Professional Documents
Culture Documents
Ebook Fundamentals of Aeroacoustics With Applications To Aeropropulsion Systems 1St Edition Xiaofeng Sun Online PDF All Chapter
Ebook Fundamentals of Aeroacoustics With Applications To Aeropropulsion Systems 1St Edition Xiaofeng Sun Online PDF All Chapter
Applications to Aeropropulsion
Systems 1st Edition Xiaofeng Sun
Visit to download the full and correct content document:
https://ebookmeta.com/product/fundamentals-of-aeroacoustics-with-applications-to-a
eropropulsion-systems-1st-edition-xiaofeng-sun/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
https://ebookmeta.com/product/photovoltaic-systems-fundamentals-
and-applications-abou-jieb/
https://ebookmeta.com/product/fundamentals-of-analysis-with-
applications-atul-kumar-razdan/
https://ebookmeta.com/product/fundamentals-of-engineering-
tribology-with-applications-1st-edition-harish-hirani/
https://ebookmeta.com/product/building-integrated-photovoltaic-
thermal-systems-fundamentals-designs-and-applications-1st-
edition-huiming-yin/
https://ebookmeta.com/product/building-integrated-photovoltaic-
thermal-systems-fundamentals-designs-and-applications-1st-
edition-huiming-yin-2/
https://ebookmeta.com/product/medical-image-synthesis-methods-
and-clinical-applications-imaging-in-medical-diagnosis-and-
therapy-1st-edition-xiaofeng-yang-editor/
https://ebookmeta.com/product/modeling-of-dynamic-systems-with-
engineering-applications-2nd-edition-clarence-w-de-silva/
https://ebookmeta.com/product/plasmonic-catalysis-from-
fundamentals-to-applications-1st-edition-pedro-h-c-camargo/
FUNDAMENTALS OF
AEROACOUSTICS WITH
APPLICATIONS TO
AEROPROPULSION SYSTEMS
This page intentionally left blank
FUNDAMENTALS OF
AEROACOUSTICS WITH
APPLICATIONS TO
AEROPROPULSION SYSTEMS
Elsevier and Shanghai Jiao Tong
University Press Aerospace Series
XIAOFENG SUN
Fluid and Acoustic Engineering Laboratory,
School of Energy and Power Engineering,
Beihang University, Beijing, China
XIAOYU WANG
Fluid and Acoustic Engineering Laboratory,
Research Institute of Aero-Engine,
Beihang University, Beijing, China
Academic Press is an imprint of Elsevier
125 London Wall, London EC2Y 5AS, United Kingdom
525 B Street, Suite 1650, San Diego, CA 92101, United States
50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom
© 2021 Shanghai Jiao Tong University Press. Published by Elsevier Inc.
No part of this publication may be reproduced or transmitted in any form or by any means,
electronic or mechanical, including photocopying, recording, or any information storage and retrieval
system, without permission in writing from the publisher. Details on how to seek permission,
further information about the Publisher’s permissions policies and our arrangements with
organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be
found at our website: www.elsevier.com/permissions.
This book and the individual contributions contained in it are protected under copyright by the
Publisher (other than as may be noted herein).
Notices
Knowledge and best practice in this field are constantly changing. As new research and experience
broaden our understanding, changes in research methods, professional practices, or medical
treatment may become necessary.
Practitioners and researchers must always rely on their own experience and knowledge in
evaluating and using any information, methods, compounds, or experiments described herein.
In using such information or methods they should be mindful of their own safety and the safety of
others, including parties for whom they have a professional responsibility.
To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors,
assume any liability for any injury and/or damage to persons or property as a matter of products
liability, negligence or otherwise, or from any use or operation of any methods, products, instructions,
or ideas contained in the material herein.
Library of Congress Cataloging-in-Publication Data
A catalog record for this book is available from the Library of Congress
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
ISBN 978-0-12-408069-0
Preface vii
v
vi Contents
Index 533
Preface
Aeroacoustics had its beginnings in the early 1950s thanks to the pioneering
work of Sir James Lighthill about aerodynamic sound. Since then, this dis-
cipline has concentrated on understanding noise generation by either turbu-
lent fluid motion or aerodynamic forces interacting with surfaces. In
particular, emphasis has also been placed on aeronautical applications for
noise reduction of both aircraft and aeropropulsion systems. At present,
there are already several books available that introduce basic knowledge
and progress in its various aspects. It is noted, however, that the most prom-
inent noise sources of modern aircrafts originate from the aeropropulsion
system in many situations; meanwhile, the thermoacoustic oscillations
occurring in the combustion chamber of the aeroengine has received great
attention on its prediction and control, which is equally related to aeroa-
coustics. Therefore, it is necessary to have a book dedicated to aeroengine
acoustic problems.
As is well known, there are various scales, such as different vortex struc-
tures, entropy, and pressure fluctuations in moving flow media, which make
it extremely difficult to accurately calculate and analyze all flow and acoustic
field details under the condition of multiscales. There are thus two ways to
get work done. One method is Lighthill’s acoustic analogy theory, which is
still widely applied in various aerodynamic noise calculations, especially for
rotating source prediction. The other method is numerical techniques like
computational aeroacoustics (CAA), which suppresses both dissipation and
dispersion on the basis of the differential schemes. There is no doubt that
CAA has shown great potential for the study of jet noise and other flow-
induced acoustic problems. Still, given the limits of current computational
capabilities, acoustic computation for a problem of practical interest is still
out of reach by directly solving Navier-Stokes equations, particularly for
a numerical simulation of rotating or moving sound source problems in asso-
ciation with aeroengine noise generation.
With reference to the context, the aim of this book is to develop a unified
framework to handle the acoustic problem of aeropropulsion systems for the
design phase of an aeroengine; meanwhile, it is also intended for scientists,
engineers, and graduates who are interested in modeling and doing it right.
Therefore, the core parts of this book are still based on the acoustic analogy
theory or the unified solution approach of flow and acoustic fields under the
vii
viii Preface
Xiaofeng Sun
Xiaoyu Wang
Beihang University
Beijing, China
February 2020
This page intentionally left blank
CHAPTER 1
Fundamentals of Aeroacoustics with Applications to Aeropropulsion Systems © 2021 Shanghai Jiao Tong
https://doi.org/10.1016/B978-0-12-408069-0.00001-4 University Press. 1
Published by Elsevier Inc.
2 Fundamentals of aeroacoustics with applications to aeropropulsion systems
ρ Density
v Velocity
p Pressure
S Entropy
!
p 0 , v 0 , ρ0 Mean flow quantities
!
p 0 , v 0 , ρ0 Fluctuations
q An external volume flow source within the fluid
c0 Speed of sound
f An externally applied volume force
U Mean flow velocity
E Acoustic energy density
!
I Acoustic energy flux
τ Time associated with emission of sound wave
t Time associated with the arrival of sound wave at observation point
G Green’s function, i.e., fundamental solution of wave function
G0 Free-space Green’s function
R Distance between the sound source and the observation point
ψ Velocity potential
M Mach number, M ¼ U/c0
Tij Lighthill stress tensor
whereρ, v, p, and S stand for fluid density, velocity, pressure, and entropy,
respectively, whereas q and f denote an external volume source and an
externally applied volume force, respectively. There are four unknowns
in Eq. (1.2), and an additional equation is required to close the equation.
Assume that:
ρ ¼ ρðp, SÞ,
thus,
∂ρ ∂ρ
dρ ¼ dp + dS: (1.3)
∂p S ∂S p
∂p
Considering c02 ¼ Dt ¼ 0, therefore
, and for an isotropic motion, DS
∂ρ S
it follows from Eq. (1.3) that:
∂ρ 1 ∂p
+ v rρ ¼ 2 + v rp : (1.4)
∂t c0 ∂t
Applying the following relations:
p ¼ p0 + p0 , v ¼ v0 + v0 , ρ ¼ ρ0 + ρ0 , S ¼ s0 + S0 ,
and the steady Euler’s equation:
8
>
> r ρ0 v0 ¼ 0
<
ρ0 v0 rv0 ¼ rp0
, (1.5)
>
> v0 rS0 ¼ 0
: 2
c0 v0 rρ0 ¼ v0 rp0
the basic equations of sound propagation in moving media can be derived as:
8
>
> ∂ρ0
>
> + r ðρ0 v0 + v0 ρ0 Þ ¼ ρ0 q
>
> ∂t
>
> ∂v 0
>
> 0 0
+ v0 rv + v rv0 + ρ0 v0 rv0 ¼ rp0 + f
< ρ0
∂t
(1.6)
>
> ∂s0 0 0
>
> + v rs + v rS ¼ 0
>
> 0 ∂t
0
0
>
> ∂ρ ∂p0
>
> 0
+ v rρ0 + v0 rρ ¼ 0
+ v0 rp0 + v0 rp0
: c02
∂t ∂t
The results are actually consistent with linearized aerodynamic equa-
tions. Under different simplified conditions, the equations can be used to
Basic equations of aeroacoustics 5
y2
U(y 2)
y1
study the sound propagation in ducts. For example, if a parallel shear flow is
considered as shown in Fig. 1.1, along with the assumption that
v0 ¼ i U ½rðy2 , y3 Þ, ρ0 ¼ constant, p0 ¼ constant, S0 ¼ constant: (1.7)
Substituting Eq. (1.7) into Eq. (1.6) yields:
8
>
> 1 D0 p 0 0
>
>
> 2 Dt + ρ0 r v ¼ ρ0 q
>
< 0 0
c
D0 v dU
ρ0 + i vr g ¼ rp0 + f , (1.8)
>
> Dt dr
>
>
>
> D 0s
0
: ¼0
Dt
where
qffiffiffiffiffiffiffiffiffiffiffiffiffi
D0 ∂ ∂ rr
¼ +U , vr ¼ v 0 , gðy2 , y3 Þ ¼ jrr j, r ¼ y22 + y23 :
Dt ∂t ∂y1 jrr j
D0
Applying to the first equation of Eq. (1.8), and imposing r to the
Dt
second equation in Eq. (1.8) yields:
1 D20 p0 dU ∂vr D0 q
r2 p0 + 2ρ0 g ¼ r f ρ0 : (1.9)
2
c0 Dt 2 dr ∂y1 Dt
Because:
0
D0 v 0 ∂v ∂v0 D0 dU ∂vr
r ¼r +U ¼ ðr v0 Þ + g ,
Dt ∂t ∂y1 Dt dr ∂y1
6 Fundamentals of aeroacoustics with applications to aeropropulsion systems
dU ∂ dU dU ∂vr ∂ dU
r i vr g ¼ vr g ¼ g + vr g
dr ∂y1 dr dr ∂y1 ∂y1 dr
dU ∂vr
¼ g ,
dr ∂y1
and for uniform flow, v0 ¼ U ¼ constant, Eq. (1.9) can be written as:
1 D20 p0 D0 q
r2 p0 ¼ r f ρ0 ¼ γ: (1.10)
c02 Dt 2 Dt
This is the basic equation used to study sound propagation in ducts with
uniform flow [11]. When U ¼ 0, the resulting equation becomes:
1 ∂2 p0 ∂q
r2 p0 ¼ r f ρ0 : (1.11)
c02 ∂t2 ∂t
This is the basic equation for classical acoustics.
or
∂E !
+ r I ¼ 0:
∂t
Eq. (1.14) means that, no matter what form of definition of sound energy
flux is given, it must satisfy the conservation law. However, as far as moving
media are concerned,it is still very difficult to give a reasonable definition
Basic equations of aeroacoustics 7
for a general case [12, 13]. But for an isentropic irrotational flow, sound
energy flux and sound intensity can be given by:
!2 !
p0 ρ0 ρ0 u 0* !
E¼ + + ρ u v0 , (1.15)
2ρ0 2
0
! p * ! * !
I¼ + u v 0 ρ0 u + ρ0 v 0 : (1.16)
ρ0
It can be verified that Eqs. (1.16) and (1.15) satisfy the conservation law
!
in Eq. (1.14). When v 0 ¼ 0, the relevant equations reduce to Eqs. (1.12) and
(1.13) for classical acoustics.
1 ∂2 ψ
r2 ψ ¼ ρ0 q0 ðτÞδ½y y0 ðτÞ: (1.17)
c02 ∂τ2
R(x,t)
R( x,t e)
Source at time t
Fig. 1.2 Geometrical relation of a moving point source.
8 Fundamentals of aeroacoustics with applications to aeropropulsion systems
where γ(y, τ) is the source term, whereas G0(y, τ/x, t) denotes Green’s func-
tion for free space, which can be expressed as:
1 R
G0 ¼ δ τt + : (1.19)
4πR c0
Hence the solution of Eq. (1.17) is:
! ρ0 q0 ðτ∗ Þ
ψ t, x ¼ , (1.20)
4πR∗ ð1 Mr ∗ Þ
where
Ri V
R ¼ jx y0 ðτÞj, Mr ¼ , Ri ¼ xi y0i ,
R c0
and τ∗ is the root of the retarded time equation g ¼ τ t + jx y0(τ)j/c0
(noting that τ∗ can have several different choices, and superscript “*”
denotes the value of the corresponding quantity at time τ∗), whereas the
derivative of g is:
dg 1 ½x y0 ðτÞ dy0 R Ri
¼1 ¼1 V¼1 Vi ¼ 1 Mr
dτ c0 jx y0 ðτÞj dτ R c0 R c0
(1.21)
Consequently, the solution of sound pressure is:
∂ψ ρ0 q0 0 ðτ∗ Þ ∂τ∗ ρ0 q0 ðτ∗ Þ ∂ ∗
p0 ¼ ¼ ½R ð1 Mr ∗ Þ: (1.22)
∂t 4πR∗ ð1 Mr ∗ Þ ∂t 4πR∗2 ð1 Mr ∗ Þ2 ∂t
In consideration of
∂τ∗ ∂ R∗ 1 ∂R∗ 1 ∂R∗ ∂τ∗ 1 Ri ∗ ∂τ∗
¼ t ¼1 ¼1 ¼1 ðVi Þ
∂t ∂t c0 c0 ∂t c0 ∂τ∗ ∂t c0 R∗ ∂t
∂τ∗
¼ 1 + Mr ∗
∂t
Basic equations of aeroacoustics 9
i.e.,
∂τ∗ 1
¼ , (1.23)
∂t 1 Mr ∗
and
∂ ∗ ∂R∗ ∂τ∗ ∂Mr ∗ ∂τ∗
½R ð1 Mr ∗ Þ ¼ ð1 Mr ∗ Þ R∗ ,
∂t ∂τ∗ ∂t ∂τ∗ ∂t
where
∂R∗ Ri ∗ V
¼ ¼ c0 Mr ∗
∂τ∗ R∗
and
∂Mr ∗ ∂ Ri ∗ ∗ Ri ∗ ∂Mi∗ ∗ 1 ∂ ∗ ∂ ∗
¼ Mi ¼ + Mi 2 R∗ Ri Ri ∗ R
∂τ∗ ∂τ∗ R∗ R∗ ∂τ∗ R∗ ∂τ∗ ∂τ∗
Ri ∗ ∂Mi∗ c0
¼ M ∗2 Mr∗2
R∗ ∂τ∗ R∗
we obtain
∂ ∗ Ri ∗ ∂Mi ∗ M ∗2 Mr ∗
½R ð1 Mr ∗ Þ ¼ + c0 (1.24)
∂t 1 Mr ∗ ∂τ∗ 1 Mr ∗
Substituting Eqs. (1.23) and (1.24) into Eq. (1.22) yields:
2 (1.25)
ρ0 q0 ðτ∗ Þc0
3 M ∗ Mr ∗
4πR∗ ð1 Mr ∗ Þ
2
Eq. (1.25) denotes the sound field generated by a monopole source with
arbitrary motion. For a far-field solution, the third term including 1/R∗2 can
be ignored. Compared with the first term and second term, the solution is
thus expressed as:
0 ρ0 ∂q0 q0 ðτ∗ Þ Ri ∗ ∂Mi
p ðx, tÞ ¼ + , (1.26)
4πR∗ ð1 Mr ∗ Þ2 ∂τ τ¼τ∗ ð1 Mr ∗ Þ R∗ ∂τ τ¼τ∗
and the near-field solution becomes:
ρ0 c0 q0 ðτ∗ Þ ∗
p0 ðx, tÞ ¼ 3 Mr M ∗2 : (1.27)
4πR∗2 ð1 Mr ∗ Þ
10 Fundamentals of aeroacoustics with applications to aeropropulsion systems
(1.29)
For an observer at time t, the solution of Eq. (1.29) may give two differ-
ent source radiation timeτ; thus, the distance between the observer and
source can be expressed as:
R ¼ c t τ
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
! !2
M0 ðx1 V0 tÞ ½M0 ðx1 V0 tÞ2 + ð1 M02 Þ x V0 t i
¼ :
1 M02
(1.30)
Obviously, only the positive real root of Eq. (1.30) has an exact physical
meaning. On the other hand, for subsonic motion, i.e., V0 < c0, Eq. (1.30)
has only one real root, and R+ is a physical solution. This can also be seen
with the help of the geometrical relation described in Fig. 1.3. Therefore,
when the point source is rectilinearly moved in subsonic speed, the expres-
sion for the sound field reduces to:
ρ0 q00 ðτÞ ρ0 q0 ðτÞV0
p0 ðx, t Þ ¼ 2 + ð cos θ M0 Þ: (1.31)
4πRð1 M0 cos θÞ 4πR ð1 M0 cos θÞ3
2
Basic equations of aeroacoustics 11
R=c 0(t–τ)
x
R
x–v0 t
v 0τ θ
v0 t
Besides, when the moving speed is zero, i.e., V0 ¼ 0, the previous equa-
tion further becomes:
ρ0 q00 ðτÞ
p0 ðx, tÞ ¼ : (1.32)
4πR
Eq. (1.32) is the expression of the sound field generated by monopole for
classic acoustics.
When the point source moves in supersonic speed, i.e., V0 > c0,
Eq. (1.30) may have two positive real roots. This means that the observer
can receive two signals at time t, which radiate from two different positions.
But the condition for two positions of R is:
! !2
½M0 ðx1 V0 tÞ2 + 1 M02 x V0 t i > 0, (1.33)
or
pffiffiffiffiffiffiffiffiffiffiffiffiffiffi
M02 1 jx1 V0 t j
< !
(1.34)
M0 !
x V 0t i
It is seen that the left side of Eq. (1.34) denotes the cosine relation of the
pffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Mach angle, i.e., α ¼ cos 1 M02 1=M0 . As shown in Fig. 1.4, the right
side represents the cosine relation of the angle between the moving direction
and the connecting line from the observer
i to the vertex of the Mach cone, i.e.,
h
! !
δ ¼ cos 1 ðx1 V0 t Þ= x V0 t i . As shown in Fig. 1.5, Eq. (1.34) can
be expressed as δ < α. This indicates that the condition that R has two
positive real roots is that the observer must be located inside the Mach cone.
12 Fundamentals of aeroacoustics with applications to aeropropulsion systems
Q
a
R+ R–
q–
q+
–
t t t
Fig. 1.4 Geometrical relation for an observer receiving two signals.
Fig. 1.5 Observer within the Mach cone will hear the sound coming from two different
points.
If the condition is satisfied, the observer can hear two different signals at time t
from different positions as shown in Fig. 1.5.
When the point source is moving at supersonic speed, the acoustic field
can described as:
ρ0 q00 ðt R + =c0 Þ ρ0 q00 ðt R =c0 Þ
p0 ðx, tÞ ¼ +
4πR + ðM0 cosθ + 1Þ
2
4πR ðM0 cos θ 1Þ2
ρ0 V0 q0 ðt R + =c0 ÞðM0 cos θ + Þ
3 (1.35)
4π ðR + Þ2 ðM0 cosθ + 1Þ
ρ0 V0 q0 ðt R =c0 ÞðM0 cos θ Þ
+ :
4π ðR Þ2 ðM0 cos θ 1Þ3
It is seen that, compared with the subsonic case, the sound pressure in
Eq. (1.35) has some additional terms. In addition, if the strength of sound
source q0(τ) is constant, the sound pressure in far field will be approximately
Basic equations of aeroacoustics 13
1 ∂2 Ai
r2 A i ¼ fi ðτÞδ½y y0 ðτÞ: (1.37)
c02 ∂τ2
Its solution is similar to that of the monopole source mentioned earlier.
According to generalized Green’s function formula, the solution of
Eq. (1.37) can be expressed as:
fi ðτÞ
Ai ¼ s (1.38)
4πRð1 Mr Þτ¼τ∗
For simplicity, in the following content we drop the superscript “*,” so it
should be noted that the corresponding values at τ∗ are assigned to those rel-
evant variables. One obtains:
0 ∂ fi ðτÞ
p ðx, tÞ ¼
∂xi 4πRð1 Mr Þ
(1.39)
1 ∂fi ðτÞ ∂τ fi ðτÞ ∂½Rð1 Mr Þ
¼ + ,
4πRð1 Mr Þ ∂τ ∂xi 4π ½Rð1 Mr Þ2 ∂xi
∂τ
where ∂x i
and ∂½Rð1M
∂xi
r Þ
are all derivatives of the compound function. In
terms of the relevant derivative rule, the following relations can be obtained:
∂τ ∂ Rðx, τÞ 1 ∂ ∂ ∂τ
¼ t ¼ ½Rðx, τÞτ¼const + ½Rðx, τÞx¼const ,
∂xi ∂xi c0 c0 ∂xi ∂τ ∂xi
and
∂τ Ri
¼ : (1.40)
∂xi c0 Rð1 Mr Þ
14 Fundamentals of aeroacoustics with applications to aeropropulsion systems
Similarly,
∂½Rð1 Mr Þ ∂ ∂ ∂τ
¼ ½Rð1 Mr Þτ¼const + ½Rð1 Mr Þx¼const : (1.41)
∂xi ∂xi ∂τ ∂xi
The first term of Eq. (1.41) can be written as:
∂ ∂R ∂Mr Ri ∂ Rj
½Rð1 Mr Þτ¼const ¼ ð1 Mr Þ R ¼ ð1 Mr Þ RM j
∂xi ∂xi ∂xi R ∂xi R
Ri Ri Rj Ri
¼ ð1 Mr Þ Mj δij ¼ Mi :
R R R R
(1.43)
1 fi Ri
+ 1 Mi fi Mi :
2
4πR2 ð1 Mr Þ2 ð1 Mr Þ R
In view of different influences for the terms related to 1/R and 1/R2, the
solution of a far sound field is:
Basic equations of aeroacoustics 15
0 1 Ri ∂fi fi ðτÞ Ri Rj ∂Mj
p ðx, tÞ ¼ + , (1.44)
4πc0 Rð1 Mr Þ R ∂τ ð1 Mr Þ R R ∂τ
2
From Eq. (1.44), if the required solution of a far sound field is zero, the
relevant conditions are that the dipole source must be steady, i.e.,
∂ fi/∂ τ ¼ 0, and at the same time there is no accelerating motion for the
source, i.e., ∂Mj/∂τ ¼ 0.
For uniform rectilinear motion of a dipole source, this means that
y0(τ) ¼ V0τ, M1 ¼ M0 ¼ V0/c0, and Mr ¼ RR1 Vc00 ¼ M0 cosθ, where θ
is the included angle between the moving direction and observer point,
whereas Mr is the projection of the Mach number of the moving direc-
tion with respect to the direction of the observer point. Eq. (1.43) thus
reduces to:
f10 ðτÞcos θ f1 ðτÞð cos θ M0 Þ
p0 ðx, t Þ ¼ 2 + : (1.46)
4πc0 Rð1 M0 cosθÞ 4πR2 ð1 M0 cosθÞ3
This is the solution of sound pressure for a point dipole source with uni-
form rectilinear motion when the moving direction of the dipole is consis-
tent with the pole of the sound source. When V0 ¼ 0, Eq. (1.46) becomes:
f10 ðτÞcos θ f1 ðτÞ cos θ
p0 ðx, tÞ ¼ + : (1.47)
4πc0 R 4πR2
This is the expression of sound pressure for a longitudinal dipole.
As for a transversal dipole that is perpendicular to the moving direction,
considering R2/R ¼ sin θ and fiMi ¼ 0, Eq. (1.43) reduces to:
0 f20 ðτÞsin θ f2 ðτÞsin θ 1 M02
p ðx, t Þ ¼ + : (1.48)
4πc0 Rð1 M0 cosθÞ2 4πR2 ð1 M0 cosθÞ3
When the moving speed of the source is zero, this formula will become:
f20 ðτÞsin θ f2 ðτÞ sinθ
p0 ðx, tÞ ¼ + : (1.49)
4πc0 R 4πR2
This is also a classical expression for a point dipole source.
16 Fundamentals of aeroacoustics with applications to aeropropulsion systems
1 dR 1 ðM0 cos θ ÞM0 M0 cos θ
¼ 2 M02 + ¼ :
c0 dt M0 1 M0 cosθ 1 M0 cosθ 1
It can be seen that, for the supersonic case, the range of receiving fre-
quency at the observer point corresponds to two different intervals, i.e.,
ω0 ω0
∞ < ω < , < ω < + ∞:
1 M0 M0 + 1
The results also indicate that the observer will receive different frequen-
cies simultaneously if the sound source moves at supersonic speed. The
interaction between different frequencies will cause dramatic pressure
fluctuations.
It can be verified that [14], no matter how the sound source moves,
supersonically or subsonically, the sound radiation power in far field is:
Es
E¼ 2, (1.55)
ð1 M02 Þ
where Es represents the sound power for a static source; if the source is har-
monic with the frequency f0, its expression is:
π ðq0 f0 Þ2
Es ¼ : (1.56)
2ρ0 c0
However, because:
D20 p0 2
0 D0 G D0 D0 p0 0 D0 G
G 2 p ¼ G p
Dτ Dτ2 Dτ Dτ Dτ
0
∂ D0 p 0 D0 G ∂ D0 p0 0 D0 G
¼ G p +U G p ,
∂τ Dτ Dτ ∂y1 Dτ Dτ
it can be shown that:
ð 2
D20 p0 0 D0 G
G 2 p dy
vðτÞ Dτ Dτ2
ð ð
∂ D0 p0 D0 G ∂ D0 p0 D0 G
¼ G p0 dy + U G p0 dy
vðτÞ ∂τ Dτ Dτ vðτÞ ∂y1 Dτ Dτ
ð ð
d D0 p0 D0 G ! D0 p0 D0 G
¼ G p0 dy + ðiU VS Þ n G p0 dSðyÞ:
dτ vðτÞ Dτ Dτ SðτÞ Dτ Dτ
Basic equations of aeroacoustics 19
Observer
x
Source
y
v c(τ) v(τ)
Moving body
S(τ)
Fig. 1.6 Sound radiation from bounded source region.
20 Fundamentals of aeroacoustics with applications to aeropropulsion systems
where
Tij ¼ ρui uj eij + δij p c02 ρ ,
and
0 i¼j
δij ¼ :
1 i¼
6 j
After linearizing the equation, the wave equation with fluctuating den-
sity as a variable can be given in the form of:
2 0
∂2 ρ0 2 2 0 ∂ Tij
c0 r ρ ¼ , (1.70)
∂t 2 ∂yi ∂yj
where ρ0 ¼ ρ ρ0, whereas Tij0 ¼ ρuiuj eij + δij[(p p0) c20(ρ ρ0)] is
called Lighthill’s turbulence stress tensor.
Eq. (1.70) is Lighthill’s equation on the basis of the concept of acous-
tic analogy. However, it is difficult to find its analytic solution because
the two sides of Eq. (1.70) all contain unknown variables. The only pos-
sibility is to directly solve Eq. (1.70) using a numerical approach. How-
ever, this equation actually deals with various flow scales; it was found
that the numerical solution with such a multiscale problem was also very
difficult in a long period of time. However, in terms of Lighthill’ theory,
the right term of Eq. (1.70), i.e., the source term, can be independently
obtained either by numerical simulation with a single-flow scale or by an
experimental approach. With a known source term, the sound pressure
can be given by solving the conventional wave equation. In fact, Light-
hill’s acoustic analogy offers a very successful application for various prac-
tical problems.
ðT ð
0
1 ! ∂2 Tij0
ρ t, x ¼ 2 G dydτ
c0 vðτÞ ∂yi ∂yj
T
ðT ð
∂ Vn0 ∂ 0 0 ∂ Vn0 ∂
+ G + ρ ρ + G dSðyÞdτ:
SðτÞ ∂n c02 ∂τ ∂n c02 ∂τ
T
(1.71)
It is known that:
0
∂ ∂Tij ∂ 0 ∂G ∂2 Tij0 ∂2 G
G Tij ¼G Tij0 ,
∂yi ∂yj ∂yj ∂yi ∂yi ∂yj ∂yi ∂yj
and using ni ∂y∂ i to replace ∂n∂ ,the application of the divergence theorem
results in:
ð 0 ð
∂ ∂Tij ∂Tij0 !
G dy ¼ G n dSðyÞ,
vðτÞ ∂yi ∂yj SðτÞ ∂yj
ð ð
∂ 0 ∂G ∂G !
Tij dy ¼ Tij0 n dSðyÞ:
vðτÞ ∂yj ∂yi SðτÞ ∂yi
Hence,
ðT ð
! 1 ∂2 G
0
ρ t, x ¼ 2 Tij0 dydτ
c0 vðτÞ ∂yi ∂yj
T
ðT ð
1 ∂ 2 0 ∂ρ0
+ 2 Gni c0 δij ρ + Tij0 + Vis dSðyÞdτ (1.72)
c0 SðτÞ ∂yj ∂τ
T
ðT ð
1 ∂G ∂G
2 nj Tij0 + c02 δij ρ0 + Vjs ρ0 dSðyÞdτ,
c0 SðτÞ ∂yi ∂τ
T
0
where ni ∂y∂ j δij ρ0 ¼ ni ∂ρ
∂yi . Replacing the second and third terms on the right
side of Eq. (1.72) using Lighthill’s stress tensor, the corresponding solution
becomes:
ðT ð ðT ð
! 1 ∂2 G 1 ∂G
0
ρ t, x ¼ 2 Tij0 dydτ + 2 fi dSðyÞdτ
c0 vðτÞ ∂yi ∂yj c0 S ðτ Þ ∂yi
T T
(1.73)
ðT ð
1
2 ni hi dSðyÞdτ,
c0 SðτÞ
T
Basic equations of aeroacoustics 23
where
fi ¼ nj eij ni ðp p0 Þ, (1.74)
which denotes force per unit area acting on the body. Besides,
0
∂ s ∂ρ
ni hi ¼ Gni ρui uj eij + δij ðp p0 Þ + Vi
∂yj ∂τ
(1.75)
∂G ∂G
+ nj Vjs ρ0 + ρui uj :
∂τ ∂yi
In terms of Eq. (1.67), it is known that:
∂ρui uj ∂eij ∂p ∂ρui
+ ¼ :
∂yj ∂yj ∂yi ∂t
Introducing this relation into Eq. (1.75) yields:
∂ρui ∂ρ0 ∂G ∂G
hi ¼ G Vis + ρui uj + ρ0 Vis : (1.76)
∂τ ∂τ ∂yj ∂τ
Assume the body is impenetrable, so this means niui ¼ niVsi . Naturally,
∂ρui ∂ρ0 ∂G ∂G
ni hi ¼ ni G Gui + ρui uj + ni ui ρ0 :
∂τ ∂τ ∂yj ∂τ
Notice that ρ0 ¼ ρ ρ0, hence:
∂ρui G ∂ρ ∂G ∂G
ni hi ¼ ni Gui + ρui uj ni ρ0 ui :
∂τ ∂τ ∂yj ∂τ
Substituting Eq. (1.66) into the previous equation yields:
∂ρui G ∂ρuj G ∂G
ni hi ¼ ni + ui ni ρ0 ui : (1.77)
∂τ ∂yj ∂τ
The application of Leibniz’s rule results in:
ð ð ð
d ∂ρuj G ∂ ∂ρuj G ∂ρuj G
dy ¼ dy + ni ui dSðyÞ, (1.78)
dτ vðτÞ ∂yj vðτÞ ∂τ ∂yj SðτÞ ∂yj
and by applying the divergence theorem, we have:
ð ð
d ∂ρuj G ∂ρui G ∂ρuj G
dy ¼ ni + ui dSðyÞ: (1.79)
dτ vðτÞ ∂yj S ðτ Þ ∂τ ∂yj
24 Fundamentals of aeroacoustics with applications to aeropropulsion systems
Hence,
ðT ð ðT ð
! 1 ∂2 G 1 ∂G
0
ρ t, x ¼ 2 Tij0 dydτ + 2 fi dSðyÞdτ
c0 vðτÞ ∂yi ∂yj c0 SðτÞ ∂yi
T T
(1.80)
ðT ð
1 ∂G
+ ρ0 Vn dSðyÞdτ,
c02 SðτÞ ∂τ
T
whose physical explanation has been given during the derivation of gener-
alized Green’s function formula.
Eq. (1.80) is a basic expression including the effect of a solid boundary
on sound generation. Among which, the first term denotes the sound
source due to the moving volume belonging to a quadruple source,
whereas the second term represents the source due to the force belonging
to dipole source, and the last term stands for the source due to the displace-
ment of the moving body belonging to a monopole source. This equation
can be applied to any aeroacoustic problem with the effect of a solid
boundary.
Especially, for the third term, the partial derivative with respect to time
needs to be moved out of the integration. Substituting Ψ ¼ Vi ∂G ∂xi into
Eq. (1.60) results in:
ðT ð ðT ð
d ∂G ∂ ∂G
0¼ Vi dydτ ¼ Vi dydτ
T dτ vc ðτÞ ∂xi T vc ðτÞ ∂τ ∂xi
ðT ð (1.83)
∂G
Vj nj Vi dSðyÞdτ,
T SðτÞ ∂xi
where the second term takes a negative sign because nj is the outward normal
vector of v(τ) and the outward normal vectors of vc(τ) and v(τ) oppose each
other. Again, in terms of the divergence theorem, the third term of
Eq. (1.82) can be expressed as:
ðT ð ðT ð
∂G ∂G
ρ0 Vn dSðyÞdτ ¼ ρ0 ni Vi dSðyÞdτ
SðτÞ ∂τ SðτÞ ∂τ
T T
(1.84)
ðT ð
∂ ∂G
¼ ρ0 Vi dydτ:
vc ðτÞ ∂yi ∂τ
T
Moreover,
ðT ð
∂G
ρ0 Vn dSðyÞdτ
SðτÞ ∂τ
T
(1.85)
ðT ð
∂ ∂G ∂G ∂ ∂G
¼ ρ0 Vj + Vi + Vi dydτ:
vc ðτÞ ∂yj ∂τ ∂xi ∂τ ∂xi
T
26 Fundamentals of aeroacoustics with applications to aeropropulsion systems
For a rigid body, the projection of the velocity on the body is zero, i.e.,
∂Vi
∂yi ¼ 0, hence,
∂ ∂G ∂G ∂ ∂G
Vj + Vi + Vi
∂yj ∂τ ∂xi ∂τ ∂xi
∂G
2
∂Vi ∂G ∂2 G ∂Vi ∂G ∂2 G
¼ Vj + Vj + Vj Vi + + Vi :
∂yj ∂τ ∂yj ∂xi ∂yj ∂xi ∂τ ∂xi ∂τ∂xi
Considering ∂G ∂G
∂xi ¼ ∂yi , it is shown that:
∂ ∂G ∂G ∂ ∂G ∂G DV i ∂2 G
Vj + Vi + Vi ¼ Vj Vi , (1.86)
∂yj ∂τ ∂xi ∂τ ∂xi ∂xi Dτ ∂xj ∂xi
Notice that the integration of any function f(τ) with δ function gives:
ðT X f ðτi Þ
f ðτÞδ½gðτÞdτ ¼
dgðτi Þ , (1.88)
i
T dτ
Basic equations of aeroacoustics 27
where τi is the ith root determined by the retarded time equation g(τi) ¼ 0,
whereas:
R jx2yj
g ðτ Þ ¼ τ t + ¼τt + : (1.89)
c0 c0
Its derivative is:
dg 1 R dy R
¼1 ¼ 1 M, (1.90)
dτ c0 R dτ R
where R ¼ x 2 y, M ¼ V
c0 . Integrating Eq. (1.87) with respect to time gives:
ð
0
! 1 ∂2 Tij0
ρ t, x ¼ dξ
4πc02 ∂xi ∂xj vðt0 Þ Rj1 Mr j τ¼τe
ð
1 ∂ fi
2 dSðξÞ
4πc0 ∂xi Sðt0 Þ Rj1 Mr j τ¼τe
ð (1.91)
1 ∂ ρ0 ai
2 dξ
4πc0 ∂xi vc ðt0 Þ Rj1 Mr j τ¼τe
ð
1 ∂2 ρ0 Vi Vj
+ dξ,
4πc0 ∂xi ∂xj vc ðt0 Þ Rj1 Mr j τ¼τe
2
where Vn0 denotes the dot product of the outward normal vector of vc(τ)and
the moving velocity of the body, i.e., Vn0 ¼ Vn.
When the body is static, i.e., a 5 M 5 V 5 0, then Eq. (1.91) becomes:
ð
0
! 1 ∂2 Tij0 R
ρ t, x ¼ y, t dy
4πc02 ∂xi ∂xj vðt0 Þ R c0
ð (1.92)
1 ∂ fi R
2 y, t dSðyÞ,
4πc0 ∂xi Sðt0 Þ R c0
which was presented by Curle [3]. Now it is named Curle’s equation.
For some specific cases, its far-field solution is interesting. Consider any
time-correlation function A(τ), similar to the integration function in
Eq. (1.91), by introducing
^ AðτÞ
A¼ ,
RjC { j τ¼τe
where C† ¼ 1 Mr stands for the Doppler factor. Notice that:
∂g ∂g ∂τe
+ ¼ 0: (1.93)
∂xi τe ∂τe x ∂xi
It is shown that:
∂τe Ri
¼ :
∂xi c0 C { R τ¼τe
whereas for a specific function R, i.e., the distance from the observer to the
source point, it is shown that:
∂R Ri Ri ∂R ∂yj Ri
¼ ¼ :
∂xi R c0 C R ∂yj ∂τe RC {
{
^
In this case,
^
for any function similar to A, when R ! ∞,it is easy to
verify that ð∂A=∂xi Þτe ¼ OðR2 Þ. Therefore, the ^
first and second order par-
tial derivatives with respect to xi for function A are:
^
∂A Ri ∂ A
¼ { {
+ O R2 ,
∂xi c0 C R ∂τ jC j τ¼τe
2
Basic equations of aeroacoustics 29
and
^
∂A Ri Rj ∂ 1 ∂ A
¼ 2 { 3 + O R2
∂xi ∂xj c0 C R ∂τ C { ∂τ jC { j τ¼τe
∂2 ρ0 2 ∂2 ρ0 ∂2 Tij0
c ¼ , (1.96)
∂τ2 0
∂y0i ∂y0j ∂y0i ∂y0j
30 Fundamentals of aeroacoustics with applications to aeropropulsion systems
where the Lighthill’s stress tensor Tij0 ¼ ρui0 uj0 eij + δij[(p p0) c20 ρ0 ] is
defined by the relative velocity
u0i ¼ ui δ1i U
In practical problems, it is more convenient to use a ground-fixed coor-
dinate. It is shown that Eq. (1.96) can be written by:
D20 ρ0 2 ∂2 ρ0 ∂2 Tij0
c ¼ , (1.97)
Dτ2 0
∂yi ∂yj ∂yi ∂yj
where
D0 ∂ ∂
¼ +U :
Dτ ∂τ ∂y1
Eq. (1.97) is named a generalized Lighthill equation, which is an exten-
sion of Lighthill’s equation in moving media.
To derive the solution of generalized Lighthill’s equation, consider
sound generation by a moving body with velocity Vs and volume vc(τ)
bounded by S(τ). Now the right term of Eq. (1.97) is assumed to be known,
and in terms of generalized Green’s function formula, it is shown that:
ðT ð
0
1 ! ∂2 Tij0
ρ t, x ¼ 2 G dydτ
c0 vðτÞ ∂yi ∂yj
T
ðT ð
∂ Vn0 D0 0 0 ∂ Vn0 D0
+ G + ρ ρ + G dSðyÞdτ,
SðτÞ ∂n c02 Dτ ∂n c02 Dτ
T
where Green’s function satisfies Eq. (1.57) and the causality condition of
Eq. (1.58), besides, Vn0 ¼ V0 n,V0 ¼ Vs iU, and ∂n ∂
¼ ni ∂y∂ i . Applying
the same approach within Section 1.1.3 yields:
ðT ð ðT ð
! 1 ∂2 G 1 ∂G
0
ρ t, x ¼ 2 Tij0 dydτ + 2 fi dSðyÞdτ
c0 vðτÞ ∂yi ∂yj c0 S ðτ Þ ∂yi
T T
(1.98)
ðT ð
1
ni h0i dSðyÞdτ,
c02 SðτÞ
T
where
∂ h 0 0 i D0 ρ0
h0i ¼ G ρui uj eij + δij ðp p0 Þ + Vi0
∂yj Dτ
(1.99)
D G ∂G
+ Vi0 ρ0 + ρui 0 u0j
0
:
Dτ ∂yj
Basic equations of aeroacoustics 31
Hence,
ð ð ð
d ∂Gρu0j D0 ∂Gρuj
0
∂Gρu0j
dy ¼ dy + Vn0 dSðyÞ
dτ vðτÞ ∂yj vðτÞ Dτ ∂yj SðτÞ ∂yj
ð
D0 Gρu0i 0
∂Gρu0j
¼ ni + ui dSðyÞ:
SðτÞ Dτ ∂yj
This expression is zero after the integration with respect to time τ.
Therefore, it is shown that only the last term of Eq. (1.101) remains after
some mathematical treatment, and Eq. (1.98) becomes:
ðT ð ðT ð
! 1 ∂2 G 1 ∂G
0
ρ t, x ¼ 2 Tij0 dydτ + 2 fi dSðyÞdτ
c0 vðτÞ ∂yi ∂yj c0 SðτÞ ∂yi
T T
(1.102)
ðT ð
1 D0 G
+ 2 ρ0 Vn0 dSðyÞdτ,
c0 SðτÞ Dτ
T
where Vn0 is the projection of the relative velocity in the normal direction of
the body. Eq. (1.102) is a general expression for a moving body with flow
media. The first term denotes the sound caused by a volume source belong-
ing to a quadruple source, the second terms stands for the sound due to the
32 Fundamentals of aeroacoustics with applications to aeropropulsion systems
force on the body belonging to a dipole source, and the last term represents
the sound due to the displacement of the volume of a moving body belong-
ing to a monopole source.
References
[1] J. Lighthill, On sound generated aerodynamically I: general theory, Proc. Roy. Soc. A
211 (1952) 564–587.
[2] J. Lighthill, Early development of an “acoustic analogy” approach to aeroacoustic the-
ory, AIAA J. 20 (1982) 449–450.
[3] N. Curle, The influence of solid boundaries upon aerodynamic sound, Proc. Roy. Soc.
A 231 (1955) 505–514.
[4] J.E. Ffowcs Williams, D.L. Hawkings, Sound generation by turbulence and surfaces in
arbitrary motion, Philos. Trans. R. Soc. Lond. Ser. A 264 (1969) 321–342.
[5] S.K. Lele, Compact finite difference schemes with spectral-like resolution, J. Comput.
Phys. 103 (1992) 16–42.
[6] C.K.W. Tam, J.C. Webb, Dispersion-relation-preserving finite difference schemes for
computational acoustics, J. Comput. Phys. 107 (1993) 262–281.
[7] T. Colonius, S.K. Lele, Computational aeroacoustics: progress on nonlinear problems
of sound generation, Prog. Aerosp. Sci. 40 (2004) 345–416.
[8] C.K.W. Tam, Recent advances in computational aeroacoustics, Fluid Dyn. Res.
38 (2006) 591–615.
[9] E. Envia, A.G. Wilson, D.L. Huff, Fan noise: A challenge to CAA, Int. J. Comput.
Fluid Dyn. 18 (6) (2004) 471–480.
[10] N. Peake, A. Parry, Modern challenges facing turbomachinery aeroacoustics, Annu.
Rev. Fluid Mech. 44 (2012) 227–248.
[11] M. Goldstein, Aeroacoustics, McGraw-Hill International Book Company, 1976.
[12] C. Morfey, Acoustic energy in non-uniform flows, J. Sound Vib. 14 (1971) 159–170.
[13] M. Myers, Transport of energy by disturbances in arbitrary steady flows, J. Fluid Mech.
226 (1991) 383–400.
[14] M. Lowson, The sound field for singularities in motion, Proc. Roy. Soc. A 286 (1965)
559–572.
CHAPTER 2
Fundamentals of Aeroacoustics with Applications to Aeropropulsion Systems © 2021 Shanghai Jiao Tong
https://doi.org/10.1016/B978-0-12-408069-0.00002-6 University Press. 33
Published by Elsevier Inc.
34 Fundamentals of aeroacoustics with applications to aeropropulsion systems
Both of these papers did not predict the absolute amplitude of a radiated
sound field. This instead was accomplished by Gutin [4] in 1948. He pointed
out that, based on the linearized acoustic analysis, the steady aerodynamic
force exerted on each propeller blade can be represented by a dipole distri-
bution on the blade surface. To generate the same pressure distribution
caused by the rotating blades, it is assumed that the stationary disk surface
is covered with the sound sources radiating with specific phase and strength.
Then, by conducting the Fourier analysis for the solution, one can obtain the
amplitudes of the noise due to the torsion and the thrust, respectively. It is
worth mentioning that, although the solution was obtained under many
simplifications, the basic ideas revealed in this work still have a great effect
on the subsequent researches.
With the development of computers in the 1950s, more complicated
numerical computations could be realized. Hence, many simplifications
are not necessary anymore.
Hubbard [5] further developed Gutin’s theory to calculate the fluctuat-
ing amplitude and sound field at the points that are one chord-length away
from the blade tip. Also, he conducted some parameter studies and com-
pared the theoretical calculation with the experimental results. Moreover,
Garrick and Watkins [6] generalized Gutin’s theory to consider the effect
of forward flight. In their approach, the sound field was calculated by an
integration over the propeller disk surface, this method was further simpli-
fied as the integration over an effective ring, thus the formulae for the near
field as well as the far field could be obtained. The analysis for a specific pro-
peller indicates the maximum noise point would move forward as the for-
ward flight speed increases.
Furthermore, Watkins and Deming [7] extended the work of Garrick
and Watkins to investigate the influence of spanwise and chordwise load dis-
tribution on the harmonic distribution as well as the total sound pressure
level. According to the calculation and analysis, with the radial distribution
of thrust and torque prescribed, the noise distributions showed negligible
difference compared with the results obtained through an effective radius
simplification. Therefore, it is appropriate to concentrate propeller force
on an effective radius to take into the effect of real load distribution when
concerning the noise variation trend. Inevitably, this simplification will
bring about certain errors in the prediction of total sound pressure level.
On the other hand, it was found that the chordwise load distribution had
Another random document with
no related content on Scribd:
“We’ll take this one,” and there were four thousand credits in the
box.
Throw one take one, and there were eight thousand.
The eight became sixteen; then thirty-two; and the dealer lost his
urbanity completely. He looked just plain ugly.
“Maybe that’s enough for now,” Joan suggested. “After all, we
don’t want to take all the man’s money.”
“Tightwad’s trick, huh? Quit while yer ahead?” the dealer
sneered. “Why’n’cha let ’er ride just once more?”
“If you insist, we will,” Cloud said, “but I’m warning you it’ll cost
you thirty two more M’s.”
“That’s what you think, Buster—I think different. Call your play!”
“We’ll take it!” Cloud snapped. “But listen, you clever-fingered jerk
—I know just as well as you do that the top card is the king of clubs,
and the one below it is the trey of diamonds. So, if you want to stay
healthy, move slowly and be damned sure to lift just one card, not
two, and take it off the top and not the bottom!”
Glaring in baffled fury, the dealer turned up the king of clubs and
paid his loss.
At the next table the results were pretty much the same, and at
the third. At the fourth table, however, instead of pyramiding, they
played only single M-bills. They lost—won—lost—lost—won—lost—
won—lost. In twenty plays they were only two thousand credits
ahead.
“I think I’ve got it, Joan,” Cloud said then. “Coming up—eight, six,
jack, five, deuce?”
“Uh-uh. I don’t think so. Eight, six, jack, three, one, I think. The
trey of spades and the ace of hearts. A two-and-one shift with each
full cycle.”
“Um . . . m. Could be . . . but do you think the guy’s that smart?”
“I’m pretty sure of it, Storm. He’s the best dealer they have. He’s
been dealing a long time. He knows cards.”
“Well, if you’re done passing out compliments, how about calling
a play?” the dealer suggested.
“QX. We’ll take the eight for one M . . . and it is the eight, you
notice . . . let it ride . . . throw the six—without looking, of course . . .
we’ll take the jack for two M’s. . . .”
The host, accompanied by no less a personage than the
manager himself, had come up. They stood quietly and listened as
Cloud took three bills out of the box, leaving one, and went on:
“The next card is either a five or a trey. That M there is to find out
which it is.”
“Are you sure of that?” the manager asked.
“Not absolutely, of course,” Cloud admitted. “There’s one chance
in approximately fourteen million that both my partner and I are
wrong.”
“Very good odds. But since you lose in either case, why bet?”
“Because if it’s a trey, she solved your system first. If it’s a five, I
beat her to it.”
“I see, but that isn’t necessary.” The manager took the remaining
cards out of the rack, and, holding them carefully and firmly, wrapped
the M-note tightly around them. Then, picking up the two small
stacks of played cards, he handed the whole collection to Cloud, at
the same time signalling the dealer to go ahead with his game. “We’ll
be smothered in a crowd very shortly, and I would like very much to
play with you myself. Will you, sir and madame, be gracious enough
to continue play in private?”
“Gladly, sir,” Joan assented, at Cloud’s questioning glance. “If it
would not put you out too much.”
“I am delighted,” and, beckoning to a hovering waiter, he went on:
“We will have refreshments, of course. In uniform, you might possibly
prefer soft drinks? We have some very good Tellurian ginger ale.”
“That’d be fine,” Cloud said, even while he was thinking at the
Lensman in contact with his mind: “Safe enough, don’t you think? He
couldn’t be thinking of any rough stuff yet.”
“Perfectly safe,” Nordquist agreed. “He’s just curious. Besides,
he’s in no shape to handle even the Vortex Blaster alone, to say
nothing of the task force he knows would be here two hours after
anything happened to either of you.”
The four strolled in friendly fashion to the suggested private
room. As soon as they were settled:
“You said the top card would be either a five or a trey,” the
manager said. “Shall we look?”
It was the trey of spades. “Congratulations, Joanie, a mighty
swell job. You really clobbered me on that one.” He shook her hand
vigorously, then handed the bill to the manager. “Here’s your M-note,
sir.”
“I couldn’t think of it, sir. No tipping, you know. . . .”
“I know. Not a tip, but your winnings. I called the play, remember.
Hence, I insist.”
“Very well, if you insist. But don’t you want to look at the next
one?”
“No. It’s the ace of hearts—can’t be anything else.”
“To satisfy my own curiosity, then.” The manager flipped the top
card delicately. It was the ace of hearts. “No compulsion, of course,
but would you mind telling me how you can possibly do what you
have just done?”
“I’ll be glad to,” and this was the simple truth. Cloud had to
explain, before the zwilniks began to suspect that they were being
taken by an organized force of Lensmen and snoopers. “We aren’t
even semi-habitual gamblers. The lieutenant-commander is Doctor
Joan Janowick, the Patrol’s ace designer of big, high-speed
electronic computers, and I am Neal Cloud, a mathematical analyst.”
“You are ‘Storm’ Cloud, the Vortex Blaster,” the manager
corrected him. “A super-computer yourself. I begin to see, I think . . .
but go ahead, please.”
“You undoubtedly know that random numbers, which underlie all
games of chance, must be just that—purely random, with nothing
whatever of system or of orderliness in their distribution. Also that a
stacked deck, by definition, is most decidedly not random. We were
kicking that idea around, one day, and decided to study stacked
decks, to see how systematic such distributions actually were. Well
—here’s the new part—we learned that any dealer who stacks a
deck of cards does so in some definite pattern; and that pattern,
whether conscious or unconscious, is always characteristic of that
one individual. The more skilled the dealer, the more complex,
precise, complete, and definite the pattern. Any pattern, however
complex, can be solved; and, once solved, the cards might just as
well be lying face up and all in sight.
“On the other hand, while it is virtually impossible for any dealer
to shuffle a deck into a really random condition, it can approach
randomness so nearly that the patterns are short and hence very
difficult to solve. Also, there are no likenesses or similarities to help.
Worst of all, there is the house leverage—the sevens of hearts,
diamonds, and clubs, you know—of approximately five point seven
seven percent. So it is mathematically certain that she and I would
lose, not win, against any dealer who was not stacking his decks.”
“I . . . am . . . surprised. I’m amazed,” the manager said. He was,
too; and so was the host. “Heretofore it has always been the guest
who loses by manipulation, not the house.” It is noteworthy that
neither the manager nor host had at any time denied, even by
implication, that their games of “chance” were loaded. “Thanks,
immensely, for telling me. . . . By the way, you haven’t done this very
often before have you?” the manager smiled ruefully.
“No.” Cloud smiled back. “This is the first time. Why?”
“I thought I would have heard of it if you had. This of course
changes my mind about wanting to deal to you myself. In fact, I’ll go
farther—any dealer you play with here will be doing his level best to
give you a completely random distribution.”
“Fair enough. But we proved our point, which was what we were
primarily interested in, anyway. What’ll we do with the rest of the day,
Joan—go back to the ship?”
“Uh-uh. This is the most comfortable place I’ve found since we
left Tellus, and if I don’t see the ship again for a week it’ll be at least
a week too soon. Why don’t you send a boy out with enough money
to get us a chess kit? We can engage this room for the rest of the
day and work on our game.”
“No need for that—we have all such things here,” the host said
quickly. “I’ll send for them at once.”
“No no—no money, please,” the manager said. “I am still in your
debt, and as long as you will stay you are my guests. . . .” he
paused, then went on in a strangely altered tone: “But chess . . . and
Janowick . . . Joan Janowick, not at all a common name . . . surely
not Past Grand Master Janowick? She—retired—would be a much
older woman.”
“The same—I retired for lack of time, but I still play as much as I
can. I’m flattered that you have heard of me.” Joan smiled as though
she were making a new and charming acquaintance. “And you? I’m
sorry we didn’t introduce ourselves earlier.”
“Permit me to introduce Host Althagar, assistant manager. I am
called Thlasoval.”
“Oh, I know of you, Master Thlasoval. I followed your game with
Rengodon of Centralia. Your knight-and-bishop end game was a
really beautiful thing.”
“Thank you. I am really flattered that you have heard of me. But
Commander Cloud. . . ?”
“No, you haven’t heard of him. Perhaps you never will, but
believe me, if he had time for tournament play he’d be high on the
Grand Masters list. So far on this cruise he’s won one game, I’ve
won one, and we’re on the eighty fourth move of the third.”
The paraphernalia arrived and the Tellurians set the game up
rapidly and unerringly, each knowing exactly where each piece and
pawn belonged.
“You have each lost two pawns, one knight, and one bishop—in
eighty three moves?” Thlasoval marveled.
“Right,” Cloud said. “We’re playing for blood. Across this board
friendship ceases; and, when dealing with such a pure unadulterated
tiger as she is, so does chivalry.”
“If I’m a tiger, I’d hate to say what he is.” Joan glanced up with a
grin. “Just study the board, Master Thlasoval, and see for yourself
who is doing what to whom. I’m just barely holding him: he’s had me
on the defensive for the last forty moves. Attacking him is just like
trying to beat in the side of a battleship with your bare fist. Do you
see his strategy? Perhaps not, on such short notice.”
Joan was very willing to talk chess at length, because the fact
that Fairchild’s Chickladorian manager was a chess Master was an
essential part of the Patrol’s plan.
“No . . . I can’t say that I do.”
“You notice he’s concentrating everything he can bring to bear on
my left flank. Fifteen moves from now he’d’ve been focused on my
King’s Knight’s Third. Three moves after that he was going to
exchange his knight for my queen and then mate in four. But, finding
out what he was up to, I’ve just derailed his train of operations and
he has to revise his whole campaign.”
“No wonder I didn’t see . . . I’m simply not in your class. But
would you mind if I stay and look on?”
“We’ll be glad to have you, but it won’t be fast. We’re playing
strict tournament rules and taking the full four minutes for each
move.”
“That’s quite all right. I really enjoy watching Grand Masters at
work.”
Master though he was, Thlasoval had no idea at all of what a
terrific game he watched. For Joan Janowick and Neal Cloud were
not playing it; they merely moved the pieces. The game had been
played long since. Based upon the greatest games of the greatest
masters of old, it had been worked out, move by move, by chess
masters working with high-speed computers.
Thus, while Joan and Storm were really concentrating, it was not
upon chess.
Chapter 14
[3] A few months later, Joan did send him the full
game, which white of course won. Thlasoval studied
it in secret for over five years; and then, deciding
correctly that he never would be able to understand
its terrifically complex strategy, he destroyed the
tape. It is perhaps superfluous to all that this game
was never published. E.E.S.
Chapter 15