably more universally used than any
other. The first cost is reasonable and the power consumption is low. It is
460 THE GAS LOOP
widely used for lorizontal movement of coal and, to a certain extent, for in¬
clined runs. Because of the varied conditions under which they are operated,
each individual belt conveyor installation should be
M given a thorough engineering analysis so that an eco-
MoQnttic nomical and, at the same time, adequate installation
will be had. A fair average speed of all belt conveyors
is 91.4 m/min. From ol - 91 4 m/mm is the usual range for
coal conveyors because higher speeds tend to shatter the
Tramp Iron discharge end. A minimum speedof 91A m/znin
hopptr—
CooiJ^
Coot-^.
should be used if the belt discharges over a tripper (Fig.
’ifi 12-6). There is no economy in running at a given speed
Fig. 12-15
Fig. 12-15 Iron
Iron load; it is better to reduce the speed and
separator. the belt full. Both flat and troughed belt conveyors
can handle coal, but the troughed are more frequently
used because their capacity is about 100% greater per inch width of belt. With
the proper design of troughed carriers and correct alignment of head and foot
pulleys, guide idlers are not necessary.
Troughed belt conveyors also handle coal successfully
at slants up to about 20®, but with capacity reduced 5% VToout
I spout \ \
to 10% as compared with horizontal conveyors. When in- \\30^X \\
dined, the conveyor must be equipped with a no-reverse
mechanism to prevent its running backward in case it is ?
stopped under load. The belt itself is made of multiple-ply ® *
canvas duck surfaced with rubber covers fur friction and [ ^ ^j
resistance to abrasion, ^s tl;^e belt itself will cost nearly i
ha^f tha^ conveyor installation, expert advice should a-s mm welded steel
be sought for its selection. The canvas may be varied in const r.
numbers of plies and in weight of the duck used (28, 32, ^ig, 12-16 Coal
36, and 42 ) so that it will have sufficient strength for gate,
tight side tension, be firm enough to prevent sagging be¬
tween carriers, and yet be flexible enough to pass around pulleys readily. There
is no practical limitation to length of these conveyors and, if well raain-

Table 12-1. BELT CONVEYOR PRACTICE

Minimum Pulley.
Weights of Material
Diameter— «m
(per ply per cm width Carrier Spacing
(Plain drive)
perm)
No. of plies
Belt width Spacing
28 duck, 0.0119 kg
6 8
32 duck, 0.0isa kg
36 duck, 0.014S kg
2 106.7 80.5-45.7 1JB2
42 duck, 0.0165 kg
.4 121.0 50.8-70.2 1.07
(1.0 mm) rubber cover,
0.010S kg
187*2 91-41524 0.91
Idlers,
8.05
COAL CONVEYORS
/
462 THE GAS LOOP
tained, they will transport millions of tons of coal before the belt needs re¬
placement.
Estimating Conveyor Requirements. Although, as previously mentioned,
the advice of conveyor specialists must be sought for final layout work, the
power plant engineer must make some decisions as to type, capacity, etc. In
estimating power requirements of various conveyors, the empirical relations of
Table 12-2 may be helpful, but estimates of space and support requirements
will have to be implemented by references to manufacturers' data.
The symbols in Tables 12-2 and 12-3 have the following meanings:
A, Working area of each flight, m* .
6, Belt width,cm, or bucket pitch,m.
Hj S X sin of the angle of inclination of the conveyor.
K, Horsepower Constant, see Table 12-3.
L, Conveyor length,®.
Ny Drivesliaft speed, rpin.
jS, Lineal speed of conveyor, m/min.
Ty Conveyor capacity,tonnes coal per hr.
w^y Individual bucket load, kg coal.
Wfy Weight of single strand, chains and flight, kg per m.
Table 12-2. CONVEYOR CAPACTTY AND POWER

Maximum Capacity Horsepower Required

Conveyor Type Tonnes j^er Hi- to Driveshaftf

Belt conveyor* 0.000404 W L_±_«:72 jo + T]t

9000 ‘ ^ *

Bucket elevator 0.07 b 1.8 X (Theo. hoisting power)

Flight conveyor* 48.055 AS 0.000198 WiLS 4- 0.002567 TL
Screw conveyor* kg jier Rev vs Dia 0.001 L(1.331 K + 7.338 T)

Die, cm 1^16 18.34 20.32 25.4

kg 0.136 0.544 1,5 8>63
30.48 35.56 40.04
5.0 9.53 12.25

*For horizontal conveyors. Add 77//273.7hpfor rise' on inclined conveyors. Belt con¬
veyors can be used up to 20° slope with little los.s m capacity.
t This is power to conveyor drivesliaft. For motor jiower allow foi losses as follows:
For each pair gears, sprockets, sheaves:
Cut gears Cast gears Steel roller chain Plain chain V-helt
5% 10% 5% 10% 5%
For each manufactured speed reducer unit:
Helical gear type, 5%; worm gear type, 10-20%.
t Add tripper power where pertinent. See Tabic 12-3.

Example 1: The size and drive shaft power for a 12.2 m flight conveyor will be
estimated. It is inclined 20° and will carry 18.1 tonnes of coal per hr at 80.5 m/min.
From: Capacity = 48.055 A3, the

working area of flights = 18.1/48.055 x3615= 0.01235 m*

PULVERIZED COAL SYSTEMS 453
Table 12^. HORSEPOWER CONSTANTS FOR BELT CONVEYORS •

Belt width
Con} K Tripper Hp Belt Type

35.6 22.3 U.6I+O.OOSO T 5-ply, 28 duck

40.6 23.8 0.61+0.0039 T 5-ply, 28 duck
45.7 28.3 0.79+0.0089 T 5-ply, 28 duck
80.8 31.3 0.88+0.0046 T 5-ply, 28 duck
'1 87.2 1.03+0.0046 T 5-ply, 32 duck
76.. 60.6 1.43+0.0066 T 5-plyj 36 duck
91.4 1.82+0.0056 T 6-ply. 32 duck
106.7 74.4 2.10+0.0061 T 6-ply, 32 duck
121.9 89.3 2.49+0.0067 T 6-ply, 42 duck
137.2 3.04+0.0078 T 6-ply, 42 duck
162.4 114.6 3.864.0.0089 T 8-ply, 42 duck

On account of partial ’filling, conbicler working area to he 50% of flight area. Then

Flight area = 0.0247 ma

Select 30.48 cm x 10.16 cm rectangular flights.
Strand weight Wf assumed to be 17.86 kg/m.
Again referring to Table 12-2,
Power for horizontal conveyor = 0.000198x 17.86x12.2x30.5+0.002567x18.1x12.2 =*
1.88 hp
Power for lifting =18.1 x 30.6 »in 207273.7 = 0.69 hp

Total power to drive sprocket 1.88 + 0.69 = 2.67 hp (say 3 hp motor)

Example 2: Principal data for a horizontal belt conveyor will be estimated. Given
T =4.j.4,L =76.2, iS = 91.4
From Table 12-2, Capacity =0.000404852, go

lielt width = V45.4/(0.000404 x 91.4) = 35.08 cm, say 36.6 cm belt

76.2+46.72
I'ower = --- [(0.06 X 22.3 x»M) + 46.4] - 2^1^^
9UU0

Allow for trip])cr 0.61+0.0039x46.4 ==. 0.79 hp

Total to drive pulley == 3.06 hp
Assume speed reduction in drive consisting of 2 sets cut gears and 1 set cast gears
(Fig. 12-7). For this arrangement, add 20% for drive losses. Motor power = 3.06 X
1.20 = 3.67. Next larger standard size is 5 hp. From Table 12-3, belt specification is
5-ply, 28 This with two 1.6mm. rubber covers will weigh (5x0.0119+2x0.0102)
36.6 « 2.84 kg per m or 433.5 kg for the conveyor.
Pulley diameter estimated 71.1 om from Table 12-1.
Pulley speed 91.4/0-71 hr =41 rpm
Over-all speed reduction required from 1750 rpm motor = 1750:41 or 43:1.
C^j»fijL4;pll spacing, 1.22 moc; idlers, 3.O6 moc.

/12-4 J^ulverized Coal Systems. The burning of coal in pulverised form

h^^^^b^ailpreviously discussed. To prepare the coal for this system of combustion
454 THE GAS LOOP
is, of course, more difficult and costly than by the other methods of firing. Like
most other improvements the use of pulverized coal as a boiler fuel is attended
by some specific disadvantages that tend to limit the range of applications in
which pulverized coal is the sujKTior system of combustion. In general, the
field of application is the large power
plant; hence it is important in the
electric power station. As part of gas
loop study, it is pertinent to examine
the system which supplies the finely
ground coal to the burner.
There are two systems of prepara¬
tion of pulverized coal. One is the
central system (bin system); the
other, the imit system. A central pul¬
verizing system employs a limited
number of larger capacity pulverizers
at a central point to prepare coal for
all the burners. Driers, if required,
are conveniently installed at this
point. From the pulverizers the coal
is transported to a central storage bin
where it is deposited and its trans¬
Fig. 12-18 Elements of pulverized coal porting air vented from the bin
system. through a ^^cyclone.^^ This bin may
contain from 12 to 24 hours^ supply
of pulverized coal. From the bin the coal is metered to the burners by motor-
driven feeders of varied design. Primaiy air, added at the feeders, floats the
coal to the burners.
Central system advantages arc:

1. The pulverizing mill may work at constant load because of the storage capacity
between it and the burners.
2. The large storage is a jirotection against interruption of fuel supply to the
burners.
3. Offers good control of coal fineness.
4. More latitude in the arrangement and number of burners is allowed the de¬
signers.
5. The boiler aisles are unobstructed.

Central system disadvantages are:

1. Additional cost and complexity of a coal transportation system.

2. Central preparation may require a separate building.
3. Driers are usually necessary.
4. Fire hazard of quantities of stored pulverized coal.

Most pulverized coal plants are now being installed with unit pulverizers.
The unit system is so-called from the fact that each burner or burner group
and the pulverizer constitute a unit. Crushed coal is fed to the pulverizing mill
at a variable rate governed by the combustion requirements of the boiler and
PULVERIZED COAL SYSTEMS 465
furnace. Primary air is admitted to the mill and 1>ecomes the transport air
which carries the coal through the short delivery pipe to the burner. This air
may be preheated if mill drying is desirable.
Unit system advantages are:
1. It is cheaper than the central system.
2. It allows direct control of combustion from the pulverizer.
3. Coal which would require drying in order to function satisfactorily in the central
system may usually be employed without drying in the unit system.
4. In a replacement of stokers, the old conveyor and bunker equiiiment may be used
5. There is no complex transportation system.

Unit system disadvantages are:

1. The mills operate at variable load, a condition not especially conducive to best
results.
2. With load factors in common practice, total mill capacity must be higher than
for the central system.
3. Firing aisle is obstructed with pulverizing equipment, unless the latter is rele¬
gated to a basement.

Coal is pulverized in order to increase its surface cxi)osure, thus promoting

rapid combustion without using large quantities of excess air.^In a typical unit
system, lump coal, crushed to uniform size, is continuously supplied to the
pulverizer hopper, whence it is withdrawn by a feeder and sent to the pulver¬
izing section. Rate of combustion is controlled at the feeder. Pulverization
being accomplished, the coal is next swept out of the mill and floated to the
burner, located in the furnace wall, by admitting enough of the combustion air
at the pulverizer to accomplish air-bome transportation. This is called primary
air, in distinction to secondary air wdiich is supplied directly to the burner. The
primary air may vary from as little as 10% to almost the entire combustion
air requirements, depending on the load and type of pulverizer.
The pulverizer is the most important part of a pulverized coal system. The
degree of fineness to which it is expected to reduce the coal has been mentioned.
To accomplish this on a large commercial scale wdth variable coal moisture
content is a task whose successful accomplishment bears tribute to the inge¬
nuity of equipment designers. In general, pulverizers (sometimes called mills)
may be classified as attrition or impact types. To these might be added ‘‘shear¬
ing” types; however, tbe latter might be considered a form either of the at¬
trition type, impact type, or both. The impact mills generally have some
attrition action present and, conversely, while attrition may be the primary
action of a mill, almost invariably impact is present as a secondary action;
consequently, a mill classification really represents a statement of the most
prominently used principle.

Impact Mills Attrition Mills

Ball mills Bowl mills
Hammer mills Ball and race mills

A general description applicable to any pulverizer would include the means

and method of size reduction, of fineness regulation, of air introduction and
456 THE GAS LOOP

coa/
Z -'Win€tbf0-sp€ta
fk€e/0r cfriye
initf
ifrom ,,-Po//-pressure
fktdtr} spring
|M, ''-Xb/Ac/ra/wjcie
! - Qe/Justrnenfroe/
spout
"-Orine/ing ro//
Grinding nng

'^'//ewtvingsweeps
\'-Revo/¥ing bow/
" 'Tramp-iron
discharge

B
''Pyrites remotia/pipe
Spira/-f/ow To exhauster and burner Preheated
classifier ,-From coa/ feeder airduct
»Sound-proof

Automatic
shutoff
damper

if /
\l ' / / *■ - A- Hdter(oiMtdbearingj

\ Vuer^size and raw coa/

"Oyer-size return to mi//

A—^Ball and Race Mill C—^Ball Mill

B—^Bowl Mill D—Hammer Mills
Fig. 12-19 Coal pulverizers.
PULVERIZED COAL SYSTEMS 467
movement, of feed, and of drive. Pulverizers are driven by electric motors with
the feeders either actuated by the main drive or by a small d-c motor, de¬
pending on the type of control used. Integral fans provide movement of air
to float the pulverized coal through the short duct connecting with the
burner. Air separation is frequently employed to control fineness, but some use
is made of ‘Vhizzers” or ^^spinners” (throw-out vanes for oversizes), entrap¬
ping ribs or vanes, centrifugal force, etc.
Fineness of the pulverized coal is important since it affects the incomplete
combustion loss, tube fouling, performance of flue gas cleaning systems. To
increase the fineness of pulverization (as measured by the percent of a sample
passing 75 micron screen) is to increase the cost of pulverization, to decrease
maintenance* to improve efficiency of combustion, and to reduce duty of dust
collectors. The fineness chosen is economic when the incremental cost of further
increasing the fineness exceeds the corresponding summation of the resulting
savings. This varies somewhat with the size of the unit, but is in excess of
75^0 fineness. It i^ equally important that the remaining 25% not contain a
substantial portion of large i)ieces. Thus a requirement of 98% or more through
a 300 micron screen is also imposed.
Fig. 12-19 shows the ball and race pulverizer, the designation referring to
the appearance of the grinding elements themselves. This is a low-speed unit
in which grinding pressure is maintained by adjustable springs. The bowl mill
shown in Fig. 12-19 grinds the coal between a whirling howl and rolls mounted
on pivoted axes. Coal fed into the center is thrown by centrifugal force against
the sides o? the bowl w’here it is pulverized between the sides of the bowl and
the grinding ring. The fine and intermediate sizes are picked up from the top
by an air current and carried into the separator above for classification.
The ball mill operates something like a foundry tumbling barrel. The large
rotating drum (100-200 rpm) contains a quantity of iron balls mixed with the
coal. As the drum turns some balls are carried upward to be dropped on the
coal w'hile others, remaining in the agitated mixture, grind the coal at random
between them. The coal is fed into one end and reduced in size by this action
until it can be swept out of the mill by a current of air.
Hammer mills have swinging hammers or bars, into the path of which is
fed the coal to be pulverized. Fig. 12-19 show’^s hammer mills, one having a
single impact stage followed by attrition action, the other having two impact
stages.
The very nature of the action in a pulverizer insures that there will be
considerable wear and that a great deal of energy will be consumed per tonne
of coal. Hammer faces, balls, fan blades, and the like need periodic replacement.
By employing hard alloys and by designing so that the labor and time of re¬
placement are minimized, costs of maintenance are reduced. The life of the
wearing parts these days extends through the pulverization of thousands of
tonnes of coal. The amount of energy input per tonne of coal varies with the
grindability of the coal. Low grindability indexes, such as that of anthracite
(4), are indicative of the greatest pulverizer wear and energy consumption.
The average bituminous coal index is around 60, but there may be considerl^ble

*Tube lancing, slag cleaning, soot blowing.

458 THE GAS LOOP

Kennedy-Van Saren Mfg. ^ Engr, Cotp.

Fig. 12-20 Boiler equipped with water walls and fired by unit pulverizers.

variation from this in individual cases. The kw hr of energy required for the
conveying and pulverizing of a tonne of bituminous coal varies from 11 lo 33; a
fair average being 22 kw hr per tonne. Anthracite might use 38.6 kw hr per tonnCL
12-5 Oil and Gas Supply Systems. Gas, The cost of storing gaseous
fuel is too high for storage to be economically practical in power plant work.
Nature has gas m storage in the form of gas deposits at high pressure in gas
sands. When wells are drilled into these strata the gas can be drawn off through
pipe lines—some of which extend hundreds of miles cross-country to customers,
domestic and industrial. Friction in the pipe line flow consumes fluid pressure
so that gas compressor stations must be located at intervals of 32 to 80 km.
along the lines in order to renew the pressure and continue the flow. When this
pipe line gas finally arrives at a boiler plant, it will have a fluid pressure in
excess of that customarily used in the combustion equipment, and one or more
pressure reducing stations are installed. The plant layout, insofar as a gas
OIL AND GAS SUPPLY SYSTEMS 459
supply system is involved is, therefore, quite simple. From pressure reducing
and metering points it flows to the gas burners where combustion is effected as
was described in Chapter 10.
Gas burning systems are(l)low pressure, i.e.,0-0.07 kg/cm*ga,and(2)highrpr-
essure i.e., 0.35-1.41 kg/cm^ga. It may be practical to install the gas meter on therfi -
nal reduced pressure inhigh-pressure systems,but an intermediate pres8ure(033-*l41
kg/cm*ga) is preferable for meteringwith low-pressure systems in order to reduce
gas volume and meter size. To meter gas exactly the plant gauging station
should have recording pressure and temperature gauges so that gas meter
readings can be corrected for these variables. The remainder of the system con¬
sists of the branching pipe ^'tree’^ to the various burners, with valves, drains,
boiler gas meters, and by-passes.
Oil. Unlike gas, fuel oil compares favorably with coal in storage bulk. Oil
will contain about twice the heating value of coal per cubic metreof storage.
This ratio is favorable, but the cost of storage volume is not, since coal can
be piled on the ground with little previous preparation,, whereas oil must be
contained in tanks. After oil arrives at a ])lant, the operating expense of
storage and reclamation arc likely to be less than for coal by virtue of the
fluidity and the pumping method of transportation.
Fuel oil is transported in barges, railway tank cars, and motor trucks.
Boiler plants that can be located on a railway system will usually receive fuel
in tank cars. On U. S. railroads these are30,283; 37,854 or 45,425 litres capacity.
Barge loads are much greater.Highwaytruck-trailerunitsareaboutlS,927litres.
It is desirable to have sufficient tankage so that several days* operating supply
remains in other tanks wlien a tank is empty and ready for fill by railroad or
truck delivery. Because of differences in production methods it is not con¬
sidered essential or economic to store several months* supply, but rather to
provide only against delivery failures originating from flood, storm, and the like.
Present-day demands for motor fuel are great and, in meeting them, whether
by distillation or cracking, refiners accumulate large stores of heavy oil suitable
only for furnace combustion. While the burner equipment is somewhat more
elaborate and expensive than for light oils, where properly chosen it will bum
the heavy (No. 6) oil efficiently and cleanly. Hence attention is focused on
this fuel since supplies are abundant and cost is reasonable.
Table 12-4. DIMENSIONS OF BULK STORAGE TANKS

Capacity Plate Thickness Weight

Diameter I^ength
m m (litr^) (mm) m
2.44 2.64 11860 6.36 1844
2.44 6.08 23720 6.35 2906
2.44 7.67 36344 6.35 3967
2.44 10.06 46969 6.36 6028
3.05 5.08 37067 7.M 4273
8.06 7.67 56229 7.94 6804
8.05 10.06 78892 7.94 7885
8.06 12.66 91668 7.94 8868
3.05 16.04 109716 7.94 10899
460 the gas loop
The regulations of the National Board of Fire Underwriters require oil
storage tanks to be constructed of steel, wrought iron, or concrete, the latter
not being permitted for oils lighter than 35° Baume. When the storage tanks
are located above ground and are liable, in case of breakage, to overflow and
endanger surrounding property, each tank should be protected by a continuous
embankment or dike. Inside storage is much more hazardous than outside and
should be considered only when there is no possibility of outside storage.
Then the tanks should not be located above the lowest story or basement of
the building and should also be located below the level of any piping to which
they may be connected. The Underwriters' rules state that the normal gross
capacity of such tanks shall not exceed 18927 litres in ordinary buildings nor
56781 liters in fire-resistive buildings. However, under special conditions, and
where the oil is stored in a specially constructed room,! 89270liters may be stored.
Temperature degrees fohrenheit
50 60 70 80 90 100 110 120 130140150160 170
Easy pumping fluidity

A Cent atomizers
B Steam

While light oils such as Nos. 2 or 3 are sufficiently fluid at all normal tem¬
peratures, heavy oil must be heated both for pumping and atomizing. Pumping
temperatures are dependent on viscosity. The chart in Fig. 12-21 will permit
selection of a pumping temperature. Atomizing temperatures are dependent on
burner design and manufacturers' recommendations govern. The chart may be
employed to determine required preheating for pumping and atomizing oils
of different viscosities. It is entered with the viscosity and temperature of the
viscosity rating. These determine a point on the chart from which projections
parallel to the slope lines will intersect the desirable pumping and atomizing
fluidity lines. The abscissa at the intersection is the preheat temperature.
Example 1: Principal data will be estimated for an oil supply system to be used
with a 304-hp ‘^package” boiler, fired with No. fi fuel oil, Be 12®, viscosity rating 155
SSF at 50® C* Pressure atomizing type burner. Estimated capacity factor, 33%.
Delivery by tank truck biweekly. Minimuni oil reserve to be a supply for one week.
OIL AND GAS SUPPLY SYSTEMS 461
This type of steam generator can be expected to get better than 75% thermal
efficiency. Using 75% as the average efficiency, the maximum fuel flow is calculated as
follows:
Eq 5-8: Qh « 42450+98-04(12~-10) - 42,264 J/g*"
140
Eq 5-2: S.G. = ——— = 0.985
^ 130 + 12
„ 804x8S23x4.187x J0» .
Fuel flow - (i.7Sx42264 x 60xlU«r “ »*” P*'
Fuel used per 2-wefek period = 5.657 x 60 x 24 x 14 x 33% » 37635 !

Were three 2.44 mx 5.08 m tanks to be installed, each would take the full load of a
truck-trailer transport. No. 1 could he in use while No. 2 lay empty awaiting delivery.
No. 3 would contain somewhat more than the specified reserve.
Since 62 baybolt Seconds Furol equals 600 Seconds Universal, this oil is 1500 SSU.
Enter Fig. 12-21 at 1500 and 60®C. Locate the intersection and project upward to
the left to find that tank oil temperature should be at least 84.4®C. Projection downward
to the right to line C shows that the oil should be preheated to 91.rc before atomiza¬
tion.
Tank heating is done with a steam or hot water coil immersed in the oil,
located so as to surround the suction opening. Heating for the burner is done
by electric and steam heaters in series. The electric heater under thermostatic
switch control comes into action only w^hile starting up, after which the steam
heaters override it and the thermostat keeps the electric circuit disconnected.
These heaters are generally located as near the burner as practical.

Fig. 12-22 Typical oil supply systems.

Since automatic oil burners commonly return a large part of the quantity,
drawn from the tank, back to the tank, and since all of the suction is heated,
the burner oil heater ordinarily supplies all the heat needed to keep the whole
system sufficiently fluid. Exceptions are the very long suction lines, exposed
piping and/or tanks in cold climates, long shut downs, as over week-ends, etc.
In such cases patented electric heating which warms the piping with electric
current is helpful. This system needs to have all piping well insulated and uses
special electric supply to produce the heating currents, utilizing the pipe itself as
the conductor. Another method is to install a steam or hot water line with the
oil lines inside a blanket of heat insulation.
 462 THE GAS LOOP
The major field of application of oil burning has been the nonutility field,
generally the small plant, often with a single burner. Oil supply then becomes
relatively simple since all the automatic burners have in-built oil pumps and
can draw the oil from storage provided it is reasonably located, say not over 100
ft away horizontally or morethanS.OS mstatic lift. Otherwise transfer pumps
are required. With multiple burners installed, transfer pumps will create maxi¬
mum flexibility of oil supply.
12-6^ Flue Gas Cleaning. The products of combustion of coal-fed fires con¬
tain particles oTsolid matter floating in suspension. This may be smoke or dust.
If smoke, the indication is that combustion conditions were faulty, and the
proper remedy is in the design and management of the furnace. If dust, the
particles' are mainly fine ash particles called ^‘fly-ash” intermixed with some
quantity of carbon-ash material called “cinder.” Pulverized coal and spreader
stoker firing units are the principal types causing difficulty from this source.
Other stokers may produce minor quantities of dust but generally not enough to
demand special gas cleaning equipment. The two mentioned are troublesome be¬
cause coal is burned in suspension—^in a turbulent furnace atmosphere—and
every opportunity is offered for the gas to pick up the smaller particles and
sweep them along with it.
Any sample of atmosphere will be found to contain some matter other than
gaseous, but, of course, ^ widely varying percentages, dependent on the time
and place of collection^he power planiL_engiiiecr is inteiMed mainly in the
quantity and effect of the solid content of the products of combustion his equip¬
ment discharges from the plant chimney. Dust collection is a comparatively new
subject to the power plant engineer, for until pulverized coal was introduced,
the main nuisance that could be created was smoke. Nowadays there are often
legal restrictions on the dust content that may be emitted and there is always the
possibility that adjacent residents may bring damage suits resulting from the
fall of ash or cinder on their premises. Consequently dust collection is often of
importance even in the absence of such things as municipal ordinances.
Nomenclature, As the highly specialized field * of dust collection is liable to
be unfamiliar to the reader, it may be well to mention some of the methods of
measurement. The size of dust particles is measured in microns. The micron is
one millionth of a meter. As an indication of the scale of this measure, the
diameter of a human hair is approximately 80 microns. Typical classification of

Fio. 12-23 Typical particle sizes. A. Flue gas particles and ranges of collecting equip¬
ment. B. Typical distribution of particle size in products of combustion.
* Boiler plant chimney dust is but a small segment of this field. Dust collection is
practiced in industries such as petroleum refining, food products, cement, foundries, smelt¬
ers, etc.
FLUE GAS CLEANING 463
particles by name is given in Fig. 12-23, but it must be imderstood that ,the
limits shown are, for the most part, arbitrary. A critical characteristic of dust is
its Settling Velocity in still air. This is proportional to the product of the square
of micron siae and mass density. Manufacturers often assume a specific gravity
of 2.0 for flue gas dust. Typical values:18.3cm/minatl0microns,!8.3m/mm at
100 microns. With settling velocitv of 61 cm/min or less the dust will be carried
from 1.6 to 4.8 km in a light breeze, from an average chimney height.TheGas Lo-
oding is the weight of dust particles per cubic foot of flue gas. It can range from
4.6 to 27.6 grams per m* with typical operating conditions represented by the
minimum value.
The removal of dust and cinders from flue gas can usually be effected to the
required degree by commercial dust collectors. These can be classified^ as me-
ohanical ajid electrical. The mechanical collectors are subdivided into wet and
dry types. Wet types. callecTscrubbers. operate with water sprays to wash dust'
from the air. Such large quantities of wash water are needed for central station
gas washing that this system is seldom used. It also produces a waste water that
may require chemical neutralization before it can be discharged into natural
bodies of water. Hence the common mechanical dust collector is the dry type,
which can be subdivided according to operating principle as follows:
^_Gravitational Separators. Act by slowing down gas flow so that particles
remain in a chamber long enough to settle to the bottom. Not very suitable
because of large chamber volume needed.
Inertial Separators. Act by rapid change of direction of gas, which cannot
l^e^llowed completely by the heavier particles. Common forms are the baffle,
the louvre, and the cyclone separators. Baffle separators are frequently impror
vised within a boiler setting in-order to drop the large cinders from the^ases.
Because volume is less than that of the other types^ louvre and cyclone separa¬
tors are favored. The former is a high-speed gas type in which a small portion
of the gas carries the bulk of the dust into a secondary chamber where velocities
are low enough for a combination of gravitational and inertia forces to make the
separation. Meanwhile the main gas flow passes out the side of the direct-flow
chamber. A separator of this type is shown in Fig. 12-24. The cyclone is a
separating chamber wherein high-speed gas rotation is generated for the purpose
of “centrifuging'^ the particles from the carrying gases. Usually there is an
outer downward flowing vortex which turns into an inward upward flowing
vortex. Involute inlets and sufficient velocity head pressure are used to produce
the vortices. As multiple, small-diameter vortices with higher pressure drops
appear to have high cleaning efficiency, that type is now being exploited. Skim¬
ming cyclones shave off the dust at the periphery of the vortex, along with a
small portion of the gas flow. This concentrated flow is then led to a secondary
chamber for final separation.
Some of the aforementioned principles are shown in Fig. 12-24. There are
numerous proprietary patents in this field. Though the underlying^ principles
may be few and simple, they are capable of various ingenuities in application,
each with sufficiently 'original features to have been patentable. Also, while
some separate the dust from the gq.a mor^ <*ompletely tbkiLxrfiiftrs, the draft lOflS
requirements differ, and the more efficient the gas cleaning, the higher the price
in draft. Thus baffle and louvre types may be builttooperate(ml2.7to25.4inm
464 THE GAS LOOP
water static pressure loss, but cyclones must have from 38 to 76iiiii)Jn addition
to the occasional patented features, much specialized knowledge on gas velocities
and decelerations, particle inertia, and the like must go into collector design,
E^trical PredpitatoT. This separation principle consists of imposing an
electncal charge on the particles as they pass near a collecting electrode of
^posite polarity. Gas is made to flow between grounded collecting electrodes
between which are suspended the highly charged ionizing wires. The particles
are attracted to the collecting electrodes and stick there until removed by being
jarred loose with electrode rapping or some like method.
By means of transformers the pressure ot the a-c supply is raised sufficiently
high, then made unidirectional either by synchronous rotarv switching or
vacuum tubes. The d-c voltage is hold at 30,000 to 60,000 v, dependent on
electrode spacing, particle size, etc. For good results this voltage needs to ap¬
proach the flashover voltage between electrodes.

Buell Engr. Co Wc'rtmi Precipitation Corp.

louvered Cyclono Electrostatic^

Fig. 12-24 Representative dust collectors.

Although the draft loss of the electrical precipitator is the least nljdLfonQs,
its operation requires a steady input of electrical energy which can well be much
more than the extra fan power used.to boost the gases through mechanical col¬
lectors. Also the. first JCQst is high. However^ in spite of costs, it is frequently
employed with pulverized coal-fired power stations for its effectiveness on very
fine ash particles is superior to that of any other type. As is typical in the dust
collection field, this principle is applicable to general industrial problems, and
flue gas cleaning is a small segment of its field of use. Also it is proprietary,
having been originally developed by Cottrell.
Installation. Dust collectors are installed between the boiler outlet and the
chimney, usually on the chimney side of thefair heat^ if there is one. There
'^uld be some advantages from the standpoint of ^ater cleanliness were the
collector to be put ahead of it; however, the practice seems tp be to follow with
the collector, and use spotJbJowers to keep the heater surfaces, jclean. Where
FLUE GAS CLEANING 466
there is more than one boiler, the practice is to use an individual collector for
each boiler. In somlp cases a low resistance inertial and an electrostatic precipi¬
tator have been installed in scries, again with pros and cons as to which should
be ahead of the other. Generally the mechanical type is placed first in the gas
flow. Another characteristic of interest in a combination is the variation of col¬
lection efficiency with gas flow. As flow increases, the. electrostatic efficiency de-
cjreascs, tlLe^clonc efficiency increasegL
Collectors arc bulky. The space they consume and the necessary breeching
connections arc costly—and difficult to provide, for, as an after thought. Hence
the need for a collector should receive careful consideration during the original
power plant layout.
Reinjection, The dust and cinders which are collected by this equipment fall
intojiioppers from which they must be removed by some system. Two altema
tives are: (1) dispose to a fly-ash bin with a pneumatic ash transport system;
(2) reinject the collection into the furnace, borne upon an air jet furnished by
a reinjection fan capable of about25.4
cm water static pressure. The first al¬
ternative is proper where there is but
little carbon in the dust—as in pul¬
verized coal plant fly-ash, but the
second may be desirable with spreader
stokers, for the material collected
may have sufficient carbon content
that the boiler and furnace efficiency
might be raised as much as 1% by Fio. 12-25 Typical fractional efficiencies
completing the combustion. However, A. of dust collectors.
the desirability of reinjection is a
matter on which the industry is not of one mind. Consider what happens when
the ash and cinder from a mechanical collector are sent back to the furnace.
Probably only a part of the ash gets permanently mixed with the ash on the
grates; the rest recirculates to the collector and back to the furnace. This proc¬
ess reduces the average micron size and increases the gas loading until an
equilibrium is established where the collector passes superfine fly-ash to the
chimney. Used in this way the collector merely assures minimum carbon loss
and a fly-ash state suitable for good atmospheric dispersion. The obvious an¬
swer to a need both for no-carbon-loss and no-fly-ash-discharge is to follow
the mechanical collector with a Cottrell whose collections are not reinjected.
But it is equally obvious that this is costly and not likely to be adopted unless
local conditions absolutely demand it. Two-stage mechanical collectors which
reinject coarse particles and send fines to storage have been built.
Efficiency, The absolute efficiency of a dust collector is the percentage of en¬
tering solids that will be removed by the collector. Some manufacturers prefer
to rate their equipment on an efficiency curve, as illustrated in Fig. 12-26. For
example, the curve for cyclpne., type showa .91%.^.15..JThis can be in¬
terpreted as meaning that this collector will remove 91% of the solid matter
over a limited range centered on 15 microns, say the 10 to 20 micron group of
dusts.
When variable-load operating conditions are considered, it will be seen that
466 THE GAS LOOP
dust collection is a very inexact science. Gas volumes, gas loadings, and size
distribution all vary and are not built with regulating features; the specifications
and design must be pointed at some assumed ^'normal operating condition^' and
acceptance tests conducted at that load. Since the potential nuisance value of
dust-laden fliie gas increases with boiler loading, some high-load condition is
selected as the specification point and not an average load.
Actual performance is determined by testing, commonly with the Bagtest
Sampler shown in Fig. 12-26. This is an apparatus for withdrawing a sample of
dusty gas and filtering out the dust. Weighing of the filter before and after a
timed collection period, together with data on relative area of sampling nozzle
to flue gas passage, will suffice to establish the collection efficiency if the gas
passage is traversed both at the inlet and discharge of the dust collector, load

Fig. 12-26 Dust sampler.

remaining steady meanwhile. In order to obtain average samples the duct cross-
sectional area must be subdivided into elements and a reading taken at the
geometrical center of each. Furthermore for a true sample the gas velocity into
the sampling tube must be the same as that of the surrounding gas flow. This
considerably complicates the testing, for preliminary pitot-static traverse must
be made and then the rate of flow into the sample nozzle adjusted for the same
velocity. This explains the reason for some of the components shown on the
sampler.
Example 1: The results of a test* for collector performance will be calculated
based on assumed data. Methods of measuring gas flows in ducts are given elsewhere,t
so it will be assumed that a pitot-static test has yielded a veloeityofAO.67 m/sec in the
80.5 omx 182.9 om inlet to the collector,and 7.01 m/seojn the 91.4 omx91.4 oro breeching to
which the collector discharges. Flue gas temperature in main flow 266^ in sampler 138T>
Sampling nozzle area, 5.074 om* I orifice area, 3.252 cm* with taps located so that
the coefficient is 61%.

First the differential manometer reading for proper rate of flow at the nozzle will be

determined. Where double data appears in brackets, it is |"dowM^e^ J*

Required rate of flow into supply nozzle = 6.574 [ 7M ']-[SS] om*/geo.

Bequired rate of flow through orifice t = Jcni»/.ec.

♦ The conduct of a test is the subject of an ASME Code. An examination of the Code
will reveal that this example greatly abbreviates the procedure.
tSec 12-9.
t This is a temperature correction. Pressure correction is considered to be negligible.
 ASH HANDLING 467
Note that: Orifice fiow area X coefficient X ideal velocity

Ideal velocity V^*, where H = - ILUv

In the above y is differential reading on the manometer in ^ of water and 0.881 is

gas density (Fig. 12-44). Combining these relations, the following equation is developed
for guidance in adjusting the apparatus.

y =114060 X (volume of orifice flow)* cm water Q535 t

Hence the apparatus must be adjusted for manometer readings of 114060| 10*6 J or

[low]®” "'***'■•
Asslfme that the Bagtest Sampler flow was so adjusted and that samples were taken
for 30-min runs, showing collections of 0.0816 kg upstream 1 and 0.0154 kg downstream.
Then
0.0816^0.0164
Collector efficiency =
0.0816
- 81%.

Rate of flow of solids into sampler = ^ x gO [ “ [^OOM J ***“*

Gas Loading = 0,0453 x 106/5947 » 7.62 grains per m6 entering collector and 0.0086x
106/3907 am 2.2 grams per m* leaving.
80l5x 182-9 60 x 7.62
Total flow of solids from boiler =10.67 x—— X —= 2.72 kg per min.

Hopper collection = 2.72x60x81% ~ 132.2 kg per hr.

12-7 Handling^ All coal has more ordess ash. Combustion of the coal
is attended by the necessity of providing some means of removing the ash
which is deposited in the ash hopper. All the ash should be in this hopper, but,
as a matter of fact, from 5 to 40% of it leaves the furnace with the gases carried
in suspension. The constant endeavor is to reduce the percentage of ash particles
leaving with the gas because their collection and handling are more difficult than

Remoining fly osh

□
Steam
generator
Fly osh
Eb Fly ost>rdmovol (Portion
fPv^- Stock sproys
suspended

*1!
^ Electrical precipitation*
m gos streoml
i.WetBoffles
t Trops 5 centrifugol seporotere
All or bulk Speciol bloded fon
of oih to
I ut r
Molten osh
Convtyor system 1 .Continuous flow
2.Periodicolly
Oischorge to Topped >oiid osh
L Ash dump.
2.R.R Cor^ 1 Hydroulicfill 1 Hond raking
2 Settling tonk ^Conveying system
B.eorps 3 Dryoshpit 2 Grovity Dump
^ 4 j^otor truck 4 Ash bunker
3 Pneumotic conveyor
4 Steom jet conveyor
5 Ashcors 5cort»
6 Wheel borrows

Fio. 12-27 Outline of ash disposal equipment.

♦ Here the reader may wish to review velocity head theory. If so, see Ex 1, Sec 12-9.
468 THE GAS LOOP
if they were in the ash hopper. Ash handling is a ipiajor problem because, first,
the ash is dusty, hence irritating and annoying to handle; second, it forms
clinkers by fusing together in large lumps which must be broken before given
to any reasonably sized conveying equipment; third, ash is abrasive and will
wear all conveyor parts on contact with it if there is any relative motion.
Ash disposal systems are designed for intermittent or continuous operation.
An ash-handling system will consist of a means for removing ash from the
furnace and loading it onto a conveyor system, the conveyor to deliver the ash
to fill or storage, and a means of disposing of stored ash.
Ashes can be raked from ashpits to boiler room floor, then either shoveled
into wheelbarrows, cars, etc., or raked to gratings where they will fall into a
carrying system. If there is an ash basement with ash hopper, the ash can fall
directly into an ash car or conveying system. See Fig. 12-27 for an outline of
ash-handling methods.
Ashes are generally disposed of to railroad cars or to trucks for local fills.
Loading of them may be by hand (small plants), by grab bucket from ashpit, or
by gravity from elevated storage bin to which ashes arc delivered by bucket ele-

::-i,_M—i,_/
\. W W / \ flTlSI /
FunwriM

\!iS/

'MfKIIIIII

I*

B§oumo«t*Bifch Co,

Fio. 12-28 Steam-jet, vacuum-type ash conveyor.

Am PREHEATER 473
CALCmATIONS FOR FIUI COEFFlCUiNTB, h

'^Side X«r SuU

Strwm dmwity ^ m.4*0«0.4W (Fig. 12-44) Stream density ® 68.0*b1.040
StrMun BMMI d«M^VMO.Oa07 Film density @ 108.8*aa0.875 • P ■* 0.0801
Stavua n -Sl.O X (Fig. 0-6) Film M -28.4 X lO*'
Stream k •AonS Film k -0.0878
Cp - 0.25 X«.81-14ai Cp - 0.24 X 0.81i.lt.864
Vt - MW/(|00OxO.60»).ll»m/aM t«w - 1.10x0.00-1.07 kB/M
r. - 1.07/(0J07X1.040)-8.48 a^tao.
18.600 Re - 0.0048x8.ax0.0801/88.4xlO-« -18,880
Pr -8,854X88.4X 10^/®^^ -0.780
.3600
Nu - 0.023 X 18,i00<" X0.788> < > 88J Nu - 0.33 X12.880»'«X0.7S0>'> - 84.0
K, -51-8 X 0.0l78/76.t;(10-6 - 88.06 5. 84.0 X 0.0878/0.004.8 - 84«i

Over-all U - P*hMot-dagC

A check on the original assumption of 88^ drop on the gas side is next made by equat¬
ing heat transfer to heat release. Log mtd and total heating surface are required.

0^ -287.3-15.9-272; 370.9-111.9-259 ; 0 - - aosM

860
Total surface of 35 76.2 *r>m diameter X 3.66 m tubes isJO^m*
3600wgCp,Al = VAO
M -12.57 X30.66X 265/(3600 X 1.19 X 0.25) = 96.4“C

This is practically equivalent to the initial a.ssiimj>tion: however, a recalctilation is

indicated since the difference found between gas and .'tir film coefficients of heat trans¬
fer does not justify the assumption that was made about the average temperature of
the tube.
RecQJjerative air preheaters are generally enclosed in rectangular sheet steel
casings. The recuperative type may lx* composed of steeltubes,fromS0.8to76inin
in ^ameter,io a parallel bank; or altem.ntely, of rectangular flat plates spaced
from 12.5 to 25 mm apart, leaving alternate air and gas passages. The
regenerative air preheater is sometimes built as a static double-chambered cas¬
ing with tubular elements, and provided with valves to alternate the gas-air
flow. However, the valvclc.ss rotary regenerative heater is more often seen jn the
power plant field. The Ljungstrom heater, a jiroiwictaiy design, is illustrated.
The elements of this type of heater are masses of metal in the form of crimiJed
or cellulai steel sheet built into the foiin of a cylindrical rotor whose shaft is
mounted so that rotation carries its elements, alternately, through air and flue
gas passages. The rotor turns at about 3 rpm, and while a motor is required,
which is a disadvantage compared to recuperative types, its power requirements
are very small. The casing in which this rotor is set is divided by diaphragms
and seals, so that gas may flow continuously through one side of it, and air
through the other.
For continuous operation of an air heater, arrangements must be made to
dean the heating surface, particularly the gas surface, this being done by pro¬
viding openings for the use of brushes and steam lances, or by permanent soot-
474 THE GAS LOOP
blower installations. Hoppers are frequently provided to accumulate fly-ash
and soot at the bottom of the heater. There have been cases of frequent heating
surface replacements because of corrosion. This difficulty is occasioned mainly
by the condensation of vapor and sul¬
fur trioxide to form a corrosive acid
at the cool end of the heating surface
on the gas side. The presence of sul¬
fur trioxide has a tendency to in¬
crease the partial pressure of the
vapor, and so raise the temperature
at which condensation can occur. To
avoid this difficulty minimum outlet
gas temperatures are set and main¬
tained by passing a part of the air
around the heater at light boiler loads
when the outlet gas temperatures are
lowest.
12-9 Draft. Combustion requires
oxygen—and therefore air. To move
this air through the fuel bed and to
produce a flow of the gaseous prod¬
ucts of combustion out of the fur¬
Power
nace, then through the boiler, econo¬
Fig. 12-31 Regenerative air preheater. mizer, etc., requires a difference of
pressure equal to that necessary to
accelerate the gases to their final velocity, plus friction head losses. This dif-
ference of pressure is called draft whether measured above or below a^wos-
pheric pressure. At a pressure less than atmospheric, the draft is called a
vacuum, whereas it is a plenum if above atmospheric pressure. Fig. 12-32
shows typical pressures in the gas loop. A close scrutiny of the manometers,
schematically depicted, will serve to show how the air, once having been given
a big boost by the forced draft fan has the plenum subject to continual attri¬
tion by draft-consuming components of the gas loop until not only does the
plenum disappear, but an increasing vacuum appears along the path of the
gas loop. This is finally reduced to what th,e chimney can overcome by means
of an induced draft fan. Fans and chimneys produce pressure in the positive
direction; friction, turbulence, fuel bed resistance, and the like consume draft
and produce decreasingly positive or increasingly negative pressures.
Measurements of draft serve not only to define the resistances to gas
flow, but will also, in a comparative way, indicate the rate of flow. Thus
the difference in manometer readings between the first pass of the boiler and
the boiler outlet could be employed in a suitable instrument to indicate rate
of gas flow.
The range of pressures, both plenum and vacuum, appearing in the
gas loop are such as to be directly and accurately measurable by water-
loaded manometers. The simple U-tube manometer is an instrument which
balances fluid heads in a glass tube so that readings may be taken by com¬
paring the registry of menisci on a scale mounted alongside the tube.
DRAFT 475
Usually one end of the U-tube is open to the atmosphere, the other being
connected to a region of vacuum or plenum. The vertical component of
menisci displacement is a measure of the vacuum or plenum. A displace¬
ment of ^ cm of a liquid of ti;g/cm* density represents a pressure of fayg/caAA
more sensitive and easily read instrument is made when one leg of the

Fig. 12-32 Typical pressures in the gas loop.

U-tube is inclined at a small angle to the horizontal and the other made
comparatively large in bore. The liquid displacement will then be almost
entirely in the small tube. A considerable meniscus travel will be produced
by a very small pressure difference.
Although the manometer is a simple primary instrument for draft
measurement, routine operating needs are better met by a dry type of draft

ADAPTOR fNCLINEO LEG TYPE TYPE

Fig. 12-33 Apparatus for velocity traverse and static pressure with a single man¬
ometer. If the duct carries a plenum, the leads to the manometer are reversed. When
the static pressure exceeds the capacity of the sensitive manometer, a U-tube manometer
is added, as shown.
476 THE GAS LOOP
gauge, one which will give a pointer reading on a prominent scale and re¬
quire little or no maintenance through the years. However, such gauges are
calibrated to read “cm (inches) of water.”
For some purposes the draft pressure must be used as g/cm* or kg/cm*.
To convert a draft gauge reading, multiply the gauge reading in cm by the
density of water in g/cm* to obtain pressure in g/emK
It frequently happens that gas flow is slowed down and a pressure built
up or vice versa. The equation used to relate the pressure increment to gas
velocity is derived from Bernoulli’s principle, assuming incompressible flow.
7^2 _ 7^2 = X 2gH (12-2)

where Fi and V2 are the gas velocities before and after applying a driving
pressure of if m of gas.
C = a velocity coefficient, generally nearly unity.
Since gauge readings are in cm of water rather than m of gas, a
conversion is necessary.

Fig. 12-34 Draft loss through fuel beds.

Example 1: The draft pressure consumed in giving 176.7*0 air, under 14 cm water
plenum, a velocity increase from 0 to 15.24 m/Bec will be determined.
As Vi is 0, and C is assumed == 1, Eq (12-2) can be rendered: H « V2^/2g « 11.8S.iki
of air. The process of converting the pressure head, represented by 11.8S m of air at
14cm. water plciniin and J76,7’*C, into cm water follows next. Static pressure of the
air «= 1.034 + 1*048 kg/om*
Density of the air - 1.048 x 10* /29.8 x 449.7 .0.M14B kg/m>
Let Vw be the cm of a water column equivalent to 11.83 m of air

IQif^ - 11.88 X 0.80146

Vw "> 0.948 om water

The total draft of a gas loop is the arithmetical sum of all incremental
draft losses in the scries circuit. A classification of these increments of draft
is:
DRAFT 477
Di « Velocity head, V*/2jr. Since velocity and pressure are frequently partially
interchanged along the gas conduit, it is the final velocity of gas exit from
the chimney that should be considered in the total draft requirements.
Dt » Fuel bed resistance. Applies to grates and stokers. It varies with the rate
of combustion and is roughly indicated by Fig. 12-34. Spreader stokers
are not shown, since much of the coal is burned in suspension and does
not load up the grates. Draft resistances of spreader stokers are of the
order of 0.635 om or less.
D* = Draft loss caused by friction and turbulence of gas flow throu^ the
equipment of the gas loop. Equipment manufacturers’ data should be
consulted whenever available.
Boiler and superheater. Depends on mass flow, tube and baffle arrangement.
May be from 0.381 om to 5.08 om water.
Economizer. i.87 om to 8.81 om water.
Air heater. 3.54 om to 5.08 om water.
Dust collector. 0.354 om to 13.7 om. water. Highly efficient dust collection by
mechanical action is paid for by high draft losses; however, the required draft
decrea8(\s rapidly if gas cleaning specifications are eased slightly.

D4 Draft loss due to friction in air ducts, gas breechings, chimney.

This loss may be computed with the following equation:

D4 *= —4r water (12-3)

lO \2gR /
where H « Length of conduit, m
R = Hydraulic radius of cross section, m (Area divided by perimeter.)
V = Gas velocity, m/sec,
d = Gas density, kg/m*
/ — Friction factor as follows:*

Air against steel, 0,005; against masonry and concrete, 0.007.

Flue gas against steel, 0.014; against masonry and concrete, 0.014.

Example 2: The draft loss in a sheet metalhot air duct55.88 om x 56.88 om square
by 91.44 m long will be detenninedwith the use of the foregoing friction equation. Air
velocity,609.6 m/min; temperature, 93.8M; pressure, 10.16cm water plenum. Run of duct
includes 3 right-angle and 2 45-degree turns.
The draft loss is the sum of friction plus an allowance for loss in the turns of the

*For rectangular passages increase these values by 15%. Add 0«L27 cm water for
each 90* bend.
Numerical values of / appear in technical literature from time to time. Many investiga¬
tors have experimented with gas friction* or attempted correlation of experimental data
pertaining to /. Reported values range between the wide limits of 0.004 and 0.016. Steel
and brick or concrete air ducts should have different coefficients because of the difference
in smoothness. In the case of breechings and ehimne3r8 there is not so much dffierence on
account of the tendency of both brick and steel surfaces to soot up to approximately the
same conditions.
 478 THE GAS LOOP
duct. Equivalent right-angle turns = 3-l-2x%==4. At 0.127 tm right-aagie
bend,1088=: 0.127 i-0.508 em water.

Density of air at 08.8*0 and 0.01016 kg/om* plenum is now calculated from the

general gas law PV = wRT.

P » 1.084 -f 0.01016 i. 1.04416 kg/omi

Density = P/RT = 10441.6/29.8 x 866.8 » 0.9728 kg/m*

Hydraulic radius, R = —^ -■ 16.97 om •- 0.1897 m

’ 4 X ».88
0.9728 115% X 0.006 X 91.44
7) -- X-^ Q 500 „ 2.489 om water
10 6 X 9.816 X 0.1897 ^

When all the draft losses of the gas loop are summed up, the total repre¬
sents a fluid pressure the equal of which must be created by “draft pro¬
ducers.” These are generally fans or chimneys, although occasionally steam
and air jets are used for boosting other systems of draft or for localized
draft aid.
Some of the draft nomenclature of the gas loop is covered by the fol¬
lowing definitions.
Natural draft The fluid pressure difference created by confining a column
of heated gas, as by a chimney. See Fig. 12-43.
Forced draft, A plenum on the air before it reaches the combustion zone.
Created by fans.
Induced draft, A pressure rise from a greater to a lesser vacuum, created
in the gas loop between steam generator outlet and chimney by means of a
fan.
Balanced draft, A combination of forced and induced, or forced and
natural draft, so chosen that the gas loop pressure is nearly atmospheric at
the furnace.
Required draft. That draft pressure required to overcome the sum of the
draft losses at any specified rate of flow, i.e., Di 4- D2 + Ds + D4.
Available draft. That draft which the fans and/or chimney of the gas
loop can produce at any specified rate of flow.
Chimney draft decreases somewhat with increasing gas flow but not
greatly. Fan-produced draft varies according to the Fan Characteristic,
which generally shows increasing delivery pressure with decreasing flow. A
system whose available draft was not at least equal to the required draft
at maximum expected rate of flow would be one in which the draft deficiency
limited the output .of the steam generator to less than its capacity. A
properly engineered design would provide some excess available draft at
maximum expected rating. At lesser loads the excess draft available would
always increase and the gas loop pressures are then brought into balance
by (1) adjustable dampers or (2) fan speed control.
The Rate of Flow in the gas loop may be established by:
1. Direct means^ generally a traverse of some strategic cross-section in
GAS CONDUITS 470
the duct or breeching, using pitot-static tube and inclined leg manometer.
Temperature readings are also necessary.
2. Indirect means, employing the principle of mass continuity. Complete
analysis of the fuel and flue gas are necessary, but no gas velocity measure¬
ments are taken. Fuel consumption rate is necessary, however, for the mass
flow comes from a calculation of Gas-Fuel ratio, which is then multiplied by
the rate of fuel flow.
Example 3: At a time when coal was being fired at the rate of ]805.Slkgper hr a
traverse of theOl .4 om x i21.9 om breeching was made by pitot-static tube. Other data,
as follows, were simultaneously recorded. The gas flow will be calculated by both meth¬
ods previously mentioned.
Coal ultimate analysis: C, 79.5; H2,5.1; O2, 5.0; N2,1.6; S, 1.0; H2O, 2.8; Ash, 5.0.
No combustible in refuse. Orsat: CO2, 10.5; O2, 9.5; N2, 80.0.
Flue gas 2d(W. 5.08 em water vacuum. Average of square roots of manometer read¬
ings on pitot-static traverse, 0.767 om i/*
Assume MW of gas, 30; R, 28.8 m/kf
Direct method. Assume pitot-static coefficient = 1. Neglect effect 6.08 om vacuum
on density. From the general gas law, or Fig. 12-44, density = 0.68660 kg/m^
To use Eq 12-2, change manometer reading to feet of gas thus:
10 X 0.767*
8J869 m gas.
^-53866“

Then V = lVi«.«s x 8.369 » 18J1 m/M.

01 4 X 121.0
Gas Flow *= AVd «= /qo x Too ^ ^ 0.68650 x 8600 « 85280 kg per hr

Indirect meihod. Employ Eq 5-17 for dry gas.

X 10.5 + 8 X 9.5 -I- 7 X 80.
Dry flue gas
3 X 10.5 -) (0.795 + f X 0.01)
+ t X 0.01 = 19.05 kg per kgcoal
Water vapor - 9 X 0.051 + 0.028 = 0.48 kg per kgcoal
Total gas flow = (19.05 + 0.48) X 1805.8 = 85260 kg per hr

12-10 Gas Conduits. The conduits connecting the various pieces of

equipment of the gas loop, thereby providing a passage for the gas or air
as it flows from one to the other, are either ducts or breechings, depending
on whether air or products of combustion are conveyed. The term jlue is
sometimes used interchangeably with breeching. The passages may be cir¬
cular or rectangular, made of steel, masonry, or concrete. The rectangular
steel type is most frequently seen in power plant practice.
Air Ducts. The design of a duct is a compromise between first cost and
cost of energy loss from friction. This means a compromise between cross-
sectional area and velocity. Table 12-5 contains recommendations for ve¬
locities in air ducts. Such data are in common use since most ducts are too
short to warrant exact analysis of economic size. Branch ducts should
have velocities not over two-thirds those given in the teble.
Draft losses were discussed in Sec 12-9, and equations and methods set
forth there may be used for duct calculations. A rough approximation is to
480 THE GAS LOOP
Table 12^. RECOMMENDED VELOCITIES IN AIR DUCTS

Static pressure Velocity Static pressure Velocity

em (in) water m/min cm (in) water m/min (4)m)

5.08 (3) 549 (1800) 12.7 (5) 869 (3850)

(H) 810 (3000) 14 (5J) 914 (300(n
7.M (I) 671 (2300) 15.34 (6) 945 (8100)
9M (Si) 782 (3400) 16.51 (61) 991 (8350)
10.1« (4) 777 (3550) 17.8 (7) rOSl (8850)
U.4S (41) 823 (3700)

Based on air atU.l%i760nunVelocities may be increased 0.e5i%(l%) for eacha 190 m

(500 lt)altitudeatid 0.0%(1%) for each5*C (10^) over 31.1*0 (70*F).

allow 0.022 cm loss per 3 mof run where the duct friction is a small part of
the total draft requirements. Air ducts, in general, extend from the outlet
of a forced draft fan to the stoker or burner plenum chamber. Where air
heaters are used the air duct is necessarily in two sections, viz., the cool
section, fan to heater, and the hot section, heater to combustion equipment.
The latter should be insulated against heat loss and may possibly require
special design for thermal expansion. Bends should be long radius if possible.
Where of short radius, or square, internal guide vanes should be provided in
large (60cm -f) ducts. The duct walls should not be lighter than:
0.644 mm metal up to 122 cm wide.
0.812 mmmetal up to 152 cm wide.
1.02 mm metal up to 183 cm wide.
1.29 mm metal up to 213 cm wide.
Breechings, Velocities in breechings are not usually made lefis than 6
nor over 15.2 m/sec atratedflow.If possible, the gas should be made to flow with
approximately the same velocity as it has in the preheater or economizer
The gas passages of modern plants offef so many variations that little
may be said of them in the way of generalization. A breeching, as such, is
often practically nonexistent, while auxiliary heat-absorbing surfaces have
occasioned almost as many variations of gas passage design as there are
plants. A factor of no little importance is the trend to fewer and larger boiler
units which, in some cases, are serviced by individual stacks. Some of the
devices employed today, all of which must be incorporated in the flue gas
circuit between boiler, and stack if they are installed, are:
1. Economizer.
2. Air preheater.
3. Induced draft fan.
4. Dust collector.
For control of air temperatures, for continuous plant operation while equip¬
ment repairs are carried out, and for part load adjustments, these devices
are frequently by-passed, the by-passes themselves adding much to the
expense and difficulty of design of the gas passages.
 ana oiir-neater miet
draft.**
@ Droft measurement ot
this poir^t gives “air- .
heater outlet d^’bndl
“dust-cdlectw inlet drof r
® Draft measurement at
this point gives*tiust'
C^lector <S^et draft*
and serves os^induceoH
droft'fan indicator”
® IVessure msosurwent at
this point gives “air- I
heoter mlw pressure **1
and serves OS ”forced|
droft>fan indicator.*'
® Pressure mcosurement ot
this point gives **air-
hearer outlet pressure”
and**«vindboK pressure”
Poioer
Fig. 12-35 Location of draft measurements.

of its carrymg a few hundred cubic metres of gas per kilogram of coal burned,
the breeching arrangeAient is not entirely flexible. Large boiler breechinp,
particularly, may be difficult to dispose of within the boiler room, and are
frequently installed on the roof. The breeching arrangement is stUl further
dependent upon the method of supporting the stacks. Obviously the breech-
ing would be different were a battery of boilers to be connected to stadcs
under conditions such as:
 482 THE GAS LOOP
1. Stack mounted on boiler room roof and carried by roof structure^ or by a special
steel framework within the boiler room extending from a firm foundation to the base
of the stack.
2. Stack supported by a short solid foundation at ground level. The breeching entry
approximately at ground level permits lower headroom in the boiler house, and the
mounting of auxiliary heat-absorbing surfaces and fans near boiler floor grade.
3. Stack carried on high masonry pedestal allowing breeching to be carried hori¬
zontally into it at the level of gas outlet from the boiler.
Problems to’be met and overcome by a breeching are numerous, but ap¬
parently corrosion is one of the most troublesome. Constructed of sheet iron
or steel plate, the breeching is prey to the corrosive gases it may contain and
to stresses set up by expansion and contraction occasioned by varying
temperature. Hence, expansion joints which permit change of length of the
breeching are needed and may be either slip joints packed with a high-tem-
perature material or flexible creases which also serve as stiffeners. Large
breechings must be stiffened by angle-iron or in some other fashion, since
the 4^8 mm to 7.9 mm steel plate is too thin to be sufficiently rigid alone.
Corrosion will be least where no condensation takes place in the gas
loop; therefore the breeching should be insulated where surface radiation
from it would reduce the temperature of some of the gas below the dew
point. In the absence of mechanical draft, an exposed breeching should be
insulated to allow the chimney to develop the maximum draft of which it
is cai)able, by delivering the gas to it as hot as the gas left the boiler. Also,
especially if mechanical draft is not employed, the draft resistance of the
breeching should be kept to the minimum by eliminating bends, abrupt
changes of area, and interior roughness. Other factors important to the
successful breeching are elimination of gas and air leaks, and the obtaining
of a completed structure of minimum weight consistent with the meeting of
other requirements.
Many elements enter into the determination of breeching size, some of
them being assumptions or allowances which are to permit the structure to
be successful at operating conditions which might be said to be abnormal.
The foremost of these are the excess air allowed for, and the gas velocity
assigned to this operating condition. Given percent excess air, flue gas tem¬
perature, coal analysis, and infiltration allowance, the flue gas volume may
be found by a suitable application of combustion theory.
The proper design of ducts and breechings should include provision for:
(a) evpansion, (b) insulation, (c) support, (d) rigidity. Expansion and in¬
sulation may be computed, following methods and data presented in Chapter
14. The construction needs to be stiffened, and this is generally accomplished
with angles welded or riveted circumferentially around the breeching or
duct at intervals to suit the individual case. Some types of expansion joints
are themselves excellent transverse stiffeners.
Insulation, The preheated air duct and the breeching must be well in¬
sulated to conserve heat and to render the vicinity habitable. Molded block
insulation securely attached to the steel is the usual practice. The material
commonly used has been 85% magnesia, although o^er materials are be¬
ginning to be competitive at present. The magnesia is magnesium carbonate
GAS CONDUITS 483
mixed with binder. Being able to withstand temperatures up to 316*C ' ^th-
out deterioration it is suitable for all air duct work and can be applied
directly to the steel. With a metal spacing material such a V-rib expanded
metal lath between it and a breeching wall, the 85% magnesia is suitable
for the majority of breeching coverings. Occasionally, however, there are
sections between boiler and air heater which require some more highly re¬
fractory material, or which must be double layered with a material such as
diatomaceous silica (which can withstand at least 843'G) next to the
breeching, followed by an outer magnesia layer.

Fig. 12-36 Details of breeching construction.

Insulations are usually applied to the outside of breechings, but occasion¬

ally they line the breeching. This is particularly desirable for outdoor in¬
stallations and may result in a lower total cost compared to outside insulated
and weather-proofed work. However, the exterior insulation is definitely
cheaper on indoor work. Table 12-6 shows the heat loss through single
layer breeching insulation in vertical position. A glance at Table A-11,
Appendix, will show that the vertical surface is an approximate average of
top and bottom horizontal surfaces, so Table 12-6 may be used to represent
the four sides of rectangular breechings. Double layer insulation can be
calculated by methods set forth in Chapter 14.
Table 12-6. HBAT LOSS THROUGH 96% MAOHBSIA INSULATIOI^ HATING
12.7 mm ASBESTOS CEMENT FINISH
[koal per m* per hr. Air amumed 22.2%) (7S*C)]

Insulation Ten^^erature Difibrenee, Hot SorCaee to Air ^ (*F)

X IIIUAXICENS
nun (in) S54S (100) in (200) 166.5 (800) 222 (400) 277.5 (500)

2S.4 (1) 107.6 260.9 860.8 466.5

88.1 (1|) 18S.1 189.6 261.8 889.1
fiO.8 (S) 06.8 148.0 206.6 266.1
68.5 m 80.6 128.6 169.8 219.2
76.S m 67.0 104.2 148b8 186.1
61.8 90.6 124.8 161.9
101.6 (4) 51.5 80.0 110.7 148.9

In all cases it is necessary to employ special means to hold the insulation

on securely and to give it a neat hard finish. The metal spacers should be spot
welded to the duct or attached to‘welded-on studs and the insulation secui^y
wired to it. Also large breechings should have angle clips welded to the top
484 THE GAS LCX)P
and bottom comers and along the center of the bottom, punched bO that
tic wires may be attached. After the block insulation is applied it is wrapped
with wire netting which is raised a bit from the insulation surface with thin
spacers so that the final hard finish of an asbestos cement will be locked
into position.
12-11 Draft by Fans. Mechanical draft fans have an important place
in the engineering of power plants. Without fans, the high rates of heat
transfer now possible could not have been realized nor would the thick fuel
beds of underfeed stokers have been possible. Mechanical draft may be
classified as forced or induced, the former having the combustion air placed
under a plenum, the latter referring to gas movement produced as the result
of a vacuum. The forced draft fan draws in air from the atmosphere and
delivers it through air ducts either directly to the combustion equipment
or to the air inlet of an air preheater. Forced draft alone is undesirable.
Furnace doors or ports may not be opened without an outflow into the
boiler room. Furnace gases escape through all joints and cracks in the set¬
ting. There is more “soaking up” of heat by the furnace walls. However,
some package oil-burning boilers use it successfully.
Conversely a system entirely vacuum in character will have considerable
dilution of the products of combustion by infiltration of boiler room air
through the setting and at any point in the external flue gas passages where
the casings are not airtight. Most hand-fired boilers, also a few stokers, and
gas burning units are of this nature. Induced draft is created by chimneys
and by fans located in the gas passage on the chimney side of the boiler
and its auxiliaries. The logical arrangement is to employ both vacuum and
plenum in such proportions that the furnace pressure is nearly atmospheric,
then the effect of small leaks in the setting is negligible. In a balanced draft
system, the controls are usually set to maintain about 2.5 mm water vacuum
over the fuel bed or in the furnace.
The centrifugal fan is a machine for moving large volumes of air against
a small plenum. It has .a bladed wheel rotating in a stationary scroll casing.
Air admitted around the hub flows into the moving blades and is thrown
outwards into the scroll casing. The drive is most frequently an electric
motor, either direct-connected or belted, although sometimes steam e];igines
or turbines are used.
The energy used by a fan impeller goes into bearing friction, flow work,
compression, velocity head, and heating. It is commonly assumed in fan
work, where the pressures are usually less than 25.4 cm water plenum or
vacuum, that the fluid is incompressible. This assumption, together with
considering heating as a mechanical-type loss, simplifies the work equailon
to

W = vAp H-—- kgm per kg air flow (12-4)

in which W « Work added to the air, kg m per kg.

V » Specific volume of the air or gas, m* per kg
Ap » Static pressure increase, kg/m*.
Vi, Vi « Velocity, m per see*
DRAFT BY PANS 485
From Eq 12-4 we see that the work, and consequently the powerj im¬
parted to the air consists of two components, viz., static pressure and velocity.
An ordinary manometer reading on the inlet and outlet of a fan gives the
static pressures relative to the atmosphere. Their difference is the working
static pressure of the fan. Similarly an efficient pitot tube turned into the
air stream furnishes an indication of the sum of static pressure and velocity
head on the attached manometer. This is called total pressure, or dynamic
pressure.
If the fan sucks air directly from the surroimding atmosphere this total
pressure is the dynamic draft produced. If there is an inlet duct, the before-
and-after pressures must be read and subtracted or taken from a manometer
that is differentially connected.

Let Ap. = Static draft, cm water

Apt » Total draft, cm water
Q » Gas flow, cfm.
Then from Eq 12-4 the following can be derived:
Total air hp = QApt /450 (12-5)

Static air hp = /450 (12-6)

The fan efficiencies are expressed as follows:

Total fan mech eff, fin — ^ (12-7)

iShait np

c,. r cf Static air hp

Static fan eff, tiu - . (12-8)
Shaft hp
In fan practice the term manometric efficiency is employed to describe
the effectiveness of a fan in producing the draft pressure of which its design
is theoretically capable.
Draft actually produced
Manometric eff — (12-9)
Theoretical draft
Some manufacturers base this efficiency on a radial blade standard. In
effect, this is to assume a theoretical draft standard of u^/g. On this basis, the
manometric efficiency (better referred to as manometric ratio) of a good
plate fan is 0.65; of a forwardly curved multivane, 1.12; of a backwardly
curved multivane, 0.385. For high-speed motor or turbine drive, low mano¬
metric ratios are desirable.
Draft fans are designated as plate (paddle wheel), multivane, or propel¬
ler type. The propeller type is seldom used, since it develops but little
static pressure. The plate fan is employed to some otent, especially with
en^e drive, but the multivane centrifugal fan is the common type for
bol^ forced and induced draft. Multivane fans are constructed with blades
which are radial, backvaardly curved, or forwardly curved. The blade curva¬
tures are most imp<»tant in deteimixung tte fan characteristics.
486 THE GAS LOOP
The effect of blade curvature may be studied with the aid of Fig. 12-37,
in which u = peripheral speed of the blades, while V and Vr represent, re¬
spectively, the absolute and relative velocities as the air leaves the blade.
Obviously Vr must increase with increased rate of discharge. In the
vector diagrams the vectors V' represent the higher discharge rate and one
may see that for a given increase in V/ the vector V' is greater than V in
two of the types, abont the same in the other. Thus by study of these dia¬
grams one may understand why the backwardly curved type possesses a
self-limiting demand for power and a limited rate of discharge, whereas
power increases rapidly with discharge for the forwardly curved type. But

Bockwardly curved Forwordiy curvtd Rodiol

Fig. 12-37 Basic blade forms for fan wheels.

since V is largely diffused into plenum in the scroll case, it is apparent that
lower values of u will produce the same plenum in a forwardly curved fan.
Backwardly curved blade wheels are generally selected for forced draft
service because the high speed is suitable for standard motor drive. The
power demand is self-limiting, and the static efficiency is high. These fans
may be satisfactorily operated in parallel.
Induced draft fans operate in gas of much higher temperature and may
handle gases laden with dust. Forwardly curved blade wheels run at the
lowest speed to develop a given pressure, hence are frequently chosen for
induced draft service so that the centrifugal stresses in the wheels will be
least. Low speeds, together with absence of dust-gathering tendency, mini¬
mize out-of-balance vibrations. The forward curvature reduces the blade
depth, but gives a large inlet opening for the gas. The rising horsepower and
pressure characteristics of this type usually render it unsuitable for parallel
operation, for it can overload its dnver under abnormal working conditions.
Induced draft service is exacting in requiring heavy-duty construction and
is frequently met by a modified radial blade.
Basic operating conditions of forced draft service are:

1. The fan handles cool, clean air.

2. The fan location can be wherever convenient. Ducts carry air from fan to plenum
chambers.
3. General use is made of backward curving, high speed multivane fans.
4. Required draft consists of air duct, preheater, and fuel bed resistance.

Induced draft service is not so simple and direct as forced draft. Its basic
operating conditions are:
DRAFT BY FANS 487
1. The fan handles hot gases, often from fOO*—480*0*^
2. The gases often contain soot which fouls the blades, or ashes and cinder which
wear them.
3. Required draft consists of the sum of gas friction loss through furnace, boiler,
superheater, economizer, dust collectors, and air preheater. Chimney may assist.
4. Location is fixed, somewhere between boiler outlet and chimney.
6. The handling of hot gases requires more expensive construction, such as shield¬
ing or water-cooling of bearings, etc.
6. Vacuum costs more to create than plenum so the induced draft is used alone
only with installations where little or no fuel bed resistance is encountered.
Fan manufacturers test their product under standard conditions and
publish the results in tables or as performance charts. The power plant en¬
gineer’s work is then the fitting of his own particular conditions with the aid
of these charts or tables, making the proper corrections where his data differ
from the manufacturer’s test conditions. The purchase of large fans is com-

1 ^Static prwtsura ] □
1
“ Effldlancy=^ —
7J
-

8a
mmKi>POW 1
i ^

LJ
r
0 -1is *210 *45 *60
0*liv*ry~Tbousond m/min Oalivary—Thousond m/min Oalivary-Thousond m/min
Fon with bockwordiy Fon with forwordly Fan with rodlal
curved blading curved btoding blodok

Fig. 12-38 Comparative performance characteristics of centrifugal fans. All rated

666.84 (80.000 elm) at 16.24 om (6") SPhut not all same speed or sise.

monly made upon the basis of performance guarantees. The generalized per¬
formance curves of Fig. 12-38 are for constant speed, since that approximates
the usual condition of use. One is able to predict the performance at other
shaft speeds and gas conditions, making use of such data. From fan theory,
assuming efficiency to remain constant, it is possible to formulate a number
of equations covering the before-and-after-a-change situation. £q 12-10 rep¬
resents all of them.

Let Q B fan capacity, volume of flow.

W “ fan capacity, weight of flow.
P ■* horsepower.
D » draft.
N «■ rotative qpeed.
d » doudty of the air.
488 THE GAS LOOP
Also, let A and B represent any pair of these. Then

t-[fm (12-10)

X Y

The manner in which Eq 12-10 is implemented to represent a specific

case is demonstrated by the following example.
Example 1: A certain centrifugal fan which has received a rating of I450ai*/iiun
at 15M em of water, 1080 rpm« and4.4<0 is to be employed at 7.62 om water with air
heated tai07..2*0 Although operating at the same efficiency point, the rotative speed must
be altered, a different discharge will ensue, and the shaft power will decrease.
Let B in Eq 12-10 be the draft 2), then in succession let A be Q, iV, and P. The ex¬
ponents being selected from the table, these three equations are deviised: *

Qi*Li>2J UJ N, IdJ P2 L2>2J UJ

X ^ = 2*« X 1.37-'« ^ = 2»« X 1.37-»«
Qj Ni Ps
whence Q2 = 1200 mS/min N2 = 890 rpm; P2 = 0.415Pi.
The flow-pressure relation of the system of equipment and ducts in which
a fan is maintaining the flow is called the system resistance characteristic.
This has the form approximately of

D. = KQ2 (12-11)

When the fan is running, the flow Q is that singular value existing at the
intersection of the fan characteristic curve and the system characteristic.
This is made the full rating of the fan if the system characteristic is definitely
known. Otherwise a fan of characteristic shown in Fig. 12-39 as Fan No. 2
(which might be another fan or Fan No. 1 at higher speed) is selected.
Then, were the system characteristic correctly evaluated, the surplus draft
oa' would be dissipated by damper action or speed control. The efficiency of
Fan No. 1 would be better than Fan No. 2 for a position at point a; hence
the allowance aa" if used should be made as small as possible by careful in¬
vestigation of the gas loop resistances.
The effect of variable load on this component of the power plant is to

♦The necessary D and d ratios are: ^ » 2; $

Di 76.2 oi S72
DRAFT BY FANS ^
require operation along the D. characteristic. At some fractional load, for
example, the fan should operate at point b, producing Qb and Db- There are
three methods in common use for attaining this new state with Fan No. 1.
Inlet vane control. This is the most used method in the forced draft
field. The inlet vanes are located on the inlet to the fan and by adjustment
can change the direction of air entering the wheel. Over limited ranges of
control, say from 60% to 100% rated volume, there is not much throttling;
rather, the inlet vanes reduce the volume of air handled by reducing the
speed of the fan wheel relative to the entering air. At still smaller flows the
nearly closed vanes have considerable throttling action and the performance
is not much different from plain damper control.

Fig. 12-39 Variable load performance of a centrifugal fan.

Speed control. Fans may be driven by turbines, variable-speed motors,

or constant-speed motors with variable-speed couplings—^hydraulic, mag¬
netic, or mechanical. This control is, of course, far more expensive in firet
cost than other forms, but the resulting performance as depicted in Fig.
12-39 is quite favorable if the fan is operated much at fractional loads be¬
low 60% of rated capacity. This type of control provides more advantage
on induced draft service, where inlet vane control is not very expedient.
Theoretically, the action of speed variation is to move along the D«
curve at constant efficiency. From Eq 12-10, by considering A = D, B = Q
and di = d^t

D - D.(Q/Q.)« = Constant X Q*

This is obviously the equivalent of Eq 12-11. However, it must be re¬

membered that system resistance is only approximately parabolic, and small
corrections will still be required by the variations.
A compromise solution is damper control with two-speed motor drive.
The two-speed motor is much cheaper than variable-sp^ control and re¬
duces the damper losses at low flows. Another variation is the double-motor
drive—one freewheeling or being disengaged at part load.
490 THE GAS LOOP
Example 2; The characteristic fan curves shown in Fig. 12-40 are the draft and
efficiency curves for a cold air fan rated at 1416 ms/min at 18.7 om static pressure.
The operating conditions under which this fan would produce 860 m^/minat 10,16 cm
static pressure will be investigated. A constant efficiency parabola is passed through
the proposed new operating point and carried to the orinnal operating curve in order
tr determine what, the efficiency will be. The constant emciency curve is Eq 12-10 set
up as follows:
A/A = (A/Qi)*
Since one point on the plot of this relation is known, i.e., iu.l6 cm at 860 m*, the
constant efficiency line may be drawn, extending to an intersection with the original
fan characteristic. This is point ar, for which the corresponding original efficiency is
read to be 72%. All points along the plotted line, including the new operating point,
have this efficiency. By writing Eq 12-10 with N and Z), and noting that, at point
X, iV = Ni, D = 14.48 om^ the necessary new speed is found.

Damper control. This is the simplest method and least expensive in first
cost; also, the least efficient on account of the irreversible, entropy-increasing,
action of throttling flow. Nevertheless it is a common method of control of
induced draft in power plants. The damper is usually located in the system
on the boiler side of the fan. The fan outlet is
an unrestricted discharge to the chimney. Since
damper control merely imposes a controllable
pressure drop in the system, the fan can produce
point b (Fig. 12-39) conditions by operating at
point c with the damper increment be added to
the system resistance.
Construction. Centrifugal draft fans consist
primarily of a Rotating bladed wheel enclosed in
a spiral-shaped sheet-metal housing. The wheel
may be single-entry, meaning that air enters the
Fig. 12-40 Change of oper¬ interior of the wheel from one side only, or
ating conditions. double-entry. The latter form is employed for
large volume fans where the necessary wheel
width is so great that it is expedient to feed in the gas or air from each end
of the wheel. The shaft on which the wheel is mounted turns in bearings which
are supported on brackets fastened to the housing or on pedestals independ¬
ently mounted on the foundation. Bracket support is the cheaper and is suit¬
able for close-coupled motor drives. The pedestals are preferable for induced
draft fans and for forced draft fans driven by turbines or speed controllers.
The drive may be V-belt, but direct drive is usual, there being interposed
between driver and fan a flexible coupling.
The National Association of Fan Manufacturers had adopted four standard
fan classifications in order to meet the varied operating conditions of the field
with fewer standard designs. The classification is based on maximum total pres¬
sure, i.e., static and velocity head. The total is practically equivalent to static
pressure for outlet velocities less than610 m/minr^but at high duct speeds the
static pressure is more limited.
Bearings of fans should be sealed ball bearings where possible. Air-
DRAFT BY FANS ^ 491
Table 12-7. TRADE PRACTICE IN OPERATING LIMITS
OF CENTRIFUGAL FANS*
Class I—115.86 mm water maximum total pressure
Class II— 171.46 mm water maximum total pressure
Class III— 847.66 mm water maximum total pressure
Class IV—Greater than 847.66 mm total pressure
^At 81.1*0 and 760 mm Hg. Correct required total pressure before selecting fan class
by multiplying by ratio of density (at 81.1*0, 760 mm) to actual working density.

cooling disks or water-cooling jackets are used to protect the bearings of

induced draft fans.
Installation, In determining the required capacity of draft fans, it is
well to allow for a quantity of air considerably in excess of that normally
anticipated so that boiler capacity need not be impaired during those periods
when automatic combustion control equipment is in maladjustment. The
excess capacity provided is necessarily a judgment factor. Sometimes the
gas loop is designed to carry 100% excess air abnormally. Meeting this
requirement naturally penalizes the normal operating efficiencies. Conse¬
quently, designers often restrict the overload capacities of the mechanical
draft equipment. However, forced draft fans should have flow capacity at
least 10% and induced draft 20% in excess of normal conditions. Illustrating
the effect of abnormal gas flows- on the fan, note in Fig. 12-41 the normal
operating condition for a certain induced
draft fan is b, at which time the system
is at point a and ab is a damper-resist¬
ance. The power required is P. Now con¬
sider the effect of an abnormal condition
introducing Qx excess flow for which the
new system resistance is a' and damper
effect a'b'. Obviously the damper is nearly
wide open and fan pressure has been re¬
duced to 6'. But power reqi^rements have
increased to considerably above the
rated power. The driver must be able to Fig. 12-41 Fan capacity allowances.
meet this demand. Furthermore if the
system had by-passes, as around air heaters, the system resistance might
change to curve and in the absence of limit controls the system flow
could increase to point c with damaging overload power Px demanded of
the fan drive. These possibilities must be investigated by the plant designer.
The location of induced draft fans is somewhat restricted by the needs
of the remainder of the flue gas circuit. They are frequently required to be
mounted high in the boiler room, on steel framework supports. The fans
are relatively bulky and slow speed. Their noise is not particularly objection¬
able.
Forced draft fans are inherently of higher speed, usually with open inlets.
Hence their noise level is quite high, and the plant designer shoidd isolate
them if at all possible. The author has seen many plants where the forced
draft fan was placed close beside the boiler, probably with the idea of re-
ducinff the forced draft duct work to a minimum. The price paid for this
492 THE GAS LOOP
minor advantage is always: (1) noise levels in the boiler room too high for
ordinary conversation; (2) drafty condition in the boiler room—^nice for
summer heat, but objectionable in cold weather. It is suggested that an
isolated fan room ought to be provided with possibilities of drawing air
either from outside or inside the plant, at the option of the operators. Forced
draft duct work generally comes in under the boiler floor in stoker plants,
being terminated under the grates, in the ash hopper or special plenum
chamber, depending on the type of stoker. The burner box is the terminus
of the air duct on pulverized coal and oil burning equipment, and such duct

B. F. Stwrtevant Co.

Fig. 12A2 An installation of two motor-dnven induced-draft fans 6ftl3.9

(2SO.OOO ofm) at li' (27 9 om) BP* 219.1*0 (42inr)

work is usually overhead and exposed. If an air heater is used, the fan
location is dictated chiefly by gas flow, not air flow, consideration. Hence
this always affects the air duct location and frequently the forced draft
fan itself.
Fan testing. Fan tests are seldom performed by the plant engineer unless
as acceptance tests. Then pitot-static equipment is generally employed, and
there should be some provision for test openings. Experience indicates that
the velocities in ducts and breechings vary across the flow, so that the area
should be divided into elements. Rectangular sections are divided into small
squares and readings taken at the center of each. Circular ducts are divided
into equal-area rings, and test stations are set at opposite mid-radii of each
CHIMNEYS 493
ring. It is the square roots of vclocity^head readings that are averaged for
the calculation of flow velocity, not the readings themselves.
On tests where it is not possible to hold the rpm constant, corrections to
readings must be made before comparing results with constant-speed guaran¬
tees. If N is the rated speed and Nx the test speed, multiply test volume by
N/Nx, test pressure by (N/Nx)^, and test powers by (N/Nx)^ before com¬
paring with the ratings.
12-12 Chimneys. Whether to produce draft by fan or chimney may be
an open question, necessitating a complete comparison of the characteristics
of each before an answer can be given. Briefly, their comparative character¬
istics are:
1. Chimney may have to be high enough to produce sufficient draft for
reasons foreign to draft.
2. Higher rates of combustion arc possible with mechanical draft.
3. Mechanical draft is more readily controlled to meet varying load
conditions.
4. Mechanical draft is independent of atmospheric conditions.
5. Chimneys, once erected, cost nothing for operation, and have a very
low depreciation rate.
The chimney is provided primarily to produce a certain available draft
at the flue entry. But, in addition, there must be sufficient draft to overcome
frictional losses in the chimney itself. These losses are proportional to the
square of the velocity which, in turn, is proportional to cross-sectional area
of the chimney for any given flow of gas. Hence, the problem of chimney
diameter is more than the mere assumption of a velocity comparable to that
used in actual practice; it should be such that the diameter and height it
indicates result in a chimney of the least cost. Deinlein has pointed out
that a combination whose diameter multiplied by its height was the least
of all workable combinations would be the most economical. Ordinarily, the
velocities range between 6.land 15.24 m/iec,but, foi economical design, height
cannot be separated from velocity and diameter.
Most chimney equations are based on rational derivations, but in at¬
tempts to simplify them to where direct substitution is possible, various fac¬
tors have had to be introduced to care for the many variables that enter into
a problem of this sort and, as a result, these equations have the aspect of
rule-of-thumb proportioning. The author believes it better to work from
the fundamental data through the various logical steps, thus illustrating
basic features of the solution.
A chimney produces draft by virtue of an extremely simple principle of
thermodynamics. When the gas is heated, it expands in volume and decreases
in density, in which condition it may be displaced by a more dense gas.
Fig. 12-43 shows how this principle is incorporated in a chimney, (a) Air
movement is produced by heating one leg of a U-tube. (b) Comparative
weights of equal columns of air (2S.9'C)and flue gas (260X1), the difference
producing the draft, (c) The U-tube is replaced by two stacks connected
at their bases. The heat of combustion produces a flow of air down the cool
stack and gas up the heated stack, (d) Progressing toward the conventicmal
form of power house chimney, the air stack is brought around the chimney,
494 THE GAS LOOP
producing the same flow conditions as in (c). (e) If the diameter of the air
stack is made exceedingly large, we have the chimney standing alone in the
atmosphere, but performing as in (c) because of the flow action produced by
the heat not absorbed in the boiler.
The light hot flue gas is confined by the chimney column and the draft
produced is proportional to (1) the height and (2) the difference in density
of air and gas. Therefore the draft of a chimney is, in an elementary way,
expressed by
D= Height X difference in density of flue gas and air

j t
1 I
sQl
I
<

y I ■ w. J
I'.i
'M'..
p Air

Fig. 12-43 Evolution of chimney action.

When this equation is written for Height = SD m ^ incorporating certain

factors needed to convert draft to inches of water, and allowing for cooling
and friction in the chimney, it has the form:

D„ • Jr(J, ~ i.) - 0 007578 4, (12-12)

where D = Available draft per 30 m of chimney, cm water.

K = 2.7 for brick chimneys, and 2.4 for steel stacks,
d* = Density of air, kf fmP
dg ~ Density of flue gas,
V = Gas velocity in. the chimney, m per sec
Qg = Gas flow, mP/wo
Eq 12-12 is based on / = 0.014. To use any other friction coefficient /',

= K(d, - d.) - 0M\Zfd, JB (12-13)

^Qt
The actual height of a chimney is obtained by dividing required chimney
draft by and multiplying by 30 The required chimney draft is the sum
of all friction losses external to it, plus the velocity head loss of gas dis-
CHIMNEYS 405
charging from the chimney, less the effective draft furnished by fans or Jets.
Velocity head is given by
Di = 0.004 YHg cm water (12-14)
The sea-level air density can be calculated by -gas laws or read from
Fig. 12-44. It having been found that the density of the products of com¬
bustion of ordinary grades of coal is but little affected by the coal or by the
excess air used, the gas density chart was prepared on the assumption of a
gas molecular weight of 30. This assumption is sufficiently accurate for most
chimney calculations, but should be replaced by actual combustion calcula¬
tions if (1) the gas is unusually cool, thus reducing the differential density
(da — dg), or (2) if the fuel is lignite, natural gas, or other fuel with a vari¬
able composition considerably different from bituminous coal. Since petro-

Air 1«mperoture« C
23^2

Fig. 12-44 Variation of air and flue gas densities with temperatures.

leum derivatives have a fairly constant chemical analysis it was also pos¬
sible to supply a curve for fuel oil fires. The charted densities are for
sea-level locations. Density corrections to both d* and dg must be made for
chimneys located at altitudes. Up to 1829 m above sea level, it can be as¬
sumed that the barometric pressure decrement is 25 mw Hg per 300 m*
The draft of a chimney is affected by atmospheric temperature, and its
diameter by the excess air assumption. In boiler plants used mainly for
heating, the maximum loads ^(hence maximum required draft) come when
atmospheric temperatures are low, so that atmospheric temperature effects
are in the favorable direction. In order to be conservative, chimney heights
are commonly designed to give the maximum required draft with l5.6Ti
ambient temperature. In southern climates and for cases where maximum
loads can occur in hot weather, and where gas temperatures are low (say
 496 THE GAS LOOP
232^> or less) the design assumption of atmospheric temperature should be
carefully weighed.
The quantity of gas flow, Q„ is directly affected by the excess air as*
sumption. Depending on the combustion equip*
«S2^ ment, and the size and type of plant, the oper*
ISO'
%Sfalic
ating range of excess air is from 0 to 100^ (or
: I2S -Hatpr««»ur»
N|- more). Central station pulverized coal combus¬
s New eparating jlOriginai |

E
I point" Irating
Solnt - ,ao £ tion represents one extreme, using from 5 to
N 5 20% excess air, while hand-fired grates, or small
U-
stokers in poor adjustment, represent the other.
•ffieloncy curvo
Uo« Central station designs rarely are critical on
T excess air assumptions used for chimneys, since
gas movement is secured mainly by fans. In
Oalivary •Theuftand m^/inin cases where insufficient chimney would adversely
affect the remainder of the gas loop equipment,
Fiq. 12-45 Determination of it is well to allow 100% excess air in chim¬
economic chimney propor¬ ney proportioning, even though normal opera¬
tions. tions are expected to be carried out with much
less.
Example 1: The combinations of diameter and height of a chimney required for
a specific case will be determined over a range that includes the minimum product.
Assume that the chimney is to be constructed of bnck, proportioned to carry 48.42
of gas and maintain a static draft of t.007 om water. Qas density given as 0.6776
kg/m’; atmospheric density, 1.171 kg/m*.
Let V be the velocity of gas in the chimney. Then by Eq 12-14 the required draft is
2.007 - 0.004 X 0.0776V* — 2.007 + 0.00271 F* cm water.
0.007678 X 0-6776 F*/*
Eq (12-12): . 2.7(1.171 - 0.6776) — ---om per 80 m
48.421/*
Dfo - 1.882 ~ 0.00078793 F*^ cm per 80 m.
2.007 + 0.00271 F^
Chimney height X 80m (1)
1.882 — 0.00078708 F*/*
From the above it is seen that the higher the gas velocity, the higher the chimney.
Next a relation between diameter and velocity is developed.

Inside chunney diameter

Where Q* « 48.42 m*/90€> chimney diameter « 7.868 F-i'^m (2)

Eqs (1) and (2) give height and diameter as functions of chimney gas velocity. By
assuming various velocities, the corresponding heights and diameters are obtained.
For example, if V = 6J m/sae
2.007 -f. 0.00871 X 87.21
Height - X 80 «■ 60.08 m.
1.882 .. 0.0007S708 X 82.004
Diameter 7.86^47 8.184 m
Height and Diameter » 60.08 x 6.184 «■ 160.8
Other values of V were assumed and the resulting products plotted as shown in
Kg. 12-46.
CHIMNEYS 497
The miniiniim product is found at 7.MS m/Mo gas veloetty. Subsiiiutiiig V«74tt6
in Eqs (1) and (2).
Height = 64.tS m ; Diameter = 2.793 m
The height is '^above-the-grates”; hence, the actual structural height of the
chimney might be different. The diameter is usually taken as that inside the top
of the chimney. Lower sections may increase in diameter due to need for tajier, but
since the effect is to reduce gas velocity and friction, neglect of inside taper is on the
side of adequate draft.

Use of the minimum product as an index to economic porportions of a

masonry chimney has this basis. The wall volume of a hollow cylinder is
proportional to the product of length and diameter. Masonry costs are pro*
portional to masonry volume. Although a chimney is not a straight cylinder,
the thinner smaller sections cost more per unit volume in place than the
thicker because they are toward the top of the chimney where the unit cost
of construction is more than for sections closer to ground level.
Chimneys which are to serve small boilers that may, nevertheless, re¬
quire about the same draft as larger ones are likely to be too slender struc¬
turally when proportioned by Deinlein’s rule. In such cases the
minimum requirements of structural stability govern the diam¬
eter, and the gas velocity is taken to be less than the economic
value. Analysis of the masonry chimney as a structural column
subject to side load from wind was given in Chapter 4.
Tall chimneys are usually built of special, radial-shaped
bricks. To a smaller extent, reinforced concrete is also employed
in chimney building. .
Steel chimneye, ordinarily called stacks, are designed both as
self-supporting and as guyed structures. The self-supporting type
is found most often where, on account of the use of induced draft
fans, the chimney serves only as an “exhaust pipe” and can be
comparatively short. Such stacks are usually fully lined with a
refractory lining. They are cylindrical in shape except for those
in which an efficient diffusion of velocity into pressure is sought Fig. 12-46.
Evase stack.
for in the stack itself. These shapes, called evase stacks, are illus¬
trated in Fig. 12-46. The guyed steel stack is used where minimum chimney
cost is sought, especially where gas flow volume is small. These can be built of
thinner plates than the self-supporting and are generally left unlined. Both
reinforc^ concrete and steel chimneys can have tension to windward under
side load; hence the theory of partial cracking at extreme loadings as de¬
veloped in Chapter 4 does not apply; rather the customary structural analy¬
ses of loaded beams may be employed.
Although chimneys have been built of common rectangular bricks, per¬
forated radial blocks are now in general use for this type of masonry work.
The dead air spaces act as heat installation, the blocks are rapidly laid,
and weigh less per cubic msM in the wall than common brick. These blocks
are semi-refractory and are hard-burned. If properly made, they will ring
when held separately and struck.
Typical data on radial block construction are summarised as follows:
498 THE GAS LOOP
Material, Semi-refractory clay, hard-burned.
Common colors. Buff and red.
Water absorption. 5-10% weight increase on 24-hr immersion.
Net area. Perforations about 35% of gross area.
Density of masonry. 1032 kg/mS
Minimum crashing strength, based on net area, 281.2 kg/cm*
Maximum working compression, based on gross area, 19.69 kg/cm*
Mortar. 1:2:5 by volume of cement, hydrated lime, and sand.
The lower parts of chimneys should be protected by a lining supported so
as to be free to expand and contract separately from the structural column.
Unless some unusually high temperatures are anticipated, the lining of
radial brick chimneys may also be radial brick. Otherwise, say above
538^ C f the lining should be fully refractory and extend from 9.2 to 15.2 m
above the breeebmg entrance.
Chimney accessories. While the chimney is primarily a cylindrical struc¬
ture it must have some extras in order to function properly and last in¬
definitely. These are covered by the following listing.

Stchon Ihtouyhflut

American Chimney Carp.

Fig. 12-47 Masonry chimney.

PROBLEMS 607
20. Find the coefficient of film conductance for gas flowing through a 60.8 Air
heater tube having 4.76 nun wall thickness. Terminal temperatures, 309*0» 960*C. Gas
velocity, 10.7 m/sac. Average tube temperature, 282.2*0.
21. Test data on a 8047 m* plate type heater are: Air flow, 166 716 kg per hr.
Air: in, 87.8*; out, 226.1"; gas, in,836.7*; out, 182.2*, all "O. Calculate the over-all
coefficient of conductance.
22. A plate type recuperative air heater is to be arranged with sufficient surface
to cool gas from 882.2* to 287.8*0 at the same time heating air from 10* 0. Air: gas, by
weight = 0.95. Both gas and air passages are rectangular;2'6 cm x 76 om.Mass flow
of gas, 29,290 kg per hr-m* Predict the over-all conductance from the thermodynamic
parameters.
23. Assuming that the results of Prob. 19 are 146 plates and 2.76 m height, find the
gas and air outlet temperatures when, at part load, G =19630 kg per hr-m*, t[ =826.7*0.
Same A-F ratio and same jicrcentage of air heated.
24. What pressure, cm water, will give air at 93.3*0 and 10.2 om static pressure a
velocity increase from 0 to 731.6 m/min?
25. Find the draft loss in rectangular breeching 2.59 mx3.66 m x 30.48 mlong when
196 m8/aec flue gas of 0.769 kg/m® density are flowing.
26. What is the friction loss, om of water,in a 2.44 no difiuneter chimney 61 m high
up which gas is flowing at 9.14m/8eo? Gas density,0.6776 kg/m®.
27. Flue gas (7^ = 28.3) enters a tapering breeching at 16.24 m/aec, 6-0 om water
vacuum304.4*C.Breechmg is short and well insulated. It tapers from 1.72 m* at inlet
to 8.02 m* at exit. Estimate the exit vacuum, assuming incompressible flow.
28. A concrete air tunnel which carries atmospheric air is composed of two
sections, one 61 om x 76.2 cm x 16.76 m long, followed by one 61 om x ^.7 om x 9.14 m
long. Flow through the large section is 311.4 m®/min4ihen the flow divides andi66.7 m^/
min continue on through the small section. Find the draft loss over the whole tunnel
length. There are one and two 45* turns in the tunnel.
29. A concrete air tunnel, plan as sho’\yn, is
to carry 233.2 m®/minair from A to B. Find om
water air plenum at B. AC-12.19roof 61 omx76.2 om
tunnel. CB =9.14 m of 61 om x 45.7 om tunnel.
Fic 12-29P
30. The products of combustion of 636 kg
Pocahontas coal per hr pass through a breeching
at 260*0 and 7.62 m/seo. Excess air, 50%. Assume square section breeching and find its
dimensions. Sea level.
31. p]stimate tho required cross-sectional area of the air ducts serving this boiler
plant. 1, 2, and 3 are water tube boilers using
A.-C
West Virginia bituminous coal. = 31400
J/g*. Excess air, simultaneously: 1, 70%; 2, -L». 1 rr^
269 4 n?^ 232 3 m*
70%; 3, 100%. 1 200*/. 2 200 V. 3
roiinfi r alin9
32. Estimate the required cross-sectional
75 V. eff 70 V. •M
area of the flue gas passages of the plant men¬
tioned in Prob. 31. Gas velocity, 9.14 m/aeo.,
temp., 287.8*0.
33. An air duct conveys 92990 kg air per hr between an air -heater and a
pulverized coal burner. Ah- temp., 107.2*0, delivering to burner at 6.08 om water. Length
27.13 m with two 90* bends. 83.1 mm magnesia heat insulation, (a) Find dimensions of duct,
if square, (b) Find total air head at heater, (c) Estimate heat loss per hour.
34. The temperature of air in a76.2 omx 150,8 om air duct is l08.3*GPitot-sta*tic explora¬
tions give 8.76 urn pitot head, 7.87 omst a tic head. Barometer, 789,1 mm Hg. What is the
weight of air flow, kg per min ?
35. Determine the m®/miii free air (21.1*0,760 mm Hg) required of a Fp fan which is

^Multiply by 0.239 to obtain in koal/kg

 508 THE GAS LOOP
to supply a stoker with enough air to be 50% excess when burning 680 kg bituminous
coal per hour.
36. Consider Fig. 12-39 to represent a certain ID fan performance on 21.1* air
at 1140 rpm. The rating point 7.62 omdraft, 609*7 m3/miii,With change of speed, could
this fan he used to produce «as4.8 m’/min at287.8*0.7.62 omdraft. What damper action,
if any, is allowable?
37. An FD fan when driven by a constant-speed motor produces a pressure of
20.B2omwater when Q=:i4i6.amS/miiiWhatpressurewould it produce on same air if Q
were made 2266.4 tr^/min 7 What would be the relative speed and power?
38. A boiler plant is expected to bum 749.3 kg of 81760 J/g* coal per hr with
44% excess air. The gas loop is designated in
Fig. 12-38P. The boiler draft loss ial. 14 cm at this
rate of combustion, and the fuel bed resistance
is 8.26 om water. Find the draft available at the
base of the chimney. Find the pressure drop re¬
quired Cl the boiler outlet damper in order to
Fig. 12-38P have 0.25 cm furnace pressure.
39. Find the fan drive horsepower to be recommended for the installation of
Fig. 12-38P assuming that the total efficiency of the fan is 70%, at rated capacity.
AO. A cold air fah which has been operating against a static draft 7.62 omat 900rpm
isshiftedto 6 newd\Av wherel5.24 om plenum is desired. Find the new speed, power, and
air delivery, for constant efficiency and density.
41. Find the power required and delivery of a fan at 1200 rpm when its 1080 rpm
characteristics are 1501 m»/mm,26.4 cm static pressure, 111.5 hp.Temperature constant at
280*0 . Efficiency constant.
42. A certain FD fan delivers to a system whose K (Eq 12-11) is 1,488xlO-^t.
A static pressure of 8.89 om water is wanted at the outlet of this system at a rated
flow of 1188 m*/min.If this fan’s characteristics at 1126 rpm are as shown in Fig. 12-40,
and an allowance ofo.64 cm damper Ap is made, what are the required speed and drive
power for the fan ?
43. Calculate the data for and plot a curve covering optimum combination of
Dia. X Ht. vs Dia. of a masonry chinmey which is to produce an available draft of
2.41 om water-{-velocity head. Gas temperature, 315.6*0 ; atmospheric temperature, ♦,4*C.
Fuel is 2948 kg per hr of bituminous coal burned with 100% excess air. Qh == 30240
J/g* Location, 467 m above sea l^vel,
44. The products of combustion of a designated coal from Table 6-4 are discharged
through a guyed steel stack 91.4 cm inside dia. X 22.9 m high (above breeching en¬
trance). Gas temperature in breeching 816.6*0, atm 16.6*0, sea level. Rate of firing
816.5 kg per hr. Find the available draft when .(a) 60% and (b) 100% excess air is
used.
45. Plot the height-diameter combinations against chimney gas velocity for a
masonry chimney serving a boiler in which the fuel is natural gas (Monroe Field,
Table 5-4). Combustion equipment can burn 1246 per hr. Natural draft.
Excess air, 16%. Lowest flue gas temperature on day of maximum atmospheric
temperature ( 36*0 ) is 218.3*. Sea level. Gas measured at 762 mmHg and 15.6*0. Draft
losses: Burner,0.608 om; boiler,o.889 om;breechmg,0.254 omHsO.The calculated products
of combustion of a mol pf this gas are CO2,0.9786; HjO, 1.861; O2,0.286; N2,8.394 mols.
46. Two 185.8 m2 boilers, each with 2.18 mx 1.68 m grate area, will have a natural
draft stoker, using Indiana bituminous coal. Maximum rating, 150%. Excess air,
100%. Expected boiler efficiency, 75%; flue gas temperature, 279.4*0; boiler loss,
0.711 om;breeching loss,0.669 om. fmd the most economical proportions by Deinlein’s
Rule. Do these seem to 1^ reasonable for structural stability? Atm, 16.7*0.

^Multiply by 0.289 to obtain value in keal/kg

tFor .D^ in om.
PROBLEMS 509
47. The products of combustion of 685 kg per hr Pocahontas coal are to pass
through a breeching at 260*C and 7.6 m/aeof then through a masonry chinmey|6.6mhigh
and 1.58 m inside diameter at the top. Sk level location. 50% excess air. 81.1*0 abnos-
phere. (a) Find the required dimensions of a square breeching, (b) Find the available
chimney draft.
 CHAPTER 13

THE FEEDWATER LOOP

13-1 Function of the Water Loop. The water loop might be said
to have as its purpose the conversion of condensate into hot boiler feed at
a pressure sufficiently above that of the boiler to cause the correct flow into
the boiler, that is, correct from the standpoint of feedwater demand under
variable load operation. The elements of this loop in the portion extending
between the condenser and the boiler are shown in Fig. 13-1. As is seen there,
a pump must first remove the condensate from the condenser. Make-up is
then added to counteract loss of the working medium throughout the cycle
from blowdown, leaks, etc, Tlie condensate is heated, put under pressure
by a boiler feed pump, then further heated. Its condition then is one having
a temperature approaching, and a pressure exceeding, that of the boiler
water. It is finally regulated for the desired flow into the boiler.
The remainder of the w^ater loop consists of the flow at diminishing pres¬
sure and enthalpy, but in the form of vapor, from boiler through turbine to
condenser. This part of the water loop has been covered in Chapters 10 and
11, and our present concern is with that portion diagrammed in Fig. 13-1.
There the equipment is generalized. Details of actual systems vary greatly,
and often involve a complexity not suggested by the diagram. The technical
problems of design and operation of this phase of steam plant engineering
are conveniently grouped under:
1. Water Treatment. Water quality, and operating difficulties originating
in inferior water. Remedies and preventative treatment.
2. Water Heating. The reason for heating has been alluded to previously
in this volume, especially in Chapter 8.* Here we deal with details of heaters.
3. Water Pressurizing and Regulation. Pumps and feedwater regulators.
At high boiler pressures the feedwater treatment should be adequate to
continue the surfaces of the boiler in approximately the same condition as
when new. Once in the boiler, the water is first heated to saturation tem¬
perature, then evaporated at the point of contact with heated tube surface.
In general, the steam is free of all impurities the water might have contained
(except dissolved gas). Impurities are left in the boiler water whose concen¬
tration thereby increases. The point of evaporation being the tube surface,
there is every opportunity for the impurities to deposit on these surfaces as
a scale. When untreated feedvvater produces enough scale on the boiler sur-
• 510
WATER CONTAMINATION—ITS EFFECT 611
faces to interfere with heat transfer, or when/it contains elements, which
either corrode or alter the strength of the boiler metal, feedwater treatment
is necessary.
The higher the rate of lieat transfer, the more important it becomes to
keep that surface scale-free, because the scale can both reduce the steaming
capacity and cause overheating of the tubes. As steam generation goes to
higlu*r pressures and rates of evaporation, the feedwater problem becomes
mor(‘ important and more intricate, even to the extent of requiring specialized
j)hysieal and chemical talent for its solution.

Though it is primarily installed to promote higher thermal efficiency, the

heating of boiler feedwater introduces threefold benefits into plant operation.
1. Heat which would otherwise be wasted, if diverted into feedwater,
represents fuel savings. Less obvious, but nonetheless tangible, is the im¬
provement in efficiency by regeneratively heating the water loop.
2. Thennal stress which can be induced in boiler surfaces by impinge¬
ment of cold water streams is minimized or eliminated.
3. Heating feedwater increases the steaming capacity per square metre of
installed boiler capacity. This becomes of increasing importance at high
pressures, for then hi is a substantial or even major portion of ftg.*
One sees that feedwater heating increases thermal efficiency and boiler
capacity. Balancing the cost of the water loop equipment against these
savings is a factor employed by designing engineers to decide on the extent
of heating the circumstances warrant. However, some heating is almost al¬
ways justified.
13-2 Waiter Contamination—Its Effect. Natural waters usually con¬
tain dissolved salts and gases, and some organic and inorganic material in
suspension. They rarely are neutral in reaction. The dissolved salts are
*At 9S.5 kg/cmi ab» ht is 6U%; hu, 486%.
512 THE FEEDWATER LOOP
chiefly the carbonates, sulfates, and chlorides of calcium, sodium, and mag¬
nesium, and occasionally some iron, aluminum, or silica salts. Oxygen and
carbon dioxide are the gases. The suspended matter is usually alumina and
silica in the form of mud and silt or, if organic, sewage and industrial wastes.
When the impurities find their way into the water of a boiler, they re¬
main there until (1) they are neutralized by some intentionally produced
internal chemical reaction; (2) they are removed by boiler water blowdown;
or (3) they produce operating difficulties which, if continued unabated, may
finally lead to disuse of the boiler or to explosion. The troubles caused by
the feeding of water of undesirable quality arc scaling, corrosion, foaming,
()riming, and embrittlement.

Scaling. Primarily scaling results from the decrease of the solubility of

some salts with increase of temperature. One chemical mechanism used to
explain scaling is a reaching of chemical saturation by the water in the
boiler, then a beginning of precipitation with sedimentation forming a layer
of scale on the heating surfaces and loose i)recipitate in drums. But more
likely, heating surface scale is produced by ciystallization of scale-forming
salts from a locally supersaturated layer of water lying on the heating
surface. This forms an incrustation at the point of evaporation.
Scale is due mainly to salts ot calcium and magnesium and, to a lesser extent,
to silicates. Calcium is the principal offender, particularly calcium sulfate. Cal¬
cium and magnesium bicarbonates are broken down by moderate heating (ICKf
C) into the relatively insoluble monocarbonate and CO2. The reactions are:
l T
Ca(HCO,)2 + heat CaCO, + COj + HjO
I t
Mg(HCO,)2 + heat MgCO, + CO, + H,0
WATER CONTAMINATION—ITS EFFECT 613
Magnesium sulfate is fairly soluble in hot water, and chlorides are 'usually
sufficiently soluble not to be troublesome. Sodium salts are so highly soluble in
hot water that not only are they nonscale-forming, but some chemical treat¬
ments intentionally replace Ca and Mg with Na. But unfortunately calcium
sulfate has solubility varying from 3200 ppm in!5.6®Cwater to 55 ppm in 227*
C water, and 27 ppm at 316^ .
(^^caling may take place in boiler drums or tubes, heater tubes, and feedwater
p^ng. Its effect on the piping system is to choke the flow, requiring an increase
of pressure to maintain water delivery. Its effect on heat transfer surfaces is to

Hartford Steam BoUer Inspection is Insurance Co

Fjg. 13-3 When a boiler drum lets go! General view of a wrecked boiler room sug¬
gesting the destructive forces set loose after lack of feedwater treatment, or incorrect
feedwater treatment has done its work.

decrease the transfer. The accumulation of scale may become so thick that the
temperature drop from gas to water is principally through the scale. When this
condition occurs, overheated tubes, blistering, and rupturing may be expected.
The scales that are tightly adherent and tough are the worst from the stand¬
point of removal. Others are adherent but soft, whereas still others are brittle
enough to, be removed by tube hammering cleaners. Boiler water conditioning
often seeks to precipitate the impurities in the boiler as a soft sludge which can
be removed by blowing down. Antiscale treatment consists of removing the
scale-forming elements or replacing them with extremely soluble salts.
When scale has formed, tubes are cleaned with water- or electric-powered
rotary brushes and cutters which are pushed through the tubes during boiler
514 THE FEEDWATER LOOP
overhauls. Some large, high-capacity boilers, however, have long multi-bend,
small-diameter tubes such that working out the scale with tube cleaners is not
practicable. Also, the maximum thickness of scale permitted in these high-duty
boilers is so little that cutting it out would not leave tubes sufficiently clean.
Rather, scale is dissolved with weak acid baths. Primarily the operational in¬
tent is to prevent scale.
Corrodon, By corrosion is meant the destructive conversion of metal into
(ftfldes or salts. It may occur any place in the water loop, but is most feared in
the steam generator because that is a '‘i)ressurc-fired vesser’ containing large
quantities of saturated water. But heaters, pumps, and piping are also attacked.
Corrosion is due to an acid condition of water, or to oxygen, carbon dioxide, or
chlorides. The most serious factor in corrosion is the dissolved oxygen. The per¬
missible limit of oxygen content varies with the acidity of the water and the
amount of scale on the tubes but should not exceed 0.5 cc per liter. Even this

Hartford Steam Boiler Inspection ir Insurance Co.

(A) (B)

Fig. 13-4 Results of unsuitable boiler water. (A) Extreme cases of scaling in water-
tube boiler tubes. (B) Caustic embrittlement cracking of a boiler blow-off flange.

would not be permissible in high-pressure units. Oxygen enters a closed system

through make-up, condenser leakage, open surge tanks, condensate pump pack¬
ing, etc. The corrosion may be a general loss of metal over the whole tube surface
or a locaUzed action. The latter is the more serious as it produces pitting and
grooving.(To prevent corrosion the boiler water is maintained alkaline and, if
necessary, the feedwater is processed to reduce the oxygen content to a safe
value.
Actually corrosion is a complex subject—and a highly important one be¬
cause of the immense annual damage for which it is responsible. It is probably
the most frequent cause of damage to power boilers. The steam and condensate
lines it may attack are often located in such inaccessible sites that replacement
is costly and time-consuming.
Generally, most corrosion is of an electrochemical nature involving an
anode, a cathode, and an electrolyte. In the water loop, iron will go into ionic
solution (pure water is an excellent solvent) because a steel surface is anodic,
the water will be a weak electrolyte, and a cathodic reaction will follow any dis-
WATER CONTAMINATION—ITS EFFECT 516
ruption of the hydrogen film that plates out on the ferrous surface when the
iron goes into solution. Dissolved oxygen, which becomes very active as the
water temperature increases, will unite with the atomic hydrogen, forming H2O
and producing the necessary cathodic reaction for progressive corrosion.
Oxygen corrosion occurs typically as small pits and depressions, often
covered with a crust or scab of black Fe304. For one reason or another, oxygen
attack is often localized rather than being a general wastage over a large area.
In such cases deep pits are formed and rapid failure follows.
Corrosion fatigue has recently been discovered. This is different from
mechanical fatigue, although repetitive stressing (as by thermal strains caused
by irregular condensate flow in steam conduits) is necessary, as well as galvanic
conditions. Fatigue and corrosion seem to act simultaneously, having an accel¬
erating effect on each other and contributing a definite reduction in fatigue
endurance.
There is no positive way to discover and assess corrosion damage other than
to remove the surfaces from service and examine them carefully.
Boiler operators who have been treating feedwater for absolute prevention
of scale sometimes protect against corrosion by applying protective coatings to
the internal surfaces of boilers and economizers. The requisites of a successful
coating are toughness and elasticity so as to give with the expansion of the
metal and maintain an unbroken film, chemical stability at actual operating
temperatures, a negligible, resistance to heat transfer, and a semi-permanent
adherence to the metal surface.
Foaming and priming. Foaming refers to that condition of boiler operation
where a stable foam is produced. It may or may not be accompanied by priming,
which is the production of wet steam or, in the aggravated case, slugs of water.
Wet steam is indicative of faulty operating conditions in a power boiler; slugs
of water are liable to be very destructive to piping, engines, or turbines. Prim¬
ing can be produced by other causes than foaming, for instance, carrying too
high a water level, insufficient disengagement area, or a pulsating steam demand
that overtaxes the boiler steam storage. Priming is also frequently caused by a
disturbed disengagement area—for instance, on high-capacity boilers where
tubes entering drums from water walls discharge at such high velocities as to
disturb the water surface.
The source of foaming resides in the condition of the boiler water itself. Too
high a concentration of dissolved salts is a frequent cause of foaming. Foaming
results also from saponification of the boiler water through mixture of oil or
grease with the alkali. Floating organic matter is another source of foam. When
foaming is due to concentration of salts in the water, the condition is relieved
by altering the treatment or by blowing down more of the concentrated water.
Normally, a steam boiler without drum internals will produce from 0.5 to
1.6% moisture in the steam in the form of a mist or fog. This is not permissible
in the high-temperature, high-pressure power plant. There the steam generator
has special internal purifiers in the drum. Then less than 0.1% moisture can be
normal for the steam. Leakage past faulty internals could be the cause of ab¬
normally wet steam, as well as foaming or priming.
Whatever the cause, the result is carryover, that is, the presence of impurity¬
bearing droplets of water in the steam flow. As this passes through a superheater
616 THE FEEDWATER LOOP
the water is vaporized, leaving the solid carryover as a deposit on superheater
tubes, or a dust which will float with the steam as far as the turbine stage where
the expanding steam enters the saturated steam regime. There it may form
troublesome deposits on the turbine blading. Carryover can be qualitatively
measured by recording conductivity meters which determine the micromhos of
the saturated steam flowing past electrodes located in a steam pipe. Quantita¬
tive determination (ppm solids) requires the withdrawal, condensation, and
analysis of steam samples. Foaming is the most common source of carryover on
boilers equipped with drum internals. Tests have sometimes revealed foam
blankets of 305 mm or more thickness above the drum water level.
Table 13-1. FACTORS AFFECTING CARRYOVER*
Mechanical Conditions Water Conditions Operating Conditions
Boiler design Source and makeup Rating
Drum sizes Concentration Changes in rating
Number of drums Alkalinity Pressure
Drum internals Organic matter Water level
Circulation Suspended matter Changes in level
Radiant vs. convection Chemical feed Blowing flues
heating surface Inherent foammess Blowing safety valve
♦P. B. Place, “Carryover Problems,’^ Combustion, March 1947.

EmbriMement Although this is the rarest of all boiler “diseases,^^ it cannot

be said to be so rare as to be unimportant. A serious feature of embrittlement is
that, when failure occurs, it may come as a disastrous explosion, because embrit¬
tlement affects the drums and its presence is not detectable except on minute
scrutiny. Embrittlement is attributed to the presence of a certain concentration
of sodium hydroxide in the absence of inhibiting agencies. The steel loses its
toughness and cracks appear along the seams below the water line. They gen¬
erally run from rivet to rivet, following the intercrystalline structure. In cases
of embrittlement it has always been found that the feedwater was high in
sodium bicarbonate which broke down into sodium carbonate in the boiler and
partially hydrolized. The reaction
T
NajCO, + HOH CO* + 2 NaOH
assumes considerable proportions at the water temperatures in high-pressure
boilers. Prevention of embrittlement consists of reducing the causticity or add¬
ing inhibiting agents. Formerly sodium sulfate was considered to be an inhibitor.
Recent researches have cast doubt on the reliability of sulfate, meantime indi¬
cating sodium nitrate as an efficient inhibitor of caustic embrittlement. This is
added to maintain in the boiler a ratio of sodium nitrate to total alkalinity (as
NaOH) of 0.3 minimum. It can be steadily added to feedwater either by a
chemical proportioner or by doses to the boiler drum via an internal chemical
feeder (Fig. 13-9). Water testing will control frequency of dosages required.
Embrittlement is most likely to occur in riveted drum boilers. It is not un¬
known in welded drums, a vulnerable spot being where the tube end is rolled into
the drum. Concentration of caustic at a highly stressed point is a forerunner of
embrittlement. Where embrittlement is a likelihood, boiler inspectors are care¬
ful to examine vulnerable points, even resorting to acid etching, and inspecting
WATER TESTS AND ANALYSES 517
with magnifying lens. A test of great reliability, (designed to show whether em¬
brittlement conditions exist, is the U. S. Bureau of Mines* embrittlement detector
shown in Fig. 13-5. The detector is de¬
signed to produce conditions of stress in Sfeam
the test bar and concentrations of boiler ^ Boaerumferin
water at the stressed area. The only con¬
clusion that c^n be drawn from the test A^usf/nff
screw~~'
is that (under accelerated conditions)
Concenfrof^
the water is, or is not, embrittling in sofution
character. It is not a perfect test, but the
best one currently available. Where seri¬
ous conditions exist the specimen usually
fails within a month.
13-3 Water Tests and Analyses.
The water loop equipment iias been
shown by Fig. 13-1 to be an assortment
of devices to heat, pressurize, and regu¬
late the flow of liquid water. They func¬
Y^ieroui
tion in series on the water flow. The
chemical condition of the water—what it Fig. 13-0 Embrittlement detector
carries in solution or suspension—is not a (patented). (Courtesy Power.)
matter confined to any specific piece of
equipment, but is measured at different points for different purposes. Pressure,
temperature, and quantity measurements of the water which suffice to define it
physically are simple of execution because of the liquid state. Chemically, the
determination of state is more complex—^but equally important. This section is
mainly descriptive of tests for chemical state.
The purity of the working fluid is of extreme importance to the safety and
life of steam generators. The water involved may be:
1. Boiler water. Actual liquid contents of the boiler, of which that in the
water drum is typical,
2. Blowdown. Water that is drained from the boiler. Depending on the point
of blowdown connection, this water may have equal or greater impurity con¬
tent than the boiler water.
3. Feedwater. That water which is introduced into a boiler to replace the
evaporation.
4. Raw water. Natural water with such impurities as the earth has originally
endowed it.
6. Make-up of losses. May be raw water or treated raw water.
If natural waters available were pure H2O, all the foregoing four items
would be nearly chemically identical. But pure water is not to be had from
natural sources; even waters from municipal systems, ordinarily considered
potable, aie not fit for continuous, uncontrolled use as feedwater.
Groundwater (wells and springsl has picked up dissolved hardness in se€|(p-
ing through rock strata but is usually clear.
Surface water (streams and lakes) has picked up dissolved hardness and
insoluble organic and inorganic matter in suspension. Accordingly it is turbid.
In general, natural waters need softening for the removal of dissolved solids
and sometimes clarification for the removal of turbidity.
 518 THE FEEDWATER LOOP
For the control of water conditioning on an engineering basis, it is necessary
to make certain ehemical tests on various water samples, the nature of the tests
depending upon the system of treatment used and the extent to whieh condition¬
ing reagents are required. Oj)cratorh in steam plants have found it desirable to
maintain a system of test adjustment, since in the absence of information
furnished by such tests, wahu* conditions are unknown. In larger power plants,
especially in connection witli high-pressure operation, it is essential that an
appropriate control system be instituted and that suitable tests be made at
frequent intervals on representative samples from the water loop. These tests
can be expected to supply pertinent information on such important items as;
1. Variation in mineral characteristics of the raw water supply.
2. Percentages of condensate and make-up in the feedwater.
3. Efficiency of deaeration.
4. Condition of the boiler water with respect to treatment adjustment.
5. Condition of the boiler water with respect to blow-down adjustment.
6. Correction of feedwater and boiler water for corrosion control.
7. Steam purity.
8. Return line corrosion.
9. Condenser leakage.
10. Contamination of returns from process units.
When external sofrening systems are used, test control is necessary in order
properly to adjust chemical dosage or to determine regeneration requirements.
Samples are readily withdrawn from the w’ater loop on the discharge side of
the condensate pump up to temperatures of
Waftrsomplt i 82®-93® C. At higher temperatures the
sample would need to be passed through a
flash cooler before being liberated to atmos¬
pheric pressure. Engineers use portable
shell-and-tube coolers for this purpose.
These are piped to the desired sampling
points and connected to cold w’ater supply
with garden hose. Permanent installations
are made as a boiler auxiliary in case it is
decided that water control through sample
analysis will be part of boiler operations.
In collecting samples, clean glass con¬
tainers are indispensable. Collection of sam¬
ples is sometimes a critical operation—for
example, those used for oxygen determina¬
tion. A rubber tube is attached to the cooler
outlet and placed in the sample bottle,
l2.7 Cootiog woftr outift which is in a bucket. The bottle is then filled
Cochrane Corp, to overflowing, and additional sample water
Fig. 13-6 Section through flash is flowed through it until the bucket watey
cooler for water sampling. level is above the bottle mouth. Then the
ground glass stopper is inserted, under wa¬
ter. Even with this precaution, and with the sample thoroughly chilled leaving
the cooler, the actual oxygen test should be made within the hour.
WATER TESTS AND ANALYSES 519
In the usual small plant, measurements of feeciwater quantity are ma’de only
at the rare intervals when the plant is being tested. In such cases actual weigh¬
ing is often used. Continuous flows are weighed by having tWo scale tanks with
manual or automatic means of directing the flow from one tank to the other. In
most large plants, and in some of the more efficiently operated small ones, a
continuous record of feedwater flow is kept. The make-up water is most easily
measured when cold by standard cold-water meters, either positive displace¬
ment, orifice, or venturi. Hot water measurement is more severe service, but
meters are available for this too.
A complete quantitative analysis of feedwater or boiler water is a matter for
the chemical laboratory. Rapid tests which may be performed at the plant are
essential to control of the water of large high-pressure, high-capacity boilers.
An experienced chemist can set up these tests in a simplified form so that an
intelligent station attendant can be instructed in their performance and can,
from them, maintain boiler water in a suitable condition.
Hardness. Hardness is due to dissolved salts which make it difficult to obtain
soap suds in the water. The simplest hardness test is the soap test. Hardness is
determined by mixing increments of a standard soap solution wdth the w^ater.
The sample is shaken after each addition and when lather is produced that will
last 5 mins the number of ml soap solution used determines the hardness.
TT j TT r. ~ f^o) X 1000
Hardness H, as ( aCO, = — ppm (13-1)
ml of sample
where S — ml soap solution used; So = ml for lather in distilled water.
One U.S. degree is the equivalent of 17.11 mg calcium carbonate per
liter or 17.11 ppm by weight. Hardness by soap test is not an accurate indi¬
cator of feedwater quality as it indicates such a soluble salt as sodium chloride
as being equally hard as calcium sulfate,
A large quantity of dissolved CO2 in the water sample interferes with the
soap test and should be eliminated if present.
The type of hardness that can be removed by heating to the boiling point is
temporary hardness. This is also called ‘‘carbonate hardness,’' because it is
effective in purifying water containing calcium bicarbonate or magnesium bicar¬
bonate in solution. The reactions have previously been given. The boiling is
generally conducted at atmospheric pressure (100®Cl . Heaters using steam at
a few kg gauge can also be considered to treat for temporary hardness if
their internals include a filter for holding the precipitated CaCOs and MgCOs.
Waters containing salts wdiich precipitate only at higher temperatures are
said to have permanent hardness. This hardness is removable by suitably
chosen chemical reaction. Waters having less than 10 degrees hardness are con¬
sidered soft, while those having more than 25 degrees are considered very hard
and will always require treatment.
pH Scale. Pure water is tasteless, odorless, colorless, and “neutral,” Natural
and commercial waters are never pure. Because water is a most universal sol¬
vent, and because of ionization, either OH"" or H+ ions will predominate, caus¬
ing either an alkaline or acidic condition. A scale, called pH, has been devised to
measure the intensity of acidity or alkalinity of a solution. An exact definition
of the pH number is that it is the logarithm of the reciprocal of the hydrogen
620 THE FEEDWATER LOOP
ion concentration (grams per liter). Just as temperature is measured in degrees,
so is acidity or alkalinity measured in pH values from 1 to 14. pH 1 is strongly
acid, pH 14 strongly basic, and pH 7 a neutral solution. pH 5 is 10 times as acid
as pH 6; pH 4 is 10 times as acid as pH 5 and 100 times as acid as pH 6. As a
corollary of the law of mass action in dissociation, OH’" X H+ = 10*"^^. For
instance, if H+ is 10’"^, then OH“" is 10"’® gram ions per liter and pH 9 desig¬
nates the solution which, of course, is alkaline as OH ions predominate. Control
by pH has been widely adopted, and experience has demonstrated the pH
values which should be maintained. Control to hold alkalinity between pH 10
and pH 11 is a common practice. Although it might, at first, seem that pH 7 was
the desideratum, the fact is that iron will go into solution in pure water far more
rapidly than in alkaline water held between pH 10 and 12.
Alkalinity. Alkalinity is the condition of a predominance of OH"” ions over
H+ in a solution. There are uses of alkalinity as detected by methylorange and
phenolphthalein indicators and reported as hydroxide alkalinity, carbonate
alkalinity, etc. Using standardized technique, the sample is colored with
phenolphthalein and titrated with acid to an end point. From the titration
quantity, alkalinity as ppm CaCOa is determined. Let this be P. The same
sample is then colored with methylorange and again titrated with acid to an
end point. The total acid used defines total alkalinity as j)pm CaCO^. Let this
be M. Then with i)pm of hardness, H, and the data contained in Table 13-2,
certain useful information as to the nature of water hardness comes to light.

Table 13-2. ALKALINITY RELATIONS

Where water analysis shows: Then these quantities are expected:
Total Tempo- Perma¬
Free Carbon- Bicar- rary nent
Caustic ates bonates Hardness Hardness
P* = 0 P* = 0 0 0 M 0 0
P = 0 H < M* 0 0 M H 0
P = 0 H = M 0 0 M H 0
P = 0 H > M 0 0 M M H-M
P < M/2 P = 0 0 2P M-2P 0 0
P < M/2 H < M - 2P 0 2F M-2P H 0
P < M/2 P = M - 2P 0 2P M-2P H 0
P - Af/2 0 2P 0 0 0
P> M/2 2P--M 2(M - P) 0 0 0
P ^ M M 0 0 0 H

*P phenolphthalein alkalinity, M = methylorange alkalinity, ff = hardness (£q

13-1); all in ppm as CaCOa.

Dissolved O2 and CO«. These two corrosion accelerators are detected and
quantitatively measured by a chemical titration routine wherein the quantity of
titrating reagents is used to determine the dissolved gases in ppm. Both tests
must be carried out soon after collecting the sample or their validity will be
under suspicion. The oxygen test commonly employed is known as the Winkler
test. Special portable apparatus has been designed to facilitate carrying out this
test rapidly in the plant where the sample is drawn.
Other chemical tests for boiler water include those for chlorides, sulfates,
WATER TESTS AND ANALYSES 521
sulfite, and nitrates. Calorimetric or gravimetric methods are employed’for
residual phosphates, silica, oil, and dissolved solids.
Dissolved Solids, Primary gravimetric tests for dissolved solids involve time
and equipment requirements that place these methods outside the scope of
routine boiler water testing. However, special types of hydrometers and elec¬
trical conductivity meters are available for making field tests of the mineral
content dissolved in water. Most of these are sufl&ciently accurate for plant
work. Tolerance limits for TDS (Total Dissolved Solids) in boiler water re¬
quested by boiler manufacturers are set down in Table 13-3.

Table 13-3. MAXIMUM BOILER WATER CONCENTRATIONS

RECOMMENDED BY AMERICAN BOILER
MANUFACTURERS' ASSOCIATION
Boiler Steam Boiler Water
Pressure Concentration
(kg/omS) {not over-ppm)
Olo ii 3500
OVGitt vplosi. 3000
over 31 upto 4St 2500
over 42 upto 52 2000
over 62 npto 68 1500
over 63 «pio 70 1250
over 70 upio 105j6 1000
Alkalinity not to exceed 20% of total concentration.

Complete Water Test. Natural or raw water is the supply most frequently
subjected to a full analysis. On account of ionization, it is impossible to de¬
termine by analysis of .water the nature of each compound that was dissolved in
it. However, testing can detennine the dissolved constituents by ionic measure¬
ments so that, although the results cannot state how much of the compounds

Table 13-4. PROPERTIES OF COMMON IONS

POSITIVE IONS NEGATIVE IONS

Name Formula 7. W,* Eq, W. Name Formula 7. W.* Eq, W.

Aluminum A1+++ 27.0 9.0 Bicarbonate HCO,- 61.0 61.0
Ammonium NH4^ 18.0 18.0 Carbonate CO,— 60.0 30.0
Barium Ba+-^ 137.4 68.7 Chloride Cl- 35.5 35.5
Calcium Ca++ 40.1 20.0 Fluoride F- 19.0 19.0
Hydrogen. H+ 1.0 1.0 Iodide I- 126.9 126.9
Ferrous iron Fe-^+ 55.8 27.9 Nitrate NO,- 62.0 62.0
Ferric iron Fe+++ 55.8 18.6 Hydroxide OH- 17.0 17.0
Magnesium Mg++ 24.3 12.1 Phosphate 3 PO,— 95.0 31.7
Manganese Mn++ 54.9 27.5 Phosphate 2 HPOr- 96.0 48.0
Potassium K+ 39.1 39.1 Phosphate 1 Ha>Or 97.0 97.0
Sodium Na+ 23.0 23.0 Sulfate sor- 96.1 48.0
Sulfite 80,” 80.1 40.0

♦ I. W. = Ionic weight.
622 THE FEEDWATER LQOP
such as calcium sulfate or magnesium carbonate was in solution, the plus ions
(cations) and the negative ions (anions) may be separated out and measured.
Because of this fact it is reasonable to report dissolved water impurities as ions,
in parts per million by weight. Some of the common positive and negative ions
of water analyses are shown in Table 13-4. The equivalent weights given there
are the weights that contain or will “match up with^^ a unit weight of hydrogen.
For example, the molecular weight of calcium is 40, but its valance is two com¬
pared to hydrogen^s one. Therefor'^, its equivalent weight is 20.
Of considerable assistance in arriving at the treatment requirements of an
analyzed water is an arrangement of the soluble ions in anodic and cathodic
equivalents per million, *'epm.^’ To obtain an epm, divide the ppm of an ionic
substance by the equivalent weight. However, some analyses are composed of
'^ppm as CaCOa^^ then epm is the ppm as CaCOa divided by the equivalent
weight of CaCOa.
For a perfectly analyzed water the sum of the plus epm’s should equal that
of the minus epm s; however, since ionic water analysis technology is not usually
able to achieve this accuracy, the tester usually reports some -f-Na or —NOa
sufficient to make the quantitative balance. A breakdown of an ionic water
analysis into positive and negative epm is shown in the next example.
Example 1: The following ionic analysis of a water sample is in ppm. Equivalents
per million are obtained by dividing with equivalent weights taken from Table 13-4.

Ionic, as reported +€jym —eprti

Ca 29.0 . .1.445
Mg 7.5 . .0.620
HCOa 98.0 . . l.(J05
SO4 22.0 . . 0.458
NO3 2.1 . . 0.034
Cl 9.2 . . 0.259
2.065 2.356
A balance is provided by adding epm of 2.356 — 2.005 = -f0.291 for Na. Ppm
of Na = 0.291 X 23.0 = 6.8. We would therefore add Na 6.8 to the ionic column, and
0.291 to the d-epm column, thereby bringing anions and cations into balance.
13-4 Deconcentration (or Blowdown). In plants using raw water or
chemically treated water for make-up, there is a steady accumulation of dis¬
solved solids in the boiler water. Even where make-up is by evaporation, the
boiler water concentration will increase, although slowly. There is an upper
limit beyond which concentrations of total dissolved solids may produce trouble¬
some foaming and carry-over. In individual cases local conditions may demand
even lower concentrations. C;ontrol of concentration is obtained by two methods,
both involving draining off some of the concentrated boiler water.
1. Periodic manual blow-off. Employed in smaller plants and where ppm of
solids in feedwater is low.
2. Continuous flow deconcentration, usually accompanied by heat-saving.
The quantity of blowdown that will maintain a given status of the boiler
water is readily determined by a balance of solids in and out of the boiler. Fig.
13-7 is provided to illustrate the Quantities involved.
DECONCENTRATION (OR BLOWDOWN) 623
Every boiler must have a manually operated blov^-off valve connected to the
mud ring or mud drum. If this is opened at intervals some boiler water can be
blown out, carrying not only dissolved and suspended impurities, but also any
precipitate or silt that may have gravitated to the bottom of the boiler. The
water so removed contains the saturation heat of the liquid which, of course, is
entirely wasted. If the same quantity of water were removed in a slow but con¬
tinuous stream, there could be equivalent deconcentration with simultaneous
salvaging of the heat in the blowdown via exchange with the incoming feed.

This is also schematically illustrated. Symbols on the diagram are defined as

follows:

Ws == Weight of steam evaporated during a given interval

Wb = Weight of blowdown required in same interval.

m = Fraction of make-up in entire feedwater. (Remainder taken to be

condensate.)

iSb — Concentration maintained in boiler, ppm.

Sm ~ Concentration in make-up water, ppm.

Si = Concentration in feedwater, ppm.

ti, to « Heat exchanger blowdown water temperatures.

ti\ to' Heat exchanger feedwater temperatures.

A balance of solids gives the equations:

iShWa * mSjnCWg Wb) (13-2)

SbWB « SfiWB + Wb) (13-3)

624 THE FEEDWATER LOOP
These equations can be used in various ways to solve the blowdown prob¬
lem.*
For example, a solution of Wb/Ws gives the rate of blowdown as a fraction
of steam generated. It will be shown how a knowledge of boiler water and feed-
water concentrations can be used to determine the proper interval between
manual blowdowns. In far too many small boiler plants this is done by an
assumed routine derived from fireman '^experienced’ and usually results in more
water than necessary being blown down. That this is an error on the side of
safety does not alter the fact that it may be costing the owner several hundred
dollars a year. The cost of a sensitive hydrometer suitable for determining
boiler water ppm is a small fraction of this.

Example 1: Suppose a water tube boiler with steam drum 1220 mm x 1820 mm
long has normal w^ater at mid-drum and is equipped with a 250 mm gauge glass.
Operating pressure, kg/cm* ga. Let it be required to determine the proper interval
between blowdowns, assuming that a “half gauge'' will be the quantity blown off. The
boiler has a steam flow meter of the indicating-integrating type. Given = 12 ppm,
= 2500 ppm.
The volume of w^ater removed is approximately a rectangular volume 125 mm x
1220 mm x 1830 mm, or 0.28 m*. Its density is 881 kg/m^ at the operating pressure; hence
a halt gauge is 247 kg of boiler water. Now with Eq 13-3, 12(lVg + 247) » 2500 x 247.
Wq = 51203 kg steam flow between blowdowns. When the steam flow integrator on
the meter shows 51203 kg above the reading at the time of the last blowdown,
another half gauge blowdown is in order. For a safety allowance m plants w'here Si>
is only occasionally checked, probably it would be better to use 50% to 75% of the
calculated TTg for blowdown guidance, then adjust it to a schedule routine in boiler
operation.
A manual blow-off system is an essential boiler attachment. It must be in¬
stalled even if continuous blowdown is practiced. Continuous blowdown equip¬
ment is installed only if justified on the basis of heat lecovery, or some tangible
advantage to boiler operations. By a development from Eq 13-3, we can say that
the annual saving due to heat recovered from continuous blowdown is:

Annual saving = 0.001 hWa X C X dollars

— Of

where kWa = Estimated annual evaporation, thousands of l^g

C = Dollar value of 10* Joules in the feedwater.
Ahf = Heat saved per Joulcs/kg between and •
The blowdown required to keep the Si02 content of boiler water low enough
so that vaporized Si02 will not promote turbine operating troubles may be more
than that to maintain maximum water concentrations on a total ppm basis. This
should be checked for all high-pressure cases. Eqs 13-2 and 13-3 are valid if
Sby Sfy Sjn are taken to be ppm of Si02.
Much high-pressure operating experience is needed to fix allowable Si02

* Note that had m been the make-up as a fraction of TFg.

= S^imW^ + Wj,)
WATER TREATMENT 525
limits in boiler water. Some present recommendations are roughly indicated by
the following equation; however, future experience may well alter these.

£^b(8i) 161.88 — 0.513 X (saturation temperature) ppm

Furthermore, plans for use of continuous deconcentration equipment should
be checked to see where and how heat can be re¬
moved from the blowdown water. Since it should
be cooled well below 93.3® C to eliminate flashing,
the heat cannot always be delivered to feedwater,
for in a regenerative cycle the feed leaving the last
heater might already be 149® C or more. Some heat
might be delivered to the high-temperature feed,
and the remainder to preheating the make-up. This
is implied by the equipment shown dotted in Fig.
13-7. Also flash tanks could be installed and heat¬
Fig. 13-8 Proportions of a
ing accomplished (to 100® C) by flashed steam.
blow-off tank. (Volume A
Example 2: Given annual amortization rate of should be about twice that
20%, the justifiable investment cost of a continuous of expected maximum blow
blowdown system fora 48 kg/om* ga boiler will be caloa- per boiler.)
lated.Data : Average rate of eteam generation 2.76 x 10*
kg daily at 0.72 efficiency. The coal, which costs $ 9.37 per tonne, has 27.9x 10* J per kg.
Make-up at 16.6® C,m = 0.03, = 200, 8,n(Ri) = 2.
From data of this chapter, 8^, = 2000, 8b(Si) = 32, Sf/(S^ — St) =

Wb/Wb = 0.03 X 200/(2000 - 0.03 X 200) = 0.00301 for TDS control.

Wb/Wb = 0.03 X 2 (32 — 0.03 X 2) = 0.00188 for Si control.
Thus total dissolved solids are found to determine the rate of deconcentration.

ieWs = 2.76 X 10* X 866 x 10"^ *= 1004 x 10* kilo kg per year.
C » 9.37 X 10*/(1000 X 27.9 x 10« x 0.72) = $ 9.37/279 x 72 per lO* Joules.
ti * 264®C ; assume to =27®0; then Aht = 993.2 x 10* Joules.

Justifiable investment ®=(l/20%)(lt)04 xlO* x ^75^575 ^ 0.0031 x 983.2 xl0*)=* $7003.

The cool effluent from a continuous blowdown system may be directly

wasted, but the hot rush of fully saturated water following opening of the
manual blow-off is dangerous and must be flashed and cooled before it can be
discharged to the sewer or to surface drainage. Blow-off pits and tanks are in¬
stalled for this purpose. These receive the blow-off water, let it flash, vent the
vapor to atmosphere, and discharge the cooled water to drainage.
13-5 Water Treatment. Treatment can be mechanical, thermal, and
chemical. Examples of mechanical treatment are sedimentation, filtration, and
interior painting. Distillation and deaerative heating are thermal processes.
Chemical treatment may consist of chemical reactions with prepared reagents
to cause 'precipitation or maybe exposure to a zeolite in order to obtain an ion
exchange. Ion exchange is always carried out on make-up or feedwater before it
enters the boiler, but precipitation is practiced both external to and internally in
the boiler water.
526 THE FEEDWATER LOOP
Internal treatmenf^ means the addition of chemical reagents to water al¬
ready in the boiler. This is usually done by periodic additions of the chemical
which is placed in a feeder attached to the boiler drum or shell and so piped
that, after the feeder is filled and closed, feedwater pressure can be admitted to
force it slowly into the boiler drum. Pot feeders are available from manufactur¬
ers; also, one may readily be constructed by a piping contractor, as shown in
Fig. 13-9. Openings A and B are connected so as to let
feedwater pressure either inject the chemical directly
to the boiler drum or to the suction side of a feed pump.
The former is preferable where there are two or more
boilers, each with its own feeder. Internal treatment
may then be individualized for each boiler. As the
reagents are usually more dense than water, they are
displaced downward out of the feeder by connecting B
to the boiler and A to the feedwater header. Thus the
pressure drop across the feedwater regulating valve is
applied to the feeder. Tricocks enable the progress of
feeding to be estimated, gauge glasses being useless for
this purpose on account of internal incrustations by
the chemical. The line to the boiler should be orificed
or includ(‘ a small valve such as 3.2 mm or 6.4 mm so
that the rate of reagent feed will be restricted and the
feeding period extend over an hour or more.
Internal treatment is widely employed by opf'rators
of small boilers, and by others whose feedwater prob¬
lem is not intensified by the presence of large ppm of
dissolved salts in the water. It has also been employed
to give supplementary treatment where external soft¬
ening is the principal treatment. However, the plant
Fig. 13-9 Attachment operator should not expect any good results unless he
to feed chemicals for adjusts the treatment, from periodic analyses of boiler
internal treatment water samples. It is not advisable to use internal treat¬
ment for high pressure (over 21 kg/cm*)boilers or for
those fed wdth water of high solids content, say over 15 pi)m.
Some of the chemicals employed internally are:
Sodium hydroxide. To establish alkaline pH, and precipitate magnesium. *
Sodhnn carbonate. To precipitate calcium.
Sodium aluminate. To coagulate finely divided precipitates into a floe; also
to precipitate calcium and magnesium.
Sodium 'phosphates. To precipitate calcium in form favorable to riddance
through blowdown. The trisodium phosphate reaction is

i
3 CaCOs + 2 Na8P04 Ca3(P04)2 + 3 Na^COg

Sodium sulphite. To complete deoxidation.

Colloids. Protective colloids such as tannin; and reactive organic forms to
make a fluid sludge of suspended particles.
WATER TREATMENT 627
f

Acid, To control alkalinity. A t3rpical reaction:

Na«CO, + H2SO4 Na2S04 + HaCO,.
Also to be classed as internal treatment is the employment of electrolytic
apparatus of the sacrificial anode type. Such apparatus is installed in the boiler
drum below the water line and consists in part of a special metallic anode which
is slowly consumed during the period when it is protecting the boiler against
scaling and oxygen corrosion. That such a protective system could be successful
is logical considering the electrolytic nature of both oxygen corrosion and salt
solubility. Advantages are: (1) a self-inherent adjustment to the water need,
obviating control water analyses; (2) simplicity; and (3) minimum attention.
Disadvantages are: (1) cost of replacement anodic cartridges; (2) possibility of
unperceived exhaustion of the anode; and (3) limitation to boiler pressures less
than 24.5 kg/cm* approximately.

Table 13-5. SUMMARY OF WATER TREATMENT

Limit of Tolerance^ Kind of

Impurity Trouble in Feedwater Treatment

0, Corrosion 0.03 cc per liter Chemical deoxidation^

Thermal deaeration

CO, Corrosion 3 ppm Thermal deaeration

Ca, Mg, salts Scale 20 (low pressure) to 2 (high External softening

pressure) ppm as CaCOs Internal softening

Fe Scale Trace Ion exchange

SiO, Carryover 40 (low pressure) to 5 (high Deionization

Scale pressure) ppm

Na alkalinity Embrittlement 5 ppm as CaCOs at high press. Acid neutralizer

Foam Variable at low press.^

TDS» Priming See Table 13-3 for tolerance Deconcentration

Foam limits in boiler water

Turbidity Sludge 5 ppm or less Coagulation

Sediment Sedimentation
Filtration

ou Foam 1 5 ppm or less Coagulation

Filtration
Surface blow-off

^ These are an average of practice. Quantitative statements alwa3r8 debatable.

•To impart excess 30 ppm sodium sulfite in boiler water.
• Total dissolved solids in boiler water.
•’ABMA specification: Not to exceed 20% of TDS.
628 THE FEEDWATER LOOP
External water softeners are divided into precipitation and ion exchange
types. There are a few rather generally used reactions for precipitation and a
great many specialized patented treatments which can be prescribed for unusual
and difficult cases of water treatment. It could not be expected that all of these
could be covered in a book of this character, nor would it be feasible since
highly specialized knowledge of water chemistry is needed to understand many
of them. We shall mention some of the treatments that are frequently used be¬
cause they can economically cope with the average water problem. The reader
faced with an unusual and serious water problem needs the consultative services
of a specialist in this field.

On separator
Vent condenser ~ -
-'Exhaust inlet
Adjustable or/f/ce
^Steom otonuzer

Wash-
water
return

* Phospi Chemical agitator Sludge ' ,

tank btomotf solve Nan-siliceous '
Phasphote - Ume-soda tank filtering material
pump
Power

Fig. 13-10 Hot process lime-soda ash softener, with deaerator section.

A precipitation softener embodies the principle of using calculated quantities

of soluble reagents to react with the hardness in the raw water. Two treating
tanks are used in the intermittent system, one supplying treated water to feed
service while the other is receiving its charge of chemicals and water or main¬
taining a quiescent condition so that the precipitate may settle out. On the other
hand, water flows continuously from inlet to outlet in the continuous type of
softener. Reagents are added at the inlet, and what precipitate does not settle
out in the reaction tank is removed by filtration.
It is well known that chemical reactions are accelerated by increased tem¬
peratures. This fact forms the basic advantage of the hot process softeners
wherein the raw water is heated nearly to the boiling point before the reagents
are added. The reaction is very rapid and the character of the precipitate formed
is such that sedimentation is greatly improved. Lime (CaO or Ca(OH)2) and
soda ash (NaaCOa) are frequently used. The lime precipitates the carbonate
WATER TREATMENT 629
hardness; and soda ash, the sulfate hardness. Soditim phosphate is also*em¬
ployed in external softeners, for the hot process lime-soda treatment is fre¬
quently followed with a slight phosphate reaction to obtain near-zero hardness.
The lime-soda treatment reactions are presumed to be:
i
Ca(HCO,), + Ca(OH)2 2 CaCOa -f 2 H2O
i i
Mg(HCO,), + 2 Ca(OH)2 2 CaCOs + MgCOH)* + 2 HjO
i
MgCb + Ca(OH)2 Mg(OH)2 -f CaCU

CaS04 + NajCOs CaCO» + Na2S04

i
CaCb + Na,CO, CaCO, + 2 NaCl
i
MgS04 + NajCO, MgCOa + Na2S04
i i
MgCOa -f Ca(OH)2 Mg(OH)2 + CaCO,
A hot-process external softener is shown in Fig. 13-10. Raw water which
enters through a float-controlled regulating valve passes through a chemical
proportioning orifice, a vent condenser, and a preheater, after which it receives
the lime and soda ash. The precipitation occurs as the hot water moves down¬
ward to the bottom of the reaction chamber. Here it drops the coarse precipitate,
then flows upward to the jet type deaerator. Deaerated water then flows
through a filter bed, after which it is ready for boiler feed or make-up storage.
The reagent storage and proportioning equipment is at the left of the reaction
chamber, the filter with its backwash system for cleansing the bed being to the
right. The following example shows how an ionic water analysis, reduced to ±
epm form, is used to calculate the required quantity of lime and soda ash.
Example 1: A water analysis has been reported as: Ca, 62.8; Mg, 15.45; HCO2,
200.5; SO4, 30.0; Cl, 13.81; all in ppm. It is desired to fix the quantity of 96% CP
soda ash and 87% CP lime needed for treatment of S785 liters of this water, leaving an
excess of 17J1 mg per liter soda in the treated water.
This calculation is simplified if done on an epm basis wherein the hardness is
matched by the same epm of reagent. Reactions previously cited indicate that the
lime is supplied to soften magnesium and bicarbonate hardness, with soda ash for
calcium.
First step is to convert the analysis to epm. Divide by the equivalent weights from
Table 13-4. This should give a balance of positive and negative ions, but one may
usually expect a discrepancy because exact ionic analyses are difficult to achieve
experimentally. Transferred to epm, the water analysis becomes:
Ca+^3.14; Mg-^M.28; HCO," 3.29; SO4- 0.625; Cl", 0.39; all in epm.
Reactions: 2 HCO," + CafOH), CaCO, + 2 HjO + CO,"
Mg++ + CaCOH), Mg(OH), + Ca++
Ca-H- + NaaCO, CaCO, + 2 Na++
Lime needed: The HCO, and Mg are 3.29 + 1.28 =» 4.57 epm. The ppm of
CafOH), “ epm hardness X Eq. wt. CafOH), ■» 4.57 X 37 »10O.O j^tni.aiiioe 17.11 mg/
liter ••I7.11ppm,unie quantity per $7851water— 17.U x 189 x 8785/17.11 xO.STxlO**" ^*798 hg«
630 THE FEEDWATER LOOP
Soda ash needed: The Mg reaction will produce Ca in addition to that in the water
axial)rsis. However, the HCOs reaction produces CO3 which is instrumental in precipitat¬
ing calcium, thus reducing the amount of Na2C03 required. Total Ca == 3.14 + 1*28 «
4.42 epm. Net Na2C08 required = 4.42 — 3.29 = 1.13 epm. The ppm of Na2C08 «
epm X Eq. wt. = 1.13 X 53 ~ 59.8 ppm. Since 17.11 mg/liter exoeas soda « 17.1
(50.8 17 1)8786
ppm, soda ash quantity per 1000 gal water = 0.803 kg

Ion-exchange Processes. There arc materials which will exchange their base
radical with anions in solution in water brought into contact with them. If an
exchange of calcium for sodium might thus be accomplished, the action would
be a water-softening one. The word zeolite is associated with this action.
“Zeolite’^ was originally applied to certain natural sands, but artificially pre¬
pared porous materials have supplanted natural zeolite. The most used form is
sodium zeolite, Na2(AloSi208); simplified symbol, Na2Z. Recently organic zeo¬
lites have been iiroduced, containing no silica. These are carbonaceous or resin¬
ous, but like inorganic zeolite they exchange the base radical with water perco¬
lating tlirough a bed. Typical sodium zeolite actions on hard water arc:
CaS()4 Na2Z —> Na2v>04 C^aZ

Mg(HC()3)2 + NaaZ 2 NaHCO., + MgZ

From the foregoing it is seen that, as hard water flows into a zeolite bed and
emerges as soft water, there is a loss of sodium and gain of calcium by the
zeolite. Eventually the bed will become exhausted and fail to soften the water,
h'ortunately, zeolite possesses another desirable property—the regeneration of
its strength through backwashing with a suitabk sodium solution. If a calcium
and magnesium loaded zeolite bed is washed with brine (sodium chloride), it is
thus regenerated:
C^aZ + 2 Nad NazZ +
MgZ 4- 2 Nad NazZ -f Mgd.
An advantage of ion exchange over precipitation is noted from these reactions,
namely, that hardness is disposed of as a solution instead of a solid requiring
filtration.
Th(' operation of a zeolite softener is explained by Fig. 13-11. Normally raw
water enters the zeolite tank at the top, although the system can be by-passed
while backwashing, if necessary. It flows downward through the bed, trading
hardness for Na ions. There is no precipitation, no reduction of TDS; hence no
follow-on filter is needed. The figure also shows arrangements for the periodic
backwashing with brine needed by this system.
C'onsidcr that the organic zeolite is represented by H2Z. Typical softening ac-
tion is as follows:
CaS04 + HzZ CaZ + H2SO4
Backwashing an exhausted bed requires an acid for hydrogen exchange:
TaZ + H2SO4 HzZ + CaS04
The softened water is now acidic, requiring neutralization. It also makes it
mandatory to build the softener of corrosion-resistant material. In some in-
WATER TREATMENT 631
stances raw water treatment is proportioned between Na2Z and H2Z softeners
and the effluents blended. As one is alkaline^ the other acid, the mixed product
can be of chosen pH value if the raw water proportioning is correctly done.
As the hydrogen zeolite process leaves acids, including H2CO8, in the
effluent, a demineralization is possible if the zeolite is followed by degasification
to remove CO2 and anion exchange (another type of resinous zeolite exchanger)
to absorb the acids. This is a chemical alternate to evaporation as a de¬
mineralizing action.
Silica Removal. This substance may be present in water in either the
crystalloidal or the colloidal^ forms. Water analysis convention is to express it
as Si02. As a colloid it is in suspension, finely divided, and yields to coagulation
and filtration. In the soluble, crystalloidal form it is extremely difficult to re¬
move.

Power

Ficj. 13-11 Ion exchange system—zeolite

Presence of silica in boiler water has become increasingly serious as boiler

pressures trend upward Silica, being vaporized by high water temperatures,
goes over into the turbine with steam and mostly deposits on turbine blades to
cause unbalance and reduction of steam flow area. For this reason very iow
concentrations of silica are allowable in boiler vrater. High-pressure boiler
operators will control blowdown by silica concentrations rather than by TDS.
Studies seem to have demonstrated that, although 200 ppm silica is allowable in
14kg/cm*boilers,only6-10 ppm‘can be tolerated in boiler water at88 kg/cm* and
have no silica troubles. Silica in solution is amphoteric and in several other ways
unpredictable. Ordinary treatment methods leave it nearly unscathed. Hence it
is the recipient of special interest from those associated with the operation of
high-pressure boiler plants. Many special processes are under development for
silica control.
Summarizing the engineering of power plant water conditioning, it is seen
that the problem may vary from the simple to the complex. Often it may be
solved in more than one way. Competitive methods are not of equal cost or up-
632 THE FEEDWATER LOOP
keep and do not require the same degree of skill to operate. Not only must the
engineer have a good prediction of chemical results of alternate methods, but
lie must be prepared to analyze chemical ojicrating costs, installation costs, and
the prospects of the operating staff being sufficiently skillful to give proper
supervision and control to the treatment system.
13-6 Feedwater Heating. AVe shall now turn to details of the heating
equipment implied by Fig. 13-1. As the diagram shows, this equipment may be
the extraction steam heaters of the regenerative vafior cycle; also, it may in¬
clude low- and high-lc\Q\ heat-saving equipment. The low and high refer to
relative temperature levels at which the heat is available. The low-level heat
must be added to the feedwater stream near the condenser where the water
itself is cool Between the low and high heat levels there may be water heating
as a by-product of treatment. Thus it is with deaerators and evaporators.
Feedwater heaters arc divided into two classes, the contact and surface
heaters. The economizer and part of the boiler constitute actual water heating
surfaces; however, it is customary to refer only to equipment obtaining heat
from steam as ^‘feedwater heaters.” Live (boiler) steam is used in heaters only
when the primary purpose is evaporation or the thermal precipitation of hard¬
ness external to the boiler.
The arrangements of equipment in the water loop are innumerable and no
comprehensive survey can be given here. A review of certain of the illustrations
of Chapter 8 will show typical examples.
Where it has been possible to «ipend the maximum amount for heat recovery
equipment, oil coolers, generator coolers, etc., are incorporated into the feed
loop. The generator cooler is employed when the generator windings are cooled
by the closed system, using either air or hydrogen. Condensate may be taken
directly from the hotwell and pumped through the heat exchanger of the cooling
gas circulating system. In order to insure adequate cooling of the windings in
hot weather, or during light loads, the condensate cooled heat exchanger is often
supplemented by a raw water-cooled exchanger section.
13-7 Contact Heaters. Open and jet heaters constitute this class. The open
heater is ordinarily built up in rectangular form, but heaters for other than
near-atmospheric pressure are constnicted in cylindrical form of steel plate.
The open heater is provided with tiers of trays, designed to break up the flow
of water, delivered by gravity from a distributing pipe or trough, into a multi¬
tude of small cascading streams which present a large surface to the steam. It
is possible to heat water to the temperature of saturated steam entering the
heater if there are no noncondensable vapors. Heating is by direct conduction
from steam to water so that the effectiveness as a heater is not adversely
affected by scale accumulation. A float-regulated valve admits the cold water
required to supplement the returns in drips, condensate, and other uncontrolled
feed supply.
In addition to its heating function the ordinary open heater performs the fol¬
lowing important services:
1. Acts as a storage reservoir for heated feedwater.
2. Precipitates temporary hardness and removes it by a filter bed in the
heater.
3. Provides a convenient receiver for condensation from various sources.
CONTACT HEATERS 633
4. Removes oil from engine or pump exhaust. (M6st open heaters have oil
separators.)
The modem open heater ^'induces"' the steam into it through a connection to
the exhaust line, by the slight pressure reduction created by condensation of the
induced steam. The steam is frequently a portion of the exhaust of an engine.
In condensing plants, steam for the open heater is obtained from steam-driven
auxiliaries, such as stoker and fan engines, or from
the exhaust of a noncondensing auxiliary turbogener¬
ator. The auxiliary unit is used to supply a part of
the energy to station auxiliaries in accordance with
the feed heating demand for its exhaust steam, the
balance of auxiliary energy being obtained from the
main generator.
The jet heater is similar to a low-level parallel-
flow jet condenser mounted atop a storage tank. Wa¬
ter is heated by spraying it in fine drops from nozzles
in the rim of water belts which surround the mixing
space. The nozzle area is proportioned to discharge
the maximum flow with a velocity sufficient to throw
a good spray. Nozzles are arranged in two or more
nozzle belts so that, for fractional flows, groups of
nozzles can be manually or automatically cut out of
service. A few central power stations have used re¬
generative jet heaters
Swartwout Co.
The open type of contact heater should be used
for small- and medium-sized plants, but the jet heater Fig 13-12 “Open” type
is well suited to capacities above 90,000 kg water per of contact heater.
hr. Compared to the surface heater it has advantages
of simplicity and compactness. No cascading of drips is necessary, but a booster
pump is required in the feed line between each heater.
Deaerating Heaters. The role of oxygen as a corrosive agent has already
been explained. It is desirable to reduce the oxygen content of feedwater to zero,
or practically so. Heating of water to 104-110® C with subsequent agitation is
capable of freeing the water from this gas. While deactivation has been accom¬
plished by chemical reaction and by water flashing,the trend is toward traytype
oontact heating using steam at 0.35-0.7kg/cm* gainoontactmixing* By operating
at a plenum the liberated gas can simply be released to the atmosphere. On all
but small-sized units the released air-vapor mixture first passes through a vent
condenser so that finally only the gas is discharged.
The form of one of these deaerating heaters is shown in an accompanying
figure. The heater unit is generally mounted on a horizontal storage tank, or the
storage section of the heater is made sufficiently deep to fulfill the needs of a
reservoir of deaerated water.
The feedwater courses first through the vent condenser, then to the spray dis¬
tributor, from which it is projected upwardly, falling in a uniform shower over
the entire width of the beating tray stack. After descending through the nest of
heating trays and then through the air separating trays, it passes from the tray
compartment into the storage space of the deaerator.
534 THE FEEDWATER LOOP
The steam enters through a nozzle in the side of the shell, filling the entire
space between the shell and the tray compartment. It then fiows downward
through perforations in the top plate of the tray compartment, meeting the
water sprayed upward from the distributor. The steam continues down through
the heating tray stack and air separating tray stack.
Practically all of the condensation of the steam occurs in the spray distribu¬
tor space and in the heating tray stack, the proportion of steam drawn through
the lower or air separating trays being determined by the capacity of the vent
condenser.
From the bottom of the air separating tray stack the vent steam, together
with the separated gases, passes to the vent condenser. The vent condenser drips

Fig. 13-13 Diagram of tray-type deaerating heater.

are returned through a sealed connection to the tray stack, where they are de¬
aerated.
Merely rendering gases insoluble by heating water to boiling temperature
does not of itself eliminate molecules and bubbles of gas in the mass of water
and, unless they are separated, they will enter into solution again as soon as the
temperature is reduced and the pressure increased. In order to escape from the
mass of water, gas molecules must diffuse through the surface film surrounding
the particle of water. The rate of diffusion is proportional to the concentration of
gas within the water particle, to the surface area, and to the time of exposure.
Diffusion is, therefore, much more rapid from the small particles of water con¬
tinually mixed, agitated, and broken up by successive layers of trays than from
appreciable masses of still water. The repeated agitation and breaking up of the
water screen as it passes over the deaerating trays serve to bring the gas mole-
CONTACT HEATERS 635
cules to the surface rapidly and eliminate the need of the gas to move from the
interior of each water particle to the surface film by diffusion.
The atomizing deaerator does the job without trays and is therefore liked
where the water is very corrosive; also where the deaerator is not stationary as
in maritime practice.
In the atomizing deaerator the water is first heated to within 1 or 2 degrees
of the temperature of the steam by spraying the water into a steam atmosphere
from nozzles spring-loaded to give a uniformly high spraying velocity at all
rates of flow. The heated water then encounters a high-velocity steam jet and is
atomized into a mist, presenting greatly increased total surface to contact with
the steam. The gases pass almost instantaneously into the enveloping steam

Exhaust

Levels.
LNormol operating
2.LOW woter okirm
5Emergencj^

BOILER FEED
PUMP

Fig.13-14 Connections to a deaerating heater. Suitable for a small boiler plant.

Condensate and feedwater tank need to be mounted above the pumps so as to provide
the required positive suction head.

atmosphere from the small droplets into which the water is subdivided by the
atomizing action. The deaerated water drops into a storage or pump supply
chamber, while the steam, little of which has been condensed in the atomizing
process, flows to meet the incoming water in a spray heater chamber, from which
the liberated gases and a small amount of steam are withdrawn to a vent con¬
denser.
At the same time that oxygen is eliminated, free CO2 can also be removed
completely if the water is slightly alkaline. By first treating the water with
enough acid to decompose the bicarbonate, and then deaerating, bicarbonate
CO2 can likewise be removed.
A good deaerator will reduce oxygen to 0.03 cc per liter or less. As the
Winkler test yields O2 in ppm, it is necessary to know that 0.7 cc per liter = 1
ppm when comparing performance with the usual form of guarantee, expressed
as cc O2 at 0® C, 760 mm.
536 THE FEEDWATER LOOP
Terminal difference in a heater or deaerator in good condition should be
less than 3® C. The steam and water thermometers on deaerators frequently
show the same temperature (within the tolerance of these instruments). The
following equation expresses the conserv^'ation of heat in contact heaters. As
these can be well insulated against heat leaks, an assumption of no heat loss is
reasonable. With nomenclature as in Fig. 13-12:

- I,) - h,o) (13-4)

also {u\ + w’8)/w’8 = 1 + (/i — h{o)/(to — ^i)4.19 (13-5)
Example 1: A steam turbine-dnven boiler feed pump is to exhaust into a
deaerating heater to heat feedwater. It is desired to determine whether the pump
exhaust will be sufficient The boiler has maximum rate of evaporation of 0Mkg
per hr, dry and saturated at7 kg/cm* gawDafa; Pump operating head, 105 m water;
turbine steam rate at 0.35 kg/cm* ga back pressure=30 kg/bhp-hr.Water from condensate
tank, 85* C. Assume pump efficiency = 0 50, and terminal difference = 0® C.

105 x 0800 1 .
Steam flow through turbine = L 00 x 80 x 75 x 0.5 J *“

Exhaust enthalpy’ will be approximately 2708 -^ * 3080 J/g at the

deaerator. " 30xl0»
Deaerator will work with steam at about 0.35 kg/om* ga and 108.0^ C saturation
temperature.^^^ -|- w,)/w, =l + (2680 - 466)/(108.9 - 86)4.1»= 22.9

Since -I- = 0800 kg per hr, liig needed is 0800/22.9 = 207 kg steam per
hr. The turbine exhaust has been estimated to bel58.0kgper hr; hence in the absence
of any other auxiliary exhaust steam live steam from the boiler should be sent through
a reducing valve to supply the difference.

13-8 Surface Heaters. The surface heaters are divided into steam tube
and water tube types. Steam tubes are ordinarily used in evaporators, but most
extraction heaters are of the water tube type. These heaters can also be divided
into straight tube and bent tube (U tubes and steam coils), and into single- or
multi-pass. The surface heater is used when water is to be heated under pres¬
sure without direct contact with the steam. Problems associated with good
longitudinal distribution of steam and removal of noncondensable gases are as
important as in condenser practice. Noncondensable gases are vented from the
heater to the condenser through a small vent line. Sometimes, to avoid all
possible load on the air removal apparatus, the heaters supplied with steam
above atmospheric pressure are vented to atmosphere through a small vent
condenser.
The ordinary closed heater is housed in a cylindrical shell of steel, well in¬
sulated on the outside. Steel or copper alloy tubes are rolled into tube sheets,
one of which is fixed, the other floating. The floating tube sheet allows free
expansion and contraction and permits the use of interchangeable tubes. The
vertical type is also used. It offers savings in required floor space, but takes
more headroom. The rating of a closed heater should be expressed in terms of
m* surface and rate of heat transfer; or as the quantity of water heated per hr
from initial to final temperature. Heat transfer in the water tube heater may
 SURFACE HEATERS 537

.4 *'

ElUott Co
Fi<^ 13-15 Turbine room of a central station showing teed water storage and deaerat¬
ing heater

A, Multi-paM extntotion heater B» Condensate cirauier C, Condensate drcuner by-pass valve D» Bleeder
we check valve hydraulic control Boat chamber E, Reli^ \a've F, Pressure gage. G, air vent line
H, Condensate drainer vent line I, Feed water by pass line J, Cuscade dnp line from next higher pressure
heater. K, Two-pen recording thermometer. L, Steam inlet. M, Water inlet. N, Water outlet.

Fig. 13-16 Connections to the extraction heater.

 638 THE FEEDWATER LOOP
be treated in the same manner as for the surface condenser, which it resembles
in construction and principle.
The mathematical treatment of this category of heat transfer was developed
in Sec 9-10. When using the condenser equation for heaters, f/, is numerically
much higher than in condenser practice. Being of the order of29295>*the scale
factor will be less significant than the velocity factor in heater design.
Water velocities employed vary from 0.914 to 3.048m/scc, having about the sa¬
me range as in condenser practice.Hereagain economics may be the deciding factor
in selection of coefficient of heat transfer. The higher the water velocity, the
better the heat transfer and the cheaper the heater. At the same time, however,
the friction losses and costs of moving the water through the tubes will in¬
crease. The proper compromise is that which will make the annual heater cost,
including amortization and water pumping costs, a minimum.

1» 1M 23S 370 3tS 310 iOS «S0

ACTUAL STEAM TO TURBINE >
1000 kg/lir 1000 kg/hr

Frtme Movers uommitteet N.E.LA.

Fig. 13-17 Performance of surface-type extraction heater for variable load conditions.
(94,000-kw turbine, 5th-stage heater )

Terminal differences in these heaters are of the order of 3®-6® C for design
purposes. However, actual tests may show more because it is possible that
operating conditions (proper venting, cleanliness, etc.) are not all that could
be expected.
Steam-tube heaters cannot be so readily analyzed as the water-tube type,
due to the uncontrolled circulation of water. The mean temperature difference
of steam-tube heaters is generally of higher order than for water tubes, rang¬
ing from 28® to 84® G. Heat transfer rates vary between 975 to 1955 kcal/m^-lw-
deg C if the tubes are well drained.
The variable load performance of a regenerative heater is shown in Fig.
13-17. As turbine load increases, all quantities tend to increase more or less
linearly. This applies to extraction flow, water velocity, temperatures, and
coefficient of heat transfer. This is a high-pressure heater, i.e., on the boiler side
of the feed pump. Economic studies indicate slightly lower water velocities,
and coefficients of heat transfer should be employed in low-pressure heaters.
Some of the details of installation of closed heaters are shown in accompany¬
ing illustrations. The several heaters of a large regenerative cycle are usually
mounted on a floor below the turbine operating floor. Extraction steam nozzles

♦ koal/mS-hr-d^ O
EVAPORATORS 569
open downward from the turbine casing; hence nobe of the auxiliary piping
needs appear in the turbine room itself. However, there is plenty of it in the
space below. Equipment and piping layouts for that region require careful ex¬
ecution in detail.

Example 1: It is desired to verify proposed specifications of an extraction heater

for heating ntWOOkg water per hr from4i.4^Cto 85.e*CfcUsing extracted steam at 87.8*
C saturation temperature. Proposed design is4-pa88, withS04l8mm Xi.87 mm gauge
Admiralty tubes, 8,66 mlong. Desired water velocity about t.t m/aec
Let the principal dimensions be calculated from Eq 9-19 and those in Sec 11-14. A
comparison of calculated dimensions with the proposed surface may be sufficient
verification.

Eq 9-13:# » (85.6 - 44.4)/ln .n «, Itj* c.Then heat transfer is solved with

87.8 — 86.6
the use of Eqs 9-19 to 9-22.
At the mean water temperature of 60P C, Lw is calculated to be 6005 keat/inS-br«d«g
OL f7t 08.03/0.001845 ^ 75288 koal/ifi5.hr-deg C. Amuming » 20205,
lao 7 . a X 0.0101 . 0.0101 1
^ ’ Ay L 7*2»8(4».«1»1 + O.OU0) SSIW Vo.<S!H J
After reduction to » S1W02 — 6,618*9/A.; with the help of Fig. 9-12,
g/Ay 444.84 kMl/ml-hr
« » 118400(86.62 — 44.40) — 4*69008 keel/hr.
Required siuface A. °° 4668008/44464 104.82 m>-
With Eq 11-20,119400 8600 X lOOOn x 0.0002166 x 1.2
Then n 120 tubes per pass, or 480 for 4-pass heater.
With Eq 11-19, 104.82.0.06869 k «014 L - 9.66 m.
Comparison of these calculations with the proposal indicates that the heater should
do the job, provided it is kept clean and vented.

13-9 Evaporators. Evaporation treatment of feedwater has the advantage

of requiring but little boiler blowdown of deconcentrating character and the
disadvantage of using steam to supply the evaporative heat. The disadvantage
is not marked in the regenerative cycle because the evaporator is readily
adapted to the feed heating system, with the heat in the steam used by it be¬
ing put into the feedwater stream. How this is done is shown in Fig. 8-13.
Essentially, the function of the evaporator is to produce, from raw water, vapor
that can be condensed to distilled water for boiler feed make-up. Evaporators
were first used in marine practice as a matter of necessity, but their use has
extended to land practice, especially the central station.
An evaporator system may be single effect, in which the steam is produced
from one-evaporatoc, or multiple effect, in which the steam is produced from
several evaporators in series. In a multiple-effect system the vapor from one
evaporator becomes the heating steam in the succeeding. Unusual conditions
met in industrial or steam heating plants may require so large a fraction of
make-up as to warrant double, triple, or quadruple effect evaporators. The
central generating station ordinarily employs a single effect. The ratio (vapor
produced) / (steam used) is about 0.8 for the single effect, 1.5 for the double
effect, and 2.5 for the triple effect system. Evaporator feed is usually preheated
to increase evaporator capacity.
540 THE FEEDWATER LOOP
Evaporators are classed as film, flash, or submerged tube types. The first
and last are steam-tube types; in the former the raw water trickles over the
hot tubes; in the latter the tubes are entirely surrounded by the water being
evaporated. The flash type produces steam by dropping the pressure on water
at the saturation temperature. The excess heat flashes part of the water into
steam, then the remainder is drawn off, reheated, and again flashed. The suc¬
ceeding remarks about evaporators refer to the more common submerged tube
type.
neetrte

Fig. 13-18 Connection of extraction heater to turbine nozzle.

In small or moderate capacities the evaporator consists of a number of

helical coils on vertical axes, enclosed in a steel or cast-iron shell, with provision
for control of water level, and ample vapor storage. For moderate and large
capacity the transfer surface is ordinarily a bank of horizontal tubes, slightly
inclined for condensate drainage. The surface requirements are a function of 6 ,
which, in turn, is controlled by the pressure of the vapor.
Since the evaporator is applied to the power plant for the purpose of pro¬
ducing distilled water for boiler feeding, thereby preventing scale in the boiler,
it is essential that the scale be deposited in the evaporator. Therefore it is im¬
portant that the disposition of scale be given attention in evaporator design.
EVAPORATORS 541
A serious difficulty encountered in evaporator operation is the steady loss of
evaporating capacity as scale accumulates on the water side. To maintain heat
.transfer and reduce evaporator outage, some makes have been designed to be
self-cleaning, either shedding the scale by tube flexure or scraping it off. The
decrease of heat transfer due to scale accumulation has been shown to be*
l/f72 =:A + Bt (13-6)
in which A and B are constants and r is the time elapsed since the last cleaning.
If, for an evaporator already installed, U is determined at two different times
since cleaning, the constants can be evaluated and, knowing the cost of cleaning
the evaporator, an economic study can be made to determine the optimum
period of time between evaporator cleanings.

Fig. 13-19 Evaporator paralleled with reducing valve for supply of industrial ste .m.

To serve its purpose completely, an evaporator should produce clean steam;

i.e., there should be no solids in the vapor. Like the boiler, the evaporator will
produce carryover of suspended solids in the steam when it is foaming or
priming. The evaporator should not prime if it is of the requisite capacity, but
it will foam if the concentration of soluble salts becomes too great. To main¬
tain the concentration below the foaming point it is necessary, therefore, that
the evaporator be blown down periodically. The amount of blowdown can be
calculated through the use of equations in Sec 13-4.
There can be but little similarity in the application of the evaporator to
the small generating station, the large central station, and the industrial plant.
Steam is supplied from an extraction point in the regenerative cycle, but it
must be live steam or auxiliary turbine exhaust in the simple Rankine cycle.
When considerable quantity of low-pressure steam is used in process, the evap¬
orator can be installed in parallel with a reducing valve.
Example 1: Fig. 13-20 represents a double effect evaporator and its auxiliary
piping. Preheated raw water is fed into both effects. This same quantity is recovered
as distilled water in the drains of the second effect arid the condenser. Let it be
desired to find the quantity of live steam at iskg/en^ aibiKd^i needed to secure the
evaporation of ISMftiigof distillled water. Vapor side pressure of first effect,6.tkg/omSab;
* Badger, ^'Evaporators for Boiler Feed Make-up Water,'* Trans, FSP 50-45.
ll
WATER PUMPS 543
of second effect,2.8 kg/cm* ab. Preheated feed t-emperature, 98.8® C- Evaporator
blowdown, 3% of feed.
Designate feed to tii>t ellVcl H\i; 1o .mtoiuI, A'-^umc* dry .saturated vapor
and no condensite d(*pr(*s>i()ii. The following heal and weight balance.^ are written
for an input of 1kg live steam;
1. I0»(292l - 842) 103(2759 - 413)0.97W’'w» -f 10»(685 -- 413)0,
2. 10«(2769 - e85)0.97Trv,i « 10«(2721 - 413)0.071^^, + 10®(549 ~ 413)0.031^^1..
3. Evaporator feed jierkglive .'.^team = ll’xv, + Ibw-
A solution of these eejuations is 11 \v, == 0.911 kg, IPw. = O.SlHkg.
13608
Live steam pei 13608 kg water - Qg,, ^ q = 7*^™ kg.

Distilled water pioduced with 7870 kg st(‘am = 13608 x 0.97 + 7870 « 21070 kg-
13-10 Water Pumps. The wat(‘r looj) i.'- aelivated by pumps Tlie regener¬
ative cycle with its exlrartiuii iKatei'., deaeiator. and other auxiliario presents
numerous recpiin'UK'ut^ lor moving water Irom one placa* to another, well
as of increasing its pre^ssure. The futiction oj a putHp /.s* /o add /a the pre^nure
existing on a liquid an inert nient sufficient to the required neri ire. This s< rviee
may be the iirodiiction oi a \elucity or the uvia’eoming o) Irietion or (‘xternal
pressure. Water pumps are ust'd tor condensing v\atei. (‘omhaisate, boiler feed,

F(»od kiiichtnery Corf).

Fig. 13-21 Two-stage centrifugal innn]) mounted on eonimon base with single-stage
steam turbine.

heater drain, booster flows, .*<1111^) drain, and other services. The principal types
in pow'er plant service are ieei]>rocating, rotary, and centrifugal pumps. Jet
pumps as represented by the boiler injector are comparatively unimportant in
this field, though sometimes used as emergency feed for small boilers. Rotary
pumps are all types operating on pure rotation, excluding centrifugal force as
pressure-building action. The principal example is the turbine-tyi>e pump.
Reciprocating and rotary pumps are used to some extent in smaller fiower
plants and heating plants; centrifugals, in small and large plants. Reciprocat¬
ing pumps are capable of the very highest jiressures, but are too bulky for
 644 THE FEEDWATER LOOP
moving large volumes of water. By multi-staging, the centrifugal pumps may
be caused to produce pressures suitable for feeding modem power boilers, in¬
cluding those operating at up to 105 kg/cm* steam pressure. However, their field
of maximum discharge capacity occurs at lower pressures.
At very small flows, but for boiler pressures, the centrifugal design is in¬
ferior to others.
Essential data for the selection of anypumpincludetb»headminetrea^capaci^^
inliter^min and properties of the Zigrwid such as viscosity;temperature,corrosive¬
ness, grittiness. Secondary data concerning the pump equipment are speed of
rotation, power required, and first cost. Because of the importance of the pumps
to successful operation of plants, and of the small portion of total cost repre¬
sented in pump equipment, the cost of pumps should not be made the most
important factor in their selection.
The general requirements of a water pump installation are the provision
of foundation, accurate mounting of the pump and driver on it, connection of
suction and discharge pipes of adequate size without stressing the pump casing,
and provision of the desired form of discharge control.
The foundations of high-speed rotary and centrifugal pumps should be
solid and rigid—^indicating concrete as the material. For foundation construc¬
tion and installation of the bedplate,
see Sec 4-7. These machines are sold
mounted on a cast-iron or welded steel
uu bedplate of suflicient length to accom¬
TP- toper gouflo
Offset misoligmont Angular misalignment Carrect alignment
modate the driver. Because of the
direct connection to the motor or tur¬
Fig. 13-22 Alignment of coupling. bine, the pump manufacturer is usu¬
ally requested to buy the driver,
mount it on the base, and be responsible for the performance of the entire unit.
However, the alignment of the coupling between driver and pump should be
verified after installation; also after pump or motor overhaul this alignment
needs re-establishment. The adjustment is made with shims, slotted feet, and
the like, while alignment checks are made as suggested in Fig. 13-22. When hot
fluids are involved (turbine drive and/or hot water pump), the final alignment
should be made with the machine hot.
Designers usually arrange to bring water to a pump in a suction line that
is one pipe size larger than the pump opening. The change in pipe size is made
with reducer fitting, not a pipe bushing. The same can be done with discharge
piping, but is not as beneficial there. Suction piping is of particular importance
as it is the most frequent source of trouble, apart from misalignment. Particu¬
lar care should be taken to prevent air leakage into the suction line if the
suction head is negative, and to avoid piping alignment that leaves air pock¬
ets. Whether suction is positive or negative, all pumps should be piped writh
a check valve and gate valve in the discharge line, and a gate valve in the suc¬
tion. The latter is for the purpose of isolating the pump for inspection or repair.
A pressure gauge ought to be connected to the discharge close to the pump; one
on the suction is also desirable.
T^ere the pump operates normally with negative suction head, as do con¬
densing water pmnps, some means of initial priming may be necessary. Vacuum
WATER PUMPS 545
pumps and jet ejectors are extensively used to priiiie large centrifugal pumps.
In some cases a valved by-pass around the discharge check valve will suffice
to prime a pump with water from a discharge header, if there is a foot valve
or check valve on the suction.
The layout of pumps in design drawings requires mainly dimensional in¬
formation on the bedplate and the suction and discharge openings. Points of
connection of bedplate drains, priming openings, casing drains, etc., may be
indicated, but not necessarily dimensioned, for the connecting piping is small
size and can be properly fitted in the field. Fig. 13-23 illustrates the type of
pump manufacturer’s information needed by the power plant designer in laying
out a pump installation.

Fig. 13-23 Typical dimensions required for pump layout. (All dimmrioiis In mm upr
leas otherwise speoi/led) Note: Pump illustrated is two-stage centrifugal suitable for
small boiler feed pump. Capacity about 190 l/miii at 107 m, 3600 rpm.

The purpose of a pump is to move a quantity of water against a pressure.

Essential to any quantitative discussion of performance, capacity, etc., are
the following teims and definitions.

Q = Pump discharge, 1/min (1000 litres or dm* 1 m*)

H = Head, ra of water of specified density. At ordinary temperatures, \ kg

= 1 litre; 1000 kg = 1 m*; and 1 kg/cm* « 10 m; however these con¬
stants should not be used in hot water calculations unless the density
error can be countenanced.

dw = Water density, kg/m*. Reciprocal of Vt from steam tables.

Static head is the height (usually in metre) of the surface of the water above
the gauge point.
Pressure head is the static head plus gauge pressure on the water surface
plus friction head.
Velocity head is the head required to produce a flow of the water.
Dynamic head i& the pressure head plus the velocity head. Except for water
velocities considerably above average, or for large volumes handled at low
heads, the velocity head can be neglected.
Total {sliSSm*] Head is the reading of a true gauge at the pump {SSSSS**} con¬
verted to metres of liquidand referred to datum, plus the velocity head at the point
546 THE FEEDWATER LOOP
of gauge attachment. If the suction head is less than atmospheric it is negative.
This is because the existence of a suction lift causes the gauge to read vacuum,
i.e., a negative gauge pressure.
Pump operating head is the algebraic dif¬
ference of discharge and suction heads. This
can also be called total dynamic head and
abbreviated TDH.
Net positive suction head, abbreviated
NPSH, is defined as the dynamic pressure of
the liquid at pump suction,kg/on* ab, less
the saturation pressure corresponding to the
temperature at the same point, converted to
metres head of the liquid.
Speafic speed of a centrifugal pump im¬
peller is the rotative speed at which a geo¬
metrically similar impeller would run if it
were of such size as to raise 75 kg of water
per second against one metre head.
0.0149ATVQ
\ (13-7)
A ih pump -'peed, rpm, and II is tht‘
TDH in metres per stage^Q is the discharge 1/m,
Fk.. Ja-24 Pump workiiiji: uiulci Specific speed ls iiscdul in defining the suc¬
suction lift <iTifl atmoqihenc dis¬ tion limitation of impellers of different de¬
charge signs.
Qd H
Wakr horsepower, whp - 45000OO

Pump (fficiency. rjj, ~ ^^hp/l hp (13-9)

Example 1: The drive requirementof the pumping installation diagrammed in
Fig 13-24 are <uiiK/(d Data Velocity in Miction hne,2in/8eo; discharge hno,8m/8©o
Fiiftion estiniitt in auction line, including loot \ d\e and entrance loss,0,65m, in
(h^dnrge line,0.76 m. Flow is to be 3800 1/m cold water, = 0 65. Using centcr-line
ot pump a" datum, total head calculated \^Miming gauge attachments 800mm bel¬
ow datum:
Dischaige head = 4.5 + 3.0 -f- 0,75 + 3*/8llf 8*0 m
Suction head = —2,5 -f- 0,3 — 0,66 -f 2*/3jf » —2,65 m

Pump opei a ting head (difference) 8,65 m

. 8800 x 1000 x 8.^
Drive must supply jiowcr of 4560^000" x 0 66 ““ 1U2 np

13-11 Reciprocating Pumps. The reciprocating pumps are best in the

field of high-pressure, moderate-capacity pum])ing. Feeding water into the
smaller ))ower boilers *is service in this category. The principal types used are
the direct-acting steam pump and the triplex power ])innp.
Direct-acting steam pump. As the sectional view in Fig. 13-*26 shows, steam
RECIPROCATING PUMPS 547
acts against its piston, thereby transmitting force to a water piston and pres¬
surizing the water. In boiler feed service the water goes back to the same
region from which the driving steam was drawn; hence the steam piston must
be larger than the water piston to overcome the effect of hydrostatic head,
losses, and inertia. If p is the boiler gauge pressure, pi, the static plus friction
pressure on the water end, pt the friction loss in the steam line, and Dw, D,
the diameters of water and steam pistons:

D ^
■^w If P — Pf

The foregoing cannot be an equality, for unless the left-hand term exceeds the
right there is no surplus force for accelerating the water at the beginning of
each pumping stroke. In order to insure positive pump action, Dg/Dw is made
about 1.6 for ordinary boiler feed service and 2.5 for low-pressure feeds.

hammanMl iron
padung rmgt
Worthington Pump and Machinery Carp,

Fig. 13-25 Vertical section through one side of a duplex direct-acting pump.

On the steam end it will be observed that the power is derived, as in an

engine, from the fluid pressure of steam against a piston. Steam action is con¬
trolled by a simple slide valve having no steam or exhaust lap. Consequently,
the pump uses steam nonexpansively (rectangular indicator card) and 8tc.am
rates are comparatively very high. This, however, is necessary, for the pump
would stall part way through an expansion stroke unless the D,/Dw ratio were
much larger than manufacturers consider practicable.
These pumps are always double-acting and are manufactured as simplex or
duplex units. Many consider the valve gear of the duplex pumps (wherein the
rod motion in one cylinder valves the steam cylinder on the other) somewhat
548 THE FEEDWATER LOOP
more reliable and easily set than on the simplex unit where the rod motion of
a cylinder must throw its own steam or pilot valve.
There are several sliding parts requiring lubrication. Steam pistons are
lubricated by injecting steam cylinder oil into the steam. Valve gear is lubri¬
cated by hand oiling, by grease cups, and by force-feed oilers. There are pack¬
ing glands on the steam and water piston rods and on the steam valve rods.
The steam piston is sealed with hammered iron piston rings; the water piston,
with a set of composition packing rings appropriate for the service. The water
must provide whatever lubrication that end gets. All this adds up to more in¬
spection and maintenance required than for centrifugal pumps.
Against the disadvantages of oily exhaust steam, low thermal efficiency, and
maintenance can be set the following advantages: (1) simplicity of design and

FEEDWATER

Fig. 13-26 Control of direct-acting steam pump m boiler feed service.

‘^understandability^^ of pumping action by the average mechanic. (2) Flexi¬

bility in operation. (3) No need for separate driving motor or turbine. (4) Fairly
quiet and can set on simple light foundation. (5) Not likely to become mis¬
aligned. (6) Comparatively low initial cost up to capacity of several hundred
liters per minute.
Being a complete pumping unit the direct-acting pump is simple to install.
A hydrostatic oiler may be needed on the steam line. There should be a good-
quality globe valve for stopping and starting. The supply line should include
a steam separator. Steam cylinders and throttle valve should have drain lines
to sewer. Any inverted bends in the exhaust line also need draining. A simple
and convenient discharge* control is accomplished by using a pump governor
valve This is an automatic regulating valve which will throttle the steam to
the pump when the feedwater line pressure increases due to closing. In effect,
it is flow control through speed control.
Except for the effect of slip, thet is, leakage past piston packing or through
discharge valves of the water pump, the delivery of a slow-speed reciprocating
pump equals the piston displacement. Any pump can be operated at speeds high
enough to impair the volumetric efficiency. Commercial speeds for rated capacity
are, therefore, set low enougl> to give the water time enough to fill the cylinder
on each suction stroke. Commercial pump speeds are approximately:
RECIPROCATING PUMPS 549
rp»1.38/X m/minof piston speed (13-11)
where L » stroke in mm.
This is for use with cold water and should be reduced by a multiplier factor
if hot water is pumped This factor is 0.85 for 32.2®Cwater, 0.71 at65.5®Cand
0.55 at 204,4®C. It can be considered to vary linearly with temperature. Also
the NPSH should be suflSciently high. Because of the pulsating character of
reciprocating pump suction, close control of NPSH is not as practical as for
steady flow pumps. In its stead we have the recommended normal suction lifts
of Fig. 13-27.
——

^5*

_1
- r-^
Bas»d on Slandords of
Hydraulic ln»lilute. These —^
suction conditions ore *normal\
i.e. guoronteeable. With favorable
conditions,1«2m more suction
might be developable. Dotted .
portion not covered by HydroulicX
Institute Siondords.

5"
submergence lift
pump suction m water static heod

Fig. 13-27 Commercial suction practices for duplex direct-acting pumps.

The discharge capacity of positive displacement equipment is the piston

displacement multiplied by volumetric efficiency, where the latter includes the
effects of slip and piston rod displacement.

Q= l/min (13-12)
where rix = Volumetric efficiency. Should be more than 0.90 for pumps in
good order, not overspeeded.
PI) = Piston displacement, cm*. per min per pump.
Manufacturing practice is to build pumps of cylinder diameters which ad¬
vance by 6.4 mm increments from 50.8 mm to 139.7 mm and by 25.4
mm increments to 406.4 mm. Strokes increase by 30.8 mm increments
from 50.8 mm to 406.4 mm, plus 63.5 mm and 88.9 mm sizes. Only those combi
nations offering suitable ratios arc usable for, boiler feeding.
Example 1 : The discharge oapsoifcy of a duplex ataam boiler feed pump 228.6 x
133.35 x 254 mm, oxwrating at normal speed, will be found. Water temperature^ 93.3*0.
These dimensions represent D,x Dw xZ», mm. From £q 13-11, oorreoted for 93.8*0, Vp
ai0.66xl.38V‘2dl»l^*3 m/min. Fb of duplex pumpa6 2xl3.l43iv(14.5x 1001/4 ■■449900
om^/min. Assuming ^v»0.00, 0«0.23x0.90x449900/331»405 l/min. Spasd in strokes
per min per cylinder ■> jj*^ xl000
650 THE FEEDWATER LOOP
The d-a steam pump has a high steam rate because it does not expand the
steam. On account of initial condensation and clearance the steam consumed
exceeds the product of piston displacement and density of the throttle steam.
Steam rates vary from 45 to 135 kg per hp hr. i/t is so low that thermal
performance is gauged by a term called 'pinnp duty, defined as the kgm work
done in the water cylinders per million Joules chargeable to the steam end. The
chargeable Joule/gram steam is throttle enthalpy less heat of the liquid at ex¬
haust pressure. Pump duty is also expressed as kgm water work per 1000 kg
steam.
Esmtiiple 2 : A test was performed on a 152 mm x 102 mm x 152 mm d-a duplex pump
for the purpose of determining its operating condition and performance. In 30 min of
steady operation the pump used 45.2 kg steam (determined by condensing exhaust at
O.Ol kg/cm» ga) and made 1254 strokes per cylinder. Atmospheric pressure, 0.98 kg/om* ab.
Throttle steam, 3.2 kg/cm* ga at 0.974 quality. Mean effective pressures : left, 2.7 kg/om>;
right, 2.6 kg/cm*. Average strokes : left, 140 mm; right, 120.7 mm. Piston rod, 31.8 mm.
Water pumped, 2164 kg; temperature, 17.2®C. Discharge head. 52.3 m ; suction, 0.65 m
vacuum.
2164(52.3-1
Pump duty = =0.00112 kgm per 10* Joules
1000x45.2(2684

/ 140 120.7 ^ 7t(i5.2«-~3.18») 1354 I

HE il»p = ^2.7xjooo- ■+2.6X jq5q
^ 2 ^30x60x75"" ^•‘WWihp

/ 140 120.7 \ x(16.2*~3.18*) 1254 ^ I

WE ihp =/ 2.7X 1000 j X— =0,558
4 2 30x60x75
- 0.558 ihp

Pump power ~ sum of head end and water end — 1.141 ihp
Water power = 2164 (52.3 — (-0.65))/(30 X 60 X 75) = 0,849 whp.
Combined hydraulic-mechanical efficiency = 0.849/ 1.141 = 0.74.
Steam rate = 45.2 X 60/(30 X 1 141) = 79.3 kg per ihp hr
Let V = Volumetric displacement of water piston per two strokes.
V = 4067.3 cm*. (exact), or 4244 cm* (neglecting rod effect).

From Eq (13-12), i/v =

JPjy OtRA
{J!*! *} X 1264/(30 x 2) = 70.81/inin
Hence tj. = 231 X IQ-l/Islio} = 0 S27 or 0.792, depending on whether the eflSciency
18 not or is to include the loss of delivery due to piston rod displacement.
Triplex Power Pumps, The reciprocating pump is also used in the form of a
three-cylinder, single-acting, motor-driven pump when discharge pressure is
moderately high to high, and where the inefficiency of steam pumps wasting
their exhaust or of small-sized centrifugal pumps is to be avoided. Triplex
pumps are three-cylindered in order to produce overlapping deliveries and
minimize pulsation. They are single-acting to simplify the mechanical linkage
connecting crankshaft to piston and to avoid rod stuffing boxes. These pumps
are quite efficient, but control is a problem since either the driver must be
operated at variable speed or the portion of normal capacity not needed at
part-load by-passed back to suction by relief or regulating valves. If an excess
pressure regulating valve is placed in a by-pass, and operated by the pressure
TURBINE PUMPS ,551
drop across the feedwater regulator, it must be of the spring-closing instead
of spring-opening type. With by-pass control the part-load efficiency will vary
approximately proportional to the net discharge.
Small boilers arc supplied by motor-driven triplex pumps that can be put
under float switch control and so operate on an on-off basis. No feedwater
regulator valve is used and boiler water level fluctuates between high- and low-
water limits set by the float switch.
Pump speeds must be held down to those which will not result in impaired
volumetric efficiency. This is represented by the empirical formula,*

N = 907L-1/* rpm (13-13)

where L = stroke in mm.
The schematic illustration of a triplex pump shows speed reduction by
gearing the motor shaft to the crank¬
shaft. This is also made with V-belt
drive. The illustrated pump is plunger
type, hut power pumps are also built
with trunk pistons having pins to take
the connecting rod.
13-12 Turbine Pumps. This is the
name given to a class of rotary, non-
positive displacement pumps having
some importance in the small boiler
field. It is generally seen as a single-
stage pump, for with a singk' vaned
wheel it can produce a pressure incre¬
ment of 7 kg/cm* or more. This pump
employs impact, diffusion, and some
centrifugal force, but is not a true cen¬
trifugal pump. Water enters at the
periphery and is carried around to the outlet by the vanes, being repeatedly
forced out into the free channel (see Fig. 13-29) only to return for re-engage-
ment with the vanes. Thus the pressure is built up regenera-
tively from inlet to outlet. Considerable churning exists and
therefore the efficiency is low (20%, rising to 40 or 50% in
the larger capacities). However, small centrifugal pumps
accomplish about the same waste of energy without being as
productive of pressure per stage.
The characteristics of the turbine pump meet well the
boiler feed requirements of the small boiler operating with
Fig. 13-29 Mul- Steam pressures up to 14 kg/cm* This includes 3.5-10.51%/cm*
ti-vane or “tur¬ package oil burner units (described in Chapter 10). Largo
bine” type pump. numbers of these are in service, and many new ones are sold
annually. Most of them are fed by “pump sets’^ which con¬
sists of a motor-driven turbine pump, a feed tank with float valves for auto¬
matic make-up, and a float switch to attach to the boiler at the normal water

* For cold water. Make temperature correction for hot water.

662 THE FEEDWATER LOOP
level. As steam is generated and the water level drops to a set level, the float
closes switch contacts and starts the motor. After the pump has raised the boiler
level to an upper limit, the float opens the switch and stops the motor. This
system has the virtues of simplicity
and low cost, and since the combus¬
tion control is also generally of the
on-off form, another intermittently
operating electrical auxiliary is no
disadvantage. Up to 1 hp of electrical
power the float switch may handle
the motor directly, but larger motors
need a relay between float contacts
and motor power circuit.
13-13 Centrifugal Pumps. The
basic parts of this pump are a rotat¬
ing impeller and a casing to sur¬
round it and direct water flow. Water
let into the center or ^^eye’^ of the im¬
peller is quickly rotated and dis¬
charged from the periphery of the
impeller with velocity and pressure
more than it had at the eye. This
naturally creates a force on the im¬
peller opposing rotation, accounting
for the toique which must be contin¬
Fairbanks Morse Co uously supplied to the impeller shaft
Fig. 13-30 Boiler feed unit, with turbine by its driver. Without entering into
pump. the theory of the centrifugal pump, it
may be stated that a workable theory
has been derived but that this theory, in its simple form, is incomplete due to
certain indeterminate flow conditions in the impeller. The centrifugal pump is
a velocity machine; that is: its pumping action requires, first, the production
of a water velocity; second, the conversion of velocity to pressure head. The
velocity is given by the rotating impeller, the conversion accomplished by
diffusing guide vanes or in a volute casing surrounding the impeller. The dif¬
fuser pump is the more efficient, especially on high heads, but is the more costly.
It is quite common to use the multi-stage volute type where the heads are be¬
low 35 kg/cmiWith few exceptions, single-stage pumps are of the volute type.
Other bases of classification are: (1) the direction of the pump shaft, hori¬
zontal or vertical; (2) method of constructing the casing for access; horizontally
split, or barrel type with removable ends and stage separators; (3) extended or
close-coupled. When pump impeller is mounted on the motor shah and casing
is bolted to the motor end-bell, the pump is said to be ^'close-coupled.'^ There
arc, of course, numerous differences in construction detail which the engineer
must examine in any critical appraisal of competitive offerings. Some of these
arc the method of balancing the end thrust which occurs with single suction
impellers, how mechanical and hydraulic wear can be offset by replacement of
wearing parts such as shaft sleeves and wearing rings, the design of packing
CENTRirUGAL PUMPS 568
glands, arrangements for cooling, draining, etc. Casings are cast iron where the
pressure permits; otherwise they are carbon steel. It is standard practice to
fit pumps with forged steel shafts and cast bronze impellers, but for high-tem¬
perature service, stainless steel is more corrosion-resistant.

Inacrsoll-Rand Co.

Fig. 13-31 Two-stage horizontal centrifugal pump Design suitable for boiler feed
service up to 5675 1/m at 335 m TDK and 3500 rpm.

I/ubrication is simple on most centrifugal pumps, there being only the two
shaft bearings to require it. Ring oiling from reservoirs cast in the bearing
pedestal and grease-lubricated ball bearings are usual. Cooling may be needed
lor high-temperature feed pump bearings.^ In the high-pressure regenerative
cycle, power station boiler feed requirements constitute severe service for a
pump. End thrusts are high and oil-lubricated thrust bearings such as the
554 THE FEEDWATER LOOP
Kingsbury are used. Such installations may have a more complex lubricating
system, including oil eiiculation and cooling
Capacity Ls a function of iinpellei width, diameter, and sjieed. Head is
determined by impeller diameter and speed, and efficiency results from proper
integration of impeller and casing shapes. The variable discharge performance
of a pump is usually tested with cold water at constant shaft speed, resulting
in a set of curves similar in nature to those jireserited for the centrifugal fan
in Fig. 12-38. Where it is desired to show' jierfonnancc data for a line of geo¬
metrically similar punijis, with different impeller diameters reiiresented, a plot
like Fig. 13-32 is often used. This is a head-diNcliarge graph for the several
For Single - Suction Pumps with Shaft Through Eye of impeller
15 20 25 30 35 40 50 60

Fig 13-33 Limiting ^jiecific bjieed.s for centrifugal iiiunp suction lifts (Ha^'edon
, 29,4 and sea level )

impellers, all at the same speed with pow’er and efficiently noted at oaeh head-
discharge coordinate. Then all lines of the same efficiency and the same bhp
are connected to display these characteristics as contours For other speeds
similar graph sheets would be necessary; however, for limited variation the
law of similitude may be invoked. Head is proportional to capacity to N,
and power to (A")®.
Specific speed is an indication of impeller type, the same as it indicates the
runner type of hydraulic turbines. Designers employ it to meet different con¬
ditions of H, Q, and N\ jiower plant engineers should understand how to use
it to check an installation for permissible suction heads to avoid cavitation.
A pump of low specific speed will operate safely with greater suction lift than
one of higher speed. With low suction lifts (high NPSH) a pump of higher
CENTRIFUGAL PUMPS 565
specific speed can be used—which often is a cheaper pump. Fig. 13-33 repre¬
sents operating experience on the upper limit oi N, for which centrifugal
pumps can be considered to be safe from cavitation difficulties, viz., noise,
vibration,and pitting. The chart basis is a sea-levelatmosphere and 29.4"^water.
Example 2: Given driver speed of 3550 rpm, the chart will be employed to
determine highest, safe, specific speed for a multi-stage boiler feed pump to handle
1893 I/min against a total head of 167.6 m of 121.1^0 water. Suction head available is
2.56 kg/om* ab •
The S.G. of 121,1 water is 0.942 referred to cold water. Its vapor pressure is
2.089 kg/om< ab.
Then NPSH = (2.66 -2.09)10= 4.7 ra cold water.
Chart is based on 29.4®C water, for which vapor pressure is 0.042. At sea level,
suction pressure, absolute = 1.03x10—induction lift. With 29.4®C water NPSH =
(1.03 X 10 — m-suction 1 ifb)—0.042 x 10 - 9.88—suction. Therefore< suction 1 ift on chartss9.88
-NPSH-9.88 -4.70 = 5.18 m. = 32.6 — 15.2 = 17.4 ft.
The head per stage is now determined, assuming 2 and 3 stages. For 2-8tage,
ff = 167.6 x 0.942/2 = 78.9 m cold water. Likewise, if 3-8tage, ^=52.7 m.
Enter chart with suction lift and head.
2-stage Ng = 1180; 3-stage Ng = 1520
Next solve the specific speed equation for pump speed.
2- stage JV = 1180 x 4.74 x 78.9»'«/VT893==3405rpm
3- stage = 1520 x 4.74 x 52.7»/Vlj^*=3241 rpm
The 2-stage pump uses a speed nearer the specified drive.
Capoc'ity I/min double sucfion pumps _
20 =«

.'TBSiBiiiaiairiiii
o y;s;^';^-^:s5ssssssass:s;^^:ss^ • BBiaBlliar-U.U.i.MI

■BBiiiSiiiSBSSSSSSSrSBiiiiiiif^fiiiiiiiBi^^'jj-ujj.'-^ij-’j
0*9rT-rr i iJ.U.i.BaHHHaiBKaBBIIIII’^flllllii»|R0^i2{Sjg>2{2j{jf\
tti 11 SMiaiaiHBBr:aiBiaBBisr'4iaisiiiiniR^BaBa8iiaisi!ii*.:.'

0*5 wCv"'I mS ■■ mm a^ BiBBBBUB^It^^trii iiiii BiBiBBBaaiii Biiiiiiiii

fdblx' ^sSlSlBBBiBiM'^iifMaMPvaiMBBlMlirMStlllllllllMMBBBBBailBIIIIIIIII
I I ■■■■ v;^MMaMMiaiBIB^4SIIISIIIRIIIlilMMBBBBaBliBSIIIi>*
Q.^^aBiiSiiiiSSSvSMB&aii^Baiiiiiiiiiiiiiiiil3MliB»i**<«<»iii

4000

Copocity l/min-single auction pumpa

Hydraulic IfutiMe

Fio. 13-34 Minimum positive suction head for centrifugal condensate pumps*
666 THE FEEDWATER LOOP
The chart may also be used for altitude cases by adding to the specified
suction lift, or deducting from the specified suction head, 0.1 m per lOOm alti¬
tude above sea level.
Condensate Pump, Centrifugal pumps are almost exclusively given the job
of condensate pumping. For one thing, the self-regulating character of sub¬
mergence control mentioned in the treatment of condenser auxiliaries simplifies
many condensate pumping installations, particularly those from vacuum re¬
gions.
Condensate pumps are a special class of centrifugals. They are generally
required to operate on minimums of net positive suction head, frequently when
pumping hot condensate. The flow into the pump through the suction nozzle is
whatever the NPSH, used as H in the equation Q = A yJ2gti, will provide.
Since submergence of inlet is a prominent factor in NPSH it acts to regulate
discharge.
It follows, of course, that installation mistakes are easily and frequently
committed when the significance of suction head is not understood. The recom¬
mendations shown in Fig. 13-34 are based on wide experience in this field. It
will be noted that the inflow to the pump at a given NPSH varies as the square
of the pump speed.
The single-stage, double-suction pump has the widest range of application
and can serve for all condensate drainage except those cases with high discharge
pressures. Then 2-stage axially balanced pumps are needed.
Another application of the centrifugal
condensate pump is the integrated conden¬
sation pumping set, illustrated in Fig.
13-35. These sets are employed to receive
assorted flows of condensate from traps,
heating systems, etc., and deliver the con¬
densate to the feedwater tank. On a com¬
mon base will be mounted a motor-driven
centrifugal pump, a receiver tank of steel
or cast iron, and the necessary valves,
strainers, controls. Usually the control is
"Nash Engineering Co. a simple, float-operated switch that starts
Fig. 13-35 Condensation unit. Con¬ the motor when the tank is nearly full and
sists of cast-iron receiver tank, tank- stops it when the tank has been nearly
mounted centrifugal pump, and float pumped out. By venting the tank, hot trap
switch. discharges are cooled to 100®G through
flashing; hence the pump must be de¬
signed to operate on a few cm suction submergence, although pumping hot
water.
Boiler Feed Pump, The multi-stage centrifugal pump has few competitors
foHboiler feeding in the large power station. There is no auxiliary more vital
to the continued operation of the station. Let the boiler feed pump begin to
misbehave and operators spring to action, endeavoring to bring standby equip¬
ment into service and avoid a plant shutdown. That standby equipment will
be provided is axiomatic to the central station and is nearly universal practice
elsewhere. The modern, high-duty steam generator might be out of water in a
CENTRIFUGAL PUMPS 567
minute or less were the feed to fail completely. At' such times an immediate
shutdown is the only alternative to service from standby feed pumps.
The maximum capacity needed in this pump is theoretically the maximum
rate of evaporation of the boilers it feeds, but here is an auxiliary which ought
to be specified oversize. Where steam demand is fairly steady and the ratio
of boiler water capacity to evapora¬
tion rate is conservative, then a
margin of 50% over the maximum
rate of evaporation would be ample
capacity. Contrariwise, if the steam
drum water capacity is a small frac¬
tion of the evaporative rate, and
heavy load swings are expected, then
prudence dictates sizingthe pump even
more conservatively. A boiler rated
at 22700 kg per hr could and would
meet brief demands for 22700 to
36200 kg steam if there were that
demand. This would not likely
happen in an electric power station,
but could for institutional, industrial,
or central heating loads where load
diversity was temporarily missing.
There is probably nothing more un¬
nerving to experienced boiler opera¬
tors than to see the water level disap¬ Fig. 13-36 Working conditions for a boiler
pear into the bottom of the gauge feed pump.
glass with the feed valve wide open.
True, it may be only a temporary load peak and the pump will quickly gain
command of the situation, but these philosophical attitudes are conspicuous by
their absence when operators face empty water gauge glasses.
The total operating head of a boiler feed pump is calculated as the differ¬
ence between discharge and suction heads. Velocity head is negligible. The
power plant boiler feed pump will be handling hot water under a positive suc¬
tion head composed of hydraulic head to the water level in the feedwater supply
or surge tank plus gauge pressure, if any, on that water level, less friction in
pipe, fittings, strainers, and valves. Discharge head will be the sum of hydro¬
static head to the boiler water level, piping and fittings friction, boiler gauge
pressure, and pressure drop through the feedwater regulator. All should be
expressed in metres of water at the pumping temperature.
Example 3: A simple problem in boiler feed pump calculation is presented in
Fig. 13-36. Data: Pipe and fittings friction losses, 1.5min suction line and 2.6m in
delivery line;* pressure drop through regulating valve at open position, 0.14 kg/om*
water temperature, . Maximum discharge, 473 l/min.
Using Vf from tables, the density of 93.8^ water referred to the 15.6^ standard
(where lOm—1 kg/cm> ) is 1.001 / 1.03S 0.964. Datum will be pump center-line.

* Methods of computmg pipe friction will appear in the next chapter.

558 THE FEEDWATER LOOP
Discharge head » 6.1+2.64.(l44.0.14)10/0.9e4 » 155.Sm
Suction head *= 9.1—1.5 « 7.6 m

Calculated pump operating head = 147.7 m

Specified pump operating head will exceed this by whatever safety allowance the
engineer wants to throw in to meet effect of initial frictional discrepancies^ aging of
pipe line, lack of allowed suction head, variation of excess pressure control, etc.
Engineers will wish to put in from 10% to 25% allowance here. Using 15%, the
specified head here would be 170 m water at 93.3'^C.
Now, if performance curves of the pump were available, the driving power could
be computed. Assuming it was foimd tlmt at 4731/min and 170 m the pump efficiency
will be 0.75, the power required is:
478xi^xl70
bhp = ^ = 1.001^ ^^

The installation therefore calls for a 25-hp motor.

On account of increased operating head and lowered efficiency of the ma¬
chines at part load, almost as much power may be needed at half load as at full.
It is also very important in this class of pump to provide sufficient positive
suction head. A pump which operated quite satisfactorily on 110°C water feeding
a boiler at 7 kg/cm* ga when provided with 3 m suction submergencp; was en¬
tirely unable to produce flow when supplied with 82.2®C water from a sump
necessitating about 0.9 m suction lift.

Capacity m I/min

Hydraulic ItuHHdc

Fig. 13-37 Net positive suction head. Centrifugal hot water pumps—sin^e suction.
These curves serve as guides in determining the net positive suction head for hot water
pumps and do not necessarily represent absolute minimum values. Compiled from data
by representative companies. The curves apply to water temperatures up to 100®C,
For temperatures above 100®C use temperature correction chart. For speeds within
db 25% of those shown correct capacity according to rpm \/jpm = Constant.
CENTRIFUGAL PUMPS 559
The data of Fig. 13-37 were compiled by the Hydi^aulic Institute from data
by representative companies. It is provided as a reliable indicator of minimum
NPSH on the boiler feed pump suction. This graph is for single suction pumps.
A set of curves for double suction pumps is similar except that, at maximum
discharge, they rise to about 70% of the NPSH's given for single suction.
Example 4: Water is heated to saturation( 1.407 kfr/om< ab, 108^9^) in adeaerator,
then delivered by gravity to a boiler feed pump rated at 284 l/min at 10.5 kg/om* ga
pressure. It is desired to determine the required suction submergence height^ H. 0.3 m
head will be allowed for pipe friction in the suction line. N — 3500' rpm.
From Fig. 13-37, NPSH = 8matioo®C. Correction for 108.9^0»b.24 m making
NPSH at 108.9®C, 3.24 m 8.0. of 108.9* water = ^ =0.953.
1.0507
_ , vapor pressure
NPSH s suction head ---
1.407x10 1.407x10
--o;^ +H-0.3- -5^553—
H = 3.54 m. Allowing some margin of assurance, the submergence will be put at 4 m.
In this example H was simply the NPSH -h friction, but only because the water was
saturated, and at the same temperature in both deaerator and pump suction.
Steam turbines and electric motors are employed to drive the centrifugal
boiler feed pump. Motors are direct-connected; so are the steam turbines,
though large, turbine-powered units will also be found with geared turbines.
This is so the turbine may operate at higher speed than the pump in the in¬
terest of better efficiency. Turbines are the mechanical drive type, usually non¬
condensing, with exhaust steam directed to feed heating or otherwise usefully
absorbed. Motors may need to have special features of high-temperature in¬
sulation of the windings and drip proof frames.
Centrifugal pumps expected to be operated in parallel should have a droop¬
ing pressure characteristic throughout the operating range and possess the
same percent AQ over the operating range.
Control of discharge is largely dependent on the drive employed. Motor
drives are essentially constant speed so that the pump is caused to operate
back and forth along its constant-speed, head vs discharge characteristic by

Fio. 13-38 Two systems of centrifugal boiler feed pump control with governed differ¬
ential pressure across feedwater regulator.
 660 THE FEEDWATER LOOP
throttling the discharge. Turbine drives are readily speed-controlled by pres¬
sure-operated throttle valves in series with the turbine speed governor. These
two systems are comi)ared in Fig. 13-38.
Speed control is premised on the shift of the H-Q characteristic by change
of pump speed. Imagine a feedwater system resistance characteristic which is
composed of a ])ressure head and a variable frictional resistance. Constant
excess pressure control is presumed to be applied to the feed regulating valve.
Then the system 11-Q relation is represented by F, whereas that for the pump
is N. The piimj) will always operate in hydraulic condition represented by the
intersection of these two curves. Thus if N in diagram a is normal rated turbine
speed set by its speed governor, then would be the pump characteristic
needed to supply rated discharge Qi. The surplus head A A' represents the al¬
lowance mentioned in Example 4 as a contingency should
OlFFERENTrAL the actual system resistance turn out to be F' instead ot
PRESS. LEADS the estimated F. At part loads Q2, Qs, the pump charac¬
teristic will be shifted to .V2, -V3 in order to achieve requi¬
site hydraulic balance. This shifting is carried out by
turbine governing wherein an excess pressure governor
takes over. The feedwater system is usually set up with
excess pressure control on the feed regulator.* A turbine
throttle valve to go in series with the speed governor
would be similar to Fig. 13-39. Pressure leads from the
high- and lo^v-pressure side of the feedwater regulator, or
alternately from feed and steam headers, are connected
to the opposite sides of a diaphragm. The low pressure is
aided by the spring, and the system comes into balance at
a differential pressure that can be adjusted by spring
Fig. 13-39 Excess compression.Differential pressures of 1.5 to 3.5 kg/cm* are
pressure regulating used with the conventional feedwater regulators. When the
valve feedwater regulator p.ai’tially closes, it tends to increase
the pressure differential. This, transmitted to the turbine
pressure governor, partially closes the steam valve and the turbine slows down
to such speed as will again adjust the excess pressure to that value preset by
the spring compression. .
Throttling control is shown in Fig. 13-386. Disregarding the normal small
speed regulation of the motor, there is only one H-Q characteristic to work with.
Now to deliver Qi, Q2, Qa, the control system must produce system resistances
Fj, F2, F3. But as the basic system resistance is F, an extra controlled resistance
must be inserted in the discharge line. This takes the form of a water line valve
automatically controlled by the same pressure differential used in system a.
AA' is the necessary throttling at rated discharge resulting from the same pre¬
cautionary pump specificationf that produced AA' in Fig. 13-38a. If F turns
out to have been a correct prediction of system resistance, the water valve pro¬
duces pressure drops BB' and CC' to regulate for flows Q2 and O3. The throt-

* Discussed in the following section.

t Refers to use of a characteristic F' in specifying the pump rather than F which is the
computed resistance.
CENTRIFUGAL PUMPS 561
tling valve to put in series with the feedwater regulator is similar in design to
Fig. 13-39, that is, a spring-opening design. /
When a pump is operated at a small fraction of its rated output, yet pre¬
serving rated speed (case ^), the efficiency is quite' low. One with characteristics
of Fig. 13-386, with exceeding 70% at rated discharge, probably would be
down to 15% at 1/lOth load These large losses will heat the water passing
through the pump by several degrees, sometimes too much. Then there is always
a possibility that the feed valve might close completely for some time, during
which the motor input at r^p = 0% would rapidly heat the water, churn it
into steam, and possibly overheat the pump to the point of mechanical seizure

Fig 13-40 Connections tor feed control with constant-speed centrifugal boiler feed-
water pump.

Pipe plug

From discharge
(int of pump

Orifice Oiameter

Fig. 13-41 Dimensions and capacity of by-pass orifices. (Courtesy Southern Power
and Industry,)
662 THE FEEDWATER LOOP
To prevent this a small orifice should always be installed to by-pass that mini¬
mum flow of water required to cool the pump from the discharge back into
the feedwater cycle. The orifice must be quite small, should be corrosion proof,
and have a strainer in front of it, or be installed so that its interior can be
rodded from time to time to verify that it is serviceable. The hydraulic loss
occasioned by this small by-pass is relatively insignificant; its protective feature
is valuable. If the orifice flow is sufficient to prevent more than 8.3®C rise in
the water temperature in the pump, it is said to be adequate protection against
flashing and overheating in the first-stage impeller.

Example 3; A1703 ]/min pump which operates against 366 m water TDH at 100^
has efficiency of 6% at 94 l/min and 15%, at 1801/min Its by-pass orifice needs are to
be determined.
Each kg of water flowmg through the pump has 366 kgm of work imparted
to it, equivalent to 3689 J. The turbulence energy absorbed as a temperatur. rise is
3589 . This amoimts to 68190 J per kg at 941/min flow and 20348 J per kg at
189 l/min The specific heat at 100%; being 4.187 J per g-deg we have 16.rC rise at 94
1/min, 4.8”C at 189 l/min. Interpolating, at 8.3'*C rise a flow of 160 l/min is indicated. From
Fig. 13-41, it is seen that the orifloe size to use is 7.1 mm.

Installation, Some factors in centrifugal pump installation have been men¬

tioned. To these might be added the following, which are specially pertinent
to the high-temperature, multi-stage centrifugal boiler feed pump. The elevnted
temperature of the water being pumped creates an expansion problem. The
large-capacity, high-pressure pump will be joined to pipe lines of considerable
rigidity, whose own expansion may, if not allowed for, twist the pump casing
causing misalignment, wear, or breakage. Hence the connecting piping should
be independently supported during installation with the final companion
flanges exactly lining up at the pump before the bolts are inserted and tightened.
Although expansion bends and loops can be built into the discharge system in
order to produce flexibility, there is a limit to this on the suction side because
of the effect on NPSH. However, a short, direct, rigidly anchored suction ap¬
proach is dangerous. Sometimes an expansion joint has to be placed in such
a line. Large, high-temperature pumps ought also to be provided with a
warm-up, manually controlled by-pass so that the pump may be in thermal
equilibrium before it is started.
Desirable instrumentation includes pressure gauges on suction and dis¬
charge, and on pressure pilot lines to regulator valves; suction and discharge
thermometers; feedwater flow meter; and an alarm system for low flow or low
pressure. Of course, the boiler water column low-water alarm will eventually
report insufficient feedwater, but precious minutes are saved by the earlier
alarm from a pump discharge warning system. Not all the foregoing are con¬
sidered necessary for every pump installation; however, the pressure gauges
should be a ‘‘must.”
Operation and Tests, The centrifugal pump is mechanically simple, yet
easy to damage if not operated correctly. Turbine-driven pumps can always be
started slowly and gradually brought up to speed. Motor-driven pumps are
either started at full speed or, with large motors, on one reduced starting speed.
FEEDWATER REGULATION S63
The higher the pressure, and the higher the water temperature, the more the
care and time to be taken in starting up. One should be very careful that no
valves are closed in the suction line, and that the pump casing is vented dii!ring
the start. Operational supervision in¬
volves watching pump instruments and
listening to the pump sound. Failure pf
the pump to operate on its normal head-
discharge characteristic, vibration, or un¬
usual noises are evidences of impending
trouble that can be detected by watchful
supervision, often far enough in advance
of enforced shutdown to enable the owner
to get the necessary parts on hand for
quick repairs.
Pump tests are made to measure
quantities necessary to plot the charac¬
teristic curves of total head, bhp, and
eflSciency against discharge. The bhp of Fig. 13-42 Factors and Ft for Eq
installed and operating pumps must come 13-14. (Courtesy Southern Power and
from analysis of the driver performance, Industry.)
whether turbine or motor. Laboratory or
factory testing of pumps is carried out with cold water. The results are corrected
to actual operating conditions with these rules:
1. Power consumption is directly proportional to specific gravity of the
water.
2. Adjust NPSH of the test to be equivalent to that available under actual
operating conditions.
3. If test speed is not operating speed, a speed correction is applied to
power.
4. Efficiency is corrected by the formula
Vpt
Corrected tjp (13-14)
Vpt + (I "" Vpt)FnPt
where F„,F, = Correction factors. Fig. 13-42.
Vpt = Test efficiency of pump.
In addition to correcting test results, this equation will be of assistance in
prediction of required power to drive a pump at some lesser speed than that for
which an efficiency is already known.
13-14 Feedwater Regulation. Air, fuel, and water are the three vari¬
ables entering into the production of steam. (Control of the supply of fuel and
air was described in the last chapter. We now come to the final element of the
feedwater loop—^the feedwater regulator. The feedwater regulator is the gov¬
ernor of the feedwater supplied to the boiler.
Soon after steam boilers came into use it was discovered that major damage
resulted if they boiled dry; also that the engine might be wrecked by water
passing over with the steam. Naturally great care was taken to prevent the
water level in the boiler from passing below or above the safe linrits.
564 THE FEEDWATER LOOP
When pressure boilers first came into use, no regard was taken of any exact
water level—the only object was to keep water in the drum. This naturally
brought about the establishment of a mean water level, and consequently when
gauge glasses were introduced, as a matter of course the glass was placed with
its middle at this mean water level. In the case of the water tube boiler, the
center-line of the drum was selected. The establishment of this mean water level
at the middle gauge was purely arbitrary; it merely represented a point midway
between the two extremes; it gave the water tender something to aim at, for
avoiding the extremes.
Under modern conditions it is necessary for feedwater to fiow into the
boiler almost as rapidly as the steam flows out; and since boilers are approach¬
ing the flash type, it is plain that the feedwater regulation should be automatic.
It cannot be done successfully by hand; a close adjustment of boiler water
level cannot be accomplished manually by operators.
It will be seen, however, that maintaining a fixed water level is not advisable.
Nowadays our high boiler ratings, high furnace temperatures, and small, water-
storage space call for a very sensitive water level adjustment; every change in
load must be met with a change in feedwater input. Otherwise, the boiler will
surely be operating at about 4% or 5% below the efficiency which can be had
with correct regulation.
Even in the small plants there is no excuse for controlling feedwater input
by hand. In our efforts to produce greater efficiency, the number of devices and
instruments for the boiler room has greatly increased, and the small plants
which show the highest eflBciency are the ones which are making the fullest use
of these improvements.
There are several makes of feedwater regulators on the market, most of
them either float or thermostatically operated. Continuous regulators have
become the standard type for large boilers. Small boilers are frequently fed by
an intermittent system based on automatic start-and-stop of the motor-driven
feedwater pump.
The principles of continuous-type regulators are schematically pictured
in Fig. 13-43 and are described as follows:
a. The float regulator has a float ^.hamber piped to the boiler drum. It is installed
opposite the normal water level of the boiler so that level can be duplicated in the
float chamber. As this level changes, the notion of the float mechanically opens or
closes a balanced regulating valve in the feedwater lint thus governing the inflow
of feedwater. On some noncontinuous float regulators the float motion actuates a
switch for pump motor control.
b. The generator type of regulator has a vapor generator whose central pipe is
connected to the boiler above and below the water line. Surrounding this is the finned
generator chamber communicating, through connecting tubing, with the diaphragm
chamber of the spring-closing regulating valve. This region is filled with water to a
level in the generator chamber determined by the manufacturer. It is a closed region
with all joints pressure-tight. Steam above the water level in the central tube is
condensed by heat transfer to the generating chamber where some of the water is
vaporized at a pressure which is determined by the rate of heat leakage from the fins
and the height of the water line in the central tube. This pressure, transmitted to the
diaphragm of the valve, produces a stem movement sufficient to govern the fiow of
water to the boiler.
FEEDWATER REGULATION 565
c. The thermostatic tube re^^ulator is actuated by slight changes in the length of
a long inclined tube in which the water level of the boiler ii^ duplicated by proper pipe
connections to the boiler drums. Because of the flat angle of installation, a small
change in boiler water level creates larger changes in the portion of this tu^ that is
exposed to steam. Therefore, its average temperature varies with water level. The small
linear thermal expansion that follows is mechanically multiplied by a lever whose end
movement is utilized to operate the balanced regulating valve.
Normally, these regulators actuate the feed valve directly, but cases arise
where it is desirable to operate the valve by a relay under the control of a
primary element such as the thermostatic tube. Relay operation is recommended
for high-pressure boilers (40 kg/cm^and over) and for large diameter valves (say

b ^'opor prttiurt typt

e.Th«rmostotic cspontion typp

Fio. 13^ Three generic types of boiler feedwater regulators for simple level control.
566 THE FEEDWATER LOOP
127 mm and larger) operating with high differential pressure. The use of an air-,
oil-, or water-operated relay provides the necessary power to position the valve
accurately and rapidly, thus permitting the primary element to operate more
sensitively.
Regulators designed to maintain a constant feedwater pressure at the
pump by controlling the steam supply to it have been in use a long time. When
close feedwater regulation is necessary an excess, or differential, pressure regu¬
lator is used to maintain constant pressure across the control valve. Engineers
will appreciate the fact that the amount of water that may be passed by a valve
is dependent not only on the area of the opening, but upon, the pressure drop
across the valve. The practice of modern engineering is to control the water
pressure differential in the individual feed line to each boiler so that there will
be a uniform water pressure drop across each regulator. The excess pressure
regulator is installed at the inlet side of the water level controller. Acting prac¬
tically as a reducing valve, it maintains at its outlet a pressure which is a
delate amount higher than the pressure at the controller outlet.
I In order to take advantlige of the thermal storage between high- and low-
water level in the boiler drum, the regulator should gradually increase or de¬
crease the feed on slowly increasing or decreasing steam demands, but it should
decrease the feed during rapid increase of steam demand and increase it when
the demand decreases rapidly, for in that way the fluctuation of boiler water
level can offset the time lag of combustion response to changing steam require¬
ments. For illustration, suppose that the regulator shown in Fig. 43-c is regu¬
lating the water level at middle gauge. The expansion tube is half full of water
and half full of steam, giving normal valve opening which produces a normal
feed. There is a sudden increase in steam demand, and rapid withdrawal of
stored steam is followed by a rapid drop in pressure and bubbling up of escap¬
ing steam at a rapid rate. This raises the water level and contracts the tulbe
which decreases the valve opening. The decrease of cold feed input allows the
boiler to steam at maximum rate A continued heavy steam demand gradually
lowers the water level in the boiler and tube, expanding the tube and opening
the control valve wide enough to pass the required flow.
Modem boiler operating conditions have necessitated adding to standard
control equipment an element which is sensitive to the flow of steam and which
is auxiliary to the ordinary water level regulator The auxiliary control is
exerted by means of a steam pressure drop, such as that existing across the
superheater, which drop is roughly proportional to the square of the steam flow.
As shown by Fig. 13-44, the feedwater regulating valve is operated by the
combined effect of two controls, water level and steam flow. In this particular
system the steam flow element is a diaphragm unit actuated by pressure drop
through the superheater, but the differential pressure could be taken across an
orifice in the steam line if there were no superheater. The diaphragm balances
the flow-induced differential pressure against a spring force. At any time, then,
it assumes a position proportional to steam flow. The two-element regulator
being responsive to flow as well as level is able to anticipate changes in level
and can produce a more stable boiler water level when the load is fluctuating
than can level control alone.
The reader might now review the contents of the last two chapters to grasp
 PROBLEMS 667

Fig. 13-44 A two-element feedwater regulating system.

fully the double loop idea as outlined in Fig. 9-1. He will discover that, not only
has the equipment been described, but, what is more important from the stand¬
point of power plant engineering, its relation to the remainder of the loop, and
the behavior and control of each loop on variable load (the fundamental operat¬
ing condition) has been treated in as much detail as space permitted.

PROBLEMS
1. The feedwater to a boiler is 92% condensate and 8% make-up containing 270
ppm solids. What weight of solids enters the boiler per hr at a 22680 kg per hr steam
evaporation?
2. In a test for hardness of a water sample, 14,0 ml soap solution were needed in
a 60-ml sample of water. Lather factor, 1.5 ml. Phenolphthalein and methylorange
alkalinities were determined to be 0 and 150 ppm respectively. What percentage of
the hardness is of the permanent type?
3. The hardness of a feedwater is reported as 4.5 U.S. Degrees. Phenolphthalein
alkalinity, 50 ppm; methylorange alkalinity, 190 ppm. Describe the character of the
alkalinity and hardness of this w^ater.
4. Write the following water analysis on a balanced epm basis: Ca, 55.0; Mg, 19.0;
Na, 10.0; HCO3, 220.0; SO4, 29.0; Cl, 9.0; all in ppm.
5. Write the following w^ater analysis on a balanced epm basis: Ca, 29.0; Mg, 7.6;
Na, 12.0; HCO3, 98.0; SO4, 22.0; NO3, 2.1; Cl, 9.2; all in ppm.
6. Determine the quantity of each reagent for lime-soda treatment of 8800 litow
of water of either problem 13-4 or 13-5 assuming lime to be 90% pure, soda ash
86% pure.
668 THE FEEDWATER LOOP
7. Find the hardness in U.S. Degrees'of a water whose ionic analysis in ppm is:
Ca, 58.0; Mg, 17.0; Na, 44.0; HCOa, 178.0; SO4, 44.0; NOa, 2.2; Cl, 91.0.
8. The water of Prob. 13-7 is used as make-up for a boiler evaporating an average

of 272156 kg steam per day. Make-up, 40%; remainder condensate. What weight of
Ca(OH)2 and NaaCOa would constitute a month's supply for water treatment?
9. Estimate suitable proportions of a blow-off tank for the boiler of Example 1,
Sec 13-4 and sketch it approximately to scale.
10. The ] 370 mm di». steam drum on a boiler is 2440 mm long and has a 260 mm
8>^uge glass at mid-drum level. Find the maximum steam generation that could be
oared for by a blowdown of half a water gauge each 8-hr shift. Pressure, 17.5 kg/om* ga
Sf = 150 ppm; Sb, 2000 ppm.
11. The annual production of a boiler is estimated to be 2 x 10^ kg of 16.6 kg/om* ga
steam. Efficiency: 10 kg equivalent evaporation per kg coal. Coal cost $6.50 per tonne.
Assume that 85 < 3500 ppm; m = 0.03; 8^ = 400. What annual dollar saving is pos¬
sible from a continuous blowdown system which will cool the effluent 66.6®C ?
12. Find 1/min of continuous blowdown required to maintain boiler water concen¬
tration within ABMA limits for this case: 850 hp boiler operated at 150% rating. 28
kg/om* ga, 316% feed at 138% Feed is 30% make-up, 70% condensate. 8fm»120.
13. Determine the maximum sum that can be economically invested in a continuous
blowdown system for an 66 kg/om> ga plant whose steam generator averages 463600 kg
daily at 75% over-all effioienoy. Fuel is ooal of 12 x 10* Joules costing $8.76 per tonne.The
49^0 feedwater is half condensate and half 16^0 make-up. 8f = 32. Base estimate on 20%
annual amortization. Diagram the system.
14. The exhaust of a mechanical drive turbine is estimated to be 1270 kg per hr at
14 kg/om* ga and 2850 J/g. How much feedwater will this heat per hr with a 2.7*0
terminal difference (^d®) T Cold water temperature, 26.7®C.
15. Ten percent of heated feedwater is exhaust steam (0 kg/cm* ga, 90% dry) condensed
in an open heater; 90% condensate return at 37.8*C. Find the heated water tempera¬
ture and the terminal difference. Describe the heater situation had the returns been
at 140"
16. Water at 60*0 will be deaerated in a heater at 0.7 kg/om* ga steam saturation pres¬
sure. <d=2.7®0. Steam is supplied through a reducing valve from a header where the
steam is at 10.5 kg/om* ga dry and saturated. How many kg per hr steam will be drawn
into the heater when the rate of outflow of deaerated water is 6304/min 7
17. Determine the principal specifications of a surface heater assigned from one of
the extraction heaters calculated in Ex 1, Sec 8-5. Tubes to be 19mm by 1.02 mm copper
alloy in low-pressure heaters. 1.6 m/seo; U, «29205.
18. The conditions of steam extraction to a surface heater are: 2650 J/g 1.1 kg/
-cm* ab. Water flow is 13470 kg per hr heated from 20.6® to99®C Specify principal
dimensions of this heater, given desired tube length about 2,4 mxl6.8 mm x 1.27 mm
tubes of copper alloy ; F^—1.6 m/seo, U% =29296.
19. Lay out the two tube sheets and water Box partitions for the heater of Prob. 18,
on the basis that it is six-pass with 18 tubes per pass.
20. During a test an extraction heater condensed 557.6 kg steam per hr and heated

17890 kg from or to 78®C. The heating surface is such that A = 19.7 m*and F^
2.11 m/seo . Steam pressure, 0.8 kg/cm* ab, 94*0; condensate, 82*0 . Find the td, V, and
thermal efficiency.
21. Evaporation of 2268 kg of raw water is the desired performance of a single-stage
evaporator. How much live steam will it use 7 Live steam at 7.3 kg/om* ab, 18.3%
superheat. Raw water at 15.6*0. 5=33.33*0 on the heating surface. 5% blowdown,
22. Solve Prob. 13-21 were the evaporator a two-effect type. Record the necessary

evaporator condenser vapor pressure.

PROBLEMS 569
23. Calculate the maximum quantity of distilled water that can be obtained from
the operation of a multiple effect evaporator operating between 7.0 kg/om* ab dry and
saturated live steam state and 1.03 kg/om* ab oondenaer vapor pressure. Kegleot blow-
down # about 21. PC. Consider that the input
is 0.46 kg of steam.
24. An evaporator is inserted into the
heat balance of an industrial plant as
shown in Fig. 13-24P. A make-up of 10%
is required and this is supplied by the
evaporator. Calculate boiler feedwater tem¬
perature, and all flows (kg per hr) for a
generator load of 100 kw. =* 2635 J/g.

25. Half of the steam sent by an indus¬

trial steam plant to the factory is returned
as condensate. An evaporator, located as
in Fig. 13-25P, supplies the other half of
feedwater flow. Calculate the boiler feed-
water temperature. Neglect blowdown and
heat losses.
26. Were the raw water of Prob. 4 (or 5) to be fed to an evaporator in which the
concentration of TDS should not exceed 7000 ppm, what rate of continuous blowdown
should be practiced? Average evaporation is 1814 kg per hr.
27. Find the cold water TDH for a pump through which 4731/mof 03.3^C water
are passing. Discharge, 50 mm dia; suotkm. 75
mm. Discharge pressure is 2.5 kg/om* ga mea*
BuredO.6 m above pump center-line. Suction,
0.4 kg/omSga measured 0.45 m belowoenter-line.
28. Calculate drive hp for puinpingl708
1/mincold water to a tank. Suction at 127
mm Hg vacuum, delivery at 5.3 kg/om* ga,
both measured close to pump. Tjp bsO.OS.
29. Estimate discharge rate(I/min) and
maximum suction lift of a 95 mm x 127 nun
X152 mm duplex, direot-aoting steam pump.
iQy »0.90. Sea levehCould this be a boiler feed
pump?
30. Select suitable dimensions for a du¬
plex direct-acting steam pump for the following boiler service. 265 l/min against 3.8
kg/om* ab boiler preMure. Water at 03.3^. Boiler water level, 5.5 m above pump.Installation
of suction in accordance with Fig. 13-27. Altitude, 457 m. Water cylinder D = L.
= 0.90. Neglect velocity head and pipe friction. Find whp.
31. Results of a test on a 254 mmxl52 mmx305 mm duplex direot-aoting pump are
given. Find r}y, whp, **dutv.'' Test data : Time, 60 min. Steam used 567 kg of 14 kg/om**
ab saturated steam. Exhaust, 1.03 kg/cm* ab. Speed, 38 strokes per min.per cylinder. Water
pumped, 22680 kg. TDH. 76 m of 82.F*C water.
32. What size of 1750-rpm motor should be used to drive a l00mmx200mmsingle.^
acting triplex power pump ? o.88, TQm, 0.90, tQv, 0.95. What V-belt pulley diameters
should bo used ? Water-temperature. 110®0, TDH *24.6 kg/om*.
33. Draw the and the Q-P curves of the 305 mm impeller pump charted in Fig.
13-32 if it were directly connected to a 1200-rpm motor. What motor size will be
needed?
670 THE FEEDWATER LOOP
34. What turbine speed would be required if the 305 mm impeller pump of Fig. 13-32
were to be driven so that I5t4 I/m could be delivered at 84 m head? Hint: Plot 1750-
rpm curve on Q-H plane, then add others (by the laws of similitude) until one passes
through the specified Q-// point.
35. A 2-stage centrifugal pump delivers 190 l/min of 107®C water against 76 m head
at 3500 rpm. What is its specific speed? What head could a geometrically similar pump
produce if operated at 3000 rpm at 151 l/minflow?
36. Determine the minimum practicable static head of hot water on the suction of
a double-suction, centrifugal, heator-condensate-pump. Water at 85"C, 92530 kg per
hr flow. Heater pressure, 0.7 kg/cm* ab ; discharge at 2.8 kg/om* ab. Neglect pipefriction.
37. The constant-speed (3500-rpm) performance of a multi-stage centrifugal pump,
corrected to 121®C , is as follows:

Q, l/min: 379 757 1130 1614 1898* 2271

H, m: 661 652 640 616 582 618
%• 25 45 59 68 72 70

Diagram the connections for using this pump as a boiler feeder. Suction from overhead
water tank pressurized to 2.5 kg/cm* ab and heated to 12PO.NPSH in accordance with
Fig. 13-37. Control system: FW regulator valve plus speed control of the driving tur¬
bine. Boiler is 44.6 kg/om< ga. Discharge static head, 9 in. Friction head 2.8 kg/cm>, FW
regulator working differential pressure a:2.1 kg/cra*. Calculate (a) the static head on the
pump suction, (b) the head at point A. Fig. 13*38a.
38. For the same plant as described in Prob. 13-37, construct a scaled diagram
similar to Fig. 13-38a, showing all lines except N^. B is to be at 11361/m .
39. Repeat Prob. 13-37, except adapt the pump to throttling control, i.e., constant-
speed motor, differential pressure valve in series with the FW regulator, and by-pass.
40. For the same plant as described in Prob. 13-39, construct a scaled diagram simi¬
lar to Fig. 13-38b, showing all lines except F3. B is to be at 1136 l/m. Calculate the
required orifice flow for cooling at zero feedwater flow and select an orifice size from
Fig. 13-41.
41. Determine the orifice flow and select an orifice diameter (Fig. 13-41) for a
by-pass line around a throttle-controlled boiler feed pump.TDH = 244 m, = 5% at
37 l/min, 15% at 114l/min20% at 228 l/min.
42. Suppose that the pump charactenstic data of Prob. 13-37 were for cold water
instead of hot. Then what would the efficiency of this pump have been at:
(a)l893 l/min; 3500 rpm; 121®C water. (b)li36 l/min;3120 rpm; lfI*C water.

* Rated point.
 CHAPTER 14

THE PIPING SYSTEM

14-1 Pipe System Classification. After a study of the vapor cycle and its
equipment, a person could hardly fail to note the multitude of flow lines re¬
quired to connect the individual pieces of equipment to make them the homo¬
geneous working unit constituting the well designed power plant. Flows of
liquids, gases, or vapors are through pipes comprising the piping system of the
plant. Piping, valves, and fittings are indispensable necessities in all power
plants, especially the steam plant. The piping incidental to the installation of
a power plant may not appear, offhand, to be of primary importance, yet its
selection and arrangement is deserving of the most careful study. The best
possible selection can be made of mechanical equipment, yet the plant may be
inferior from the standpoint of economy, operation, appearance, and reliability
if the piping is not equally well selected. Then too, with advancing temperatures
and pressures, and with the increasing complexity of the central station cycle,
the cost of the piping system has become one of the major items of expense and
its selection and arrangement one of the major items of design.
Service conditions of portions of the piping system are frequently rigorous.
Feedwater piping may be expected to carry pressuresS.S-Tkg/cm^or more, higher
than the boiler pressure; steam temperatures as high as 425-540® are met in
some plants; corrosive and grit-laden liquids are sometimes pumped; moisture-
laden steam may have to be handled at high velocity; and some of the water
lines may have to resist water hammer as well as hydrostatic pressure. Obvi¬
ously, these various service conditions call for differing construction if the
least expensive system is to be built. Piping systems may be classified on the
basis of the fluid carried, for example (1) steam, {2\ cold water, (3) hot water,
(4) oil, etc., but a more useful classification is based on the service conditions,
as follows:
1. High-pressure superheated or saturated steam.
2. High-pressure drip piping.
3. Low-pressure steam piping.
4. Boiler feedwater piping.
5. Heater piping.
6. Blow-off piping.
7. Condensate piping.
8. Hot and cold water service piping.
671
572 PIPING SYSTEM
9. Circulating water piping.
10. Oil piping (lubricating, fuel, insulating).
11. Instrument piping.
12. Compressed air piping.
13. Soot blower piping.
14. Drainage piping, etc.
14-2 Commercial Pipe. Pipe is either wrought or cast. Wrought pipe may
be welded or seamless. The material may be iron, carbon steel, alloy steel,
wrought iron, brass, or copper. There is very little use of cast pipe in power
plants—^most pipe being wrought steel, with wrought iron in fair demand for
corrosion resistance—as in condensate lines.
Cast Iron, This pipe is principally used for drainage or for resistance to
corrosion and abrasion. Various weight classifications, for heads to 244 m water,
are listed under American Waterworks Association Standards.
Wrought Steel, Most power plant piping is the low-cost and strong wrought
steel pipe. The resistance of wrought steel pipe to corrosion is increased by a
protective coating of zinc. Such pipe is designated as '^galvanized.'' Small water
pipe is often galvanized, but large pipe and gas, steam, air, or oil pif)e are
^'black.” Wrought pipe is manufactured by welding long plates, called skelp,
with butt or lap weld; or by piercing and drawing a seamless pipe from a billet.
Butt-welded pipe has small area of contact at the weld. Its inherent weakness
limits its use to sizes smaller than 76.2 mm. Pipe is not lap welded in sizes smaller
than 50.8 mm. Large OD pipe is made by hammer welding and by spiral riveting.
Alloy Steels, These are steels which owe their special properties to alloying
elements other than carbon. Such steels are required for temperatures in excess
of 455®C and may be beneficially employed above 400®C. Carbon-molybdenum
alloy was used for many years, but recent experience with graphitization* at
very high temperatures has directed attention to the advantages of other alloys.
Chromium molybdenum is being specified for the.highest temperatures, as it has
been found that carbon steel is subject to graphitization above 427*ti C-Mo
steel above 477®C . Cr-Mo steels have Mo. with Cr ranging from 1 to 5%,
depending on the temperature to be met. Very high pressure and temperature
pipe of large size is made by turning and boring solid forgings.
Wrought Iron. This is a two-component metal consisting of iron permeated
with 1% to 3% of finely divided and uniformly distributed iron silicate. The
silicate is distributed throughotit the base metal in the form of threads or fibers.
(Jntil the fourteenth century wrought iron, was the only product made from iron
ore. Thus there has been ample time to Verify that this is a metal of superior
resistance to most forms of corrosion. In power plants it is favored for conden¬
sate, feedwater, and blow-off lines. It is not advisable to employ it when pres¬
sure exceeds 17.5 kg/cm*.
A superficial examination is not always able to distinguish wrought iron from
wrought steel pipe. Wrought iron manufacturers mark their product by some
means such as spiral color stripe. A ragged fibrous fracture—or a crumbly chip
instead of a long shaving from a pipe threader—^will identify wrought iron.
* Graphitization is a condition where the carbon in a steel continuously exposed to high
temperature changes to graphite and may concentrate at critical spots, definitely weaken¬
ing the steel in tension.
COMMERCIAL PIPE 573
Since wrought steel pipe is frequently called ‘‘wrpught iron'^ by material men
and contractors, one should specify ^‘genuine wrought iron^^ when it is w'anted.
Copper and Brass, There are several compositions used for brass pipe, all of
which contain over 60% copper, the remainder being mostly zinc. Brass pipe is
cold drawn to size and annealed. It may be threaded and is rigid, compared to
copper alone. The latter is generally employed as tubing where flexibility is de¬
sired or where some special quality of the fluid is detrimental to brass.
The use of brass and copper pipe is limited by its cost to piping in which
flexibility, appearance, or resistance to certain forms of corrosion are important.
Brass and copper piping is principally used for oil lines and occasionally for
sections of boiler feed pipes. Brass pipe should not be used above 21 kg/cm*
or 204®G.
Some of the standards and specifications applying to the more common pip¬
ing materials are given in Table 14-1.
In order to meet various service conditions economically, steel pipe is manu¬
factured in several wall thickness classifications. The American Standards
Association has established a standard, ASAB36.10, for a rational system of pipe
wall thicknesses to supersede the common designations* of “standard,” “extra
strong,” and “double extra strong.” In this standard Schedule Numbers were
introduced to designate pipe weight. These numbers are an approximation of the
equation:
Schedule Number = lOOOp/S (14-1)
where p = Gauge pressure carried, kglcm\

S = Working stress,kg/cm*. See Table 14-1.

These numbers are made to agree closely with regularly used commercial sizes,
the wall thickness of Schedules 40 and 80 being identical with those of the old
Standard and Extra Strong lists up through 203.2 mm nominal size* "Double
extra strong” is pipe with wall thicknesses double those of “extra strong.*'
Pipes are sized by their nominal inside diameter for sizes up to3(>4.8 mm. When
the pipe wall is thickened to provide additional strength, the increment is added
to the inside so that, the outside diameter remaining constant, the same size of
thread and fittings can be used for all weights. Above 304.8 mm the sizes are
based on outside diameters; for example, 406.4 mm OD by 9.53 mm. Pipe is
marketed in random lengths of 4.88 m to 6.71 m, 6.4 m being a common length.
Seamless sftee) pipe in the smaller sizes can be obtained inverybng lengths (7.62-
10.67 m).
A table of pipe size standards, incorporating schedule numbers, will be
found in the Appendix. The tabulated thicknesses are average wall dimensions.
Customarily steel pipe specifications permit the actual pipe wall thickness to
vary from the nominal size by ± 12%%.
Many formulae have been devised for calculating the gauge working pressure
in a pipe of given size, material, and thickness. One of these, of considerable
usage, is the Barlow formula which is based on a factor of safety of 8 and
applies to thin-walled pipe.
d% « 4pDo/Su mm. (14-2)
♦These, however, are used for genuine wrought iron pipe.
 Table 14-1. STANDARDS AND SPECIFICATIONS FOR POWER PIPING
(Abstracted. Consult ASA and ASTM standards for full information***)

SB

•
Material ASTM Specification

•s^

1
1
H
J*
J
1

1!
1S 1
18*'
A-120 Furnace welded and

«!

s
seamless drawm

“
.
Carbon steel

CO
^
1 &
i 5$

Pm
1
00
QD

J:
a
1
if
'Z
2
•S
S
t
<?
i

<

«
*■4
-
wm

>>%
6
J
OQ

%
1
§
i
04

II
§3
si
3 E
■S.S
i
M

®
00

ai
i i
" i
1
M1

PO
1
s

O
1
0
^
Hi
«
CD «
% t

^
s
to me
•0

U
01
a
1
1
1
i

574
cdii
es
3
8

e«5
« lis
|*Ss

1
■
l-gS
J,Si2
i
t* to

1
«e 00

I 0
00
0
0

1
5
Pf
8
1

!
1
J-
s
g
1
1

s 3
1 r-

•t
U
1
a
1
•i
s

1
s
»o

3
b 31«

I'i 1
1
oL
«

1
fm
CD
•

ii
i 1

1/5
cl)
1
X

i
1
1
5
E
S

1- a ,f
DO

6
»o
1
r-
CD

8.
i
§ § 1 i

5
j
1
!

1
tjr
1S-°
i
i
iO
cn
l>
i

<o
o
t
e

1
1
cl)
o
11
fsa
8
S
i

<
g
1"
Sj3
If
ID
o

f&
8
1-4
CD CD

1
CD
CD 'Ol #•

4ii4
a 1
is 1

*c
b.
•c

CD
§'

CD

i 1
s
00
di «e

»Uoy * ^^**'^^*** weld; LW «= lap weld; FW, RW « electric fusion or resistance weld; S •» seamless; A, B « ductility grades; li Cr • \\% Cr-Mo
AM B31.1. Code for power piping Stresses given include the necessary factor of safety, for these metals yield from 3516-7031 kff/om^ ultiinate strength.
Where exact values are required, interpolation from Oiis table may be 10% in error.
PIPE CONNECTIONS 576
However, the ASA Code for power pressure pipin(t gives the following, inot
limited to thin wall:

In Eqs 14-2 and 14^3 the symbols are:

dt = Minimum pipe wall thickness, mm.
p = Working pressure, kg/cm* ga.
Do = Outside diameter, mm.
Su = Ultimate tensile strength, kg/cm*.
S = Working stress (Table 14-1).
C = 1.27 for pipe25.4mm and smaller; 1.65 for 31.8 mm and larger.
(Allowances for cutting, corrosion, etc.)
Esamplc 1 : The neoessaTy specification for a 152.4 mm pipe to carry steam at 4d.2 kg/
em* ga» 426.7'*C is to be found. Table 14-1 indicates the use of seamless carbon steel pipe
A-106. Qf Qrades A and B, the latter is of greater tensile strength, but Grade A is more
ductile, i.e., bends and cuts better. The working stress for Grade A at 426.7*^ is obtained
from the table, =687.7 kg/cm*. Using Bq 14-1, Schedule Number^ 1000 x 49.2/587.7=83.8.
Ordinarily the next higher Schedule Number is chosen, but as this would require using
120 rather than 80, the latter is tested for safety with Eq 14-3. Nominal pipe thickness
for 152.4 mm Schedule 80 is 10.97 mm. Outside diameter is 168.28 mm. The minimum thick-
ness (alloiriag for manufacturing toleranoe>=10.97— 0.125 x 10.97 =9.60 mm. Then
49.2x168-28
9.60 « +1.65
25+0.8x49.2

S * 501 kg/om<

Since this is less than the maximum safe stress, Schedule 80 pipe of A-106, Grade A,
manufacture is the specification.

14-3 Pipe Connections. Adjacent sections of pipe are connected together

in various ways but all connections can be grouped under these four headings:
1. Packed joints such as leaded bell-and-spigot, or plain end coupling.
Mainly used for low pressures, soil pipe, drainage, large OD pipe.
2. Screwed joints such as couplings and unions. Generally used for sizes less
than 101.6 mm, but large sizes are marketed.
3. Flanged joints with companion fianges either loose or screwed, shrunk,
riveted, or welded to the pipe. Flanged fittings are generally called for in the
larger sizes of pipe; and for high-pressure, high-temperature work.
4. Welded joints. Welds made by the fusion process (as opposed to hammer
welds) using gas or metal arc welders. The first cost of the welded line is an
advantage in large pipe sizes. Can be used on high-pressure and temperature
lines.
The type of joint selected for use depends on several factors, viz., size of
pipe, pressure and temperature, ease of removal of sections between joints, ex¬
pected future maintenance, labor available for installation, and first cost. The
676 THE PIPING SYSTEM
type of joint used influences the method selected for depicting piping dimensions
on drawings. Note Fig. 14-1.
Screwed Joints, The most common way of making a screwed joint is to use
wrought couplings. All fittings for screwed pipe are threaded to conform to the
American Standard for pipe threads.
Threads should be smooth and clean, go
together easily, and remain pressure-
tight. Each length of threaded pipe as
purchased normally comes equipped with
one coupling. The standard coupling is
right handed, but right and left couplings
can be had where the pipe is not to be
turned. More frequently, unions are pro¬
vided when easily broken joints are
wanted.
Pipe measurements are made end to
end of pipe. When it becomes necessary
to calculate the length of a piece of
Fig. 14-1 Details showing dimension¬ screwed pipe, as for instance in calculat¬
ing of piping ing the length to cut for a given offset,
one needs to know how far, on the aver¬
age, the pipe will enter the fitting, otherwise a pure guess will be made and the
pipe as cut may or may not give the desired pipe line configuration. Fig. 14-2
gives the approximate lengths of thread that will enter the fitting m making up
a tight joint.
Pip# E
SIZ0 mm
12-7 12-7
19-1 14*3
25'4 17*5
31-8 17*5
38*1 17-5
50*8 19-1
76-2 25*4
101*6 28-6
127-0 31*8
152*4 33-3

Fig. 14-2 American Standard Pipe Thread ^with normal engagement length E re¬
quired to make tight joints.

Example 1: Let it be required to determine the length of the pipe to be cut for a
0*985 m offsef in a 7fl.2 mm line, standard pipe.
The layout, shown in Fk. 14-1, can be draijrn after taking the dimension B from
Table A-8 for a 76.2 mm standard 45^ ell and the dimension E from Fig* 14-2* Baa65.1 mm;
Fs25.4mm. The length required is seen to be j[>—2(B—Al)* The offset isO.QB6m4-s
L « « 1.365 m.
Therefore, the required length is 1.365 -2(0.0551 -0.0254)b: 1.306 m.
PIPE CONNECTIONS 577
Flanged Joints, Low-pressure piping larger than 152.4 mm and high-pressure
piping are, in the majority of cases, coimected by companion flanges. Flanges
are drilled to a standard templet and drawn tightly together by means of flange
bolts. Alloy steel bolting to conform to ASTM specifications should be used
when pressures exceed 11.25 kg/cm* or
temperatures exceed 232^^ but below these
limits commercial bolting can be used.
Flanges are manufactured in a number
of styles, the more important of which are
shown. It will be noted that some are de¬ Coupimg Union

signed to be screwed to the pipe, others Fig. 14-3 Screwed joints.

welded on, still others being loose. With
the latter, the necessary pressure-sealing gaskets are held between flanged and
faced ends of the pipe itself, while in the other cases the gaskets are pressed
between flange faces. The loose flange joint is called a lap joint or a Van Stone
joint. Its advantage is that the pipe does not need to be turned in order to bring
bolt holes in companion flanges into register. Often a line with tight flange joints
is terminated with a lap joint in order to simplify connection to a fitting or
valve. A.blank flange is a standard flange in which the bolt holes have not been
drilled. A blind flange is a solid disk, drilled for bolting, and used for temporary
end closure of flanged lines.

SOME STANDARD PIPE FLANGE FACES

XL. .1. =a. ^

Screwed Welded Welding neck Lapped stub end
PIPE—FLANGE ATTACHMENTS
Fig. 14-4 Companion flanges.

As it is expensive to lap-grind joint faces to a fluid tight condition, gaskets

are inserted between the faces of companion flanges or pipe laps to provide the
seal. When flange bolts are tightened the gasket material flows into the minor
surface irregularities and seals the fluid in the pipe. If flange bolts stretch or
gasket continues a slow yield, flanges may later begin to leak and require a
further tightening—or renewal of gasket. Special high-temperature joints such
as the Sargol have very smooth finish. No gasket is used, but sealing lips are
provided which may be welded together to stop leakage.
678 THE PIPING SYSTEM
Gaskets are generally cut from red rubber, sheeting for air, steam, and water
lines up to 100®C. Oil lines require something other than rubber, such as fiber.
For high-temperature lines, such as high-pressure steam, the gasket will usually
be wire-inserted asbestos or asbestos-corrugated metal sandwich. Thin, corru¬
gated copper is satisfactory only if the flange
surfaces are lap-ground to a smooth plane sur¬
face.

k Very high pressure-temperature lines are

gasketed by corrugated iron, soft steel, or Monel
S’l metal disks, also by ingot iron rings. Special
flange faces must be provided for the ring joint.
Fig. 14-5 Proportions of Its advantage is the concentration of pressure,
flanges. Raised face is optional. making heavy-duty seals without excessive bolt¬
t=1.59mfn in 10.68- and 81.1 kg/ ing loads.
cmv Standards, 0.35 mm on others. Welded Joints. Welded joints are made by
See Fig. A-7 for dimensions. bringing the two pipe ends within a specified
small distance of each other, then joining them
by a high-temperature fusion of weld metal from a welding rod and .the pipe
metal. This butt welding process is the one generally employed, although others
are used on occasion. A good butt weld is stronger than the pipe itself, requires
no gasket, flange, or bolts, and is free from future maintenance. It cannot be
broken for disassembly as can the screwed and flanged joint.
The speed and cost of making welded joints are in their favor for larger pipe
sizes, for higher pressures, and for lines that are to be covered with heat insula¬
tion. Welded pipe lines are the only satisfactory conduits for extremely high-
pressure and high-temperature steam. Thus the main steam lines of a modem

Butt Weld
For Metal Thickness of
H mm or less
90* Nozzle Weld
90” Nozzle Weld
Standard Type
Reinforced Type

Bute Weld
For Metal Thideness
greater than lOw"*

Welded Coigdinf WeldedPad Angular Nozzle Weld Mlar Nozzle Weld

Standard Type einforced Type
Crane Co.
Fig. 14-6 Types of welded joints.

power station will be of welded construction because of the deficiencies of other

methods of jointing, whereas pipe lines elsewhere in power plants and industry
may be so joined because it has been found to be the cheaper method. Manu¬
facturers have developed lines of fittings and valves for use with welded joints.
Welded connection valves are to use with extreme steam conditions, for most
designers prefer flanged valves in welded lines if gasketing difficulties are not
expected.
PIPE CONNECTIONS 679
The advantages of welded joints are, besides low cost: (1) neat appearance,
(2) no gaskets to blow out, (3) additions and changes readily made with cutting
torch, and (4) insulation is easily applied. Many pipes which are screwed or
flanged have the small branches and connections (instrument piping, auxiliaries,
etc.) welded on. On the other hand, in spite of procedure control, much still
depends on the skill of the welder.
This fusion welding can be made with either the gas flame or electric arc
method. Electric welding predominates where preheating the pipe or annealing
the weld is not required. To obtain welds of uniform strength and ductility, the
equipment, materials, and welding procedure must be expertly selected and con¬
trolled. An outline of procedure control for fabrication of welded piping might be
as follows:
1. Check of the welders.
a. Experience.
b. Qualification tests.
2. Selection and inspection of material.
a. Use commercial iron or steel pipe of good weldable quality.
b. Have welding rods confonn to specifications for such.
c. Gases and apparatus.
3. Design and layout of welded joints.
4. Preparation of material for welding.
See that cutting or other preparation of the joint is correct and that it is cor¬
rectly assembled and clamped preparatory to welding.
5. Organization and welding technique.
a. Qualifying the designers and operators.
b. Specifying the actual technique of welding.
6. Inspection and test.
a. To check the foregoing items of procedure.
b. To test the strength of individual welds.

It is advisable to specify conformance to one of the codes for the construc¬

tion of welded pressure piping. These codes set up very comprehensive standards*
for welding procedures, and the qualifications of welders to use these procedures.
ASME, ASA, and American Welding Society have these standard codes. On the
average construction project the supervising engineer may elect to qualify the
welders for the low-pressure, pipe-welded joints, but not for pressure piping.
For that he will usually avail himself of an existing service in the form of welder
qualification carried on by the insurance firms specializing in the damage in¬
surance of boilers and pressure piping. These firms will qualify a welder and
issue him a certificate of limited duration. By requiring a “certified” welder,
the engineer has some assurance that the joints in important piping are made by
workmen of proven competence.
Butt welds are generally used unless very high stresses are expected at the
joint. The pipe ends should be squared accurately and beveled to an angle of
46® to within 1.6 mm. of the inner wall. New pipe is purchased with ends
beveled. Field joints in cut pieces are generally beveled by flame cutting with
the oxygen torch. The abutting edges are then separated to a specified distance
and tack welded. The joint is then ready to be welded.
In thick-walled, i.e., high-pressure, pipes preheating before welding and
580 THE PIPING SYSTEM
stress relieving afterward are desirable to insure the development of full joint
strength.
Considerable concern has been given to evidences of graphitization at the
welded joints of high-temperature steam piping since the first failure, a few
years ago, of such a joint. Some graphitization has been found in carbon steel
used above 425®G and carbon molybdenum steel above 480®C. It is an erratic
phenomenon, not necessarily reproducing itself under similar or 'Equivalent con¬
ditions. It has, however, induced many companies operating high-temperature
steam lines to begin a program of periodic sampling of joint metal. There has
been developed a weld probing machine that removes a half moon sample from
the joint in such a manner as not to penetrate through the pipe wall. The
sample is adequate for microscopic examination and bend testing. The hole is
safely filled with new weld metal. A badly graphitized joint needs to be cut out
and a new section of pipe welded in. Some companies believe that slightly
graphitized metal can be rehabilitated by redissolving the carbon with induction
heating of the joint.

Elbow
FLANGED
Cross
PATTERN
h
Ecctntric
rtductr
^ o
Elbow Tot

Strsst Shoptd
Tts f Ibow Lotsrol 45* f Ibow nippits Soddle
SCREWED PATTERN WELDING PATTERN
Fig. 14-7 Some of the common pipe fittings.

Inspection of Joints, The inspector of new construction needs to examine

joint making while the work is in progress, and test the finished piping for tight¬
ness—as a system or in subsections.
During construction the material and pressure classification of screwed and
flanged joint material must be verified for compliance with specifications. The
performance of thread cutting equipment (for clean sharp threads) should be
checked. Workmen should be made to ream or file the burrs left on pipe interiors
after cutting. Poorly threaded joints can sometimes be tightened by caulking the
threads with punch and hammer, but specifications and inspection should forbid
it. Pipe should enter fianges and couplings the full distance indicated by Fig.
14-2. Flange bolts should be of proper style, material, and length. Gaskets should
be checked for conformance with specifications and suitability to fiange faces.
Inspection should confirm that they are installed concentrically on the flange
face.
Welding is inspected to determine if beveling, separation, and alignment are
correct, if proper welding rod is on the job, if metal is being laid on in suitable
number of passes, and if the finished weld has the “right look.” '
When a section of piping is ready for test, the ends are blanked off, and any
PIPE CONNECTIONS 681
equipment, such as pressure-reducing valve diaphragms, that might be daiqaged
by over-pressure is removed or protected. The pipe is filled with water and
hydrostatic pressure applied to the extent specified (often 150% of worliing
pressure for 24-36 hr) and the line made tight at this condition. During such a
test the joints of welded pipe should be tested by striking the pipe on each side
with hammer blows, as should the weld metal itself.
Simple tests such as these are of no value in detecting small flaws in pipe or
weld. The safety problem is acute in high-pressure, high-temperature alloy
steel pipe and welded joints. An internal ^‘look^^ into the metal is about the only
nondestructive means of checking for internal cracks. For several years it has
been possible to X-ray fabricated piping in the manufacturer’s shops, but X-ray
examination of field welds is impracticable on account of the bulky equipment
and the restricted space in which pipe is so frequently located. Instead, radio-
graphic inspections are carried out by gamma-ray penetration of the weld.
Radium was used for this purpose, but now the source of gamma rays for in¬
dustrial radiography is usually radioactive cobalt 60. In a capsule, this is
centered in the interio of the pipe at the weld while X-ray films are wrapped
around the outside over the weld. Exposures of from one to several hours (de¬
pending on w’all thickness and diameter) will so mark the film that, when
developed, sub-surface cracks and other defects may be detected by the trained
examiner. This inspection is expensive and requires special training and care on
the part of the inspection technicians; hence it is used mainly for extreme
service conditions where protection through unnecessarily high safety factors
would be prohibitive.
Recently an ultrasonic method of exploring the integrity of joints and pipes

Fiq. 14-8 Installation of welded pipe fittings.

682 THE PIPING SYSTEM
has been tried with indications that it will prove to be a safe, low-cost means for
detecting interior flaws. It involves transmitting ultrasonic vibrations into the
section under test and comparing the transmitted and reflected waves on an
oscilloscope screen.
14-4 Pipe Fittings. The piping system, in fulfilling its function of provid¬
ing a flow path for the liquids and vapors of a power plant, is rarely a straight
run of pipe between two points. Flows are joined, parted, started, stopped, and
regulated in the piping system. Only occasionally is it possible to take a ‘^crow
flight” path between end connections; the common run of pipe must follow
configurations of equipment, walls, floors, beams, etc. Fittings and valves,
properly incorporated in the pipe system, enable it to meet these varied service
conditions.
In general, fittings consist of the pieces required to make turns, junctions,
and reductions. The straight size fittings are the 45° and 90° elbows, the tees,
crosses, Y’s, laterals, and reducers. These fittings may also be had in special
reducing sizes. The fitting pattern varies greatly with the type of joint. All forms
of fittings are available for both screwed and flanged joints; also many of them
for welded connections. Ordinarily, fittings are chosen for the same style of
joint as is used in the straight pipe runs. Brass and bronze fittings are used with
brass pipe, cast iron for working pressures up to 17.6 kg/cm*, malleable iron for
10.6 kg/cm*, and forged steel for high temperatures and pressures.
Table 14-2. FITTINGS CLASSIFICATION BY MATERIAL,
WORKING PRESSURE, AND JOINTS

Wrought steel
Cast Malleable Wrought Cast
Pressure class iron iron iron Carbon Alloy steel

Schedules 40. 80, IGO Welded Special

order

1.76 kg/om* Flanged

8.8 kg/om* Flanged Welded

& Screwed (Special)

10.6 kg/cm* Screwed Flanged

17.6 kg/om> Flanged Welded

& Screwed (Special)

21.1 kg/om* Flanged

28.12, 42.2. 63.3,106.5.

k 140.6 kg/om* Flanged

Higher pressures require forged steel

Note: Screwed bronze fittings and solder-type bronze fittings are also available, as are
stainless steel and Monel for special service.
COMMON VALVES 583
Fig. 14-7 shows a few of the many types of fM/tings that are manufactured
for use in pipe systems, and Table 14-2 gives a r^um6 of the materials and
pressure groups in which they are produced. Tables of dimensions of standard
fittings are relegated to the Appendix.
14-5 Common Valves. Valves start, stop, and regulate fluid flow. The com¬
mon pipe line valve consists of a body to house it and give a means of connect¬
ing it to the pipe, a valve seat, a valve, a valve stem, a top, a means of operating
the stem—^usually screw and hand wheel—and a packing to prevent leakage
around the stem as it emerges from the top. Valves are furnished for either
screwed or flanged connection. Inside, or female, threads are standard on
screwed valves, and if male threads are wanted they must be so specified.
Valves are sized by the nominal inside diameter of the pipe to which they are
connected. They are made of brass, bronze, malleable iron, cast iron, cast steel,
and forged steel. Cast and forged steel are employed in high pressure-tempera¬
ture service. Valve sealing materials (sometimes constituting the valve seat,
sometimes attached to the valve face) include rubber for cold water, brass and
bronze for ordinary temperatures, stainless steel, Monel metal, and various
other alloys for high temperatures.
Valves most used are the ordinary hand-operated globe and gate valves, and
the check valves. These are classified thus:
1. Globe valves (straight and angle).
a. Inside screw; outside screw.
b. Screw bonnet top; bolted yolk top.
2. Gate valves (straight and angle).
a. Rising stem; nonrising stem.
b. Wedge valve (split and solid); parallel seat valve.
3. Check valves (lift and swing types).
a. For vertical pipe,
b. For horizontal pipe.
The globe valves do not allow a line to drain completely; also, they offer
more frictional resistance than gate valves. They are frequently used in very
small lines (both water and steam) and where the valve is to be used for throt¬
tling, as they can be closely regulated and the seats which are liable to be cut
away in throttling service can be more easily replaced than in gate valves.
Gate valves are used in large pipe lines, in high-pressure steam lines, and in
all service where small friction loss is wanted. They should always be fully
shut or open and not be used for throttling the flow.
Gate valves differ chiefly in the action of the stem, or spindle, and the shape
of the gate, i.e., wedge or parallel. The inside screw rising stem is the simplest
construction. The position of the stem indicates whether the valve is open or
closed. However, the room needed for stem travel may not exist in crowded
quarters, especially if the valve is a large one, and the nonrising stem is better.
Also the packing wear is less. Life and maintenance are superior if an outside
screw rising stem design is used. Threads are not in contact with the fluid and
stem is packed against longitudinal motion only.
The disks of gate valves are split or solid wedge, and parallel double disk.
The solid wedge is the simplest and least expensive to build. It is therefore the
most common type for steam and water, particularly in the smaller sizes. On
584 THE PIPING SYSTEM
larger valves the split wedge with its self-adjusting feature not only maintains
a tighter closure, but is also freer from sticking closed. Parallel-seat, double¬
disk type of seat is especially jam-free as disks spread apart to the seats on
final turn of the wheel. Being more expensive than the split wedge, they are
less frequently selected. Some gate valves have the valve seat machined in the
metal of the valve body; others have renewable seats, screwed in. Fig. 14-9A is
the former; 9C, the latter. Integral seats may be refaced a limited number of

A
Inside Screw Inside Screw Inside Screw Outside Screw and Yoke
Non-Rising Spindle Traveling Spindle Non-Rising Spindle Rising Spindle
Solid Wedge Split Wedge Double Disc—^Parallel Double Disc—Taper
Seat Seat

Jenkins Bros.
£ F G H
One-piece bonnet Union bonnet 'Bolted bonnet. Outside screw and yoke
Renewable compositiou Regrinding bevel disc R«iewable composition Stainless steel plug and
disc witii throttling nut disc. Spider seat ring. seat ring

Fig. Type Seat Top Connection

A Gate Solid wedge . Inside screw, non-rising spindle Screwed

B Gate Split wed^ . Inside screw, rising spindle Screwed
C Gate DOTble disk, parallel seat. Outside screw, non-rising spindle Caulked
D Gate Double disk, taper seat... Outside screw, rising spindle Flanged
E Globe Composition disk. One pieee, screwed bonnet Screwed
F An^ Alloy bewl disk... Screwed usiion bonnet Screwed
G Angle Composition disk. Bolted bonnet Flanged
H Globe Plug . Outside screw & ytdk bonnet Screwed

Fio. 14-9 Common forms of gate and globe valves.

COMMON VALVES 585
times, whereas gate and seats may be renewed as many times as wanted in the
renewable type, whose life is therefore indefinitely long.
Globe valve designs feature different styles of the valve body top (which
guides and packs the stem) and different seat designs. The tops are classified as
screwed bonnet or bolted bonnet. The former are always inside screw type; the
latter may be inside or outside screw. All are rising stem. The same advantages
and defects prevail with integral seats as were cited for gate valves. The better
globe valves have renewable seat rings and renew¬
able valve disks or plugs. Valve elements are con¬
ical plugs or flat disks.* The cone types may be
reground in place when leaking, while renew disks
of vulcanized rubber, phenolic resins, soft metal, or
special composition are inserted to rehabilitate flat
disk types. The one-piece screwed bonnet type is Fio. 14-10 Swing and lift
most common, but regrinding seats is easier with check valves.
the union screwed bonnet. Bolted bonnets are em¬
ployed in the larger sizes with outside screw and yolk type used if it is desired
to keep the fluid from the screw.
Angle valves have the same internal construction as globe valves. They can
be used as an alternate to a globe valve plus an elbow and, as such, are cheaper
and have less fluid friction.
Materialwise, common valves are manufactured as:
1. Bronze valves. Noncorrosive, very malleable.
2. Iron valves. Iron body, bronze mounted (IBBM); or all iron.
3. Cast steel. High strength carbon or alloy steel with special high-tempera¬
ture duty seat trim.
Pressure ratings are given in Table 14-3. Do not assume that the maximum
sizes are available for the maximum listed pressures.
14-6 Special Valves. Many valves made by manufacturers must be classi¬
fied as special, although they are offered as standard catalogue items. This dis¬
tinguishes them from the common valves of Sec 14-5 and also reflects the fact of
their occasional use compared to gate and globe valves. Only a limited selection
of these special valves may be taken from manufacturers^ catalogues for inclu¬
sion here, the choice being based on their importance to the power plant field.
Safety Valves, These are primarily the boiler safety valves which constitute
the ultimate line of defense against the occurrence of hazardous steam pressures
in the boiler. Their importance is reflected in the considerable section of the
ASME Boiler Construction Code devoted to their specification and installation.
Here we shall describe the valve itself, and in the following section the selection
and installation.
The power boiler safety valve must be a spring-loaded valve, one which holds
the valve against its seat by means of a compression coil spring. The adjusted
compression in the spring determines the pressure at which the valve will open
and relieve over-pressure. The design of the ^^huddling chamber” in the seat
determines that steam suflScient to produce a certain pressure drop must be
exhausted before the valve will close. The design is such that the valve opens
* Needle valves for close regulation of small flows.
686 THE PIPING SYSTEM

Table 14-3. RANGE OF BODY MATERIALS, PRESSURE AND

TEMPERATURE RATINGS, AND SIZES OP VALVES

Maximum High Temperature

Cold
Working Maximum Maximum Range of
Body Metal Pressure Pressure Temp. Sizes
Gate Valves (kg/om*) (kg/cm*) (•C) (mm)

Brass 70.3 24.6 288 6.4 to 2S4

Xion 84^ 17.« 204 6.4 to 1219
Cast Steel 258.1 42.2* 538* 50.8 to 864
Forged Steel 210.0 40.1* 638 12.7 to 50.6

Maximum High Temperature

Cold !
Working Maximum Maximum Range of
Body Metal Pressure Pressure Temp. Sizes
Globe Valves (kg/om*> (kglom*) ro) (mm)

Brass 56.2 24.6 288 8.2 to 162.4

Iron 3&2 ire 282 «.4to4M.4
Cast Steel 103.5 4B.2 * 688* 60.8 to 804,8
Forged Steel 147.6 T0.»* 688* 8.2 to 60.8

Schutte-’Koerting YamaU^Waring Farria

Stop and check Blow-oflf (seatless) Pop safety
Fig. 14-11 Three special boiler valves.
* Higher pressure—temperature ratings with cast alloy steel.
SPECIAL VALVES 687
quickly and remains wide open until the pressure has fallen the pre-set amount
(generally 2-4%), when it quickly*closes. There is no floating and wire drawing
which could cut the valve or its seat. This is a distinguishing feature, for com¬
mon relief valves do not possess it.
Belief Valves. The relief valve is a form of safety valve, but usually intended
for less severe service and of less importance from the safety viewpoint. Relief
valves are applied to air, water, and steam lines, also to tanks, heaters, etc.
Among them could be mentioned the back pressure valves and atmospheric re¬
lief valves.
BUnv-ofl Valves. These, together with their connected lines, are to rid the
mud drums of sediment accumulations, to drain the boiler, to reduce concentra¬
tion of boiler water, and to provide a means for rapidly lowering the boiler
water level in case the feedwater regulator becomes deranged or hand regulation
has been careless. Blowdown valves fall into three classifications: (1) seatless
plunger type valves, (2) quarter-turn cock and shutter valves, and (3) seat and
disk valves. Most safety codes specify tandem boiler blow-off valves for safety,
and an additional reason for using tandem valves is the added insurance against
leakage, for should a slight leak develop in a single valve, it would quickly in¬
crease in amount and soon represent considerable waste of heat.
The valve nearest the boiler is the guard valve; the other is the normal
blow-off and should be opened last and closed first when blowing down. Where
a slow-opening valve and a quick-opening cock dre used as the tandem com¬
bination, the cock is the guard valve. Only one boiler should be blown down at a
time. The operator should be taught to remain at the valve until it is closed in
order to reduce the danger of forgetting that the valve is open.
A “must” for plant superintendents is to check the blow-off valves of a
boiler for closed position and secure them that way before sending men into a
drum for inspection or repairs. This precaution applies to plants where other
live boilers are connected to the same blow-off header. Failure to do this has
been the cause of fatal accidents.
Control and Regulating Valves. These are for water level, steam pressure,
water flow, etc. Some makes are thermostatically controlled, others mechani¬
cally, as by a float, others by pressure either steam, hydraulic, or pneumatic.
The actuation of the pressure-operated valves is obtained from diaphragm,
sylphon, or loaded piston. Steam pressure reducing valves and regulating valves
are used largely in heating work, but also, to a limited extent, in the power
plant for supplying steam to auxiliaries, heaters, air ejectors, etc., when these
should have steam at less than header pressure.
Pressure-regulating valves are found in great variety because of differing
requirements as to: (1) the fluid—^liquid, gas, or steam; (2) the sensitivity—
close pressure regulation vs. less expensive types having greater variation of
the controlled pressure; (3) high or low ratio of final to initial pressure; (4)
pilot-operated or direct-operated (large diaphragm type); (5) if pilot-operated
may have internal or external pilot valve; (6) single seat or double seat—
double seat being more sensitive, in better balance, but single seat being neces¬
sary for service where there could be possible dead end, i.e., times of no low-
pressure steam flow.
Boiler Outlet Valves. These special valves are of the stop-check or automatic
588 THE PIPING SYSTEM

Leslie Swariwout Leslie

(B) (C)
Fig. 14-12 Representative regulating valves.
(A) Pilot-operated pressure-reducing valve. The regulated pressure (downstream) is ad¬
mitted against the lower side of a diaphragm and is balanced against an adjustable
spring pressure. Diaphragm moves the pilot valve to proper position to let upstream
steam pressure against the piston face in the magnitude required to open the main
valve for that flow of steam that will maintain the desired downstream pressure. Other
manufacturers may build pru^s with separated pilots where the pilot is joined with the
main valve and with downstream pressure through small pipe or tubing.
(B) Direct-acting pressure-reducing valve. '^Direct acting” means that no pilot valve is
used. The regulated pressure, applied against the upper side of the ^ring-loaded dia¬
phragm, directly actuates the main valve. This is a double-seated, spring-opening valve.
(C) This temperature regulating valve has an internal pilot regulating the fraction of
upstream pressure placed against the upper surface of the diaphragm, the lower side
of which receives spring-assisted downstream pressure. This diaphragm regulates the
S
osition of the main valve. The pilot valve position is adjusted by a thermostat
iaphragm which balances gas or vapor pressure set up in the thermostat bulb against
an adjustable spring pressure.

nonreturn type so as to prevent one of boilers in parallel receiving backflow

from the others should its pressure become substandard. Fig. 14-11 shows a
frequently used type of stop and check valve. With the stem in the position
shown the valve acts as a one-way, or nonreturn, valve, steam flow being per¬
mitted in the direction shown by the arrows. When the stem is screwed down it
holds the valve tightly against its seat, .thus completely shutting off the flow of
steam until such time as the stem is again screwed up, releasing the valve. Such
valves are not used for regulating purposes.
Valve Accessories. Among the equipment that could be so classified are (1)
by-passes, (2) mechanical operators, and (3) motor drives. Large gate valves
should have small by-passes by means of which the pressures on the two sides
of the gate can be equalized before it is opened. Globe valves are often modified
so as to be balanced automatically in all positions, closed or open.
Valves high off the operating floor, if not too large, can bc opened and closed
by hand chain. The valve is mounted with the stem horizontal. Handwheel is
replaced with sprocket wheel over which is placed a loop of chain extending to
the operator’s reach.
BOILER SAFETY VALVES 689
Valves are sometimes equipped with hydraulic or electrical opening and
closing units. On some lines this is done for the sake of convenience on accotmt
of the large size of the valve or its inaccessibility, but on high-pressure steam
lines electric motor control is to enable the operator to stop the flow from a re¬
mote control station. With the increasing application of high pressures and
superheats in power stations, the need for safe and positive means for controlling
the operation of steam lines and headers is emphasized. Central stations offer
many opportunities for the application of motor-operated valves for both shut¬
down and sectionalizing service.
Provision is made for hand operation, if necessary. Control systems for
motor-operated valves are worked out in considerable detail. Most control
systems permit the operation of the valve from one or two main control stations,
and emergency closing from a number of emergency points. Indicating lamps
are used at the control stations to indicate position of the valves, i.e., closed or
open.
14-7 Boiler Safety Valves. The selection and installation of these special
valves are largely governed by the ASME Boiler Code. Conformance with this
code is a requirement imposed by the boiler insurance companies so that,
whether there is a local safety code covering this matter or not, the owner who
insures his property against the hazard of boiler explosion must comply. For
that reason significant paragraphs of the ASME Code are here summarized.
1. All power boilers must have a safety valve. All power boilers having 46 m*
or more heating surface must have at least two safety valves on the boiler drum and
at least one on the superheater outlet header.
2. Safety valves must be mounted close to the boiler drum, with stem vertical.
3. If mounted singly on the drum, the smaller safety valve must have at least half
the relieving capacity of the larger. If mounted on the arms of a Y branch fitting, both
valves must be of the same capacity, though one may be set for a lower pressure than
the other.
4. No steam outlet connection is permitted on a safety valve nozzle.
5. If there is a superheater, 75% of the total relieving capacity for the steam gen¬
erator must be in the drum valves.
6. Setting of the highest-set safety valve on the boiler shall not exceed 103% of
the design pressure of the steam generator. The range of pressure (pop-off) between
lowest and highest set valve shall not exceed 10% of the set value of the highest
pressure valve.
7. Discharge capacity. Manufacturers’ rating of a valve is set at 90% of its actual
test steam discharge, in kg per hr, this to be secured without inlet pressure rising more
than 3% above popping pressure. Let this be IFb.v. Also let = Total required re¬
lieving capacity for the boiler, leg per hr.

Wnut « Maximum rated steam generating capacity, kg per hr.

Ai n Total heating surface of boiler (excluding water wall) exposed to products of
combustion,
Af « Total heating surface of water walls and their headers exposed to products of
combustion, m*.
Then Wv ^ WmBx is one Code criterion for relieving capacity. Another is
Wr Cjiii -f* C%At (14-4)
690 THE PIPING SYSTEM
The larger Wr so determined governs the choice.
In Eq 14-4, C is a constant with these values:

Stoker Burner
Firing by
r, c. Cl c,
Water tube boiler 89.06 68.50 48.82 78.12
Fire tube boiler 34.18 30.06 48.82 68.85

After Wt is determined, the safety valve sizes can be selected from manu¬
facturers’ rating tables, in accordance with the requirement, consistent with
above Code specifications.
Wr g (14-5)
In case more than one safety valve is used, the smaller one can be set to pop
at the desired maximum pressure and the larger at 0.35 to 0.7 kg/cm* higher. Since
the safety valve blows down a few kg
cm*ga pressure before closing, a smaller
“vernier'^safety valve giving less press¬
ure drop between pop and close is in¬
stalled, usually on the superheater
outlet, though sometimes on the boiler
lead, for the purpose of giving partial
relief to the high pressure, warning the
attendants of high pressure, prevent¬
ing overheating of superheater tubes,
and possibly forestalling popping of
the main safety valves and the result¬
ant waste of high potential heat.

Example 1: This example will illustrate the selection and choice of setting of safety
valves, given operating conditions, and the following line of safety valves.

Size, mm : 88,1x60.8 60.8 x 76.2 63.6x101.6 76.2x101.6 101.6x162.4

Rating, kg per hr,

at 7.08 l^/omi ga: 780 1932 8125 4418 11,575

Data : Boiler design pressure 11.88 kg/om* g», operating (drum) pressure 7.08 kg/om^ ga,
water tube boiler, stoker-fired. Heating surfiMe. 269.4 mt, water wall, 46.45 m*. Maximum
rate of evaporation, 8165 kg per hr. Assume 3 valves are to be used.

From the above, WIomx ^ 0165 and Wr = 8165 ; alternately, Wr » 89.06^69.4^

46.45)•|-58.59 x 46.45»ll,480. Use Wr = 11,480 kg per hr smce Eq 14-4 gives the
greater value.

If the low-set valve is set to open at 7.88 kg/ozn*, then the highest valve is limited to
BOILER SAFETY VALVES Ml
•pproximAtoly 10%, or 0.74 kg/om^ higher. From this the settings nre speoifted nt 7.88^ 7.7fii,
and 8.12 kg/otn* ga, which oomplies with all the Code reqairements heretofore mentionedL
The following combinations from the available valves provide the Requisite capacity,
but only B meets the specification, item 3, quoted before.
A. One 38.1 y 50.8 mm set for 7.38 kg/om3, one 101.6 x 152.4 mm set for 7.75 kg/oiii9.
Total TTg.v. = 12*314 ka per hr.
B. One 63.5 X 101.6 mm set for 7.38 kg/om* one 76.2 x 101.6 mm set for7.75 kg/om* and
one 76.2x101.6 mm set for 8.12 kg/em*.
Total W^g.v. ^ 11,961 kg per hr.
Example 2: Assume that the unit of Ex 1 now has a superheater for producing
steam of 98.3^0 superheat, all other data remaining the same. Superheater pressure
loss at full rating is 0.35 kg/om*. A complement of safety valves is to be specified.
Since the superheater valve usually is given the lowest setting, it probably should be
set between 6.89 and 7.17 kg/om* (superheater header pressure is 6.68 kg/om*) if we want to
use 7.38 kg/om* on the low-set drum valve, for the difference between settings on super*
heater and low-set drum valves should be about 0.35 kg/om*. Hence 7.03 kg/om> is taken
as the setting. Proceeding on the assumption of two drum valves, one would be set at
7.38 kg/cm< and the other not to exceed 103% of 11.88 kg/om*, but also not to produce
more than 10% difference between highest *and lowest. Ton percent of 7.75 kg/em* is
0.78 kg/om< ; hence, low and high settings of 7.03 and 7.75 kg/cm* come within this limit.
Drum valves must total 0.75x11,430 »8573 kg per hr capacity. This indicates two
76.2x101.6 mm valves; total TTg.v. -•8836kg per hr. Steam rating of the superheater
valve is 11430—8836=2594 kg per hr. Manufacturers issue correction factors which
indicate the relative discharge capacity if used with superheat. For the valves used in
this and Ex 1, the manufacturer's correction for 7.03 kg/cm*, 93.3^, is 0.90. The equiva*
lent saturated steam capacity required of the superheater valve is therefore 2594/0.90, or
2882 kg per hr (63*5 x 101*6 mm valve)

The safety valve specification is s One 63.5 x 101.6 mm valve on the superheater, set
for 7.08 kg/om* ; one 76.2x101.6 mm drum vidve set for 7.38 kg/om^; one 76.2x101,6 am
drum valve set for 7.76 kg/cm*. 1f,.v.»3125 x.90.f.4418 +4418«11.649 kg per hr. (Note :
Discharge ratings at 7.03 kg/om^ ga have been used but were discharge tables available*
the actual settings the valves would determine their rating.)
The safety valves are sometimes arranged to discharge upward in the boiler
room. However, the noise, humidity, and possibility of scalding someone inad¬
vertently near the discharge when the valve opens are good reasons for convey¬
ing the discharge above the roof in a pipe; in fact, open discharge ought not to
be used for any boiler operating above 3.5 kg/cm*.
Safety valve discharge piping should be independently supported and be
free of contact with the safety valve discharge nozzle. It should project through
flat roofs far enough to eliminate the possibility of the unwary looking into it
and must be open (no rain hoods). Rain and condensation can be caught by a
drip elbow fitting on the discharge nozzle, which fitting is piped for drainage.
There must be no chance that part of the discharge piping weight could come
to rest on the drip elbow, for that could strain the valve body. Even if not over¬
strained, this can cause the valve to increase its blowdown range.
Discharge piping should be straight if possible; at least right-angle bends
should be avoided, and where bends are necessary special lateral bracing is
needed to withstand steam thrust. Preferably each valve’s discharge should be
piped separately; however, a manifold is often necessaiy. Its ai'ea should be
fully equal to the sum of all inlets into the manifold. The designer must con-
 THE PIPING SYSTEM
sider expansion of discharge piping as the steam hits it and provide the necessary
flexibility and clearance that will prevent stressing the valve.
14-8 Thermal Expansion. Pipes have the property possessed by materials
of expanding with increase of temperature. If they were constrained to a fixed
length, a reaction equivalent to the force required to compress the pipe through
a deformation equal to the prevented expansion would be set up. For all but
very short steam lines this force is too large to incorporate in the piping system.
The same force would be present, theoretically, in the short line, but the supports
would have enough elasticity to take the small expansion. In long lines the ex¬
pansion is permitted by the use of suitable joints or bends. The expansion of a
straight run of pipe can be treated as pure linear expansion resulting from
temperature change. The following equation is applicable to problems of ex¬
pansion.
AL = KLM (14-6)
where AL = Change (m) in length of a pipe, L m long.
At = Temperature change, ®C.
K — A constant.* Values to use for K are:
Cast iron, 10.62 X
Wrought iron, 12.42 X 10^
Carbon steel, 12.6 X 10~®
Alloy steel, 12.96 X 10^
Brass, 17.46X10^
Esample 1: What is the extensioa of a straight run of Sted pipe SBJSi m long 7 Installa¬
tion temperature 4.4**0, normal operation on steam at 8.70 kg/om* ga, saturated steam.
The saturation temperature = 176.7*Cf a*—176.7—4 4>■172.3^
A1^»0.0000126x172.3 x 22.86-0.0496 m-4.96 om.
Podbed expansion joints are used up to 17.6 kg/cm* saturated steam. High
temperature has a deteriorating effect on packing; however, packed joints have
been designed for high temperature by protecting the packing by air-cooled

Fig. 14-14 Packed slip joints for expansion. (Single-end joint shown for welded con¬
nection; double-end joint is flang^.)
* There is considerable variation in the coefficient of expansion reported from different
sources. Probably the data of Holbum and Day, Smithsonian Physiccl Tables, 7th Ed., will
be best for vexy accurate calculations of expansion.
THERMAL EXPANSION 608
sleeves. Expansion joints take up expansion at one point by allowing relative
motion of the two sections of pipe connected by the joint. Usually one pipe end
is anchored by a rigid connection to the body of the joint, but occasionally the
double slip joint in which both pipe ends are free to move in the joint is used.
Fig. 14-14 illustrates the packed expansion joint.
Low-pressure expansion joints rely on flexibility of a bellowslike section
or the slight movement of screwed pipe threads in the fittings, which is the
action in a swing joint These are shown in Fig. 14-15. Copper bellows joints are
often used to connect large low-pressure conduits such as exhaust piping, turbine
to condenser, etc.
When expansion is to be taken by the flexibility of the pipe itself, various
forms of pipe bends as illustrated in Fig. 14-16 may be used. This way of caring
for expansion is free of the temperature-pres¬
sure limitations of the expansion joints and o
also of any maintenance work such as the
repacking of joints. Consequently, it has been
the standard for boiler and turbine leads and
for long runs of high-pressure piping of all
sorts. Its principal drawbacks are the added
friction losses, the expense of fabrication (most
bends are special jobs), and the space required. Fig. 14-15 Low-pressure expan¬
Originally these bends were designed by em¬ sion joints.
pirical rules, but more recently the progress in
this field has led to rational methods of stress calculation in preference to rules
of thumb. A stress-strain study of pipe bends is important because:
1. The reaction of the pipe on turbines, boilers, headers, etc., must be known
and kept within bounds.
2. The anchorages should be calculated from known reactions.
3. The safe stress of the pipe metal should not be exceeded m the bends.
4. The flanged connections of the bends to tangents should not transmit
moments large enough to spring leaks or damage joints.
The use of higher steam pressures and temperatures has complicated the
design of expansion bands. Heavier, stiffer pipe is used, while the higher tem¬
peratures produce more expansion and decrease the elastic strength of metal.
Loops in larger diameter pipe are quite expensive. This warrants the application
of considerable engineering to the economic design of expansion bends. The
design problem is primarily one of stress in a deflected structural member, but
a number of conditions peculiar to this field (such as the fattening of the pipe
bend to an oval shape while subject to expansion forces) make the mathematical
analysis highly complicated. Investigations have been made and special methods
of analysis have been evolved * which, however, are set forth in a series of
charts or tables for which space is not available in this book. Several pipe

W. Shipman, 'ODesdgn of Steam Piping to Care for Expansion,’' Tram. A8ME, Vol.
61, No. 62.
F. Peiter and M. J. Fish, ^^Method for Determination of Reactions and Stresses in Ex¬
pansion Pipe Bends,” Combustion, December 1938.
Crocker, Piping Handbook.
Wirt, Smith, and Cope, Manual for Design of Piping for FlextbUUy by Use of Graphs.
 THE PIPING SYSTEM

Double Offset
Exponsion U Bend

'f 1/^ i\ 1

Exponsion U Bend

Double Offset
UBend
—
T T
Advisable Mimmum
Size of Pipe R R Length of Length of
Inside Advisable Minimum Radius to which Tangent, or Tangent, or
Diameter Radius of Pipe Can Be Bent, m Strai^t Pipe Straight Pipe
mm Bends • on Ends of on Ends of
m Extra Heavy Benda^m Bends t m
Standard Pipe Pipe
UX 0.981 0.254 0.178 0.305 0.076
fU 0.457 0.306 O.S0I 0.305 0.076
Mi 0.533 0.356 ojm 0.305 0.00
0.610 0.305 0.38 0.10
m 0.762 <L508 0.381 0.405 0.130
IM.# 0.014 0.610 0.457 0.455 0.16
iOU 1.310 0.813 0^10 0.51 0.30
tH 1.524 1.016 03U 0.61 0.30
aoli 1.820 1.373 1.210 0.61 0^

Fig. 14-16 Fabricated pipe bends

Crane Co,

Fio. 14-17 Three methods of laying out expansion U bends.

THERMAL EXPANSION 596
fabricating manufacturers have excellent engineering sections in their catalogues
covering this same problem. The reader is referred io this literature for equa¬
tions and methods which are beyond the scope of this chapter. When a consider¬
ation of the expansion problem by these more involved methods is not war¬
ranted, the dimensions of an expansion bend can be determined from Table
14-4.
Cold spring is the initial reversed stress applied to piping during erection by
cutting the lengths a little short and joining them by tensile forces. When it is
later heated, the pipe will overcome the cold spring and go into the normal com¬
pressive and bending stresses expected from thermal expansion. Any initial cold
spring, up to half the computed expansion, is helpful in reducing thermal stress.
Example 2: Illustrating the layout of a double offset expansion U bend to accom¬
modate 88.9 mm expansion in a Schedule 40,101.6 mm steel pipe carrying 7.03 kg/om* steam»
it is considered that the use of Table 14^ is reasonable in this case.
First divide the expansion by 5 according to the footnote of the table. This gives

Table 14-4. EXPANSION CARED FOR BY WROUGHT STEEL

QUARTER BENDS- mm.
(The Crane Co.)

The above values, based on actual tests, are recommended as allowing a good safety
factor.
Cold springing has not been taken into account in preparing this tAhle.
The above values are for 90* bends. For **U” bends, multiply by 2. For single offset
bends or expansion **U” bends, multiply by 4. For double offset expansion bends or
circle bends, multiply by 5.
 596 THE PIPING SYSTEM
17.8 mm. Use this in the table for a 101.6 mm pipe. By interpolation, the bend radius is
0.864 m. Cheeking this against Fig. 14-16 it is seen to be above the minimum.*
A tangent length of 0.381 m is selected, then the bend oan belaid out by one of the
methods of Fig. 14-17.
Expansion joints are now being made in welded piping by building up a
flexible loop such as that shown in Fig. 14-18, using standard turns manufactured
especially for welded piping. The chart is based upon a flexure theoryf and is
constructed so that the maximum fiber stress in the loop, due to bending, will be
approximately 1125 kg/cm*. This value is conservative for low carbon steel to
232® C but should be decreased at higher temperatures.
Since fiber stress is proportional to expansion, the chart may be used for
different maximum fiber stress by making a suitable correction of expansion.

Welded expan&ion loop

5 6 7 8 9 10 1214161820 25 30SS404S506070 100 200 300 400 500 000

B-in Pipe Diameters Z.-in Pipe Oitmetifi
Ttibe Tum$ Inc,

Fig. 14-18 Welded expansion loop design chart.

*The ^'advisable*’ radius in Fig. 14-16 is used in cases where expansion is not the
governing quantity.
t Crocker and McCutchan, “Frictional Resistance and Flexibility of Seamless-Tube
Fittings Used in Pipe Welding,” ASME, FSP paper, 53-17, December 1930.
HEAT INSULATION 697
For example, an expansion of 100 mm in a 152.4 mm pipe line at 482.2*^0,,for
which maximum fiber stress is taken to be 843.7 kg/cm*, would be entered on
the chart at ^ X —- «s0.89 pipe diameter.
152.4 843.7
The lengths W and H can be limited to any construction requirement, or W
even reduced to 0 by use of a return bend. When using the chart, for the solution
of an expansion problem, the amount of expansion is first calculated by equa¬
tion. This requires knowing the distance between anchor points, the tempera¬
tures, and the material of the pipe. Either IF or is then arbitrarily selected,
depending on the limiting conditions of construction. The one not arbitrarily
selected can then be found from the chart. All dimensions in the chart are ex¬
pressed in nominal pipe sizes.
Example 3: Design a welded expansion loop for the pipe line of Ex 1, when W is
limited to 1.22 m and the pipe size is 1.52.4 mm. First, all dimensions must be converted
into pipe diameters, sinoe all dimensions on the chart are so expressed.

,
Nominal pipe size is
22.86
L = (Pjg2 ~
152*4 mm

diameters.

Expansion = = 0.32 pipe diameter.

15.2
1 22
W= = 8 pipe diameters.

From the intersection of TF = 8 and L = 150 on the chart, project horizontally to 0.32
expansion, then vertically downward, reading H from the horizontal axis, if = 15 pipe
diameters. Hence H = 15x0.152«i2.28 m .

14-9 Heat Insulation. Bare pipes carrying steam or water at temperatures

considerably above atmospheric lose much heat to the atmosphere. Several
experimenters have determined the loss of
heat from bare pipe. Heilman^s data is
shown in Fig. 14-19. The calories per hr loss
from bare pipe may not, on first thought,
seem to amount to much, but if it be
remembered that this loss is suffered
throughout the year, and that, in the case
of the boiler leads at least, the calories so
lost are high potential heat and therefore
more valuable than the average kcal in a kg
of steam, it will be understood wdiy prac¬
tically every hot pipe in the modern power
plant is covered. Cold pipes are also in¬
Ttmptroturc difftrcncc, cylinder surface
sulated to keep heat out, but tliough to ombient air, deg C
insulation for this service is common in
Fig. 14-19 Coefficient of heat trans¬
refrigeration plants, it is rare in the power
fer for the discharge of heat from the
plant. By keeping heat in the hot lines not surface of horizontal iron cylinders.
only is there a conservation of calories which
have at considerable expense and trouble been transferred to the fluid, but also
there is the avoidance of an uncomfortably overheated atmosphere in the
698 THE PIPING SYSTEM
vicinity of the pipe. Besides the pipe itself, fittings, valves, ducts, boiler drums,
tanks and heaters are insulated.
A good pipe covering should, of course, be non-conducting. Commercial
insulations are materials of very low conductivity. Insulation should be able
permanently to withstand the temperature to which it will be subjected; that is,
it should be stable and resist deterioration over the working life of the pipe. It
should be easily molded and applied and have the requisite mechanical strength.
No insulation commercially procurable will overload the pipe by its dead weight.

Fig. 14-20 Insulation details. Pipe with double layer of insulation and canvas jacket.
Insulation of a high-temperature valve

for density is not one of the attributes of a good insulator. In fact, the non-con¬
ducting properties seem chiefly to be derived from the presence of large numbers
of air cells. The materials most commonly used are asbestos, “magnesia” (car¬
bonate), cork, hair felt, wool felt, rock wool, and diatomaceous earths. Most
commercial insulations are either built up from corrugated asbestos paper, or
laminated asbestos paper artificially roughened to produce air spaces, or are
molded, or felted with asbestos, or blanketed. A very common and effective
insulation for temperatures upto315®Cisthe molded “85% magnesia,” so called

Table 14-5. TRADE STANDARDS FOR 85% MAGNESIA

SECTIONAL PIPE INSULATION

Double Double
Pipe Standard Standard Pipe Standard Standard
size thick thick size thick thick
( mm.) ( mm.) ( mm.) ( mm.) (mm.) ( mm.)

12,7 22 49 102.4 29 6»
19.1 22 49 177.8 83 64
25.4 22 49 208.8 32 64
SIJ 22 49 228.6 32 64
S8.1 22 49 254 82 64
00.8 26 65 804.8 88 76
88.6 26 55 355.6 38 76
76.8 26 55 406.4 88 76
88.9 26 57 457.2 88 76
101.6 29 57 506 88 76
114.8 29 59 609.6 88 76
U7 29 59 762 88 76
HEAT INSULATION 599
because it is 85% carbonate of magnesium and 15% binder. Pipe insulation for
higher temperatures should have an inner layer of some special high-tempera-
tiire insulation, since 85% magnesia alone will (let(‘rioral(‘. Standard insulating
sections of magnesia are 3 ft in length, held together by canvas jackets and
metal bands. Fittings and flanges are covered by specially molded forms or by
plastic insulating cement. Sometimes the flanges are left bare, but that practice
wastes so much heat on superheated steam lines that flanges also are covered.
Magnesia insulation has been the standard for medium temperature sec¬
tional insulation, but is being challenged by several excellent competitors. The

0-15

cn
*8
^ 0*12

; 0-09

> 0*06

> 0*03

0-0

Curve Serv, limit Material

hot face,
1 .. . -.688 . ... Felted asbestos fiber
2. ... 149. ... Corrugated asbestos paper
3. .... 538 or. ... Glass wool in blanket form
315 if Glass wool molded with thermosetting binder
4. .... 816 . ... Diatomaceous earth bonded with asbestos hbre
6..315. ... 85% magnesia molded sectional insulation
6.. 638 . ... Hydrous calcium silicate bonded with asbestos fiber
Fio. 14-21 Conductivity of some commercial pipe insulating materials. (As reported
by manufacturers of same.)

trade practice of firms supplying 85% magnesia preformed pipe insulation is to

mold it to Standard and Double Standard thickness. Other molded pipe insula¬
tion is generally offered in thickness varying by 127 mm increments from 25.4
mm to 101.6 mm and in sections 0.61 or 0.91 m long.
Molded sectional insulation is applied to the pipe with wire loops or by past¬
ing a flap of lightweight cloth which backs the insulation. This then receives a
cover treatment according to the ideas of the plant designer or owner. Medium-
duty pipe covering for a power plant could be specified as follows:
^^Apply 85% magnesia sectional insulation of Standard thickness tightly
butted together, each section being held in place by not less than three loops of
1.29 mm iron wire. Enclose the insulation in a jacket of 8-canvas, sewed over
 THE PIPING SYSTEM
resin-sized paper. Paint the canvas with one coat of glue sizing, after which
paint it with two coat'^ of first quality lead and oil paint of specified color.’’
In some cases flanges and valve bodies are covered; in others they are left
bare for maintenance convenience. Supply firms manufacture several grades of
insulating cement for the purpose of molding in place an insulation over irregu¬
lar surface. These can be finished with pasted canvas or troweled with hard-
finishing insulating cement.
Blanket insulation, such as rock wool, comes in a roll, sandwiched between
mesh reinforcement to give it body. It is cut to fit the pipe, wrapped snugly on
and ^\iIed in place. After that it receives a finish of plastic insulating cement
and a cover to suit.
Manufacturers publish ^^insulation efficiency” tables for the various standard
thicknesses of their different grades of insulation. Such tables usually present
percent efficiency against two variables: pipe size and temperature difference
— to- ^^efficiency” of an insulating material is expressed as the percent
heat saved by using the insulation, compared to what would have been lost had
the pipe been left bare.
Like many other powder plant design problems, the amount of insulation to
apply is, basically, an economic problem. The cost of the covering must be
weighed against the saving of heat. The economic studies are tedious because
of the large number of variables, the effects of which must be included in the
analysis. The chart of Fig. 14-22 shows what these variables are and arranges
them so that, if they are known, or can be estimated, the economic insulation
thickness can be quickly found.
Heat Loss. Transfer of heat through pipe insulation is a case of conduction
through a curved wall and the discharge of heat by convection and radiation
fiom the outside suiface. The inner surface is presumed to be at the same
t(*mperature as the fluid in the pijie. The final temperature is that of the ambient
atmosphere. The equation for heat loss, derived from Eq 9-9, is

(14-7)

An empirical equation* for coeflScient of heat transfer from the usual outer
surface, a canvas jacket, is
fc«i761.3/£D-"(15l.4-»)] kcal per hr-in*-deg C (14r8)
/„ <o = Inner and outer surface temperatures of the insulation, *C,
U = Ambient atmospheric temperature, ®C,

e ^ to-U.
ri, r% = Inner and outer radii of the insulation, m.
D = Outer diameter of pipe covering, m.
h = Conductivity of insulation, kcal per hr-m*dcg G.
*“Heat Losses from Bare and Covered Wrought Iron Pipe at Temperatures up to
4m}(800* F),” Tran». A8ME. 1922.
HEAT INSULATION 601

Fta. 14-22 Chart for economical thickness of insulation. (From Heat Iiutdatinft
Practice in the Modem Steam Plant, L.B. McMillan, Mechgnicdl Engineering, May,
602 THE PIPING SYSTEM
Eq 14-8 applies to insulation located in “stiir* air; that is, the convection
is produced only through the natural levity of heated air.
In case there are two layers of insulation of differing conductivities, the
denominator of Eq 14-7 becomes

{ki is for the inner layer, k2 the outer, while n is at to^.)

Example 1 : An 203.2 mm pipe carrying steam at 282.2^0 through an atmosphere at
32.2^0 should be heat insulated. Fig. 14-22 is to be used to estimate the eoonomio thickness
of a type 5 (Fig. 14-21) pipe covering whose cost is 49. ^ per board metre. Daia ; opera¬
tion, 7200 hr per year ; cost of heat, 168.74:per 10* koal; fixed charges at 20%,
An approximation of li from Fig. 14-21 would be 0.066. Now enter the
chart at 7200 hr, and proceed graphically to 1.587, to 250°, to 0.402, to 20%, and finally to
203.2 m^ pip® sise, where the required insulation thickness is found to be 76.2 mm.

Example 2: The heat loss per metre of pipe, and the insulation efficiency will be
calculated for the case in Ex 1.
An assumption for will be necessary. After some preliminary trials, this is taken
to be 64.4°C. Then mean temperature of the insulation is (54.4+282.2)/2 = 168.3°0, at
which k = 0.066.
A 203.2 mm pipe is 219.08 mm OD, so Z>as219.08*f 2x76.2» 371.48 mm; r|al85.7 mm •
ri»109.6 mm. 6«:64.4—32.2i-22.2°C.
h=-761.3/[0.37150d9(i6i.4_22.2)]«7-128 kcal per hr.m*-deg C

282.2-32.2
Q = 164.1 koal per hr-m*
1 / 0.1867 \ 1
0.066 \ ■■■ ■ ■ 0.1096 / 7.128
Verifying the assumption of to =64.4°; q = 7.128x22.2^168.2 discharged from cover.
while
282.2-54.4
»163.3

0.066 0.1095

transmitted through the covering. These are in close enough agreement.

Heat loss per metre3sl63.3xxx0.3715=sl78.7 kcal per hr.
Coefficient for bare pipe at 6 = 250°C is 20.75.
Heat loss per metre bare pipe * 20.76 x 260xxx0.2191ai3670.3 koal per hr.
^ ^ . 3570.3-178.7
Insulation efficiency = —- —0.953.
So70.3

The principles of pipe insulation apply also to the insulation of tanks, such
as feedwater heaters, storage, etc. Much the same materials are used as for
pipe covering. Method of application may be somewhat different, as the binding
of the insulation closely to the tank may be more of a problem, especially under
the belly of the tank, where the insulation will have the tendency to sag away
from the tank surface, due to its own weight. Lacing wire attached to welded-on
clips or studs and tied in with circumferentially applied wire netting will make
a permanently attached insulating jacket. Tanks may be covered with molded
DETERMINING PIPE SIZE 603
segmental blocks, narrow rectangular blocks, or blanket insulation. The final
covering is usually hard finish insulating cement, and some prefer to apply a
cover of canvas over that.
A development in the insulation field is the use of reflective insulation. This
is composed of a series of aluminum alloy sheets, separated at intervals by
spaced blocks providing multiple dead air spaces. Advantages claimed are
relative cleanliness during construction,* expansion without cracking the sur¬
face, and smooth, clean finished appearance without paint. This has been used
throughout a power station, i.e., to cover pipes, tanks, ducts, and breechings.
14-10 Determining Pipe Size. Most of the problems of this nature arising
in power plant design will involve pipes carrying only water or steam; hence
this section is devoted mainly to problems of water and steam lines. Engineers
will appreciate that pipe size is not determined alone by weight or volume to be
transported. For instance, there is no one size that mwi be selected to carry
15m* steam per min. At 1500 m/min velocity a pipe of 0.01 m* cross-sectional
area is required ; at 3000 m/min it is 0.005 m*. But when both velocity and
flow volume are specified, the pipe size is fixed, according to the relation
Volume = Velocity X cross-sectional area
The law of continuity is more useful than the above-mentioned relation, for
mass flow is a constant anywhere in a pipe having no side-branched flows.
Weight flowing = Velocity X area X density (14-9)
To determine the correct size for a pipe of relatively short length and moder¬
ate cost (typical of interior plant piping) knowing the quantity of fluid to be
carried, the usual practice is to assume a velocity based on satisfactory prior
experience and to determine the area from Eq 14-9. Diameters can then be cal¬
culated and the pipe of the nearest or, in some cases, the next larger standard
size selected. Table 14-6 is furnished as a guide to representative practice in
flow velocities. Having so determined the pipe size, resulting friction losses can
be computed by methods to follow. In some cases a predetermined maximum al¬
lowable friction loss sets the pipe size rather than average flow practice.
Table 14-6. AVERAGE PRACTICE IN FLOW VELOCITIES, M/MIN
General water flow 61-152
Plant water services (misc) 91-183
Pump lines, discharge 91-183
Pump lines, suction 61-158
Boiler feed lines 152.183
High-pressure saturated steam 1829-3048
Low-pressure saturated steam 1219-1829
High-pressure superheated steam 8048-4572
High-vacuum exhaust steam 6096-7315

Long runs of pipe, or shorter ones of expensive, thick-walled, alloy steel

high-pressure pipe may create the need for an economic analysis of optimum
pipe size. The higher the velocity, the smaller the required size of pipe, but un-
♦ Insulators tend to shower a plant interior with dust and spots from cutting insulation
and applying plastic cement.
604 THE PIPING SYSTEM
fortunately the friction loss increases about as the square of the velocity. In an
economic analysis a search is made for that pipe size for which the annual fixed
cost, including insulation if used, plus the annual operating cost resulting from
friction loss is a minimum. Power plant pipe sizing does not often warrant a
least-cost economic analysis.

1: The size of Schedule 40 pipe which will carry 5443 kg steam per hr
at 8.79 kg/om* ga, 0.97 dry, will be found with the aid of the table of average flow velocity.
This steam has a specific volume of 0.1947 m> per kg. Cross-sectional area for the
pipe, assuming velocity of 2135m/min is

A = 5443 x 0.1947/60 x 2135s0.008258m>=82 58om*.

The mside diameter of a 101.6 mm pipe ie

I14.30-'2x6.02»89.56mm. Area = 82.13 em*.

This IS obviously the ncarebt pipe size, for the velocity is almost exactly 2136 m/min.

One of the most thoroughly investigated subjects in engineering is the fric¬

tion of fluid flow in pipes. Formulae are sufficiently numerous to bewilder the
beginner, and dozens of manufacturers^ catalogues carry data sections wherein
are excellent tables or charts of friction factors. Most practicing engineers have
a favorite, well-thumbed source of such data, usually sufficiently extensive to
obviate any need for calculation on their part. With that practice this author
has no quarrel, but he cannot find the space in this volume to include this hand¬
book feature. Furthermore, a few basic calculations practiced by the neophyte
will bring to light the influential factors in fluid friction. Having mastered these
then let him locate the short-cut tables or charts to suit. Here we intend to sub¬
mit a basic equation for friction of liquid flow, and one for steam flow.
The following symbols are employed:
H = Friction head, m of the fluid.
Ap = Friction loss, kg/cm*
/ = Coefficient of friction.
L = Pipe length, m
V = Flow velocity, m/scc.
D = Internal diameter of pipe, m.
z = Fluid viscosity, centipoises.
d = Density of the fluid, kg/m*.
8 = Specific gravity, referred to water.
g = 9.81 m per sec*.
For the flow of liquids:
H = 2fLV^/gDm (14-10)
DETERMINING PIPE SIZE 605
For the flow of steam or air:
Ap = /LdFV5000aDkg/cm* (14-11)
These are two comparatively simple equations, but each requires implemen¬
tation with a coefficient of fnction which unfortunately is subject to several
influencing factors. Since it is not a constant, investigators have been at some
pains to express it as a function of the controlling quantities, and therein has
arisen the multiplicity of formula offerings.
Table 14-7. FRICTION FACTORS, /, FOR WATER FLOW
IN IRON AND STEEL PIPE

New Pipe Old Pipe

Diameter of 0.914 m/4ea 1.829 m/sao 3.048 m/Mo 1.829 m/sec 3.948 m/seo
Pipe, mm (in) (3fp8) (6 fps) (10 fpB) (6fpe) (10 fpe)

26.4 (1) 0.0075 0.0061 0.0099 0.0091

60.8 (2) 0.0071 0.0059 0.0097 0.0088
76.2(3) 0.0070 0.0058 0.0096
101.6(4) 0.0066 0.0057 0.0093
127 (5) 0.0064 0.0056 0.0090 0.0084
162.4(6) 0.0063 0.0055 0.0085 0.0082
202.2 (8) 0.0060 0.0054 0.0082
264 (10) 0.0058 0.0081
304.8(12) 0.0056 0.0078 BHI
In the field of liquid flow, investigators have found it necessary to place
fractional exponents on V and D, or to express / as a function of V and D.
Here we follow the latter school. In this field also there are two distinct fric¬
tional cases, i.e., idscous flow and turbulent flow. These require different
coefficients. Only very slow velocities are viscous in water. One criterion for
determining the characteristic of flow is to evaluate z/DVS. A flow is turbulent
if z/DVS < 1163.5; otherwise, it is viscous.
Friction coefficients* for turbulent liquid flow in clean pipe arc:
Copper, brass, and lead. / = 0.00181 + 0.0011787{2;/D7S)® ««« (14-12)
Iron and steel. f = 0.0035 + 0-0007562(«/DFS)o «M 14-13)
For viscous liquid flow in all pipes:

/ - 0000016026 (s/DFS) (14-14)

On account of the frequency of cases of water flow in the 0.914 to 3.048 m/sec
range, a table is included, giving the / factor against F and D.
For the flow of steam and air in pipes: f

/ = 0.0054 + 0.0465 WDVd) (14-15)

To use these equations for the prediction of friction loss in a given piping
configuration, two supplementary studies arc required. The first is the value of
’'‘Wilson, McAdams and Setzler, Journal Industrial and Enomeering Chemistry, Feb¬
ruary 1922.
t McAdams and Sherwood, Mechanical Engineering, October 1996.
606 THE PIPING SYSTEM
viscosity, z] the second, a method of allowing for the effects of fittings and
valves. The first item is supplied by Eq 14-16 and Table 14-8; the second, by
Fig. 14-23.
Viscosity of steam z =0.00894+3*6 X10*(14-16)

in which t = Steam temperature, ® G. Also, see Fig. 9-6.

Table 14-8. VISCOSITIES—CENTIPOISES

Tempera-
ature, Crude Oil Fuel Oil Fuel Oil Lubricating Lubricating
(•0) Water 15 Deg. B 24 Deg. B Bunker C Oil, Turbine Oil, Engine

(1S.9) 1.0 5000 50 100-250 300

(».l) 1.0 3000 40 6600 70-180 200
fM.7) 0.9 1000 30 100
(M.2> 0.8 600 25 80
(VIA) 0.7 450 18 1420 35-75 60
I
(4S.») 0.65 300 17 50
(*«.«> 0.6 200 15 605 21-46 40
(84.4) 0.5 120 12 25
(«0.0) 0.45 90 10 20
(06.6) 0.42 70 8 220 12-24 18
(S3.3) 0.3 8

The following examples will illustrate some of the cases of frictional flow
these formulae will solve.
Bsampl« 2 : The statio head at a certain station on a 101.6 mnu Schedule 40, clean,
iron, cold water pipe is 60.96 m. The static head is to be determined for a station 152.4 m
^rther along the pipe in the direction of flow, and where the pipe is 0.14 m higher than
station 1. Between the stations are three elbows. Flow rate, 13251/min.
The 101.6 mm pipe has a 102.4 mm internal diametei; Bay 2> = 102 mm.
Flow area A = n/4x0.1023»0.007886 m*.
Flow velocity V = 1326/(1O^x 0.007886 x 60)sB.S m/sec.
From Table 14-7, get / = 0.0058, while with Fig. 14-23 determine that pipe length
equivalent to 3 elbows is 10.06 m. Then use Eq 14-10.
H =: 2 x0.0058x(152.4+10.06)x2.8>/9.81x0.l02»14.U8 m

Static head at station 2 = 60.96—9.14—14.08—37.74 m

Example 3: In this example a pipe size is to be detennined such that 668 I/min cold
water flow will not suffer more than 0.069 kg/cm* loss per 30 m of pipe.
Flow volume, Q = 668/10*x 60—0.009467 n^/seo .
H = 0.69 m when L = 30 m • Assume / = 0.006.

V -.a.009467/-^Da_0.0ia05/i)*iii/aeo.

Using Eq 4-10; 0^-SxO.OMxMx(O.OiaOS/i)>)*/».8IZ> .

D » 96 mnuSelect a 101.6 mm. |dpe. D 101 mni •
A more exact value for / is now possible. / « 0.0066.
DETERMINING PIPE SIZE 607
mm ^ *»
The velocity m a I0i.6iiim pipe will be 0.00949T / - X (0402)“** 1.180m/«ae.

// = 2 X 0 0065 X 30 X 1.160 V 0.81 X 0.102 = 0.6281 m

Ap 0.062 kg/em*

Example 4: This is an cximple of the dotermination of steam pipe size needed to

pass a given flov\ without exceeding i specified pressure loss Given flow of |1464^^^
of steam at STl.l^C m a pipe of 80.68 m total equivalent length Average pressure,OftjH
kg/omO ab; Ap limited too.likg/om* Schedule 80 pipe

From tables v =* 0.00608 m^/kg; hence d « 15.894kg/m*.

Flow rate = 146 160 X 0.00608 /3600 « 2.1524 m»/aeo

Fig 14-23 Resistance of valves and fittmgs to flow of fluids. Note. Valve types shown
do not include check valves The resistance of swing checks is approximately the same
as for the eloae return bend, while that of vertical hft check valves is approximated by
the open globe valve.
608 THE PIPING SYSTEM

Flow velocity V = 2.024/3.3407/I>* m/seo.

As an approximation of /, use 0.0054 from Eq 14-15. Then, since Ap*0.14Eq 14-11

becomes
0.14^*0.0054 X 36.58 x 15.394(3.3407/D*)*/5000 X 9.81I>.
D ~ 0.3458 in*i345.8 mm.

D for 355.6 mm OD pipe is 355.6-~(2x 19.05)»317.5 mm.

D for 406.4 mm OD pipe is 406.4—(2 x 21.4) =*363.6 mm .
The 406.4 mm. OD size is selected since it will limit the pressure loss to less than
UJ 4 kg/om*.
V « 3.3407/0.3636<»25.27m/8eo.
The correct / is next calculated, using z =» 0.00894+3.6 x 371.1 x 10^ * 0.0223.
/ = 0.0054+0.0465 x 0.0223/(0.3636 x 25.27x15.394) =30.00541.
Ap 3= 0.00541 X 36.58 X 15.394 x 25.272/(5000 x 9.81 x 0.3636)-0.109 kg/oi^*.

Example 5: The flow of a viscous liquid is illustrated by a calculation of the rate

of oil flow through a 76.2 mm. Schedule 40 iron pipe. Assume a Bimker C oil of 0.962
S.G at 48.9''C to be flowing through pipe of 41.15 m equivalent length, with an expend
able head of 1.62 m of oil and let it be required to find the rate of flow.
Expecting the flow to be viscous in nature, use Eq 14-14.
With a viscosity from Table 14-8, / = 0.000016026 x 605/(0.0762 x 0.962F)=0.1322/F.
0 1322
From Eq 14-10,1.62=*2x ~j^x41.16F*/(9.81 x0.078) ; r=0.107 m/sec.

The viscous condition is now verified: 605/0.078x0.107x0.962>1163.5,

Oil flow = 0.078* X 0.107x1000 x 0.962 x 60 —29,6 kg per min.

14-11 Pipe Support. Steam and water pipes are usually supported inter¬
mittently by hangars, brackets, rollers, etc. This results in the pipe acting as a
continuously loaded beam over several supports. However, the configuration and
support of a pipe line might be such that it bore more resemblance to a beam
with free ends. Thus there are t\^o extremes of end-fixity assumption. Weight
of the pipe, its covering, and its contents will produce a bending stress that may
need calculation and comparison with safe values (Table 14-1). The deflection
between supports may need calculation, for steam lines must be graded down
enough to avoid pockets of condensation.
The average gradient is the span in metres
divided by the difference in elevation at
the supports and is expressed as “1 in
-In general, when condensate
flows with the steam the gradient should
Fig. 14-24 Pipe gradient. not be flatter than 1 in 240. Conden¬
sate flow against the steam should be
avoided if possible, but where necessary the gradient should not be less than
1 in 96, and steam velocities should be held^ to moderate values. Further¬
more, the downstream support needs to be placed below the upstream at
least enough to produce a horizontal tangent or, preferably, a positive angle as
PIPE SUPPORT 609
shown in Fig. 14-24. This criterion will require that V exceed 4y, Most dei^igners
will prefer to accentuate the drainage by making Y equal 6 to 8 times y.
The following equations are for the maximum deflection, y, and the maxi¬
mum beam tensile stress, S, in pipe lines for either free or fixed ends at the
supports.
For free ends: y ~ 5wL*/3SEI cm. (14-17)
also, Si = wL^Do/161 kg/cm*. (14-18)
For continuous spans, y = wL^/(ZMEI) (14-19)
also, Si == wL^Do/241 (14-20)
These are theoretical formulae, based on the elastic flexure theory.*
Symbols represent the following quantities:
w mt Weij^t of pipe, covering, and contents, kg per cm.
L » Distance between supports, cm.
I M Moment of inertia, 0.0491(Do^ — Di^) cm*.
Do, Di » Pipe diameters, cm.
B ■■ Modulus of elasticity of the pipe; kg/cm* at operating temperature.
B * (20.5—0.00760x10® kg/cm* t being*C. Valid for carbon steel up
to315^C, and alloy steel to 427*C. Consult reference sources for
higher temperatures.
A pipe wall may be stressed by the simultaneous action of:

1 . Dead weight deflections. Si of Eqs 14-18 and 14-20.

2 . Internal fluid pressure:

Hoop tension, S2 == pD,/2dt kg/cm* (14-21)

Longitudinal tension, Sz = (14-22)

3. Torsion and bending, S4, arising from the action of pipe bends, and other
configurations, taking up thermal expansion.
Stress of the Si form is not likely to be important where S4 stress predomi¬
nates, for the pipe bends are ordinarily closely and carefully supported. Only
in large pipes will Ss assume an important magnitude, and only in pipes con¬
taining considerable fluid pressure will hoop tension become of moment in stand¬
ard weight pipe. Therefore, except for the use of high-pressure, closely de¬
signed pipe bends, the rational analysis of which has already been set outside
the scope of this book, the principal pipe stresses are Si and S2. These occur
across normal planes and are independently considered (no combined stress
calculations). Each is required to be less than the safe working stress, S. A
determination of schedule number by formula is usually assurance of safety

’*'One series of tests on Schedule 40 pipe indicated that tlic test deflections approached
values ealoulated from Bq 14*17 for pipes smaller than 60.8 mm, and Bq 14-19 for larg^
pipes.
610 THE PIPING SYSTEM
against hoop stress. Only in the case of unusually long spans in low-pressure
piping is >Si liable to exceed 82-
EsmmfUm 1: The pipe epeoinoatioxiB of a 42.2 kg/om* ga, 843^, horiaontal eteam
line are : 152.4 mm ASA 106-A pipe, welded, covered with insulation weighing 0.071 kg/om
run. The pipe wall dimension, support spacing, and drainage gradient are to be found.
Condensate same direction as steam flow.
Table 14-1 gives S = 843.7 for this grade of pipe. Calculated Schedule number =
1000 X 42.2/843.7 = 50. Next determine whether Schedule 40 or 80 must be selected.
The minimum wall thickness for Schedule 40 is 17.11—0.125 x 7.11 = 6.22mm.
Stress, by Eq 14-3 is760 kg/om*: hence, Schedule 40 is strong enough for the internal
pressure.
The next step is to determine whether /Si or y governs the span by comparing com¬
puted Si with /S. A 152.4 mm. Schedule 40 pipe weighs 28.3 kg/m has Do^ 168.28 mm.,
Di —155.84 mm. 7 —1036.6 cm* These properties are based on dt -6.22 mm min. Water
contents = 7 (IMI)* X0.001 0.184kg/em, w * 0.283 + 0.071 + 0.184 — 0.538 kg per
4
om.Pipe is taken to be full of water in calculating span, for conditions might arise that
the pipe became filled. UsTng Eq 14-20,/Si i« 843.7—0.538 X 15,831^/24X1036.61583 cm.
Span L, for Si = iS, is, therefore, 15.28 m
Eq 14-19: y = 0.538 x 152t*/(884 x 1036.61^. After calculating E to be IM X 10*
kg/oih*, 1/ =4.06 am. Minimum 7 » 4 x 4.06 — 16.24om, representing 1528/16.24 or 1 in
93j gradient. A gradient of 1 in 300 is ample for drainage, so obviously deflection, not
bending stress, governs the span.
For drainage gradient calculations have pipe empty, and 6y,w^ 38.8/100 4* 0l071
= 0.884 kg/em . Assume L - 015^1040^ and 1115 om ; then
f 915 *1*/ r 0.348 *1
y »0.354x1 I /(384X 1036.6X 17.9xl0«)=j 0.581 I cm

Y = 6y and average gradient = L/Y =

r 438 1 'r9.15 )
Gradients are, therefore, 1 inJ 298 v for the i
10.4 L m spans.
I 242 J (
11.15 J
This shows that a satisfactory span is 10.4 m, with Y — 3.49 cm.
While the arrangement of supports for plant pipes can be delegated to the
installation foreman by appending to piping drawings some note such as
“Pipe supports to be located in field,” this practice can be recommended only
for small-sized pipes of limited temperature range. Certainly the large and the
hot pipes should receive careful attention as to support, anchorage, expansion,
and drainage, and the drawings should show method of support in sufficient
detail that these important matters are designed in the office and not during
field erection.
Direction of expansion, weights to be supported, convenient points of at¬
taching hangars to building structure, etc., are correlated to provide a plan of
support. Manufacturers have many different types of pipe supports as regular
catalogue equipment, a small sampling of which appears in Fig. 14-25. Points
of anchorage must be sufficiently rigid not to yield under the pipe thrust, while
intermediate supports must be able to accommodate pipe movement either by
long hangar rods, pipe rolls, or slides. Sway bracing and vibration dampers
are required on occasion. Spring or counter-weighted supports will be needed
where the thermal expansion would tend to lift the pipe off a fixed support.
DRIPS AND DRAINS 611

a b C d
Fig. 14-25 Forms of pipe supports, (a) Spring hanger, (b) Bracket-supported anchor
chair, (c) Adjustable roll support, (d) Bracket-supported hangers.

Heavy vertical runs of pipe are supported at the base by anchorage fittings
and at intermediate points by spring suspension hangars.
14-12 Drips and Drains. All steam lines must be adequately drained of
condensation. Even superheated steam lines need drainage since condensation
forms during the warming up period, and while the line is hot but having no
flow. Also, slugs of water may come over from the boiler. Fig. 14-26 shows
methods of caring for the condensation in a high-pressure steam main. The
points to be drained are the low points in the line, separators, drip pockets,
globe valves, and valves in vertical lines. Draining of a high-pressure Steam

Fig. 14-26 Methods of caring for the condensation in high-pressure steam lines. Pipe
drawn to larger scale than floor level equipment.
G12 THE PIPING SYSTEM
line requires that the drainage equipment be able to draw off the condensate
and hold back the steam. This can be done by manually operating valves on
a drip tank, by a steam trap, and by a pump. These methods are also illustrated.
The standard code for pressure piping contains some specifications on the
installation of drips and drains which can be summarized thus:
1. Drip lines to be taken from all points where condensate may collect and
drain lines from points that will drain all water-bearing equipment. Each drip
or drain line to be manually valved.
2 Pressure steam should be drained through traps.
3. Drips from steam regions at different pressures should be drained
through different traps.
4. If several traps discharge into a common header which may have some
pressure, each trap discharge line should have a hand and check valve.
6. Trap discharge lines should be protected against freezing.
6. A point of discharge to the atmosphere should be properly safeguarded
for human protection

Fig. 14-27 Steam line accessories (A) Steam or oil separator. (B) Steam trap.

To collect the condensate from the bottom of steam lines, a drip pocket
made of pipe fittings as shown in Fig. 14-26 is desirable. When located where
dirt and scale are likely to settle, it becomes a sediment pocket and should be
provided with a valved discharge to waste. The steam separator will not only
divert a condensation stream from the piping, but will also remove suspended
drops or slugs of water.
The principles upon which separation is based are: (a) reverse current,
(b) centrifugal force, and (c) wet baffles. The separator should provide for an
enlarged path for the steam since it has been found that separation is more
effective at lower steam velocities.
The simple separator shown in Fig. 14-27A uses a reversing chamber for
gravitational separation. Units with multiple baffles for the steam to flow over
can separate more of the finer drops of water from the steam. However, they
are more expensive, create more line pressure drop, and may be an unnecessary
refinement in many cases. Separators ought to be placed in all prime mover
steam leads.
Exhaust steam lines from engines and reciprocating steam pumps contain
DRIPS AND DRAINS 613
considerable oil as a result of the method of lubrication. If this steam is to be
used in heating systems, to heat feedwater, or for industrial processes, the oil
content is objectionable. An oil separator, built much on the same principles as
the steam separator, should form part of the exhaust line from such equipment.
Steam Traps. Condensate that is to be removed from a steam region where
there is gauge pressure must be passed through some automatic valve that will
let the condensate pass but hold back the steam. The alternate to this is to
pass it through a fixed orifice which, if large enough to carry the maximum
condensate load, will be blowing steam at lower rates of condensation. How¬
ever, this may be satisfactory for draining condensate that is produced only
during warm-ups of pipe lines.
Removal of condensate from a vacuum region by traps is possible only if
there is a region of still greater vacuum for the trap discharge because a trap
works on a pressure difference.

§
Floot Bucket Inverted bucket Thermoetotle Exponeion

Fig. 14-28 Types of steam traps.

Probably because of the variety of circumstances under which condensate

is removed (low, high, or variable pressure; small or large flow; airfree or
otherwise; etc.), a large and somewhat bewildering assortment of traps is
built and offered by the many manufacturers of these devices. Some of the
principles of operation* will be described here, but the reader is directed to
manufacturers* bulletins for explanations of details of construction and opera¬
tion. The major principles seem to be float, thermostatic, and expansion action.
Float traps arc classifiable as continuous or intermittent. The following brief
descriptions are to be associated with Fig. 14-28.
1. Continuous float traps. These are primarily a float-operated valve, quite
simple in principle and operation. They are suitable for low pressure, but
should have a thermostatic air vent auxiliary if air can accumulate in the
trap. At high pressures and light loads the valve, floating on its seat, may be
cut and channeled by the discharge.
2. Intermittent float traps. The bucket trap is a well-known example, this
being seen in upright and inverted bucket types. The upright bucket floats on
the incoming condensation and holds the discharge valve closed until the accu¬
mulating water rises in the trap body far enough to spill into the bucket. When
nearly full, the bucket loses its buoyancy and sinks, pulling open the discharge
valve. The steam pressure can then force the water from the bucket through
the valve into the discharge. The bucket can then float again and close the dis¬
charge. Separate air vent valves must be provided if air enters the trap.
^ No claim is made of an all-inclusive survey.
614 THE PIPING SYSTEM
The inverted bucket type of trap vents both the condensate and air through
the main valve. With the bucket down the discharge valve is open, and water
flows out of the trap until the following steam reaches it. The steam rises up
under the bucket, making it buoyant and causing it to rise. This closes the
discharge valve. Condensate then rises in the bucket, the steam condenses or
leaks out through a small upper hole, and eventually the bucket weight sinks
it and opens the valve for another cycle of discharge. The bucket must have
some provision to keep air from accumulating in it, but no separate air vent
is needed in the body of the trap. Upright buckets prime themselves; inverted
buckets may have to be filled with water when first installed in order to start
working.
3. Thermostatic traps. A temperature-sensitive element is used to detect
whether steam or condensate surrounds it. Therefore the condensate must be
a little cooler than the steam. It would not do to install such a trap close to
the point of condensation—a condensation cooling leg is needed. However, for
draining heating coils, heaters, kettles, and the like, the thermostatic principle
is very practical and the traps have the advantage of passing water or air
equally well.
Temperature elements can be gas-filled bellows, bimetallic strips, or other
devices which respond to temperature difference with a mechanical movement.
Some thermostatic traps must be used only at the appropriate saturation tem¬
perature for which the element is designed; others have compound elements
that enable them to operate effectively over a range of pressures. Depending on
mechanical design, such traps may tend to become either continuous discharge
or semi-intermittont,
4. Expansiov, or orificej traps. Such traps have a flash chamber, or expan¬
sion chamber, between tw’o restrictions in the flow line. They are built with
and without auxiliary discharge valves operated by flashed
steam. If possessing no moving parts, it is the choking action
of an increased volume of hot-condensate-flashed-steam
mixture through the outlet orifice or maze that controls flow.
Under ^^no-load^^ conditions steam will pass through, and
such traps are not suited to some drainage applications.
However, they have the advantage of passing air and water
equally well; and some have no moving parts or valves.
The size of a trap needs to be determined by an estimate
of the maximum rate of condensation (exclusive of warm-up
if there is a by-pass valve), followed by a selection from
trap manufacturers^ rating tables. Traps should, if possible,
be installed below the unit; however, most traps will also
operate if placed above the unit they are draining, provided
the pressure differential is suflBcient to lift the water and
Fig. 14-29 Steam operate the trap. In such cases install a check valve in the
trap with bypass. trap inlet line. Traps are generally by-passed with a manual
valve to take care of initial condensation, and to keep the
steam unit in service when the trap is unserviceable, or being repaired. Notice
in Fig. 14-29 that three valves are required. Every trap should have a dirt
strainer installed just ahead of it in the inlet line. Some prefer to place the
DESIGNING THE PIPING SYSTEM 616
strainer inside the by-pass; others, outside. Some users connect a small line and
globe valve to the trap on the discharge side of the 'Valve and use it for a »tell-
tale. If pressure steam blows to the atmosphere when the tell-tale is opened,
then some trap maintenance is needed. A surprising quantity of steam can be
lost daily through a leaking trap.
14-13 Designing the Piping System. The following procedure is sug¬
gested as a guide to the successive steps to be undertaken in the design of a
piping system, large or small. Start with a diagrammatic representation of the
flows—a flow diagram—then:
1. Select the weight classification and the materials to be used in the
various sections of the piping system.
2. Calculate the diameters required and specify the nominal sizes.
3. Select and determine the thickness of insulation.
4. Provide for expansion.
5. Provide for draining condensation from steam lines and water from
water lines.
6. Make design studies and examine the layout for clearances.
7. Examine for accessibility and convenience of operation.
8. Consider the visual effect. Will the piping have a clean-cut appearance,
or will it appear jiiml)lod and amatcurKh?
9. Provide for support, and in so doing do not fail to follow up the drainage
scheme decided upon (item 5).
10. Make final drawings.
Selection of pipe weights and adherence to the selection will be facilitated
if some thought be given to classifying the piping according to the service it is
to render and establishing, at the beginning of the design, the weight to suit
the needs of each service. A pipe and fittings schedule is a good form in which
to keep such data. Against different services (feedwater, main steam, etc.)
such a schedule would list the pipe specification, at what pipe size joints change
from screwed to flanged, flange type, gasket type, fittings specification, etc.
The schedule can be retained as a permanent reference, w^hich can be very
convenient at times of alteration, addition, or repair.
Problems of pipe diameter have already been discussed. Most power plant
piping is too limited in length to warrant an economic study. More frequently,
the solution used is based on the assumption of velocities which have been
shown by experience to be liberal estimates.
The effectiveness of a piping system—its appearance, servidpability, ease
of operation—depends largely on the ability of the designer to visualize the
situation and to select the best arrangement to cope with it. Arrangement of
large or important pipe should not be left to the installation crew. The de¬
signer should not use the obvious in pipe design but should make several studies
of possible arrangements. The clearances he must consider are not only those
between pipes and pipe flanges and walls, beams, etc., but also the disassembly
clearances of the equipment to which the pipe is connected or past which it runs.
Convenience in operation is not to be overlooked, especially where piping is
congested and valves are thickly strewn. In those vicinities the design should
show the exact position of each valve even to the orientation of the valve stem.
Furthermore the designer must view the whole design from the standpoint of
616 THE PIPING SYSTEM
two other groups of meiii i.e. the installation and maintenance crews. For
instance, valves and expansion joints should be placed where they can be re¬
packed, strainers should be bypassed, etc. The design should permit ordinary
construction methods as far as possible.
Design of the pipe supports may well be left until the last on all but large
or important lines. So many varieties of support are marketed that supports
to meet the demands of the individual problem are readily selected. One thing
must be remembered: the success of the drainage system of steam lines depends,
in a large measure, on exact alignment of the pipe supports and this fact may
influence the designer to select the more expensive adjustable supports in some
cases.
All valves should be as accessible as it is possible to arrange them. This is
an important factor contributing to convenience of operation, and avoidance
of mistakes in closing or opening valves.
No rules can be formulated to insure good appearance. Nicety of judgment
and design technique on the part of the designers, ability to visualize what the
appearance will be when completed, and a natural sense of proportion are re¬
quired.
For operating convenience valves should be tagged with a label showing
their function in the system, and where there is a possibility of some delay in
the selection of the proper valve to operate in times of stress, a schematic flow
diagram of the piping and valves should be conspicuously mounted nearby.
Identification of power house piping by its painted color is embodied in the
ASME Report on the Identification of Power House Piping.
Steam division.a—High pressure.white
b—^Exhaust system. buff
Water division . c—Fresh water, low pressure. blue
d—^Fresh water, high pressure, boiler
feed lines.blue and white
e—^Salt water piping. green
Oil division . f—^Delivery and discharge .brass or bronze
yellow
Pneumatic division .g—^AU pipes . gray
Gas division .h—City lighting service. aluminum
i—Gas engine service. black, red flanges
Fuel oil division. j—^All piping.black
Refrigerating system .k—^White and green stripes alternately on flanges and fit¬
tings, body of pipe being black.
Electric lines and feeders .. 1—^Black and red striped alternately on flanges and fittings,
body of pipe being black.
High-Pressure Piping. The lead from boiler to turbine is the most important
element of high-pressure piping. It is characterized by a liberal use of long
radius bends and by thick insulation. Supports should be close together and
rigidly braced. Although the unit system of boiler and turbine arrangement is
growing in popularity, the main steam header continues a prominent feature
in most plants. These headers are of forged, welded, and cast-steel construction.
Typical arrangements are illustrated. The simplest arrangement is the single
header into which all boilers feed and from which all prime movers obtain their
DESIGNING THE PIPING SYSTEM 617
supply. Sectionalizing valves, normally open, increase reliability. Still more
reliable supply is obtained through the use of the ring, or loop header which’ also
is sectionalized by stop valves. These header valves and their counterparts on
the boiler leads are often motorized for remote control. On account of its cost
the ring header is infrequently used.
The centralized single header (or spider) system offers the further advantage
of bringing the header valves and fittings together for centralized support and
control. The manifolds used in the unit system of high-pressure piping are cross
connected but the cross connections are normally closed. Determination of the
type of high-pressure steam system to use rests mainly with the character of
the plant—the number of boilers and turbines, and their placement.

tfm Fup SumySrsriM-

I LOOP Fuo M<ir« Mom

AvroMor/c Coptool

atMie •SecrioMuup camMutn Srsrttt

UttiHMs Mmpeeo er Mmmps
n BontP
^ Tupamt
Secnofounm oo Toponu
Feeo Atotoomo opo exeess
Pptssmg VoLVti

Fig. 14-30 Schematic diagrams of typical steam headers and feedwater mains.

Feedwater Piping. Feedwater piping offers somewhat less difficult problems

than the high-pressure steam piping but its continuous functioning is no less
important. The feedwater line should be covered with insulation and given good
anchorage against vibration. It is not necessarily the same weight pipe as the
boiler lead but many engineers prefer to make it so. However, due to its smaller
size, it may be screwed instead of flanged. The ASME Boiler Construction Code
states that the feed pipe shall be provided with a check valve near the boiler
and with a valve or cock between the check valve and the boiler, and when
two or more boilers are fed from a common source, there shall also be a globe
valve on the branch to each boiler, between the check valve and the source of
supply. In addition to the valves mentioned, the branch lines will contain the
feedwater regulator valves when automatic regulators are installed. The regu¬
lating valve should be bypassed and the bypass line located so its valve may be
operated manually from the firing floor in times of derangement or inspection of
the feedwater regulator.
Blow-off Piping. Blow-off piping should be of a weight corresponding to
618 THE PIPING SYSTEM
the working steam pressure. The lines are usually flanged and uninsulated.
Steel fittings are used. Long radius ells and laterals are used instead of standard
ells and tees because of the high velocity of the water in the blow-off line.
The blow-off is disposed of to sewers, to the atmosphere, or to canals, pondp,
lakes, etc.
Where the boiler pressure exceeds 7.03 kg/cm* either two slow-opening
valves or one slow-opening valve and one quick-opening valve or cock should
be used in the blow-off line. Cocks having the plug held in place by a nut or
other fastening at the bottom should not be used for blow-off service. Such
fastenings^can become loosened or defective without being observed and permit
the plug to be blown out. Only cocks that have a guard or gland for securing
the plug should be used.
Where the blow-off piping passes through the boiler setting, ample spacing
should be provided around the pipe to allow for movement due to expansion
and contraction of the blow-off line and the boiler to which it is connected.
Expansion, and contraction stresses imposed upon a pipe, held tightly in the
setting wall or rigidly anchored a short distance from the setting can weaken
the blow-off line. It may be caused to crack at the threads of the mud drum
saddle.
14-14 Pipe Drawings. Originally there must be a conception of the
principal features of fluid flow in a plant. For example, the regenerative power
plant would be based on a flow plan set forth by drawings such as Fig. 8-13.
From the flow plan a schematic diagram of all major piping (omitting some¬
times the drip, drain, by-pass, and control, or pilot, lines) can be drawn. Such
a diagram should pay scant attention to the relative physical location of the
equipment, but should make it easy to trace flow patterns and understand
the connections of the plant. Therefore, valves and tees are important in this
drawing, elbows are not. The schematic should (1) clarify the basic plan of
the plant to all interested readers, (2) let the contractor bidder pick off all
required quantities, types, and sizes of equipment of the piping system except
for elbows and the lengths of pipe needed, and (3) be of tangible assistance
to plant operators. Schematic arrangements in isometric form can frequently be

High-pressure steam.
Tee.
" " ' Low-pressure steam.
H©l- Tee looking up.
«.i». Hot water.
■ O—
Tee looking down.
———— Returns—steam or water.

Air vent. Gate valve.

-II- Flanges.
Globe valve.

Union.
Angle valve.

Elbow.
Elbow looking up.
-S- Diaphragm valve.
Oi- Elbow looking down. Check valve.

Fio. 14-31 Symbols for single-line piping drawings.

PIPE DRAWINGS 619
used to good advantage as a preliminary to final drawings of the piping system.
Following the schematic, sealed pipe drawings are made in whatever detail
the job seems to require or justify.
Drawings of a piping system consist essentially of a plan and one or more
elevations. These may be supplemented by detail drawings of small portions,
drawn to a large scale. Isometric drawings are rarely used, except where three
dimensional configuration of some particularly crowded region is to be shown
on one view. The piping system of almost any plant is sufficiently extensive to
warrant drawing a special set of piping drawings with the piping shown heavy
(in function) y and the remaining equipment shown light (out of function) or
not shown at all. Method of drawing and dimensioning flanged piping has been
covered. Screwed piping is mostly small so that the steam fitters can cut and
fit such piping from a single line drawing. Screwed pipe so drawn may or may
not be dimensioned, depending on the individual case.

C||lXlll35t)^Cl|^
Globe Gate Check
Fig. 14-32 Symbols for ordinary flanged valves.

Pipe can be drawn in single-line diagram, or in double-line outline. Both

systems should not appear on the same run of piping but can appear on the
same drawing with the large, main piping two-line and the auxiliary piping
single line. Symbols to use on single line drawings are shown in Fig. 14-31.
Important piping, especially flanged-joint lines where flange face to face
dimensions are given, is drawn to scale as exactly as possible.
Actual valve outlines can be drawn in on flanged designs but to eliminate
this tedious and usually valueless work the valve syiUfeols shown in Fig. 14-32
are used. The final end dimension of a run of flanged piping is to the flange
face of the equipment served by the pipe. Now since this flange face is dimen¬
sioned to the equipment's center-line in practically all machine layouts, there
can be a column line to column line check on the dimensions of a pipe run if
the method of locating the equipment by building column lines as outlined in
Chapter 4 is followed.
Fig. 14-33 shows a piping plan drawing for the main high-pressure steam
piping in a power station boiler and turbine room. Note the in-function status
of the pipe, the out-of-function status of equipment and structure. Column-line
reference dimensioning and flanged pipe dimensioning may be studied with
this drawing.

PROBLEMS
1. Determine the specifications of material and wall thickness for a KHLS pipe
^ CMry steam at a state selected from the following: (a) 28.1 fcgim^gt. iie^5 (b)
26.71ig/oiiiSgai8aturat^; (c)84.4kg/om*fi^61(ra; (d) 5M kg/mbS ga 484*C»
Fig. 14-33 Boiler and turbine room piping plan—^high-pressure steam piping.
PROBLEMS 621
2. Would Schedule 120,203.2 mm pipe made to A53-SA Specification be acceptable
on a line operating at 17.6 kg/om* ga 400*C?
3. Specify the pipe required to carry 272,200 kg steam per hr at 87.9 kg/om* ga638*t?,
with velocity approximating 3050 m/min
4. What maximum working pressure is advisable in an 457.2 mm OD, Schedule 40
pipe A53-SA, operations not to exceed 232.2^ ?
5. A certain pipe line covered with heat insulation has exposed flanges. These are
found to be steel and to measure 279.4 mmdiax33.8imm thick, and to have 8 bolts. With¬
out removing any of the insulation, it is desired to determine whether it would be
proper to put 24.6 kg/om* ga steam through this line. Assume line is carbon steel with
flanges properly selected for the pipe weight.
6. Detail to scale a joint of 203.2 mm Schedule 80 piping consisting of a welded elbow
and two 1.22 m tangents terminated by 28.1 kg/om^raised face slip-on-and-weld type flanges.
7. Detail to scale a 1.37 m offset in a 152.4 mm screwed pipe line (including the 45*
elbows).8.8kg/om» W8P. Specify correct cut length for the offset pipe.
8. Repeat Prob. 7 but for (a) flanged joint piping with lap joint faces and (b) ail
welded joints.
9. Fig. 14-9P gives the configuration of a certain pipe
line in isometric. Detail it to scale in plan and elevation. Pipe
is 254 mm with 28.1 kg/cm* raised face flanges. Fxo. 14-9F
10. Repeat Prob. 9, assuming 101.6 mm screwed piping' for 17.6 kg/om^ worJong
pressure with flange end connections.
11. Repeat Prob. 9, assuming 203.2 mm
welded pipe, 21.1 kg/om* working pressure.
12. The configuration of a certain
pipe line in isometric is seen in Fig. 14-
12P. Detail it to scale in plan and eleva¬
tion for 28.1 kg/om* class, raised face
flanged joints.
13. Repeat Prob. 12, but assume
Schedule ^ welded piping except for end
connections and valve, which are flanged.
14. A portion of a manufacturer's
safety valve rating table is reproduced
herewith. From this line specify 3 safety
valves and settings suitable for use on a
stoker-fired, water-tube boiler generating at 17.6 kg/om* ga Boiler heating surface.
427.4 m8 including 55.7 mt water wall area. Steam rating, 10886 kg per hr; design pressure,
21J kg/omS ga

Valve size: mm 38.1x50.8 50.8x76.2 63.5x101.6 76.2x101.6 101.6x152.4

Rated Wg y i ' 16.9 kg/omS ga 4196 6895 9843 15241 ssaoi
kgperhr J 17.6 kg/om* ga 4364 7149 10206 15830 24113
at pres- 1 18.3 kg/oms ga 4527 7439 10614 16447 2d02B
sures of 1 19.0 kg/cm> ga 4695 7711 10986 17056 25836
15. A unit to produce steam at 17.6 kg/oms ga 139*0 superheat will be rated at 68039
kg per hr. Pulverized coal, 1203 m* boiler, surface of which 576 m> are m water
walls. 0.7 kg/omt loss in superheater. Select 3 drum valves and one superheater valve.
Valve discharge factor for iii*C superheat = 0.83. Using ratings from Prob. 14, specify
sizes and settings of the four valves.
16. A 61 m long straight steam line is anchored one end and terminates in a packed
expansion joint at the other. It holds steam at 10.5 kg/cm* ga. Expansion joints canbe
622 THE PIPING SYSTEM
bought in various capacities generally in increments of 102 mm of slip. If the minimum
cold temperature is o°C what slip length should th^ purchased expansion joint have?
Show by dimensioned sketch how the steam fitters should set the slip when measuring
to cut pipe for the connection on a day when the pipe temperature is 16.6*X3.
17. A264min Schedule 80 A-53 steel pipe is clamped between two rigid supports 6.1
m apart when its temperature isl6,6*C. A liquid at93.3®Cis then pumped through the
line. What thrust ^rce is developed at the reactions?
18. Design and detail to scale a double offset expansion TJ bend (type Fig. 14-17)
in a pipe suitable for conveying 680139 kg per hr superheated steam at tU kg/om* ab»
871.1*0. Installation at 26.7 «c; no cold spring. Pipe line is straight between anchors
located 46.7 m apart.
19. Design and detail to scale a double offset expansion U bend (Type 2, Fig. 14-17,
A =4j064ni) in a 254 mmC-Mo steel pipe forsro^ 63.3 kg/om* flanges. Straight run be¬
tween anchors =30.5 m. Expected installation temperature, 26.7®C; expected minimum
temperature,-l7.8®c.Cold spring of % total expansion, corrected for temperature on day
of installation, is to be applied. What cold spring will steam fitters actually apply?
2Q. Solve Prob. 19 for a welded U bend expansion joint having W (Fig. 14-18)
= 1.62 m.
21. Design a welded U bend to absorb the thermal expansion in an 24.4 m length of
203.2inoiA1068teelpipe(SA)operated at343k3*C. Installation temperature 16.6^3, no cold
spring. Assume H (Fig, 14-18) = 3.06 m.
22. How much condensation is produced per hour by heat leakage from a Schedule
40 steel pipe855.6mmODX26.9m long. Saturated steam at 17j6 kg/em^ab; ambient
temperature|i6«6*X7. Cover is Double Standard thick sectional magnesia.
23. Find the kcal heat loss per hr per mirom a louemm pipe line carrying a fluid at
I76.7*0when covered withaimmof insulation No. 6, Fig. 14-21. Ambient temperature,
-6.7^C.
24. Steam at60t2 kg/om3,464.4°Cis flowing through a254 mmx30.5inipipe line at the
rate of 40*823kg per hr. What thickness of insulation (diatomaceous earth with asbestos
fibers) is needed if the heat leakage from this pipe is to be reduced to 0.015% of the
heat flowing? 16.6*C ambient temperature.
25. Solve Ex 24 if a surface temperature of 48.9®C is the criterion of insulation
thickness rather than heat loss.
.26. Suppose the pipe line of Prob. 24 had this covering instead of the one given:
61 mm insulation No. 1 over which is 51 mm insulation No. 5. Find the kcal per hr
per m beat loss.
27. A 203.2 mm main steam header is to receiveian'^economic’' thickness of insulation
No. 4, Fig. 14-21. Steam temperature, 37l.l*C; header in continuous use. Insulation
cost 79^ per bd m; steam cost, 200 4 per 10® kcal. Fixed^c^rges, 12%. Find the
thickness and calculate the heat loss per m of pipe.
28. If the answer to Prob. 27 is 236.6 kcal per hr pep^^’what is the insulation
‘^efficiency’"?
29. A 203.2 mm pipe carries saturated steam at 7.03 kgifom* ga. Find the heat loss
per hr per 30 m length if the pipe is (a) bare, (b) covered with Standard Thick
Magnesia. Ambient temperature, 87.8*C.
30. Using average flow practice determine size of pipe needed to convey one of
the following fluids.
(a) 22,680 kg per hr saturated steam at 12.7 kg/om* ga.
(b) 17,100 kg per hr wet steam at 3,7 kg/cm* ab, .96 dry.
(c) 16,876 kg per hr of feedwater at I87.8°C.
(d) 9,072 kg per hr saturated steam at 0.36 kg/om* ga.
31. Find the least size of Schedule 80 pipe that will convey 1825 l/min water a short
distance, including four 90® elbows and one gate valve, without Ap exceeding 0.86 kg/emP.
 BIBLIOGRAPHY

(For Chapter 1, see end of Chapter 1.)

CHAPTER 2
Application of Diversity Factor, H. B. Gear, Proc. Natl. EL Light Assoc.y 1915.
The Electric Power Industry, McGraw-Hill, 1949.
Westinghouse Engineer, January 1950.
CHAPTER 3
Economics of Public Utilities, L. R. Nash, McGraw-Hill, 1932.
The Electric Power Industry, McGraw-Hill, 1949.
Power Economics for Engineering Students, Pittsburgh Printing Co., 1939.

CHAPTER 4
Semi-Outdoor Steam Plants, Gourdon, Friend, and Elliott, Combustion, April 1946.
How to Design a Foundation, G. Steven, Power, February 1947.

CHAPTER 5
Fuel Oils Commercial Standard, U. S. Dept. Commerce, Govt. Printing Office.
Diesel Fuels, Standard Oil Co., N. J., 1939.
The Processes of Combustion in a Furnace, Henry Kreisinger, Combustion, November
1929.
U. S. Bureau of Mines, Bull. Nos. 22, 85, 193; Tech, papers Nos. 93, 158; Information
Circulars Nos. 6888, 7538.
Fuel Test Codes, ASTM.
» CHAPTER 6

Diesel and Gas Engine Power Plants, G. C. Boyer, McGraw-Hill, 1943.

Diesel Engineering Handbook, L, H. Morrison, Diesel Publications, Inc., 1943.
Modern Diesel Engines, Power, April 1948.
Combustion of Diesel Fuel, M. A. Elliott, Trans. SAE, July 1949.
Stationary Diesel Engine Lubrication, Socony-Vacuum Oil Co., 195L
Diesel Engines—Fuels and Lubricants, Sinclair Refining Co., 1949.
Significance of Diesel Exhaust Gas Analysis, Holtz and Elliott, Trans. ASME, Febru¬
ary 1941.
Preliminary Cooling Tower Selection, Foster Wheeler Co., Heat Engineering, March-
April 1949.
Test Code for LC. Engines, ASME.
CHAPTER 7
Theory and Design of Gas Turbines and Jet Engines, E. T. Vincent, McGraw-Hill,
1950.
An Appraisal of Gas Turbines for Power Plants, A. G. Christie, CombvMtion, Novem¬
ber 1947.
667
668 BIBLIOGRAPHY
Convenient Gas Properties and Charts, C. J. Walker, Ai^ME paper 50-F28f 1950.
A Comparison of Gjis and Steain Turhines, O. A. F. Mnnzinjrer, Combustion, Septem-
' her 1948.

CHAPTER 8

Present and Future Trend of Thermal Prime Movers, C. W. E. Clarke, Combustion,

July 1949.
The Steam Turbine Regenerative Cycle, J. K. Salisbury, General Electric C'o. (Collec¬
tion of 3 reprints.)
Reheat Cycle (Symposium), Trans. ASME, August 1949.
Topping at Sherman Creek, H. Knecht, Combustion, February 1947.
Modem Mercury-Unit Power-Plant Design, Hackett and Douglass, Trans. ASMS.
January 1960.
Theory of Incremental Rates, Steinburg and Smith, Electrical Ejigweering, March
1934.
CHAPTER 9

Beat Transmission Between Fluids and Solids, W. H. McAdams, Mechanical Engineer¬

ing, July 1930.
Radiant Heat Transmission, H. C. Hottel, Mechanical Engineering, July 1930.
Elements of Heat Transfer and Insulation, Jacob and Hawkins, John Wiley and Sons,
1942.
Nuclear Fuel for Power Production, Flagg and Gross, General Electric Review, March
1952.
Estimation of Radiant Heat Exchange in a Boiler Furnace, Orrok and Ortsay, Com¬
bustion, April 1938.

CHAPTER 10
Combustion Engineering, 0. deLorenzi, Combustion Engineering-Superheater Co., Inc.,
1947.
Steam Generation, Power, December 1946.
Economic Factors Involved m Selection of Industrial Boilers, Patterson and Riker,
Combustion, August 1949.
Refractories, Power, June 1950.
Factors Affecting Superheat Control, M. Frisch, Heat Engineering, December 1950.
Fuels and Firing, Power, December 1948.
Test Code for Stationary Steam Generating Units, ASME.
Development of Pulverized Coal Firing, C. G. R. Humphreys, Combustion, September,
October, November 1948.
Design and Operation of Spreader Stokers, W. S. Major, Combustion, July 1949.
The Problem of Generating Pure Steam at High Pressures (abstract), Frisch and
Lorenzini, Heat Engineering, May 1960,

CHAPTER 11

Stationary Steam Engine Lubrication, Socony-Vacuum Oil Co.

Steam Turbine Lubrication, Socony-Vacuum Oil Co.
Steam Turbines, Church, McGraw-Hill, 1950.
Steam Turbines and Their Cycles, Salisbury, John Wiley and Sons, 1950.
Steam Turbines, Power, December. 1945.
Relative '^Engine Efficiencies'^ of Large Steam Generator Units, Warren and Knowlton,
Trans, ASME, February 1941.
Modem Extraction Turbines, L. E. Newman, Power Plant Engineering, January,
February, March, April 1945.
 Entropy kca(/kg deg,C

0«34
Fmj. A-- Molher diastram for nirrciiry vai>or.

661
 Vnthalpy, Joult per grae Butropy, Joulta per graa deg C

8p TOl
Preaa Teep Sat Sat Sat Sat Sat Vaper
Kc/c«>ab C Liquid Brap Vapor Liquid Srap Vapor a^As

0*0007 111*00 16*010 297*105 312*614 0.0476 0.7716 0*8102 227

0*0014 126*60 17*510 206*830 314*350 0*0027 0*7426 0*7963 118
0*0021 136*68 18,766 206*658 315*424 0.0558 0*7257 0*7816 81
0*0028 142*40 10*685 206*532 316*217 0.0580 0*7176 0*7717 62
o*ooas 147*76 20*418 296*430 316.848 0.0507 0*7043 0.7640 60
0*0002 157*80 21.802 296*242 318.043 0.0630 0*6873 0*7603 34
0*0070 165*40 22*827 206.100 318*027 0*0654 0.6752 0*7386 26
0*0140 184.60 25*437 205.737 321.174 0.0711 0*6461 0*7173 13*6
0*0210 106.62 27*072 206*511 322.584 0*0747 0*6201 0.7037 0.3
0*0281 205*54 28*282 205*314 323*626 0*0772 0*6170 0*6942 7.1
0*0302 212*75 29*233 205*207 324*440 0*0702 0*6076 0*0868 6*75
0*0422 218*80 30.073 205*005 325*168 0*0800 0.6009 0*6808 4*86
0*0402 234*00 30*780 204*006 325.775 0*0823 0*6733 0*6757 4.21
0*0563 228*00 31*401 204*009 326*310 0*0835 0.5878 0*6713 3.72
0*0033 232*70 31*969 204.832 326.701 0*0946 0*5828 0*6674 3.33
0*0703 236*50 32*468 204.760 327*820 0*0856 0*5784 0*6640 3.02
0*1400 263.74 35.087 204*267 330*264 0*0023 0*5402 0*6410 1.58
0*2109 270.50 38*237 293*053 332*100 0.0064 0*5320 0*6284 1.09
0*2812 202.14 30.916 203.716 333*632 0*0003 0*5108 0*6101 0*84
0*3515 302.05 41 *266 293.526 334*701 0,1017 0*5103 0.6121 0*68
0*4216 .310.67 42.410 203*364 335.774 0,1037 0*5026 0*0062 0*68
0*4022 318.15 43*396 203*227 336.623 0*1063 0*4960 0*6013 0*60
0*5625 324.72 44*275 203*102 337,377 0*1068 0*4003 0*5071 0.44
0*6328 325,15 45.000 202*088 338*068 0*1080 0*4852 0*5033 0*40
0,703 336.10 40*787 202.888 338*675 0*1003 0*4808 0*0001 0*36
1.406 374.50 50*856 202*167 343.022 0*1172 0*4613 0*0680 0*10
2*100 309.22 54*142 201*607 345*830 0*1222 0*4338 0*5560 0*13
2.812 418.00 56*627 201*343 347*070 0*1258 0*4216 0*5473 0*103
3.515 433.40 58*622 201*065 349.677 0*1288 0*4120 0*5406 0*084
4.318 446.50 60*336 200*806 351*142 0*1300 0*4041 0*5351 0.071
4.722 458.00 61*837 200*500 352.426 0.1320 0*3074 0*5304 0*062
6*625 468.15 63*172 200*300 353*571 0*1347 0*3017 0*5265 0*063
6*328 477.50 64*384 200*824 354*608 0*1303 0*3866 0*6220 0*050
7.031 486.00 65*482 290.066 356*543 0*1378 0*3812 0*6109 0.046
7.734 493.85 66.514 289*016 356*420 0*1300 0*3780 0*6170 0.042
8*437 501 .30 67*465 289*778 357,243 0*1407 0*3742 0*6145 0.030
9*140 508.10 68,421 280.648 368.000 0*1414 0.3706 0*6122 0*036
0*843 514,60 69*104 280.527 358*720 0*1424 0*3676 0*6000 0*034
10.546 520.80 69.980 289.410 350*400 0.1434 0*3646 0*5080 0*032
11*250 526.60 70*746 289*299 360*044 0*1443 0*3618 0*6060 0*030
11*952 532.15 71.464 289*104 360*660 0*1452 0*3501 0*5043 0*028
12*655 536.05 72*148 280.094 361*242 0*1460 0*3666 0*6027 0.087
13.358 341.50 72*780 289*003 361*784 0.1468 0*3644 0*5011 0*026
14*061 547.35 73.400 288.010 362*310 0*14751 0*36217 0*40068 0*025
15*810 588.85 74*007 288,602 363*508 0*14918 0*34698 0*40624 0*022
17*577 569.55 70*266 288*402 364*747 0*15086 0*34241 0*49327 0*020
10*.334 570.35 77*307 288*306 365*815 0*15228 0*33827 0*40066 0*019
21*002 588.45 78.670 288*136 366*810 0*15362 0*33442 0*48812 0*017
24*607 605.20 80*822 287*819 368*041 0*10697 0*32776 0*48372 0*016
28*133 620.22 82*724 287*540 370*264 0*15800 0*32100 0*47096 0*013
31.638 634.00 84*400 287*288 371*761 0*16004 0*31670 0*47674 0*012
35*164 646.65 86*076 267*042 373*118 0*16166 0*31214 0*47480 0*011
42*184 699,35 88*058 886*612 375*570 0*16463 0*30414 0*46878 0.000
40*215 689 •'Bo 01*503 286.232 377*735 0.16710 0*20736 0*46460 0*008
56*246 707.85 03*804 285.880 370*678 0*16046 0*20146 0*46091 0*007
63*276 724.55 05*802 280*775 381*466 0*17146 0.28687 0*46772 0*006

specific volume of liquid mercury at lOO'^C =» 0*000078 m^/kg. Volumetric coefficient

of expansion « 0*000182 per deg. C.
Fig. A-3 Thermodynamic properties of saturated incrcur>' vapor.*
• Reprinted from paper co-sponsored by Research Committee on Properties of Gases and Gas Mixtures,
Heat Transfer and Applied Mechanics Division, ASME, for presentation 1949 Annual Meeting.

662
663
* Carbon monoxide. t Sulphur dioxide. % C. F. Sulphur dioxide C. F. » cubic feeU

Fio. A-4 Combustion properties of fuel.

 Atemk Nom* •! Alomie Atomic Nomo of Atomic
Symbol Ntimbor Elomonf Woigkl Volone* Symbol Numbor Elomonf WoigKt Voloneo

H 1 Hydrogon I.OOBO 1 Pd 46 PoIIadTum 106.7 2.4

Ho 2 Holium 4.0D3 0 Ag 47 Silver 107.880 1
U 3 Lithium 6.940 1 Cadmium 112.41 2
Cd 48
Bo 4 Borylllum 9.02 2 3
In 49 Indium 114.76
B S Coron 10 82 3 Sn Tin 2.4
50 118.70
C 6 Corboo 12.010 4 3.5
Sb 51 Antimony 121.76
DIomond To 52 Tellurium 127.61 2.4 6
Gropkito
1 53 Iodine 126.92 1
N 7 Nitrogon 14.001 lA
Xe 54 Xenon 131.3 0
O B Oxygon I6 00C 2 Ct Cesium 13291 1
55
F 9 Fluorino 19.00 1 2
Bo 56 Boiium 137.36
Nc 10 Noon 20J83 • La 57 3
Lanthanum 138 92
No II Sodium 22997 1 3.4
Ce 58 Cerium 140 13
Mg 12 Mognotlum 24J2 2
Pr 59 Praseodymium 140 92 3
Al 13 Aluminum 26.97 2
Nd 60 Neodymium 144.27 3
Si 14 Silicon 28.06 4
15 Fkospkorou* 3098 lA 11 61 Illinium 144 50 3
F
block Sm 62 Somorlum 150.43 3
red Eu 63 Europium 152 0 3
16 Sulfur 32.06 Gd 64 Gedoiinlum 156 9 3
2.4A Tb 65 Terbium 159.2 3
17 Cklorino 35457 1 Dy 66 Dysprosium 162.46 3
II Argon 39.944 ? Ho 67 Hoimium 163.5 3
19 Potostium 39.096 1 3
Er 68 Erbium 167.2
20 Calcium 40 08 2 Tm Thulium 169.4 3
69
21 Scandium 45 10 1 3
Yb 70 Ytterbium 173 04
22 Titanium 47.90 4 3
Lu 71 Lutecium 174.99
23 Vanadium 50.95 3A Hf 72 Hafnium 178.6 4
24 Chromium 52.01 2.3.6 To 73 Tentelum 180.88 5
25 Mongonoto 54 93 2,3.7
W 74 Tungsten 183 92 6
26 Iron 55.85 23 Re 75 Rhenium 18631 7
Co 27 Cobalt 58 94 23 Osmium 190 2 6.8
Os 76
Ni 2B Nickol 58.69 2.^
Ir 77 Iridium 193 1 4
Cu 29 Copper 63.57 •3
Pt 78 Plotinum 195 23 2.4
Zn 30 Zinc 65 38 2
Au 79 Geld 197.2 1.3
Go 31 Gallium 69 72 3
Go 32 Germanium 72.60 4 Hg 80 Mercury 200AI 1.2
33 Arsenic 74.91 3.5 Tl BI Thallium 204.39 1.3
At
34 Selenium 78.96 2.4,6 Pb B2 Lead 207 21 2.4
So
35 Bromine 79.916 1 Bi B3 Bismuth 209.00 3.5
Br
Kr 36 Krypton 83.7 0 Po B4 Polonium 210.0 2.4.6
Rb 37 Rubidium 85.48 1 Ab 85 Alobamlne 221.0 B
Sr 31 Strontium 87 A3 2 Rn 86 Radon 222.0 0
Y 39 Yttrium 88.92 3 Vi 87 Virglnium 2223 1
Zr 40 Zirconium 91.22 4 Ro 88 Radium 226.05 2
Cb 41 Columbium 92.91 5
Ac 89 Actinium 227.0 3
Mo 42 Molybdenum 95.95 4.6
Th 90 Thorium 232.12 4
Mo 43 Mesurium 97.8 2,3.7
Pa 91 Protactinium 231.00 5
Ru 44 Ruthenium 101 7 6.8
U 92 Uranium 238.07 4.6
Rh 45 Rhodium 102 91 3

Fig. A-5 Table of elements.

664
 i_j j | | | ||
Decimal ihicknesses are average. Expect variation of :±:12.5% wall thickness.

Fig. A-6 Pipe Standards.

Fig. A>7 Steel flange dimensions*, mm, (Abstract of ASA, B16e, Standard)!) = dia.,
B = bolt circle dia., T = thickness**, N = munber of holes, d = bolt hole dia.
• See Fig. 14^. 106.5 kg/om» and 176.8 kg/om* oImms and some pipe mam ave
omitted.
Thicknem, T, mdudee raised ftboe of 1.68 mm in 10.8 kg/om* and 81.1 kg/ena^
Standards ; does not include raised fooe in the others.
 90 Elbow Lone Radius
Elbow

Oast Iron>.d«8 kg/om9 WSPCiASS^

Size

25.4
B 88.90
B

127.00
C

44.46
D

107.96
E
146.05
F

44.45
G9 H

11.11
Si 75 95.26 189.70 60.89 117.48 158.76 44.45 12.70
88.10 101.60 152.40 67.16 127.00 177.80 60.80 14.29
50.80 114.30 166.10 68.60 162.40 203.20 63.60 127.00 16.88
63.60 127.00 177.80 76.80 177.80 241.30 63.60 139.70 17.46
76.20 139.70 196.85 76.20 190.60 254.00 76.20 162.40 19.05
88.90 162.40 216.90 88.90 216.90 292.10 76.20 166.10 20.64
101.60 165.10 228.60 101.60 228.60 804.80 76.20 177.80 23.81
127.00 190.60 260.35 114.30 254.00 24S.90 88.90 203.20 28.81
152.40 203.20 292.10 127.00 279.40 368.30 88.90 228.60 25.40
203.20 228.60 366.60 139.70 342.90 444.50 114.30 279.40 28.58
254.00 279.40 419.10 165.10 406.40 520.70 127.00 304.80 30.16
804.80 304.80 482.60 190.60 482.60 622.30 139.70 366.60 31.75
165.60 856.60 546.10 190.60 583.40 686.80 152.40 406.40 84.93
406.40 381.00 609.60 203.20 696.90 762.00 165.10 457.20 86.51
467.20 419.10 673.10 215.90 636.00 812.30 177.80 482.60 89.69
608.00 467.20 736.60 241.30 698.50 889.00 203.20 508.00 42.82
609.60 568.80 868.60 279.40 812.80 1028.70 228.60 609.60 47.63
762.00 636.00 1054.10 381.00 084.25 1244.60 264.00 762.00 53.98

CAat*St«al-10.<S kg/cn2 WSP Glass s&d Cast Iron«»17.6 kg/cm^ WSP Class^

25.4 101.60 127.00 50.80 123.83 165.10 50.80 17.46

31.T6 107.95 139.70 63.50 133.35 184.15 57.15 19,05
38.10 114.30 152.40 69.85 155.58 215.90 63.50 30,64
50.80 127.00 165.10 76.20 165.10 228.60 63.50 127,00 23,23
63.50 139.70 177.80 88.90 190.50 226.70 76.20 139,70 25,40
76.80 152.40 196.85 88.90 209.55 279.40 76.20 152,40 28,60
88.90 165.10 215.90 101.60 228.60 317.50 76.20 165.10 30.16
101.60 177.80 228.60 114.30 254.00 342.90 88.90 177,80 31,75
127.00 203.20 260.35 127.00 279.40 381.00 101.60 203,20 34,93
152.40 215.90 292.10 139.70 317.50 444.50 127.00 228,60 36.51
803.20 254.00 355.60 152.40 381.00 530.70 139.T0 279.40 41,28
264.00 298.10 419.10 177.80 444.50 609.60 152.40 304.80 4T.63
304.80 303.20 482.60 203.20 520.70 658.80 165,10 355.60 50,80
355.60 361.00 546.10 215.90 584.20 787.40 100.50 406.40 53,98
406.40 419.10 609.60 241.30 647.70 876.30 203.20 457.20 57,15
457.20 457.20 673.10 254.00 711.20 952.50 482.60 60,33

Fig. A-9 Dimensions of standard flanged fittings—straight sizes.®*

Footnotes will be found at end of table.

667
 Cast Stssl-ai.l kf/asi« WSf Class<r /

Slse

12.7
A B C D B P
HH H

10.1 For sls4iS below 50.8 an u se dlaens;Ions of tbs 42.2 kjS/on^ olasis.
25.4 Where 1 •95 aa raised face la requlired. subtract 4.76 na froa
31.8 A. B. C, E, and F.
38.1
50.8 127.00 165.10 76.20 165.10 228.60 63.50 127.00 22.23
63.5 139.70 177.80 88.90 190.50 266.70 63.50 139.70 25.40
76.2 152.40 196.90 88.90 209.55 279.40 76.20 152.40 28.58
88.9 165.10 215.90 101.60 228.60 317.50 76.20 165.10 30.16
101.6 177.80 228.60 114.30 254.00 342.90 76.20 177.80 31.75
127.0 205.20 260.40 127.00 279.40 381.00 88.90 203.20 34.93
152.4 215.90 292.10 139.70 317.50 444.50 101.60 228.60 36.51
203.2 254.90 355.60 152.40 381 .OO 520.70 127.00 279.40 41.28
254.0 292.10 419.10 177.80 444.50 609.60 139.70 304.80 47.63
304.8 330.20 482.60 203.20 520.70 698.50 152.40 355.60 50.80

Cast Stss 1-28.1 ks/oa^ WSP Class*’

12.7
19.1
25.4
31.8
38.1 For sls« s below ; .01.6 na i ise dlaeni lions of i 2.2 kg/oi class.
50.8
63.5
76.2
88 ..9
101.6 203.20 139.70 254.00 406.40 114.30 209.55 34.93
127.0 228.60 152.40 279.40 425.15 127.00 234.95 38,10
152.4 247.65 158.75 317.50 476.25 133.35 254.00 41 .28
203.2 298.45 171.45 381.00 565.15 146.05 304.80 47.63
254.0 336.55 196.85 444.50 654.05 158.75 342.90 53.08
304 .8 381.00 222.25 520.70 755.65 165.10 387.35 57.15

Cast Stssl—42.2 kg/oa^ WSP Class'*’^

12.7 82.55 50.8 95.25 146.05 44.45 127.00 14.29

19.1 95.25 63.50 117.48 171.45 50.80 127.00 15.88
25.4 107.95 63.50 123.83 184.15 57.15 127.OO 17.46
31 .8 114.30 69.85 133.35 203.20 63.50 127.00 20.64
38.1 120.65 76.20 155.58 228.60 69.85 127.00 22.23
50.8 146.05 107.95 165.10 260.35 88.00 152.40 25.40
63.5 165.10 114.30 190.50 292.10 88.90 171.45 28.58
76.2 177.80 127.00 209.55 323.85 101.60 184.15 31.75
88.9 190.50 139.70 228.60 355.60 114.30 196.85 34.93
101.6 215.90 152.40 273.05 419.10 114.30 222.25 38.10
127.6 254.00 177.80 330.20 495.30 152.40 260.35 44.45
152.4 279.40 190.50 335.60 533.40 165.10 287.75 47.63
203.2 330.20 215.90 419.10 622.30 177.80 336.55 55.56
254.0 393.70 241.30 508.00 749.30 203.20 587.35 63.50
304.8 419.10 254.00 558.80 800.10 215.90 612.75 66.68

Fig. Af»9 Continued

 Slg9 A B 1 c 11 B 1L E 1 1 r 1i OB 1 H

25.4'
31.8
38.1 For glsdIt btlow 76.2 ■■ Qtt dlatntloat of 1435.5 kg/oa^ olatt.
50.8
63.5^
76.8 190.50 139.70 241.30 368.30 114.30 196.85 38.10
88.9 215.90 152.40 273.05 419.10 114,30 222.25 41.28
101.60 228.60 165.10 292.10 444.50 139.70 234.95 44.45
127.00 279.40 190.50 349.25 533.40 165.10 285.75 90.80
152.40 304.80 203.20 381.00 571.50 165.10 311.10 55.56
203.20 368.30 228.60 469.90 698.90 190.90 374.65 63.50
254.00 419.10 254.00 546.10 787.40 215.90 425.45 69.85
304.80 482.60 279.40 609.60 863.60 228.60 450.85 79.38

Cast Stoel-lOS.S kg/on WSP Claes

25.4 127.00 88.90 149.23 228.60 63.50 127.00 38.58

31.8 139.70 101.60 158.75 254.00 j 76.20 146.05 28s58
38.1 152.40 107.95 127.80 279.40 88.90 158.75 31.75
50.8 185.15 120.65 215.90 336.55 101.60 184.15 38.10
63.5 209.55 133.35 244.48 387.35 114.30 209.55 41.28
76.2 234.95 146.06 266.70 438.15 127.00 234,95 47.63
101.6 373.05 184.15 311.15 488.95 152.40 273.05 53.98
127.0 336.55 222.25 374.65 590.55 190,50 349.29 73.03
152.4 352.43 238.13 393.70 631.83 206.38 368.30 82.55
203.2 415.93 276.23 482.60 758.83 231.78 431.80 02.08
254.0 499.30 304.80 584.20 914.40 260.35 514.35 107.95
304.8 569.15 336.55 673.10 1035.05 i 304.80 584.20 123.83

Fig, A-V Continaed.

o DiWMioM ar« In Thin tnbl* i« nn nbbrcvantiron of tho Standard, which eoworo larger einea
than here entered, and which includee 1,76-, 176,6~and 281,2 kg/en* claeeea,
b All 8,8 kg/en* elaaa oast iron flangss hare a platn face.
All 17,6 kg/en* claos east iron flanges hare a 1,69 nn raised face. This raised face is included in
all dinensioas except S,
d A raised face of 1,68 nsi is included in all diaenslons exoept D, When flanges other than 1,59 wm raised
face are used, the center to flange edge diaensions shall renain unchanged.
^ A raised face of 6,36 wm is inolnded in all diaensions except 0 and H. Where flanges other thas these
hawing a 6,36 aa raised face are uaed, the center to flange edge diaensiona shall resain unchanged,
/ The cast steel fitting pressure ratlnga incorporated in this table are based on the following toaperatnreot

10.6 ki^oa*.260,0*c
21.1 k^ea*.•398.9*e
28.1 k^ca^.90e.0*c
42.2 ki/ea«.482.2*c
63.3 kf/cB*.482•2*c
105.6 ki^ea*.482.2*e

Working pressures aay be raised if the tenperature is lower than the abowe. Conwerooly, higher tOBpomturoi
can be aecenaodated only if the pressure is lower. Fitting aanufaeturers are prepared to furnish pressure class
reeoaasndatieas tor stoaa cenditlons differing conoidombly froa those quoted.
g Redueers and eccentric reducers for all reductions use the sane face to face diaenoions giwon in the abowe
table of diasnaiona for the larger opening.

669
 fO'CLk AVtlX KCTUKHtCNB TSB RCOUCBR STUB CNO CAB

Sis«>HUi
DMg BB B E
BB H

■g
2S.0 38.1 38.10 50.80 101.60 50.80 38.10
31.8 47.63 47.63 50.80 101.90 63.50 38.10
38.1 57.15 28.58 57.15 63.50 101.60 73.03 38.10
50.8 76.20 34.03 63.50 79.20 152.40 02.08 38.10
63.5 05.25 44.45 76.20 88.00 152.40 104.78 38.10
76.8 114.30 50.80 85.73 88.00 152.40 127.00 50.80
88.8 133.35 57.15 05.25 101.60 152.40 130.70 63.5
101 .• 152.40 63.50 104.78 101.60 152.40 157.16 63.50
127.0 190.90 70.38 123.83 127.00 203.20 185.74 76.20
102.4 228.60 05.25 142.88 130.70 203.20 215.00 88.00
203.2 304.80 127.00 177.80 152.40 203.20 260.88 101.60
254.0 381.00 158.75 215.00 177.80 254.00 323.85 127.00
308.4 457.20 100.50 254.00 203.20 254.00 381.00 152.40
350.6 OD 304.80 412,75 165.10
406.4 OD 304.30 460.00 177.80
457.2 304.80 533.40 203.20
506.0 304.80 584.20 254.00
600.6 304.80 602.15 266.70

Fig. A-10 Standard (Schedule 40) and extra heaw (Scdiedule 80) weldinj? fitting
dimension.^, inches.

670
 i

I
8
I

i
I

'C
S3
o3
s.

013
%

I
ecature difference between pipe and air.)
 Barometric
Pressure
1*033 kg/cm^

0 0004 0.008 0.012 0.016 0.020

kg of Moisture per kg of dry oir

672
ENTHALPY-kcal/kg (BTU/LB)
674
HroffAUUc Raoms
 Courtesy G B Warren and P H Knowlton

Frr, A-15 Efficiencieb of a-c generators

Fio. A-18 Temperature-pressure relation for very low-pressure saturated steam.
(0*0 mercury.)
Efficiency 8 Power foctor at ro^ed food

Amps locked rotor current

Rated horsepower of motor
Characteristics of three-phase, 60-cycle» induction motors

Motor power formuioe

D-c motor, Shaft power • i7El/735-5 hp
Single-phase o-c motor. Shaft power ■ 9EIF/733S hp
Three-phase a-c motor, Shoft power ■17El f/735.5 ho
Voltoge iose in motor lends
0-c motor, ^E”IR volts
Single-phase a-c motor, 5£»Irf volts
Three-phose a-c motor, AE^'^IRF volts

Where*
I • Line current, omperes
E • Line voltoge
t? * Motor efficiency
F ■ Power factor
R • rL, in which r« ohms per m of single conductor, ond
L« length of single conductor from supply line, m

Stondard motor sizes to 50 hp

i i I I H 2 3 S 7i 10 IS 20 25 30 40 SO

Stondord synchronous speeds - rpm

3600 1800 1200 900 720 600 514 450 400 360
Fig. A-19 Electric Motor Data.
677
 INDEX

Accumulator, hydraulic, 56 Barometric condenser, 414

steam, 54 Base exchange water softener, 530-531
A. C. power, measurement of, 651, 652 Baum4 scale, 119
Air, combuscion, 137 Belt conveyor, 445-446, 449-450, 452
excess, 137, 144 Belting, 450
filter, 173,175 Binary vapor cycle, 246
Air-fuel ratio, 139, 144, 159 Blade velocity ratio, 385
Air leakage (condenser), 423 Bleeder turbine. See Extraction.
Air preheater, 470-474 Blowdown, 522
Alkalinity, water, 520 control, 524
Altitude, derating for, 160 piping, 617
Ammeter, 651 tank, 525
Ampere, 13 valve, 586, 587
Analysis, feedwater, 519, 522 Boiler, availability, 292
fuel, 127 capacity, 294
proximate, 125, 127, 134, 135 classification, 297
ultimate, 127, 128 controlled circulation, 293
Anchor bolts. 111 economic surface, 28^287
API scale, 119 feed pump, 556-562
Apron conveyor, 444 furnace, 313-318
Architecture, power plant, 86, 89 high pressure, 263
Ash analysis, 135 horsepower, 294
Ash conveyor, losses, causes of, 349
classification, 467 meter, 647-650
hydraulic, 469-470 trim, 326
pneumatic, 468 Boiler Code, ASME, 290
ASME Code, Boiler, 290 Boiler efficiency, definition, 296
boiler test, 348, 350, 351 Boiler feed pump, 556-562
safety valves, 589 control, 559-560
turbipes, 408 required positive suction head, 558
ASME Short Code (Boiler), 351 Boiler heat balance, calculation of, 353-355
Atmospheric relief valve, 405, 409,429 Boilers, bent tube, 299
Atomic energy, 259,260 fire tube, 297,301-302
Atomic Energy Commission, 6, 14, 259 water tube, 297, 303-306
Atomizing burner, 344 Brake horsepower, 163, 367
Auxiliaries, condenser, 426-430 Breeching, gas, 4SO-483
Diesel, 169 Brick estimating, 90
Auxiliary drive turbine, 401 Brick,^radial chimney, 100, 497-498
Availability, boiler, 292 specification, 498
BS, 285
Baffles, 294, 304 Bunker, 444-445
Bagtest sampler, 466 Burner, oil, 343-347
Balanced draft, 478, 501 pulverized coal, 340-342
Barlow formula, 573 Bypass, trap, 614
Barometer corrections, 412-413 Bypass orifice, pump, 561
680 INDEX
Calcium, 512 Column lines, 91
Calorific value, fuel, 12 "Combined” efficiency, 367, 380
Calorimeter, bomb, 131-132 Combustion, 135-145, 663
gas, 133 approx, calculations, 144
Capacity factor, 42 Diesel engine, 158-161
Capability, 399 equipment (summary), 333
turbine, 377, 378 nature of, 258-259
Carbon dioride. See COa meters. process, 136
Carnot, 211, 220, 247" requirements, 333
Carryover, 515 sample examples, 140-144
Caustic embrittlement, 616 Combustion control, 499-503
Centipoise, 606 elements of, 501
Centrifugal fan, 484-487 Compression, unaflow engine, 373
Centrifugal fans, commercial classes of, Compression ignition, 155, 156
491 Compression ratio, 153, 155
Centrifugal pump, 552-563 Concrete, 90
Cetane number, 120, 160 Condensate depression, 418
Chain grate stoker, 336 Condensate pump, 427, 556
Charts, 630 Condensation, heat transfer, 274-277
progress, 24 Condensers, contact, 413-417
psychrometric, 672 surface, 417-423, 431-433
Check valve, 583, 585 Condensing water. See Circulating water
Chemicals (water treatment), 526 Condition line, turbine, 235, 239, 248, 388,
Chezy formula, 674 673
Chimney, 96, 493-499 Conductance, coefficients of, 276, 277
concrete, 99 Conduction formulae, 267
masonry, 100-105 furnace wall, 318-320
proportions, 100 heat, 265, 266-267
stability, 101-105 Conductivity, coefficient of
steel, 99 refractories, 319
Chimney draft, calc, for, 494-495 furnace insulation, 319
principle of, 494 pipe insulation, 599
Chimney foundation, 107 Conductivity (heat insulation), 399, 599
Chlorination, 431 Continuous blowdown, 523
Circulating water ^condenser), 427-429 Control, air-fuel ratio, 502
COa meters, 645-647 centrifugal pump, 427, 559-560
Coal, commercial sizing, 130 combustion, 499-503
properties, 124-131 feedwater, 563-567
purchasing specifications, 130 steam turbine, 375-376, 391-394
ranks, 126 Controllers, combustion, 504
storage, 441-445 Controls, draft fan, 489-490
typical analyses, 127 oil burner, 345-347
Coal bunker, 444-445 turbine, 404
Coal conveying, Convection, heat, 268-270, 273-277
classification, 442 Convection coefficient, calculation of, 278
equipment, 445-453 280
system design, 452 Conveyor design, coal, 449-453
Coefficient of conduction, 267, 319, 599 stoker, 336
furnace wall materials, 319 Cooling tower, 180-181
pipe insulations, 599 Cooling svstems, engine, 177-178
Coefficients of convection, separation of, Corliss, 371
433 Correction, orifice flow, 639
Cold spring, 595 Corrections, steam rate to guarantees, 411
INDEX 681
Correspondence, engineering, 27 Drag scraper, 443
Corrosion, 514-515 Drainage devices, pipe, 611-615
Cost elements in electric rates, 66 Drainage gradient, pipe, 608-609
Creep, 264 Drawings, power plant, 26, 27
Customer costs, 69 Drum internals, 327, 329
Cutoff ratio, 153 Dryness factor, 9
Cycle, Carnot, 211 Ducts, air, 479-480
closed, 204 velocity in, 480
conventional, 366 Duplex pump, 547
Diesel, 153 Duration, load, 40
gas turbine plant, 195, 202 Dust collectors, 462-466
1. C. engine, 150-151 sampler, 466
mercury vapor, 246 Duty, pump, 550
Rankine, 212 Dynamic head, 545
regenerative, 218
reheating, 236 Economizer, 329-331
steam engine, 365 EDR (Equivalent direct radiation), 295
two vs. four, 157 Effectiveness, heat exchanger, 202
Cycles, vapor, 209-212 Efficiency, Diesel engine, 163
Cyclone burner, 341( dust collector, 465
Cyclone dust separator, 463-464 gas turbine, 194, 197
generator (elec.), 675
Deaerator, 533-536 motor, 677
Deconcentration of boiler water, 522-525 steam generator, 348
Deinlein’s rule, 493 steam turbine, 379, 380
DEMA, 163, 184, 185 variable load, 46
Demand factor, 41 Efficiency formulae, boiler, 296
table of, 42 Carnot cycle, 211
Demineralization, 531 Diesel engine, 163
Density, air and flue gas, 495 electric motor, 677
fuel oil, 118, 119 fan, 485
Depreciation, 71 gas turbine, 196, 197
legal rates, 73 generator, 228
state practices, 74 pump, 546
Desuperheater, 325 Rankine cycle, 213, 214
Diagram factor, 366, 368 regenerative cycle, 219, 220
Diameter ratio, 638 reheating cycle, 237
Diesel cycle, 153 steam engine, 367
Direct-acting pump, 546-550 steam turbine, 379, 380
Disengagement area, 294 Ejector, steam jet, 425
Dissociation, 259 Electric distribution system, 28, 31, 41
Dissolved solids, 521 Electric rates, 65
Distribution system, electric, 28, 41 Electrical measurements, 651-653
Diversity factor, 4) Electrical precipitator, 464
table of, 42 Electron, 6
Draft, 474-479 Elements, table of, 664
components, 477 Elevator, coal, 446, 449,452
definitions, 474, 478 Embrittlement,- 516
drawings, power plant, 26, 27 Emergency trip valve, 376, 394, 405, 406
Draft fans, specification of, 491 Energy, definition, 1
Draft gauge, 636 nuclear binding, 7
Draft loss, fuel bed, 476 Energy equivalent of mass, 14
measurements, 481 Energy costs, direct, 68
682 INDEX
'‘Engine” efficiency, 367, 381-386 Flyash, 462
Engine vs. turbine, 360 Flywheel, drive from engine, 362
Engines, I. C., 149-153 Foaming, 515
Enthalpy, definition, 10 Footing, wall, 106
Entropy, definition, 10 Forced draft, 478, 486
epm, 5^ Foundation bolts. 111
EPRS, 286 Foundation, mass of, 108
Equivalent evaporation, 295 turbine, 110
Equivalent radiant heating surface, 285 Foundations, 105-111
Equivalent weight, 522 Four cycle, 157
Equivalent per million, 522 Friction factor, 604, 605, 607
ERS, 285 gSLS, 477
Evaporative cooling, 179-182 Friction loss, 227
Evaporator, 539-542 chimney, 494
Evase stack, 497 condenser, 428
Excess air, 137, 144 conduits (large) 674
Excess pressure regulator, 560 duct, 477
Exhaust, I. C. engine, 174-176 pipe, 604-608
Expansion, pipe, 592 water, 428-429
ratio of, 365 “from and at,” 295
thermal, 264 Fuel, engine, 118
Expansion bends, pipe, 593-596 definition, 116
Expansion joint, pipe, 592-593 Fuel ignition, Diesel, 156-157
Extraction, steam from turbine, 218, 226- Fuel oil, furnace, 122
227, 375, 380, 396 Fuels, occurrence, 3
Furnace arch, 337
Factor of evaporation, 295 Furnace, boiler, 313-320
Fan characteristics, 487 temperature, 317
control, 489-490 wall, 309
efficiency,'^485 wall types, 315-316
power, 4^ Furol viscosity, 119, 461
types, 486-487
Fans, draft, 484-492 Gas, fuel, 116
Feedwater, definition, 517 analysis, 127
heaters, 532-539 burners, 346
piping systems, 617 heating value test, 133-134
regulators, 563^67 loading (dust), 463
treatment summarized, 527 loop, description, 440-441
Firing quality, coal, 130 meter, 642
Fission, 260, 261 occurrence, 116
Fittings, pipe, 580, 582, 666, 667-669, 670 supply system, 458-459
Fixed costs, 66 turbine, 191-192
Fliinged fittings, 667-669 turbine plants, x)eak load, 53
Flanges, 577-678 Gate valve, 583, 584
standard dimensions, 665 Generator efficiency, 185-186, 228,380, 675
Flash cooler, 618 Globe valve, 583, 584, 585
Flight conveyor, 446, 448, 452 Go/emor, engine, 158, 364
Flow coefficient, 639 turbine, 391-394
Flow meter corrections, 639 Grindability, coal, 125
Flow meters, 643-644 Grouting, 112
Flue gas, quantity of, 140-143, 355, 478- Guarantee, correction for steam rate, 410,
479 411
analysis, 645-647 Guarantees, turbine, 410-411
INDEX 683
Hanger, pipe, 608-611 Ignition, 136
Hardness, water, 519 compression, 155, 156
Headers, boiler, 297 temperature, 259
steam pipe, 617 Increment rate, 249
Heads, definition of pump, 545-546 Indicated horsepower, 163, 367
Heat balance, calculations, 187-188 Indicator card, 161,162, 184,185, 365, 366,
definition, 210 369
Diesel, 164 Induced draft, 478, 487
method, 352 Industrial heat-power balance, 241
Rankine cycle, 212-216 Initial condensation, 366, 371
regenerative cycle, 230-233 Injection, Dien 1 oil, 156
steam generator, 350-355 Input-output method, 352
steam turbine, 389 Inspection, piping, 580-582
Heat conductivity. See Conductivity and Instrument, location, 625, 629
Coefficient of conduction. selection, 653-654
Heat Exchange Institute, 277, 423 Instrumentation, general, 624
Heat insulation, 597-602 Instruments, classified, 626-628
breeching, 483 Insulation, heat, 319, 597-603
specifications, 599-600 “Intemal’' water treatment, 526
Heat loss, bare pipe, 597, 671 (also see Ions, 521
Losses). Iron pipe, wrought, 572
Heat rate, 249, 380 Isotope, 6
Heat release rates, 314
Heat stream, 210, 216, 234 Jet- condenser, 414
Heat transfer, 264-266 Jet ejector, steam, 425
air preheater, 471-473
boiler, 282-284, 291, 292, 307, 310, 311 kB, 294
economizer, 330 Kem radius, 102
feedwater heater, 538-539
pipe covering, 600-602 Larry, coal, 442, 447, 451
radiant, 13, 282-284, 310, 318, 600 Leakage, condenser air, 423
surface condenser, 420-422 Lighting, plant, 93
waterwall, 310-311 Lime-soda water softener, 528-529
Heaters, feedwater, 222-226, 217, 532-539, Load curve, 37, 38, 40, 47, 49
537 Load dispatching, 50
Heating value, fuel, 12 Load factor, 40
oil, 121, 124 Load, variable, 45
Homologous equations, fan, 488 Log mtd, 272
pump, 554 Losses, boiler heat, 349
Hopkinson rate, 77 steam engine, 366
Horsepower, boiler, 294 steam turbine, 384
Diesel, 163 Lubrication, I. C. engine, 172-174
steam engine, 366, 367 pump, 548, 553
Hotwell, 417 steam engine, 363
HRT boiler, 300 turbine, 376, 402, 407-408
Humidity, 181, 672
Hunting, governor, 393 Mach number, 194
Hydraulic Institute, 554, 555, 558, 559 Magnesia insulation, 483, 598
Hydraulic ash sluice, 469-470 Manometer, 474-476, 638
radius, 477 Mass flow, 273, 322
Hydroelectric power, 51 Master regulator, 503
Materials, building, 91
Identification code, pipe, 616 Maximtim demand, 41
684 INDEX
Mean effective pressure, 365, 366 Performance, Diesel engine, 161-165
Mean temperature difference, 271-273 part load, 397-400
Mechanical drive turbine, 400-401 steam engine, 364-370
Mercury, h-s chart, 661 steam generator, 351
properties, 662 steam turbine, 379-390
Mercury cycle, 244-247 Performance evaluation, 655-656
Metering control, 502 Permanent hardness, 519
Micron size, definition, 462 Petroleum fuel, properties, 118
Mills, coal pulverizing, 455-457 pH scale, 519
Moderator, 260, 261 Pil)e, commercial, 572-575
Moisture effect, 385, 387 standards, 665
Mollier Diagram, steam, 16 Pipe fittings, screwed, 576
Mols, 138-139 flanged, 577
Motor, electric, 677 welded, 578
standard sizes, 677 Pipe «^lze, required, 603
Muffler, I. C. engine, 175-176 Pi])e bends, 503-596
Multiple effect, 539 draw mgs, 61S-620
flange.s, 577-578
insulation, 598
Nat. Assoc. Fan Mfgr.s, 490
joints, 575-581
Natural draft, 478
sizing, 573
Neutron, 6
standards, 574
No-load factor, 397, 308
strength, 573, 575, 609
Nozzle flow, ejector, 432
supports, 608-611
part load, 396
thread, standard, 576
NPSH, 546
Piping, «team engine, 363
Nuclear energy, 259-261
steam turbine, 402, 403
Nucleus, atomic, 6
turbine oil, 404
Nusselt number, 269
system, general, 571
systems, design of, 615-618
Oil burner, 343-347 Piston speed, 368
controls, 345-347 Pitot tube, 475
Oil, fuel, 116 Pivoted bucket conveyor, 446, 447
fuel supply system, 171 Plutonium, 261
lubricating, 174 Polytrojiic jirocesses, ecjuation'-, 676
preheat for fuel, 460 Positioning <'ontrol, 503
turbine lubricating, 408 Potentiometer, 633
Oil supply system, 458-461 Pots, condensation, 640
Oil qualities, 118-120, 174 Powder, electrical, 14
Open heater, 533 Power plant engineering, definition, 20
Orifice, 638, 641 Power pump, 550
Orifice flow formulae, 639 Prandtl number, 269
Orrok, 277, 284 Preferred standards, 377, 378
Orsat, 645 Preheat requirements, oil, 460
Outdoor-type plant, 96 Pressure, boiler, 263
Overfeed stoker, 334 instruments, 634
measurement, 634-636
Parallel operation, 393 ratio, optimum, 197
Part load, draft fans, 489 regulating valve, 566, 587, 588
steam turbine, 248, 396-400 scales, 9
Pass, condenser tube, 420 temperature chart, steam, 676
Peak load, carrying, 51-57 Prime mover, 359
Percent rating (boiler), 295 Priming, 515
 INDEX
Product's of combustion, 13f> Regenentivif c>clc, 216-227
Profits, 70 steam turbines for, 226, 227
Propcrt\, physical, 8 part load m, 248
]>roce‘-v, working, 14, 67f> Regenerative hcatcT, air, 470, 474
Proton, 0 Regulating valve, 587, 588
ProxiiUMtc anaivsi'-, coil, 12r), 127, 134, 135 Reheat, gas turbine plan*, 202
Pmchromctrir chart, 072 stage, 385
J^ulv(‘nzo(l coal, 340 Reheat factor, 390
contra! v& unit sjstoin, 454-455 R(‘heat vayior cycle, 234-237
fineness, 457 Reinjection, cinder, 465
see also Gnndability Relatee humidity, 181
tran‘*portation, 454-455 Re'^rv(‘ eajiaeitv, plant, 48
I^ilvcrizer, coal, 455-457 Rf'sorvoir at orifice plate, 640
Piun]), air, 425, 426 Re\ nolds number, 269
condensate, 427 Ringelmann chart, 145
characteristics, 553 Rotarv cup biiriHT, 343
fuel oil, 170 Rotor, turbine*, 374
governor, 548, 5f)0 Ruth's accumulator, 54
installation of, 544, 562
layout, 545
test 563 Safe 1 leanng ])OW(‘r, 105
Piinijis 543-563 Safety val\(, 586, 589-592
Purifier, steam, 329 Sa\bolt viseosit\, 1J9
I*\ rometrv, 631 Scaling, 512-514
Schedule* numl)e*r, 573
Qualification of welder^, 57M Screwed fittings, 666 ,
Quality, 9 Securities Exchange* Conmu-siem, 30
Separator,
Radial brick, 497, 498 flvash, 463-464
Ua<liant energy, 282 oil, 326
Radiant superheater, 322 steam, 612
Radiation, 281-285 Setting, boiler, 312
energy, 13 Settling \elorit\, flvash, 463
(formula), 318 Silencer, I (' engine (*\haust, 175-176
loss (boiler chart), 353 Silica (feedwater), 531
RadiU'*, hydraulic, 477 Sinking funel, 71
Rankine cycle, 212-216 Skip hoist, 44t>, 447
engine, 366 Slagging factor, 285
Rates, electric, 65 Slip, pump, 548
types of electric, 77 Smoke, 145
Rating, steam engino, 370 Smoke meters, 647
steam turbine, 377 Soap test, 519
Ratio, comiiression, 153 Soil, bearing power, 105
cutoff, 153, 154 Soot blowef, 327-31^
Ratio of expansion, 365 Specifications, 25
apparent, 365 Specific heat, 137, 138
Reactor, nuclear, 260, 262 Specific speed, 546, 554, 655
Reciprocating pumps, 546-551 Speed regulation, 371, 393
Records, power plant, 655-656 Spiral conveyor, 44tb 448, 452
Reentry turbine, 401 Spontaneous combustion in sMorwl coid,
Recuperative heater, 470 442
Refractories, 316, 320 Spreader stoker, 335
Regeneration, gas turbine plant, 201, 202 aSF, 119
686 INDF
SSU, 119 Superheat control, 324-326
Stack, 497 steam, 320-326
Stage efficiency, 386-388, 390 Superheater, 322, 323, 324
Starting, steam turbine, 405, 406 Superheater heat transfer, 278-280
gas turbine, 199 Superposition, thermal, 238
I. C. engine, 177 Support, pipe, 608-611
Static draft, 485 Surface condenser, 417-423, 431-433
Static head, pump, 545-546 Surface heater, feedwater, 536-539
Steam consumption, oil burner, 344 Symbols, flow diagrams, 660
reciprocating pump, 547, 550 gas turbine plant, 200
steam engine, 369, 370, 372 pipe drawings, 618, 619
steam turbine, 399 steam plant, 660
Steam engine, 360-373 Synchronous speed of motors, 677
Steam flow meter, 638-644 System, see system in question.
Steam generator, 290
Steam rate, 367, 371, 379, 380, 399
engine, 369, 370 Tanks, blowoff, 525
Steam sp)eed, 362 fuel, 169, 170
Steam temperature, high vacuum, 676 storage, 459
Steam turbine, advantages, 374 Taps, pipe (flowmeter), 640
blade friction, 384, 385, 386 TDH, 546
bleeder, 226, 227 TDS, 521
classification, 376 Temperature, high, 264
compound, 375 Temperature difference, heat transfer, 271-
condition line, 388, 673 273
effect of change of throttle state, 396, Temperature measurement, 630-634
397 Temperature scales, 9
elements, 374 Temporary hardness, 519
engine efficiency, 382 Terminal temperature difference, 227, 229
extraction valve gear, 402 415, 422
gland, 375 feedwater heater, 536, 538
governor, 391, 396 Test, condenser, 431-433
heat rate, 380 pump, 563
in parallel, 393 turbine, 408-412
losses, 384 Test stations, boiler, 350
lubrication, 376, 407-408 Tests, draft kn, 493
moisture loss, 386, 387 water, 517-524
no-load steam, 397, 398 Thermal efficiency. See Efficiency.
overload, 376, 377 Thermocouples, 167, 631, 632, 633
part load, 396-400 Thermometer types, 632
prediction of condition line, 248, 388 Total draft, 485
preferred standards, 263, 378 Tramp iron separator, 448, 450
regenerative, 226, 227 Trap installation, 614
stage efficiency, 385 Traps, principles of, 613
standard sizes, 263, 378 steam, 613-614
starting and stopping, 405-406 Traveling grate stoker, 336
steam rate, 379 Trim, boiler, 326
tests, 408-412 Triplex pump, 550-551
thermal efficiency, 379-390 Tube cleaning, condenser, 431
velocity ratio, blades, 386 Tube sheets, 419
Willans line, 397 Tubes, condenser, 418-419
Stokers, 334-339 Turbine. See Steam turbine.
Supercharge, engine, 160-176 Turbine foundation, 110
fDEX
Turbine installation, 402-404 Volatile matter, coal, 127, KU
pump, 552 V-notch meter, 637
tyi^s, 375, 376 Volt, 13
Turbine vs. engine, 360 Voltmeter, 651
Turbulence, 136, 341 Volumetric efficiency, 549
Two-cycle, 157
Water, natural impurities in, 511-512
U-bends, 593-596 Water column, 327
Ultimate analysis, coal, 127, 128 Water loop, described, 510-511
Unaflow engine, 371-373
Water horsepower, 546
Underfeed stoker. 338-339
softeners, 528-531
Union, pipe, 577 testing, 517-524
Uranium, 260 treatment, 525-532
U. S. degree (hardness), 519
Waterwall, boiler, 306-311
Use factor, 43
circulation, 306
Useful lives, 73
heat transfer, 310
Utility company, 29, 30
Watt, 13
Watthour meter, 76, 78
Vacuum, 412-413
Wattmeters, 651-652, 653
Vacuum breaker, 409
Weighing, coal, 448, 449
pump, 425
Welded pipe, 578, 579
Valves, common, 583-586
fittings, 670
Van Stone joint, 577
Welding control, pipe, 579
Vapor cycle. See Cycle.
Willans line, 367, 397, 399, 400
Variable load, 45
Wind power, 5
‘ elocity, average pipe flow, 603
Wind pressure on chimney, 98
\ elocity by pitot static, 475-476
Wind stress, chimney, 101
Velocity head, 477
Work, mechanical, 10
yena contracts, 640
Wrought iron pipe, 572
’mturi meter, 637
iscosity, data, 270, 606
fuel oil, 119 Zeolite, 530-531