Visit https://ebookfinal.

com to download the full version and

explore more ebooks

Axial flow compressors a strategy for

aerodynamic design and analysis 1st Edition
Ronald H. Aungier

_____ Click the link below to download _____

https://ebookfinal.com/download/axial-flow-compressors-
a-strategy-for-aerodynamic-design-and-analysis-1st-
edition-ronald-h-aungier/

Explore and download more ebooks at ebookfinal.com

Here are some suggested products you might be interested in.
Click the link to download

Design and Analysis of Centrifugal Compressors 1st Edition

Rene Van Den Braembussche

https://ebookfinal.com/download/design-and-analysis-of-centrifugal-
compressors-1st-edition-rene-van-den-braembussche/

Practical stress analysis in engineering design 3rd ed

Edition Ronald Huston

https://ebookfinal.com/download/practical-stress-analysis-in-
engineering-design-3rd-ed-edition-ronald-huston/

Practical Stress Analysis in Engineering Design 3rd

Edition Ronald Huston (Author)

https://ebookfinal.com/download/practical-stress-analysis-in-
engineering-design-3rd-edition-ronald-huston-author/

Microelectronic Circuits Analysis and Design 2nd Edition

Muhammad H. Rashid

https://ebookfinal.com/download/microelectronic-circuits-analysis-and-
design-2nd-edition-muhammad-h-rashid/
Jet Propulsion A Simple Guide to the Aerodynamic and
Thermodynamic Design and Performance of Jet Engines 2nd
Edition Cumpsty
https://ebookfinal.com/download/jet-propulsion-a-simple-guide-to-the-
aerodynamic-and-thermodynamic-design-and-performance-of-jet-
engines-2nd-edition-cumpsty/

Network Flow Analysis 1st Edition Michael W. Lucas

https://ebookfinal.com/download/network-flow-analysis-1st-edition-
michael-w-lucas/

Patent Strategy for Researchers and Research Managers 3rd

Edition H. Jackson Knight

https://ebookfinal.com/download/patent-strategy-for-researchers-and-
research-managers-3rd-edition-h-jackson-knight/

Flow Networks Analysis and optimization of repairable flow

networks networks with disturbed flows static flow
networks and reliability networks 1st Edition Michael T.
Todinov
https://ebookfinal.com/download/flow-networks-analysis-and-
optimization-of-repairable-flow-networks-networks-with-disturbed-
flows-static-flow-networks-and-reliability-networks-1st-edition-
michael-t-todinov/

Theories Of Comparative Politics The Search For A Paradigm

Reconsidered Second Edition Ronald H. Chilcote.

https://ebookfinal.com/download/theories-of-comparative-politics-the-
search-for-a-paradigm-reconsidered-second-edition-ronald-h-chilcote/
Axial flow compressors a strategy for aerodynamic design
and analysis 1st Edition Ronald H. Aungier Digital Instant
Download
Author(s): Ronald H. Aungier
ISBN(s): 9781860584220, 1860584225
Edition: 1
File Details: PDF, 3.56 MB
Year: 2003
Language: english
 AXIAL-FLOW
COMPRESSORS
A STRATEGY FOR AERODYNAMIC DESIGN AND ANALYSIS

Ronald H. Aungier

NEW YORK ASME PRESS 2003

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 © 2003 by The American Society of Mechanical Engineers
Three Park Avenue, New York, NY 10016
ISBN: 0-7918-0192-6

Co-published in the UK by Professional Engineering Publishing Limited,

Northgate Avenue, Bury St Edmunds, Suffolk, IP32 6BW, UK
ISBN: 1-86058-422-5

All rights reserved. Printed in the United States of America. Except as permitted under the
United States Copyright Act of 1976, no part of this publication may be reproduced or dis-
tributed in any form or by any means, or stored in a database or retrieval system, without
the prior written permission of the publisher.

Statement from By-Laws: The Society shall not be responsible for statements or opinions
advanced in papers . . . or printed in its publications (B7.1.3)

INFORMATION CONTAINED IN THIS WORK HAS BEEN OBTAINED BY THE AMERI-

CAN SOCIETY OF MECHANICAL ENGINEERS FROM SOURCES BELIEVED TO BE
RELIABLE. HOWEVER, NEITHER ASME NOR ITS AUTHORS OR EDITORS GUARAN-
TEE THE ACCURACY OR COMPLETENESS OF ANY INFORMATION PUBLISHED IN
THIS WORK. NEITHER ASME NOR ITS AUTHORS AND EDITORS SHALL BE RESPON-
SIBLE FOR ANY ERRORS, OMISSIONS, OR DAMAGES ARISING OUT OF THE USE OF
THIS INFORMATION. THE WORK IS PUBLISHED WITH THE UNDERSTANDING THAT
ASME AND ITS AUTHORS AND EDITORS ARE SUPPLYING INFORMATION BUT ARE
NOT ATTEMPTING TO RENDER ENGINEERING OR OTHER PROFESSIONAL SER-
VICES. IF SUCH ENGINEERING OR PROFESSIONAL SERVICES ARE REQUIRED,
THE ASSISTANCE OF AN APPROPRIATE PROFESSIONAL SHOULD BE SOUGHT.

For authorization to photocopy material for internal or personal use under circumstances
not falling within the fair use provisions of the Copyright Act, contact the Copyright Clear-
ance Center (CCC), 222 Rosewood Drive, Danvers, MA 01923, Tel: 978-750-8400,
www.copyright.com.

Library of Congress Cataloging-in-Publication Data

Aungier, Ronald H.
Axial-Flow compressors : a strategy for aerodynamic design and analysis / Ronald
Aungier.
p. cm.
ISBN 0-7918-0192-6
1. Compressors—Aerodynamics. 2. Compressors—Design and construction. I. Title.

TJ267.5.C5 A94 2003

621.5’1—dc21 20020385720

Cover photo: Courtesy of Elliott Turbomachinery Co., Inc., Ebara Group

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 To Anne

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a
 TABLE OF CONTENTS

Preface xi

1 Introduction 1
1.1 Axial-Flow Compressor Basics .........................................................3
1.2 Basic Velocity Diagrams for a Stage................................................5
1.3 Similitude and Performance Characteristics...................................7
1.4 Stage Matching and Stability........................................................11
1.5 Dimensionless Parameters.............................................................13
1.6 Units and Conventions ..................................................................14

2 Thermodynamics 17
2.1 First and Second Laws of Thermodynamics................................18
2.2 Efficiency.......................................................................................20
2.3 Fluid Equation-of-State Fundamentals.......................................22
2.4 The Caloric Equation of State .....................................................24
2.5 Entropy and the Speed of Sound................................................25
2.6 The Thermal Equation of State for Real Gases ..........................26
2.7 Thermodynamic Properties of Real Gases ..................................30
2.8 Thermally and Calorically Perfect Gases .....................................31
2.9 The Pseudo-Perfect Gas Model ...................................................32
2.10 Component Performance Parameters ........................................33
2.11 Gas Viscosity .................................................................................37
2.12 A Computerized Equation-of-State Package .............................37

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 vi • Table of Contents

3 Fluid Mechanics 41
3.1 Flow in a Rotating Coordinate System .........................................43
3.2 Adiabatic Inviscid Compressible Flow...........................................46
3.3 Adiabatic Inviscid Compressible Flow Applications .....................48
3.4 Boundary Layer Analysis................................................................50
3.5 Two-Dimensional Boundary Layer Analysis..................................51
3.6 Axisymmetric Three-Dimensional Boundary Layer Analysis........54
3.7 Vector Operators in Natural Coordinates.....................................57

4 Axial-Flow Compressor Blade Profiles 59

4.1 Cascade Nomenclature ................................................................60
4.2 NACA 65-Series Profile.................................................................62
4.3 Circular-Arc Camberline...............................................................65
4.4 Parabolic-Arc Camberline ............................................................66
4.5 British C.4 Profile..........................................................................68
4.6 Double-Circular-Arc Profile..........................................................69
4.7 NACA A4K6 63-Series Guide Vane Profile ...................................70
4.8 Controlled–Diffusion Airfoils.......................................................71
4.9 Blade Throat Opening .................................................................73
4.10 Staggered Blade Geometry .........................................................75

5 Two-Dimensional Blade-to-Blade Flow Through

Cascades of Blades 77
5.1 The Blade-to-Blade Flow Problem ................................................79
5.2 Coordinate System and Velocity Components .............................81
5.3 Potential Flow in the Blade-to-Blade Plane .................................83
5.4 Linearized Potential Flow Analysis ...............................................92
5.5 The Time-Marching Method .........................................................96
5.6 Blade Surface Boundary Layer Analysis......................................107
5.7 Summary.......................................................................................113

6 Empirical Performance Models Based On

Two-Dimensional Cascade Tests 117
6.1 Cascade Geometry and Performance Parameters....................119
6.2 Design Angle of Attack or Incidence Angle .............................121

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 Table of Contents • vii

6.3 Design Deviation Angle.............................................................125

6.4 Design Loss Coefficient and Diffusion Factors .........................128
6.5 Positive and Negative Stall Incidence Angles...........................134
6.6 Mach Number Effects.................................................................136
6.7 Shock Wave Loss for Supersonic Cascades................................138
6.8 Off-Design Cascade Performance Correlations ........................141
6.9 Blade Tip Clearance Loss............................................................146
6.10 Shroud Seal Leakage Loss..........................................................147
6.11 Implementation, Extensions and Alternate Methods .............149

7 Meridional Through-Flow Analysis 153

7.1 Meridional Coordinate System ...................................................154
7.2 Inviscid, Adiabatic Flow on a Quasi-Normal...............................157
7.3 Linking Quasi-Normals ................................................................162
7.4 Repositioning the Stream Surfaces.............................................164
7.5 Full Normal Equilibrium Solution ...............................................165
7.6 Simplified Forms of the Through-Flow Analysis ........................167
7.7 Annulus Sizing .............................................................................169
7.8 Numerical Approximations .........................................................171

8 End-Wall Boundary Layer Analysis 175

8.1 Historical Development of End-Wall Boundary Layer Theory .....177
8.2 The End-Wall Boundary Layer Equations ...................................181
8.3 The Boundary Layer Velocity Profile Assumptions ....................183
8.4 Empirical Models for Entrainment and Wall Shear Stress .........184
8.5 The Blade Force Defect Thicknesses ...........................................187
8.6 Seal Leakage Effects for Shrouded Blades .................................191
8.7 Boundary Layer Jump Conditions ...............................................193
8.8 Solution Procedure ......................................................................194
8.9 Typical Results ..............................................................................195

9 Aerodynamic Performance Analysis 199

9.1 Geometry Considerations............................................................200

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 viii • Table of Contents

9.2 Cascade Performance Considerations.........................................203

9.3 Stall and Compressor Surge Considerations...............................204
9.4 Approximate Normal Equilibrium Results..................................207
9.5 Full Normal Equilibrium Results..................................................211
9.6 Concluding Remarks ....................................................................213

10 Compressor Stage Aerodynamic Design 215

10.1 Dimensionless Performance Parameters.................................217
10.2 Application to Stage Design....................................................219
10.3 Blade Design.............................................................................221
10.4 Selecting the Stage Performance Parameters ........................222
10.5 Selecting the Swirl Vortex Type...............................................229
10.6 Free Vortex Flow ......................................................................230
10.7 Constant Reaction Vortex Flow...............................................235
10.8 Constant Swirl and Exponential Vortex Flow.........................242
10.9 Assigned Flow Angle Vortex Flows .........................................245
10.10 Application to a Practical Stage Design .................................245
10.11 A Repeating Stage Axial-Flow Compressor............................251
10.12 A Computerized Stage Design System ...................................257

11 Multistage Axial-Flow Compressor Aerodynamic

Design 259
11.1 The Basic Compressor Design Approach ..................................261
11.2 Aerodynamic Performance Specifications................................262
11.3 Blade Design ..............................................................................264
11.4 Refining the Compressor Design ..............................................266
11.5 An Axial-Flow Compressor Design Example.............................268
11.6 The Distribution of Stage Performance Parameters................272
11.7 The Swirl Vortex Type ................................................................280
11.8 Risks and Benefits ......................................................................284

12 Quasi-Three-Dimensional Blade Passage Flow

Field Analysis 287
12.1 Quasi-Three-Dimensional Flow .................................................289
12.2 Hub-to-Shroud Flow Governing Equations ..............................291

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 Table of Contents • ix

12.3 Numerical Integration of the Governing Equations................294

12.4 Repositioning Stream Surfaces .................................................297
12.5 The Hub-to-Shroud Flow Analysis.............................................298
12.6 Coupling the Two Basic Flow Analyses .....................................299
12.7 Boundary Layer Analysis............................................................302

13 Other Components and Variations 309

13.1 Adjustable Blade Rows ..............................................................311
13.2 The Exhaust Diffuser..................................................................316
13.3 The Scroll or Collector ...............................................................322
13.4 Reynolds Number and Surface Roughness Effects...................328
13.5 The Axial-Centrifugal Compressor............................................328

Answers to the Exercises 333

References 349

About the Author 355

Index 357

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a
 PREFACE

Books on compressor aerodynamics approach the subject in various ways

depending on the intended audience. Introductions to the fundamentals are
available to students and newcomers to the field. The experienced aerodynami-
cist can find more comprehensive overviews of the core technologies that review
alternate approaches and summarize important contributions that have signifi-
cantly advanced the state of the art. Users of compressors can find books that
address the alternatives and critical issues of particular interest to the applica-
tion, selection, procurement and operation of compressors. But in recent years,
the aerodynamicist seeking detailed information and practical guidance on the
effective application of basic technology to compressor aerodynamic design and
analysis has been less fortunate.
When I started working in this field, books of that type were available,
although they were somewhat dated. The Centrifugal Compressor Stage (Fergu-
son, 1963), Axial Flow Compressors (Horlock, 1958) and Aero-Thermodynamics
and Flow in Turbomachines (Vavra, 1960) were my particular favorites, and they
provided valuable guidance during my early efforts to formulate compressor
aerodynamic design and analysis systems. Unfortunately, their modern equiva-
lents have simply not appeared. With his usual insight, Cumpsty (1989) recog-
nizes and discusses this change in emphasis in recent books in his preface to
Compressor Aerodynamics. He attributes it to the expanded roles of the computer
and of proprietary industrial research and development, which have combined to
make it rather impractical for authors to provide a general and detailed descrip-
tion of modern compressor aerodynamic design and analysis methods. Several
years ago, similar reasoning led me to consider a different type of book to meet
this need. If a general description of aerodynamic design and analysis methods is
no longer practical, a detailed description of a specific aerodynamic design and
analysis system might be a viable alternative. Although unsure of how it might be
received, I decided to write a book on centrifugal compressor aerodynamics and
managed to find a publisher willing to try something different (Aungier, 2000).
The feedback received since my book was published confirms that there is inter-
est in this type of book as well as a significant audience.
Aerodynamic design and analysis systems evolve through a trial-and-error
process while we painfully learn what capabilities should be included, how to use
them effectively and how they can interact efficiently to support the overall
process. The conversion of basic technology into a working design and analysis

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 xii • AXIAL-FLOW COMPRESSORS

system is a very complex process involving many decisions and false starts. Along
the way, all developers expend a lot of time and effort formulating virtually iden-
tical models for phenomena fundamental to the process. These critical details are
quite important to aerodynamicists who must develop, maintain or improve an
aerodynamic design and analysis system. The present book approaches axial-
flow compressor aerodynamics in a manner similar to that of my previous book
on centrifugal compressors. The approach used is a description of a comprehen-
sive aerodynamic design and analysis system in sufficient detail so that readers
can readily implement the complete system or any of its components.
This proved to be more difficult for axial-flow compressors than was the case
for centrifugal compressors. Centrifugal compressor stage designs tend to be
fairly unique and specific to the design objectives. Most current design and analy-
sis systems share many common features and are used in a wide range of appli-
cations. The basis for axial-flow compressor design is far more varied and
application-dependent, often based on proprietary information that is not avail-
able in the open literature. Indeed, it is now quite common to find designs based
on proprietary and customized airfoil families, such as the popular controlled
diffusion airfoil. The geometry and performance characteristics of these propri-
etary airfoils are well known only to the organizations that developed them.
There are also inherent differences in the technology used on industrial axial-
flow and centrifugal compressors. The development of a completely original
industrial axial-flow compressor design is relatively rare. These compressors are
almost always unique, one-of-a-kind designs that must rely on variations of stan-
dard components to minimize risks while maintaining acceptable development
and manufacturing costs. Hence, the variety of application experience is far more
limited for the axial-flow compressor design and analysis system than was the
case for the centrifugal compressor system.
I decided that the desired objectives could still be achieved by adopting the
classical design approach based on the systematic application of standard airfoil
families to develop the blade geometry used in the compressors. The basic prin-
ciples of the design process described here remain applicable when proprietary
airfoil families are in use. But it is likely to be necessary to adapt them to reflect
the specific geometry and performance characteristics of those airfoils. Aerody-
namic performance prediction accuracy is established by comparing predictions
with experimental data for several axial-flow compressors. That established per-
formance prediction accuracy is then used to demonstrate the effectiveness of
the overall design and analysis system. A substantial number of design examples
are included to illustrate the use of this design and analysis system, as well as to
provide some evaluation of alternate design approaches suggested in the litera-
ture, or which I have found to be effective.
Considerable care is taken to provide complete and detailed descriptions of
this comprehensive aerodynamic design and analysis system for axial-flow com-
pressors. The basic principles of thermodynamics and fluid mechanics required
are presented in a form particularly well-suited to the axial-flow compressor
application. Well-defined empirical models are used to augment these basic prin-
ciples to address the essential problem areas of performance analysis, stage
design, compressor design and internal flow analysis. Descriptions of numerical
methods used are included as well as other critical considerations important to

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 Preface • xiii

readers who may wish to apply these methods. In a few cases where components
are common to both my centrifugal and axial-flow compressor design and analy-
sis systems, I refer to Aungier (2000) rather than repeat some rather lengthy and
detailed descriptions that will be of interest only to readers who choose to imple-
ment those specific methods.
Some important topics have received inadequate treatment or have been com-
pletely omitted. Surge and stall are discussed only in the context of estimating
the expected limits of stable operation, while noise and blade vibration are not
discussed at all. I prefer to limit myself to topics on which I can offer at least
some original ideas. I always feel uncomfortable when presenting ideas obtained
almost entirely from others, even when I have considerable confidence in the
sources. I much prefer that readers obtain such information from qualified
authors, even if it does display my own limitations.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 Chapter 1

INTRODUCTION

Compressors are commonly classified as either positive displacement or dynamic

compressors. The positive displacement compressor achieves its pressure rise by
trapping fluid in a confined space and transporting it to the region of higher pres-
sure. The dynamic compressor develops its increase in pressure by a dynamic
transfer of energy to a continuously flowing fluid stream. There are two basic
types of dynamic compressors: axial-flow compressors and centrifugal (radial-
flow) compressors. The flow streamlines through rotating rows in an axial-flow
compressor have a radius that is almost constant, whereas they undergo a sub-
stantial increase in radius in a centrifugal compressor. For this reason, the cen-
trifugal compressor can achieve a much greater pressure ratio per stage than the
axial-flow compressor. But the axial-flow compressor can achieve a significantly
greater mass flow rate per unit frontal area. Figure 1-1 compares normalized dis-
charge pressure, P, versus flow rate, Q, for these two compressor types to illus-
trate the differences in their performance characteristics. The axial-flow
compressor approximates a variable pressure ratio—constant flow machine,
whereas the centrifugal compressor is closer to a constant pressure ratio—vari-
able flow machine. The performance data displayed in Fig. 1-1 are for a single-
stage centrifugal compressor and a five-stage axial-flow compressor, both of
which have about the same design pressure ratio. This demonstrates the superior
pressure ratio-per-stage capability of the centrifugal compressor. Traditionally,
the centrifugal compressor has been the more rugged and lower-cost type, while
the axial-flow compressor has offered better efficiency. Those differences have
become much less significant in recent years due to advances in technology, par-
ticularly with regard to efficiency. Presently, the compressor type selected is more
likely to be based on the performance characteristics, size and cost that is best
suited to the application.

NOMENCLATURE

a = sound speed
C = absolute velocity
cp = specific heat at constant pressure
cv = specific heat at constant volume
H = total enthalpy and compressor head

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 2 • AXIAL-FLOW COMPRESSORS

h = static enthalpy
k = ratio of specific heats = cp / cv
M = Mach number = C / a
ṁ = mass flow rate
N = rotation speed (rpm)
P = pressure
Q = volume flow rate = ṁ / ρt
R = Gas constant and stage reaction
Re = Reynolds number
r = radius
T = temperature
U = local blade speed, ωr
W = relative velocity
β = flow angle
γ = stagger angle
δ = Pt0 / Pref
η = efficiency
θ = polar (tangential) coordinate and Tt0 / Tref
µ = viscosity
ρ = density
φ = stage flow coefficient
ψ = stage work coefficient
ω = rotation speed (radians/second)

FIGURE 1-1 Performance of Dynamic Compressors

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 Introduction • 3

Subscripts

d = parameter at compressor discharge

ref = reference thermodynamic conditions
rev = reversible thermodynamic process
t = total thermodynamic condition
z = axial component
θ = swirl (tangential) component
0 = parameter at guide vane inlet or compressor inlet
1 = parameter at guide vane exit and rotor inlet
2 = parameter at rotor exit and stator inlet
3 = parameter at stator exit

Superscripts

′ = relative condition

1.1 AXIAL-FLOW COMPRESSOR BASICS

Figure 1-2 illustrates the basic configuration of an axial-flow compressor. The first
blade row shown is an inlet guide vane (IGV) to develop the swirl (tangential)
velocity for which the first rotating row (R1) was designed. If the first rotating row
is designed for no inlet swirl, the inlet guide vane will normally be omitted. This is
followed by a series of stages (two in this illustration), where a stage refers to a
rotating row, or rotor, in combination with its downstream stationary row, or sta-
tor (e.g., R1 and S1). The rotor row imparts energy to the fluid by increasing the
swirl velocity. The stator row removes the swirl developed by the rotor to convert
kinetic energy to static pressure and to establish the proper swirl velocity for the
flow to enter the next rotor. Typically, an exit guide vane (EGV) follows the last
stage to remove any residual swirl velocity to convert that kinetic energy to static
pressure. Like the inlet guide vane, this row may be omitted if the last stator is
designed to remove all of the swirl velocity. Although not shown on the figure, a
diffuser-combustor (gas turbine) or diffuser-collector (industrial compressor) will
follow the exit guide vane to recover as much kinetic energy as possible, as well as
to direct the flow to its intended destination. Similarly, an inlet passage will pre-
cede the inlet guide vane. This can range from a smooth axial bell-mouth inlet to
a complex side inlet, depending on the compressor’s application.
Figure 1-3 illustrates the blade profiles for a stage viewed on a polar stream
surface between adjacent blades. The rotor row is rotating with a velocity, U = ωr,
where ω is the rotation speed and r is the radius. Viewed in a frame of reference
rotating with the rotor, the upstream velocity, W, is referred to as the relative
velocity. The rotor deflects the flow such that the velocity in the stationary frame
of reference of the stator (the absolute velocity), C, is properly aligned to enter
the stator row. This process repeats in subsequent stages, with each stage adding
energy to the fluid to achieve the overall pressure ratio required.
Axial-flow compressor design strategies are quite varied. Gas turbine com-
pressors are normally intended for use in many identical units. Extensive design

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 4 • AXIAL-FLOW COMPRESSORS

FIGURE 1-2 Axial-Flow Compressor Configuration

FIGURE 1-3 Polar Surface View of a Stage

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 Introduction • 5

and development programs are common and may include unique blade designs
for all blade rows. By contrast, industrial compressors are usually designed
specifically for a unique application, rarely involving any duplicate units. Here, a
repeating stage approach is more common, where one or more basic stage
designs are used for all compressors. Aerodynamic and mechanical flexibility are
obtained by minor adjustments that do not compromise the basic stage’s per-
formance. Blades may be scaled to longer and thicker blades for mechanical
integrity with a corresponding adjustment of the number of blades per row to
preserve aerodynamic similarity. Blades usually must be restaggered, i.e., rotated
on their base to change the stagger angle, γ, to achieve different performance lev-
els. Otherwise, the intended duty would normally require a non-integer number
of stages. Often the inlet guide vane and some of the stator blades may be
adjustable so they can be restaggered by a control system while the machine is in
operation to broaden the compressor’s application range. This approach allows
these “one-of-a-kind” compressors to be designed within practical cost. It also
allows each compressor’s design to be based on a well-established performance
history. This is important, since these compressors cannot be confirmed by per-
formance testing until after they are manufactured.
Figure 1-2 illustrates normal cantilevered blades that are attached at the root,
with a clearance between the blade tip and the adjacent end-wall. Figure 1-4
shows a different style often used for stator blades. Here, a shroud band is
attached to the blade tips to connect them together. This is often done for reasons
of mechanical integrity. To reduce fluid leakage from the blade discharge back to
the blade inlet, seal fins are normally attached to the shroud band. These provide
a reduced clearance to retard leakage, yet are thin enough to minimize damage in
the event that a rotor shaft excursion or “rub” causes the seals and shaft to come
into contact. To minimize damage to the shaft, the stator blades and stator
shrouds, the seal fins will be sacrificed in the event of a rub.

1.2 BASIC VELOCITY DIAGRAMS FOR A STAGE

The construction of velocity diagrams is a very useful concept for axial-flow com-
pressor design. Here they will be used to illustrate the velocity vectors entering
and leaving blade rows in a stage. It will be necessary to use both absolute and
relative velocities, where relative velocities are viewed in a frame of reference
rotating with the compressor’s rotation speed, ω. Designating the relative and
absolute tangential velocities as Wθ and Cθ, respectively, the two are related by

Wθ = Cθ − ω r (1-1)

where r is the local radius. The axial components of velocity are identical in both
frames of reference, i.e.,

Wz = Cz (1-2)

Therefore, the absolute and relative velocities are

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 6 • AXIAL-FLOW COMPRESSORS

FIGURE 1-4 Shrouded Stator Blades

C = Cz2 + Cθ2
(1-3)

W = Cz2 + Wθ2
(1-4)

The absolute and relative flow angles are designated as β and β′, respectively.
They are defined by

tan β = Cθ / Cz (1-5)
tan β ′ = Wθ / Cz
(1-6)

Figure 1-5 illustrates the velocity diagrams for an inlet guide vane. The flow
enters the guide vane with no swirl, i.e., Cθ 0 = 0, C0 = Cz0. The guide vane deflects
the flow by an angle, β1. If Cz1 is known, this defines the swirl velocity compo-
nent, Cθ1. Then Eq. (1-1) is applied in vector form to subtract ωr from Cθ1 to
define the swirl velocity component in the relative frame, Wθ1 and the relative
flow angle, β′1. Hence, the complete velocity diagram for the entrance to the
downstream (rotating) rotor blade row is known. Figure 1-6 shows the velocity
diagram construction for the rotor blade row. The inlet velocity diagram is the
same as that determined for the guide vane exit. The rotor blade deflects the flow
in the relative frame of reference from β′1 to β′2 to produce the discharge swirl
velocity, Wθ2. If Cz2 is known, Wθ 2 can be computed. Then vector addition of ωr to

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 Introduction • 7

FIGURE 1-5 Guide Vane Velocity Diagrams

Wθ2 yields the absolute swirl component Cθ 2. Hence, all velocity components and
the flow angle in the absolute frame of reference can be computed to define the
inlet conditions for the downstream (stationary) stator blade row. Construction
of the stator and exit guide vane blade row velocity diagrams is accomplished in
a similar fashion and will be left as an exercise for the reader. The important
thing to note is that construction of this simple velocity diagram is a fundamen-
tal technique commonly used by turbomachinery aerodynamicists to convert
between absolute and relative flow conditions. Here, Cz has been treated as
known. In practice, values of Cz may be specified design conditions from which
the flow passage areas will be computed to conserve mass. This will be referred to
as the design mode. Alternatively, Cz may be computed from basic mass and
momentum conservation for specified passage areas and the mass flow rate. This
will be referred to as the analysis mode.

1.3 SIMILITUDE AND PERFORMANCE CHARACTERISTICS

Similitude or similarity is one of the most useful concepts in turbomachinery

aerodynamics. Two turbomachines are completely similar if the ratios of all

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 8 • AXIAL-FLOW COMPRESSORS

FIGURE 1-6 Rotor Blade Velocity Diagrams

corresponding length dimensions, velocity components and forces are equal

(Sheppard, 1956). If two turbomachines are completely similar, it is possible to
present their performance on a common performance map by selecting appropri-
ate equivalent performance parameters. Equivalent performance requires that the
two compressors have similar velocity diagrams throughout. To maintain similar
velocity diagrams while conserving the mass flow rate, ṁ, the ratio of gas density-
to-inlet density must also be similar throughout. This means that inlet volume
flow, Q0, is the relevant equivalent flow rate parameter, where

˙ / ρt
Q=m (1-7)

The local axial velocity is actually given by

˙ / ( A0ρ0 )
Cz0 = m (1-8)

where A is the passage area and ρ0 is the inlet gas density, which is unknown.
But the exercises in Chapter 2 will show that the ratio ρ0 / ρt0 is a function of Cz0,

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 Introduction • 9

where ρt0 is the known inlet total gas density. Hence, unique velocity diagrams
are associated with a unique Q0 / A0, but can correspond to many values of ṁ/A0
simply by altering ρt0. For this reason, all dynamic compressors are commonly
referred to as volume flow machines. The exercises in Chapter 2 will show that
the requirement for density ratio similarity requires that the Mach numbers be
similar throughout, where the Mach number is the ratio of fluid velocity to the
local sound speed, a. It will be shown that the ratio of a0 / at0 is, itself, a function
of the Mach number, so the unknown a0 can be replaced by the known inlet total
sound speed, at0. Figure 1-7 shows an equivalent performance map based on
these requirements. The flow parameter used is volume flow normalized by the
inlet total sound speed, at0, and the inlet area, A0. This ensures that the inlet
axial Mach numbers will be similar. Three performance characteristics, or speed
lines, are shown for three different rotation speeds: N, multiplied by a charac-
teristic diameter, D, and normalized by the inlet total sound speed. This will
ensure similarity of the tangential Mach numbers. If two axial-flow compressors
are geometrically similar, and use the same working fluid, this performance map
will apply to both machines. This, in turn, ensures that the pressure ratios will
be the same for both, so pressure ratio is a reasonable choice for the other per-
formance parameter. The situation becomes more complicated if the two com-
pressors use different working fluids. In that case, complete similarity usually
cannot be achieved, since different working fluids may produce different gas
density (or specific volume) ratios for the same blade row velocity diagram. This
“volume-ratio” effect will compromise similarity after the first rotor row, since

FIGURE 1-7 An Equivalent Performance Map

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 10 • AXIAL-FLOW COMPRESSORS

the differences in density will necessitate differences in axial velocity to conserve

mass. Indeed, even with identical working fluids, non-ideal gas behavior may
compromise similarity. True equivalent performance is assured only for working
fluids that obey the perfect gas equation of state discussed in Chapter 2, which
can be expressed as:

P = ρRT (1-9)

where P is pressure, T is temperature and R is the gas constant. An exercise in

Chapter 2 will also show that achieving equivalent density ratios requires that the
two working fluids have the same ratio of specific heats, k = cp / cv. Since axial-
flow compressors are widely used in aircraft engine gas turbines, discussions on
this topic often overlook these more subtle effects. That application of compres-
sors deals exclusively with air as the working fluid, which is very nearly a perfect
gas, offering little chance of any volume-ratio effect. Consequently, equivalent
performance maps must be used with caution when a perfect gas model with a
constant value of k cannot approximate the working fluid.
Similarity is also compromised when the two compressors operate at sub-
stantially different Reynolds number, Re. Reynolds number is a measure of the
inertia forces to viscous forces, Re = ρCL/µ, where L is a characteristic length
and µ is the gas viscosity. The Reynolds number directly affects wall friction,
which can alter the compressor’s performance. In most cases, effects of the
Reynolds number are small enough to be neglected. When that is not the case,
suitable Reynolds number corrections (e.g., Wassell, 1968) may be applied to
adjust the performance.
A common use of equivalent performance maps is to define performance of a
specific compressor that is operated at various inlet total thermodynamic condi-
tions. For that case, D and A0 are constant and can be omitted on the map.
Another common use is to relate performance of different compressor frame
sizes, derived by directly scaling the geometry. Then, D and A0 are included so the
map defines the change in speed, N, needed to preserve Mach number equiva-
lence and the flow rate supplied for each pressure ratio on a speed line.
The equivalent flow rate used in Fig. 1-7 is the true similarity parameter. When
the application is to a perfect gas with constant R and k, Eq. (1-9) can be used to
derive alternate equivalent flow parameters, i.e.,

Q / at0 ∝ m ˙ θ /δ
˙ Tt0 / Pt0 ∝ m (1-10)

where the sound speed has been replaced by a perfect gas relation from chapter 2,

a = kRT (1-11)

and θ and δ relate inlet total conditions to reference conditions (Tref, Pref), such as
standard atmosphere conditions, i.e.,

θ = Tt0 / Tref (1-12)

δ = Pt0 / Pref (1-13)

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 Introduction • 11

Similarly, the equivalent speed parameter can be replaced, using

N / at0 ∝ N / Tt0 ∝ N / θ (1-14)

The alternate flow rate and speed parameters in Eqs. (1-10) and (1-14) are com-
monly used, but have less fundamental significance than Q0 / at0 and N / at0.
Similarly, compressor head, ∆Hrev, can be used in place of pressure ratio,
where head is defined as the total enthalpy increase required to produce the
actual pressure rise by an ideal, reversible process, i.e.,

dp
∆Hrev = ∫ ρ
(1-15)
rev

It can be shown that the appropriate equivalent head parameter is ∆Hrev / (at0)2.
The use of an equivalent head is common practice for centrifugal compressors,
but is much less common for axial-flow compressors.
Figure 1-7 supplies only part of the performance information required. In
addition to the pressure ratio and flow produced, it is necessary to know the
work required to drive the compressor. Hence, a second equivalent perform-
ance map is required to completely define the compressor’s performance. The
most common parameter for this purpose is efficiency, η, defined as the com-
pressor head or ideal (no loss) total enthalpy rise divided by the actual total
enthalpy rise, i.e.,

η = ∆Hrev / ∆H (1-16)

Alternate reversible processes that can be used to define ∆Hrev and η are dis-
cussed in Chapter 2. Figure 1-8 shows an equivalent efficiency map to be used in
conjunction with Fig. 1-7. In some cases, it may be appropriate to use the exit
static thermodynamic conditions rather than total values to define η and PR. This
is appropriate when the kinetic energy available at the compressor discharge
serves no useful purpose for the specific application to which the compressor will
be applied.

1.4 STAGE MATCHING AND STABILITY

Each blade row in a compressor will achieve its best performance for a specific
inlet flow angle where losses are minimum. Basically, the designer seeks to
“match” succeeding blade rows such that all operate close to their optimum inlet
flow angles at a specific operating condition, commonly called the compressor’s
design point or match point, defined by the design flow rate and design speed.
Hence, at design speed, losses can be expected to increase and performance to
deteriorate as the compressor operates farther from its design flow rate. At flow
rates less than the design flow rate, losses will increase to a point that the pres-
sure-flow rate characteristic reaches a maximum. At lower flow rates, the char-
acteristic will have a positive slope, which is theoretically unstable. The onset of

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 12 • AXIAL-FLOW COMPRESSORS

FIGURE 1-8 An Equivalent Efficiency Map

this severe unstable operation is commonly called surge. The limit of stable oper-
ation is referred to as the surge line as illustrated in Fig. 1-7. Surge is a very com-
plex phenomenon, which depends on the entire system, not just the compressor.
So associating it with a maximum on the pressure–flow rate characteristic is an
oversimplification, but a useful one. In some cases, an approach to zero-slope
near surge is evident, such as for speed line N3 in Fig. 1-7. In other cases, the
compressor may experience an abrupt stall, such that the characteristic appears
to be quite steep at surge, similar to speed line N1 in Fig. 1-7. This is mainly
because the drop in pressure with reduced flow is so abrupt that it cannot be
resolved in a performance test. Indeed, estimation of the onset of surge during
the design phase is based more on the expected blade loading limits at the onset
of stall than on the predicted shape of the pressure-flow characteristics. Simi-
larly, at flow rates greater than the design flow rate, the increase in loss will even-
tually result in no rise in pressure. This condition is commonly referred to as
choke, although it may be caused by large losses due to off-design operation
rather than a true aerodynamic choke condition.
When the compressor is operated at off-design speeds, operation at different
Mach number levels will compromise the stage matching, similar to the volume-
ratio effect mentioned previously. Consequently, it is unlikely that all stages will
be close to their optimum operating conditions at any flow rate for off-design
speeds. Rather, optimum performance will occur at the flow rate offering the
best compromise on stage matching. Performance will deteriorate for flows dif-
ferent from this optimum, much as described for the design speed performance.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 Introduction • 13

The more speed deviates from the design speed, the greater the compromise of
the stage matching. In general, at speeds lower than the design speed, the front
stages are required to supply a greater portion of the rise in pressure while the
rear stages become less effective. The inverse is true for speeds greater than the
design speed. This stage mismatching can be alleviated to some degree if some
of the stationary blade rows are adjustable during operation. Closing some of
the stationary rows (i.e., increasing their stagger angles) in a controlled fashion
will shift the optimum matching condition to lower speeds to reduce the mis-
match at low speeds.
The Mach number level has a definite influence above and beyond its pro-
nounced effect on stage matching. As a blade row’s inlet Mach number increases,
its low-loss operating range will decrease. At sufficiently high values, the blade
row will start to experience aerodynamic choke in the blade row to significantly
reduce its maximum flow capacity. Even the minimum loss levels will increase
when the inlet Mach number becomes high enough to produce shock waves that
are strong enough to induce boundary layer separation or to produce significant
bow shock losses. Consequently, as the equivalent speed increases, pressure-flow
characteristics become steeper, with less flow range from surge to choke as illus-
trated in Fig. 1-7, and the maximum achievable efficiency can be limited by Mach
number levels, similar to speed line N1 in Fig. 1-8.

1.5 DIMENSIONLESS PARAMETERS

In addition to the dimensionless parameters associated with similitude, axial-

flow compressor design is often based on a number of other useful aerodynamic
dimensionless parameters. To introduce these parameters, it is necessary to
anticipate some results that are developed in more detail in Chapters 2 and 3. In
particular, the total enthalpy rise for simple axial flow through a rotor blade row
is expressed by the well-known Euler turbine equation, i.e.,

∆H = U (Cθ 2 − Cθ1) (1-17)

where U = ω r is the local blade speed and H is the total enthalpy. Similarly, it will
be shown in Chapter 2 that the static enthalpy, h, is related to H by

H = h + 12 C 2 (1-18)

It is often useful to introduce dimensionless stage performance parameters

expressed for a “repeating” stage, i.e., a stage designed to be followed by another
identical stage. This means that the velocity diagrams for the rotor inlet (station
1) and the stator exit (station 3) must be identical. Then, the stage work coeffi-
cient, ψ, can be defined as

ψ = ∆H / U 2 = (Cθ 2 − Cθ1) / U (1-19)

where all data correspond to a constant, mean radius, or “pitch line” for the
stage. The stage flow coefficient, φ, is defined by

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 14 • AXIAL-FLOW COMPRESSORS

φ = Cz1 / U (1-20)

The stage reaction, R, is defined as the fraction of the stage static enthalpy rise
that occurs in the rotor, i.e.,

R = ( h2 − h1) / ( h3 − h1) (1-21)

Substituting P for h in Eq. (1-21) yields an alternate definition of reaction, in

terms of static pressures, that is sometimes used. Requiring Cz to be constant
through the stage, Eqs. (1-17), (1-18) and (1-21) can be combined to yield

R = 1 − (Cθ 2 + Cθ1) / (2U ) (1-22)

In Chapter 10, it will be seen that parameters φ, ψ and R provide useful guidance
for stage design. Stage design involves defining blade geometry that will produce
the desired performance. These dimensionless performance parameters define
performance in a form general to any stage design problem. They are normally
used to specify the performance objectives the stage should achieve at its mean
radius or pitch line. While there are no fixed rules for selecting values for them,
preferred values can normally be established based on the design goals most
important to the designer, supported to some degree by simple logic. For exam-
ple, 50% reaction stages (R = 0.5) are quite common, prompted mainly by the
intuitive judgment that it is best to share the flow diffusion load equally between
the rotor and the stator. Once specified, these parameters can be used to define
the stage velocity diagrams from which the blades can be designed. For example,
Eqs. (1-19) through (1-22) can be combined to yield

tan β1′ = −(ψ / 2 + R) / φ (1-23)

tan β2′ = (ψ / 2 − R)/φ (1-24)
tan β1 = (1 − R − ψ / 2)/φ (1-25)
tan β2 = (1 − R + ψ / 2)/φ (1-26)

and the velocity diagrams for the stator exit and rotor inlet are identical for a
repeating stage. These parameters have defined the velocity diagrams at the
pitch line only. It is necessary to supply additional design specifications and use
fundamental fluid dynamics relations to generate the velocity diagrams at other
radial locations.

1.6 UNITS AND CONVENTIONS

This book assumes consistent units throughout, such that the reader may use any
set of consistent units preferred. For historical reasons, many turbomachinery
organizations do not use consistent units, often using different units for different
disciplines such as aerodynamics and thermodynamics. For example, it is not

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 Introduction • 15

uncommon to find energy terms, fluid velocity and equation-of-state parameters

expressed in inconsistent units, necessitating conversion factors in expressions
such as Eq. (1-18). It will be left to the reader to recognize the need for those con-
version factors. Flow angles and blade angles will be measured from the merid-
ional plane, i.e., a plane of constant polar angle, θ, in a cylindrical coordinate
system. These angles, and the associated swirl velocity components Cθ and Wθ,
are considered to be positive in the direction of rotation. While the nomenclature
is reasonably consistent throughout, the wide range of topics covered does not
permit unique symbols for every parameter. Consequently, each chapter will
include its own list of nomenclature to avoid confusion.

EXERCISES

1.1 An axial-flow compressor is to be operated with a different working

fluid, which can be modeled as a perfect gas, but has values of gas
constant, R, and ratio of specific heats, k, that are different from the
normal working fluid. Develop new equivalent speed and flow rate
parameters, in terms of ṁ, Tt0 and Pt0 that will ensure Mach number
equivalence at the compressor inlet.
1.2 Free vortex stages with β1 = β3 = 0 are often used for axial-flow com-
pressors. Derive an expression for work coefficient as a function of
reaction for this type of stage. If the stage is also to have 50% reaction,
specify the range of values for ψ and β′1 that can be used. If the result-
ing stage is to be used as a repeating stage in a multistage compressor,
what type of inlet and exit guide vanes will be needed?
1.3 All dimensions of the compressor producing the performance map
shown in Fig. 1-7 are scaled by a factor of 1.2 and both compressors are
operated with the same inlet conditions and working fluid. If operating
points for the original compressor are denoted as NA and QA, develop
expressions for equivalent operating conditions NB and QB for the
scaled compressor. If the original compressor operates at a speed of
3,600 rpm, what speed must be used for the scaled compressor? How
much additional flow capacity will the scaled compressor have?
1.4 What scale factor should be applied to the compressor producing the
performance map shown in Fig. 1-7 to increase the compressor’s flow
capacity by 20% for the same inlet conditions and working fluid?
What adjustment in speed will be needed?

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 Chapter 2

THERMODYNAMICS

This chapter highlights some of the fundamental concepts from thermodynamics

that are essential to the aerodynamic design and analysis of axial-flow compres-
sors. These concepts provide the basis for defining and evaluating the energy
transfer processes and for modeling the state properties of the working fluid.
Unlike the centrifugal compressor, axial-flow compressors are most often applied
to rather ideal working fluids, the most common being air. But the application of
axial-flow compressors to non-ideal working fluids is becoming more common.
Consequently, a basic description of non-ideal gas modeling is included. This
description is sufficient for application of the techniques, but is less detailed than
this author’s previous description for centrifugal compressors (Aungier, 2000).
Readers interested in more detail are referred to that earlier reference or to
Aungier (1995, 1998).

NOMENCLATURE

A = Helmholtz energy, dA = -PdV

a = sound speed and gas constant defined in Eq. (2-32)
b = gas constant defined in Eq. (2-33)
C = velocity
c = gas constant defined in Eq. (2-34)
cp = specific heat at constant pressure and pressure recovery coefficient
cv = specific heat at constant volume
H = total enthalpy
h = enthalpy
k = ratio of specific heats = cp / cv
M = molecular weight
ṁ = mass flow rate
n = exponent in Eq. (2-34), defined in Eq. (2-35)
P = pressure
q = specific heat transfer
q̇ = heat transfer rate
R = gas constant
RU = universal gas constant
s = specific entropy

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 18 • AXIAL-FLOW COMPRESSORS

T = temperature
u = specific internal energy
V = specific volume
w = specific work input
ẇ = power input rate
z = compressibility factor
η = efficiency
µ = gas viscosity
ρ = gas density
ω = acentric factor
ξ = parameter defined in Eq. (2-69)
–
ω = loss coefficient

Subscripts

ad = adiabatic-reversible (isentropic) process

c = critical point parameter
d = discharge condition
i = inlet condition
p = polytropic process
r = reduced parameter (normalized by its critical point value)
ref = reference condition
t = total condition
0 = condition ahead of an inlet guide vane
1 = condition ahead of a rotor blade row
2 = condition following a rotor blade row

Superscripts

0 = condition where the ideal gas model applies

2.1 FIRST AND SECOND LAWS OF THERMODYNAMICS

The first law of thermodynamics covers the basic principle of conservation of

energy. The first law can be applied to compressors with one restriction. Since a
compressor is an open system, steady flow is the only case to which the first law
is applicable. If ẇ is the power input to the compressor and q̇ is the heat transfer
between the compressor and its surroundings, the first law of thermodynamics
can be written

q˙ + w ˙ ∆[u + 1 C 2 + P / ρ]
˙ =m (2-1)
2

where ṁ is the mass flow rate, u is the specific internal energy, C is velocity, P is
pressure, ρ is density and any change in potential energy due to gravitational

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 Thermodynamics • 19

force is neglected. The term “specific” designates a parameter measured per unit
mass of fluid. The terms in brackets specify the internal energy, kinetic energy
and the flow work, P/ρ. The term flow work refers to the work necessary to move
the fluid across the boundaries of the system. Only when steady flow crosses the
system’s boundaries does P/ρ provide a direct measure of the flow work, which
restricts Eq. (2-1) to the steady flow case. The normal practice in fluid dynamics
applications is to combine the internal energy and flow work terms into a single
parameter called the enthalpy, h.

h = u+ P / ρ (2-2)

The above thermodynamic conditions are simple state variables or static condi-
tions. It is useful to introduce a special class of thermodynamic conditions
referred to as total or stagnation conditions. A total thermodynamic condition is
defined as the value of a parameter that will exist if the fluid is brought to rest
with no transfer of heat or external work, i.e., all kinetic energy is completely
recovered. Total conditions will generally be designated with a subscript, t. Total
enthalpy occurs so often that an exception will be made and H will be used
instead of ht. Total enthalpy is given by

H = h + 12 C 2 (2-3)

Heat transfer can normally be neglected for flow through a compressor. There are
obvious exceptions, such as when a heat exchanger is included in the system. If
heat transfer is neglected, the flow is called adiabatic and Eqs. (2-1) through (2-3)
combine to yield

˙ =m
w ˙ ( Hd − Hi ) (2-4)

where the subscripts i and d refer to the compressor’s inlet and discharge condi-
tions, respectively. Equation (2-4) is the basic energy equation for steady, adia-
batic flow through a compressor.
The second law of thermodynamics introduces the concept of a reversible
process. A process is referred to as reversible if the system and its surroundings
can be returned to their original states after the process has occurred. If that is
not the case, the process is called irreversible. Processes influenced by heat trans-
fer or friction effects are common examples of irreversible processes. The specific
entropy, s, is defined as

dqrev
ds = (2-5)
T

where T is the temperature, q is the specific heat transfer and the subscript, rev,
designates a reversible process. The second law of thermodynamics can be
expressed as

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 20 • AXIAL-FLOW COMPRESSORS

∆s ≥ 0 (2-6)

Hence, an adiabatic, reversible process is also a constant entropy or isentropic

process. Indeed, it is fairly common practice in turbomachinery to use the terms
adiabatic and isentropic interchangeably, although the latter is clearly the more
precise term. Entropy provides a fundamental measure of the irreversibility of a
process, i.e., the inefficiency or losses associated with the process. If the first law
of thermodynamics is applied to a fluid element in a closed system,

( dq)rev = Tds = du + dw = du + PdV (2-7)

where w is the specific work and V = 1/ρ is the specific volume. Equations (2-2)
and (2-7) provide a fundamental thermodynamic equation for entropy that is
valid for any process, i.e.,

Tds = dh − VdP (2-8)

2.2 EFFICIENCY
The aerodynamic quality of a compressor or a component of a compressor is
commonly measured in terms of efficiency, which is a measure of actual per-
formance relative to an ideal performance that would be achieved by some
reversible process. Figure 2-1 illustrates a typical enthalpy-entropy diagram for a
compressor or a portion of a compressor that includes at least one rotor row,
such that work has been done on the fluid. The inlet conditions are designated as
Pti, Tti, si and Hi. The compressor does work on the fluid to produce discharge
conditions Ptd, Ttd, sd and Hd. Note that static and total conditions are, by defini-
tion, related by a reversible process. Hence, there is no difference between total
and static entropy and the subscript, t, can be omitted for s. One measure of effi-
ciency is to compare the actual process to an ideal adiabatic (isentropic) process.
As seen in Fig. 2-1, an isentropic process could produce the change in total pres-
sure with an enthalpy rise ∆Had, commonly referred to as the adiabatic head. The
actual process required an enthalpy rise of ∆H. Since lines of constant pressure
always diverge on an h-s diagram, ∆H is always larger than ∆Had for a non-isen-
tropic process. Hence, the adiabatic or isentropic efficiency, ηad, is defined as

∆Had
ηad = (2-9)
∆H

where ∆Had is given by

Ptd dP
∆Had = ∫ ; (s = const) (2-10)
Pti ρ

This development of efficiency has considered a compressor operating between

inlet and discharge total conditions. Hence it is often called the total-to-total

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 Thermodynamics • 21

adiabatic efficiency. It is often the case that the fluid kinetic energy at the com-
pressor discharge serves no useful purpose to a specific application. In that case,
Ptd may be replaced by the static pressure, Pd, in Eq. (2-10) to yield the total-to-
static adiabatic efficiency. The additional substitution of Pi for Pti yields the
static-to-static adiabatic efficiency. Some care is required to understand which
basic definition is being used when interpreting efficiency data. It is not uncom-
mon for the term adiabatic efficiency to be used for any of the above three types
without qualification.
Adiabatic efficiency is the most common definition used for axial-flow com-
pressors. But it has a definite weakness as a means of evaluating the aerody-
namic quality of a design. As illustrated in Fig. 2-1, constant pressure lines
diverge on an h-s diagram. This means that two compressors having the same
basic aerodynamic design quality, but operating at different pressure ratios,
will have different adiabatic efficiencies. Hence, adiabatic efficiency is not par-
ticularly useful to an aerodynamic designer seeking to evaluate the true aero-
dynamic quality of a compressor or a stage design. Another consequence of
this thermodynamic effect is that the adiabatic head of a multistage compres-
sor is not equal to the sum of the stage adiabatic heads. Polytropic efficiency is
a more useful definition, which eliminates this undesirable thermodynamic
effect. Polytropic efficiency is sometimes referred to as the “small-stage” or
“true aerodynamic” efficiency. Instead of using a path of constant entropy as
the reversible path, polytropic efficiency uses a path of constant efficiency
defined by

FIGURE 2-1 An Enthalpy-Entropy Diagram

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 22 • AXIAL-FLOW COMPRESSORS

1 dP
ηp = (2-11)
ρ dh

where ηp is defined such that the path passes through the two end points of the
process, e.g., (Pti, Hi) and (Ptd, Hd). This is straightforward in principle, but less so
in practice. For many years, common practice was to approximate a polytrope by
a path defined by P/ρ e = constant. This approximation is appropriate for ideal
gases, but can introduce large errors for non-ideal gases. Models that use poly-
tropic efficiency with non-ideal gases have been reported by Shultz (1962) and
Mallen and Saville (1977). Huntington (1985) studied the problem in detail and
showed that the Mallen-Saville model yields excellent accuracy. Huntington pro-
posed a slightly better model by including an intermediate point on the path, but,
for convenient use, this had the disadvantage of requiring a numerical method.
Hence, this writer employs the Mallen-Saville model in all cases. This model uses
an empirical path defined by

ds
T = constant (2-12)
dT

Equations (2-8) and (2-12) can be combined to yield the polytropic head, ∆Hp, as

∆H p = ∆H − ( sd − si )(Ttd − Tti ) / ln(Ttd / Tti ) (2-13)

to provide the total-to-total polytropic efficiency, ηp, by

∆H p
ηp = (2-14)
∆H

Extension of Eqs. (2-13) and (2-14) to total-to-static and static-to-static efficiency

is analogous to adiabatic efficiency as previously discussed. Basically, it is now
possible to employ polytropic efficiency with no more difficulty than adiabatic
efficiency. It is only necessary to have an appropriate equation of state to use
either model.

2.3 FLUID EQUATION-OF-STATE FUNDAMENTALS

Thermodynamics contributes one of the fundamental governing equations for

compressor aerodynamic design and analysis, commonly called the equation of
state. To be more precise, there are actually two equations of state required. The
first is the thermal equation of state, which supplies a relationship among the
fundamental state variables, typically in the functional form P = P(ρ, T). The sec-
ond is the caloric equation of state, which relates the energy content of the fluid
to state variables, typically in the functional form h = h(T, P) or u = u(T, P). These
equations of state may be derived from kinetic theory or statistical mechanics, or
they may be developed empirically from experimental data.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 Thermodynamics • 23

Axial-flow compressors are usually applied to rather simple fluids, the most
common being air. Consequently, books on axial-flow compressors usually pro-
vide a very limited discussion of the equation of state applicable only to very ideal
working fluids. But the application of axial-flow compressors to very non-ideal
fluids is becoming more common, so this chapter provides a broader discussion.
A fluid is considered to be an ideal or thermally perfect gas if P, T and ρ can be
related by the simple linear thermal equation of state

P = ρRT (2-15)

where R is a constant for the fluid. The gas constant, R, is related to the universal
gas constant, RU, and the fluid’s molecular weight, M

R = RU / M (2-16)

where RU = 8314 Pa-m3/(kmole-°K) in metric units. All working fluids exhibit

non-ideal behavior under appropriate conditions. Figure 2-2 is a schematic of a
pressure-enthalpy diagram for any working fluid. It is seen that the fluid may be
liquid, vapor or both at various state points. Clearly, it will not be possible for a
thermally perfect gas equation of state to model all possible state points. This is
true even when the equation is restricted to the vapor phase. Figure 2-2 illus-
trates the location of the fluid’s critical point, which is defined as the highest

FIGURE 2-2 A Pressure-Enthalpy Diagram Schematic

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 24 • AXIAL-FLOW COMPRESSORS

temperature at which liquid and vapor can coexist. Experimental measurements

of critical-point properties (Tc, Pc, ρc) are commonly made and almost never
conform to Eq. (2-15). Normally, the thermally perfect gas equation of state can
provide a reasonable approximation if T is much greater than Tc and P is much
less than Pc. At pressures well above Pc, the fluid is said to be in the supercritical
regime. Again, it is known from experiment that supercritical fluids almost
always show non-ideal gas behavior.
Yet the role of the thermally perfect gas equation of state is far more significant
than just providing a useful approximation. Indeed, at sufficiently low values of
density, all fluids follow the perfect gas law. This provides a dramatic simplifica-
tion to the process of modeling non-ideal gases. Under conditions where the gas is
thermally perfect, it can be shown that the energy content of a fluid is independ-
ent of pressure, i.e., h0 = h0(T) and u0 = u0(T), where the superscript, 0, designates
conditions where the fluid is thermally perfect. This means it is always possible to
define the caloric equation of state as a function of only one state variable.

2.4 THE CALORIC EQUATION OF STATE

From the previous discussion, it is seen that the caloric equation of state can be
specified in a general form for use in either an ideal or non-ideal gas model.
Under low-density conditions where the gas is thermally perfect, the specific
heats at constant pressure and at constant volume are defined as

 ∂h0 
c0p (T ) =   (2-17)
 ∂T  P

 ∂u0 
c0v (T ) =   (2-18)
 ∂T  V

For a thermally perfect gas, it can be shown that

c0p (T ) − c0v (T ) = R (2-19)

The caloric equation of state can be specified by supplying either c0p(T) or c0v(T)
and using Eqs. (2-17) through (2-19), i.e.,

T
h0 (T ) = h0 (Tref ) = ∫ c0p (T )dT (2-20)
Tref

T
u0 (T ) = u0 (Tref ) = ∫ c0v (T )dT (2-21)
Tref

where h0 and u0 can be assigned any desired values at a reference state point (Tref,
Pref). Specific heat correlations as a function of temperature are readily available
for most fluids of interest (e.g., Ried, et al., 1977, 1987; and Yaws, 1999). For the

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 Thermodynamics • 25

special case where the specific heats are constant, the gas is referred to as a calor-
ically perfect gas. In that case, Eqs. (2-20) and (2-21) can be directly integrated to
yield analytical equations, i.e.,

h0 (T ) = h0 (Tref ) + c0p (T − Tref ) (2-22)

u0 (T ) = u0 (Tref ) + c0v (T − Tref ) (2-23)

Most books on axial-flow compressors limit attention to thermally and calori-

cally perfect gases. This simplifies the writer’s task by reducing some concepts to
simple, analytical forms. This practice is avoided in this book because it results in
some unfortunate consequences. Even experienced turbomachinery engineers
have occasionally applied these simplified concepts beyond their limits of valid-
ity, resulting in some unfortunate conclusions. It has often encouraged develop-
ers of computerized aerodynamic analyses to incorporate special thermodynamic
relations that are limited to this special class of fluids. A numerical analysis
rarely benefits from these simplified methods, and it can be very difficult to gen-
eralize such an analysis later, should that become necessary. It is quite possible to
have hundreds of such relations scattered through an analysis, presenting a for-
midable problem when attempting to generalize it.

2.5 ENTROPY AND THE SPEED OF SOUND

An entropy equation is essential to all aspects of compressor aerodynamic design
and analysis. The efficiency definitions introduced previously in this chapter all
require an entropy equation. Even the simplest aerodynamic analyses require an
entropy equation to convert between total and static thermodynamic conditions.
This conversion is an isentropic process, accomplished by imposing a change in
enthalpy (the kinetic energy) at constant entropy. An entropy equation is also
required to impose a non-isentropic loss mechanism such as a total pressure loss.
It is useful to develop the entropy equation consistent with the caloric equation
of state defined for a thermally perfect gas. This will be needed for the non-ideal
gas model described later in this chapter. For any pressure where the gas is ther-
mally perfect, the specific entropy is given by Eq. (2-8), i.e.,
T
c0p (T )
s0 (T , P ) = s0 (Tref , Pref ) + ∫ T
dT − R ln( P / Pref ) (2-24)
Tref

where s0 can be assigned any desired value at any reference state point (Tref, Pref).
For a calorically perfect gas, this simplifies to

s0 (T , P ) = s0 (Tref , Pref ) + c0p ln(T / Tref ) − R ln( P / Pref ) (2-25)

The speed of sound is also an essential parameter governing the performance

of a compressor. From fundamental thermodynamics, the speed of sound, a, for
any gas is given by

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 26 • AXIAL-FLOW COMPRESSORS

 ∂P   ∂P 
a 2 =   = k  (2-26)
 ∂ρ  s  ∂ρ  T

where k is the ratio of specific heats.

k = cp / cv
(2-27)

For a thermally perfect gas, this yields

a0 = kRT (2-28)

2.6 THE THERMAL EQUATION OF STATE FOR REAL GASES

The general thermal equation of state for a real gas is

P / ( ρRT ) = z(T , P ) (2-29)

where z is the compressibility factor. This equation applies to any fluid, with z = 1
for the special case of a thermally perfect gas. The compressibility factor can be
obtained from generalized tabular data (e.g., Nelson and Obert, 1954; and Pitzer
et al., 1955). But that is rarely done today since many excellent real gas equations
of state are available. The real gas equation of state directly provides the many
other thermodynamic parameters required and yields much better computational
speed when used in numerical methods. The simple two-parameter equations-of-
state are a good choice for general aerodynamic design and analysis. They offer
good accuracy, excellent computational speed and easy access to the required gas
property data for almost any working fluid or fluid mixture. The Redlich-Kwong
equation (Redlich and Kwong, 1949) and various modifications to it (Aungier,
1994, 1995; Barnes, 1973; Soave, 1972; and Wilson, 1966) are recognized as being
among the most accurate of the two-parameter equations. Aungier (1994, 1995)
evaluated these five equations of state in considerable detail for twelve different
compounds over a wide range of temperatures and pressures. Specific emphasis
was placed on covering a wide range of accentric factors, ω, since that parameter
is used by the various modified Redlich-Kwong equations to improve the predic-
tion accuracy. Accentric factor (Pitzer et al., 1955) is defined as

ω = − log10 ( Pv / Pc ) − 1 ; T / Tc = 0.7 (2-30)

where Pv is pressure on the vapor saturation line and the subscript c designates a
critical point property (Fig. 2-2). Table 2-1 lists the compounds investigated,
together with their accentric factors. Based on this evaluation, it was concluded
that only the original Redlich-Kwong equation and Aungier’s modified Redlich-
Kwong are suitable for general turbomachinery aerodynamic design and analy-
sis. The modified equations of Barnes, Soave and Wilson all showed improved

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 Thermodynamics • 27

Table 2-1 List of Compounds

Compound ω
Ammonia 0.2550
Carbon Dioxide 0.2250
Ethylene 0.0868
Helium –0.464
Hydrogen –0.220
I-Butane 0.1848
Methane 0.0080
N-Pentane 0.2539
Nitrogen 0.0400
Propane 0.1520
Refrigerant R134a 0.3254
Steam 0.3440

accuracy over the original equation for the types of compounds and thermody-
namic property ranges for which they were developed. But they were found to
have serious deficiencies for ω < 0, and they exhibited reduced accuracy for
supercritical fluids (i.e., P > Pc, Fig. 2-2). Consequently, only the original Redlich-
Kwong equation and Aungier’s modified form will be described here. The original
Redlich-Kwong equation is

RT a
P= − (2-31)
V − b V (V + b) Tr

where Tr = T / Tc is the reduced temperature, and

a = 0.42747 R2Tc2 / Pc (2-32)

b = 0.08664 RTc / Pc (2-33)

Equations (2-32) and (2-33) are derived from the thermodynamic stability condi-
tion, which requires that the first and second partial derivatives of P with respect
to V must both equal zero at the critical point. For its application, this equation
requires only the critical temperature and critical pressure for the fluid (and the
caloric equation of state data). It offers very good accuracy over a wide range of
thermodynamic conditions. It does have one well-known deficiency near the crit-
ical point for the fluid. For any fluid, this equation yields z = 1/3 at the critical
point, which is not typical of most fluids. Basically, this equation should never be
used for points close to the critical point.
Aungier’s modified Redlich-Kwong equation of state is

RT a
P= − (2-34)
V − b + c V (V + b)Trn

where a and b are given by Eqs. (2-32) and (2-33), respectively, n is given by

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

 28 • AXIAL-FLOW COMPRESSORS

n = 0.4986 + 1.1735ω + 0.4754ω 2 (2-35)

and the constant, c, is calculated directly by applying Eq. (2-34) at the critical
point with all critical point properties specified. It is included to remove the defi-
ciency in the original equation for points near the critical point. Aungier’s equa-
tion requires two more specified parameters than does the original equation, i.e.,
ω and either Vc or zc. Note that if c = 0 and n = 0.5, Aungier’s equation reduces to
the original equation. If a = b = c = 0, the thermally perfect gas equation of state
is obtained. This makes it very easy to develop a computerized equation-of-state
package for use in any aerodynamic analysis, which can offer all three equations
of state.
Figure 2-3 illustrates the benefits of Aungier’s modified Redlich-Kwong equa-
tion over the original form. This figure shows a comparison of the prediction
accuracy of the two equations for about 25% of the data points considered in the
evaluation in Aungier (1994, 1995). Tabular (P, V, T) data from the literature were
used for that purpose. The two equations of state were used to predict P from the
tabular values of T and V, which were compared to the tabular value of P to com-
pute the error. It is seen that Aungier’s model provides a significant improvement
in prediction accuracy for a vast majority of the points considered. Aungier
(1994, 1995) notes that Aungier’s model resulted in about a 50% reduction in the
root-mean-square error for the complete set of data considered. This writer uses
Aungier’s model for all real gas problems, but maintains the original model as a

FIGURE 2-3 State-Point Prediction Accuracy

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 01/05/2016 Terms of Use: http://www.asme.org/about-a

Exploring the Variety of Random
Documents with Different Content
pronouncing the hated name. “I think I’ll go to the post office and
send this back in a registered parcel.”
She quitted the room abruptly, leaving her friend to some very
sharp-edged thoughts.
 XLIX
THE days that followed were dark and difficult for Mame. Nor did
Lady Violet find them particularly easy. At heart she was kindly and
honest and she could not help fixing upon herself a good deal of the
blame for what had occurred. She it was who had introduced this
little marauder; she had been wilfully and stupidly blind to the
consequences; and her only excuse was that not for a moment could
she believe that Bill would be so weak. Yet she had deliberately
thrown them together. She had bestowed upon Mame a spurious
eligibility. This bitterly humiliating business was an object lesson in
the sheer folly of playing the fool.
It was decidedly painful all round. First of all, poor Mame really
suffered. The part she had undertaken to play was superhumanly
big. Very few girls could have gone back in that way on their whole
philosophy of life; and to Violet’s good heart it was hateful to have to
ask her to do it. She must have loved Bill, perhaps as no other
woman was likely to do, to nerve herself to a sacrifice so high.
Much diplomacy was called for in the days that followed. Bill could
be a stubborn fellow. His family had always humoured his whims. It
looked at first as if there would be no handling him. Irresponsible,
not to say flabby, as his nature was, had he had the wit to realise
clearly that poor Mame had got a blow over the heart, he would not
have taken the thing lying down.
Violet, hating intensely the rôle her own folly had doomed her to
play, yet proved herself, when fairly put to it, a consummate
tactician. Bill must not guess, must not come near guessing, how
much this good and brave child really cared for him. He must be
rather a “dud,” his sister thought, not to see it for himself. Instinct
ought to have told him how the little brick was steeling her heart.
Even while Violet threw dust in his eyes with a subtlety and a
success for which she loathed herself, she yet clung to the
paradoxical view that had Bill been worthy of Mame he would have
been less obtuse. “She can’t bear your losing the Towers and giving
up the Army.” That phrase was Violet’s strongest weapon. It
admitted two interpretations and his sister was not in the least proud
of Bill when he allowed it to suggest the wrong one. “Of course she
considers the gilt will be off the gingerbread.” That was
Machiavellian. The end seemed to justify the means; at any rate
Violet had so persuaded herself; but it really was rather low down.
And she could not, womanlike, help resenting Bill’s denseness and
lack of character which made the unpleasant task of throwing dust
in his eyes so much less difficult than it ought to have been.
The weeks went by and Mame set herself stoically to forget. There
was pride in her as well as grit. She was determined to stand up to
life and make something of the mighty difficult business of living it.
To stanch her wounds she threw herself into her work with new
ardour. Back of everything was rare common sense. She must have
the strength to bear the self-inflicted blow without flinching.
One morning, however, just before Christmas, an incident occurred
that reopened the closing wounds. Mame and Celimene were
discussing the make-up of the weekly cable to New York, when
Celimene said with an odd change of tone, “There’s one bit of news
that may not be without interest on the other side. It’s not yet
announced, so we shall be the first in the field.”
“What is the news?” asked Mame keenly. Her flair for a choice tit-bit
was not less than of yore.
“A marriage has been arranged, the date of which will be shortly
announced, between the Marquis of Kidderminster and Gwendolen,
the only child of Giles Childwick, Esquire, and Mrs. Childwick,
formerly of Treville, New Jersey.”
“Oh!” Mame gave a little gasp. Celimene saw her turn very white.
“We are none of us worthy of you, my dear. In the end you’ll find
yourself well rid of people like us.” Violet’s tone had a note of pain
that for her was something new. Life had not been exactly a bed of
roses lately. She had discovered, a little late in the day perhaps, that
she had a conscience. A share of the hurt she was inflicting had to
be borne by herself.
When Mame was able to speak she said: “You’d have been good
enough for me. I like you all. You are some of the nicest folks I’ve
met.” The whimsical frankness of this good child brought back the
laugh, if not too readily, to the worldly wise Violet. What a piece of
luck, they both had a sense of humour!
“We owe you more than perhaps you realise.” Violet did not find the
words easy, but they had to be said. “You showed how terribly
dangerous was delay as far as that brother of mine is concerned. He
might have been snapped up by some pretty little prowler, with a
nice taste in dicky birds, who short of death and destruction could
never have been persuaded to unclasp her claws.”
“That’s so,” assented Mame. “And you’re wise to have hustled on
that marriage.” And then whimsically, to ease the aching of her
heart: “Some other little cat might have got that bird.”
She was so real and so true, this brave child, that Violet felt she
would like to have taken her in her arms and hugged her. “My
mother and I both realise that you have done a big thing. She has
sent you all sorts of kind messages. We are going to see that
Gwendolen plays the game so far as you are concerned. And no
matter how long you stay over here you will always have friends.”
“Gwendolen will make him a good wife.”
“Yes, she will. She is a very sensible girl, with a strong will. I am
sure she will keep him on the rails. And she really cares for him.”
“Perhaps he’ll get to care for her after a while.”
“I think he may. Gwendolen is a very good sort. But I’m afraid Bill’s
feelings don’t run deep.” And the note of pain crept again into the
voice of his sister.
 L
THE marriage ceremony was fixed for the second week of the new
year. All London was invited. The Childwicks liked to do things well,
and however well they did them made no impact on their wealth.
They were really rich people and of late years their stock had been
going continually up.
Still it was generally felt that Gwendolen had done very well. She
was a cut and polished jewel, but a setting was needed for her.
What choicer setting could mortal girl desire than an old and
distinguished marquisate and that beautiful and historical place
Warlington Towers? These things would supply the essential
background her millions lacked. Then, too, Bill was a popular young
man. His feelings, as his sister said, might not run deep, but in his
way he was immensely attractive. Everybody liked Bill. Even his
fecklessness was a point in his favour. And he would be all the better
for having a shrewd and clever American wife to keep him up to the
mark.
Mame received an invitation from the Childwicks; her friends saw to
that. It was duly accepted; yet when the day came she did not feel
equal to facing the music. Lady Violet did her best to persuade her
to come to the church and to the reception in Berkeley Square, until
she realised that it would not be kind to persist. Good, brave little
puss! Those quaint furry paws had been rather nastily trapped. She
was still suffering. For all her wonderful grit, when the day came she
could not help groaning a bit over the throbbing of her wounds.
It went to Lady Violet’s heart to leave her behind. But there was no
help for it. The child could not face the music. And it was hardly
reasonable to expect that she should.
Even as Lady Violet stood in her regalia, which emphasised her fine
points, and took a tender and affectionate leave of poor Mame, she
could not help saying in her frank way: “I’m not going to enjoy this
one bit. We’re shown up too badly. I wish now I hadn’t interfered.”
It was no more than the truth. She felt on a low plane. Compared
with her friend and business partner she had a sense of inferiority
which was new and decidedly unwelcome.
“You were quite right, honey.” Mame was brave and magnanimous to
the end. “Had I been you I’d just have done that. It’s nature. And it’s
no use trying to go against nature.”
But Mame’s eyes were so tragic, that her friend, without venturing to
say another word, made a sort of bolt for the door and for the lift
beyond it. No, Lady Violet did not feel that she was going to enjoy
her day.
The day, for Mame, was very far from being one of enjoyment either.
A great deal of it was spent walking about London. Now the pinch
had really come it was the go-getter who mounted to the saddle and
took command. The practical side of a dual nature was not slow to
inform her that she had gone back on herself. She had betrayed, at
the beck of a mere whim, all that she had stood for.
Life had always been against her until a few months ago. She had
had to fight so hard for bare existence that she might at least have
had the sense to ensure the future when opportunity arose. But no,
she had denied her luck. There was a Tide. Well, the Tide had come
and she had deliberately ignored it. A great position, social security
had been offered her. She had even accepted her chance and then,
just for a whim, had passed it on to her enemy and rival.
The go-getter spared her nothing. It had a royal time. She had been
a fool. The lot of the fool was suffering. “You’ve broken a paw,
honey. Always you’ll be a little mucker now. You’ll never be able to
fix on anything again. No more will you be able to mark down your
bird and fly at it. You’ll be unsettled for the rest of your days. Not
always will you be brisk and quick; your looks, such as they are, and
they’ve never been much to bank on, honey, are going already.
These folks in London, England, who mocked at you and pulled your
leg, you could have handed them a haymaker. But no, you must get
fanciful and highfalutin.”
Yes, the go-getter with the sardonic voice and a flow of rather
second-rate conversation had a royal time. He spared her nothing.
She was unworthy of herself, of her breed, of her clan. Idealism. The
hearty fellow landed his simp of a partner a blow on the point. If
some of these honest-to-God Americans didn’t watch out, Idealism
was going to be their ruin. You didn’t catch the dyed-in-the-wool
Britisher playing around with flams of that kind. He was a pretty
successful merchant, the Britisher, but he was content to leave
idealism to other people. As a business man he was a model for the
world. And why? Because with all his lip service he knew how to
keep the soft stuff apart from the hard.
Poor Mame! She trudged half the day about the streets of a hostile
city. With that pressure upon her spirit it was impossible to stay
quiet indoors. Her very soul seemed to ache. Every syllable was true
that was whispered in her overwrought ear. She had gone back on
all she had ever stood for. They had put one over on her, these hard-
roed Britishers. Yes, she must be cuckoo. You came over to pull big
stuff, whispered the relentless voice. And by cripes, Mame Durrance,
you’ve pulled it!
Tired out at last with tramping the West End parks and squares, she
took a little food in a restaurant in a by-street of Soho. There she
was likely to meet no one she knew. She was not in a mood to face
her kind. As she ate her soup a nostalgia came upon her. After all,
she was in a strange land, an alien. Their ways were not her ways;
they had a different viewpoint; their method of doing things was not
the same. She began to long for the sight and the sound of the
homely, hearty, warm-blooded folks she had known; the folks who
spoke the same language as herself, in the curious drawl that lately
she had been taking such pains to get rid of.
Her thoughts went back to the land where she belonged. In the
bitterest hour she had known since she had started out from her
home town to see life, she had a craving for the friendly easy-
goingness of her own kind. She had crowded a lot of experience into
her European pilgrimage; in certain ways her luck had been truly
remarkable. Mame Durrance had made good at her job; but this
evening with a very large size “black monkey” upon her, she had a
sudden yearning for the larger and freer air of her native continent.
Miserably unhappy she returned to the flat about nine o’clock. Lady
Violet had been home from the revels, had inquired for her
anxiously, but had changed her dress and gone out to dinner.
Evidently she was making a day of it. Mame was not sorry. She had
no wish to be caught in this mood. Yet she had no desire for bed.
She would not be able to sleep if she turned in. The ache in her
heart was terrible. If she could not learn to subdue it, for the first
time in her life she would be driven to take a drug.
Suddenly her eye lit on a package on the writing table. It bore the
label of a New York publisher and was addressed to herself.
Perfunctorily she tore off the wrapper. A novel in a gay jacket was
revealed. It was called Prairie City; and the name of the author was
Elmer Pell Dobree.
Mame’s heart leaped. Coincidence has an arm notoriously long, but
nothing could have been more timely than the arrival of this book.
Her thoughts rushed back to the source of its being. And as if to
speed them on their way, a letter had been enclosed in the parcel.
Characteristically it ran:
Dear Mame,
You are such a girl now, among your slick friends in
London, and you are pulling such big stuff with your
weekly columnising, that I daresay you have forgotten
your obscure Cowbarn epoch, and the junk whose early
chapters you had the honour (sic) of typing for your first
and most distinguished (sic) editor, with which the boob
 filled in his spare hours. You used to tell him how good
these opening chapters were and the boob used to believe
you. The consequence is ... well, this is the consequence.
By the way, it was clever you that invented the title, after
we had shaken a leg—and you were always some dancer
—at that dive at the top of Second Street, one wet
afternoon.
Well, Mame, that title was lucky. Prairie City has made
good. It has, if I may say it modestly, made quite
surprisingly good. ’Tis hardly six weeks since it was first
issued here by Allardyce, Inc., but already it’s gone big.
Early next month their London house will publish it, and if
it can repeat in England what it has done in New York, the
undersigned Elmer Pell Dobree is a permanent whale. So
go to it, Mame, you little go-getter. Corral the big drum
and get around the comic town of your adoption—worse
luck!—and see if for the sake of old times—good times
they were, too, I’ll tell the world—you can’t put one over,
in the name of an old friend, on the doggone Britisher.
P.S. When do we see you again here? As I’ve told you
more than once, this continent is a poorer place for your
absence. If you’ve left us for keeps it’s a great, great
shame; and there’s one young man who won’t forgive
you.
When, at a decidedly late hour, or, rather, an early one, Lady Violet
returned from the festivities, she invaded Mame’s bedroom. She
wanted to see if the child had come back and that she was all right.
In spite of the excitements of the day her thoughts had been pretty
constantly with her little friend. She was a good deal concerned
about Mame. And now it was something of a relief to find her
propped up in bed, simply devouring a book with a red cover.
It was so good to see a smile on a countenance which a few hours
ago looked as if it would never smile again, that the intruder
exclaimed, “Why, why, whatever have you there?”
“Elmer’s written a book.” A glow of excitement was in Mame’s tone
and in her eyes. “It’s all about Cowbarn and the folks we used to
know.”
“True to life, I hope.” Mame’s friend, gazing at her furtively, sought
to read what lay behind that rather hectic air.
“Better than life ever was or ever will be. It just gets you feeling
good. And it keeps you feeling good all the time.”
“He must be a clever man, your friend Elmer P.”
“Clever is not the name for Elmer.” Mame spoke excitedly. “A genius
—that baby! All the Cowbarn folks are in it. I’m in it. And I’ll say he’s
let me down light.”
“I hope he’s made you the heroine of the piece. Anyhow, he ought
to have. You are fit to be the heroine of the best piece ever written.”
“Say now, honey,” expostulated Mame in the way her friend had
learned to love, “that’s where you get off. Heroine-ism and that
fancy jake is no use to me. I’ll never be able to get away with it. But
Elmer has let me down light.”
“You must let me read it.”
Mame laughed. She actually laughed. “Why, you shall so. It’s the
goods. That boy has an eye to him. He can see into things. And he
knows a lot about human nature, does that boy.”
Already Lady Violet was feeling a lively sense of gratitude towards
the famous and legendary Elmer P. The poor child was transformed.
Her own people, for whose homely and abounding kindliness she
had in her misery been longing, were alive in those magic pages.
Yes, they were alive and they were dancing, Mame declared. And
half America was dancing with them.
The mirth of the simple creatures she loved so well had lifted a
weight from her heart. The relief might only be temporary, but Lady
Violet was very willing to do homage to the wizardry of Elmer P.
“There’s his letter.” Mame tossed it excitedly across the bright green
expanse of counterpane.
Lady Violet read with a smile. And then suddenly there came a
mischievous little clutch at her heart. Yes, why not? It was an idea.
The brilliantly clever woman of the world again glanced furtively at
Mame. This intoxicating moment, which was doing so much to heal
the child and to keep her sane, must, if possible, be held. But how?
Like all things in time it was fleeting, transitory. With the coming of
daylight it would surely pass. A pitiless January dawn would throw
her back upon hard and cruel reality. Yet this moment of happiness
might be extended; perhaps there was a chance of making it
permanent.
She fixed her wise eyes upon those of the feverish Mame. “Yes, my
dear, we’ll go round with the big drum. We’ll corral the Press all
right. If Prairie City doesn’t knock London endways it shall not be
our fault.”
“You can bet your life it won’t be!”
“Well, write to-morrow and tell him so.”
“I will.”
“And tell him, my dear, in a postscript, strongly underlined, that if he
will do the one thing you ask you’ll guarantee the success of Prairie
City this side the Atlantic.”
Mame was all ears. “What’s that, honey?”
“He must come over himself as soon as ever he can. We’ll promise
the very best that can be done for him in the way of a good time.
After all there is no advertisement for a book quite equal to the
person who wrote it.”
Mame gave a chuckle of pleasure. Sure, it was an idea. Why had not
she thought of it herself?
“Yes, hon, if Elmer comes that’ll fix it.”
But would he come? That stern question at once invaded Mame’s
mind.
“We’ll make him,” said Lady Violet.
“He’s not easy to make do anything he don’t want to. And he’s pretty
busy these days and rather important, too.”
“We’ll get at him somehow.” Lady Violet had an arch look. “For his
own sweet sake,” she added artfully.
 LI
THE next day they laid their clever heads together and wrote a letter
to Elmer P. It was appreciative but peremptory. He must mail his
latest photograph at once. And he had better send that old green
album, too, with all those snapshots in it of Cowbarn, Iowa,
including some interesting views of the exterior and the interior of
the office building of the good old Independent.
Mame gave a solemn undertaking to get out the big drum and to go
around with it. She and her friends, in fact, would organise such a
publicity campaign as would astonish even a booster like himself. But
he, too, must be ready to do his bit. By the time London had the
book in its hands, it would be dying to see the author. He must come
over, if only for a week, he must, he must! That final imperative
word was three times underscored.
The thirteen days that followed the letter’s posting were rather
anxious ones. Lady Violet feared that when Mame’s excitement had
time to cool a bad reaction would set in. By judicious fanning,
however, she was able to keep the flame alive; sufficiently, at any
rate, to provide a highly necessary distraction.
Elmer’s reply came promptly. The urgent skill of the summons had at
least stirred him to that. Moreover, in the pompous language of the
British House of Go-getters, which Mame had heard for herself from
the Ladies’ Gallery, the answer was in the affirmative. The great man
actually promised to come to London. His coming, moreover, should
synchronise with the British publication of his book, which since he
wrote last had sold three more editions in these United States.
“He doesn’t let the grass grow!” Mame crowed her triumph. It was
disinterested triumph. She had nothing to gain, as far as she knew,
by the coming of Elmer, except in the way of humble-minded
ministry to his rapidly growing fame. “One of the up-and-coming
ones is Elmer, for all he’s so quiet. Tell me, Vi, what do you think of
his photograph?”
“Is it like him?”
“He’s changed some, I’ll say, since I left him in the editor’s office at
Cowbarn, twenty months ago to-morrow.”
“A clever face.” A good face, too, Lady Violet might have added. She
certainly saw something oddly attractive in the fair, open
countenance of Elmer P.
“I guess there’s something better than clever in it.” Mame gazed
critically at the photograph.
“Well, dear child, I guess there is, too.”
Her friend looked at her tenderly and then laughed to herself softly.
 LII
THEY lost no time in getting to work. First they went to Henrietta
Street and called upon Allardyce, Inc. Taking wise men into their
counsels, they started to plan a campaign whose aim was the
making of Elmer Pell Dobree a household name in Britain. They
infected Allardyce, Inc., with their own enthusiasm; not perhaps
such a difficult process. Every mail was bringing news of breaking
records across the water. Prairie City was the best in its kind since
Mark Twain. Indeed some of the highbrows thought really and truly
it was better.
The publishers were taken with Mame. All the natural zip of the
booster-born sprang to the surface as she lightly gave off her ideas
for the big drum. Very bright some of these ideas were. There was
one in particular which appealed to these shrewd men.
Unfortunately, to carry it out would cost money. But as smart Miss
Amethyst Du Rance, a writer herself, by the way, declared, quoting
from a favourite calendar that had once adorned the wall at the back
of her typewriter in that identical office which soon would be famous
over the breadth of two continents, If you want to make Omelettes
you’ve got to break Eggs.
This boom was going to cost money. No use burking that fact in
political economy. But it was going to be worth it. Lady Violet
Treherne—that very distinguished-looking girl who had accompanied
the corking little Miss Du Rance into the back parlour: Celimene, by
the way, of the Morning News—was of that opinion, too. Prairie City
was the goods. It was the big stuff. Every dime spent on boosting it
would earn a dollar.
Allardyce, Inc., of the Allardyce Building, East Forty-ninth Street,
New York City, U.S.A., and 1-a Henrietta Street, London, England,
not to mention 16 Rue de la Paix, Paris, France; 39 Stratton Street,
Johannesburg, S.A., and 105 Victoria Avenue, Melbourne, Australia;
in short, wherever the honest mother tongue is spoken, Allardyce,
Inc., decided to fall for Miss Du Rance and her little campaign. Still it
was going to cost money.
“It’ll be worth it all the time.” The air of Miss Du Rance was already
victorious. “You do your bit and we’ll do ours. That’s all we ask.”
The head of the London branch of the well-known firm personally
bowed the two ladies into their taxi. He had not been so impressed
in years. Full of vim this Miss Du Rance. Portentously full of pep. No
wonder they made good in cradle-rocking Britain, when they came
over, these one-hundred-per-cent little ladies from the U.S. And the
cunning minx kept back a few grains of the pep for her final shot. As
she offered a hand in parting, at its most fashionable angle, to
Allardyce, Inc., she said in her new and careful Mayfair manner:
“When Elmer Pell Dobree arrives in this country, Lady Violet Treherne
will give a luncheon for him at the Savoy Hotel. All the most worth-
while folks in London will be invited to meet him. I tell you, sir,
although you mustn’t tell the world just now, she has already
arranged with her friend the Prime Minister, if he happens to be
disengaged at the mo-ment, to attend the gathering and to give an
ad-dress on the Value of Literary Art in International Relations.”

They left Allardyce, Inc., balanced on the extreme edge of the kerb
of Henrietta Street, staring after their departing chariot. As they
drove off to luncheon at the Ladies Imperium, Lady Violet said, “My
child, I rather think you’ve clicked.”
Mame felt rather that she had. The happy feeling was confirmed,
moreover, a little later in the week when a second letter from Elmer
P. was delivered in Half Moon Street. In it, that now famous man
positively undertook to be at the Savoy Hotel on February 10, always
providing the Olympic in which he had booked a passage came in on
time. He hoped to stay a fortnight in London. But that, he feared,
must be his limit. For just now he was living a forty-eight-hour day in
New York.
“You can bet your life that’s so,” was Mame’s approving comment.
 LIII
ELMER came and saw London, England. And the ancient burg gave
him a real good time. He went here, there and everywhere; his
photograph was in all the papers; columns were written about his
book. There was a brilliant luncheon at the Savoy. Lady Violet kept
her promise. Big-wigs attended it, including her father’s old friend
the Prime Minister, who seized the occasion to deliver a most
significant address on the Value, Etc., which was cabled verbatim all
over the English-speaking world.
No young author, since the art of writing was invented, ever had a
more generous reception in the great metropolis. A modest, rather
shy, young man, he was inclined at first to be overwhelmed by it.
But the undefeated Mame, who met him at Euston, who took him to
his hotel, who gave him continual advice, saw to it that he wasn’t.
For the honour of Cowbarn, Iowa, he must stand right up to his job.
It was her task to see that he did so without flinching and she duly
performed it. She mothered him through receptions and tea parties;
she toted him around; and the bewildered and breathless Elmer
hardly knew whether to be more impressed by the storm his coming
had aroused or by the manner in which Mame rode it.
Nothing in the whole of London astonished him quite so much as
Mame’s transformation from her chrysalis Cowbarn period. Her
clothes, her style, her English accent fairly tickled him to death.
Then the friends she had made! She appeared to hob-nob with half
the swells in Britain and to have them feeding from the hand.
Elmer had many surprises in these crowded and glorious days. But,
shrewd and cool American citizen that he was, he managed to keep
a perfectly level head. For the life of him he couldn’t imagine what
all the fuss was about; or at least if he had an inkling of the reason
for it, he could not understand how Mame had contrived it all. She
had evidently had the luck to strike some very powerful backers.
Even before landing in England, he had surmised that such was the
case. The mysterious Celimene, of the weekly news-letter, had
proved to be so highly informed in social matters that her value had
been clearly demonstrated in New York. Her name had been given
him in confidence before he came over; and he was mighty keen to
meet her.
They might be said to challenge each other’s curiosity. But their
meeting not only fulfilled their hopes of one another; it was the
beginning of a friendship. One could not help liking the author of
Prairie City. He was a well-set-up young man; and behind the dry
shrewdness and the determination to get there, qualities
characteristic of Mame herself, were genuine kindliness and modesty.
His rise to fame had been less sudden than it seemed. It had been
prepared for and earned. He owned to thirty-one years of life. They
had not been easy years, but they had made him the man he was.
Lady Violet was glad that Elmer answered fully to Mame’s description
of him as “a regular fellow.” There was something about him that
inspired confidence. Whether it was a certain slowness of speech
which implied depth of mind, a vein of real grit, or the charming air
of diffidence with which he wore the fame that so deservedly was
his, she instinctively felt that here was what Mame called a he-man.
This was well. She had a plan in that sagacious mind of hers. But
the carrying out of it depended upon Elmer P. himself. Unless he
could pass the test, and a pretty severe one, that a thorough woman
of the world felt bound to impose, the fine scheme was doomed
from the outset. His bearing, however, in those crowded days in
which they saw a good deal of each other, convinced this friend that
rumour had not over-painted him. Undoubtedly the young man
deserved the position his talents had won. Beneath a surface a little
stiff and formal at first, and, the critic thought, none the worse for
that, was a warmth of heart and a balance of nature which enabled
him to pass his examination with flying colours.
As much time as Elmer could spare from his exceedingly numerous
engagements was devoted to Half Moon Street. From the first
afternoon he went there to drink tea, with reviewers and people of
influence in the world of letters, he took a great liking to the place.
For one thing he was made to feel so much at home. The presence
of Mame guaranteed that. She was quite unspoiled in spite of the
English accent, which to Elmer’s secret delight was apt to wear a
little thin in places. He was no end of an observer, the author of
Prairie City. Back in the Cowbarn days there was something in Mame
that had appealed to him; and in this new orientation she was still
the Mame he had liked, smiled at just a little, and yet admired.
Wonderful how she had been able to get away with it; yet he was
not really surprised. He had always known that his little
stenographer had a lot in her.
Everybody was so friendly in Half Moon Street. They seemed to take
quite a personal pride in his success; they seemed to treat it almost
as a part of their own. During the hours he spent there Mame and
her friend Lady Violet were always devising fresh schemes for Prairie
City. The boom was growing daily. But it must get bigger and bigger.
Had he been their own brother they could not have done more.
One afternoon, when Elmer had been in London a week, he came
rather early and happened to catch Lady Violet alone. Mame had
gone, at the call of duty, to the première of a new play. In this rather
providential absence, which yet did not owe quite so much to
providence as appeared on the surface, Lady Violet seized the
chance to have a private talk.
“So you are leaving us a week to-day?”
Elmer confessed that was his intention.
“If you find us all as complacent as our perfectly absurd newspapers
you won’t be sorry.”
Elmer had the tact to ignore the vexed question of the British
newspaper. “I’ll be sorry enough,” he said with simple sincerity. “You
are just giving me the time of my life.”
“Your book is so delightful. Every fresh reader is one friend more for
the man who has written it.” Lady Violet yielded to none in point of
tact. Over that course few could live with her. “But I do hope you
realise,” she went laughingly on, “that, although you are your own
best asset, as of course every true author has to be, you have also
had a very clever and enthusiastic friend to pull the strings over
here.”
Elmer realised that.
“One doesn’t say your success might not have been as great without
her; but it could hardly have come so soon.”
Yes, Elmer was sure.
“The way that dear child has worked for you has simply been
splendid. Had she written the book herself, I don’t think she could
have been prouder of it. She literally bullied your publishers into
boosting you—you know what even the best publishers are!—she
bullied me into corralling the Prime Minister—it was a rare bit of luck
getting him to come and make that speech—and it was her idea,
wasn’t it, that you should come over here and let us see you?”
Elmer felt all this was true. But gallantly he wanted to include Lady
Violet herself in the big bill of his gratitude.
“Please keep it all for Mame. That good child deserves every bit. She
has worked for you like a demon. As good as gold, as true as steel.
And she is quite cast down that you are leaving us next week.”
Unluckily there was no help for it. But Elmer P., like most people of
true genius, was simple at heart. He responded to the piping. Mame
Durrance—in the mouth of her former employer the accent fell upon
the first syllable of her surname rather than upon the second—
deserved all the luck there was in the world. She was as real as they
made them; and she was able to think of others.
Lady Violet drove that right home. “I, of all people, have reason to
know it. She is capable of big things, that dear child. Some day,
when you come and see us again, as of course you will, I may tell
you a little story about her.”
Elmer could not help a feeling of subtle flattery. It is difficult for
rising young men to resist such a feeling when they find themselves
tête-à-tête with an accomplished woman of the world. Lady Violet
was quite as intriguing as any of the Fifth Avenue queens, with one
or two of whom he was beginning to get acquainted. Mame had had
amazing luck to put herself in so solid with this fascinating woman.
It was true that Elmer personally owed a lot to Mame, a peerless
little go-getter, but it was also true that Mame for her part owed
much to this brilliant daughter of a famous statesman who in his day
had done a great deal for the English-speaking world.
The clock ticked pleasantly on to five o’clock and then quite as
pleasantly to 5:15. But no Mame. Lady Violet feigned surprise. Then
she glanced at an imposing array of cards on the chimneypiece. Oh,
yes, she remembered! This was the at-home day of the wife of an
influential editor. Mame had evidently kept it in mind even if Lady
Violet herself had forgotten. “She’s gone there to boost Prairie City,”
said her friend with a smile. “Wherever she goes now she boosts
Prairie City; and at night she boosts it in her dreams.”
A quarter past five already! Elmer Pell Dobree rose with a start. At
5:30—Lady Violet must really excuse him—he was due in Aldwych to
orate to the Journalists’ Circle on the Coming of the American Novel.
He was sorry to go, but Mame had fixed that on him at about twelve
hours’ notice, and as his good fairy declared it would mean another
edition, he supposed he must stand up, a hero, and face it.
“You must.” Lady Violet had what Elmer privately described as a
Gioconda smile. Here was guile, here was subtlety or the author of
Prairie City was not a judge of such matters. How intriguing she was.
Gee! she had a power of making your blood course quicker.
“I’ve one favour to ask, Lady Violet.” Elmer was moving to the door.
“Can you and Mame spare to-morrow evening to dine with me
quietly at the Savoy?”
Lady Violet took up a little red book from the writing table. A glance
revealed that by the courtesy of providence to-morrow night was
free. She could not answer for Mame, but to the best of her
recollection there was reason to think the little go-getter would not
be in action that evening.
So then it was arranged, a little dinner, just the three of them, for
the next evening, which was so providentially free, at the convenient
hour of eight. Unless, of course, Mame, which somehow Lady Violet
felt was hardly likely, telephoned him to the contrary.
 LIV
THE little dinner was capital. In every small but considered detail it
could not have been nicer. Elmer P., as the world looks for in one of
his eminence, was growing to be a judge of food and wine. Also the
shrewd dog knew how to choose his company. On his right sat Lady
Violet, on his left was Mame. Over and beyond this pair of friends
and boosters was vacancy, the limitless inane, at least so far as
those three minds were concerned, although at other tables sat
persons not without importance in their way.
In return for delicate food and dry champagne Elmer received high
entertainment from the lively tongues of his charming guests. Both
were observers of the human comedy, yet they observed it in the
right way. There was nothing in their talk that was spiteful or
backstairs, or beneath the dignity of human nature. Their aptness,
wit, and general information, their opinions upon books, plays, music
and the world at large gave the host a mental punch from the hors
d’œuvres to the comice pear and the crème de menthe, for which
crude liqueur both ladies confessed a partiality.
Elmer had had his triumphs, in the last week or so particularly, but
frankly he doubted whether he had ever enjoyed a meal like this. It
was so gay. And there was the glamour of new experience. His life
had suddenly been touched to newer and finer issues.
When the coffee appeared, at the end of the meal, Lady Violet drank
hers quickly. Then quite unexpectedly she rose. She would have to
fly. There was a musical party she had promised to attend. A stupid
affair, but it was the call of duty.
Mame and Elmer were pressing in their entreaties for their amusing
friend to remain, but she was not to be seduced from the true path.
Besides, as she laughingly said, it was a perfect night of stars. And
this being a sufficiently rare occurrence for London, she hoped that
Elmer—if she might use his Christian name?—would walk with Mame
through Trafalgar Square, along Pall Mall, up the Haymarket, across
the Circus and down the full length of Piccadilly. She ventured to
prescribe that route, because a little bird had whispered that if Elmer
duly followed it he might look for a very pleasant surprise, for which
Mame was alone responsible.
This was all so enigmatic that Elmer might have been tempted to
disbelieve Lady Violet. But he knew she was no trifler. Emphatically
she was one of the people who did not make promises unless they
were able to deliver the goods.
“We’ll go along and try it, anyway,” conceded Elmer the polite. And
being something of an altruist into the bargain: “But you’ll come
with us, won’t you? We can’t lose you!”
Lady Violet’s refusal was amusingly definite. She was late already;
she must fly. Besides, there was an even more cogent reason. That,
however, she was careful not to disclose to Elmer P. Dobree. As that
homme du monde moved a bit ahead of her to the restaurant door
to see her into her cloak and her taxi, she bent to Mame’s ear and
whispered urgently, “My child, if you don’t put one over on him to-
night I’ll never speak to you again.”
 LV
A BLITHE twenty minutes or so after Lady Violet had “flown,” Elmer
and Mame decided to get a move on. For one thing Elmer’s curiosity
had been tremendously piqued by the surprise that had been
predicted. What could it be? Lady Violet, he supposed, was just
pulling his leg. Yet he didn’t think so really; he knew she was not the
kind of person to break a solemn promise. Still there was nothing to
deduce from the attitude of Mame. The aider and abettor of Lady
Violet was giving nothing away. The stars were very bright, the air
for the time of year quite balmy, the pavements of London were dry
as a bone. All the conditions, therefore, were favourable for outdoor
exercise. Indeed, as Elmer said, or it may have been Mame who said
it, the evening was just ideal for the purpose.
Mame put on her lovely new cloak trimmed with fur, or at least
Elmer put it on for her. Then Elmer got into his overcoat and clapped
on his smart gibus, which gave him such a look of distinction, that a
loafer cadging for pence just beyond the courtyard of the hotel
promptly addressed him as Captain.
The mutt got the coppers all right. It was not so much that Elmer
was susceptible to that kind of flattery, as that just now he was not
in a mood to refuse anything to anybody. He was moving about this
evening in an enchanted world.
At every step they took in it now, the world through which they were
moving seemed to grow more entrancing. For one thing there was a
powerful magic in the stars. The strip of moon, too, as Mame
remarked, seemed to be trying to put one over on them. She made
this observation while they were in the act of steering each other
across the perilous vortex from Northumberland Avenue to Morley’s
Hotel, and nearly barging into more than one of their compatriots in
the process.
However, they crossed in safety. Then they crossed again by the
National Gallery and sauntered gaily along until they came to that
great landmark in Mame’s adventurous life, the Carlton Hotel. She
gave a long look at it as they went by. Even on this night of marvels
she could not pass that consecrated spot without a sense of
amazement and gratitude.
They turned up by the Haymarket, according to plan, and then
slowly rounded the corner into Piccadilly Circus. And then it was in
this identical moment that the goods were delivered in the most
unexpected and convincing way. The surprise that had been
solemnly promised Elmer appeared right before his eyes.
A flaming electric sign winked letter by letter from the starlit sky.

PRAIRIE CITY

BY ELMER PELL DOBREE

The Book
All the World
Is Reading
“Gee!” gasped Elmer. The secret had been carefully kept; he had not
an inkling! A surprise indeed, a masterpiece of boosting.
Mame’s voice rose in triumph. “Say, listen, Elmer. I’ll tell the world
this is where we put one over on London, England.”
Down Piccadilly they walked on air. No word passed. But to keep in
touch with himself and the mundane realities Elmer took Mame’s
hand. These were sublime moments. Suddenly, high above the
famous street, the sign flamed out again.
“Say, listen, Mame,” began Elmer hoarsely. But even with all his
genius to help him he did not know how to end, so he merely
squeezed her hand.
The dear little go-getter, how slick she was! But she was also
something much better than slick. She was fine and true. A minute
they stood gazing at the recurring sign in all its brilliancy and then,
life being too wonderful to stand still in it, they moved on hand in
hand.
Sure it was destiny they should be walking thus, four thousand
miles, four thousand solid miles, from the dear funny old spot in
which they had walked last. If only Cowbarn, Iowa, could see that
sign. The book all the world was reading; the book that had
immortalised the Folks. Would they recognise themselves in all their
humour and their quiddity?
When speech was possible between them, which was not until they
were near the precincts of the Ritz, it was Mame who dared. “Elmer,”
her voice was very soft, “I’m feeling pretty good about our book.”
She said “our book.” “There’s not one word we’ll ever have to wish
away. The folks aren’t saints, the folks aren’t, but there ain’t a line
that’s mean. There’s nothing to make ’em sorry. Some of the stories
you might have told you didn’t tell. Some of the things you might
have said you didn’t say. Elmer, I’m feeling pretty good about that
book.”
Elmer, too, was feeling pretty good. In fact so good was Elmer
feeling, that for all he was fully launched in the realm of letters, he
still couldn’t find a word. Not one word. But like all young men of
force and originality he enjoyed a certain power of action. Quite
suddenly, without premeditation, he put one over on Mame. In the
dark shadow cast by the Ritz arcade, he kissed the little go-getter.
Mame was thrilled by the sheer audacity of the assault. But there
was the authority of the book all the world was reading that the
heart of woman is a queer thing; so she just didn’t mind at all.
However, she did not speak again, until hand in hand they had
convoyed each other past a line of pirates in the guise of taxi
cabmen, whom the law allows to range themselves in a row opposite
Devonshire House. Nay, she didn’t speak until they had passed the
end of Half Moon Street, the other side of the road, and on by the
railings of the Green Park.
It was when they halted to gaze at the bright windows of the Ladies
Imperium that Mame’s soft voice was heard.
“That’s the hen club I belong to.”
Elmer was impressed.
“It’s the Chickest hen club in London.”
Elmer guessed it was.
“Cocks are not admitted. If they was,”—Mame said was, yet she
knew perfectly well it wasn’t grammar,—“I’d take you right in and
buy you a cocktail.”
This was a little too much for the author of Prairie City. Such un-
American play upon words was the palpable fruit of mental stress,
but in combination with the magic of the stars it was a little too
much for Elmer P. Dobree.
“Now, then, Mame Durrance, can that.” And then immediately
opposite those flaming windows and in the lee of the park railings
which hid them well, he kissed her again with rapturous violence.
After this stimulating episode they moved slowly along by the Green
Park. They still walked hand in hand; even now it didn’t seem safe to
let go of each other. But when they came to the Quadriga, that
symbol of victory significantly poised on the top of the park gates,
which is much and justly admired, they stopped and gazed up.
They gazed up at the Quadriga by the royal light of the stars. Their
hands were locked in each other’s. The eternal verities caught them
suddenly. Mighty, mighty forces were flowing through and over,
through and over, this brief and transient, this pitifully brief and
transient, life of man.
“Elmer!” It was Mame’s voice, but hardly more than a whisper, it was
so solemn and so hushed. “Imagine, Elmer, you and me—”
But Elmer said nothing. With a queer tightening of the breast he
continued to gaze upwards to the symbol of victory on the top of the
park gates.
THE END
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

ebookfinal.com