0% found this document useful (0 votes)
25 views376 pages

Elementary Thermodynamics by Faires

This document is the third edition of 'Elementary Thermodynamics' by Virgil Moring Faires, aimed at providing a comprehensive yet concise resource for short courses in thermodynamics. It covers essential topics such as thermodynamic properties, conservation of energy, and the behavior of ideal gases, while offering a clear structure that allows instructors to select relevant content for their courses. The book includes a variety of problems and references to support the learning process, making it a valuable tool for both educators and students in the field of mechanical engineering.

Uploaded by

r zero k 1 be
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
25 views376 pages

Elementary Thermodynamics by Faires

This document is the third edition of 'Elementary Thermodynamics' by Virgil Moring Faires, aimed at providing a comprehensive yet concise resource for short courses in thermodynamics. It covers essential topics such as thermodynamic properties, conservation of energy, and the behavior of ideal gases, while offering a clear structure that allows instructors to select relevant content for their courses. The book includes a variety of problems and references to support the learning process, making it a valuable tool for both educators and students in the field of mechanical engineering.

Uploaded by

r zero k 1 be
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

ELEMENTARY

THERMOD YNAMICS
ELEMENTAR Y
THERMOD YNAMICS

VIRGIL MORING FAIRES


Professor of Mechanical Engineering,
North Carolina State College r

Third Edition

THE MACMILLAN COMPANY • New York


Third Edition ©\ M Faires 1957

AJI rights reserved —no pvt of this book may be reproduced in any
form without permission in writing from the publisher except by a
reviewer who wishes to quote brief passages in connection with a
review written for inclusion in magazine or newspaper
Printed in the United States of America

first Printing

Previous editions copyright 1936 lUJS^nd 1918 by V M Faires

Library of Congress catalog card number 57-0893


PREFACE TO THE THIRD EDITION
This book, extracted from my Thermodynamics ,
is intended for short
courses. Its feature is that its explanations are not shortened, but are as
complete and clear as in a full length work. Thus this approach is expected
to appeal to those who wish to offer a short theoretical course by omitting
topics, rather than by covering hurriedly many topics. Even though the
book is relatively short, too much material is probably included for the
shortest courses, in which case, the teacher has some choice of topics to
be covered.
While ideal gases are covered first (so as not to confuse the student with
the consequences of the phenomena of" evaporation and condensation), most
of the basic theory has been covered at the outset; and there is a clearer
indication of which principles and methods apply to gases and which to
vapors. I think that the result of the reorganization and the changed
viewpoint will be that the student leaves the short course with a more
significant comprehension of the subject.
The problems are placed at the rear; somewhat more than enough for
a course. The teacher may be interested in noting that many of the prob-
lems in Problem s on Thermodynamics by Faires, Brewer, Simmang are
appropriate to this book, and that moreover this problem book contains
a set of steam tables and other useful reference material.
All comments, suggestions, and notices of errors will be gratefully received.
Since it is always difficult to decide what matter to include in a book for a
short course in thermodynamics, opinions on this subject will be welcomed.

VMF
SYMBOLS
With few exceptions, the ASA standard symbols have been used, but the
choice of symbols has been influenced by those used in the ASHAE Guide
and in Keenan and Kaye Gas Tables.

A area; mass number,


8 acceleration.
tB volumetric percentage of a component in a mixture; represents
a heat unit,
O a constant.
C) Cv molar specific heats at constant pressure and at constant
volume.
° clearance ratio in engines and compressors; specific heat,
Btu/lb-°F; c p specific heat at constant pressure; c„, specific
,

heat at constant volume; c n polytropic specific heat,


,

t) diameter.
a dimension; distance.
fejR,
general symbol for energy; total stored energy of a system; E c,

electrical energy; E
chemical energy and energy chargeable
c,

against an engine in obtaining the thermal efficiency; Ef ,

energy dissipated in friction (not a different kind of energy)


Ea ,
available energy; E„, unavailable energy; E t, total stored
energy.
cu thermal efficiency; eb brake thermal efficiency; e,-, indicated
,

thermal efficiency; ex., combined or over-all thermal efficiency;


e„, nozzle efficiency.
tq force or total load; FA ,
angle factor in radiation; FE ,
emissivity
factor.
acceleration of gravity; g0 , standard acceleration of gravity,
Jq'Q total enthalpy of w lb. of substance; H
m enthalpy of a mixture,
,

specific enthalpy; h/, specific enthalpy of saturated liquid;


h g specific enthalpy of saturated vapor; h Sg change of specific
, ,

enthalpy during evaporation; h S 2 enthalpy of saturated vapor


,

at state 2, etc.; h, enthalpy of one mol of substance; h°,


enthalpy in standard state (77°F).
vii
vtu SYMBOLS
h film coefficient.
hp horsepower, bhp, brake horsepower, thp, indicated horsepower,
fhp, friction horsepower
/ correction factor for initial velocity in nozzles
J Joule’s constant (~ 778)
K kinetic energy.
k the ratio cjc.
k conductivity
L distance, length, stroke of piston, represents unit of length,
L", stroke m
inches
M Mach number
M molecular weight, lb /mol, Jl/„ molecular w eight of gas .Y, etc
9TC momentum
m mass in slugs, percentage steam bled in regenerative cycles,
exponent in pVm = C, which defines the pV relation at the end
points of an irreversible process
N number of anything, number of moJs of a gas, number of port er
cycles per minute completed by an engine
n revolutions per minute (rpm) polytropic exponent
,

P potential energy, represents the unit of a pound


p unit pressure (psi or psf), p n mean effective pressure (mep),
,

Pmi, indicated mep, p„B, brake mep, p„, stagnation or impact


pressure, p„ relative pressure
Q heat, Q Ai heat added, (?u, heat rejected, Qr radiated heat, etc
q heating value, gi, lower heating value, higher heating value
R specific gas constant, pv/T R, umv ersal gas constant
r radius, ratios, reheat factor, rk lsentropic compression ratio,
,

rc , cutoff ratio, r p pressure ratio, r„ expansion ratio


,

S total entropy
s specific entropy, 8/, specific entropy of saturated liquid, s„,

specificentropy of saturated vapor, s/„, change of specific


entropy during vaporization, sg2 specific entropy of saturated
,

vapor in state 2, etc


T absolute temperature, usually degrees Rankine, Ta stagnation
temperature
t temperature usually jn degrees Fahrenheit, time, t„, stagna-
tion temperature
U total internal energy, over-all transmittance
u specific internal energy, it/, for saturated liquid, for satu
rated vapor, u/t , change during vaporization, u/ 2 , for a satu-
rated liquid in state 2, etc
internal energy of one mol of substance
total volume, Vd, displacement volume
v specific volume, i„ relative volume
SYMBOLS tx

v volume of one mol.


v velocity; speed.
TF work; TFr, indicated work; WB ,
brake work; ]V K ,
combined
work TF/, flow work; TF„, pump work.
;

w mass in pounds; mass flow per unit time; W;, specific fuel con-
sumption; Wb, brake steam rate; w„ indicated steam rate; wk ,

combined steam rate.


x quality of a two-phase system.
y percentage (or fraction) of liquid in a two-phase system,
z altitude; potential energy of one pound, ft-lb./lb.
Z atomic number.
a (alpha) constant in specific heat equation; absorptivity,
d (beta) constant in specific heat equation coefficient of thermal expan-
;

sion; bulk modulus.

y (gamma) constant in specific heat equation; coefficient of performance;


specific weight; angle.
6 (delta) angle.
e (epsilon) effectiveness; emissivity.
7j (eta) efficiency ratios; engine efficiency; combustion efficiency; %,
brake engine efficiency ij c compression efficiency (adiabatic if
; ,

not qualified); j^, discharge coefficient; ij„ indicated engine


efficiency; 7 combined engine efficiency; ij m mechanical effi-
j;., ,

ciency; jj„, nozzle velocity coefficient; ti p, propulsive efficiency,


pump efficiency; y r effectiveness of regenerative process, effi-
,

ciency of reaction blades; turbine stage efficiency; ij„ volu-


metric efficiency.
0 (theta) represents unit of temperature; angle.
y. (mu) degree of saturation; absolute viscosity,
ir (pi) 3.1416 . . .

p (rho) density,
cr (sigma) Stefan-Boltzmann constant,
r (tau) time; represents a unit of time.
4> (phi) relative humidity; angle; used to mean junction of.

co (omega) humidity ratio; angle.


A (delta) indicates a difference or a change of value, At = change of
temperature or difference in temperatures, in accordance with
the context.
ABBREVIATIONS

A/F air fuel ratio cated horsepower, fhp


AIChE American Institute of friction horsepower
Chemical Engineers ICE internal combustion
ASA American Standards engine
Association ID inside diameter
ASME American Society of imep indicated mean effective
Mechanical Engineers pressure
ASHAE American Society of kg kilogram
Heating and Air Con Jew kilowatt
ditioning Engineers LMTD loganthnuc mean tem
atm atmospheres a unit of perature difference
pressure in natural logarithm
BDC bottom dead center (base e)
bhp brake horsepower LOCE law of consen ation of
bmep brake mean effective energy
pressure log logarithm to the base 10
Btu British thermal unit LP low pressure
cfm cubic feet per minute mep mean effective pressure

cgs centimeter gram second mev million electron volts


system mph miles per hour
cm centimeter OD outside diameter
cpm cycles per minute psf pounds per square foot
cps cycles per second psi pounds per square inch
da dry ait psia pounds per square inch
dg dry gas absolute
electron volt psig pounds per square inch
F/A fuel air ratio gage
fpm feet per minute rpm re\ olutions per minute
fps feet per second rps revolutions per second
fps 2 feet per second second TDC top dead center
gpm gallons per minute TEL tetraethyl lead
HP high pressure SAE Society of Automotive

P horsepower blip brake Engineers


horsepower, lhp mdi V vapor
X
CONTENTS
Preface y

Symbols
1. vii

Abbreviations x

Thermodynamic Properties 1

1. Introduction. 2. The System and the Working Substance. 3.


The Pure Substance. 4. Properties and State. 5. Systems of
Units. 6. Pressure. 7. Specific Vo lume and D ensity. 8. Tem-
perature. 9. Process. 1(3. Cycle s. II. Closure.

2. Conservation of Energy 15

12. Introduction. Measuring Energy. 14. Mechanical Po-


13.
tential Energy. 15. Mechanical Kinetic Energy. 16. Inte rnal
Energy. 17. Work. 18. Work of a Nonflow System. 19. Heat.
20. Flow Work. 21. Other Forms of Energy. 22. Classification
of Systems. 23. Continuity of Mass. 24. Conservation of En-
ergy. 25. Energy Equations for Closed Systems. 26. Steady
Flow Energy Equation. 27. Enthalpy. 28. Steady Flow Equa-
tion with Enthalpy. 29. Applications of the Steady Flow Equa-
tion. 30. Example. 31. Power. 32. Perpetual Motion of the
First Kind. 33. Closure.

3. Energy Relations 37

34. Introduction. 35. Specific H eat. 36. Constant Volume Spe-


cificHeat. 37. Constant Pressure Specific Heat. 38. Reversibil-
ity. 39. Entropy. 40. Temperature-Entropy Coordinates. 41.

Sp dV for Reversible Steady Flow Processes. 42. Other Energy


Relations. 43. Molar Specific Heats. 44. Variable Specific
Heats. 45. Example. 46. Mean Specific Heats. 47. Steady Flow
of More than One Mass Stream. 48. Variable Flow Process. 49.
Frictional Energy. 50. Energy Equation for Flow of Incompres-
sible Fluids. 51. Closure.
xt
CONTENTS
4 The Ideal Gas 55

52 Introduction 53 Boyle's Law 54 Charles’ Law 55


Equation of State 0/ an Ideal Gas 56 The Gas Constant 57
Joule s Law 58 Internal Energy of an Ideal Gas 59 Enthalpy
of an Ideal Gas 60 Example 61 Relation between cp and e,
62 Entropy Change of an Ideal Gas 63 Closure

.J^-Processes of Ideal Gases 67

64 Introduction 65 Constant Volume Process 66 Example


Constant Volume 67 Constant Pressure Process 68 Example
Constant Pressure 69 Isother mal Process 70 Adiabatic Proc-
ess— Isentropic Process 71 Irreversible \diabatic with Work
72 Example 73 Polytropic Process 74 Example Polytropic
Process 75 Effect of Varying n in Polytropic Equations 76
Throttling Process 77 Isentropic Process with "\ anable Specific
Heats 78 Example 79 Closure

Cycle Analysis and the Second Law 90

80 Introduction 81 Elements of a Cycle 82 Work of a Cycle


83 Thermal Efficiency 84 The Carnot Cycl e 85 Carnot Cj
cle with Ideal Gas as the W orking Substance 86 Mean EfTectiv e
Pressure 87 Ratio of Expansion Ratio oi Compression Pressure
Ratio 88 Analysis of Cycles 89 Example 90 The Ericsson
Cycle 91 Reversible Cycle 92 The Reversed Carnot C> clc
93 Reversible Engine Most Efficient 94 Example 9o Exter
nal Irreversibility 96 Therm odyn amic Temper ature 97 \ ar
iou s^ Stat e ment s of the Second ^Law Thermody namics 98
Available and Unavailable Energy 99 Example 100 Entropy
from the Carnot Cycle 101 Clausius Inequality 102 Entropy
Production 103 Example Loss of Available Energy with Heat
Transfer 104 Example Heat Exchanger 105 Equilibrium
106 Third Law of Thermodynamics 107 Perpetual Motion of
the Second Kind 108 Closure

7 Compression and Expansion of Gases 122

109 Introduction 110 The Indicator Card 111 Mean Effec


tive Pressure and Ihp 112 W ork of a Compressor 113 Work
from Conventional Diagram 114 Clearance and Clearance Vol-
ume 115 W ork of the Conventional Card with Clearance 116
Tree Air 117 Capacityand Volumetric Efficiency US Conven
tional Volu metric Efficiency 119 Preferred Compressi on Curv es
120 Example 121 Efficiencies 122 Multistage Compression
CONTENTS xiii

123. Energy Diagram for Two-Stage Compression. 124. Example.


125. Gas Tables. 126. Example. 127. Closure.

8. The Gas Turbine and Turbo-Jet 150


12S. Introduction. of a Simple Gas Turbine
129. Operation
Power Plant. Standard Brayton Cycle. 131. Inter-
130. Air
mediate Temperature for Maximum Work. 132. Brayton Cycle
with Fluid Friction. 133. Air-Cycle Efficient with Fluid Friction.
134. Example. 135. Heating Value of Fuels. 136. The Combus-
tion Process. 137. Regenerative Heating —
Ideal Cycle. 138. Ef-
fectiveness of Regenerator. 139. Other Variations of the Brayton
Cycle. 140. Jet Propulsion. 141. Work from the Impulse Mo-
mentum Principle. 142. Efficiencies of Turbo-jets. 143. Closure.

9. Internal Combustion Engines 171

144. Introduction. 145. The Four-Stroke Cycle. 146. The Otto


Cycle. Standards of Comparison. 148. Clearance
147. Ideal
Volume. 149. Example. 150. Diesel Cycle. 151. Fuel Cutoff
Ratio. 152. The Two-Stroke Cycle. 153. Brake Power. 154.
Example. 155. Thermal Efficiencies. 156. Engine Efficiencies.
157. Actual Thermal Efficiency. 158. Mechanical Efficiency.
159. Volumetric Efficiency. 160. Closure.

10. Liquids and Vapors 191

161. Introduction. 162. Vaporizing a Liquid at Constant Pres-


sure. 163. Vapor Tables. 164. The pv and Ts Planes. 165. In-
ternal Energy. 166. Properties of a Wet Mixture. 167. Prop-
ertiesof Superheated Vapors. 168. Compressed Liquid. 169.
Examples: Enthalpy of Compressed Water. 170. Other Vapors.
171. The Ts Diagram Including the Solid Phase. 172. Closure.

11. Processes of Vapors 209

173. Introduction. Constant Pressure Process. 175. Con-


174.
stant Volume Process. and Isentropic Processes.
176. Adiabatic
177. Example: Isentropic Process. 178. Lost Work and Unavail-
able Energy. 179. Example: Irreversible Adiabatic Process. 180.
Isothermal Process. 181. Polytropic Process. 182. Throttling
Process. 183. Calorimeters. 184. Mollier Diagram. 185. Heat
Exchanger Process. 186. Output of Boilers. 187. Closure.

12. Vapor Cycles .and Engines 222

1SS. Introduction. 189. Ideal Cycle, Ideal Engine. 190. Ran-


kine Cycle and Engine. 191. Rankine Efficiencies. 192. Mep
CONTENTS
and Ste3in Rates 193 Efficiencies in Terms of Steam Rates
194 Effect of End Conditions on the Thermal Efficiency of Ran-
kine Engine 195 Incomplete-Expansion Engine 196 Exam-
ple 197 Enthalpy of Exhaust Steam 19S Example 199
Loss Due to Friction of Flow 200 Example Lost Work by
Throttling and Heat Loss. 201 Modern Cycles 202 The
Ideal Reheat Cjcle 203 Regeneration 204 Regenerate e Cy-
cle 205 Idea] Reheat Regenerative Cjcle 206 Converting
Mass into Energy 207 The Reactor 20S Closure

^13 Nozzles 255

209 Introduction 210 Flow through Nozzles 211 Stagnation


Properties 212 Mass of Flow through Nozzle 213 Example
Air Nozzle 214 Equilibrium Flow of a Vapor 215 Example
Steam Nozzle Equilibrium Flow 216 Variation of Velocity,
Specific Volume, and Area of Section of a Nozzle 217 Critical
Pressure in a Nozzle 218 Values of Critical Pressures 219
Throat Velocity 220 Example Throat Size 221 Supersatu
rated Flow 222 Example Supersaturated Flow 223 Over-
expansion and Underexpansion 224 Losses Nozzlesm 225
Coefficient of Discharge 226 Estimated Actual Volume 227
Example 228 Diffuser 229 Ejector 230 Closure

Reversed Cvcle 276

231 Introduction 232 Carnot Cycle for Refrigeration 233


Conclusions from the Carnot Cycle 234 Definitions and Rat
mgs 235 Vapor Compression System 236 Displacement of
the Compressor 237 Example 238 Refrigerants 239 Ab-
sorption Systems of Refrigeration 240 Variations of Basic Refng
erating Cycles 241 Liquefying Gases 242 Reversed Cycle for
Heating 243 Closure

15 Gxseous Mixtures 293

244 Introduction 245 Characteristic Equation of Ideal Gas in


Terms of Mol» 240 Mixtures of Gases 247 Dalton's Law of
Partial Pressures 248 Molecular Height and Gas Constant for
Mixture 249 Example 250 Energy Changes of a "Mixture of
Gases 251 Gas-Vapor "Mixtures 2o2 Dew Point 253 Rela
tive Humidity 254 Humidity Ratio 255 Wet-Bulb Tem-
perature 256 Adiabatic Saturation Process 257 Enthalpy of
Superheated Steam at Low Pressure 25S Example 2o9 En-
thalpy and Total Heat of an Mr-Steam Mixture 260 Example
261 Internal Energy and Entropy 262 Mixtures Other than
Air-Steam. 263. Example: Fuel-Air AExture. 264. Example:
Constant Total Pressure (Steady Flow) . 26-5. Closure.

315

266. Introduction. 267. Conduction. 268. Fourier's Law. 269.


Variation of Thermal Conductivity. 270. Conduction through a
Plane Wall. 271. Film Heat Transfer from
CoeScient. 272.
Fluid to Fluid. Curved "Wall. 274.
273. Conduction through
Example. 275. Logarithmic A lean Temperature Difference. 276.
Thermal Radiation. 277. Stefan-Boltzmann Law. 278. Closure.

;t of References 333

PEND IX
A_ Problems 339
B. Table XIV. Useful Constants 362
XXL Saturation Properties of Freon 12 363
XXL Properties of Superheated Ereon 12 363
XXTL Saturated Ammonia 364
XXHL Superheated Ammonia 365
Jlollier Diagram for Steam following 366
Psychrometric Chart following 366

Ljtdex 367
ELEMENTARY
THERMODYNAMICS
1

THERMODYNAMIC PROPERTIES

Introduction. Thermodynamics is ihal branch of physics which treats


of various phenomena of energy, and especially of the laios of transformations
of heat into other forms of energy, and vice versa. Examples of such trans-
lormations are: the process of converting heat into work, of converting
electricity into heat (because of a resistance), and any chemical reaction
(heat of reaction). In engineering, the science is conventionally divided
into two parts: chemical thermodynamics, which is devoted largely to reac-
tions and solutions, and the thermodynamics concerned with the production
and use of work and power. This division is fortuitous, as the areas merge
at many points. Moreover, there are aspects of the science that may not be
included in either of the foregoing parts, notably the thermodynamics of
atomic processes, those processes in which the structure of the atom is
changed.
This book is the story of power, the demand for which is increasing at
geometric rates. Using the basic principles, we shall study the theory of
internal combustion engines, gas turbines, air compressors, steam turbines,
refrigeration, air conditioning, and a few miscellaneous topics of related
interest such as the flow of fluids and heat transfer. Insofar as possible,
we shall apply the theory to actual engineering situations shortly after the
theory has been developed. This plan of study should engender the reader's
interest and result in a more intimate understanding of the implications of
the theory. It is assumed in the writing that the reader is familiar with
the contents of an introductory book on engineering physics and that he
can handle simple calculus.
Thermodynamics takes the macroscopic or large-scale view rather than
the microscopic (1).* The science has been developed without regard to

* Italicized Arabic numerals designate a reference at the end of the book. In many
1
2 THERMODYNAMIC PROPERTIES [C7t 1

the structure of matter A


pressure gage indicates the average pressure of a
gas at rest in a container
Since a finite quantity of a gas is composed of a
tremendous number of molecules about 4 5 X
10"° in one cubic inch at
atmospheric temperature and pressure each moving with its independent
velocity it may occur (microscopically) that the pressure at a point is
momentarily quite high (or low) because by chance a number of high
velocity (or low velocity) molecules have just happened to strike that point
However the area of that part of the instru ment which the molecules
a ctually strike is so large compared with the area of the point, t hat the
Instrument gives a statisTleafanswer that is t it measur es the force ev erted
by the pounding of a large number of molecules^ by t ot a relatn ely sm alf per
c entago of the total numbe r The actual number striking the instrument is
large enough to provide an adequate statistical sample Hence we safely
conclude that the same pressure would be indicated at any and all other
positions in the container (effect of grav ity negligible) Similarly the tem
perature which indicates something of the level of energy in a body is a
statistical measurement obtained from the action of a large sample of
molecules (the macroscopic view) and it tells nothing of the energy of a
single molecule (the microscopic view —
really submicroscopic)
In engineering applications we" often find it expedient to neglect some
macroscopic effects The temperature of a substance in a container is
virtually a certain value but we may knowingly neglect the effect of a
localized spot where the temperature is significantly higher or lower Since
the engineering o f thermody na mics is the art of applying the science ot
therm ody namics, o ne's Judgment and experience are used to decide whethe r
or not such macroscopic alffm'tmt'US may be saieiy ignore^
At the outset oi our study we should become familiar with certain
concepts at the foundation of the subject

"2 The System and the Working Substance A system is that portion of
the universe an atom or a galaxy or some certain quantity of matter, w hich
we specifically wash to study * IIts a region enclosed by specified boundaries
or by imaginary but definite mental boundaries We isolate a sy stem or region
in our minds because we wish to study transformations of energy occurring

cases the reference « ill be an acknowledgement of the use of an idea or other information
The interested reader should find elaboration or a related discussion in tl e w ork cited
* As Humpty Dumpty said to Alice in Wonderland \M en / use a word it means

just what I choose it to mean neither more nor less It is only natural that as a
science develops there should be troubles with the meanings of words On the whole
the science of thermodynamics is well developed but unfortunately there are several
words which do not jet have a angle precise and generally accepted definition In
general the reader should expect to find now and then the same w ord being used by
various writers to mean different things and different words to mean the same thing
and he should be alert to this situat on E F Obert in an ASEE paper on Teaching
the concepts of state properties boundaries systems
etc (2) has highlighted variations in
terms
the definitions of some important thermodj namic
§ S] THE PURE SUBSTANCE S

within the boundaries, and the passage of energy (if any) across the sys-
tem’s boundaries, to or from the surroundings, which passage may or may
not be accompanied by the movement of matter across the boundaries. The
region about the system, outside of the boundaries, is called the surround-
all

ings or environment. The free body in analytic mechanics is a system,


in which the mode of anal 3'sis is based on Newton’s laws of motion. The
principal mode of analysis in thermodynamics is on the basis of energy laws.
•S*.The working substance in thermodynamics is a fluid in which energy can be

stored and from which energy can be removed. The purpose of storing and
removing energy is to bring about desired energy transformations, such
as transforming heat into work. A large part of this study is devoted to
learning about this particular transformation. Examples of working sub-
stances are: steam in a steam turbine, air in an air compressor, air-and-fuel
mixture in an internal combustion engine, and water in a hydraulic turbine.^
A working substance, or a certain mass of working substance, may be
(and in this book usually is) taken as the system. However, the word
system has a more general connotation. J^The system which the mind has
surrounded with imaginary boundaries may be an electric motor, a storage
batten', or a shaft and propeller combination on an airplane.-] The mind can
place the boundaries of the system wherever desired, as easily in one position
as another. If a block is slid i ng down a rough incline, we may take the
block alone as the s ystem, or the block and the incline to gether. In another
situation, the system may be the turbine (which would include effects at the
bearings) or it may be the region within the turbine through which the work-
ing substance passes. In the case of a reciprocating engine, the system may
be taken as the fluid in the cylinder; it may be the fluid plus the piston and
piston rod (which would enclose the frictional effects at glands and cylinder
walls) it may be the whole engine, or the whole power plant.
;
Whatever
the sj'stem is, it is important to maintain a precise mental picture of its
limits or boundaries. While considering a particular problem, the reader
should make a conscious effort to define the boundaries and then to keep
them in mind.
In a closed system, matter does not cross the boundaries. Energy may,
however, flow into or out of this .system . An open system is one where
matter passes across its boundaries. Energy may also pass across the bound-
aries, either with the flow of mass or separately. The mass of matter in an
open system may vary with time or be constant. A special case with which
we shall frequently deal is one in which the rate of flow of mass into the
system is constant and equal to that leaving the system (constant mass in
the system), called a steady flow system.

3. The Pure Substance. We shall consider that a pure substance is a


single substance which retains an unvarying molecular structure or a fixed
solution of homogeneous substances, each of which retains an unvarying molecu-
4 THERMODYNAMIC PROPERTIES [Ch 1

far structure (?) Thus a system of pure oxygen is a pure substance, as is


dry air (m the gaseous state) which is largely a solution of oxygen and
nitrogen with fixed percentages of each component A system of tee water
and water vapor may be considered to be a pure substance, since the m olecu-
la r structure in all parts is the same (substantially so, although there are
other water molecules than HjO) E xamples of syste ms which are not pure
substances are a mixt ure of oil and water, because they anil not go into
solution, liquefi ed dry air in conta ct with its vapor because, since the con
densing temperatures (boiling points) of oxygen and nitrogen are different,
the relatne percentages of oxygen and nitrogen in the liquid and in the
vapor would change— that is, the solutions before and after partial condcn
sation would not bo the same, also, a system which includes any chemical
process such as combustion would not be a pure substance during the
process, because the molecular structures before and after the process are
different
In genera! a pure substance may exist in any of three phases the 6ohd
phase, the liquid phase and the gaseous phase Under certain conditions
(§ 230) all three phases may coexist Melting is the change of phase from
solid to liquid Heat must be added to cause melting The change in the
re\ ersed direction liquid to solid is freezing or solidifying The change of
phase from the liquid to the gaseous phase is called vaporization, and the
liquid is said to vaporize (or boil) The change from vapor (gaseous phase)
to liquid is condensation, and during the process the vapor is said to be
condensing
Not all substances pass through these three phases, some normally pass
directly from the solid to the gaseous phase a change of phase called sub-
limation Moreover many substances which ordinarily pass through the
three phases may sublimate under certain conditions For example a piece
of ice exposed to the atmosphere at temperatures below 32°F wall sublimate,
and given time, mil pass entirely into the atmosphere as w atcr vapor (steam)
Dry ice, which is solid carbon dioxide, sublimates as you would ordinarily
use it, while it receives heat
In our study of thermodynamics we shall usually be dealing with a
pure substance in fluid form A fluid ts either gas or liquid, or both The
significance of the pure fluid to our study that any particular condition of a
is

fluid at rest is completely defined by tu>o independent properties, provided there


are no effects from motion graxity capillarity electricity or magnetism
(S) This brings us to the matter of how to define the condition of a pure
substance

4 Properties and State In order to compute changes of energy which


have occurred m
a system or working substance we must be able to express
the behavior of the system in terms of descriptive characteristics, which are
called properties Properties which are familiar to the reader from prior
§ Jf\ PROPERTIES AND STATE 5

study include pressure p, temperature T, density p, and specific volume v,


each of which is to be discussed shortly.
The condition (state) of a pure substance in liquid or gaseous form is
defined by two independent properties. If these two properties are stipu-
lated, the state of the substance is set. By this, we mean that all other
properties of the substance, have certain particular values whenever the
substance is in this particular state. There are a number of properties
of interest and utility, but the working list for the heat-power engineer
includes those mentioned above, together with specific internal energy,
specific enthalpy, and specific entropy. (These terms are studied later.)
Now repeating the idea of the first sentence of this paragraph, we may say
that if the pressure and specific volume of a certain pure substance are
stipulated, the temperature will be a certain value, as will the specific values
of internal energy, enthalpy, and entropy. No
matter what happens to the
substance (no chemical change, though), be compressed, heated, expanded,
it

or cooled, if it is returned to the stipulated pressure and volume, the other


properties also return to values identical, respectively, with their original
values. See Fig. 1.
We sometimes speak of two kinds of properties,
intensive and extensive. Intensive properties arc
independent of the mass ( extent ) of (he substance.
Extensive properties are total values, such as the
total volume, and total internal energy. We note
that the volume, for e xample. Js.a.properly-Qf
t otal

the total m ass, of the system and we shall have fre-


quent occasion to deal with extensive properties of
Fig. 1. A substance
this kind. On the other hand, the integsi.vo.prop- whose state is repre-
erty tem perature is indcpende iit jf .thejnass. sented by point 1 has a
To describe a pure fluid system completely, we temperature 2V If the
pressure and volume are
specify the substance (steam, nitrogen, etc.) and varied as defined by the
its mass, in addition to the two independent path 1-A-2-B-1, return-
properties. ing to their original
values, the temperature
Consider for a moment the repeated qualification, also returns to its original
independent properties. As you know, the density value 2V
is the reciprocal of the specific volume; hence,

these properties are not independent of each other. During the boiling of a
liquid, the pressure and temperature of the liquid-vapor mixture are not
independent; the boiling temperature is a certain value for a particular
substance, depending upon the value of the pressure (as we shall learn in
detail later).
From your study mathematics you have learned that two coordinates
of
(the values of x y) locate (or define) a point which is known to be in a
and
given plane (the xy plane). Three coordinates locate a point in space.
So it is with the properties of a pure substance. We look upon these proper-
6 THERMODYNAMIC PROPERTIES [Ch 1

ties as being coordinates locating a point ( defining a state) and we may


picture this point, or any number of state points, on \anous planes, for
example, the pressure-volume plane (Hg 1), the temperature-entropj
plane, etc Any two properties may be used Because, as described,
two properties locate a point in a plane, these properties are called point
functions.

5. Systems of Units. Isaac Newton* made the momentous statement


to the effect that the acceleration of a body is directly proportional to the
force acting on it and inversely proportional to mass, and systems of
its

units hat e been based on it e\ er since Rearranging this statement for a


proportionality constant of unity, we say

Force = mass X acceleration,


(a) F = ma,
in which force is defined in terms of mass, and mass m terms of force We
know that acceleration has the dimensions of length per unit of time squared,
L/r1 Recall that a dimension an attribute of something in general
is

terms, thus, length L is also an attribute of \olume L* Units are dimen-


sional characteristics expressed in terms of defined quantities The length
of a foot isaccurately defined (in terms of the International meter) Other
common English units of length are inches, rods, \ ards, and miles Defined
units of time are seconds, minutes, hours, days etc In English units, a
1
unit quantity of acceleration is one foot per second second (ft per sec

or fps1 )
Now we may say from equation (a), that a unit force is one which pro-
duces unit acceleration in a body of unit mass A consistent” system 1

of units may be built up on this statement If a unit mass is taken as a


pound we call the corresponding unit force a poundal, and we say

(b) F — ma, poundal —* lb„ X "“s'


{CORRESPONDING TO A rOLND suss]

where the sign — » is taken to mean has the units (or dimensions) of The
engineeris accustomed to using the pound as the unit of force, in which

case the unit of mass is a slug.

(C)

* Sir Isaac Newton (1042 1727) is often credited with being the greatest scientist of
all times Born of farmer parents, he was soon exercising his mechanical aptitude,
devising a water clock and a sun dial during his years in grammar school Two sears
after graduating from Cambridge, he had discovered the binomial theorem started
inventing calculus, experimented with color, and speculated on gravity few of his A
achievements the reflecting telescope the composite nature of sunlight a science of
optics, the invention of a thermometer (this is before thermodynamics became a science),

and the most monumental the laws of gravitation He himself credited hi3 scientific
successes to hard work and patient thought
:

§ S] SYSTEMS OF UNITS 7

where vn is the mass in slugs. A consistent system of units may also be


derived on the basis of a pound force.
When an “We used six pounds of steam,” or “The weight
engineer says,
of the body
pounds,” it is most likely that he means a mass of six
is six

pounds. When he says, “The push of the spring on the piston is six
pounds,” he means a force of six pounds. There is ordinarily little confusion
until a problem arises in which the engineer is consciously involved with sys-
tems of units. To rationalize the matter, consider the following statements

1 . One must know the difference between mass and force. A pound mass is an
absolute quantity of matter. By absolute quantity, we mean that one pound of matter
is one pound of matter regardless of where it is located —in outer space at zero
gravity, for example.
2. A pound mass located at a point
of standard gravity g 0 is subjected to a force
of gravity of one pound. To find the mass of a body, we measure the force of
gravity at a point of standard gravity g 0 and compare this force with that on a body
of known mass. Or, we compare the unknown mass with a known mass on balance
scales.
3. In classical mechanics, “weight of” means the force of gravity on, but as previ-
ously mentioned, engineers generally mean an absolute quantity of matter (mass)
when they say weight (but not in stating potential energy —
more on this later.)
There are two units of mass commonly used in consistent English systems, a
4.
pound and a slug (—» lb/-sec. 2 /ft.). We can and do on occasion use ounces, tons,
kilograms, grams, etc., for mass.
There are two units of force in consistent use in the English systems, a pound
5.

and a poundal (—> lb,„-ft./sec. 2 ). The poundal is not used very often. We can
and do on occasion use ounces, kips, tons, kilograms, grams, dynes, etc., for force.
6. The same unit, pound, is used for both force and mass. This should not cause
any more confusion than using the foot-pound or inch-pound to measure both
moment and energy. However, it is essential to know the concepts of moment and
energy and of force and mass.

In the study of mechanics, the pound is the unit of force, and therefore in
order to maintain a consistent system of units, the slug (—> lb/-sec. 2 /ft.) is
the unit of mass. In thermodynamics, by virtue of tradition and the fact
that the pound mass cancels from the energy equations, we use a mixed or
inconsistent system of units.That is, we use the pound for the unit of both
force (and pressure p, lb. per sq. ft., or psf) and mass. No harm is done, as
we shall see. In general, the mathematics of the science of thermodynamics
can be developed independently of the quantity of matter (mass) because it
deals with properties. However, in studying heat transfer, we again must
be consistent in units; the need for consistency will be explained in Chapter
25.
Standards of measure are maintained by governments, so that the magni-
tude of a pound does not change. At a point where the force of gravity
is not the standard value of g0 = 32.174 fps
2
this standard quantity of,

matter, one pound, would not “weigh” one pound on a spring balance,
8 THERMODYNAMIC PROPERTIES [Ck 1
because a spring balance measures the force of gravity Since gravity is
less than standard at high altitudes, a pound of mass ueighs less than one
pound there Consider Newton’s law with respect to a given mass and
note that the force on the mass is proportional to the acceleration let
u>o lb/ be the force of gravity on this mass at a locality of standard gravity

(it will also be themass m


pounds because it is being weighed at standard
gravity), and let w lb/ be the force of gravity on this mass at some other
locality where the acceleration of gravity is g, then

(d) or w w. - 9m ,

where m = u„/gt (—* lb/-sec ’/ft ) the mass m slugs (equation (c)] Also
from (d), we see that the mass in slugs ism = w/g where w is the force of
gravity on the mass m pounds and g is the acceleration of gravity at the
same locality which may be anywhere within the gra\ itational field of the
earth Said another way If the mass m of a body in slugs is multiplied by
the local acceleration of gravity g, the result is the local force of gravity
w lb on the body also
(e) Mass m slugs - standard
„ ^ ,n pounds
acceleration of gra\ ity, g„

because gra\ itational force u» lb and mass in pounds are identical at standard
gravity
Since there is little variation in the force of gravity on the surface of the
earth we may consider g equal to g. for practical!} all engineering purposes,
and for slide-rule calculations wc may use g « g„ « 32 2 fps ! We shall
assume that the reader has mastered the concepts of mass and force, and
therefore we shall not use subscripts (lb/ and Jbm ) to designate whether force
or mass is intended except occasionally in analyzing units When it is
said that there are so many pounds of substance the reader should under
stand that an absolute quantity of matter is intended (unless it is clear from
the context that the force of gravity is meant) From here on, with few
exceptions, let

to = mass in pounds m= mass in slugs

be rewarding to pa} close attention to units throughout this study


It will
We shall rationalize the units of vanous quantities as they are discussed
and return to the subject in the chapter on heat transfer

6 Pressure Pressure, which is the force per unit area, is one of the
most useful thermodynamic properties because it is easily measured directly
(Extremely accurate measurements of anything are difficult ) In terms of
the kinetic theory of matter, the pressure of a fluid is due to the change of
momentum of the molecules when they strike the boundaries of the system
§ 6] PRESSURE 9

(walls of the container). It is a macroscopic observation, as mentioned in


§ 1.
In practice, pressure above and
below atmospheric are determined by
means of a pressure gage (Fig. 2) or a
manometer. The dial of a pressure
gage is marked to read the gage pres-
sure, usually in pounds per square
inch (psi). The gage pressure is the
difference between the pressure inside
of a vessel and the atmospheric pres-
sure outside. Thus, to find the abso-
lute pressure when this pressure is
above atmospheric, add the atmo-
spheric pressure to the gage reading;
that is,

(f) Absolute pressure (psia)


= atmospheric pressure Courtesy Crosby Steam Gage and Valie Co.,
+
gage pressure (psig). Boston

Fig. 2. Movement for a Bourdon


The units of pressure in each of the Pressure Gage. This picture shows the
movement in one type of pressure gage,
foregoing terms must of course be the known as the single-tube gage. The
same. Hence be convenient to
it will fluid enters the tubethrough the threaded
recall the relations of the various connection. As the pressure increases,
the tube with an elliptical section tends
units of pressure in common use by
to straighten, the end that is nearest the
the engineer. Barometers (for meas- linkage moving toward the right. The
uring atmospheric pressure)* and linkage causes the sector to rotate. The
sector engages a small pinion gear.
vacuum gages (for measuring pres- The index band moves with the pinion
sures below atmospheric) usually give gear. The whole mechanism is of course
the pressure in inches of mercury. enclosed in a case, and a graduated dial,
from which the pressure is read, is placed
Manometers may give the pressure in under the index hand.
inches of mercury, in inches of water,
or in terms of the height of column of any fluid that may be used.

* After Evangelista Torricelli (1608-1647) discovered the pressure of the atmosphere,


Otto von Guericke (1602-1686) invented the air pump and set about producing a vacuum.
He finally succeeded after having made two hemispheres, known as the Madgeburg
hemispheres, which were capable of withstanding atmospheric pressure. Before a large
audience of notables, von Guericke placed his hemispheres together and soon had most
of the air pumped from the inside. A horse was hitched to each hemisphere, and try as
they might, the two of them could not pull the hemispheres apart Additional horses
were hitched until there were thirty in all and still the hemispheres could not be pulled
apart. The people, who knew nothing of the pressure of the atmosphere, were astounded
when von Guericke broke the vacuum and the hemispheres fell apart. Had not von
Guericke been a public official, renowned for his wisdom and kindness, his magic might
have resulted in no good for him. Other scientists of the time were persecuted, even
killed, for less.
10 THERMODYNAMIC PROPERTIES [Ch 1

A manometer (Fig 3 any column) measures the pressure by


fluid in the
balancing the force of gravity gm on the part d of the fluid column plus
atmospheric pressure against the pressure p in the system If precision is
needed, the manometer reading taken at a location where gravity is not
standard should be corrected in accordance with equation (d) If the
column is mercury the pressure in inches of mer-
Open to Atmosphere-
cury taken at a point of standard gravity (for a tem
perature of about C0°r) is converted to pounds per
square inch as follows
*

psi \ lb
(g) (Inches Hg) ^0 491
m Ilg/ sq in
Fig 3 Manometer
and temperature may be
Corrections for gravity
made when desirable Since the pressure p in the equations of this book
will usually be in pounds per square foot (psf), a conversion from psi to psf
is a regular requirement

m n)(*«ih>
lb_
ft*

E\en when tho conversion unit cancels from the equation it is advisable
to write it down (and then cancel it), because the habit of conversion is
important
If the absolute pressure is less than atmospheric pressure the gage reading
is sometimes spoken of as the vacuum pressure, or the vacuum In this
instance, the absolute pressure is found from
(l) Absolute pressure (psi a)
- atmospheric pressure - gage pressure (psig)

The gage still measures the between the pressure inside the system
difference
and the pressure of the atmosphere Standard atmospheric pressure is
29 92 in Hg or 14 G96 p«ia (use 14 7 for slide rule work) Remember to
convert gage pressure to absolute pressure
7 Specific Volume and Density The density p of any substance is its

mass per unit volume


mass lb_ slugs
0) Density =
volume it 1
“TIT
If the mass w
measured
is m
pounds and the volume I in cubic feet then
density is p = xo/V lb per cu ft The specific volume v is the volume of a
unit mass, say, cubic feet per pound and is the reciprocal of the density,
v = V/w = 1/p Unless otherwise stated, the density will be in pounds per
cubic feet.

8 Temperature. The temperature of a body is its thermal state considered


with reference to its power of communicating heat to other bodies {4) Our
§ 8} TEMPERATURE 11

sense of feel responds to temperature. Roughly speaking, objects at lower


temperatures than body temperature (98.6°F) feel cool or cold; objects at
higher temperatures feel warm or hot. However, for engineering and
scientific work, the temperature must be designated more precisely with
reference to some scale. A scale of temperature is an arbitrary thing. The
most common reference points are the ice point (the temperature at which
pure water freezes) and the boiling point for water at standard atmospheric
pressure. By way of an example of a scale, if the temperature zero is assigned
to the ice point and 100 to the boiling point, the scale obtained is the centi-
grade or Celsius scale used in scientific work. If the temperatures assigned
to the ice and boiling points are 32 and 212, respectively, we obtain the
the Fahrenheit scale, used in engineering and everyday measurements in
this country. Thus, between the freezing and boiling points, there are
100 degrees on the centigrade scale and 180 degrees on the Fahrenheit scale
(
18
%oo = %), giving the relations
(k) tc = %{t, - 32),
(l) t, = %tc + 32,

where tc and If are the temperatures on the centigrade and Fahrenheit


scales, respectively. Because there is more than one in use, the scale of a
temperature is always stated: as 212°F or 100°C, where F means Fahrenheit
and C centigrade.*
Thermodynamics requires the use of absolute temperature (or thermo-
dynamic temperature), which is measured from a point of absolute zero.
More will be said on this subject later (§ 97). At this time, we shall simply
accept the location of absolute zero on the Fahrenheit scale as —459.69°
(below 0°F). Use 460 in slide-rule work. Absolute temperatures T on the
Fahrenheit scale are called degrees Rankine (°R);

(m) T°R = f° F + 459.69 « t° F + 460.

For example, 60°F is the same temperature as 520°R, approximately.


Absolute temperatures on the centigrade scale are called degrees Kelvin,
written °K, and absolute zero is — 273.16°C. Thus,

(n) T°K - t°C + 273.16.


For example, 40°C is the same temperature as 313°K, approximately.
Notice the symbolization; lower case t for Fahrenheit (or centigrade) and
upper case T for Rankine (or Kelvin). Engineers in English-speaking
countries generally use Fahrenheit and Rankine degrees.

* Galileo invented a thermometer in 1592, but it did not have a well-founded scale.
Gabriel Fahrenheit of Amsterdam, Holland, was the first (in 1720) to devise an instru-
ment that indicated temperature in degrees, choosing the ice and boiling points of water
as 32° and 212°, respectively. The centigrade scale was introduced in 1742 by Anders
Celsius (1701-1744), a Swedish astronomer and professor at Uppsala.
22 THERMODYNAMIC PROPERTIES [Ck 1

Temperatures are measured in a number of ways, always through a


change m
some other characteristic of the substance to be measured

1, Change tn volume Nearly everyone is familiar in a qualitative way, at least


with the phenomenon of substances (say, mercury or a gas) that expand with
increase in temperature If the amount of expansion in a particular case ts corre-
lated with the freezing and boiling points of water, and the change in volume divided
into 100 or 180 equal parts, the instrument could be used to "read" temperatures

Liquids used include (5) mercury (— 38°F to about G00°F or 900°F with nitrogen
above the mercury) alcohol (— I00°F to about 300°F), and pentane (— 300°F to
about 70°F) Glass begins to soften at about 900°F.
2 Change tn pressure II a gas is confined to a constant volume its pressure
will go up as its temperature increases, and the change m
pressure can be correlated
with temperature change See Fig 2
3 Change tn electrical resistance The electrical resistance of metals increases
almost m direct proportion to temperature increase Thus the measured change
in resistance of a particular piece of wire can be converted to temperature change
Metals used include nickel, copper (0°F to 250°F), and platinum (at high tempera-
ture and in high precision instruments)
— Change in electrical potential The device that measures temperature by the
electromotive force (emf) is called a thermocouple * It operates by virtue of the
phenomenon occurring when two wires of different materials are joined together
at their ends, with different temperatures existing at the two junctions Under
such a condition, an electromotive force exists which is very nearly directly propor
tional to the temperature difference between the two junctions The potentiometer
which measures the electromotive force can have a scale which reads temperature
directly One junction of the thermocouple is kept at a reference temperature
which may be the room temperature This is one of the favorite means of measuring
temperature Combinations of metals used include (5) copper and constantan
(—300 to G50°F) iron and constantan (—300 to 1500°F), chromel and alumel
(-300 to 2200°F)
5 Optical changes The instrument for this measurement is called an optical
pyrometer It is sighted at the hot body whose brightness is compared with the
brightness of an adjustable and calibrated source of light within the instrument
When the brightness of the instrument light has been adjusted to be the same as the
body whose temperature is desired the instrument “reads” the body temperature
There are several types of optical pyrometers

Obtaining precise measurements of any property is not easy, and even in


wdvwaxy votk where great precision is unnecessary, readings
should be made with care Also, one must know of the characteristics and
eccentricities of the instrument being used in order to avoid taking false
readings In short, if you are concerned with the actual taking of data,
you should do further study, for instance, references (5), (6), (7), (5), (3),
y. Thomas J Seebeck (1770-1831), who discovered the thermocouple (1821), was born
in Estonia Although he was educated as a Doctor of Medicine (Gottingen College,
Germany), he chose to lecture and experiment in the physical sciences For his invention
of the thermocouple, he was given a citation by the Pans Academy of Science^
§ 11] CLOSURE IS

and others, including handbooks. We shall deal further with temperature


after having built a background for better understanding.

'UT' Process. If any property (or all properties) of a substance (§ 4) changes,


the substance is said to have undergone a process; there has been a change of
state. Actual processes involve changes in all or nearly all properties, but
we study thermodynamics via ideal counterparts in which often one of the
properties remains constant. For example, if a change
p
of state occurs during wliich the pressure does not
change (1-2, Fig. 4), the working substance is said to
undergo a constant pressure process (isobaric) ; if the
volume of a particular mass remains constant (ah,
Fig. 4), but other properties change, the process is
called a constant volume process (isometric). Proc-
esses will be studied later in some detail in several
chapters. F&4 - Processes.

Cycles. When a fluid in a particular state passes through a series of


processes and returns undergoes a cycle. The processes
to its initial state, it
may be ones which have names, as constant volume
and constant pressure, or they may be series of
state change without names, as in Fig. 1. The
path of a process may be any random path
1-A-2-B-1 ;
only necessary that the substance
it is

return to state 1 in order to have completed a

cycle. Cycles are studied in detail after a study of


processes.

Fig. 1. Repealed. 11. Closure. Much of what is said in this


chapter is a review of topics viiich have been studied
in previous courses. However, do not let this deceive you. It will likely
be necessary in this course to know this material better than you have ever
known it before. Inasmuch as most failures in school subjects are due to
a lack of understanding of the English language, and inasmuch as an
understanding of succeeding chapters depends on knowing the meanings of
the technical words in this chapter, review this chapter and each chapter
hereafter with the aim of being sure that you understand the new words,
most of which will be in bold -face italics for easy identification.
It is necessary for the beginner in thermodynamics to study regularly and
earnestly each assignment. In many subjects, it may seem feasible, if not
desirable, for you to study intensively at irregular intervals in order to
perform acceptably where your work is graded. However, on average, such
At the outset of the
tactics are sure to result unfortunately in this subject.
course,you are confronted with conceptions, most of them new conceptions,
which you will use off and on throughout the remainder of the work. More-
U THERMODYNAMIC PROPERTIES [Ch 1

ox er, inmanx instances, these conceptions build up, one on the other, so
that without a clear understanding of the earlier x\ orL it is not onlj iraprac
ticable but also impossible to achieve a clear understanding of the later
work Stud} each assignment diligentlj and learn as much as jou can
before class explanation is gn en
a Then use the classroom period to but-
tress jour original efforts and to clear up the hazj parts You must not
just understand the words of jour teacher, but you must make the ideas he
explains a part of your tporting knowledge You will find profitable /reguen l
quid renews made on jour own imtiatne
While the subject of thermodjTiamics is not an easj one neither is it
difficult if taken and mastered bit bj bit Moreen er, students who want to
become engineers find the course attractix e and manj of them \ oluntanlj
remark that thermodj-namics is exceptional!} interesting
2

CONSERVATION OF ENERGY

Energy ( E) is inherent in all matter. Indeed, there


12. Introduction.
is evidence that suggests that matter and energy are fundamentally
scientific
the same thing, or at least that mass may be converted into energy and
energy into mass. According to Einstein, whose hypothesis (1905) was one
step in the development of atomic power, the relation between mass and
energy is (energy) = (mass) (velocity of light) 2 where a consistent system of
,

units is used. Since the velocity of light squared is a very large number,
the energy of a particular mass is very large; for example, one pound of
matter is equivalent to nearly 39 X —
10 12 Btu but this is not to say that it
will ever be possible to convert a particular mass of any substance entirely
into energy. (See Chapter 7.)
By energy, we mean a certain something which appears in many different
forms, forms which are related to each other insofar as conversion can be
made from one form of energy to another. This is admittedly not a defini-
tion. We are unable to define the general term energy (E) in a simple way,
but we can define with precision the various forms in which it appears.
This chapter considers the forms of energy with which we shall be concerned
in this study and it names some of the other forms.

13. Measuring Energy. The total amount of energy that a system con-
tains cannot be determined. We are accustomed therefore to measuring
energy above some arbitrary datum. This practice is quite satisfactory,
since in engineering we need know only the change in energy. Thus, we
consider the potential energy of one pound of water to be the change in
potential energy that this pound of water would undergo in falling from the
reservoir to the power plant. The level of the power plant is the datum
level. In like manner, other forms of energy are measured above some
so-called datum stale.

15
16 CONSERVATION OF ENERGY [Ch 2

Energy' is a scalar quantity, not a vector quantity Velocity,


a \ector
quantity has direction as well as magnitude Energy has only magnitude
The energy of a system of bodies is simply the sum of the energies of the
individual bodies The total energy of a single system or substance is the
sum of the magnitudes of the various forms of energy (such as mechanical
kinetic energy, molecular energy, chemical energy) which the system or
substance possesses
You arc familiar with many units of energy from your study of physics
defined before the law of eonsen ation of energy was accepted Expon
menters with heat defined the British thermal unit (Btu) as the amount of
heat to raise the temperature of one pound of water 1°F This Btu is
slightly different at different temperatures, hence the mean Btu Yso of
the heat required to raise the temperature of one pound of pure water from
32° to 212T under standard atmospheric pressure, 14 GOG psia has been
widely used Of course, experimenters with work had the foot pound as a
unit of energy, and then the early workers in electricity came up with their
units of energy When it was realized that heat and work and electricity
were manifestations of the same concept, energy a tremendous amount of
experimentation was done to relate the foot-pound to the Btu Since
experiment cannot by its nature give a single unique answ er (experiment can
say that the odds are so many that a value is between certain limits) we
finally decided simply to define the relations between the units just as an
American yard is defined m
terms of an international meter (metric system)
maintained at Sevres Trance As currently defined then, 1 Btu is equiva
lent to 778 172 ft-lb Because of Joule s* work in establishing the relation
between heat and work, we call the conversion constant Joule's constant
and designate it by J,
J = 778 172 ft lb /Btu;

use 778 in slide rule w ork See the Appendix for other conversion constants

14 Mechanical Potential Energy The mechanical potential energy or


gravitational energy of a body energy due to its position or elevation
is its

with respect to the earth A weight at


some distance z above the floor wall
give up a certain amount of energy upon impact if it is allowed to fall freely
to the floor The change of mechanical potential energy of the body at the
instant of impact is equal to the product of the force of gravity on the body
[mg see equation (d), p 8] and the vertical distance z through which the
* James Prescott Joule (1818-1889) an English scientist was educated pnvatelj by
a tutor being at one time a student assistant to Dalton (§ 198) His researches were
significant in the budding science of electricity as well as in the science of thermodynamics
in which he established two important fundamental principles the equivalence of heat
and work and the dependence of the internal energy change of a perfect gas upon tem
perature change ($ 57) As a result of this work the modern kinetic theory of heat
superseded the caloric theory of beat Joule once remarked I believe I have done two
or three little things but nothing to make a fuss about
§ 15} MECHANICAL KINETIC ENERGY 17

force acts. For a body with finite dimensions, z is the distance moved by
the body’s center of gravity. Thus, the mechanical potential energy P,
due to elevation and the change of potential energy A P = P 2 — Pi, are

(IA) P = mgz =
^ gz ft-lb.,

(IB) P = — z ft-lb. /lb.


(fo

(IC) AP = mg{zi — z{] ft-lb., or — (z 2 — zj) ft-lb./lb.,



where m is the mass in slugs

[— > lb/-scc.-/ft.; mgz —> (lb/-sec. 2 /ft.)(ft./scc. 2 )(ft.) = ft-lb/]

and g is the local acceleration of gravity; w0 may be considered as either the


mass in pounds or the force of gravity at a location of standard gravity g0 .

In actual engineering systems, we generally deal with a rate of flow; for


instance, w lb. per sec., V cu. ft. per min. (cfm), in which case, the change of
potential energy is in foot-pound per second or per some other unit of time.
The context makes clear the time unit involved. Inasmuch as the accelera-
tion of gravity varies little on the surface of the earth, it is generally accurate
enough in engineering problems to consider g « g„, and get

(a) P « wz and AP = Pi —P « x w(z 2 — zj ) ft-lb.,


~ Zl)
or AP ~ Btu,

where w is the mass in pounds of the body whose potential energy P or


change AP is desired, and where z ft. is the elevation of the body’s center of
gravity above the desired datum level. Equation (a) is generally accurate
enough, but note particularly the true value in equation (1). We say that
a body has mgz ft-lb. of energy stored in it when we mean that it has mgz
ft-lb. of potential energy with respect to the datum z ft. away.

A change of elevation is a change of state. For a fluid to have a change


in the property z, it must move. Thus, to define a change of state of a body
which changes elevation, specify not only the changes in two indepen-
dent thermodynamic properties, such as p and v, but also the change in z.
Observe that mechanical potential or gravitational energy exists by virtue
of the force of attraction between two bodies, the earth and the body which
is said to have potential energy.

Mechanical Kinetic Energy. A body of mass m (any unit, but only


16.
whose parts are moving with a speed of v fps
slugs result in ft-lb.) all of
possesses kinetic energy K, as it is derived in mechanics, (48), of

< 2A > Tr = mv-


K ~r r “Tt —X
lb/-sec. 2 ft.

sec^J
2
\ , ..
ff - lb
"
18 CONSERVATION OF ENERGY [Ck 2

w ith the earth as the reference body Since m slugs is the mass w m pounds
divided by the standard acceleration of gravitj g0 [equation (e)
p 8]
hav c the energj stored m a body in the form of mechanical kinetic energy as

-(2B) or ft lb /lb or
g Btu

and the change of kinetic energy is

(2C) AA.-Aj — Ai-~ (ui‘ - Ui l


) ft lb or (ut’ - m*) Btu
iv here u is the speed in feet per second If you care to remember it

2 gj « 50 000

This equation (2) is applicable to moving fluids liquids or gases as well as


to rigid bodies Notice that u / (2g0) is in foot pounds per pound of mass
where the force pound in the foot pound should not cancel the mass pound
because the units would then lose their significance To define a change of
state of a fluid which changes speed sjiecify not only the changes in two
independent thermodynamic properties such as p and v but also the
initial and final speeds In thermodj namic practice w is likely to be a
rate of flow in which case the time unit is made clear by the context

16 Internal Energy Matter is composed of an aggregation of molecules


which are moving continuously but haphazardly Since the molecules
have mass we conceive of them as having a kinetic energy called internal
kinetic energy analogous to that of anj more tangible body m motion
Tho total internal kinetic energj arises principal lj from (1) the transla
tional motion of the molecules (2) the rot ational motion of the mole cules
and (3) a vibratory motion of the atoms within the molecules^ Some
molecules consist of one atom (for example helium and argon) The
molecular kinetic energy of such substances called monatomic substances
is almost wholly due to tho translational motion of the molecules first

because the vibration of atoms within the molecule relative to one another is
impossible since there is only one atom and second because the kinetic
energy of rotation is negligible since the mass of the molecule (atom) is
concentrated so close to the axis of spin (that is the moment of inertia of
the molecule with respect to the axis of rotation is negligible)
In a diatomic molecule which is one made up of tw o atoms (for example
oxygen Oj and nitrogen N*) the atoms may move to and fro relative to
each other in a vibration while the molecule is undergoing a translation and
a rotation Moreover the atoms are removed somewhat from an axis of
spin so that the rotational energy may be appreciable These conditions
are true of polyatomic molecules in general Thus especially at high tem
peratures the internal kinetic energy of molecules of actual gases with more
§17] WORK 19

than one atom includes a significant amount of rotational and vibratory


energy. (See Fig. 19.)
In addition to the internal kinetic energy, substances have an internal
'potential energy; that is, an internal energy due to the relative positions of

its molecules and the attraction of these molecules for one another. We
know that a change in the mechanical potential energy of a body occurs
when the elevation of the body relative to the earth as a datum is changed;
the force of gravity is acting to aid or hinder this motion. The change of
potential energy results from a force of attraction between at least two bodies
which change position with respect to one another. Since there is an attrac-
tive force between molecules, we can see, by analogy with the concept of
mechanical potential energy, that if anything happens to increase or
decrease the average distance between molecules, there will be a change in
the internal potential energy, which is that portion of the internal energy
due to the configuration of the molecules. As an illustration of such a
change, imagine one pound of water being boiled. As water, it occupies
about 0.016 cu. ft. If the pound of water is entirely evaporated at atmos-
pheric pressure and the steam collected, it will be found that the one pound
of steam occupies about 27 cu. ft., a volume about 1700 times that occupied
by the water. Since the number of molecules in the pound of water is
virtually the same as the number
in the pound of steam, it follows that the
molecules must be much more widely separated in the steam than in the
water. To move these molecules apart against their attractive forces
requires a large amount of energy, which is retained in the steam as part of
of the steam’s store of energy. When and if the steam is condensed, this
same amount of internal potential energy will be given up.
The sum of the internal kinetic energy and the internal potential energy is
*
called the internal energy which is energy stored within a body or substance
by virtue of the activity and configuration of its molecules and of the -vibra-
tion of the atoms within the molecules. We do not know how to find the
absolute quantity of internal energy in any substance; however, what must
be known in engineering is not the absolute quantity but the change of
internal energy. Above any convenient datum,
u Btu per lb. represents the specific internal energy (1 lb.),
U = wu Btu represents the internal energy of w lb.,
A u = «2 — «i and A (7 = U 2 — Ui, the change of internal energy.
In thermodynamic practice, w is often a rate of flow, in which case the time
unit is made clear by the context.

17 Work. For work IF to be done, there must be a force acting on a


.

body that moves. The work of a force, F, may be defined as the displace-
*
writers use the term internal energy to mean the total energy stored within a
Some
*
In this hook, it niij nlu/n s mean the molecular energy as defined.
63 's fern (3) This
usage has become well entrenched.
20 CONSERVATION OF ENERGY [Ch 2

merit of a body (considered as a pirtich) multiplied bj the component Fx


dx
of the force in the direction of the displacement dll =
Fx dx ft lb This
definition provides a basic unit of energy the fool pound when the force is
measured inpounds and the displacement in feet
It is important to observe that work is energj in transition that is it
exists only when a force is ov ercommg a resistance (which maj be inertia
only in an ideal sjstem) and it is being done only when a force is moving
through a distance When the point of application of the force ceases to
nun e there is no work Contrast the concept of work with that of internal
energj The internal energj is energj which the body contains On the
other hand a bodj never contains work The body or system may have
the capacitj to do work or work may be done on the sj stem For example
work flows into a sjstem but after it is in it is not work it maj have
been converted into internal energj (We study what happens m detail
later) Similarlj as derived in mechanics work done on a bodj maj be
corn erted into kinetic energj and vice v ersa
Work is a form of energj which can be conv erted 100% into other forms in
the absence of friction of anj kind and for that reason it is the most valuable
form For example it can all be conv erted into electricity or kinetic energj
and electricity and kinetic energj can be fullj converted into work in ideal
fnctionless sj stems In fact we could with logic take electncitj os a form of
work (10) since it too is fullj available energj in transition but not so
with kinetic energy which is comparable m kind with internal energj m
that it is stored in a bodj or sj stem W ork can be conv erted 100% into heat
(§ 19) friction or no friction friction hastens the com ersion of work into
heat
Since work can be measured at various points m a machine adjectives
must be appl ed m order to saj what work is intended Manj thermo-
djnamic computations give the ideal work U of the fluid that is the work
which the fluid would do (or have done on it) if there were no friction or
other losses during the process Ideal w ork is done by or on the fluid
In a machine which manufactures work the actual work which the
fluid does is alwajs less than the ideal and the actual work of the fluid is
greater than the work deliv ered by the shaft of the engine In a reciprocat-
ing engine such as a Diesel engine the actual work of the fluid in the cj lin
der [system (1) Fig 5] is called the indicated work 11 1 when it is taken for a

complete cycle of ev ents m the cylinder because it may be determined with


an indicator (§ 119) Note that energy 11 1 is crossing the boundary of
sjstem (1) The actual work which passes a boundarj at C is less than IFr
because of friction at the piston rings and cj Under walls at the packing
glands at B and at the p n near C In the engine sjstem (2) the work Wb
which crosses the boundary (that is delivered bj the shaft) is smaller than
TT i because of friction at the crank pm and at the crankshaft bearings (and
other miscellaneous frictional losses for example the fanning of the air
18] WORK OF A NONFLOW SYSTEM 21

by the flywheel and This work delivered by the shaft is commonly


crank).
called brake work Wn
because it was first measured by a brake which
absorbed all the output of the engine via friction. Other names for brake
work include shaft work and delivered work.

Fig. 5. Work in a Reciprocating Engine. Observe that shaft work (or any kind of
work) is energy in transition in the sense that it exists by virtue of the movement of an
element of a machine against a resistance or, in general, of the movement of a force
through a distance. Shaft work may be transformed into kinetic energy and stored
in the attached rotating parts to a limited extent System (2) is outlined by the large
dotted rectangle.

Since work may be done either by or on the system, it.is convenient to use
plus and minus signs to designate its direction of “flow.” Let
Wo rk done hi/ the system be nosi ivp. (as a steam turbine); an outflow of energy;
t,

Work done on the system be negative (as an air compressor) an inflow of energy.
;

The symbol W
stands for work done by
or on one pound or any number of pounds
of substance; the mass involved is defined
by the context. Also, there may be a time
unit involved. Typical units for IF, which
will be clear from the context, include ft—lb.,
ft-lb. per min., Btu, Btu per sec., hp,

hp-hr., etc.

18. Work a Nonfiow System. Let


of
the system be a quantity of an expansible
fluid, such as a gas or vapor, enclosed within
a cylinder and piston (Fig. 6). This is a
closed system in which nonflow processes
may occur. The volume of the fluid is Fi
and its pressure is pi. If we consider the
state of the fluid on the pV plane (mean- Fig. 6. Closed System— Work
ing that the coordinates are of Expansion. The boundaries of
p and F), the
particular coordinates the system expand and contract
t and Fi locate the
p with the movement of the piston.
point 1 (Fig, G). If the working substance
expands and moves the piston against a variable resistance, work will
be done by the fluid. In a typical practical process of this sort, the
pressure drops and the state of the substance changes as suggested by
28 CONSERVATION OF ENERGY [Ch 2

the curve which curve is called the path of the state point
1-C-/-2, This
curve represents a process of some land Since the pressure and therefore
the force on the piston is varying it is necessary to integrate fFx dx in
order to find the work Consider a change of state from e to / (Fig 6)
so small that the pressure is essentially constant during the change The
force acting on the piston wall be the pressure times the area of the piston,
r* = pA The distance that the piston moves is a differential quantity
dL, and the work for this infinitesimal motion of the piston is force times
distance,
dW - CpA)dL = p(A dL) « p dV,
where A dL is a differential \olumc equal to dV The total work done in
the reversible nonflow process is

(3) II - ['pi I tt-lb,


tneVEHSIRLB, NONM.OW rROCESS]

where the units must be p lb per sq ft (psf) and V cu ft to give work m


foot pounds We shall go into the matter of a reversible process later, so
let it suffice here to saj that there is no
friction in a reversible process, mechanical
1
or fluid, everything is on an ideal basis and
Piston
the work of the fluid fp dV appears as
shaft work
j
lArra^lj
Piston Rod -* Notice the differential area befc (Fig G)
whose width is dl and height p The
^-System ,

magnitude of this area is p dV and the sum


of all these differential areas is the total
area under the curve 1 e-/-2 when the
integration is made from state 1 to state 2
Thus we have the area equal to fp dV and
1 II — fpdV therefore,the area "under
process on the pV plane

m
fez be n V
the curve of the
represents the
reversible process
work done during a nonflow

Fig 6 Repeated The work done and the area under the
curve on the pV plane depend upon the
location of the path l-e-f-2 (Tig G), that is upon the re/atiort between

p and V = /(F)] Although states 1 and 2 are particular points and


[p
Pi P2and Vi have particular values the change of state may occur
, Vi,
along any other path l-x~2 Thus the work done has changed because the
path of the state point has changed the work depends upon the process —
It is not a property, as pressure and volume and internal energy Since
such properties are dependent only on the location of the state point, they
are called point functions All the properties listed in § 4 are point func-
;

§ 19] HEAT 23

tions,depending only on the state of the substance. It is correct to say that


the change of volume AF = F 2 —
Fi, but the work done cannot be said to
be TF 2 — TFi; TF is not a point function.
When
the function p = /(TO is known or assumed, the integration can be
made. For example, Boyle’s law is that pV = C — piVi — p 2 F 2 etc., ,

when the temperature of an ideal gas is constant; C is a constant. Substi-


tuting p = C/F, into equation (3), we get the work during a frictionless
process at constant temperature as

(b) TF = J~pdV = C J ^=C In


||
= Pl V , In Ij.

[GOOD FOR REVERSIBLE PROCESS OVI.V WHEN BOVI.e’s L.VU HOLDS]

The units in equation (b) are foot-pounds or foot-pounds per some unit of
time, the time unit being associated with F, as efm. The convention of
signs defined for IF in § 17 was chosen because it accords with the results of
the integral fp dV that is,

fp dY is positive when nonflow work is done by the system,


jp dV is negative when nonflow work is done on the system.

This integration maj' be made for any kind of expansible substance, given
the pV relation.

19. Heat. We shall use the word heat Q to mean energy in transition
moving ) from one body or system to another because of a temperature difference
(:

.*
between the bodies or systems This transfer of energy occurg-by conduction,
by radiation, and by convection. We have confidence on the basis of per-
sonal experience that heat flows from a hot body to a cold body. Also, it is
easy to accept the notion that if two bodies are at the same temperature, heat
does not “flow.” When two bodies are in thermal communication (that is,
ineffective insulation between them) and in states such that no heat goes from
one body to the other, they are said to be in thermal equilibrium. It logically
follows that if two bodies are in thermal equilibrium with a third body the two ,

are in equilibrium with each other; this statement is known as the Zeroth law
of thermodynam ics ( 1 ). Thus, when two bodies are in thermal equilibrium
with the same environment they are in equilibrium with each other and
there will be no net interchange of heat between them.
Conduction of heat occurs because the faster moving molecules in the
hotter part of a body communicate (by impacts) some of their energy to
adjacent molecules. Since temperature is a measure of the molecular
activity, we are reminded that heat flows from the hotter to the cooler por-
tion of a body, that the more active molecules lose some of their energy to

* In layman’s parlance and in some physics books, heat is used to mean molecular

energy, that which we here call internal energy. If you are accustomed to this notion of
heat, you must make an effort to replace it with the foregoing definition.
H CONSERVATION OF ENERGY [Ch 2

the less active molecules Heat is conducted through solids, liquids, and
gases
Radiant heal consists of electromatic waves waves of the same nature as
light and radio waves We may consider that the radiant energy which
enters or leaves a system is included m
the symbol Q, but it is better to
give other forms of radiated energy a distinguishing symbol, reserving
Q for
heat Typical wave lengths of radiant energy in meters are as given in
parentheses {12) cosmic rays (10 14 and shorter) gamma rays (10 ‘MO u
),
X-rays (10 “-10“*), ultraviolet (10 *-10’), usiblclight (4 X 10~ 7-7X 10“’),
-4
infrared or heat (10~*-10 ), microwaves, radar (10 s-10 ‘), television and
TM radio (1-10), short-wave radio (10-10 AM radio (10M0*), maritime
2
),
communications (10M04)
All bodies radiate heal (§ 416), so a transfer of heat by radiation occurs
because a hot body emits more heat than it receives or a cold body receiv es
more heat than it emits
Convection occurs because a moving fluid picks up energy from a warm
body and delivers energy to a colder body Tor example the air surround-
ing the furnace of a hot air heating system receives heat by radiation and
conduction This heated air being lighter, will rise and circulate through
the house, giving up energy by radiation and conduction to keep the house
and con ton ts warm When this series of events happens, we say that heat
has been convected although the energy is not classified as heat while it is
being transported but only while it is being received or discharged In the
systems which we shall study, the convectcd energy will generally not be
considered as £pat This energy quantity will be cared for by energies
associated with the stream
Observe that heat, like work, is energy in transit, it is a concept of that
something which moves across a boundary into or out of a sj stem by virtue
of a driving potential which we call temperature Heat Q which flows into a
system may result in an increase of internal energy of the system or in work
being done by the system but after it enters the system it is no longer^ heat
Work is 100% available for conversion into other forms of energy, whereas
heat cannot continuously be converted 100% into work even in the most
perfect engine imaginable this is a significant distinction betw een these two
,

kinds of energy in transition


It will be made clearer soon (§§ 35-37) that heat, like work, is a path
function By its nature, Qis a change, we never say Qi — Q\ because such
a difference holds only for point functions (state properties) The umts of

Q are any convenient energy umts in this book usually Btu or Btu per
unit of time, either for one pound or any number of pounds

20 Flow Work. If a substance is flowing into or out of a system being


studied, a certain energy quantity is mvolvcd, as the substance crosses the
boundaries of the system, called the flow work or flow energy Imagine a
§ 21 ] OTHER FORMS OF ENERGY 25

fluid, liquid or gas, flowing in a pipe (Fig. 7).


Let some small quantity
of thissubstance be on the point of crossing the boundary and entering the
system. In order for it to get into the system, work must be done on it in
an amount sufficient to move
against the resistance (at pressure p) offered
it

by the system. The F is p.-l, and the work done against


resisting force
this resistance in pushing a quantity of fluid of length L across the boundary
is FL = pAL = pV, where V = AL is the volume of fluid pushed across the
boundary. An energy quantity equal to pV thus crosses the boundary and
enters the system. Similarly, there is an
energy quantity pV leaving the system
while the substance passes outward across
the boundary. Tliis idea is generalized
for a movement of any volume V across
any boundary against constant resistance _. _ ,

when we may say that the energy’ of flow


crossing such boundary is pV, where p and V arc values at the boundary.
This energy a point function, IF/ = pV, because it depends on par-
is

ticular values of p and I' at some boundary; therefore for boundaries 1


and 2, we may say

(c) AH'/ = W f,
~ IF/, = P=F 2 - P,F, ft-lb.,
All'/ = lff/2 — ll'/i = p v 2 — p ,Vi ft-lb./lb.,
2

where p is the absolute pressure in pounds per square fool, V — wv cu. ft. is
the volume of the total mass of substance crossing a boundary (usually
during a particular time, as cfm), and v is the specific volume (for unit mass
per unit time).
Since the energy’ to cause flow in a pipe does not create itself, we may

think of the flow work IF/ as originating, for example, in a pump which is
located upstream, or in the potential energy’ of a reservoir. However, it
isgenerally not pump work (but it may be) it exists because energy’ is being
;

expended somewhere, somehow, to cause a movement of a fluid across a


boundary.

21. Other Forms of Energy. Manifestations of energy occur in many


forms with which we shall have little or no concern in this book. However,
the laws of thermodynamics apply to all forms of energy and are often useful
in specialized fields of study. Recapitulating first, we have defined the
energy forms:

P, potential energy’; stored; change is P2 — Pi;


K, kinetic change is /C 2 — Ki\
energy’; stored;
U, u, internal eneigy; stored; change is U 2 — U 1 or u 2 — Ui;
Wf, flow work; in transition; change (between boundaries) is TF/ 2 — n; W
TF, work; in transition; change is path function, Jp dV in ideal nonflow;

Q, heat; in transition; change is a


path function.
26 CONSERVATION OF ENERGY [Ck 2

Then there is electricity E, (which we combined work IF* on


shall call
occasion), chemical energy Et (resulting from a change m
molecular struc-

ture as in combustion of fuels), electromagnetic emanations (light, radio,
brain naves ( ?), etc), acoustic energy (sound waves), electrochemical

energy a source of electricity (storage battery), nuclear energy (resulting
from a change in the structure of the atom’s nucleus and a conversion of
mass into energy), energj stored due to surface tension, and others In
subsequent discussions, all energj forms mentioned in this paragraph are
assumed to be absent unless specifically included, for example, when it is
said that an equation applies to all reversible processes, it is understood, m
the absence of a contrarj statement, that the processes are concerned only
With the energj quantities listed through Q at the bottom of p 25

22 Classification of Systems Sj stems may be classified in different


wajs The definitions in this article will serve our purpose As we have
learned, a closed system is one t n
tchtch mass docs not cross tIs

boundaries energy may (and does


in the sjstems we studj here)

cross the boundaries in the form
of work and/or heat Within a
closed sjstem the working sub-
stance maj or may not hav e mo-
tion If, for example, all of the

working substance, steam and


water in a power plant is taken
as a sjstem, the sjstem is a
Courtesy Power Aw J ork City
closed one See Fig 8, but note
Fig 8 Steam Power Plant The source of that the sj-stem referred to here
the heat added to the steam is in the entering
fuel and air stream at A in the form of chemical
is inside the various pipe lines
energy Operation Water is evaporated into and containers However, the
steam in the boder, steam flows to the prune working substance is continu
mover (engine) and some of its energy is con-
verted into work IV,,,, leaving the system, ouslj moving, and at any instant
(

exhaust steam enters the condenser and is it will have stored in it some
condensed to water by circulating cooling water, total kinetic energj and some
in and out at C and D, where heat is rejected
from the cycle, the two pumps pump the water total potential energj as well as
back into the boiler and it repeats the Cycle internal energy Each of these
The energy required by the pumps is (JV«, energj quantities as thej applj
and Wm) The net work of the cycle is
W= - W* to a particular small mass of sub
stance varies from point to point
as the mass circulates through the cycle The total energj within such a
system is its stored energy E, the sum of all potential, kinetic, and interna!
energies
A closed system of a working substance which is not flowing, as for
§ 22] CLASSIFICATION OF SYSTEMS 27

instance the system in Fig. 6, p. 21, is a nonflow system, and the processes
which such a S 3'stem undergoes are called nonflow processes. Excluding
turbulence, which may exist in a nonflowing system, the total stored energy
is the total internal energy (P = 0).

An open system is one in which mass crosses its boundaries; transit energy,
work and heat, may or ma.y not be crossing the boundaries at the same time.
There may be several mass streams, as in a steam power plant where the
mass streams are (1) the entering fuel, (2) the entering air to burn the fuel,
both at A Fig. 8, (3) the products of combustion leaving at B, (4) the circu-
,

lating cooling water for condensing the exhaust steam from the turbine,
entering at C, leaving at D. There are a number of common examples of
systems involving one substance; for example, a steam turbine (Fig. 9)
where the steam enters the system at 1 through a valve and leaves at 2.
In the turbine, the steam speeds up in going through nozzles and gains
kinetic energy which is converted into work as the steam passes through
the turbine blades.
The type of open system which we shall study most frequently is a steady
flow system, which has the following characteristics: (1) the rate of flow
(pounds per second, for example) of each mass stream entering and leaving
the system is constant; this means that there is neither accumulation nor
diminution of mass within the system; (2) there is neither accumulation nor
diminution of energy within the sj'stem (the rate of flow of heat Q and work
IF are constant) for example, the turbine of Fig. 9 delivers work and radiates
;

Fig. 9. Turbine as a System.

heat at a constant rate while operating under steady flow conditions; (3)
the state of the working substance at any point in the system remains con-
stant. For most practical purposes, a reciprocating engine may be analyzed
as a steady flow system when the state of the working substance varies
uniformly and periodically through the same series of states; for example, an
28 CONSERVATION OF ENERGY [CA 2
internal combustion engine maj be considered as a stead} state s}stem
when it burns the same amount of fuel during each c}de, when the tem-
perature of each batch of exhaust is always the same, when the heat earned
nvvaj b> the cooling water is the same during each c} cle of events, etc
Unsteadj flow s} stems arc of several forms, each one characterized b>
accumulation or diminution of either mass or energj, or both, within the
s} stem Tor example (1) fluid maj flow outward onl} from a s}stem (as a
system of air in a tank from which air flows to “pump up” a tire), (2) fluid
maj flow inward onlj (as in refilling the air tank with the discharge valve
closed), (3) fluid maj be flowing both inward and outward but at different
rates so that the amount of fluid within the s}stcm increases and decreases,
or (4) the engine is w arming up (increasing pow er) or cooling off (decreasing
power) By waj of illustration of all these ideas consider an automotive
engine W
hen it is started cold much of the energj from the fuel is used to
heat the metal parts and the cooling water W hilc the car is getting under
waj varying quantities of fuel and air enter the engine
,
If the car speed is
maintained constant on level ground and if the environment (temperature
and wind speed, principal!}) remains constant the engine eventual]}

reaches stead} flow operation even though the temperature at a particular
viemit} within the sjstem (inside the cjhnder for instance) fluctuates, each
ejele of fluctuation is the same as the preceding ones If the car now goes
up hill at the same speedmore work must be done b} the engine more
working substance must and the engine “gets hotter
flow into the engine
If the car goes down hill at the same speed less work is done bj the engine
less fuel is used, and the engine runs "cooler” and not in stead} state
While stead} flow and stead} state are cvidentl} difficult to achieve in a
moving automobile, tests and analjses of automotive engines are generatl}
made under virtual stead} flow conditions notwithstanding It is easj to
approach stead} flow in a stationarj internal combustion engine on an over-
all basis

23 Continuity of Mass Implied in the foregoing discussion is the law


of the continuity of mass, also called con
servation of mass, which says that mass
ts indestructible Within the bounds of
our experience, this law bolds except m
atomic processes w here mass is corn erted
into energj Also, there is speculation
that mass maj be created m the “vacant'

space of the umverse one atom of hjdrogen per jear in some several mil
*
lion cubic feet of space
In Fig 7, let the av erage v elocity of the fluid at section C m
the pipe line

*See Fred Ilojle, The Mature of the Unit erse for an interesting account written for
the layman
§ 25) ENERGY EQUATIONS FOR CLOSED SYSTEMS 29
be v fps; then the quantity of fluid flowing is Av cu. ft. per sec., where A
sq. ft. is the cross-sectional area. If the specific volume is v cu. ft. per lb.,
then the mass rate of flow is w = Av/v lb. per sec. If the area of the cross
section changes, the law of the continuity of mass shows that
A Aiv 2
(4A) w = mi lb./sec.
Vi a2
(4B) w = AiVipi = A 2V 2 P 2 lb./sec.,

forv fps, A sq. ft., v cu. ft. per lb., and p lb. per cu. ft. Equation (4) says
that the mass of flow past boundary 1 is equal to the mass passing boundary
2 —
assuming no accumulation or diminution of mass between sections
1 and 2.

24. Conservation of Energy. The law of the conservation of energy


states that energy can he neither created nor destroyed.* It is a law based on
physical observations and not subject to mathematical proof. In its
is
application to energy transformations on this earth, there is no known excep-
tion, except as mass is converted into energy, and vice versa, and it is there-
fore an accepted principle and a reliable guide.
Historically, the first law of thermodynamics stated that work and heat
are mutually convertible (see § 82), but the tendency now is to broaden the
statement to include all forms of energy: one form of energy may he converted
into another.

26. Energy Equations for Closed Systems. The law of conservation of


energy may be stated in several different working forms. For example, for
any kind of a system, it should be evident, if energy is neither created nor
This law is not an idea which burst suddenly upon the scientific world. After scien-
*

tists began working with energy, it was years before the law was comprehended. Benja-
min Thompson (Count Rumford), 1753-1814, who has been called an arrogant and insuf-
ferable genius, really discovered the equivalence of work and heat in the course of
manufacturing cannon (1797) by boring solid metal submerged in water. He was
intrigued by the water boiling because of the mechanical work of boring, yet no heat had
been added to the water. He convinced himself, but not the world, that the then accepted
caloric theory of heat (a theory which supposed heat to be a substance without mass)
did not explain all known phenomena of heat, and that work and heat were in some
manner related phenomena. In his words, “Is it possible that such a quantity of heat
as would have caused five pounds of ice cold water to boil could have been furnished by
so inconsiderable a quantity of metallic dust merely in consequence of a change in its
capacity for heat?” Other experimenters later discovered more evidence, until some
fifty years after Rumford’s cannon experiments, Joule, with assistance from Lord Kelvin,
showed conclusively that mechanical work and heat are equivalent. (We are now just
a few years from Joule. See the footnote on p. 10. Considering the age of the earth
in billions of years and the age of man as about 1,000,000 years, thermodynamics is a

new science as is all science). Rumford was teaching school in Rumford, Mass, (now
Concord, N. II.) when he met, wooed, and won a wealthy widow. He was sympathetic
with the “other” side at the time of the Revolutionary Wnr and decided that it would
be smart to leave Boston with the British, which he did, deserting his wife and daughter.
Though he wanted to, ho never returned to America, but he gained fame and honors in
Europe and England.
f

32 CONSERVATION OF ENERGY [Ch 2


or solving for Q,

(7B) Q = AU + AW + A K + AP + W,
[steady flow energy equation]

where iH terms are jn the same units and a]] A v alues are the second state
property minus the first state property Since the foot-pound is a rela-
tively small unit for most of our purposes, we use the Btu more often and
Joule's constant J to show the conversion where necessary Also, (7) may
represent one pound or any number of pounds of fluid flowing during some
elapsed time You should fix in mind the meaning of (7), the method of
determining the v alue of each term (or the change in value, such as AU), and
learn these things so well that you will remember and recall them as long as
you practice engineering We know that the potential energy is

(ug/g.)z
P = J Btu

and AP is given by AP = P« — P, = (wg/gt)(t t — Zi)/J ~ to (zj — z )/J Btu,


x

that the kinetic energy K= wv 2


/{2g*J) Btu and
m(vi* - -Pi
1

AK
)

2goJ ‘

that the flow work is IP/ = pV/J Btu and A If'/ = (p t Vj — P\V\)/J Btu,
or (psoj — pi0i)/7 Btu per lb The convention of signs is the same as
explained for the nonflow equation (6)
The working substance of some systems may be at a higher temperature
than the environment m
which event, heat Q will naturally flow outward
Or the flow wall be inward if the working substance is at a lower temperature
than the environment In certain situations, heat is intentionally added or
abstracted concurrently with the natural passage of heat, as a steam m
boiler In most systems where heat undesirably moves or out, the m
amount of such heat is often negligibly small compared with other energy
quantities, hence, w c often think of Q as a planned transfer

27 Enthalpy A property which is found to be frequently useful to

engineers and scientists is one called enthalpy (//, h) pronounced en thal -py
It is a composite property applicable to all fluids and is defined by

(8A) h = u + Btu/lb and H= U+ Btu for w lb ,

where H— wh It is seen to be a point function (or property) because it is


defined in terms of point functions only The change in specific enthalpy is

Apv
(8B) Ah = Au -f
IT’
hi - hi
PiPi
'

J
§ 28] STEADY FLOW EQUATION WITH ENTHALPY S3

and in terms of differentials,

d
(8C) dh = du + ^ J
V)
- = du + V
J 4 V
+ J

Enthalpy* being a point function, f dh = h 2 — hi, from state 1 to state 2.


Since the absolute quantity of internal energy not known, the absolute is

value of enthalpy is unknown. However, it usually matters little, because


what we wish to know is the change of enthalpy. Enthalpy may be meas-
ured above any convenient datum.

28. Steady Flow Equation with Enthalpy. Inasmuch as the steady flow
equation (7) includes the terms

u + Wf = u +y and U+ IF/ = U+

we may replace these two terms with their equivalents, h and H. As we


shall learn later, it will be as easy or easier to determine the change in
enthalpy as it is to find the change in internal energy (not to mention pv).
It is also true that we shall seldom be concerned with the change in potential
energy AP; it is not to be considered unless specifically included or unless it
obviously belongs. The change of potential energy in a hydraulic power
plant is the significant change, but in heat engines, elevation differences are
relatively minor; or I\ ~ P 2 With these two revisions, the steady flow
.

equation becomes

(9A) Ki + hi + Q = IC 2 + A* + TF,
(9B) ~ +h + 1 Q = ~ + h + IF Btu.
2 [AP - 0]

In applications of equation (9), other terms may be zero (A K or Q or TF


or Ah). One of the objectives of this study is for the reader to learn to omit
energies which should be omitted and to include the ones which count.
There are several other arrangements of the steady flow equation which are
advantageous under one circumstance or another, including
(9C) Q = h2 - hi +K - 2 Ki + IF = Ah + AK + IF,

(9D) Q = Am + ^ + AK + TF,

(9E)

(9F)
dQ

dQ =
= du + ^yl+ dK + dW
dh + dK + dW.
= du +
^ + -^ + V
dK + dW,

[ant steady flow, AP = 0J

Equations (9C) and (9F) are important forms. If A IC


= 0, we have
(6B), 30. If IT — 0,
dQ — dh -f- dW] compare with the nonjiow equation p.

* Other names in use for this property are total heal and heal content, both
of which

should be avoided.
S4 CONSERVATION OF ENERGY [CA 2

we get dQ — dh + dK As for a nonflow process (§ 2d) Q and


are net
II
transfers of energy but as seen II ^ Jp dl here Practice writing the
equation (9) m different arrangements just to get acquainted with it
(See § 42 )

29 Applications of the Steady Flow Equation In order to apply any


energj equation to a device it ts necessary to know what the deuce does and
more or less how it does it One reason for this is that not all of the energj
terms in an energy equation (0) or (7) are applicable to all devices and
machines One or more terms usuallj drop out because they are irrelevant
or inconsequential Thus the first step in the solution of a thermodymamic
problem is to determine bj whatever means possible the operation of
the thing with which the problem is concerned ^ ith this knowledge jou
are in a position to decide or tentatively decide which energies are to be
considered
Consider a turbine gas or steam It receives a stream of fluid more or
less steadily at high pressure which then expands to a low pressure doing
work, In stationary practice this turbine will be fairly well insulated
that is Q « 0 From a practical point of \ lew the speed at which the fluid
approaches the turbine is not very different from the speed at which it
departs which is to say that there is not much change in kinetic energj
Thus the change of kinetic energy may be negligible Ah « 0 Now draw
an energy diagram Fig 210 considering all forms of energj but omitting
those which are not apropos or using the full steadv flow equation write
the energy equation for this part tular application It is best to do both
Thus from the energy equation (9C) we may write
(d) ir -Ah - Ah or II —Ah h, - h, Btu lb
(0 01 \Q » a/v 01

The same from energy diagrams


results are obtained
Tor another illustration consider a nozzle on a jet engine (see Fig 219
§ 322) V nozzle is a device which receives a fluid guides its expansion to
some lower pressure converting some of the entering energj into kinetic
energj at exit (Iig 12) No shafts are turned
so that II = 0 Also Q
~ 0 because the time of
passage of a particle of mass through the nozzle
is so short (a fraction of a second) that there is not

Fig 12 Energy D a enough time for the heat loss Q to be significant


gram for Nozzle The sys as compared to other energy quantities From
tern is bounded by the
internal surface of the
the energy diagram Fig 12 or from (9C) we get
nozzle and imaginary »
- - hi - h s Btu/Ib
planes at its extremities
,
' e' Ah Ah - (hi hi)
sections 1 and 2 [Q - 0 H 0)

which says that the increase in kinetic energy in the nozzle is equal to the
decrease of that useful property enthalpy In many nozzles the entering
§ SO] EXAMPLE So
liaetie energy is negligible. K z ~ 0 and ±K ~ K«. If this is so. equation
(e) becomes


© ^j = hi hi Btu/lb.

[Ki =0. <3=0, vr = 01

The reader probably observes that, even in these two simplest applica-
tions. one must have some engineering knowledge in order to judge the
appropriateness of the various energies. Such elementary knowledge (and
much morel is working knowledge for the engineer. Other applications
will be made here and there throughout the book. The student should
make an energy diagram for each problem solved.
SO. Example. An air compressor takes in air at 15 psia and discharges it at
100 psia: r t = 2 cu. ft. Ib. and r-_ = 0.5 cu. ft. Ib. The increase of internal energy
is A0 Btu Ib. and the work is 70 Btu Ib.: IP = 0 and \K = 0. How much heat is
transferred?
soimoy. The fo-ms of energy to be considered are Sow work, internal energy,
work, and of course heat. For I Ib. of substance.

_ (15)(1AA)(2)
rr~
W/1 ~ Pitt
5.55 Btu Ib.
~J~

IT,. = ££:- = qQ0)a^)(Q-3) = 9.25 Btu lb.


! '~
J (7S
Air, = TT/» - vrn = 9.25 - 5.55 = -r-3.7 Btu Tb.

Since work must be done or. the air to compress it. the work is negative,

IT = —70 Btu lb.:

and since the internal energy increases. A a ~ 11 — fit ---


-f-AO Btu / lb. Using
these various values in equation (9D). we solve for Q and find

Q = An -r ATT/ -j- TT = (vAO) -h (-r-3.7) -f- (-70) = -26.3 Btu Ib..

where the negative sign indicates that heat is rejected by the system. Some air
comnressors have water jackets for the purpose of cooling the air during compression:
hence the negative sign would be expected.
Since it is important to be able to utilize energy diagrams, we shall repeat the
solution from this viewpoint. The procedure is to decide upon what the system is.
say. the air on Its way through the compressor. Fig-
ure 13 sets the boundaries and shows what is known. 'U - 5-.V: n>2 =°25 -

The internal energy may be measured above the


datura 0 ? the entering air. which gives = 0 and R'=70 m
u- = AO, because «. —
in = 4-AO. At this stage, it !',=0* ,=J0

may not be known whether heat Sows in or out.


If we choose to show heat Q as transferred in. as in
Fig. IS.
Fix. 13. a plus sen for the answer would say that

The direction of energy flow as shown is correct.


38 ENERGY RELATIONS \Ch 3

points 1 and 2 between the states


for different processes Therefore, the
specific heat c as defined m
equation (10B) depends on the process, and there
are an infinite number of possible processes Tor example, the specific heat
of a constant volume process is different from that of a constant pressure
process Moreover, even in a particular process, the specific heat is not
constant (§ 44), but luckily it can usually be expressed as a function of
temperature so that equation (10B) may be integrated readily Tor small
\anations of temperature, the specific heat may be considered constant
Therefore, if the mean or instantaneous value is used, equation (10B) may
be integrated to give

(11) Q - wef dT = wc(Tt — 7\) Btu,


l CONSTANT SPECIFIC HEAT]

between the limits of states 1 to 2 where the value of c is taken according to


the process

36 Constant Volume Specific Heat There are two


processes whose specific heats are especially important and
Constant
useful constant and constant pressure c„ Con-
volume c.

1 sider first a nonflow constant volume system (Fig 14)


Since 11 — 0 the heat transferred into the system must go
Fig Con
14 into internal energy and the nonflow equation for unit
stant Volume mass gives Q. — A u where the subscript v suggests that
System
the volume remains constant Thus we have
(b) Q, — AU *= wc,(Tt - Ti) Btu or Q, = Au — c,(T t — T\ ) Btu/lb
[constant e.|

and a definition of c, in terms of point functions as

(12) c, - or c. - Bttt/lb-°R

[average value) [instantaneous value]

The second f6rm is recognized as a partial differentiation of u with respect

to T with v constant but we shall not adopt the partial


notation at this time Tor the ideal gas to bo studied
soon, it is not necessary for the volume to remain constant
in order for equation (12) to be true See Table I for
some values of c»

37 Constant Pressure Specific Heat Let the system


be a nonflow system as before, and add heat while the
pressure is maintained constant (Fig 15) The pressure
is balanced by a constant load F (Fig 15) on a fnctionless,
Con~
movable piston
*
For an average value of cp the heat is slant Pressure
1
(c) Qp = wc v AT Btu or Qp = c, AT Btu/lb , System

§ S7] CONSTANT PRESSURE SPECIFIC HEAT 39

Table I. CHARACTERISTIC CONSTANTS FOR GASES


(a) For the gases marked (a), the values of c r are instantaneous values at
540°R = S0°F taken from spectroscopic data. Then c„ was found from c r = c p — R/J,
and k from c p /c r These values are for zero pressure, but the}’ are suitablj' accurate
.

for ordinary computations at ordinary pressures.


(b) For the gases marked (b), the values of c P and k were taken from the Inter-
national Critical Tables, Vol. V, for standard atmospheric pressure and 15°C (59°F).
Then c- was computed from c„ = c v /k.
The gas constant R for each gas was computed from Ii = p/(pT), where p = 1/v
was taken from the International Critical Tables, Yol. Ill, for standard atmospheric
pressure and 32°F. M= molecular weight; c p and cr are in Btu per lb-°R. To
obtain C, Btu per mol, use Ct = CF — 1.986.

Air
Gas

(a)
M

28 970
HD B
0.24 0.1715 1.4
k

6.95 53.3
R MR

1545
Carbon monoxide (CO) (a) 28.010 0.2487 0.1779 1.39S 6.97 55.1 1543
Hydrogen (H ; ) (a) 2 016 3.421 2.4354 1.405 6.9 767 1546
Nitric oxide (NO) (a) 30.008 0.2378 0.1717 1.384 7.14 51.4 1542
Nitrogen (N-) (a) 28.016 0.2484 0.1776 1.4 6.96 55.1 1543
Oxygen (0 : ) (a) 32 0.2193 0.1573 1.394 7.02 48.25 1544

Argon (A) (b) 39.95 0.125 0.0749 1.668 4.99 38.7 1544
Helium (He) (b) 4.003 1.25 0.754 1.659 5.00 386 1544

Carbon dioxide (CO*) (a) 44.01 0.202 0.157 1.29 8.89 34.9 1536
Hx’drogen sulfide (H:S) (a) 34.08 0.328 0.270 1.21 11.18 44.8 1523
Nitrous oxide (N-O) (a) 44.02 0.211 0.166 1.27 9.29 34.9 1536
Sulfur dioxide (SO*) (a) 64.06 0.154 0.123 1.25 9.87 23.6 1512

Acetvlene (C:Hi) (a) 26.04 0.409 0.333 1.23 10.65 58.8 1529
Ethane (C-He) (a) 30.07 0.422 0.357 1.18 12.69 50.8 1524
Ethvlene (C*H<) (a) 28.05 0.374 0.304 1.23 10.49 54.7 1532
Isobutane (CiHio) (a) 58.12 0.420 0.387 1.09 24.4 25.8 1499
Methane (CH<) (a) 16.04 0.533 0.409 1.30 8.55 96.2 1539
Propane (CiHs) (a) 44.09 0.404 0.360 1.12 17.81 34.1 1504

this heat goes into internal energy and into work being done in raising the
weight F and overcoming the pressure of the surroundings. For one pound
the work is (§ 18),

,
(d)

w = J Jf pdv = fP J[ dv
1 ,
= P( V 2

J

y l)
=
~T
- —
PNl

[pressure constant only]

Then the energy equation, dQ = du + p dv for 1 lb., applied to a constant

pressure process gives

(e) = U2 _ Ml + VJjl - Pp = h, - h, = Ah = c P (T , - Ti) Btu/lb.,


Qp
4° ENERGY RELATIONS [Ch 3

where u + pv/J has been replaced by h, the specific enthalpy Thus ne


see that cp may be defined in terms of point functions as

(13) = or c' = (^), Btu ^° R


[average value] [instantaneous value]

There are advantages in some advanced thermodynamic problems in having


the definitions of c, and c„ in terms of point functions Moreover, since
work entering the systems in Tigs 14 and 15 may bring about the same
terminal state, equations (12) and (13) free the definitions of these specific
heats of dependence on heat Nevertheless the concept of heat per unit
mass °R is very useful and holds for processes in which there is no fluid
friction See Table I for values of c„ for a few common substances
Air is so often the working substance that the student should memorize
its specific heats at normal atmospheric temperatures

air cp « 0 24, and c. - 0 1715 Btu/lb °R,

or Btu per lb "F The ratio of the specific heats c p and c, appears so often in
thermodynamic equations that it is given a symbol k

(14) k =& > 1

The value of k always greater than unity because for a particular tem-
is

perature difference AT, Ah = Au + A pv is always greater than Au that is


to say that the heat in the constant pressure process not only increases the
internal energy but also does work From equations (b) (e) and (14) we
notice that Ah/Au - k
38 Reversibility A process or a cycle which is reversible in every
respect is one which is as perfect as the mind can conceive Inasmuch as
thermodynamic analyses treat first of events which are reversible (ideal)
in some respect, it behooves us to search for the significant meanings of the
word reversibility (There are many metaphors in the language which show
that mankind has long realized the inevitable irreversibility of time and
*
actual events )

Adiabat c Walls
^ \
Consider a nonflow system of an ex-
pansible fluidG in a cylinder (Fig 16)
By
\°$E$CX (X ) A/
c \\ surrounded by adiabatic watts
definition, adiabatic means that the
JJ
transferred heat is zero (Q = 0), and
Flywheel' ^ adiabatic walls are ones which are per-

Fig 16 fectly insulated (There is no such


thing ) Imagine that the piston is an
infinitesimal movement past head-end dead center position (piston at its
* No use crying over spilled milk All the king s horses and all the king s men could
not put Ilumpty Dumpty together again
§ 38} REVERSIBILITY 41

farthest leftward position) and that the fluid G is at a pressure above atmos-
pheric. If the engine is free tomove, G expands and does work at the
expense of its internal energy (Q = AU +IF = 0). If the start is from rest,
this work can be used to speed up the flywheel, storing kinetic energy in it.
When the expansion of G is completed at the end of the stroke, the kinetic
energy in the flywheel is a maximum. Now if this process is reversible, the
kinetic energy in the flywheel is just enough to compress G back to its original
«i, T i), where the engine again comes to rest.
state (p t , This event will
not happen because no actual process is reversible.
() External irreversibility is some irreversibility external to the system,
when the system is the working substance; for example, friction at the bear-
ings A, B, and C, between the piston and cylinder walls, and between the
atmosphere and the rotating members; all absorb some of the IF output of
the system. Instinctively you know that you cannot use the “frictional
energy” to recompress the fluid to its original state (which is one of the

things said by the second law of thermodynamics see Chapter 7).
Another form of external irreversibility is due to the flow of heat through
the containing -walls, inasmuch as an adiabatic wall is a fiction of the mind.
Since the temperature of the substance changes during the expansion, there
is certain to be a temperature difference AT with the surroundings. Heat is
that energy which is transferred because of a temperature difference, and if a
temperature difference exists, heat is necessarily transferred. We say that
the flow of heat from a body at high temperature to a body at low tem-
perature is irreversible because all experience suggests the improbability of
its flowing in the opposite direction. As Clausius* said in 1850, “It is

impossible for a self-acting machine unaided by external agency to move


heat from one body to another at a higher temperature." (Tins is a state-
ment of the second law of thermodynamics, Chapter 7.) We see then that
heat can be transferred reversibly only when AT = 0 and that the transfer
approaches reversibility as AT approaches zero.
() Internal irreversibility is irreversibility caused by fluid friction within
the system, or an internal event which is irreversible, such as the mixing
it is

or diffusion oftwo or more gases. You can easily imagine that a large and
fast movement of the piston would result in a sudden expansion of G (Fig.
16) and momentary differences in the properties, say, pressure and tem-
perature, in various parts of the fluid. That part adjacent to the piston
would expand first. Because of the lower pressure in the vicinity of the
piston, gas would rush in from the higher pressure regions, there would be
*Rudolf Julius Emmanuel Clausius (1S22-1S8S), born in northern Germany, a pro-
fessor of physics, was a genius in mathematical investigations of natural phenomena.
He elaborated and restated the work of Carnot, deducing the principle of the second law
of thermodynamics. His mathematical work in optics, electricity, and electrolysis was
significant. James Clerk Maxwell credits him with being the founder of the kinetic
theory of gases, on the basis of which Clausius made many important calculations, one
of which was the mean free path of a molecule. He is the author of an exhaustive
treatise on the steam engine, wherein he emphasized the then new conception of entropy.
43 ENERGY RELATIONS [Ch S
whirlpools and eddies, and consequently fluid friction These effects can
be neither ascertained m
detail nor completely controlled While this inter-
nal motion exists, the fluid not in an equilibrium condition
is Some of the
energy which might have been delivered as work remains stored in the sub-
stance In order for this phase of the operation to be rev ersible, the piston
would have to move very slowly (at an infinitesimal speed in the limit),
the substance must pass through a series of equilibrium states, which is to
say that there is no fluid turbulence of any degree At any instant, all parts
of the system must be at the same temperature and pressure While
internal rev crsibility cannot be achic\ ed, actual processes can approach it
closely It is internal reversibility that we most often assume in this study
is II =
For example, the work of a fluid fp dV when the nonflow process is
internally reversible (p is the instantaneous pressure in all parts of the
system), whether there is or is not external reversibility
The internally reversible process is a controlled process, as contrasted
with an uncontrolled expansion, as in opening a valve and letting the fluid
flow out without doing work On the other hand, a guided expansion in a
nozzle maj closely approach a rev enable process, because the kinetic energy
of the jet can be reconverted in such a manner as to bring the substance
back close to its original state in the absence of fluid friction and turbu-
lence during the compression process
To summarize If, after a process consisting of a continuous series of equi-
librium stales is completed, the substance can Ik made to retrace m
the reverse
order the various states of the original process, if all energy quantities to or
from the surroundings can be returned to their original states ( everything as it

was before the process), then the process is externally and internally reversible
The conditions for reversibility are (1) processes controlled through a series
of equilibrium states (no fluid friction), (2) no mechanical friction, and (3)
no temperature difference during transfer of heat Perhaps we should add
a fourth condition, that there is no diffusion For example, if gases A and
B are separate but at the same temperature and pressure, they will gradually
mix (diffuse) when the separating partition is removed This is an irre-
versible process which could be construed as being covered by condition (1)
In engineering, we usually care for irrev crsibihtes or convert ideal results into
actual results by the use of ratios and efficiency numbers which are based
on experience

39 Entropy. has come to introduce another property, entropy


The time
(en'-tro-py) We shall happens to be of great
discuss this property, winch
utility, in more detail you have become better acquainted with it
later after
Entropy (s, S) is zero at absolute 2 ero temperature, but unless tabulated
absolute values of entropy are available, we generally compute only changes
of entropy and w e generally define it in terms of a change Tor an internally
reversible process, the change of entropy of a substance receiving (or deliver-
§ J^o] TEMPERATURE-ENTROPY COORDINATES 43

mg) heat is defined by

(15 A) dS =
^ or AS =
f ^ Btu/°R.
= -
(15B) As s2 s, = J ^ Btu/lb-°R,
[REVERSIBLE TROCESS]

where dQ
is the heat transferred [total amount in equation (15A), for one

pound equation (15B)] at the temperature T; AS is the total change of


in
entropy for w lb.; As is the change of specific entropy. If the integrations
are made from 1 to 2, the signs of the answers obtained will always accord
with the conventions set up in Chapter 2; that is, an increase is positive, a
decrease negative.
In processes where the specific heat is a known finite value, we may use
dQ = wc dT and evaluate the change in entropy as

C
(f) AS = w -^!- Btu/°R; or
J
AS = wc = wc In
(g)
J^ y Btu/°R,
[constant specific heat]

between the limits of states 1 and 2, where the specific heat is the proper
average value for the kind of process involved; for example, for a reversible
constant pressure process, AS = c v In (T 2 /T i) Btu per lb-°R.
Considering equation (15), AS = J dQ/T, it appears that S is not a
point function since it is defined in terms of a path function Q yet ;
we truly
stated that entropy is a property. As will be shown later, dividing dQ by T
results in a property.

40. Temperature -Entropy Coordinates. If equation (15) is rewritten as

(15C) Q = JT dS Btu, or

Q — J T ds Btu/lb.

and compared with W


— Jp dV, we observe that
T, like p, is an intensity factor and (s, S), like ( v ,

V), is an extensive factor. We see that the product


of T and dS, during a reversible process, gives heat
and that this product represents an area on the TS
plane (Fig. 17). The integration of T dS from 1
Fig. 17. TS Plane. A
to 2 (Fig. 17) gives the area under the curve down differential area is T dS,
to absolute zero temperature, area m-l-2-n, which recognized as dQ.
represents the amount of heat transferred during
the process 1-2. a useful attribute of entropy and makes the
This is

TS or Ts plane a favorite one for depicting processes and cycles. Since


U ENERGY RELATIONS [C/t 3

dQ — dU + p dV/J = T dS, we can write

00

which can be integrated for certain circumstances (§ G2)


The process must be reversible for either of the areas, fp dV or JT dS, to
hav e significance Irreversible processes will be shown dotted as a reminder
that the areas under the curves do not represent work or heat

41 fp dV for Reversible Steady Flow Processes It might be supposed


that the integral fpdV, which gives the work for a nonflow process has
some significance in steady flow processes Let us n rite the energy equation
as established for reversible nonflow processes as follows

(0 Q - \U - fpdl ft-lb ,

and the steady flow equation as

(j) Q ~ - All , + AK + A P + »r ft-lb

We recall that the change of internal energy depends only on the location of
the state points and 2 that it is independent of the nature of the process
1

connecting the points and that therefore the change is the same between any
two states no matter whether the process is reversible or irreversible non-
flow or steady flow Limiting our discussion to reversible processes we
recall that the heat added in an internally rev ersiblc process is expressed in
general by wfcdT where c has a definite and characteristic value depending
upon the kind of process only That is wfc dT has the same value for say
a constant pressure process no matter whether this process is nonflow or
steady flow Thus the left hand sides of the two foregoing equations are
identical for a particular rev ersiblc process Since the right hand sides of
equations (i) and (j) are equal to the same thing they must be equal to
each other,

(k) fp dT = ATT / + A K + AP + TT ft-lb

In other words, the integral fp dV evaluates the total change in the mecham
cal energy terms of the steady flow equation for reversible processes Equa
tiOTi {\) vs therefore an energy equation applying to either reversible hotvJTow.
or steady flow processes In differential form, (i) becomes

(16) dQ -TdS = dU + Btu or dQ - T da = du + Btu/lb ,

| ANT RE1FRSIBLE 1 ROCESS]

where p dV is defined by equation (k) Note that equation (k) reduces to


the nonflow relation fp dV = IF when the terms associated with motion are
taken out
§ 42} OTHER ENERGY RELATIONS 45

Other Energy Relations. With equation (16) established for gen-


42.
eral use and with the definition of enthalpy and other energy quantities, a
variety of energy equations can be obtained. Some are particularly useful
and significant. By way of illustration, we differentiate H = U + pV/J,
and get

(1) dH = dU + 2^1 + Btu.

But dU + p dV/J — dQ = T dS from equation (16); so for reversible


processes, equation (1) becomes

V P
(m) dH = dQ + or dQ = dH
J
,

or
v
(17) TdS = dH - and Tds = dh- -^,
[ant reversible process! [for one pound]

where dQ = T dS (and T ds) has been used and where the integration would
be from state 1 to 2. The integral fV dp represents an area (Fig. 18), the
area “behind” the curve. Equation (17) says this area is

(n) - j J~ V dp =+ j V dp = (T dS - dH) = Q — AH Btu.

[reversible process]

ideal situations, the area repre-


If Q should be zero, as it is often assumed in
sents AH. Now in equation (9E), which is
V
(o) dQ = du + + -^- + dK + dW
let du + p dv/J = dQ, and get

(p) — J~ v dp = W 4" AK ft-lb./lb.


and
Area for fV dp.
- J V dp = W 4 AK ft-lb.
2 Fig. 18.
Area 1-2 -a-b
represents the
i

[reversible steady flow, A P = 0] work of a reversible steady flow


process when A
— 0 and K
As mentioned before, the potential energy is
&P = 0.

seldom, if ever, significant in a heat engine,


dP = 0. Therefore, we may say that the area — fV dp represents the work
J
of a steady flow process when AK = 0, or it represents AK when W — 0, and
in both cases it represents —AH if Q = 0. Observe that fV dp is a
positivenumber when integrated from 1 to 2 (Fig. 18), the work being done
by the substance. In order to make the integration, one must
know the
40 ENERGY RELATIONS [Ch 8

relation between p and V (as pV n C). Practice arranging energy


equations now

43. Molar Specific Heats. It is often convenient to use specific heats


for a mol of substance A mol is pounds, where M is the molecular M
weight of the substance Tor example, a mol of oxygen is 32 lb The
corresponding -values are called molar specific heats C, C = Me Btn per
mol-°R;

(q) Cp = Mcp ,
C. = Me,
44. Variable Specific Heats.* As previously mentioned, specific heats in
general are not constant They not only vary markedly with tempera-
ture, but also with pressure The specific heat values to be given below
are for zero pressure, but they are used for finite pressures in the absence of
better data and serve very well for most engineering situations The
specific heats of vapors close to temperatures w here they will condense vary
markedly w ith pressure Tor gases such as air at very high pressures and at
temperatures below, say, 1000°F, the variation of specific heat with pressure
is large {15) At ordinary atmospheric temperatures, the values for gases
at zero pressure are beginning to be questionable if accuracy is important
when the pressure exceeds, say, 200 psia However, specific heats for zero
pressure are successfully used in analyzing internal combustion engines
where pressures are much higher than 200 psia (but the temperatures are
higher also and pressure has less effect at high temperature above 2000T; —
The curves of Tig 19 give instantaneous values of the molar specific heat
at constant pressure These curves can also be used to estimate approxi-
mately the mean values of C„ between any two temperatures A constant
Volume specific heat is obtained from C. = Cp/k, or better from

C, = Cr -l 980 Btu/mol-°R

As previously explained, the internal energy of a gas is largely due to the


translational energy of the molecule, to the energy of rotation, and to the
energy of vibration of the atoms within the molecule At high tempera-
tures, the vibrational energy increases markedly, and since only the trans-
lational energy is measured by the temperature, the specific heat must
increase to account for the absorption of the energy which goes to increase
the vibrations Observe how abruptly the curve for COt rises, this suggests
that the vibrational energy of a polyatomic molecule undergoes a consider-
able change at ordinary temperatures Trom this theory, it follow's that
the specific heats of monatomic gases, in which the molecule has practically
only translation energy, do not vary with temperature
It is interesting to recall that in accordance with the kinetic theory of

* In short courses, it may be desired to omit part of or all the remainder of this chapter,
except perhaps § 51
;

§ /t Jt ] VARIABLE SPECIFIC HEATS 47

1.4

1.3

1 2
a
s
d-

l.i

500 1000 2000 3000 -1000 5000


Temperature, Degrees Rankine

Plotted from data in Keenan and Kaye, Gas Tables (John Wiley)

Fig. 19. Molar Specific Heals at Constant Pressure. The molar specific heat at
constant volume is C, = CT - 1.986 or C r = C r /h Btu per mol, where values of k are
on the right-hand ordinate. Divide C„ (or C v) by the molecular weight M to obtain
c r Btu per lb.

gases, which you studied in physics, the specific heats of gases in the absence

of vibration of the atoms have the values:

3R 57?
Monatomic ~ k = 1.C67
gases, cv
2
Cp
2T
5R 111
=
Diatomic gases, 2./’
Cp ~ 2.7’
fc 1-4;

37? 47?
Polyatomic gases, Cp ~ J ’ k = 1.333;
T’
where R is called the gas constant (Chapter 4) Compare these values with
.

those in Table I, which come from experiment.


(b) Nil,, ammonia 17 03 c, - 0 3G3 + 2 577710'
- 1 319r*/10* - 22 47rV10'
(540-1800°It) R/J - 0 no
(c) II,, hj droRcn 2 010 c, = 2 857 + 2 8077710* CP = 5 70 + 5 787710'
4- 9 92/7’ 1/ * + 20/r i/*

(540- 1000°Il) R/J- 0980

(b) C,H«, ethylene 28 03 c, - 0 0965 + 5 787 /10‘ C, = 2 706 + 1627710'


- 9 977’VIO* -279 OT'VIO’
(540 2700°R> R/J = 0 0731

(b) C*II, acetylene 26 02 cp - 0459 + 0 9372710 4 i * 11 94 + 24 3727 10'


- 2 89 X 10‘/7’» - 75 2 X io'/r>
(500 2300°R) R/J = 0076

48
§ 46] MEAN SPECIFIC HEAT 49
The empirical equations of Table II make it possible to integrate; for
example,

= cdT
(r) « - cdT and As f
T
[per pound]
1 J

(s) f CdT and AS = f CdT


’ [per mol]
J T
thus allowing for the variation of specific heat with temperature. To find
the specific heat at constant volume for each substance in Table II, subtract
R/J from the equation for c p that is, the coefficients of the T terms remain
;

the same; c„ = c p — R/J. To find the molar specific heat at constant


volume, subtract 1.986 from Cp ;
that is, Cv = Cp — 1.986.

45. Example. What is the change of entropy of ait which is heated at constant
pressure from t, =
40°F to = 1540°F?
solution. In Table II,
W'e see that the constant pressure specific heat of air is
given in terms of absolute temperature (T, = 500°R, 7’ 2 = 2000°R) by an equation
of the form

(t) cp = a + (3T + yT* Btu/lb-°R


-4
where a = 0.219,(3 = 0.342 X 10 ,
and 7 = —0.293 X 10~ s . From As = / dQ/T,
we have

— I
r
5
?- £•(=? + »" + *")
7*2 -12000

= [«\nT + l3T + y Y \ 500


200(1 'v
= a In yr~ + (3(2000 - 500) +| (2000 2 - 500 2 ),

from which As = 0.35 Btu per lb-°R.

46. Mean Specific Heat. The use of equations for specific heats becomes
rather tedious at times. To ease the burden of computations, tables and
charts of properties, such as entropy and enthalpy, of many common sub-
stances are available: air, nitrogen, oxygen { 14 ), ammonia, freon, steam,
etc. (See the material at the back of Problems on Thermodynamics by
Faires, Brewer, and Simmang.) In the absence of tabulated properties,
equations of specific heats are useful in many practical applications where
accuracy is desired. If a number of repetitive calculations are to be made
for approximately a constant temperature range, the mean specific heat
c and C may be used:

= fc dT
or r_ ICdT
(u) c
Ti - Ti T,- Ti
For quick estimations, the mean values as obtained from curves such as
Fig. 19 are satisfactory.
(e) CII», methane 1C 03 I c, -=0 211 + G 257710‘ C, 3 38 + 100 27/10*
- 8 287710* - 132 77710*
0 1237

(e) CjHj, ethane 30 05 c, = 0 0731 + 7 087710* C, = 2 195 + 212 77/10*


- 11 37710* - 3407710*
(540-2700 o R) R/J «= 00653
(e) C<H]«, 7i butane 58 OS c, = 0075 + 6 947710* C, = 4 30 + 4037/10*
- 11 777’VIO’ - 6837710*
(540-2700°R) R/J = 0 0331
(e) CiHg, propane 44 06 c, = 00512 + 7 277/10* Cp * 2 258 + 3207/10*
- 12 327710' - 5437710*
(540~2700°R) R/J *= 0 0439
§ 46} MEAN SPECIFIC HEAT 49
The empirical equations of Table II make it possible to integrate; for
example,

cdT and f cdT


(r) Q - As = [per pound]
! J T

CdT and AS = f
CdT
(s) Q = [per mol]
J 1 T

thus allowing for the variation of specific heat with temperature. To find
the specific heat at constant volume for each substance in Table II, subtract
R/J from the equation for c p that is, the coefficients of the T terms remain
;

the same; Cp = c p — R/J. To find the molar specific heat at constant


volume, subtract 1.986 from Cp ; that is, C r = Cp — 1.986.

45. Example. What is the change of entropy of ait which is heated at constant
pressure from li =
40°F to = 1540°F?
solution. In Table II, we see that the constant pressure specific heat of air is
given in terms of absolute temperature (T, = 500°R, T2 — 2000°R) by an equation
of the form

(t) ^= or + $T + yT- Btu/lb-°R


where a = 0.219, /S = 0.342 X 10-Sand 7 = -0.293 X 10~ 8 . From As = / dQ/T,
we have

^2 “12000
[
a In T + PT + 7 yJ 500
2000 -v
= a In -ggg + jS(2000 - 500) + | (2000 - 500=),2

from which As = 0.35 Btu per lb-°R.

46. Mean Specific Heat. The use of equations for specific heats becomes
rather tedious at times. To ease the burden of computations, tables and
charts of properties, such as entropy and enthalpy, of many common sub-
stances are available: air, nitrogen, oxygen (14), ammonia, freon, steam,
etc. (See the material at the back of Problems on Thermodynamics by
Faires, Brewer, and Simmang.) In the absence of tabulated properties,
equations of specific heats are useful in many practical applications where
accuracy is desired. If a number of repetitive calculations are to be made
for approximately a constant temperature range, the mean specific heat
c and C may be used:
dT
(u) or C = fC-
Ti Ti

For quick estimations, the mean values as obtained from cun es such as r

Fig. 19 are satisfactory.


50 ENERGY RELATIONS [Ch 3

47. Steady Flow of More than One Mass Stream W e recall the stipu-
lations for stead} flowthat there be no change in either the total mass or the
total energ} within the sj stem In tins exent, the law of conservation of
energ} «ays that energ} entering js equal to energ} leaving With this
knowledge, it is just as eas> to wnte the energ} equation for an} number of
substances as for one, but it takes more time and the equation looks more
formidable However, b} using E to represent the total energ} associated

Fig 20 Energy Diagram Steady Flow — Two Fluids


with 1 lb of a particular stream t -w + pi-f-A-f-/* — A-f-A-f/*
and b\ using onl} two streams for illustration the resulting equation looks
harmless enough Thus for fluid 4, Fig 20 we ha\e

Entering EA - Au +
i Aj, + PAU Leaving EAi = hA t + Kai + Pm,
and for fluid B,

Entering Eu =
i A«i + Kbi + Pbi, Leaving F bj — Ab» + As- + Pbi

The potential energ} P raaj be omitted as in Fig 20 or included as appro-


priate The energ} equation becomes

(v) tPj£Jji 4- v?bEbi +Q~ + xcbEbi + IT Btu

where Q and IF are net values and the comention of signs is as pre\iou«l}
explained

48 Vanable Flow Process There are cares m engmeenng where the


quantit} of mass and energ} within a s}stem do not remain constant
Typical of these would be a fluid flowing into a tank (and not out) or out of a
tank (and not in) When the flow is \anable one must consider what
happens during a particular period of time, or w hat happens while a certain
mass lea\ es or enters the s}stem, or ret up the energ} equation in terms of
differentials Let E consist m
general of all stored energ} quantities in J lb
of the system at an} instant In terms of the ones w e ha\ e been considering,
E=u +K+ P Btu per lb (Note that this is not the same E as in the
previous article) In the beginning then (Fig 21), there are tr t Ei Btu of
.

§ 1,8] VARIABLE FLOW PROCESS 51

energy within the system; at the end, there are w«E 2 Btu. During an infini-
tesimal time interval, the change of energy stored in the system is d(wE)
Now apply the law of the conservation of energy, as stated in equation (5),
to Fig. 21, and find

Energy entering = change in stored energy + energy leaving.


(w) + K, + P,) + dQ — d(wE) + dw/(hf
div,(h, + K + P/) dW,
/ -f-

where i suggests the initial state at entry and / the state of the working
substance at the final (exit) boundary. The mass balance gives

Mass entering — mass leaving = change of mass within.


(x) dw,— dwf = u 2 — w h ’

where w 2 — W\ may be positive (increase) or negative (decrease). The fore-


going two equations are good for two unknowns, but in many actual prob-
lems, assumptions must be made and special data collected in order to be

dw, (h,+K,+Pi’•>
£i Btu/lb. dW
7T~ !(/] at Start
lb.
E« Btu/lb.
lb at End I

c k 7 ^7
d (wE) Change
pr dwf (hf+Kf+Pf )

"System
dQ I

Fig. 21. Energy Diagram — Variable Flow.


able to solve equation (w) for an unknown. If, for example, the system is a

tank of gas from which there is an outflow only, the term dw,()i, + K, + P,),
which is the energy accompanying the inflow of dw, lb. of substance, would
be dropped, and dW = 0 because IF is the kind of work which turns shafts;
Pf = 0; d(wE) = d{wu); and dQ = 0 if the time element is short. Not-
withstanding the foregoing simplifications, it is still true that for each dw f
flowing out, there is a different value of h/ and K/, and hence there would be
difficulty in summing the terms dwf {lij + K f ) over a finite time. (It
could be further assumed that the fluid which remains in the tank under-

goes a constant entropy process a process to be described later.) Observe
that if the properties of the system (pressure and temperature, say) are
measured before and after the event, and if the amount of outflow (or
inflow) is known, the change of internal energy A wit — w 2 u 2 — wpii can be
found, as explained in detail later.
The problem is somewhat simpler inward only from a source
if the flow is

of substance where h, = 0, as from a pipe line in


and K, are constant and P,
which substance is flowing steadily (more or less). The student should by
now be able to set up an energy diagram and obtain the energy equation for
this case with little difficulty. The stored energy E becomes internal
52 ENERGY RELATIONS [Ch 3

energy u because kinetic energy K


of a gas in a tank is virtually zero and
potential energy P is irrelevant See references 3, 10,13, and 22

49 Frictional Energy. One frequently hears the term “frictional energy' 1

or its equivalent, but it is not a form of energy different from those forms
already defined here To allow for energy losses due to friction, engineers
generally use cfficicncj numbers which ha\e been accumulated through
experience This practice is satisfactory and will be explamed as the occa-

sion arises m later pages However, we may profit now from a brief con-
sideration of the nature of this energy
If the frictional loss is caused by the rubbing together of two sohd parts,
with or without lubrication, as in a bearing, what happens is that work is
expended to excite the molecules in the vicinity of the surfaces being rubbed,
raising their temperatures This is to say that work is converted into
internal energy of the parts (of a bearing say) Trom the hotter surfaces
heat is conducted through the parts, and if the environmental temperature is
lower than that of the parts, this energy is radiated and convectcd awaj
from the parts as Q That portion of this energy carried off by the oil will
also be given off as heat when the oil is cooled Thus the frictional energy
m this situation eventually appears as heat —
heat dissipated into the
atmosphere, winch is such an immense reservoir that it does not notice the
difference except locally
If the frictional loss is that caused by fluid friction, the substance has
more internal energy at the end of the process than it would hav e had in the
absence of fluid friction If the substance also gets hotter than its environ-
ment, at least a part of the “frictional energy” eventually leaves as heat to
the surroundings In all events, frictional energy, in effect is eventually
spilled into the atmosphere at the expense of work (no form of energy is
finer than work)
Suppose m Tig 5, we let 27/ represent the frictional losses in system (2)
at bearings E, D, C, etc ,
we see that IF/ = 11 s + 27/ In Tig 22, IF, is

the shaft work and 27/ is the frictional loss in the


bearings, etc of a steady flow machine If

V Q = 0, except as 27/ is heat, the energy equa-


tion is

hi -h Ki - h, + K* + IF, + EJf
where the work IF, which in the absence
fluid
of faction could have all been shaft work, is
W = IF. + 27/

50 Energy Equation for Flow of Incompressible Fluids The reader has


probably derived the Bernoulli equation in his study of fluid mechanics,
using a force analysis Heat and mtemal energy were not involved for a
good historical reason Bernoulli, a contemporary of Newton, derived the
§ 50} ENERGY EQUATION FOR FLOW OF INCOMPRESSIBLE FLUIDS 53

equation around 1700, whereas the law of conservation of energy was not

accepted until about 1850 150 years later. However, the energy law
throws additional light on the Bernoulli equation. Consider the stead}' flow
energy equation in the following form,

Q = Am + A (pv) + A K + AP + TF ft-lb./lb.

as applied to Fig. 23, an open system of fluid flowing in a pipe between the
boundaries 1
2. In this system, IF = 0 and Q can be negligibly small
and
if there temperature difference between the inside and outside, or if
is little

the line is well insulated. Use v = 1/p, where p lb. per cu. ft. is the density.
An incompressible fluid in frictionless flow with Q = 0 will undergo no
change of internal energy, An = 0 Between the boundaries 1 and 2, the
energy relation then becomes

A f?) + AK + AP = 0,

2
P2 Pi ur HL =
(y)
pi
|

2<7„
+ Z2 _ 2i o ft-lb./lb.,
P2

in which g/g„, as
it modifies z, is taken as unity [equation (a), § 14], Ber-
equation for the flow of incompressible fluids without friction is
noulli’s
sometimes written in the form of equa-
tion (y). Observe that if the force
pound and the mass pound are canceled
(ordinarily not advisable), each term can
be interpreted as a head in feet, the name
usually used in hydraulics. The vr/ (2g„)
term is called the velocity head p/p is
,

called the pressure head. Ifthe system


of Fig. 23 should lead to a hydraulic
turbine at 2, the events would be that
some of the potential energy Az would Fig. 23. Energy Diagram — Fluid
be converted into kinetic energy and then
some of the kinetic energy would be converted into work in the turbine.
The reason that all the change of potential energy is not converted into
kinetic energy is that friction prevents it. If Q is negligible during the
flow of an incompressible fluid, the energy loss due to friction appears in the
internal energy of the fluid. Thus, using all the energy terms shown in Fig.
23, we have

(2)
P= _ V}
+ Hi'
2
- Hi!
2g„
+ 2 - 2l =
. -(?/. - Wl ) ft-lb./lb.
P2 Pi ga
This one form of Bernoulli’s equation with friction if we designate w» — iq
is

as the friction head, which is the sum of the changes of “all heads,” a net loss.
Recall also that Q = 0 and IF = 0. If Q is not equal to rero, the equation
54 ENERGY RELATIONS [Ch 8

(z) can be made to balance by making the friction head on the nght hand
side arbitrarily equal to the left hand side (rather than equal to w 2 — «i)
51 Closure would be superb if ever} thing about thermodynamics
It
could be said at once and be done with Not only is this impossible but
it is also often difficult to write about one idea without involving another
Thus we have referred to se\ eral notions which are to be explained in detail
later Perhaps these notions are not essential to the purposes of these first
three chapters Nevertheless if the reader will review these chapters a
time or two within the next few weeks he will no doubt find meanings which
he missed on first study Specific heats entropy and the energ} relations
are basic to further study
W e have armed at two important mathematical relationships which are
applicable to all reversible thermodynamic processes namely

dQ-iu + e* and dQ-ih-’ff


[equat on (16)] Iequation (17)]

In each of these equations dQ - T ds for a reversible process


4

THE IDEAL GAS

52. Introduction. We shall now study the characteristics of the perfect


or ideal gas. Since the mathematics of an ideal gas is relatively simple, it is
fitting that this be the first working substance to be examined in detail.
Early experiments on air resulted in laws which we now recognize as being
characteristic of an ideal gas (§§ 53 and 54). That is, air is almost an ideal
gas in its most common states, as are all the monatomic and diatomic gases;
which is one practical reason for studying this hypothetical substance.
While no actual gas conforms exactly to the tenets of an ideal gas, many of
them conform so closely that ideal-gas calculations give good engineering
answers. Hence, this ideal substance is a practical one too.
Some substances in a gaseous state deviate substantially from ideal-gas
laws. In general, when this is true the temperature of the substance is
close to the boiling point of the liquid; that is, the substance is not too far

from the point of condensation at least as compared with nitrogen, for
instance, whose boiling point at 14.7 psia is — 321°F. A substance which is
at or near its condensation point is called a vapor. Thus, actual “ideal”
gases are either highly superheated or are vapors at a very low pressure (low
density).
Since ideal-gas laws are usually inaccurate for vapors, which are imperfect
gases, we must use either more elaborate mathematical procedures or tabu-
lated values and charts of properties. There is no distinct line of demarca-
tion dividing imperfect gases and ideal gases, but the criterion
simple: is

If the ideal gas laws yield sufficiently accurate results, the substance is
considered as an ideal gas; otherwise, it is a vapor or an imperfect gas. A
substance departs from the action characteristic of ideal gases when its
molecules are so close together that the forces of attraction between mole-
cules are large enough to affect significantly their behavior (§ 16). In one
55
66 the IDEAL GAS [Ch 4

condition, a substance may


act as an ideal gas, jn another, it may not For
instance, steam, as it exists in the atmosphere, is at such a low density that
it accords well with ideal-gas laws, steam in a steam power plant does
not
An ideal gas is ideal only in the sense that it conforms to the simple per-
fect-gas laws It might lack many characteristics which we would care to
assign to it, for instance, a greater heat conductivity

53 Boyle’s law. Robert Boyle ( 1Q27—IG91 ), during the course of experi-


ments with air, observed the following relation between the pressure and
volume If the temperature of a given quantity of gas is held constant, the
volume of gas varies inversely with the absolute pressure during a change of
state ( Boyle’s law).* In mathematical form, if a gas is in a condition repre-
sented by state 1 (Tig 24), and undergoes a change of state at constant tem-
perature ( isothermal process ) to state 2, then

18 or
<
> g-f;,
PiVi = PiVt, or p7 = C
(BOYLE 8 I AW T- Cl
The equation of the curve 1-2 (Fig
24) is thus
pV =
C, where <7 has some particular positive value
Fig 2i Boyle's Law
This curve is seen to be an equilateral hyperbola
A point such as 1 is a
state point (xy = C), and since C may have any one of an
The curve
joining 1 and 2 is theinfinite number of values, it is one of a family of
path of the stale point
curves
as the state of the sub
stance changes and is Observe that we shall use C in a generic sense to
spoken of as a processmoan a constant, but it will be a different constant
in different equations, that is, for example, stating m
Boyle’s law mathematically, we write t = C and pV = C, which the two m
C’s are different constants, t is not equal to pV
The experiments of Boyle were not so precise as modern science can make
them More accurate observations reveal that no actual gas follows this
law exactly Hence, w c consider Boyle’s law as being a defining character-
istic of an ideal gas
>
54 s Charles’ Law. About a hundred years after the discovery of Boyle’s
law, two Frenchmen, Jacques A Charles (174&-1823) and Joseph L Gay-
Lussab (1778-1850), each without knowledge of the other’s work, discovered
the law which we usually call Charles' law This law is in two parts

1 If the pressure on a particular quantity of gas is held constant then vnth any
change of slate, the tolume will i wry directly as the absolute temperature
* Edme Marcotte, ft Frenchman independently discovered this same principle ftt
approximately the same time that Bojle did Although Manottc therefore, is due
fully as much ctedit for the discover} , the law is more commonly called Boyle s in thie
country
§ 54] CHARLES’ LAW 57
2. If the volume of a particular quantity of gas is held constant, then, with any change
of slate, the pressure will vary directly as the absolute temperature.

Equations (19) and (20), which are mathematical expressions of Charles’


law,may be considered as other characteristics of the perfect gas, usable how-
ever for gases which are nearly perfect. As before, consider a change of
state for some gas, point 1 to point 2 [Fig. 25(a)], this time using the absolute
temperature T and volume V as coordinates, and letting the pressure at 1
equal to the pressure at 2, pl = p2 . Then we may write

Fi Ti
(19) or
Vt Ti
[CHARLES’ LAV, p ~ C]

The volume V in this law is any volume, including specific volume v, of a


constant mass of gas. Next, using the coordinates T and p [Fig. 25 (b)],
letting Vi = Vt, we have

(20 )
Pi
Pi
_ Ti
Tt
or —
Ti
Pi
=—
Ti
Pi
or

[CHARLES’ LAW, V = C]

The equations (19) and (20) are seen to be straight lines ( y = Cx) which
pass through the origin with slopes C. However, since all actual gases

Fig. 25. Charles’ Law. In (a), the line 1-2 represents a constant pressure process
foran ideal gas as it appears on the TV plane. In (b), the line 1-2 represents a constant
volume process as it appears on the Tp plane.

undergo changes of phase at low temperatures, this law does not hold down
to absolute zero. Charles’ lawis applicable to gases through only a limited

range of temperature.
Each of the parts of Charles’ law is used as a basis for measuring tem-
perature. If the most nearly ideal gases are used, accurate temperature
indications may be obtained from a constant-volume gas thermometer, in
which Ti/T 2 = P 1 /P 2 , or from a constant-pressure gas thermometer, in
which T 1 /T 2 = V 1/V 2 . Also, experiments on nearly perfect gases furnish
an important clue to absolute zero temperature. For example, the slope of
the line in Fig. 25(a) is

Ti - T _ AT _ At_ _ r
2
(a)
Vr - V 2 AV AV
Experimental data for air at p = 14.696 psia are: at I2 — 32°F, the specific
58 THE IDEAL GAS [Ch 4

volume is Vt — 12 39 cu ft ; at h - 100°F, e, = 14 1 cu ft. Then


100-32
- - 39 7G,
P, r, 14 1 12 39
T = Cv = 39 7Gp

This is the equation of the line on Tig 23(a) for air and the data grxen
Thus, the absolute temperature at 32*T (where r = 12 39) is

T= (39 70) (12 39) = 492 G° abs ,

which a rough estimation of the absolute temperature on the Fahrenheit


is

scale corresponding to the freezing point of water, 32°F If the data for
some other gas were used as aboxe, we should find a slightly different abso-
lute zero point, because exeiy gas deviates to some extent from Charles’
law Thus, we may ha\ e as mam absolute scales as ga^es Howex er, u«ing
more refined technique than the method gixen above and usng a nearly
perfect gas (helium at low pressure), we find that 32°F is 491 69°R See
§§ 8 and 9G

55 Equation of State of an Ideal Gas. An equation of state of a sub-


stance is one which relates three properties of the substance for example,
pressure as a function of x olume and temperature p = /(l T) Satis- ,

factory equations of state for rapors are complicated (Chapter 14), but the
one for an ideal gas is simple What we wish is a relation between p, i,
and T that relates any two states whatsoexer Such a relation may be
obtained by combining Boyle’s law with either part of Charles’ law, or by
combining the two parts of Charles’ law
Let an ideal gas be in state 1 (Fig 2G), and
let the state be changed at random 1-5-2, until

it isrepresented by point 2 Let the ma«s of


gas be one pound, for convenience so, the cor-
responding i olume is the specific x olume e
Xow through point 1, draw a horizontal Line of
indefinite extent representing a constant pres-
sure process on the pT plane, through point 2,
draw a vertical line of indefinite extent repre-
senting a constant x olume process The lines
representing the-e processes intersect at point a Applying Charles’ laxx
and noting that i a — Vt and p„ = p>, we find from the constant pressure
process 1-a,

and from the constant \ olume process 2-a,


§ 56] THE GAS CONSTANT 59
From equations (b) and (c), we find

or, by transposing terms,

(a)
Pi»i
-™— = V&i n
ts— = It, a constant.
J 1 I 2

Since points 1 and 2 were selected at random, it follows that a like relation
could be derived for any other pair of points, say, 1 and 3; thus

PlVl _ P3V3
Ti r,‘
Consequently, this expression pv/T must be constant for a particular gas.
Since the constant R is associated with a unit mass of substance, it is desig-
nated as the specific gas constant, ordinarily shortened, however, to the
gas constant. It is expressed in general terms as pv/T = R, or

(21) pv = RT and pV = wRT,


[1 LB.] [w LB.]

where V = wv. Equation (21) is the equation of state or the characteristic


equation of a perfect gas. From it may be obtained Boyle’s and Charles’
laws by letting T, p, and V be successively constant. For the gas constants
R given in Table I, always use p in pounds per square foot (psf), V in cubic
feet, and T in degrees Rankine in equation (21).

56. The Gas Constant. The value of R for any gas may be determined
from accurate experimental observations of simultaneous values of p, v, and
T. Thus, at 32°F and standard atmospheric pressure, the specific volume of
air is 12.39 cu. ft. Hence,

pv (14.7)(144)(12.39)
(e) K„ 53.3 ft-lb./lb-°R,
T (460 + 32)

the approximate gas constant for air only. Air is so frequently used that the
student should memorize this number. Gas constants for other gases are
given in Table I.
Consider the units of R. From equation (e) above, we may write
(pressure unit) (volume unit/unit mass)
K absolute temperature

Thus, R computed to be in (tons/sq. ft.)(cu. meters)/lb-°K; that is,


could be
any pressure unit, volume unit, and temperature scale could be used.
From another point of view, the unit of pv is an energy unit,

2
(lb./ft. )(ft.
3
) = ft-lb.
60 THE IDEAL GAS [C/t 4
Therefore, R could be expressed m terms of any energj unit,

energy unit
R t
(mass) (absolute temperature)

Common English units for the specific gas constant are

R— » ft lb /lb-°R, and Btu/lb-°It

constant for a mol (M lb ) of substance


The gas is often needed Multi-
ply both sides of pv = RT by and get M
(() MR =

According to Axogadro, an Itahan (177G-185G) All ideal gases at a particu-


lar pressure and temperature have the same number of molecules in a given
volume ( Avogodro’s law) Since the molecular weight is an index to the
weight of the molecule, it follows that the density p, say, pounds per cubic m
foot, is proportional to the weight of the molecule, or

Pm Me
(g)
Py Mj
where p, and p„ are the densities, and M, and M y are the molecular weights of
the gases A' and T, respectively As the densities are m\ ersely proportional
to the specific \olumes, pt /p„ = = Me/M „, or

(h) M*. - MyV,

We have shown that the molecular weight times the specific -volume for gas
X is equal to the product of these numbers for gas Y, the gases being at the
same pressure and temperature Since gases X
and Y may be any two gases,
it follows from Avogadro’s law that the product Mv must be the same for all
ideal gases at any particular temperature and pressure Therefore, in
equation (f), which applies to an ideal gas, the volume Mv vanes inversely
as the pressure (Boyle) and directly as the absolute temperature (Charles)
Since Mv is the same for all gases (A\ ogadro), it follows that is the same MR
constant for all gases This gas constant is called the universal gas con-
stant ~R and its accepted value is (14)

7C = 1545 32 ft lb /mol °R,

use 1545 for slide-rule work Now compare this value of R with values MR
given in Table I, p 39, and observe that a typical value for the more nearly
perfect gases is not too different from 1545 This is a convenient constant
to have in mmd, because one can often decide from memory the molecular
weight of a substance and then find the gas constant in ft-lb.
0
per lb- R from
R= 1545/M.
§ 57] JOULE’S LAW 61

For our purposes, the unit of the molecular weight number is lb. per M
mol. (In the European system of units, used in scientific work here, it
might be M
gm. per mol, which we call a gram-mol, or M
kg. per mol.)
From equation (f), the units of MR = R are

T (pressure umt) (volume unit) energy unit


W 5SPR '

Values of the universal gas constant R in various units are given in the
Appendix (23). When the universal gas constant is in terms of any cgs
units, as 82.06 atm-cc per (g-mol)-°K, the mol is a gram-mol; that is, for
example, 32 grams of 0 2 not 32 lb.
;

As you would expect by now, the actual value of Mv, called the mol
volume, is not constant because gases are not ideal. At 14.696 psia and
32°F, typical values of Mv are: hydrogen (H), 359 cu. ft.; helium (He),
358.8 cu. carbon dioxide (C0 2 ), 356.4 cu. ft.; ammonia (NH 3 ), 353.5
ft.;

cu. ft. The standard mol volume (14.696 psia and 32°F) for the ideal gas
may be taken as 359 cu. ft.

57. Joule’s Law. Joule arranged two copper containers, as shown diagram-
matically in Fig. 27, in a bath of water. The water was m an insulated
vessel, so that the w hole apparatus could
be considered as an isolated system, without
thermal contact with outside bodies. One
of the containers held air at a pressure of 22
atm [(22) (14.7) psia], and the other container
was evacuated as nearly as possible. After
all parts had reached thermal equilibrium,

the valve was opened. The air rushed into


the evacuated container, and the whole mass
of air finally came to rest at a pressure of 11
atm. Joule observed that there was no
change in the temperature of the water
surrounding the containers, a simple observation that leads to an important
deduction.
There are several facts to be noted. First, although there is a momentary
flow of air, giving rise to energy terms characteristic of flow, the net effect
due to flow is zero, since the gas is at rest in the initial and final states.
Consequently, the experiment is considered in relation to the simple energy
equation. Second, inasmuch as there is no connection by which energy may
leave the system as work, TF = 0. Third, because the temperature of the
water did not change, there must have been no flow of heat from the air in
the containers to the water, or from the water to the air, Q — 0. Fourth,
the temperature of the air in the containers did not change, since otherwise,
62 THE IDEAL GAS [Ch 4

due to the temperature difference there would ha\e been a transfer of


heat *
Now from the simple energ} equation AU — Q — 11 = 0 The pres
sure changed the xolume changed but the internal energ} did not change
Consequently we conclude that the internal energ} of this gas is not a func
tion of the pressure or xolume However since the temperature did not
change Joule deduced that the change of internal energy is a function of the
temperature change The italicized statement is known as Joule s law
"While the evidence from Joule s experiment leading to this deduction is not
conclusn e the law is true for an ideal gas and can be proved mathematical!}
for a gas conforming to the characteristic equation pi — RT Moreover
the law maj be explained from the viewpoint of the kinetic theorj of gaaes
As no have learned (§ 1G) the molecules in a gas are so far from one another
that the forces of attraction between molecules arc \er> small Even
though the distance between the molecules is changed b} a considerable
amount the change of potcntiaC mternaf energ} wifi be negligible small
because the attracts e forces arc so small I or ordinar} expansions or com

pressions of gases then the change of potential internal energ} is practicall}


zero Consequently an} internal energ} change must be a change of the
kinetic internal energ} the internal energ} due to the motion of the mole-
cule Since temperature is a measure of the molecular kinetic energ} it
follows that the change of internal energ} is a function of the temperature
change

68 Internal Energy of an Ideal Gas VI ith respect to the nonflow

constant xolume s}stem of §30 wc observed that the total effect of the
inflow of energ} was to increase the internal energ} of the system and we
found c, = (du/dT), [equation (12)] applicable to an} substance It can

be shown that for an ideal gas the limitation of constant \olume is not
necessary Therefore we ma> write c, — du/dT and

(22) An - jc, dT, A U = wfc.dT


[iDEVL UAS AN V KIND OF FRO(-E«s]

Tor actual gases A U approaches a function of T onl} as the pressure


approaches zero but equation (22) often gixes a good estimate of Ab for an}
gas If it is permissible to take the specific heat as constant integration of

* More precise experiments conducted later bj Joule and \\ illiam Thomson (Lord
Kelv in) show cd a charge of temperature of the air X porous plug « as used at the \ ni\ c
and extreme care was taken to prevent heat transfer to or from an external source Thus
real gases do have a change of temperature with a cl ange in pressure The rate of
change of temperature w ith pressure under t! e conditions of this experiment is called the
Joule Thomson coefficient Mathematically the Joule-Thomson coefficient = (dT/dp)i
where the subscript k means that the enthalpy is constant In a perfect gas this eoeffi
iC ent is zero Thus Joules law is true onlj for ideal gases is nearlj true for nearly
deal gases is decided]} in error for tapors The magnitude of the Joule-Thom«on eoeff
cient measures the degree of departure of tl e gas from the ideal
§ 60} EXAMPLE 63
equation (22) gives

G) At: = U« - Ui = «c,(T 2 - T 1) Btu,


[CONST ANT SPECIFIC HEAT]

which is the internal energy change of an ideal gas whenever the temperature
changes from TitoT* and c* is constant.

59. Enthalpy of an Ideal Gas. In § 37, we found c p ( dh/dT) p , equa- =


tion (13). As for the case of internal energy, it can be shown that the con-
stant pressure limitation is not necessary for an ideal gas. Therefore,
from cp = dh/dT. we get

(23) Ah = f cp dT, AH = w cp dT,


j
[ideal gas, a\t eis'd of process]

applicable to actual gases in most engineering processes. For constant


specific heat,

(k) AH = H« — Hx —icc p (Ti Ti) — = wcp AT Btu,


[constant specific heat]

which is the enthalpy change of an ideal gas whenever the temperature

changes from Ti to T« and cp is constant.

60. Example. A gas in the initial state of pi = 75 psia and Fi = 5 cu. ft. under-
goes a process to p. = 25 psia and V. = 9.6S cu. ft., during which the enthalpy
decreases 62 Btu. The specific heat at constant volume is c, = 0.745 Btu/lb-°R.
Determine (a) the change of internal energy, (b) the specific heat at constant
pressure, (c) the gas constant R.
solution. In general, if a method of solution is not immediately apparent, it is a
good idea to write down on your work sheet all that you can think of which might
have a bearing, including the most fundamental relations. If this is done for this
problem, the work sheet should have on it AU — icc-(T~ — T i) and pV = wRT;
with these two equations there to look at, it might occur to you to use pV = wRT in
order to eliminate the temperature in the equation for At'. [See part (c).] How-
ever, the gas constant R is unknown and this solution cannot be made. If you
have written the fundamental definition of enthalpy change, AH — AU + A(pF),
a solution for (a) is evident.
(a) Substitute AH = — 62 Btu because the change is a decrease. Then

_ 6o = AU “
K25X9.6S) -
7/8
(75)(5)jI44 = ^ ^
or AU = —62 -p 24.6 = —37.4 Btu. a decrease of U as indicated by the negative
sign. Be sure to remember to convert psi to psf (144) and to have all energy terms
in the same units (778).
(b) We notice from equations (j) and (k) above (constant k) that

= 1.655 = k,
Au cv —37.4
Cp = 1.655c, = (1.655) (0.754) = 1.25 Btu/lb-°R
64 THE IDEAL GAS [Ch 4

(c) Now we may use Tt = (p l V,)/(wR) and T = t {paVj)/(io/2) m the internal


energy equation and find

&V ~ wc.tr, - Ti) = | (ptVi - p,7,) Btu,

-37 4 =,
9^1 [(25)(9 08) - (75)(5)]144,

R= 386 ft-lb /lb-°R

note If the substance known, these constants c„ and R can be found in the
is

literature, as in Table I Comparing the answers found with the values in Table I,
we see that the gas in this problem is helium if it is a single gas. However, such a
problem as this might be a mixture of gases with constants different from those of
any single gas Note that equation (24) below could be used to find R in this
example
ANOTHER NOTE The Way to study examples is with pencil and paper. You
solve, or try to solve, the example before or after reading the solution

61. Relation between cr and c„ The relations derived below will be


repeatedly useful Starting with the definition dH — dU + d(pV)/J, we
substituted// = wcp dT,dL — wc,dT, nndp7 = wRT [or d(pV) = wRdT]
and find
wR dT
we, dT ~ wc, dT + J '

(24) or cp - j Btu/lb-°R

This equation says that the difference between the specific heats is equal to
the gas constant in Btu units (§ 50), not only true for ideal gases, but also
substantually true for real gases Now, using k = cP/cv or c„ =* kcv [sec
equation (14), p 40], we get from equation (24) kc» — c„ = R/J, or

= R
(25) c -
J{k 1)

Since = kcv , we find from equation (25)

KR
(20) -
J(k 1)

Recall that k varies with the temperature (Fig 19)


To get the relation between molar specific heats, we have Cr ~ Mctl
Cv = Mc„ and MR - R Thus, equation (24) yields

(27) Me, - Me. = or C, - C. =j- 1 986,

that is, the difference between these molar specific heats is equal to the
universal gas constant in Btu units, the same constant for all gases,

154
^78 = 1 986 Btu per mol-°R
V —
§ 62] ENTROPY CHANGE OF AN IDEAL GAS 65
(See § 56.) From equations (25) and (26), we have

(m) Cv =
Ti 1.986
and
, „ = Ml _ 1.986fc
- - C„
J(k 1) k 1 .nk - 1) k- r
62. Entropy Change of an Ideal Gas. In § 41, the energy equation was
set up in the form dQ = dU + p dV/J, where for

Steady flow, /p d = Al Vt -f AK + AP + IF,


Nonflow, Jp dV = W ft-lb.

If we let dQ = T dS and dU — wcv dT, we get an important form of the


energy equation;

dQ = T dS =
pdV
(28) wc, dT + [rDEAL gas]

Also using the enthalpy change of equation (23), § 59, and equation (17),
p. 45, we write

dQ = T dS = wcp dT —
V dp
(29) [ideal gas]

It will be worth while for the reader to practice deriving these two equations
from the various fundamental definitions which
have been given. By dividing through equations
(28) and (29) by T, we have two values of dS in
terms of point functions only. Since entropy is a
point function, <S 2 — <Si in Fig. 28 is a certain def-
inite number no matter whether the path followed
is l-d-2, or 1-5-2, or any other path, reversible or
irreversible.Equations to relate the entropies at
any two points are particularly desirable inasmuch
as entropy has been defined only for a reversible
process, AS = f dQ/T. Fig. 28. Entropy Change.
First, divide equation (28) byT; use
p/T = wR/V from pV = wRT (note that the results are restricted to ideal
gases), and get

J
,a
f dS = wJ
f
— +~j
Cv dT wR '
f dV

If the specific heat is constant, we integrate from 1 to 2 and get


'Ft Wl? Vo
(n) AS = S 2 - Si = wc r In jd H j- In pr Btu/°It.

[ideal gas, constant c r ]

While it is true that Q = JT dS and IF = Jp dFfor reversible processes only,


the foregoing integration [equation (n)J in terms of point functions is no longer
dependent upon the path, as are Q and IF. However, with points 1 and 2 as
66 THE IDEAL GAS [CA 4

shown m Tig 28, the first


term of equation (n) is the change of entropy along
a reversible constant \ olume line l-o (S. — S t = / dQ/T = uc,j dT/T),
and the second term is the change along a reversible constant temperature
hne a 2 (§ 69), S„ — St 4* St — = St — Si (Tig 28) In other word*,,
equation (n) and those that follow actually giv e the entropj change along
two reversible paths joining the points
Similarly from equation (29), after dividing bv T and using V/T = toR/p,
we hav e

[ I I)F vl cvs]

If the specific heat is constant, integration from 1 to 2 yields

(o) AS - S,- Si = ire, In £


It
-^
J
In 2* Etu/"R
pi
||DE.VI 0\S CONSTANT C,J

It mn> be useful to recall that

— log^ - +log and In V = log, V - 2 3 logm V

Using various known relations (pi = tvRT ep — c, = R/J etc ) man)


different forms of equations for AS can be obtained These variations will
be left for the reader to find os needed for example AS in terms of p and
U (no T) maj be found from equation (n) as

(p) AS = ire. In -f ire. In jr* Btu °R

JlDEVL CVS CONSTANT C, AND C,|

The equations of this article (§ 62) are pnmanl) for an ideal gas as
explained In most engineering situations, allowing for the \anation of
specific heats will result m
working answers for an) gas

63 Closure. The aanous relations evolved in this chapter appl) to


ideal gases only When )Ou think of Bojle, think T — C and pi =C,
when you think of Charles, think p = C and I /T = C and also \ — C and
p/T = C Appl) pV = wRT on\) to monatomic (He A) or diatomic ga«es
Other g3-s.es. are much too imperfect, except at Ion pressures Joule s law
(AU — trc. AT, and also AH = tee, AT) applies onl) to an ideal gas (or for
all practical purposes a nearly ideal gas)
: :

PROCESSES OF IDEAL GASES

64. Introduction. Knowing the characteristics of one working substance,


the ideal gas, we are in a position to study processes of such a substance,
after which we shall be able to investigate cycles in detail. Since graphical
representations are of inestimable value in the solution of problems, the
reader should always sketch the processes on the pV and TS planes.
Area “under” curve on pV plane, fp dV, represents the woik of a reversible nonflow
process,
Area “behind” curve on pV plane, —JV dp, represents A K + IF for a reversible
steady flow process,
Area “under” curve on TS plane, JT dS, represents heat of a reversible process.

It is worth while to know the energy equation in the forms

wc dT = wc„ dT + and wc dT = wcp dT —


[AFTER EQUATION’ ( 28)1 [AFTER EQUATION (29 )]

where the specific heat c is chosen in accord with the kind of process and the
substance.
For all ideal gas processes (constant specific heats)

Af7 = wc v (T2 - T,), and A H = wcp {T2 ~ T,).

For all ideal gas reversible processes (constant specific heats)

T, (except T = C.
S ivc n
|
c accor(j s with process]’

_ . /except T = C.
Q — wc{ 2 i)
|c acoorcj s with process
67
PROCESSES OF IDEAL GASES [Ck 5

For all reversible nonflow processes

ir = fpdv
Positn e values of A U, A II, and AS mean increases negativ e \ alues, decreases,

Positive Q is heat added, negntn c, heat rejected Positn e IP is work done


by, negative, work done on the system W ith some of the material of this
chapter, } ou are already familiar from preceding chapters, but it is necessary
to organize the knowledge with respect to processes

65 Constant Volume Process On the pV plane, this process, which


is also called on isomelrtc process, is represented by a \ertica! line 1-2
[Fig 29(a)] On the TS plane, a constant % olume line slopes upw ard toward
the right, the equation of the curve being obtained from f dS — uc.j dT/T
for constant c„ or
S - uc. In T + C,

where C is a constant of integration As show n in Tig 29, the gas is being


heated up, at least, energy is flowing into the system

V S
(a) V-C pV Plane (b) \ C, TS Plane

Fig 29 Reversible Isometric Process On the pV plane, there is no area under 1-2,
[pdV = 0

In the solution of problems, it is frequently necessary to recall the p, V, T


relations for the process at hand For F = C, we think immediately of
Charles’ law and have

E* _ Tt
(a) [I = C, CHARLES, LAW]
Pi Ti

An mtvnuflfy tb\ essvVAe eoastant \ oluwve protest vs owe \w vvhveh the rise m
temperature is due to heat only and

(b) Q = wc,(Tt - Ti)

Since dV = 0, the work ir = JpdF = 0


The change of entropy is

Ti
AS - ucr In
(c)
y
T

§ 66] EXAMPLE: CONSTANT VOLUME gg


From the energy equation, we observe that

(d) Q = AU + or Q = AU.

For a reversible steady flow isometric process (it could happen to an


incompressible liquid), we have the energy equation from equation (7A),
p. 31, in the form

Q = AU + AW, + AK + A P + W.
Since Q = AU, this equation yields

TF = - (ATI’/ + A K + AP).
Example Constant Volume. A 10-cu. ft. vessel of hydrogen at a pressure of
66. :

305 psia is stirred by paddles (Fig. 30) until the pressure becomes 40(1 psia. Deter-
mine (a) the change of internal energy, (b) the work, (c) the transferred heat, and
(d) the change of entropy of 1 lb. of the system. The system is the hydrogen in
the tank, which is perfectly insulated (adiabatic walls).
solution, (a) We know that AU = tcc v A for any kind of process of an ideal
gas, but we have no temperatures. Theiefore, eliminate tempera tuies with
pV = wRT and find

(2.4354) (10)
AU = ^(P ,- Pl ) (400 - 305)(144),
767

orAU = 434 Btu, where c„ and R have been taken from Table I and where we
remembered (144) that p must be in pounds per square foot.
(b) Although fp dV — 0 because dV = 0, the
work is not zero. We have repeatedly warned
that TF = fp dV only for a reversible process.
The kind of constant volume process that has been
discussed up to this point is one which is internally
reversible; that is, the process was brought about
by the addition or abstraction of heat. While
the heating process is irreversible externally, it

is which is added
reversible internally in that heat
can be made to flow out again (in both directions
Fig. 30. Irreversible Con-
as heat). However, the work flow into the system stant Volume Process. V= C;
cannot be reversed. We shall learn later how part
Q = 0.
of thisenergy can be got out as work but the flow —
of energy as work, dissipated as fluid friction, is irreversible.
Since the walls do not permit flow of heat, the nonflow energy equation becomes

Q = AU + TF = 0, or TF = -AU,
or TF = —434 Btu,where the minus sign says that the work is done on the system.
note. is to be used in a motor to cause work to flow into the system.
If electricity
Fig. 30 shows the hard way—driving a paddle wheel. The easy way would be to
install an electric resistance coil inside of the system and feed electricity directly
to it. Since electricity is 100 convertible into work, it can be classified as work TF
%
70 PROCESSES OF IDEAL GASES [CA 5

insofar as all our energj equations are concerned, but what lea\es the resistance
coil is heat Think about this until it is dear Evidently , the specified «tate
change in this problem can be brought about entirely bj heat entireh b\ work
or bv part heat and part work in anj proportions
(c) The heat Q = 0 bj statement of the problem
(d) 4Ve have defined specific heat as the amount of heat to be added to one
pound of substance to change its temperature 1°F But no heat is added in this
example However, since the entropv change of the '} «tcm is a function of state
properties onlj , we maj refer to equation (n), p 65 and its antecedents and get

a = Tt ,
R, ~ r,
“ .T,
A* ,
c. In + In c ' ,n
Y,
j
The term (R/J) In (tj/ti) drops out because it/n ~ 1 and In 1 = 0 U«mg Charles'
law, Tt/Ti ” Pi/Pi in this expression, we find

As - e. In & = 2 4354 In = 0 C59 Btu lb-°R,


Pi 30o

an mcrea«e within the system There are not enough data gi\ en to determine the
total ma«s and total entropj involved

67. Constant Pressure Process V constant pressure process, al'o


called an i sobaric process, is a change of «tate dunng which the pressure

remains constant It may be rev ersible or irreversible nonflow or flow


In anj cfl«e, for an ideal gas,
r
Tt
W I t

r,

1

i.
1
” T t
[p C CH WILES lAW)

On the pV
plane (Fig 31), the process is represented bv a horizontal line,
and on the TS plane the rev ersible process is represented b} the cune
obtained from the indefinite integral of dS = dQ/T, or

S = tre, In T + C,

where C is In the reversible process 1-2 (Fig


the constant of integration
31), heat is being added to the system and work is being done bj it If the

state point should have a leftward movement, as from 2 to 1 (Fig 31), heat
and w ork done on the s} stem would be indicated
is rejected (See § 64 )
Notice the similarity of appearance of the constant volume and con
stant pressure lines on theTS plane (Fig 31) The area under the constant
pressure hne must be larger than the area under the constant volume
fine for two particular temperatures 7'i and Tj, because cp is larger than
c, ( dQ = c dT) Thus, to accord with this fact, it is essential for the con
stant pressure hne to be less steep than the constant v olume hne within a
particular temperature range
(a) jp dV. For constant pressure,

(f) fpdV — pjdV = p{Vi — FJ - WRIT, - T ,) ft-lb,

Ia\T fluip] [iPEVL GV*1


§ G7] CONSTANT PRESSURE PROCESS 71

where the lastterm is obtained by use of ;!'


7
= will' and applies to an ideal
gas only. Thus, for a nonflow process,

IT = fpdV = p(F ; - I',) ft-lb.

For a steady flow process, equation (k), § 41 ,


shows that (with A P = 0)

= — = All + AK +
jp (I \
p( l ; I i) / II

= ;>(!’: - Vi) + A K + IF.

Since the terms p(F ; — 1 ',) cancel each other, the shaft work, if any, is
IF = —A A." for any fluid in steady flow at constant pressure. The same
conclusion is arrived at for dp = 0 in Jv dp in equation (p), §42; to wit,
IF = -A K.

v s
(a) p-C. pV Plane (b) />=C, TS Plane

Fig. 31.Reversible Isobaric Process. Observe that the shaded area of figure (a)
is equal to TT = p(V. — V,), the work for a nonflow process, which is the same as the
result obtained from //> dV.

(b) Heal. Using a constant c p in dQ = iccr dT, we find for an internally


reversible process,

(g) Q = icc p JdT = icc r (T? — Ti) Btu.

The cnergj- for a change of state from 7\ to Tz could be supplied as work


(see paddle, § GO), in which case the process is irreversible. In any reversi-
ble process, flow or nonflow, Q — AU + / (p dV)/J and for p ;
— C, this is

(h) Q = U~ — Ui + pzV? — piVi = Hz — Hi,


[any fluid, FLOW OR NONFLOW, 7> = C]

which shows that the heat transferred in any reversible constant pressure
process is A//, which for an ideal gas is A H = wcp AT = Q as in equation
(g). If the process is stead}' flow (with AP = 0),

Q - AH + AK + IF.
[any fluid, any process]

If this flowoccurs with little or no change in kinetic energy (AK = 0) and


with no shaft work (IF = 0), as it does in all kinds of heat exchangers, then,
72 PROCESSES OF IDEAL GASES [Ch S
even with the pressure varying,

(i) Q = AH Q = vx,{Tt - Tt )
[any fluid, STEADY FLOW PROCESS [ideal GAS, steady flow,
ak = ap = ir = o] ak = ap = w = o]

This equation holds whether the process is reversible or irreversible when the
stated conditions are met See equation (h) again. The energy equation
for p =* C is (<2 = AU + Jp dV/J)

UK, AT = wc, AT + Btu


\CP AND Ca CONSTANT, IDEAL OAS]

68 Example: Constant Pressure. Anidealgas, for which R = 3S6 and k = 1 659,


undergoes a reversible isobanc process during w hich 500 Btu are added to 5 lb of
gas The initial temperature is I00 a F. Determine (a) tt, (b) All, (c) AS, (d) AV,
and (e) W
for a nonflow process.
solution The specific heats which will be needed, are found from equations
(25) and (26) (The gas is not stated )

R 3S6
c* “ - " 0 754 Btu/Ib-°R
J(k 1) 778(1 659 - 1)
kR (1 G59)(3SG)
cp “ J(k - ” 25 Btu/lb-°R
1) 778(1 659 - 1)
1

(a) The heat is 500 — uc,(T t — 2\), or

or fj = 180°F
(b) Since AH = uc,AT then AH = 500 Btu [This is also the heat in any
steady flow process, under certain conditions See equation (l) ]
(c) The change of entropy is

T, 640
AS = wc. In
y= (5)(1 25) In
^- 0 83 Btu/°R

(d) The change in internal energy is

AU = wc,(Tt - Ti) = (5)(0 754RG40 - 560) = 301 G Btu

(e) The work may be found from IF = p(t>t — Vi) * uR{T, — Ti) ft-lb ,
but it is
easily found from the nonflow energy equation

IF = Q — AU = 500 - 301 6 = 199 4 Btu

69 Isothermal Process. An isothermal process is one earned out at


constant temperature, T = C Unless stated otherwise, we shall mean a
reversible isothermal process whenever the term is used Since T = C, we
think immediately of Boyle’s law
pV ~ C, or PjFi = piYi =C [T = C, boyle]
[any REVERSIBLE OR IRREVERSIBLE rEOCESS]

The pV and TS curves of this process are shown in Fig 32, where, on both
planes, the
§ 69} ISOTHERMAL PROCESS 73
State point has a component motion toward the right;
Heat is added, work is done by the system.
If the process is run in the opposite direction, the

State point has a component motion toward the left;


Heat is rejected, work is done on the system.
One can conceive of a process for which the state point has a component
motion on one plane in the opposite direction to the motion on the other

plane but not for the isothermal process.

(a) T — C, pV Plane (b) T = C, TS Plane

Fig. 32. Observe that the differential area in (a),


Reversible Isothermal Process.
P dV, is a differential work quantity dW in a nonflow system. When the volume
increases, as 1 to 2, the process is called an expansion. When the volume decreases, it
is said to he a compression.

Since U and II are functions of temperature only for an ideal gas, and
since T ~ C, it follows that U= C and H — C; or AC7 =
0 and AH = 0.
Thus, for an ideal gas, we could (but we will not) call this process a constant-
internal-energy process or a constant-enthalpy process.
(a) Jp dV. This integration has been made (§ 18) using p = C/V.

J
ydV = C
f^
= ViVr In
^ = wRT g In ft-lb.,

where, in the last form, we have used p\Vi — wRT and F 2/Fi = pi/p 2 .

For the nonflow process, then

(j) IF = piFi In = gj-' ln Btu '

Yj.
[reversible nonflow, ideal gas]

For the steady flow process at constant temperature (A P = 0) we may


use equation (p), § 42, and obtain

p 1 F 1 ln = AK +
— j V dp = IF ft-lb.,

from which one may if the other terms are known.


solve for IF The states
1 and 2 are at the boundaries of the system where the substance enters and
leaves.
U PROCESSES OP IDEAL GASES [Cft 5

(6) Heat Since heat is transferred


but the temperature does not
change, the specific heat of the process is infinite, c = «. . With df = 0

Fig 32 Repeated

we hue Q — cdT — («)(0), an indeterminate Therefore, to find Q,


w e must u«e the energj equation

Q = iU + J pdV or Q - I.H -j 1 ip
J j
PiTi, Ft wRT pi
flO Q-jjpiv-
r\ f
Ej-'T,'—
1 ,

'"]r,
T,
Bu ,

[ANY REVERSIBLE ISOTHERMAL, PLOW OR NON PLOW IDEAL OAS]

This equation applies as well to a steady flow isothermal as to a nonflow, as


you realize when you fp recall that dV is equal
the mechani- to the sum of all
cal energj terms in the steadj flow equation given in § 41 Equation (k)
says that the w ork of a nonflow isothermal process is exact!} compensated
bj heat — without increase or decrease of internal energj
The change of entropy is

(1) iS- lap|Btu/“R

Compare with the rectangular area under 1 2 on the TS plane (fig 32)


70 Adiabatic Process Isentropic Process An adiabatic process ts one
tn uhick no heat is transferred, Q = 0 An isentropic process ts defined as
one of constant entropy In a reversible adiabatic Q = 0 as in anj adiabatic,
but also T dS = 0, since Q = T dS m
a reversible process We know that
T is not zero hence, dS = 0 In this book, the w ord isentropic shall mean a
*
reversible adiabatic
In any nonflow adiabatic, reversible or irreversible anj substance,
Q = AU -f IF = 0, or
(30) w= —&U = Ui-Ut or ir = - T)
2 Btu
[ant fluid! [ideal gasJ

taken to mean only constant entropy and an irreversible


* Occasionally, isentropic is
constant entropy process which is not adiabatic is easilj conceded though it is most
unlikely to be of any use
§ 70\ ADIABATIC P2.0CESS ISEHTROPIC P3.0CBSS —
-C StScdv fioic cdiccctic revt-WoIe or irreversible. suv fluid (with
Q = 0 and AP = 0 j
(31 1 TT = — AH-XKi.
and if AET happens to be negligible, as it often is.

(32 ) ft = —
-A H = Hi
-an n -— t H,
ti-_ or IT = irc.(Ti — T-_) Btu.
[a-VT FLUID. ±E = 0 [ideal GAp

If IT = 0. as in a nozzle. A K = —AH'. or

KTTT ITTTx
(33) — Hi — Hi = wc^Ti — T~ I Btu.
2qj 2c r /
[j. N'T FLUTD. W = 0. (IDEAL GAS.

'We see that thework done in any nonfiow adiabatic [equation (30)1 is at the
expense of the internal energy. In any adiabatic steady flow [equation (32) J.
the work is usuaEy largely at the exoense o: a decrease in enthalpy.
(c) Relation between p, V, end T. If ire wish to obtain the relation
between p and V. we use the energy equation and pi" = irRT. eliminating
the temperature. For a reversible process.

(m)

l dts
, _
= ~
0 = arc. di — p dV
,

J
Differentiate pr = RT and get
p dT -r T* dp = vrR dT,
or
dV T dp
dF _ p -r
mi?

Put this value of dT into equation (mi and 2nd

p dT c-p dT -r c-T dp
J R
Multiplying both sides of this equation by R and dividing by pT. we have

_ RdT _ dT c- Cp
J T~^T or —
Since R J -f- c. = c~ 61). and since e-
(.§
c- — k. we have

k
F
= —
P
IfI- 15 taken as a constant (say. the mean value i. we have now arrived at
an integrable form.* Thus, between two states I and 2.
-r- it-

5‘rtr § i i for 2 treatment of the isen tropic process t^i-h varo


.
Ec heats.
76 PROCESSES OF IDEAL GASES [Ck 6

or

_Mn|!_l„(£) = ln»
Taking the antilogarithm of the last two expressions, we find

(a) Qr) = or pxV? = piVf

or, since states 1 and 2 were chosen at random along any isentropic line,

(34) pV* = C,

the equation of an isentropic process for an ideal gas in pV coordinates


The corresponding cur\e is similar in general appearance to the equilateral

(a) 5 - C, pV Plane (b) S = C, TS Plane

Fig 33 Isentropic Process For a comparison of the isothermal and the isentropic
curves, Isothermals 1 -<j and b-1 are shown dotted Point a is related to 2 in being
on the constant volume line through 2 Points 1 and 6 are also on a constant volume
line Observe that the isentropic line on the pV plane is steeper than the isothermal,
also that there is no area under the curve 1-2 on the TS plane, which is in accordance
with the definition that 0 = 0 and S “ C

hyperbola [Tig 33(a)] With equation (n), we can derive the relation
between T and V, and the relation between T and p Trom the equation of
state,piVi/Ti = piVi/Tt we get

Pi _ Vi Tt
Pi ViTi

which, substituted into equation (n), gives

Now, from equation (n),


§ 70] ADIABATIC PROCESS—ISENTROPIC PROCESS 77
Substituting this value of Fi /T% in equation (35), we find

Equations (35) and (36) will be useful in problem work and in deriving
special forms of other equations. All the foregoing p, V, T relations apply
to an ideal gas.
(&) Work. We have found work from the energy equations (30), (31),
and (32), all of which are very useful, but we must also learn how' to obtain
the work from the pV plane —area under the curve is nonflow work. Since
pVl = C [Fig. 33(a)], we use p = C/V h in p dV and get

Setting the first C = pTs* and the second C = p\Vfl, we find

pi V. »V »V '
ft-Ib. - Btu,
(37)
f l :
[whenever pT 71 = C
]

which is, then, the work TF


a nonflow isentropic process. This integral
of
may be converted to temperatures by using pV = wRT and it can be shown
to be identical to equation (30) when both are in the same units.
In a steady-flow isentropic (A P = 0) with pVL — C, we get from equation
(P), §42,

(P) ~ f V dp = —C
= ^-p VA = &K +
k (P* v
TF.

The elementary area V dp is shown in Fig.


34; let C = p l/k V or V = Cp uk . If the V
change of kinetic energy is negligible f;g 34
. isentropic Process—
,

(A K= 0), equation (p) gives


Steady Flow.

(38) IF = - V dp = ^V^-PiV,) ft lb _
J steady flow, AX' =
[reversible 0]

If TF = 0, we get AK = —fV dp. Summarizing, we have

(q) TF = -JFdp and AK = —JT dp.


[steady flow, A K = 0] [steady flow, W = 0]

The negative sign for the JT dp is there because in integrating with respect
to pressure from 1 to 2, the direction of integration is downward (negative)
for an expansion and the fV dp turns out to be a negative number. But
TF done by the system, as in Fig. 34, is here taken positive; hence, we use
78 PROCESSES OF IDEAL GASES [Ch 5

— JP dp to accord with the convention of signs already defined The


student should work through the details of all integrations made in this
text

71. Irreversible Adiabatic with Work.* If, during an adiabatic process

in a machine such as a turbine, there is fluid friction and turbulence, the


work done is less than that during a corresponding lsentropic expansion
The process is internally irreversible because of frictional effects The end

B D

(a) (b)

Fig 35

states of the adiabatic with work arc represented by 1 D (Tig 35) where D
maj be any where between B and 2 The less the loss from irreversibility,
the closer 1 -D approaches the reversible 1-B
In a general way, we shall use pnme marks to indicate actual values and
actual points, for example ll = actual work done (the work of the fluid
r


unless otherwise indicated that work which can be converted into shaft
work the absence of mechanical friction) MI - actual change of
m
enthalpy, 2 = the actual end point when the corresponding ideal end point

s s
(a) Ex pi nsion (b) Compression

Fig 36 Irreversible Adiabatic Frocesses

is 2Since the energy equations in general form apply to any kind of proc-
ess, we already have useful forms in § 70 However, rewriting equations
(30), (32), and (33) in the pnme notation, which indicates that
some
irreversibility effects are involved and referring to Fig 36, we get

shorten the course one may touch lightly on irreversible processes


* If it is desired to
and omit § 71, also omit 77 and 78 and other articles on irreversible processes
; ;

§ 71] IRREVERSIBLE ADIABATIC WITH WORK 79


(30A) IF' = U, - Uv IF' = v)c v {T\ - 7V);
[any nonflow system, Q = 0] [ideal gas]

(32A) W = Hr- Hr IF' = wcp (Tr - TV)


[steady flow, any fluid, [ideal gas]
<3=0, AK = 0]

(3 3 A) AK' = '-gj - = ffi - A/V = wcp (Ti - TV).


Dj
|STEADY FLOW, ANY FLUID, [IDEAL GAS]
<3=0, IF = 0]

Irreversible adiabatic processes always occur with increasing entropy.


In Fig. 36(a), an ideal adiabatic expansion from pi to p 2 ends at 2, but the
actual expansion ends to the right at 2'. Likewise, the isentropic com-
pression [Fig. 36(b)] ends at 2; the actual expansion at 2' at greater entropy
than 2.
(a) Relations between p, V, and T. Actual expansions occur which
accord reasonably with an equation of the form pV m = C. If an actual
exponent m
can be determined and if the fluid is an ideal gas so that we may
use pV/T = C, we get

[ideal gas]

after equations (35) and (36). These relations hold for any irreversible
change of state to which pVm = C applies, whether it is adiabatic or not,
but the fluid must be an ideal gas, pv = RT.
( b ) Change of Entropy. The change of entropy may be evaluated along
any reversible path connecting the points. In Fig. 35, AS from 1 to 2 is
equal to AS from B to 2, in this case along a constant pressure line. The
equations of § 62 may be used to compute the change of entropy, as con-
venient; for example, equation (n)

ac — wcv i
=- . wR . V2
AS In -[ =- In j T -

i 1 J VI

Using F 2/Fi = in this equation, we get


~m)
Ti.wR. (T /(1
AC = wc v
AS
.
In y- + -j- In bp-JY t

,
T^ wc v {k - 1) T2
wc p In
,
In
T ^+ _m ,
1 T
(k — m\ T2
AS — wc
.
In
\1 - m
v I

T i

which gives the change of entropy in terms of temperatures, often a con-


80 PROCESSES OF IDEAL GASES [Ch S

\ emcnce * Remember that in irreversible processes, Jp dV is not work and


STds is not heat. Irreversible adiabatics are closely approached in some
centrifugalpumps and blowers, in steam and gas turbines, and in recipro-
cating compressors and engines
72 Example. There are compressed 3500 cfm of air from p, - 14 5 psia and
t> = 75°F to p a = 29 psia Determine the work for each of the following processes
(a) nonflow lsentropic (b) nonflow irieversible adiabatic to a temperature of 190°F
(650°R), (c) steady flow I'entropic for which u, = 40 fps and u, = 120 fps (d)
steady flow adiabatic to a temperature of 190°F, velocities as in (c) (e) If the
irreversible processes accord with the relation p\ * = C, what is the value of ml
(f) What is the change of entropy 7

solotiov The temperature after i«entropic compression is Fig 36(b),

The amount of air involved is

PiFi (14 5)(144)(3500)


ic 250 Ib/min
iWi (5SS)(63S)

(a) Tor a nonflaw isentropic

IF - —AU = UC,(7\ - T) = t (256) (0 1715) (535 - C25) -3950 Btu/min


This w ork is for the case w here all the air is in a c> lmdei and the piston compresses it
It docs not include work that is necessary to get the air into and out of the evlinder

s s
(a) Expansion (b) Compression
Fig 36 Repeated

(b) For the nonflow lrieversible adiabatic, Fig 36(b)

TF = uc,(T: — T, ) = (256) (0 1715)(535 - 650) = -5050 Btu/min


(c) For a steady flow piocess 0 = 0 and s = C,

ir- -AK-tji - + *>-*.)


40 s 120* I
' + ° 24(535 ~ 625 = _5G0° Btu /min
[ 5gJ > J

This work includes that necessary to get the air into and out of the cylinder

* We might note the similarity of cj(fc m)/(l »«)] — — m equation (r), which is not a
specific heat, to the polytropie specific heat equation cj(t — n)/(l — »)], § 73(b) The
polytropic is defined as a reversible process and 0 Q
§ 73] POLYTROPIC PROCESS 81

(cl) For the irreversible steady flow adiabatic, -AH' = wcp {Tt. - 7\_>) and

r= 256 [~j - + 0.24(535 - 650)] = -7130 Btu/min.

(e) To find m, \\ e have

Tz (th _ 650 / 25 \ («-')/»


_
~ \pj ~ __
"
Ti 535 \14d5/

Using logarithms (or log log slide rule),

m — 1
_
~ log (650/535)
m log (25/14.5)’
°r m— 1.553.

(f) To find the entropy change of the system, use one of the basic equations of

§ 62 or use the derived form, equation (r), p. 79, vc hich is

AS wc v In - (2o6)(0.171o)
_ In - - 2.35 Btu/°R.
__ m) Ti ( j j 553 ) 35

73. Polytropic Process. A polytropic process, Fig. 37, is an internally


reversible process, which conforms to the relation

(39) pV n = C or pi TV = p*TV, [any fluid]

where n is any constant. Since many actual processes follow closely the
relation in (39), this discussion is especially important. Because p7 n = C

V S
(a) (b)

Fig. 37. Reversible Polytropic Process.These curves show the general appearance
n is between 1 and k. Since, in theory, the
of a polytropic process where the value of
polytropic exponent may have any value from n = — 03 to n = curves sloping in
any general direction may be obtained (see Fig. 39). In practice, the value of ti does
not vary greatly from k.

is mathematically the same as p7* = C, all the mathematic consequences of


§ 70 apply here when k = n. Thus, the relation between T and 7 is

(40)
^ = (p) = (jp [ideal gas]

Between T and p, we have


rp / \ (n—D/n
IDEAL GAS J
(41) jr = (“J t
82 PROCESSES OF IDEAL GASES [Ck 5
If a process is known to be polytropic, and
if two state points are defined

(say, pressures and volumes are known), the \alue of n can be found by use
of logarithms Writing equation (39) as p t /p 2 = (Vs/Fi)* and taking
logarithms, we find

(s) _ log (Pi/Pi )


log (V t/\\)
Equations (40) and (41) can be handled similarly
(a) fp dV and fV dp. Corresponding to equation (37), we have

(42) W- j pdV -
[NOVrLOW, ANT rLLID]
7
-
” r* ~
^ [IDEAL CAsJ
H lb I?''' " C1

Also by inspection of equation (38), wc can write

w- v dp - "feD
~* v,) [pv-.q
(t)
- J ft-ib

(steady now, AA « aP = 01

Equation (t) is also the value of AK when 11' = 0 and A7 = 0J

Air, + AK + ir - Jp d 1 and AA + H' = -JV dp


| STEADY FLOW At* - 0|

(b) Heat I or any reversible process, nonflow or steady flow,

dQ -dU+^
Tor the integral of p dV, use the value in equation (42) and use

A (7 = wc t (Tt - T t) = wc, AT
Then

Q-wc.iT,- T,) + “» (—T~tr~) - r ‘)

Using c„ + Ii/J = cp and k — c p /c„ we find

Q - w - T,) - wc. - T) Btu


(43)
(j^£) fr. t

Now suppose we the specific heat of a polytropic process, that is,


call c*
the amount
it is of heat added or abstracted during the process to change
one pound by 1°F The heat then becomes Q = wCn(Tt — Ti) Com-
paring this equation with equation (43), we can write

(44) c- = c
-(r^)’
the polytropic specific heat The \ alue of c» maj v arj infinitely, inasmuch
as n may vary infinitely Observ e also that c„ is negative w hen l >n> 1
§ 74] EXAMPLE: POLYTROPIC PROCESS 83

A negative specific heat means that heat is rejected by the substance even
though the temperature increases, or that heat is added even though the
temperature decreases. A process of heat rejection with simultaneous
temperature increase actual!}' occurs in the usual air compressor. The work
done on the air raises the temperature of the air (stores some internal
energy), yet at the same time, in order to obtain a lower final temperature
at the end of compression, cooling water is circulated around the cylinder to
cam- some heat away. The area under the curve on the TS plane [Fig.
37(b) or 38(b)] represents Q in Btu units.
The change of entropy of a reversible polytropic process is

(u) c =
AS
I
/
dQ
~Y
= w f 1 — dT
cn
In
Ti
[ cr.
constant]

Notice, for example, that by using the values of T«/T i given by equa-
tions (40) and (41) in equation (u), AS may be expressed in terms of other
properties.

74. Example : Polytropic Process. Five pounds of an ideal gas are compressed
polytropically from a pressure of 15 psia and 40°F to 85.3 psig. The gas constant is
R — 50, c v and the change of state is according to pV 5 = C. Find (a) the
— 0.25, l

initial volume, (b) the final temperature, (c) the work for nonflow and steady flow'
processes, (d) the change of internal energy, (e) the transferred heat, (f) the change
of entropy.

Fig. 38.

solution', (a) See Fig. 38. The student should cultivate the habit of making
sketches for every problem. From the characteristic equation

V1 = 'EMi = (5) (50) (40 + 4 60 = ) -- fl cu


Pi (144)(lo)

(b) From equation (41), Tt = T (pi/ViYn ~ l)/n Assume standard


l - atmospheric
pressure of 14.7 psia. Then p* = 85.3 + 14.7 = 100 psia, and

T, - (500) (500) (1.549) = 774°R = 314°F.


(f>)'
84 PROCESSES OF IDEAL GASES [Ch 5
(c) The nonflow work is W= (p,P, -p l Fj)/(l - n) The volume I, must
be calculated or eliminated Usmg pi — uRT we have

it... - "«<?;_-*> - ^wxm-aio) _ _ 22s m tHb _ _ 293 Bla

represented by the area a 2 b (Fig 38) in foot-pound units


1 The negative sign
indicates work done on the substance The work of a stead) flow pioeess in which

AK «= 0 and AP = 0 is « times tl e nonflow work Compare equations (42) and (t)

Therefore
If (steady flow) (1 3)(-293) * -381 Btu
rerepresented in foot-pound units b> area c 1 2-d (Fig 38)
(d) To find c, use c, — c, R/J or

-
c, = 0 25 - ^= 0 25 00043 = 0 1857 Dtu/lb-'R

Then
A 17 - uc,(Tt - Ti) - (5)(0 1857) (774 - 500) +254 Btu
The positive sign indicates an increase in internal energy
(e) The value of k = c,/e, 0 25/0 1857 = 1 346 Tie polytropic specific

heat is

c*

Q =
= e.

uc.(Tt
(*
~ -
)
Ti)
=
-
0 1857

(5)(— 0 02S45)(774
= -0 02845
- 500) = -39 Btu

The negative sign indicates heat rejected by the substance As a check on this

calculation use ti e nonflow energy equation

Q AU + B = +254 - 293 = -39 Btu (check)

The area under the curve 1 2 on the TS plane (Fig 38) represents Q Figure 38
is not drawn to scale
(f) The change of entropy is

AS = wcK in ~
2 1
(5}(— 0 02845) In ~
7
ouu
-0 0624 Btu/°R.

The negative sign indicates a deciease of entropy


§ 75] EFFECT OF VARYING n IN POLYTROPIC EQUATIONS 85

Varying n in Polytropic Equations. Polytropic processes


75. Effect of
are inclusive in that the equations for the change of state of all the fore-
all

going reversible processes can be obtained from pV" = C by choosing proper


values of n.

p — C; n =0; then pV° = C or p = C, a constant pressure process. This value


of n — 0 substituted in polytropic equations (39), (40), (42), (43), and
(44) will
give the correct forms for the constant pressure process.
V= C; n — «. Then from pV" = C, we have p 1/n V = C, or

pi/* y _ oy _ Qi
p
or V= C”, a constant volume process. This value of n = « substituted in the
polytropic equations (41), (42), (43), and (44) will yield the correct forms for the
constant volume process.
Let n = k, then pV t = C, which is recognized as the equation for an isentropic
process. This value of n = k substituted wherever n appears yields the correct
forms for the isentropic process.
Let n = l, then pV = C, which is recognized as Boyle’s law and the equation of

an isothermal process. This value of n = 1 yields the correct result in equations (39)
and (41). However, when n = 1 is substituted in equations (42), (43), or (44), the
resulting expressions are indeterminate.

Fig. 39. Effect of Varying n. Expansions or compressions are imagined to take


place from point 1. Notice that all positive values of n give curves in the
some common
second and fourth quadrants on the pV plane (a); that positive values of n may produce
curves in all four quadrants on the TS plane (b). Notice too that curves with values
of n between 1 and ft will fall in the second and fourth quadrants on the TS plane and
within a very narrow region on the pV plane.

important to be able to sketch various curves of thermodynamic


It is
processes on the pV and TS planes, and sketch them about as they would
appear if they were actually plotted to scale. Figure 39 will be of assistance
in learning to make these sketches and it shows also the effect of varying n.
Remember that, although they appear much alike, the isentropic curve on
86 PROCESSES OF IDEAL GASES [Ch 5
the pV plane than the isothermal curve, and that, on the TS
is steeper
plane, the constant volume curve is steeper than the constant pressure curve
when both are drawn between the same temperature limits
76. Throttling Process. A throttling process is taken as an adiabatic
flow process (Q = 0) from one region to another at lower pressure, during
which no w ork is done It is a free expansion, as in opening the valve on a
compressed air tank, or the household faucet It is usually analyzed for a
period during which the flow is stead} If A P = 0, as well as IF = 0 and

Q = 0, the steady flow energy equation becomes

lit + A', = lit +K t

Possibly somewhere m the stage of a throttling expansion, the kinetic energy


K might be relatn cly large In practical application howe\er, the end
conditions for the substance are virtual)} stntionar} or Ki *s Kt In cither
event,

(46) lit = //, or uc p (Ti - Ti) = 0,


[ant fluid] [ideal gas]

represent defining conditions for a throttling process If the substance is

an ideal gas, T\ T\ as well as Il x = lit, but not otherwise


= (See foot-
note, p 02 )
T = C, piFi = p t Fi for an ideal gas (Fig 35) Imper
Since
feet gases generall} undergo a temperature decrease as a result of throttling,
corresponding to some end conditions such as 1-C [rig 35(b)]

v s
(a) (b)

Fig 35 Repeated (1-2 = Throttling Process)

In Tig 35, the curve 1 -B represents a reversible adiabatic (S = C),


during which the maximum work is obtained from the energy which the
substance had in state 1 The throttling adiabatic 1-2, which produces no
work, is shown dotted to remind us that the areas under the curves no
longer represent work and heat, that is, the points along the curve are not
a sequence of equilibrium points and the dotted curve 1-2 does not neces-
sarily represent the path of the state point The expansion ts irreversible,
and unorganized for the production of work and the conservation of available
energy
The change of entropy, which will always be an increase, can be computed
\

§ 7.9] EXAMPLE S7

from any one of (lie equations of §02 (bo sure to refer Imck in order to
become oriented), say, equation (n) which reduces to

(v) A.S - w f
in ]'-? Btu/°li,
j

because 'J\ — T« and In = 0. Conq)arc equation (v)


1 wit.li the entropy
change of a reversible isothermal process [equation (1), § GO].

77. Iscntropic Process with Variable Specific Heats.’


For the case of 1'

constant specific heat, wc found the relation between p and V of an isen-


tropic process (ideal gas) to be p n «= C [e(|uation (34)]. If the specific
heats vary, k also varies; so, to solve the iscntropic process under this
condition,we must start again with fundamental relations. If pv «= RT,
wo have dU — wcr d'V; and the energy equation dQ «= dll -|- p dV/J
becomes wc,. dT — —pdV/J. Substituting p — wJIT/V into this last
equation and separating the variables, we find for w = lb. I

ii i r 2 R
(w) ~ 7 1,1
Vi ./

To integrate the left-hand side of this equation, substitute the proper value
of r„ as obtained from Tabic If or elsewhere. There is often a convenience
in using mol values, Cv - Me, and R ~ MR, for which equation (w)
becomes, in Blu per mol,

fn dT ~ R fdV Vt _ Vi
(x) -1.080 In 1.080 In -

J T J J V fT F,’

where the universal gas constant in Htu units is Vf/./ = J /Sd fi/778 — 1.080
(p. 00). After the left-hand side of either equation (w) or (x) has been
integrated, one may solve for the temperature IT?, the usual unknown, by
trial and error, as illustrated in the following example.

78. Example. The molar specific heal of a mixlmo of gases in an internal com
bastion engine lias been found to be given by (see Chapter J 2, § 202)
4
3.107' 3^8 7' 2 2027.5 (80.2) (JO ) 37.1
Cv 7.500 + Kb 10 " ]<
**"
.

J'2
'
J’ 1/2
I3tu/mol-°R.

The eompression ratio of a nonflow isentiopic eompicssion in this engine is (J; that is,
V2 F
/0 and the initial teinpeiatme is
«= t
=> 1d0' I'’. Compute the tompoiatuio ti ,

after isentiopic eompicssion.


solution. The value of the riglil-lmmi side of equation (x) is

(y) 1 .080 In «= 1 .080 In 0 - 3.550.


I 2

* In sliort courses, if may be desired to oinil part or all of the remainder of this chapter,
except perhaps § 70.
§ 79] CLOSURE 89

value. Then, a range is defined between which the answer should be found. After
several guesses, u e find for T? = 1180°R that

fl>80 dT
3.559,
goo T
which is the desired value; therefore t~ = 1180 — 480 = 720°F.

79. Closure. Thermodynamics is exacting but fascinating. The begin-


ner prone to depend unduly on memory for details with which to solve
is

problems, whereas it is easier and more profitable to classify the knowledge


in such a way as to see the repetitive utility of the more basic laws. For
example, in any nonflow process, the law of conservation of energy is
expressed by
dlF = dQ - dU,

where dt/ = wc t dTwhcnjw = RT;ii the process is reversible, dW = pdV/J


and dQ = wc dT, where c depends upon the kind of process.
In the steady flow process (A P = 0),

dW = dQ-dH- A K,
where dll = wc p dT when pv = RT dQ is as stated for the nonflow process,
;

and dW — — J"F dp/J — dQ — dll when dK — 0. The tabulation of


equations in Table III is for the purpose only of highlighting many similari-
ties and emphasizing differences.
Recall that energy is a scalar quantity, so that should there be two or more
successive processes, the total work will be the algebraic sum of the works
for the individual processes, and the heat transferred will be the algebraic
sum of the heats of the processes. This brings us to cycle analysis, Chap-
ter 6.
92 CYCLE ANALYSIS AND THE SECOND LAW [Ch 6

that without subscript, 7T applied to a cycle is understood to mean the


is,

net work If it is negative, it is the net input w orh


Equation (4G) embodies the classical definition of the first law (§ 24),
namely that work and heat are mutually com ertible Arming at equation
(46) as we have, we see the necessary conditions in order for the first law,
as stated in this brief form, to hold If there ts no change in stored energy
during cyclic operation of a system, which may be flow or nonflow, the net
amount of energy crossing the boundary as heat is equal to the net amount of
energy crossing the boundary as work
Note in passing that when the working substance is taken as the system,
the net w ork does not lm oh e the losses from mechanical friction in bearings,
etc It is the fluid work Tor a power generating machine, the actual shaft
work is less than If, and for a power-consuming machine, the actual shaft
work is greater than IF, the difference is a loss due to mechanical and fluid

friction
We may and do say
IF - 2Q and IF -Q -
A Q R,
[algebraic sum] 1 arithmetic difference!

meaning the same thing m each case If you use the \ anous equations for
process heat, as deriv ed in this book, the result will automatical!) be positive
when heat is added and negative when heat is rejected When we write
IF *=2Q, we mean an algebraic sum where rejected heats ha\e negatne
signs If we write the rejected heat as a positi\ e number then it must be
subtracted from the heats added, IF — Qa ~ Qk There is no difference,
simply different wajs of saying the same thing Since both viewpoints
are handy the reader should master both

Thermal Efficiency Efficiency may well be thought of in its sim-


plest form, output divided bv input The output of a power cycle, which is
a thermodynamic cycle for the production of power is the net work, the
input is the heat added to the working substance from an external source of
heat Qa in Tig 41 The net
work IF,.* — IT m is taken as the
output of the power cy cle because
some of the gross w ork IT out is used
to supply the energy N „ to e dm
the pump Therefore, the ther-
mal efficiency of a power cycle is

20
Qi
Ftg 41 Power Cycle Qa > Qa Observe that equation (47) applies
only to a cycle However, if the cycle is performed wholly within the
engine, as we shall imagine for ideal internal combustion engines, then
equa
\ion (47) gives the efficiency of the engine as well as that of the cycle
) . 1

§ 85] CARNOT CYCLE WITH IDEAL GAS 95

of equation (48) that for the particular temperature limits of operation, it is


the highest conceivable thermal efficiency. Equation (48) is the efficiency
of the Carnot and other reversible cycles only. Most of the cycles which we
shall study are not externally reversible ones.

85. Carnot Cycle with Ideal Gas as the Working Substance. While the
rectangle represents the Carnot cycle on the TS plane, the cycle cannot
be shown on the pV plane without a knowledge of the properties and
the equation of state of the substance. We are familiar with the ideal gas
whose equation of state is pV = wRT. The pF equation of the isothermal
process is pV — C (§ 69); for the isentropic process, it is pVk = C (§ 70).
With known constants, these equations may be plotted to get the cycle.
(See Fig. 44.) It will be instructive to analyze the Carnot cycle for an ideal
gas, because themethod of attack is the same for all cycles.

(a) Work from the TS Plane. By this ex-


pression, we mean to find an equation for work
as S Q, that is in terms of heats. For the
isentropics, Q = 0; for the isothermals,
Q - (wRT/J § Applying this
In (F./Fj), 69.
basic form to the process ah (Figs. 42 and 44)
and to the process cd, we have

n
Qa = wRTij
~J~
F(,

Va
n
Q* =
toRT,
__
Vi
ln _,
.

wRT
= XQ = wRT i Ft - . Vi
y+
,
IF In 1,1
J ./ F Fig. 44. Carnot Cycle.
, , Tr , wRT i 1 wRT Ideal Gas. Not to scale.
--Inrr- See Fig. 4G.
I i

From equation (35) for an isentropic change of state,

rr, \ l/tt-t) !/(*-!>


F_c _ _ (Tt and > - (£)
Ft " \rj \T,0
whence
Fe Ft '
(f)
Vi F
Substituting this value of Ve /Vd in equation (e), we find

l
(g) If (T i
- Te) -y In y Btu,
*/ * a

which is represented by the enclosed area on the TS plane. The thermal


efficiency is

IF _ (T, - T-)(y}R/J) lnJIVU _ Ti — Tt


c ' (iL'Rf.I)Ti In (F»/F.) Ti

which, as it should be, is the same result as that given by equation (4 Si.
96 CYCLE ANALYSIS AND THE SECOND LAW [Ch 6

(6) Work from the pV


Plane We recall that areas under the curve on
the pV plane represent the work of the reversible nonflow process
(/p dV)
The net w ork of a senes of processes is the algebraic sum of the works of the
mdmduat processes These work quantities are positive when the state
point has a component motion toward the right, as abc (Fig 45), negative
when there is a component motion of the state point toward the left The
result is that the sum, the net w ork, is represented by the enclosed area on
the pV plane in foot-pound units (abed, Fig 45) Applying the equations
for work found in the preceding chapter,

~ V
Hr - P,V, and = tf r ’

[isothermal]
In
£ lt's .

(isENTiioricJ

to this particular cycle (Tig 44 or 45) wc got IP = SIP,

(b) W - P.V. In + p'V,\n % +


Equations (d) (g) and (h) should reduce to identical forms (except for
units), because each represents the work of the Carnot cj clc Tor example,
m note
(h), that p»P* = p.P. (Boyles law) and pt V, = pjVd The work
terms of the iscntropics are therefore equal but of opposite sign and cancel
each other Also as already proved P*/P. = VJVt Therefore, with
pV — wltT, equation (h) becomes

ir = kRT, In - wRT, In - - T,)wR lb


(i)
]J-‘ jr-* IT, In
^ ft

the same as equation (g) except for units The student should show that
for an ideal gas, equations (d) and (g) arc the same
Even though there is no practical significance to the Carnot engine, the
cycle itself is most important as the measure of the limit of efficiency for
other ideal or actual cycles It also serves to establish an absolute zero
temperature that is absolute in the sense that no loner temperature can be
conceived (§ 96)

86 Mean Effective Pressure Non flow cycles of which the Carnot is

an illustration take place w ithin a cylinder, w ork being obtained through the
action of a variable pressure on a mo\ able piston The work done during
the cycle is represented by the enclosed area on the p V plane abed (Tig 45)
This area may be reproduced in the form of a rectangle area efgh — area
abed The length of the rectangle is a volume Vd, called the displacement
volume, which is seen to be the volume swept by the piston in one stroke
(Tig 43), Vd = Ve — Va The height of the rectangle is a pressure pm
called the mean effective pressure, and known in abbreuated form as the
mep Since the rectangle is the same area as the cycle area, it too represents
§ 86] MEAN EFFECTIVE PRESSURE 97

W to some scale; hence, the mep is that average pressure which, acting
through one stroke, will produce on the piston the net work of a single cycle.
In equation form, the mep is

(49) Pm = y^
work/cycle (say, ft-lb.) lb
3
~ 2

displacement/stroke (ft. ) ’ ft.

or IF = pm VD .

This equation shows that the higher the mean


effective pressure p m the greater is the work for
,

a particular size of engine (a given Vd). Inas-


much as cost is some function of the amount of
material used, and because mere bulk is usually
a disadvantage, a high mep is desired. Con- Fig. 45. Mean Effective
sequently, it is doubtful that a Carnot engine
which operates on a gas would be useful in
practice even if it could be made operable, because the mep corresponding to

reasonable expansions and compressions is so low that the size of the engine
as compared with modern engines would be excessively large. [See § 89,
part (h).] Furthermore, a comparison of the mean effective pressure with
the maximum pressure of the cycle is revealing. The various mechanical
parts of an engine must be designed maximum forces coming upon
for the
them, forces which will be related to the maximum pressure. So, even if
the mep of a cycle, any cycle, is relatively high, the cycle may be impractica-
ble if the maximum pressure is too great as compared with the average net
pressure (mep), because the large maximum pressure necessitates massive
parts for the engine. Thus, the mean effective pressure is a significant
characteristic of cyclesfrom a practical point of view.
While the units of equation (49) must be consistent, they need not be on
the basis of one cycle. More often, it is convenient to base a computation
on the number of cycles completed per minute N, in which case the work is
in, say, foot-pounds per minute. In order to match foot-pound per minute,
the displacement volume must be in cubic feet per minute (cfm), obtained

from
/ volume in cu. ft. swept\ / number of cycles
D c m _
— piston in one stroke/ \completed per min u-
Also, the displacement is obtained from the dimensions of the engine,

(j) V D = ~~ L cu. ft., or VD = LN cfm,

[one cycle] [iV CYCLEs/MlN.)

where D is the diameter of the cylinder and L the length of stroke, both in
feet, and N is the number of cycles completed
in one minute. To be safe
98 CYCLE ANALYSIS AND THE SECOND LAW [CA 6
m regard to units, always use the work in foot pounds and the -volume in
cubic feet, then convert the resulting pounds per square foot to pounds per
square inch, the unit engineers usually use The mcp of the Carnot cjcle
is the value of IF from equation (i) in foot pound units divided bj VD or
87 ,

_ (T. - T t)wR In (Vb /Va)



P"
y _ y Jcvhnot]

Ratio of Expansion, Ratio of Compression, Pressure Ratio The


ratio of expansion during an> particular process is the ratio of the volume Fi
r
at the end of the expansion divided bj the lolumc T i at the beginning of
expansion Tor instance, the ratio of isothermal expansion (or the iso-
thermal expansion ratio) in Fig 4G is V*/V\ the ratio of lsentropic expansion
is r* *=» Vi/ Vt, the over-all ratio of cxpausion is Vi/Vi

1000‘R
1 | -———92

j>r.
4° -- *3

V S
(a) (b)

Fig 46 The enclosed areas on each plane represent work. The unit of pV areas
is ft lb
,
of TS areas, Btu Since the Btu is 778 times larger than the foot pound it
would be inconvenient to show these areas to relative size This pv diagram is drawn
approximately to scale for the data in § 89 Notice that the enclosed area is relatively
long and slender

The ratio of compression is the ratio of


the initial olume V, div ided bj the
t

final i For instance the lsentropic compression ratio m Fig 4G


olume 1%
is r* = Vi/Vi Observe that each of these ratios is greater than one
The pressure ratio is also usually written ns a number greater than one,
and is therefore the highest pressure divided bj the lowest pressure of a
particular process, pt/pi for the lsentropic process 2-3 (Fig 46)

88 Analysis of Cycles In the anal} sis of gas c> cles wc shall be particu
iarly interested in (1) the heat supplied to the c}cle, (2) the heat rejected
(3) the net work, (4) the efficiencj, and (5) the mean effective pressure
Other items will command our attention in special cases All ideal ejeles
r
are composed of internally reversible processes, so that Jprfl is nonflow
wrork, dQ = cdT and As = / dQ/T
,

Example The working substance of a Carnot cjcle is air, considered as an


89
ideal gas At the beginning of isothermal expansion, w e find p t = 300 psia 1 1 = 5
cu ft , and fi = 540®F The ratio of isothermal expansion is r, — 2 (— Vt/Vi =
'

§ 89] EXAMPLE 99
Fa/ TV) and the isentropic compression ratio rt = 5 (= F</F, = Vz/V : ). Find (a)
the sink temperature and the pressure at each comer of the cycle, (b) the change of
entropy during an isothermal process, (c) the heat supplied to the cycle, (d) the
heat rejected, (e) the horsepower developed if the volume Fx is 5 cfm, (f) the effi-
ciency, (g) the volume V 3 and the over-all ratio of expansion, (h) the mep.

solution, (a) See Fig. 46. For T, = 540 -f 460 = 1000°R, /; = 1.4, and
for rk — V,/Vi — 5, we get

T =
t
^ - 525°R,

or U = tz = 65°F. The pressures are (Fig. 46),

fr y
= Vi\Yj
t 300
= 31.6 psia;
.

pt = 5FI
/FA 300
= ~2 ~ =
.

Pz = Pi
\y-J
lo0 psia;

/F.V 150 .

pz = Pz (jT,/ = gTl = la -8 P S “.

(b) The change of entropy is

AS_ _
. wR Vz piV
- j In Vi - JTi ,
t

1"
Fs
Y
_ (300) (144) (5)
(778)(1000)
2 “ °’ 192 ° m
Btu/ R ’
t

an increase for 1-2, a decrease for 3-4.


(c) The heat supplied during 1-2 is

Qi.z = In vw = (Ss - S,)T, = (0.1925) (1000) = 192.5 Btu.

(d) The heat rejected during 3-4 is

Qz-t = In = -05: - -SriF: = - (0.1 925) (525) = -101 Btu.

(e) The work is ZQ, or

TF = 192.5 - 101 = 91.5 Btu.

for 5 cu. ft.If we consider, for example, that one cycle is completed each minute
with F> 5 cu. ft., the work is 91.5 Btu/min. and hp
= = 91.5/42.4 = 2.16. See
Appendix for conversion constants.
(f) The efficiency is TV/Qa or ( T — T= ),'Tx,
x
or

91 5 1000 — 825
= ~ -
«
I9 T5 = 47 5 ^’
- or e
1000
4 '-°5c-

(g) For the volume Fa, we have

V. = 2Fx = 10 cu. ft., and Fa = 5F; = 50 cu. ft.

The over-all ratio of expansion is

' - ft - (ft) (ft) ’ <5)(2 > 10 -



100 CYCLE ANALYSIS AND THE SECOND LAW [Ch 6
(h) The mean effective pressure 13

jr (778) (91 5)
P- - = 11 psi
V o (50 5) (144)

As practical engines go a low mep characteristic of Carnot engines In


this is

general a Carnot cycle encloses an area on (he p\ plane that is small for its pressure
range (Fig 46) Refer to § 86 for the significance of this low pressure

90 The Ericsson Cycle * There are several other rev ersible cy cles (§ 91)
(conceivably reversible that is) but we shall describe only one other the
Ericsson cycle The reason for doing this is that the Ericsson cycle mcor
porates a feature of particular importance in large modern, steam power
plants and in other cycles a regenerative effect
Knowing that a reversible cycle is the most efficient Ericsson set out to
design an air engine to operate on such a cycle As in the Carnot cycle
heat is added and rejected at constant temperatures so that the transfer of
heat can theoretically occur with a temperature difference A T which
approaches zero and reversibility In Fig 47 heat is received from an
external source at T along ab
i The hot air then leaves the cylinder and
passes through a regenerator while the pressure remains constant be
The regenerator is a chamber
in which energy can be transferred from the
gas and stored In practice a regenerator of the kind intended here may
contain a labyrinth of brickwork or it may be filled with mesh wire One
end of the regenerator is at a temperature T, and the other end at T j and
there is a gradual temperature gradient from one end to the other
To get the energy into storage reversibly we imagine the gas entering the
regenerator at a temperature 7\ AT [Fig 47(c)J so that heat begins to
+
be transferred to the contents of the regenerator at A As the gas passes
through the regenerator temperature drops gradually to Tt + AT at
its

the exit B and at all points in the regenerator the gas temperature is higher
than the adjacent content temperature by an amount AT which is zero in
the ideal rev ersible case

* John Ericsson (1803 1889) born a Swede methodical stubborn irascible des goer
and inventor extraordinary had a pass on for machinery In England he built a loco-
motive to compete with Stephenson s Rocket Ylthough the Rocket won the prize on
tractive effort and dependability Ericsson s locomotive establ shed the amaz ng speed
cecoed. of. 3/1 tojjIl Thft Lnndnn Timrji saui It seeme/L indeed to fly presenting, one of
the most sublime spectacles of human ingenuity and human daring the world ever
beheld Because he was unable to interest anyone in England mbs screw propeller

a substitute at the time for paddle wheels he came to the United States to promote it
lie is the designer of the Monttar wh ch defeated the Confederate iron clad the Memmae
He des gned and bu It a hot air engine for a boat launched in 1853 sunk shortly after
wards in a squall The weight of the engine for the power developed was tremendous
and the boat was not capable of carry ng it in rough weather He vv as such a prol fic
producer that space does not perm t even a list of his invent ons and des gns During
the last few years of his life he was a recluse in New York City irate at the elevated a
disbeliever in the telephone
§ 90] THE ERICSSON CYCLE 101

After this storage operation, the gas rejects heat at constant temperature
to an external receiver, the sink; that is, there is an isothermal compression
of the gas cd. Now this gas reenters the regenerator at B [Fig. 47(c)] at a
temperature T« — AT, and being slightly cooler than the contents, it receives
heat and continues to receive heat until the gas leaves at a temperature
T\ — AT at A. As in the previous passage in the opposite direction, there
is always a temperature difference of AT between the gas and adjacent
contents. In the limit, as AT becomes zero, the whole operation becomes
reversible. It is essential to observe that the heat picked up by the gas in
moving from A [Fig. 47(c)] is the same as that stored when the gas
B to
passed through from A to B. This heat does not involve an external source;

Fig. 47. Ericsson Cycle. This cycle consists of two isothermal processes ab and cd
and two constant pressure processes be and da.

it isan interchange of heat within the system. Now if the source of heat
and the sink are at temperatures T and T 2 respectively, all processes are
i ,

reversible externally and internally.


Thus, with a system consisting of the gas and the regenerator, no matter
crosses the boundaries, and the energy crossing consists of Qa, Qr, and II

Qa = VLBlEi In It Btu, along ab.


J ' a

Qp. = —
wRTt
J
=
,
In
Va
w-
Vc
= wRTi
Jf
— In
Y
>d
e ,
Btu, along ,
cd.

From Charles’ law, we have


Vd T2
£
V„
= £
T j
and
F„ Ti
fc
Vd
It
F„

Using this relation, we find the work to be

(k) IF = SQ = (Ti — Ti)


^ In
^ Btu,
102 CYCLE ANALYSIS AND THE SECOND LAW [C/i 6
and the thermal efficiency is

TV Tt - T t
€ ’
Qi Ti

which the same as that of the Carnot cycle, the highest possible efficiency
is

for any working substance operating between temperatures T i and Tt

(a) (b) (c)

Fig 47 Repealed

The work may also be taken from the p\ plane by algebraically summing the
work quantities for the processes of the cycle

- - vy,
(1) ir - p.V. In -pi + p,(I , 1 .) + p«I « In
£ + Ml .

in foot pounds The student should check this equation with Fig 47 and
show that it is the same as equation (k), except for units The mean
effect!! e pressure is

where Vo = V| — I < Ericsson engines ha\ e been u=ed to a limited


extent but the mep is too low for it to be a competitor of modem reciprocal
mg engines
The Stirling cycle is a regeneratn e cycle similar to the Ericsson cycle,
except that it is composed of two isothermal and two constant tolume
pToce3sea, the itgeaerat^e action gccuttois dwnwg the eewstawt \oluuafi

processes

91 Reversible Cycle Id a reversed cycle, the net work is an energy


input, the net heat is rejected <?* Qs and the ci cle is a heat pump t\e
>
refrained from defining a reversible cycle until the reader had a chance to
learn something about cy cles Now we may sa\ that a reversible cycle is one
composed of processes which are reiersible internally and externally there is
no friction and heat is transferred with an infinitesimal temperature drop
§ 92] THE REVERSED CARNOT CYCLE 103

The consequences of this definition are that heat is received while the work-
ing substance is same constant temperature as the source; heat is
at the
rejected while the working substance is at the same constant temperature
as the sink (AT nearly zero), except ns reversible heat interchanges occur
within the system in A process which receives heat from an
a regenerator.
external source while temperature varies can be made externally reversi-
its

ble with the use of reversible engines, but if such a process is part of a cycle,
the cycle will not qualify as a reversible cycle as we have defined it. Any
cycle with such a process (exclusive of reversible engines, which may be con-
ceivably included) will have a lower thermal efficiency than (T i — Tj)/T\.
The following discussion will explain further.

92. The Reversed Carnot Cycle. If a cycle is a reversible cycle, the


energy effects all take place in the opposite sense, but (he paths traversed

Fig. 48. Reversed Cycle— Heal Pump. This figure, with certain devices not shown,
represents a reversed cycle. Note that the work is supplied to the system, that heat
is rejected at the higher temperature, and that heat is added at the lower temperature.
The net work is »' = - W,„ = 2<? = Q A - Qh, a negative number which indi-
cates that work is done on the working substance.

by the stale point arc identical with the paths before the cycle was reversed.
Customarily, in a power cycle, the state point is pictured as moving in a
clockwise sense on the pV and TS planes. In a reversed cycle, the state
point conventionally moves in a counterclockwise sense. We can reverse
cycles, make them go counterclockwise, which are not reversible in the
defined sense. For instance, cycles which receive and reject heat while
the temperature of the working substance varies can be made to “run” in
either direction. But since the external irreversibility is inevitable in a
natural heat transfer with varying temperature, such a cycle, though
reversed, is not a reversible cycle, which is ideal in all respects.
Reversed cycles are used for two purposes: (1) to provide a cooling effect
machine) and (2) to provide a heating effect (a heat pump,
(a refrigerating
Chapter 14). system receives work from the outside, heat can be
If the
made to flow into the system from the cold reservoir and flow from the
106 CYCLE ANALYSIS AND THE SECOND LAW [CA 6

that the assumption that engine I is more efficient than engine R is absurd

and impossible
Engine I could be taken as another re\ ersible engine as easily as not, and
the same sort of reasoning as abo\e would lead to the conclusion that one
reiersible engine cannot be more efficient than another ichen both engines operate
beiueen the same temperature limits all reiersible engines operating betneen
the same temperature limits haie the same thermal efficiency namely
(r, - sro/r,

94 Example Suppose a cycle is composed of the following processes from


Pi = 20 psia and 200 “T 1 lb /sec of air is compressed poly tropically with
pi 1 * «* C until pt *= 100 psia it is then cooled at constant pressure until <> t,

nn isothermal process completes the cy cle Determine (a) the heat added (b) the
heat rejected (c) the work and horsepower and (d) the mep (e) Write the equa
tion for the work from the p\ plane in terms of the state names in Fig 51
sottrriov This example illustrates the kind of knowledge you should have
about analyzing eyefes T/ie eyefe defined is prohabfy like none which has ever
been used but it is eas> to analvze it The first tiling to do is to construct the
cycle (freehand) on the pi and TS planes This you should do from the problem
statement alone before studying Fig 51
A poly tropic compression cur\e slopes up-
ward toward the left on the pi plane and
also on the TS plane n hen 1 < n < l (Fig
39 p 85) Therefore such a curie ab can
be draw n of indefinite extent on both planes
(Fig 5J) Now draw a constant pressure
line of indefinite extent intersecting the
This locates point 2
pol> tropic Now the
problem states that cooling occurs from 2
therefore the \olume and entropy decrease and point 3 is located somewhere
as shown relative to point 2 The final process is one m which T = C so
this curve is drawn through 3 Where it intersects the poly tropic a6 is located
point 1 It usually does not matter that the various points in the cycle
are not located to scale We
nc»« see that positive Q (heat added) occurs along

3 1 negative Q along 1-2 and 2-3 see the TS plane and the component motion
of the state point right and left Likew lse there i» positiv e w ork during 3-1 and
negative work during 1 2 and 2-3 The cycle goes counterclockwise (II should be
negative) and is therefore a reversed cycle (but not a reterstWe one)

To find the items asked for we shall need fefs = to ft '•ec )

-«ra
1
yOT, (M 3K6ffl) ,,(E!),12 2
Vpi/
m)-2«ch
V100/
p, (20) (144)

(a) The heat added along 3 1 is

Q. - - 72 6 Btu/sec
§ S3] TEERIIOD'TTAMIC TEMPERATURE 107
(b) The heat rejected is the sum of (ir = 1 Ib.,see.)

«-- * “•
(HO w. - n
= (0.1715) (S63 - 660) = -34 .5 Btu sec..

Q-.-z = tre/Fj - F = 0J24C66O - S63) = -45,7 Btu/sec.


;)

O s = — 34.S — 45.7 = — S3 .5 Btu sec.


(c) Tbe r-ork is T.Q. or

W = 72.6 — S3.5 = — 10.9 Btu 'sec.


104X60 = 15 , fcD _

Tbe negative sizn for IT indicates that this much work mitt be supplied to the
system to keep it running.
(a) Tbe nep is

TT _ 10.9 X 77S 6.03 Dsi.


r - Yt
t (12 2. - 2.44) (144)
Observe how low this pressure is.
(e) The work equation in terms of the — J p tfF around the cycle is often useful.
We should be able to write it easily whenever the cycle is defined, and although it
serves no particular purpose in this problem, we shall write it for the practice. For
Fig. 51, we have

IT = m p.(r _ r.) 4- p*Vt In


1
U-Ib,
i — r,
3
f-;
[

which is the algebraic sum of the work quantities. The siuns are automatically
negative where they should be when numbers are inserted. The value of IT obtained
from the foregoing equation should check with that in part
(c). when' expressed in the same units. In important cal- T
culations. all available checks should be made.

95. External Irreversibility. You have probably


observed that the analyses of cycles is on the basis of
internal reversibility. Suppose a Carnot engine re-
ceives heat from a source at Te while the temperature
in the engine is TV where Te > Ft (Fig. 52). Simi-
Iarlv.
- the svstem at F- discharges
w heat to a sink at Tc . __ „ Eztemcl
Tig. 52.
__
Ts > Tc . The eSciency of the engine is (Fj — F;)/T>. irreversiirlily.
as usual But if the heat were transferred without tem-
perature drop from Te to 7% and from T. to F,.. the eSciency would be
(Te — Tc)'Te . which, even without numbers inserted, is seen to be greater
than before.

9S.Thermodynamic Temperature. The absolute temperature that we


have been using is a thermodynamic temperature, which is independent
108 CYCLE ANALYSIS AND THE SECOND LAW [Ck 6

of the tliermometnc substance It is an energy scale of temperature


deduced by Lord Kelvin * Suppose a quantity of heat Q is supplied to (

a Carnot engine A (Fig 53) at a temperature level 0! on some unknown scale


Heat Q 2 is rejected at temperature 0* Since the Carnot cj clc is a rectangle
on the 0 S plane, we sec from Fig 53 that

Let the heat Qt be rejected b> engine 4 to another Carnot engine B (Fig
53), which then discharges heat Qt at 0j As
before,
4 Qi = 0s

Qt o,’
k
and so on for anj number
of engines If each
engine A, B docs the same work, then
C, etc
0i
— 0* *= 0s — Thus we may define a
0j etc
degree of temperature as the temperature change
*7 of the working substance a Carnot engine under m
specified conditions as illustrated bj Fig 53
These conditions might be adjusted so that any
Fig S3 Kelvin Tem desired number of reversible engines are interposed
perature
between temperature level 0i and that level at
which 0. = 0 Such a scaleis based on energy quantities and is independent

of the properties of the substance The cflicicncj of a Carnot engine on


this scale is
Qi — Qn 0i — 0j ,
02
(p)
0i
[any cycle] [carnot]

William Thomson (Lord Kelvin) (1821 1907) who was a professor of physics at
*
Glasgow University is credited b> some as being the greatest English physicist Cer
tainly ho possessed a rare combination of talents His early education was received from
his father who also w as a professor at Glasgow University As a j outh he was robust an
active participant in athletics and student affairs at Cambridge jet he was most distm
guished mhis studies and before his graduation at age 21 from Cambridge I e had cstab
lishcd an enviable reputation in scientific circles by lus original contributions An evccl
lent mathematician a genius at inventing and designing laboratorj apparatus and
models he claimed that he could not understand his own ideas until he saw them at
work jo znpdris JIc contributed most to the science of Ihcj-rooiJyjjsjnirs having cstab
lishcd a thermometric scale of absolute temperatures which is independent of the prop-
erties of anj gas hav ing aided in establishing the first law of thermodjTiam cs on a firm
foundation and having state! significantly the second law He was the inventor of
some fifty-siv instruments and mach nes and in add tion to all this he was interested in
the arts and was himself a musician He was knighted for lus indispensable services in
laying the first successful transatlantic cable and later was made a peer Baron Kelvin
of Lnrjr He vigorously denounced the abs ird ridiculous time-w asting brain-destroy
mg British system of weights and measures favoring the metric system He received
honorary degrees from nearly every important university in Europe was elected a mem
ber of every foreign academy of science and art He probably received more honors for
and recognition of his achievements during his 1 fetime than has any other scientist

§ 96] THERMODYNAMIC TEMPERATURE 109


as compared to equation (48) already found.

[carjtot]

Since each of these equations refers to the efficiency of the same reversible
engine,

ZL-
(q)
Ti

which is to say that in the previous work with the Carnot cycle, a thermo-
dynamic scale of temperature, one where 8 /B« = Qj/Q, in a reversible
;

engine, was implicit, although it had not been defined. The temperature
degree on the 6 scale is not necessarily the same as on the Rankine (or
Kelvin) scale, but can be so chosen by assigning the corresponding number
of degrees between the freezing and boiling points of water. (Read § 8
again.)
Suppose that a Camot engine operates between the limits of 01 and
6r
.
= 0 (Fig. 53). Its efficiency would be
— 0i - Br 0>
e 100 %.
e, 0i

Ifa Carnot engine discharges heat at some negative value of 0, say, —6',
then its efficiency would be

= 0. - (-00 = 0i d- B'
e > 100 %,
0i 0i

or some number greater than 100%. If this were true, the engine would be
discarding more energy than it received, doing work at the same time; that
is, it would be creating energy, a violation of the first law of thermodynamics.

Thus the energy scale of temperatures is an absolute scale in a special sense


zero on the energy scale is the lowest temperature conceivable, because a
lower temperature violates the first law.
Inasmuch as the Camot engine cannot be used for actual temperature
measurements, the question arises as to how thermodynamic temperatures
can be determined. Also, considering that temperature is measured through
the change of some other property, we might conclude (with Kelvin) that
the trick can be accomplished if we use a substance for which pv is propor-
tional to T. Then, for example, if v is held constant, Ap is proportional to
AT. Such a substance is an ideal gas, pv = RT. That is, the ideal gas
temperature scale (Charles' law) and the energy scale are the same. (See
§ 85.) Although there is no ideal gas, both hydrogen and helium approach
ideal, getting closer as the pressure is reduced. First choice for calibration
purposes a constant-volume helium thermometer, with corrections made
is

to give thermodynamic temperatures. Current working limit for gas


thermometers is about 2000°F.
HO CYCLE ANALYSIS AND THE SECOND LAW [Ch 6

In order to calibrate actual temperature-measuring devices, there needs


to be agreement on some reference temperatures, guide points along the
"ay By way of illustration, the following primary temperatures, at stand
ard atmospheric pressure, are rather accurately known (34) boiling point
of oxygen, — 182970°C, boiling point of sulfur, 444 00°C, melting point of
SCO 8°C, melting point of gold, 10G3 0°C A number (some 20 or
silver,
more) of other secondary temperatures are accepted internationally for
example, melting point of mercury, —38 87°C, melting point of tin 231 9°C,
melting point of platinum, 17C9°C, melting point of tungsten, 338Q°C
It may be of interest to know that the lowest temperature so far reached in
the laboratory is about 0 01°K

97 Various Statements of the Second Law of Thermodynamics The


second law, like tlio first, is a consequence of experience and logic It is
founded on the work of Carnot A macroscopic phenomenon which violates
the second law has never been observed hence we say that it is most
improbable that a violation will c\er be obsened There are so man}
facets to this law that it is being stated in different ways below (recall
Clausius, too, § 93)
Ivelvm It ts impossible by means of inanimate material agency to derne
mechanical effect from any portion of matter by cooling it below the temperature
of (he coldest of the surrounding objects
Kelvm-Planck (IS) It is impossible to construct an engine which while
operating m
a cycle produces no effects except to do work and exchange heat
with a single reservoir (See Fig 50(c) ]

All spontaneous processes result in a more probable state

The net result of any actual process ts an increase in entropy of the system
and its surroundings considered together (not necessaril} on increase in each)
This statement and the preceding one arc the most general in their implica
tions, but they are not so obviously based on experience
No actual or ideal heat engine operating m cycles can conierl into work all

the heat supplied to the working substance it must discharge some heat into a
naturally accessible sink Because of this aspect, the second law is often
aptly referred to as the law of degradation of energy
Certain of the foregoing statements overlap to the extent of saying the
same thing in different ways, some apply especially to the production of
power from heat, some are more rigorous and general than others, but each
one contains a significant idea that the reader should make a part of his
store of knowledge Those who enjoy exercises m logic may prove that if
one statement is true, another is necessarily true (I) Several of these
aspects of the second law will be discussed further

98 Available and Unavailable Energy Since one of the more important


endeavors of an engineer is to obtain the maximum amount of work from a
process or cycle and to use the minimum work in reversed operations, it
§ 98] AVAILABLE AND UNAVAILABLE ENERGY 111

behooves us to inquire into conditions which ideally produce the maximum


work. We already know wdiat this is for a reversible engine, but inasmuch
as most engines do not propose to be reversible, even ideally, we need to
set up standards for other situations. From the two expressions of the
Carnot efficiency,

and e = 1 - Qr
Qa
we write
Qr _ T 2
(r) 01 Qr
Qa aV
[CARNOT, OTHER REVERSIBLE CYCLES]

Q R is the heat that must be discharged by a reversible engine,


This value of
the most perfect conceivable, operating between '1\ and T z Since no
.

imaginable engine could convert a larger portion of heat into work,


we say that this Qn is unavailable energy. None of the unavailable

energy can be converted into work unless a colder sink becomes available.
See § 97.
Equation (r) is useful only when the heat QA is added at constant tem-
perature. To get a more general expression, let the heat be added in accord-
ance with some internally reversible path ab (Fig. 54) and consider an
infinitesimal Carnot cycle 1-2-3-4. The working substance is a closed
system. Since the width of this cycle dS is very
small, heat is effectively added at constant tem-

perature along 4-1; call this temperature T.


The sink at temperature T 2 is some natural heat
reservoir, such as the atmosphere, a river, or a
lake. In practice, over a short period of time,
the sink temperature is constant; that is, the
temperature of such a vast reservoir as the atmos-
phere, the ultimate repository, is not noticeably
affected by all the heat discharged from all
engines. Let the natural sink temperature be
designated by T„. Now, in equation (r), let Q A = dQ, Qr = dQ R T =
, i T,
and T z = T0 a constant, and get
,

dQn dQ —•
(s) or Qr = T„
To T J
In words, this equation says that when heat Q is transferred to a system
whose temperature T is variable (or constant), the portion Qr as defined by
equation (s) is the minimum heat that must be rejected as heat in case we
try to obtain work from the heat Q. For this reason, we call this particular
rejected heat the unavailable energy. Thus, the unavailable energy E u may
be defined as that portion of any transferred heat which cannot be converted
.-

112 CYCLE ANALYSIS AND THE SECOND LAW [Ch 6

into w ork when the heat is used in a fnctionless engine for which all processes
are ideal Since the Q& of equation (s) has this special significance we
shall designate it by Eu Moreover recalling
that f dQ/T = AS for a reversible process (§ 39),
we write

(50) E„- T.J^ - 3% AS,

where AS the sj stem s change of entropy dur


is

lag the transfer of heat Q Available energy E,


may now be found as the transferred heat minus
*
the una\ ailable energy

(51) Ea - Q — Eu - Q — T, AS

Available energy is energy -which is comerhble 100% into work in the absence
of trrcicrstbihites, and m this case it is that portion of the heat transferred to a
system during an internally reiersilde process which
work in a closed cycle
is available for coniersion into
£
and its amount is given by equation (51) whether
or not the system remains at constant temperature
while receiving the heat After heat has been
/
added ab (Tig 54), the at ailable cnergj area abed 0
Ta
can bo realized as work by completing the cycle **
/ // / /
with twoiscn tropics he and da and an isothermal
/ r~ T»
cd The area edef represents the una\ ailable
energy The actual w ork obtained from any heat - m
Q will always be less than the available energy — s„ s#— s
99 Example One pound heated at con
of air is Fig 55 Notice that
stant pressure in a closed sj stem from 4, — 140°F to when the sink temperature
is lowered from T to T
tb = 540T Determine (a) the heat supplied (b) the
the unavailable energy de
change of entropy (c) the unavailable energy for sink creases the available part
temperatures of 40°F and — 60°F (d) the value of increases, conclusion dis

Q k/T0 for 40°F and fori„ = — 60°F (e) the maxi charge heat at the lowest
practicable temperature of
mum possible work which might be done by the heat the system Also if all
found in (a) when the sink temperature is t. = 40°F the heat were transferred
solution (a) The heat added is represented by at the highest tempera
area abmn (Fig 55) T„ = G00°R and Tb = 1000°R ture, along gb instead of
ob, ttt'ftiu extra bxnA wa
Qa = c„(Tb - TJ = (024) (1000 - 600) = 96 Btu/Jb agb. Is available energy
conclusion transfer heat

(b) Si,
— Sa = As Cp In jr
while the system is at the
highest practicable tem

= 0 24 In
^ 0 123 Btu/lb °R

is often used synonymously with ideal work of a cycle


perature

*The term available energy


but if applied indiscriminately to cycles which are not externally
available energy ia

reversible it has a number of mean ngs As defined by equation (51) it has one precies
and unique mean ng with respect to heat Q for a particular sink temperature T,
§ 100] ENTROPY FROM THE CARNOT CYCLE 113

(c) For t0 = 40°F = 500°R, the unavailable energy is (area cdnm )


Eu = T0 As = (500) (0.123) = 61.5 Btu/lb.;
and for t„ = — 60°F = 400T1, (Eu represented by area efnm)
Eu = T'0 As - (400) (0.1 23) = 49.2 Btu/lb.

(d) For each sink temperature,

= = = °‘ 123 = As Btu /lb-°R.


500 400

Note that the heat rejected to the sink divided by the constant sink temperature
is always the increase of entropy of the sink.
(e) The maximum possible work which can be obtained from this Q A = 96 Btu
is the available energj- Ea ,
which, at t„ = 40°F, is (area abed, Fig. 55)

Ea = Q A - Eu = 96 - 61.5 = 34.5 Btu/lb.

note. If this air had been heated at constant pressure in an irreversible manner
bj- an electric coil (or bj- paddle-wheel work), the entropy change of the system
would be as computed; because entropj- change is dependent on the end states
and can be reckoned b>- an intemallj- reversible path at constant pressure in this —
example. However, if the computed entropj- change is brought about bj- heat
from a reservoir in the surroundings, the reservoir undergoes a decrease in entropj-
that offsets to some extent the gain in entropj- of the sj-stem. If the heating is
done with electricitj-, which is ideallj- 100% convertible into work, the decrease of
entropj- of the surroundings is zero, and the gain of entropj- of the system is the
net gain. The greater the specific entropj- gain, the more irreversible the process.

from the Carnot Cycle. It is seen from equations (s) and


100. Entropy
(50) that bj- considering theCarnot cj'cle in terms of differentials, •we end
up with the expression j dQ/T, and it is this integral for an internally rever-
sible process which was used to define entropy AS. We maj- now demon-
strate that entropy is a point function. Write equation (s) in this form,

dQ dQ R
(t)
T T0
The integration of (t) may be written (permitting T„ to varj- in the general
case)

<»> /¥-/¥ = 2 /¥=°> [reversible cycle]

where Z/ dQ/T represents the algebraic sum Z of the integrals around the
cj-cle, which instead of one, maj- be composed of several processes during

which heat is added, and several during which heat is rejected. When the
cycle is composed of intemallj- reversible processes, the sum of these inte-
grals is equal to zero, as shown. Since a property is a characteristic which
is alwa\T s the same in a particular state 1, (u) maj be accepted as proof that
T
i

m CYCLE ANALYSIS AND THE SECOND LAW [CA 6

entropy a property, 2 (AS) = 2/ dQ/T = 0 which says that S returns to


is
its original value no matter what various cycles of reversible processes are
performed from state 1 It can also be shown that the entropy returns to
its initial value in a cycle composed of irreversible processes

101 Clausius’ Inequality For the cycle of internally reversible proc


esses abed (rig 50), which is any cycle chosen at random, the system’s
increase in entropy during a6 is balanced bj the system's decrease cd, and

[ dQ ~ QR V f dQ ^’ I pAT »
J ~T ~T Lj I ~T
as previously found Now
suppose that the expansion from b is irreversible
instead of re\ ersible, along some path be', and for convenience, let be' be
adiabatic (Q — 0) Since the process be' is irreversible less work is done
than in process be, therefore, more energy remains in the sjstem than when
the expansion is reversible and point c' is consequently to the right of point
c Thus since more fieat Q'R is rejected after the irreversible expansion,
Q'n/To is a larger number than Qr/T, Then

Qh _ dQ
<
(v)
/¥ T0 ~ II :
o,

and it is less than zero (a negativ e quantity) because Q R /T, is a larger num
her than J dQ/T In other words this equation says that the sjstem
discharged (AS ) more entropy than it rcceiv ed (AS,) from the surroundings
it is producing entropy (AS,) Combining the
ideas of equations (u) and (v), we can say m
general that for the sj stem

dQ- n
< 0
(0
II rjy |

which is known as Clausius’ inequality, and


where the mathematical signs mean that the
algebraic sum of dQ/T is taken over the entire
cycle This development docs not provide a
general proof of the Clausius inequality («ee 3,
Fig 56 Irreversible Proc
148), but if one has confidence in the second
ess in Cycle
law, acceptance should be easy

102 Entropy Production The net result of irreversibility is an increase


in entropy (AS,, Fig 56) This increase of entropy due to irreversibility ts

called the entropy production or entropy growth, and often expressed as a


is

rate (Btu per °R mm ,


for example) From Fig 56, vve may write the
equation,

<w> S-^+AS,- 0
;

§ 102} ENTROPY PRODUCTION 115

where A S p is the entropy production during process be’, and the entire inter-
nal production of entropy during this particular cycle (Fig. 56), because the
other processes are internally reversible.
For a cyclic operation, theentropy of the system returns to its initial
value Sa every time completed. That is, the internal change of
a cycle is

entropy is zero for irreversible as well as reversible cycles. The entropy


production during a cycle is the growth of entropy of the surroundings.
We may say in accordance with the second law that
(^) AjSgyatem "4“ AtSgurroumlincs = 6,

where the equals sign applies only when the processes are internally and
externally reversible. For all actual processes, therefore, we can write
A*Sp AlSgyatctn 1 AiSaurroundings

(y) A Sp — AS, s.,te m + A S, Mrcc + AS,*,],


where A S p is the net increase in entropy and A,S, yltC m = 0 for a cycle and the
other changes are algebraic (AS, 0Utc „ is negative when it supplies heat,
positive when it receives heat in a reversed cycle). Since for any process,
the actual change of entropy of the system is A S' ^ f dQ/T, where the

equals sign applies for internal reversibility, we may write

j^+
00 AS' = AS,,

the actual change of entropy, applicable to a process or a series of processes.


If a system’s entropy change is a loss, the gain of entropy of the sur-
roundings will be greater than the loss, resulting in a net gain. Thus,
entropy, like time, is unidirectional and always increasing, and moreover,
the change of entropy is a measure of the change of unavailable energy (T0 AS)
^the greater the entropy change because of irreversibility, the greater the
increase of unavailable energy and the greater the loss of available energy.
Energy in the form of work is available energy; kinetic energy, potential
energy, and electricity are available energies, because in a perfect engine,
these forms can be converted 100% into work. Thus, using electricity for
heating water, for heating metal for heat treatment, etc., is using the highest
grade of energy (100% available) for a job that can be done by low-grade
energy. Nevertheless, such use of high-grade energy is often economic in
industry or is so convenient that the purchaser is willing to pay the extra
cost. As implied by the second be graded. The energy
law, energy may
from a source at high temperature is of higher grade (larger percentage of
available energy if used directly in a heat engine) than energy from a source
at a low temperature. See example, § 103.
Ultimately, the available energy which is evolved becomes unavailable
energy. All actual eventually dissipated by friction, or
work produced is

the equivalent, and becomes energy in the sink. For example, all the
m CYCLE ANALYSIS AND THE SECOND LAW [Ch 6

millions of horsepower produced in automotive engines arrive at the sink


level via friction between the vehicle and the air, friction associated with the
engine (Chapter 9), and friction at numerous other points, the remainder
of the energy released by the fuel reaches the sink via the exhaust, the air
across the radiator, etc High grade energy is continuously being degraded
See degradation of energy § 97 ,

The entropy production is a measure of how irrev ersible a process or cycle


is, the greater the enfropj production, the more irreversible the process
And with Clausius, the entropj of an isolated sjstem tends to increase to a
maximum (at the dead state), the energ} of an isolated system remains
constant

103 Example Loss of Available Energy with Heat Transfer A source at 1600°R
is available for transfer of heat at the rate of 3200 Btu/min to a sjstem at 800°R
If these temperatures remain constant and if the sink temperature is 600°R deter
mine (a) the decrease of entropy of the
source (b) the original available energy,
(c) the entropy production accompanj
ing the heat transfer (d) the increase of
unavailable energy, (e) the available en*
ergj after heat transfer (See Fig 57 )
solution (a) Since the temperature
is constant f dQ/T = Q/T, and we find

--SWBmm
a decrease or negative change with ro>
spect to the source because Q is flowing
out of the source
(b) The unavailable part of this heat
Fig S7 Irreversibility with Heat = 1200
Transfer at 1000°R is T'AS,b (000) (2)
Btu/nnn The available energ} area
abhk is (taking Q as positive for this computation)

- Q - Eu = 3200 - 1200 - 2000 Btu/ram (02 5%)

(c) The entrop} entering” the system receiving Q at a constant S0O°R is

! dQ/T = Q/T, or

AS., - %- - +4 Etu/"R min

The entrop} production because of the irreversible heat transfer is

AS, = AS* + ASt/ « -2 +4 - 2 Btu/°R nun

(d) The increase in unavailable energ}, area Imgc, is

AEu = T.AS, = (600) (2) = 1200 Btu/°R min


§ 104} EXAMPLE: HEAT EXCHANGER 117

(e) At 800°R, the available energy, area efmh, is

Ea = 3200 - T0 AScf = 3200 - (600) (4) = 800 Btu/°R-min.,

which is only 25% of Q = 3200 as compared with 62.5% before transfer. Another
and telling point of view is that the loss of available energy is

2000 - 800
60%
2000

of the original available energy, a horrifying percentage. Let’s repeat that idea:
the process of heat transfer is 100% efficient as far as the first law is concerned (all
the heat leaving the source was received by the system), but only 60% efficient with
respect to the second law!
These results emphasize the desirability of using energy in an engine at as high a
temperature as possible and of transferring heat with as small a temperature differ-
ence as practicable. The rapidity with which heat is transferred is pioportional to
the temperature difference, other things being equal. Hence, the difference must
not be too small, because would then take too long to transfer the required heat
it

(or a very large heat transfer surface would become necessary). Also, it should be
easy to visualize from Fig. 57 that the lower the temperature at which real engines
discharge heat, the more efficient the engines are likely to be. Many steam power
plants are more efficient in the winter than in the summer because the sink tempera-
ture T„ is lower in the winter. The lowest economic temperature T„ is one which
is naturally at hand. If the discharge temperature T„ were lowered by refrigeiation,
there would be a net loss because the work to run the refrigeration cycle would be
more than the work gained from the power cycle as the second law says.—
104. Example: Heat Exchanger. In a heat exchanger (Fig. 58) 100 lb./min. of
water (c = 1) be heated from 140°F
are to
240°F (700°R) by hot gases w„h t -* - Gas - w„ha
(600°R) to
(c p = 0.25) which enter this exchanger at 440°F 1
-*>
^
w* h i“ a er — *-ww h z 2.
(900°R) and flow at the rate of 200 lb./min. q
Compute the change of unavailable energy of
la9ram ’
each fluid for a sink temperature of 40°F (500°R), Heat Exchanger^"
the net increase in entropy, and the loss of avail-
able energy. Consider this heat exchanger as an isolated system with no exchange of
energy with the surroundings while the materials flow through.
solution. The changes in kinetic energies are likely to be negligible and no
work is done. Therefore, the energies entering and leaving the system are the
enthalpies of the flowing fluid, as shown in Fig. 58. Now we may think of two
subsystems, the gas and the water. In accordance with the first law, the heat flow
from the gas is numerically equal to the heat flow into the water, and the algebraic
sum is equal to zero.
of these heats For the water, A w = wcAT, very nearly H
closeenough for low-temperature heat exchangers (accurate means of getting Ah for
water are given in Chapter 10). For the gas, A H„ — wcp AT. Therefore, we have
(Fig. 59)

Q water “I” Qgas — 0, Or Qgas = Qwater*


AH„ = Q water = wc(T« - T,) = (100)(240 - 140) = 10,000 Btu/min.
A Hp = Qm = wcp (Tb - Ta) = (200) (0.25) (’A - 900) = -10,000,
,
,

118 CYCLE ANALYSIS AND THE SECOND LAW [Ch 6

from which 7i = 700°H, the 'econd temperature of the gas which is necessary to
satisfy the first law The entropj changes are
y. t ooo
=
AS, tcc. In - (200)(0 25) In
^_ - 12 57 Btu/°R mm
the negative «ign indicating a decrease of entropv, and

AS* = ire In —
Ti
100 In
700
= +1 o 4 Btu/°It nun ,

the positive sign indicating an increase of entropj Of the 10000 Btu transferred
from the gas we have for ab,

E., = T. AS = (500) ( — 12 57) = — C2S5 Btu/min, a decrease


E„ = —10000 + G2S5 “ —3715 Btu/min a decrease
,

Of the 10000 Btu received bj the water, we find for I 2

T\ £„ - (o00)(15 4)
900*R = 7700 Btu/min an increase ,

£.. = 10 000 - 7700


- 2300 Btu/min an increase

The entropj production is the net change of


entropv

AS, - AS, i + AS* +15 4 + (-12 57)


= +2 ^ Btu °R mm
the po«itive «ign showing that the net change
ia an increase The loss of available energj
(increase of unavailable energ\ sometimes
called irrfrcrnbt/tfy) is

Fig S9 Entropy Production — Heat = T,ASr — (o00)(2 *^3)

Exchanger =1415Btu/nun

See the area done on the TS plane (Fig 59) which represents this energj

105 Equilibrium W e shall not u«e the space to giv e an evtensiv e treat
ment of equilibrium but a brief discussion will be helpful in promoting
understanding (/, S) Among the various hinds of equilibrium, we have
the follow ing
Thermal equilibrium (§ 19), for which the sjstem is at the same tempera
ture as the environment (W e also speak of two or more bodies as being in
thermal equilibrium with each other when thej are at the same temperature,
and of a single mass as being m
internal thermal equilibrium when all parts
of the mass are at the same temperature ) Mechanical equilibrium which
means that no part of the sjstem accelerating (~F = 0 etc ) and that the
pressure within the sv«tem is the same as in the environment Chemical
equilibrium, w hich means that the sj otem does not tend to undergo a chemi
cal reaction, the matter in the svstem is said to be inert If all three of
§ 108] CLOSURE 119

these equilibrium conditions exist, the system is said to be in thermodynamic


equilibrium (1 ) or in the dead state (S). A system is in the dead state if it

will not spontaneously undergo a process; and its entropy is a maximum.


There are various meanings attached to the word equilibrium. We speak
of equilibrium states of a substance, meaning something which we might
call internal equilibrium. That is, the substance in a system may not be in
thermodynamic equilibrium with the surroundings, yet it may be at an
equilibrium state. That is to say the pressure and temperature are the
same at all points in the substance; there is no acceleration of the whole or
any part of the substance, and the system is chemically inert. Typically,
the substance is at rest. This internal equilibrium is an implicit require-
ment for any reversible process. So, we have been continually dealing with
substances at various equilibrium states. Other meanings of equilibrium
will be taken up as needed.

106. Third Law of Thermodynamics. We


have so far dealt only with
difference of entropy. made in §§65 and 67,
Indefinite integrations were
but no means of determining the constants of integration were at hand. In
most thermodynamic problems, change in entropy from one state to another
is sufficient. However, when a chemical reaction occurs, the final substance
is different from the initial substance, and in this case, it is often a con-

venience to have absolute values of entropy; for example, in computing


Gibbs function h — Ts before and after the reaction.
Important generalizations come gradually and contributions toward the
third law have been made by many: Einstein, Nernst (heat theorem),
Boltzmann, Planck (28). The third law may be stated: At absolute zero
temperature, the entropy of a substance in some crystalline form becomes zero.
Since such a statement may be a naive expectation, there must be experi-
mental verification. Since there are now many experiments which justify
the law, it is being used with greater confidence.

107. Perpetual Motion Second Kind. A proposed machine which


of the
violates the second law of thermodynamics is called a perpetual motion
machine of the second kind. If it can be shown that such a machine violates
the second law by producing work while taking heat from just one reservoir
or if it can be shown that it would produce more work from a given supply of
energy than the available energy, the machine will not operate as proposed.

108. Closure. Neither the Carnot nor Ericsson cycles are seriously con-
sidered today as a means of manufacturing power. The Carnot cycle is a
monumental conception which, as we have seen, has led to important deduc-
tions regarding the limitations of any power machine. The Ericsson cycle
(and the Stirling) involves an idea, regenerative heating, universally used
in large, modern, steam-power plants, although in a form somewhat differ-
ent from that in Ericsson’s engine. Also, in this chapter, we have learned
120 CYCLE ANALYSIS AND THE SECOND LAW [Ch 6

of the essential elements ofany cycle, and we have learned how to analyze
cycles Before proceeding the student should satisfy himself, by working
problems concerning other cycles, that this lesson in analysis is mastered
Notwithstanding the visionary or impractical aspect of these cycles in them
sclv es, it should be evident that the study of them should yield results of
great practical value
We write W W
= Qa — Qr or = 2Q = Qa Qr as may be convenient
+
for a particular purpose, where, in the first instance, Qg is a positive num
ber, and m the second instance, Qg is a negative number Note that both
expressions mean the same thing and you w ill not be confused
It is one thing to sense the irreversibility of events in general and another
to measure the degree of irreversibility, as we are able to do for some events
with the use of the concept of entropy We have noted that whenever a
process or cycle deviates from a rev crsible process or cycle, the amount of
work that is delivered decreases below that for reversibility (or the amount
of work that must be put into the system for an actual reversed cycle is
greater than for the corresponding ideal)
In closing this chapter, w e might point out that the thermal efficiency of a
machine might be low because of a small temperature range, but good
because there is little loss of av ailablc energy or it might be high because of a
large temperature range, and yet not be so good because of a large loss of
available energy
Courtesy Johnson Publishing Co., Cleveland

Archimedes and His Spiral Pump. The source of power 2000 years ago was man.

121
7

COMPRESSION AND EXPANSION


OF GASES

109 Introduction Now we move to some applications Gases at pres-


sures abov e atmospheric arc a common requirement in industry The com-
pression process is an integral part of refrigeration c> cles and gas turbine
cycles Most widely useful is compressed air m operating air engines and
pneumatic tools such as hammers and drills, spraying paint, cleaning by air
blast, operating air hoists, pumping water by air lift, and in a host of other
jobs Although this discussion is concerned specifically with the compres-
is the only substance we have so far learned to handle,
sion of a gas, which
the basic energy equations and borne of the derived ones under specified
conditions apply as w ell to any compressible substance

110. The A widely useful means of studying the per


Indicator Card.
formance of reciprocating engines (internal combustion engines, compres-
sors,etc) is the indicator card or indicator diagram. This "card” is a
pictured record of the variation of pressure and volume of the working sub-
stance in a cylinder as the piston reciprocates The record is obtained with
an indicator, a sample of which is shown in Fig 01 Other types of indica-
tors are used for high-speed engines A typical indicator card for an air
compressor is shown in Fig GO The area of the card is found by using
the plammeter, Fig G2
Although the indicator card is a closed pressure-volume diagram, it is
not a thermodynamic cycle A compressor diagram shows the details of
what necessary for one process in a cycle, picturing the movement of gas
is

into (4-1)and out of (2-3) the cylinder and the re-e\pansion process (3-4)
(Fig GO). However, the indicator card and the thermodynamic cycle have
m
,

Ill] MEAN EFFECTIVE PRESSURE AND IHP 123

one thing in common: areas represent work. The work obtained from an
indicator diagram is called indicated work Wi, which is the work done on the
substance in getting it into the cylinder, compressing it,and discharging it
from the cylinder.
The valves (Fig. 63) in an air compressor typically operate on a difference
in pressure, which explains the waviness in the lines 2-3 and 4-1, Fig. 60.
A relatively large pressure difference is
necessary to start the opening action,
because of friction and inertia, so that
there is generally an abrupt opening
followed by a fluttering. The intake
valve does not open until some pres-
sure a little below atmospheric is
reached. Then it often flutters, as
after 4, Fig. 60, producing the wavy
portion of the suction line 4-1. Notice
that the intake pressure is slightly
below atmospheric. The compression
Fig. 60. Actual Indicator Card,
1-2, which is often close to an adiabatic
Compressor.
process, continues until a pressure
greater than the delivery pressure is reached, at which point the discharge
valve opens. Here again, valve fluttering occurs and the discharge line is
wavy.
For most miscellaneous uses, compressed air, though hot on delivery, is
cool when used. There is time enough before use for heat to be transferred
to the atmosphere, until the compressed air is almost at atmospheric tem-
perature. Since higher-pressure air cannot “hold” as much water vapor as
lower-pressure air (temperature the same, Chapter 15), moisture condenses
as the air cools. Therefore, in a small system, delivery of ar to a receiver
(Fig. 64), equipped with a drain device is likely to be the situation. In
large systems, it will be desirable to let the discharged air flow through an
aftercooler which is simply a heat exchanger wherein water is used to cool
the air and rid it of excess moisture before it gets into the distribution
system. An aftercooler and an intercooler, Fig. 72, are much the same.
See § 119 for cases where aftercooling is not desired.

111. Mean Effective Pressure and Ihp. The actual mep is found from
the indicator card. The indicator, Fig. 61, is fitted with a spring of a cer-
tain scale. If the scale is 100 lb. in the case of an indicator, we mean that a
vertical movement of 1 in. of the stylus indicates a pressure of 100 psi.*
Therefore the actual heights on the indicator card multiplied by the scale
of the indicator spring will be the actual pressures in pounds per square
inch (neglecting errors inherent in the indicator). Thus, to find the actual
* This statement is true when a standard piston is used in the indicator. If an under-
sized piston is used, a common practice, the scale stamped on the spring must, of course
be modified.
,,

m COMPRESSION AND EXPANSION OF GASES [Ck 7

mep, find the area of the indicator card in square inches, divide the area by
the length of the card l m
inches, Tig 60, and obtain the average height of
card in inches This a\ erage height multiplied by the scale of the indicator
spring is the a\ erage pressure, called the indicated mep (p„r)

.
^"*r
_ (area of indicator card) (scale of indicator spring)
length of indicator card

This the pressure acting on the face of the piston during one stroke, which
is

results in the same work as produced by the actual pressures throughout all
events of the indicator card Now reread § 86 again

Fig 61 Indicator Sleam enters through the connection to the cylinder and acts on
a small piston, whose motion is opposed by the indicator spring The greater the pres
sure from the cylinder, the more the piston moves The piston rod actuates the stylus,
through a straight line linkage, and the indicator card is traced The cord is attached to
the drum, which moves m
phase (the motion is reduced w
magnitude) with the engine
piston (and crosshead) No indicator paper is shown on the drum-

To get horsepow er from the mep, divide the work m foot pounds per min
ute by 33 000 The force corresponding to a pressure of Pmi psi is p«iA lb
where Asq in is the piston area This force acts through a distance equal
to the stroke of the piston, L ft , so that the w ork per stroke is pmjAL ft
lb

which is also the w ork for one complete card If the number of cards per
minute (or the number of cycles per minute) is N, the work per minute is

pmiALN = p„iLAN ft-Ib per mm


The indicated horsepower is therefore

lhp =* p m ,LAN
(52) ’

33,000
§ 111 I iiEi^T EFFECTIVE PRESSURE AUD EBP 125

where obtained from the indicator card as explained above. This


p-_- is

equation is mep equation and also the p-L-A-X equation.


often called the
The beginner frequently makes an error in applying equation (52) because,
as derived. .1 is in square inches (to
match pressure in psi) but the strobe
L is in feet. However, it is quite
correct to use square feet and pounds
per square foot together. Observe
carefully that, in general,
.
'
-
X is
.
not the
.
Covrterj CroA-j Sj.cr.v. Gere and ralee Co.,
Boston
r
numDer o: revolutions per minute but
the number of cards (or cvcles) com- _ Fig-62. Plcmmeter. Point F is Sred.
, , . ,
Pomt P is moved around the boundaries
D.e,ed pe*. minute. These happen lO 0 f tie area to be measured. Wheel D
be the same for a single-cylinder, turns with the movement of P, and after
the complete outline has been traced, the
sinale-actins comoressor. Ifthesinale
“ , .
“ , * . r magnitude of the area may be read from
cyhnder IS double-acting. .\ = 2 n. the dial G and the vernier E.
where rpm. In a single-acting
r. is

engine, the working substance acts on one side of the piston; in a dmible-
cciir.g engine, the working substance acts on both sides of the piston.

We have already defined the symbol p„ as the mep of an ideal cycle (or
indicator diagram) and hp as ideal horsepower: hence, the mep equation as

Courtesy lYortrtr.r'on Co-p. r


Hcmvrt, .V. J.
Fig 63. Freiherr Yclccfcr
Gcscrud Yctcrr Compressors.
The strips of ribhou steel
are shown pressing against
the valve guard in the open
position. When the pres-
sure on the valve-seat side
of the valve is less than that
on the guard side, the strips
lie Sat ou the ground seats,
closing the valve. Intake
and discharge valves of this Ccrj.rt.em Ir.rcrcoU-Ttr-r.d Co.,
type are identical. York
Fig. 63. Receiver.
ISO COMPRESSION AND EXPANSION OF GASES [Ck 7

applied to ideal engines is

(53) hp
p = vJjAN
33,000

When the mep


equation is applied to the crank end of a double-acting
cylinder, Tig 65, the A in equation (53) should be the area of the piston
minus the area of the piston rod, which is the net area over which the mep
acts In run-of-the-null compressor tests, engineers often neglect to correct
for the area of the piston rod

packing, piston, piston rod, crosshead, crosshead guide, connecting rod, water jackets on
beads and cylinders, indicated work, shaft work, camshaft for ICE Observe Ihe articu
lafed construclion for the connecting rods, a type of construction also used in radial air-
craft engines Notice that the ICE is single acting (work done on one face of the piston),
that the compressor is double acting (work done on two faces of the piston)

The shaft horsepower, also called the brake horsepouer (bhp), is another
important pow er term This bhp is the horsepow er w hich must be delivered
to the compressor shaft to drive the compressor, and it is larger than the
ihp of the compressor by the amount of frictional losses in bearings, packing,
etc , Fig 65

112 Work of a Compressor The reciprocating as well as the rotary


types of compressors can be considered as steady flow machines on an over
all basis, in which case the work from equation (9) is

IF = A, - ht + Ki - K, 4- Q.
b

§ 112 ] WORK OF A COMPRESSOR 127

There is generally little difference between the entrance and exit velocities,
so that for w’ lb.,

(54) W = w'Oh - h t) + Q = -A H + Q.
[any substance, any steady flow process, AK = 0]

If the substance is an ideal


2 ). gas, hi
If the process is — hi = c p {Ti — T
internally reversible, Q = 0 or Q =
In compressor problems, other fc dT.
forms of the work equation are more convenient.
(a) Work for Adiabatic and Isentropic Compressions. Review § 70.
If the process is adiabatic, Q = 0 and IF = —Ah Btu per lb.; or for a flow
through the compressor of w' lb., we have for constant specific heat

[any adiabatic, ideal gas, AK = 0]

From equations (26) and (36), we have


a-n/k
kR Ti
(c) Cp — - and
J(k 1) Ti

These values and piV[ — w'RT t substituted into the work equation give
t_i,/t

(55) W = J(kw'RTi
- 1 k)
Yg?Y
\pJ
- ii
X
J
= fcp^»
/(i - k) .w
- Btu,
[iSENTBOPIC ONLY", IDEAL GAS, AK = 0]

where V\ is the volume measured at pi and T i corresponding to the mass w'.


( )
Work for Polytropic Compression. Review § 73. We recall that the
poly tropic is defined as a reversible process, that Q = w'c„ AT and
c„(k — n)
Cn ~
(1 -»)
for an ideal gas; that is,

w'c„(k — n) T
Q = w'c„(T 2 — =
Ti)
1 — n irr') Btu.
During polytropic compression from atmospheric temperatures, this value
of Q is normally negative. Using known values in equation (54), we get
w'c v (k — n)Ti (Ti \
W = —w'cpTi ^ 1 -n \Ti)
c v (k — n)
Btu.
I —n
Putting the term in the brackets over a common denominator (1 — n),

using c = kc v and c p — c„ = R/J, this equation becomes


p

W = nw'RTi (Ti A npiV'i


1 [AK = 0]
(56)
J{ 1 - ») \Ti ) J(1 — n)
>
d
m COMPRESSION AND EXPANSION OF GASES [CA 7

the latter form being obtained by using T t/Ti = (pi/p y n~ 1),n andi

w'RTi = Pl V[

This equation seen to be the same as equation (55) except for n in place
is
of K, good for an ideal gas when AK = 0
(c) Work for Isothermal Compression Review § G9 If the tempera-
ture is constant for an ideal gas, AA = 0 and = Q For an isothermal W
process, Q- p\Vi In (Vt/V } ) ft-Ib ,
and piVi = pjFj, or

(57) II [a£ _oi


« p2 J Pi

is the volume of w’ lb at pi and T t


where, as before, V'v
( ) Workfor Irreversible Adiabatic Compression Review §71 Actual
rotary tjpes of compressors consume work approximating the irreversible
adiabatic in steady flow In equation (54) let the actual final state be
represented by 2 ,
11 =* te'(Ai — Aj ) and find

(d, ir - - Tt) - - l)’


(fr
where we have used the ideal gas relation c, = kR/[J(k - 1)1 Usuall}
the temperature T j is measured in the actual compressor unless from
experience with a particular type of compressor, we have some approximate
pT relation such as (see § 71)

s-fer"
Substituting this value for Tt /T i in (d) we get

<•> w ~° ]

Since this equation is for an irreversible compression it represents an actual


work quantity Thus, we may say of m that it is that number used as
shown which results in equation (e) giving the value of the actual fluid
work W—but only when AK
-
= 0

113 Work from Conventional Diagram A conventional diagram is an


idealized indicator card or pV
diagram which can be used to determine the
work of a compressor The analysis from the diagram is not so revealing
thermodynamically, but it is helpful in other ways
Consider first a conventional diagram which suitably portrays the work of
a reciprocating compressor without clearance, Fig C6 The area under
4-1 represents the work done on the piston during the intake stroke and
the area under 2-3 represents the work done on the substance in pushing
(delivering) it from the cylinder For purposes of illustration let the com-
§ 118] WORK FROM CONVENTIONAL DIAGRAM 129
pression curve be represented by the equation pV = C (which could be a n

reversible polytropic but not necessarily so; just pVn = C). Since the
is

work is represented by the enclosed area 1-2-3-4, the algebraic sum of the
areas under each line of the card is the work to some scale of the conventional
compressor.

_ P2V, ~ P.F, p2(Fj _ yj pi(7i _ Vi)


w + +
_
= P 2 V2 - P 1 F1 _ T7 _ Tr
P 2 V2 + ,
piVi,
^

since F 3 and F 4 are equal to zero. Reduced to a common denominator, the


foregoing expression becomes
Discharge

w_ n(p 2 F 2
1
-
— n
piV 1)
Valve

HT
P“
m i

which is also —JV dp for pV" = C. Intake?!*


yaive -
LU
Since F2 /F1 = (pi/p 2 )
1/n
,
this equation 4r v
becomes P
,r Delivery line
1>/n 3
(f) IF - ”PlFl l] ft-lb.
\> Compression
[enclosed area for p n = C]
\^-Iine
4
Suction .
^ .

If n = k, we see by comparison with |


line
T7 J
equation (55) that equation (f) gives the
V
work when there is isentropic compres-
Fig. Conventional
66. Diagram
sion. Similarly, comparing equation (f)
without Clearance.
with equation (56), we see that if n is

for a reversible polytropic, we obtain the work for a polytropic compressor


except of course that the units in equation (f) are foot-pounds instead of
Btu as in the other equations. The Vi in equation (f) is the same as F,
and means the volume passing through the compressor as measured at pi and
T 1 If the substance is a gas, one may use wRT for p\V in equation (f).
. x

Comparing Fig. 66 with Fig. 18, p. 45, you see immediately that the
enclosed area 1-2-3-4 is equal to — JV dp. This integration for pVk = C
is on p. 77. As a purely mathematical relationship, — JV dp for pVn = C
is identical with equation (f) above.
worth noting that equation (f) represents the area of any diagram
It is
bounded by the zero volume line (p axis), two constant pressure lines, and a
curve of the form pV n = C. All the work equations for compressors so far
given in this chapter should give a negative number, inasmuch as they are
on an algebraic basis and the work is done on the substance (enters the
system). For those who must make repeated computations of compressor
work, tables are available which give the value of the pressure ratio to its
exponent, (p 2 /pi) '~ 1,/* of equation (55), or which give the entire term in the
(

bracket ( 14 82). ,
ISO COMPRESSION AND EXPANSION OF GASES \Ch 7

114. Clearance and Clearance Volume. Displacement volume is defined


(§ 80) as the v olume swept by the face of the piston in one stroke In order
to be sure that the piston docs not strike the cylinder head at the end of the
stroke and that there is room for valves, a clearance volume is essential in
reciprocating compressors There is another consideration for clearance
volume combustion engines, but in compressors, the smallest
in internal
practical clearance volumes are desirable (§ 118). Since, as we shall see in
the next article, the power consumption is theoretical!} independent of the
amount of clearance, there would be no point in increasing manufacturing
costs significantly just to obtain a smaller clearance The ratio

, .
_ clearance volume
displacement \ olume

iscalled the clearance ratio, the percentage clearance, or just the clearance
Values of this ratio in practice v ary from about 3 %
in some large reciprocat
ing compressors to more than 12% in others, with most values falling
between 8% and 12%.

116 Work of the Conventional Card with Clearance. The events of


the diagram with clearance are the same as those for the case of no clearance,
Discharge except that since the piston does not
force all the air from the cylinder at the

pressure pi, the remaining air at point 3,


Tig G7, must re-expand to the intake
pressure process 3-4 before intake
starts again at 4 Since the expansion
3-4 involves only a relatively small mass
the value of n on the expansion curve
has little effect on the results and it is

therefore taken the same for both c»™


pression and expansion curv es, although
actually the values are different With
V
Fig 67 Conventional Diagram out clearance, the volume of air taken
with Clearance The greater the into the cylinder is equal to the displace-
clearance (= Fj), the less air drawn ment volume As seen from Fig 67 for
»<- VO the diagram with clearance, the volume
of airdrawn into the cylinder is Vi — V^ = V[ and is less than the displace-
ment volume Vd
To find the work of the diagram with clearance, we imagine it to bemade
up of two diagrams, a-1-2-6 and a-4 3 6 Each of these diagrams is similar
in all respects to the diagram of the compressor without clearance, Fig
hence, equation (f) may be applied to each of them The work of the dia-
gram 1 2-3-4 will be equal to the work a-1 2 b minus w ork a
r
4-3-6 Thus
for 1-2-3-4 we find
§ 116 ] FREE AIR 131

IF = npiVi i
1—71
n Pl (Fi - F„)
(n— l)/n

1—71
- 1

since p K = pi and p 3 =p 2. For V[ = Fx F 4, the work equation becomes


l)/n

c) n'- - >] 1 ft-lb.,


= [fe)“
which again the same as equation (56) except for units; F', is the volume of
is

air drawn mass of air passing through the compressor, corre-


in; w' is the
sponding to volume Fj. The conclusion is that the amount of work necessary
to compress a particular mass of air under given conditions is independent of
the clearance, which is perfectly true for the conventional diagrams. In
the real compressor, however, there are additional frictional effects. The
displacement must be greater with clearance than without, for a particular
capacity; this means a larger machine that will cost more and have greater
mechanical friction.

116. Free Air. Free normal atmospheric conditions in a


air is air at
particular geographical location. A
particular volume of free air at an
altitude of 5000 ft. does not have the same mass as the same volume at sea
level. At the higher altitude, the barometric pressure is lower, Fig. 68,

Fig. 68.Variation of Atmospheric Pressure with Altitude. Useful in estimating the


mass compressor will deliver at various altitudes. Enter chart at the ordinate,
of air a
which represents altitude. Move rightward to the curve, as along the dotted line, then
downward to the abscissa and read the normal atmospheric pressure at the corresponding
altitude.
p
182 COMPRESSION AND EXPANSION OF GASES [CA 7

and a given mass of air occupies a greater volume It is to be inferred too


that the volume of free air vanes with the temperature In the absence of a
specified temperature, 68°F may be used

117. Capacity and Volumetric Efficiency. The capacity of a compressor is


the actual quantity of gas delivered, as measured by an orifice at intake
pressure and temperature, expressed in cubic feet per minute (cfm) The
actual volumetric efficiency of a reciprocating compressor is the ratio

.
. _ capacity of compressor
Vi .cw.i ’
displacement in cfm

where the displacement computed as explained in § 86, equation (j)


is

The value of the actual volumetnc efficiency, which ranges from G5% to
85%, is obtained only from test of the actual compressor
118
Conventional Volumetric Efficiency An equation for the volumetric
found from the conventional diagram, is useful in estimating
efficiency, as
actual values, but more important, it accents certain factors on which

volumetric efficiency depends The volume of air measured at intake


conditions on the conventional card, Fig 69, is
r, = V, - V 4 , or

Jj „ Xlj Vi
,

'

V~D V

From process 3-4,

Also, Vi = Vd cVp, where cV is the clearance volume Vt and


+ c is the

per cent clearance expressed as a decimal Therefore w e have

7, - Vi Vd + eVo - cVp^/PiY'"

(58)

which the conventional volumetric efficiency


is The actual volumetric
efficiency is different from the conventional volumetric efficiency because of
fluid friction of flow (the pressure in the cylinder is less than the pressure of
free air), and because the cylinder walls, being relatively hot, heat the
incoming hot air can occupy a given space)
air (less mass of Since, in

equation than pi, the volumetric efficiency decreases as


(58), pi is greater
the clearance increases, and as the volumetnc efficiency decreases, the
capacity decreases The clearance may become so large that no air is dis-
charged by the compressor This characteristic is often used to control the
§ 119] PREFERRED COMPRESSION CURVES 133

output of a compressor by increasing the clearance when a reduced output is


desired. Also observe from equation (58) that the volumetric efficiency
goes down as the pressure ratio p 2 /pi goes up. Neither clearance nor
volumetric efficiency is a reliable indicator of quality. The user is most
concerned about the actual power consumed for the desired capacity.

119. Preferred Compression Curves. Inasmuch as the isentropic curve


1-2', than the isothermal
Fig. 70, is steeper
1-2, it takes more work to compress and
deliver the air when the compression is isen-
tropic than when the compression is isother-
mal, the difference being represented by the
shaded area. Compression curves with values
of n between 1 and k will fall within the
shaded area. Thus we see that the work
necessary to drive the compressor decreases as Fig. 70. Comparison of Work
the value of n decreases. Note that between for Isothermal and for Isen-
tropic Compression.
specified pressures, Fig. 70,

Work, isentropic process < work, isothermal process,


Work, isentropic compressor > work, isothermal compressor.

Polytropic compression and values of n


less than k are brought about by
around the cylinder, Fig. 71. The
circulating cooling water, Fig. 65, or air
cooling water or air picks up heat because the work has raised the sub-
stance’s temperature above that of the environment. In small cheap

Courtesy I ngersoll-Rand Co., .\eir York

Fig. 71. Air-Cooled Compressor. This is a small two-stage compressor driven by


an electric motor via a V-belt drive. on the left.
A part of the tinned intercooler is visible
134 COMPRESSION AND EXPANSION OF GASES [Ch 7

compressors the cooling tv ill be inadequate and the value of n will be 1 35


or higher Under favorable circumstances a v alue of n - 1 3 or less may
be expected A alues from l 2a to 1 3 represent the best results for water
jacketed compressors
m
A low \alue of n a compressor is not necessarily desirable The best
compression process depends upon what the compressed stuff is used for
Stuart and Jackson (31) have discussed this question fully TVe note that
the adiabatic process results in an increase of enthalpy (by the amount of
the work done) Hence if the compressed substance is to be used in a
gas turbine for example the available portion of the energy involved is
later available for w ork in the turbine and less heat w ould have to added
in the combustion chamber (Chapter 8 )
Onthe other hand even though the substance is air used driving m
miscellaneous devices the increase of energy of the air does not help beca lse
the air ordinarily loses this energy as heat to the surroundings before it is
used Hence in this situation isothermal compression would be best
thermodynamically In the case of atmospheric air w hich always has some
water vapor in it an actual isothermal compression w ould probably not be
desired even if it could be attained because of the troubles which would
accompany the condensation of vapor which inevitably takes place See
aftcrcooler § 110

120 Example A 14xl5-in double-acting air compressor whose clearance is


4% runs at 150 rpm and takes air at 14 psia and S0°F (po nt 1) Discharge is at
5G ps a after a compression w hich is taken as poly trop c with n = 1 3 The atmos
phene pressure and temperature arc 14 7 ps a and 70°F respectively (a) Est mate
the free air from the convent onal volumetric efficiency For the substance as it
passes througt the compressor determine (b) the heat transferred (c) the change of
enthalpy (d) the convent onal horsepo ver (See Fig GO )
solution (a) Because tl e compressor is double-acting the number of diagrams
completed per m nute is A 2n = 2 X 150 = 300 vhere n rpm The d S
placement is (sizes of eng nes are al vays given
bore X stroke = 14x15)

KD =
Vd (L)(\)
W14W 300) 401cfra
\i (4) (1728)

With a volumetr c efficiency of

1
Fig 69 Repeated - ’
= 92 4%
«= 104 0 04 0 924

we find the volume drawn ra per m nute as


I j (0 924) (401) 370 cfm

measured at 14 psia and 80°F at point 1 If the subsenpt o refers to atmospheric


§ 121} EFFICIENCIES 135
a h*, FG have p.YJT. = pi T \/T\ ; or the approximate volume of free air is

PiV’iT. (14) (370) (530)


Tr
To " = = 346 cfm
,

^>7 ImokujT -

(b) To get the heat, we might first find the temperature after compression.

(n—l)/n
T = - T, = 540 )'" = 744°R =
te) ( 1 284°F.

The mass of air delivered is

_ ViV[ (I4)(144)(370)
= 25.9 lb./min.
RTy (53.3) (540)

Hence, for c, — 0.1715 for air and k = 1.4, and for

_
— (0.1715)(1.4 - 1.3)
=
cn —0.0572 Btu/lb.,
(I - 1.3)
we get

Q = ir'c„ AT = (25.9) (-0.0572) (744 - 540) = -302 Btu/min.,

the negative sign indicating heat rejection.


(c) The change of enthalpy for c p = 0.24 is

AH = w'CpAT = (25.9) (0.24) (744 - 540) = 1268 Btu/min.

(d) The work, equation (54), is

TF = -AH +Q= -1268 + (-302) = -1570 Btu/min.

Using 42.4 Btu/hp-min., we get

hp = = 3 ' hp -
42A

The use of equation (56) will result in the same answer, but having found Q, the
foregoing computation is easier.

121. Efficiencies. The mechanical efficiency of a compressor is

...
_ ihp of the compressor _ Wi
^ v" ~~
bhp of the compressor WB
If the compressor is driven by a steam or internal combustion engine, the
mechanical efficiency of the compressor system is (32)

_ ihp of compressor
*
.

' Vm, —
driving engine
180 COMPRESSION AND EXPANSION OF GASES [CA 7

Tor rough estimation and pedagogical purposes we may estimate the horse-
power (flip) used to overcome friction in the compressor as

(l) fhp = 0 10oVd *'\

where V a is the piston displacement in cubic feet per minute The brake or
shaft horsepower of the compressor may be estimated by adding this fric
horsepower to the indicated horsepower of the compressor
tionftl
The adiabatic compression efficiency ij is the theoretical power required
as obtained from a conventional diagram with isentropic compression
dntded by the actual indicated power of the compressor

m Ve
_ work (or hp) of comcntional card isentropic compression
' '
indicated work (or ihp) of compressor

which is the compression efficiency most commonly used The isothermal


compression efficiency is similarly defined except that the numerator of
(m) would be the work of the conventional card with isothermal com
press* on The overall efficiency is the product of the mechanical efficiency
and the compression efficiency v = Thus the adiabatic overall effi
ctency is

(n) Va « [eq (j) or (k)] [cq (m))


hp corresponding to isentropic compression

(ihp of engine) or (bhp of compressor)

where the denominator depends on whether the compressor is driven by an


engine or an electric motor The ASME Test Code (30) calls the ratio in
equation (n) the shaft efficiency when the denominator is the brake pover

122 Multistage Compression The xolumetnc efficiency is affected


not only by the clearance but also by the ratio of pressures pi/pi [equation
(58)] Thus if the compression is earned out in two or more cylinders
(Fig 72) the volum/itnc efficiency of the multicylinder machine will be
greater than that of a single-cylinder machine of the same clearance and the
same total pressure range Moreover since the air gets very hot when com
pressed to high pressures the final temperatures may be so high as to cause
trouble with the lubrication of the cylinder and piston Furthermore it is
of
possible to sa\e considerable power by using two or more stages instead
one if the final pressure is above about 60 to 100 psi if the installation is
permanent and if the required displacement is greater than about 300 cfm
§ 122] MULTISTAGE COMPRESSION 137
To obtain the saving in power, we use an intercooler to lower the air tem-
perature between stages, Fig. 72.
The following discussion applies to both Fig. 73 and Fig. 74, in which
are pictured the events of the conventional indicator cards of a two-stage
machine, with the high-pressure (HP) card superposed on the low-pressure
(LP) card. Suction in the LP cylinder begins at A and the volume V[ is

Courtesy Ingersoll-Rand Co., Xew York

Fig. 72. Two-Stage Compressor, Electric Drive. The synchronous motor is mounted
on the compressor shaft and therefore does not have separate shaft and bearings. This
drive has a high power factor and efficiency. It is said that about 95% of the output of
the motor is utilized in the cylinders. The intercooler is “two-stage.” Cold water enters
intercooler at HP end, and after circulating through the LP end of the intercooler, goes
to the cylinder jackets. Air flows around baffles; see arrows. Two-stage compressors
are made with the cylinders at an angle (Fig. 71), cylinders at right angles (one hori-
zontal and one vertical), and cylinders in a tandem. In sizes over 1000 hp, this com-
pressor is made in twin tandem style, two more cylinders added opposite to those shown,
but with a combination of LP and HP cylinders in tandem arrangement. Output is con-
trolled by clearance pockets, two on each end of each cylinder. See Fig. 67. Com-
pressors are often driven by reciprocating steam engines on the same shaft.

drawn compression 1-2 occurs, and the LP cylinder then discharges the
in;
air along 2-B. This discharged air passes through an intercooler, being
cooled by circulating cold water. The temperature to which the air is
cooled is governed by the temperature of the cooling water available. It is
relatively easy to bring the temperature of the air to within 20°F of the
temperature of the water. In the conventional analysis, it is commonly
assumed that the air has the same temperature upon entering the IIP
cylinder as it had upon entering the LP cylinder. This assumption places
point 3, the end of the suction stroke E -3 in the HP cylinder, on an isother-
1S8 COMPRESSION AND EXPANSION OF GASES [Ck 7

mal through 1 (Fig 73 and 74)


lino The air is compressed to the final
pressure along 3-4, discharged along 4 -F, after which the clearance air
re-expands, r~E The same mass of air is involved at points 1, 2, 3, and 4,
also the mass of air discharged is equal to the mass drawn into the LP
cylinder, if leakage is neglected and if steady flow has been achieved

V s
(a) (b)
Fig 73, Conventional Cords, Two Stage, No Pressure Drop In (a), 1-M-G-H out-
lines the conventional card for single stage compression to p t for a given per-cent clear-
ance If the per-cent clearance is the same in both cylinders, re-expansion in the high-
pressure cylinder starts at some point F instead of C Re-expansion in the low pressure
cylinder starts at B, where V B — Vo For a two-stage machine, suction starts at A
For a single-stage machine, suction starts at H
The capacity of the two-stage com
pressor Is greater than that of the smgle-stage by the amount Vh — Va Observe that
it is possible to make the clearance so large that no air would be delivered The work
saved by the two-stage compression is represented by the shaded area 2-M-4-3 Only
the thermodynamic processes (not the suction and discharge) are shown on the TS plane
in (b) Isentropic compressions should be 1-a and 3-b If the LP cylinder had an

isothermal compression, the state point would follow the dotted line 1-3.

The equation (h) gives the work of an indicator card like 1-2-B-A or
2-4-F-E, and only for diagrams like these We may therefore apply this
equation to each of these cards, and the total work of the two-stage compres-
sor will be the work of the LP stage plus the work of the IIP stage, thus
= mass

^
(it)' passing through),

(o) IF - - l] + [(gp" - l}
\ r

j 122] MULTISTAGE COMPRESSION 120

Typically, the design ba‘ed on the a— umption that the «ame amount
i-'

of work
done in each cylinder. Thi~ condition al-o happen* to rc.-ult in
i-.

minimum work for compressing a particular man of air.* Thu*, for the

Conventional Cards, Two-Stage, with Pressure Drop. The only chance from
Fig. 74.
Fig. 73 is thepressure drop from 2 to 3 in the intercooler. The cards ore otherwise con-
ventional in all respects; the compression is taken as polytropic. The shaded area
between B and 3 in Cat represents lost (or repeated! work due to the pressure drop.
The process through the intercooler is steady flow but irreversible (therefore shown
dotted/.

rav of 7’ t = 7 and of }>: - J>s = P-. ‘ay, we have (lie work of the first

stage equal to that of the second statr<\ or

nir'/rr, - 1
I - n

from which we find

_ p«. or V‘ ~ (piPtV n
(P). " ,

I Pi V*
of no prr -• *’ ifrf'p !* r **' r\ ij'jrl* r< for
* The, miv Ik’ p'ow-rl for the ;

bttimr ,m equation 'o />; » JU ~ }', r. % r'.ru! !** n,t jv ;ro to


r, - T , i,y J

mimed, Ti I: difh rr ntiato H from f'pta •AH to r ~. iro


*'0 V ith -
be <1/ v f
r* t*i :i,.‘ i

and fj'mtr tic r< ".it to r* *o. Tic ••Mu** of


* tiitm j»
p, ,

~ r
‘1HC that found by 'T 11 '
11 **"IC to tf.nt
p.p,'' th<
(if 4
140 COMPRESSION AND EXPANSION OF GASES [Ck 7

which the proper value for the intermediate pressure for the conditions
is

specified (Fig 73) A pressure drop in the intercooler could be spread on


each side of this ideal value
The conventional cards show very clearly how work is saved, the saving
being indicated by the area 2-JF-4-3, Tig 73(a) Actual indicator diagrams
taken from a two stage machine arc shown in Fig 75
For three or more stages of compression (more than three stages arc not
often used), themethod of analysis is similar to that given for the two stage
machine, and since it is anticipated that the reader wall have no difficulty
in finding the work of the conventional cards, no
further detail will be given here Let p, = the
intermediate pressure between the LP cylinder and
the intermediate cylinder, p„ = the intermediate
pressure lie tween the intermediate cylinder and tho

IIP cylinder, p, = the initial or intake pressure,


and let pi = the final discharge pressure Then,
either by differentiation or by the condition of equal
works we find the optimum values

(q) V* = (P.’P/)'" and p* = (pp/


2)'",

winch arc the \alucs of the discharge pressures


from the low (p,) and intermediate (p») cylinders
for minimum total work of the conventional cards
without pressure drops in intercoolers Three
stages may show a net saving in costs when the
final pressure is above about 150 psi, the precise

transitional pressure being governed of course, by local conditions Tor


very high pressures, many stages are used Tor example, one installation
pumping hydrogen and nitrogen to 15,000 psi uses seven stages

123. Energy Diagram for Two-Stage Compression. It wall be worth


while to keep our thermodynamics in perspective with an energy diagram of
the system which consists of tho air between the entry and exit boundaries
In Fig 70, the subscripts have the same meanings as before, I ig 73 and
74, except that subscript 5 indicates the state of the substance as it leaves

the receiver or aftcrcooler We take each device in the group as an


may
energy diagram and obtain the same work equations as before, after the
method of § 2 12, except that the number for the work will be a positive
number inasmuch as it has l>ccn assumed as flowing in the correct direction
in Fig 70 Similarly, Q is shown in its correct direction To be specific,
the law of conservation of energy applied to the LP cylinder, with &K
=0
as usual, gives

TFlf « h- ft, + Q Btu/lb


§ 124 EXAMPLE
]
m
where the number for each symbol is substituted as a positive number and
Ws.p is positive. Similarly, for the intercooler, we have (AK = 0)
Q 2-3 = hi — h3 and Q 2-3 = Ah = h3 — h 2,
[energy diagram] [conventional]

where the value of Q 2-3 from the energy diagram will be a positive number
since it is taken in its true direction, and the value from the conventional
form [see equation (i), p. 72] will be negative, the sign indicating that heat is
flowing from the system. The transferred heat quantities shown in Fig. 76
are obtained on test by observing the temperatures of the water entering
and leaving each device. With these heats known, we may use the system

Fig. 76. —
Energy Diagram Two-Stage Air Compressor. Any processes, reversible or
irreversible. There is a choice to be made in sketching energy diagrams: whether to
show energy flux (0 and W) in their known directions and handling each one as a positive
number, or whether to show them in conventional directions (opposite to the directions
shown here), and let each one be an algebraic number, in which case the sign obtained
from the basic form of the equation takes care of the situation. If there is any doubt as
to which direction W
or Q flows, use the conventional forms. The best plan is to know
well enough what you are doing so that you axe at home with either idea.

of the air passing through, outlined by heavy dashes, and write the equation
for energy entering equal to energy leaving and get the total work as

W — hi — /ii -f- Qlv + Qi— 3 d" Qnp Btu/lb.,

where the Q’s are reduced to Btu per pound of substance flowing through
the compressor. Since temperature is an easy property to determine
(but not so easy if precision is important), the energy diagram approach
is frequently appropriate in test work. If the substance is an ideal gas,
Ah = cp AT.
124. Example. A two-stage, double-acting, air compressor, operating at 150
rpm, takes in air at 14 psia and 80°F. The size of the LP cylinder is 14x15 in., the
stroke of the HP cylinder is the same as that of the LP cylinder, and the clearance
of both cylinders is 4%. The LP cylinder discharges the air at a piessure of 56 psia.
The through the intercooler and enters the HP cylinder at 80°F and a
air passes
pressure of 53.75 psia, after which it is discharged from the compressor at a pressure
of 215 psia. The value of n in both cylinders is 1.3. Neglect the effect of the piston
rods on the crank end. Atmospheric pressure and temperature are 14.7 psia and
70°F, respectively, (a) For the gas in the intercooler, how much heat is removed
and what is its change in available energy? (b) What should be the diameter of
the HP cylinder? (c) What conventional horsepower is required? (d) Determine

the change of available energy of the gas as it passes through the HP cylinder.
142 COMPRESSION AND EXPANSION OF GASES [Ck 7

sornnov Comparison of this example with that of § 1 20 show s that the Lp stage
in this problem is identical with the single stage of § 120 Therefore from § 120
we obtain the following additional data (see Fig 77)

to' = 25 9 lb /mm T t = 744°R V[ _ 370 cfm


ij,(LP) = 92 4% Ar = 300 cards/min , hp a 37

(a) The heat removed is Q = All (see also Fig 70) and the air is cooled from
744®R to S0°F - 540°R

Q - WcP {Ti - Tt) = (2o 9){0 24) (540 - 744) 120S Btu/min

The change of a\ ailable cnerg j isQ — T.As (equation (51) p 112) ai d Q = Ah


during steady flow through the intercooler (AA. — 0) Thus for the air

F. « ht - A, - r.(«, - Si) Btu/lb

Let T0 = 530*11 the atmospheric temperatuie and As maj be obtained from


T ds = cp dT - (t dp)/J as [see equation (o) p GO]

.
= Tt « pj Fj —
fJ.
As i
c. In s
J t J/
.
In
pt
,
c p In
J t
+ J, In pi
- -0 24 In In j— - -0 0711 Btu/lb-'R

Thence using Aft > All /w' *» —1268/25 9 — 49 Btu/lb

E. = -49 - (530)(— 0 0744) - -9 G Btu/lb

or (— 9 0)(25 9) = -248 5 Btu/min a decrease Oidinatilv there is little or no


opportunity to use this available energy The net loss in the intei cooler is not

V a b cd s
(a) lb)
Fig 77

so great ns computed because the cooling water gains availability However, in


the end the heat ends up in the sink
(b) The volume I » passing through the cvhndcr Fig 77 is that correspond HP
ing to the weight 25 9 lb /min

*
^, = £59X533X540)
Pi (53 75) (144)

The conventional volumetric efficiency of the HP cylinder is


V" ‘

- 1 + c - »(g) - 1 « - 0<M (|%)‘ ‘


- 0924,
=

§ 125] GAS TABLES 1J,S

the same as for the LP stage, because the pressure ratio is the same (as well as c),
56/14 = 215/53.75 = 4. Since the HP displacement is

by definition of volumetric efficiency,


7m
V, S
~ “ 0.924 = io4 2 cfm -

we have

Vd " = Ct~) (|f)


(300) = 104.2 cfm,

from which D — 0.595 ft. =7.14 in., or say, 7)4 in., the diameter of the HP cylinder.
(c) We have already found the conventional power required for the LP cylinder
(37 lip). For the HP cylinder, we have
n -' )/n
np*V\ \(p*\
IF =
J(. 1 - n) LVtV
(1.3) (53.75) (144) (96.3)
—1570 Btu/min.,
(778) (1 - 1.3)

or 1570/42.4 = 37 hp. Therefore, the total horsepower required is 37 + 37 = 74 hp


(Actual horsepower would be of the order of 85-100.)
(d) Since the pressure ratio (4), the value of n (1.3), and the initial temperature
(80°F) are the same in LP and HP cylinders, the final temperature
is the same,

= T« —
744°R, Fig. 77. For 1 lb. and for c„ = —0.0572 (from §120), the
change of entropy of the air during the leversible polytropic process is fc„ dT/T,

— = = —0.0572 = —0.0193 Btu/lb-°R.


s4 s3 c„ In
y In

Now the input of available energy to the air is the work IF (as a positive number);
but some of this input was lost with the transfer of heat Q = —302 Btu/mm
(from §120). The amount of the loss is

Q - TAS = -302 - (25 9) (530) (-0.0193) = -37 Btu/min.

Then the net increase of available energy is IF — 37 = 1570 — 37 = +1533


Btu/min.

125. Gas Tables. were assumed to be constant in the


Specific heats
foregoing discussion. accuracy is important, the effect of the
However, if

variation of specific heat should be included. This can be done: by using


equations for c P such as those in Table II, if they are available; by using a
,

known average value for the temperature range involved, as from Fig. 10;
by using a chart of the properties of the substance (see the h nlhnlpij-L ntropy
Chart for Air in Problems on Thermodynamics'* or by using tabulated proper-
ties of substances. (See Chapter 10.) The Keenan and Kaye G as 7 ablcs
contain tabulated properties of air, of certain products of combustion,

nitrogen, oxygen, carbon dioxide, hydrogen, carbon monoxide, and aho


many other useful tables Since these tables are becoming a common item

* By rnires, Brewer, Siminang


144 COMPRESSION AND EXPANSION OF GASES [CA 7

in an engineer’s library, we
them briefly
shill wntc
Tables IV and
of
herewith are short extracts from the Keenan and Kaye tables The
sy mbols h and u m the air table stand for enthalpy and internal energy as
usual, but they are psuedo-nbsolute \ alues computed from

h ~ c,dT and u = h — j- = h — -j-'


Jo
where the integration is from absolute zero temperature to T Thus, the
JcP dT between limits of 1 and 2 is simply the difference of the values read
from the table h» — hi All the other symbols in Table I\ represent point
functions too The sy mbol c illed the entropy function, is defined by

- cp dT
4>
/: T
The \ alue of giv en in the air table unlike ^ as gi\ en in all the other tables
isthe basic value minus one If one \aluc of £ obtained from the table is
subtracted from another the difference is the definite integral from 1 to 2 of
JcpdT/T Again from T ds =» cp dT — vdp/J, equation (29), we ha\e
2
ep dT R.Pi R, n Ps
f
-V- jin-- jta-
. , ,
(0 4,
-i,
Thus if the change of entropy between any two states is desired determine

it inaccordance with equation (r) (There is a table of values of the log


term in Iveenan and Kay e ) For a constant pressure process only As - A t>
If the entropy is constant we get from equation (r)

- = or = j In pn
4>i
j In <f>

(entropy constant]

the defining relation for p r which is called the relative pressure,

p, = antilog ~ = e*«*
All values of p, in the tables are ones obtained from the foregoing equation
modified bv a constant factor in order to have a convenient range of num
bers m the end For an lsentropic process.

(s) Ei,
Pi

where the subscript s is a reminder that the entropy is constant for this
,

§ 126] EXAMPLE H5
relation. The value of the relative volume v in the tables is computed from

(t)

where the units of R were taken so that v r is the specific volume in cubic feet
per pound when p r is the pressure in psia for tables based on 1 lb. From
equation (t), we see that

the constant entropy carries over from the definition of p r .

If the values of the foregoing properties of air are desired on a mol basis,
multiply by 28.970 (use 29 for slide-rule work), the equivalent molecular
weight of air. The student may interpolate in our tables, but he should
realize that the interval between tabulated values
is too large for accuracy.
While, as stated, these various properties are accurate only at low pressures
(ideal gases), the error involved for pressures of several hundreds of psi is
negligible. Use Tables IV and V in accordance with any of the fundamental
energy equations previously presented.
The properties_of the other substances given in other Keenan and Kaye
tables are on a pound-mol basis (h Btu/mol in Table V, for instance).
Thus, an enthalpy value in Table V should be divided by the corresponding
molecular weight if it is desired in Btu per pound.
The internal energy in Btu per mol is found from the definition of enthalpy
h = u pv/J = u
+ HT Btu per mol, where R = 1.986 Btu per mol-°R.
+
Thus, to get the internal energy of the gases from Table V, subtract the value
given in the last column RT from the enthalpies h at the
same temperature. T

126. Example. Let one pound of air be compressed adia-


batically and in steady flow from 14 psia and 525°R with a
compiession ratio of 5. The compression efficiency is 75%.
Determine the work done, the discharge temperatuie and
pressure, the increase in entropy and in unavailable energy
for a sink temperature of 500°R. What value of m
in Tz'/Ti = (Fi/F 2 ') m_1 defines the relations of the end
states?
solution. The actual work of this problem is obtained by
first finding the isen tropic work and then applying
1-2, Fig. 78, Fig. 78.

the efficiency. By interpolating in our Table IV, the reader


should check the table values used below, which are taken diiectly fiom Keenan
and Kaye and are therefore somewhat more accurate. We have at 525°R,

J„ = 125.47 Btu/lb., Pn = 1.2560,


t> rl = 154.84, <t>
= 0.59403.
. ,, .

Table V. ENTHALPY BTU/MOL ) OF GASES AT LOW PRESSURE


(.

Taken with permission from Keenan and Kaye, Gas Tables, John Wiley; values
of h Btu/mol; Tt = 30
1.9S6 Btu/mol- R; 53G.69 R = 459.69 + 77°F.

Prod* Prods. . .

Temp. Ni O: HtO CO: Ht CO 400 '"'c 200 ^ rIJ


°R 23 016 32 18 016 44 01 2 016 28 01 Ideal Ideal ~ HT

500 3 ,472 2 3 .466 2 3,,962 0 3 .706 2 3,386 1 3.472 1 3,.486 7 3 ,511 2 993
520 3,,611 3 3,,606 1 4.,122 0 3 .880 3 3.523 3 3,611 2 3 .627 4 3 ,653 7 1 ,033
536 69 3,,727 3 3,,723 0 4.,255 8 4,,027 5 3.638 1 3,727 3 3,,744 G 3,.772 7 1 .067
540 3,,750 3 3,,746 2 4, 282 4 4,,056 8 3.660 9 3.750 3 3.,768 0 3,,79G 3 1 .072
560 3,,889..5 3,,880 C 4.,442 8 4 .235 8 3.798 8 3.8S9 5 3 .909 2 3 .939 4 1 .112

5S0 4,,028 7 4,,027 3 4.,603 7 4 .417 2 3,037 1 4,028 7 4.,050 4 4 .082 7 1 .152
600 4,,167 9 4,,168 3 4.,764 7 4 ,600 9 4,075 6 4.16S 0 4.,191 9 4 .226 3 1 ,192
700 4,,864 9 4,.879 3 5,,575 4 5 .552 0 4,770 2 4.S6G 0 4,,901 7 4 ,947 7 1 .390
£00 5,,564 4 5,,602 0 c,,396 9 6 ,552 9 5.467 1 5,568 2 5 ,G17..5 5 ,G7G 3 1 .589
900 6,,268 1 6,,337 9 7,,230 9 7 ,597 6 G, 165 3 G.276 4 6,,340,.3 G .413 0 1 ,787

1000 6,,977 9 7,,0S7 6 8.,078 9 8 .682 1 6,864 5 6.992 2 7 ,072 1 7 .159 8 1 ,9S0
1100 0 7,,850 4 8.,942 0 9 .802 6 7,564 6 o .185
7,,695 7,716 8 7,,812 9 7 ,91G 4
1200 8,,420 0 8,,625 8 9.,820 4 10 ,955 3 8,265 .8 8,450 8 8,,563 4 8 ,GS3 G o ,3S3
1300 0,,153 9 9,,412 9 10, 714 5 12 .136 9 8,968 7 9.194 G 9,,324 1 9 ,461 7 2 ,5S2
1400 9 ,896 9
, 10,,210 4 11.,624 8 13 .344 7 9. 673 8 9,948 1 10,,095 0 10,,250 7 o .780

1500 10,,648 9 11,,017 1 12.,551 4 14 .576 0 10,381 5 10,711 1 10,,875 G 11 .052 o 2 .979
1600 11,,409 7 11,,832 5 13,,494 9 15,,829 0 11,092 5 11.483 4 11,,665 6 11,,859 6 3 .178
1700 12,,178 9 12,,655 6 14. 455 4 17,.101 4 11,807 4 12.264 3 12,,464 3 12,.678 G 3 .376
1800 12,,956 3 13,,485 8 15. 433 0 18,,391 5 12,520 8 13.053 2 13,.271 7 13 .507 0 3 .575
1900 13,,741 6 14,.322 1 16, 427 5 19 .697 8 13.250 9 13.849 8 14,,087 2 14 ,344 1 3 ,773

2000 14,,534 4 15 .164 0 17.,439 0 21 .018 7 13.9SO 1 14.G53..2 14,,910 3 15 ,189 3 3 .972
2100 15,,334 0 16,,010 9 18.,466 9 22 352 7 14.714 5 15.463 3 15,,740 5 16,,042 4 4 .171
2200 16,,139 8 16,,862 6 19.,510 8 23 ,699 0 15.454 4 16.279 4 1G,,577 1 1G,,902 5 4 .309
2300 16,,951 2 17,,718 8 20.,570 6 25,,056 3 16.199 8 17,101 0 17,,419 8 17,,769 3 4 .5GS
2400 17,,767 9 18,,579 2 21. 645 7 26,,424 0 16.950 G 17.927 4 18,,268 0 18,,642 1 4 ,766

2500 18,,589 5 19,,443 4 22. 735 4 27 ,801 2 17,707 3 18.758 8 19,,121 4 19,,520 7 4 .965
2600 19,,415 8 20,,311 4 23,,839 5 29,,187 1 18.469 7 19.594 3 19.,979. 7 20.,404 G 6 .164
2700 20,,246 4 21..182 9 24. 957 2 30,,5SI o 19.237 8 20.434 0 20. 812 8 21.,293 8 5 .362
2800 21,,031 1 22,,057 8 26, OSS 0 31 982 8 20.011 8 21.277 2 21. 709 8 2° ,187 5 5,,561
2900 21,,919 5 22,,936 1 27.,231 2 33.,391 5 20.791 5 22,123 8 22 531 4 23!,086 0 5,.759

3000 22,,761 5 23 .817 7 28.,3SG 3 34,,806 6 21.576 9 22,973 4 23. 456 G 23.,9SS 5 5,,958
3100 23,,600 8 24 ,702 5 29.,552 8 36, 227 9 22,367 7 23.820 0 24. 335 5 24.,895 3 G,,157
3200 24,,455 0 25,,590 5 30. 730 2 37. 651 7 23.164 1 24 681 2 25. 217 8 25. 805 6 6,,355
3300 25,,306 0 26.,481 6 31.,918 2 39,,0S6 7 23,965 5 25.539 0 26. 102 9 26.,719 2 6,,554
3400 26,,159 7 27,,375 9 33. 116 0 40.,523 6 24,771 9 26,399 3 20, 991 4 27 636 4 0,,752

3500 27,.015 9 23 ,273 3 34,,323 5 41 .965 2 25.5S2 9 27.2G1 8 27, 8S2 9 28, 556 8 6,,951
3600 27,.874 4 29 .173 9 35 540 1 43,,411 0 26.39S 5 28.126 G 29. 479 9 7 ,150
3700 28,,735 1 30,,077 5 36,,765 4 44.,860 G 27.218 5 28.993 5 30 406 0 7 ,348
3800 29,.597 9 30,,981 1 37. 998 9 46, 314 0 28.042 8 29.862 3 31, 334 8 7 517
3900 30 ,462 S 31 .893 6 39.,240 o 47,.771 0 2S.871 1 30.732 9 32 266 o 7 ,745

32 .806 1 49,,231 4 29,703 5 31.605 o 33. 199 6 7 ,944


4000 31 .329 4 40.,489 1
4100 32 ,193 0 33 .721 6 41.,745 4 50,.695 1 30.539 8 32,479 1 8 143
4200 33 ,068 1 34 .639 9 43.,00S 4 52,.162 0 31,379 8 33.354 4 8. 341
4300 33 .939 9 35 ,561 1 41,,278 0 53,,632 1 32,223 5 34.231 2 8 510
4400 34 .813 1 36 ,485 0 45.,553 9 55,,105 1 33,070 9 35.109 2 8. 738

4500 35 6S7 8
. 37 .411 8 46.,835 9 56 ,581 0 33.921 G 35. OSS 6 8. 937
4600 30 .503 8 33 .341 4 48, 123 6 5S 059 7 34.775 7 36.869 3 9 136
4700 37 .441 I 39 .273 G 49.,416 9 59,.541 1 35.633 0 37,751 0 9. 334
4800 38 319 5 40..203 6 50. 715 5 G1 024 9 36 403 4 3S.633 9 9. 533
4900 39 ,199 1 41 ,146 1 52. 019 0 62 511 3 37.356 9 3Q.517 S 9. 731

5000 40 .079 8 42 ,0S0 3 53.,327 4 64.,000 0 35.223 3 40.402 7 9. 930


5100 40 .061 C 43 .029 1 54 510 3 65 490 9 39 002 8 41 2SS 6 10. 12^
5200 41 SSI 4 43 .974 3 55. 957 4 66 981 0 39 965 1 42.175 5 10. 327
5300 42 ,72S 3 44 .022 n 57 278 7 68 479 1 40 810 2 43 063 2 10. 526
5100 43 .612 9 45, S72 r
!
5S. 603 9 69. 975 9 41.717 7 43.951 S 10. 724
,

148 COMPRESSION AND EXPANSION OF GASES {Ck 7

For a compression ratio of 5 during an iscntropic process,

©.=
Locate this value of i,j in the air table and read other needed properties at point 2
" inch is on an iscntropic line, Fig 78

t iL'n Tt = 9S9 5°R,* ht = 23S 37 Btu/lb , p„ = 11.837

Then, from equation (54), $ = C,

IF. = A, — A, = 125 47 - 23S 37 =* -112 9 Btu/lb,


work done on the air Also from equation (54), the actual adia
batic work is h, — hr, Fig 78, from the definition of adiabatic
compression effieienej (§ 121), the actual indicated work is
IF' =* (Ai — A»)/ij t Fquating these two \alues of the actual
work, v» c ha\e

Ftg 78 Repeated
.
hi - I
hr ^
h\ — — hi1
*=
112 9
0?5 -
in- 17
12o 47 - ?
hr,

from which hi = 275 97 Btu/lb For this \alue of A, we interpolate to find

Tt = 1139 6°It and = 0 78317

Now we ha\e from pi/T = C (keep the stated compression ratio h/t t = 5)

£
T
p.tm
tTr
or - TOS! .
T \Vr
(I'30 6)(5)(14)
525
_ , 6, g

the discharge pressure Note that the relative pressure at state 2', Fig 78 is

irrelevant because 2' is not on the i«entropic line ftom 1 (However, we can find

the pressure at 2 from

” 14 " 132 p " a


f, % (ni) '

which is pi for an isentropic compression ratio of 5 ) The increase in entropj to 2


equation (r) is

As •* - - ?ln ££ = 0 78317 - 0 59403 - = 0 0259 Btu/lb °E,


<#>*• 4>i
J pi 778 H
and (§ 9S) the increase in unavailable energy which is that portion of the 100%
available energj (work) input that becomes unavailable, is (T. — 500°R, Fig 78),

T. As = (500) (0 0259) = 12 95 Btu/lb

In order to find m for the given compression ratio of 5, we have

Tr = (U)" 1
_ 11396 _
r, \VrJ 525

* It will interest you to note tliat for the same compression ratio (5) and the same low
temperature (525°R) the high temperature is 1000°R § 89, compared
with 989 5 K m
here Constant specific heats and k = 1 4 were used in 5 89
§ 127] CLOSURE 149

or rn = 1.4S1. entropy change were computed from equation (r), § 71, for
If the
this value of m,would be somewhat different from 0.0259 found above, because
it

c„ is taken constant in equation (r). However, if mean values of c, and k for the
particular temperature range are used, the foregoing answer for As will be closely
approximated. Similar remarks are appropriate with respect to the work computed
from equation (e) of § 112.

127. Closure. The reader should fix in mind the difference between an
engine diagram on the pV plane and a thermodynamic cycle. The intake
and exhaust of an engine may be shown on the pV plane because changes of
pressure or volume, or both, are involved. A constant pressure intake is

not a constant pressure heating. no more than a


Idealized, the intake is

change of location of a body of substance with no change of thermodynamic


properties. While it is important to understand what an engine diagram is,
be sure to observe how the law of conservation of energy is simply applied to
any engine or process (such as the heat exchanger).
It may have occurred to the reader that the air compressor plus the air
engine plus the atmosphere (sink) plus another heat reservoir (source) can
be combined to form a thermodynamic cycle. Such a cycle is actually used
for refrigeration purposesand for the purpose of manufacturing power,
depending upon the way the heat flows during certain processes. An air
cycle for developing power, the gas turbine, is the subject of the next

chapter. For additional information on fans, compressors, and blowers, see


references (31 to 36 )
8

THE GAS TURBINE AND


TURBO-JET

128 Introduction The first gas turbine to produce useful work was
probably a w indmill w herein there is no prccomprcssion and no combustion
The we think of the name today
characteristic features of a gas turbine as
include a compression and a heat addition (or combustion) process
process
These features are not new although a practical machine is a relatively
recent development Joule and Bray ton* independent!} proposed the c\ cle
which is the ideal prototype of the actual unit An unsuccessful turbine
unit was built as far back as 1872 and by 190b a unit winch produced net
power had been built (39) There were two principal obstacles to be over
come as revealed by thermodynamic analysis Such an analysis showed
that in order for practical amounts of power to be deh\cred (1) the tern
perature at the beginning of expansion must be high (until a few years ago
the highest permissible temperatures were about 700 S00°F) and (2) the
compressor must operate at a high efficiency Metallurgical developments
in recent 3 cars (for example the use of an expensiv e alloy of cobalt chro-
mium nickel for the turbine blades of the J47) are raising the highest per
missible temperatures (sometimes lfiOOT and more 2000°F if short life is
acceptable as for some military purposes) A better knowledge of aero-
dy namics has been responsible for improving the efficiency of both the centnf
ugal and axial flow compressors An axial flow compressor is much the
reverse of a turbine but it takes a more precise knowledge of aerodynamic
characteristics to design an efficient compressor (38) Gas turbines driven
by the exhaust of internal combustion engines have long been u«cd for super
* See footnote
p 16 George Bray ton a contemporary of Otto vas a Boston eng
necr However his engme was a reciprocating type rather than a turbine
150
§ 130] AIR STANDARD BRAYTON CYCLE 151

charging such engines (turbo-superchargers). Of further promise on the


temperature problem is the use of ceramic-metal combinations.

129. Operation of a Simple Gas Turbine Power Plant. Air enters the
compressor at condition 1, Fig. 79. After compression, it enters the com-
bustors, some of it going around the outside of the combustion chamber
proper and the remainder furnishing oxygen for burning the fuel, which is
continuously injected into the combustion chamber. Because of their
temperature rise, the gases expand (Charles’ law) and enter the turbine in
state 3, Fig. 79. After expansion through the turbine, the exhaust to the
atmosphere is in some condition 4. In an ordinary power plant arrange-
ment, the work of the turbine TF, is great enough to drive the compressor
1 F„ and deliver brake work W
B to drive, say, a generator or propeller;
TF, = TFb -f TFC An .external source of power is needed to start a gas
turbine unit. Figure 80, a cutaway of a turbo-prop engine, shows in some

Fig. 79. Diagrammatic Layout of a Gas Turbine Unit. The temperatures and pres-
sures given are typical of actual values.

detail the actual appearance of the simple gas turbine plant, where the
excess of power produced by the turbine drives the propeller (instead of a
generator). Additional driving force (thrust) is obtained in this applica-
tionfrom the change of momentum of the gases leaving the tail pipe as
compared with the entering momentum of the air (perhaps 15 to 20%
“jet” propulsion).

Standard Brayton Cycle. In Fig. 79, let the combustor be


130. Air
changed to a heat exchanger and let the same amount of heat be added
to the air at constant pressure from an external reservoir of
heat. Then,
let the exhausted air at 4 be led through another heat
exchanger via which
the
heat is transferred to the sink. Following the dotted path in Fig. 79,
same air, cooled to its original intake temperature, now re-enters the com-
pressor at 1 and starts the cycle anew. This is a closed cycle, spoken of as
the equivalent air-standard cycle, and it is pictured on the pV and
TS planes
Such closed systems for a gas turbine, with air and other work-
in Fig. 81.
ing substances, have been built and operated, but the open system is cur-

rently receiving the more attention. In the ideal cycle, the compression
rejection 4-1
1-2 and expansion 3-4 are isentropic; the heat supply 2-3 and
Courtesy Pratt it IJ hitney Aircraft East Hartford Conn

Fig SO Turbo prop Engine The compressor is a 13 stage anal flow, the turbine a
3 stage, 4200 rpm Turbo prop engines are well adapted for airplane speeds of some
400-450 mph and for "middle” distances (up to about 1500 mi ) The anal flow com
pressor presents a smaller frontal area for a particular capacity than the centrifugal type-
advantageous on an airplane

definition of tho pressure ratio rp = pi/p (§ 87) and the


t
Tp relation for an
iscntropic, w e have

whence Tt/Ti = T 3 /T* Rearranging we get

T< _T t Ti _ , Tt _ Ti - Ti = T, - Tt
h T, °r 2, It T
3
Ti

T, - Tj T,
W Ti — T t
Ti
ft Ts

Using this relation in equation (b), we get the thermal efficiency of the
Brayton cycle as
§ 130} AIR STANDARD BRAYTON CYCLE 153

From this equation, we may conclude that the efficiency of the Brayton
cycle increases as T« increases and as T decreases. Now let the compression
ratio for isentropic compression be rt — Vi/V i, by definition (§ 87). Then
the TV relation for the isentropic is

Using rp from equation (c) and rt from equation (f), in equation (e) we find

1
(g) 1 - t
_ =
— 1 - ( t-D/i
ri:

The pressure ratio rp is more commonly used for gas-turbine units .than is

the compression ratio rt . Equation (e) reminds one of the Carnot efficiency,
but it is not so. A Carnot cycle operating through the same temperature

Fig. 81. Brayton Cycle. Also known as the Joule cycle.

limits, T z and Ti, Fig. 81, has a greater efficiency, (Tz — Ti)/T z . An
examination of equation (g)suggests that to improve the thermal efficiency
of the gas turbine, it is necessary to increase the compression ratio, a sur-
mise which is but which must be qualified
strictly true for the ideal cycle,
for actual cycles, as we shall see later. which theOne of the facts of life
gas-turbine engineer must contend with is a temperature ceiling, as previ-
ously stated. With this limitation, we would not likely choose the most
efficient Brayton cycle, because the mep and the amount of work obtainable
from a particular size of engine affects the decision. In Fig. 81(b), let the
cycle under consideration be 1-d-e-f, wherein the compression ratio (1 to
d) is large and the efficiency of the cycle is approaching the Carnot efficiency.
Although diagram is not drawn to any particular scale, it is easy to see
this
that the is becoming quite small.
work done Also, if the compression ratio

decreases, as to 1 -a, the work of the cycle l-a-b-c becomes quite small (and
the efficiency is lower too).
166 THE GAS TURBINE AND TURBO JET [CA 8

and rearrange to get

~ Tir «-W/Ve \( _ 1 \
\7t - 1 ,
- T, (//*-«'* - \
1 )/„«/ r„(l
-'wk
J
Notice that the last parentheses enclose a term v Inch is the efficiency of an
ideal Brayton cycle with a pressure ratio
of r p Tlic purpose of arriving at this
form is to show that the efficiency of the
actual cycle depends upon the high and
Ion temperatures as n ell as on the pres-
sure ratio If it is desired to know the
maximum efficiency for a particular tem
perature range, differentiate the fore-
going equation with respect to r„ and
equate to zero There is little or nothing
to be done about Ti, so let it be say,
540°R N on for a particular turbine in
,

let temperature (say, 1200°F = lGGOTt),


assume various pressure ratios and plot
Fig 83 Efficiency vs Pressure a cur\e You "ill find a curve such as
Ratio (39) Each curve is for a par-
the one labeled 1200T in Fig 83 which
ticular value of Tt , as labeled, and for
the simple cycle corresponding to points up an important fact For each
Fig 82 Combustor efficiency is turbine inlet temperature, there ts a certain
100 % pressure ratio which results maximum m
thermal efficiency the actual thermal efficiency does not go up
that is

indefinitely with pressure ratio as in the case of the ideal cycle See the
dotted curve in Tig 83

134 Example The intake of the compressor of an air standard Brayton cycle
is 40 000 cfm at 15 psia and 40T The compression ratio r* = 5 and the tempera
tnre at the turbine inlet is 1440°F h-eglect the pressure drop between compressor
and turbine and let the exit pressure of the turbine be 15 psia (a) For the ideal

cycle determine the net horsepower output and the thermal efficiency (b) For

the case of an engine efficiency rj, = 85% and a compression efficiency r)e — 83%
(these are almost as high as they go) compute the net output and the thermal
efficiency What is the percentage reduction in power? {<0 Determine the mep
of the ideal cycle (d) What is the available energy in the exhaust air from the ideal
cycle with respect to a sink at p 0 = 15 psia and Ta = 500°R?
solution Refer to the ideal cycle 1 2-3 4 Tig S4 (See also Fig 82 ) In
the compressor the high temperature is not so high but that we may assume that
k »
1 4 as for cold air Thus the pressure at the end of compression is (answers
within brackets are as obtained from Keenan and Kaye (14) —check them from
Table IV]

Pt = Pl = (15) (5)
1 * = (15) (9 5) 142 5 psia [142]
§ 134 ] example 157
and we note that the pressure ratio is r v = 9.5 [9.45], See Fig. S3. This problem
may now be solved by using the gas tables, by using the air chart (in the problem
book), or by using the equations and comput-
ing desired property changes. If the proper-
ties are computed, average c p and k for the
approximate temperature range (or variable
specific heat) should be used. (However, the
cold-air values would illustrate the thermody-
namic procedure, if one wishes to practice
using thermodynamics the easiest way.) We
Fig. 84.
shall use the air chart as far as possible and
check by the Gas Tables [indicated by brackets]. The work of the compressor is

W e = -(h. - h t) = -(227 - 119) = -10S Btu/lb. [-107.7]

The turbine e\-pansion is from 142.5 psia [142] and 1440°F = 1900°R (point 3,
Fig. 84) to 15 psia.

IF, = h, - h, = 477 - 255 = 222 Btu/lb. [222.6]


TFMt = TF = 222 - 108 = 114 Btu/lb. [114.9]

For a mass of air of

pF = (15)(144)(40,000)
3240 lb./ min.,
RT (53.3) (500)

we find the net ideal horsepower as

hr = Qxm® = 4z.4
8710 [8790].

The energy chargeable against the air cycle is (Fig. S2)

Q.< = h, - hz = 477 - 227 = 250 Btu/lb.;

hence, c = W/Qa = 114/250 = 45.6%. [46%]


(b) The actual works and the corresponding net horsepower output are

-10S =
IF' = = -130 Btu/lb., IF,' = (0.85) (222) 1SS.7 Btu/lb.
0.83

IF' = 1SS.7 - 130 = 5S.7 Btu/lb., hp = = 4490 lip.

The actual work delivered by the shaft would be of the order of less than this 2%
value —allowing for the mechanical efficiency. To get the actual thermal efficiency,
we need the enthalpy at state 2', Fig. S2, in order to compute Qa — ht — hy.
From the value of 11/ above,

IT/ = hi - hi = -130, or hi = 119 + 130 = 249 Btu/lb.


IF' 5S.7
,
d~ ~
Q\ 477 - 249

= 114 - 5ST
Perccntace loss 48.5%.
111
158 THE GAS TURBINE AND TURBO-JET [Ch 8

This large loss from the work of the ideal cjcle ignores losses from incomplete
combustion and from the pressure drop through the combustor and other passages
(c) The ‘displacement" of the ideal cj cle is the largest volume minus the smallest
Vi —Vt To keep the numbeis small, use one pound, i t — » 2 Hence,

- = 12 34 cu ft /lb , vt = 2 468 cu ft /lb,

[26),

where in computing i, the \alue of T t has been estimated from the air chart as
1070°Il [1055] The mep of the ideal cjcle Fig 84, is

W IF (114) (778)
25 8 psi
P-i
VD l« — \ 2 (2G4 - 247)(144)

Compare with mep


obtained for the Carnot cjcle (11 psl, § 89) with the same
the
compression ratio Of course the actual mep which is a hypothetical number fora
turbine unit inasmuch as a turbine does not provide an indicator card, is only some
51 5% ol the ideal [(58 7/114){25 8) = 13 29) Because of otherlosses in the actual
engine not considered in these calculations, the brake mep of the engine would be even
less than 13 29 psi This low \ alue of the mep can be tolerated an airplane engine m
onlj because of the high rotative speeds which may be used, 4200 rpm for the turbo-
prop engine of Tig SO 7000 rpm and more for a turbo-jet engine Fig 92 Some
gas-turbinc units turn 36 000 rpm some faster The sometimes close agreement
between the answers obtained from the chart and those from the gas tables is
fortuitous because the small chart cannot be read accurately
(d) The heat that would be rejected to the sink at T 0 = T, from state 4 Fig 84,
is h 4 - h. = h, - h u of which Ta (u - s.) - Ta (s ( - s,) is unavailable (§98)

(The sink state has been taken the same os state 1 ) Therefore, the available por-
tion of the rejected heat is

E = a hi ~ h, — T.(s { — s 9)

The entropy term is approximately

s, - So = e„ In ~ = 0 24 In = 0 182 Btu/lb-
a
It [0 1814]

Since h t is the same ns h we have t

Eo = 255 - 119 - 500 X 0 182 = 45 Btu/Ib

The high temperature of the exhaust and this availability suggests that something
further might be gained from the energy of the exhaust, although it is unlikely that
more work can be obtained directly from this energy However, see § 137.

135 Heating Value of Fuels. The air standard is an invaluable tool m


cycle analysis, inasmuch as the effects of changes in variables can be
easily
answer,
appraised If the effect is reduced to a number or percentage, the
while not being precise, is often closely indicative of w hat actually happens
cycle
because the actual cycle vanes somewhat in proportion to the air
§ 136] THE COMBUSTION PROCESS 159

However, there a question as to how much energy to charge against the


is

actual engine in obtaining its thermal efficiency the higher heating value—
or the lower heating value.
The healing value of a fuel is that amount of heat given up by the products
of combustion on being cooled to the initial temperature after complete
combustion at constant pressure (or constant volume). This is not a
single simple number because of the different ways in which the test may be
run and because of H
2 0, which is formed from some fuels. Fuels used are
commonly hydrocarbons, such as fuel oil, kerosene, and gasoline, the chemi-
cal formula of which is in the form CJI,,. When these fuels burn (react with
oxygen), the hydrogen forms 2 0. H
If the products of combustion are hot
(above about 125°F), this H 2 0 is vapor (steam); if the products have been
cooled to normal atmospheric temperatures, the H 2 0 is condensed, or
largely condensed, and the H 2 0 is water. During condensation it gives up
the latent heat of evaporation. Thus, considering this factor only, we see
that there may be at least two heating values for fuels containing hydrogen,
the higher heating value qh when the H 2 0 formed from the fuel is condensed,
and the lower heating value qi when the fuel is burned so that the H 2 0
does not condense/ Since tests are run sometimes at constant volume and
sometimes at constant pressure, this gives two more heating values.
In the actual engine, the exhaust gases are quite hot and the steam does
not come close to condensing. Since this is so, it is reasoned that it would
be unfair to the engine to charge against it the higher heating value; hence,
the tendency is to use the lower healing value at constant pressure in computing

the thermal efficiency of actual gas-turbine engines. (The practice in this


country has been to use the higher heating value for the same purpose in
computing the thermal efficiency of internal combustion engines, but there
is some tendency now to use the lower. With such a confused state of
affairs, courtesy demands that the kind of heating value used always be
stated.)

136. The Combustion Process. For pedagogical reasons, we shall adopt


a simple approach to the combustion process at this time/ As a close
approximation, we may take the lower heating value q, at constant pressure
as the energy to be released by the ideal chemical reaction. Thus, for
iv/ lb. fuel per lb. air and for one pound of entering air, Fig. 85, we have

(o) iv/qi + h, = (1 + W/)h p + Q Btu/lb. air, [Arritox 1

where h, lb. of air is the initial enthalpy of the air, h P Btu per lb. of
Btu per
products the enthalpy of the departing products, both measured from the
is

same datum as the heating value qi which is usually 77°F and where Q is the
heat transferred (shown as positive if departing). In this application, (he
value of is small enough that the combustion can often be considered ns
Q
* Sec the author's Thermodynamic* (Chapter 13) for a little more detail.
Q

160 THE GAS TURBINE AND TURBO JET [Ch 8

adiabatic Q - 0 The approximation invoked in (o) relates to the fact


that the energy entering with the fuel is wf (Ec hf), in which Ec may be +
called the stored chemical energy and hs is the enthalpy of the fuel its m
entering state Since hf is relatnely small w,(E e hf) is closely approxi +
mated by ui/jj Thus for adiabatic combustion (Q =0), we have

+ W/)hp — A, =
^ (P)

For the
(1

air standard (reactants


Wfiji

and prod
*7*1
ucts the same), one could ignore the
h (l
quantity of fuel and use the further

Fig 85 Simplified Energy Dio approximation


gram — Combustor
(q) Ah — fcp dT = Wfqi

The foregoing equations implj that all the fuel is burned and that all of the
energj released appears in the enthalpy of the outgoing products, that is
that the combustor cfficicncj ij/ is 100% Me may define combustor
efficicncj as

r \ _ actual energy taken on by the gases (IIp — II,)


' ' ti>
energj released during ideal and complete combustion

Reasonable approximations are in order The actual heat added in the air
standard ejele of Fig 82 is h t — h t (mass of fuel neglected) If the ideal

energy released is to,qi then the efficiency is 17/ = (A* — Aj )/v>/qi If the

combustor efficiency is 17/ and if the combustion process is adiabatic the


result is thatunburned fuel in the amount of (1 — v/)v>t passes out of the
combustor with a heating value of q Btu per lb fuel thus an energy term
t

leaving the combustor should be added to Tig 85, the amount of m


Unburncd energy of fuel «= (1 — v/)u>fifr Btu/lb air, 10 0]

and the mass 1 + W/ times h p should be modified accordingly that is A, is


multiplied by 1 plus the mass of fuel actually burned The combustor
efficiency at rated loads or thereabouts should be greater than 95% (38)
We may now w rite the actual thermal efficiency as

_ actual work TP
energy chargeable against the cycle, E c w/qt

w'here W/ is the actual amount of fuel supplied If TP = IT's the brake

work the result would be the brake thermal efficiency If w/ is in lb

fuel per hp hr then v)/qi Btu/hp hr is called the heat rate, and

_ 2544
C ” (w/ffi)

The pressure drop in the combustor should fall between the limits of 2%
to 10% of the entering pressure
§ 138] EFFECTIVENESS OF REGENERATOR 161

137. Regenerative Heating—Ideal Cycle. Having observed that the


temperature of the turbine exhaust 4, Fig. 86, is higher than the temperature
at the end of compression 2, we might happen to
think of applying Ericsson’s notion of regeneration.
In this event, the exhaust gas at 4 and the dis-
charged air at 2 could each be led to a heat
exchanger (regenerator) so that the hot exhaust 4
gives up heat to the air 2. (See the regenerator in
Fig. 87.) Theoretically, if the heat exchanger were
large enough and the flow were slow enough, the air
from the compressor could be heated reversibly
to temperature 4 at state b, Fig. 86, while the
exhaust cools to temperature 2 at state a (§ 90).
Some of the formerly discharged heat hi — h a is
exchanged within the system and the heat to the
sink is now h a — hi. Moreover, it is necessary to Fig. 86. Regeneration
add only the heat equal to h, — hi,, instead of —Ideal Cycle. With the
h% — hi as formerly. Consequently, less fuel is s axis at absolute zero,
area i-a-e-f = 2-b-e-d,
needed and this additional piece of equipment each representing heat.
should materially increase the efficiency of the ideal
cycle, which it does. From Fig. 86, we find the thermal efficiency as
(IF = 2Q, Tt = T it and Ta = Ti)

2Q c p (Ti — Ti) + c p (T i — Ti)


- Ti - Ti
= 1
Qa cp (Ti - T h) Ti - T t

(S)
_ . _ Ti ( Ti/Ti - A = Ti
Ti\ 1 - Ti/Tj

~
where we have used Tb/Ti = T,/T< = rpa J)/h [equation (c), § 130]. With
a fixed initial temperature T h equation (s) shows that with a regenerator,
the thermal efficiency increases as T, increases and decreases as the pressure
ratio increases. Note in contrast that without the regenerator [equation
(g)[ the cycle efficiency increases as the pressure ratio increases. With
regeneration, the cycle 1 -a-b-c, Fig. 81, approaches the Carnot efficiency.
Regeneration is impossible with cycle l-d-e-f, Fig. 81.

138. Effectiveness of Regenerator. Study Figs. 87 and 88 until the


regenerative action is clearly in mind. The state points with the prime
marks indicate actual points, except that pressure drops have not been
shown. There will certainly be a pressure drop of both the air and the
exhaust gases in flowing through the regenerator, and it is important to keep
this pressure drop small. The resistance of the regenerator to flow could
easily be so great as to offset the theoretical gain of efficiency due to regen-
102 THE GAS TURBINE AND TURBO-JET [C/t 8

eration As a consequence, regenerators are likely to be relatively large


and costly
The ejfectuencss of the regenerator is defined as

_ actual amount of heat transferred


amount that could be transferred reversibly
Considering the actual points 2 and 4', Fig 88, a reversible transfer of
heat (§ 90) w ould result in the air being heated from 2' to d, and the exhaust

Fig 87 Diagrammatic Layout with Regeneration

gases being cooled from 4' to c Actually how c\ or, the air is heated only to
,

some state d and the gases are cooled to some state c' Thus the effective-
ness of the regenerator in terms of the states
show n in Fig 88 is

which applies with or without pressure drop in

the heat exchanger (regenerator) if d' is the

actual state the departing air (because


of
<?., — h d t for steady flow, A A — 0) and
—h
with a negligible mass of fuel If the products

are considered to have the same properties as


air and if the variation of specific heat is to be

Fig 88 Imperfect Re ignored use Ah = cp AT and cancel cp If the


generation without Fluid analjsis of the cycle is being made for the
Friction^ hre& e 2 d f =
ac tual substances we note that the denominator
of equation (u) applies to the products of com
bustion If the prior discussion is understood, the student should have no
trouble in writing equations for heats added and rejected and for the work
of a cycle with regenerative heating
At a particular ceiling temperature, the efficiency curve plotted against
pressure ratio for a Brayton cycle with regenerator and with fluid friction
rises to a peak at some pressure ratio and then decreases Such a pressure
ratio is about 3 5 for a turbine inlet temperature of 1500°F
§ 1S9 ] OTHER VARIATIONS OF THE BRAYTON CYCLE 163

139. Other Variations of the Brayton Cycle. Intercooling in the com-


pression process is used to save work, theoretically just as described in

Fig. 89. Intercooling and Regeneration. The properties shown are typical values as
rounded off from actual test data. Consider w/ as lb. fuel per lb. air. (See Fig. 90.)

§§ 122 and This feature, together with regeneration, is diagram-


123.
matically pictured in Fig. 89. The various energy quantities can be written
by inspection of the individual systems in Fig. 89. For negligible changes in
kinetic energy (Figs. 89 and 90)

IF' = hi - ha ' +h- /i 2 - Btu/lb. air,

T V[ = h3 — he Btu/lb. products,

but the substance in the turbine is the


product of combustion (not air) and the
mass flow in the turbine is 1 + wf lb. for
each pound of air entering the compres-
sor, where w/ — lb. fuel per lb. air.
Therefore, we have
W' = e (1 + wf)(h, - he)
— IF' + IF' Btu/lb. of entering air.

Q'n — ha ' — hb + (1 + w/)(he — hi)


Btu/lb. of enteung air.

The compression process may be broken


into more than two stages if it should
be economic and advantageous. S

In addition to intercoohng during com-


Fig. 90. Intercooling , Regenera-
pression, the turbine may be divided into tion, and Reheating . The cycle 1 -a'~
two turbines. In between these two tur- shows the Ts repre-
bines, we may arrange to burn more sentation of the cycle in Fig. 89. The
reheating is e'/, so that the two tur-
fuel. That is, the gases pass through
bine expansions are 3-e' and/g'.
another combustor and are reheated to
approximately the original maximum temperature. This idea, together
with intercooling and regeneration, is depicted on the Ts plane in Fig. 90.
,

164 THE GAS TURBINE AND TURBO-JET [Ch 8

In this figure, the dotted the second combustor proc-


line e'f represents
ess and fq' represents the second turbine expansion A common plan
is to design the high-pressure turbine

(3 e') with just enough power to drive


the compressor, and to use the lower
pressure turbine (fq’) on a separate shaft
to drive the generator (or do other work),
but separate shafts are not necessary m
order to use this idea If the plan is
used, how ever, the w ork of the HP tur
bine is equal to the w ork of compression
plus whatever energy Ef is needed to
overcome friction m
these two compon-
ents, Fig 90,

(1 + «v)(A, - h,)
= K — hi 4- kx — h + E{ Bt u/lb air,

where Wf may
be taken as zero if you
wish to ignore the mass of fuel, which is
of the order of 1% of the mass of air in
gas turbines
Fig 90 Repeated
140 Jet Propulsion Let us consider
a turbo-jet engine, Fig 92 If this engine can drive a plane at a speed of
«, fps in still air, the rclatn e effect is the same

as if the engine is on a stationary test block and


receives air at an initial speed of u, fps If the
entrance to the compressor is properly designed it

utilizes the kinetic energy of the entering air to


compress it, that is, the entrance is a diffuser
(§ 228) This compression is called the ram effect
(when the engine is moving through the air) and is
pictured by to, Fig 91, as an lsentropic compres
Sion If the air is moving slowly at point o, 0 — 0 K
just as it enters the compressor, it carries energy
h„, and
(v) ho » h, +K , Btu/lb Fig 91 Ideal Tut
bo jet Cycle The ki-
where K{ the initial kinetic energy (or, for the
is netic energy represented
moving plane, the Kinetic energy of the air stream by iade is relative to the
corresponding to the speed of the plane, which is to engine, which may be
moving
say that it is the kinetic energy relative to the

plane) The enthalpy A„ for the stream at rest, is called the stagnation
enthalpy Since the diffuser is not 100% efficient in converting kinetic
energy into enthalpy, the actual nse in pressure is not so large as the
§ UO] JET PROPULSION 165

isentropic rise. Let o' (not shown on Fig. 91 and somewhat below o be
)
the actual state to which the air is compressed; then the

Ram or pressure coefficient = —


Po-P,
that is, the ratio of the actual pressure rise to that which would occur during
isentropic compression.
At compression begins in the compressor and is completed at b (Fig.
o,

91). The
process in the combustor is the same as before, and gases enter
the turbine in state c, expanding to d and doing enough work abed to drive
the compressor. (The areas in Fig. 91 are not to scale.) Leaving the
turbine in state d, the gases expand in the nozzle according to the energy
relation (Fig. 91)
lid — he = Ke — Kd\ [nozzle]

that is, the drop in enthalpy Ah is converted into kinetic energy AK. If
the gases enter the nozzle at low velocity, which is likely, we may let Kd = 0.
All our discussion on the assumption that the jet gases expand in the
is

nozzle to exactly atmospheric pressure at that location. If the engine is


stationary, the energy diagram is as shown in Fig. 92. No net work is
;
i'
2 = 75 psia />
3
= 67psia ;j
J = 27psia]

Courtesy General Electric, Schenectady, N. Y.


Fig. 92. Turbo-jet Engine. The energy diagram ignores the mass of fuel. There
are 12 stages of compression, r* = 5.05. Normal thrust on test is 4730 lb. at 7630 rpm
with fuel consumption of 1.04 lb. per hr-lb. cruising thrust is 3700 lb. at 7000 rpm with
;

fuel consumption of 1.03 lb. per hr-lb. The temperatures and pressures given are
roughly typical.

done. The work and


quantities W t W
c are an interchange of work within

the system. If an airplane moving at a speed of u, fps in


the engine is in
still air, level flight, an observer on the plane “sees” the same energy dia-

gram. In this situation, the velocieties u, and v e are initial and exit veloci-
ties relative to the plane, the velocities as the observer sees them. The
“heat added” to the air in the combustors is the energy released by the
chemical reaction of the fuel and oxygen, as this observer sees the power
system, and according to our approximation is taken as Q — wt qi, where W/
is the mass of fuel (per pound of air, if the energy equation is set up on this

basis; note that 1 + Wf « 1 lb.):

h, + K, + wf qi
= (I +wf)(K + e h e) Btu/lb. air.
,

J6G THE GAS TURBINE AND TURBO JET [CA 8

Since tins equation does not contain a w ork term w e may resort to another
principle to find the work

141 Work from the Impulse-Momentum Principle The stream moving


past turbine blades does work by \irtue of the continuous change of momen
turn of the stream as you ha\c learned in jour course on mechanics The
force exerted bj an airplane or ship propeller exists by virtue of the change
of momentum of the fluid
brought about bj the propeller The propulsiv e
force of jets exists for thesame reason Trom your mechanics you recall
a = di/dt and F = va/g B which latter equation is Newtons law that the
resultant force r m any direction is equal to the mass times the acceleration
in that direction Together these relations result in
(w) rdt - (tc/ff.) di

which is the famous impulse^ dt) momentum(tr di/ga) principle where


w/g» is the mass in slugs If the mass rate of flow is constant we have
from equation (w)

(59) F-'-2(v.-„) or
+
F , (£l »/>* -
g. g0 g.

where upon integration the time interval is taken as At = 1 sec (that is


w is the constant mass rate of flow in pounds per second) u, and v fps arc
the exit and initial \ elocities respcctiv clj Yllo \ mg for the addition of fuel

to the stream we find the second equation where u„ lb per sec of air and
wf lb per see ol fuel are flowing steadily The right hand side of the frst
equation (o9) is the rate of change of momentum of a stream in stead} flow
undergoing a velocity change of Au 1\ e recall that momentum is a vector
quantity (^S) but the vectors aro in the same direction in jet engines In
applying equation (59) to a stream passing through a mowng body an
airplane use the velocities as those relatuc to the moitng bodg Since the
mass of fuel used in a turbo-jet engine is of the order of 1 %
of the mass of air
small error is introduced if the mass of fuel is ignored m computing F (A
large amount of air in excess of that needed for combustion is necessary to
keep the temperature rise within limits ) Therefore the propulsive force of
a jet engine is approximately

(59) r = " (v, -u ) lb

which isthe force of the fluid on the plane From one viewpoint F is the
force necessary to accelerate the stream from a relative velocity of v to a
relative velocity of v, but action and reaction are equal so it is also a force
on the plane If the plane is moving with a speed of fps the work done
by r lb is Fvf ft lb per sec or from equation (59)

(*) W = Fv p
f = (v, — u) ft lb /sec
§ 141 ] WORK FROM THE IMPULSE-MOMENTUM PRINCIPLE 167

where v, is the initial relative velocity of the air


with respect to the plane.
If w lb. per sec. is the actual mass of air and
the velocities are actual relative
if

velocities, this equation gives the actual propulsive work, or rate of work, in
moving the plane at a constant speed v p (mass of fuel neglected). Or it
would be the instantaneous rate of work if the plane should be accelerating
with an instantaneous speed of Up. Orient yourself to the fact that no work
is done by F at v p = 0; and that for a particular value of the thrust F, the

greater u p the greater the work.


,
Thus, a jet engine producing a thrust force
of 5000 lb. in the direction of motion, would be developing horsepower as
follows, for example:
-- (5000) (550) ....
A
At
.
n
vp = 550 ,
fps,
,
hp = ~r = 5000;
550
(oQ °° )(1100)
At vp = 1100 fps, hp = = 10,000;

(550 fps «= 370 mph; 550 ft-lb. perhp-sec.) At an altitude where p = 8psia,
T = 500°R, and for a plane speed of 500 fps and u e = 2500 fps, the 5000-lb.
thrust would require a flow of about 111,000 cfm; check it for yourself, using
equation (59) and pV = wRT. A reciprocating internal combustion engine
can be connected to a dynamometer; then its horsepower output can be
measured and reduced to standard conditions. Applying a known effi-
ciency of transmission and propeller, we may convert this power into
thrust at appropriate plane speeds. Thus, as we shall soon see, the economy
of the reciprocating engine may be expressed in terms of parameters like
pounds of fuel per horsepower-hour, called specific fuel consumption. In view
of the distinctive characteristic of jet engines described above (power is
zero when the engine is at rest), other ratios for expressing the fuel consump-
tion are used, one of which is
lb. fuel/hr.
(y) Specific fuel consumption = '

lb. thrust

At a particular speed and thrust, the specific fuel consumption may be


expressed in terms of some work unit like horsepower-hour for the purpose of
comparison with engines which deliver shaft work. Jet engines are tested
at rest and the initial air stream velocity u, = 0. From equation (59), we
see that since the thrust depends on v„ it is different under static test than
when the engine is in motion. Therefore, any value of specific fuel con-
sumption or of thrust should include the conditions to which the value
applies. A typical fuel rate is 1.08 lb. per hr. of fuel for 1-lb. thrust at sea
level static test.
The solution of problems on turbo-jets should follow suggestions presented
for the gas-turbine engine. With certain specified data, one would start
at the entrance state and work through the cycle, piece by piece, either for
the ideal cycle, or for the actual cycle given the appropriate efficiency
numbers.
,

168 THE GAS TURBINE AND TURBO-JET [Ch 8

142. Efficiencies of Turbo-jets. The propulsive efficiency y, is defined as


the work, of the propulsiv e force dmded by the energy converted to work or
kinetic energy (100% available energy ) in the system The work of the
propulsiv e force for the plane flj ing in still air, v = u,, and for to = 1 lb
p
per sec is, from equation (x),
~
ir = ft . lb /sec_lb
ff.

The energy developed in the system which 100% available may be found
is

in either of two ways First, the observer on the plane sees the increase in
relative kinetic energy but no work, for one pound,
1,1 ~ v’* + ~
(z) E. - - ( r* *- ',>)
ft lb /scc-lb

Second, the observer on the ground “secs” absolute zero kinetic energy
entering the system (still air) an absolute kinetic energy leaving the system
of Vt*/ (2g0)t w here v* is the absolute v clocity’ of the departing gases,

Vi = v. — 1

and he “secs" work 11 being done in moving the plane against a resistance,
that is in overcoming air resistance, etc Thus the 100% available energy
m sight is

- v Pr
. g 2 g.

. K-r^ + .
i>>
fllb/sec-ib,

the same (The ground observer also “sees” hi - h, enthalpy


as before
entering the sy stem, A 2 = h. enthalpy departing, and he sees a decrease in
the stored energy of the sy stem of wrfi Btu The reader should sketch this
energy diagram with IT leaving the system and show that (v, = v p)

A. - A'. TT + A's,
where Ai is and the other terms are as
the absolute exit kinetic energy
previously defined However, the impulse-momentum principle is usually
employed for computing W, as explained above ) Now we write the expres-
sion for the propulsiv c efficiency as

_ IT _
~ v p (v, — v p )2g0 _ 2v p /v,
Vp ~ E* + vp )(v, — v p ) + vp /v/
g e (v, 1

where Ea is defined by (z) This equation shows that this efficiency’ is zero

when vp = 0 and is 100% when vp = v« Since the condition vp = v, is the


condition for zero available energy generated (equation (z)], there would be
no work done to drive the plane at 100% efficiency According to this
equation, high efficiencies are obtained when the speed of the jet u, is a little
§ US] CLOSURE 169

larger than the plane speedvp but in this event the rate of flow of air would
,

necessarily be very large [equation (59)] in order to obtain large thrust,


which means a larger engine.
The thermal efficiency of the jet engine is the propulsive work W divided
by the energy Ee chargeable against the engine, taken as E =
c w,qi ;

e _ _ JV_ _ wv p {v — vp) = Fvp


e

E c wfqi Jw f qi
where W/ is lb.fuel per lb. air and qt is Btu per lb. of fuel. The field of jet
engines is so recently developed that one finds a number of different effi-
ciency ratios defined —and different names for the same ratio. We shall
define one more efficiency, the engine efficiency for the whole unit considered
as the engine,

— IE! — actual work


ffifVi
' ' ^ W work of the corresponding ideal cycle or ideal engine’

where the ideal cycle in this application is ibce, Fig. 91, and the corre-
sponding work may be found in any one of several ways already explained.
The thermal efficiency of turbo-jet engines is likely to be low as compared
with reciprocating types, but, especially for airplanes, there are some
offsetting advantages, such as the smaller frontal area to produce less air
resistance and less weight per horsepower developed in normal flight. Also,
some speed, say about
since the efficiency of a propeller drops rapidly after
400-450 mph, the turbo-jet drive actually becomes more efficient than a
reciprocating-engine-propeller drive at some high speed. In commercial
planes, speeds of 500-600 mph with turbo-jet drives appear to be economic
on long range flight. At and above plane speeds of Mach 1, the velocity of
sound in the air surrounding the plane, jet propulsion is being used. The
practical limiting speed of travel with turbo-jet engines is expected to be
about Mach 1.5 to 2.
Other means of jet propulsion than the turbo-jet are ram jets and rockets.
The ram jet becomes practical v hen the speed of body is high, well above
Mach 1 (say, Mach 3, or about 2000 mph). At these high speeds, the ram
effect produces enough compression of the entering air to develop a cycle as
previously described, the ideal prototype being the Brayton cycle. Typical
data for a speed through air of 2000 fps are: combustion temperature,
3600°F; jet temperature exhaust, 2200°F; jet exhaust velocity, 4000 fps.
The German V-l buzz bomb was propelled by a ram-jet engine in which the
firing was intermittent, giving rise to the buzzing. The rocket is a jet
propulsion device which not only carries its own fuel but also the reactant
for the fuel, as liquid oxygen. Speeds up to 11,000 fps are known to have
been obtained with hydrogen as the fuel and liquid oxygen as the reactant.

143. Closure. The student of thermodynamics is generally interested in


actual performance data, and typical values have been indicated in the
170 THE GAS TURBINE AND TURBO-JET [Ch 8

text and on illustrations in this chapter Other data, taken at random from
the literatdre, are given below Design \ alues for gas turbines (56)

Compression efficiency = 85%, Engine efficiency = 83%,


Combustion pressure loss «= 3% Nozzle efficiency = 98%,
Combustion efficiency =98% Regenerator eflectneness = 50%
Ram coefficient = 85%, Regenerator pressure loss = 5%,
Heating value tfr= 18 550 Btu/lb Intercooler picssure loss ~ 3%
Some characteristic data as giv cn l one of the General Electric bulletins
arc

Fuel oil Simple Cycle Regenerative Cycle

Rating 5300 kw 4000 kw


Heat rale 18 950 Btu/k" hr 1C COO Btu/kw hr
Thermal efficiency 18% 20 5%
Air required (80°F) 114 lb /sec 97 lb /sec
R eight 72 5 tons 200 tons
j

Inasmuch ns there are books written on the subjects of gas engines and
jet engines, it is easily understood that this presentation is necessarily

limited in scope Development of gas turbine engines for various purposes


(for example, as an automotive drive) is being carried on actively in many
laboratories, and progress has been and probably will continue to be excep-
tionally rapid It is easy to understand that the various ideas of this
chapter might be combined in many different ways, giving a number of
modifications of the Brayton cycle Moreover, combinations of the gas-
turbine engines w ith other apparatus, such as steam cy cles, might become
advisable Since this is so, the beginner s endetv or should be to master each
idea so that he can intelligently analyze any combination or so that he
might perhaps devise a useful combination
The gas turbine and the steam turbine are much the same For further
readings on gas turbines and jet propulsion, see references (38—17)
9

INTERNAL COMBUSTION
ENGINES

144. Introduction. The internal combustion engine is relatively new.


The earliestattempts to build such an engine were based on the use of gun-
powder. Barsanti and Mattcucei built a frcc-piston engine in 1857, which
operated as follows: An explosion drove a piston vertically upward. As it
started down under the action of gravity, it engaged a ratchet which was so
connected as to turn a shaft. Such a clumsy machine was doomed to fail-
ure, although Otto and Langen successfully marketed a number of free-
piston engines about 1807. In 1SG0, Lenoir proposed and built an engine
without compression. This engine drew in a charge of gas and air at atmos-
pheric pressure for half a stroke, at which point the mixtures was burned.
The resulting rise in pressure provided the motive force to complete this
stroke, return the piston to the end of the next stroke to exhaust the burned
gases, and to bring the piston again to the point of the burning of the new
charge. While this engine was used for a while, its efficiency was too low
for it*to be an economic source of power.
Although Beau de Rochas, a Frenchman, worked out the theory and gave
the conditions for high efficiency in 18G2, it remained for Nicholas A. Otto
(1832-1891) to build a successful engine in 1876 after he had independently
invented the same cycle. This engine was called the silent Otto engine, but
the word “silent” should not be taken in a literal sense. Otto was bom in
Holzhausen, Germany, and was a partner in a gas-engine manufacturing
plant at the time of his famous invention.
To circumvent Otto’s patents, Sir Dugald Clerk, born in Glasgow in
1854, invented the two-stroke-cycle engine, which was
first exhibited in 1881.

171
172 INTERNAL COMBUSTION ENGINES (C/t 9
In these early stages of the internal combustion engine, rotative speeds of
the order of 200 rpm were typical The German Gottlieb Daimler (1834-
1899) was the first to conceive of small, relatively high-speed engines for
greater power from a particular size, say 1000 rpm (vs 4000 rpm and more
for today's automotive engines), and he made them w ork by improved hot-
bulb ignition The “high-speed” engine made the automobile a practicable
idea
We have seen that the highest temperature in the gas-turbine cycle
occurring in the combustor, is sharply limited The combustor is subjected
to this operating temperature continuously On the other hand, since the
reciprocating internal combustion engine (ICE) is subjected to the highest
temperature intermittently, just after the fuel has been fired, temperature
has not been much of a problem We may use air-fuel ratios such that the
amount of air is close to that which is ideally required for combustion and
let the temperature go where it may Since the highest temperature exists
for only a small portion of the cycle, the interval during the remainder of
the cycle can be used for aster or far air coaling the cylinder in order to
prevent the metal from becoming dangerously hot
As you know energy is supplied to the internal combustion engine by
,

the combustion of a fuel within the cylinder The widespread use of inter-
nal combustion engines in automobiles, on the farm, m industrial plants,
on ships, in power plants, is common knowledge Because these engines
are used so much, the ideal cycles for them are particularly significant The
fuels used are natural or manufactured gas, gasoline, kerosene, oil, etc,
alcohol, and others The most common fuels are gas, gasoline, and fuel oil

145. The Four-Stroke Cycle. The four-stroke cycle


one wherein four
is

strokes of the piston, two revolutions, are required to complete a cycle


The sequence of events, pictured in Fig 93, are the same for any four-
stroke ICE, namely

1 A suction stroke, drawing fuel and air into an Otto engine, § 146, or drawing
air only into a Diesel engine, § 150,
2 A compression stroke, Fig 93(b),
3 Ignition of a fuel already in the cylinder, as by a spark plug, or the self ignition
of fuel, which ideally is injected into the cylinder at the end of the compression
stroke (the burning of the fuel releases energy for use by the system),
4 An expansion stroke or power stroke, during which positive work is done, and
5 An exhaust stroke during which most of the products of combustion are
pushed from the cylinder, then the cycle repeats

The end positions of a piston in any reciprocating machine are called


head-end dead center, or in the automotive industry the top dead center
(TDC) [Fig 93 (a) and (c)], and the crank end dead center or bottom dead
center (BDC) [Tig 93 (b) and (d)] Figure 94 reveals some of internal
construction of an automotive engine
(a) (b) (c) (d)
Suction Strobe Compression Power Stroke Exhaust Stroke
Begins Stroke Begins Begins Begins

Fig. 93. Four-Stroke Cycle. This diagrammatic representation shows a spark plug
which ignites the fuel after compression. The same sequence of events occurs in the
4-stroke-cycle Diesel engine, except that the air is compressed to a temperature high
enough to cause the fuel to burn without spark ignition.
1

174 INTERNAL COMBUSTION ENGINES [Ck 9

146 The Otto Cycle The Otto cycle, which is the ideal prototype of
most small internal combustion engines, is one wherein it is imagined that

the combustion process takes place instantaneously at top dead center to gue a
constant lolume combustion of the fuel (or constant \ olume process of heat
added in the cquiv alcnt air cycle) The Otto engine may be analyzed either
as a flow device or as a closed cj cle
We note that ideally (no pressure drops, etc ), the suction stroke 0
and discharge stroke 1-0, Fig 95, cancel one another, so to speak The
positive work under 0-1 is equal to the negative work under 1 0, and these
works correspond to the ideal flow works done in getting the air into and
out of an open system Moreover
the ideal open bj stem and the ideal
closed system reject thesame amount
of heat, according to the first law,
because in each case, if the cycles are
comparable, the same heat is added
and the same work is done, hence
from Qa — Qr = H’, each must
reject the same heat Q R to the sink
(a) (b) Therefore we see that the analysis
of an ideal open air cycle is the
Fig 95 Otto Cycle
same as that of an ideal, closed air

cycleand such an analysis is called an air -standard analysis


The Otto cycle is shown on the pV and TS planes in Fig 95 and the

reader should relate these processes to the events in the actual engine (Fig
93) 1 2 is the compression process which the ideal case is, as usualm
an lsentropic process, 2 3 is the instantaneous heating of the air (com
bustion) at constant volume, 3-4 is the expansion which is ideally lsentropic,
and 4-1 is the instantaneous rejection of heat at constant volume (equivalent
to a valve opening at 4 with heat rejected to the atmosphere after the gases
have left the engine in the open cycle) During nonflow constant volume
processes, Q = AU, regardless of working substance Since it is common
practice to analyze these cycles for constant specific heats, we have, for this
assumption and for the closed cycle Fig 95(b),

Qa « Ui - Ut = wc,(T t - Tr) Btu,


Q„ = Ui - Vi = wc.(Ti - T,) wc,(7\ - Ti Btu

The net work II is 2Q so that

(a) V, - Ut “ {Ua - Ur) = uc,(Ts - Tz) - wcv (T4 - Ti) Btu

The thermal efficiency of the Otto cycle is

IF U, - U* - (U« - U0 wc,(P* - T - t) ttc.( P« - Ti),


“ Qa ~
__
e
U - U * 2 vlc,{1 s — 2 2)
147] IDEAL STANDARDS OF COMPARISON 175
or

= — Ti - Ti'
(b) e 1 [c p AND Cc CONSTANT]
T3 - T,‘

The expressions in terms of internal energies are basic; but the analysis is
continued for the air standard. To simplify equation (b), use the TV rela-
tion for an isentropic process, equation (35). Thus, T,/T z = (F3/F4)* -1
and T,/T 2 = (Yt/Vi)*-1 ;
or, since V = V2 and V = V h
3 4

t - t
‘ •
(ftr
- r

(r?r'’ ^ or-
Substituting these values of T 4 and T\ into (b), we find

T3 (V'./V i)*- - T, (F2/F,) 1 " 1


1 1

(c)
, fV.y-
t z -t 2 \v ) t

Let the adiabatic compression ratio V \/V 2 be represented by the symbol


Tk\then for constant specific heats,

61 ) e = 1 — 1
r ^- 1
(

Since we may wish to study something about the cycle related to pressures
and volumes (mep, for example), we should keep in practice in writing work
equations from the pV plane; thus, from Fig. 104(a),

= P2V * ~ V lVl ~ 3
(d) TF
y
+ ft-lb.,

which, with pV = wRT and c v = (R/J)/(k — 1), can be shown to be the


same as equation (a) except for units.

147 . Ideal Standards of Comparison. In equation ( 61 ), we have arrived


at an important characteristic of the Otto cycle, to wit, that its efficiency

with constant specific heats depends only on the value of k and the com-
pression ratio tk (Fig. 96 ).The efficiency of the real engine is subject to
many other variables, of course, including simple thermodynamic factors
such as the initial temperature and the temperature at 3 (amount of heat
released). Nevertheless, one of the principal aims in the development of
spark-ignition engines over the years has been to increase the compression
ratio, whose valuenow limited largely by the detonation characteristics
is

of the fuel. If use the air standard as a basis of comparison, we may


we
consider the air as “cold air,” k = 1.4, and the corresponding standard
is called the cold-air standard. Let rt = 6 and let the actual value of the
thermal efficiency be e’ = 21% by test; then we might say that the actual
3

176 INTERNAL COMBUSTION ENGINES [Ch 9

02i «2I_
41/0
- l/r* (l X-l/l> a <

as efficient as the ideal engine


The cold air standard is unfair to the engine, howev er, because operation
is inev itablj with rather hot gas most of the tune To keep the computa-
tions simple, some value of i which is rough I the average throughout the
j
cjdc is sometimes assumed, and the ejele is anal} zed as explained above
except that L is some value like 1
with the corresponding [equation
(25)] v alue of c. (and c, if needed,
Dic«cl cj cle) If a hot air v alue of l
la u<=ed, the standard is «aid to be a
hot-air standard Again, let r* = 6
and — 21% b\ te't, then as be-
c'
fore,except that f =13, we maj
sav that the actual engine is

0 21
- l/r,<* »
0 21 .

1 / 0°
-
1

as efficient as the ideal This higher


Compression Ratio
ratio of the efficiencies is a better
Fig 96 Efficiency Vs Compression
indicator (than 41 %) of the possible
Ratio Otto Cycle The solid curve is for
cold air, k = 1 4, dotted curve for hot air, margin of improv ement as the actual
k — 12. The test values, computed on engine might be made to approach
the lower heating value and taken at
the ideal, and it is a fairer indicator
random from the literature, suggest that
actual efficiency tends to improve with of how good the actual engine is
ideal efficiency Another nltematn e is to integrate
fc,dT with % anable specific heat
equations, Table II A much less tedious approach 13 to u«e the air table,

p 146, and obtain a variable specific-heat air standard


Still another alternative is to consider the actual fuel-air mixture and
the products of combustion, including the phenomenon of dissociation,
ifan} This method results in the most reali-tic standard of companion,
and ccrtainlj there could be little reason left for the real engine to complain
about being compared with such a standard, which we shall call the teal
mixture standard. Such calculations as are involved would be quite
tedious and time consuming were it not for available aids—charts giving
properties of real mixtures allowing for dissociation, and the Gas Tables
(wluch do not account for dissociation)
§ 149] EXAMPLE 177

148. Clearance Volume. The compression ratio is varied by varying the


clearance volume, which is the volume of the combustion space when the
piston is on TDC position, Fig. 95. It is usually expressed as the clear-
ance fraction or per-cent clearance, Thus, the clearance volume c. is

cVd, where VD is the displacement volume. The compression ratio is

+ cVd +C
(e) t ,. = V
2
^= V° cVd -
1

from which the clearance can be


computed from the compression
ratio, orthe compression ratio can
be found from the clearance.

149. Example. An ideal Otto engine


with 25% clearance operates on one '

(jj)
pound of air with k = 1.3. In Fig.
Fig. 95. Repeated.
95, pi = 14 psia, t3 = 120°F, and
U = 4740°F. (a) What (b) Find U, pi, and p 3
is the displacement volume? .

(c)Find Qa, Qr, and e. (d) Find the mep of the ideal C3’de and the percentage
approach to perfection (commonly called engine efficiency), if the actual thermal
efficiency is 24%.
solution, (a) First find Fi and rk .

wRT i (53.3) (580)


=
Vi = 15.33 cu. ft.
Pi (14) (144)
(1 +c) _ 125 ~ _
Tk = " 0.25 _ 5 “ Yl.
Vt
From this we find

Vi = —= ri 5
= 3.066 cu. ft.

Vd = Vi - V2 = 15.33 - 3.066 = 12.26 cu. ft.

(b) Use the TV relation for an isentiopic process and Charles’ law.

P
i i w 2/
or Ti = (580) (5> *-*) = 940°R = 480°F.

From piV\l = P 2 V 2S
1

Pi = Pi = (14) (5*
3
)
= 113.4 psia.
(J;)

Fiom Charles’ law (T 3 = 4740 + 460 = 5200°R),

Pa = Pi (jr) = (113.4) (7^) = 627 psia.

(c) First, we must find c, coriesponding to the value of k = 1.3.

„ __J?
~ “
53.3
-
J{k - 1) (77S) (1.3 1)
178 INTERNAL COMBUSTION ENGINES [Cft 9
Qa = v.c.(Tt - T*) = (0 22S)(5200 - 940) 971 Btu

r T

-(rT‘
Qa - uc,(Tt - T ) = (0 22S)(5S0 -
a 3210) = -600 Btu
SO 971- 600
“ 3S2%
Qa 971

3S 3% (check)

Tl_ (Qa (0 3S3)(971)(778)


<d) P“ “ 1 b “ 1 1> “ (12 20)(144)
1G4 p«i

actual thermal efficiency 0 24


p = —
__ - 62 7%
ideal theimal efficiency 0 383

Me ha\e non solved example* of three cycles with the same i«entropic compression
ratio r* — 5 the Carnot §89 the Brnyton § 134, and the Otto herewith Let us
compare the corresponding meps even though they are not computed on the «ame
basis (Carnot cold air B raj ton variations of specific heat allowed for, Otto hot
air) Carnot 11 psi Brajton 262 pi Otto 164 psi Recalling that the relative
sizes of the engines are rough! j inversely proportional to the
meps for a particular
power and speed we «ec immediately the advantage of the Otto cycle at relatively
low speeds The possibility of running gas turbines at much higher speeds than
the reciprocating ICC compensates for the relatively low mep of the Bray ton cycle
Actual meps are naturally lc*s than those of the ideal cycles

150. Diesel Cycle Rudolf Diesel* was primarily interested m develop-


ing an internal combustion engine to operate on coal as a fuel The final
outcome, however was a four stroke-cycle engine (§ 145) in which air only
is taken into the cy linder on the suction stroke and a liquid fuel is later

injected, the injection starting theoretically at the end of the compression


stroke and continuing at such a rate that burning proceeds at constant
pressure 2 3 Fig 97 Otherwise the cycle operates as the Otto cycle does
and the air standard (closed) cycle is 1 2-3-4 Fig 97 The ideal open
air cycle would be the same except that 0 1 is the suction and 1-0 is the
discharge Thus, for constant specific heats w e hav e

Qa = wcp (7 1 — Tj) Btu


Qr = tt c,(T x
~ TJ = - Ti) Btu
II - ZQ = itc p {Ti - T) - t uc.(T, - Btu
tt c,(T 4 - T,) T 4 - Ti
(f) c
Qa c,(rs ~2.) m t -Tt)

This expression (f) is usable but it may be placed in a more convenient and
* Rudolf Diesel (1858 1913) born in Pans of German parents who later moved to
London because of the Franco German War (L870) educated in Germany obtained in
1893 a patent on the type of engine which now bears his name After some difficulty in
financing the project he built an engine which blew up at the first injection of fuel
D esel narrowly escaped be ng killed Four years of tedious and costly experiment
elapsed before he produced a successful engine He inexplicably disappeared in 1913
while crossing the English Channel during a storm
:

§ 150] DIESEL CYCLE 179

revealing form by eliminating the temperatures. We may express three of


the temperatures in terms of the fourth, say, in terms of 7V Thus, along
the isentropic 1 - 2 Fig. 97, T«/Ti = (Fi/Fj)* -1
, However, F,/F 5 is .

defined as the compression ratio, n. Therefore

Along the constant pressure fine 2-3, Charles’ law holds and = F /F
3 2.

Let F 3 /F 2 = r c a ratio termed the


,

fuel cutoff ratio. We find then


(h) T3 =

by using equation
T,
^ For the
= Tr

(g).
1 l
l
~'rc

isentropic processT4/T3 — 3-4,


{VffVtf-K But from 2-3, F, = V S
(a) (b)
(T,/T 2 )V2 = rc F2 Using this
.

value of F 3 the value of T 3 from


,
Fig. 97. Diesel Cycle. Between the
same temperature limits, the constant vol-
equation (h), and using the defini- ume curve on the TS plane is steeper than
tion of the compression ratio the constant pressure curve (§67). How-
namely rk = F /F2 we
1 ,
get ever, both curves get steeper as the tem-
perature increases. Thus, the temperature
1
at 3 is so high that the constant pressure
Tt = =
(i) J\
(0 TV,*. curve is steeper than the constant volume
curve, which is at the lower temperatures
between 1 and 4.
Substituting into equation (f) the
values of TV Tz and T
, t just found, we have

(62)

Observe that this expression for the efficiency of the Diesel cycle differs
from that of the Otto cycle [equation (61)] only in the bracketed factor. This
factor is always greater than 1, because rc is always greater than 1. Thus,
for a particular compression ratio n, the Otto cycle is more efficient. (See
also Fig. 98.) However, if the compression ratio is too high in an Otto
engine, a knocking occurs due to self-ignition of the fuel. Since the Diesel
engine compresses air only, the compression ratio is higher than in an Otto
engine. Therefore, an actual Diesel engine with, say, r* = 15 is more
efficient than an actual Otto engine with n = 9. In passing, we may note
the relation between the compression ratio n, cutoff ratio rc and expansion ,

ratio r, (F« = Fi, Fig. 97)


180 INTERNAL COMBUSTION ENGINES [Ck 9

Study of equation (62) shows that as re increases, the bracketed factor


increases, and the efficiency decreases (Fig 99) Therefore, the lower
fuel cutoff ratios are conducive to higher efficiencies but larger ratios result
m greater power However, there is a limit to the amount of fuel which

S S V
(a) (b) (c)

Fig 98 Comparison of Olio and Diesel Cycles These cycles may be compared in
many different ways and these sketches, which are qualitative with respect to areas may
be of interest In (a), they are sketched for the same compression ratio and the same
heat added, that is, area under 2 3 (Otto) is equal to area under 2 m (Diesel) We see
that the Diesel rejects more heat n 1 than the Otto does 4 1, a visual demonstration
that the Otto is more efficient But the compression ratio is not the same m the two
cycles as they are actually used, so (b) and (c) are sketched for the same temperature
and pressure d after combustion (which is not true either, except by chance) At any
rate, you can see how the areas are affected

can be injected without excessive ‘ smoking”, hence the compromise usually


is such that cutoff seldom occurs later than 10% of the stroke, correspond

ing to a cutoff ratio of about 2 4 usually earlier


As in the Otto cj cle the value of k in the cold air standard is I 4 Lower
values, say about 1 35 would be used in the hot-air standard A preferred

Fig 99 Diesel Efficiency Vs Cutoff Ratio — Constant Compression Ratio


standard of comparison would be one based upon an analysis of the real
mixture in the engine and accounting for the variation of specific heat
An end-view cross section of a four stroke Diesel engine is shown m Fig 100
§ 151} FUEL CUTOFF RATIO 181

Rocker Arm Valve


Springs
Water "
Cooling nV^-Waler
®^Cooled
Exhaust
Manifold

Intake

Valve

Injection Nozzle

Piston

Manifold

Wrist Pin

Connecting
Rod

Crank Pin

Courtesy Ingersoll-Rand Co ,
New York

Fig. 100. Four-Stroke Diesel Engine. Study the picture for detail. A hole at the top
of the drilled connecting rod serves to spray the piston head with oil for cooling. The
bore and stroke are 10^x12 in., assembled with 5 to 8 cylinders; 375 to 600 bhp; super-
charged, 670-900 bhp; supercharged with cooling of supercharged air, 600-1000 bhp; all
at 720 rpm. Figure 103 is an external view of this engine.

151. Fuel Cutoff Ratio. The question arises as to what is the ideal Diesel
cycle corresponding to a real engine operating at a particular load. The
point 2 at the end of isentropic compression is defined by the compression
ratio, which is a characteristic number of the engine and is usually known.
To locate state 3, Fig. 97, we note that the first law applied to the process
2-3 gives

(Stored energy) 2 + energy entering = (stored energy) 3 + energy departing


as work,

„ =
+ lUfEc) + w/)«3 4 p3(1 + wf)v - 3 p«vi
iiz +, ,
{io/hf
, , N
(1
x ,

j
>

(k) hz + Wfh/ + W/E = c (1 + W/)lf3>


where h 2 is the specific enthalpy of the gases in cylinder at 2, wf is lb fuel
,

182 INTERNAL COMBUSTION ENGINES [Ch 9

per lb air, ft, is the enthalpy of the entering fuel (A, is often negligible m
tthicli case it may be dropped), Ee
v T is the chemical energy released by
J
2 3 P ,c / combustion m
Btu per lb of fuel
\ V Y and h } is the specific enthalpy of
\ \. // 4
the products of combustion The

— — OVc enthalpies ftj, A,, ft a and the energy


o
-
if |
should be reckoned from the
v s same datum (14) when you are
(a) (b) accounting for the real mixtures
Fig 97 Repeated With the proper aids in the way of
tables or charts (14, 51), equation
(k) is easy to use to solve for A,, then other state properties at 3 can be
determined However, the air standard can be closely defined if in equa
tion (k), W/E, = 10,
we let where q Btu per lb is the lower heating \alue
17 , t

of the fuel at constant pressure, andif wc let the other terms be represented

by the approximation w T (h } — Aj )

(l) Wrf, » w T (h, — hi) = UrSCpdT = u tcp (Ti - Tt),


(air standard]

whore tir = 111, -t- v>/t the total mass of working substance, which may be on
the basis of wa » I lb of air, and where cP in the final form is some mean
value for the cycle based on the hot air standard For the air standard
Tt is computed from the properties at 1 Since from Charles’ law

Vi/Vi = Tt/Tt the cutoff ratio for the cor-


responding ideal cycle is re = V x /Vt - T 3 /Tt

152 The Two-Stroke Cycle Many small


gasoline engines and many Diesels large and
small, operate on a two stroke ejele Since the
exhaust stroke in the real engine is for the purpose
of scavenging the cylinder (ridding the cylinder of
the products of combustion), it is only necessary 101 Tw0 str0,te

y
to provide other means of scavenging in order to
be able to complete the cycle in two strokes (one revolution) An idealized

sort of indicator card is shown in Tig 101 Compression has been com
pleted at point c, combustion occurs cd, followed by an expansion In the
two stroke cycle exhaust begins early at some point e and sca\ enging is
accomplished by blowing air (or air and fuel) into the cylinder To allow
time for scavenging or for the introduction of fuel mixtures the valves
usually remain open until the piston has mov cd to some position correspond
mg to b, where compression begins
In four-stroke cycles, there is one power stroke for two revolutions, in
two-stroke cycles, there is one power stroke for one revolution However,
§ 153] BRAKE POWER 188

the two-stroke-cyele engine, instead of developing 100% more power,


develops only some 70% to 90% more than the four-stroke-cycle engine of
the same displacement, because of (1) poorer scavenging, (2) a smaller
mass of combustible mixture in a given size of cylinder, (3) a greater loss
of unburned fuel, (4) a small power consumption in compressing the air
which scavenges the cylinder, and (5) the loss of pressure by early exhaust.
Compressed air for scavenging is obtained from crankcase compression, as
in some outboard motors and lawn-mower engines, or from separate blowers.

153. Brake Power. The name brake work came about because in the
early days of small, slow-speed engines the power output was dissipated in
the friction of a brake. Brakes are used for this purpose over suitable
still

ranges of power and speed. A type known as a prony brake is shown dia-
grammatically in Fig. 102. When the brake is clamped to the flywheel,

the frictional force F (= F/2 + F/2 at the brake shoes) tends to turn the
brake with the flywheel. However, the knife edge on the beam rests on
scales and prevents motion of the brake. The force P, which is weighed by
the scales, consists of the reaction produced by friction and a portion of the

weight of the brake, called the tare unless there is a counterweight to
balance the brake about the centerline of the flywheel. To find the tare,
support the brake on an edge at B and weigh the tare on the scales. Thus,
the net force on the scales produced by the frictional moment is (P — tare).
The information obtained from a prony brake test is used to compute the
brake horsepower. Thus, the frictional force F acting through one revolu-
tion of the flywheel does work W
= (FvD)/ 12 ft-lb., where the flywheel
diameter Dexpressed in inches (usual practice). The work multiplied
is

by n revolutions per minute gives the work in ft-lb. per min. Dividing this
result by 33,000 converts to horsepower; thus

,
FwDn F2irrn _ Tn\
(63) tip
- ( 12 )(3 3j ooO)
“ (12) (33,000) 63,000’

where T = Fr = FD/2 in-lb. is the frictional torque on the flywheel. The


sum of the moments of the forces on the brake about the center of the fly-
184 INTERNAL COMBUSTION ENGINES [CA 9

wheel gives

(m) rr = ~= (P- tare)L = T,

where L m inches moment arm of the force P Thus knowing the


is the
dimension L on the brake we maj observe the value of P indicated by the
scales and the number of rpm of the engine and calculate the brake horse-
power (bhp) from equation (G3) Sometimes the tare is neglected some
times it is balanced by a counterweight
Other instruments for measuring shaft work are a hydraulic brake and a
dynamometer Because the ordinary type of indicator (Fig 62) has too
much inertia to respond accurately at high speed it became customary to
measure the power output of automotive and other similar engines by
dynamometer only This practice brought about the common use of the
brake mean effective pressure (bmcp) p m a which is simply the pressure
computed from the mep equation (53) using bhp instead of ihp (review
§§ 80 and 111),
33 000 bhp
(53C) PmB =
LAN
While dynamometer routine industrial practice high speed
tests are still

indicators utilizing the oscilloscope and also optical and photographic


effects have been developed so that it is now possible to obtain from high
speed engines the valuable information revealed by an indicator card

164 Example At 3000 rpm a six cylindei four stroke gasoline engine 3>£x4 in
(always bore X stroke) develops SO bhp Mhat is the bmcp’
solution To find the number of cycles nr jiowei strokes per minute N we note
that there are 2 revolutions cycle for each cylinder or cycle/cylinder one m
revolution which is G X 3 cyclcs/icvolution for the six cylinder engine there
fore N— (3) (3000 rpm) = 9000 cpm From equation (53C)

(33 000) (80)


VmB 106 psi
(M 2 )fr 3 25’/4)(9000)

165 Thermal Efficiencies The thermal efficiency e — TF/Q^ has already


been defined a number of times In general terms for a power cycle or
engine
work output of the system
(47A) Thermal efficiency =
energy chargeable against the system

Tor ideal cycles and engines the numerator and denominator are ideal
values which are defined for each ideal system taken up In this expression
the work may be any energy which is 100% available electricity, for
exam
several
pie There is no one actual thermal efficiency, because there are
§ 155} THERMAL EFFICIENCIES 185

places where work and power may be measured (Fig. 103). You know
about the indicated and brake horsepower. The combined work or com-
bined power, measured by electrical instruments on the instrument panel,
say, in kilov atts (kw), is the output of the generator. If we let Q'A represent

Courtesy Ingersolt-Rand Co ,
New York

Fig. 103. Meaning of Indicated Work, Brake Work, and Combined Work ; Wi, Wb, W f:

the actual energy to be charged against a system, we have three thermal


efficiencies, the indicated e„ the brake e b , and the combined e> as follows

(47B)
Wi
b
_ WB h
_ WK
Q'a Q'a Q'a

The symbol Q'A will be defined for each actual engine analyzed. As you
know from preceding chapters, we charge the heating value of the fuel
against the gas-turbine unit and the internal combustion engine, preferably
the lower, W/Qi, where qi is Btu per lb. of fuel* and wf is in units suitable for
the units of the numerator of equation (47B). Thus, if TF is in Btu per lb.

of air, then wf must be lb. of fuel per lb. of air; etc. the specific If tty is

fuel consumption in pounds of fuel per unit of work, the term wf qi Btu per
uni of work is called the heal rate, which is defined as the energy supphed
t,

to the system per unit of work. The umt of work is generally the hp-hr.
(2544 Btu) or the kw-hr. (3412 Btu). From these values, we obtain other
important equations for thermal efficiency:

2544 / Btu/hp-hrA 3412 / Btu/kw-hr. \


(64) or
W/qt \Btu/hp-hr./’ wfqi \Btu/kw-hr./
or
2544 2544
(64A)
Wffll’ wp/n'

where the specific fuel consumption wu is in lb. per ihp-hr. and w,h is in

* The higher heating value is frequent!}' used in this country for ICE (See § 157.)
180 INTERNAL COMBUSTION ENGINES [Cft 9

lb per bhp hr If w/t is in lb per kwr-hr output of the generator, we have


3412
(64B) ek =
w/tqi

156 Engine Efficiencies. The engine efficiency rj has been defined in


connection with gas-turbine units (§§ 132 and 142), m general terms, let it
be*
actual work of a sys tem
(00) Engine efficiency ij
work of the corresponding ideal stem
Applied to reciprocating internal combustion engines, this definition results
in three engine efficiencies corresponding to the three actual w orks, Wu IT®,
and IF* Thus for the brake engine efficiency T, b the indicated engine ,

efficiency ij,, and the combined engine efficiency t we have rj


,

(OOA) ij, “ Wj ijb = Wb and


j
17 *
= W jc

where the ideal work IF is in the same units as the numerator and is com
puted for the corresponding ideal system The corresponding ideal system
for an Otto engine is one which has the same compression ratio as the actual
engine and the same energy supplied ( Qa - Q\) For a Diesel engine, the
corresponding ideal system is an ideal Diesel cycle whose r* and Qa arc the
same as for the actual In each case, one must of course, decide upon what
standard of comparison to use, cold air hot air, variable specific heat, or
real mixture In those cj cles w here the actual Q'a and the ideal Qa are the
same, as in Otto and Diesel ejeles (but not gas turbine-unit cycles), the
engine efficiencies are also ratios of the thermal efficiencies (Tf = cQa)

[\\ HEN Q a - Qa)

Since W = VmVD [equation (49), p 97] we get other ratios from equation
(60) as follows

utbiwi T> m \% th«. of. tfea lors.'supy?/ v&eaJ. wui thu. vucs/tcatora are
determined from equation (53) See equation (53C), § 153 The reader
can find other ratios which give the engine efficiency, for example, horse-
powers and fuel rates
Since the engine efficiency can often be estimated closely from previous
experience w ith a certain kind of engine, it is a convenient design factor for
use in determining the size of an actual engine to produce a specified amount
* This is different from the previous editions of this book, but it accords with ASME
^

158 ] HECHAinCAL EETICIEit'CY 187

of power. The engine efficiency was applied to a turbine in ? 132. equation


(n).

Thermal Efficiency. As mentioned in Chapter 8 on gas tur-


157. Actual
bines, there some question as to what heating value to use in computing
is

the efneiency. Although the lower heating value is commonly used in


other countries, the practice in this country has been to use the higher heat-
ing value, which accords with the argument that the engine is at fault if it
cannot use any part of the latent heat of the H : 0 formed by combustion.
However, the fact remains that none of the ICO is likely to condense in the
ICE: hence, it is said to be unfair to the engine to charge it with the higher
value. Consider the definition of thermal efficiency, (Q A — Qr)/Qa-
There is agreement that the QA in the numerator should be the lower heating
value: that is, the lower value is used to determine the temperature at 3
after combustion. Xow in order to be consistent thermodynamically, we
should use the lower value for the Qa in the denominator too. It is sug-
gested that in this course the lower heating value be used for all combustion
engines, as now done by some ICE designers —unless the instructor directs
otherwise. Thus, the thermal efficiency is the work divided by the energy
E= wi'ii chargeable against the engine,

TF
(P) e = ;

qi

where a dimensionless ratio and the numerator and demoniminator must


e is
therefore have the same units. For example, if wf is the specific fuel con-
sumption in lb. of fuel per hp-hr.,avalue commonly obtained on test from the
total fuel used in a particular time when the engine is steadily delivering
a certain horsepower, if rp Btu per lb. of fuel is the lower heating value of
the fuel, then

_ 2544 (Btu/hp-hr.) _ 2544


(q) /lb. fuel\ x . / Btu \ “77*
\hp-hr. / \lb. fuel/

158. Mechanical Efficiency. The mechanical efficiency -q is a number


which tells of the mechanical losses in a machine. For the generator (Fig.

103) it includes internal electrical looses;

(r) [GENXEATOP.l

For a reciprocating engine of any type which delivers work,


'
.
= ,! c
(s)
^
1J„ [P.ECIPEOCATINO ENGINE)
~Jy~'

See j 121 for i;-, for a compressor. The foregoing works may be expressed
in any convenient work or power unit or in terms of any numbers which are
188 INTERNAL COMBUSTION ENGINES [CA 9

proportional to the IF’s, but remember that engine efficiency is a dimension-


less ratio The difference (lhp) — (bhp), orWi — \VB, represents the loss
duo to mechanical friction of the moung parts of the engine, expressed in
horsepower, called the frietton horsepower (fhp ),fhp = (1
it is ij*)(ihp) —
Be sure to note that mechanical efficiency is not a fixed number characteristic
of the machine but that it depends upon operating conditions, especially
output, speed, and lubrication

159. Volumetric Efficiency. There are a number of factors which


account for power loss m an internal combustion engine, among which are
the obuous mechanical-friction loss and the loss which accompanies the
heat rejected to the cooling medium, which is the cooling water in the auto-
mobile engine Another factor is that combustion is not instantaneous as
assumed in an ideal Otto ejele (nor at constant pressure as assumed the m

Diesel ejele), nor is combustion complete some unburned components
escape in the exhaust A significant factor is the time and work imohed
in pumping the working substances into and out of the cj lmder, a loss which
becomes re!ati\ ely large at high speeds This is a fluid friction loss There
is also a loss in multic} hnder engines because the manifold is imperfect,

the same mixture (air-fuel ratio) is not dcli\ ered to e\ er> c> lmder
Other factors remaining the same, the power obtained from an engine
which draws in air and fuel depends upon the mass of combustible mixture

drawn into the cjhndcrs gnen a mixture with the correct air-fuel ratio—
and aiij thing which reduces the mass of fuel entering the engine reduces the
power output below what could ha\e been obtained Tor example (1) in
the real engine, we haxe found that, because of the fluid faction of flow or
throttling around the \al\es and in the passages, the suction pressure is less
than atmospheric pressure, and therefore the mass of gas ( w = pV/RT) is
less atmospheric pressure were maintained, (2) the internal surfaces
than if

and passages of the engine are rclatn cl} hot, so that the mixture is heated
as it In accordance w ith Charles’ law the increase
passes into the cylinder ,

m temperature further reduces the weight of mixture that the gnen dis-
placement can contain, (3) the gases in the clearance space of the real engine
are at a pressure aboic atmospheric at the end of the exhaust stroke and must
expand during the suction stroke to the intake pressure before a new charge
begins to enter, (4) the pressure of the atmosphere decreases nvth aJii-
tude, so that the mass of mixture drawn in at high altitudes is still further
decreased below that w hich w ould be drawn in at sea level
The mass of charge brought into the c} lmder is sometimes defined m
is a ratio
terms of volumetric efficiency (that is what we call it, § II 7, but it
of masses m
an ICE), which is defined as (49)

. mass of air draw n into the engine ,


(t) n» — masg Qf air ^at w ould occupy the displacement \ olume at p„ and T,
§ ISO I CLOSURE 189

where p and Tc are the air pressure and temperature at intake, say, in the
:

test room. The value of the denominator is w = (PzTd)/(RcT<,). The


numerator and denominator in equation (t) must be in the same units
Usually we use mass per engine cycle or mass per minute. The displace-
ment volume per engine cycle is the volume swept out by the piston in one
stroke: this number, multiplied by the number of cycles per minute, gives the
displacement per minute (§ S6). In a particular engine, volumetric effi-
ciency is affected by the speed. Suppose, for example, that the valve timing
has been adjusted for maximum volumetric efficiency at 2000 rpm. Then,
in general, as the speed increases, the volumetric efficiency decreases because
o: the greater throttling effect (fluid friction') at higher speeds. The volu-
metric efficiency as deflned by equation (t) can be made greater than unity
by use of a supercharger.
160. Closure. There are economic and advantageous uses for all kinds
of prime movers. The Otto, or spark-ignition, type is especially suited to
low power (less than several hundred horsepower) where rotative speeds up
to say 4000-6000 rpm are adaptable. Large Otto engines are more prone to
detonation than small ones, because the flame front has farther to travel in
large combustion chambers, so that this is a factor in limiting the size of
such engines. Diesel or compression ignition engines overlap Otto engines
on size, because they bum a cheaper fuel. For this reason, they are widely
used on trucks and buses although they are more expensive per horsepower.
Diesels are used in much larger units than Otto engines (up to several
thousand horsepower, maybe SOOO). and are excellent power plants for
marine use. locomotives, and small electric generating stations. Internal
combustion engines are appropriate for mills which operate on a seasonal
basis, because of the ease of maintenance and start-up. To keep per-
spective. the ICE does not compete in central station power plants, in the
largest of which, single compound steam-turbine units of more than 100.000
hp are often found.
190
10

LIQUIDS AND VAPORS

161. Introduction. Up to this point, we have considered substances


which were entirely gaseous or entirely However, one of the most
liquid.
common situations in practice concerns a mixture of a liquid and its vapor,
called a two-phase system, such as the water and steam in a boiler. Fre-
quently, either the liquid is being evaporated or the vapor is being con-
densed. Also, it often happens that a mixture of a vapor with some liquid
suspended in it enters a process or emerges from one. In any event, we
must be able to determine the properties of liquid-vapor mixtures, which is
the subject of this chapter.
In case you have not studied the preceding chapter, which is not essential
for an understanding of the remainder of this book, we should note the
difference between a vapor (imperfect gas) and an ideal gas. When any
gas is cooled, it approaches a state where it will begin to condense. When
the substance is about to condense or when the pressure is relatively high,
or both, the molecules are relatively close together, so that the molecular
forces are large and the volume of the molecules themselves is a significant
portion of the total space containing the substance. In an ideal gas, the
molecules exert no forces on one another and have zero volume. The out-
come is that, except at quite low pressures, the substances which we call
vapors behave very differently from ideal gases, and equations of state which
express their actions are likely to be complicated. To avoid the use of
these complicated equations, engineers have developed tables which give
significant properties of several substances. For illustrative purposes,
we shall usesteam most often, but the methods outlined apply to other
imperfect gases for which thermodynamic properties are available.
If a vapor is heated to a temperature far from that at which condensation
will occur, or if the pressure is reduced to low values, it will act nearly in
accordance with the ideal gas laws and may be treated as a perfect gas.
191
192 LIQUIDS AND VAPORS [Ch 10

162 Vaporizing a Liquid at Constant Pressure. To understand the states


.

in which a liquid, a vapor, and a mixture of a liquid and vapor may exist,
consider the phenomenon of a liquid being heated w hile the pressure remains
constant Let there be one pound of water m
a cylinder, Fig 104(a), and
letthe weight w be such that the pressure on the water is 100 psia Suppose
further that the temperature of the water is 32°F Now let heat be added
to the water The temperature increases, and during most of the time the
volume of the liquid increases The increase in volume causes the piston
and weight to move up, thus, work is done in moving the piston against this
pressure This work, however, is only a very small portion of the heat
added to the liquid during the rise in temperature, that is, the temperature
change substantially is a measure of the change of internal energy

Liquid

(a) Liquid (b) I lquid (c) Vapor (d) Superheated


only and vapor only vapor

Fig 104. Heating at Constant Pressure

The temperature of the liquid is soon such that it begins to boil The
temperature at which a liquid boils depends upon the pressure on it For
each pressure, there is a precise temperature that marks the boiling point of a
particular liquid This temperature is called the saturation temperature,
and when a liquid is at this temperature, it is called a saturated liquid
The saturation temperature is a function of the pressure As >ou know, water
boils at 212°F when the pressure is atmospheric, 14 696 psia At 100 psia,
water boils at 327 81 °r At 100 psia, ammonia boils at 56 05°F These
data are taken from vapor tables (§ 103) Another characteristic of the
boiling process is that the temperature of the liquid and vapor remains
constant at the saturation temperature as long as there is any liquid pres
ent * Thus, while the water in the cylinder of Fig 10t(b) and (c) is evapo-
rating, the temperature remains at 327 8l°F
In Fig 104(b), part of the water has evaporated The c> hnder has in it a
mixture of steam and water A mixture of a vapor and its liquid is called a
wet mixture or a two-phase system. The state of a wet mixture at a given
pressure is expressed by its quality, or percentage moisture The quality x
of a mixture is the per cent by weight which is vapor Thus, if the quality
is x = 75%, then m one pound of mixture, 0 75 lb is vapor and 0 25 lb
is

liquid The percentage moisture is the per cent by w eight of the mixture

* This statement assumes a condition of internal equilibrium and is substantially true,


are
but actually, small differences of temperature in the various parts of the mixture
hound to exist
:

§ 163 ] VAPOR TABLES 193

which is liquid, so that 25% moisture means that, in one pound, 0.25 lb.
is liquid and 0.75 lb. is vapor.
If the transfer of
heat to the mixture continues, all the liquid will eventu-
ally be evaporated [Fig. 104(c)]. At the point when the last drop of liquid
is evaporated, the temperature of the vapor is the saturation temperature.
Vapor which is at the saturation temperature and 100% quality is known as
saturated vapor. (The mixture in Fig. 104(b) is one of saturated water and
saturated steam and is therefore also called a saturated mixture.)
Finally, if added to the vapor, its temperature will rise,
further heat is

its volume and the vapor is said to be super-


will increase [Fig. 104(d)],
heated. Superheated vapor is vapor at any temperature above the satura-
tion temperature. thermal equilibrium exists within the supei heated
If
vapor, there can be no liquid in it. To define the state of a superheated
vapor, we usually* state the pressure and the temperature. Often we speak
of the degrees of superheat, which is the difference between the actual
temperature of the superheated vapor and the saturation temperature for
the existing pressure. For example, suppose the steam is at a pressure of
100 psia and a temperature of 500°F Since the saturation temperature of
steam at this pressure is 327.81°F (see above), the degrees of superheat are
500 - 327.18 = 172.19°.

163. Vapor Tables. The computations of the properties of vapors are


based upon the data of extensive and carefully conducted experiments.
Different sets of vapor tables may give slightly different values for the
properties of a saturated vapor or liquid at a particular pressure. Such
from which the
differences arise because of variations in experimental data
values in the tables are computed. However, so many data have now been
accumulated for steam that we can say confidently for most values that the
true property lies between certain limits which are fairly* close together.
Both the values quoted on pp. 196-199 from Keenan and Keyes (71) and
the steam tables quoted in the problem book fall within internationally
adopted tolerances. The Keenan and Keyes values are the preferred ones
in this country.
The nature of these tables is best explained by quoting briefly from the
contents. The first two columns of Tables VI and VII are corresponding
saturation values of p and l. Beginning with the third column

ty is the volume of 1 lb. of saturated liquid, the specific volume of the water at the
stated pressure or temperature;
Cf- is the change of volume undergone when 1 lb. of water evaporates to 1 lb. of
steam;
ty is the specific volume of the steam;
hi is the enthalpy of of water, the specific enthalpy;
1 lb.

h/„ is the change of enthalpy during the vaporization of 1 lb. of liquid;

hc is the specific enthalpy of (1 lb.) saturated steam;


194 LIQUIDS AND VAPORS [CA 10

«/ isthe entropy of 1 lb of saturated liquid,


S/„ is the change of entropj during the evaporation of 1 lb of liquid,
$„ is the specific entropj of saturated vapor
Uj and u, are the specific internal energies of saturated liquid and saturated vapor
respectively (in Table VII onlj)

In connection with these tables note that the volume of the liquid V/
increases as the pressure and temperature increase However, the change
of volume is practically independent of the change in pressure, unless the
change in pressure is verj large This statement is on the assumption that
the liquid is nearly incompressible a satisfactorj assumption for the average
practical problem It follows that the noted increase in volume is due to
the increase m temperature, so
liken the specific volume of (he liquid is desired Ike tohime
corresponding to the actual temperature should be found

no matter vv hat the pressure maj be * The tables show that the volume of
the saturated vapor i, is equal to the volume of the saturated liquid i /plus
the change in volume during vaporization i/,, that is

(a) ia = v, 4- v/f

By definition the enthalpy h, of saturated liquid at 32°F is zero (Table


\ I) Since the enthalpy of anj substance is defined b> A “ u + pv/J the
internal energj «/ for the saturated water at 32°F is

. 0 00026 Btu

1 he negativ e sign indicates mcrclj that the measurement is below the chosen
datum
e see from Table VI that if the pressure is 0 08854 psia water will boil
at 32T To bring about this evaporation w e must add heat to the water
the amount needed for saturated water being h fa — 107o8 Btu per lb
(Table \I) when the pressure is maintained constant The heat trans
ferred to one pound of saturated liquid to evaporate it is often called the
latent heat of evaporation, or simplj the latent heal Observe from Tables
I / and VII that the latent heat of steam Aft decreases as the saisratxxt pres-
sure and temperature increase and that

(b) ht — hf + hu
As in the case of enthalpy the datum of entropy is saturated water at
32°F, where S/ - 0 Observe that
(c) Se - S/+ s fg

* in error and
At very high pressures the ass imption of racompressib bty of water js
some liquids are more compressible than water See Table I\ and 5 168
§ 164} THE pv AND Ts PLANES 195

164. The pv and Ts Since we shall now be dealing with mixtures


Planes.
of liquids and vapors, be convenient to draw on the pv and Ts planes
it will

lines that mark the boundaries of regions representing the various phases.
From the values in the tables, we may plot points through which may be
drawn curves which are called the saturated liquid line and the saturated
vapor line or dry vapor line. See Fig. 105 for the method. We commonly
use the terms liquid line and vapor line for short, and they are so labeled in
Fig. 105, but the word saturated is always understood. The liquid line
meets the vapor line on both pv and Ts planes at the critical point (§ 172).

Fig. 105. Liquid and Vapor Lines on pv and Ts Planes. The curves are obtained as
follows: Plot the pressures against v/ (as taken from vapor tables) to find the saturated
liquid line maz in (a). For example, let the pressure be 1642.9 psia; the corresponding
Vf = 0.0236 cu. ft. (Table VI). Lay off these values in (a) and locate point a. Other
points are plotted in a similar manner. A smooth curve through these points will give
the liquid line. Points on the vapor line zbc in (a) are found by plotting pressure against
v„ as taken from the tables. Point b in (a) is for p = 1000 psia and v„ = 0.4466 cu. ft.
(Table VII). The curves on the Ts plane are found by plotting T°R against S/ for the
liquid line ndz and T°R against s„ for the vapor line zeh. See points d and e in (b) and
compare the coordinates of these points with values in Table VI. The volume scale in
(a) has been distorted because the volume of the liquid is so very small as compared with
the volume of the vapor at low pressures.

Whenever a point such as r, which represents the state of a substance, lies

within the curves mzc, Fig. 105(a), and nzh, Fig. 105(b), the substance is a
wet mixture, part liquid and part vapor. If the state point is on the liquid
line, for example, a or d in Fig. 105, the substance is a saturated liquid. If
the state point is on the vapor line, for example, h or e, the substance is a
saturated vapor. Finally, whenever the substance is a superheated vapor,
its state point will lie to the right of the vapor line; for example, point k.

Observe how the saturated vapor line on the pv plane flattens out at low
pressures, indicating that thevolume increases at an increasing rate at low
pressures. Refer to steam tables for quantitative data. In the solution of
problems concerning vapors, the Ts plane is particularly helpful. The
student should acquire the habit of sketching one or both of these planes for
Table VI SATURATED STEAM TEMPERATURES*
1

Table VII. SATURATED STEAM: PRESSURES

Ans. Specific Intern ai*


Press. Volume Enthalpy Entropy Energy

Lb. Fahr.
Sat. Sat. Sat. Sat. Sat. Sat, Sat.
Evap. Evap.

m
Sq. In. Liquid Vapor Liquid Vapor Liquid Vapor Liquid Vapor

P t *7 Tff h mm s ff V/ Ug
1.0
2.0
30
101.74
12G.0S
141.48
0,01614 333.6
0.01G23 173.73
0.01G30 118.71
69.70
93.99
109.37
mm
1013.2 1122.6
ilkWi
0.2008
5

1.6855
1.9782
1.9200
1.8863
69.70
93.98
109.36
1044.3
1051.9
1056.7
4.0 152.97 0.01G36 90.63 IKiMJ 100G.4 1127.3 0.2198 1.6427 1.8625 120.85 1060.2
5.0 1G2.24 0.01G40 73.52 130.13 1001.0 1131.1 0.2347 1.6094 1.8441 130.12 1063.1

6.0 170.06 0.01G45 G1.98 137.96 996.2 1134.2 0.2472 1.5820 1.8292 137.94 1065.4
7.0 17G.83 0.01G49 53.64 144.76 992.1 1136.9 0.2581 1.5580 1.8167 144.74 1067.4
8.0 182.86 0.01G53 47.34 150.79 988.5 1139.3 0.2G74 1.5383 1.8057 150.77 1069.2
9.0 188.28 0.01G5G 42.40 156.22 985.2 1141.4 0.2759 mmwM 1.7962 156.19 1070.8
10 193.21 0.01659 33.42 161.17 982.1 1143.3 0.2835 1.5041 1.7876 161.14 1072.2

14.69G 212.00 0.01G72 26.80 180.07 970.3 1150.4 0.3120 1.4440 1.7560 180.02 1077.5
15 213.03 0 01G72 2G.29 181.11 909.7 1150.8 0.3135 1.4415 1.7549 181.06 1077.8
20 227.9G 0 01G83 20.0S9 196.16 960.1 1156.3 0.3356 1.39G2 1.7319 196.10 1081.9
25 210.07 0.01G92 1G.303 20S.42 952.1 IMiM 0.3533 1.3606 1.7139 208.34 1085.1
30 250.33 0.01701 13.74G 2 IS. 82 945.3 1164.1 0.3680 1.3313 1.6993 218.73 1087.8

35 259.28 0.01708 11.898 227,91 939.2 1167.1 0.3807 1.3063 1.C870 227.80 1090.1
40 2G7.25 0.01715 10.49S 230.03 933.7 1169.7 0.3919 1.2S44 1.G7G3 235.90 1092.0
45 274.44 0 01721 9.401 243.3G 928.6 1172.0 0.4019 1.2050 1.6669 243.22 1093.7
50 281.01 0.01727 8.515 250.09 921.0 1174.1 0.4110 1.2474 1.G585 249.93 1095.3
55 287.07 0.01732 7.787 256.30 919.6 1175.9 0.4193 1.2316 1.6509 256.12 1096.7

60 292.71 0.0173S 7.175 262.09 915.5 1177.6 0.4270 1.2168 1.G438 261.90 1097.9
G5 297.97 0.01743 G.G55 2G7.50 911.6 1179.1 0.4342 1.2032 1.G374 2G7.29 1099.1
70 302.92 0.0174S G.20G 272.61 BOTE] 1180.6 0.4409 1.1906 1.0313 272.38 1100.2
75 307.G0 0 01753 5.816 277.43 901.5 1181.9 0.4472 1.1787 1.6259 277.19 1101.2
80 312.03 0.01757 5.472 282.02 1183.1 0.4531 1.1G7G 1.G207 281.7G 1102.1

316.25 0.01761 5.168 28G.39 897.8 118-4.2 0.4587 1.1571 1.6158 2SG.1 1102.9
BiTi^Bi 320.27 0 0I7GG 4.89G 290.56 894.7 1185.3 0.4G41 1.1471 1.G1I2 290.27 1103.7
324.12 0 01770 4.G52 294.56 891.7 1186.2 0.4692 1.137G 1.G0G8 294.25 1104.5
327.81 0.01774 4.432 298.40 8S8.8 1187.2 0.4740 1.I28R 1.C02G 298.08 1105.2
110 334.77 0.01782 4.049 305.66 883.2 1188.9 0.4832 1.1117 1.5948 305.30 1100.5

120 341.25 0 01789 3.728 312.44 877.9 1190.4 0.491G 1.0962 1.5878 1107.G
130
TfiHl 347.32 0.0I79G 3.455 318.81 872.9 1191.7 0.4995 1.0817 1.5812 1 QEJ 110S.G
353.02 0.01802 3.220 324.82 8GS.2 1193.0 0.5069 if-:-: a 1.5751 *
afcia 1109.G
150 35S.42 0.0IS09 3.015 330.51 SG3.G 1194.1 0.5138 1.5691 330.01 1110.5
3G3.53 0.01815 2.834 335.93 859.2 1195.1 0.5204 1.5G40 335.39 1111.2

170 3G8.41 0.01822 2.675 341.09 854.9 1196.0 0.5266 1.0324 1.5590 340.52 1111.9
180 373.0G 0.01827 2.532 346.03 1 19G.9 0.5325 1.0217 1.5542 345.42 1112.5
190 377.51 0.01833 2.404 350.79 846.8 1197.6 0.5381 1.0116 1.5497 350.15 1113.1
381.79 0.01839 2.288 355.36 843.0 1198.4 0.5435 2.0018 1.5453 354.08 1113.7
250 400.95 0.01865 1.8438 376.00 825.1 twim 0.5675 0.9588 1.5263 375.14 1115.8

417.33 0.01890 1.5433 393.84 809.0 1202.8 0.5879 0.9225 1.5104 392.79 1117.1
350 431.72 0.01913 1.32G0 409.69 794.2 gKjMjl 0.6056 0.8910 1.49GG 408.45 1118.0
KTiTi^K,^ 444.59 0.0193 1.1G13 424.0 780.5 1204.5 0.G214 0.8630 1.4844 422.6 1118.5
450 456.28 0 0195 1.0320 437.2 767.4 0.635G 0,8378 1.4734 435.5 1118.7
4G7.01 0.0197 0.9278 449.4 755.0 1204.4 0.6487 0.8147 1.4034 447.6 1118.6

550 476.94 0.0199 0.8424 460.8 743.1 IWWl 0.6608 0.7934 1.4542 458.8 1118.2
486.21 0.0201 0.7G98 471.6 731.6 t'iiK&i 0.6720 0.7731 1.4454 4G9.4 1117.7
t 494.90 0.0203 0.7083 481.8 720.5 0.G826 0.7548
0.7371
1.4374
1.4296
479.4
488.8
1117.1
1116.3
503.10 0.0205 0.6554 491.5 709.7 1201.2 0.6925
800 518.23 0.0209 0.5G87 509.7 688.9 1198.6 0.7108 0.7045 1.4153 500.6 1114.4

900 531.98 0.0212 0.5006 526.6 668.8 1195.4 0.7275 0.6744 1.4020 523.1 1112.1
1000 544.61 0.021G 0.445G 542.4 649.4 1191.8 0.7430 0.64G7 538.4 1109.4
u !

r rfl
1200 567.22 0 0223 0.3G19 571.7 611.7 1183.4 0.7711 0.5956 IfKit
Tv 5G6.7 1103.0
1400
1500
587.10
596.23
0 0231
0.0235
BamE
0.27G5
598.7 574.7 1173.4
1167. 9
0.79C3
0.8082
0.5491
0.5269
Kvkl ill 592.7
1.3351 605.1
1095.4
1091.2
611.6 556.3

2000 G35.82 0.0257 K£fj£2§ 671.7 463.4 1135.1 0.8619 0.4230 1.2849 662.2 1065.6
1020.3 0.9731 0.1885 1.1G15 783.4 972.7
tww G95.3G
705.40 t'X'Hvl
0.0858 802.5
0.0503 902.7
217.8
0 902.7 1.0580 0 1.0580 872.9 872.9

197
Table Vm SUPERHEATED STEAM

m
) 2

Table VUE. SUPERHEATED STEAM ( Continued

Asa. Press. TcifPEEATUEE —DEGREES FaHREJTHEIT


Lb./Sq. In. 500° 550° 600° 620° 640° 660° 680° 700° 800° 900° 1000°
(Sat .Temp.)

x 1.1231 1.2155 1.3005 1.3332 1.3652 1.3967 1.4278 1.4584 1.6074 1.7516 1.8928
450 k 1233.4 1272.0 1302.8 1314.6 1326.2 1337.5 1348.8 1359.9 1414.3 2467.7 1521.0
(456.23) 3 1.5095 1.5437 1. 5735 1.5845 1.5951 1.6054 1.6153 1.6250 1.6699 1.7108 1.74S6
X 0.9027 l.OSOO 1.1591 1.1893 1.2188 1.2478 1.2763 1.3014 1.4405 1.5713 1.6996
500 h 1231.3 12G6.8 1298.6 1310.7 1322.6 1334.2 1345.7 1357.0 1412.1 1460.0 1519.6
(467.01) a 1.4919 1.5280 1.5588 1.5701 1.5810 1.5915 1.6016 1.6115 1.6571 1.6982 1.7363
X 0.8852 0.9686 1.0131 1.0714 1.0989 1.1259 1.1523 1.1783 1 3038 1.4241 1.5414
550 h 1223.7 1261.2 1294.3 1306.8 1318.9 1330.8 1342.5 1354.0 1409.9 1464.3 1518.2
(476.94) a 1.4751 1.5131 1.5451 1.5568 1.5680 1.5787 1.5890 1.5991 1.6452 1.6868 1.7250

X 0.7947 0.8753 0.9463 0.9729 0.9988 1.0241 1.0489 1.0732 V WfJ 1.4006
600 h 1215.7 1255.5 1289.9 1302.7 1315.2 1327.4 1329.3 1351.1 mtlfri J 1516.7
(486.21) a 1.4586 1.4990 1.5323 1.5443 1.555S 1.5667 1.5773 1.5S75 1.7147
J
I'
X 0.7277 o o rt 0.8177 0.8)11 0.8639 0.8860 0.9077 1.0103 1.1082 1.2024
700 h 1243.2 1280.6 1294.3 1307.5 1320,3 1332.8 1345.0 1403.2 1459.C 1513.9
(503.10) a 1.4722 1.5034 1.5212 1.5333 1.5449 1.5559 1.5665 1.6147 1.6573 1.6963

X 0.6154 0.6779 0.7006 0.7223 0.7433 0.7635 0.7833 0 8763 0 9633 1.0470
£00 h 1229.8 1270.7 1285.4 1299.4 1312.9 1325.9 1338.6 1398.6 1455.4 1511.0
(513323) » 1.4467 1.4863 1.5000 1.5129 1.5250 1.5366 1.5476 1.5972 1.6407 1.6801

r 0.5264 0.5873 0.6089 0.6294 0.6491 0.6680 0 6863 0.7716 0.8506 0.9262
000 h 1215 0 1260.1 1275.9 1290.9 1305.1 1318.8 1332.1 1393.9 1451.6 1503.1
(S3 1.9 S) a 1.4216 1.4653 1.4800 1.4938 1.5066 1.5187 1.5303 1.5814 1.6257 1.6656

X 0.4533 0.5140 0 5350 0.5516 0.5733 0.5912 0.6034 0.6878 0.7604 0.S294
1000 h 1198.3 1248.8 1265 9 1281.9 1297.0 1311.4 1325.3 1389. 1448.2 1505.1
(544.61) a 1.3961 1.4450 1-4610 1.4757 1.4893 1.5021 1.5141 1.5670 1.6121 1.6525

X 0.4532 0.4738 0.4929 0.5110 0.5281 0.5145 0.6191 0.7503


1100 h 123S.7 1255.3 1272.4 1288.5 1303.7 1318.3 13S4.3 1444.5 1502.2
(556311) a 1-4251 1.4425 1.4583 1.4728 1.4S62 1.49S9 1.5535 1.5995 1.6405

X 0.4016 0.4222 0.4410 0 4586 0.4752 0.4909 0.5617 0.6250 0.6843


1200 h 1223.5 1243.9 1262.4 1279.6 1295.7 1311.0 1379.3 1440.7 1499.2
(567.22) a 1.4052 1.4243 1.4413 1.4568 1.4710 1.4843 1.5409 1.5879 1.G293

r 0.3174 0.3390 0.35S0 0.3753 0.3912 0.4062 0.4714 0.5S05


1400 h 1193.0 1218 4 1240 4 1260.3 1278.5 1295.5 1369.1 1433.1 1493.2
(5S7.10) * 1.3639 1.3877 1.4079 1.4258 1.4419 1.4567 1.5177 125606 1.6093

r 0.2733 0.2936 0.3112 0.3271 0.3417 0.4034 0.4553 0.5027


1600 h 11S7.8 1215.2 1238.7 1259.6 1278.7 135S.4 1425.3 14S7.0
(601.90) a 1.3489 1.3741 1.3952 1.4137 1.4303 1.4964 1.5476 1.5914

V 0.2407 0.2597 0.2760 0.2907 0 3502 0.39 86 0.4421


1800 h 1185.1 1214.0 1238.5 1260.3 1347.2 1417.4 14SQ.8
(621.03) a 1.3377 1.3638 1.3855 1.4044 1.4765 1.5301 1,5752

t 0.1936 0.2161 0.2337 0.2489 0.3074 0.3532 0.3935


2000 h 1145.6 1184.9 1214.8 1240.0 1335.5 1409.2 1474.5
(635.82) a 1.2945 1.3300 1,3564 1.3783 1.4576 1.5139 1.5603

V 0.1484 0.1686 0.2294 B130 0.3061


2500 h 1132.3 1176.8 1303.6 13S7.8 1458.4
(668.13) » 1.2687 1.3073 1.4127 1.4772 1.5273

V 0 0984 0.1760 0 2159 0.2176


2000 h 1060.7 1267.2 1305.0 1441.8
(695.36) a 1.1966 1.3690 1.4439 1.49S4

V 0.25S3 @3311 022238


3200.2 h 1250.5 1355.2 1434.7
(705.40) > 1 .3503 1.4S74

199
)

200 LIQUIDS AND VAPORS [Ch 10

Table IX. COMPRESSED WATER


Taken with permission from Table 4 of Keenan and Kejcs, Thermodynamic
Properties of Steam, published by John Wiley <L Sons, Ine

Temperature •r - 32* 100* 200* 300° 400° 500* 600*


O 08854 O 9492 11 520 07 013 247 31 080 f 1542 9
Saturated 0 010022 ) 010132 0 01CC34 ) 017449 )018639 ) 020432
liquid hy - 0 07 97 167 90 209 59 374 07 487 82 617 0
•j " 0 0 12948 0 293S2 0 43094 0 56638 0 68871 0 8131

Abt press
(Sat temp
-1 1 -1 1 -1 1 -1 1
200 (A
- +0 01 +0 54 +0 41 +0 23
<381 70) (» - s/) I0» +0 01 -0 05 -O 21 -0 21

»l)10‘ —5 7 -6 1 -5 4 -0 0 -8 7 -6 4
1000
(544 01)
(A
(i
-

M
«/)10'
+2
+0
09
15
+2 70
-0 53
+2
-1 20
21 +1
-1
75
04
+0
-2
84
00
-0
-1 41
14

»/)10* -11 0 -9 0 -10 8 -13 8 -19 5 -27 8 -32 6


2000 (h - hi) +6 07 +5 31 +4 51 +3 04 +2 03 -2 5
(035 82) (« - •/) 10* +0 22 -1 18 -2 39 -3 42 -4 57 -5 58 -4 3

-10 3 -14 7 -10 0 -20 7 -30 0 -47 I -87 9


3000
(095 30)
(5 - M +0
+0
00
2S
+7
-I 79
88 +0
-3
70
50
+5
w5
49
12
+3
-7
33
03 -9 42 12 4

each problem, showing in each case the liquid and \apor lines, drawn
freehand

166. Internal Energy. Ordinarily, vapor tnbles do not gi\c internal


energies, yet tlie> arc needed now and then Values of internal energy are
computed from it = ft - pv/J Tor a saturated liquid, we may write

. h, - Ej-' Btu/lb ,

where the \ allies of ft/ p, (saturation pressure) arc taken from the tables
and
in accordance with the actual temperature of the liquid Howe\er, unless
the temperature is quite high, the corresponding saturation pressure will be
so low that the term p,i//J is negligible Customarily, we therefore con-
sider that the internal energy of the liquid at low pressures is equal to the
enthalpy of the liquid, u, ~ hr
The internal energy of saturated taper is

«. Btu/lb,

where ft* and ve are found m


the tables according to the pressure p (or the
corresponding temperature) Tor example, at 100 psia, the internal energy
of saturated uapoi is (compare with Table VII)
§ 166] PROPERTIES OF A WET MIXTURE 201

^ = U87 2 _ (100)(144)(4.432) = n05 2 Bfcu/lb

166. Properties of a Wet Mixture. Consider any liquid under some pres-
sure p psia and at 32°F. If p is greater than the saturation pressure cor-
responding to 32°F, the liquid is not saturated and is called a compressed
liquid. If the datum of enthalpy be taken as zero for a saturated liquid, then
thiscompressed liquid does not have zero enthalpy, since h — u + pv/J.
However, the pv/J term is small unless the pressure is quite large, and for the
purpose of a visual aid in remembering the properties, we may consider it as
negligible. In this event, the enthalpy of
the liquid at t, Fig. 106, is nearly zero,
and if it is heated at constant pressure to
the boiling point (saturation temperature)
atto, the area under the constant-pressure

curve Im on the Ts plane represents heat


Q = Ah for p = C. That is, the area
under Im represents closely the enthalpy
of the liquid at to. The s coordinate of to
is the entropy of the liquid exactly. Now
if part of the liquid is evaporated, the
increase in any property is a part of the
change which that property undergoes
during complete vaporization. Take any
Fig. 106. Properties of Wet
property for illustration, say, s. The
Mixture.
entropy of a two-phase system represented
by point 1, Fig. 10G, is the entropy s/ of the liquid phis the fraction z of
the liquid evaporated (x = quality) (imes the entropy sfa of evaporation, or

(d) Si = s/ + xs /0 .

In terms of the percentage moisture y = 1 — x, start at point n. The


entropy at 1 is equal to the entropy s a of saturated vapor at n minus the
fraction y of the liquid which has not evaporated limes the entropy of
evaporation, or

(e) si = s„ — ys /0 .

All properties of a two-phase system can be obtained by similar reasoning.


For one pound,
(f) h = h„ - ylifgi h = hf + xhfg,
(g) v = v„ - yv/ a ,
v = vf + xv /0 .

[man quality] [low quality]

The expressions with y give more accurate slide-rule answers when the qual-
ity high (above 75%) the expressions with x gives more accurate answers
is ;

when the quality is low (below 25%). There is little difference in the intcr-
203 LIQUIDS AND VAPORS [Ch 10

mediate range If the internal energy of a wet mixture is needed, compute


h and v from equations (f) and (g), and then u — h — pi/J Btu per lb
It takes two coordinates to locate a point in a plane Since, in the wet
region, the pressure and temperature both locate the same line it is necessary
to give still another property in order to define the state point The
entropy or the volume will serve, but it is common practice to define the
state of the two-phase sjstem by giving cither the pressure or temperature
and the qualitj or moisture content

167. Properties of Superheated Vapors Modern vapor tables usually


give the necessarj properties of superheated vapors In Table Mil for
steam, we find the v alues of ,h, and s the v alues of h and $ being measured
i

abov e saturated water at 32°F In order to define the state of a substance


in the superheat region, it is necessarj to specifj tw o properties Anj tw o
maj be used, but the most common practice is to g« e the pressure and
temperature
At low pressure, when the state of a superheated vapor falls outside of
the limits of the superheat table the vapor maj be treated as an ideal gas
For practice in interpolation and to note the differences in values in other
tables the student should solve the following example using the tables m
Problems on Thermodynamics if available

Examples (a) What is the interna! energj (above saturated liquid at 32°F) of
steam at 100 p«ia and COOT’
solution From Table \ III, we find h — 1329 1 and i = 6 218

„ . * - „// - ,3* .
- _ mt Btu/ b ,

(b) What the entropv of steam at 297 psia and 512°F 7


is

solution This problem illustrates the method of making a double interpolation


Bj comparing the temperature gi\ en with the saturation temperature corresponding
to 297 lb we note that the steam is superheated From more complete tables
than Table VIII we hav e taken the follow mg values of entropj

At 500°F At 520°F
For p = 295 s = 1 5725 For p = 295 a = 1 5847
For p - 300 s = I 5701 For p 300 a 1 oS24

Difference “ 0 0024 Difference = 00023


(HX00024) =000090 2
( i) (0 0020) -00009
For p - 297, s = 1 5715 For p = 297 s = 1 5S3S

At p = 297 p'i
For f 500°, s = 1 5715
For l = 520° a = 1 5S3S
Difference 0 0123
(
12
5o)(0 0123) = 0 0074
For t 512° s — 1 5789 Ans
204 LIQUIDS AND VAPORS [Ch 10

However, at low temperature and pressure, water acts very much as an


incompressible fluid, in which case the points b, c, and B, become quite close
together These three points are defined by an isothermal compression ah,
an iscntropic compression ac, and a constant volume compression aB As
these points approach each other at low pressures, their properties approach
equality One cannot say exactly where a line should be drawn, but for
pedagogical purposes, let us say that for pressures and temperatures of steam
below 400 psia and ^D0°F, use properties of saturated liquid, do not correct
for the compressed-liquid effect

If a liquid is pumped re\ersibly in steady flow to a higher pressure, the


work is (A K = 0,AP = 0)

(h) 11 = — (AA)„ [equation (32)1

where the subscript s indicates constant entropy This change can be


evaluated accurately with the properties of subcooled liquid In the
absence of known properties, we can obtain a good estimate of equation
(h) for liquids which are nearly incompressible by the relation from equation
(17), § 42 (Q - 0)
V
(i) Ah = vdp - ~j^> I 8 - c l

j J
Observe that v A p is the rectangular area naBm, Fig 107 (a) ,
a = t>a — »i

an d Ah = ha — h, — ha — h/t where the subscript / indicates properties of

saturated liquid Thus, from equation (i), we have

(j) h, - A, + ^ - h, + Mp
-
fr-ei

in which hf anu i>/ are taken for saturated liquid at the temperature of the
liquid before compression, is the actual pressure, and p, is the saturation

pressure The point B, at a temperature slightly higher than that at a,


§ 170} OTHER VAPORS 205

isnot very far from c, the state after isentropic compression, and the approxi-
mation of equation (j) will usually be better than no allowance for the
subcooling effect.

169. Examples : Enthalpy of Compressed Water, (a) Given water at 100°F and
3000 psia. Wlint is its enthalpy?
solution. From Table IX, we get

h- h; = 7.88, or h= 67.97 + 7.88 = 75.85 Btu/lb.

Note that this is at state b, Fig. 107.


water at 100°F (before compression) is compressed isentropically
(b) If saturated
to 3000 psia, what is the final enthalpy?
solution. From the chart in Keenan and Keyes, we read h c — h!a — 8.9 Btu/lb.,
or
K= 07.97 + 8.9 = 76.87 Btu/lb.

See Fig. 107. If you do not have the chart referred to, try the approximation
given by equation (j); Vj
= 0.01613 (Table VI);

(0.01613) (3000 - 0.949) (144)


hu 07.97 + = 67.97 + 8.96 = 76.93 Btu/lb.
778

This value good agreement with the one obtained from the chart, but note that
is in
the temperature is low. At high temperatures, water becomes relatively compressi-
ble and the agreement between the results would not be so good.

Other Vapors. Properties of many other vapors are available in


170.
tables and charts (see the problem book). Before using an unfamiliar set of
tables, study them carefully. There are different arrangements, different
nomenclature, and different symbols. They are not all based on the same

Table X. SATURATED AMMONIA

Abs.
Specific volume Heat Content (Enthalpy) Entropy
Pres.

Temp,
°F
Latent
Psi Liquid Vapor Liquid Vapor Liquid Vapor
V Vf va hf kg S/ Sff
h ta

-60 5.55 0.02278 44.73 -21.2 589 6 610.8 -0.0517 1 4769

-40 10.41 0.02322 24.86 0.0 597.6 597.6 0.0000 1.4242


0 30.42 0.02419 9.116 42.9 611.8 568.9 0.0975 1.3352
20 48.21 0.02474 5.910 64.7 617.8 553 1 0.1437 1.2969
40 73.32 0.02533 3.971 86.8 623.0 536.2 0.1885 1.2618
60 107.6 0.02597 2.751 109 2 627.3 518.1 0.2322 1.2294
80 153.0 0.02668 1.955 132 0 630.7 498.7 0.2749 1.1991
90 180.6 0 02707 1.661 143.5 632.0 488.5 0.2958 1.8146
100 211.9 0.02747 1.419 155.2 633.0 477.8 0.3166 1.1705
208 LIQUIDS AtrD VAPORS [Ch 10

datum state, which does not matter one does not become confused with
if
signs For example, the properties of refrigerants, substances which are
used m refrigerating cycles, are generally measured from a state of saturated

Fig 108 Various Fluids on Ts Plane

liquid at — 40°F—as in X for ammonia, extracted from Tables of


Table
Thermodynamic Properties of Ammonia, U S Bureau of Standards * Refng
eration temperatures go below — 40°F in some industrial processes in which
case the enthalpy and entropy of saturated
liquidare negative The negative signs
mean only that measurement is below the
datum Carry the sign into the basic equa-
tions and stick to the rules of algebra
The saturated liquid and vapor lines for
several substances arc shown in Fig 108, as
plotted for one pound Notice the varia-
bility of the latent heat of evaporation, pro-
portional to the distance between the liquid
and vapor lines ata particular temperature
Sulfur dioxide, carbon dioxide, ammonia
and freon 12 are refrigerants (H*0 is also
used as a refrigerant) Mercury (JFfg) is
Fig 109 Saturation Curve
used to generate power in turbines in so-
for Acetic Acid
called binary-vapor cycles Mercury has
a small latent heat per pound, which means that the work which can be
obtained per pound between given temperatures is small relative to that
* Bulletin No 142 may be obtained from the Superintendent of Documents Govern-
ment Printing Office, Washington, DCIn view of its reasonable price students are
urged to obtain a copy
§ 171} THE Ts DIAGRAM INCLUDING THE SOLID PHASE 207
which can be obtained from steam, but the saturation pressure of mercury
is low at high temperatures.

Xot all saturated-vapor curves slope downward toward the right. Some
show a double curvature, for example, benzene; and several such curves
slope downward toward the left as, for example, the saturated-vapor
curve of acetic acid (Pig. 109). If a substance with this characteristic
undergoes an isentropic expansion, it becomes dryer or more highly super-
heated, whereas other substances discussed here become wetter or lose
superheat.

171. The Ts Diagram Including the Solid Phase. The solid phase is of
no concern in power generating cycles, but it is significant in many other
waj's. If heat is extracted at constant pressure from a superheated vapor

Fig. 110. Ts Diagram with Solid Phase. Simplified and distorted. Notice that the
triple point on this plane is a line.

at a, Fig. 110, first the superheat is removed, ab, then the vapor is con-
densed, bd. After the liquid is cooled below its saturation temperature at
d, it is a compressed liquid; and it begins to freeze at this particular pressure
when the state point passes e; solidification is ef (melting is fe). If heat is

extracted from the solid at/, Fig. 110, its temperature decreases along some
path fg. The triple line (32.02°F for steam) is the temperature at which the
solid, liquid, and vapor phases can exist together in equilibrium. Below
the triple line, also called the triple point, the solid sublimates to the vapor
phase. The area under nr is the latent heat of fusion (enthalpy of fusion)
at the pressure of the triple point. The area under fe is the heat of fusion
at the pressure p.
Any three properties of a substance may be used to construct a solid, or
surface in space. A Tps surface is shown in Fig. 111. Notice that what is
shown on the Ts plane, Fig. 110, is the projection on that plane of the sur-
face in space (although Figs. 110 and 111 are not drawn to the same scale).
A constant pressure line ab ... g and other points are lettered correspond-
ingly in both illustrations. Table 5 of Keenan and Keyes (71) gives prop-
erties of UK) in the solid-vapor phase.
208 LIQUIDS AND VAPORS [Ch 10
172. Closure. The purpose of this chapter is to acquaint the reader with
the characteristic phenomena of liquids and vapors and the use of the \apor
tables Carefully study available vapor tables If you should happen to

Fig 111 Tps Surface Simplified After (72) The surface, done in detail for a
particular substance, is the locus of all equilibrium states

be using two different tables for the same substance, differences in a particu-
lar property should not alarm you However never mix values from differ
ent tables in a particular problem because this practice may lead to a serious
error
You perhaps recall from your study of physics, that the critical tempera
ture, Te in Fig 111 ,
is the temperature abo\ e which a gas cannot be liquefied
Representative critical temperatures are

Ammonia, 270°F, Methane, — 11G°F,


Carbon Dioxide, 88°F, Nitrogen, 227°R,
Freon 12, 232°F, Octane, 5G5°F,
Helium, 9 33°R, Oxygen, 278°R,
Hydrogen, 59 8°R, Sulfur dioxide, 315°F
11

PROCESSES OF VAPORS

173. Introduction. The basic energy relations for the processes as defined
for perfect gases also hold for vapors. In fact, all previous energy equations
in terms of the general symbols ]V, Q, H, h,U, u, K, P, apply to any sub-
stance under the circumstances specified. The equations derived from the
assumption of an ideal gas ( pv = RT, Joule’s law, etc.) do not hold. Thus,
the reader’s problem is largely one of classifying in his own mind the con-
ditions which do and do not apply to vapors and of learning to solve energy

equations with the use of vapor tables which is the purpose of this chapter.
Remember that the areas on the pv plane under the curve of an internally
reversible process represent fp dv, even though the integral has not been
made, and that this area is the work of a nonflow process. The area behind
this same curve, to the p axis, is the fv dp. The areas on the Ts plane under
the curve of an internally reversible process represent heat transferred to
(state point moving rightward) or from (state point moving leftward) the
system. There is no change in the convention of signs previously estab-
lished. The energy equations dQ = du + p dv/J and dQ = dh — v dp/J
may be applied to any internally reversible processes and, for convenient
reference, the steady flow equation may be written

(9) W = hi — hi + Ki — K, + Q,

in which K — v 2 / (2g„J), the kinetic energy, and there is no change in


potential energy, A P = 0.

174. Constant Pressure Process. (See § 67.) To define a process com-


pletely, it is necessary to give the location of two points, with a description
of the path joining these points, or enough information to find two coordi-
nates of each of these two points. When the defining relation is known
(say, pressure constant), this relation is a coordinate of each of the two points
209
1 ,

210 PROCESSES OF VAPORS [C7i 11

m questionIn addition to this defining relation another property at


each point is needed Suppose in this case that point 1 is in the wet region
and point 2 in the superheat region Fig 112 Tilth the pressure known a
point in the wet region is generallj defined bj giv ing its quality x or per
centage moisture y (Point 1 could be defined also bj gi\ mg the entropj or

Tig 112 Reversible Constant Pressure ( Isobanc ) Process The area a 1 2 b (a) m
represents the work done m
a nonflow process. The area 1 2 n in (b) represents the m
transferred heat

the internal energj ) Similarlj point 2 in the superheat region is generallj


but not necessarilj defined bj giving its temperature (pressure knot n)
The work of a reversible nonflovv process at p C is
(a) Jp di — p( i
— 1 ) ft lb /lb of v apor

where the specific v olume i * of the superheated apor is taken directlj from
v

the superheat tables tj is the specific volume of the wet vapor

ti = («» — y/»)i = 0/ +
The heat transferred is

„ fpdt Piia p fi
Q — Am + — +
.
,
-
-j - it ui ,
—j j
[ant reversible process] [p C]

(b) Q = h hi Btil/lb —
[flow or nonflow p CJ

where hs is taken from the superheat tables (for Fig 112) and hi is found
from h, - (h/ 4- x,h/e), or (h, - yh/t ) The change of internal energj
is found from the enthalpies

<<=) (h, - £/) (*,


- Btu/lb

176 Constant Volume Process (See § 65 ) Choose point I in the


superheat region point 2 in the wet region Fig 113 In the solution of
nroblems it is often necessarj to u^e the defining relation of the process in
§ 176] ADIABATIC AND ISENTROPIC PROCESSES 211

this case,

(d) vt = a. = (v„ — y«Vf 0)» — (vf + .r«v/ 0 ) 2 cu. ft./lb.

From this equation, the quality may


be found, given v,. Knowing this
.r 2

quality and an additional property, determine other properties of the vapor

Fig. 113. Reversible Constant Volume ( Isometric ) Process. The area m-2-l-n
represents the transferred heat and also Au in this process. The dotted line abc repre-
sents a constant pressure line.

at 2. Or if is known, the volume i>« = Vj can be found, after which it is

usually possible to find in the superheat tables other properties at 1.

Since the fp dV is zero, the energy equntion is

(e) Q — An — — Hi = ~ (^h ~ Btu/lb.

Some tables give values of the internal energy.

176. Adiabatic and Isentropic Processes. (See § 70.) For the adiabatic
process including the isentropic, <3 = 0, the nonflow energy equation
Q = Ait + TP gives

(f) W= —A u or TF = Mi — m2 or IF = Mi — ux,
[.VO.VF1.OW] [itEVEItSIBI.E] [lUnEVEBSIBLE]

where the u’s are generally obtained from u = h — pv/J and the states

Fig. 114. Isentropic Process.

are as shown in Fig. 114. The steady flow' equation with Q — 0 and AP = 0
becomes
t

212 PROCESSES OF VAPORS \Ch 11

(g) W= -(Ah4* AK) or W=


hi — ht, or W = hi - h 3
[STEADY PtOW] [REVERSIBLE, AK = 0] [IRREVERSIBLE,
AK = 0)

An example of steady flow to which this equation applies is a steam turbine


If the process is reversible (lsentropic),

(h) «i =
** (s„ = —
ys/ a)t = ( «/ xsj,) 3f +
[reversible nonflow or steady flow]

in which it is presumed that state 2 is in the iv et region, Fig 1 14


177 Example 1 lsentropic Process Three pounds per second of steam expand
iscntropicnlly from pi = 300 psia nnd i x = 700°F to /, = 200°F Find x, and the
works for steady flow (AK 0) and nonflow processes

solution See Fig 114 From Tabic VIII, we find *i = 1 6751, h t ** 1308 3,
and t»i = 2 227 From Table VI, we get p 3 = 11 526 psia and
8/2 = 0 2038, h, - 1C7 90, v n = 0 016C3,
= 1 4821, h/„2 = 977 9, v,„ - 33 62,
8,2 = 1 7762, h ,2 = 1145 9 v„ = 33 64

(One often saves time by writing down all the table values at once ) Using st ~ s3
= (s» - A 8/.)* we have
1 6751 = 1 7762 - 1 4824yj, whence, yi = 6 82% and *, = 93 18%
hi = (h , - yktg)i - 1145 9 - (0 0682) (977 9) = 1079 2 Btu/Ib

The work of the steadj flow lsentropic is

IF hi - hi - 1368 3 - 1079 2 = 289 1 Btu/lb ,


or 867 3 Btu/sec

for 3 lb /sec Next, find the internal energies at states 1 and 2

«, - A, - tf - 13G8 3 — lamwoxv .
V, - - y,v,„ -- <0 0682X33 62) - 31 35 cu
33 64 ft /lb

«, - 1079 2 - (» 526)^X3135) , m2 3 Btu/lb


7 <8

The nonflow lsentropic work is

IF = mi - w s = 1244 6 - 1012 3 = 232 3 Btu/lb.

or 696 9 Btu/sec for 3 lb /sec.


§ 179 ] EXAMPLE: IRREVERSIBLE ADIABATIC PROCESS 213

178. Lost Work and Unavailable Energy.


As a simple and convenient
partial measure comings of an actual work-producing process or
of the short
cycle, we may use lost work, which is
defined as the ideal work IF of the cor- S' "'n. id
responding process (or cycle) minus the / Pl \/ \
actual work IF' of the process (or cycle) ;
/ \/
(i) Lost work = TF — TF'.
~ 2
~
(rosiTiiE n oiiK.s] T„ d 'V/fc

For steady flow adiabatic processes 1-2 Lost Work = 2-2-n-m


Decrease A £T„=
//
—*//
and 1-2' between pressures p i and ,
Increase A£r„=7 A,=dci!m s 2
pi, Fig. 115, IF = hi — hi and TF'
r>
/// -V
m n s
= hi — hi’, for A K — 0; then
Fig. 115. Lost Work and Unavail-

(j) Lost work = hi — hi — (hi — h?) able Energy. Irreversibility results


= hi' — hi Btu/lb.,
in increasing entropy; point 2' is
to the right of point 2. If T„ = TV
with 2' as shown, the increase of
which is seen to be represented by the unavailable energy T„ As is less than
area under the constant pressure line the lost work by the triangular area
a-b- 2'. If 2' is in the wet region and
2'-6-2. Note that the increase of un-
T„ =Ti = T-', the increase in una-
available energy T„ As as pre-
A/?„ = vailable energy is the same as the lost
viously defined is good measure of
also a work. If T„ < T; as shown, the loss
of available energy is always less
the effect of irreversibility when com-
than the lost work.
pared with the original available energy,
but it should not be confused with lost wc rk. Read the caption to Fig. 115.

179. Example: Irreversible Adiabatic Process. Let the data be the same as in

§ 177 except that an irreversible adiabatic expansion occurs in a steam turbine to


state 2' on the saturated vaporline,£»' = 100%.
T s. J What is the horsepower output? W'hat per-
S j
/ Pl \/\ centage of the ideal work is lost? See Fig. 116.
/ ~ \ ' solution. The work of the irreversible ex-
/ \\ / pansion (A K= 0) is

IF' = hi -h? 1368.3 - 1145.9


= 222.4 Btu/lb.
Work Lost Because
of Irreversibility, - and 42.4 Btu/hp-min.,
For a flow of 3 Ib./sec.
Area m-2-2^n
we get
(3))(22|4)(6_0) =
Fig. 116.
942 h p .

42.4

The lost work is IF - IF' = 289.1 - 222.4 = 66.7 Btu/lb., vhich is

= 23.1 %
of the ideal work between the given pressures.
,

316 PROCESSES OF VAPORS \Ch 11

main If a sample of this high-pressure steam is throttled to a lower pres-


sure, then, within limitations, it will become superheated, point 2 In
the superheated condition, we can measure its pressure and temperature
thus locating precisely point 2 With the pressure and temperature known,
the enthalpy ht maj be obtained from the superheat tables Then
(t) ht = (hg - yihft )i Btu/lb

from which Xi = —
y t maj be found bj looking up h a and h /t correspond-
1
ing to some known main line pressure p lt Fig 119
The drawback in using equation (t) is that the actual temperature and
pressure at 2 is nearlj always such
that a double interpolation in the
superheat tables is necessary If the
highest degree of accuracy is not re-
quired, the computation maj be sim-
plified with the use of the specific heat
of steam at constant pressure In
Fig 119 we note that the enthalpj at
2 is equal to the enthalpj of saturated
Fig 119 Throttling Process (Irre-
steam h 0 plus the enthalpj to super-
versible) (Area a-b-d-l-e) (area heat the steam from g to 2 The heat
a b c 2 f) because these areas represent, to superheat the steam Q* may be
respectively, h, and hi (closely)
computed from Q, k = ucp At because
the points g and 2 are on a constant pressure curie For the conditions
encountered in a calorimeter c P for steam maj be taken as 0 48 Btu per
lb °F Thus for one pound,
(u) ht == ht 4- Cp At — ha» + 0 48 At,t
where At.h = degrees of superheat at point 2, Fig 119 and is obtained from

At,K = (actual temperature at 2) — (saturation temperature corresponding to p,)

183 Calorimeters In finding the quality of the steam as e\plained


in § 182, to flow into an instrument known as the
we allow a sample of steam
throttling calorimeter, Fig 120 thence to the atmosphere Before tern
peraturc readings are taken at the calorimeter the steam should flow long
enough to obtain internal equilibrium that is, long enough for all parts to
be heated to a temperature w hich remains constant To maintain adiabatic
conditions the instrument should be well insulated and for reliable results,
the steam should have at least 10° of superheat m the calorimeter If the

steam is mitiallj very wet, it maj not become superheated in the calorimeter
in which case the calorimeter readings are meaningless, the temperature
and pressure being those at saturation This condition is illustrated bj
process tnn. Fig 119 The minimum quality that can be reliably measured
by the throttling calorimeter depends upon, the initial pressure pi (the final
§ 183] CALORIMETERS 217

Amend, New Yorl.

Fig. 120. Throttling Calorimeter. The sampling Fig. 121. Separating


pipe, which is in the steam main with holes facing Calorimeter. The wet
upstream, should be as far as possible downstream steam, entering at the top
from elbows and valves in order for the sample to be through the valve, under-
representative, preferably in a vertical steam line. goes a sudden reversal of
The steam flows into the inner chamber. A ther- direction of motion when
mometer gives the temperature. The pressure is it strikes the baffle plate
virtually atmospheric. Flowing out of the inner 14. This causes the water,
chamber at the top, the steam turns downward into which has a greater inertia,
the annular space and leaves the instrument at the to separate from the mix-
bottom. The hot steam surrounding the inner ture. The water collects
chamber (a patented feature) is an aid in insulating in the inner chamber and
against loss by radiation. can be measured by the
gage glass 10. The steam
moves upward from the
inner chamber 3 and down-
ward in the annular space 4.
pressure p< being nearly atmospheric), and varies from about 97% for
Pi ~ 50 psig to 94% for =
600 psig.
If the steam is too wet to be measured by a throttling calorimeter, a
separating calorimeter, Fig. 121, or a combination separating and throt-
tling calorimeter may be used. In the separating calorimeter, the moisture
is separated from the steam. The amount of collected water is determined.
The corresponding mass of dry steam is calculated from flow equations
applied to a nozzle (see Chapter 13) or read from a gage; or the steam leaving
218 PROCESSES OF VAPORS [Ck
the nozzle may be condensed and weighed The qualitj is then computed
from the formula

_ (mass of drj steam)


x
(mass of water) + (mass of drj steam)
Incomplete separation of water and steam and inaccuracies m the devices
and methods used to find the w eight of flow of drj steam will result an m
incorrect e\aluation of the quality

184 Mollier Diagram Using anj two point functions of substances


such as temperature and entropj we maj construct to scale a diagram con
taming a senes of lines each representing a constant pressure another senes
where each line represents a particular \olume and another senes for con
stant qualitj etc On the Ts
plane Fig 122 hat e been plotted
set eral constant pressure constant
tolume and constant quaht) (also
superheat) lines Thus a po nt
maj be located bj any tw o of these
coordinates after which all other
properties pictured maj be read
from the diagram The degree of
accuracy of the reading will depend
upon the spacing of the lines and
the size of the diagram
Tig 122 Temperature Entropy Diagram More contement than the Ts
for Steam The point a for example repre diagram for the solution of certain
sents steam at a pressure of 100 psia and a problems in practice is the Moll er
quality of 80% These two coordinates
could be used to locate a and then the diagram a chart on w hich enthalpj
volume could be determined from the is the ordinate and entropj the
constant volume lrnes in th s case between
abscissa On this chart Fig 123
o = l 5433 and v — 6 816 (There are not
enough constant volume lines on this figure are plotted a senes of constant
tor accurate interpolation ) pressure lines a series of constant
qualitj and superheat lines and
a senes of constant temperature The constant temperature lines of
lines

course coincide with the constant pre sure lines in the wet region (below
the saturated \apor line) but bend t< yard the right awaj from the con
stant pressure lines in the superheat region 1 our steam tables contain a
Moll er chart but one which shows a section similar to that marked off by
the dotted lines of Fig 123
The Mollier chart is most useful in connection with steadj flow processes
The constant qualitj lines are labeled according to their per cent of moisture
y — 1 — x The constant pressure hues in Fig 123 are straight in the wet
region the break shown being the result of a change of scale for entropy
See the examples in the caption to Fig 123 for the method of using the chart
Enthalpy

Enthalpy-Entropy ( Mollier ) Diagram for Steam. Example. Steam in the


Fig. 123.
main is at p a = 100
psia. The sample in the calorimeter is at 14.7 psia and 240°F.
What is the moisture content t/ 0 of the steam in the main?
Solution. From the conditions in the calorimeter, locate point b at the intersection of
the curves pi. = 14.7 psia and = 240°F. Since the throttling process into the calorim-
It,

eter is a constant enthalpy process, h n = fe, move along a horizontal line until the pres-
sure line pi — 100 is reached, and locate point a. Read the answer t/„ = 2.6%. If the
initial pressure is too high, pi > p n ~ 1800 psia, expansion does not reach superheat at
14.7 psia and a throttling calorimeter could not be used.
Example. Steam enters a turbine at a pressure of 100 psia and 400°F, point 1, and
then expands isentropically to a pressure of 6 psia. What work is done if A K = 0?
Solution. Using the properties of the entering steam, locate point 1 at the intersection
of the curves pi = 100 psia and h = 400°F. Now follow a constant entropy (vertical)
line until the 6-psia line is reached, and locate point 2. Move to the left ordinate from
point 1 and read hi = 1227 Btu per lb. ; move to the left ordinate from point 2 and read
hi = 1011 Btu per lb. Then equation (g), § 176, gives W= 1227 —
1011 = 216 Btu/lb.
Also, we read the moisture at 2; y 2 = 11.9%.
219
220 PROCESSES OF VAPORS [Ch 11

185 Heat Exchanger Process. The typical heat exchanger is a steady


flow device, an open system, involving two substances, one rejecting heat
and the other receiving the heat which is being rejected The system may
be either one or both substances No work as shaft work is done in these
devices and usually the change in kinetic energy of either substance is quite
small (&K « 0) and A P = 0 For these conditions, the energy equation
(9), repeated at the beginning of this chapter, applied to one substance,
y letds
(w) Q *= Aft — ht — hi Btu/lb

which is seen to be the same as for a constant pressure process, equation (b)
For a steady flow process as defined, equation (w) applies for either constant
pressure (a reversible flow) or for the case where the pressure decreases

in the downstream direction (as it actually does an irreversible flow)
Descriptions of several heat exchangers are given in the next chapter,
boilers, air heaters, economizers, feedwater heaters, condensers, etc

186. Output of Boilers. Big hydroelectric power plants notwithstanding,


the great preponderance of electric energy used in this country is developed
in steam power plants Therefore, we may appropriately give a few
moments attention to boilers which produce the steam for this power The
output of boilers is often expressed as the number of pounds of steam deliv-
ered per hour However, since steam at different boiler pressures and
temperatures contains different amounts of energy, the number of pounds of
steam does not constitute an absolute unit It is evident that an absolute
unit of capacity must measure the amount of heat transferred to the steam
The ASME recommends a unit of 1000 Btu per hr called a kilo Btu (kB)
,

per hour, or a unit of 1,000,000 Btu per hr called a mega Btu (mB) per hour
Let hi be the enthalpy of the water as it reaches the boiler (or economizer if
one is used), let h t be the enthalpy of the steam delivered by the boiler (or
superheater if one is used), and let uv be the pounds of steam delivered by
the boiler (or superheater) per hour Then, the total transferred heat in
Btu per hour is w^ht — hi), equation (w), and the output of the boiler is

(x) Boiler output = ^


“^lOOO
For accurate results, one should consider the enthalpy of the entering water

h\ as being that of compressed liquid (§ 168) The pressure on the water is


taken as boiler pressure, yet the temperature is below saturation tempera-
ture If the weight of steam uj. is the maximum amount that the boiler can
produce, equation (x) gives the maximum capacity
A unit of boiler output or boiler capacity that has been commonly used
for years is the boileT horsepower. Although, at one time, one boiler horse-
power produced roughly one horsepower m
the prime mov er, today, because
of the greater efficiency of modern prime movers, there is no relation
betw een
§ 187] CLOSURE 221

a boiler horsepower and the horsepower of an engine. Since this is true,


the name boiler horsepower is misleading, it causes much confusion in the
mind of the novice, and its use is one boiler
discouraged. Specifically,
horsepower is the evaporation of 34.5 hour from and at
lb. of water per
212°F. By from, and at is meant from saturated water at 212°F into satu-
rated steam at 212°F. For a saturation temperature of 212°F, the pressure
is 14.696 psia. The amount of heat required to evaporate one pound of
water from and at 212°F is hf „ = 970.3 Btu, p. 196; hence, the amount of
heat for 34.5 lb. of water is (34.5) (970.3) = 33,475.3 Btu. Consequently,

One boiler horsepower = 33,475.3 Btu/hr.,

of heat transferred to water and steam. We see then that the output or
capacity of a boiler may be expressed as
Wf, 0 (/i 2 — hi)
(y) Boiler horsepower (bo. hp) ’
33,480

where wto ,
h 2 and hi are as defined for equation (x). Although the unit
,

boiler horsepower is frowned upon, its use is so widespread that the engineer

must know its meaning.

187. Closure. The work of this chapter and the preceding one lays the
foundation for the study of vapor engines and vapor
cycles. In solving
problems, the student should aim to achieve facility in the use of the vapor
tables and of the charts. On this account, it is advisable to find both vapor-
table and chart solutions of problems whenever both methods apply.
Since we have met with a number of adiabatic processes, it may be useful
to summarize certain features in tabulated form. The following relations
are true for either a perfect gas or a vapor:

reVersible
Nonflow( { W= -Au,
(Q - °) {i rrevers ible
>
[ ^ ^—Au,

12

VAPOR CYCLES AND ENGINES

188. Introduction The Carnot


cycle, which consists of two isothermals
and two on any substance, hence, the limits it sets are
lsen tropics, operates
as appropriate to a \ apor cycle as to any, and all the conclusions based on
the Carnot cycle are valid Nevertheless, in practice, it is advisable to use
as a standard of comparison a cycle which is less perfect, but which accords
more closely with the actual e\ ents of a particular cycle, just as we used the
Otto cycle as an approximation of the events of a gasoline engine and as we
used the Brayton cycle as the ideal prototype of the gas turbine

189. Ideal Cycle, Ideal Engine In one important respect, the cycles for
vapors will differ from those for the internal combustion engines, namely
allthe ideal processes of the cycles for internal combustion engines were
conceived of as taking place vnthm the engine, whereas in the vapor cycles,
the working substance is passed successively through distinct and separate
devices, such as a boiler, an engine, a condenser, a pump Because of this
difference, it is often advisable to consider separately the deal cycle and the
i

ideal engine .* The ideal cycle includes all the processes which occur
those in the boiler, the engine, the condenser etc The ideal engine is

concerned only with the processes associated directly w ith the engine, those
of delivering the substance to the engine, of expansion in the engine, of
exhaust from the engine, including any transfer of heat w hile the substance
ism the engine The practical reason for this distinction lies in the desire to
charge losses to the proper pieces of equipment Tor example, a boiler may
generate steam at 400 psia pressure and 700°F, but this steam, due to
friction of flow and radiation from the steam line, may arrive at the throttle
of the prime mover at a pressure of 370 psia and 060°F In this instance,
* The word engine, unless otherwise defined by its context, should be understood to
include all appropriate types of prime movers
222
§ 190 ] RANKINE CYCLE AND ENGINE 223

there is a distinct loss between boiler and engine. The manufacturer of the
engine would justly object to charging the engine with the higher-pressure,
higher-temperature steam of the boiler, since the engine never receives such
steam. Thus, in this illustration, the ideal engine would have one efficiency
based upon the condition of the steam as delivered to it, and the ideal cycle
would have another efficiency based upon the condition of the steam as
delivered by the steam generator. The other
principal difference between the
ideal engine and the ideal cycle of a steam-power plant is that the output
(work) of the ideal cycle is charged with the work of pumping the feedwater
into the boiler, because this expenditure of energy necessarily occurs in the
performance of the cycle, but the ideal (or actual) engine is considered
independently of the pump.

Rankin e Cycle and Engine. The Rankine cycle, or complete expan-


190.
sion cycle,was the first accepted standard of comparison for steam power
plants and is often so used today. (See Fig. 40, p. 91.) Figure 124 shows

a b v in ns
(a) (b)

Fig. 124. Rankine Cycle. The volumes of the liquid in (a) and the temperature rise
ZB in (b) are greatly exaggerated.

the processes on the pv and Ts planes. The steam leaves the boiler in the
state 1, moves to the engine -without loss, expands isentropically
in the ideal
engine to point 2, and passes to the condenser (or to the atmosphere).
Circulating water condenses the steam to a saturated liquid at 3, from which
stateit is isentropically pumped into the boiler in a subcooled liquid state B.

Note the irreversible process of mixing cold water at temperature Ib with the
hot water in the boiler at temperature U — h. The compressed liquid at B
is heated until it becomes saturated at 4, after which it is evaporated to
steam at 1, and the cycle begins again.
If the steam is superheated before

it leaves the steam generator, the corresponding Rankine cycle would be


e-/-3-B-4-e.
,

224 VAPOR CYCLES AND ENGINES [Ch 12

The Rankine* cycle may be analyzed either by considering the cycle


as a whole or by studying each of the individual pieces of equipment,
as
show n in Fig 125 The cycle as shown in Fig 124 has heat added at con-
stant pressure B-4-1 (or steady flow, AK = 0, § 185),

(a) Qa = hi - h B Btu/lb

represented by the area m-B-4-l-2-n on the Ts plane. Verify equation (a)


from Fig 125, where in the ideal cycle h a = hi, that is, there is neither friction
of flow nor heat loss Q m
any pipe line of the ideal cycle The state point is

(a) (b)

Fig 124 Repeated

moving leftward only from 2 to 3, Fig 124 (b), so the rejected heat as a
positive number (constant pressure or steady flow with AK = 0) is

(b) Qr = h* — h 3 Btu/lb ,

represented by the area n-2-3-m, Fig 124 (6) Since 1F,„ = Qa — Qr, the
net work of the cycle is

(c) T7„ e , — hi — kB — (fit — h t) => hi — h2 + h3 — hB Btu/lb ,

represented areas 1-2-3-B-4, Fig 124


by the enclosed The value of
hB =h 3 + W,
equation (e) below] is the enthalpy of the liquid after an
[see
lsentropic compression from a saturated (or near saturated) state 3; for
water, it can be found from Keenan and Keyes {71) (see their Fig 3) In
* William John M
Rankine (1820-1872), a professor at Glasgow University, was con-
temporary with those giants of thermodynamics Joule Maxwell, Kelvin, Clausius It
was during the lifetime of and largely due to the efforts of these five men that the laws
and the science of thermodynamics were formulated and interpreted Rankine had no
small part in these developments He was a man of versatile genius and a prolific con
tnbutor to the engineering and scientific literature of his day Among his published
writings are included an important textbook on mechanics, a civil engineers' manual,
and works on water supply, the steam engine shipbuilding, and many other subjects of
a technical nature, not to mention Songs and Fables for he was also —
a composer of music
and a vocalist of good talent Rankine s modification of Gordon’s column formula is
in wide use today
§ 190] RANKINE CYCLE AND ENGINE 225

the absence of Keenan and Keyes Steam Tables, about the only thing one
can do is to estimate hs — hi from equation (17) for T ds = 0 and v = C, as
in § 168:

(d) hB - h3 =
J -f
V
= ^
V
J [appkoxijute]

as a positive number, where ps is the pressure on the discharge side of the


pump, and in the general form, vf is the volume of saturated liquid and p, is

the saturation pressure at the temperature of the liquid entering the pump.
Considering the energy diagram of the pump as a system in Fig. 125,
we see that if the heat Qp is negligible, as it usually is and as it is in the ideal
case, the ideal work of the pump is

(e) TFp = ( hB — h 3), Btu/lb.,

represented (in foot-pound) by the area 3 -B-c-d, Fig. 124(a); and equation
(d) gives an approximate value of the ideal W p . The right-hand side of

Energy Diagrams for Steam Power Plant. The radiation losses Q, Q, and
Fig. 125.
Qp There would be some heat loss from the water line cd, but it
are usually small.
would be quite small. The line mn is not really a line at all because the condenser is
attached directly below the turbine. In this course, ignore Q’s which are not specified
or implied.

equation (d) is the rectangular area whose width is Vz and whose height is dc
in Fig. 124(a). The actual pump work TFp can be obtained if the efficiency
rj
P is known:

(f) TFp = (/lB


- h ^’ Btu/lb.,
Vp

where the subscript s is a reminder that, strictly, the change of enthalpy is

for constant entropy. At high pressures and temperatures where water is

significantly compressible, equation (d) maybe seriously in error as a


percentage of the true value of h B — hi, but the error as a percentage of the
work Rankine cycle will be small.
of the
Since other ideal vapor cycles contain the equivalent of one or more
Rankine cycles, it will be advantageous to put equation (c) into a form that
will shorten future analyses. First we note from the energy diagram of the
226 VAPOR CYCLES AND ENGINES [Oh IS

turbine (which in the ideal case of an isentropic expansion is the Rankine


engine (Fig 125), or from § 176, that the work is

(65) IF = —Ah = hi — hi Btu/lb ,

[ra.nki.ne engine A K — 0 As = 0]

represented by the area c-l-2-d, Tig 124(a), and to a more natural scale by
area 6-1-2-a, Fig 126 When equation (65) is applied to the engine, point 1

Fig 126 Work of Rankine Engine la (a), area 12 a-b - p Vi/J (m - u,) - +
piVi/J * hi - hi Btu per lb Observe that this area is that of an indicator card for a
complete-expansion reciprocating engine without clearance Only the expansion process
appears on the Ts plane, as 6een in (b)

is the state at the entrance to the engine and point 2 is the state at the
exhaust pressure after an lsentropic expansion Comparison of equations
(65) and (c) shows that the work of the cycle II „„ is equal to the gross work
IF minus the pump w ork p W
This is also revealed by the energy diagram
of all the HiO in the cycle (Tig 40, p 91) Thus, we observe that the work
of a Rankine engine (A K= 0, As = 0) is the initial minus the final enthalpy,
and the work of the Rankine cycle is this decrease of enthalpy minus the
pump work
(66) TF„« = IF — IF„ = hi ht — — TFP Btu/lb
[rankine cycle)

Besides the difference of the pump work another significant difference is as


follows

, /point 1 is steam leaving steam generator


state of
y 18 1 point 2 is on a constant entropy line from 1
/point 1 is state of steam entering engine
^ ine l point 2 is on a constant entropy line from 1

At low pressures, the pump work is negligible as compared with the work of
the cycle and can often be omitted But in modern central station plants
w ith thepressures becoming higher and higher, it should be counted The
decision as to whether or not to include TF„ in equation (66) is a matter of
judgment, a decision which the engineer makes, based on experience For
academic purposes, we may take an arbitrary dividing line Omit pump
§ 192] MEP AND STEAM RATES 227
work for pressures at or below 400 psia; include it for pressures above
400 psia. It is not always advisable to consider AK = 0 for the turbine
(89).

191. It a nk in e Efficiencies. The thermal efficiency of the Rankine cycle


is

_ linnet hi — ho — (Kb — A3)


Qa hi — Kb
where the subscripts refer to Fig. 124. Since the pump work W p = hB — hi
[equation (e)], we may substitute h B = hi + W p into the foregoing equa-
tion and obtain the thermal efficiency of a
Rankine cycle as

(67) e = = hi — h2 — WT
Qa hi — hi

In setting up the expression for the effi-

ciency of steam engines and turbines, we


must decide concerning the quantity of
energy E c that should be charged against
the engine. Custom and the ASME Test
Code require that the engine be credited
with the enthalpy of saturated liquid at the
exhaust pressure. It is argued that the water from the condensation of the
steam with an appreciable enthalpy may be returned to the boiler without
further loss. Moreover, the steam prime mover is not designed to obtain
work from and it would therefore be unfair to charge the enthalpy of
liquids,
the liquid against the engine. This practice of crediting the engine
with the enthalpy of saturated liquid at the exhaust condition is followed
whether the engine exhausts to a condenser, whence the condensate is
returned to the boiler, or whether the engine exhausts to the atmosphere with
the consequent actual loss of the enthalpy of the liquid. Therefore, the
thermal efficiency of the ideal Rankine engine is

W hi — hi
( 68 ) e ~ K ~ hi - hr ’

where the enthalpy of the throttle steam, hi is the enthalpy of the steam
hi is
leaving the engine, and hr, which is the same as hi in Fig. 124, is the enthalpy
of saturated liquid at the exhaust pressure. Comparison of equations (67)
and (68) shows that the efficiencies of the ideal cycle and engine are the same
W
whenever the pump work P is negligible, since h 3 = A/ 2 but only then. —
192. Mep and Steam Rates. The mean effective pressure of the Rankine
engine is the work in foot-pound units divided by the displacement in cubic-
)

228 VAPOR CYCLES AND ENGINES [<?A 12


foot units, that is,

J(k i
- hi)
psf,
»2

where v 2 is taken as the effective displacement per pound of steam of the


Rankine engine w ithout clearance, the volume at state 2, Fig 126
The water rate or steam rate is the number of pounds of steam supplied to
an engine per unit of work The unit of work is usually either the horse-
power-hour or the kilowatt-hour; therefore the steam rate is expressed as
pounds per horsepower-hour or pounds per kilowatt-hour. Let the work IF
be in Btu per pound, and note that there are 2544 Btu per hp-hr Then, the
steam rate w is
2544 (Btu)/(hp hr lb steam
(G9A) W = ’
IF (Btu) /(lb steam) hp-hr
3412 lb steam
(GOB) **
IF kw-hr

As we have learned, the work output of a steam engine may be measured at


any one of three points (Fig 103, p 185) It follows that for the actual
engine, there may be stated three steam rates, one each based upon the
indicated w ork IF/, brake work IF*, or combined w ork 1F K If actual steam
rates are known, we may find actual works from them Thus, let u>, = indi-
cated steam rate in pounds per indicated horsepower-hour, wt, = brake
steam rate m pounds per brake horsepower-hour, and wt = combined steam
rate in pounds per kilowatt-hour output of the generator Then from equa-
tion (69), we can have

TF/ = Btu/lb

3412
,
IFb = ^ Btu/lb ,

Wk =
...
Btu/lb
u>*

Itshould be remarked that the relative performances of two or more prime


movers cannot be determined from the steam rates alone An engine or
turbine operating through a large pressure range may have a relatively small
steam rate, while the engine operating through a small pressure range may
have a relatively large steam rate, yet the latter engine may more nearly
approach perfection, the approach being measured by the engine efficiency
Moreover, the Btu input to the latter engine to produce a unit of work may
be less than the Btu input to the engine operating on the large pressure
range Thus, while the steam rate is a convenient number, we must use
the thermal and engine efficiencies and heat rates for reliable indices of
absolute and comparative performances

193. Efficiencies in Terms of Steam Rates. The heat rate of an engine


or power plant is the input of energy per unit of work For a power plant
§ 193] EFFICIENCIES IN TERMS OF STEAM RATES 229

as a whole, usually measured as the higher heating value of the fuel


it is

consumed Btu per kw-hr. of work (electricity) output.


in
The energy charged against the engine is Ec = hi — hf> Btu per lb.
(§ 191). If the engine has a steam rate of w lb. per hp-hr., the input energy
chargeable against the engine is

(g) Heat rate = wQii — hn ) Btu/hp-hr.,

where the heat rate may


be for the ideal engine (use w), or it may be the
indicated heat rate (use w, and get Btu per ihp-hr.), or the brake heat rate
(use Wb and get Btu per bhp-hr.) If the combined steam rate is Wk lb. per
kw-hr., this value in equation (g) gives the heat rate in Btu per kw-hr.
Since the thermal efficiency of the engine is the work out divided by the
energy chargeable as the input, and since there are 2544 Btu per hp-hr.,

.
_ (Btu of work)/(hp-hr.) _ 2544
.
e ~ ~
. (Btu of chargeable energy)/(hp-hr.) w(hi — hji)’

where hi is the enthalpy of the throttle steam and li/i is the enthalpy of
saturated liquid at the exhaust condition. Now the ideal engine corres-
ponding to any actual engine is one wherein (1) the pressure and quality
(or superheat) at the beginning of isentropic expansion in the ideal engine
(point 1, Fig. 126) is the same as the pressure and quality (or superheat)
at the throttle of the real engine, and (2) the pressure at the end of isentropic
expansion in the ideal engine is the same as the exhaust pressure of the real
engine. It follows that hi and h12 in equation (h) are the same for both
actual and ideal engines. Therefore, equation (h) will yield the ideal ther-
mal efficiency, the indicated thermal efficiency e,-, or the brake thermal
efficiency depending upon whether the steam rate is, respectively, the
e&,

w,-, or the brake wb


ideal w, the indicated Also, if w> is lb. per kw-hr., the . :

combined thermal efficiency is


3412
n 1 Bk =

WkQli h; 2 )

The engine efficiency n not only is used to measure the approach of the
actual engine to the ideal but is also useful in estimating the performance
of a newly designed engine where old data on the efficiencies of similar
engines are available. We recall (§ 156) the definition of engine efficiency
as

. .
_ actualwork of a system
' ' 71
work of the corresponding ideal system

Review § 156. Since equation (69) is true, we may note that

Wi w
— Wb w Wk W
(j)
W ~ w! Vb
Wr ~ wb
~TT7
>
Vt ~ w
280 VAPOR CYCLES AND ENGINES [Ch 12

In each expression, TF and w are values for the proper ideal engine In
these ratios, the steam rates of the numerator and denominator must of
course be in the same units, either both lb per hp hr or both lb per kw hr
The work quantities may each be in Btu per pound of steam, or any energy
unit per unit of time, such as a horsepower
The mechanical efficiency is the ratio of the brake work to the indicated
work, or 17 „ = Wb/Wx — bhp/ihp, etc

194 Effect of End Conditions on the Thermal Efficiency of Rankme


Engme Though there are limitations to this statement, we ha\e found
by experience that when certain changes bring about an increase in the
efficiency of the ideal cycle, analogous changes m the actual cycle generally
increase the actual thermal efficiency This is one reason for the importance
of the ideal cj cles For the Rankine engine, a numerical study of equation
(G8) will show that the following changes increase the Rankine efficiency

1 If the condenser temperature


is loitered (Fig 127) the heat rejected mil be less
the work will be greater and therefore the efficiency mil be increased Nature,
however precisely defines the limit of im
provement that may be obtained by this
means In order to condense the steam,
some cooling medium at a temperature lower
than that of the steam in the condenser must
be available for carrying away the rejected
heat Since we know from the second law
that there will be no net gain if artificial

coohng is reported to we must be satisfied

with a condenser temperature somewhat (say


20°) above that of condensing water from

Fig 127 Effect of Operating


mere lakes etc The best condensers are
Pressures on the Efficiency of the so well designed that we may confidently
Rankine Cycle A lowering of the expect no great thermal improv ement in this
final pressure from 2 to d increases phase of the cycle but efforts at improve-
the Rankine work by the amount of
ment are always being made by someone
the area 2 d k 3, reduces the heat
rejected from 2 3 b c to dkac For It is interesting to note that condensing
the increased boiler pressure at m, plants commonly operate more effieientlv in
it is assumed that the maximum tem- the winter than in the summer because of
perature remains the same, tm — t>
the lower cooling water temperature in the
Subcoolmg effects are ignored
winter A particular plant reports a 10%
greater power production in winter (condenser pressure of 1 in Hg abs ) than in
summer (condenser pressure 2 5 in Hg abs )
2 If the boiler pressure is raised, the evaporation line is raised say from 4-1 to
em Fig 127, and the work area mnke is greater than l-d k-1 True more heat is
transferred to the working substance at the higher pressure but a greater proportion
of the heat supplied is conv erted into work Accordingly we find that the efficiency
of actual plants using higher pressures is greater than that of plants using lower
pressures, which accounts in part for the modem trend toward high pressures
§ 195 INCOMPLETE-EXPANSION ENGINE
] 231

Mooney (10) has computed Rankine efficiencies for pressures up to 3000 psia with
the always on the saturated vapor line (point 1 is saturated steam) and
initial state

finds that the ideal efficiency reaches a maximum at pi 2500 psia, when the
exhaust pressure is about 2 psia. Above about 2500 psia, the efficiency decreases
(when the steam is initially saturated).
3. Although it is not evident from a Ts diagram, the use of superheated steam,
for example, cycle e-/-3-4-e, Fig. 124(b), improves the thermal efficiency as com-
pared with a cycle with the same pressure limits and with saturated steam at the
beginning of expansion. In fact, the improvement in the actual thermal efficiency
is usually greater than that in the corresponding ideal cycle. Small particles of
water in steam moving at high speed past the blades of the turbine, besides having
an eroding effect on the blades, reduce the efficiency of the transformation of energy.
Observe from Fig. 127 that the higher the pressure for a given initial temperature,

the wetter the steam at the exhaust point n versus point d. Consequently, at
higher steam pressures, superheat becomes more desirable in order to reduce the
degree of wetness in the Ion -pressure stages of the turbine. Superheated steam
improves the efficiency of the actual reciprocating steam engine because it reduces
the loss from initial condensation. For these reasons, modern central station power
plants invariably use superheated steam.

195. Incomplete -Expansion Engine. Since complete expansion is seldom


obtained in a reciprocating steam engine, a complete-expansion engine is not
the fairest standard of comparison for it. The events for an incomplete-
expansion engine without clearance are shown in Fig.
128. The engine draws in steam at constant pres-
sure along 6-1; the steam expands isentropically to
some pressure p 2 at 2, the exhaust valve opens,
;

steam flows from the cylinder until the pressure drops


to the exhaust pressure 3, after which the steam is
pushed from the cylinder along 3-a. The enclosed
area l-2-3-a-6 represents the work of the ideal incom-
Fig. 128. Diagram
plete-expansion engine. To get the work equation
for Incomplete-Expan-
into the most satisfactory final form, it will be advisa- sion Engine. This dia-
ble to find the area 1-2-F-6 (which was found to be gram may be thought of

hi — h 2 in Fig. 126 for Q = 0) and add to it the area


as a conventional indi-
cator card for an engine
2-3 -a-F, which is v 2 (p 2 — Thus we have without clearance.

(k) W= hi - 62 +
~ ^ Btu/lb.
Only 1-2 is a thermody-
namic process involving
constant mass.

In finding the thermal efficiency of this ideal engine, we credit the engine
with the enthalpy of the liquid at the exhaust pressure (§ 191), and use
E c = hi — hfz Therefore
m _ hi — h2 + v 2 (p 2 - v*)/J
w 6
hT=T7i '

where hf3 is the enthalpy of saturated liquid at the temperature or pressure at


232 VAPOR CYCLES AND ENGINES [Ch 12

point 3, Fig 128 Steam rates and .engine efficiencies are found as previ-
ously explained Typical values of the indicated engine efficiency for 100-hp
simple engines arc single-valve engine, 56%, Corliss engine, 65%, unaflow
engine, 73% These values tend to increase as the size of the engine
increases, for example, a 1000 hp unaflow engine may be expected to have
an engine efficiency of about 83%
The mean effectne pressure of the ideal incom-
plete expansion engine is p M = W/VD or

Vt
F '

where both work and displacement are values per


pound of steam The volume v 2 Fig 128, is the
,

Fig 128 Repeated. maximum required in the cylinder when one pound
of steam is admitted
To determine the operating conditions of an ideal engine corresponding
to some real engine, let pi (ideal engine) be equal to the throttle pressure of
the real engine, let the quality (or superheat) at 1 in the ideal engine be
equal to the quality (or superheat) at the throttle of the real engine, let p t
(ideal engine) be equal to the release pressure of the real engine, and let
Pi (ideal engine) be equal to the exhaust pressure of the real engine Of
course, the ideal expansion 1-2 is lsentropic

196 Example Steam at p, - 160 psia and h — 400°F is delivered to the


ihrottle of an engine The steam expands to 30 psia, where release occurs Exhaust
s at 15 psia A test of the engine showed a steam consumption of 21 8 Ib/ihp-hr
ind a mechanical efficiency of 85% Find (a) the ideal work and ideal thermal
ifficiency, (b) the ideal steam rate (c) the brake and indicated works, (d) the brake
thermal efficiency, (e) the brake engine efficiency, (f) the mep of the ideal engine
ind the indicated mep See Fig 129

Fig 129 Incomplete Expansion Engine

solution (a)The student should check all values taken from the steam tables

pp 196-199 The properties at 2 are found by using s, = s 2


'

§ 197} ENTHALPY OF EXHAUST STEAM 233


= ’Jt-Vch
1.590S = 1.G993 - 1.3313y :
!/: = S.15%, Xj = 91.85%.
hz = 11G-U - (0.0S 15) (945.3) = 10S7 Btu/lb.
eg = 13.740 - (0.0S15)(13.729) = 12.01 ru. ft./lb.

)r = h, — h z -f (p ; - Pi)
J
= 1217.0 - 10S7 + (30 - 15) (144) = 105.8 Btu-lb.

The ideal thermal efficiency is

IT 105.8
€ — 10 %.
(hi — A/a) (1217.0-181.il)

(b) The ideal steam rate is

2544 2544
15.4 lb./hp-hr.
Ti' 1G5.S

(c) The indicated work is

IF, = = 110.7 Btu/lb.


ic, Z 1 .8

The brake work is

IF B = 1)„W, = (0.85) (1 10.7) = 99.2 Btu/lb.


ir B 99.2
(d) Ci - = 9.59%.
(h, - h /3 ) (1217.G-1S1.1I)
H' B _ 99.2
(e) Vi = 00
11' 105.8

(0
tii
Ideal mep = -
ideal
y —
work = — = (TzgIKuT) =
-^7work
ideal (1G5.S)(77S) _
^
Indicated mep =
j"«»^ork = (Jj _ = 49.9 psi.

197. Enthalpy of Exhaust Steam. The reciprocating engines may be


analyzed as steady flow machines; thus, Fig. 130 may represent any engine,
where IF and IF' are works of the fluid, ideal and actual, and lig and hr are
the enthalpies of the exhaust, ideal and actual. If the process is adiabatic
and A K — 0, energy balances give
1 It' (ideal)
- AA' c 0 II"( actual)
(m) hz = hi IF,
[iDF.U. ENGINE) ^~® 1
h* (ideal)
,
Steady Fio-.v A = '(act*jT3)
[

(n) hr = hi - IF'
= Fig. 130. Enthalpy of Exhaust.
[actual engine. Q = 0, ah' OJ

If there is a lo=s of heat from the engine. Q may be included without difficulty
for the actual engine. Equation (m) applies to any ideal engine and may be
used to locate state c as pictured in Fig. 129 (h ; in (m) = h,].
,

S34 VAPOR CYCLES AND ENGINES [CA 12

In equation (n), W'the work done by the fluid


is With certain test
data, IF' is unknown but may be approximated For reciprocating steam
engines, IF' is approximately the indicated uorl IF/ The small difference
(IF' — IF/) is duo to radiation (winch can be corrected for, if known) and to
mechanical frictional effects, especially between the piston and the cylinder
In the turbine, the mechanical losses are small, so that the w ork of the fluid
IF' is approximately the same as the brake tu>rk IF* The small difference
(U" — IFjj) is largely accounted for in friction at the bearings and packings
Adapting equation (n) to these t\\ o machines, w e hav e

(70) II', = hi- Btu/Ib


(reciprocating steam engine]

(71) h, » h, - IF, - h, - ~ Btu/lb


[turbine]

From another point of view, and especially in the design procedure, we may
use a turbine efficiency iji, which is defined as the actual work of the fluid
IF' divided by the corresponding Bankine work IF In general, between
any two states 1 and 2 on an iscntropic line, wo hare

wrherc state 2' is the actual state after the expansion Fig 120(b) From
(72A),

(72B) h2 = hr - n«(hi - h,) Btu/lb

The turbine efficiency an engine efficiency, and is sometimes so called,


is

but it is not precisely either an indicated or brake engine efficiency If


the exhaust steam is superheated, the temperature and pressure readings
define the state 2' If the exhaust steam is wet, then the value of h 2 must
be estimated by computation In well-insulated machines, the radiation
loss Q should not exceed 1 of hi %
198 Example. For the engine defined in the example of § 190, determine the
enthalpy of the exhaust from the ideal engine, and the enthalpy and quality of the
exhaust from the actual engine Also find the lost work of the actual engine (§ 178)
and the increase of unavailable energy for a sink temperature of 2% = 550°R
solution. From § 196, we take the following data hi — 1217 G, TF = 165 8,
Wi = 116 7 From Table VII, forp, = 15psia h = 1150 8, h/, — 969 7,se — l 7549,
sfl) = 1 4415, f.« = 213 03°F. See Fig 129
'

§ 199} LOSS DUE TO FRICTION OF FLOW 235

Ideal engine, K= hi - W = 1217.6 - 165.8 = 1051.8 Btu/lb.


Actual engine, K- = h, - W, = 1217.6 - 116.7 = 1100.9 Btu/lb.
h e — [h 0
'
y e'h/o) r.

1100.9= 1150.8 — 969.7y or y = 5.15%, c i, e, xc > = 94.85%.


Lost work = ideal work — actual work = 165.8 — 116.7 = 49.1 Btu/lb.

Also, the lost work is h t — h„ Fig. 129. Check this. The increase of unavailable
-

energy during the process 1-e' is T0 As as usual, where As = s e > — s e With the .

moisture contents at e and e ' the values of s may be computed at those states;
,

whence T As may
0 be computed. However, in this example, since point e' is in
the wet region, the areas which represent lost
work and increase of unavailable energy are
each rectangles of width ee'; therefore A Eu and
lost work are proportional to the heights of the
lespective rectangles, that is to the absolute
temperatures. The saturation temperature
at 15 psia is 213.03°F « 673°R. With
T0 = 550°R,

550
( 673
= = 40.1 Btu/lb.
Fig. 129. Repeated.

199. Loss Due to Friction of Flow. Internal friction causes a process to


be internally irreversible, and actual motion of a fluid cannot occur without
such friction. There is therefore an inevitable loss in any pipe line, a loss
which is affected by the magnitude of such variables as the velocity of flow,
density, and viscosity of the fluid. If the fluid is superheated steam, the
states can be defined by pressure and temperature readings. Then, con-
sidering the steam in the pipe line as the system (Fig. 125), we get

(o) h„ = Q+ hi,

when A K — 0. (Note that A P, the change of potential energy, may become


significant when the fluid is a liquid.) Equation (o) may be used to deter-
mine Q when states a and 1 are known. The throttling process is sometimes
used to govern steam engines and turbines, in which event the throttling
may be large as the governing mechanism acts to reduce the opening of a
valve in the steam passageway when the load on the engine is lightened.
Because of its destructive effect on available energy, full-throttling govern-
ing is used only on small prime movers.
The lost work due to the change of state from a to 1, Fig. 131, is taken as
the Rankine work from state a (= h a — hb) minus the Rankine work from
state 1 (= h — i
/to), or

(P) Lost work„_i = ha — h — /ii + h 2 Btu/lb.


, ,

236 VAPOR CYCLES AND ENGINES [C% 12

If Q= 0, as it nearly is in a
well insulated pipe, the process approaches an
adiabatic throttling process, constant enthalpy, represented in Fig 131
by aA In this case,

(q) Lost work. A = A. — k> - hA + hB - hs - Btu/lb

since ha = hA

200 Example Lost Work by Throttling and Heat Loss Leaving a steam gen
erator at p. — 1800 psia and 700°F, steam enters a turbine at p, = 1400 psia and
640°F Fig 131 The condenser pressure pi — pi, — l psia (a) Determine the
lost work for T\ = 561 7°R (b) Determine the lost work if the throttling is
adiabatic, process aA, Tig 131

Fig 131 Losses in Pipe Line The only event occurring w


the engine (turbine)
which is shown on the Ts plane is the expansion process, aA represents the throttling
in the pipe line, Q = 0, ol represents throttling plus a heat loss The effects are very Q
much exaggerated m order to separate the points a 1, and A for clarity

solution (a) From Tables \ II and VIII pp 196-199 we find

pa = 1800 psia p, 1400 psia p2 1 p*ia


700° F t, = 640°F (, = 101 74°F

K = 1260 3 A, 1240 4 s, = 1 97S2


Sa = 1 4044 Si = 1 4079 s c = 1 8456
h. = 1106 0
A/, = 1036 3

To obtain the properties at b Tig 131, we have

sa ~ (s, — S'**/.)*

1 4044 = 1 9782 - 1 8456> or yb - 31 1%


kt = 1106 - (0 311) {1036 3) = 784 Btu/lb

To check into the situation, use A. — 1260 3 and determine the state at A after an
adiabatic throttling In Table % in note that for h A = 1260 3 and p A = W00 psia
the temperature is 660°F Since the gnen temperature is 640°F, some heat has
§ 201) MODERN CYCLES 237

been lost and the situation in this problem corresponds to a-1 and 1-2, Fig. 131.
Thus, at state 2, we have

Si = s2 = 1.4079 = 1.9782 - 1.8456!/ 2 , or y2 = 30.9%.


h2 = 1106 - (0.309) (1036.3) = 786 Btu/lb.

The Rankine works are

W ab = ha -h= 1260.3 - = 476.3 Btu/lb.


784
TFi- 2 = hi - h2 = 1240.4 - = 454.4 Btu/lb.
786
Lost work „_j = 476.3 - 454.4 = 21.9 Btu/lb.

The lost work for an adiabatic throttling process aA, Fig. 131, is h B — h.
(b)
A quick check in the steam tables establishes the fact that state B is in the wet
region so that h B — h
is represented by the rectangular area mbBn. Therefore

hB -h= T b (s B - sb ) = Tb (s A - sa) = 561.7(1.4258 - 1.4044) = 12 Btu/lb.,

the lost work for adiabatic throttling. The data in this example were chosen for
convenience in the use of Table VIII and are not representative of ordinary operating
conditions.

The Rankine and incomplete-expansion cycles are no longer important


standards of comparison for large central station power plants. They are
important, however, for many small plants in which reheating and regen-
erative feedwater heating are not used. Moreover, the Rankine cycle is

basic in that all modern cycles may be considered as being made up of two
or more Rankine cycles. In the various ideal cycles, all processes are con-
sidered as internally reversible and/also as adiabatic except where there is

an intentional transfer of heat, as in the boiler and condenser.

201. Modern Cycles. Some designers consider the maximum permissible


moisture while the steam is in the turbine blades as 12%, because the water
particles erode the blades, resulting in expensive maintenance. At a par-
ticular upper temperature, the moisture in the steam after an isentropic
process (or other adiabatic work process) increases as the initial pressure
increases (Fig. 133). The highest temperature which can be used is limited,
because metals become weaker as the temperature increases above a certain
value (about 600-700°F for ordinary steel). High temperatures of 1000°F
or somewhat more have been common, practice (in large installations) for
several years. Some
designs are currently for 1150°F (5); where the extra
cost is justified, temperatures as high as are being used (7). Given MOOT
a particular temperature ceiling, a point is reached where the steam in the
low-pressure stages becomes too wet. To avoid this condition, we may
resuperheat the steam after it has partly expanded through the turbine, a
238 VAPOR CYCLES AND ENGINES [Ck 12

step which also improves the thermal efficiency (§ 194) The correspond
ing cycle is called the reheal cycle
The following cycles represent modern modifications of the simple
Rankine cycle for the purposes of improving the efficiency and avoiding
excessive moisture in the low pressure regions Notice that m
each case
the relation between the engine and cycle works is cycle work plus pump
work is equal to engine work and that the energy chargeable against the
engine to obtain its thermal efficiency is the heat added m
the corresponding
ideal cycle plus the pump work of the cycle (§ 191) Sec reference 169 for
more detail
Many engineering problems on steam are satisfactorily solved with the
use of the Molher diagram, which is generally a great timesaver The
student should solve all problems suitable to the diagram by both steam
tables and diagram in order to develop facility in the use of each

202 The Ideal Reheat Cycle Figure 132 is a diagrammatic layout


showing the principal features of the reheat cycle and Fig 133 shows the
cycle on the Ts plane Compare these illustrations carefully, and before
proceeding with the study note the locations of points 1, 2, 3, 4, 5 and B,
which have the same meanings on the two diagrams
Heating occurs in the boiler and superheater along B G 1 Fig 133
After the lsentropic expansion 1 2 the steam is led from the turbine and is
reheated 2 to 3 After another lsentropic expansion 3-4 the steam is
condensed 4 5 At 5 the saturated liquid is pumped into the boiler as a
compressed liquid condition B As we learned m § 190 the work of a
Rankine engine minus the total pump work is the net work of the cycle
See Fig 132 We shall use this approach to the cycle analyses of this

Diagrammatic Layout for Reheating Fig 133 Reheat Cycle Mois


Fig 132 —
Plant The system is all the HjO , W.»( — W, tuie at 4 is less than at a, pump
- “SO work 6 B considerably magnified
Q« Q*»ti
§ 203] REGENERATION 239

chapter, although of course TFnet = ZQ. The gross work, as indicated in


Fig. 133 for the equivalent of two Rankine engines 1-2 and 3-4, is

(r) W = hi — h» +hi — h 4 Btu/lb.,


[engine]

the gross work for the ideal cycle as shown on Fig. 133. The ideal cycle

net work is

(73) IFcet = h\ — hi + hz — hi — TF P Btu/lb.,


[cycle]

where (Fig. 132) TF P = T7 pl + TF p2 = hD - h, the total pump work. The


heat added (Fig. 133) is

Qa = hi — hB + h, —h 2

(s) Qa = hi — h$ hz — hi — 1FP Btu/lb.,

of the ideal cycle is


where h B = hs + TF P . Thus, the thermal efficiency

1F»« _ hi — hj + hz — hi — TF P
^ ^ e
Qa hi — hs -{- hz — hi — 1F P

In the absence of Keenan and Keyes Steam Tables, the pump work of the

ideal cycle may be estimated (see § 190) as

'7 -—
" 5(P
(t) TF P = hB — hi ~ Btu/lb.

1300 psig or more, but


Rankine efficiency should be 3% or better for pi
too low In fact, at unreason-
may be less if the reheating pressure is taken
the reheat efficiency is less than le
ably and unlikely low' reheat pressures,
Rankine efficiency (5).

it is true
observed in the Rankine cycle, and
in
203. Regeneration. We
subcooled water » plunged ruto
the reheat cycle just described, that cold
;

the hot water ol the boiler, there to mi* irrevcrably with.the hot bate l m
water could be heated to bo ler
some way could be devised whereby the
interchange of beat w, h,n the system as he m
temperature by a reversible
mivmg would be prevented .and the
this irreversible
Edison cycle « 90), effic.ency as high as that
Cam
resulting ideal cycle may have an
w ater may be accomplished is shown
"/'plan by which this heating of the
240 VAPOR CYCLES AND ENGINES [Ck 12

diagrammatically Fig 134, m m


which the turbine rotor is surrounded by a
hollow casing Let the steam at the beginning of expansion be saturated
(point 1, Fig 135) The feedwater from the condenser is to be pumped
through the hollow part of the turbine casing (Fig 134), back to the boiler

Fig 134 Regenerative Cycle Diagram Fig 13S Perfect Regener


mafic The working substance is the system alive Heating Pump work
effects ignored Smce curve
1 2 parallel to 4 3, the
is
changes of entropy ob and mn
are eipial

The condensate enters the hollow part in condition 3, Fig 135 On the
inside of the casing is the steam just about to be exhausted in condition 2
Now ,
mo\ cs along the casing toward the high-
as the condensate or feedwater
pressure end of the turbine,it picks up heat from the steam on the other side

through a perfectly conducting partition Considering the history of a drop


of water we may imagine this drop picking up heat at such a rate that it is
always at the temperature of the steam opposite it on the other side of the
casing This is the re\ crsible manner of transferring heat Thus one
pound of water will be heated gradually along 3-4 and at the same time one
pound of the steam in the turbine will lose gradually a like amount of heat
during expansion 1 2 Such heating is called regenerative heating The
water now enters the boiler in a saturated condition 4 Smce the heat
rejected during 1-2 (area 1 b~a 2) is equal to the heat supplied during 3-1

(area m
3-4 n), and since this transfer of heat is simply an interchange
within the sy stem, these heat quantities do not enter into the expressions for
work and efficiency Thus, the net work is 2 Q, where the heat quantities
are those transferred to or from an external source W e may write
ITt = XQ = r,(5i - Si) - Ti(Si - ss).
Qa - Ti(s - s t t)

Since Sj — St is equal to Si — (the curves 1-2 and 4-3 are parallel), the

efficiency is the same as for the Carnot cycle,

- r, - ay
c
i i
§ 204} REGENERATIVE CYCLE 241

The lesson from the Carnot cycle, Chapter8, is that to obtain the best
heat must be added at the highest temperature. If all the heat
efficiency,
cannot be added at the highest temperature, any step that increases the
average temperature at which the system receives heat will result in improve-
ment. That is what actual regenerative feedwater heating does increases —
the average temperature at which heat is received from an external source.

204. Regenerative Cycle. Even if the foregoing regenerative cycle could


be practically approximated, it is unlikely that it would be used because of

the low quality of the steam during the latter stages of the expansion near
point 2, Fig. 135. The same effect of regeneration could be obtained by
bleeding or extracting small quantities of steam at various points during the
expansion, utilizing as fully as possible the energy of the bled steam rather
than absorbing the required energy from all the steam. In this way, the
major portion of the steam, that part which continues to expand and to do
work, is not subjected to excessive condensation. Figure 136 shows dia-
grammatically a regenerative cycle with three open heaters (O.H.). Figure

Fig. 137. Regenerative Cycle.


The pump works (6-B e, etc.) are
considerably magnified. Mass of
water heated: from G to 7 is
Fig. 136. Diagrammatic Layout for Regener- (1 — m, — m» — m 3) lb.; from 7
ative Heating. to 8 is (1 — mi — m 2) lb.; from 8
to 9 is (1 — mi) lb.; from 9 to 10
is 1 lb.

137 is a representation on the Ts plane of this same cycle. To simplify the


analysis, consider one pound of steam, which enters the turbine in state 1,
having arrived there without loss from the boiler. After an expansion to
state 2, mi 1 lb.) is bled for feedwater heating.
lb. (a fractional part of The
remainder of the steam (1 — in 0 lb. continues to expand in the turbine to
state 3, where another quantity of steam m 2 lb. is bled for feedwater heating.
The remainder, (1 — ?Wi —m 2) lb., continues isentropic expansion. At
state 4, vi 3 lb. is bled for heating, so that there is left in the turbine (1 — m 1

— vii — m z ) which expands further and passes to the condenser. This


lb.,

weight of steam is condensed and the condensate is pumped to heater (1),


*4* VAPOR CYCLES AND ENGINES [Ch 12

6 to Be, where it is heated from Be to 7, Fig 137, by mixing with the m, lb


of steam bled for the purpose and condensed from 4 to 7 Since the heater
is assumed to bean open heater, the m } lb of condensed steam mixes with the

(1 — mi — ms — m*) lb of condensate from the condenser, so that the


amount of water pumped, 7 to Bj, to heater (2) is (1 — mi — m t) lb In
heater (2), this water is heated from condition B 7 to condition 8, Fig 137,
by the condensation of m 2 lb from 3 to 8 Leaving heater (2) is (1 — m t) lb
of water, which is pumped 8 to B» and heated from to 9, Fig 137, by the
condensation of m, lb from 2 to 9 in heater (3) The one pound of water
which now leaves heater (3) is pumped 9 to B» into the boiler.

First wo shall find equations from which the fractional amounts of steam
bled, mi, m2 , m j, may be found In so doing, w e consider only the law of
conservation of energy and we neglect the fact that the water enters the
heater slightly subcooled, that is, we shall consider 6 and Be, 7 and B 7 etc , ,

as coincident points Tor each heater, the heat added to the water is equal
to the heat given up by the condensation of the bled steam Thus wo haxe

Heat to w ater = heat from steam ,

(u) Heater (1), (1 — mi — m* — mj)(Aj — he ) = ms(h t — k 7 ),


(v) Heater (2), (1 — mi — m 2 ){/ig — h 7 ) = m 2 (ft 3 — h»),
(w) Heater (3), (1 — mO(A> — h») — mi(ft 2 — he),

where, by way of explanation, ht — ht is the heat given up by the condensa-


tion of one pound of steam from 2 to 9 and m 3 is the mass of steam condensed,
ht — ha ~ ht — hj3% = ht — he — W
fi is the heat required to
heat one pound
of water from 8 to 9 and (1 — mi) is the mass of water involved Having
found the various specific enthalpies at 2, 3, 4, 5, 6, 7, 8, and 9, we may use

equations (u) and (v) and (w) in this order to find mi, m 2 , and } m
Next observe that this regenerative cj cle may be conceived of as being
made up of a series of Ranhine cycles, 1-2-9-10, 2-3-8-9, 3-4-7-8, and 4-5-6-7
§ 204] REGENERATIVE cycles 243

As we have learned, it is easy to write directly the expression for the gross or
engine work for each of these imaginary cycles, but we should note that the
mass of steam involved in each of the imaginary cycles is not one pound.
Referring to Fig. 137 for the mass involved in each case, we find that the
work is
(x) IF„n = hi — hi + (1 — mi){h 2 — h- + (1 —
,) mi - m 2 )(h 3 — h t)
+ (1 — nii — m* — m 3 )(h — t /i 5) Btu/lb. throttle steam,

which is the work of the regenerative engine with three stages of regenerative
heating. The net work of the cycle becomes
(75) TF nel = hi — h + (1 — wii)(^2 — h +
2 4} (1— mi — m 2 )(h 3 — h 4 )
+ (1 — nil — mi — mz){hi — h$) — ZTFP Btu/lb. throttle steam.
The heat supplied from an external source to the ideal cycle is Qa — h — lino 1

where h B g — ho + W
P 4, Figs. 136 and 137. Notice that 1F P 4 represents
the pump work of the fourth feedwater pump. Thus we have

Qa = hi — ho — IFpi Btu/lb.

From 6 to 9, there an internal interchange of energy which is not part


is

of Qa, the heat supplied from an external source. The thermal efficiency of
the ideal cycle is the ideal work from equation (75) divided by Qa; thus
we have
rrrm TF [fromeq. (75)]
(/6)
.
6 -
-hi-ho- w;7'
The total pump work STFP may be found from the following if Keenan
and Keyes Steam Tables are available:
(y) STFp = (1 — mi —m — 2 mi)(h B c — hQ + (1 — wii — m 2)(h B 7 — h 7 )

+ (1 — mi) (has — h$) + h B — hg Btu/lb. o

ofsteam reaching the throttle. The individual pump works can be approxi-
mated by a constant volume compression as before. For example, the
pump work from p 9 to pi is
(z) = - « v h ~
TFpi hB9 ho
^ Btu/lb.

Moreover, a satisfactory approximation for the total pump work STFp, since
the total of this energy is not large, is that for one pound of saturated liquid
pumped at constant volume from to p thus,

(a) STFp = p.PA Btu/lb.

At every extraction point, Fig. 137, there should be a break in the expan-
sion line and the saturated-vapor line (since it is plotted for one pound)
to account for the departing steam. However, these breaks are not shown,
2 U VAPOR CYCLES AND ENGINES [CA 12

because the} generally confuse rather than elucidate The simplest


plan then is to sketch the diagram as though one pound of steam were
involved throughout, and indicate upon this diagram the weights imohed
between the various points as in Fig 137 Although steam departs at
various points during the expansion the remaining steam expands lsen
tropicill} Consequent!} ,
we find the quaht} at each point on the expan
sion line just as though one pound expanded all the vvaj , the specific entropy
is constant (Fig 137)

#1 = St = St = St = $s

If a point is wet region as point 4, *« - (s„ — yt sfg) t The student


in the
should observe carefull} the method of anal} sis of this regeneratne c}cle
in order that he ma> be able to set up the proper equations for an} number

of stages without further aid


Thec}cle just described would
be termed an ideal regeneratne
C} clewith three stages of feedwater
heating Actual c} cles in use hat e
from one to nine stages The
number of stages used is purel} an
economic matter The use of one
stage results in a decided improve-
ment in the thermal efficienc} (Fig
138) and would justif} a relatnel}
large expenditure for equipment
The increment improvement due
to the addition of a second heater
Courtesy Babcock A Wilcox Co \etc 1 ork is not so great as that due to the
138 Saving in Heal Bate from
first heater with further diminish
Fig
Regenerative Heating These curves are mg returns as additional heaters
for the operating conditions shown They arc added Four stages of feed
indicate that for a particular number of
water heating become profitable for
heaters the saving is a maximum at a cer-
tain temperature rise of the feedwater for pressures 600 psia
of The eco-
example, using 4 heaters give a maximum nomic number of stages increases
saving of about 12 3%
of the corresponding
as the initial pressure and tempera
Rankme cycle when the feedwater increases
340 °F ture increase See Fig 138

205 Ideal Reheat -Regenerative CycleIn order to gain the advantage


of drier steam in the low pressure stages of the turbine and to gain the
significant increase in thermal efficienc} that accompanies regenerative
feedwater heating reheat and regeneration are often used together For
illustration we have chosen a three-stage regenerative c}c!e with reheat

occurring at the first bleedmg point A diagrammatic la} out is shown in


Fig 140 and the Ts diagram Fig 141 mThe various points of the two
figures have corresponding numbers Let one pound of throttle steam be the
Fig. 141. Reheat-Regenerative
Cycle. The pump works (7 -B 7 ,
etc.) are considerably magnified.
Mass water heated: from 7 to 8
of
is (1 — — 2 — m 3) lb.; from
mi m
Fig. 140.Diagrammatic Layout for Reheat- 8 to 9 is (1 — mi — mf) lb.; from
Regenerative Plant. The working substance is 9 to 10 is (1 — mi) lb.; from 10 to
the system. 11 is 1 lb.
ms VAPOR CYCLES AND ENGINES [CA 12

umt of mass At point passes to the open heater (3) and the
2, nti lb
remainder (1 —
mi) lb passes to the reheater Steam from the reheater
returns to the turbine in condition 3, expands to 4, w here 3 lb of steam is m
bled for heater (2) From point 4, the remaining steam (1 t ) lb
— —m
expands to point 5, where m» mi — t — mi) lb
lb is bled Finally, (1 — m
completes the expansion and is condensed to a saturated liquid at 7. The
regenerative heating has already been described For each heater only
enough steam is bled to heat the w ater to the temperature of the steam at
the bleeding point. The heat added to the condensate m the first heater

Fig 140 Repeated Fig 141 Repeated

(1) is (A* A bt ) Btu per lb of condensate, but the approximation A s — A 7 IS


-
probably satisfactory Thus, equating energy quantities, we have as before
(Fig 141),

Heat from steam = heat to water


m,(hi — A s ) = (1 — mi — m 3 — m,)(A g — A 7 ),
m 2 (ht — ha) = (1 — m, — mj)(A» — h»),
mi(A* — Aio) =* (1 — mi)(Aio — A,),

from which values of mi, m


and mj for the ideal cycle are obtained As
2,

before, we imagine this cycle to be composed of a series of Rankme cycles,


1-2-10-11, 3-4-9-10, 4-5-S-9, and o-b-7-3. Fig 141 Since the net work tA
the ideal cycle is equal to the gross work minus the total pump workW
21 V w we get

(77) = li - h, + (1- m )(*, - 1.) + (1 - «i - m,X*i - M


,

+ (1 — mi — m 2 — m )(A — A«) — 2TF„ Btu/lb throttle steam


3 s

The total pump work 2TF* is given by equation (y) of § 204 {but see equation
(a)] for the three-stage heating The heat supplied to the cycle is

(78) Qa= hi- Am + (1 - mi)(A - 3 AO - W pt Btu/lb throttle steam


;

§ 206] CONVERTING MASS INTO ENERGY 247

In equation (78), the term (1 — mi)(h 3 — hi) represents heat supplied in the
reheater. The thermal efficiency of this cycle is evidently

TT [from eg. (77)]


fbi „ =
Qa [from eq. (78)]

If a reader understands this analysis, he will be able to obtain proper equa-


tions for any combination of reheaters and regenerative water heaters.
A large turbine designed for one reheat and five extractions is shown in
Fig. 139.
206. Converting Mass into Energy. We have previously mentioned that,
according to Einstein’s equation E — me 2
the amount of energy equivalent
,

to a mass one pound is 39 X 10 12 Btu, a misleading number because at


of
this time we do not know how to convert and control more than about 0.1 %
of a fissionable mass into energy, much less any other mass. If the process
of using this energy from mass is via thermal processes or heat engines, the
second law introduces certain additional limitations. Moreover, there is
perhaps some law not yet formulated concerning the availability of mass for
conversion into energy. The first observed fission of an atomic nucleus
was in 1939. As a matter of interest, we may inquire briefly into the mech-
anism by which power is obtained from fission of the atoms, and start by a
review of some definitions.
An atom is the smallest unit of a particular chemical element. An element
is a basic kind of matter (such as oxygen and uranium, of which 100 have

been found in nature or have been artificially created). The current


picture of the atom, which is substantiated by imposing experimental evi-
dence, is that it has a nucleus about which electrons (negatively charged
particles) move, much as planets move about a sun. The electrons have a
relatively small mass, so that practically all the mass of an atom is in its
nucleus. The nucleus is composed of protons, which are positively charged
particles, and neutrons, which are neutral particles, both of virtually the
same mass. A normal atom of a particular element has a certain number of
negatively charged electrons, which is the atomic number Z, moving about a
nucleus in which there is the same number of protons (positively charged)
so the atom is neutral. The atomic number identifies the element. The
isotopes of a particular element have the same number of electrons and
protons (the same atomic number), but they differ in mass because they have
different numbers of neutrons. The total number of protons and neutrons in
the nucleus is called the mass number A, which is the integer nearest the
atomic weight. Thus, the atomic weight of the most common isotope of
uranium is 238.07, and its mass number A is therefore 238, which is the total
number of protons and neutrons in its nucleus. Uranium 235, the kind
which fissions, has 235 protons and neutrons in its nucleus, but the same
number of protons (and electrons) as U 238. To identify a particular atom,
the following symbolization is used:
248 VAPOR CYCLES AND ENGINES [Ch 12
X= chemical symbol of substance,
A = mass number (of atom),
I Z= atomic number

You recall that the atomicu eight of oxygen is arbitrarily taken as 16 by


the chemist and since oxygen has three isotopes the \ alue of 16 is the aver
age value for oxjgen as it is normally found Since the physicist must
distinguish between the isotopes, he has assigned the mass of 16 to the
lightest and most abundant (about 99 8%) isotope ( s 0 16 ), which results m
atomic ueighls of the other two isotopes of oxygen as 17 0045 (,0 W) and
18 0049 (aO 1 *), and an average atomic weight somewhat greater than 16
(16 0044) Thus the scale of relative atomic weights as used bv the
nuclear phjsicist is slightly different from the established chemical scale
A few values of atomic numbers Z and mass numbers A are
Z A Z A
Americium (Am) 95 243 Banum (Ba) 56 137
Boron (B) 5 10 Carbon (C) 6 12
Cesium (Cs) 55 133 Hjdrogen (H) 1 1
krypton (Kr) 36 84 Lithium (Li) 3 7
Nitrogen (N) 7 14 Oxjgen (0) 8 16
Plutonium (Pu) 94 239 Rubidium (Rb) 37 85
Sodium (Na) 11 23 Uranium (U) 92 238

By way of illustration we see from this tabulation and equation (c) that the
physicists designation of uranium 238 is »4 U* ,S uranium 235 would be
«U »
!

We obtain energy from a chemical reaction (combustion) as a result of a


change to a new molecular structure The mass equivalent of the energy
released (heat of combustion) is so small as to go undetected Nuclear
energy is obtained (or consumed) when there are changes in the structure of
the atom's nucleus In particular we obtain energy by virtue of the
splitting of an atom of a certain substance into two atoms which are two
different substances called fission fragments (Energy is also obtainable
from certain of the light atoms by a fusion process as in the hydrogen bomb )
There are only three atoms whose nuclei we are now able to split with a
resulting advantageous output of energy uranium 235 the rare isotope
occurring in nature uranium 233, which is an isotope formed when neutrons
are captured by thorium 232 and plutonium 239 which is formed when an
atom of uranium 238 captures a neutron When one of these atoms splits
it changes into two atoms of different elements, and m
order to produce
energy, the stored energy (or mass) of the fission fragments must be less than
the stored energy (or mass) of the original atom Now we are using the
words mass and energy in the same sense In other words, the mass which
"disappears' on fission reappears as energy, in accordance with Einstein s
equation
The unit of energy used in nuclear physics is the electron volt (e\) which
250 VAPOR CYCLES AND ENGINES [Ck 12

165 are very rarely formed, the most likely ratio of the atomic numbers
of
the fragments being approximately 3 to 2 (85)
A schematic picture of how a chain reaction can occur is shown Fig 142m
If the first neutron n, fissions the nucleus Nj, there results the two high-
energy fragments F and say the two neutrons n* and n s Possibly each of
these neutrons fissions other nuclei N 2 and N*, which in this illustration
produces five more neutrons Then each of these fiv e neutrons are a\ ail
able though not so likely, to fission fi\ e more nuclei At anj rate, there is a
possible multiplication effect If on a\ erage more than one of the released
neutrons fissions a nucleus, something will e\ entuallj burn up, or there will

To Other
Nuclei
Perhaps

Fig 142 Chain Reaction Explosive

be an explosion w hen the multiplication rate is fast enough For a steady


chain reaction the neutrons from one split nucleus sooner or later split one
other nucleus which is the condition to be maintained in a reactor for a
power plant in a steady state If less than an a\ erage of one released neu
tron per fission splits another nucleus the reaction rate decreases To
ha\c a sustaining reaction there is also the matter of a entreat mass, which
we shall not go into except to saj that if the mass is too small so man} of
the emitted neutrons will escape before the} find a nucleus to split that the
average of future fissions becomes less than one fission per fission

207 The Reactor. The present plan of obtaining power from atomic
energy is to ha\ e a reactor in which the chain reaction proceeds to provide
means of using the released energ} for heating a working substance, and then
to use the w orking substance in a heat engine There are a number of differ-
ent reactors including the pressurized water t}pe, the boiling water type,
the fast breeder, and the homogeneous t}pe but space does not permit
complete descriptions here The basic components of an} power reactor
include

1 Core A core of fissionable niatenal in an amount greater than the critical


mass is nece«sar}
2 Control rods Control rods are made of a material with a high affimti for
neutrons hen the^e rods are full} inserted the} absorb a large proportion of the
neutrons from fissioning and the rate of fissioning is reduced to a low xalue When
the rods are entirel} withdrawn the rate of fissioning accelerates The control
rods can be automatical!} moxed in and out to maintain the fissioning at the desired
rate Materials used include cadmium, boron and hafnium
§ 207] THE REACTOR o 51

3. Coolant. A coolant is used to pick up and carry away the energy released by
the reaction.
4. Moderator. There may or may not be a moderator, which is any substance
used for the purpose of slowing the neutrons to thermal energies. By thermal
energy is meant energies of the order of those possessed In gaseous molecules
(0.025 ev), in contrast with the relatively very high cneigt (speed) which a neutron
possesses on emission (2 mcv). These very high-speed neutrons are not so likely
to cause fission as the slow neutrons; thus, to use as many neutrons as possible for
we arrange to have them traverse a modeiator (carbon or water, for example)
fission,

where they gradually lose energy, reaching a range of kinetic cncigy called thermal
energy. A reactor which contains a moderator is called a thermal reactor; one which
docs not is called a fast reactor, because it depends upon the high-speed neutrons to
cause fission.
5. Shielding. Shielding is used to protect life against the lethal radiation of
gamma There are, of course, many other items
rays and beta particles (electrons).
involved, such ns pumps, heat exchangers, and instalments, which arc not concerned
with this discussion.

A boiling reactor is shown in Fig. 143, in which the water within the reactor
serves as both a moderator and a coolant. The pi e«sures shown are typical

Fig. 1J3. Boiling Reactor.The fuel is encased in "rods” (long slender cylinders'
in this reactor,but sometimes it is incorporated in a circulating liquid. These rods arc
supported in hexagonal zirconium coolant channels, which form a honeycomb pattern.
The feature of this reactor is that the coolant channels extend above the reactor core.
The resulting chimney effect improves circulation (and heat transfer'. For nbout two-
thirds of the length of the core, the water is subcoolcd (no steam bubbles), the conse-
quence being that the reaction is not impeded by steam bubbles.
252 VAPOR CYCLES AND ENGINES [CA 12

but and pressure and temperature drops are ignored in the


illustrative only,
following discussion The primary steam moves into the steam drums A
and B at 600 psia, thence to the turbine, preferably via a superheating
furnace (fired by conventional fuels), because turbines have a better effi-
ciency when they are operated on superheated steam In the upper
(§ 308)
part of the rcactoi, the water will be saturated at the temperature of
480 21°F (Table VII), corresponding to GOO psia If some of this water
undergoes a throttling process (§ 182) to, say, a pressure of 350 psia, the
result is a mixture of a little steam and much water, A/e o = (A/
+
zA/e)jso
This process is said to flash steam, hence, the name flash lank The satu-
rated steam at 350 psia goes to the turbine, entering it at the stage where
the pressure is 350, expanding to condenser pressure The 350 psia water
at a saturation temperature of 431 72°F is pumped back into the reactor
The reason for the extra paraphernalia of the flash tank, etc , lies in the
fact that steam among the fuel rods reduces the transfer of energy because
a particular volume of steam picks up less energy than the same volume
of water Hence, not so much steam can be generated if any of the space
about the fuel rods is occupied by steam By throttling saturated water,
which the reactor produces efficiently, additional steam, albeit at a lower
pressure, is obtained (The throttling process is destructive of available
energy and perhaps someone can devise a way to avoid it
In practice, the plan is to provide most of the turbine governor adjustment
on the 350-psia steam Then, if there is a demand for more power, the
governor allows more steam to enter the turbine from the flash tank which
reduces the pressure in this tank somewhat, which results in a greater flow
of water into the flash tank to provide the additional steam needed for the
greater load In the meantime, the flow of GOO psia steam remains more or
less constant With a steady flow of the GOO psia steam, additional steam is
not generated in the core, which action, if it happened, would reduce the
output of the reactor just at a time when the load increases (88) Another
mitigating influence is that the flash pump, in maintaining the water level
in the flash tanks, pumps more subcooled water into the reactor, this reduces
the numberof steam bubbles within the core, resulting m
an increase of
output of the reactor There are several other types of reactors
The amount of electricity generated from nuclear reactors is expected to
increase rapidly over the next several years One prediction is that by
1980 about half of the newly installed electric power will be obtained with
nuclear steam generators

208 Closure. There are so many ways in which various ideas may be
combined for the purpose of producing power that there is no alternative but
to understand the basic principles In this connection, a review of the dis
cussion of the second law would be appropriate, notice the example of § 200
The reader may have noticed that the power cycle w ith the nuclear reactor
§ 20S] CLOSURE 258

utilized saturatedsteam and also that superheated steam is advantageous


thermodynamically and practically. Because of the advantages of super-
heated steam, it is reasonable to suppose that land-based nuclear power
plants will use furnace superheaters if it continues to be infeasible to super-

heat with nuclear energy. In other cases, in order to protect the turbine
blades from erosion by the entrained water particles moving at high speed,
the blades are surfaced with a very hard alloy, such as stellite. This
shielding of the blades with hard material is also practiced now for those
blades in the low-pressure stages subjected to wet steam.
The advantages of high temperature have been emphasized. The
highest steam temperature used in commercial power plants at this time is
probably 1150°F. As an example of current ideas, the new Philadelphia
Electric plant generates 275,000 kw, using steam at 5000 psi and, eventually,
1200°F. There are nine stages of regenerative heating (feedwater tem-
perature of 565°F) and two reheats, each to 1050°F.
At least in the large units operating on modern cycles, there is an optimum
initial pressure at which the heat rate is a minimum for a particular size of
unit. This pressure for a 100,000-kw turbo-generator, one reheat, is about
2800 psia (103). The reason for this phenomenon is not theoretical but is
associated with the low initial volume at high pressures. The volume of
flow can be too small for good efficiency. So, as the initial pressure is
increased, a point is reached at which the theoretical gain in efficiency from
higher pressure more than offset by the loss of turbine efficiency from low
is

volume of flow.For a further reduction of heat rate at pressures above the


critical pressure of steam (3206 psia), the units must be quite large in order
to obtain adequate flow volumes. A 200,000-kw unit with two reheats
does not show a minimum heat rate for pressures up to 5500 psia (103).
13

NOZZLES

209. Introduction. The manner in which work originates in a gas or


vapor turbine is first the expansible fluid expands through a nozzle, during
:

which process some of the fluid’s initial energy is converted into kinetic
energy; then the issuing jet of high-velocity fluid passes across turbine blades
which have been designed to change the momentum of the stream. The
consequence of the change of momentum is a force (Newton) which does
work turning the turbine shaft. This chapter covers the flow of expansible
fluids through nozzles, omitting the flow of such fluids through flow meas-
uring devices, such as venturi meters. * Since the approach is from the
point of view of the energy relations, the reader should be informed that
the general subject of fluid mechanics says much more about the phenomena
of flow. Moreover, this discussion does not cover the flow of nonexpansible
fluids (liquids).
Since we shall be dealing with both gases and vapors, this chapter pro-
vides an excellent opportunity to clarify in your mind the differences in the
methods of handling these substances.

210. Flow through Nozzles. A nozzle is a device (open system) used for
the purpose of guiding the expansion of a substance to a state where the
kinetic energy of the substance is relatively large. There are two types,
convergent-divergent nozzles (Fig. 144) and convergent nozzles (Fig. 145).
They may have any cross-sectional shape to suit the application. Some-
times the section is circular. The elements of the surface of the divergent
section (Fig. 144) are generally straight, for convenience in manufacture.
The smallest cross-sectional area of a convergent-divergent nozzle is called
the throat (Fig. 144). The customary manner of proportioning a con-
vergent-divergent nozzle is to determine the throat area, as explained later;
*
See reference (ISO) for a thermodynamic treatment of this kind of flow.
266
256 NOZZLES [Ch IS
to provide a well rounded entrance, to choose a nozzle length L, Fig 144,
such that the flare of the sides of the divergent section is within good limits
in accordance with experience 'With respect to the flare, too large an
angle 6, Fig 144, above about 12-15°, results in excessive turbulence and
consequent irreversibility The nozzle will be excessively long if 0 is too
small, say, less than about 6°
A convergent nozzle with a short section of parallel elements Fig 145(b),
tends to discharge a better formed jet than that in Fig 145(a) Convergent
nozzles as used in turbines are commonly foil nozzles, that is they are
formed by stationary blades which are small foils
For any nozzle, the shaft work is zero II =0 The length of time that a
particular mass of substance is within the nozzle is a small fraction of a

Fig 144 Convergent Divergent Not Fig 145 Convergent Notiles


tie The pressure drop per unit of length
is not constant (See $ 216 )

second, therefore for practical purposes the flow is adiabatic The only
pertinent energies are therefore the internal energy and flow w ork (enthalpy )
and the kinetic energy K, as shown in the energy diagram of the system in
Fig 144 Letting K= vt/fpgj), we have for steady flow, dK = —dh, or

<*>
%j + h '-£? + h ’ B w b '

[ant fluid Q = 0, II =0]

applicable to ideal or actual expansions when state 2 is the ideal or actual


state at the exit section The initial v elocity u, is usually difficult to deter-

mine and sometimes it is negligible (K, « 0), in winch case,

(b) vj - \2gJQii - h t)] in = 223 8 (A, - W' 1


,

[gas OR VAPOR Vi negligible]

applicable toany expansible fluid, w here 2g»J ~ 50 000 and (2g*J) us *» 223 8
These constants are convenient ones to memorize for current use As
necessary, we use the conservation of mass equation (4), § 23,

Aivi AsUi
A

§ 211 ] STAGNATION PROPERTIES 257


from which

(c)
= (iS) ps
f ’

in which the cross-sectional area and v cu. ft. is the specific


.4 sq. ft. is
volume. Substituting the value of Vi from equation (c) into equation (a)
and solving for u», we get
1/2
_r 2 gJVn-h) ] ,
2 fps ’
|_1 - [-W(Aii’ 2)] 2
J
[gas or vapor, ideal or actual flow]

where the states


1 and 2 are at any two sections 1 and 2, wherever taken.
Comparing equations (79) and (b), we see that the factor

cares for the effect of initial velocity and it rnay be kept in mind as a cor-
rection factor applicable to equations obtained later; p lb. per cu. ft. is
the density. For circular channels, it may be convenient to note that
A,/ i = (Dj/DO 2 .

211. Stagnation Properties. Stagnation properties, which will be useful


in this chapter, are those which a moving stream of fluid would have if it were
brought to rest by an isentropic compression. If a moving stream with a
kinetic energy K
Btu per lb. and enthalpy h Btu per lb. is brought to rest
isentropically, the resulting kinetic energy is zero and the enthalpy is h„.
At the entrance to the nozzle (Fig. 144), h„ = hi -f- K\. Thus, we see that
h 0 includes the effect of initial velocity. Refer to § 140, which deals with
stagnation enthalpy relative to a body moving in a fluid at rest.
substance is a flowing vapor for which vapor tables are available,
If the
and the state where the enthalpy and kinetic energy are h and
if is defined K
(say, pressure and temperature in the superheated region), the properties in
the stagnation state are easily found from

ho = h -f- K or h0 — h —K and sa = s,

which locate the state in the tables. If the substance is an ideal gas, the
property relations of an isentropic process apply; repeated for convenience
from § 70,

The ideal gas equation of state applies, p0 v0 = RT


C where T0 is the stagna- ,

tion temperature; p, v, and T in the foregoing equations are values at any


other state for which the entropy is the same and in which the kinetic
258 NOZZLES [Ch 18
energy is K Since Ak = CpAt for an ideal gas, we have

(e> c,(T. -T) - K or T.-T +


2 gjc„
In terms of stagnation enthalpy, the energy equation for a steady flow noz-
zle is obtained from equation (a) as

(80) K= ht +
Ki, or u, = 223 8(k - k ,)»'*,
[gas or vapor, » — C or *s*C()=0J
where h0 = hi + Vi*/ (2g*J) Equation (80) is the same as equation (79)
when the initial velocity is significant and it is the same as equation (b)
when Ui is negligible

212. Mass of Flow through Nozzle. The mass rate of flow is computed
from equation (4),

(4) w— — lb /sec ,

where v fps, obtained from equation (80), and v cu ft per lb are values at a
section whose area is A sq ft This equation may be applied at any section
of the nozzle, but it is most often applied to the throat and exit sections
The method of finding the volume v = v* at any section depends upon
whether the substance is an ideal gas or a vapor For an ideal gas an m
iscntropic expansion,

m "‘“fey "•*(?-;) (fy


where pi and iq may be used in place of pa and i„, because

Pol o* = PlVl
k
= p %»
!*

213 Example Air Nozzle . At the rate of 5 lb /sec air at 200 psia and COT ,

enters a nozzle with negligible velocitjand expands isentropically to 15 psia Deter-


mine the final velocit} specific volume, and the area of the exit section
,

solution The final temperature is

Using Ah = cP A7’, we have fiom equation (80),

u, = 223 8 V(0 24) (520 - 248) = 1810 fps

The specific volumes are

4
= Vl = (0 902) = 6 1 cu ft /lb
G )

§ 216] VARIATION OF VELOCITY 259


From equation (4), we get for w = 5 lb./sec.,

wv 2 _ (S) (6,1)
A 2 = 0.01085 sq. ft., or 2.425 sq. in.
u2 1810

Flow of a Vapor. When a vapor


214. Equilibrium starts its expansion
in a nozzle from the superheat region,
as at 1, Fig. 146, and follows an isen-
tropic path l-a-2, it should start con-
densing at a where the process line
crosses the saturated vapor line, and it
should become progressively wetter as
the expansion proceeds to 2. If this
should be true (which it is not § 221), —
the vapor-liquid mixture would at all
times be an equilibrium mixture and _.
,i - , , , . Fig. 146. Equilibrium Expansion.
the expansion would be an equilibrium
expansion. Hence, for internal equilibrium conditions,

St = S2 — (Sg — UlS/g) 2,
hi = (h„ - yjif„) 2 and v2 = (v a ~ y 2 vf0 ) h
where s may be s0 if stagnation properties are known.
i These equations,
together with equations (4) and (80), give the solution to problems in which
equilibrium conditions are assumed.

Example: Steam Nozzle, Equilibrium Flow. Steam enters a nozzle with


215.
negligible velocity at 100 psia and 400°F, flowing at the rate of 10 lb./sec. If it
expands isentropically to 10 psia, what should be the area of the exit section (process
1-2, Fig. 14G).

solution. From the steam tables,

hi = 1217. =
s, 2 1.7876 hf, 2 = 982.1
Si = 1.5908 =
s/,, 1.5011 IV: = 38.42
a, = .3.008 Ki = 1143.3 V/gz = 38.4

Si = S: = 1.5908 = 1.787G - 1.5041?/:, = 13.05%.


A: = 1143.3 - (0.1305)(9S2.1) = 1015.1 Btu/lb.
Vi 38.42 - (0.1.303) (38.4) = 33.4 cu. ft./Ib.
U2 223.S(/i i
- A:) ,/s = 223.8(1217.0 - 1015.1)'/= = 31S0 fps.
wi>! (10) (33.4) ( 144
A-_ _ = 15.1 sq. in.
u- 31S0

216. Variation of Velocity, Specific Volume, and Area of Section of a


Nozzle. be revealing to investigate the manner in which the velocity,
It will
specific volume, and section area vary as the pressure drops rough ath
nozzle, an investigation which may be made for either a gas or vapor. The
method of procedure for a gas is as shown in detail in § 213, those calcula-
tions being repeated for several different p : ’s. The answers arc tabulated
below Fig. 147, in which the values are plotted.
* 60 NOZZLES [Ch IS
The most notable feature, true for vapor as well as gas nozzles, is that the
area decreases to a minimum., already named the throat and then increases,
that is, the nozzle must be converging diverging Since w is the same for all
sections, the A/w curve would look like the A curve, Fig
147, then since

Fig 147 Specific Volume Area and v/v Plotted Against Pressure Vapors and
other gases flowing through nozzles have attributes quite similar to these curves plotted
for air Observe that the velocity may be very large the maximum in this case being
1810 fps — 1259 mph It is apparent that the nozzle for this expansion must have a
convergent portion followed by a divergent portion This illustration is plotted for a
uniform pressure drop (abscissa) However, the pressure drop in the nozzle is not
constant per unit length of nozzle There is no reason for it to be

Pi * V, A,
Wi>i u>/i4j*
Psia Cu ft /lb Fps Sq m
200 0 962
176 1069 483 1 678 466
150 1 18 700 1212 693
125 1346 887 1 091 669
106 1 SIS 1020 1 070 673
100 1 68 1060 1072 671
50 2 69 1431 1 302 563
25 4 24 1675 1 821 395
15 6 1 1810 2 425 297

Mi Diq It

A/w — v/v the w/ A — v/v


curve naturally moves to a maximum and then
dawn The reason that these curves move as they do is that the early m
part of the expansion the velocity increases at a greater rate than the
specific volume but later the specific volume increases at the more rapid
rate Study the curves
; |

§ 217] CRITICAL PRESSURE IN A NOZZLE 261

Note that it is necessary for the area to increase after the throat section if
the expansion (volume increase) is to continue. In the tabulation of Fig.
147, the minimum area is 1.07 sq. in. at p 2 = 106 psia. If this nozzle, cut
off at the throat section, should discharge into a passage of I.07-sq. in.
area, Fig. 148, there would be no further expansion, and in the absence of
frictionand heat the air would continue to move in this passage at a pressure
of 106 psia and v = 1020 fps for an indefinite distance. If the pressure p 2
at the end of the pipe B, Fig. 148, is less than 106 psia, there is a sudden,
uncontrolled expansion and the air leaves the pipe in a chaotic condition,
the degree of chaos increasing as the pressure drop increases. If the pressure
p 2 were 100 psia, that is, any pressure a little lower than 106, much of the
acquired kinetic energy could be con-
verted into work in turbine blades.
On the other hand, if the pressure p 2
were 15 psia, the confusion in the
stream would be so great that much of
the acquired kinetic energy would be
lost in internal friction before it could
do work on turbine blades. When the
Fig. 148. The properties of the fluid
expansion is guided, as suggested by do not change from, say, C to D.
the dotted lines at B, not only is the
kinetic energy acquired at C conserved, but additional kinetic energy is also
generated during the further expansion, all of which is available for work on
turbine blades.
Now suppose that the pressure p 2 is greater than 106 psia, say, 175 psia.
Then in the absence of friction, this pressure would build up in the pipe line
theoretically to the section C that is, p, would be 175 psia. If this should be
true, the velocity acquired at C would
be 483 fps (instead of 1020 fps) as
shown in the tabulation under Fig. 147, but to obtain the same flow w, the
area at C would have to be increased to 1.578 sq. in. See also § 223 and
Fig. 151.

217. Critical Pressure in a Nozzle. Since the pressure at C, Fig. 148,


will not go below p = 106 psia, called the critical pressure, the conditions
t

at the throat deserve further study. Let us find an equation for w in terms
of the properties of an ideal gas and study the result. First, we have the
following transformations:

= 223.8(/i„ - h 2 y r- = 223.8 [c p (T0 - T 2 )]


u-

= 223.8 cP T0 (’-£)]"
i
[ideal gas]

{fc-D/tn

(g) C! = 223.8 {c„r. [l - (g) fps,

[rDEAI, GAS, A = C
NOZZLES [Cft 18
where 223 8 = y/2gt,i Using this equation and t> s = It TJpf) (pa/p 2) 1 *
(
from equation (f) in w = Av/o, we find

-fenr-
(h) , 993 e Isi
r-fenr^
[ideal gas, a = C]

which the mass of ideal gas flowing lsentropically across a nozzle section
is
of area At where the pressure is p t If w is plotted against pt/p„ with all
other terms constant, the curve Oba, Fig 149, is obtained Thus, for a
particular nozzle with At constant, the value of w from equation (h) has a
maximum value at b, a value which occurs when
(*+!)/»

fer-fe)
isa maximum, and the pressure p 2 which makes this term a maximum,
,

can be found by equating its differential with respect to pt to zero, thus,

d[{pt/poY' k ~ (p*/p„y
k+ »'k
]

dpt

1 -
-

(**) (0 C4 )
°'

from which

(rli)
*-1 /k
Since T /T„ -
t (pi/p0 ) ,
equation (81) yields

(.) T‘- T -COVLT]


-{kTi) [.

The pressure at the throat p, is called the critical* pressure and can be pre-
dicted for an ideal gas in an ideal expansion (and quite closely for any
gaseous expansion) by equation (81), the ratio p,/Po is the critical pressure
ratio, T t in equation ( 1) is the temperature at the throat after an lsentropic
vsvtnfscv!, IS wr/t wA’f -iaS-ra Aj bm tha insult^
he concludes that A varies with the pressure ratio function in the brackets
The differential of the bracketed terms equated to zero yields the condition
of minimum area, which is the same as equation (81) In other words, the
section of the critical pressure and the section of minimum area coincide
which is what you had no doubt already concluded This being so the
mass of flow is govern d by the area of the throat (and also the initial state),
The word critical 13 overworked
§ 218 ] VALUES OF CRITICAL PRESSURES 203

the size of the sections of the nozzle subsequent to the throat are irrelevant
to the mass flowing.
As was pointed out in connection with Fig. 148, the pressure will not fall
below p in a nozzle that has no divergent section; therefore, the dotted
t

portion of the curve bO of Fig. 149 is a


physical impossibility; the flow rate does
not fall off as the pressure goes below the
critical p,, but remains constant,
pressure
and is represented by curve abc, Fig. 149.
Summarizing, we may say

1. The flow is less than the maximum attain-


able if the final pressure is greater than the
Fig. 149. Discharge from Noz-
critical pressure p,, and zle. The curve Oba is for a con-
2. The flow is the maximum for all final pres- stant area A - of the exit section as
sures equal to or less than p,. obtained from equation (h).
Curve abc represents the actual
The final state defines the area of the discharge, which reaches a maxi-
mum at the critical pressure ratio
final section. If, in the tabulation of Fig.
pt/p and then remains there along
,

147, the final pressure is 25 psia, the area of be. Curve abc represents equa-
the exit section should be 1.821 sq. in., as tion (h) when A- is allowed to vary
so as always to correspond to p~.
seen in the tabulation; if the final pressure
is50 psia, the exit area should be 1.302 sq. in. In both cases, the dis-
charged mass would be the same.
Because the use of equations (g) and (h) conceals some thermodynamic
fundamentals, the student is advised to avoid them except for special studies,
such as in this article, and except for the purpose of deriving special forms of
equations for repeated applications to particular situations.

218. Values of Critical Pressures. Observing that the curves for A and
V/v, Fig. 147, do not change abruptly at the critical pressure, we correctly
conclude that the throat area is not sensitive to the true value of the critical
pressure; that is, k in equation (81) may vary somewhat from its true value
without significant effect on the nozzle design. For this reason, equation
(81), which strictly applies to an isentropic expansion, is commonly used
for computing critical pressures. The actual expansion is practically
adiabatic, and because a correctly designed and well-made nozzle is highly
efficient, the expansion is only slightly* irreversible. For ideal gases with
k — 1.4 (which means diatomic gases at temperatures not too far from
atmospheric), the critical pressure is

The critical pressure of air for an initial stagnation pressure of 200 psia is
(0 53) (200) = 106 psia (see Fig. 147). For k = 1, the critical pressure ratio
,

*e4 NOZZLES [Ch 18


is0 606, for k = 1 67, pt/p» — 0 487 Note the relatively small variation
Other values in common use are

Superheated steam, A = 13, p, = 0 545p t ,


Supersaturated steam, £ = 13 p, = 0 545 p,
(moderate pressures]
Wet steam k = 1 13, p, = 0 58pi

Keenan and Keyes gives values of k for superheated steam

219 Throat Velocity If equation (g) is applied to the throat section,

P* = Pi, we obtain an expression for the throat velocity Take these steps
1 Substitute Pt/jPo from (81) into equation (h) and simplify,
2 Substitute •\/2gd for 223 8
3 Substitute for Cp = Rk/\J(k — 1)], which is equation (26), §61, and
simplify,
4 Substitute for T„ its value from equation (l), T, = (l + l)Tt/2, and
simplify to find
1

- (S^)
'*
(k) v, = IgARTd'" - fps,

where pi ** l/vt lb per cu ft is the density and p,v, = RT, The velocity
given by equation (k) is the sonic velocity or the acoustic velocity (Mach 1),

which the velocity of sound in a gas whose specific gas constant is R ft-lb
is

per lb-°R, whose c v/cv = k, and whose temperature is T, Perhaps you


recall from your physics that Laplace found that the velocity of sound in a
gas is uj = (g00v) Ui where /9 is the lsentropic bulk modulus and vis the specific
volume of the gas By definition, the bulk modulus is the pressure (stress)
change dp divided by the accompanying volume change per unit volume
(strain) dv/v, or

wheie the negative sign recognizes that the volume decreases as the pressure
increases To get dp/dv for an lsentropic process of an ideal gas differenti-
ate pv* — C and find —dp/ (dv/v) = pk = 8 Then substitute pk for 8 in

v, = (goB »)
1/1 = (?*fcpy) 1/S,

the velocity of sound in a gas This expression is seen to be the same as


the second form in equation (b) It is often convenient to express velocities

in terms of Mach numbers M, which is the ratio of the actual velocity of the
substance (or of a body in the substance) divided by the velocity of sound in
the substance, that is,

(m) M "
». toJtpK)
1 '1
§ 221] SUPERSATURATED FLOW 265

Mach numbers greater than unity are supersonic velocities; Mach numbers
less than unity are subsonic velocities. We note that in a convergent noz-
zle, the velocity of the substance is always subsonic; in a convergent-

divergent nozzle, the velocity is M


< 1 up to the throat, Mach 1 at the
throat, and M > 1 in the divergent part.

220. Example: Throat Size. What should be the throat area for the steam flow
is, p, = ICO psia, ti - 400°F,
defined in the example of
215; that § p 2 = 10 psia,
10 lb./sec., isentropic process?
solution. The throat pressure for superheated steam (from § 218) is

Pi = (0.545)(160) = 87 psia (nearly).

Taking values from the steam tables for 87 psia, we find

s, = s, = 1.5908 = 1.0139 - or
1.153 y,, y, = 1.99%.
h, = 1184.6 - = 1166.8 Btu/lb.
(0.0199) (896.5)
v, = 223.8(1217.6 - 1106.8) ' 2 = 1592 fps.
1

v, = 5.055 - (0.0199) (5.037) = 4.95 cu. ft./lb.


(10)(4.95)(144)
= 4.47 sq. in.,

which is the necessary throat area to pass 10 lb./sec. of steam in an ideal expansion.

221. Supersaturated Flow. As previously mentioned, the time that it

takes for a particular molecule of substance to pass through a nozzle is

(a) Ts Plane (b) hs Plane

Fig. ISO. Ideal Supersaturated Flow.

short (of the order of 0.001 sec.). Therefore it should not be too surprising
that condensation does not start at a, Fig. 150, where the expansion line
1-a-c crosses the saturated vapor line. Instead, the steam remains a vapor
until some point reached when condensation suddenly occurs. Such
c is
steam as that between a and c is called supersaturated steam and the state
of this steam is spoken of as a inctastab k stale, meaning that it would take

more than a very small action to cause the steam to achieve internal equi-
266 NOZZLES [Ch IS
librium states along ac —say, an action as big as a spray of
fine water drops
into the stream ahead of point a (not that this result is desired) The actual
pressure at the supersaturated state e is p 6 Fig 150, but the saturation
,

pressure corresponding to the temperature tc is pj fFig 150(a)] which is the


pressure that would ha\e existed at c if an equilibrium expansion had
occurred to U The ratio of these pressures, p6/pj, is called the degree of
supersaluration or supersaturation ratio
A number of studies haxe been made as to the limit of supersaturation
states, a limit named the IPi/son line (Fig 150) Bmmc
and \\ oods (101)
located the Wilson line through the points »/ = 3 5%, p - 10 psia, and

y = 2G%, p = 5 psia Yellott and Holland locate the Wilson line at

approximately GO Btu below the saturated \apor line on the Molher dia-
gram, that is ac, Fig 150 (b) is equal to about GO Btu Vs you would
expect, this so-called line is actually a region, condensation does not begin
at a precisely predictable point
Supersaturated steam is denser than steam in equilibrium at the same

pressure The of supersaturated steam is less than the xelocity of


i el o city’

steam in equilibrium,both ha\ mg expanded from the same initial condition


Since the density increases by a larger percentage than the \ eloeity decreases
there will be a greater discharge (tc — lu/t — Avp) when the steam is
supersaturated than when it is in equilibrium Because of tins phenomenon
the discharge coefficients are sometimes greater than unity that is, the
actual discharge is sometimes greater than that computed for an ideal
equilibrium flow (§ 235)
The ideal expansion of supersaturated steam is found to be closely defined
by pr* = C, where k — 1 3 in the region of moderate pressures From the
energy equation (17), §42, Tds = dh - i dp/J for a reiersible process
and for ds = 0, we have dh = v dp/J, and the integration of dp can bet

made for a known pi relation This integral w ith pe* = C has been used at
various places in this text equation (38) and below, § 70, equation (t),
§ 73, equation (f), § 113, etc Since dh = -dh from equation (a), we haxe
]

§ 222 EXAMPLE: SUPERSATURATED FLOW 267

dK = —v dp in foot-pound units and

[supersaturated steam, superheated steam, ideal gas]

the velocity of supersaturated steam (or of an ideal gas) after an isen-


tropic expansion when the integration is made from the stagnation state,
K = AK =
2 V2 2/2g„ ft-lb. per lb. If we use p\ and iq in equation (82) in
place of p 0 andv0 then the correction factor I for initial velocity, [equation
,

(d)] should be used when the initial velocity is significant. Equation (82)
applies to any reversible expansion where pv l = C (or pv n — C), which of
course includes the ideal gas. It would be easy to show that equations (82)
and (g) are identical. Equation (82) may also be applied to superheated
steam over a moderate drop in pressure when the corresponding average
value of k is used. Supersaturation is likely to persist to beyond the throat
even though the steam initially has very little or even no superheat.
The procedure in solving problems for an ideal expansion of super-
saturated steam would involve computing

where V\ (or v0) is taken from the steam tables for p\ and h (or p„ and t0 ),
computing U 2 from equation (82), and then using w = Av/v lb. per sec.
The throat pressure in a convergent-divergent nozzle is as given in § 218.
Our knowledge of the flow of steam in converging-diverging nozzles is not
all to be desired, and space is short here. For the purposes of this course,
you may assume that the mixture of water and vapor leaving such a nozzle
is in an equilibrium state (if it is more than GO Btu below the 100% quality

line on the Mollier diagram, § 223).

222. Example: Supersaturated Flow. On the basis of supersatuiation, find the


throat area of a nozzle for the conditions given in §§ 215 and 220; p = ICO psia, t

l\ — 400°F,
p : = 10 psia, 10 Ib./sec., isentropic process, Ui 0. «
solution. From §§ 215 and 220, nc take the data: t>i = 3.00S cu. ft., p, = 87 psia.
Then
v, = v, = 3.00S
(f r 4.81 cu. ft.,

which is observed to be smaller than for the equilibrium flow (4.95 cu. ft. in § 220)
even though the steam is barely across the saturated vapor line. From equation
(82), we get

v ‘=
1(2) (32.2) (ICO
\
X 144)(3.00S)(1.3)
f3^1 Manr
,

NOZZLES [Ch IS
or u{ — 1590 fps For 10 lb per sec

wo, _ (10) (4 81) (144)


,
1
_ = 4 36 sq in
v, 1590

This is a better answer than the one in § 220 (4 47 sq in ) The difference between
the answers would have been greater if the expansion had been farther below the
saturated vapor line As a matter of interest we might note that if this much area
were needed in a small turbine it would be divided among several nozzles

223 Overexpansion and Underexpansion If the exit section of tne


nozzle istoo large over expansion occurs, that is, the substance expands to

some pressure below the pressure in the discharge region and then rises to
the discharge pressure If the exit section is too small under expansion
occurs, that is the pressure at the exit section is greater than that the dis- m
charge region and as a result there is a free and turbulent expansion after
the substance leaves the nozzle Although either overexpansion or under-
cxpansion results in a loss of available energy such operating conditions
cannot be avoided at times because of the necessity for varying the output
the steam turbine is not always producing the same power the turbo jet
engine is not always exerting the same thrust
The curves m
Fig 151 are for a convergent divergent nozzle 10 in long
and show what happens to the pressure in a steam nozzle for various back

After Btnnte and Woods (101)

Fig 151 Effect of Varying Discharge Pressure

pressures The nozzle is designed for a pt — 0 20 p„ (closelj) The throat


pressure is nearly 0 55p„ Notice that vv hen the back pressure is p2 81 p« =0
there is no effect on the throat pressure and therefore no effect on the
amount of flow The same amount of steam is discharged from this nozzle for
pt = 0 81p0 as for p 2 - 0 20p„ (The difference between this case and that
discussed in connection with Fig 148 is that there is a diverging channel
§ 22Jf] LOSSES IN NOZZLES 200

after the throat in this instance, thus permitting further expansion.) The
curve for 0.82 indicates that the back-pressure effect has just reached the
throat at a. For = 0.83p o the flow through the nozzle is definitely less.
,

For all final pressures greater than 0.82p„, the pressure falls to a minimum
near the throat and then rises to the discharge pressure, as for p : = 0.87p„.
The break is where condensation occurs in the
be in the curve, Fig. 151,
supersaturated steam at This phenomenon is seen to be accompanied
b.
by a small pressure jump. Notice that even when the back pressure is
0.47p„ (nearly as much as the throat pressure), the steam expands (to f) in
accordance with the increasing area of the channel through as much as 80%
of the length of the nozzle. Where the back pressure is much greater than
the design exit pressure, there is a shock wave somewhere beyond the throat,
where the velocity is supersonic before the shock and is subsonic afterwards,
as at de, Fig. 151; and as the substance passes through the wave, there is a
sudden pressure rise, as seen in the various curves of Fig. 151.

224. Losses in Nozzles. The nozzle efficiency e„ is defined as the actual


kinetic energy of unit mas3 at discharge divided by the ideal kinetic energy
per unit mass, each measured to the same final pressure. Let 2' be the
actual final state, Figs. 146 and 152; then

Kr _ - 1*7(2gj)
w
, *
e "~_ K ~
2
h.
(ho -
hr-

h t ).
_
(hi - h 2 ), + Ki
where the subscript s is a reminder of constant entropy; Ki = vi t/(2grJ).
In a convergent-divergent nozzle, K\ is nearly always relatively negligible
and can be omitted; this is also true when there is a substantial expansion in

Fig. 146. Repeated. Fig. 152. Gas Expansion. If v, = 0, state


1 is the same as the stagnation state.

a convergent nozzle. If the nozzle fils the expansion, the efficiencies are
high, ranging from some 92% for long convergent-divergent nozzles to
98-99% for convergent nozzles. The efficiency of convergent-divergent
nozzles up to the throat is similarly some 98-99% for well-made ones.
The nozzle coefficient or velocity coefficient tj„ is the ratio of the actual
velocity divided by the ideal velocity,
270 NOZZLES [Ch 18
_ Vt _ [20»/(A. - ht)] x '* r~
(0)
- A*).] 1/*
“ Ve '
[2ff*/(k

Engineers experienced in this phase of design know from


past experience the
approximate nozzle efficiency or nozzle coefficient to expect under certain
circumstances, and this knowledge can be used for the purpose of computing
hz from equation (n) or (o) Notice that the actual \ elocity is now given by

(P) w* = [e.(2gJ){K - hi)]"* - u„[2 gJ(h„ - h t) ]>'* fps

Similarly, any equation giving v* may be multiplied by ij„ = yfc to get vi


Another approach to the actual velocity for a gas or superheated steam or
is to assume that the actual expansion follows a relation
supersaturated steam
pv m = C (Review § 71 now ) Since this is an irre\ersible process, we are

s
Fig 15Z Repealed.

not justified in using J Aft = ft) dp for p>"* = C Howe\er we may arbi-
trarily ask ourselves what value of m in the exponent of the pressure ratio
actual velocity at 2', and find the
Pi/pa in equation (82) would yield the
answer from the following equality

U, - y/elvi

The m is not substituted for in front of the brackets for two reasons, first,
fc

by comparing equations (82) and (g), we see that these k’s are involved in an
ideal gas relation, cv = KR/[J(k
- 1)] and second it is easier this way
Thus

« 4-fen— fen m for a known efficiency It is all right


from w hich one may find a value of
to use pi instead of p« m equation (r)
§ 226 } ESTIMATED ACTUAL VOLUME 271

The shaded area work due to the irreversi-


in Fig. 146 represents the lost
bilities in In either Fig. 146 or 152, the lost work is hr — hi.
the nozzle.
Irreversibility is present not only because of the normal fluid friction on the
sides of the nozzle and on itself, but also in the case of steam because of the
sudden change from supersaturation to equilibrium, and if water particles
are present, because of the different velocities of vapor and water. The
vapor overtakes and impinges upon slower water particles. These impacts
are irreversible. In Fig. 146, the area under 2 '-b represents the further loss
from turbulence as the steam passes over the blades; some point b represents
the condition of the steam leaving the turbine stage.

226. Coefficient of Discharge. The ideal rate of flow is related to the


actual rate of flow through a nozzle (or other flow device, such as an orifice)
by a coefficient of discharge y,i, defined as

, .
_ actual mass of flow per unit time
^ Vd
ideal mass of flow per unit time

The value of this coefficient can be determined by test on a particular device.


Also, it can be estimated, sometimes with considerable accuracy, from past
data on similar devices.
In the case of the flow of supersaturated steam through nozzles, there is

some confusion as to whether to compute the ideal flow [the denominator


of equation (s)] on the basis of equilibrium flow or on the basis of super-
saturated flow. Hence, one should always specify which basis is used.

226. Estimated Actual Volume. For most practical purposes in adiabatic


expansions, it can be assumed that the actual volume at some pressure p 2 is
equal to the ideal volume after an isentropic expansion, unless the efficiency
during the expansion is much lower than it normally is in a nozzle. Perhaps
in converging-diverging nozzles, the use of the calculated “actual” volume
at the exit may
be worth while. At any rate, methods of computing w 2 are -

given below. This volume is needed in order to be consistent in the con-


tinuity equation, because in w' = A i-vr/vr, the values must be simultaneous
ones.
(a) Ideal Gas. If the substance is an ideal gas, equation (n) for the
efficiency may be written

ho — ht _ c p (T„ TV)
" ho - hi - c p (To - T 2y

from which 7V can be computed for a known or assumed efficiency. Then,


given p 2 = pi, we get v? = RTi'/pi. Or, if the exponent in pv m = C for
>

an irreversible adiabatic is known, v? = Vi(pi/pi) 1/m .


m NOZZLES [Ck 13
(b) Equilibrium Flow If a vapor expands through a nozzle in eqm
librium, the efficiency equation (n),

hi
ht

can again be used to compute hi at any desired point 2' Now with ht
and pi = pi, we may determine all properties at 2' from the vapor tables
If point 2' is in the wet region, the wetness is computed from

hi = (hp — then vi = vB — y 2 v/e cu ft /lb

(c)Supersaturated Flow. The easiest way to compute the actual volume


during a eupcrsaturation flow is first to find the value of from equation m
(r) and then determine vg = v t (p 2 /pi) l/m. This is the same thing as com-
puting Vi from e»(p tv t — piiq) = p t vt — ptv t

227 Example Compute the actual flow through the nozzle of § 222, super-
saturated conditions, if the efficiency up to the throat is 98 What is the coeffi- %
cient of discharge? The data are pi - 160 psia, h «= 400T, p, = 10 psia, initial
velocity negligible
solution Trom § 222, we have p, — 87 psia m = 3 008 cu ft /lb v, ,
- 1590 fps
and A, = 4 36 sq in Letting £ = 13, and using equation (r), we get

-['-ri-er
0 9S[1 - («*««)• v»i) - 1 -
The value of the bracket on the left was found in § 222 as 0 131 Simplifying and
using the log log scales on jour slide rule find (m — 1 )/m = 0 220 and m = 1 29
Then

4 82 cu ft /lb

(0 98)
l/,
{1590) = 1570 fps
A,x>\ (4 30) (1570)
9 88 lb /sec
(144) (4 82)

This mass will pass across the exit section of the nozzle accordance with the m
general indications of Fig 151 The coefficient of discharge based on supersatura
tion is rid = 0 88/10, or 98 8% (The computations for a problem like this should
be made with logarithms and a calculating machine )

228 Diffuser. A diffuser receives a fluid at a (relatively) high velocity


and discharges it at a higher pressure and lower velocity the reverse of the —
nozzle process That is, some of the initial kinetic energy, which is 100%
available, is used as the w ork of compression The same conditions 0 = 0 ,

and W = 0, are applied to diffusers, Fig 153, to get the same energy relation,
equation (a),
Ml*
§ 229] EJECTOR 273

except that often the final velocity (and Kj) is negligible. Letting K-, = 0
and integrating jv dp = J (h — 2 hj) for pv
k = C as before (Fig. 18), we get

(t) ur = {poVo - pivi),

where p 0 is the stagnation pressure, which is virtually equal to the discharge


pressure when the final velocity is relatively small. The manipulation of
this equation is left to the reader inasmuch as the forms which it may take
We are able to maintain a fine control over
will parallel those for a nozzle.
the expansion process in a nozzle, with the result that the efficiency is high
(see Chapter 6). But if the initial velocity is high, the efficiency of a diffuser
will be less than that of a similar nozzle, because of the greater turbulence
in the diffuser. The venturi is quite efficient because the pressure change
is small. Let the critical pressure ratio be pi/pi, defined by equation (81);
then by way of comparison:

Nozzle Diffuser

If 21 <— Vt
i if 2i >2< (m<P}\
Pi Pi Pl Pl \P2 P,J
u2 > sonic velocity, converging- V! > sonic velocity, converging-
diverging nozzle. diverging diffuser.

If — > El, If 21 <2-


Pi pi Pi Pi
u2 < sonic velocity, converging Ui < sonic velocity, diverging
nozzle. diffuser.

In each instance, pi the initial pressure and p 2 is the final pressure. Since
is

the actual process an irreversible adiabatic, an appropriate value of m


is

in pxT- = C can be found after the manner explained for the nozzle. It
is convenient to use a diffuser efficiency e d, (Fig. 153),

(h 2 — hi),
(u) ed = - hi ’
h2 -

in which, strictly, states 2 and 2' are stagnation


states, ideal and actual. Thus, the initial velocity
necessary to compress to a state 2' is

u( = [2 gj(h,- - hi)]
1 '*

Fig. 153. Diffuser. where 1-2 is an isentropic path.

The diffuser is also an integral part of the injector or


229. Ejector.
ejector. These two terms are applied to essentially the same instrument.
The operation of an injector or ejector is as follows: Steam enters the usual
274 NOZZLES [ Ch IS
type of divergent nozzle at A, Tig 154 At B, the pressure is low and the
is high
velocity The low pressure in the region B causes the gas or liquid
to be sucked into the device, and the momentum of the fast-moving jet of
steam carries the substance into the diffuser (through C) where the velocity
decreases and the pressure increases until the exit section D is reached

fa) (b)
Fig 1S4 Ejector. AB is a steam nozzle, BD is the diffuser Substance being
pumped enters at C The injector, used as a feedwater pump, is similar to this device,
and it pumps the water to a higher pressure at D than exists at A

Ejectors are commonly used to rid evacuated spaces (steam condensers, for
example) of non condensable gases such as air, and also as pumps, called jet
pumps, in vacuum refrigerating systems A single-stage ejector can main-
tain a vacuum of some 20 in Hg, two-stage, up to about 29 3 Hg, three m
stage, up to 29 9 in Hg, and a four-stage ejector can maintain absolute
pressures of the order of 1 mm
Hg Since the thermal efficiency of the
ejector is low, rotary or reciprocating pumps are generally preferred except
where large volumes of gases or vapors arc to be handled

230. Closure One of the most common errors made by students is in


the units of A and p, which should be square feet and pounds per square
foot, respectively, unless conversion constants are used or unless the units
cancel, as ina ratio (pj/pO
Maybethe reader has wondered about “the velocity” of a stream flowing
in a channel, inasmuch as there are different velocities at different point*
Thus, we have been obviously thinking of the velocity as a mean or average
value, otherwise computations of w from u) — Av/v are meaningless
Looking through the literature, you will find many different equations
which purport to give the mass flow through a nozzle under certain circum-
stances Nearly all the useful ones can be derived from equation (h), p
262, with possibly the inclusion of a coefficient of discharge If you like

mathematical problems, it will be fun to simplify equation (h) for particular


conditions For example assume that maximum flow occurs, that is, that
as defined by equation (81) and show that
p 2 is the throat pressure
- ,)
'
2 V /(t
2 gak T
= A p0 t
,k + 1/ Uw + dJ lb /sec

for any ideal gas, or that for air,


Courtesy Johnson Publishing Co , Cleveland

James Wall and His Steam Engine.

275
§ 233 ] CONCLUSIONS FROM THE CARNOT CYCLE 277
internally reversible and it approaches external reversibility as the tem-
perature differences T - T0
2 and T, - 1\ approach zero.
Since the refrigeration of this cycle is the
heat added Qa, we find

(b) Refrigeration = Qa = 2T(s 0 - sd ).

The work is S Q or Qa — Qrt, and will be a


negative number if expressed in its fundamen-
tal form, because Q R > QA in a reversed
cycle. The negative sign also shows that net
work done on the substance. Observing
is

these facts and remembering that the signs


Fig. 155. Reversed Carnot
given to energy changes are arbitrary, we may
Cycle.
as well write the expression for the work as a
positive number and avoid the inconvenience of a negative sign. Accor-
dingly, we use
IF = Qr — Qa
and find
(c) Work = IF = T-,(s b - s e) - T x (s a - sd ) = (T 2 - T,)(s a - sd ),

where Sb — s is seen to be equal to s„ — s d


c . Then the coefficient of per-
formance COP from equations (b) and (c) is
/ _ refrigeration _ T i

^ ~ work T2 — T\
[CARNOT OR OTHER REVERSIBLE CYCLE]

From a previous discussion of the Carnot cycle, we realize that the coefficient
of performance of a refrigerating cycle as expressed by equation (d) is the
highest possible coefficient for all cycles operating between the temperature
limits of Ti and T 2 In this sense, equation (d) gives us a standard of com-
.

parison; yet, as in power cycles, we find it desirable in practice to set up


other ideal cycles more nearly approaching the actual events.

233. Conclusions from the Carnot Cycle. There are certain general con-
clusions which can be drawn from the Carnot cycle.

1. It is desirable that the work to activate the cycle be a minimum, since it must

come from an outside source and be paid for.


2. The work will be reduced as the temperature 7\, at which heat is discharged,
this temperature
is lowered. It follows that we should take all practical steps to keep
down. The lowest temperature Tc Fig. 155, that is attainable by a natural coolant,
,

such as the atmosphere or the water from a nearby lake or river, is the most eco-
nomical. There is thus a limit to improvement of a refrigerating cycle at this point.
3. The work will as the temperature T x Fig. 155, is increased.
be reduced
Thus, ,

temperature
it is economical of energy carry on the desired refrigeration at as high a
lo
as possible. To freeze water, temperatures below 32°F
There are limits here too.
278 THE REVERSED CYCLE [Ch. U
are e«ential To cool air for some purposes, much higher temperatures may be
u«ed
4 For particular temperature limits, the heat exchanges should take place at
constant temperature for the maximum possible effect This condition is approach-
able in the event of \ apor refrigeration, but not when a gas is the refrigerant

234. Definitions and Ratings. The Btu is too small a unit to be con-
venient in rating commercial refrigerating plants The larger unit used in
refrigerating practice is roughly defined as the number of Btu required to
freeze one ton of water at 32°F into ice at 32°F The heat of fusion of ice
is very nearly 144 Btu per lb therefore, to freeze 2000 lb there must be
;

abstracted (2000) (144) = 2SS.OOO Btu This precise unit, 2S8.000 Btu,
is the definition of a standard ton of refrigeration

To specify capacity we must know how long it takes to perform a particu-


laramount of refrigeration One standard commercial ton of refrigeration
isdefined as 288,000 Btu absorbed at a uniform rate during 24 hr When
the engineer speaks of a ton of refrigeration, he generally means a standard
commercial ton Thus, on an hourly basis, a ton of refrigeration is

28 |°°°
= 12,000 Btu/hr ,
or . 20o Btu/min

Ifwo say that a plant is a 10-ton plant, we arc stating the commercial rating
which is the same thing as saying that the refrigeration may be at the rate
of (10) (200) = 2000 Btu per min
A common method of expressing the actual efficiency of a compression
system (§ 235) is to gn e the horsepower used per commercial ton of refrigera-
tion This number, which is not a dimensionless number, is obtained of
course by test of the actual machine A relation between the horsepower
per ton and the coefficient of performance is easily obtained from the defini-
tion of the coefficient of performance
Let the refrigeration be given as N
tons = 200.Y Btu per min Let the
horsepower required for this N
tons of refrigeration be represented by
hp, then, the corresponding w ork is 42 4 hp Btu per min (42 4 = 25
?6o)
Thus, the coefficient of performance COP is

,
<•>
,
cop = .
T - —^— m
refrigeration _ 200.V

from which 11' — (refrigeration ) / and the horsepower per ton of refrigera-

tion is
hp 200 _ 4 72
w A' 42 4y y

This expression may be used for either actual or ideal cy’cles

235 Vapor Compression System. The most common method of securing


refrigeration is by a 1 apor-comprcssion system, so called because the vapor
§ 235} VAPOR COMPRESSION SYSTEM 279

(working substance) is carried from the low-temperature region to the high-


temperature region by compressing it. The devices necessary to carry out
the vapor cycle are simple and are represented diagrammatically in Fig.
156. (See also Figs. 160 and 161.) In the ideal case, all flow is of course
without friction, except flow through the expansion valve, and all processes
except those in the condenser and evaporator (cold room) are adiabatic.
Figure 157 shows the reversed idealized vapor cycle on the Ts plane with
numbers corresponding to those on Fig. 156. Starting at state 1, the
vaporous refrigerant enters the compressor, which may be either a centri-
fugal or a reciprocating machine. The refrigerant is at a low temperature
as it enters the compressor, but inasmuch as the compression is actually
nearly adiabatic 1-2', Fig. 157, and ideally isentropic 1-2, it leaves the

Fig. 156. Compression System for Vapors. Fig. 157. Refrigeration


Cycle.

compressor at some pressure p 3 such that the saturation temperature cors


,

responding to this pressure is above the normal sink temperature T0 Thi- .

relation of temperature must exist in order that the substance may reject heat
to the natural receiver. Leaving the compressor in condition 2 (or 2'), the
vapor enters coils in the condenser. As a rule, the heat of condensation is

carried away by steam condenser, but sometimes


circulating water, as in the
circulating air absorbs the rejected heat Qn The condenser removes the
superheat (if any), the latent heat of vaporization, and generally subcools
the liquid a small amount, say from/ to 3, Fig. 157. In state 3 as it leaves
the condenser, the liquid enters an expansion valve, which is a throttling
valve separating the region of higher pressure from that of lower pressure.
This valve is used to regulate, either manually or automatically, the flow
of refrigerant. For the throttling process 3-4 through the expansion valve,
h z = hi. In order that hi, at the lower pressure, be equal to h 3 at the
higher pressure, a small amount of the liquid must vaporize. At 4 then,
there is a very wet mixture. The wet mixture enters the evaporator
280 THE REVERSED CYCLE [Ch JJ)

in state 4, absorbing heat Qa from (doing refrigeration on) the surroundings


process 4-1 The surroundings may be a cold room, as the inside of the
household refrigerator, or another substance In commercial ice manu-
facture, cold brine circulates about the cans containing water, taking heat
from the water for the purpose of freezing it The brine then flows through
the evaporator, where it is cooled again, thence to return to pick up more
heat from the water m
the cans W hen air is to be cooled, it generally
circulates directly about the e\ aporator coils
The energy diagram of Fig 156 shows that Qa — Qr = W, as usual If
the system is operating in a steady state, the process in the condenser and m
the evaporator is a .steady flow process in which A K — 0, IF = 0, and

Fig 156 Repeated Fig 157 Repeated

Q - Aft I§ 67 equation ( 1 )]
Thus either for constant pressure as m Fig
157 or for steady flow (A, = A 4 ),

(g) Qa = hi — ht — hi — As Btu/lb [refrigeration]

The heat rejected carried aw ay b> the cooling w ater is

(h) Qr = hi - As Btu/lb

As a positive number, the work is

(1) W= Qr - Qa = A, - A, - (A, - A,) — hi — hi Btu/lb

Now Aj — Ai is observed to be the difference of the enthalpies at the ends


of the lsentropic compression (si = si) From our study of the Rankine
cycle, we are reminded that this energy is represented closely b> the area
1 2 3-e-l, Fig 157 The area m-4-1 n represents the refrigeration The
area n-2/-3 q represents the heat rejected in the condenser The value of
Aj is taken as that for saturated liquid at temperature 1%, the small amount of
subcooling being alw ays quite negligible The coefficient of performance of
§ 2S7] EXAMPLE 281

the ideal vapor cycle for refrigeration is

(i) 7 = h2 — ~
hi ho hi

Since the actual compression process is irreversible (with increasing


entropy) to some final state 2', Fig. 157, we have

(k) IF' = h,’ -In- Q Btu/lb.,

from equation (9A) § 28, for A K = 0, with the signs of the terms on the
right-hand side changed in order to give a positive sign for TF' (in this
chapter only). If the compression is adiabatic, Q = 0. The compression
efficiency is 77 = 1F/1F'.

236. Displacement of the Compressor. For a particular refrigerating


capacity, the size of the compressor depends upon the number of pounds of
refrigerant that must be circulated per unit of time to obtain the desired
refrigerating effect, and upon the specific volume of the substance at the
intake pressure of the compressor. Suppose the plant is to have a capacity
of N
tons; then the rate of refrigeration is 200 Btu per min. Now if the
refrigeration per pound of refrigerant is In — In Btu, Fig. 157, then the
mass of refrigerant circulated is

Btu /min. 200iY


(1 ) W ~ — lb./min.
Btu/lb. In In

The condition 1, Fig. 157, is known or assumed, so


of the refrigerant at
that the specific volume may
be determined. Evidently (w lb./min.)
iq

times (iq cu. ft. /lb.) is the needed displacement volume Vo cu. ft. per min.
for 100% volumetric efficiency. Review § 117 again. If the volumetric
efficiency is p r , we have
v, /_200 N_\
(m) cfm.
Vt \h 1 — hi

Practical values of the volumetric efficiency should usually fall within the
range 05-85%. The factors affecting volumetric efficiency in vapor com-
pressors are much the same as those discussed in § 1 17. Also, the types of
compressors used are reciprocating or rotary. Most of the discussion of
Chapter 7 applies to refrigerant compressors, except as an ideal gas is
postulated.

237. Example. compressor receives wet vapor at 10°F and com-


An ammonia
presses it The temperature at the expansion valve
to a saturated state at 190 psia.
is 85°F. The compressor is double acting, 12x14 in., it runs at 200 rpm, and its
volumetric efficiency at normal operating conditions is 78%. Let the compression
efficiencybe i) = 80%. For the ideal cycle, determine (a) the coefficient of per-
formance 7, (b) the tons of refrigeration N, and (c) the horsepower per ton. On
the basis of the actual fluid work, determine (d) 7, (e) N, (f) hp/N, and (g) the
temperature at 2', Fig. 158, if the actual compression is adiabatic.
V

282 THE REVERSED CYCLE [Ck 14


solution From the NH* tables in the back of this book or the complete tables
of the U S Bureau of Stmdards, Bulletin No 42, ive find for state 1 at 10°F, state 2
at 190 psia, and state 3 at 860 F

h/i = 53 8 s,i = 1 3157 A oJ = hi = 632 4


A,„ = 501 1 »/, = 0 0244b sti = st = 11802
«/, = 0 1208 »,i = 7 304 A, = k4 = 137 8

From *i = s, ne get

1 1802 - 0 120S + 1 1 949x1, or z, = 88 7%


A, = 53 8 + (0 887) (5G1 1) = 551 5 Btu/lb
Fig 1SS 0, = 0 0244 + (0 887) (7 28) = G 48 cu ft /lb
(a) For TF = Aj — Ai *> 632 4 — 551 5 = 80 9 Btu/lb and the refrigeration
Q* « A, - Ai = 551.5 - 137 8 - 413 7 Btu/lb we have ,

_Qa _ 413 7
7 IF 80 9
5 11

(b) The displacement in cfm is, for 2n diigrams/min in a double-acting com


piessor,

7r{144){14)(2 X 200)
366 cfm
(4)(1728)

For t], = 78% the volume V\ drawn t n per minute is V[ - (0 78) (3GG) cfm, and for
0i = 6 48 cu ft /lb , the mass of refrigerant circulated per minute is

V\ (0 78)(306)
44 1 lb /mm
»i 0 48

Thus, the total refrigeration is (44I)(413 7) Btu/mm Dividing this value by


200 Btu/nnn ton gives

(c) The horsepower for 44 1 lb /min of refrigerant is

(44 1)(80 9)
k 84 2 hp
42 4

and the horsepower per ton is 84 2/91 2 = 0 942


(d) The actual work ol the fluid is = IF/17 = 80 9/0 80 = 101 W 1 Btu/lb,
and for other conditions the same,

413 7
=
101 1

(e) If the actual rate of circulation of refrigerant remains the same as in the ideal
cycle (the horsepower of the motor must be greater than before), the tons of refrig-
erant are the same as before, 91 2 tons
§ 238 } REFRIGERANTS . 283
(f) The actual fluid horsepower is the ideal hp divided by the compression effi-
ciency, or hp' = S4.2/0.80 = 105; thus

hp' _ 105
— 1.152.
ton 91.2

Including mechanical and other losses, the actual bhp per ton will be somewhat
greater than this value.
(g) For an adiabatic steady flow compression 1-2', the work as a positive number
is hf — hi — (h~ — Ai),/0.80, or

hi’ = hi + gQ^‘ = 551.5 + = 652.6 Btu/lb.

In the complete tables at 190 psia and h = 652.6, we find the temperature at 2' as
h- = 122.4°F.

238. Refrigerants. Table XI shows a comparison of some character-


istics of popular refrigerants. Some of the desirable qualities of a refrigerant
are:

1. Refrigerants are preferably nontoxic, so that in case of leakage no one is in

danger of injury. This attribute is of paramount importance in air conditioning


systems and home refrigerators, for example, but toxic refrigerants are tolerable in
commercial installations where reasonable precautions are taken. The Carrenes,
Freons (trade names) and carbon dioxide are not toxic, but the others in Table XI
are in more or less degree. Methyl chloride is not only toxic, but also is practically
odorless; where this refrigerant is used, a warning agent (acrolein) which is irritating
to the eyes and nose should be added.
2. Refrigerants should be economical, both in initial cost and in maintenance.
Maintenance problems include: controlling leakage (there is less trouble with leakage
of large molecules than of small ones) providing adequate lubrication (the refrigerant
;

should not react with the oil to destroy its lubricating qualities); and avoiding
corrosion (the refrigerant should not corrode the materials which it contacts).
Also the refrigerant should be readily available for recharging the system when
necessary.
3. Refrigerants should be nonflammable. A number of hydrocarbons have been
and are used as refrigerants, examples of which in Table XI are butane and pro-
pane. These and others (ammonia, methyl chloride, etc.) constitute a fire and
explosion hazard. The other refrigerants in Table XI are nonflammable.
4. Refrigerants should have a high latent heat at the evaporator temperature
and a low specific volume. The type and size of compressor
(see Fig. 108, p. 206)
isa function of these physical traits. If the latent heat is high (see 3 Table NH ,

XI), much refrigeration is done by each pound of substance circulated; if in addi-


tion the specific is low, the volume of substance to be circulated and there-
volume
fore the size of compressor and passageways are small. Notice that when the
ideal displacement volume V D is small, reciprocating compressors are feasible; when
V „ is large (Carrene 1, F 11, and F 113), centrifugal compressors (which can be
run at high speed) become necessary. To some extent, however, the total amount
of refrigeration to be done may affect the choice of the type of compressor to be
§ 239] ABSORPTION SYSTEMS OF REFRIGERATION 285

being at or above atmospheric pressure in the evaporator; for example, butane is


approximately at atmospheric pressure when the saturation temperature is at 30°F,
Freon 114 at 39°F, and sulfur dioxide at 14°F. Also important is a low pressure
ratio p : /pi, because of the effect of this ratio on the work necessary for compression.

G. While the foregoing attributes are perhaps the most significant, there are

miscellaneous other characteristics which are desirable (124): good thermal con-
ductivity for rapid heat transfer, wetting ability, inertness (the refrigerant should
not react in any way with the materials it touches), stability (the refrigerant should
not break down into different matter of smaller molecules), low viscosity (for ease
of movement), high critical temperature, and a high dielectric strength (in her-
metically sealed units where the refrigerant contacts the motors). Also, it stands
to reason that the refrigerant should not solidify at any temperature in its cycle.

The traditional refrigerants are still widely used; ammonia in industrial


and commercial refrigeration; carbon dioxide, because of its low critical

temperature of S7.S°F, is used principally in the northern latitudes and for


low temperature refrigeration (solid CO*, known as dry ice, is widely used
for cooling purposes) sulfur dioxide is extensive^ used in industrial applica-
;

tions because of its advantageous thermodynamic traits, but it is toxic.

The “newer” refrigerants (for example, the Frcons, often designated by


the letter F and a number, as F
appeared when physical chemists found
12)
that chlorine or fluorine atoms could be substituted for hydrogen atoms in
certain hydrocarbon molecules (the methane and ethane series). When
substances evolved which were neither flammable nor toxic and had satura-
tion temperatures and pressures suitable for refrigerants, they were found
to be more desirable in mam" situations than the traditional refrigerants.
They are extensively' used and can be manufactured with nearly any desired
property, though not necessarily^ with all the properties desired. For
household refrigerators and air conditioning, F 11, F 12, F 114, and Car-
rene 7 are common.
A significant commercial advantage of the line of newer refrigerants is
that a manufacturer may' use the same compressor for different capacities
by changing the refrigerant and installing a different size of motor. For
example, for a particular displacement, a greater refrigerating effect is
obtained from F 22 than from F 12; greater power is also required for the
F 22.

239. Absorption Absorption systems are char-


Systems of Refrigeration.
acterized by the absorbed by an absorbent on the
fact that the refrigerant is

low-pressure side of the system and is given up on the high-pressure side.


The advantage derived from these additional processes is that the work
supplied to the system needs to be only enough to pump a liquid from the
low-pressure to the high-pressure region of the system. We have had occa-
sion to note several times before that the work to pump a particular mass of
gas or vapor is much larger than the work to pump a like amount of liquid.
,

m THE REVERSED CYCLE [Ch U


depicted m Fig For the larger temperature differences between the
160
condenser and evaporator, a regenerative process may be worth the money
In operation, the first expansion valve loners the pressure part way, e to r
The saturated vapor, indicated by the quality at r, Fig 160(b), mixes with
the superheated vapor b (or b ') resulting in the condition c at entry to the
second stage of compression Compression cd (or
then occurs, whence cd')
the refrigerant, 1 + m lb ,
enters the condenser lb of vapor While the m
leaves the flash chamber, the 1 lb moves through the
of remaining liquid
second expansion valve hi and enters the evaporator where the refrigeration
occurs Thus, in the diagram of Fig 160, it is intended that calculations
be based on one pound of refrigerant (per any unit of time) doing refrigera-
tion Actual computations will show that it is more efficient than would be

Fig 160 Two Stage Compression with Regenerative Inlercoohng The discharge
from the low pressure stage at b may be circulated through the flash chamber with a
resulting lower temperature at e

the case for single stage compression from a to the pressure line pd and
throttling from e to u An energy balance for the mixing operation at k,
Fig 160(a), is

(n) h b 4- mh n = (1 + m)he

In the flash chamber, Fig 160(a),

(o) (1 + m)h. — h, + mh n
Also, the refrigeration Q*. and the total w ork IF are

(p) Qa — ha — h t Btu/lb ,

(q) IF = IF, + IF* = h- A. + (1 + m)(hd - K) Btu/lb

both per pound m


the evaporator
If the cold body is to be maintained at a very low temperature, it may be
advisable to use two or more separate refrigerating systems with different
refrigerants, as suggested by Fig 161, where the intermediate heat exchanger
acts as both a condenser and an evaporator If Wi is the mass of refrigerant
§ 241] LIQUEFYING GASES 289
circulated per unit of time in the low-temperature cycle (this is determined
by the amount of refrigeration required), then the amount of refrigerant

Fig. 161. Cascade Refrigerating System.

w 2 for the other cycle must be such as to make a heat balance in the con-
denser-evaporator ;

(r) Wi(h b — h c) = w 2 (h —
e h,).

In this system, called the cascade system, the number of stages may be
more than two. The successive refrigerants should be such that the triple-
point temperature of the “higher” one is lower than the critical temperature
of the “lower” refrigerant.

241. Liquefying Gases. In order for a gas to exist as liquid, its tempera-
ture must be reduced below its some values
so-called critical temperature,
of which are given in § 172, p. 208. For example, nitrogen must be at
or below a temperature of — 233°F. There are various means used to attain
temperatures of this order, one of which is to use the cascade idea (Fig. 161).
N
One cascade system designed to produce liquid 2 is made up of four refriger-
ating systems in series, the refrigerants being NIi 3 C 2 4 CEL, and 2 in
,
H ,
N ,

that order. Similar arrangements are used to liquefy other gases. One
method of producing oxygen is to liquefy air and then separate the oxygen
by fractional distillation.
Some “permanent” gases can be liquefied with equipment simpler
of the
than the cascade refrigerating system of four refrigerants. The simplest
method of doing this is the Linde process, diagrammed in Fig. 162. Gases

Fig. 162. Linde System of Gas Liquefaction.


290 THE REVERSED CYCLE [Ch U
"Inch soltdifx at a temperature higher than the desired liquid temperature
nre remoxcd from the make-up, for example, cleaning the air of CO* and
IIjO before it is, added to the liquefy ing s\ stem The compressor A dehx ers
air for example, at procures of 100 to 200 atm If the di charge tempera
turn at B i» greater than that of the naturally a\ ailable «mk, the air passes
through a cooler BC, which loners, its temperature to approximately that
of the environment Then it pn-^es CE through a regenerate e heat
exchanger D, n herein its temperature is lowered to some xalue in the
x lcimtx of its critical temperature to about — 22l°F for air The expansion
xnlxe E throttles the gas to atmo phene pressure Ah = 0, after which some
of it is liquid and some is \ apor The liquid about 10^ of the fluid enter-
ing F after steady state has been attained is drawn off at H as de ired and
the vapor pas.es GJ back through the heat exchanger acting as the heat
receiver for the heat gixen off from C to E The \apor at J, together with

F i? 162 Repeated

the make-up re-enters the compressor If instead of an expansion valve

E in engine is u cd and work is produced more liquid per unit of work


expended on the system is produced \t the low temperatures ltnohed
lubrication difficulties practicilhpreclude a reciprocating engine but a
turbine max be u ed Oillx a part of the flow is likelx to be expanded
through an engine and this part at its low temperature is u ed for a further
regeneratne tooling of the pirt which goo dim th to the expan ion ^ ahe
The reider who has followed this text from the beginning hould haxe little
trouble in writing equations for mass balances and energy b dances from
Fig 162 as desired Notice the processes 1 2-3-4 on the Ts plane
In order for a gas to become cooler on passing through an expan ion x ahe,
the value of its Joule-Tliom on coefficients (dT dp)* mu t be poutiy e
(^ee footnote on p 62 ) For this to be so the temperature of the gas mu t
be below its maximum mrrrwn temperature which i- roughly a reduced
temperature of 5 that is the lmer-ion temperature is approximately fixe
times the critical temperature (1) The inxer-ion temperature is that
temperature aboxc which the Joule-Thomson coefficient is negatnc below
w Inch it is poutn e l uallx the xalue of the inversion temperature pre ents
no problem becau e it is high enough to be out of the wax
‘ Ilowexer
it mu t be considered m hquefxing hxdrogen inxer-ion temperature of
3ol°R, and helium, whose imer-ion temperature is 42 5°R Throtthng of
§ 242) REVERSED CYCLE FOR HEATING 291

these gases atnormal temperatures results in a temperature increase. To


liquefy helium,we cool it below 42.5°R with liquid hydrogen, which has
been obtained by cooling H 2 below about 351°R with liquid nitrogen.
242. Reversed Cycle for Heating. When the reversed cycle is used for
heating, it is popularly called a heat 'pump (all reversed cycles are heat
pumps). As usual, the Carnot cycle furnishes a simple means of under-
standing the flow of energy and provides the highest conceivable standard
of performance. In Fig. 163, consider the working substance in some con-
dition c, a few degrees above room temperature, and let it expand isentropi-
cally to a temperature U, below outside atmospheric temperature. The work-
ing substance may now be led through outside coils and heat will flow from the
outside air (or earth or water) into the substance, process da. In condition
a, the substance is compressed to a temperature 4 above room temperature,
,

e f s
Fig. 163. Healing Cycle. This cycle approaches external reversibility as the AY’s
approach zero.

and led into inside coils; so heat now flows from the substance into the
room, process be. For the heating cycle, we have

Q a = Ti(sa — s d), heat taken from the sink;


Q r = T 2 (sa — Sd), heat rejected to the room;
W= {Ti — Ti)(s a - sd ), work for the cycle;
,
.
_ output _ heating effect _ T2
S
^ input work T2 — T\

the highest possible coefficient for the specified temperatures. If a coeffi-

cient ofperformance is 7 = 5, and if the work is 100 Btu, the heat rejected
to the room will be 500 Btu. We see that this system provides a means of
delivering an amount of heat much in excess of the heat equivalent of the
work expended. Note that the coefficient of performance of a reversed cycle
depends upon whether the objective (output) is cooling or heating.
There are several commercial machines available which are designed to
operate as air conditioners both summer and —
winter cooling the inside air
during the summer and heating it during the winter, by merely throwing a
292 THE REVERSED CYCLE [CA 14

switch See Fig 164 During the cooling season, the outside coil B acts
as the condenser, receiving the discharge from the compressor The coil A
is the ev aporator and it cools the house air which flows ov er it For heatmg,
the coils A and B swap functions, the coil A becomes the condenser, adding
heat to the house air which circulates ov er it, the coil B is the evaporator,
picking up heat from the outside

(a) Cooling (b) Hevtmg


Courtesy General Electric Schenectady, A 1

Fig 164 Heat Pump for Heating and Cooling

It is not often that the rev crscd-cj cle machine is an economic means of
heating alone, but where refrigeration is needed for other purposes such as
air conditioning, it will be easier to justify its use Favorable factors
include low rates for electricity, a mild climate, the availability of rela
tivcly warm water as a source of heat, and, in a negative way, expensive
heating fuels The unit operates more efficiently if the summer condenser,
coil B, is cooled by water In small installations, the extra cost of water
may more than offset the loss of efficiency which goes along with air cooling
243 Closure As usual m chapters dealing with a specialty, only the
basic elements of refrigeration are giv en here One must refer to boohs and
current literature devoted to the subject for more detail [See references

(114 *«?) 1
15

GASEOUS MIXTURES

244. Introduction. Since you have already worked many problems


with air as the working substance and since air is a mixture of gases, you
already know how to handle mixtures of ideal gases given the appropriate —
characteristics ( R ,
c p , c„, etc.) of Review §§ 43 and 56 now,
the mixture.
noting Avogadro’s law, the universal gas constant, the meaning of a mol, etc.
The symbol M represents molecular weight; thus, x is the molecularM
weight of gas X. A mol of a substance is lb. or M M
gm. of the substance.
These mols are different and to distinguish between them, we say pound mol
(lb. mol) and gram mol (gm. mol or g-mol). However, we shall use only
the pound mol in this book and will therefore speak of mols without the
defining adjective. It will be helpful now to think of the units of as M
(a) •
M— > lb. /mol; N mols have a mass of NM lb.,
where N is the number of mols of substance with molecular weight M.

A mixture of gases is considered to have an equivalent or apparent molecu-


lar weight; thus, air is said to have a molecular weight of 28.97. If the
equivalent molecular weight of a mixture of gases is known, the specific gas
constant R is found from
1545
R =
(b)
M ’

where 1545 = R ffc-lb. per mol-°R is the universal gas constant.

245. Characteristic Equation of Ideal Gas in Terms of Mols. Recall


that at a specified pressure and temperature, the volume of one mol of one
ideal gas is the same as the volume of one mol of any other ideal gas, § 56.
From pv = RT, we get pMv - MRT = RT, where M is the molecular
29S
GASEOUS MIXTURES [Ck 15
weight of the gas or mixture of gases The \ olume of one mol is i — l/i?
and if there are \ mols of gas the ideal gas equation becomes

(c) pi -HT or pi - \RT - 1545\r


I* mol] [\ mols}

where 1 is olume p is in p«f and T in °R


the total \ Equation (c) is applied
to a single component or to a mixture of gases Applied to component Y
we haxe px =* A JlTm where the partial pressure of the gas Y
p x is

Arx is the number of mols of I« and Tm applj to the mixture and the
component

246 Mixtures of Gases The individual gases or vapors in a mixture


are called constituents or components The description of the mixture is
guen b} a volumetric analysis or bx a gravimetric analysis Sometimes
one anal} sis is useful sometimes the other hence w e must be able to con
from one to another w ith ease
\ ert
The lolumetnc analjsis expresses the amounts of the components in the
mixture b> the percentages of the total \ olume which each component
would occupy if the xanous ga«es were placed in separate compartments at
the pressure p m and temperature Tm of
the mixture That is in Fig 16o imag
me gases \ 1 and Z separated b} par
titions at b and c each one at pressure p„
and temperature T m Thus each one
occupies a certain percentage or fraction
of the entire x olume represented bj Bz
B, and B (the svrobol B suggesting
bulk.) Nowthepartitionsareremoxed
if

Fg 16S Total Volume is V. and the ga^es mix wesaj that the xolu
XVhen Vm l cu ft. the volume of metric percentages of the ga=es A 1
gas Xis B of gas Y B, of gas ZB ^^ are ^
important to note that if the xessel in Fig 16o contains one mol of
^ B
It is
the total mixture the mols of gases \ 1 and Z are B z and B That is
the xolumetnc fraction B z is the number of mols of gas \ in a mixture
which total* one mol Bx -f B¥ B = 1 The i olumetnc percentage Br
+
is the number of mols of \ in a mixture w hich totals 100 mols

Bt + B, +B 100

when all gases are at the same temperature In terms of the actual known
mols of each gas N x \ „ and N we write

- \,+ \f + N,
§ 2!tT\ DALTON’S LAW OF PARTIAL PRESSURES 295
where iV m is the total number of mols in the mixture. Notice that

N Nj,
(d) Bx = N x By =
Nm and B = Nm
m
z
Nm
by volume of the constituents X, Y, and Z, respectively.
are the percentages
The volumetric percentage is also defined by

^ _ cu. ft. of gas X at p m and Tm


cu. ft. of mixture at p m and Tm


and so on for gases Y and Z.
The gravimetric analysis describes the mixture by giving the percentages
by weight (mass) of each constituent. For example, the gravimetric per-
centage or fraction of gas X is
mass of gas X
g _
total mass of the mixture

For a mixture of gases X, Y, and Z, the gravimetric analysis is Gx of X, %


Gv % of Y, and G.% of Z, each expressed as a percentage of the mass; G
also stands for the fractional part of the mass.

247. Dalton’s Law of Partial Pressures. It was John Dalton (1766-1844)


who first stated that the total pressure pm exerted by a mixture of gases (or
vapors) is the sum of the pressures which each gas (or vapor) would exert
were it to occupy the vessel alone. The pressure wdiich one gas in a mixture
exerts is called its partial pressure. Thus, if p z p v: and pz represent respec-
,

tively the partial pressures of the mixed gases X, Y, and Z, Dalton’s law
states

(83) pm - px + p„ +p + x

.

[T„= Tx = = T„ Vm=Vx = V„ = V,\


The characteristic equation (c) of an ideal gas in terms of the number of
mols is pV = 15451VT. Applying this equation to a component and to X
the mixture, we get

(e) PxVm — 1545Ar I 7’„ and pm Vm = 1545 N m Tm .

By division with equations (e), we have

px vm _ i545iy,r, vi =
Nm = b
?Ijl

p m Vm 1545 Nm T m pm

In the foregoing, we used the following principles: If the mixture is in inter-


nal thermal equilibrium all gases are at the same temperature T m ; each gas
occupies the same volume Vm ; we recognized that the number of mols x of N X
divided by the total number of mols N
m is the volumetric percentage; and
,

296 GASEOUS MIXTURES [Ch IS

v.o have arrived at the quite useful relation that the volumetric fraction B
is equal to the ratio of the partial pressure divided by the total pressure, or

(84) pz = B zpm , py ~ B up m , p, ~ B,p m ,

true for ideal gases and useful for actual gases

248. Molecular Weight and Gas Constant for Mixture Given Nz or


B x mols of X, N„ or B„ mols of Y, and N, or B, mols of Z, the corresponding
masses are

(1) (2)
N,M. lb B Z MZ lb /mol mixture
NV M, lb B y My lb /mol mixture
AM/, lb B, 1/, lb /mol mixture
Total = ZNM lb 2 BM lb /mol mixture (Mm)

If B* + Bv + B, = 1 in column (2), there is one mol of mixture and the mass


in pounds per mol is the molecular weight [equation (a)], thus, XBM = M m ,

the equivalent molecular w eight of the mixture By similar reasoning, there


are Nm = Nz 4- N„ 4- N, mols in column (1), and Mm = 2NM/Nm or ,

(f) il/„ = BZM Z + B M V + B.M. = IBM lb /mol,


V

Mm
,, N M Z + NyMy + N.M. = ZNM IU
Z
^
(g)
--
-jj- lb /mol,

where, in 2 BM
and 2 NM, B and N apply to the individual gases Also,
equations (G5) and (60), we see that 2BAf = 1/ {ZG/M) hence

w °” d «- = ^ftib/ib-R

249 Example A mixture of 64 lb of O, 281b ofN, and 16 lb of Hs is at 140°F


and 15 45 psia What volume does the mixture occupy?
solution The easiest w a> to find the volume (if one docs not know the standard
mol volume) is to determine first the number of mols of mixture thus

64 lb 28 ib =
No = 2 mol Oi, Nh = 1 mol Nj
32 lb /mol 28 lb /mol
1C lb —
Nn 8 mol Hj,
2 lb /mol

or total number of mols is 11

V - 1545 AT-T. = (1545)(11)(600) = 417Q ft


(15 45) (144)
]

§ 250 ENERGY CHANGES OF A MIXTURE OF GASES 297

Energy Changes of a Mixture of Gases. If we take the heat trans-


250.
ferred to orfrom one pound of a mixture dQ = cm dT as the sum of the trans-
ferred heats for the components, whose masses are Gx Gy and G z lb. per lb. , ,

of mixture, we have

Cm dT = GjCx dT + G^Cy dT + G c 2 z dT,


or

(85) cm = GjCz +Gc u u -j- GzCz = 2(7c,

where Cr, c.Jt and c- Btu per lb. are instantaneous or average specific heats of

gases X, Y, and Z, respectively, for some particular process, and C„ is the


specific heat of the mixture for the same process. Equation (85) is also true
for variable specific heats.
The molal
or molar specific heat C„ Btu per mol of a mixture is the sum
of the products of the individual molal specific heats and the corresponding
fractional part of the mol;

(i) Cm = B XC X + B VCV + B C = XBC X X Btu/mol,

where Cm is the molal specific heat of the mixture for a particular process
when Cx C„, and C, (etc.) are the molal specific heats of the constituents
,

for the same process ( Cx is the specific heat of a mol of gas X, and B- is the
fractional part of the mol of gas X, for example). Also C m = mcm M .

It is logical to expect that the total internal energy of a mixture of gases


is the sum of the internal energies of the components, energy being a scalar
quantity;

(j) um = Gxu x + Gytiy + Gu = z z ZGu Btu/lb.

And using a bar over the symbol to distinguish values for a mol, we have
(k) um = Bx~ix + BfUy + BxTi = z ZBu Btu/mol
of mixture. It is sometimes convenient to work with a known total number
of mols iV; thus

(l) Xm Um
+ NxUx = 2-Nu,
= NxUx + NyUy

where u is the internal energy per mol, N m = Nx + N + N the number y z is

of mols in the mixture, and Nx A and N (etc.) represent the number of


,
r
„, z

mols of each individual gas. We remember that Nx/N m = B Notice x.

that equations (f) and (i) can also be written in terms of the total number of
mols, after equation (1).

Equations analogous to (j), (k), and (1) can be written for enthalpy.
Gibbs’ principle is entropy of a mixture is the sum of the entropies
that the total

of the individual constituents of the mixture. Hence, we may write equations


for entropy s and s analogous to equations (j), (k), and (1). Note, however,
298 GASEO0S MIXTURES [CA 15
that two gases which are separated by a partition are allowed to mix by
if

remoi ing the partition, an increase of entropy results because the diffusing
process is irreversible Treating each component separately is often the
convenient way, but if many calculations are to be made for a particular
mixture it vv ould be easier to get the specific heats (i R etc ) of the mixture
and treat the mixture as we have already learned to treat air

251 Gas-Vapor Mixtures There are many instances where the engi
neer must deal w ith mixtures of gases and vapors that is mixtures which m
one component may condense and settle from the mixture or an additional
amount may ex&poratc into the gaseous mixture There is the mixture of
air and gasoline \ apor from the carburetor of an automobile engine In the
products of combustion of hydrocarbon fuels there is a relatively large
percentage of water vapor much of which will condense if the products
are cooled to atmospheric temperatures There are many industrial
processes of gaseous mixtures some of whose components may be easily
condensable Of more general interest is normal atmospheric air which
invariably contains some water vapor usually superheated steam You
ha\w inferred by vvoxx that the expression a mixture of gases and vapors
means that one or more components of the mixture may condense in more
or less amount during a process which the mixture is undergoing or at
least one component is acting so unlike an ideal gas as to make some of the
ideal gas equations inapplicable Mostly we shall be concerned below
with mixtures where some condensation or evaporation occurs during the
process
All the gases in the mixture the noncondensing components we shall
consider as the dry gas abbreviated dg The condensable component shall
be spoken of as the vapor abbreviated v Sometimes in discussing air
mixtures vv e shall use dry air da to mean the noncondensing components
of the atmosphere There are certain laws and principles which we shall
assume to be true Try to fix the following in mind
1 Dalton s law (§ 247) the sum of the partial pre«sures of the constituents
is equal to the total pressure of the mixture (liquid is not included of course)
Pm = p* p, V>
+ + +wl ere the constituents are Y 1 and Z
2 The volumetric percentage of vapor is A,/(A. \ ,) = + the volumetric
percentage of dry gas is \»/(N» + A(») — Y J\ m where mots of vapor
\ , «= mols of dg \„ total mofs of mixture (Equation (d) § 246 ]
3 The partial pressure of a constituent in a mixture of ideal gases is equal to its
fraction by volume (or mol fraction) multiplied by the pressure p„ of the mixture
(total pressure) § 247 e have

(m) p, = B,p m = and P'

partial pressures of the


where A, — mols of vapor A, = mols of dg p, an p, are 1

vapor and dry gas respectively p* — total pressure (See § 246


§ 252] DEW POINT 299
4. Each gas or vapor in a mixture occupies the same volume; V
m = Vx = V„, etc.
The volume of any one of the constituents at its partial pressure is the volume of the
mixture. Often this principle affords the easiest way to find the volume of a mixture.
5. Each gas and vapor in a mixture is at the same temperature internal thermal —
equilibrium obtains; Tm = T x = Tv etc. ,

6. The density of a mixture is the sum of the densities of the individual con-

stituents; p m = px + py + p + z
' '
' .

7. The per cent or fraction by weight (gravimetric percentage)


of a particular
constituent in a mixture is the density of the constituent divided by the density of
the mixture, Px!Pm, Pyl Pmt etc.

252. Dew Point. Imagine a container A, Fig. 166(6), in which there is


superheated vapor (say, H 2 0), state 1, Fig. 166(a). If this container is
placed in colder surroundings so that heat flows outward, the superheated

S
(a) (b)

Fig. 166. Dew Point and Relative Humidity.

vapor is cooled at, say, constant pressure, path 1-c. If heat is further
rejected after arrival at state c, where the vapor is saturated, some of the
vapor will condense. The state c is the condensation point or dew point at
constant pressure with respect to the initial state 1. Thus, any superheated
vapor has a dew point which depends on the initial state and the process by
which the saturated vapor line is reached.
In atmospheric air, the H 2 0 is generally superheated steam. If this air
is cooled at a total constant pressure, the steam cools at constant pressure

(all components of the mixture being taken as ideal gases). Some tem-
perature is eventually reached below which condensation of the H 2 0 in
the air occurs; this temperature is said to be the dew 'point of the air, but from
the preceding paragraph we see that it is the dew point of the H 0 in the air.
2

As explained later, we have means of determining the partial pressure p vl of


the H 2 0 vapor in the air. Knowing p vl we find the dew point from the ,

vapor tables by looking up the saturation temperature corresponding to


Pvi] and that is it. (You have often noticed drops of water on the cold
windowpane in the winter because the adjacent air has been cooled below
its dew point.)
If atmospheric air is cooled at total constant pressure below its dew
point c (say, to state 6, where the temperature is 6.), some of the vapor
j )

800 GASEOUS MIXTURES [Ch 15


necessarilj condenses but the vapor which remains at th is saturated and
at
a pressure lower than p, t the vapor pressure p rh is the saturation pressure
corresponding to the temperature U During cooling below the dew point
at constant total pressure the partial pressures do not remain constant
If p„ is the total pressure of the mixture
and p, is the vapor pressure the pressure
of the dry gas p„ is taken as
p„ - p m p. -
We shall generally assume that the liquid
formed by cooling a gas vapor mixture
below its dew point settles out which it will
do gi\ en time and that the vapor left is
saturated vapor (Atmospheric mists con
tarn small bits of liquid HjO which means
Fig 166(a) Repeated
that the vapor is wet vapor

263 Relative Humidity Consider again the container A Fig 1G6


with superheated vapor in it Suppose a small amount of liquid is injected
and the sjstem is manipulated in such a way as to maintain constant tem
perature (This can be done by letting the sjstem come mto thermal
equilibrium with its environment and then giving it time to do so again
after additional liquid has been injected ) Perhaps the first injection is
such an amount that it all v aponzes and the final equilibrium state is a
If just the right amount of liquid is introduced and vaporized at constant
temperature the equilibrium state becomes saturated v apor at d What
the foregoing discussion suggests is that a saturated vapor state can be
reached at constant temperature by evaporating more liquid mto the space
A Fig 16G The ratio of the v apor pressures at 1 and d p, /p,i is called
the relative humidity <p Notice that the relative burn dity of anj super
heated vapor can therefore be expressed as the actual pressure of the vapor
divided b the saturation pressure corresponding to the temperature of the
vapor
Returning to a gas vapor mixture we conclude that if additional liquid is

vaporized into the mixture (at constant temperature and total pressure
pm) the vapor becomes less superheated its partial pressure mcreases and
if the process is continued the vapor in the gas becomes saturated vapor
In applying these notions to atmospheric air w e saj that air which con
tains saturated steam (state deb or anj other on the saturated vapor
curve) is saturated air but we mean that the team is saturated Howev er
such air (or other gas) is saturated with steam (or other vapor) in the
sense that there cannot be an increase m
the amount of steam (or i apor) in the

air (or gas) as long as the total pressure and temperature of the mixture remain
the same
The partial pressure of the steam in the a r is ordinarily a fraction of a
pound per square inch (0 5069 ps a for saturated air at 80°F less if the
254] humidity ratio 801

“air” is not saturated; see your steam tables). At such low pressures, the
molecules are so far apart that they exert a negligible force on one another
and occupy negligible space; thus, the steam acts very much like an ideal
gas,

pv = RT, pV ~ NRT = 154 5NT, and p ~ pRT,

where N is the number of mols. We shall find the density p = 1/v of con-
venience in this chapter. Thus, for the vapor at states 1 and d, we have
Pvi/Ptd ~ (pviRvT)/ {pdRvT), and the relative humidity is

Vvd
( 86 ) ,

(f)
= Pci » Pel = }

Pvd Pvd Vvl

and where the approximation is often


where the subscripts refer to Fig. 166,
suitably accurate at low pressures. For steam, the ideal gas laws yield
reasonably good results when its pressure is below 1 psia. At temperatures
below 32°F, the partial pressures of saturated steam are given in Keenan
and Keyes (71). Accurate properties of air-steam mixtures are found in
the ASHVE Guide (127).

254. Humidity Ratio. We shall find it convenient to base calculations


on a unit mass of dry gas because the mass of the vapor, and therefore
(dg.),
of the mixture, often varies; but the mass of dry gas remains constant.
Thus, a convenient term is the mass of vapor per pound of dry gas, called the
humidity ratio w in connection with air-steam mixtures; w = w v /wa (The .

name may be extended to include other mixtures.) This value can be


found from
p„ lb. vapor/cu. ft. _ lb. v.
{fi ' )
lb. dg./cu. ft. lb. dg.’

in which p„ = is the partial pressure of the dry gas;


pJ(R„Tf), where pa
pa = pm — density of saturated vapor is p„ = l/v v in which v v is
Pv The ,

taken from vapor tables. The density of superheated vapor can be obtained
from equation (86) when the relative humidity is known; p»i = </>p vtj, where
Pvd is the density of saturated vapor at the actual temperature of the super-
heated vapor, h and T, Fig. 166. Since p = 1/v, we also have a = vjv*.
When the vapor is in a state approximating an ideal gas, we may use
P = p/(RT) in equation (87) to get

/_% _ Pv _ (Pv) (RpTf) pA = PA jb. v-


V J "
Pa (R,TV ) (pa ) pA RAm ~ p/) lb. dg.

where for internal equilibrium, T0 = Tv and where the partial pressure of


,

the gas is equal to the pressure of the mixture minus the pressure of the
vapor, p a = p m - pv Equation (n) may be applied to gas-vapor mixtures
.
802 GASEOUS MIXTURES [Ch 15
in any state, provided that the partial pressure of the vapor is low Another
useful form of equation (n) is obtained bj solving for
p*,

(o)
P^uR,
P.
Ra + uR,'

where the units of p, and p m are the same


Equation (o) is applied to a par-
d or c, Pig 16G If equations (n) and (o) are applied to
ticular state, as 1 or
an steam mixture, we use Ra = Rt = 53 3 and R. =* 154 %8 = 85 7 for
air
HjO, whose molecular weight is approximately 18, and find

53 3p, _ 0 G22p, pmu


(P) " 85 7p.
“ P’
__
“ 0CjJ+^ i

(AIR STEAM MIXTURE ONLY]

where w is pounds of vapor per pound of dry gas The percentage by vol-
ume, or the mol fraction of vapor in the mixture, is [w lb v /(AT. lb /mol)
= mol of vapor, 1 lb da /(A/, lb /mol) = mol of air]

R* g/18 also B, - Si
w/18 + 1/28 97’ Pm

256 Wet-Bulb Temperature The relati\e humidity of atmospheric


air,that is the relative amount of steam in the air, is easily found via an
experimental determination of a wet-bulb temperature tu The actual
temperature of the air is called the dry-bulb temperature tj, because it is
of course read from a thermometer whose bulb is drj The instrument
used for the purpose of determining is called a psychromeler, of which

there are several types The commonest one is a sling ps> chrometer, which
consists of two thermometers attached to a handle so that thej may be
easily whirled about the axis of the handle One thermometer has on its
bulb a wet gauze and is called a wet-bulb thermometer The temperature
indicated by the wet bulb thermometer depends upon the rate of evapo-
ration of moisture from the wet gauze The heat necessary to evaporate
this moisture is supplied by the air passing over the gauze (see § 256)
Consequently, this air (and water) are cooled below the atmospheric
temperature, the amount by which it is cooled, id — i** being called the
wet-bulb degression
The rate of evaporation of the water depends in part on the amount of
steam already in the air If the air is saturated, none of the w ater on the

gauze evaporates because the air is already a saturated mixture with respect
to the steam, the wet bulb and drj bulb temperatures are the same The
less “moisture” (steam) carried by the air, the more that
must be evapo-
rated in order to result m
saturation, hence, the rate of evaporation being
§ 256] ADIABATIC STATURATION PROCESS 803

greater, the wet-bulb temperature will be lower. The higher the relative
humidity, the slower the rate of evaporation and the higher the wet-bulb
temperature. With the wet-bulb and dry-bulb temperatures, we may
enter a psychromelric chart, Fig. 107 (see Problems on Thermodynamics
for aworking chart), and find the amount of steam in the air, the relative

Dry-Bulb Temperature, *F

Form of Psychromelric Chari. This chart may be entered with various


Fig. 167.
data. Suppose the wet- and dry-bulb temperatures are known; find the wet-bulb
temperature at b, the dry-bulb at a, and follow constant temperature lines until they meet
at state 1. Move to right or left to c and read values of u and p,\ estimate the volume
\

between the lines d and e; estimate between lines e and/; follow constant wet bulb
<f>

to g and read value of total heat.

humidity, and other useful information. In the absence of a psychrometric


chart, or if the atmospheric pressure is significantly different from standard
(14.G9G psia), the wet- and dry-bulb temperatures may be used in the
equations for an adiabatic saturation process (§ 25G), an application of usual
principles of thermodynamics. This process is an irreversible one, path

Fig. 168. Adiabatic Saturation on Ts Plane.

dw, Fig. 1GS. Note that the wet-bulb temperature is above the dew
point c and that the amount of moisture in the air is continuously increasing
from d to w.

266. Adiabatic Saturation Process. Processes which are approximately


adiabatic saturation and similar steady flow processes are often encountered
804 GASEOUS MIXTURES [Ch 15
in practice The manner
in which the adiabatic saturation process occurs
is suggested by Fig 169(a) For convenience the basis of computations is
one pound of dry air If the water is not initially at temperature tw it
becomes U* when steady state has been attained If the beginning the m
water is hotter than U, it cools off by supplying some energy to evaporate
the water If the water is colder than lv it is warmed by energy transferred
from the air At the steady state condition the energy used to evaporate
the water comes from the entering air (let the make up water be added at
temperature <„)

(q) Energy given up by entering I _ f latent heat of the evaporated


air in cooling from U to t a \ \ water at
(adiabatic system = air and water AK =» 0]

It is clear from this word equation why the process is called adiabatic
However, for purposes of an energy diagram, we may take the boundaries
of the system as outlined with heavy dashes in Fig 169(a), equating energy

Evaporation* Water to W ck (Evapor&t on)


(a) Adiabatic Saturation (b) Wet Bulb Thermometer
Fig 169 The vessel in (a) is long enough that ample time is allowed for the air end
water to reach equilibrium with respect to one another (which is at the adiabatic satura
tion temperature t„) The process could be helped along by spraying the water into the
air stream

entering to energy leaving the system (no change of stored energy) Each
point of view results in the same equation From the energy diagram of
Fig 169, we have
(r) had + Wdh,d + ( w u>
— Ud)h/W = ha„ -f- Btu/lb da ,

where Ki — enthalpy of da at dry bulb tj


I lb

haw = enthalpy of 1 lb da at wet bulb t„

h,d= enthalpy of 1 lb steam as it exists in the atmosphere at td (see § 257)


haw= enthalpy of 1 lb saturated steam at wet 6uf6
k/w= enthalpy of 1 lb saturated water at wet bulb
Uw ~ humidity ratio at the wet bulb f. (saturated air) u„ = pjp* where
pa = 1/p, v, being taken at f„
b>d — humidity ratio of the original air at td ai* — ud = mass of HO
2

evaporated lb v/lb da

We note in equation (r) that haw — h/w hj a —


h/l!a the change of — =
enthalpy (latent heat) during evaporation at l„ and we use Wwh,,w
for
.

§ So 7] ENTHALPY OF SUPERHEATED STEAM AT LOW PRESSURE 305


— h r ~) Then, solving for the humidity ratio a:* of the original
atmospheric air, we get (see Fig. 16S for the state points of the steam)

(SS) cjj = '

% — lb. v./lb. da.,

where — h.-> = — 0.24(f^ — t~). the change in specific enthalpy for the
dry air. Knowing the total pressure p., and the humidity ratio aj at anv
state, the corresponding partial pressure of the steam may be computed
from equation (o) or (p). Observe that the conditions under which equa-
tion (SS) applies are those which define the adiabatic saturation process: to wit,
the process steady flow, the total pressure remains (virtually) constant,
is

the heat transferred is zero, and the temperature of the make-up water is

the wet-bulb temperature t...


The process in connection with a wet-bulb thermometer is seen to be much
the same as the adiabatic saturation process [Fig. 169(b)]. The wet-bulb
reading is affected by radiation to the thermometer, by the velocity of air
relative to the thermometer, by the design of the instrument, and by the
rate of diffusion of the evaporated water into the air stream (131). For-
tunately, under ordinary circumstances, the heat absorbed by radiation
from the surroundings is nearly compensated by providing a rapid move-
ment of air over the bulb (convection), so that the wet-bulb temperature is a
good approximation of the adiabatic saturation temperature. Thus the
adiabatic saturation process is nearly one of constant iret-bulb temperature.
The psychrometrie chart provides a quick answer to the same problem to
which equation (SS) would be applied. But since the engineer is interested
in learning principles, he will master the use of equation (SS), which is the
basic energv equation for the construction of psychrometrie charts. W. H.
Carrier developed from equation (SS) an expression for computing the partial
pressure of the steam in atmospheric air, after making some simplifying
assumptions. Such a derivation welds the following equation,

, x
= (P™ Psfi)(td f'r) r .
Icabejees equation-],
(s) p,
2S30 - 1 44 1

which is different from Carrier’s original equation only in the constants.*

257. Enthalpy of Superheated Steam at Low Pressure. Since steam


tables do not gives enthalpies of superheated steam at pressures below 1 psia,
another means of obtaining h.j in equation (SS) is needed. Having observed
that the steam in the air may be treated as an ideal gas with little error, we
recall that the enthalpy of an ideal gas is a function of temperature only

(§ 59). Thus, the easiest way to determine the enthalpy of superheated


steam at pressures below 1 psia is to use the enthalpy of saturated steam at
* Ecraation (s) is derived in some detail in the earlier editions reference { 158 ).
)

S06 GASEOUS MIXTURES [Ch IB

the same temperature In Fig 170, for instance,

(t) i
= h ab Btu/lb v

This same principle may be applied to other vapors when the vapor pressure
is low An equation which gives satisfactory answers for the enthalpy of
steam in atmospheric air is

(u) h, = 1061 + 0 444^ Btu/lb v.,

where lj is the dry bulb temperature, °F

258 Example The dry and wet bulb temperatures of air are found to be
td «= 83°F and t„ « C8°F The barometer is p„ = 29 4 in Hg Determine (a) the
humidity ratio, (b) the partial pressure of the vapor (from theory and from Carrier s
equation) (c) the relative humidity, (d) the dew point, (e) the density of the air,
(f) the density of dry air at the same pm and
V 83‘F tj (Refer to Fig 170 steam table values are
f,„-68°F from Keenan and Keyes
60 LUTI 0 V (a) Humidity Ratio Let the
Dew Point Temperature state be represented by I, Fig 170 First
obtain talues needed for the solution of
equation (88) From the steam tables at
t w = 68°F, h/ta — 1055 5 Btu/lb. and
— 3604 Btu/lb At 83°F,

Fig 170 h, = 1097 9 Btu/lb,

which is closely the value of h , i To get


use h, i
=
find p w and1097 9
At = 68°F, the partial pressure of the steam is p^ - 0 339 psia (from steam
tables) and v, = 925 9 cu ft /lb or p.„ = 1/925 9 lb /cu ft The pressure of the
,

dry air at saturation, state to, Fig 170 is (p„ *= (0 491) (29 4) = 14 45 psia)

P~ = pn ~ p.» = H 45 - 0 339 = 14 1 1 1 psia

P™

which is the humidity ratio after adiabatic saturation state w Fig 170 Using
these various \ ulues in equation (88) we get the humidity ratio of the original air
as

(0 015)0055 5) - (0 24) (83 - 68)


v da,
0 01151b /lb
1097 9 - 36 04
or (0 0115) (7000 grams/lb ) - 80 5 grams v /lb da

(b) Partial Pressure of Vapor According to equation (p), the vapor pressure

pv i Fig 170 is

o), p„ 0 01 15) (14 45)


— _ ( = 0 262 psia
p.i
0 622 + «i 0 622 + 00115
§ 259] ENTHALPY AND TOTAL HEAT OF AN AIR-STEAM MIXTURE 307
From Carrier’s equation (s), we have

Pcl
_ T (P" If)

2830 - 1.441,-

_ U d '^
'
(14.11)05)
°‘ 2615 psia.
.

2830 - (1.44) (68)

(c) Relative Humidity. As defined by equation (S6), the relative humidity is

0.262
* “ ^- _
Pvl
r .
,
~ 46-9
Ptb 0.5588

where p = 0.558S is the saturation pressure corresponding to the dry bulb tem-
perature h - S3°F.
(d) Dew Point. At c, the vapor pressure is p TC = p tl = 0.262 psia. The satura-
tion temperature corresponding to this pressure is found in the steam tables as
60.6°F, the dew point.
(e) Density of Mixture. The density of the mixture is the sum of the densities
of the constituents; p„ = p„-f- p r The pressure and density of the dry air in
-

state 1. Fig. 170, are

Pa i — Vn — Pai = 14.45 — 0.262 = 14.188 psia.

Pal ~
Pal
~ (14.188K144) _
- .
°-°' 06 CU ’ ft/lb -
RaTa (53.3) (543)

From equation (86), we get the density of the vapor at 1 as

Ph = d>Ptb
Ugh
==? =
Oil. ±
0.00081 Ib./cu. ft.,

where Ly = 577.4 cu. ft./lb. is the volume of saturated vapor at state b, Fig. 170.

p„ = Pai + Pai = 0.0706 + 0.00081 = 0.07141 lb./cu. ft.

(f) Density of Dry Air. For dry air at 29.4 in. Hg = 14.45 psia and 83°F, the
density is

Pa_ _ (14.45) (144)


P“ “ _ ~ 0.0719 lb./cu. ft.
RaTa (53.3) (543)

It is interesting to note that dry air at a particular temperature and pressure is

heavier than atmospheric air at the same temperature and pressure, 0.0719 >
0.07141. For practice, the student should check the foregoing answers by a psy-
chrometric chart, noting that the chart is constructed for a p m = 14.696 psia. It is

possible for chart values to be corrected for other than standard atmospheric
pressure (IS, 127), but the innate inaccuracies in reading a chart are often greater
than those due to an atmospheric pressure which differs little from standard.

259. Enthalpy and Total Heat of an Air-Steam Mixture. It is the prac-


computations to reckon the enthalpy of dry air from
tice in air-conditioning
0°F and the enthalpy of steam from the usual 32°F datum. Thus, the
enthalpy of a gas-vapor mixture in general may be taken as

(89) hm = ha + oih = v cp t + oj h r Btu/lb. dg.,


S08 GASEOUS MIXTURES [Ch 15
where the vapor component Msmeasured from its usual datum, cp - 0 24Btu
per lb for air, t is in degrees Fahrenheit, « is the humidity ratio
Tables are available (127) which afford the easiest means of determining
the properties of mixtures of air-steam at p„ = 14 69G psia If the air-
steam mixture is not saturated, a new term is needed, degree of saturation
n, which is defined as the humidity ratio of the actual air divided by the
humidity ratio of saturated air at the
same dry bulb temperature and total pres-
sure, or in terms of the state points on
Fig 170,
Dew Point Temperature
(V) ft
— 6>i/w6

Together w ith other properties, the tables


gn e the enthalpy of dry air and a property
designated by h„, Btu per lb da , which is
Fig 170 Repeated the difference between the enthalpy of
saturated air at a particular temperature
minus the specific enthalpy of dry air at the same temperature and total
pressure,

ha, = h ut — hirj = ajigb Btu per lb da

Wehave«ife„i = uth, i(«j/wi) =


ph a, when h tl = A4& the subscripts
ujigin = ,

being as defined in Fig 170 This approximation is good at air tempera-


tures U < 150°F (127), so that (h a = specific enthalpy of dry air)

(w) k„ = ha + fih a , Btu/lb da

There another property which


is is commonly used in air conditioning

called total heat Til, defined by*

(x) TH = hav, + — 0 24t„ + Btu/Lb da ,

where h, w is the enthalpj of saturated steam and


the humidity ratio, is

both at the wet-bulb temperature t» Actually, equation (1) is the enthalpy


of a saturated mixture at the wet-bulb temperature, h m„ = haw ,A fW +
Notice that since total heat is a function of the uet-bulb temperature only, it
can be included on a psychrometric chart without complication (Fig 1C7)
The total heat is nearly equal to the enthalpy of a mixture, so that where fine
accuracy is not important, as in some air-conditioning problems, it is
accurate enough to use Q « A Til instead of Q — AA in steady flow proc-
esses where W K
= 0 and A = 0 Observe that the right-hand side of
equation (r), p 304, is the so-called total heat, that the enthalpj of the given
mixture on the left-hand side is A*, = had +
Udh v i, w here the subscripts refer
to Fig 169 Thus, total heat maj be corrected to enthalpy by subtracting

•The language is somewhat confused here This property is sometimes incorrectly


exiled enthalpy
§ 262] MIXTURES OTHER THAN AIR-STEAM S09

the term (&v — uglify,;

(y) hm = TH — (uw — o>d )h/w = haw + oiwKw — (o>w — ud)hfW Btu/lb. da.

Compare equations (r), (89), and (x).

260. Example. For the mixture in the example of § 258 (t d = 83°F, l w = 68°F),
determine the enthalpy by equation (89) and by the tables in the ASHAE Guide
(127), and also compute the total heat for comparison.
solution. Using equation (89), we get

hm = (0.24) (83) + (0.0115)(1097.9) = 32.55 Btu/lb. da.


hm = 32.55/1.0115 = 32.2 Btu/lb. mixture,

where 1097.9 = h„ ~ h rt (Fig. 170) is taken from the steam tables and is the enthalpy
of saturated steam at the dry-bulb temperature; h ti = k tb .

From the tables in the ASHVE Guide (127) for 83°F, we get co b = 0.02471; cor-
responding to which fi
— a>i/«! = 0.0115/0.02471 = 0.466; and
,

hm = ha + uha, = 19.942 + (0.466)(27.1) = 32.57 Btu/lb. da.,

where the values of h a and h„ have been taken from the tables. This answer is a
close check of the preceding answer.
The total heat of the mixture is

TH = (0.24) (68) + (0.015)(1091.5) = 32.69 Btu/lb. da.,

where 1091.5 *= h„ at t„ = 68°F and «„ = 0.015 from § 258. Check this answer
against a psychrometric chart and note that it is somewhat different from the
computed values of h.

261 Internal Energy and Entropy. The internal energy of a mixture is


.

the sum of the internal energies of the constituents, and as in the case of
enthalpy, it is not necessary to measure the individual internal energies from
the same datum; thus,

(z) um = u0 + om v = cv t + w (h„ - Btu/lb. dg.,

where u a for the dry gas is measured above 0°F and the vapor properties are
found as usual from vapor tables.
The change of entropy for a mixture is the sum of the changes for dry gas
and vapor considered separately. Compute these changes in accordance
with principles already established.

Mixtures Other than Air-Steam. The foregoing principles apply


262 .

to any gas-vapor mixture, but vapor properties are not always at hand.
Chemistry handbooks will contain enough data for certain purposes, data
such as latent heats, specific heats, saturation pressures, and temperatures.
SIO GASEOUS MIXTURES [CA 15
The cun es of Fig 171 are typical of a\ ailable data The use of this infor-
mation is illustrated by the following Example

Fig 171 Pressure! and Temperatures oj Saturated Vapors To use Enter chart
for, say, a known temperature of 100°F Move toward the right along a constant-
temperature Une until the desired curve is reached. Move down to read the corre-
sponding saturated-vapor pressure Following the dotted line for octane, find a ptes
sure of 0 61 psia.

253 Example: Fuel-Air Mixture A fuel air mixture in an intake manifold con-
(C,H, a ) and air, is at 122°F and a total pressure of p„ = 12 psia
sisting of octane
Determine the pounds of air per pound of fuel (a) if the mixture is saturated with
respect to the octane, (b) if the relative humiditj of the octane is 21%, but fiu=t
determine the dew point at this humiditj
solution (a) From Fig 171 the partial pressure of the saturated vapor of

octane at 122°F is found to be p, t = 0 95 psia See Fig 172 The molecular


weight of octane is 1 1 4 14 from which
its gas constant ma\ be taken as

1545
RKtm = = 13 5
114 14

From the characteristic equation of an


ideal ga« we find the approximate den-
«it\ of octane vapor to be

. pi (0 95)(144)(1)
Pw '
RT (13 5)(5S2)
= 0 0174 lb /cu ft

The partial pressure of the air at «tate 3 is 12 — 0 95 = 11 05 psia, whence the


densitj of the air is

pY = (11 0o)(144)(l)
= 0 0513 lb /cu ft
P« i
RT (53 3)(oS2)
§ 264] EXAMPLE: CONSTANT TOTAL PRESSURE (STEADY FLOW)
Then, the mass of octane vapor per pound of air (or the humidity
ratio) is

0.0174
u3 = Pra = 0.339 lb. v./lb. da.
Pal 0.0513

This ratio is equivalent to 1/0.339 = 2.95 lb. of air/lb. of fuel.


Such a mixture is
not likely in practice because there is not enough air to complete the combustion
of the fuel.
(b) Since the dew point is that temperature at which the vapor becomes saturated
during a constant pressure cooling, we find first the partial pressure of the octane
from

$ = = 0.21 _ -Erl.
Pel 0.95’

or pt i = 0.2 psia (Fig. 172). From Fig. 171, the corresponding saturation tem-
perature is U = 68°F, the dew point.
The density of saturated vapor at 122T has been found as p e3 = 0.0174 Ib./cu. ft.
Using equation (86), — pd/pcz, we have tf>

Pc i = 4>Pd = (0.21) (0.017 4) = 0.00363 lb./cu. ft.

The partial pressure of the air with the vapor in condition 1 is 12 — 0.2 = 11.8 psia
whence
(11.8) (144)
= 0.0548 lb./cu. ft.
(53.3) (582)

J_ _ pa i _ 0.054S
15.1 lb. air/lb. v. (fuel).
oil Pci 0.00363

This ratio is typical of mixtures found in internal combustion engines.

264. Example Constant Total Pressure (Steady Flow).


: An atmospheric air mix-
ture in state 1 has the following properties: p„ = 29.92 in. Hg, tj = 100°F, and

oil= 0.0295 lb. v./lb. da. is cooled to GOT, state 2, Fig. 173, and then
This air
heated to 85°F, state 3. Determine the relative humidity of the original mix-
(a)
ture. (b) How much moisture is deposited during the cooling to GOT? (c) How-

much heat is removed during cooling? (d) For a flow of 100,000 cfm, how many
tons of refrigeration are ideally required? (e) What is the volume of the original

100,000 cfm after cooling? (f) 110131 is the relative humidity of the air in state 3

(Fig. 173)?

note. events of this problem are fairly typical of the events in summer air
The
conditioning of habitations. Since the various answers may be obtained writli the
aid of a psychrometric chart, the reader should use the chart and check all results.
solution, (a) To find the relative humidity, first find the partial pressure p r]
of the vapor from equation (p);

PcI
_ OllPr, (0.0295)(14.7)
0.666 psia.
0.622 + on 0.622 + 0.0295
i * )

GASEOUS MIXTURES [CA 15


From the steam tables, we find p,* = 0 9492 p*i3 for 100°F and then

= 70.2%
P 0 9492

(b) The condensed HjO


is obtained bj
subtracting w, from u, To get an idea
of where state 2 for the H,0 is, we enter
the “team tables and find the dew point c
(Fig 173) corresponding to p.i = 0 666 as
f, = S8 4°F Thus, the air has been cooled
Fig 173 below the dew point and HjO has been con-
densed At 60°F, p,, = 0 2563 psia is
taken from the tables, then from equation (p), we get

to, -— =
“ 0 622p,i (0 622) (0 2563) _
~ OOUOcIb v/Ib _ .
dn
Pm - P.i 14? -0M3
The condensation is therefore

w, - to, - 0 0295 - 0 01105 = 001845 lb \ /lb da

(For the 100 000 cfm, this condensation amounts to

(6770)(0 01845)
14.85 gpm,
8 4 lb /gal H,0
which isequivalent to 892 gal /hr See part (d) for computation of 6770 lb /mm
(c) The heat transferred at constant pressure or a steadj flow process where m
IF = 0 and AA. = 0 is 0 = AA The enthalpj of the vapor in the original air is
h.i ~ h# - 1105 2 Btu/lb \ at 100°F, or

to,ft„ = (0 0295)(1105 2) = 32 6 Btu/lb da

At state 2, the enthalpj of the H,0 should account for the condensed steam, that is,

A,, = to, A,, + (w, — u,)A/,


= (0 01 105) (10SS) + (0 01845) (2S 06)
= 12 -f 0 518 = 12 518 Btu/lb da

where h t , and A/, correspond to a temperature of 1, = 60°F Now usmg equation


(89), we find

Q = Ah,, - c,(ft ti) + A.i


-
— A, —
= 0 24(60 - 100) + 12 518 - 32 6 = -29 7 Btu/lb da

The heat remo\ ed per pound of original mixture is 29 7/1 0295 = 28 85 Btu/lb
mixture This \alue of <3 should be closely approximated by A 777
(d) If there are 100,000 cfm of atmospheric air, there are also 100 000 cfm of dn
air and 100,000 cfm of i apor The mass of 100 000 cfm of drj air flowing is

(147 - 0 666)(144)(100 000)


^‘ =2-1^**
R.T (53 3) (560)
= 67701b/min
t
8I4 GASEOUS MIXTURES [Ch 16

For example, a cubic inch of crushed and prepared condition


silica gel in its
exposes some 50,000 sq ft (127) While silica gel and activated alumina
have a particular affinity for HjO, activated carbon prefers organic vapors
and is used, for example, in refrigerators to reduce objectionable vapors
with odors After they have become saturated, both absorbents and
adsorbents are reactnaled by heating them
16

THE TRANSFER OF HEAT

We have repeatedly used transferred heat Q and


266. Introduction.
some heat exchangers without discussing the details of the phe-
described
nomena by which heat is transferred. The purpose of this chapter is to
give the reader a general idea of the nature of the problems and to provide
an introductory background for further study. The developments in this
are too extensive to be covered fully in a book on thermodynamics.
field
Heat exchanger is a name of general connotation applied to any device
which effects a transfer of heat from one substance to another. One way in
which this transfer of heat is brought about is by the irreversible mixing of
cold and hot fluids in some exchangers, such as open feedwater heaters, jet
condensers, desuperheaters, wherein an equilibrium condition is estimated
by equating the energy lost by one fluid to the gain in energy of the other
fluid. In other cases (Figs. 174, 175), heat is transferred from one fluid on
one side of a wall or partition to and through the wall to another fluid on
the other side. Numerous heat exchangers already mentioned come to
mind; as examples, we have: steam boilers, surface condensers, evaporators,
closed feedwater heaters, and the automobile radiator. Those factors
which determine the coefficient of heat transfer between a fluid and the
adjacent surface will not be discussed in detail; and we shall have little to
say about radiation. The idea is to obtain a general understanding of the
mechanisms of heat transfer.

267. Conduction. Methods of heat transfer are defined in § 19, which


should be read again now. We recall that heat is conducted from one part
of a system to another because faster moving atoms or electrons or molecules
in the hotter part induce by impacts an increased activity of adjacent atoms
or electrons or molecules (increase their kinetic energy) thus heal flows
;

(there is a heat flux) from the hotter to the colder parts.


815
),

3*° THE TRANSFER OF HEAT [Ch 16

Heat may be conducted at uneven rates the rate increasing or decreasing,


that is, the temperature difference in the path of heat flow may be increasing
or decreasing This condition is an unsteady state, as during the warm up
period of an engine, and it must be analyzed \ia infinitesimal heat flows
dQ/dr, where dr is the duration of time during which heat dQ passes a cer-
tain section However, this study will be confined to systems in a steady

Courtesy Brown Fmlube Co Elyria 0


Fig 174 Heat Exchanger, Extended Surface Used for heating viscous oils—
reducing the pumping work Oil enters from the oil tank at the open left end Steam
circulates through the tubes, two-pass flow through tubes, single pass flow through the
shell See Fig 176(b)

state, wherein each point of the system remains at a constant temperature


and heat transfer is constant at a rate of Q energy units per unit of tune
Also it is assumed that the flow of heat is unidirectional

268 Fourier’s Law Tor steady state, unidirectional flow, Fourier’s


equation (1822) gives the heat conduction as

(00) Q - -kA Btu/hr,


(J^
where Q Btu per hr is the heat conducted across a surface of A sq ft

through a w all thickness of dL in and with a temperature drop of di°F


,

through the distance dL The ratio dl/dL is called the temperature gradient

Courtesy Foster Wheeler Corp !feu> York


,
Courtesy Brown r intube Co Elyria 0
(a) Circumferential extended surface (b) Longitudinal extended surface

Fig 17S Extended Surface Elements

along the path The negative sign in equation (90) is used because the
temperature decreases in the direction of heat flow (that is / dt = h
- h is
convenience)
a negative number and the negative sign makes Q positive for
The symbol k represents the thermal conductivity which is the amount
*

unit area (sq ft


of heat (Btu per hr ) transmitted in unit time across
Familiarity With the subject will
* Do not confuse the symbol k with k c,/c.
context
make it easy to attach the correct meaning to the symbol by its
§ 289\ VARIATION OF THERMAL CONDUCTIVITY 317
through unit thickness (both in. and ft. are used) for unit temperature
change (IT 7 ). Any set of units for k. as defined by equation
(90), may
be used, but the foregoing units are the usual ones. Values of k given and
used in this chapter are for a 1-in. thickness; or

u Btu-in.
(a)
ft 5 -°F-hr."

If the thickness L is in feet, then

(b) ( Btu-in. ^
)( ^ ^ Btu-ft.
e
\ft'- F-hr.y'Vl2 in.) * ft--°F-hr.”

It will soon be evident that extraordinary care must be exercised in the


matter of units in heat transfer, and because units are diverse in the litera-
ture, the reader is cautioned to check thoroughly the units of all factors
entering into any equation.

269. Variation of Thermal Conductivity. Thermal conductivity varies


widely, just as electrical conductivity does. Study the values in TableXII.
Xot only is there great variation between materials, but particular materials
may also have widely different conductivities in different states. For
example, k for aluminum at 212°F — 672 'R is about 1440 Btu-in. per
c C
ft : -hr- F, but at about 1S R it goes up to about 35,000 Btu-in. per ft : -hr-°F

(118). However, the conductivity of solids varies so little with the more
usual temperatures that a single average or typical value may be used with
small error. In most applications, the thermal resistivity, which is the
reciprocal of the conductivity, is so small for metals as compared with other
resistances to heat flow that a small variation of k from its true value has
little effect on the over-all conductance or resistance. If the relevant
temperatures are extreme, high or low, it is advisable to look into more
detailed sources for test values. As we shall see, the largest resistance is in
the fluid film adjacent to the solid. King (11)9) makes the following general-
izations concerning conductivities of solids:

1. The conductivities of allhomogeneous solid materials are relatively high;


practically all good (heat) insulators are porous, cellular, fibrous, or laminated
materials.
2. In general, the conductivity increases with the density and elasticity.

3. Kith rare exceptions, the conductivity of insulating materials increases very


materially with the temperature.
4. The absorption of moisture greatly impairs the insulating value of porous
materials.

The and gases are more sensitive to temperature


conductivities of liquids
changes. Moreover, the difficulty of eliminating convection currents in
tests on liquids and gases has somewhat complicated the determination of
conductivities. For relatively small temperature variations, the con-
v ,

818 THE TRANSFER OF HEAT [Ch 16

ductmty of solids, liquids, and gases may


be assumed to vary lincarly-with
the temperature For this assumption, the ft equation (90) may be m
taken as the arithmetic average for the temperature range involved (or the
value of ft for the mean temperature), and the integration made with ft
constant If the vanation is not linear, and a plot of ft values is available,
the mean value between tivo temperatures may be estimated from the curve
(as suggested for specific heats in connection with I lg 19) Lacking other
data, interpolate between the values given in Table XII for ft at the average
temperature of the conducting body.

270. Conduction through a Plane Wall. After a steady state of unidirec-


tional heat flow has been reached m
a single homogeneous material whose
thermal conductivity ft is constant, the temperature gradient dt/dL for a

plane wall is constant (a straight line not true when ft varies with tempera-
_ .
ture) Accordingly, integration of
equation (90) gives
L
Q jQ dL = -k A dt,

A (4 - 4)
ft
(C) Q Btu/hr
L

where 4 and 4 are the surface tem


peraturcs of a partition such as A, Fig
17G and ft is the average value for
the given conditions We sec that the
rate of flow of heat depends (1) directly
upon the temperature difference (poten-
tial) between the two surfaces of the

wall, (2) directly upon the area of sur


face A through which transmission
F‘? Temperature Gradients
7 Composite Wall occurs, (3) indirectly upon the thick-
ness L of the wall, and (4) directly
upon the value of the thermal conductivity ft Applying equation (c) to a
composite wall made up of three homogeneous materials A, B, C, Fig 176,

we have
fti A (4 - 4) ft 2 A(4 ~ 4) ft, A (4 - 4)
(d) Qa = Qb Qc = Btu/hr
L t

Solving for the temperature difference from each of these equations, and
noting that Qa = Qb ~ Qc = Q Btu per hr for steady-state flow, we find

QLi —4 QU
4 = and 4
.

(e) x~. '


ftj.-l

Next, equating the sum o£ the left-hand terms to the sum of the right hand
)

Table XII. CONDUCTIVITIES


2
The units of k are Btu-in./ft -hr-°F; at atmospheric pressure; k for a solid changes
little with pressures below 200 psi, but significant changes of k for liquids and vapors
may accompany pressure changes. The values are from various sources, here
selected largely from McAdams (187). Straight-line interpolations are permissible
between the temperatures given. Nearby extrapolations may give satisfactory
results.

Temp. Temp.
Material °F k Material °F h

Solids Liquids ( continued


Aluminum 32 1400 Kerosene 68 1.03
392 1490 167 0.97
Aluminum piston alloy 0-400 1290* Petroleum oil, average 68 1.0
Asbestos, 29 lb./ft. s -200 0.865 Sodium 212 590
Asbestos, corrugated, 410 550
4 plies/in. 300 0.828 Sulfur dioxide 5 1.53
Bearing metal, white 68 164 68 1.33
Brickwork, low density 68 5 Water 32 4.1
Cost iron, grey 0-400 360' 200 4.7
Copper, pure 32 2690’ 300 4.75
212 2616 620 3.3
Concrete, 1-4 dry 5.4
Cork board 86 0.3 Gases
Glass window 3 6-7.4 Air -148 0.109
Gold 64 2028 32 0 168
Graphite 32 1165 572 0.312
392 910 Ammonia -58 0.107
32 0.151
Gypsum 68 3
Gypsum 3.3 212 0.23
plaster
Magnesia (85%) 100-300 0.43* Carbon dioxide -58 0 077
32 0 101
Mineral wool (glass
86 0.27 212 0 154
and rock wool)
68 242 Freon F12 32 0.057
Monel
212 0.096
Plaster on wood lath,
Hydrogen 32 1 06
%-in. total thickness 2.5
0-400 312* 572 2.04
Steel
70 0.34 Nitrogen 32 0.167
Wallboard, insulating
86 0.32 572 0 306
Wood, balsa
1.44 Oxygen 32 0.17
Oak, maple 59
1.05 212 0.226
White pine 59
Steam 212 0 163
932 0 394
Liquids
3.13 Sulfur dioxide 32 0.06
Ammonia 68
0.0827
3.48 212
140

* Average for temperature range given.

819
$

Table Xffl. CONDUCTANCES AND TRANSMITTANCES


The units of k/L and U
are Btu per hr-ft*-°F. These values are intended as
representative, suggesting the order of magnitude in the various situations They
are not to be u«ed in actual design unless it is known that thej apply. Since the
values of U
are not particularized, the reference area is not meaningful However,
if these values are u«ed for problem work, let the reference area be the internal tube
or pipe area m
such cases (a) From McAdams {187). (b) From ASHAE {127)

k
Construction and Materials U
i

Air space, *4 m
or more in width (b) 1 10
Air space, 5* m
or more in width, bounded by aluminum foil (b) 0 46
Ammonia condenser, 2z3-in double pipe, water inside at v = 6 fps
Nil* in annular space, — 3 5°F, clean (a) 320
Asphalt shingles (b) 6 50

Brick wall, 8 in thick, plaster inside (b) 0 46


Brick veneer, frame wall with wood sheathing, \ in plaster on
gypsum lath (b) 0 27
Brick veneer as above, plus 2 in mineral wool insulation (b) 0 097
Concrete blocks, 8 in , hollow gravel aggregate (b) 1 00

Feedwater heaters, closed, steam condensers, free convection (a) 50-200


" forced conv ection
, (a) 150-800
Insulating board, in thick (b) 0 66
Heat exchanger air in tube, condensing steam outside tubes (on
outside surface area) (a) 8
Heat exchanger, cooling oil with water in tubes 50

Steam condensing, to air, free conv ection (a) 1-2


" , forced convection (a) 2 10
Steam condensing, to boiling water, free convection (a) 300-800
Steam condensing, to liquid oil, free convection (a) 10-30
11
forced convection (a) 20-60
,

Superheaters, steam, free convection (a) 1 6-2


“ , forced convection (a)
Water to gas and liquid to gas (hot water radiators, air coolers,
economizers steam boilers), free convection (a) 1-3
“ , forced convection (a)
Water to water, free conv ection (a) 25-60
“ , forced convection (a) 150-300

32G
§ 271 ) FILM COEFFICIENT 321

terms of these equations, we get

(f)

or
A (U - Id) _ A At
(g) Q = U r
Li/k\ + L /A +
2 2 Lt/k 3 X(L/k)
'

for the composite wall in which the temperature drop from surface to surface
is = la - L for the wall of Fig. 176; observe that in this usage, the
At (At
At is the first temperature minus the second, counting in the direction in
which heat flows). It is readily seen that if another section were added to
the wall, the only change necessary in equation (g) would be the addition of
another term L A /k t to the sum of the other L/k’s and the interpretation of
t d as being the temperature of the final surface. This sum of L/k values is
simply represented by 2 L/k, as shown.
For any one section of the wall (Fig.
176), C = kA/L in equation (d) is
called the conductance, which is seen
to include the effect of size and shape
of the conducting body. The conduct-
ance is the rate of heat transfer per
unit temperature difference, Q/At.
The reciprocal of the conductance is
the thermal resistance, R — L/(kA)
= At/Q. The conductance per unit
area, or the unit conductance k/L, is
often given for particular bodies. See
Table XIII.) The unit resistance (per
unit area) is L/k. For the composite
plane wall (Fig. 176), the conductance
from surface to surface is A/H(L/k)
[equation (g)], the unit conductance l/Z(L/k), and the resistance is
i.

MWk)/A.
Film Coefficient. On each side of the composite wall of Fig. 176 is
271.
a the nature of which is of no concern at the moment. On the hot
fluid,
side of the wall, the fluid is hotter than the surface at some temperature
<1 > t a On the cold side, the fluid is colder than the surface, its tem-
.

perature being h < td. Thus, through thin films of the fluids adjacent to
the surfaces, there are temperature drops ti — ta and t d — t 2 The unit rate
.

of heat flow through these films is called the film coefficient (and other names
such as film, conductance, surface conductance ) and is represented by h, whose
units in this book are Btu per hr-ft 2 -°F ; that is, h is the rate of heat flow
(Btu in 1 hr.) through an area of 1 sq. ft. when the temperature potential
across the film is 1°F. Its magnitude depends on variables which are too
S22 THE TRANSFER OF HEAT [Ch 16
manj' to be considered at time At this moment, we are interested in
this
the film coefficient in order to obtain a picture of the over-all, fluid to-fluid
flow of heat From the definition of h, we see that the heat passing across
the films of Fig 176 is

(h) Q= AiA(f, — U) and Q = - /,) Btu/hr

As examples of the surface coefficients, we hax e from ASHAE (127),

Inside building walls, still air, design value, h = 1 65 Btu/hr-sq ft-°F,

Outside walls, 15-mph wind, de-ign \alue, h = 6 00 Btu/hr-*q ft-°F,


Eiaporating refrigerants m
tube, typical value, h - 200 Btu/hr-=q ft-°F,
Condensing *team m tube, typical value, h = 2000 Btu/hr-«q ft-'F

Actual x alues of h may be quite different from the foregoing *

272, Heat Transfer from Fluid to Fluid, If it were simple and convenient
to measure surface temperatures, equation (g) would be the one to use
Howex er, in practice, the fluid temperatures are easy to find, unless the speed
of the fluid is high (152), and it i& therefore desirable to express the heat flow
in terms of these temperatures To find (he corresponding equation, solve
for the temperature differences in equation (h), adding them to those m
equation (e) as previously explained, and obtain

„ AUi-U) n . -

Generalizing from this equation, we wnte for the «tead\ -state, unidirectional
flow of heat,

< 91A >

where Q is the rate of heat transfer from fluid to fluid through a wall of area
A sq ft When the temperature difference bet w een the fluids is St = f i — U
Si? is the total thermal resistance, and the total unit resistance to heat flow is

(,) +
in which S(l/h) is the sum of all the surface-layer Quid resistances, and
Z(L/k) is the sum of all the unit resistances of the materials of the wall
The ox er all unit resistance is represented by 1/C and the reciprocal of this
oxer-all resistance is called the fxansnuttance U, also the over-all coefficient
of heal transfer and the over-all (unit) conductance. f Since from equation
* Do not confuse the symbol h with that for enthalpy h

t Terms ending in -in(y designate characteristics of materials, normally independent


of «ixe or shape, sometimes called specific properties Examples are conduchntj and
resistivity Terms ending in -anee designate properties of a particular object, depending
not only on the material, but also upon size and shape sometimes called total quantities
m
Examples are crmdudsnee and transmittance Terms ending -ton designate time rate of
§ 273] CONDUCTION THROUGH CURVED WALL 323

(i), U = l/[S(l/h) + 2 (L/k)], equation (91A) may be written

(9 IB) Q — UA At Btu/hr.

See § 275. Table XIII shows some representative values of U.


It will be helpful to draw an analogy between the flow of heat and the
flow of electricity. You recall Ohm’s law as I = E/R where 7 is the cur- ,

rent flowing (analogous to the rate of heat flow), E is the electromotive force
or electrical potential (analogous to the temperature difference or thermal
potential which
the “driving force” for heat), and R is the resistance to
is

the flow of electricity (analogous to the thermal resistances of 1/hA and


L/hA). Note that the total thermal resistance is the equivalent of a series
connection of electrical resistances, in which case the total resistance is the
sum of the series connected resistances; R — R 3 R 2 Rz + + + • •

,

where R\, R?, R 3) etc. are the individual resistances. Compare with Fig.
176, p. 321. For R constant, the greater the voltage drop, the greater the
current flow (the greater the temperature drop, the greater the heat flow).
If the total resistance is increased by adding more resistances in the denomi-
nator of equation (91A), or by increasing the size of one or more resistance,
heat flows at a slower rate. This effect is the desired one in providing insu-
lation for steam lines, cold-storage rooms, etc. On the other hand, if a
greater rate of heat flow is desired, as in apparatus designed for the purpose
of transferring heat, we endeavor to reduce the resistance.
If one resistance
isvery much larger than any of the others, there will be little benefit derived
from reducing any of the resistances except the largest; thus, if the largest
resistance cannot be reduced, practically the best conditions have been
attained.
The engineer has to keep in mind that the transmittance may decrease
significantly during the operation of a heat exchanger because of the accu-
mulation of deposits on the transmitting surfaces. On the other hand, the
transmittance of insulated walls may increase due to the deterioration of the
insulating material. These materials often lose a large part of their insu-
lating value because they become wet from seepage or from moisture
deposited from air cooled below the dew point.

Conduction through Curved Wall. Since the area through which


273.
heat flows in a curved wall is not constant, we must reconsider Fourier’s

equation (90) in this connection. Consider a thick cylinder, Fig. 177, for
which the temperature on the inside surface is ta , the temperature on the
outside surface is t b and the thermal conductivity is k.
,
The heat flows
radially, say, from the inside to the outside, and in doing so, a given quantity
of heat passes across larger and larger areas, since the cylindrical area

the process of transfer Examples are conduction and transmission. Transmission,


. . .

transmissivity, transmittance usually refer to transfer by one or more of the processes of


conduction, convection, and radiation.” From A.S.A. Standard, Z 10.4-1943.
z

824 THE TRANSFER OF HEAT {C7i 16


increases with the radius of the cylinder Consider a length of cylinder z
and take a \ery thin element of the cjlinder of thickness dr with a radius of
r inches The area of this thin cylindrical surface is 2irr The change in .

temperature across dr is a differential amount dt Thus Fourier’s equation


gives

0 - dr

Separating the variables and integrat-


ing ne get

-2rrzk J\l,
Q J
Qln^ - 2irzk{ta - /„),

2 ir-k(U - h) = 2 rrzh At
In (r./r.) In (/).//>.)'
Fig 177 Curved Wall The ends
are insulated so that there Is no longi
tudinal flow of heat where r. is the outside radius of the
pipe r is the inside radius and r»/r, =
D„/D, In equation (j) note that the resistance for the curved wall is

Af _ In (r./r.)
(k) R _ ~
Q 2 rzk

Setting up an equation for the heat flow through a composite curved wall
such as an insulated pipe is simple when it is done by summing resistances
2 li in accordance with equation (01 A) 0 = At/2R Consider Fig 178
which represents a pipe A with insulation Y wherein the inside film coeffi

Fig 178 Composite Curved Wall


] —
§ 273 CONDUCTION THROUGH CURVED WALL 325

cient is ft, and the outside value is ft„; the temperature of the fluid on the
inside is U, which, in the following discussion, is greater than U, the tem-
perature of the fluid on the outside. The film resistances are (assuming no
film resistance at the intermediate area A)
1
K, — *
“t : and R0
A, ft. A 0h0
Applying equation (k) to X and Y (see the dimensions in Fig. 178
D /Da = n/ra ), we
b get their resistances as

_ In (Db/Da) In (D e/D b)
,

x and
2 -zk z 2 TTZky

Thus, the heat flow is Q = Al/ZR, or

1 ,
In {D„/D a ) In (D c /D b) 1 ’

A, ft,- 2 TTzk x 2-Kzky A 0 h0

or generalizing for any number of laminated cylinders, each with certain


inside and outside diameters of D and D„, with a conductivity of t ft, and
z ft. long, we have for a steady state

Al
(92A)

where At = h — l2 is
Q =

u +V Ay
the temperature drop from fluid to fluid, 2(1/Aft)
In {Do/Di)
2-zft

is

the sum of all film resistances in the path of heat flow, and the other term in
the denominator sums all other resistances. The heat is flowing radially
only.
important note: The value of ft in equation (92) must have the units
Btu-ft per hr-ft 2 -°F in order to get heat in Btu per hr. Compare with the
units given in Table XII.
In applying Q = U A At to curved walls, the area A becomes a con-
venient reference area; for a single pipe, either the outside or the inside pipe
area. Inasmuch as a certain amount of heat Q is passing through the wall
under steady-state operation no matter which area is the reference area, we
have
(92B) Q = UoA'At or U iJi , LiO,

wherein, by comparison with equation (92A)

UcAo = UiAi =
In {Do/D,)
2wzk

Thus, if an over-all transmittance U is given for a curved wall, the corre-


sponding reference area should be stated.
826 THE TRANSFER OF HEAT [Ch 16
274 Example. An insulated steam pipe, located where the ambient tempera-
ture is 90°F, has an internal diameter of 2 in and an external diameter of in
1\i
The outside diameter of the corrugated asbestos insulation is 5 in and the surface
coefficienton the outside is A, = 2 Btu/hr-sq ft-°F, a value intended to include
the effect of radiated heat where the movement of the air is that due to natural
circulation (free convection) On the inside, the steam has a temperature of
300°F and ft, = 1000 Btu/hr-sq ft-°F. (This value of A, is commonly used for
saturated or uet steam flowing m a pipe Note that this v alue corresponds to a low
resistance to heat flow, so that some inaccuracy here has little effect on the trans-
mittance ) Compute (a) the heat loss per foot of pipe length, and (b) the surface
temperature on the outside of the insulation
8OLUTI0V (a) Refer to Fig 178 Trom Table XII, we find
ft, = 312 Btu in /hr-ft*-°F for steel
ft„ = 0 828 Btu-m /hr-ft ,-°F for the corrugated asbestos insulation

Converting the inch unit to feet, we ha\e ft, = 312/12 and ft, = 0 828/12 The
true value of ft for steel will vary with the composition of the steel, but Bince the

Fig 178 Repeated

resistance of any steel to the flow of heat is relatively small in a case like this, what-
ever error is involved will be seen to have negligible effect after numerical results

have been obtained First, compute the various resistances per foot of pipe length

The area of a cylinder is irDz, where z = 1 ft is its length and the area is to be in
square feet (2/12 ft in Ri, 5/12 ft m R t )
12
Inside film R i
= 0 00191 hr-°F/Btu
A, A, (* 2 ) ( 1000 )
In (D./A) In (2 5/2)
Pipe Ri " 0 001365 hr-°F/Btu
2«ft (2t) (312/12)
In (Dp/D,) = In (5/2 5)
Insulation R» 1 6 hr-°F/Btu
2irzft (2t) (0 828/12)

Outside film A
A
§ 275] LOGARITHMIC MEAN TEMPERATURE DIFFERENCE 327

Summing these resistances, we have R = 1.985275 hr-°F/Btu, where it is evident


that the last several digits are not significant. Observe the small effect of the first
two resistances. For At = 300 — 90 = 210°F, we get

Q = ^ = 106 Btu/hr. for each foot of length.

(b) The heat Q = 106 Btu/hr. flows through each cylindrical lamination and
will be equal to

temperature difference between two sections


q _ sum
of the resistances between the same sections

For a temperature difference to the outside surface of h — tc = 300 — tc


,
Fig. 178,
and a resistance to the outside surface of + R2 + Rs — 1.603, we get

106 = ~
1.603

from which tc — 130°F, a safe value for the outside surface.

276. Logarithmic Mean Temperature Difference. Our discussion of the


transfer of heat through wallsfrom one fluid to another has been on the
tacit assumption that the hot fluid remains at a constant temperature fa
and the cold fluid remains at a con-
stant temperature fa. However, in
many instances, either the cold fluid
or the hot, or both, undergo a change
of temperature in passage through
the heat exchanger.
For purposes of explanation, con-
sider a double-pipe type of exchanger
(Fig. 179). Let the hot fluid flow
through the annular space and the
cool fluid through the inner pipe. As
the temperatures of the fluids change, Fig. 179. Double-Pipe Arrangement.
Observe the notation. One end of the
the difference between the tempera-
exchanger is designated as section A
tures of the fluids changes. Thus we and the other end, section B. In this
find different differences between the illustration, a counterflow arrangement,
IcA isthe initial temperature of the cold
temperatures of the hot and cold fluids
fluid and ihA is the Anal temperature of
at different sections of the exchanger. the hot fluid.
In such cases, the logarithmic mean
temperature difference (LMTD) represented by Afm is used in place of ,

ti — ti. Since the rate of change of temperatures of the substances is not


constant, the LMTD is not the same as the arithmetic mean, as shown in
equation (93) below.
At this time, we shall consider two types of flow', parallel flow, in which
case the fluids flow in the same direction through the heat exchanger, and
828 THE TRANSFER OF HEAT [Ck 16

counterflow, where the fluids flow in opposite directions In a general


way, the variation of temperatures is as indicated in Fig 180 In case of
parallel flow, both fluids enter at section A, and as they pass to section B
their temperatures approach one another At and near section A (Fig
180(a)] the temperature difference is a maximum, and consequently the
rate of flow of heat and the rate of decrease of temperature difference
are a maximum As the temperatures of the fluids approach one
another, the rates of change of the temperatures decrease and the curves
flatten out In Fig 180(b), a counterflow arrangement, the cold fluid
enters at section B, leaves at A In contrast to parallel flow, heat transfer
is taking place between the fluids at the moment when each is in its cold
cst state and when each is in its hottest state The transfer of heat in
a counterflow exchanger tends to conserve available energy and makes

Fig 180 Temperature Variations, Parallel Flow and Counterflow In each case, the
temperature difference at section A Is At* >» f** — t, A and at section B the tem-
,

perature difference is OIb = Ub — Ub

possible a higher final temperature of the fluid being heated than obtainable
with parallel flow
To find the equation for the mean temperature difference consider a
differential area of transmitting surface dA in a counterflow exchanger
As the cold fluid flows past this area, its temperature increases a differential
amount dle due to the transfer of a differential quantity of heat dQ the
amount of which is dQ = wcce dt c where wc is the mass of cold fluid flowing
,

in a unit time and ce is the specific heat of the cold fluid Assuming no
heat loss from the exchanger and no change of kinetic energy we conclude
that the heat loss of the hot fluid dQ = ic*c* dt h is the same as the gain of
the cold fluid Note that the f d/* is a negative number, so that
dlc + dU = d(At)

is the change in the difference of temperatures, Jd(Ai) = Ah — Ati (Fig

180) across a differential area dA of infinitesimal length dL between sections

1 and 2 From the definition of the transmittance U, wo have also

dQ = UdAAt
f l

§ 276 ] THERMAL RADIATION 329


Using the conclusions of the foregoing discussion, we get

die + dk = d(M) = dQ f-i


\WcC c
i-)
IChCh/
= u dA A t(—
\WcCc
L\
Wh C,J
where dQ is a positive number and the negative sign on the right-hand side
isnecessary to produce an airthmetic difference. Rearranging the foregoing
equation, we have

wherein the transmittance U and specific heats c are taken as constant.


Integration of this equation gives

Mb = = HA - EaL = _ UA _ ua
Jn
At A
_i n
A[fi WtCc wh Ch (
\WhCh WcC
\
c )
Now using Q — UA At„ = wc c c ((c a — (c b) = wh ch (th A — thB) in this equa-
tion, we find -

in —1 = A ~ = ^hA ~ ~ (l hB ~ Qb)
;
‘At B M„ M„ Mm
from which
A lA - Afs
(93! M„ =
In (,Ma/M b )’

the value of the log mean temperature difference between sections A and
B when the difference in temperature at A is AtA , and the difference at B is
A B . not matter which end of the heat exchanger is taken as section
It does
.4 B the same result will be obtained from equation (93). A
and which ;

similar analysis of the case of parallel flow will yield the same equation (93).
Hence this equation applies to either case and it serves also whenever the
temperature of either fluid is constant, as during evaporation or condensa-
tion. When AIa is equal to or nearly equal to Alo, use the arithmetic aver-
age for At m Where there is cross flow, as around baffles in heat exchangers,
.

and where there are fins, the mean temperature difference is not given by
equation (93); nor in case w, c, or U varies. For such other situations, see
McAdams (137). Equation (91) may now be written,
(9 1C) Q = UA A m .

In taking values of U for curved walls from the literature, notice whether
they are based on the inside A, or outside A„ surfaces.

276. Thermal Radiation. AE bodies radiate heat. If two bodies were


completely isolated in a vacuum from all other bodies, but not from each
other, the colder body would radiate heat to the hotter body, and the hotter
S30 THE TRANSFER OF HEAT [Ch 16
body would radiate heat to the colder bodj For purposes of explanation
let all the radiant heat from the hotter body go to the
colder body and all
the radiant heat from the colder body to the hotter body Then the colder
body would be heated because due to its lower temperature it radiates less
heat than it receives from the body with the higher temperature Radiant
heat or thermal radiation is an emanation of the same nature as light and
radio waves Like other wa\cs mstriking a body some of them will be
reflected (the fraction being indicated by the reflectivity) some will be
absorbed (the fraction being indicated by the absorptivity) and if the body
is transparent to the naves some will be transmitted through the body (the

fraction being given by the transmissivity) The sum of the reflectivity


plus absorptn lty plus transmissivity is equal to 1 or 100% Bodies which
are opaque to light are for the most part opaque to radiant heat (true of
most solids) For opaque substances the energ} of the radiant heat is
either absorbed or reflected w which case the reflectivity plus absorptuitj
is equal to one At the other extreme some gases and some glass transmit
nearly all the radiant heat Brightly polished metals are such good reflec
tors that most of the radiant heat may be reflected
A body which absorbs all the impinging radiant heat is called a black
body, a hypothetical conception m
w hich the absorptiv lty is unity A black
body is also the best radiator At a particular body temperature actual
bodies radiate less heat than a black body The ratio of the radiation from
an actual body to the radiation from a black body is called the emisswily <
The emissiMty is not a constant property but usually it increases with the
temperature of the radiating body
Ivirchoff s law may be stated as follows the absorptivity a and the
emissivity t of a bodj arc the same when the body is m thermal equi
librium with its surroundings From this statement we can see that the
emissivitj of highly polished metals maj be quite low
The nearest approach to a black body is obtained by a hollow \ essel
penetrated only bj a small pm hole through which radiant heat maj pass to
the inside Once inside little of this radiant heat has a chance to be
reflected or rcradiated back through the pin hole Thus for practical

purposes this vessel absorbs all the energy entering the hole

277 Stefan-Boltzmann Law This law states that the amount of radia
tion from a black body is proportional to the fourth pow er of the absolute
temperature Qr = <r 4 T* where tr ~ (0 1713) (10 ) is the Stefan Boltz
8

mann constant when Q is m Btu per hour and A sq ft is the radiating area
of the black body w hose surface temperature is T°R As usual this equa
tion for an ideal situation must be modified to care for actual situations As
pre\iously stated net heat transferred by radiation is the result of an inter
change of radiation the radiation from the hot body to the cold bodj
minus the radiation from the cold body Also the effective emissivities
;

§ 278 ] CLOSURE 831

and the fact that all the radiation from a certain source will not strike the
surface must be considered. Introducing these corrections into the Stefan-
Boltzmann law, we have

(94) <3, - o.mw [Q4)' - (it


mj J
Btu/hr.,

where Fa is a view factor, sometimes called an angle factor, which allows for the
average solid angle through which one surface “sees” the other; that
is, it allows for the exchange of radiation which is direct (not reflected

from intermediate surfaces)


Fe is a factor which depends on the individual emissivities e (epsilon) and, in
some instances, on the ratio of the areas.
A is the area in square feet of a surface.

Further detail on radiation is beyond the intended scope of this book.

278. Closure. At this stage of development of the science of heat trans-


fer, the computation of the heat flux is dependent largely on experimental

results, as indicated in the previous discussions. On this account, it is often


desirable to search the literature for experiments related to the particular
situation in which one is interested; or to verify one’s design by experiment.
Ifthe effects of contaminants in the fluids and on the surfaces are included,
the over-all coefficients vary widely for seemingly similar circumstances,
which suggests a thoughtful approach to actual designs.
LIST OF REFERENCES
This is not a bibliography. It includesbooks and papers from which acknowledge-
ment to specific material could be made, but it also includes works where the help

was less tangible. These references should be helpful to those who wish to pursue
further study.

1. Zemansky, M. W., Heal and Thermodynamics, McGraw-Hill.


2. Proceedings of the 1951 Summer School for Thermodynamics Teachers;
Michigan State University.
3. Keenan, J. H., Thermodynamics, John Wiley.
4. Maxwell, James C., Theory of Heat, Van Nostrand.
5. Steam, 37tli ed., The Babcock & Wilcox Co.
6. Baker, Ryder, Baker, Temperature Measurements in Engineering, John Wiley.
7. Combustion Engineering, Combustion Engineering-Superheater, Inc.
8. ASME Codes on Instruments and Apparatus.
9. Shoop and Tuve, Mechanical Engineering Practice, McGraw-Hill.
10. Mooney, D. A., Mechanical Engineering Thermodynamics, Prentice-Hall.
11. Symposium on “High Temperatures,” Scientific American, Vol. 191, Sept.,
1954.
12. Shortley and Williams, Elements of Physics, Prentice-Hall.
13. Kiefer, Kinney, Stuart, Principles of Engineering Thermodynamics, 2d ed.,

John Wiley.
14. Keenan and Kaye, Gas Tables, John Wiley.
15. Ellenwood, Kulik, and Gay, The Specific Heat of Certain Gases over Wide Ranges
of Pressures and Temperatures, Cornell Eng. Exp. Sta., Bull. No. 30.
16. Paitington and Schilling, The Specific Heal of Gases, Van Nostrand.
17. Spencer and Flannagan, “Empirical heat capacity equations of gases” Jour.
Am. Chem. Soc., Vol. 64, p. 2511.
18. Spencer and Justice, “Empirical heat capacity equations for simple gases,”
Jour. Am. Chem. Soc., Vol. 56, p. 2311.
19. Sweigert and Beardsley, Empirical specific heal equations based upon spec-
troscopic data, Ga. Inst, of Tech., Eng. Exp. Sta. Bull. No. 2.
20. Chipman and Fontana, “A new approximate equation for heat capacities at
high temperatures,” Jour. Am. Chem. Soc., Vol. 57, p. 48.
21. Spencer, Jour. Am. Chem. Soc., Vol. 67, p. 1859.
22. Obert, E. F., Thermodynamics, McGraw-Hill.
23. Dodge, B. F., Chemical Engineering Thermodynamics, McGraw-Hill.
24. Physical Tables, 1954; Smithsonian Institution.
25. Sears, F. W., Thermodynamics, The Kinetic Theory of Gases, and Statistical
Mechanics, Addison-Wesley.
26. Darrieus, G., “The rational definition of steam turbine efficiencies,” Engineer-
ing, Vol. 130, p. 195.
sss
, ,

334 LIST OF REFERENCES


27. Keenan, J H “A steam , chart for second law analysis," Meek Eng Vol 54,
p 195
28 Lewis and Randall, Thermodynamics and the Free Energy of Chemical Sub-
stances, McGraw-Hill
29 Dole, M, Introduction to Statistical Thermodynamics, Prentice Hall

30, Keenan, J ,
H
“Availability and irreversibility thermodynamics,” British m
Jour App Phys Vol 2, p 183
31 Stuart and Jackson, “The analysis and evaluation of compressor performance,”
Meek Eng Vol 76, p 287.
32 Compressed Air Handbook, Compressed Air and Gas Institute
33 Wilson and Crocker, “Fundamentals of the Elhott-Ly sholm Compressor,”
Meek Eng , Vol 68, p 514
34 Baumeister, Jr , T , Fans, McGraw-Hill
35 Church, A II , Centrifugal Pumps and Blowers, John Wiley
36 ASME Test Code for Compressors and Exhausters, PTC 10 1949
37 Gill, T T Air and Gas Compression, John Wiley
,

38 Lee, John F , Theory and Design of Steam and Gas Turbines, McGraw-Hill
39 Rettaliata, J T, "The gas turbine,” Allis Chalmers Electrical Renew, Sept,
1941, Mar, 1942
Dec,
40 Zucrow, M
J , Principles of Jet Propulsion, John Wiley
41 Dusmberre, G , M
Gas Turbine Power, International Textbook
42 Vincent, E T The Theory and Design of Gas Turbines and Jet Engines, Mc-
,

Graw-Hill
43 Godsey and Young Gas Turbines for Aircraft, McGraw-Hill
44 Fraas, A P Combustion Engines McGraw-Hill
,

45 Jennings and Rogers, Gas Turbine Analysis and Practice McGraw-Hill


46 Cohen and Rogers Gas Turbine Theory, Longmans Green
47 Sorenson, II A Gas Turbines, Ronald Press
48 Faires and Chambers, Analytic Mechanics, Macmillan
49 Obert E F Internal Combustion Engines 2d ed International Textbook
, ,

50 Taylor and Taylor, The Internal Combustion Engine rev cd International


Textbook
51 Hersey, Eberhardt, and Hottel, “Thermodynamic properties of the working
fluid in internal combustion engines ” SAE Jour \ol 39, p 409 ,

52 Lichty, L C Internal Combustion Engines, McGraw-Hill


,

53 Rogowske, A R Elements of Internal Combustion Engines, McGraw-Hill


,

54 Maleev, V L Internal Combustion Engines 2d ed McGraw-Hill


,

55 Poison, J A Internal Combustion Engines, 2d ed John Wiley


, ,

56 Streid D D , “Gas-turbine fundamentals 'Meek Eng Vol 68 p 127


57 Dickson, J , “The Diesel engine for rail transportation, ’
Mech Eng Vol 57,

p 547
58 Automotive Engine Test Code, General Motors
59 Lewis and von Elbe Combustion Flames and Explosions Academic Press
New York
60 Jost W ,
(translated by H O Craft), Explosion and Combustion Processes

l» Gases McGraw-Hill
61 NACA Report No 476 616
LIST OF REFERENCES 335

62. Selected Values of Properties of Hydrocarbons, National Bureau of Standards,


Cire. C 461.
63. Hottel, illiams, Satterfield, Thermodynamic Charts for Combustion Processes,
Parts I and II, John Wiley.
64. Su and Chang, “A generalized van der Waal’s equation of state for real gases,”
Industrial and Engineering Chemistry, Vol. 38, p. 800.
65. Su, G. J., “Modified law of corresponding states for real gases,” Industrial
and Engineering Chemistry, Vol. 38, p. 803.
66. Nelson and Obert, “Laws of corresponding states,” AIChE Jour., Vol. 1,
p. 74.
67. Nelson and Obert, “Generalized compressibility charts,” Chem. Eng., Vol. 61,
p. 203.
6S. Nelson and Obert, “Generalized pvT properties of gases,” ASME Trans.,
Vol. 76, p. 1057.
69. Hall and Ibele, “Compressibility deviations for polar gases,” ASME paper,
1954 Annual Meeting.
70. Bird, Hirschfelder, Curtiss, “Theoretical calculation of the equation of state
and transport properties of gases and liquids,” ASME Trans., Vol. 76, p. 1011.
71. Keenan and Keyes, Thermodynamic Properties of Steam, John Wiley.
72. Peebles, Jr., J. H., Thesis, Tulane Univ.
73. Kreisinger and Purcell, “Some operating data for large steam generating
units,” ASME Trans., Vol. 50.
74. Philo, F. G., “Comparative performance of a large boiler using oil and natural-
gas fuels, ASME Trans., Vol. 54.
75. Barnard, EUenwood, and Hirshfeld, Ileal-Power Engineering, Vols. II and III,
John Wiley.
76. Gaffert, G. A., Steam Power Stations, McGraw-Hill.
77. Karlsson and Holm, “Heat transfer and fluid resistances in Ljungstrom
regenerative-type air preheaters,” ASME Trans., Vol. 65.
78. Simmang, Saadeh, Short, “Heat transfer in the rotating-element air preheater,”
Mech. Eng., Nov., 1954.
79. Morse, F. T., Power Plant Engineering and Design, Van Nostrand.
SO. ASME Rules for Construction of Power Boilers.
81. Zerban and Nye, Steam Power Plants, International Textbook.
82. Powell, S. T., Boiler Feed Water Purification, McGraw-Hill.
S3. Matthews, F., Boiler Feed Water Treatment, 3d ed., Chemical Publishing Co.
84. Ellenwood and Mackey, Thermodynamic Charts, John Wiley.
85. Murray, R. L., Nuclear Engineering, Prentice-Hall.
86. Glasstone and Edlund, The Elements of Nuclear Reaction Theory, Van Nostrand.
87. Ridenour, L. N., et al., Modern Physics for the Engineer, McGraw-Hill.
88. G-E’s New Dual-Cycle Boiling Reactor, General Electric Co.
89. Bern.-, C. H., “Steam turbine testing,” Mech. Eng., Nov., 1935.
90. Croft, T.,Steam Engine Principles and Practice, McGraw-Hill.
91. Peabody, C. H., Thermodynamics of the Steam Engine and Other Heat Engines.
John Wiley.
92. The Universal Unafiow, and other publications of Skinner Engine Co.
93. Thurston, R. H., .4 Manual of the Steam Engine, John Wiley.
,

336 LIST OF REFERENCES

94 Allen, R C , “Steam turbine luston and highlights," AUu-Chalmers Elec-


trical Reneu.
95 Emmet, W L R , “Status of the Emmet mercurv apor process " Mech Eng
No\ , 1937
9G Sheldon, L A , “Properties of mercury \apor,” ASVE Trans Vol 4G, p 272 ,

97 Ilackett, H
N , “Mercun steam power plants ” Mech. Eng \ol 73, p 559 ,

98 Krieg, E H., “Superposition,” Meek Eng Sept 1936. ,

99 Shapiro, A H , The Dynamics and Thermodynamics of Compressible Fluid Flour,


2 vol , Ronald Press
100 Binder, R C , Fluid Mechanics and Adianced Fluid Dynamics and Fluid
Machinery, Prentiee-IIall
101 Binnie and Woods, “The pressure distribution in a com ergent-dn ergent
nozzle " Proc Inst Mech Enp.lol 13S, p 2G0
102 Yellott and Holland, “Condensation of steam in dnergmg nozzles,” AS ME
Trans , Vol 59, p 171
103 Neblett, R S , “Steam turbine-generators of the future,” General Electric
Renew, Vol 5S, No 4, p 8
104 ASME
Report of the ASMS
Special Research Committee on Fluid Meters
105 ASME Flow Measurements (1940), Part 5, Power Test Codes
106 Stodola-Lowenstein Steom and Gas Turbines, McGraw-Hill
107 Church, Jr, E F Steam Turbines McGraw-Hill ,

10S Salisbury J , Steam K


Turbines and Their Cycles McGraw-Hill
109 Shrotzki and \opat Steam and Gas Turbines, McGraw-Hill
110 Wood well, J E “Efficiencj test of a 15, 000-kilowatt turbogenerator,” Power,
,

Vol 61, p 128


111 Goudie, IV JSteam Turbines Longmans Green
112 Franck Sr, C C (W estinghouse Electric), Super pressure steam turbines,
a paper presented to Southeastern Electric Exchange Sept 1955 ,

1 13 New man Keller, La ons, and M ales Modem T urbmes, John Wilej
114 Refrigerating Data Book, Am Soc Refrig Eng
115 Woolnrh and Bartlett Handbook of Refrigerating Engineering Van Xost rand
116 Jordan and Pne«ter Refrigeration and Air Conditioning Prentice-Hall
117 Raber and Hutchinson Refrigeration and Air Conditioning Engineering John
Wdea
118 Maelntire II J Refrigeration Engineering John Wile\
,

1 19 Jennings and Lew is, Air Conditioning and Refrigeration International Textbook
120 Sparks, N
R Theory of Mechanical Refrigeration McGraw-Hill
,

121 Mojer and Fittz, Refrigeration McGraw-Hill


122 Penrod E B “A renew of some heat pump installations ” Mech Eng Vol
,
,

69,p 639
123 Penrod, E B , "Sizing earth heat pumps ” Refng Eng Vol 62, No 4 p 57
124 Thompson, R J , Properties and characteristics of refrigerants presented at
7th annual corn ention of Refrigeration Sen ice Engineers Soc
125 Penrod, Baker Ievj, Chung, Unit of Kentucky Heat Pump A’o. 1, Eng
Exp Sta Bull No 30 Umv of Kj
126 Scatchard, Epstein, VaTburton Coda, “Thermodanamic properties satu-
rated liquid and a apor of ammonia-water mixtures,” Refng Eng , lol 53,
No 5, p 413
LIST OF REFERENCES 337
127. ASHAE Guide (published annually).
12S. Carrier, Cherne, Giant, Modern Air Conditioning, Heating and Ventilating,
Pitman.
129. Greene, Jr., A. M., Principles of Healing, Ventilating, and Air Conditioning,
John "Wiley.
130. William Goodman, Air Conditioning Analysis, Macmillan.
131. Arnold, J. H., “The theory of the psyehrometer,” Physics, Vol. 4.
132. London, Mason, Boelter, “Performance characteristics of a mechanically in-
duced draft, couuterflovs packed cooling towtT,” ASME Trans., Yo\. 62, p. 41.
,

133. Lichtenstein, J., “Performance and selection of mechanical-draft cooling


towers,” ASME Trans., Vol. 63, p. 779.
134. Foster Wheeler Corp., “ Preliminary cooling ton er selection,” Heat Engineering,
Vol. XXIV, Xo. 2, p. 26.
135. Engineering Manual for Industrial Cooling Towers, The Marie}' Co.
136. Moss, J. F., Cooling tower performance for air conditioning systems, Bui. No. 110,
Texas Eng. Exp. Sta., College Station.
137. McAdams, W. H., Heat Transmission, McGraw-Hill.
138. Brown and Marco, Introduction to Heat Transfer, McGraw-Hill.
139. Croft, H. 0., Thermodynamics, Fluid Flow and Heal Transmission, McGraw-
Hill.
140. Dodge and Thompson, Fluid Mechanics, McGraw-Hill.
141. Jakob and Hawkins, Elements of Heat Transfer and Insulation, John Wiley.
142. Hall, X. A., Thermodynamics of Fluid Flow, Prentice-Hall.
143. General Discussion of Heat Transfer, AIME-ASME, Proc., 1951.
144. Jakob, M., Heat Transfer, Vols. I and II, John Wiley.
145. Ingersoll and Zobel, Heat Conduction, McGraw-Hill.
146. Schack, Smith, and Partridge, Industrial Heal Transfer, John Wiley.
147. Dusinberre, G. M., Xumerical Analysis of Heat Flow, McGraw-Hill.
148. Lee and Sears, Thermodynamics, Addison-Wesley.
149. King, W. J., “The basic la us and data of heat transmission,” Mech. Eng.,
Vol. 54, April, 1932.
150. Hilsenrath and Touloukian, “The viscosity, thermal conductivity, and Prandtl
number for air, O?, X*,NO, CO, C0 2 H»0, He and A,” ASME Trans., Vol. 76,
,

p. 967.
151. Boelter, Cherry, Johnson, Martinelli, Heat Transfer, Univ. of Calif. Press.

152. Hottel and Kalitinsky, “Temperature measurement in high-velocity air

streams,” ASME Trans., Vol. 67, p. A-25.


153 Hottel, H. C., “Radiant heat transmission,” Mech. Eng., Vol. 52, p. 699.
154. Hottel and Mangelsdorf, “Heat transmission by radiation from non-luminous
flames,” AIChE Trans., Vol. 31, p. 517.
155. Hottel, H. C., “Radiation from non-luminous flames,” ASME Trans., Vol. 57,
p. 463.
156. Keyes, F. G., “Summary of viscosity and heat conduction data for He, A, H 2,

0 Xs, CO, CO:,


2, s O, andHair,” ASME
Trans., Vol. 73, p. 589.
157. Hall and Ibele, “The tabulation of imperfect-gas properties for air, nitrogen,
and oxygen,” ASME Trans., Vol. 76, p. 1039.
158. Eckert, E., Introduction to the Transfer of Heal and Mass, McGraw-Hill.
159. Faires, V. M., Thermodynamics, Macmillan.
APPENDIX A. PROBLEMS
vote 1 If desired, additional problems, together with convenient tables of properties and
trorl mg charts, are mailable in Problems on Thennodjnamies byFaires, Brewer, Simmang,
published by The Macmillan Co
1.
NOTE 2 Unless otherwise staled in these problems, the atmospheric pressure should be
talen as 14 ~ psia ISO 92 in Hg)
note 3 The answers given are usually slide-rule answers and are subject to inherent
errors

THERM OD I V mic PROPERTIES


1. Work, energj and moment (or
,
m psi, and in psf 9 (c) Is pressure an
torque) are all measured in foot-pounds intense, e or extensive propert} 9
or similar units of space and force Maj Suppose pressure is being measured
7.
we add energN and moment 9 Explain b}a column of water whose densit} is
full}
62 lb /cu ft Explain how to com ert a
2. Weight and mass mai both be meas- column height z in feet to psi and deter-
ured in pounds If the w eights gi\ en are mine the corn ersion constant
a= obtained on Mt Ramer, i= it correct to 8 Steam at a pressure of 150 psia and
add the weights in pounds and masses in a temperature of 400°F occupies a x olume
pounds 9 Is there an} circumstance w here of 3 223 eu ft /lb (a) What is its densit}
such a sum would gne either the total in lb /cu ft and m
slugs/cu ft 9 (b)
mass or the total weight 9 Discuss fulli Coni ert the pressure to inches Hg and
3. A. bod} weighs 1000 lb at standard feet of water
(60°F) (c) Coni ert the
grant} What would it weigh on top of temperature to °R, °K, and °C Is tem-
Pikes Peak (g ~ 32 12 fps”) 9 at sea lei el perature an mtensne or an extensive
at the equator (g = 32 26 fps 2 9 at sea )
properti 9

lei el at the North Pole (g = 32 09 Ans (a) 0 31, 0 009G5, (b) 306, 5085,
fps*) 9 (c) 860, 478, 204 4

4. Fne mas-es are as follows m, is 9 The same as 8 except that the steam
500 gm of mass, m» weighs 800 gm in i is at 200 psia, 500°F, 2 726 cu ft /lb
weighs 4 poundals, m t weighs 32 2 lb and , 10. What are the units of the constant
m, i= 2 slugs The weights are at standard 0 96 mD = 0 96(a?) 2 where 1) is in mph,
1 ''
,

grai tty What is the total mass expressed a in ft /sec and s m ft 9


(a) m
pounds, (b) in slugs, (c) in grams 9 Ans mph /fps
Ans (a) 99 59 lb (b) 3 I slugs, (c)
,
11. A certain equation in heat transfer
45,200 gm is h = C AT” ?i6 /(D° 2 7 ° ), where h is
,
m
6. V 10-cu ft tank contain” 3 22 lb of the heat transfer coefficient for free con-
a gas II hat is the density of the gas m icction in Btu/ft 2-hr-°R, AT is a tempera-
a slugs /cu ft 9 -4ns 0 01 ture difference in °R, I) is a characteristic
6 (a) If the pressure gage on a tank
dimension in inches, and T is temperature
reads 20 in Hg gage, what t« the absolute in
C
R What are the units, if ani of C 9 ,

pressure in inches Hg, in p”i, and in psf 9


Ans Bt u-in » 2 /ft 2-hr-°R 085 1

(b) If the gage reads 20 in Hg laeuum, 12-20. These numbers ma} be used for
what is the absolute pressure in inches Hg, other problems

2. CONSERVATION OF ENERGY
note Consider all energy problems from two viewpoints, the energy equation and the
energy diagram

21. An inientor claim® that he will be time


his engine in a specified An engi-
able to get 3,890,000 ft-lb of work from neer miestigatmg the claim finds that
339
840 APPENDIX A PROBLEMS
during the same time, fuel is supplied to the work per lb Convert to hp Is it
the engine which releases 4400 Btu of heat done on or by the substance?
What conclusion does the engineer draw 4ns +5 85 Btu/lb
concerning the inventor’s claims? What
29 An compressor (an open system)
air
is the basis of this conclusion?
receives 600 lb /min of air at 14 4 psia and
22 A closed system which is undergoing a volume of 2 cu ft /lb The air
specific
cyclic changes of state receives heat at the flows steadily through the compressor and
rate 13 550 Btu/min , and discharges is discharged at 100 psia and 0 4 cu ft /lb

17,110 Btu min of heat Over a com- The initial internal energy of the air is

plete cycle there is no change of stored 12 Btu/lb at discharge


, the internal
energy which is entirely internal energy energy is 47 Btu/lb The cooling water
The only other energy crossing the bound- circulated around the cylinder carries
ary is work How much work is done? away 33 Btu/lb of air The change m
Is it done on or by the system? kinetic energy is a 2 Btu/lb increase
A ns —59 33 Btu/sec AP — 0 Sketch an energy diagram
Compute the work (by or on?) in Btu/lb
23 A closed system undergoes a process and in hp —72 Btu/lb
Ans
during which it receives 2250 Btu of heat
and docs 3220 Btu of w ork What is the 30 Steam enters the blades of a single-
change of stored energy? stage impulse turbine with a \elocity of
1400 fps and leaves with a velocity of
Ans -755,000 ft-lb
4 CO fps Assuming that there is no change
24 A closed system consisting of helium in ft = u +
pv/J and no transferred heat,
iscompressed by 373 Btu of work while apply the energy equation and calculate
the internal energy increases 299 Btu the work delivered to the blades per pound
How much heat, if any, flows? in or out? of steam flowing Compute the hp for a
flow of 42 4 lb /min Sketch an energy
25. Assuming that there are no heat
diagram indicating the flow of energy
effects and no frictional effects find the
An* 34 9 Btu/lb
kinetic energy and velocity of a 3220-lb
body (mass of water) after it falls 778 ft 31 An air compressor picks up air for
from rest Start with the steady flow which = 20 Btu/lb and dcli\crs it
ft

equation, deleting energy terms which are when ft = 70 Btu/lb While passing
irrelevant Ans Uj - 223 7 fps through the compressor the air rejects
10 Btu/lb Neglect the change of kinetic
26 Water leaves a nozzle at 72 mph energy and compute the work (in Btu and
If the jet is directed vertically upward, hp) for compressing 50 lb /min Apply
how far above the nozzle will the water go, the energy equation and then show the
ideally? This is an open system without energy diagram An* — 3000 Btu/min
frictional effects Start with the steady
flow equation, deleting energy terms which 32 (a) A boiler is a steady flow device
are not applicable Ans 173 5 ft Let the system be the II 2 0 between entry
and exit Entering energy consists of the
27 A
64,400 lb airplane
(a) is traveling internal energy of the w ater the flow w ork
at 1000 fps (682 mph) How much is its and the heat transferred to generate the
kinetic energy in hp-hr ? (b) If it noses steam Departing energy consists of in-
vertically upward at this speed with power ternal energy and flow work The change
off and in a vacuum (no friction) at stand- of kinetic energy is generally negligible
ard gravity, through what vertical dis Show an energy diagram and set up an
tance would it move? expression lor the beat to generate \ lb oi
An* (a) 1 288,000 Btu, (b) 2 94 mi steam (b) Water enters a boiler with
ft = 128 Btu/lb The steam leaving has
28During a steady flow process the internal energy it* * 1111 Btu/lb and
pressure of the working substance drops w’ork passing the boundary of
does flow
from 200 to 20 psia the velocity increases
W, x = 83 9 Btu/lb What is the trans
from 200 to 1000 fps, the internal energy
f erred heat?
of the open system decreases 25 Btu/lb ,

and the volume increases from 1 to 10 cu 33 (a) Apply the energy equation to a
ft /lb No heat
transferred
is Sketch turbine where the transferred heat is negli
an energy diagram and apply the steady gible, and set up the expression for the
flow equation for 10 lb /sec to determine work of 1 lb of steam Show the energy
APPENDIX A. PROBLEMS S41
diagram, Steam
enters a turbine with
(b) is negligible, what is the final velocity?
h = 1107.5 Btu/lb. It leaves with hi (b) Considering the process in the turbine
= 9S5 Btu/lb.; AK
= 0. What is the as a whole where the change of kinetic
work done in one hour when the steam energy is zero, determine the work in
(lows through at the rate of 1 lb. /sec.? Btu/min. and in hp.
What is the corresponding hp?
37. Jetengines provide a thrust of
Ans. 703,920 Btu/hr.
10,000 lb. on a plane moving at 500 mph.
Steam enters a turbine with a veloc-
34. What is the rate at which work is being
ity of 50 fps and a value of h = 1292 , expended on the plane in Btu/sec., in hp,
Btu/lb. The exit velocity is 117 fps and in kw, and in hp-hr./sec.?
hi = 1098 Btu/lb. Loss by radiation is
13 Btu/lb. (a) What is the work in 38. (a) Determine the /pdF for a
36. process during which the pressure remains
Btu/min. and hp if the amount of steam
constant and evaluate it in Btu for a pres-
flowing is 2 lb. /sec.? (b) The change of
kinetic energy is what percentage of Ah?
sure of 778 psi during a volume change
Sketch an energy diagram. from 2 to 10 eu. ft. What does this
integral evaluate for a reversible nonflow
Ans. (a) 21,690 Btu/min.
process? (b) Using the Jp dV, determine
A combustor on a gas turbine,
the nonflow work while the volume remains
Fig. 80, p. 152, receives air with an en-
constant.
thalpy of 208 Btu/lb. and, at the same
time, fuel which carries with it 323 Btu/lb. 39.A nonflow process occurs for which
of air, mostly chemical energy which is p = V* + 8/F psia, where V is in cu. ft.
released by combustion. The heat loss is If the volume expands from 5 to 10 cu. ft.,
M Btu/lb. of air passing through, and the what is the work? Ans. 55.1 Btu.
amount of energy leaving in the exit stream 40. A process occurs during which
in the form of unburned fuel is 16 Btu/lb.
p = 6F +
20 psia while the volume V
of air. What is the enthalpy of the de- changes from 2 to 3 cu. ft. What is the
parting gases? Sketch an energy diagram.
/p dV in Btu? What is this in a nonflow
Ans. 501 Btu. process?
36. A
gas in steady flow at the rate of
41. The same as 40 except that pF"
1000 lb. /min. performs a process in which = 778, where p is in psia, F in cu. ft.,
<?=0 and the enthalpy decreases 19.2 and n = 1.28.
Btu/lb. (a) If this process occurs in a
nozzle where IF = 0, what is the change 42-60. These numbers may be used
of kinetic energy? If the initial velocity for other problems.

3. ENERGY RELATIONS

61. The heatby 10 lb. of each


rejected necessary to change the temperature of
of the following gases 200 Btu: nitrogen,
is 1 lb. through this temperature range while

helium, nrgon. For each gas, what is the the pressure is constant. What percent-
change of temperature (a) if pressure age error would there be in using the value
remains constant, (b) if volume remains of c v from Table I in this computation?
constant? 66. The constant pressure specific heat
62. While the pressure remains con- of a certain gas is c r = 0.3 +
0.0037’
stant, 3 lb. of a gas are reversibly cooled Btu/lb., where T is °It, applicable from
from 340°F to 60°F. Assuming constant 0°F to 600°F. The corresponding c, =
cP ,
determine the chnnge of entropy if the —.3 + 0.0037’.
Determine the heat to
gas is (a) air, (b) argon, (c) helium. raise the temperature of 3 lb. from lOT
• las. (a) -0.31 Btu/”F, (b) -0.161 to 5I0°F (a) at constant pressure, (b) at
Btu/°F, (c) -1.61 Btu/°F. constant volume, (c) Compute the mean
specific heat for the specified temperature
63. The same ns 62 except that F = C.
range. Ans. (a) 3S25 Btu, (b) 2925 Btu.
64. From the curves of Fig. 19, estimate
the mean specific heat of O- between 66. IIow much heat is necessary to raise
1000’F and 4000°F and find the heat the temperature of 10 lb. of CO: from
5

S42 APPENDIX A PROBLEMS


400 °F to 4000T during a constant pres- the work? (c) Determine AH (d) In
sure process if tegrate - [V dp for p\ k = C and deter
e, = 0 276 + 1 4687 /10 - 0 994/T /*
s 1 mine its magnitude (e) What is Qf
Consider equation (m), § 42
4n* 10,590 Btu 4ns (a) A K
= 62 Btu, (e) Approx zero
67. What are the instantaneous specific 61 For the conditions of 40, namely,
heats at constant pressure and at constant - 6F + 20 psia, F, = 2, and F, = 3
p
volume, Btu/xnol, Btu/lb and ft lb /lb , cu ft, make the integral — fVdp and
at a temperature of 3SI0*F for each of the evaluate it in Btu What does this mte
following gases COj, N» air 7 By what gral represent when the change of kinetic
percentages do these values vary from energy is negligible?
those at normal atmospheric temperatures 7
62 If
(a) p = —28 8F 900 psia +
68 If the temperature of a substance while volume changes from 10 to
the
remains constant at 280°F while it recones 20 cu determine fpdV and — JF dp
ft ,

1480 Btu of heat in an internally reversible by integrations Sketch the curve on the
60
process, what is AS (a) for air, (b) for pF plane and designate the areas which
hjdrogen 7 Ant (a) 2 Btu/T represent these integrals Name the en
69 a reversible process occurs such
If ergies which these integrals represent
that T = 100 + 3Q0S* what heat, rev crsi- (b) If the system should be a diatomic gas
bly transferred, will change the entropy what heat is transferred 7
from 0 5 to 0 C Btu/T 7 Between what 4ns (a) 8GG Btu, 800 Btu (b) 1031 Btu

two temperatures does this process take 63 (a) During a certain reversible
place 7 Roughly sketch this process on process of 5 49 lb of argon Q = —164 5
the TS plane Arts 19 1 Btu Btu and A// — 21G5 Btu IVhat are the
The /p dr « 29 000 lb/min in a ft
values of -/l dp, At/,and JprfF 7 (b)
steady flow process of helium during which If this is a nonflow process what is the
74
work (on or by) 7 (c) If the temperature
the pressure changes from 75 to 25 psia
whilo the volume changes from 5 to 9 68 range is 40°F to 35G°F what is the mean
cfm (a) If this process occurs in a nozzle
specific heat of the process 7
4ns (a) -381 Btu 130 Btu -291
determine the change of kinetic energy and
the exit \clocity when A, - 0 Consider Btu (c) -0 005 Btu/lb °R
equation (k) §41 (b) If this process is 64-70 These numbers may be used for
applied to a turbine with AA’ - 0, what is other problems

4 THE IDE L GAS t

71. The temperature an ideal gas re-


of (b) hydrogen, (c) carbon dioxide (d)
mains constant while the pressure changes oxygen 4ns (c) 0 574 lb /cu ft

from 14 7 psia to 100 psig (a) If the An automobile tire contains 3730
initial volume was 3 2 cu ft , what is the cu in of air at 32 psig and 80°F (a)
final volume? (b) If the mass of gas is What mass of air is in the tire' (b) In
0 2 lb , determine the change of density operation this air temperature increases
Ans (a) 0 410 cu ft , (b) 0 425 lb /cu to 145°F If the tire is inflexible what is
ft increase the resulting percentage increase in gage
72 Hydrogen 400°R occupies 30
at pressure? (c) What mass of the 145°F air

cu ft (a) If the volume is increased to must be bled off to reduce the pressure
90 cu ft at constant pressure, what is the back to its original value 7
final temperature in °K and °C 7 (b) If 4ns (a) 0 504 lb (b) 17 5%, (c) 0 0o4 lb
,

the pressure is next tripled while the 76 A tank 6 in in diameter and 48 in


volume remains constant what is the long contained acetylene at 250 psia and
temperature in °F? 80°F After some acetylene was used
4ns (a) GG7 7 K (b) 3140T the pressure was 200 psia and the tempera
73 Compute the density of the follow- ture was 70°F (a) What percentage of

ing gases at 100 psia and GOT (a) air. the acetylene was used? (b) What v olume
appendix a. problems SJfS

would the used acetylene occupy at 14.7 81. Using the universal gas constant R
pda and GOT? and molecular weights, determine the gas
Ans. (a) 18.5%, (b) 2.38 cu. ft. constant in (a) ft-lb./lb-°R, (b) Btu/lb-°R,
76. A spherical balloon is 40 diam-
ft. in (c) Btu/mol-°R, and (d) atm-ft. 3 /mol-°R:

eter and surrounded by air at GOT and for air and hydrogen.

29.92 in. Hg abs. (a) If the balloon is Ans. (c) 1.986, (d) 0.73.

filled with hydrogen at a temperature of During a process of 4 lb. of air, the


82.
TOT and atmospheric pressure, what total temperature increases from 40°F to 340°F.
load can it lift? (b) If it contains helium Find (a) A U, (b) A H. (c) If Q — 0, what
instead of hydrogen, other conditions are the numerical values of jp dV and
remaining the same, what load can it lift? — fV dpi See §51.
(c) Helium is nearly twice as heavy as Ans. (a) 205.5 Btu, (b) 288 Btu.
hydrogen. Does it have half the lifting
83. During a certain process of helium,
force? Ans. (a) 2380 lb.
the state changes from p\ = 75 psia and
77. An engineer that for
estimates Vi = 5 cu. ft. to p 2 = 25 psia and V 2 =
proper ventilation of an industrial plant 9.68 cu. ft. What are A U and AH?
40,000 cfm of air are needed when the Ans. —37.4 Btu, —62 Btu.
outside temperature is 100T and the
atmospheric pressure is 14.3 psia. What
84. For a certain ideal gas, R — 35
ft-lb./lb-°R and k = 1.3. It is at 60 psig
volume of air is being handled as measured
and 90T. (a) What are c p and c,? (b)
under standard conditions of 60T and
What volume will 5 lb. of this gas occupy?
14.7 psia?
(c) If 20 Btu are added at constant volume,
78. A
tank contains air at 800 psia and what are the resulting temperature and
80T. After the removal of an amount of pressure?
air which measures 700 cu. ft. at standard Ans. (a) cr = 0.195, (b) 8.94 cu. ft.,
atmospheric pressure and GOT, the pres- (c) 117T and 63.7 psig.
sure is found to be 300 psig and the tem-
86. The same as 84 except that It — 77.8
perature 80T. What is the volume of the —
22 cu.
and k 1.4.
tank? Ans. ft.
86. The gas constant of a certain ideal
79. A closed vessel contains A
2 V =
gas is 400 ft-lb./lb-°R; c, = 0.8 Btu/lb.;
cu. ft. of air at GOO psia and 100T. This
vessel connects with B
which contains an id = 5 lb. From the state p = 100 psia
t

and Vi = 40 cu.
the gas undergoes a
ft.,
unknown volume of air Vb at 20 psia and
constant pressure process until the tem-
40°F. When
the valve separating these
perature becomes 1000°R. Find (a) c p
vessels opened, internal equilibrium of
is
and k, (b) A U, (c) A H, (d) Q, (e) A S, and
the mixture is established at 250 psia and
GOT. What is Vb? Ans. 2.67 cu. ft. (!) the jp dV. What does this integral
represent in a nonflow process? Does the
80. The volume of a Gxl2-ft. tank is nonflow energy equation balance?
339.2 cu. ft. It contains air at 200 psig Ans. (b) 2848 Btu, (c) 4680 Btu, (e)
and 75°F. How many
1-cu. ft. drums can 1832 Btu.
be filled G5°F if it is assumed
to 50 psig and'
87. The same as 86 except that 7? = 77.8
that the air temperature in the tank re-
mains at 75°F? The drums have been
and c v = 0.2.

sitting around in the atmosphere which is 88-100. These numbers may be used for
at 14.7 psia and 65°F. Ans. 992. other problems.

5. PROCESSES OF IDEAL GASES

note. In each of the following problems, indicate clearly whether the work is done on
or by the substance, whether the heal is transferred to or from the substance, and whether there
is an increase or a decrease of internal energy, enthalpy, and entropy. Moreover, a sketch of
the process on the pV
and TS planes and energy diagrams should be included all solutions m
101. Ten cu. ft. of air at 300 psia and ume. What is (a) the final pressure, (b)

400T is cooled to MOT at constant vol- the work, (c) the change of internal energy,
3U APPENDIX A PROBLEMS
(d) the transferred heat the change of
(e) (e) If the process nonflow and internally
is
enthalpy (f) the change of entropy? reversible what is the work’ (f) If the
Ans (a) 219 5 psia (c) —420 Btu (e) process is steady flow with Ik — 0 and
-587 Btu (f) -0 581 Btu/“It Ah. = —5 Btu what is Q?
102 600 Btu of heat are added to
If
Ans (a) -463 Btu (b) -648 Btu (e)
5 lb of helium at constant volume when -185 Btu
the initial temperature is 100°F, find (a) 108 A certain gas with c, = 0 529 and
the final temperature (b) AH (e) AS (d) R “ 96 2 ft lb /lb °R expands from 5 cu
AU (e) If (t) If this state change of the ft and SO'F to 15 cu ft while
p - 15 5
system is brought about by paddle work psia *= a constant Compute (a) Tt (b)
instead of heat what work is required and AH (c) A U (d) AS (e) For an internally
what are the changes in the properties* reversible nonflow process what is the
Draw an energy diagram Qualitatively [pdf and the work? (f) For an irrever
how do the entropy changes of the sur sible nonflow process during which 11 = 20
roundings compare for the reversible and Btu what is Q 7 (g) For a steady flow
irreversible processes F xplain process in which JF = 0 and AK = — 12
Ann (a) 232 5*F, (b) 828 Btu (c) Btu what is Q*
0 796 Btu/°R Ant (a) I160°F (b) 122 8 Btu (c>
106
94 Btu (d) 0 125 Btu/°Jl, (e) 28 7 Btu
103 Same as 102 except that the gas is
one for wl ich I? - 77 8 and k - 1 4 109 The heat transferred to a certain
quantity of air during a constant pressure
104 The pressure on 3 lb of air is de
process is 48 Btu What are AU All
creased at constant xolumc from 200 to
and 11 ? Ant 11 - 13 8 Btu
50 psia hat is AS?
Ant -0 713 Btu/°R 110 During an isothermal process at
88°F the pressure on 8 lb of air drops
The pressure on 5 cu ft of hjdro
from 80 psia to 5 psig For an internally
gen being heated at constant volume
reversible process determine (a) the Jp dV
changes from 40 psia to 200 psig Find (a)
the work (b) AU (c) Q (d) All (e) As
and the work of a nonflaw process (b) the
Btu/lb "It (f) If this constant volume
-/I dp and the work of a steady flow
process during which A K - 0 (c) Q (d)
process is irreversible and Q — 0 H — 399
Btu of paddle work what arc AU and At
AU and AH (e) AS
Ans (a) 421 Btu (e) 0 768 Btu/°R
of the hydrogen system* Draw an energy
diagram Qualitatively how do the cn 111 While its temperature remains
tropy changes of the surroundings compare constant at 80°F 2 5 lb of a certain gas
for the reversible and irmersible proc whose cp — 0 28 and c "02
Btu/lb °R
csscs’ Explain rejects 300 Btu pi = 85 psia Determine
Ana (b) 399 Btu (d) 500 Btu (e) (a) l i V , and pi (b) IF for a reversble
4 09 Btu/lb °R nonflow process (c) AS (d) A V and AH
(e) the change of kinetic energy if (he
106 The temperature of 5 lb of air
process is a reversible steady flow during
decreases from 300° F to 40°F w hile p — C which IF = —350 Btu
Compute (a) AU (b) All (c) Q (d) AS
(e) Jpdl (f) —fl dp (g) For an inter 112 During the isothermal expansion
nally reversible nonflow process what is of 5 lb of air the gain in entropy is 0 552
the work* nonflow process is
(h) If the units W
hat is (a) the expansion ratio
irreversible with IV — 75 Btu what is Q Fi/Fi (b) the ratio pj/pi 7 Ans (a) 5
for the same state change (energy dia 113 Trom a state defined by 300 psia
gram)? (i) If the process is steady flow 100 cu ft and 240°F helium undergoes
with IF = 0 and AK = +12 Btu what an iscntropic process to 0 3 psig Find (a)
is the heat* (j) Does the energy equation Fj and t t (b) A U and AH (c) /p IT (d)
Q = AH /l dp/J balance’ — JV dp (e) Q and AS What is the
Ann (b) -312 Btu (d) -0 5025 work (f) if the process is nonflow (g) if the
Btu/°R, (e) 89 2 Btu process is steady flow and AK = +10 Btu?
Ans (a) 608 cu ft 247°F (b)
107 While the pressure remains con-
-5870 -9740 Btu (g) 9730 Btu
stant at 100 psia the volume of a sj stem
of air changes from 20 cu ft to 10 cu ft 114 The same as 113 except that the
What are (a) AU
(b) AH
(e) Q (d) As? gas 13 methane
" K 3

APPENDIX A. PROBLEMS 845


116.
During an isentropic process of Q — A// — fV dp/J.
(f) What are the
4 lb. of air, the temperature increases from works of a reversible nor, flow process and
40°F to 340°F. Compute (a) AU and AH, of a reversible steady flow during which
(b) fpdV, (c) -fVdp, (d) Q and AS. A = —20 Btu? Does the area “be-
What is the work (e) if the process is non- hind” the curve represent this work?
flow? (f) if the process is steady with Explain.
AK = — 10 Btu?
For an irreversible (g) Ans. (a) 1.295, (b) 130, 216.5 Btu, (e>
adiabatic process from the same initial -164.5 Btu, (f) -294.5, -361 Btu.
state to the same final pressure, the final 122. A polytropic process of air from
temperature is 400°F. What are the
150 psia, 300°F, and 1 cu. ft. occurs to
works of the nonflow and steady flow
pt = 20 psia in accordance with pF = C. 1

( AK = —10) processes and the change of


Determine (a) f 2 and Fj, (b) AU, AH,
entropy? Is fp dV/J — — AU? Find m, and AS, (c) fp dV and — JV dp. (d)
p. 79, and evaluate the integral for Compute the heat transferred from the
pF" = C. polytropic specific heat and check by
Ans. (a) 205.5, 288 Btu, (g) -247, energy equations Q = AU
—335.5 Btu. fp dV/J and +
Q = AH — fV dp/J. (e) Find the non-
116. The same as 116 except that the flow work, and (f) the steady flow work
gas is argon. for AK = 0.

117. From and 4 lb.


Ans. (a) 17°F, 4.71 cu. ft., (b) —25.85,
85.3 psig 0.5 cu. ft.,
-36.2 Btu, +0.0142 Btu/°R, (d) 8.62 Btu,
of an ideal gas, for which cp = 0.124 and
= 0.093 Btu/lh-°R, undergo an isen- (e) 34.45, 44.8 Btu.
Ce
tropic process to a volume of 3 cu. ft. The nonflow work of reversibly
123.
What are (a) 7? and f2 , (b) A U and AH, (c) compressing a certain amount of air in
fpdV, (d) -fVdp, (e) Q and AS? (f) accordance with pF 32 = C is 60,000 1

What is the work if the process is nonflow? ft-lb. Compute (a) AU, (b) AH, (c) Q
(g) the process occurs in a nozzle.
If from the heat equation, (d) If a
specific
IF = 0, what is AK? Also compute reversible steady flow with AK = 0
l) 2 if K
is negligible,
i (h) The foregoing (instead of nonflow) takes place between
expansion between the stated pressures is the same what is the work?
states,
actually an irreversible adiabatic with a Ans.61.8 Btu, (b) 86.5 Btu,
(a) (c)
final temperature of 240°F. Compute -15.45 Btu, (d) -102 Btu.
IF' (nonflow) and AK' as asked for above. 124. The same as 123 except that the
Is fp dV/J = -AU? Find m, p. 79, gas is methane.
and evaluate the integral for pV = C. Ans. (a) 81.8 Btu, (b) 106.6 Btu, (c)
/Ins. (a) 24.1, 194°F, (b) -50.2, 67 5.12 Btu.
Btu, (h) m = 1.3.
126. If 1 lb. of air is throttled from 60
118. The same as117 except that psig to 12 psia, what is the change of
cp = 0.7 and c„ = 0.5 Btu/lb-°R. entropy? Ans. 0.1253 Btu/lb-°R.
During a polytropic process, 10 lb.
119. 126. The same as 126 except that the
of an ideal gas, whose R = 40 and c„ = gas is helium.
0.25, changes state from 20 psia and 40°F
127. Given a system A (see figure) which
to 120 psia and 340°F. Determine (a) n,
is 0.1 lb. of air at p, = 14.7 psia and
(b) AU and AH, (c) AS, (d) Q, (e) fp dV,
(f) —fV dp. (g) If the process steady
is

flow during which AK = 0, what is IF?


What IF = 0?
is AK if (h) What is IF
for a nonflow process?

120. The same as 119 except that


R = 155.6 and cp = 0.7.

121. A polytropic process of argon


occurs from 14.7 psia and 50 cu. ft. to
125 psia and 9.6 cu. ft. Find (a) n, (b) Problem 127.
AU and AH, (c) fp dV, (d) -fV dp. (e)

Compute the value of Q from the specific tt = 80°F. The movable piston, whose
heat equation and then check it from the area is 30 sq. in., does not permit the pas-
energy equations Q = AU fp dV/J and + sage of matter or heat. The only heat is
346 APPENDIX A PROBLEMS
Qa added to system A until the spring has done in displacing part of the mass in
been compressed from its free length (zero system B across its boundaries’ (e)
force) an amount L *= 6 in after which Sketch the pi cun e for system A What
the 6tate of A is defined by pt and it work is done by svstem A’ (f) What is
System B includes part of the atmosphere the final temperature of system A and its
and the spring whose scale is 500 lb /in change of internal energy’ Let the aver
Po = 14 7 psia and f» = 80°F which age specific heat at constant volume be
values do not change Make energy dia 0 2 Btu /lb °F (g) What is the heat
grams and write the energy equations (a) added? (h) W hat mass of air is displaced
for system A between the states 1 and 2 from the system B’
and (b) for system B while A changes from An* (c) 0 964 Btu (e) 1 249 Btu (f)
1 to 2 in the form of statement (5B) § 25 79 8 Btu (g) 81 Oo Btu
(e) How much energy is stored in the 128 140 These numbers may be used
spring (Btu)? (d) IV hat work (Btu) is for other problems

^^laE n IL15IS l\D the second l nr

141 During a Carnot cycle the work of isothermal expans on and the overall
ing substance receives 500 Btu at 540°F ratio of expansion (g) p„
and rejects heat at 140°F Find (a) the Ans (b) 64 Btu (g) 333 ps
thermal efficiency (b) the work (c) the
rejected heat (d) AS during the isothermal 146 Air is made to pass through a
processes Stirling cycle which conssts of two iso
Ans thermal processes and two regeneralue
(a) 40% (b) 200 Btu (d) 0 5
Btu/®R constant volume processes At the be
ginning of isothermal expansion p. - lOo
142 V Carnot power cycle operates on psia l „ - 2 cu ft and l, t
- 600®F
2 lb of air between the limits of 70°F and The ratio of o thermal expansion is
i

500®F The pressure at the beginning of 1 /l . - 1 5 and the minimum tempera


isothcrn al expans on is 400 psia and at tucc is (> 80°F Calculate (a) AS dur
the end of isothermal expansion is 185 psig ing the isothermal processes (b) Qa (<*) Qr
Determine (a) tl e volume at the end of the (tl) net If (e) e and (f) mep
l otl crmal compression (b) AS during an In* (b) 15 75 Btu (e) 49 1% (f)

isothermal process (c) Qa (d) Qr (c) B 41 7 psi


(f) e (g) the ratio of expansion during iso
thermal heating and the ov erall ratio of 146 From 20 psia and 80®F 10 cu ft of
expansion (h) the mop oxygen are isothermally compressed 1 2
to 80 psia next heat, is supplied at con
Ans 7 82 cu ft
(a) (b) 0 0948 (c)
stant pressure 2 3 unt I the entropy returns
91 Btu (d) 50 3 Btu (h) 15 88 psi
to its orig nal v aluc after which a reversi
143 During a single Carnot cycle the ble explosion 3 1 closes the cycle (a)
< hinge of entropy of the w orking substance Sketch this cycle on the pi and TS planes
during the addition of 400 Btu of I cat is and give equations for Qa Qr If e and
0 01 Btu/°R The efFcitncy of the cycle p„ (b) Find numer cal values of I T,
is 30% Determine (a) the temperatures Pj AS, j net II (from pi and T<S planes)
(®F) at which heat is transferred (b) the e and p«. (c) If the rate at which the gas
rejected 1 eat (c) How man} cj ties must » circulated through the cycle is 10 cfm
be completed per minute for an output at 1 what horsepower is being develope 1®
of 100 hp? Ans (b) AS, , = -0 095 Btu/®R H
-114 Btu e = 18 2% p m = 8 18 psi
144 Air is used as a working substance (c) 0 2G9 hp
in an Ericsson cycle Properties at the
beginning of isothermal expansion are 147 Ten cu ft of helium at 20 psia and
100 psia 5 cu ft and 540°F For a ratio 80®F are compressed isentropically 1 2 to
of isothermal expansion of 2 and a mini 80 psia The helium is tl en expanded
mum temperature of 40°F find (a) AS dur polytropically 2 3 with n — 1 35 to the
ing the isothermal process (b) Qa (c) Qr initial temperature Aw isothermal rejec
(d) net II (e) e (f) the volume at the end tion of beat 3 1 returns the hel urn to the
APPENDIX A. PROBLEMS 847
initial state, (a) Sketch this cycle on the (c) the net work, (d) If this cycle is used
pV and TS planes and give equations for for heating, what the COP?
is

Qa, Qr, IF (from pV and TS planes), e, Ans. (a) 92 Btu, (b) 8G°F, (c) -17.2
and p m .(b) Find numerical values for T ,, Btu, (d) 6.35.
F 2 pa, Qa, Qr, AU1-2, AUa-i, AU1-2-3-1,
,
163. A reversed Carnot cycle acts as a
net IF, and AS 3 .,.
heat pump to deliver 4600 Btu/min. to a
Ans. (b) pa = 9.52 psi, Qa = 36.55
heated space at 75°F, having received heat
Btu, Af/j.s = 41.4 Btu, IF = 9.1 Btu. from a 5°F cold body, (a) What are IF
148. An engine operates on 0.3 lb. of (in hp) and the COP? (b) If this same
air as a working substance, which is ini- system is used as a refrigerating machine,
tially at 25 psia and 100°F. There occurs what is its COP? (c) Prove that the COP
a constant volume heating 1-2 to 125 psia, of a reversed cycle used for heating, COP),,
is the COP of the same cycle used for
after which there is an isentropic expansion
2-3 to 25 psia. Finally, a constant pres- cooling, COP,, plus one; i.e., COPh =

sure process 3-1 returns the air to the COP, + 1.

initial condition, (a) Sketch this cycle


Ans. (a) 14.15 hp, 7.67, (b) 6 67.
on the pF and TS planes and give equa- 164. The layman sometimes points out
tions for Qa, Qr, IF (from pF and TS that there are about 19,000 Btu of heating
planes), e, and (5) Find numerical value in a pound of gasoline, that his
values for these quantities. automobile engine produces, say, 4000 Btu
Ans. (b) Qa = 115.3 Btu, IF = 28.3 work per pound of fuel, and that
of useful
Btu, p„ = 28.4 psi. when the engine is perfected, it will pro-
duce almost 19,000 Btu of work per pound
149. One lb. of argon at 20 psia and
of fuel. Refute this idea.
200°F (state 1) is compressed poly tropi-
cally with pF 1! = C to 100 psia (state 2). 166. From a source at a constant tem-
It is next cooled at constant pressure to perature of 3000°F, there are transferred
the initial temperature (state 3), then 2000 Btu to a system which accepts this
expanded isothermally to the initial state, heat at a constant temperature of 300°F.
(a) Sketch this cycle on the pF and TS The available sink temperature is 80°F.
planes and give equations for Qa, Qr, IF Determine the change of entropy (a) of
(from pF and TS planes), and p„. (b) the source, (b) of the system. What is the
Find the numerical values of Fi, Fa, Fa, net change? (c) Compute the available
Qa, Qr, IF, and p„. part of the 2000 Btu at the source tem-
Ans. (b) Fa = 2.32 cu. ft., Qa = 52.9 perature and at the system temperature.
Btu, IF = —8.2 Btu, p m = 6.24 psi. What is the change of available energy?
Does this check 7 'oAS? (d) Suppose that
150. In a reversed cycle, there is a work the system received the 2000 Btu of energy
input of 100 Btu/min., a work output of as paddle work (instead of from the
50 Btu/min., while 150 Btu/min. are source). What arc the net increases in
rejected. What is the heat added to the unavailable energy and entropy?
system and the COP if the cycle is used for Ans. (b) 2.057 net change, (c) 1108 Btu.
(a) refrigeration, (b) heating? (c) What
168. While the pressure is constant,
horsepower is required to drive the cj’cle?
Ans. (a) COP = 2, (b) COP = 3.
5 lb. of air at 140°F receive 1200 Btu of
heat, (a) If the sink is at 80°F, how much
161. A Carnot engine has a thermal of the heat supplied is unavailable? (b)
efficiency at 25% as a power engine. It is What cycle would convert all of the avail-
reversed and does 900 Btu of refrigeration, able energy into work (i.e., sketch on TS
(a) Determine the work and COP. (b) If plane)? What is its thermal efficiency?
this system is used for heating, what is the (c) Under what circumstance may all of
COPl Is more work required than in (a)? this transferred heat be converted into
Ans. (a) COP = 3, (b) COP = 4. ideal work? (d) If the energy received
by the system, 1200 Btu, had its origin as
162. The COP of a reversed Carnot cycle
paddle work (instead of heat), what would
is 5.35 when the refrigeration is done at
6 be the increase of unavailable energy?
0 F. The change of entropy during the
Ans. (a) 636 Btu, (b) 47%.
hothermal processes is 0.2 Btu/°R. Find
(a) the amount of refrigeration done, (b) 167. Steam at 45G°F with a specific
the temperature at which heat is rejected, heat of cp = 0.6 Btu/lb-°R is to be heated
848 APPENDIX A PROBLEMS
to 708°F in a heat exchanger by heat from temperature from 263 “F to 384°F (This
a gas whose c, = 0 24 and whose initial “preheats” the water on its way to the
temperature is 1500°F The rate of flow boiler and thus saves heat in the boiler
)
of the gas is 52 5 lb /sec , and of the steam If the system processes are internal^
25 lb /sec Let the flow be reversible reversible (no friction of flow), what is (a)
and compute (a) the final temperature of the net change of entropy due to the ir
the gas, (b) the change in entropy of the reversible heat flow and (b) the net increase
steam, (c) the change in entropj of the gas, of unavailable energy for a sink tempera-
(d) the net change of entropy (e) \\ hat ture of 40°F? (c) flow much available
is the net loss of a\ a liable energy of the energy is lost to the sink in the gas leaving
system for TV = 530°R? (f) How does the economizer (at 38G°F)? (d) If the
the entropy change in any actuat process? hot gas (767°F) is discharged to the sink
Ans (a) 1200-F, (d) 1 54 Btu/°R-sec ,
(without going through the economizer)
(e) 816 Btu/sec what is the loss of available energy?
158 \n economizer which may be Consider that the heat is rejected at con-
taken as a thermally isolated system and stant pressure Has available energy
heat exchanger, has a through flow of been conserv ed by use of the economizer?
Ans (b) 7 G X 10« Btu/hr (d) 106 X
405,800 lb /hr of water and 536 000 lb /hr
10* Btu/hr
of flue gas whose cr = 0 24 Btu/lb-°R
The heat from the cooling of the gas from 169-170 These numbers may be used
7C7“F to 386°F goes to raise the water for other problems

7. COMPRESSION AND EXP N SION OF GASESI

note In general, tl uould be beet to uork these problems by using constant specific heats
and then to check the results by using the air table, p 146 While the differences tn the
ansuers so obtained for Ike low temperatures invoked should be small some difference is to
be expected

171 \n indicator with a GO-lh spring going situations, would it take more or less
is used for cards from a double-acting or the same work to compress the same
12x16 in compressor running at 200 rpm volume of hydrogen’
The length of all cards is 3 5 in the aver Ans <i = 442°F,
(a) 143 5 hp, (b)
age area of the head-end cards is 2 55 139 5 hp, 109 5 hp (d) 159 hp
(c)
sq m and of the crank end cards 2 5 173 Hydrogen enters a compressor at
sq in The diameter of piston rod is 2 in 80°F after a compression according to
Compute the indicated mep s on each side pV l ** = C (a) If 20 lb /mm are han
of the piston the corresponding lhp s, and died what is the conventional horsepower
the total ihp of the compressor? (b) Would it take

172 The capacity of a compressor is


more or less or the same work to compress
= 1000 cfm measured at intake where the same mass of air between the same
I i'
temperatures? Ans (a) 544 hp
pi — 14 psta and h = 80°F Discharge
is at 84. psia. with. negligible change of 174 A centrifugal air compressor han
kinetic energy (a) For iscntropic com- dies 5000 cfm measured at intake where
pression determine the mass of air de- pi = 14 psia and <i = 75®F Discharge
livered (lb /min ), the temperature at the is at 40 psia A ~ 0 What is the
K
end of compression, and the conventional work if the process is (a) an irreversible
work and horsepower (b) The same as adiabatic for which m= 1 48, (b) lsen
(a) except that the compression is poly tropic, an internally reversible poly
(c)
tropic with re = I 35 (c) The same as tropic with n = 1 32? (d) What is the
(a) except that the compression is iso increase of availability of the air in (a)
thermal (d) The same as (a) except that and in (b)?
the process is an irreversible adiabatic Ans (a) -
18 400 Btu/min , (b)
with m = 1 45 (e) In the various fore -15 850 Btu/min , (c) -15 500 Btu/min
K 5

APPENDIX A. PROBLEMS S/,0


176.
Measured at intake of 14.6 psia anee is 8%, A = 0, and 42.4 lb. /min
and 80°F, 28,000 efm of air are pumped to are to be delivered. Find (a) the required
28.5 psia. At discharge the velocity is shaft horsepower, (b) the displacement ns
100 fps and the temperature is 200°F. If obtained from the conventional volumetric
the initial velocity is negligible and the efficiency, (c) Comment on the magni-
process is an irreversible adiabatic one, tude of the discharge temperature. Would
find the horsepower required to drive the there be danger of igniting the cylinder
compressor, mechanical friction neglected. lubricant? Is the volumetric efficiency
Ans. 1400 hp. good or bad? What alternative would
176. Same as 176 except that at dis- you suggest as a hotter design for actual
charge pi = 38 psia and h = 270°F.
u«e? Ans (a) 225 blip, (b) 905 cfm.

177. Consider that


3500 cfm of air at 181. The same as 180 except that the
gas is helium.
14.5 psiaand 75°F (the state at intake and
also the dead state) enter a compressor 182. An electric motor is delivering
and are discharged at 25 psia and 100°F. 170 hp to the shaft of a double-acting
The initial velocity is zero and the final reciprocating compressor which is turning
velocity is 120 fps. The process is an 270 rpm and compressing propane from
irreversible adiabatic following pV m - C. 14.7 psia and 90°F. Bore by stroke is
Find m, (b) AH, (c) IF, (d) AS. (e)
(a) 20x12 in.; volumetric efficiency is 80%;
What the increase of availability of the
is adiabatic compression efficiency is 78%;
air during compression? (f) What would AK — 0. Using the value of the fric-
be the increase of availability if the com- tional horsepower as given by equation
pression were iscntropic between the (I), | 121, estimate (a) the i hp of the
stated pressures? In which case is the compressor, (b) the mechanical efficiency,
increase of availability of the system (c) the iscntropic work of the correspond-
greater? Is this what you expected? Is ing conventional card, Btu/min., (d) the
the greater work of the irreversible com- mass of gas delivered, (c) For the fore-
pression compensated fully by its greater going conditions, compute the discharge
increase of availability? Discuss. temperature after iscntropic compression.
Ans. (a) 1.555, (b) 7070 Btu/min., Ans. (a) 149 ihp, (c) 8090 Btu/min.,
(c) —7144 Btu/min., (d) 2.39 Btu/°R, (e) 408°F.
(e) 5790 Btu/min.
183. A two-stage compressor without
178. A
reciprocating air compressor with clearance receives 80 Ib./min. of air at
a elearanee of 4.5% is to handle 1000 cfm 14 psia and G0°F and delivers it at 120 psia
of free air when atmospheric conditions The compression is polytropic, pF = f, 1

are p„ = 14.6 psia and l„ = 7G°F. At and the intercooler cools the air back to
the end of the suction stroke, the air is at G0°F. For each piece of equipment.
Pi = 14 psia and I, — 95°F. For a dis- AK = 0. Find (a) t : t ( and the optimum
, ,

charge pressure of 65 psia, for n = 1.33 intermediate pressure, (b) the conven-
and AK = 0, what should be the displace- tional work, (c) the heat transferred in the
ment of the compressor? various processes (sketch on TS plane),
Ans. 1200 cfm. (d) the iscntropic horsepower for a single-
179. A reciprocating air compressor stage machine, (e) the corresponding per-
with a 6% clearance pumps 50 lb. /min. of centage saving for the two-stage machine,
air from 14.7 psia and 80°F to 50 psig. (f) the mass of water to he circulated
The compression process is polytropic with through the intercooler if its temperature
n = 1.33. With respect to the system of rise is 15°F.

50 lb. /min., compute (a) Ti, (b) All, (c) Ans. (a) 41 psia, (b) 16! hp, (r) 4169

Q, (d) IF in hp. (e) Determine the Btu/min., (d) 199.5 hp, (e) 17.8%, (f)
required displacement ns based on the 18S lb./inin.
conventional volumetric efficiency. 184. (n) A compressor
delivers 50 lb./
Ans. (a) 320°F, (b) 2SS0 Btu/min., min. of receiving it at G5°F and com-
air,
(d) 78.2 hp, (e) 775 cfm. pressing it isentropirally to 786°F. Draw
180. A reciprocating compressor is ex- the energy diagram and determine the
pected to have an adiabatic compression work, For an adiabatic compression
(b)
efficiency of 77% when pumping air from 75%, what is the actual horse-
efficiency of
1 1.7 psia and 65°F to 300 psia. The clear- power and the temperature at the end of
)

S50 APPENDIX A PROBLEMS


the actual compression (same state 1)? loss of Ea by the surroundings and the
(c) During the actual compression, the net loss of £«?
system (50 lb /min ) undergoes an increase A ns (a) 8660 BUi/min , (b) 272 hp.
of available energy and the surroundings 1030°F, (c) 10,430, 11,520, 1000 Btu/min
experience a loss of available energy 185-190. These numbers may be used
What arc the gain of E„ by the system, the for other problems

S. THE GAS TURRUSE Ai\D TURBO-JET


note Because the air table alltnis for the variation of specific heat and because of the
high temperatures tn gas turbines, the results are much more accurate if the air (able is used
Houever, there is something to be said for the practice in using specific heats Both methods
of solution are desirable Answers are for table solutions unless otherwise indicated tn
the statement

191. (a) If the highest permissible tem- and a maximum temperature of 1100°F
perature in a gas turbine is 1640“F and For 1 lb of air in an ideal c>clc, determine
ti = 70°F, what compression ratio and (a) the approximate (that is for con
pressure ratio result in the maximum ideal slant c v ) temperature and pressure at the
work (constant specific heats) 7 (b) For end of compression for maximum work,
the ratios found in (a), what arc the net (b) the heats added and rejected, (c) the
work and thermal efficiency? (Note that works of the compressor and turbine, (d)
while ptfpt ” pi/pi, it is not true that tbc net work (from 1F — 1FC and 2(3), (c)
(

vi/vt — vi/vi when specific heats vary the thermal efficiency (f) If 1 lb /min
(o) If the compression ratio found above of fuel oil (yi = 17,400 Btu/lb ) is eon
is doubled, what would be the work, and sumed, what is the corresponding ideal
thermal efficiency of the ideal air cycle amount of air and of air and fuel lb /min 7
7
(minimum and maximum temperatures What is the consequent net work in hp
the same)? What percentage is this work Ans (a) 458°F, 95 7 psia (d) 66 35
of the maximum work 7 Btu/lb , (f)107 lb /mm 169 5 hp
,

Ans (a) 5 44 10 52, (b) 120 9 Btu/lb ,


47 2%, (c)e = 58 4% 196 The same as 194 except that the
maximum temperature is 1540°F
192 A Brayton cycle has an adiabatic
compression ratio of 4, the process begin- 196 Let the basic data be as given in
ning at 80°F and 15 psia At the end of 194 The compression efficiency is 82%,
(i = 1080°F the engine efficiency is 84%
the heating, Considering Determine
the working substance as 1 lb of cold air (a) the actual enthalpy k, after compres
(fc is constant at 1 4) and using specific sion (b) the actual enthalpy h, after
heats compute (a) U and pj (b) the works expansion, (c) the net work IF , (d) the
of the compressor and turbine, and the net actual thermal efficiency What is the
work, (c) Qa, Qr, ~Q (d) Find the mep ratio of the actual e' to ideal e T
from the work computed above Also
197 The following data correspond
write an equation for the mep as obtained
approximately to those for a General
from the pV plane and solve it for p„
Electric gas turbine 4690 bhp qi = 17 400
(e) Is this cycle the one for which the work
Btu/lb intake at 14 5 psia and 80°F,
is a maximum for the specified temperature
,

pressure ratio =* 6, turbine inlet tempera


limits? Explain (f) If the flow of air -
ture = 1400°F, compression efficiency
is 100 000 cfm at the initial state, what
ideal horsepower is being developed?
83%, engine efficiency = 85% For the
air cycle with friction, find (a) actual net
Ans (a) 480°F, 104 2 psia, (b) IF =
work of the fluid 11 hr, and «' (ignore the
61 2 Btu/lb , (d) 17 9 psi, <f) 10 850 hp
effect of the mass of fuel) (b) fit , (« , (c)
193 The same as 192 except that the the percentage of ideal net work lost by
working substance is 1 lb of hot air with virtue of the imperfect processes in com-
an average ft = 1 32 pressor and turbine (d) Assume that
fluid
194 The design of a gas turbine is to be the shaft (brake) work is 95% of the
based on an intake at 14 7 psia and 8Q°F, work computed above and determine the
APPENDIX A. PROBLEMS 851

air How, Ib./sec., (e) fuel u-cd, ll)./lir. and and I, = I12°R. The engine thrust is
ll)./I)hp-lir., (f) the heal rate, Btu/hhp-hr. 0200 lb.; air flow is 79 lb. /sec.; the ram
coefficient is S0%; r r ~
1.2 in compressor;
198. In a pa turbine unit, nir from the
1
!
turbine inlet temperature is 1500°F;
compre—or enters the eomhustor at I00°F engine efficiency of turbine is 75%. Find
and leaves at 15I0°I\ Of a fuel with (a) tlie propulsive power, (b) the exhaust
i/i = IS, 000 Btu/lb., 5 %
passes through
jet velocity relative to the plane. For the
unhurned The heat loss from the com- ideal cycle, compute (e) the pressure at
bustor is S00 Btu/lb. of fuel. Sketch an
entry to the compressor (stagnation effect),
enerpv diagram shou ing a complete energy
(d) ideal compressor work, (e) ideal ex-
halanee. Now ignoring the enthalpy of haust temperature of the turbine, (f) the
the liquid fuel and the effect of the mass ideal enthalpy drop in the jet norrle and
of fuel, and treating the products ns though the ideal relative velocity at exit from jet
they have the properties of nir, determine nozzle (expansion to atmospheric pressure
the amount of fuel per pound of nir, and from zero initial velocity).
the efficiency of the combustor. .Ins. (a) 5900 2330 fps, (c)
lip, '(b)
.Ins. 0.0177 lb., 90.5%. (d) 00.95 Btu/lb, (e) 1277°F,
7.71 psia,
A jet-oneiric driven plane is moving
199. (f) 2710 fps.

at a constant speed of 700 mph at an 200-210. Th esc number!) may be used


altitude of 00,000 ft. win re = -LOG p-ia for other problems.

o. i\Ti:nx.it. comiwstiox rxcrxr.s

NOTE, Unless othenri’e tinted, Ihr answers In the following problem! arc for some average
constant specific hentt. II owner, wlnre appropriate, it would hr interesting and instrnctnc
to rohe thrni also In/ use af the air table (the lariahte specific heat air standard), p. t/,0. If
the Gas Tables arc at hand, ter the examples in the had: for the method of nulling a real
mixture analysis (rxrlusiir of dissociation effects), if it is desired to male such an analysis.
In an Otto air cycle, the computed temperature at the end of the heat addition process is
lil.cly to go beyond the top limit of the air table for normal air /furl ratios.

211. An Otto cycle operates on 0.1 ment per cycle nnd the mcp. (f) If this
lb./scc. of air from 13 psia and I30°F at is a G-cylinder, -1-strokc-cyclc engine,
the beginning of compression, state 1. turning at 300 rpm, determine the number
The temperature at the end of combustion of cycles completed per minute nnd the
C
is 5000 R; compression ratio is 5.5; hot ideal horsepower.
nir standard, /: = 1.3.
r
(a) Find c„ l i, pr, A ns. (a) 9.09, (c) G510°It, H70 psia,
Pi, Fj, ti, and p,. (b) Compute (d) c = •19.5%., (c) p„ = 210 psi, (f) 811
Qa, Qr„ IF (from TS and pV planes), c, hp.
and the corresponding horsepower.
An*. (a) pi = GOo psia, 7\ = 3000’R, 214. The same as 213 except that the
(b) Qa = 91. G Btu/sec., c — 10%, 51.8 hp. air table is to be used in the solution (do
not me = 1.311. Compare answers.
212. What would lie the efficiency of the
An*, G529°R (obtained by
(n) 8.08, (c)
cycle in 211 if the uorkirig substance had
extrapolation of unabridged table), 1300
the properties of cold helium at all times?
psia, (d) c - -18 3%, (e) pm = 212 psi,
Is there any advantage in an Otto cycle
(f) 835 hp.
of such a substance?
213. From
state 1 of 100°F, 13.8 psia, 215. A Diesel cycle operates with a com-
and 1 cu.the compression in an Otto
ft., pression ratio of 13.5 and with the cutoff
cycle is to 250 psia. The heat supplied occurring nt G%
of the stroke. State 1 is
per cycle is 80 Btu. Using an average defined by 11 psia, 140°F. For the hot-
h = 1.31 for the hot nir standard, find (a) air standard with k = 1.34 nnd for an
the compression ratio, (b) the percentage initial 1 cu. ft., compute (a) fj, pi, F», h,
clearance, (c) T, and p, at the end of com- pi, Fi, pi, and l,, (b) Qa, Qa, (c) IF com-
bustion, (d) Qa, IF, and c, (e) the displace- puted from both TS nnd pV planes, (d)

352 APPENDIX A PROBLEMS


e and mep (e) For a rate of "circulation (c) the brake and indicated mep’s (d)
of 1000 cfm, compute the horsepower What is the heat rate, Btu/bhp-hr m and
An* (a) pi = 427 psia, /, = 805 °F, in Btu /min 7
(c) 10 Btu, (d) p„ = 58 6 psi, (2) 237 hp Ans (a) e. = 36%, (b) ij, = 774%,
216 In an ideal, single-cylinder, 2- (c) pwj, = 134 pst, (d) 15,700 Btu/min
stroke-cycle Diesel engine, operating at 220 A
12 5xl6-in , single-cylinder, 2
300 rpm, compression is from 14 7 psia, stroke-cycle gas engine develops 40 bhp
80’F, 1 43 cu ft to 500 psia Then at 200 rpm while using a natural gas with a
16 Btu/cycle is added asheat Makccom- lower heating value of 1000 Btu/cu ft
putations for the cold air standard and Its compression ratio is 6 and its mechant
find (a) p, V, T
at each corner of the cycle, cal efficiency is 81%, and at the stated
(b) TF, (c) e and the mep, (d) hp load, the heat rate is 10,000 Btu/bhp-hr
Ans (a) T, = 2120°R, (b) 9 7 Btu, Let the hot air Otto cycle, k = 1 33, be
(c) 60 7%, 39 9 pst, (d) CS 6 hp the standard of comparison Determine
217. The name as 216 except that com- the brake and indicated thermal efficien-
putations are to be made for hot air with cies, engine efficiencies, and mep’s

k = 133 An* Ci ** 31 5%, >i. = 70 6%, p m , -


218. The charge in a Diesel engine con-
sists of 0 015 lb of fuel, whose lower heat- 221. A 12 5xl3-in , single-acting 2-

ing value is 17,400 Btu/lb and 1 2 lb of strokc-cycle oil engine is running at 326
,

air and products of combustion From rpm An indicator with a


spring 350 lb

t,
o
100 F, the compression ratio is 12 produces a card with an average area of
(a) Compute the cutoff ratio of the corre- 0 39C sq in and length of 2 97 in Crank
sponding hot air standard when c, = 0 27 case compression is used for scavenging
Btu/lb (b) What is the thermal effi- An indicator card from the crank case
ciency and the work done’ has an area of 0 5 sq in and a length of
Ans (a) 2 78, (b) 47%, 368 Btu/charge 2 97 in when the spring scale is 12 lb
of fuel Mechanical efficiency has been determined
as 8o% Compute (a) the mep of the
219 A
6-cylinder, 3 5 £x3 C*m automo- working cylinder and of the crank case,
tive engine with a compression ratio of 8, (b) the corresponding lhps and net ihp,
shows a fuel consumption of 0 45 lb per (c) the bhp
bhp-hr at 3000 rpm, bhp — 113,fhp = 20 An* (b) net ihp «= 58 4, (c) 49 6 hp
(let lhp = bhp + Let the standard
fhp)
222 The mep of an ideal hot-air
for comparison be the hot
air standard
with k =
1 3 The heating value of the (L = 1 34) Diesel cycle is 100 psi pi = 14
fuel is qi — 18,500 Btu/lb Compute (a) psia, tu = 12 What is r,7 An* 2 47
the mechanical efficiency, *4, and e., (b) 223-230 These numbers may be used
the brake and indicated engine efficiencies. for other problems

10 LIQUIDS AND VAPORS

NOTE given to the problems on steam are based on Steam Tables published
The answers
by Combustion Engineering Inc If other tables are used some differences in answers are
to be expected Combustion Engineering may be willing to furnish their tables for loan to
students A topy tff tfcew {obits vs found vn FrwbUivns. ww. Thermodynamics la same
of the problems, also in later chapters, the data are chosen for a comement fit
unlh these
tables

231 (a) What are the temperature, 232 (a) A 10-cu ft drum contains
volume, enthalpy, entropy, and internal saturated steam at 8o 3 psig What are
energy of 3 lb of saturated steam at the temperature and mass of steam in the
350 psia? (b) The same as (a) except drum 7 Determine the enthalpy, entropy,
that the steam is at 350 psia and 520°F and internal energy (b) The same as (a)
(c) The same as (a) except that the steam except that the drum contains 3 lb of
is wet with 10% moisture steam (c) The same as (a) except that
the mass is 1 241 lb of gteam
APPENDIX A. PROBLEMS S5S
233. For practice in interpolation, find lb. (c) The same as (a) except that
the enthalpy, volume, entropy, and h = 629.9 Btu/lb.
internal energy of 1 lb. of steam at 462 psia 238. (a) Compressed liquid water is at
and G25°F.
3000 psia and 200°F. For 1 lb., what are
234. (a) Steam at 500°F has a specific its enthalpy, entropy, and volume? (b)
entropy of 1.815 Btu/lb-°R. What are Saturated water at 200°F is pumped to
its pressure, specific volume, and en- 3000 psia. Find the approximate increase
thalpy?
236. (b) The same as (a) except that of enthalpy (pump work), using fv dp.
a = 1.4 Btu/lb-°R, (c) The same as (a) Show these points on a large-scale Ts
except that s = 1.433 Btu/lb-°R. plane.
(a) Steam at 120 psia has a specific 239. The same as 238 except that the
enthalpy of 1410 Btu/lb. What are its temperature is 600°F.
temperature, specific volume, entropy,
240. (a) Water at 3000 psia and 300°F
and internal energy? (b) The same as enters a steam generator and leaves as
(a) except that h = 1190.6 Btu/lb. (or
superheated steam at 3000 psia and 800°F.
1190.4 for Table X). (c) The same as
How much heat is added per pound? (b)
(a) except that h = 1000 Btu/lb.
Saturated water at 300°F is pumped to
236. (a) Ammonia at 100°F has a 3000 psia and then heated to steam at
specific volume of 2.8 cu. ft./lb. Deter- 3000 psia and 800°F. How much heat is
mine its pressure, enthalpy, entropy, and added? See equation (j), § 1G8. (c) The
internal energy, (b) The same as (a) same (b) except that p — 85.3 psig.
as
except that v = 1.419 cu. ft./lb. (c) The (d) Is it advisable in either (b) or (c) to
same as (a) except that v = 1.2 cu. ft./lb. account for the enthalpy increase during
237. (a) Ammonia at 140 psia has an
compression of the liquid?
enthalpy of 650 Btu/lb. What are its Ans. [from K
& K] (a) 992.1, (b) 993,
(c) 1159.2 Btu/lb.
temperature, volume, entropy, internal
energy, and degrees of superheat? (b) 241-260. These numbers may be used
The same as (a) except that h — 600 Btu/ for other problems.

11. PROCESSES OF VAPORS


reminder. The student should sketch all processes on the Ts and/or pv planes and
sketch energy diagrams, whether or not the problem says so.

261. An internally reversible isobarie 253. Ammonia at 20 psia with a specific


process occurs from water at 200°F and enthalpy of 136.8 Btu/lb. changes state
160 psia to 600°F. (a) For 1 lb., deter- at constant pressure until the temperature
mine Ah, As, Ac, and Au. (b) If the becomes 0°F. For 10 lb. of NHj, find (a)
process is nonflow, what are the jp dv/J, xi, AH, AF, AS, and AU. (b) What are
IF, and Q? Check by the nonflow energy the work and heat of a nonflow system?
equation, (c) For a steady flow process (c) What are the work and heat of a steady
through a steam generator with A — 0, K flow system of the N H > flowing through
what are the Jv dp and Q1 What is IF? an evaporator in which A K = 0?
Ans. (a) A« = 1042.7, (b) IF = 113.4, Ans. (a) AF = 114.8 cu. ft., (b) 425
(c) Q = 1156.1 Btu/lb. Btu, 4787 Btu.
262. Steam with an entropy of 1.302 264. Five pounds of steam occupy a
Btu/lb-°R undergoes a constant pressure 14.02-cu. ft. tank at 190 psia. After some
process at 400 psia until the final entropy cooling of the tank, its pressure drops to
is 1.6393 Btu/lb-°R. (a) Determine y u Compute AS, and AU.
80 psia. (a) yt,
Ah, An, Au. (b) If the process is steady (b) What is IF? What amount of heat is
flow through a heat exchanger wherein transferred? (c) At what pressure was
AK = —5 Btu/lb., how much heat is the steam in the tank dry and saturated?
transferred? (c) If the process is nonflow, Ans. (a) 7/2 = 49%, A U = —2309, (c)
what are fp dv/J, IF, and Q? What 161.9 psia.
statement of the law of conservation of
energy best describes the energy balance 266.Ammonia at 20 psia and Xi = 20%
in this on c? is trapped in a vessel and receives heat
854 APPENDIX A PROBLEMS
until its pressure rises to 160 psia For nonflow compression, what is the work?
1 lb , find (a) h As, Ah, and A u (b) (e) What is the work during steady flow
What are IF and Q1 (c) If the volume if A K— 0’ (d) For a compression effi-
of the vessel is 15 cu ft , what is the moss ciency [in (c)J of 78 2% and 4 lb /min
in it before and after this process? flow, what is the actual v ork (in hp) and

256 During an lsentropic process, 5 the actual discharge temperature of the


lb
of steam are compressed from 10 to 150 Nil,?
psia The initial total entropy is Si — -4ns j/i= 5 83%, A u = 107 4 Btu/
(a)

7 8490 Btu/°R (a) Compute the prop-


lb, (c) -1319 Btu/lb, (d) 15 9 hp,
erty changes AH, A V, and AU 260'F
(b) If the
process is steady flow with A = —100 A 260 During an irreversible adiabatic
Btu, what work is required? (c) What process from 250 psia and 420°F to p, =
arc the work and heat for a nonflow 5 3 in Ilg abs the entropy increases 0 3
,

process? Sketch the process on the Ts Btu/lb c Tt, A — 0 Calculate (a) y, and
K
plane after the solution is completed Ah, (b) IF and Q, (c) the work lost due to
Ana (a) y, - 14 46%, AU = 847 5 irreversibility, and (d) the increase of
Btu, (b) -865 5 Btu una\ ailable energy for a sink at 80°F
Ana (b) II ~ 125 7 Btu/lb, <c) 179
267. If 6 lb /sec of steam pass through
Btu/lb (d) 162 Btu/lb
an isentropic process from 200 psia and
450’T to 5 psia, find (a) {/>, AH, AU, and 261 During an isothermal compression,
AS (b) If the process is nonflow, what steam at 80 psia and 600°F rejects 781
are Q and 11 and fpdV/J ’ (c) If the Btu/lb (a) Compute x. Ah, and Am
process is steady flow with II — 0, what (b) If the process is nonflow, what is the
is AA? Check by Mother chart showing work’ (c) IVhat is the work of a steady
sketch of solution (d) If the process is flow process mth AA “ 0?
steady flow with AA — 200 Btu/sec A ns (a) i, - 45%, (b) -392 7 Btu/lb
what is IF’ The horsepower? (e) If (i) -315 5 Btu/lb
the process is an irreversible adiabatic 262 Saturated steam at 400°F expands
with the entropy increasing to a* = 1 8437 isothermally to 60 psia For 10 lb , (a)
Btu/lb °R, what are the nonflow work what arc AS, AH and A 17? Determine
and the steady flow work (A - 200 K Q and IF (b) for a nonflow process, (c)
Btu/sec )? for a steady flow with AA — 0
258 An lsentropic process of 10 lb of Ana (a) = 253 1 Btu, (b) Q -
AU
steam occurs from 175 psia and 4 231 1603 Btu
cu ft /lb to 20 psia Find (a) y, or f,, 263 Saturated steam at 130 psia ex
(b) All, A U, Q, and AS (c) What is the panda polytropically to 230°F in accord
work of a nonflow process? (d) For ance with pvl 11 = C (a) Compute i. Ah
A A = 0, what is the work of a steady flow As, and Au Now ’sketch the (process on
process’ Check by the Mollier chart, the Ta plane to match these computations
showing sketch of solution (e) If the (b) If this expansion occurs nonflon in a
process ts an irreversible steady flow steam engine cylinder, what arc IF and
adiabatic from the same initial state with Q during the process? (c) Consider the
the entropy increasing 0 0847 Btu/lb “R process as steady flow and apply the energy
what is the work (A A = 0)? (f) How equation to determine an expression for IF
much work is lost because of the irrevcrsi Solve for TF from this equation (AA - 0)
bikty? (g) In (c) what is the increase of and from [v dp Do they check’ (d)
unavailable energy for a sink temperature What is the change of availability of the
of 70°F? (h) Show areas on the 1 s plane steady flow s}stem for to = 70°F’ Does
which represent respective!} the lost work the unavailable energy of the surroundings
and the increase of unavailable energy increase or decrease’
Ana (b) AU = -1704 Btu, (d) 2336 /Ins (a) t>, = 17 cu ft /lb Au =
Btu, (e) 1631 Btu, (f) 705 Btu, (g) 449 -134 2 Btu/lb ,
(b) 117 2, -17 Btu/lb,
Btu (c) 135 Btu

259 Ammonia in a refrigeration ma- 264 Steam at 480 psia and 1000°F
chine is compressed iscntropically from expands polytropically to 80 psia and
For 360°F The mass is 10 lb (a) Deter
20 psia to 200 psia and 200°F 1 lb ,

find (a) yi, Ah, An, Q, and As (b) For a mine n, AH, All, fpdv, and —fir dp (b)
APPENDIX A. PROBLEMS 855
What are Q anti IE for a nonflotv process? versibly, until saturated vapor at
it is all
(c) Set up the energy equation and energy 150 psia. The mass 3 lb. (a) Deter-
is
diagram for a steady flow process. What mine n and the change of properties AH,
is Q for this process? If A K
= 10 Btu/ A U. (b) What is the nonflow work? (c)
lb., v, hat is the work? What does — fv dp What is the steady flow work if AK = — 15
represent in this process? Show a numeri- Btu? (d) How much heat is transferred?
cal check. If the flow is 10 lb. /min., into or out of the system?
what is the horsepower? (d) What is the Ans. (a) n — 1.09, A U = 116 Btu, (b)
unavailable portion of the heat 'trans- -216 Btu, (c) -221 Btu, (d) -100 Btu.
ferred for a sink at 500°R? Is it an in-
crease or decrease? Show area on sketch. 267. A throttling calorimeter receives
What is the change of availability of the a sample of steam from a steam main, in
266. in steady flow?
system which the pressure is 100 psia. After
Ans. (a) n = 1.49, AU
= -2.381 Btu, throttling, the steam is at 14.7 psia and
250°F. What the percentage moisture
fp dv = 111,500 ft-lb./lb., (b) -946, 1435
is

Btu, (c) 48.1 hp, (d) -421, -2664 Btu. in the original steam? Solve by using
(a) the superheat tables, (b) the saturated
Starting from 30 psia and 10°F, steam tables and the specific heat of steam,
ammonia is compressed to 160 psia during
(c) the Mollier chart, making a sketch to
an internally reversible process for which show your solution.
pi/
1 255
- = C. (a) Calculate Vt, Ah, Au, As.
Use the nearest whole degree for /?. (b) 268. The same as 267 except that
Find IF and Q for a nonflow process, (c) throttling occurs from 160 psia to 15 psia
Find IF (from [v dp) and Q for a steady and 250°F. Ans. 3.06%.
flow (A K = 0). How do these values of
269. In a refrigerating system, liquid
Q compare? If the rate of flow is 20 lb./ ammonia enters the expansion (throttling)
min., what power is required? (d) What
is the change of availability of the steady
valve at 150 psia and 70°F. If it emerges
at 25 psia, compute (a) x - and As; (b)
flow system when To — 500°R?
also the loss of available energy during
Am. (a) U — 210°F, Au = 74.7 Btu/
throttling.
lb., (b) -83.6, -8.9 Btu/lb., (c) 49.5 hp.

266. Ammonia at 20 psig and 5% liquid 270-280. These numbers may be used
(y = 5%) is compressed, internally re- for other problems.

12. VATOR CYCLES AND ENGINES


REMINDER. Show sketches for each problem depicting the processes on some convenient
plane and sketch energy diagrams for steady flow problems. Use the Mother chart as a
check on the steam table solution wherever possible.

281. (a) A Rankine engine receives 1 ib. (a) For a Rankine engine operating
saturated steam at 110 psia and exhausts between these limits, compute its work,
it at 15 psia. For 1 lb., calculate the work, thermal efficiency, steam rate, and mep.
(b) Suppose the exhaust from the fore- (b) Considering that a Rankine cycle
going engine is passed without loss through occurs between the same limits, determine
a steam separator, which removes all the Qa, Qr., net TF, and e. How much is the
liquid. Suppose norv that the remaining ideal pump work? Indicate areas on the
saturated steam at 15 psia enters another Ts plane which represent Qa, Qr., and IF.
Rankine engine and expands to 1 psia. (c) What mass of steam is required for a
How much work is done by 1 lb. ? How net output of 50,000 kw?
much for of original steam?
1 lb. (c) Ans. (a) 492.1, 37.16%, 5.17 Ib./hp-
Does it appear that the work to be ob- hr., 7.11 psi, (b) e = 37.09%, IF* — 1.605
tained by carrying the expansion of steam Btu/lb., (c) 347,000 lb. /hr.

below atmospheric pressure is significant? 283. The same as 282 except that the
Is a low pressure of 1 psia feasible? expansion is from 2000 psia and 1100°F
Ans. (a) 145.6 Btu/lb., (b) 170, 151.1. to 110°F.

282. Steam is generated at 540 psia and 284. A turbo-generator unit consumes
75Q = F and condensed at 00°F. Consider 250,000 lb. /hr. ofsteam while delivering
856 APPENDIX A PROBLEMS
30 000 kw The steam expands from isas follow s head-end area is 2 17 eq in ,
800 psia and 900 °F to 1 psia (a) Con- crank end area is 2 34 sq in, average
sidering an ideal engine, determine the length of all is 3m, scale of indicator
thermal efficiency, steam rate, and mep spring is 100 lb , piston rod diameter is
(b) Considering 1 lb of steam an ideal m 1}^ m At this load, the mechanical
cycle between the same limits, compute efficiencywas 86% and the steam rate
the pump work, heats added and rejected, was 23 /bhp hr
lb (a) For the ideal
and the thermal efficiency (c) Consider- engine, compute IF and e For the actual
ing the actual engine, determine the com- engine, determine (b) e6 , u>,, and ij, , (c)
bined steam rate, combined work, com- ihp and the total steam flow lb /hr m
bined thermal efficiency, and the combined A ns (a) 21 1%, (b) 10 2%, 19 75 lb /
engine efficiency Also, estimate the ihp hr , 56%, (c) 127 7 ihp, 3410 lb /hr
enthalpy of the actual exhaust if the 288 A turbine whose steam rate is 6 1
generator efficiency is 93%
lb /bhp-br receives steam with an en-
Am (a) 38 9%, 6 34 lb /kw-hr , 10 7
thalpy of 1507 5 Btu/lb Radiation loss
psi, (b) Qa = 1382 8 Btu/lb 38 8%, (c) ,
from the turbine is 18 Btu/lb of steam
W K = 410 Btu/lb , = 76 2%, h. ~ and the condenser pressure is 5 3 in Hg
1013 9 Btu/lb abs (a) If Ah — 5 Btu/lb of steam,
285 In an incomplete expansion engine, what are the approximate enthalpy and
steam expands from 1G0 psia and 136 45 quality of the actual exhaust? (b) What
degrees of superheat to 30 psia which is is the stagnation enthalpy of the exhaust
the point release The exhaust pressure if the entering kinetic energy is zero 7 (c)
is 15 psia At full load, the actual engine If the total steam flow is 610,0001b /hr,
has a steam rate of 22 3 lb /bhp hr and what horsepower is being delivered?
ij«— 80% (a) For the ideal engine, Ant (a) 1067, 94 7%, (b) 1072 Btu/lb
compute the efficiency, steam rate, and 289 Saturated steam leaves a steam
mep (b) For the actual engine, deter- generator at 190 psia Because of fric
mine the brake thermal efficiency, the tional losses in the pipe and throttling at
indicated engine efficiency, and the heat
the turbine entrance, the pressure drops
rate in Btu/bhp-hr
adiabatically to 140 psia The sink tem
Ans (a) 16 4%, 14 22 lb /hp hr 73 ,
perature is 77°F What is the increase of
psi, (b) 10 45%, 79 7%, 24,400 Btu/bhp hr (a) the entropy and (b) the unavailable
286 A compound steam engine with energy 7 (c) What is the change of avail
incomplete expansion receives saturated ability? (d) What is the loss of Rankine

steam at 150 psia expands it to 15 psia work during this process 7 Indicate sig
where release occurs Exhaust is to a nificant areas on your Ts diagram
condenser at 104°F The actual engine Am (b) 36 75, (c) -10 75 (d) 21
uses 2700 lb /hr of steam, it has a brake Btu/lb
engine efficiency of 60% and a mechanical 290 Steam is delivered by a steam
efficiency of 85% The electric generator generator at 570 psia and 500°F Because
which it drives has an efficiency of 92% of frictional losses in the pipe and throt
(a) For the ideal engine, find the thermal tling at the turbine at reduced load the
efficiency and mep For the actual pressure drops adiabatically to 470 psia
engine, determine (b) the brake work in at entry to the nozzles The condenser
Btu/lb and hp the broke thermal effi- is at 104°F, brake engine efficiency is 72%
ciency, (c) the indicated work, thermal generator efficiency is 94%, work of ideal
efficiency, and steam rate, (d) the com-
engine is 388 2 Btu/lb sink temperature
,

bined thermal efficiency and steam rate is 80°F Solve by tables with a Mollier
(e) What is the approximate enthalpy of chart check (sketch) Determine (a) the
the actual exhaust Bteam? loss of available energy due to fluid fric-
Am (a) 20 35%, 54 psi, (b) 148 5 bhp, tion in arriving at the nozzles, (b) the
12 2%, (c) 14 35%, <d) 11 2%, 27 1 lb / approximate enthalpy of the actual
kw hr , (e) 1033 2 Btu/lb exhaust, (c) the loss of available energy
287. A double acting, 12x15 in , simple during passage through the turbine and
steam engine, turning at 200 rpm, expands the total loss (d) the overall loss of
steam from 185 psia and 400°F to 15 psia Rankine work
where the exhaust valve opens Exhaust Am (a) 9 77, (b) 940 2, (c) 102 9,
is at 5 psia Indicator card information 112 67, (d) 117 5 Btu/lb
APPENDIX A. PROBLEMS 857
291. Steam enters an ideal reheat tur- expansion to a condenser temperature of
bine at 1-tOO psia and 900°F and exhausts 90°F. Compute V/ r Qa, Q n and c. , ,

at 1 psia and a quality of 87%. The ideal Ams. 192 (approx.), 1112.1, G28 Btu/
295.
work is 051 Btu/lb. What is the ideal lb., 13.5%.
thermal efficiency? Draw an energy Indicate the following rehcat-
diagram of engine and reheater.
rcgenerativc cycle on the Ts plane and
Ans. -12.1%. set up equations for (a) the amount of
292. In a reheat cycle, steam enters the extracted steam at each point, (b) the
turbine at G30 psia and G50°F, expands heat added, (c) the thermal efficiency:
to 180 psia, whence it parses through a one reheat with two stages of regenerative
rchcater emerging at 700°F. It now feedwater heating after the reheat.
expands to the condenser temperature of 29S. Inan ideal reheat-regenerative
101°F. For the ideal cycle, compute cycle, steam expands from 1700 psia and
B'y, Qa, Or, and c.
700°F to 800 psia where some steam is
Ant. 1.91, 1121.8, 900.G Btu/lb., 3G.G%. extracted for feedwater heating and the
293. In an ideal regenerative cycle, remainder is reheated to 750°F. The
steam generated nt 520 psia and 900°F;
is steam expands again to 145 psia where
it then expands to 255°F and 32.53 psia additional steam is extracted for feed-
where m lb. are extracted for feedwater water heating, and the remainder expands
heating. Condensation is at 81°F. For to the condenser pressure of 1.202 psia.
1 lb. of throttle steam, find the fraction Set up the equations needed on the basis
extracted, the total pump work, the of 1 lb. and compute the following: (a)
enthalpy of water entering boiler, and the percentages of extracted steam, (b) Qa, Qr,
thermal efficiency. and net work (find net work by "ZQ and
Ana. 0.151, 1.51, 225.1, 10.9%. check by sum of Rankine engine works
294. In an ideal rehcat-regonorntive minus pump work), (c) thermal efficiency,
(d) Draw an energy diagram of the engine
cycle, steam at 1G50 psia and 7S0°F
expands to 130 p=ia, whence all of it passes only and show an energy balance ns a
check.
through a rchcater and re-enters the
turbine at G00°F. It then expands to
A ns. (a) 20.9%, 17.8%, (b) net IF =
393, (c) 15.7%.
135 psia, at which point some is bled for
regenerative feedwater heating. The re- 297-310. These numbers mny be used
mainder of the throttle flow continues its for other problems.

13. NOZZLES
311. A
substance flows nt the rate of 313. A large user of natural gas (let
2 through a nozzle from 110 psia
lb. /sec. Ti = 9G.2, h = 1.3) receive, it at 100 psia
and 500°F to 80 psia. The expansion is and 80°F but uses it at 15 psia. It is pro-
isentropic. the initial velocity is zero
If posed that the gas be used in a turbine to
(state 1 is the stagnation state), find the generate power. If it is decided that each
temperature and area at the exit section nozzle is to allow a flow of l.G Ib./see.,
when the substance is (a) steam, (b) air. determine the pressure, temperature, ve-
(c) The same as (b) except that the initial locity, and area (a) at the throat and (b)
velocity is 200 fps. Show energy diagram. at the exit section for an isentropic expan-
Ana. (a) 380°F, 1.0G8 in. 5 (b) 821. 5°F,
,
sion (ideal gas); initial velocity negligible.
0.871 in. 5 (air tables), (c) 0.833 in. 5 Would the state of the exhaust be such
that it could be used for cooling air for air
312. A nozzle expands steam in equilib-
conditioning purposes? Discuss.
rium from a stagnation state of 110 psia
Ans. (a) 4G8.5°It, 0.965 in. 5 , (b)
and 100°F to 5 psia. Throat area is
0.32G sq. in. Let p c = 0.515pi. For an
348.1^, 1.58 in. 5
ideal expansion, determine (a) the mass 314. Helium flows through a nozzle
rate of flow, and (b) the area of the exit whose exit area is and changes
2 83 sq. in.

section. state from 100 psia and 90°F to 80 psia.


Ana. (a) about 1800 lb./hr., (b) 1.39 The nozzle efficiency is 97 % and the initial
in. 5 velocity is negligible. What are (a) the
358 APPENDIX A PROBLEMS
exit temperature, specific volume, and ble velocity at 1050 psia and 650“F and
velocity? (b) the rate of flow, (c) the expands to 450 psia The actual exit
value of m
in p,F,» = p t tm, (d) the rate
V temperature is 400°F and the rate of flow
of increase of unavailable energy? is 2 lb /sec Determine (a) the actual
Ans (a) 508 5°R, 1703, 1687 fps, (b) exit velocity and nozzle area, (b) the
7030 !b /hr , (c) 1 54, (d) 91 7 Btu/hr nozzle efficiency, (c) the value of in m
315 A nozzle, whose throat area is pv m — C which defines the pressure-
1 sq in , expands steam from 200 psia volume relation at the end states, (d) the
and 400 aF to 50 psia The expansion is 317of available
loss energy for a sink tem-
into the supersaturated state Tor a perature of 40°F
nozzle efficiency of 95% and negligible Ans (a) I960 fps, 0 153 in *, (b) 96 4 %
initial velocity, compute (a) the actual (c) 1 255 (d) 3 2 Btu/scc

discharge, and (b) the exit area Assume The same as 316 except that the
that the actual specific volume is equal working substance is helium and the
to the ideal (closely so) actual exit temperature is 510°F
Ana (a) 10,230 lb /hr , (b) 1 293 in * 318-330 These numbers may be used
316 Steam enters a nozzle with negligi- for other problems

n. TIIE REVERSED C\CLE

331 A refrigerating cycle operates be (d) Freon 12, (e) lf«0 Are all of the
tween an evaporator temperature at 10°F answers reasonable? For COt
and a condenser pressure of 120 psia “ 98 4 hf sfl*F — 35 6 Btu/lb , p,4o*f ™
Entering the compressor the refrigerant, 0 145 cu ft /lb For SO, h„o’r -
which is ammonia, is 100% vapor and its 185 37, h„ 0‘r -* 40 05 Btu/lb , =
temperature on the high pressure side 2 887 cu ft /lb
of the expansion valve is G0°F For 1 lb Ans (a) 40 5 cfm, 6 67x6 67 in
in an ideal cycle find (a) the work (b)
334 A single acting twin cylinder,
the COF, (c) the rate of refrigerant circula-
12x12 in compressor running at 150 rpm,
tion for a 5 ton capacity, lb /min (d) ,
takes in saturated vapor at 1°F and dis
the heat rejected to the condenser water,
charges it at 180 psia Saturated liquid
Btu/min , (e) the horsepower per ton,
enters the expansion valve The com
(f) the piston displacement for a volumetric
pressor s volumetric efficiency is82% The
efficiency of 80% (g) For an adiabatic brake work is 1 25 times the ideal The
compression efficiency of 75% but ignor-
water to be frozen is at 8DT and the manu
ing other cycle losses compute the actual
factured ice is at 16°F for icc c = 0 5
horsepower per ton of refrigeration
Btu/lb-°F If NIIj is the refrigerant,
Ans (a) 67 6, (b) 7 48 (c) 1 98, (d)
determine (a) the capacity of the system
1135, (e) 0 631, (f) 18 1 cfm, (g) 0 842
(b) bhp per ton, (c) the tons of ice manu-
332 The same as 331 except that the factured in 24 hr if there are no radiation
refrigerant is Freon 12 and pi = 80 psia losses
(instead of 120) An* (a) 50 8 tons, (b) I 94, (c) 36 5
333 The temperature in a refrigerating 335 The same as 334 except that the
system at the expansion valve is 80°F refrigerant is Freon 12 and pi = 115 psia
The evaporator is at 40°F The sub- (after compression)
stance leaves the evaporator, entering
the compressor, as saturated vapor only 336 A reversed vapor-compression cycle
The cooling capacity is to be 25 tons is to be used for heating The maximum
demand is expected to be 600 cfm of
40°F
For comparison purposes consider the
volumetric efficiency of the compressor outside air heated to 85°F The tempera-
as 100%, let L/D = 1, let n = 150 rpm ture in the evaporator is to be 25°F and
let the compressor be double acting the Freon 12 refrigerant is pumped to
Compute the required displacement in 150 psia having entered the compressor
as saturated vapor Liquid enters the
cfm and the bore D
and stroke L when
(a) What rate
the refrigerant is (a) NHj, (b) SO,, (c) CO,, expansion vah e at 100°F
APPENDIX A. PROBLEMS 359
of refrigerant circulation needed, lb./
is compressor at 130°F and is compressed to
min.? Also determine (b) the horsepower saturated vapor at 220°F. The refriger-
input for an ideal/actual work ratio of ant enters the expansion valve at 210°F.
70%, (c) the actual COP, (d) the cost of The capacity is 20 tons. For an ideal
heating at 2 cents/kw-hr. (i) when the cycle, determine (a) the COP, (b) the
heat is obtained from the reversed cycle, rate of circulation of refrigerant, (c) the
(ii) when the heat is obtained from an compressor displacement for a volumetric
electrical heating element. efficiency of 80%.
337. The inhabitants of a planet where Ans. (a) 6.26, (b) 4.75 lb. /min., (c)

the outside temperature is 200°F decide 844 cfm.


to maintain a room at 150°F and use H«0 338-360. These numbers may be used
as the refrigerant. The steam enters the for other problems.

15. GASEOUS MIXTURES


note. The student should check computations where possible by a psychromelric chart,
thereby becoming familiar with both the basic theory and the commercial manner of obtaining
a solution. Abbreviations: d.b., dry bulb; w.b., wet bulb; da, dry air; dg, dry gas; v, vapor.

361. Atmospheric air at standard pres- 75°F (typical of home heating), (a)
sure and 85°F d.b. has a relative humidity What isthe relative humidity in state 2?
of 65%. Find (a) the partial pressures (b) How much moisture must be added
of the vapor and the dry air, (b) the to the air in the second state to bring its
humidity ratio, (c) the dew point, (d) the relative humidity to 40% (state 3)?
volume of the mixture, cu. ft./lb. da. Convert to gallons per hour for 10,000 cfm
Ans. (a) 0.388, 14.312 psia, (b) 0.01685, of atmospheric air (not typical of home
(c) 72°F, (d) 14.1. heating to bring in this much outside air).
362. Atmospheric air at 80°F has a
Name all states involved in your solution
on your Ts diagram, (c) What is the dew
humidity ratio of 0.016 Ib./Ib. da. Deter-
point of the air in state 3?
mine (a) the vapor pressure, (b) the rela-
Ans. (a) 14.5%, (b) 27 gal./hr., (c)
tive humidity, (c) the dew point.
49°F.
Ans. (a) 0.369 psia, (b) 72.8%, (c)
70.49°F. 366. The following processes are similar
to those in summer air conditioning.
363. In a condenser for a steam turbine Atmospheric air at 100°F and 70% relative
where the temperature 92°F, there are
is
humidity is cooled to 60°F (state 2) and
0.11 lb. da./lb. v. What is the vacuum delivered to a room where 80°F (state 3)
pressure, referred to a barometer of 29.92
is maintained. The total pressure remains
in. Hg? Ans. 28.3 in. Hg.
constant at 14.7 psia. Solve this problem
364. The state of atmospheric air is by chart only; include all details of your
defined by 27.6 in. Hg abs., 72°F d.b., chart readings on a sketch. Find (a)
58°F w.b. (a) Using the adiabatic satura- and TH ,, (b) w 2 and TTh. (c) For an
tion process, determine the humidity ratio, atmospheric air flow of 500 cfm, compute
the vapor pressure, the relative humidity, the condensed H-0 in lb. /min., and the
and the dew point, (b) Check the vapor heat transferred 1-2. How many tons
pressure by Carrier’s equation, (c) What of refrigeration are needed?
is the density of the atmospheric air? Ans. (c) 0.73 lb. /min., 6 tons.
What would be the density of dry air at 357. The
following processes are typical
the same total pressure? Which is of those summer air conditioning.
in
heavier? (d) Check this solution by the Atmospheric air, at 90°F d.b., $ = 70%,
psychrometric chart. Show sketch. and the barometer at 28.5 in. Hg, is cooled
Ans. (a) p, = 0.1706 psia, (c) p n =
in a steady flow process to 50°F (state 2),
0.0686 lb./cu. ft. after which it is reheated (perhaps by heat
C
366. Atmospheric air (29.92 in. Hg flow into the building) to 73 F d.b.
barometer) is heated at constant pressure (state 3), all at constant total pressure.
from 36° F and 60% relative humidity to Determine (a) the relative humidity and
860 APPENDIX A PROBLEMS
humidity ratio in state 2 (b) the heat rate of condensation for a flow of 1000 cfm
rejected 1 2 (A// including k/ at state 2) measured at state 1
(c) the change of total heat 1 2 (Find An* v /lb dg
(a) 0 01313 lb 71%
the wet bulb temperature from the psy 108°F, (b) 2 03 lb /min
chrometrie chart ) Is ATH a reasonable
359 It is desired to process atmospheric
approximation of A// for engineering pur
air at state 1, 92°F and 60% relative
poses? (d) How much moisture is taken
humidity to state 2 75°F and 40% rela
out of the air? (e) What arc the relative
tive humidity (a) Sketch the necessary
humidity and humidity ratio state 3? m processes on the Ts plane using refngera
An* (a) 0 00801 lb HjO/lb da (b) tion and determine the temperature to
-25 64 Btu/lb da (c) -25 84 Btu/lb which the atmospheric air must be cooled
da (d) 101 gr (e) = 44 3% (b) How much moisture is removed?
358 The volumetric analysis of a mix- (c) What are the wet bulb temperatures

ture of gases is 98% of Ni and 2% of 11*0 at states 1 and 2?


The mixture is cooled at a constant total An* (from chart) (a) 49°F (b) 0 01245
pressure of CO 1 psia from 120°F to 80®F lb /lb da (c) 80°F 59 5°F

Determine (a) the original humidity ratio 360-370 These numbers may be used
relative humidity and dew point (b) the for otherproblems

16 THE TRANSFER OF IIEAT

371 A 4 in thick wall has an outside film coefficients are 20 Btu/ft*-hr °F on


surface temperature of 90°F and conducts inside (aluminum) 3 Btu/ft*-hr °F on the
125 Btu/hr ft 'from the inside out What outside (a) Determine the overall trans-
is the temperature of the ms de surface mittance (b) What is the resistance
if the material of the w all is (a) grey cast in hr ft* °F/Btu? (c) What percentage
iron ft - 360 (b) dry concrete ft «» 5 4 change in U would occur if the aluminum
(c) oak * — 1 44 Btu in /ft’ hr °F 7 (d) copper and steel were omitted m
its cal
If all three of the walls are put together in culation? Would it make any significant
series determine the conductance from difference if steel were substituted for the
outs de surface to inside surface and the aluminum and copper which are excellent
inside surface temperature conductors?
An* (a) 91 39°F (b) 182 6°F, (c)
437"F (d) 532 F 374 (a) A residence is built ns follows
brick veneer on frame walls with in of
372 (a) A 10x30-ft wall of a building
plaster 1400 sq ft glass in walls (7=09
is made up of 8 in of brick with in of
Btu/hr ft*-°F 340 sq ft ceiling with
plaster on the inside If the temperatures
transmittance of 0 65 Btu/hr ft*-"F 2000
arc 10°F on the outside and 75°F on the
sq ft floor with transmittance of 0 3 Btu/
inside determine the conductance of the
hr ft* °F 2000 sq ft Inside temperature
wall the transmittance and the rate of
is to be 75°F Lowest expected outside
heat transmisson The conductivities temperature is 10°F (let this be the attic
arc brick ft = 5 plaster ft = 3 3 Btu Take the temperature
temperature too)
in /ft*~hr °F (b) The same as (a) except What is
difference for the floor as 40°F
that there is an air space o( over $£ m the heat loss from the house? (b) The
between the plaster and the brick (c)
same as (a) except that there is 2 in of
The air space reduces the heat flux by rock wool insulation mthe wall and 3 in
what percentage? in the ceding (£7e = 008 Btu/hr ft* °F
An* (a) 0 57 0 396 Btu/hr ft* *F now) (c) W hat is the percentage reduc
7740 Btu/hr (b) 0 375 0 292 Btu/hr ft*- tion of the heat loss due to the insulation*
°F, 5690 Btu/hr (c)2C5% (d) Compute the temperature of the ms de
373 A partition is made up of 1 m of wall surface in (a) and (b)
aluminum (ft = 1400) 1 of corrugated m
asbestos (ft = 0 828) ra of copperM 375 The same as 374 except that the
(ft = 2690) 1 in of a r space and J4 of m cooling loads for an outside temperature
steel (ft = 312 Btu in /ft*-hr °F) The of 100 F are desired Let the attic tem
APPENDIX A. PROBLEMS sei

perature be 120°F and the temperature parallel flow. Which exchanger would be
differential on the floor be 15°F. How the more expensive?
many tons of refrigeration are required? ,4ns. (a) 387°F, (b) 321.5°, 193.5°, (c)

376. Feedwater is to enter an econo- 3890 ft. 2 , 6460 ft.


2

mizer at 160°F, leave at 3G0°F. The flue 377. (a) A feedwater heater, for which
gases (cp = 0.242) enter at 800°F and the transmittance is 350 Btu /hr-ft 2 -°F,
there are 2 lb. of flue gases per pound of uses condensing steam at 20 psia for heat-
water. The water flow is to be 50,000 ing 85,000 lb. /hr. of water from 60°F to
and the transmittance is expected
lb. /hr. 215°F. What transmitting surface is
to be about 8 Btu/hr-ft 2 -°F based on the required? (b) After some fouling of the
inside tube surface. Determine (a) the tubes, the transmittance decreases to
exit temperature of the gases, (b) the 305 Btu/hr-ft 2-°F. For the area found in
LMTD for counterflow and for parallel (a), compute the exit temperature of the
flow, (c) the transmitting area (inside tubr feedwater after fouling.
surface) needed for counterflow and foe Ans. (a) 621 ft. 2 (b) 208°F. ,

note. There are not 377 problems in this booh. See blank numbers at the end of each
chapter.
APPENDIX B. TABLES
Table XIV USEFUL CONSTANTS
Abbreviations atm = atmospheres, eal = calorie cm = centimeter gm
= gram
gmol = gram-mol, kcal = kilo calone kg - kilogram kJ = kilo joule km =
kilometer kw = kilowatt, m
= meter mol = lb mol nt - watt others as usual
23 login N •= log. A = In Ar
Table XV. SATURATION PROPERTIES OF FREON 12

Temp Abs Pres Volume Enthalpy Entropy


°F psi cu ft /lb Btu/lb Btu/lb-°R

i V '

Vf VQ h, hfq hg s/ Sg

-10 19 19 0 01091 1 973 6 37 69 82 76 19 0 01462 0 16989


0 23 85 0 01103 1 609 8 52 68 75 77 27 0 01932 0 16888
1 24 36 0 01104 1 577 8 74 68 64 77 38 0 01979 0 16878
10 29 34 0 01116 1 324 10 68 67 65 78 33 0 02395 0 16798

20 35 74 0 01130 1 099 12 86 66 52 79 38 0 02851 0 16719


30 43 15 0 01144 0 919 15 06 65 36 80 42 0 03301 0 16648
40 51 67 0 01159 0 774 17 27 64 16 81 43 0 03745 0 16586
50 61 39 0 01175 0 655 19 51 62 93 82 44 0 04184 0 16530

60 72 43 0 01191 0 558 21 77 61 64 83 41 0 04618 0 16479


70 84 89 0 01209 0 478 24 05 60 31 84 36 0 05048 0 16434
80 98 87 o 01228 0 411 26 36 58 92 85 28 0 05475 0 16392
90 114 49 0 01248 0 355 28 71 57 46 86 17 0 05900 0 16353

100 131 86 0 01269 0 308 31 10 55 93 87 03 0 06323 0 16315

Table XVI. PROPERTIES OF SUPERHEATED FREON 12*

Saturation properties m italics

80 PSI A (66 21° F) 100 PSIA (80 76° F)

Temp V h 5 V h s
°F
1

0 5068 84 00 0 16450 O 4067 85 So O 16389

Mm
1]
m 0 5127
0 528
84
86
64
32
0
0
16571
16885

mm
,

110
0 543
0 557
0 572
87
89 64
91 24
98 0
0
0
17190
17489
17782
0 419
0 431
0 444
86 96
88 69
90 41
0 16685
0 16996
0 17300

115 PSIA (90 SHF) 150 PSIA (109 45° F)


Temp
°F
0 8537 86 20 0 16352 0 2697 87 80 0 16281

100 0 365 87 94 0 16665


no 0 376 89 71 0 16978 0 271 87 91 0 16299
120 0 387 91 46 0 17283 89 80 0 16629
130 0 398 93 20 0 17581 0 289 91 66 0 16947

*
Copyright 1955 and 1956, E. I, du Pont de Nemours & Co ,
Inc. Reprinted by
permission.
863
Table XVII SATURATED AMMONIA'

-63 11 0 02271 49 31 -0 0599 1 4857


-41 34 0 02319 25 81 -0 0034 1 4276
-27 29 0 02350 17 67
-25 0 0 02357 16 66
-20 0 0 02369 146S
-17 20 0 02375 13 70
-16 64 0 02378 13 50
-150 0 02381 12 97
-14 78 0 023S2 12 90
-100 0 02393 11 50
-8 40 0 02397 11 0S6
- 796 0 0239S 10 96
- 50 0 02406 10 23
- 3 40 0 02110 9 853
-0 57 0 02418 9 236
00 0 02419 9116
50 0 02432 8 150
5 52 0 02433 8 067
5 89 0 02434 7 991
9 07 0 02443 7 452
10 0 0 02446 7 304
11 66 0 02451 7 047
15 00 0 024G0 6 562
20 0 0 02474 5 910
21 67 0 02479 5 710
25 00 0 02488 5 334
40 0 0 02533 3 971
56 05 0 02584 2 952
60 0 0 02597 2 751
66 02 0 02618 2 476
68 00 0 02625 2 393
70 0 0 02632 2 312
70 53 0 02634 2 291
78 81 !
0 02664 1 994
80 0 0 02668 1 955
850 0 02687 1 801
86 00 0 02691 1 772
88 00 0 02698 1 716

89 78 0 02706 1 667
90 0 0 02707 1 661
92 00 0 02715 1 609
93 13 0 02720 1 581
96 34 0 02732 1 502
97 90 0 02738 1466
100 90 0 02751 1 400
110 00 0 02790 1 217

~ extracted uy peiuuooiw** '

U S Bureau of Standards Bulletin Iso 142


*

Table XVm. SUPERHEATED AMMONIA


Absolute Pressure, psi. (Saturation temperature in italics)

Temp. 15 20 25
°F. -37.29° —16.64° - 7.96°
r * 3 V h * r h 0

Sat. 17.67 €024 13938 13.60 606.2 1.3700 10.96 609.1 1.3516
-20 1S.01 606.4 1.4031
-10 18.47 611.9 .4154 13.74 610.0 1.3784
0 18.92 617.2 1.4272 14.09 615.5 1.3907 11.19 613.8 1.3616
10 19.37 622.5 .43S6 14.44 621.0 .4025 11.47 619.4 .3738
20 19.82 627.8 .4497 14.78 626.4 .4138 11.75 625.0 .3855
30 20.26 633.0 .4604 15.11 631.7 .4248 12.03 630.4 .3967
40 20.70 638.2 .4709 15.45 637.0 .4356 12.30 635.8 .4077
50 21.14 643.4 1.4S12 15.78 642.3 1.4460 12.57 641.2 1.4183
60 21.58 64S.5 .4912 16.12 647.5 .4562 12.84 646.5 .4287
70 22.01 653.7 .5011 16.45 652.8 .4662 13.11 651.8 .4388
80 22.44 65S.9 .510S 16.78 658.0 .4760 13.37 657.1 .4487
90 22.88 664.0 .5203 17.10 663.2 .4856 13.64 662.4 .4584
100 23.31 669.2 1.5296 17.43 668.5 1.4950 13.90 667.7 1.4679
110 23.74 674.4 •53SS 17.76 673.7 .5042 14.17 673.0 .4772
120 24.17 679.6 .5478 18. 0S 678.9 .5133 14.43 678.2 .4864
130 24.60 684.8 .5567 18.41 684.2 .5223 14.69 683.5 .4954
140 25.03 690.0 .5655 18.73 689.4 .5312 14.95 688.8 .5043
150 25.46 695.3 1.5742 19.05 694.7 1.5399 15.21 694.1 1.5131
160 25.88 700.5 .5827 19.37 700.0 .5485 15.47 699.4 .5217
170 26.31 705.8 .5911 19.70 705.3 .5569 15.73 els
In .5303
ISO
1SO
200
26.74
27.16
27.59
711.1
716.4
721.7
.5995
.6077
1.6158
20.02
20.34
20.66
710.6
715.9
721.2
.5653
.5736
1.5817
15.99
16.25
16.50
Ei
720.8
.5387
.5470
1.6552
220 28.44 732.4 .6318 21.30 732.0 .5978 17.02 731.6 .5713
240 21.94 742.8 .6135 17.53 742.5 .5870
260 18.04 753.4 , .6025

30 35 40
- 0.67° S.S9° 11.66°

Sal. 9236 611.6 15364 7.991 613.6 1.3236 7.047 6164 1.3125
10 9.492 617.8 1.3497 8.078 616.1 1.32S9
20 ,9.731 623.5 .3618 8.287 622.0 .3413 7.203 620.4 1.3231
30 9.966 629.1 .3733 8.493 627.7 .3532 7.387 626.3 .3353
40 10.20 634.6 .3845 8.695 633.4 .3646 7.568 632.1 .3470
50 10.43 640.1 1.3953 8.895 638.9 1.3756 7.746 637.8 1.3583
60 10.65 645.5 .4059 9.093 644.4 .3SG3 7.922 643.4 .3692
70 10.88 650.9 .4161 9.2S9 649.9 .3967 8.096 648.9 .3797
SO 11.10 656.2 .4261 9.4S4 655.3 .4069 8.268 654.4 .3900
90 11.33 661.6 .4359 9.677 660.7 .4168 8.439 659.9 .4000
100 11.55 666.9 1.4456 9.869 666.1 1.4265 S.609 665.3 1.409S
110 11.77 672.2 .4550 10.06 671.5 .4360 8.777 670.7 .4194
120 11.99 677.5 .4642 10.25 676.8 .4453 8.945 676.1 -42SS
130 12.21 682.9 .4733 10.44 052.2 .4545 9.112 681.5 .4381
140 12.43 6SS-2 •4S23 10.63 6S7.6 .4635 9.278 686.9 .4471
150 12.65 693.5 1.4911 10.82 692.9 1.4724 9.444 692.3 1.4561
160 12.87 698.8 •499S 11.00 69S.3 .4811 9.609 697.7 .4648
170 13. OS 704.2 •50S3 11.19 703.7 .4897 9.774 703.1 .4735
ISO 13.30 709.6 .5168 11.38 709.1 .49S2 9 93 s
. 70S.5 .4820
190 13.52 714.9 .5251 11.56 714.5 .5066 10.10 714.0 .4904
200 13.73 720.3 1.5334 11.75 719.9 •
1.5148 10.27 719.4 1.4987
220 14.16 731.1 .5495 12.12 730.7 .5311 10.59 730.3 .5150
240 14.59 742.0 .5653 12.49 741.7 .5469 10.92 741.3 .5309
260 15.02 .5808 12.86 752.7 .5624 11.24 752.3 .5465
280 15.45 764.1 .5960 13.23 763.7 .5776 11.56 763.4 .5617
300 11.88 774.6 .5766

* Extracted by permission from Tables of Thermodynamic Properties of Ammonia,


U» S. Bureau of Standards Bulletin No. 142.

S65
)

Table XVIII SUPERHEATED AMMONIA ( Continued

3b 6
ENTROPY
Index

A Otto cycle, 174


regenerative gas turbine, 161
Abbreviations, x variations of Brayton cycle, 163
Absolute pressure, 9 Air-steam mixtures, 302, 304
Absolute temperature, 11, 107 (see Gas-vapor mixtures)
Absolute zero, 58, 107 Ammonia, 364, 365
by energy scale, 107 as refrigerant, 283, 284
by gas law, 58 ol Ts plane, 206
Absorbent, 285, 313 properties of, 205, 304, 365
Absorption refrigeration, 285 Apparent molecular weight, 293
Absorptivity, 330 Archimedes’ pump, 121
Acceleration, 6 Argon, constants for, 39
of gravity, 7 Articulated construction, 126
standard value, 7 Atmospheric pressure, 9
Acetic acid, 206 standard, 10
Acoustic energy, 26 variation with altitude, 131
Acoustic velocity, 264 Atomic energy, 247
Adiabatic compression efficiency, 13 Atomic number, 247
gas turbine, 154 Atomic weight, 247
Adiabatic overall efficiency, 136 Automotive engine, 173
Adiabatic process, 74, 211, 215 Available energy, 110, 112, 141
compressor, 127 loss of, 116, 117
irreversible, 78, 86, 128, 211, 213 Avogadro's law, 60, 293
isentropic, 74 Axial-flow compressor, 152
nonflow, 74
between properties 76
relations B
steady flow, 75, 128
throttling, 86, 215 Barometer, 9
vapors, 211 Beau do Rochas, 171
work of, 77 Bernoulli, 52
Adiabatic saturation process, 303 equation, 53
vs. wet bulb, 304 Beta particles, 249
Adiabatic wall, 40 Black body, 330
Adsorbent, 313 Bleeding (steam), 241
Aftercooler, 123 Boiler horsepower, 221
Air, properties of, 146 Boilers, output of, 302
Air compressors, 126, 133, 137 Boiling point, 11, 192
preferred compression curves, 133 Boltzman, L., 1 19
Air-fuel mixture, dew point, 310 and second law, 119
gas turbine, 160, 104 Bore-stroke ratio, 1S4
partial pressures, 310 Boundaries, 2
relative humidity, 311 Bourdon pressure gage, 9
Air-standard analysis, 174 Boyle's law, 56
Brayton cycle, 151 Brake engine efficiency, 1S6, 229
Brayton with friction, 155 Brake mep, 1S1
Diesel cycle, 178 Brake power, 1S3, 22S
gas turbine, 151 Brake steam rate, 22S
367
1

868
INDEX
Brake thermal efficiency 185 229 of vapor cycle 281
in terms of steam rate 229 Cold air standard 175
Brake work 21 185 188 228 Cold body 90 105
Brayton George 150 Combined engine efficiency 186 229
Brayton cycle 151 154 Combined thermal efficiency 185 229
maximum work 154 in terms of steam rate 229
variations of 163 Combined work 185 228
with fluid friction 155 Combustion 159
with regeneration 161 energy equation 159
British thermal unit 16 beats of, 159
Bulk modulus 264 Combustor 160 162 163
Butane 39 energy diagram 160
pressure drop in 160
C Component 294
Compressed air uses of 122
Calorific value Compressed liquid 203
(see Healing talue) Compressed water 200 203
Calorimeter, for quality of steam Compression stroke 172
216 Compression efficiency compressor 136
Capacity of compressor 132 154 281
Carbon dioxide 39 Compression ratio 98 176
as refrigerant 283 281 and efficiency 1-6
on Ts plane 206 in terms of clearance 177
Carnot N L 8 93 turbo jet eng ne 165
principle 104 Compression system of refrigeration 279
Carnot cycle 93 Compressor 126 132
conclusions on refrigeration 276 capacity of 132
efficiency of, 9 displacement of 132 281
heating 291 example 134 141 281
of ideal gas 95 for vapors 281
on pi plane 94 95 single-stage 126 128 132
on TS plane 93 277 two stage 130 141
operation of 94 work of 126 128 130 141
refrigeration 277 Condensation in gas-vapor mixture 312
reversed 103 277 Condenser 306 478
Carrene 283 284 heating cycle 292
Carrier s equation 305 in power cycle 222 225
Cascade refr gerating system 289 in refrigerating cvclc 279
Celsius 11 Condens ng vapors 4
Chain reaction 250 Conductance 321
Charactcrist c equation 59 293 table of values 320
(see Equation of state) unit 321
of ideal gas m
mols 294 Conduction 23 315
Charles law 56 liquidsand gases 319 322
Chemical energy 26 solids 318 323
Chemical equilibrium 118 through curved wall 323
Clausius R J E 41 through plane wall 381
inequality 114 Conductivity 316
Clearance volume 17® ta'D'ie oi Til'S)

air compressor 130 \ ariation of 3 17


internal combustion engine 177 Conservation of energy 29
Clerk Sir D igald 171 law of 29
Closed system 3 26 Conservation of mass 28
energy equation 31 Constant pressure process 38 70
Coefficient of discharge 271 of gas 70
Coefficient of heat transfer of gas vapor mixture 311
(see Transm tlance) of vapor 209
coefficient of performance 104 276 Constant xolume process 38 68
of heating cycle 104 291 irreversible 69
INDEX 369
of gas, 68 Degree of superheat, 193
210
of vapor, Degree of supersaturation, 266
Constituent, 294 Delivered work, 21
Continuity of mass, 28 Density, 10
Convection, 24 of mixture, 299
Conventional diagram, 128 units of, 10
work for compressor, 128, 130, 138 Dew point, 298
Conventional indicator card, fuel-air mixture, 310
for single-stage compressor, 128, 129 pressure constant, 299, 312
for two-stage compressor, 138, 139 Diatomic gases, 18, 47
Conventional volumetric efficiency, air Dichlorodifluoromethane, 284
compressor, 132 Dichloromethane, 284
Convergent nozzles, 255, 256 Diesel, Rudolf, 178
Convergent-divergent nozzle, 255, 256 Diesel cycle, 178
Correction factor for initial velocity, 257 comparison with Otto, 180
Counterflow (heat exchanger), 328 efficiency, 179
mean temperature difference, 329 efficiency vs. cutoff ratio, 180
Critical mass, 250 Diesel engine, 181
core, 250 fuel cutoff ratio, 181
Critical point, 195 heat rate, 185
properties at, 208 thermal efficiency, 185
Critical pressure, nozzles, 261 Diffuser, 272
values of, nozzles, 263 ram jet, 164
Critical temperature, 206,208 vs. nozzle, 273
Curved wall, composite, 324 Discharge from nozzle, 268
heat through, 323 mass of, 258
Cutoff ratio, 179 Displacement, 96
for ideal Diesel cycle, 181 comparison for different refrigerants, 284
variation of efficiency with, 180 for gases, 96, 97
Cycle, 13 for vapors, 281
analysis of, 90, 98, 106 of compressor, 132
Brayton, 151 Displacement volume, 96, 97, 281, 284
Carnot, 93, 95, 98 vapor compressor, 281, 284
closed, 26 Divergent nozzle, 255
Diesel, 178 Double-acting engine, 125, 126
elements for, 91 Double-pipe heat exchanger, 327
Ericsson, 100 Dry-bulb temperature, 302
incomplete expansion, 231 Dyne, 7
Otto, 174
power, 92 E
Rankine, 233
refrigerating, 103, 276 Effectiveness, regenerator, 161
regenerative, 100, 241 Efficiency, 92, 135
reheat, 238 adiabatic compression, 136, 154
reheat-regenerative, 244 brake thermal, 185, 229
reversed, 103 combined thermal, 185, 229
reversible, 102 engine, 155, 186, 229
Stirling, 102 indicated thermal, 185, 229
thermal efficiency of, 92 in terms of steam rate, 228
work of, 91 isothermal compression, 136
mechanical, 135, 187, 230
D nozzle, 269
of actual gas turbine, 160
Daimler, Gottlieb, 172 of Brayton cycle, 153
Dalton’s law, 295 of Carnot cycle, 94
Datum, for measuring energy, 15 of combustor, 160
Dead-center position, 172 of Diesel cycle, 179
Degradation of energy, 110, 116 of diffuser, 273
Degree of saturation, 308 of gas turbine with friction, 155
S70 INDEX
Efficiency (Continued) Energy scale of temperatures 108
of Otto cycle, 175 Engine effic ency 155 186 229
of Itankine cycle 227 brake 186 229
of turbo-jet 168 comb nod 186 229
overall 136 gas turbine 155
shaft 136 indicated 186 229
xolumetric 132 turbo jet eng ne 169
Einstein 15 119 247 values of 1/0
mass and energj 15 247 Enthalpy 32
Ejector 273 -entropy diagram 219
Electrical energy G of compressed liquid 204
Electrochemical energj 26 of compressed water 200 203
Electromagnetic energj 26 of exhaust 233
Electron 247 of gas-vapor mixt ire 306 307
Element 247 of ideal gas 63
of Ion pressure superheated steam 30o
Emissivity 330 l>
of mixture of gases 07
Energy 15
of steam from nozzle 269
changes of 16 30 297
of vaporizat on 193 194
conservation 29
of various gases 14
conversion constants 362
of net mixture ‘>01
flow work 24
stagnation 154 1C4
forms of 25
use in steady flow 33
bc&V 1Z
Enthalpy of exha ist 233
internal 18
for nozzle 269
into mass 247
Entropy 42 97 194
mechanical kinetic 17
change for ideal gas 64 79
mechanical potential 16
eq lations for (gas) 64
transitional 20 23 2o
from Carnot cycle 111 113
work 19
increase of 79 114
Fncrgy diagram 31 of gas vapor mixture 309
adiabatic saturation 301
of H
O 196 199
bo ler unit 225 of in xture of gases ^97
combustor 160 103 of let m
xture 201
compressor 35, 141 163
proluction 114 118
condenser 225 relation to unavailable energy 110 115
flu d flow 53
revers ble process 43
gas turbines 151 163 Environment 3
heat exchanger 117 Equal on of state 58 293
nozzle 34 ideal gas 58
pipe line 225 Equlibrum 23 118
pump 225 chemical 1 18
regenerative cycle 241 in nozzle 259
reheat cj cle 238 internal 119 192
reheat-regenerative cycle 245 mechanical 118
steady flow engine 233 states 119
steam pow er plant 225 thermal 23 1 18
two fluids 50 thermodynam c 1 19
two stage compressor 141 163 Equilibrium expans on (nozzle) 259
turbo jet engine 165 volume during 272
variable flow 51 Equ valent molecular weight 993
with friction 52 Ericsson J 100
Energy equations 30 45 cycle 100
ad abatic saturation 304 Evaporator 279 286 288 289 292
for enthalpy of exhaust 233 Exchanger heat
incompressible fluid 52 (see Heal exchanger*)
nonflow 30 Expans on ratio 98
simple 30 Expansion valve 279 288 289
steady flow 31 33 45 Extended surface (heaters) 316
INDEX 871
External irreversibility, 41 Gas tables, 146, 147
Extracting steam, 241 entropy function, 144
properties, 145
F relative pressure, 144
relative volume, 145
Feather valve for compressor, 125 Gas turbine, 150, 152
Feedwater pump, 225 air standard, 151
work of, 225, 239, 243 compression efficiency, 154
Film coefficient, 321 efficiency, 153
steam, 326 efficiency vs. pressure ratio, 156
Film conductance. 321 heat rate, 170
Film coefficient)
(see operation of, 151
First law of thermodynamics, 29 performance data, 170
Fission fragments, 248 rotative speeds, 158
barium, 249 turbine efficiency, 155
krypton, 249 with fluid friction, 155
Flash tank, 251 with regeneration, 161
Flow, Gas-vapor mixtures, 298
(see Steady flow and Variable flow ) adiabatic saturation process, 303
Flow energy, 24 density, 299, 307
Flow work, 24 dew point, 299, 307
Fluid, 4 enthalpy, 305, 306, 309
Fluid friction in nozzle, 269 entropy, 309
Force, units of, G fuel-air,310
Four-stroke cycle, 172 humidity ratio, 301, 305, 306, 312
Diesel, 178 internal energy, 309
Fourier’s law, 316 partial pressures, 298, 302, 305, 306
applied to curved wall, 323 relative humidity, 300, 307
applied to plane wall, 318 total heat, 307, 309
Free air, 141 volumeof, 299, 313
Free body, 3 wet-bulb temperature, 302, 304
Freon 12, 283, 284, 363 Gay-Lussac, J- L., 56
on Ts plane, 206 Generator (refrigeration), 286
Friction, Gibbs function, 119
effect on entropy, 236 Gravimetric analysis, 294, 299
of flow, 235 conversion to volumetric analysis, 296
Friction head, 53 Gravitational energy, 16
Frictional energy, 52 Gravity, standard, 7
Frictional horsepower, 188
Fuel consumption, 185 H
gas turbine, 159
specific, ICE, 185 Heat, 23, 315
specific, jet engine, 167 conduction, 23, 315
Fuel cutoff ratio, 181 convection, 24
equations for (gas), 88
G exchangers, 315, 316
latent, 194, 207
Gage pressure, 9 of combustion, 159
above atmospheric, 9 radiation, 24, 330
below atmospheric, 10 sink, 90
Gamma rays, 249 Heat balance
Gas constant, 59 (see Energy diagram)

computation of, 59 Heat capacity, 37


for mixture of gases, 296 (see Specific heals)

specific, 59 Heat content, 33 (footnote)


table of, 39 Heat exchangers, 220, 315
units of, 59, 60, 61, 362 absorption refrigeration, 285
universal, 60, 293 counterflow in, 327
Gas engine, 126 energy diagram, 117, 141
872
INDEX
Heat exchangers (Cont nved) processes 67
extended surface 316 specific heats 64
loss of available energy in 117 JdeaJ work 20
parallel flow m
327 (see name of ideal cycle)
process in 220 Impulse momentum 160
types 315 turbine work from 166
Heat pump 102 104 291 292 Incomplete expansion engine 231
coefficient of performance 104 291 exhaust enthalpy 233
Ileat rate 160 185 228 mep 232
brake 229 thermal efficiency 231
combined 229 Incompress blc fluid flow 52
gas turbine 160 170 Indicated engine efficiency 186 229
ICE 185 Indicated horsepou er 123
indicated 229 Ind catcd steam rate 228
regenerative engine 244 Indicated thermal efficiencj 186 229
steam engine 228 Indicated nork 20 123 185 228
steam power plant 228 Indicator 124
Heating cycle 291 Indicator card 122
Heating value 158 187
(see Conventional tad color card)
at constant pressure 159 for single-stage air compressor 123
at constant volume 159 for two stage air compressor 140
for computing efficiency 159 Initial vclocitj in nozzle correction for
higher 159
257
lower 159 185
Injector 273
High pressure power plants 253
Instantaneous specific heats 47 48
II gher heating v alue 159
Intercooler 137
Horsepower 36
heat transfer in 141
boilcT 221
in Braj ton cycle 163
hour 36 362
pressure drop tl ro igh 170
minute 36 362
Internal combustion engines 171 181
per ton of refrigeration 278
actual thermal effic ency 185
Hot air standard 170
automotive 1"3
Hot body 90 lOo
compression rat o 175 179
Hoyle Fred 28
development of 171
Humidity relatne 300
first law analysis 181
(see Helatite humidity)
supercharging 189
Humidity ratio 301
throttl ng 189
air steam mixture 302 305 30G
volumetric efficiencj 188
fuel in air 311
Internal combustion t irbine
(sec Gas It rb nc)
Internal energj 18 200
Ico point 11 kinetic 18
of gas-vapor mixture 309
Ideal cycle 222
(see name of cycle) of ideal gas 62
vs ideal engine 226 of liquid 200
Ideal eng ne 222 of mixture of gases 297

vs ideal cycle 226 of superheated vapor 202


of vapor 200
Ideal gas 55
Avogxdro s law 60 of wet mixture 202
Boyle 8 law 56 potential for gas 19
Charles law 56 Internal irrevers bil tj 41
enthalpy 63 Interpolation in steam table 202
entropy change 65 Irrevers bihty 40 116
equation of state 58 external 10”
equations 88 Irreversible adiabatic process 78
gas constant 59 362 in compressor 128
internal energy 62 value of m
in 79
Joule s law 61 work of 79
INDEX 373
Irreversible process, 78, 86 Linde system of gas liquefaction, 289
adiabatic, 78 Liquefying gases, 289
entropy change in, 64, 79 Liquids,
for vapors, 212, 213, 216 and vapors, 191
Isentropic process, 74, 211 compressed, 200, 203
for vapor, 211 percentage, 201
ideal gas, 74 saturated, 192
in compressor,127 strong, 286
258
in nozzle, subcooled, 203
steady flow, 75 wenk, 286
with variable specific heat, 87 Logarithmic mean temperature difference,
Isobaric process, 70, 210 327
(sec Constant pressure process) Lost work, 213
Isometric process, 68, 211 due to friction of flow, 235
(sec Constant volume process) due to heat loss, 236
Isothermal compression efficiency, 136 in nozzles, 269
Isothermal process, 56, 72 in steam engine, 235
in compressor, 128 Lower heating value, 159, 185
perfect gas, 72
vapor, 214 M
Mach number, 169, 264
J
Macroscopic view, 1

Manometer, 10
Jet propulsion, 164
Mariotte, Edme, 56
efficiency, 169
Mass, 6-8
engine, 165
conservation of, 15, 28
work, 166
flow in nozzle, 258
Joule, James Prescott, 16, 29, 62, 153
into energy, 15, 247
constant, 16
number, 247
Joule cycle, 153
Matter, composition of, 247
Joule-Thomson coefficient, 62, 290
Maxwell, James C-, 41
Joule’s law, 61
Mean effective pressure, 96, 123
brake, 184
K equation, 97
indicated, 124
Keenan and Kaye, 144 of Rankine engine, 228
Keenan and Keyes, 193 Mean specific heat, 49
Kelvin, Lord, 29, 62 Mean temperature difference, 327
biography, 108 Mechanical efficiency, 187, 230
second law, 110 compressor, 135
temperature scale, 108 generator, 187
Kilo Btu, 220 Mechanical equilibrium, 118
Kilowatt, 36, 362 Mechanical friction, 52
-hour, 362 Mechanical kinetic energy, 17
Kinetic energy, 17 Mechanical potential energy, 17
Kirchoff’s law, 330 Mega Btu, 220
Methyl chloride, 283, 284
L Microscopic view, 2
Mixtures,
Latent heat, 194 density of, 299, 307
of evaporation, 194 gas constant for, 296
of fusion, 207 of gases, 294
Law, of gases and vapors, 298
first, 29 of liquid and vapor, 193
of conservation of mass, 28 pressure of, 295
of degradation of energy, 110, 116 specific heats of, 297
second, 41, 110 volume of, 294, 295, 296
third, 119 Moderator, reactor, 251
374
INDEX
Moisture
Optical pyrometer 12
m mr 299 Otto NA 171
in steam 192 201 Otto cycle 174
percentage 201
(see Internal combust on engine)
Mo) 46 61 293 air standard 174
Mot volume 61 clearance 177
Molar gas constant comparison with Diesel 180
(see Universal gas constant)
efficiency 175
Molar specific heat 46 47 297 efficiency vs compression ratio 376
Molecular weights 39 48 296 example 177
apparent 293 four stroke 172
equivalent 293 two stroke 182
of air 293 Over all eoeffic ent heat transfer 322
of gas mixtures 293 2% Over all conductance 322
some values 0! 39 48 Over all effic ency compressor 136
Molecules 2 (sec Combined thermal efficiency)
Mother diagram 218 219 at rear Overcxpans on (nozzle) 268
Momentum 166
Monatomic gases 18 47 P
Multistage compression 136
Parallel flow (heat exchanger) 327
N mean temperature difference 329
Partial pressure 295
Neutron 247 of vapor in mixture 301 302
Newton Sir Isaac 6 vs volumetric fraction 296
New ton s laws 6 Path of state po nt 22
Nonflow processes 67 209 Percentage clearance 130 177
energy equation 30 and displacement volume 132
work for 21 of air compressors 130
Nonflow sj stem 21 of internal combustion eng nes 177
energy equation 30 Percentage moisture 192
work 21 use of 201
Noz2le 255 Perfect gas 55
actual volume 271 (see Ideal gas )
air 258 Perpetual motion
convergent 255 256 of first kind 36
critical pressure in 261 of second kind 119
divergent 255 256 Phases of substances 4
effect of d scharge pressure 261 268 Planck 110
efficiency 269 Plan meter 122 125
energy diagram 34 256 Plutonium 248
flow through 268 Point function 6 22
ideal exit velocity, 257 259 Polytropic process 81
losses in 269 effect of varying n S5
mass flow through 258 ideal gas 81
ram jet engine 165 in compressor 127
supersaturated flow 265 267 vapor 215
*
throat of 255 Poly tropic specific heat 82
variation of properties in 259 Potent al energy 17
Nozzle coefficient 269 internal for gas 19
Nozzle efficiency 2f9 Pound mass 6 7
Nuclear energy 26 247 Poundal 6
Nuclear power plant 254 Power 36
Nucleus 247,250 relation between units of 362
Power cycle 92 103
number per minute 97
Pow er plant 26
Ohm slaw 323 Power stroke 172
Open system 3 27 31 finding number of 184
1

INDEX 375
Pressure, 8 Radioactive decay, 249
absolute, 9 Ram coefficient, 165, 170
coefficient, 165 Ram effect, 164
gage, 9 Rankme, Wm. J. M., 224
saturation, 197, 205. 310 Rankme cycle, 223
units of, 10 efficiency of, 227
Pressure head. 53 net work, 224
Pressure ratio, 9S pump work of, 225
critical, 262 with incomplete expansion, 231
Pressure-volume plane. 5 Rankme engine, 226
of vapor, 195 efficiencies of, 227
processes on, 6S-S1, S5, 210-215 heat rate, 228
Process. 13, 56 improving efficiency of, 230
adiabatic. 74, 211, 215 mep of, 22S
adiabatic saturation. 303 steam rate, 22S
constant pressure. 3S, 70, 20°, 311 Rankine temperature, 1
constant volume. 3S. 6S. 210 Ratio of compression, 9S
gas-vapor mixtures. 29S Ratio of expansion, 9S
isentropic. 74, 211 Reactor, 250
isothermal. 56. 72, 214 boiling, 251
nonfiow, 74 fuel rods, 251
polvtropic, SI, S5, 215 Real-mixture standard, 176
steady flow, 75 Receiver, 125
throttling, 86, 215 for compressed air, 123
Prony brake, 183 of heat, 90
Propane, 39 Rectifier (absorption refrigeration), 286
as refrigerant, 2S4 References, list, 333
Properties, Reflectivity, 330
extensive, 5 Refrigerants, 206, 283
intensive. 5 toxic, 283
of a substance, 4 Refrigeration, 103
of air, 146 absorption, 2S5
of ammonia, 205, 364, 365 Carnot, 103, 277
of compressed water, 203 cascade, 2S9
of gases, 147 coefficient of performance, 103, 277,
of steam, 196-199 2S1
of superheated vapor, 19S. 202 cycle, 103, 277, 2S0
of two-phase system, 201 rating of, 278
of wet mixture, 201 two-stage, 28S
on thermodynamic surface, 20S units for measurement. 27S
Propulsive efficiency. 16S vapor, 27S
Proton, 247 Regeneration, 100, 203
Psychrometer, 302 in refrigerating cycle, 2SS
Psychrometric charts. 303, at rear Regenerative cycle, 100. 241
Pump, efficienev, 225 for gas turbine, 161
work, 225, 239, 243 pump work for, 243
Pure substance, 3 saving with, 244
Pyrometer, 12 Regenerator, 101, 161
effectiveness of, 161, 171
Q pressure loss, 171
Reheat cycle, 202
Quality, 192 pump work for, 239
determination of, 216 Reheat engine, gas, 163
use of, 201 Reheat-regenerative cycle, 244
turbine for, 245
R Relative humidity, 299, 300
air-octane, 311
Radiant heat, 24, 330 air-steam, 307, 312
wave lengths, 24 Reservoir of heat, 105
376
INDEX
Res stance 321 323 ratio 40 47
curved wall 323 325 relation between for ideal gases 64
insulated steam line 326 theoretical values of 47
thermal vs electrical 323 units of 38
unit 321 values of 39 48
Resstivity 317 variable 46 47 48
Reversed cycle 92 103 276 variation of 47
used for heating KM ^91 Specific humidity (now called Hun idity
Reversible cycle 102 ratio)
engine 104 Specific volume 10 193 201
moat efficient 104 nozzle 271
process 42 Stagnation properties 257
steady flow 44 45 enthalpy 154 164
Reversibility 40 nozzle 2o7
external 41 Standard gravity 7
internal 41 Standard ton of refrigeration 278
State 4
S of substance, 5
path 22
Saturated air 300 points 5
Saturated gas 301 Steady flow 27 31
Saturated liquid 192 applications 34
line 105 energy equation 32 33 50 137 304
Saturated mixture 193 gas vapor mixtures 311
Saturated steam 196 197 of incompressible fluid 52
Saturated \npor 193 of liquids 52
lae 19o of more than one stream 50
Saturation curves 206 process for gases 71 “3 75 77 79 8"*
Saturation temperature 192 processes for v a pors 211 213 214 215
of alcohol benzene octane 310 mo 2° l
of ammonia 20o 304 36a system 27
of Freon 12 303 Steady state 91
of steam 196 197 Steam 195
Scavenging 182 enthalpy at low pressure 30o
Second law of thermodynamic* 41 110 film coefficient SM
Secbeck T J 12 Mother diagram of 219 at rear
Separating calorimeter 217 properties of 190-199
Shaft efficiency 136 temperature-entropy diagram of 218
Shaft work 21 126 Steam power plant 26
Simple energy equation 30 modern cy cles for 237
Single acting engine 125 126 Steam rate 223
S ngie-pass flow 316 and exhaust enthalpy 233
S nk 90 lOo brake 223
Slug 8 combined 2°S
Sol d phase 207 curve of 244
Some velocity 264 efficienciesterms of 00 9
m
Source of heat 90 indcatcd 228
Specific fuel consumption 167 185 Steam turbine 245
Specific heat capacity 37 (see Turbine)
Specific heats 37 Stefan Boltzmann law 330
constant pressure 38 Stirling cycle 10°
constant volume 38 Stored energy 17 18 19 25 26
curves of 47 Strong liquid 286
equations for 37 Subcooled liquid 203
instantaneous 38 40 47 48 Sublimation 4
mean 49 Subsonic velocity 265
molar 46 297 Suction stroke 172
of mixtures of gases 297 Sulfur d oxide 39
polytropic 82 on Ts plane 206
INDEX 377
Superheat, degrees of, 193 Temperature-entropy plane, 43
Superheated steam, 188 of vapors, 195, 206
at low pressure, 305 Temperature gradient, 318
Superheated vapor, 193, 202 Thermal conductivity, 316
flow through nozzle, 2G8 (see Conductivity)
Supersaturation, 265 Thermal efficiency, 92, 184
degree of, 266 actual for ICE, 187
expansion, 2G5, 267 brake, 185, 229
Supersonic velocity, 265 combined, 185, 229
Surface conductance, 321 Diesel, 179
(see Film coefficient) gas turbine (see Gas turbine)
Surroundings, 3 indicated, 185
Symbols, jet engine, 169
list of, pit Otto, 175
for vapors, 193 Rankine, 227
Systems, Thermal energy, 251
classification, 26 Thermal equilibrium, 23, 118
(see Cycles ) Thermal radiation, 330
closed, 3, 26 (see Radiation)
constant pressure, 38 Thermal resistance, 321
constant volume, 38 (see Resistance)
definition, 2 Thermal resistivity, 317
open, 3 Thermocouple, 12
Thermodynamic equilibrium, 119
T Thermodynamic surface, 208
Thermodynamic temperature, 107
Tables, Thermodynamics,
characteristic constants for gases (I), 39 definition, 1
characteristics of refrigerants (XI), 281 first law of, 29

compressed water (IX), 200 second law of, 110


conductivities (XII), 319 third law of, 119
conversion constants (XIV), 362 Thermometers, 12
enthalpy of gases (V), 147 Third law, 119
ideal gas formulas (III), 88 Thompson, Benjamin, 29
properties of air (IV), 146 Thomson, Win., 62, 108
saturated ammonia (X), 205 (see Kelvin)
saturated ammonia (XVII), 364 Throat, 255
saturated Freon 12 (XV), 363 of nozzle, 256, 264
saturated steam: pressures (VII), 197 pressure at, nozzle, 262, 268
saturated steam: temperatures (VI), 196 size, 265, 268
superheated ammonia (XVII), 365 temperature at, nozzle, 262
superheated Freon 12 (XVI), 363 velocity at, nozzle, 264
superheated steam (VIII), 198 Throttling calorimeter, 216
transmittances (XIII), 320 Throttling process, 86, 215, 279
variable specific heats (II), 48 for gases, 86
Tare, for prony brake, 183 for vapors, 215
Temperature, 10 in calorimeter, 216
absolute, 11 in nuclear steam generator, 251
absolute zero, 58 loss ofwork due to, 236
centigrade, 11 on Mollier diagram, 219
deg. Kelvin, 11 vapor refrigeration, 279
deg. Rankine, 11 Thrust force, 166
energy scale of, 108 Ton of refrigeration, 278
Fahrenheit, 11 horsepower per, 278
limiting, in steam power, 237 Torque, 184
measuring, 12 Torricelli, E., 9
of constituents in mixture, 299 Total heat, 33 (footnote)
Temperature-entropy diagram for steam, example, 309
of steam and air mixture, 307
218
378 INDEX
Total heat (Continued) processes of 209
psychrometnc chart 303 at rear on 278 287
refrigerat
Transferred heat 23 superheated 193 202
(see Heat) symbols for properties of 193
Transit onal cnergj 20 23 tables 193 196 199 205
Transmittance 322 Vaporization 4 192
table of values of 320 enthalpy of 193 194
Transmissivity 330 V anable flow 50
Triple point 207 208 energy diagram 51
Turbine effic cncy 155 23-4 Variable specific heats 46
Turbine 245 air standard 176
as system 27 curves of 47
energy equation appl cd to 226 equations for 48
enthalpy of exhaust from 233 examples 49 87
gas 150 lsentropic with 87
work 234 Velocity coeffic ent nozzle 269
Turbo jet engine 164 V doc ty from nozzle 256
efficiency 168 169 actual 270
energy diagram 164 V clocity variation nozzle 260
fuel consumption 167 V ibrat on of atoms 18
pressure coefficient 1G5 V lew factor rad ation 331
ram coefficient 164 165 Volume
thrust 165 166 drawn m 130 134
work 166 of constituents in mixture 299
Turbo prop engine 152 of steam in nozzle 271
Two phase system 191 192 of wet mixture 201
Two stage compressor 130 specific 10
energy diagram 141 V olumetnc analysis 294 298
example 141 conversion to grav metr c analys s 296
intermediate pressure for 139 relation to partial pressure 295 298
work for 138 Volumetric effic eney 132 136
l>
Two stroke cycle 182 air compressor 13
con\ rational 132
D internal combustion eng ne 188
in two stage compress on 136
Unavailable energy 110 118 in vapor con press on 281
comparison with lost work 235 von Guer eke Otto 9
increase of 114
relation to sink temperature 112 W
Underexpansion (nozzle) 268
Unit conductance 321 Water 194
(see Conductance ) (see Mo sture)
Unit resistance 321 compressed 200 203
(see Resistance) properties 19G 197
Units 6 VV ater rate 228
consistent system 6 VV eak hqu d 286
conversion constants 362 VV et bulb depress on 302

Universal gas constant 60 203 362 W et bulb temperat ire 302 304
Uranium 248 in total heat 308
Wet mixture 192
properties of 201
VV llson 1 ne 266
Vacuum pressure 10 Work 19
Vapor 55 191 brake 21
critical point of 195 208 convention of s gns 21
cycles 222 cycle v s eng ne 222
distinction from gas 55 delivered 21
gas mixtures 298 flow 24
for revers ble nonflow process 21
line 195
INDEX S79
from impulse-momentum, 166 shaft, 21
ideal, 20 units, 21
indicated, 20 Working substance, 3, 90
lost, 213, 235, 23G, 2G9
of compressor, 12G, 129, 131, 138
of cycle, 91
of steady flow process, 127 Zeroth law, 23

You might also like